The LysR-type transcription factor PacR is a global regulator of photosynthetic carbon assimilation in Anabaena

For Peer Review Only The LysR - type transcription factor PacR is a global regulator of photosynthetic carbon assimilation in Anabaena Journal: Environmental Microbiology and Environmental Microbiology Reports Manuscript ID: EMI-2014-1511.R1 Manuscript Type: EMI - Research article Journal: Environmental Microbiology Date Submitted by the Author: 22-Jan-2015 Complete List of Authors: Picossi, Silvia; Consejo Superior de Investigaciones Científicas, Instituto de Bioq´uímica Vegetal y Fotosíntesis Flores, Enrique; Consejo Superior de Investigaciones Científicas, Instituto de Bioq´uímica Vegetal y Fotosíntesis Herrero, Antonia; Consejo Superior de Investigaciones Científicas, Instituto de Bioquímica Vegetal y Fotosíntesis Keywords: bacteria, environmental signal/stress responses, gene expression/regulation Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 1 The LysR-type transcription factor PacR is a global regulator of photosynthetic 1 carbon assimilation in Anabaena 2 3 Silvia Picossi, Enrique Flores and Antonia Herrero* 4 5 Instituto de Bioquímica Vegetal y Fotosíntesis, Consejo Superior de Investigaciones 6 Científicas and Universidad de Sevilla, Américo Vespucio 49, E-41092, Seville, Spain. 7 8 9 *Corresponding author. Tel.: +34 954489522. Fax: +34 954460165. E-mail address: 10 [email protected] 11 12 Keywords: ChIP; Cyanobacteria; Oxygenic phototrophy; Photoprotection; RuBisCo 13 14 Running title: Photosynthetic carbon assimilation regulator 15 16 Accession link to data: 17 http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?token=afcxwkacxpydfax&acc=GSE58861 18 19 20 21 22 Page 1 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 2 Summary 23 Cyanobacteria perform water-splitting photosynthesis and are important primary 24 producers impacting the carbon and nitrogen cycles at global scale. They fix CO 2 25 through ribulose bisphosphate carboxylase/oxygenase (RuBisCo) and have 26 evolved a distinct CO 2 concentrating mechanism (CCM) that builds high CO 2 27 concentrations in the vicinity of RuBisCo favoring its carboxylase activity. 28 Filamentous cyanobacteria such as Anabaena fix CO 2 in photosynthetic 29 vegetative cells, which donate photosynthate to heterocysts that rely on a 30 heterotrophic metabolism to fix N 2 . CCM elements are induced in response to 31 inorganic carbon limitation, a cue that exposes the photosynthetic apparatus to 32 photodamage by over-reduction. An Anabaena mutant lacking the LysR-type 33 transcription factor All3953 grows poorly and dies under high light. The rbcL 34 operon encoding RuBisCo is induced upon carbon limitation in the wild type but 35 not in the mutant. ChIP-Seq analysis was used to globally identify All3953 targets 36 under carbon limitation. Targets include, besides rbcL, genes encoding CCM 37 elements, photorespiratory pathway, photosystemand electron transport-38 related components, and factors, including flavodiiron proteins, with a 39 demonstrated or putative function in photoprotection. qRT-PCR analysis of 40 selected All3953 targets showed regulation in the wild type but not in the mutant. 41 All3953 (PacR) is a global regulator of carbon assimilation in an oxygenic 42 photoautotroph. 43 44 45 Page 2 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 3 Introduction 46 As the organisms that developed oxygenic photosynthesis, cyanobacteria have played 47 a crucial role in Earth's history and the evolution of life in our planet. Indeed, the 48 production of O 2 as a result of cyanobacterial activity was responsible for the oxidation 49 of the Earth's atmosphere about 2.5-2.3 billion years ago (Lyons et al., 2014). 50 Furthermore, all the extant plastids of eukaryotic algae and plants are of cyanobacterial 51 origin. Cyanobacteria were the first organisms to link the activity of the two types of 52 photosystems (PSI and PSII), which allowed the generation of high electrochemical 53 potential, and to combine them with a H 2 O-splitting complex. Nowadays, most 54 cyanobacteria are phototrophs relying on oxygenic photosynthesis to generate ATP 55 and reducing equivalents for the fixation of CO 2 and the assimilation of inorganic 56 nitrogen. Indeed, they are responsible for an important fraction of the primary 57 productivity in the Earth's oceans, where they are important CO 2 and N 2 fixers, thus 58 impacting the C and N cycles at a global scale (Knoll 2008; Price et al., 2008). 59 The enzyme responsible for the bulk of CO 2 fixation in the biosphere is ribulose-60 1,5-bisphosphate carboxylase/oxygenase (RuBisCo), which has a relatively low affinity 61 for CO 2 and, moreover, it can also accept O 2 as a substrate. Compensating for this 62 relatively low performance, RuBisCo is considered the most abundant enzyme on 63 Earth. As a carboxylase, RuBisCo catalyzes the first step of the Calvin-Benson-64 Bassham (CBB) cycle, i.e., the incorporation of atmospheric CO 2 into ribulose-1,5-65 bisphosphate to give two molecules of 3-phosphoglycerate. As an oxygenase, it 66 catalyzes the incorporation of O 2 into ribulose-bisphosphate, which produces 2-67 phosphoglycolate that leads to photorespiration with a subsequent loss of fixed C and 68 energy. To increase the efficiency of CO 2 fixation, cyanobacteria have developed a 69 distinct CO 2 concentrating mechanism (CCM) constituted by inorganic carbon (C i ) 70 transporters that incorporate bicarbonate and CO 2 into the cell, and a proteinaceous 71 compartment, the carboxysome, where RuBisCo, together with carbonic anhydrase, is 72 Page 3 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 4 confined (Price et al., 2008; Cameron et al., 2014). Three high-affinitiy bicarbonate 73 transporters (the ABC-type Cmp, also called BCT1, and the Na + -dependent BicA and 74 SbtA), and two CO 2 transporters (the high-affinity NDH-I 3 and the low-affinity NDH-I 4 ) 75 have been identified in different cyanobacteria (see Price, 2011). Through CCM, Ci in 76 the form of CO 2 can concentrate in the vicinity of cyanobacterial RuBisCo, to allow high 77 specific activity for production of 3-phosphoglycerate to levels much higher than in 78 plants (Cameron et al., 2014). Indeed, components of cyanobacterial CCM have been 79 transformed in tobacco, with the result of improved specific activity of CO 2 fixation, 80 which represents a step towards improved photosynthesis in plants (Lin et al., 2014). 81 In unicellular cyanobacteria, CCM elements are regulated by C i availability. 82 Especially genes encoding C i transporters are induced, whereas the structural genes 83 for carboxysome components and the rbcL/S genes encoding RuBisCo are only 84 moderately responsive (see Price et al., 2008; Cameron et al., 2014). In chemotrophic 85 bacteria the process of CO 2 fixation and the response to C i limitation are usually 86 controlled by LysR-type transcriptional regulators (LTTRs). The genes encoding the 87 enzymes of the CBB cycle are usually found in clusters regulated by CbbR factors, 88 which constitute a sub-family of LTTRs (Gibson and Tabita, 1996). In unicellular 89 cyanobacteria, a number of CbbR homologs have been characterized. CmpR is an 90 activator of the cmp genes (Nishimura et al., 2008), whereas CcmR (aka NdhR) acts as 91 a transcriptional repressor of multiple genes encoding other C i transporters (e.g., Figge 92 et al., 2001; Wang et al., 2004). A third type of CbbR-like protein, the activator of the 93 RuBisCo genes, has not yet been identified in cyanobacteria. 94 Filamentous heterocyst-forming cyanobacteria, of which Anabaena sp. PCC 95 7120 (hereafter Anabaena) is a model organism, additionally have the capacity for cell 96 differentiation to turn some O 2 -evolving photosynthetic cells of the filament into 97 heterocysts, which are heterotrophic cells especialized in the fixation of atmospheric 98 N 2 . Thus, Anabaena is a truly pluricellular bacterium with different cell types specialized 99 in different nutritional tasks that exchange nutrients and regulators and contribute to 100 Page 4 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 5 the performance of the filament as the organism unit (see Flores and Herrero, 2010). 101 Nitrogen assimilation and heterocyst differentiation in Anabaena are regulated by the 102 global transcription factor NtcA, which responds to the cellular C-to-N ratio, and the 103 heterocyst-specific transcription factor HetR (see Herrero et al., 2013). The Anabaena 104 genomic sequence includes three genes annotated as CbbR-like LTTRs (Kaneko et 105 al., 2001). Of these, all0862 has been identified as cmpR, and its product activates the 106 expression of the cmp operon (alr2877-alr2880) and the cmpR gene itself in response 107 to C i limitation (López-Igual et al., 2012). Notably, this regulation is effected in 108 combination with NtcA, thus revealing a mode of co-regulation by C and N availability 109 (López-Igual et al., 2012). Furthermore, in Anabaena the rbcLXS operon encoding 110 RuBisCo, which is moderately induced under C i limitation (López-Igual et al., 2012), is 111 repressed in the heterocysts (Madan and Nierzwicki-Bauer, 1993), a regulation likely 112 exerted by NtcA (Ramasubramanian et al., 1994; Picossi et al., 2014). 113 Here we have identified the CbbR-homolog All3953 as the activator of the 114 RuBisCo-encoding operon in Anabaena. We have determined the All3953 regulon by 115 ChIP-Seq, which has revealed that All3953 is a global regulator for C i assimilation 116 genes and genes involved in protection of the photosynthetic apparatus against 117 oxidative damage that are regulated by Ci availability. 118 119 Results 120 All3953, an RbcR-like factor in Anabaena sp. PCC 7120 121 To gain insight into the LTTR All3953 in Anabaena, the expression of the all3953 gene 122 under different growth conditions was analyzed by northern blot. Three bands of 123 hybridization corresponding to transcripts of ca. 1.9, 1.6 and 1.3 kb, respectively, which 124 appeared similarly represented before and after Ci limitation could be observed (Supp. 125 Fig. 1A). On the other hand, all3953 expression did not significantly respond to N 126 depletion (Supp. Fig. 1B). The latter result was consistent with previous global 127 Page 5 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 6 transcriptional studies (Ehira and Ohmori, 2006; Flaherthy et al., 2011) and with the 128 lack of binding associated to all3953 of the N-control transcriptional regulator NtcA 129 upon combined nitrogen depletion (Picossi et al., 2014). 130 131 Isolation and characterization of an all3953 mutant 132 To study the role of All3953, a mutant strain bearing an inactivated version of all3953 133 was constructed (see Experimental procedures). In strain CSS74 most of the all3953 134 gene was deleted and the C.S3 gene cassette (encoding Sm/Sp resistance) was 135 introduced to facilitate segregation and maintenance of the mutation in Anabaena 136 (Supp. Fig. 2A). As a control, strain CSS77 bearing the C.S3 gene cassette in the 137 Anabaena plasmid alpha was also constructed. For cis complementation, a wild-type 138 version of all3953 was transferred to strain CSS74 in an integrative plasmid, 139 generating strain CSS74C (Supp. Fig. 2A). 140 Strain CSS74 exhibited poor growth under standard growth conditions with 141 ammonium as a nitrogen source, and it formed short filaments in liquid medium, 142 whereas strain CSS74C behaved similarly to CSS77 (not shown). To quantify the 143 deleterious effect of the all3953 mutation, growth rates were calculated in liquid 144 medium under different illumination and C i -supply conditions. Growth rate of the control 145 strain CSS77 was highest (0.892 days -1 ) under high light (HL) and high carbon (HC), 146 and was about 30% lower under the other tested conditions (Fig. 1A). In contrast, 147 growth of strain CSS74 was severely affected under HL HC conditions (growth rate ca. 148 70% lower than that of the control) (Fig. 1A), under which it ended up dying after about 149 5 days (Fig. 1B). Under HL and low carbon (LC) or low light (LL) HC, the defect was 150 close to 30% with regard to the control, although after prolonged incubation the mutant 151 was more severely affected under HL LC. The defect was the smallest (ca. 15%) under 152 LL LC conditions (Fig. 1A,B). In solid medium, growth of strain CSS74 was similar, and 153 similar to that of the control, in the presence of ammonium, nitrate or no combined 154 Page 6 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 7 nitrogen under LL LC conditions, whereas under HL LC a severe growth defect was 155 observed with regard to the control with any of the tested nitrogen sources (not shown). 156 In summary, the lack of All3953 had a deleterious effect on growth, especially under 157 HL (and HC) conditions. 158 To further characterize the growth defect of the all3953 mutant, the rate of 159 oxygen evolution using CO 2 as a final electron acceptor was measured under different 160 illumination conditions in the CSS74 mutant in relation to the control strain CSS77. 161 When exponentially-growing cells were incubated for 24 h under LL HC, the oxygen 162 evolution rate was slightly lower in strain CSS74 in comparison to CSS77 (88 and 163 105 µmol O 2 ·[mg Chl] -1 ·h -1 , respectively). Under HL the difference between the two 164 strains was larger (167, for CSS77, and 110, for CSS74, µmol O 2 ·[mg Chl] -1 ·h -1 ). 165 166 Effect of all3953 mutation on rbcLXS expression 167 To test the effect of the all3953 mutation on the expression of the rbcLXS operon, 168 northern blot analysis was performed with RNA isolated from cells of the control and 169 mutant strains grown with HC and transferred to LC conditions. After 1 h incubation 170 with LC, a ca. 2-fold increase in the amount of the rbcLXS transcripts could be 171 observed in the control strain. No induction could be detected in the mutant (Fig. 2). In 172 the complemented strain CSS74C the expression of rbcLXS increased in LC similarly 173 to the control (Fig. 2). These results indicated that the induction of the rbcLXS operon 174 upon C i deficiency was dependent on All3953 and that the defect in strain CSS74 was 175 exclusively due to the lack of All3953. 176 177 ChIP-Seq analysis of the All3953 targets 178 To determine the DNA targets of All3953 at a genomic level, we used chromatin 179 immunoprecipitation followed by high-throughput sequencing analysis. To this end, we 180 constructed a strain (CSS57) expressing from the all3953 promoter a version of 181 Page 7 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 8 All3953 C-terminally fussed to TAP-tag (Rigaut et al., 1999), as well as a control strain 182 (CSS68) expressing the TAP-tag alone under the control of the all3953 promoter 183 (Supp. Fig. 2B, see Experimental procedures). Immunoprecipitation was carried out 184 using cells of strains CSS57 and CSS68 grown with ammonium as the nitrogen source 185 under HC conditions and incubated for 3 h with ammonium under LC conditions. 186 The analysis of the sequences resulted in a total of 142 All3953 binding 187 regions, of which 127 were located in the chromosome, 10 in plasmid alpha, three in 188 plasmid beta and two in plasmid gamma. Each binding region was ascribed to one or 189 two genes according to the location (midpoint) of the region, and the relative location 190 with respect to the assigned gene was also indicated (Table 1 and Supp. Table 1). A 191 total of 175 genes were ascribed to the 142 binding regions. The binding regions were 192 mostly located upstream of the ascribed genes (72%), whereas 21% were intragenic 193 and 7% were located downstream of genes. The results of the ChIP-Seq analysis are 194 available at GEO accession number GSE58861 195 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc= GSE58861). 196 The 175 ascribed genes were classified according to their functional category 197 (Table 2). Remarkably, there were 19 genes encoding proteins related to 198 photosynthesis and respiration, and 21 genes encoding regulatory proteins, including 199 All7179, a SigB homolog. The rest were mostly genes encoding hypothetical or 200 unknown proteins (42%), but also genes encoding proteins involved in translation, in 201 biosynthesis of amino acids and cofactors, prosthetic groups and carriers, in transport, 202 and in other cellular processes. Table 3 highlights All3953-binding regions related to 203 photosynthesis and respiration among which, confirming our results of gene 204 inactivation, the gene encoding the large subunit of the RuBisCo (rbcL; binding region 205 #37) is included. The fact that a high number of genes involved in photosynthesis and 206 C fixation, including rbcL, were identified as targets of RbcR suggests that this protein 207 is a global transcription factor for photosynthetic C assimilation. 208 Page 8 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 15 pCSE120 [S.K3/L.HEH2 (BamHI)/C.S3 (BamHI); nomenclature as in (Elhai and Wolk, 374 1988)], was inserted obtaining plasmid pCSS162. Plasmid pCSS162 was transferred to 375 strain PCC 7120 by conjugation (Elhai et al., 1997). Exconjugants resistant to Sm and 376 Sp, which had the ∆all3953::C.S3 construct integrated by double recombination were 377 selected, obtaining strain CSS74. The segregation of the mutation was tested by PCR 378 (Supp. Fig. 2) with primers all3953-14, all3953-15, all3953-17 and all3953-20. 379 To construct a control strain expressing Sm r and Sp r plasmid pCSS163, a 380 derivative of plasmid pCSEL24 (Olmedo-Verd et al., 2006) containing the C.S3 gene 381 cassette, was transferred to Anabaena by conjugation. Exconjugants that had the 382 pCSS163 integrated in the alpha plasmid of Anabaena were selected, obtaining the 383 strain CSS77 (Supp. Fig. 2). 384 To complement the all3953 mutation of strain CSS74, a DNA fragment 385 encompassing the whole all3953 gene and sequences upstream from it was amplified 386 by PCR using the primer pair all3953-24/all3953-25, both including EcoRI sites, and 387 strain PCC 7120 DNA as the template. This fragment was cloned in the EcoRI site of 388 the mobilizable Nm r encoding vector pRL424 (Elhai and Wolk, 1988) producing plasmid 389 pCSS164, which was transferred to strain CSS74 by conjugation followed by selection 390 for Nm r . The genomic structure of the exconjugants in the all3953 region (Supp. Fig. 2) 391 was corroborated by PCR. 392 To construct a strain expressing All3953-C-TAP, the all3953 gene (including the 393 upstream region) was PCR-amplified with primers all3953-11 and all3953-12 and DNA 394 of PCC 7120 as the template. The TAP-tag was PCR-amplified with primers TAPtag-1 395 and TAPtag-2 using DNA of plasmid pBS1479 as the template (Puig et al., 2001). The 396 two PCR products were digested with SalI and ligated, after which the ligation product 397 was used as the template for an overlapping PCR using primers all3953-11 and 398 TAPtag-2. The PCR product was digested with PstI and ligated to the mobilizable 399 vector pCSV3 (Valladares et al., 2011) digested with PstI, rendering plasmid pCSS107. 400 Page 15 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 16 To construct a control strain with the TAP-tag under the control of the all3953 401 promoter, a 0.4-kb region upstream of all3953 was PCR-amplified using primers 402 all3953-11 and all3953-18 and DNA of pCSS107 as the template. The PCR product 403 was digested with SalI and ligated to the PCR-amplified TAP-tag digested with SalI, 404 after which the ligation product was used as the template for an overlapping PCR using 405 primers all3953-11 and TAPtag-2. The PCR product was digested with PstI and ligated 406 to PstI-digested pCSV3 to give plasmid pCSS157. Plasmids pCSS107 and pCSS157 407 were transferred by conjugation to strain PCC 7120 and single Sm r Sp r recombinants 408 were selected, obtaining strain CSS57 and CSS68, respectively. Western blots using 409 Peroxidase-Anti-Peroxidase Soluble Complex (PAP, Sigma-Aldrich) were performed to 410 ensure that the two strains expressed the TAP-tag (Supp. Fig. 2B). 411 412 Chromatin immunoprecipitation 413 Cells of strains CSS57 growing exponentially (3-5 µg Chl·ml -1 ) in the light (80 414 µE·m -2 ·s -1 ) in medium supplemented with 2 µg·ml -1 Sm and Sp, in HC conditions were 415 incubated with LC for 3 h. Formaldehyde was then added to the cultures to a final 416 concentration of 1%, and the cultures were incubated for 15 min. Glycine was added at 417 125 mM final concentration and the incubation was continued for 5 min to stop the 418 fixing reaction. The cells were then filtered, washed with cold TBS (20 mM Tris-HCl, pH 419 7.4, 140 mM NaCl) and collected in tubes (25 ml of culture per tube). The pellets were 420 frozen in liquid nitrogen and stored at -20°C until used. Pellets corresponding to about 421 25 ml of culture were resuspended in 500 µl of lysis buffer (50 mM HEPES-KOH, pH 422 7.5, 140 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% sodium deoxycholate, 423 supplemented with Mini EDTA-free protease inhibitor cocktail [Roche]) and, after 424 addition of 150 µl of glass beads (acid-washed, 425-600 µm [Sigma]), the cells were 425 broken in a multivortexer at 2,000 rpm for 1 h at 4°C. The cell lysates were collected by 426 centrifugation and the extracts were subjected to sonication to shear the DNA to about 427 300-bp fragments (60 cycles of 10 s, 20 s ice, 15% amplitude, in a Branson Digital 428 Page 16 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 17 Sonifier). After centrifugation to eliminate cell debris, the whole-cell extracts were 429 stored at −20°C or immediately used for immunoprecipitation. 430 Immunoprecipitation of DNA was carried out as described (Picossi et al., 2014), 431 with some modifications. Whole-cell extracts were prepared at 4 mg·ml -1 of total protein 432 with lysis buffer (in 500 µl total volume). A 50-µl sample was taken as the input sample, 433 and the extracts were incubated with 15 µl IgG-conjugated Dynabeads (about 6 µg 434 IgG) at 4ºC with rotation for 12-14h. The washes of the Dynabeads, as well as the 435 elution of the immunoprecipitated material, the crosslinking reversion and the isolation 436 of the DNA were performed as in (Picossi et al., 2014). 437 438 Massive sequencing of the immunoprecipitated DNA 439 Input and ChIP DNA samples were sent for sequencing to the Functional Genomics 440 Core Facility of the Institute for Research in Biomedicine, Barcelona (Spain). Next 441 generation sequencing was carried out using Illumina’s sequencing technology. ChIP 442 DNA Sample Prep Kit (Illumina) was used for library preparation. Libraries were loaded 443 at 8 pM concentration into the flow cell using the Cluster Station running recipe V7 with 444 the Single-Read Cluster Generation Kit v4 (all Illumina). The flow cell was loaded into 445 the Genome Analyzer II and samples were sequenced for 120 nucleotides from a 446 single end using the Sequencing Kit v5 and recipe v8 (all Illumina). Manufacturer’s 447 recommendations were strictly followed. Illumina sequencing data were pre-processed 448 with the standard Illumina pipeline version 1.5 and sequences were aligned to the PCC 449 7120 genome (http://genome.microbedb.jp/cyanobase/Anabaena) with the Bowtie 450 software 0.12.5 (Langmead et al., 2009). The analysis of the results were carried out 451 using the Triform algorithm (Kornacker et al., 2012) as in (Picossi et al., 2014). The 452 sequences in the ChIP samples of strain CSS68 (TAP control) were used as the 453 background of the sequences found for strain CSS57 (All3953-TAP), and thus to 454 determine the specific binding regions of All3953-TAP in the genome of Anabaena. 455 Page 17 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 18 The binding regions were visualized and analyzed using the UCSC Microbial Genome 456 Browser (Scheneider et al., 2006). They were ascribed to one or two genes, in case it 457 was not possible to ascribe them to only one, and classified as upstream of the gene, if 458 the midpoint of the binding region was located upstream of the start of the gene, 459 internal, if the midpoint of the binding region was inside the gene, or downstream, if the 460 midpoint of the binding region was located downstream of the end of the gene to which 461 it had been ascribed. 462 463 Northern and qRT-PCR analyses 464 Isolation of total RNA from Anabaena was done as described previously (Mohamed 465 and Jansson, 1989). Northern analysis was performed as described previously (López-466 Igual et al., 2012). 467 For qRT-PCR, 750 ng of DNA-free RNA samples were used for all the PCR 468 primer pairs. For the RT reaction, the Quantitech Reverse transcription kit (Qiagen), 469 with the Random Hexamer Primer mix (100 ng per sample) (Bioline) was used. The 470 cDNA produced was diluted 7.5 times to use 2 µl of cDNA per PCR reaction. PCR was 471 done using the Quantimix Easy SYG Kit (Biotools) (SYBR green I) in a iCycler iQ Multi-472 Color Real Time PCR Detection System (Bio-Rad). The abundance of a transcript in 473 the RNA sample was calculated as: abundance= 2^[Ct(sample)-Ct(control)], where the 474 RNA sample of the control strain CSS77 in HC condition (0) was used as the control. 475 476 Oxygen evolution 477 2-ml samples of exponentially grown cultures in HC LL or HC HL conditions were used 478 to measure O 2 evolution with an O 2 electrode calibrated with culture medium and 479 Na 2 S 2 O 4 as the reducing agent. O 2 production was measured in the light (400 480 µE·m -2 ·s -1 ) after a seven-minute incubation in the dark. 481 482 Page 18 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 19 Acknowledgments 483 This work was supported by grants BFU2010-17980 and BFU2013-44686-P from the 484 Spanish government, co-financed by FEDER. The authors are grateful to K. Kornacker 485 for carrying out the Triform and CisFinder analyses, and to H. Auer and I. Pons, from 486 the Functional Genomics Core Facility of the IRB, Barcelona (Spain); to F. Monje-487 Casas and Yagut Allahverdiyeva-Rinne for critical reading of the manuscript and to M. 488 J. Huertas and A. Torrado for help with O 2 production measurements. 489 490 REFERENCES 491 492 Black, T.A., Cai, Y., and Wolk C.P. (1993) Spatial expression and autoregulation of 493 hetR, a gene involved in the control of heterocyst development in Anabaena. Mol 494 Microbiol 9: 77-84. 495 Cameron, C.J., Sutter, M., and Kerfeld, C.A. (2014) The carboxysome: Function, 496 structure and cellular dynamics. The Cell Biology of Cyanobacteria, eds. Flores E, 497 Herrero A (Caister Academic Press, Norfolk), pp. 171-188. 498 Ehira, S., and Ohmori, M, (2006) NrrA, a nitrogen-responsive response regulator 499 facilitates heterocyst development in the cyanobacterium Anabaena sp. strain PCC 500 7120. Mol Microbiol 59: 1692-1703. 501 Elhai., J., Vepritskiy, A., Muro-Pastor, A.M., Flores, E, and Wolk, C.P. (1997) Reduction 502 of conjugal transfer efficiency by three restriction activities of Anabaena sp. strain 503 PCC 7120. J Bacteriol 179: 1998-2005. 504 Eisenhut, M., Ruth, W., Haimovich, M., Bauwe, H., Kaplan, A., and Hagemann, M. 505 (2008) The photorespiratory glycolate metabolism is essential for cyanobacteria and 506 might have been conveyed endosymbiotically to plants. Proc Natl Acad Sci USA 507 105: 17199-17204. 508 Elhai, J., and Wolk, C.P. (1988) A versatile class of positive-selection vectors based on 509 the nonviability of palindrome-containing plasmids that allows cloning into long 510 polylinkers. Gene 68: 119-138. 511 Ermakova, M., Battchikova, N., Allahverdiyeva, Y., and Aro, E.M. (2013) Novel 512 heterocyst-specific flavodiiron proteins in Anabaena sp. PCC 7120. FEBS Lett 587: 513 82-7. 514 Figge, R.M., Cassier-Chauvat, C., Chauvat, F., and Cerff, R. (2001) Characterization 515 and analysis of an NAD(P)H dehydrogenase transcriptional regulator critical for the 516 survival of cyanobacteria facing inorganic carbon starvation and osmotic stress. Mol 517 Microbiol 39: 455-68. 518 Flaherty, B.L., Van Nieuwerburgh, F., Head, S.R., and Golden, J.W. (2011) Directional 519 RNA deep sequencing sheds new light on the transcriptional response of Anabaena 520 sp. strain PCC 7120 to combined-nitrogen deprivation. BMC Genomics 12: 332. 521 Flores, E., and Herrero, A. (2010) Compartmentalized function through cell 522 differentiation in filamentous cyanobacteria. Nat Rev Microbiol 8: 39-50. 523 Gibson, J.L., and Tabita, F.R. (1996) The molecular regulation of the pentose 524 phosphate pathway in proteobacteria and cyanobacteria. Arch Microbiol 166: 141-525 150. 526 Herrero, A., Picossi, S., and Flores, E. (2013) Gene Expression during heterocyst 527 differentiation. Adv Bot Res 65: 281-329. 528 Page 19 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 20 Kaneko, T., Nakamura, Y., Wolk, C.P., Kuritz, T., Sasamoto, S., and Watanabe, A. 529 (2001) Complete genomic sequence of the filamentous nitrogen-fixing 530 cyanobacterium Anabaena sp. strain PCC 7120. DNA Res 8: 205-213. 531 Knoll, A.H. (2008) Cyanobacteria and Earth history. The Cyanobacteria: Molecular 532 Biology, Genomics and Evolution, eds. Herrero A, Flores E (Caister Academic 533 Press, Norfolk), pp. 1-19. 534 Kornacker, K., Rye, M.B., Handstad, T., and Drablos, F. (2012) The Triform algorithm: 535 improved sensitivity and specificity in ChIP-Seq peak finding. BMC Bioinformatics 536 13: 176. 537 Langmead, B., Trapnell, C., Pop, M., and Salzberg, S.L. (2009) Ultrafast and memory-538 efficient alignment of short DNA sequences to the human genome. Genome Biol 10: 539 R25. 540 Lin, M.T., Occhialini, A., Andralojc, P.J., Parry, M.A.J., and Hanson, M.R. (2014) A 541 faster Rubisco with potential to increase photosynthesis in crops. Nature 513: 547-542 550. 543 López-Igual, R., Picossi, S., López-Garrido, J., Flores, E., and Herrero, A. (2012) N 544 and C control of ABC-type bicarbonate transporter Cmp and its LysR-type 545 transcriptional regulator CmpR in a heterocyst-forming cyanobacterium, Anabaena 546 sp. Environ Microbiol 14: 1035-1048. 547 Lyons, T.W., Reinhard, C.T., and Planavsky, N.J. (2014) The rise of oxygen in Earth's 548 early ocean and atmosphere. Nature 506: 307-315. 549 Madan, A.P., and Nierzwicki-Bauer, S.A. (1993) In situ detection of transcripts for 550 ribulose-1,5-bisphosphate carboxylase in cyanobacterial heterocysts. J Bacteriol 551 175: 7301-7306. 552 Maddocks, S.E., and Oyston, P.C. (2008) Structure and function of the LysR-type 553 transcriptional regulator (LTTR) family proteins. Microbiology 154: 3609-3623. 554 Mohamed, A., and Jansson, C. (1989) Influence of light on accumulation of 555 photosynthesis-specific transcripts in the cyanobacterium Synechocystis 6803. Plant 556 Mol Biol 13: 693-700. 557 Nierzwicki-Bauer, S.A., Curtis, S.E., and Haselkorn, R. (1984) Cotranscription of genes 558 encoding the small and large subunits of ribulose-1,5-bisphosphate carboxylase in 559 the cyanobacterium Anabaena 7120. Proc Natl Acad Sci USA 81: 5961-5965. 560 Nishimura, T., Takahashi, Y., Yamaguchi, O., Suzuki, H., Maeda, S., and Omata, T. 561 (2008) Mechanism of low CO 2 -induced activation of the cmp bicarbonate transporter 562 operon by a LysR family protein in the cyanobacterium Synechococcus elongatus 563 strain PCC 7942. Mol Microbiol 68: 98-109. 564 Olmedo-Verd, E., Muro-Pastor, A.M., Flores, E., and Herrero, A. (2006) Localized 565 induction of the ntcA regulatory gene in developing heterocysts of Anabaena sp. 566 strain PCC 7120. J Bacteriol 188: 6694-6649. 567 Picossi, S., Flores, E. and Herrero, A. (2014) ChIP analysis unravels an exceptionally 568 wide distribution of DNA binding sites for the NtcA transcription factor in a 569 heterocyst-forming cyanobacterium. BMC Genomics 15: 22. 570 Pollari, M., Ruotsalainen, V., Rantamaki, S., Tyystjarvi, E., and Tyystjarvi, T. (2009). 571 Simultaneous inactivation of sigma factors B and D interferes with light acclimation 572 of the cyanobacterium Synechocystis sp. strain PCC 6803. J Bacteriol 191: 3992-573 4001. 574 Puig, O., Caspary, F., Rigaut, G., Rutz, B., Bouveret, E., Bragado-Nilsson, E., Wilm, 575 M., and Seraphin, B. (2001) The tandem affinity purification (TAP) method: a 576 general procedure of protein complex purification. Methods 24: 218-229. 577 Price, G.D., Badger, M.R., Woodger, F.J., and Long, B.M. (2008). Advances in 578 understanding the cyanobacterial CO 2 -concentrating-mechanism (CCM): functional 579 components, Ci transporters, diversity, genetic regulation and prospects for 580 engineering into plants. J Exp Bo. 59: 1441-1461. 581 Price, G.D. (2011) Inorganic carbon transporters of the cyanobacterial CO 2 582 concentrating mechanism. Photosynth Res 109: 47-57. 583 Page 20 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 21 Ramasubramanian, T.S., Wei, T.-F. and Golden, J.W. (1994) Two Anabaena sp. strain 584 PCC 7120 DNA-binding factors interact with vegetative celland heterocyst-specific 585 genes. J Bacteriol 176: 1214-1223. 586 Rigaut, G., Shevchenko, A., Rutz, B., Wilm, M., Mann, M., and Seraphin, B. (1999) A 587 generic protein purification method for protein complex characterization and 588 proteome exploration. Na. Biotechnol 17: 1030-1032. 589 Rippka, R., Deruelles, J., Waterbury, J.B., Herdman, M., and Stanier, R.Y. (1979) 590 Generic assignments, strain stories and properties of pure cultures of 591 cyanobacteria. J Gen Microbiol 111: 1-61. 592 Schaefer, M.R., and Golden, S.S. (1989) Differential expression of members of a 593 cyanobacterial psbA gene family in response to light. J Bacteriol 171: 3973-3781. 594 Scheneider, K.L., Pollard, K.S., Baertsch, R., Pohl, A., and Lowe, T.M. (2006) The 595 UCSC archaeal genome browser. Nucleic Ac Res 34: D407–D410. 596 Valladares, A., Rodríguez, V., Camargo, S., Martínez-Nöel, G.M., Herrero, A., and 597 Luque, I. (2011) Specific role of the cyanobacterial PipX factor in the heterocysts of 598 Anabaena sp. strain PCC 7120. J Bacteriol 193(5): 1172-1182. 599 Wang, H.L., Postier, B.L., and Burnap, R.L. (2004) Alterations in global patterns of 600 gene expression in Synechocystis sp. PCC 6803 in response to inorganic carbon 601 limitation and the inactivation of ndhR, a LysR family regulator. J Biol Chem 279: 602 5739-5751. 603 604 605 606 Page 21 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 22 Figure legends: 607 Fig. 1. Growth of the all3953 mutant. A, The growth rate constant (µ = ln2/t d , where t d is 608 the doubling time) was calculated from the increase of protein content determined in 609 0.2 ml samples of cultures. The table shows the mean and standard deviation from 3 610 independent cultures of each strain and condition. ∆all3953 is strain CSS74; CSS77 is 611 a control strain that carries the Sm/Sp-resistant determinant in a wild-type background. 612 B, Samples of cultures were photographed after 5 days of incubation under the 613 indicated conditions. HL, high light; LL, low light; HC, high carbon; LC, low carbon. 614 615 Fig. 2. Expression of rbcLXS in the ∆all3953 mutant and complemented strain. 616 Northern analysis carried out with RNA from strains CSS77 (control) CSS74 (∆all3953) 617 and CSS74C (CSS74 complemented) was isolated from cells grown with HC (0) and 618 incubated for 1h (1) with LC. The membranes were hybridized with an internal 619 fragment of the rbcL gene (upper panels) and, as a loading and transfer control, of the 620 rnpB gene (lower panels). Arrowheads point to the main transcripts detected with the 621 rbcL gene probe (approximate sizes are indicated). 622 623 Fig. 3. Consensus All3953 binding sequence and rbcL promoter. A, The primary 624 consensus motif based on 142 high confidence CSS57 ChIP-Seq peak sequences is 625 shown with indication of the probability of occurrence of each base along the 22-nt 626 sequence. W is A or T; Y is C or T; B is C, G or T. B, Structure of the rbcLXS promoter 627 region. The transcription initiation point of the operon (+1) and the -10 and -35 boxes 628 (from Nierzwicki-Bauer et al., 1984) are indicated in red. The NtcA-binding site 629 (GTAN 8 TAC) is indicated in green, and the three putative binding sites for All3953 (Box 630 I, Box II and Box III) are indicated in blue. 631 632 Page 22 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 23 Fig. 4. qRT-PCR analysis of the expression of selected photosynthesis and respiration-633 related All3953 gene targets. Transcriptional response of the indicated genes to C i 634 limitation in the control (CSS77) and ∆all3953 mutant (CSS74) strains was 635 investigated. RNA was isolated from cells grown with 10 mM NaHCO 3 -supplemented 636 medium bubbled with 1%CO 2 in air (0) incubated for 1 h (1) or 3 h (3) in NaHCO 3 -free 637 medium bubbled with air. Bars represent the mean transcript levels (± standard 638 deviation) in three independent experiments. 639 640 Page 23 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 1 The LysR-type transcription factor PacR is a global regulator of photosynthetic 1 carbon assimilation in Anabaena 2 3 Silvia Picossi, Enrique Flores and Antonia Herrero* 4 5 Instituto de Bioquímica Vegetal y Fotosíntesis, Consejo Superior de Investigaciones 6 Científicas and Universidad de Sevilla, Américo Vespucio 49, E-41092, Seville, Spain. 7 8 9 *Corresponding author. Tel.: +34 954489522. Fax: +34 954460165. E-mail address: 10 herr[email protected] 11 12 Keywords: ChIP; Cyanobacteria; Oxygenic phototrophy; Photoprotection; RuBisCo 13 14 Running title: Photosynthetic carbon assimilation regulator 15 16 Accession link to data: 17 http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?token=afcxwkacxpydfax&acc=GSE58861 18 19 20 21 22 Page 24 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 8 upon C i deficiency was dependent on All3953 and that the defect in strain CSS74 was 182 exclusively due to the lack of All3953. 183 184 ChIP-Seq analysis of the All3953 targets 185 To determine the DNA targets of All3953 at a genomic level, we used chromatin 186 immunoprecipitation followed by high-throughput sequencing analysis. To this end, we 187 constructed a strain (CSS57) expressing from the all3953 promoter a version of 188 All3953 C-terminally fussed to TAP-tag (Rigaut et al., 1999), as well as a control strain 189 (CSS68) expressing the TAP-tag alone under the control of the all3953 promoter 190 (Supp. Fig. 2B, see Experimental procedures). Immunoprecipitation was carried out 191 using cells of strains CSS57 and CSS68 grown with ammonium as the nitrogen source 192 under HC conditions and incubated for 3 h with ammonium under LC conditions. 193 The analysis of the sequences resulted in a total of 142 All3953 binding 194 regions, of which 127 were located in the chromosome, 10 in plasmid alpha, three in 195 plasmid beta and two in plasmid gamma. Each binding region was ascribed to one or 196 two genes according to the location (midpoint) of the region, and the relative location 197 with respect to the assigned gene was also indicated (Table 1 and Supp. Table 1). A 198 total of 175 genes were ascribed to the 142 binding regions. The binding regions were 199 mostly located upstream of the ascribed genes (72%), whereas 21% were intragenic 200 and 7% were located downstream of genes. The results of the ChIP-Seq analysis are 201 available at GEO accession number GSE58861 202 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc= GSE58861). 203 The 175 ascribed genes were classified according to their functional category 204 (Table 2). Remarkably, there were 19 genes encoding proteins related to 205 photosynthesis and respiration, and 21 genes encoding regulatory proteins, including 206 All7179, a SigB homolog. The rest were mostly genes encoding hypothetical or 207 unknown proteins (42%), but also genes encoding proteins involved in translation, in 208 Page 31 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 9 biosynthesis of amino acids and cofactors, prosthetic groups and carriers, in transport, 209 and in other cellular processes. Table 3 highlights All3953-binding regions related to 210 photosynthesis and respiration among which, confirming our results of gene 211 inactivation, the gene encoding the large subunit of the RuBisCo (rbcL; binding region 212 #37) is included. The fact that a high number of genes involved in photosynthesis and 213 C fixation, including rbcL, were identified as targets of RbcR suggests that this protein 214 is a global transcription factor for photosynthetic C assimilation. 215 A CisFinder analysis of the primary consensus motif was carried out based on 216 142 high-confidence ChIP-Seq peak sequences (Fig. 3A). The consensus motif found 217 has a dyad-symmetry architecture and matches the consensus of LysR-recognition 218 binding sites (RBS) (T-N 11 -A) (Maddocks and Oyston, 2008), as well as the consensus 219 binding sites proposed for CbbR factors (TNA-N 7/8 -TNA). The primary motifs identified 220 in the central 100 nt of the binding regions are indicated in Supp. Table 1 (for some 221 binding regions more than one motif have been identified). 222 223 Expression analysis of some All3953 targets in Anabaena 224 To corroborate our ChIP-Seq analysis and to support the notion that All3953 is indeed 225 a global regulator for C fixation genes, the expression levels of some of the 226 photosynthesisand C-fixation-related target genes, its response to C i limitation and its 227 dependence on All3953 was further analyzed by qRT-PCR (Fig. 4). Strains CSS77 and 228 CSS74 were grown in ammonium and HC under standard light conditions (80 229 µE·m -2 ·s -1 ) at 30ºC up to the exponential phase. They were then transferred to medium 230 with ammonium under LC conditions. As previously shown by northern analysis, 231 transcription levels of the all3953 gene did not significantly change after C i deprivation 232 in the control strain CSS77. As expected, all3953 transcript levels were not detectable 233 in CSS74, corroborating that the all3953 mutation was segregated in this strain. The 234 rbcL gene (alr1524) was 4.6-fold induced at 3 h after transfer to LC in strain CSS77, 235 Page 32 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 10 whereas no induction was observed in CSS74, thus corroborating the dependence on 236 All3953. Interestingly, under HC the expression of rbcL in the mutant was higher than 237 in the control strain, suggesting that besides as an activator under LC, All3953 could 238 be acting as a repressor of rbcL under HC conditions. 239 ORF all4446 (flv4) was highly induced (about 100-fold), and its level was 240 maximum 1 h after the shift to LC (Fig. 4). In the all3953 mutant, the basal transcription 241 of all4446 in HC was about 8-fold lower than in the control strain, and it was only 242 slightly induced (2-fold) upon the shift to LC. all3891 (flv1A) was induced about 6-fold 3 243 h after the shift to LC in the control strain, whereas only a 2-fold induction was 244 observed in the all3953 mutant. all1304 (bicarbonate transporter homolog) and alr4156 245 (NdhF homolog) were both highly induced (up to 30and 20-fold, respectively) upon 246 transfer of strain CSS77 to LC. In contrast, no induction of alr4156 and only a small 247 induction of all1304 took place in strain CSS74. Expression of alr4592 (psbAIII) 248 increased about 5-fold upon C i deprivation in CSS77, but did not appreciably change in 249 CSS74. The alr0223 (NdhA homolog) gene was about 2-fold induced under C i 250 deprivation in CSS77 but not in CSS74. Finally, the expression of alr1004 (alanine-251 glyoxylate aminotransferase) was repressed by 5-fold under LC conditions in CSSL77 252 but not in CSS74. These results confirm that expression of the above studied genes is 253 regulated, either positively or negatively, by All3953. 254 255 Discussion 256 We have identified the LTTR All3953 as the activator of the RuBisCo-encoding genes 257 in the cyanobacterium Anabaena sp. PCC 7120. All3953 appears to activate the rbcL 258 operon under C i limitation and to repress it when C i is abundant. An LTTR factor 259 regulating the expression of the rbcL operon has not, to our knowledge, been 260 described in any cyanobacterium. The expression of all3953 does not respond to C i 261 limitation (Supp. Fig. 1) and, indeed, no binding region of All3953 was found adscribed 262 Page 33 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 11 to all3953. Thus, the all3953 gene seems to belong to the non-autoregulated LTTRs. 263 All3953 shares 28 and 26% identical residues with NdhR from Synechocystis sp. PCC 264 6803 and Synechococcus sp. PCC 7002 (CcmR), respectively. 265 By ChIP-Seq analysis of a strain bearing a TAP-tagged version of All3953, we 266 have determined 142 All3953-bound regions at 3 h after transfer from high to low C i 267 conditions, which have been assigned to 177 genes. Apart from genes encoding 268 unknown or hypothetical proteins, the larger category is of genes encoding regulatory 269 proteins, including the transcriptional regulators Alr0353 (a LTTR) and All4500 (CRP-270 like), the two-component response regulator All3348, the two-component sensor 271 histidine-kinase All1145 and the group 2-sigma 70-type sigma factor All7179 (Supp. 272 Table 1). Interestingly, when comparing to Synechocystis sigma factors, All7179 273 (SigB4) is more similar toa homolog of Synechocystis SigB, which, along with SigD has 274 been shown to be important for PSII recovery in this unicellular cyanobacterium (Pollari 275 et al., 2009). All3953 also binds upstream of genes patS and hetN, whose products 276 regulate heterocyst differentiation (Supp. Table 1). The fact that All3953 binds to the 277 promoter region of genes encoding other regulatory proteins suggests a wide role of 278 this protein in the physiology of the organism. 279 In the promoter region of the rbcL operon we have found three putative binding 280 sites for All3953, Box I, Box II, and Box III (Fig. 3B) that resemble the consensus 281 recognition sequence found by Cisfinder analysis (Fig. 3A). It is conceivable that, like in 282 some other LTTRs these boxes combine repression and activation sites. In this regard, 283 binding of All3953 to Box III, overlapping gene promoter elements, could be related to 284 the repression of rbcL observed under high C i (Fig. 4). On the other hand, as 285 mentioned above, the rbcL operon is repressed in heterocysts by the global 286 transcriptional regulator NtcA, for which a binding site is found overlapping the operon 287 TSP (Ramasubramanian et al., 1994; Picossi et al., 2014) (Figure 3B). It is conceivable 288 that NtcA binding in these differentiated cells interfere with All3952-mediated activation. 289 Page 34 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 12 Besides being the activator of the RuBisCo genes, All3953 targets include 290 genes involved in other processes related to carbon assimilation, such as C i transport 291 (all1304, encoding an homolog of the BicA bicarbonate transporter; alr4156 encoding a 292 homolog of the NdhF3 subunit of the NDH-1 3 CO 2 uptake system; and alr0869 293 encoding a homolog of the NdhF4 subunit of the NDH-1 4 CO 2 uptake system); 294 components of the carboxysome shell (all0868, putative ccmK) and 2-295 phosphoglycolate metabolism (alr1004 and alr2873, possibly related to 296 photorespiration [Eisenhut et al., 2008]). Notably, All3953 targets include also a 297 number of genes encoding photosystem components, such as alr5154 (psaA, 298 encoding the PSI core protein PsaA), alr3727 (psbAII, encoding a component of form II 299 of the PSII core protein PsbA [D1]), alr4592 (psbAIII, encoding another component of 300 form II of PsbA) and alr1216 (PSII 12 kD extrinsic protein PsbU), and genes related to 301 PS activity. In the latter group are alr4149 (biliverdin reductase, putatively involved in 302 phycobilisome -PSII antennasynthesis), and genes that can participate in 303 photosynthetic electron transfer, such as alr0223 and alr0348 (ndhA and ndhD, 304 subunits of other putative NADH dehydrogenases), alr1576 (dehydrogenase subunit), 305 all0737 (thioredoxin reductase), all1365 (CytM cytochrome), all4148 (ferredoxin I), 306 all3891 and all4446 (flavodiiron proteins Flv1A and Flv4, respectively). (Besides in CO 2 307 uptake, NdhF3 and NdhF4 can also participate in electron transfer.) 308 Reports on gene expression regulation by C i are scarce for Anabaena. 309 However, in the unicellular cyanobacterium Synechocystis sp. PCC 6803 310 transcriptomic analysis has already shown down-regulation of some genes encoding 311 polypeptides of PSI and PSII complexes as well as of phycobilisome components, 312 upon transfer to C i limitation conditions, likely as an adaptation to lower assimilatory 313 power demand, and up-regulation of some PSII core polypeptides, interpreted as 314 adaptation to conditions of shortage of electron acceptors that could lead to 315 photodamage and increased turnover of PS core components (Wang et al., 2004). Our 316 analysis extends the array of photosynthetic genes responding to C i limitation, 317 Page 35 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 13 remarkably to include the core PSI reaction center psaA gene. Moreover, except for 318 alr1004 that responds negatively, all the Anabaena photosynthetic genes mentioned 319 above increase expression upon the shift to C i limitation. 320 Noteworthy, for the majority of photosynthetic gene targets of All3953 in 321 Anabaena, a function in protection against reactive oxygen species, which can be 322 generated by C i limitation or exposure to HL, has either been described or could be 323 predicted. Thus, the psbAII and psbAIII genes are induced under HL in the unicellular 324 Synechococcus sp. PCC 7942, and cells cultured under HL showed more form II, and 325 less form I (encoded by psbAI), of D1 compared to cells under LL (Schaefer and 326 Golden, 1989). Regarding flavodiiron proteins, genes, flv1A, flv3A, and specially flv2 327 and flv4 of Anabaena have been shown up-regulated in vegetative cells in low Ci, and 328 flv1A and flv3A also in high light, whereas flv1B and flv3B are expressed exclusively in 329 heterocysts (Ermakova et al., 2013). Whereas Flv1A and Flv3A appear involved in 330 photoreduction of oxygen to water by removing excess electrons from PSI through 331 NAD(P)H dehydrogenases, Flv2 and Flv4 could have a role in photoprotection of PSII 332 under low Ci (Ermakova et al., 2013). Regarding photorespiration, it has also been 333 considered to have a role in removal of excess O 2 (Eisenhut et al., 2008). To the best 334 of our knowledge, the regulator responsible for the response to C i availability of any 335 photosynthetic gene has not been identified in cyanobacteria (oxygenic phototrophs). 336 Our ChIP-Seq and expression analysis indicate that All3953 is a regulator of 337 photosynthetic genes in Anabaena. 338 The growth rate of Anabaena is highest under HL HC conditions (Fig. 1), 339 implying that this cyanobacterium has mechanisms to get profit of HC while 340 counteracting HL stress. The all3953 mutant strain CSS74 exhibited a growth defect in 341 all the conditions tested, but especially under HL, where it ends-up dying (Fig. 1). This 342 shares the idea that in Anabaena All3953 is required to cope with HL stress. The effect 343 of the lack of All3953 seems more detrimental in relation to impaired photoprotection 344 than to impaired C i scavenging (preference of the mutant for LL LC over HL HC 345 Page 36 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 14 conditions). Indeed, even in LL, the mutant seems to perform slightly better with LC 346 than with HC (Fig. 1). Although LC could suppose a limitation of electron acceptors, an 347 increased rate of photorespiration under LC together with the fact that some of the 348 All3953 targets that could cope with excess oxygen are residually induced upon the 349 transfer from HC to LC in the CSS74 mutant (Fig. 4) could contribute to the preference 350 of this strain for LC over HC, especially under HL. 351 Our results show that in Anabaena the responses to C i availability include 352 regulation of genes encoding elements of CCM and RuBisCo, but also of 353 photosynthetic genes to adjust generation of assimilatory power while preserving the 354 photosynthetic apparatus from oxidative damage, which is specially relevant in 355 oxygenic phototrophs. Because All3953 is a transcriptional regulator globally 356 coordinating these responses, we have named it PacR (Photosynthetic assimilation of 357 carbon Regulator). 358 359 Experimental procedures 360 Strains 361 Anabaena sp. strain PCC 7120 was grown photoautrophically at 30°C with illumination 362 (80 µE·m -2 ·s -1 ) in liquid BG11 0 medium (Rippka et al., 1979) supplemented with 3 mM 363 NH 4 Cl, 6 mM TES buffer and 10 mM NaHCO 3 and bubbled with a mixture of CO 2 and 364 air (1% v/v) (HC). Other conditions used were no NaHCO 3 supplement and bubbling 365 with air (LC); 12 µE·m -2 ·s -1 (LL); 175 µE·m -2 ·s -1 (HL). For growth on plates illumination 366 was 9 µE·m -2 ·s -1 (LL) or 34 µE·m -2 ·s -1 (HL). For the mutants generated in this work, 367 antibiotics were used at the following concentrations: Sm, 2 µg ml −1 ; Sp, 2 µg ml −1 ; and 368 Nm, 25 µg ml −1 for bubbled cultures; and Sm, 5 µg ml −1 ; Sp, 5 µg ml −1 ; and Nm, 369 40 µg ml −1 for cultures in solid medium. 370 371 Strain construction 372 Page 37 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 15 To construct a mutant of the all3953 gene, the 5' and 3' end of the gene, along with the 373 flanking regions, were PCR amplified using chromosomal DNA of PCC 7120 as the 374 template and primers all3953-14 (BglII) and all3953-15 (SalI), and primers all3953-16 375 (SalI) and all3953-17 (PstI), respectively (all primers are specified in Supp. Table 2). 376 The PCR products were digested with SalI and ligated. The resulting mixture was used 377 as a template for overlapping PCR with primers all3953-14 and all3953-17. The new 378 PCR product was digested with BglII and PstI and ligated to the BglII-PstI-digested 379 pRL271 (Black et al., 1993), obtaining plasmid pCSS161. Plasmid pCSS161 was 380 digested with SpeI and the 2-kb Sm r Sp r gene cassette C.S3, excised with XbaI from 381 pCSE120 [S.K3/L.HEH2 (BamHI)/C.S3 (BamHI); nomenclature as in (Elhai and Wolk, 382 1988)], was inserted obtaining plasmid pCSS162. Plasmid pCSS162 was transferred to 383 strain PCC 7120 by conjugation (Elhai et al., 1997). Exconjugants resistant to Sm and 384 Sp, which had the ∆all3953::C.S3 construct integrated by double recombination were 385 selected, obtaining strain CSS74. The segregation of the mutation was tested by PCR 386 (Supp. Fig. 2) with primers all3953-14, all3953-15, all3953-17 and all3953-20. 387 To construct a control strain expressing Sm r and Sp r plasmid pCSS163, a 388 derivative of plasmid pCSEL24 (Olmedo-Verd et al., 2006) containing the C.S3 gene 389 cassette, was transferred to Anabaena by conjugation. Exconjugants that had the 390 pCSS163 integrated in the alpha plasmid of Anabaena were selected, obtaining the 391 strain CSS77 (Supp. Fig. 2). 392 To complement the all3953 mutation of strain CSS74, a DNA fragment 393 encompassing the whole all3953 gene and sequences upstream from it was amplified 394 by PCR using the primer pair all3953-24/all3953-25, both including EcoRI sites, and 395 strain PCC 7120 DNA as the template. This fragment was cloned in the EcoRI site of 396 the mobilizable Nm r encoding vector pRL424 (Elhai and Wolk, 1988) producing plasmid 397 pCSS164, which was transferred to strain CSS74 by conjugation followed by selection 398 for Nm r . The genomic structure of the exconjugants in the all3953 region (Supp. Fig. 2) 399 was corroborated by PCR. 400 Page 38 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 16 To construct a strain expressing All3953-C-TAP, the all3953 gene (including the 401 upstream region) was PCR-amplified with primers all3953-11 and all3953-12 and DNA 402 of PCC 7120 as the template. The TAP-tag was PCR-amplified with primers TAPtag-1 403 and TAPtag-2 using DNA of plasmid pBS1479 as the template (Puig et al., 2001). The 404 two PCR products were digested with SalI and ligated, after which the ligation product 405 was used as the template for an overlapping PCR using primers all3953-11 and 406 TAPtag-2. The PCR product was digested with PstI and ligated to the mobilizable 407 vector pCSV3 (Valladares et al., 2011) digested with PstI, rendering plasmid pCSS107. 408 To construct a control strain with the TAP-tag under the control of the all3953 409 promoter, a 0.4-kb region upstream of all3953 was PCR-amplified using primers 410 all3953-11 and all3953-18 and DNA of pCSS107 as the template. The PCR product 411 was digested with SalI and ligated to the PCR-amplified TAP-tag digested with SalI, 412 after which the ligation product was used as the template for an overlapping PCR using 413 primers all3953-11 and TAPtag-2. The PCR product was digested with PstI and ligated 414 to PstI-digested pCSV3 to give plasmid pCSS157. Plasmids pCSS107 and pCSS157 415 were transferred by conjugation to strain PCC 7120 and single Sm r Sp r recombinants 416 were selected, obtaining strain CSS57 and CSS68, respectively. Western blots using 417 Peroxidase-Anti-Peroxidase Soluble Complex (PAP, Sigma-Aldrich) were performed to 418 ensure that the two strains expressed the TAP-tag (Supp. Fig. 2B). 419 420 Chromatin immunoprecipitation 421 Cells of strains CSS57 growing exponentially (3-5 µg Chl·ml -1 ) in the light (80 422 µE·m -2 ·s -1 ) in medium supplemented with 2 µg·ml -1 Sm and Sp, in HC conditions were 423 incubated with LC for 3 h. Formaldehyde was then added to the cultures to a final 424 concentration of 1%, and the cultures were incubated for 15 min. Glycine was added at 425 125 mM final concentration and the incubation was continued for 5 min to stop the 426 fixing reaction. The cells were then filtered, washed with cold TBS (20 mM Tris-HCl, pH 427 7.4, 140 mM NaCl) and collected in tubes (25 ml of culture per tube). The pellets were 428 Page 39 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 17 frozen in liquid nitrogen and stored at -20°C until used. Pellets corresponding to about 429 25 ml of culture were resuspended in 500 µl of lysis buffer (50 mM HEPES-KOH, pH 430 7.5, 140 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% sodium deoxycholate, 431 supplemented with Mini EDTA-free protease inhibitor cocktail [Roche]) and, after 432 addition of 150 µl of glass beads (acid-washed, 425-600 µm [Sigma]), the cells were 433 broken in a multivortexer at 2,000 rpm for 1 h at 4°C. The cell lysates were collected by 434 centrifugation and the extracts were subjected to sonication to shear the DNA to about 435 300-bp fragments (60 cycles of 10 s, 20 s ice, 15% amplitude, in a Branson Digital 436 Sonifier). After centrifugation to eliminate cell debris, the whole-cell extracts were 437 stored at −20°C or immediately used for immunoprecipitation. 438 Immunoprecipitation of DNA was carried out as described (Picossi et al., 2014), 439 with some modifications. Whole-cell extracts were prepared at 4 mg·ml -1 of total protein 440 with lysis buffer (in 500 µl total volume). A 50-µl sample was taken as the input sample, 441 and the extracts were incubated with 15 µl IgG-conjugated Dynabeads (about 6 µg 442 IgG) at 4ºC with rotation for 12-14h. The washes of the Dynabeads, as well as the 443 elution of the immunoprecipitated material, the crosslinking reversion and the isolation 444 of the DNA were performed as in (Picossi et al., 2014). 445 446 Massive sequencing of the immunoprecipitated DNA 447 Input and ChIP DNA samples were sent for sequencing to the Functional Genomics 448 Core Facility of the Institute for Research in Biomedicine, Barcelona (Spain). Next 449 generation sequencing was carried out using Illumina’s sequencing technology. ChIP 450 DNA Sample Prep Kit (Illumina) was used for library preparation. Libraries were loaded 451 at 8 pM concentration into the flow cell using the Cluster Station running recipe V7 with 452 the Single-Read Cluster Generation Kit v4 (all Illumina). The flow cell was loaded into 453 the Genome Analyzer II and samples were sequenced for 120 nucleotides from a 454 single end using the Sequencing Kit v5 and recipe v8 (all Illumina). Manufacturer’s 455 Page 40 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 24 (GTAN 8 TAC) is indicated in green, and the three putative binding sites for All3953 (Box 641 I, Box II and Box III) are indicated in blue. 642 643 Fig. 4. qRT-PCR analysis of the expression of selected photosynthesis and respiration-644 related All3953 gene targets. Transcriptional response of the indicated genes to C i 645 limitation in the control (CSS77) and ∆all3953 mutant (CSS74) strains was 646 investigated. RNA was isolated from cells grown with 10 mM NaHCO 3 -supplemented 647 medium bubbled with 1%CO 2 in air (0) incubated for 1 h (1) or 3 h (3) in NaHCO 3 -free 648 medium bubbled with air. Bars represent the mean transcript levels (± standard 649 deviation) in three independent experiments. 650 651 Page 47 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only Table 1. Results of the ChIP-Seq analysis of All3953 binding to DNA at 3 h after C i limitation *Some of the binding regions were ascribed to more than one gene (see Supp. Table 1). Binding regions found Genes ascribed Position of the binding region with respect to the gene upstream internal downstream Chromosome 127 157 118 29 10 Alp ha 10 13 8 4 2 Beta 3 3 0 3 0 Gamma 2 2 0 1 1 Total 142 175 126 37 13 Page 48 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 1 Table 2. Functional categories of the genes ascribed to the All3953 binding regions. Functional category Number Amino acid biosynthesis 5 Biosynthesis of cofactors, prosthetic groups, and carriers 4 Cell envelope 2 Cellular processes 5 Central intermediary metabolism 3 DNA replication, recombination and repair 3 Energy metabolism 4 Other categories 19 Photosynthesis and respiration 19 Purines, pyrimidines, nucleosides and nucleotides 2 Regulatory proteins 21 Translation 9 Transport and binding proteins 4 Unknown and hypothetical proteins 75 Page 49 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only Table 3. All3953-binding regions assigned to photosynthesis and respiration-related genes. BR NLQ LOC GENE POSITION FUNCTION ¶ ST START SEQ 7 172,84 239694 alr0223* upstream NADH dehydrogenase subunit 1; NdhA + 239712 AGGTATTAGTTTAACTAATGTT 12 38,58 398948 alr0348 upstream close NADH dehydrogenase subunit 4; NdhD + 398912 AACAATTCCTTAATATAATGTA 19 62,81 859596 all0737 upstream thioredoxin reductase - 859596 ATTCATAAAAGCGTTTTATATC 25 238,45 997693 all0868 upstream CO 2 concentrating mechanism protein CcmK + 997698 ATGTATAAGTTTTATTAATATG alr0869 upstream NADH dehydrogenase subunit 5 27 66,90 1175262 alr1004* upstream alanine--glyoxylate aminotransferase + 1175259 GTATATAGGCGATCATTATGGC 31 36,15 1433154 alr1216 upstream photosystem II 12 kD extrinsic protein PsbU + 1433147 AAATATTGTGAGCATTAATAAG 34 385,11 1547013 all1304* upstream bicarbonate transporter - 1546894 GTGCATTTGCAATAGTTATTAT 35 31,96 1620674 all1365 downstream cytochrome CytM + 1620645 CGTAATAAATTTTAATCATCAT 37 63,26 1785455 alr1524* upstream RbcL + 1785446 ACTTATGCCATTTCTTGATATA 38 204,06 1843221 alr1576 upstream far dehydrogenase subunit - 1843218 AGTAATAACTGCTACTTATTAC 61 29,78 3499605 alr2873 internal 3' end possible glycerate kinase - 1843218 CAAAATTAAACTGTCTAATTTC 86 170,25 4499861 alr3727 upstream photosystem II protein D1 (psbAII) + 4499883 GTATATATATTTTAGTAATATT 91 155,06 4693128 all3891* upstream flavoprotein (flv1A) + 4693099 ATTTATAAGTTTTACTTAAGCT 98 166,48 4993006 all4148 upstream ferredoxin I - 4993375 AACCATAAATTTTTCTAATAAC 99 61,47 4999503 alr4156* upstream NADH dehydrogenase subunit 5; NdhF + 4999503 AAAGATAAATTTGCCTTATTTA 108 163,25 5332512 all4446* upstream flavoprotein (flv4) - 5332500 AATAATAAATTTTACTAATAAA 112 233,71 5489698 alr4592* upstream photosystem II protein D1 (psbAIII) + 5489979 CTATATAGTTTTTACTCATATT 121 51,04 6151146 alr5154 upstream photosystem I core protein A1 + 6151133 GGACATAAGTTTTACGAATTGT *Genes whose expression has been studied by qRT-PCR ¶ Functions are as specified in cyanobase (http://genome.microbedb.jp/cyanobase/Anabaena). BR: binding region, NLQ: -logQvalue , LOC: chromosome location of the midpoint of the binding region, ST: DNA strand, START: chromosome location of the 5’ end of the putative binding sequence of All3953 (SEQ). Page 50 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only Figure 1 262x350mm (300 x 300 DPI) Page 51 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only Figure 2 262x350mm (300 x 300 DPI) Page 52 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only Figure 3 297x420mm (300 x 300 DPI) Page 53 of 54 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only Figure 4 297x420mm (300 x 300 DPI) Page 54 of 54 Wiley-Blackwell and Society for Applied Microbiology