- Email: [email protected]
Contributions of fermentative acidogenic bacteria and sulfate-reducing bacteria to lactate degradation and sulfate reduction
Available online at www.sciencedirect.com
Chemosphere 72 (2008) 233–242 www.elsevier.com/locate/chemosphere
Contributions of fermentative acidogenic bacteria and sulfate-reducing bacteria to lactate degradation and sulfate reduction Yangguo Zhao a,b, Nanqi Ren a,*, Aijie Wang a a
School of Municipal and Environmental Engineering, Harbin Institute of Technology, Harbin 150090, China b College of Environmental Science and Engineering, Ocean University of China, Qingdao 266100, China Received 17 September 2007; received in revised form 20 January 2008; accepted 22 January 2008 Available online 10 March 2008
Abstract The roles of fermentative acidogenic bacteria and sulfate-reducing bacteria (SRB) in lactate degradation and sulfate reduction in a sulfidogenic bioreactor were investigated by traditional chemical monitoring and culture-independent methods. A continuously stirred tank reactor fed with synthetic wastewater containing lactate and SO2 4 at 35 °C, 10 h of hydraulic retention time was used. The results showed that sulfate removal efficiency reached 99%, and sulfide and acetate were the main end products after 20 d of operation. 16S rRNA gene based clone libraries and single-strand conformation polymorphism profiles demonstrated that the proportion of SRB increased from 16% to 95%, and that Desulfobulbus spp., Desulfovibrio spp., Pseudomonas spp. and Clostridium spp. formed a stable, dominant community structure. The decreasing COD/SO2 4 ratio had little effect on the community pattern except that Pseudomonas spp. and Desulfobulbus spp. increased slightly. The addition of molybdate to the influent significantly changed the microbial community, sulfate removal efficiency and the pattern of end products. Clostridium spp., Bacteroides spp. and Ruminococcus spp. became the dominant community members. The main end products switched from acetate to ethanol and then to propionate with the oxidation– reduction potentials increasing from 420 to 290 mV. A lactate degradation pathway was deduced: lactate served as the electronic donor for Desulfovibrio spp., or was fermented by Clostridium spp. and Bacteroides spp. to produce propionate or ethanol, which were subsequently utilized by Desulfobulbus spp. and Desulfovibrio spp. The acidotrophic SRB oxidized part of the acetate finally. Ó 2008 Elsevier Ltd. All rights reserved. Keywords: Community structure; Lactate; Molybdate; Single-strand conformation polymorphism
1. Introduction The presence of sulfate-reducing bacteria (SRB) often causes the failure of anaerobic high-sulfate wastewater digestion due to their ability to out-compete methaneproducing bacteria (MPB) for similar substrates (Parkin et al., 1990). Therefore, early reports on the ecology of SRB were primarily focused on the relationship between SRB and MPB or hydrogen-producing acetogenic bacteria (HPAB). SRB have been frequently shown to out-compete *
Corresponding author. Tel./fax: +86 451 86282008. E-mail address: [email protected] (N. Ren).
0045-6535/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2008.01.046
MPB or HPAB for substrates under most conditions (Parkin et al., 1990; Uberoi and Bhattacharya, 1995), especially in the acidogenic phase of two-phase anaerobic digestion systems (Reis et al., 1988). During the acidogenic phase, the most abundant microbial group, as well as SRB, is the fermentative acidogenic bacteria (FAB) (Kalyuzhnyi et al., 1998), not MPB. The FAB play a key role in degrading macro-organics to hydrogen, ethanol and volatile fatty acids (VFAs), which are then utilized by SRB to reduce sulfate. The symbiotic relationship between SRB and FAB has been examined (Kalyuzhnyi et al., 1998); although no direct microbiological evidence has been presented previously. For example, some researchers have shown that
234
Y. Zhao et al. / Chemosphere 72 (2008) 233–242
changes in the COD/SO2 4 (C/S) ratio and the addition of molybdate affected the sulfate removal efficiency (SRE), which was ascribed to changes in the microbial community structure (Cadavid et al., 1999; Vossoughi et al., 2003). Nemati et al. (2001) concluded that addition of molybdate could suppress the activity of SRB, but would not affect the overall composition of the microbial community. As such, further study using microbiological techniques is required to elucidate the actual relationship between the SRB and FAB in degrading organic substances. At a low concentration, molybdate is a metabolic inhibitor of SRB and shows little effect on other microbes (Smith and Klug, 1981; Sørensen et al., 1981). Hence, it has been commonly used to investigate the relationship between the SRB and MPB in previous research. Smith and Klug (1981) and Nemati et al. (2001) showed that 0.095 mM and 0.2 mM molybdate was enough to suppress the production of H2S by pure cultures and by eutrophic lake sediments. In this study, 0.1–0.4 mM molybdate was used to inhibit SRB in a bioreactor in order to illuminate the relationship between SRB and FAB using the polymerase chain reaction (PCR)-single-strand conformation polymorphism (SSCP) technique. SRB can utilize more than 100 different organic substances, although lactate has been shown to be the preferred electron donor (Song et al., 1998; Kaksonen et al., 2004). FAB, such as some members of genera Clostridium and Bacteroides, can ferment lactate to propionate, acetate, ethanol and hydrogen (Macy et al., 1978; vander Wielen et al., 2002). Hence, lactate is a favorite substrate for both SRB and FAB and therefore was used in the present study. This study used traditional chemical monitoring and the culture-independent PCR-SSCP method to demonstrate the relationship between SRB and FAB, and their contributions to lactate degradation. 2. Materials and methods 2.1. Reactor operation and analytical methods A continuously stirred tank reactor (CSTR) with a 1.35 l capacity and a 1 l working volume was used. Seed sludge was obtained from Shuangcheng moat sediment (Heilongjiang, China), which had suspended solids and volatile suspended solids of 166 and 27 g l1, respectively. The sludge was incubated in the bioreactor at 35 °C, with a 10 h of hydraulic retention time and a stirring speed of 200 rpm. The bioreactor operation involved three phases based on the composition of the influent. During startup (Phase I), the synthetic wastewater contained 4000 mg COD l1 of 1 lactate, 2000 mg sulfate (SO2 and a small quantity of 4 )l (NH4)2HPO4 at a final COD:N:P ratio of 100:5:1. After startup, COD in the influent decreased to 1000 mg l1 for 15 d (Phase II) and then returned to Phase I. Subsequently, molybdate was fed to the influent with an increasing gradient of 0.1–0.4 mM (Phase III). Sludge from the CSTR was periodically sampled for total DNA extraction.
2 and pH in the bioreacThe COD, alkalinity, SO2 4 , S tor were determined using standard methods (APHA, 1998). Ethanol and VFAs, including acetate and propionate, were analyzed as described previously (Ren et al., 1997). The oxidation–reduction potential (ORP) in the reactor (Eh) was calculated by the equation Eh = Ec + 249.1 mV, where Ec is the observed ORP measured by an acidity voltmeter (pHS-25, Shanghai Analytical Apparatus Corporation, Shanghai China) and 249.1 is the potential value of the saturated calomel electrode.
2.2. DNA extraction and PCR amplification Total DNA was extracted from 0.25 g (wet weight) activated sludge with a PowerSoil DNA kit (MoBio Laboratories, CA USA) according to manufacturer’s instructions. Finally, the total DNA was suspended in 100 ll of 2 mM Tris–HCl (pH 8.0). The concentration and purity of the DNA were estimated by agarose gel electrophoresis and ultraviolet spectrometry (Beckman Coulter DU800, CA USA). Two sets of primers were used to amplify the partial 16S rRNA genes of the bacterial community and SRB group of d-Proteobacteria for SSCP analysis, respectively. For Bacteria, the forward primer BSF8/20 and the reverse primer BSR534/18 (Edwards et al., 1989) were utilized, corresponding to the 8–27 bp and 534–515 bp of the 16S rRNA gene of Escherichia coli. The second primer pair, SRB385F (Amann et al., 1990) and 926R (Schwieger and Tebbe, 1998), corresponded to the 385–402 bp and 926–907 bp of the 16S rRNA gene of E. coli. The SRB385F primer is group-specific for SRB of the d-Proteobacteria as well as some Gram-positive bacteria. Both reverse primers were phosphorylated. For microbial composition analysis, the bacterial 16S rRNA gene based clone libraries were constructed with the primer pair BSF8/20 and 926R using the sludge on days 0, 19 and 47, respectively. All primers were acquired from Invitrogen (Shanghai, China). Each sample was amplified using four 25 ll volumes in PCR tubes containing 1PCR buffer with Mg2+, deoxynucleoside triphosphate solution (200 lM each), primers (0.6 lM each), and 0.125 U of EX Taq DNA polymerase (Takara, Dalian China). Approximately 5 ng of genomic DNA was added to each PCR mixture. A GeneAmp thermal cycler (Perkin–Elmer, MA USA) was used for PCR using the following program: 94 °C for 5 min, followed by 30 cycles at 94 °C for 40 s, 50 °C for 30 s and 72 °C for 40 s with a final extension step at 72 °C for 10 min. The purity and quantity of PCR products were estimated by running a 1% agarose gel and comparing their brightness to the quantitative marker DL2000 (Takara, Dalian China). 2.3. Construction of 16S rRNA gene based clone libraries The PCR products of day 0 (seed sludge), day 19 (LM5) and day 47 (LM12) were purified with an agarose gel recov-
Y. Zhao et al. / Chemosphere 72 (2008) 233–242
ery kit (NucleoSpinÒ Extract II, Macherey-Nagel, Germany) and then cloned into the pMD19-T vector (Takara, Dalian China). Positive clones were screened by blue-white selection and PCR methods (Sambrook et al., 1989) and were then sequenced using M13 universal primers with a model 3730 sequencer (Applied Biosystems, CA USA). Similarity searching of nucleotide sequences was performed using the Blast program at the NCBI website (http:// www.ncbi.nlm.nih.gov/). The sequences were aligned and the phylogenetic and molecular evolutionary analyses of the aligned sequences were conducted using the neighborjoining method of MEGA version 4.0 (Tamura et al., 2007). 2.4. SSCP profiles To obtain single-stranded DNA from PCR products, the phosphorylated strand was removed by k exonuclease digestion. k exonuclease (20 U; New England Biolabs, MA USA) was mixed with 30 ll of the PCR product in a total volume of 100 ll. The reaction mixture was incubated at 37 °C for 2 h. Protein was removed by phenol–chloroform extraction as described by Sambrook et al. (1989), and the single-stranded DNA was resuspended in 20 ll of ddH2O. To generate an SSCP pattern, 10 ll of the singlestranded product was combined with 5 ll SSCP loading buffer (Schwieger and Tebbe, 1998), and denatured at 98 °C for 5 min and immediately cooled on ice. The product was separated on a 0.67 MDE gel matrix (Cambrex Bio Science, ME USA) at a voltage of 300 V for 24 h at 20 °C (PowerPac 1000; Bio-Rad Laboratories, CA USA). The DNA was then silver stained as described by Bassam et al. (1991), and the electrophoretogram was scanned with a Powerlook 1000 (UMAX Technologies Inc., TX, USA). 2.5. Recovering, sequencing and analyzing the SSCP band sequences Recovery of the single-stranded DNA from the bands was performed using the method described by Schwieger and Tebbe (1998). The eluted DNA was reamplified using the same primer sets and conditions described above, purified using an agarose gel recovery kit (NucleoSpinÒ Extract II, Macherey-Nagel, Germany), and cloned into the pMD19-T vector (Takara, Dalian China). Five clones per band were sequenced using the M13 universal primers by the model 3730 sequencer (Applied Biosystems, CA USA). The integrated phylogenetic tree was obtained by combining different trees using the reference sequences. 2.6. Nucleotide sequence accession numbers The nucleotide sequences obtained in this study have been deposited in the GenBank nucleotide sequence database. The partial 16S rRNA gene sequences of the seed sludge are under accession numbers DQ088228 to
235
DQ088273 and DQ444158 to DQ444188, and the sequences from the sludge of days 19 and 47 are under accession numbers DQ443880 to DQ443926. Both groups of SSCP sequences are under accession numbers DQ325465 to DQ325513. 3. Results 3.1. Sulfate removal efficiency Fig. 1 shows the performance of the sulfidogenic CSTR after 64 d of operation. The SRE reached 98% after day 15. The ORP decreased slowly to 410 mV and effluent alkalinity increased to 4000 mg l1. The end products changed from propionate (accounting for 70% (w/w) of the end products) to acetate (95%). A drop in the C/S ratio resulted in a decline in the SRE and the effluent alkalinity with a slight increase of the ORP. However, this had little effect on the pattern of end products and the effluent pH. By contrast, the addition of molybdate significantly changed the composition of end products and the ORP, as well as the SRE and the effluent alkalinity. When molybdate was added, the SRE decreased to 6% (day 51) while the ORP climbed to approximately 290 mV. Ethanol and propionate became the main end products. Finally, acetate became the primary end product when molybdate was removed from the influent. 3.2. Community succession of bacteria and SRB-like groups The SSCP profiles of the bacterial community and the SRB group are shown in Figs. 2 and 3. The integrated phylogenetic tree based on the SSCP band sequences and the library sequences is shown in Fig. 4. The bacteria appeared to compete for better ecological niches during startup (day 0–19). Some of the dominant populations in the seed sludge were washed out and new populations were gradually enriched. The drop in C/S ratio had little effect on the bacterial community structure except for some SRB (Fig. 3). Addition of molybdate drastically changed the microbial community structure with new populations becoming dominant (Fig. 2). Fig. 5 demonstrates the distribution of the SSCP band sequences, which were obtained from the integrated phylogenetic tree. The sequences with similarity to members of the phylum Firmicutes and the sub-phylum d-Proteobacteria predominated in both SSCP groups. 3.3. Composition and diversity of the microbial community The microbial composition of the seed sludge and the activated sludge of days 19 and 47 are shown in Fig. 5. Three 16S rRNA gene clone libraries obtained 77, 20 and 27 sequences, respectively. The diversity of microbial populations was highest in the seed sludge and distributed within 9 known phyla or sub-phyla (phylogenetic tree not shown). The phyla Bacteroidetes and Firmicutes accounted for 29% and 18% of the community, respectively.
Y. Zhao et al. / Chemosphere 72 (2008) 233–242 COD/SO42- = 0.5
a
80
600
60 400 40 200
20 0
0 influent alkalinity
-1
Alkalinity (mg l )
6000
effluent alkalinity
influent pH
effluent pH 9 8 7 6 5 4 3 2 1 0
b
5000 4000 3000 2000 1000
-1
Volatile fatty acid (mg l )
0
acetic acid 2500
ethanol
propionic acid
ORP -250
c
2000
-1
sulfide concentration 800
Sulfide concentration (mg l )
COD removal efficiency
pH
Removal efficiency (%)
sulfate removal efficiency 100
Molybdate
-300
1500
-350
1000
-400
500
-450
0
Oxidation-reduction potential (mV)
236
-500 0
5
10
15
20
25
30
35
40
45
50
55
60
Time (d) COD/SO2 4
Fig. 1. Performance of the sulfidogenic CSTR. The = 0.5 indicates that the C/S ratio decreased from 2 to 0.5 by reducing the COD to 1000 mg l1; ‘molybdate’ indicates the addition of 0.1–0.4 mM of molybdate to the influent.
Fig. 2. SSCP analyses of PCR amplified 16S rRNA gene sequences of the V1–V3 regions. The lanes (0–59) indicate that bacterial community samples were taken periodically during operation of days 0–59. The COD/SO2 4 = 0.5 indicates that the C/S ratio decreased from 2 to 0.5 by reducing the COD to 1000 mg l1; ‘molybdate’ indicates the addition of 0.1–0.4 mM of molybdate to the influent. Solid triangles (N) indicate the sequenced bands and the corresponding band names are shown to the right.
The sub-phyla a-Proteobacteria and d-Proteobacteria accounted for 17% and 16%, respectively. On 19th day, SRB of the d-Proteobacteria became the dominant popula-
tion and 19 sequences of 20 resembled SRB. However, none of 27 sequences obtained when molybdate was added to the influent on day 47 were similar to SRB (Fig. 4).
Y. Zhao et al. / Chemosphere 72 (2008) 233–242
237
Fig. 3. SSCP analyses of PCR-amplified 16S rRNA gene sequences from SRB-related bacterial communities taken periodically from the bioreactor activated sludge. See the legend Fig. 2 for further details.
4. Discussion 4.1. Influence of C/S ratio and molybdate on sulfate removal Research has demonstrated that the main reaction for sulfate reduction with lactate as the electronic donor by SRB is Eq. (1) (Rabus et al., 2000). In this study, the probable sulfate-reducing pathway was deduced based on the accumulation of VFAs from 16 to 19 d. As shown in Fig. 1, sulfate reduction within 4 d averaged 4.8 g l1 d1, and thus, based on Eq. (1), produced 6 g l1 d1 of acetate. However, the actual accumulation of acetate over the same period of time was only 4.8 g l1 d1, less than the predicted 6 g l1 d1. This indicates acetate-utilizing SRB (Eq. (2)) (Rabus et al., 2000) or lactate-oxidizing SRB (Eq. (3)) (Rabus et al., 2000) may exist in the bioreactor. 2CH3 CHOHCOO þ SO2 4 þ ! 2CH3 COO þ 2HCO 3 þ HS þ H
DG ¼ 160 kJ mol1 sulfate
ð1Þ
CH3 COO þ SO2 4 ! 2HCO3 þ HS
DG ¼ 48 kJ mol1 sulfate
ð2Þ
2CH3 CHOHCOO þ 3SO2 4 þ ! 6HCO 3 þ 3HS þ H
DG ¼ 85 kJ mol1 sulfate
ð3Þ
The C/S ratio should be above 0.67 if the sulfate is completely reduced (Lens et al., 1998). When COD becomes a limiting restricted factor, the SRE would drop sharply. In
this study, the SRE decreased to about 20% when the C/S ratio changed to 0.5. Therefore, for biological treatment of wastewaters containing high-sulfate levels and low concentrations of COD, such as acid mine drainage, addition of extra COD is required. Ethanol is a preferred electronic donor for SRB in addition to lactate (Kaksonen et al., 2004). When molybdate was added to the system, the SRE decreased and the activity of SRB was inhibited. Ethanol appeared and increased in the end products, which suggests that FAB ferment lactate to produce ethanol (Eq. (4)) and that SRB utilize ethanol as an electronic donor (Eq. (5)) in the bioreactor and they cooperated in lactate degradation. When molybdate was absent, the SRE continued to drop and almost all SRB were inhibited on day 51. However, the end product pattern dramatically changed from ethanol to propionate. This conversion could be ascribed to the activity of the FAB but not of the SRB, because SRB were inhibited throughout. Some of the FAB, especially the ethanol-type fermentative hydrogen-producing bacteria, are severely affected by the ORP of the bioreactor (Ren et al., 2002). If the ORP is above 300 mV, the ethanol-type fermentation would transform into mixed-acid-type fermentation. This type of fermentation also took place in the bioreactor upon startup when the ORP was higher. The conversion of end products could be attributed to the succession of functional microbial populations, which will be discussed further. During mixed-acid-type fermentation, the main end products were propionate and acetate, with the former twice the latter. This implies that the FAB are able to perform the reaction in Eq. (6) (Willis et al., 1997).
238
Y. Zhao et al. / Chemosphere 72 (2008) 233–242
Fig. 4. Phylogenetic tree based on the partial 16S rRNA gene sequences obtained from clone libraries and the SSCP bands. Both library and SSCP band sequences are displayed. The clone library sequences are bolded and are designated by ‘LM5’ (day 19) or ‘LM12’ (day 47) plus sequence numbers; the SSCP band sequences are bolded and are designated by ‘Eu-b’ or ‘Srb-b’ plus the band numbers. The remaining 16S rRNA gene sequences were obtained from the GenBank database. The tree was built using the MEGA 4.0 program with the neighbor-joining method. Numbers on the nodes indicated the bootstrap support values (1000 bootstrap replicates); bootstrap values below 50% are not presented. The scale bar indicates 5% sequence divergence.
Y. Zhao et al. / Chemosphere 72 (2008) 233–242 100
Chloroflexi Planctomycetes Fusobacteria Gemmatimonadetes -Proteobacteria -Proteobacteria -Proteobacteria -Proteobacteria Bacteroidetes Firmicutes
Distribution (%)
80
60 40 20
SRB SSCP
bacterial SSCP
LM12 (d 47)
LM5 (d 19)
seed sludge (d 0)
0
Fig. 5. The proportion and distribution of 16S rRNA gene sequences obtained from clone libraries and SSCP bands, respectively. The pattern representing the different phyla or sub-phyla is shown on the right.
SRE increased with the decrease in propionate and ORP. Therefore, we conclude that some propionate-utilizing SRB were enriched in the bioreactor. As Eq. (7) (Lien et al., 1998) shows, when propionate is consumed, a quantity of acetate is produced. CH3 CHOHCOO– þ H2 O ! CH3 CH2 OH þ HCO 3 DG ¼ 13:8 kJ mol1
ð4Þ
þ CH3 CH2 OH þ SO2 4 ! 2HCO3 þ HS þ H þ H2 O
DG ¼ 66:4 kJ mol1 sulfate
ð5Þ
3CH3 CHOHCOO– ! 2CH3 CH2 COO– þ CH3 COO– þ CO2 DG ¼ 54:9 kJ mol1
ð6Þ
4CH3 CH2 COO– þ 3SO2– 4 ! 4CH3 COO– þ 4HCO–3 þ 3HS þ Hþ DG ¼ 51 kJ mol1 sulfate
ð7Þ
To summarize, the end product pattern conversion from acetate to ethanol was a result of SRB inhibition. The conversion from ethanol to propionate was ascribed to the increase of ORP; and the reduction of propionate was attributed to the enrichment of propionate-utilizing SRB. 4.2. The roles of sulfide in the bioreactor The sulfide produced in the bioreactor could be recovered as sulfur by physical or biological processes or used to bind heavy metals for the treatment of acid mine drainage. In this system, the sulfide played several additional roles. Based on the method to determine COD, sulfide acts as part of the COD and 1 mol of S2 equalizes approximately 64 g of COD (Isa et al., 1986). Sulfide was one of the strongest reducing agents in the bioreactor and maintained the ORP at a low level. When the SRB were inhibited and the amount of sulfide decreased, the ORP climbed slowly from 422 to 286 mV, which
239
resulted in the conversion of fermentation types. In contrast, when molybdate was removed, the sulfide concentration increased rapidly with the drop in ORP (until 421 mV) (Fig. 1). Hence, the ORP and sulfide concentration were negatively correlated and sulfide was necessary for sustaining the higher activity of SRB in this sulfidogenic system. The sulfide was also important for alkalinity as well. Two mol of alkalinity are produced reduced according to Eqs. (1), (2), (3), per mol SO2 4 (5), and (7). Consequently, sulfide positively correlated with the effluent alkalinity. In addition, sulfide inhibits the activity of microbes (Kuo and Shu, 2004), and different microbial groups have different levels of sensitivity to sulfide (Rabus et al., 2000). Sulfide toxicity is associated with the undissociated form which is permeable to the cell membrane. Total inhibition of growth is generally obtained at concentrations below 550 mg free H2S l1 (Kalyuzhnyi et al., 1998). In this study, when the SRB were inhibited and the sulfide concentration was low, FAB were metabolically active and grew quickly due to freedom from sulfide inhibition and demonstrated two types of fermentation. During bioreactor operation with high sulfide levels (310 ± 213 mg l1), most of the sulfide was as HS due to the high pH (pH 8) (Kalyuzhnyi et al., 1998; Mora-Naranjo et al., 2003). The toxicity of HS to SRB and other microbes is less than that of H2S (Mora-Naranjo et al., 2003). Therefore, FAB and SRB were able to cooperatively degrade lactate and reduce sulfate in the presence of high concentrations of sulfide. 4.3. Succession of microbial populations As shown in Figs. 2–4, at the end of startup, most of the dominant seed sludge populations had been gradually washed out and a stable community pattern, containing bands Eu-b1, Eu-b5, Eu-b16, Eu-b18 and Eu-b21, was formed. A Blastn search showed that these band sequences were similar to Desulfobulbus spp. (Eu-b1 and Eu-b21), Pseudomonas putida (Eu-b5) and Desulfovibrio desulfuricans (Eu-b18 and Eu-b16). Comparing the community composition with the SRE, it showed that the community comprising these dominant populations had as much as a 99% SRE. Desulfobulbus spp., Desulfovibrio spp. and Desulfuromonas thiophila (Srb-b24) were all incomplete oxidizing SRB and could not use acetate as an electronic donor (Finster et al., 1997). Desulfotomaculum kuznetsovii (Eub11) was a thermophilic SRB and could oxidize many types of VFAs during sulfate reduction (Nazina et al., 1988). Some members of genus Desulforhabdus (Srb-b1, Srbb15) were also SRB, which could oxidize formate, acetate, propionate and ethanol during sulfate reduction (Oude Elferink et al., 1995). These results provided microbiological evidence of acetate utilization and confirmed the existence of Eqs. (1) and (2). The decreased C/S ratio had little influence on the bacterial community pattern (Fig. 2), while it did affect
240
Y. Zhao et al. / Chemosphere 72 (2008) 233–242
the SRB-like populations (Fig. 3). The Desulfobulbus spp. (Fig. 3, Srb-b18, Srb-b20–23, similar to Desulfobulbus rhabdoformis (Lien et al., 1998) with a 97–99% identity) appeared and were enriched. Hence, compared with other populations, SRB were apt to be disturbed by the changes in the C/S ratio. As such, a high and steady C/S ratio should be maintained when treating highsulfate wastewater in view of the characteristics of the SRB populations. The addition of molybdate evidently altered the dominant populations, which differs from the results of Nemati et al. (2001). The primary SRB bands (Eu-b1, Eu-b21) gradually died out, while some subordinate FAB bands increased on days 43–47 (Fig. 2). From days 51–55, new dominant population patterns were formed, which differed from those of days 43–47 when the SRB populations were inhibited. The new dominant populations were mainly comprised of Clostridium spp. (Eu-b4, Eu-b70 ), Bacteroides spp. (Eu-b14) and Ruminococcus spp. (Srb-b9, Srb-b16), and all were able to ferment lactate to produce propionate and acetate via the propionate pathway (Eq. (6)) (Macy et al., 1978; vander Wielen et al., 2002). This might be the primary reason for the conversion of the end product pattern from ethanol to propionate (Fig. 1). The decrease in propionate could be for different reasons. First, ORP dropped to 400 mV due to the fact that the SRB recovered and the concentration of sulfide rose; this condition would favor FAB producing ethanol but not propionate. And ethanol could be utilized by some members of Desulfovibrio to reduce sulfate (Willis et al., 1997). Second, the amount of SRB, especially the genus Desulfobulbus, increased. Desulfobulbus spp. could use propionate, lactate and ethanol as electronic donors during sulfate reduction (Stams et al., 1984; Lien et al., 1998). Okabe et al. (2003) found that Desulfobulbus spp. were the dominant populations in a sulfate-rich wastewater biofilm, and played a key role in propionate oxidation and sulfate reduction. Laanbroek and Pfennig (1981) also demonstrated that the SRB which degraded propionate in freshwater and seawater sediments, were Desulfobulbus spp. One SSCP sequence (Eu-b24), which was similar to the sulfur oxidizing bacterium Sulfuricurvum kujiense, was also identified within the bacterial community. This bacterium oxidized sulfur or sulfide to form sulfate (Kodama and Watanabe, 2003), and consequently would participate in the sulfur cycle in the sulfidogenic bioreactor. In addition, the SRB SSCP profile detected a sequence that resembled the acetotrophic SRB, Desulforhabdus amnigena (Srb-b1, b15), which uses acetate as the sole carbon and energy source. Based on the bioreactor’s performance and the bacterial and SRB population analysis, we deduced the possible decomposition pathway of lactate (Fig. 6). Lactate was first fermented by Clostridium spp. and Bacteroides spp. to produce propionate or ethanol, which served as electronic donors for sulfate reduction by Desulfobulbus or Desulfovibrio spp. Alternatively, the lactate was
Fig. 6. Presumptive lactate degradation pathway in this sulfidogenic system.
directly oxidized to acetate by Desulfovibrio spp. Some of the incomplete oxidizing product acetate would be oxidized by the acetotrophic D. amnigena to form CO2 and H2O. 4.4. The relationship between sulfate removal and the amount of SRB The relative quantity of SRB reflects the sulfate-reducing power of the system. From a practical point of view, the SRE of similar systems could be improved by enriching SRB-related populations. On 19th day, the SRE was 99%, and 95% of the 16S rRNA library sequences were similar to SRB (with the exception of LM5-7, similar to Clostridium ganghwense with 95% identity, Fig. 4). SSCP band sequence analysis showed that Pseudomonas spp. (Eu-b5) were also present in the bioreactor at the same time and that Clostridium spp. were not the dominant band. On 47th day, the SRE was suppressed to 21% and the SSCP profiles showed that the bands corresponding to Desulfobulbus spp. were still present (Eu-b1 in Fig. 2, Srb-b3–4 and Srb-b19–23 in Fig. 3). However, none of the 25 library sequences resembled SRB. Therefore, there was a discrepancy between the SSCP and library sequences. This is most likely because the 16S rRNA gene based clone library could only detect the dominant populations, especially if the sequence coverage was low. The SSCP profiles were probably able to detect lower abundance populations. The 16S rRNA gene sequences of days 19 and 47 and the SSCP band sequences were combined and used to build
Y. Zhao et al. / Chemosphere 72 (2008) 233–242
a phylogenetic tree (Fig. 4). The sequences in the libraries and the SSCP bands both contained the following phyla or sub-phyla: d-Proteobacteria, Chloroflexi, Firmicutes, Bacteroidetes. The SSCP sequences also contained the SRB genus Desulfotomaculum (Eu-b11) as well as Desulfobulbus and Desulfovibrio. Desulfotomaculum belongs to the low G + C gram-positive group. The seed sludge 16S rRNA gene library also detected many other SRB genera. The phylogenetic tree constructed with the SSCP band sequences was better for investigating the diversity of microbes in this study than the tree obtained from the 16S rRNA gene library sequences. This suggests that the use of the SSCP profile combining band clones may be a viable substitute for the more traditional gene clone library techniques. Acknowledgements This work was supported by the National Natural Science Foundation of China (Grant Nos: 30470054, 50208006, and 50678049) and the China Postdoctoral Science Foundation (Grant No: 20070410266). In addition, the authors thank Dr. Joy Van Nostrand at the Institute for Environmental Genomics, University of Oklahoma, for her excellent editorial assistance. References Amann, R.I., Binder, B., Chisholm, S.W., Olsen, R., Devereux, R., Stahl, D.A., 1990. Combination of 16S rRNA targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl. Environ. Microb. 56, 1919–1925. APHA, 1998. Standard Methods for the Examination of Water and Wastewater, 20th ed. American Public Health Association, Washington, DC, USA. Bassam, B.J., Caetano-Anolles, G., Gresshoff, P.M., 1991. Fast and sensitive silver staining of DNA in polyacrylamide gels. Anal. Biochem. 196, 80–83. Cadavid, D.L., Zaiat, M., Foresti, E., 1999. Performance of horizontalflow anaerobic immobilized sludge (HAIS) reactor treating synthetic substrate subjected to decreasing COD to sulfate ratios. Water Sci. Technol. 39 (10–11), 99–106. Edwards, U., Rogall, T., Blo¨cker, H., Emde, M., Bo¨ttger, E.C., 1989. Isolation and direct sequencing of entire genes. Characterization of a gene coding for 16S ribosomal RNA. Nucl. Acids Res. 17, 7843–7853. Finster, K., Coates, J.D., Liesack, W., Pfennig, N., 1997. Desulfuromonas thiophila sp nov., a new obligately sulfur-reducing bacterium from anoxic freshwater sediment. Int. J. Syst. Bacteriol. 47, 754–758. Isa, Z., Grusenmeyer, S., Verstraete, W., 1986. Sulfate reduction relative to methane production in high-rate anaerobic digestion: microbiological aspects. Appl. Environ. Microb. 51, 580–587. Kaksonen, A.H., Plumb, J.J., Franzmann, P.D., Puhakka, J.A., 2004. Simple organic electron donors support diverse sulfate-reducing communities in fluidized-bed reactors treating acidic metal- and sulfate-containing wastewater. FEMS Microbiol. Ecol. 47, 279–289. Kalyuzhnyi, S., Fedorovich, V., Lens, P., Hulshoff Pol, L., Lettinga, G., 1998. Mathematical modeling as a tool to study population dynamics between sulfate reducing and methanogenic bacteria. Biodegradation 9, 187–199. Kodama, Y., Watanabe, K., 2003. Isolation and characterization of a sulfur-oxidizing chemolithotroph growing on crude oil under anaerobic conditions. Appl. Environ. Microb. 69, 107–112.
241
Kuo, W.C., Shu, T.Y., 2004. Biological pre-treatment of wastewater containing sulfate using anaerobic immobilized cells. J. Hazard. Mater. 113, 147–155. Laanbroek, H.J., Pfennig, N., 1981. Oxidation of short-chain fatty acids by sulfate-reducing bacteria in freshwater and marine sediments. Arch. Microbiol. 128, 330–335. Lens, P.N.L., Visser, A., Janssen, A.J.H., Pol, L.W.H., Lettinga, G., 1998. Biotechnological treatment of sulfate-rich wastewaters. Crit. Rev. Environ. Sci. Tec. 28, 41–88. Lien, T., Madsen, M., Steen, I.H., Gjerdevik, K., 1998. Desulfobulbus rhabdoformis sp nov., a sulfate reducer from a water–oil separation system. Int. J. Syst. Bacteriol. 48, 469–474. Macy, J.M., Ljungdahl, L.G., Gottschalk, G., 1978. Pathway of succinate and propionate formation in Bacteroides fragilis. J. Bacteriol. 134, 84–91. Mora-Naranjo, N., Alamar-Provecho, C., Meima, J., Haarstrick, A., Hempel, D.C., 2003. Experimental investigation and modeling of the effect of sulfate on anaerobic biodegradation processes in municipal solid waste. Water Sci. Technol. 48 (4), 221–227. Nazina, T.N., Ivanova, A.E., Kanchaveli, L.P., Rozanova, E.P., 1988. Desulfotomaculum kuznetsovii sp. nov., a new spore-forming thermophilic methylotrophic sulfate-reducing bacterium. Mikrobiologiya 57, 823–827. Nemati, M., Mazutinec, T.J., Jenneman, G.E., Voordouw, G., 2001. Control of biogenic H2 S production with nitrite and molybdate. J. Ind. Microbiol. Biot. 26, 350–355. Okabe, S., Ito, T., Satoh, H., 2003. Sulfate-reducing bacterial community structure and their contribution to carbon mineralization in a wastewater biofilm growing under microaerophilic conditions. Appl. Microbiol. Biot. 63, 322–334. Oude Elferink, S.J., Maas, R.N., Harmsen, H.J., Stams, A.J., 1995. Desulforhabdus amnigenus gen nov. sp. nov., a sulfate reducer isolated from anaerobic granular sludge. Arch. Microbiol. 164, 119–124. Parkin, G.F., Lynch, N.A., Kuo, W.C., Van Keren, E.L., Bhattacharya, S.K., 1990. Interaction between sulfate reducers and methanogens fed acetate and propionate. J. Water Pollut. Con. F. 62, 780–788. Rabus, R., Hansen, T., Widdel, F., 2000. Dissimilatory sulfate- and sulfurreducing prokaryotes. In: Dworkin, M., Falkow, S., Rosenberg, E., Schleifer, K.-H., Stackebrandt, E. (Eds.), The Prokaryotes: An Evolving Electronic Resource for the Microbiological Community, 3rd ed. Springer-Verlag, New York (Online). Reis, M.A.M., Goncalves, L.M.D., Carrondo, M.J.T., 1988. Sulfate reduction in acidogenic phase anaerobic digestion. Water Sci. Technol. 20 (11–12), 345–351. Ren, N., Wang, B., Huang, J.-C., 1997. Ethanol-type fermentation from carbonhydrate in high rate acidogenic reactor. Biotechnol. Bioeng. 54, 428–433. Ren, N., Zhao, D., Chen, X., Li, J., 2002. Mechanism and controlling strategy of the production and accumulation of propionic acid for anaerobic wastewater treatment. Sci. China Ser. B 45, 319–327. Sambrook, J.E., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. Schwieger, F., Tebbe, C.C., 1998. A new approach to utilize PCR – single strand-conformation polymorphism for 16S rRNA gene-based microbial community analysis. Appl. Environ. Microb. 64, 4870–4876. Smith, R.L., Klug, M.J., 1981. Electron donors utilized by sulfatereducing bacteria in eutrophic lake sediments. Appl. Environ. Microb. 42, 116–121. Song, Y.-C., Piak, B.-C., Shin, H.-S., La, S.-J., 1998. Influence of the electron donor and toxic materials on the activity of sulfate reducing bacteria for the treatment of electroplating wastewater. Water Sci. Technol. 38 (4–5), 187–194. Sørensen, J., Christensen, D., Jørgensen, B.B., 1981. Volatile fatty acids and hydrogen as substrates for sulfate-reducing bacteria in anaerobic marine sediment. Appl. Environ. Microb. 42, 5–11. Stams, A.J.M., Kremer, D.R., Weenk, G.H., Hansen, T.A., 1984. Pathway of propionate formation in Desulfobulbus propionicus. Arch. Microbiol. 139, 167–173.
242
Y. Zhao et al. / Chemosphere 72 (2008) 233–242
Tamura, K., Dudley, J., Nei, M., Kumar, S., 2007. MEGA4: Molecular evolutionary genetics analysis (MEGA) software version 4.0. Mol. Biol. Evol. 24, 1596–1599. Uberoi, V.U., Bhattacharya, S.K., 1995. Interaction among sulfate reducers, acetogens, and methanogens in anaerobic propionate systems. Water Environ. Res. 67, 330–339. vander Wielen, P.W., Rovers, G.M., Scheepens, J.M., Biesterveld, S., 2002. Clostridium lactatifermentans sp. nov., a lactate-fermenting
anaerobe isolated from the caeca of a chicken. Int. J. Syst. Evol. Micr. 5, 921–925. Vossoughi, M., Shakeri, M., Alemzadeh, I., 2003. Performance of anaerobic baffled reactor treating synthetic wastewater influenced by decreasing COD/SO2 4 ratios. Chem. Eng. Prog. 42, 811–816. Willis, C.L., Cummings, J.H., Neale, G., Gibson, G.R., 1997. Nutritional aspects of dissimilatory sulfate reduction in the human large intestine. Curr. Microbiol. 35, 294–298.
Chemosphere 72 (2008) 233–242 www.elsevier.com/locate/chemosphere
Contributions of fermentative acidogenic bacteria and sulfate-reducing bacteria to lactate degradation and sulfate reduction Yangguo Zhao a,b, Nanqi Ren a,*, Aijie Wang a a
School of Municipal and Environmental Engineering, Harbin Institute of Technology, Harbin 150090, China b College of Environmental Science and Engineering, Ocean University of China, Qingdao 266100, China Received 17 September 2007; received in revised form 20 January 2008; accepted 22 January 2008 Available online 10 March 2008
Abstract The roles of fermentative acidogenic bacteria and sulfate-reducing bacteria (SRB) in lactate degradation and sulfate reduction in a sulfidogenic bioreactor were investigated by traditional chemical monitoring and culture-independent methods. A continuously stirred tank reactor fed with synthetic wastewater containing lactate and SO2 4 at 35 °C, 10 h of hydraulic retention time was used. The results showed that sulfate removal efficiency reached 99%, and sulfide and acetate were the main end products after 20 d of operation. 16S rRNA gene based clone libraries and single-strand conformation polymorphism profiles demonstrated that the proportion of SRB increased from 16% to 95%, and that Desulfobulbus spp., Desulfovibrio spp., Pseudomonas spp. and Clostridium spp. formed a stable, dominant community structure. The decreasing COD/SO2 4 ratio had little effect on the community pattern except that Pseudomonas spp. and Desulfobulbus spp. increased slightly. The addition of molybdate to the influent significantly changed the microbial community, sulfate removal efficiency and the pattern of end products. Clostridium spp., Bacteroides spp. and Ruminococcus spp. became the dominant community members. The main end products switched from acetate to ethanol and then to propionate with the oxidation– reduction potentials increasing from 420 to 290 mV. A lactate degradation pathway was deduced: lactate served as the electronic donor for Desulfovibrio spp., or was fermented by Clostridium spp. and Bacteroides spp. to produce propionate or ethanol, which were subsequently utilized by Desulfobulbus spp. and Desulfovibrio spp. The acidotrophic SRB oxidized part of the acetate finally. Ó 2008 Elsevier Ltd. All rights reserved. Keywords: Community structure; Lactate; Molybdate; Single-strand conformation polymorphism
1. Introduction The presence of sulfate-reducing bacteria (SRB) often causes the failure of anaerobic high-sulfate wastewater digestion due to their ability to out-compete methaneproducing bacteria (MPB) for similar substrates (Parkin et al., 1990). Therefore, early reports on the ecology of SRB were primarily focused on the relationship between SRB and MPB or hydrogen-producing acetogenic bacteria (HPAB). SRB have been frequently shown to out-compete *
Corresponding author. Tel./fax: +86 451 86282008. E-mail address: [email protected] (N. Ren).
0045-6535/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2008.01.046
MPB or HPAB for substrates under most conditions (Parkin et al., 1990; Uberoi and Bhattacharya, 1995), especially in the acidogenic phase of two-phase anaerobic digestion systems (Reis et al., 1988). During the acidogenic phase, the most abundant microbial group, as well as SRB, is the fermentative acidogenic bacteria (FAB) (Kalyuzhnyi et al., 1998), not MPB. The FAB play a key role in degrading macro-organics to hydrogen, ethanol and volatile fatty acids (VFAs), which are then utilized by SRB to reduce sulfate. The symbiotic relationship between SRB and FAB has been examined (Kalyuzhnyi et al., 1998); although no direct microbiological evidence has been presented previously. For example, some researchers have shown that
234
Y. Zhao et al. / Chemosphere 72 (2008) 233–242
changes in the COD/SO2 4 (C/S) ratio and the addition of molybdate affected the sulfate removal efficiency (SRE), which was ascribed to changes in the microbial community structure (Cadavid et al., 1999; Vossoughi et al., 2003). Nemati et al. (2001) concluded that addition of molybdate could suppress the activity of SRB, but would not affect the overall composition of the microbial community. As such, further study using microbiological techniques is required to elucidate the actual relationship between the SRB and FAB in degrading organic substances. At a low concentration, molybdate is a metabolic inhibitor of SRB and shows little effect on other microbes (Smith and Klug, 1981; Sørensen et al., 1981). Hence, it has been commonly used to investigate the relationship between the SRB and MPB in previous research. Smith and Klug (1981) and Nemati et al. (2001) showed that 0.095 mM and 0.2 mM molybdate was enough to suppress the production of H2S by pure cultures and by eutrophic lake sediments. In this study, 0.1–0.4 mM molybdate was used to inhibit SRB in a bioreactor in order to illuminate the relationship between SRB and FAB using the polymerase chain reaction (PCR)-single-strand conformation polymorphism (SSCP) technique. SRB can utilize more than 100 different organic substances, although lactate has been shown to be the preferred electron donor (Song et al., 1998; Kaksonen et al., 2004). FAB, such as some members of genera Clostridium and Bacteroides, can ferment lactate to propionate, acetate, ethanol and hydrogen (Macy et al., 1978; vander Wielen et al., 2002). Hence, lactate is a favorite substrate for both SRB and FAB and therefore was used in the present study. This study used traditional chemical monitoring and the culture-independent PCR-SSCP method to demonstrate the relationship between SRB and FAB, and their contributions to lactate degradation. 2. Materials and methods 2.1. Reactor operation and analytical methods A continuously stirred tank reactor (CSTR) with a 1.35 l capacity and a 1 l working volume was used. Seed sludge was obtained from Shuangcheng moat sediment (Heilongjiang, China), which had suspended solids and volatile suspended solids of 166 and 27 g l1, respectively. The sludge was incubated in the bioreactor at 35 °C, with a 10 h of hydraulic retention time and a stirring speed of 200 rpm. The bioreactor operation involved three phases based on the composition of the influent. During startup (Phase I), the synthetic wastewater contained 4000 mg COD l1 of 1 lactate, 2000 mg sulfate (SO2 and a small quantity of 4 )l (NH4)2HPO4 at a final COD:N:P ratio of 100:5:1. After startup, COD in the influent decreased to 1000 mg l1 for 15 d (Phase II) and then returned to Phase I. Subsequently, molybdate was fed to the influent with an increasing gradient of 0.1–0.4 mM (Phase III). Sludge from the CSTR was periodically sampled for total DNA extraction.
2 and pH in the bioreacThe COD, alkalinity, SO2 4 , S tor were determined using standard methods (APHA, 1998). Ethanol and VFAs, including acetate and propionate, were analyzed as described previously (Ren et al., 1997). The oxidation–reduction potential (ORP) in the reactor (Eh) was calculated by the equation Eh = Ec + 249.1 mV, where Ec is the observed ORP measured by an acidity voltmeter (pHS-25, Shanghai Analytical Apparatus Corporation, Shanghai China) and 249.1 is the potential value of the saturated calomel electrode.
2.2. DNA extraction and PCR amplification Total DNA was extracted from 0.25 g (wet weight) activated sludge with a PowerSoil DNA kit (MoBio Laboratories, CA USA) according to manufacturer’s instructions. Finally, the total DNA was suspended in 100 ll of 2 mM Tris–HCl (pH 8.0). The concentration and purity of the DNA were estimated by agarose gel electrophoresis and ultraviolet spectrometry (Beckman Coulter DU800, CA USA). Two sets of primers were used to amplify the partial 16S rRNA genes of the bacterial community and SRB group of d-Proteobacteria for SSCP analysis, respectively. For Bacteria, the forward primer BSF8/20 and the reverse primer BSR534/18 (Edwards et al., 1989) were utilized, corresponding to the 8–27 bp and 534–515 bp of the 16S rRNA gene of Escherichia coli. The second primer pair, SRB385F (Amann et al., 1990) and 926R (Schwieger and Tebbe, 1998), corresponded to the 385–402 bp and 926–907 bp of the 16S rRNA gene of E. coli. The SRB385F primer is group-specific for SRB of the d-Proteobacteria as well as some Gram-positive bacteria. Both reverse primers were phosphorylated. For microbial composition analysis, the bacterial 16S rRNA gene based clone libraries were constructed with the primer pair BSF8/20 and 926R using the sludge on days 0, 19 and 47, respectively. All primers were acquired from Invitrogen (Shanghai, China). Each sample was amplified using four 25 ll volumes in PCR tubes containing 1PCR buffer with Mg2+, deoxynucleoside triphosphate solution (200 lM each), primers (0.6 lM each), and 0.125 U of EX Taq DNA polymerase (Takara, Dalian China). Approximately 5 ng of genomic DNA was added to each PCR mixture. A GeneAmp thermal cycler (Perkin–Elmer, MA USA) was used for PCR using the following program: 94 °C for 5 min, followed by 30 cycles at 94 °C for 40 s, 50 °C for 30 s and 72 °C for 40 s with a final extension step at 72 °C for 10 min. The purity and quantity of PCR products were estimated by running a 1% agarose gel and comparing their brightness to the quantitative marker DL2000 (Takara, Dalian China). 2.3. Construction of 16S rRNA gene based clone libraries The PCR products of day 0 (seed sludge), day 19 (LM5) and day 47 (LM12) were purified with an agarose gel recov-
Y. Zhao et al. / Chemosphere 72 (2008) 233–242
ery kit (NucleoSpinÒ Extract II, Macherey-Nagel, Germany) and then cloned into the pMD19-T vector (Takara, Dalian China). Positive clones were screened by blue-white selection and PCR methods (Sambrook et al., 1989) and were then sequenced using M13 universal primers with a model 3730 sequencer (Applied Biosystems, CA USA). Similarity searching of nucleotide sequences was performed using the Blast program at the NCBI website (http:// www.ncbi.nlm.nih.gov/). The sequences were aligned and the phylogenetic and molecular evolutionary analyses of the aligned sequences were conducted using the neighborjoining method of MEGA version 4.0 (Tamura et al., 2007). 2.4. SSCP profiles To obtain single-stranded DNA from PCR products, the phosphorylated strand was removed by k exonuclease digestion. k exonuclease (20 U; New England Biolabs, MA USA) was mixed with 30 ll of the PCR product in a total volume of 100 ll. The reaction mixture was incubated at 37 °C for 2 h. Protein was removed by phenol–chloroform extraction as described by Sambrook et al. (1989), and the single-stranded DNA was resuspended in 20 ll of ddH2O. To generate an SSCP pattern, 10 ll of the singlestranded product was combined with 5 ll SSCP loading buffer (Schwieger and Tebbe, 1998), and denatured at 98 °C for 5 min and immediately cooled on ice. The product was separated on a 0.67 MDE gel matrix (Cambrex Bio Science, ME USA) at a voltage of 300 V for 24 h at 20 °C (PowerPac 1000; Bio-Rad Laboratories, CA USA). The DNA was then silver stained as described by Bassam et al. (1991), and the electrophoretogram was scanned with a Powerlook 1000 (UMAX Technologies Inc., TX, USA). 2.5. Recovering, sequencing and analyzing the SSCP band sequences Recovery of the single-stranded DNA from the bands was performed using the method described by Schwieger and Tebbe (1998). The eluted DNA was reamplified using the same primer sets and conditions described above, purified using an agarose gel recovery kit (NucleoSpinÒ Extract II, Macherey-Nagel, Germany), and cloned into the pMD19-T vector (Takara, Dalian China). Five clones per band were sequenced using the M13 universal primers by the model 3730 sequencer (Applied Biosystems, CA USA). The integrated phylogenetic tree was obtained by combining different trees using the reference sequences. 2.6. Nucleotide sequence accession numbers The nucleotide sequences obtained in this study have been deposited in the GenBank nucleotide sequence database. The partial 16S rRNA gene sequences of the seed sludge are under accession numbers DQ088228 to
235
DQ088273 and DQ444158 to DQ444188, and the sequences from the sludge of days 19 and 47 are under accession numbers DQ443880 to DQ443926. Both groups of SSCP sequences are under accession numbers DQ325465 to DQ325513. 3. Results 3.1. Sulfate removal efficiency Fig. 1 shows the performance of the sulfidogenic CSTR after 64 d of operation. The SRE reached 98% after day 15. The ORP decreased slowly to 410 mV and effluent alkalinity increased to 4000 mg l1. The end products changed from propionate (accounting for 70% (w/w) of the end products) to acetate (95%). A drop in the C/S ratio resulted in a decline in the SRE and the effluent alkalinity with a slight increase of the ORP. However, this had little effect on the pattern of end products and the effluent pH. By contrast, the addition of molybdate significantly changed the composition of end products and the ORP, as well as the SRE and the effluent alkalinity. When molybdate was added, the SRE decreased to 6% (day 51) while the ORP climbed to approximately 290 mV. Ethanol and propionate became the main end products. Finally, acetate became the primary end product when molybdate was removed from the influent. 3.2. Community succession of bacteria and SRB-like groups The SSCP profiles of the bacterial community and the SRB group are shown in Figs. 2 and 3. The integrated phylogenetic tree based on the SSCP band sequences and the library sequences is shown in Fig. 4. The bacteria appeared to compete for better ecological niches during startup (day 0–19). Some of the dominant populations in the seed sludge were washed out and new populations were gradually enriched. The drop in C/S ratio had little effect on the bacterial community structure except for some SRB (Fig. 3). Addition of molybdate drastically changed the microbial community structure with new populations becoming dominant (Fig. 2). Fig. 5 demonstrates the distribution of the SSCP band sequences, which were obtained from the integrated phylogenetic tree. The sequences with similarity to members of the phylum Firmicutes and the sub-phylum d-Proteobacteria predominated in both SSCP groups. 3.3. Composition and diversity of the microbial community The microbial composition of the seed sludge and the activated sludge of days 19 and 47 are shown in Fig. 5. Three 16S rRNA gene clone libraries obtained 77, 20 and 27 sequences, respectively. The diversity of microbial populations was highest in the seed sludge and distributed within 9 known phyla or sub-phyla (phylogenetic tree not shown). The phyla Bacteroidetes and Firmicutes accounted for 29% and 18% of the community, respectively.
Y. Zhao et al. / Chemosphere 72 (2008) 233–242 COD/SO42- = 0.5
a
80
600
60 400 40 200
20 0
0 influent alkalinity
-1
Alkalinity (mg l )
6000
effluent alkalinity
influent pH
effluent pH 9 8 7 6 5 4 3 2 1 0
b
5000 4000 3000 2000 1000
-1
Volatile fatty acid (mg l )
0
acetic acid 2500
ethanol
propionic acid
ORP -250
c
2000
-1
sulfide concentration 800
Sulfide concentration (mg l )
COD removal efficiency
pH
Removal efficiency (%)
sulfate removal efficiency 100
Molybdate
-300
1500
-350
1000
-400
500
-450
0
Oxidation-reduction potential (mV)
236
-500 0
5
10
15
20
25
30
35
40
45
50
55
60
Time (d) COD/SO2 4
Fig. 1. Performance of the sulfidogenic CSTR. The = 0.5 indicates that the C/S ratio decreased from 2 to 0.5 by reducing the COD to 1000 mg l1; ‘molybdate’ indicates the addition of 0.1–0.4 mM of molybdate to the influent.
Fig. 2. SSCP analyses of PCR amplified 16S rRNA gene sequences of the V1–V3 regions. The lanes (0–59) indicate that bacterial community samples were taken periodically during operation of days 0–59. The COD/SO2 4 = 0.5 indicates that the C/S ratio decreased from 2 to 0.5 by reducing the COD to 1000 mg l1; ‘molybdate’ indicates the addition of 0.1–0.4 mM of molybdate to the influent. Solid triangles (N) indicate the sequenced bands and the corresponding band names are shown to the right.
The sub-phyla a-Proteobacteria and d-Proteobacteria accounted for 17% and 16%, respectively. On 19th day, SRB of the d-Proteobacteria became the dominant popula-
tion and 19 sequences of 20 resembled SRB. However, none of 27 sequences obtained when molybdate was added to the influent on day 47 were similar to SRB (Fig. 4).
Y. Zhao et al. / Chemosphere 72 (2008) 233–242
237
Fig. 3. SSCP analyses of PCR-amplified 16S rRNA gene sequences from SRB-related bacterial communities taken periodically from the bioreactor activated sludge. See the legend Fig. 2 for further details.
4. Discussion 4.1. Influence of C/S ratio and molybdate on sulfate removal Research has demonstrated that the main reaction for sulfate reduction with lactate as the electronic donor by SRB is Eq. (1) (Rabus et al., 2000). In this study, the probable sulfate-reducing pathway was deduced based on the accumulation of VFAs from 16 to 19 d. As shown in Fig. 1, sulfate reduction within 4 d averaged 4.8 g l1 d1, and thus, based on Eq. (1), produced 6 g l1 d1 of acetate. However, the actual accumulation of acetate over the same period of time was only 4.8 g l1 d1, less than the predicted 6 g l1 d1. This indicates acetate-utilizing SRB (Eq. (2)) (Rabus et al., 2000) or lactate-oxidizing SRB (Eq. (3)) (Rabus et al., 2000) may exist in the bioreactor. 2CH3 CHOHCOO þ SO2 4 þ ! 2CH3 COO þ 2HCO 3 þ HS þ H
DG ¼ 160 kJ mol1 sulfate
ð1Þ
CH3 COO þ SO2 4 ! 2HCO3 þ HS
DG ¼ 48 kJ mol1 sulfate
ð2Þ
2CH3 CHOHCOO þ 3SO2 4 þ ! 6HCO 3 þ 3HS þ H
DG ¼ 85 kJ mol1 sulfate
ð3Þ
The C/S ratio should be above 0.67 if the sulfate is completely reduced (Lens et al., 1998). When COD becomes a limiting restricted factor, the SRE would drop sharply. In
this study, the SRE decreased to about 20% when the C/S ratio changed to 0.5. Therefore, for biological treatment of wastewaters containing high-sulfate levels and low concentrations of COD, such as acid mine drainage, addition of extra COD is required. Ethanol is a preferred electronic donor for SRB in addition to lactate (Kaksonen et al., 2004). When molybdate was added to the system, the SRE decreased and the activity of SRB was inhibited. Ethanol appeared and increased in the end products, which suggests that FAB ferment lactate to produce ethanol (Eq. (4)) and that SRB utilize ethanol as an electronic donor (Eq. (5)) in the bioreactor and they cooperated in lactate degradation. When molybdate was absent, the SRE continued to drop and almost all SRB were inhibited on day 51. However, the end product pattern dramatically changed from ethanol to propionate. This conversion could be ascribed to the activity of the FAB but not of the SRB, because SRB were inhibited throughout. Some of the FAB, especially the ethanol-type fermentative hydrogen-producing bacteria, are severely affected by the ORP of the bioreactor (Ren et al., 2002). If the ORP is above 300 mV, the ethanol-type fermentation would transform into mixed-acid-type fermentation. This type of fermentation also took place in the bioreactor upon startup when the ORP was higher. The conversion of end products could be attributed to the succession of functional microbial populations, which will be discussed further. During mixed-acid-type fermentation, the main end products were propionate and acetate, with the former twice the latter. This implies that the FAB are able to perform the reaction in Eq. (6) (Willis et al., 1997).
238
Y. Zhao et al. / Chemosphere 72 (2008) 233–242
Fig. 4. Phylogenetic tree based on the partial 16S rRNA gene sequences obtained from clone libraries and the SSCP bands. Both library and SSCP band sequences are displayed. The clone library sequences are bolded and are designated by ‘LM5’ (day 19) or ‘LM12’ (day 47) plus sequence numbers; the SSCP band sequences are bolded and are designated by ‘Eu-b’ or ‘Srb-b’ plus the band numbers. The remaining 16S rRNA gene sequences were obtained from the GenBank database. The tree was built using the MEGA 4.0 program with the neighbor-joining method. Numbers on the nodes indicated the bootstrap support values (1000 bootstrap replicates); bootstrap values below 50% are not presented. The scale bar indicates 5% sequence divergence.
Y. Zhao et al. / Chemosphere 72 (2008) 233–242 100
Chloroflexi Planctomycetes Fusobacteria Gemmatimonadetes -Proteobacteria -Proteobacteria -Proteobacteria -Proteobacteria Bacteroidetes Firmicutes
Distribution (%)
80
60 40 20
SRB SSCP
bacterial SSCP
LM12 (d 47)
LM5 (d 19)
seed sludge (d 0)
0
Fig. 5. The proportion and distribution of 16S rRNA gene sequences obtained from clone libraries and SSCP bands, respectively. The pattern representing the different phyla or sub-phyla is shown on the right.
SRE increased with the decrease in propionate and ORP. Therefore, we conclude that some propionate-utilizing SRB were enriched in the bioreactor. As Eq. (7) (Lien et al., 1998) shows, when propionate is consumed, a quantity of acetate is produced. CH3 CHOHCOO– þ H2 O ! CH3 CH2 OH þ HCO 3 DG ¼ 13:8 kJ mol1
ð4Þ
þ CH3 CH2 OH þ SO2 4 ! 2HCO3 þ HS þ H þ H2 O
DG ¼ 66:4 kJ mol1 sulfate
ð5Þ
3CH3 CHOHCOO– ! 2CH3 CH2 COO– þ CH3 COO– þ CO2 DG ¼ 54:9 kJ mol1
ð6Þ
4CH3 CH2 COO– þ 3SO2– 4 ! 4CH3 COO– þ 4HCO–3 þ 3HS þ Hþ DG ¼ 51 kJ mol1 sulfate
ð7Þ
To summarize, the end product pattern conversion from acetate to ethanol was a result of SRB inhibition. The conversion from ethanol to propionate was ascribed to the increase of ORP; and the reduction of propionate was attributed to the enrichment of propionate-utilizing SRB. 4.2. The roles of sulfide in the bioreactor The sulfide produced in the bioreactor could be recovered as sulfur by physical or biological processes or used to bind heavy metals for the treatment of acid mine drainage. In this system, the sulfide played several additional roles. Based on the method to determine COD, sulfide acts as part of the COD and 1 mol of S2 equalizes approximately 64 g of COD (Isa et al., 1986). Sulfide was one of the strongest reducing agents in the bioreactor and maintained the ORP at a low level. When the SRB were inhibited and the amount of sulfide decreased, the ORP climbed slowly from 422 to 286 mV, which
239
resulted in the conversion of fermentation types. In contrast, when molybdate was removed, the sulfide concentration increased rapidly with the drop in ORP (until 421 mV) (Fig. 1). Hence, the ORP and sulfide concentration were negatively correlated and sulfide was necessary for sustaining the higher activity of SRB in this sulfidogenic system. The sulfide was also important for alkalinity as well. Two mol of alkalinity are produced reduced according to Eqs. (1), (2), (3), per mol SO2 4 (5), and (7). Consequently, sulfide positively correlated with the effluent alkalinity. In addition, sulfide inhibits the activity of microbes (Kuo and Shu, 2004), and different microbial groups have different levels of sensitivity to sulfide (Rabus et al., 2000). Sulfide toxicity is associated with the undissociated form which is permeable to the cell membrane. Total inhibition of growth is generally obtained at concentrations below 550 mg free H2S l1 (Kalyuzhnyi et al., 1998). In this study, when the SRB were inhibited and the sulfide concentration was low, FAB were metabolically active and grew quickly due to freedom from sulfide inhibition and demonstrated two types of fermentation. During bioreactor operation with high sulfide levels (310 ± 213 mg l1), most of the sulfide was as HS due to the high pH (pH 8) (Kalyuzhnyi et al., 1998; Mora-Naranjo et al., 2003). The toxicity of HS to SRB and other microbes is less than that of H2S (Mora-Naranjo et al., 2003). Therefore, FAB and SRB were able to cooperatively degrade lactate and reduce sulfate in the presence of high concentrations of sulfide. 4.3. Succession of microbial populations As shown in Figs. 2–4, at the end of startup, most of the dominant seed sludge populations had been gradually washed out and a stable community pattern, containing bands Eu-b1, Eu-b5, Eu-b16, Eu-b18 and Eu-b21, was formed. A Blastn search showed that these band sequences were similar to Desulfobulbus spp. (Eu-b1 and Eu-b21), Pseudomonas putida (Eu-b5) and Desulfovibrio desulfuricans (Eu-b18 and Eu-b16). Comparing the community composition with the SRE, it showed that the community comprising these dominant populations had as much as a 99% SRE. Desulfobulbus spp., Desulfovibrio spp. and Desulfuromonas thiophila (Srb-b24) were all incomplete oxidizing SRB and could not use acetate as an electronic donor (Finster et al., 1997). Desulfotomaculum kuznetsovii (Eub11) was a thermophilic SRB and could oxidize many types of VFAs during sulfate reduction (Nazina et al., 1988). Some members of genus Desulforhabdus (Srb-b1, Srbb15) were also SRB, which could oxidize formate, acetate, propionate and ethanol during sulfate reduction (Oude Elferink et al., 1995). These results provided microbiological evidence of acetate utilization and confirmed the existence of Eqs. (1) and (2). The decreased C/S ratio had little influence on the bacterial community pattern (Fig. 2), while it did affect
240
Y. Zhao et al. / Chemosphere 72 (2008) 233–242
the SRB-like populations (Fig. 3). The Desulfobulbus spp. (Fig. 3, Srb-b18, Srb-b20–23, similar to Desulfobulbus rhabdoformis (Lien et al., 1998) with a 97–99% identity) appeared and were enriched. Hence, compared with other populations, SRB were apt to be disturbed by the changes in the C/S ratio. As such, a high and steady C/S ratio should be maintained when treating highsulfate wastewater in view of the characteristics of the SRB populations. The addition of molybdate evidently altered the dominant populations, which differs from the results of Nemati et al. (2001). The primary SRB bands (Eu-b1, Eu-b21) gradually died out, while some subordinate FAB bands increased on days 43–47 (Fig. 2). From days 51–55, new dominant population patterns were formed, which differed from those of days 43–47 when the SRB populations were inhibited. The new dominant populations were mainly comprised of Clostridium spp. (Eu-b4, Eu-b70 ), Bacteroides spp. (Eu-b14) and Ruminococcus spp. (Srb-b9, Srb-b16), and all were able to ferment lactate to produce propionate and acetate via the propionate pathway (Eq. (6)) (Macy et al., 1978; vander Wielen et al., 2002). This might be the primary reason for the conversion of the end product pattern from ethanol to propionate (Fig. 1). The decrease in propionate could be for different reasons. First, ORP dropped to 400 mV due to the fact that the SRB recovered and the concentration of sulfide rose; this condition would favor FAB producing ethanol but not propionate. And ethanol could be utilized by some members of Desulfovibrio to reduce sulfate (Willis et al., 1997). Second, the amount of SRB, especially the genus Desulfobulbus, increased. Desulfobulbus spp. could use propionate, lactate and ethanol as electronic donors during sulfate reduction (Stams et al., 1984; Lien et al., 1998). Okabe et al. (2003) found that Desulfobulbus spp. were the dominant populations in a sulfate-rich wastewater biofilm, and played a key role in propionate oxidation and sulfate reduction. Laanbroek and Pfennig (1981) also demonstrated that the SRB which degraded propionate in freshwater and seawater sediments, were Desulfobulbus spp. One SSCP sequence (Eu-b24), which was similar to the sulfur oxidizing bacterium Sulfuricurvum kujiense, was also identified within the bacterial community. This bacterium oxidized sulfur or sulfide to form sulfate (Kodama and Watanabe, 2003), and consequently would participate in the sulfur cycle in the sulfidogenic bioreactor. In addition, the SRB SSCP profile detected a sequence that resembled the acetotrophic SRB, Desulforhabdus amnigena (Srb-b1, b15), which uses acetate as the sole carbon and energy source. Based on the bioreactor’s performance and the bacterial and SRB population analysis, we deduced the possible decomposition pathway of lactate (Fig. 6). Lactate was first fermented by Clostridium spp. and Bacteroides spp. to produce propionate or ethanol, which served as electronic donors for sulfate reduction by Desulfobulbus or Desulfovibrio spp. Alternatively, the lactate was
Fig. 6. Presumptive lactate degradation pathway in this sulfidogenic system.
directly oxidized to acetate by Desulfovibrio spp. Some of the incomplete oxidizing product acetate would be oxidized by the acetotrophic D. amnigena to form CO2 and H2O. 4.4. The relationship between sulfate removal and the amount of SRB The relative quantity of SRB reflects the sulfate-reducing power of the system. From a practical point of view, the SRE of similar systems could be improved by enriching SRB-related populations. On 19th day, the SRE was 99%, and 95% of the 16S rRNA library sequences were similar to SRB (with the exception of LM5-7, similar to Clostridium ganghwense with 95% identity, Fig. 4). SSCP band sequence analysis showed that Pseudomonas spp. (Eu-b5) were also present in the bioreactor at the same time and that Clostridium spp. were not the dominant band. On 47th day, the SRE was suppressed to 21% and the SSCP profiles showed that the bands corresponding to Desulfobulbus spp. were still present (Eu-b1 in Fig. 2, Srb-b3–4 and Srb-b19–23 in Fig. 3). However, none of the 25 library sequences resembled SRB. Therefore, there was a discrepancy between the SSCP and library sequences. This is most likely because the 16S rRNA gene based clone library could only detect the dominant populations, especially if the sequence coverage was low. The SSCP profiles were probably able to detect lower abundance populations. The 16S rRNA gene sequences of days 19 and 47 and the SSCP band sequences were combined and used to build
Y. Zhao et al. / Chemosphere 72 (2008) 233–242
a phylogenetic tree (Fig. 4). The sequences in the libraries and the SSCP bands both contained the following phyla or sub-phyla: d-Proteobacteria, Chloroflexi, Firmicutes, Bacteroidetes. The SSCP sequences also contained the SRB genus Desulfotomaculum (Eu-b11) as well as Desulfobulbus and Desulfovibrio. Desulfotomaculum belongs to the low G + C gram-positive group. The seed sludge 16S rRNA gene library also detected many other SRB genera. The phylogenetic tree constructed with the SSCP band sequences was better for investigating the diversity of microbes in this study than the tree obtained from the 16S rRNA gene library sequences. This suggests that the use of the SSCP profile combining band clones may be a viable substitute for the more traditional gene clone library techniques. Acknowledgements This work was supported by the National Natural Science Foundation of China (Grant Nos: 30470054, 50208006, and 50678049) and the China Postdoctoral Science Foundation (Grant No: 20070410266). In addition, the authors thank Dr. Joy Van Nostrand at the Institute for Environmental Genomics, University of Oklahoma, for her excellent editorial assistance. References Amann, R.I., Binder, B., Chisholm, S.W., Olsen, R., Devereux, R., Stahl, D.A., 1990. Combination of 16S rRNA targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl. Environ. Microb. 56, 1919–1925. APHA, 1998. Standard Methods for the Examination of Water and Wastewater, 20th ed. American Public Health Association, Washington, DC, USA. Bassam, B.J., Caetano-Anolles, G., Gresshoff, P.M., 1991. Fast and sensitive silver staining of DNA in polyacrylamide gels. Anal. Biochem. 196, 80–83. Cadavid, D.L., Zaiat, M., Foresti, E., 1999. Performance of horizontalflow anaerobic immobilized sludge (HAIS) reactor treating synthetic substrate subjected to decreasing COD to sulfate ratios. Water Sci. Technol. 39 (10–11), 99–106. Edwards, U., Rogall, T., Blo¨cker, H., Emde, M., Bo¨ttger, E.C., 1989. Isolation and direct sequencing of entire genes. Characterization of a gene coding for 16S ribosomal RNA. Nucl. Acids Res. 17, 7843–7853. Finster, K., Coates, J.D., Liesack, W., Pfennig, N., 1997. Desulfuromonas thiophila sp nov., a new obligately sulfur-reducing bacterium from anoxic freshwater sediment. Int. J. Syst. Bacteriol. 47, 754–758. Isa, Z., Grusenmeyer, S., Verstraete, W., 1986. Sulfate reduction relative to methane production in high-rate anaerobic digestion: microbiological aspects. Appl. Environ. Microb. 51, 580–587. Kaksonen, A.H., Plumb, J.J., Franzmann, P.D., Puhakka, J.A., 2004. Simple organic electron donors support diverse sulfate-reducing communities in fluidized-bed reactors treating acidic metal- and sulfate-containing wastewater. FEMS Microbiol. Ecol. 47, 279–289. Kalyuzhnyi, S., Fedorovich, V., Lens, P., Hulshoff Pol, L., Lettinga, G., 1998. Mathematical modeling as a tool to study population dynamics between sulfate reducing and methanogenic bacteria. Biodegradation 9, 187–199. Kodama, Y., Watanabe, K., 2003. Isolation and characterization of a sulfur-oxidizing chemolithotroph growing on crude oil under anaerobic conditions. Appl. Environ. Microb. 69, 107–112.
241
Kuo, W.C., Shu, T.Y., 2004. Biological pre-treatment of wastewater containing sulfate using anaerobic immobilized cells. J. Hazard. Mater. 113, 147–155. Laanbroek, H.J., Pfennig, N., 1981. Oxidation of short-chain fatty acids by sulfate-reducing bacteria in freshwater and marine sediments. Arch. Microbiol. 128, 330–335. Lens, P.N.L., Visser, A., Janssen, A.J.H., Pol, L.W.H., Lettinga, G., 1998. Biotechnological treatment of sulfate-rich wastewaters. Crit. Rev. Environ. Sci. Tec. 28, 41–88. Lien, T., Madsen, M., Steen, I.H., Gjerdevik, K., 1998. Desulfobulbus rhabdoformis sp nov., a sulfate reducer from a water–oil separation system. Int. J. Syst. Bacteriol. 48, 469–474. Macy, J.M., Ljungdahl, L.G., Gottschalk, G., 1978. Pathway of succinate and propionate formation in Bacteroides fragilis. J. Bacteriol. 134, 84–91. Mora-Naranjo, N., Alamar-Provecho, C., Meima, J., Haarstrick, A., Hempel, D.C., 2003. Experimental investigation and modeling of the effect of sulfate on anaerobic biodegradation processes in municipal solid waste. Water Sci. Technol. 48 (4), 221–227. Nazina, T.N., Ivanova, A.E., Kanchaveli, L.P., Rozanova, E.P., 1988. Desulfotomaculum kuznetsovii sp. nov., a new spore-forming thermophilic methylotrophic sulfate-reducing bacterium. Mikrobiologiya 57, 823–827. Nemati, M., Mazutinec, T.J., Jenneman, G.E., Voordouw, G., 2001. Control of biogenic H2 S production with nitrite and molybdate. J. Ind. Microbiol. Biot. 26, 350–355. Okabe, S., Ito, T., Satoh, H., 2003. Sulfate-reducing bacterial community structure and their contribution to carbon mineralization in a wastewater biofilm growing under microaerophilic conditions. Appl. Microbiol. Biot. 63, 322–334. Oude Elferink, S.J., Maas, R.N., Harmsen, H.J., Stams, A.J., 1995. Desulforhabdus amnigenus gen nov. sp. nov., a sulfate reducer isolated from anaerobic granular sludge. Arch. Microbiol. 164, 119–124. Parkin, G.F., Lynch, N.A., Kuo, W.C., Van Keren, E.L., Bhattacharya, S.K., 1990. Interaction between sulfate reducers and methanogens fed acetate and propionate. J. Water Pollut. Con. F. 62, 780–788. Rabus, R., Hansen, T., Widdel, F., 2000. Dissimilatory sulfate- and sulfurreducing prokaryotes. In: Dworkin, M., Falkow, S., Rosenberg, E., Schleifer, K.-H., Stackebrandt, E. (Eds.), The Prokaryotes: An Evolving Electronic Resource for the Microbiological Community, 3rd ed. Springer-Verlag, New York (Online). Reis, M.A.M., Goncalves, L.M.D., Carrondo, M.J.T., 1988. Sulfate reduction in acidogenic phase anaerobic digestion. Water Sci. Technol. 20 (11–12), 345–351. Ren, N., Wang, B., Huang, J.-C., 1997. Ethanol-type fermentation from carbonhydrate in high rate acidogenic reactor. Biotechnol. Bioeng. 54, 428–433. Ren, N., Zhao, D., Chen, X., Li, J., 2002. Mechanism and controlling strategy of the production and accumulation of propionic acid for anaerobic wastewater treatment. Sci. China Ser. B 45, 319–327. Sambrook, J.E., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. Schwieger, F., Tebbe, C.C., 1998. A new approach to utilize PCR – single strand-conformation polymorphism for 16S rRNA gene-based microbial community analysis. Appl. Environ. Microb. 64, 4870–4876. Smith, R.L., Klug, M.J., 1981. Electron donors utilized by sulfatereducing bacteria in eutrophic lake sediments. Appl. Environ. Microb. 42, 116–121. Song, Y.-C., Piak, B.-C., Shin, H.-S., La, S.-J., 1998. Influence of the electron donor and toxic materials on the activity of sulfate reducing bacteria for the treatment of electroplating wastewater. Water Sci. Technol. 38 (4–5), 187–194. Sørensen, J., Christensen, D., Jørgensen, B.B., 1981. Volatile fatty acids and hydrogen as substrates for sulfate-reducing bacteria in anaerobic marine sediment. Appl. Environ. Microb. 42, 5–11. Stams, A.J.M., Kremer, D.R., Weenk, G.H., Hansen, T.A., 1984. Pathway of propionate formation in Desulfobulbus propionicus. Arch. Microbiol. 139, 167–173.
242
Y. Zhao et al. / Chemosphere 72 (2008) 233–242
Tamura, K., Dudley, J., Nei, M., Kumar, S., 2007. MEGA4: Molecular evolutionary genetics analysis (MEGA) software version 4.0. Mol. Biol. Evol. 24, 1596–1599. Uberoi, V.U., Bhattacharya, S.K., 1995. Interaction among sulfate reducers, acetogens, and methanogens in anaerobic propionate systems. Water Environ. Res. 67, 330–339. vander Wielen, P.W., Rovers, G.M., Scheepens, J.M., Biesterveld, S., 2002. Clostridium lactatifermentans sp. nov., a lactate-fermenting
anaerobe isolated from the caeca of a chicken. Int. J. Syst. Evol. Micr. 5, 921–925. Vossoughi, M., Shakeri, M., Alemzadeh, I., 2003. Performance of anaerobic baffled reactor treating synthetic wastewater influenced by decreasing COD/SO2 4 ratios. Chem. Eng. Prog. 42, 811–816. Willis, C.L., Cummings, J.H., Neale, G., Gibson, G.R., 1997. Nutritional aspects of dissimilatory sulfate reduction in the human large intestine. Curr. Microbiol. 35, 294–298.