A sporulation-specific, sigF-dependent protein, SspA, affects septum positioning in Streptomyces coelicolor

Abstract

The RNA polymerase sigma factor SigF controls late development during sporulation in the filamentous bacterium Streptomyces coelicolor. The only known SigF-dependent gene identified so far, SCO5321, is found in the biosynthetic cluster encoding spore pigment synthesis. Here we identify the first direct target for SigF, the gene sspA, encoding a sporulation-specific protein. Bioinformatic analysis suggests that SspA is a secreted lipoprotein with two PepSY signature domains. The sspA deletion mutant exhibits irregular sporulation septation and altered spore shape, suggesting that SspA plays a role in septum formation and spore maturation. The fluorescent translational fusion protein SspA-mCherry localized first to septum sites, then subsequently around the surface of the spores. Both SspA protein and sspA transcription are absent from the sigF null mutant. Moreover, in vitro transcription assay confirmed that RNA polymerase holoenzyme containing SigF is sufficient for initiation of transcription from a single sspA promoter. In addition, in vivo and in vitro experiments showed that sspA is a direct target of BldD, which functions to repress sporulation genes, including whiG, ftsZ and ssgB, during vegetative growth, co-ordinating their expression during sporulation septation.

Fig 1

2D-PAGE analysis of spore proteins identifies SspA (marked with an arrow) in the wild-type spores but not in the spores of the sigF mutant. Details of sample preparation and 2D gel-electrophoresis are described in Experimental procedures.

Fig 2

The sspA mutant generates irregularly sized spores. A. Wild-type M145 and the sspA mutant strains were grown in minimal medium for 72 h and the spores were viewed using phase-contrast microscopy. B. S. coelicolor wild-type M145 and the sspA mutant K55 were grown in SFM medium for 72 h and the spore morphology was assessed using scanning electron microscopy. Arrows mark spores longer than the wild-type average. The scale bar represents 2 μm. C. Histogram of the length and width of mature spores from S. coelicolor M145 (black) and the sspA mutant, K55 (white) grown in SFM medium for 72 h and viewed using phase-contrast microscopy. Measurements of ∼ 500 spores were analysed using the excel package. D. Transmission electron microscopy of the wild-type and sspA mutant spores, which were cultivated as in (B). Scale bars represent 1 μm.

Fig 3

Septum formation is irregular in the sspA mutant. A. Early spore chains of the wild-type M145 and the sspA mutant, K55, were viewed using laser-scanning confocal microscopy after staining the cell wall with WGA-Alexa488 (green; top panel) and the chromosomes with propidium iodide (PI; red; bottom panel) after growth for 48 h in SFM medium alongside microscope coverslips to allow viewing of aerial development and sporulation. Yellow arrows mark aberrant septum formation in the sspA mutant. B. Histogram of the length and width of pre-spores within spore chains of the wild-type (black) and sspA mutant (white) strains. Samples were generated and viewed as in (A). Measurements of ∼ 550 spore compartments were analysed using the excel package.

Fig 4

Transcription of sspA is sigF dependent. A. S1 nuclease analysis was performed using RNA samples collected at distinct developmental stages of S. coelicolor wild-type M145 (left) or sigF mutant (right) strains. Three different probes specific to sspA, sigF and sigNP2 were used, this latter as a control to assess RNA integrity in the samples. sp indicated sporulating samples as was assessed by microscopy. B. Forty micrograms of RNA from 72-h-old wild-type (M145) samples, grown on SFM medium, were used together with a radioactively labelled probe containing the presumed transcriptional start for sspA together with a sequencing ladder to identify the exact transcriptional start point. C. In vitro run-off transcription was performed by RNA polymerase core enzymes (lanes 1 and 3) and RNA polymerase holoenzyme containing His-SigF (lanes 2 and 4) using the templates ctc (lanes 1 and 2) or sspA (lanes 3 and 4). The 155 nt ctc-specific transcript and the 54 nt long sspA transcripts are marked with arrows. The sizes of the marker DNA fragments generated using pUC19 digested with Sau3AI (M) are shown in basepairs. D. Sequence of the probe used for S1 nuclease analysis is shown. The starts of horizontal single arrows mark the 5′ ends of the oligonucleotides used. An asterisk marks the transcription initiation site that coincides with the translational start site. Translated sequences are italicized. The double headed arrows mark the regions protected by cell extracts (below the sequence; see also Fig. 6) or by purified BldD (above the sequence; see also Fig. 7) in DNase I footprinting assays. Grey highlighting indicates the most highly conserved nucleotides of the consensus BldD target sequence (den Hengst et al., 2010). E. Putative promoter sequence of sspA. The promoter of sspA is compared to the SigF-dependent promoter, whiEP2 together with the predicted consensus for SigB of S. coelicolor (Lee et al., 2004) and the established consensus for SigB of B. subtilis (Petersohn et al., 1999). Asterisks mark the transcription start points. The −10 and −35 promoter sequences are underlined and the BldD box (den Hengst et al., 2010) is highlighted in grey.

Fig 5

Monitoring SspA expression during sporulation. A. S. coelicolor M145/pK37, carrying sspAP–egfp, where Egfp is expressed from the sspAP promoter, was grown for ∼ 72 h in SFM medium alongside a coverslip and the samples were viewed by laser-scanning confocal microscopy. The green fluorescence image is above the overlaid image with the bright-field view. Scale bar represents 2 μm.B–D. S. coelicolor M145/pK39, expressing SspA–mCherry from pK39 integrated at the native chromosomal location of sspA, was grown for 72 h (B and D) or 44 h (C) in SFM medium and the samples were viewed by laser-scanning confocal microscopy. B. The red fluorescence image (left) is shown together with the overlaid image with the bright-field view (right). C. The bright-field view (left), the red fluorescence image (middle) and the overlaid image (right) are shown. Arrows mark the positions of developing septa. D. Samples were stained using WGA-Alexa488 (green, middle). The red fluorescence image (left) is shown together with the overlaid image (right). Scale bars represent 2 μm.

Fig 6

The sspA promoter is targeted by a putative repressor. A. Electrophoretic mobility shift assay. A radiolabelled DNA fragment containing the sspA promoter (lane 1) was incubated with increasing amounts (5 and 10 μg of protein) of cell extract from samples collected at an early developmental stage comprising of vegetative mycelium of wild-type S. coelicolor (lanes 2–3), or of the bldD mutant (lanes 5–6). The specificity of the protein–DNA interaction was confirmed by using 10 μg of protein extracts from the wild-type strain (as in lane 3) with the addition of a 50-fold excess of cold probe (lane 4). B. DNase I footprinting of the sspA promoter. The probe (as in A) was treated with increasing amount of DNase I either in the absence (lanes 1–2) or in the presence of 20 μg of protein extracts from wild-type S. coelicolor extracts (lanes 2–4). The generated fragments were analysed next to a sequencing ladder. The protected region is marked by a bar.

Fig 7

BldD targets the sspA promoter. A. BldD ChIP-chip data for the 8 kb region spanning the sspA locus in wild-type S. coelicolor (black circles) and the bldD mutant (grey squares). DNA obtained from immunoprecipitation of BldD was labelled with Cy3 and hybridized to DNA microarrays together with a total DNA control that was labelled with Cy5 (den Hengst et al., 2010). Data generated by den Hengst were retrieved from the Gene Expression Omnibus (GSE23401) and were plotted as Cy3/Cy5 ratios (y-axis), as a function of chromosome location around sspA (x-axis). B. DNase I footprinting analysis of BldD binding to the promoter region of sspA. 5′ end-labelled probes were incubated in the presence of 0, 0.5, 1.0 or 2.0 μM BldD and subjected to DNase I footprinting analysis as described in den Hengst et al. (2010). Footprints are flanked by Maxam and Gilbert sequence ladders (AG). Protected regions are marked by bars.