Professional Documents
Culture Documents
Biotechnology of Vitamins, Pigments and Growth Factors
Biotechnology of Vitamins, Pigments and Growth Factors
Biotechnology of Vitamins, Pigments and Growth Factors
Edited by
ERICK J. VANDAMME
Laboratory of General and Industrial Microbiology,
State University of Ghent, Belgium
No responsibility is assumed by the Publisher for any injury and/or damage to persons or
property as a matter of products liability, negligence or otherwise, or from any use or
operation of any methods, products, instructions or ideas contained in the material herein.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval
system, or transmitted in any form or by any means, electronic, mechanical, photocopying,
recording, or otherwise, without the prior written permission of the publisher.
To Mireille
PREFACE
Vitamins and related growth factors belong to the few chemicals with a
positive appeal to most people; the name evokes health, vitality, fitness,
strength .... each one of us indeed needs his daily intake of vitamins, which
should normally be provided via a balanced and varied diet. However, current
food habits or preferences, or food processing and preservation methods do
not always assure a sufficient natural daily vitamin supply, even for a healthy
human being; this is all the more true for stressed or sick individuals. Although
modern society is seldom confronted with the notorious avitaminoses of the
past, they do still occur frequently in overpopulated and poverty- and
famine-struck regions in many parts of the world. Apart from their in-vivo
nutritional-physiological roles as growth factors for man, animals, plants and
micro-organisms, vitamin compounds are now being introduced increasingly as
food/feed additives, as medical-therapeutical agents, as health-aids, and also
as technical aids. Indeed, today an impressive number of processed foods,
feeds, cosmetics, pharmaceuticals and chemicals contain extra added vitamins
or vitamin-related compounds, and single or multivitamin preparations are
commonly taken or prescribed.
These reflections do indicate that there is an extra need for vitamin supply,
other than that provided from plant and animal food resources. Most added
vitamins are indeed now prepared chemically and/or biotechnologically via
fermentation/bioconversion processes.
Similarly, other related growth factors, provitamins, vitamin-like com-
pounds, i.e. special unsaturated fatty acids, gibberellins and certain pigments-
some of which are increasingly used in agriculture, food/feed production or
processing and as health aids-are equally important biotechnological pro-
ducts, where production via microbial fermentation or micro-algal bioconver-
sion is now applied industrially. Indeed, biotechnology-based on bacteria,
yeasts, fungi and micro-algae-here again has been very instrumental in
procuring sufficient amounts of several of these valuable complex molecules via
vii
viii Preface
natural processes, although for certain products there is fierce competition with
chemical synthesis. Abundant and excellent literature is available on the
chemical properties, biochemistry, nutritional aspects and clinical aspects of
vitamins and related products, while that on microbial synthesis and produc-
tion methods is rather scarce or difficult to find.
In this respect, this book intends to assemble useful information on the
(potential) industrial synthesis of economically important vitamins, growth
factors and pigments, with emphasis on biotechnological aspects including
microbiology, genetics, biochemistry and bioprocess technology; so far, such
information is scattered widely in the scientific literature: for some products,
only secrecy and sparse data are available, other excellent volumes deal with
only one specific vitamin compound, while others then stress the chemical
synthesis processes. Therefore, I felt that there was a scientific need for a
biotechnological survey of the world of vitamins and related compounds
synthesis.
The help of several colleagues and friends in suggesting potential authors for
difficult-to-get chapters has been invaluable in constructing a rather com-
prehensive volume; so was the positive interaction with all my contributors. In
this respect, I am particularly indebted to:
Dr S. Anderson, Genentech, USA; Dr G. C. Barrere, Rhone-Poulenc,
France; Dr E. Cerda-Olmedo, University of Seville, Spain; Dr D. De Buyser,
N. V. Vandemoortele, Belgium; Dr. D. Defterdarovic, Pliva, Yugoslavia;
Professor Dr A. L. Demain, MIT, USA; Dr J. Florent, Rhone-Poulenc,
France; Dr A. Furuya, Kyowa Hakko Kogyo Co., Ltd, Japan; Dr H. Nelis,
University of Ghent, Belgium; Dr L. Segers, Orffa, Belgium; Dr S. Shimizu,
Kyoto University, Japan; Dr G. Smits, Tiense Suikerraffinaderij, Belgium;
Professor Dr Y. Tani, Kyoto University, Japan.
The editor also thanks the staff of Elsevier Science Publishers Ltd.
I suspect that my wife, Mireille, could only have withstood my 'mental
absence', strengthened with multivitamin preparations, although my sole
vitamin-shot was her encouraging and moral support during this biotechnologi-
cal enterprise.
E. J . VANDAMME
CONTENTS
Preface . . . . . . vii
List of Contributors . xi
Water-Soluble Vitamins
9. Microbial Synthesis of Vitamin Bl (Thiamine) (A. Iwashima) 137
10. Microbial Production of Vitamin B2 (Riboflavin) (T. Kutsal &
M. T. Ozbas) . . . . . . . . . . . . . . . . . . . . . 149
11. Microbial Production of D-Ribose (K. Sasajima & M. Yoneda) 167
12. Pantothenic Acid (Vitamin B s), Coenzyme A and Related
Compounds (S. Shimizu & H. Yamada). . . . . . . . . . 199
13. Microbial Production of Vitamin B6 and Derivatives (Y. Tani) 221
ix
x Contents
Index. 431
LIST OF CONTRIBUTORS
D. BLECHSCHMIDT,
Friedrich Schiller University Jena, Department of General Microbiology,
DDR-6900 Jena, Neugasse 24, GDR
L. J. BOROWITZKA,
Western Biotechnology Ltd, 2-6 Railway Parade, Bayswater, WA 6063,
Australia
M. A. BOROWITZKA,
Algal Biotechnology Laboratory, School of Biological and Environmental
Sciences, Murdoch University, Murdoch, WA 6150, Australia
B. BRUECKNER,
Friedrich Schiller University lena, Department of General Microbiology,
DDR-6900 lena, Neugasse 24, GDR
E. CERDA-OLMEDO,
Departmento de Genetica y Biotecnia, Universidad de Sevilla, Apartado
1095, Sevilla, Spain
A. P. DE LEENHEER,
Laboratoria voor Medische Biochemie en voor Klinische Analyse,
Rijksuniversiteit Ghent, Harelbekestraat 72, B-9000 Ghent, Belgium
V. DELle,
PLIVA Pharmaceutical, Chemical, Food and Cosmetic Industry, Research
Institute, Zagreb, Yugoslavia
G. FERNI,
Farmitalia Carlo Erba, Via dei Gracchi 35, 20146, Milano, Italy
A. FURUYA,
Kyowa Hakko Kogyo Co. Ltd, Tokyo Research Laboratories, 3-6-6
Asahi-machi, Machida-shi, Tokyo 194, Japan
L. GAROFANO,
Farmitalia Carlo Erba, Via dei Gracchi 35, 20146, Milano, Italy
A. GREIN,
Farmitalia Carlo Erba, Via dei Gracchi 35, 20146, Milano, Italy
A. IWASHIMA,
Department of Biochemistry, Kyoto Prefectural University of Medicine,
Kawaramachi-Kirokoji, Kamigyo-ku, Kyoto 602, Japan
Y. IZUMI,
Department of Agricultural Chemistry, Kyoto University, Kyoto 606, Japan
xi
xii List of Contributors
T. KUTSAL,
Hacettepe University, Chemical Engineering Department, 06532 Beytepe,
Ankara, Turkey
P. MARGALITH,
Department of Food Engineering & Biotechnology, Technion-Israel
Institute of Technology, Haifa, 32000, Israel
H. J. NELIS,
Laboratoria voor Medische Biochemie en voor Klinische Analyse,
Rijksuniversiteit Ghent, Harelbekestraat 72, B-9000 Ghent, Belgium
M. T. (hBAS,
Hacettepe University, Chemical Engineering Department, 06532 Beytepe,
Ankara, Turkey
K. SASAJIMA,
Central Research Division, Takeda Chemical Industries Ltd, 2-17-85
Jusohonmachi, Yodogawa-ku, Osaka 532, Japan
G. SCHNEIDER,
Institute of Plant Biochemistry, Academy of Sciences of the GDR, DDR-
4050 Halle, Weinberg 3, GDR
G. SEMBDNER,
Institute of Plant Biochemistry, Academy of Sciences of the GDR, DDR-
4050 Halle, Weinberg 3, GDR
S. SHIMIZU,
Department of Agricultural Chemistry, Kyoto University, Kyoto 606, Japan
C. SPALLA,
Farmitalia Carlo Erba, Via dei Gracchi 35, 20146, Milano, Italy
D. SUNIC,
PLIVA Pharmaceutical, Chemical, Food and Cosmetic Industry, Research
Institute, Zagreb, Yugoslavia
K. TAKAYAMA,
Kyowa Hakko Kogyo Co. Ltd, Tokyo Research Laboratories, 3-6-6
Asahi-machi, Machida-shi, Tokyo 194, Japan
Y. TANI,
Research Center for Cell and Tissue Culture, Faculty of Agriculture, Kyoto
University, Kyoto 606, Japan
E. J . VANDAMME,
Laboratory of General and Industrial Microbiology, Faculty of Agricultural
Services, State University of Ghent, Coupure Links 653, B-9000 Ghent,
Belgium
D. VLASIC,
PLIVA Pharmaceutical, Chemical, Food and Cosmetic Industry, Research
Institute, Zagreb, Yugoslavia
H. YAMADA,
Department of Agricultural Chemistry, Kyoto University, Kyoto 606, Japan
M. YONEDA,
Corporate Strategy, Takeda Chemical Industries Ltd, 2-17-85
Jusohonmachi, Yodogawa-ku, Osaka 532, Japan
Chapter 1
1 HISTORICAL
The start of the history of vitamins can be traced back to 400 Be, when
Hippocrates reported that eating liver could cure night-blindness. Much later,
in the 16th century, the therapeutical effects of lemon juice against scurvy or
scorbut became known; scorbut had inter alia caused the loss of 100 crew
members on Vasco da Gama's journey around Cape Hope. The English ship
doctor James Lind studied this disease further and described in 1757 in his
book A Treatise of Scurvy the beneficial effect of eating fresh vegetables and
fruits in preventing it. For another nutritional deficiency disease (already
mentioned in 1762 by Oviedo), the Italian doctor Francesco Frapoli used the
name pellagra (pella = skin; agra = rough). In the 19th century in Japan, the
Hikan child disease (keratomalacia and xerophthalmia) was successfully
treated by including ale-fat, cod liver oil or chicken liver in the diet. Trousseau
discovered that cod liver oil and also, direct sunlight, had a curing effect on
rickets, a disease already well described by Whistler in 1645. In the Far-East,
when hulled rice was replaced by de hulled or polished rice, a sharp increase in
the occurrence of beriberi was observed. In 1897 Eijkman observed that
poultry fed with polished rice developed polyneuritis, a disease very similar to
human beriberi. This disease could also be prevented and cured by feeding
rice and the silver fleece of the rice kernel. Grijns in 1901 hypothesised that
beriberi was caused by a protecting factor, which was obviously lacking in
dehulled rice. We now know that all these diseases are a result of nutritional
vitamin deficiencies (Machlin, 1984).
Around 1910, Hopkins in the UK and Osborne & Mendel in the USA
initiated modern vitamin research and developed a theory, stating that diseases
such as scurvy, pellagra, rickets, beriberi, etc., are the result of a lack of
certain essential food components. The Polish chemist Casimur Funk isolated
in 1912 from rice bran a beriberi-preventing compound, displaying chemical
1
2 E. J. Vandamme
properties of an amine; this led him in 1912 to coin the name vitamin for this
type of compounds.
In 1913, McCollum & Davis demonstrated a liposoluble factor A in butter
fat and egg yolk, and in 1915, a water-soluble factor B was found in
wheat-germ. It was Drummond in 1920 who named the fat-soluble factor,
vitamin A; the water-soluble anti-beriberi factor was named vitamin B; the
water-soluble anti-scorbut factor vitamin C. In 1925, the fat-soluble anti-rickets
factor was named vitamin D. After 1930, discovery and isolation of several
other vitamins followed quickly and their structure, nutritional and chemical
properties and synthesis were studied in great detail in the following decades.
These aspects of vitamins and growth factors are compiled in several
excellent standard references (Goodwin, 1963; Sebrell & Harris, 1971; De
Luca, 1978; Machlin, 1984; De Leenheer et al., 1985; Diplock, 1985; Chytil &
McCormick, 1986; Adrian, 1988; Friedrich, 1988).
Table 1
Survey of Vitamins and Growth Factors
(continued)
4 E. J. Vandamme
Table l---contd.
The vitamin content was originally and still often is expressed in Interna-
tional units (IU), which relate to the biological activity of a certain amount of
pure vitamin towards a specific test animal.
Presently, it is common practice to express the vitamin content as mg or f-tg
per 100 g of material; conversion factors for IU values into weight units are also
given in Table 1. Recommended daily intake per person per day (FAO/WHO
data) is also given.
Quantitative assays for vitamin content, in foodstuffs, concentrated mix-
tures, synthetic formulations, tissues, blood, fermentation broths, etc., are
very important and ever more sophisticated techniques (e.g. HPLC) are
introduced; for several vitamins, i.e. B 12 , biotin, etc., a microbiological
bioassay is still the method of choice (Freed, 1966; Berg & Behagel, 1972; De
Leenheer et al., 1985).
3 PHYSIOLOGICAL FUNCTIONS
Table 2
Water-Soluble Vitamins and their Corresponding Coenzymes
Vitamin Coenzyme Group Transfer
Nicotinamide Nicotinamide-adenine Hydrogen
(B3 or PP) dinucleotide (NAD+)
Nicotinamide-adenine Hydrogen
dinucleotide phosphate
(NADP+)
Riboflavin Flavine adenine mono- Hydrogen
(B2) nucleotide (FMN)
Flavine adenine di- Hydrogen
nucleotide (FAD)
Pyridoxine Pyridoxalphosphate Amino-group, decar-
(B6) (PLP) boxylation
Folic acid Tetrahydrofolic Formyl
(Bg) acid
Biotin Biotin (biocytin, Carboxyl
(H) E-N-biotinyllysine)
Pantothenic Coenzyme A Acyl
acid (Bs)
Thiamine Thiaminepyrophos- Cz-aldehyde, decar-
(B 1 ) phate (TPP) boxylation
Cyanocobalamin B12-coenzyme Carboxyl
(B 12)
Table 3
Important Biochemical Actions of Fat-Soluble Vitamins and of Vitamin C
Palleroni et a/., 1978; Institute of Food Technologists, 1980; Wong & Koehler,
1981). Indeed, in addition to the well-known provitamin-carotenoids, a range
of anthraquinone pigments, chlorophylls and several others have been demon-
strated in bacteria, yeasts, fungi and algae and are attractive as natural
colorant.
In this respect, more emphasis should be given to screening and research on
other natural pigment synthesis via safe micro-organisms; this would open up a
wider application area in agriculture, food, feed, chemicals, cosmetics and
pharmaceuticals.
Fungal gibberellins are already widely used as plant growth-regulators in
agriculture and horticulture and in the brewing industry (Jefferys, 1970;
Palmer, 1974; Brueckner et a/., this volume).
Details about technical applications of certain vitamins, pigments and
gibberellins are given in the corresponding chapters in this volume.
The staple food of man, including cereals, rice, potato, vegetables, fruits, milk,
fish, meat, eggs, forms his basic source of vitamins and growth factors. Adequate
nutrition should thus supply this daily vitamin need, which however increases
with physical exercise, pregnancy, lactation, active growth, reconvalescence,
drug abuse, stress, air pollution, etc. Pathological situations (intestinal
malresorption, stressed intestinal flora, liver/gall diseases, drug, antibiotic or
hormone treatment, enzyme deficiencies), can also lead to vitamin shortages
despite a sufficient intake. Malnourishment in many countries of the world also
asks for direct medical remediations, combined with diet adjustment. Vitamin-
enriched and medicated feed is used worldwide to procure healthy livestock.
However, derived concentrates or extracts from these vitamin-rich natural
food products find relatively little use in the food, feed, pharmaceutical or
cosmetic industry (except for w-3-fatty acids, vitamin E, etc.). Some of the
reasons are:
(a) the average daily intake of vitamins does not always seem to be supplied
solely via these natural products;
(b) the level of the natural plant/ animal source vitamins is usually relatively
low and fluctuates drastically;
(c) their organoleptic presentation and shelf-life is often far from optimal;
(d) vitamins are labile molecules during the process of preservation, storage
or preparation of foodstuffs and are generally very sensitive to pH, heat,
light, oxygen, etc.; water-soluble vitamins are easily lost by aqueous
extraction or manipulation of foods.
These factors have led to the industrial manufacturing of most vitamins and
growth factors. Current world production of important vitamins is given in
Table 4.
At the moment, several vitamins are produced only or mainly chemically
8 E. I. Vandamme
Table 4
Survey of Vitamins and Growth Factors-Production and Application
(continued)
Vitamins and Related Compounds via Micro-organisms 9
Table 4--contd.
a Listing of vitamin and growth factor producing companies or countries: 1. Abbott (USA); 2. BASF
(FRG); 3. Biocrops Ltd (UK); 4. China; 5. Coors Biotech Inc. (USA); 6. Cyanotech Inc (USA); 6a.
Dainippon Ink & Chemicals (Japan); 7. Degussa (FRG); 8. Duphar (The Netherlands); 9. Eastman-
Kodak (USA); 10. Efamol Holdings Ltd (UK); 11. Eisa Co. (Japan); 12. Eli Lilly (USA); 13. E. Merck
(FRG); 14. Farmitalia-Carlo Erba (Italy); 14a. Fujisawa (Japan); 15. Genex Corp. (USA); 15a. Gist
Brocades (The Netherlands); 16. Glaxo (Sefton Chern) (UK); 17. Henkel (FRG); 18. Hoffmann-
Laroche (Switzerland); 19. ICI (UK); 20. Koor Foods (Israel); 21. Kyowa Hakko Kogyo (Japan); 22.
Martek Inc. (USA); 23. Medimpex (Yugoslavia); 24. Merck S & D (USA); 25. Microbial Products Inc.
(USA); 26. Nippon Oil & Fats Co. Ltd (Japan), 27. Nippon Zeon (Japan); 28. Pao Yeh (Taiwan); 29.
Pierrel (Italy); 30. Pfizer Inc. (USA); 31. Pliva (Yugoslavia); 32. Q. P. Corporation (Japan); 33.
Rhone-Poulenc (France); 34. Riken Vitamin (Japan); 35. RMC (Evening Primrose Oil Company Ltd)
(UK); 36. Roussel-UcLaF (France); 37. Shiseido Co. Ltd (Japan); 38. Sturge Biochemicals (UK); 39.
Sumitomo (Japan); 40. Synergen Inc. (USA); 41. Takeda (Japan); 42. Tanabe-Kongo (Japan); 43.
Tanabe (Japan); 44. USSR; 45. Vanetta (Italy); 46. Western Biotechnology Ltd (Australia); 47. Yuki
Gosei (Japan); 48. Yodo Gowa-Duphar (Japan).
REFERENCES
Adrian, J. (1988). Parat Dictionary of Food and Nutrition, VII. Ellis Horwood Series in
Food Science and Technology. Ellis Horwood, Chichester, UK.
Berg, T. M. & Behagel, H. A. (1972). Semi-automated method for microbiological
vitamin-assays. Appl. Microbiol., 23,531-42.
Borowitzka, M. A. & Borowitzka, L. J. (eds) (1988). Micro-algal Biotechnology.
Cambridge University Press, Cambridge, UK.
Brueckner et al. (1989). Fungal gibberellin production. In Biotechnology of
Vitamins, Pigments and Growth Factors, ed. E. J. Vandamme. Elsevier Science
Publishers, London, pp. 383-429
Chytil, P. & McCormick, D. B. (eds) (1986). Vitamins and Coenzymes. Methods in
Enzymology, Vols 122, 123. Academic Press, New York, London.
De Leenheer, A. P., Lambert, W. E. & Deruyter, M. G. M. (eds) (1985). Modern
Chromatographic Analysis of the Vitamins. Marcel Dekker, New York, Basel.
De Luca, H. F. (ed.) (1978). The fat-soluble vitamins. In Handbook of Lipid Research,
Vitamins and Related Compounds via Micro-organisms 11
1 HISTORICAL
v
c::
....Vo
L
111
U
I
C!l.
205·071 I
III
c::
111
....
L
I
vv
v iii c::C::
c:: Vv
U
c:: :c 0
0LL
111 1::111 111111
fo 105·931 )( v UU
I I
111 c::
III
.c c:: V c:: c:: c:: v C!l.C!l.
<{ ~
o
N
~ ~ ~
0
-0L I
.- U
I
U\.!!!
c:: 0 0 U M
~ c:: c:: c:: 111
~ U Ol~
c:: ~ ~
c:: c::
c::
jj
I
:> :> :>
:> d
6790~==~==~=====;~~====~~==~==~~~~====t=========
o 5 10 15
Time (min)
Fig. 1. Typical HPLC analysis of D. salina J3-carotene extract.
salt works. These ecosystems and the ponds are inhabited by wild-type D.
salina, non-carotenogenic Dunaliella species, predatory protozoa, halophilic
bacteria and brine shrimp (Borowitzka et al., 1986; Moulton et al., 1987b). In
the presence of this mixture of organisms, maintenance of cultures dominated
by the carotenogenic D. salina is the first priority of pond management. In an
environment dominated by the wild-type D. salina, introduction of an
18 L. J. Borowitzka & M. A. Borowitzka
'improved' D. salina strain is extremely difficult, and has not been reported on
a commercial scale. In the large, open-air ponds used for the commercial
culture of D. salina the major control of the culture available to the operator is
salinity (Borowitzka et al., 1986), although some additional control may be
exercised by manipulating nutrient concentrations.
In more isolated systems, as for example, concrete or plastic-lined raceway
ponds and tubular photobioreactors, wild-type D. salina and other competing
organisms may be excluded and the introduction of improved strains can be
achieved, at the cost of higher investment in plant and equipment.
Strain improvement, both in terms of growth characteristics and product
yield, can be achieved using mutagenesis and selection programmes. Improved
strain breeding using mating strains as in Chlamydomonas, a closely related
alga, is also possible (Craig et al., 1988). Further into the future, and awaiting
additional research, is the use of cell fusion techniques or the genetic
engineering manipulations developed for yeast cells (e.g. Spencer & Spencer,
1983).
In the laboratory, D. salina may be isolated into unialgal and axenic culture
either by repeated streaking on agar plates or by isolating clonal cultures from
single cells. The most commonly used medium for D. salina and related algae
is modified Johnson's medium (Borowitzka, 1988), although a range of other
media can also be used, as long as the salinity is appropriately adjusted with
NaCI.
Laboratory cultures are usually maintained on agar slants or in liquid
cultures, and sub-cultured every 1-2 months. Care must be taken to avoid
excessive drying out of the high salinity media. We have also successfully
maintained Dunaliella species, frozen in liquid nitrogen in our laboratory for
periods in excess of 12 months. In the field, stock cultures are best maintained
at saturating salinities where the likelihood of contamination by other
organisms is minimised.
Table 1
Summary of the Influence of Various Environmental Factors on
the Biomass and Jj-Carotene Content of Dunaliella salina. + =
Stimulating Effect; - = Inhibitory Effect; 0 = No Effect
5 BIOMASS PRODUCTION
1
I HOLDING PONDS
I I HARVESTERS
I
1
1
CULTURE MEDIUM
BIOMASS
(BRINE)
I
1
l DRY EXTRACT fJ-CAROTENE
!
CONCENTRATE
1
PURIFY
1
MILL
1
Dunaliella salina FORMULATION
powder containing
3% fJ-Carotene Suspensions in
Vegetable oil
(2%, 10%,20%,30%)
Water Dispersibles
Water Solubles
--- Final
Quality control
-- .....
PACKAGING
J PACKAGING
J
Fig. 3. Flow chart of the Dunaliella salina fJ-carotene production process used by
Western Biotechnology Ltd.
fJ-Carotene Production with Algae 21
7 BIOLOGICAL PROPERTIES
8 CHEMICAL SYNTHESIS
10 ECONOMICS
Just as the companies in the new industry guard the confidentiality of their
processes, costs of commercial production are not available. Moulton et aZ.
(1987a), however, present a bioeconomic model for the {:1-carotene production
process using D. salina.
Synthesised {:1-carotene is sold for a minimum of US$300 per kg {:1-carotene,
and at higher prices depending on the formulation. 'Natural' {:1-carotene
24 L. I. Borowitzka & M. A. Borowitzka
commands higher prices, with the highest price attainable for the nutritional
supplement application. Market price for 'natural' IJ-carotene is volatile and
in 1987-88 world market demand exceeds supply.
If a price range for formulations of natural IJ-carotene of US$500-1000 is
assumed, and IJ-carotene represents 10% of D. salina biomass, production
costs for D. salina should not exceed US$50-100 per kg. In fact, costs must be
significantly lower, to account for losses at each processing step, capital
expenses, marketing, packaging and distribution costs.
REFERENCES
Aasen, A. J., Eimhjellen, K. E. & Liaaen-Jensen, S. (1969). An extreme source of
p-carotene, Acta Chim. Scand., 23,2544-5.
Avron, M., Ben-Amotz, A. & Edelstein, S. (1987). Feed supplement. UK patent
application 2 189 675A.
Bauernfeind, J. C. (ed.) (1981). Carotenoids as Colourants and Vitamin A Precursors.
Academic Press, New York, pp. 938.
Ben-Amotz, A. (1980). Glycerol production in the alga Dunaliella. In Biochemical and
Photosynthetic Aspects of Energy Production, ed. A. San Pietro. Academic Press,
New York, pp. 191-208.
Ben-Amotz, A. & Avron, M. (1982). The potential use of Dunaliella for the production
of glycerol, p-carotene and high protein feed. In Biosaline Research: A Look to the
Future, ed. A. San Pietro. Plenum Publishing Corporation, New York, pp. 207-14.
Ben-Amotz, A. & Avron, M. (1983). On the factors which determine the massive
p-carotene accumulation in the halotolerant alga Dunaliella salina. Plant Physiol.,
72,593-7.
Ben-Amotz, A., Katz, A. & Avron, M. (1982). Accumulation of p-carotene in
halotolerant algae: purification and characterisation of p-carotene-rich globules
from Dunaliella bardawil (Chlorophyceae). J. Phycol., 18,529-37.
Ben-Amotz, A., Edelstein, S. & Avron, M. (1986). Use of the p-carotene rich alga
Dunaliella bardawil as a source of retinol. Br. Poult. Sci., 27,613-9.
Bloch, M. R., Sasson, J., Ginzburg, M. E., Goldman, Z., Ginzburg, B. Z., Garti, N.
& Perath, A. (1982). Oil products from algae. US patent no. 4.341 038.
Borowitzka, L. J. (1981). The microflora. Adaptions to life in extremely saline lakes.
Hydrobiologia, 81, 33-46.
Borowitzka, M. A. (1988). Algal growth media and sources of algal cultures. In
Micro-algal Biotechnology, ed. M. A. Borowitzka & L. J. Borowitzka, Cambridge
University Press, Cambridge pp. 456-65.
Borowitzka, L. J., Borowitzka, M. A. & Moulton, T. P. (1984). The mass culture of
Dunaliella salina for fine chemicals: from laboratory to pilot plant. Hydrobiologia,
116/117, 115-2l.
Borowitzka, L. J., Moulton, T. P. & Borowitzka, M. A. (1986). Salinity and the
commercial production of beta-carotene from Dunaliella salina. Nova Hedwigia,
Beih., 83, 224-9.
Borowitzka, M. A. & Borowitzka, L. J. (1988a) Dunaliella. In Microalgal
Biotechnology, ed. M. A. Borowitzka & L. J. Borowitzka. Cambridge University
Press, Cambridge, pp. 27-58.
Borowitzka, M. A. & Borowitzka, L. J. (1988b) Limits to growth and carotenogenesis
in laboratory and large-scale cultures of Dunaliella salina. In Algal Biotechnology,
ed. T. Stadler, J. Mollion, M-C. Verdus, Y. Karamanos, H. Morvan & D.
Christiaen. Elsevier Applied Science Publishers, Barking, pp. 371-82.
fJ-Carotene Production with Algae 25
Chen, B. J. & Chi, C. H. (1981). Process development and evaluation for algal glycerol
production. Biotech. Bioeng., 23, 1267-87.
Craig, R., Reichelt, B. Y. & Reichelt, J. L. (1988). Genetic engineering of
micro-algae. In Micro-algal Biotechnology, ed. M. A. Borowitzka & L. J.
Borowitzka. Cambridge University Press, Cambridge, pp. 415-55.
Curtain, C. C. & Snook, H. (1983). Method for harvesting algae. US patent no. 511
135.
Curtain, C. c., West, S. M. & Schlipalius, L. (1987). Manufacture of p-carotene from
the salt lake alga Dunaliella salina; the scientific and technical background. Aust. J.
Biotechnol., 1, 51-7.
Drokova, I. G. & Dovhorouka, S. I. (1966). Carotene-formation in Dunaliella salina
Teod. under the effect of some carbon sources. Ukransk. Bot. Zhour., 21, 59-62.
Federov, V. D., Maksimov, V. N. & Kromov, V. M. (1968). Effect of light and
temperature on primary production of certain unicellular green algae and diatoms.
Fiziol. Rast., 15, 640-5l.
Gibor, A. (1956). The culture of brine algae. Bioi. bull., Woods Hole, 3,223-9.
Grant, B. R. (1968). The effect of carbon dioxide concentration and buffer system on
nitrate and nitrite and assimilation by Dunaliella tertiolecta, J. Gen. Micro., 54,
327-36.
Kessler, J. O. (1982). Algal cell harvesting. US patent no. 4324067.
Kessler, J. O. (1985). Hydrodynamic focusing of motile algal cells. Nature, 313,
208-10.
KI1tui, H. (1981). Industrial and commercial uses of carotenoids. In Carotenoid
chemistry and Biochemistry, ed. G. Britton & T. W. Goodwin. Pergamon Press,
Oxford, pp. 309-28.
Klausner, A. (1986). Algaculture: Food for thought. Biotechnology, 4,947-53.
Lerche, W. (1937). Untersuchungen iiber die Entwicklung und Fortpflanzung in der
Gattung Dunaliella. Arch. fur Protistenk., 88, 236-9.
Loeblich, L. A. (1972). Studies on the brine flagellate Dunaliella salina. PhD thesis,
University of California, San Diego.
MacKinney, G. & Chichester, C. O. (1960). Biosynthesis of carotenoids. In
Comparative Biochemistry of Photoreactive Systems, ed. M. B. Allen. Academic
Press, New York, pp. 205-15.
Massyuk, N. P. (1965). Carbonate and bicarbonate as stimulators of growth and
carotene accumulation in Dunaliella salina Teod. Ukransk. Bot. Zhour., 22, 18-22.
Massyuk, N. P. (1966). Mass culture of the carotene-bearing alga Dunaliella salina
Teod. Ukransk. Bot. Zhour., 23, 12-19.
Massyuk, N. P. & Abdula, E. G. (1969). First experiment of growing carotene-
containing algae under semi-industrial conditions. Ukransk. Bot. Zhour., 26,21-7.
Mathews-Roth, M. M. (1982). Medical applications and uses of carotenoids. In
Carotenoid Chemistry and Biochemistry, ed. G. Britton & T. W. Goodwin.
Pergamon Press, Oxford, pp. 297-307.
Mayer, H. & Isler, O. (1971). Total syntheses. In Carotenoids, ed. O. Isler.
Birkhauser, Basle, pp. 328-575.
Mil'ko, E. S. (1962). Study the requirement of two Dunaliella spp. in mineral and
organic components if the medium, Moscow University Vestnik. Biologya, 6, 21-3.
Mil'ko, E. S. (1963). Effect of various environmental factors on pigment production in
the alga Dunaliella salina. Mikrobiologiya, 32,299-307.
Mironyuk, V. I. & Einor, L. O. (1968). Oxygen exchange and pigment content in
various forms of Dunaliella salina Teod. under conditions of increasing NaCl
content. Gidrobiol Zhournal, 4,23-9.
Mohn, F. H. (1988). Harvesting of micro-algal biomass. In Micro-algal Biotechnology,
ed. M. A. Borowitzka & L. J. Borowitzka. Cambridge University Press, pp.
395-414.
26 L. J. Borowitzka & M. A. Borowitzka
1 INTRODUCTION
Humans, like most other animals, need carotenoids but cannot synthesize
them. Our main suppliers are the fruits and vegetables, with lesser, mostly
indirect contributions from fungi, algae, and bacteria. Carotenoid production
is rarely the main objective of a culture, but colourful products rich in
carotenoids are obtained from various plants and algae. The oldest example is
saffron (Crocus sativus) , still planted in La Mancha, Spain, and elsewhere for
the stigmas of its flowers. Many fields of central Mexico become bright orange
in the summer with the flowers of cempasuchil (Tagetes ten uifolia , a kind of
marigold), whose dried petals are fed to chicken for meat and egg yolk colour.
Annatto is extracted from the seeds of a tropical tree, Bixa orellana. The most
familiar carotenoid-rich product is probably paprika, a powder made by
grinding red peppers (Capsicum annuum).
The carotenoids may be scarce in the diets of humans and animals for
various reasons. Many staple foods and feeds, including most of the cheap
sources of carbohydrates and fat, are poor in carotenoids, as witnessed by their
lack of colour. The long winters of many countries halt the natural production
of carotenoids, and these are relatively unstable and may not survive in stored
foods and feeds.
Organic chemists have devised several complete syntheses, which now
supply considerable amounts of f3-carotene and other purified carotenoids. The
current trend towards natural food colours and the discovery of salutary effects
of the carotenoids, beyond their important role as provitamin A, stimulate the
demand and press towards a biological production.
The fungi can hardly be considered as traditional food colourants, but the
relative ease of cultivation and the many possibilities for physiological,
genetical, and industrial manipulations make them attractive to biotech-
nologists. Several species are potential sources of carotenoids, and a large-
scale process has been based on Blakeslea trispora.
27
28 E. Cerda-Olmedo
It seems that the Soviet Union is the only country that uses Blakeslea
trispora for the industrial production of carotenoids; the yearly production is
estimated at about 200 kg purified fJ-carotene and another 4 Mg fJ-carotene in
the form of mycelia used as an additive to animal feed (A. A. Dmitrovskii,
pers. comm., 1988).
The abundant literature on carotenoids has been the subject of many books
and reviews. The monumental and indispensable treatise edited by Isler (1971)
is complemented by newer reviews (Feofilova, 1974; Goodwin, 1976, 1980,
1988). The practical aspects are explained in great detail by Bauernfeind
(1981). Updates are provided by the International Symposia on Carotenoids,
published every 3 years as a special issue of Pure and Applied Chemistry.
There are specialized and recent reviews on the influence of external factors on
carotenogenesis in the Mucorales (Lampila et al., 1985a), on Phycomyces
(Cerda-Olmedo & Lipson, 1987), on the regulation of carotenogenesis
(Bramley & Mackenzie, 1988), on the metabolism and metabolic effects of the
carotenoids (Goodwin, 1986), and many other subjects.
2 FUNGAL CAROTENOIDS
Most fungi produce no carotenoids at all and many of the others contain a
single major carotenoid. Although the pathway intermediates may be present
in amounts widely variable with the strain and the culture conditions, the
carotenoid composition of the fungi is usually much simpler than that of the
photosynthetic organisms. Here are a few selected examples of fungal
carotenoids. fJ-Carotene predominates in the Mucorales Blakeslea trispora,
Phycomyces blakesleeanus and Choanephora cucurbitarum, in the yeast
Rhodotorula aurea, in Aspergillus giganteus (EI-Jack et al., 1988) and various
species of Penicillium; neurosporaxanthin is the end-product of Fusarium
aquaeductuum (Bindl et al., 1970); neurosporaxanthin and fJ-carotene, those
of Gibberella fujikuroi (Avalos & Cerda-Olmedo, 1987), Neurospora crassa
and Verticillium agaricinum (Valadon & Mummery, 1973); torularhodin and
fJ-carotene, those of the yeasts Rhodotorula rubra, R. minuta, and
Rhodosporidium diobovatum; fJ-carotene, lycopene, and fJ-zeacarotene are
found in different strains of the smut Ustilago violacea (Will et al., 1984);
canthaxantin in the mushroom Cantharellus cinnabarinus; astaxanthin, in the
yeast Phaffia rhodozyma. See Goodwin (1980) for other fungi and additional
references.
Figure 1 shows the structure of the carotenoids that have just been
mentioned, with their likely biosynthetic pathways. The carotenoids, like all
terpenoids, are synthesized from hydroxymethylglutaryl-coenzyme A, which is
first converted to mevalonic acid. The specific part of the pathway begins with
the condensation of two molecules of geranylgeranyl pyrophosphate to form
phytoene, a colourless carotene. Four dehydrogenations transform phytoene
into lycopene, the pigment of red tomatoes. Two cyclisations convert lycopene
----.. ------ ~ ~
Ivcopene
~-/
V-carotene torulene
J
5'
;:
COOH
~
f) - carotene " torularhodin a~
~
;:
c
~
;
~
~
~.
.DH
"
Fig. 1. Some fungal carotenoids with their possible biosynthetic pathways. The horizontal arrows indicate dehydrogena-
tions, the vertical arrows, cyclisations, and the tilted arrows, oxidative reactions.
~
30 E. Cerda-Olmedo
3.1 Light
Blue illumination stimulates the production of carotenoids in many fungi, as it
does in various bacteria, algae, and plants (reviews, Harding & Shropshire,
1980; Rau, 1983; Rau & Schrott, 1987). This phenomenon (photo-
carotenogenesis) does not occur in many fungi, for example, in Blakeslea
trispora (Sutter, 1970).
Photocarotenogenesis presumably protects the cells against the deletereous
effects of very strong illuminations and avoids waste in the dark or in dim light,
but is of limited interest to biotechnologists, because bright illumination is not
practical in large cultures and the final concentrations of carotenoids are not
very high. The examples given in Table 1 prove this point, but should not be
compared with each other, because they were obtained under very different
conditions, most under non-saturating light exposures.
In many species of the Mucorales the contact zone of mycelia of opposite sex
becomes bright yellow (Blakeslee, 1904). The sexual stimulation is easily
observed in mated cultures (mixed cultures of strains of opposite sex) (Barnett
et al., 1956; Anderson et al., 1958). The beta-carotene content increases 5-15
times over the level found in separate cultures, according to various reports.
Sexual stimulation in the Mucorales occurs in the absence of physical contact
when the interacting hyphae are separated by a membrane (Burgeff, 1924). An
enhanced carotenogenesis occurs in strains of Blakeslea trispora grown under
such conditions (Barnett et al., 1956). A race between chemists of several
major biotechnological companies led to the isolation of chemicals, the
trisporic acids, present in mated cultures and able to stimulate carotenogenesis
in single cultures (Caglioti et al., 1966). The trisporic acids are industrially
useless because the increases in p-carotene content do not compensate their
cost.
Table 1
Examples of Photocarotenogenesis in some Fungi
a The numerical values are the total concentrations of the carotenoids detected in
illuminated cultures, in Ilg/g dry mycelial weight.
32 E. Cerd4-01medo
Since trisporic acids are made from p-carotene (Austin et a/., 1970), the
activation of carotenogenesis by trisporic acids in the Mucorales provides a
regulatory loop for increased sexual stimulation. Another likely role of the
newly synthesized p-carotene is to make sporopollenins, tough polymers found
in the cell walls of the zygospores (Gooday et a/., 1973) and of other
structures. Lack of trisporic acids and sporopollenins explains the sexual
impotence of mutant Mucorales devoid of p-carotene (Sutter, 1975; Khabrova
& Zhdanov, 1979), but not that of the carS mutants, which overproduce it. The
trisporic acids have no effect on carotenogenesis or morphogenesis in fungi
other than the Mucorales.
4 CHEMICAL STIMULATION
5 GENETICAL STIMULATION
5.1 Phycomyces
phytoene
~
p8
p8
.. \
pB carRA carB
pB I
pR
pR .. ~
f)-carotene
~ \
TRANSCRIPTION
I
I
,
\
\ I
I
~" ~ ~ ///
/
y
pS pA
I carD
---C±l-PD~
--8-PC~
9=1===F
care
carS
Fig. 2. The genetics of carotenogenesis in Phycomyces blakesleeanus (Cerda-Olmedo,
1987). Four copies of phytoene dehydrogenase (the product of gene carB) and two
copies of lycopene cyclase (one of the products of gene carRA, closely linked to carB)
convert phytoene into beta-carotene (above). The product of gene carS and a product
of gene carRA combine with beta-carotene to repress the synthesis (center). Two other
genes, care and carD, regulate in opposite ways the action of carS (below).
in gene carS are epistatic over those in gene carD: the double mutants do not
surpass single carS mutants and are, therefore, of no practical value. There are
superproducing strains that considerably exceed the J3-carotene content of the
carS mutants, but they have not yet been published.
The highest published J3-carotene level in Phycomyces blakesleeanus
corresponds to intersexual heterokaryons with carS mutations and balanced
lethals (Murillo et aI., 1978). They contain 25 mg J3-carotene per g dry weight,
some 500 times more than the wild type. Their genetic make-up provides
constitutive sexual activation and removal of end-product inhibition, and
therefore, they do not respond to retinol or J3-ionone. They do not respond to
light or dimethyl phthalate either, although no provision is made for
endogenous activation of the mechanisms normally triggered by these agents.
36 E. Cerda-Olmedo
When certain Phycomyces strains are grown in the dark, most of the
fJ-carotene appears in a red complex, similar to the main carotenoprotein of
many plants, which stabilizes fJ-carotene and changes its colour in an attractive
way. These strains contain a carS mutation and a cytoplasmic carE mutation
(De la Concha & Murillo, 1984). They accumulate about 100 times more
fJ-carotene than the wild-type and no attempt has been made to improve
their yield.
Some genetic constructs contain considerable amounts of pathway inter-
mediates. Lycopene is accumulated by carR mutants, defective in lycopene
cyclase. Constitutive sexual activation and partial removal of end-product
inhibition led to strains with about 15 mg lycopene per g dry weight (Murillo et
al., 1978). A high phytoene content (9 mg/g dry weight) was obtained by
growing a carB mutant, defective in phytoene dehydrogenase, with dimethyl
phthalate in the light (Bejarano & Cerda-Olmedo, 1989). Other intermedi-
ates can only be produced as part of carotene mixtures: phytofluene and
~-carotene in some leaky carB mutants (Eslava & Cerda-Olmedo, 1974;
Bejarano et al., 1987); y-carotene in certain heterokaryons (De la Guardia et
al., 1971).
The genetic analysis of carotenogenesis is feasible in many fungi, but has been
carried out in few, and only to limited extents (see, for example, Khabrova &
Zhdanov, 1979, for Blakeslea trispora; Harding & Turner, 1981, for
Neurospora crassa; Will et al., 1984, for Ustilago violacea; Avalos &
Cerda-Olmedo, 1987, for Gibberella fujikuroi). Certain mutants of Gibberella
are deep orange even when grown in the dark and contain up to 1.7 mg
neurosporaxanthin per g dry weight, but they are not analogous to Phycomyces
mutants because there is no end-product regulation in Gibberella.
6 INDUSTRIAL PRODUCTION
6.1 Blakeslea
The only fungal production of carotenoids that has been developed to the
industrial stage is that of fJ-carotene with mated cultures of Blakeslea trispora.
After detailed studies of media and additives, Ciegler and his co-workers at the
US Department of Agriculture scaled up their cultures to a volume of 11 litres
(Ciegler et al., 1963a). The process has been reviewed in detail (Ciegler, 1965;
Hanson, 1967). The production phase (Ciegler et al., 1963a) consists in the
joint culture at 28°C for 3 days of mycelia of two wild types of opposite sex in
liquid medium (20 g ground maize, 40 g cotton-seed meal, 30 g white grease or
vegetable oil, 30 g deodorized kerosene, 50 g citrus molasses, and 0·2 mg
thiamine·HCI per litre). The fermented product contains about 17 mg fJ-
Production of Carotenoids with Fungi 37
6.2 Phycomyces
Production has been studied only under laboratory conditions. The genetic
work was carried out on minimal agar; the wild type contains about 0·040 mg
p-carotene per g dry weight and the superproducing strains described above,
about 25 mg (Murillo et al., 1978). The mycelial dry weight is 10 g/litre in both
cases. We have observed that various simple media, some composed of cheap
subproducts of biological industries, increase the biomass to about 40g dry
mycelia per litre and maintain its p-carotene concentration. Carotenogenesis
suffers if the cultures are agitated, whether they are wild type (Lilly et al.,
1960) or superproducers, and therefore surface cultures are the most
productive.
6.3 Application
7 THE PROSPECTS
REFERENCES
Engel, B. G., Wiirsch, J. & Zimmermann, M. (1953). fIber den Einfluss von
beta-Jonon auf die Bildung von beta-Carotin durch Phycomyces blakesleeanus.
Helv. Chim. Acta, 36, 1771-6.
Eslava, A. P. (1987). Genetics. In Phycomyces, ed. E. Cerda-Olmedo & E. D. Lipson.
Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, pp. 27-48.
Eslava, A. P. & Cerda-Olmedo, E. (1974). Genetic control of phytoene dehydrogena-
tion in Phycomyces. Plant Sci. Lett., 2,9-14.
Eslava, A. P., Alvarez, M. I. & Cerda-Olmedo, E. (1974). Regulation of carotene
biosynthesis in Phycomyces by vitamin A and beta-ionone. Eur. J. Biochem., 48,
617-23.
Feofilova, E. P. (1974). Pigmenty mikroorganizmov. Nauka, Moscow.
Feofilova, E. P. & Bekhtereva, M. N. (1976). Effect of vitamin A on carotene
biosynthesis by Blakeslea trispora (in Russian). Mikrobiologia, 45, 557-8.
Feofilova, E. P., Ushanova, A. E. & Ivanova, G. B. (1982). Effect of dimilin on
carotene production by Blakeslea trispora (in Russian). Mikrobiologia, 51,
267-71.
Garber, E. D., Baird, M. L. & Weiss, L. M. (1978). Genetics of Ustilago violacea. II.
Polymorphism of color and nutritional requirements of sporidia from natural
populations. Bot. Gaz., 139,261-5.
Gogek, C. J. (1968). Stabilization of crude carotene-containing mycelium. J. Agric.
Food Chem., 16, 730-34.
Gooday, G. W., Fawcett, P., Green, D. & Shaw, G. (1973). The formation of fungal
sporopollenin in the zygospore wall of Mucor mucedo: a role for the sexual
carotenogenesis in the Mucorales. J. Gen. Microbiol., 74, 233-9.
Goodwin, T. W. (Ed.) (1976). Chemistry and Biochemistry of Plant Pigments, 2nd edn,
two vols. Academic Press, London.
Goodwin, T. W. (1980). The Biochemistry of the Carotenoids. 2nd edn., Vol. 1, Plants.
Chapman and Hall, London/Academic Press, London.
Goodwin, T. W. (1986). Metabolism, nutrition, and function of carotenoids. Ann. Rev.
Nutr., 6,273-397.
Goodwin, T. W. (Ed.) (1988). Plant Pigments. Academic Press, London.
Goodwin, T. W. & Griffiths, L. A. (1952). Studies in carotenogenesis. 5. Carotene
production by various mutants of Phycomyces blakesleeanus and Phycomyces nitens.
Biochem. J., 52, 499-501.
Govind, N. S. & Cerda-Olmedo, E. (1986). Sexual activation of carotenogenesis in
Phycomyces blakesleeanus. J. Gen. Microbiol., 132,2775-80.
Hanson, A. M. (1967). Production of pigments and vitamins. In Microbial Technology,
ed. H. J. Peppler. Van Nostrand-Rheinhold, Princeton, pp. 222-50.
Harding, R. W. & Shropshire, W. (1980). Photocontrol of carotenoid biosynthesis.
Ann. Rev. Plant Physiol., 31,217-38.
Harding, R. W. & Turner, R. V. (1981). Photoregulation of the carotenoid biosynthe-
tic pathway in albino and white collar mutants of Neurospora crassa. Plant Physiol.,
68,745-9.
Harding, R. W., Huang, P. C. & Mitchell, H. K. (1969). Photochemical studies of the
carotenoid biosynthetic pathway in Neurospora crassa. Archs Biochem. Biophys.,
129,696-707.
Hsu, W. J., Yokoyama, H. & Coggins, C. W. (1972). Carotenoid biosynthesis in
Blakeslea trispora. Phytochem., 11,2985-90.
Isler, O. (Ed.) (1971). Carotenoids. Birkhauser Verlag, Basel.
Johnson, E. A. & Lewis, M. J. (1979). Astaxanthin formation by the yeast Phaffia
rhodozyma. J. Gen. Microbiol., 115,173-83.
Johnson, E. A., Conklin, D. E. & Lewis, M. J. (1977). The yeast Phaffia rhodozyma as
a dietary pigment source for salmonids and crustaceans. J. Fish. Res. Board Can.,
34,2417-21.
Production of Carotenoids with Fungi 41
Johnson, E. A., Villa, T. G., Lewis, M. J. & Phaff, H. J. (1978). Simple method for
isolation of astaxanthin from the basidiomycetous yeast Phaffia rhodozyma. Appl.
Environ. Microbiol., 35, 1155-9.
Khabrova, A. M. & Zhdanov, V. G. (1979). Study of sex defective mutants of
Blakeslea trispora. Mikrobiologia, 48, 1055-9.
Krzeminiski, L. F. & Quackenbush, F. W. (1960). Stimulation of carotene synthesis in
submerged cultures of Neurospora crassa by surface-active agents and ammonium
nitrate. Archs Biochem. Biophys., 88,64-7.
Lampila, L. E., Wallen, S. E. & Bullerman, L. B. (1985a). A review of factors
affecting biosynthesis of carotenoids by the order Mucorales. Mycopathologia, 90,
65-80.
Lampila, L. E., Wallen, S. E., Bullerman, L. B. & Lowry, S. R. (1985b). The effect of
strain and type of whey on the production of beta-carotene and other parameters.
Lebensmittel-Wissenschaft-Technologie, 18, 366-9.
Lampila, L. E., Wallen, S. E., Bullerman, L. B. & Lowry, S. R. (1985c). The effect of
illumination on growth and beta-carotene content of Blakeslea trispora grown in
whey. Lebensmittel-Wissenschaft-Technologie, 18,370-73.
Lilly, V. G., Barnett, H. L. & Krause, R. F. (1960). The production of carotene by
Phycomyces blakesleeanus. West Virginia Univ. Agr. Exper. Sta., Bulletin 441T,
Morgantown.
Mackinney, G., Nakayama, T., Buss, C. D. & Chichester, C. O. (1952). Carotenoid
production in Phycomyces. 1. Am. Chem. Soc., 74,3456-7.
Mackinney, G., Nakayama, T., Chichester, C. O. & Buss, C. D. (1953). Biosynthesis
of carotene in Phycomyces. 1. Am. Chem. Soc., 75, 236-8.
Murillo, F. J. & Cerda-Olmedo, E. (1976). Regulation of carotene synthesis in
Phycomyces. Molec. Gen. Genet., 148, 19-24.
Murillo, F. J., Calder6n, I. L., L6pez-Diaz, I. & Cerda-Olmedo, E. (1978).
Carotene-superproducing strains of Phycomyces. Appl. Environ. M icrobiol., 36,
639-42.
Murillo, F. J., Torres-Martinez, S., Arag6n, C. M. G. & Cerda-Olmedo, E. (1981).
Substrate transfer in carotene biosynthesis in Phycomyces. Eur. 1. Biochem., 119,
511-16.
Ninet, L. & Renaut, J. (1979). Carotenoids. In Microbial Technology, 2nd edn,
Vol. 1, ed. H. J. Peppler & D. Perlman. Academic Press, New York, pp.
529-44.
Ninet, L., Renaut, J. & Tissier, R. (1969). Activation of the biosynthesis of carotenoids
by Blakeslea trispora. Biotechnol. Bioeng., 11, 1195-210.
Rau, W. (1983). Photoregulation of carotenoid biosynthesis. In Biosynthesis of
isoprenoid compounds, ed. J. W. Porter & S. L. Spurgeon. John Wiley, New York,
pp. 123-57.
Rau, W. & Schrott, E. L. (1987). Blue light control of pigment biosynthesis-
Carotenoid biosynthesis. In Blue Light Responses, Vol. I, ed. H. Senger. CRC
Press, Boca Rat6n, Florida, pp. 43-63.
Reyes, P., Chichester, C. O. & Nakayama, T. O. M. (1964). The mechanism of
beta-ionone stimulation of carotenoid and ergosterol biosynthesis in Phycomyces
blakesleeanus. Biochim. Biophys. Acta, 90, 578-92.
Roncero, M. I. G. & Cerda-Olmedo, E. (1982). Genetics of carotene biosynthesis in
Phycomyces. Curro Genet., 5, 5-8.
Suarez, T. & Eslava, A. P. (1988). Transformation of Phycomyces with a bacterial gene
for kanamycin resistance. Molec. Gen. Genet., 212, 120-23.
Sutter, R. P. (1970). Effect of light on beta-carotene accumulation in Blakeslea trispora.
1. Gen. Microbiol., 64, 215-21.
Sutter, R. P. (1975). Mutations affecting sexual development in Phycomyces blakes-
leeanus. Proc. natn. Acad. Sci. USA, 72, 127-30.
42 E. Cerda-Olmedo
I GENERAL ASPECTS
1 HISTORY
1934). Pioneering work on photosynthetic bacteria was carried out by Van Niel
& Smith (1935).
20
r~
2
20
,.'
.' 3'
~ 2'
B 3
20'
~. 2' 4
f
1 ""
•• OH
2
HO
3
-",
HO
4
0 0
5
0
OH
6
HO
9
OCH3
10
11 ...,
0
Structure Trivial name Basic structure Semi-systematic name Systematic name (IUPAC)
(Fig. 1)
3 BIOLOGICAL PROPERTIES
4 INDUSTRIAL APPLICATIONS
5 PRODUCING MICRO-ORGANISMS
7 REGULATION OF BIOSYNTHESIS
Table 2
Micro-organisms of Potential Practical Interest for the Production of
Carotenoids
Species Carotenoid(s) produced
1. Micro-organisms specifically designed as pigment sources
1. Sources of carotenes
A. Fungi
Blakeslea trispora Lycopene
B. Non-photosynthetic bacteria
Streptomyces chrestomyceticus, Lycopene
subsp. rubescens
2. Sources of xanthophyllsa
A. Green algae
Spongiococcum excentricum Lutein
Chlorella pyrenoidosa Lutein
B. Fungi
Dacrymyces deliquescens Lutein
C. Non-photosynthetic bacteria
Flavobacterium sp. Zeaxanthin
Streptomyces chrestomyceticus, Unidentified xanthophylls
var. aurantioideus
Mycobacterium phlei Unidentified xanthophylls
3. Sources of monocyclic ketocarotenoids
A. Non-photosynthetic bacteria
Deinococcus radiophilus-radiodurans- Derivatives of 4-keto-
radiopugnans y-carotene
Mycobacterium smegmatis Derivatives of 4-keto-
y-carotene
4. Sources of bicyclic ketocarotenoids
A. Cyanobacteria
Anabaena variabilis Canthaxanthin
Aphanizomenon flos-aquae Canthaxanthin
Nostoc commune Canthaxanthin
B. Green algae (N-deficiency)
Dictyococcus cinnabarinus Canthaxanthin
Haematococcus pluvialis Astaxanthin
C. Fungi/yeast
PhajJia rhodozyma Astaxanthin
D. Non-photosynthetic bacteria
Brevibacterium KY -4313 Canthaxanthin
Rhodococcus maris / Mycobacterium Canthaxanthin
brevicale 32-MCT
Mycobacterium lacticola Astaxanthin
Brevibacterium 103 Astaxanthin
5. Sources of retrocarotenoids
Pseudomonas extorquens Rhodoxanthin
II. Micro-organisms designed for applications other than
pigment production
A. Fungi/yeast
Rhodotorula, Rhodosporidium sp. Torulene,
(red yeasts) torularhodin
B. Non-photosynthetic bacteria
Methylotrophs Miscellaneous
C. Photosynthetic bacteria
Rhodobacter capsulatus Spheroidene,
Spheroidenone
a Ketocarotenoids excluded.
50 H. J. Nelis & A. P. De Leenheer
OCH,
~
1 IS'
~ JA
lc .--e-~
3
Both the pigment patterns and the amount of carotenoids in the cells may be
altered by changes in the composition of the growth medium and by the
influence of physical factors such as light, temperature and oxygen. Variation
of the growth conditions of an organism with the aim to optimize its carote-
noid content sometimes has a rational basis (e.g. generation of biosynthetic
intermediates), but very often is rather empirical. From the numerous effects
reported in the literature (for a compilation, see Goodwin, 1980) only a few
Microbial Production of Carotenoids 51
7.1.1 Algae
Z 1. 4 Photosynthetic bacteria
8 STRAIN IMPROVEMENT
Various bacterial mutants have been prepared that have a defect in the
structural gene for one of the carotenogenic enzymes (Spurgeon & Porter,
1983). Three types of mutants with altered carotenoid biosynthesis are of
potential practical interest.
9 FERMENTATION OF CAROTENOID-PRODUCING
MICRO-ORGANISMS-GENERAL
II FERMENTATION-MICRO-ORGANISMS SPECIFICALLY
DESIGNED FOR PIGMENT PRODUCTION
1 CAROTENES: LYCOPENE
1.1 General
presence in poultry feeds enhances the color of the egg yolk (Marusich &
Bauernfeind, 1981).
1.2 Fungi
As discussed in this book (Chapter 3), the fungus Blakeslea trispora (Muco-
rales) synthesizes massive amounts of p-carotene when the two sexual forms of
this species are mixed in the same culture. However, there are several
approaches to block the p-carotene biosynthesis at the lycopene level.
1. 2. 3 Mutants
Mutants could be prepared which lack the ability to convert lycopene to
p-carotene. However, Blakeslea trispora lends itself reluctantly to mutation
(Ninet & Renaut, 1979).
2 XANTHOPHYLLS
2.1 General
Among the xanthophylls, the hydroxylated derivatives lutein and zeaxanthin in
particular are of high interest as key components in natural pigmenters for
56 H. J. Nelis & A. P. De Leenheer
In the early 1960s, alga meals were extensively studied for broiler and yolk
pigmentation, as alternatives to other xanthophyll sources like com and alfalfa
meal. So far the green alga Spongiococcum excentricum represents the only
example of a micro-organism industrially exploited for the production of
natural xanthophyll (lutein). For some time, the dried fermentation meal of
this species, containing 1·6-2·4 g of xanthophylls per kg dry mass was
marketed in the US under the name of A-Zanth (Grain Processing Co.,
Muscatine, Iowa) (Anonymous, 1962). The alga was grown under hetero-
trophic conditions, at 28°C, in a medium containing glucose, com steep liquor,
urea and mineral salts (Ciegler, 1965; Hanson, 1967). Incremental feeding of
nutrients was required because Spongiococcum does not tolerate high sugar
concentrations. Another relative disadvantage is that this alga grows slowly in
liquid medium so that ample time must be allowed for inoculum development
(Hanson, 1967). Lutein was produced at a rate of 294 mg/liter in 9 days.
Of all other green algae tested Chlorella sorokiniana (pyrenoidosa, 7-11-05)
in particular showed great promise for further development. Cell yield and
xanthophyll content of this species, grown under heterotrophic conditions in
30-liter fermenters, were optimized by Theriault (Ciegler, 1965; Theriault,
1965; Hanson, 1967). Dry cell weights in excess of 100 g/liter and a total
xanthophyll content of 467-512 mg/liter were obtained from 230 to 260 g
glucose/liter, in 168 h illuminated batch-type fermentations (25°C). Apart from
glucose, the medium constituents were urea, phosphate, MgS0 4 ·7H2 0 and
trace elements. The yield was further increased in continuous feed runs to
302 g cells/liter and 650 mg xanthophyll/liter, obtained from 520 g/liter of
glucose. Again, maximum concentrations of xanthophylls resulted from using
low initial levels of nutrients, followed by incremental feeding of additional
medium ingredients.
In a patent assigned to F. Hoffmann-La Roche, another, thermophilic strain
of Chlorella pyrenoidosa (A 119 Lagos, ATCC 14860) grown at 35°C was used.
The culture medium consisted of cerelose, crude invert molasses, ethanol,
potassium nitrate, amino acids, guanidine nitrate, MgS04 ·7H2 0, phosphates,
acetate, EDTA salts and thiamine (Wendall et ai., 1964). Although the total
carotenoid content amounted to 7·4 mgt g dry weight, the cell yield (28 g/liter
in 4 days) was lower than in the process of Theriault.
Microbial Production of Carotenoids 57
2.3 Fungi
Jamikorn, 1956; Hochmannova et al., 1968). More recent patents describe the
production of 0·3 g 'oxycarotenoids' /liter in a medium containing beet
molasses and urea (Anonymous, 1966-1967, 1969-1970b; Ninet & Renaut,
1979).
3 MONOCYCLIC KETOCAROTENOIDS
3.1 General
Monocyclic, mono-ketocarotenoids have not been systematically studied so far
for application as pigmenters in animal feeds. One of the reasons may be that
they are not widespread in nature. However, the color (Amax 470-480 nm) and
chemical properties (keto group in conjugation with the polyene chain) of
these compounds resemble those of their bicyclic counterparts, e.g. canthaxan-
thin. Hence they could theoretically become useful if a sufficient supply could
be secured.
3.2 Non-photosynthetic Bacteria
Keto-derivatives of y-carotene and dehydro-y-carotene have been demon-
strated in radio-resistant pink tetracocci, e.g. Deinococcus radiophilus and
radiodurans (formerly designated as Micrococcus) (Bamji & Krinsky, 1966;
Lewis & Kumta, 1973). Our own (unpublished) investigation of Deinococcus
radiopugnans ATCC 19172 (formerly Micrococcus roseus) also suggested the
presence of monocyclic ketocarotenoids in this related species. A systematic
study has been devoted to optimizing the growth during fermentation of
tetracocci (Shapiro et al., 1977). The iron content of the medium proved to be
a crucial factor in this regard. Various compounds increasing the availability of
this element, e.g. hydroxamic acids and hemin, acted as growth promoters.
Although no quantitative data on the pigments are available, the mere
presence of these unique ketocarotenoids and the exotic nature of the bacteria
appear to warrant further study.
Similar monocyclic ketocarotenoids have been reported in a Mycobacterium
smegmatis strain, grown on hydrocarbons (Tanaka et al., 1968a,b». The effect
of various nutrients and added compounds (amino acids, detergents) on
growth and pigmentation was examined. The optimal medium contained 2% of
an alkane (preferably n-hexadecane), a mixture of ammonium sulfate and
urea, or, alternatively, ammonium phosphate, as a nitrogen source, a number
of mineral salts and trace elements as well as histidine. However, the
maximum carotenoid yield was disappointingly low (0·8 mg/liter). A non-
specific part of this consisted of 4-keto-y-carotene and its mono-hydroxy- and
mono-methoxy derivatives.
4 BICYCLIC KETOCAROTENOIDS
4.1 General
Canthaxanthin and astaxanthin are widely used as pigmenters in poultry and
fish feeds (see above). Although both compounds are manufactured syntheti-
Microbial Production of Carotenoids 59
cally, their high market value makes any suitable natural source potentially
attractive. Canthaxanthin is found in substantial quantities in Crustacea
(Castillo et al., 1982) and the feathers of birds (Brush, 1981), but not in higher
plants. The fungus Cantharellus cinnabarinus, the first discovered natural
source of canthaxanthin (Haxo, 1950) can be dismissed from a practical point
of view. Astaxanthin occurs in the feathers of birds (Brush, 1981) and flower
petals of Adonis annua (Seybold & Goodwin, 1959) and is abundant in
Crustacea (Castillo et aI., 1982) and fish (Simpson et al., 1981). Apart from
crustacean products (shrimp meal, crayfish waste), no producing organism has
been used commercially (Marusich & Bauernfeind, 1981; Simpson et al.,
1981).
Although strictly speaking the retrocarotenoid rho do xanthin also belongs to
the ketocarotenoids, this compound is discussed in a separate paragraph (see
Section 5).
4.2 Cyanobacteria
8141lg/g dry weight) in the presence of tomato waste. This can be rationalized
in terms of the uptake by the yeast cells of carotenoid precursors. Typical
astaxanthin yields are of the order of 2 mg/liter, obtained in a 6O-h
fermentation.
effects were noted from combining the optimal carbon source, octadecane or
(cheaper) paraffin mixtures enriched (50-90%) in this compound, with certain
branched alkanes (pristane) or hydrocarbon mixtures rich in cycloalkanes.
Ammonium phosphate, ammonium acetate and sodium nitrate were nearly
equally effective as nitrogen sources. Yeast extract was slightly superior to malt
extract as a growth factor. The incorporation in the medium of chemicals
supposedly acting as precursors of carotenoid biosynthetic intermediates had
little or no effect. Examples of such compounds include acetate, pyruvate,
farnesol, geraniol, leucine and various sources of acetyl CoA. In addition,
Brevibacterium KY-4313 proved refractory to the action of a variety of
chemicals, which are reportedly carotenogenic in other micro-organisms,
particularly Blakeslea trispora: nitrogenous heterocyclic compounds, vegetable
oils, /3-ionone, retinol, aromatics. However, in non-hydrocarbon medium
(Brain Heart Infusion Broth) a number of substances were found to promote
substantially carotenoid biosynthesis, particularly alcohols (propanol and
isopropanol), glycerol and retinol. The latter caused accumulation of /3-
carotene rather than canthaxanthin. Vegetable oils greatly stimulated growth
but suppressed carotenoid formation.
2. Preparation of mutants: Following treatment of Brevibacterium KY-4313
with the mutagen NMU (Voznyakovskaya & Daraseliya, 1972), we isolated a
colony that distinguished itself from the bulk of the orange ones by its reddish
hue. This 'new', presumably mutant strain, contained about twice as much
canthaxanthin as the parent strain, but also more echinenone. Retreatment of
the 'new' strain with the same mutagen again permitted the isolation of a
slightly more pigmented colony. The latter was subsequently used for all
optimization work.
5 RETROCAROTENOIDS: RHODOXANTHIN
5.1 General
By virtue of its deep red color, rhodoxanthin, which has been approved as a
food additive in certain countries, might be envisaged as an alternative to
astaxanthin. The compound occurs in autumn leaves (Ida, 1981), fish (Kat-
suyama & Matsuno, 1979) and the feathers of birds (Brush, 1981).
1 INTRODUCTION
Section 11,5). Attempts have been made to optimize the carotenoid yield
from selected methylotrophs by altering the culturing conditions. A
Corynebacterium sp. XG was found to form most pigment on methanol,
dimethyl amine and methylamine as carbon sources (Dumenil et al., 1983).
Other medium ingredients included ammonium sulfate, minerals and yeast
extract. Light, pH variation and the nature of the growth factor did not
appreciably affect pigmentation. Protaminobacter ruber, another facultative
methylotroph, produced 2 mg/liter of an unidentified pink carotenoid in a
medium containing 1,2-propanediol as a carbon substrate and methionine and
riboflavine as moderately effective carotenogenic agents (Shimizu et al., 1982).
Both examples demonstrate the interest in methylotrophic bacteria for their
carotenoid composition. Although their primary biotechnological potential
clearly lies in the production of vitamin B12 and biomass, the presence of
useful pigments again adds extra potential value to these organisms.
4 PHOTOSYNTHETIC BACTERIA
4.1 General
toxicity (provided the sodium molybdate had been eliminated), the uncertainty
about the exact pigment composition precludes, for the time being, the use of
Rhodobacter capsulatus as a pigmenter for chickens.
2 ASSAY METHODS
3 CHEMICAL SYNTHESIS--BIOCATALYSIS
1 2
Fig. 4. Reduction reactions involved in the preparation of an optically active key
intermediate in the synthesis of (3R,3/R)-zeaxanthin. A, enantioselective hydrogena-
tion of double bond, catalyzed by baker's yeast; B, chemical reduction; C, reduction of
keto group, catalyzed by baker's yeast (unwanted side reaction). 1, desired reaction
product (4R, 6R); 2, unwanted reaction product (4S, 6R).
72 H. J. Nelis & A. P. De Leenheer
HO~ HO~
OH 0
REFERENCES
Andrewes, A. G. & Starr, M. P. (1976). (3R, 3'R)-Astaxanthin from the yeast Phaffia
rhodozyma. Phytochem., 15,1009-11.
Andrewes, A. G., Phaff, H. J. & Starr, M. P. (1976). Carotenoids of Phaffia
rhodozyma, a red-pigmented fermenting yeast. Phytochem., 15, 1003-7.
Anonymous (1962). Algae-derived xanthophylls for pigmentation control. Feeds
Illustrated, November issue, 24-25.
Anonymous (1966-1967). Oxycarotenoid preparations. Microbiology Abstracts, Section
A, 2, A1056.
Anonymous (1969-1970a). Microbiological production of carotenoid-containing addi-
tives for animal feeds. Microbiology Abstracts, Section A, 5, A200l.
Anonymous (1969-1970b). Production of bacterial preparations with a high content of
oxy-carotenoids. Microbiology Abstracts, Section A, 5, A2292.
Arpin, N., Lebreton, J. & Fiasson, J.-L. (1966). Recherches chimiotaxinomiques sur
les champignons. II. Les caroteno"ides de Peniophora aurantiaca (Bres.)
(Basidiomycete). Bull. Soc. Mycol. France, 82,450-59.
Ascenzi, J. M. & Cooney, J. J. (1975). Action of visible light on enzymes in cell
envelopes of Micrococcus roseus. Photochem. Photobiol., 21, 307-1l.
Bamji, M. S. & Krinsky, N. I. (1966). The carotenoid pigments of a radiation-resistant
Micrococcus species. Biochim. Biophys. Acta, 115, 276-84.
74 H. I. Nelis & A. P. De Leenheer
Johnson, E. A., Lewis, M. J. & Grau, C. R. (1980b). Pigmentation of egg yolks with
astaxanthin from the yeast Phaffia rhodozyma. Poultry Sci., 59, 1777-82.
Katsuyama, T. & Matsuno, T. (1979). Isolation and identification ofrhodoxanthin from
the fish Tilapia nilotica. Bull. lap. Soc. Scient. Fish., 45, 1045.
Kester, A. S. & Thompson, R. E. (1984). Computer-optimized normal-phase high-
performance liquid chromatographic separation of Corynebacterium poinsettiae
carotenoids. I. Chromatogr., 310, 372-8.
Klaui, H. (1982). Industrial and commercial uses of carotenoids. In Carotenoid
Chemistry and Biochemistry, ed. G. Britton & T. W. Goodwin, Pergamon Press,
Oxford, pp. 308-17.
Klaui, H. & Bauernfeind, J. C. (1981). Carotenoids as food colors. In Carotenoids as
Food Colorants and Vitamin A Precursors, ed. J. C. Bauernfeind. Academic Press,
New York, pp. 48-317.
Kobayashi, M. (1972). Utilization of photosynthetic bacteria. In Fermentation Technol-
ogy Today (Proceedings of the Wth International Fermentation Symposium), ed. G.
Terui, pp. 527-31.
Kobayashi, M. & Kurata, S. (1978). The mass culture and cell utilization of
photosynthetic bacteria. Process Biochem., 13,27-30.
Kobayashi, M. & Tchan, Y. T. (1973). Treatment of industrial waste solutions and
production of useful by-products using a photosynthetic bacterial method. Water
Res., 7, 1219-24.
Koronelli, T. V., Kalyuzhnaya, T. V. & Rozynov, B. V. (1981). Pigmented arctic
mycobacteria and their role in the oxidation of hydrocarbons. Mikrobiologiya, 50,
167-70; Microbiology (English translation), SO, 123-5.
Koronelli, T. V., Pahlavuni, I. K. & Rozynov, B. V. (1982). A pigmented amphiphilic
biopolymer from marine mycobacteria. Mikrobiologiya, 51,873-5.
Koronelli, T. V., Pakhlavuni, I. K., Voroybyeva, I. A. & Balashov, S. P. (1987). The
lipopeptidocarotenoid complex and cells of Rhodococcus maris studied by spectros-
copy. Mikrobiologiya, 56,890-91.
Korthals, H. J. & Steenbergen, C. L. M. (1985). Separation and quantification of
pigments from natural phototrophic microbial populations. FEMS Microbiol. Ecol.,
31,177-85.
Koyama, Y., Takii, T., Saiki, K. & Tsukida, K. (1983). Configuration of the carotenoid
in the reaction centers of photosynthetic bacteria (2). Comparison of the resonance
Raman lines of the reaction centers with those of 14 cis-trans isomers of f3-carotene.
Photobiochem. Photobiophys. 3, 139-50.
Krinsky, N. I. (1979). Carotenoid protection against oxidation. Pure Appl. Chem., 51,
649-60.
Lem, N. W. & Glick, B. R. (1985). Biotechnological uses of Cyanobacteria.
Biotechnol. Adv., 3, 195-208.
Leuenberger, H. G. W. (1985). Microbiologically catalyzed reaction steps in the field of
vitamin and carotenoid syntheses. In Biocatalysts in Organic Syntheses, ed. J.
Tramper, H. C. van der Plas & P. Linko, Studies in Organic Chemistry, Vol. 22.
Elsevier, Amsterdam, pp. 99-118.
Levi, J. D., Shennan, J. L. & Ebbon, G. P. (1979). Biomass from liquid n-alkanes. In
Economic Microbiology, Vol. 4, Microbial Biomass, ed. A. H. Rose, Academic
Press, New York, pp. 361-419.
Lewis, N. F. & Kumta, U. S. (1973). Lycopene accumulation in pigmented radio-
resistant Micrococci grown in presence of nicotine. FEBS Lett., 30, 144-6.
Liaaen-Jensen, S. (1978). Chemistry of carotenoid pigments. In The Photosynthetic
Bacteria, ed. R. K. Clayton & W. R. Sistrom. Plenum Press, New York, pp.
233-47.
Lichtfield, J. H. (1979). Production of single-cell protein for use in food and feed. In
Microbial Technology, Vol. I, 2nd edn, ed. H. J. Peppler & D. Perlman. Academic
Press, New York, pp. 93-155.
Microbial Production of Carotenoids 77
Ninet, L. & Renaut, J. (1979). Carotenoids. In Microbial Technology, 2nd edn, Vol. I,
ed. H. J. Peppler & D. Perlman. Academic Press, New York, pp. 529-44.
Ninet, L., Renaut, J. & Tissier, R. (1969). Activation of the biosynthesis of carotenoids
by Blakeslea trispora. Biotechnol. Bioeng., 11, 1195-210.
Okagbue, R. N. & Lewis, M. J. (1981). Mixed culture of Bacillus circulans WL-12 and
Phaffia rhodozyma on different carbon sources: yeast-wall lytic enzyme production
and extractability of astaxanthin. Biotechnol. Lett., 5,731-6.
Okagbue, R. N. & Lewis, M. J. (1984a). Vse of alfalfa residual juice as a substrate for
propagation of the red yeast Phaffia rhodozyma. Appl. Microbiol. Biotechnol., 20,
33-39.
Okagbue, R. N. & Lewis, M. J. (1984b). Inhibition of the red yeast Phaffia rhodozyma
by saponin. Appl. Microbiol. Biotechnol., 20,278-80.
Okagbue, R. N. & Lewis, M. J. (1984c). Autolysis of the red yeast Phaffia rhodozyma:
a potential tool to facilitate extraction of astaxanthin. Biotechnol. Lett., 6,247-50.
Okagbue, R. N. & Lewis, M. J. (1985). Influence of mixed culture conditions on
yeast-wall hydrolytic activity of Bacillus circulans WL-12 and on extractability of
astaxanthin from the yeast Phajfia rhodozyma. l. Appl. Bacteriol., 59,243-55.
Pachlavuni, I. K. (1987). A carotenoid-containing complex from Rhodococcus maris:
some properties and structure. Abstracts of the 8th International Symposium on
Carotenoids, Boston, 27-31 July, 1987, p. 12.
Peterson, W. J., Evans, W. R., Leece, E., Bell, T. A. & Etchells, J. L. (1958).
Quantitative determination of the carotenoids in yeasts of the genus Rhodotorula. l.
Bacteriol., 75, 586-9l.
Ratledge, C. (1970). Microbial conversions of n-alkanes to fatty acids: a new attempt to
obtain economical microbial fats and fatty acids. Chem. Ind., 843-54.
Ratledge, C. (1982). Microbial oils and fats: an assessment of their commercial
potential. In Progress in Industrial Microbiology, Vol. 16, ed. M. J. Bull. Elsevier,
Amsterdam, pp. 119-206.
Rau, W. (1983). Photoregulation of carotenoid biosynthesis. In Biosynthesis of
Isoprenoid Compounds, Vol. 2, ed. J. W. Porter & S. L. Spurgeon. Wiley, New
York, pp. 123-57.
Riiedi, P. (1985). HPLC-a powerful tool in carotenoid research. Pure Appl. Chern.,
57,793-800.
Sakurai, H., Kato, K., Sakai, T., Masuda, Y. & Kuriyama, T. (1971). Studies on
microbial utilization of petroleum. I. Separation and characterization of carotenoids
produced by a species of Brevibacterium in hydrocarbon media. Bull. Chem. Soc.
lapan, 44,481-84.
Saperstein, S., Starr, M. P. & Filfus, J. A. (1954). Alterations in carotenoid synthesis
accompanying mutation in Corynebacterium michiganense. l. Gen. Microbiol., 10,
85-92.
Sato, K., Veda, S. & Shimizu, S. (1977). Form of vitamin B12 and its role in a
methanol-utilizing bacterium, Protaminobacter ruber. Appl. Environ. Microbiol.,
33,515-2l.
Schlegel, H. G. (1959). Conversion of carotenoids to oxycarotenoids by Mycobacterium
phlei. l. Bacteriol., 77, 310-16.
Schmidt, K. (1978). Biosynthesis of carotenoids. In The Photosynthetic Bacteria, ed. R.
K. Clayton & W. R. Sistrom. Plenum Press, New York, pp. 729-50.
Schwartzel, E. H. & Cooney, J. J. (1970). Isolation and identification of echinenone
from Micrococcus roseus. l. Bacteriol., 104,272-74.
Schwartzel, E. H. & Cooney, J. J. (1972). Isolation of 4'-hydroxyechinenone from
Micrococcus roseus. l. Bacteriol., 112, 1422-24.
Schwartzel, E. H. & Cooney, J. J. (1974a). Action of light on Micrococcus roseus. Can.
l. Microbiol., 20, W15-2l.
Schwartzel, E. H. & Coon~y, J. J. (1974b). Isolation and characterization of
pigmentation mutants of Micrococcus roseus. Can. l. Microbiol., 20, 1007-13.
Microbial Production of Carotenoids 79
1 INTRODUCTION
24
HO HO
1 2
where clouded skies prevail throughout most of the year not enough vitamin D
is biosynthesized. The migration into cities and the increase in indoor
professions led to a decline in the synthesis of the natural vitamin and created
a demand for a dietary supplement that would protect the human body from
rickets.
In rickets, the organic matrix of new bone is not mineralized. This is due to
the body's inability to calcify the collagene matrix of the growing bone and
results in large areas of uncalcified bone (osteoid). The resultant lack in
rigidity of bones leads to the ends becoming entwisted and bent. Also, tooth
enamel formation may be affected. Children are specially prone to this disease.
Even with an optimal intake of calcium and phosphorus in vitamin D deficient
children, the concentration in the blood serum of these minerals is very low.
Vitamin D increases the calcium and phosphate absorption and restores the
mineral balance which results in the appropriate calcium phosphate deposition
in the bone.
The use of liver oil as a prophylactic against rickets, as well as the
importance of light as an antirachitic factor, have been recognized since before
CH3
HO HO
7-Dehydrocholesterol Ergosterol
1 1
./"
HO HO
Cholecalciferol (D3) Ergocalciferol (D 2 )
Fig. 1.
The Biotechnology of Ergosterol 83
the end of the 19th century. Steenbock & Black (1924) showed that irradiation
of foods led to the generation of the vitamin. We now know that irradiation
may lead to two types of vitamin 0: (1) By irradiation of ergosterol,
ergocalciferol (or O 2 ) is formed while (2) irradiation of 7-dehydrocholesterol
leads to the production of cholecalciferol (or D 3 ), see Fig. 1. This chapter will
deal chiefly with the production of ergosterol, the provitamin O 2 •
Although yeasts are known as producers of ergosterol, the sterol was first
isolated from rye ergot (sclerotia of C/aviceps purpurea). It is from this plant
disease that the name ergosterol was derived (Tanret, 1889). Many organisms
were later screened and examined for their sterol content. While procaryotes
were found to be practically devoid of true sterols, eucaryotic micro-organisms
84 P. Margalith
could be divided into a number of categories with regard to the role sterols
play in their cell biology:
(1) Organisms in which sterols are an important constituent of the cell
material:
(a) Organisms that synthesize their own sterols.
(b) Organisms unable to synthesize their sterols, their requirements
being met by environmental sources.
(2) Organisms which do not require sterols for vegetative growth, but
need such compounds for the differentiation of certain sexual or
asexual reproductive organs.
(3) Organisms which have no known requirements for sterols but which
may incorporate such compounds if suitably exposed to them (Mar-
galith, 1986).
Screening for organisms that are good producers of ergosterol was obviously
directed to the first category of organisms that synthesize their own sterols.
Although sterols are widespread among photosynthesizing plants including the
fast-growing microalgae, phytosterols of green plants usually do not comprise
ergosterol. The search was therefore limited to eucaryotic heterotrophs,
amongst these the Mucorales (Zygomycotina), the Ascomycotina (including
the yeasts and yeastlike fungi), the Basidiomycotina (in the mushrooms) and
many of the Fungi Imperfecti.
In most cases ergosterol is not the only sterol to be found, it is usually
accompanied by other related sterols such as zymosterol (3) fungisterol (4) or
ergosta-5,7,22,24(28)-tetraenol (5). We shall return to these sterols when
discussing the biosynthesis of ergosterol. The quantitative aspects of ergosterol
formation and distribution have been also studied. There seems to be no great
variation in the sterol content of fungi. According to Appleton et al. (1955),
molds have an average of c. 0·6% sterols on a dry weight basis. Exceptional
higher values were recorded for Penicillium westlinghi (2·2%).
Not surprisingly, yeasts have been the object of many workers interested in
the biotechnology of ergosterol and its biosynthesis. The ease with which
yeasts can be grown and the familiarity of many microbiologists with these
organisms in various food industries, in addition to their being an excellent
source for most vitamin Bs, have made yeasts the organism of choice for the
production of provitamin D.
The first to study the ergosterol content of yeasts were Bills et al. (1930).
They examined 29 strains from 13 species of 5 genera: Endomyces, Nadsonia,
Mycoderma, Saccharomyces, and Zygosaccharomyces. Values of 0·2-0·3%
(dry weight) have been recorded. Saccharomyces carlsbergensis (strain ATCC
2345) had the highest ergosterol level (2·4%). Usdin & Burell (1952) found
2·7% in Rhodoturula gracilis. A systematic study for ergosterol in yeasts was
carried out by Dulaney et al. (1954) who examined 146 cultures of the
following genera: Candida, Diploascus, Hansenula, Nematospora,
The Biotechnology of Ergosterol 85
HO
3
HO
4
HO
5
Saccharomyces, Schwanniomyces, Torulopsis, Cryptococcus, Endomyces,
Kloeckera, Pichia, Saccharomycodes, Sporobolomyces, Debaryomyces,
Endomycopsis, Nadsonia, Rhodotorula, Schizosaccharomyces, Taphrina. With
the exception of Saccharomyces, levels of ergosterol never exceeded 0·4%.
The best organism was Saccharomyces cerevisiae for which 3·9% was reported.
Additional studies were made by Appleton et al. (1955) who reached similar
conclusions. They confirmed the advantage of the genus Saccharomyces for the
formation of ergosterol. There seems to be considerable variation between
species or even strains. However, generally speaking, brewer's yeast had the
highest level of sterols, followed by baker's yeast. A method for the
production of high sterol yeasts (10% dry weight) was patented by Dulaney
(1957).
Not all reports deal with the nature of the sterols found in yeast. In most
cases where this was done the ratio of ergosterol to other sterols was 8: 2.
ACETATE
MEVALONATE
1
1
FARNESYLPYROPHOSPHATE
1
SQUALENE
1
SQUALENE EPOXIDE
1
LANOSTEROL
1
ZYMOSTEROL
I
FECOSTEROL
1
EPISTEROL
1
ERGOSTA-7 ,22,24(28)TRIENOL
1
ERGOST A-5,7 ,22,24(28)TETRAENOL
!
ERGOSTEROL
Fig. 2.
lanosterol (6). Interestingly, this compound is also formed during sterol
synthesis in animal tissues and is not found in green plants which synthesize the
related cycloartenol (7) as the first cyclization product.
HO HO
6 7
88 P. Margalith
HO HO H 9
The pattern described above is not the only possible pathway for the
production of yeast ergosterol. Sterol synthesis, in general, does not follow a
rigid pattern but rather a more flexible framework where many alternative
pathways may take place. (Fryberg et al., 1973). Because of the complexity of
sterol biosynthesis and the diversity in the timing of certain key reactions (in
some strains (A~ B~ C, while in others A~ C~ B), the regulation of sterol
synthesis has not been sufficiently studied. Of the many enzymic reactions
potentially capable of a regulatory function, hydroxy-methyl-glutaryl-CoA
reductase is probably the rate-controlling enzyme of sterol biosynthesis
(Qureshi et al., 1980).
6 FERMENTATION
Very few publications deal with the ergosterol fermentation. In the 1950s a
number of papers were published dealing with the production of ergosterol by
The Biotechnology of Ergosterol 89
various yeasts and fungi. Since yeasts with strong fermentative metabolism
were found to be the best producers of ergosterol, most of the studies
concentrated on the fermentation of S. cerevisiae strains. Using a molasses-
cornsteep medium, cells with an ergosterol content of 7-10% could be
obtained. Cell yields of 30-40 g/liter medium and a sterol content of
3-4 g/liter broth were found in shake flask experiments after 4 days of growth
at 28°C (Dulaney et al., 1954). Good yields were also obtained with S.
fermentati (EI Refai & EI Kady, 1969). Little has been published about
fermentation in larger volumes. Nor has the patent literature revealed
important new information on the industrial fermentation of the provitamin.
Since the early patent of Dulaney (1957) a number of patents were granted to
several companies in various countries. Most of these deal with extraction
procedures for the isolation and crystallization of ergosterol from the biomass
(e.g. Nelson, 1959). Few patents describe a modification of the fermentation
procedure by the incorporation of precursors to the medium. The addition of
isopentenol (0·92%) and Na-polyphosphate (0·3%) to the growth of S. uvarum
UFO 1264 at 26°C for 96 h, yielded 2·1 % ergosterol (dry weight) (Japanese
Patent 8158496, 1981). Organisms described in the patent literature comprise:
S. cerevisiae, S. carlsbergensis, Candida tropicalis, Fusarium spp and
Trichoderma spp. Yields (ergosterol on basis of dry weight or volume of
medium) described in the patent literature are generally far lower than those
reported in scientific publications. A number of patents have been granted for
the growth of yeasts in medium containing hydrocarbons as sole carbon source,
e.g. Candida petrophilum ATCC 20226 (Horiguchi et al., Japanese Patent
7392586, 1973). Also here ergosterol yields were rather low (see Table 1 for
list of patents).
Isolation of ergosterol from biological sources usually involves the extraction
of the fatty component, saponification and re-extraction with an ether. Direct
extraction of yeast is not sufficient, usually a preliminary digestion with hot
alkali is included prior to the extraction.
Provitamin D3
Table 1
Some Patents for the Production of Ergosterol
HO
10
8 METABOLIC TRANSFORMATIONS
Vitamin D formed in the skin or absorbed by ingestion through the gut wall, is
transported by the bloodstream into the liver where it is hydroxylated at C-25
to yield 25-hydroxycalciferol which is the main circulating form of Vitamin D.
It is transported on a specific a-globulin to the kidney where further
hydroxylation takes place at C-1 or C-24, apparently in response to the calcium
levels in the blood. Low phosphate levels also stimulate 1,25-
dihydroxycholecalciferol (11) production which, in turn, enhances calcium
absorption. The C-1 hydroxylation is the critical step which enables the vitamin
to assume its hormonal activity in regulating the calcium transport as well as
bone calcium mobilization. This important reaction is carried out in kidney
mitochondria by a mixed function mono-oxygenase involving cytochrom P-450.
OH
/;k
HO
11
Regulation of vitamin D metabolism appears to be controlled by the levels
of calcium, phosphate and the parathyroid hormone in blood serum. Below
9·5 mg/ml of Ca2 +, 25-dihydroxy vitamin D is formed, above that level
24,25-dihydroxycalciferol is produced with the discontinuation of the la-
hydroxylase activity. The enzyme is also stimulated by the homeostatic
parathyroid hormone which causes phosphate diuresis in the kidney and
enhances 1,25-dihydroxycholecalciferol production. The parathyroid hormone
and 1,25-dihydroxycholecalciferol act at the bone site cooperatively in the
stimulation of calcium mobilization from bone. Other metabolites of vitamin D
92 P. Margalith
are also formed. However, they are less potent and their metabolic significance
is not yet fully understood.
HO
12
The Biotechnology of Ergosterol 93
REFERENCES
1 INTRODUCTION
The fat-soluble vitamin, vitamin E, has received great attention for its
usefulness as an antioxidant in clinical and nutritional applications. The
demand for the vitamin has increased year after year. The supply of
tocopherols has been limited to the chemically synthesized racemate of
a-tocopherol or a mixture of a-, f3(y)- and 6-tocopherols extracted from
vegetable oils.
Tocopherols are indispensable components of the lipid bilayer of biological
membranes; decrease in their content brings about structural and functional
damage to the membranes. It is generally accepted that the stabilizing effect of
tocopherols on the membranes is mainly related to three functions: (1)
reaction with lipid peroxide radicals, (2) quenching of reactive molecular
oxygen, and (3) ordering (i.e. restriction of molecular mobility) of the
membrane lipid bilayer by formation of tocopherol complexes with fatty acids.
The highest physiologically active form among the various tocopherols and
tocotrienols so far identified is D-a-tocopherol.
Since a preliminary survey on the occurrence of tocopherols in micro-
organisms (Green et al., 1959), its presence has been demonstrated especially
in chlorophyll-containing organisms, though not in all of them (Taketomi et
al., 1983). Early, a tocopherol-like antioxidant was found in yeasts (Forbes et
al., 1958) and a small amount of a-tocopherol was detected in cells of baker's
yeast (Diplock et al., 1961). a-Tocopherolquinol and a-tocopherolquinone
were found in the algae, Euglena gracilis, in bacteria and in yeasts (Hughes &
Tove, 1982). Ruggeri et al. (1985) reported enhancement of a-tocopherol
production in temperature-stressed cells of E. gracilis Z grown photo-
heterotrophically. However, a study on industrial production of these com-
pounds has not been performed so far.
Recently, Tani & Tsumura (1989) have initiated a study aiming towards
95
96 Y. Tani
Rl R2
CH J CH J a- Tocopherol
CH J H B-Tocopherol
H CH J y- Tocophero 1
H H 0-Tocophero 1
Rl R2
CH J CH 3 a-Tocotrienol
CH J H B-Tocotrienol
H CH J y- Tocotri eno 1
H H 6-Tocotrienol
3 BIOSYNTHESIS OF VITAMIN E
It has been accepted that tocopherols are synthesized through two routes, the
tocopherol route and tocotrienol route. Homogentisic acid, a precursor of
tocopherols, which was derived from the shikimate pathway (the biosynthetic
pathway of aromatic amino acids), incorporates phytyl pyrophosphate or
geranylgeranyl pyrophosphate. When phytyl pyrophosphate is incorporated,
tocopherols are formed and the following additional methylation using
S-adenosyl-L-methionine as the methyl donor occurs to form various types of
tocopherols (tocopherol route). When geranylgeranyl pyrophosphate is in-
corporated, tocotrienols are formed with a similar methylation. Subsequently,
tocotrienols can be transformed to tocopherols. NADPH is the reducing
coenzyme for the reduction of the side chain (tocotrienol route). In Spinach
chloroplasts, the enzymes involved in the a-tocopherol synthesis were found in
the chloroplast envelope fraction. The prenyltransferase and methyltransferase
activities were found to be tightly bound to the chloroplast envelope
membrane. Thus, it is strongly suggested that the tocopherol biosynthesis is
related to the occurrence of chloroplasts or plastids.
might repress their own biosynthesis and thus increase the biosynthetic flow to
a-tocopherol. Homogentisate is a direct precursor of the tocopherol biosynth-
esis. Among prenyl alcohols, which might be involved in the tocopherol
biosynthesis in the form of pyrophosphate to be incorporated to the phytyl side
chain, isopentenyl alcohol was slightly effective on a-tocopherol production.
When these compounds were added simultaneously to the a-tocopherol
production medium, a combination of 0·1 % homogentisate and 0·01 %
L-tyrosine was the best for a-tocopherol production. Moreover, it was found
that higher productivity can be obtained when this mixture was added to the
culture medium only at 5th day of cultivation rather than at the beginning.
Feeding of ethanol to the medium together with peptone, when glucose was
almost consumed, gave higher production of a-tocopherol than that of
glucose-peptone in spite of the same cell yield.
Supplementation with ethanol and peptone, and homogentisate and L-
tyrosine during cultivation was performed finally. As shown in Fig. 2, the
specific production of a-tocopherol increased markedly when the mixture was
fed at 5th, 7th and 9th days of the cultivation. On the other hand, when only
ethanol and peptone were fed, the amount of a-tocopherol in the culture broth
increased according to increase in cell mass but the intracellular content of
a-tocopherol was almost the same as that of the non-supplemented culture.
(8) (C)
,
100
+ ,
,
a;-
...
....
......
~'"
.....
20 .~
..
50
a... ......
.&:
'"
a. "0
0
u 10 .~
....
0
I
>.
"
0
0 5 10 15 0 S 10 lS 0 5 10 15
Cultivation Time (day)
Table 1
Progress of a-Tocopherol Production by E. gracilis Z
"'co~
~CHO +
Wittig reaction
hydrogenation
hydrolys"
Ho4c
Fig. 3. Synthesis of (2R, 4'R, 8'R)-a-tocopherol by coupling optically active chroman
aldehyde with a C I5 phosphonium salt derived from natural phytol.
Vitamin E Production 101
R
I
I •
(4* :C 1 : (4- (5
Fig. 4. Construction of an optically active C 14 side chain from S-fJ-hydroxyisobutyric
acid.
j
~COOH
HO I
~COOR
CD/
HO~OH
I B' I B
A'~ -- A~
-
hyctroly."
cyclzatlon
i/ i
c~
Fig. 7. Bioconversion leading to optically active C 5 -synthons.
Vitamin E Production 103
leaoH HO COOH
HO*
I ----
~ OH
I
I
,
I
I
I
I
I
I
t'
H~
¥O~OH
Fig. 8. Synthesis of the optically active chroman moiety from trimethylhydroquinone
and (S)-citramalic acid.
REFERENCES
Shigeoka, S., Onishi, T., Nakano, Y. & Kitaoka, S. (1986). The contents and
subcellular distribution of tocopherols in Euglena gracilis. Agric. BioI. Chern., 50,
1063-5.
Taketomi, H., Soda, K. & Katsui, G. (1983). Results of screening test in tocopherols in
microbial realm. Vitamins (Japan), 57, 133-8.
Tani, Y. & Tsumura, H. (1989). Screening of tocopherol-producing microorganisms
and a-tocopherol production by Euglena gracilis Z. Agric. Bioi. Chern., 53, 305-12.
Chapter 7
& H. YAMADA
S. SHIMIZU
1 INTRODUCTION
The n-6 and n-3 families of PUFAs are distinguished by differing positions of
the double bond closest to the methyl terminal group of the fatty acid chain
105
106 S. Shimizu & H. Yamada
n-7
c:::::::.~ --~ ~~~~ c::::::~~ ~~
16:1 16:2 18:2 18:3 20:3
Fig. 1. Pathways for the biosynthesis of PUFAs of the n-7 (palmitoleic), n-9 (oleic),
n-6 (linoleic) and n-3 (a-linolenic) families. PUFAs include the n-7, n-9, n-6 and n-3
families, which are defined by the position of the double bond closest to the methyl end
of the fatty acid molecule. Desaturation occurs toward the carboxyl end of the molecule
and chain elongation at the carboxyl end, leaving the methyl end unaltered. Usually,
PUFAs of each family are not inconvertible. GLA, r-linolenic acid; DHGLA,
dihomo-r-linolenic acid; ARA, arachidonic acid; EPA, eicosapentaenoic acid; DHA,
docosahexaenoic acid.
(see Fig. 1). Linoleic acid (9,12-cis-octadecadienoic acid, 18: 2 n-6) and
a-linolenic acid (9,12,15-cis-octadecatrienoic acid, 18: 3 n-3) are found abun-
dantly in plants. They are synthesized de novo in higher plants and together
represent nearly all the polyenoic fatty acids present. Linoleic acid comprises
about 50% of the total fatty acids in corn, soybean, sunflower-seed and
cotton-seed oils and over 70% of the fatty acids of safflower oil. a-Linolenic
acid represents over 50% of the total fatty acids of linseed and perilla oils. In
Oenothera biennis (evening primrose) and O. lamarckiana, however, linoleic
acid is further desaturated to form y-linolenic acid (6,9,12-cis-octadecatrienoic
acid, 18: 3 n-6), which comprises about 10% of the total fatty acids in their
seeds. Mosses (Hartmann et al., 1986), algae (Seto et al., 1984) and protozoa
(Erwin & Bloch, 1964), some fungi (Gellerman & Schlenk, 1979; Yamada et
al., 1987, Shimizu et al., 1988a,1989a) and some marine bacteria (Yazawa et
al., 1988) can further add a 2-carbon unit to these C-18 PUFAs to form C-20
PUFAs.
Animals cannot synthesize fatty acids with double bonds at either the n-3 or
n-6 positions, and therefore are dependent upon dietary sources for these fatty
acids. Most animals, however, possess enzyme systems for further desaturation
and chain elongation at the carboxyl end of these fatty acid molecules. The
resulting further desaturated or longer-chain derivatives such as y-linolenic
acid, dihomo-y-linolenic acid (8,1l,14-cis-eicosatrienoic acid, 20: 3 n-6),
arachidonic acid (5,8,1l,14-cis-eicosatetraenoic acid, 20: 4 n-6), eicosa-
pentaenoic acid (5,8,1l,14,17-cis-eicosapentaenoic acid, 20:5 n-3, or EPA)
and docosahexaenoic acid (4,7,10,13,16,19-cis-docosahexaenoic acid, 22: 6
Microbial Production of PUFAs 107
n-6
Linoleic acid Most vegetable oils Essential fatty acid, minor
9,12-cis-18: 2 component of most tissues
y-Linolenic acid (GLA) None Major fatty acid of evening
6,9,12-cis-18: 3 primrose seed; used as an addi-
tive of feed, cosmetics, etc.
Dihomo-y-linolenic acid None Precursor of 1-group of prosta-
8, 11,14-cis-20 : 3 glandins
Arachidonic acid Meat, liver, brain Major component of most mem-
5,8,11 , 14-cis-20 : 4 brane phospholipids; precursor
of 2-group of prostaglandins
Docosapentaenoic acid None Frequently found in tissues
4,7, 10,13, 16-cis-22 : 5 deficient in n-3 PUF As
n-3
a-Linolenic acid Some vegetable oils Minor component of tissues
9,12,15-cis-18: 3 (soy, linseed), leafy
vegetables
Eicosapentaenoic acid Fish, shellfish, algae Precursor of 3-group of prosta-
(EPA) 5,8,11,14,17- glandins; prevent thrombosis;
cis-20:5 used as pharmaceuticals and a
feed additive
Docosahexaenoic acid Fish, shellfish, algae Minor component of membrane
(DHA) 4,7,10,13,16,19- phospholipids in retinal photo-
cis-22: 6 receptors, sperm, etc.
n-9
Oleic acid Animal and vegetable
9-cis-18: 1 fats Major component of many tissues
Eicosatrienoic acid None Accumulated in total essential
5,8,11-cis-20: 3 fatty acid deficiency
n-7
Palmitoleic acid None Prevent apoplexy
7-cis-16: 1
108 S. Shimizu & H. Yamada
3 BIOSYNTHESIS
The biosynthesis of linoleic acid and a-linolenic acid in higher plants, lower
plants and some micro-organisms takes place via common metabolic C-18 fatty
acids, i.e. stearic acid and oleic acid, from acetate fragments produced during
the catabolism of carbohydrates (Fluco, 1974). Linoleic acid and a-linolenic
acid are usually the terminal products of fatty acid synthesis in higher plants.
These two fatty acids are converted to two distinct families of long-chain and
more unsaturated fatty acids through the n-6 and n-3 routes, respectively, as
shown in Fig. 1. In these routes, the successive chain elongation and
desaturation take place toward the carboxyl end of the fatty acid molecule.
Thus, the members of the n-6 family have six carbons after the last double
bond in the molecule, and the members of the n-3 family have three carbons in
the terminal structure. The first step of the biosynthesis of PUFAs is the
desaturation at the ~6-position of linoleic acid or a-linolenic acid to yield
y-linolenic acid or 6,9,12,15-cis-octadecatetraenoic acid (18: 4 n-3),
respectively (~6-desaturation). The ~6-desaturation is followed by the chain
elongation to the respective C-20 PUFAs, i.e. dihomo-y-linolenic acid and
8,1l,14,17-cis-eicosatetraenoic acid (20:4 n-3). The ~5-position of dihomo-y-
linolenic acid in the n-6 route is then desaturated to yield arachidonic acid
(~5-desaturation). Usually, arachidonic acid is the terminal product of the n-6
route. The ~5-desaturation on 20: 4 n-3 fatty acid in the n-3 route results in
the formation of EPA, which is followed by chain elongation and desaturation
to yield DHA. It is usually the end product of the n-3 route.
The n-6 and n-3 routes have been suggested to share the same enzymes or
enzyme systems for the transformations of C-18 PUFAs to C-20 PUFAs
(Brenner, 1974). Therefore, linoleic acid and a-linolenic acid compete with
each other for the same enzymes. In animals, the n-6 family is usually favored
in this competition, forming arachidonic acid. Neither isolation nor detailed
characterization of enzymes responsible for the desaturation and chain
elongation reactions involved in both routes has been performed yet except for
~6 desaturase from rat liver (Okayasu et al., 1981) . Desaturation reactions
require CoA, A TP, Mi+ and NAD(P)H as cofactors, suggesting that
desaturation enzymes are a kind of oxygenase and that only fatty acyl-CoA
derivatives are substrates for the enzymes. Elongation enzymes have also been
suggested to utilize only acyl-CoA derivatives as substrates.
The members of these two families of fatty acids constitute nearly all the
PUFAs in most living organisms. However, small amounts of different families
of fatty acids have been detected. 5,8, ll-cis-eicosatrienoic acid (20: 3 n-9) is
derived from oleic acid (9-cis-octadecenoic acid, 18: 1 n-9) through the n-9
route which involves the following successive reactions: desaturation at the ~6
position of oleic acid to form 6,9-cis-octadecadienoic acid (18: 2 n-9), followed
by its carbon chain elongation to 8,1l-cis-eicosadienoic acid (20:2 n-9) and
desaturation at the ~5-position of the latter to 20: 3 n-9 fatty acid (Fig. 1).
7,10,13-cis-eicosatrienoic acid (20: 3 n-7) is derived from palmitoleic acid
Microbial Production of PUFAs 109
(9-cis-hexadecenoic acid, 16: 1 n-7) through the n-7 route, as shown in Fig. 1.
These data indicate that there are at least four families of PUFAs in
organisms.
----- "I
Ciliates;Des;y-lin Ameba;Desjy-lin I
Green Algae;X;a-lin
Cryptomonads;Des(?);a-lin
Red Algae;Des;a-lin
..-----
Eubacteria;
An;Des;O
Actinomycetes;Des;O
Photosynthetic Bocteria;An;O
Blue-green Algae;Des;a-lin
Yamada et al. (1987, 1988) assayed productivities of C-20 PUFAs in about 300
fungal isolates from natural sources which are capable of growing on an agar
plate containing stearic acid or elaidic acid as the main carbon source for
growth. They also tested more than 600 stock strains of a wide variety of
micro-organisms, i.e. bacteria (23 genera), actinomycetes (4 genera), yeasts
(20 genera), filamentous fungi (45 genera) and basidiomycetes (11 genera).
Bacteria, actinomycetes, yeasts and basidiomycetes, in general, did not
produce detectable amounts of C-20 PUFAs intracellularly. Most C-20 PUFA
producers were found to be filamentous fungi belonging to the orders of
Mucorales and Entomophthorales. Through this screening, they found that 60
strains of Mortierella and 4S isolates from natural sources produce large
amounts of C-20 PUFAs of the n-6 family (i.e. arachidonic acid and
dihomo-')I-linolenic acid) together with C-18 PUFAs of the same family
(mainly ')I-linolenic acid). Most of the PUFA-producing isolates were found to
belong to the genus Mortierella. It should be noted that all of these Mortierella
strains found as C-20 PUFA producers belong to the subgenus Mortierella.
Neither the stock cultures nor isolates belonging to the subgenus Micromucor
showed any detectable accumulation of C-20 PUFAs, although they produced
n-6 C-18 PUFAs such as ')I-linolenic acid. The arachidonic acid contents of
most of these strains accounted for more than 1S% of the total extractable
fatty acids. These values represent more than SO% of the PUFAs and are
particularly high when compared with those in the case of C-18 PUFAs (Table
Table 3
Comparison of Arachidonic Acid Productivities in Mortierella Fungi and their Mycelial Fatty Acid CompositionG
Mycelial Total 20:4 20:4 16:0 18:0 18:1 18:2 18:3 20:3 20:4 20:5 others
mass FA content yield
(mg/ml) (mg/g) (mg/g) (mg/ml)
M. hygrophila IFO 5941 7·0 183 32-8 0·23 24·7 2·9 3704 9·5 5·5 1-6 17·9 0 0·5
M. zychae CBS 652·68 7·7 120 19·6 0·15 23·2 12-6 29·3 8·7 6·0 3·3 16·3 0 0·6
M. elongata CBS 121·71 8·1 61 13·2 0·11 15-4 14·0 30·3 6·7 6·0 3·8 21-7 0 z.t
M. elongata IS-5 AKU 3999 6·9 166 39·2 0·27 14·5 8·0 34·0 7-6 6·0 2·6 23·6 0 3·7
M. parvispora 2S-13 AKU 3994 8·0 201 19·7 0·16 7-8 9·0 56·3 5·3 6·9 1-9 9·8 0 0
M. schmuckeri NRRL 2761 8·0 205 25-4 0·20 19·9 12-4 37-1 704 4-9 4·9 12-4 0 1·0
M. alpina IS-4 AKU 3998 9·5 318 151·7 1-44 17-9 5-9 11·3 9·8 4-1 3-3 47·7 0 0
M. alpina 20-17 AKU 3996 9·4 277 109·7 1-03 15-8 5·3 12·0 18·2 4·8 2·3 39·6 0 0·7
M. alpina CBS 250-53 6·2 124 33-6 0·21 14·5 5·9 27-8 11-4 704 4·0 27-1 0 1-9
M. alpina 1-83 AKU 3995 9·4 300 143·7 1·35 18·6 4-8 12·3 8·9 4-1 3-4 47-9 0 0
M. alpina CBS 219·35 5·5 139 31-1 0·17 11·2 4·0 30·5 14-4 10·9 4-1 22-4 0 1-7
~
~
W
114 S. Shimizu & H. Yamada
3). No detectable amount of free fatty acids was found in lipid fractions
extracted with chloroform-methanol, suggesting that the PUFAs produced are
present as triglyceride and/or phospholipid forms . Through this screening,
they obtained three isolates, which were taxonomically identified to be
Mortierella alpina IS-4 , M. alpina 20-17 and M. elongata IS-5, as potent
producers of arachidonic acid.
Subsequently, they investigated the cultural conditions for arachidonic acid
production with the three Mortierella strains (Yamada et al., 1987,1988). These
three fungi utilize not only glucose but also glycerol, maltose, n-hexadecane
and n-octadecane as carbon source for the arachidonic acid production. As
shown in Table 4, all the strains produced more than 1 g/liter of arachidonic
acid under 5-liter bench scale fermentor conditions. Based on the results of
studies on individual and combined factors affecting arachidonic acid produc-
tion, they selected M . alpina IS-4 as the most promising producer. Using a
2000-liter fermentor, 22·5 g/liter mycelia (dry weight) containing 44·0%, by
weight, of the lipids, in which arachidonic acid comprised 31 ·0% of the total
fatty acids, was produced under the conditions of intermittent feeding of
glucose and 10 days cultivation at 28°C (Fig. 3(a» . They also reported that the
aradchionic acid in the harvested mycelia can be specifically enriched when
the mycelia are allowed to stand for a further few days at room temperature.
The arachidonic acid content of the resultant mycelia reached nearly 70% of
the total fatty acids (Fig. 3(b». Fractionation of the lipids in the mycelia
demonstrated that triglycerides (70·1 %) and phospholipids (23,4%) were their
(s)
(b) 20:3 '8'218~
30 3r-------------A 20:~ \'S'3G '6~
l' ::: ~tlMm
10 I i }
9 § }{
1 ill ,8:1
o 2 , 6 8 to o 50 100
CultlVOtlon time (cloys) FA composition (%)
Fig. 3. Production of arachidonic acid by M. alpina IS-4 under the optimal culture
conditions. Mortierella alpina IS-4 was cultivated in a 2000-liter fermentor containing
1400 liters of a medium containing 2% glucose and 1% yeast extract, pH 6·0 at 28°C.
Glucose was added to the medium as shown in (a). Changes in the mycelial fatty acid
composition during growth are shown in (b). 16 : 0, Palmitic acid ; 18 : 0, stearic acid;
18 :2, linoleic acid ; 18 : 3G, y-linolenic acid ; 20:3, dihomo-y-linolenic acid; 20:4 or
Ara , arachidonic acid.
Microbial Production of PUFAs 115
o~L1
n·he,ana n.tocophero'
Na.,SO. I_L~
,I ~';;:I.~I;;·:~§":"
mil' 1
~entr"ug..
mil ' 2
f"ter evaporator
Fig. 4. Flow sheet for the production of fungal oil of high arachidonic acid content.
major components, and about 50% of the arachidonic acid was present in the
phospholipid fraction.
The lipids containing arachidonic acid can be obtained as an oil from the
mycelia of M. afpina 1S-4 in a good recovery (80-90%) through the following
successive steps: separation of mycelia by filtration , drying, crushing by ball
mill, extraction of the lipids with n-hexane , removal of insoluble materials by
centrifugation, decolorization and deodorization with active charcoal , and
concentration . The resultant purified oil contains myristic acid (0·2%, by
weight), palmitic acid (7·0) , palmitoleic acid (0·1), stearic acid (2·8), oleic acid
(6·6), linoleic acid (5·9), y-linolenic acid (3 ·9), 11-cis-eicosenoic acid (0·9),
11 ,14-cis-eicosadienoic acid (1 ·3), dihomo-y-linolenic acid (3·9) and arachido-
nic acid (67·4) . The overall process is outlined in Fig. 4. The arachidonic acid
in the purified oil can be isolated as the methyl or ethyl ester in a good
recovery after successive transesterification, liquid-liquid partition chromatog-
raphy and high-performance liquid chromatography.
I
2.0~
J 1.0 CII
II)
o
o ~
~2.0 "
• .c!J
1 20 :
"';;-1.0
<"c
...J
"o
l:
I
I
I
,
I
I
Time (days)
acid, suggesting that all the arachidonic acid-producing fungi potentially have
the ability to produce large amounts of this fatty acid.
Shimizu et al. (1989a) reported that the mycelial dihomo-y-linolenic acid
content of M. alpina 1S-4 increases, with an accompanying marked decrease in
its arachidonic acid content, on cultivation with sesame oil. They suggested
that this unique phenomenon is due to specific repression of the conversion of
dihomo-y-linolenic acid to arachidonic acid by the oil. The effective factor(s)
causing this phenomenon was found to be present in the non-oil fraction of the
oil, after fractionation of the oil with acetone. In a study on optimization of the
culture conditions for the production of dihomo-y-linolenic acid by M; alpina
1S-4, a medium containing glucose, yeast extract and the non-oil fraction was
found to be suitable for the production. Under the optimal conditions in a
SO-liter bench scale fermentor, the fungus produced 2·17 g/liter of dihomo-y-
linolenic acid (107 mg/ g dry mycelia) (Fig. S). This value accounted for 23·1 %
of the total mycelial fatty acids. The mycelia were also rich in arachidonic acid
(S3·S mg/g dry mycelia, 11·2%). Other major fatty acids in the lipids were
palmitic acid (24·1%, by weight), stearic acid (7·0), oleic acid (20·1), linoleic
acid (6·6) and y-linolenic acid (4·1).
4.4 Eicosapentaenoic Acid (EPA)
4.4.1 EPA -Production under Low Temperature Growth conditions
Yamada and co-workers (Shimizu et al., 1988a; Yamada et aI., 1988) investi-
gated the growth conditions causing compositional changes in mycelial fatty
Microbial Production of PUFAs 117
acids using the arachidonic acid producers obtained through their screening
studies, because they could not find any strains capable of accumulating a
detectable amount of EPA under their screening conditions (see Table 3).
They found that lowering the cultivation temperature caused the additional
accumulation of a PUFA with five double bonds, EPA, by all the arachidonic
acid producers tested, as shown in Table 5. In all cases, cultivation at low
temperature significantly increased the mycelial phospholipid content. For
example, the mycelial lipids extracted from M. alpina 1S-4 grown at 12°C
contained 47·6% (by dry weight) of phospholipids and 35·7% of triglycerides.
More than 60% of the EPA accumulated in the mycelia was found in the
phospholipid fraction. On the other hand, the lipid from the mycelia grown at
28°C contained only 21·6% of phospholipids, the triglyceride fraction compris-
ing 73·5% of the total extractable lipids. It should be noted that most of these
arachidonic acid producers grow well at low temperature (6-16°C), and yield
enough mycelia in simple growth media.
Subsequently, they studied the mechanism involved in this unique phenome-
non. The results of experiments with cell-free extracts of M. alpina 1S-4
demonstrated that the enzyme(s) that catalyzes the formation of EPA is
produced even when the fungus is grown at high temperature, but that the
reaction(s) yielding EPA does not take place at high temperature, as shown in
Table 6. Since all the EPA-producing Mortierella strains accumulate PUFAs of
the n-6 family (i.e. y-linolenic acid, dihomo-y-linolenic acid and arachidonic
acid) in their mycelia and do not accumulate PUFAs of the n-3 family other
than EPA, they suggested that an n-6 PUFA, probably arachidonic acid, may
be a precursor of EPA. If this is the case, an enzyme(s) or enzyme system
catalyzing the methyl-end directed desaturation of arachidonic acid (a17-
desaturation) may be activated on cold adaptation. The resultant EPA may be
necessary for maintaining a proper membrane fluidity in a low temperature
environment (Shimizu et al., 1988a).
Among various arachidonic acid-producing strains, they selected M. alpina
20-17 as the most promising EPA producer. The fungus was found to
accumulate about O· 5 g/liter (26·6 mgt g dry mycelia) of EPA on cultivation
for 7 days at 12°C. This value accounted for 11 % of the total fatty acids in the
extracted lipids. Other major fatty acids in the lipids were palmitic acid (6·4%,
by weight), stearic acid (4·8), oleic acid (3·2), linoleic acid (3·1), y-linolenic
acid (4·5) and arachidonic acid (63·8) (Shimizu et al., 1988b).
Table 5
Formation of EPA and Changes in Mycelial Fatty Acid Composition in Mortierella Fungi at Various Growth temperaturea
M. exigua 12 5·45 9·8 1·5 22·5 9·9 6·0 3·2 37·5 7·0 2·6 ~
IFO 8571 28 0 22·3 7·2 33·1 10·2 4·7 3·6 18·0 0 0·9
M. parvispora 12 6·97 10·1 4·0 43·3 7·0 8·7 2·7 15·3 5·2 3·7
20-24 AKU 3997 28 0 13·2 3·7 54·6 5·5 5·4 1·7 15·5 0 0·4
a For details, See Shimizu et al. (1988a).
b Values are given in mg/g dry mycelia.
C Values are given in weight %. Any fatty acid of the n-3 family other than EPA was not detected. For abbreviations of fatty
Table 6
Enzymatic Formation of EPA by a Cell-Free Extract of M. alpina 1S-4
Grown at 28°C'
5 CONCLUSION
Considerable progress has been made in biotechnological production of
PUFAs during the past years. Areas of progress include: (a) the finding of
120 S. Shimizu & H. Yamada
REFERENCES
Borowitzka, M. A. & Borowitzka, L. J. (1988). Micro-algal Biotechnology. Cambridge
University Press, Cambridge, UK.
Brenner, R. R. (1974). The oxidative desaturation of unsaturated fatty acid. Molec.
Cell. Biochem. 3,41-52.
Dyerberg, J. (1986). Linolenate-derived polyunsaturated fatty acids and prevention of
atherosclerosis. Nutr. Rev. 44, 125-34.
Erwin, J. & Bloch, K. (1964). Biosynthesis of unsaturated fatty acids in micro-
organisms. Science, 143, 1006-12.
FIuco, A. J. (1974). Metabolic alterations of fatty acids. Ann. Rev. Biochem., 43,
215-41.
Fukuda, H. & Morikawa, H. (1987). Enhancement of ')I-linolenic acid production by
Mucor ambiguus with nonionic surfactant. Appl. Microbiol. Biotechnol. 27, 15-20.
Gellerman, J. L. & Schlenk, H. (1979). Methyl-directed desaturation of arachidonic
acid to eicosapentaenoic acid in the fungus, Saprolegnia parasitica. Biochim.
Biophys. Acta, 573, 23-30.
Hansson, L. & Dostalek, M. (1988). Effect of culture conditions on mycelial growth
and production of ')I-linolenic acid by the fungus Mortierella ramanniana. Appl.
Microbiol. Biotechnol., 28, 240-46.
Hartmann, E., Beutelmann, P., Vandekerkhove, 0., Euler, R. & Kohn, G. (1986).
Moss cell cultures as sources of arachidonic acid and eicosapentaenoic acids. FEBS
Lett., 198,51-5.
Holloway, P. W. (1983). Fatty acid desaturation. In Enzymes, 3rd edn., Vol. XVI, ed.
P. D. Boyer, pp. 63-83. Academic Press, New York.
Horrobin, D. F. & Huang, Y.-S. (1987). The role of linoleic acid and its metabolites in
the lowering of plasma chloesterol and the prevention of cardiovascular disease. Int.
J. Cardiol., 17, 241-55.
Needleman, P., Turk, J., Jakschiik, B. A., Morrison, A. R. & Lefkowith, J. B. (1986).
Arachidonic acid metabolism. Ann. Rev. Biochem., 55,69-102.
Numa, S. (Ed.) (1984). Fatty Acid Metabolism and its Regulation. Elsevier,
Amsterdam.
Okayasu, T., Nagao, M., Ishibashi, T. & Imai, Y. (1981). Purification and partial
characterization of linoleoyl-CoA desaturates from rat liver microsomes. Arch.
Biochem. Biophys. 206,21-8.
Microbial Production of PUFAs 121
Rahm, J. J. & Holman, R. T. (1971). Essential fatty acids. In The Vitamins, Chemistry,
Physiology, Pathology, Methods, 2nd edn, Vol. III, ed. W. H. Sebrell Jr & R. T.
Harris, pp. 303-39. Academic Press, New York.
Seto, A., Wang, H. L. & Hesseltine, C. W. (1984). Culture conditions affect
eicosapentaenoic acid content of Chlorella minutissima. J. Am. Oil Chem. Soc., 61,
892-4.
Shimizu, S., Shinmen, Y., Kawashima, H., Akimoto, K. & Yamada, H. (1988a).
Fungal mycelia as a novel source of eicosapentaenoic acid, Activation of enzyme(s)
involved in eicosapentaenoic acid production at low temperature. Biochem.
Biophys. Res. Commun., 150, 335-41.
Shimizu, S., Kawashima, H., Shinmen, Y., Akimoto, K. & Yamada, H. (1988b).
Production of eicosapentaenoic acid by Mortierella fungi. J. Am. Oil Chem. Soc.,
65,1455-9.
Shimizu, S., Akimoto, K., Kawashima, H., Shinmen, Y. & Yamada, H. (1989a).
Production of dihomo-y-linolenic acid by Mortierella alpina IS-4, J. Am. Oil Chem.
Soc. 66, 237-41.
Shimizu, S., Kawashima, H., Akimoto, K., Shinmen, Y. & Yamada, H. (1989b).
Microbial conversion of an oil containing a-linolenic acid to an oil containing
eicosapentaenoic acid. J. Am. Oil Chem. Soc. 66, 342-7.
Show, R. (1965). The occurrence of y-linolenic acid in fungi. Biochim. Biophys Acta,
98,230-37.
Suzuki, O. (1985). y-Rinorensan no biseibutsu seisan (Production of y-linolenic acid by
micro-organisms). Hakko to Kogyo (Fermentation and Industry), 43, 1024-31.
Suzuki, O. (1988). Production of y-linolenic acid by fungi and its industrialization. In
Proceedings of the World Conference on Biotechnology for the Fats and Oils
Industry, Hamburg. American Oil Chemists' Society, Champaign, Illinois, pp.
110-16.
Suzuki, O. & Yokochi, T. (1986). Production of y-linolenic acid by fungi. J. Am. Oil
Chem. Soc., 63, 434.
Wagner, A. F. & Folkers, K. (1964). The essential fatty acid group. In Vitamins and
coenzymes. Interscience Publishers, New York, pp. 389-406.
Yamada, H., Shimizu, S. & Shinmen, Y. (1987). Production of arachidonic acid by
Mortierella elongata IS-5. Agric. BioI. Chem. 51, 785-90.
Yamada, H., Shimizu, S., Shinmen, Y., Kawashima, H. & Akimoto, K. (1988).
Production of arachidonic acid and eicosapentaenoic acid by micro-organisms,
Proceedings of the World Conference on Biotechnology for the Fats and Oils
Industry, Hamburg. American Oil Chemists' Society, Champaign, Illinois, pp.
173-7.
Yazawa, K., Araki, K., Okazaki, N., Watanabe, K., Ishikawa, C., Inoue, A., Numao,
N., & Kondo, K. (1988). Production of eicosapentaenoic acid by marine bacteria. J.
Biochem., 103, 5-7.
Chapter 8
Y. TANI
Research Center for Cell and Tissue Culture, Faculty of Agriculture,
Kyoto University, Kyoto 606, Japan
1 INTRODUCTION
2 CHEMISTRY OF MENAQUINONES
The chemistry of vitamin K was established by the schools of Dam, Doisy and
Karrer, who succeeded in 1939 in isolating two compounds with vitamin K
activity, namely MK-7 (vitamin K 2 (35» from putrefied fish meal and phyllo-
quinone from alfalfa meal (Fig. 2). The structure of phylloquinone was proven
to be 2-methyl-3-phytyl-l,4-naphthoquinone, and that of MK to be 2-methyl-3-
multiprenyl-l,4-naphthoquinone. The multiprenyl side chain of MK is of
variable length from 5 to 70 carbon atoms. A number of derivatives of the
basic structure also occur in nature; this is reflected in the modification of ring
substitutents at the C-2 or C-3 position or both.
Phylloquinone is a yellow viscous oil and MKs are light yellow microcrys-
talline plates. Melting points of MKs vary from 35 to 62°C on the length of the
multiprenyl side chain. These are soluble in ether, petroleum ether, benzene,
n-hexane and acetone, slightly soluble in methanol and alcohol, and insoluble
in water. These are stable to air and heat, but very unstable for alkali and UV
irradiation. The absorption maxima are 243,248,261,270 and 325 nm with the
molecular extinctions coefficient of 18000-19000 at 248 nm.
(a) (b)
Fig. 2. Chemical structure of vitamin K. (a) Menaquinone-n (MK-n); (b) phyllo-
quinone.
Vitamin K2 and Vitamin Kl Production 125
3 BIOSYNTHESIS OF MENAQUINONES
Intense research based on microorganisms during the last two decades has
clearly proven the MK biosynthetic pathway as summarized by Bentley &
Meganathan (1982).
The biosynthetic pathway of MK in bacteria is now believed to consist of the
biosynthesis of demethyl MK by polyprenylation of 1,4-dihydroxy-2-
naphthoate, which is formed as the first naphthalenoid intermediate from
shikimate through chorismate(isochorismate) and o-succinylbenzoate, by ca-
talysis of a membrane-associated transferase and its subsequent methylation.
The methyl group is derived from S-adenosylmethionine. Biosyntheses of MK
and aromatic amino acids share a common route until chorismate. The
immediate precursor, shikimate (chorismate), is incorporated to non-carboxyl
carbon atoms of 2-ketoglutarate to form the naphthoquinone nucleus. The
biosynthetic pathway of MK can now be summarized as shown in Fig. 3, to
which recent experimental results are added.
HO~COOH
,
H O Y 8hlklmata HOOC~H
,
Mavaloneta
HOO'--oOCOOH
HZ~
HO Chorlemeta
6-
OOH
, ~~g
,
Olmethylallyl-PPI l.opentenyl-PPI
Hoar.. OOH
H.<f° O COOH COZt
Z Ieochorleo TPP
meta
, ,.CHZCHZCOOH ~P
..... HO-",
6
TPP
COOH Oerenyt-PPI
2-8ucclnyH-h clroxr-
2,4-cyctohexa
COCil CH eOOH
2 Z
c~rtele
- - , _ I.opentenyt-PPI
,
,
Feme.y""""
,
, _ (n-3)xl.opentenyI-PPI
Polypr...yI-PPI
; Oemethylmenaqulnone-n
Manaqulnone-n
120 12i
()
0
EJ MK-4 I
a>
100 0 MK-6 100,
E
:::::
a> C
! 80 8 01)
loC C
.. 0
:f ()
loC
60 6
:v I
I
:f
-
III
340
..
I
I 4 !D
:u
20
...
.. ,.
l 2
1:
a>
a:
r1l r I
Fig. 4. Improvement of menaquinone production by Flavobacterium sp. 238-7. The left
and right bars indicate the amount of MK in mg/liter of culture broth and mg/g of dry
celis, respectively, the open and shaded bars denoting MK-6 and MK-4, respectively.
The amounts of MK produced by each strain indicated are the values when grown on
the basal medium, except that K3-1sa was grown in the optimized medium and K3-1S b in
the optimized medium supplemented with cedar wood oil.
Vitamin K2 and Vitamin K) Production 131
10
20
~
eft ~
.
.§ ~
0
'a 0
u .g
..
::a
ie
'a
0
a.
laC C)
::Ii! 10
~ 20
£ ~
~
•u:::a
'a 0
0
'a
0
S
a :5
laC
::E 10 5
.
~
0
C)
productivity increased about 20% compared to that without the detergent, and
the total amount of MK was 39 mg per liter of culture broth.
Figure 6 shows the fermentation course of MK production on cultivation in
the presence of 0·05% Rikanon VA 5012. The cell mass reached a maximum
level, about 10 g by cell weight per liter of culture broth, after 30 h cultivation,
and then gradually decreased. MK-6 was mainly produced intracellularly and
reached about 12 mg per liter of culture broth after 30 h cultivation. A small
amount of MK-6 was detected in the culture filtrate at the late stage of the
cultivation. On the other hand, the excretion of MK-4 occurred from the
beginning of the cultivation and increased in the later logarithmic phase. The
extracellular production of MK-4 reached its maximum after 48 h cultivation.
The level of intracellular MK-4 was low but increased with increasing
cultivation time. The total amount of MK-4 reached about 25 mg per liter of
culture broth at 48 h cultivation, and the amount of total MK was 39 mg liter.
MK-4 excretion occurred from the onset of the cultivation and the excreted
MK was only MK-4. These facts showed that the excretion of MK-4 was not
due simply to cell lysis. The MK-4leakage might be due to a change in the cell
membrane structure, as a result of exposure to the detergent during growth.
The role of MK-4 in the mutant strain remains unknown, it might be that
MK-4 is not related to the respiratory function, since the growth and growth
rate of strain K3-15 did not decrease when most of the MK-4 had been
excreted.
REFERENCES
A.IwASHIMA
Department of Biochemistry, Kyoto Prefectural University of Medicine,
Kyoto, Japan
1 HISTORICAL
Thiamine was discovered in the course of a search for an agent that would cure
beriberi. The conquest of beriberi began in 1885 when Takaki (1885)
practically eradicated the disease among the Japanese navy by introducing fish,
vegetables, meat and barley into the diet. In 1897 Eijkman (1897) showed that
an experimental polyneuritis in fowl, which closely resembled the polyneuritic
symptoms of beriberi, could be produced by feeding the birds on a diet of
polished rice. When they were fed on unpolished rice they did not develop the
disease. In 1926 Jansen & Donath (1926) isolated a crystalline hydrochloride of
the antineuritic factor from rice bran.
The structure of thiamine, elucidated in the mid-1930s, was established
by Williams (1936) as 3-(4-amino-2-methyl-5-pyridimidinylmethyl)-5-(tJ-
hydroxyethyl}-4-methylthiazolium chloride hydrochloride (Fig. 1). The name
thiamine derives from the chemical nature of the vitamin, in that it has the
thiazole (sulfur-containing) ring attached to a pyrimidine ring with an amine
group. Thiamine was first synthesized in 1936 by Williams & Cline (1936) from
the pyrimidine and thiazole moieties of thiamine. Within a year, Lohmann &
Schuster (1937) isolated thiamine pyrophosphate (TPP) from yeast and showed
that it was the cofactor for the decarboxylation of pyruvic acid. In addition to
thiamine and TPP, thiamine monophosphate (TMP) and thiamine triphosphate
(TIP) have been found in living organisms. These phosphate esters of
thiamine are also shown in Fig. 1.
The double-salt from thiamine with hydrochloric acid (C 12H 17N 4 0SCI HCI;
molecular weight, 337·28) consists of monoclinic plates in rosette-like clusters
137
138 A. Iwashima
N
I
N
H1C"",,:x
I
NH2
U- 1!:
CH2~~ 4' CH,CH,OR
H,
Thiochrome
R= -H Thiamine
Q
R= -P-OH
I
TMP CH,
OH W:-Ofl
R=
0
II
0II
-P-O-P-OH
(JH OHI
TPP NJ.
- CHi' Nt::
HJCyNyNH'lr--s
""
0 0
~ C,H,O-P-O-P-OH
HJ 6H 6H
0 0 0
R= -P-O-P-O-P-OH TTP Hydroxyethylthlamlne pyrophosphate
ClH elH
OH
Fig. 1. Structures of thiamine and its related compounds.
with slight thiazole odor (Merck Index, 1983). It is soluble in water and
methanol, slightly soluble in ethanol, and practically insoluble in ether,
benzene and chloroform. In aqueous solution thiamine is most stable between
pH 2 and 4, but unstable at alkaline pH (Yurugi et aI., 1979). It is
destroyed by heat if the pH of the solution is above 5·5: the rate of destruction
increases with the pH. Under dry conditions thiamine is stable to heating at
100°C for 24 h. The absorption maximum of thiamine in the UV light is
affected by pH, and it is 243 nm at pH 2·3 (the molar absorption coefficient:
10·6 X 103 ), and 235 and 265 nm at pH 7·4 (Yurugi et al., 1979).
TPP chloride (commercially available in the dried form) is stable when
stored cold in the dark (Yurugi, 1975). In aqueous solution it is stable at
pH 2-6 and O°C for 6 months, but it partially decomposes to TMP and
thiamine when allowed to stand for several months at pH 5 and 38°C. TIP is
stable under dry conditions, whereas it is unstable in aqueous solution,
especially to heat, and decomposes to TMP and TPP. In alkaline solution TIP
decomposes to TMP and inorganic phosphate. TMP, which is obtained by
hydrolysis of TIP and TPP, is quite stable at acidic and neutral pH. Thiamine
and thiamine phosphate esters are quantitatively converted to thiochrome (Fig.
1) and its phosphate esters which are compounds exhibiting intense blue
fluorescence. This property observed in early studies of thiamine oxidation
serves as the basis for the chemical assay for thiamine (Barger et al; 1935).
Hydroxyethylthiamine pyrophosphate (Fig. 1), a TPP-activated aldehyde
intermediate of the enzyme reactions, it is also known to be present in
micro-organisms and mammalian tissues, especially in muscles (Morita et al.,
1968).
3 PRODUCING MICRO-ORGANISMS
Thiamine, as well as other vitamins, is produced by micro-organisms in
extremely small amounts. Usually it is not formed in great excess over the
Microbial Synthesis of Vitamin B 1 139
thiamine. Washed cell suspensions of E. coli ATCC 9637 (15-20mg dry cells)
synthesizes 210 ng of thiamine per mg dry weight from 10 IlM each of the two
moieties in the presence of 0·2% glucose at pH 7 for 1 h at 37°C (Iwashima et
al., 1968). With the cells derepressed thiamine biosynthesis by adenine or
thiamine auxotrophs grown in the presence of limited amount of thiamine the
production of thiamine is much more enhanced (Kawasaki et al., 1969). The
cells suspension of baker's yeast also has the capability of joining the
pyrimidine and thiazole moieties of thiamine to form thiamine in the presence
of glucose (Ashida, 1942). In S. cerevisiae, thiamine synthesis from both
moieties added leads to 20-40 times higher contents of intracellular thiamine
in comparison with normal growth conditions (Silhankova, 1985b).
4.1 Biosynthesis
Nl.f1~.'
CH,CH.OH
H.
_7, N
H.
9
CH.CH.O-P-OH
6H
~
PPt
N
N NH, N
Hlcy:x
CHz'~
13: CH,CH.O-f-OH
9
H. OH
Fig. 2. The biosynthesis of thiamine from its pyrimidine and thiazole moieties.
Microbial Synthesis of Vitamin B 1 141
by compounds that precede AIR on the purine pathway, indicating that AIR is
the last common intermediate on the route to the purines and to the
pyrimidine moiety of thiamine. The biosynthetic pathways of hydroxy-
methylpyrimidine in micro-organisms differ between bacteria and yeast. In
bacteria, hydroxymethylpyrimidine is synthesized from AIR, and N-1, C-4 and
C-6 of the pyrimidine are derived from the nitrogen, C-1 and C-2 of glycine,
respectively (Estramareix & Lesieu, 1969; Estramareix, 1970; White &
Rudolph, 1979), and the C-2 from formate (Kumaoka & Brown, 1967;
Estramareix & Lesieu, 1969). These results suggest that the imidazole ring is
opened between C-4 and C-S. The remaining three carbon atoms of hydroxy-
methylpyrimidine (C-S, C-7 and C-8) originate from the ribose part of AIR
which is thus the precursor of all the carbon atoms of this pyrimidine
(Estramareix & Therisod, 1984). Since N-3 and the amino group at C-4 have
been recently shown to be derived from the amide-N atom from glutamine in
E. coli and S. cerevisiae (Tazuya et ai., 1987a), the origin of almost all of the
pyrimidine moiety of thiamine is now known for bacteria, although the
reactions whereby AIR is converted to the pyrimidine remain unknown. In
yeast, on the other hand, C-4 of the pyrimidine originates from formate (David
et ai., 1966) and no significant incorporation of the nitrogen atom and C-2 of
glycine into the pyrimidine is found (Linnett & Walker, 1968; White &
Spenser, 1979). Furthermore, the incorporation of C-3, C-4 and C-S of
pentulose into C-6, C-S and C-7 of the pyrimidine has been recently reported
in S. cerevisiae (Grue-S0rensen et ai., 1986). The origin of C-2, C-8 and N-1 of
hydroxymethylpyrimidine remains to be clarified in the biogenesis of the
pyrimidine moiety in yeast.
Relatively little has been discovered regarding the biosynthetic origins of the
thiazole moiety of thiamine, but some progress has been made in recent years
toward the identification of its precursors. In E. coli and S. typhimurium the
direct utilization of C-2 and the nitrogen atom of tyrosine for the formation of
the thiazole ring was demonstrated (Estramareix & Therisod, 1972; Bellion et
ai., 1976; White & Rudolph, 1978). On the other hand, no incorporation of
C-2 of tyrosine into thiamine is observed in yeast, whereas significant
incorporations of C-2 and the nitrogen atom of glycine into C-2 and N of the
thiazole are shown (Linnett & Walker, 1968, 1969). These observations seem
to indicate that bacteria and yeast use different amino acids as precursors of
the C-2 and nitrogen portion of the thiazole ring, implying that the pathways in
these two micro-organisms are different. Further incorporation studies with
deuterated carbohydrates and related compounds showed that the precursor of
the S-carbon chain (C-4, C-S, C-6, C-7 and C-8) of the thiazole moiety of
thiamine in E. coli is a S-carbon sugar derived from pyruvate and triose
phosphate (White, 1978), which might then react with tyrosine and a sulfur
compound in an undefined series of steps to yield hydroxyethylthiazole. This
has been supported by the incorporation of 1-·deoxY-D-threo-pentulose into the
S-carbon chain of the thiazole without carbon-carbon bond cleavage in E. coli
(David et ai., 1981). In yeast, the incorporation pattern leads to the inference
142 A.lwashima
that the 5-carbon chain may not be derived from pyruvate but from
2-pentulose, which is generated from the hexose precursors by oxidative as
well as non-oxidative pentose phosphate pathway (White and Spenser, 1982).
The precursor of the sulfur atom of the thiazole ring is still not known with
certainty, but several experimental observations suggest that either cysteine or
H 2 S which could be produced from cysteine, is the most likely precursor of the
sulfur atom of hydroxyethylthiazole in bacteria and yeast (Bellion & Kirkley,
1977; Tazuya et al.; 1987b).
4.2 Regulation
5 ASSAY METHODS
5.1 Microbiological Assays
Assay methods based on use of the thiamine requiring fungus Phycomyces
blakesleeanus and thiamine requiring strains of several bacteria such as
Staphylococcus aureus, E. coli and Lactobacillus species, and S. cerevisiae
have been used (Thomas, 1966).
5.2 Chemical Assays
Chemical assay methods are usually preferred to microbiological methods since
such analyses can be performed rapidly and they are usually more reliable for
routine determination. Although a number of chemical assay methods for
thiamine are known, only the coupling reaction with diazotized p-
aminoacetophenone (Melnick & Field, 1937) and thiochrome reaction have
been developed as precise methods of assay. In recent years, however, the
thiochrome method has been used almost exclusively. As described above,
thiamine in alkaline solution is oxidised quantitatively to thiochrome. As
oxidizing reagents potassium ferricyanide (Hennessy & Cerecedo, 1939) and
cyanogen bromide (Fujiwara & Matsui, 1953) have been routinely used.
Thiochrome formed from non-phosphorylated thiamine is extracted with
isobutanol. Since thiochrome fluoresces intensely under UV illumination, it
can be measured readily in a fluorimeter (excitation 365 nm; emission 430 nm).
Phosphorylated thiamines can be determined after their hydrolysis to thiamine
by phosphatase. Hydroxyethylthiamine gives thiochrome by oxidation with
alkaline ferricyanide, but not with cyanogen bromide. Thiamine gives thio-
chrome by either oxidizing agent, so that this difference in the oxidation
property is used for the simultaneous determination of thiamine and hydroxy-
ethylthiamine (Morita et al., 1968). For the differential determination of
thiamine and its phosphate esters in biological materials electrophoresis, paper
chromatography, thin-layer chromatography and ion-exchange chromatog-
raphy can be used to separate thiamine compounds from each other (Yurugi et
al., 1979). In recent years, high-performance liquid chromatography (HPLC)
has been developed for simultaneous determination of thiamine and its
phosphate esters and assay procedures have been established (Kawasaki &
Sanemori, 1985). The oxidation of thiamine compounds in samples can be
carried out either before the chromatography (precolumn derivatization
procedure) or after the chromatography (postcolumn derivatization proce-
dure). Therefore, the precolumn derivatization procedure is essentially the
HPLC of thiochrome and its phosphate esters. A mixing coil with proportion-
ing pump as an additional equipment for thiochrome reaction is required for the
postcolumn derivatization procedure. The intensity of the fluorescent products
is measured with a spectrophotofluorometer. The minimum detection by these
methods has been reported to be 0·1 pmol or 33·7 pg as thiamine hydro-
chloride (Sanemori et al., 1980).
144 A.lwashima
6 BIOLOGICAL PROPERTIES
6.1 Metabolism
7 CHEMICAL SYNTHESIS
Thiamine has been synthesized in two basic ways (Matsukawa et al., 1970).
The condensation method, first adapted to the synthesis of thiamine by
Williams & Cline (1936), consists of the condensation of the pyrimidine with
thiazole derivatives. In another method, which was devised by Todd & Bergel
(1937), the pyrimidine moiety is synthesized with an appropriate side chain
conductive to the formation of the remaining structure. Matsukawa (1953)
synthesized thiothiamine as an intermediate for thiamine synthesis. This is
formed by the condensation of 4-amino-5-aminomethyl-2-methylpyrimidine,
CS 2 and 3-acetyl-3-chloro-1-propanol. Thiothiamine is a water-insoluble crys-
talline compound and is readily converted by oxidation into thiamine in good
yield.
REFERENCES
1 INTRODUCTION
2 HISTORICAL
Fig. 1. Riboflavin.
Riboflavin (C 17H zoN40 6), crystallizes from 2 N acetic acid, alcohol, water or
pyridine in orange-yellow needles. Its molecular weight is 376·36. The
decomposition point is 278-282°C, and it darkens at about 240°C. The vitamin
is odorless and has a bitter taste. Riboflavin is soluble in water only to the
extent of 10-13 mg in 100 ml at 25-27·5°C, 19 mg in 100 ml at 40°C, and
230 mg in 100 ml at 100°C. The vitamin dissolves in ethanol and is slightly
soluble in amyl alcohol, cyclohexanol, benzyl alcohol, phenol and amyl
acetate, but is insoluble in ether, chloroform, acetone and benzene. Although
alkali dissolves the vitamin well, these solutions are unstable.
Neutral aqueous solutions of riboflavin have a greenish-yellow color. The
absorption spectrum shows characteristic absorption maxima at 475, 446,
359-375, 268 and 223 nm. The absorption in the visible part of the spectrum is
used for quantitative determination of riboflavin (Wagner-Jauregg, 1972).
Neutral aqueous solutions of riboflavin display intense yellowish-green fluores-
cence, with a maximum at 565 nm which can be used for quantitative
determination of the vitamin. The fluorescence vanishes on the addition of acid
or alkali, optimum fluorescence occurs at pH 3-8.
Riboflavin is an amphoteric compound. Its dissociation constants are
Ka = 6·3 X 1O- 1z and KfJ = 0·5 X 10-5 , the isoelectric point corresponds to a
pH of 6·0.
The basic factors affecting the stability of riboflavin in food are heat, light
and the reactions which occur in cells during the storage. Neutral solutions of
riboflavin can be sterilized by autoclaving for a short time: only slight
destruction occurs by heating to 120°C for 6 h. No appreciable destruction of
the vitamin can be observed during the cooking of food since it is relatively
heat stable, but alkali decomposes riboflavin rapidly (Wagner-Jauregg, 1972;
Heimann, 1980). When milk in bottles is exposed to sunlight, 85% of its
riboflavin content is destroyed within 2 h.
Riboflavin is stable against air and most common oxidizing agents, but when
reduced by reducing agents such as sodium hydrosulfite or hydrogen, riboflavin
readily takes up two hydrogen atoms to form the colorless and non-fluorescent
1,1O-dihydro compound, leucoriboflavin, which can be reconstituted by shak-
ing an aqueous solution with air. This reduction and reoxidation is also useful
Microbial Production of Vitamin B2 151
R = D-Ribityl
III IV
Fig. 2. The biosynthesis of riboflavin.
That lumazine III alone is the precursor of riboflavin (IV) in organisms such
as E. ashbyii and A. ashbyii has been established by biochemical studies. All
the carbon atoms of the o-xylene moiety of riboflavin are derived from the
lumazine III, two molecules of which are converted by riboflavin synthesis into
one molecule of riboflavin (IV) and 4-ribitylamino-5-amino-2,6-dihydroxy
pyrimidine. In this biochemical reaction the lumazine (III) simultaneously
functions as a donor and acceptor of a C 4 unit which consists of the two methyl
groups and the C atoms 6 and 7. There is also a similar mechanism for
riboflavin formation in plants (Wagner-Jauregg, 1972).
Goodwin et al. showed independently that amino acids provide carbon units
and not nitrogen in the biosynthesis of riboflavin. These workers used E.
ashbyii and found that L-threonine, L-serine or L-tyrosine stimulated riboflavin
synthesis whereas L-glutamate, L-aspartate and L-asparagine stimulated both
growth and riboflavin synthesis; L-cystein on the other hand inhibited both
(Goodwin, 1963a,b; Robinson, 1966).
6 STRAIN IMPROVEMENT
The changes in the culture media and the selection of high riboflavin-producing
mutant cultures cause the most important increases in riboflavin productivity.
In 1950, Pfeifer et al. suggested that some improvement in fermentation
productivity was possible as a result of culture selection. Paralleling the pilot
plant fermentations, a special study was carried out in the laboratory by
Pridham of the NRRL fermentation division to obtain and test natural and
induced variants of Ashbya gossypii NRRL Y -1056 for riboflavin production.
One of the natural variants was chosen as superior to the original strain on the
basis of shaker-flask experiments and was tested in a series of pilot plant
fermentations. This particular strain gave consistently higher yields of
154 T. Kutsal & M. T. Ozbas
riboflavin than did the standard NRRL Y -1056. The results indicate the
desirability of isolating high-producing strains at regular intervals in order to
maintain riboflavin yields at the highest possible values (Pfeifer et al., 1950).
7 FERMENTATION/BIOCONVERSION PROCESS
7.1 Inoculum
As reported in Section 4, a great number of cheap natural materials and
industrial wastes have been used for riboflavin fermentation by these Ascomy-
cetes (Hickey, 1954; Goodwin, 1959; Robinson, 1966; Gutcho, 1973; Ozbas,
1985). In order to produce very large amounts of riboflavin, an appropriate
combination of these substrates can be chosen and if necessary, several
minerals can be added to the medium.
The generally accepted inoculating procedure is to use a 0·25-1·0%
inoculum from an actively growing 24-28 h culture (Goodwin, 1959).
7.2 Medium
In laboratories, cultures of E. ashbyii and A. gossypii are grown on solid agar
media in slant tubes and Petri dishes. The composition of the culture medium
is as follows (g/liter): glucose, 20·0; peptone, 5·0; yeast extract, 5·0; malt
extract, 5·0; MgS04 ·7H2 0, 0·2; K2 HP0 4 , 0·2. The -pH of the medium is
adjusted to the desired value with 0·5M H 2 S04 (Ozbas et al., 1984, 1986a,b).
Both liquid and solid media are sterilized at 121°C in an autoclave before
inoculation. The initial substrate concentrations can be changed by making the
amount of initial carbon in the medium equivalent to 20·0 g/liter glucose while
the amounts of other constituents are kept constant.
0.15
.
//:--..-
, \ 3.00
,
L:
-
0
0 E
-~0.10 2.00~
::L.
~o It:
QI
~
0
, '~i
0 0
6.0 6.5 7.0 7.5
pH
Fig. 3. Effect of initial pH on the specific growth (I-') and riboflavin production (v)
rates for E. ashbyii and A. gossypi (30°; stirring rate 100 rpm; SGO = 20·0 g/liter; . , E.
Ashbyii; 0, A. gossypii.
::.
/:#:~_o_o 0.14
0.12
8.0
7.0
~Jo\I// "\
-
0.10 _ 6.0
....
1ot -
0.08 ~ 5.0 ::r:
It: Co
U
0.06 4.0
_
2.0
0.04 3.0
1.0 / \ 0.02 2.0
0.15 3.00_
.J::.
b
E
t7l
";".c 0.10 2.00-;;::
'~~\~o-r-o
01
1- ",-
e
.,.
I~I .'\.~
0.05 1.00
--. I
0 0 0
20 25 30 35 40
T (oC)
Fig. 5. Effect of temperature on the specific growth and riboflavin production rates for
E. ashbyii (stirring rate, 100 rpm; initial pH = 6·5; 0, SGo = 20·0 g/liter; . , SSo =
20·0 g/liter).
0.30
lei 4.00
0.20
/;or" '\ 300~
:/,/" \ \ 2DO!
0.10 ,./ ~:~ i;
/0. . . . . 0"'0 ~ 1.00
?
'r'-0
1----°
-----0
0~2~0----~2~5~--~3~0----~35~----4~0~0
T (DC)
e
Fig. 6. Effect of temperature on the specific growth and riboflavin production rates for
A. gossypii (stirring rate, 100 rpm; initial pH = 6·5; 0, Sao = 10·0 g/liter, Sso =
1O·0g/liter; e, Sao = 20·0 g/liter).
production rates and riboflavin yields are given in Table 1 (Ozbas & Kutsal,
1986a,b).
7. 3. 4 Fermentor experiments
The effects of agitation and aeration rates on the formation of riboflavin were
observed at the range of 300-1200 rpm and 70-330 cc/min, respectively for
Eremothecium ashbyii. Optimum values were reached as Il = 0·176 h- 1 and
v = 4·45 X 10- 3 g/liter per hat 750-1000 rpm and Il = 0·181 h- 1 and v = 3·69 X
10-3 g/liter per h at 220 cc/min. At these optimum conditions, adding the
sunflower oil to the medium resulting in 5·0 g/liter glucose and 15·0 g/liter
sunflower oil showed Il to be 0·281 h- 1 and v = 5·34 X 10- 3 g/liter per h. These
results suggested that in addition to pH, temperature and initial substrate
concentrations, agitation and aeration are also important control parameters
on the riboflavin production (Ozbas et al., 1984).
Table 1
Effects of Various Substrates on Specific Growth (It) and Riboflavin Production (v) Rates and Riboflavin Yields (YP/SCa)
for E. ashbyii and A. gossypii (30"C, Stirring Rate: 100 rpm, Initial pH = 6·5)
....
VI
\C
160 T. Katsal & M. T. Ozbas
7.4 Flow-Sheet
2 3 4 5
Sulphuric Evaporator
acid
~
~.
~
\)'
s·
8 "tI
-
Fermentor ~
Mixing -._- ~g.
tank . :s
~
~
Water out
10
I·
b:I
N
water in
Air compressor
Fig. 7. Flow sheet for riboflavin production by fermentation (Pfeifer et al., 1950). 1, Glucose; 2, com steep liquor;
3, animal stick liquor; 4, sulphuric acid; 5, soybean oil; 6, caustic soda; 7, inoculum tank; 8, air filter; 9, sterilizer;
10, cooler. 0'1
......
162 T. Kutsal & M. T. Ozbas
9 (BIO)-ASSA Y METHODS
in body fluids and tissues suitable for clinical and nutritional surveys is based
on the ciliate protozoan Tetrahymena pyriformis (Pearson, 1967; Kirk &
Othmer, 1968).
10 BIOLOGICAL PROPERTIES
The principal forms of riboflavin which exist in living cells are riboflavin
5'phosphate (FMN) and flavin adenine dinucleotide (FAD). Both these
phosphates are protein bound and 60-90% of the total riboflavin in natural
products is present in the form of the dinucleotide. Riboflavin performs its
biological functions in a number of different enzyme systems. Two derivatives
of riboflavin, FMN and FAD, serve as prosthetic groups and combine with
specific protein enzymes which catalyze oxidation-reduction reactions in cells
(Pearson, 1967). The good sources of riboflavin are milk, egg, liver, heart,
kidney, green, leafy vegetables, apricot, tomato, beef and poultry meats
(Wagner-Jauregg, 1972).
11 CHEMICAL SYNTHESIS
Riboflavin is essential for the growth and normal health of animals as well as of
man. A lack of riboflavin in human or animal diets causes the formation of
certain oral, cutaneous and corneal lesions. Vitamin B2 is widely used in the
food enrichment, pharmaceutical and feed supplement industries (Kirk &
Othmer,·1968).
USP riboflavin for therapeutic purposes is administered orally in tablet form
or by injection as a sterile aqueous solution, with niacinamide or other
Microbial Production of Vitamin 8 2 165
solubilizing agent added. All of the vitamin B complex and most of the
multivitamin preparations also contain riboflavin (Kirk & Othmer, 1968).
Sterile, supersaturated solutions of riboflavin in normal saline have been
employed for intravenous administration (Wagner-Jauregg, 1972).
Some of the cereal products such as flour and bread are enriched with iron,
thiamine and niacin as well as riboflavin to compensate for the loss of these
nutrients that occurs in the processing of wheat (Kirk & Othmer, 1968).
Riboflavin can be synthesized by most higher plants, yeasts, and lower fungi,
and bacteria. The tissues of higher animals are unable to synthesize this
vitamin. For the enrichment of poultry and livestock feeds, riboflavin is usually
added at concentrations of 2-8 g/ton, depending on the species, age and
purpose. The supplement of riboflavin improves health, growth, tissue repair,
and reproduction of the animals and also has an economic importance in
poultry raising and egg production (Kirk & Othmer, 1968; Wagner-Jauregg,
1972).
The present world consumption of riboflavin is approximately 1·25 million
kg for human and animal use combined (Lago & Kaplan, 1981).
Commercial fermentation processes for production of riboflavin or riboflavin
concentrates are relatively recent, having been developed in the past 40 years.
In 1946 processes using the ascomycete A. gossypii were started. Manufac-
turers using the microbiological process are Merck and Co., Inc. (USA) and
BASF (FRG). Companies manufacturing riboflavin by chemical synthesis
include Hoffmann-LaRoche Inc. (USA and Switzerland), Takeda Chemical
Industries Ltd (Japan), Pfizer, Inc. (USA), and E. Merck (FRG) (Perlman,
1979).
REFERENCES
Ainsworth, G. C. & Sussman, A. S. (1965). The Fungi, Academic Press, New York,
pp. 37,492.
Brit. Pat. 621,468 (April 11 1949) (to Commercial Solvents Corp.)
Demain, A. L. (1972). Riboflavin oversynthesis, In Annual Reviews of Microbiology
Vol. 26, ed. C. E. Clyton, S. Raffel & M. P. Starr. George Benta Inc., USA, pp.
369-88.
Goodwin, T. W. (1959). Production and biosynthesis of riboflavin in micro-organisms.
In Progress In Industrial Microbiology, ed. D. J. D. Hockenhull. Heywood &
Company, London, pp. 139-177.
Goodwin, T. W. (1963a). Vitamins. In Biochemistry of Industrial Micro-organisms, ed.
C. Rainbow & A. H. Rose. Academic Press, London, pp. 151-8.
Goodwin, T. W. (1963b). Riboflavin and related compounds. In The Biosynthesis of
Vitamins and Related Compounds, ed. T. W. Goodwin. Academic Press, London,
pp.24-35.
Gutcho, S. J. (1973). Chemicals by Fermentation, Noyes Data Corporation, Chemical
Technology Review No. 19, New Jersey, pp. 323-25.
Heimann, W. (1980). Fundamentals of Food Chemistry, Ellis Horwood, Chichester, pp.
212-14.
166 T. Kutsal & M. T. Ozbas
1 HISTORY
Table 1
Occurrence of o-Ribose in Nature
Table l--contd.
Roberts et al., 1963), and in the cell walls of Proteus mirabilis (Gmeiner, 1977;
Gmeiner et al., 1977) and Vibrio parahaemolyticus (Miyano et al., 1983) as a
component of lipopolysaccharides and of various Gram-positive bacteria as a
component of ribitol teichoic acid (Ward, 1981). It is also a component of
riboflavin (Karrer et al., 1935a; Kuhn et al., 1935a; von Euler et aI., 1935).
D-Ribonic acid was the oxidation derivative of D-ribose produced by pseudo-
monads (Foster, 1944; Lockwood & Nelson, 1946; Weimberg, 1961). Another
oxidation product, D-riburonic acid, was found in the cell wall polysaccharides
of Rhizobium meliloti (Amemura et al., 1981). 5-Methylthioribose is a com-
ponent of 5'-methylthioadenosine (Mandel & Dunham 1912; Suzuki et at.,
1914, 1924) which has an important role in the metabolism of adenosylthi-
omethionine (Schlenk & Smith, 1953). Glutamyl D-ribose 5-phosphate was
found as an abnormal storage substance in the brain and kidney of a patient
suffering from renal failure and neurologic deterioration (Williams et al.,
1986). D-Ribose 5-phosphate is an intermediate in the pentose phosphate
pathway (Wood, 1985) which plays an important role in the production of
D-ribose by the tkt mutants of Bacillus species as will be described later.
D-Ribose-l-phosphate and 5-phosphoribosyl I-pyrophosphate take part in
purine and pyrimidine nucleotide biosynthesis (Hartmann, 1970). o-Ribose
1,5-bisphosphate was found in human erythrocytes (Vanderheiden, 1970).
The chemical and physical properties of o-ribose and its derivatives have
already been described in detail by Jeanlotz & Fletcher (1951), Overend &
172 K. Sasajima & M. Yoneda
"O~C"
" " " 0"
" "
Fig. 2. Formula of o-ribose.
Stacey (1955) and Windholz et al. (1983). The formulae of o-ribose is shown in
Fig. 2. The chemical and physical properties of o-ribose and its derivatives are
as follows.
o-Ribose C5 H lO0 5 ; mol. wt 150·13, C 40·00%, H 6·71; 053·29%. Melting
point of crystalline o-ribose is 87°C (Levene & Jacobs, 1909). Crystals are
plates or needles. o-Ribose shows mutarotation, final [«]b - 23'7° (4,5% in
water) (Phelps et al., 1934). Soluble in water, slightly insoluble in alcohol.
Anhydroribose C5 H s0 4; m.p. 229-230°C (Bredereck et al., 1940).
«-Aniline-N-o-ribopyranoside; m.p. 125-127°C, [«m + 63·4°~ + 48·6°
(C = 1·0% in pyridine) (Berger & Lee, 1944; Berger et al., 1944).
Benzylphenylhydrazone ClsH22N204; chrysanthemum-like crystals, m.p.
128°C (Sasajima & Yoneda, 1971).
p-Bromophenylhydrazone; m.p. 166-167°C (van Ekenstein & Blanksma,
1913; Steiger, 1936)
Methyl-o-riboside; crystals from ethyl acetate, m.p. 83-84° C, [«]~ - 113·6°
(p = 3) (Windholz et al., 1983).
Phenylosazone C17H2oN'403; yellow needles from pyridine + water, m.p.
163-164°C.
The infrared absorption of o-ribose from 8 to 15 microns was measured by
Kuhn (1950). A polarographic study of o-ribose was described by Cantor &
Peniston (1940); and crystallographic properties were reported by Keenan
(1926).
O-Ribose 1-PhosPhate]
r--_ _ _ _ L__~> o-Xylulose 5-phosphate o-Ribose 5-phosphate ----+ [
5-Phosphoribosyl
1-pyrophosphate
a~
0-
1 §.:
-- --"-- -- ---.- --- I
Glyceraldehyde 3-phosphate Sedoheptulose 7-phosphate
1
Transketolase
1 ~
§o
Purine nucleotide
~
Pyrimidine nucleotide ~
~
s:
o-Fructose 6-phosphate o-Erythrose 4-phosphate A ~
1 ~
Shikimic acid ~
Lipopolysaccharide
1
L-Tryptophan
L-Tyrosine
TCA cycle
L-Phenylalanine
CoO
......
Vitamin K -..J
....,
Folic acid
Fig. 3. Pentose phosphate pathway and related pathwayso
174 K. Sasajima & M. Yoneda
50-RIBOSE-PRODUCING MICRO-ORGANISMS
The first o-ribose accumulation was found in the culture medium of Penicillium
brevi-compactum (Simonart & Godin, 1951) during studies on the biosynthesis
of the components of benzene derivatives (Godin, 1953a,b). Since o-ribose
was detected by paper chromatography, the amount of accumulated o-ribose
was unknown. About 10 years later, an unidentified bacterium was reported to
accumulate both o-ribose and D-ribose 5-phosphate (Suzuki et al., 1963).
Pseudomonas reptilivola was also reported to accumulate o-ribose in a culture
medium at the rate of 0·12 mg of o-ribose per ml (Saito & Sugiyama, 1966).
All the micro-organisms used in these studies were wild-type and not much
o-ribose was accumulated. However, the amount of o-ribose produced by the
tkt mutants of Bacillus species was remarkable (Sasajima & Yoneda, 1971) and
the procedure, with subsequent improvement, has been used to commercially
produce o-ribose for riboflavin synthesis for 15 years. A detailed description of
the isolation and improvement of o-ribose-producing tkt mutants of Bacillus
species is given in the following section.
steps at which the yield of o-ribose production was remarkably improved were
the following; (1) a mutation of the regulation of D-glucose dehydrogenase
(EC 1.1.1.47) synthesis (Yokota et al., 1979; Yokota & Sasajima, 1981) and
(2) an asporogenous mutation (Sasajima et al., 1972, 1985b; Sasajima, 1976;
Sasajima & Yoneda, 1984)
D-Glucose dehydrogenase in Bacillus species, is known to be synthesized at
stage III of sporulation in parallel with morphological changes during spore
formation (Warren, 1968), exists only in spores (Fujita et al., 1977) and has the
role of supplying energy during germination (Otani et al., 1986). The initiation
of sporulation is also known to occur only after the carbon source is depleted.
In the improved mutant strain, the enzyme was found in vegetative cells while
no morphological changes concerning sporulation occurred (Yokota et al.,
1979; Yokota & Sasajima, 1981). It was presumed that o-glucose was
converted to o-ribose in the mutant strain through two metabolic pathways,
the non-phosphorylated pathway (from o-glucose to the pentose phosphates
via D-gluconate) and the phosphorylated pathway (the oxidative branch of the
pentose phosphate pathway) (Fig. 4). The gluconate pathway, probably,
contributed to the high D-ribose productivity of the improved strain. However,
there was also the possibility of o-gluconate as well as o-ribose accumulation.
The ratio of o-ribose and o-gluconate production depended on the partial
pressure of oxygen during fermentations, i.e. oxygen deficiency led to
D-gluconate rather than o-ribose accumulation (Sasajima & Yoneda, 1984).
D-Ribose accumulation by the mutant strain was about 60 mg/ml (Sasajima et
al., 1972, 1985b; Sasajima, 1976; Sasajima & Yoneda, 1984).
The second-step asporogenous mutation improved the capability of o-ribose
production up to 70 mg/ml (Sasajima et al., 1972, 1985b; Sasajima, 1976;
Sasajima & Yoneda, 1984). In this case, the mechanism for the improvement
was not clear. It might be a mutation by which carbohydrate metabolism is
changed in the direction of higher-level o-ribose accumulation. The high-yield
D-Glucose - - - - - - - o-Gluconate
1
o-Glucose-6-phosphate -
1
6-Phospho-o-gluconate
1
0- Ribulose-5-phosphate
1
0- Ribose-5-phosphate
1
D-Ribose
Fig. 4. Pathways of o-ribose formation from o-glucose in high-yield mutant strains.
Microbial Production of D-Ribose 179
mutant strain has been used for 15 years to produce o-ribose for riboflavin
synthesis on an industrial scale.
6 FERMENTATION
6.1 Inoculation
In the case of Penicillium brevi-compactum (Simonart & Godin, 1951), an
unidentified bacterium (Suzuki et al., 1963), and Pseudomonas reptilivora
Microbial Production of D-Ribose 181
Table 2
Inoculum Medium of Bacillus Species
(Saito & Sugiyama, 1966), the cells grown on bouillon agar slants were directly
inoculated into the fermentation medium. In the case of the tkt mutants of
Bacillus species (Sasajima & Yoneda, 1971), the inoculum culture incubated at
37°C was used. The inoculum medium composition is shown in Table 2.
Adenosine and guanosine can be excluded from the medium in the case of
high-producing mutant strains whose requirements were compensated by
further reversion mutations.
The composition of the fermentation media used for o-ribose production are
shown in Tables 3, 4, 5 and 6 (Simonart & Godin, 1951; Suzuki et ai., 1963;
Saito & Sugiyama, 1966; Sasajima and Yoneda, 1971).
Suzuki et ai. (1963) described that the Mn2+ ion stimulated o-ribose
accumulation while o-ribose 5-phosphate was accumulated more in the absence
of Mn2+ ion in the case of an unidentified bacterium. Fe2+ and Zn 2+ ions are
also effective in stimulating o-ribose accumulation.
In the case of Pseudomonas reptilivora, o-ribose accumulation from various
carbon sources other than o-glucose was observed (Saito & Sugiyama, 1966).
o-Mannose, o-fructose and o-gluconic acid were found to be as good substrates
as o-glucose.
Table 3
Fermentation Medium of Penicillium brevi-
compactum
Table 4
Fermentation Medium of an Unidentified
Bacterium
Constituent Concentration (%)
D-Glucose 10·0
KHZP04 1·0
K zHP04 1·0
MgS0 47HzO 1·0
CaClz2H2 0 0·01
Urea 0·6
Yeast extract 1·0
In the case of Bacillus species, it was also found that various carbon sources
were useful for D-ribose accumulation as shown in Table 7 (Sasajima &
Yoneda, 1971). D-Glucose, D-mannose, sorbitol, D-mannitol, maltose, and
lactose were good substrates. Various kinds of natural nitrogen sources as well
as dried yeast are useful (Table 8) (Sasajima & Yoneda, 1971). Corn steep
liquor-based medium was used for a large-scale production (Asai et ai., 1978).
It was necessary to complement aromatic amino acids to the medium in the
case where corn steep liquor was used as nitrogen source (Asai et ai., 1978).
Table 5
Fermentation Medium of Pseudomonas
reptilivora
Table 6
Fermentation Medium of Bacillus Species
Constituent Concentration (%)
D-Glucose 12·5; 15·0
(NH4)2S04 1·5
CaHP04 0·5
Ca3 (P04 )z 0·5
CaC03 1·0
Dried yeast 2·5
(pH 7·0)
Microbial Production of D-Ribose 183
Table 7
D-Ribose Formation from Various Carbon Sources
Table 8
Efficiency of Various Nitrogen Sources for D-Ribose
Formation
7
pH 6 pH
5
"'C
C
..,~
"'C ;f. 10
QI ~
~ QI
-
::> '"
0
E v D-Ribose
::>2
VC)
:;: I
QlO
.8"iii
.- ::>
5
a::"'C
I .;;;
0::'
L -_ _ _ _L...'-_ .. _........--.----1r-
o 50 100
Time (hr)
shown in Fig. 5 (Sasajima et al., 1972, 1985b; Sasajima, 1976; Sasajima &
Yoneda, 1984).
As already described in the section on 'Producing micro-organisms', the
improved strains have the ability to produce n-gluconate as well as n-ribose.
n-Ribose is made from n-glucose through two pathways: one is a non-
phosphorylated pathway via n-gluconate and the other is a phosphorylated
pathway via n-glucose 6-phosphate (Fig. 4). Both n-gluconate and n-glucose
6-phosphate are converted to 6-phospho-n-gluconate, the common intermedi-
ate of both pathways, and finally converted to n-ribose.
Appropriate culture conditions are required for the smooth conversion of
n-gluconate to 6-phospho-n-gluconate. Otherwise, n-gluconate accumulates in
the culture medium resulting in a reduced production of n-ribose (Asai et al.,
1978; Sasajima & Yoneda, 1984). Effective factors are the pH of media, an
oxygen concentration in the medium enough for three oxidative reactions in
the pathway from n-glucose to n-ribose (Fig. 4) (Sasajima & Yoneda, 1984),
and the temperature of the culture (Asai et al., 1978). The maximum yield of
n-ribose was noted at 36·6°C when exponential growth cells were used as the
inoculum (Asai et al., 1978).
Ekenstein & Blanksma, 1913; Steiger, 1936) or the arylamide riboside (Berger
& Lee, 1944; Berger et al., 1944). After subsequent purification, o-ribose was
recovered by hydrolysis (Jeanlotz & Fletcher, 1951; Overend & Stacey, 1955).
In the case of preparing o-ribose by enzymatic hydrolysis of 5-amino-4-
imidazole-carboxamide-riboside with a nucleosidase, o-ribose was separated
and purified from the other product, 5-amino-4-imidazole-carboxamide, by
column chromatography with ion-exchange resin Dowex 50 (H+) (Sano et al.,
1977b). Finally, o-ribose was obtained as the lyophilized form.
Saito & Sugiyama (1966) purified o-ribose from the culture medium of
Pseudomonas reptilivora by first separating microbial cells by centrifugation
and subsequent column chromatographies with chromatopile and strong-base
anion resin (borate form). Finally, they obtained o-ribose in crystalline form
from the syrup.
Sasajima & Yoneda (1971) purified and isolated o-ribose from the culture
medium of Bacillus species tkt mutant by removing microbial cells, column
chromatography with carbon, yeast treatment to assimilate the residual
substrate o-glucose and excluding metal and anionic ions with ion exchange
resins, amberlite IR120 and amberlite IRA400. o-Ribose was crystallized by
adding ethanol to the syrupy solution.
Hough et al. (1948, 1949) described the separation of o-ribose from
coexisting sugars by partition chromatography with a powdered cellulose
column.
9 CHEMICAL SYNTHESIS
CHO
I
sodium
HCOH
amalgam .. I
HCOH
I
HCOH
I
CH20H
o-ribono-y-Iactone o-ribose
Fig. 6. Chemical synthesis of D-ribose according to the method of van Ekenstein &
B1anksma (1913).
Hr[l
--
Zn dust
CH
I
acetic aCid) HCOAc 0
H10
CH
AC
2
I
o-arabinose triacetyl-J3-o- diacetyl-o-arabinal
arabino-pyranosyl
bromide
Hr[l
CH
I H 0
HCO
CHO
I
HCOH
I
...:.pe_r_be_nz_oi_c_ac_id+)
HCOR
I
H10H
CH 2
I
HCOH
I
CH 20H
o-arabinal o-ribose
Fig. 7. Chemical synthesis of D-ribose according to the method of Gehrke & Aichner
(1927).
Microbial Production of D-Ribose 187
CH2 0H
HtOH
HOtH
reduction I
HC-O mel"period"le)
II ""
HCOH/CHC6H s
.
H 2 C-O
4,6-benzylidene-o-glucose 4,6-benzylidene-o-glucitol 2,4-benzylidene-o-erythrose
-
CH
HC-O/' "---CH 3 HC-O/' "CH 3
I I .
HOCH - CH 3S0 2 0CH
I I
HCOH HCOH
I I
HC-O....... y,,/CH 3 HC-O, £H3
I /CH I "Cli .
H 2 C-O "---CH 3 H 2 C-O/ "CH 3
1,2: 5,6-diisopropylidene-o-glucose 3-0-mesyl-l,2: 5,6-diisopropylidene-o-glucose
CHO CHO CHO
I I I
-
HCOH CH 3S0 2 0CH HCOH
I . I I
CH 3S02 0CH HCOH - HCOH
. I I J
HCOH HCOCHO HCOH
I I I
HCOH CH 20H CH 2 0H
I
CH 20H
3-0-mesyl-o-glucose 2-0-mesyl-4-0-formyl-D-arabinose o-ribose
preparation and was used to prepare o-ribose for the chemical synthesis of
riboflavin (Karrer et al., 1935b; Kuhn et al., 1935b). The process includes
arabinal as an intermediate as shown in Fig. 7.
The formation of o-ribose from o-glucose was described by Sowden (1950)
and Smith (1955) as shown in Figs 8 and 9.
Stroh et al. (1964) compared the above four methods and concluded that the
best yield was obtained by the method of Smith (1955). The overall yield was
24%.
o-Ribose has long been prepared for the chemical synthesis of riboflavin from
o-glucose by one of the chemical processes described above. However, a
microbial production process was developed in the early 1970s (Sasajima et al.,
1972, 1985b; Sasajima, 1976; Sasajima & Yoneda, 1984). The procedure, using
the tkt mutant of Bacillus species is economical and prevents environmental
pollution by avoiding the use of mercury, which is essential for the electrolytic
reduction required in the chemical process (Harris et al., 1972). The industrial
microbial production of o-ribose for the chemical synthesis of riboflavin started
in 1973; it has given substantial results for about 15 years. Some companies in
the world have been using this process to produce o-ribose for the synthesis of
riboflavin, which is used not only for pharmaceuticals but also for animal feed
additives worldwide in large quantities.
o-Ribose itself has been also used to treat some cases of heart diseases. It
was suggested that administered o-ribose bypasses the pentose phosphate
pathway and increases the size of the phosphoribosylpyrophosphate pool
resulting in stimulation of nucleotide synthesis, maintenance of adenine
nucleotide levels, and protection of the myocardium in various heart diseases
(Zimmer et al., 1984). Exercise-induced muscle pain and stiffness due to
myoadenylate deaminase deficiency was also successfully treated with o-ribose
(ZOllner et al., 1986).
D-Ribose is a natural component of biologically important substances such as
nucleic acids and coenzymes. It has been also found in many nucleoside
antibiotics. Nucleoside analogs have been developed as antiviral agents. It is
hoped that o-ribose can be used in the chemical synthesis of drugs to treat viral
and bacterial infectious diseases in the future.
REFERENCES
Abou-Sabe, M., Pilla, J., Hazuda, D. & Ninfa, A. (1982). Evolution of the D-ribose
operon of Escherichia coli B/r. J. Bacteriol., 150, 762-9.
Adler, J. (1969). Chemoreceptors in bacteria. Science, 166, 1588-97.
Aksamit, R. & Koshland, D. Jr (1972). A ribose binding protein of Salmonella
typhimurium. Biochem. biophys. Res. Commun., 48, 1348-53.
Aksamit, R. R. & Koshland, D. E. Jr (1974). Identification of the ribose binding
protein as the receptor for ribose chemotaxis in Salmonella typhimurium.
Biochemistry, 13, 4473-8.
Alphen, W. van, Lugtenberg, B. & Berendsen, W. (1976). Heptose-deficient mutants
of Escherichia coli K12 deficient in up to three major outer membrane proteins.
Molec. Gen. Genet., 147,263-9.
Amemura, A., Hisamatsu, M., Ghai, S. K. & Harada, T. (1981). Structural studies on
a new polysaccharide, containing D-riburonic acid, from Rhizobium me/iloti IFO
13336. Carbohydr. Res., 91,59-65.
Ames, G. F.-L., Spudich, E. N. & Nikaido, H. (1974). Protein composition of the
outer membrane of Salmonella typhimurium: effect of lipopolysaccharide mutations.
J. Bacteriol., 117,406-16.
Anderson, A. & Cooper, R. A. (1970). Biochemical and genetical studies on ribose
catabolism in Escherichia coli. J. Gen. Microbiol., 62,335-9.
Asai, T., Doi, M., Kono, T. & Fukuda, H. (1978). Kinetic study on the production of
D-ribose by Bacillus sp. J. Ferment. Technol., 56,91-5.
Austin, W. C. & Humoller, F. L. (1934). The preparation of I-ribose. J. Am. Chem.
Soc., 56, 1152-3.
Avery, O. T., MacLeod, C. M. & McCarty, M. (1944). Studies on the chemical nature
of the substance inducing transformation of pneumococcal types. J. expo Med., 79,
137-58.
Baddiley, J., Thain, E. M., Novelli, G. D. & Lipmann, F. (1953). Structure of
coenzyme A. Nature, Lond., 171, 76.
Baker, B. R. & Schaub, R. E. (1953). Achromycin. Synthetic studies. III. Synthesis of
3-amino-D-ribose, a hydrolytic product. J. Am. Chem. Soc., 75, 3864-5.
Barnett, J. A. (1976). The utilization of sugars by yeasts. Adv. Carbohydr. Chem.
Biochem., 32, 125-234.
Bayer, M. E., Koplow, J. & Goldfine, H. (1975). Alterations in envelope structure of
heptose-deficient mutants of Escherichia coli as revealed by freeze-etching. Proc.
natn. Acad. Sci. U.S.A., 72,5145-9.
Beaven, G. R., Holiday, E. R., Johnson, E. A., Ellis, B., Mamalis, P., Petrow, V. &
Sturgeon, B. (1949). The chemistry of anti-pernicious anaemia factors. Part III.
5: 6-Disubstituted benziminazoles as products of acid hydrolysis of vitamin B 12 • J.
Pharm. Pharmac., 1,957-70.
Bentley, H. R., Cunningham, K. G. & Spring, F. S. (1951). Cordycepin, a metabolic
product from cultures of Cordyceps militaris (Linn.) Link. Part II. The structure of
cordycepin. J. Chem. Soc., 1951,2301-5.
Berezovskii, V. M. & Rodionova, E. P. (1953). Transformations and synthesis of
carbohydrates. VII. Synthesis of D-ribose. Sbornik Statel Obshchel Khim., 2,
939-43; Chem. Abstr., 49,6838 (1955).
Berger, L. & Lee, J. (1944). Arylamine-N-glycosides. Part II. Arylarnine-N-pentosides
and complex salt formation studies. J. Org. Chem., 11, 84-90.
Berger, L., Solmssen, U. V., Leonard, F., Wenis, E. & Lee, J. (1944). Arylamine-N-
glycosides. Part III. Hydrolysis of arylamine-N-pentosides and the preparation of
crystalline D-ribose. J. Org. Chern., 11,91-4.
Bernstein, I. A. (1953). Fermentation of ribose-C4 by Lactobacillus pentosus. J. BioI.
Chem., 205, 309-16.
Bial, M. (1902). Die Diagnose der Pentosurie. Deut. med. Wochschr., 28,
253-4.
190 K. Sasajimo & M. Yoneda
Bial, M. (1903). Ueber die Diagnose der Pentosurie mit dem von mir angegebenen
Reagens. Deut. med. Wochschr., 29,477-8.
Borman, A. H., Jong, E. W. de, Thierry, R., Westbroek, P., Bosch, L., Gruter, M. &
Kamerling, J. P. (1987). Coccolith-associated polysaccharides from cells of
Emiliania huxleyi (Haptophyceae). J. Phycol., 23, 118-23.
Bredereck, H. & Rothe, G. (1938). Nucleinsauren, VI. Mitteil: Darstellung der
Nucleoside durch enzymatische Hydrolyse der Hefenucleinsaure; zugleich ein
Beitrag zur Darstellung der d-Ribose. Ber., 71,408-11.
Bredereck, H., Kothnig, M. & Berger, E. (1940). Uber die d-Ribose (Darstellung
einer krystallisierten Anhydroribose). Ber., 73,956-62.
Brink, N. G. & Folkers, K. (1949). Vitamin B 12 . VI. 5 ,6-Dimethylbenzimidazole , a
degradation product of vitamin B12 . J. Am. Chem. Soc., 71, 295l.
Brink, N. G. & Folkers, K. (1950). Vitamin B 12 . X. 5 ,6-Dimethylbenzimidazole , a
degradation product of vitamin B 12 . J. Am. Chem. Soc., 72, 4442-3.
Brink, N. G., Holly, F. W., Shunk, C. H., Peel, E. W., Cahill, J. J. & Folkers, K.
(1950). Vitamin B 12 . IX. 1-a-Ribofuranosido-5,6-dimethylbenzimidazole, a de-
gradation product of vitamin B 12 . J. Am. Chem. Soc., 72, 1866.
Buchanan, J. G., Johnson, A. W., Mills, J. A. & Todd, A. R (1950a). The isolation of
a phosphorus-containing degradation product from vitamin B 12c • Chem. Ind., 1950,
426.
Buchanan, J. G., Johnson, A. W., Mills, J. A. & Todd, A. R. (1950b). Chemistry of
the vitamin BI2 group. Part I. Acid hydrolysis studies. Isolation of a phosphorus-
containing degradation product. J. Chem. Soc., 1950, 2845-55.
Caldwell, D. R & Newman, K. (1986). Pentose metabolism by Bacteroides ruminicola
subsp. brevis strain BA. Curro Microbio/., 14, 149-55.
Cantor, S. M. & Peniston, Q. P. (1940). The reduction of aldoses at the dropping
mercury cathode: estimation of the aldehyde structure in aqueous solutions. J. Am.
Chem. Soc. 62, 2113-2l.
Caputto, R, Leloir, L. F., Cardini, C. E. & Paladini, A. C. (1950). Isolation of the
coenzyme of the galactose phosphate-glucose phosphate transformation. J. Bioi.
Chem., 184, 333-50.
Chargaff, E. & Lipshitz, R (1953). Composition of mammalian desoxyribonucleic
acids. J. Am. Chem. Soc., 75, 3658-6l.
Chargaff, E., Vischer, E., Doniger, R, Green, C. & Misani, F. (1949). The
composition of the desoxypentose nucleic acids of thymus and spleen. J. Bioi.
Chem., 177,405-16.
Cherbuliez, E. & Bernhard, K. (1932a). Recherches sur la graine de croton. I. Sur Ie
crotonoside (2-oxy-6-amino-purine-D-riboside). Helv. Chim. Acta, 1S,464-7l.
Cherbuliez, E. & Bernhard, K. (1932b). Remarques sur Ie crotonoside. Helv. Chim.
Acta, IS, 978-80.
Curtis, S. J. (1974). Mechanism of energy coupling for transport of D-ribose in
Escherichia coli. J. Bacterio/., 120,295-303.
David, J. & Wiesmeyer, H. (1970). Regulation of ribose metabolism in Escherichia
coli. I. The ribose catabolic pathway. Biochim. biophys. Acta, 208,45-55.
Davis, A. R, Newton, E. B. & Benedict, S. R (1922). The combined uric acid in beef
blood. J. Bioi. Chem., 54,595-9.
Dickens, F. (1938a). Oxidation of phosphohexonate and pentose phosphoric acids by
yeast enzymes. I. Oxidation of phosphohexonate. II. Oxidation of pentose
phosphoric acids. Biochem. J., 32, 1626-44.
Dickens, F. (1938b). Yeast fermentation of pentose phosphoric acids. Biochem. J., 32,
1645-53.
Dills, S. S., Apperson, A., Schmidt, M. R & Saier, M. H. Jr (1980). Carbohydrate
transport in bacteria. Microbiol. Rev., 44,385-418.
Microbial Production of D-Ribose 191
Dilworth, M. J., Arwas, R., McKay, I. A., Saroso, S. & Glenn, A. R. (1986). Pentose
metabolism in Rhizobium leguninosarum MNF300 and in cowpea Rhizobium
NGR234. J. Gen. Microbiol., 132,2733-42.
Dische, Z. (1962a). Color reaction of ketoses with carbazole and sulfuric acid. In
Methods in Carbohydrate Chemistry, Vol. I, ed. R. L. Whistler, M. L. Wolfrom, J.
N. BeMiller & F. Shafizadeh. Academic Press, New York, London, pp. 481-2.
Dische, Z. (1962b). Color reactions of pentoses. In Methods in Carbohydrate
Chemistry, Vol. I, ed. R. L. Whistler, M. L. Wolfrom, J. N. BeMiller & F.
Shafizadeh. Academic Press, New York, London, pp. 484-8.
Doly, J. & Mandel, P. (1967). Mise en evidence de la biosyntbese in vivo d'un
polymere compose, Ie polyadenosine diphosphoribose dans les noyaux de foie de
poulet. C. R. hebd. Slanc. Acad. Sci., Paris, Ser. D., 264,2687-90.
Duncan, M. J. (1981). Properties of tn5-induced carbohydrate mutants in Rhizobium
meliloti. J. Gen. Microbiol., 122,62-7.
Eidels, L. & Osborn, M. J. (1971). Lipopolysaccharide and aldoheptose biosynthesis in
transketolase mutants of Salmonella typhimurium. Proc. natn. Acad. Sci. U.S.A.,
68,1673-7.
Ekenstein, W. A. van & Blanksma, J. J. (1913). Over o-ribose. Chem. Weekblad., 10,
664.
Euler, H. von & Vestin, R. (1935). Zur Kenntnis der Wirkungen der Co-Zymase. Z.
physiol. Chem., 237, 1-5.
Euler, H. von & Schlenk, F. (1937). Co-Zymase. Z. physiol. Chem., 246,64-82.
Euler, H. von, Karrer, R., Malmberg, M., Schopp, K., Benz, F., Becker, B. & Frei, P.
(1935). Synthese des Lactoflavins (Vitamin B2 ) und anderer Flavine. Helv. Chim.
Acta, 18,522-35.
Euler, H. von, Karrer, P. & Usteri, E. (1942). Der Zucker der Cozymase. Helv. Chim.
Acta, 25, 323-5.
Foster, J. W. (1944). Microbiological aspects of riboflavin. III. Oxidation studies with
Pseudomonas riboflavina. J. Bacteriol., 48,97-111.
Fujita, Y., Ramaley, R. & Freese, E. (1977). Location and properties of glucose
dehydrogenase in sporulating cells and spores of Bacillus subtilis. J. Bacteriol., 132,
282-93.
Galloway, D. R. & Furlong, C. E. (1977). The role of ribose-binding protein in
transport and chemotaxis in Escherichia coli K12. Archs. Biochem. Biophys., 184,
496-504.
Gehrke, M. & Aichner, F. X. (1927). Uber das Arabinal. Ber., 60,918-22.
Gerber, N. N. & Lechevalier, H. A. (1962). 3'-Amino-3'-deoxyadenosine, an
antitumor agent from Helminthosporium sp. J. Org. Chem., 27, 1731-2.
Gmeiner, J. (1977). The ribitol-phosphate-containing lipopolysaccharide from Proteus
mirabilis, strain D52. Investigations on the structure of O-specific chains. Eur. J.
Biochem., 74, 171-BO.
Gmeiner, J., Mayer, H., Fromme, I., Kotelko, K. & Zych, K. (1977). Ribitol-
containing lipopolysaccharides from Proteus mirabilis and their serological relation-
ship. Eur. J. Bioc;hem., 72, 35-40.
Godin, P. (1953a). Etude du metabolisme ternaire de Penicillium brevi-compactum. I.
Analyse chromatographique du milieu de culture. Biochim. biophys. Acta, 11,
114-18.
Godin, P. (1953b). Etude du metabolisme ternaire de Penicillium brevi-compactum. II.
Analyse chromatographique de la solution glucosee substitutee au milieu de culture
initial. Biochim. biophys. Acta, 11, 119-21.
Gomez, L. D. & Wallace, J. W. (1986). Flavonoids of Phlebodium. Biochem. Syst.
Ecol., 14,407-8.
Greenberg, D. M. (1969). Biosynthesis of amino acids and related compounds. In
192 K. Sasajima & M. Yoneda
Metabolic Pathways, Vol. III, ed. D. M. Greenberg. Academic Press, New York,
London, pp. 237-315.
Groarke, J. M., Mahoney, W. C., Hope, J. N., Furlong, C. E., Robb, F. T., Zalkin,
H. & Hermodson, M. A. (1983). The amino acid sequence of o-ribose-binding
protein from Escherichia coli K12. I. Bioi. Chem., 258, 12952-6.
Guarino, A. J. & Kredich, N. M. (1963). Isolation and identification of 3'-amino-3'-
deoxyadenosine from Cordyceps militaris. Biochim. biophys. Acta, 68,317-19.
Hackman, R. H. & Trikojus, V. M. (1952). The composition of the honeydew excreted
by Australian coccids of the genus Ceroplastes. Biochem. I., 51, 653-6.
Harris, R. S., Wagner-Jauregg, T., Horwitt, M. K. & Witting, L. A. (1972).
Riboflavin. In The Vitamins, Vol. V (2nd edn), ed. W. H. Sebrell & R. S. Harris.
Academic Press, New York, London, pp. 1-96.
Hartman, S. C. (1970). Purines and pyrimidines. In Metabolic Pathways, Vol. IV, ed.
D. M. Greenberg. Academic Press, New York, London, pp. 1-68.
Hazelbauer, G. L. & Adler, J.' (1971). Role of the galactose binding protein in
chemotaxis of Escherichia coli toward galactose. Nature New Bioi., 230, 101-4.
Heath, E. C., Hurwitz, J., Horecker, B. L. & Ginsburg, A. (1958). Pentose
fermentation by Lactobacillus plantarum. I. The cleavage of xylulose 5-phosphate
by phosphoketolase. I. Bioi. Chem., 231, 1009-29.
Hershey, A. D. & Chase, M. (1952). Independent functions of viral protein and nucleic
acid in growth of bacteriophage. I. Gen. Physiol., 36,39-56.
Hockett, R. C. & Hudson, C. S. (1934). Improvements in the preparation of
o-arabinose from calcium gluconate. I. Am. Chem. Soc., 56, 1632-3.
Hoeksema, H. & Baczynskyj, L. (1976). A new metabolite from Streptomyces
hygroscopicus, II. Identification as 1-deoxy-o-threo-pentulose. I. Antibiot., XXIX,
688-91.
Hoffman, J., Lindberg, B., Hofstad, T. & Lygre, H. (1977). Structural studies of the
polysaccharide antigen of Eubacterium saburreum, strain L.452. Carbohydr. Res.,
58,439-42.
Hofstad, T. & Lygre, H. (1977). Composition and antigenic properties of a surface
polysaccharide isolated from Eubacterium saburreum, strain L452. Acta Pathol.
Microbiol. Immunol. Scand. Section B, 85, 14-17.
Holiday, E. R. & Petrow, V. (1949). Vitamin B12 as a 5:6-dimethylbenziminazole
derivative. I. Pharm. Pharmac., 1,734-5.
Hough, L., Jones, J. K. N. & Wadman, W. H. (1948). Application of paper partition
chromatography to the separation of the sugars and their methylated derivatives on
a column of powdered cellulose. Nature, Lond., 162, 448.
Hough, L., Jones, J. K. N. & Wadman, W. H. (1949). Quantitative analysis of
mixtures of sugars by the method of partition chromatography. Part IV. The
separation of the sugars and their methylated derivatives on columns of cellulose. I.
Chem. Soc., 1949,2511-16.
!ida, A., Harayama, S., lino, T. & Hazelbauer, G. L. (1984). Molecular cloning and
characterization of genes required for ribose transport and utilization in Escherichia
coli K-12. I. Bacteriol., 158,674-82.
Irvin, R. T., Chatterjee, A. K., Sanderson, K. E. & Costerton, J. W. (1975).
Composition of the cell envelope structure of a lipopolysaccharide-defective
(heptose-deficient) strain and a smooth strain of Salmonella typhimurium. I.
Bacteriol., 124,930-41.
Izumori, K., Rees, A. W. & Elbein, A. D. (1975). Purification, crystallization, and
properties of o-ribose isomerase from Mycobacterium smegmatis. I. Bioi. Chem.,
250, 8085-7.
Jeanlotz, R. W. & Fletcher, H. G. Jr (1951). The chemistry of ribose. Adv. Carbohydr.
Chem. Biochem., 6, 135-74.
Microbial Production of D-Ribose 193
Nogami, I., Kida, M., Iijima, T. & Yoneda, M. (1968). Studies on the fermentative
production of purine derivatives. Part I. Derivation of guanosine and inosine-
producing mutants of a Bacillus strain. Agric. Bioi. Chem., 32, 144-52.
Ochoa, S. & Rossiter, R. J. (1939). Flavin-adenine-dinucleotide in rat tissues.
Biochem. J., 33,2008-16.
Ogata, K. (1976). History and prospects. In Microbial Production of Nucleic Acid-
Related Substances, ed. K. Ogata, S. Kinoshita, T. Tsunoda & K. Aida. Kodansha
Ltd, Tokyo and John Wiley & Sons, New York, London, Sydney, Toronto, pp.
xiii-xviii.
Ogata, K., Kinoshita, S., Tsunoda, T. & Aida, K. (1976). Microbial Production of
Nucleic Acid-Related Substances. Kodansha Ltd., Tokyo and John Wiley & Sons,
New York, London, Sydney, Toronto.
Otani, M., Ihara, N., Umezawa, C. & Sano, K. (1986). Predominance of gluconate
formation from glucose during germination of Bacillus megaterium OM B1551
spores. J. Bacteriol., 167, 148-52.
Overend, W. G. & Stacey, M. (1955). Chemistry of ribose and deoxyribose. In The
Nucleic Acids, Vol. I, ed. E. Chargaff & J. N. Davidson. Academic Press, New
York, pp. 9-80.
Perlman, D. (1979). Ketogenic fermentation processes. In Microbial Technology, Vol.
II, ed. H. J. Peppler & D. Perlman. Academic Press, New York, San Francisco,
London, pp. 173-7.
Petersson, G. (1974). Gas-chromatographic analysis of sugars and related hydroxy
acids as acyclic oxime and ester trimethylsilyl derivatives. Carbohydr. Res., 33,
47-61.
Phelps, F. P., Isbel, H. S. & Pigman, W. (1934). Mutarotation of beta-d-ribose and
beta-I-ribose. J. Am. Chern. Soc., 56, 747-8.
Postma, P. W. & Lengeler, J. W. (1985). Phosphoenolpyruvate: carbohydrate
phosphotransferase system of bacteria. Microbiol. Rev., 49,232-69.
Postma, P. W. & Roseman, S. (1976). The bacterial phosphoenolpyruvate: sugar
phosphotransferase system. Biochim. biophys. Acta, 457,213-57.
Rebers, P. A. & Heidelberger, M. (1959). The specific polysaccharide of type VI
Pneumococcus. I. Preparation, properties and reactions. J. Am. Chem. Soc., 81,
2415-19.
Rebers, P. A. & Heidelberger, M. (1961). The specific polysaccharide of type VI
Pneumococcus, II. The repeating unit. J. Am. Chem. Soc., 83,3056-9.
Roberts, W. K., Buchanan, J. G. & Baddiley, J. (1963). The specific substance from
Pneumococcus type 34(41). The structure of a phosphorus-free repeating unit.
Biochem. J., 88, 1-7.
Roseman, S. (1969). The transport of carbohydrates by a bacterial phosphotransferase
system. J. Gen. Physiol., 54, 138s-180s.
Sable, H. Z. (1950). Phosphorylation of ribose and adenosine in yeast extracts. Proc.
Soc. exptl. Bioi. Med., 75,215-19.
Sable, H. Z. (1952). Pentose metabolism in extracts of yeast and mammalian tissues.
Biochim. biophys. Acta, 8,687-97.
Saier, M. H. Jr (1977). Bacterial phosphoenolpyruvate:sugar phosphotransferase
systems: structural, functional, and evolutionary interrelationships. Bacteriol. Rev.,
41,856-71.
Saito, N. & Sugiyama, S. (1966). D-Ribose formation by Pseudomonas reptilivora.
Agric. BioI. Chem., 30,841-6.
Sano, K., Yokozeki, K. & Mitsugi, K. (1977a). Enzymatic production of 5-amino-4-
imidazole-carboxamide from 5-amino-4-imidazole-carboxamide-riboside by Bacillus
thiaminolyticus. Optimal conditions for the enzyme formation and the enzymatic
reaction. Agric. Bioi. Chern., 41,2331-4.
Sano, K., Yokozeki, K. & Mitsugi, K. (1977b). Screening of micro-organisms
Microbial Production of D-Ribose 195
1 INTRODUCTION
2 CHEMISTRY
Sodium o-pantothenate I White, hygroscopic crystals. Decomposed by acids and bases. Solutions
are stable between pH 5 and 7. mp 122-124°C; [aJ~ + 27·1° (c = 2).
CHlOH For solubility, see calcium pantothenate
I I
HOCH2C-CHCONH(CH2)2COONa
I
CH 3
4H 16NaN0 5 MW: 241·21
Soluble in water; insoluble in ethanol. Unstable to bases. Free acid is
4' -Phosphopantothenic acid (Ba salt) unstable. [a J~ + 9·0° (c = 3· 3). King & Strong (1951); Okada et al.
o CH 3 0H (1967).
/I I I
HO-P-OCH2C-CHCONH(CH2)2COOH
I I
OH CH 3
4H 16NO HP MW: 313·27
Pantothenoyl-L-cysteine (Sa salt) Soluble in water, methanol, moderately soluble in ethanol; insoluble in
ether. Unstable to acids and bases. [ll']D - 14° (c = 2). Ohta et al.
CH,OH COOH
II I (1967).
HOCH2C-CHCONH(CH2hCONHCHCH2SH
I
CH,
C12H22N206S MW: 322·38
4' -Phosphopantothenoyl-L-cysteine (Sa salt) Soluble in water; slightly soluble in alcohol. Unstable to acids and bases.
o CH,OH COOH Easily oxidized in air. [ll']~ 0° (c = 2). Baddiley & Mathias (1954);
II I .I I Nagase (1967).
HO-r-OCH2T-CHCONH(CH2hCONHCHCH2SH
OH CH,
C12H2,N209PS MW: 416·42
CH 3 2
4' -Phosphopantetheine (Sa salt) Soluble in water; slightly soluble in ethanol; insoluble in ether. Unstable
o CH 30H to acids and bases. Easily oxidized in air. [ll']~ + 13·3° (c = 2·25);
II I .I Oxidized form, [ll']~ + 12·2°. Acetobacter suboxidans factor. Baddiley
HO-P-OCH 2C-CHCONH(CH2hCONH(CH 2hSH & Thain (1953); Moffatt & Khorana (1961); Nagase (1967).
I I
OH CH 3
CIIH23N207PS MW: 358·35
(continued)
Table.l-(contd.)
Dephospho-coenzyme A (Li salt) Soluble in water, methanol; insoluble in acetone. Unstable to acids and
bases. For ref., see coenzyme A.
o CH,OH
" I . I
HO-P-OCH 2C-CHCONH(CH 2hCONH(CH 2hSH
I I
9 CH, i:NH2
HO-f=0 <N I N
~tJ N
J
OH OH
~IH36N7016P,S
MW: 767·55
o OH
I
HO-P~
HO/~
4' -Phosphopantetheine-S-sulfonate (Ca salt) Soluble in water. [a]~ + 7·2° (c = 1·95). Bifidus factor. Yoshioka &
o CH,OH Tamura (1971).
II I· I
HOrOCH2r-CHCONHCH2CH2CONHCH2CH2SS0,H
OH CH,
CIIHnOlON2S2P MW: 438·41
(continued)
Table l-contd.
4' -O-(!3-glucopyranosyl)-D-pantothenic acid Soluble in water. [c:r]ii - 18· 20 (c = 1·0). Growth factor of Leuconostoc.
Amachi et at. (1971).
CH 20 H lH3?H
f{ o OCH 2-C-CHCONHCH 2CH 2COOH
OH I
HO CH2
OH
C44H2S014N MW: 791·68
OA-6129A (Na salt) Unstable to acids and bases. [c:r]ii + 11.60 (c = 1·0). Antibiotic produced
OH CH 1 by Streptomyces sp. OA-6129. Yoshioka et al. (1983).
I I'
}x£ ;/ SCH2CH2NHCOCH2CH2NHCOCH-r-CH20H
CH 3
COOH .
C2oH31N107S MW: 457·54
Vitamin B s , Coenzyme A and Related Compounds 205
!• 4 '-~tetheine •,
I : ' pantetheine •'
: : - -pantothenic aci.d---\ I
: :-pantoic acid ~ I !
:: : : i
~H <j>H CH.C;>H
0=6-o-~-oCH'-{:-CHCo-NHCH2CH2CO-NHCH2CH2SH
Ho-~ ~~~
NHz_L
from the air. It breaks down into fJ-alanine and pantoic acid by alkaline
hydrolysis. The latter forms a lactone, i.e. D-( - )-pantoyl lactone, readily in
acid solution or upon heating. Acid hydrolysis of pantothenic acid gives
fJ-alanine and pantoyl lactone. Pantothenic acid is soluble in water, ethyl
acetate, dioxane, acetic acid, ether and amylalcohol, but is insoluble in
benzene and chloroform.
The structure of pantothenic acid contains a single asymmetric center, so
that it is optically active; only the natural D-( + )-isomer has vitamin activity.
The absolute configuration of the vitamin has been defined as R (Hill & Chan,
1970). The conformation of the vitamin has also been reported (Fritz & LOwe,
1962).
The calcium salt of pantothenic acid can be obtained as needle crystals from
methanol. It is moderately hygroscopic and is rather more stable to heat, air
and light than the free acid. It is soluble in water and glycerol and slightly
soluble in alcohol and acetone. Reviews by Baddiley (1955) and Wagner &
Folkers (1964) summarize early studies on the chemistry of pantothenic acid.
Table 1 lists several naturally-occurring derivatives of pantothenic acid. They
can be grouped into three types based on their chemical structures, i.e. simple
pantothenate derivatives, pantetheine derivatives in which cysteamine (or its
analogs) attaches by an amide linkage and coenzyme A derivatives in which
the pantetheine is adenosylylated (see Fig. 1).
Pantothenyl alcohol, an alcohol analog of pantothenic acid, is also a
pharmaceutically important unnatural derivative. Unnatural analogs of pan-
tothenic acid and coenzyme A were reviewed by Shimizu (1970).
3 BIOSYNTHESIS
CH,COCOOH 7 ~jH
~ CH"'r-COOH
CH,COCOOH CH,-c=O
pyruvic acid a-aoetolactic acid
yH,OH
HOCH,c-CHCONHCH,cH,aJOH
tlUD-pantothenic acid
\?
H,NCNHCH,CH.a>OH
N-carl:>am:>yl-Il-alanine
- 1 3
dihydrouracil ..L uracil
malonylsemialdehyde
3.1 /I-Alanine
Three routes to p-alanine have been reported. Several micro-organisms have
been reported to form p-alanine by a--decarboxylation of L-aspartic acid (Fig.
2, reaction 1). Confirmatory evidence for this conversion was provided by
Williamson & Brown (1979), who purified (to apparent homogeneity) from
extracts of Escherichia coli an enzyme that catalyzes the a--decarboxylation of
L-aspartic acid to yield p-alanine and CO2 • They also reported that the enzyme
is missing in a mutant of E. coli that requires either p-alanine or pantothenate
as a nutritional factor, but is present in the wild-type strain and in a revertant
strain of the mutant. It has also been suggested that p-alanine is produced by
decarboxylation of N-carbamoyl-p-alanine formed from uracil on the basis of
the observation that mutants of Salmonella typhimurium lacking the ability to
degrade uracil require N-carbamoyl-p-alanine, p-alanine or pantothenate as a
nutritional factor (Fig. 2, reactions 2, 3 and 4) (West et al., 1985). p-Alanine
may also be produced by transamination of malonylsemialdehyde produced
from propionic acid (Fig. 2, reaction 5 or 6), because enzyme activity
catalyzing this conversion was detected in several micro-organisms. However,
there have been no further studies concerning this reaction.
Vitamin B 5 , Coenzyme A and Related Compounds 207
3.3 Coenzyme A
The pathway for the biosynthesis of coenzyme A from pantothenic acid,
L-cysteine and A TP as shown in Fig. 3 was first demonstrated by Brown
(1959a,b) based on his studies with Proteus morganii and the early observa-
tions in the 1950s with pig liver and other organisms. Later, the validity of this
pathway was confirmed by Abiko and co-workers who separated and charac-
terized the enzymes involved in this pathway in rat liver (Abiko, 1975).
The first step in this pathway is the phosphorylation of pantothenic acid (Fig.
3, reaction 1) by pantothenate kinase (EC 2.7.1.33). The enzyme has been
purified and characterized from rat liver (Abiko et al., 1972) and
208 S. Shimizu & H. Yamada
r---" X
AlP ~ Pantothenic acid
I
: P-Pontothenic acid Pantothenoylcysteine
.~:
: AlP_J2 +Cysteine 6l
:lSi P-Pantothenoylcysteine Pantetheine
.!;!
~,
~lP
~
-g: P-Pantetheine
1!:
,QI L
AlP» "
If': Dephospho-CoA
+-
L5
I
: AlP_
!... __________________ CoA _________ 1I
I
Fig. 3. The pathway for the biosynthesis of coenzyme A from pantothenic acid,
L-cysteine and ATP. For chemical structures of each compound, see Table l.
Abbreviation used: CoA, coenzyme A.
Optical ________
resolution
Me Me
~ ~CHg~-~~CHO ~TOOMe
I
MeONa 4MeOH C;::OOEt
COOMe MeONa COOEt
Me MeONii
2 x )CHCHO
Me Me
T'K". [:~CHCH-~~CHO
~COCOOMe
1
~CLHLCHO ~
)CHCOCOOH +
j
Me
Meuet5Me0
'v, HCOOMe + MeOH
HCHO
'CH 0
,.fe 0
Me Me OH
Me
~~/ ___________
-t---\°o Microorganism
•• Me7---\-H
~_?O ~ -alanine • (R)-(+)-pantothenic acid
o 0
Fig. S. The reaction pathway for the chemicoenzymatic synthesis of 0-( - )-pantoyl
lactone.
D-( - )-pantoyl lactone in the reaction mixture reached 18·2 g/liter with a
molar yield of 90·5% (optical purity, 94·4% e.e.). This unique conversion
proceeds through the successive reactions as follows: (1) the enzymatic
oxidation of L-( + )-pantoyl lactone to ketopantoyl lactone (the same enzyme
as that in N. asteroides has been suggested to be the responsible enzyme for
this oxidation); (2) the rapid and spontaneous hydrolysis of ketopantoyl
lactone to ketopantoic acid, and (3) the enzymatic reduction of the ketopantoic
acid to D-pantoic acid. The enzyme catalyzing this reduction seemed to be
ketopantoic acid reductase, because R. erythropolis cells could not utilize
u: eo
OH carbonyl
00
reductase OH
0
(1)
L-PL DH (D-PL forming)
I H
(2)
carbonyl
L-PL° reductase
(L-PL forming} J(PL D-PL°
I f
Spontaneous Chemical
(3)t<PH 7) (HC1)1 (5)
Ho
OH
J(PA
MOH
reductase ,
(4)
OH OH OHOH
KPA D-PA
Fig. 6. Reactions involved in the enzymatic transformation to 0-( - )-pantoyl lactone.
Abbreviations used: L-PL, L-( + )-pantoyllactone; D-PL, 0-( - )-pantoyllactone; KPL,
ketopantoyllactone; KPA, ketopantoic acid; D-PA, o-pantoic acid.
212 S. Shimizu & H. Yamada
4.2 Coenzyme A
The production methods for coenzyme A roughly fall into chemical and
microbial ones. The chemical methods have been reviewed by Shimizu (1970)
and Mautner (1970). They are not practical due to their complexity.
Therefore, commercial production is carried out by microbiological methods.
Extraction of coenzyme A from yeast cells has been performed since the early
1950s. Usually, cells of baker's or brewer's yeasts, which are relatively rich in
coenzyme A, are used as the source. Later, an efficient enzymatic method
using Brevibacterium ammoniagenes cells as the catalyst was developed.
These microbial methods were reviewed by Shimizu & Yamada (1986).
A successful enzymatic method using the biosynthetic route of coenzyme A
from pantothenic acid, L-cysteine and ATP was first reported by Ogata et al.
(1970), who found that B. ammoniagenes has all five enzymes necessary for the
biosynthesis of coenzyme A in high activities. The above three substrates,
when added to a reaction mixture containing the bacterial cells, were
converted to coenzyme A with a satisfactory yield (2-3 g/liter). They also
found that the same organism can accumulate coenzyme A directly in the
culture medium on addition of pantothenic acid, L-cysteine and AMP,
adenosine or adenine in the presence of a surfactant, cetylpyridinium chloride,
and high levels of glucose (usually 10%), K2 HP0 4 and MgS0 4 ·7H2 0. Under
optimal conditions, the amount obtained was 5·5 g/liter. Most coenzyme A in
the medium was present in the disulfide form due to the vigorous shaking
during the reaction. After treatment of the culture filtrate with Duolite S-30,
charcoal and Dowex 1 (Cl-), and the reduction of the disulfide, the very pure
thiol form was obtained in a high yield. The mechanism of this coenzyme A
production has been suggested to be that shown in Fig. 7 (for details, see
Shimizu et at., 1979a,b).
In order to improve the product yield further, the regulation mechanism of
the biosynthesis was investigated. As described in the previous section, the
biosynthesis of coenzyme A in B. ammoniagenes is mainly controlled by the
feedback inhibition of pantothenate kinase by coenzyme A. This was the main
problem in the practical production, because the over-produced coenzyme A
itself stopped the biosynthesis. Two methods to abolish this feedback
inhibition have been reported.
A synthetic scheme in which the reaction is initiated by the condensation of
4' -phosphopantothenic acid and L-cysteine or the transadenosylylation of
Villlmin B s , Coenzyme A and Related Compounds 213
~ Cysteine
Ar4dhenicacid
PRPP AMP
~ne
Aderiine CoA
Fig. 7. Reaction sequences of coenzyme A production with Brevibacterium am-
moniagenes under A TP-generating conditions. Abbreviations used: PRPP, 5-
phosphoribosyl-l-pyrophosphate; CoA, coenzyme A.
5 ASSAY METHODS
The current world capacity of calcium pantothenate production and its demand
are presumed to be about 4000 and 3600-4000 tons/year, respectively. It is
mainly used as an additive of animal feed (about 3000 tons/year) and as a
pharmaceutical product (about 600 tons/year). Pantothenyl alcohol is used as a
source of panthothenate activity for pharmaceutical vitamin products. Pan-
tothenyl alcohol itself has no pantothenate activity; in fact, it is a competitive
growth inhibitor of several pantothenate-requiring lactic acid bacteria. How-
ever, it has been demonstrated to be quantitatively converted to pantothenic
acid in the animal body, and to be equivalent to pantothenic acid in man.
Pantethine, the disulfide of pantetheine, and coenzyme A are also used as
pharmaceutical products in several countries. They have been suggested to be
effective in reducing cholesterol level, curing fatty liver, and related diseases.
Some sulfonate derivatives of pantetheine or coenzyme A (Bifidus factors),
such as 4'-phosphopantetheine-S-sulfonate, which were originally isolated
from carrot roots have been shown to be growth factors of Bifidobacterium
(Yoshioka & Tamura, 1971). Addition of the bifidus factors to dried milk for
infants has been suggested to be useful in improving the quality of the milk. A
carbapenem antibiotic, OA-6129A, (Table 1) produced by Streptomyces sp.
OA-6129, may be an interesting example suggesting a new use of the vitamin
as building block for its synthesis (Yoshioka et al., 1983).
ACKNOWLEDGEMENT
Research from the authors laboratory was supported, in part, by a Grant in aid
of Scietific Research from the Ministry of Education, Science and Culture of
Japan.
Vitamin B s , Coenzyme A and Related Compounds 217
REFERENCES
1 INTRODUCTION
RJ
HO CH2 0R2
,9'4 5
/""
~ 1 61
CH 3 N
Pyridoxine: R\ = CH 20H, R2 = H
Pyridoxal: R\ = CHO, R2 = H
Pyridoxamine: R\ = CH2NH2, R2 = H
Pyridoxine 5'-phosphate: R\ = CH20H, R2 = P0 3 H 2
Pyridoxal 5'-phosphate: Rl = CHO, R z = P03HZ
Pyridoxamine 5'-phosphate: Rl = CHzNHz, Rz = P03H2
Fig. 1. Chemical structure of vitamin B 6 •
as mineral acids or aqueous alkali, hot or cold, do not affect the vitamin. With
ferric chloride, pyridoxine reacts as a phenolic substance giving a reddish
brown coloration. In alkaline solution, pyridoxine on treatment with 2,6-
dichloroquinone chlorimide gives an immediate blue color fading to reddish-
brown.
Pyridoxine hydrochloride occurs as white platelets, melting point 204-206°C
with decomposition. The free base melts at 160°C. The compound is optically
inactive. Both base and hydrochloride readily sublime without decomposition.
The hydrochloride is freely soluble in water but sparingly in alcohol and
acetone. The base is soluble in methanol. Rapid destruction of pyridoxine by
light occurs in neutral and alkaline solutions. In 0·1 M hydrochloric acid there is
very little destruction.
The tautomeric properties of pyridoxine are well illustrated by the changes
in its UV absorption produced by varying the hydrogen ion concentration. The
single maximum at 292·5 nm at pH 2 diminishes in intensity at pH 4·5, and
concomitantly a new maximum appears at 327·5 nm. This latter band increases
in intensity when the pH is changed to 6·75, and the 292·5 nm maximum
disappears but a new band appears at 256·0 nm. When the pH is further raised
to 10·2, both bands increase in intensity and shift to shorter wavelengths.
The existence of other forms of pyridoxine was recognized as a result of the
comparison of microbiological assays on extracts of natural materials with the
values based on chemical and animal assays.
When pyridoxine was treated with mild-oxidizing agents, a marked increase
in l?io-activity towards the micro-organism, Lactobacillus casei, was observed.
Autoclaving of pyridoxine in the presence of the assay medium or amino acids,
greatly increased the activity of pyridoxine towards the test organism,
Streptococcus faecalis R. The products formed by treating pyridoxine with
animating agents and mild-oxidizing agents were, respectively, the amino and
aldehyde derivatives of pyridoxine. The compounds are pyridoxamine (2-
methyl-3-hydroxy-4-aminomethyl-5-hydroxymethylpyridine) and pyridoxal (2-
methyl-3-hydroxy-4-formyl-5-hydroxymethylpyridine) .
The availability of pyridoxal in synthetic form permitted identification of
Microbial Production of Vitamin B6 223
Glycolate
D-Glutamate 2-Ketoglutarate
1
Glycolaldehyde
Pyridoxamine-P • .
Pyndoxal-P •• \. .J. Pyn'doxme-
. P
1
5-Pyridoxic acid lactone
1
4-Pyridoxic acid
1 1
2-Methyl-3-hydroxy-5-formyl-
5-Pyridoxic acid
pyridine-4-carboxylate
1
a-Hydroxymethyl-a' -(N-acetylamino- 1
methylene )succinate 2-Methyl-3-hydroxypyridine-
1
4,5-dicarboxylate
1
a-(N -acetylaminomethylene)-
succinate
1
Succinate semialdehyde + acetate
+NH 3 +C02
CH3
I
CH2
I
C=O CH3
I I
!0
-4'
....
""-
9
3
12 0.6 '"...
~
....iii ,.,
01
.......
~
..
""- II)
00
<I
~
"tI
"tI 8 0.4 ....01
~ .......u
....0 ...
'",qI ~
"
'-'
0
Cultivation time (h)
Fig. 4. Production of vitamin B6 by Flavobacterium sp. 238-7.
pyridoxine using cells or cell-free extract of the organism (Tani et al., 1969). In
particular, an appreciable amount of pyridoxal-P was formed by use of
phenylphosphate as the phosphoryl donor under alkaline conditions. Under
the optimal reaction conditions, 0·4 g of pyridoxal-P was formed from 2 g of
pyridoxine incubated with phenylphosphate and cells for 24 h.
_jp~'OHCH,-O
H°fr~OH
I OH
H3 C ::"'N
Pyridoxine 4' -a-D-glucoside
Fig. 5. Chemical structure of pyridoxine glucoside.
REFERENCES
1 HISTORICAL
In 1927, Boas described a skin injury produced in rats by feeding raw egg
white, as well as the occurrence in various foodstuffs of a 'protective factor X'
that prevented and cured this injury (Boas, 1927). Gyorgy et al. (1939) tracked
down this protective factor and called it vitamin H after the German Haut for
skin. Kogi & Tonnis (1936) isolated a yeast growth factor, biotin vitamin B8
from egg yolk in the form of its crystalline methyl ester and Kogi (1937)
determined its empirical formula. Biotin was later shown to be identical to
vitamin H, the protective factor X, and to coenzyme R. Melville et al. (1942)
determined the correct structure of biotin, which was later confirmed by total
chemical synthesis in the Merck laboratories (Harris et al., 1943, 1944a,b,
1945) and verified by X-ray crystallography (Traub, 1956; Trotter & Hamilton,
1966).
Studies on the biosynthesis of biotin started in 1937 with studies on the
nutritional requirements of micro-organisms (Mueller, 1937a, b; du Vigneaud
et al., 1942; Eakin & Eakin, 1942; Dittmer & du Vigneaud, 1944; Dittmer &
du Vigneaud, 1944; Lilly & Leonian, 1944; Stokes & Gunness, 1945).
Subsequently, through radiochemical and genetic-biochemical studies (Tatum,
1945; Pontecorvo, 1953; Ryan, 1956), part of the biosynthetic pathway was
proposed. Thereafter, a hypothetical pathway was established by Okumura et
al. (1962a,b) based on their investigations of the microbial production of
glutamic acid using biotin-requiring bacteria. Ogata et al. (1965a,b) and
Iwahara et al. (1966a,b,c) made extensive investigations of biotin biosynthesis
and verified the main pathway described by Okumura et al. (1962a,b) using
growing and resting cells: pimelic acid - 7-keto-8-aminopelargonic acid
(KAPA)-7,8-diaminopelargonic acid (DAPA)-dethiobiotin (DTB)-
biotin (Izumi & Ogata, 1977). Through the extensive studies of three groups,
Eisenberg (1973), Pai (1971) and Izumi (Izumi & Ogata, 1977), the enzymic
231
232 Y. Izumi & H. Yamada
system involved in the biosynthetic pathway from pimelic acid to DTB had
been established by 1975.
In the 1950s it became clear that biotin was involved as a cofactor in many
biochemical processes, especially a number of carboxyl transfer reactions.
Investigations on the function of biotin in the reaction mechanisms were
obstructed by the fact that biotin is active only when covalently attached to an
e-amino group of a lysyl residue of the enzyme (Kosow et al., 1962), as was
suggested earlier by the isolation of biocytin (e-N-biotinyllysine) from yeast
extract (Wolf et al., 1952).
Wakil et al. (1958) and Wakil & Gibson (1960) showed that highly purified
acetyl-CoA carboxylase, isolated from avian liver extracts, was enriched with
respect to biotin and sensitive to inhibition by avidin, suggesting the tight
binding of the vitamin to the carboxylase and its probable role in acetyl-CoA
carboxylation. Lynen et al. (1959) made the interesting observation that the
A TP- and Mi+ -dependent carboxylation of free biotin was catalyzed by a
bacterial p-methylcrotonyl-CoA carboxylase. This model reaction led to the
formation of an unstable carboxybiotin derivative which was identified as
1'-N-carboxybiotin (Knappe et al., 1961; Lynen et al., 1961). Subsequently, it
was demonstrated with several biotin enzymes that the site of carboxylation of
the enzyme-bound prosthetic group was the same as that of free biotin (Moss
& Lane, 1971).
The existence of separate subsites for the catalysis of each partial reaction
has been unequivocally demonstrated with acetyl-CoA carboxylase of
Escherichia coli (Fall & Vagelos, 1972; Polakis et al., 1974). These investiga-
tions have shown that there are three dissimilar subunits involved in catalysis:
(a) the biotin carboxyl carrier protein (CCP), (b) the biotin carboxylase (BC),
which catalyzes the carboxylation of the CCP biotin, and (c) the carboxyl
transferase (Cf), which catalyzes the carboxyl transfer from the CCP biotin to
acetyl-CoA. Wood's group (Wood & Barden, 1977) has unequivocally
demonstrated with transcarboxylase that each partial reaction is catalyzed by a
separate subunit and that a distinct CCP subunit is present.
Major portions of the amino acid sequence of the CCP from transcar-
boxylase of Propionibacterium shermanii and acetyl-CoA carboxylase from
Escherichia coli have been determined (Wood & Barden, 1977). Up to now,
the sequences of the portions around biotin of the CCP or the corresponding
domains have been elucidated with almost all the biotin enzymes including
chicken liver acetyl-CoA carboxylase (Takai et al., 1987), human pyruvate
carboxylase (Lamhonwah et al., 1987) and human propionyl-CoA carboxylase
(Lamhonwah et al., 1987). Recently, Takai et al. (1988) have deduced for the
first time the complete amino acid sequence of a biotin enzyme, chicken liver
acetyl-CoA carboxylase.
* asymmetric carbon
(a)
pressed in many ways: d-biotin, d-( + )-biotin, D-biotin and D-( + )-biotin.
Here, we describe it as (+ )-biotin, and unless otherwise stated, the term
'biotin' indicates ( + )-biotin. The structure of crystalline biotin has also been
investigated by X-ray diffraction techniques and is now known precisely
(Traub, 1956; Trotter & Hamilton, 1966; Stallings & de Titta, 1985). X-ray
crystallographic analysis of (+ )-biotin and of the carboxybiotin derivative
described previously revealed that the bicyclic ring system has a boat-like
configuration (Fig. 1(B». The planar ureido ring projects upward at an angle
of 62·0° with respect to the plane (plane A) formed by the four carbon atoms
of the tetrahydrothiophene ring. Another plane comprising S-1, C-2, and C-5
tilts upward at an angle of 37 ·6° with respect to plane A (Bonnemere et al.,
1965). Because of the cis orientation of the ureido ring with respect to the
aliphatic side chain, C 6 resides approximately 2·8 A from the 3' -N of the
ureido ring (Traub, 1956). Repulsion between the 3'-N and C-6 positions is
thought to cause the greater-than-anticipated C 3-Cz-C6 angle of 119°
(Traub, 1956).
There are seven other forms which can be chemically synthesized: (-)-
biotin, (+)- and (- )-epibiotin (cis-form); (+)- and (- )-allobiotin, and
( + )- and ( - )-epiallobiotin (trans-form).
The crystals of ( + )-biotin are colorless, fine long needles: m.p. 232-3°C,
[{1']22=91° (c=1, 0·1NNaOH). The isoelectric point is pH3·5. The pH of
0·01 % aqueous solution is 4·5. Solubility in water is about 22 mg/100 ml at
25°C and it is more soluble in hot water or in dilute alkali. Solubility in 95%
ethanol is about 80 mg/100 ml at 25°C. It is insoluble in other common organic
solvents. It is stable in air, in a wide temperature range, and up to about pH 9.
It is also stable after autoclaving in 6 N H 2S04 at 120°C for 1 h. Accordingly,
most of the naturally occurring biotin bound to protein can be hydrolyzed to
free forms by such autoclave treatments without any problem.
Biotin combines with avidin, a protein in raw egg white, and streptavidin, a
protein produced by an actinomycete to become inactive.
Various chemical reactions with biotin and their products are shown in
Fig. 2.
234 Y. Izumi & H. Yamada
?
Jl.. NH2
?
)l...N
f "MoO,
(HOI
Q(CH24
, COOH
HN~NH ~-<::HCOOOCH, HNt\H , , thienyl·"aleric acid
Q t(CH3J2S0 4
s CH,l.CO'l'H 5 CH,l.COCI A "MoO.
CHCOOH biotin acid '-- HN NH ~ H2N NH2 (HNO,)
N-flIotlnyi ft chloride sOO--- I--l -- H - HOOC(CH2'.COOH
ammo a c l d / CH,N,/
l.A).
5
COC(,
(CH 2'.COOH
l.A).
5 (CH2,.COOH adipic acid
Q NH, JL / biotin diaminoblotln
HN)l...NH HN NH ~ H, (Rane, Ni)
l...S~CH,l.CO::-- V(CH,>.CONHNH 2
biotin azide biotin hydrazide
Using about 1000 strains of molds, bacteria and actinomycetes, Ogata et al.
(1965a,b) and Iwahara et al. (1966a,b) investigated the accumulation of
biotin-vitamers when pimelic acid was added to media as a precursor. They
found a promoting effect of pimelic acid in a large number of strains. On
addition of pimelic acid to the medium of Bacillus sphaericus IFO 3525,
20-200 Ilg/ml of 'total biotin', which is the amount of biotin-vitamer given by
the bioassay using Saccharomyces cerevisiae and includes DTB, KAP A and
DAPA as well as biotin, was accumulated. This was several hundred times
more than has hitherto been reported in micro-organisms. The main com-
ponent of the biotin-vitamers formed from pimelic acid was (+ )-DTB.
Afterwards, Yamada et al. (1983) found this strain also produced relatively
large amounts (0·8-1·1Ilg/ml) of biotin from pimelic acid, as well as from
DTB, under optimized conditions.
Ogino et al. (1974a,b) demonstrated a production method using an
n-paraffin-utilizing bacterium from a new compound, oL-cis-tetrahydro-2-oxo-
4-n-pentylthieno-(3,4-d)-imidazoline (oL-TOPTI), which is a biotin analog
having a methyl group instead of a carboxyl group of the biotin molecule.
TOPTI was chemically synthesized from Nt ,N3 -dibenzyl-oL-cis-tetra-
hydrothieno-(3,4-d)-imidazoline-2,4-dione (compound VII in Fig. 6) via a
Grignard reaction with n-pentyl magnesium bromide, dehydration in the
Microbial Production of Biotin 235
4.1 Biosynthesis
Pimelic acid
Pimelyl CoA
7-Keto-B'amino-
pelargonic acid
(KAPA)
78-0iamino-
pelargonic acid
(OAPA)
Oethiobiotin
(OTB)
Biotin
positions of DTB takes place with the loss of two hydrogen atoms at
C-1 and C-4 and without the loss of hydrogen atoms at C-2 or C-3. These
results suggest that unsaturation does not occur at C-2 or C-3.
Table 2 (Izumi et al., 1981) shows that among the strains of bacteria and
yeasts tested for activities of the four biotin biosynthetic enzymes, only B.
sphaericus IFO 3525, a DTB producer as described previously, showed
significant activities for all four enzymes.
Microbial Production of Biotin 237
Table 1
Properties of Biotin Biosynthetic Enzymes
(a) Pimelyl-CoA synthetase of Bacillus megaterium
Optimum temperature 32°C
Optimum pH 8·0
Km value: pimelic acid 2·7 x 10- 4 M
CoA 5·5 x 10- 4 M
ATP 1·5 x to- 3 M
Mi+ 1·5 X to- 3 M
Substrate specificity Pimelic acid
(Other dicarboxylic
acids were inert)
Nucleotide requirement ATP, ADP
Metal ion requirement Mi+, Mn2+
Inhibitors EDTA, a, a' -dipyridyl,
o-phenanthroline
(cOlllinued)
238 Y. Izumi & H. Yamada
Table l---contd.
ti
NH2 NH2
Biotin diamino-
S COOH carboxylic acid (10)
4.3 Biodegradation
Biotin is known to be degraded by molds, yeasts and bacteria via f3-oxidation
of the side chain of the molecule (Izumi & Ogata, 1977; Tanaka et aI., 1988).
The biotin-degrading bacterium, Mycoplana sp. No. 166, formed bisnorbiotin,
a compound having two less carbon atoms in its side chain than biotin, from
Table 2
Biotin-vitamer Producing Abilitiesa and Biotin Biosynthetic Enzyme Activities of Various Yeasts and Bacteria
Yeasts
Saccharomyces kloeckerianus IFO 0016 trt 0·71 tr tr 0·06 0·21
Lipomyces starkeyi IFO 0678 0·015 0·02 tr 2·90 tr 0·07 ~
~.
Sporobolomyces salmonicolar IFO 0374 0·085 0·49 0·23 1·30 tr 0·06
C3
Sporobolomyces salmonicolor IFO 1038 0·068 0·30 0·57 1·80 tr 0·03
Sporobolomyces coprophilus IFO 1442 0·100 0·16 0·14 0·31 tr 0·01
s·0-
Rhodotorula glutinis IFO 0415 0·066 ~
0·12 tr 0·10 tr 0·03
-a
Bacteria t
Escherichia coli AKU 007 0·018 0·40 1-13 2·45 tr 0·03 '"S·
Klebsiella pneumoniae IFO 12059 tr 0·30 tr 0·18 0·08 0·03 :lI
Enterobacter aerogenes IFO 12010 0·021 tr 0·18 0·08 tr 0·02 .l;.
Alcaligenes faecalis IFO 3160 tr 1·58 0·21 0·16 tr 0·06 ~
5·
Bacillus megaterium NI 8100 tr 1·90 4·55 tr tr 0·10 S·
Bacillus roseus lAM 1257 tr 0·87 0·52 tr 0·15 0·01
Bacillus sphaericus IFO 3525 0·067 20·0 1·05 4·42 0·03 1·24
Brevibacterium divaricatum NRRL 2311 tr tr tr tr 1·25 0·03
Pseudomonas graveolens IFO 3460 tr 4·30 tr tr tr 1·44
a The amounts of biotin-vitamers produced by organisms grown with pimelic acid.
b The amount of biotin-vitamer given by the bioassay using Lactobacillus plantarum, which includes biotin and biotin
sulfoxide.
C 1, pimelyl-CoA synthetase; 2, KAPA synthetase; 3, DAPA aminotransferase; 4, DTB synthetase.
dTrace.
~
240 Y. Izumi & H. Yamada
o o
~ ~
N N
H
N N
CH3 ~ CH3
H3C (CH2)3CHCOOH lS).(CH2)3 CHCOOH
a-Methyldethiobiotin a-Methylbiotin
5-(2-Thienylhl- Amic/enomycin
valerie acid
H2NO(CH2hCHCOOH CH3
NHCOCHCHCH2 CH3
NHCH3
Stravidin
Fig. 4. Various biotin antimetabolites.
Microbial Production of Biotin 241
6 FERMENTATION
6.1 Dethiobiotin Production by B. sphaericus IFO 3525 (Ogata, 1970)
Medium: 10 g peptone, 5 g Casamino acids (Difco), 100 g soybean meal, 20 g
glycerol, 1 g K 2HP0 4, 0·5 g KCI, 0·5 g MgS0 4·7H20, 10 mg FeS04·7H20,
242 Y. Izumi & H. Yamada
as DAPA, then the biotin-vitamers (e.g., biotin, biotin sulfoxide and DTB)
are eluted with 0·012 M formic acid, and lO-ml fractions are collected. Biotin
concentrations are quantitatively determined by microbiological assays with
Saccharomyces cerevisiae and Lactobacillus plantarum.
8 BIOASSAY METHODS
There are three methods for the determination of biotin and its vitamers:
microbiological, chemical and enzymatic assay methods. Table 3 summarizes
the characteristics of the three methods. The microbiological assay has been
most commonly used among the three methods. This assay uses the micro-
organisms Lactobacillus plantarum ATCC 8014 (Scheiner, 1985),
Saccharomyces cerevisiae ATCC 7754 (Gyorgy, 1967) and Bacillus subtilis
AKU 236 (Iwahara et al., 1966c). In addition, a biotin-requiring mutant of E.
coli, C162, (bio B-, His-) has also been used as a test of the growth-
promoting activities of biotin-vitamers (Salib et al., 1979; Osakai et al., 1986b).
These micro-organisms have biotin-vitamer specifi.cities for growth, which
enables us to assay them differentially. Table 4 shows the microbiological
activities of various biotin-vitamers when assayed with L. plantarum, S.
cerevisiae and B. subtilis. The microbiological assay also has the advantage of
high sensitivity. However, one of the main disadvantages of the method is that
it requires the time-consuming cultivation of assay organisms and the laborious
preparation of washed cell suspensions. Tanaka et al. (1987a) have developed
improved assay methods for biotin using lyophilized cells of L. plantarum and
E. coli C162 and glycerol-suspended cells of S. cerevisiae. Figure 5 shows the
standard curves by the paper disk plate method and turbidimetric method
using such cells. Those lyophilized or glycerol-suspended cells, which are
preserved at -20°C, can be used for the assay for more than 1 year or half a
year, respectively.
9 BIOLOGICAL PROPERTIES
Feeding a diet either low in biotin or, more frequently, using a diet amended with
with raw egg white, produced biotin deficiencies in the rat, chicken, poultry, pig,
monkey, man and other species (Gyorgy & Larger, 1968). Chicken and poultry,
however, require an outside source of biotin, even when egg white is replaced
by other proteins. The effectiveness of egg white in producing biotin deficiency
is due to the presence of the glycoprotein avidin, which forms a complex with
biotin that renders the biotin unavailable to the host. Avidin also forms tight
complexes when biotin is the prosthetic group of an enzyme, resulting in the
loss of enzyme activity. The biotin-binding capacity is present not only in the
Table 3
Assay Methods of Biotin and Its Related Compounds
Table 4
Microbiological Activities of Various Biotin-Vitamers
egg whites, but also in the egg yolks of avian species and the turtle (Moss &
Lane, 1971). Since avidin is heat-labile, prolonged heating of egg white
denatures the avidin and destroys its biotin-binding capacity. Another biotin-
binding protein similar to avidin, streptavidin, is produced by a Streptomyces
strain (Chaiet & Wolf, 1964).
Biotin deficiency produces dermatitis and perosis in chicken and poultry
(Gyorgy, 1968); alopecia, seborrheic skin changes, spasticity of the hind legs
and cracks in the feet of pigs. The activities of the biotin-dependent enzymes
are also decreased (Whitehead, 1981). These enzymes are involved in
carboxylation, transcarboxylation, and decarboxylation reactions, and function
in the vitally important metabolic processes of glucose and fat synthesis (Moss
& Lane, 1971; Wood & Barden, 1977; Whitehead, 1981). Among the most
important enzymes are acetyl-CoA carboxylase, pyruvate carboxylase and
propionyl-CoA carboxylase.
It was generally believed that the combination of biotin in the feed
Microbial Production of Biotin 247
(A)
E
~4
III
c
0
N
.c
i0
bI
03
~
j
is
2
(B) 5 (e)
2 4
'"ci3
£
.c
i
&2
Fig. 5. Standard curves of microbiological assays of biotin. (A) Paper disc plate (agar
diffusion) method using lyophilized cells of L. plantarum (x) and E. coli (e) and
glycerol-suspended cell of S. cerevisiae (0). (B) Turbidimetric method using lyophilized
cells of L. plantarum. (C) Turbidimetric method using glycerol-suspended cells of S.
cerevisiae (0) and lyophilized cells of E. coli (e). From Tanaka et al. (1987a).
10 CHEMICAL SYNTHESIS
/t"
° /t"
° /c"
9
ToaH ~r ~r (a) ~HR ~HR (b) R~ RN 7R (e)
~R (d) R~ ~R (e)
CH=CH _ CH-CH - - CH-CH - - HC--CH __ Ht-CH - HC-CH-
tOOH taaH taoH taaH tOOH taoH taaH ot to AcatH)a
R=benzyl "0/ "0
o
• • IV V Ac=CH 3Ca
VI
RN
)<NR
Ht-tH
Hzt Co
"S/
(j)
Resolution
O-Camphorsulfonate
of (+)-IX
XIU XIV
Fig. 6. Flow sheet of one chemical process for the industrial preparation of ( + )-biotin.
(a) Benzylamine; (b) phosgene; (c) acetic anhydride; (d) Zn, a mixture of acetic acid
and acetic anhydride; (e) H 2S, HCI; (f) C2H50(CH2hMgBr; (g) H 2, Raney nickel; (h)
HBr, acetic acid; (i) silver o-camphorsulfonate; (j) isopropanol; (k) sodium di-
ethylmalonate; (I) HBr. From Goldberg & Sternbach (1949).
groups present (in substituted form) in the biotin molecule, i.e. the meso-
configuration in diaminosuccinic acid derivatives corresponding to the cis-
structure of the two amino groups in a ring compound such as biotin.
Moreover, since resolution is carried out at the intermediate stage (XII-XIII)
it permits the direct production of the optically and biologically active biotin,
( + )-biotin.
<" s~. ~
s COOCH3
H~H"
-H COOCH3
---.. H
-H
H_
~
H-·
-H
OOCH3
-H-
~COOCH
"H
-tI -
H 3
(t)-Biotin
o 0 OR bR
'>< IX: R=H
N3 N3 AcNH HNAc
X: R=CHlSO Z
VII XI
Fig. 7. Stereospecific synthesis of ( + )-biotin from D-mannose. From Ohrui & Emoto
(1975).
the aldehyde yield compound (IV) which has a side chain-like biotin. The
treatment of (IV) with NaOCH3 in methanol, followed by the reduction of the
resulting aldehyde (V) with NaBH4 produces (VII). Treatment of (VII) with
NaS affords a tetrahydrothiophene derivative (VIII) which is converted via
(XI) to (X). Treatment of (X) with NaN3 gives a diazido compound (XI).
Catalytic reduction of the azido groups of (XI) in a mixture of methanol and
acetic anhydride gives a diacetoamido derivative (XII). Treatment of (XII)
with Ba(OHb followed by the treatment with phosgene affords ( + )-biotin.
Thus, (+ )-biotin is synthesized from D-mannose in a good yield, because a
five-membered ring consisting of isopropyridene, which protects the hydroxyl
groups of the sugar, is used to fix the molecular conformation during the
intramolecular substitution reactions.
REFERENCES
Scheiner, J. (1985). Biotin. In Methods of Vitamin Assay, 4th edn, ed. J. Augustin, B.
P. Klein, D. A. Becker & P. B. Venugopal. John Wiley & Sons, New York, pp.
535-53.
Scheiner, J. & DeRitter, E. (1975). Biotin content of feedstuffs. J. Agric. Food Chem.,
23, 1157-62.
Spence, J. T. & Koudelka, A. P. (1984). Effects of biotin upon the intracellular level of
cGMP and the activity of glucokinase in cultured rat hepatocytes. J. Bioi. Chem.,
259,6393-6.
Stallings, W. & Titta, G. T. de (1985). Crystallographic investigations of biotin and
carboxybiotin derivatives. Ann. N. Y. Acad. Sci., 447, 152-68.
Stokes, J. L. & Gunness, J. (1945). Microbial activity of synthetic biotin, its optical
isomers, and related compounds. J. BioI. Chem., 157, 121-6.
Svejcar, J. & Homolka, J. (1950). Experimental experiences with biotin in babies. Ann.
Paediat., 174, 175-93.
Szybalski, E. H. & Szybalski, W. (1982). A physical map of the Escherichia coli bio
operon. Gene, 19,93-103.
Takai, T., Wada, K. & Tanabe, T. (1987). Primary structure of the biotin-binding site
of chicken liver acetyl-CoA carboxylase. FEBS Lett., 2U, 98-102.
Takai, T., Yokoyama, C., Wada, K. & Tanabe, T. (1988). Primary structure of chicken
liver acetyl-CoA carboxylase deduced from cDNA sequence. J. Bioi. Chem., 263,
2651-7.
Tanaka, K. (1981). New light on biotin deficiency. N. Engl. J. Med., 304,839-40.
Tanaka, M., Yamamoto, H., Izumi, Y. & Yamada, H. (1986). Purification and
properties of biotinyl-CoA synthetase from Mycoplana sp. No. 166. Archs Biochem.
Biophys., 251,479-86.
Tanaka, M., Izumi, Y. & Yamada, H. (1987a). Biotin assay using lyophilized and
glycerol-suspended cultures. J. Microbiol. Meth., 6,237-46.
Tanaka, M., Izumi, Y. & Yamada, H. (1987b). Enzymatic assay for biotin using
biotinyl-CoA synthetase. Agric. BioI. Chem., 51, 2585-6.
Tanaka, M., Izumi, Y. & Yamada, H. (1988). Microbial metabolism and production of
biotin. Vitamin (Japan), 62,305-15.
Tatum, E. L. (1945). Desthiobiotin in the biosynthesis of biotin. J. Bioi. Chem., 160,
455-59.
Thoene, J., Baker, H., Yoshino, M. & Sweetmen, L. (1981). Biotin-responsive
carboxylase deficiency associated with subnormal plasma and urinary biotin. N.
Engl. J. Med., 304,817-20.
Trainor, D., Parry, R. J. & Gitterman, A. (1980). Biotin biosynthesis. 2. Stereochem-
istry of sulfur introduction at C-4 of dethiobiotin. J. Am. Chem. Soc., 102, 1467-8.
Traub, W. (1956). Crystal structure of biotin. Nature, Lond., 178,649-50.
Trotter, J. & Hamilton, J. A. (1966). The absolute configuration of biotin.
Biochemistry, 5,713-4.
Vesely, D. L. (1982). Biotin enhances guanylate cyclase activity. Science, 216,1329-30.
Vesely, D. L., Wormser, H. C. & Abramson, H. N. (1984). Biotin analogs activate
guanylate cyclase. Molec. Cell. Biochem., 60, 109-14.
Vigneaud, V. du, Dittmer, K., Hauge, E. & Long, B. (1942). The growth-stimulating
effect of biotin for the diphtheria bacillus in the absence of pimelic acid. Science, 96,
186-7.
Wakil, S. J. & Gibson, D. M. (1960). Studies on the mechanism of fatty acid synthesis.
VIII. The participation of protein-bound biotin in the biosynthesis of fatty acids.
Biochim. Biophys. Acta, 41, 122-9.
Wakil, S. J., Titchener, E. B. & Gibson, D. M. (1958). Evidence for the participation
of biotin in the enzymic synthesis of fatty acids. Biochim. Biophys. Acta, 29, 225-6.
Watanabe, K., Yano, S. & Yamada, Y. (1982). The selection of cultured plant cell lines
producing high levels of biotin. Phytochemistry, 21,513-6.
256 Y. Izumi & H. Yamada
Whitehead, C. C. (1981). The assessment of biotin status in man and animals. Proc.
Nutr. Soc., 40, 165-72.
Wolf, D. E., Valiant, J., Peck, R. L. & Folkers, K. (1952). Synthesis of biocytin. J.
Arn. Chern. Soc., 74, 2002-3.
Wood, H. G. & Barden, R. E. (1977). Biotin enzymes. Ann. Rev. Biochern., 46,
385-413.
Yamada, H., Osakai, M., Tani, Y. & Izumi, Y. (1983). Biotin overproduction by biotin
analog-resistant mutants of Bacillus sphaericus. Agric. Bioi. Chern., 47, 1011-6.
Yamada, H., Osakai, M. & Izumi, Y. (1984). Formation of biotinyl-CoA synthetase,
the first enzyme involved in microbial biotin degradation. Agric. BioI. Chern., 48,
2039-45.
Chapter 15
1 INTRODUCTION
From the chemical point of view, the term vitamin B12 is synonymous with
cyanocobalamin, which is by far the most important component of the large
family of the cobalt corrinoids. In this review, however, as in most publica-
tions, the word vitamin B12 is given a very broad meaning so as to include all
the cobalt corrinoids of the cobalamin group. Other cobalt corrinoids which
can be considered analogous or precursors of cobalamins are also included.
Discovered in 1948 and identified as the anti-pernicious anemia factor,
vitamin B12 has been thoroughly investigated from several points of view,
including biochemical significance, biosynthesis in micro-organisms, production
in commercial amounts and applications. The B12 biosynthetic pathway in
micro-organisms is now largely known and good production processes are
available and will be described in this review.
Due to its implications in extremely important biochemical reactions,
vitamin B12 originated wide interest in large and useful applications in human
and animal health care. Many analogues and derivatives were obtained and
studied to find new activities with particular emphasis towards B12 molecules
endowed with anti-cancer activity. Unfortunately, none of these substances
were of particular interest and the application of vitamin B12 remained limited
to anti-pernicious anemia activity, polyvitaminic specialties and as a feed
supplement in husbandry. The interest in the development of vitamin B12
reached a peak in the 1960s and then decreased slowly.
At present, a well consolidated but steady market and good production
processes are in the hands of a very limited number of companies. Like other
vitamins, B12 found its niche and from the economic point of view it can now
be considered as a 'mature product'.
2 HISTORICAL
The elucidation of the chemical structure of vitamin B12 is one of the most
outstanding examples of the successful collaboration of the chemist, the X-ray
crystallographer, and the biologist. The chemistry of this complex molecule has
been reviewed by those actively connected with the research and their
interpretation of the X-ray crystallography (Brink et al., 1954; Hodgkin et al.,
1955, 1957) and the degradative studies leading to the confirmation of the
structural formula, proposed by Dr Hodgkin and associates, should be
consulted for a guide to the literature (Folkers & Wolf, 1954; Wolf & Folkers,
1954; Bonnett et al., 1955; Folkers et al., 1957).
Ten years were spent from the first efforts of Woodward and Eschenmoser
until a full chemical synthesis was achieved (Krieger, 1973; Maugh, 1973). It
turned out to be a very difficult and it involved a group in Zurich, and a group
in Cambridge (UK) with more than 100 people. The synthesis requires about
70 steps and is of no value for industrial purposes. Today, vitamin B12 is
exclusively obtained by fermentation processes utilizing high-producing micro-
organisms or, less frequently, sewage.
Vitamin B12 (or cyanocobalamin) belongs to the large family of the cobalt
corrinoids exhibiting the general formula presented in Fig. 1. The molecule of
cobalamins, which are the most interesting cobalt corrinoids, is formed by the
following entities: a macrocycle in planar position, constituted by 4 reduced
pyrrol rings linked through their N -atoms to a cobalt atom in central position
(corrin) and showing 6 side chains (3 acetamide and 3 propionamide residues).
Almost perpendicular to the macrocycle a nucleotide, formed by phosphate,
ribose and 5,6-dimethylbenzimidazole is linked, through a coordination bond,
to the cobalt atom as shown in Fig. 2. The group of the cobalamins includes
cyanocobalamin or true vitamin B 12 , hydroxocobalmin or vitamin B 12 a,
methylcobalamin or mecobalamin and 5'-desoxyadenosilcobalamin or cob am-
mide or coenzyme B12 of Barker (Barker et al., 1958), characterized by a
cyano, hydrozyl, methyl or 5'-desoxyadenosyl radical respectively (Fig. 2).
Other cobaltcorrinoids with different heterocyclic bases (like substituted
benzimidazoles or purines), have been found in various micro-organisms,
either spontaneously or after the supply of the corresponding base to the
fermentation medium. Pseudovitamin B12 and factor III, in which the base is
adenosine and 5-hydroxybenzimidazole respectively, are the most frequently
encountered. They can be considered analogues of vitamin B 12 . A list of these
compounds is reported by Perlman (1959) and by Marvyn & Smith (1964).
Natural analogues designated as 'incomplete' are also known, for instance
factor B which is devoid of the ribose-phosphate moiety. All the known
analogues showed low activity as growth factors for vertebrates and no
therapeutic effect but they are very potent growth factors for the micro-
organisms. Cyanocobalamin, hydroxycobalamin, methylcobalamin and vitamin
260 C. Spalla et al.
NH,COH,C CH,CH,CONH,
CH,
NH,COH,C CH,
~HCOCH,H,C
--------------------~
Cobalamins
x y Trivial name
C obamam ide or
coenzyme Bu:
ens. (Pierce et al., 1949, 1950). This compound is also unstable to light and
temperature and in the presence of cyanide originates the more stable, well
known cyanocobalamin. Methylcobalamin is the active form by which vitamin
B12 is involved in methionine biosynthesis in Escherichia coli from which it was
first isolated (Guest et al., 1962). As the other native forms mentioned above,
it is also unstable to light.
The B12 corrinoids show the best stability at pH ranging from 4 to 6; they are
sensitive to ascorbic acid, to thiols, to gamma rays. They are endocellular
products and are extracted from the cells by treatment with water, neutral
buffer solution, lower alcohols or aqueous acetone. When natural corrinoids
like vitamin B12 coenzyme, hydroxycobalamin or methylcobalamin are to be
obtained, the extraction must be performed in the dark. Such a procedure is
not needed when cyanocorrinoids are wanted. In fact, the addition of cyanide
to the cell material transforms the unstable coenzyme forms to the stable
cyano-form. This procedure is usually applied in industry. A comprehensive
review of the chemical and physical properties of vitamin B12 corrinoids has
been written by Friedrich (1975).
Table 1
Vitamin B12 Production by Different Strains
Strain Yield References
(mg/liter)
and the culture conditions may be, it is necessary to add to the medium some
elements essential for vitamin biosynthesis, like cobalt ions and frequently
5,6-dimethylbenzimidazole. The addition of potential precursors such as
glycine, threonine, c>-aminolevulinic acid and aminopropanol proved some-
times to be beneficial.
Investigations on the biosynthesis of the vitamin B12 molecule had mainly the
objective of elucidating the biosynthetic pathway leading to the tetrapyrrole
macrocycle and that leading to the incorporation of the nucleotide to form
vitamin B 12 .
With few exceptions all authors agree with the general pathway shown in
Fig. 3. Two molecules of c>-aminolevulinic acid condense to form por-
phobilinogen which is the basic unit of the macrocycle. Latest studies identify
in uroporphyrinogen III the point where the biosynthesis of vitamin B12 and
that of porphyrins diverge. One of the most striking structural differences
between the corrin and the porphyrin ring system is the lack in the former of
a methene bridge between one of the pairs of pyrroline-type rings.
Even if the sequence of methylations in the course of the biosynthesis of the
corrin ring remains unknown, early work from Shemin's laboratory showed
that all of the methyl groups are derived from methionine (Bray & Shemin,
1963; Shemin & Bray, 1964).
Substances endowed with a strong stimulatory effect on the production of
vitamin B12 are betaine and choline. Ansbacher et al. (1949) reported that out
of all the microbial processes available in 1949, the best ones were charac-
terized by the disappearance of choline from the medium.
Miller & Putter, quoted by Demain et al. (1968), discovered that betaine
264 C. Spalla et al.
Glycine
----------1
CH, units •
C02+
Cobyrinic acid
amlnopropanol
NH,
5' -deaxyadenosine
5,6-0imethyl-
benzimidazole
r
Riboflavin
Fig. 3. General pathway for the biosynthesis of cobalamins (Florent & Ninet, 1979).
20
18
16
14
- 12
e
~10
<.
N 8
eli
c: 6
°eItJ
+-
:;: 4
2
00 1 2 3 4 5
Additive (mg Iml)
Fig. 4. EtIect of betaine and choline on vitamin BJ2 production by Pseudomonas
denitrificans (Demain et aI., 1968).
raw material for vitamin Bt2 production. Even in this case, however, addition
to the medium of an extra amount of betaine exerts a stimulatory effect on the
produced vitamin Bt2 which increases from 90 to 140 Ilg/ml. Under the same
conditions betaine can be substituted by choline with similar results provided
the medium be suitably buffered with CaC03 •
As far as the biochemical mechanism· of the stimulation of vitamin B12
production by betaine and choline is concerned, it has been demonstrated that
it is not due to the methyl groups' donation to the corrin ring of vitamin B t2 , as
methionine only is the precursor of these 'extra' methyl groups (Demain &
White, 1971). According to Demain et al. (1968) and White & Demain (1971),
betaine is an absolute requirement also for porphyrin over-production and it is
needed by P. denitrijicans for over-production of both corrins and porphyrins.
Although the site of betaine action in the biosynthetic pathway is unknown,
these data point out that betaine may be a positive effector for some early
common step of corrin and porphyrin synthesis.
During the last 30 years, several reviews dealing with research on vitamin B 12
production have been published (Perlman, 1959, 1967; Prescott & Dunn, 1959;
Marvyn & Smith, 1964; Friedmann & Cagen, 1970).
Microbial Production of Vitamin B12 267
I@ I®
~nloO,<Jg/ml)~(l50,<Jg/ml)
Fig. 5. Mutation and selection program for the improvement of vitamin B12 production
by Pseudomonas denitrificans. The mutagens employed are the following ones: 1,
ultraviolet; 2, ethyleneimine; 3, nitrosomethyluretane; 4-7, N-methyl-N-nitro-N-
nitrosoguanidine; 8,10, ultraviolet-bromouracil; 9,11, psoralene near UV light (Garo-
fano & Merli, unpublished data).
The main objectives of the research work according to Florent & Ninet
(1979) have been:
to acquire a better knowledge of the biosynthetic pathway in order to
support biochemical and genetic studies;
to improve strains which turned out to be the most useful ones (propioni-
bacteria and Pseudomonas);
to replace traditional sugars by more economical nutrients in the media;
to improve extraction and purification techniques.
STOCK CULTURE
Lyophilized with skim milk
1
MAINTENANCE CULTURE
Test tube with medium (a) incubated 4 days at 30°C
1
SEED CULTURE - FIRST STAGE
2-liter Erlenmeyer flask with 0·4 liter of medium (b) incubated 2 days at
30°C without agitation
1
SEED CULTURE - SECOND STAGE
30-liter stainless steel fermentor with 10 liters medium (c), incubated 24 h
at 30°C without aeration and with frequent adjustment of pH to 6·5 with
aqueous NH 4 0H solution
1
PRODUCfION CULTURE
SOO-liter stainless steel fermentor with 340 liters of medium (d) sterilized
40 min at 120°C, inoculated with 7 liters of second-stage seed culture and
incubated at 30°C, during the first 80 h under slight nitrogen pressure (no
aeration) and with slow agitation, then during the next 88 h with agitation
and aeration (2 m3 h): pH adjusted to 7·0 by aqueous NH 4 0H addition
along the whole fermentation
Fig. 6. Vitamin B12 from Propionibacterium shermanii. Semi-pilot plant scale fe~IlI:en
tation process. Medium (a): tryptone, 10 g; yeast extract, 10 g; filtered tomato JUice,
200 ml; agar, 15 g; tap-water to 1 liter; pH adjusted to 7·2. Medium (b): medium (a)
without agar. Medium (c): corn-steep liquor, 20 g; dextrose, 90 g; tap-water to 1 liter;
pH adjusted to 6·5. Medium (d): corn-steep liquor, 40 g; dextrose (sterilized separ-
ately), 100 g; cobalt chloride, 20 mg; tap-water to 1 liter; pH adjusted to 7·0. (Adapted
from Florent & Ninet, 1979).
Microbial Production of Vitamin B 12 269
(DBI) are the most useful ones. Since aeration favours DBI formation, the use
of a two-stage culture is advisable in order to obtain the highest yields. In the
first stage, anaerobic culture is run to almost total depletion of sugar which
promotes the growth of the bacteria and cob amide biosynthesis. Then, in the
second stage, an aeration shift leads to DBI biosynthesis and conversion of
cobamide to cobalamin. The two stages can be carried out batchwise in the
same tank or continuously in two connected fermentors (Speedie & Hull,
1960). Almost no information is available on strain preservation and culture
maintenance conditions.
The fermentation media consist mainly of glucose or inverted molasses
(50-100 g/liter) with small amounts of ferrous, manganous and magnesium
salts in addition to cobalt salts (60-100 mg/liter), of buffering or neutralizing
agents and nitrogenous compounds. Among these, yeast preparations, casein
hydrolysates and other traditional sources are currently used. Corn steep
liquor (50-70 g/liter) is frequently preferred since it supplies some lactic and
pantothenic acids which enhance the growth of the bacteria. Generally, the
temperature of the culture is 30°C and the pH is controlled at around 6·5-7·0.
Reports indicate yields of 25-40 mg/liter of vitamin B12 by propionibacteria.
Figure 6 gives an example of a current process using P. shermanii.
Several Pseudomonas species are vitamin B12 producers but currently the most
used species is P. denitrificans (Miller & Rosenblum, 1960). In contrast to the
propionibacteria fermentation, that of Pseudomonas is characterized by the
fact that its growth parallels cobalamin biosynthesis under aerobic conditions
throughout the whole fermentation process.
Accordingly, fermentation is conducted with aeration and agitation in a
single tank, batchwise or continuously. The strain requires traditional nutri-
ents, such as yeast extract, sucrose and several mineral salts in the growth
medium. In order to enhance vitamin B12 production, the medium has to be
supplemented at the beginning of the culture with 10-25 mg/liter of DBI and
40-200 mg/liter of cobaltous nitrate (McDaniel, 1961; Daniels, 1970).
Likewise, betaine and to some extent choline, have favorable effects in
activating some biosynthetic steps as mentioned before. Glutamic acid
stimulates the growth of bacteria (Daniels, 1966; Koike & Hattori, 1975).
Owing to its low costs and high betaine and glutamic acid content, beet
molasses are preferentially used in industrial fermentations at a 60-120 g/liter
concentration in conjunction with 2-5 g/liter of ammonium phosphate and
some oligoelements. The optimum temperature is 28°C, with a pH around 7·0.
By mutation and selection of proper strains, vitamin B12 production rose
from 5 to 120-140 mg/liter within the last 20 years (Long, 1962; Lago &
Demain, 1969; Ferni & Pennella, unpublished data). The process is outlined in
Fig. 7. In Table 2 the composition of typical media for vitamin B12 production
with selected mutants of P. denitrificans is reported.
270 C. Spalla et al.
STOCK CULTURE
Lyophilized with skim milk
1
MAINTENANCE CULTURE
Test tube with medium A incubated 4 days at 28°C
1
SEED CULTURE-FIRST STAGE
2-liter round-bottomed flask with 0·6 liter of medium E inoculated with a
working culture slants and incubated 48 hr at 28°C on rotating shaker at
120 r.p.m. with 90 mm excursion.
1
SEED CULTURE-SECOND STAGE
40-80 liter stainless steel fermentor with 25-50 liter medium E inoculated
with 1-1·2% seed culture first stage, incubated 25-30h at 32°C, 200
r.p.m. agitation 0·2 bar pressure, 0·5 v/vm' aeration.
1
PRODUCTION CULTURE
500-liter stainless steel fermentor with 300-liter medium P inoculated with
5% seed culture second stage, incubated 140-160 hr at 32°C, 0·2 bar
pressure. Impeller speed regulated in order to assure sufficient oxygen
during the first phase of growth (20-30 hr) and then to maintain constant
the oxygen consumption at the level reached during the first phase. The
fermentation lasts 140-150 h and gives a production of about 120-
130,ug/ml
Fig. 7. Vitamin B12 from Pseudomonas denitrificans. Pilot plant scale production. For
media composition see Table 2. (Femi & Pennella, unpublished data.).
Microbial Production of Vitamin B12 271
Table 2
Media for the Production of Vitamin B12 with Pseudomonas denitrificans
Media (g/liter)
A E P
Sugar beet molasses 100 70 105
Sucrose 15
Betaine 3
Choline
Ammonium phosphate 2 0·8
Ammonium sulfate 0·25 2 2·5
Magnesium sulfate 0·2 0·2 0·2
Zinc sulfate 0·02 0·02 0·08
5-6 dimethylbenzimidazole 0·005 0·025
Cobalt chloride
Difeo agar 20
Tap water 1000 toOO 1000
pH 7·2 7·2 7·2
8.1 Hydroxocobalamin
During the last 30 years the extraction processes have been improved many
times; they usually include the following main steps: solubilization of cobala-
mins and conversion to cyanocobalamin and isolation of a crude product, 80%
pure (utilizable directly for animal feeding), followed by purification to a
95-98% level (for medical use). Usually the whole broth or an aqueous
suspension of harvested cells is heated at 80-120 C for 10-30 min at
D
10 BIOASSAY METHODS
!
Filtrate Extraction 3 times with 350 ml of a cresol
and carbon tetrachloride mixture (1: 2
ratio)
!
Organic extract I Addition of 500 ml n-butanol
Extraction 3 times with 100 ml water
!
Aqueous extract Extraction 3 times with 30 ml of a cresol
and carbon tetrachloride mixture (1: 2
ratio)
!
Organic extract II Addition of 200 ml acetone and 100 ml
ether
!
Pure vitamin B12
Fig. 8. Crystalline vitamin B12 isolation process (adapted from Florent & Ninet, 1979).
274 C. Spalla et aI.
& Hall, 1979; Beck, 1979). Also the application of HPLC for rapid corrinoid
determination has been reported (Jacobsen et al., 1979).
11 BIOLOGICAL PROPERTIES
The biological properties of vitamin B12 are many and this has never been
found for any other natural product; interesting biochemical, haematological,
neurological and oncological aspects in man have been reported. This
physiological versatility can be explained by the fact that vitamin B12 is
involved in the functioning of numerous enzymes, which have vital roles in
assuring a normal, undisturbed evolution of the metabolism.
against both the greediness of intestinal micro-organisms and loss in the stools
and is safely transported down to the gastrointestinal tract until it reaches a
section of the distal ileum, where it is reabsorbed by the intestinal mucosa.
This process is mediated by specific receptors of the mucosa cells for the
intrinsic-extrinsic-factor complex. Within the intestinal mucosa cells the
intrinsic-extrinsic-factor complex is split, and vitamin Bt2 is now transported
across the cell for reabsorption from its vascular contact side.
The IF (intrinsic factor of Castle; Castle et al. 1930) is a glycoprotein of
55 000-60 000 mol. wt showing variable sugar moieties according to the animal
species bearing it. The IF of pigs and man are very similar and this fact helped
much the elucidation of the absorption mechanism it plays. It is assumed that
IF possesses two receptor sites, one for the vitamin and the other for the
microvilli of the epithelial cells of the ileum. The absorption takes place at a
neutral pH and in presence of Ca or Mg ions (Shinton, 1972). As for the
formation of the IF-vitamin B12 complex, the presence of 5,6-benzimidazole
seems essential (Friedrich, 1975). There are other vitamin Bt2 binder proteins
isolated from the blood serum, named transcobalamins (TC). Among them TC
I and TC II are the most important ones. The biological turnover of vitamin
Bt2 in human liver is much lower than for other soluble vitamins, being its
half-life over 12 months.
Vitamin Bt2 deficiency causes pernicious anaemia, a disease which in the past
usually led to death within 1 month after detection and inevitably within 3 years.
As mentioned before, classical observations and experiments revealed the
cause of the disease in the lack of intrinsic factor produced in the stomach.
This factor was necessary for the reabsorption of vitamin B 12 . The anaemia is
hyperchromic, i.e. there is more haemoglobin available than red cells.
Maturation and division of erythroblasts are disturbed and the resulting cells
are too few and too big, i.e. macrocytic and megalocytic. Similar deficiencies
are seen in the neutrophyls which are hypersegmented and diminished in
number. These anomalies can be reversed by parenteral injection of extrinsic
factor (vitamin B 12) which immediately releases the inhibition of cell matura-
tion and corrects the anaemia in a few days. Patients affected by pernicious
anaemia show a series of disorders like bone marrow depression, transverse
myelitis (myelopathy), peripheral neuropathy and subacute combined de-
generation (Reynolds, 1979). A report on clinical diseases related to de-
ficiencies of vitamin B12 transport proteins is given by Hitzig et al. (1979).
Vitamin B12 deficiency is revealed best by its serum level, since experiments on
the presence and determination of this compound showed that plasma and
serum are primarily indicative of this disorder, other tissues or organs being
affected only later on (Sullivan, 1970). The normal Bt2 serum level is
200-900 pg/ml and its total binding capacity is about 500-1100 pg/ml so that
278 C. Spalla et al.
the serum is saturated by only about 60% with vitamin B 12 . A serum B12 level
of about 200 pg/ml is considered critical. Clinical deficiency shows serum
concentrations of 150-80 pg/ml. Patients affected by leukemia show significant
higher serum levels due to a TC I increase. The presence of several cobalamins
in the plasma and the prevalence among them of methylcobalamin was
demonstrated first by Lindstrand & Stahlberg (1963) and Lindstrand (1964).
The optimal requirement of a healthy person for vitamin B12 is 0·5-1/-lg daily
according to Herbert (1968). The normal daily turnover of vitamin B12 is
2·5/-lg. In order to have a reabsorbtion of l/-lg of vitamin from an oral dose of
20/-lg, an amount of IF factor is required capable of binding 5-10 /-lg.
A vitamin B12 nutritional-dependent deficiency is rare and observable only
after about 10-20 years; it usually occurs in persons who live on a pure
vegetable diet. Oral treatment of vitamin B12 deficiency caused by the absence
or a defective function of IF allows a low level of reabsorbtion varying from 1
to 3% of a dosage of 100·000 /-lg. Therefore, a maintenance dose of about
300 /-lg daily of cyanocobalamin at intervals of 3-4 days is suggested.
According to Berlin et al. (1968) even higher doses (5OO-1OO0/-lg daily) are
suggested. Tablets of 10 and 100 mg are available. In case of parenteral
treatments in presence of minor vitamin B12 deficiencies a dosage of
500-1000 /-lg of hydroxycobalamin injected 3-4 times at regular intervals
within a year is suggested (Heinrich, 1970). Neuropathies caused by vitamin
B12 deficiency need massive dosages of the order of 1000/-lg daily for the first
week followed by 1000 /-lg twice a week for several months administered by
injection (Friedrich, 1975). The assumption that vitamin B12 lacks side-effects
even at high doses has to be revised.
Vitamin B12 is available in numerous forms and grades. As the US market is
still using almost exclusively cyanocobalamin, the major portion of B12
material in the world keeps being distributed in that form. However, in
Europe hydroxycobalamin is preferred since it shows a better uptake in the
liver, less urinary excretion and a more sustained serum level than cyanocobal-
amin (Heinrich, 1970) and permits a less frequent dosage regimen than needed
for the cyano form. Hydroxycobalamin is often combined with cob amide
(vitamin B12 coenzymes).
The pharmaceutical use of vitamin B12 varies considerably. In the United
States injections of vitamin B12 used as a general boost against tiredness as well
as a remedy for miscellaneous aches and pains are virtually unknown; in many
other countries, particularly in France, Italy and Spain the above-mentioned
indications are the predominant ones. On the other hand, the US is by far the
single largest market on a per capita basis, for vitamin B12 used in dietary
supplement formulations; straight vitamin B12 tablets are available in US stores
in strengths of 100 /-lg. Vitamin B12 finds a fairly large use as feed supplement.
Generally it is dosed into all animal feeds in Europe and the USA with the
exception of ruminants. The dosage levels are of 10-30 mg/ton of feed for
poultry, pigs and for calves as milk replacer.
The overall world market of vitamin B 12 , including both pharmaceutical and
Microbial Production of Vitamin B12 279
animal feed use, is around 3 tons/year which is not expected to increase in the
near future. The bulk price varies from 3 to 4 US $/g. The main producers of
this vitamin are Rhone-Poulenc and Roussel-Uelaf (France), Glaxo (UK),
Merck (USA) and Medimpex (Yugoslavia).
ACKNOWLEDGEMENT
The help of Anna Tomassoni in the translation and typing of the text is
acknowledged.
REFERENCES
Ansbacher, S., Hill, W. H., Tieman, J. W., Downing, J. F. & Caldwell, J. H., (1949).
Microbially synthesized APF (animal protein factor). Fedn Proc., 8, 180.
Barker, H. A., Weisbach, H. & Smith, R. D. (1958). A coenzyme containing
pseudovitamin B 12 . Proc. natn. Acad. Sci. U.S.A., 44, 1093-7.
Barrere, G., Geneste, B. & Satatier, A. (1981). Fabrication de la vitamine B 12 ,
l'amelioration d'un procede. Pour La Science, Novembre 1981, 56-60.
Beck, R. A., (1979). Essential prerequisites for the analysis of cyanocobalamin in
biochemical complex samples using radiometric competitive binding assays. In
Vitamin B 12 , ed. B. Zagalak & W. Friedrich. W. de Gruyter, Berlin, New York,
pp.675-80.
Beck, W. S. (1968). Desoxyribonucleotide synthesis and the role of B12 in erythropoi-
esis. In Vitamins and Hormones, Vol. 26, ed. R. S. Harris, G. G Wool, J. A.
Loraine & K. V. Thimann. Academic Press, New York, London, pp. 413-42.
Begley, J. A. & Hall, C. A. (1979). Effect of residual extract products and the type of
binders (R or IF) on serum vitamin B12 levels by radioisotope dilution assay. In
Vitamin B 12 , ed. B. Zagalak & W. Friedrich. W. de Gruyter, Berlin, New York,
pp.655-6.
Berlin, H., Berlin, R. & Brante, G. (1968). Oral treatment of pernicious anaemia with
high doses of vitamin B12 without intrinsic factor. Acta Med. Scand., 184,247.
Bemhauer, K. & Friedrich, W. (1954). Uber die vitamine der B13 gruppe. Angew.
Chem., 66,776-80.
Bonnett, R., Cannon, J. R., Johnson, A. W., Sutherland, L., Todd, A. R. & Smith, E.
L. (1955). Structure of vitamin B 12 . Nature, Lond., 176,325-8.
Bray, R. C. & Shemin, D. (1963). On the biosynthesis of vitamin B 12 . J. Bioi. Chem.,
238, 1501-8.
Brink, c., Hodgkin, D. C., Lindsey, J., Pickworth, J., Robertson, J. H. & White, J.
G. (1954). Structure of vitamin B 12 . Nature, Lond. 174, 1169-7l.
Cannata, J. J. B., Focesi, A., Mazumder, R. & Ochoa, S. (1965). Metabolism of
propionic acid in animal tissues. XII properties of mammalian methylmalonyl-
coenzyme A mutase. J. BioI. Chem., 240, 3249-57.
Cardinale, G. J., Dreyfus, P. M., Auld, P. & Abeles, R. H. (1969). Experimental
vitamin B12 deficiency: its effect on tissue vitamin B 12-coenzyme levels and on the
metabolism of Methylmalonyl-CoA. Archs Biochem. Biophys., 131,92-9.
Cary, C. A., Hartman, A. M., Dryden, L. P. & Likely, G. D. (1946). An unidentified
factor essential for rat growth. Fedn Proc., 5, 128.
280 C. Spalla et aI.
Kaczka, E. A., Wolf, D. E. & Folkers, K. (1956). Vitamin B12 analogs and processes
for preparing the same. US Patent 2,738,302, Merck and Co., Inc.
Kaken Kagaku, K. K. (1965). Japanese Patent 65/4,440.
Kato, R, Hayashi, M. & Kamikubo, T. (1968). Isolation of 5,6-
Dimethylbenzimidazolylcobamide coenzyme as a cofactor for glutamate formation
from Acetobacter suboxydans. Biochim. biophys. Acta, 165,233-7.
Kellermeyer, R W., Allen, S. H. G., Stjernholm, R & Wood, H. G. (1964).
Methylmalonyl isomerase. IV. Purification and properties of the enzyme from
propionibacteria. J. Bioi. Chem., 239,2562-9.
Koike, 1. & Hattori, A. (1975). Growth yield of a denitrifying bacterium Pseudomonas
denitrificans under aerobic and denitrifying conditions. J. Gen. Microbiol., 88, 1-10.
Kojima, 1., Salo, H., Maruhashi, K. and Fujiwara, Y. (1976). Abstr. Int. Ferment.
Symp. 5th, 1976, No. 7-04, p. 134.
Krieger, J. H. (1973). Vitamin B 12 : the struggle toward synthesis. Chem & Eng. News,
51, 16-19, 25, 27, 28.
Kyowa Hakko Kogyo Co. Ltd (1970). Vitamin B12 by fermentation. Japanese Patent
70/36159.
Lago, B. D. & Demain, A. L. (1969). Alternate requirement for vitamin B12 or
methionine in mutants of Pseudomonas denitrificans, a vitamin B12 producing
bacterium. J. Bacteriol., 99,347-9.
Lengyel, Rajarshi & Ochoa, S. (1960). Metabolism of propionic acid in animal tissues.
VI Mammalian methylmalonyl isomerase and vitamin B12 coenzymes. Proc. natn.
Acad. Sci. U.S.A., 46, 1312-19.
Lillie, R J., Denton, C. A. & Bird, H. R. (1948). Relation of vitamin B12 to the
growth factor present in cow manure. J. BioI. Chem., 176, 1477-8.
Lindstrand, K. (1964). Isolation of Methylcobalamin from natural source material.
Nature, Lond., 204, 188-9.
Lindstrand, K. & Stahlberg, K. G. (1963). On vitamin B12 forms in Human Plasma.
Acta Med. Scand., 174,665.
Long, R A. (1962). Production of vitamin B 12 . U.S. Patent 3,018,225, Merck and Co.,
Inc.
McDaniel, L. E. (1961). Production of cobalamins. US Patent 3,000,793: Merck and
Co., Inc.
McGinnis, J., Stevens, J. M. & Groves, K. (1947). The in vitro synthesis of a chick
growth promoting factor in hen feces. Poultry Sci., 26,432-3.
Marston, H. R, Allen, S. H. & Smith, R M. (1961). Primary metabolic defect
supervening on Vitamin B\2 deficiency in sheep. Nature, Lond., 190, 1085-91.
Marvyn, L. & Smith, E. L. (1964). The biochemistry of vitamin BI2 fermentation.
Progr. Industr. Microbiol., 5, 152-201.
Maugh, T. H. (1973). Vitamin B 12 : after 25 years, the first synthesis. Science, 170,
266-7.
Miller, 1. M. & Rosenblum, C. (1960). Production of vitamin B\2 using delta-
aminolevulinic acid. US Patent 2,939,822: Merck and Co., Inc.
Nakao, Y., Hisano, K., Kanemaru, T. & Yamano, T. (1974). Vitamin B12 production
by cultivation of Rhodopseudo monas and Corynebacterium bacteria in hydrocarbon
medium. Japanese Patent 74/15796: Takeda Chemical Ind. Ltd.
Ohmori, H., Ishitani, H., Sato, K., Shimizuki, S. & Fukui, S. (1971). Vitamin BI2
dependent glutamate mutase activity in photosynthetic bacteria. Biochem. biophys.
Res. Commun., 43, 156-62.
Pantskhava, E. S., & Bykhovsky, V. Y. (1966). Prod. Mikrobnogo Sin., Akad. Nauk
Latv. SSR Inst. Mikrobiol. 81-94: Chem. Abstr., 67 (1967) 10335.
Perlman, D. (1959). Microbial synthesis of cobamides. Adv. Appl. Microbiol., 1,
87-122.
Perlman, D. (1967). Microbial production of therapeutic compounds. In Microbial
Microbial Production of Vitamin B12 283
ABBREVIATIONS
1 INTRODUCTION
Orotic acid was first discovered in cow's milk in 1905 and was later found to
accumulate as an intermediate in the biosynthetic pathway of pyrimidine
nucleotides in a variety of mutants of micro-organisms. On the other hand, a
similar substance-which was a growth factor for rats, chickens and lactic acid
bacteria-had been isolated from distillers' dried solubles and had been known
as vitamin B 13 . It was proved in 1953 that orotic acid and vitamin B13 are
identical. Orotic acid is generally formed in mammals, so it is not a vitamin in
a strict sense.
Several reviews have been published on the production of orotic acid
(Furuya, 1976; Enei, 1984; Kuninaka, 1986) and the biosynthesis of pyrimidine
compounds (O'Donovan & Neuhard, 1970; Shiio, 1972).
In this chapter, recent aspects of the microbial production of orotic acid by
the de novo pathway will be described.
2 HISTORICAL
Orotic acid has a complex history. It was isolated from the whey of cow's milk
(Biscaro & Belloni, 1905) and subsequently found to be present in the milk of
* Present address: Central Research Laboratory, Shikishima Baking Co. Ltd, 3-5,
Shirakabe, Higashi-ku, Nagoya, 461 Japan.
285
286 K. Takayama & A. Furuya
various mammals. It was not until 1930 that the chemical structure of orotic
acid was perfectly known (Bachstez, 1930). Further investigations on the
compound were carried out late in the 1940s. Mitchell and co-workers (1947,
1948) observed that one of the pyrimidine-requiring mutants of Neurospora
utilized orotic acid instead of the pyrimidines, while the others, which did not
utilize orotic acid, accumulated large quantities of orotic acid in a culture
medium. Afterwards, isotopic experiments led to the conclusion that orotic
acid was the important intermediate in the biosynthetic pathway of pyrimidine
nucleotides.
On the other hand, a growth factor for rats and chickens, was discovered
from distillers' dried solubles (DDS) and named vitamin B13 (Novak & Hauge,
1948). Moreover, Wright et al. (1950) observed that DDS was also effective in
promoting the growth of Lactobacillus bulgaricus and orotic acid could
substitute for DDS. Subsequently, evidence has been given that vitamin B13
was identical to orotic acid (Manna & Hauge, 1953).
The accumulation of orotic acid by mutants of several species of bacteria was
reported, but the amounts accumulated were very low. In 1961, the accumula-
tion of orotic acid by a uracil-requiring mutant of Micrococcus glutamicus
(synon. Corynebacterium glutamicum) (a glutamic acid-producing micro-
organism), was reported (Tanaka et al., 1961; Kinoshita & Tanaka, 1963;
Konishita et al., 1963). This was the first report aimed at the industrial
production of orotic acid by fermentation procedure. Studies on the microbial
production of orotic acid have mainly been reported by Japanese researchers.
The industrial production of orotic acid by fermentation started in the
middle of the 1970s in Japan.
3.1 Chemistry
H~~
~COOH
H~ H
HOH'd HO OH
(a) (b)
Fig. 1. Chemical structure of orotic acid (a) and orotidine (b).
Microbial Production of Vitamin B13 287
Table 1
Properties of Orotic Acid and Oritidine
Property Orotic acid Orotidine
Molecular weight 156·10 288·22
Melting point 340-45°C Turns brown near 200°C
but failed to melt at 400°C
Absorption max. 282 nm 268nm
E 7680 (pH 4·4-7·2) 9570 (0·1 N HCl)
Solubility Soluble in water Soluble in hot water,
about 1·7 mg/ml lower aliphatic alcohols
magnesium, calcium and ammonium salts of orotic acid are scarcely soluble in
water.
3.2 Assay
4 PRODUCING MICRO-ORGANISMS
Bacteria
Aerobacter aerogenes pyr- OA Brooke et al. (1954)
Escherichia coli pyr- OA Yates & Pardee (1956)
Escherichia coli [AUR added] OA,Ura Skoda & Sorm (1958)
Escherichia coli [AUR added] OA,OMP,UMP Handschumacher (1958)
Serratia marinorubra ura- OA Belser (1961) ~
Corynebacterium glutamicum ura- (OPRT) OA Tanaka et al. (1961)
Bacillus subtilis ura- OR Konishi et af. (1963) ~
~
Corynebacterium glutamicum ura- + ade- OA,OR Nakayama et af. (1965) ~
Corynebacterium rathayi ura- OA,OR Nakayama et af. (1965) ""~
Brevibacterium ammoniagenes ura- OA,OR Nakayama et af. (1965) R-
Bacillus subtilis ura- OA,OR Nakayama et al. (1965) ;t.
Escherichia coli amino acid- OA ~akayama et al. (1965)
Brevibacterium ammoniagenes ura- (OPRT) OA Skodova et al. (1969a,b) .,~
Corynebacterium sp. ura- OA Skodova et al. (1969b) ~
Escherichia coli wild (OMP dec) OA Machida et af. (1969,1970) ""
Arthrobacter paraffineus ura- OA,OR Kawamoto et al. (1970)
Streptomyces showdoensis ura- (OMP dec) OA,OR Ozaki et al. (1972)
Escherichia coli ARs (OPRT) OA Shimosaka et al. (1984)
Bacillus subtilis ura- (OMP dec) OA,OR Doi et al. (1988)
Fungi
Neurospora sp. pyr- OA Mitchell et al. (1948)
Penicillum commune [NaF added] OA Sugimoto et af. (1962)
Candida tropicalis ade-/hyp- OA Watanabe et al. (1968)
Candida lipolytica ura- OA,OR Takayama et af. (1971)
ARs , adenosine sensitive
Microbial Production of Vitamin B13 289
0 0
HN:\ HNJ
o o/I OAN COOH OAN
HO-t°.
~H, 9H, HNACH,
~
I I
HN/ C ~O~ ~O~
o O.. C.... N/CH-COOH O"C'N/CH -COOH ~
/I O)'N:tCOOH ~
NH,-C-OPO,H, H H H OH OH OH OH ~
CO 2 [CPS] [ATC] [DHO] [DHOdeh] [OPRT] [OMPdec]
~ O"1P
~
+ ~ Carbamyl T Carbamyl Di hydro- Orotate • UMP R0-
~H3 1 Phosphate Aspartate oro tate ~
ATP ~
Aspartate t ~Ii:I
Ornithine NH, OR UDP
~ HOOC-CH,-tH-COOH
Cit.rull ine l
NH,
UTP
oII 0 0 O~~
II II N
Arginine HO-P-O-P-O-P-OC~'O
I I I
OH OH OH l
OH OH CTr
Fig. 2. Synthesis of pyrimidine nucleotide.
Microbial Production of Vitamin B13 291
repression by uracil compounds to the six enzymes and the feedback inhibition
by UMP to CPS were remarkably lost.
It has been observed that great differences exist for the regulation of the
pathway as a whole, especially for ATC, even though a common pathway is
found in all organisms (O'Donovan & Neuhard, 1970).
6 FERMENTATIVE PRODUCTION
6.1 Fermentation
Most orotic acid-accumulating strains require uracil for growth, hence the
concentration of uracil in media severely affects the accumulation of orotic acid
as well as cell growth. Figure 3(a) and (b) shows the remarkable effects of
uracil on the accumulation of orotic acid by C. glutamicum and B.
ammoniagenes, respectively (Nakayama et al., 1965). Maximal accumulations
were obtained at concentrations of uracil which allowed the micro-organisms
to grow to about half-maximal levels. Similar results were obtained with
uracil-requiring mutants of other micro-organisms (Brooke et aI., 1954;
Tanaka et al., 1961; Konishi et al., 1963; Kawamoto et aI., 1970; Ozaki et al.,
1972; Doi et al., 1988).
For the accumulation of orotic acid by a mutant of Candida tropicalis,
addition of aspartic acid to the medium is effective. The amino acid has been
shown to be a metabolic precursor of orotic acid, and the accumulation of
3.0 ( a) (b)
2.8
_ 2.5
e _ 2.4
......
If
-., 2.0
e
..
......
~2.0 1.0
.,
" "
,
~
~ 1.6 0.8
~o 1.5 ..o
o
."
Fig. 3. Effect of uracil on orotic acid accumulation. (a) C. glutamicum KY 9824, (b) B.
ammoniagenes KY 7349 . • , Orotic acid, /':,; orotidine; 0, Growth.
Microbial Production of Vitamin B13 293
Fig. 4. Bacterial cells and crystals of orotate in the orotic acid fermentation broth.
(scanning electron micrograph).
294 K. Takayama & A. Furuya
(a)
CO,Et
H,C" NH,
I + I
EtO,CCO
HN-?
CSMe
ro
NH
Ho,clN)"o
H
(b) EtO CH CO
CO,Et
NH
/ + I '--+ EtO,CHC
CO
=f.
0
.
I
NH
--+
CO,H NH,
I
HO,CHC=C~
1--+
_co
, , --- N--":::-O 'N-
H,N H H
7 CHEMICAL SYNTHESIS
8 APPLICATION
Orotic acid has been used as a hepatic drug. However, it may be contrarily
said that large doses of orotic acid result in liver disturbance, because of the
imbalance between purine and pyrimidine compounds.
Microbial Production of Vitamin B 13 295
REFERENCES
1 INTRODUCTION
2 HISTORY
Table 1
Some Historical, Production and Economical Data on L-Ascorbic Acid Development
Time period Events Production Price References
(years) (t/yr) (US$jkg)
3.1 Nomenclature
Names previously used for L-ascorbic acid were: cevitamic acid, hexuronic
acid, redoxon, scorbutamin, vitamin C and L-xylo-ascorbic acid. Throughout
this review, the name L-ascorbic acid will be used.
In 1965 the trivial name ascorbic acid or L-ascorbic acid was recommended
by the IUPAC-IUB Commission on Biochemical Nomenclature as a name for
vitamin C. The systematic name for L-ascorbic acid is L-threo-hex-2-enonic acid
y-lactone or L-threo-hex-2-enomo l,4-lactone (IUPAC-IUB, 1965).
302 V. Delic, D. Sunic & D. VlaJic
6
H~"
OH
,~
0
-0
H5 0CH0'20H
q HO H H --
HO OH
or 4 __ , =0 (b)
HO OH HH~O'"
'-- H 0
H"'" H"'" _ _ =0
HO
HO OH
(a)
3.2 Properties
Since fruit and vegetables are the main natural sources of L-ascorbic acid,
biosynthetic pathways in plants were subjected to extensive studies. It has been
shown that L-ascorbic acid is a product of hexose phosphate metabolism
including conversion of D-glucose or o-galactose to L-ascorbic acid (Fig. 2).
Microbial Production of Vitamin C 303
I
UDP-o-Glucose
I
1
UDP-Glucuronic acid
I
o-Galacturonic acid
1
o-Glucuronic acid
D-Glucuronic acid
1
L-Gulonic acid
1
L-Galactonic acid
1
L-Gulonic acid
1
L-Gulono-y-Iactone
~
j j
L-Gulono-y-Iactone L-Gulono-y-lactone
1
L-Galactono-y-Iactone
1
L-2-Keto-gulono- y-lactone
1
L-2-Keto-galactono-y-lactone
1
j
L-Ascorbic acid L-Ascorbic acid
(a) (b)
Fig. 2. Proposed routes of L-ascorbic acid biosynthesis in plants (a) and animals (b),
involving inversion of configuration; UDP-uridine diphosphate.
After 1896, when Bertrand for the first time described microbial oxidation of
sorbitol to sorbose by the micro-organism Bacterium xylinum (subsequently
classified as Acetobacter xylinum) (Bertrand, 1896), more than 30 years passed
until this microbial reaction was used as a biooxidation step in L-ascorbic acid
synthesis. Afterwards, research in the field of microbial biosynthesis of
L-ascorbic acid intermediates became more intensive and further knowledge
accumulated quickly.
During the past 50 years there have been a lot of attempts to find ways of
microbial conversion of intermediates in L-ascorbic acid synthesis which could
compete with the Reichstein-Griissner synthesis from an economical point of
view. Different micro-organisms were screened for their ability to carry out
reactions on carbohydrates as well as methods for new biosynthetic steps in
those routes. These reactions, micro-organisms and media are listed in Table
2. L-Ascorbic acid intermediates biosynthetic routes, including micro-organisms,
are shown in Fig. 3.
In this review we present names of micro-organisms as reported in the
original articles. Since their first presentation in the literature, the taxonomy of
these micro-organisms has undergone many changes due to their morphologi-
cal, cultural and biochemical properties, e.g. Acetobacter suboxydans has been
renamed Gluconobacter oxydans subsp. suboxydans. t For actual names of
micro-organisms see Bergey's Manual of Systematic Bacteriology (1984). Most
of the micro-organisms mentioned here are mutants with blocked side
reactions in 2-keto-L-gulonic acid biosynthetic routes from carbohydrates,
which could utilize alternative carbon sources.
t 'Names included in the Approved List of Bacterial Names are the only names which
are nomenclaturally valid as at the 1st January, 1980. All other names which have
appeared in the literature prior to 1st January, 1980 are nomenclaturally invalid'
(Skerman et af., 1980).
Table 2
Micro-organisms and Media Used in Biosynthesis of L-Ascorbic Acid Intermediates
(ATCC 21914) ~
Acetobacter Glucose- 5-ketogluconic Glucose 10%, 33 h Stubbs et al.,
suboxydans acid 1940
~
s::..
I:
(90%) g.
Acetobacter sp. Glucose- 5-KDG Glucose 10-15; yeast 0·75-1·0%; Teramoto et al., ;:s
(89%) 26°C, 7 days 1946 ~
Acetobacter o-Glucose- 2,5-DKG Glucose 10; yeast ext. 0·5; glycerol 0·5; MgS04 x 7H2O Stroshane &
meianogenus (93%) 0·5; CaC03 1%; pH7·0. 25-26°C, 300 rpm, 3 days Perlman, 1977 ~::!
ATCC9937 s·
Acetobacter o-Glucose-2,5-DKG 2 Liters of medium (pH 6·0) containing glucose 110; corn Kita & Hall, (j
cerinus (95%) steep liquor 0·5; (NH 4)zHP0 4 0·58; KH 2P0 4 1·5; 1981b
IFO 3263 MgS04 x 7H 20 0·05; urea 0·5 g/liter; CuS0 4 x 5H2O
1; nicotinic acid 0·3 mg/liter. 1; nicotinic acid
o· 3 mg/liter. 28°C, 36 h, 1700 rpm, aeration o· 75 v/v x
min. An additional 55 g glucose/liter was added at 20 h
and the pH was maintained at 5·5
Erwinia sp. o-glucose- Ca-2,5-DKG o-Glucose 5·8; com steep liquor 1-13; (NH4)zHP04 O'56; Sonoyama et al.,
SHS 2629001 (94·5%) CaC03 18·4; p-2000 antifoam 0·02%; pH 6·8. 1982
der. from 1·86 m3 /10 m3 ; 160 rpm; 3·6 m3 air/min; 28°C; 26 h.
SHS2006 During fermentation, 2304 kg 50% (wt/wt) o-glucose ...,
(ATCC 31626) was added 0
-..,J
(continued)
~
Table 2-contd.
Micro-organisms Conversion Media, conditions References
(yield)
Corynebacterium Ca-5-DKG- Ca-2-KLG o-Glucose 2; com steep liquor 3; NaN0 3 0·345; KH2P04 Sonoyama et ai,
sp. (92·5) 0·067; p-2000 antifoam 0·00167%; ZnS04 X 7H20 4·9; 1982
SHS 752001 MnCl2 x 4H20 0·8; thiamine hydrochloride 0·22; Ca-
der. from o-pantothenate 0·17 mg/liter; pH 6·9; 28°C; 160 rpm;
SHS 0007 1·18 air/min. NaN03 , o-glucose and Ca-2,5-DKG
(ATCC 31090) added during fermentation
Citrobacter 2,5-DKG-2-KLG 100 ml medium contg. cerelose 2; (~)2HP04 1; Kita & Hall, :"'
freundii (30%) KH2P04 1; MgS04 X 7H20 0·5; beet molasses 2 and 1981a t::::I
~
ATCC6750 glycine 0·2 g/liter; pH 6·7. After 22 h 15 ml of an ~
Acetobacter cerinus ferm. broth contg. 15-20% 2,5- !='
DKG was added 28°C; addnl. 52 h; pH 6·5
Erwinia o-glucose- 2-KLG Glucose 3; glycerol 20; yeast ext. 5; peptone 5; CaC03 Anderson et al. , §
7·5 g/liter; pH 7·0. 1 g/liter 2-KLG formed 1985 ~
herbicola (33%)
Estell et al., 1985 Ro
Erwinia o-glucose- 2-KLG 500 ml of medium contg. glucose 10; o-mannitol 20; com Hardy et al. , !='
citreus (50%) steep liquor 10; yeast ext. 1·5; casamino acids 10; 1987 ~
ERl116 K2HP04 4; KH2P04 1; ~Cl1; CaCI2 0·01 and ~
<"I,
K2S04 2·6 g/liter with three further additions of
10 g/liter o-glucose. 28°C; 800 rpm; over 60 h
Candida Galactonic- L-ascorbic Galactonic acid or galactono-y-Iactone 0·5; com steep Roland et al. ,
norvegensis acid acid liquor 0·25; ~CI 0·1; glycine 0·7; MgS04 x 7H2O 1985
KCC MF 42 (8-9%) 0·05; Na-glutamate 0·2; EtOH 1·5%; trace metals;
pH 4.2. 300C, stirring, aeration, 48 h. 0·43 g/liter L-
ascorbic acid formed
Microbial Production of Vitamin C 309
PseudogluconOb~cter Si!cch.ilroketog~
~y~
CHO
G. m.,.nog'''' fO HO}COOH
HO OH
1
~P.ut;d' ~X.
r
~p'Ut;d.
tran"u"ns :: H
HO
~ CH,OH CHZOH
L - Sorbosone l-Gulonic aCid
CH20H CH20H
HO
f f _HO
OH
HO
OH
~~
HO
OH
HO
OH
HO
0
l H O ~C:O~H
diacetone gulonlc .. cid gulonlc acid 2-Keto-L- Clcid
diace-tone gulonate
~glnosa
Po mildenber9li
~~HO~OH
COOH
p tluorescens
p.dlerug~
~g~
® G mel.ilnogenus
OH
HO
CH20H CH20H
L - Idose L - Idonic
acid
<;Qrynebacterlum
5-Keto-D- Brevibacterium
ILao_'d_-=:=:'n='~=~'=S~=-~= ;'='~=:__
{
Arthrobact~r
Micrococcus
~P...b):'lococcus
glueo,;, Bacillus
Erwmia
Gluconobact~r
~
Acetobacter cerlnus freundir
Crtrobact~r
AcetomonCiis albosesamae
CHZOH
2,5-0rKeto-O-
gluconrc acid
Erwima cltreus
Erwinla herbrcola
It seems likely that researchers in the People's Republic of China are able to
produce 2-KLG from L-sorbose (10 g/liter) over L-sorbosone on a commercial
scale with a 90% yield using bacteria of the genera Gluconobacter and Bacillus
(Kieslich, 1984; Lu et al., 1985; Ning et al., 1988).
Fujiwara et al. (1987) have used the G. oxydans strain, which has a high
activity of L-sorbose dehydrogenase, to produce 2-KLG. After a 4-day
cultivation of G. oxydans U-13 in the L-sorbose medium 2-KLG was formed
with a 64% yield. Also, G. oxydans U-13 was able to produce 2-KLG during
cultivation in o-sorbitol medium.
Recently, Nogami et al. (1987) reported the capability of Pseudo-
gluconobacter saccharoketogenes IFO 14464 (novel genus, isolated from soil
origin) and a concomitant bacterium Bacillus megaterium IFO 12108 to
produce 2-KLG in a high yield (76%) during cultivation in a 2 m 3 fermenter
containing 1000 liters of L-sorbose medium. Bacillus megaterium, which was
not directly involved in the oxidation process, promoted the growth of the
oxidative strain (P. saccharoketogenes) and increased the yield of 2-KLG. (Fig.
3). Compared with a pure culture of the oxidative strain the mixed culture was
able to oxidize higher concentrations of L-sorbose to 2-KLG in a shorter period
of time. The bacteria of some other genera could be successfully used instead
of B. megaterium, e.g. Pseudomonas, Proteus, Citrobacter, Enterobacter,
Erwinia, Xanthomonas and Flavobacterium.
11~ ~~';7ri."IP'
(ATCC 31(90)
l28"'C. .ah
[j 10.600 ml 0/ pn:-
seed medium in a
!
2·ntrc=fluk
~
:.:.:: :::.::::
.....:.. :
.20 ,••<> of the
seed tncdium in
l ~ml rermentor
1860 tilfa of 1M
ferment. medium in
100mJ {ermen.or
inoculated with the
cnlirc conterU of 4DO I:llrCl oIlhc:
the seed fcrmc:nlc:r fennent. medium in
IO-m 3 fennenlor
inoculated with the
entire content of
28"C' 26 h. ,60 rpm; tl'M: JCCd fennenu~:t
1..
3.6 m3 Iit/m.!n. (&Ilk
prcuure 1.5 .... /an 2;
tdd.ition of D1lucosc
28"C. 18h . '60 rpm;
1-18 m1 lir/min:
as a hydrogen
fermen ted broth:
328 m, Ca·2.5-DKGlmL do"".
Add.itJon of SOS. after 6 I'll
the IQlal viabk; cell
populldon docrc.&$Cd
!Torn 5 • 109 t. 9. ,03
c:cUsiml.
Table 3
Media Used in the Biosynthesis of Calcium 2,S-Diketo-o-Gluconate and
Calcium 2-Keto-L-Gulonate by Erwinia sp. SHS 2629001 and
Corynebacterium sp. SHS 725001
Media Concentration
(g/liter)
Erwinia sp. Corynebacterium sp.
Agar slant
o-glucose 5·0
glycerol 5·0
yeast extract 5·0 5·0
peptone 3·0 5·0
KH2 P04 1·0 1·0
MgS04 x 7H2 O 0·2 0·2
agar 20·0 20·0
Preseed medium
o-glucose 10·0
glycerol 5·0
yeast extract 5·0
peptone 5·0
com steep liquor 30·0
NaN0 3 1·0
KH2 P0 4 1·0 1·0
MgS04 x7H2 O 0·2
p-200 antifoam 0·1
Seed medium
o-glucose 10·0
glycerol 5·0
com steep liquor 30·0 20·0
NaN0 3 2·0
KH2 P04 1·0 1·0
MgS04 x7H 2 O 0·2
p-2000 antifoam 0·1 0·05
Fermentation medium
o-glucose 58·0 20·0
com steep liquor 11·3 30·0
NaN0 3 3·45
KH2 P0 4 0·67
(NH4)2HP04 5·6
CaC03 184·0
p-2000 antifoam 0·2 0·0167
ZnS04x 7H2 O 4·9 mg/liter
MnCl2 x 4H2 O 0·8 mg/liter
thiamine hydrochloride 0·22 mg/liter
Ca-0- Pantothenate 0·17 mg/liter
316 V. De/if, D. Sunil & D. Vlalif
These new technologies and processes significantly reduce the cost of bio-
synthetic production (Aiba et al., 1973; Rehm & Reed, 1981; Karube et al.,
1984; Kulbe & Knopki, 1986).
These methods rely on preparation of biocatalysts such as immobilized
enzyme(s), parts of cells or whole cells (Klein & Wagner, 1978; Scott, 1987;
Brodelius and Vandamme, 1987). The biocatalysts can be used for the
production of various substances in a single-step enzyme reaction (6-
aminopenicillanic acid from penicillin G; fructose from glucose, etc.). The
application of biocatalysts for multi-step enzyme biosynthesis (for instance,
antibiotics) is rudimentally described in the literature (Morikawa et al., 1979,
1980; Vandamme, 1984). This interesting area represents a new type of
biosynthesis, and as such this approach to L-ascorbic acid synthesis should not
be neglected in the future.
Martin & Perlman (1976) described conversion of L-sorbose to L-sorbosone
by biocatalysts. They used immobilized cells of native Gluconobacter melano-
genus IFO 3293 entrapped in polyacrylamide gel. Authors concluded that
conversion depended on the concentration of monomer used. The higher the
monomer concentration, the more L-sorbosone accumulated. Different results
were found with immobilized lyophilized cells. In this case a very active
preparation of biocatalysts was obtained with low monomer concentration due
to the impaired permeability of lyophilized cells, allowing diffusion of the toxic
monomers into the cytoplasm. Fujiwara et al. (1987a, b) described isolation,
purification and characterization of two enzymes (L-sorbose dehydrogenase
and L-sorbosone dehydrogenase) from Gluconobacter oxydans UV-10 involved
in conversion of L-sorbose to 2-KLG via L-sorbosone.
The cells of Gluconobacter oxydans strain No.6 VNIVI capable of oxidizing
o-sorbitol to L-sorbose were entrapped in polyacrylamide gel. The highest
oxidative activity of such a biocatalyst, reused five times, was 3·3 g of L-sorbose
per liter per h (Pomortseva & Krasil'nikova, 1983).
The production of L-sorbose by cells of Gluconobacter suboxydans (ATCC
621) immobilized in polyacrylamide gel was carried out in a continuous
process. The entrapped cells of this strain were capable of almost completely
converting o-sorbitol into L-sorbose at a rate of about 7 kg/m3 per h during a
long period of time (Stefanova et al., 1987).
Bailey et al. (1985) developed a mathematical model on K-carrageenan-
immobilized cells of Acetobacter suboxydans oxidizing ethyl alcohol to acetic
acid. This model, which predicted the cells, substrate and product concentra-
tion in reaction with the biocatalyst, could possibly be used for biooxidative
studies in the L-ascorbic acid pathway.
Studies on enzyme participation in the pathways of o-sorbitol oxidation were
done by Shinaga-,a et al. (1982). They described the solubilization, purification
and characteristics of o-sorbitol dehydrogenase from the membrane fraction of
Gluconobacter suboxydans var. a IFO 3254.
Recently, Sonoyama & Kobayashi (1987) reported two 2,5-DKG reductases
(reductase I and II) isolated from Corynebacterium sp. SHS 752001 mutant
318 V. De/ie, D. Sunie & D. V/aIiC
Molecular biology techniques developed in the last 15 years and the possibility
of gene manipulation in vitro (recombinant DNA technologies-rDNA),
particularly in microbial cells, resulted in the creation of hybrid features and
the non-conventional production of several substances in bioreactors. Gene
cloning was extended to many different micro-organisms, showing diversity of
application in a wide range of products like hormones, vaccines, enzymes,
amino acids, vitamins, antibiotics, etc. (Old & Primrose, 1985; Primrose,
1987). The classical example is cloning of genes for human insulin in the
bacterium Escherichia coli (Goeddel et aZ., 1979) and the production of this
hormone by fermentation (Frank & Chance, 1983). These techniques were
also usefully applied in the construction of improved microbial strains for
producing penicillin G-acylase, an enzyme involved in the hydrolysis of
penicillin G to 6-aminopenicillanic acid (Mayer et al., 1979; Vandamme, 1984;
Garcia & Buesa, 1986).
-
2.5- DKG red. gene
HindW
S
Transfectron ot E. colt JMIOI
w ith heterodupr;;-;;~ ' 2RF
(A)
ECOR!
mil 12 6 f Screening for phage
containing deletion
e)
m
I
Hind
E. CORl e
PS~ H;nd m
pBR 322
(B)
--
Handm
-----Ligation
Transformation J,I ~
------
MM294.
select ion for AmpR (olonies
I .
(C)
5 .6kb~
8
/ o.56kb
SR322 lE~:tR!
SphI
L1gation
I
Transfo,.mat ion of "",,!!,!Ill>1!....!lS=~. selection for Tet R cOlOnies
7 INDUSTRIAL PROCESS
The first chemical synthesis was reported by Reichstein et al. (1933; 1934).
L-Ascorbic acid was synthetized from L-xylose over L-xylosone but this route
was never commercialized. It is interesting that when Reichstein and co-
workers published the first synthesis of L-ascorbic acid, the correct structure
was not yet known.
The most important synthesis of L-ascorbic acid for the manufacturing
processes is that reported by Reichstein & Griissner (1934), well known in
literature as Reichstein's second synthesis (Fig. 3, route 1).
This synthesis from o-glucose included the inversion of the glucose chain,
which means the rearrangement of the molecule where C-1 of o-glucose
became C-6 of L-ascorbic acid. The main steps of the Reichstein-Griissner
synthesis were: reduction at C-1 (o-glucose- o-sorbitol) followed by microbial
biooxidation at C-5 (o-sorbitol- L-sorbose), acetone treatment (L-sorbose-
L-sorbose diacetone), oxidation at C-6 (L-sorbose diacetone- 2-keto-L-gulonic
acid (2-KLG) diacetone), hydrolysis (2-KLG diacetone- 2-KLG), esterifica-
tion (2-KLG- 2-KLG methyl ester) and lactonization (2-KLG methyl
ester- L-ascorbic acid). o-Sorbitol was oxidized to L-sorbose with Acetobacter
xylinum in a 60% yield. Later on A. suboxydans became the micro-organism
of choice due to its high effective oxidative ability. The modified Reichstein-
Griissner synthesis is a very efficient process for L-ascorbic acid synthesis
and served as the basis for modern industrial production.
From an industrial point of view, the Reichstein-Griissner synthesis has two
crucial features. The first is low cost of raw material, e.g. o-glucose which is a
starting compound, and second, fully protected intermediate L-sorbose diace-
tone for the second oxidation which precludes other side reactions.
Many chemical and technological modifications have improved the
efficiencies of each step. This multi-step chemical synthesis, which includes
micro-organisms for oxidation in one step, remains till now the principal and
the most economical process for industrial production of L-ascorbic acid. The
process steps including microbial oxydation are outlined in Fig. 6.
o-Sorbitol obtained by the hydrogenation of o-glucose was oxidized by
Acetobacter suboxydans into the key intermediate L-sorbose. L-Sorbose is then
condensed with acetone to form sorbose diacetone which is oxidized to a
diacetone derivative of 2-keto-L-gulonic acid (2-KLG). After hydrolysis and
esterification 2-KLG diacetone gives 2-KLG methyl ester which is converted to
L-ascorbic acid by cyclization. The procedure requires about 1·7 kg of
L-sorbose per kg of L-ascorbic acid with an overall yield of 66% (Jaffe, 1984).
324 V. Delil, D. Sunil & D. V[alil
D-GLUCOSE
!
D- SORBITOL
~
r-------------------------,
FERMENTATION
Acetob~cter Suboxy'd~ns
Filtered air
L-SORBOSE
~
L - SORBOSE DIACETONE
~
2 - KETO - L- GULONIC ACID
l
L- ASCORBIC ACID
Fig. 6. Synthesis of L-ascorbic acid including biooxidation of D-sorbitol to L-sorbose by
Acetobacter suboxydans.
7.2 Product recovery and purification
Since L-ascorbic acid is widely distributed in different plant and animal tissues,
as well as in crystalline form added to a variety of products (e.g. in food,
drinks, etc.), the determination and quantification of its content deserve a
special analytical approach.
Apart from these difficulties L-ascorbic acid is easily subjected to oxidation
by moderate oxidants, thus yielding dehydroascorbic acid which frequently
interferes with L-ascorbic acid determination.
As analytical details on L-ascorbic acid determination are not within the scope
of this review, readers interested in this field of L-ascorbic acid are recommended
to a very pertinent review by Pelletier (1985). The main analytical methods will
be mentioned here only briefly.
Bioassay as a method was not frequently used for L-ascorbic acid determina-
tion. This method based on guinea-pig protection against scurvy, was further
applied to determine if certain L-ascorbic acid derivatives or other compounds
exhibited anti-scurvy activity.
Most determinations are based on colorimetric methods using blue dye
2,6-dichloroindophenol (Hughes, 1983). In the presence of L-ascorbic acid, dye
is reduced, resulting in colourless or pink colour solution. Direct titration of
L-ascorbic acid with 2,6-dichloroindophenol is at present the most frequently
used oxido-reduction method applicable to diverse types of samples yielding
reproducible results.
Murty & Rao (1979) used iodine salts for the determination of L-ascorbic
acid. In this case different dye indicators could be used (naphthol blue black,
Amaranth, Brilliant Ponceau 5R) in titration mixture, liberating iodine which
in the presence of starch yielded a blue colour. Reduction of ferric, cupric and
mercuric ions by L-ascorbic acid has been the basis of simple and sensitive
methods for measuring L-ascorbic acid. Corresponding ferrous, cuprous or
mercurous coloured complexes with different substances were used as a
measure of L-ascorbic acid.
Pelletier & Brassard (1977) developed a method for the determination of the
total content of L-ascorbic acid, dehydroascorbic acid and diketogluconic acid
in the samples on the basis of coupling with 2,4-dinitrophenyl hydrazine.
Gas-liquid chromatography of trimethylsilyl L-ascorbic acid derivatives was
also adapted for quantitative determination of L-ascorbic acid in different
samples (Schlack, 1974).
326 V. De/ie, D. Sunie & D. V/aIie
Table 4
Main World Producers of L-Ascorbic Acida
10 CONCLUSION
ACKNOWLEDGEMENTS
REFERENCES
Geiger-Huber, M. & Galli, H. (1945). Uber den Nachweis der L-Ascorbinsaure ais
Stoffwechsel-produkt von Aspergillus niger. Helv. Chim. Acta, 28, 248-50.
Geigert, J., Hirano, D. S. & Neidleman, S. L. (1981). High performance liquid
chromatographic method for the determination of L-ascorbic acid and D-isoascorbic
acid. J. Chromatogr., 206, 396.
Goeddel, D. V., Kleid, D. G., Bolivar, F., Heyneker, H. L., Yansura, D. G., Crea,
R., Hirose, T., Kraszewski, A., Itakura, K. & Riggs, A. D. (1979). Expression in
Escherichia coli of chemically synthesized genes for human insulin. Proc. Natl.
A cad. Sci. U.S.A., 76, 106-10.
Gray, E. B. (1947a). 2-Keto-gulonic acid and its salts. US Patent No.2 421611.
Gray, E. B. (1947b). 2-Keto-gulonic acid and its salts. US Patent No.2 421612.
Grindley, J. F., Payton, M. A., Van de Pol, H. & Hardy, K. G. (1988). Conversion of
glucose to 2-keto-L-gulonate, an intermediate in L-ascorbate synthesis, by a
recombinant strain of Erwinia citreus. Appl. Environ. Microbiol., 54, 1770-5.
Hardy, K., Van de Pol, H., Grindley, J. & Payton, M. A. (1987). Production of a
vitamin C precursor using genetically modified organisms. WO 87/00863.
Haworth, W. N. & Hirst, E. L. (1933). Ascorbic acid. Chem. Ind. (London), 52,
645-6.
Hirst, E. L. (1939). The structure and synthesis of vitamin C and its analogs. Fortschr.
Chem. Org. Naturst., 2, 132-59.
Hirst, E. L., Percival, E. G. V. & Smith, F. (1933). Ascorbic acid. Nature, Lond., 131,
617.
Holloway, B. W. (1986). Organization of the Pseudomonas genome. In Proceedings of
the 5th International Symposium on the Genetics of Industrial Micro-organisms, ed.
M. Alacevic, D. Hranueli & Z. Toman. Pliva, Zagreb, pp. 393-401.
Huang, H. T. (1962). 2-Keto-L-gulonic acid. US Patent No.3 043 749.
Hughes, D. E. (1983). Titrimetric determination of ascorbic acid in commercial liquid
diets with 2,6-dichloroindophenol. J. Pharmac. Sci., 72, 126-9.
Isono, M., Nakanishi, K., Sasajama, K., Motizuki, K., Okazaki, H. & Yoshino, H.
(1968). 2-Keto-L-gulonic acid fermentation. I Paper chromatographic characteriza-
tion of metabolic products from sorbitol and L-sorbose by various bacteria. Agric.
Bioi. Chem., 32,424-31.
IUPAC-IUB (1965). Commission on Biochemical Nomenclature. Biochim. Biophys.
Acta, 107,4.
Jaffe, G. M. (1984). Ascorbic acid. In Encyclopedia of Chemical Technology, Vol. 24,
ed. R. E. Kirk & D. F. Othmer. John Wiley & Sons, New York, pp. 8-40.
Kahn, M. S. & Manning, R. F. (1988). Process for the preparation of keto-gulonic acid.
European Patent No. 276 832.
Kanzaki, T. & Okazaki, H. (1970). 2-Keto-L-gulonic acid fermentation. IV L-Sorbose
metabolism in Pseudomonas aeruginosa. Agric. BioI. Chem., 34, 432-6.
Karube, I., Suzuki, S. & Vandamme, E. J. (1984). Antibiotic production with
immobilized living cells. In Biotechnology of Industrial Antibiotics, ed. E. J.
Vandamme. Marcel Dekker, New York, Basel, pp. 762-80.
Katznelson, H., Tanebaum, S. W. & Tatum, E. L. (1953). Glucose, gluconate and
2-ketogluconate oxidation by Acetobacter melanogenus. J. Bioi. Chem., 204,43-59.
Kieslich, K. (1984). Present state of biotechnological productions of pharmaceuticals.
In Proceedings of the 3rd European Congress on Biotechnology, Vol. IV. VCH,
Weinheim, pp. 39-72.
Kita, D. A. (1979). Fermentation process for converting L-gulonic acid to 2-keto-L-
gulonic acid. US patent No.4 115 812.
Kita, D. A. & Hall, K. E. (1981a). Preparation of 2-keto-L-gulonic acid. US Patent No.
4245049.
Kita, D. A. & Hall, K. E. (1981b). Process for producing 2,5-diketo-gluconic acid. US
Patent No.4 263 402.
330 V. Delii, D. Sunii & D. VlaIiC
process for producing 2-keto-L-gulonic acid from sorbose. European Patent No. 278
447.
Nogami, I., Shirafuji, H., Oka, M. & Yamaguchi, T. (1987). A method for producing
2-keto-L-gulonic acid. European Patent No. 221707.
Obata, Y., Nara, K., Tarni, K., Mochizuki, K. & Isono, M. (1975). L-Ascorbic acid.
Japan Patent No. 75 22113.
Oga, S., Sato, K., Imada, K. & Asano, K. (1972). Fermentative manufacture of
2,5-diketogluconic acid. Japan Patent No. 72 38193.
Ogawa, T., Uzawa, J. & Matsui, M. (1977). A carbon-13 NMR study on the
conformation of L-ascorbic acid in deuterium oxide. Carbohydr. Res., 59, 32-5.
Okazaki, H., Kanzaki, T., Doi, M., Nara, K. & Motizuki, K. (1968). 2-Keto-L-gulonic
acid. II. Identification of metabolic products from sorbitol. Agric. BioI. Chem., 32,
1250-5.
Okazaki, H., Kanzaki, T., Sasayama, K. & Taketa, Y. (1969). 2-Keto-L-gulonic acid.
III Evaluation of the pathway of sorbitol metabolism in Gluconobacter melanoge-
nus. Agric. BioI. Chem., 33,207-11.
Old, R. W. & Primrose, S. B. (1985). Principles of Gene Manipulation, 3rd edn.
Blackwell Scientific Publications, Oxford.
Palleroni, N. J. (1986). Isolation and characterization of R' plasmids in Gluconobacter
oxydans. In Proceedings of the 5th International Symposium on Genetics of
Industrial Micro-organisms, ed. M. Alacevic, D. Hranueli & Z. Toman. Pliva,
Zagreb, pp. 415-24.
Pelletier, O. (1985). Vitamin C (L-Ascorbic and dehydro-L-ascorbic acids). In Methods
of Vitamin Assay, 4th edn, ed. J. Augustin, B. P. Klein, D. A. Becker & P. B.
Venugopal. John Wiley & Sons, New York, Chichester, Brisbane, Toronto,
Singapore, pp. 303-47.
Pelletier, O. & . Brassard, R. (1977). Determination of vitamin C (L-ascorbic acid and
dehydroascorbic acid) in food by manual and automated photometric methods. J.
Food Sci., 42, 1471.
Perlman, D. (1959). Preparation of 2-keto-L-gulonic acid by Pseudomonas aeruginosa.
US Patent No.2 917 435.
Pomortseva, N. V. & Krasil'nikova, T. N. (1983). Okislenie D-sorbita v L-sorbozu
immobilizovanimi kletkami Gluconobacter oxydans. Khim.-Farm. Zh., 17,721-5.
Primrose, S. B. (1987). Modern Biotechnology. Blackwell Scientific Publications,
Oxford.
Ramakrishnan, C. V. & Desai, P. J. (1956). Effect of addition of iron, cobalt and
ascorbic acid to the medium on the synthesis of ascorbic acid in molds. Current Sci.
(India), 25, 189-90.
Razumovskaya, Z. G. (1962). Obzori puti izpolzovanija mikroorganizmov v sinteze
vitamina C. Mikrobiologija, 31, 172-8.
Razumovskaya, Z. G. & Vasil'eva, O. A. (1956). Oxidation of glucose by acetic acid
bacteria. Ser. Bioi. Nauk, 41, 57-66.
Rehm, H. J. & Reed, G. (1981). Biotechnology, Microbial Fundamentals. Verlag
Chemie, Weinheim, Deerfield, Florida, Basel.
Reichstein, T. & Grussner, A. (1934). Eine ergiebige Synthese der 1-Ascorbinsaure.
(C-Vitamin). Helv. Chim. Acta, 17,311-28.
Reichstein, T., Grussner, A. & Oppenauer, R. (1933). Synthese der d- und
I-Ascorbinsaure (C-Vitamin). Helv. Chim. Acta, 16, 1019-33.
Reichstein, T., Grussner, A. & Oppenauer, R. (1934). Synthese der Ascorbinsaure und
verwandter Verbindungen nach der Oson-Blausaure-Methode. Helv. Chim. Acta,
17,510-20.
Robakis, N. K., Palleroni, N. J., Despreaux, C. W., Boublik, M., Baker, C. A.,
Churn, P. J. & Claus, G. W. (1985a). Isolation and characterization of two phages
for Gluconobacter oxydans. J. Gen. Microbiol., 131,2467-73.
332 V. Delic, D. Sunic & D. VlaJic
Sonoyama, T., Kageyama, B., Yagi, S. & Mitsushima, K. (1987b). Biochemical aspects
of 2-keto-L-gulonate accumulation from 2,5-diketo-D-gluconate by Corynebacterium
sp. and its mutants. Agric. BioI. Chem., 51, 3039-47.
Stefanova, S., Koseva, M., Tepavicharova, I. & Beshkov, V. (1987). L-Sorbose
production by cells of the strain Gluconobacter suboxydans entrapped in a
polyacrylamide gel. Biotechnol. Lett., 9,475-7.
Stroshane, R. M. & Perlman, D. (1977). Fermentation of glucose by Acetobacter
melanogen us. Biotechnol. Bioeng., 19, 459-65.
Stubbs, J. J., Lockwood, L. B., Roe, E. T., Taubenkin, B. & Ward, G. E. (1940).
Ketogluconic acids from glucose. Bacterial production. Ind. Eng. Chem., 32,
1626-31.
Stubbs, J. J., Lockwood, L. B., Roe, E. T. & Ward, G. E. (1943). 5-keto-gluconic acid
by fermentation of glucose. US Patent No.2 318 641.
Svirbely, J. L. & Szent-Gyorgyi, J. (1933). Vitamin C. Biochem. J., 27,279-85.
Szent-Gyorgyi, A. (1928). Function of peroxidase systems and chemistry of the adrenal
cortex-carbohydrate derivat. Biochem. J., 22, 1387-409.
Takeda Chemical Industries Ltd (1964). Manufacture of 2-keto-L-gulonic acid. French
Patent No. 1376741.
Takeda, R., Isono, M., Nakanishi, T. & Sasajima, K. (1964). 2-Keto-L-gulonic acid.
Japan Patent No. 64 2463.
Tengerdy, P. R. (1961a). Redox potential changes in the 2-keto-L-gulonic acid
fermentation. I Correlation between redox potential and dissolved oxygen con-
centration. J. Biochem. Microbiol. Tech. Eng., 3,241-53.
Tengerdy, P. R. (1961b). Redox potential changes in the 2-keto-L-gulonic acid
fermentation. II Relationship betwen redox potential and product formation. J.
Biochem. Microbiol. Tech. Eng., 3,255-60.
Teramoto, S., Yagi, R. & Hori, I. (1946). Utilization of oxidative fermentation. IV
Basal studies of manufacturing 5-keto-gluconic acid. J. Ferment. Technol., 24, 22-6.
Thompson, R. Q. (1987). Peroxidase-based colorimetric determination of L-ascorbic
acid. Analyt. Chem., 59, 1119-21.
Tsukada, Y. & Perlman, D. (1972a). The fermentation of L-sorbose by Gluconobacter
melanogen us. I General characteristics of the fermentation. Biotechnol. Bioeng., 14,
799-810.
Tsukada, Y. & Perlman, D. (1972b). The fermentation of L-sorbose by Gluconobacter
melanogen us. II. Inducible formation of enzyme catalyzing conversion of L-sorbose
to 2-keto-L-gulonic acid. Biotechnol. Bioeng., 14,811-8.
Tsukada, Y. & Perlman, D. (1972c). The fermentation of L-sorbose by Gluconobacter
melanogen us. III Investigation of the metabolic pathway from sorbose to 2-keto-L-
gulonic acid. Biotechnol. Bioeng., 14, 1035-8.
Ullmanns, (1983). Encyklopiidie der technischen Chemie, Vol. 23. Verlag Chemie,
Weinheim, Basel, pp. 685-92.
Vandamme, E. J. (1984). Biotechnology of Industrial Antibiotics. Marcel Dekker, New
York, Basel.
Wakisaka, Y. (1964a). 2,5-dioxogluconic acid by fermentation process. Japan Patent
No. 64 14493.
Wakisaka, Y. (1964b). Carbohydrates oxidation by Pseudomonas albosesamae. nov. sp.
I Taxonomic studies on a newly isolated bacterium P. albosesamae. Agric. Bioi.
Chem., 28,369-74.
Wakisaka, Y. (1964c). Carbohydrate oxidation by Pseudomonas albosesamae. nov.
sp. II Chemical structure of the product on glucose and gluconate oxidation.
Agric. BioI. Chem., 28,819-27.
Werneke, J. M., Sligar, S. G. & Schuler, M. A. (1985). Development of broad-
host-range vectors for expression of cloned genes in Pseudomonas. Gene, 38,
73-84.
334 v. Delil, D. Sunil & D. VlaIil
1 INTRODUCnON
ATP has the chemical structure shown in Fig. 1, and a molecular weight of
507·2. The absorption maxima are 257nm at pH2 and 259nm at pH7, with
the molecular extinction coefficients of 14·7 and 15·4, respeCtively. It is rather
stable in neutral solution but unstable in acidic solution; 60% of phosphorus is
liberated by heating at 100°C for 10 min in 1 M HCl. The breakage of P-O
bonds at y and fJ positions produce free energy of 7-8 kcal (30-40 kJ).
ATP and its related compounds can be determined quantitatively by several
methods.
A convenient paper chromatographic method is as follows. The sample on
filter paper is developed in a solvent system containing 5 vol. of iso-butyric
acid and 3 vol. of 0·5 N ammonium hydroxide. Each compound on paper is
detected under UV irradiation. Absorbing parts are cut off and eluted with
5 ml of 0·1 M hydrochloric acid for about 15 h at room temperature. Absor-
bancy at 260 nm is measured by a spectrophotometer and the amounts of
compounds are calculated from standard curves. Since inosine and AMP are
developed on the same position with this solvent system, these compounds
were calculated by measuring absorbancy at 250 and 260 nm.
High voltage paper electrophoresis using two kinds of buffers, 0·1 M borate,
pH 9·4, and 0·2 M acetate, pH 3·5, is carried out at 3 kV for 30-60 min.
Detection and determination of nucleic acid-related compounds are the same
as the paper chromatography.
High performance liquid chromatography is carried out with a Hitachi
liquid chromatograph system 655 under the following conditions: column,
M&S PACK C18 (4·6 mm i.d. x 150mm); mobile phase, 0·2M potassium
phosphate buffer, pH 6·0 (solvent A), and methanol (solvent B); gradient, 2%
of solvent B (0-6 min) and 10% of solvent B (6-20 min); flow rate,
1·0 ml/min; injection volume, 5 Ill; detection, absorbancy at 260 nm. Figure 2
demonstrates the elution pattern of a reaction mixture to phosphorylate
adenosine by a methylotrophic yeast, which will be described below.
ATP and ADP are determined by enzymatic methods using
o 0 0
II II II
HO-PY-O-Pf3-O-p a -O-H2 CS' 0
I I I
OH OH OH
OH OH
Fig. 1. Chemical structure of ATP .
Microbial Production of A TP 339
::
o II 10 111 20
min
Fig. 2. High performance liquid chromatography of A TP and its related compounds in
the reaction mixture.
of inorganic phosphorus and the release of the inhibited glycolysis in high salt
concentrations by AMP.
This method was improved to permit easier cell preparation, and by adding
higher concentrations of the substrate (Tanaka & Hironaka, 1972). They
obtained 15·7 g of ATP·Na2 from 10 g of adenosine in 1·2 liters of the reaction
mixture after treatment of the yeast cells with a surfactant, Cation S, and the
addition of manganese chloride.
Further improvement of the productivity was studied in detail by
Tochikura's group. High amounts of ATP (202 mM, 102·5 g/liter) and ADP
(37 mM, 15·8 g/liter) were obtained at 35-h reaction at 28°C in the reaction
mixture consisting of 300 g dried yeast cells per liter, 0·4 M adenosine, 1 M
inorganic phosphate and 1·5 M glucose initially and with feeding of 0·6 M
inorganic phosphate and 0·5 M glucose at 3-h incubation periods.
The mechanism of the A TP production by this method is as follows:
By adding one-and-a-half times equation (3) to the sum of equations (1) and
(2), equation (4) is derived:
adenosine + 1· 5 glucose + 3Pi ~ A TP + 3ethanol + 3C02 (4)
Screening to find an enzyme source other than baker's yeast has also been
carried out. The AMP-phosphorylating activity was searched for among 59
strains of 17 genera of yeasts, and Debaryomyces nilssonii and Hansenula
jadinii were found to be good ATP producers. These yeasts could phosphory-
late 80-100% of AMP (lOmM) to ATP. It was also found that a hydrocarbon-
assimilating yeast, Candida sp., possessed the ability to form A TP from AMP
in high yield; the activity was markedly affected by the culture conditions for
cell production.
The mechanism of phosphorylation was further studied with respect to
the hexokinase isozyme (Kimura et al., 1980). Dried cells of Hansenula jadinii,
that had been cultured aerobically with acriflavine, contained three hexokinase
isozymes and metabolized glucose to produce A TP in the presence of a high
concentration of inorganic phosphorus. The cells cultured aerobically without
acriflavine contained two hexokinase isozymes and could not metabolize
glucose under the same conditions. Two of the isozymes of the yeast cultured
with acriflavine were similar to isozymes of the yeast cultured without
acriflavine. The third isozyme was resistant to a high phosphorous concentra-
tion. Therefore, the isozyme was considered to be responsible for regeneration
of ATP through glycolysis and phosphorylation of nucleotides in the A TP-
Microbial Production of A TP 341
producing system.
Recently, genes of hexokinase and glucokinase of Saccharomyces cerevisiae
were cloned and the yeast cells harbouring the hybrid plasmid could
phosphorylate glucose about three times as effectively as the original strain
(Fukuda et al., 1984). These cells completely converted adenosine (130 mM)
into A TP, consuming glucose rapidly by the enhanced glycolytic pathway.
The yeast cell method could also be applied to the production of CfP, UTP
and GTP from the corresponding mononucleotides in high yield.
There has been extensive work on bioreactors which use immobilized enzymes,
resting cells and living cells as catalysts (Kennedy, 1987).
A TP regeneration systems so far studied are classified into four categories;
chemical synthesis, whole cells (mainly yeasts), organelles, and cell-free
enzymes. Chemical synthesis was eliminated because it uses an excess of
342 Y. Tani
by the yeast extract which was added to prevent inactivation. Another role of
the yeast extract was to act as an energy-rich substance which is necessary to
initiate the reaction.
ATP for reaction (1), adenylate kinase, is initially supplied from the
endogenous pool in the protoplast and then by reaction (2), oxidative
o H;o2 If. \ (. \
2 ~ NAD+ NADH NAD+ NADH
Microbody
H20
ATP
Fig. 3. ATP-producing system from AMP by methylotrophic yeast.
346 Y. Tani
Adenine CH 3 0H
Adenosine
1
J\
HCHO
1
AMP
\ NATyto'~H
ADP
H' cO 2
ATP ADP O2
H'"
Fig. 4. ATP-producing system from adenosine or adenine by methylotrophic yeast.
Table 1
Free Energy Changes of Reactions Involved in A TP
Production from Adenosine
Reaction - Ll.G(kcallmol)
Table 2
ATP Production by Biochemical System
System Raw material A TP(g/liter) Efficiency * Molar yield (%)
Culture of Adenine, 1·57 20
B. ammoniagenes glucose
Cell of Adenine, 10·9 0·10 83
B. ammoniagenes glucose
Cell of yeasts Adenine, 102·5 0·34 50
glucose
Cell of E. coli AMP, 14·0 0·14 92
glucose
Chromatophore of ADP, 0·4 2·0 80
R. rubrum light
Chromatophore of ADP, 200·0
R. capsulata light
Cell of blue- ADP, 0·17 56·0 17
green alga light
Acetate kinase ADP, 10·0 100
acetyl-P
Adenylate kinase ADP 0·24 1·3 60
Cell of AMP, 30·0 1·5 60
C. boidinii methanol
Adenosine, 100·0 4·0 77
methanol
Adenine, 4·1 0·20 40
methanol
* ATP produced (g) by 1 g of dry cell weight, protein or chlorophyl.
produced by 1 g of dry cell weight, protein or chlorophyll, and molar yield are
compared. The highest amount is obtained by yeast cell methods using
glycolysis (substrate level phosphorylation) and oxidative phosphorylation.
The efficiency of A TP productivity for cells is more than 10 times higher in the
latter than that in the former. The efficiency is quite high in the system using
photophosphorylation, but the amount of A TP produced by these methods is
quite low and the substrate needed is ADP.
REFERENCES
Asada, M., Yanamoto, K., Nakanishi, K., Matsuno, R., Kimura, A. & Kamikubo, T.
(1981). Long term continuous ATP regeneration by enzymes of the alcohol
fermentation pathway and kinases of yeast. Eur. J. App!. Microbiol. Biotechnol.,
12, 198-204.
Fujio, T. & Furuya, A. (1983). Production of ATP from adenine by Brevibacterium
ammoniagenes. J. Ferment. Techno!., 61,261-7.
Fujio, T. & Furuya, A. (1985). Effects of magnesium ion and chelating agents on
enzymatic production of ATP from adenine. Appl. Microbiol. Biotechno!., 21,
143-7.
350 Y. Tani
Fukuda, Y., Yamaguchi, S., Hashimoto, H., Shimosaka, M. & Kimura, A. (1984).
Cloning of glucose phosphorylation genes in S. cerevisiae by the KU-method and
application to ATP production. Agric. BioI. Chem., 48,2877-81.
Kennedy, J. F. (Ed.) (1987). Enzyme Technology, Vol. 7A, Biotechnology, eds. H. J.
Rehm & G. Reed. VCH, Weinheim, FRG.
Kimura, A., Tatsutomi, Y., Fukuda, H. & Morioka, H. (1980). Effect of acriflavine on
the hexokinase isozyme pattern of a yeast, Hansenula jadinii. Biochim. Biophys.
Acta, 629,217-24.
Lutwak-Mann, C. & Mann, T. (1935). Uber die Verkettung der chemischen Umsetz-
ungen in der alkoholichen Garung. I. Mitteilung: Bildung uns Spaltung der
Adenosintriphosphorsaure und deren Zusammenhang mit den Vorgangen der
Zuckerspaltung. Biochem. Z., 281, 140-56.
Ostern, P., Baranowski, T. & Terszakowec, J. (1938). Uber die Phosphorylierung des
Adenosins durch Hefe und die Bedeutung dieses Borgangs flir die alkoholische
Garung. II. Metteilung, Hoppe-Seyler's Z. Physiol. Chem., 251,258-84.
Shimosaka, N., Fukuda, Y. & Kimura, A. (1981). Application of plasmid to ATP
production by E. coli. Agric. Bioi. Chem., 45, 1025-7.
Tanaka, A. & Hironaka, J. (1972). Studies on enzymatic production of ATP. Agric.
Bioi. Chem., 36,867-9.
Tanaka, H., Sato, Z., Nakayama, K. & Kinoshita, S. (1968). Production of nucleic
acid-related substances by fermentative processes. Part XV. Formation of ATP,
GTP and their related substances by Brevibacterium ammoniagenes. Agric. BioI.
Chem., 32,721-6.
Tani, Y. & Yonehara, T. (1985). ATP production from adenosine or adenine by a
methanol-utilizing yeast. Candida boidinii (Kloeckera sp.) No. 2201. Agric. BioI.
Chem., 49, 637-42.
Tani, Y., Mitani, Y. & Yamada, H. (1982). Utilization of Cccompounds: phosphoryla-
tion of adenylate by oxidative phosphorylation in Candida boidinii (Kloeckera sp.)
No. 2201. Agric. Bioi. Chem., 46, 1097-9.
Tani, Y., Mitani, Y. & Yamada, H. (1984a). ATP production by protoplasts of a
methanol yeast, Candida boidinii (Kloeckera sp.) No. 2201. Agric. BioI. Chern., 48,
431-7.
Tani, Y., Mitani, Y. & Yamada, H. (1984b). Preparation of ATP-producing cells of a
methanol yeast, Candida boidinii (Kloeckera sp.) No. 2201. J. Ferment. Technol.,
62,99-101.
Tani, Y., Yonehara, T., Mitani, Y. & Yamada, H. (1984c). ATP production by
sorbitol-treated cells of a methanol yeast, Candida boidinii (Kloeckera sp.) No.
2201, J. Biotechnol., 1, 119-27.
Tochikura, T., Kuwahara, M., Yagi, S., Okamoto, H., Tominaga, Y., Kano, T. &
Ogata, K. (1967). Fermentation and metabolism of nucleic acid-related compounds
in yeasts. J. Ferment. Technol., 45, 511-29.
Vandamme, E. J., Vanloo, J., Machtelinckx, L. & De Laporte, E. (1987). Microbial
sucrose phosphorylase: fermentation process, properties and biotechnical applica-
tions. Adv. Appl. Microbiol., 32.
Yamaguchi, S., Fukuda, Y., Shimosaka, M. & Kimura, A. (1984). Phosphorylation of
AMP to ATP by dried Escherichia coli B cells, J. Ferment. Technol., 62,29-33.
Yonehara, T. & Tani, Y. (1986). Comparative studies on ATP production from
adenosine by a methanol yeast. Agric. BioI. Chem., 50,899-905.
Yonehara, T. & Tani, Y. (1987). Highly efficient production of ATP by a methanol
yeast, Candida boidinii (Kloeckera sp.) No. 2201. J. Ferment. Technol., 65,255-60.
Yonehara, T. & Tani, Y. (1988). ATP production by a methanol yeast, Candida
boidinii (Kloeckera sp.) No. 2201: effects of sorbitol treatment and zinc on cell
structure as to ATP production. Agric. BioI. Chem., 52,909-14.
Chapter 19
ADENOSYLMETHIONINE, ADENOSYLHOMOCYSTEINE
AND RELATED NUCLEOSIDES
S. SHIMIZU & H. YAMADA
1 INTRODUCTION
2 CHEMISTRY
The methyl group of the methionine unit in the molecule of Ado Met is
activated by the positive charge on the adjacent sulfur atom, which makes it
very reactive. Thus, AdoMet is used as a methyl donor in versatile biological
Adenosylmethionine, Adenosylhomocysteine and Related Nucleosides 353
3 BIOSYNTHESIS
Amino acid modifications Chemical structures of each amino acid moiety (R) are as shown below left
NH2
~JyN~
':~ ~
HO OH ~
§.
Accumulated in Candida utilis when grown with D-methionine (Schlenk et al., 1978); I:j.
S-Adenosyl-D-methionine I:
CH 3 shows methyl donor activity in some transmethylation systems (Nakamura & Schlenk, Roo
• I 1976) ;t:
HOOC-CH-(CH 2)2-S-CH2-
I + ~
NH2 ~
~
S-Adenosyl-L-ethionine Accumulated in baker's yeast (Schlenk et al., 1965) and molds (Kusakabe et al., 1974)
CH 2-CH3 when grown with L-ethionine
I
HOOC-CH-(CH 2)2-S-CH2-
I +
NH2
S-Adenosyl-2-methyl-DL-methionine Accumulated in Candida utilis when grown with 2-methyl-DL-methionine (Schlenk et
CH] CH3 al., 1978); shows methyl donor activity in some transmethylation systems (Nakamura
I . I & Schlenk, 1976)
HOOC-C-(CH 2)2-S-CH2-
I +
NH2
S-Adenosyl-o-homocysteine Synthesized chemically and shows inhibitor activity in some transmethylation reactions
• (Borchardt & Wu, 1974)
HOOC-CH-(CH2h-S--CH2-
I
NH2
;...
S-Adenosyl-L-cysteine Synthesized by lupin seed and beef liver AdoHcy hydrolases (Guranowski & ~
::r
Jakubowski, 1983)
HOOC-CH-CH 2-S--CH 2- ~
I
NH2 I§.
Sinefungin Isolated from Streptomyces griseoius as an antifungal antibiotic (Hamill & Hoehn, s·
1973); a potent inhibitor of some transmethylation reactions (see Ueland, 1982) .';..."
HOOC-CH-(CH2h-CH-CH,-
I 1-
NH2 NH2 ~
~
A9145C A metabolite of Sinefungin (Hamill & Hoehn, 1973); a potent inhibitor of some s::
transmethylation reactions (see Ueland, 1982) ~
HOOC-CH-(CH2h-CH-CH= ~
I I '"~
NH2 NH2 s·
5' -Deoxy-5' -S-isobutyl-thioadenosine §'"
Chemically synthesized as an antiviral agent (Hildesheim et ai., 1971); an inhibitor of >:>..
H2N~
homocysteine, the ribose moiety is modified as shown.
,
~
CH-CH 2-CH 2-S--CH z base
HOOC/ ~
U\
""
(continued) U\
HO OH
....
VI
Table l-contd. 0"1
S-Formycinyl-L-homocysteine Synthesized by beef liver, lupin seeds and bacterial AdoHcy hydro lases (Guranowski et
at., 1981; Shimizu etal., 1984a, b; Shimizu & Yamada, 1988)
NH z
N~N\
~NJ--C~
S-Nebralinyl-L-homocysteine Synthesized by beef liver, lupin seeds and bacterial AdoHcy hydro lases (Guranowski et
at., 1981; Shimizu et at., 1984a, b; Shimizu & Yamada, 1988)
H
~
N~N
~
§.
~NJ-) N·
I:
R-
S-Tubercidinyl-L-homocysteine Synthesized by beef liver, lupin seeds and bacterial AdoHcy hydrolases (Guranowski et
aI., 1981; Shimizu et aI., 1984a). A potent inhibitor of various transmethylation ~
NH z reactions (Coward et at., 1974; also see Ueland, 1982) ~
(j) I
N N
S-2-Chloroadenosyl-L-homocysteine Synthesized by bacterial AdoHcy hydrolase (Shimizu et aI., 1984a, b; Shimizu &
Yamada, 1988); 2-Chloroadenosine shows a potent inactivator activity toward
AdoHcy hydrolase (Shimizu et aI., 1984a; see also Ueland, 1982)
NJ:-N
CIAN~~
S -It'-Methyladenosyl-L-homocysteine A potent inhibitor of some transmethylation reactions (Hoffman, 1978); Synthesized by
H-N-CH 3 liver, lupin seeds and bacterial AdoHcy hydrolases (Hoffman, 1978; Guranowski et
at., 1981; Shimizu et at., 1984a)
ex)
S-l-Methyladenosyl-L-homocysteine Synthesized by bacterial AdoHcy hydrolase (Shimizu et al., 1984a)
HC NH z
~
, "~~N~ f}
5
"""NJ-N ~
S-Inosyl-L-homocysteine Synthesized by beef liver, lupin seeds and bacterial AdoHcy hydrolases (Guranowski et i
OH ai., 1981; Shimizu et al., 1984a) ";;.
5·
;:s
s·
-"
7JyN~ ~
"""NJ-N ~
S-3-Deazaadenosyl-L-homocysteine Synthesized by beef liver and lupin seeds AdoHcy hydrolases (Guranowski et al., 1981); ~
s:
3-Deaza-adenosine is a potent competitive inhibitor of AdoHcy hydrolase
(Guranowski et al., 1981) ~
~~ ~
..
~
N~C~~ ~.
S-2-Aza-3-deazaadenosyl-L-homocysteine Synthesized by beef liver and lupin seeds AdoHcy hydrolases (Guranowski et ai., 1981)
~
~
NH z Ei""
~
1:1.
..,~
~:X) ~
N..,C ~
S-8-Azaadenosyl-L-homocysteine Synthesized by beef liver and lupin seeds AdoHcy hydrolases (Guranowski et al., 1981)
.~
NH z
N~\ w
(continued) ~
~NJ...N
~
00
Table l---contd.
S-8-Aminoadenosyl-L-homocysteine Synthesized by beef liver and lupin seeds AdoHcy hydrolases (Guranowski et al., 1981)
NH2
N~N
~~TJ- )-NH2
N N
S-(5' -Deoxyadenosine N'-oxide- 5') Synthesized by beef liver, lupin seeds and bacterial AdoHcy hydrolases (Guranowski et
-L-homocysteine aI., 1981; Shimizu et al., 1984a, b; Shimizu & Yamada, 1988)
~
NH2
~
O'-NJ)-N) §.
N·
I:
~N~ R-
S-Pyrazomycynyl-L-homocysteine Synthesized by beef liver and lupin seeds AdoHcy hydrolases (Guranowski et al., 1981); ~
pyrazomycin is a potent inactivator of AdoHcy hydrolase (see Ueland, 1982) ;;,<
o
H2N~N~ l
T~~'/;
HO C
S-Aristeromycinyl-L-homocysteine Synthesized by beef liver and lupin seeds AdoHcy hydrolases (Guranowski et al., 1981);
aristeromycin is a potent competitive inhibitor of AdoHcy hydrolase (Guranowski et
NH2 al.,1981)
HOCH 2
0:) ~
HO OH
Adenosylmethionine, Adenosylhomocysteine and Related Nucleosides 359
':Hz
lY)
....2
¥l>+r%-~-CH~
Hj:. 0
N~N,; C~2./ S-Adenosylmethylthlopropylamlne
~_J--N ~ l
j--J~ 1<H3
I«l OH
ATP
~- -~-CHz-CHr~
~,
HO OH
N:.JLN
S-Adenosylmethionine ~N~I)
HJC-!i-Di2
~
¥,
l?=~
~
HO OH
I
Methylthloadenoslne
HOOC'CH-CHz-CHrS-CH6
L
Adenine lIethylthlor!bO$e-1P
HOOH
,._~ Methionine S-Adenosylhomocysteine L s-Ribosylhomocysteine
5-AdenOSYUJomlysreine hydrolase
CH3- Homocysteine
~
~ Adenosine
,.
\ Ribose
,'
I
AdoHcy and the enzymatic degradation of AdoHcy is inhibited (de la Haba &
Cantoni, 1959). The accumulated AdoHcy blocks some metabolically impor-
tant AdoMet-dependent transmethylation reactions, because AdoHcy is a
potent inhibitor of many AdoMet-dependent transmethylases (Usdin et al.,
1979; Borchardt et al., 1986). Furthermore, AdoHcy hydrolase is an adenosine
binding protein (Hershfield & Kredich, 1978)\ Various adenosine analogs
including adenosine have been shown to inactivate this enzyme irreversibly. As
the result of this inactivation, the cellular level of AdoHcy increases, which
also results in inhibition of transmethylation reactions. Thus, the hydrolysis of
AdoHcy by AdoHcy hydrolase is a key step for the regulation of the activated
methyl cycle and AdoHcy itself is thought to be a key regulator compound for
transmethylation reactions (Cantoni & Chiang, 1980; Ueland, 1982; Borchardt
et al., 1986).
The route which yields adenine and S-ribosyl-L-homocysteine for the
degradation of AdoHcy is present in only few bacteria, especially in Entero-
bacteriaceae (Duerre, 1962; Miller & Duerre, 1968; Shimizu et al., 1984a,
1988). This route involves the following two reactions: cleavage of glycosyl
linkage of AdoHcy by AdoHcy nucleosidase (EC 3.2.2.9), yielding adenine
and S-ribosyl-L-homocysteine, and hydrolysis of S-ribosyl-L-homocysteine to
L-homocysteine and ribose by S-ribosylhomocysteine hydrolase (EC 3.3.1.3)
(see Fig. 1).
3.2. Decarboxylation of AdoMet
Decarboxylation of AdoMet is a key step in the biosynthetic pathway for
spermidine and spermine. This reaction is catalyzed by AdoMet decarboxylase
(EC 4.1.1.50) (Tabor & Tabor, 1984). The decarboxylated AdoMet, S-
adenosylmethylthiopropylamine, donates its aminopropyl group to putrescine
to form spermidine, or spermidine to spermine in the presence of
aminopropyltransferases. On transfer of the aminopropyl group, the decar-
boxylated AdoMet itself is converted to MeSAdo, which has been demonstr-
ated to be a potent inhibitor of the aminopropyltransferases and AdoHcy
hydrolase (Ueland, 1982; Schlenk 1983). The inhibition by MeSAdo is relieved
by the enzymatic cleavage of MeSAdo to adenine and 5-deoxy-
methylthioribose-l~phosphate by MeSAdo phosphorylase (EC 2.4.2.28).
Thus, this enzyme, as well as AdoHcy hydrolase, plays an important role in
regulation of the activated methyl cycle. This enzyme activity has been
detected in the organisms having AdoHcy hydrolase, but not in the bacteria
having AdoHcy nucleosidase instead of AdoHcy hydrolase. In such bacteria,
MeSAdo is hydrolyzed to adenine and ribose by AdoHcy nucleosidase
(Shimizu et al., 1988) (see Fig. 1).
3.3 AdoHcy Hydrolase
As mentioned above, AdoHcy hydrolase is the most important enzyme for the
regulation of the activated methyl cycle. Hydrolysis of AdoHcy by this enzyme
Adenosylmethionine, Adenosylhomocysteine and Related Nucleosides 361
Table 2
Properties of AdoHcy Hydrolase of A. faecalis·
Mol. wt.
Sedimentation equilibrium 280000
Gel filtration 290000
HPLC 290000
SDS-PAGE 48000
Sedimentation velocity (S20.w) 9·79
Number of subunits 6
pI 4·7
Km (mM) AdoHcy 0·014
adenosine 0·050
DL-homocysteine 1·0
Vm (Ilmol/min per mg) Hydrolysis 546·5
Synthesis 7·47
Optimum temperature (0C) 50
Optimum pH Synthesis 8·0-10·5
Heat stability CC) 45
pH stability 6·0-10·0
COOH-teiminal amino acid Tyr
Other substrates Formycin A, neburalin,
adenosine N' -oxide,
tubercidin, inosine, etc.
Inhibitor Adenosine 5' -carboxylate,
5'-deoxyadenosine, etc.
Inactivator Arabinosyladenine, 2'-
deoxyadenosine,5'-deoxy-
adenosine, adenosine, etc .
• See Shimizu et al. (1984a) for details.
362 S. Shimizu & H. Yamada
mass (55 (00). The function of NAD+ in the reversible reaction was estab-
lished by Palmer & Abeles (1979). The prokaryotic enzyme efficiently
catalyzes thioether formation between L-homocysteine and several novel
nucleosides, most of which are known as antibiotic, antiviral or antitumor
agents (for chemical structures, see Table 1), yielding the corresponding
S-nucleosidyl-L-homocysteines. This fact suggests that the enzyme can be used
as the catalyst for the production of not only AdoHcy but also these novel
nucleoside derivatives. The enzyme is a characteristic adenosine binding
protein as well as mammalian enzymes. Properties of the bacterial enzyme are
summarized in Table 2.
4 PRODUCTION OF ADOMET
Since AdoMet was first prepared from L-methionine and A TP with a liver
preparation containing methionine adenosyltransferase (Cantoni, 1953), there
have been several reports for the preparation of Ado Met. Schlenk and
co-workers (Schlenk & DePalma, 1957; Schlenk et ai., 1965) pointed out that
several common yeasts, such as Saccharomyces cerevisiae and Candida utiiis,
might be suitable sources of AdoMet. They reported that AdoMet is
accumulated in growing or washed cells in the presence of relatively large
amounts of L-methionine. Several molds were also reported to be suitable
sources of AdoMet (Kusakabe et ai., 1974). However, the cellular contents of
AdoMet were not satisfactory in these micro-organisms.
i o
•U100 0
~
"a 0
QI
"- 0 0
§
:8
c 0
•
~ •
c
g• 50 aa
-;
0
u
•
~
0 /j.
"a
C /j. A
~
i
16 0 0.05 0.10
Methlon"'e adenosyltransferase activity
(pmo1/30 m .../mg)
12
'l
i 5 ._ ..\ ' - / / .. _... _., ............. _.,... -..., ............_._.-
r
i 0"",
--
0'1 /
E 10 50 0\ / E
/i\ Y- 0'1
Qj
::.: ~ E
10
;,<~\
0 I '0
'0 c
« I 0
8 I 40 / 0 ..c
~ / W
a a /
::2
OJ
u
l:
6 ~30 /
£/ /
~I
x E /
/
Vl
QI
L.. ..c / 300 ~
u
0 I
Q
3 / A-·A, C
4 ~ 20 A"'/ "'A
U
I
'., .f
0'1
L..
0 0
::2 I /
200
OJ I I '\
u
~
I I ~
2 10 I i A C
C A' 100 0u
0 I OJ
;§ 9
)e _e_e",
L
0
0 0 ~
0 2 3 4 5 6 7
Time (days)
Fig. 3. Typical time course of AdoMet production by S. sake K-6. S. sake K-6 was
cultivated in a lO-liter bench scale fermentor with 5 liters of a medium containing 10%
sucrose, 1% yeast extract, 1·8% urea, 1% L-methionine, 0·2% glycylglycine, 0·4%
KH2P04, 0·01 % MgS04·7H20, biotin (2,Ltg/ml) and 10 ml of a mixture of metal salts
and trace elements (20 g of CaCI2·2H20, 0·5 g of MnS04'5H20, 0·5 g of FeS04·7H20,
1·0 g of ZnS04'7H20, 0·01 g of CuS04·5H20, 0·01 g of CoCI2·6H20, 0·01 g of H 3 B03 ,
0·01 g of Na2Mo04 and 0·01 g of KI in 1 liter H 20), pH 6·3. Cultivation was carried out
at 28°C with aeration at 1 vvm and agitation at 300 rpm. Ethanol (50 ml) was added to
the medium at the time indicated by arrows in the figure. MgS0 4·7H20 (0·25 g) was
further added to the medium on the 3rd day of the cultivation. See Shiozaki et al.
(1986) for details.
364 S. Shimizu & H . Yamada
without any decrease in intracellular level of AdoMet. They also pointed out
that addition of glycylglycine, biotin and CaCl2 to the culture medium
increases cellular AdoMet content (Shiozaki et al. , 1986). Under the optimal
culture conditions in a lO-liter bench scale fermentor , S. sake K-6 produced
lO·8 g/liter of AdoMet (Fig. 3). Almost all the AdoMet was accumulated in
vacuoles in the cells (Fig. 4), the extracellular concentration being very low.
The maximum AdoMet content of cells reached 260 mgt g dry cells; this
compares favorably with the commercially available yeast cells rich in AdoMet
which contain only about 20 mg/g solids.
Leakage of AdoMet from vacuoles leads to its rapid decomposition to
MeSAdo and it is important to retain vacuolar integrity and high concentra-
tions of Ado Met in order to produce satisfactorily large amounts of the
nucleosides. Although it is not yet clear why sake yeasts show notably high
ability to accumulate AdoMet, one possible explanation is suggested by the
data shown in Fig. 2, where methionine adenosyltransferase activities are
compared among various kinds of yeasts. The results reveal a positive
correlation between accumulation of Ado Met and the enzyme activity.
Another interesting characteristic of this yeast is that it is able to grow well in
the presence of high concentrations of L-methionine. L-Methionine has been
reported to markedly inhibit the growth of common Saccharomyces yeasts
(Takahashi & Takahashi, 1968). This may also be an important characteristic
for the high accumulation of Ado Met.
Another approach to produce AdoMet was reported by Gross et al. (1983) ,
who synthesized 7·8 mmol of AdoMet on incubation of a reaction mixture
(850 ml) containing 20 mmol of AMP, 0·5 mmol of A TP, 40 mmol of phos-
phoenol pyruvate, lO mmol of MgCI2 , 150 mmol of KCl, 20 mmol of L-
Adenosylmethionine, Adenosylhomocysteine and Related Nucleosides 365
L-Methlonine
acids (i.e. salts with citric acid, tartaric acid, malic acid, succinic acid, etc.),
salt with ascorbic acid, and cyclodextrin complexes of AdoMet+HSOi,
AdoMet+HSOi . (H2 S04 )n , etc. Among these, the p-toluenesulfonate salts and
cyclodextrin complexes seem to be considerably stable. The double salt with
p-toluenesulfonic acid and H 2S04 has been used to treat various liver diseases
in Italy and some other countries.
Another problem recently revealed is that most commercially available
AdoMet products contain 10-20% of the inactive ( + )-enantiomer, the source
of which is unknown (Stolowitz & Minch, 1981). Matos et al. (1987) reported
that the ( - )-enantiomer is the sole product of the enzymatic reaction with E.
coli methionine adenosyltransferase. This suggests that conversion of the
( - )-enantiomer to the ( + )-enantiomer might take place during isolation or
storage.
5 PRODUCTION OF AdoHcy
not be utilized as the substrate. To overcome this, they screened for microbial
strains capable of synthesizing AdoHcy from D-homocysteine. Several
Pseudomonas strains were found to have this ability. When washed cells of P.
putida were used as the catalyst, AdoHcy synthesis from adenosine, and each
of the three forms of homocysteine, i.e. L-, D- or DL-homocysteine, proceeded
at the same reaction rate. In each case, the product was the L-form of AdoHcy.
They demonstrated that, although the P. putida AdoHcy hydrolase as well as
the A. faecalis enzyme catalyzes only the condensation of adenosine and
L-homocysteine, the bacterium also produces a racemase which catalyzes the
interconversion of the D- and L-homocysteine isomers but not that of the
AdoHcy produced. For practical purposes, A. faecalis cells are still superior to
P. putida cells in that they show high AdoHcy hydrolase activity and low
adenosine deaminase activity. In the mixed cell system containing the two
strains, the reaction rate greatly decreased when the content of P. putida cells
was increased. Therefore, they used P. putida cells just as a source of the
racemase. With 200 mM adenosine and 200 mM DL-homocysteine, the molar
conversion to AdoHcy was 86%, when a mixture of A. faecalis cells
(36 mg/ml) and P. putida cells (4 mg/ml) was used as the catalyst (Table 3).
The reaction mechanism for this system is considered to be as shown in Fig. 6
(Shimizu & Yamada, 1984; Shimizu et al., 1986b).
Table 3
AdoHcy Synthesis with Varying Concentrations of Adenosine and L- or DL-
Homocysteine with Cells of A. faecalis or Mixed Cells of A. faecalis and P. putida
AdoHcy
hydTo/ase
(A.faecalls)
L-Homocysteine + Adenosine ~ " AdoHcy
:
1:R.cemase
l(P.putlda)
D-Homocysteine
Fig. 6. Reaction mechanism for the synthesis of AdoHcy from adenosine and
DL-homocysteine.
5.2 Chemical Methods
Two routes have been used generally for the chemical synthesis of AdoHcy.
One involves (1) protection of the 2'- and 3'-hydroxyl groups of adenosine
using an isopropylidene protecting group; (2) activation of the nucleoside
5'-position by formation of the 5'-tosylate; (3) condensation of the intermedi-
ate 5'-tosylate with S-benzyl-L-homocysteine; and (4) removal of the
isopropylidene protecting group using dilute acid (Baddiley & Jamieson, 1955;
Borchardt & Wu, 1974). The other route involves replacement of the
5'-hydroxyl group of the adenosine by chlorine followed by the condensation
of the resulting 5'-chloro-S'-deoxyadenosine with L-homocysteine (Borchardt et
al., 1976). Condensation of unprotected adenosine with a protected homocy-
steine in the presence of trialkyl phosphine was also reported (Serafinowski,
1985). These standard procedures have been used for the synthesis of a broad
spectrum of AdoHcy analogs.
o 1 2 3 4 5 6 7 8 9 10 11
Number of repeat
Fig. 7. Synthesis of S-nucleosidylhomocysteines with K-carrageenan entrapped cells of
A. faecalis. See Shimizu & Yamada (1988) for details. Adenosine (Ado), formycin A
(For), adenosine N1-oxide, (Ado-N-oxide) and 2-chloroadenosine (Cl-Ado) were used
as nucleoside substrates.
7 CONCLUSION
ACKNOWLEDGEMENT
REFERENCES
1 INTRODUCTION
The name 'coenzymes' has been given to a group of dissociable, low molecular
weight active components of enzymes which catalyze transfer of chemical
groups, hydrogen or electrons. Coenzymes, defined in this way, usually
function as cosubstrates for enzymes to couple two otherwise independent
reactions. Thus, they can be regarded as transmitters of metabolites. In a
wider sense, any catalytically active, low molecular weight, non-protein
components of enzymes are also included in the coenzyme group. This
definition includes coenzymes which are covalently bound to enzymes as
prosthetic groups. In this case, the coenzyme and the apoenzyme together
make up the active holoenzyme.
Many coenzymes are biosynthesized from vitamins and also contain a
nucleoside (or nucleotide) moiety in their molecules. The relationships of some
coenzymes to vitamins and metabolic functions are listed in Table 1. Some
coenzymes are produced on an industrial scale and are used as pharmaceuti-
cals. For example, ATP, COP-choline, FAD, NAO, coenzyme A, S-
adenosylmethionine, coenzyme Q1O, etc., are used for muscular dystrophy,
metabolic diseases, clouding of consciousness, cerebral surgery, hepatic or
renal dysfunction and so on. They are also used for analysis and research in
clinical chemistry, biochemistry, metabolism and pharmacology. Because
several coenzymes are already mentioned in the preceding individual chapters
describing the corresponding parent vitamins, this chapter outlines a few of the
remaining ones. A previous review on coenzymes (Shimizu & Yamada, 1986)
seen from a biotechnological point of view may be useful for further
understanding.
373
w
Table 1 -...I
.;..
Classification, Metabolic Functions and Sources of Coenzymes
:j
Ul
376 S. Shimizu & H. Yamada
2 NICOTINAMIDE COFAcrORS
3 PYRROLOQUINOLINE QUINONE
Table 2
Systems for Regeneration of NAD(P)H from NAD(P)"
a Modified from Chenault & Whitesides (1987). G-6-P, glucose-6-phosphate; 6-PG, 6-phosphogluconate.
Other Vitamin-related Coenzymes 379
systematic search has been reported. POO has been shown to have a
growth-promoting effect on several bacteria (Ameyama et al., 1984a; Shimao
et al., 1984).
Several bacteria have been reported to excrete PQO into their culture
medium. Especially, methylotrophic bacteria seem to be favorable sources for
PQQ production. Reported fermentation yields range from 0·01 to 10 ,ug/ml
(Ameyama et al., 1984b). From a culture filtrate (1100 liters) of
Hyphomicrobium X grown with 0·5% methanol, a yield of 1385 mg POO was
obtained through the procedures involving absorption with Amberlyst A21
anion exchanger and elution from it (Duine et al., 1987).
4 COENZYME 0
Coenzyme Q (CoQ, ubiquinone) is a 2,3-dimethyl-5-methyl benzoquinone
with an isoprenoid side chain composed of dihydroisoprene units. There are
various types which are designated according to the number of isoprene units
in the side chain: Co06 , COQ7, CoOs, Co0 9. COQIO, etc. CoQ6 -COQIO are
widely distributed (the name 'ubiquinone' is from its ubiquitous occurrence).
Co0 1 -CoQs were found in several bacteria and yeasts, such as Rhodospirillum
rubrum, Escherichia coli and Saccharomyces cerevisiae (Friis et aI., 1966;
Daves et aI., 1967; Bentley & Meganathan, 1987). The occurrence of CoO u ,
C00 12 and Co0 13 was also reported (Natori & Nagasaki, 1979, 1980). Among
these homologs, CoO IO has been extensively studied as to practical production,
because it is the natural coenzyme in humans and it is used in therapeutic
treatment for diseases of such tissues as heart and muscle.
An ethionine-resistant mutant derived from Agrobacterium sp. has been
reported to be a potent producer of CoO lO . This mutant produced two types of
morphologically distinguishable colonies, rough and white, and smooth and
yellow, on an agar plate. Only the former showed excellent production
(4·1 mg/ g dry cells), but changed easily to the latter with lower productivity
(2·8 mg/g dry cells) during cultivation or on transfer. The concentration of
ammonium ions in the production medium and aeration or agitation during the
cultivation were found to be important factors affecting cellular COQlO content.
Under optimum cultivation conditions, a value of 5·1 mg/g dry cells or
211 ,ug/ml of culture broth was attained with this mutant (Kuratsu et al.,
1984a, b, c). A facultative methylotroph, Protaminobacter ruber, produced
1· 52 mg/ g dry cells on cultivation with methanol as sole carbon source for
growth. In this case, no other CoO homologs than COQlO were detected in the
cells (Natori et aI., 1978). Paracoccus denitrificans (Matsumura et al., 1983)
and tobacco cells (Ikeda et al., 1981) have also been shown to be potent
producers of CoO lO .
Microbial production of several CoO homo logs other than COQlO has also
been studied. An n-alkane-utilizing yeast, Candida tropicalis, produced
4·7 mg COQ9/g dry cells (52,ug/ml of culture broth) on cultivation in a
medium containing n-alkanes with successive addition of p-hydroxybenzoate as
380 S. Shimizu & H. Yamada
REFERENCES
B. BRUECKNER, D. BLECHSCHMIDT
Friedrich-Schiller-University lena, Dept. of General Microbiology, DDR-6900
lena, Neugasse 24, GDR
&
G. SEMBDNER, G. SCHNEIDER
Institute of Plant Biochemistry, Academy of Sciences of the GDR, DDR-4050
Halle, Weinberg 3, GDR
1 HISTORICAL SURVEY
Gibberellins (GAs) exhibit a bifacial nature, on one hand they are products of
the secondary metabolism in fungi, especially Gibberella fujikuroi, on the
other hand they are biologically active, endogenous hormones in the higher
plants. From the chemical point of view, gibberellins represent a group of
diterpenoids having a typical tetracyclic ring system that, according to the
nomenclature (Rowe, 1968) is called ent-gibberellane (Fig. 1).
The trivial nomenclature attributes gibberellins to a numbered A series
(GAl ... GAx). This system was originally dedicated to the plant hormone
status of gibberellins, because in addition to having the typical structure and
the natural occurrence, the gibberellins have to be biologically active in certain
bioassays (MacMillan & Takahashi, 1968). Currently, the group of gibberellins
(A-series) comprises altogether 72 different GAs; 26 of them were isolated
from fungal sources (Bearder, 1980; Muromtsev & Agnistova, 1984; Takahashi
et al., 1986). Also 3-0-acetyl-GA3 was found to be a genuine product of the
fungus (Schreiber et al., 1966). The structures and some physical data of fungal
gibberellins are compiled in Table 1.
Gibberellins fall into two groups: ~o-gibberellins, which possess the
complete diterpenoid skeleton, and C 19-gibberellins, which have biogenetically
lost the C-20 resulting in a 19,1O-y-lactone structure that is typically for GAs
with high bioactivity. Other inevitable features of physiologically important
GAs are the 6-carboxy function and the 16(17) methylene group.
The most prominent representative GA is GA3 (=gibberellic acid, ent-
3£t', 10, 13-trihydroxy-20-nor-gibberella-1(2), 16(17)dien-7 ,19-dicarboxy-19,1O-
lactone) because of its high abundance in microbial fermentation but also
because of its high bioactivity.
Due to the structural diversity and the manifoldness of functional groups
GAs cover a broad range of polarity, e.g. hydrophilic and lipophilic properties.
HOJ11y[},~~
(dec.)
CH, COOH 2
GA 2 o H
C'9H 26 0 0 350 255 +12 0
HO~OH
(dec.)
@
GA, C'9H 22 0 0 346 234-236 +920
(dec.)
~o nn OH
HO - H CH 2
CH, COOH
@
GA. C'OH24O, 332 214-216 _3 0
(dec.)
co
HO 2 (H
H ,
CH 3 COOH
HO· @~
CH,
H
(DOH
-(II
2
202 (dec.)
W
GAo C,oH2.04 316 208-211 -220
,. H CH,
CH, COOH
~O"
GAIO
CH 3 (DOH (H 3
W
GAil C'OH22O, 330 242-244 + 11 0
~O
H CH z
CH, COOH
HOOC""\~
CH,
H
COOH
CH
z
"@
HOOC\""
(H,
H
COOH
CH z
Table l-contd.
HO~
HOO(~ H
(H 3 (DOH
(H,
~ytL,
GA" 0
(H 3 (DOH
#
GAI6 C I9H z.06 348 157-165
=0
H =
(H 3H DOH (H,
®CH,
HOOC CH3 COOH
HO DC
~CH'
CH 3 COOH
HO~CH'
HOOC CH
3
(OOH
@
GA37 C 2oH z6 O, 346 228-232
CO
HO ~ H CH,
CH 3 COOH
(H 3H COOH (H,
366 174-182
(continued)
Table l-contd.
348
Ho.@,.H
~o
HO :; H (H 2
(H 3 (OOH
"4P",
GAs. 348 243-246
4htt"
GA" 364 260-263
H H (H
(H, COOH 2
H 364 am.
HO
(O()-j
~ H
364 147-150
H~(H
~IIIOH
(H, COOH 2
3-0-
~
~H
acetyl- ~O
380 226
GA, AcO" ,,,,OH
H (H_
(H, COOH '
glbbe ri c ac Id
HO @~" (H,
H
(OOH
(H 2
,~~ ~
HO
~
"~O
H
- H
II... OH
(H 2
~
H(k~H
~gO("· H
lu.. OH
(H 2
rn, (OOH ~ (OO~
iso-GA, d icarboxyllc acid
H H
H; Lewis Ac id
0)
OOH 100 D C n C H3
CH 2 = - 0
H
H+
fLCH' n'CH;
b)
H H
c)
U CH, H+
14CH,
OH
Fig. 3. Typical reactions of ring C/D site of gibberellins.
~H3
Ho-~-CH2COOH mevalonic acid
CH 2CH2 0H
ATP---1
ATP-i
ATP-i
J
(opp
~ 3.3'- dimethylallyl pyrophosphate
1PP-i
(Copp geranyl pyrophosphate - - monoterpenes
IPP-1
abscisic acid
farnesyl pyrophosphate ~
- - squalene __ sterols
~I
~
I ~
OPP
geranylgeranyl pyrophosphate
phytoe ne -
,ChlorOPhyuesters
0(. tocopherol
phylloqUInone
plas toquin one
carotenolds
~I ;;
H
,# OPP
copalyl pyrophosphate :::::::::
______ macrocyclic dlterpenes
,H - - ent-isokaurene
!
B
17
18
~c~
HJw~'
H C"' ,"Ho,~".
~3
H
e nt- kaurenol
.. H
Hf "'-(H 2 0H
!
~
(H3 -c."'~
H ent- kaurenal
H -
Hf ~(HO
~ l
~~
',~H
H
c."'~
ent-kaurenoic ac id
-
___.~ ~§H
H : t~7oc. ~ H ~
OH - .
18-dih'Jdroxy-
kaurenolide
Hf ~OOH HOH 2 ( --C 0-0
1
~
H ~(:'0, ~_ CH z
a H OH ent-1x..hydroxy - - -... ~ H
OH ent- 60(,70( -
H3(
!
">~~OOH kaurenoic acid H3 C %.~OOH OH dihydroxy-
kaurenoic acid
Hi
~"" "(OOH
GA" - old.hyd.
ring A occurs when the aldehyde oxidation step is reached; tricarboxylic acids
like GA13 , GA25 , and GA41 are not converted to Cwgibberellins. Feeding
experiments using 14C_ and 13C-Iabelled precursors demonstrated that C 20 is
lost as CO2 (see Graebe, 1987).
The main product of GA biosynthesis in Gibberella fujikuroi is GA3, which
is formed from G~ via GA7 by 1,2-dehydrogenation (G~-+ GA7) and
13-hydroxylation (GA7-+ GA3). The alternative pathway leading through GAl
to GA3 is a minor one and may be due to non-specificity of the dehydrogenat-
ing enzyme. However, in Sphaceloma manihoticola GA4 is the main product
Fungal Gibberellin Production 395
R= H R = OH
erl,
'" GA 12 GA'4~GA2
.D
(\-\,
GAlS GA 3-
(hydroxy acid) ( hydroxy ac id)
R
'"
1
u
1 1
0\,
GA36
GA 25 II \A~GA13 41
GA,s GAS7
and GA7, GA3 and GAl are not formed. In Gibberella fujikuroi 13-
hydroxylation is a major reaction in conversion of Cwgibberellins such as GA4
( -+ GAl) and some subsequent products derived from GA4 by hydroxylation
at position 10 (GA I6 ), 1 (GAS4) and 20 (G~7)' respectively. It is interesting to
note that 13-hydroxylation occurs only at the end of the pathway and is
catalyzed by soluble enzymes, whereas 3P-hydroxylation is microsomal and
occurs at the beginning of the pathway (GA12-aldehyde-+ GAI4-aldehyde).
The oxidation of GAwaldehyde to GA l4 also proceeds in fungal microsomes
(Hedden, 1983). Opposite to the fungus in higher plants two further pathways
starting with early 13-hydroxylation (GAS3) or 3p,13-dihydroxylation (GAlS)
at the ~o-level are operating (see Graebe, 1987).
Both the amount and the type of gibberellins formed by the fungus are
dependent on the genetic constitution of the strain used (see Section 5) and the
fermentation conditions applied (see Section 6). Though rapid progress has
been made in the biosynthetic routes and the enzymes catalyzing conversion
steps, only little is known about the metabolic regulation of GA biosynthesis.
This holds true especially for fungal systems, whereas some more insights have
been gained during the last few years into the physiological and biochemical
control of GA biosynthesis in higher plants (see Graebe, 1987; Sembdner et
ai., 1987).
For the first time the genetics of the fungus Gibberella fujikuroi was
investigated by Gordon, who was occasionally successful in crossing some of
the strains of his large collection. He could show that the fungus was
heterothallic (Gordon, 1960). Physiological-genetic studies on gibberellin
production presented evidence for the presence of two genes that control
different steps in the gibberellin biosynthetic pathway. Genetic studies were
difficult at first because production of perithecia was erratic and a rare process.
A number of natural and synthetic media were tested which would support the
development of the sexual stage of the fungus. Spector & Phinney (1966, 1968)
found a medium containing stems of Citrus medica stimulated perithecial
production. When fungal strains of opposite mating type were grown on stems
of this tree, perithecia were regularly produced within 3-6 weeks. The number
of ascospores present in an ascus varied, a maximum number of eight spores
was observed. Mating between a high and low gibberellin-producing strain
gave tetrads that segregated 2: 2 in terms of the total amount of produced
gibberellins. One of the analyzed tetrads was of special interest, since one of
the four strains produced high amounts of G~, GA7 and GA9 but no GAl
and GA3 (Spector & Phinney, 1968). Genetic studies with this strain resulted
in the identification of two genes, gl and g2. Gene gl is responsible for the con-
trol of the overall gibberellin production, and gl-mutants neither produced
Fungal Gibberellin Production 397
gibberellins nor metabolized those added. A second pair of alleles, gene g2,
controls 13-hydroxylation since g2+ strains produced the whole spectrum of
gibberellins, whereas g2 strains did not synthesize GAl or GA3 but accumu-
lated GA4 and GA7.
As indicated above, the best wild strains such as Gibberella fujikuroi ACC
917 produce about 1 g GA3 per liter of culture filtrate in an optimized medium,
and this was the yield in the early years of GA manufacturing. Current yields
are reported to be many times higher (Martin, 1983). Increased yields and
production rates are the result of improvement in strains and in fermentation
processes. Strain development of gibberellin-producing organisms had received
a great deal of attention. Since most of this research has been done at the ICI
Pharmaceutical Division and other industrial laboratories, it is mostly un-
published because of its proprietary nature. Like other natural products the
techniques included a combination of direct selection for high-producing
strains and mutagenesis by several different agents. Conventional mutation
processes have been published by Soviet workers (Imshenetsky & Ulyanova,
1962; Erokhina & Sokolova, 1966; Erokhina & Efremov, 1970). Mutants with
increased GA-titers of up to 60% were found after UV irradiation, fast
neutrons, and gamma treatments from spores or mycelia in the case of
non-sporing organisms (Erokhina & Sokolova, 1966).
A mutant strain resulting from UV and ethylenimine treatments gave yields
of approximately 2 g/liter in a medium with plant oil as the sole carbon source
(USSR Patent 440408). Another way of increasing yields is the screening of
mutants with blocked carotenoid biosynthesis as a competing biosynthetic
pathway with the same precursor mevalonate. Avalos & Cerda-Olmedo (1987)
found that strains with particularly low basal levels of neurosporaxanthin,
which usually represents about 75% of total carotenoids, are good gibberellin
producers.
In general, Gibberella fujikuroi is a highly suitable organism for the
induction and isolation of mutants. The uninucleate microconidia commonly
allow the expression of recessive mutations (Avalos et al., 1985). Besides
gibberellic acid, the last few years' commercial interest has been concentrated
on the production of GA4 and GA7. Gibberellin A3-producing strains can be
switched over to increased production of the precursors of GAl and GA3 by
increasing the pH value to the range of pH 6-7·5 (Jefferys, 1970). In
g2-mutants the hydroxylation at C-13 is blocked and consequently, relatively
large amounts of GA4 and GA7 could be isolated without separation from GA3
and GAl' Bearder (1983) described such a g2-mutant R-9, which did not
produce GAl and GA3. Mutation and selection for increased product
formation are probably the most important factors in improving the yield of
gibberellins. There is only a small number of publications concerning methods
of parasexual recombination to obtain diploid strains in Gibberella fujikuroi.
Crossing both wild strains and mutants as parents Calam et al. (1973) found
different recombinants, some of which gave increased yields of gibberellic acid,
compared with the already high-yielding parents.
398 B. Brueckner et al.
6.1 Inoculum
combinations of fast and slowly utilized carbon sources (Brit. Patent 919 186;
Darken et al., 1959). Darken et al. (1959) obtained yields of 880 mg gibberellic
acid per liter in a 7-day period using a basal fermentation medium with com
steep liquor, ammonium sulfate, potassium dihydrogen phosphate and a
mixture of glycerol (20 g/liter), glucose (10 g/liter), and lactose (20 g/liter).
Moderate, but economically reasonable yields of gibberellin were produced
on molasses, sulfite liquors, and skimmed milk (FRG Patent 1081402;
Maddox, 1977). The Soviet workers have successfully used plant oils, e.g.
sunflower oil (Muromtsev et al., 1968; Muromtsev & Agnistova, 1984; USSR
Patent 440 408). They compared growth and gibberellin formation on mediums
with adequate amounts of oil and sucrose. The fat was introduced into the
initial nutrient medium as a single additive while sugar was supplemented to
the liquid in fractional doses. In the oil-containing medium significantly more
biomass was formed, and the production phase continued for a long time by
high mycelium productivity (Muromtsev & Agnistova, 1984). Sucrose con-
sumption for the gibberellic acid synthesis (as calculated per carbon) was
2·3-3·6 times higher than oil consumption. The results in Fig. 7 display the
increased yields with plant oil in comparison with sucrose. Besides plant oils,
hydrocarbons (Rehm, 1980) and fatty acids (Muromtsev & Dubovaya, 1964)
have also been tested for their ability to support growth and gibberellin
production.
Very important for the gibberellin fermentation are the quality and quantity
of nitrogen. Favorable nitrogen sources are ammonium sulfate, ammonium
chloride and slowly assimilable sources such as glycine, ammonium tartrate
and natural sources of nitrogen (Jefferys, 1970). The productivity on am-
monium acetate was considerably lower than on the other N-sources. It seems
possible that the stimulation in productivity by natural sources of carbon and
nitrogen such as plant meals, plant oils and com steep liquor could be
attributed to the content of precursors.
In submerged culture significant production of gibberellins starts only at the
time of nitrogen exhaustion. Production was greater the higher the value of
initial nitrogen. Further increases in nitrogen concentrations lead to a decrease
of the mycelium productivity because of the decreasing efficiency of oxygen
transfer (Borrow et al., 1964). Therefore, the selection of both the initial
concentration of nitrogen and of an optimal C/N-ratio is very important.
Besides carbon and nitrogen sources magnesium, potassium, phosphate and
sulphate were all needed. Trace element requirements would be met by
impurities in commercial media (Vass & Jefferys, 1979). For a maximum
production of GA Jefferys (1970) recommended a temperature of 31-32°C for
growth and for product formation 29°C. The airflow-agitation regime should be
as vigorous as possible.
Borrow et al. (1964) showed that the specific growth rate and the gibberellin
yield are fairly constant over the range of pH 3·5-6·5. However, the
composition of the gibberellin mixture produced depends on the pH value. In
a wild-type culture of Gibberella Jujikuroi the normal end-product of the
400 B. Brueckner et aJ.
, I
, a) 0
, <l
0 ~
9
0 I
/ 6 ____6 I
'-...
Ol
2000 <l- 5
.C-
01
5,5 .... ,
,
"\ 6
/~
"en
E
c::;;
55 ~
E
E .- 4 I 5,0
\
\ A/ OJ 45 c::;;
OJ
<{
M
3 I
a.
\
,,/~, ~ /
""
0 0
-'"
l:J 1 000 ;::-3 4,5 V............... \ /,/ -", ./0 35 ~
.--\-~~
Y '---'- -----
u /t:i. c::;;
/ \. / 0----,,0
2 4,0 30 25
.i
0 .y /0 20
-t!;
200 3,5 '0 0 10 15
3 5 7 9 11 13 15 17 19
days
0 .q
"
b) q
,
I
0 I
, '-... '-...
9 2 ODD- <l-5 S,S ""
E 55 E
Ol
->-
.C- "- c::;;
'-...
en Vl
- 4 '.... ./' - . . . x::-:6~::s;:::~:::::::,A_6 OJ
E I 5,0 '--- / ~ 0 0 en 45 c::;;
OJ
). ./A /o~ "-
~ 0
/Y ',
3 I "- 0
<{
M
a. / X
'6 ..... ..- -'"
// /
l:J 1 000 ;::-3 4,5 ./ li '_ . . . . . . . . . . 0 '--- 35 CI
'0
'\ c::;;
2 4,0 A 0 ,--- 3D 25
o 0/ 20
200 3,5 D~o 10 15
3 5 7 9 11 13 15 17 19
day s
Fig. 7. Growth and GA3 production by Fusarium moniliforme (strain F6) with different
carbon sources: (a) medium with glucose (fed batch); (b) medium with plant oil (after
Muromtsev & Agnistova, 1984).
at., 1964). It is well known that the Q~ of the fungus decreases when the cells
switch to storage phase metabolism and the oxygen transfer capacity is no
longer saturated. That is why the initial nitrogen concentration, optimum
medium composition and nutrient feeds must be selected in such a way that
some further proliferation and, therefore, a mixed linear growth and storage
phase occurs after the balanced phase (Jefferys, 1970).
The gibberellin fermentation has a long history of development. Early trials
of the Japanese workers were based on surface cultivation, and low yields
(40-60 mg/liter) after prolonged periods of incubation were reported (Rehm,
1980).
In shake-flask experiments with the same media 200 mg/liter GA3 was
obtained. Optimization of medium composition, especially of the initial
nitrogen and carbon concentrations, led to a further increase of GA yield up to
1 g/liter. The early gibberellin production, prior to 1961, used a purely batch
technique. Later fermentations are batch-fed processes (Vass & Jefferys,
1979). From this time onward the feed regimes have been developed using
natural and synthetic carbohydrates and nitrogenous nutrients in the feedstock.
The carbohydrate feed rate also must be carefully restricted to limit the storage
activity of the fungus without lowering the rate of GA production. In the
patent literature it was said that some precursor feeds, e.g. mevalonic acid,
kaurene and kauranol, increased the yield of gibberellin A3 (Brit. Patent
957634). However, it seems unlikely that the obtained increases have any
economic interest because of the high cost of these compounds.
Holme & Zacharias (1965) and Bu'Lock et at. (1974) described the
production of gibberellin A3 in glycine-limited continuous culture conditions
and found maximum gibberellic acid synthesis at a growth rate of 0·005 h- 1 •
But because of the little specific rates of synthesis and the long period of
fermentation the chemostate culture has a theoretical importance only.
Besides submerged fermentation techniques employed throughout the
world, solid state fermentation is known to produce the metabolites in most of
the cases at high yield, especially if molds are involved (Lindenfelser &
Ciegler, 1975). First information on the cultivation of the gibberellin-producing
fungus on com grains was published by Focke et at. (1967). Indian workers
compared the production of GA3 by solid state fermentation and submerged
fermentation. Based on adequate carbohydrate contents they found that in the
case of solid medium (wheat bran) the accumulation of GA3 was 1·6 times
higher (Kumar & Lonsane, 1987). The degree of biomass formation varied
between 9·7 and 11·9 g/kg commercial wheat bran, the maximum quantity of
GA3 accumulated at the end of a incubation period of 7-8 days was 1·2 g/kg
solid substrate.
Up to now, there is no information on scaling up the extension of the solid
state fermentation technique for technical purposes in GA3 production on an
industrial scale. Probably this is due to the problems with the process control,
especially with oxygen transfer and moisture control on an enlarged scale.
Fungal Gibberellin Production 403
The total cost of a fermentation process comprises many factors, namely the
cost of raw materials, fixed costs, costs for separation and purification of the
product and so on. Process improvements, especially increase of the yield,
include strain improvement, optimization of the medium composition and
extension of the production phase. Conditions which lead to increases in the
overall rate of production, for example, by decreasing the percentage of
by-products, are also very important.
Vass & Jefferys (1979) calculated the changes in production data over a
period of 18 years. The process development obtained resulted in the
introduction of new strains, transition from a purely batch technique to
batch-fed process and modifications in medium compositions (Fig. 6.2). Only
those process changes which have shown a cost reduction or yield improve-
ment in laboratory experiments were implemented.
As can be seen in Fig. 8 the producing strains of Gibberella fujikuroi (A-D)
were developed during this period. For example, the introduction of strain D
shows an improvement in volumetric production rate of 35% in comparison
with strain C. The authors defined the term volumetric production rate as the
150
100
I
E
i
50 §
§ ~
dsrlilll
~
§
E
§
E I
'1=
Medium
Strain
Fig. 8. Changes in productivity of gibberellic acid up to 1976. Relative values have
been calculated from process data of Imperial Chemicals Industries. Horizontally
striped histograms indicate yield (mol), open histograms production rate (mol liter- 3 ) ,
and vertically striped histograms volume-specific production rate (molliter3 (t + tr )-I),
where tr denotes the time for turn-round (after Vass & Jefferys, 1979).
404 B. Brueckner et al.
average rate of change of concentration of product with time, where the time
includes the period for vessel cleaning and preparations for new fermentation.
On the other hand, in the period from the introduction of strain D 1970 to
1976 an improvement in the volumetric production rate of more than 60% was
achieved by variations of the medium components in both the initial batching
and in the feed-liquor and by control of the supply of assimilable nitrogenous
nutrients for the growth period. The feed regime is also controlled during the
course of fermentation to nearly maintain a state of oxygen depletion (Vass &
Jefferys, 1979).
It is interesting that although the total fermentation costs per batch have
increased by 50% during these 18 years, the unit production costs decreased
with the increase of gross yield per batch up to 1700%.
Vass & Jefferys (1979) have mentioned that at the end of fermentation a
greater percentage of available carbohydrates from the feed was metabolized
to cell storage components rather than gibberellins. The decrease of these
losses of expensive components seems to be a target for further improvements.
During the fermentation the gibberellins are secreted into the culture medium.
Therefore, the fermentation of Gibberella fujikuroi normally results in the
production of a mixture of gibberellins among which GA3, GA4 and GA7
predominate. GA3 is the most abundant gibberellin.
Prior to isolation of gibberellins, the mycelium should be separated from the
broth by means of filter press, centrifuge, or drum filter with a filter aid.
Afterwards, the gibberellins can be prepared from the broth by adsorbing on
activated charcoal, by various types of ion exchange processes or by liquid-
liquid extraction. But it is very difficult to describe the processes in use on a
large commercial basis because these processes are kept as company secrets
and are not published. This is why the product recovery and purification of
gibberellins can only be described by a general discussion of the kinds of
technology available and published in literature and in patent specifications.
The adsorption on active carbon (e.g. charcoal) was the first method for the
extraction of GA3 developed by Japanese researchers. Probably it is also used
in commercial processes. After the adsorption the elution of the gibberellins is
carried out by percolation with water-miscible solvents, e.g. methanol or
acetone. The solvents may be used pure or mixed with ammonia (Jefferys,
1970).
However, this method has some disadvantages:
(a) it requires large amounts of solvents;
(b) the complete recovery of the gibberellins from the carbon is not
possible;
(c) the carbon can be used just once;
(d) the purification is poor.
Fungal Gibberellin Production 405
resins by the use of synthetic sorption resins with non-ionic character and with
unpolar or middle-polar surfaces of the type Amberlite XAD-4, XAD-2 or
XAD-7. Gibberellins are caught quantitatively by resins of this type, whereas
the impurities and by-products are bound only to a negligible extent. The
gibberellins may be eluted by aqueous solutions of acetone, methanol or
ethanol. A further removal of impurities is possible by the choice of selective
elution media. The advantage of this procedure is based on the fact that the
solvent mixtures for the desorption of the gibberellins and of the impurities are
identical. They only differ in their concentrations. For example, the desorption
of the gibberellins is carried out with 35-80% methanol and with acetone
>70% or <45%, respectively; by comparison the desorption of the impurities
is possible by elution with 80-95% methanol and 45-70% acetone.
The third way of gibberellin recovery is the liquid-liquid extraction of the
culture filtrate with water-immiscible solvents. These methods were developed
as the yields in the fermentation broths were increased. The solvents mostly
used are esters such as ethyl acetate or butyl acetate and ketones such as ethyl
methyl ketone or methyl isobutyl ketone. However, n butanol or esters can be
used either alone or in mixtures. Ethyl acetate appears to be used as solvent on
plant-scale processes. Acid conditions (pH 2-4) are needed for the extraction.
The subsequent recovery of the gibberellins from the solvent is effected by
adsorption on solid sodium or potassium bicarbonate or by the buffer-solvent
processes being based on the relative solubilities of the free acid in solvent and
of the sodium or potassium salt in an aqueous phase. The extraction
efficiencies described in the literature varied between 40 and 100% (Jefferys,
1970). An important disadvantage of the liquid-liquid extraction is the large
demand for solvents and their recovery.
Definitely, it can be judged that the cost for each of the three procedures of
the recovery described is much higher than the fermentation cost itself because
the yields of the recovery processes on the large commercial scale are still
relatively low. The recovery cost is the main reason for the expensive price of
the product. The procedures mentioned up to now relate to the separation of
the gibberellins from the culture medium. But the culture broth contains a
mixture of gibberellins amongst which GA3, G~ and GA7 predominate. It is
desirable to separate the gibberellins not only from the culture broth but also
from each other because the gibberellins GA3 and G~/GA7 have different
plant gJ,:owth-regulatory effects, and, thus, the compounds have distinct and
specific practical applications.
The separation of GA3 and a mixture of G~/GA7 is possible by a
liquid-liquid extraction which is based on the different partition coefficients of
GA3 and G~/GA7 between certain organic solvents and water within a
particular pH range. For example, the culture filtrate is extracted with an
organic water-immiscible solvent (preferably ethyl acetate) at a pH value
between 4 and 8·5. The result is an extract rich in G~/GA7 and an aqueous
broth rich in GA3. The precipitation of G~/GA7 from the solvent is difficult
because other substances inhibit this process. The leI patent EP 83 306 682
Microorganism Manufacturing process Product recovery Product
Gibberella fujikuroi
(Saw.) Wr.
~ MY',li.m ~P7'ti., i : Mycelium I
= Fusarium moniliforme
(Sheld.)
8 ASSAY METHODS
8.1 General
Table 1
Experimental Conditions for Fluorimetric Analysis of
some Fungal Gibberellins (AE, Exciting Wavelength, AF,
Fluorescence Wavelength)
8.3 Bioassays
Table 3
Sensitivity of GA Bioassays
Assay Minimum GA3 dose Range of linear
concentration log dose-response
giving significant
response
Dwarf rice 0·1 ng/plant 0·1-100 ng/plant
'Tan-ginbozu'
microdrop variant
Root application 10- 8 _10- 7 mol/liter 10- 7 _10- 5 mol/liter
Dwarf maize--d 1 5 ng/plant 5-10,Ltg/plant
Dwarf pea--cv. 'Meteor' O· 5-1·0 ng/ plant 1 ng-lO ,Ltg/plant
Lettuce bypocotyl- 5 x 10- 7 mol/liter 5 x 10- 7 _10- 4 mol/liter
cv. 'Arctic'
a-amylase; barley 10- 10_10- 9 mol/liter 10- 9 _10- 7 mol/liter
endosperm
Fungal Gibberellin Production 411
Response
o
. \1/ 10 I 00 \
logOose
I
1000
-111''-
0,7~X1~1.1
,/ ~
----~I~/-'-__+_I -+1--
I \ 1 ~~~~~tati
Lp 0n
70 100110
70~X2~110
Fig. 10. Schematized log dose-response curve. Deviation ranges (Yl/Xl; Y2/X2) pre-
sented in logarithmic and linear scale.
use as a bioassay. However, only three types of bioassays have become widely
used due to their ease of performance, reliability, sensitivity and range of
response. These are the dwarf mutants, the hypocotyl and the a-amylase
bioassays. The very extensive literature dealing with GA bioassays including
complete descriptions of their performance is covered by a number of reviews
and monographs (see Bailiss & Hill, 1971; Reeve & Crozier, 1975, 1980;
Graebe & Ropers, 1978; Hoad, 1983; Bergner, 1988; Bernhardt, 1988).
Some dwarf mutants of various plant species are known to be deficient in
endogenous gibberellins and, therefore, can be normalized by exogenous GA.
Such GA-sensitive mutants are favoured bioassay objects. Most important are
special mutants of Zea mays L., Oryza sativa L. and Pisum sativum L., the
dwarfism of which is controlled by a single recessive gene. The GA sensitivity
of those mutants is due to a block in GA biosynthesis (see Section 3)
preventing the formation of GAl necessary for normal stem growth. GA
biosynthesis can be blocked either at a very early stage (d 5 mutant of maize, dx
mutant of rice and na mutant of pea) or immediately before GAl formation (d l
mutant of maize, dy mutant of rice and Ie mutant of pea). Mutants belonging to
the first type respond to a broad spectrum of different gibberellins, and the
second type is sensitive only to GAs having a 3f3-0H group, like GAb GA3 ,
G~, GA7 , etc. Dwarf rice bioassays are widely used either with a microdrop
or root application technique. Both dx-mutants (cv. 'Tan-ginbozu') and
dy-mutants (cv. 'Katake-Tamamishikai' and 'Waito-C') are applied. Rice
seedlings are pre-germinated and-after application of the test solution-
412 B. Brueckner et al.
8.4 Immunoassays
9 BIOLOGICAL PROPERTIES
The bakanae disease syndrome in rice plants, including the typical overgrowth
symptoms, led to the discovery of gibberellins as the active metabolites of the
pathogenic fungus Gibberella fujikuroi (see Section 1). Apparently, the
gibberellins are part of the substantial interactions between the parasite and
the host and, therefore, might have some functional role in disease develop-
ment. However, there is only one piece of evidence that gibberellins have
some physiological significance in Gibberella fujikuroi. Nakamura et al. (1985)
found a promotive effect of GA3 on conidial germination and elongation of
414 B. Brueckner et al.
activities or they are inactive in animal systems. The acute toxicity on animals
is very low. The tolerance limit to GA3 is much higher than to antibiotics; for
mice it is given as 6300 and 25 ()()() mg/kg after intravenous and oral
application, respectively (Schwartz et al., 1983). The acute toxicity for
G~/GA7 on mice is also>500mg/kg (Abdel-Rahman et al., 1977). Neither
in animals nor in tissue cultures were teratogenic or carcinogenic effects
observed over a period of years. Pharmacological studies have been done with
GA3 using predominantly mice, rats and guinea-pigs. Among the results
reviewed by Schwartz et al. (1983), there are stimulating effects on the body
weight in various animals, influences on endocrinologic functions and on the
functions of the macrophaga and changes in mitotic activities as well as
protective effects against pathogenic infections, toxicants and X-rays.
10 CHEMICAL SYNTHESIS
10.2 Conversion
Most of the endogenous plant GAs occur in traces only. Thus the microbiolog-
ical source of GAs is of high importance in order to get access to substantial
amounts of these substances. Commonly the main fungal GAs like GA3, G~
and GA7 act as starting material for preparing CwGAs, and GAl3 for
Cw-GAs. Figure 11 compiles the main routes in GA conversion, for references
see Takahashi & Yamaguchi (1983). Sometimes a single step like hydrogena-
tion leads easily to another GA (GA3~ GAl> GA7~ G~). Other conver-
sions need a complex sequence of reactions, e.g. GA3~ GA29 •
10.3 Labelling
In principle, many of the reactions used in conversion of GAs (see above) are
suitable for introducing stable or radioactive labels into GAs (see Takahashi &
Yamaguchi, 1983). For example, hydrogenations with deuterium or tritium
lead efficiently to labelled analogues. A widely used method for double
labelling of GAs consists of oxidatively removing the exocyclic methylene
group followed by refixing it with 13C, 3H2 -Wittig reagent.
Sometimes biotransformation by Gibberella fujikuroi or plant enzymes are
included in modifying initial products. Other possibilities for introduction of
3H are the catalytic hydrogen exchange with GA3-type gibberellins (Nadeau &
Rappaport, 1974) or the 6-H exchange on the 6-aldehyde level (Lischewski,
1985).
atoms like N, e.g. as l-azido-, l-amino-GAs (Voigt & Adam, 1982) or amino
acid conjugates (Adam et ai., 1977). Physiologically relevant glucosyl conjug-
ates of GAs have also been prepared. Thus a series of GA-O-glucosides and
GA-glucosyl esters were synthesized (see Schneider, 1983).
From the 72 gibberellins only gibberellic acid (GA3) and, to a lesser extent,
mixtures of GA,. and GA7 have found practical use. Sometimes it is difficult to
calculate the financial return to the grower for the application of gibberellins.
For example, climatic effects, environmental factors and differences in varietal
response can effect or veil the efficiency of the hormones applied so that a
reliable calculation of the economic value is connected with many elements of
uncertainty. Moreover, it has been difficult to gather accurate information
separating actual commercial use of gibberellins from wishful thinking or
potential uses.
The gibberellins are the most widely used group of native plant hormones.
In 1980 the world production of gibberellins for commercial use was estimated
to be 12-15 tons (Martin, 1983). Countries with a good tradition in the
production of gibberellins are the USA (Abbott Laboratories, Eli Lilly), the
UK (Imperial Chemical Industries) and Japan (Takeda Chemical Industries).
Further countries producing more or less large amounts of gibberellins are the
USSR, Hungary, Poland and the People's Republic of China. Well-known
trade names are Pro-Gibb, Promalin, Gib-Sol, Gib-Tabs, Activol, Berelex,
Auxilin, Gibersib and Gibreskol. The preparations often contain GA3 and
small quantities of other gibberellins (At, A 4 , A7 and the artefact iso-GA3)
together with spreader activators, effervescent mixtures or desiccants. At
present, practical uses of gibberellins are most extended in the brewing
industry, in increasing both yields and quality of grapes and in stimulating the
growth of sugar cane.
In the routine commerical malting of barley the use of GA3 is a widespread
method because the development of malt is a costly, time-consuming step in
brewing beer. Normally, malt is developed by weighing a certain quantity of
barley seed into a tank where germination occurs. During this process en-
dogenous gibberellins passing from the germinating embryo to the aleurone
layer of the seed induce the synthesis of hydrolytic enzymes, especially
a-amylase and diastase. These enzymes degrade endosperm food reserves such
as cell wall carbohydrates, storage protein and a limited quantity of starch. The
resulting liquor is then drained from the steep for further processing. Small
additions of GA3 accelerate the production and release of the enzymes which
degrade the hemicellulosic-protein-starch complex into essential brewing ma-
terials such as sugar, peptides and amino acids. Today the average steeping
and germinating time amounts to 7-10 days. It can be shortened to 1-3 days
by the addition of GA3. In commerical use GA3 is applied as an aqueous
Fungal Gibberellin Production 419
set, and to treat frost damage of pear (Martin, 1983). In the UK a mixture of
GA3 and naphthalene acetic acid is registered for use on Cox's Orange Pippin
applies to improve fruit set and yield (Looney, 1979). In the UK, the
Netherlands and Canada GA3 is recommended for use on rhubarb. The
treatment with the hormone breaks the dormancy in some early cultivars and
increases stick yield from plants emerging from dormancy in all cultivars.
GA3 is also effective in breaking dormancy of potatoes. The treatment with
GA3 is particularly useful in climates where two potato crops are grown in one
season (e.g. in India), by breaking the dormancy of seed potatoes saved from
the first crop. Moreover, GA3 hastens shoot emergence and increases the
number of stems produced and daughter tubers formed (Rappaport, 1980).
The ability of GA3 to stimulate flowering and to increase the flower number
in a wide range of vegetable species has been utilized in the production of seed
from poor-bolting lettuce cultivars. GA3 (10 ppm) applied at 4- and 8-leaf stage
increased the seed yield about 4-7-fold compared with the unsprayed and not
beheaded control (Martin, 1983)_
Yields of certain cultivars of hops are improved and the number of cones are
increased up to 35% after spraying with GA3 (Palmer, 1974).
In Mediterranean countries and in South Africa GA3 is used commercially
for the production of globe artichokes. By spraying the hormone onto young
plants it is possible to increase the head number per plant and to advance the
harvest by several weeks, especially at low temperatures which limit bud
growth. In such a way total crop yields can be increased by up to 40% without
affecting earliness (Thomas, 1985).
Besides these uses, on a large commercial scale numerous possibilities of
GA3 application are known in crops with a limited market size: GA3 can be
used for the early harvest of chicory, cabbage, mint, parsley and spinach, for
the increase of the yields of beans, celery, endives, peppers, watercress and
tomatoes under glasshouse conditions, for the seed production of carrots,
celery and onions, for the improvement of the germination of celery and
onions, for a better fruit set of blueberries and plums, for the increase of the
fruit size of lemons and melons and for the induction of parthenocarpic
development of many fruits such as black and white currants, figs, straw-
berries, apples, pears and tomatoes (Martin, 1983).
In addition to GA3, the gibberellins GA4 and GA7 are going to reach a
greater utilization. It is supposed by numerous experts that their importance
will rise further in the near future, because the physiological potencies of GA 4,
GA7 and GA3 differ from each other in some important respects. For example,
GA4/GA7 have the remarkable ability to modify the sex expression of
cucurbits. Treatment of gynoecious cucumbers with GA4/GA7 (e.g. ProGibb)
can enhance male flower formation, and this phenomenon can be used
commercially for producing pollen parents in gynoecious cucumber fields
(Rappaport, 1980).
In celery seeds, thermodormancy can be broken by GA4/GA7 treatment
Fungal Gibberellin Production 421
(Biddington & Thomas, 1978). This procedure has some importance for the
UK with a celery production of about 41000 tons (Thomas, 1985).
In the USA the russet incidence of the apple cultivar 'Golden Delicious' is
reduced by spraying with GA4/GA7 (Meador & Taylor, 1987). In Italy five to
six treatments of GA4 applied 8-15 days after full bloom reduced russeting of
the 'Golden Delicious' apple by 70%.
A recent addition for the growth-regulator treatment of apples is a
combination product of a mixture of GA4+ 7 and the cytokinines N6 _
benzyl adenine (Promalin). This treatment increases the ratio of length to
diameter of 'Golden Delicious' apples, resulting in an apple which better meets
market expectation with respect to shape (Looney et al., 1979). In forestry,
promoting influences of GA4/GA7 on the flowering and the seed production of
conifers were observed in field experiments (Rappaport, 1980). For example,
the strobili production in western hemlock (Tsuga heterophylla) and the
flowering in white spruce (Picea glauca) were promoted (Pharis et al., 1986;
Rottink, 1986). These important economic forest trees have a poor natural
regeneration because of the infrequency of good seed years. Thus there is a
need for a reliable method to increase and stabilize seed production. However,
a true commercial scale application will not be possible before sufficient
amounts of GA4/GA7 are available.
Finally, it seems to be worthwhile to mention some outlooks for new and
unconventional fields of gibberellin application. Japanese patents describe the
addition of gibberellins to a cosmetic cream for the decolorizing of freckles
(Japan Patent 58103307 and Japan Patent 58077 808) and to a hair tonic for
the promotion of the growth of hair (Japan Patent 8059906).
By adding GA3 to the food of fattened bulls, the daily increase in weight of
these animals was increased by 18%. The GA3 stimulated the protein,
carbohydrate and fat metabolism (Shamberev et al., 1985). Furthermore, GA3
stimulated the growth of lactic acid bacteria resulting in an increase in biomass
yield of 4-8-fold. This effect of GA3 will possibly gain importance for the
silage of green fodder (Erzinkyan, 1981). It is to be expected that the
commercial interest in the gibberellins will grow further, especially as this
group of compounds has very low mammalian toxicity and occurs naturally in
many vegetable foodstuffs.
In spite of the applications described above, the scale of gibberellins is
relatively low compared with the market of the herbicides, defoliants and
desiccants. Some important factors limiting use of gibberellins are:
(a) the relatively high cost and the low availability, especially for GA4+ 7;
(b) the action over a long period of time in contrast to the herbicides which
are expected to produce a rapid effect;
(c) potential side-effects, such as inhibition of flowering in fruit trees for a
long period of time after treatment;
422 B. Brueckner et al.
12 CONCLUSIONS
REFERENCES
Gancheva, V. & Dimova, Ts. (1984). Biosynthesis of gibberellins. II. Influence of the
quantity and age of inoculum on the biosynthesis of gibberellins from the strain
Fusarium moniliforme IM-11. Acta Microbiol. Bulgarica, 14,74-9.
Gordon, W. L. (1960). Distribution and prevalence of Fusarium moniliforme Sheld
(Gibberella fujikuroi (Saw.) Wr. producing substances with gibberellin-like biologi-
cal properties. Nature, Lond., 18(J, 698-700.
Graebe, J. E. (1987). Gibberellin biosynthesis and control. Ann. Rev. Plant Physiol.,
38,419-65.
Graebe, J. E. & Ropers, H. J. (1978). Gibberellins. In Phytohormones and Related
Compounds--a Comprehensive Treatise, Vol. 1, ed. D. S. Letham, P. B. Goodwin
& T. J. V. Higgins. Elsevier, Amsterdam, Oxford, New York, pp. 107-204.
Hartsuck, J. A. & Lipscomb, W. N. (1963). Molecular and crystal structure of the
di-p-bromobenzoate of the methyl ester of gibberellic acid. J. Am. Chem. Soc., 85,
3414-19.
Hedden, P. (1983). In vitro metabolism of gibberellins. In The Biochemistry and
Physiology of Gibberellins, Vol. 1, ed. A. Crozier. Praeger, New York, pp. 99-149.
Hedden, P. (1986). The use of combined gas chromatography-mass spectrometry in the
analysis of plant growth substances. In Modern Methods of Plant Analysis, Vol. 3,
Gas Chromatography/Mass Spectrometry, ed. H. F. Linskens & J. E. Jackson.
Springer, Berlin, Heidelberg, New York, Tokyo, pp. 1-12.
Hedden, P. (1987). Gibberellins. In Principles and Practise of Plant Hormone Analysis,
Vol. 1, ed. L. Rivier & A. Crozier. Academic Press, London.
Hoad, G. V. (1983). Gibberellin bioassays and structure-activity relationships. In The
Biochemistry and Physiology of Gibberellins, Vol. 2, ed. A. Crozier. Praeger, New
York, pp. 57-94.
Holme, T. & Zacharias, B. (1965). Gibberellic acid formation in continuous culture.
Biotechnol. Bioeng., 7,405-15.
Hori, S. (1898). Some observations on 'bakanae' disease of the rice plant. Mem. Agr.
Res. Sta. (Tokyo), 12, 110-19.
Imshenetsky, A. A. & Ulyanova, O. M. (1962). Experimental variability in Fusarium
moniliforme Sheld leading to the formation of gibberellins. Nature, Lond., 195,
62-3.
Jacobsen, J. V. (1983). Regulation of protein biosynthesis in aleurone cells by
gibberellin and abscisic acid. In The Biochemistry and Physiology of Gibberellins,
Vol. 2, ed. A. Crozier. Praeger, New York, pp. 159-87.
Jefferys, E. G. (1970). The gibberellin fermentation. Adv. Appl. BioI., 13,283-316.
Jensen, E., Crozier, A. & Monteiro, A. M. (1986). Analysis of gibberellins and
gibberellin conjugates by ion-suppression reversed-phase high-performance liquid
chromatography. J. Chromatogr., 367,377-84.
Jones, R. L. & Moll, C. (1983). Gibberellin-induced growth in excised lettuce
hypocotyls. In The Biochemistry and Physiology of Gibberellins, Vol. 2, ed. A.
Crozier. Praeger, New York, pp. 95-128.
Karssen, C. M. & Lacka, E. (1986). A revision of the hormone balance theory of seed
dormancy: Studies on gibberellin and/or abscisic acid-deficient mutants of
Arabidopsis thaliana. In Plant Growth Substances, ed. M. Bopp. Springer, Berlin,
Heidelberg, New York, Tokyo, pp. 315-23.
Kaufman, P. B. & Dayanandan, P. (1983). Gibberellin-induced growth in Avena
internodes. In The Biochemistry and Physiology of Gibberellins, Vol. 2, ed. A.
Crozier. Praeger, New York, pp. 129-57.
Kawanabe, Y., Yamane, H., Murayama, T., Takahashi, N. & Nakamura, T. (1983).
Identification of gibberellin A3 in mycelia of Neurospora crassa. Agr. BioI. Chern.,
47, 1693-4.
Keith, B., Srivastava, L. M. & Murofushi, N. (1979). Agric. BioI. Chem. (Tokyo), 43,
141-3.
Knoefel, H. D. (1988). Immunologische Methoden. In Methoden zur
426 B. Brueckner et aI.
431
432 Index