Blackwell Science, LtdOxford, UKPCEPlant, Cell and Environment0016-8025Blackwell Science Ltd 2003?

2003 266915928 Original Article Differential expression of LTP genes in pepper H. W. Jung et al.

Plant, Cell and Environment (2003) 26, 915928

Three pathogen-inducible genes encoding lipid transfer protein from pepper are differentially activated by pathogens, abiotic, and environmental stresses*
HO WON JUNG1, WOONBONG KIM2 & BYUNG KOOK HWANG1

Laboratory of Molecular Plant Pathology, College of Life and Environmental Sciences, Korea University, Anam-dong, Sungbuk-ku, Seoul 136701, Korea and 2Department of Biochemistry and Molecular Biology, Michigan State University, East Lansing, MI, 48823, USA

ABSTRACT
The three cDNA clones, CALTPI, CALTPII, and CALTPIII, corresponding to pepper lipid transfer protein (LTP) genes were isolated from a pepper (Capsicum annuum) cDNA library from hypersensitive response (HR) lesions of leaves infected with Xanthomonas campestris pv. vesicatoria. The CALTP genes are well conserved in their coding region with 5772% identity at the amino acid level, but display 7283% identity at the nucleotide sequence level. The transcripts of the three CALTP genes differentially accumulated in pepper leaf, stem, and fruit tissues infected by X. campestris pv. vesicatoria, Phytophthora capsici and Colletotrichum gloeosporioides. The CALTP genes were also strongly induced in the systemic, upper leaves after immunization on lower leaves by either pathogenic or nonpathogenic bacteria. In situ hybridization results showed that the CALTPI mRNA was localized in phloem cells of vascular tissues in pepper leaf, stem and fruit tissues after pathogen infection. CALTPI and CALTPIII genes were predominantly expressed in various pepper tissues infected by pathogens, while infection by P. capsici and C. gloeosporioides did not induce the transcription of the CALTPII gene. Ethylene, methyl jasmonate and abscisic acid induced CALTPI and III gene expression in pepper leaves. Drought, high salinity, low temperature and wounding stresses also induced the expression of the CALTPI and CALTPIII genes in a similar manner. In contrast, only high salinity induced the CALTPII expression that was not generally affected by abiotic and other environmental stimuli. When compared with each other and with LTPs from other plants, CALTPI is more distantly related than CALTPII and CALTPIII sequences, indicating that the three pepper CALTP genes represent two different classes. These results thus show that CALTPI and CALTPIII genes, although

different in sequence structure, are transcriptionally activated in pepper tissues by pathogen infection as well as abiotic and environmental stresses. Key-words: Capsicum annuum; Xanthomonas campestris pv. vesicatoria; environmental stress; lipid transfer protein; plant hormone; systemic expression.

INTRODUCTION
Plant lipid transfer proteins (LTPs, PR-14), a class of small, soluble, and basic proteins, were initially identied in spinach leaves based on their ability to facilitate the transfer of various lipids between membranes and articial vesicles in vitro (Kader, Julienne & Vergnolle 1984; Nishida & Yamada 1985). However, whether or not LTPs act in vivo as intracellular transporters of lipids between organelles was later questioned, because they are generally secreted (Sterk et al. 1991) and externally associated with the cell wall (Segura, Moreno & Garcia-Olmedo 1993; Pyee & Kolattukudy 1995). Other functions have been ascribed to LTPs, including involvement in the epicuticular wax layer or cuticle biosynthesis, and participation in embryogenesis (Sossountzov et al. 1991; Sterk et al. 1991; Thoma et al. 1994; Pyee & Kolattukudy 1995; Clark & Bohnert 1999). Furthermore, LTPs bind fatty acids and acyl coenzyme A with high afnity, which suggests that they may function as fatty acid and acyl coenzyme A carrier proteins (Arondel et al. 1990; stergaard et al. 1993). Plant LTPs are encoded by a small multigene family in a diversity of plant species (Kader 1996; Clark & Bohnert 1999). The presence of multiple members of the LTP family may be a consequence of gene duplications and subsequent sequence variation, while those residues crucial to function are conserved (Dean, Pichersky & Dunsmuir 1989; Vignols et al. 1997). Although the nucleotide and amino acid sequences have considerable divergences, several features are highly conserved in the encoded LTPs, including eight cysteine residues to form four disulphide bonds (Kader 1996). All plant LTPs identied so far also possess a Nterminus signal peptide that likely directs the proteins to the endoplasmic reticulum (ER), along with [K/HDEL] ER 915

Correspondence: Byung Kook Hwang. Fax: + 82 2 925 1970; e-mail: bkhwang@korea.ac.kr *Nucleotide sequence data have been deposited in EMBL/GenBank database under accession numbers AF208832, AF208833, and AF208834. 2003 Blackwell Publishing Ltd

916 H. W. Jung et al. retention signal at the C-terminus, which is expected that LTPs may enter the secretory pathway (Sterk et al. 1991). Since the tertiary structure of LTPs from maize was initially predicted to be all b-sheet (Tchang et al. 1988), the crystal structure of LTPs from rice as well as solution structures of LTPs from maize, barley and rice have been determined (Han et al. 2001). Structural studies predict that plant LTPs contain an internal hydrophobic pocket that is large enough to accommodate a fatty acid or lysophospholipid (Gomar et al. 1998). LTPs may play a signicant role in plant responses to pathogen attack, high salinity, drought and low temperature (Hughes et al. 1992; Torres-Schumann, Godoy & Pintor-Toro 1992; Molina & Garcia-Olmedo 1993; Trevino & OConnell 1998). In particular, it was demonstrated that barley LTPs, synergistically with thionins, could inhibit the growth of bacterial and fungal pathogens in infected plants (Molina & Garcia-Olmedo 1993). The transcription of the ltp4 gene was shown to be induced nine-fold in barley plants 12 h after fungal pathogen infection. Some LTPs in radish, maize, and grape also were found to exhibit antimicrobial activities in vitro (Terras et al. 1992; Molina, Segura & Garcia-Olmedo 1993; Salzman et al. 1998). Furthermore, these LTPs were distributed over the plant surface and in the vascular tissues exposed to pathogens at a high concentration that could inhibit sensitive pathogens (Molina et al. 1993; Segura et al. 1993; Pyee, Yu & Kollattukudy 1994). More recently, overexpression of barley LTP2 protein in transgenic tobacco and Arabidopsis has been demonstrated to enhance resistance to Pseudomonas syringae pv. tabaci and P. syringae pv. tomato, respectively (Molina & GarciaOlmedo 1997). LTPs have also been identied as major allergens of Rosaceae fruits, such peach, apple and apricot (SanchezMonge et al. 1999). They are highly concentrated in the peel of fruits (Fernandez-Rivas & Cuevas 1999). LTPs are very stable allergens that showed the strong resistance to pepsin digestion and heat treatment (Asero et al. 2000). The wide distribution of LTPs in plants suggests that a potential role of these proteins as plant panallergens that are capable of inducing severe and near fatal responses in allergenic individuals (Ballmer-Weber 2002) A variety of expression patterns and genetic complexicity of LTPs suggest that a specic member of the gene family may be involved in a particular biological function. Furthermore, the secretion of some LTPs and the presence of a signal peptide sequence indicate that LTPs may have another function, yet to be elucidated, to mediate intracellular lipid trafc, as indicated from in vitro assays (Kader 1996). Whether or not multiple biological functions of various LTPs overlap in plants also remains to be elucidated. Moreover, gene-specic probes have rarely been used to monitor differential expression patterns of LTPs in pathogen-infected plants (Molina et al. 1996). In the present study, we have isolated three LTP genes (CALTPI, II, and III) from pepper cDNA library constructed from the hypersensitive response (HR) lesions of leaves infected by avirulent strain Bv5-4a of Xanthomonas campestris pv. vesicatoria. Using these pepper CALTP gene-specic probes, the spatial expression of the three CALTP genes was examined in organs of pepper plants. The differential expression of three CALTPs was characterized in various tissues of pepper in responses to pathogen infection, chemical stimuli or environmental stresses. We further examined the expression patterns of the three CALTP genes in systemic leaves of pepper plants immunized by either pathogenic X. campestris pv. vesicatoria or non-pathogenic Pseudomonas uorescens. The cellular localization of CALTP transcripts in pepper leaf, stem, and red fruit infected by plant pathogens were also analysed using in situ hybridization technique.

MATERIALS AND METHODS Plant, pathogens and inoculation

Pepper plants (Capsicum annuum L. cv. Hanbyul) were raised in a growth room at 28 C with a 16 h photoperiod under white uorescent light (80 mmol photons m-2 s-1), as previously described (Kim, Hwang & Park 1989). Inoculation of pepper plants with the virulent strain DS1 and avirulent strain Bv5-4a of X. campestris pv. vesicatoria was carried out by inltration of bacterial suspension into the abaxial side of fully expanded leaves of pepper plants at the four-leaf stage, as described previously (Jung & Hwang 2000a, b). To estimate whether or not pepper lipid transfer protein genes are systemically expressed in pepper plant, bacterial suspensions were inltrated into the lower leaves of pepper plants at the two- and six-leaf stages using a 3 mL syringe without a needle. Non-pathogenic P. uorescens strain ATCC 13525 grown in Kings B medium were also inltrated into lower leaves. The inoculated plants were incubated for 18 h in a humid chamber to provide the 100% relative humidity required for disease development, and then placed in the growth room at 28 C. The stems of pepper plants at the two-leaf stage were inoculated by wrapping them with a small piece of cotton soaked in the zoospore suspension (105 zoospores mL-1) of the virulent isolate S197 and the avirulent isolate CBS178.26 of Phytophthora capsici, as previously described (Jung & Hwang 2000b). The inoculated plants were placed in the growth room at 28 C. Green (unripe) and red (ripe) fruits were inoculated with conidial suspension (105 conidia mL-1) of Colletotrichum gloeosporioides, as previously described (Kim, Oh & Yang 1999). Leaf, stem, and fruit samples harvested from the infected pepper plants were immediately frozen in liquid nitrogen, and stored at -70 C until isolation of total RNA.

Treatment with plant hormones and environmental stresses

For treatment with ethylene, pepper plants at the four-leaf stage were enclosed in sealed 500 mL glass bottles and the ethylene was injected to yield a nal concentration of 5 mL L-1. The 100 mM methyl jasmonate in sterile distilled

2003 Blackwell Publishing Ltd, Plant, Cell and Environment, 26, 915928

Differential expression of LTP genes in pepper 917 water was applied as a spray to pepper plants at the fourleaf stage. The plants were enclosed in a vinyl bag for exposure to methyl jasmonate vapours. The plants also were sprayed with 100 mM abscisic acid (ABA) or 5 mM salicylic acid. For the drought-stress treatment, watering was withheld until the plants were visibly wilted. For NaCl treatment, pepper plants removed from the soil at the four-leaf stage without mechanical wounding were acclimated to NaClfree distilled water for 1 d, and transferred to 1, 10, 100 or 200 mM NaCl solution. The treated pepper plants were placed in a growth room. For low-temperature stress, the plants were transferred to an incubator in which the temperature was set at 4 C. Leaves of pepper plants at the sixleaf stage were scraped using sterile needles for wounding. Leaf samples were taken at various time intervals after treatments. Leaf samples were immediately frozen in liquid nitrogen and stored at -70 C until isolation of total RNA. CALTPII, and CALTPIII were dot-blotted onto Hybond N+nylon membrane (Amersham), and the membranes were probed with the gene-specic probes. Prehybridization, hybridization and washing were carried out under the same conditions for RNA gel blot analysis

Genomic DNA isolation and gel blot analysis

Capsicum annuum genomic DNA was prepared from young leaves, as previously described by Jung & Hwang (2000b). The high molecular weight genomic DNA was recovered by spooling. Twenty-micrograms of genomic DNA was digested with EcoRI, XbaI, HindIII, or EcoRV, according to the protocols described by Sambrook, Fritsch & Maniatis (1989). After ethanol precipitation, completely digested genomic DNA was re-suspended in 10 mM Tris and 1 mM ethylenediaminetetraacetic acid (EDTA) prior to gel electrophoresis. The DNA fragments were separated on 0.8% agarose gel. The DNA was transferred to Hybond N+ (Amersham) membrane, and hybridized to 32P-labelled gene non-specic or gene-specic probes at 65 C in 5% (w/ v) dextran sulphate, 0.25 M disodium phosphate pH 7.2, 7% (w/v) sodium dodecyl sulphate (SDS), and 1 mM EDTA. The membranes were washed twice with 2 SSC and 0.1% SDS for 10 min each at room temperature and nally several times with 0.1 SSC and 0.1% SDS for 15 min each at 65 C.

Sequence analysis of pepper lipid transfer protein cDNA genes

The three lipid transfer protein genes, used in this study, were isolated from the cDNA library made from the mRNA of pepper leaves infected with an avirulent strain of X. campestris pv. vesicatoria using differential hybridization techniques, as described in our previous study (Jung & Hwang 2000a). The cDNA genes were sequenced on ABI310 DNA sequencer (Applied Biosystems Inc., Foster City, CA, USA) using Thermo-cycle sequenase kit (Amersham, Little Chalfont, UK) with T3 or T7 primers according to the manufacturers instruction. DNA sequence analysis was performed using PC/Gene computer software and BLAST and ORF nder network services at the NCBI web site (Altschul et al. 1997). The cDNA genes were designated CALTPI, CALTPII, and CALTPIII (Capsicum annuum lipid transfer protein gene). Their nucleotide and deduced amino acid sequence data have been deposited in the EMBL/GenBank database under accession numbers AF208832, AF208833, and AF208834, respectively. Phylogenetic analysis of those sequences was subsequently performed through comparison with sequences in the BLAST databases using a clustal method in DNASTAR software program (DNASTAR, Madison, WI, USA).

RNA isolation and gel blot analysis

Total RNA was extracted from pepper leaves, stems, root, owers, and fruits by the guanidinium thiocyanate method (Chomczynski & Sacchi 1987). The concentration and integrity of total RNA in individual extracts were determined by UV absorbance and staining with ethidium bromide, respectively. Total RNA (20 mg) was denatured by heating at 65 C for 10 min in a formaldehyde gel-loading buffer, separated by electrophoresis on 7.4% formaldehyde gels, and transferred to nylon membranes (Hybond N+, Amersham) (Sambrook et al. 1989). RNA was cross-linked on the blots by UV illumination. The gene-specic regions of CALTPI, CALTPII and CALTPIII were 32P-labelled with a random prime kit (Boehringer Mannheim, Germany). Prehybridization and hybridization was performed, as described above. RNA gel blot analysis was repeated three times with similar results.

Preparation of gene-specic and non-specic probes for CALTPI, CALTPII, and CALTPIII
DNA restriction fragments of 5-terminal translation region of the three lipid transfer protein cDNA genes were used as gene-specic probes: a 207 bp EcoRI/HindIII fragment for CALTPI, a 174 bp EcoRI/PstI fragment for CALTPII, and a 170 bp EcoRI/XbaI fragment for CALTPIII in pBluescript SK() cloning vector. The region of CALTP genes used as a gene non-specic probe was the cDNA sequence representing a major portion of the sequences encoding the mature LTPs. Approximately 20 ng of denatured cDNAs of CALTPI,

In situ hybridization
Pepper leaf, stem, and fruit tissues cut into small pieces were xed and embedded as described by Lee et al. (2000). The embedded materials were stored at 4 C until used for in situ hybridization. In situ hybridization of mRNA in pepper leaf, stem, and fruit tissues was carried out following the methods of Hause et al. (1996) and Lee et al. (2000). Thin sections (10 mm in thickness) were prepared and attened on

2003 Blackwell Publishing Ltd, Plant, Cell and Environment, 26, 915928

918 H. W. Jung et al. microscope slides coated with poly L-lysine. After adhesion of the sections to the slides and removing the wax with xylol, the sections were re-hydrated by serially diluted ethanol. The sections were transferred sequentially to 0.01 M Tris-HCl (pH 8.0), 1% bovine serum albumin in 0.01 M Tris-HCl for 10 min at 37 C, 1 mg proteinase K mL-1 in 50 mM EDTA for 30 min at 37 C, and then 0.25% acetic anhydride in 0.1 M triethanolamine (pH 8.0) for 10 min at room temperature. The gene-specic probe of CALTPI was labelled with deoxigenin (DIG) high prime labelling kit (Boehringer Mannheim) according to the manufacturers recommendations. Tissue sections on the slides were prehybridized and hybridized at 42 C in 50% formamide, 2 SSC, 150 mg tRNA mL-1, and 0.5% blocking reagent (Boehringer Mannheim). After hybridization, the sections were washed twice in 50% formamide, 2 SSC for 10 min at 42 C, 2 SSC for 10 min at 42 C, and then in distilled water for 5 min at room temperature. A DIG-detection kit (Boehringer Mannheim) was used to detect the probe. Control tests were carried out without DIG-labelled CALTPI gene-specic probe or on the uninoculated pepper tissues. PII and CALTPIII also have two charged residues, aspartic acid (Asp, D) at position 42 and arginine (Arg, R) at position 43, which have been hypothesized to interact with the phosphate group of a phospholipid in other LTPs (Tchang et al. 1988; Kader 1996; Clark & Bohnert 1999). As shown in Fig. 1, these LTP gene-specic residues were conserved in previously reported genes of other plant species. Two of the four conserved proline residues (position 28 and 85) were substituted with Asn and Ser in the CALTPI cDNA sequences. Moreover, CALTPI cDNA does not contain a putative signal peptide region based on the sequence analyses. The two aspartic acids in residues are substituted for glutamic acid (Glu, E), which is a conserved substitution (Fig. 1). The predicted polypeptide sequences of the three CALTPs cDNA encoding putative pepper lipid transfer protein were compared with the previously sequenced nonspecic lipid transfer proteins from other plant species (Fig. 1). CALTPII exhibits 72% amino acid identity to CALTPIII and 79 and 80% identity to a tobacco LTP and a tomato LTP, respectively. CALTPIII also shares similar amino acid identities to tobacco and tomato LTP. CALTPII and CALTPIII have 5864% identities to LTPs of beet and spinach. However, CALTPI shares only 57 and 58% sequence identity to CALTPII and CALTPIII, respectively, and low levels of sequence identity (4063%) to each of the known LTPs of tobacco, tomato, beet, spinach and rice. The three CALTPs of pepper plants share lower levels of sequence identities (4145%) to Arabidopsis LTPs than those of other dicot plant species. The CALTPs also share 3943% sequence identity to the allergenic plant LTPs of apple, peach, apricot and carrot. A phylogenetic tree constructed for the C. annuum LTPs and closely related amino acid sequences available in the GenBank is shown in Fig. 2. The tree obtained shows three distinct phylogenetic groups among the LTP amino acid sequences. Within group A, all of the allergenic plant LTP sequences including apple, apricot, peach and prune cluster together. The Gramineae LTP genes of maize sorghum and rice also cluster together in group B, whereas the Solanaceae LTP genes are present in the group C cluster. In particular, pepper CALTPI sequence has a long phylogenetic distance from pepper CALTPII and CALTPIII sequences. Moreover, pepper CALTPII sequence also clustered together with CALTPIII sequence. The phylogenetic analysis reveals that at least two gene subfamilies may exist in pepper genome.

RESULTS Isolation and analysis of pepper LTP cDNA clones

Some cDNA clones corresponding to pepper LTP genes were differentially expressed in leaves undergoing HR, and isolated from a pepper cDNA library using a array-based differential hybridization technique (Jung & Hwang 2000a). Subsequent restriction enzyme mapping and DNA sequencing analyses revealed that the pepper cDNA represent three different genes, designated CALTPI, CALTPII, and CALTPIII (Capsicum annuum lipid transfer protein). The three LTP cDNA clones CALTPI (accession no. AF208832), CALTPII (accession no. AF208833), and CALTPIII (accession no. AF208834) that have an apparent open reading frame are 588, 717 and 685 bp in length, respectively (data not shown). The overall amino acid sequence identities between clones are 57% (CALTPI and CALTPII), 58% (CALTPI and CALTPIII) and 72% (CALTPII and CALTPIII). Interestingly, the nucleotide sequences of the 3-untranslated region of CALTPII and CALTPIII are almost identical, wherreas the 3-untranslated region of CALTPI is very different. The predicted polypeptides of CALTPII and CALTPIII contain a typical hydrophobic signal peptide region at the N-terminus, and the signal peptide is predicted to be cleaved between positions 24 (Ala) and 25 (Leu) in both gene products (Von Heijne 1986; Trevino & OConnell 1998). CALTPII and CALTPIII cDNA sequences code for eight cysteine (Cys) and four proline (Pro) residues and a valine residue at the sixth position of the mature peptide in highly conserved regions, as previously reported for other LTPs (Kader 1996) (Fig. 1). The mature proteins of CALT-

Generation of three CALTP gene-specic probes

Based on a comparison with each of the nucleotide sequences (data not shown), cDNA fragments partially encompassing the 5 untranslated regions (UTR) of CALTPI, CALTPII, and CALTPIII were used as genespecic probes to examine the patterns of pepper LTP gene expression. The gene-specic probes were prepared by digestion of each of pepper LTP cDNA homologues using different restriction enzymes (Fig. 3a) The sizes of probes

2003 Blackwell Publishing Ltd, Plant, Cell and Environment, 26, 915928

Differential expression of LTP genes in pepper 919

Figure 1. Comparison of the deduced amino acid sequences of CALTPI, CALTPII, and CALTPIII genes encoding putative lipid transfer proteins from pepper with lipid transfer proteins from tobacco (accession no. D13952; Masuta et al. 1992); tomato (accession no. U81996; Plant, Cohen & Bray 1991); beet (accession no. Q43748; Nielsen et al. 1996); spinach (accession no. M58635; Bernhard et al. 1991); rice (accession no. Z23271; Vignols et al. 1994); Arabidopsis LTP2 (accession no. Q9S7I3, Clark and Behnert 1999); Arabidopsis LTP1 (accession no. AF159798; Arondel et al. 2000); apple LTP (accession no. CAB96874; Diaz-Perales et al. 2001); peach LTP1 (accession no. CAB96876; Diaz-Perales et al. 2001); apricot LTP1 (accession no. P81651, Conti et al. 2001); and carrot (accession no. P27631; Sterk et al. 1991). Underlines indicate position of putative leader cleavage site in pepper lipid transfer proteins of CALTPII and CALTPIII. Dots mark the eight conserved cysteine residues to form four disulphide bridges. Boxes indicate identical amino acid residues. Dashes indicate the spacing in amino acid sequences that was done by proper alignment. The bold letters indicate highly conserved amino acid residues among amino acid sequences of plant lipid transfer proteins.

CALTPI, CALTPII, and CALTPIII were 207, 174, and 170 bp, respectively. The extent of cross-hybridization of the CALTP probes was examined by dot blot analysis (Fig. 3b). These probes did not cross-hybridize each other under the hybridization conditions used, indicating that these cDNA fragments could be used as gene-specic probes.

CALTP genes, multiple bands were detected in the digests of the genomic DNA (Fig. 4a). As shown in Fig. 4b, using the CALTP gene-specic probes, CALTPI hybridized to the two bands in the digests, whereas CALTPII and CALTPIII hybridized to only one band. These results suggest that each of the three CALTP cDNAs has single or two copy genes in the pepper genome and pepper LTPs are composed of a gene family.

Genomic organization of CALTP genes

To investigate the number of genes corresponding to each of CALTP cDNAs in pepper genome, genomic DNA was digested with EcoR1, Xba1, HindIII or EcoRV, and the resulting fragments were hybridized with the CALTP probes. When hybridized with a gene non-specic probe for

Organ-specic expression of CALTP genes in pepper tissues

Transcripts of the CALTPI gene were highly expressed only in ower and green (unripe) fruit of pepper (Fig. 5). CALTPII mRNAs were detected in all of tested organs except

2003 Blackwell Publishing Ltd, Plant, Cell and Environment, 26, 915928

920 H. W. Jung et al.

Figure 2. A phylogenetic tree of pepper

CALTP and other plant LTP genes. The tree was built using the amino acid sequences indexed in GenBank/EMBL BLAST databases. The length of each pair of branches represents the distance between sequence pairs, while the units at the bottom of the tree indicate the number of substitution events.

root and red fruit. Whereas CALTPII expression was low at these organs, an elevated level of CALTPII was found in green fruit. Expression of CALTPIII gene was observed in the stem and green fruit tissues of pepper, but was not detected in other tested pepper organs.

Differential expression of CALTP genes by pathogen infection

Expression of three CALTP genes was monitored over 30 h following inltration of pepper leaves with either a virulent (DS1) or avirulent (Bv5-4a) strain of X. campestris pv. vesicatoria (Fig. 6a). CALTPI transcripts were very weakly detected or absent in compatible interaction from 18 to 30 h after inoculation, but were strongly induced at 24 h after inoculation in incompatible interaction. This induction at 24 h was no longer detected within 6 h as shown in Fig. 6a, suggesting that CALTPI is subjected to a rapid turnover. In contrast to CALTPI, CALTPII and CALTPIII mRNA expression was highly induced in both compatible and incompatible interactions upon the bacterial infection. In particular, a strong accumulation of CALTPII mRNA already occurred in the incompatible interaction at 6 h after inoculation with X. campestris pv. vesicatoria. CALTPIII transcripts were rst detected in leaves at 12 h after inocu-

lation with the avirulent strain. As shown in Fig. 6a, the CALTP genes were both more rapidly and strongly transcribed in incompatible interaction than in compatible interaction. RNA gel blot analysis of the three CALTP genes was performed in pepper stems at various time intervals after inoculation with virulent S197 or avirulent isolate CBS178.26 of P. capsici (Fig. 6b). In the compatible interaction, CALTPI mRNA was not detected in stem tissues until 48 h after inoculation with the virulent isolate. However, the transcription of CALTPI was induced at 36 h after inoculation in the incompatible interaction. This induction at 36 h was signicantly reduced after 48 h, as similarly observed for CALTPI in an incompatible interaction with X. campestris pv. vesicatoria (Fig. 6a). In the case of the CALTPII gene, the transcripts remained at low levels in stem tissues of the healthy or infected plants. However, CALTPIII gene expression was signicantly higher in the incompatible interaction, in comparison with the compatible interaction. RNA gel blot analysis also was performed in green or red pepper fruit inoculated with C. gloeosporioides (Fig. 6c). At 5 d after inoculation, the green fruits showed typical circleshaped symptoms, but the red fruits did not produce any symptoms (Kim et al. 1999) (data not shown). In the inter-

2003 Blackwell Publishing Ltd, Plant, Cell and Environment, 26, 915928

Differential expression of LTP genes in pepper 921

(a)

(b)
Figure 4. Genomic DNA gel blotting analyses of CALTPI, CALTPII and CALTPIII genes encoding lipid transfer protein in pepper plants. (a) Hybridization of fragments of genomic DNAs digested with EcoRI, EcoRV, HindIII, or XbaI, using a gene nonspecic probe for CALTP genes. (b) Hybridization of fragments of genomic DNAs digested with EcoRV using each of the genespecic regions of the three CALTP genes. Twenty micrograms of genomic DNA was digested with the restriction enzymes and separated in 0.8% agarose gel, followed by Southern hybridization. The numbers on the left indicated the length of the fragments in kilobases.

Figure 3. DNA gel blot analysis showing the specicity of genespecic probes of three pepper lipid transfer protein cDNA clones. (a) Restriction enzyme map analysis of three lipid transfer protein cDNA clones. Grey boxes indicate the positions of the genespecic hybridization probes generated by restriction enzyme digestion. (b) Dot blot hybridization of the full-length cDNA of CALTPI, CALTPII, and CALTPIII with the gene-specic probes.

accumulated in the systemic upper third leaves at 3 d after immunization on lower leaves, but not in the mockimmunized, systemic upper leaves. The systemic expression of the CALTP mRNA drastically declined in upper fourth, fth and sixth leaves.

actions between green fruit and anthracnose fungus, transcription of the three CALTP genes was slightly induced in the diseased green fruit, although the CALTP genes were constitutively transcribed in healthy green fruit (Fig. 6c). In particular, expressions of CALTPI and CALTPIII genes were more strongly detected in the green fruit than CALTPII from 24 h after inoculation. However, the level of transcripts of CALTPII was gradually decreased and all CALTP genes drastically disappeared in green fruits at 72 h after inoculation. In red fruits infected by the anthracnose fungus, only CALTPI mRNA was detected at 72 h after inoculation, whereas there was no induction of CALTPII and CALTPIII transcripts.

Systemic expression of CALTP genes in pepper leaves

CALTP mRNA accumulation was analysed in the upper leaves of pepper plants whose lower (rst and second) leaves were inoculated with avirulent strain Bv5-4a of X. campestris pv. vesicatoria. As shown in Fig. 7a, transcripts of the three CALTP genes, were strongly induced and

Figure 5. RNA gel blot analysis of mRNA expression of CALTPI, CALTPII AND CALTPIII in various tissues of pepper plants. Ten micrograms of total RNA was separated in each lane. The blots were hybridized with each gene-specic probe of CALTPI, CALTPII and CALTPIII. The equivalence of RNA loading among lanes of agarose gel and complete transfer of RNA to membrane was demonstrated by ethidium bromide staining of RNA on membrane before hybridization.

2003 Blackwell Publishing Ltd, Plant, Cell and Environment, 26, 915928

922 H. W. Jung et al.

(a)

(b)

(c)

Figure 6. RNA gel blot analysis of transcript levels of CALTPI, CALTPII, and CALTPIII genes in pepper tissues inoculated with various plant pathogens. (a) CALTP mRNA levels in pepper leaves at various time intervals after inoculation with virulent strain DS1 and avirulent strain Bv5-4a of Xanthomonas campestris pv. vesicatoria at the six-leaf stage. (b) CALTP mRNA levels in pepper stems at various time intervals after inoculation with virulent isolate S197 and avirulent isolate CBS178.26 of Phytophthora capsici at the two-leaf stage. (c) CALTP mRNAs levels in green (unripe) fruit and red (ripe) fruit at various time intervals after inoculation with Colletotrichum gloeosporioides. Ten micrograms of total RNA from each sample was separated in each lane. The blots were hybridized with the gene-specic probes of pepper lipid transfer proteins CALTPI, CALTPII and CALTPIII cDNAs. The equivalence of RNA loading among lanes of agarose gel and complete transfer of RNA to membrane was demonstrated by ethidium bromide staining of RNA on membrane before hybridization.

To determine more detailed systemic expression patterns of CALTP mRNA, lower rst leaves of pepper plants at the two-leaf stage were immunized with either the virulent and avirulent strains of X. campestris pv. vesicatoria or non-pathogenic Pseudomonas uorescens. RNA gel blot analysis was performed in local (inoculated) and systemic (uninoculated) leaves of pepper plants using a CALTPIII full-length cDNA probe (Fig. 7b). CALTPIII transcripts were strongly expressed in the lower rst leaves inoculated

with the virulent strain DS1. Although CALTPIII mRNA was considerably induced in systemic, upper second leaves at 1 h after immunization, the transcripts gradually disappeared thereafter. Infection of lower leaves by avirulent strain Bv5-4a also strongly induced the accumulation of CALTPIII mRNA in the systemic upper leaves as well as the local lower leaves. In particular, the transcript accumulation in systemic upper leaves was remarkably high at 20 h after infection of the avirulent strain Bv5-4a on lower

2003 Blackwell Publishing Ltd, Plant, Cell and Environment, 26, 915928

Differential expression of LTP genes in pepper 923

(a)

(b)

Figure 7. Systemic expression of CALTP genes in pepper plants immunized with Xanthomonas campestris pv. vesicatoria, or Pseudomonas uorescens. (a) Expression of CALTPI, CALTPII and CALTPIII genes in systemic, upper leaves at 3 d after immunization of lower rst and second leaves with avirulent strain Bv5-4a of X. campestris pv. vesicatoria at the six-leaf stage. The blots were hybridized with different gene-specic probes of pepper lipid transfer proteins. (b) Expression of CALTPIII gene in local (rst) and systemic (second) leaves at various time intervals after immunization of lower rst leaves with virulent (DS1), avirulent (Bv5-4a) strains of X. campestris pv. vesicatoria, and nonpathogenic Pseudomonas uorescens strain ATCC13525 at the two-leaf stage. Control experiments with non-treatment (control) and mock treatment were performed to conrm the no expression of CALTPIII in both local (L) and systemic (S) leaves. The 685-bp EcoRI/ XhoI fragment of pepper CALTPIII full-length cDNA insert was used as a probe. The equivalence of RNA loading among lanes of agarose gel and complete transfer of RNA to membrane was demonstrated by ethidium bromide staining of RNA on the membrane before hybridization.

leaves. Interestingly, immunization of pepper lower leaves with non-pathogenic P. uorescens was also effective in triggering the accumulation of CALTPIII mRNA in the local, lower and systemic upper leaves. The transcript accumulation occurred rapidly in both leaves at the early time from 3 to 7 h after immunization. The CALTPIII transcripts remained at relatively high levels in systemic upper leaf tissues up to 20 h after the immunization, whereas the

transcripts were at low levels in local leaves after 10 h of infection.

In situ localization of CALTP mRNA in pepper tissues

The in situ hybridization technique was used to localize the CALTPI mRNA expressed in the pepper leaf, stem, or red

Figure 8. In situ localization of CALTPI transcripts in the pepper leaf (a, b), stem (c, d), and red fruit (e, f) tissues. (a), (c), and (e) are
uninfected, healthy pepper tissues. (b) Systemic upper, second leaf tissues at 20 h after immunization of lower rst leaves with avirulent strain of Xanthomonas campestris pv. vesicatoria. (d) Pepper stem tissues at 48 h after inoculation with avirulent isolate of Phytophthora capsici. (f) red fruit tissues at 72 h after inoculation with Colletotrichum gloeosporioides. Bars = 50 mm. Co, collenchyma cell; LE, lower epidermis; M, mesophyll cell; P, phloem; PM, parenchyma mesophyll; SM, sponge mesophyll; UE, upper epidemis; Vs, vascular bundle; X. xylem. 2003 Blackwell Publishing Ltd, Plant, Cell and Environment, 26, 915928

924 H. W. Jung et al. fruit tissues infected with the pathogenic bacteria, oomycete, and fungi (Fig. 8). No hybridization signals of CALTPI mRNA were detected in any organelles of uninfected healthy leaf, stem and red fruit tissues (Fig. 8a, c & e). After immunization of the lower rst leaves with avirulent strain of X. campestris pv. vesicatoria, a heavy deposit of CALTPI mRNA occurred only within the phloem area of vascular tissue and in the epidermal tissue in upper second leaves (Fig. 8b). The CALTPI transcripts were localized in the phloem cells of pepper stem tissue infected by avirulent isolate of P. capsici, but not in xylem vessels (Fig. 8d). In the red fruits infected by C. gloeosporioides, CALTPI gene was strongly expressed within phloem cells in the vascular bundles of fruit tissues (Fig. 8f). No staining was observed in either the inoculated tissue or the uninoculated healthy plant tissue without CALTPI probe (data not shown). stitutively expressed in untreated, control leaves and the CALTPII transcription was slightly decreased with the lapse of time after ABA treatment. However, treatment with salicylic acid (5 mM) did not induce elevated level of the three CALTP transcripts, except for a strong induction of CALTPI transcription at 24 h after treatment (Fig. 9d). CALTPI and CALTPIII mRNA began to accumulate at 1 d after drought and remained at a constant level to 4 d after the appearance of wilting (Fig. 10a). Transcripts of CALTPII were weakly detectable in untreated leaf tissues, slightly increased at 1 d drought and remained at low levels thereafter (Fig. 10a). Sodium chloride triggered a relatively strong accumulation of all three CALTP mRNA in pepper leaf tissue from 6 to 24 h after treatment (Fig. 10b). The 100 mM NaCl treatment was effective in inducing high levels of all three CALTP transcripts. After 2 h of cold treatment, CALTPI and CALTPIII transcripts also were induced in leaf tissues and remained at a constant level up to 24 h (Fig. 10c). In contrast, cold treatment did not affect the levels of CALTPII transcripts during the cold acclimation. As shown in Fig. 10d, CALTPI mRNA was not detected in pepper leaves during 24 h after wounding. Only the CALTPIII transcripts were elevated in wounded pepper leaves, reaching the maximum level at 18 h after wounding.

Induction of CALTP genes by abiotic stimuli

The CALTPI and CALTPIII mRNAs began to accumulate in pepper leaves at 6 h after ethylene treatment (5 mL L-1) and remained at constant levels up to 24 h (Fig. 9a). In the case of the CALTPII gene, the transcripts gradually declined 18 h after ethylene treatment. For the methyl jasmonate treatment (100 mM), expression patterns of the three CALTP genes were similar to those of ethylene treatment (Fig. 9b). Abscisic acid (100 mM) induced the transcription of CALTPI and CALTPIII at 12 h after treatment and the transcription of the two genes slightly increased at 18 and 24 h (Fig. 9c). In contrast, CALTPII gene was con-

DISCUSSION Structural analysis and organ-specic expression of CALTP genes

The three novel non-specic lipid transfer protein genes CALTPI, CALTPII, and CALTPIII were isolated from the

(a)

(b)

(c)

(d)

Figure 9. RNA gel blot analysis of the expression of CALTPI, CALTPII, and CALTPIII genes in pepper leaves in response to various abiotic stresses of ethylene (a), methyl jasmonate (b), abscisic acid (c), and salicylic acid (d). Total RNA was extracted from pepper leaves at various time intervals after treatment with ethylene (5 mL L-1), methyl jasmonate (100 mM), abscisic acid (100 mM), and salicylic acid (5 mM). Ten micrograms of total RNA from each sample were separated in each lane. The blots were hybridized with different gene-specic probes of the pepper lipid transfer proteins. The equivalence of RNA loading among lanes of agarose gel and complete transfer of RNA to the membranes were demonstrated by ethidium bromide staining of RNA on membrane before hybridization.
2003 Blackwell Publishing Ltd, Plant, Cell and Environment, 26, 915928

Differential expression of LTP genes in pepper 925

(a)

(b)

(c)

(d)

Figure 10. RNA gel blot analysis of expressions of CALTPI, CALTPII and CALTPIII genes in leaves treated with drought (a), NaCl (b), low temperature (c), and wounding (d). Total RNA was isolated from pepper leaf tissues at various time intervals after treatment. Twenty micrograms of total RNA was separated in each lane. The blots were hybridized with different gene-specic probes of these CALTPs. The equivalence of RNA loading among lanes of agarose gel and complete transfer of RNA to membrane were demonstrated by ethidium bromide staining of RNA on membrane before hybridization.

cDNA library constructed from pepper leaves infected with avirulent strain Bv5-4a of X. campestris pv. vesicatoria (Jung & Hwang 2000a). The deduced amino acid sequences of CALTPII and CALTPIII share a number of characteristics common to all known plant LTPs, including a low molecular weight, a basic pI, eight cysteine and four proline residues at the conserved positions, and a N-terminal signal peptide. Moreover, homology in the nucleotide sequences of these two cDNAs extend through the 3-UTR, and unique DNA hybridization pattern, suggesting that they may occur via alternative splicing of a LTP gene, as in the case of maize LTP genes (Arondel et al. 1991). Some of the conserved regions in LTP genes may play a signicant role in their structure and function. The charged portion DRK and the hydrophobic amino acid sequence I/LPA/NM/TCGVNIPY were suggested to function as possible sites of the interaction for the lipid phosphate group and the acyl chains, respectively (Tchang et al. 1988; Torres-Schumann et al. 1992). In particular, the predicted CALTPI protein does not contain a putative signal peptide, due to eight amino acid substitutions or deletions in the conserved positions. The hydrophobic amino acid residues in CALTPI were not found in CALTPII, CALTPIII and other plant LTP genes, which suggest that CALTPI gene may have a different function, as compared to the other LTPs. When based on amino acid sequence comparisons of the three CALTPs with LTP gene family members from other species, the sequence of CALTPI was less closely related to predicted amino acid sequences of LTP genes isolated from Solanaceae than to those of CALTPII and CALTPIII. The phylogenetic tree constructed from the amino acid sequence alignment of LTPs showed that plant LTPs were divided into two major groups, that is, pepper CALTPI and other LTPs (Figs 1 & 2). In the group including CALTPII and CALTPIII, the CALTPIII was also denitely separated from CALTPII and other Solanaceae LTPs as well as other plant families. These

phylogenetic analyses indicate that CALTPI, CALTPII and CALTPIII are novel cDNA clones encoding different isoforms of the plant non-specic lipid transfer protein. Interestingly, some LTPs of the allergenic plants, such as peach, apricot, and apple (Sanchez-Monge et al. 1999) clustered together, but have long phylogenetic distances from Solanaceae LTPs including pepper LTPs. These ndings suggest that pepper LTPs may not be plant panallergens, although allergenic LTPs are a family of small polypeptides involved in plant defence against pathogens (Kader 1996). Organ- and tissue-specic expression of LTP genes has been demonstrated in several plant species (Sossountzov et al. 1991; Sterk et al. 1991; Clark & Bohnert 1999). The expression of the LTP genes was found to be active in the aerial plant tissues, including leaves, stems, shoot meristem and owers, but not detected in the roots of various plants (Kader 1996). In pepper plants, expressions of the three CALTP genes were differentially regulated in various organs, which suggest that the CALTP genes, especially CALTPII genes, are specically expressed in certain tissues during the development of leaf, stem, ower and fruit when biosynthesis of the cuticular membrane is required (Sterk et al. 1991). A role for LTPs in cutin biosynthesis, specically affecting the transport of cutin monomers through the extracellular matrix, has been proposed (Sterk et al. 1991).

Differential expression and in situ localization of CALTP genes and mRNA in the diseased tissues
Infection of pepper plants by X. campestris pv. vesicatoria or P. capsici affected accumulation of the three CALTP transcripts in different ways (Fig. 6). In particular, the transcription of the CALTP genes was more rapidly and strongly induced in the incompatible than the compatible interactions. These results suggest that increase in three

2003 Blackwell Publishing Ltd, Plant, Cell and Environment, 26, 915928

926 H. W. Jung et al. CALTP transcripts concomitant with bacterial and oomycetes infection may play a signicant role in the expression of resistance response to X. campestris pv. vesicatoria and P. capsici in pepper leaves and stems, respectively. The responses of CALTPs to plant pathogen infection in pepper leaves and stems are consistent with the earlier observations of expression of pepper basic PR-1, pepper b-1,3-glucanase and other plant PR genes (Ryals, Uknes & Ward 1996; Jung & Hwang 2000b; Kim & Hwang 2000). In the interaction of pepper unripe, green fruit with C. gloeosporioides, we did not observe any dramatic changes in the level of CALTP transcripts by the anthracnose fungal infection. These results suggest that the lipid transfer protein genes may not be involved in triggering the anthracnose disease response in unripe, green fruits of pepper. It is postulated that during the formation of necrotic lesions in some plantpathogen interactions, a signal triggers the signal transduction pathway that initiates a defence response to pathogen attack (Ryals et al. 1994). In our present studies, the three CALTP transcripts accumulated in the systemic upper leaves of pepper plants exposed on the lower leaves to virulent, avirulent or nonpathogenic bacteria (Fig. 7). These results suggest that cellular signalling processes may activate defence responses at systemic levels upon pathogen infection by systemically inducing the CALTP genes, especially CALTPI and CALTPIII. Thus, CALTP may function in signalling for SAR in pepper plants. The identity and relative expression levels of SAR genes have been documented to vary among plant species, partly due to different evolutionary or breeding constraints (Kessmann et al. 1994). More recently, a putative lipid transfer protein has been demonstrated to be involved in systemic resistance signalling in Arabidopsis mutant DIR1 (Maldonado et al. 2002). Based on our present SAR results and Arabidopsis DIR1 study, we postulate that LTPs may function as either cosignal or translocator for release of the mobile signals during SAR response in plants. The spatial expression of CALTPI genes, which did not contain unique N-terminus signal peptide sequences, was observed in leaves, stem, and red fruit tissues of pepper infected with plant pathogens (Figs 6 & 8). In particular, CALTPI transcripts were expressed within phloem cells in the vascular bundles of pepper leaves, stems, and red fruits. It is interesting to note that the transcripts of CALTPI were highly expressed in the vascular bundles and the epidermal cells of systemic upper leaf tissues after immunization of lower leaves with X. campestris pv. vesicatoria (Fig. 8b). Previous works indicated that LTP transcripts are expressed in the epidermis cell layer and the lipid transfer proteins are localized extracellularly in the cuticle and wax layers of various plant tissues (Sterk et al. 1991; Thoma et al. 1994; Pyee & Kolattukudy 1995). In particular, maize LTP transcripts were localized in vascular tissues of the coleoptiles and leaves as well as the epidermal cell layer (Sossountzov et al. 1991). More recently, using different LTP promoters of potato to regulate the expression of a reporter gene, one member of the LTP gene family has been demonstrated to direct expression in the phloem surrounding the nodes of young plants (Horvath et al. 2002). The LTP expression has been postulated to predominantly occur in the epidermal tissue of aerial parts of plants, suggesting the postulated roles in defence related to external wax deposition. In our present study, the expression of CALTPI in the vascular and epidermal tissue of pepper may play a signicant role in the production of cuticular waxes in epidermal cells.

Induction of CALTP genes by plant hormones and environmental stresses

In our previous (Jung & Hwang 2000a) and present studies, exogenous salicylic acid treatment did not accumulate elevated CALTP transcripts in pepper leaves, which was inconsistent with the expression patterns of rice LTP mRNA in the SA-treated coleoptiles (Vignols et al. 1997). These results suggest that SA may not be an effective signal molecule capable of inducing the transcription of CALTP genes. As shown in Fig. 9, transcription of the CALTPI and CALTPIII genes was strongly induced in pepper leaves treated with either ethylene or methyl jasmonate, which suggests that ethylene and jasmonic acid may function in the signal transduction pathway to activate CALTP gene expression. However, exogenous ethylene treatment did not stimulate the expression of CALTPII and rather decreased the CALTPII gene expression. Several stress-induced LTP genes have been isolated and characterized from various plant species, since LTP (TSW12) gene in tomato was shown to respond to salinity (Torres-Schumann et al. 1992; Molina & Garcia-Olmedo 1993; Colmenero et al. 1997; Trevino & OConnell 1998). The three LTP genes from drought-tolerant wild tomato (Lycopersicon pennellii) were induced in ABA-treated leaves (Trevino & OConnell 1998). In our present study, transcripts of CALTPI and CALTPIII were strongly accumulated in pepper leaves after ABA treatment, whereas transcription of CALTPII declined in the pepper leaves after ABA treatment (Fig. 9c). These results suggest that ABA may serve as a signal molecule in mediating the induction of CALTPI and CALTPIII genes, but not CALTPII. The application of ABA to vegetative tissues was found to mimic diverse effects of drought, low temperature, and salt stresses on plants (Thomas, Vivekananda & Bogue 1991). In the present study, drought, high salt, low temperature and wounding triggered accumulation of CALTPI and CALTPIII transcripts in a similar manner. In contrast, drought and low temperature had little or no effect in CALTPII gene expression in the leaf tissues. The accumulation of CALTPI and CALTPIII transcripts by environmental stresses may be associated with a putative role of LTPs in cuticle biosynthesis (Sterk et al. 1991), because the cuticle plays a signicant role in the water balance of plants (Lemieux 1996). Moreover, induction of CALTP gene expression under environmental stresses may contribute to

2003 Blackwell Publishing Ltd, Plant, Cell and Environment, 26, 915928

Differential expression of LTP genes in pepper 927 the repair of stress-induced damage in membranes by increasing the cuticle thickness. Although CALTPI and CALTPIII genes can be developmentally expressed or specically up-regulated in different organs of pepper, their expression may be mediated by a variety of biotic and abiotic stimuli. As clearly demonstrated in the present study, CALTPI and CALTPIII genes may be involved in defence-related genes inducible to inhibit pathogen growth as well as to adapt environmental stresses in pepper. These differential expression of the pepper lipid transfer proteins encoded by CALTPI, II and III suggests that CALTP proteins may play signicantly different roles not only in the developmental physiology of pepper, but also in the local defence reactions, the induction of systemic resistance, and the adaptation to the environmental stresses. Taken together, our results suggest that pepper CALTPI and CALTPIII genes with distinctly different sequence structures may be transcriptionally regulated in a similar manner by pathogen infection and abiotic stresses. Although CALTPI and II maintain a distinct tissue- and organ-specicity in their expression, the various LTPs may have functional overlap under specic physiological conditions. Finally, the phylogenetic analysis and the diversity of expression patterns of CALTP genes suggest that pepper CALTP genes could be divided into two classes, and additional copies or even new classes may exist in pepper genome. Further isolation and characterization of LTP genes will provide insights into the role of the multifunctional LTP multigene family in pepper.
Asero R., Mistrello G., Roncarolo D., et al. (2000) Lipid-transfer protein: a pan-allergen in plant derived food that is highly resistant to pepsin digestion. International Archives of Allergy and Immunology 122, 2032. Ballmer-Weber B.K. (2002) Lipid transfer protein as a potential panallergen. Allergy 57, 873875. Bernhard W.R., Thoma S., Botella J.R. & Somerville C.R. (1991) Isolation of a cDNA clone for spinach lipid transfer protein and evidence that the protein is synthesized by the secretory pathway. Plant Physiology 95, 164170. Chomczynski P. & Sacchi N. (1987) Single-step methods of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Analytical Biochemistry 162, 156159. Clark A.M. & Bohnert H.J. (1999) Cell-specic expression of genes of the lipid transfer protein family from Arabidopsis thaliana. Plant Cell Physiology 40, 6976. Colmenero-Flores J.M., Campos F., Garciarrubio A. & Covarrubias A.A. (1997) Characterization of Phaseolus vulgaris cDNA clones responsive to water decit: identication of a novel late embryogenesis abundant-like protein. Plant Molecular Biology 35, 393405. Conti A., Fortunato D., Ortolani C., Giuffrida M.G., Pravettoni V., Napolitano L., Farioli L., Perono Garoffo L., Trambaioli C. & Pastorello E.A. (2001) Determination of the primary structure of two lipid transfer proteins from apricot (Prunus armeniaca). Journal of Chromatography B: Biomedical Sciences and Applications 756, 123129. Dean C., Pichersky E. & Dunsmuir P. (1989) Structure, evolution and regulation of RbcS genes in higher plants. Annual Review of Plant Physiology and Plant Molecular Biology 40, 415439. Diaz-Perales A., Garcia-Casado G., Sanchez-Monge R., GarciaSelles F.J., Barber D. & Salcedo G. (2002) cDNA cloning and heterologous expression of the major allergens from peach and apple belonging to the lipid-transfer protein family. Clinical and Experimental Allergy 32, 8792. Fernandez-Rivas M. & Cuevas M. (1999) Peels of Rosaceae fruits have a higher allergenicity than pulps. Clinical and Experimental Allergy 29, 12391247. Gomar J., Sodano P., Sy D., Shin D.H., Lee J.Y., Suh S.W., Marion D., Vovelle F. & Ptak M. (1998) Comparison of solution and crystal structures of maize nonspecic lipid transfer protein: a model for a potential in vivo lipid carrier protein. Proteins 31, 160171. Han G.W., Lee J.Y., Song H.K., Chang C., Min K., et al. (2001) Structural basis of non-specic lipid binding in maize lipid-transfer protein complex revealed by high resolution X-ray crystallography. Journal of Molecular Biology 308, 263278. Hause B., Demus U., Teichmann C., Parthier B. & Wasternack C. (1996) Developmental and tissue-specic expression of JIP-23, a jasmonate inducible protein of barley. Plant Cell Physiology 37, 641649. Horvath B.M., Bachem C.W.B., Trindade L.M., Oortwijn M.E.P. & Visser R.G.F. (2002) Expression analysis of a family of nsLTP genes tissue specically expressed throughout the plant and during potato tuber life cycle. Plant Physiology 129, 1494 1506. Hughes M.A., Dunn M.A., Pearce R.S., White A.J. & Zhang L. (1992) An abscisic acid-responsive, low temperature barley gene has homology with a maize phospholipid transfer protein. Plant, Cell and Environment 15, 861865. Jung H.W. & Hwang B.K. (2000a) Isolation, partial sequencing and expression of pathogenesis-related cDNA genes from pepper leaves infected by Xanthomonas campestris pv. vesicatoria. Molecular PlantMicrobe Interaction 13, 136142. Jung H.W. & Hwang B.K. (2000b) Pepper gene encoding a basic b-1,3- glucanase is differentially expressed in pepper tissues

ACKNOWLEDGMENTS
The authors wish to thank S. C. Lee for in situ hybridization experiment and Dr Laura Gilliland and Dr Scott Sattler for critically reading the manuscript. This work was nancially supported by the Korea Science & Engineering Foundation (grant no. R01- 2000-000-00086-0) and the Crop Functional Genomics Center of the 21st Century Frontier Research Program funded by the Ministry of Science and Technology of Republic of Korea (grant no. CG1224). H.W.J. is the recipient of a predoctoral fellowship from Korea University in 2003.

REFERENCES
Altschul S.F., Madden T.L., Schaffer A.A., Zhang J., Zhang Z., Miller W. & Lipman D.J. (1997) Gapped BLAST and PSIBLAST: a new generation of protein database search programs. Nucleic Acids Research 25, 33893402. Arondel V., Tchang F., Baillet B., Vignal F., Grellet F., Delsney M., Kader J.C. & Puigdomenech P. (1991) Multiple mRNA coding for phospholipid-transfer protein from Zea mays arise from alternative splicing. Gene 99, 133136. Arondel V., Vergnolle C., Cantrel C. & Kader J. (2000) Lipid transfer proteins are encoded by a small multigene family in Arabidopsis thaliana. Plant Science 157, 112. Arondel V., Vergnolle C., Tchang F. & Kader J.C. (1990) Bifunctional lipid-transfer: fatty acid-binding proteins in plants. Molecular and Cellular Biochemistry 98, 4956.

2003 Blackwell Publishing Ltd, Plant, Cell and Environment, 26, 915928

928 H. W. Jung et al.

upon pathogen infection and ethephon or methyl jasmonate treatment. Plant Science 159, 97106. Kader J.C. (1996) Lipid transfer proteins in plants. Annual Review of Plant Physiology and Plant Molecular Biology 47, 627654. Kader J.C., Julienne M. & Vergnolle C. (1984) Purication and characterization of a spinach-leaf protein capable of transferring phospholipids from liposomes to mitochondria or chloroplasts. European Journal of Biochemistry 139, 411416. Kessmann H., Staub T., Hofmann C., Maetzke T., Herzog J., Ward E., Uknes S. & Ryals J. (1994) Induction of systemic acquired disease resistance in plants by chemicals. Annual Review of Phytopathology 32, 439459. Kim K.D., Oh B.J. & Yang J. (1999) Differential interaction of a Colletotrichum gloeosporioides isolate with green and red pepper fruits. Phytoparasitica 27, 97106. Kim Y.J. & Hwang B.K. (2000) Pepper gene encoding a basic pathogenesis-related 1 protein is pathogen and ethylene inducible. Physiologia Plantarum 108, 5160. Kim Y.J., Hwang B.K. & Park K.W. (1989) Expression of agerelated resistance in pepper plants infected with Phytophthora capsici. Plant Disease 73, 745747. Lee Y.K., Hippe-Sanwald S., Lee S.C., Hohenberg H. & Hwang B.K. (2000) In situ localization of PR-1 mRNA and PR-1 protein in compatible and incompatible interactions of pepper stems with Phytophthora capsici. Protoplasma 211, 6475. Lemieux B. (1996) Molecular genetics of epicuticlar wax biosynthesis. Trends in Plant Science 1, 312318. Maldonado A.M., Doerner P., Dixon R.A., Lamb C.J. & Cameron R.K. (2002) A putative lipid transfer protein involved in systemic resistance signaling in Arabidopsis. Nature 419, 399403. Masuta C., Furuno M., Tanaka H., Yamada M. & Koiwai A. (1992) Molecular cloning of a cDNA clone for tobacco lipid transfer protein and expression of the functional protein in Escherichia coli. FEBS Letters 311, 119123. Molina A. & Garcia-Olmedo F. (1993) Developmental and pathogen-induced expression of three barley genes encoding lipid transfer protein. Plant Journal 4, 983991. Molina A. & Garcia-Olmedo F. (1997) Enhanced tolerance to bacterial pathogens caused by the transgenic expression of barley lipid transfer protein LTP2. Plant Journal 12, 669675. Molina A., Diaz I., Vasil I.K., Carbonero P. & Garcia-Olmedo F. (1996) Two cold-inducible genes encoding lipid transfer protein LTP4 from barley show differential responses to bacterial pathogens. Molecular and General Genetics 252, 162168. Molina A., Segura A. & Garcia-Olmedo F. (1993) Lipid transfer protein (nsLTPs) from barley and maize leaves are potent inhibitors of bacterial and fungal plant pathogens. FEBS Letters 316, 119122. Nielson K.K., Nielson J.E., Madrid S.M. & Mikkelsen J.D. (1996) New antifungal proteins from sugar beet (Beta vulgaris L.) showing homology to non-specic lipid transfer proteins. Plant Molecular Biology 31, 539552. Nishida I. & Yamada M. (1985) Semisynthesis of a spin-labeled monogalactosyldiacyglycerol and its application in the assay for galactolipid transfer activity in spinach leaves. Biochimica et Biophysica Acta 813, 298306. stergaard J., Vergnolle C., Schoentgen F. & Kader J. (1993) Acyl-binding lipid-transfer proteins from rape seedlings, a novel category of proteins interacting with lipids. Biochimica et Biophysica Acta 1770, 109117. Plant A.L., Cohen A. & Bray E.A. (1991) Nucleotide sequence and spatial expression pattern of a drought- and ABA-induced gene of tomato. Plant Physiology 97, 900906. Pyee J. & Kolattukudy P.E. (1995) The gene for the major cuticular wax-associated protein and three homologous genes from broccoli (Brassica oleracea) and their expression patterns. Plant Journal 7, 4959. Pyee J., Yu H. & Kollattukudy P.E. (1994) Identication of a lipid transfer protein as the major protein in the surface wax of broccoli (Brassica oleracea) leaves. Archives of Biochemistry and Biophysics 311, 460468. Ryals J., Neuenschwander U.H., Willits M.G., Molina A., Sterner H.Y. & Hunt M.D. (1996) Systemic acquired resistance. Plant Cell 8, 18091819. Ryals J., Uknes S. & Ward E. (1994) Systemic acquired resistance. Plant Physiology 104, 11091112. Salzman R.A., Tikhonova I., Bordelon B.P., Hasegawa P.M. & Bressan R.A. (1998) Coordinate accumulation of antifungal proteins and hexoses constitutes a developmentally controlled defense response during fruit ripening in grape. Plant Physiology 117, 465472. Sambrook J., Fritsch E.F. & Maniatis R. (1989) Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring. Harbor Laboratory Press, Cold Spring Harbor, NY. USA. Sanchez-Monge R., Lombardero M., Garcia-Selles F.J., Barber D. & Salcedo G. (1999) Lipid-transfer proteins are relevant allergens in fruit allergy. Journal of Allergy and Clinical Immunology 103, 514519. Segura A., Moreno M. & Garcia-Olmedo F. (1993) Purication and antipathogenic activity if lipid transfer proteins (LTPs) from the leaves of Arabidopsis and spinach. FEBS Letters 332, 243 246. Sossountzov L., Ruiz-Avila L., Vignols F., et al. (1991) Spatial and temporal expression of a maize lipid transfer protein gene. Plant Cell 3, 923933. Sterk P., Booij H., Schellekens G.A., Van Kammen A. & De Vries S.C. (1991) Cell-specic expression of the carrot EP2 lipid transfer protein gene. Plant Cell 3, 907921. Tchang F., This P., Stiefel V., et al. (1988) Phospholipid transfer protein: full-length cDNA and amino acid sequence in maize. Journal of Biology Chemistry 263, 1684916855. Terras F.R.G., Goderis I.J., Van Leuven F., Vanderleyden J., Cammue B.P.A. & Broekaert W.F. (1992) In vitro antifungal activity of a radish (Raphanus sativus L.) seed protein homologous to nonspecic lipid transfer protein. Plant Physiology 100, 1055 1058. Thoma S., Hecht U., Kippers A., Botella J., De Vries S. & Somerville C. (1994) Tissue-specic expression of a gene encoding a cell wall-localized lipid transfer protein from Arabidopsis. Plant Physiology 105, 3545. Thomas T.L., Vivekananda J. & Bogue M.A. (1991) ABA regulation of gene expression in embryos and mature plants. In Abscisic Acid: Physiology and Biochemistry (eds W.J. Davies & H.G. Jones), pp. 125135. Bios Scientic Publishers, Lancaster, UK,. Torres-Schumann S., Godoy J.A. & Pintor-Toro J.A. (1992) A probable lipid transfer protein gene is induced by NaCl in stems of tomato plants. Plant Molecular Biology 18, 749757. Trevino M.B. & OConnell M.A. (1998) Three drought-responsive members of the nonspecic lipid-transfer protein gene family in Lycopersicon pennellii show different developmental patterns of expression. Plant Physiology 116, 14611468. Vignols F., Lund G., Pammi S., Tremousaygue D., Grellet F., Kader J.C., Puigdomenech P. & Delseny M. (1994) Characterization of a rice gene coding for a lipid transfer protein. Gene 142, 265270. Vignols F., Wigger M., Garcia-Garrido J.M., Grekket F., Kader J.C. & Delseny M. (1997) Rice lipid transfer protein (LTP) genes belong to a complex multigene family and are differentially regulated. Gene 195, 177186. Von Heijne (1986) A new methods of predicted signal sequence cleavage sites. Nucleic Acids Research 14, 46834690. Received 18 July 2002; received in revised form 28 November 2002; accepted for publication 24 December 2002

2003 Blackwell Publishing Ltd, Plant, Cell and Environment, 26, 915928