Journal of Bioscience and Bioengineering

VOL. 118 No. 4, 426e433, 2014

www.elsevier.com/locate/jbiosc

Comparison of sulfate-reducing and conventional Anammox upow anaerobic

sludge blanket reactors
Ergo Rikmann,1, * Ivar Zekker,1 Martin Tomingas,1 Priit Vabame,1 Kristel Kroon,1 Alar Saluste,1
Taavo Tenno,1 Anne Menert,1 Liis Loorits,2 Sergio S.C. dC Rubin,3 and Toomas Tenno1
Institute of Chemistry, University of Tartu, 14a Ravila St., 50411 Tartu, Estonia,1 Tallinn University of Technology, 5 Ehitajate St., 19086 Tallinn, Estonia,2 and Centro Nacional de
Investigaciones Biotecnologicas, CNIB, Calle Alcides Arguedas 429, Cala Cala, Cochabamba, Bolivia3
Received 15 November 2011; accepted 21 March 2014
Available online 23 May 2014

Autotrophic NHD
4 removal has been extensively researched, but few studies have investigated alternative electron
D
acceptors (for example, SO2L
4 ) in NH4 oxidation. In this study, sulfate-reducing anaerobic ammonium oxidation (SRAO)
and conventional Anammox were started up in upow anaerobic sludge blanket reactors (UASBRs) at 36 (0.5)
C and 20
2L
(0.5)
C respectively, using reject water as a source of NHD
or NOL
2 , respectively, were applied as electron ac4 . SO4
ceptors. It was assumed that higher temperature could promote the SRAO, partly compensating its thermodynamic
disadvantage comparing with the conventional Anammox to achieve comparable total nitrogen (TN) removal rate.
3
Average volumetric NHD
4 LN removal rate in the sulfate-reducing UASBR1 was however 5e6 times less (0.03 kg-N/(m
day)) than in the UASBR2 performing conventional nitrite-dependent autotrophic nitrogen removal (0.17 kg-N/(m3 day)).
However, the stoichiometric ratio of NHD
4 removal in UASBR1 was signicantly higher than could be expected from the
extent of SO2L
4 reduction, possibly due to interactions between the N- and S-compounds and organic matter of the reject
water. Injections of N2H4 and NH2OH accelerated the SRAO. Similar effect was observed in batch tests with anthraquinone-2,6-disulfonate (AQDS). For detection of key microorganisms PCR-DGGE was used. From both UASBRs, uncultured
bacterium clone ATB-KS-1929 belonging to the order Verrucomicrobiales, Anammox bacteria (uncultured Planctomycete
clone Pla_PO55-9) and aerobic ammonium-oxidizing bacteria (uncultured sludge bacterium clone ASB08 Nitrosomonas) were detected. Nevertheless the SRAO process was shown to be less effective for the treatment of reject water,
compared to the conventional Anammox.
2014, The Society for Biotechnology, Japan. All rights reserved.
[Key words: Sulfate-reducing ammonium oxidation; Upow anaerobic sludge blanket reactor; Humic matter; Anammox intermediates; Autotrophic
NH
4 removal]

Anammox-based technology for biological nitrogen removal has

several advantages over conventional nitrication-denitrication
since it needs less energy for aeration and no additional organic
carbon is required due to autotrophic processes involved (1,2).
Currently, the Anammox-related technology has the potential to
achieve a neutral or even positive energy balance in complete
wastewater treatment cycle (3e5) if, following to anaerobic treatment (biogas production), UASBRs are applied in the nitrogen
removal stage (Driessen et al., 14th European Biosolids and Organic
Resources Conference, 2010). Hence, this reactor type was selected
in the study. The metabolic versatility of Anammox organisms,
involving use of various substrates and electron acceptors has been
shown (6). Although there is no genome information available on
the ability of Anammox bacteria to consume SO2
as electron
4
acceptor, SO2
reduction
by
them
has
been
experimentally
4
observed. A Planctomycetes bacterium named Anammoxoglobus

2
sulfate, capable to oxidize NH
4 into NO2 using SO4 as an electron
acceptor, was isolated in 2008 from an enrichment culture (7).
Sulfate-reducing ammonium oxidation (SRAO) process was rst

* Corresponding author. Tel.: 372 5691 2374; fax: 372 737 5264.
E-mail addresses: ergo.rikmann@ut.ee, rikmannster@gmail.com (E. Rikmann).

assumed by Fdz-Polanco et al. (8), who proposed a summary

equation describing the two-staged process (Eq. 1), which has later
been complemented with a possibility for sulde formation noted
previous literature (6,9,10) (Eq. 2):
2
2NH
4 SO4 /S0 N2 4H2 O

DG0 46 kJ=mol

2


8NH
4 3SO4 /4N2 3HS 12H2 O 5H

22kJ=mol

(1)

D G0
(2)

In anaerobic, sulde-saturated sediments of mesophilic springs

carbohydrate fermentation and sulfur reduction are possible
mechanisms employed by heterotrophic Planctomycetes for growth
and survival (11). The alternative electron acceptors such as SO2
4
may provide opportunities to reduce the need for aeration in the
nitritation step preceding the Anammox process. This is especially
important in the case of wastewaters with high content of both Nand S-compounds, such as the ones generated in oil reneries, sh
canning, production of fertilizers (12), yeast factories etc. For
wastewaters with high content of total nitrogen (TN), sulfate and
organics, simultaneous removal of COD and nitrogen can thus be
achieved in the anaerobic phase of treatment. Simultaneously the
accumulation of toxic H2S is avoided. If the SRAO process gives

1389-1723/$ e see front matter 2014, The Society for Biotechnology, Japan. All rights reserved.
http://dx.doi.org/10.1016/j.jbiosc.2014.03.012

VOL. 118, 2014

COMPARISON OF SULFATE-REDUCING AND CONVENTIONAL ANAMMOX

satisfactory results, dosage of sulfate (as Na2SO4, for example) or, in

some cases even mixing the nitrogen-rich wastewater with sulfaterich wastewater from food or fermentation industry instead of
applying a pre-nitritation step would make the Anammox process
even more energy-efcient.
The aims of this study were to achieve quick start-up of the
Anammox UASBRs and to compare the SRAO process with the
conventional Anammox process (inoculated with the same seed)
using a real wastewater. Since addition of low-molecular quinoid
analogs of humic matter has been reported as an option to increase
TN removal efciency of denitriers (13), the effects of AQDS were
also researched.
MATERIALS AND METHODS
Reactor congurations, seeding procedure
In this research 0.75 L thermostated UASBR1 (for SRAO) at 36 (0.5)
C and 1.5 L volume UASBR2 (for con

ventional Anammox) at 20 (0.5) C were operated in parallel (Fig. 1). The inuent
was fed by periodically switched-on peristaltic pumps (SEKO, Italy). The hydraulic
retention time (HRT) kept was one to two days.
2
Inuent
The effect of NO
as Anammox electron acceptors was
2 vs. SO4
studied by feeding the reactors with reject water from anaerobic digestion of
municipal wastewater sludge obtained from Tallinn municipal wastewater treatment plant. The latter was diluted with tap water. For the UASBR1, SO2
4 was added to
2
the inuent as the solution of K2SO4 keeping an approximate molar NH
4 =SO4 ratio
of 2. For feeding of the UASBR2 and batch tests, nitrite for the Anammox reaction was

provided by adding NaNO2 to ensure the inuent molar NO

2 =NH4 ratio close to 1.32
as the optimum for Anammox reaction. Reject water contained Anammox microorganisms e Planctomycetales bacterium clone P4 (GenBank ID: DQ304521) and
sufcient amounts of micro- and macronutrients for Anammox propagation (14).
The inuent of reactors had the following average ratio of carbon and nitrogen
compounds: COD/TN 0.78:1 (range 0.39:1e1.10:1). The biodegradability of
inuent expressed as COD/BOD7 was 1.95:1 (range 1.82:1e2.03:1).

Seeding of UASBRs
Both UASBRs were seeded with anaerobic sludge containing Anammox bacteria, obtained from the facility treating wastewater of the

Salutaguse Yeast Factory (Salutaguse, Estonia) rich in both SO2

4 and NH4 . After the
inoculation, the volatile suspended solids (VSS) of UASBR1 and UASBR2 were 1.87 g/
L and 1.10 g/L, respectively. The average TN removal rate in the UASB reactor of the
Salutaguse yeast factory wastewater treatment facility was around 4.80 kg-N/(m3
day).
Batch assays
To study the effect of a model compound of a quinone
analog for humic matter (HM) e anthraquinone-2,6-disulfonate disodium salt
(AQDS) e on TN and SO2
4 removal rate by Anammox bacteria, batch assays were
conducted at 20 (0.5
C). The same temperature was selected for batch tests with
2
NO
2 and with SO4 as electron acceptors in order of better comparability of batch
tests at least with UASBR2. Stable temperature was maintained using a water bath
thermostat (Assistent 3180, Glaswarenfabrik Karl Hecht GmbH, Germany). Batch
tests were performed using sludges from UASBR1 and UASBR2, maintaining a VSS
concentration of 1.8e2.0 g/L. (NH4)2SO4-water solution was used as a synthetic
Anammox medium in case of UASBR1 and NH4CleNaNO2-water solution in case of

427

UASBR2. Acidic solution (3 mL) and 3 mL of alkaline solution of micronutrients were

added into the substrate of batch tests along with the 40 mL of macronutrients
solution as described in previous literature (12,15).
The procedure of the batch tests and analytical methods has been described in
detail in previous literature (14). Data and statistical analyses were performed by the
MS Excel 2010 Analysis ToolPak. Homogeneity of group variances and the difference
between group means were checked using the F-test and the two-way t-test,
respectively. The level of signicance was set at a < 0.05.
PCR-DGGE, sequencing and phylogenetic analysis
The PCR-DGGE was
performed as described in previous literature (12,15). PCR for sequencing was
performed with the BigDye Terminator v3.1 Cycle Sequencing Kit (Life
Technologies Corporation, USA). The sequences acquired were compared to the
available database sequences via a BLAST (Basic Local Alignment Search Tool)
search from the GenBank (http://www.ncbi.nlm.nih.gov/genbank/). The samples
from the treatment facility of Salutaguse Yeast factory were pyrosequenced at the
Integrated Systems Biology Centre of Tallinn University of Technology. Universal
8F and 357R primers were used for the PCR amplication of the V2eV3 hyper
variable regions of 16S rRNA genes. The 357R primer included additionally a
unique sequence tag to barcode each sample (Plaza et al., Proceedings of the IWAWEF Spec. Conf. Nutrient Recovery and Management, Miami, FL, 2011). Sequences
obtained from PCR-DGGE analysis were compared with 16S rDNA sequences of
related species. Phylogenetic tree showing these relationships was constructed
with MEGA software version 5.0.

RESULTS AND DISCUSSION

Description of operation of the reactors Operation of the
reactors was divided into periods based on the characteristics of
efuent quality, HRT, loading rates (and in the nal, IV period,
dosage of intermediates as discussed below). The main parameters
of operation of the UASBR1 (SRAO) and the UASBR2 (Anammox) are
given in Table 1. The UASBRs were started up with HRTs of one day.
Selection of the HRT was based on comparison with other studies
(9,16). Surprisingly, the seeding sludge showed a more rapid
adaptation in the UASBR2 than in the UASBR1, and in the latter
the TN removal rates were signicantly (p-value <0.05) lower
than in the UASBR2 (see Table 1, Figs. 2 and 3).
Considering
disproportionally
high
ratio
of
2
DNH
4  Nconsumed =DSO4  Sconsumed , comparing with the ratio
emanating from Eq. 1 or 2, the NH
4 content in the inuent for the
UASBR1 was doubled between days 41e100 while the SO2
4 content
was retained unchanged (Table 1, Fig. 2a). For the UASBR1, the
increased concentration of NH
4 in the inuent had no obvious effect
on promoting the TN removal, although higher substrate concentrations have been reported to facilitate the SRAO reaction in earlier
studies (9). The data presented in Table 1 show that at comparable
ammonium loading rates, nearly complete deammonication was
achieved in UASBR2 while in the UASBR1 it remained less than 1/3.

FIG. 1. Scheme of the UASBR1 (performing SRAO) and the UASBR2 (performing conventional Anammox process).

428

RIKMANN ET AL.

J. BIOSCI. BIOENG.,

TABLE 1. The basic parameters of the ow-through experiments.

Parameter

UASBR1 SRAO

UASBR2 Anammox

300

0,50

Sulfate-S, influent UASBR1

Sulfate-S, effluent, UASBR1

Average

0.43
0.5
0.23

1.10
20
7.5

250

Stdev
0.3
1
0.5

1
17
0.09
0.02
85
70
71
64
19.1/1

0.2
9
0.03
0.01
28
23
6
9

1
75.4
0.22
0.21
80.5
42

0.2
21
0.04
0.06
14.23
39.5

1
18
0.15
0.02
134
119
82
78
26.8/1

0.2
15
0.01
0.02
24
33
7
5

1
70
0.37
0.25
80.42
1.15

1
25
0.17
0.04
158
125
123
110
8.2/1

0.3
11
0.04
0.02
46
42
32
31

1
70
0.43
0.25
123.5
9.8

2
30
0.11
0.04
221
155
193
154
6.7/1

0.4
8
0.01
0.01
22
18
16
26

2
69
0.40
0.28
178.23
4.6

0.4
2.9
0.11
0.077
47.00
3.4

24
0.13
0.03
155
119
123
106
13.3/1

10
0.03
0.01
32
30
18
20

70
0.37
0.25
120
11

14.8
0.1
0.08
50
nd

Ammonium-N, influent UASBR1

0,40

Ammonium-N, effluent, UASBR1

200

Sulfide-S, influent UASBR1

0,30

Sulfide-S, effluent, UASBR1

150
0,20
100
0,10

50

0,00

0
0
0.2
18.3
0.04
0.04
13.22
0.41

0.3
20
0.09
0.07
38.63
5.4

250

50

100

150
200
Days of operation

250

300

350

250

300

350

Ammonium-N, influent UASBR2

Ammonium-N, effluent UASBR2
Nitrite-N, influent UASBR2

200

Nitrite-N, effluent UASBR2

Nitrate-N, effluent UASBR2

150

100

50

0
0

TN loading rate after day 86 for the UASBR1 was lower due to
diminished NH
4 values in reject water. Efuent TN values of both
reactors were unstable during this period (Figs. 2a and b), but the
TN removal rate remained low only in the UASBR1. After day 239
(period IV), stable and efcient TN removal was achieved in the
UASBR2; also the performance of the UASBR1 was more stable than
in earlier periods.
After day 239 in the UASBR1 and day 259 in the UASBR2, the
HRT was set to 2 days to facilitate a better process performance. For
the same purpose, Anammox intermediates were injected into the
reactors. Finally, a more stable performance of the UASBR1 with
fewer uctuations in the TN removal efciencies and rates than in
the UASBR2 was achieved (Table 1, Figs. 2 and 3).
Nevertheless, the process efciency in the UASBR1 remained
somewhat lower compared to other researches performed with
synthetic wastewaters. NH
4 removal efciencies of 40e45% have
been achieved (9,17e19), while Fdz-Polanco et al. (8) has reported a

Sulfide conc, mg S/L

Stdev

Sulfate/Ammonium conc,
mg N/L, mg S/L

1.87
36
8.11

Ammonium, Nitrite, Nitrate conc, mg N/L

Average
VSS of seeding sludge, g/L
Temperature,
C
pH
Period I, start-up (days 0e40)
HRT, days
TN removal efciency, %
TN loading rate, kg-N/(m3 day)
TN removal rate, kg-N/(m3 day)
NH
4  N (inuent), mg N/L
NH
4  N (efuent), mg N/L
SO2
4  S (inuent), mg/L
SO2
4  S (efuent), mg/L
2
NH
4  Nconsumed =SO  Sconsumed
Period II, days 41e100
HRT, days
TN removal efciency, %
TN loading rate, kg-N/(m3 day)
TN removal rate, kg-N/(m3 day)
NH
4  N (inuent), mg N/L
NH
4  N (efuent), mg N/L
2
SO4  S (inuent), mg/L
2
SO4  S (efuent), mg/L
2
NH
4  Nconsumed =SO4  Sconsumed
Period III, days 101e239
HRT, days
TN removal efciency, %
TN loading rate, kg-N/(m3 day)
TN removal rate, kg-N/(m3 day)
NH
4  N (inuent), mg N/L
NH
4  N (efuent), mg N/L
SO2
4  S (inuent), mg/L
SO2
4  S (efuent), mg/L
2
NH
4  Nconsumed =SO4  Sconsumed
Period IV, from day 239 onwards
HRT
TN removal efciency, %
TN loading rate, kg-N/(m3 day)
TN removal rate, kg-N/(m3 day)
NH
4  N (inuent), mg N/L
NH
4  N (efuent), mg N/L
SO2
4  S (inuent), mg/L
2
SO4  S (efuent), mg/L
2
NH
4  Nconsumed =SO4  Sconsumed
Entire experimental period
TN removal efciency, %
TN loading rate, kg-N/(m3 day)
TN removal rate, kg-N/(m3 day)
NH
4  N (inuent), mg N/L
NH
4  N (efuent), mg N/L
2
SO4  S (inuent), mg/L
2
SO4  S (efuent), mg/L
2
NH
4  Nconsumed =SO4  Sconsumed

50

100

150
200
Days of operation

FIG. 2. Process performance of the UASBR1 (SRAO) and the UASBR2 (conventional
Anammox) reactors. (a) Dynamics of ammonium-N and sulfate-S in the UASBR1.
Symbols: open triangles, inuent ammonium-N; closed triangles, efuent ammoniumN; open diamonds, inuent sulfate-S; closed diamonds, efuent sulfate-S; open circles,
inuent sulde-S; closed circles, efuent sulde-S. (b) Changes in the inuent and


efuent NH
4  N and NO2  N, NO3  N concentrations in the UASBR2. Symbols:
open triangles, inuent ammonium-N; closed triangles, efuent ammonium-N; open
circles, inuent nitrite-N; closed circles, efuent nitrite-N; closed diamonds, efuent
nitrate-N.

30e55% total Kjeldahl nitrogen (TKN) removal in treatment of

vinasse wastewater.
Bioprocesses in the UASBR1 (SRAO)
The SRAO is a 2-staged
process, with formation of nitrite in the rst stage as an intermediate according to Eq. 1, giving a molar ratio of
2
DNH
4  Nconsumed =DSO4  Sconsumed 2 (7e9,17e19). The SRAO
processes was also reported to occur without participation of
Anammox bacteria, as discovered by Jing et al. (18) isolating the
Bacillus benzoevorans strain ASR. Controversially, the work of
Sabumon (20) similarly to our UASBR1 has shown a much higher
ratio (Table 1). Examining the bioprocesses in the facility treating
wastewater of the Salutaguse yeast factory (Salutaguse, Estonia)


rich in both SO2

and NH
4
4 , partial nitrication of NH4 to NO2
using SO2
as
an
oxidant
was
assumed
(21).
Syntrophic
4
relationship between NH
sulfate-reducing and
4 -oxidizing,
Anammox bacteria could make thermodynamically not favorable

2
oxidation of NH
to
4 to NO2 coupled with reduction of SO4
elemental sulfur (Eq. 1) possible, facilitated by the presence of
easily degradable organics (11,22). In the present case the
occurrence of sulfate reduction with organics (Eq. 3) had a
limited extent as indicated by only a low reduction of COD
(average 8%) in UASBR1 due to mainly recalcitrant nature of
organics, unfavorable for most sulfate-reducing bacteria, and low
sulde concentration.


SO2
4 organics/HS HCO3 H2 O

(3)

COMPARISON OF SULFATE-REDUCING AND CONVENTIONAL ANAMMOX

a 100

80

300
70

250
200

60

150
100

50

429

NO2-N control
NO2-N with humic substance
NH4-N control
NH4-N with humic substance
NO3-N with humic substance
NO3-N control

50

40

Hydrazine conc, g/L

90

TN conc, mg N/L

500
TN control
450
TN with humic substance
Hydrazine with humic substance 400
Hydrazine control
350

N conc mg N/L

VOL. 118, 2014

30

20

10

50
40

0
2

4
Time (h)

4
Time (h)

d
Sulfate control

500

270

Ammonium-N control

265

Ammonium-N with humic

substance
Hydrazine with humic substance

400

Hydrazine control

300

450

800
Sulfate with humic
substance

NH4+ conc, mg N/L

780
SO42- conc, mg/L

260

760

350

250

255

740

200

250

720

245

700

240

150

Hydrazine conc, g N/L

100
50

10

15

20

25

0
0

10

15

20

25

Time (h)

Time (h)

FIG. 3. Time-dependent changes in concentrations of nitrogen species and SO2

4 and hydrazine production in batches with UASBR sludges in the presence and absence of humic
substance. (a) UASB1: closed squares, TN in control; open triangles, TN with humic substance; closed diamonds, N2H4 with humic substance; open circles, N2H4 in control. (b)




UASB1: crosses, NO
2 in control; open diamonds, NO2 with humic substance; pluses, NH4 in control; closed triangles, NH4 with humic substance; open squares, NO3 with humic
2

substance. (c) UASB1: closed squares, SO2

4 control; open triangles, SO4 with humic substance. (d) UASB2: closed squares, NH4 control; open triangles, NH4 with humic substance;
closed diamonds, N2H4 with humic substance; open circles, N2H4 in control.

mechanism of O2 generation in anoxic medium by oxygenic bacteria was described by Ettwig et al. (25), although the presence of
anoxic oxygenic bacteria in the engineered systems has not been
reported yet. As sulde concentrations in the efuent of UASBR1
were very low (not exceeding 0.1 mgS/L) it indicated to a rapid
sulde oxidation into S0 or SO2
4 .
Interactions between HM present in the reject water, nitrous
and sulfurous compounds can also contribute to the above2
mentioned effects on NH
4 oxidation. Lower SO4 reduction than
assumed could be due to partial re-oxidation of S0 or HS into SO2
4
taking place via sulfur-utilizing denitrication/denitritation. This
D 2
alters the nal balance DNH
4 consumed = SO4 consumed vratio,
increasing it. Presence of denitrifying sulfur-oxidizing Sulfurimonas
denitricans DSM 1251 in the seeding sludge as well as in the
reactor provides clear evidence in favor of this denitrication
mechanism. Sulfur-utilizing denitrication/denitritation reactions
include (26):
2

2
3
4
S0 6 5 NO
3 5 H2 O/SO4 5 N2 5 H
=

(4)

DG0 548 kJ=mol

2
S0 2NO
2 /SO4 N2

DG0 676 kJ=mol

2

7
4
4
HS 8 5 NO
3 5 H /SO4 5 N2 5 H2 O

(5)
(6)

DG0 739kJ=mol
2

7
4
4
HS 8 3 NO
2 3 H /SO4 3 N2 3 H2 O

DG0 981 kJ=mol

Pyrosequencing, on the contrary, indicated the presence of

several phyla of sulfur-cycle bacteria (see Table 2). The relative
abundance of genus Kosmotoga (phylum Thermotoga) in the sludge
of UASBR1 was in the range of 20%e75%. The presence of these
thermophilic, obligately anaerobic heterotrophs has recently been
reported also in mesophilic environments (23) as well as their
growth stimulation by reduction of elementary sulfur (24) produced in the SRAO reaction. As mentioned above, conventional
Anammox is the terminal stage of SRAO. DGGE-PCR indicated the
presence of Planctomycetaceae in the UASBR1. However, nitrite used
by conventional Anammox can be generated in other pathways
than the initial stage of SRAO including aerobic NH
4 oxidation or
NO
respiration
(if
the
latter
is
present).
3
Even if other possible electron acceptors for NH
4 oxidation
3
[NO
x , Fe , DO (at concentration 0e0.2 mg O2/L in the inuent)]
were taken into consideration, the ratio of oxidized NH
4  N still
remained higher than it can be concluded from consumption of
electron acceptors. Organic matter in wastewater may inuence the
anaerobic oxidation of NH
4 . Sabumon (20) proposed that heterotrophic facultative anaerobes when being previously affected by
the conditions of oxidative stress generate H2O2 under anaerobic
conditions. H2O2, while detoxied by catalases, can provide O2 for
NH
4 oxidizers. The assumption on formation of reactive oxygen
species (ROS) in bacterial respiration involving organic carbon and/
or any electron acceptor under anoxic conditions to be further
detoxied by catalases/superoxide dismutases is, however, objected in another work (25). Additional O2 if generated in the microenvironments in detoxication of ROS can be used by NH
4
oxidizing bacteria for producing NO
2 , which, in turn, can be
consumed by Anammox bacteria. Recently, a novel intra-aerobic

(7)

430

RIKMANN ET AL.

J. BIOSCI. BIOENG.,
TABLE 2. Bacterial strains identied in the seeding sludge and in the reactors.

Name of species
Seeding sludge
Uncultured bacterium clone Inoculum14
(Porphyromonadaceae), GenBank: HM008286
Unclassied Planctomycetaceae uncultured
bacterium JJB347, GenBank: GQ143799
Uncultured Verrucomicrobia bacterium
Sulfurimonas denitricans DSM 1251
Methylobacterium spp. (strains EI189, EI187,
AKB-2008-OT7, ED323, SEMIA 6407)
UASBR1 SRAO
ATB-KS-1929 (order Verrucomicrobiales),
GenBank: EF686989
Uncultured Verrucomicrobiales bacterium clone
De2102, GenBank: HQ183974
Uncultured Planctomycete clone Pla_PO55-9,
GenBank: GQ356109
Uncultured bacterium clone Dok23,
GenBank: FJ710742.1
Uncultured bacterium clone A1-F6_M13R Thiobacillus,
GenBank: GU083403.1
Uncultured sludge bacterium clone ASB08 Nitrosomonas,
GenBank: FJ947122.1
Uncultured bacterium clone 081203-OL-PVP22:1-9
Desulfobulbaceae, GenBank: FJ823212.1
UASBR2 Anammox
ATB-KS-1929 (order Verrucomicrobiales),
GenBank: EF686989
Uncultured Verrucomicrobiales bacterium clone
De2102 16S ribosomal RNA gene, partial sequence,
GenBank: HQ183974
Uncultured Planctomycete clone Pla_PO55-9 16S
ribosomal RNA gene, partial sequence,
GenBank: GQ356109
Uncultured Planctomycetales bacterium clone P4
16S ribosomal RNA gene, GenBank: DQ304521.2
Uncultured sludge bacterium clone ASB08
Nitrosomonas, GenBank: FJ947122.1
Uncultured bacterium clone KIST-JJY016 Nitrospira,
GenBank: EF594049.1
Uncultured bacterium clone B17_926R Thiobacillus,
GenBank: HQ141362.1

Primer

Determination method

Reference

GC-BacV3f/907r

PCR-DGGE

32

GC-BacV3f/907r

PCR-DGGE

32

Eub27f/Eub1492rPla46F-GC/Amx368R
8F/357R with 454 specic sequencing
primer parts
8F/357R with 454 specic sequencing
primer parts

PCR-DGGE
Pyrosequencing

33e35
454 platform manufacturer
(Roche) protocol
454 platform manufacturer
(Roche) protocol

GC-BacV3f/907r

PCR-DGGE

33e35

Pla46F-GC/Amx368R

PCR-DGGE

32

Eub27f/Eub1492r and Pla46f/Amx368r

PCR-DGGE

32

GC-BacV3f/907r

PCR-DGGE

32

8F/357R with 454 specic sequencing

primer parts
8F/357R with 454 specic sequencing
primer parts
8F/357R with 454 specic sequencing
primer parts

Pyrosequencing

454 platform manufacturer

(Roche) protocol
454 platform manufacturer
(Roche) protocol
454 platform manufacturer
(Roche) protocol

Pyrosequencing

Pyrosequencing
Pyrosequencing

GC-BacV3f/907r

PCR-DGGE

33e35

Pla46F-GC/Amx368R

PCR-DGGE

32

Eub27f/Eub1492r and Pla46f/Amx368r

PCR-DGGE

32

Eub27f/Eub1492r and Pla46f/Amx368r

PCR-DGGE

32

8F/357R with 454 specic sequencing

primer parts
8F/357R with 454 specic sequencing
primer parts
8F/357R with 454 specic sequencing
primer parts

Pyrosequencing

454 platform manufacturer

(Roche) protocol
454 platform manufacturer
(Roche) protocol
454 platform manufacturer
(Roche) protocol

Sulfur- and sulde-utilizing denitrication, consuming NO

3, a
by-product of Anammox bacteria (Eqs. 4 and 6), would be benecial
for TN removal, ensuring NO
3 reduction in the efuent. On the
other hand, sulfur-utilizing denitriers may compete with Anammox bacteria for available NO
2 (Eqs. 5 and 7). This might be one of
factors contributing to low efciency in the UASBR1.
Bioprocesses in the UASBR2 (conventional Anammox) The
extent of heterotrophic denitrication in the UASBR2 can be estimated from COD reduction, assuming that low DO concentration in
the inuent had a negligible effect on removal of recalcitrant organics. The measured COD values of both inuent and efuent were
corrected taking into consideration the calculated oxygen demand
needed for oxidation of NO
2 present in the wastewater. Nitrite
values were subtracted from COD values measured and the
resulting difference was used in mass balance calculations. In the
following estimation of the extent of denitrication we assume that


NO
2 and NO3 NOx  N were consumed in denitrication equimolarly. The values of corrected COD reductions in the UASBR2
were in the range of 61e80 mg O2/L (average 70 mg O2/L); the
corresponding NO
x  N reduction in heterotrophic denitrication
based on the aforementioned assumption was in the range of
18e23 mg N/L (average 20 mg N/L). Genera of denitriers present
in the UASBR2 included, inter alia, Paracoccus and Comamonadaceae. Based on TN removal efciencies we can conclude that heterotrophic denitrication constituted 6e12% (average 10%) of total
TN removal rate in the UASBR2.

Pyrosequencing
Pyrosequencing

The fraction of nitrogen removed by the Anammox process can

be calculated by subtraction of NO
x  N removed in heterotrophic
denitrication from total TN removal. The estimated TN removal
rate by the Anammox process was 88e94% (average 90 %) of total
TN removal rate.
In the UASBR2 technical setup water dripping (see Fig. 1) was

applied in which case part of feed NH
4 was oxidized into NO2

through water recirculation to save NO2 amounts to be added into
feeding tank. Water dripping also provided protection from NO
2

inhibition, enabling oxidation of excess NO
2 into less toxic NO3 .
The extent of the aerobic ammonium oxidation can be calculated by
subtraction of anaerobically oxidized NH
4  N from the difference
between inuent and efuent NH

N. Ammonium-oxidizing
4
bacteria (AOB), including genus Nitrosomonas, contributed to nearly

40 % of total NH
4 oxidation, while 60 % of NH4  N oxidation was
achieved in Anammox pathway. At the same time the concentra
tions of NH
4 and NO2 in the efuent were very low, obviously due
to the joint activity of the AOB and the nitrite oxidizing bacteria
(NOB).
NO
3 was formed in the UASBR2 either as a by-product of the
Anammox reaction or a product of oxidation of NO
2 performed by

NOBs. The calculated extent of direct NO
2 oxidation into NO3 can
be estimated from the difference between inuent NO

N
and
2

efuent NO
2  N by subtracting from it the amount of NO2 utilized
in the Anammox process and in the heterotrophic denitrication,
and adding the amount of NO
2 formed in the aerobic oxidation of

NH
4 performed by AOB-s. The calculated amount of NO3 formed

VOL. 118, 2014

COMPARISON OF SULFATE-REDUCING AND CONVENTIONAL ANAMMOX

can be derived by summarizing the calculated amount of NO

2

directly oxidized into NO
3 and the calculated amount of NO3
formed emanating from the stoichiometric equation of the Anammox reaction. Subtracting from this result the calculated amount of
NO
3 used in heterotrophic denitrication results in the calculated
amount of NO
3 formed in the bioprocesses in the UASBR2. This
calculated value for the amount of NO
3 formed is well comparable
with the value calculated as the actual difference between inuent
and efuent NO
3 (both in a range of 110e160 mg N/L).
In the UASBR2, the concentration of NO
3 in the efuent (Fig. 2b)
was more than twice higher than the Anammox stoichiometry
would predict. NOBs adaptation to low DO concentrations
(0e0.2 mg/L) limited the TN removal in UASBR2, while its effect on
UASBR1 was rather insignicant (small NO
3 concentrations in the
efuent). NOBs belonging to Nitrospira spp. may be responsible for
the disproportionally high NO
3 levels of efuent. On the other
hand, oxidation of excess NO
2 protected the biomass of UASBR2
from NO
2 inhibition.
The role of hydrazine and hydroxylamine
When intermediates N2H4 and NH2OH were not added into the reactors
N2H4 was still detected in the efuents of both reactors during the
period IV (days around 200, around 0.01 mg N/L and 0.13e0.21 mg
N/L in UASBR1 and UASBR2, respectively), indicating the Anammox
activity. Injections of hydrazine at low dosages (0.44 mg N/L) into
UASBR1 since day 268 showed a TN removal efciency increase
over 30%. Consumption of SO2
also increased and the ratio of
4
2
NH

N
=SO

S
consumed
consumed decreased (Table 1). Since day
4
4
301, N2H4 and NH2OH (5.50 mg N/L) of each was applied to both
reactors. The dosage selection was based on prior batch tests
(data not shown). By 24 h, the both substances were mostly
consumed in UASBR1 with residual NH2OH concentrations
ranging from 0.07e0.16 mg N/L and the residual N2H4
concentrations 0.36e3.15 mg N/L in the efuent. Residual NH2OH
and N2H4 concentrations in the efuents of UASBR2 were
0.09e0.17 mg N/L and 0.13e0.22 mg N/L respectively. The higher
rate of consumption of intermediates in UASBR2 indicated higher
activity of Anammox microorganisms in this system as compared
to UASBR1.
Interactions between organic matter, N- and Scompounds Organics in landll leachates and liquid fractions
from dewatering of anaerobic digestates are largely present in the
form of humic matter (HM; approximately 2/3 of organics expressed
as TOC in reject water used in our experimental work). Effect of HM
on nitrogen removal is still scantily researched. However, HM has
been shown to inuence both denitrication (12) and Anammox
(Williams, S. et al., Microbial ecology of Anammox organisms: do
organics drive Anammox? The 14th International Symposium
on Microbial Ecology, ISME14, 2012). If wastewater contains

431

S-compounds in addition to N-compounds and organics (including

HM), the network of interactions becomes even more complex.
HM, present in the wastewater either in an oxidized (functionally
active groups e quinones) or reduced form (hydroquinones),
behaves as a redox mediator both biologically and abiotically,
amplifying the effect of small amounts of O2 that can penetrate
into an anaerobic reactor medium. HM and its quinoid analogs can
serve as terminal electron acceptors during the anaerobic
oxidation of organic compounds and inorganic compounds such as
HS (oxidizing it into S0/polysuldes) (27), S2 O2
(oxidation
3
2
product SO2
(into Fe3) (28). On the contrary, reduced
4 ) or Fe
HM and hydroquinones can serve as electron donors for the

anaerobic reduction of NO
2 , NO3 , N2O (into N2) (12,29,30).
Batch assays AQDS increased the production of N2H4 and the
TN removal rate in SRAO comparing to control tests without AQDS.
Better TN removal rates were achieved with sludge from UASBR2
than compared to the UASBR1 (Fig. 3). TN removal rates for UASBR2

sludge using a molar ratio of NO

2  N=NH4  N 1:32 with AQDS
added (87 mg/L to receive the same TOC concentration as in reject
water) and without AQDS were 3.80 and 2.61 mg N/(g VSS h),
respectively. In comparison, for UASBR1 the TN removal rate with
AQDS was 0.36 mg N/(g VSS h) while being 0.30 mg N/(g VSS h)
without it. The SO2
removal rate (with UASBR1 sludge) was
4
different with and without AQDS added being 0.56 and 0.37 mg
SO2
4 /(g VSS h), respectively (Fig. 3c).
Our batch tests showed that AQDS may inuence the performance of both conventional (nitrite-utilizing) and sulfate-reducing
Anammox.
PCR-DGGE and pyrosequencing
The seeding sludge for both
reactors originally contained Anammox organisms (unclassied
Planctomycetaceae uncultured bacterium) in addition to unclassied Porphyromonadaceae and uncultured Verrucomicrobia bacterium (Table 2). Sludge samples from the treatment facility of
Salutaguse yeast factory were pyrosequenced, since this was the
origin of the sludge used. The presence of S. denitricans DSM
1251 was revealed. These bacteria are benecial for eliminating
nitrate (produced by Anammox bacteria), enhancing thus the
quality of the efuent water of the system. Methylobacterium spp.
(strains EI189, EI187, AKB-2008-OT7, ED323, SEMIA 6407) were
also found in the seeding sludge of the reactors. Methylobacterium
is a facultative methylotroph that grows on methylamine,
methanol, and C2, C3, and C4 compounds. Methylobacterium forms
a strong cohesive mat at carrier/water interfaces, which promotes
biolm formation (31). These bacteria might have facilitated also
the formation of strong aggregates in UASBR1.
From both UASBRs, uncultured bacterium clone ATB-KS-1929
belonging to the order Verrucomicrobiales, uncultured Planctomycete clone Pla_PO55-9 e an Anammox bacterium, and uncultured

FIG. 4. Phylogenetic neighbor-joining tree, reecting the relationships between identied sequences [uncultured Planctomycetaceae bacterium clone JJB347 (GenBank: GQ143799)
and uncultured bacterium clone ATB-KS-1929 (GenBank: EF686989)] and 16S rDNA sequences of other known bacteria belonging to the phyla Planctomycetes and Verrucomicrobia.

432

RIKMANN ET AL.

sludge bacterium clone ASB08 Nitrosomonas e an aerobic

ammonia oxidizing bacterium were detected by PCR-DGGE (Fig. 4).
Some bacteria involved in sulfur metabolism (uncultured bacterium clone A1-F6_M13R Thiobacillus and uncultured bacterium
clone 081203-OL-PVP22:1-9 Desulfobulbaceae) were present in
the UASBR1 (Table 2). In the UASBR2, Nitrosomonas strain GenBank:
FJ947122.1 was detected in addition to some Nitrospira strains.
Using the pyrosequencing technique additionally the following
OTUs (operational taxonomic units) with 98e100% condence and
relative abundance >0.6% were detected in UASBR1: Bacteria, Thiobacillus, Betaproteobacteria, Anaerolineaceae, Aminobacterium,
Xanthomonadales, Kosmotoga, Thermomonas, Flavobacteriaceae,
Gemmatimonas, Desulfobulbaceae and Sulfurimonas. In the UASBR2
the variety of OTUs was different: representatives of Arenimonas,
Xanthomonadales, Rhodocyclaceae, Thermomonas, Anaerolineaceae,
Paracoccus, Bacteroidetes, Rhodobacteraceae, Petrimonas, Diaphorobacter, Comamonadaceae, Thiobacillus were found.
In conclusion, Verrucomicrobiales and Planctomycetales persisted in both UASBR1 (SRAO) and UASBR2, with more species
present in UASBR2. Our results indicate that Verrucomicrobiales
may possibly be involved in some nitrogen-removing process.
However, species present in the inoculum were replaced by species possibly originating from reject water over time. The electron
acceptor had a major role on the composition of microbial consortia in reactors both fed with reject water: in the UASBR1, sulfur
cycle microorganisms were dominant, while the UASBR2 was
dominated by nitrogen-cycle microorganisms. Combining PCRDGGE and pyrosequencing techniques provides comprehensive
information about microbial ecosystems in different reactor
environments.
The results of this study indicate that nitrogen could be removed
from the real wastewater and synthetic wastewater using either

NO
2 (more efciently) or SO4 (less efciently) as an electron
acceptor in AnammoxeUASB reactors inoculated with the same
seeding material. We conclude that the SRAO process does not fully
justify itself for nitrogen removal from reject water, compared with
conventional Anammox. However, according to previous studies
(8,18,19), the SRAO process may give satisfactory results for
wastewater, containing simultaneously high concentrations of nitrogen, sulfate and easily biodegradable organics. Further studies
with different wastewaters are needed to assess the applicability of
the SRAO process.
From our results of operation of two UASBRs inoculated with
the same seeding material it appeared that the SRAO process was
rapidly started up for the treatment of supernatant from sludge
digestion, although it was much less efcient than the conventional Anammox process despite higher temperature (36 (0.5)
C)
applied. The SRAO process took place as one reaction of the
multiple complex interactions between N-compounds, S-compounds and organics (containing quinoid groups), resulting in a
signicantly higher removal ratio of NH
4 than SRAO stoichiometry
predicts. The presence of denitrifying sulfur-oxidizing microorganism S. denitricans DSM 1251 in the seeding sludge and in the
reactor provided the evidence in favor of this denitrication
mechanism. Methylobacterium detected might have facilitated the
formation of strong aggregates in UASBR1. Small amounts of N2H4
were naturally present in the reaction medium of UASBRs. In
UASBR2 higher concentration of N2H4 was detected than in the
efuent of UASBR2, well correlating with higher TN removal rates
represented in UASBR2. Injections of intermediates and AQDS, a
low-molecular quinoid analog model compound for humic substance had an ameliorating effect only on the performance of the
SRAO. From both UASBRs, uncultured bacterium clone ATB-KS1929 belonging to the order Verrucomicrobiales, Anammox bacteria (uncultured Planctomycete clone Pla_PO55-9) and aerobic
ammonium-oxidizing bacteria (uncultured sludge bacterium clone

J. BIOSCI. BIOENG.,
ASB08 Nitrosomonas) were detected. Our results indicate that
Verrucomicrobiales may be involved in some nitrogen removal
process. In the UASBR1, sulfur cycle microorganisms were dominant, while the UASBR2 was dominated by nitrogen-cycle
microorganisms.
ACKNOWLEDGMENTS
The research was supported by the Estonian target-nanced
research project Processes in macro- and microheterogeneous and
nanoscale systems and related technological applications
(SF0180135s08) and by the Estonian Science Foundation research
project Alternative ways of anaerobic ammonium oxidation process and the ways of its usage (ETF 9370). BiotaP LLC, Estonia and
Madis Metsis, Triin Lillsaar and Jaak Simm from the Integrated
Systems Biology Centre, Tallinn University of Technology are
acknowledged for the pyrosequencing results. Salutaguse Yeast
Factory (subsidiary of Lallemand Inc.) is acknowledged for fruitful
cooperation.
References
1. Jetten, M. S., Niftrik, L. V., Strous, M., Kartal, B., Keltjens, J. T., and Op den
Camp, H. J.: Biochemistry and molecular biology of Anammox bacteria, Crit.
Rev. Biochem. Mol. Biol., 44, 65e84 (2009).
2. Jetten, M. S. M., Horn, S. J., and Van Loosdrecht, M. C. M.: Towards a more
sustainable wastewater treatment system, Water Sci. Technol., 35, 171e180
(1997).
3. Kartal, B., Kuenen, J. G., and van Loosdrecht, M. C. M.: Sewage treatment with
Anammox, Science, 328, 702e703 (2010).
4. Terada, A., Zhou, S., and Hosomi, M.: Presence and detection of anaerobic
ammonium oxidizing (Anammox) bacteria and appraisal of Anammox process
for high-strength nitrogenous wastewater treatment, Clean Technol. Environ.
Policy, 13, 759e781 (2011).
5. Ma, B., Zhang, S., Zhang, L., Yi, P., Wang, J., Wang, S., and Peng, Y.: The
feasibility of using a two-stage autotrophic nitrogen removal process to treat
sewage, Bioresour. Technol., 102, 8331e8334 (2011).
6. Strous, M., Kuenen, J. G., Fuerst, J. A., Wagner, M., and Jetten, M. S. M.: The
Anammox case e a new experimental manifesto for microbiological ecophysiology, Antonie van Leeuwenhoek, 81, 693e702 (2002).
7. Liu, S., Yang, F., Gong, Z., Meng, F., Chen, H., Xue, Y., and Furukawa, K.:
Application of anaerobic ammonium-oxidizing consortium to achieve
completely autotrophic ammonium and sulfate removal, Bioresour. Technol.,
99, 6817e6825 (2008).
8. Fdz-Polanco, M., Fernndez, N., Uruea, M. A., Garci, P. A., and
Villaverde, S.: Combining the biological nitrogen and sulfur cycles in anaerobic
conditions, Water Sci. Technol., 44, 77e84 (2001).
9. Zhang, L., Zheng, P., He, Y. H., and Jin, R. C.: Performance of sulfate-dependent
anaerobic ammonium oxidation, Sci. China Ser. B Chem., 52, 86e92 (2009).
10. Schrum, H. N., Spivack, A. J., Kastner, M., and DHondt, S.: Sulfate-reducing
ammonium oxidation: a thermodynamically feasible metabolic pathway in
subseaoor sediments, Geology, 37, 939e942 (2009).
11. Elshahed, M. S., Youssef, N. H., Luo, Q., Najar, F. Z., Roe, B. A., Sisk, T. M.,
Bhring, S. I., Hinrichs, K.-U., and Krumholz, L. R.: Phylogenetic and metabolic diversity of planctomycetes from anaerobic, sulde- and sulfur-rich Zodletone Spring, Oklahoma, Appl. Environ. Microbiol., 73, 4707e4716 (2007).
12. Rikmann, E., Zekker, I., Tomingas, M., Tenno, T., Menert, A., Loorits, L., and
Tenno, T.: Sulfate-reducing anaerobic ammonium oxidation as a potential
treatment method for high nitrogen-content wastewater, Biodegradation, 23,
509e524 (2012).
13. Aranda-Tamaura, C., Estrada-Alvarado, M. I., Texier, A.-C., Cuervo, F.,
Gmez, J., and Cervantes, F. J.: Effects of different quinoid redox mediators on
the removal of sulde and nitrate via denitrication, Chemosphere, 69,
1722e1727 (2007).
14. Zekker, I., Rikmann, E., Tenno, T., Menert, A., Lemmiksoo, V., Saluste, A., and
Tenno, T.: Modication of nitrifying biolm into nitritating one by combination
of increased free ammonia concentrations, lowered HRT and dissolved oxygen
concentration, J. Environ. Sci., 23, 1113e1121 (2011).
15. Zekker, I., Rikmann, E., Tenno, T., Lemmiksoo, V., Menert, A., Loorits, L.,
Vabame, P., Tomingas, M., and Tenno, T.: Anammox enrichment from reject
water on blank biolm carriers and carriers containing nitrifying biomass:
operation of two moving bed biolm reactors (MBBR), Biodegradation, 23,
547e560 (2012).
16. Jeanningros, Y., Vlaeminck, S. E., Kaldate, A., Verstraete, W., and
Graveleau, L.: Fast start-up of a pilot-scale deammonication sequencing batch

VOL. 118, 2014

17.

18.

19.

20.

21.

22.

23.

24.

25.

COMPARISON OF SULFATE-REDUCING AND CONVENTIONAL ANAMMOX

reactor from an activated sludge inoculum, Water Sci. Technol., 61, 1393e1400
(2010).
Yang, Z., Zhou, S., and Sun, Y.: Start-up of simultaneous removal of
ammonium and sulfate from an anaerobic ammonium oxidation (Anammox)
process in an anaerobic up-ow bioreactor, J. Hazard. Mater., 169, 113e118
(2009).
Jing, C., JianXiang, J., and Ping, Z.: Isolation and identication of bacteria
responsible for simultaneous anaerobic ammonium and sulfate removal, Sci.
China Ser. B Chem., 53, 645e650 (2010).
Zhao, Q.-L., Li, W., and You, S.-J.: Simultaneous removal of ammonium-nitrogen and sulfate from wastewaters with an anaerobic attached-growth
bioreactor, Water Sci. Technol., 54, 27e35 (2006).
Sabumon, P. C.: Effect of potential electron acceptors on anoxic ammonia
oxidation in the presence of organic carbon, J. Hazard. Mater., 172, 280e288
(2009).
Zub, S., Kurissoo, T., Menert, A., and Blonskaja, V.: Combined biological
treatment of high-sulfate wastewater from yeast production, Water Environ. J.,
22, 274e286 (2008).
Koplimaa, M., Menert, A., Blonskaja, V., Kurissoo, T., Zub, S., Saareleht, M.,
Vaarmets, E., and Menert, T.: Liquid and gas chromatographic studies of the
anaerobic degradation of bakers yeast wastewater, Procedia Chem., 2,
120e129 (2010).
Nesb, C. L., Kumaraswamy, R., Dlutek, M., Doolittle, W. F., and Foght, J.:
Searching for mesophilic Thermotogales bacteria: Mesotogas in the wild,
Appl. Environ. Microbiol., 76, 4896e4900 (2010).
Nunoura, T., Hirai, M., Imachi, H., Miyazaki, M., Makita, H., Hirayama, H.,
Furushima, Y., Yamamoto, H., and Takai, K.: Kosmotoga arenicorallina sp. nov.
a thermophilic and obligately anaerobic heterotroph isolated from a shallow
hydrothermal system occurring within a coral reef, southern part of the
Yaeyama Archipelago, Japan, reclassication of Thermococcoides shengliensis as
Kosmotoga shengliensis comb. nov., and emended description of the genus
Kosmotoga, Arch. Microbiol., 192, 811e819 (2010).
Ettwig, K. F., Butler, M. K., Le Paslier, D., Pelletier, E., Mangenot, S. M. M.,
Kuypers, M. M., Schreiber, F., Dutilh, B. E., Zedelius, J., de Beer, D., and other

26.

27.

28.

29.
30.

31.

32.

33.

34.

35.

433

12 authors: Nitrite-driven anaerobic methane oxidation by oxygenic bacteria,

Nature, 464, 543e548 (2010).
Li, W., Zhao, Q.-L., and Liu, H.: Sulde removal by simultaneous autotrophic
and heterotrophic desulfurizationedenitrication process, J. Hazard. Mater.,
162, 848e853 (2009).
Heitmann, T., Goldhammer, T., Beer, J., and Blodau, C.: Electron transfer of
dissolved organic matter and its potential signicance for anaerobic respiration
in a northern bog, Glob. Change Biol., 13, 1771e1785 (2007).
Van der Zee, F. P., Bisschops, I. A. E., Blanchard, V. G., Bouwman, R. H. M.,
Lettinga, G., and Field, J. A.: The contribution of biotic and abiotic processes
during azo dye reduction in anaerobic sludge, Water Res., 37, 3098e3109 (2003).
Lovley, D., Fraga, J. L., Coates, J. D., and Blunt-Harris, E. L.: Humics as an
electron donor for anaerobic respiration, Environ. Microbiol., 1, 89e98 (1999).
Coates, J. D., Cole, K. A., Chakraborty, R., OConnor, S. M., and
Achenbach, L. A.: Diversity and ubiquity of bacteria capable of utilizing humic
substances as electron donors for anaerobic respiration, Appl. Environ.
Microbiol., 68, 2445e2452 (2002).
Eller, G. and Frenzel, P.: Changes in activity and community structure of
methane-oxidizing bacteria over the growth period of rice, Appl. Environ.
Microbiol., 67, 2395e2403 (2001).
Muyzer, G., De Waal, E. C., and Uitterlinden, A. G.: Proling of complex microbial population by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplied genes coding for 16S rRNA, Appl. Environ.
Microbiol., 59, 695e700 (1993).
Lane, D. J.: 16/23S rRNA sequencing, pp. 177e204, in: Stackebrandt, E. and
Goodfellow, M. (Eds.), Nucleic acid techniques in bacterial systematics. Wiley,
Chichester, UK (1991).
Neef, A., Amann, R. I., Schlesner, H., and Schleifer, K. H.: Monitoring a
widespread bacterial group: in situ detection of planctomycetes with 16S rRNAtargeted probes, Microbiology, 144, 3257e3266 (1998).
Sa nchez-Melsi, A., Ciliz, J., Balaguer, M. D., Colprim, J., and Vila, X.: Development of batch-culture enrichment coupled to molecular detection for
screening of natural and man-made environments in search of Anammox bacteria for N-removal bioreactors systems, Chemosphere, 75, 169e179 (2009).