METAGENOMIC ARRAY ANALYSIS OF AN ENHANCED BIOLOGICAL PHOSPHORUS REMOVAL SLUDGE ENRICHED WITH ACCUMULIBACTER

INTRODUCTION Previously, metagenomic analysis was performed on our lab-scale acetate-feed sequencing batch reactor highly enriched with Accumulibacter (Garcia Martin et al. 2006). Nearly complete genome sequence from the dominant Accumulibacter strain was obtained, many functional genes were identified, and metabolic pathways were reconstructed. The objective of this study was to investigate which biochemical pathways are expressed and regulated during enhanced biological phosphorus removal (EBPR) by detecting transcripts using a metagenomic array. MATERIALS AND METHODS Medium density oligonucleotide Combimatrix microarrays were generated with probes targeting most metabolic genes from Accumulibacter. Samples were collected from the same reactor, from which the samples for metagenomic analysis of Accumulibacter had been previously taken. RNA was extracted using the RNeasy mini kit (Qiagen) with bead beating and on column DNase digestion. The mRNA was enriched by removing most of 16S and 23S rRNA and further polyA tailed using mRNA-ONLY™ Prokaryotic mRNA Poly(A)-Tailing Kit (EPICENTER). Subsequently, cDNA was synthesized and further in vitro transcribed, with 5-(3-aminoallyl)-UTP incorporated into the resulting antisense RNA, using TargetAmp™ 1-Round aRNA Amplification Kit (EPICENTER). One of two fluorescent dyes (Alexa Fluor 555 or 647) was coupled to the modified UTP. Purified dyecoupled antisense mRNA was hybridized to blocked arrays as recommended by the manufacturer (Combimatrix). The spots were imaged using a Molecular Devices Axon4000B scanner and data was extracted using GenePix Pro 6.0 software. When two fluorescent dyes were applied on the same microarray, Lowess normalization was performed to remove signal intensity-dependent variation. RESULTS AND DISCUSSION A dye swap experiment was first conducted to determine if there was any bias associated with labeling. The antisense mRNA generated from the same sample was subjected to Alexa Fluor 555 and 647 labeling separately and then applied on the same microarray. After Lowess normalization, the signal intensity-dependent variation was removed, indicated by the M-A plot in Figure 1. The nearly 1:1 ratio of normalized signal intensities from both dyes indicated no significant bias associated with labeling procedure (Figure 2).To detect genes that are expressed during an EBPR cycle, samples were collected from anaerobic and aerobic phases, combined and labeled with Alexa Fluor 647. The expression of some genes in glycolysis/glucogenesis pathways and the tricarboxylic acid cycle were detected. Consistent with the proposed carbon metabolism model for EBPR, Acetyl-CoA acetyltransferase gene that is associated with the formation and degradation of polyhydroxybutyrate (PHB) were highly expressed. In Accumulibacter, the gene encoding guanosine 3′,5′-bis-diphosphate (ppGpp) synthase (relA), is located in a cluster with polyphosphate kinase (ppk) and exopolyphosphatase (ppx) (Garcia Martin et al. 2006). The elevated levels of ppGpp were shown to inhibit ppx activity and thus favor polyP formation in model organisms (Kornberg et al, 1999). Our preliminary data suggest that relA and ppk were expressed, implying that similar regulation of polyP metabolism may be operating in Accumulibacter. Interestingly, genes associated with defense mechanism against bacterial virus were also highly expressed, indicating the persistent viral predation pressure in the reactor.To determine genes that are differentially expressed between aerobic and anaerobic phases, antisense mRNA generated from aerobic and anaerobic samples were labeled with Alexa Fluor 555 and 647 respectively. The cutoff of 3-fold change (corresponding to M value of ±1.585) was used to indicate differential expression. Preliminary results showed that expression levels of most genes, especially those with high expression levels, were not significantly different between the aerobic and anaerobic phases (Figure 3). IMPLICATIONS AND ON GOING WORK From this study, we confirmed some genes that were previously hypothesized to be involved in EBPR metabolism were expressed. Currently, we are investigating Accumulibacter gene expression under disturbed and adverse conditions, aiming at identifying the key gene(s) whose expression level is closely related to EBPR performance, so that reverse transcription quantitative PCR can be further designed to monitor transcription level of the key gene(s) and provide a more complete understanding of the genetic and biochemical mechanisms responsible for EBPR.