ENVIRONMENTAL BIOTECHNOLOGY

Characterization of soybean β-expansin genesand their expression responses to symbiosis,nutrient deficiency, and hormone treatment

Xinxin Li & Jing Zhao & Thomas C. Walk & Hong Liao

Received: 16 May 2013 /Revised: 15 August 2013 /Accepted: 18 August 2013 /Published online: 10 October 2013# Springer-Verlag Berlin Heidelberg 2013

Abstract Expansins are plant cell wall-loosening proteinsencoded by a superfamily of genes including α-expansin,β-expansin, expansin-like A, and expansin-like B proteins.They play a variety of biological roles during plant growthand development. Expansin genes have been reported in manyplant species, and results primarily from graminaceous mem-bers indicate that β-expansins are more abundant in monocotsthan in dicots. Soybean [Glycinemax (L.)Merr] is an importantlegume crop. This work identified nineβ-expansin gene familymembers in soybean (GmEXPBs) that were divided into twodistinct classes based on phylogeny and gene structure, withdivergence between the two groups occurring more in intronsthan in exons. A total of 887 hormone-responsive and environ-mental stress-related putative cis-elements from 188 familieswere found in the 2-kb upstream region of GmEXPBs .Variations in number and type of cis-elements associated witheach gene indicate that the function of these genes is differen-tially regulated by these signals. Expression analysis confirmedthat the family members were ubiquitously, yet differentiallyexpressed in soybean. Responsiveness to nutrient deficiencystresses and regulation by auxin (indole-3-acetic acid) andcytokinin (6-benzylaminopurine) varied among GmEXPBs .

In addition, most β-expansin genes were associated with sym-biosis of soybean inoculated with Rhizobium or abuscularmycorrhizal fungi (AMF). Taken together, these resultssystematically investigate the characteristics of the entireGmEXPB family in soybean and comprise the first reportanalyzing the relationship of GmEXPBs with rhizobial orAMF symbiosis. This information is a valuable step in theprocess of understanding the expansin protein functions insoybean and opens avenues for continued researches.

Keywords β-expansin . Gene family . Expression pattern .

Soybean

Introduction

Plants have evolved tough and flexible, but sometimes fairlyrigid, layer cell walls to protect tissues from environmentaldamage as well as to maintain structural stability. During celldivision and elongation, the cell wall must be extended andproperly form the cellulose–hemicellulose network linked viahemicellulosic tethers. Expansins are proteins that play impor-tant roles in rapid pH-dependent plant cell wall extension or“acid growth” through the loosening of wall matrix polymersand modification of the linkage between adjacent cellulosemicrofibrils (Cosgrove 1989; McQueen-Mason et al. 1992).These activities and products were initially identified incucumber hypocotyls in which extension of heat-inactivatedcell walls was restored in vitro (Cosgrove 1998, 2000).

Expansins are encoded by a superfamily of genes delimitedas α-expansin (EXPA), β-expansin (EXPB), expansin-like A(EXLA), or expansin-like B (EXLB). The proteins have asignal peptide in their N-terminus, indicating that expansinproteins are secreted. EXPAs and EXPBs have a series ofconserved cysteine (Cys) residues in the N-terminal region, aconserved His-Phe-Asp (HFD) motif in the center region and

Electronic supplementary material The online version of this article(doi:10.1007/s00253-013-5240-z) contains supplementary material,which is available to authorized users.

X. Li : J. Zhao :H. LiaoState Key Laboratory for Conservation and Utilization of SubtropicalAgro-Bioresources, Root Biology Center, South China AgriculturalUniversity, Guangzhou, People’s Republic of China

T. C. WalkGolden Fidelity LLC, St. Louis, MO, USA

H. Liao (*)Root Biology Centre, South China Agricultural University,Guangzhou 510642, People’s Republic of Chinae-mail: hliao@scau.edu.cn

Appl Microbiol Biotechnol (2014) 98:2805–2817DOI 10.1007/s00253-013-5240-z

several conserved tryptophan (Trp) residues in the putativecellulose-binding domain in the C-terminal region (Choi et al.2006; Cosgrove 1998). The conserved domains of EXLAs andEXLBs have diverged from each other as well as from EXPAsand EXPBs, with loss of the HFD motif in the center regionbeing most notable (Choi et al. 2006).

Expansin activity is most often associated with cell wall-loosening in growing cells, but the precise mode of actionremains a mystery. Previous research has revealed that plantcell walls extend or relax by the process of molecular “creep,”in which expansin proteins are hypothesized to disrupt thebonding of glycans to the microfibril surface (McQueen-Mason and Cosgrove 1994) or weaken the non-covalentbinding between wall polysaccharides (Cosgrove 2000).Recently, a novel activity of grass pollenβ-expansin frommaizewas found to be solublization of wall matrix polysaccharideswith weak dependence on temperature, and thereby indicating anon-enzymatic process (Tabuchi et al. 2011). Furthermore, inother tests, expansions were found to not have hydrolytic orother enzymatic activities (Li and Cosgrove 2001; McQueen-Mason et al. 1993; McQueen-Mason and Cosgrove 1995).

Expansin genes have been reported to play a broad range ofroles in diverse aspects of plant growth and development in awide variety of species. Based on searches of Arabidopsis andrice databases, as well as a review publication (Sampedro andCosgrove 2005), it is evident that EXPBs are more abundantin monocots than in dicots. It may be because the compositionof primary cell walls in monocots differs significantly fromthose in dicots (Carpita and Gibeaut 1993). EXPBs were firstknown as group-1 grass pollen allergens with expansin activity.These allergens, e.g., Zeam1, are hypothesized to aid fertiliza-tion by loosening the cell walls of the stigma and style, andthereby facilitating penetration of the pollen tube (Cosgroveet al. 1997). Beyond the group-1 allergens, research has shownthat EXPBs are involved in many other biological processes.The first barleyEXPB gene,HvEXPB1 , is a critical factor in theinitiation of root hair development (Kwasniewski and Szarejko2006). TaEXPB23 , an expansin gene isolated from wheatcoleoptiles (Gao et al. 2008), enhances drought resistance intobacco plants (Xing et al. 2009). TaEXPB23-overexpressingtransgenic lines sustain structural integrity under water stressbetter than wild-type plants (Li et al. 2011). It has been shownthrough promoter deletion that RHE motifs are necessary forroot hair-specific expression genes, such as HvEXPB1 frombarley andOsEXPB5 from rice (Won et al. 2010). Analysis ofEXPB2 from maize implicates ZmEXPB2 as a factor in theelongation of maize roots with possible involvement in plantresponses to environmental stimuli (Kam et al. 2005).

To date, a handful of reports provide limited information onEXPBs in dicots. One example is theArabidopsis EXPB gene,AtEXP -β1 , which confers sensitivity to salt stress by anunknown mechanism in ectopic expression analysis (Kwonet al. 2008). In soybean, only two EXPBs have been reported.

One is a cytokinin-induced mRNA (Cim1 ) that may accumu-late to high levels in suspension-cultured soybean cells(Crowell 1994), and is regulated by a post-transcriptionalmechanism (Downes and Crowell 1998). Further studies havedemonstrated that cytokinin and auxin act synergistically toinduce the accumulation and proteolytic processing of Cim1 ,and that Cim1 proteins are extensively modified by post-translational N-glycosylation and proteolysis (Downes et al.2001). The other reported soybean β-expansin, GmEXPB2 ,was cloned from a phosphate (Pi) starvation-induced soybeancDNA library. It is a critical root β-expansin gene that isintrinsically involved in root system architecture responsesto P, Fe, and water deficiency (Guo et al. 2011).

In short, most work on expansin genes focuses onArabidopsis (Li et al. 2002) or graminaceous crops, such asrice (Magneschi et al. 2009; Shin et al. 2005), maize (Wu et al.2001), and wheat (Liu et al. 2007; Sun et al. 2005). Little isknown about these genes and products in most dicots,including soybean. In this study, we report the systematicinvestigation of the entire β-expansin gene family members(GmEXPB), including sequence analysis of genes and pro-teins, similarity within soybean and with EXPBs from otherspecies, gene structural analysis, and promoter cis -elementidentification. We also analyze tissue-specific expression pat-terns, responses to nutrient deficiency or hormone treatment,and effects of symbiosis using quantitative real-time PCR(qRT-PCR). All results demonstrate possible functions ofEXPBs in cell growth and extension of soybean under avariety of growth conditions, as well as in regulating plantresponses to nutrient deficiency and phytohormone signaling.

Materials and methods

Database search and molecular sequence analysisof β-expansin genes in soybean

The amino acid sequences derived from the reported soybeanβ-expansin genes GmEXPB1 (Cim1) (Crowell 1994; Downesand Crowell 1998; Downes et al. 2001) and GmEXPB2 (Guoet al. 2011) were used to BLAST search soybean at thephytozome website (http://www.phytozome.net), which yieldedfive soybean β-expansin genes. Further BLAST searches werethen conducted using these five amino acid sequences at thesame website, which led to a total of nine predicted β-expansingenes in the soybean genome. Furthermore, a BLAST searchin the phytozome database was also carried out using theamino acid sequences of six Arabidopsis β-expansinsobtained from The Arabidopsis Information Resource(http://www.arabidopsis.org/). No more predicted β-expansinswere found, possibly reflecting that the total number ofβ-expansin family members in the soybean genome is nine.The nine soybean β-expansin genes were named GmEXPBs

2806 Appl Microbiol Biotechnol (2014) 98:2805–2817

(1 to 9) according to their positions on the chromosomes,except GmEXPB1 (Cim1) and GmEXPB2 , which were num-bered based on previous identification and are located on chro-mosome 01(Glyma01g16140.1) and 10 (Glyma10g24080.1),respectively. Genome sequences of GmEXPBs were retrievedfrom the phytozome website. Calculations of protein molecularweight and theoretical isoelectric point (pI) values wereperformed using the Compute pI/Mw tool (http://www.expasy.org/tools/pi_tool.html). A phylogenetic tree of GmEXPBs,AtEXPBs, OsEXPBs, two homologous β-expansins fromMedicago as well as the corresponding α-expansins wasconstructed based on entire protein sequence alignments usingClustalX (Larkin et al. 2007) and the neighbor-joining methodwith 1,000 bootstrap replicates in the MEGA 4.1 program(http://www.megasoftware.net/mega4/mega41.html). Signalpeptides and maximum cleavage site probability in the aminoacid sequences was determined using the SignalP 3.0 server(http://www.cbs.dtu.dk/services/SignalP/). The genomicstructures of soybean β-expansin genes were analyzed atthe Gene Structure Display Server (http://gsds.cbi.pku.edu.cn/index.php). The putative cis-elements of GmEXPBs werepredicted using the website of plant cis-acting regulatoryDNA elements (PLACE) (http://www.dna.affrc.go.jp/PLACE/signalscan.html).

Plant materials and growth conditions

Soybean genotype HN66 seeds were surface-sterilized in10 % H2O2 for 1 min, rinsed with distilled water, and germi-nated in glass plates containing filter paper saturated by 1/2strength nutrient solution. One week after germination, theseedlings were transplanted into full-strength nutrient solutionthat has been modified by our lab (Li et al. 2012). Theseedlings were grown for 28 days, and then roots, stems,leaves, and flowers were separately harvested. At 39 days,young pods and seeds were separately harvested. All tissuesamples were stored at −80 °C for RNA extraction.

For expression analysis of GmEXPBs in root nodules,1-week-old soybean seedlings were inoculated withBradyrhizobium sp. BXYD3 and transplanted into lownitrogen (100-μM N added) nutrient solution. BXYD3 isan effective rhizobial strain (Cheng et al. 2009), and itscollection number is CGMCCNo.: M207123 in China Centerfor Type Culture Collection. The nodules were harvested at50 days after inoculation and were used for RNA extractionand qRT-PCR analysis.

For the abuscular mycorrhizal fungi (AMF) infectionexperiment, soybean seeds were germinated in sand culture withor without the inoculants, which consists of rhizosphere soilcontaining Glomus mosseae spores, mycelium, and colonizedroot fragments of maize. The AMF strain was bought fromthe International Culture Collection of Vesicular ArbuscularMycorrhizal Fungi (http://invam.wvu.edu/, Morgantown,

WV, USA), and multiplied in our lab using maize plants.The seedlings were irrigated with nutrient solution containing5-μM KH2PO4 every day and the roots were harvested50 days after inoculation for RNA extraction and qRT-PCR analysis. In addition, parts of the roots were used toevaluate colonization by analysis of the expression of twoAMF-induced soybean phosphorus transporters GmPT10and GmPT11 (Tamura et al. 2012).

To scrutinize possible impacts of nutrient deficiency onGmEXPBs expression, 10-day-old seedlings pre-cultured infull-strength nutrient solution as described above were thentreated under low nitrogen (−N), low phosphorus (−P), lowpotassium (−K), or low iron (−Fe) conditions for 14 days.For −N treatment, KNO3 and Ca(NO3)2 were replaced by750-μM K2SO4 and 1,200-μM CaCl2, respectively. For −Ptreatment, KH2PO4 was replaced by 125-μM K2SO4. For −Ktreatment, KNO3 and KH2PO4 were replaced by 1,500-μMCa(NO3)2 and 250-μMNaH2PO4, respectively. For −Fe treat-ment, Fe-EDTA was fully withdrawn. Plants continuouslygrown in the full-strength nutrient solution were sampled asthe control treatment (CK). Each treatment had three biologicalreplicates. Leaves and roots were separately collected for totalRNA extraction and qRT-PCR analysis. The soybean genePLDZ (Glyma20g38200, Accession No. KF542816) wasused as a marker gene for low P responsiveness, as were NiR(Glyma02g14910, Accession No. KF542817) for −N, HAK(Glyma03g42480, Accession No. KF542818) for −K, and IRT(Glyma07g34930, Accession No. KF542819) for −Fe.

To test the effects of auxin (indole-3-acetic acid, IAA) andcytokinin (6-benzylaminopurine, 6-BA) on the expression ofGmEXPBs, soybean seedlings were germinated on germina-tion paper with full-strength nutrient solution and treated with0.5-μM IAA or 0.5-μM 6-BA for 4 days. Plants continuouslygrown in full-strength nutrient solution without phytohor-mone addition were used as CK. Leaves and roots wereseparately sampled for total RNA extraction and qRT-PCRanalysis. Each treatment had three biological replicates.

RNA isolation and qRT-PCR analysis

High-quality total RNA was isolated from soybean roots,nodules, mycorrhizal roots, stems, leaves, flowers, youngpods, and seeds using RNAisoTM Plus reagent (Takara Bio,Otsu, Shiga, Japan) according to themanufacturer’s instructions.In order to avoid amplification from genomic DNA, the RNAsamples were treated with RNase-free DNase I (Invitrogen,Grand Island, NY, USA). Then, the first cDNA strand wassynthesized from total RNA using oligo d(T), dNTPs, andMMLV-reverse transcriptase (Promega, Madison WI, USA)according to the protocol from the supplier. qRT-PCR wasperformed in a 20-μL volume containing 2-μL 1:50 dilutedreverse transcription product, 0.6-μL specific primers (0.2-μMfinal concentration), 6.8-μL ddH2O and 10-μL SYBR® Premix

Appl Microbiol Biotechnol (2014) 98:2805–2817 2807

EX Taq™ (Takara Bio, Otsu, Shiga, Japan). The qRT-PCRreactions were carried out in a Rotor-Gene RG-3000 (CorbettResearch, Mortlake, NSW, Australia) machine and reactionprocesses were similar to those described by Li et al. (2012).The elongation factor 1α (TefS1) gene (Accession No. X56856)was used as the housekeeping gene to calculate the relativetranscript abundance as the ratio of the expression value of thetarget gene to that of TefS1 . The specific primers for detectingtarget genes were listed in Supplementary Table S1.

Date analysis

The data from qRT-PCR analysis were normalized in eachexperiment. All data were analyzed statistically using MicrosoftExcel 2007 (Microsoft Company, USA) for calculating meanand standard error.

Accession numbers of GmEXPBs

The accession numbers for each soybean β-expansin genesfamily numbered from GmEXPB1 (Cim1) to GmEXPB9 areU03860, EU362626, JQ303249, JQ303250, JQ303251,JQ303252, JQ303253, JQ303254, and JQ303255, respectively(Supplementary Table S2).

Results

Identification and characterization of soybean β-expansingene family members

A total of nine putative GmEXPBs were identified in thephytozome website (http://www.phytozome.net/). The full-length cDNAs and amino acid sequences of GmEXPB3 ,GmEXPB5 and GmEXPB9 were obtained using the NCBIORF finder web application (http://www.ncbi.nlm.nih.gov/gorf/gorf.html), while all of the others were available in thephytozome website (Supplementary Table S2). Nine genesencoding soybean β-expansin proteins were detected andnew members were named GmEXPB3 through GmEXPB9 .All new GmEXPBs were named according to their chromo-some localization, while the previously identified GmEXPB1(Cim1 ) and GmEXPB2 are localized on chromosome 1 and10, respectively (Supplementary Table S2). All GmEXPBsexhibit a high similarity in coding region length and calculatedmolecular weight, but varied in theoretical pI values. Thetheoretical pI values of the mature β-expansin proteins rangedfrom 4.87 to 9.11. Signal peptide prediction by Signal P 3.0indicates that all β-expansin genes contain N-terminal secre-tory signal peptides with lengths ranging from 23 to 31residues (Supplementary Table S2).

A multiple sequence alignment was conducted to examineamino acid sequence similarity and to identify conserved

motifs. Each GmEXPB product contains previously identifiedhighly conserved regions (Cosgrove 1998), all GmEXPBshave seven conserved C (Cys) residues in the N-terminal halfof the protein, an HFD motif in the center region, and fourconserved W (Trp) residues towards the C-terminus (Fig. 1a).

Full-length cDNAs and genome sequences ofGmEXPBs wereused to investigate gene structures on the Gene Structure DisplayServer (http://gsds.cbi.pku.edu.cn/index.php). The result was thatallGmEXPBs have four exons and three introns, yet they can stillbe divided into two classes. The first class includes GmEXPB1,GmEXPB2, GmEXPB3, GmEXPB5, and GmEXPB6, with totalintron length varying from 1,164 to 1,793 bp, and an intron phasepattern as 1, 0, and 2 from 5′ to 3′. The remaining fourGmEXPBsbelong to the second class with an intron phase pattern of 1, 2, and0 from 5′ to 3′, and with one or more long introns, especiallyGmEXPB7, which has a total intron length of 4,131 bp (Fig.1b and Supplementary Table S3). This is more than doublethe longest total intron length of genes in the first group. It isalso notable that all GmEXPBs in the first class containequivalent second-exon length, and all GmEXPBs in thesecond class have equivalent second-, third-, and fourth-exon lengths (Supplementary Table S3).

To determine the evolutionary relationships among the β-expansin family members in soybean, a phylogenetic tree wasconstructed by neighbor-joining analysis in MEGA 4.1 (http://www.megasoftware.net/mega4/mega41.html) including the fullEXPB protein sequences from Arabidopsis, rice, soybean, andtwo Medicago EXPBs. As shown in Fig. 1c, α-expansins havediverged significantly from β-expansins. Within β-expansins,nine GmEXPBs, six AtEXPBs, and two Medicago β-expansinproteins were sorted into two distinct clades with rice β-expansins intermixed, suggesting that β-expansins in monocotsand dicots shared common ancestry and the divergence of β-expansins between two clades occurred prior to the divergence ofmonocots and dicots.Most of the testedβ-expansins belonged toclade I, which included 5 of 9 soybeanβ-expansin proteins, 17 of19 rice β-expansin proteins, 4 of 6 Arabidopsis β-expansin

�Fig. 1 General characteristics of the nine identified GmEXPBs . aAlignment of amino acid sequences of the β-expansins in soybean.Sequence alignment was performed with the ClustalW program.Identical and similar amino acids were shaded in black and gray,respectively. Positions of the highly conserved Cys residues, the HFDdomain, and Trp residues were indicated. b Gene structures ofGmEXPBsfrom soybean. White boxes represent exons and thin lines representintrons. Numbers above introns indicated intron phases. Phase 0, 1, and2 introns were defined as introns starting before the first, after the first,and after the second nucleotide of a codon, respectively. c Phylogeneticanalysis of α- and β-expansins. This phylogenetic tree including 9EXPBs from soybean (Glycine max), 6 from Arabidopsis thaliana, 19from rice (Oryza sativa ), and 2 from Medicago truncatula wasconstructed by the neighbor-joining method with 1,000 bootstrapreplicates in the MEGA 4.1 program. β-expansin proteins were sortedinto two different clades (I and II), and the α-expansins had divergedconsiderably from β-expansin genes as shown in this tree

2808 Appl Microbiol Biotechnol (2014) 98:2805–2817

A

Exon Intron UTR

0 1 2 3 4 55’ 3’

GmEXPB1 (Cim1)

GmEXPB2

GmEXPB5

GmEXPB6

GmEXPB3

GmEXPB4

GmEXPB7

GmEXPB8

GmEXPB9

B

GmEXPB1 (Cim1)GmEXPB2GmEXPB3GmEXPB4GmEXPB5GmEXPB6GmEXPB7GmEXPB8GmEXPB9

GmEXPB1 (Cim1)GmEXPB2GmEXPB3GmEXPB4GmEXPB5GmEXPB6GmEXPB7GmEXPB8GmEXPB9

GmEXPB1 (Cim1)GmEXPB2GmEXPB3GmEXPB4GmEXPB5GmEXPB6GmEXPB7GmEXPB8GmEXPB9

GmEXPB1 (Cim1)GmEXPB2GmEXPB3GmEXPB4GmEXPB5GmEXPB6GmEXPB7GmEXPB8GmEXPB9

GmEXPB1 (Cim1)GmEXPB2GmEXPB3GmEXPB4GmEXPB5GmEXPB6GmEXPB7GmEXPB8GmEXPB9

GmEXPB1 (Cim1)GmEXPB2GmEXPB3GmEXPB4GmEXPB5GmEXPB6GmEXPB7GmEXPB8GmEXPB9

474940414647413532

979990919697918582

147149139140145147140134131

197199188190195197190184181

247249237237242247237231228

277277267267272277267261258

20 40

60

C

80 100

C C C 120 C

C

140HFD

160 180 200

220 240

260 280

WW W

W

Appl Microbiol Biotechnol (2014) 98:2805–2817 2809

proteins, and both of the tested Medicago β-expansin proteins.In clade I, four soybean β-expansin proteins, GmEXPB1,GmEXPB2, GmEXPB3, GmEXPB5, and GmEXPB6, werevery close, while GmEXPB3 separated into a subgroup withbothMedicago β-expansin proteins. The smaller clade II includ-ed a pair of β-expansins from Arabidopsis , AtEXPB1 andAtEXPB3, a pair from rice, OsEXPB16 and OsEXPB17, andfour from soybean, GmEXPB4, GmEXPB7, GmEXPB8, andGmEXPB9. The lack of Medicago representation in clade IIindicates that either this genus has lost these members throughevolution or there remain undiscovered β-expansins in

Medicago . This classification of soybean β-expansins basedupon protein phylogeny is consistent with grouping based ongene structures (Fig. 1b, c).

In a search of the PLACE database (http://www.dna.affrc.go.jp/PLACE/), a total of 887 putative cis-elements from 188families were found in the 2-kb upstream regions of the ninesoybean GmEXPBs by searching the PLACE database (http://www.dna.affrc.go.jp/PLACE/, Supplementary Table S4). Thesecis-elements were not distributed uniformly amongGmEXPBs ,so a complex pattern of differential expressions among tissues ortreatments might be expected. Among these identified putative

OsEXPB15 (Os04g0552000) OsEXPB18 (Os05g0246300)

OsEXPB12 (Os03g0645000) OsEXPB14 (Os02g0658600) OsEXPB5 (Os04g0552200)

OsEXPB11 (Os02g0658800) GmEXPB1 (Glyma01g16140) GmEXPB6 (Glyma11g17160)

GmEXPB2 (Glyma10g24080) GmEXPB5 (Glyma10g24120)

GmEXPB3 (Glyma03g03980) Mt4g064780.1

Mt4g036120.1 AtEXPB5 (At3g60570)

AtEXPB4 (At2g45110) AtEXPB6 (At1g65681)

AtEXPB2 (At1g65680) OsEXPB1a (Os03g0106500) OsEXPB1b (Os03g0106900) OsEXPB13 (Os03g0106700)

OsEXPB10 (Os03g0106800) OsEXPB9 (Os10g0548600)

OsEXPB8 (Os03g0102500) OsEXPB2 (Os10g0555700)

OsEXPB7 (Os03g0102700) OsEXPB3 (Os10g0555900)

OsEXPB4 (Os10g0556100) OsEXPB6 (Os10g0555600)

OsEXPB17 (Os04g0530100) AtEXPB1 (At2g20750)

AtEXPB3 (At4g28250) OsEXPB16 (Os02g0639500)

GmEXPB4 (Glyma06g44930) GmEXPB7 (Glyma12g12350)

GmEXPB8 (Glyma12g33070) GmEXPB9 (Glyma13g37390)

OsEXPA10 (Os04g49410) AtEXPA7 (At1g12560)

OsEXPA17 (Os06g01920) OsEXPA9 (Os01g14660)

OsEXPA11 (Os01g16770) GmEXP1 (Glyma17g37990)

OsEXPA1 (Os04g15840) OsEXPA15 (Os03g06020)

OsEXPA12 (Os03g06000) OsEXPA13 (Os02g16730)

OsEXPA2 (Os01g60770) AtEXPA2 (At5g05290)

AtEXPA8 (At2g40610) AtEXPA1 (At1g69530) AtEXPA10 (At1g26770) AtEXPA5 (At3g29030)

AtEXPA3At2g37640) AtEXPA9 (At5g02260)

GmEXP2 (Glyma07g15910) AtEXPA4 (At2g39700) AtEXPA6 (At2g28950)

0.1

EXPB

EXPA

I

II

C

Fig. 1 (continued)

2810 Appl Microbiol Biotechnol (2014) 98:2805–2817

cis-elements, 35 were found in the upstream regions of allsoybean β-expansin genes, 39 were only located in the up-stream of one soybean β-expansin gene, and 24 were associ-ated with multiple genes in only one class (either class I or II)(Supplementary Table S4). Moreover, base on analysis inPLACE (http://www.dna.affrc.go.jp/PLACE/signalscan.html),13 cis-elements are known for tissue-specific gene expression,and nine are sites for recognition of response regulators,such as transcriptional activators, DNA-binding proteins andpolyadenylation signals. Furthermore, 10 cis-elements arerevealed to be environmental stress-related motifs, and twoare hormone-related elements. Notable examples include (1)ACGTATERD1, which is required for etiolation-inducedexpression; (2) GATABOX, GT1CONSENSUS, IBOXCORE,and INRNTPSADB, which are mainly light-responsive regula-tory elements; (3) MYBCORE, MYCCONSENSUSAT, andWBOXNTERF3, which are chiefly responsive to waterstress; (4) WBOXATNPR1, which binds transcription factorsrelated to defense responses (Chen and Chen 2002), and (5)GT1GMSCAM4, which plays roles in pathogen- and salt-induced SCaM-4 gene expression in soybean (Park et al.2004). In regards to hormone-responsive elements, ERELEE4mediates ethylene-induced expression (Tapia et al. 2005), andWRKY71OS modulates gibberellin (Zhang et al. 2004) andabscisic acid (Xie et al. 2005) signaling pathways.

Tissue-specific expression patterns of soybean β-expansingenes

To clarify the tissue specificity of soybean β-expansingenes, qRT-PCR analysis were performed to determine rel-ative transcript values in six tissues, including roots, stems,young leaves, flowers, young pods, and seeds. The resultsshowed that the expression of each GmEXPB varied amongtissues. Three class II genes, GmEXPB4 , GmEXPB7 , andGmEXPB9 , were expressed in all the tested tissues(Supplementary Fig. S1). GmEXPB7 was predominantlyexpressed in leaves and roots, while GmEXPB4 andGmEXPB9 were mainly expressed in leaves and flowers.Expressions of the other six GmEXPBs showed tissue spec-ificity (Table 1). GmEXPB1 and GmEXPB2 were mainlyexpressed in roots. GmEXPB3 and GmEXPB5 were mainlyexpressed in flowers and seeds, respectively. GmEXPB8was expressed at the highest levels by a wide margin,particularly in pods and seeds, and it was still abundant inroots and flowers in comparison to the other GmEXPBs . Onthe other hand, GmEXPB6 was relatively rare compared tothe other GmEXPBs , but it still appeared to play a role inroots and shoots. SevenGmEXPBs were expressed in roots,stems, pods, and seeds, while three and six GmEXPBs wereexpressed in leaves and flowers, respectively. These resultsindicate that individual tissues incorporate the coordinatedfunctioning of multiple β-expansin genes.

Expression patterns of soybean β-expansin genes in soybeannodules and mycorrhizal roots

To investigate how the nine soybeanβ-expansin genes respondto nodulation and mycorrhization, qRT-PCR analysis was car-ried out using total RNA from nodules and mycorrhizal roots.The results showed considerable effects of nodulation on theexpressions of GmEXPBs (Fig. 2a). Two soybean β-expansingenes that were not detected in roots, GmEXPB3 andGmEXPB5 , were found in nodules. On the other hand,GmEXPB2 , which was predominantly expressed in roots, wasnearly undetectable in nodules. GmEXPB4 , GmEXPB5 , andGmEXPB6 were up-regulated in nodules compared to roots(Fig. 2a, Supplementary Fig. S1). In contrast, GmEXPB8 wasfound in nodules, but at much lower levels than in roots. Formycorrhization effects, the transcription of soybean AMF in-ducible phosphate transporters GmPT10 and GmPT11 wereused as markers to verify AMF colonization of roots (Fig. 2b).As shown in Fig. 2c, in mycorrhizal roots, the expression wasenhanced for fiveGmEXPBs and reduced for two others. Usinga significance threshold of a twofold change, the transcriptionlevels of GmEXPB2 and GmEXPB9 were significantly up-regulated by 3.19- and 4.22-fold, respectively. No other expres-sion changes exceeded the significance threshold. GmEXPB3and GmEXPB5 were not detected in mycorrhizal roots.

Transcriptional changes of GmEXPBs in response to nutrientdeficiency

To investigate the possible roles of GmEXPBs in adaptation todifferent nutrient deficiencies, soybean plants were treated withfour nutrient stresses,−N, −P, −K, and −Fe, for 14 days. SoybeanNiR, PLDZ , HAK , and IRT were used as marker genes for −N,−P, −K, and −Fe deficiency, respectively. As expected, NiR was

Table 1 Most abundant GmEXPB mRNAs in six different tissues

Root Stem Leaf Flower Pod Seed

GmEXPB1(Cim1) +

GmEXPB2 +

GmEXPB3 +

GmEXPB4 + +

GmEXPB5 +

GmEXPB6 + +

GmEXPB7 +

GmEXPB8 + +

GmEXPB9 + +

In each tissue, the fold change between the most highly expressed andsecond most highly expressed GmEXPB was calculated as previouslydescribed (Li et al. 2012). A threshold of twofold was used to determinethe most abundant GmEXPB in each tissue. If the difference between themost highly and second most highly expressed GmEXPB was less thantwofold, then they both were considered abundant and labeled with +

Appl Microbiol Biotechnol (2014) 98:2805–2817 2811

significantly suppressed in both roots and shoots under −Nconditions, while PLDZ, HAK , and IRT were highly expressedin the −P shoots, −K roots, and −Fe roots, respectively(Supplementary Fig. S2). Results from qRT-PCR analysis usinga significance threshold of a twofold change in relative expres-sion showed that each GmEXPB responded to one or morenutrient stresses, and that responses varied among GmEXPBs(Table 2 and Supplementary Fig. S2). For −N treatment,GmEXPB8 and GmEXPB9 were induced in roots, andGmEXPB3 was induced in both roots and shoots. Decreasesin shoot transcript levels in response to N deficiency wereobserved for GmEXPB1 and GmEXPB7 , while, otherGmEXPBs showed no alteration in response to N treatment.In response to P deficiency, seven of nineGmEXPBs displayedaltered expression in at least one tissue (Table 2 andSupplementary Fig. S2). GmEXPB3 was significantly inducedin shoots, whileGmEXPB2 andGmEXPB7 were preferentiallyincreased in roots. Two class II genes, GmEXPB8 andGmEXPB9 , were enhanced in both roots and shoots, whiletwo class I genes, GmEXPB1 and GmEXPB6 , had decreasedexpression levels in roots and shoots, respectively. In the −Ktreatment, GmEXPB2 , GmEXPB5 , and GmEXPB9 displayedno change at transcription levels in either roots or shoots,while GmEXPB3 was more active in both roots and shoots(Table 2 and Supplementary Fig. S2). GmEXPB6 appearedto be up-regulated and down-regulated in shoots and roots,respectively. GmEXPB4 had increased transcription levels inshoots, while GmEXPB7 and GmEXPB8 were decreased in

45

0

90

135

180

1 2 3 4 5 6 7 8 9

AR

elat

ive

expr

essi

on v

alue

500

250

375

125

0

GmPT10

Rel

ativ

e ex

pres

sion

val

ue

B

1 2 3 4 5 6 7 8 9

Fold

cha

nge

of e

xpre

ssio

n

C

GmEXPBs

GmEXPBs

GmPT111600

800

1200

400

0

Shoot-M +M

Shoot

-M +MRoot

-M +M

Root

-M +M

Fig. 2 Expressions of GmEXPBs in response to symbiosis with rhizobiaor AMF. a The transcription levels of GmEXPBs in nodules. Seven-day-old seedlings were inoculated with rhizobia, Bradyrhizobium sp. BXYD3for 50 days. Mature nodules were harvested for RNA extraction and qRT-PCR analysis. Numbers from 1 to 9 under the x-axis represent GmEXPB1to GmEXPB9 . Expression values were relative to the housekeeping geneTefS1 . Each bar represents the mean of four biological replications withstandard error. (b) Expressions of two AM-inducible phosphate trans-porters GmPT10 and GmPT11 in the AMF-colonized soybean roots.Shoots and roots of soybean seedlings grown in sand culture with (+M)or without (−M)AMF inoculation for 50 days under low P conditions wereseparately harvested for expression analysis. Expression values were rela-tive to the housekeeping gene TefS1 . c The fold changes of GmEXPBexpressions between mycorrhizal colonized (+M) and control (−M) roots.The negative value represents higher expressions of GmEXPBs innoninoculated roots than in inoculated roots. A threshold of twofold changein relative expression was used to determine significance of responsivenessof GmEXPBs to AMF inoculation. GmEXPB3 and GmEXPB5 were notdetected in roots regardless of inoculation or not. The numbers from 1 to 9under the x-axis representsGmEXPB1 toGmEXPB9 . Each bar representsthe mean of four biological replications with standard error

Table 2 Responses of GmEXPBs to deficiency of nitrogen, phosphorus,potassium, or iron

Root Shoot

−N −P −K −Fe −N −P −K −Fe

GmEXPB1(Cim1) − − − − − − − −GmEXPB2 +

GmEXPB3 + + +++ +

GmEXPB4 − + − −GmEXPB5

GmEXPB6 − − − − +

GmEXPB7 + − − − − −GmEXPB8 + + − + − − − −GmEXPB9 + + − + − −

10-day-old soybean seedlings were treated with N (−N), P (−P), K (−K),or Fe (−Fe) deficiency. Seedlings grown in the full nutrient solution wereused as controls. The fold changes of the expression of GmEXPBsbetween stress treatments and controls were calculated. The expressiondifferences were classified into four groups according to their responsesto abiotic stresses as follows: (1) not induced or inhibited (no symbols),(2) induced/inhibited (+/− ≥two- and <tenfold change), (3) highlyinduced/inhibited (++/−− ≥10- and <100-fold change), or (4) veryhighly induced/inhibited (+++/−−− ≥100-fold change) (Li et al. 2012).A minus sign represents higher expressions of GmEXPBs in controlconditions than in nutrient deficiency treatment

2812 Appl Microbiol Biotechnol (2014) 98:2805–2817

shoots. In addition, GmEXPB1 expression was prominentlydecreased in roots. Interestingly, in response to Fe deficiency,six GmEXPBs appeared negatively regulated in either roots orshoots, and none were up-regulated (Table 2 and SupplementaryFig. S2). GmEXPB2, GmEXPB3, and GmEXPB5 , all in class I,showed no responses to Fe deficiency.

Expression of GmEXPBs in response to IAA and 6-BA

Previous studies showed that GmEXPB1 (Cim1 ) is acytokinin-regulated soybean pollen allergen gene (Crowell1994; Downes and Crowell 1998; Downes et al. 2001), andGmEXPB2 is induced by 0.5-μM IAA (Guo et al. 2011). Inorder to examine the responses of other GmEXPBs to IAAand 6-BA, soybean seedlings were grown in nutrient solutioncontaining 0.5-μM IAA or 0.5-μM 6-BA for 4 days.

As shown in Fig. 3a, four class I GmEXPBs responded toIAA treatment. All were induced, with GmEXPB1 ,GmEXPB3 , and GmEXPB6 showing increased expressions inshoots, and GmEXPB2 being induced in roots. On the otherhand, significantly decreased expression in response to IAAwas also observed. Three genes, GmEXPB4 , GmEXPB7 , andGmEXPB9 , from class II was decreased in shoots, and twogenes, GmEXPB1 and GmEXPB6 , from class I was inhabitedin roots. In response to 6-BA treatment, eight GmEXPBs hadaltered expressions (Fig.3b). In the 6-BA-treated roots, ex-pressions of GmEXPB1 , GmEXPB2 , and GmEXPB6 wereincreased dramatically, while GmEXPB4 was increasedslightly. In the 6-BA-treated shoots, expressions of all thegenes in class II, GmEXPB4 , GmEXPB7 , GmEXPB8 , andGmEXPB9 decreased while the expression of GmEXPB3increased from undetectable to low expression. GmEXPB5

Rel

ativ

e ex

pres

sion

val

ue

GmEXPB240

0

30

20

10

5.0

0

4.5

3.0

1.5

GmEXPB324

0

18

12

6

GmEXPB4

8

0

6

4

2

GmEXPB6 20

0

15

10

5

GmEXPB720

0

15

10

5

GmEXPB820

0

15

10

5

GmEXPB9

CK IAA CK IAA CK IAACK IAA

GmEXPB1

GmEXPB4160

0

120

80

40

GmEXPB3

0

1.6

1.2

0.8

0.4

Rel

ativ

e ex

pres

sion

val

ue

CK 6-BA CK 6-BA CK 6-BA CK 6-BA

60

0

45

30

15

GmEXPB2

GmEXPB61600

0

1200

800

400

GmEXPB7100

0

75

50

25

0

80

60

40

20

GmEXPB8 GmEXPB9100

0

75

50

25

B

A

1200

0

900

600

300

GmEXPB1

40

0

30

20

10

Fig. 3 Expression analysis ofGmEXPBs after treating withIAA (a) and 6-BA (b). Four-day-old soybean seedlings weretreated with 0.5-μM IAA or0.5-μM 6-BA for 4 days, and theseedlings without phytohormonetreatments were used as controls(CK). Roots and shoots wereseparately sampled for qRT-PCRanalysis.White bars stand for thevalues from roots and black barsfor shoots. GmEXPB5 was notdetected in both roots and shoots,and therefore not present here.Expression values were relative tothe housekeeping gene TefS1 .Each bar represents the mean ofthree biological replications withstandard error

Appl Microbiol Biotechnol (2014) 98:2805–2817 2813

was not detected in either shoots or roots treated with eitherIAA or 6-BA.

Discussion

Plant cell walls have important functions in cell division andexpansion, as well as in plant structural integrity. Some reportshave found that the plant cell wall is also involved in responseto environmental stimuli (Yen et al. 2001). However, relatedinformation is scarce. Previously, we found that GmEXPB2functioned as an intrinsic component of root architecturalchanges to abiotic stress in soybean (Guo et al. 2011). In thepresent study, the GmEXPB family was studied in a moresystematic and comprehensive fashion.

A total of nine soybean β-expansin genes were identified ina genome-wide search based on sequence similarity. TheseGmEXPBs can be categorized into two distinct classes basedupon phylogenetic analysis including β-expansins from mono-cot and dicot species (Fig. 1c), as well as by variations in geneintron phases (Fig. 1b), which has been associated with varia-tions in the structure and function of corresponding proteins(Ruvinsky and Watson 2007). Class I contains GmEXPBs 1 , 2 ,3 , 5 , and 6 , while class II includes GmEXPBs 4 , 7 , 8 , and 9 .This classification is supported by expression profiles, withparticularly consistent responses in group II GmEXPBs inresponse to signaling molecules (Fig. 3). The sum of theseresults suggests that specific biological functions might be cate-gorized in two groups in accordance with the gene classification,though further experiments are required to test this hypothesis.

The qRT-PCR results with nine GmEXPBs from six soybeantissues in this study exhibited variations in expression patternsamong GmEXPBs that confirms expression patterns reportedpreviously in global transcription studies using next generationsequencing (Libault et al. 2010; Severin et al. 2010). However,there are a couple of points of differences. GmEXPB8 andGmEXPB9 transcript levels in leaves were completely oppositeto the results from Libault et al. (2010) and Severin et al. (2010).In the current study, GmEXPB8 was nearly undetectable in the28-day-old leaves (Supplementary Fig. S1), but it expressed inthe shoots after 14 days under nutrient stresses or 4 days withhormone treatments (Supplementary Fig. S2, Fig. 3). In contrast,it was detected in the leaves tested by Libault et al. (2010) andSeverin et al. (2010). GmEXPB9 was most abundant in theleaves in this study (Supplementary Fig. S1), but was not detect-ed in the leaves by Libault et al. (2010) or Severin et al. (2010).The different expression patterns found for the twoGmEXPBs inleaves may be caused by using different soybean varieties, orsampling at different developmental stages, procedures or bias,treatments, or growth conditions. Notwithstanding these specificdifferences, GmEXPBs expression patterns appear to be stableand reproducible, and have been largely confirmed in threeindependent studies. This study contributes by focusing on

GmEXPBs, whereas the previous reports dealt with globaltranscriptome profiles.

Expression profiles forGmEXPBs suggest potential roles forindividual members. GmEXPB3 and GmEXPB5 were foundnearly exclusively in flowers and seeds, respectively, whichimplied that they are generative specific β-expansin genes andplay roles in pollen and young seed development. On the otherhand, GmEXPB1 and GmEXPB2 appear to be vegetative spe-cific and involved in cell division and elongation during root,leaf, and pod growth and development. Furthermore, themRNA accumulation of GmEXPB4 , GmEXPB6 , GmEXPB7 ,GmEXPB8 , andGmEXPB9 among the six tissues demonstratesthat they are expressed in generative and vegetative tissues,which indicates that each of these genes might function in thedevelopment of several organs. Further experiments with for-ward and reverse genetic approaches are needed to test theseconclusions and to determine specific roles for each GmEXPB .

The current work also reveals that expansions mediateplant hormone effects on growth through regulation of cellwall-loosening and cell elongation. Several publications sup-port this conclusion. For example, maize EXPB2 plays a rolein the elongation of maize roots regulated by indole-3-aceticacid (Kam et al. 2005). Also, expression of rice β-expansingenes has been correlated with rapid elongation of deepwaterrice internodes regulated by GA treatment (Lee and Kende2001), andGmEXPB1 (Cim1) is a cytokinin regulated soybeanpollen allergen gene (Crowell 1994; Downes and Crowell1998; Downes et al. 2001). Finally, GmEXPB2 was inducedin the previous work by treating plants with 0.5-μM IAA (Guoet al. 2011). In this study, the transcription levels of all identi-fied soybean GmEXPBs were studied in response to IAA and6-BA. The results show that other than GmEXPB5 which isseed-specific, all GmEXPBs are either up-regulated or down-regulated by 0.5-μM IAA or 0.5-μM 6-BA in roots or shoots.Furthermore, analysis of the promoters of GmEXPBs revealshormone-related cis-elements (Supplementary Table S4). Theexistence of these elements in association with all nineGmEXPBs and their potential roles in regulating gene expres-sion indicate that signaling molecules, such as IAA and 6-BA,are important regulators of soybean β-expansin genes through-out plant growth and development.

One set of factors that affect plant growth and development isabiotic stress, including nutrient deprivation. Stress-tolerantplants develop various strategies to survive in adverse conditions,including regulation of genes controlling cell development andmetabolism. Expansins have been isolated and characterizedunder stress conditions, and several studies have indicated thatexpansins are associated with plant adaptation to environmentalstresses by regulation of cell growth and extension (Geilfus et al.2010; Noh et al. 2009;Wu et al. 1996; Xu et al. 2007). However,there is little known about expansin responses to nutrient defi-ciency. Here, we investigated the transcription levels ofGmEXPBs in response to withholding of several nutrients. Our

2814 Appl Microbiol Biotechnol (2014) 98:2805–2817

results showed variations among GmEXPBs in response to eachof the nutrient stress. This suggests that not only are expansinproteins involved in altered growth and development underabnormal conditions, but that roles vary according to thenutrient and expansin involved. GmEXPB2 is highly inducedby −P treatment, which is consistent with previous results(Guo et al. 2011). Except for GmEXPB5, which was notconsistently detected in soybean roots and shoots, all of theother GmEXPBs also responded to deficiencies of N, P, K,or Fe. Interestingly, six GmEXPBs were negatively regulatedunder Fe starvation conditions, and Fe deficiency is knownto affect critical enzymes in plant growth and extension(Briat et al. 2007). The current results with Fe, along withN, P, and K, demonstrate that GmEXPBs play key rolesunder stress conditions and further understanding of themolecular mechanisms of nutrient stress responses associatedwith expansins may lead to the development of novel strat-egies for improving soybean productivity.

Another area where plant growth and development is alteredis in symbiotic associations. Many legumes respond toRhizobium inoculation by developing new organs known asnodules. Due to expansin mediation of cell wall extension, theyare likely to be involved in the growth of various plant organs,including nodules. Several reports have associated expansinswith root nodule symbiosis. The expression of MaEXP1 , aMelilotus alba expansin gene, is enhanced in roots within hoursafter inoculation with Sweetclover—Sinorhizobium meliloti , aswell as in nodules (Giordano and Hirsch 2004). An expansingene identified as specific for pea nodules, was localized innodule apoplasts, as well as in the infection thread walls(Sujkowska et al. 2007). Our results showed that all nineGmEXPBs are expressed in soybean nodules to varying de-grees (Fig. 2a), and each might participate in nodule morpho-genesis. None appear specific for nodules. Rather, nodulation insoybean appears to proceed through cooption and redirection ofcell wall extension pathways involving GmEXPBs that areotherwise utilized in many tissues and growth phases.

Another symbiotic association that is not specific to legumesinvolves AMF. Many vascular plants are able to develop symbi-otic associations with AMF. It has been reported that an expansinhas been localized to cucumber root cells inoculated with AMF,and proposed that this wall-loosening protein is directly involvedin the accommodation of the fungus by infected cortical cells(Balestrini et al. 2005). In addition, partial transcriptional silenc-ing of EXLB from tomato was correlated with a reduction inAMF spore formation in transgenic roots (Dermatsev et al.2010). Our results support the conclusion that EXPBs are in-volved in mycorrhizal formation. Two GmEXPBs in this studyhad significantly higher transcript levels in response to AMinfection compared to expression in control roots (Fig. 2c).Four other soybean expansin genes had altered expression inresponse to AMF inoculation, but the responses were belowthe threshold for significance. Conclusive evidence of the

number of soybean expansions involved in mycorrhizal sym-biosis and the mechanisms of their involvement requiresfurther study focused on this specific topic.

In summary, systematic and comprehensive isolation andcharacterization of the soybean GmEXPBs were performed,along with descriptions of their molecular features, expressionpatterns in vegetative and reproductive tissues, and responsive-ness to nutrient deficiency, hormone treatments, and symbioticassociations. The results reveal the existence of nine GmEXPBsthat can be divided into two distinct subgroups based on se-quence similarity and gene structural features, which is furthersupported by associations with specific cis-elements andexpression profiles. Some GmEXPBs are largely expressedin generative organs, and others are primarily found in vegeta-tive tissues. Eight of the nine soybean expansin genes wereshown to be differentially responsive to nutrient deficiencies orphytohormones. All GmEXPBs were expressed in nodules,though three predominated, and two were enhanced in AMroots under P-limited conditions. Further understanding of themolecular mechanisms of expansins responses to nutrientstresses, phytohormones, and symbiosis with Rhizobium orAMF may lead to insights to improving soybean production.

Acknowledgments This work was jointly supported by the grants fromthe National Natural Science Foundation of China (Grant No. 31000931and 31025022) and the National Key Basic Research Special Funds ofChina (2011CB100301). We thank Mr. Chengchen Li, Mr. QiangqiangCui andMr. YongxiangGuo for their sampling assistance.We are gratefulto Ms. Lu Qin for help with bioinformatic analysis.

References

Balestrini R, Cosgrove DJ, Bonfante P (2005) Differential location ofα -expansin proteins during the accommodation of root cells toan arbuscular mycorrhizal fungus. Planta 220(6):889–899

Briat JF, Curie C, Gaymard F (2007) Iron utilization and metabolism inplants. Curr Opin Plant Biol 10(3):276–282

Carpita NC, Gibeaut DM (1993) Structural models of primary cell wallsin flowering plants: consistency of molecular structure with thephysical properties of the walls during growth. Plant J 3(1):1–30

Chen CH, Chen ZX (2002) Potentiation of developmentally regulatedplant defense response by AtWRKY18, a pathogen-inducedArabidopsis transcription factor. Plant Physiol 129(2):706–716

Cheng FX,CaoGQ,WangXR, Zhao J, YanXL, LiaoH (2009) Isolation andapplication of effective nitrogen fixation rhizobial strains on low-phosphorus acid soils in South China. Chinese Sci Bull 54(3):412–420

Choi D, ChoHT, Lee Y (2006) Expansins: expanding importance in plantgrowth and development. Physiol Plantarum 126(4):511–518

Cosgrove DJ (1989) Characterization of long-term extension of isolatedcell-walls from growing cucumber hypocotyls. Planta 177(1):121–130

Cosgrove DJ (1998) Cell wall loosening by expansins. Plant Physiol118(2):333–339

Cosgrove DJ (2000) Loosening of plant cell walls by expansins. Nature407(6802):321–326

Cosgrove DJ, Bedinger P, Durachko DM (1997) Group I allergens ofgrass pollen as cell wall-loosening agents. Proc Natl Acad Sci U S A94(12):6559–6564

Appl Microbiol Biotechnol (2014) 98:2805–2817 2815

Crowell DN (1994) Cytokinin regulation of a soybean pollen allergengene. Plant Mol Biol 25(5):829–835

Dermatsev V, Weingarten-Baror C, Resnick N, Gadkar V, Wininger S,Kolotilin I, Mayzlish-Gati E, Zilberstein A, Koltai H, Kapulnik Y(2010) Microarray analysis and functional tests suggest the involve-ment of expansins in the early stages of symbiosis of the arbuscularmycorrhizal fungus Glomus intraradices on tomato (Solanumlycopersicum). Mol Plant Pathol 11(1):121–135

Downes BP, Crowell DN (1998) Cytokinin regulates the expression of asoybean β-expansin gene by a post-transcriptional mechanism.Plant Mol Biol 37(3):437–444

DownesBP, Steinbaker CR, Crowell DN (2001) Expression and processingof a hormonally regulated β-expansin from soybean. Plant Physiol126(1):244–252

Gao Q, Zhao MR, Li F, Guo QF, Xing SC, Wang W (2008) Expansinsand coleoptile elongation in wheat. Protoplasma 233(1):73–81

Geilfus CM, Zorb C, Muhling KH (2010) Salt stress differentially affectsgrowth-mediating β-expansins in resistant and sensitive maize(Zea mays L.). Plant Physiol Bioch 48(12):993–998

Giordano W, Hirsch AM (2004) The expression of MaEXP1, a Melilotusalba expansin gene, is upregulated during the sweetclover-Sinorhizobium meliloti interaction. Mol Plant Microbe In 17(6):613–622

Guo WB, Zhao J, Li XX, Qin L, Yan XL, Liao H (2011) A soybean β-expansin gene GmEXPB2 intrinsically involved in root systemarchitecture responses to abiotic stresses. Plant J 66(3):541–552

Kam MJ, Yun HS, Kaufman PB, Chang SC, Kim SK (2005) Twoexpansins, EXP1 and EXPB2, are correlated with the growth anddevelopment of maize roots. J Plant Biol 48(3):304–310

Kwasniewski M, Szarejko I (2006) Molecular cloning and characteriza-tion ofβ-expansin gene related to root hair formation in barley. PlantPhysiol 141(3):1149–1158

Kwon YR, Lee HJ, Kim KH, Hong SW, Lee H (2008) Ectopicexpression of Expansin3 or Expansin β1 causes enhancedhormone and salt stress sensitivity in Arabidopsis . BiotechnolLett 30(7):1281–1288

Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA,McWilliam H, Valentin F, Wallace IM,Wilm A, Lopez R, ThompsonJD, Gibson TJ, Higgins DG (2007) Clustal W and clustal X version2.0. Bioinformatics 23(21):2947–2948

Lee Y, Kende H (2001) Expression of β-expansins is correlated withinternodal elongation in deepwater rice. Plant Physiol 127(2):645–654

Li CC, Gui SH, Yang T, Walk T, Wang XR, Liao H (2012) Identificationof soybean purple acid phosphatase genes and their expressionresponses to phosphorus availability and symbiosis. Ann Bot 109(1):275–285

Li F, Xing SC, Guo QF, Zhao MR, Zhang J, Gao Q, Wang GP, Wang W(2011) Drought tolerance through over-expression of the expansingene TaEXPB23 in transgenic tobacco. J Plant Physiol 168(9):960–966

Li LC, Cosgrove DJ (2001) Grass group I pollen allergens (β-expansins)lack proteinase activity and do not causewall loosening via proteolysis.Eur J Biochem 268(15):4217–4226

Li Y, Darley CP, Ongaro V, Fleming A, Schipper O, Baldauf SL,McQueen-Mason SJ (2002) Plant expansins are a complexmultigene family with an ancient evolutionary origin. Plant Physiol128(3):854–864

Libault M, Farmer A, Joshi T, Takahashi K, Langley RJ, Franklin LD, HeJ, Xu D, May G, Stacey G (2010) An integrated transcriptome atlasof the crop model Glycine max , and its use in comparative analysesin plants. Plant J 63(1):86–99

Liu Y, Liu DC, Zhang HY, Gao HB, Guo XL, Wang DM, Zhang XQ,Zhang AM (2007) The α- and β-expansin and xyloglucanendotransglucosylase/hydrolase gene families of wheat: molecularcloning, gene expression, and EST data mining. Genomics 90(4):516–529

Magneschi L, Kudahettige RL, Alpi A, Perata P (2009) Expansin geneexpression and anoxic coleoptile elongation in rice cultivars. J PlantPhysiol 166(14):1576–1580

McQueen-Mason S, Cosgrove DJ (1994) Disruption of hydrogen bond-ing between plant cell wall polymers by proteins that induce wallextension. Proc Natl Acad Sci U S A 91(14):6574–6578

McQueen-Mason S, Durachko DM, Cosgrove DJ (1992) Two endoge-nous proteins that induce cell wall extension in plants. Plant Cell4(11):1425–1433

McQueen-Mason SJ, Cosgrove DJ (1995) Expansin mode of action oncell walls—analysis of wall hydrolysis, stress relaxation, and bind-ing. Plant Physiol 107(1):87–100

McQueen-Mason SJ, Fry SC, Durachko DM, Cosgrove DJ (1993) Therelationship between xyloglucan endotransglycosylase and in-vitrocell wall extension in cucumber hypocotyls. Planta 190(3):327–331

Noh SA, Park SH, Huh GH, Paek KH, Shin JS, Bae JM (2009) Growthretardation and differential regulation of expansin genes in chilling-stressed sweet potato. Plant Biotechnol Rep 3(1):75–85

Park HC, Kim ML, Kang YH, Jeon JM, Yoo JH, Kim MC, Park CY,Jeong JC, Moon BC, Lee JH, Yoon HW, Lee SH, Chung WS, LimCO, Lee SY, Hong JC, ChoMJ (2004) Pathogen- and NaCl-inducedexpression of the SCaM-4 promoter is mediated in part by a GT-1box that interacts with a GT-1-like transcription factor. Plant Physiol135(4):2150–2161

Ruvinsky A, Watson C (2007) Intron phase patterns in genes: preserva-tion and evolutionary changes. Open Evol J 1(1):1–14

Sampedro J, Cosgrove DJ (2005) The expansin superfamily. GenomeBiol 6(12):242

Sun QX, Lin Z, Ni ZF, Zhang Y, Yao YY, Wu HY (2005) Isolation andcharacterization of 18 genes encoding α- and β- expansins in wheat(Triticum aestivum L.). Mol Genet Genomics 274(5):548–556

Severin AJ, Woody JL, Bolon YT, Joseph B, Diers BW, Farmer AD,Muehlbauer GJ, Nelson RT, Grant D, Specht JE, Graham MA,Cannon SB, May GD, Vance CP, Shoemaker RC (2010) RNA-Seq Atlas of Glycine max : A guide to the soybean transcriptome.BMC Plant Biol 10:160

Shin JH, Jeong DH, Park MC, An G (2005) Characterization and tran-scriptional expression of the α-expansin gene family in rice. MolCells 20(2):210–218

Sujkowska M, Borucki W, Golinowski W (2007) Localization ofexpansin-like protein in apoplast of pea (Pisum sativum L.) rootnodules during interaction with Rhizobium leguminosarum bv.Viciae 248. Acta Soc Bot Pol 76(1):17–26

Tabuchi A, Li LC, CosgroveDJ (2011)Matrix solubilization and cell wallweakening by β -expansin (group-1 allergen) from maize pollen.Plant J 68(3):546–559

Tamura Y, Kobae Y, Mizuno T, Hata S (2012) Identification and expres-sion analysis of arbuscular mycorrhiza-inducible phosphate trans-porter genes of soybean. Biosci Biotech Bioch 76(2):309–313

Tapia G, Verdugo I, Yañez M, Ahumada I, Theoduloz C, Cordero C,Poblete F, González E, Ruiz-Lara S (2005) Involvement of ethylenein stress-induced expression of the TLC1 .1 retrotransposon fromLycopersicon chilense Dun. Plant Physiol 138(4):2075–2086

Won SK, Choi SB, Kumari S, ChoM, Lee SH, Cho HT (2010) Root hair-specific EXPANSIN B genes have been selected for Graminaceaeroot hairs. Mol Cells 30(4):369–376

Wu YJ, Meeley RB, Cosgrove DJ (2001) Analysis and expression of theα-expansin and β-expansin gene families in maize. Plant Physiol126(1):222–232

Wu YJ, Sharp RE, Durachko DM, Cosgrove DJ (1996) Growthmaintenance of the maize primary root at low water poten-tials involves increases in cell-wall extension properties,expansin activity, and wall susceptibility to expansins. PlantPhysiol 111(3):765–772

Xie Z, Zhang ZL, Zou XL, Huang J, Ruas P, Thompson D, Shen QJ(2005) Annotations and functional analyses of the riceWRKY gene

2816 Appl Microbiol Biotechnol (2014) 98:2805–2817

superfamily reveal positive and negative regulators of abscisic acidsignaling in aleurone cells. Plant Physiol 137(1):176–189

Xing SC, Li F, Guo QF, Liu DR, Zhao XX, Wang W (2009) Theinvolvement of an expansin gene TaEXPB23 from wheat in regu-lating plant cell growth. Biol Plantarum 53(3):429–434

Xu J, Tian J, Belanger FC, Huang B (2007) Identification and character-ization of an expansin gene AsEXP1 associated with heat tolerancein C3 Agrostis grass species. J Exp Bot 58(13):3789–3796

Yen SK, Chung MC, Chen PC, Yen HE (2001) Environmentaland developmental regulation of the wound-induced cell wallprotein WI12 in the halophyte ice plant. Plant Physiol127(2):517–528

Zhang ZL, Xie Z, Zou XL, Casaretto J, Ho THD, Shen QXJ (2004) Arice WRKY gene encodes a transcriptional repressor of the gib-berellin signaling pathway in aleurone cells. Plant Physiol 134(4):1500–1513

Appl Microbiol Biotechnol (2014) 98:2805–2817 2817