FEMS Microbiology Letters 158 (1998) 61^67

The anaerobic oxidation of hydrazine:

a novel reaction in microbial nitrogen metabolism
Jos Schalk a , Hege Oustad b , J. Gijs Kuenen a , Mike S.M. Jetten a; *
a
Kluyver Laboratory for Biotechnology, Delft University of Technology, Julianalaan 67, 2628 BC Delft, The Netherlands
b
Technical University of Trondheim, Department for Biotechnology, Trondheim, Norway

Received 14 October 1997; revised 28 October 1997; accepted 28 October 1997

Abstract

Hydrazine is rarely found as an intermediate in microbial nitrogen conversions. In this study the conversion of hydrazine by
the anaerobic ammonium oxidation (Anammox) culture, in which hydrazine has been proposed as an intermediate, was
investigated. This study demonstrated the biological nature of hydrazine conversion by the Anammox culture. In batch
cultures with hydrazine it was observed that 3 mol N2 H4 was disproportionated to 4 mol NH‡ 4 and 1 mol N2 . Hydrazine with
nitrite as an electron acceptor showed a conversion of 3 mol N2 H4 and 4 mol NO3 2 to 5 mol N2 , with a specific activity of 5.5
nmol min31 (mg volatile suspended solids)31 . Addition of hydrazine to a biofilm reactor for 80 days showed that it was not
possible to grow Anammox with hydrazine.
ß 1998 Federation of European Microbiological Societies. Published by Elsevier Science B.V.

Keywords : Anaerobic ammonium oxidation; Hydrazine; Nitrogen removal ; Nitrite ; Bio¢lm reactor; Nitri¢cation

1. Introduction bic conditions, hydrazine can be converted to N2 via

Hydrazine is a hazardous compound, which can
be readily oxidized. It is used as a rocket fuel, as
an antioxidant or as an intermediate for the produc-
tion of explosives and pesticides [1]. Hydrazine is a
strong reducing agent, which reacts easily with metal
ions [2]. In the presence of catalysts like Cu(II) and
phosphate ions hydrazine radicals are formed via
a one-electron oxidation, which ultimately are
converted into ammonia and dinitrogen gas
(2N2 H4 C2N2 H3 CN4 H6 CN2 +2NH3 . Under aero-
* Corresponding author. Tel.: +31 (15) 2781193; Fax:
+31 (15) 2782355; E-mail: M.Jetten@STM.TUDelft.NL
a two-electron oxidation (N2 H4 CN2 H2 CN2 ) [1,3].
In contrast to the numerous chemical reactions,
hydrazine is rarely observed as an intermediate in
microbial nitrogen conversions. Hydrazine is pro-
posed to be an enzyme-bound intermediate in the
reduction of dinitrogen gas to ammonium by nitro-
genase [4]. In Nitrosomonas europaea cells, hydrazine
can be used as an external energy source, e.g. for the
de novo synthesis of polypeptides [5]. Under anaero-
bic conditions hydrazine can be used as an electron
donor for the reduction of nitrite to NO and N2 O
[6]. In addition, it was shown that the enzyme hy-
droxylamine oxidoreductase (HAO) from N. euro-
paea was capable of converting hydrazine to dinitro-
gen gas [7,8]. A di¡erent enzyme system, the R2

0378-1097 / 98 / $19.00 ß 1998 Federation of European Microbiological Societies. Published by Elsevier Science B.V.
PII S 0 3 7 8 - 1 0 9 7 ( 9 7 ) 0 0 5 0 1 - 6

FEMSLE 7929 5-1-98

62 J. Schalk et al. / FEMS Microbiology Letters 158 (1998) 61^67
Fig. 1. Possible reaction mechanism for anaerobic ammonium
oxidation. Hydrazine (N2 H4 ) is formed by the oxidation of am-
monium with hydroxylamine. Hydrazine is then oxidized to dini-
trogen generating four reduction equivalents, which can be used
for the reduction of nitrite to hydroxylamine [12].
subunit of ribonucleotide reductase from Escherichia
coli, was shown to catalyze a disproportionation of
hydrazine into dinitrogen gas and ammonium [9].
Recently a novel process was discovered in which
hydrazine occurred as an intermediate. In this proc-
ess, called anaerobic ammonium oxidation (Anam-
mox), nitrite is reduced to dinitrogen gas with am-
monium as the electron donor [10]. Inhibition studies
showed that this process had a biological nature [11].
It was possible to enrich the anaerobic ammonium
oxidizing microorganisms in an medium containing
ammonium, nitrite and bicarbonate as the sole car-
bon source. Based on results from 15 N-labelling stud-
ies a novel pathway for the Anammox process has
been proposed [12]. In this reaction mechanism am-
monium and hydroxylamine are combined to form
hydrazine. Hydrazine itself is then converted to dini-
trogen gas, generating four reduction equivalents. It
is postulated that these reduction equivalents are
used for the reduction of nitrite to hydroxylamine,
as shown in Fig. 1 [12].
Since hydrazine is a strong reductant and was ob-
served as an intermediate in the Anammox process,
we investigated whether the Anammox culture is
able to use hydrazine as a substrate for growth. In
order to gain more knowledge about the physiology
of the hydrazine metabolism in Anammox, the re-
search was focussed on three topics. The ¢rst aim
was to con¢rm that the hydrazine conversions in
Anammox were of a biological nature. The second
topic was to investigate the hydrazine conversions by
the Anammox culture in batch experiments with and
without additional electron acceptors. Finally, the
capability of Anammox to grow on hydrazine in
bio¢lm reactors was investigated.
FEMSLE 7929 5-1-98
2. Materials and methods
2.1. Origin of biomass and cultivation
An Anammox culture, grown in a £uidized bed
reactor on synthetic medium containing 45 mM am-
monium and nitrite, was used for batch culture and
continuous culture experiments. The £uidized bed
reactor was fed with an anaerobic autotrophic syn-
thetic medium [10]. Before use, the biomass was ho-
mogenized and washed with 20 mM anaerobic
KHCO3 bu¡er (pH 7.5).
Alcaligenes faecalis TUD (LMD 89.147), an or-
ganism capable of denitri¢cation and heterotrophic
nitri¢cation, was cultivated in an acetate-limited
chemostat at 30³C and pH 8.0 with a dilution rate
of 0.05 h31 [13]. Saccharomyces cerevisiae T23D was
grown aerobically in a glucose-limited chemostat
with a dilution rate of 0.1 h31 at 30³C and pH 5.0
[14]. SHARON sludge was obtained from a labora-
tory-scale-nitrifying reactor fed with synthetic me-
dium [15].
2.2. Anaerobic batch culture experiments
Anaerobic batch cultures were grown in 50 ml
serum bottles at 30³C. Each bottle contained 10 ml
of 20 mM KHCO3 bu¡er (pH 7.5) and approxi-
mately 20 mg volatile suspended solids (VSS) of
Anammox biomass, except for the incubation with
hydrazine alone, when approximately 75 mg VSS
of Anammox was used. The bottles sealed with
butyl-rubber stoppers were made anaerobic with
an Argon/CO2 (95/5%) mixture. Hydrazine
(H4 N2 WH2 SO4 ), ammonium ((NH4 )2 SO4 ), nitrite
(NaNO2 ), and nitrate (NaNO3 ) solutions were sub-
sequently added with a syringe. N2 O was added in
the headspace. The initial NH‡ 3
4 , N2 H4 , NO2 , NO3
3
and N2 O concentrations were approximately 3 mM.
The pH was adjusted to 7.5 with 1 M Na2 CO3 . Con-
trol experiments without biomass were performed
simultaneously.
For the experiments with gamma-irradiated cells,
50 ml serum bottles with Anammox sludge, nitrify-
ing sludge, A. faecalis and S. cerevisiae cells were
exposed to 25 kGy of radiation for 7 h using a
57
Co source (Gammaster, Ede, The Netherlands),
before use [11].
 J. Schalk et al. / FEMS Microbiology Letters 158 (1998) 61^67
2.3. Bio¢lm reactors with hydrazine
2.3.1. Operation of the bio¢lm reactors
Two small bio¢lm reactors (height 25 cm, volume
330 ml) were operated at 30³C for 10 weeks using an
anaerobic autotrophic mineral medium [10]. The pH
of the medium was kept constant at pH 7.5 by £ush-
ing continuously with an Argon/CO2 (95/5%) mix-
ture during the experiment [16]. Both reactors were
inoculated with approximately 0.25 g VSS of Anam-
mox biomass. To establish a bio¢lm the feed of am-
monium and nitrite was increased in a stepwise man-
ner from 2 to 10 mM at a £ow rate of 0.16 ml min31 .
After 3 weeks, 1 mM of ammonium was replaced by
0.1 mM hydrazine in one of the reactors. The hydra-
zine was supplied from a separate bottle to avoid
chemical decomposition. When the hydrazine con-
centration fell below the detection limit (1 WM), the
in£owing concentration was increased stepwise,
reaching 0.9 mM at day 70. The second bio¢lm re-
actor was used as a control, and was fed with 10 mM
of ammonium and nitrite.
2.3.2. Activity measurements in the bio¢lm reactors
To determine the activity of the Anammox bio-
mass in both reactors, the reactors were run in batch
mode for 6 h with ammonium and nitrite. The initial
concentration of both ammonium and nitrite was
6 mM. Samples were collected at appropriate time
intervals and analyzed for the di¡erent nitrogen
compounds.
2.4. Analytical procedures
Nitrate, nitrite, hydroxylamine and ammonium
were determined colorimetrically as previously de-
scribed [11]. Nitrous oxide and dinitrogen gas forma-
tion were quanti¢ed using a GC 8340 model (Fisons
Table 1
Conversion rates (nmol min31 (mg VSS)31 ) of hydrazine and nitrite by active and Q-radiated inactive cells from di¡erent sources in batch
cultures
Inactive cells after Q-radiation
N2 H4
A. faecalis 6 0.01
S. cerevisiae 6 0.01
Nitrifying sludge 6 0.10
Anammox sludge 0.39
FEMSLE 7929 5-1-98
63
Instruments, Interscience, Breda, The Netherlands)
equipped with a thermal conductivity detector and
an electron capture detector [13]. Hydrazine was de-
termined colorimetrically by means of the method
described by Watt and Crisp [17]. Dry weight was
determined as previously described [16].
3. Results
3.1. Determination of the biological nature of
hydrazine conversion in Anammox
Batch experiments with active cells and cells inac-
tivated by Q-radiation from four di¡erent microor-
ganisms with hydrazine and nitrite as substrates
were performed to determine if hydrazine conversion
was of a biological nature. No signi¢cant conversion
of hydrazine was observed in three batch experi-
ments with cells inactivated by Q-radiation. The in-
activated Anammox sludge showed low activity com-
pared to active Anammox cells. Active A. faecalis
and S. cerevisiae were also not able to convert hy-
drazine and nitrite (Table 1). On the other hand,
Anammox showed high conversion of hydrazine
and nitrite, and the nitrifying sludge showed low
conversion of hydrazine and nitrite under anaerobic
conditions. The hydrazine conversion rate of the
Anammox culture was approximately 10-fold higher
than the nitrifying sludge.
3.2. Anaerobic batch culture experiments with
hydrazine
Since Anammox showed a high hydrazine conver-
sion, the oxidation of hydrazine with and without
nitrite, nitrate and nitrous oxide as electron accept-
ors was investigated in anaerobic batch experiments.
Active cells
NO3
2 N2 H4 NO3
2
6 0.01 6 0.01 6 0.01
6 0.01 6 0.01 6 0.01
6 0.01 0.47 1.80
0.19 4.23 5.46
64 J. Schalk et al. / FEMS Microbiology Letters 158 (1998) 61^67

Fig. 2. The conversion of hydrazine in a batch culture with an Fig. 3. The conversion of hydrazine and nitrite in a batch culture
Anammox culture (closed symbols) and without Anammox (open by the Anammox culture. Symbols : (b) N2 H4 ; (R) NO3 2 ; (F)
symbols). Symbols : (b and a) N2 H4 ; (F and E) NH‡
4. NH‡4.

Each experiment was repeated at least twice. The
standard deviation was not more than 25%. When
the culture was provided with hydrazine alone, am-
monium was formed (Fig. 2). The conversion rate
for hydrazine was 1.6 nmol min31 (mg VSS)31 and
for the ammonium formation a rate of 1.8 nmol
min31 (mg VSS)31 was calculated. Incubations with
hydrazine and nitrite as the electron acceptor showed
a rapid conversion of both nitrite and hydrazine
(Fig. 3). When all the nitrite was consumed, hydra-
zine was disproportionated into ammonium and di-
nitrogen gas. No hydroxylamine, nitrous oxide or
nitrate was detected. The rates of nitrite and hydra-
zine conversion were 4.2 and 5.5 nmol min31 (mg
VSS)31 , respectively. Nitrate and nitrous oxide could
also serve as electron acceptors (not shown). Since it
was shown that hydrazine reacts easily with metal
ions [2], control experiments without sludge were
performed. No spontaneous chemical conversion of
any compound was observed during such experi-
ments.
3.3. Bio¢lm reactors with hydrazine
After the short-term batch experiments, the long-
term e¡ect of hydrazine on the Anammox culture
was studied. For this purpose anaerobic bio¢lm re-
actors were operated for 80 days with the Anammox
culture. The reactors were ¢rst fed with ammonium
and nitrite to produce su¤cient biomass as bio¢lm.
In£owing NH‡ 3
4 and NO2 concentrations were in-
creased from 2 to 10 mM in 20 days. At that time
the NH‡ 3
4 and NO2 removal rates of the Anammox
bio¢lm was 9.0 and 9.5 WM min31 at day 20, respec-
tively (Table 2), leaving some residual ammonium,
but no nitrite (Fig. 4). At day 23, 1 mM NH‡ 4 was
replaced by 0.1 mM N2 H4 and increased in a step-
wise manner up to 0.9 mM in 10 days. In the ¢rst 20

Table 2
Anammox activity (WM min31 ) in a bio¢lm reactor, fed with 10 mM NH‡ 3
4 and 10 mM NO2 before and after the addition of N2 H4

Day N2 H4 in (mM) N2 H4 out (mM) Activity (WM min31 )

NH‡
4 conversion NO3
2 conversion NO3
3 formation

20 0 0 9.0 9.5 1.5

43 0.8 0 N.D. N.D N.D
70 0.9 0.1 6.0 11.0 0.7
80 0.9 0.2 1.2 1.7 ^*
The Anammox activity in the bio¢lm reactors was measured in batch mode for 6 h with ammonium and nitrite. The initial concentration of
both ammonium and nitrite was 6 mM. Hydrazine was added at day 23.
N.D., not determined.
*Below detection limit.

FEMSLE 7929 5-1-98

 J. Schalk et al. / FEMS Microbiology Letters 158 (1998) 61^67 65

Fig. 4. The e¡ect of hydrazine on the anaerobic ammonium oxidation in a bio¢lm reactor. The bio¢lm reactor was fed with 10 mM NH‡
4
‡
and NO3 3
2 . N2 H4 was added at day 23. Symbols : (b) N2 H4 in; (a) N2 H4 out; (O) NO2 out; (E) NH4 out.

days after the hydrazine addition, the bio¢lm was
able to convert and tolerate increasing amounts of
hydrazine (100% conversion, Fig. 4). Conversions of
ammonium and nitrite did not change signi¢cantly.
From day 45 the hydrazine and ammonium conver-
sion rate decreased. The ammonium removal, which
is correlated to the Anammox activity, dropped to
6.0 WM min31 at day 70, while the nitrite conversion
increased up to 11.0 WM min31 (Table 2). A very
small amount of nitrous oxide was detected. Activity
measurements at day 80 showed a greatly reduced
capacity for both ammonium and nitrite removal.
Apparently, in spite of its capability to convert hy-
drazine, the (mixed) Anammox culture cannot be
grown on hydrazine alone. A second attempt to
grow Anammox in a bio¢lm reactor on hydrazine
also failed. The control reactor, which was fed with
10 mM NH‡ 3
4 and NO2 performed constantly during
the whole experiment.
4. Discussion
In microbial nitrogen conversions hydrazine is
rarely observed as an intermediate. Previously, batch
experiments with Anammox sludge showed that hy-
drazine accumulated when the culture was fed with
hydroxylamine [12]. However, it was not known if
this phenomenon had a biological nature. The ex-
periments described in this article with active cells
and cells inactivated with Q-radiation from four dif-
ferent microorganisms with hydrazine and nitrite
showed that Anammox was able to convert hydra-
zine biologically (Table 1). A. faecalis and S. cerevi-
siae did not show any conversion of hydrazine and
nitrite. The nitrifying sludge, on the other hand,
showed some hydrazine conversion. This could be
explained by the presence of about 70% of Nitroso-
monas bacteria in the nitrifying sludge (Jetten, un-
published results). Nitrosomonas contains a large
amount of hydroxylamine oxidoreductase (HAO),
an enzyme catalyzing the oxidation of hydroxyl-
amine to nitrite. This enzyme can also convert hy-
drazine to dinitrogen gas [7,8]. However, the conver-
sion rate of hydrazine by the nitrifying sludge was
about ten times lower than the rate observed in the
Anammox culture (Table 1). A. faecalis, being a het-
erotrophic nitri¢er with HAO activity, seems not
capable of oxidizing hydrazine, in contrast to the
autotrophic nitri¢er N. europaea. The low conversion
of hydrazine observed in the batch experiment with
Q-radiated inactive Anammox culture may be caused
by incomplete inactivation of the hydrazine-convert-
ing enzyme, since a small amount of ammonium was
produced. Chemical decomposition of hydrazine was
not observed in the batch experiments [1], because
the hydrazine concentration in the batch experiments
with A. faecalis and S. cerevisiae and in the batch
experiments without cells did not decrease (Table 1).
The Anammox culture was also capable to convert
hydrazine in the absence of other electron acceptors
(Fig. 2). These results con¢rmed observations made

FEMSLE 7929 5-1-98

66 J. Schalk et al. / FEMS Microbiology Letters 158 (1998) 61^67
earlier in our laboratory [12]. According to the con-
version rate of hydrazine and formation rate of am-
monium the following equation could be derived:
3N2 H4 +4H‡ C4NH4 ‡ +N2 . The disproportionation
of hydrazine to ammonium and dinitrogen was
also found for the R2 subunit of the enzyme ribonu-
cleotide reductase from E. coli [9]. In batch experi-
ments with hydrazine and nitrite a diauxic substrate
conversion was observed (Fig. 3). Nitrite was ¢rst
reduced with hydrazine as the electron donor. After
all the nitrite was converted, disproportionation of
hydrazine to ammonium and dinitrogen was ob-
served, as shown in Fig. 2. According to the conver-
sion rates of hydrazine and nitrite during the ¢rst
5 h the following equation could be derived:
4NO3 ‡
2 ‡ 4H ‡ 3N2 H4 !5N2 ‡ 8H2 O. To resolve
the possible mechanism by which hydrazine is con-
verted with and without additional electron accept-
ors, 15 N-labelling studies should provide more infor-
mation.
The disproportionation of hydrazine to ammo-
nium and dinitrogen, as well as the conversion of
hydrazine and nitrite are thermodynamically favour-
able reactions (vGo P = 3181 kJ and 3311 kJ per mol
N2 H4 , respectively). Therefore the capability of the
Anammox culture to grow on hydrazine was tested.
This long-term e¡ect was studied with bio¢lm reac-
tors. During the ¢rst 20 days after the addition of
hydrazine it was observed that the Anammox culture
was able to convert and tolerate 1 mM hydrazine
(Fig. 4). The results from the activity measurements
(Table 2) showed that the Anammox activity de-
creased according to the rate of ammonium removal,
while the rate of nitrite conversion increased. The
increase in nitrite conversion is probably due to the
increase of denitri¢cation at the expense of decaying
biomass. Since denitri¢cation is a faster process than
the Anammox process [18], this will result in a lower
Anammox activity, as was observed after 80 days
(Table 2 and Fig. 4). Thus it is concluded that in
the long run hydrazine was toxic for the Anammox
culture, as was observed for most forms of life [1].
Although Anammox could not grow on hydrazine,
the enzyme responsible for the conversion of hydra-
zine is still very interesting. Whether one enzyme is
responsible for the disproportionation of hydrazine
into dinitrogen gas and ammonium or two di¡erent
enzymes are involved needs to be investigated. It is
FEMSLE 7929 5-1-98
possible that an enzyme similar to that observed in
the R2 subunit of ribonucleotide reductase from E.
coli with a dinuclear iron centre is present in Anam-
mox, or an enzyme similar to that of HAO from N.
europaea. More studies need to be performed to
solve the mechanism of hydrazine conversion in
Anammox.
References
[1] Moliner, A.M. and Street, J.J. (1989) Decomposition of hy-
drazine in aqueous solutions. J. Environ. Qual. 18, 483^
487.
[2] Demadis, K.D., Malinak, S.M. and Coucouvanis, D. (1996)
Catalytic reduction of hydrazine to ammonia with MoFe3 S4 -
polycarboxylate clusters. Possible relevance regarding the
function of the molybdenum-coordinated homocitrate in ni-
trogenase. Inorg. Chem. 35, 4038^4046.
[3] Sykes, A.G. (1970) Some reactions of ions and molecules of
non-metallic elements in aqueous solutions. In: Kinetics of
Inorganic Reactions, 2nd Edn., pp. 195^197. Pergamon, Ox-
ford, UK.
[4] Dilworth, M.J. and Eady, R.R. (1991) Hydrazine is a product
of dinitrogen reduction by the vanadium-nitrogenase from
Azotobacter chroococcum. Biochem. J. 277, 465^468.
[5] Hyman, M.R. and Arp, D.J. (1995) E¡ects of ammonia on the
de novo synthesis of polypeptides in cells of Nitrosomonas
europaea denied ammonia as an energy source. J. Bacteriol.
177, 4974^4979.
[6] Remde, A. and Conrad, R. (1990) Production of nitric oxide
in Nitrosomonas europaea by reduction of nitrite. Arch. Mi-
crobiol. 154, 187^191.
[7] Nicholas, D.J.D. and Jones, O.T.G. (1960) Oxidation of hy-
droxylamine in cell-free extracts of Nitrosomonas europaea.
Nature 185, 512^514.
[8] Hooper, A.B. and Terry, K.R. (1977) Hydroxylamine oxido-
reductase from Nitrosomonas: Inactivation by hydrogen per-
oxide. Biochemistry 16, 455^459.
[9] Han, J.Y., Swarts, J.C. and Sykes, A.G. (1996) Kinetic studies
on the hydrazine and phenylhydrazine reductions of the Es-
cherichia coli R2 subunit of ribonucleotide reductase. Inorg.
Chem. 35, 4629^4634.
[10] van de Graaf, A.A., de Bruijn, P., Robertson, L.A., Jetten,
M.S.M. and Kuenen, J.G. (1996) Autotrophic growth of anae-
robic, ammonium-oxidizing micro-organisms in a £uidized
bed reactor. Microbiology 142, 2187^2196.
[11] van de Graaf, A.A., Mulder, A., de Bruijn, P., Jetten, M.S.M.,
Robertson, L.A. and Kuenen, J.G. (1995) Anaerobic oxida-
tion of ammonium is a biologically mediated process. Appl.
Environ. Microbiol. 61, 1246^1251.
[12] van de Graaf, A.A., de Bruijn, P., Robertson, L.A., Jetten,
M.S.M. and Kuenen, J.G. (1997) Metabolic pathway of anae-
robic ammonium oxidation on basis of 15 N-studies in a £uid-
ized bed reactor. Microbiology 143, 2415^2421.
 J. Schalk et al. / FEMS Microbiology Letters 158 (1998) 61^67 67

[13] Otte, S., Grobben, N.G., Robertson, L.A., Jetten, M.S.M.
and Kuenen, J.G. (1996) Nitrous oxide production by Alcali-
genes faecalis under transient and dynamic aerobic and anae-
robic conditions. Appl. Environ. Microbiol. 62, 2421^2426.
[14] de Jong-Gubbels, P., Vanrolleghem, P., Heijnen, S., van Dij-
ken, J.P. and Pronk, J.T. (1995) Regulation of carbon metab-
olism in chemostat cultures of Saccharomyces cerevisiae grown
on mixtures of glucose and ethanol. Yeast 11, 407^418.
[15] Jetten, M.S.M., Horn, S.J. and van Loosdrecht, M.C.M.
(1997) Towards a more sustainable municipal wastewater
treatment system. Water Sci. Tech. 35, 171^180.
[16] Strous, M., van Gerven, E., Ping, Z., Kuenen, J.G. and Jetten,
M.S.M. (1997) Ammonium removal from concentrated waste
streams with the Anaerobic Ammonium Oxidation (Anam-
mox) process in di¡erent reactor con¢gurations. Water Res.
31, 1955^1962.
[17] Watt, G.W. and Chrisp, J.D. (1952) A spectrophotometric
method for the determination of hydrazine. Anal. Chem. 24,
2006^2008.
[18] Jetten, M.S.M., Logemann, S., Muyzer, G., Robertson, L.A.,
de Vries, S., van Loosdrecht, M.C.M. and Kuenen, J.G.
(1996) Novel principles in the microbial conversion of nitro-
gen compounds. Antonie van Leeuwenhoek Int. J. Gen. Mol.
Microbiol. 71, 75^93.

FEMSLE 7929 5-1-98