Identification of cyanobacterial non-coding RNAs by comparative genome analysis

Abstract

Background: Whole genome sequencing of marine cyanobacteria has revealed an unprecedented degree of genomic variation and streamlining. With a size of 1.66 megabase-pairs, Prochlorococcus sp. MED4 has the most compact of these genomes and it is enigmatic how the few identified regulatory proteins efficiently sustain the lifestyle of an ecologically successful marine microorganism. Small non-coding RNAs (ncRNAs) control a plethora of processes in eukaryotes as well as in bacteria; however, systematic searches for ncRNAs are still lacking for most eubacterial phyla outside the enterobacteria.

Results: Based on a computational prediction we show the presence of several ncRNAs (cyanobacterial functional RNA or Yfr) in several different cyanobacteria of the Prochlorococcus-Synechococcus lineage. Some ncRNA genes are present only in two or three of the four strains investigated, whereas the RNAs Yfr2 through Yfr5 are structurally highly related and are encoded by a rapidly evolving gene family as their genes exist in different copy numbers and at different sites in the four investigated genomes. One ncRNA, Yfr7, is present in at least seven other cyanobacteria. In addition, control elements for several ribosomal operons were predicted as well as riboswitches for thiamine pyrophosphate and cobalamin.

Conclusion: This is the first genome-wide and systematic screen for ncRNAs in cyanobacteria. Several ncRNAs were both computationally predicted and their presence was biochemically verified. These RNAs may have regulatory functions and each shows a distinct phylogenetic distribution. Our approach can be applied to any group of microorganisms for which more than one total genome sequence is available for comparative analysis.

Figure 1

Small RNAs in marine Cyanobacteria. About 10 μg of total RNA from Prochlorococcus strains MIT 9313 (MIT), SS120 (SS1) and MED4 (MED) and from Synechococcus sp. WH 8102 (WH8) was analyzed by staining a 10% polyacrylamide gel with ethidium bromide (center) and by Northern blot hybridization with DNA-oligonucleotides directed against known RNA molecules such as scRNA (ffs gene product), the separate 5' and 3' ends of tmRNA and, as controls, tRNASerin and 5S rRNA. Two distinct precursors of the 5S rRNA were detected. Selected bands have been labeled by arrows in the hybridization and in the gel picture and their sizes (nt, nucleotides) are indicated.

Figure 2

Pipeline for comparative prediction of non-coding RNAs. (a) Intergenic sequences (IGRs) longer than 49 base-pairs were gathered from four Prochlorococcus and Synechococcus genomes and locally aligned using BLASTN. An overview of the intergenic sequences is given in Additional data file 2 (Table S4). Because of the initial asymmetric local alignment using BLASTN (see Figure 2b for a summary of significant BLASTN hits between the strains Prochlorococcus MED4 (MED), MIT 9313 (MIT), SS120 (SS) and Synechococcus WH 8102 (WH)), all candidate sequences were reverse-complemented. Redundancy in this data set was reduced by unifying those hits from each genome that showed a reciprocal overlap of 85% or greater. This candidate set was used as both query and subject in another local alignment step (BLASTN considering only the query strand as possible subject strand). Sequences that directly produced a significant blast hit (E-value ≤ 10^-10), or were connected by a chain of such hits, were gathered into clusters ('single-linkage clustering'). Both genome strands were screened; thus, the pipeline produced 310 pairs of clusters in both forward and reverse complementary orientation. After an additional unification step of overlapping sequences within each cluster, the resulting clusters and their complement clusters were scored using ALIFOLDZ [33]. (b) The number of BLASTN high-scoring segment pairs for each query and subject combination of intergenic regions is given for a BLASTN E-value cut-off of 10^-5 and after import of high-scoring segment pairs with an E-value of 10^-10 or lower (in parentheses). MIT, Prochlorococcus strain MIT 9313; SS, Prochlorococcus strain SS120; WH, Synechococcus sp. WH 8102, MED, Prochlorococcus strain MED4.

Figure 3

Experimental screen for the presence of an RNA-coding gene in the guaB-trxA intergenic region. (a) Sequence alignment of the guaB-trxA (guaB: sequence not shown, located upstream of yfr1) intergenic region visualises the conserved yfr1 gene labeled by the bar above the alignment and its transcriptional initiation site in three of the analyzed strains (MED, MED4; MIT, Prochlorococcus strain MIT 9313; WH8, Synechococcus sp. WH 8102) but not in Prochlorococcus strain SS120 (SS1). Transcriptional initiation sites (TIS) and the deduced -10 elements are indicated. (b) Northern blots show a signal for Yfr1 at a size of 54, 56 and 57 nucleotides (nt) for MED4, WH 8102 and MIT 9313, respectively. No signal with RNA from SS120 confirms the absence of this gene in this strain, as was predicted from the sequence data. (c) Predicted secondary structures of Yfr1 in MED4, MIT 9313 and WH 8102 by MFOLD [59].

Figure 4

Test of transcript accumulation of Yfr1-7 from MED4 (MED) under different conditions. The left side shows the Northern hybridizations for which the following conditions were used: nutrient depletion (phosphate (P-), nitrogen (N-), iron (Fe-)); blue light for three hours (3 h); controls under blue (Blue), white (White) and no light (Dark); oxidative stress mediated by the application of 3-(3,4-dichlorophenyl)-1,1-N-N'-dimethylurea (DCMU); low (15°C) and high (30°C) temperatures; and high light intensity (50 μE). For comparison, 5S rRNA was hybridized as an internal standard and the mRNA of gene PMM3822n which, with a length of approximately 250 nucleotides, was taken as an example for a small mRNA. Additional controls by quantitative RT-PCR for the genes isiB (Fe), glnA (N), pstS (P) and hli8 (high light) [data not shown] were carried out to confirm the effects of nutrient depletion or high light. The amounts of these mRNAs were enhanced by a factor of 79.7 (isiB), 5.8 (glnA), 2.8 (hli8) and 4.0 (pstS) under the respective treatment compared to standard conditions (data not shown). Yfr6 shows an inconstant signal; for example, at cold, blue/white light, N-, Yfr2 to Yfr5 were hybridized with the consensus oligonucleotide y_gen (Figure 5). The band intensities were quantified and normalized to the amount of 5S rRNA as an internal standard (right).

Figure 5

Comparison of Yfr2, Yfr3, Yfr4 and Yfr5 from MED4. (a) Sequence comparison of the yfr2 through yfr5 coding regions of MED4. Transcriptional initiation sites (TIS) and the deduced -10 elements are indicated. The location of specific oligonucleotide probes y2aM, y3aM, y4aM and y5aM used in Figure 5b and in 5' RACE and of the y_gen consensus probe used in Figure 4 is indicated by the lines with black diamonds on the ends on top of the alignment. (b) Signals for the four individual non-coding RNAs (ncRNAs) were detected in Northern blots using probes y2aM, y3aM, y4aM and y5aM. These probes have a minimum of five mismatches to their non-target ncRNAs, making cross-hybridizations impossible. The numbers indicate transcript lengths in nucleotides. (c) Prediction of secondary structure of MED4 Yfr2 by MFOLD [59].

Figure 6

Characterization of a gene encoding Yfr6. (a) Sequence alignment of the region containing the transcriptional initiation site (TIS) and the first 97 transcribed nucleotides of Yfr6 from MED4 and SS120. The alignment begins with the TATA element (in red) preceding the mapped first transcribed guanidine (labelled by an arrow). (b) In Northern blots, a signal for the predicted Yfr6 was detected for 244 and 239 nucleotides in total RNA from MED4 (MED) and SS120 (SS1), respectively, but not in RNA from WH 8102 (WH8) and MIT 9313 (MIT). (c) Comparison of Yfr6 secondary structures using ConStruct version 3.0a [60] The base pairing probability is colour-coded from light yellow (low) to red (high). Missing positions are indicated by a dash. The predicted RNA structures were obtained by RNAfold at 24°C. Both sequences were equally weighted (1.0). The consensus was calculated based on the predicted optimum structures. Default parameters were used for all other options.

Figure 7

Characterization of Yfr7. (a) Sequence alignment of Yfr7 from four marine cyanobacteria. The 5' end transcriptional initiation site (TIS) was mapped for Yfr7 from MED4 and SS120. (b) In Northern blots, a signal for the predicted Yfr7 was detected with RNA from all four strains and seven additional strains from the marine cyanobacterial radiation: Prochlorococcus NATL2A (N2), MED4 (MED), SS120 (SS1), MIT 9313 (MIT), MIT 9312 (9312), MIT 9211 (9211), MIT 9215 (9215) and Synechococcus WH 8102 (WH8), WH 7803 (7803), WH 8020 (8020), RS9906 (9906). The high-light-adapted Prochlorococcus strains are labelled in red, low-light-adapted strains in blue, and Synechococcus strains are colour-coded in green; M indicates the marker lane. Numbers indicate lengths of RNA markers in nucleotides. (c) Prediction of secondary structure of the Synechococcus WH 8102 ncRNA Yfr7.

Figure 8

A putative gene encoding the RNA chaperone Hfq can be predicted in two of the four marine cyanobacteria investigated here. (a) The dapF-leuS intergenic region in Synechococcus WH 8102 (WH8) and Prochlorococcus MIT 9313 (MIT) is, at 298 and 297 nucleotides, respectively, relatively long and contains a short reading frame for a putative hfq gene. In Prochlorococcus SS120 (SS1) and MED4 (MED), this region is only 123 and 108 nucleotides, respectively. (b) Sequence comparison of putative Hfq proteins from the three cyanobacteria Synechocystis PCC 6803 (ssr3341 gene), Synechococcus WH 8102 (WH8) and Prochlorococcus MIT 9313 (MIT). Hydrophobic residues within the Sm1 and Sm2 motifs [29] are indicated by an H.