Molecular ecology of anaerobic granular sludge grown at different conditions

E. Daz, R. Amils and J.L. Sanz

Centro de Biologa Molecular, Universidad Autnoma de Madrid, 28049 Espaa (E-mail: joseluis.sanz@uam.es; eediaz@cbm.uam.es; ramils@cbm.uam.es) Abstract Qualitative and quantitative diversity of microorganisms present in anaerobic granular sludges fed with different substrates, as well as the structure of these granules have been studied using uorescent 16S rRNA-targeted in situ hybridization and electron microscopy. The granules showed a multi-layered structure, in which both densely packed and loose micro-colonies, channels and holes could be observed. Only bacteria were found in the outer shell of the granules, while both archaea and bacteria were detected in the inner core. Although high cell density was found in the granules (more than 1011 cells/gram, determined by DAPI-stain) only a low percentage of cells was able to hybridize with the rRNA-targeted probes. Signicant quantitative and qualitative differences were observed in the composition of granules fed with different substrates (formate, acetate at high and low concentrations, propionate, sucrose, starch and peptone). Bacterial cells were mostly gram-positives. Active proteobacteria were scarce in the granules exposed to VFA. Syntrophobacteria became dominant in the propionate-grown biomass. Concerning methanogenic archaea, Methanosaeta was the predominant species using complex substrates or low acetate concentration fed granules, while Methanosarcina and members of Methanobacteriales were predominant in the granules grown at high concentration of acetate or formate, respectively. Other Methanomicrobiales and Methanococcales, have been detected in the anaerobic granular sludge in the conditions used in this work. Keywords Anaerobic granular sludge; FISH; granulation; molecular ecology; molecular probes; UASB

Water Science and Technology Vol 48 No 6 pp 5764 IWA Publishing 2003

Introduction

Around 75% of the approximately 2,000 anaerobic treatment systems presently in operation world wide correspond to Upflow Anaerobic Sludge Bed (UASB) reactors or to new configurations based on the same principle (EGSB, IC). In all of them, microorganisms form compact aggregates up to 24 mm in diameter, granular sludge, with high sedimentation velocity and high methanogenic activity. Due to the huge structural complexity of granular sludge and the many trophic interactions among the microbial populations required to transform complex organic matter into biogas (CO2 and CH4), the microbial ecology (taxonomy, colonization, topological distribution of microbes, etc.) of this microecosystem is still not well understood. Molecular ecology techniques, such as 16S rRNA gene cloning and sequencing (Amann et al., 1995), Fluorescence In Situ Hybridization (FISH) (Amann, 1995; Amann et al., 1995), and Denaturing Gradient Gel Electrophoresis (DGGE) (Muyzer et al., 1993; Muyzer and Smalla, 1998), are nowadays the most powerful tools available to assess the diversity, abundance and distribution of microorganisms in natural and engineered ecosystems, superceding the restrictions and bias of conventional microbiology techniques (isolation, plate-counting, etc.) (Amann et al., 1995; Pace, 1996). The use of molecular ecology techniques to the study of anaerobic granular sludge is quite recent and allows the main prokaryotic populations developed in the granule and its interactions to be identified (Santegoeds et al., 1999; Sekiguchi et al., 1999; Chan et al., 2001; Liu et al., 2002). The importance of this approach to the study of anaerobic water treatment processes is reflected in the special issue recently published by Water Research (2002). In this paper we present the microbial composition of anaerobic granular sludges studied using FISH, TEM and SEM.

57

Methods
Biomass and reactor experimental set-up

All the experiments were carried out using granular sludge from a brewery full scale UASB reactor (Mahou, Guadalajara, Spain) as inoculum and control. Because the different bacterial populations developed in the granules are dependent on the type of pollutants present in the wastewater, the following substrates have been evaluated: formate, acetate at both high and low concentrations, propionate, sucrose, starch and peptone. Assays were carried out in 1 litre batch reactors. The reactors were fed by spike doses to keep the desired substrate concentration constant. Each reactor was monitored for 3 months.
Fluorescent in situ Hybridization (FISH)

58

E. Daz et al.

The protocols described by Amann (Amann et al., 1990, 1995) were used for the hybridization experiments. The samples were fixed, just after their collection, in 4% paraformaldehyde, subsequently washed in phosphate buffer saline (PBS) and preserved in PBS-ethanol at 20C until use. For the hybridization procedure, the samples were fixed on a multi-dish slide at 46C and dried in ethanol (50, 80, 100%). Formamide concentrations used for hybridization (2 h, 46C) and NaCl concentrations used for washing (15 min, 48C) are listed in Table 1. The probes, marked at the 5-end with fluorescein or Cy3, were purchased from Genotek (Barcelona, Spain).
Microscopy and cell counting

The total cells present in the samples were determined by counting 4,6-diamin phenylindol (DAPI) stained cells. Representative samples of the granules (0.03 g of wet weight granules in 1 ml of PBS: ethanol 1:1) were sonicated 3060 s at middle power and 1 cycle/s. To quantify the cells, the cell suspensions were diluted 1/50 for DAPI-counting and 1/5 for hybridized cell-counting. Both DAPI-stain and hybridize cells were examined with an epifluorescent microscope Zeiss Axioskop. Ten fields (corresponding to 5001,000 DAPI-

Table 1 DNA probes used in this work, with indications of the target site, specicity and hybridization/washing conditions
Probe Sequence (53) Formamide (%)/NaCl (mmol l1) Specicity

EUB338 ARC915 NON338 ALF968 BET42a GAM42a SRB385 DSS658 DSV698 BAC1080 SYN835 LGC354A HGC69A MEB859 MC1109 MG1200 MSSH859 MS1414 MX825

GCTGCCTCCCGTAGGAGT GTGCTCCCCCGCCAATTCCT ACTCCTACGGGAGGCAGC GGTAAGGTTCTGCGCGTT GCCTTCCCACTTCGTTT GCCTTCCCACATCGTTT CGGCGTCGCTGCGTCAGG TCCACTTCCCTCTCCCAT GTTCCTCCAGATATCTACGG GCACTTAAGCCGACACCT GCGGGTACTCATTCCTG TGGAAGATTCCCTACTGC TATAGTTACCACCGCGT GGACTTAACAGCTTCCCT GCAACATAGGGCACGGGTCT CGGATAATTCGGGGCATGCTG TCGCTTCACGGCTTCCCT CTCACCCATACCTCACTCGGG TCGCACCGTGGCCGACACCTAGC

35/80 20/225 35/80 20/225 35/80 35/80 35/80 60/15.6 35/80 20/225 35/80 20/225 25/159 25/159 35/80 5/636 35/80 35/80 20/225

Bacteria Archaea None (negative control) Alpha-proteobacteria Beta-proteobacteria Gamma-proteobacteria Sulfate-reducing bacteria Desulfosarcina, Desulfococcus Desulfovibrio Bacteroides Syntrophobacter sp. Low GC Gram-positive High GC Gram-positive Methanobacteriales (except for Methanothermaceae) Methanococcales Methanomicrobiales (except for Methanosarcinaceae) Methanosarcinaceae Methanosarcinaceae (except for Methanosaeta) Methanosaeta

stained cells) were counted to determine the average number of cells per sample. For granular structural studies, the specimens were microscopically examined with a Radiance2000 confocal laser scanning microscope (CLSM) equipped with He-Ne lasers and a reversed microscope Zeiss Axiovert S100. Transmission electron microscopy was done according to the protocol described by Grotenhuis et al. (1991) using a JEM1010 microscope. Scanning electron microscopy was performed as described by Alphenaar et al. (1994) with a Philips XL30 microscope.
Results and discussion

E. Daz et al.

Two different approaches were used to study the microbial diversity present in the granule: i) hybridization with 16S rRNA-targeted probes (FISH) and both transmission (TEM) as well as scanning (SEM) electron microscopy of thin sections of the whole granule, in order to reveal the spatial distribution of THE microorganisms, and ii) hybridization of crushed granules to quantify the different microbial groups (FISH).
Structure and microbial distribution of a granular sludge

TEM of granular sludge showed different structural features: densely packed microcolonies, loose colonies, channels and holes. Clearly defined micro-colonies of a single microbial species appear in the granules, together with overlapping heterogeneous colonies (Figure 1: A, B), probably as a result of distinct metabolic specificities and interactions (hydrogen or other metabolic products) of different microbial populations. The holes and channels might be related to the growth of the granule and the required diffusion of substrates and products. The presence of low electron dense cells (ghost microorganisms) is noteworthy (Figure 1: C). These forms could be resting or dead cells and their presence may explain the low percentage of active microorganisms detected by FISH. FISH, TEM and SEM microscopy of the granular sludge revealed a multi-layer structure (Figure 2). A primary spatial distribution of archaea and bacteria could be observed by FISH. Bacteria always appear in the outer shell of the granule, while archaea are always located in its interior (Figure 2b), as was also described by Sekiguchi et al. (1999). This feature reflects the distribution of well characterized facultative aerobic bacteria in the outer layer which consume oxygen (higher metabolic activity) and prevent its diffusion into the granule, thus protecting the strict anaerobes (especially methanogenic archaea and sulfate-reducing bacteria) from oxygen toxicity. Inside the granule, bacteria and archaea are arranged in micro-colonies (Figures 1 and 2).

Figure 1 TEM of granular sludge. A: colony of a single microorganism. B: heterogeneous colonies made of several microorganisms, with a channel between them. C: low electron density cells. Amplications: A and B: 1,500; C: 4,000

59

60

E. Daz et al. Figure 2 Multi-layered structure of the granular sludge shown and SEM (left) and confocal microscopy and FISH (right) using the universal probes for Bacteria (Eub338-Fluorescein, green) and Archaea (Arch915Cy3, red)
Quantitative microbial composition of granular sludge under different feeding conditions

The number and types of relationships among the total microorganisms, bacteria and archaea are presented in Table 2 as a function of the granule type (pollutant, size and/or color of the granule). There is a great variation of the active populations of microorganisms determined by the history of sludge. First of all, a remarkably high concentration of microorganisms is found in the granular sludge. The granules did not grow inside the reactors thus no significant differences in the total cells number between different substrates were detected. However, it is noteworthy that less than 10% of the total number of microorganisms present in the granule were active in most of the samples (cells with high content in rRNA that are able to hybridize versus those stained with DAPI, regardless of whether they are active or not) (Table 2). This is a somewhat surprising result. A higher percentage of active cells in granular sludge from an EGSB treating brewery wastewaters has been described (Gonzlez-Gil et al., 2001). Probably, the low percentage of active cells in our granules is due to the low methanogenic activity (0.5 g COD/l*d) of the granular sludge sampled from a brewery full-scale UASB reactor used in this work. These results suggest that most of the microorganisms in the sludge correspond to resting forms (Figure 1, C). In fact, when the control crushed granular sludge was used as inoculum to develop biofilms on glassslides, most of the cells grown on the biofilms gave positive hybridization signals with the
Table 2 Quantication of the total number (DAPI-stained) and active (hybridized) cells in granular sludges
Substrate (reactor) Total cells (DAPI) cells/gram of sludge Conc. (g COD/l) Bacteria (Eub338) Archaea (Arch915) Active (%) Bacteria (%) Archaea (%)

cells/gram cells/gram of sludge of sludge

Formate (R1) Low Acetate (R2) High Acetate (R3) Propionate (R4) Sucrose (R5) Starch (R6) Peptone (R7) Control Grey Black Brown

2.61011 2.31011 2.51011 2.21011 2.31011 2.51011 2.51011 2.71011 3.51011 2.31011 2.51011

4.0 0.2 5.0 4.0 9.0 11.0 12.0

0.00 1.0108 7.4108 1.1109 9.3109 1.21010 5.7109 9.7109 1.31010 8.7109 5.3109

7.6109 5.6109 8.0109 1.21010 1.31010 1.01010 4.8109 1.41010 1.41010 8.8109 1.91010

2.9 2.5 3.5 6.2 9.7 8.8 4.2 8.8 7.7 7.5 9.6

0.0 1.8 8.5 8.2 41.4 54.3 54.2 40.8 49.3 49.9 21.9

100.0 98.2 91.5 91.8 58.6 45.7 45.8 59.2 50.7 50.1 78.1

Grey, black and brown are different types of granules (differences in color, morphology, size, sedimentation velocity and abundance) found in the control sludge. Control granules were xed in the brewery plant

fluorescent probes, indicating that the majority of the cells were metabolically active (data not shown). The metabolic pathway of the substrate appears to influence the number of active cells present in the granule: lower numbers for the volatile fatty acids than for more complex substrates (starch, sucrose, sludge control from brewery, etc.). The low number of labeled cells obtained with olive oil must be related to its poor availability (insolubility). Moreover, the relationship between bacteria and archaea depend on the type of substrate used. In the case of volatile fatty acids, located at the end of the anaerobic digestion pathway, only active archaea were detected. Using not only domain but group specific probes for Bacteria (Figure 3A) and Archaea (Figure 3B), significant quantitative differences were obtained in the populations developed in the reactors fed with different substrates. As can be expected, the microbial diversity increases with the complexity of the substrates. For example, while acetate can be directly assimilated by the aceticlastic methanogens, diverse microorganisms are required for the complete degradation of complex organic matter to CH4 and CO2. In addition, different problems have to be considered when interpreting the results. In some cases the fluorescent signal of the probe does not allow cells that give a positive hybridization signal to be clearly distinguished from those that have not, making the results unclear. For instance, the specific probe for

E. Daz et al.

Figure 3 Microbial composition of anaerobic granular sludge exposed to different substrates detected with different group specic oligonucleotide probes for Bacteria (A) and Archaea (B). The black bars show the differences among the % of cells hybridized using the universal probe for Bacteria (A) and Archaea (B) and the different group-specic probes tested (negative values were obtained when the addition of hybridized cells using different probes was higher than the number of hybridized cells labeled with the universal probe). The difference between the cells hybridized with MSSH859 (Methanosarcinaceae) and MX825 (Methanosaeta) corresponds to Methanosarcina cells

61

62

gram-positives did not uniformly label whole cells, or the brightness of probe MA1414 was very intense when used to hybridize Methanosarcina but also other rod-shapes cells gave less intense positive hybridization, and as a consequence the cell count can be overestimated. In other cases, mainly for VFA and bacteria, the percentage of active cells was very low (i.e. 1.1% of active bacteria in the granules grown at low acetate concentration), thus special care must be taken in the interpretation of the results. The control granule is the only system that presents all the bacterial groups tested in this work. Diversity decreases when the granules are exposed to a variety of substrates ranging from macromolecules (starch, peptone) to VFA, in which typical proteobacteria (-, -, -proteobacteria) are scarce (with the only exception in the reactor being fed at high acetate concentration). The formategrown granules are an extreme case, since active bacteria are absent. Interestingly enough, syntrophobacteria (propionate-oxidizing bacteria) become dominant in the propionategrown biomass. On the other hand, gram-positive bacteria accounted for a high percentage of the total hybridized bacteria. A similar result was found by Liu (Liu et al., 2002). 50% of the bacterial clones obtained from granular sludge were affiliated to the low G+C grampositive group of bacteria. In the case of methanogenic archaea, we did not nd positive hybridization signals with either MC1109 or with MG1200 probes. Araujo (Araujo et al., 2000) has also reported negative hybridization results using both probes in biolms developed using granular sludge from a domestic sewage treatment plant as inoculum, and Liu (Liu et al., 2002) found only 1% of Methanococcales. MC1109 is a specic probe for Methanococcales, which have, until now, only been isolated from marine environments. MG1200 is specic for methanogens belonging to the families Methanomicrobiaceae, Methanocorpusculaceae and Methanoplanaceae (included in the order Methanomicrobiales) limited to the use of H2 and formate as substrates. Most of the members of this group require a relatively high sodium ion concentration. Thus, the absence of this type of methanobacteria in the granular sludge would not be surprising. On the other hand, due to the apparent low specicity of the MS1414 probe, probes MSSH859 and MX825 were used to test all the Methanosarcinaceae species, and to discriminate between Methanosarcina and Methanosaeta. The granules fed with complex substrates presented higher diversity of methanogenic archaea, similar to what was observed with Bacteria. In this case, Methanosaeta was the predominant species. It is important to underscore that the reactor fed with formate presented a total prevalence of microorganisms hybridizing with the Methanobacteriales probe (MEB859), whose morphology resembled that of Methanobrevibacter spp. or Methanobacterium spp., with almost total absence of active bacteria. Both observations are consistent considering that formic acid is directly metabolized by methanogenic archaea. Other meaningful examples are the characteristics of the biomass obtained in the reactors fed with acetate, in which 90100% of the active microorganisms were archaea. In this case all the cells stained with the universal archaeal probe also hybridized with the Methanosarcinaceae probe (MSSH859). The sludge exposed to high concentrations of acetate presented a high proportion of Methanosarcina (probe MS1414, Figures 3B and 4A), while members of this genera were scarcely detectable in the reactor fed with low concentrations of acetate. However, in this case a high level of hybridization was obtained with the specific probe for Methanosaeta (MX825, Figures 3B and 5B). These results agree with the kinetic parameters for both species of aceticlastic methanobacteria. Methanosarcina has a higher growth rate for acetate that Methanosaeta (max 0.21 d1 versus 0.11 d1), although its affinity for acetate is lower (Ks 4.02 mmol l1 versus 0.44 mmol l1). Thus, the predominance of Methanosaeta at low concentrations of acetate can be linked to its high affinity for this substrate. Methanosarcina became dominant if the acetate concentration increased to several grams per litre.

E. Daz et al.

E. Daz et al.

Figure 4 Microbial diversity of an anaerobic granular sludge fed exclusively with acetate. 4A: high concentration of acetate. Methanosarcina (FISH: MS1414-Cy3). 4B: low concentration of acetate. Prevalence of Methanosaeta-like microorganisms (SEM)

Conclusions

Molecular techniques (FISH) and electron microscopy (TEM and SEM) are very useful tools to study complex micro-cosmos such as anaerobic granular sludge. Using these techniques the following conclusions can be drawn. 1. Granules have a multi-layer structure. In the outer layers only bacteria are present. Densely packed micro-colonies, loose colonies (both pure and heterogeneous), channels and holes can be observed in the inner layers. 2. FISH using domain and group-specific probes, showed significant differences in the microbial populations developed using different substrates, although only a low percentage of cells were active. Microbial diversity (both bacteria and methanogenic archaea) increased with the complexity of the substrates. 3. Gram-positive bacteria accounted for a high percentage of the total hybridized bacteria. -, -, and -proteobacteria were scarce in VFA-cultivated granules. Active bacteria were absent in formate-grown granules. Syntrophobacteria became dominant in the propionate-grown biomass. 4. Methanosaeta was the predominant archaeal species using granules exposed to complex substrates or low acetate concentration. Methanosarcina and members of Methanobacteriales were the majority in the granules cultivated with a high concentration of acetate and formate respectively. Methanococcales and Methanomicrobiales, with the exception of members of the family Methanomicrobiaceae, were never observed in the different incubation conditions used in this work.
Acknowledgements

This work was partially supported by a grant from the Ministerio espaol de Ciencia y Tecnologa to Jos L. Sanz, project number REN2001-2980-C02-02/HID. E. Daz is a fellow of the Agencia Espaola de Cooperacin Internacional (Ministerio de Asuntos Exteriores).
References
Alphenaar, P.A., Groeneveld, N. and van Aelst, A.C. (1994). Scanning electron microscopical method for internal structure analysis of granular sludge. Micron, 25(2), 129133. Amann, R.I., Binder, B.J., Olson, R.J., Chisholm, S.W., Devereux, R. and Stahl, D.A. (1990). Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl. Environ. Microbiol., 56(6), 19191925. Amann, R.I. (1995). In situ identification of microorganisms by whole cell hybridization with rRNAtargeted nucleic acid probes. In: Molecular Microbial Ecology Manual, D.L. Akkermans, J.D. van Elsas and F.J. de Bruijn (eds.). Kluwer Academic Publishers, Dordrecht, 3.3.6, 115. Amann, R.I., Ludwig, W. and Schleifer, K.H. (1995). Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol. Rev., 59, 143169.

63

Araujo, J.C., Brucha, G., Campos, J.R. and Vazoller, R.F. (2000). Monitoring the development of anaerobic biofilms using fluorescent in situ hybridization and confocal laser scanning microscopy. Wat. Sci. Tech., 41(12), 6977. Chan, O.-C., Liu, W-T. and Fang, H.H.P. (2001). Study of microbial community of brewery-treating granular sludge by denaturing gradient gel electroforesis of 16S rRNA gene. Wat. Sci. Tech., 43(1), 7782. Gonzlez-Gil, G., Lens, P.N.L., van Aelst, A., Versprille, A.I. and Lettinga, G. (2001). Cluster structure of anaerobic aggregates of an expanded granular sludge bed reactor. Appl. Microbiol. Biotechnol., 67, 36833692. Grotenhuis, J.T.C., Smit, M., van Lammeren, A.A.M., Stams, A.J.M. and Zehnder, A.J.B. (1991). Localization and quantification of extracellular polymers in methanogenic granular sludge. Appl. Microbiol. Biotechnol., 36, 115119. Liu, W.-T., Chan, O.-C. and Fang, H.H.P. (2002). Characterization of microbial community in granular sludge treating brewery wastewater. Wat. Res., 36, 17671775. Muyzer, G. and Smalla, K. (1998). Application of denaturing gradient gel electrophoresis (DGGE) and temperature gradient gel electrophoresis (TGGE) in microbial ecology. Mini Review. Antonie van Leeuwnhoek, 73, 127141. Muyzer, G., de Waal, E.C. and Uitterlinden, A.G. (1993). Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. App. Environ. Microbiol., 59(3), 695700. Pace, N.R. (1996). New perspective on the natural microbial world: molecular microbial ecology. ASM News, 62(9), 463470. Raskin, L., Stromley, J.M., Rittmann, B.E. and Stahl, D.A. (1994). Group specific 16S rRNA hybridation probes to describe natural communities of methanogens. App. Environ. Microbiol., 60(4), 12321240. Santegoeds, C.M., Damgaard, L.R., Hesselink, G., Zopfi, J., Lens, P., Muyzer, G. and de Beer, D. (1999). Distribution of sulfate-reducing and methanogenic bacteria in anaerobic aggregates determined by microsensor and molecular analyses. Appl. Environ. Microbiol., 65(10), 46184629. Sekiguchi, Y., Kamagata, Y., Nakamura, K., Ohashi, A. and Harada, H. (1999). Fluorescence in situ hybridization using 16S rRNA-targeted oligonucleotides reveals localization of methanogens and selected uncultured bacteria in mesophilic and thermophilic sludge granules. Appl. Environ. Microbiol., 65(3), 12801288. Water Research, 36(2), (2002). Special Issue: Modern Scientific Tools in Bioprocessing.

64

E. Daz et al.