Septal Junctions in Filamentous Heterocyst-Forming Cyanobacteria

Septal Junctions in Filamentous Heterocyst-forming 1" Cyanobacteria 2" 3" Enrique Flores1, Antonia Herrero1, Karl Forchhammer2 and Iris Maldener2 4" 5" 1Instituto de Bioquímica Vegetal y Fotosíntesis, CSIC and Universidad de Sevilla, 6" Américo Vespucio 49, E-41092 Seville, Spain 7" 2Department of Microbiology/Organismic Interactions, University of Tübingen, Auf der 8" Morgenstelle 28, D-72076 Tübingen, Germany 9" 10" Corresponding author: Flores, E. ([email protected]). 11" 12" Key words: Anabaena; cell envelope; cyanobacteria; intercellular communication; 13" multicellularity. 14" 15" 16" Abstract 17" In the filaments of heterocyst-forming cyanobacteria, septal junctions that traverse 18" the septal peptidoglycan join adjacent cells allowing intercellular communication. 19" Perforations in the septal peptidoglycan have been observed, and proteins involved 20" in the formation of such perforations and putative protein components of the 21" septal junctions have been identified, but their relationships are debated. 22" 23" 24" " 2" The N2-fixing Cyanobacterial Filament 25" Some cyanobacteria grow as chains of cells (filaments or trichomes) that can be 26" hundreds of cells long. The cyanobacteria bear a Gram-negative type of cell envelope, 27" and the cyanobacterial filament consists of individual cells surrounded by their 28" peptidoglycan layers but enclosed in a continuous outer membrane that defines a 29" continuous periplasm [1]. Under nitrogen-limiting conditions, the filaments of 30" heterocyst-forming cyanobacteria contain two cell types with specialized functions: the 31" vegetative cells that carry out oxygenic photosynthesis and the heterocysts that perform 32" nitrogen fixation (Figure 1A). Growth of the filament as the organismic unit depends on 33" the coordinated activity of these cell types in a simple example of multicellularity. An 34" intercellular exchange of regulators and nutrients takes place in the filament, and the 35" structures and mechanisms involved are now actively being investigated. Here we wish 36" to provide a clarifying view of the current knowledge on septal communication 37" structures in the cyanobacterial filament, highlighting issues that need further research. 38" 39" Intercellular Molecular Exchange 40" Heterocysts differentiate from vegetative cells, and heterocyst differentiation is 41" subjected to regulation by intercellularly transferred inhibitors produced by developing 42" and mature heterocysts [2]. Additionally, in the diazotrophic filament, the vegetative 43" cells provide the heterocysts with reduced carbon, including sugars, and the heterocysts 44" feed the vegetative cells with fixed nitrogen in the form of amino acids [1]. Two 45" possible pathways have been discussed for intercellular molecular exchange in 46" heterocyst-forming cyanobacteria: the continuous periplasm and direct cell-cell 47" connecting structures [1]. Intercellular communication involving the continuous 48" periplasm would require transport across the cytoplasmic membranes of vegetative cells 49" " 3" and heterocysts, and cytoplasmic membrane transporters required for optimal 50" diazotrophic growth have indeed been identified [1]. On the other hand, the use of 51" fluorescent tracers, including calcein, to probe intercellular molecular exchange has 52" shown that a rapid intercellular exchange of the tracer takes place in the filament, and 53" that this exchange has properties of diffusion [3]. This observation suggests the 54" presence of structures directly connecting adjacent cells in the filaments. 55" 56" Septal Junctions and Septal Peptidoglycan Nanopores 57" Structures observed by electron microscopy apparently joining adjacent cells in the 58" filament (Figure 1B,C) have been known for years as “microplasmodesmata” [1, 2], 59" then termed “septosomes” [4] and now “septal junctions” to better reflect their possible 60" analogy to metazoan gap junctions [5, 6]. To mediate intercellular molecular transfer, 61" these structures must traverse the septal peptidoglycan. Consistently, perforations 62" termed “nanopores” have been observed in isolated septal peptidoglycan [7]. The septal 63" “channels” recently observed by electron tomography [8] would correspond to such 64" nanopores. A peptidoglycan amidase of the AmiC type has been identified to be 65" responsible for nanopore formation [7]. Furthermore, a novel type of peptidoglycan-66" binding protein, SjcF1, influences their size [6]. Both proteins have a preferential 67" location in the septal regions of the filaments and corresponding mutants exhibit 68" impaired intercellular exchange of calcein [6, 9]. 69" 70" Putative Septal Junction Proteins 71" The septal junctions likely contain protein [4]. In the model heterocyst-forming 72" cyanobacterium Anabaena sp. strain PCC 7120 (Figure 1), SepJ (also known as FraG), 73" FraC, and FraD are integral membrane proteins that, as shown with GFP fusions, are 74" " 4" located at the cell poles in the intercellular septal regions of the filament, with SepJ 75" being located" in" more" focused" regions than FraC and FraD (see [10, 11] and 76" references therein). Mutants lacking these proteins have a decreased number of 77" nanopores and are impaired in the intercellular exchange of fluorescent tracers, relating 78" these proteins to the septal junctions that traverse the nanopores [3, 5]. SepJ has a long 79" extra-membrane section containing large coiled-coil motifs [10], which are known to 80" participate in protein-protein interactions. This led to the hypothesis that SepJ proteins 81" from adjacent cells interact and contribute to the formation of septal junctions, implying 82" that the extra-membrane section of SepJ is periplasmic [10]. FraD has an extra-83" membrane section that, as shown by immunogold-labeling, is located in the septum 84" between adjacent vegetative cells [11], making FraD a possible component of septal 85" junctions. FraC and FraD are encoded in an operon (fraCDE) that is strongly conserved 86" in filamentous cyanobacteria and, based on the similar phenotype of their mutants, 87" could work together [11]. 88" The extra-membrane section of SepJ has now been proposed to be cytoplasmic, 89" which would imply that SepJ is not a component of septal junctions [8]. Instead, it was 90" proposed that SepJ is a docking protein for septal channels [8]. Because the C-terminus 91" of SepJ is most likely cytoplasmic [10, 12], a periplasmic or cytoplasmic location of its 92" N-terminal extra-membrane section will depend on the number (odd or even, 93" respectively) of transmembrane segments in its integral membrane section. Different 94" protein topology prediction programs render diverse numbers (from 9 to 11) of 95" transmembrane segments for Anabaena SepJ [6]. Furthermore, a particular program, 96" TMHMM (http://www.cbs.dtu.dk/services/TMHMM/), predicts 9, 10 or 11 97" transmembrane segments for the SepJ protein from different heterocyst-forming 98" cyanobacteria whose genomic sequence is available. Thus, in the absence of structural 99" " 5" information, the number of transmembrane segments in SepJ and, hence, its topology 100" are uncertain. Nonetheless, a SepJ topology with a periplasmic N-terminal extra-101" membrane section is supported by available experimental evidence: the SepJ extra-102" membrane section interacts with the peptidoglycan-binding protein SjcF1 [6] and with a 103" periplasmic domain of the divisome protein FtsQ [12]. Additionally, immunogold-104" labeling of the SepJ coiled-coil domain (detected in a strain overexpressing SepJ) 105" clearly indicates a preferential localization in the septa between vegetative cells [8]. 106" Localization in a ring, whose position is similar to that of a Z ring, of GFP fused to the 107" extra-membrane section of SepJ has been interpreted to suggest a cytoplasmic location 108" [8]. Nevertheless, the interaction of SepJ with FtsQ [12] suggests rather localization in 109" the periplasm of the SepJ extra-membrane section fused to the GFP. The possible 110" mechanism of translocation into the periplasm is however unknown and represents an 111" important research objective. More generally, investigating the topology of SepJ is 112" imperative. 113" Summarizing, SepJ and FraD are candidate components of the septal junctions, 114" and FraC is a likely companion of FraD. Interestingly, FraC also appears to interact 115" with the nanopore-related, peptidoglycan-binding protein SjcF1 [6], further indicating a 116" relationship between putative septal junction proteins and nanopores. Whereas a similar 117" decrease (by about 90%) in the number of nanopores in sepJ and fraCD mutants [5] 118" suggests that SepJ and FraCD together are needed to make a normal number of 119" nanopores, differential impairment in the intercellular transfer of different fluorescent 120" tracers (calcein, 5-carboxyfluorescein and the sucrose analog esculin) in those mutants 121" suggests that septal junctions with somewhat different specificities are present in the 122" cyanobacterial filament [5, 11]. 123" 124" " 6" Heterocyst-vegetative Cell Junctions 125" In the heterocyst-vegetative cell septum, in which the heterocyst “neck” contacts the 126" polar region of a vegetative cell (Figure 1C), SepJ has a distinct location: whereas SepJ-127" GFP is seen as a single fluorescent spot in septa between vegetative cells, two spots are 128" seen in the heterocyst-vegetative cell septa [1, 8, 10; see Figure 1D]. This observation 129" suggests a re-localization of SepJ that may accompany the differentiation of the 130" heterocyst neck. In the heterocysts, immunogold-labeling has been reported to place the 131" SepJ coiled-coil domain surrounding the polar region where the cyanophycin granule (a 132" multi-L-arginyl-poly [L-aspartic acid] reserve polymer), lost during sample preparation, 133" is normally located [8]. We consider that this connection of SepJ with the cyanophycin 134" granule needs corroboration. If, as discussed earlier, the SepJ extra-membrane section 135" has a periplasmic location, localization of the coiled-coil domain towards the 136" cyanophycin granule in the heterocyst neck could imply that the heterocyst-polar 137" cyanophycin granule is located inside a compartment topologically equivalent to the 138" periplasm. Such compartment ought to be surrounded by a membrane, reminiscent of 139" the thylakoid lumen surrounded by the thylakoid’s photosynthetic membranes. 140" Although the presence of a membrane specifically surrounding the heterocyst-polar 141" cyanophycin granule has not been described, this granule is frequently seen in electron 142" microscopy surrounded by an electro-dense layer of unknown composition [9-11]. 143" Investigating the fine structure of the heterocyst neck would thus be of interest. 144" 145" Concluding Remarks 146" Multicellularity requires cell-cell attachment and communication, and the septal 147" junctions discussed here appear to represent a unique development in evolution that 148" contributes to multicellularity in cyanobacteria. Three putative protein components of 149" " 7" the septal junctions, SepJ, FraC, and FraD, have been identified, but the precise 150" composition of the septal junctions and their possible interactions with peptidoglycan 151" and peptidoglycan-related proteins remain to be fully explored. Finally, the special 152" construction of the heterocyst-polar regions, including the heterocyst-vegetative cell 153" septa, remains intriguing and deserves further research. 154" 155" Acknowledgments 156" Research was supported by grant numbers BFU2014-56757-P (EF) and BFU2013-157" 44686-P (AH) from Plan Nacional de Investigación, Spain, co-financed by the 158" European Regional Development Fund, and by Deutsche Forschungsgemeinschaft, 159" grant number SFB766 (IM, KF). 160" 161" 162" 163" References 164" 165" 1 Flores, E., and Herrero, A. (2010) Compartmentalized function through cell differentiation in 166" filamentous cyanobacteria. Nat. Rev. Microbiol. 8, 39-50. 167" 2 Kumar, K., Mella-Herrera, R.A., and Golden, J.W. (2010) Cyanobacterial heterocysts. Cold 168" Spring Harb. Perspect. Biol. 2(4), a000315. 169" 3 Mullineaux, C.W., Mariscal, V., Nenninger, A., Khanum, H., Herrero, A., Flores, E., and 170" Adams, D.G. (2008) Mechanism of intercellular molecular exchange in heterocyst-forming 171" cyanobacteria. EMBO J. 27, 1299-1308. 172" 4 Wilk, L., Strauss, M., Rudolf, M., Nicolaisen, K., Flores, E., Kühlbrandt, W., and Schleiff, E. 173" (2011) Outer membrane continuity and septosome formation between vegetative cells in the 174" filaments of Anabaena sp. PCC 7120. Cell. Microbiol. 13, 1744-1754. 175" 5 Nürnberg, D.J., Mariscal, V., Bornikoel, J., Nieves-Morión, M., Krauß, N., Herrero, A., 176" Maldener, I., Flores, E., and Mullineaux, C.W. (2015) Intercellular diffusion of a fluorescent 177" sucrose analog via the septal junctions in a filamentous cyanobacterium. MBio 6(2), e02109-14. 178" 6 Rudolf, M., Tetik, N., Ramos-León, F., Flinner, N., Ngo, G., Stevanovic, M., Burnat, M., 179" Pernil, R., Flores, E., and Schleiff, E. (2015) The peptidoglycan-binding protein SjcF1 180" influences septal junction function and channel formation in the filamentous cyanobacterium 181" Anabaena. MBio 6(4), e00376-15. 182" " 8" 7 Lehner, J., Berendt, S., Dörsam, B., Pérez, R., Forchhammer, K., and Maldener, I. (2013) 183" Prokaryotic multicellularity: a nanopore array for bacterial cell communication. FASEB J. 27, 184" 2293–2300. 185" 186" 8 Omairi-Nasser, A., Mariscal, V., Austin, J.R. 2nd, Haselkorn, R. (2015) Requirement of Fra 187" proteins for communication channels between cells in the filamentous nitrogen-fixing 188" cyanobacterium Anabaena sp. PCC 7120. Proc. Natl. Acad. Sci. USA 112, E4458-4464. 189" 9 Berendt, S., Lehner, J., Zhang, Y.V., Rasse, T.M., Forchhammer, K., and Maldener, I. (2012) 190" Cell wall amidase AmiC1 is required for cellular communication and heterocyst development in 191" the cyanobacterium Anabaena PCC 7120 but not for filament integrity. J. Bacteriol. 194, 5218-192" 5227. 193" 10 Flores, E., Pernil, R., Muro-Pastor, A.M., Mariscal, V., Maldener, I., Lechno-Yossef, S., 194" Fan, Q., Wolk, C.P., and Herrero, A. (2007) Septum-localized protein required for filament 195" integrity and diazotrophy in the heterocyst-forming cyanobacterium Anabaena sp. strain PCC 196" 7120. J. Bacteriol. 189, 3884-3890. 197" 11 Merino-Puerto, V., Schwarz, H., Maldener, I., Mariscal, V., Mullineaux, C.W., Herrero, A., 198" and Flores, E. (2011) FraC/FraD-dependent intercellular molecular exchange in the filaments of 199" a heterocyst-forming cyanobacterium, Anabaena sp. Mol. Microbiol. 82, 87–98. 200" 12 Ramos-León, F., Mariscal, V., Frías, J.E., Flores, E., and Herrero, A. (2015) Divisome-201" dependent subcellular localization of cell-cell joining protein SepJ in the filamentous 202" cyanobacterium Anabaena. Mol. Microbiol. 96, 566-580. 203" 204" Figure 1. The Filament of an N2-fixing Heterocyst-forming Cyanobacterium, 205" Anabaena sp. Strain PCC 7120. (A) Optical micrograph showing fragments of 206" Anabaena filaments consisting of vegetative cells and heterocysts (some indicated by 207" arrows). (B) Electron micrograph of a portion of a filament of Anabaena showing the 208" septum between two vegetative cells in which thin structures perpendicular to the 209" cytoplasmic membranes of the adjacent cells are visible (white arrows). These 210" structures are known as septal junctions and thought to join the adjacent cells. (C) 211" Electron micrograph of the junction between a heterocyst (top) and a vegetative cell 212" (bottom) where septal junctions are visible (white arrows). The polar region of the 213" heterocyst known as the ‘heterocyst neck’ is indicated (Het neck). The place of the 214" " 9" cyanophycin granule (a cell inclusion that serves as a nitrogen reservoir), lost during 215" sample preparation, is seen as a split white space in the heterocyst neck and close to it. 216" (Samples prepared and electron micrographs taken as described [7, 9].) (D) Fragment of 217" a filament of Anabaena sp. strain CSAM137 containing vegetative cells and a 218" heterocyst (Het). Strain CSAM137 is Anabaena sp. strain PCC 7120 bearing a sepJ-gfp 219" gene fusion [10]. Bright field (top) and GFP green fluorescence (bottom) are shown. 220" The GFP fluorescence is observed as single spots in the septa between vegetative cells 221" (single arrow), a localization that identifies SepJ as a possible component of septal 222" junctions, and as two spots in the heterocyst-vegetative cell septa (double arrow). 223" (Micrographs taken as described [10].) All micrographs are from the authors’ 224" laboratories. 225"