Skip to main content
NCBI home page
iScience logoLink to iScience
. 2021 Jan 5;24(1):101977. doi: 10.1016/j.isci.2020.101977

Impaired cell-cell communication in the multicellular cyanobacterium Anabaena affects carbon uptake, photosynthesis, and the cell wall

Sergio Camargo 1, Dena Leshkowitz 2, Bareket Dassa 2, Vicente Mariscal 3, Enrique Flores 3,∗, Joel Stavans 1,4,∗∗, Rinat Arbel-Goren 1
PMCID: PMC7797909  PMID: 33458622

Summary

Cell-cell communication is an essential attribute of multicellular organisms. The effects of perturbed communication were studied in septal protein mutants of the heterocyst-forming filamentous cyanobacterium Anabaena sp. PCC 7120 model organism. Strains bearing sepJ and sepJ/fraC/fraD deletions showed differences in growth, pigment absorption spectra, and spatial patterns of expression of the hetR gene encoding a heterocyst differentiation master regulator. Global changes in gene expression resulting from deletion of those genes were mapped by RNA sequencing analysis of wild-type and mutant strains, both under nitrogen-replete and nitrogen-poor conditions. The effects of sepJ and fraC/fraD deletions were non-additive, and perturbed cell-cell communication led to significant changes in global gene expression. Most significant effects, related to carbon metabolism, included increased expression of genes encoding carbon uptake systems and components of the photosynthetic apparatus, as well as decreased expression of genes encoding cell wall components related to heterocyst differentiation and to polysaccharide export.

Subject Areas: Microbiology, Omics

Graphical Abstract

graphic file with name fx1.jpg

Open in a new tab

Highlights

  • •

    SepJ and Fra septal proteins are key for communication in Anabaena filaments

  • •

    Spatial expression of a regulatory gene is affected when those proteins are missing

  • •

    The transcriptomes of Anabaena wild-type and septal protein mutants were mapped

  • •

    Deletion of those proteins leads to global and specific changes in gene expression

Microbiology; Omics

Introduction

Multicellularity in an organism is predicated on the molecular transfer of metabolites, nutrients, and signals that enables communication between its cells, division of labor, and cooperation (Herrero et al., 2016). Due to the many evolutionary advantages it confers, multicellularity arose independently and in many contexts other than metazoans or plants, such as in slime molds, ciliates, and prokaryotes (Lyons and Kolter, 2015). One of the earliest examples of multicellularity is furnished by filamentous cyanobacteria such as Anabaena sp. strain PCC 7120 (hereafter Anabaena) (Herrero et al., 2016). The basic structure of an Anabaena filament consists of a one-dimensional array of cells (Herrero et al., 2016; Flores et al., 2018), each delimited by a cytoplasmic membrane surrounded by a peptidoglycan mesh. An outer membrane spans the entire filament, defining a continuous periplasmic space (Mariscal et al., 2007; Wilk et al., 2011). Filament integrity is maintained by both cell wall components, the outer membrane and the peptidoglycan mesh (Burnat et al., 2014). Septal junctions, which likely traverse the septal peptidoglycan through nanopores, join adjacent cells and facilitate cell-cell communication (Flores et al., 2018). Furthermore, septal junction-related proteins are important for filament integrity (Flores et al., 2007; Nayar et al., 2007; Merino-Puerto et al., 2010).

Under nitrogen-replete conditions, all cells in an Anabaena filament carry out both photosynthesis and nitrogen assimilation from combined nitrogen sources. Such a filament grows by the division of each and every cell along its length into two daughter cells. However, when combined nitrogen sources become scarce, some cells along a filament differentiate into heterocysts, micro-oxic cells specialized in atmospheric nitrogen fixation. The differentiation of a vegetative cell into a heterocyst involves large morphological and metabolic changes that are controlled by a specific program of gene expression (Flores et al., 2019). This program is orchestrated by two main regulators: the global transcription factor NtcA and the differentiation-specific factor HetR (Herrero et al., 2016; Flores et al., 2019). Heterocysts supply neighboring vegetative cells with products of nitrogen fixation, e.g., glutamine and β-aspartyl-arginine dipeptide, whereas vegetative cells supply heterocysts with reduced carbon compounds, such as glutamate, alanine, and sucrose (Herrero et al., 2016). Individual heterocysts are separated by intervals of about 10–15 vegetative cells where photosynthesis takes place, forming thus quasi-regular developmental patterns. Heterocysts lose the ability to divide, and developed filaments grow by division of vegetative cells. When the vegetative cell interval between two heterocysts becomes large enough, a new intercalary heterocyst grows in its midst, thereby preserving an intrinsic pattern scale. How are the development and behavior of Anabaena filaments affected when cell-cell communication is impaired is the overarching question we address in the present work.

Previous studies have addressed this question by inactivation of genes encoding proteins involved in the formation of septal junctions between adjacent cells. It has been proposed that septal junctions are at least of two types (Merino-Puerto et al., 2011), comprising the septal proteins SepJ, FraC, and FraD (Mullineaux et al., 2008; Merino-Puerto et al., 2010, 2011; Mariscal et al., 2011; Ramos-León et al., 2018). Inactivation of the sepJ gene results in slow intercellular molecular transfer of calcein (Mullineaux et al., 2008; Mariscal et al., 2011), and SepJ has been implicated in the transfer of an inhibitory morphogen that impedes the formation of contiguous heterocysts (Rivers et al., 2014; Corrales-Guerrero et al., 2015). FraC and FraD have been suggested to be involved in the transfer of nutrients such as sucrose between cells (Nürnberg et al., 2015). FraC and FraD are thought to work together (Merino-Puerto et al., 2011) and indeed have been shown to be associated with the septal junctions that have been visualized by electron cryotomography (Weiss et al., 2019). Interestingly, whereas ΔsepJ mutants are arrested at an early stage of heterocyst differentiation upon nitrogen deprivation (Flores et al., 2007; Nayar et al., 2007), ΔfraC/ΔfraD mutants form heterocysts that are active while presenting lower nitrogenase activity than the wild-type strain (Merino-Puerto et al., 2010, 2011). Neither the ΔsepJ mutant nor the ΔfraC/ΔfraD mutant can grow diazotrophically.

These findings led us to enquire about the behavior of filaments in which cell-cell communication is perturbed further, by studying the characteristics of a triple mutant of Anabaena with a ΔsepJΔfraCΔfraD genetic background. Previous studies have shown that under nitrogen-poor conditions, ΔsepJΔfraCΔfraD mutant filaments display a marked fragmentation phenotype upon nitrogen deprivation (Nürnberg et al., 2015). In addition to exhibiting an altered septum structure with thinner septa and a denser peptidoglycan layer, strain ΔsepJΔfraCΔfraD is characterized by an impaired intercellular diffusion of small fluorescent molecules and a reduction in the frequency of nanopores in intercellular septal cross walls (Nürnberg et al., 2015). As its parental sepJ mutant strain, the ΔsepJΔfraCΔfraD mutant cannot form heterocysts under nitrogen-deprived conditions and cannot grow diazotrophically.

In an attempt to understand better the extent to which the addition of the ΔfraC/ΔfraD deletions to the ΔsepJ genetic background perturbs further the physiology of Anabaena, we endeavored to tease apart the differences in behavior between the ΔsepJ and ΔsepJΔfraCΔfraD mutant filaments, both under nitrogen-replete and -poor conditions. Using single-cell, time-lapse microscopy, we characterized developmental spatial patterns of expression from the hetR promoter and autofluorescence in Anabaena strains with wild-type (WT), ΔsepJ, and ΔsepJΔfraCΔfraD genotypes, both under nitrogen-replete and -poor conditions, and measured their cell growth and absorption spectra. To shed light on the observed behavior, we studied the differences between the transcriptomes of these strains using RNA sequencing (RNA-seq) methods (Transparent Methods). We found that perturbations to cell-cell communication in this model organism led to significant changes in expression of genes encoding specific metabolic pathways as well as various cellular structures.

Results

Contrasting patterns of hetR expression in septal junction mutant backgrounds under nitrogen-replete and -poor conditions

To pinpoint the differences in behavior between Anabaena strains with ΔsepJ and ΔsepJΔfraCΔfraD genotypes, we focused on the expression of the master regulator of development hetR and characterized statistically patterns of expression from its promoter in strains bearing a PhetR-gfp reporter fusion and the autofluorescence of photosynthetic pigments. We show in Figure 1 phase contrast, GFP fluorescence, and autofluorescence images of filaments of strains carrying the PhetR-gfp construct in WT, ΔsepJ, and ΔsepJΔfraCΔfraD genetic backgrounds, just immediately (0 h) and 18 h after nitrogen deprivation. At this time, the filament with the WT background displayed well-defined heterocysts, as evidenced by a high GFP fluorescence intensity, low autofluorescence, and large size as in previous studies (Di Patti et al., 2018). In contrast, neither ΔsepJ nor ΔsepJΔfraCΔfraD mutant background filaments exhibited heterocysts, but GFP fluorescence intensity was significantly larger in clusters of contiguous cells. The autofluorescence intensity in these cell clusters was negatively correlated with the fluorescence intensity from the PhetR-gfp reporter fusion, consistent with the known loss of autofluorescence in cells that start differentiation. Hence, heterocyst differentiations starts in sepJ mutants as previously reported (Flores et al., 2007; Nayar et al., 2007).

Figure 1.

Figure 1

Open in a new tab

Behavior of septal protein deletion mutant filaments under nitrogen-replete and -poor conditions

Phase contrast, fluorescence intensity from PhetR-gfp expression (green), autofluorescence from photosynthetic pigments (red), and their overlay, for strains carrying the PhetR-gfp fusion in WT, ΔsepJ, and ΔsepJΔfraCΔfraD (TM) genetic backgrounds, at 0 and 18 h after nitrogen deprivation. Scale bar, 10 μM.

However, the autofluorescence of ΔsepJ cells was visibly lower than the autofluorescence of the WT cells, whereas that of the ΔsepJΔfraCΔfraD cells was higher. This observation is captured quantitatively in Figure 2A, where we show the mean autofluorescence intensity in cells of both ΔsepJ and ΔsepJΔfraCΔfraD mutant backgrounds relative to WT background values, both immediately (0 h) and 18 h after nitrogen deprivation. To investigate this behavior further, we measured UV-visible absorption spectra of cultures (Figure 2B). Consistent with the autofluorescence results presented above, absorption spectra profiles displayed the same trends. The spectrum of the ΔsepJΔfraCΔfraD mutant background was higher than that of the WT background in the presence of nitrate for almost all the spectral range, whereas the spectrum of the ΔsepJ background was somewhat lower, particularly in the region where chlorophyll a (Chl) absorbs blue light, around 440 nm. The same trends were maintained under nitrogen deprivation, but with some differences in magnitude. After 18 h nitrogen deprivation, the differences between the spectra of the ΔsepJΔfraCΔfraD and WT backgrounds were larger mainly in the regions of Chl (centered at 440 and 680 nm), whereas the differences between the spectra of the ΔsepJ and WT backgrounds were more significant in the phycocyanin region from 550 to 650 nm. These results suggest that the two septal junction mutants respond differently to nitrogen deprivation, and hence to the carbon-to-nitrogen balance that is key in the physiology of Anabaena (Herrero and Flores, 2019).

Figure 2.

Figure 2

Open in a new tab

Differences in the behavior of the autofluorescence between strains with ΔsepJ and ΔsepJΔfraCΔfraD genetic backgrounds

(A) Mean autofluorescence intensity from photosynthetic pigments in ΔsepJ and ΔsepJΔfraCΔfraD (TM) cells normalized by the mean in WT cells measured from microscope images. Autofluorescence intensity ratios were measured at 0 h (pink) and 18 h (cyan) after nitrogen deprivation. For each strain, mean autofluorescence intensity was measured as indicated in the Transparent Methods, from more than 500 cells in 20 filaments. Error bars denote standard errors of two independent experiments, with at least two replicates each.

(B) Absorption spectra of intact cells in WT (blue), ΔsepJ (orange), and ΔsepJΔfraCΔfraD (green) cultures in the presence of nitrate (top panel) and 18 h after nitrogen deprivation (bottom panel). For each strain, absorption spectra were measured as indicated in the Transparent Methods from two independent experiments with three replicates each.Error bars denote the corresponding standard errors. Exponentially growing cultures were adjusted to OD750 = 0.4 in growth medium.

We also characterized how deletion of septal proteins affected filament growth under nitrogen-replete conditions. As absorption spectra varied between the three strains, we did not use total Chl concentration for the measurement of growth. Instead, we determined the number of cells in given filament segments directly as a function of time from our microscope observations and calculated the cell doubling time. We obtained 24 ± 2 h for the WT (similar to a value reported for liquid cultures, Pernil et al., 2008) and 36 ± 1 h and 38 ± 3 h for the ΔsepJ and ΔsepJΔfraCΔfraD backgrounds, respectively, indicating that the main difference in growth was between WT filaments on one hand and the septal mutant backgrounds ΔsepJ and ΔsepJΔfraCΔfraD on the other. Therefore, the sepJ deletion, common to both, seems to be especially responsible for the decrease in growth rate.

To characterize the statistical differences between expression patterns and spatial correlations of PhetR-gfp expression in ΔsepJ and ΔsepJΔfraCΔfraD filaments, we plotted scatterplots of the fluorescence intensities f(n+1) of the n+1-th cell in a filament as a function of the fluorescence f(n) of its nearest neighbor n along all filaments, as shown for the different strains in Figure 3. Under nitrogen-replete conditions, filaments of the three strains displayed a highly correlated behavior, with nearest-neighbor cells tending to exhibit similar fluorescence intensities, yielding significant positive correlations as measured by the Pearson correlation coefficients P (top panels, Figure 3). Note that the value of P for ΔsepJ filaments was somewhat smaller than for WT and ΔsepJΔfraCΔfraD filaments. Nitrogen deprivation brought about a large reduction in correlation in the three strains (middle panels, Figure 3), but particularly in the WT background, as a result of the large differences in fluorescence intensities between heterocysts and their nearest-neighbor vegetative cells. In contrast, the reduction in the ΔsepJ and ΔsepJΔfraCΔfraD backgrounds was smaller, due to the high expression from PhetR-gfp in clusters of contiguous cells. Differences between patterns of expression in ΔsepJ and ΔsepJΔfraCΔfraD backgrounds were more evident if one considers the correlation between second neighbors along filaments (f(n+2) as a function of f(n), bottom panels in Figure 2B): whereas the Pearson correlation for the WT and ΔsepJ filaments was essentially zero, filaments of the ΔsepJΔfraCΔfraD background retained a non-negligible degree of correlation (P = 0.24 ± 0.04). These results suggest that the lateral inhibition in the ΔsepJΔfraCΔfraD background was weaker than in the ΔsepJ background, resulting in more homogeneous cells. Furthermore, the results in this section suggest that under nitrogen deprivation, the effects of the ΔsepJ and ΔfraCΔfraD mutations may be non-additive and that the combined effect of deletion of sepJ and fraC-fraD partially preserves positive correlations such as those observed under nitrogen-replete conditions.

Figure 3.

Figure 3

Open in a new tab

Correlated fluctuations along filaments

Scatterplots of the fluorescence intensity from PhetR-gfp expression in cell pairs along filaments of WT, ΔsepJ, and ΔsepJΔfraCΔfraD (TM) genetic backgrounds. Top row: scatterplots of the fluorescence intensity of the n+1-th cell in a filament, as function of the fluorescence intensity f(n) of the n-th cell, just after nitrogen deprivation (0 h). Middle row: f(n+1) versus f(n), 18 h after nitrogen deprivation. Bottom row: f(n+2) versus f(n), 18 h after nitrogen deprivation. Shown in all scatterplots are Pearson correlation coefficients P, with errors calculated from 10,000 bootstrap samples.

RNA-seq analysis

To understand better the effects of the deletions of the septal protein-encoding genes, we carried out RNA-seq experiments and compared the transcriptomes of the WT, ΔsepJ, and ΔsepJΔfraCΔfraD strains, both immediately (0 h) and 18 h after nitrogen stepdown. At this time, transcripts belonging to intermediate (6–12 h) and late (12–24 h) stages in the expression of genes from the network controlling heterocyst differentiation should be well represented (Flores et al., 2019). Although the fragmentation process giving rise to short filaments is known to commence right after nitrogen deprivation (Nürnberg et al., 2015), fragmentation was not complete 18 h after nitrogen stepdown, and filaments such as those shown in Figure 1 were observed. Analysis of four replicates for each condition was carried out to improve accuracy, after removal of ribosomal RNA (Transparent Methods). In total, 259.6 million sequence reads (up to 61 nt long) were mapped to the genome, and only reads with unique mapping, 54.5 million, were considered for further analysis. As an initial characterization and check of our experimental design, samples were clustered using principal-component analysis (PCA). All genes with a sum count above 5 reads in all samples were used to carry out the PCA and predict the relatedness between samples. As shown in Figure S1, samples belonging to the different strains and grown under different conditions segregated into well-separated groups, and biological replicates clustered tightly together. No outlier values were detected, so all the replicates were used for further analyses. The first component PC1 explains 25.2% of the variance between samples and segregates these samples according to nitrogen availability. The second component PC2 explains 16.6% of the variance and shows a clear difference between the mutants and the WT strain, as well as between the mutant strains themselves (Figure S1).

Global analysis of changes resulting from acclimation to N depletion and deletion of septal protein genes

To identify unique patterns of transcriptional activity in the mutant strains relative to the WT, as well as between the different mutant strains themselves, we performed an unsupervised k-means clustering analysis of the data. As we detected a few cases of strong changes in expression of plasmid genes that could reflect genome rearrangements (e.g., deletions), we used the regularized log transformation values only for chromosomal genes that were differentially expressed (differential expression genes [DEG]). The RNA-seq data for the whole genome and for the DEG are found in Table S1. A k-means analysis was carried out both for nitrogen-replete and -poor conditions, taking into account the four biological replicates for each strain. To estimate a minimal value of clusters that show common expression profiles, we used the elbow method (Syakur et al., 2018). Although the method yields k = 4 as an optimal number of initial clusters (Figure S2), we found more informative to group the total of 3,712 genes into six clusters (k = 6), as shown in Figure 4A.

Figure 4.

Figure 4

Open in a new tab

K-means clustering of differentially transcribed genes in WT, ΔsepJ, and ΔsepJΔfraCΔfraD (TM) strains in the presence of nitrate (+N) and under nitrogen deprivation (-N)

(A) Normalized values of 3,712 chromosomal genes were used to run an unsupervised k-means analysis, setting k = 6. The legend shows the z-scores. The values were mean centered, and the colors scaled from −3 to 3 standard deviations.

(B) Network of differentially expressed genes and their association with enriched COG terms. Differentially expressed genes were clustered into six groups (using the K-means algorithm), based on their expression profile in the different strains (A). Enrichment analysis revealed 574 of these differentially expressed genes (represented by arcs) to be associated with enriched COG terms. Each gene is assigned to a gene expression cluster (numbers) based on its expression profile and is associated with an enriched COG term (labels). Note that in group 2 no COG terms were detected in this analysis. Enrichment analysis was performed using GProfiler (with a significant threshold of p < 0.05), and the network was constructed using CytoScape 3.7.2.

Cluster 1 includes genes that were expressed at high levels under nitrogen-replete conditions and were downregulated under nitrogen deprivation, whereas cluster 4 includes genes that exhibited the opposite behavior, namely, genes that were expressed at low levels in nitrate-replete conditions and were upregulated under nitrogen deprivation (Figure 4A). Cluster 2 includes genes that in the WT were expressed independently of nitrogen availability and that were downregulated in both the ΔsepJ and ΔsepJΔfraCΔfraD mutants. Cluster 3 includes genes that in general were less expressed in the WT independently of nitrogen availability and were upregulated in both mutants. We attribute the effects of gene expression of clusters 2 and 3 to the ΔsepJ deletion, which is common to both mutants. Cluster 5 includes genes that were upregulated only in the ΔsepJΔfraCΔfraD mutant, independently of the nitrogen state, whereas cluster 6 represents the opposite behavior, namely, it includes genes that were downregulated only in the ΔsepJΔfraCΔfraD mutant. These last two clusters yield information about the effects of deletion of fraC and fraD. Most of the genes belonging to clusters 1 and 4 have been well studied during the last decades (Herrero et al., 2016; Flores et al., 2019). A glance at the data (Table S1, DEG_with_clusters sheet, filtering by cluster number) reveals that cluster 1 includes genes whose products are part of Photosystem II (e.g., psbV, psbC), Photosystem I (e.g., psaX, psaA, psaB), and phycobilisomes (e.g., pecA, pecB, pecC), as well as genes related to nitrate assimilation (e.g., nrtD, nrtC, nrtA, nirA, narB). In contrast, cluster 4 includes genes related to heterocyst differentiation (e.g., hesA, hetC, hepB, devA, patS, hetR) and other regulatory genes (e.g., patB, pknC, sigA).

The response to sepJ or fraC-fraD deletion of genes belonging to clusters 2, 3, 5, and 6 was previously unknown. To shed light on the behavior of these genes, we tested whether a category included in the Cluster of Orthologous Groups (COG) annotation was overrepresented in these clusters (Tatusov et al., 2000). For some of these genes, specifically for 574 differentially expressed genes, we obtained positive results that were used to construct a network that relates the different clusters with enriched COG categories (Figure 4B). Cluster 1 was enriched in five categories, related primarily with the transport and metabolism of amino acids, lipids, nucleotides, inorganic ions, and coenzymes, which exhibited higher expression under nitrate conditions. Cluster 4 showed enrichment in the Cell wall/membrane/envelope biogenesis and Signal transduction mechanisms categories. Genes in cluster 3 related to changes in Cell wall and Energy production categories were independent of nitrogen availability, whereas the genes in cluster 5 seem to be at an intermediate situation between clusters 3 and 1, as they belong to categories common to both clusters. We did not observe any enrichment in the case of genes belonging to group 2, and therefore this cluster is not represented in the network of Figure 4B. Enrichment in only one category (Mobilome) was observed in the case of cluster 6.

We additionally performed a gene ontology (GO) enrichment analysis using Ontologizer (Bauer et al., 2008) (Transparent Methods) and found in all clusters 2,999 enriched GO terms (see results in Table S2). Of these, 916 terms are found in cluster 5 and 401 terms in cluster 6, which gives further insight on the effects of the fraC-fraD deletions on the transcriptional landscape (clusters 5 and 6; Figure 4A). The results of this enrichment analysis for clusters 5 (upregulated genes) and 6 (downregulated genes) in the ΔsepJΔfraCΔfraD mutant compared with the ΔsepJ mutant show that upregulated genes included genes related to general cellular processes such as photosynthesis, development, and some amino acid and nucleotide biosynthesis pathways (for p value < 0.05, Table S2). Consistent with detection of photosynthesis-related genes, some encoded gene products were related to metal and cofactor binding as well as to electron transport. In contrast to upregulated genes, downregulated genes were related to a lower number of specific processes including nucleoside metabolism.

Identifying specific changes in transcription

To obtain a more refined picture of the transcriptional landscape and to identify pathways that are enriched in gene groups more than would be expected by chance, we performed a functional enrichment analysis of our RNA-seq data. DEG from comparisons between the ΔsepJ or ΔsepJΔfraCΔfraD mutants and the WT strain under both nitrogen-replete and -poor conditions were used for this enrichment analysis. Significantly enriched pathways are shown in Figure S3, based on either the COG (Tatusov et al., 2000) or the CyanoBase (Nakao et al., 2010) annotations.

Globally, when we compared both mutants against the WT strain under nitrate conditions, we observed some effects common to both mutants. Thus, both mutants exhibited upregulation of genes belonging to the Inorganic Ion Transporters category (Figure 5). Among those genes, transporter genes related to bicarbonate (HCO3-) uptake, which were among the most strongly affected, are considerably upregulated. These genes include those encoding the ABC transporter Cmp and the MFS low-affinity Na+-bicarbonate co-transporter BicA (All1304) as well as its neighbor All1303, a putative Na+/H+ antiporter that may support BicA activity (the expression changes of these genes can be consulted in Table S3). The MFS high-affinity Na+-bicarbonate co-transporter, SbtA (All2134), was also upregulated in response to sepJ inactivation (Table S3). Genes in the all1843-all1837 cluster, which encode another putative Na+/H+ antiporter, were also highly upregulated (Figure 5). Finally, all0626, also highly upregulated, is a predicted Ca2+/H+ antiporter. All the components of the three known HCO3- transporters in Anabaena were upregulated with a fold change in the range 4–81 (log2FC of 2.13–6.35). We note that in addition to changes in the expression of carbon transporters (Figure 5), some changes in a wide repertory of transporters of other inorganic ions also took place (Figure S4).

Figure 5.

Figure 5

Open in a new tab

Hierarchical clustering heatmaps showing relative expression level of DEG in some selected enriched categories

Regularized-logarithm transformation expression values of DEG in each of four replicates for every strain and condition were used as input values. The color scale shows the differences between minimum (blue = 0) and maximum (red = 1) values in each row independently. Symbols + and − indicate whether there are significant differences in the comparison of the mutants with the wild-type strain or not, in this order: ΔsepJ versus WT (nitrate), ΔsepJΔfraCD versus WT (nitrate), ΔsepJ versus WT (no-nitrate), and ΔsepJΔfraCΔfraD versus WT (no nitrate). Heatmaps of additional genes pertaining to the Inorganic ion transport and Cell wall/membrane/envelope biogenesis categories are shown in Figure S4.

Under nitrate conditions we observed, on the other hand, some differences between the mutant strains in categories related to Energy Production. For example, whereas the ΔsepJ mutant exhibited downregulation of phycobilisome and ATP synthase genes, the ΔsepJΔfraCΔfraD mutant did not show this behavior (except for cpcF), and instead exhibited upregulation of some genes of the Photosystem II category and NADH dehydrogenase (NDH-1 or photosynthetic complex I, Schuller et al., 2020) (Figure 5). Within the Photosystem II category, expression changes were found for psbAII, psbAIII, and psbAIV genes that code for D1:2 isoforms of the D1 protein, which plays a key role in the PSII reaction center. The replacement of the native D1:1 isoform takes place under different stresses, such as under UV radiation (Sicora et al., 2006). Other genes such as all4000, all4001, (isiA) and all4002, whose products are homologous to the CP43 protein, and which play an essential role in the transmission of light energy to the photosynthetic reaction centers, also appeared in this category. (Note that isiA was placed by the automatic annotation as related to Photosystem II, although it is normally associated with Photosystem I). A very significant difference between the triple mutant and the ΔsepJ mutant under nitrate conditions was a 2.4- to 7.2-fold upregulation (2.3- to 4.2-fold upregulation in -N) of genes in the alr1576-alr1586 cluster (see Table S1). Some of these genes encode extracytoplasmic proteins related to polysaccharide and sugar metabolism, including glycoside hydrolases, oxidases, and two proteins similar to xylose isomerase (xylose is found in the envelopes of heterocyst-forming cyanobacteria, Cardemil and Wolk, 1979).

Under nitrogen deprivation, the most statistically significant effect that was detected in the mutants was a strong downregulation of structural genes related to the Cell wall/membrane/envelope biogenesis category. Although this effect was present in both mutants, the repression of genes in this category was stronger in the ΔsepJΔfraCΔfraD mutant (Figure 5). Among these structural genes, we found many related to the synthesis of the extra-cell envelope that surrounds heterocysts, which is composed of the heterocyst-specific polysaccharide (Hep) and heterocyst-specific glycolipid (Hgl) layers (Flores et al., 2019). Most of the hep genes located in the cluster alr2823-alr2841, also known as “HEP island,” were downregulated in this category (Huang et al., 2005), as well as other genes such as alr3698 (hepB) and alr3699, located outside of this region and also required for the formation of this layer (Wang et al., 2007). Some genes related to the synthesis of the Hgl layer (hgl genes) were also downregulated and were found in the Heterocyst differentiation category (Figures S3B and 5), as it was also the case for devC and devA genes that are involved in the deposition of this layer (Black et al., 1995; Fiedler et al., 1998) and hglK that is involved in septal maturation (Black et al., 1995; Arévalo and Flores, 2020). Among the structural genes that were downregulated, the gene cluster (all4420-all4430) is quite interesting because it showed this behavior in the ΔsepJΔfraCΔfraD mutant but hardly in the ΔsepJ mutant (Figure 5). Some of the genes of this cluster encode glycosyl transferases or extracellular polysaccharide biosynthesis/transport proteins. We report on other genes within this category that displayed changes in their expression in Figure S4.

Finally, some metabolic effects previously detected in the nitrate-grown mutants remained stable after nitrogen deprivation (see Figure S3B, Energy Production and Conversion, in both mutants, and ATP synthase, Photosystem II, and NADH dehydrogenase in the ΔsepJΔfraCΔfraD mutant). In addition, new upregulated gene categories related with protein synthesis, fatty acid metabolism, and CO2 fixation arose in the ΔsepJΔfraCΔfraD mutant under nitrogen deprivation. This last category includes the rbcS and rbcL genes that code for the small and large subunits of the CO2-fixing enzyme Rubisco, the ccmM and ccmN genes that are related to carboxysomes (Turmo et al., 2017), and the prk gene that codes for phosphoribulokinase, a key enzyme in the Calvin-Benson-Bassham cycle and an important point of regulation for CO2 fixation. In contrast, phycobilisome genes that under nitrate conditions were downregulated in the ΔsepJ mutant, appeared upregulated under nitrogen deprivation (Table S3).

Changes in expression of HetR-dependent genes

Given the importance of the HetR regulator in the heterocyst differentiation process, we checked whether the expression of HetR-regulated genes is altered in the septal protein mutants. For this purpose, we selected from our data genes that are known to be expressed under nitrogen deprivation in a HetR-dependent way (Brenes-Álvarez et al., 2019). In this work, genes were classified as early or late, depending on the specific time during heterocyst differentiation in which they displayed maximal expression. Most of these genes were well represented in our transcriptome and appeared upregulated under nitrogen deprivation in the WT strain. Septal mutants displayed changes in the expression of both early and late genes. Although we did not detect significant changes in hetR nor patS expression, both the ΔsepJΔfraCΔfraD and the ΔsepJ mutants displayed downregulation of ntcA in a combined nitrogen-independent way (Figure 6). Interestingly, under nitrogen deprivation the ΔsepJΔfraCΔfraD mutant displayed downregulation of patX, a widely distributed gene in cyanobacteria that appears to have a role very similar to that of patS during heterocyst differentiation in Anabaena (Figure 6) (Elhai and Khudyakov, 2018). Finally, upregulation of patA under nitrogen deprivation was observed in both mutants (Figure 5). PatA regulates the levels or activity of HetR (Risser and Callahan, 2008) and was recently shown to interact with SepJ and with some elements of the divisome (Valladares et al., 2020).

Figure 6.

Figure 6

Open in a new tab

Heatmaps showing relative expression level of DEG related with heterocyst differentiation process, in the early and late stages of the process

DEG whose expression is induced under nitrogen deprivation and in a HetR-dependent way (Brenes-Álvarez et al., 2019) are showed in this analysis. Regularized-logarithm transformation expression values of these genes in each of four replicates for every strain and condition were used as input values. The color scale shows the differences between minimum (blue = 0) and maximum (red = 1) values in each row independently. Symbols + and − indicate whether there are significant differences in the comparison of the mutants with the WT strain or not, in this order: ΔsepJ versus WT (nitrate), ΔsepJΔfraCΔfraD versus WT (nitrate), ΔsepJ versus WT (no nitrate), and ΔsepJΔfraCΔfraD versus WT (no nitrate).

Validation of RNA-seq results by real-time quantitative polymerase chain reaction

To provide independent evidence for the validity of the gene expression results identified in the RNA-seq analysis, we carried out real-time quantitative polymerase chain reaction (RT-qPCR) measurements of two of the most differentially expressed genes in our analysis: cmpB, which is one of the most upregulated genes in both the ΔsepJ mutant and the ΔsepJΔfraCΔfraD mutant independently of the nitrogen source, and hepC, which conversely, is one of the most downregulated genes. cmpB codes for a permease component of the bicarbonate (HCO3-) uptake Cmp complex (López-Igual et al., 2012), whereas the product of hepC resembles an UDP-galactose lipid carrier transferase (Huang et al., 2005). The expression profiles of these genes displayed the same tendency as that detected by RNA-seq, with only minor quantitative differences between the two methods (Figure 7).

Figure 7.

Figure 7

Open in a new tab

Validation of differentially expressed genes by real-time PCR

Comparison of RNA-seq (red) and RT-PCR (blue) measures of changes in expression of cmpB (positive changes) and hepC (negative changes) genes in the ΔsepJ mutant and the ΔsepJΔfraCΔfraD (TM) mutant against the wild-type strain in the presence of nitrate and after nitrogen deprivation. All data are shown as means ± standard error (n = 3).

Discussion

Heterocyst-forming cyanobacteria, organisms of ancient origin, represent one of the first successful attempts at creating a multicellular unit of selection in the history of life on Earth (Schirrmeister et al., 2013). Multicellularity and division of labor strongly hinge on cell-cell communication (Herrero et al., 2016). It is therefore of interest to test the extent to which disruption of intercellular communication affects the biology of the organism. In the present work, we studied the effects of altering the molecular transfer of nutrients and signals between cells by deleting selected septal proteins in Anabaena filaments, a model system of heterocyst-forming (multicellular) cyanobacteria in which the number of neighbors of each cell is well defined.

Although some differences in sugar transport, septal structure, and fragmentation between the WT, ΔsepJ, and ΔsepJΔfraCΔfraD genetic backgrounds have been already described (Nürnberg et al., 2015), our experiments shed light on additional aspects in which these strains differ. In terms of growth, the two mutant strains grow similarly under nitrogen-replete conditions, but at a lower rate than the WT. On the other hand, cellular autofluorescence levels and absorbance of pigments of ΔsepJ filaments are lower than those of the WT, whereas those of ΔsepJΔfraCΔfraD are higher, irrespective of whether these levels are measured before or 18 h after nitrogen deprivation. We note that some phycobilisome genes are downregulated already in the presence of nitrate in the sepJ mutant. The differences in autofluorescence as well as absorbance spectra between the three genetic backgrounds may also reflect different degrees of degradation of phycobilisomes, e.g., by NblA-dependent degradation (Baier et al., 2004), especially between the two mutant backgrounds (i.e., decreased phycobilisome degradation in the triple mutant when compared with the sepJ mutant). Phycobilisomes play a role in the antenna complex for photosynthesis and as a source of stored nitrogen (Bienert et al., 2006), and our RNA-seq results show that many genes within the Phycobilisome category were downregulated under nitrogen deprivation and the expression of some genes was altered in one or the two mutants analyzed (Figure 5).

The characterization of spatial patterns of expression of GFP from the promoter of the hetR gene, which encodes the master regulator of cell differentiation in Anabaena, also sharpens the differences between the behaviors of ΔsepJ and ΔsepJΔfraCΔfraD filaments. Nearest-neighbor Pearson correlations are high in WT, ΔsepJ, and ΔsepJΔfraCΔfraD filaments under nitrogen-replete conditions, whereas the weaker correlation between nearest-neighbor cells in ΔsepJ filaments relative to WT filaments is consistent with impaired transfer of a PatS-derived signal, as observed previously (Corrales-Guerrero et al., 2015; Mariscal et al., 2016). In contrast, under nitrogen-poor conditions, heterocysts are formed only in WT filaments, giving rise to off-diagonal contributions to the scatterplot of expression due to lateral inhibition and increased promoter activity of hetR. This results in a significant drop in correlation in WT filaments when compared with those of ΔsepJ and ΔsepJΔfraCΔfraD backgrounds. Correlations between second neighboring cells capture better the difference between expression patterns in ΔsepJ and ΔsepJΔfraCΔfraD filaments: whereas the Pearson coefficient is essentially zero for ΔsepJ filaments, correlations in ΔsepJΔfraCΔfraD filaments are still significant. Note that in contrast to nearest-neighbor pairs, which include sister cells resulting from cell division, correlations between second neighboring cells are much weaker. One may hypothesize that the smoother, graded expression patterns observed in ΔsepJΔfraCΔfraD filaments are consistent with a compensation in intercellular molecular transfer in the ΔsepJΔfraCΔfraD compared with ΔsepJ background filaments. Such compensation might result from the absence of a FraD-containing plug that gates some septal junctions (Weiss et al., 2019).

The differences between filaments of the three genetic backgrounds in spatial patterns of expression and absorption spectra were more marked at late stages following nitrogen deprivation, in which the expression of genes controlling heterocyst differentiation should have taken place. This motivated an exploration of the transcriptome landscapes under both nitrogen-replete and -poor conditions. The transcriptomes revealed that deletion of septal proteins causes effects in the expression of several sets of genes, likely as a result of impaired cell-cell communication. A striking effect of sepJ inactivation and, further, of the combined inactivation of sepJ, fraC, and fraD is upregulation of bicarbonate uptake transporters including an ABC transporter and Na+-bicarbonate co-transporters, accompanied by Na+/H+ antiporters that may be important to support the activity of the co-transporters. This may indicate a state of sensing C starvation in the mutants, resulting in induction of the transporters. The all4444-all4445-all4446 operon encoding vegetative cell flavodiiron proteins is also upregulated in the ΔsepJ and ΔsepJΔfraCΔfraD strains (Table S3), and this operon is known to be induced by CO2 limitation (Allahverdiyeva et al., 2015; Picossi et al., 2015). These flavodiiron proteins are involved in the photo-protection of Photosystem II, which along with the NDH-1 complex is also upregulated in the mutants. It is noteworthy that this effect is noticeable particularly under nitrogen-replete conditions. How inactivation of the septal proteins leads to these effects is unknown, but it may be related to the impairment in the intercellular transfer of sucrose that appears to occur in the mutants (Nürnberg et al., 2015), which may result in uneven distribution of photosynthate along the filament. Formation of septal junctions, which mediate intercellular molecular exchange, is impaired in the sepJ and fraC-fraD deletion mutants (Flores et al., 2018).

There are genes whose expression in the septal protein mutants is different from the WT but similar in both mutants (clusters 2 and 3), and can therefore be ascribed to the effects of sepJ deletion common to both mutants, whereas there are genes whose expression is primarily affected by the fraC-fraD deletion (clusters 5 and 6). Most of the genes in these clusters have not been annotated, and their function is unknown. The most striking effect in the ΔsepJΔfraCΔfraD strain when compared with the ΔsepJ strain is induction of the alr1576-alr1586 gene cluster, in which some genes encode enzymes that may catalyze exopolysaccharide and sugar modifications.

In line with previous results, we observed significant remodeling of the WT transcriptome following nitrogen deprivation (Flaherty et al., 2011), and in particular changes in the transcription of heterocyst differentiation genes controlled by the master regulator HetR (Brenes-Álvarez et al., 2019). As shown in Figure 6, the changes in HetR-controlled genes classified as representing early and late differentiation stages are similar for the WT, and timing differences may be accounted for by the different conditions of our experiments. These genes also highlight the differences between the transcriptomes of the ΔsepJ and ΔsepJΔfraCΔfraD mutant strains. Among these differences, we detected changes in early genes that could affect heterocyst differentiation patterns, such as lower expression of patX that codes for a polypeptide similar to PatS and that can assist in lateral inhibition during heterocyst pattern formation (Elhai and Khudyakov, 2018), and overexpression of patA, whose product affects the levels and activity of the HetR regulator (Risser and Callahan, 2008).

Our results showing downregulation of heterocyst differentiation genes (some of which are included in the Cell wall/membrane/envelope biogenesis category) in the ΔsepJ and ΔsepJΔfraCΔfraD strains are consistent with the known block in heterocyst differentiation caused by inactivation of sepJ (Flores et al., 2007), which as expected is evident only under N deprivation. Importantly, heterocyst differentiation starts in the sepJ mutants, but it is blocked after a thin Hep layer is deposited (Flores et al., 2007; Nayar et al., 2007). However, not all the genes included in the Cell wall/membrane/envelope biogenesis category have been previously related to heterocyst differentiation. This is the case, to the best of our knowledge, for the all4420-all4430 gene cluster that appears to be involved in the production of extracellular polysaccharides and is specifically downregulated by deletion of fraC-fraD. Filament integrity in Anabaena is dependent on both cell wall components, the peptidoglycan mesh and the outer membrane (Burnat et al., 2014), and filament fragmentation by inactivation of sepJ or fraC-fraD is dependent on the activity of cell wall amidases (Bornikoel et al., 2017). Thus, our results showing downregulation, mainly under nitrogen deprivation, of Cell wall/membrane/envelope biogenesis genes suggest that a decreased level of the encoded proteins may contribute to the fragmentation of filaments previously reported (Nürnberg et al., 2015). Again, downregulation may result from a decreased intercellular molecular transfer in the mutants, in this case related to an impaired transfer of regulators (Rivers et al., 2014; Corrales-Guerrero et al., 2015; Mariscal et al., 2016).

Taken together, our results show how deletion of sepJ, fraC, and fraD genes that code for proteins that have a role in the formation of septal junctions (Flores et al., 2018) can have a broader impact on gene expression in Anabaena than previously thought. In addition to the more obvious effects of disruption of exchange of metabolites and signals between heterocysts and vegetative cells in a developed filament (Nürnberg et al., 2015), perturbations to cell-cell communication also have clear effects under nitrogen-replete conditions in which cells are undifferentiated. This underscores the benefits of intercellular communication even when all the cells along a filament may be functionally equivalent. A pooled metabolism facilitated by cell-cell communication may correct imbalances that may arise in one cell in isolation. Septal proteins can affect gene expression in two ways. On one hand, they play an important structural role, forming part of a highly organized and regulated superstructure such as the septum. Deletion mutants have an altered septum structure, and this alteration can conceivably influence processes that take place on this structure, such as peptidoglycan synthesis (Nürnberg et al., 2015). On the other hand, these proteins are crucial in the transfer of metabolites and signals between the cells that make up Anabaena filaments. Differences in the intercellular transfer of different molecular species (e.g., of regulators of heterocyst differentiation and nutrients such as sucrose) could account for most of the differences observed between the transcriptomes of these mutants, as well as between each mutant and the wild-type strain. Among these differences, we observed changes in genes that belong to different categories, especially genes related to heterocyst differentiation, the cell wall, inorganic ion transport, carbon fixation, and photosynthesis. This underscores the fundamental function of septal junctions in a multicellular cyanobacterium such as Anabaena, going far beyond a role in the heterocyst differentiation process, participating in other processes such as the coordination of photosynthetic metabolism between the cells that are part of the same filament.

Limitations of the study

Deletion of septal proteins can have profound effects on a cyanobacterial organism of filamentary structure such as Anabaena. The present study sheds light on these effects, both under nitrogen-rich and -poor conditions, by investigating a sepJ deletion mutant and a sepJ/fraC/fraD triple deletion mutant. We found both global and specific effects in different mutants and nitrogen availability conditions. The transcriptome landscape mapping yields a comprehensive, broad analysis of these effects, whereas the resulting datasets invite further investigation to elucidate in-depth, specific effects of intercellular junction integrity on gene expression. Detailed studies to annotate and elucidate the function of genes, particularly those differentially affected in the triple mutant compared with the sepJ mutant, are called for to pinpoint at higher resolution the differences between the two mutants and the non-additivity of the sepJ and fraC/fraD deletions. For nitrogen deprivation our study was limited to 18 h after nitrogen stepdown. It will be interesting to explore the transcriptome landscape later, when mutant filaments are fully fragmented. Detailed studies of the transcriptomes at other times after nitrogen deprivation may also unravel other differences between the growth and survival of the sepJ and triple mutants that are not obvious at 18 h. Last, the effects of septal protein deletions on plasmid genes have not been analyzed due to possible genomic rearrangements. Future studies could address this aspect of gene expression and explore further the unique resources we are sharing with the community.

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to the Lead Contact, Joel Stavans (joel.stavans@weizmann.ac.il).

Materials availability

Strain CSVM12 will be made available upon request.

Data and code availability

The RNA-seq data generated in this study have been deposited in NCBI's Gene Expression Omnibus and are accessible through GEO Series accession number GSE161085 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE161085). The published article includes all the bioinformatics analysis reported in this study.

Methods

All methods can be found in the accompanying Transparent Methods supplemental file.

Acknowledgments

We thank Antonia Herrero for stimulating discussions and Shlomit Gilad for construction of libraries of cDNA and sequencing. Work in Seville was supported by grant BFU2017-88202-P from Plan Estatal de Investigación Científica y Técnica y de Innovación, Spain, co-financed by the European Regional Development Fund, to E.F. Work in Rehovot was supported by a Siegried and Irma Ullman Professorial Chair to J.S., and by the Minerva Stiftung http://www.minerva.mpg.de/ to J.S.

Authors contribution

Conceptualization R.A.-G., E.F., and J.S.; Methodology V.M., R.A.-G., and J.S.; Validation D.L., B.D., R.A.-G., S.C., and J.S.; Formal Analysis D.L., B.D., S.C., R.A.-G., E.F., and J.S.; Investigation R.A.G. and S.C.; Resources V.M., D.L., B.D., E.F., S.C., R.A.-G., and J.S.; Data Curation D.L., B.D., R.A.-G., and J.S.; Writing – Original Draft S.C., R.A.-G., and J.S.; Writing – Review & Editing S.C., D.L., B.D., E.F. R.A.-G., and J.S.; Visualization S.C., D.L., B.D., R.A.-G., and J.S.; Supervision E.F., R.A.-G., and J.S.; Project Administration R.A.-G. and J.S.; Funding Acquisition E.F. and J.S.

Declaration of interests

The authors declare no competing interests.

Published: January 22, 2021

Footnotes

Supplemental Information can be found online at https://doi.org/10.1016/j.isci.2020.101977.

Contributor Information

Enrique Flores, Email: eflores@ibvf.csic.es.

Joel Stavans, Email: joel.stavans@weizmann.ac.il.

Supplemental information

Document S1. Transparent methods and figures S1–S4
mmc1.pdf (931.1KB, pdf)
Table S1. RNA-seq data for the whole genome and for differential expression genes (DEG), related to Figure 4A
mmc2.xlsx (14.4MB, xlsx)
Table S2. Gene ontology (GO) enrichment analysis of all chromosomal genes using Ontologizer, related to Figure 4A
mmc3.xlsx (406KB, xlsx)
Table S3. Comparison of upregulated and downregulated genes between the wild-type and mutants under nitrogen-rich and -poor conditions, related to Figure 4A
mmc4.xlsx (161.6KB, xlsx)

References

  1. Allahverdiyeva Y., Isojärvi J., Zhang P., Aro E.-M. Cyanobacterial oxygenic photosynthesis is protected by flavodiiron proteins. Life. 2015;5:716–743. doi: 10.3390/life5010716. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Arévalo S., Flores E. Pentapeptide-repeat, cytoplasmic-membrane protein HglK influences the septal junctions in the heterocystous cyanobacterium Anabaena. Mol. Microbiol. 2020;113:794–806. doi: 10.1111/mmi.14444. [DOI] [PubMed] [Google Scholar]
  3. Baier K., Lehmann H., Stephan D.P., Lockau W. NblA is essential for phycobilisome degradation in Anabaena sp. strain PCC 7120 but not for development of functional heterocysts. Microbiology. 2004;150:2739–2749. doi: 10.1099/mic.0.27153-0. [DOI] [PubMed] [Google Scholar]
  4. Bauer S., Grossmann S., Vingron M., Robinson P.N. Ontologizer 2.0 - a multifunctional tool for GO term enrichment analysis and data exploration. Bioinformatics. 2008;24:1650–1651. doi: 10.1093/bioinformatics/btn250. [DOI] [PubMed] [Google Scholar]
  5. Bienert R., Baier K., Volkmer R., Lockau W., Heinemann U. Crystal structure of NblA from Anabaena sp. PCC 7120, a small protein playing a key role in phycobilisome degradation. J. Biol. Chem. 2006;281:5216–5223. doi: 10.1074/jbc.M507243200. [DOI] [PubMed] [Google Scholar]
  6. Black K., Buikema W.J., Haselkorn R. The hglK gene is required for localization of heterocyst-specific glycolipids in the cyanobacterium Anabaena sp. strain PCC 7120. J. Bacteriol. 1995;177:6440–6448. doi: 10.1128/jb.177.22.6440-6448.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bornikoel J., Carrión A., Fan Q., Flores E., Forchhammer K., Mariscal V., Mullineaux C.W., Perez R., Silber N., Wolk C.P., Maldener I. Role of two cell wall amidases in septal junction and nanopore formation in the multicellular cyanobacterium Anabaena sp. PCC 7120. Front. Cell. Infect. Microbiol. 2017;7:1–15. doi: 10.3389/fcimb.2017.00386. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Brenes-Álvarez M., Mitschke J., Olmedo-Verd E., Georg J., Hess W.R., Vioque A., Muro-Pastor A.M. Elements of the heterocyst-specific transcriptome unravelled by co-expression analysis in Nostoc sp. PCC 7120. Environ. Microbiol. 2019;21:2544–2558. doi: 10.1111/1462-2920.14647. [DOI] [PubMed] [Google Scholar]
  9. Burnat M., Schleiff E., Flores E. Cell envelope components influencing filament length in the heterocyst-forming cyanobacterium Anabaena sp. strain PCC 7120. J. Bacteriol. 2014;196:4026–4035. doi: 10.1128/JB.02128-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Cardemil L., Wolk C.P. The polysaccharides from heterocyst and spore envelopes of a blue-green alga. Structure of the basic repeating unit. J. Biol. Chem. 1979;254:736–741. [PubMed] [Google Scholar]
  11. Corrales-Guerrero L., Tal A., Arbel-Goren R., Mariscal V., Flores E., Herrero A., Stavans J. Spatial fluctuations in expression of the heterocyst differentiation regulatory gene hetR in Anabaena filaments. PLoS Genet. 2015;11:e1005031. doi: 10.1371/journal.pgen.1005031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Elhai J., Khudyakov I. Ancient association of cyanobacterial multicellularity with the regulator HetR and an RGSGR pentapeptide-containing protein (PatX) Mol. Microbiol. 2018;110:931–954. doi: 10.1111/mmi.14003. [DOI] [PubMed] [Google Scholar]
  13. Fiedler G., Arnold M., Hannus S., Maldener I. The DevBCA exporter is essential for envelope formation in heterocysts of the cyanobacterium Anabaena sp. strain PCC 7120. Mol. Microbiol. 1998;27:1193–1202. doi: 10.1046/j.1365-2958.1998.00762.x. [DOI] [PubMed] [Google Scholar]
  14. Flaherty B.L., Van Nieuwerburgh F., Head S.R., Golden J.W. Directional RNA deep sequencing sheds new light on the transcriptional response of Anabaena sp. strain PCC 7120 to combined-nitrogen deprivation. BMC Genomics. 2011;12:332. doi: 10.1186/1471-2164-12-332. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Flores E., Pernil R., Muro-Pastor A.M., Mariscal V., Maldener I., Lechno-Yossef S., Fan Q., Wolk C.P., Herrero A. Septum-localized protein required for filament integrity and diazotrophy in the heterocyst-forming cyanobacterium Anabaena sp. strain PCC 7120. J. Bacteriol. 2007;189:3884–3890. doi: 10.1128/JB.00085-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Flores E., Picossi S., Valladares A., Herrero A. Transcriptional regulation of development in heterocyst-forming cyanobacteria. Biochim. Biophys. Acta Gene Regul. Mech. 2019;1862:673–684. doi: 10.1016/j.bbagrm.2018.04.006. [DOI] [PubMed] [Google Scholar]
  17. Flores E., Nieves-Morión M., Mullineaux C. Cyanobacterial septal junctions: properties and regulation. Life. 2018;9:1. doi: 10.3390/life9010001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Herrero A., Flores E. Genetic responses to carbon and nitrogen availability in Anabaena. Environ. Microbiol. 2019;21:1–17. doi: 10.1111/1462-2920.14370. [DOI] [PubMed] [Google Scholar]
  19. Herrero A., Stavans J., Flores E. The multicellular nature of filamentous heterocyst-forming cyanobacteria. FEMS Microbiol. Rev. 2016;40:831–854. doi: 10.1093/femsre/fuw029. [DOI] [PubMed] [Google Scholar]
  20. Huang G., Fan Q., Lechno-Yossef S., Wojciuch E., Wolk C.P., Kaneko T., Tabata S. Clustered genes required for the synthesis of heterocyst envelope polysaccharide in Anabaena sp. strain PCC 7120. J. Bacteriol. 2005;187:1114–1123. doi: 10.1128/JB.187.3.1114-1123.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. López-Igual R., Picossi S., López-Garrido J., Flores E., Herrero A. N and C control of ABC-type bicarbonate transporter Cmp and its LysR-type transcriptional regulator CmpR in a heterocyst-forming cyanobacterium, Anabaena sp. Environ. Microbiol. 2012;14:1035–1048. doi: 10.1111/j.1462-2920.2011.02683.x. [DOI] [PubMed] [Google Scholar]
  22. Lyons N.A., Kolter R. On the evolution of bacterial multicellularity. Curr. Opin. Microbiol. 2015;24:21–28. doi: 10.1016/j.mib.2014.12.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Mariscal V., Herrero A., Nenninger A., Mullineaux C.W., Flores E. Functional dissection of the three-domain SepJ protein joining the cells in cyanobacterial trichomes. Mol. Microbiol. 2011;79:1077–1088. doi: 10.1111/j.1365-2958.2010.07508.x. [DOI] [PubMed] [Google Scholar]
  24. Mariscal V., Nürnberg D.J., Herrero A., Mullineaux C.W., Flores E. Overexpression of SepJ alters septal morphology and heterocyst pattern regulated by diffusible signals in Anabaena. Mol. Microbiol. 2016;101:968–981. doi: 10.1111/mmi.13436. [DOI] [PubMed] [Google Scholar]
  25. Mariscal V., Herrero A., Flores E. Continuous periplasm in a filamentous, heterocyst-forming cyanobacterium. Mol. Microbiol. 2007;65:1139–1145. doi: 10.1111/j.1365-2958.2007.05856.x. [DOI] [PubMed] [Google Scholar]
  26. Merino-Puerto V., Mariscal V., Mullineaux C.W., Herrero A., Flores E. Fra proteins influencing filament integrity, diazotrophy and localization of septal protein SepJ in the heterocyst-forming cyanobacterium Anabaena sp. Mol. Microbiol. 2010;75:1159–1170. doi: 10.1111/j.1365-2958.2009.07031.x. [DOI] [PubMed] [Google Scholar]
  27. Merino-Puerto V., Schwarz H., Maldener I., Mariscal V., Mullineaux C.W., Herrero A., Flores E. FraC/FraD-dependent intercellular molecular exchange in the filaments of a heterocyst-forming cyanobacterium, Anabaena sp. Mol. Microbiol. 2011;82:87–98. doi: 10.1111/j.1365-2958.2011.07797.x. [DOI] [PubMed] [Google Scholar]
  28. Mullineaux C.W., Mariscal V., Nenninger A., Khanum H., Herrero A., Flores E., Adams D.G. Mechanism of intercellular molecular exchange in heterocyst-forming cyanobacteria. EMBO J. 2008;27:1299–1308. doi: 10.1038/emboj.2008.66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Nakao M., Okamoto S., Kohara M., Fujishiro T., Fujisawa T., Sato S., Tabata S., Kaneko T., Nakamura Y. CyanoBase: the cyanobacteria genome database update 2010. Nucleic Acids Res. 2010;38:D379–D381. doi: 10.1093/nar/gkp915. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Nayar A.S., Yamaura H., Rajagopalan R., Risser D.D., Callahan S.M. FraG is necessary for filament integrity and heterocyst maturation in the cyanobacterium Anabaena sp. strain PCC 7120. Microbiology. 2007;153:601–607. doi: 10.1099/mic.0.2006/002535-0. [DOI] [PubMed] [Google Scholar]
  31. Nürnberg D.J., Mariscal V., Bornikoel J., Nieves-Morión M., Krauß N., Herrero A., Maldener I., Flores E., Mullineaux C.W. Intercellular diffusion of a fluorescent sucrose analog via the septal junctions in a filamentous cyanobacterium. mBio. 2015;6:e02109. doi: 10.1128/mBio.02109-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Di Patti F., Lavacchi L., Arbel-Goren R., Schein-Lubomirsky L., Fanelli D., Stavans J. Robust stochastic Turing patterns in the development of a one-dimensional cyanobacterial organism. PLoS Biol. 2018;16:e2004877. doi: 10.1371/journal.pbio.2004877. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Pernil R., Picossi S., Mariscal V., Herrero A., Flores E. ABC-type amino acid uptake transporters Bgt and N-II of Anabaena sp strain PCC 7120 share an ATPase subunit and are expressed in vegetative cells and heterocysts. Mol. Microbiol. 2008;67:1067–1080. doi: 10.1111/j.1365-2958.2008.06107.x. [DOI] [PubMed] [Google Scholar]
  34. Picossi S., Flores E., Herrero A. The LysR-type transcription factor PacR is a global regulator of photosynthetic carbon assimilation in Anabaena. Environ. Microbiol. 2015;17:3341–3351. doi: 10.1111/1462-2920.12800. [DOI] [PubMed] [Google Scholar]
  35. Ramos-León F., Arévalo S., Mariscal V., Flores E. Specific mutations in the permease domain of septal protein SepJ differentially affect functions related to multicellularity in the filamentous cyanobacterium. Microb. Cell. 2018;5:555–565. doi: 10.15698/mic2018.12.661. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Risser D.D., Callahan S.M. HetF and PatA control levels of HetR in Anabaena sp. strain PCC 7120. J. Bacteriol. 2008;190:7645–7654. doi: 10.1128/JB.01110-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Rivers O.S., Videau P., Callahan S.M. Mutation of sepJ reduces the intercellular signal range of a hetN-dependent paracrine signal, but not of a patS-dependent signal, in the filamentous cyanobacterium Anabaena sp. strain PCC 7120. Mol. Microbiol. 2014;94:1260–1271. doi: 10.1111/mmi.12836. [DOI] [PubMed] [Google Scholar]
  38. Schirrmeister B.E., De Vos J.M., Antonelli A., Bagheri H.C. Evolution of multicellularity coincided with increased diversification of cyanobacteria and the Great Oxidation Event. Proc. Natl. Acad. Sci. U S A. 2013;110:1791–1796. doi: 10.1073/pnas.1209927110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Schuller J., Saura P., Thiemann J., Schuller S., Gamiz-Hernandez A., Kurisu G., Nowaczyk M., Kaila V. Redox-coupled proton pumping drives carbon concentration in the photosynthetic complex I. Nat. Commun. 2020;11:1–7. doi: 10.1038/s41467-020-14347-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Sicora C.I., Appleton S.E., Brown C.M., Chung J., Chandler J., Cockshutt A.M., Vass I., Campbell D.A. Cyanobacterial psbA families in Anabaena and Synechocystis encode trace, constitutive and UVB-induced D1 isoforms. Biochim. Biophys. Acta. 2006;1757:47–56. doi: 10.1016/j.bbabio.2005.11.002. [DOI] [PubMed] [Google Scholar]
  41. Syakur M.A., Khotimah B.K., Rochman E.M.S., Satoto B.D. Integration K-means clustering method and elbow method for identification of the best customer profile cluster. IOP Conf. Ser. Mater. Sci. Eng. 2018;336:012017. [Google Scholar]
  42. Tatusov R.L., Galperin M.Y., Natale D.A., Koonin E.V.( The COG database: a tool for genome-scale analysis of protein functions and evolution. Nucleic Acids Res. 2000;28:33–36. doi: 10.1093/nar/28.1.33. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Turmo A., Gonzalez-Esquer C.R., Kerfeld C.A. Carboxysomes: metabolic modules for CO2 fixation. FEMS Microbiol. Lett. 2017;364:176. doi: 10.1093/femsle/fnx176. [DOI] [PubMed] [Google Scholar]
  44. Valladares A., Velázquez-Suárez C., Herrero A. Interactions of PatA with the divisome during heterocyst differentiation in Anabaena. mSphere. 2020;5 doi: 10.1128/mSphere.00188-20. e00188–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Wang Y., Lechno-Yossef S., Gong Y., Fan Q., Wolk C.P., Xu X. Predicted glycosyl transferase genes located outside the HEP island are required for formation of heterocyst envelope polysaccharide in Anabaena sp. strain PCC 7120. J. Bacteriol. 2007;189:5372–5378. doi: 10.1128/JB.00343-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Weiss G., Kieninger A., Maldener I., Forchhammer K., Pilhofer M. Structure and function of a bacterial gap junction analog. Cell. 2019;178:374–384. doi: 10.1016/j.cell.2019.05.055. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Wilk L., Strauss M., Rudolf M., Nicolaisen K., Flores E., Kühlbrandt W., Schleiff E. Outer membrane continuity and septosome formation between vegetative cells in the filaments of Anabaena sp. PCC 7120. Cell. Microbiol. 2011;13:1744–1754. doi: 10.1111/j.1462-5822.2011.01655.x. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Document S1. Transparent methods and figures S1–S4
mmc1.pdf (931.1KB, pdf)
Table S1. RNA-seq data for the whole genome and for differential expression genes (DEG), related to Figure 4A
mmc2.xlsx (14.4MB, xlsx)
Table S2. Gene ontology (GO) enrichment analysis of all chromosomal genes using Ontologizer, related to Figure 4A
mmc3.xlsx (406KB, xlsx)
Table S3. Comparison of upregulated and downregulated genes between the wild-type and mutants under nitrogen-rich and -poor conditions, related to Figure 4A
mmc4.xlsx (161.6KB, xlsx)

Data Availability Statement

The RNA-seq data generated in this study have been deposited in NCBI's Gene Expression Omnibus and are accessible through GEO Series accession number GSE161085 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE161085). The published article includes all the bioinformatics analysis reported in this study.

Articles from iScience are provided here courtesy of Elsevier

RESOURCES