In a sulfate reducing process, increasing loading rates and sulfide accumulation may induce population changes resulting in decreasing effectiveness of the process. Thus, the relationship between microbial metabolism changes and population dynamics was studied. An upflow anaerobic sludge blanket reactor was operated at different sulfate loading rates (SLR), from 290 to 981 mg SO4 − S/L d at a constant carbon/sulfur ratio of 0.75. When the SLR was increased, the total organic carbon and sulfate consumption efficiencies decreased to nearly 30% and 25%, respectively. The acetate and propionate yields increased with increasing SLR and 385 ± 7 mg sulfide-S/L d was reached. The ecological indices, determined by random amplified polymorphic DNA and denaturing gradient gel electrophoresis techniques, diversity and evenness were found to be constant, and similarity coefficient values remained higher than 76%. The results suggest that the microbial population changes were negligible compared with metabolic changes when SLR was increased. The sulfide accumulation did not modify the microbial diversity. The sequencing of 16S rRNA genes showed strains related to sulfate reducing, fermentation, and methanogenesis processes. The results indicated that the decreasing of effectiveness, under the experimental conditions tested, was dependent more on operational parameters than microbial changes.

INTRODUCTION

The biological sulfate reducing process (SRP) is a dissimilative anoxic respiratory process, where sulfate is reduced to sulfide using electron donors present in the environment. The SRP may be used to remove sulfate from wastewater generated from many industrial processes. The application of the SRP in the treatment of chemical, mining, and galvanic industrial wastewaters is increasing (Mohammadi et al. 2010; Sánchez-Andrea et al. 2014). A limitation of the biological treatment of sulfate-rich wastewaters is the sulfide accumulation that may inhibit microbial respiratory pathways (Muyzer & Stams 2008). Thus, the effectiveness of the SRP decreases as sulfide increases. The decreasing effectiveness might also be associated with changes in microbial populations due to modifications in environmental conditions. This is relevant in wastewater treatment, since operational performance can be a function of changes in microbial composition and/or changes in operational parameters.

The population dynamics have been studied in order to relate the microbial findings to the performance of the bioreactor (Zhao et al. 2008; Briones et al. 2009). However, there are few studies that analyze the microbial structure under steady-state (SS) conditions; of these studies, results can appear contradictory. For example, Fernández et al. (1999) observed changes in the microbial community of a methanogenic reactor while the respiratory process was apparently at SS. Conversely, LaPara et al. (2002) observed a stable microbial community during SS. The apparent contradiction between said studies might be attributed to numerous factors including: design and operational parameters, type of microbial species, diversity of microbial communities, and so on. Therefore, to understand the population dynamics, more studies during SS conditions are needed. Likewise, studies about the relationship between the community structure and the respiratory pattern in sulfate reducing reactors are still scarce. Dar et al. (2007), using ethanol and sulfate in an upflow anaerobic sludge blanket reactor (UASB) reactor, obtained a sulfate consumption efficiency of 93% with four different strains of sulfate reducing bacteria. Kaksonen et al. (2004) reported sulfate consumption efficiency of 77–95% in a fluidized-bed reactor fed with ethanol and with a relatively more diverse consortium of sulfate reducing bacteria. These studies reached a high efficiency but with different microbial diversity and composition. This suggests that the operational parameters such as: bioreactor design, residence time, loading rates, pH value, etc. might be just as important as the microbial diversity and composition.

Microbial analysis is needed to identify the relationship between metabolism and microbial populations of the system. The use of molecular biology techniques is useful for studying the structure and dynamics of microbial populations. Random amplified polymorphic DNA (RAPD) has been used on activated sludge (Xie 2004), and denaturing gradient gel electrophoresis (DGGE) has also been used for studying microbial populations (Kaksonen et al. 2004). Nevertheless, if both population dynamics and metabolic behavior studies are simultaneously carried out, it would be possible to obtain evidence for establishing whether the metabolic profiles are mainly associated with microbial changes or environmental conditions. The use of molecular biology methods as tools for estimating the ecological indices (richness, biodiversity, evenness, and similarity) as well as physiological response variables, such as efficiency and yield values, could be a more complete strategy to study a microbial consortium under SS conditions.

The aim of this work was to assess the effect of different sulfate and lactate loading rates in sludge from a UASB reactor on the microbial community dynamics and its metabolic profile under SS sulfate reducing conditions. Ecological indices were used to determine whether the changes in substrate consumption efficiencies and yield values are linked with one of or both microbial community and environmental changes.

METHODS

UASB reactor

A UASB reactor of 1.5 L inoculated with sulfate reducing sludge was fed at five sulfate loading rates (SLR, mg SO4 − S/L d: 290, 436, 654, 817, and 981). The sludge source was from the vinasse and brewery industry. The carbon/sulfur (C/S) ratio used was close to the stoichiometric value (0.75) at each SLR. The reactor was operated at 32 °C for 202 days, one day of hydraulic retention time, and fed with media F1 and F2. F1 contained sulfate source (Na2SO4) while F2 provided the electron source (sodium lactate). Definition and formulation of mineral media are shown in Table 1. NH4Cl was used as nitrogen source and adjusted to a carbon/nitrogen (C/N) ratio value of 50. The pH of FI and F2 was adjusted to 7 ± 0.1. Inorganic carbon measured as HCO3, organic carbon, acetate, propionate, CO2, CH4, sulfate, dissolved sulfide, and total and volatile suspended solids (VSS) contents were determined as described in Garcia-Saucedo et al. (2008). At the pH used in this study, HCO3 is the main chemical species (around 85%) and CO2 is 15%. Lactate-C was determined as the difference of total organic carbon (TOC) minus acetate-C and propionate-C. Statistical analyses were conducted with the Number Cruncher Statistical System 2001 software (Hintze 2001). The respiratory process was evaluated by means of substrate consumption efficiencies (E, mg substrate consumed/mg substrate fed) and yield products (Y, mg product/mg substrate consumed).

Table 1

Definition and formulation of mineral media: medium F1 and F2

Mineral media (mg/L) 
F1 F2 
NH4Cl 16–46 CoCl2 0.5 
MgCl2 CaCl2 
NaMoO4 0.5 CuSO4 
FeCl3 NiCl2 0.8 
Na2SeO3 ZnCl2 0.5 
  KH2PO4 2,000 
Mineral media (mg/L) 
F1 F2 
NH4Cl 16–46 CoCl2 0.5 
MgCl2 CaCl2 
NaMoO4 0.5 CuSO4 
FeCl3 NiCl2 0.8 
Na2SeO3 ZnCl2 0.5 
  KH2PO4 2,000 

DNA amplification by RAPD

Sludge homogeneous samples were withdrawn from the UASB reactor at each SLR. Total DNA was extracted from 500 mg of sludge by means of an UltraClean™ Soil DNA Isolation Kit (MO BIO Laboratories, Solana Beach, CA, USA). Total DNA of the community was analyzed. Polymerase chain reaction (PCR) amplification reactions were carried out in a volume of 50 μL, containing 1× magnesium free buffer, magnesium chloride 2.5 mM, nucleotide mix (200 μM of each dNTP), 2.5 units of Taq DNA polymerase (Promega, USA), 40 pmol of each primer (Table 2), and 200 ng of sludge DNA. The primers were selected considering previous studies (This et al. 1997; Quednau et al. 1998; Ruimy et al. 2001; Xie 2004). DNA amplification was performed with a Mastercycler thermocycler (Eppendorf, Hamburg, Germany) as follows: 30 cycles for 45 s at 94 °C, 45 s at 40 or 45 °C (Table 2), and 90 s at 72 °C. The amplified products were analyzed by agarose gel electrophoresis, stained with ethidium bromide and visualized in a Gel Doc 2000 UV transilluminator by using Quantity One software (Bio-RAD, Hercules, CA, USA). Reactions and gels were done in duplicate.

Table 2

List of the random primers used for RAPD

Code Sequence (5′–3′) Guanine + cytosine (%) T annealing (°C) T reference (°C) 
F01 ACG CGC CCT 77 45 36a 
A09 GGG TAA CGC C 70 45 38b 
LE CTG CTG GGA C 70 40 36c 
6.1 ACC CGG TCA C 70 45 36d 
6.2 TTC GAG CCA G 60 45 36d 
6.3 GTG AGG CGT C 70 45 36d 
6.9 TGG ACC GGT G 70 45 36d 
Code Sequence (5′–3′) Guanine + cytosine (%) T annealing (°C) T reference (°C) 
F01 ACG CGC CCT 77 45 36a 
A09 GGG TAA CGC C 70 45 38b 
LE CTG CTG GGA C 70 40 36c 
6.1 ACC CGG TCA C 70 45 36d 
6.2 TTC GAG CCA G 60 45 36d 
6.3 GTG AGG CGT C 70 45 36d 
6.9 TGG ACC GGT G 70 45 36d 

PCR and DGGE analysis of 16 rRNA genes

The V6–V8 regions of the bacterial 16S rRNA genes were amplified using universal bacterial primers 968GC-f and 1401-r (Nübel et al. 1996). The V2–V3 regions of archaeal 16S rRNA genes were amplified using the primers A109 (T)-f and 515GC-r (Roest et al. 2005). PCR amplifications were performed as described by de Bok et al. (2006) but with annealing temperatures of 54 and 53 °C for bacterial and archaeal DNA amplification, respectively (CG1-96 thermocycler, Corbett Research, Sydney, Australia). The amplification product was separated by DGGE on a DCode universal mutation detection system (Bio-RAD Laboratories, Hercules, CA, USA) (Cervantes et al. 2003), on 6% (w/v) polyacrylamide gels, using denaturant gradients of 30–55% and 40–55% for bacterial and archaeal amplicons, respectively. The 100% denaturant contained 7 M urea and 40% (v/v) formamide. The DGGE gels were stained with silver nitrate and visualized using EpiChemi3 Darkroom (UVP Bioimaging Systems, Upland, CA, USA), a Labworks Image Acquisition, and Analysis Software 4.0. DGGE bands were excised and reamplified using the previously described primers. The PCR products were purified (Wizard® SV gel and PCR Clean-Up System, Promega, Madison, WI, USA), visualized by agarose gel electrophoresis and sequenced using ABI Prism Big-Dye Terminator 3.1 Ready Reaction Cycle Sequencing on an ABI Prism 3100 Genetic Analyzer (Applied Biosystems, Warrington, UK). The DNA sequences were analyzed using Chromas (http://www.technelysium.com.au/chromas.html) and BioEdit (http://www.mbio.ncsu.edu/BioEdit/bioedit.html) programs. The identification of the partial sequences was estimated using BLAST algorithm (Basic Local Alignment Search Tool, http://www.ncbi.nlm.nih.gov/BLAST). The presence of chimerical sequences was checked with the Ribosomal Database Project Chimera Check program (http://rdp.cme.msu.edu).

Nucleotide sequence accession numbers

Sequences obtained in this study were deposited in GenBank under accession numbers EU287856–EU287864, EU287866–EU287867, and EU443144–EU443149.

Estimation of ecological indices

The richness, S, represents the total number of quantifiable and reproducible bands in a lane. Diversity index, H, is the measure of species diversity in a given community. H takes into account both abundance and evenness of species that are present in the microbial community. A high H value indicates a greater biodiversity. Likewise, if the species are evenly distributed, then the H value will be high. The term H is defined as H = –ΣPi lnPi, where Pi is the relative abundance of the bands in a lane and is calculated as Pi = ni/N, ni is the band intensity for individual bands, and N is the sum of intensities of bands in a lane (Shannon & Weaver 1963). The evenness index, J, indicates the distribution of the abundance of the different bands. J index values closer to 1 means no predominance of species within the microbial population. The J index was calculated as J = H/lnS. Band surface and pixel intensity were taken as a surrogate measure of band abundance. The similarity coefficient, C, between two populations was calculated as described by Yang et al. (2000). The C has a range from 0 to 1 with higher values indicating greater similarity than lower ones. Images were analyzed by Image J 1.37v software (http://rsb.info.nih.gov/ij/).

RESULTS AND DISCUSSION

UASB reactor performance

Profiles of sulfur and carbon compounds in the effluent of the UASB reactor are shown in Figure 1. Steady-state SRP was established at each SLR when the variation coefficient for sulfate consumption, sulfide, and HCO3 production rates was less than 11%. The production rate of HCO3, propionate, and acetate from lactate increases as SLR increases (Figure 1). The bicarbonate yield values decreased by approximately 30% at the higher SLR, whereas acetate and propionate yields increased significantly by a factor of 7 and 17, respectively (α, 0.05). Methane production was scarce, as it represented less than 7% of the TOC consumed (Figure 1).

Figure 1

Operational and performance data of the sulfate reducing sludge in the UASB reactor at five SLR, mg SO4 − S/L d. (a) mg/L d of carbon: (□) lactate-C influent; (Δ) HCO3 − C; (●) propionate-C; (○) acetate-C, and (-) methane-C. (b) mg/L d of sulfur: (□) sulfate-S influent; (▪) sulfate-S effluent, and (○) dissolved sulfide-S. I, II, III, IV, and V are SLR: 290, 436, 654, 981, and 817, respectively. The arrows represent samples of sludge taken for molecular analysis under SS conditions on days: 57, 65, 71, 79, 87, 101, 144, 165, 183, 193, and 202.

Figure 1

Operational and performance data of the sulfate reducing sludge in the UASB reactor at five SLR, mg SO4 − S/L d. (a) mg/L d of carbon: (□) lactate-C influent; (Δ) HCO3 − C; (●) propionate-C; (○) acetate-C, and (-) methane-C. (b) mg/L d of sulfur: (□) sulfate-S influent; (▪) sulfate-S effluent, and (○) dissolved sulfide-S. I, II, III, IV, and V are SLR: 290, 436, 654, 981, and 817, respectively. The arrows represent samples of sludge taken for molecular analysis under SS conditions on days: 57, 65, 71, 79, 87, 101, 144, 165, 183, 193, and 202.

The mass balance indicated that 90% of SO4 − S fed was detected in the output as residual sulfate plus dissolved sulfide produced by SRP. The dissolved sulfide-S increased proportionally while the SLR increased to a maximum of 385 ± 7 mg sulfide-S/L d (Figure 1). The dissolved sulfide yield values were close to 1. Within the ranges of SLR fed, the VSS concentration increased from 3.6 to 7.3 g/L. It must be mentioned that TOC consumption efficiency gradually decreased from 93 ± 3 to 66% ± 2 as SLR increased from 290 to 981 mg SO4 − S/L d, respectively. Sulfate consumption efficiency decreased from 60 ± 5 to 45% ± 4 in the range of 654–981 mg SO4 − S/L d. The consumption efficiency decrease is related to the acetate and propionate accumulation from the lactate fermentation. Therefore, it may be said that both lactate fermentation and lactate mineralization by SRP were simultaneously carried out. This behavior was also seen by García-Saucedo et al. (2008), and it was assumed that accumulation might possibly be related to a high Monod constant (Ks) value for the volatile fatty acids. Thus, the decrease in the SRP efficiency was not due to sulfide inhibition, but to a possible lower specific consumption rate of acetate and propionate due to a high Ks value.

Relationship of the sludge performance and microbial community at different SLR

In order to evaluate the population dynamics exposed to the different loading rates of sulfate and lactate and its possible relation to the physiological respiratory response of the consortium, the microbial community was analyzed. All samples were collected from the UASB reactor under SS conditions at each SLR and the ecological indices were used in order to detect changes in the microbial populations. This study was divided into two parts: (1) the analysis of the microbial populations during SS at one SLR (290 mg SO4 − S/L d) for a period of 48 days and (2) the analysis of the microbial populations during changes in the SLR. The banding patterns of bacterial and archaeal communities obtained by DGGE at different SLR and days are shown in Figure 2. The presence of 11 and 8 distinguishable bands in archaeal and bacterial communities, respectively, was observed. Few differences in position, intensity, and number of bands were noted. H and J values remained constant: 1.45 ± 0.03 and 0.89 ± 0.02 for the bacterial community, and 2.33 ± 0.09 and 0.97 ± 0.01 for the archaeal community, respectively, during a period of 48 days in SS at 290 mg SO4 − S/L d (lanes 2–7). The results found by RAPD technique using ecological indices (Table 3) were coincidental to those obtained with the DGGE technique at 290 mg SO4 − S/L d and 217 mg lactate-C/L d, whereas the similarity coefficient (C) values were higher than 0.97 (Table 4). Thus, these results indicated stable microbial communities during SS conditions, as the ecological indices were constant. Similar results were also observed by LaPara et al. (2002) treating pharmaceutical wastewater during SS.

Table 3

Estimation of richness (S), biodiversity (H), and evenness (J) indices of total community by RAPD at each SLR (mg SO4 − S/L d) in SS

SLR S VC H VC J VC 
290–436 5.5 ± 0.11 1.99 1.64 ± 0.02 1.3 0.96 ± 0.005 0.6 
654–981 6.32 ± 0.13 2.18 1.83 ± 0.02 1.28 0.99 ± 0.001 0.12 
SLR S VC H VC J VC 
290–436 5.5 ± 0.11 1.99 1.64 ± 0.02 1.3 0.96 ± 0.005 0.6 
654–981 6.32 ± 0.13 2.18 1.83 ± 0.02 1.28 0.99 ± 0.001 0.12 

VC – variation coefficient (%).

Table 4

Similarity coefficients (C) of total DNA sequence by RAPD at each SLR (mg SO4 − S/L d)

Samples SLR 290 290 290 290 290 290 436 654 981 981 817 
290a           
290          
290         
290        
290 0.98 0.98 0.98 0.98       
290 0.97 0.97 0.97 0.97 0.98      
436 0.88 0.88 0.88 0.88 0.89 0.91     
654 0.81 0.81 0.81 0.81 0.82 0.84 0.92    
981 0.74 0.74 0.74 0.74 0.76 0.77 0.85 0.93   
10 981 0.74 0.74 0.74 0.74 0.76 0.77 0.85 0.93  
11 817 0.73 0.73 0.73 0.73 0.74 0.76 0.84 0.91 0.98 0.98 
Samples SLR 290 290 290 290 290 290 436 654 981 981 817 
290a           
290          
290         
290        
290 0.98 0.98 0.98 0.98       
290 0.97 0.97 0.97 0.97 0.98      
436 0.88 0.88 0.88 0.88 0.89 0.91     
654 0.81 0.81 0.81 0.81 0.82 0.84 0.92    
981 0.74 0.74 0.74 0.74 0.76 0.77 0.85 0.93   
10 981 0.74 0.74 0.74 0.74 0.76 0.77 0.85 0.93  
11 817 0.73 0.73 0.73 0.73 0.74 0.76 0.84 0.91 0.98 0.98 

a1–6 are samples of sludge at 290 mg SO4 − S/L d within a period of 48 days at SS.

Figure 2

DGGE profiles of 16S rRNA amplicons obtained from UASB reactor; (a) archaeal and (b) bacterial DGGE profiles. Lane numbers indicate the SLR (mg SO4 − S/L d) applied. Identified DGGE bands are labeled to the right. M, marker (lane 1); A, archaeal community; B, bacterial community. Samples from UASB reactor were withdrawn at different days: 57 (lane 2), 65 (lane 3), 71 (lane 4), 79 (lane 5), 87 (lane 6), 101 (lane 7), 144 (lane 8), 165 (lane 9), 183 (lane 10), 193 (lane11), and 202 days (lane 12).

Figure 2

DGGE profiles of 16S rRNA amplicons obtained from UASB reactor; (a) archaeal and (b) bacterial DGGE profiles. Lane numbers indicate the SLR (mg SO4 − S/L d) applied. Identified DGGE bands are labeled to the right. M, marker (lane 1); A, archaeal community; B, bacterial community. Samples from UASB reactor were withdrawn at different days: 57 (lane 2), 65 (lane 3), 71 (lane 4), 79 (lane 5), 87 (lane 6), 101 (lane 7), 144 (lane 8), 165 (lane 9), 183 (lane 10), 193 (lane11), and 202 days (lane 12).

When the SLR was increased (lanes 7–12, Figure 2), at 436 mg SO4 − S/L d, the bacterial community showed an increase in two bands (B2 and B5), while the archaeal community did not show changes irrespective of the SLR. B2 band was assigned to a strain related to Clostridium propionicum (96%), which is known as a fermenter of lactate (Janssen 1991) (Table 5), and this increase might be related to the lactate loading rate increasing to 327 mg lactate-C/L d. The relative abundance of this strain was constant in the range of 436–981 mg SO4 − S/L d (327–736 mg lactate-C/L d). For the B5 band, no such relationship could be made; it might be possible that there was more than one strain present at the B5 in the gel preventing proper sequencing. The H index values with DGGE (1.8 ± 0.16 and 2.38 ± 0.005 for bacterial and archaeal communities, respectively) and RAPD (Table 3) remained without important changes, suggesting a diverse microbial community irrespective of the environmental changes.

Table 5

16S rRNA genes retrieved from archaeal and bacterial cells in a UASB reactor and similarity percentages to the closest related sequence in the NCBI database

Positiona Accession no. Closest related sequences in database (NCBI) Accession no. % Similarity 
Archaea 
 A1 EU287863 Uncultured Methanosaeta sp. AY692056 99 
Methanosaeta concilii AB212062 99 
 A2 NA    
 A3 NA    
 A4 EU287866 Uncultured Methanosarcinaceae archaeon EF420174 99 
Methanothrix soehngenii X51423 99 
 A5 EU287864 Uncultured Methanosarcinaceae archaeon EF420174 99 
Methanothrix soehngenii X51423 99 
 A6 EU287860 Methanobacterium beijingense DQ649302 98 
 A7 EU287862 Uncultured archaeon DQ088782 94 
Methanobacterium beijingense DQ649302 93 
 A8 Sb Uncultured Methanosaeta sp. AY692056 99 
 A9 EU287859 Uncultured Methanosaeta sp. AY692056 99 
 A10 EU287856 Uncultured Methanosaeta sp. AY692056 99 
 A11 EU287861 Uncultured Methanosaeta sp. AY692056 99 
Bacteria 
 B1 Sc Uncultured bacterium AY340824 100 
Halothiobacillus kellyi AF002671 93 
 B2d EU443148 Uncultured bacterium AY667258 99 
Clostridium propionicum X77841 96 
EU443149 Uncultured bacterium AY667258 98 
Clostridium aminobutyricum X76161 93 
 B3 EU287857 Uncultured bacterium AY340824 100 
Halothiobacillus kellyi AF002671 93 
 B4 EU287858 Uncultured bacterium AY340824 99 
Halothiobacillus kellyi AF002671 93 
 B5 NA    
 B6 EU287867 Desulfovibrio sp. L7 EF055877 98 
 B7d EU443146 Desulfobacca acetoxidans AF002671 95 
EU443147 Desulfatibacillum aliphaticivorans AY184360 93 
 B8d EU443144 Desulfobacca acetoxidans AF002671 94 
EU443145 Desulfuromonas alkaliphilus DQ309326 94 
Positiona Accession no. Closest related sequences in database (NCBI) Accession no. % Similarity 
Archaea 
 A1 EU287863 Uncultured Methanosaeta sp. AY692056 99 
Methanosaeta concilii AB212062 99 
 A2 NA    
 A3 NA    
 A4 EU287866 Uncultured Methanosarcinaceae archaeon EF420174 99 
Methanothrix soehngenii X51423 99 
 A5 EU287864 Uncultured Methanosarcinaceae archaeon EF420174 99 
Methanothrix soehngenii X51423 99 
 A6 EU287860 Methanobacterium beijingense DQ649302 98 
 A7 EU287862 Uncultured archaeon DQ088782 94 
Methanobacterium beijingense DQ649302 93 
 A8 Sb Uncultured Methanosaeta sp. AY692056 99 
 A9 EU287859 Uncultured Methanosaeta sp. AY692056 99 
 A10 EU287856 Uncultured Methanosaeta sp. AY692056 99 
 A11 EU287861 Uncultured Methanosaeta sp. AY692056 99 
Bacteria 
 B1 Sc Uncultured bacterium AY340824 100 
Halothiobacillus kellyi AF002671 93 
 B2d EU443148 Uncultured bacterium AY667258 99 
Clostridium propionicum X77841 96 
EU443149 Uncultured bacterium AY667258 98 
Clostridium aminobutyricum X76161 93 
 B3 EU287857 Uncultured bacterium AY340824 100 
Halothiobacillus kellyi AF002671 93 
 B4 EU287858 Uncultured bacterium AY340824 99 
Halothiobacillus kellyi AF002671 93 
 B5 NA    
 B6 EU287867 Desulfovibrio sp. L7 EF055877 98 
 B7d EU443146 Desulfobacca acetoxidans AF002671 95 
EU443147 Desulfatibacillum aliphaticivorans AY184360 93 
 B8d EU443144 Desulfobacca acetoxidans AF002671 94 
EU443145 Desulfuromonas alkaliphilus DQ309326 94 

NA, no accession number.

The sequences obtained were not of sufficient quality to be analyzed.

Escherichia coli positions 110–482 and 969–1367.

aPositions of the amplicons in the DGGE gel (Figure 2).

bSimilar to A9 (accession no. EU287859).

cSimilar to B3 (accession no. EU287857).

dPossible chimera, suggesting that two microorganisms could be present.

The J index values were close to 1 with DGGE (0.94 ± 0.02 and 0.99 ± 0.01, for bacterial and archaeal communities, respectively) and RAPD (Table 3), indicating that irrespective to the sulfate and lactate loading rate, there was no predominant species within the microbial population. However, in Figure 2, higher relative intensities of B6 and B8 bands in the bacterial community related to SRP, and A8 and A9 bands in the archaeal community were observed. The B6 and B8 bands were related to Desulfovibrio sp. (98%), an acetate producer from lactate, and Desulfobacca acetoxidans (94%), an acetate consumer (Oude Elferink et al. 1999). The A8 and A9 bands were related to Methanosaeta sp. (99%) (Table 5).

However, some strains clustering with γ-Proteobacteria (corresponding to B1, B3, and B4 bands, Figure 2) showed phylogenetic similarity with sulfur and sulfide oxidizing bacteria, such as Halothiobacillus kellyi (93%) (Table 5). This strain has been detected in sulfate reducing reactors (Kaksonen et al. 2004). The microbial analysis also showed hydrogenotrophic methanogenic microorganisms, which have been detected in anaerobic digesters (de Bok et al. 2006) and several acetoclastic methanogens. Despite having methanogenic microorganisms, results show no significant methane production at any loading rate (Figure 1), suggesting that the operating conditions defined by SRP (stoichiometric C/S ratio) had a greater impact on the overall process. Similar results were also observed by Raskin et al. (1996) where methanogens were present in the reactor but methane production was negligible when sulfate was added.

Therefore, when the SLR increased from 290 to 981 mg SO4 − S/L d, output TOC increased while the S, H, and J indices in total community, archaeal, and bacterial communities had minor changes. C values by RAPD and DGGE were higher than 73% (Table 4) and 81%, respectively, at each SLR increase. These changes observed in the ecological indices (S, H, J, and C) were negligible compared with the metabolic changes. The increase in TOC output seems to be strongly related to physiological and biochemical properties of the sludge, rather than the microbiological changes. The results obtained suggest that the accumulation of acetate and propionate at higher loading rates might be related to a possible lower specific consumption rate of acetate and propionate from lactate fermentation. The increases in sulfate and lactate loading rates showed a negligible effect on both bacterial and archaeal communities; however, there was one instance where a strain, related to C. propionicum, changed when the loading rate was increased. This change might be related to the lactate concentration; further increases in lactate had no measureable impact in the microbial community. Likewise, sulfide accumulation appeared to have no impact on microbial diversity as indicated by the H index. This observed trend could be attributed to numerous factors including, but not limited to: relatively low sulfide/biomass ratio, sulfide transport limiting biomass structure (i.e., biogranules), and pH operating range. According to these results, there was no intimate relationship between changes in metabolism profiles and the microbial population changes, despite the sulfate and lactate loading rate changes. Thus, the results of the present work suggest that more attention should be paid to operating conditions than microbial composition for better operation of wastewater bioreactors. This finding is important to consider in the startup and operation of wastewater treatment reactors like the UASB. Likewise, the present work is one of the few studies that show a steady-state SRP related to a stable microbial community.

CONCLUSIONS

In spite of increasing sulfate and lactate loading rate, minor changes in the ecological indices S, H, and J were observed, while at the same time, C values remained higher than 73 and 81%. On the contrary, the respiration pattern had significant changes as the efficiencies decreased, and acetate and propionate yield values increased. The changes observed in the ecological indices were negligible compared with the metabolic changes of the process. The analysis of results suggests that performance in the bioreactor, for the ranges of sulfate and lactate loading rates and operational parameters tested, was impacted more by operating conditions than by microbial changes.

ACKNOWLEDGEMENTS

This research was partially supported by Consejo Nacional de Ciencia y Tecnología (CONACyT) grant SEP-CONACyT, CB-2005-C01-49748-Z. Citlali García Saucedo was a PhD student financed by CONACyT.

REFERENCES

REFERENCES
Briones
A. M.
Daugherty
B. J.
Angenent
L. T.
Rausch
K.
Tumbleson
M.
Raskin
L.
2009
Characterization of microbial trophic structures of two anaerobic bioreactors processing sulfate-rich waste streams
.
Water Res.
43
,
4451
4460
.
Cervantes
F. J.
Duong-Dac
T.
Ivanova
A. E.
Roest
K.
Akkermans
A. D. L.
Lettinga
G.
Field
J. A.
2003
Selective enrichment of Geobacter sulfurreducens from anaerobic granular sludge with quinones as terminal electron acceptors
.
Biotechnol. Lett.
25
,
39
45
.
de Bok
F. A. M.
van Leerdam
R. C.
Lomans
B. P.
Smidt
H.
Lens
P. N. L.
Janssen
A. J. H.
Stams
A. J. M.
2006
Degradation of methanethiol by methylotrophic methanogenic archaea in a lab-scale upflow anaerobic sludge blanket reactor
.
Appl. Environ. Microbiol.
72
,
7540
7547
.
Fernández
A.
Huang
S.
Seston
S.
Xing
J.
Hickey
R.
Criddle
C.
Tiedje
J.
1999
How stable is stable? Function versus community composition
.
Appl. Environ. Microbiol.
65
,
3697
3704
.
Garcia-Saucedo
C.
Fernández
F. J.
Buitrón
G.
Cuervo-López
F. M.
Gómez
J.
2008
Effect of loading rate on TOC consumption efficiency in a sulfate reducing process: sulfide effect in batch culture
.
J. Chem. Technol. Biotechnol.
83
,
1648
1657
.
Hintze
J. L.
2001
Number Cruncher Statistical Systems 2001
.
Jerry L. Hintze
,
Kaysville, Utah
.
Mohammadi
M.
Man
H. C.
Hassan
M. A.
Yee
P. L.
2010
Treatment of wastewater from rubber industry in Malaysia
.
Afr. J. Biotechnol.
9
,
6233
6243
.
Muyzer
G.
Stams
A. J. M.
2008
The ecology and biotechnology of sulphate-reducing bacteria
.
Nat. Rev. Microbiol.
6
,
441
454
.
Nübel
U.
Engelen
B.
Felske
A.
Snaidr
J.
Wieshuber
A.
Amann
R. I.
Ludwig
W.
Backhaus
H.
1996
Sequence heterogeneities of genes encoding 16 rRNAs in Paenibacillus polymyxa detected by temperature gradient gel electrophoresis
.
J. Bacteriol.
178
,
5636
5643
.
Oude Elferink
S. J.
Akkermans-van Vliet
W. M.
Bogte
J. J.
Stams
A. J. M.
1999
Desulfobacca acetoxidans gen. nov. sp. nov., a novel acetate-degrading sulfate reducer isolated from sulfidogenic granular sludge
.
Int. J. Syst. Bacteriol.
49
,
345
350
.
Raskin
L.
Rittmann
B. E.
Stahl
D. A.
1996
Competition and coexistence of sulfate-reducing and methanogenic populations in anaerobic biofilms
.
Appl. Environ. Microbiol.
62
,
3847
3857
.
Roest
K.
Heilig
H. G. H. J.
Smidt
H.
de Vos
W. M.
Stams
A. J. M.
Akkermans
A. D. L.
2005
Community analysis of a full scale anaerobic bioreactor treating paper mill wastewater
.
Syst. Appl. Microbiol.
28
,
175
185
.
Sánchez-Andrea
I.
Sanz
J. L.
Bijmans
M. F. M.
Stams
A. J. M.
2014
Sulfate reduction at low pH to remediate acid mine drainage
.
J. Hazard. Mater.
269
,
98
109
.
Shannon
C. E.
Weaver
W.
1963
The Mathematical Theory of Communication
.
University Illinois Press
,
Urbana
.
This
P.
Cuisset
C.
Boursiquot
J. M.
1997
Development of stable RAPD markers for the identification of grapevine rootstocks and the analysis of genetic relationships
.
Am. J. Enol. Vitic.
48
,
492
501
.
Xie
B.
2004
Analysis of the effect of zinc ions on activated sludge microbes by random amplified polymorphism DNA (RAPD)
.
Int. J. Environ. Pollut.
21
,
533
546
.