Sulfate-reducing bacteria in anaerobic bioreactors

The treatment of industrial wastewaters containing high amounts of easily degradable organic compounds in anaerobic bioreactors is a well-established process. Similarly, wastewaters which in addition to organic compounds also contain sulfate can be treated in this way. For a long time, the occurrence of sulfate reduction was considered to be undesired. However, there are some recent developments in which sulfate reduction is optimized for the removal of sulfur compounds from waste streams. In the treatment of wastewaters which contain sulfate and organic compounds, sulfate reduction interferes with methanogenesis. Both mutualistic and competitive interactions between sulfate-reducing bacteria and methanogenic bacteria have been observed. Sulfate reducers can compete with methanogens for substrates such as hydrogen, formate and acetate, and with acetogens for substrates such as propionate and butyrate. On the other hand sulfate reducers can also assist propionate- and butyrate-degrading acetogens by acting as hydrogen scavenger, and in the absence of sulfate some sulfate reducers are even able to grow fermentatively or in syntrophic association with methanogens.Thus far it has been difficult to steer the wastewater treatment process in sulfate-fed bioreactors in the desired direction (i.e. in the direction of sulfidogenesis or of methanogenesis). Therefore, the aim of the research presented in this thesis was to investigate the effect of sulfate on the presence and activity of sulfate reducers and their acetogenic and methanogenic counterparts in sulfate-fed anaerobic bioreactors, in order to get a better grip on the treatment process.Acetate is a key intermediate in the anaerobic degradation of organic material. Thus far, information about the competition between sulfate reducers and methanogens for acetate in anaerobic bioreactors has been scarce, and contradictory. Furthermore, information on the type of acetate-degrading sulfate reducers in reactor sludge was not available, which made predictions over the outcome of competition between sulfate reducers and methanogens difficult.Therefore, the research first focused on the characterization of acetate-degrading sulfate reducers which are dominantly present in sulfidogenic granular sludge. This led to the isolation and characterization of two thus far unknown acetate-degrading sulfate reducers, now named Desulforhabdus amnigenus strain ASRB1 and Desulfobacca acetoxidans strain ASRB2. Desulforhabdus amnigenus was isolated from granular sludge of a pilot-scale upflow anaerobic sludge bed (UASB) reactor treating papermill wastewater, while Desulfobacca acetoxidans was isolated from a lab-scale UASB reactor fed with acetate and sulfate. In the pilot-scale reactor the COD/sulfate ratio (g/g) was approx. 1.1, and 75% of the degraded COD was degraded via sulfate reduction. The lab-scale reactor was operating at an excess of sulfate (COD/sulfate ratio (g/g)= 0.6), and all acetate was degraded via sulfate reduction. Both acetate-degrading sulfate reducers were isolated from the highest serial dilutions of sludge which showed growth on acetate and sulfate. This indicated that these bacteria are the dominant acetate-degrading sulfate reducers in the two respective sludge-types. Based on 16S rRNA analyses both sulfate reducers phylogenetically cluster with the delta subdivision of the Proteobacteria , but they are not closely related to each other. There are large differences between the physiological characteristics of D. amnigenus and D. acetoxidans . D. amnigenus is a substrate generalist, which besides acetate, can use a wide variety of other substrates, such as propionate, butyrate, lactate, H 2 +CO 2 , and alcohols, while D. acetoxidans is a substrate specialist, which only utilizes acetate as a carbon and energy source.A special characteristic of D. amnigenus is its ability to isomerize butyrate to isobutyrate (Chapter 5), a property which was thus far not described for sulfate reducers. Nuclear Magnetic Resonance (NMR) studies with 13C-labelled butyrate showed that isobutyrate was formed by a migration of the carboxyl group, conform the butyrate isomerization reaction reported for methanogenic consortia. Further investigations showed that the capacity of D. amnigenus to isomerize butyrate was not unique among sulfate reducers. Several other butyrate-degrading sulfate reducers, including Desulfobacterium vacuolatum , and Desulfoarculus baarsii , were to a lesser extent also capable of butyrate isomerization.Important factors for the outcome of the competition for acetate between methanogens and sulfate reducers are the acetate degradation properties of the bacteria involved.To examine the competition for acetate between D. amnigenus , D. acetoxidans , and acetate-degrading methanogens, the kinetics of acetate degradation was studies for both sulfate reducers and the kinetic properties were compared with those of acetate-degrading methanogens (Chapter 4). The Michaelis-Menten parameters for D. amnigenus (K m = 0.2-1 mM, Vmax = 21-35 mmol min -1g protein -1), and D. acetoxidans (K m = 0.2-1 mM, Vmax = 29-57 mmol min -1g protein -1) were in the same range or slightly better than those of most Methanosaeta sp. (K m = 0.4-1.2 mM, Vmax = 32-170 mmol min -1g protein -1). The same applied for the Monod kinetic parameter m max and the acetate-degradation threshold, which were 0.14-0.20 day -1and <15 mM for D. amnigenus , 0.31-0.41 day -1and <15 mM for D. acetoxidans , and 0.08-0.69 day -1and 7-69 mM for Methanosaeta , respectively. Compared with Methanosarcina sp. (K m = 3.0 mM, threshold= 0.19-1.2 mM, m max =0.46-0.69 day -1), D. amnigenus and D. acetoxidans had much better K m and threshold values, but only D. acetoxidans had a comparable m max value. Based on the acetate kinetic parameters of D. amnigenus and D. acetoxidans compared to those of Methanosaeta sp. and Methanosarcina sp., it can be predicted that D. amnigenus can slowly outcompete Methanosaeta sp., while D. acetoxidans can even outcompete Methanosarcina sp.Generally, Methanosaeta sp. are the dominant acetate-degrading methanogens in methanogenic bioreactors under acetate-limiting conditions. Such acetate-limiting conditions also prevailed in the papermill UASB-reactor. This could explain why D. amnigenus became the dominant acetate-degrading sulfate reducer in the papermill UASB-reactor, and also why a long adaptation time was needed before sulfate-reduction became the dominant process in this reactor. Unfortunately, the fact that D. amnigenus is a substrate generalist makes it difficult to draw clear conclusion about the exact role of D. amnigenus in the sludge, because batch studies with mixed substrates indicated that acetate was one of the least preferred substrates of D. amnigenus (Chapter 4). Although it is known that carbon substrates which provoke diauxy under batch conditions are used simultaneously under carbon limited conditions, it cannot be excluded that the presence of substrates, such as lactate, propionate, or ethanol, could have a negative effect on the acetate degradation rate of D. amnigenus in the sludge.From the characterization studies presented above it became clear that reactors treating different wastewaters also contained different sulfate-reducing populations. Two factors which play an important in the microbial composition of sludge are the composition of the organic components in the wastewater and the COD/sulfate ratio of the wastewater. In order to investigate the effect of these factors on the presence and activity of sulfate reducers and their acetogenic and methanogenic counterparts in sludge, good sludge characterization methods are indispensable. Fortunately, to date several sludge characterization methods are available, including conventional methods such as Most Probable Number (MPN) methods, but also more modern methods such as polar-lipid fatty acid (PLFA) analyses, specific PCR amplifications, and 16S rRNA dot blot hybridizations. In Chapter 6, 7 and 8 the sludge from various methanogenic and sulfidogenic reactors was characterized using the different sludge characterization methods. This research did not only improve our knowledge on the effect of wastewater composition and COD/sulfate ratio on the microbial sludge composition, but it also showed the advantages and disadvantages of the different characterization methods.Very useful methods for the species, genus or group-specific detection of microorganisms in sludge are the PCR amplification method, and the dot blot hybridization method. This is shown in Chapter 6 describing the development of a 16S rRNA oligonucleotide probe, probe ASRB1, for the species-specific detection of D. amnigenus in sludge. If this probe was used in dot blot hybridization studies, D. amnigenus could still be detected if approx. 0.1 to 0.001 ‰ of the total bacterial sludge population was D. amnigenus . If the probe was used as a PCR primer the sensitivity was even 10 times higher. It is not possible to use the dot blot hybridization method for the exact quantification of the number of bacterial cells in the sludge, because it is based on the rRNA content in the cell. For D. amnigenus as shown that this rRNA content was affected by the growth rate and the growth phase, and that it ranged from 19 fg cell -1in slow-growing cultures to 90 fg cell -1in fast-growing cultures. This also indicates that the detection threshold of the dot blot hybridization method for fast-growing cells is approximately 5-fold lower than for slow-growing cells.Many sludge-types were analyzed for the presence of D. amnigenus . Unfortunately this bacterium could only be detected in the sludge from the pilot-scale reactor from which is was originally isolated, and in the sludge which was used as seed-sludge for this pilot-scale reactor. This seems to indicate that D. amnigenus plays no important role in other sulfidogenic bioreactors. This could be due to the absence of D. amnigenus in the seed sludge of the sulfidogenic bioreactors.In Chapter 7 granular sludge from a full-scale UASB reactor was studied, by using the 16S rRNA dot blot hybridization method in combination with MPN estimates. The wastewater which was treated in the UASB-reactor contained mainly starch, acetate, propionate, butyrate and formate, and had a COD/sulfate ratio of 9.5. Under these conditions acetate seemed to be mainly degraded by Methanosaeta -like bacteria, while propionate was the preferred substrate for sulfate reduction.However, the Desulfobulbus -like propionate-degrading sulfate reducers in the sludge, competed with Syntrophobacter -like bacteria, for the available propionate. Hydrogen and formate were probably mainly degraded via methanogenesis by members of the order of Methanobacteriales . Dot blot hybridization studies of the MPN enrichments showed that the hydrogen, formate and butyrate-degrading sulfate reducers in the sludge could not be characterized with the 16S rRNA probes available to date. The same was true for syntrophic butyrate degraders. This clearly demonstrates that, although the dot blot hybridization method is very useful for sludge characterization studies, it does not (yet) give a complete picture of the total sludge composition, and it should be used in combination with other methods to avoid missing important groups of microorganisms in the sludge.A method which also gives insight in the sludge composition is the PLFA method. In Chapter 8 this method was combined with the dot blot hybridization method to study the sulfate-reducing and acetogenic population of several methanogenic and sulfidogenic sludges. The results show that PLFA analyses of the sludge were useful to obtain a quick general impression of the total bacterial sludge composition, but were less suitable for an accurate characterization and quantification of the sulfate-reducing population in the sludge. This was due to the lack of selective biomarkers for these bacteria. The combined results of the PLFA analysis and 16S rRNA dot blot hybridizations showed that presence of sulfate reducers in the sludge was not dependent on the presence of sulfate in the wastewater. This may be explained by the syntrophic and/or fermentative capacities of some sulfate reducers in the absence of sulfate. Desulfobulbus sp. seemed to be important in reactors with carbohydrates and/or volatile fatty acids containing wastewater. In the presence of sulfate these bacteria could play a role in propionate degradation, while in the absence of sulfate they might play a role in lactate and ethanol fermentation. Desulfobacter sp. did not seem to be important for acetate degradation in any of the sulfate-fed reactors. Apparently, most Desulfobacter sp. are typical marine organisms. In the acetate and sulfate-fed reactor, Desulfotomaculum acetoxidans -like bacteria seemed to play a role in acetate degradation.The results which were presented in this thesis improved our knowledge of the effect sulfate on the microbial sludge population in anaerobic reactors. As already mentioned it has been, and still is, difficult to steer the wastewater treatment process in sulfate-fed bioreactors in the direction of sulfate reduction or methanogenesis. However, from this thesis a few conclusions can be drawn which are useful for optimization of the reactor process, and for future sludge research.There appears to be a thight competition between acetate-degrading sulfate reducers and acetate-degrading methanogens (mainly Methanosaeta sp.) in anaerobic reactor sludge. It will therefore take a long time before acetate-degrading sulfate reducers have outgrown acetate-degrading methanogens, even if there is an excess of sulfate.In sulfate-limited bioreactors sulfate reducers compete with each other for the available sulfate. Under sulfate limitation acetate seems to be one of the least, and propionate one of the most favoured substrates for sulfate reduction.Based on 16S rRNA analyses Desulfobulbus -like bacteria can be found in high numbers in many reactors, even in methanogenic reactors.There is a large variation in the microbial composition of granular sludge. The choice of a seed sludge for a new reactor will probably not only affect the initial, but also the final purification efficiency of the reactor.All sludge characterization methods have their advantages and disadvantages. The best picture of the microbial sludge composition can be obtained by combining as many of them as possible.