Interactions of PatA with the divisome during heterocyst differentiation in anabaena

Interactions of PatA with the Divisome during Heterocyst Differentiation in Anabaena Ana Valladares, a,b Cristina Velázquez-Suárez, a,b Antonia Herrero a,b a Instituto de Bioquímica Vegetal y Fotosíntesis, CSIC, Seville, Spain b Universidad de Sevilla, Seville, Spain ABSTRACT The Anabaena organismic unit is a filament of communicating cells. Under conditions of nitrogen scarcity, some cells along the filament differentiate into heterocysts, which are specialized in the fixation of atmospheric N 2 and provide the vegetative cells with N 2 fixation products. At a certain stage, the differentiation process becomes irreversible, so that even when nitrogen is replenished, no return to the vegetative cell state takes place, possibly as a consequence of loss of cell division capacity. Upon N-stepdown, midcell FtsZ-rings were detected in vegetative cells, but not in differentiating cells, and this was also the case for ZipN, an essential protein that participates in FtsZ tethering to the cytoplasmic membrane and divisome organization. Later, expression of ftsZ was arrested in mature heterocysts. PatA is a protein required for the differentiation of intercalary heterocysts in Anabaena. The expression level of the patA gene was increased in differentiating cells, and a mutant strain lacking PatA exhibited enhanced FtsZ-rings. PatA was capable of direct interactions with ZipN and SepF, another essential component of the Anabaena Z-ring. Thus, PatA appears to promote inhibition of cell division in the differentiating cells, allowing progress of the differentiation process. PatA, which in mature heterocysts was detected at the cell poles, could interact also with SepJ, a protein involved in production of the septal junctions that provide cell-cell adhesion and intercellular communication in the filament, hinting at a further role of PatA in the formation or stability of the intercellular structures that are at the basis of the multicellular character of Anabaena. IMPORTANCE Anabaena is a cyanobacterial model that represents an ancient and simple form of biological multicellularity. The Anabaena organism is a filament of cohesive and communicating cells that can include cells specialized in different tasks. Thus, under conditions of nitrogen scarcity, certain cells of the filament differentiate into heterocysts, which fix atmospheric nitrogen and provide organic nitrogen to the rest of cells, which, in turn, provide heterocysts with organic carbon. Heterocyst differentiation involves extensive morphological, biochemical, and genetic changes, becoming irreversible at a certain stage. We studied the regulation during heterocyst differentiation of several essential components of the Anabaena cell division machinery and found that protein PatA, which is required for differentiation and is induced in differentiating cells, interacts with essential cell division factors and destabilizes the cell division complex. This suggests a mechanism for establishment of commitment to differentiation by inhibition of cell division. KEYWORDS Cyanobacteria, bacterial multicellularity, cell differentiation, cell division Anabaena sp. strain PCC 7120 (here referred to as Anabaena) is a model of filamentous cyanobacteria able to undertake a developmental program to form a multicellular diazotrophic bacterium. In response to nitrogen shortage, some cells of the filament differentiate into cells specialized in the fixation of atmospheric nitrogen (N 2 ), Citation Valladares A, Velázquez-Suárez C, Herrero A. 2020. Interactions of PatA with the divisome during heterocyst differentiation in Anabaena. mSphere 5:e00188-20. https://doi .org/10.1128/mSphere.00188-20. Editor Grant R. Bowman, University of Wyoming Copyright © 2020 Valladares et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International license. Address correspondence to Antonia Herrero, [email protected]. Received 5 March 2020 Accepted 7 May 2020 Published RESEARCH ARTICLE Molecular Biology and Physiology crossm May/June 2020 Volume 5 Issue 3 e00188-20 msphere.asm.org 1 20 May 2020 on October 1, 2020 at USE/BCTA.GEN UNIVERSITARIAhttp://msphere.asm.org/Downloaded from called heterocysts, which are distributed in a semiregular pattern of two heterocysts separated by an average of ca. 10 vegetative cells along the filament. Thus, the diazotrophic filament is composed of two cell types that exchange nutrients and regulatory molecules: the vegetative cells, which fix CO 2 and transfer C-rich compounds (sugars) to heterocysts, and the heterocysts, which fix N 2 and donate N-rich compounds (amino acids) to the vegetative cells (1). Additionally, at least one morphogen, PatS, is also transferred from the differentiating cells to inhibit differentiation of its neighbors, thus contributing to the setting of the heterocyst distribution pattern (2, 3). Several septal proteins, including SepJ, FraC, and FraD (4), have been implicated in the formation of the septal junctions that bridge the cytoplasms of neighboring cells and provide cohesion and intercellular communication along the filament. Vegetative cells and heterocysts differ in multiple morphological and physiological features as the result of largely different gene expression programs that are established in each cell type during the differentiation process, including multiple events of transcriptional regulation (the process that has been most extensively studied) but also events of posttranslational regulation, which are currently being dissected (5). Two principal transcriptional regulators are required for differentiation: the cAMP receptor protein (CRP)-family type NtcA global factor, which mediates regulatory responses to changes in the C-to-N balance of the cells (6), and HetR, which mediates gene activation in the differentiating cells (7). Heterocysts are terminally differentiated cells, which implies that beyond a certain point in the differentiation process no reversion to the vegetative cell entity takes place, even when the cue that triggers differentiation, i.e., nitrogen deficiency, is abrogated (8, 9). In the mature diazotrophic filament, old heterocysts are thought to undergo senescence and finally go dead. Irreversibility of heterocyst differentiation can be related to loss of cell division capacity, which is assumed to take place roughly coincidently with the commitment to differentiation. FtsZ has been reported to occur at lower levels in mature heterocysts than in vegetative cells (10, 11). However, the mechanism responsible for an asymmetric distribution of FtsZ in the diazotrophic filament is not known. Other possible regulatory events that may result in inhibition of cell division during heterocyst differentiation are also unknown. Several genes that are required for heterocyst differentiation have been related to the step of commitment. This is the case with hetC, encoding an ABC-type exporter (12) localized in the heterocyst poles that is involved in transferring the PatS morphogen to neighboring cells (13), and with hetP and some hetP homologs (14). Judging from the morphological effects of their inactivation or overexpression, two other factors that regulate heterocyst differentiation, HetF and PatA, might affect cell division. Thus, inactivation of the hetF gene leads to the presence of enlarged and elongated cells, and ectopic overexpression of hetF leads to the presence of cells smaller than those of the wild type (WT) throughout the filament (15). The PatA protein of Anabaena bears a CheY-like phosphoacceptor domain in its C terminus as well as a so-called PATAN domain of undetermined function (16). Inactivation of patA results in a phenotype of a low frequency of heterocysts that, moreover, are mainly found at the filament ends and in consequent poor growth under diazotrophic conditions (17), whereas patA overexpression increases the frequency of heterocysts (18). To gain insight into the mechanism of inhibition of cell division during heterocyst differentiation, we have monitored the expression and localization of initial cell division factors and PatA along the filament during the differentiation process. We have also studied the effects of patA deletion on the localization of the FtsZ-ring, as well as possible interactions between PatA and proteins involved in cell division or intercellular communication. RESULTS Expression and localization of FtsZ during heterocyst differentiation. To study the expression from the ftsZ gene promoter, we generated strain CSAV43, bearing a copy of the ftsZ promoter fused to gfp-mut2 in the native gene locus and keeping an Valladares et al. May/June 2020 Volume 5 Issue 3 e00188-20 msphere.asm.org 2 on October 1, 2020 at USE/BCTA.GEN UNIVERSITARIAhttp://msphere.asm.org/Downloaded from intact copy of the gene expressed from the native promoter (Fig. 1A). The spatiotemporal pattern of expression of P ftsZ -gfp was monitored in cells of strain CSAV43 grown with nitrate and transferred to medium lacking combined nitrogen, which triggers heterocyst differentiation (Fig. 1B). At the onset of N deprivation, filaments showed similar levels of green fluorescent protein (GFP) fluorescence in all the cells. Then, certain cells along the filament differentiated to heterocysts, so that after 24 h many regularly spaced mature heterocysts, recognized by loss of red fluorescence (due to dismantlement of photosystem II) and by cell morphology (including increased cell size and the presence of refringent polar granules made of cyanophycin, a product of N 2 fixation), could be detected. At that time, most heterocysts exhibited GFP fluorescence similar to that in neighboring vegetative cells (the mean level of fluorescence intensity in heterocysts was ca. 1.1 times that in vegetative cells, as counted over 40 heterocysts and 341 vegetative cells). After 48 h, GFP fluorescence in heterocysts became noticeably lower that in vegetative cells (the mean level of fluorescence intensity in heterocysts was ca. 0.7 times that in vegetative cells, as counted over 40 heterocysts and 255 vegetative cells). Thus, the ftsZ promoter was downregulated in heterocysts well after the differentiation process has been completed. We have also studied the localization of FtsZ upon N-stepdown in Anabaena strain CSSC19, which expresses a ftsZ-gfp-mut2 gene fusion placed at the ftsZ locus and which keeps an intact copy of ftsZ also expressed from the native promoter (19)(Fig. 2A). As shown in Fig. 2B, at 18 h after nitrate removal, GFP fluorescence could be detected in midcell rings in all the cells of the filament, including differentiating cells (the mean level of fluorescence intensity in proheterocysts was ca. 1.2 times that in vegetative cells, as counted over 9 proheterocysts and 126 vegetative cells). By 22 h, proheterocysts still exhibiting high GFP fluorescence, together with immature heterocysts (recognized by a thickened cell envelope) showing low GFP fluorescence, could be detected (the mean levels of fluorescence intensity in proheterocysts and immature P FIG 1 Expression of the Anabaena ftsZ gene promoter. (A) Schematic of the genome structure in the ftsZ region of strain CSAV43 (P ftsZ -gfp) in comparison to PCC 7120 (WT). The gray trace represents an inserted plasmid encoding resistance to Sm and Sp (see Materials and Methods for details). (B) Filaments of strain CSAV43 grown in BG11 medium were transferred (at a cell density of 2 ␮ g Chl/ml) to BG11 0 (no combined nitrogen) medium (N 2 ) and incubated under culture conditions. After the times indicated, filaments were observed by confocal microscopy and photographed. Time 0 denotes the start of incubation in BG11 0 . GFP fluorescence (green), cyanobacterial autofluorescence (red), and bright-field images are shown. Arrowheads point to heterocysts. The magnification is the same for all micrographs. PatA-Divisome Interactions in Anabaena May/June 2020 Volume 5 Issue 3 e00188-20 msphere.asm.org 3 on October 1, 2020 at USE/BCTA.GEN UNIVERSITARIAhttp://msphere.asm.org/Downloaded from heterocysts were ca. 0.9 and 0.2 times that in vegetative cells, respectively, as counted over 23 proheterocysts, 20 immature heterocysts, and 238 vegetative cells). After 42 h, GFP fluorescence was high in vegetative cells and low in mature heterocysts (which exhibited conspicuous polar granules) (the mean level of fluorescence intensity in heterocysts was ca. 0.2 times that in vegetative cells, as counted over 11 heterocysts and 255 vegetative cells). Thus, Z-rings appear to be lost during the process of heterocyst differentiation. Downregulation of ZipN during heterocyst differentiation. ZipN is an FtsZ- interacting protein that contributes to FtsZ tethering to the cytoplasmic membrane and is an essential part of the Z-ring in Anabaena (20). To study the localization of ZipN during heterocyst differentiation, we generated Anabaena strain CSAV39, which bears a version of the zipN gene 5=fused to the sf-gfp gene (encoding superfolder GFP) in the zipN locus (Fig. 3A). Upon nitrate withdrawal, GFP fluorescence localized in midcell rings and in intercellular regions could be detected in vegetative cells but not in either proheterocysts (see 18-h data in Fig. 3B) or heterocysts (see 22-h and 42-h data). This shows that the ZipN protein levels are negatively regulated during heterocyst differentiation. Morphological features of a patA mutant. Strain UHM101, an Anabaena mutant in which the coding region of the patA gene was deleted, forms predominantly terminal heterocysts (21) (see Fig. 4A). Here, we studied cell and filament size in strain UHM101 in comparison to wild-type Anabaena. Cell area was significantly larger in the patA mutant than in the wild type, both after incubation with nitrate (mean cell area, 8.47 ␮ m 2 for the wild type [WT] and 10.65 ␮ m 2 for the patA mutant) and, in particular, after incubation in the absence of combined nitrogen (mean vegetative cell area after 4 days P FIG 2 Localization of FtsZ during heterocyst differentiation in Anabaena. (A) Schematic of the genome structure of strain CSSC19 (ftsZ-gfp-mut2)(19) in comparison to PCC 7120 (WT). The gray trace represents an inserted plasmid encoding resistance to Sm and Sp. (B) Filaments of strain CSSC19 grown in BG11 medium were transferred (at a cell density of 2 ␮ g Chl/ml) to BG11 0 medium and incubated under culture conditions. After the times indicated, filaments were observed by confocal microscopy and photographed. GFP fluorescence (green), cyanobacterial autofluorescence (red), and bright-field images are shown. Arrowheads point to proheterocysts (yellow), immature heterocysts (orange), and mature heterocysts, exhibiting polar refringent granules (white). The magnification is the same for all micrographs. Valladares et al. May/June 2020 Volume 5 Issue 3 e00188-20 msphere.asm.org 4 on October 1, 2020 at USE/BCTA.GEN UNIVERSITARIAhttp://msphere.asm.org/Downloaded from under diazotrophic conditions, 6.69 ␮ m 2 WT and 9.76 ␮ m 2 patA mutant; mean vegetative cell area after 9 days, 6.74 ␮ m 2 WT and 9.64 ␮ m 2 patA mutant) (Fig. 4A and B). In addition, filaments were considerably longer in the mutant than in the wild type (Fig. 4C). Expression and localization of PatA. The expression of the patA gene has been described as increasing upon N-stepdown, but always taking place at low levels (17, 22). To study the spatiotemporal pattern of patA expression during differentiation, we generated Anabaena strain CSAV45, which expresses a fusion of the patA gene promoter to gfp-mut2 in the patA locus (Fig. 5A). Fluorescence was monitored in filaments of strain CSAV45 grown with nitrate and incubated during different times after removal of nitrate (Fig. 5B). Only low levels of GFP fluorescence were detected in strain CSAV45, although even at time zero (nitrate growth), the level of fluorescence was noticeably higher than in wild-type Anabaena (not bearing the gfp gene). Upon N-stepdown, GFP fluorescence increase was detected after 24 h, in cells that still resemble proheterocysts and immature heterocysts. By 48 h, GFP fluorescence in mature heterocysts had decreased to basal levels. Thus, activation of patA expression during heterocyst differentiation seems to be transitory. (Given the low expression level, we cannot rule out the possibility that the patA gene was activated earlier than the time at which accumulation of GFP was detected.) To study the localization of PatA, we generated two strains bearing patA fusions to sf-gfp inserted in the native patA locus and maintaining an intact patA gene: strain CSAV35 bears a patA gene 3=fused to sf-gfp and strain CSAV41 a patA gene 5=fused to sf-gfp (Fig. 6A). In media supplemented with nitrate, both strains, CSAV35 and CSAV41, grew well, and their filament morphology was similar to that of the wild type (Fig. 6C). In the absence of combined nitrogen, both strains were able to grow, although growth was somewhat impaired in comparison to that seen with the wild type (Fig. 6B). Strains CSAV35 and CSAV41 were able to differentiate terminal and intercalary heterocysts (Fig. 6C), although, with regard to the wild type, differentiation was somewhat delayed, and the frequency of intercalary heterocysts was lower in P FIG 3 Localization of ZipN during heterocyst differentiation in Anabaena. (A) Schematic of the genome structure of strain CSAV39 (sf-gfp-zipN) in comparison to PCC 7120 (WT). (B) Filaments of strain CSAV39 were treated as described in the legend for Fig. 2, observed by confocal microscopy, and photographed. Arrowheads point to proheterocysts (yellow), immature heterocysts (orange), and mature heterocysts that exhibit polar refringent granules (white). The magnification is the same for all micrographs. PatA-Divisome Interactions in Anabaena May/June 2020 Volume 5 Issue 3 e00188-20 msphere.asm.org 5 on October 1, 2020 at USE/BCTA.GEN UNIVERSITARIAhttp://msphere.asm.org/Downloaded from CSAV35 than in CSAV41 or the wild type. In both the CSAV35 and CSAV41 mutant strains, occasional aberrant cells could be observed, and those cells were more noticeable under conditions of rapid growth (see Fig. 6C). This suggests some degree of interference with cell division, which might have resulted from the presence of two patA copies. GFP fluorescence was monitored in strains CSAV35 and CSAV41. Only occasional weak signals were detected, and, in both strains, they were seen as condensed spots in the polar regions of heterocysts. These signals were absent from strain PCC 7120 (lacking GFP) (Fig. 7). Localization of FtsZ in a patA mutant background. To test whether the mutation of the patA gene had any effect on the conformation of FtsZ-rings, we generated strain CSAV38, which bears in UHM101 (patA background) the same ftsZ-gfp-mut2 gene fusion present in strain CSSC19 (wild-type background; see above) (Fig. 8A). In nitrate-grown filaments, both strain CSSC19 and strain CSAV38 showed GFP fluorescence in midcell rings. Notably, the intensity of GFP fluorescence was stronger in CSAV38 than in CSSC19 (the mean midcell fluorescence intensity in CSAV38 was ca. 1.6 times that in CSSC19, as FIG 4 Cell and filament size in a patA mutant. Strains PCC 7120 (WT) and UHM101 (patA mutant) grown in BG11 medium (NO 3 ⫺ ) were transferred to BG11 0 medium (N 2 ) and incubated under culture conditions. At time zero and after 4 and 9 days on incubation in BG11 0 medium, filaments were observed with a bright-field microscope and photographed. (A) Filaments after 4 days of incubation in BG11 0 . Purple arrowheads indicate intercalary heterocysts and orange arrowheads terminal heterocysts. (B) Photographs were used for determination of cell area (vegetative cell area in the case of N 2 cultures) as described in Materials and Methods. A total of 200 to 250 cells of each strain were measured under each condition. A box plot representation of the data is shown. Mean values are represented by black dots. (C) Filament length was counted as the number of cells per filament over 50 to 60 filaments for each strain and condition. Filaments with more than 100 cells were counted as representing 100 cells. WT, green; patA mutant, pink. Valladares et al. May/June 2020 Volume 5 Issue 3 e00188-20 msphere.asm.org 6 on October 1, 2020 at USE/BCTA.GEN UNIVERSITARIAhttp://msphere.asm.org/Downloaded from counted over 50 cells of CSAV38 and 59 cells of CSSC19; Student’s ttest, P⫽10 ⫺17 ) (see Fig. 8B for representative images), suggesting that the deletion of patA has a positive effect on the formation or stability of the Z-ring. Upon N-stepdown, strain CSAV38, similarly to its parental UHM101 strain, produced mostly terminal heterocysts. After incubation in the absence of combined nitrogen, the levels of fluorescence intensity in vegetative cells of CSSC19 and CSAV38 were similar (the mean midcell fluorescence intensity in CSAV38 was ca. 0.9 times that in CSSC19, as counted over 51 cells of CSAV38 and 55 cells of CSSC19; Student’s ttest, P⫽0.09). However, a noticeable difference P P P P P P FIG 5 Expression of the Anabaena patA gene promoter. (A) Schematic of the genome structure of strain CSAV45 (P patA -gfp) in comparison to PCC 7120 (WT). The gray trace represents an inserted plasmid encoding resistance to Sm and Sp (see Materials and Methods for details). (B) Filaments of strain CSAV45 grown in BG11 medium were transferred (at a cell density of 2 ␮ g Chl/ml) to BG11 0 medium and incubated under culture conditions. After the times indicated, filaments were observed by confocal microscopy and photographed. Time 0 denotes the start of incubation in BG11 0 . GFP fluorescence (green), cyanobacterial autofluorescence (red), and bright-field images are shown. Arrowheads point to proheterocysts (yellow), immature heterocysts (orange), and mature heterocysts that exhibit polar refringent granules (white). PatA-Divisome Interactions in Anabaena May/June 2020 Volume 5 Issue 3 e00188-20 msphere.asm.org 7 on October 1, 2020 at USE/BCTA.GEN UNIVERSITARIAhttp://msphere.asm.org/Downloaded from between the two strains was that CSAV38 did not exhibit intercalary cells devoid of GFP fluorescence, as observed in CSSC19 (Fig. 8C; see also Fig. 2). PatA interactions with Anabaena cell division factors. Given the apparent effects of patA alterations on cell morphology and detection of the cell division ring, we used bacterial adenylate cyclase two-hybrid system (BACTH) analysis to test possible direct interactions of PatA with identified components of the Anabaena Z-ring, namely, the proteins FtsZ, ZipN, Ftn6, and SepF. Also, because of the apparent polar localization of PatA in the heterocysts, we tested interactions with Anabaena septal junction proteins SepJ, FraC, and FraD. For that, all four possible versions of PatA (T18-PatA, PatA-T18, T25-PatA, and PatA-T25) were generated. A strong interaction of T18-PatA with T25- ZipN was detected (as already published [20], T25-ZipN was the only ZipN fusion that we were able to clone). A clear interaction was also detected involving T18-PatA and SepF-T25. Finally, a weak but significant interaction was detected between T25-PatA P FIG 6 Growth and morphology of strains expressing GFP fusions to PatA. (A) Schematic of the genome structure of strain CSAV35 (patA-sf-gfp) and strain CSAV41 (sf-gfp-patA) in comparison to PCC 7120 (WT). The gray trace represents an inserted plasmid encoding resistance to Sm and Sp (see Materials and Methods for details). (B) Cultures in BG11 medium (containing NO 3 ⫺ ) were used to inoculate flasks with BG11 0 medium (at a cell density equivalent to 1 ␮ g Chl/ml), which were incubated under culture conditions and photographed after 4 days. Strain UHM101 is a patA mutant (see the text). (C) Filaments from flasks inoculated with 0.5 ␮ g Chl/ml and photographed after 43 h. Purple arrowheads indicate intercalary heterocysts, orange arrowheads terminal heterocysts, and red brackets stretches of aberrant cells in the filament. The magnification is the same for all micrographs. Valladares et al. May/June 2020 Volume 5 Issue 3 e00188-20 msphere.asm.org 8 on October 1, 2020 at USE/BCTA.GEN UNIVERSITARIAhttp://msphere.asm.org/Downloaded from and SepJ-T18 (Fig. 9). No significant interaction was detected involving C-terminal fusions to PatA (i.e., neither PatA-T18 nor PatA-T25), suggesting that these fusions could lead to the production of PatA versions that are not active for BACTH analysis. The interaction between PatA and ZipN was also assessed by copurification assays. A version of ZipN that included a 6⫻His tag fused to its N-terminal end and a version with PatA N-terminally fused to a Strep-tag were generated. Cell-free extracts of E. coli expressing each of these proteins alone or mixed together were incubated and then passed through a His-select column, and the presence of Strep-PatA in the eluted fractions was assessed by immunoblotting with antibodies against the Strep-tag and by matrix-assisted laser desorption ionization–time of flight (MALDI-TOF). As shown in Fig. 10, Strep-PatA, as revealed by immunoblotting, was tightly retained in the column only when it had been previously incubated and loaded together with His-ZipN. In the fractions that, according to Western blotting results, contained His-ZipN and Strep- PatA, MALDI-TOF analysis rendered PatA protein sequence coverage of 34% and a Mascot score of 177 (protein scores greater than 93 were considered significant; P⬍0.05). For ZipN, sequence coverage of 12% and a Mascot score of 102 were obtained. Taken together, the results of the BACTH and copurification analyses indicate that PatA interacts with ZipN. DISCUSSION In the filamentous cyanobacterium Anabaena,heterocysts specialized in the fixation of atmospheric nitrogen differentiate in a semiregular spatial pattern under conditions of nitrogen scarcity. Heterocysts are terminally differentiated cells that do not divide or P FIG 7 Localization of PatA during heterocyst differentiation. (A) Schematic of the genome structure of strain CSAV35 (patA-sf-gfp) and strain CSAV41 (sf-gfp-patA) in comparison to PCC 7120 (WT). (B) Filaments of strains PCC7120, CSAV35, and CSAV41 grown in BG11 medium were transferred to BG11 0 medium and incubated under culture conditions. After 22 h, filaments were observed under a fluorescence microscope and photographed. GFP fluorescence (green) and bright-field images are shown. Arrowheads point to heterocysts. The magnification is the same for all micrographs. PatA-Divisome Interactions in Anabaena May/June 2020 Volume 5 Issue 3 e00188-20 msphere.asm.org 9 on October 1, 2020 at USE/BCTA.GEN UNIVERSITARIAhttp://msphere.asm.org/Downloaded from and PatS in heterocyst pattern formation. J Bacteriol 188:1808–1816. https://doi.org/10.1128/JB.188.5.1808-1816.2006. 22. Bastet L, Boileau C, Bédu S, Janicki A, Latifi A, Zhang CC. 2010. NtcA regulates patA expression in Anabaena sp. PCC 7120. J Bacteriol 192: 5257–5259. https://doi.org/10.1128/JB.00640-10. 23. Wang Y, Xu X. 2005. Regulation by hetC of genes required for heterocyst differentiation and cell division in Anabaena sp. strain PCC 7120. J Bacteriol 187:8489–8493. https://doi.org/10.1128/JB.187.24.8489-8493 .2005. 24. Sakr S, Thyssen M, Denis M, Zhang C-C. 2006. Relationship among several key cell cycle events in the developmental cyanobacterium Anabaena sp. strain PCC 7120. J Bacteriol 188:5958–5965. https://doi .org/10.1128/JB.00524-06. 25. Yang Y, Huang XZ, Wang L, Risoul V, Zhang CC, Chen WL. 2013. Phenotypic variation caused by variation in the relative copy number of pDU1-based plasmids expressing the GAF domain of Pkn41 or Pkn42 in Anabaena sp. PCC 7120. Res Microbiol 164:127–135. https://doi.org/10 .1016/j.resmic.2012.10.010. 26. Rippka R, Deruelles J, Waterbury JB, Herdman M, Stanier RY. 1979. Generic assignments, strain histories and properties of pure cultures of cyanobacteria. J Gen Microbiol 111:1–61. https://doi.org/10.1099/00221287-111-1-1. 27. Mackinney G. 1941. Absorption of light by chlorophyll solutions. J Biol Chem 140:109–112. 28. Olmedo-Verd E, Muro-Pastor AM, Flores E, Herrero A. 2006. Localized induction of the ntcA regulatory gene in developing heterocysts of Anabaena sp. strain PCC 7120. J Bacteriol 188:6694–6699. https://doi .org/10.1128/JB.00509-06. 29. Elhai J, Vepritskiy A, Muro-Pastor AM, Flores E, Wolk CP. 1997. Reduction of conjugal transfer efficiency by three restriction activities of Anabaena sp. strin PCC 7120. J Bacteriol 179:1998–2005. https://doi.org/10.1128/ jb.179.6.1998-2005.1997. 30. Valladares A, Rodríguez V, Camargo S, Martínez-Nöel GMA, Herrero A, Luque I. 2011. Specific role of the cyanobacterial PipX factor in heterocysts of Anabaena sp. strain PCC 7120. J Bacteriol 193:1172–1182. https://doi.org/10.1128/JB.01202-10. 31. Black TA, Cai YP, Wolk CP. 1993. Spatial expression and autoregulation of hetR a gene involved in the control of heterocyst development in Anabaena. Mol Microbiol 9:77–84. https://doi.org/10.1111/j.1365-2958 .1993.tb01670.x. 32. Karimova G, Dautin N, Ladant D. 2005. Interaction network among Escherichia coli membrane proteins involved in cell division as revealed by bacterial two-hybrid analysis. J Bacteriol 187:2233–2243. https://doi .org/10.1128/JB.187.7.2233-2243.2005. 33. Ramos-León F, Mariscal V, Frías JE, Flores E, Herrero A. 2015. Divisomedependent subcellular localization of cell-cell joining protein SepJ in the filamentous cyanobacterium Anabaena. Mol Microbiol 96:566–580. https://doi.org/10.1111/mmi.12956. 34. Napolitano M, Rubio MA, Camargo S, Luque I. 2013. Regulation of internal promoters in a zinc-responsive operon is influenced by transcription from upstream promoters. J Bacteriol 195:1285–1293. https:// doi.org/10.1128/JB.01488-12. Valladares et al. May/June 2020 Volume 5 Issue 3 e00188-20 msphere.asm.org 16 on October 1, 2020 at USE/BCTA.GEN UNIVERSITARIAhttp://msphere.asm.org/Downloaded from