R-DeeP/TripepSVM identifies the RNA-binding OB-fold-like protein PatR as regulator of heterocyst patterning

Nucleic Acids Research , 2025, 53 , gkae1247 https://doi.org/10.1093/nar/gkae1247 Advance access publication date: 19 December 2024 RNA and RNA-protein complexes R-DeeP / TripepSVM identifies the RNA-binding OB-fold-like protein PatR as regulator of heterocyst patterning Manuel Br enes-Álvar ez 1 , * , Halie R. Ropp 1 , Dimitr ios P apagiannidis 2 , Clement M. Potel 2 , F r ank St ein 2 , Ing eborg Sc holz 1 , Claudia St eglic h 1 , Mikhail M. Savitski 2 , Agustín Vioque 3 , Alicia M. Muro-Pastor 3 and Wolfgang R. Hess 1 1 Genetics and Experimental Bioinformatics, Faculty of Biology, University of Freiburg, Schänzlestr. 1, 79104 Freiburg, Germany 2 European Molecular Biology Laboratory (EMBL), Meyerhofstr. 1, 69117 Heidelberg, Germany 3 Instituto de Bioquímica Vegetal y Fotosíntesis, Consejo Superior de Investigaciones Científicas and Universidad de Sevilla, Avenida Américo Vespucio 49, 41092 Sevilla, Spain * To whom correspondence should be addressed. Tel: +49 761 / 203 2708 Email: [email protected] Abstract RNA-binding proteins (RBPs) are central components of gene regulatory netw orks. T he differentiation of heterocysts in filamentous cyanobacteria is an example of cell differentiation in prokaryotes. Although multiple non-coding transcripts are in v olv ed in this process, no RBPs ha v e been implicated thus far. Here we used quantitative mass spectrometry to analyze the differential fractionation of RNA–protein comple x es after RNase treatment in density gradients yielding 333 RNA-associated proteins, while a bioinformatic prediction yielded 311 RBP candidates in Nostoc sp. PCC 7120. We validated in vivo the RNA-binding capacity of six RBP candidates. Some participate in essential physiological aspects, such as photosynthesis (Alr2890), th ylak oid biogenesis (Vipp1) or heterocy st differentiation (P rpA, PatU3), but their association with RNA w as unkno wn. Validated RBPs Asl3888 and Alr1700 were not pre viously characteriz ed. Alr1700 is an RBP with two oligonucleotide / oligosaccharide-binding (OB)-f old-lik e domains that is differentially expressed in heterocysts and interacts with non-coding regulatory RNAs. Deletion of alr1700 led to complete deregulation of the cell differentiation process, a striking increase in the number of heterocy st-lik e cells, and was ultimately lethal in the absence of combined nitrogen. These observations characterize this RBP as a master regulator of the heterocyst patterning and differentiation process, leading us to rename Alr1700 to PatR. Gr aphical abstr act R-DeeP in Cyanobacteria In vivo validation (PNK assays) Characterization of novel RBPs Western-Blot Autoradiography -+ -+ UV 3xFLAG-PatR KDa 70 55 40 35 15 25 10 RBPs with novel protein domains and architecture 6 new RBPs validated in vivo Some RBPs are involved in cell differentiation in cyanobacteria 333 RNA-associated proteins detected Size gradient + RNases MS Normalized protein amount non-RBP RBP RBP Fractions - RNases Introduction Filamentous cyanobacteria are considered one of the oldest multicellular organisms on Earth ( 1 ). Some of these organisms, such as Nostoc sp. PCC 7120 (hereafter Nostoc ), also known as Anabaena sp. PCC 7120, have the ability to differentiate a specialized cell type devoted to nitrogen fixation, the heteReceived: May 31, 2024. Revised: November 15, 2024. Editorial Decision: December 2, 2024. Accepted: December 4, 2024 ©The Author(s) 2024. Published by Oxford University Press on behalf of Nucleic Acids Research. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (https: // creativecommons.org / licenses / by-nc / 4.0 / ), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact [email protected] for reprints and translation rights for reprints. All other permissions can be obtained through our RightsLink service via the Permissions link on the article page on our site—for further information please contact [email protected]. Downloaded from https://academic.oup.com/nar/article/53/3/gkae1247/7928523 by Biblioteca Universidad de Sevilla user on 27 February 2025 2 Nucleic Acids Research , 2025, Vol. 53, No. 3 rocyst. Under nitrogen-replete conditions there are no heterocysts, but when the filaments encounter nitrogen starvation, cell–cell competition following a one-dimensional Turing pattern induces the genetic program for cell differentiation only in some vegetative cells. The developmental program takes about 24 h to complete, providing an excellent experimental model for the analysis of cell differentiation in bacteria. Heterocysts are terminally differentiated, non-dividing cells dedicated to nitrogen fixation, and their morphology and physiology have evolved to create a microoxic environment that protects and supports nitrogenase activity ( 2 ). During diazotrophic growth, vegetative cells continue to fix CO 2 through photosynthesis, while heterocysts fix N 2 . Therefore, both types of cells must exchange metabolites to maintain the growth of the filament as a whole, providing one of the simplest examples of a true multicellular organism with two cell types cooperating together ( 3 ). NtcA and HetR have been recognized as the main transcriptional regulators of this differentiation process ( 4 ). However, in striking parallel to more complex eukaryotic organisms, genome-wide analyses have suggested the involvement of hundreds of non-coding RNAs [antisense RNAs (asRNAs) and small RNAs (sRNAs)] in this process ( 5 ,6 ). Several sRNAs are regulated by the availability of nitrogen themselves and were designated n itrogen s tress-i nduced R NAs (NsiR). Some have been characterized in more detail ( 7–9 ) and are involved in the reshaping of heterocyst metabolism ( 9 ) or in the core regulation of the developmental process ( 8 ). NsiR1 is expressed at a very early stage in the cells undergoing differentiation ( 10 ) and inhibits, as an asRNA, the translation of the overlapping hetF messenger RNA (mRNA) ( 11 ) encoding a protease crucially involved in heterocyst differentiation ( 12 ). As a trans -acting sRNA, NsiR1 regulates the expression of alr3234 ( 8 ), encoding a HetP-like protein involved in the regulation of commitment to heterocyst differentiation ( 13 ). Regulation through prokaryotic non-coding RNAs frequently depends on RNAbinding proteins (RBPs) such as RNA chaperons or matchmakers, and extensive networks of post-transcriptional regulation have been elucidated ( 14 ), but little is known about the interaction of RBPs and their functions in cyanobacteria. In fact, homologs to prominent RNA chaperones in proteobacteria, such as CsrA, ProQ, FinO or Hfq do not exist in cyanobacteria or do not bind RNA ( 15 ). RBPs play a central role in the physiology of cells. The ribosomes, the RnpB-RnpA (RNase P) complex, the signal recognition particle, the transfer -messenger -RNA-SmpB complex, or the vast diversity of CRISPR systems are well-known ribonucleoprotein complexes in prokaryotes. RBPs are key regulators of gene expression because they determine the fate of all RNA molecules in the cell ( 16 ), including the regulation of mRNA stability or the spatial localization of mRNAs in the cell ( 17 ). However, only a few pieces of information on the role of RBPs in the physiology of cyanobacteria are known. A family of RBPs containing an RNA recognition motif (Pfam: PF00076), similar to certain plant RBPs, was discovered in cyanobacteria almost 30 years ago ( 18 ). Recently, they have been shown to be essential for the targeting of mRNAs encoding photosystem subunits near the thylakoid membrane, the site where photosynthetic proteins should be inserted ( 19 ). Several experimental screens for the identification of RBPs have been developed in recent years. They rely on the fractionation of cellular ribonucleoprotein complexes using either glycerol / sucrose gradients and ultracentrifugation ( 20 ,21 ) or size exclusion chromatography ( 22 ). In Grad-seq the further analysis of the components of each fraction using RNA-seq and mass spectrometry (MS) allows the prediction of RBPs and their RNA targets ( 23 ). GradR / R -DeeP ( 24 , 25 ) use an RNase-treated sample to identify proteins whose position in the gradient shift after RNA removal. It is important to note that GradR / R-DeeP identifies not only RBPs, but also proteins that are part of RNA–protein complexes, although they do not directly interact with the RNA component ( 25 ). In this work, we have used GradR / R -DeeP ( 24 , 26 ) to identify new RBPs expressed in Nostoc cell filaments that undergo heterocyst differentiation. Since GradR and R-DeeP are similar approaches, only R-DeeP will be used in the following. We have identified 333 RNA-associated proteins that may be part of RNA–protein complexes. The in vivo validation of some of the candidates opens a new world of riboregulation in these organisms, as some RBPs have described functions not related to RNA metabolism, or were completely uncharacterized, such as Alr1700 (PatR), a protein with two OB-fold-like domains that affects the cell differentiation process. Materials and methods Strains and growth conditions The different Nostoc strains used in this work ( Supplementary Table S1 ) were grown photoautotrophically with constant shaking under standard conditions (30 ◦C and 50 μE illumination) in liquid BG11 medium ( 27 ). Filaments were collected by centrifugation at 3270 ×g for 5 min at room temperature (RT), washed and then resuspended in BG11 0 (medium without a combined nitrogen source) to induce heterocyst differentiation. Nostoc derivative strains bearing Sm R Sp R or Nm R plasmids were grown in liquid in the presence of 2 μg / ml streptomycin (Sm) and spectinomycin (Sp) each, or 5 μg / ml neomycin (Nm). R-DeeP experiment; cell lysis, gradient preparation and fractionation Triplicate 200 ml Nostoc liquid cultures were grown photoautotrophically under standard conditions in BG11 medium until they reached a concentration of 3 μg chlorophyll / ml. Heterocyst differentiation was induced as described above. After 26 h of cultivation, the filaments were harvested by centrifugation and resuspended in ice-cold lysis buffer (20 mM Tris-HCl pH 7.5, 150 mM KCl, 10 mM MgCl 2 , 1 mM dithiothreitol) supplemented with protease Inhibitor [cOmplete ethylenediaminetetraacetic acid (EDTA)-free, Roche]. From this point on, the samples were kept at 4 ◦C. Each replicate was divided into equal sample tubes. The control samples in which the RNA should be preserved were supplemented with 8 μl of Ribolock RNase Inhibitor. To all tubes, 100 μl of a mixture of 0.1 and 0.5 mm glass beads were added. Cells were mechanically lysed in a pre-chilled Precellys homogenizer (Bertin Technologies) using nine cycles of 10 s shaking at 6000 rpm followed by a pause of 5 s. Cell debris and glass beads were removed by centrifugation at 1500 ×g for 2 min. Next, 1.25% β-DM was added to the supernatant to solubilize membrane proteins by vortexing for 30 s. Membranes were pelleted by centrifugation at 21000 ×g for 15 min. Next, 100 μl of RNase A / T1 mix (Thermo, 2 mg / ml RNase A and 5000 U / ml RNase T1 stock) was added to the RNase-treated extracts, while 100 μl of RNase buffer (50 mM Tris-HCl pH 7.4 and 50% glycDownloaded from https://academic.oup.com/nar/article/53/3/gkae1247/7928523 by Biblioteca Universidad de Sevilla user on 27 February 2025 Nucleic Acids Research , 2025, Vol. 53, No. 3 3 erol) was added to the control extracts. Control and RNasetreated samples were incubated on ice for 20 min. Linear sucrose gradients were prepared on Ultra-Clear Beckman tubes using a Gradient Master 108 forming system (Biocomp) and two solutions of lysis buffer supplemented with either 10% or 40% sucrose. A total of 600 μl of the control and RNase-treated total extracts were loaded carefully onto the gradients. Macromolecular complexes were separated by ultracentrifugation of the lysates in a swingingbucket rotor (Beckman SV40 Ti) for 16 h at 285000 ×g . Fractions of equal volume (600 μl) were collected using a Piston Gradient Fractionator (Biocomp). The protein concentration of each fraction was determined using the Lowry procedure ( 28 ). Northern and western blot analysis of fractions RNA was extracted using hot phenol ( 29 ) with modifications ( 6 ) from 300 μl of each fraction. The RNA was fractionated on a 10% urea-polyacrylamide (PAA) gel, stained with ethidium bromide and electroblotted onto a HybondN + membrane (Amersham) at 1 mA per cm 2 for 1 h. The distribution of RnpB, Yfr1 and NsiR8 was analyzed by northern hybridization using single-stranded radioactively labeled RNA probes transcribed in vitro from polymerase chain reaction (PCR)-generated templates (see Supplementary Table S2 for primers). Radioactively labeled probes were generated using [ α32 P]-UTP and the Maxiscript T7 In vitro transcription kit (Thermo Fisher Scientific). The distribution of NsiR1, 4.5S and 5S along the gradient fractions was also analyzed by northern hybridization using end-labeled oligonucleotides (see Supplementary Table S2 for primers). Oligonucleotides were labeled with [ γ32 P]-ATP and polynucleotide kinase (PNK) (Thermo). Hybridizations were performed as previously described ( 30 ). The signals were detected with a Typhoon FLA 9500 (GE Healthcare). For western blot analysis, 35 μl of each fraction were fractionated in 15% sodium dodecyl sulfate (SDS)-PAA gels and transferred to a nitrocellulose membrane (Amersham). As a control, identically prepared gels were stained with Coomassie Brilliant Blue G-250. The membranes were blocked in 5% skim milk in 1% Tween in phosphate-buffered saline (T-PBS) for 1 h at room temperature and incubated with rabbit polyclonal antibody anti-RplA (#AS111738,1:5000, Agrisera) or mouse monoclonal anti-FLAG coupled to Horseradish peroxidase (HRP) (#A8592,1:2000, Sigma) for 16 h at 4 ◦C. The membranes were washed three times with T-PBS, 5 min each. Membranes previously incubated with anti-RplA were incubated with the secondary antibody anti-rabbit IgG-HRP (#A8275, 1:10 000, Sigma) for 1 h at room temperature and then washed three times with T-PBS. Signals were detected with western-blot ECL spray (Advanta). Sample preparation for MS and measurements Samples for proteomic analysis were prepared as previously described ( 31 ), with minor modifications. A total of 1% Nlauroylsarcosine was added to equal volumes of samples along the gradient (fractions 1 and 2 were combined) and proteins were digested following a modified SP3 protocol ( 32 ). Specifically, 20 μl samples were incubated with 40 μl bead suspension (Thermo Fischer Scientific, Sera-Mag Speed Beads, 4515– 2105-050250, 6515–2105-050250) in 2.5% formic acid and 50% ethanol (15 min, RT, 500 rpm). Beads were then washed four times with 70% ethanol and proteins were digested overnight (RT, 500 rpm) with trypsin and LysC (200 ng of each enzyme in 5 mM chloroacetamide, 1.25 mM tris(2carboxyethyl)phosphine in 100 mM 4-(2-hydroxyethyl)-1piperazineethanesulfonic acid buffer pH 8, per 5 μg protein). Peptides were eluted from the beads, dried under vacuum, and resuspended in water. Ten microlitre samples (up to 5 μg per sample) were labeled with 50 μg of TMT18plex (Thermo Fisher Scientific) reconstituted in 4 μl acetonitrile. Labeling reaction (1 h, RT, 500 rpm) was quenched with 4 μl 5% hydroxylamine (30 min, RT, 500 rpm) and samples of the same gradient were pooled. The samples were then desalted with solid phase extraction by loading onto a Waters OASIS HLB μElution Plate (30 μm), washing twice with 100 μl of 0.05% formic acid, and eluting with 100 μl of 80% acetonitrile. Samples were then dried under vacuum and resuspended in 20 mM ammonium formate (pH 10.0). Offline high pH reverse phase fractionation was performed using an Agilent 1200 Infinity high-performance liquid chromatography system equipped with a quaternary pump, degasser, variable wavelength ultraviolet (UV) detector (set to 254 nm), Peltier-cooled autosampler and fraction collector (both set at 10 ◦C for all samples). The column was a Gemini C18 column (3 μm, 110 Å, 100 ×1.0 mm, Phenomenex) with a Gemini C18, 4 ×2.0 mm SecurityGuard (Phenomenex) cartridge as a guard column. The solvent system consisted of 20 mM ammonium formate (pH 10.0) (Buffer A) and 100% acetonitrile as mobile phase (Buffer B). The separation was accomplished at a mobile phase flow rate of 0.1 ml / min using the following linear gradient: 100% Buffer A for 2 min, from 100% Buffer A to 35% Buffer B in 59 min, to 85% Buffer B in a further 1 min and held at 85% Buffer B for an additional 15 min, before returning to 100% Buffer A and re-equilibration for 13 min. Forty-eight fractions were collected along with the LC separation and subsequently pooled into 12 fractions. Pooled fractions were dried under vacuum centrifugation, reconstituted in 10 μl 1% formic acid, 4% acetonitrile prior liquid chromatography–mass spectrometry analysis. An UltiMate 3000 RSLC nano LC system (Dionex) fitted with a trapping cartridge ( μ-Precolumn C18 PepMap 100, 5 μm, 300 μm internal diameter ×5 mm, 100 Å) and an analytical column (nanoEase™ M / Z HSS T3 column 75 μm × 250 mm C18, 1.8 μm, 100 Å, Waters) was used. Trapping was carried out with a constant flow of 0.05% trifluoroacetic acid at 30 μl / min onto the trapping column for 6 min. Subsequently, peptides were eluted via the analytical column with a constant flow of solvent A (3% dimethyl sulfoxide (DMSO), 0.1% formic acid in water) at 0.3 μl / min with an increasing percentage of solvent B (3% DMSO, 0.1% formic acid in acetonitrile). The outlet of the analytical column was coupled directly to a QExactive plus (Thermo) mass spectrometer using the Nanospray Flex™ ion source in positive ion mode. The peptides were introduced into the QExactive plus via a Pico-Tip Emitter 360 μm outer diameter ×20 μm internal diameter; 10 μm tip (CoAnn Technologies) and an applied spray voltage of 2.3 kV. The capillary temperature was set at 320 ◦C. Full mass scan was acquired with a mass range of 375–1 200 m / z in profile mode with resolution of 70 000. The filling time was set at a maximum of 250 ms with a limitation of 3 ×106 ions. Data-dependent acquisition was performed with the resolution of the Orbitrap set to 35000, with a fill time of 120 ms and a limitation of 2 ×105 ions. A normalized collision energy of 30 was applied. The isolation window Downloaded from https://academic.oup.com/nar/article/53/3/gkae1247/7928523 by Biblioteca Universidad de Sevilla user on 27 February 2025 4 Nucleic Acids Research , 2025, Vol. 53, No. 3 of the quadrupole was set to 0.7 m / z . A dynamic exclusion time of 30 s was used. The peptide match algorithm was set to ‘preferred’ and charge exclusion ‘unassigned’, charge states 1, 5–8 were excluded. MS2 data was acquired in profile mode. Bioinformatic analysis of R-DeeP data Bioinformatic analysis was performed in R. Only proteins with at least two identified peptides were kept for the analysis. In the case of small proteins ( < 100 aa), a more relaxed filter was applied and a single identified peptide was considered sufficient to keep the protein in the analysis. Furthermore, only proteins identified in at least one pair of control and RNasetreated gradients were kept. Batch effect removal and a variance stabilization normalization method were applied to each fraction using the limma ( 33 ) and vsn ( 34 ) packages. An ingradient normalization was also performed, so that the sum of each protein along a gradient was set to 100%. A differential expression-like approach was applied to each fraction using the limma package. The presence of a protein in multiple replicates was taken into account using the parameter weights for the adjustment of the linear model through the lmFit function. The false discovery rate (fdr) was calculated using the fdrtools ( 35 ) package. Proteins with a log 2 fold change ≥1 and fdr ≤0.05 in at least one fraction were considered as hits. Proteins with a log 2 fold change ≥0.58 and fdr ≤0.02 were considered candidates. An R package with the Nostoc annotation was created using the AnnotationForge R package. Sequences and description of the Nostoc proteome were downloaded from NCBI and UNIPROT. The Gene Ontology (GO) terms were downloaded from QuickGO ( https://www.ebi.ac.uk/ QuickGO/annotations ). The molecular weight and isoelectric point for each protein were calculated using the Expasy Compute pI / Mw tool ( https:// web.expasy.org/ compute _ pi/ ). GO-term enrichment of the RNA-associated proteome was performed using the clusterProfiler ( 36 ) R package and a q value ≤0.05 was used as cutoff. In order to group proteins with similar sedimentation profiles, a soft clustering approach was applied to the mean of the control gradients by Mfuzz ( 37 ). An m parameter of 1.14 was selected and a confidence > 0.85 was used as the threshold for a protein to be considered in the core of a cluster. The log 2 fold change of each fraction for the 333 RNA-associated proteins was clustered using hierarchical clustering. Silhouette was used as a metric to determine the best number of clusters in both cases. Bioinformatic prediction of RBPs We used a modified version of TriPepSVM ( 38 ) to predict RBPs in cyanobacteria. Instead of using a reference proteome and its evolutionary close proteomes for the generation of positive and negative datasets, we selected proteomes of cyanobacteria with diverse morphologies and lifestyles (see Supplementary Table S3 ). In addition, we also included the proteome of two well-studied Gram-negative bacteria, Esc heric hia coli K12 and Salmonella typhimurium LT2. All proteomes were downloaded from UNIPROT. The assembly of the positive and negative training datasets was performed following the TriPepSVM pipeline ( 38 ). We used the KeBABS ( 39 ) package to perform a 10-k-fold validation in a grid search using the following combination of parameters: positive class weight, negative class weight, k-mers and cost. The balanced accuracy score was used to select the best combination of parameters. An RBP prediction was performed using the following parameters: positive class weight = 2.7, negative class weight = 0.05, cost = 1 and k-mer = 3. A threshold of 0.25 was selected in the support vector machine (SVM) scores of the proteins for classification as potential RBP, resulting in 311 potential RBPs. The prediction of intrinsically disordered regions in the predicted RBPs was performed using IUPred2A ( 40 ). Regions with an average disorder score > 0.5 were considered not globular regions. A sliding window programmed in R was used to count the frequency of tripeptides in globular or disordered regions of the proteins. A curve was smoothed using locally estimated scatterplot smoothing (LOESS) regression (span = 0.2) to analyze the tendency of tripeptides enriched in RBPs to localize in intrinsically disordered regions. To investigate the evolutionary conservation of the RNAbinding potential of Nostoc proteins, homologs were predicted for the entire Nostoc genome using ‘Genome Gene Best Homologs’ from the IGM / MER Portal of the IMG database ( 41 ), 55 cyanobacterial genomes (see Supplementary Table S4 ), a minimum percentage of identity of 40% and an e-value of 0.05. A prediction of RBPs for all homologs was carried out using the parameters specified above. Construction of mutant and reporter strains To generate a patR strain, two overlapping fragments were amplified by PCR using genomic DNA with oligonucleotides 669 and 670 and oligonucleotides 671 and 672, respectively (see Supplementary Table S2 for oligonucleotides). The resulting products were then used as templates for a third PCR with oligonucleotides 669 and 672 resulting in deletion of the sequences corresponding to patR and the generation of a unique XhoI site between the two amplified fragments. The fragment was digested with BamHI at the sites provided by oligonucleotides 669 and 672, and cloned into BamHI-digested sacB - containing Sm R Sp R vector pCSRO ( 42 ), yielding pMBA62 (see Supplementary Table S5 for plasmids). An Nm R cassette was excised from pRL278 ( 43 ) as a SalI-XhoI fragment and cloned into the Xho site of pMBA62, yielding pMBA66, which was transferred into Nostoc by conjugation ( 44 ). Genomic DNA was extracted from clones exhibiting sucrose resistance and sensitivity to neomycin. Segregation of the clones was analyzed by PCR using genomic DNA and oligonucleotides 738 and 739. Clones without the patR gene and no patR wild-type chromosomal copies were named patR . The plasmid pMBA79 was constructed to express patR from the rnpB promoter. A PCR fragment was amplified using oligonucleotides 738 and 739, digested with NsiI and XhoI and cloned in NsiI and XhoI-digested pMBA20 ( 45 ). pMBA79 or two reporter plasmids [pSAM301 ( 46 ) and pSAM270 ( 5 )] were transferred into patR strain by conjugation with selection of Sm R Sp R cells. Two plasmids for the expression of proteins fused with 3x-FLAG were designed (see the Supplementary Table S5 for the plasmid descriptions). Fragments containing sgfp fused to the 3xFLAG sequence were amplified by PCR using as template pMBA96 ( 8 ) and oligonucleotides 32 and 34 (3xFLAG N-terminal fusion) or 33 and 35 (3xFLAG C-terminal fusion). The resulting products were NsiI / SacIdigested and cloned into NsiI / SacI-digested pMBA20 ( 45 ), yielding pMBA131 and pMBA132, respectively. The rnpB Downloaded from https://academic.oup.com/nar/article/53/3/gkae1247/7928523 by Biblioteca Universidad de Sevilla user on 27 February 2025 Nucleic Acids Research , 2025, Vol. 53, No. 3 5 promoter and sfGFP of pMBA132 were exchanged by upstream sequences plus the coding sequences of the candidate RBPs. DNA fragments were amplified by PCR using as template genomic DNA and oligonucleotides 129 and 130, 136 and 137, 148 and 149, 151 and 152, 156 and 157, 166 and 167, 194 and 195 or P03 and P04, resulting in the amplification of upstream sequences plus the coding sequences of alr2809 , vipp1 , patU3 , prpA , trpB , rnpA , alr2890 and asl3888 , respectively. Fragments were PstI / XhoIdigested and cloned into PstI / XhoI-digested pMBA132, yielding pMBA166, pMBA170, pMBA177, pMBA179, pMBA182, pMBA188, pMBA211 and pMB2, respectively. All plasmids contained the upstream sequences plus the coding sequences of the selected genes fused to the 3xFLAG sequence and were designed for the integration into the Nostoc alpha plasmid. DNA fragments were amplified by PCR using as template pMBA166, pMBA170, pMBA177, pMBA179, pMBA182, pMBA188 or pMBA211, forward oligonucleotides 131, 138, 150, 153, 158, 168, 196 and 19 as reverse oligonucleotide, respectively. These fragments were digested with SacI and cloned into SacI-digested pCSV3 ( 47 ), yielding pMBA167, pMBA171, pMBA178, pMBA180, pMBA183, pMBA189 and pMBA213, respectively. These plasmids contain the upstream sequences plus the coding sequences of alr2809 , vipp1 , patU3 , prpA , trpB , rnpA and alr2890 fused to 3xFLAG sequence and allowed the insertion of the fragments at the native chromosomal loci expressing the fusion proteins from their native promoters. Plasmids pMBA167, pMBA171, pMBA178, pMBA180, pMBA183, pMBA189, pMBA213 and pMB2 were transferred into Nostoc by conjugation with selection of Sm R Sp R cells. An N-terminal fusion of PatR to the 3xFLAG epitope was generated. A fragment containing the promoter and 5  -UTR of patR was amplified by PCR using genomic DNA as template and oligonucleotides 57 and 58. The resulting product was digested with PstI and NsiI and cloned into PstI / NsiIdigested pMBA131, yielding pMBA140. A fragment containing the coding region of patR was amplified by PCR using as template genomic DNA and oligonucleotides 53 and 54. The product was digested with XhoI and SacI and cloned into XhoI- / SacI-digested pMBA140, yielding pMBA149, which was transferred into patR by conjugation with selection of Sm R Sp R cells. In order to generate a patR promoter reporter fused with gfpmut2 , a fragment containing the promoter region of patR (positions 2038795f to 2039043f) was amplified by PCR using genomic DNA as template and oligonucleotides 978 and 979. The resulting fragment was digested with ClaI / XhoI and cloned into ClaI / XhoI-digested pSAM270 (5), yielding pMBA117, which was transferred into Nostoc by conjugation with selection of Sm R Sp R cells. A markerless HetR -3xFLA G strain was generated. A fragment containing the promoter, 5  -UTR and coding region of hetR was amplified by PCR using as template genomic DNA and oligonucleotides 49 and 70. The resulting fragment was PstI / XhoI-digested and cloned into PstI / XhoI-digested pMBA132, yielding pMBA154. A fragment containing the sequences downstream of hetR was also amplified by PCR using genomic DNA as template and oligonucleotides 50 and 51. The fragment was digested with SacI at the sites provided by the oligonucleotides and cloned into SacI-digested pMBA154, yielding pMBA157 (only clones with the correct orientation of the downstream region were kept). A fragment containing the full construct (promoter, 5  -UTR, hetR coding region, 3xFLAG, and downstream region) was finally amplified by PCR using pMBA157 as template and oligonucleotides 47 and 52. The resulting fragment was digested with BamHI at the site provided by the oligonucleotides and cloned into BamHI-digested pCSRO ( 47 ), yielding pMBA160, which was later transferred into the hetR mutant CSSC2 ( 48 ) by conjugation. Sm S Sp S clones exhibiting resistance to sucrose and able to grow again diazotrophically were selected and sequenced to confirm the presence of a hetR copy fused to the 3xFLAG inserted at the hetR locus. PNK assays The PNK assays were carried out as previously described ( 49 ) with some modifications. A total of 400 ml of cultures from reporter strains expressing 3xFLAG protein fusions were grown in BG11 supplemented with the appropriate antibiotics until they reached a concentration of 2.5 μg chlorophyll / ml. Heterocyst differentiation was induced as described above. After 24 h of cultivation, half of each culture (200 ml) was UV-treated (1.6 J / cm 2 ) while gently shaking to induce RNA– protein crosslinking keeping the sample on ice (the remaining half was also placed on ice). Cells were pelleted by centrifugation at 3270 ×g for 5 min. The pellets were resuspended in NT-P buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, 0.05% Tween, pH 8.0) supplemented with protease Inhibitor (cOmplete EDTA-free, Roche). Cells were lysed as described for the R-DeeP experiment. Glass beads were removed by centrifugation at 1500 ×g for 2 min. Next, membranes were pelleted by centrifugation at 21 000 ×g for 30 min at 4 ◦C. A total of 2 ml of clarified lysate per sample (crosslinked and non-crosslinked) was incubated with 80 μl of anti-FLAG M2 magnetic beads (Sigma) prewashed five times in NT-P buffer. From this point on, all the washing steps were performed with 2 ml of buffer. The samples were incubated at 4 ◦C while gently rotating for 1 h. Beads were washed twice with high-salt NT-P buffer (50 mM NaH 2 PO 4 , 1 M NaCl, 0.05% Tween20, pH 8.0), once with of cold NT-P buffer and once with Benzonase buffer (50 mM Tris-HCl, 1 mM MgCl 2 , pH 8). The beads were resuspended in 100 μl benzonase buffer supplemented with protease inhibitor and 25 U of Benzonase nuclease (Sigma). The samples were incubated at 37 ◦C for 10 min, while shaking at 900 rpm. After this, the beads were washed once with high-salt NT-P buffer and twice with FastAP buffer (10 mM Tris-HCl, 5 mM MgCl 2 , 100 mM KCl, pH 8). The beads were resuspended in 150 μl of FastAP buffer supplemented with protease inhibitor and 2 U of Fast Alkaline Phosphatase (Thermo). Samples were incubated at 37 ◦C for 30 min, shaking at 900 rpm. The beads were washed once with high-salt NT-P buffer and twice with PNK buffer (50 mM TrisHCl, 10 mM MgCl 2 , 0.1 mM spermidine, pH 8). The magnetic beads were resuspended in 100 μl of PNK buffer supplemented with protease inhibitor, 10 μCi of [ γ32 P]-ATP and 1 μl of PNK (Thermo). Samples were incubated at 37 ◦C for 30 min, while shaking at 900 rpm. The beads were washed once with high-salt NT-P buffer and twice with NT-P buffer. For the elution of the protein–RNA complexes, the beads were resuspended in 35 μl of Tris-buffered saline (TBS) (50 mM TrisHCl, 150 mM NaCl, pH 7.5) supplemented with 0.5 μg / μl FLAG peptide (Sigma) and incubated at 4 ◦C for 15 min, while gently shaking at 900 rpm. After separation of beads from the supernatant using a magnetic rack, protein loading dye was Downloaded from https://academic.oup.com/nar/article/53/3/gkae1247/7928523 by Biblioteca Universidad de Sevilla user on 27 February 2025 6 Nucleic Acids Research , 2025, Vol. 53, No. 3 added to supernatant (5 ×concentrated loading dye: 25 mM Tris-HCl pH 6.8, 25% glycerol, 10% SDS, 50 mM dithiothreitol and 0.05% bromophenol blue). After 5 min of denaturation at 95 ◦C, the samples were separated on 15% SDS-PAA gels and electroblotted onto nitrocellulose membranes (Amersham). Radioactive ink was added to the corners of the membrane to facilitate the overlay of the size markers. Radioactive signals were detected using a Typhoon FLA 9500 (GE Healthcare). After detection of the radioactive signal, the membranes were blocked in 5% skim milk in T-PBS for 1 h and subsequently used for western blotting using anti-FLAG antibodies as described above. CLIPseq experiment Three biological replicates of wild type (WT) and 3xFLAGPatR strain were cultivated in 200 ml of BG11 supplemented with the appropriate antibiotics to a concentration of 2.5 μg chlorophyll / ml. Heterocyst differentiation was induced as described above. After 24 h of cultivation, cells were irradiated with UV light (1.6 J / cm 2 ) to crosslink the proteins to their interacting RNA. The purification of the RNA–protein complexes was performed as explained above for the PNK assay with some modification. No Ribolock RNase Inhibitor was used during the lysis steps to allow the endogenous RNases to perform a mild digestion of the RNAs not protected by the RBP. After incubation of the clear lysates with 100 μl of antiFLAG M2 magnetic beads (Sigma), the beads were washed three times with high salt NT-P buffer and twice with cold NT-P buffer. To validate the progress, 20 μl of beads were analyzed following the PNK assay protocol. The other 80 μl of beads were resuspended in 50 μl of TBS buffer supplemented with 0.5 μg / μl FLAG peptide (Sigma) and 2 μl of Ribolock RNase Inhibitor for elution of the RNA–protein complexes. Beads were incubated at 4 ◦C for 15 min, while shaking at 900 rpm. Proteins were digested in 400 μl proteinase K buffer (50 mM Tris-HCl, pH 6, 5 mM EDTA and 0.5% SDS) with 400 μg of proteinase K for 30 min at 37 ◦C. Another 400 μl of proteinase K buffer + 9 M urea was added for an additional incubation of 30 min at 37 ◦C. RNA fragments were isolated using phenol-chloroform and chloroform extractions. RNA was later treated with 2 U of TURBO DNase (Life Technologies) to eliminate DNA traces, followed by RNA cleanup using RNA Clean & Concentrator columns (Zymo Research). Synthesis of complementary DNA (cDNA) was facilitated by SMARTScribe reverse transcriptase (TaKaRa), which adds the loop adapter at the 3  -end in the first-strand synthesis step and the SMART oligo at the 5  -end in the subsequent template switching and extension step. The loop adapter ( Supplementary Table S2 ) was prepared as described ( 50 ), and annealing of the loop adapter to RNA was achieved by incubating both at 72 ◦C for 3 min and 42 ◦C for 2 min. All components were added to this mixture, including the SMART oligo ( Supplementary Table S2 ), 100 U SMARTScribe reverse transcriptase (TaKaRa) and 40 U of Ribolock RNase inhibitor, and the reaction was incubated at 42 ◦C for 90 min and 70 ◦C for 10 min, followed by cooling to 12 ◦C. After cDNA synthesis, the samples were excised from 3% Nusieve agarose gels and purified with the NucleoSpin Gel and PCR Clean-up Kit (Macherey-Nagel), using NTC buffer to solubilize the gel slices. The cDNA was amplified in 22 PCR cycles with Phusion high-fidelity DNA polymerase (New England BioLabs) as described in ( 50 ) using Illumina 8-nt index primers for both sides, and PCR samples were purified using the Agencourt AMPure XP system (Beckman Coulter). All individual amplified cDNA libraries were pooled in equimolar ratios and size selected between 100 and 320 bp using the Pippin Prep system (Sage System) and subsequently purified using the Agencourt AMPure XP system (Beckman Coulter). The quality of RNA and DNA was analyzed on a Fragment Analyzer parallel capillary electrophoresis system (Agilent). Sequencing of libraries was performed on a NextSeq 2000 sequencer with a 100 bp single-end read mode at the Core Unit Systems Medicine of the University of Würzburg. Quality control analysis of the reads was carried out using FASTQC ( https://www.bioinformatics.babraham.ac.uk/ projects/ fastqc/ ). Adapter primers were removed using cutadapt ( https:// cutadapt.readthedocs.io/ en/ stable/ ). Reads were mapped to the Nostoc genome using bowtie2 in mode ‘very-sensitive-local’ ( 51 ). Bam files containing only aligned reads were used for peak calling using PEAKachu ( https: // github.com/ tbischler/ PEAKachu ). Only peaks with a log 2 fold change > 2.3 and fdr < 0.0001 were used for downstream analysis. For visualization purposes, the number of reads aligning to a position of the genome was normalized using the number of mapped reads of each library. Bedgraph files were generated by BamCoverage ( 52 ). Growth assay and time-course expression of 3xFLAG-PatR Cells from liquid Nostoc cultures, patR and patR+ were grown in BG11 medium under standard conditions as explained above. After 5 days, they were collected and resuspended in BG11 0 at OD 750 = 0.1. Five-fold serial dilutions were prepared and 10 μl of each dilution was plated on BG11 0 plates containing ammonium (2.5 mM NH 4 + and 12 mM N-tris (hydroxymethyl) methyl-2-aminoethanesulfonic acid-NaOH buffer (pH 7.5), nitrate (17 mM NO 3 −) or lacking combined nitrogen (N 2 ). Growth was analyzed after 13 days of incubation at 30 ◦C. To analyze heterocyst differentiation by optical microscopy, liquid cultures of Nostoc , patR and patR + growing photoautotrophically in BG11 under standard conditions were harvested, washed and resuspended in media without combined nitrogen source (BG11 0 ) as described above. After 24 h of cultivation, samples of patR were taken and heterocysts were stained with alcian blue as previously described ( 53 ). After 7 days of cultivation, visualization of heterocysts was not possible in patR because this strain did not grow diazotrophically. Liquid cultures of the strain 3xFLAG-PatR were grown in BG11 under standard conditions, cells were harvested by centrifugation, and heterocyst differentiation was induced as explained above. Samples were taken at different time points after combined nitrogen removal. Cells were mechanically lysed as explained above and 30 μg of total soluble protein was separated on a 15% PAA-SDS gel and transferred to a nitrocellulose membrane. The membrane was stained with Ponceau Red [0.1% ( w / v ) in 5% acetic acid] as loading control, blocked in 5% skim milk in T-PBS for 1 h and subsequently used for western blotting using anti-FLAG antisera as described above. Microscopy and image analysis Filaments stained with alcian blue were visualized using an Olympus BX60 microscope. Fluorescence of Nostoc filaments carrying plasmid pMBA118 growing for 5 days on top of soDownloaded from https://academic.oup.com/nar/article/53/3/gkae1247/7928523 by Biblioteca Universidad de Sevilla user on 27 February 2025 Nucleic Acids Research , 2025, Vol. 53, No. 3 7 lidified BG11 or BG11 0 (nitrogen-free) medium supplemented with appropriate antibiotics was analyzed by a Nikon CFI Plan apochromat VC 60 ×1.4 oil immersion objective and DIC N2 condenser attached to a Nikon Eclipse T1 confocal microscope. The samples were excited at 488 nm and the fluorescent emission was filtered by a HQ filter set (green fluorescent protein (GFP), 500–550 nm window) and a Cy-5 filter set (photosynthetic pigment autofluorescence, 663–738 nm window). For quantification purposes, all images were taken with the same settings. Fluorescence of WT and ΔpatR filaments carrying plasmids pSAM301 or pSAM270 growing for 5 days on top of solidified BG11 or BG11 0 (nitrogenfree) medium were analyzed using a Leica HCX PLAN-APO 63 ×1.4 NA oil immersion objective attached to a Leica TCS SP2 laser-scanning confocal microscope. Samples were excited at 488 nm by an argon-ion laser and the fluorescent emission was monitored by collection across windows of 500–538 nm (GFP) and 630–700 nm (photosynthetic pigment autofluorescence). For quantification purposes, the WT and ΔpatR strains were grown in the sample plate and images were taken with the same settings. Images were analyzed with Fiji ( 54 ). Phylogenetic analysis of PatR and conservation of patR promoter sequences A BLASTP search was performed using default parameters and the Nostoc PatR protein sequence (WP044521042.1) as a query. Because 1658 homologs were found, we decided to use only 107 homologs from genomes associated with reference cyanobacteria ( 55 ) for the downstream analysis. A multiple alignment was performed using ClustalW. The best distance model was selected using MEGAX ( 56 ). The evolutionary history was inferred by using the Maximum Likelihood method and Jones et al . w / freq. model ( 57 ). A total of 1000 iterations of bootstrapping were carried out to infer the confidence of the branches. Initial tree(s) for the heuristic search were obtained automatically by applying Neighbor-Joining and BioNJ algorithms to a matrix of pairwise distances estimated using the JTT model, and then selecting the topology with the higher log likelihood value. A discrete Gamma distribution was used to model evolutionary rate differences between sites (five categories; +G, parameter = 1.1415). The rate variation model allowed for some sites to be evolutionarily invariable ([+I], 1.40% sites). Structural analysis Alphafold prediction for PatR (AF-Q8YWB5-F1), Slr1122 (AF-P72645-F1) and PMM0486 (AF-Q7V2J0-F1) were downloaded from the AlphaFold Protein Structure Database. Sequences for the conserved domains of PatR (residues 70– 158 and residues 198–279, respectively) were folded by the ESMFold method and compared with the structure of other proteins through the Foldseek webserver( 58 ). The visualization of the PatR structure, the calculation of the Coulombic electrostatic potential (ESP) and the superimposition of structural homologs were performed by ChimeraX ( 59 ). Results Nostoc R-DeeP to fractionate multimeric complexes of a membrane-rich cyanobacterium Triplicate cultures of Nostoc were grown for 26 h in medium without a combined nitrogen source to obtain filaments containing vegetative cells, pro-heterocysts, and mature heterocysts for R-DeeP / GradR analysis ( 24 ,26 ). In this technique, all macromolecular complexes in the cell are fractionated in a density gradient followed by high-resolution proteomics of 19 fractions taken from each gradient. Gradients of RNasetreated samples were compared against those of untreated samples in triplicates. If a protein was part of an RNA–protein complex, the complex should be disassembled after RNase treatment and a shift of the protein position in the gradient should be observed (RNA-associated proteins) (Figure 1 A). A critical step in the success of the approach is the preservation of the native protein–protein and RNA–protein complexes. Since cyanobacteria, as photosynthetic organisms, are rich in pigments, the fractionation of their macromolecular complexes resulted in colorful gradients reflecting the abundance of large photosynthetic complexes (Figure 1 B). We used 10–40% sucrose gradients, which have a resolution between 0 and 900 KDa. If the Nostoc proteome was completely disassembled into monomeric proteins, no proteins should be detected in fractions beyond fraction 5 ( Supplementary Figure S1 A). However, proteins were measured in higher fractions ( Supplementary Figure S1 B) and SDS-PAA gels stained with Coomassie blue showed highly expressed proteins, such as RuBisCO or glutamine synthetase, appearing in higher fractions, consistent with their association into multisubunit complexes. Furthermore, the overall distribution of proteins did not appear to be affected by RNase treatment, excluding unspecific effects of treatment ( Supplementary Figure S1 C and D). In order to test the integrity of the RNA in the untreated samples, RNA was extracted from the different fractions and separated in denaturing urea-PAA gels. Several intact highly expressed RNAs were directly identified (Figure 1 C). We analyzed by northern blot hybridization the distribution of six previously known non-coding RNAs; 5S ribosomal RNA (rRNA), RnpB, 4.5S, Yfr1 ( 7 ), NsiR8 ( 6 ) and NsiR1 ( 8 ). The position of an RNA in the gradients did not depend on its size, but rather on its interaction with large protein complexes. This fact can be observed in the different distributions of the approximately equal-sized RNAs 4.5S (signal recognition particle RNA) and 5S rRNA (Figure 1 D). The detection of NsiR1 and NsiR8, two sRNAs expressed only in heterocysts, confirmed the capacity of detecting putative heterocyst-specific ribonucleoprotein complexes. This was a relevant aspect regarding the sensitivity of the analysis, because after heterocyst differentiation, there is on average only one heterocyst per 15 vegetative cells. Finally, as a positive control, the change in the position of the ribosomal RBP RplA after RNase treatment was tested by western blot (Figure 1 E). As a negative control, we generated a strain expressing the master transcription factor of heterocyst differentiation HetR ( 60 ) fused to a 3xFLAG epitope tag under control of its native promoter. We prepared gradients with this strain following the same procedure as for the WT. The absence of a shift for HetR -3xFLA G confirmed the specificity of our approach (Figure 1 E). Tandem mass tags ( 61 ) were used for the parallel detection and relative quantification of proteins by MS. We pooled fractions 1 and 2; thus, 18 samples were measured per gradient. These measurements identified 2638 proteins in at least one pair of control-RNase-treated gradients, which represent 44% of the Nostoc proteome. Out of 139 proteins previously defined as being expressed exclusively in heterocysts ( 5 ), 40 were detected in this study (29%).The overall in-gradient distribution of protein profiles showed correlation scores of 0.92– Downloaded from https://academic.oup.com/nar/article/53/3/gkae1247/7928523 by Biblioteca Universidad de Sevilla user on 27 February 2025 8 Nucleic Acids Research , 2025, Vol. 53, No. 3 2 CA B D Gradient Control + RNases MS Normalized protein amount Control + RNases Fractions 26h N Size (nt) RplA snoitcarFsnoitcarF Fractions Fractions 123456 78910111213141516171819 700 600 500 400 300 200 100 RnpB 4.5S tRNAs 5S 1234 56 78 910111213141516171819 5S RnpB 4.5S Yfr1 NsiR8 NsiR1 HetR-3xFLAG 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 40 35 25 40 35 25 Control + RNases 40 35 40 35 EKDa KDa non-RBP RBP RBP Fractions Figure 1. Nostoc R-DeeP experiment. (A) Overview: Lysates from heterocyst-containing filaments are fractionated in sucrose density gradients f ollo w ed by MS proteomics. Gradients from RNase-treated and untreated samples are compared. An RNA–protein complex should dissociate after RNase treatment and a shift in the position of the respective proteins in the gradient should be observed. (B) Gradient tubes after centrifugation and fractions after elution in collection tubes. A representative pair of gradients is shown. The different colors correspond to abundant pigment-protein complexes, containing orange carotenoids, blue ph y cobilins or green chlorophylls. (C) Separation of RNA from 19 fractions on a 10% urea-PAA gel. RNA was extracted from 300 μl of each fraction of one untreated gradient out of three. All extracted RNA was loaded onto the gel. The position of abundant RNA species is indicated for orientation. The position of size markers is indicated on the left. (D) Distribution of selected non-coding RNAs analyzed by northern blot. RNA was extracted from 300 μl of each fraction of a representative untreated gradient, separated on a 10% urea-PAA gel and after blotting h ybridiz ed with radiolabeled probes f or the indicated non-coding RNAs. (E) Distribution of RplA and HetR-3xFLAG along untreated and treated representative gradients. Western blot analyses were performed using 35 μl of each fraction separated on a 15% SDS-PAA gel and antisera against RplA or the FLAG epitope. The position of size markers (in kilodaltons) is indicated on the right. The results from one of three gradient pairs are displayed. 0.94 between the replicates, indicating good reproducibility ( Supplementary Figure S2 ). To compare the distribution of proteins with different expression levels, an in-gradient normalization was performed, so that the sum of each protein along a gradient was set to 100%. The overall quality of the approach can be observed in the co-sedimentation of proteins belonging to essential ribonucleoprotein complexes such as 50S and 30S ribosomal subunits, RNA polymerase or one of the Nostoc CRISPR systems ( 62 ) (Figure 2 A–D). In addition to the housekeeping sigma factor SigA, three alternative sigma factors SigE, SigD and SigI were detected (Figure 2 D). SigA showed a clear co-sedimentation pattern with the RNA polymerase core while the alternative sigma factors did not. Only a small subpopulation of the alternative sigma factors may interact with the RNA polymerase core. In the case of SigE, a heterocyst-specific factor ( 63 ), only 2% of the overall protein amount was detected in each of the fractions were the RNA polymerase core co-sedimented ( Supplementary Data S2 ). To solubilize membrane proteins, we used n-dodecyl β-Dmaltoside, a mild detergent commonly used to study photoDownloaded from https://academic.oup.com/nar/article/53/3/gkae1247/7928523 by Biblioteca Universidad de Sevilla user on 27 February 2025 Nucleic Acids Research , 2025, Vol. 53, No. 3 9 RplA (Alr5301) RplB (All4212) RplC (All4215) RplD (All4214) RplE (All4203) RplF (All4201) RplI (All0579) RplJ (Alr5302) RplK (Alr5300) RplL (Alr5303) RplM (All4188) RplN (All4205) RplO (All4198) RplQ (All4190) RplR (All4200) RplS (Alr5297) RplT (Alr3428) RplV (All4210) RplW (All4213) RplX (Asl4204) RplY (Alr4785) RpmA (Asl0146) RpmB (Asl2630) RpmC (Asl4207) RpmG (Asl4452) RpmI (Asr3427) RpmJ (Asl4194) A 080 % Protein 19 18 17 16 15 14 13 12 10 11 9 8 7 6 5 4 3 1_2 Fractions Rps1_1 (Alr1078) Rps1_2 (All0136) RpsB (All4792) RpsC (All4209) RpsD (Alr2737) RpsE (All4199) RpsF (All4802) RpsG (All4339) RpsH (All4202) RpsI (All4187) RpsJ (All4336) RpsK (All4192) RpsL (All4340) RpsM (All4193) RpsN (All3969) RpsO (Asl0749) RpsP (Asr1953) RpsQ (Asl4206) RpsR (Asl4451) RpsS (Asl4211) RpsT (Asr1592) RpsU1 (Asr0742) B 040 % Protein 19 18 17 16 15 14 13 12 10 11 9 8 7 6 5 4 3 1_2 Fractions C 030 % Protein RpoA (All4191) RpoB (Alr1594) RpoC1 (Alr1595) RpoC2 (Alr1596) RpoZ (Asr4648) SigA (All5263) SigD (Alr3810) SigE (Alr4249) SigI (All2193) 19 18 17 16 15 14 13 12 10 11 9 8 7 6 5 4 3 1_2 Fractions D 030 % Protein 19 18 17 16 15 14 13 12 10 11 9 8 7 6 5 4 3 1_2 Fractions All1472 All1474 Csx7 (All1475) Csx10 (All1477) Cas10 (All1479) PecA (Alr0524) PecB (Alr0523) PecC (Alr0525) ApcA (Alr0021) ApcB (Alr0022) ApcC (Asr0023) ApcF (All2327) ApcE (Alr0020) CpcA (Alr0529) CpcB (Alr0528) CpcC (Alr0530) CpcD (Asr0531) CpcG2 (Alr0535) CpcL (Alr0536) CpcG1 (Alr0534) CpcG4 (Alr0537) ApcD (All3653) PsbA1 (Alr4866) PsbA3 (Alr4592) PsbB (All0138) PsbC (Alr4291) PsbD (Alr4290) PsbE (Asr3845) PsbH (Asl0846) PsbL (Asr3847) PsbO (All3854) PsbP (All3076) PsbQ (All1355) PsbU (Alr1216) PsbW (All0801) PsbY (Asr1025) Psb27(All1258) Psb29 (All0646) PsaA (Alr5154) PsaB (Alr5155) PsaC (Asr3463) PsaD (All0329) PsaF (All0109) PsaJ (Asl0108) PsaE (Asl4319) PsaK2 (Asr5289) PsaM (Asr4657) PsaL1 (All0107) E 050 % Protein 19 18 17 16 15 14 13 12 10 11 9 8 7 6 5 4 3 1_2 Fractions 19 18 17 16 15 14 13 12 10 11 9 8 7 6 5 4 3 1_2 Fractions NdhA (Alr0023) NdhB (All4883) NdhD (Alr3957) NdhF (Alr3956) NdhG (Alr0225) NdhH (Alr3335) NdhI (Alr0024) NdhJ (All3840) NdhK (All3841) NdhM (All1732) NdhN (Alr4216) NdhO (Asr4321) NdhQ NdhS (Asr0654) NdhV (Alr3297) H 030 % Protein K 19 18 17 16 15 14 13 12 10 11 9 8 7 6 5 4 3 1_2 Fractions CoxA1 (Alr0951) CoxB1 (Alr0950) CoxA2 (Alr2515) 030 % Protein SdhA (All2970) SdhB (All0945) SdhC (All3341) I 19 18 17 16 15 14 13 12 10 11 9 8 7 6 5 4 3 1_2 Fractions 030 % Protein J 19 18 17 16 15 14 13 12 10 11 9 8 7 6 5 4 3 1_2 Fractions PetA (All2452) PetB (Alr3421) PetC (All2453) 025 % Protein L 040 % Protein AtpA (All0005) AtpB (All5039) AtpC (All0004) AtpE (All5038) AtpD (All0006) AtpF (All0007) AtpG (All0008) AtpI (All0010) AtpH (Asl0009) 19 18 17 16 15 14 13 12 10 11 9 8 7 6 5 4 3 1_2 Fractions 035 % Protein 19 18 17 16 15 14 13 12 10 11 9 8 7 6 5 4 3 1_2 Fractions UreB (Alr3668) UreC (Alr3670) G F 025 % Protein NifH (All1455) NifD (All1454) NifK (All1440) 19 18 17 16 15 14 13 12 10 11 9 8 7 6 5 4 3 1_2 Fractions Figure 2. In-gradient distribution of proteins that are part of large macromolecular comple x es in Nostoc . Heatmap representation of the mean normalized abundance of each detected protein in three untreated gradients. The sum of a protein abundance along the gradient is normalized to 100%. Proteins that are part of RNA–protein complexes are shown as follows: (A) 50S ribosomal subunit, (B) 30S ribosomal subunit, (C) RNA polymerase, (D) Subtype III-D CRISPR complex. Proteins that belong to multiprotein complexes involved in photosynthesis, nitrogen metabolism or the respiratory chain are also shown: (E) Photosynthetic complexes, (F) Nitrogenase, (G) Urease, (H) NADH dehydrogenase, (I) Succinate dehydrogenase, (J) Cytochrome b 6 f, (K) Terminal respiratory oxidases, (L) ATP synthase. Components of each complex with similar distributions are framed in red. In panel (E), a different color palette for the frames is used and a photo of the gradients is displayed above the heatmap to show the colors of the different fractions. Ph y cobilisome components containing ph y coerythrocy anin, alloph y cocy anin or ph y cocy anin are framed in orange, black and blue, respectiv ely. Co-sedimentation of the PSII and PSI core is also highlighted in dark green and light green, respectively. Downloaded from https://academic.oup.com/nar/article/53/3/gkae1247/7928523 by Biblioteca Universidad de Sevilla user on 27 February 2025 16 Nucleic Acids Research , 2025, Vol. 53, No. 3 A D C WT ΔpatR ΔpatR+ NH4 + NO3 - NO3 - N2 WT ΔpatR ΔpatR+ WT ΔpatR ΔpatR+ B ΔpatR N2 (24h) WT ΔpatR ΔpatR+ ΔpatR N2 (7 days) -10 10 ESP ΔpatR + PnsiR1 N2 E NO3 - NO3 -N2 TWTWΔpatR PnsiR1PhetR ΔpatR Figur e 6. P atR regulates heterocy st differentiation. (A) Gro wth on solid media. Cells from liquid cultures of the indicated strains gro wn in the presence of nitrate were collected and resuspended in BG11 0 at an OD 750 = 0.1. Fivefold serial dilutions were prepared and 10 μl of each dilution plated on BG11 0 plates containing ammonium (NH 4 + ), nitrate (NO 3 −) or lacking combined nitrogen (N 2 ). Pictures were taken after 13 da y s of incubation at 30 ◦C. (B) Presence of heterocyst-like cells in patR . Upper panel: bright-field image of filaments of patR growing in liquid medium 24 h after removal of combined nitrogen. Heterocysts and heterocyst-like cells are stained with alcian blue. Bottom panel: Growth in liquid media of indicated strains 7 days after combined nitrogen remo v al. (C) Confocal fluorescence images of filaments from WT and patR strains carrying nsiR1 and hetR cell-type specific promoter fusions to gfp. Filaments were grown on top of medium containing nitrate (NO 3 −) or without a combined nitrogen source (N 2 ). Green channel (GFP fluorescence) and magenta channel (autofluorescence) are shown. Higher magnification (D) . Scale bars, 25 μm. (E) PatR str uct ure predicted by Alphafold (AF-Q8YWB5-F1). Left panel: Coulombic ESP of PatR str uct ure calculated by ChimeraX. Positively charged surfaces (blue) and negatively charged surfaces (red) are shown. Right panel: core of the PatR protein str uct ure. The two tandem conserved OB-fold-like domains are highlighted; domain 1 in red (residues 70–158) and domain 2 in blue (residues 198–279). Downloaded from https://academic.oup.com/nar/article/53/3/gkae1247/7928523 by Biblioteca Universidad de Sevilla user on 27 February 2025 Nucleic Acids Research , 2025, Vol. 53, No. 3 17 A WT 3xFLAG-PatR KDa 70 55 40 35 15 25 10 WB WT 3xFLAG-PatR PNK B 0 1000 2000 3000 4000 5000 Axrt1500rla 000750006500055 PatR_1 PatR_3 PatR_2 WT1 WT2 WT3 PatR_1 PatR_3 PatR_2 WT1 WT2 WT3 yfr1 Genomic position Genomic position Normalized reads C 4271500 4272500 4273500 50 40 30 20 10 0 Normalized reads 134567891112 hetF PatR_1 PatR_3 PatR_2 WT1 WT2 WT3 Genomic position 2232500 2233500 600 500 400 300 200 100 0 D Normalized reads all1871 alr1870 E 1368000 1369000 all1164 ffs asl1165 250 200 150 100 50 0 Normalized reads Genomic position PatR_1 PatR_3 PatR_2 WT1 WT2 WT3 201 nsiR1 Figure 7. RNA targetome of PatR. (A) PNK assay of 3xFLAG-PatR (positive control) and WT (negative control) samples used for CLIP-seq. Western blot analysis (WB) and autoradiography (PNK) are shown. Size markers are indicated on the left. ( B –E ) Distribution of reads mapped in relevant regions of RNA obtained from three biological replicates of WT (blue) or 3xFLAG-PatR (green). The genomic regions of yfr1 (B) , nsiR1 array (C) , all1871 (D) and ffs (E) are shown. Genes are represented by arrows together with the gene name or ID. Genomic coordinates denote the position on the Nostoc chromosome. The scale indicates the number of mapped reads per nucleotide position normalized by the total number of reads mapped in each library. The individual nsiR1 instances are displayed in panel (C) with numbers 1–12. seem striking, it has also been observed in the GradR / R-DeeP analyses of Salmonella or human samples ( 24 ,25 ). One possible explanation is the precipitation of a protein when its RNA counterpart is not present. In the heterologous purification of some RBPs, such as CRISPR proteins, it was observed that the coexpression of their target RNA was required to stabilize the proteins in soluble form ( 81 ). However, the right-shift could also reflect a regulatory mechanism (see the discussion below about Vipp1). The prediction of RBPs usually relies on the search for conserved RNA-binding domains. However, the ability to bind RNA may rely on more basic biochemical properties. As an example, previously known RBPs are usually enriched in positively charged residues that may facilitate their interactions with the negatively charged RNAs, a tendency that is observed in Figure 3 B. In order to consider subtler patterns, we used a modified version of TriPepSVM ( 38 ), an algorithm that is trained with known RBPs or non-RBPs sequences, according to the databases, and considers their tripeptide content for RBP prediction. The advantage of this method is the ability to predict new RBPs whose RNA-binding sites could be located in intrinsically disordered regions. In fact, we observed a slight tendency of the characteristic RBP tripeptides to be located in disordered regions in Nostoc (Figure 3 C). This tendency was previously reported for the human proteome, but not for the proteomes of E. coli and S. typhimurium ( 38 ). The importance of disordered regions in RBPs is a recent field of research ( 82 ) since some RBPs are assembled in liquid-liquid phase separation droplets ( 83 ), including an RNA helicase in cyanobacteria ( 84 ). There was only a partial overlap between the shifting proteins and the prediction of TriPepSVM (Figure 3 D), indicating complementarity between the two methods. The shifting proteins not predicted as RBP could be part of RNA–protein complexes but may not bind RNA themselves. This fact could be supported by the low pI of the proteins from this group (Figure 3 B). The presence of previously known RBP with no apparent significant shift has also been reported ( 24 ). Weak and transient RNA–protein interactions or the R-DeeP resolution limits ( Supplementary Figure S5 ) could be the reasons for these false negatives. Because the use of size exclusion chromatography in SEC-seq has an improved resolution ( 22 ), it may be a good approach for future research on RBPs appearing in the low-molecular-weight fractions. Downloaded from https://academic.oup.com/nar/article/53/3/gkae1247/7928523 by Biblioteca Universidad de Sevilla user on 27 February 2025 18 Nucleic Acids Research , 2025, Vol. 53, No. 3 Some of the proteins not significantly shifting but predicted as RBPs might be true RBPs. Two metabolic enzymes related to nitrogen metabolism, CphA1, cyanophycin synthetase ( 85 ), and UreB, an urease subunit ( 86 ), had a visible slight shift in our gradients just below our strict cutoff, but in contrast had very good SVM scores not only in Nostoc but in all cyanobacterial proteins tested (Figure 3 E and F). They could be false positives predicted by TriPepSVM, but a recent report found both proteins as interacting partners of YlxR, an RBP in Synechocystis that interacts with the RNA subunit of RNase P ( 87 ) . The detection of new RNA-binding capacities in many metabolic enzymes has also been reported in S. typhimurium ( 80 ), and may represent a direct link between responses to nutrient availability and post-transcriptional gene regulation. All four proteins with significant shifts and high SVM scores (Alr2890, Vipp1, PrpA and Asl3888) had the ability to interact with RNA in vivo (Figure 4 C–E), confirming the specificity of our combined approach. However, we also obtained very clearly labeled bands for PatU3 and PatR (Figure 4 I and J), two proteins with low SVM scores in most of the cyanobacterial proteomes. This result suggests that some of the 282 proteins that shifted but were not predicted to be RBPs (Figure 3 D) could be true RBPs. It is important to note that SVMs are trained with current knowledge and, therefore, totally new domains or interaction patterns could be missed. The newly validated RBPs have previously annotated functions related to essential physiological processes such as photosynthesis and heterocyst differentiation. The Alr2890 homologs are strictly conserved in cyanobacteria and plant chloroplasts. Although Alr2890 contains a putative S4 RNAbinding domain and was therefore tagged as ‘RNA-binding’ in the UNIPROT database, Sll1252, its close homolog in Synechocystis , was found in PSII preparations, is necessary for the proper activity of PSII ( 69 ) and is involved in the coordination of electron transport between plastoquinone and the cytochrome b 6 f complex ( 88 ). The authors hypothesized a regulatory function for Sll1252 ( 69 ), but the exact mechanism has remained unknown. Here, we confirm that the Nostoc ortholog Alr2890 is an RBP, suggesting that it is bifunctional. The case of Vipp1 (All2342) is even more striking, since it was the only validated RBP that had a shift toward higher fractions after RNase treatment (Figure 4 F). Oligomers of Vipp1 are essential for maintaining the integrity of the thylakoid membrane ( 70 ) and Vipp1 was recognized as a bacterial ancestor of the eukaryotic ESCRT-III membrane remodeling system ( 89 ). In cyanobacteria, the protein continuously moves between a diffused fraction and discrete foci at the thylakoid membranes ( 90 ), which are more abundant after high light stress ( 91 ) and consist of Vipp1 oligomers ( 70 ). The trigger that causes the polymerization of the protein is unknown. Based on our observation that Vipp1 migrated in our gradients after RNase treatment to multiple fractions corresponding to higher molecular weight, it is tempting to speculate that it is the presence of an RNA that sequesters Vipp1 in monomeric form. After the removal of this hypothetical RNA (or degradation in our treatment), Vipp1 is able to form oligomers. Proteins involved in heterocyst differentiation were also validated as new RBPs. PrpA was first described as a phosphatase that affects heterocyst differentiation ( 71 ). Here, we have demonstrated the in vivo RNA-binding capacity of PrpA, which fits the report that a closely related homolog in T richor - mus variabilis ATCC29413 is part of an RNA repair system ( 92 ). In contrast, no link to RNA has been reported for PatU3, a protein expressed specifically in heterocysts ( 93 ) that is involved in the regulation of heterocyst formation ( 12 ) and cell division ( 12 , 94 , 95 ). Our mutation of patR highlighted the fundamental role that this RBP plays in the heterocyst differentiation process (Figure 6 ). The absence of patR caused an alteration of heterocyst patterning and differentiation (Figure 6 ). patR filaments had a higher tendency to differentiate heterocysts, since even growing in plates with nitrate, a condition in which the differentiation process should not occur, the strain had a higher expression of nsiR1 and hetR (Figure 6 C). After combined nitrogen removal, this tendency for cell differentiation generated an unnaturally high number of heterocyst-like cells (Figure 6 C), sometimes control of the patterning is altered and generates doublets and triplets of heterocyst-like cells (Figure 6 D). It is intriguing that this phenotype of Mch was reported for cells lacking PatU3 ( 93 ), also characterized as RBP here. This higher frequency of heterocyst-like cells did not improve diazotrophic growth, but rather resulted in an inability to grow diazotrophically at all. An explanation for this observation is the inability of the heterocyst-like cells to reach the final steps of maturation, including the assembly of functional nitrogenase. This explanation is supported by the absence of polar granules, an indicator of mature functional heterocysts. The CLIP-seq data provide insight into the possible regulatory mechanism of PatR, although the presence of multiple RNA targets complicates the interpretation. PatR seems to be an important hub for the interaction with several different sRNAs ( Supplementary Data S10 ). Among them, Yfr1 was the statistically most enriched sRNA (Figure 7 B). Yfr1 is highly expressed in all cells of the filaments irrespective of the nitrogen status and, in Nostoc, regulates the translation of mRNAs encoding enzymes for the synthesis and remodeling of the cell wall ( 7 ). In fact, an altered expression of yfr1 generates fragmented filaments that cannot support the flow of nitrogen compounds between vegetative cells and heterocysts ( 7 ), a fragmentation that is also observed in the patR strain at longer time points after combined nitrogen removal. The interaction between Yfr1 and PatR is also supported by the presence of Yfr2 in the PatR targetome ( Supplementary Data S10 ), a nitrogen-regulated sRNA that traps Yfr1 and prevents its regulatory effects ( 7 ,96 ). The identification of the heterocyst-specific sRNA NsiR1 (Figure 7 C) among the interacting partners of PatR may explain the specific function of PatR in the regulation of the heterocyst differentiation process. nsiR1 is specifically expressed in young heterocysts, which could be coherent with a higher expression of patR in these cells (Figure 5 D) and, given that this sRNA modulates heterocyst differentiation through its interaction with important nodes of the heterocyst regulatory network ( 8 ,11 ), it is tempting to speculate that altered expression of NsiR1 may generate a deregulation of the whole cell differentiation process. The interaction between Yfr1 or NsiR1 and PatR detected by CLIP-seq is supported by different experimental data. First, there is a clear correlation in the distribution of the RBP (Figure 4 J) and these two sRNAs (Figure 1 D) in our density gradients. Second, the small shift of PatR after RNase treatment (Figure 4 J) suggests an interaction with small RNAs. Furthermore, these interactions seem to be specific because other highly transcribed non-coding RNAs with similar distributions in the gradients, such as the 4.5S Downloaded from https://academic.oup.com/nar/article/53/3/gkae1247/7928523 by Biblioteca Universidad de Sevilla user on 27 February 2025 Nucleic Acids Research , 2025, Vol. 53, No. 3 19 RNA (Figure 1 D), were not enriched in our CLIP-seq data (Figure 7 E). The strong phenotypic effects suggest PatR as a key regulator of heterocyst maturation. Interestingly, there is also a parallel to HetR, the transcriptional master regulator of heterocyst differentiation. The hetR gene is transcribed from four discernible start sites ( 97 ). Apparently, patR expression decreased after combined nitrogen removal (Figure 5 A and B), but our fluorescence confocal analysis of single cells revealed a more complex picture (Figure 5 C). Expression of this gene appeared to be repressed from early stages of heterocyst differentiation only in cells that were not undergoing the cell differentiation process (Figure 5 D). Although we analyzed the transcription from the first transcriptional start site (TSS) of patR (Figure 5 A), four more transcriptional start sites were identified for this gene ( 72 ) ( Supplementary Figure S9 ). The presence of multiple transcriptional start sites, sometimes with paradoxical regulations that create complex expression patterns along the filament, has also been observed for other differentiation-relevant genes in Nostoc ( 98 ) . In the case of patR transcriptional start sites, the usual pattern for TSS usage in heterocysts, i.e induction in the WT after nitrogen limitation but not in the hetR mutant, is not clearly detected ( 72 ). Therefore, the reduced transcription in the vegetative cells may be achieved by repression of the same TSSs that are used in the heterocyst, but by a specific repressor that is expressed only in the vegetative cells after nitrogen limitation. The wide phylogenetic conservation of PatR suggests that it is also important for non-heterocystous cyanobacteria. This result is coherent with a more general function that may involve its interaction with Yfr1, a universally conserved sRNA in cyanobacteria ( 6 ). The protein has a structural architecture that contains two OB-fold-like domains combined in tandem (Figure 6 E). A similar arrangement has been described for a ssDNA binding protein in Deinococcus r adiodur ans ( 99 ). Although the presence of PatR homologs in unicellular cyanobacteria could be paradoxical considering its effects on heterocyst differentiation, homologs of other proteins essential for heterocyst differentiation, such as HetR, also exist in non-heterocyst-forming cyanobacteria ( 100 ). PatR and its homologs probably have some essential function related to the physiology of all cyanobacteria, but the insertion of two additional α-helices in the homologs of heterocyst-forming cyanobacteria ( Supplementary Figures S7 and S8 ) may have allowed the particular regulatory role of PatR in heterocyst differentiation. The exact regulatory mechanism(s) involving PatR and its targets are highly interesting topics for future work. Data availability Source data are provided as a Source Data file. MS raw data is deposited at the ProteomeXchange Consortium ( http: //proteomecentral.proteomexchange.org ) via the PRIDE partner repository( 101 ) under the identifier: PXD050404. CLIPseq data is deposited in GEO under the identifier GSE280056. All code involving the bioinformatics workflow and the shiny app is available on Github ( https:// github.com/ manbrealv/ RDeeP-Nostoc ) and Zenodo ( https:// doi.org/ 10.5281/ zenodo. 14265455 ). A shiny app was programmed to facilitate the accessibility and visualization of the data: https://sunshine. biologie.unifreiburg.de/RDeePNostoc/. Supplementary data Supplementary Data are available at NAR Online. A c kno wledg ements The authors thank Prof. Enrique Flores and Prof. Antonia Herrero for sharing the ΔhetR CSSC2 strain. We also thank Marc Broghammer for help in cloning the Asl3888-3xFLAG fusion, Karsten Voigt and Edith Ams for help in the implementation of the shiny app on the webserver, Panagiota Arampatzi and Thorsten Bischler for sequencing the CLIP-seq RNA libraries and Marcus Ziemann for providing assistance in the visual representation of CLIP-seq data. Author contributions : M.B.A. designed the project. M.B.A., A.M.M-P ., M.M.S. and W .R.H. secured funding. The R-DeeP approach, PNK assays, phylogenetic analysis and construction of the mutants and reporter strains was performed by M.B.A. D.P ., C.P ., F .S. and M.M.S. performed and interpreted the proteomic analysis. Bioinformatic analysis of R-DeeP data was performed by F .S., D.P . and M.B.A. The modified version of TriPepSVM was implemented by H.R.R. and M.B.A. Interpretation of the R-DeeP data and SVM output was performed by M.B.A. and W.R.H. Growth analysis and microscopy experiments for patR were performed and interpreted by A.V., A.M.M-P. and M.B.A. CLIP-seq experiment was carried out by I.S., C.S. and M.B.A., and analyzed by M.B.A. M.B.A. and W.R.H. wrote the manuscript with the input of all authors. Funding Alexander von Humboldt-Stiftung (to M.B.A.); Deutsche Forschungsgemeinschaft [HE 2544 / 20-1 to W.R.H.]; Paul G. Allen Frontiers Group (to M.M.S.); Agencia Estatal de Investigación [PID2022-138128NB-I00, MCIN / AEI / 10.13039 / 501100011033, FEDER, European Regional Development Fund, UE to A.M.M-P]. We acknowledge support by the Open Access Publication Fund of the University of Freiburg. Conflict of interest statement None declared. References 1. Lyons, N.A. and Kolter, R. (2015) On the evolution of bacterial multicellularity. Curr. Opin. Microbiol., 24 , 21–28. 2. Zeng, X. and Zhang, C.C. (2022) The making of a heterocyst in cyanobacteria. Annu. Rev. Microbiol., 76 , 597–618. 3. Muro-Pastor, A.M. and Hess, W.R. (2012) Heterocyst differentiation: from single mutants to global approaches. Trends Microbiol., 20 , 548–557. 4. Flores, E. , Picossi, S. , Valladares, A. and Herrero, A. (2019) Transcriptional regulation of development in heterocyst-forming cyanobacteria. Biochim. Biophys. Acta Gene Regul. Mech., 1862 , 673–684. 5. Brenes-Álvarez, M. , Mitschke, J. , Olmedo-Verd, E. , Georg, J. , Hess, W.R. , V ioque, A. and Muro-Pastor, A.M. (2019) Elements of the heterocyst-specific transcriptome unravelled by co-expression analysis in Nostoc sp. PCC 7120. Environ. Microbiol., 21 , 2544–2558. 6. Brenes-Álvarez, M. , Olmedo-Verd, E. , V ioque, A. and Muro-Pastor,A.M. (2016) Identification of conserved and Downloaded from https://academic.oup.com/nar/article/53/3/gkae1247/7928523 by Biblioteca Universidad de Sevilla user on 27 February 2025 20 Nucleic Acids Research , 2025, Vol. 53, No. 3 potentially regulatory small RNAs in heterocystous cyanobacteria. Front. Microbiol., 7 , 48. 7. Brenes-Álvarez, M. , V ioque, A. and Muro-Pastor, A.M. (2020) The integrity of the cell wall and its remodeling during heterocyst differentiation are regulated by phylogenetically conserved small RNA Yfr1 in Nostoc sp. strain PCC 7120. mBio , 11 , e02599–19 8. Brenes-Álvarez, M. , V ioque, A. and Muro-Pastor, A.M. (2022) The heterocyst-specific small RNA NsiR1 regulates the commitment to differentiation in Nostoc . Microbiol. Spectr., 10 , e0227421. 9. Álvarez-Escribano, I. , Brenes-Álvarez, M. , Olmedo-Verd, E. , Georg, J. , Hess, W.R. , V ioque, A. and Muro-Pastor, A.M. (2021) NsiR3, a nitrogen stress-inducible small RNA, regulates proline oxidase expression in the cyanobacterium Nostoc sp. PCC 7120. FEBS J., 288 , 1614–1629. 10. Ionescu, D. , Voss, B. , Oren, A. , Hess, W.R. and Muro-Pastor, A.M. (2010) Heterocyst-specific transcription of NsiR1, a non-coding RNA encoded in a tandem array of direct repeats in cyanobacteria. J. Mol. Biol., 398 , 177–188. 11. Brenes-Álvarez, M. , Minguet, M. , V ioque, A. and Muro-Pastor,A.M. (2020) NsiR1, a small RNA with multiple copies, modulates heterocyst differentiation in the cyanobacterium Nostoc sp. PCC 7120. Environ. Microbiol., 22 , 3325–3338. 12. Xing,W .Y ., Liu,J., Zhang,J.Y., Zeng,X. and Zhang,C.C. (2022) A proteolytic pathway coordinates cell division and heterocyst differentiation in the cyanobacterium Anabaena sp. PCC 7120. Proc. Natl Acad. Sci. U.S.A., 119 , e2207963119. 13. V ideau, P. , Rivers, O.S. , Hurd, K. , Ushijima, B. , Oshiro, R.T. , Ende, R.J. , O’Hanlon, S.M. and Cozy, L.M. (2016) The heterocyst regulatory protein HetP and its homologs modulate heterocyst commitment in Anabaena sp. strain PCC 7120. Proc. Natl Acad. Sci. U.S.A., 113 , e6984–e6992. 14. Wagner, E.G. and Romby, P. (2015) Small RNAs in bacteria and archaea: who they are, what they do, and how they do it. Adv. Genet., 90 , 133–208. 15. Schuergers, N. , Ruppert, U. , Watanabe, S. , Nürnberg, D.J. , Lochnit, G. , Dienst, D. , Mullineaux, C.W. and Wilde, A. (2014) Binding of the RNA chaperone Hfq to the type IV pilus base is crucial for its function in Synechocystissp. PCC 6803. Mol. Microbiol., 92 , 840–852. 16. Holmqvist, E. and Vogel, J. (2018) RNA-binding proteins in bacteria. Nat. Rev. Micro., 16 , 601–615. 17. Fei, J. and Sharma, C.M. (2018) RNA localization in bacteria. Microbiol. Spectr., 6 , 10.1128. 18. Mulligan, M.E. , Jackman, D.M. and Murphy, S.T. (1994) Heterocyst-forming filamentous cyanobacteria encode proteins that resemble eukaryotic RNA-binding proteins of the RNP family. J. Mol. Biol., 235 , 1162–1170. 19. Mahbub, M. , Hemm, L. , Yang, Y. , Kaur, R. , Carmen, H. , Engl, C. , Huokko, T. , Riediger, M. , Watanabe, S. , Liu, L.N. , et al. (2020) mRNA localization, reaction centre biogenesis and thylakoid membrane targeting in cyanobacteria. Nat. Plants , 6 , 1179–1191. 20. Smirnov, A. , Förstner, K.U. , Holmqvist, E. , Otto, A. , Günster, R. , Becher, D. , Reinhardt, R. and Vogel, J. (2016) Grad-seq guides the discovery of ProQ as a major small RNA-binding protein. Proc. Natl Acad. Sci. U.S.A., 113 , 11591–11596. 21. Riediger, M. , Spät, P. , Bilger, R. , Voigt, K. , Ma ˇ cek, B. and Hess, W.R. (2021) Analysis of a photosynthetic cyanobacterium rich in internal membrane systems via Gradient profiling by sequencing (Grad-seq). Plant Cell , 33 , 248–269. 22. Chihara, K. , Gerovac, M. , Hör, J. and Vogel, J. (2022) Global profiling of the RNA and protein complexes of Esc heric hia coli by size exclusion chromatography followed by RNA sequencing and mass spectrometry (SEC-seq). RNA , 29 , 123–139. 23. Smirnov, A. , Schneider, C. , Hör, J. and Vogel, J. (2017) Discovery of new RNA classes and global RNA-binding proteins. Curr. Opin. Microbiol., 39 , 152–160. 24. Gerovac, M. , El Mouali, Y. , Kuper, J. , Kisker, C. , Barquist, L. and Vogel,J. (2020) Global discovery of bacterial RNA-binding proteins by RNase-sensitive gradient profiles reports a new FinO domain protein. RNA , 26 , 1448–1463. 25. Caudron-Herger, M. , Rusin, S.F. , Adamo, M.E. , Seiler, J. , Schmid, V.K. , Barreau, E. , Kettenbach, A.N. and Diederichs, S. (2019) R-DeeP: proteome-wide and quantitative identification of RNA-dependent proteins by density gradient ultracentrifugation. Mol. Cell , 75 , 184–199. 26. Caudron-Herger, M. , Wassmer, E. , Nasa, I. , Schultz, A.S. , Seiler, J. , Kettenbach, A.N. and Diederichs, S. (2020) Identification, quantification and bioinformatic analysis of RNA-dependent proteins by RNase treatment and density gradient ultracentrifugation using R-DeeP. Nat. Protoc., 15 , 1338–1370. 27. Rippka, R. , Deruelles, J. , Waterbury, J.B. , Herdman, M. and Stanier,R.Y. (1979) Generic assignments, strain histories and properties of pure cultures of cyanobacteria. Microbiology , 111 , 1–61. 28. Markwell, M.A. , Haas, S.M. , Bieber, L.L. and Tolbert, N.E. (1978) A modification of the Lowry procedure to simplify protein determination in membrane and lipoprotein samples. Anal. Biochem., 87 , 206–210. 29. Mohamed, A. and Jansson, C. (1989) Influence of light on accumulation of photosynthesis-specific transcripts in the cyanobacterium Synechocystis 6803. Plant Mol. Biol., 13 , 693–700. 30. Ziemann, M. , Reimann, V. , Liang, Y. , Shi, Y. , Ma, H. , Xie, Y. , Li, H. , Zhu, T. , Lu, X. and Hess, W.R. (2023) CvkR is a MerR-type transcriptional repressor of class 2 type V-K CRISPR-associated transposase systems. Nat. Commun., 14 , 924. 31. Mateus, A. , Hevler, J. , Bobonis, J. , Kurzawa, N. , Shah, M. , Mitosch, K. , Goemans, C.V. , Helm, D. , Stein, F. , Typas, A. , et al. (2020) The functional proteome landscape of Esc heric hia coli . Nature , 588 , 473–478. 32. Hughes, C.S. , Moggridge, S. , Müller, T. , Sorensen, P.H. , Morin, G.B. and Krijgsveld,J. (2019) Single-pot, solid-phase-enhanced sample preparation for proteomics experiments. Nat. Protoc., 14 , 68–85. 33. Ritchie, M.E. , Phipson, B. , Wu, D. , Hu, Y. , Law, C.W. , Shi, W. and Smyth,G.K. (2015) limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res., 43 , e47. 34. Huber, W. , von Heydebreck, A. , Sültmann, H. , Poustka, A. and V ingron, M. (2002) Variance stabilization applied to microarray data calibration and to the quantification of differential expression. Bioinformatics , 18 , S96–S104. 35. Strimmer,K. (2008) Fdrtool: a versatile R package for estimating local and tail area-based false discovery rates. Bioinformatics , 24 , 1461–1462. 36. Yu, G. , Wang, L.G. , Han, Y. and He, Q.Y. (2012) ClusterProfiler: an R package for comparing biological themes among gene clusters. OMICS , 16 , 284–287. 37. Kumar, L. and Futschik, M.E. (2007) Mfuzz: a software package for soft clustering of microarray data. Bioinformation , 2 , 5–7. 38. Bressin, A. , Schulte-Sasse, R. , Figini, D. , Urdaneta, E.C. , Beckmann, B.M. and Marsico, A. (2019) TriPepSVM: de novo prediction of RNA-binding proteins based on short amino acid motifs. Nucleic Acids Res. , 47 , 4406–4417. 39. Palme, J. , Hochreiter, S. and Bodenhofer, U. (2015) KeBABS: an R package for kernel-based analysis of biological sequences. Bioinformatics , 31 , 2574–2576. 40. Erd ˝ os, G. and Dosztányi, Z. (2020) Analyzing protein disorder with IUPred2A. Curr. Protoc. Bioinformatics , 70 , e99. 41. Nordberg, H. , Cantor, M. , Dusheyko, S. , Hua, S. , Poliakov, A. , Shabalov, I. , Smirnova, T. , Grigoriev, I.V. and Dubchak, I. (2014) The genome portal of the Department of Energy Joint Genome Institute: 2014 updates. Nucleic Acids Res. , 42 , D26–D31. 42. Merino-Puerto, V. , Mariscal, V. , Mullineaux, C.W. , Herrero, A. and 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. Downloaded from https://academic.oup.com/nar/article/53/3/gkae1247/7928523 by Biblioteca Universidad de Sevilla user on 27 February 2025 Nucleic Acids Research , 2025, Vol. 53, No. 3 21 43. Black, T.A. , Cai, Y. and Wolk, C.P. (1993) Spatial expression and autoregulation of hetR , a gene involved in the control of heterocyst development in Anabaena . Mol. Microbiol., 9 , 77–84. 44. Elhai, J. and Wolk, C.P. (1988) Conjugal transfer of DNA to cyanobacteria. Methods Enzymol., 167 , 747–754. 45. Olmedo-Verd, E. , Brenes-Álvarez, M. , V ioque, A. and Muro-Pastor,A.M. (2019) A heterocyst-specific antisense RNA contributes to metabolic reprogramming in Nostoc sp. PCC 7120. Plant Cell Physiol. , 60 , 1646–1655. 46. Muro-Pastor,A.M. (2014) The heterocyst-specific NsiR1 small RNA is an early marker of cell differentiation in cyanobacterial filaments. mBio , 5 , e01079–14. 47. Valladares, A. , Rodríguez, V. , Camargo, S. , Martínez-Noel, G.M. , Herrero, A. and Luque, I. (2011) Specific role of the cyanobacterial PipX factor in the heterocysts of Anabaena sp. strain PCC 7120. J. Bacteriol., 193 , 1172–1182. 48. Camargo, S. , Valladares, A. , Flores, E. and Herrero, A. (2012) Transcription activation by NtcA in the absence of consensus NtcA-binding sites in an Anabaena heterocyst differentiation gene promoter. J. Bacteriol., 194 , 2939–2948. 49. Stenum, T.S. , Kumar, A.D. , Sandbaumhüter, F.A. , Kjellin, J. , Jerlström-Hultqvist, J. , Andrén, P.E. , Koskiniemi, S. , Jansson, E.T. and Holmqvist,E. (2023) RNA interactome capture in Esc heric hia coli globally identifies RNA-binding proteins. Nucleic Acids Res., 51 , 4572–4587. 50. Pfeifer-Sancar, K. , Mentz, A. , Rückert, C. and Kalinowski, J. (2013) Comprehensive analysis of the Corynebacterium glutamicum transcriptome using an improved RNAseq technique. BMC Genomics , 14 , 888. 51. Langmead, B. and Salzberg, S.L. (2012) Fast gapped-read alignment with Bowtie 2. Nat. Methods , 9 , 357–359. 52. Ramírez, F. , Ryan, D.P. , Grüning, B. , Bhardwaj, V. , Kilpert, F. , Richter, A.S. , Heyne, S. , Dündar, F. and Manke, T. (2016) deepTools2: a next generation web server for deep-sequencing data analysis. Nucleic Acids Res. , 44 , W160–W165. 53. Olmedo-Verd, E. , Flores, E. , Herrero, A. and Muro-Pastor, A.M. (2005) HetR-dependent and -independent expression of heterocyst-related genes in an Anabaena strain overproducing the NtcA transcription factor. J. Bacteriol., 187 , 1985–1991. 54. Schindelin, J. , Arganda-Carreras, I. , Frise, E. , Kaynig, V. , Longair, M. , Pietzsch, T. , Preibisch, S. , Rueden, C. , Saalfeld, S. , Schmid, B. , et al. (2012) Fiji: an open-source platform for biological-image analysis. Nat. Methods , 9 , 676–682. 55. Shih, P.M. , Wu, D. , Latifi, A. , Axen, S.D. , Fewer, D.P. , Talla, E. , Calteau, A. , Cai, F. , Tandeau de Marsac, N. , Rippka, R. , et al. (2013) Improving the coverage of the cyanobacterial phylum using diversity-driven genome sequencing. Proc. Natl Acad. Sci. U.S.A., 110 , 1053–1058. 56. Kumar, S. , Stecher, G. , Li, M. , Knyaz, C. and Tamura, K. (2018) MEGA X: molecular evolutionary genetics analysis across computing platforms. Mol. Biol. Evol., 35 , 1547–1549. 57. Jones, D.T. , Taylor, W.R. and Thornton, J.M. (1992) The rapid generation of mutation data matrices from protein sequences. Comput. Appl. Biosci., 8 , 275–282. 58. van Kempen, M. , Kim, S.S. , Tumescheit, C. , Mirdita, M. , Lee, J. , Gilchrist, C.L.M. , Söding, J. and Steinegger, M. (2024) Fast and accurate protein structure search with Foldseek. Nat. Biotechnol., 42 , 243–246. 59. Goddard, T.D. , Huang, C.C. , Meng, E.C. , Pettersen, E.F. , Couch, G.S. , Morris, J.H. and Ferrin, T.E. (2018) UCSF ChimeraX: meeting modern challenges in visualization and analysis. Protein Sci., 27 , 14–25. 60. Huang, X. , Dong, Y. and Zhao, J. (2004) HetR homodimer is a DNA-binding protein required for heterocyst differentiation, and the DNA-binding activity is inhibited by PatS. Proc. Natl Acad. Sci. U.S.A., 101 , 4848–4853. 61. Werner, T. , Sweetman, G. , Savitski, M.F. , Mathieson, T. , Bantscheff, M. and Savitski, M.M. (2014) Ion coalescence of neutron encoded TMT 10-plex reporter ions. Anal. Chem., 86 , 3594–3601. 62. Hou, S. , Brenes-Álvarez, M. , Reimann, V. , Alkhnbashi, O.S. , Backofen, R. , Muro-Pastor, A.M. and Hess, W.R. (2019) CRISPR-Cas systems in multicellular cyanobacteria. RNA Biol., 16 , 518–529. 63. Mella-Herrera, R.A. , Neunuebel, M.R. , Kumar, K. , Saha, S.K. and Golden,J.W. (2011) The sigE gene is required for normal expression of heterocyst-specific genes in Anabaena sp. strain PCC 7120. J. Bacteriol., 193 , 1823–1832. 64. Golub, M. , Hussein, R. , Ibrahim, M. , Hecht, M. , Wieland, D.C.F. , Martel, A. , Machado, B. , Zouni, A. and Pieper, J. (2020) Solution structure of the detergent-photosystem II core complex investigated by small-angle scattering techniques. J. Phys. Chem. B , 124 , 8583–8592. 65. Zheng, L. , Li, Y. , Li, X. , Zhong, Q. , Li, N. , Zhang, K. , Zhang, Y. , Chu, H. , Ma, C. , Li, G. , et al. (2019) Structural and functional insights into the tetrameric photosystem I from heterocyst-forming cyanobacteria. Nat. Plants , 5 , 1087–1097. 66. Adir, N. , Bar-Zvi, S. and Harris, D. (2020) The amazing phycobilisome. Biochim. Biophys. Acta Bioenerg., 1861 , 148047. 67. Collier, J.L. and Grossman, A.R. (1992) Chlorosis induced by nutrient deprivation in Synechococcus sp. strain PCC 7942: not all bleaching is the same. J. Bacteriol., 174 , 4718–4726. 68. Valladares, A. , Herrero, A. , Pils, D. , Schmetterer, G. and Flores, E. (2003) Cytochrome c oxidase genes required for nitrogenase activity and diazotrophic growth in Anabaena sp. PCC 7120. Mol. Microbiol., 47 , 1239–1249. 69. Inoue-Kashino, N. , Kashino, Y. , Orii, H. , Satoh, K. , Terashima, I. and Pakrasi,H.B. (2011) S4 protein Sll1252 is necessary for energy balancing in photosynthetic electron transport in Synechocystis sp. PCC 6803. Biochemistry , 50 , 329–339. 70. Gupta, T.K. , Klumpe, S. , Gries, K. , Heinz, S. , Wietrzynski, W. , Ohnishi, N. , Niemeyer, J. , Spaniol, B. , Schaffer, M. , Rast, A. , et al. (2021) Structural basis for VIPP1 oligomerization and maintenance of thylakoid membrane integrity. Cell , 184 , 3643–3659. 71. Zhang, C.C. , Friry, A. and Peng, L. (1998) Molecular and genetic analysis of two closely linked genes that encode, respectively, a protein phosphatase 1 / 2A / 2B homolog and a protein kinase homolog in the cyanobacterium Anabaena sp. strain PCC 7120. J. Bacteriol., 180 , 2616–2622. 72. Mitschke, J. , V ioque, A. , Haas, F. , Hess, W.R. and Muro-Pastor,A.M. (2011) Dynamics of transcriptional start site selection during nitrogen stress-induced cell differentiation in Anabaena sp. PCC 7120. Proc. Natl Acad. Sci. U.S.A., 108 , 20130–20135. 73. Brenes-Álvarez, M. , V ioque, A. and Muro-Pastor, A.M. (2023) Nitrogen-regulated antisense transcription in the adaptation to nitrogen deficiency in Nostoc sp. PCC 7120. PNAS Nexus , 2 , pgad187. 74. Buikema, W.J. and Haselkorn, R. (1991) Characterization of a gene controlling heterocyst differentiation in the cyanobacterium Anabaena 7120. Genes Dev. , 5 , 321–330. 75. Yoon, H.S. and Golden, J.W. (1998) Heterocyst pattern formation controlled by a diffusible peptide. Science , 282 , 935–938. 76. Callahan, S.M. and Buikema, W.J. (2001) The role of HetN in maintenance of the heterocyst pattern in Anabaena sp. PCC 7120. Mol. Microbiol., 40 , 941–950. 77. Álvarez-Escribano, I. , Brenes-Álvarez, M. , Olmedo-Verd, E. , V ioque, A. and Muro-Pastor, A.M. (2020) The nitrogen stress-repressed sRNA NsrR1 regulates expression of all187 1, a gene required for diazotrophic growth in Nostoc sp. PCC 7120. Life (Basel) , 10 , 54. 78. Lhee, D. , Ha, J.S. , Kim, S. , Park, M.G. , Bhattacharya, D. and Yoon,H.S. (2019) Evolutionary dynamics of the chromatophore genome in three photosynthetic Paulinella species. Sci. Rep., 9 , 2560. Downloaded from https://academic.oup.com/nar/article/53/3/gkae1247/7928523 by Biblioteca Universidad de Sevilla user on 27 February 2025 22 Nucleic Acids Research , 2025, Vol. 53, No. 3 79. Queiroz, R.M.L. , Smith, T. , V illanueva, E. , Marti-Solano, M. , Monti, M. , Pizzinga, M. , Mirea, D.M. , Ramakrishna, M. , Harvey, R.F. , Dezi, V. , et al. (2019) Comprehensive identification of RNA–protein interactions in any organism using orthogonal organic phase separation (OOPS). Nat. Biotechnol., 37 , 169–178. 80. Chu, L.C. , Arede, P. , Li, W. , Urdaneta, E.C. , Ivanova, I. , McKellar, S.W. , Wills, J.C. , Fröhlich, T. , von Kriegsheim, A. , Beckmann, B.M. , et al. (2022) The RNA-bound proteome of MR S A reveals post-transcriptional roles for helix-turn-helix DNA-binding and Rossmann-fold proteins. Nat. Commun., 13 , 2883. 81. Özcan, A. , Pausch, P. , Linden, A. , Wulf, A. , Schühle, K. , Heider, J. , Urlaub, H. , Heimerl, T. , Bange, G. and Randau, L. (2019) Type IV CRISPR RNA processing and effector complex formation in Aromatoleum Aromaticum . Nat. Microbiol., 4 , 89–96. 82. Corley, M. , Burns, M.C. and Yeo, G.W. (2020) How RNA-binding proteins interact with RNA: molecules and mechanisms. Mol. Cell , 78 , 9–29. 83. Nandana, V. and Schrader, J.M. (2021) Roles of liquid-liquid phase separation in bacterial RNA metabolism. Curr. Opin. Microbiol., 61 , 91–98. 84. Whitman, B.T. , Wang, Y. , Murray, C.R.A. , Glover, M.J.N. and Owttrim,G.W. (2023) Liquid-Liquid phase separation of the DEAD-box cyanobacterial RNA helicase redox (CrhR) into dynamic membraneless organelles in Ssynechocystis sp. strain PCC 6803. Appl. Environ. Microb., 89 , e0001523. 85. Picossi, S. , Valladares, A. , Flores, E. and Herrero, A. (2004) Nitrogen-regulated genes for the metabolism of cyanophycin, a bacterial nitrogen reserve polymer: expression and mutational analysis of two cyanophycin synthetase and cyanophycinase gene clusters in heterocyst-forming cyanobacterium Anabaena sp. PCC 7120. J. Biol. Chem., 279 , 11582–11592. 86. Argall, M.E. , Smith, G.D. , Stamford, N.P. and Youens, B.N. (1992) Purification and properties of urease from the cyanobacterium Anabaena cylindrica . Biochem. Int., 27 , 1027–1036. 87. Hemm, L. , Miucci, A. , Kraus, A. , Riediger, M. , Tholen, S. , Abdelaziz, N. , Georg, J. , Schilling, O. and Hess, W.R. (2024) Interactors and effects of overexpressing YlxR / RnpM, a conserved RNA binding protein in cyanobacteria. RNA Biol , 21 , 1–19. 88. Balaga, R.R. , Itoh, F. , Chauhan, S. , Mandal, M. , Krishna, P.S. , Suzuki, I. and Prakash, J.S.S. (2023) Sll1252 coordinates electron transport between plastoquinone and cytochrome b 6 / f complex in Synechocystis PCC 6803. Genes , 14 , 2151. 89. Liu, J. , Tassinari, M. , Souza, D.P. , Naskar, S. , Noel, J.K. , Bohuszewicz, O. , Buck, M. , Williams, T.A. , Baum, B. and Low, H.H. (2021) Bacterial Vipp1 and PspA are members of the ancient ESCRT-III membrane-remodeling superfamily. Cell , 184 , 3660–3673. 90. Junglas, B. and Schneider, D. (2018) What is Vipp1 good for? Mol. Microbiol., 108 , 1–5. 91. Bryan, S.J. , Burroughs, N.J. , Shevela, D. , Yu, J. , Rupprecht, E. , Liu, L.N. , Mastroianni, G. , Xue, Q. , Llorente-Garcia, I. , Leake, M.C. , et al. (2014) Localisation and interactions of the Vipp1 protein in cyanobacteria. Mol. Microbiol., 94 , 1179–1195. 92. Chan, C.M. , Zhou, C. and Huang, R.H. (2009) Reconstituting bacterial RNA repair and modification in vitro . Science , 326 , 247. 93. Zhang, W. , Du, Y. , Khudyakov, I. , Fan, Q. , Gao, H. , Ning, D. , Wolk, C.P. and Xu, X. (2007) A gene cluster that regulates both heterocyst differentiation and pattern formation in Anabaena sp. strain PCC 7120. Mol. Microbiol., 66 , 1429–1443. 94. Y in, L. , Zheng, Z. , Li, Y. , Li, X. , Cheng, D. , Dong, C. , Liu, Y. and Zhao,J. (2023) PatU3 plays a central role in coordinating cell division and differentiation in pattern formation of filamentous cyanobacterium Nostoc sp. PCC 7120. Sci. China Life Sci., 66 , 2896–2909. 95. Liu, J. , Xing, W .Y . , Liu, B. and Zhang, C.C. (2023) Three-dimensional coordination of cell-division site positioning in a filamentous cyanobacterium. PNAS Nexus , 2 , pgac307. 96. Lambrecht, S.J. , Kanesaki, Y. , Fuss, J. , Huettel, B. , Reinhardt, R. and Steglich,C. (2019) Interplay and targetome of the two conserved cyanobacterial sRNAs Yfr1 and Yfr2 in Prochlorococcus MED4. Sci. Rep., 9 , 14331. 97. Rajagopalan, R. and Callahan, S.M. (2010) Temporal and spatial regulation of the four transcription start sites of hetR from Anabaena sp. strain PCC 7120. J. Bacteriol. , 192 , 1088–1096. 98. Muro-Pastor, A.M. , Brenes-Álvarez, M. and V ioque, A. (2017) A combinatorial strategy of alternative promoter use during differentiation of a heterocystous cyanobacterium. Environ. Microbiol. Rep., 9 , 449–458. 99. Bernstein, D.A. , Eggington, J.M. , Killoran, M.P. , Misic, A.M. , Cox, M.M. and Keck, J.L. (2004) Crystal structure of the Deinococcus r adiodur ans single-stranded DNA-binding protein suggests a mechanism for coping with DNA damage. Proc. Natl Acad. Sci. U.S.A., 101 , 8575–8580. 100. Elhai, J. and Khudyakov, I. (2018) Ancient association of cyanobacterial multicellularity with the regulator HetR and an RGSGR pentapeptide-containing protein (PatX). Mol. Microbiol., 110 , 931–954. 101. Perez-Riverol, Y. , Bai, J. , Bandla, C. , García-Seisdedos, D. , Hewapathirana, S. , Kamatchinathan, S. , Kundu, D.J. , Prakash, A. , Frericks-Zipper, A. , Eisenacher, M. , et al. (2022) The PRIDE database resources in 2022: a hub for mass spectrometry-based proteomics evidences. Nucleic Acids Res. , 50 , D543–D552. Received: May 31, 2024. Revised: November 15, 2024. Editorial Decision: December 2, 2024. Accepted: December 4, 2024 ©The Author(s) 2024. Published by Oxford University Press on behalf of Nucleic Acids Research. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (https: // creativecommons.org / licenses / by-nc / 4.0 / ), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact [email protected] for reprints and translation rights for reprints. All other permissions can be obtained through our RightsLink service via the Permissions link on the article page on our site—for further information please contact [email protected]. Downloaded from https://academic.oup.com/nar/article/53/3/gkae1247/7928523 by Biblioteca Universidad de Sevilla user on 27 February 2025