NsiR3, a nitrogen stress-inducible small RNA, regulates proline oxidase expression in the cyanobacterium Nostoc sp. PCC 7120

NsiR3, a nitrogen stress-inducible small RNA, regulates proline oxidase expression in the cyanobacterium Nostoc sp. PCC 7120 Isidro ´ Alvarez-Escribano 1 , Manuel Brenes- ´ Alvarez 1 , Elvira Olmedo-Verd 1 , Jens Georg 2 , Wolfgang R. Hess 2 , Agustı´n Vioque 1 and Alicia M. Muro-Pastor 1 1 Instituto de Bioquı´mica Vegetal y Fotosı´ntesis, Consejo Superior de Investigaciones Cientı´ficas and Universidad de Sevilla, Sevilla, Spain 2 Genetics and Experimental Bioinformatics, Faculty of Biology, University of Freiburg, Freiburg, Germany Keywords heterocyst; NtcA; post-transcriptional regulation; PutA; regulatory RNA Correspondence A. Vioque, Instituto de Bioquı´mica Vegetal y Fotosı´ntesis, Consejo Superior de Investigaciones Cientı´ficas and Universidad de Sevilla, Avda. Am´ erico Vespucio 49, Sevilla 41092, Spain Tel: +34-954489519 E-mail: [email protected] (Received 14 May 2020, revised 17 July 2020, accepted 10 August 2020) doi:10.1111/febs.15516 NsiR3 (nitrogen stress-inducible RNA 3) is a small noncoding RNA strongly conserved in heterocyst-forming cyanobacteria. In Nostoc sp. PCC 7120, transcription of NsiR3 is induced by nitrogen starvation and depends on the global nitrogen regulator NtcA. A conserved NtcA-binding site is centered around position −42.5 with respect to the transcription start site of NsiR3 homologs, and NtcA binds in vitro to a DNA fragment containing this sequence. In the absence of combined nitrogen, NsiR3 expression is induced in all cells along the Nostoc filament but much more strongly in heterocysts, differentiated cells devoted to nitrogen fixation. Co-expression analysis of transcriptomic data obtained from microarrays hybridized with RNA obtained from Nostoc wildtype or mutant strains grown in the presence of ammonium or in the absence of combined nitrogen revealed that the expression profile of gene putA (proline oxidase) correlates negatively with that of NsiR3. Using a heterologous system in Escherichia coli, we show that NsiR3 binds to the 50-UTR of putA mRNA, resulting in reduced expression of a reporter gene. Overexpression of NsiR3 in Nostoc resulted in strong reduction of putA mRNA accumulation, further supporting the negative regulation of putA by NsiR3. The higher expression of NsiR3 in heterocysts versus vegetative cells of the N 2 -fixing filament could contribute to the previously described absence of putA mRNA and of the catabolic pathway to produce glutamate from arginine via proline specifically in heterocysts. Post-transcriptional regulation by NsiR3 represents an indirect NtcA-operated regulatory mechanism of putA expression. Database Microarray data are available in GEO database under accession numbers GSE120377 and GSE150191. Introduction In some filamentous cyanobacteria, the acclimation to nitrogen deficiency involves differentiation of heterocysts, a cell type specialized for the fixation of atmospheric nitrogen [1]. Growth at the expense of N 2 involves complete metabolic remodeling so that two different cell types with different metabolic capabilities Abbreviations 2-OG, 2-oxoglutarate; dRNASeq, differential RNA sequencing; FC, fold change; GEO, Gene Expression Omnibus; GOGAT, glutamine oxoglutarate aminotransferase; GS, glutamine synthetase; GSA, glutamate γ-semialdehyde; LB, Luria Broth; OAA, oxaloacetate; OAC, ornithine–ammonium cycle; P5C, Δ 1 -pyrroline-5-carboxylate; Sm, streptomycin; Sp, spectinomycin; sRNA, small RNA; β-Asp-Arg, βaspartylarginine; STRR, short tandemly repeated repetitive; TCA, tricarboxylic acid cycle. 1614 The FEBS Journal 288 (2021) 1614–1629 ª2020 Federation of European Biochemical Societies (vegetative cells and heterocysts) cooperate to achieve growth of the filament as a whole [2]. The differentiation of heterocysts involves complex transcriptional changes both in vegetative cells and in heterocysts not only to achieve morphological differentiation of heterocysts but also to undergo metabolic adaptations required for growth at the expense of N 2 [3,4]. Small RNAs (sRNAs) constitute a relevant class of post-transcriptional regulators involved in the adaptation of bacterial metabolism to different environmental situations [5]. sRNAs can coordinate changes in gene expression in response to environmental stresses. Similar to the observations made in other groups of bacteria, cyanobacteria exhibit abundant transcription of noncoding RNAs, including antisense RNAs and sRNAs, and transcription of many of them is regulated in response to nitrogen availability [6–9]. Several nitrogenregulated noncoding RNAs have been identified and characterized in the filamentous, heterocyst-forming cyanobacterium Nostoc sp. PCC 7120 [3,10–14]. Furthermore, heterocyst-specific transcription of noncoding RNAs has been described [3,11,15,16], and some of those transcripts affect the process of heterocyst differentiation [17] or the regulation of enzymatic activities whose levels must be adjusted specifically in heterocysts versus vegetative cells [13]. NsiR3 is an sRNA whose expression is induced upon nitrogen deprivation and depends on NtcA, the global nitrogen regulator in cyanobacteria [9,11].It was originally described as conserved in 27 heterocystforming cyanobacteria but absent in nonheterocystous strains [11], suggesting a possible role of NsiR3 related to heterocyst function. However, no regulatory target has so far been identified for NsiR3. Computational prediction of mRNAs regulated by a given sRNA can be misleading because the interactions between sRNAs and their targets take place through short, often discontinuous complementary sequences [18]. Experimental identification of possible targets based on simple analysis of differences in expression changes upon alteration of the amount of the sRNA is in many cases impeded by secondary effects. Furthermore, such approaches rely on the assumption that the post-transcriptional regulation exerted by the sRNA results in clear changes in the amount of the target mRNA, which is not necessarily true [19]. For this reason, the combination of computational and more sophisticated experimental approaches is often more productive (reviewed in Ref. [20]). In this work, we show that upon nitrogen deprivation, expression of NsiR3 is more strongly induced in heterocysts than in vegetative cells. By using a novel approach based on correlation of expression analysis we identify putA, that encodes proline oxidase, as a target of NsiR3 because of the strong negative correlation with NsiR3. Finally, we demonstrate the interaction between NsiR3 and the 50-UTR of the putA mRNA, and the negative regulation exerted by NsiR3 on the expression of putA in Nostoc sp. PCC 7120. Results NsiR3 is conserved in heterocyst-forming cyanobacteria, and its transcription is induced upon nitrogen stress The nitrogen-regulated transcription of NsiR3 was discovered in a previous dRNASeq experiment [9]. Phylogenetic conservation of NsiR3 homologs across cyanobacterial genomes was later described [11], and we have now identified NsiR3 homologs in additional cyanobacteria, confirming that NsiR3 appears restricted to heterocyst-forming strains (Fig. 1). In Nostoc sp. PCC 7120, NsiR3 is transcribed downstream of all4558 and in the opposite orientation (Fig. 2A). The predicted secondary structure contains two possible stem–loops acting as transcriptional terminators (Fig. 2B); the second one (T2) is also found in closely related cyanobacterial strains and is formed by STRR (short tandemly repeated repetitive) imperfect 7-nt repeats that are present in genomes of filamentous cyanobacteria usually in intergenic regions but in some case within coding sequences. Their function, if any, is unknown [21,22] (Fig. 2C). Expression of NsiR3 is induced upon nitrogen stress (Fig. 2D,E). Induction upon removal of combined nitrogen is quick, and the amounts of NsiR3 remain high up to 24 h. Transfer from ammonium-containing medium to nitrate-containing medium also induces NsiR3 expression but to lower levels and the induction is transient (Fig. 2D,E), reflecting the fact that nitrogen stress in these conditions is quickly relieved by derepression of the nir operon which encodes nitrate reductase and nitrite reductase, the enzymes involved in assimilation of nitrate [23,24]. Consistent with the predicted secondary structure, two RNAs, of 45 and 115 nucleotides, were detected by northern hybridization with the NsiR3 probe. The 45-nucleotide RNA (NsiR3S) is of the expected size from the experimentally determined transcriptional start site (TSS) for NsiR3 [9] to the first predicted terminator, T1. The major detected band is 115 nucleotides long (NsiR3L) and matches the calculated length from the TSS to the second predicted terminator, T2, further downstream (Fig. 2B). NsiR3L is the most abundant of the two species of NsiR3; therefore, it was used in all the experiments described here. 1615The FEBS Journal 288 (2021) 1614–1629 ª2020 Federation of European Biochemical Societies I. ´ Alvarez-Escribano et al.sRNA regulation of proline oxidase Fig. 1. Alignment of NsiR3 RNA. The sequences of NsiR3S from different cyanobacteria were aligned with CLUSTAL OMEGA. Nucleotide positions conserved in the 70 strains where NsiR3 was found are indicated by asterisks. Each sequence is indicated by the organism name followed by the GenBank accession number (www.ncbi.nlm.nih.gov/genbank). The NsiR3 sequence from Nostoc sp. PCC 7120 is highlighted in bold. 1616 The FEBS Journal 288 (2021) 1614–1629 ª2020 Federation of European Biochemical Societies sRNA regulation of proline oxidase I. ´ Alvarez-Escribano et al. NtcA directly regulates NsiR3 expression Transcription of NsiR3 in Nostoc sp. PCC 7120 depends on NtcA [9,11]. Alignment of the promoter region from all 70 identified nsiR3 sequences shows that a putative conserved NtcA-binding sequence appears centered around position −42.5 in 67 of them (Fig. 3), which is compatible with direct activation of NsiR3 transcription by NtcA. To verify the interaction between NtcA and the promoter region of nsiR3, we performed an electrophoretic mobility shift assay with purified NtcA Fig. 2. The nitrogen stress-inducible (NsiR3). (A) Schematic representation of the region encoding NsiR3 in Nostoc sp. PCC 7120. The two flanking genes are indicated. The bent arrow represents the transcriptional start at position 5452083f, and the stem–loops represent the two transcriptional terminators of NsiR3. The position of the NtcA-binding site is also indicated. (B) Secondary structure model of NsiR3 from Nostoc sp. PCC 7120 based on the consensus obtained with RNAalifold from the alignments in Fig. 1and C this image. The heptanucleotide repeats (STRR) are framed. (C) The sequences of NsiR3 encoded in those strains that could have a second transcriptional terminator (T2) further downstream T1 were aligned using CLUSTAL OMEGA. The predicted seed region is highlighted in yellow. Accession numbers are indicated in Fig. 1. (D) Nitrogen-responsive expression of NsiR3 in Nostoc sp. PCC 7120. Expression was analyzed by northern blot in cells grown in the presence of ammonium and transferred to medium containing no source of combined nitrogen (N 2 ) or containing nitrate (NO 3) for the number of hours indicated. The upper panel shows hybridization to the NsiR3 probe. The lower panel shows hybridization to a probe for 5S RNA used as loading and transfer control. Sizes (nt) are indicated on the left. The experiment was repeated four times with similar results. The northern blot containing the highest number of time points is shown. (E) Quantification of NsiR3L in the blot shown in (D) upon nitrogen removal (black, N 2 ) or upon nitrogen removal followed by nitrate addition (red, NO 3). 1617The FEBS Journal 288 (2021) 1614–1629 ª2020 Federation of European Biochemical Societies I. ´ Alvarez-Escribano et al.sRNA regulation of proline oxidase Fig. 3. Alignment of the promoter sequences of nsiR3. The sequences upstream nsiR3 were aligned with CLUSTAL OMEGA. Nucleotide positions fully conserved are indicated by asterisks. The consensus sequence is represented with WebLogo [49]. The likely −10 elements are boxed, and two experimentally determined TSS from Nostoc sp. PCC 7120 [9] and Nodularia spumigena CCY9414 [50] are circled in red. Accession numbers are indicated in Fig. 1. 1618 The FEBS Journal 288 (2021) 1614–1629 ª2020 Federation of European Biochemical Societies sRNA regulation of proline oxidase I. ´ Alvarez-Escribano et al. protein and a DNA fragment extending from positions −118 to +31 with respect to the TSS of nsiR3 that includes the possible NtcA-binding sequence. In the presence of NtcA, a retarded complex was observed with the fragment containing the wild-type promoter sequence (Fig. 4A). However, the fraction of fragment bound by NtcA was strongly (ninefold) reduced (Fig. 4 A,B) when the assay was performed with a DNA fragment containing a version of the nsiR3 promoter with the NtcA-binding site mutated (GTG changed to CAC). These results demonstrate the binding of NtcA to a specific sequence of the nsiR3 promoter. NsiR3 accumulates differentially in heterocysts but is not essential for diazotrophic growth It was previously shown that NsiR3 transcription is dependent on NtcA but independent of HetR [9,11], the master regulator of heterocyst differentiation. Therefore, we expect that NsiR3 transcription would be induced in all cells of the filament upon nitrogen stress. To characterize the pattern of expression of NsiR3, we prepared a plasmid containing a transcriptional fusion of the NsiR3 promoter to the gfp gene (pSAM341, see Table S4). The plasmid was introduced in Nostoc sp. PCC 7120 by conjugation, and expression of GFP was analyzed by confocal fluorescence microscopy (Fig. 5). Filaments growing in ammoniumcontaining medium had very low green fluorescence, while filaments growing in medium lacking combined nitrogen showed fluorescence in all cells, but peaks of green fluorescence were associated with cells that were differentiating as heterocysts, as indicated by their lower red autofluorescence. Interestingly, induction of the nsiR3 promoter occurs early in heterocyst development because significant green fluorescence was already observed in immature heterocysts, still having substantial red fluorescence (not shown). The conservation of NsiR3 in heterocyst-forming cyanobacteria and its NtcA-dependent expression suggests that NsiR3 is involved in some aspect of the regulation of the response to nitrogen stress in these complex cyanobacteria that includes differentiation of specialized cells. To explore the possible function of NsiR3, we constructed a Nostoc strain lacking NsiR3 (ΔnsiR3). This strain grew similarly to wild-type in media containing ammonium and nitrate, or lacking combined nitrogen (Fig. 6); therefore, the differentiation of functional heterocysts is not compromised in the absence of NsiR3. NsiR3 is a regulator of putA (alr0540) Bacterial sRNAs impact their target mRNAs often by repressing the initiation of translation via binding to their 50-UTRs [5]. Hence, the mRNAs do not associate with ribosomes and may become more vulnerable to riboendonucleases, leading to reduced mRNA level. To identify potential targets of NsiR3 whose expression would be negatively regulated by NsiR3 in Nostoc sp. PCC 7120, we decided to search for genes whose expression profile showed a negative correlation with the expression of NsiR3. For this purpose, we have used a novel approach based on correlation analysis of data obtained from hybridization of high-density microarrays [3]. In order to identify those genes with negative correlation with NsiR3, we analyzed data obtained from hybridization of RNA samples extracted from the ΔnsiR3 mutant grown in ammonium-containing medium (two replicates) or from the ΔnsiR3 mutant after 8 h of incubation in medium lacking combined nitrogen (two replicates). In addition, in order to increase the Fig. 4. NtcA binds to the nsiR3 promoter. (A) Electrophoretic mobility shift assays showing binding of purified His-tagged NtcA protein to a DNA fragment containing the wild-type promoter of NsiR3 from Nostoc sp. PCC 7120 (left) or a mutated version altered in the positions indicated in red (right). Three times more probe was used in the assay with the mutant fragment than with the wild-type fragment. A representative experiment is shown. (B) Quantification of the autoradiogram presented in panel (A). The fraction of probe bound by NtcA was calculated for the wild-type (black) or the mutated (red) version. 1619The FEBS Journal 288 (2021) 1614–1629 ª2020 Federation of European Biochemical Societies I. ´ Alvarez-Escribano et al.sRNA regulation of proline oxidase sensitivity of the correlation analysis, previously available data from hybridization of the same microarray platform with other 20 samples (both from the wild-type strain and from a hetR mutant strain either in the presence of ammonium or after different periods of nitrogen deprivation) [3] were included in the analysis. The expression data from the 24 samples (Table S1) were analyzed as described [3], and a correlation table was obtained for NsiR3 (Table S2). The probe with the strongest negative correlation with NsiR3 (−0.943) was that of gene alr0540 (putA), encoding proline oxidase. Moreover, another probe corresponding to the 50-UTR of alr0540 had a strong negative correlation as well (−0.884). Figure 7A shows the expression profile of NsiR3 and putA in the 24 samples hybridized to the arrays. We also confirmed by northern blot the opposite Fig. 5. Expression pattern of the nsiR3 promoter in Nostoc filaments. Confocal fluorescence images (red channel, green channel, both channels merged, and brightfield image) of Nostoc filaments carrying the gfp gene under the control of the nsiR3 promoter (plasmid pSAM341) and growing on top of medium containing ammonium, NHþ 4(A), or lacking any source of combined nitrogen, N 2 (B). Quantification of the signals for the green channel (GFP) is shown for the segments indicated by a blue line in the corresponding bright-field images. All images were acquired with the same sensitivity settings so that intensities can be compared. Scale bars, 20 µm. Fig. 6. The ΔnsiR3 strain can differentiate heterocysts. Cells were grown in the presence of nitrate, washed with BG11 0 , and resuspended at 0.1 μg chlorophyllmL −1 . Fivefold serial dilutions of liquid cultures of wild-type or four independent ΔnsiR3 isolates were prepared. Ten microlitre of the cell suspension and of three fivefold serial dilutions were plated on BG11 0 plates containing ammonium (NHþ 4), nitrate (NO 3), or no source of combined nitrogen (N 2 ). Pictures were taken after 10 days of incubation at 30 °C. 1620 The FEBS Journal 288 (2021) 1614–1629 ª2020 Federation of European Biochemical Societies sRNA regulation of proline oxidase I. ´ Alvarez-Escribano et al. expression dynamics of putA with NsiR3 that was deduced from the co-expression analysis (Fig. 7B). The results described above strongly suggest that the expression of putA might be regulated by NsiR3. In fact, we could predict with IntaRNA [25] an interaction between NsiR3 and the 50-UTR of putA mRNA (Fig. 8A). NsiR3 would bind to positions −5to−17 with respect to the initiation codon of putA, occluding the ribosome-binding site. To verify the interaction between NsiR3 and the 50UTR of putA, we used a heterologous reporter system [26] in which the 50-UTR plus the first 60 nt of the putA coding sequence was translationally fused to the gene encoding superfolder GFP (sfgfp). This construct was co-expressed in E. coli either with NsiR3 or with a control, unrelated RNA. The initiation codon of putA was changed from GTG to ATG for optimal expression in E. coli. The fluorescence of cells carrying sfgfp fusions significantly decreased when NsiR3 was co-expressed, indicating a negative effect of NsiR3 on expression of sfGFP (Fig. 8B). To verify the interactions between NsiR3 and the mRNA of putA, we created a mutated version of NsiR3 altered in position 14 (G to C, NsiR3 Mut14) (Fig. 8A). The mutation in NsiR3 eliminated the interaction between NsiR3 and the mRNA of putA (Fig. 8B). We designed a compensatory mutation in the 50-UTR of putA (Comp-14*17) that would restore the interaction with NsiR3 (Mut14) (Fig. 8A). When NsiR3 (Mut14) was combined with the mutated version of the 50-UTR of putA, the reduction of fluorescence was restored (Fig. 8B). All together, these results verify the interaction of NsiR3 with the 50-UTR of putA at the positions predicted by IntaRNA. In consequence, an inhibitory effect of NsiR3 on putA in the in vivo context of the heterologous E. coli system is supported. NsiR3 represses the expression of putA in Nostoc sp. PCC 7120 To analyze the regulatory effect of NsiR3 on putA in Nostoc sp. PCC 7120, we generated a Nostoc strain Fig. 7. Negative correlation of putA and nsiR3 expression. (A) Expression level of nsiR3 (red) and putA (alr0540, black), in the 24 RNA samples hybridized to the microarrays used for the co-expression analysis. (B) putA expression after removal of combined nitrogen. RNA was extracted from Nostoc cultures at different times after nitrogen removal and subjected to northern blot. The filter was hybridized with probes for putA (top), nsiR3 (middle), and rnpB (bottom), that was used as loading control. The experiment was repeated three times with similar results. The northern blot containing the highest number of time points is shown. (C) Quantification of the blot shown in (B). The amount of putA mRNA (black) and NsiR3 (red) is expressed as percentage of the maximum. For putA, the full-length transcript (arrow in B) was used in quantification. Sizes of ribosomal RNAs are indicated in (B). 1621The FEBS Journal 288 (2021) 1614–1629 ª2020 Federation of European Biochemical Societies I. ´ Alvarez-Escribano et al.sRNA regulation of proline oxidase with controlled expression of NsiR3 by complementing strain ΔnsiR3 with a plasmid containing nsiR3 under control of the copper-inducible petE promoter (ΔnsiR3+P petE ::nsiR3). Accumulation of the mRNA of putA was analyzed by northern blot in the wild-type, ΔnsiR3, and ΔnsiR3+P petE ::nsiR3 strains at different times after nitrogen removal and copper addition (Fig. 9A). A stronger reduction in the amount of the full-length putA mRNA was observed in the ΔnsiR3+P petE ::nsiR3 strain than in the wild-type, in agreement with the observation that ΔnsiR3+P petE :: nsiR3 accumulated about 40 times more NsiR3 than the wild-type (Fig. 9A,B). The deletion of nsiR3 in the ΔnsiR3 strain resulted in levels of the putA mRNA that were not significantly reduced after 24 h in the absence of combined nitrogen. These results indicate a strong negative correlation between the amount of NsiR3 and the amount of the putA mRNA. As a control, we hybridized the same filter with a probe for gene agrE encoding the bifunctional enzyme arginine dihydrolase/ornithine cyclodeaminase that acts upstream proline oxidase in the catabolic pathway from arginine to proline [27]. The dynamic of changes in the amount of agrE mRNA upon nitrogen removal was different to that of putA. In the wild-type strain, whereas the amount of putA mRNA was reduced to about 40% after 24 h in the absence of combined nitrogen, no significant change was observed in the case of agrE. In addition, the levels of agrE mRNA did not correlated to changes in the amount of NsiR3 (Fig. 9), as predicted for a mRNA that is not a target of NsiR3. Discussion In this work, we characterize NsiR3, a highly conserved sRNA in heterocystous cyanobacteria. NsiR3 was identified in the genomes of seventy heterocystforming cyanobacteria but was not found in unicellular or filamentous strains that do not develop heterocysts. This distribution of the nsiR3 gene, in contrast to that of other NtcA-regulated sRNAs such as NsiR4, also found in unicellular strains [10], suggests a relevant function related to the specific metabolic traits of heterocyst-forming strains. nsiR3 is not transcribed in the presence of ammonium, but its transcription is strongly induced upon transfer to nitrogen-free medium, suggesting a possible function in the response to nitrogen stress. Furthermore, nsiR3 transcription is also induced upon transfer from ammoniumto nitrate-containing medium. However, in this case the induction is lower and transient. This can be explained because the nitrogen stress is quickly relieved upon induction of the nitrate assimilation pathway in the presence of nitrate [23]. We show that the global regulator of nitrogen assimilation, NtcA, binds directly to the nsiR3 promoter, explaining the previous observation that NtcA is required for expression of NsiR3 [9,11]. By means of a fusion to the gfp gene, we show that expression of NsiR3, although induced in all cells of the filament under nitrogen deprivation, is much stronger in heterocysts than in vegetative cells. This observation might be explained by the higher concentration of NtcA present in the heterocyst versus in vegetative cells [28,29]. Alternatively, the nsiR3 promoter Fig. 8. Verification of NsiR3 interaction with the 50-UTR of putA using an in vivo reporter system. (A) Predicted interaction between NsiR3 and the 50-UTR of putA mRNA according to INTARNA [25]. Nucleotides are numbered with respect to the start of the coding sequence (initiation codon is underlined). The ribosome-binding site in the putA mRNA is highlighted in yellow. A mutation introduced in NsiR3 at position 14 (G to C, Mut14) and the corresponding compensatory mutation in putA 50-UTR position −17 (C to G, Comp-14*17) are indicated in red and blue, respectively. (B) Fluorescence measurements of E. coli DH5αcultures bearing combinations of plasmids expressing different versions of NsiR3 and putA::sfgfp fusions. Plasmid pJV300 (encoding a control RNA) was used as control. The data are presented as the mean standard deviation of cultures from eight independent colonies after subtraction of fluorescence in cells bearing pXG-0 (***P<0.0001, Student t-test). 1622 The FEBS Journal 288 (2021) 1614–1629 ª2020 Federation of European Biochemical Societies sRNA regulation of proline oxidase I. ´ Alvarez-Escribano et al. Pseudomonas stutzeri A1501. Proc Natl Acad Sci USA 113, E4348–E4356. 34 Rippka R, Deruelles J, Waterbury JB, Herdman M & Stanier RY (1979) Generic assignments, strain histories and properties of pure cultures of cyanobacteria. Microbiology 111,1–61. 35 Sambrook J & Russell DW (2001) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. 36 Muro-Pastor AM, Valladares A, Flores E & Herrero A (1999) The hetC gene is a direct target of the NtcA transcriptional regulator in cyanobacterial heterocyst development. J Bacteriol 181, 6664–6669. 37 Urban JH & Vogel J (2007) Translational control and target recognition by Escherichia coli small RNAs in vivo.Nucleic Acids Res 35, 1018–1037. 38 Wright PR, Richter AS, Papenfort K, Mann M, Vogel J, Hess WR, Backofen R & Georg J (2013) Comparative genomics boosts target prediction for bacterial small RNAs. Proc Natl Acad Sci USA 110, E3487–E3496. 39 Elhai J & Wolk CP (1988) Conjugal transfer of DNA to cyanobacteria. Methods Enzymol 167, 747–754. 40 Merino-Puerto V, Mariscal V, Mullineaux CW, Herrero A & Flores E (2010) Fra proteins influencing filament integrity, diazotrophy and localization of septal protein SepJ in the heterocyst-forming cyanobacterium Anabaena sp. Mol Microbiol 75, 1159–1170. 41 Cai Y & Wolk CP (1990) Use of a conditionally lethal gene in Anabaena sp. strain PCC 7120 to select for double recombinants and to entrap insertion sequences. J Bacteriol 172, 3138–3145. 42 Buikema WJ & Haselkorn R (2001) Expression of the Anabaena hetR gene from a copper-regulated promoter leads to heterocyst differentiation under repressing conditions. Proc Natl Acad Sci USA 98, 2729–2734. 43 Mohamed A & Jansson C (1989) Influence of light on accumulation of photosynthesis-specific transcripts in the cyanobacterium Synechocystis 6803. Plant Mol Biol 13, 693–700. 44 Vioque A (1992) Analysis of the gene encoding the RNA subunit of ribonuclease P from cyanobacteria. Nucleic Acids Res 20, 6331–6337. 45 Altschul SF, Gish W, Miller W, Myers EW & Lipman DJ (1990) Basic local alignment search tool. J Mol Biol 215, 403–410. 46 Madeira F, Park YM, Lee J, Buso N, Gur T, Madhusoodanan N, Basutkar P, Tivey ARN, Potter SC, Finn RD et al. (2019) The EMBL-EBI search and sequence analysis tools APIs in 2019. Nucleic Acids Res 47, W636–W641. 47 Lorenz R, Bernhart SH, H¨ oner zu Siederdissen C, Tafer H, Flamm C, Stadler PF & Hofacker IL (2011) ViennaRNA package 2.0. Algorithms Mol Biol 6, 26. 48 Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W & Smyth GK (2015) limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res 43, e47. 49 Crooks GE, Hon G, Chandonia J-M & Brenner SE (2004) WebLogo: a sequence logo generator. Genome Res 14, 1188–1190. 50 Voss B, Bolhuis H, Fewer DP, Kopf M, M¨ oke F, Haas F, El-Shehawy R, Hayes P, Bergman B, Sivonen K et al. (2013) Insights into the physiology and ecology of the brackish-water-adapted Cyanobacterium Nodularia spumigena CCY9414 based on a genome-transcriptome analysis. PLoS One 8, e60224. Supporting information Additional supporting information may be found online in the Supporting Information section at the end of the article. Table S1. Differentially_expressed_genes_all_samples. Table S2. Pearson_correlation_of_NsiR3_and_all_genes. Table S3. Strains. Table S4. Plasmids. Table S5. Oligonucleotides. Table S6. Sequences. 1629The FEBS Journal 288 (2021) 1614–1629 ª2020 Federation of European Biochemical Societies I. ´ Alvarez-Escribano et al.sRNA regulation of proline oxidase