Zur (FurB) is a key factor in the control of the oxidative stress response in Anabaena sp. PCC 7120

For Peer Review Only Zur (FurB) is a key factor in the control of the oxidative stress response in Anabaena sp. PCC 7120. Journal: Environmental Microbiology and Environmental Microbiology Reports Manuscript ID: EMI-2014-0941 Manuscript Type: EMI - Research article Journal: Environmental Microbiology Date Submitted by the Author: 16-Jul-2014 Complete List of Authors: Sein-Echaluce, Violeta; University of Zaragoza and BIFI, Biochemistry and Molecular and Cell Biology González, Andrés; University of Zaragoza and BIFI, Biochemistry and Molecular and Cell Biology Napolitano, Mauro; CSIC, Instituto de Bioquímica Vegetal y Fotosíntesis Luque, Ignacio; CSIC, Instituto de Bioquímica Vegetal y Fotosíntesis, Barja, Francois; UNIVERSITE DE GENEVE Unité de Microbiologie, Peleato, Maria Luisa; University of Zaragoza and BIFI, Biochemistry and Molecular and Cell Biology Fillat, Maria F.; University of Zaragoza and BIFI, Biochemistry and Molecular and Cell Biology Keywords: bacteria, ecophysiology, environmental signal/stress responses, gene expression/regulation, metabolic networks, metals, microbial genetics Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 1 2 Zur (FurB) is a key factor in the control of the oxidative stress 3 response in Anabaena sp. PCC 7120. 4 Violeta C. Sein-Echaluce 1 , Andrés González 1 , Mauro Napolitano 2 , Ignacio Luque 2 , 5 Francisco Barja 3 , M. Luisa Peleato 1 and María F. Fillat 1* 6 7 1 Departamento de Bioquímica y Biología Molecular y Celular, Universidad de 8 Zaragoza, 50009 Zaragoza, Spain; Instituto de Biocomputación y Física de Sistemas 9 Complejos (BIFI), Universidad de Zaragoza, 50009 Zaragoza, Spain. 10 2 Instituto de Bioquímica Vegetal y Fotosíntesis, CSIC-Universidad de Sevilla, E-41092 11 Sevilla, Spain. 12 3 Microbiology Unit, Botany and Plant Biology Dept., University of Geneva, Ch. Des 13 Embrouchis 10, CH-1254 Jussy-Geneva, Switzerland. 14 15 16 17 18 19 * For correspondence: María F. Fillat. Department of Biochemistry and Molecular and 20 Cell Biology. Faculty of Sciences. University of Zaragoza. Pedro Cerbuna, 12. 50009- 21 Zaragoza, Spain; Tel. (+34) 976761282; Fax (+34) 976762123. E-mail [email protected]. 22 23 24 25 Running title: Role of Zur in the oxidative stress response in Anabaena 26 27 28 29 Page 1 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 2 SUMMARY 30 31 Iron and zinc are necessary nutrients whose homeostasis is tightly controlled by 32 members of the FUR superfamily in the cyanobacterium Anabaena sp. PCC7120. 33 Although the link between iron metabolism and oxidative stress management is well 34 documented, little is known about the connection between zinc homeostasis and the 35 oxidative stress response in cyanobacteria. Zinc homeostasis in Anabaena is controlled 36 by Zur, the FurB paralogue. When overexpressed in Escherichia coli, Zur (FurB) 37 improved cell survival against oxidative stress. Zur also protected DNA against 38 hydroxyl radical damage in vitro. In order to investigate the possible correlation 39 between Zur and the oxidative stress response in Anabaena, zur deletion and zur- 40 overexpressing strains have been constructed and the consequences of Zur imbalance 41 evaluated. The lack of Zur increased sensitivity to H 2 O 2 , whereas an excess of Zur 42 enhanced oxidative stress resistance. Both mutants displayed pleiotropic phenotypes, 43 including alterations on the filament surfaces observable by scanning electron 44 microscopy, reduced content of endogenous H 2 O 2 and altered expression of sodA, 45 catalases and several peroxiredoxins. Transcriptional and biochemical analyses unveiled 46 that the appropriate level of Zur is required for proper control of the oxidative stress 47 response and allowed us to identify major antioxidant enzymes as novel members of the 48 Zur regulon. 49 50 Page 2 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 3 INTRODUCTION 51 52 Reactive oxygen species (ROS) are unavoidable by-products of aerobic metabolism that 53 can damage several cellular sites including iron-sulfur clusters, cysteine and methionine 54 protein residues, lipids and DNA (Chiang and Schellhorn, 2012). The generation of 55 toxic radicals is enhanced by a source of biometals including iron and zinc that, at the 56 same time, are essential nutrients for the vast majority of organisms (Stohs and Bagchi, 57 1995). Iron is a constituent of a wide range of proteins involved in photosynthesis, 58 respiration, nitrogen metabolism, defense against oxidative stress, DNA biosynthesis 59 and gene regulation, among others (Cornelis and Andrews, 2010). In spite of being the 60 fourth most abundant element in the Earth’s crust, the high reactivity of ferric iron with 61 oxygen to form insoluble oxides and hydroxides, makes this nutrient scarcely 62 bioavailable (Andrews et al., 2003). Unlike iron, zinc is only a trace metal in the Earth’s 63 crust. Nevertheless, zinc is involved in a huge number of biological processes and its 64 chemical properties make this metal a staple part of many proteins and enzymes, where 65 it plays structural, catalytic or regulatory roles. Although under physiological conditions 66 zinc is not a redox active metal, its capacity to bind and to protect free sulfhydryl groups 67 in proteins as well as the involvement of zinc proteins in the biosynthesis of low 68 molecular weight thiols links zinc homeostasis to the maintenance of the intracellular 69 redox status (Ma et al., 2009; Eide, 2011). Furthermore, the occurrence of a variety of 70 redox zinc switches coupled to thiol/disulphide exchanges enables a connection between 71 redox status and zinc metabolism (Maret, 2006). While iron toxicity lies in the ability of 72 this metal to effectively catalyze the Fenton reaction, zinc toxicity has been attributed to 73 its propensity to interact adventitiously with thiol groups of many proteins, especially 74 those involved in electron transport (Mills et al., 2002). Consequently, metal 75 concentrations inside the cells must be finely tuned not only for adjusting metal uptake 76 Page 3 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 4 to cell requirements, but also for keeping redox homeostasis in order to minimize 77 oxidative stress. In prokaryotes, this control is carried out by different families of 78 metalloregulators, which act in a coordinated way regulating metal homeostasis and 79 preventing cell oxidative damage (Ma et al., 2009). One of the most important families 80 of metalloregulators is constituted by the FUR (Ferric Uptake Regulator) proteins that 81 can be divided into different functional classes according to the signal they sense: Fur 82 (iron sensing proteins), Zur (zinc), Mur (manganese) and Nur (nickel). The PerR and Irr 83 subfamilies, which also belong to the FUR family, sense oxidative stress and haem 84 levels, respectively (Lee and Helmann, 2007). 85 Because of their photosynthetic metabolism, iron requirements of cyanobacteria are 86 much larger than those of heterotrophic microorganisms (Shcolnick et al., 2009). On the 87 other hand, the generation of reactive oxygen species by fortuitous electron transfer to 88 oxygen during the photosynthetic process enhances the risk of damage to iron-sulfur 89 clusters compromising cyanobacterial metabolism. Accordingly, metal homeostasis 90 mechanisms and the oxidative stress defenses are tightly coordinated in cyanobacteria, 91 where FUR proteins play major role. 92 Anabaena sp. PCC 7120 is a nitrogen-fixing cyanobacterium that in the absence of 93 combined nitrogen is able to differentiate heterocysts distributed semi-regularly in the 94 filament (Flores and Herrero, 2010). The genome of Anabaena sp. PCC 7120 encodes 95 three FUR proteins, previously named as FurA, FurB and FurC (Hernandez, 2004). 96 FurA is the master regulator of iron homeostasis and couples iron status with both 97 nitrogen metabolism and the oxidative stress response (Lopez-Gomollon et al., 2007b; 98 Lopez-Gomollon et al., 2007a; Gonzalez et al., 2010; González et al., 2011; González et 99 al., 2012). FurC has been identified as a PerR protein in Anabaena sp. PCC 7120 100 (Yingping et al., 2014), while FurB controls zinc homeostasis acting as a Zur protein 101 Page 4 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 5 (Napolitano et al., 2012). Zur (FurB) binds to DNA in a zinc-dependent manner and 102 represses transcription of target genes under zinc-sufficient conditions. Zur controls a 103 regulon which includes genes encoding putative metallochaperones (e.g. All4722, 104 All1751), paralogues of zinc metalloproteins (e.g. All4725/HemE, All4723/ThrS), 105 components of plasma membrane ABC transport systems (e.g. ZnuABC), and several 106 outer membrane proteins (e.g. Alr3242, Alr4028) (Napolitano et al., 2012). Similar 107 targets have been found or predicted to be regulated by Zur in other cyanobacteria 108 (Barnett et al., 2012) as well as in non-photosynthetic bacteria, such as Escherichia coli 109 (Patzer, 2000), Staphylococcus aureus (Lindsay and Foster, 2001), Bacillus subtilis 110 (Fuangthong and Helmann, 2003) and Mycobacterium tuberculosis (Maciag et al., 111 2006), among others. 112 Besides its role as a metal regulator, Zur from Anabaena sp. has been shown to enhance 113 cell survival under oxidative stress conditions when it is overexpressed in E. coli. In 114 addition, in vitro assays have shown the ability of this protein to unspecifically bind 115 DNA, protecting it against both, oxidative damage and DNaseI digestion (López- 116 Gomollón et al., 2009). Hence, a dual role for Anabaena sp. Zur has been previously 117 suggested depending on the protein concentration into the cell. At low concentrations, 118 Zur works as a transcriptional regulator binding to the promoters of target genes in a 119 specific manner. At higher concentrations of the protein, maybe induced by oxidative 120 stress, Zur would bind unspecifically to DNA, protecting it from oxidative damage 121 (López-Gomollón et al., 2009). 122 Correlation between oxidative stress and iron starvation has been well established in 123 cyanobacteria (Latifi et al., 2005; Shcolnick et al., 2009). However, the elements linking 124 zinc homeostasis to oxidative stress management by the cell remain to be identified. In 125 this study, the molecular bases of the connection between Zur and the oxidative stress 126 Page 5 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 6 response have been investigated. Our analyses of the phenotypes of two Anabaena sp. 127 derivative strains, a zur deletion mutant and a zur-overexpressing strain, show that 128 changes in zur expression levels deeply affect cyanobacterial phenotype, including 129 alterations in septum morphology and the organization of the outmost cell layers, 130 among other features. Transcriptional and biochemical assays led to the identification of 131 key genes involved in the oxidative stress response as novel members of the Zur 132 regulon. Those results establish a direct connection between the control of zinc 133 metabolism and the regulation of the antioxidant defenses in Anabaena sp. PCC 7120. 134 135 RESULTS 136 137 ∆ ∆∆ ∆ zur and zur overexpressing mutants of Anabaena sp. PCC 7120 exhibit a pleiotropic 138 phenotype 139 To achieve a better understanding of the alternative functions of Zur in Anabaena sp., a 140 zur overexpressing strain (VCS2770) was generated and its phenotype evaluated in 141 comparison with those from a zur deletion mutant (∆zur) and the parental wild type 142 Anabaena sp. PCC 7120. Photoautotrophic growth under standard culture conditions of 143 ∆zur was slower than that observed in the wild type strain (doubling time 10 days 144 versus 8.4 days). Conversely, the zur-overexpressing strain VCS2770 doubled in only 145 7.8 days. (Figure 1). Despite the zur-overespressing strain exhibited higher chlorophyll 146 content than ∆zur and the wild type control, photosynthetic and respiratory activities 147 were similar in the three Anabaena strains. (Table 2). 148 Cyanobacterial morphology of exponentially growing cultures was visualized using 149 different microscopy techniques. Bright-field and fluorescence microscopy analyses 150 showed that there were not noticeable differences in filament length and intrinsic 151 Page 6 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 7 fluorescence between the wild type and the derivative strains. Under absence of 152 combined nitrogen (BG11 0 ), cultures of ∆zur and VCS2770 strains displayed similar 153 heterocyst development patterns than those observed in Anabaena sp. PCC 7120 (data 154 not shown). However, scanning electron microscopy (SEM) analyses showed that ∆zur 155 cells displayed a different shape and appeared to be connected by narrower septa 156 compared to Anabaena sp. PCC 7120 and VCS2770 (Figure 2A). A severe disruption of 157 ∆zur filaments after treatment for transmission electron microscopy (TEM) studies in 158 comparison to Anabaena wild type and VCS2770 strains was also observed (not 159 shown). Those features, as well as the release of phycobiliproteins of the photosynthetic 160 antenna in ∆zur when cultures were left to decant overnight without bubbling (Fig. 2B) 161 could be indicative of a more fragile junction between cells. On the other hand, the cell 162 surface of the VCS2770 strain was visibly affected (Figure 2A), suggesting that not 163 only the slime sheath enclosing filaments might be influenced by zur overexpression, 164 but also the organisation of the outer membrane. 165 Oxidative stress tolerance in Anabaena sp. is strongly influenced by Zur expression 166 levels 167 Prior to the identification of FurB as a Zur regulator in Anabaena sp., it was found that 168 overexpression of FurB/Zur in E. coli increased its tolerance to ROS (López-Gomollón 169 et al., 2009). In the present study, we sought to investigate whether the expression levels 170 of Zur could influence tolerance to oxidative stress imposed by exogenous hydrogen 171 peroxide (H 2 O 2 ) in Anabaena cells. As shown in Figure 3, the Anabaena sp. strain 172 lacking Zur was much more sensitive to oxidative challenge than its parental strain. In 173 contrast, Zur overexpression increased cyanobacterial tolerance to hydrogen peroxide as 174 it was observed when this protein was overproduced in E. coli. 175 Page 7 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 8 ∆ ∆∆ ∆ zur and VCS2770 derivative strains exhibit altered superoxide dismutase (SOD) and 176 catalase activities, as well as diminished H 2 O 2 contents 177 To gain more insights about the mechanism underlying tolerance to H 2 O 2 , SOD and 178 catalase activities, as well as the endogenous content of H 2 O 2 were measured in the 179 three cyanobacterial strains. As shown in Fig. 4, the ∆zur strain showed increased SOD 180 and catalase activities (about 120% and 156%, respectively) compared to Anabaena sp. 181 PCC 7120. However, catalase activity appeared diminished in the zur-overexpressing 182 strain VCS2770 compared to the Anabaena sp. wild type (70%), while SOD activities 183 were similar in both strains. Surprisingly, the amount of endogenous H 2 O 2 dropped 184 dramatically in both, ∆zur and VCS2770 Anabaena strains, whose values were less than 185 10% of this from the parental Anabaena sp. control (Figure 5). Those results suggest 186 that the reduction in the amount of endogenous H 2 O 2 in ∆zur and VCS2770 strains takes 187 place through different pathways. 188 Changes in Zur levels affects the Anabaena sp. oxidative stress response machinery 189 These results prompted us to investigate a potential connection of Zur with the 190 transcription of main genes related to oxidative stress tolerance. Genes under study were 191 the two Mn-catalases (alr0998 and alr3090) encoded by Anabaena sp. PCC 7120, 192 superoxide dismutases and peroxiredoxins prxA, gct1 and gct3. Since FurB/Zur was 193 previously described as a DNA protecting protein (López-Gomollón et al., 2009), 194 transcription of several genes coding for DNA-binding proteins related to the oxidative 195 stress response, namely dpsA, hanA and all4145 (probable DNA-binding stress protein) 196 was also analysed. Finally, because of the implication of furA and furC (perR) in 197 cyanobacterial redox homeostasis, the influence of zur expression in these paralogs was 198 investigated. Changes in mRNA levels were determined by semi-quantitative reverse 199 Page 8 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 15 in Anabaena sp. PCC 7120, showing an interesting overlap between the FurA and Zur 350 regulons. In summary, those results evidence that Zur is a multifunctional regulatory 351 protein that connects zinc metabolism to oxidative stress management in Anabaena sp. 352 PCC 7120. 353 EXPERIMENTAL PROCEDURES 354 355 Bacterial strains and culture conditions 356 Bacterial strains used in this study are described in Table S1. Anabaena sp. PCC 7120, 357 the zur-overexpressing derivative mutant VCS2770 and the deletion-insertion mutant 358 ∆zur were grown photoautotrophically in BG-11 medium (Rippka et al., 1978) at 28°C 359 under a constant illumination of 50 µE m − 2 s – 1 . Culture medium was supplemented 360 with neomycin 50 µg ml -1 for strain VCS2770 and with streptomycin and 361 spectinomycin 2-5 µg ml -1 for strain ∆zur. Cultures were performed using Erlenmeyer 362 flasks at a constant shaking of 120 r.p.m. 363 Escherichia coli strains used for cloning procedures were grown at 37°C in Luria- 364 Bertani medium, supplemented with the appropiated antibiotics at the following 365 concentrations: kanamycin 50 µg ml -1 , ampicillin 50 µg ml -1 , chloramphenicol 30 µg 366 ml -1 , streptomycin 25 µg ml -1 and spectinomycin 100 µg ml -1 . 367 368 Construction of the Anabaena sp. PCC 7120 derivative strains 369 The zur deletion-insertion strain was described elsewhere (Napolitano et al., 2012). The 370 zur-overexpressing strain was constructed as follows. Chromosomal DNA was 371 extracted from Anabaena sp. PCC 7120 (Cai and Wolk, 1990) and used as a template to 372 amplify the zur gene, using primers 2770FurB_up and 2770FurB_dw (Table S2 ) . These 373 primers contained the restriction sites for BamHI and NdeI enzymes, so that the PCR 374 product was double digested and cloned into those restriction sites in the pAM2770 375 Page 15 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 16 shuttle vector (Lee et al., 2003). The resulting plasmid, pAM2770::zur, contained the 376 zur gene downstream the petE promoter, which is inducible by copper (Buikema and 377 Haselkorn, 2001). The construction was sequenced to verify that the cloning procedure 378 was successful. Plasmid pAM2770::zur was transferred to Anabaena sp. PCC 7120 by 379 triparental mating (Elhai et al., 1997). Three exconjugant clones were cultured in BG-11 380 medium to an optical density of 0.5 at 750 nm and then collected to test the 381 overexpression of Zur by Western blot (Figure S1). As previously described, the 382 amount of copper in BG-11 medium was enough to activate the petE promoter 383 (Gonzalez et al., 2010). Therefore, no additional copper was added to enhance the 384 overexpression of zur. Finally, the exconjugant clone with the highest expression of zur 385 was selected as the zur-overexpressing strain, named VCS2770. 386 387 Cyanobacterial growth and pigment measurements 388 In order to analyze cyanobacterial growth, all strains were cultured in Erlenmeyer flasks 389 and the optical density was measured at 750 nm every 2-3 days during 40 days. Growth 390 parameters such as the specific growth rate and doubling time were calculated as 391 previously described (Stein, 1973). Measurements were carried out using a Cary 100 392 Bio UV-visible spectrophotometer (Varian). 393 Quantification of photosynthetic pigments and total protein content was performed in 394 cultures at the exponential phase of growth. Chlorophyll a (Nicolaisen et al., 2008), 395 phycobiliproteins (Glazer, 1976) and carotenoids (Davies, 1976) were quantified as 396 previously described. Total protein content was determined by using the bicinchoninic 397 acid protein assay (Pierce). Pigment and protein content were expressed as micrograms 398 per microliter of packed cell volume (PCV), where the PCV was determined by 399 centrifuging 5 milliliters of each culture for 5 minutes at 2,000xg in a special graduated 400 tube. 401 Page 16 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 17 402 Photosynthetic and respiratory activities 403 404 Photosynthetic and respiratory activities were measured in exponentially growing cells 405 with a Clark type electrode (Oxylab model by Hansatech) at 21°C and a constant 406 shaking of 65%. Photosynthetic activity was determined by measuring the oxygen rate 407 at light saturating conditions (400 µE m − 2 s – 1 ), while the respiratory activity was 408 determined at dark conditions, covering the electrode chamber with an aluminum paper. 409 Data were processed with the provided Hansatech software and the results were 410 expressed as nmol O 2 ·ml -1 ·min -1 ·mg Chla -1 . 411 412 Purification of Zur and Western blot analyses 413 To obtain the recombinant Zur protein, the zur gene was amplified using all2473N-2 414 and all2473C primers described in Table S2 and cloned between the NdeI and HindIII 415 sites of plasmid pET 28a(+). The resulting His-tagged protein was purified using a zinc 416 affinity column (Matrix Chelating Sepharose™ Fast Flow, Amersham) and conserved 417 in a 100 mM NaH 2 PO 4 , 300 mM NaCl pH 6 solution. For Western blot analysis, 418 cyanobacterial liquid cultures were collected by centrifugation at 4°C and cells were 419 resuspended in cold phosphate buffer 50 mM pH 8. The suspension was sonicated five 420 times during 45 seconds with cooling intervals of 30 seconds and then centrifuged to 421 remove cell debris. Protein concentration in crude extracts was determined by using the 422 bicinchoninic acid method (BCA™ Protein Assay Kit, Thermo Fischer Scientific). For 423 each sample, 10-30 micrograms of total proteins were loaded and separated by 424 electrophoresis with 17% SDS-PAGE gels. Proteins were transferred to a PVDF 425 membrane (0.45 µm pore size Immobilon® transfer membrane from Millipore) and 426 immunodetection was carried out using rabbit polyclonal antibodies raised against Zur. 427 428 Catalase and superoxide dismutase activities determinations 429 Page 17 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 18 Fifty milliliters of each cyanobacterial culture were collected by centrifugation and cells 430 were resuspended in one milliliter of phosphate buffer 50 mM pH 8. The cell 431 suspension was sonicated and then centrifuged to remove cell debris. Protein 432 concentration in the extract was quantified by using the BCA method and antioxidant 433 activities were immediately determined. Catalase activity was determined as decribed 434 previously (Beers and Sizer, 1952), following the hydrogen peroxide dissociation by 435 measuring the optical density at 240 nm. Breafly, 300-600 micrograms of protein 436 extract were rapidly mixed in a quartz cuvette with hydrogen peroxide to a final 437 concentration of 20 mM. The reaction was followed spectrophotometrically at 240 nm 438 with a Cary 100 Bio (Varian) device during five minutes. Catalase activity was 439 expressed as Units per milligram of total proteins, defining a Unit as the amount of 440 enzyme that catalyzes the dissociation of 1 microgram of hydrogen peroxide per minute. 441 Superoxide dismutase (SOD) activity was determined by a modification of the method 442 by Winterbourn (Winterbourn et al., 1975), which is based on the ability of SOD to 443 inhibit the reduction of nitro-blue tetrazolium (NBT) by superoxides. Reaction mixtures 444 contained 600 micrograms of protein extract, 6.4 mM EDTA, 41 µM NBT, 2.3 µM 445 riboflavin and 23.5 µM TEMED. The control of the assay, with a maximum NBT 446 reduction, contained phosphate buffer instead of the protein extract. Optical density at 447 560 nm was measured before and after illuminating the mixtures for 10 minutes with 448 UV light. Superoxide dismutase activity was expressed as Units per milligram of total 449 proteins, defining a Unit as the amount of enzyme that inhibited the maximum reduction 450 in a 50%. 451 452 Endogenous hydrogen peroxide measurement 453 Intracellular hydrogen peroxide was determined in the cyanobacterial strains using the 454 ferrithiocyanate method (Thurman et al., 1972). Fifty milliliters of each culture were 455 Page 18 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 19 collected to obtain 50-100 milligrams of fresh cells. Trichloroacetic acid (TCA) at a 456 final concentration of 5% was added to the cells and the mixture was centrifuged to 457 remove cell debris. 800 microliters of the supernatant were mixed in a plastic cuvette 458 with 160 µl of Fe(NH 4 ) 2 (SO 4 ) 2 10 mM (1.3 mM final concentration), 80 µl of KSCN 459 2.5 M (167 mM final concentration) and 160 µl of TCA 50%. The absorbance at 480 460 nm was measured using a Cary 100 Bio spectrophotometer (Varian) to determine the 461 hydrogen peroxide content. 462 463 Hydrogen peroxide tolerance assay 464 To test the tolerance of the cyanobacterial strains to hydrogen peroxide, filaments were 465 exposed to increasing concentrations of hydrogen peroxide for 24 hours. Cultures with 466 an approximate optical density of 1.0 at 750 nm were washed once with fresh BG-11 467 medium and 200 µl of culture were displayed into each well of a microtiter plate. 468 Hydrogen peroxide was added to the wells at a final concentration of 0, 0.5, 0.7, 1 and 469 1.3 mM. The plate was incubated for 24 hours in dark conditions and 28°C. Chlorosis 470 was estimated by reading the absorbance at 620 nm with a Multiskan EX microplate 471 photometer (Thermo Fischer Scientific). 472 473 Microscopy 474 475 Bright-field and fluorescence microscopy analysis of exponentially growing cells were 476 carried out using a Nikon Eclipse 50i Epi-fluorescence microscope coupled to a Nikon 477 DXM 1200F camera. For scanning electron microscopy, cells were harvested at the 478 exponential phase of growth and fixed with 2.5% glutaraldehyde in phosphate buffer 479 (66 mM NaH 2 PO 4 .2H 2 O, 66 mM KH 2 PO 4 , pH 7) for 1 h at room temperature, washed 480 three times for 5 min each in phosphate buffer, fixed with 2% OsO 4 and washed three 481 Page 19 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 20 times for 5 min each in distilled water. Scanning electron microscopy was performed in 482 a SEM JEOL JSM 7001FA. Pictures were processed using the Photoshop 6.0 program. 483 484 Semi-quantitative reverse transcription (RT-PCR) 485 Total RNA was isolated from exponentially growing cultures as previously described 486 (Olmedo-Verd et al., 2005) and residual DNA was removed by treating the samples 487 with RNAse-free DNAseI (Roche). The successful removal of genomic DNA was tested 488 by PCR. RNA was heated at 85°C for 10 minutes and reverse transcription was carried 489 out using the SuperScript Reverse Transcriptase kit (Invitrogen) and following the 490 manufacturer’s conditions. The rnpB gene was used as an internal control to normalize 491 the amounts of cDNA in the PCR reactions. The results of the PCRs were visualized in 492 1-1.5% agarose gels stained with ethidium bromide in a GelDoc 2000 device (Bio-Rad). 493 494 Electrophoretic Mobility Shift Assays (EMSAs) 495 496 Gene promoters were obtained by PCR, using the Anabaena sp. PCC 7120 genome as a 497 template using the primers described in Table S2. To ensure the specific binding of the 498 protein to the studied promoters, the promoter of nifJ (alr1911) gene was used as a 499 competitor DNA in all reactions. Reaction mixtures with a final volume of 20 µl 500 contained 50 ng of each promoter, binding buffer (10 mM bis-Tris pH7.5, 40 mM KCl, 501 2 mM MgCl 2 ·6H 2 O, 5% glycerol), 0.05 mg/ml BSA, 1 mM DTT, 5µM ZnSO 4 ·H 2 O and 502 100-300 nM recombinant Zur. Resulting mixtures were incubated for 30 minutes at 503 room temperature and loaded into non-denaturing 6% polyacrylamide gels. Gels were 504 stained with SYBR®Safe (Invitrogen) and visualized in a GelDoc 2000 device (Bio- 505 Rad). 506 507 ACKNOWLEDGEMENTS 508 Page 20 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 21 This work has been supported by grants BFU2012-31458 from MINECO, B18 from 509 Gobierno de Aragón and Obra Social La Caixa. VCS is recipient of a fellowship from 510 Gobierno de Aragón. The authors have no conflict of interest to declare. 511 512 REFERENCES 513 514 Andrews, S.C., Robinson, A.K., and Rodríguez-Quiñones, F. (2003) Bacterial iron 515 homeostasis. FEMS Microbiology Reviews 27: 215-237. 516 Banerjee, M., Ballal, A., and Apte, S.K. (2012) Mn-catalase (Alr0998) protects the 517 photosynthetic, nitrogen-fixing cyanobacterium Anabaena PCC7120 from oxidative 518 stress. Environmental Microbiology 14: 2891-2900. 519 Banerjee, M., Raghavan, P.S., Ballal, A., Rajaram, H., and Apte, S.K. (2013) Oxidative 520 stress management in the filamentous, heterocystous, diazotrophic cyanobacterium, 521 Anabaena PCC7120. Photosynthesis Research 118: 59-70. 522 Barnett, J.P., Millard, A., Ksibe, A.Z., Scanlan, D.J., Schmid, R., and Blindauer, C.A. 523 (2012) Mining genomes of marine cyanobacteria for elements of zinc homeostasis. 524 Front Microbiol 3: 142. 525 Beers, R.F., and Sizer, I.W. (1952) A spectrophotmetric method for measuring the 526 breakdown of hydrogen peroxide by catalase. Journal of Biological Chemistry 195: 527 133-140. 528 Bihani, S.C., Chakravarty, D., and Ballal, A. (2013) Purification, crystallization and 529 preliminary crystallographic analysis of KatB, a manganese catalase from Anabaena 530 PCC 7120. Acta Crystallographica Section F Structural Biology and Crystallization 531 Communications 69: 1299-1302. 532 Bonet, B., Corcoll, N., and Guasch, H. (2012) Antioxidant enzyme activities as 533 biomarkers of Zn pollution in fluvial biofilms. Ecotoxicol Environ Saf 80: 172-178. 534 Page 21 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 22 Botello-Morte, L., Gonzalez, A., Bes, M.T., Peleato, M.L., and Fillat, M.F. (2013) 535 Functional Genomics of Metalloregulators in Cyanobacteria. Genomics of 536 Cyanobacteria 65: 107-156. 537 Buikema, W.J., and Haselkorn, R. (2001) Expression of the Anabaena hetR gene from a 538 copper-regulated promoter leads to heterocyst differentiation under repressing 539 conditions. Proceedings of the National Academy of Sciences 98: 2729-2734. 540 Cai, Y., and Wolk, P. (1990) Use of a Conditionally Lethal Gene in Anabaena sp. strain 541 PCC 7120 To Select for Double Recombinants and To Entrap Insertion Sequences. 542 Journal of Bacteriology 172: 3138-3145. 543 Cornelis, P., Andrews, S. C., eds. (2010) Iron Uptake and Homeostasis in 544 Microorganisms. Caister Academic Press, Norfolk, UK. 545 Chiang, S.M., and Schellhorn, H.E. (2012) Regulators of oxidative stress response 546 genes in Escherichia coli and their functional conservation in bacteria. Arch Biochem 547 Biophys 525: 161-169. 548 Choi, S., and Bird, A.J. (2014) Zinc'ing sensibly: controlling zinc homeostasis at the 549 transcriptional level. Metallomics 6: 1198-1215. 550 Davies, B.H. (1976) Carotenoids. In Chemistry and Biochemistry of Plant Pigments. 551 Goodwin, T.W. (ed). New York: Academic Press, pp. 38-165. 552 Divakaruni, A.V., Loo, R.R.O., Xie, Y., Loo, J.A., and Gober, J.W. (2005) The cell- 553 shape protein MreC interacts with extracytoplasmic proteins including cell wall 554 assembly complexes in Caulobacter crescentus. Proceedings of the National Academy 555 of Sciences 102: 18602-18607. 556 Eide, D.J. (2011) The oxidative stress of zinc deficiency. Metallomics 3: 1124-1129. 557 Eijkelkamp, B.A., Morey, J.R., Ween, M.P., Ong, C.L., McEwan, A.G., Paton, J.C., and 558 McDevitt, C.A. (2014) Extracellular zinc competitively inhibits manganese uptake 559 Page 22 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 23 and compromises oxidative stress management in Streptococcus pneumoniae. PLoS 560 One 9: e89427. 561 Elhai, J., Vepritskiy, A., Muro-Pastor, A.M., Flores, E., and Wolk, P. (1997) Reduction 562 of conjugal transfer efficiency by three restriction activities of Anabaena sp. strain 563 PCC 7120. Journal of Bacteriology 179: 1998-2005. 564 Flores, E., and Herrero, A. (2010) Compartmentalized function through cell 565 differentiation in filamentous cyanobacteria. Nat Rev Microbiol 8: 39-50. 566 Fuangthong, M., and Helmann, J.D. (2003) Recognition of DNA by Three Ferric 567 Uptake Regulator (Fur) Homologs in Bacillus subtilis. Journal of Bacteriology 185: 568 6348-6357. 569 Gabriel, S.E., and Helmann, J.D. (2009) Contributions of Zur-controlled ribosomal 570 proteins to growth under zinc starvation conditions. J Bacteriol 191: 6116-6122. 571 Glazer, A.N. (1976) Phycocyanins. In Photochemical and Photobiological Reviews. 572 Smith, K.C. (ed). New York: Plenum Press, pp. 71-115. 573 Gonzalez, A., Bes, M.T., Barja, F., Peleato, M.L., and Fillat, M.F. (2010) 574 Overexpression of FurA in Anabaena sp. PCC 7120 Reveals New Targets for This 575 Regulator Involved in Photosynthesis, Iron Uptake and Cellular Morphology. Plant 576 and Cell Physiology 51: 1900-1914. 577 González, A., Bes, M.T., Peleato, M.L., and Fillat, M.F. (2011) Unravelling the 578 regulatory function of FurA in Anabaena sp. PCC 7120 through 2-D DIGE proteomic 579 analysis. Journal of Proteomics 74: 660-671. 580 González, A., Bes, M.T., Valladares, A., Peleato, M.L., and Fillat, M.F. (2012) FurA is 581 the master regulator of iron homeostasis and modulates the expression of tetrapyrrole 582 biosynthesis genes in Anabaena sp. PCC 7120. Environmental Microbiology 14: 583 3175-3187. 584 Page 23 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 24 Graham, A.I., Sanguinetti, G., Bramall, N., McLeod, C.W., and Poole, R.K. (2012) 585 Dynamics of a starvation-to-surfeit shift: a transcriptomic and modelling analysis of 586 the bacterial response to zinc reveals transient behaviour of the Fur and SoxS 587 regulators. Microbiology 158: 284-292. 588 Hassan, H.M., and Sun, H.-C.H. (1992) Regulatory roles of Fnr, Fur and Arc in 589 expression of manganese-containing superoxide dismutase in Escherichia coli. 590 Proceedings of the National Academy of Sciences 89: 3217-3221. 591 Hasset, D.J., Howell, M.L., Ochsner, U.A., Vasil, M.L., Johnson, Z., and Dean, G.E. 592 (1997) An Operon Containing fumC and sodA Encoding Fumarase C and Manganese 593 Superoxide Dismutase Is Controlled by the Ferric Uptake Regulator in Pseudomonas 594 aeruginosa: fur Mutants Produce Elevated Alginate Levels. Journal of Bacteriology: 595 1452-1459. 596 Hernandez, J. (2004) Three fur homologues from Anabaena sp. PCC7120: exploring 597 reciprocal protein-promoter recognition. FEMS Microbiology Letters 236: 275-282. 598 Hernandez, J.A., Pellicer, S., Huang, L., Peleato, M.L., and Fillat, M.F. (2007) FurA 599 modulates gene expression of alr3808, a DpsA homologue in Nostoc (Anabaena) sp. 600 PCC7120. FEBS Lett 581: 1351-1356. 601 Hernandez, J.A., Muro-Pastor, A.M., Flores, E., Bes, M.T., Peleato, M.L., and Fillat, 602 M.F. (2006) Identification of a furA cis antisense RNA in the cyanobacterium 603 Anabaena sp. PCC 7120. J Mol Biol 355: 325-334. 604 Kallifidas, D., Pascoe, B., Owen, G.A., Strain-Damerell, C.M., Hong, H.J., and Paget, 605 M.S.B. (2009) The Zinc-Responsive Regulator Zur Controls Expression of the 606 Coelibactin Gene Cluster in Streptomyces coelicolor. Journal of Bacteriology 192: 607 608-611. 608 Page 24 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only 31 Table 3. Relative induction ratio of selected genes in relation to the wild type 750 strain. Data are derived from the results of the semi quantitative RT-PCR analyses. 751 Intensity of the DNA bands in the agarose gels was determined with ImageJ software. 752 Values are means of two independent determinations ± SD. 753 754 SUPPLEMENTARY MATERIAL 755 756 Figure S1. Verification of zur deletion and overexpression by Western blot. A) 757 Verification of zur deletion. Lanes contain 30 µg of Anabaena sp. PCC 7120 and 758 Anabaena ∆zur protein extracts respectively. B) Verification of zur overexpression. 759 Lanes contain 6.5 µg of protein extracts from Anabaena sp. PCC 7120, zur- 760 overexpressing strain (VCS2770) and Anabaena ∆zur respectively. Molecular weight of 761 protein marker bands is expressed in kDa. 762 763 Table S1. Bacterial strains used in this study. 764 765 Table S2. Oligonucleotides used in this study. 766 767 Page 31 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only Figure 1 0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 10 20 30 40 50 OD (750 nm) Time (days) VCS2770 WT ∆zur Page 32 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only Figure 2 A B WT VCS2770 ∆zur Page 33 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only Figure 3 WT ∆zur VCS2770 H2O2 (nM) 0 0.5 0.7 1 1.3 A B 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0 0.5 0.7 1 1.3 OD (620 nm) H2O2 (nM) WT ∆zur VCS2770 Page 34 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only Figure 4 0 0.5 1 1.5 2 2.5 3 SOD activity (U/mg) WT Δzur VCS2770 A B 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Catalase activity (U/mg) WT Δzur VCS2770 Page 35 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only Figure 5 0 0.2 0.4 0.6 0.8 1 1.2 nmolH2O2/mgFW WT Δzur VCS2770 Page 36 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only Figure 6 rnpB furB furA furC sodA sodB alr0998 alr3090 Δzur WT VCS2770 Δzur WT VCS2770 gct3 gct1 prxA dpsA hanA all4145 Page 37 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only Figure 7 PnifJ - + FurB 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 all4725 furA alr0998 alr3090 sodA gct3 prxA dpsA 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 PnifJ - + FurB Page 38 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only Table 1 Pigments Strains WT ∆zur VCS2770 Chlorophyll a 2.20 ± 0.14 2.56 ± 0.61 4.06 ± 0.45 Carotenoids 0.050 ± 0.00 0.034 ± 0.00 0.052 ± 0.00 Phycobiliproteins 24.56 ± 0.03 24.64 ± 0.45 27.60 ± 0.42 Page 39 of 41 Wiley-Blackwell and Society for Applied Microbiology For Peer Review Only Table 2 Strain Photosynthesis Respiration Total oxygen evolution WT 4.76 ± 0.17 0.68 ± 0.17 5.44± 0.17 ∆zur 4.29 ± 0.22 0.82 ± 0.15 5.11± 0.37 VCS2770 4.69 ± 0.28 0.75 ± 0.16 5.81± 0.44 Page 40 of 41 Wiley-Blackwell and Society for Applied Microbiology