THE VITAMIN REQUIREMENTS FOR GLYCEROL OXIDATION BY STREPTOCOCCUS FAECALIS P. J.

VANDEMARK1"2
Laboratory of Bacteriology, College of Agriculture, Cornell University, Ithaca, New York

Received for publication January 16, 1950

Gunsalus and Sherman (1943) have shown that oxygen is necessary for glycerol fermentation for the majority of the lactic acid bacteria. These organisms, though considered devoid of those iron enzymes usually necessary when oxygen is involved as a hydrogen acceptor, could aerobically utilize glycerol. Because of the anomaly of these observations Gunsalus and Umbreit (1945) studied the mechanism of glycerol oxidation by Streptococcu faecalis (F24) and reported that it consisted of the following complex of reactions:

(1) (2) (3) (4)

glycerol + ATP -- a-glycerophosphate + ADP a-glycerophosphate + 02 -4 triose phosphate + H202 triose phosphate -2H + ADP -. pyruvic acid + ATP pyruvic acid + 2H -. lactic acid

in which ATP and ADP represent adenosine triphosphate and adenosine diphosphate, and reaction 3 represents a summation of further reactions, essentially the same as those concerned with triose phosphate breakdown in hexose glycolysis. The present paper describes a study of the vitamin requirements for glycerol oxidation by Streptococcus faecalis (F24). This was conducted by means of a vitamin-deficiency technique similar to that used by Bellamy and Gunsalus (1944). As applied to the glycerol oxidation by Streptococcus faecalis (F24), this technique consists of growing the organism on media deficient in the various known vitamins and then determining the ability of such deficient cells to oxidize glycerol.
METHODS AND RESULTS

Culture. An inoculum of Streptococcus faecalis (F24) was prepared by transferring stab cultures of this organism from agar into a medium of 1 per cent each of typtone and yeast extract, 0.5 per cent K2HPO4, and 0.1 per cent glucose. After 10 hours' incubation at 37 C the culture was centrifuged from the medium, aseptically suspended in an equal volume of sterile distilled water, recentrifuged, and resuspended in an equal volume of sterile distilled water. Five-tenths per
1 A large portion of this work was done under the Standard Brands Fellowship in Bacteriology at Cornell University. Grateful acknowledgment is made for this support. 2 The author is indebted to Dr. W. W. Umbreit, Department of Enzyme Chemistry, Merck Institute for Therapeutic Research, Rahway, New Jersey, who gave continual guidance and advice, and in whose laboratory preliminary studies on this problem were conducted. 533

r-3o4

P. J.

VANDEMARK

[VOL. 59

cent by volume of this suspension was used as the inoculum for the experimental media. Cell suspensions were prepared, after 12 hours' growth at 37 C on the experimental media, by centrifuging the culture, then washing the cells once in a halfvolume of distilled water, and resuspending in one-tenth the growth volume of distilled water. These cell suspension were then stored at 4 C for 1 hour, a procedure that tends to inactivate those enzyme systems by which hydrogen may be transported to oxygen from triose phosphate in this strain (Gunsalus and Umbreit, 1945). Manometric techniques. The rate of glycerol oxidation was measured manometrically as previously described by Gunsalus and Umbreit (1945). The rate of oxidation was expressed as the Qo, (N), i.e., the microliters of oxygen taken up per hour per milligram of bacterial nitrogen. The cell nitrogen was calculated from the turbidity of the cell suspension, the linear relationship between the bacterial nitrogen and the density of the cell suspensions having previously been determined. Medium. A typical synthetic medium (table 1) was satisfactory for the growth of Streptococcus faecalis (F24). However, cells grown on this medium oxidized glycerol only slowly. It was found that cells grown on this synthetic medium supplemented with either 0.1 per cent Difco yeast extract or 10 per cent crude strepogenin (Wright and Skeggs, 1944) would oxidize glycerol at an appreciable rate (table 2). However, more highly purified strepogenin (Roberts and Snell, 1946), though stimulating growth, was not satisfactory for the production of cells that would actively oxidize glycerol (table 2). If Streptococcus faecalis (F24) is grown on the completely synthetic medium given in table 1, the activity of these cells on glycerol can then be stimulated by the addition of yeast extract or crude strepogenin to the cells in the Warburg flask, a result that would indicate that the effect of the factor (or factors) was not in apoenzyme formation. The substance (or substances) in yeast extract or crude strepogenin was also found to be dialyzable. It therefore became apparent that a cofactor (or cofactors) not essential for the growth of Streptococcus faecalis (F24) was necessary for the oxidation of glycerol by this organism. For convenience throughout this paper the substance necessary for glycerol oxidation will be referred to as the glycerol factor. The glycerol factor. Attempts were made to set up an assay system for the glycerol factor, but these were not entirely successful and require further study. One can, however, obtain rough estimates of the glycerol factor content of various substances by growing the organism on the base medium with or without purified strepogenin (Roberts and Snell, 1946) and then adding the substance being assayed either to the growth medium or directly to the Warburg flask with the deficient cells. The results of some of these crude assays by the latter method are shown in table 3. It will be noted that various liver fractions were good sources of the factor, yet neither vitamin B12 nor thymidine was stimulatory. Glutathione and histidine showed some ability to stimulate glycerol oxidation by deficient cells, but the amounts required were relatively large, ruling out

1950]

VITAMIN REQUIREMENTS FOR GLYCEROL OXIDATION

535

the posibility that either is actually the glycerol factor. Sodium thioglycolate and ascorbic acid failed to stimulate the glycerol oxidation. Cysteine could not be tested by adding it directly to the Warburg flask as the cysteine was spontaneously oxidized. However by the addition of this amino acid to the growth medium
TABLE 1 Medium satisfactory for growth of Streptococcus faecalis (FS4).
QUANTITY PER LITER

Acid-hydrolyzed casein.................................... Glucose...................................................

K2HPO4................................................... Sodium acetate............................................

Salts B*................................................... Sodium thioglycolate..................................... L-Cystine................................................. DL-Tryptophan........................................... Adenine sulfate............................................ Guanine hydrochloride................................... Uracil..................................................... Thiamine hydrochloride.................................. Nicotinic acid.............................................. Riboflavin ................................................. Calcium pantothenate..................................... Pyridoxamine dihydrochloride............................ Folio acid.................................................. Biotin ..................................................... pH........................................................

10.0 g 1.0 g 5.0 g 2.0 g 10.0 ml 100.0 mg 200.0 mg 200.0 mg 10.0 mg 10.0 mg 10.0 mg 2.0 mg 5.0 mg 1.0 mg 1.0 mg 2.0 mg 0.1 mg 0.01 mg

7.2-7.3

Para-aminobenzoic acid was omitted, since it is slightly inhibitory to this organism. * Salts B - MgSOj.7H,0, 10 g; NaCl, 0.5 g; FeSO407H20, 0.5 g; MnSO4.4H20, 0.5 g; water, 250 ml.
TABLE 2

The effect of various materials on the ability of Streptococcus faecalis to oxidize glycerol
ADDITIONS TO 3AS MEDIUM

(ROWTH

Qo. (N)
151 505 196 436

None ..... ................................. Crude strepogenin ................................ Charcoal-adsorbed strepogenin .................... Difoo yeast extract (10-20 mg/10 ml) ..............

111 145 146 133

it was found that the glycerol oxidation by Streptococcus faecalis (F24) was not stimulated. Other substances tested by being added to the growth medium were inositol, choline, unhydrolyzed Labco casein, Difco trypsin, yeast nucleic acid (Merck), and acid-autoclaved and alkaline-autoclaved yeast nucleic acid, not one of which was a source of the glycerol factor. As neither unhydrolyzed casein nor trypsi is an available source of the glycerol factor, whereas the tryptic digest of casein

536

'P. J. VANDEMARK

[VOL. 59

(i.e., crude strepogenin) is, it would appear that the glycerol factor is present in unhydrolyzed casein in a bound form. The observation that yeast nucleic acid, or acid- or alkaline-autoclaved yeast nucleic acid, fails to stimulate glycerol oxidation would seem to indicate that the glycerol factor is not a nucleic acid or a component of the nucleic acids tested. Also, as neither amino acids, yeast ash, nor known factors are capable of functionng as the glycerol factor, it would seem likely that an unknown factor is concerned.
TABLE 3 Known substances tested for stimulation of glycerol oxidation by resting cells
SUBSTANCE
QUANTITY PK LAS

POTmNCY

Thymidinet ................................

Wilson 10 U.S.P. liver extractt .............. Wilson liver fraction St ..................... Wilson liver fraction Lt ..................... Lilly 15 U.S.P. liver extractf ................ Vitamin B12t ...............................

0.8 2 2 2 0.01 10

mg mg mg mg

pg pg

Yeast ash .................................. Acid-hydrolyzed casein ..................... Glutamine ................................. Glutathione ................................ Histidine ................................. Tomato juice factor concentratest .......... Biocytint (concentrate No. 4698) ...........

i20

Protogent .................................

Ascorbic acid ............................... All known vitaminst........................ * The unit of the glycerol factor is arbitrarily defined as that amount of factor in 1 mg of Difco yeast extract. Potency represents the units of glycerol factor per mg of substance

mg yeast 10 mg 5 mg 5 mg 5 mg 0.2 ml (20 units) 20 pg 0.1 ml (100 units) 5 mg 0.3 ml

2 0.2 0.5 0.9 None None None None None 0.2 0.1 None None None None None

assayed. t Appreciation is expressed to the Wilson Laboratories and to the Eli Lilly Company for the various liver fractions, to Merck and Company for the crystalline vitamin B12, to Dr. W. Shive for the thymidine, to Dr. T. Woods for the tomato juice concentrates, to Dr. L. Wright for the "biocytin," and to Dr. E. Stokstad for the "protogen." $ The vitamin solution used had the following concentration of vitamins per ml: nicotinic acid, 50 pg; riboflavin, 10 psg; calcium pantothenate, 10 ,ug; thiamine hydrochloride, 20 pg; pyridoxamine dihydrochloride, 20 ,g; folic acid, 1 ,ug; and biotin, 0.1 pug.
From its presence in crude strepogenin the glycerol factor might plausibly be a peptide. However, autoclaving Wilson's 10 U.S.P. liver extract with 6 N H2S04 for 30 minutes at 120 C failed to decrease the activity of the liver extract, making it difficult to postulate a peptide. Autoclaving with 2 N NaOH for 20 minutes at 120 C destroys approximately 70 per cent of the activity of the preparations. The factor is very highly water-soluble, and on extracting water solutions with various organic solvents, e.g., n-butanol, the activity remained in the aqueous fraction. The glycerol factor could be precipitated from the aqueous solution of liver extract by the addition of 10 volumes of acetone. The factor was also readily adsorbed on various activated charcoals over a wide range

1950]

VITAMIN REQUIREMENTS FOR GLYCEROL OXIDATION

53-07

of pH. However, an adequate purification procedure for the glycerol factor is not yet available. Known vitamin requirements. In order to determine which other vtamins or coenzymes are involved in glycerol oxidation, the glycerol factor must be provided to the cells in as pure state as possible, that is, free from the other vitamins. By growing the cells deficient in the glycerol factor on synthetic medium (table 1) and then adding 0.5 ml of crude strepogenin to the Warburg flask with these deficient cells, the glycerol factor could be provided relatively free of the known vitamins. Therefore, to grow cells deficient in any given B vitamin, they were actually grown deficient in two factors, the vtamin being studied and the glycTABLE 4

The effect of various deficiencies on glycerol oxidation

TYPE OF CELLS
GROWT

Q0, (N)

REMARK

* With all vitamins .138 Deficient in thiamine .147

1,350 540

Deficient in riboflavin .21 Deficient in nicotinic acid .20 Deficient in pantothenic acid .31 Deficient in biotin .70 Deficient in folic acid .61 Deficient in pyridoxamine .68 With all vitamins plus vitamin B,2 (16 mug/10 m) .138 With all vitamins plus para-aminobenzoic acid (40 pg/10 m) .139
*

228 51

75 765 114 1,360

Deficient cells stimulated with cocarboxylase, boiled extracts of sufficient cells, and sometimes with thiamine Not stimulated Deficient cells stimulated with coenzyme 1, but not nicotinic acid Not stimulated Not stimulated Not stimulated Apparently not involved

1,445
1,340

Apparently not involved

Apparently not involved

All vitamins are the same as listed in the base medium (table 1).

erol factor, and then the deficient cells were provided with the glycerol factor by adding crude strepogenin with the cells in the Warburg flask. Streptococcus faecalis (F24) cells were grown deficient in each of the known vitamins. These deficient and sufficient cells, after being centrifuged from the growth medium, were brought to equal concentrations turbidimetrically and their activity on glycerol determined. The results of this study are shown in table 4. Thiamine, riboflavin, nicotinic acid, pantothenic acid, biotin, and folio acid are all somehow concerned in glycerol oxidation, but deficiencies of vitamins BR, Bu, and para-aminobenzoic acid cause no decrease in glycerol oxidation. In order to determine which of these vitamins are concerned directly as co factors in glycerol oxidation, attempts were made to stimulate the decreawd activities of the vitamin-deficient cells with either the specific vitamin, its co-

co8

P. J. VANDEARKC

[VOL. 59

enzyme form (if known), or a boiled extract of cells grown with sufficient amounts of the vitamin. If marked stimulation of cells deficient in any given vitamin is observed, it would seem highly probable that the vitamin serves a direct function' in the oxidation of glycerol, that is, as a coenzyme. The results of this study are included in the "remarks" column of table 4. Decreased activities due to thiamine deficiency can be stimulated with cocarboxylase or boiled extracts of thiaminesufficient cells, and occasionally by thiamine alone. Coenzyme I, but not nicotinic acid, will stimulate nicotinic-acid-deficient cells. Riboflavin-deficient cells were not stimulated by flavin adenine dinucleotide, boiled extracts of riboflavinsufficient cells, or by riboflavin. Yet riboflavin, as flavoprotein, would seem to be the substance most directly concerned in hydrogen transport in this reaction (Gunsalus and Umbreit, 1945; Gunsalus, 1947). Decreased rates of glycerol oxidation due to deficiencies of pantothenic acid, biotin, or folic acid could not be stimulated. In these cases, though these vitamins are concerned in the oxidation of glycerol, it is not certain whether their function is as a cofactor of the glycerol system or in some less direct manner, such as the formation of one or more of the enzymes of this system.
DISCUSSION

From this study it should be noted that certain factors, though not necessary in the medium for growth, are required in additional amounts for maximum metabolic activity of the organism. For example, Niven and Sherman (1944) have shown that thiamine is not a nutritional requirement of the enterococci. Similarly in these studies thiamine was not required for the growth of Streptococcu.s faecalis (F24), yet for the maxim rate of glycerol oxidation by these cells thiamine is required. Also, this organism will grow satisfactorily without a source of the glycerol factor, yet this factor must be provided for glycerol oxidation. Thiamine is generally believed to function in the decarboxylation of some a-keto acids. However, thiamine's function in glycerol oxidation would not appear to be in the decarboxylation of an a-keto acid, and thus may represent a new site in which the vitamin acts in metabolism. This is further indicated by supplementary data that show that a thiamine deficiency has no effect upon the production of lactic acid from glucose, the metabolic pathway from the triose phosphate stage being the same as that followed when glycerol is oxidized. The -action of nicotinic acid in glycerol oxidation is undoubtedly as coenzyme I or coenzyme II, functioning in hydrogen transport at the glycerol phosphate and the triose phosphate stages. The action of riboflavin is probably as part of flavoproteins concerned in the transport of hydrogen to oxygen, from the glycerol phosphate stage. The function of the glycerol factor remains to be determined. However, its function may be in the breakdown of triose phosphate to lactic acid, as cells deficient in the glycerol factor were slower in the glycolysis of glucose than cells provided various sources of the glycerol factor. The identity of this factor is as yet unknown. It would not seem to be the same as the factor required for the growth of Lactobacillus bulgaricu (William et al., 1949)', as the glycerol factor is

1950]

VITAMUIN REQUIREMENTS FOR GLYCEROL OXIDATION

5 39

present in crude strepogenin and differs in various chemical properties, e.g., water solubility. Also this factor would not appear to be identical with the Leuconostoc citrovorum factor (Sauberlich and Baumann, 1948; Snell et at., 1948), as thymidine is inactive as the glycerol factor. The glycerol factor is apparently not one of the factors that stimulate the growth of Streptococcus faecalis R when this organism is grown from a small inoculum (Ruegamer et al., 1947; Colio and Babb, 1948). Thus pending its purification and a study of its properties the glycerol factor remains unidentified.
SUMMARY

Thiamine, nicotinic acid, riboflavin, and an unidentified factor are apparently concerned as coenzymes in the oxidation of glycerol by Streptococcus faecali8 (F24). This unidentified factor is apparently not a common amino acid, mineral, or known growth factor. Pantothenic acid, biotin, and folic acid are also somehow involved in glycerol oxidation by this organism. Pyridoxamine, vitamin B12, and para-aminobenzoic acid are not essential for glycerol oxidation.
REFERENCES BELLAmY, W. D., AND GuNSALUS, I. C. 1944 Tyrosine decarboxylase. II. Pyridoxinedeficient medium for apoenzyme production. J. Bact., 50, 95-103. Como, L. G., AND BABB, V. 1948 Study of a new stimulatory growth factor. J. Biol. Chem., 174, 405-409. GuNsALus, I. C. 1947 Products of anaerobic glycerol fermentation by Streptococcus faecali8. J. Bact., 54, 239-244. GUNSALUS, I. C., AND SHERMAN, J. M. 1943 The fermentation of glycerol by streptococci. J. Bact., 45, 155-162. GUINSALUS, I. C., AND UMBREIT, W. W. 1945 The oxidation of glycerol by Streptococcus faecalis. J. Bact., 49, 347-357. NIVEN, C. E., AND SHERMAN, J. M. 1944 The nutrition of the enterococci. J. Bact., 47,
335-342. ROBERTS, E. C., AND SNELL, E. E. 1946 An improved medium for microbiological assays with Lactobacillus casei. J. Biol. Chem., 163, 499-509. RuGAmER, W. B., COOPERmAN, J. M., SPoRN, E. M., SNELL, E. E., AND ELVEHJEM, C. A. 1947 Comparative distribution of a stimulatory factor for Streptococcus faecalU R and the monkey antianemia factor. J. Biol. Chem., 167, 861-868. SAUBERLICH, H. E., AND BAUMANN, C. A. 1948 A factor required for the growth of Leuconostoc citrovorum. J. Biol. Chem., 176, 165-173. SNELL, E. E., KiTAY, E., AND McNuiT, W. S. 1948 Thymine desoxyriboside as an essential growth factor for the lactic acid bacteria. J. Biol. Chem., 175, 473-474. WILLIAMS, W. L., HOFF-JORGENSEN, E., AND SNELL, E. E. 1949 Determination and properties of an unidentified growth factor required by Lactobacillus bulgaricus. J. Biol. Chem., 177, 933-940. WRIGHT, L. D., AND SKEGGS, H. R. 1944 The growth factor requirements of certain streptococci. J. Bact., 48, 117-118.