Correspondence: Cunjiang Song, Key Laboratory of Molecular Microbiology and Technology for Ministry of Education, College of Life Sciences, Nankai University, Tianjin 300071, China. Tel./fax: + 86 22 23503866; e-mail: firstname.lastname@example.org
Activated sludge is an alternative to pure cultures for polyhydroxyalkanoate (PHA) production due to the presence of many PHA-producing bacteria in activated sludge community. In this study, activated sludge was submitted to aerobic dynamic feeding in a sequencing batch reactor. During domestication, the changes of bacterial community structure were observed by terminal restriction fragment length polymorphism analysis. Furthermore, some potential PHA-producing bacteria, such as Thauera, Acinetobacter and Pseudomonas, were identified by denaturing gradient gel electrophoresis analysis. The constructed PHA synthase gene library was analyzed by DNA sequencing. Of the 80 phaC genes obtained, 76 belonged to the Class I PHA synthase, and four to the Class II PHA synthase. Gas chromatography–mass spectrometry analysis showed that PHA produced by activated sludge was composed of three types of monomers: 3-hydroxybutyrate, 3-hydroxyvalerate and 3-hydroxydodecanoate (3HDD). This is the first report of production of medium-chain-length PHAs (PHAMCL) containing 3HDD by activated sludge. Further studies suggested that a Pseudomonas strain may play an important role in the production of PHAMCL containing 3HDD. Moreover, a Class II PHA synthase was found to have a correlation with the production of 3HDD-containing PHAMCL.
Polyhydroxyalkanoates (PHA) are biopolyesters synthesized by bacteria as intracellular storage reserves of carbon and energy (Lee, 1996a). PHAs have recently attracted a great deal of industrial attention as promising biomaterials due to their excellent biodegradability, biocompatibility and capability of being produced from renewable resources. PHAs are traditionally divided into two groups depending on their monomer chain length. Short-chain-length PHAs (PHASCL) consist of monomer repeat units of 3–5 carbon atoms, whereas medium-chain-length PHAs (PHAMCL) are composed of monomer repeat units of 6–14 carbon atoms (Lee, 1996b).
The key enzyme for PHA biosynthesis is the PHA synthase. PHA synthase is classified into four classes (Classes I, II, III and IV) based on their substrate specificity, amino acid sequence and structure (Rehm, 2003). Class I and II PHA synthases, made up of only one type of subunit (PhaC) (61–73 kDa) could produce PHASCL (e.g. in Ralstonia eutropha) and PHAMCL (e.g. in Pseudomonas aeruginosa), respectively. Class I PHA synthases preferentially utilize CoA thioesters of various (R)-3-hydroxy fatty acids comprising 3–5 carbon atoms, whereas class II PHA synthases preferentially utilize CoA thioester of various (R)-3-hydroxy fatty acids comprising 6–14 carbon atoms. Class III PHA synthases reported from Allochromatium vinosum have two types of subunits: (1) PhaC and (2) PhaE (c. 40 kDa each), which preferably synthesize PHASCL. Class IV PHA synthases, which resemble class III PHA synthases (PhaE is replaced by a 20-kDa PhaR), have been reported only in Bacillus sp.
The similarities of phaC genes in these PHA synthases is limited, and to date, no common primer set for PCR that covers all of these classes has been proposed. Yet, Solaiman et al. (2000) and Zhang et al. (2001) have proposed primer sets that cover Class II phaC from plural bacterial species. Sheu et al. (2000) have proposed a primer set that cover phaC of Classes I and II. These primer sets for phaC that can cover not all but at least certain range of bacterial species could be useful to investigate the diversity of phaC genes in the natural environment.
An interesting alternative to pure cultures utilized for PHA production until now is the use of mixed cultures, such as the activated sludge from wastewater treatment plants (WWTPs; Satoh et al., 1998; Salehizadeh & Van Loosdrecht, 2004; Dias et al., 2006). The production of PHAs by activated sludge is expected to be more economically favorable than pure culture processes since the activated sludge contains PHA-producing microorganisms within it and there is no need for sterilization and sterile fermentation systems. The enrichment of PHA producers is necessary to increase the accumulation of PHA in activated sludge. The most well-studied strategy for culture enrichment is based on the alternating aerobic availability and unavailability of carbon substrate, the ‘feast or famine’ or ‘aerobic dynamic feeding’ (ADF) conditions, which have been extensively investigated using synthetic organic acids (Serafim et al., 2004, 2008).
To date, most of the data obtained from activated sludge have been limited to the production of PHASCL (Serafim et al., 2004), such as polyhydroxybutyrate (PHB), and its copolymer with 3-hydroxyvalerate [poly(3HB-co-3HV)]. Although a few studies revealed the feasibility of PHAMCL production by activated sludge (Bengtsson et al., 2010; Lee et al., 2011), a great deal of information regarding the PHA-producing bacteria and PHA synthase gene in activated sludge in connection with the efficient production of PHAMCL is still needed.
In this study, a lab-scale activated sludge reactor was operated with an ADF mode. The changes of bacterial community during domestication were monitored by terminal restriction fragment length polymorphism (T-RFLP), and some potential PHA-producing bacteria were identified by denaturing gradient gel electrophoresis (DGGE). Furthermore, the diversity of PHA synthase genes was investigated by constructing a phaC gene clone library. The monomer compositions of produced PHAMCL were identified by gas chromatography–mass spectrometry (GC–MS).
Materials and methods
Activated sludge samples were collected from the aerobic tank of anaerobic–anoxic–oxic (A2O) processes in a full-scale WWTP in Tianjin, China, in April 2011. Samples were taken in sterile polyethylene bottles, transported to the laboratory on ice, and subjected to DNA extraction within 3 h of sampling.
Activated sludge domestication and fermentation
Activated sludge was inoculated into a sequencing batch reactor (SBR) containing 3 L of mineral salts medium (MSM). The compositions of the MSM used in SBR were as follows (g L−1): CH3CH2COONa 3.2, MgSO4 0.6, NH4Cl 0.16, EDTA 0.1, CaCl2 0.07, K2HPO4 0.092, KH2PO4 0.045, and 2 mL of trace elements solution (Serafim et al., 2004).
A schematic diagram of the SBR process is shown in Supporting Information, Fig. S1. The SBR was operated under ‘feast or famine’ conditions. Each SBR cycle consisted of 10.5 h of aerobiosis, 1 h of settling, and 0.5 h withdrawing half of the volume, which was replaced with fresh medium. The SBR was operated for 10 days at 25 °C. During domestication, 100 mL of samples were taken each day from the SBR, and subjected to batch experiments to evaluate its capability to produce PHAMCL containing 3HDD. Meanwhile, the samples from the SBR were serially diluted and spread on Luria–Bertani (LB) plates containing 1.5% (w/v) agar and 0.5 μg mL−1 Nile red stain. The plates were incubated at 30 °C for 48 h and exposed to ultraviolet light (312 nm) to detect colonies of PHA-accumulating bacteria. In each batch experiment, the fermentation medium was the same as that described above for SBR except for CH3COONa (8 g L−1) and NH4Cl (0.037 g L−1). After 16 h of fermentation, PHA was extracted from the activated sludge as described previously (Guo et al., 2011).
Genomic DNA was extracted in triplicate from all samples using a FastDNA spin kit for soil (Q-Biogene) according to the manufacturer's instructions. DNA integrity was checked by 0.8% agarose gel electrophoresis. The purity and the quantity of extracted DNA were determined by UV spectrophotometry at 260 and 280 nm. DNA extracts were stored at −20 °C.
The bacterial 16S rRNA gene was amplified by PCR from the activated sludge community DNA using the universal primers 27f (5′-AGAGTTTGATCMTGGCTCAG-3′) and 1492r (5′-CGGYTACCTTGTTACGACTT-3′) (Weisburg et al., 1991). For T-RFLP analysis, the forward primer 27f was 5′-end labeled with FAM. Three replicate PCR products for each sample were purified using a QIAquick PCR purification kit (Qiagen) and then digested with the restriction enzyme HhaI for 3 h at 37 °C in a mixture containing 8 μL of purified PCR products, 1 μL of buffer, and 1 μL of restriction enzyme (10 U μL−1). Aliquots (2 μL) of the digest were mixed with 2 μL of deionized formamide, 0.5 μL of loading buffer, and 0.5 μL of a DNA fragment length standard (GS-500LIZ, ABI). The mixture was denatured at 94 °C for 5 min and snap-cooled on ice. The fluorescently labeled terminal restriction fragments (T-RFs) were detected by electrophoresis on 7% polyacrylamide gel for 10 h at 2250 V on an automated ABI 3730XL DNA sequencer in the GeneScan mode. T-RFLP profiles were analyzed using genemarker V1.8 software. The relative abundance of T-RFs was determined by calculating the ratio between the peak height of each peak and the total peak height of all peaks within one sample. The T-RFs with a relative abundance of < 1% were regarded as background noise and were excluded from the analysis.
The V6–V8 regions of bacterial 16S rRNA gene were PCR-amplified from the genomic DNA using the primers 968f-GC (5′-AACGCGAAGAACCTTAC-3′ with a GC clamp attached to the 5′ end of the primer) and 1401r (5′-CGGTGTGTACAAGACCC-3′) (Muyzer et al., 1993). Temperature cycling was performed using a touchdown protocol (Ferris et al., 1996) with one cycle of 95 °C for 15 min, 10 cycles of 94 °C for 1 min, 53 °C (each cycle decreasing by 1 °C) for 1 min and 72 °C for 1 min, followed by 20 cycles of 94 °C for 1 min, 43 °C for 1 min and 72 °C for 1 min.
DGGE analysis was performed on a DCode universal mutation detection system (Bio-Rad). PCR products (20 μL) were loaded onto 6% (w/v) polyacrylamide gels in 1X TAE buffer using a denaturing gradient ranging from 40 to 60%; 100% denaturant solution is defined as 7 M urea and 40% (v/v) formamide (Muyzer et al., 1993). Electrophoresis was performed at 150 V for 5 h at 60 °C. After electrophoresis, gels were stained for 10 min with 1X SYBR Gold in 1X TAE buffer and photographed with the Gel Doc 2000 gel documentation system (Bio-Rad).
Some differential or dominant bands were excised from the gel, eluted in 50–100 μL of sterile TE buffer overnight at room temperature and then directly re-amplified with primer set 968f (GC free) and 1401r. The PCR products were verified again by DGGE before sequencing.
Cloning and sequencing of PHA synthase gene
The PCR reaction targeted at a part of the PHA synthase gene (phaC) was performed with the forward primer CF1: 5′-ATCAACAARTWCTACRTCYTSGACCT-3′ and the reverse primer CR4: 5′-AGGTAGTTGTYGACSMMRTAGKTCCA-3′ (Sheu et al., 2000) using a PCR mixture consisting of 0.2 mM of dNTP, 2.5 units of DNA polymerase (Fermentas), 0.03% of DMSO, and 2.5 μM of each primer. The amount of DNA used as a template was 50 ng in 100 μL of PCR mixture. The PCR mixture was pre-incubated at 94 °C for 10 min, 51 °C for 2 min, and 72 °C for 2 min. The PCR cycle consisted of 20 s of denaturation at 94 °C, 45 s of annealing at 57 °C, and 1 min of extension at 72 °C. This cycle was repeated 35 times and then incubated at 72 °C for 10 min for the final extension (Sheu et al., 2000).
The PCR products were fractionated on 1% agarose gel, and DNA from appropriate band was recovered using a DNA gel purification kit (Tiangeng). These fragments were subcloned into the pMD19-T vector for sequencing using a TaKaRa TA cloning kit. In all, 100 white colonies were screened by colony PCR using M13 forward and reverse primers. Plasmid DNA was purified from overnight bacterial cultures and sequenced by an automatic DNA sequencer ABI 373 (Applied Biosystems) at the Beijing Genomics Institute (Beijing, China).
A similarity search of nucleotide sequences was performed using the blastn non-redundant database at http://www.ncbi.nlm.nih.gov/BLAST. Multiple alignments of sequences, construction of a neighbor-joining phylogenetic tree with the Kimura 2-parameter model, and a bootstrap analysis for evaluation of the phylogenetic topology were accomplished using the clustal x program (Thompson et al., 1997) and mega 4 software (Tamura et al., 2007). The nucleotide sequences of the phaC genes have been submitted to the GenBank database.
Characterization of PHA by GC–MS
The monomer composition of PHA was determined by GC–MS after methanolytic decomposition (Braunegg et al., 1978; Satoh et al., 1994). The PHA sample (5 mg) was dissolved in 2 mL CHCl3 and subjected to methanolysis in the presence of 1.7 mL H2SO4 and 0.3 mL CH3OH at 100°C for 140 min in order to obtain the corresponding 3-hydroxyalkanoic methyl esters. The GC–MS procedure was carried out as described previously (Guo et al., 2011). The MS data was searched in the NIST database to determine the monomer structure.
Results and discussion
Operation of an SBR with activated sludge for PHA production
To increase the PHA production rate, bacteria that are capable of accumulating PHA must be enriched in the activated sludge. In this study, a SBR inoculated with the activated sludge was operated under ‘feast or famine’ conditions in which activated sludge was submitted to consecutive periods of external substrate accessibility and unavailability (Serafim et al., 2004). Under these conditions, PHA-accumulating bacteria were enriched, thereby becoming dominant in the complex microbial community. Since Nile red does not affect the growth of bacteria and the Nile red-PHA conjugates produce a strong orange fluorescence, the viable-colony staining method was used in this study to detect colonies of PHA-accumulating bacteria. The numbers of PHA-producing bacteria were 5.2 × 106 and 2.4 × 108 CFU g−1 activated sludge, respectively, at days 5 and 10. At day 10, a large number of colonies grown on the Nile red plate emitted bright orange fluorescence (Fig. S2), which indicated that the PHA-producing bacteria were effectively enriched in the activated sludge.
During domestication, the samples were withdrawn from the SBR for the batch experiments. In the initial 5 days, the amounts of PHA produced by activated sludge were low in the batch experiments, which only accounted for 3–16% of CDW. In contrast, the amounts of PHA significantly increased in the latter half of the operation of the reactor, accounting for 54% of CDW on the 10th day.
T-RFLP profiles of activated sludge bacterial community in an SBR
During domestication, it is useful to monitor the changes of activated sludge community over time. For this purpose, T-RFLP was applied for analyzing community 16S rRNA gene. The community T-RFLP profiles at different time points are shown in Fig. S3. On the 1th, 3th, 5th, 7th and 9th day the total numbers of T-RFs were 24, 17, 18, 11 and 13 T-RFs, respectively (Fig. 1). Seven T-RFs – 78, 86, 93, 99, 202, 367 and 558 bp – were common in all samples. Among them, two T-RFs, 99 and 558 bp, showed a higher relative abundance than the other T-RFs and a regular change in the relative abundance was also observed for the two T-RFs. Five unique T-RFs – 108, 192, 214, 226 and 277 bp – were found on the 1st day. A T-RF of 75 bp was found on the 5th day, and a T-RF of 563 bp on the 7th day.
In this study, the T-RFLP profiles at different domestication stages were different from each other, indicating that the structure of the bacterial community changed significantly during domestication. Moreover, the types of bacterial species (i.e. the phylogenetic diversity) decreased with the prolonged domestication, as judged by the reduction in the number of T-RFs. These results from T-RFLP suggest that some specific microbes such as PHA-accumulating bacteria may be enriched in activated sludge by domestication.
DGGE profiles of activated sludge bacterial community in an SBR
DGGE analysis of amplified 16S rRNA gene permits investigation of the spatial and temporal changes of the population in the environment, and it allows identification of the predominant species in a community. To better define the composition of the enriched mixed cultures, bacterial communities were analyzed by PCR–DGGE. The 16S rRNA V6–V8 variable regions were PCR-amplified from community DNA and further analyzed by DGGE. Bands 1–10 on the DGGE gel (Fig. 2) were sequenced. The closest matches for the resulting sequences were determined by a blast search at GenBank (Table 1). The sequence extracted from band 2 showed the highest identity with the genus Thauera. Previous studies have shown that the genus Thauera was the dominant microbial population reasonably responsible for PHASCL accumulation in activated sludge enriched by periodic feeding with a mixture of organic acids (Dionisi et al., 2005, 2006). Two DGGE bands (bands 7 and 8) appeared at days 8, 9 and 10 (Fig. 2). The sequences extracted from the two bands showed the highest similarity with the genera Pseudomonas and Acinetobacter, respectively (Table 1). It is well known that many species from the genera Pseudomonas and Acinetobacter are typical PHA producers. In particular, many Pseudomonas species possess the Class II PHA synthases and can produce a wide variety of PHAMCL (Solaiman et al., 2000; Zhang et al., 2001). Two specific genera, Dechloromonas and Nitrosomonas, which significantly correlate with the functions and performance of wastewater treatment, were also identified by DGGE analysis (bands 9 and 10). Several differential bands (bands 3, 4 and 5) in DGGE profiles were not classified into any genera.
Table 1. Sequence similarities of excised DGGE bands shown in Fig. 2
Phylogenetic analysis of PHA synthase genes in activated sludge
In general, these PHA synthase genes possess high G+C contents. DNA templates with a high G+C content usually hamper PCR amplification; the reagents formamide, glycerol, DMSO and betaine are often used as PCR additives to improve the PCR amplification of GC-rich DNA sequences (Dieffenbach & Dveksler, 1995). In this study, DMSO was added to the PCR reaction mixture. PCR with the CF1-CR4 primer set gave products with sizes around 500 bp. The sizes of the PCR products were as expected (Sheu et al., 2000). The amplicons from the day 9 sample were cloned and sequenced. In total, 80 phaC genes were obtained. A phylogenetic tree of the phaC genes was constructed based on their nucleotide sequences (Fig. 3). Clones with similarities higher than 97% were classified as the same operational taxonomic units (OTU). As shown in Fig. 3, of the 80 phaC genes found, 62 were classified into 13 OTUs (OTU 1–13). Of the 13 OTUs defined after the phylogenetic analysis, 12 were closest to the Class I PHA synthase, and the remaining one (OUT 12) was affiliated to the Class II PHA synthase. Our results suggested that Class I PHA synthase is the dominant type of PHA synthase in activated sludge, which is consistent with previous reports on the investigation of the diversity of the phaC genes in activated sludge (Michinaka et al., 2007; Ciesielski et al., 2008). Moreover, the two main types of PHA produced by activated sludge are PHB and PHBV (Salehizadeh & Van Loosdrecht, 2004; Serafim et al., 2008), which indirectly provides evidence for the predominance of Class I PHA synthase in activated sludge.
In a review summarized by Rehm (2003), 59 PHA synthases from various bacteria were extensively compared, showing that these enzymes exhibit strong similarities. Six conserved blocks, which include a putative lipase box that is the catalytic active site of PHA synthases, were identified in the amino acid sequences of the PHA synthases. After the multiple alignments of the amino acid sequences, of the total of 80 PHA synthases obtained, 72 were found to have a putative lipase box (GXCXG; data not shown). Among 80 PHA synthases obtained, 76 were closest to the Class I PHA synthase, and the remaining four were closer to the Class II PHA synthase (Table S1). Class I PHA synthase is represented by that of Ralstonia eutropha and utilizes hydroxyalkanoate unit comprising 3–5 carbon atoms, whereas Class II PHA synthase is represented by that of P. aeruginosa and utilizes a hydroxyalkanoate unit comprising 6–14 carbon atoms (Lee, 1996ab).
Production of PHAMCL containing 3HDD monomer by activated sludge
Since the first finding of PHB in 1926, more than 100 different monomer units have been identified as constituents of PHA in > 300 different microorganisms (Lee, 1996ab). Many studies have been conducted to evaluate the potential for PHA production by activated sludge using various external carbon substrates (Salehizadeh & Van Loosdrecht, 2004; Serafim et al., 2008). However, most studies focus on the synthesis of PHASCL by activated sludge, such as PHB, and its copolymer with 3HV (PHBV).
PHAMCL are considered to be much more suitable biomaterials for various applications based on their physical properties (Chen & Wu, 2005). Some Pseudomonas sp. strains, such as Pseudomonas oleovorans, Pseudomonas putida and Pseudomonas resinovorans, which are the most common PHAMCL-producing species, can accumulate PHAMCL from fatty acids (Ouyang et al., 2007). PHAMCL consisting of mainly 3-hydroxydodecanoate (3HDD) monomer showed improved thermal and mechanical properties over the conventional PHAMCL, which usually have low Tm and weak tensile strength (Liu et al., 2011). In this study, the production of PHAMCL containing 3HDD monomer by activated sludge was demonstrated for the first time. The PHA copolymer produced by activated sludge at day 10 was found to be composed of three types of monomers: 3-hydroxybutyrate (3HB), 3-hydroxyvalerate (3HV) and 3-hydroxydodecanoate (3HDD), by comparing their mass spectrogram with that of the standard in GC–MS analysis (Fig. 4). The molar percentages of 3HB, 3HV and 3HDD monomer were 62.41%, 34.17% and 3.42%, respectively, as determined by the relative abundance of each monomer in GC analysis. Previously, open mixed cultures enriched in glycogen-accumulating organisms with fermented sugar cane molasses were shown to produce PHAMCL containing 3-hydroxyhexanoate (3HHx) monomer (Bengtsson et al., 2010). Recently, the PHAMCL produced by activated sludge enriched by periodic feeding with nonanoic acid contained 3-hydroxynonanoate (3HN) and 3-hydroxyheptanoate (3HHp) monomers (Lee et al., 2011). So far, 3HDD is the longest carbon chain of monomer (C12) that has been found in the PHA produced by activated sludge.
The Class I PHA synthases listed in Table S1 were responsible for the catalysis of the polymerization of 3HB and 3HV monomers. One of the four Class II PHA synthases (clone 18) listed in Table S1 was shown to be associated with the ability of the activated sludge to produce PHAMCL containing 3HDD monomer. The PHA synthase from clone 18 showed the highest similarity (99%) with the PHA synthase 1 (PhaC 1) from Pseudomonas mendocina NK-01. The PHAMCL produced by NK-01 mainly consists of 3-hydroxyoctanoate (3HO) and 3-hydroxydecanoate (3HD) monomers (13). The other three Class II PHA synthases from clones 47, 55 and 80 showed 97–99% identity with those PHA synthases that had substrate preferences for hydroxyalkanoate monomers of 6–8 carbon atoms (Table S1). These results suggest that the PHA synthase from clone 18 may correlate with the polymerization of 3HDD monomer in the synthesis of PHAMCL containing 3HDD. In the phylogenetic analysis of the phaC genes, clone 18 was affiliated with OTU 12 (Fig. 3), and it is suggested that OTU 12 most probably corresponds to the formation of PHAMCL containing 3HDD.
The production of PHA containing 3HDD by activated sludge was more active in the last 3 days, and was much less in the first 7 days. The compositions of PHA produced in the batch experiments were determined by gas chromatograph analysis. The PHA produced by activated sludge consisted of 3HB and 3HV monomer in the initial 5 days, and no 3HDD monomer was detected. However, the 3HDD monomer was detected with a molar percentage of 0.12% and 0.38%, respectively, at days 6 and 7. Interestingly, the molar percentage of 3HDD monomer significantly increased in the last 3 days, and it reached 3.42% on the 10th day.
This observation most probably is due to the changes in the population of the PHA-producing bacteria. Our speculation was supported by the evidence below. A band (band 7 in DGGE profiles) appeared at days 8, 9 and 10, and the sequence extracted from band 7 showed the highest identity with the genus Pseudomonas. Many species from the genus Pseudomonas, such as P. putida KT2442, are well-known PHAMCL producers and can produce PHAMCL containing 3HDD monomer (Ouyang et al., 2007; Liu et al., 2011). The results from DGGE suggest that the Pseudomonas strain identified from band 7 may play a key role in the production of PHAMCL containing 3HDD.
In this study, the monomer compositions of PHAMCL were shown to be related to the types of PHA-producing bacteria and PHA synthase genes. Importantly, PHAMCL containing 3HDD monomer was produced by activated sludge. The present study provides valuable information for the directed synthesis of PHAMCL with different monomer compositions.
The authors gratefully acknowledge the financial support from the National Key Basic Research Program of China (‘973′ Program) 2012CB725204, National High Technology Research and Development Program of China (‘863′ Program) 2012AA021505, Natural Science Foundation of China grant Nos. 31070039, 31170030 and 51073081, and Nankai University Youth Teacher Foundation (No. 65012411).