lable at ScienceDirect

Plant Physiology and Biochemistry 46 (2008) 1015–1018

Contents lists avai

Plant Physiology and Biochemistry

journal homepage: www.elsevier .com/locate/plaphy

Short communication

Expression of glycine-rich protein genes, AtGRP5 and AtGRP23, induced bythe cutin monomer 16-hydroxypalmitic acid in Arabidopsis thaliana

Jong Ho Park a, Mi Chung Suh b, Tae Hyun Kim a, Moon Chul Kim a, Sung Ho Cho a,*

a Department of Biological Sciences, Inha University, Incheon 402-751, Republic of Koreab Department of Plant Biotechnology, Chonnam National University, Gwangju 500-757, Republic of Korea

a r t i c l e i n f o

Article history:Received 1 August 2007Accepted 3 June 2008Available online 20 June 2008

Keywords:Arabidopsis thalianaCutin monomerElicitorGlycine-rich proteinHydrogen peroxidePathogenesis

* Corresponding author. Tel.: þ82 32 860 7696; faxE-mail address: shcho@inha.ac.kr (S.H. Cho).

0981-9428/$ – see front matter � 2008 Published bydoi:10.1016/j.plaphy.2008.06.008

a b s t r a c t

Glycine-rich proteins (GRPs) belong to a large family of heterogenous proteins that are enriched inglycine residues. The expression of two GRP genes of Arabidopsis thaliana, AtGRP5 and AtGRP23, wasinduced by 16-hydroxypalmitic acid (HPA), a major component of cutin. The expression of AtGRP3, whichencodes a GRP protein that is structurally different from AtGRP5 and AtGRP23, was not responsive to HPAapplication. Treatment with HPA also induced expression of the pathogen-related PR-1 and PR-4 genes.Abscisic acid and salicylic acid treatments enhanced the transcript levels of AtGRP5 and AtGRP23 as wellas those of AtGRP3. It was also demonstrated that HPA effectively elicited the accumulation of H2O2 inrosette leaves of Arabidopsis. Results suggest the possible role of some species of GRPs, such as AtGRP5and AtGRP23, in response to the pathogenic invasion mediated by cutin monomers in plants.

� 2008 Published by Elsevier Masson SAS.

1. Introduction

The glycine-rich proteins (GRPs) compose a large family ofheterogenous proteins that are enriched in glycine residues atvarious proportions, occupying from 20% to 70% of the total aminoacid residues of the protein [16,21]. GRPs can be classified intoseveral groups based on their molecular structure, i.e. the contentof glycine residues and the presence of a signal peptide or RNA-binding sequence, and are thought to be involved in variousphysiological activities [8,16,21].

Members of a group of GRPs have a signal peptide with a lowmolecular mass of 10–20 kDa, and some of these GRPs have beensuggested to play a role in cell wall reinforcement or in signaltransduction of pathogen-induced defense responses [5,12–15,17,20,27]. Two of the well-studied Arabidopsis GRPs that belongto this group are AtGRP3 and AtGRP5. While AtGRP3 is mainlyexpressed in leaves and stems, AtGRP5 is more abundantlyexpressed in immature seedpods and only weakly in leaves [5].Both genes are responsive to ethylene, abscisic acid (ABA), andsalicylic acid (SA) treatments [5]. However, their functions seem tobe somewhat different from each other. AtGRP3 is believed to bea secreted protein that is located in the cell wall and has been

: þ82 32 874 6737.

Elsevier Masson SAS.

shown to bind to the extracellular domain of Wak1, a cell wallassociated receptor kinase, and it has been suggested that theinteraction of AtGRP3 with Wak1 occurs through a pathogenesis-related process [17]. AtGRP5, on the contrary, has been suggested tobe an intracellular membrane-associated protein and involved incell reinforcement to counter mechanical tensions [20]. Promoterassay of AtGRP5 revealed that it is mainly expressed in theepidermal or protodermal derived tissues [20].

Cutin monomers have been found to play a role as elicitors insignal transduction for plant defenses against fungal pathogens [2].Cutin monomers are released by the action of fungal cutinase at thetime of infection and were suggested as signals for fungal geneactivation [10]. Later evidence indicated that cutin monomers arealso involved in the defense strategies of plants. Treatment of leavesof barley and rice with cutin monomers enhanced resistanceagainst fungal pathogens [22,23]. It was also demonstrated thatcutin monomers effectively worked as H2O2 elicitors in cucumberand rice leaves [6,9], suggesting their role in the defensemechanism [9].

In this study the expression pattern of three GRP genes ofArabidopsis, AtGRP3, AtGRP5 and AtGRP23, was investigated inresponse to the treatment of leaves with a cutin monomer, as oneapproach to assess the roles of the GRPs and cutin monomers inresponse to pathogenic attack. We found that the expressions ofAtGRP5 and AtGRP23 are induced by 16-hydroxypalmitic acid (HPA),a major component of cutin. AtGRP5 and AtGRP23 were also induced

J.H. Park et al. / Plant Physiology and Biochemistry 46 (2008) 1015–10181016

by abscisic acid (ABA) and salicylic acid (SA) treatments. Resultssuggest the possible role of some species of GRPs in response topathogenic invasion mediated by a cutin monomer in plants.

2. Results

2.1. Sequence analysis

AtGRP5 was previously reported and found to be expressed inleaves and stems, and showed extensive expression in immatureseedpods [5]. AtGRP23 is a novel gene that has never been char-acterized before, and was numbered accordingly, since the latestAtGRP gene to be reported was AtGRP22 [7]. AtGRP23 is closely re-lated to AtGRP5 in structure, and their deduced amino acid se-quences are aligned with comparison to those of AtGRP3 (Fig. 1).The glycine contents of AtGRP5 and AtGRP23 are 66% and 63%,respectively, while those of AtGRP3 is only 38%, in the predictedmature polypeptides. The multiple tandem repeat of G3X or G5Xwas found throughout nearly the entire polypeptide sequence ofAtGRP5 and AtGRP23 except for the N-terminal region, but not inthe case of AtGRP3.

2.2. Elicitation of H2O2 by HPA

HPA is one of the major components of cutin in plant tissues[11,26]. It has been revealed that several cutin monomers, includingHPA, and their derivatives effectively induced the accumulation ofH2O2 in cucumber leaves [6] and in rice [9]. To determine whetherHPA would also work as an H2O2 elicitor in Arabidopsis, theproduction of H2O2 was first confirmed using in situ the dia-minobenzidine (DAB)-staining method [19,27]. DAB is polymerizedinstantly and locally upon contact with H2O2 in the presence ofperoxidase. As shown in Fig. 2, HPA induced the accumulation ofH2O2 in rosette leaves and is visualized as reddish-brown stainingin the veins and adjacent areas. The degree of staining was strongerwhen 800 mM HPA was applied, compared with 200 mM HPA.Control leaf tissue treated with liquid paraffin alone exhibited onlyweak sign of DAB staining.

2.3. Expression studies

HPA induced the expression of AtGRP5 and, to a lesser degree,that of AtGRP23 (Fig. 3). Both 200 mM HPA and 800 mM HPA werevery effective in the induction. The expression of both AtGRP5 andAtGRP23 increased from 4 h, and remained until at least 12 h afterthe cutin monomer treatment. AtGRP3, however, showed nosignificant response during the time period of the experiment,although AtGRP3 was implied to be involved in the defense-relatedresponse [17]. Treatment with the cutin monomer also effectively

Fig. 1. Alignment of the amino acid sequences derived from AtGRP5, AtGRP23, and AtGRP3indicated in gray boxes. Multiple tandem repeats of G3X and G5X motifs were found in AAtGRP23, and AtGRP3 are AY045629, AAC04494, and AAB24075, respectively.

induced expression of the pathogen-related PR-1 and PR-4 genes(Fig. 3).

The expressions of AtGRP5 and AtGRP23 were also enhanced bySA and ABA treatments (Fig. 4). Both genes exhibited elevatedtranscript levels until 24 h in response to SA. ABA induced theexpressions of AtGRP5 and AtGRP23 at 4 h, and the expression levelsdiminished thereafter. AtGRP3 was also found to be responsive toboth SA and ABA (Fig. 4; [5]). In earlier studies, AtGRP5 was shownto be also induced by ethylene, drought, and flooding [5]. Since SAmediates systemic acquired resistance to pathogens [18], it is highlylikely that AtGRP5 and AtGRP23, as well as AtGRP3, somehow takepart in the defense mechanisms.

3. Discussion

The enhanced expressions of AtGRP5 and AtGRP23 by HPA sug-gest that these genes may be involved in the defense mechanism ofArabidopsis against pathogens mediated by cutin monomers. HPAand other cutin monomers induced resistance against a fungalpathogen, Erysiphe graminis f. sp. hordei, in barley leaves [22].Treatments of plants with cutin monomers have been shown toinduce pathogen-related genes in several cases. In cultured potatocells, cutin monomers induced rapid alkalinization of the culturemedium, accompanied by the transcriptional activation of somedefense-related genes [24]. We also demonstrated that HPAinduced the expression of a lipid transfer protein gene OsLTP5 alongwith a pathogen-related PR-10 gene [9].

Recently, Chassot et al. [3] showed that altered cuticularstructure in A. thaliana plants expressing a cell wall targetedcutinase (CUTE plants), or of a bdg mutant, provides full immunityto Botrytis cinerea, a fungal pathogen that causes necrotic lesions onwild-type (WT) plants. The bre1 mutant that lacks long-chain acyl-CoA synthetase 2 activity was also resistant to B. cinerea andScleotiorum sclerotinia, and it was demonstrated that an increasedpermeability of the cuticle layer facilitated the perception ofputative elicitors [1]. However, resistance of CUTE plants wasindependent of the pathways involving SA, ethylene, or jasmonicacid, and no induction of pathogenesis-related genes took placeduring infection with B. cinerea, unlike WT plants [3]. Possiblyplants with a disturbed cuticle structure adopt an alternativestrategy in response to pathogenic invasion that does not involvethe conventional signaling pathways operating in WT plants.

The manner by which GRPs contribute to the mechanism ofdefense against pathogenic attack remains in question. AtGRP3 hasbeen shown to bind to the extracellular domain of Wak1, a cell wallassociated receptor kinase, and it has been suggested that theinteraction of AtGRP3 with Wak1 occurs through a pathogenesis-related process [17]. In tobacco, a cadmium-induced GRP (ciGRP)protein exerted its inhibitory effects against tobamovirus by

genes of Arabidopsis. Amino acid residues homologous in at least two sequences aretGRP5 and AtGRP23, but not in AtGRP3. The GenBank accession numbers of AtGRP5,

Fig. 2. In situ staining of H2O2 generated by HPA treatment in rosette leaves of Arabidopsis. HPA was dissolved in chloroform–methanol (6:1, v/v), dried under N2, and applied to theupper surface of leaves with a spatula in liquid paraffin at 200 mM or 800 mM. Liquid paraffin alone was applied to leaves as a mock treatment. After 12 h, H2O2 was detected byimmersing the leaf bases in a DAB solution (1 mg/ml, pH 3.8) for 8 h under light at 25 �C. Production of H2O2 is visualized as a reddish-brown deposit in the epidermis and veins.

J.H. Park et al. / Plant Physiology and Biochemistry 46 (2008) 1015–1018 1017

enhancing callose deposits in the vasculature, thus blocking thesystemic movement of the virus [28]. A similar callose depositionwas observed with an increase of LsGRP1 gene expression againsta fungal pathogen, Botrytis elliptica, in lily plants [14]. The structuraldifference of AtGRP5 and AtGRP23 from these proteins, however,implies that the function of AtGRP5 and AtGRP23 may not berelated to that of AtGRP3, ciGRP, or LsGRP1. AtGRP5 and AtGRP23belong to the group of GRPs with the highest content of glycineresidues, reaching over 60% of the total amino acid residues of theprotein, while glycine residues occupy less than 40% of the residuesin AtGRP3, ciGRP and LsGRP1. The multiple tandem repeat of G3X orG5X that is found throughout almost the entire region of AtGRP5and AtGRP23 lacks in these proteins where the repeat is somewhatirregular. Moreover, AtGRP5 was suggested to be an intracellularmembrane-associated protein [20], while others are presumablylocated in the extracellular matrix. A model was proposed toexplain the structure of AtGRP5, in which loops of glycine residuesare organized much like Velcro, constituting the basis for protein–protein interactions [25]. In other model, the glycine-rich domainsare arranged in b-pleated sheets, with the bulky hydrophobic sidechains exposed to one side of the b-sheet and only the hydrogen

0 12 h 12 h 4 h 4 h 4 h 12 h

AtGRP23

AtGRP5

AtGRP3

PR-1

PR-4

HPA200 µM 800 µM Mock

-Tubulin

Fig. 3. The effects of HPA treatment on the expressions of AtGRP5, AtGRP23, andAtGRP3 in rosette leaves of Arabidopsis. The expression of the b-tubulin gene was usedas a loading control. HPA was applied as described in the legend for Fig. 2.

atoms of glycine residues on the other side [4]. Whether througha protein–protein communication or as a material for structuralreinforcement, it is evident from our results that AtGRP5 could playa role in conveying the signal from the pathogenic invasion to thecells.

4. Materials and methods

4.1. Plant materials and growth conditions

Arabidopsis thaliana (Columbia ecotype) plants were grown at22 �C under a 16 h photoperiod. Tissues were collected, placed inliquid N2, and stored at �70 �C prior to use.

4.2. Treatments of ABA, SA and HPA

One hundred micromolar ABA or 5 mM SA solution was sprayedonto rosette leaves. HPA (Sigma) was dissolved in chloroform–methanol (6:1, v/v), dried under N2, and applied to the uppersurface of leaves with a spatula in liquid paraffin at 200 mM or800 mM [9]. Liquid paraffin alone was applied to the leaves asa mock treatment. Extraction of RNA was conducted 4 h and 12 hafter treatments. All the experiments were repeated at least threetimes with similar results.

4.3. Reverse transcriptase-polymerase chain reaction (RT-PCR)

Using 2 mg total RNA as the template, the reverse transcription(RT) reaction was performed with M-MLV reverse transcriptase(Promega, USA) for 1 h at 42 �C, followed by 10 min at 65 �C. Fivemicroliters of the 20 ml RT reactions was used as a template for thePCR using forward primer, 50-ACATTCTTCCACCATCCAAGCGGA-30,and reverse primer, 50-CAATGATGTCCACCACCGAAACCA-30, to syn-thesize a 413 bp fragment of the AtGRP5 cDNA; forward primer,50-ACAAACCTCGTCCATTCCTCCACA-30, and reverse primer, 50-ATGCCTTTACCAAATCCACCACCG-30, to synthesize a 427 bp fragment ofthe AtGRP23 cDNA; and forward primer, 50-AGGCTTTGGTTCTGTTGGGTCTCT-30, and reverse primer, 50-TGAGCAGCCGTTG-TAACCTCTGTA-30, to synthesize a 377 bp fragment of the AtGRP3cDNA. The PCR mixture (25 ml) contained 500 ng template, 2.5 ml10� PCR buffer [100 mM Tris–HCl, pH 8.0, 50 mM KCl, 1 mM EDTA,0.1% Triton X-100, 50% glycerol (v/v)], 50 mM primers, 200 mM dNTPs,and 2.5 U of Ex Taq polymerase (TaKaRa, Japan). Reactions weredenatured for 40 s at 95 �C, annealed for 30 s at 64.4 �C (AtGRP5) or

AtGRP5

AtGRP23

-Tubulin

AtGRP3

SA ABA0 24 h 12 h 4 h 0 24 h 12 h 4 h

Fig. 4. RT-PCR analysis of AtGRP5, AtGRP23, and AtGRP3 expression in response to SA and ABA treatments in rosette leaves of Arabidopsis. One hundred micromolar ABA or 5 mM SAsolution was sprayed onto the rosette leaves.

J.H. Park et al. / Plant Physiology and Biochemistry 46 (2008) 1015–10181018

at 58 �C (AtGRP23 and AtGRP3), and extended for 90 s at 72 �C.Twenty-five cycles were conducted in all instances.

4.4. In situ detection of H2O2

Hydrogen peroxide was detected in situ using the DAB-uptakemethod [19,27]. Leaves treated with HPA were cut at the base andplaced in 1 mg/ml solution of 3,30-DAB–HCl, pH 3.8 (Sigma, USA),for 8 h under light at 25 �C. The leaves were boiled in 96% ethanolfor 10 min to remove chlorophyll, and were then stored in 96% coldethanol. H2O2 production is visualized as a reddish-brown colora-tion in tissues.

Acknowledgments

This research was supported by grants from theinterdisciplinary research program of KOSEF (Grant No. R01-2006-000-11056-0) and from BioGreen 21, Rural DevelopmentAdministration, Republic of Korea (Grant No. 2005-0401-034593).J.H. Park, T.H. Kim, and M.C. Kim were supported by the Brain Korea21 Program.

References

[1] M. Bessire, C. Chassot, A.C. Jacquat, M. Humphry, S. Borel, J.M. Petetot, J.-P.Metraux, C. Nawrath, A permeable cuticle in Arabidopsis leads to a strongresistance to Botrytis cinerea, EMBO J. 26 (2007) 2158–2168.

[2] C. Chassot, J.-P. Metraux, The cuticle as a source of signals for plant defense,Plant Biosyst. 139 (2005) 28–31.

[3] C. Chassot, C. Nawrath, J.-P. Metraux, Cuticular defects lead to full immunity toa major plant pathogen, Plant J. 49 (2007) 972–980.

[4] C.M. Condit, R.B. Meagher, A gene encoding a novel glycine-rich structuralprotein of petunia, Nature 323 (1986) 178–181.

[5] D.E. de Oliveira, J. Seurinck, D. Inze, M. van Montagu, J. Botterman, Differentialexpression of five Arabidopsis genes encoding glycine-rich proteins, Plant Cell2 (1990) 427–436.

[6] M. Fauth, P. Schweizer, A. Buchala, C. Mrkstadter, M. Riederer, T. Kato, H. Kauss,Cutin monomers and surface wax constituents elicit H2O2 in conditionedcucumber hypocotyl segments and enhance the activity of other H2O2elicitors, Plant Physiol. 117 (1998) 1373–1380.

[7] A. Fiebig, R. Kimport, D. Preus, Comparisons of pollen coat genes acrossBrassicaceae species reveal rapid evolution by repeat expansion anddiversification, Proc. Natl. Acad. Sci. U.S.A. 101 (2004) 3286–3291.

[8] A.F. Fusaro, S.N. Bocca, R.L.B. Ramos, R.M. Barroco, C. Magioli, V.C. Jorge, T.C.Coutinho, C.M. Rangel-Lima, R.D. Rycke, D. Inze, G. Engler, G. Sachetto-Martins,AtGRP2, a cold-induced nucleo-cytoplasmic RNA-binding protein, has a role inflower and seed development, Planta 225 (2007) 1339–1351.

[9] T.H. Kim, J.H. Park, M.C. Kim, S.H. Cho, Cutin monomer induces expression ofthe rice OsLTP5 lipid transfer protein gene, J. Plant Physiol. 165 (2008) 345–349.

[10] P.E. Kolattukudy, L.M. Rogers, D. Li, C.S. Hwang, M.A. Flaishman, Surfacesignaling in pathogenesis, Proc. Natl. Acad. Sci. U.S.A. 92 (1995) 4080–4087.

[11] L. Kunst, A.L. Samuels, R. Jetter, The Plant Cuticle: Formation and Structure ofEpidermal Surfaces, in: D.J. Murphy (Ed.), Plant Lipids: Biology, Utilisation andManipulation, Blackwell, 2005, pp. 270–302.

[12] W.-C. Lin, M.-L. Cheng, J.-W. Wu, N.-S. Yang, C.-P. Cheng, A glycine-rich proteingene family predominantly expressed in tomato roots, but not in leaves andripe fruit, Plant Sci. 168 (2005) 283–295.

[13] Z.-Z. Liu, J.-L. Wang, X. Huang, W.-H. Xu, Z.-M. Liu, R.-X. Fang, The promoter ofa rice glycine-rich protein gene, Osgrp-2, confers vascular-specific expressionin transgenic plants, Planta 216 (2003) 824–833.

[14] Y.Y. Lu, Y.H. Liu, C.Y. Chen, Stomatal closure, callose deposition, and increase ofLsGRP1-corresponding transcript in probenazole-induced resistance againstBotrytis elliptica in lily, Plant Sci. 172 (2007) 913–919.

[15] A. Molina, M. Mena, P. Carbonero, F. Garcia-Olmedo, Differential expression ofpathogen-responsive genes encoding two types of glycine-rich proteins inbarley, Plant Mol. Biol. 33 (1997) 803–810.

[16] A. Mousavi, Y. Hotta, Glycine-rich proteins: a class of novel proteins, Appl.Biochem. Biotechnol. 120 (2005) 169–174.

[17] A.R. Park, S.K. Cho, U.J. Yun, M.Y. Jin, S.H. Lee, G. Sachetto-Martins, O.K. Park,Interaction of the Arabidopsis receptor protein kinase Wak1 with a glycine-richprotein, AtGRP-3, J. Biol. Chem. 276 (2001) 26688–26693.

[18] I. Raskin, Role of salicylic acid in plants, Annu. Rev. Plant Physiol. Plant Mol.Biol. 43 (1992) 439–463.

[19] D. Ren, H. Yang, S. Zhang, Cell death mediated by MAPK is associated withhydrogen peroxide production in Arabidopsis, J. Biol. Chem. 277 (2002)559–565.

[20] G. Sachetto-Martins, L.D. Fernandes, D.B. Felix, D.E. de Oliveira, Preferentialtranscriptional activity of a glycine-rich protein gene from Arabidopsis thalianain protoderm-derived cells, Int. J. Plant Sci. 156 (1995) 460–470.

[21] G. Sachetto-Martins, L.O. Franco, D.E. de Oliveira, Plant glycine-rich proteins:a family or just proteins with a common motif, Biochim. Biophys. Acta 1492(2000) 1–14.

[22] P. Schweizer, A. Jeanguenat, D. Whitacre, J.-P. Metraux, E. Mosinger, Inductionof resistance in barley against Erysiphe graminis f. sp. hordei by free cutinmonomers, Physiol. Mol. Plant Pathol. 49 (1996) 103–120.

[23] P. Schweizer, A. Jeanguenat, E. Mosinger, J.-P. Metraux, Plant Protection by FreeCutin Monomers in Two Cereal Pathosystems, in: M.J. Daniels, J.A. Downie, A.E. Osbourn (Eds.), Advances in Molecular Genetics of Plant–MicrobeInteractions, Kluwer, 1994, pp. 371–374.

[24] P. Schweizer, G. Felix, A. Buchala, C. Muller, J.-P. Metraux, Perception of freecutin monomers by plant cells, Plant J. 10 (1996) 331–341.

[25] P.M. Steinert, J.W. Mack, B.P. Korge, S.-Q. Gan, S.R. Haynes, A.C. Steven, Glycineloops in proteins: their occurrence in certain intermediate filament chains,loricrins and single-stranded RNA binding proteins, J. Biol. Macromol. 13(1991) 130–139.

[26] M.C. Suh, A.L. Samuels, R. Jetter, L. Kunst, M. Pollard, J. Ohlrogge, F. Beisson,Cuticular lipid composition, surface structure, and gene expression inArabidopsis stem epidermis, Plant Physiol. 139 (2005) 1649–1665.

[27] H. Thordal-Christensen, Z. Zhang, Y. Wei, D.B. Collinge, Subcellular localiza-tion of H2O2 in plants. H2O2 accumulation in papillae and hypersensitiveresponse during the barley-powdery mildew interaction, Plant J. 11 (1997)1187–1194.

[28] S. Ueki, V. Citovsky, The systemic movement of a tobamovirus is inhibited bya cadmium-ion-induced glycine-rich protein, Nat. Cell Biol. 4 (2002)478–486.