scieee Open visual document viewer

RNA-seq analysis of the Rhizobium tropici CIAT 899 transcriptome shows similarities in the activation patterns of symbiotic genes in the presence of apigenin and salt

Pérez Montaño, Francisco de Asís; Cerro Sánchez, Pablo del; Jiménez Guerrero, Irene; López Baena, Francisco Javier; Cubo Sánchez, María Teresa; Hungria, Mariangela; Megías Guijo, Manuel; Ollero Márquez, Francisco Javier

Abstract

Background Rhizobium tropici strain CIAT 899 establishes effective symbioses with several legume species, including Phaseolus vulgaris and Leucaena leucocephala. This bacterium synthesizes a large variety of nodulation factors in response to nod-gene inducing flavonoids and, surprisingly, also under salt stress conditions. The aim of this study was to identify differentially expressed genes in the presence of both inducer molecules, and analyze the promoter regions located upstream of these genes. Results Results obtained by RNA-seq analyses of CIAT 899 induced with apigenin, a nod gene-inducing flavonoid for this strain, or salt allowed the identification of 19 and 790 differentially expressed genes, respectively. Fifteen of these genes were up-regulated in both conditions and were involved in the synthesis of both Nod factors and indole-3-acetic acid. Transcription of these genes was presumably activated through binding of at least one of the five NodD proteins present in this strain to specific nod box promoter sequences when the bacterium was induced by both apigenin and salt. Finally, under saline conditions, many other transcriptional responses were detected, including an increase in the transcription of genes involved in trehalose catabolism, chemotaxis and protein secretion, as well as ribosomal genes, and a decrease in the transcription of genes involved in transmembrane transport. Conclusions To our knowledge this is the first time that a transcriptomic study shows that salt stress induces the expression of nodulation genes in the absence of flavonoids. Thus, in the presence of both nodulation inducer molecules, apigenin and salt, R. tropici CIAT 899 up-regulated the same set of symbiotic genes. It could be possible that the increases in the transcription levels of several genes related to nodulation under saline conditions could represent a strategy to establish symbiosis under abiotic stressing conditions.

Full text

RESEARCH ARTICLE Open Access RNA-seq analysis o he Rhizobium opici CIAT 899 ansc ip ome shows simila i ies in he ac i a ion pa e ns o symbio ic genes in he p esence o apigenin and sal F ancisco Pé ez-Mon año 1*† , Pablo del Ce o 1† , I ene Jiménez-Gue e o 1 , F ancisco Ja ie López-Baena 1 , Ma ia Te esa Cubo 1 , Ma iangela Hung ia 2 , Manuel Megías 1 and F ancisco Ja ie Olle o 1 Abs ac Backg ound: Rhizobium opici s ain CIAT 899 es ablishes e ec i e symbioses wi h se e al legume species, including Phaseolus ulga is and Leucaena leucocephala. This bac e ium syn hesizes a la ge a ie y o nodula ion ac o s in esponse o nod-gene inducing la onoids and, su p isingly, also unde sal s ess condi ions. The aim o his s udy was o iden i y di e en ially exp essed genes in he p esence o bo h induce molecules, and analyze he p omo e egions loca ed ups eam o hese genes. Resul s: Resul s ob ained by RNA-seq analyses o CIAT 899 induced wi h apigenin, a nod gene-inducing la onoid o his s ain, o sal allowed he iden i ica ion o 19 and 790 di e en ially exp essed genes, espec i ely. Fi een o hese genes we e up- egula ed in bo h condi ions and we e in ol ed in he syn hesis o bo h Nod ac o s and indole-3-ace ic acid. T ansc ip ion o hese genes was p esumably ac i a ed h ough binding o a leas one o he i e NodD p o eins p esen in his s ain o speci ic nod box p omo e sequences when he bac e ium was induced by bo h apigenin and sal . Finally, unde saline condi ions, many o he ansc ip ional esponses we e de ec ed, including an inc ease in he ansc ip ion o genes in ol ed in ehalose ca abolism, chemo axis and p o ein sec e ion, as well as ibosomal genes, and a dec ease in he ansc ip ion o genes in ol ed in ansmemb ane anspo . Conclusions: To ou knowledge his is he i s ime ha a ansc ip omic s udy shows ha sal s ess induces he exp ession o nodula ion genes in he absence o la onoids. Thus, in he p esence o bo h nodula ion induce molecules, apigenin and sal , R. opici CIAT 899 up- egula ed he same se o symbio ic genes. I could be possible ha he inc eases in he ansc ip ion le els o se e al genes ela ed o nodula ion unde saline condi ions could ep esen a s a egy o es ablish symbiosis unde abio ic s essing condi ions. Keywo ds: RNA-seq, Rhizobium opici CIAT 899, Nodula ion, Nod ac o s, Lipochi ooligosaccha ides, Apigenin, Sal s ess * Co espondence: [email p o ec ed] † Equal con ibu o s 1 Depa amen o de Mic obiología, Facul ad de Biología, Uni e sidad de Se illa, A enida Reina Me cedes n° 6, 41012 Se illa, Spain Full lis o au ho in o ma ion is a ailable a he end o he a icle © 2016 Pé ez-Mon año e al. Open Access This a icle is dis ibu ed unde he e ms o he C ea i e Commons A ibu ion 4.0 In e na ional License (h p://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons license, and indica e i changes we e made. The C ea i e Commons Public Domain Dedica ion wai e (h p://c ea i ecommons.o g/publicdomain/ze o/1.0/) applies o he da a made a ailable in his a icle, unless o he wise s a ed. Pé ez-Mon año e al. BMC Genomics (2016) 17:198 DOI 10.1186/s12864-016-2543-3 Backg ound Rhizobia comp ise a g oup o α-andβ-p o eobac e ia known o hei abili y o es ablish symbioses wi h se - e al leguminous species. The hizobium-legume in e - ac ion, cha ac e ized by an exchange o signal molecules om bo h pa ne s, culmina es in he o ma ion o spe- ci ic s uc u es, called nodules, whe e biological ni ogen ixa ion akes place [1–5]. This molecula dialogue be- gins wi h he exuda ion o la onoids by he hos legume oo s ha a e ecognized by a LysR- ype ansc ip ional egula o in he bac e ium, he NodD p o ein, which igge s he exp ession o he so-called nodula ion (nod) genes by binding o speci ic sequences, nod boxes (NB), loca ed ups eam o hese genes. Thei cogna e enzymes a e implied in he p oduc ion o lipochi ooligosaccha - ides, also known as Nod ac o s (NF), which in u ns in- duce he o ma ion o oo nodule p imo dia and play an essen ial ole in he in ec ion p ocess. A pa om la onoids, o he hizobial nod gene induce s ha e been iden i ied, such as be aines [6], bu hey a e equi ed a highe concen a ions [7]. Rhizobium opici CIAT 899 (he ea e CIAT 899) is a b oad hos - ange hizobial s ain isola ed om opical acid soils o Sou h Ame ica ha e ec i ely nodula es se e al legumes, including Phaseolus ulga is,Mac op i- lium a opu pu eum, and Leucaena leucocephala [8, 9]. Main cha ac e is ics o his s ain includes i s high ole - ance o se e al en i onmen al s esses such as high empe a u e, acidi y o salini y and i s capaci y o p odu- cing a la ge a ie y o NF in he p esence o induce la- onoids, such as apigenin [8, 10, 11]. I is ema kable ha unde acidi y o sal s ess condi ions he syn hesis o NF in CIAT 899 is also induced, esul ing in inc eased di e si y and concen a ion o hese molecules in com- pa ison o non-s essing condi ions [12, 13]. In e es - ingly, Guasch-Vidal e al. [14] demons a ed ha , e en in he absence o la onoids, CIAT 899 is able o syn he- sizing NF in he p esence o high concen a ions o sal , and he biological ac i i y o hese NF was con i med. Mo eo e , he ac i a ion unde sal s ess is independen o NodD1 [14]. I has been epo ed ha ini ial s eps o hizobium-legume symbioses a e e y sensi i e o sal s ess. Howe e , he abili y o o m oo nodules on hei hos legume species unde saline condi ions has been desc ibed o many hizobia [15]. In gene al, hizobial s ains use dis inc mechanisms o osmo ic adap a ion unde sal s essing condi ions du ing he ee-li es yle, such as he in acellula accumula ion o osmoly es and speci ic ions, modi ica ion in cell su ace polysaccha ides o he syn hesis o ce ain ABC memb ane anspo e s [15, 16]. Howe e , he syn hesis o sal -induced NF has been only epo ed in CIAT 899. Genome sequencing o CIAT 899 e ealed i e di e - en nodD genes and h ee di e en nodA genes in he symbio ic plasmid [17]. NodA ca alyzes he ans e o he a y acyl g oup om an acyl ca ie p o ein o a e - minal N-glucosamine esidue p e iously deace yled by NodB, on he chi in oligome [18]. The nodA1 gene is loca ed adjacen o nodD1, whose encoded p o ein seems o be he majo egula o o NF syn hesis upon induc ion wi h la onoid [19, 20] and oge he wi h nodBC compose an ope on esponsible o he syn hesis o he NF co e. The nodA2 gene is pa o a gene clus e including hsnT and nodFE, implied in unsa u a ed a y acid inco po a ion in o NF molecules and is loca ed close o he nodD2 gene. Cu iously, a p e ious s udy [20] showed ha , appa en ly, he ac i a ion o he ex- p ession o he nodC gene unde sal s ess is lowe in a nodD2 mu an han in bo h a nodD1 mu an and he wild- ype s ain. Finally, nodA3 is loca ed downs eam he nodD3 gene bu no o he symbio ic- ela ed genes ha e been iden i ied in i s icini ies [21]. The main objec i e o his s udy was o iden i y genes o CIAT 899 ha a e di e en ially exp essed in he p es- ence o he nod-gene induce molecules apigenin and sal by RNA-seq analysis. In addi ion, he p omo e e- gions o he symbio ic genes ha we e up- egula ed in bo h condi ions we e s udied o de e mine possible con- se ed p omo e consensus mo i s. Ou esul s showed simila pa e ns o exp ession o he di e en ially exp essed genes o he symbio ic plasmid eplicon in he p esence o bo h apigenin and sal , indica ing ha he NF syn hesis was ca ied ou ollowing he same pa h- way, independen ly o he induce molecule. The bio- logical signi icance o he CIAT 899 ansc ip omic esponse unde sal condi ion was discussed. Resul s and discussion Iden i ica ion o he di e en ially exp essed genes To iden i y di e en ially exp essed genes in he p esence o nod gene induce s six RNA-seq lib a ies we e gene - a ed om CIAT 899 g own in he p esence o apigenin (3.7 μM), sal (300 mM) o unde con ol condi ions. Two independen biological expe imen s we e ca ied ou o each condi ion, being he gene al ea u es o each un shown in Addi ional ile 1. Lib a ies we e sequenced and a ange o 54 o 210 million eads we e ob ained in each condi ion, indica ing ha simila amoun s o da a we e gene a ed independen ly o he g ow h condi ion. Th ee di e en RNA-seq me ics o quali y con ol, such as GC con en , duplica e dis ibu ion, and he dis- ibu ion o espec i e gene ic coo dina es, we e pe - o med (Addi ional ile 1). Besides, be o e all subsequen analysis, a no maliza ion o he quan i a i e da a was pe o med o a oid s a is ical de ia ions due o di e - ences in lib a y and gene ic sizes [22] (Addi ional ile 1). Da a se we e alida ed by qRT-PCR (Table 1). In all cases, posi i e co ela ion deg ees we e ob ained in old- Pé ez-Mon año e al. BMC Genomics (2016) 17:198 Page 2 o 11 change alues o he qRT-PCR and he RNA-seq da a (Addi ional ile 2). R. opici CIAT899genomecon ains6289genesdis- ibu ed among one ch omosome (3672 CDS, GeneBank numbe CP004015.1) and h ee di e en plasmids: pR CIAT899a/pA (212 CDS, GeneBank numbe CP004016.1), pR CIAT899b/pB/symbio ic plasmid (500 CDS, Gene- Bank numbe CP004017.1), and pR CIAT899c/pC (1905 CDS, GeneBank numbe CP004018.1) [17]. Di e en ially exp essed genes in each condi ion we e ob ained using he s a is ical so wa e R. Resul s e ealed 19 di e en ially exp essed genes when he bac e ium was g own in he p esence o apigenin (0.3 % o he genome) and 790 genes when he bac e ium was g own unde sal s ess condi ion (12.5 % o he genome). In he i s case, 15 genes we e up- egula ed (78.9 %) and loca ed in he symbio ic plas- mid, while he o he 4 we e down- egula ed and dis ib- u ed in he o he eplicons: ch omosome (2) and plasmids A (1) and C (3) (Fig. 1a). Wi h espec o he cul u es supplemen ed wi h sal , he majo i y o genes we e down- egula ed (646, 81.6 %); only 144 genes we e o e - exp essed ep esen ing 18.4 % o he di e en ially exp essed genes (Fig. 1b). Replicon dis ibu ion e ealed ha mos o hese genes (723; 95 up- egula ed and 628 down- egula ed) we e loca ed in he ch omosome, some in plasmid B (31; 30 up- egula ed and 1 down- egula ed) and he es in plasmid C (36; 19 up- egula ed and 17 down- egula ed) (Fig. 1b). Conside ing all genes, 17 we e di e en ially exp essed in bo h condi ions, 15 we e up- egula ed and loca ed in he symbio ic plasmid and he o he wo we e down- egula ed and loca ed in he ch omosome (Fig. 2). In addi ion, many o he genes we e di e en ially exp essed only unde saline condi ions. This could be due o a su - i al s a egy when he bac e ium aces an abio ic s ess. Simila esul s ha e been ecen ly epo ed in R. e li CE3 unde saline shock condi ions, when 6.57 % o he genes we e di e en ially exp essed, mos o hem we e loca ed in he ch omosome [16]. Func ions associa ed o esponses o nod gene-inducing molecules CIAT 899 is able o syn hesize NF no only upon induc- ion wi h la onoid bu also unde sal s ess [14]. To e alua e which unc ions a e in insic o each inducing condi ion, a unc ional en ichmen was ca ied ou o assign he s a is ically o e - ep esen ed biological p o- cesses (ac i a ed o ep essed) using da a a ailable a he Unip o da abase (Gene On ology, GO). Resul s a e summa ized in Addi ional ile 3. As expec ed, when Table 1 RNA-seq da a alida ion using qRT-PCR. Fold-change alues we e calcula ed using he ΔΔC me hod and no malized o he e e ence gene RNA 16S o 20 di e en ially exp essed genes. HP: gene ha codes o a hypo he ical p o ein Gene name Locus ag Nucleo ide ange RNA-seq qRT-PCR Apigenin Sal Apigenin Sal a aC1 RTCIAT899_CH06050 1238211_1239707 1.81 6.98 1.60 3.49 psL RTCIAT899_CH07390 1516428_1516799 −1.61 12.47 3.20 7.11 a aC2 RTCIAT899_CH14150 2846615_2847376 −2.56 −8.00 1.97 2.04 y4wE RTCIAT899_PB00575 97717_98829 8.58 12.17 15.35 16.68 y4wF RTCIAT899_PB00570 96476_97504 3.20 7.11 12.05 10.04 nodA2 RTCIAT899_PB01095 192166_192756 10.30 9.81 13.90 17.00 hsnT RTCIAT899_PB01100 192929_194854 7.27 5.94 25.67 12.50 nodF RTCIAT899_PB01105 194950_195231 13.39 13.43 11.90 13.45 nodE RTCIAT899_PB01110 195232_196440 10.37 11.94 16.74 13.78 nodA1 RTCIAT899_PB01300 235667_236257 6.69 12.00 14.18 17.83 nodB RTCIAT899_PB01305 236254_236913 11.4 19.01 8.55 16.30 nodC RTCIAT899_PB01310 236925_238283 6.69 12.00 11.39 15.94 nodS RTCIAT899_PB01315 238201_238917 7.53 12.89 6.03 10.03 nodU RTCIAT899_PB01320 238953_240680 5.39 10.18 15.29 22.86 nodI RTCIAT899_PB01325 240668_241582 6.59 12.48 7.58 11.89 nodJ RTCIAT899_PB01330 241586_242371 4.98 10.15 3.92 15.83 nodH RTCIAT899_PB01340 242956_243705 3.09 7.6 14.80 17.02 HP RTCIAT899_PB01545 281777_282742 4.27 14.37 8.88 18.11 nodM RTCIAT899_PB02710 502220_504046 2.43 5.85 5.69 12.58 HP RTCIAT899_PC04980 1059004_1060056 1.32 −4.22 1.14 −5.38 Pé ez-Mon año e al. BMC Genomics (2016) 17:198 Page 3 o 11 induced wi h apigenin, some o he di e en ially exp essed genes we e in ol ed in nodula ion and oligosaccha ide anspo (Fig. 3a). The same bio- logical p ocesses we e also ac i a ed unde sal s ess. Howe e , he p esence o sal also induced many o he biological p ocesses such as ni ogen ixa ion, chemo- axis, ca bohyd a e me abolism, ansc ip ion, ansla- ion, conjuga ion and ATP biosyn hesis (Fig. 3b). RNA-seq and p o eomics s udies pe o med on o he hizobial s ains unde saline and hea shock condi- ions e ealed ha many o hese p ocesses a e also being al e ed [16, 23, 24]. Apigenin ac i a es he ansc ip ion o nod gene ope ons and he syn hesis o indole-3-ace ic acid The CIAT 899 genome ha bo s i e di e en nod genes and h ee di e en nodA genes in he symbio ic plasmid [17]. A e he nodA1BC and he nodA2FE ope ons and he nodA3 gene o CIAT 899 ac i a ed by apigenin? T ansc ip omic analysis e ealed ha ou se s o genes we e signi ican ly up- egula ed in he symbio ic plasmid unde his condi ion. Two o hem co esponded o he nodA1 and nodA2 ope ons, he hi d, loca ed adjacen o he nodD5 gene, was in ol ed in he syn hesis o he phy oho mone indole-3-ace ic acid (IAA) [21], and he Fig. 1 Dis ibu ion o di e en ially exp essed genes. Pe cen age and loca ion o di e en ially exp essed genes (up- and down- egula ed) in R. opici CIAT 899 induced wi h apigenin (a) o sal (b). Ch: ch omosome, pA: pR CIAT899a, pB: pR CIAT899b, pC: pR CIAT899c. The numbe o di e en ially exp essed genes is indica ed unde each eplicon Pé ez-Mon año e al. BMC Genomics (2016) 17:198 Page 4 o 11 las se o genes coded o p o eins wi h unknown unc- ions (Fig. 4, Addi ional ile 4). The e o e, a leas wo main biological p ocesses we e ac i a ed in CIAT 899 when induced wi h apigenin: he syn hesis o NF (nodA1BCSUIJHPQ1Q2 and nodA2hsnTnodFE genes) and he p oduc ion o IAA (y4wEF genes). O special in e es is he elucida ion o he symbio ic ole o he se o genes wi h unknown unc ion ac i a ed wi h apigenin (RTCIAT899_PB01550 and RTCIAT899_ PB01545). In summa y, hese esul s indica e ha when induced wi h apigenin, CIAT 899 is able o syn hesize NF ha will in- duce oo nodule p imo dia o ma ion, as well as phy o- ho mones ha will a o oo de elopmen . This is consis en wi h a ecen s udy in which an inc ease in he p oduc ion o IAA is epo ed when CIAT 899 is g own in cul u es supplemen ed wi h bo h apigenin and sal [20]. A simila egula ion cascade in ol ing NodD1 and la onoids has been epo ed in Sino hizobium edii NGR234 o he syn hesis o IAA [25]. Common esponses: syn hesis o nodula ion molecules and iden i ica ion o nod boxes As p e iously men ioned in he in oduc ion, he cap- aci y o nodula e unde sal s ess condi ions has been desc ibed o many hizobial s ains [15]. Howe e , he sal - egula ed p oduc ion o NF has been only epo ed o R. opici CIAT 899. A e he CIAT 899 gene ic pa h- ways o syn hesize NF unde sal s ess simila o hose desc ibed o apigenin? The RNA-seq analysis shown in his wo k indica ed ha al hough he pa e n o symbi- o ic gene ac i a ion in he p esence o sal was simila o ha ob ained wi h apigenin (Fig. 4, Addi ional iles 4 and 5), sligh ly highe exp essions o nod genes, wi h an emphasis on he nodA1 ope on, was obse ed when he bac e ium was induced wi h sal (Table 2). Mo eo e , unde sal s ess nodM and a gene ha codes o a la- ine mononucleo ide educ ase (RTCIAT899_PB02705) we e signi ican ly up- egula ed. Al hough he up- egula ion o nodM was no de ec ed in he RNA-seq o CIAT 899 g own in he p esence o apigenin, i was de- ec ed in he qRT-PCR analysis (5.69 old-change) (Table 1). All hese esul s indica e ha he ansc ip ion o he CIAT 899 nod genes seems o espond simila ly o he p esence o ei he apigenin o sal (Table 2). The ques- ion ha emains unclea is whe he he esponses ob- se ed ollow he common NodD ac i a ion pa hway. To answe his ques ion, an in silico analysis o de ec p o- mo e mo i s was conduc ed using consensus sequences Fig. 2 R. opici CIAT 899 esponses o apigenin and sal s ess. Venn diag am showing he o e lapping o di e en ially exp essed genes in he p esence o induce molecules. The a ows indica e he numbe o up- egula ed and down- egula ed genes in each ea men , being he numbe o o e lapping in e sec ions indica i e o genes exp essed in bo h condi ions Fig. 3 Numbe o egula ed genes om each o e - ep esen ed biological p ocess (Gene On hology, GO). Biological p ocesses and numbe o genes a ec ed in bac e ial cul u es supplemen ed wi h apigenin (yellow) ao sal (blue) (b). O e - ep esen ed unc ional ca ego ies we e hose wi h a p alue in hype geome ical es in e io o 0.15 Pé ez-Mon año e al. BMC Genomics (2016) 17:198 Page 5 o 11 o nod boxes (NB) p esen in o he ela ed hizobial s ains. Thus, en po en ial NB we e iden i ied, all sha - ing he AT-N 11 -AT-N 7 -AT-N 2 -A-N 4 -AT-N 2 -ATT-N-T sequence consensus (Fig. 5). Fou o hese NB we e si u- a ed ups eam o he ou di e en se s o genes up- egula ed unde bo h inducing condi ions, namely, he ope ons nodA1BCSUIJHPQ1Q2 (NB1), nodA2hsnT- nodFE (NB2), y4wEF ope on (NB4) and he nodM gene (NB3) (Table 3, Addi ional iles 4 and 5). In e es ingly, wo o he genes p e iously iden i ied as up- egula ed wi h apigenin (RTCIAT899_PB01550 and RTCIAT899_ PB01545) we e also induced wi h sal . The exp ession o one o hese genes (RTCIAT899_PB01545) was alida ed by qRT-PCR in he p esence o bo h apigenin (8.88- old) and sal (18.11- old) (Table 1). This gene exp ession, combined wi h he ac ha bo h genes we e loca ed downs eam NB5, indica e ha hese hypo he ical p o- eins could be playing a ole in he symbio ic p ocess (Table 3, Addi ional iles 4 and 5). Finally, NB7 was lo- ca ed ups eam a gene (RTCIAT899_PB01055) up- egula ed wi h sal , bu no wi h apigenin (Table 3, Add- i ional ile 5). Al oge he , hese esul s sugges ha when CIAT 899 was induced ei he by apigenin o sal , genes ela ed o Fig. 4 Ci cula ep esen a ion o he comple e RNA-seq-based ansc ip omic da a se o apigenin and sal cul u es e sus s anda d cul u es o all eplicons o R. opici CIAT 899. Fold-change alues o consecu i e genes a e ep esen ed by linked peaks. Each peak co esponds o one gene, being o de ed acco ding o hei ela i e posi ion in he eplicon. Yellow peaks: di e en ially exp essed genes in he p esence o apigenin. Blue peaks: di e en ially exp essed genes unde sal s ess. The a ows indica e names and ela ed biological p ocesses o some di e en ially exp essed genes. NB: nod box. aCh omosome (3.8 Mb), bpR CIAT899a (0.22 Mb), cpR CIAT899b (0.55 Mb), dpR CIAT899c (2.08 Mb) Pé ez-Mon año e al. BMC Genomics (2016) 17:198 Page 6 o 11 he syn hesis o NF and IAA we e ac i a ed h ough he ecogni ion o speci ic nod boxes (Table 2). In S. edii NGR234, nine een nod boxes ha e been iden i ied and eigh een o hem we e inducible wi h la onoids ia NodD1. In addi ion, S. edii NGR234 ca ies wo copies o he nodD gene and ou o hese nod boxes a e con- olled by NodD2 [26]. Sal s ess esponse: a pa hway owa ds symbiosis? Ch omosomal RNA-seq analysis showed, unde sal s ess, an up- egula ion o genes whose p oduc s a e im- plied in he o ma ion o ibosomes (bo h 30 and 50S ibosomal p o eins coded by he pl and ps genes, e- spec i ely) and in he u iliza ion o he disaccha ide e- halose ( huAB genes), an osmo ic s abilize [27, 28]. Ins ead, down- egula ion was de ec ed in genes ha en- code p o eins in ol ed in chemo axis (cheXYAWRBYD genes) [29], o ma ion o Flp- ype pili ( ype IVb p o ein sec e ion sys em; adBCD and cpaABCDEF genes) [30], o ABC- ype anspo (se e al genes: e.g. phn genes). In addi ion, a se o genes loca ed in pR CIAT899c and in ol ed in he expo o capsula polysaccha ide ac oss he inne memb ane (genes o he amily bexC/ c B/kpsE) was also down- egula ed (Fig. 4, Addi ional ile 5) [31]. The e o e, he global esponse o CIAT 899 o sal s ess indica es ha he s ain d as ically educes he up- ake o molecules and he ene gy consump ion, educing he ATP-dependen anspo o he ABC- ype pe me- ases. Howe e , unde saline shock condi ions R. e li CE3 o e -exp esses genes ha code o many ABC ans- po e s and inc eases he ansc ip ion le els o genes e- la ed o he biosyn hesis o ehalose [16]. Ne e heless, in CIAT 899 an up- egula ion o genes in ol ed in he deg ada ion o ehalose was de ec ed. The explana ion could be ha his osmoly e ends o accumula e du ing symbiosis, bu i is oxic o plan s. To sol e his p ob- lem, bac e ia would p oduce ehalose-deg ading en- zymes [32]. These obse a ions in CIAT 899, combined wi h he p oduc ion o NF and IAA unde sal s ess, make us o mula e he nex ques ion: could he p omo- ion o he symbiosis ep esen a s a egy o CIAT 899 o ensu e nodula ion unde his abio ic s ess condi ion? To shed ligh on his ques ion we ha e analyzed he changes in biological p ocesses unde saline condi ions. Fi s , we obse ed an enhancemen in ibosomal syn he- sis, which could be ela ed o a ansla ion inc ease due Table 2 Fold-change exp ession alues o he R. opici CIAT 899 nod genes on he RNA-seq analysis. Fold-change alues ollowed by an as e isk (*) a e signi ican ly o e -exp essed Gene name/Locus ag Apigenin Sal nodD1/RTCIAT899_PB01295 1.23 1.04 nodD2/RTCIAT899_PB01070 −1.25 3.01 nodD3/RTCIAT899_PB00640 −1.21 2.1 nodD4/RTCIAT899_PB01560 1.09 1.16 nodD5/RTCIAT899_PB00560 −1.12 −1.09 nodA1/RTCIAT899_PB01300 8.73* 13.66* nodB/RTCIAT899_PB01305 11.4* 19.01* nodC/RTCIAT899_PB01310 6.69* 12* nodS/RTCIAT899_PB01315 7.53* 12.89* nodU/RTCIAT899_PB01320 5.39* 10.18* nodI/RTCIAT899_PB01325 6.59* 12.48* nodJ/RTCIAT899_PB01330 4.98* 10.15* nodH/RTCIAT899_PB01340 3.09 7.6* nodP/RTCIAT899_PB01345 2.78 6.08* nodQ1/RTCIAT899_PB01350 3.25 6.39* nodQ2/RTCIAT899_PB01355 1.28 2.54 nodA2/RTCIAT899_PB01095 10.3* 9.81* hsnT/RTCIAT899_PB01100 7.27* 5.94* nodF/RTCIAT899_PB01105 13.39* 13.43* nodE/RTCIAT899_PB01110 10.37* 11.94* nodA3/RTCIAT899_PB00645 1.15 3.36 nodM/RTCIAT899_PB02710 2.43 5.85* Fig. 5 Alignmen o NB sequences om R. opici CIAT 899. So wa e uzznuc o he EMBOSS package was used o iden i y nod box-like sequences. Hi s we e selec ed based on hei conse a ion o p e iously iden i ied NB sequences. Sea ching o hese consensus sequences was ca ied ou on he whole genome o R. opici CIAT 899. DNA sequences we e aligned using he Clus alW p og am and manipula ed wi h Boxshade a EMBne . Da k and g ay boxes indica e iden ical and simila nucleo ides, espec i ely. P omo e consensus mo i s a e ma ked wi h an as e isk on he consensus line Pé ez-Mon año e al. BMC Genomics (2016) 17:198 Page 7 o 11 o an inc ease in he ansc ip ion o nodula ion genes (a s onge nod gene up- egula ion was de ec ed in sal s ess condi ions). Second, gene al p o ein sec e ion sys- ems and su ace polysaccha ides seem o play an impo an ole in bac e ial i ness unde s essing en i onmen al condi ions, such as high salini y o empe a u e [16, 23, 24]. Howe e , in CIAT 899 we obse ed inhibi ion unde sal s ess. Rema kably, bo h cellula componen s can also be conside ed mic obe- associa ed molecula pa e ns (MAMP), which igge he plan immune sys em [33]. O he hizobial s ains in he p esence o induce la onoids modi y hei su ace molecules o a oid plan immune esponses, he eby p omo ing nodula ion [34]. Finally, when colonizing leg- ume oo s many hizobia o m mic o-colonies o bio- ilms. In some o hese hizobia common nod genes a e equi ed o he de elopmen o hese s uc u es, since he bio ilm ma ix is composed in pa by NF [35, 36]. Thus, in CIAT 899, he syn hesis o NF and he down- egula ion o chemo axis genes de ec ed unde saline condi ions could be ela ed o he o ma ion o a symbi- o ic bio ilm. In summa y (Fig. 6), hese esul s sugges ha in he p esence o sal CIAT 899 could s a egically inc ease he ansc ip ion o nodula ion genes and he syn hesis o NF o inc ease he chances o es ablish sym- biosis e en unde abio ic s essing condi ions. Conclusions Ou wo k shows ha in he p esence o he nodula ion in- ducing molecules apigenin (3.7 μM) and sal (300 mM), R. opici CIAT 899 inc eases he ansc ip ion o he same se o genes (sligh ly highe exp ession upon sal ea men ), whose encoded p o eins a e in ol ed in he syn hesis o symbio ic molecules. Besides, he sal - dependen p oduc ion o hese symbio ic molecules could be a CIAT 899 s a egy o ensu e nodula ion unde sal s ess. Me hods Cul u e condi ions and RNA ex ac ion R. opici CIAT 899 was g own o 72 h a 28 °C on yp- one yeas (TY) medium [37], supplemen ed wi h api- genin 3.7 μM o NaCl 300 mM when necessa y (induc ion o he syn hesis o NF) [14, 20]. To al RNA was isola ed using a High Pu e RNA Isola ion Ki (Roche), acco ding o he manu ac u e ’s ins uc ions. Ve i ica ion o he amoun and quali y o o al RNA samples was ca ied ou using a Nanod op 1000 spec- opho ome e (The mo Scien i ic) and a Qubi 2.0 Fluo ome e (In i ogen). Two independen o al RNA ex ac ions we e ob ained o each condi ion. Quan i a i e e e se ansc ip ion PCR Resul ob ained in he RNA-seq analysis we e alida ed by quan i a i e e e se ansc ip ion PCR (qRT-PCR) o 20 selec ed genes, which ep esen ed di e en ially and non-di e en ially exp essed genes in he p esence o api- genin and sal . To al RNA was isola ed using a High Pu e RNA Isola ion Ki (Roche) and RNAase F ee DNA Se (Qiagen), acco ding o he manu ac u e ’s ins uc- ions. This (DNA- ee) RNA was e e se ansc ibed in o cDNA using a Quan iTec Re e se T ansc ip ion Ki (Qiagen). Quan i a i e PCR was pe o med using a Ligh Cycle 480 (Roche) wi h he ollowing condi ions: Table 3 Up- egula ion o he R. opici CIAT 899 pR CIAT899b genes loca ed downs eam nod boxes. T ansc ip ional ac i a ion (4- old induc ion wi h espec o con ol cul u es) o se e al nod box con olled ope ons was demons a ed by RNA-seq da a in he p esence o bo h induce molecules. HP: gene ha codes o a hypo he ical p o ein NB (Dis ance) Cds numbe /Locus ag/Gene name Pu a i e unc ion Fold changeapigenin Fold changesal 1 (270 pb) cds216 o 227/RTCIAT899_ PB01300 o RTCIAT899_ PB01345/nodABCSUIJHPQ1Q2 NF p oduc ion 8.73 o 1.28 13.66 o 2.54 2 (250 pb) cds177 o 180/RTCIAT899_ PB01095 o RTCIAT899_ PB01110/nodA2hsnTnodFE NF p oduc ion 10.3 o 10.37 9.81 o 11.94 3 (180 pb) cds450 o 449/RTCIAT899_ PB02710 o RTCIAT899_ PB02705/nodM NF p oduc ion 2.43 o 3.18 5.85 o 8.49 4 (680 pb) cds95 o 93/RTCIAT899_PB00575 o RTCIAT899_ PB00565/y4wEF Syn hesis o IAA 8.57 o 1.63 12.17 o 3.27 5 (38 pb) cds263 o 262/RTCIAT899_PB01550 o RTCIAT899_ PB01545/HP Unknown 6.75 o 4.26 28.65 o 14.37 6 (70 pb) cds61 o 62/RTCIAT899_PB00370 o RTCIAT899_ PB00375/HP Unknown 1.80 o 2.11 2.17 o 1.64 7 (191 pb) cds170/RTCIAT899_PB01055/HP Unknown 1.18 5.68 8 (470 pb) cds81/RTCIAT899_PB00495/HP Unknown −1.05 2.58 9 (979 pb) cds173/RTCIAT899_PB01570/nodD2 T ansc ip ional egula ion −1.25 3.01 10 (292 pb) cds89/RTCIAT899_PB00540/HP a Unknown 1.13 −1.44 a The ansc ip ional o ien a ion o his gene is opposi e o he NB10 sequence o ien a ion Pé ez-Mon año e al. BMC Genomics (2016) 17:198 Page 8 o 11 95 °C, 10 min; 95 °C, 30 s; 50 °C, 30 s; 72 °C, 20 s; o y cycles, ollowed by he mel ing cu e p o ile om 60 o 95 °C o e i y he speci ici y o he eac ion. The R. o- pici CIAT 899 16S RNA gene was used as an in e nal con ol o no malize gene exp ession. The old-changes o wo biological samples wi h h ee echnical eplica es o each condi ion we e ob ained using he ΔΔC me hod [38]. Selec ed genes and p ime s a e lis ed in Addi ional ile 2. RNA sequencing Ribosomal RNA was deple ed using a MICROB Exp ess Bac e ial mRNA Pu i ica ion ki (Ambion), ollowing he manu ac u e ’s p o ocol. In eg i y and quali y o he ibosomal deple ed RNA was checked wi h Agilen Bioanalyze 2100 (Agilen Technologies). RNA sequen- cing was ca ied ou by Sis emas Genómicos (h ps:// www.sis emasgenomicos.com/web_sg/) wi h he Nex Gene a ion Sequence (NGS) pla o m Illumina using he Illumina HiSeq 2000 sequencing ins umen (Illu- mina). Ribosomal-deple ed samples we e used o gene a e whole ansc ip ome lib a ies ollowing he manu ac u e 's ecommenda ions o sequencing on his NGS pla o m. Ampli ied cDNA quali y was ana- lyzed by he Bioanalyze 2100 DNA 1000 ki (Agilen Technologies) and quan i ied using he Qubi 2.0 Fluo ome e (In i ogen). Mapping o he RNA-seq da a The ini ial whole ansc ip ome pai ed-end eads ob- ained om sequencing we e mapped agains he la es e sion o he R. opici CIAT 899 genome (h p:// www.ncbi.nlm.nih.go /genome/? e m=Rhizobium_ opi- ci_CIAT_899) using he he Li e Technologies mapping algo i hm e sion 1.3 (h p://www.li e echnologies.- com/). Low-quali y eads we e elimina ed using Pic- a d Tools so wa e e sion 1.83, emaining only high quali y eads. Assessmen o di e en ially exp essed genes Gene p edic ion was es ima ed using he cu links me hod [39] and he exp ession le els we e calcula ed using he h seq so wa e, e sion 0.5.4p3 [40]. This me hod elimina es mul imapped eads, conside ing only unique eads o he gene exp ession es ima ion. The edge me hod e sion 3.2.4 was applied o di e en ial exp ession analysis among condi ions [41]. This me hod uses a Poisson model o es ima e he a iance o he RNA-seq da a o di e en ial exp essions, and elies on di e en no malized p ocesses based on dep h global samples, CG composi ion and leng h o genes. Di e en- ially exp essed genes we e es ablished in hose genes wi h a old-change lowe o highe han−4 o 4, espec - i ely, wi h a p alue adjus o 0.7. Fig. 6 Model o he Rhizobium opici CIAT 899 ansc ip omic esponse unde sal s ess. RNA-seq s udies indica e ha in he p esence o high concen a ion o sal CIAT 899 inc ease he ansc ip ion o nodula ion genes and he syn hesis o NF. T ansc ip ional ac i a ion (+) o inhibi ion (−) o o he biological p ocesses associa ed wi h he p esence o sal suppo s he hypo hesis ha sal -dependen p oduc ion o NF could be a special R. opici CIAT 899 s a egy o ensu e nodula ion unde sal s ess condi ions. HP: gene ha codes o a hypo he ical p o ein Pé ez-Mon año e al. BMC Genomics (2016) 17:198 Page 9 o 11