Figure 1.
This pond is persistently inhabited by the investigated Polynucleobacter lineage (F10 lineage). The images show the pond in autumn (left) and winter (right), when the pond is hidden below a thick ice and snow cover. The latter situation lasts for up to six months.
Figure 2.
Annual cycles of water temperature and bacterial parameters of Pond-1.
Water temperatures (measured in depths of about 30 cm; upper graph), abundance of the entire P.n. ssp. asymbioticus assemblage (PnecC) determined by FISH and the F10 lineage based on qPCR measurements (middle graph), and relative abundances of this two taxa (bottom graph). The gray areas indicate periods of ice-cover. Note that some samples were not investigated by qPCR.
Figure 3.
Diurnal cycles of environmental parameters in three different depths of Pond-1.
Water temperature, oxygen concentration, and conductivity were measured in August 2010. The days before the recording of the diurnal cycle were characterized by cooler dull weather, while the weather conditions on the first day of measurement improved and the second day was characterized by sunny and calm weather. The arrow indicates the point in time of vertical mixing. Note that maximum temperature and oxygen values, as well as minimal conductivity values measured during the diurnal cycle were recorded in late afternoon. Since the routine sampling of the pond (Fig. 2) always took place during late-morning hours, the widths of annual variation of environmental parameters were most likely underestimated.
Figure 4.
Diversity of P. necessarius ssp. asymbioticus as represented by a culture collection of 176 strains.
The graphs show collector's curves depicting genotype numbers for the global non-redundant culture collection of P.n. ssp. asymbioticus strains (broad lines) after randomization of sampling order, as well as further extrapolations for continued collection of strains (thin lines). This non-redundant culture collection excludes all but one strain of the same genotype (ITS and glnA marker considered) isolated from the same habitat. Therefore, the collection includes only one (i.e. QLW-P1DMWA-1) of the twelve F10 lineage strains isolated from Pond-1 (compare Fig. 5). The collection consists currently of 176 strains isolated from various freshwater systems located all over the world. The sequencing of 16S rRNA genes of newly isolated strains was stopped after 53 strains. The length of the sequenced stretches of the markers 16S rRNA gene, 16S–23S ITS, and glnA were 1500 bp, 500 bp, and 600 bp, respectively. Note that genotype or species-richness predictions based exclusively on observed data usually underestimated the true richness. Randomization of sampling order was performed using the software EstimateS [75].
Figure 5.
Microdiversity of the F10 lineage population of Pond-1.
The genetic diversity of the twelve F10 lineage strains isolated from Pond-1 over a period of four years (2003–2007) is illustrated. Left panel. NJ tree based on concatenated nucleotide sequences of twelve loci (Table S4) with a total length of 9031 alignment positions. Middle panel. Only three out of 13 sequenced loci showed sequence polymorphism. The icd2 and Pnuc_1095 genes were present with two alleles differing in the sequenced parts in total in eight and seven positions (polymorphic positions), respectively. Numbers depict the total numbers of polymorphic sites (first number) and the number of non-synonymous sites (second number) among the polymorphic sites. The locus Pnuc_1240 of strain P1-Kol5 differed from all other sequenced strains in the presence of an insertion element (IS, insertion sequence). Sequences of Pnuc_1240 were not considered in the calculation of the presented phylogenetic tree. Right panel. Results from genetic fingerprinting with the independent methods RAPD (Ziemke A), ERIC, and REP-PCR. Each method resulted in different fingerprints, but each method revealed basically only two types of fingerprints (G1a and G1b). The REP-PCR fingerprint G1b* differed only weakly from G1b. Colors green and red indicate that the respective strains share or do not share, respectively, a particular trait with the genome-sequenced strain QLW-P1DMWA-1.
Figure 6.
Growth efficiencies of strain QLW-P1DMWA-1 on various tested substances.
Growth efficiencies (GE) of the genome-sequenced strain were determined for 89 substances by batch culture experiments. GE describes the percentage of carbon contained in the respective tested substance transferred by the strain to carbon in its biomass. Only ten substrates resulted in GE>5%. These substances are listed with decreasing efficiencies (top to bottom).
Figure 7.
Comparison of SOX gene clusters.
The SOX gene clusters of strain QLW-P1DMWA-1 (Burkholderiaceae, Betaproteobacteria), Bradyrhizobium japonicum (Alphaproteobacteria) and Cupriavidus metallidurans (Burkholderiaceae, Betaproteobacteria) are compared. The three clusters of the Polynucleobacter strain are represented by the genes Pnuc_1486, Pnuc_1487, and Pnuc_0799 - Pnuc_0809, and Pnuc_1154 - Pnuc_1157. Start- and end-positions of the clusters in the respective genome sequences are depicted. Shaded areas indicate if Polynucleobacter genes are closer related to Cupriavidus or Bradyrhizobium genes.
Figure 8.
Phylogenetic position of the genome-sequenced Polynucleobacter strain.
Reconstruction of the phylogenetic position of the genome-sequenced strain P.n. ssp. asymbioticus QLW-P1DMWA-1 and genome characteristics of the strains used for the phylogenetic analysis. The presented tree is based on concatenated alignments of amino acid sequences of proteins encoded by eight housekeeping genes and was constructed using the NJ method. Letters A to H refer to bootstrap values obtained for the respective nodes by three different treeing methods in analyses of nucleotide and amino acid sequence data sets. The bootstrap values are presented in Table S7.
Figure 9.
Variability of genomic traits among genome-sequenced members of the family Burkholderiaceae.
The depicted traits are the total number of ORFs, G+C content of the genomes, percentage of CDS encoding signal peptides, transmembrane domains, genes assigned to Enzyme Commission (EC) number classes, proteins assigned to protein families (Pfam), proteins assigned to KEGG categories, and proteins involved in signal transduction. For most traits, 78 genomes of strains affiliated with the family were analyzed using the IMG system [24]. The sole exception is the analysis of signal transduction genes, which was performed using the MIST2 database [69], which provided data on 37 strains affiliated with the Burkholderiaceae. For all depicted parameters, all available genomes of family members were considered but not the genome of the endosymbiotic Polynucleobacter strain STIR1. Data of strain QLW-P1DMWA-1 are indicated as red dots.
Figure 10.
Gene number of bacterial genomes versus number of encoded signal transduction genes.
The plot depicts the total number of annotated signal transduction genes in particular bacterial genomes versus the total number of genes of the respective genomes. Data on Archaea are not shown, and multiple genomes per species or species-like taxa were excluded. Strain P. necessarius QLW-P1DMWA-1 is highlighted as a yellow dot. (A) Entire data set consisting of 904 analyzed bacterial genomes. Red dots indicate bacteria obligately associated as symbionts or pathogens with a host; blue dots indicate free-living or facultative host-associated bacteria. (B) Plot of genomes of bacteria classified as free-living or facultative host-associated (shown in (A) as blue dots) with ≤100 signal transduction genes. All blue dots represent bacteria either facultatively associated with hosts (e.g., dwelling in the oral cavity or in the intestinal tract) or dwelling as free-living organisms in extreme environments (permanently anoxic, low or high temperature, low pH). The yellow (strain QLW-P1DMWA-1) and the green dots represent the only exclusively free-living taxa inhabiting non-extreme systems. Almost all of them are phototrophic (Synechococcus and Prochlorococcus) or predominantly heterotrophic taxa (Pelagibacter and Puniceispirillum) dwelling as planktonic bacteria in marine systems. The signal transduction gene data were obtained from the database MIST2 [69].
Table 1.
Basic features characterizing the genome of P. necessarius ssp. asymbioticus strain QLW-P1DMWA-1.
Figure 11.
Microcosm experiment on the influence of light on growth of bacteria from Pond-1.
Growth of a natural bacterial community originating from Pond-1 was investigated in three different light treatments (UV+VIS, VIS only, and no light). Water from Pond-1 was filtered through membrane filters with a nominal pore size of 0.8 µm (exclusion of potential predators, algae, and other larger organisms) mixed with two volumes of 0.2 µm-filtered pond water (almost free of bacteria) and incubated under in situ conditions in floating bottles covered with foil blocking all light (i.e., dark treatment), with foil blocking (>90%) UV light resulting in VIS light only (i.e., VIS only treatment), or in untreated bottles (penetration of >70% UV light), i.e. UV treatment. The bacterial community consisted initially of 48% P.n. ssp. asymbioticus and 52% other bacteria. The treatments were incubated under in situ conditions in the pond for a period of 34 hours. (A) Linear plots of results, (B) Linear regressions of semi-logarithmic plots of the results obtained for Polynucleobacter. R2 values of the regression lines were 0.999, 0.996, and 0.989 for the dark, VIS, and UV treatments, respectively. Filled circles, P.n. ssp asymbioticus; filled triangles, other bacteria; red, with UV light; blue, VIS light only, black, no light. All shown data, apart of the data of the UV treatment (red symbols and lines), represent average data and SD (bars) from three replicates. One of three parallels of the UV treatment showed no growth at all. Presentation of average data was therefore omitted and only particular data for each of the two replicates with growth are shown.
Figure 12.
Experiment on potential influence of algal exudates on growth of P.n. ssp. asymbioticus bacteria.
The influence of the presence of growing algae populations on the development of P.n. ssp. asymbioticus bacteria was investigated in batch culture experiments. Grazer-free (0.8 µm filtered) water from Pond-1 was mixed with different algal cultures representing species typically abundant in Pond-1. The treatments were incubated with continuous illumination for 15 days at a constant temperature of 16°C. The graph shows mean values of relative abundance of P.n. ssp. asymbioticus bacteria and standard deviations from three replicates. The initial total bacterial numbers in all treatments that had received algae was 4.7×106 cells mL−1, and P.n. ssp. asymbioticus bacteria comprised 23% of bacterial cells. Cell numbers of the three different algal species developed differently over the course of the experiment, but all three showed growth over periods of six to 15 days.
Figure 13.
Conceptual model on carbon fluxes utilized by the F10 lineage population in Pond-1.
The model suggests origin and transformation of major carbon sources utilized by the F10 lineage population inhabiting Pond-1. Humic substances (HS) leached from soils (1) and floating mats (2) (Schwingrasen, formed mainly of Sphagnum and Carex) are transported by percolating water to the pond. P.n. ssp. asymbioticus bacteria affiliated with the F10 lineage mainly utilize low-molecular-weight (LMW) photooxidation products of HS (3). Direct utilization of aliphatic residues of HS by enzymatic cleavage is also expected to take place (4). Another potential substrate source could be fermentation products released by bioturbation and other processes from the sediment (5) to the water column of the pond.