Molecular characterization of a cDNA encoding caffeoyl ... - CiteSeerX

J. Biosci., Vol. 22, Number 2, March 1997, pp 161-175. © Printed in India.

Molecular characterization of a cDNA encoding caffeoyl-coenzyme A 3-O-methyltransferase of Stellaria longipes XING-HAI ΖΗΑΝG† and C C CHINNAPPA* Department of Biological Sciences, University of Calgary, Alberta, Canada T2N 1N4 † Present address: USDA/ARS, Photosynthesis Research Unit, University of Illinois, Urbana, IL 61801-3838, USA. MS received 21 March 1996; revised 2 November 1996 Abstract. A cDNA clone encoding S-adenosyl-L-methionine: trans-caffeoyl-CoA 3-O-methyltransferase (EC 2·1·1·104; CCoAOMT) from Stellana longipes Goldie (long-stalked chickweed) was isolated and studied. Structural analysis of both the nucleotide sequence and the predicted amino acid sequence suggests that our cloned sequence encoded a CCoAOMT enzyme of Stellaria longipes, which shared overall structural similarity with other plant CCoAOMTs but exhibited certain distinct characteristics. Southern blot hybridization and cloning analyses indicating a small CCoAOMT gene family in the Stellana longipes genome and the absence of introns in the coding region of the cDNA- corresponding gene. Sequence variations in the coding region were found among three genotypes from geographically isolated populations. Higher levels of CCoAOMT mRNA were detected in stems and leaves than in roots. The cDNAencoded protein expressed in Eschendia coli was shown to utilize caffeoyl-CoA, but not caffeic acid or 5-hydroxy ferulic acid, as its substrate. Keywords. Caffeoyl-CoA 3-O-methyl transferase; caffeic acid 3-O-methyltransferase; cDNA; phenylpropanoid metabolism; Stellaria longipes.

1. Introduction General phenylpropanoid metabolism is involved in a wide range of fundamental aspects in plant growth and development. Many of the phenylpropanoid derivative products are also vital to the survival of plants as adaptive responses to various environmental stresses such as drought, chilling, wounding, irradiation and infection (Hahlbrock and Grisebach 1979; Hahlbrock and Scheel 1989; Bowles 1990; Dixon and Lamb 1990). Phenylpropanoid metabolism may also play a role, yet largely unknown, in plant's adaptation, phenotypic plasticity and genetic differentiation under different environmental conditions. Its regulatory pattern and kinetics could be a reflection of interaction between a plant and its surrounding environment. We are interested in understanding the physiological and molecular mechanisms involved in the responses of ecotypes of Stellaria longipes (Caryophyllaceae), a dicot herbaceous perennial, to different environmental conditions (Chinnappa and Morton 1984; Emery et al 1994; Kathiresan et al 1996). We studied an enzyme implicated in various physiological functions, S-adenosyl-L-methioninerrrans-caffeoyl-CoA 3-O-methyltransferase (EC 2·1·1·104; CCoAOMT), to investigate the possible involvement of phenylpropanoid *Corresponding author (Fax, 1403-289-9311; Email, [email protected]).

161

162

Xing Hai Zhang and C C Chinnappa

metabolism in phenotypic adaptation, population differentiation and evolution of S. longipes. CCoAOMT is an enzyme specific for the synthesis of trans-feruloyl-CoA from substrate trans-caffeoyl-CoA (Kneusel et al 1989; Pakusch et al 1989, 1991). Its product was shown to be a necessary intermediate in the biosynthesis of coniferyl alcohol—one of the three major precursors (along with p-coumaryl and sinapyl alcohols) for lignin formation within the cell walls of grasses and other herbaceous plants (Hahlbrock and Grisebach 1979; Kneusel et al 1989; Lewis and Yamamoto 1990; Ye et al 1994). The native CCoAOMT enzyme in parsley is a homodimer, existing at a moderately high level in tissues undergoing normal growth. Following elicitation, the level of CCoAOMT mRNA increases rapidly and transiently (Pakusch et al 1989; Schmitt et al 1991); this pattern of transcription response is typical of genes suggested to be involved in disease resistance in plants (Templeton and Lamb 1988; Lamb et al 1989). CCoAOMT was also suggested to take part in an alternative methylation pathway in Zinnia lignin biosynthesis (Ye et al 1994). Another O-methyltransferase, S-adenosyl-L-methionine:caffeic acid 3-O-methyltransferase (COMT; EC 2·1·1·6), has been more extensively investigated in many plants (Lewis and Yamamoto 1990; Bugos et al 1991; Edwards and Dixon 1991; Gowri et al 1991; Collazo et al 1992; Ye and Varner 1995). It is distinctly different from CCoAOMT and does not depend on a CoA-ester substrate (Pakusch and Matern 1991). CCoAOMT activity or mRNA transcription has been detected in cell suspension cultures of parsley (Petroselinum crispum, Apiaceae), Bishop's weed (Ammi majus, Apiaceae), carnation (Dianthus caryophyllus, Caryophyllaceae), safflower (Carthamus tinctorius, Asteraceae) (Pakusch et al 1989, 1991; Pakusch and Matern 1991; Schmitt et al 1991), carrot (Daucus carota, Apiaceae) (Kuhnl et al 1989) and Zinnia elegans (Asteraceae) (Ye et al 1994). The CCoAOMT cDNA sequence has been available for parsley (Schmitt et al 1991), Z. elegans (Ye et al 1994), S. longipes (Zhang et al 1995), alfalfa (Medicago sativa, Fabaceae; GenBank accession number:U20736), quaking aspen (Populus tremuloides, Salicaceae; U27116) and grape (Vitis vinifera, Vitaceae; Z54233). A partial cDNA sequence of CCoAOMT from Arabidopsis thaliana (Brassicaceae) was also reported (Zou and Taylor 1994). However, the more extensive studies of this enzyme or its coding gene have only been reported for parsley (Pakusch et al 1989,1991; Pakush and Matern 1991; Schmitt et al 1991) and Zinnia (Ye et al 1994; Ye and Varner 1995). Here, we report the study of a cDNA clone encoding CCoAOMT from S. longipes, its distinct sequence features, gene complexity, tissue-specific expression, genotypic variations in DNA and protein sequences and the substrate specificity of the expressed protein, as a part of a major investigation on the molecular evolution of this species.

2. Materials and methods 2.1

Plant material

Plants of S.longipes were originally collected from different populations (table 1). Each individual plant was designated a genotype and multiplied by vegetative propagation. The plants were maintained in growth chambers, growing annually at long-day and warm condition (22°C day, 18°C night, 16 h photoperiod; LDW) for 5 months

CCoAOMT of Stellaria longipes

163

Table 1. Genotypes of Stellaria longipes used for PCR tests.

following a 7-month treatment of short-day and cold condition (8°C day, 5°C night, 8 h photoperiod). The light density (400-700 nm) was approximately 15 0 m E m -2 s -1 . The plants were collected after growing for 3 weeks at LDW, ground in liquid nitrogen and stored at — 80°C until use. 2.2 Extraction of RNA and construction of a cDNA library Total RNA was extracted from leaves of genotype 5D as described previously (Zhang and Chinnappa 1994a). Poly(A+)-RNA was isolated from the total RNA and used to construct a cDNA library in λΖΑΡ II by reverse transcription (Zhang and Chinnappa 1994a), according to the manufacturer's manual (Stratagene). 2.3 Isolation, sequencing and analysis of cDNA clone In a process of obtaining cDNA probes as markers, 30 clones were randomly isolated from this cDNA library according to the manufacturer's manual (Stratagene) and used to hybridize with total RNAs from different tissues (leaf, stem or root) of S. longipes. Preliminary results identified a cDNA clone (c9) that exhibited a certain degree of tissue specificity of gene expression. This clone was completely sequenced and subjected to database searching. DNA sequence search was carried out using the BLAST network service from the National Center for Biotechnology Information (Bethesda, MD). The search result indicated that this cDNA clone resembles a cDNA sequence encoding parsley CCoAOMT (Schmitt et al 1991). Therefore, cDNA clone c9 was used for further study. United States Biochemical's DNA sequencing kit (version 2.0) was used for sequencing. Both strands of the cDNA were sequenced using commercial primers and synthetic oligonucleotides. Amino acid sequence was analysed using the Mac Vector™ Sequence Analysis Programs package (IBI). 2.4 Isolation of genomic DNA and Southern blot hybridization Genomic DNA was isolated from ten different genotypes (cytotypes or ecotypes; table 1) as described previously (Zhang and Chinnappa 1994a). DNA samples (10 µg

164


each) were digested overnight by restriction endonucleases (Pharmacia, 30-40 units per sample) at 37°C, separated on 1 % agarose gel, and blotted to Hybond™-N nylon membranes (Amersham). The coding region was amplified by polymerase chain reactions (PCRs) with cDNA c9 as template and oligos A and C as primers (figure 1). The DNA fragment was eluted from agarose gel and labelled with [α-32P] dCTP (Amersham). Prehybridization was done overnight at 65°C as described (Zhang and Chinnappa 1994b). After hybridization with the probe at 65°C overnight, the membranes were washed twice with 2 × SSC-1·5% sodium dodecyl sulphate (SDS) at 65°C for 15 min each exposed to Kodak XAR 5 X-ray film at — 80°C. 2.5 PCR amplification of c9 gene from genomic DNA PCRs were carried out in 50 µl of reaction mixture containing 10 mM Tris-HCl, pH 8.3, 50 mM KCl, 1·5-2·5 mM MgCl2,150 µΜ of each dNTP, 0·1 µΜ of each primer, 0·2-0·5 µg of genomic DNA or cDNA clone, and 2·5 units of Taq DNA polymerase. After an initial denaturation at 95°C for 2.5 min, the reaction mixture was subjected to amplification in a temperature cycler (RoboCycler ™ 40, Stratagene) for 30 cycles consisting 1·5 min at 95°C, 2 min at 54-56°C and 2 min at 72°C, followed by a 10 min incubation at 72°C. The sequences and positions of three primers (Α, Β and C) are shown in figure 1. Amplification products were separated by 1·5% agarose gel electrophoresis, blotted to a Hybond™-N membrane (Amersham) and hybridized at 65°C to fragments of the coding region labelled by extension of primer Β (figure 1) with Klenow DNA polymerase and [α-32P] dCT P. The membrane was washed twice with 0·5 × SSC-l% SDS at 65°C for 30 min each and exposed to X-ray film. Strong hybridizing signals were detected for all the samples, indicating that the PCR products were amplified from the CCoAOMT sequence. 2.6 Cloning and sequencing of PCR products PCR products of genomic DNA from genotypes 2990(2n = 26), 2I (2n = 52) and 5D (2n = 78) were cloned using TA cloning kit (Invitrogen). The plasmid DNA was isolated and analysed with enzyme restrictions. DNA sequencing was performed using Taq DyeDeoxy™ Terminator Cycle Sequencing Kit (Applied Biosystems) by the DNA Sequencing Lab at University of Calgary. To monitor the replication fidelity by Taq DNA polymerase, the cDNA c9 was used as a standard for each of the amplification and the sequencing reactions. 2.7

Northern blot analysis

Northern blot (20 µg total RNA per lane) was carried out as described previously (Zhang and Chinnappa 1994b). The probe was a fragment of cDNA c9 coding region (between nt 29-nt 734 by primers A and C; figure 1). The RNA blot was hybridized overnight at 60°C, washed with 4 × SSC-0·5% SDS at 60°C for 15-30 min, and exposed to X-ray film at - 80°C (Zhang and Chinnappa 1994b). Actin genes have been demonstrated to be expressed at the same level in different plant tissues (Hightower and Meagher 1985). Therefore, an actin gene clone from soybean was used as an internal standard to confirm that similar amounts of RNA were loaded in each lane.


165

Figure 1. Nucleotide and deduced amino acid sequences of CCoAOMT cDNA form S. longipes. Numbers at the top of the sequence show the nucleotide positions. The putative polyadenylation signal sequence is underlined. The sequences and positions of three primers (A, Β and C) for PCRs are presented with short vertical bars and horizontal arrows. The restriction site for EcoRV used in Southern blot hybridization is shown in bold. The GenBank/EMBL accession number for the nucleotide sequence reported in this paper is L22203.

166


2.8 Expression of cDNA c9 in E. coli E. coli stain XL-1 Blue cells harbouring the CCoAOMT cDNA c9 were grown in LB-broth with ampicillin to an optical density of 0·5. The expression was induced by addition of IPTG to a final concentration of 10 mM. After 4 h of induction, the cells were collected by centrifugation at 2,400 g and resuspended in a buffer containing 50 mM Tris-HCl, pH 8·0, 10 mM EDTA, 5 mM DTT and 10 µg/ml of leupeptin. Then, the cells were lysed by sonication (3 × 20 s) using XL-Sonicator® (Misonix) and centrifuged at 8,000 g for 10 min. The supernatant was used for SDS-polyacrylamide gel electrophoresis (PAGE) and enzyme assay. XL-1 Blue cells with or without pBluescript plasmid were used as controls. CCoAOMT activity was assayed according to Pakusch et al (1991). The specific activity is expressed in pkatal (1 pmol of substrate converted to product per second) per mg protein in the crude extract. Protein content was determined with Bio-Rad dye binding reagent using BSA as standard. 3. Results and discussion 3.1 CCoAOMT cDNA sequence comparison Complete sequence of cDNA clone c9 is shown in figure 1. It is 1013 bp long and contains a portion of 5'-noncoding region (17 nt), entire coding region (726-nt open reading frame predicting 241 amino acids) and 3'-noncoding region (270 nt). A putative translation start codon ATG initiates the largest ORF within this clone (Zhang et al 1995). The sequence flanking the start codon, ACGATGT (nt 15-21, figure 1), is similar to a consensus sequence (ACCATGG) identified as the optimal sequence for translation initiation by eukaryotic ribosomes (Kozak 1986; Lutcke et al 1987; Cavener and Ray 1991). Particularly, the base A at the - 3 position upstream from the start codon (AUG) is highly conserved among eukaryotic (including plant) mRNAs, which has been suggested to be important for translation (Kozak 1986; Cavener and Ray 1991). A putative polyadenylation signal sequence, AATAAA, appears in 107 bp downstream from the first termination codon TGA. A second stop codon TAG immediately follows the first one. There is extensive similarity of both nucleotide and deduced amino acid sequences between S. longipes cDNA c9 and the CCoAOMTs from parsley (Schmitt et al 1991), Z. elegans (Ye et al 1994), A. thaliana (Zou and Taylor 1994), alfalfa, grape and aspen, although gaps are needed for a maximum alignment (figure 2). A divergent region is found to be located in the amino-terminus (residues 2-11 for S. longipes, figure 2), which is the main cause of length variation of CCoAOMT proteins between species—241 amino acids for S. longipes and parsley, 242 for grape, 245 for Zinnia, and 247 for alfalfa and aspen. Presumably, this region has fewer evolutionary constraints and its sequence is not restricted to the enzyme activities. All CCoAOMT sequences contain 2 or 3 cysteines, which have been suggested to be involved in bridging the native dimer and/or in catalytic activity (Schmitt et al 1991). The putative CCoAOMT of S. longipes has a deduced molecular mass of 26·7 kD a and calculated isoelectric point (pI) of 5·2, which is in close agreement with those for parsley (molecular mass of 27·1 kDa, pI = 5·3), Zinnia (27·6 kDa) and aspen (27·9 kDa, pI = 5·16).


167

Figure 2. Comparison of the deduced CCoAOMT amino acid sequences from S. longipes, A. thaliana (partial sequence, Zou and Taylor 1994), P. crispum (parsley; Schmitt et al 1991), Z. elegans (Ye et al 1994), M. sativa (alfalfa; GenBank accession number: U20736), V. vinifera (grape; Z54233) and P. tremuloides (quaking aspen; U27116). Numbers on the right indicate the amino acid positions for S. longipes CCoAOMT, excluding gaps. Asterisks denote identical amino acids. Gaps (-) are used to maximize the match. Box A is the consensus domain constituting part of the cofactor S-adenosyl-L-methionine binding site. Box Β is the conserved domain found in different methyltransferases. a 14-amino acid motif unique to S. longipes CCoAOMT is underlined.

The deduced amino acid sequence of the S. longipes cDNA c9 contains two consensus motifs. One is a conserved domain, - E80 VGVFTGYS88-(box A in figure 2), which resembles the binding site (-ELGAYCGYS-) for S-adenosyl-L-methionine in animal catechol O-methyltransferases (EC 2·1·1·6; Vidgren et al 1994) as well as in other methyltransferases that use S-adenosyl-L-methionine as the methyl group donor (Willcock et al 1994). Particularly, the invariable Glycine82 has been demonstrated to be essential for the enzyme activity (Willcock et al 1994; Vidgren et al 1994). Another motif, -V177 GGIIAY183-, where two glycines (-GG-) are surrounded by hydrophobic residues, is highly similar to the domain [V(I, F)GGL(I, V)IG(A) Y(F)] found in other CCoAOMTs (box Β in figure 2). This motif has also been identified as a conserved domain in many DNA or protein methyltransferases (Ingrosso et al 1989; Lauster 1989) as well as in all reported cafeic acid 3-O-methyltransferases (EC 2·1·1·6). Furthermore, the deduced C CoAO MT proteins from S. longipes and parsley share very similar structural features throughout the sequence such as hydrophilicity, surface probability, backbone chain flexibility, amphiphilicity, antigenicity and secondary structure (data not shown). Overall, the evidence strongly suggests that the cDNA c9 encodes the CCoAOMT of S. longipes.

168


However, comparison of CCoAOMT amino acid sequences reveals a certain degree of regional divergence between S. longipes and other species. The amino acid identity between parsley, Zinnia, alfalfa, grape and aspen, if excluding the first 10-16 residues at the variable N-termini, is 92-93%. This level of identify indicates highly conserved CCoAOMT proteins, considering the broad distance of evolution between these species. However, even when the N-terminal residues are excluded from comparison, the sequence identities between Stellaria and those five species is only 55-57%. Likewise, the identity between the partial sequence of Arabidopsis and those of Stellaria and the other five species is 52-54%. Phylogenetically, Stellaria (subclass Caryophyllidae) would be closer to Arabidopsis or aspen (both belong to subclass Dilleniidae) than it is to parsley and alfalfa (both belong to subclass Rosidae), Zinnia (subclass Asteridae), or grape (subclass Malvanae) (Cronquist 1981). Therefore, the divergence of species evolution could not explain the unexpectedly high variations for CCoAOMTs between S, longipes and other plants. Thus, we propose that the S. longipes CCoAOMT studied represents a new member of the CCoAOMT gene family. Its biological roles may differ in certain ways from those of CCoAOMT in parsley which has been implicated in disease resistance (Schmitt et al 1991), or in Zinnia which has been demonstrated to be specifically involved in lignin synthesis (Ye et al 1994; Ye and Varner 1995). Further analysis of the CCoAOMT enzyme, its isoenzymes and their coding genes in Stellaria would help obtain more information about this gene family. As for the Arabidopsis CCoAOMT, the complete sequence is remained to be cloned in order to further analyse its similarity to those of other plants. It is worth to note that a 14-amino acid motif located near the carboxyl terminus (underlined residues 196-210; figure 2), which is nearly identical for other five plants, is completely different for S. longipes, and, to a less extent, also divergent for Arabidopsis. This motif may represent a unique protein domain that constitutes the distinct functionality of Stellaria CCoAOMT enzyme. However, until this motif is assigned a functional role, the correlation between the protein sequence variations and the enzymatic differences among those reported CCoAOMTs remain to be revealed. 3.2 Analyses of CCoAOMT gene in the S. longipes genome To determine the complexity of the CCoAOMT gene, genomic DNA from genotype 5D was digested with SacI, EcoRV, Bam HI, HindIII and DraI, and hybridized to the coding region of cDNA c9. The cDNA c9 sequence contains a restriction site for EcoRV (nt 333- nt 338, figure 1) and no restriction site for other enzymes. One strong and one weak band was observed in the Bam HI blot, which may indicate improper enzyme digestion or hybridization. For the blots of SacI, HindIII and DraI, there appeared two strong bands ranging in size from ca. 2 kb to 8 kb (figure 3), suggesting a two-gene family. Fragments of less intensities were also detected, indicating the existence of other CCoAOMT-like genes. For the EcoRV blot, there is a restriction site near the 5'-end of the coding region. The two strong hybridizing bands observed (figure 3) should represent two distinct genes hybridizing to the major portion of the probe, whereas the fragments of less intensities may result from the hybridization with the minor 5'-fragment of the probe. Overall, the hybridization results suggest a small CCoAOMT gene family of at least two members in the S. longipes genome. Previous


169

Figure 3. Southern blot analysis of CCoAOMT - like nucleotide sequences in the S. longipes genome.Genomic DNA (10 µg per lane) from genotype 5D was digested with SacI(S),EcoRV (E), BamHI (Β), HindIII (Η) and DraI (D), and hybridized to [32P] labelled coding region of CCoAOMT cDNA. The size standards in kb were from BamHI/HindIII-digested λ DNA.

studies show that there are either a single copy gene or a two-gene family of CCoAOMT in parsley (Schmitt et al 1991), a single gene in A. thaliana (Zou and Taylor 1994), and a multi-gene family in Z. elegans (Ye et al 1994).

170


To examine the existence of introns in the CCoAOMT gene and thus the possible variations in intron sequences from different genotypes of geographically isolated populations, PCRs were carried out to amplify the CCoAOMT-like sequences using genomic DNAs of 10 representative genotypes (table 1). Two primers, located at each end of the cDNA coding region (figure 1), cover 235 out of 241 amino acids of the CCoAOMT. Only a single size of amplified products on genomic DNA was detected by both gel electrophoresis and hybridization to the cDNA coding region, in the same size as the band of 706 bp amplified on the cDNA clone. Sequencing some of the PCR clones confirmed that the amplified genomic fragments were indeed from the CCoAOMT-coding gene (data not shown). This suggests that the cDNA-corresponding CCoAOMT gene does not contain an intron in its coding region. However, the existence of introns in other member(s) of the CCoAOMT gene family remains to be examined. The studies on parsley and Zinnia CCoAOMT cDNAs did not reveal whether they contained an intron (Schmitt et al 1991; Ye et al 1994). Recently, a parsley CCoAOMT gene sequence has been released (GenBank accession numbers: Z33878 and Z54183). It apparently does not encode the same mRNA as the one reported by Schmitt et al (1991) because of different 5'-noncoding sequences, although both predict an identical amino acid sequence. This CCoAOMT gene contains four introns, all in coding region with total intron size of 654 bp. Cloning of a partial sequence of an aspen CCoAOMT gene also reveals the existence of introns in its coding region (Xing-Hai Zhang and C C Chinnappa, unpublished data). Since the S. longipes CCoAOMT gene studied is intronless, its gene organization associated with gene evolution seems different from the ones known in parsley or aspen. 3.3

Genotypic variation in the CCoAOMT coding sequence of S. longipes

We anticipate that ecotype variation and population differentiation may be caused by, at least in part, genomic variations and therefore can be assessed by molecular means such as gene sequence divergence (Zhang and Chinnappa 1994b). Three representative ecotypes (genotype acessions: 2990, 2I and 5D) originated from geologically isolated populations were compared for their CCoAOMT coding sequences. These genotypes also represented three different cytotypes in this species complex—diploid 2990, tetraploid 2I and hexaploid 5D; table 1). Genotypic variations in the gene sequence were revealed. There were 2·3% of divergence in nucleotide sequence and 2·6% in deduced amino acid sequence between genotypes 5D and 2990. The divergence between genotypes 5D and 2I was 5·7% for nucleotide sequence and 6·0% for deduced amino acid sequence, which is about 2-fold higher than the divergence between genotypes 5D and 2990 (sequences not shown). Since Taq DNA polymerase for PCR-amplifying the target sequence has lower fidelity of DNA replication than other thermo-labile polymerases, it is possible that the sequence variations observed between genotypes might be partly due to the incorporation error during PCR amplifications. However, the sequence of the PCR clone from genotype 5D was identical to the sequence of the cDNA c9 which was made from 5D, indicating that the error rate (if any) seems unlikely to effect our conclusion. Therefore, we believe that the sequence variations in CCoAOMT coding regions do exist among three distinct ecotypes from isolated populations, and should reflect genotypic changes of the gene sequence during population differentiation. The DNA sequence variations probably resulted from the


171

plant's adaptive responses to different environmental conditions exerting on separate populations during a long period of isolation. It has been observed that in the same genotype of Petunia, the small subunit of ribulose-l, 5-bisphosphate carboxylase/oxygenase (rbcS) sequence diverges by 1020% between different members of the gene family at both the nucleotide and amino acid sequence levels (Dean et al 1987). Similar obervations have been also reported for the rbcS gene members from tomato (Pichersky et al 1986) and tobacco (Mazur and Chui 1985). The variation (2-6%) in the coding sequence of the CCoAOMT gene observed in different genotypes of S. longipes may represent another example of gene evolution under different ecological environments, and/or divergence within the gene family if these PCR clones represent different gene members. The Intraspecific variation observed in S. longipes provides a clue of the possible relationship between gene evolution and population differentiation in this species. 3.4

Expression of CCoAOMT gene in S. longipes

The gene expression of CCoAOMT in S. longipes was studied by Northern blot hybridizations. Equal amounts of RNAs from roots, stems and leaves of genotype 5D (figure 4B) were hybridized to [ 32P]-labelled coding region of cDNA c9. Al·4 kb transcript was detected (figure 4A). The concentration of CCoAOMT mRNA in extraction from stems was higher than from leaves, and much higher than from roots, suggesting tissue-preferential gene expression. This may be related to the different states of cellular metabolism in those tissues, in terms of cell division and possible lignin synthesis associated with cell wall formation. Our results are different from those of parsley where CCoAOMT mRNA levels in roots and leaves were similar (Schmitt et al 1991), and of Zinnia which showed higher enzyme activity of CCoAOMT in root than

Figure 4. Abundance of the CCoAOMT mRNA in different tissues of S. longipes. (A) Total RNA (20 µg per lane) from leaves (L), stems (S), and roots (R) was hybridized to[32P] labeled coding region of the CCoAOMT cDNA. (B) The same blot was stripped and hybridized to [32P] labelled actin gene of soybean. RNA size in kb was estimated with the RNA molecular weight markers (Boehringer Mannheim).

172


Figure 5. SDS/PAGE analysis of the E. coli-expressed cDNA c9 protein. Crude protein extracts were from cells of XL-1 Blue/pBluescript (lane 1), XL-1 Blue/c9 uninduced (lane 2), and XL-1 Blue/c9 induced by IPTG (lane 3). The putative protein translated from cDNA c9 was shown by an arrow. Protein molecular mass markers (lane M) from Bio-Rad were indicated in kilodaltons. Table 2. Expression of CCoAOMT activity in E. coli XL-1 Blue cells containing cDNA c9 construct.

in leaf or stem (Ye et al 1994). S. longipes CCoAOMT also differs from the CoAindependent lignin caffeic acid O-methyltransferase (COMT; EV 2·1·1·6) whose mRNA level was observed to be higher in roots and stems than in leaves of alfalfa, maize (Gowri et al 1991; Collazo et al 1992) and aspen (Bugos et al 1991). To provide evidence for the identity of the Stellaria CCoAOMT cDNA clone, the c9 construct was expressed in E. coli strain XL-1 Blue under the induction of IPTG. Crude protein extracts were subjected to SDS/PAGE. A unique protein was observed only in E. coli XL-1 Blue cells containing the cDNA c9 construct, regardless IPTG induction (figure 5). The molecular mass of the translated protein (ca. 28 kDa) is very similar to that of the deducted CCoAOMT protein (26·7 kDa; figure 5). The cell crude extracts were also tested for activity against caffeoyl-CoA, its assumed true substrate, and two other phenylpropanoid pathway intermediates caffeic acid and 5-hydroxy ferulic acid which are the substrates for COMT. As shown in table 2, no activity was found in the control extracts from XL-1 Blue or XL-1 Blue containing pBluescript plasmid without the insert cDNA. Four hours after the addition of IPTG, a 3-fold induction of CCoAOMT activity towards caffeoyl-CoA was detected in XL-1 Blue cells harbouring cDNA c9, compared with the uninduced cultures. Since the cDNA c9 is out of frame

CCoAOMT of Stellana longipes

173

with the lacZ gene in pBluescript, translation of the c9 protein occurs from its own start codon. The expressed protein did not exhibit any activity against cafeic acid or 5-hydroxy ferulic acid. This suggests that the cDNA c9 studied here indeed encodes cafeoyl-CoA ester-specific enzyme whose functionality is not same as that of COMT. CCoAOMT and its gene have been studied only recently in few plants. The exact physiological role of CCoAOMT in phenylpropanoid metabolism, especially in lignin synthesis and cell wall formation as well as in plant defense system, is still not clear. CCoAOMT was proposed to be involved in an alternative methylation pathway in lignin biosynthesis in Zinnia, which was particularly dominant in differentiating tracheary element (Ye et al 1994). In parsley, however, this enzyme seemed to be more related to disease resistance than to tissue lignification (Schmitt et al 1991). CCoAOMT from S. longipes, on the other hand, exhibits some unique features such as genotypic sequence variation, absence of introns, and tissue specificity of gene expression, whereas the expressed protein seems able to utilize caffeoyl-CoA as a substrate like the CCoAOMT enzymes from parsley or Zinnia. It would be essential to further study the biological functions of this CCoAOMT, such as comparison of the expressed protein with the native enzyme, analysis of enzyme specificity towards other pathway intermediates, especially different CoA-esters, and relation of the CCoAOMT activity to tissue lignification or response to environmental stresses, so that a better knowledge of the involvement of phenylpropanoid metabolism in plant adaptation, phenotypic plasticity and population differentiation can be obtained. Acknowledgements We thank Dr Μ Μ Moloney for the lab facility, Dr Ulrich Matern's lab of Universität Freiburg, Germany, for providing caffeoyl-CoA, Xiao-Lu Jin for help with cloning and sequencing of PCR products, and Laigeng Li for assisting the SDS/PAGE protein analyses. This research was funded by an operation grant from Natural Sciences and Engineering Research Council of Canada to CCC, and a thesis research grant from University of Calgary to X-H Z. References Bowles D 1990 Defense-related proteins in higher plants; Annu. Rev. Biochem. 59 873-907 Bugos R C, Chiang V L C and Campbell W H 1991 cDNA cloning, sequence analysis and seasonal expression of lignin-bispecific caffeic acid/5-hydroxyferulic acid O-methyltransferase of aspen; Plant Mol. Biol. 17 1203-1215 Cavener D R and Ray S C 1991 Eukaryotic start and stop translation sites; Nucleic Acids Res. 19 3185-3192 Chinnappa C C and Morton J Κ 1984 Studies on the Stellaria longipes complex (Caryophyllaceae)— Biosystematics; Syst. Bot. 9 60-73 Collazo P, Montoliu L, Puigdomenech Ρ and Rigau J 1992 Structure and expression of the lignin O-methyltransferase gene from Zea mays L; Plant Mol. Biol. 20 857-867 Cronquist A 1981 An integrated system of classification of flowering plants (New York: Columbia University Press) pp 20-21 Dean C, Van Den Elzen, P, Tamaki S, Black M, Dunsmuir Ρ and Bedbrook J 1984 Molecular characterization of the rbcS multi-gene family ofPetunia (Mitchel); Mol. Gen. Genet. 206 465-474 Dixon R A and Lamb C J 1990 Molecular communication in interactions between plants and microbiol pathogens; Annu. Rev. Plant Physiol. Plant Mol. Biol. 41 339-367 EmeryRJN, Reid D Μ and Chinnappa C C 1994 Phenotypic plasticityofstemelongationintwo ecotypes of Stellaria longipes: the role of ethylene and response to wind; Plant Cell Environ. 17 691-700

174


Edwards R and Dixon R A 1991 Purification and characterization of S-adenosyl-L-methionine: cafeic acid 3-O-methyltransferase from suspension cultures of alfalfa (Medicago sativa L.); Arch Biochem. Biophys. 287 372-379 Gown G, Bugos R C, Campbell W H, Maxwell C A and Dixon R A 1991 Stress responses in alfalfa (Medicago sativa L.) X. Molecular cloning and expression of S-adenosyl-L-methionine: cafeic acid 3-O-methyltransferase, a key enzyme of lignin biosynthesis; Plant Physiol. 97 7-14 Hahlbrock Κ and Grisebach Η 1979 Enzymic controls in the biosynthesis of lignin and flavonoids; Annu Rev. Plant Physiol. 30 105-130 Hahlbrock Κ and Scheel D 1989 Physiology and molecular biology of phenylpropanoid metabolism; Annu. Rev. Plant Physiol. Plant Mol. Biol. 40 347-369 Hightower R C and Meagher Ρ Β 1985 Divergence and differential expression of soybean actin genes; EMBO J.4 1-8 Ingrosso D, Fowler A V, Bleibaum J and Clarke S 1989 Sequence of the D-aspartyl/L-iso-aspartylprotein methyltransferase from human erythrocytes; J. Biol. Chem. 264 20131-20139 KathiresanA, Reid D Μ and Chinnappa C C 1996 Light- and temperature-entrained circadian regulation of activity and mRNA accumulation of 1-aminocyclopropane-l-carboxylic acid oxidase in Stellaria longipes; Planta 199 329-335 Kneusel R E, Matern U and Nicolay Κ 1989 Formation of trans-caffeoyl-CoA from trans-4-coumaroyl-CoA by Zn2+ - dependent enzymes in cultured plant cells and its activation by an elicitor-induced pH shift; Arch. Biochem. Biophys. 269 455-462 Kozak Μ 1986 Point mutations define a sequence flanking the AUG initiator codon that modulates translation by eukaryotic ribosomes; Cell 44 283-292 Kühnl Τ, Koch U, Heller W and Wellmann Ε1989 Elicitor induced S-adenosyl-L-methioninexaffeoyl-CoA 3-O-methyltransferase from carrot cell suspension cultures; Plant Sci. 60 21-25 Lamb C J, Lawton Μ A, Dron Μ and Dixon R A1989 Signals and transduction mechanisms for activation of plant defenses against microbial attack; Cell 56 215-224 Lauster R 1989 Evolution of type II DNA methyltransferases. A gene duplication model; J. Mol. Biol. 206 313-321 Lewis Ν G and Yamamoto Ε 1990 Lignin: occurrence, biogenesis and biodegradation; Annu. Rev. Plant Physiol. PlantMol. Biol. 41 455-496 Lütcke Η, Chow Κ, Mickel F, Moss K, Kern Η and Scheele G 1987 Selection of AUG initiation codons differs in plants and animals; EMBO J. 6 43-48 Mazur Β J and Chui C-F 1985 Sequence of a genomic DNA clone for the small subunit of ribulose bisphosphate carboxylase-oxygenase from tobacco; Nucleic Acids Res. 13 2373-2386 Pakusch A-E,Kneusel R Ε and Matern U 1989 S-Adenosyl-L-methionine: trans-caffeoyl-CoA 3-O-methyltransferase from elicitor-treated parsley cell suspension cultures; Arch. Biochem. Biophys. 271 488-494 Pakusch A-E, Matern U and Schütz Ε 1991 Elicitor-inducible caffeoyl-coenzyme A 3-O-methyltransferase from Petroselinum crispum cell suspensions; Plant Physiol. 95 137-143 Pakusch A-E and Matern U 1991 Kinetic characterization of cafeoyl-coenzyme ΑAspecific 3-O-methyltransferase from elicited parsley cell suspensions; Plant Physiol. 96 327-330 Pichersky E, Bernatzky R, Tanksley S D and Cashmore A R 1986 Evidence for selection as a mechanism in the concerted evolution of Lycopersicon esculentum (tomato) genes encoding the small subunit of ribulose-l,5-bisphosphate carboxylase/oxygenase; Proc. Natl. Acad. Sci. USA 83 3880-3884 Schmitt D, Pakusch A-E and Matern U 1991 Molecular cloning, induction, and toxonomic distribution of cafeoyl-CoA 3-O-methyltransferase, an enzyme involved in disease resistance; J. Biol. Chem. 266 17416-17423 Templeton Μ D and Lamb C J 1988 Elicitors and defense gene activation; Plant Cell Environ. 11 395-401 Vidgren J, Svensson L A and Liijas A 1994 Crystal structure of catechol O-methyltransferase; Nature (London) 368 354-358 Willcock D F, Dryden D Τ F and Murray Ν Ε 1994 A mutational analysis of the two motifs common to adenine methyltransferases; EMBO J. 13 3902-3908 Ye Z-H, Kneusel R E, Matern U and Varner J Ε 1994 An alternative methylation pathway in lignin biosynthesis in Zinnia; Plant Cell 6 1427-1439 Ye Z-H, and Varner J Ε 1995 Differential expression of two O-methyltransferases in lignin biosynthesis in Zinnia elegans; Plant Physiol. 108 459-467 Zhang X-H and Chinnappa C C 1994a Molecular cloning of a cDNA clone encoding cytochrome c of Stellaria longipes (Caryophyllaceae)—and the evolutionary implications; Mol. Biol. Evol. 11 365-375


175

Zhang X-H and Chinnappa C C 1994b Triose phosphate isomerase of Stellaria longipes (Caryophyllaceae); Genome37148-156 Zhang X-H, Dickson Ε Ε and Chinnappa C C 1995 Nucleotide sequence of a cDNA clone encoding caffeoyl-coenzyme A 3-O-methyltransferase of Stellana longipes (Caryophyllaceae); Plant Physiol. 108 429-430 Zou J and Taylor D C 1994 Isolation of an Arabidopsis thaliana cDNA homologous to parsley (Petroselinum crispum) S-adenosyl-L-methionine: trans-caffeoyl-Coenzyme A 3-O-methyltransferase, an enzyme in volved in disease resistance; Plant Physiol. Biochem. 32 423-427 Corresponding editor: S AMIR Κ BRAHMACHARI