Figure 1.
Profiles displaying the genome coverage by sequence reads.
A) Profiles showing the coverage of the plus (red) and minus (green) strand of the P. aeruginosa PA14 genome visualized using the Artemis genome browser [58]. The upper profile represents a pileup of the sequence reads counting the number of reads overlapping each position of the chromosome. In the lower sinister profile, only left end (5′-) positions of sequence reads were counted. The genome annotation is shown below the profiles with numbers indicating the position on the chromosome and white boxes indicating genes on the plus strand (upper row) and the minus strand (lower row), respectively. The genes within the displayed region are lasI (PA14_45940), rsaL (PA14_45950) and lasR (PA14_45960). B) Proposed model for the enrichment of transcripts ends: mRNA is isolated (1.), randomly fragmented (2.) and the first 30 nucleotides of the left ends of the fragments are sequenced on an Illumina GenomeAnalyzer (3.). Since the fragmentation of each mRNA produces identical 5′-ends and all other fragments are random, mapping leads to a stack of sequence reads at the mRNA 5′-end (4.) that can be used to annotate the TSS (5.).
Figure 2.
Gene expression under planktonic and biofilm growth conditions.
A) Venn diagram showing the genes that were differentially expressed (−2≤log2 fold change ≤2; P≤0.001) in planktonic stationary phase cultures (P-12) and biofilms growing for 24 or 48 h (B-24, B-48) as compared to late exponential phase planktonic cultures (P-4). B) Principle component analysis of absolute gene expression. The first principle component (PC 1) accounted for 58% and PC 2 for 25% of the total variation in the dataset. Symbols indicate two biological replicates of late exponential phase planktonic cultures (P-4, •), stationary phase planktonic cultures (P-12, ○), 24 h old biofilms (B-24, ▴) and 48 h old biofilms (B-48, ▵), respectively. C) Cluster analysis of the normalized gene expression for genes that were differentially regulated in P-12, B-24 or B-48 as compared to P-4. Clusters have been labeled ‘B’ for ‘biofilm specific’ (blue background), if gene expression was consistently different between the biofilm cultures (B-24, B-48) and planktonic cultures (P-4, P-12). A comprehensive list of absolute and differential gene expression for all genes included in this figure is provided in Table S2. A larger version of this heatmap including gene labels can be viewed in Figure S1.
Figure 3.
Comparison of differential regulation among different studies.
This qualitative ‘heatmap’ compares the differential expression of selected genes and gene clusters in B-48 with similar studies that have been published previously. In all cases, the expression of a biofilm culture was compared to a planktonic culture. The studies included in this analysis are Whiteley et al. [1], Waite et al. [21], Hentzer et al. [4] and Mikkelsen et al. [20]. The method abbreviations are CS – chemostat (continuous culture), 96-well – static 96-well plate cultures, agar – nitrocellulose filter on agar surface, tube – silicon tubes, array – DNA microarray analysis, seq – sequencing of cDNA.
Table 1.
Small RNAs that were differentially expressed in PA14 biofilms.
Figure 4.
Alternative transcriptional start sites of the PQS operon.
A) Model of the PQS biosynthetic operon (pqsABCDE) with the three alternative TSS indicated by arrows. The TSS are named by the next downstream gene with subscripts indicating the TSS position relative to the start of the associated gene. B) Pileup profile of the PQS operon depicting the total amount of read counts generated in this study that map to the PQS operon. The enriched TSS loci are indicated by black triangles.
Table 2.
Confirmation of predicted transcriptional start sites (TSS) by 5′-RACE.
Figure 5.
Detection of transcriptional start sites.
A) Transcriptional start sites (TSS) were categorized by their position relative to annotated genes as follows: promoter (P), up to 500 bp upstream of an annotated gene on the same strand, intragenic (I), within an annotated gene on the same strand, antisense (A), within an annotated gene on the opposite strand, orphan (O), neither ‘intragenic’ nor ‘promoter’, which means no association to any annotated gene. B) Venn diagram depicting the abundance of the different categories among 1024 TSS that were independently detected for at least two different conditions. C) Distribution of TSS in promoter regions of coding genes (class ‘P’), indicating the length of the untranslated regions (UTRs) at the 5′ end of mRNAs. The peak of the distribution is located around the −25 position. The dashed line marks the −10 nt position, which indicates the 43 leaderless transcripts (red bars) with 5′ UTRs of less than 10 nt length.