ORIGINAL PAPER

Differential expression of four soybean bZIP genesduring Phakopsora pachyrhizi infection

Murilo S. Alves1 & Zamira G. Soares1 & Pedro M. P. Vidigal2 & Everaldo G. Barros3 &

Adriana M. P. Poddanosqui4 & Luciano N. Aoyagi4 & Ricardo V. Abdelnoor4 &

Francismar C. Marcelino-Guimarães4 & Luciano G. Fietto1

Received: 22 September 2014 /Revised: 7 May 2015 /Accepted: 18 May 2015# Springer-Verlag Berlin Heidelberg 2015

Abstract Asian soybean rust (ASR), caused by the obligatebiotrophic fungus Phakopsora pachyrhizi, is one of most im-portant diseases in the soybean (Glycine max (L.) Merr.) ag-ribusiness. The identification and characterization of genesrelated to plant defense responses to fungal infection are es-sential to develop ASR-resistant plants. In this work, we de-scribe four soybean genes, GmbZIP62, GmbZIP105,GmbZIPE1, and GmbZIPE2, which encode transcription fac-tors containing a basic leucine zipper (bZIP) domain from twodivergent classes, and that are responsive to P. pachyrhizi in-fection. Molecular phylogenetic analyses demonstrated thatthese genes encode proteins similar to bZIP factors responsiveto pathogens. Yeast transactivation assays showed that onlyGmbZIP62 has strong transactivation activity in yeast. In ad-dition, three of the bZIP transcription factors analyzed werealso differentially expressed by plant defense hormones, andall were differentially expressed by fungal attack, indicatingthat these proteins might participate in response to ASR infec-tion. The results suggested that these bZIP proteins are part ofthe plant defense response to P. pachyrhizi infection, by

regulating the gene expression related to ASR infection re-sponses. These bZIP genes are potential targets to obtainnew soybean genotypes resistant to ASR.

Keywords Biotic stress . Signal transduction . Plant defense .

Asian soybean rust . Transcriptional control

AbbreviationsGmbZIP Glycine max basic/leucine zipper protein

Introduction

Phakopsora pachyrhizi is an obligate biotrophic pathogenicfungus that causes the disease commonly known as Asiansoybean rust (ASR). P. pachyrhizi primarily colonizes leaftissue, and under favorable environmental conditions, the in-fection can result in losses of up to 80 % of crop production(Patil et al. 1997). Six loci in soybean (Glycine max) havebeen identified that confer immunity [Rpp1 (Cheng andChan 1968)] or resistance [Rpp2, (Hidayat and Somaatmadja1977), Rpp3 (Bromfield and Hartwig 1980), Rpp4 (Hartwig1986), Rpp5 (Garcia et al. 2008), and Rpp6 (Li et al. 2012)] tothe fungus, through the production of reddish-brown lesions,that are easily visible on both sides of the leaf (Bromfield andHartwig 1980). Such lesions have little or no sporulation andthus reduce the capacity for dissemination of the disease(Monteros et al. 2007; Garcia et al. 2008). However, none ofthese genes has been cloned to date (Bromfield 1984;Bromfield and Hartwig 1980; Hartwig and Bromfield 1983;Hyten et al. 2007; Morceli et al. 2008).

The Rpp2 and Rpp4 genotypes are stable in Brazil andhave been found in Asia for more than 20 years (Hartmanet al. 2005); however, resistance to P. pachyrhizi has always

Electronic supplementary material The online version of this article(doi:10.1007/s10142-015-0445-0) contains supplementary material,which is available to authorized users.

* Luciano G. Fiettolgfietto@ufv.br

1 Departamento de Bioquímica e Biologia Molecular, UniversidadeFederal de Viçosa, Viçosa, Minas Gerais 36570-000, Brazil

2 Núcleo de Análise de Biomoléculas, NuBioMol, UniversidadeFederal de Viçosa, Viçosa, Minas Gerais 36570-000, Brazil

3 Universidade Católica de Brasília, 70790-160 Brasília, DistritoFederal, Brazil

4 Embrapa Soja, Londrina, Paraná 86001-970, Brazil

Funct Integr GenomicsDOI 10.1007/s10142-015-0445-0

been overcome in nature because of the high genetic diversityof the fungus (Bromfield 1984). This scenario is termed theBboom and bust^ syndrome in the literature, defined as thepathogen overcoming the resistant phenotype of the hostplant.

The destructive potential of P. pachyrhizi, along with itsrecent introduction in major soybean-producing countries inthe Western hemisphere, has generated a great deal of interestin understanding the molecular interactions of P. pachyrhizi insoybean plants and other non-host plants (Hoefle et al. 2009;Goellner et al. 2010), with the aim of identifying novel genesfor resistance to the fungi, and to obtain varieties with durableresistance.

Plants withstand pathogen attacks by activating a variety ofdefense mechanisms, including the hypersensitive response,the induction of genes that encode pathogen-related proteins,the production of antimicrobial compounds called phyto-alexins, the generation of reactive oxygen species, and en-hanced cell walls (Reymond and Farmer 1998; Kim et al.2012).

These complex response mechanisms are finely regulatedby many genes that encode regulatory proteins. Typical exam-ples of regulatory proteins are transcription factors (Huanget al. 2005; Yamaguchi-Shinozaki and Shinozaki 2005,2006; Nakashima and Yamaguchi-Shinozaki 2006;Umezawa et al. 2006). Transcription factors respond to envi-ronmental stimuli through a signaling cascade that can in-crease their concentration or activity. Consequently, these fac-tors regulate gene expression by interacting with target genepromoters. The family of factors containing a basic leucinezipper domain (bZIP) is among the transcription factors moststudied in plants. The bZIP transcription factors have beenextensively investigated for their role in the regulation ofgenes in seed storage (Chern et al. 1996; Albani et al. 1997),photomorphogenesis (Oyama et al. 1997), leaf development(Yin et al. 1997), flower development (Chuang et al. 1999),response to abscisic acid (Nakagawa et al. 1996), and biosyn-thesis of gibberellins (Fukazawa et al. 2000).

However, of the various functions of bZIP proteins, plantdefense against pathogens is one of the least studied (Singhet al. 2002; Jakoby et al. 2002; Alves et al. 2013). A classicexample of bZIP proteins are TGA factors (TGACGTCA cis-element-binding proteins) (Singh et al. 2002), which act asregulators of signaling by salicylic acid (SA) during the de-fense response in plants. Some TGA proteins interact with anankyrin repeat domain protein called non-expresser ofpathogen-related genes (NPR1), which is an important com-ponent of SA signaling in defense responses (Singh et al.2002; Jakoby et al. 2002; Alves et al. 2013).

The first bZIP studied in plants with high agronomic valuewas the G/HBF-1 soybean protein. This factor binds to cis-elements responsive to pathogen elicitors called G-box and H-box motifs. These cis-elements were found in the promoter of

the chs15 gene, which encodes a chalcone synthase protein, akey enzyme of flavonoid biosynthesis in plants (Dröge-Laseret al. 1997), which is involved in the synthesis of diterpenoidsand flavonoids in response to pathogens (Dröge-Laser et al.1997).

The expression of G/HBF-1 is not altered during the induc-tion of transcription of chs15, and G/HBF-1 is rapidly phos-phorylated in soybean cells elicited with either glutathione orinoculation with avirulent Pseudomonas syringae pv.glycinea, indicating post-translational control of G/HBF-1(Dröge-Laser et al. 1997). In grapes, a bZIP member calledVvbZIP23 has been identified as a key regulator of stressresponses, and its expression is strongly induced by drought,salinity, cold, abscisic acid, ethylene, jasmonic acid (JA), andSA (Tak and Mhatre 2012). A translation factor named PPI1,which has a different structure to other previously studiedbZIPs, was characterized in peppers (Lee et al. 2002). Its ex-pression is preferentially induced by infection with pathogensand is not altered by exposure to SA and JA (Lee et al. 2002).

There are few examples of the molecular characterizationof transcription factors that are responsive to rust fungi. Inwheat, an NAC (an acronym for NAM, ATAF, CUC) familytranscription factor, TaNAC4, which is involved in defenseagainst infection by stripe rust fungus and abiotic stresseswas identified (Xia et al. 2010). In the bZIP family, a trans-acting factor, TabZIP1, is involved in stress tolerance anddefense response against the same stripe rust fungus (Zhanget al. 2009). These studies demonstrated the involvement oftranscriptional regulators in plant defense against infection bythe fungus that causes rust.

In this work, we identified and characterized four genesthat encode bZIP family transcription factors in soybean.The expression profiles of these factors were determined indifferent tissues, in addition to their transactivation activity.Finally, we examined the possible roles of these regulatorsof defense pathways against pathogens in plants using phylo-genetic analysis and analysis of expression in response totreatment with SA and methyl jasmonate (MeJA) hormones,and the infection by P. pachyrhizi.

Materials and methods

Analysis of subtractive libraries of soybean leaves infectedwith ASR disease from the GENOSOJA database

Using the GENOSOJA consortium database (http://bioinfo.cnpso.embrapa.br/genosoja/; Benko-Iseppon et al. 2012), wescreened for genes that possibly encode bZIP transcriptionfactors that are differentially expressed during ASR. Datafrom Illumina/Solexa sequencing of subtractive libraries con-taining clones from the subtraction MOCK plants vs. inocu-lated plants of soybean cultivar PI561356 (unpublished data),

Funct Integr Genomics

separated by different P. pachyrhizi inoculation times (12, 24,48, and 192 h after infection (hai)), were used to select bZIPgenes that are differentially expressed in the response to fun-gus infection, for further functional studies.

Cloning the GmbZIP62, GmbZIP105, GmbZIPE1,and GmbZIPE2 genes into expression vectors

To clone the complete sequences of the target bZIPs into theyeast expression vector pDEST32, bZIP genes were amplifiedusing specific primers (Table 1) from complementary DNA(cDNA) extracted from susceptible soybean leaves(EMBRAPA48). The resulting amplification products werepurified using the Wizard® SV Gel and PCR Clean-UpSystem kit (Promega) and inserted by recombination into thepDONR201 entry vector using the BP Clonase Enzyme(Invitrogen). The clones obtained, pDONR201-GmbZIP62,pDONR201-GmbZIP105, pDONR201-GmbZIPE1, andpDONR201-GmbZIPE2, were used to transfer the respectivecDNAs into the expression vector pDEST32, by recombina-tion using the LR Clonase Enzyme (Invitrogen).

Yeast transactivation assay

The yeast strain AH109 (MATa, trp1901, leu23, 112, ura3-52,his3200, gal4Δ, LYS2:: GAL1UASGAL1TATAHIS3, MEL1GAL2UASGALTATA :: MELUASMEL1TATAlacZ) wastransformed with pDEST32-GmbZIP62, pDEST32-GmbZIP105, pDEST32-GmbZIPE1, or pDEST32-GmbZIPE2 and grown in SD medium (synthetic dropout) inthe absence of histidine and leucine, but supplemented with20 mM 3-aminotriazole for 3 days at 28 °C. After 3 days, theβ-galactosidase activity in the strains was measured as de-scribed by Amberg et al. (2005).

bZIP gene expression profile in soybean tissuesand in response to phytohormones

bZIP gene expression was analyzed in different tissues of thesoybean variety CD 206 at different stages of seed develop-ment. RNAwas extracted from the leaf, stem, and root in thevegetative phase using the TRIzol method (Invitrogen). Forthe seeds, fresh mass was extracted following Freitas et al.(2007) at the first, second, third, and fifth seed stages for seedsup to 75 mg and from mature seeds of more than 450 mg.

To evaluate the expression of genes in response to the ex-ogenous application of phytohormones, the plants were ini-tially germinated, and treated with 100 mM MeJA or 1 mMSAwhen they reached the V3 stage, for 0, 1, 2, 6, 12, and 24 h.As a control, plants were treated with water. The plant materialwas collected after treatment and immediately frozen in liquidnitrogen and stored at −80 °C until processing.

Infection experiments

To assess the expression of bZIP genes during infection by thefungus P. pachyrhizi, resistant (PI561356) and susceptible(EMBRAPA48) plants were infected with soybean rust.Soybean samples were grown and maintained in a greenhouseunder controlled conditions of temperature (22±1 °C) andhumidity (>60 %) and with a photoperiod of 12 h throughthe V3 stage of development (30 days). The experimentfollowed a completely randomized design with three biologi-cal repetitions, each one containing three plants, and treat-ments in a factorial structure with two factors, sampling time(12, 24, 48, and 192 hai), and genotypes (PI561356 andEmbrapa 48). The infection was induced by spraying a sporesuspension (3×105 spores/mL). As control treatments, plantsgrown in the same conditions were subjected to false inocula-tion (MOCK) for each respective time of infection. After 12,24, 48, and 192 h of inoculation, the leaves were collected,immediately frozen in liquid nitrogen, and then stored at−80 °C until RNA extraction.

cDNA Synthesis and Real-time PCR Analysis

Total RNA was extracted from tissues and cells with TRIzol(Invitrogen) and treated with 3 units of RNase-free DNase(Promega). Purified RNA was quantified spectrophotometri-cally (Thermo Fisher Scientific EVO 60) and analyzed on adenaturing 1.5 % (w/v) agarose gel. cDNA synthesis was per-formed with 4 μg of total RNA, oligo(dT), and Moloney mu-rine leukemia virus reverse transcriptase (Invitrogen), accord-ing to the manufacturer’s protocol. All of the procedures forreal-time PCR, including tests, validations, and experiments,were conducted according to the recommendations of AppliedBiosystems. Real-time PCRs were performed on an ABI7500instrument (Applied Biosystems), using specific primers(Table 2), cDNAs from the described procedure, andSYBR® Green PCR Master Mix (Applied Biosystems). Therelative quantification of each gene was determined in indi-vidual tubes. Initially, an assay was performed to determinethe efficiency of the reaction. The efficiency test was per-formed on serial dilutions of 100, 10−1, 10−2, and 10−3 ofcDNA. To verify the efficiency of the reaction, we calculatedthe R2 of the linear regression equation obtained for each gene.After 40 cycles of amplification, all samples were subjected tothermal denaturation to determine their dissociation curve(melting) and to verify the specificity of the amplifications.Samples were heated with an increase of 1 °C for 30 s from60 °C up to a limit of 94 °C. The amplification reactions wereperformed under the following conditions: 10 min at 95 °C;40 cycles of 94 °C for 15 s, and 60 °C for 1 min. For quanti-fication of gene expression in soybean cells and seedlings, weused actin as the endogenous control gene to normalize thereal-time PCR data. Gene expression in soybean tissues was

Funct Integr Genomics

quantified using the comparative Ct (cycle threshold) method:2−ΔCt comparing the Ct values between actin and bZIP genesin each tissue. The differences in the means of the relativeexpression of the genes analyzed were evaluated using the ttest at the 5 % level of probability using the software ASSISTAT (www.assistat.com/; 2013 version). Gene expressionand differences in the means of the relative expression of thebZIP genes in hormone and infection assays were analyzedusing the REST2009 software (https://www.qiagen.com/;Pfaffl 2001; Pfaffl et al. 2002), according to authors’instructions, by comparing the Ct values between untreatedand treated samples in each time point and genotype used.

Statistical analysis

All experiments were carried out with three biological repli-cates and three technical replicates. Quantitative data of yeasttransactivation assays and transcriptional profiling of bZIPgenes in soybean tissues were analyzed with the ASSISTATsoftware package (www.assistat.com/), according to authors’instructions. Quantitative data of qPCR from hormone andinfection assays were analyzed with the REST2009 software(https://www.qiagen.com/; Pfaffl 2001; Pfaffl et al. 2002),according to the authors’ instructions.

Results

Selection of GmbZIP62, GmbZIP105, GmbZIPE1,and GmbZIPE2 for functional studies of soybean bZIPsproteins

Based on the phylogenetic analyses, the bZIP protein se-quences of Glycine max were grouped according to the pro-posed model for Arabidopsis bZIP proteins, which comprisesthe following 10 classes: A, B, C, D, E, F, G, H, I, and S(Jakoby et al. 2002) (Online Resource 1). This model is usefulfor the general classification of bZIP proteins of G. max be-cause it is based on conserved domains (Alves et al. 2013). Toidentify bZIP proteins responsive to pathogens and to

determine the bZIP transcription factors differentiallyexpressed in response to P. pachyrhizi, we used data fromsubtractive libraries containing clones of inoculated plantsvs. MOCK plants subtraction (Online Resource 1). We select-ed four distinct soybean genes of the bZIP transcription factorsfamily that were differentially expressed duringASR infectionin resistant soybean cultivar PI561356 plants. The selectedproteins GmbZIP62 (ABI34659) and GmbZIP105(NP_001237027) were grouped in class C. The proteinsGmbZ IPE1 (XP_003 5433 12 ) a n d GmbZ IPE2(XP_003525005) were grouped in class E.

Transactivation activity of GmbZIP62, GmbZIP105,GmbZIPE1, and GmbZIPE2

When expressed in the S. cerevisiae strain AH109, bZIP fu-sions to the DNA-binding domain of GAL4 (BD-GmbZIP62,BD-GmbZIP105, BD-GmbZIPE1, BD-GmbZIPE2) showeddifferent transcriptional activation properties, as shown inFig. 1. The fusion BD-GmbZIP62 was able to activate theexpression of the reporter genes HIS3 and LacZ strongly,while the fusion BD-GmbZIPE1 was only able to activatethe expression of the HIS3 gene weakly. BD-GmbZIP105and BD-GmbZIPE2 were unable to induce the expression ofreporter genes in any of the tests (Fig. 1).

Transcriptional profile of bZIP genes in soybean tissues

The qPCR experiments demonstrated that the selected bZIPgenes showed different tissue expression patterns (Fig. 2).GmbZIP62 and GmbZIP105 were expressed predominantly inthe root, while GmbZIPE1 and GmbZIPE2 showed higher ex-pression levels in leaves and seeds at the mature stage (Fig. 2).The pattern of actin expression was considered stable in alltissues tested, at a probability level of 5 % (data not shown).

Induction of bZIP genes by defense phytohormones

The expressions of the bZIP genes were altered by treatmentwith hormones involved in the regulation of plant molecular

Table 1 Specific primers for amplification and cloning of bZIP genes into expression vectors

Oligonucleotide Sequence (5′-3′) Entry vector

Fw-GmbZIP62 AAAAAGCAGGCTTCACAATGACTTCCTTTTCAGCTTG pDONR201

GmbZIP62-Rv AGAAAGCTGGGTCGCGGCCGCAAGCTTCTAGTCGAGTGGCCAAATAG pDONR201

Fw-GmbZIP105 AAAAAGCAGGCTTCACAATGAATGAAATTCAAATACT pDONR201

GmbZIP105-Rv AGAAAGCTGGGTCGCGGCCGCAAGCTTTCAATTAGAAGGAGTTCCAC pDONR201

Fw-GmbZIPE1 AAAAAGCAGGCTTCACAATGACGCAATTACCTCCAAA pDONR201

GmbZIPE1-Rv AGAAAGCTGGGTCGCGGCCGCAAGCTTTCACGTTACATCCCAAACCT pDONR201

Fw-GmbZIPE2 AAAAAGCAGGCTTCACAATGGCAAATTCAAAGGGCTC pDONR201

GmbZIPE2-Rv AGAAAGCTGGGTCGCGGCCGCAAGCTTCTAAATTCTGAGAGCAGGGT pDONR201

Funct Integr Genomics

defenses (Fig. 3; Online Resource 3). After treatment with SA,a hormone that regulates defenses against biotrophic patho-gens, two bZIP genes showed increased transcription

(GmbZIP62 and GmbZIP105), and the peak was achieved2 h after application (Fig. 3; Online Resource 3). The expres-sion of GmbZIPE1 showed no changes following treatmentwith SA, while GmbZIPE2 expression decreased 1 h aftertreatment (Fig. 3; Online Resource 3). After treatment withMeJA, a compound analogous to the hormone JA that regu-lates the defenses against necrotrophic pathogens and herbi-vores, only GmbZIP105 showed increased expression 6 h af-ter treatment, while GmbZIP62 and GmbZIPE2 showed de-creased expression (Fig. 3; Online Resource 3). Again,GmbZIPE1 displayed no change in expression followingtreatment. These data show the responsiveness of these bZIPgenes to defense response hormones induced by pathogenattack. The modification of the gene expression of defensemarker genes (GmPR1, GmPR4, and GmNAC6), which areresponsive to the hormones used, showed that the treatmentswere effective in activating molecular pathways in response tobiotic stresses in the analyzed plants (Fig. 3; Online Resource3).

Gene expression profile in susceptible and resistantsoybeans during the course of infection by the fungusP. pachyrhizi

An qPCR assay was used to determine the expression profilesofGmbZIP62,GmbZIP105,GmbZIPE1, andGmbZIPE2 dur-ing infection by the fungus P. pachyrhizi in susceptible andresistant soybean genotypes, EMBRAPA48 and PI561356,respectively. The resistant genotype contains the resistancegene allele mapped near Rpp1 (Kim et al. 2012). The sampleswere collected across the infection cycle, and the success ofthe infection was confirmed by the occurrence of the tan and

Table 2 Specific primers forqPCR Oligonucleotide Sequence (5′-3′) Target gene Efficiency (%)

RT Fw-GmbZIP62 CTTACCGGATGCGATGACTT GmbZIP62 99

RT GmbZIP62-Rv TCAATGGTTGCAGTGATGGT GmbZIP62 99

RT Fw-GmbZIP105 GTTGGGAGGCATTTCTTCAA GmbZIP105 99

RT GmbZIP105-Rv CCAGGCTAGCCACTCTATGC GmbZIP105 99

RT Fw-GmbZIPE1 GACCCCAAGAGAGTCAAAAGAATC GmbZIPE1 99

RT GmbZIPE1-Rv GCTCTGATATGTATTGCAGCTTCCT GmbZIPE1 99

RT Fw-GmbZIPE2 CAACCACAGACACAGCCACAA GmbZIPE2 99

RT GmbZIPE2-Rv GCGAACTGAGATTCGAGGTCTCT GmbZIPE2 99

RT Fw-GmPR1 AACTATGCTCCCCGGCAACTATATTG GmPR1 99

RT GmPR1-Rv TCTGAAGTGGCTTTACATCGAAACAA GmPR1 99

RT Fw-GmPR4 TGCGGGTGACAAATACAGGAA GmPR4 99

RT GmPR4-Rv TGCTGCACTGATCTACGATTCTC GmPR4 99

RT Fw-GmNAC6 CCAACAAAAGCACTTGTGGCA GmNAC6 99

RT GmNAC6-Rv GGACTATTCAACTGAGCCCAAAAG GmNAC6 99

RT Fw-GmActin GAGACATCCGAGACCAGCTC Actin 99

RT GmActin-Rv AATGCCTGATGCTTCCATTC Actin 99

Fig. 1 Transactivation activity of bZIP proteins in yeast. a β-Galactosidase activity was determined in yeast cells transformed withthe constructs BD-GmbZIP62, BD-GmbZIP105, BD-GmbZIPE1, andBD-GmbZIPE2 and grown overnight, by permeabilization of cell mem-branes using chloroform and 0.1 % SDS. The thin bars indicate thestandard deviation of three separate experiments. Means indicated by anasterisk differ statistically from the control experiment (AH109 trans-formed with empty vector pDEST32) at the 5 % significance level by at test. b BD-GmbZIP62, BD-GmbZIP105, BD-GmbZIPE1, and BD-GmbZIPE2 were introduced separately in the yeast strain AH109. Thecultures were grown overnight and diluted as indicated in the figure andgrown for 3 days in medium without histidine, but supplemented with20 mM 3-aminotriazole (3AT)

Funct Integr Genomics

reddish-brown lesions, specific for compatible and incompat-ible interactions between soybean and P. pachyrhizi, respec-tively. The results are presented in Fig. 4, and in OnlineResource 3. In the interaction between the rust fungus andsusceptible host plants, the expression of GmbZIP105 wasinduced strongly in the early stages of infection (12 hai) andshowed induced expression at all time points after inoculation.GmbZIPE2 was induced in 24 hai and 48 hai, whileGmbZIPE1 was induced only at 48 hai. Interestingly, the ex-pression of GmbZIP62 was repressed strongly at 12 hai, sim-ilarly to PR1, an SA marker gene (Fig. 4; Online Resource 3).Expression peaked at 48 hai for all the bZIP genes, after whichthe levels of transcripts of GmbZIP62 and GmbZIPE1returned to the baseline level at 192 hai, while the expressionofGmbZIPE2 was repressed (Fig. 4; Online Resource 3). Theinduction of the expression of marker genes in response topathogens (GmPR1, GmPR4, and GmNAC6) showed thatthe infection was established in the analyzed plants (Fig. 4;Online Resource 3).

In the incompatible interaction between the host plant andthe fungus, and unlike the bZIP gene expression observed insusceptible plants, the expression ofGmbZIP105was inducedlater after inoculation with the fungus in resistant plants. PeakGmbZIP105 expression occurred at 48 hai, whereasGmbZIPE1 expression was induced only at 192 hai in resis-tant plants (Fig. 4; Online Resource 3). GmbZIP62 andGmbZIPE2 gene expressions were repressed strongly at al-most all time points after inoculation (Fig. 4; OnlineResource 3). Induction of the expression of marker genes in

response to pathogens (GmPR1 and GmNAC6) occurred onlyat 48 hai in this strain (Fig. 4; Online Resource 3).

Discussion

In plants, bZIP factors are master regulators of gene expres-sion in response to abiotic stresses, during seed maturation,and in flower development processes (Singh et al. 2002).Deep analysis of databases of subtractive libraries from soy-bean plants infected with ASR, coupled with phylogeneticanalyses of the bZIP family proteins in soybean led us to selectfour transcription factors that were apparently differentiallyexpressed in response to infection by P. pachyrhizi (Fig. S1.See Online Resource 1). GmbZIP62 , GmbZIP105,GmbZIPE1, and GmbZIPE2 were differentially expressedduring infection in both susceptible plants and resistant plants(Fig. 4). In addition, three of the genes analyzed (GmbZIP62,GmbZIP105, and GmbZIPE2) were differentially expressedafter treatment with hormonal mediators of defense responses(Fig. 3).

Only GmbZIP62 was able to activate transcription of thereporter genes LacZ and HIS3 strongly in yeast, reflecting thepresence of an activation domain that is highly conserved ineukaryotes (Fig. 1). GmbZIPE1 was able to weakly activatetranscription of the reporter gene HIS3, which showed thatthis protein was capable of recruiting the transcriptional ma-chinery of yeast cells. Many bZIP transcription factors use theinteraction with its leucine zipper domain to regulate the

Fig. 2 Expression profiles ofGmbZIP62, GmbZIP105,GmbZIPE1, and GmbZIPE2genes in different soybean tissues.The relative expression of thetarget genes were measured byreal-time RT-PCR in differentsoybean tissues. The expressionwas calculated using the 2−ΔCt

method, using actin as thecalibrator and normalizer. ThecDNAs used were obtained fromthree biological replicates (thinbars indicate the standarddeviation of biological replicates,n=3). R root, S stem, L leaf, S1first-stage seed, S2 second-stageseed, S3 third-stage seed, S5 seedfifth stage, SM mature seed

Funct Integr Genomics

Fig. 3 Expression analysis ofGmbZIP62, GmbZIP105,GmbZIPE1, and GmbZIPE2 inresponse to treatment with SAand MeJA in soybean leaves. Theexpression levels were calculatedusing the REST2009 software.Actin was used as the endogenouscontrol gene, and water as thecontrol treatment. The cDNAsused were obtained from threebiological replicates. Meansindicated by an asterisk differsignificantly from the controltreatment at the 1 % significancelevel

Funct Integr Genomics

affinity and specificity of binding to target DNA, a featureextensively studied in Arabidopsis bZIP proteins (Jakoby

et al. 2002). An example of the functional importance ofhomo- and heterodimerization of bZIP protein domains has

Fig. 4 Expression analysis ofGmbZIP62, GmbZIP105,GmbZIPE1, and GmbZIPE2 insusceptible and resistant soybeanplants during the course ofinfection by the fungusP. pachyrhizi. The expressionlevels were calculated using theREST2009 software. Actin wasused as the endogenous controlgene, and false inoculation(MOCK) as a control treatmentfor each time of infection. ThecDNAs used were obtained fromthree biological replicates. Meansindicated by an asterisk differsignificantly from the controltreatment at the 1 % significancelevel

Funct Integr Genomics

been shown in response to a severe shortage of nutrients inArabidopsis. In this signaling pathway, two bZIP proteinsfrom Arabidopsis, bZIP53 and bZIP1, are activated byheterodimerization with other bZIP family members. This in-teraction causes a change in the transcriptional activity of fac-tors, affecting their affinity for ACTCAT or ACGT-like cis-elements, which causes a reprogramming of primary metabo-lism in response to stress from low energy availability(Dietrich et al. 2011).

Several genes encoding bZIP transcription factors in plantsshow constitutive expression in roots, but little or no expres-sion in the stem and leaves (Schindler et al. 1992; Miao et al.1994; Katagiri et al. 1989). In this study, we demonstrated thatthe expression levels of GmbZIP62 and GmbZIP105 werehigher in roots and comparatively lower in the other tissuesanalyzed (Fig. 2). By contrast, GmbZIPE1 and GmbZIPE2were predominantly expressed in mature seeds, and onlyGmbZIPE1 showed expression in leaves (Fig. 2).

Three of the bZIP genes analyzed (GmbZIP62 ,GmbZIP105, and GmbZIPE2) were responsive to treatmentwith SA and MeJA (Fig. 3), hormonal mediators of defenseresponses against pathogens in plants. After SA treatment, thebZIP genes showed expression peaks at different times.GmbZIP62 and GmbZIP105 expression peaked 2 h after ap-plication, whereasGmbZIPE2was repressed at 1 h after treat-ment. GmbZIP105 transcription increased at 24 h after SAtreatment, showing a biphasic induction during SA treatment(Fig. 3). Treatment with MeJA caused GmbZIP105 expres-sion to peak at 6 and 12 h after application, and decrease at24 h after treatment. Meanwhile, GmbZIP62 and GmbZIPE2transcription decreased, showing a repression response to JA(Fig. 3).

Schneider et al. (2011) demonstrated that inoculation ofsoybean with P. pachyrhizi leads to a biphasic response, char-acterized by an explosion of differential expression of

responsive genes initiated within 12 h in genotypes containingthe resistance gene Rpp3. A silent period occurred at approx-imately 24–48 hai, when P. pachyrhizi continued to develop,but did not elicit host responses, followed by a second phaseof intense alteration in gene expression, which began approx-imately 72 hai (Schneider et al. 2011). The first major changein gene expression can be correlated with the formation ofappressoria and penetration of epidermal cells. The secondwave of gene expression occurs at the beginning of the for-mation of haustoria, and these processes occur in both resis-tant and susceptible plants that have compatible and incom-patible interactions, respectively. The proliferation ofhaustoria coincides with the inhibition of growth ofP. pachyrhizi in an incompatible interaction in resistant plantsand the beginning of the rapid growth of fungal hyphae incompatible interactions in susceptible plants (Schneider et al.2011). The temporal relationship between the growth ofP. pachyrhizi and the responses of the host plant are an impor-tant context in which it is possible to detail the networks ofmolecular interactions during infection (Schneider et al.2011).

Increased expression of GmbZIP105 at the initial timepoints (12 hai) of the compatible interaction between the fun-gus P. pachyrhizi and susceptible plants suggested activationof JA signaling in response to penetration of epidermal leafcells by the fungus appressorium, causing an induction ofGmbZIP105 expression, a JA-responsive bZIP gene (Figs. 3,4 and 5; Online resource 3). This process has been reported forother bZIPs in response to rust fungi (Zhang et al. 2009).Meanwhile, the JA signaling activation at the initial timepoints culminated in the suppression of GmbZIP62, as wellas GmPR1, an SA signaling marker gene, suggesting a JAantagonistic process at the initial time points of infection(Figs. 4 and 5; Online resource 3). During the hyphae forma-tion stage of infection, at 24 and 48 hai approximately, the

Fig. 5 Proposed model for the infection process caused by P. pachyrhizi,and the influence on bZIP genes expression. a Differential expression ofbZIP genes during the appressorium formation, which culminates in theactivation of the JA signaling pathway, in response to epidermal cellspenetration, causing only the activation of GmbZIP105, a JA responsivebZIP gene. b The SA signaling pathway is activated by recognition ofbiotrophic fungus elicitors, and SA responsive bZIP genes are activated(GmbZIP62 and GmbZIP105), concomitantly with activation of bZIP

genes non-responsive to SA (GmbZIPE1 and GmbZIPE2), by anunknown mechanism. c During the haustoria formation, the activationof necrotrophic pathogens-responsive genes occurs, causing theactivation of GmbZIP105, a JA-responsive bZIP gene. GmbZIPE1, anon-responsive to defense hormones bZIP gene, is activated by anunknown mechanism. In the figure, arrows indicate activation, andlines with bars indicate repression

Funct Integr Genomics

activation of all bZIP genes and SA signaling marker genesindicates a possible recognition mechanism for biotrophicfungal elicitors by the plant defense machinery, even causingthe activation ofGmbZIPE1, the bZIP gene non-responsive todefense hormones (Fig. 5; Online Resource 3). At the finaltime points (192 hai), the repression of SA signaling markergenes (GmPR1, GmPR4) suggested an activation of an antag-onistic to the SA process. Campe et al. (2014) showed thatP. pachyrhizi induces defense marker genes againstnecrotrophs in the non-host Arabidopsis thaliana, byPDF1.2 activation after both inoculation of A. thaliana leaveswith spores and upon their treatment with spore germinationfluid. This suggested JA signaling activation by secreted fun-gal effectors (Campe et al. 2014). Thus, the repression ofGmbZIP62 and GmbZIPE2, and the moderate activation ofGmbZIP105, correlates with the possible JA signaling activa-tion by secreted fungal effectors (Fig. 5).

In incompatible interactions between P. pachyrhizi and re-sistant plants, the late induction of SA signaling marker genesGmPR1 and GmNAC6 suggests a delay of SA signaling acti-vation, probably caused by modification of the plant innateimmunity machinery, a hypothesis that needs to be verifiedexperimentally. Interestingly, at 192 hai in the resistant plants,GmbZIPE1 expression was induced (Figs. 4 and 5; Onlineresource 3). As GmbZIPE1 is non-responsive to defense hor-mones, its activation seems to be solely dependent on thepresence of the fungal elicitors for the performance of its rolein plant defense machinery. This hypothesis also needs to beverified experimentally.

Infection is established late in Rpp1-resistant cultivars,most likely because of upregulated expression of genes asso-ciated with oxidative stress; therefore, the bZIP genes andmarker genes used in this study may be induced to performtheir functions only after the establishment of infection. Manystudies of the functions of bZIP proteins in response to path-ogens support the hypothesis that the biological functions ofthe bZIP factors studied here include defense against infectionby P. pachyrhizi (Alves et al. 2013). The GmbZIP62 andGmbZIP105 proteins are homologous to the G/HBF-1 andSBZ-1 proteins. Both are responsive to pathogen attack andalso show homology to the parsley bZIP factors CPRF-1,CPRF-2, and CPRF-3. These proteins bind the H-box cis-el-ement that confers a functional response to light. The H-boxcis-element is present in the chs15 promoter, which controlsthe expression of the gene encoding chalcone synthase, as dosoybean G/HBF-1 and SBZ1 (Weisshaar et al. 1991; Dröge-Laser et al. 1997). Chalcone synthase (CHS, EC 2.3.1.74) is akey enzyme in the biosynthesis of flavonoids, isoflavones, andphenylpropanoids and is directly involved in the defense path-waymediated by SA in plants (Vance et al. 1980; Dröge-Laseret al. 1997; Bhuiyan et al. 2007, 2009; Pandey et al. 2011).The H-box cis-element is a fundamental element in the pro-moter of the chs15 gene for transcriptional activation during

the defense response against pathogens by binding bZIP fam-ily transcription factors (Weisshaar et al. 1991; Loake et al.1992; Dröge-Laser et al. 1997). These data suggest a possiblefunction of GmbZIP62 and GmbZIP105 as regulators of theexpression of genes in the flavonoid biosynthetic pathwaymediated by SA. The genes that are controlled byGmbZIPE1 and GmbZIPE2 remain undetermined, and furtherstudies are needed to elucidate their roles in the defensemachinery.

Conclusions

In conclusion, our results show the differential expression ofGmbZIP62, GmbZIP105, GmbZIPE1, and GmbZIPE2 dur-ing fungal infection in both susceptible and resistant plants,which indicated the participation of these factors in plant-pathogen interactions, and also show that GmbZIP62,GmbZIP105, GmbZIPE1, and GmbZIPE2 are important can-didate genes that should be evaluated for the creation of dura-ble resistance to the fungus Phakopsora pachyrhizi insoybean.

Acknowledgments The authors thank Conselho Nacional deDesenvolvimento Científico e Tecnológico (CNPq), Coordenação deAperfeiçoamento de Pessoal de Nível Superior (Capes), Fundação deAmparo à Pesquisa do Estado de Minas Gerais (FAPEMIG), and theUniversidade Federal de Viçosa, Brazil.

Conflict of interest The authors declare that they have no conflict ofinterest.

Author contribution statement MSA and LGF conceived and de-signed research. MSA, ZGS, PMPV, AMPP, and LNA conducted theexperiments. MSA, PMPV, EGB, RVA, FCMG, and LGF analyzed thedata. MSA wrote the manuscript. All authors read and approved themanuscript.

References

Albani D, Hammond-Kosack MC, Smith C, Conlan S, Colot V,Holdsworth M, Bevan MW (1997) The wheat transcriptional acti-vator SPA: A seed-specific bZIP protein that recognizes the GCN4-like motif in the bifactorial endosperm box of prolamin genes. PlantCell 9:171–184

Alves MS, Dadalto SP, Gonçalves AB, De Souza GB, Barros VA, FiettoLG (2013) Plant bZIP transcription factors responsive to pathogens:a review. Int J Mol Sci 14:7815–7828

Amberg DC, Burke DJ, Strathern JN (2005) Assay of β-galactosidase inyeast: assay of crude extracts. Methods in yeast genetics. ColdSpring Harbor Laboratory Press, Cold Spring Harbor

Benko-Iseppon AM, Nepomuceno AL, Abdelnoor RV (2012)GENOSOJA—the Brazilian soybean genome consortium: highthroughput omics and beyond. Genet Mol Biol 35(1 (suppl)):i–iv

Bhuiyan NH, Liu W, Liu G, Selvaraj G, Wei Y, King J (2007)Transcriptional regulation of genes involved in the pathways of

Funct Integr Genomics

biosynthesis and supply of methyl units in response to powderymildew attack and abiotic stresses in wheat. Plant Mol Biol 64:305–318

Bhuiyan NH, Selvaraj G,Wei Y, King J (2009) Gene expression profilingand silencing reveal that monolignol biosynthesis plays a criticalrole in penetration defence in wheat against powdery mildew inva-sion. J Exp Bot 60:509–521

Bromfield KR (1984) Soybean rust. American PhytopathologicalSociety, St. Paul

Bromfield KR, Hartwig EE (1980) Resistance to soybean rust and modeof inheritance. Crop Sci 20:254–255

Campe R, Loehrer M, Conrath U, Goellner K (2014) Phakopsorapachyrhizi induces defense marker genes to necrotrophs inArabidopsis thaliana. Physiol Mol Plant Pathol 87:1–8

Cheng YW, Chan KL (1968) The breeding of rust resistant soybeanTainung 3. J Taiwan Agr Res 17:30–34

ChernMS, Eiben HG, BustosMM (1996) The developmentally regulatedbZIP factor ROM1 modulates transcription from lectin and storageprotein genes in bean embryos. Plant J 10:135–148

Chuang CF, Running MP, Williams RW, Meyerowitz EM (1999) ThePERIANTHIA gene encodes a bZIP protein involved in the deter-mination of floral organ number in Arabidopsis thaliana. Genes Dev13:334–344

Dietrich K, Weltmeier F, Ehlert A, Weiste C, Stahl M, Harter K, Dröge-Laser W (2011) Heterodimers of the Arabidopsis transcription fac-tors bZIP1 and bZIP53 reprogram amino acid metabolism duringlow energy stress. Plant Cell 23(1):381–395

Dröge-Laser W, Kaiser A, Lindsay WP, Halkier BA, Loake GJ, DoernerP, Dixon R, Lamb C (1997) Rapid stimulation of a soybean protein-serine kinase that phosphorylates a novel bZIP DNA-binding pro-tein, G/HBF-1, during the induction of early transcription-dependentdefenses. EMBO J 16:726–738

Freitas RL, Carvalho CM, Fietto LG, Loureiro ME, Almeida AM, FontesEP (2007) Distinct repressing modules on the distal region of theSBP2 promoter contribute to its vascular tissue-specific expressionin different vegetative organs. Plant Mol Biol 65(5):603–614

Fukazawa J, Sakai T, Ishida S, Yamaguchi I, Kamiya Y, Takahashi Y(2000) Repression of shoot growth, a bZIP transcriptional activator,regulates cell elongation by controlling the level of gibberellins.Plant Cell 12:901–915

Garcia A et al (2008) Molecular mapping of soybean rust (Phakopsorapachyrhizi) resistance genes: discovery of a novel locus and alleles.Theor Appl Genet 117:545–553

Goellner K et al (2010) Phakopsora pachyrhizi, the causal agent of Asiansoybean rust. Mol Plant Pathol 11:169–177

Hartman GL, Miles MR, Frederick RD (2005) Breeding for resistance tosoybean rust. Plant Dis 89:664–666

Hartwig EE (1986) Identification of a fourth major gene conferring resis-tance to soybean rust. Crop Sci 26:1135–1136

Hartwig EE, Bromfield KR (1983) Relationships among 3 genes confer-ring specific resistance to rust in soybeans. Crop Sci 23:237–239

Hidayat OO, Somaatmadja S (1977) Screening of soybean breeding linesfor resistance to soybean rust (Phakopsora pachyrhizi Sydow).Soybean Rust Newsl 1:9–22

Hoefle C, Loehrer M, Schaffrath U, Frank M, Schultheiss H,Hückelhoven R (2009) Transgenic suppression of cell death limitspenetration success of the soybean rust fungus Phakopsorapachyrhizi into epidermal cells of barley. Phytopathology 99:220–226

Huang J,Wang J, Zhang H (2005) Rice ZFP15 gene encoding for a novelC2H2-type zinc finger protein lacking DLN box, is regulated byspike development but not by abiotic stresses. Mol Biol Rep32(3):177–183

Hyten DL et al (2007) Map location of the Rpp1 locus that confersresistance to soybean rust in soybean. Crop Sci 47:837–838

Jakoby M et al (2002) bZIP transcription factors in Arabidopsis. TrendsPlant Sci 7:106–111

Katagiri F, Lain E, Chua NH (1989) Two tobacco DNA-binding proteinswith homology to the nuclear factor CREB. Nature 340:727–730

Kim KS et al (2012) Molecular mapping of soybean rust resistance insoybean accession PI 561356 and SNP haplotype analysis of theRpp1 region in diverse germplasm. Theor Appl Genet 125(6):1339–1352

Lee SJ et al (2002) PPI1: a novel pathogen-induced basic region-leucinezipper (bZIP) transcription factor from pepper. Mol Plant MicrobeInteract 15:540–548

Li S, Smith JR, Ray JD, Frederick RD (2012) Identification of a newsoybean rust resistance gene in PI 567102B. Theor Appl Genet125(1):133–142

Loake GJ, Faktor O, Lamb CJ, Dixon RA (1992) Combination of H-box[CCTACC(N), CT] and G-box (CACGTG) cis elements is neces-sary for feed-forward stimulation of a chalcone synthase promoterby the phenylpropanoid-pathway intermediate p-coumaric acid.Proc Natl Acad Sci U S A 89:9230–9234

Miao ZH, Liu X, Lam E (1994) TGA3 is a distinct member of the TGAfamily of bZIP transcription factors in Arabidopsis thaliana. PlantMol Biol 25:1–11

MonterosMJ et al (2007)Mapping and confirmation of the ‘Hyuuga’ red-brown lesion resistance gene for Asian soybean rust. Crop Sci 47:829–834

Morceli TGS et al (2008) Identification and validation of microsatellitemarkers linked to the Rpp5 gene conferring resistance to Asiansoybean rust. Pesq Agrop Brasileira 43:1533–1541

Nakagawa H, Ohmiya K, Hattori T (1996) A rice bZIP protein, designat-ed OSBZ8, is rapidly induced by abscisic acid. Plant J 9:217–227

Nakashima K, Yamaguchi-Shinozaki K (2006) Regulons involved inosmotic stress-responsive and cold stress-responsive gene expres-sion in plants. Physiol Plant 126:62–71

Oyama T, Shimura Y, Okada K (1997) The Arabidopsis HY5 gene en-codes a bZIP protein that regulates stimulus-induced developmentof root and hypocotyl. Genes Dev 11:2983–2995

Pandey AK et al (2011) Functional analysis of the Asian soybean rustresistance pathway mediated by Rpp2. Mol Plant Microbe Interact24(2):194–206

Patil VS, Wuike RV, Thakare CS, Chirame BB (1997) Viability of ure-dospores of Phakopsora pachyrhizi Syd. at different storage condi-tions. J Maharashtra Agric Univ 22:260–261

Pfaffl MW (2001) A new mathematical model for relative quantificationin real-time RT-PCR. Nucleic Acids Res 1:29–45

Pfaffl MW, Horgan GW, Dempfle L (2002) Relative expression softwaretool (REST) for group-wise comparison and statistical analysis ofrelative expression results in real-time PCR. Nucleic Acids Res 1:30–36

Reymond P, Farmer EE (1998) Jasmonate and salicylate as global signalsfor defense gene expression. Curr Opin Plant Biol 1:404–411

Schindler U, Beckman H, Cashmore AR (1992) TGA I and G-box bindingfactors: two distinct classes of Arabidopsis leucine zipper proteins com-pete for the G-box like element TGACGTGG. Plant Cell 4:1309–1319

Schneider KT et al (2011) Biphasic gene expression changes elicited byPhakopsora pachyrhizi in soybean correlate with fungal penetrationand haustoria formation. Plant Physiol 157(1):355–371

Singh K, Foley RC, Oñate-Sánchez L (2002) Transcription factors inplant defense and stress responses. Curr Opin Plant Biol 5:430–436

Tak H, Mhatre M (2012) Cloning and molecular characterization of aputative bZIP transcription factor VvbZIP23 from Vitis vinifera.Protoplasma 250:333–345

Umezawa T, Fujita M, Fujita Y, Yamaquchi-Shinozaki K, Shinozaki K(2006) Engineering drought tolerance in plants: discovering and tai-loring genes to unlock the future. Curr Opin Biotechnol 17:113–122

Vance CP, Kirk TK, Sherwood RT (1980) Lignification as a mechanismof disease resistance. Annu Rev Phytopathol 18:259–288

Funct Integr Genomics

Weisshaar B et al (1991) Light-inducible and constitutively expressedDNA-binding proteins recognizing a plant promder element withfunctional relevance in light responsiveness. EMBO J 10:1777–1786

Xia N et al (2010) Characterization of a novel wheat NAC tran-scription factor gene involved in defense responseagainststripe rust pathogen infection and abiotic stresses. Mol BiolRep 37(8):3703–3712

Yamaguchi-Shinozaki K, Shinozaki K (2005) Organization of cis-actingregulatory elements in osmotic- and cold-stress responsive pro-moters. Trends Plant Sci 10:88–94

Yamaguchi-Shinozaki K, Shinozaki K (2006) Transcriptional regulatorynetworks in cellular responses and tolerance to dehydration and coldstresses. Annu Rev Plant Biol 57:781–803

Yin Y et al (1997) RF2a, a bZIP transcriptional activator of the phloem-specific rice tungro bacilliform virus promoter, functions in vasculardevelopment. EMBO J 16:5247–5259

Zhang Yet al (2009) Cloning and characterization of a bZIP transcriptionfactor gene in wheat and its expression in response to stripe rustpathogen infection and abiotic stresses. Physiol Mol Plant Pathol73:88–94

Funct Integr Genomics