APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Mar. 2000, p. 1175–1182 Vol. 66, No.

3
0099-2240/00/$04.00⫹0
Copyright © 2000, American Society for Microbiology. All Rights Reserved.

Identification of Polyphosphate-Accumulating Organisms and Design of
16S rRNA-Directed Probes for Their Detection and Quantitation
GREGORY R. CROCETTI,1 PHILIP HUGENHOLTZ,1 PHILIP L. BOND,1 ANDREW SCHULER,2
JÜRG KELLER,3 DAVID JENKINS,2 AND LINDA L. BLACKALL1*
Department of Microbiology and Parasitology1 and Department of Chemical Engineering,3 Advanced Wastewater
Management Centre, The University of Queensland, St. Lucia, 4072 Queensland, Australia, and Department
of Civil and Environmental Engineering, University of California, Berkeley, California 94720-17102
Received 9 August 1999/Accepted 23 November 1999

Laboratory-scale sequencing batch reactors (SBRs) as models for activated sludge processes were used to
study enhanced biological phosphorus removal (EBPR) from wastewater. Enrichment for polyphosphate-
accumulating organisms (PAOs) was achieved essentially by increasing the phosphorus concentration in the
influent to the SBRs. Fluorescence in situ hybridization (FISH) using domain-, division-, and subdivision-level
probes was used to assess the proportions of microorganisms in the sludges. The A sludge, a high-performance
P-removing sludge containing 15.1% P in the biomass, was comprised of large clusters of polyphosphate-con-
taining coccobacilli. By FISH, >80% of the A sludge bacteria were ␤-2 Proteobacteria arranged in clusters of coc-
cobacilli, strongly suggesting that this group contains a PAO responsible for EBPR. The second dominant
group in the A sludge was the Actinobacteria. Clone libraries of PCR-amplified bacterial 16S rRNA genes from
three high-performance P-removing sludges were prepared, and clones belonging to the ␤-2 Proteobacteria were
fully sequenced. A distinctive group of clones (sharing >98% sequence identity) related to Rhodocyclus spp. (94
to 97% identity) and Propionibacter pelophilus (95 to 96% identity) was identified as the most likely candidate
PAOs. Three probes specific for the highly related candidate PAO group were designed from the sequence data.
All three probes specifically bound to the morphologically distinctive clusters of PAOs in the A sludge, exactly
coinciding with the ␤-2 Proteobacteria probe. Sequential FISH and polyphosphate staining of EBPR sludges
clearly demonstrated that PAO probe-binding cells contained polyphosphate. Subsequent PAO probe analyses
of a number of sludges with various P removal capacities indicated a strong positive correlation between P
removal from the wastewater as determined by sludge P content and number of PAO probe-binding cells. We
conclude therefore that an important group of PAOs in EBPR sludges are bacteria closely related to Rhodo-
cyclus and Propionibacter.

The removal of phosphorus (P) from wastewater can be not been unambiguously identified, and the mechanisms of P
achieved by chemical precipitation or by biological mecha- removal are unknown. Researchers have constructed biochem-
nisms in a process called enhanced biological phosphorus re- ical models that accommodate the gross chemical transforma-
moval (EBPR). The basic configuration of an EBPR-activated tions observed in EBPR processes (15, 43).
sludge plant has the influent wastewater going into an anaer- There have been many investigations attempting to match
obic zone, where it is mixed with the returned microbial bio- the metabolic performance of bacterial isolates with the bio-
mass from the secondary clarifier to form the so-called mixed chemical model suggested for EBPR. These have concentrated
liquor. This mixed liquor then flows into an aerobic zone, after mostly on isolates of the genus Acinetobacter because members
which the biomass is separated from the treated wastewater in of this genus are easily isolated from EBPR sludges (21, 26, 42)
the secondary clarifier. Polyphosphate-accumulating organ- and some isolates show some characteristics that may be im-
isms (PAOs) (40) are selectively enriched in these systems, and portant to EBPR (16, 36). However, evidence indicating that
excessive phosphate accumulation occurs in the aerobic zone. Acinetobacter may not be responsible for EBPR includes pure
Removal of a portion of the growing biomass (waste-activated culture performances not correlating with biological models
sludge) results in the net removal of P from the wastewater. (7, 38) and analyses of EBPR bacterial communities indi-
Figure 1 shows the profiles of chemical transformations rele- cating that Acinetobacter microorganisms are not present in
vant to EBPR. high enough numbers to account for EBPR (7, 14, 24, 30,
Empirical experience over the last 30 to 40 years of EBPR 41). Investigations of other EBPR-associated microorganisms
operation has permitted plant operators to more successfully are limited, although there has been some interest in gram-
conduct EBPR processes (K. J. Hartley and L. Sickerdick, positive bacteria such as Microlunatus phosphovorus (33, 39),
presented at the Second Australian Conference on Biological the gram-negative Lampropedia (35), and the Actinobacteria
Nutrient Removal from Wastewater, Albury, Victoria, Austra- and ␣ Proteobacteria (25). However, there is no general con-
lia, 1994). However, the study of EBPR microbiology is impor- sensus that these bacteria are examples of PAOs, and indeed
tant because the process does fail intermittently, PAOs have Mino et al. (31) concluded that rather than there being a single
dominant PAO, several different bacterial groups may be im-
portant. The isolation of putative PAOs is hampered by the
* Corresponding author. Mailing address: Department of Microbi- lack of an easy method of using the P removal phenotype in
ology and Parasitology, The University of Queensland, St. Lucia, 4072 isolation strategies. It is evident that more knowledge of the
Queensland, Australia. Phone: 61 7 33654645. Fax: 61 7 33654620. microbial ecology of EBPR is required before research facili-
E-mail: blackall@biosci.uq.edu.au. ties commit to in-depth investigations of particular isolates. In

1175

2 described (13). (8). Bacterial 16S rRNA gene (rDNA) clone libraries were anaerobic prepared from genomic DNA extracted from frozen A. Briefly. Bode. sequences. All clone sequences were ex- containing over 15% PO4-P in the biomass. Briefly. B. determined by reference to the coccobacillus clusters. and images of methylene blue stains were re. Each of the compiled sequences was compared to publicly avail- tative PAO in extremely efficient hyper-P-removing sludges able databases using the basic local alignment search tool (BLAST [1]) to determine approximate phylogenetic affiliations. reactor GRC fluctuated over a 12-month period. FISH cause there are no pure cultures whose 16S rRNA would bind the PAO probes.de/). formamide. Strunk. Mixed in the ARB software package. The coverslip was removed from the Feed Concn (mg/liter) slide. 13).1176 CROCETTI ET AL. using 16S rRNA. For the B sludge. Ludwig. and use with the A times. A and B. bound the PAO probes. Stuckmann. A. Schleifer. this paper. Ginhart. MLSS is in milligrams/ the software package SeqEd (Applied Biosystems. MLSS. ing a series of FISH experiments at 5% formamide increments starting at 0% Final images were prepared in Adobe Photoshop. cellular PHA (E). Sydney. c lyzed (9. ENVIRON. Two was tested by bootstrap analysis. extracellular acetate (F). B. and at regular stable operating (iii) Probe design from clone libraries. and P (8) sludges. Ludwig. and Cyto. chemical oxygen demand. acetate was not detected.05 15. primers 27f and 1492r (27) were used for PCR amplifi. Final images were prepared in Adobe Photoshop. Phylogenetic trees were constructed by to 2 liters were operated under anaerobic/aerobic cycling conditions to achieve carrying out evolutionary distance analyses on the 16S rDNA alignments. M. A representative selection of clone inserts was fully sequenced with In both conditions. GRC 28 389 18 3. Sludge PO4-P Acetate CODa:P MLSSb sludged corded. and amplified genes were cloned using a TA a cloning kit (Invitrogen. sludge. amined with the CHECK_CHIMERA program (28) to identify any chimeric directed probes.070 77 6. The performance of parameter and parsimony analysis (version 4.). d a range of primers (5).7 6. The A similar approach was used for “Microthrix parvicella” by Erhart et al. Gram designed oligonucleotides were synthesized and labeled at the 5⬘ end with the staining by the modified Hucker method was from Jenkins et al. Methylene blue staining (for polyphosphate) and Gram staining (8) were sequences which allowed their discrimination from all other reference sequences. Calif.7 cation of near-complete 16S rDNAs. conducted with two sludges (A and GRC). mixed liquor suspended solids (in milligrams per liter). M. San Diego. and the slides were stained with methylene blue. Based on comparative analysis of all sequences in cultures (sludges) from three SBRs (A. Light beled probes were evaluated with paraformaldehyde-fixed A sludge. S. sequences were compiled using liter. Clones of RFLP group representa. the optimum formamide concentrations were phaga-Flavobacterium (for probe details. and ␥ Proteobacteria. 1. A.8 s per frame with 4. B.1 tion fragment length polymorphism (RFLP) analysis using methods previously B 53 425 4. COD. The compiled sequences were aligned using the ARB software pack- age and database (O. Phylogenetic analysis of the 16S rDNA sequences was P% ⫽ (PT ⫺ Pe/MLSS) ⫻ 100 (PT ⫽ total sludge phosphate in milligrams/ performed as described previously (17). the sludge was collected and used in the study. The polyphosphate) was done as described in Eikelboom and van Buijsen (18). using EBPR. Operating data for three laboratory-scale EBPR reactors Sequential FISH and methylene blue staining was carried out by first collect- ing images from FISH preparations. T. Lenke. Under all but the lowest-stringency conditions. and cellular carbohydrate (Œ) during the anaerobic and aerobic reactor cycle stages of the P sludge of Bond et al. Enumeration distinct clusters of methylene blue-positive coccobacilli were the only cells which of ␣. http://www. and GRC) were investigated by the ARB database comprised of publicly available sequences and our in-house classical cell staining procedures and by fluorescence in situ hybridization (FISH) clone sequences. Frame size was 512 by 512 pixels. the program selected specific regions within the putative PAO (4). France). Reichel. Nonhoff. A.25 numerical sample spotted on the slide. S. (mg/liter) (mg/liter) End of Effluent (ii) Clone libraries.3 1.mikro. Therefore. New South Wales.-H. ␤ (including ␤-1 and ␤-2). (i) Microscopy of EBPR mixed cultures. Herrmann. liter).692 144 ⬍0.0b2a of PAUP* [37]). N. Scan time was 31. preparations were viewed on both a Zeiss LSM510 and a Zeiss Axiophot.48-␮s pixel dwell time. (23). Sequencing batch reactors (SBRs) with working volumes of 1 and alignments were refined manually.160 110 28 17. The robustness of the tree topology al. see Table 2) were reported as propor. and inserts from individual clones were amplified and grouped according to restric. the mounting fluid was removed by rapid washing and drying. Actinobacteria. Profiles of soluble extracellular phosphate-P (䊐). An argon laser 488-nm line and the HeNe 543-nm line were used for imaging. FIG. and operating data for three reactors are summarized in Table 1.tu-muenchen. and indocarbocyanine dye CY3 by Genset (Paris. . A 57 309 9 3. b tives were partially sequenced using primer 530f (27) and phylogenetically ana. EBPR reactors. (8). O. Vilbig. APPL. we report the phylogenetic identification of a pu. Australia). These fluorescently la- poly-␤-hydroxyalkanoate (PHB) staining was as described by Murray (32). the morphologically Samples were fixed and hybridized as reported by Bond et al. using neighbor joining with the Kimura two- reactors. Images from the Zeiss Axiophot were collected by a cooled charge-coupled device and initially processed by Kontron software (KS200). MATERIALS AND METHODS K. probe synthesis. Pe ⫽ phosphate in the effluent in milligrams/liter.biologie. (19). (8). The form- micrographs of Gram and methylene blue stains were captured on a Nikon amide concentration for optimum probe stringency was determined by perform- Microphot FXA microscope via a charge-coupled device connected to a PC. were operated to highly enrich for PAOs. aperture objective. Zeiss LSM510 confocal laser scanning microscope employed an Axiovert 100M Generally all three designed PAO probes were applied to any one individual SP inverted optical research microscope and a Plan-Neofluar 63⫻/1. SBR operation and monitoring were similar to that reported by Bond et the appropriate tool in the ARB database. Gross. Neisser staining (for Sequences were subsequently confirmed for specificity using BLAST (1). May. This was necessary be- tions of all Bacteria (according to probe EUB338 [8]) for the A sludge. PAO-specific probes (Table 2) were designed using the probe design tool Microbiological analyses. The fields from which FISH images PO4-Pc P% in had been collected were located. MICROBIOL. TABLE 1. and W.

During the teobacteria. Three PAO probes were de- division ␤-2 (i. all sludges took up large amounts spite this slight lack of specificity. 2000 PHOSPHORUS-REMOVING BACTERIA FROM WASTEWATER 1177 TABLE 2. All PAO probes are listed in Clone libraries. signed to target positions 846 to 866. 663–679 Competitor for BONE23a 35 2 GAM42a GCCTTCCCACATCGTTT 23S. probe BTWO23a helped of P. 319–336 Cytophaga-Flavobacterium 35 29 PAO462 CCGTCATCTACWCAGGGTATTAAC 16S. partially sequenced clones belonging Microscopy. signed to specifically target the SBRA220 cluster. ments using the BTWO23a probe (Table 3 and reference 8). b See Fig. In light of a recent publi- therefore a good source of PAO sequences from which specific cation systematically documenting the in situ accessibility of FISH oligonucleotides could be designed. Large clusters comprising hundreds of gram. coli num- performance EBPR systems (Table 1 and reference 8) and bering) and did not work in FISH. polyphosphate-containing (either methylene blue. 846–866 PAO clusterb 35 This study Rc988 AGGATTCCTGACATGTCAAGGG 16S. A range of sludges from logenetic groups in the A sludge clone library did not match laboratory-scale processes and full-scale EBPR plants was collected. Figure 3 A and B sludges. raising the putative probe ported in Table 3. 66. EBPR reactors. tions 28 through 1491. 1901–1918 Actinobacteria 25 34 CF319 TGGTCCGTGTCTCAGTAC 16S. which equates to ca. sludge (8) were evaluated by RFLP. the probe was rede- tives were partially sequenced. Figure 2A demonstrates the purple polyphos. clones from high-performance EBPR sludges. 1027–1043 ␤ Proteobacteria 35 29 BONE23a GAATTCCATCCCCCTCT 16S.VOL. 2B shows the typical microbial complexity of an clones from which the probes were designed and the specificity EBPR sludge by Gram staining. It is worthwhile noting that probe PAO846 (Table 2) was sen to generate 16S rDNA sequences because they were high. (iv) Use of designed probes with other sludges. The putative PAOs were broadly high- in Table 1. ever. E. probe design. Occasional tetrad-arranged of the probes. designed but not fully evaluated. 3) was members of this bacterial subdivision. clone libraries were used simply as a mechanism RESULTS to generate sequences for probe design. 50% inorganic polyphosphate. scale EBPR processes and were therefore deemed good mod. This can be ing. its lack of specificity is not surpris- by the biomass during the initial anaerobic stage. 663–679 ␤-1 Proteobacteria 35 2 BTWO23a GAATTCCACCCCCCTCT 16S. cells hybridizing to probe BTWO23a). It is recognized hybridized with the newly designed PAO probes after determining the form- amide concentration for optimum probe stringency. Group representa. additional more-specific probes are still re- were defined as hyper-P removing since they contained ⬎15% quired for the PAOs in the ␤-2 Proteobacteria group. In addition. coli numbering (12). To this PO4-P. In addition to ␤-2 Proteobacteria. E. that clone libraries may not provide quantitative data on the microbial community structure of the sample analyzed. 3. c ND. some ␥ Pro- and at the end of the anaerobic stage (Table 1). OPB9. How- characteristic arrangement in packets of four or eight cells.. 250 from the B sludge. These sludges were cho. it also targets (with no appreciated by comparing PO4-P and acetate data in the feed mismatches) members of the ␤-3 Proteobacteria. coli numbering) ␤-2 Proteobacteria while Fig. focus group for probe design. all clones from the A. 651–668 PAO clusterb 35 This study PAO846 GTTAGCTACGGCACTAAAAGG 16S. shows a phylogenetic tree of the completely sequenced (posi- phate-containing coccobacillus clusters in the GRC sludge. as the ␤-2 Proteobacteria probe (BTWO23a) was els for EBPR. since there was P release and acetate uptake teria probe (BONE23a [2]). Table 1 shows that all three SBRs were per. specifically of the sub. 988–1009 “Rhodocyclus group”b NDc This study a rRNA. and a green nonsulfur division clone. Our laboratory-scale systems exhibited carbon lighted as ␤-2 Proteobacteria by group hybridization experi- and phosphorus transformations like those found in full. it removed to the ␤-2 Proteobacteria were fully sequenced (positions 28 ⬎20 mg of PO4-P per liter from the wastewater (compare 28 through 1491. 1027–1043 ␥ Proteobacteria 35 29 HGC69a TATAGTTACCACCGCCGT 23S.or and non-EBPR clone libraries (9) and sludge clone SBRH147 Neisser-positive) coccobacilli overwhelmingly dominated the from an unpublished library were fully sequenced. end. However. dom. originally designed to target positions 841 to 857 (E. Reactor operating data are summarized Probe development. only the SBRA220 cluster was comprised exclusively of Table 3 reports the group hybridization results from the hyper. and an ad- inated the A sludge community. to the ␤-2 Proteobacteria from two previously reported EBPR negative. 462–485 PAO clusterb 35 This study PAO651 CCCTCTGCCAAACTCCAG 16S. FISH probes to E. B. but in this research.7 mg/liter in the efflu- ent) and contained 6. Note that the relative proportions of phy. fixed. However. A and B sludges teria. coli 16S rRNA (20).e. Information relevant to FISH oligonucleotides used in this study Probe Sequence (5⬘–3⬘) rRNA target sitea Specificity % Formamide Reference EUB338 GCTGCCTCCCGTAGGAGT 16S. and the overall results are re. suggesting that PAOs may be ditional probe of broader specificity called Rc988 (Fig. Two main clusters of EBPR sludge clones were cocci or “G” bacteria (6) were observed according to their observed (SBRA220 cluster and GC24 cluster [Fig. and those determined by FISH probing (Table 3). 19–35 ␣ Proteobacteria 20 29 BET42a GCCTTCCCACTTCGTTT 23S. Escherichia coli numbering) in preparation for mg of PO4-P/liter in the influent with 6. and 89 from the P cies. This became the P-removing A sludge. not determined. originally designed only as a competitor for the ␤-1 Proteobac- forming EBPR. as seen by comparing the PO4-P values at the end of the refine the identity of the PAOs to a subset of the ␤ Proteobac- anaerobic stage with those from the effluent. De- subsequent aerobic period. 338–355 Bacteria 20 3 ALF1b CGTTCG(C/T)TCTGAGCCAG 16S. A total of 281 bacterial 16S rDNA clones Table 2 with their empirically determined optimum stringen- from the A sludge.7% P (Table 1). ␤ Proteobacteria. 3]). and P sludge libraries belonging The GRC sludge was a good P-removing sludge. accessibility from 8 to 42% to 42 to 52% (relative probe fluo- .

The bar is for both panels and is 6 ␮m. Standard-arrowed orange clusters of cells match the morphology and size of the purple cells in panel A. Morphologically identified “Nostocoida limicola” II can be seen as filaments of gram-positive and gram-negative cells. (A and B) Bright-field micrographs of GRC sludge as operated according to data in Table 1. (A) Methylene blue stain. FIG. Standard-arrowed purple clusters of cells are those containing polyphosphate. Diamond-ended-arrowed pink cells match those of the blue cells in panel A. Micrographs of mixed liquors from SBRs. while diamond-ended-arrowed blue cells do not contain polyphosphate. APPL. fluorescein labeled) and a mixture of all three . (C and D) Confocal laser scanning micrographs of sludges dual hybridized with EUB338 (25 ng. 2. MICROBIOL. (B) Gram stain.1178 CROCETTI ET AL. ENVIRON.

by using this concerted hy- percent P (Fig. in the Loganholme sludge. Blue cells in Fig. CY3 labeled). These PAO probes were used with A sludge bio- During various periods of stable operations of the GRC mass. PAO probe. and S sludges (8. and expressed as a proportion of EUB338-binding cells. nuis and R. 3) are from Rhodocyclus (R. numbers of PAO probe-positive cells were counted arrangement resembled those binding the BTWO23a probe. In this preparation. and abundance of DISCUSSION those staining positive for polyphosphate by the methylene blue stain (Fig. Total clones in library NAc 281 250 89 pending on EBPR performance. and superimposed. (D) Lightly sonicated mixed liquor from an EBPR SBR (ca. ␥ bers of clusters were observed. either the Additionally. We present indirect and direct cells in Fig. each 25 ng. 2C). (C) Mixed liquor from SBR A with operating data as given in Table 1. ␤-1 and ␤-2 were determined by using BONE23a and BTWO23a probes. (% of total clones) in phylogenetic relationship. Use of designed probes. sequence from a recently published Swiss EBPR sludge (R6 P. B. 10% P in the sludge) operating at 3. likely because E. since they are dual labeled with EUB338 (green) and PAO (red) probes.5% NaCl from a study of seafood-processing wastewater. The proportion [22]) was in the group containing the full clone inserts from the of PAO probe-binding cells was plotted versus the sludge A. Proportions of major bacterial divisions in the A sludge iments. Separate images were collected for fluorescein and CY3 excitation. A positive correlation between these two bridization approach (Fig. A series of fixed sludges including ␣ Proteobacteria 12 38 (14) 32 (13) 5 (6) ␤ Proteobacteria (mostly 80 (1 ␤-1. Yellow methods like cloning and FISH. and superimposed. and P sludges (Fig. 2A are simply dried sludge suggested that they played a major role in P removal. In addition. size. 66. B. The bar is for both panels and is 4 ␮m. . the PAO probe-positive cells matched the morphology. making it appear as gran. 2A. (E) Epifluorescence micrograph of GRC sludge (Table 1) dual hybridized with EUB338 (25 ng. which stains purple. When used with full-scale and other Our approach to discovering the identity of the PAOs was to laboratory-scale EBPR sludges. and the Loganholme sludge were evaluated Actinobacteria (mostly 28 67 (24) 22 (9) 8 (9) with the designed PAO probes. B. (F) Methylene blue-stained image of the same field as in panel E. For example. Arrowed yellow cells are the PAOs. the PAO probes and the ␤-2 selectively enrich for organisms with a P removal phenotype Proteobacteria probe always bound the same cells whose mor- PAO probes (Table 2. a full-scale a activated sludge plant treating domestic wastewater with an Comments in parentheses refer to clone library data only. we demonstrated that the de- parameters is evident. Large numbers of the clusters Proteobacteria comprised 1% of the Bacteria. diamond-ended-arrowed green-colored cells are other bacteria. the clusters of PAO probe. The The closest pure-cultured bacterial relatives to the ␤-2 Proteo- activity of the sludge from the GRC reactor was high. 2F do not contain polyphosphate. arrangement. In all full-scale and laboratory. not applicable. Images were collected for fluorescein and CY3 channels. 2F) rather than evenly distributed around PAOs (called PAO probes) were designed from a group of the cells (Fig. Thus. (9) had already paraformaldehyde or the chemicals used in FISH seem to have associated bacteria of the Rhodocyclus group in the ␤-2 Pro- had an effect on the polyphosphate. moving A sludge bound the ␤ Proteobacteria probe (BET42a). 2A and C).VOL. polyphosphate. The typ. clone library data of Bond et al. De. where they bound cell clusters whose morphology and reactor. purpureus) and Propionibacter pelophilus. The bar for both panels C and D is 10 ␮m. more or fewer clusters were analyzed by RFLP present. Probes for the putative ules in the cells (Fig. ical appearance of PAOs by methylene blue staining is shown these organisms were likely ␤-2 Proteobacteria since they in Fig. the three PAO probes bound the same cells as bound the ␤-2 Proteobacteria probe (BTWO23a). 2E and F have been fixed in paraformaldehyde nance of the BTWO23a-hybridizing group in the enriched and undergone FISH. fluorescein labeled) and PAO651 (Table 2. The numerical domi- The cells in Fig. and P clone libraries as determined by have similar ribosomal higher-order structures (and therefore RFLP analysis and sequencing of RFLP group representatives similar probe accessibility profiles) due to their relatively close No. 2E are dual hybridized with EUB338 (fluorescein) evidence of the identity of the PAOs and quantitation of the and PAO651 (CY3) probes and are therefore putative PAOs. 2F). were observed in the hyper-P-removing A sludge (Fig. 2D) as in all sludges. onto microscope slides before staining. 10) were also counted. 81 ␤-2) 13 (5) 44 (18) 15 (17) the A sludge. and design and evaluation of probes from A B P these Actinobacteria clone sequences are under way. Diamond-ended-arrowed blue cells are the same green cells in panel E. A clone positive cells in other previously reported sludges like the Q. Figure 2E and F clearly demonstrate that highly related ␤-2 Proteobacteria clone sequences (ⱖ98% iden- the putative PAOs according to FISH are those containing tical) retrieved from several enriched P-removing sludges (A. 4). Cytophaga-Flavobacterium 14 83 (30) 52 (21) 45 (51) binding coccobacilli were distinct and uniform and resembled group the yellow cells and clusters arrowed in Fig. 2C and D. 10 mg of PO4-P/liter. artificially colored. TABLE 3. and P). 3). the polyphosphate. Standard-arrowed yellow cells are PAOs. with the bacteria clone sequences (Fig. te- vast majority of cells binding the EUB338 probe. Figures 2E and F show results for FISH and methylene and to evaluate this enriched culture by non-culture-dependent blue-stained sludge from the GRC reactor (Table 1). CY3 labeled). moderate num. sludge from the GRC reactor at different oper- Rhodocyclus relatives) ational periods. artificially colored. 2A). Standard-arrowed cells containing purple granules of polyphosphate are the same yellow cells in panel E. In c NA. A total of 80% of the microbial biomass in the hyper-P-re- appears as granules in the middle of PAOs (Fig. Therefore. Sludge kindly supplied by Nugul Intrasungkha. 3). Bacterial division or FISH of A sludge clone library determined Actinobacteria was the second dominant group in the A (% binding to by RFLP analysis subdivisiona probeb) sludge (Table 3). Terrabacter relatives) scale EBPR sludges examined. 25 ng. while those in Fig. the saline sludge (Fig. b Percentage of cells in the A sludge binding the EUB338 probe for all Bac- influent containing ca. coli and the Rhodocyclus-like PAOs by FISH and in the A. bound the BTWO23a probe (Table 3). teria. The redesigned probe worked well in FISH exper. teobacteria with improved P removal. PAOs in sludges. signed probes were specific for the dominant ␤-2 Proteobacteria in the A sludge. 2000 PHOSPHORUS-REMOVING BACTERIA FROM WASTEWATER 1179 rescence).

half-shaded circles are for analyses where one algorithm gave ⬎75% bootstrap support and the other gave 50 to 75%. Australia) determined in this study and sequences from publicly accessible databases. TCB. OPB9. Closed circles at nodes indicate ⬎75% bootstrap support from both analyses. and the A sludge (15. 3. cesses.. cally distinct PAOs may also exist in EBPR. SBR2. Cells containing polyphosphate also bound the very likely detecting a true PAO in the laboratory-scale pro- probe PAO651 (Fig. SBRH. and quantitative evi- P. 56% ␤ dence that bacteria closely related to Rhodocyclus and P. philus are examples of PAOs. Dashed lines indicate that the probe does not have 100% identity with the indicated sequence. 15% P in the sludge) (P⫹⫹⫹). Rubrivivax gelatinosus (D16213) was used as the outgroup in analyses but is not shown in the tree. other phylogeneti- teobacteria) were compared (plot not shown). between percent P in the sludge and numbers of PAO probe- phate. However.8% stains. Queensland. we have provided indirect. from Fig. and levels of ␤ the A sludge with PAO probes in concert with methylene blue Proteobacteria was observed when data for the P sludge (8. or a non-P-removing sludge (P⫺).1180 CROCETTI ET AL. rigorous validation of our probes in full-scale processes is ratory-scale EBPR sludges demonstrated a positive correlation now required. The coding indicates that the clone came from a hyper-P-removing sludge (ca. The specificity of the published ␤-2 Proteobacteria probe (BTWO23a) and those of probes designed in this work (PAO probes and Rc988) are shown by solid lines. An indicative correlation between increasing P removal per. Direct evidence that the PAO probes bound true PAOs binding cells (Fig. ENVIRON. Phylogenetic tree of near-complete 16S rDNA sequences obtained from sludges A. 80% ␤ Pro. Rc988 has one mismatch (at position 1009) with the SBRP112 sequence. pelo- Proteobacteria [10]). MICROBIOL. direct. B. as judged by percent P in the sludge. Subsequent in. Chromobacterium. . Chromatium spp.02 nucleotide change per nucleotide position. 45% ␤ Proteobacteria [8]). FIG. 4). Therefore. For example. The scale indicates 0. APPL. Iodobacter. SBR1. open circles indicate 50 to 75% from both analyses.1% P. In addition to specifically targeting the sequences indicated in the tree. the S sludge (12. phology was similar to those staining positively for polyphos.3% P. 2E and F). P. 4 and data on the hybridization from formance. and GC (Gold Coast.000 bootstrap resamplings. and a green nonsulfur division clone. The more vestigations with the PAO-specific probes on a range of labo. trichlorobenzene. This quantitative data demonstrates that came from sequential FISH and methylene blue-staining ex. a fair P-removing sludge (P⫹). the designed PAO probes for particular ␤-2 Proteobacteria are periments. probe BTWO23a also targets (with no mismatches) members of the ␤-3 and ␥ Proteobacteria. a good P-removing sludge (P⫹⫹). Evolutionary distance and parsimonious analyses were carried out in PAUPⴱ employing 2.

the GRC sludge at varying but stable P-removal efficiencies (Œ). R. F. P. In quence from his clone library. T.. Rev. L. An additional procedure 61:1910–1916. 1997. and S (F) sludges of Bond et al. Binder. B. E. Queensland. Investigations of the microbial ecology of enhanced bio- After the review process. and the A sludge (■) (this study). J. J. and the number and specificity of probes biological phosphorus removal activated sludges with dissimilar phosphorus designed. 7. 6. Myers. (22) logical phosphorus removal in the activated sludge process. W. 1981. 1997.. En- removing mixed microbial communities in laboratory-scale viron. M. M. Ludwig. and K. L. 1990. W. D. Brisbane. J. Our paper essentially corroborates the data University of Queensland. Water Sci. Schleifer. 3. and Pollution Control Ltd.. probes with flow cytometry for analyzing mixed microbial populations. R. P.S. Microbiol. P. and D. I. Environmental Protection Microbiol. F. Hesselmann et al. R. Amann. Blackall. Lipman. Ph. 59:143–169. S. S. Appl. Tandoi. Comparing the two reports. reactor. ACKNOWLEDGMENTS et al. sludge clone libraries. are genuine PAOs is common to both studies and suggests that these organisms may play a global role in EBPR.. The characterization and description of ADDENDUM representatives of “G” bacteria from activated sludge plants. Gavin Symonds of Carl Zeiss Australia situ visualization of high genetic diversity in a natural microbial community. Brosius. Sludges evaluated included the Q (夹). Stahl. 25:63–69. L. Keller. Gene Tyson determined some of the RFLPs on the A and B Basic local alignment search tool. C. 1996. 10. called Accumulibacter phosphatis by Hesselmann 12. Keller. 215:403–410. 1999. and probe design and application. 65:4077–4084. Microbiol. P. Agency STAR graduate fellowship and by grant BES-9612640 from 5. 11). L. Water Sci. L. Combination of 16S rRNA-targeted oligonucleotide for their valuable assistance. cation and in situ detection of individual microbial cells without cultivation. efficient biological phosphorus removal activated sludge systems. Hugenholtz. L. W.-H. Blackall. (22). Technol. Some of the work was funded by the CRC for Waste Management Environ. Appl. P. Bond. 1999. FISH with published 9. removal performance. Technol. Wagner. Blackall. Amann. and H. The was published. Blackall. Noller. 4. Technol. L. Gish. 29(7):35–42. SBRs. bacteria.. Dull. Australia. Bacterial group probes. 39(6):13– differences in the cloning strategies. Amann. H. Appl. 1995. Ludwig. Rossetti. presented in that report. Bond. the major conclusion that Rhodocyclus. assisted with preparation of the CSLM images. P. I. cloning 16S rDNA from the enriched cultures. like bacteria. cell staining (Neisser and PHB). Microbiol. 1990. J. Paul Burrell provided the SBRH147 clone se. operator of the saline 1. REFERENCES lution Control Plant and Nugul Intrasungkha. L. Mol. Christenssen. and L. Snaidr. Environ.. J. (22) was dot blot hybridization with 10. L. and L. R.VOL. However. 1994. (8. Blackall. Water Sci. Cunningham. 56:1919–1925. Microbiol. Gene organi- . including enrichment of hyper-P. L. Schleifer. Molecular identification of activated sludge foaming the National Science Foundation. R. J. Chisholm. P. a report by Hesselmann et al. Phylogenetic identifi- the Australian Government’s Cooperative Research Centres Program. L. J. A. We acknowledge all these D. L. 2. and V. D. Characterisation of enhanced quenced and analyzed. 1998. Steeter. Miller.. Andrew Schuler was supported by a U. the number of clones se. and Carl Zeiss Aus. 37:567–571. Appl. P (}).. a center established and supported under 4.. Bond. 178:3496–3500.. Olson. Erhart. L. one finds identification by a novel microbial approach. Bond. J. community structures of phosphate-removing and non-phosphate-removing activated sludges from sequencing batch reactors. and L. Bacteriol. 66. and K. M. Lett. We thank the operators of the Logan City Loganholme Sewage Treatment Plant and the City of San Francisco Southeast Water Pol. J. 11. S. 2000 PHOSPHORUS-REMOVING BACTERIA FROM WASTEWATER 1181 FIG. The identification of some of the major groups of bacteria in efficient and non- similar to those used by us. and L. Wagner. Blackall. (22) used methods 8. Bond. W. J. L. P. Hugenholtz. W. Keller. 20. J. used by Hesselmann et al.D thesis. W. Keller. Devereux. Hartman. 1995. Biol. Altschul. R. Relationship between sludge P content (percentage of dry mass) and proportion of cells binding all three PAO probes as a percentage of the EUB338 probe-positive cells. and tralia loaned the LSM510 for this purpose.. Bio-P and non-bio-P bacteria extracted nucleic acids.

PAUP*: Phylogenetic Analysis Using Parsimony (* and Zehnder. Schleifer. and L. Pauli. Hancock. and R. 1999. K. Lötter. Y. 1999. C. and A. 1994. 1981. J. Kuba. R. 4⬘. P. Wagner. R. a new 19. Bortone. Wagner.Y. biological phosphorus removal. van Loosdrecht. L. Wentzel. 339–342. G. Erhart. M. A. Wedi. M. 1987. Int. In R. M. 3877. van der Meer. Dojka. Amann. phylogenetic analysis and detection of a bacte. Woese. R. Roller. 33. Water Sci. Enhanced culture development. 15:593–600. J. Nucleic acid techniques in bacterial systematics.. sludge-culture-dependent numerical identification versus in situ identifica. Lilburn. J. 1998. Steyn. Syst. Beisker. J. A. H. and K. D. I.. Antonie Leeuwenhoek 71:109–116. E. Lemmer. 60:792–800. Ramadori. M. Burrell.). Erhart.C. H. J. Nakamura.-H. Wallner. van Loosdrecht. 1999. Overbeek.. 30(6):229–235. G. Part 26. R. P. Pace. M. FEMS Microbiol. Melasniemi.. Microbiol. R. Stante. Blackall. Microbiology diversity in a hydrocarbon. Lane. 2:119–138. T.0b2a. 24. 31:1317–1324. Marais. Environ. 1994. B. Hall. Res. 1991. Wagner. M. Evaluation of biochemical models for biological excess phos- ford. undergoing intrinsic bioremediation. J. APPL.. Werlen.. Y. T. 1997. Loewenthal. lating bacteria from sewage treatment plants with different processes for Ecol. Water Sci. 14. Environ. K. Microscopic sludge inves. Heijnen. Amann. accumulating large amounts (BPR) and non-BPR activated sludge bacterial communities with particular of phosphate. Water Environ. Ekama. p. Kanagawa. Oxford. TNO Research Institute for Environmental Hygiene. M. 1998.. J. Maidak. Microb. In Syst. E. 34. Appl. 1999. R. R. Pramanik. MICROBIOL. 1998. L. W. W. Res. Amann. and N. E. Microbial 31. M. Heijnen. 17. D. 17(12):119–125. Appl.. 115–175. Seviour. Haack. 1996.. R. 32. McCaughey. Appl. 1993. Enrichment. van Loosdrecht. G. and M. Technol. and R.. Scholten. L. Li. New 39. Keller. 27:171–173. Comparative analysis of biological phosphate removal physiological characteristics of Acinetobacter spp. Appl. Water Res. sp. Microbiological basis of phosphate removal 36. and M. p. T. P. R. wastewaters. Erhart. Tandoi. Salki- 16. 22.. Ludwig. J. J. C. 45:17–22. situ probing of Gram-positive bacteria with high DNA G⫹C content using 20. ing bacterium isolated from anaerobic/aerobic activated sludge. B.. E.. Schleifer. C. 1997. 1987. S. H. Biological phosphate removal from (Ribosomal Database Project). (ed. R. 16S/23S rRNA sequencing. Swofford. W.. Fuhs. phate removal. Y. and M. Microbiol. Garrity. Marais. A. J. The Netherlands. van Buijsen. Ludwig. Appl. and R. Daigger. C. Development of an rRNA-targeted oligonucleotide In situ identification of polyphosphate-accumulating bacteria in activated probe specific for the genus Acinetobacter and its application for in situ sludge by dual staining with rRNA-targeted oligonucleotide probes and monitoring in activated sludge. Goodfellow (ed. A combined fluorescent antibody. M. phorus removal from wastewater. I. G. R. D. Sinauer Associates. Kämpfer.6-diamidino-2-phenylindol (DAPI) at a polyphosphate-probing concen. Cellamare. W. C. Nucleic Acids Res. 29. Mass. S.. Hesselmann. J. 1986.. Microbiol. C. G. M. Environ. . 1st ed. Tilche. 28. Majone. H. Cloete. 335–338. and 25. M. Wentzel. T. R.. Malaspina. Jenkins. 13. J. Manual of methods for general bacteriology. Manual on the causes chemostat. Ecol. nov. Academic Press. polyphosphate accumulating Acinetobacter isolates in an anaerobic-aerobic 23.and genus-specific rRNA-targeted oligonucleotide probes. and P. D. Deinema.. Schwippl. 1981. J. C. R.). Bohringer.. A. Syst. and H. M. 21. T. Richard. England. 73:113–124.. 1990. J. 1991. 1988. England. W. oxidizing bioreactor. Amann. Microb. A. X. Water Res. Schon. Hugenholtz. The reliability of cold perchloric acid 1. richia coli 16S rRNA for fluorescently labeled oligonucleotide probes.. J. and W. Water South Africa 14:81–92. Pergamon Press. Ekama. Murray. Microlunatus phosphovorus gen. F.. Microbiol. and G. R. ican Society for Microbiology. Hernesmaa. R. Pergamon Press. R. G. E. V. 148:107–127. 32:2903–2912.. L. Hiraishi. Ludwig. and K.. Nakamura. Ramadori 43. S. Technol. R. Golecki. Ind. zation and primary structure of a ribosomal RNA operon from Escherichia 27. D. Water teria: problems and solutions. Blackall. and J.. Larsen. ENVIRON. tration. A new version of the RDP sludge.. Streichan.. K. Schleifer. In R. Chen.). Metabolism of microorganisms responsible for enhanced biological phos- tion using group. 20:310–318. M. Loewenthal. C. W. Phylogenetic oligonucleotide probes for the major subclasses of Proteobac- Biochemical models for enhanced biological phosphorus removal. K. M. J. Microbiology 140:2849–2858. Technol. T.-H. 32:101–121. Characterization of bacterial communities from activated 40. and A. and J. reference to Acinetobacter. C. 23S rRNA-targeted oligonucleotides. Ox. Y. M. Lewis Publishers. Some physiological characteristics of a phosphate remov- York.-H. G. Ubukata. A. Enhanced polyphosphate organism cultures in activated sludge systems. Microbiology of a nitrite. 64:4973–4982. Tiedje. L. R. 64:1878–1883. T. 33:257–265. and G. L. 42. Biological phosphorus removal by pure culture of Lampropedia spp. R. Rantanen. B. membrane filter technique for enumerating Acinetobacter in activated Schmidt. Ecol. Microbiol. p.. 1975. Yoshimi. C. 41. and L. Sunderland. 1992. 1995. Bradford. Microbiol. J. In E.-L. nov. and Delft. other methods). K. M. 1985. V. 32:3193–3207. Oldham. Polyphosphate-accumu- in the activated sludge process for the treatment of wastewater. 37. England. M. T. Chichester. Beimfohr. May. C. B. Biological phosphate removal from wastewaters. 21:300–306. M. G. A. Smolders. H. Some noja-Salonen. Parker. V. 1999. gram-positive polyphosphate-accumulating bacterium isolated from acti- Development and use of fluorescent in situ hybridization probes for the vated sludge. 22:454–465. J. Appl. Ramadori (ed. 20:1511–1521. R. tigation manual. Hahn. Wagner. K. Kawaharasaki. D. Mino. M. and K. Kamagata. Biol. Healey. M.and chlorinated-solvent-contaminated aquifer and biochemistry of the enhanced biological phosphate removal process. M. 23:567–576. 35. 15. Amer- 18. H. M. P. L. 1997. and C. Amann. Bacteriol. 1998. Appl. detection and identification of “Microthrix parvicella” in activated sludge. M. Kawaharasaki. M. J. version 4. 1998. D. D. G. R. Kerdachi.1182 CROCETTI ET AL.. extraction to assess metal-bound phosphates. W. Masuda. Tanaka. N. 1994. The behaviour of sludge. W. N. 64:3869– Water Res. and G. Fuchs. Manz. 30. G. Comeau. and A. J. D. and control of activated sludge bulking and foaming. and K. and G. Amann. J. G. Mol. Cole. J. Washington. rium that performs enhanced biological phosphate removal in activated 38. M. Olsen. Syst. Flow cytometric analysis of the in situ accessibility of Esche. Microbiol. Microbiol. P. Eikelboom. Manz. Water Sci. Stackebrandt coli.