Cloning of a New Pectate Lyase GenepelCfromFusarium solanif. sp.pisi(Nectria haematococca,Mating Type VI) and Characterization of the Gene Product Expressed inPichia pastoris

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS

Vol. 323, No. 2, November 10, pp. 352–360, 1995

Cloning of a New Pectate Lyase Gene pelC from Fusarium solani f. sp. pisi (Nectria haematococca, Mating Type VI) and Characterization of the Gene Product Expressed in Pichia pastoris1,2 Wenjin Guo, Luı´s Gonza´lez-Candelas,3 and P. E. Kolattukudy4 Neurobiotechnology Center, The Ohio State University, Columbus, Ohio 43210

Received May 31, 1995, and in revised form September 5, 1995

Antibodies prepared against a pectate lyase (PLA) produced by a phytopathogenic fungus Fusarium solani f. sp. pisi (Nectria haematococca, mating type VI) were previously found to protect the host against infection. The cDNA and gene (pelA) for PLA were cloned and sequenced. A new pectate lyase gene, pelC, was isolated from a genomic library of F. solani pisi with pelA cDNA as a probe. A 1.3-kb DNA fragment containing the pelC gene and its flanking regions was identified and sequenced. The coding region of pelC was amplified by reverse transcription–polymerase chain reaction using total RNA isolated from a pectininduced F. solani pisi culture as template. The open reading frame of pelC was predicted to encode a 23.3kDa protein of 219 amino acid residues, which shares 51% identity with PLA from F. solani pisi. No typical fungal leader peptide sequence could be identified at the N-terminus of the predicted protein sequence. The amplified pelC cDNA was expressed in Pichia pastoris yielding a pectate lyase C (PLC) with a molecular mass of 26.0 kDa and containing carbohydrates. PLC was purified to homogeneity using Mono Q anion-exchange chromatography. Purified PLC required Ca2/ for its activity and showed optimal lyase activity at pH 9.5 and 557C. Rapid drop in the viscosity of the substrate and Mono Q anion-exchange chromatography of the products generated by the lyase showed that PLC cleaved polygalacturonate chains in an endo fashion. 1 Sequence data from this article have been deposited in the GenBank data base under Accession No. U13049. 2 This work was supported by National Science Foundation Grant IBN-9318554. 3 Present address: Departamento de Biotecnologı´a, Instituto de Agroquı´mica y Tecnologı´a de los Alimentos, Jaime Roig 11, 46010Valencia, Spain. 4 To whom correspondence should be addressed at Neurobiotechnology Center, The Ohio State University, 206 Rightmire Hall, 1060 Carmack Rd., Columbus, OH 43210. Fax: (614)-292-5379.

Western blot using antibodies raised against PLA and PLC showed that PLC and PLA are immunologically related to each other. q 1995 Academic Press, Inc. Key Words: Fusarium solani f. sp. pisi; pelA; pelC; pectate lyase; PLA; PLC; Pichia pastoris.

Plant pectin, a complex heteropolysaccharide with a backbone of a-1,4-linked galacturonic acid residues interspersed with 1,2-linked rhamnose, constitutes a major component of the primary plant cell wall and middle lamella (23). Plant pathogenic fungi need to penetrate this physical barrier in order to establish infection. Enzymes that degrade pectin are frequently the first cell wall-degrading enzymes to be produced by pathogens in infected tissues (6, 16, 22). Depolymerization of pectin would not only provide a carbon source for fungal growth and development (11), but also expose other cell wall components to degradation, causing further cell wall breakdown (3). Elucidation of the role of pectinolytic enzymes in fungal pathogenesis requires an understanding of the nature of these enzymes produced by the pathogens. In culture, many plant fungal pathogens are known to produce different forms of pectic enzymes, including pectin lyases, pectate lyases, and polygalacturonases (8, 18, 19, 26, 31). The preference for highly methylated substrates differentiates pectin lyase from pectate lyase (7). Fusarium solani f. sp. pisi, a causative agent of foot rot of pea, produces both pectin hydrolases and pectate lyase when pectin is present as the sole carbon source in vitro. From the culture an exo-polygalacturonase, an endo-polygalacturonase, and an endopectate lyase were purified (8, 18, 19). Whether all of these pectindegrading enzymes are involved in pathogenesis remains unclear. When antibodies prepared against these enzymes were included in the spore suspension

352

0003-9861/95 $12.00 Copyright q 1995 by Academic Press, Inc. All rights of reproduction in any form reserved.

/ m4345f9164

10-16-95 12:11:00

arcal

AP: Archives

CLONING OF PECTATE LYASE C FROM F. solani pisi

placed on pea stem, the antibody against the lyase protected against lesion formation, whereas preimmune serum did not show any protection (8). The cDNA and gene for the lyase (pelA5) were cloned and sequenced (9). Disruption of pelA did not cause measurable decrease in the virulence of F. solani pisi on pea stem (Wenjin et al., unpublished work). Southern hybridization suggested possible presence of homologous pectate lyase genes in F. solani pisi genome. Protection by the antibodies and failure of the gene disruption to decrease virulence suggest that other lyase genes in the genome would encode lyases that might immunologically cross-react with pelA antibodies. However, other lyase genes have not been cloned and their products have not been examined. Here we report the cloning and sequencing of a novel pectate lyase gene, pelC, from F. solani pisi. The cDNA for pelC was also isolated by RT–PCR and sequenced. Characterization of the pelC product expressed in Pichia pastoris as an endopectate lyase that is immunologically related to the pelA product is reported. MATERIALS AND METHODS Fungal and bacterial strains. F. solani pisi field isolate T8 was obtained from H. D. Van Etten (University of Arizona). Single spore isolates were routinely maintained on potato dextrose agar containing 1 g/liter finely ground pea stem. F. solani pisi cultures were grown in mineral medium (12) containing 2% glucose on a rotary shaker at 200 rpm. Escherichia coli DH5a was used for construction of a F. solani pisi genomic library and propagation of plasmids. Nucleic acid isolation and analysis. Conidia (107) were inoculated in 100 ml of mineral medium containing 2% glucose and incubated for 48–72 h at 247C with shaking. Mycelia were harvested by filtration through Whatman No. 1 filters and frozen in liquid nitrogen. The samples were kept at 0807C until needed. Genomic DNA was isolated from the mycelia as described before (17). For Southern hybridization, restriction enzyme-digested DNA was fractionated on 0.8% agarose gel, transferred to nylon membrane using a vacuum blotting system (LKB 2016 VacuGene), and hybridized to a 32P-labeled DNA probe. Genomic library construction and screening. The genomic DNA from F. solani pisi was digested with XbaI and separated on 0.8% agarose gels. The DNA from gel segment corresponding to 2.5 to 20 kb was electroeluted, ligated to XbaI-digested pUC18, and used to transform E. coli DH5a cells. The library was screened by colony hybridization using labeled pelA cDNA under low stringency hybridization conditions at 607C in 61 SSC (11 SSC: 0.15 M NaCl and 0.015 M sodium citrate, pH 7.6) and 0.5% SDS. Membranes were washed twice at 607C for 30 min with 41 SSC and 0.1% SDS. DNA sequencing and sequence analysis. A genomic clone, pX56, which contained a 9.4-kb XbaI fragment and had sequence homology with pelA cDNA, was identified and isolated from F. solani pisi genomic library. A 1.3-kb HindIII–StuI fragment was subcloned from pX56 into pUC19 digested by HindIII–SmaI, yielding pPLC, and

5 Abbreviations used: EGTA, ethylene glycol bis(b-aminoethyl ether)-N,N*-tetraacetic acid; PAGE, polyacrylamide gel electrophoresis; PGA, polygalacturonic acid; pelA, pectate lyase A gene; pelC, pectate lyase C gene; PLA, pectate lyase A isoenzyme; PLC, pectate lyase C isoenzyme; RT–PCR, reverse transcription–polymerase chain reaction; SDS, sodium dodecyl sulfate.

/ m4345f9164

10-16-95 12:11:00

arcal

353

its nucleotide sequence was determined. DNA was sequenced using Sequenase 2.0 DNA sequencing kit from United States Biochemicals following the protocol recommended by the supplier. DNA strider 1.2 and the GCG packages were used to analyze and align DNA sequences. Amplification of pelC cDNA by RT–PCR. Conidia (107) were inoculated in 100 ml of mineral medium supplemented with 0.5% glucose and incubated at 247C with shaking. After glucose was depleted, citrus pectin (Sigma, 9.4% methoxy content) was added to this culture to a final concentration of 0.2%. Mycelia were harvested after 70 h of induction by filtration through Whatman No. 1 filters and frozen in liquid nitrogen. Isolation of total RNA from F. solani pisi was done as described (24). RNA was treated with amplificationgrade RNase-free DNase (BRL), reverse transcribed into cDNA with murine leukemia virus reverse transcriptase (BRL) using oligo(dT) primers. The first-strand cDNA was used directly for PCR amplification. For PCR, the sequence of sense primer was 5*-GGG GAA TTC ATG GCC TGC CTC GGA TA-3 *. The antisense primer was 5*-GGG GAA TTC CAA ACC GCT TAA GCA GG-3*. The PCR procedure consisted of a denaturation step at 947C for 5 min followed by 40 cycles of the following steps: denaturation at 947C for 30 s, annealing at 567C for 45 s, and extension at 727C for 2 min. A last elongation step was done at 727C for 15 min. The PCR product was purified from 0.8% agarose gel and cloned into pCR II vector (TA cloning kit, Invitrogen), yielding pTAC. Pichia pastoris transformation and induction of protein expression. P. pastoris growth, transformation, and induction of protein expression were done according to the user instructions provided with the expression kit by Invitrogen. A 678-bp EcoRI fragment from pTAC, containing the cDNA of pelC, was cloned into the EcoRI site of the polycloning sites of P. pastoris expression vector pHILS1, resulting in pHILS1C. P. pastoris transformants were selected on minimal dextrose plates which did not contain histidine. The alcohol oxidase gene (AOX-1) disruption mutants were selected based on their inability to grow on minimal methanol plates. Ten transformants were randomly selected to inoculate 10 ml buffered minimal glycerol medium (BMGY). The cells were harvested after 2 days of growth with shaking (200 rpm) at 307C and resuspended in 2 ml buffered minimal methanol medium and cultured for another 2 days. Pectate lyase activity in the culture supernatant was measured and the transformant which produced the highest pectate lyase activity was used to inoculate 1000 ml BMGY for large-scale expression. Pectate lyase purification. P. pastoris culture fluid was harvested by centrifugation of a 1 liter culture. To the supernatant (NH4)2SO4 was added to 75% saturation with stirring and the mixture was kept on ice for 2 h. The precipitate collected by centrifugation was dissolved in H2O, dialyzed against H2O, and lyophilized. The lyophilized enzyme was dissolved in 50 mM Tris–HCl, pH 8.5, and loaded onto a Mono Q HR 16/10 column (Pharmacia) for fast protein liquid chromatography. Pectate lyase was eluted from Mono Q with a 0 to 1.0 M linear NaCl gradient in 50 mM Tris–HCl, pH 8.5. Pectate lyase activity of aliquots of each fraction was measured and the fractions containing the most lyase activity were combined. The enzyme was stored at 0807C for future use. Enzyme assay. Pectate lyase activity was routinely determined by measuring the change in absorbance at 235 nm with 3 mg/ml PGA (Sigma) as substrate in 50 mM Tris–HCl, pH 9.5, containing 1 mM CaCl2 (8). Viscometric assays were run using an Ostwald capillary viscometer as described (8). Electrophoresis, Western blot, and detection of glycosylation. SDS–PAGE was done by the method of Laemmli (21), with a 12% separating gel and a 5% stacking gel. The GlycoTrack kit (Oxford GlycoSystems) was used to detect carbohydrate moieties in protein after SDS–PAGE and transblotting onto Immobilon P membrane as suggested by the manufacturer. For Western blot, the electrophoresed proteins were transblotted onto a nitrocellulose membrane in 10 mM cyclohexylaminopropane sulfonic acid transfer buffer (pH 10.5)/15% methanol. Blots were visualized using horseradish peroxi-

AP: Archives

´ LEZ-CANDELAS, AND KOLATTUKUDY GUO, GONZA

354

dase–protein A and the enhanced chemiluminescence system (Amersham) following the conditions recommended by the manufacturer. Antibody production and inhibition of pectate lyase activities by antibodies. The SDS–PAGE gel band corresponding to PLC (50 mg, purified) was cut out, mixed thoroughly with Freund’s adjuvant (complete for first injection, incomplete for subsequent injections), and injected subcutaneously into a rabbit; three additional booster injections of the same amount of purified lyase were given at 2-week intervals, after which blood was collected by heart puncture. The IgG fraction of the immune serum, obtained by (NH4)2SO4 precipitation, was used for subsequent experiments. The PLA antibody was prepared against PLA which was purified from culture fluid of F. solani pisi, as described (8). For enzyme activity inhibition assay, pectate lyase (15 mU) was incubated with different concentrations of immunoglobulin G in phosphate-buffered saline, pH 7.4, for 1 h at room temperature, and then the enzyme activity was measured as indicated above. Analysis of polygalacturonic acid degradation products by anionexchange chromatography. PGA (30 mg) was dissolved in 10 ml 50 mM Tris–HCl, pH 9.5, containing 1 mM CaCl2 . Purified PLC (0.2 U) was added to the solution, the reaction mixture was incubated at room temperature for 30 min, and the reaction was stopped by boiling the mixture for 10 min in a water bath. The reaction mixture was filtered through a 0.2-mm filter and loaded onto a Mono Q HR 16/ 10 column which was equilibrated with 20 mM Tris–HCl, pH 7.0, containing 100 mM NaCl. The products were eluted with a 100–400 mM NaCl linear gradient, as described (15).

RESULTS

Isolation and sequencing of pelC. To search for pectate lyase genes related to the previously cloned pelA in the genome of F. solani pisi, a library of 2.5 to 20 kb fragment of the genomic DNA from F. solani pisi, constructed in pUC18, was screened using labeled pelA cDNA probe under low stringency hybridization conditions. One clone, pX56, which showed hybridization, contained a 9.4-kb XbaI fragment. A 1.3-kb HindIII– StuI fragment from this clone hybridized with pelA cDNA. This fragment was subcloned into pUC19, yielding pPLC, and its nucleotide sequences were determined. Based on sequence alignment with pelA and conserved intron border sequences of filamentous fungi (1, 9, 10), the coding region of pelC was identified (Fig. 1). The nucleotide sequence of pelC revealed an open reading frame of 657 bp which was interrupted by two introns of 56 and 51 bp. The open reading frame would encode a protein of 219 amino acids with a calculated molecular mass of 23.3 kDa and a pI of 7.2. This protein shares 51% amino acid sequence identity with PLA from F. solani pisi. No typical fungal leader peptide sequence could be identified at the N-terminus of the protein deduced from the open reading frame. The 5* and 3* flanking regions of pelC were also sequenced. A TATATAA box was found 162 bp 5* to the ATG start codon. This sequence is homologous to the typical TATAAA motif found to be involved in the transcription of many fungal genes (1, 10). No typical polyadenylation signal (AATAAA) could be identified within the 441-bp 3* segment that was sequenced, but a similar motif,

/ m4345f9164

10-16-95 12:11:00

arcal

AACAAA, was found 175 bp downstream from the stop codon, at nucleotide 1138 (Fig. 1). RT–PCR and expression of the open reading frame in P. pastoris. Total RNA isolated from F. solani pisi culture induced by 0.2% pectin was used as template for RT–PCR. The PCR product that matched with the expected size of pelC cDNA (Fig. 2) was cloned into pCR II. The sequences of three randomly selected cDNA clones which hybridized with pelC genomic DNA matched exactly with that of the pelC genomic sequence except for the introns. The pelC cDNA insert from one clone was cloned into P. pastoris expression vector pHILS1 to yield pHILS1C. A native P. pastoris secretion signal from its acid phosphatase gene (PHO1) was used to direct PLC secretion. P. pastoris transformants harboring the vector were isolated and expression of PLC was determined by measuring the pectate lyase activity in the culture supernatant. The P. pastoris transformant which gave the highest expression was used for large-scale PLC expression and purification. Purification of the recombinant pectate lyase C. The recombinant PLC was purified to apparent homogeneity. A summary of the purification procedure is shown in Table I. When the lyase preparation obtained by (NH4)2SO4 precipitation was applied to the Mono Q column, all of the lyase activity was retained in the column. Upon application of a linear 0–1 M NaCl gradient, the lyase emerged as a single peak at about 0.22 M NaCl (Fig. 3A). SDS–PAGE of PLC obtained from this step revealed only a single Coomassie blue-staining band (at 26 kDa), indicating that the enzyme was purified to near homogeneity (Fig. 3B). Characterization of the recombinant pectate lyase C. The size of the recombinant PLC was 26.0 kDa as determined from SDS–PAGE, which was higher than the 23.3-kDa size calculated from the pelC coding sequence. This difference could be due to glycosylation, as the recombinant PLC responded positively to the GlycoTrack staining for glycoproteins (Fig. 3C). The purified recombinant pectate lyase showed a basic pH optimum of about 9.5, and the lyase activity dropped dramatically beyond pH 10 when PGA was used as the substrate (Fig. 4). With pectin as substrate, substantial lyase activity was observed at the physiologically relevant near-neutral pH. The influence of temperature on the lyase activity was tested at 57C intervals between 20 and 707C. Lyase activity increased with increasing temperature in a rectilinear manner until 557C and further increase in temperature caused a drastic decrease in activity. The optimal activity against PGA was observed at 557C (data not shown). The activity of PLC required the presence of Ca2/ (Table II). When 2 mM EGTA was added with 1 mM Ca2/, the pectate lyase activity was completely inhibited. The inhibition could be reversed by adding more Ca2/ (Ta-

AP: Archives

355

CLONING OF PECTATE LYASE C FROM F. solani pisi

FIG. 1. Nucleotide sequence of the cloned DNA fragment showing the open reading frame coding for PLC. The predicted TATA box in the 5* flanking region is boldfaced and underlined.

ble II). Monovalent and divalent metal ions could not substitute for Ca2/ except for a low level (Ç10%) of activity found with Mn2/. PLC showed a typical substrate saturation pattern with PGA resulting in a lin-

TABLE I

Purification of Recombinant Pectate Lyase C of F. solani pisi from the Culture Supernatanta of P. pastoris

Step

Total protein (mg)

Total activity (U)

Specific activity (U/mg)

Recovery (%)

Enrichment (fold)

240

291.4

1.21

100

—

24 1

99.6 74.4

4.15 74.4

34 26

3.4 61

Culture supernatant 75% (NH4)2SO4 Mono Q

FIG. 2. RT–PCR amplification of pelC cDNA. Lane 1, l DNA EcoRI–HindIII size marker; lane 2, PCR product.

/ m4345f9164

10-16-95 12:11:00

arcal

a

The starting volume of culture supernatant was 1000 ml.

AP: Archives

´ LEZ-CANDELAS, AND KOLATTUKUDY GUO, GONZA

356

FIG. 3. (A) Anion-exchange chromatography of the recombinant PLC from P. pastoris on Mono Q. (B) Coomassie blue-stained SDS–PAGE of P. pastoris culture supernatant (lane 1) and enzyme recovered from Mono Q (lane 2). Molecular mass markers (in kDa) are shown on the left. (C) Carbohydrate detection of the recombinant PLC (1 mg) using the GlycoTrack kit.

ear double-reciprocal plot (data not shown). The Km and Vmax were calculated to be 670 mg/ml and 110 U/mg enzyme, respectively.

To determine whether PLC had an endo or exo activity, the effect of PLC on the viscosity of a PGA solution was measured. The viscosity dropped by 50%, when approximately 2.6% of the total glycosidic linkages were cleaved, indicating the endo-acting nature of PLC (Fig. 5A). That PLC is an endopectate lyase was confirmed by analysis of the PGA degradation products by anion-exchange chromatography. After approximately 5% of the available bonds were cleaved, the partially digested PGA was loaded to a Mono Q column. More than 10 peaks of oligomers could be resolved from the column when a linear NaCl gradient of 100–400 mM was applied to it (Fig. 5B). TABLE II

Effect of Cations on Pectate Lyase Activity of PLC

FIG. 4. Activity of pectate lyase on polygalacturonic acid (Sigma) or pectin (Ç70% methoxy content, Sigma) at various pH values. Tris– HCl (50 mM) was used for pH 6.5–9.0 and glycine–NaOH (50 mM) was used for pH 9.0–11.0.

/ m4345f9164

10-16-95 12:11:00

arcal

Cations added

Activity (U/mg)a

Control (no addition) 1 mM Na/ 1 mM K/ 1 mM Cs/ 1 mM Co2/ 1 mM Mg2/ 1 mM Mn2/ 1 mM Zn2/ 1 mM Ca2/ 1 mM Ca2/, 2 mM EGTA 3 mM Ca2/, 2 mM EGTA

0 0 0 0 0 0 6.4 0 62.2 0 57.6

a Assay buffer: 50 mM Tris–HCl, pH 9.5, containing 3 mg/ml polygalacturonic acid.

AP: Archives

CLONING OF PECTATE LYASE C FROM F. solani pisi

357

FIG. 5. (A) Change in viscosity of the polygalacturonic acid solution due to the action of pectate lyase. Viscosity was measured with an Ostwald capillary viscometer at room temperature. The dashed line indicates 50% decrease in viscosity at approximate 2.6% bond cleavage. (B) Anion-exchange chromatographic analysis of polygalacturonic acid degradation products generated by the action of pectate lyase when 5% of the available glycosidic bonds were cleaved. PGA degradation products were loaded onto a Mono Q 16/10 HR column followed by elution with a linear NaCl gradient from 100 to 400 mM in 20 mM Tris–HCl, pH 7.0. The percentage of available bonds cleaved was calculated as described before (8).

To test whether PLC could cleave oligomers, the oligomers obtained by Mono Q fractionation of PLAtreated PGA were tested as substrates for PLA and PLC. Lyase activity obtained with the oligomers using identical units of PLA and PLC (as determined with

FIG. 6. Effect of the size of oligogalacturonides on the pectate lyase activities of PLA and PLC. Lyase activity was determined by measuring the change in absorbance at 235 nm with 40 mM of each oligomer as substrate in 50 mM Tris–HCl, pH 9.5. DP, degree of polymerization.

/ m4345f9164

10-16-95 12:11:00

arcal

FIG. 7. Western blot analysis of F. solani pisi PLA and the recombinant PLC expressed in P. pastoris, with antibody raised against PLA (A) and PLC (B). Molecular weights are shown on the left. Purified PLA and PLC were used.

AP: Archives

´ LEZ-CANDELAS, AND KOLATTUKUDY GUO, GONZA

358

FIG. 8. Inhibition of F. solani pisi PLC lyase activity with antibody raised against PLC (A) and PLA (B). The lyase activity is expressed as a percentage of the value for the control without antibody.

PGA as substrate) is shown in Fig. 6. As the degree of polymerization increased, lyase activity increased up to a DP of 7. When compared to PGA, the heptamer showed 17 and 22% lyase activity with PLA and PLC, respectively. To test whether PLC was immunologically related to PLA, Western blot analysis was done using the antibody raised against each enzyme (Fig. 7). The antibody raised against PLC severely inhibited the lyase activity of PLC, whereas the preimmune serum did not show inhibition (Fig. 8A). The PLA antibody also inhibited PLC activity, although at a much higher concentration when compared with that required to inhibit PLA (Fig. 8B). Southern blot analysis. F. solani pisi genomic DNA was digested with HindIII, PstI, SstI, and XbaI. Gel blots of the DNA fragments from different digests were hybridized with a 1.2-kb HindIII–BamHI DNA fragment from pPLC containing a partial pelC structural gene as the probe. The results showed only one major hybridization band in HindIII and XbaI digests, and two bands were found in PstI and SstI digests (Fig. 9). Since pelC contains one PstI site and one SstI site, these results suggest that the genome of F. solani pisi probably contains one copy of the pelC gene.

report we describe the cloning and sequencing of a novel pectate lyase gene, pelC, from the genome of F. solani pisi. The putative pelC product is encoded by three exons in F. solani pisi. The nucleotide sequence of the cDNA confirmed the presence of two introns in the genomic clone. The size of the introns (56 and 51 bp) in pelC, the 5* and 3* border sequences, and the putative lariat sequences are all typical of introns in genes of filamentous fungi (10). The DNA sequence surrounding the ATG translation start site of pelC (CGCCATGGC) closely resembles the Kozak sequence from filamentous fungi (CAC/AA/CATGNC) and mammals (GCCA/GCCATGG) (10, 20). The codon usage of pelC shows only a modest bias; 50 of 61 possible codons are used, but there is a strong preference for pyrimidines, especially for cytosine, in the wobble position. In general, a marked codon bias has been found in constitutively expressed or highly expressed genes, but not in genes expressed at low levels (10). The open reading frame of pelC predicts a protein (PLC) of 219 amino acids with a calculated molecular mass of 23.3 kDa, which is 2.7 kDa smaller than the molecular weight estimated based on SDS–PAGE gel. This difference seems to be due to glycosylation commonly found in secreted proteins. There is a consensus N-glycosylation site at asparagine 153, and O-linked glycosylation is also possible. In fact, glycoprotein detection using the GlycoTrack system showed the presence of carbohydrates in PLC expressed in P. pastoris. However, whether PLC produced within F. solani pisi is glycosylated is not known as we have been unable to isolate PLC from this organism.

DISCUSSION

F. solani pisi is known to produce different pectinolytic enzymes when grown in the presence of pectin as the sole carbon source (19). Two pectin hydrolases and one pectate lyase were purified from the culture filtrate and the gene encoding this lyase was cloned (8, 9, 18, 19). The lyase, which showed very little homology to the lyases previously cloned from other fungi, was induced by pectin and repressed by glucose (9). In this

/ m4345f9164

10-16-95 12:11:00

arcal

FIG. 9. Southern hybridization of restriction enzyme-digested genomic DNA from F. solani pisi with a 32P-labeled 1.2-kb BamHI– HindIII fragment from pPLC containing the coding sequence of pelC as the probe. Molecular weights (in kb) are shown on the right.

AP: Archives

CLONING OF PECTATE LYASE C FROM F. solani pisi

PLC shares 51% overall amino acid sequence identity with PLA, with the homologous regions clustered in the first three quarters (data not shown). The 5* and 3* flanking regions of pelC have little in common with the untranslated regions in pelA. Both PLA and PLC might exhibit similar three-dimensional structures. It is currently thought that proteins sharing 50% identity have highly related three-dimensional structures (2, 5). No significant homology has been found between pelC and pectinolytic enzyme-encoding genes cloned from other fungal sources (4, 13, 25, 28, 29). By analysis of pectate lyase gene sequences from E. chrysanthemi and E. carotovora and from plants, Barras et al. defined three consensus sequences in the gene products: (I) (D/ E)(G/S)-hDh--(A/G)(S/A)--hThS (wherein h stands for I, L, or V), (II) h--R-P--R-G-hH--NN-Y, and (III) (S/A/ T)--hWVDH--h (2). These consensus sequences are also found highly conserved in pectin lyases of E. carotovora and therefore were suggested as signatures of pectate/ pectin lyase activity (2). However, these sequences could not be identified in the predicted pelC product. Moreover, pelC does not have homology with the periplasmic pectate lyase gene family known to have high homology with each other (14, 30). To obtain PLC for characterization of the enzyme activity, the cDNA of pelC was amplified by RT–PCR and P. pastoris was used as the host to express it. Secretion by P. pastoris depends on the presence of a signal sequence in the expressed protein. Since the pelC coding region does not have an identifiable signal sequence, the vector pHILS1, which carries a native P. pastoris secretion signal from its acid phosphatase gene (PHO1), was used to express the cDNA of pelC. The recombinant PLC was secreted and was found to be fully functional as a pectate lyase. Expression of pelA cDNA using its own leader results in at least 25-fold higher expression than that obtained for pelC cDNA with the P. pastoris leader (data not shown). Efforts to express the open reading frame of pelC by introduction of the gene into P. pastoris have not been successful. Although introns of higher eukaryotes can be spliced out in other higher eukaryotes, P. pastoris seems to be stringent about sequence specificity for splicing in that it was necessary to use an intron-free pectate lyase gene in order to have it expressed in P. pastoris. PLC is similar to PLA in its activity on pectate. Like PLA, PLC has a basic pH optimum (pH 9.5), exhibits a dependence on Ca2/, and cleaves the glycosidic bonds in an endo-acting manner. The product distribution of PGA degradation by PLA was found to be identical to that shown for PLC in the present paper. Both PLA and PLC are also found to be immunologically related to each other, as shown by Western blot analyses and antibody inhibition assays. One significant difference between the two lyases is that PLC is more active than PLA on the more highly methylated pectins. The possible biological significance of this difference is not clear.

/ m4345f9164

10-16-95 12:11:00

arcal

359

Whether pelC is involved in the degradation of host cell wall polymers during pathogenesis is not known. So far our efforts to detect PLC protein in F. solani pisi culture filtrates and cell homogenates by Western blot analysis have not been successful. This could be due to its extremely low expression level under the growth conditions we used. Since the pelC coding sequence does not contain an identifiable signal sequence, it is possible that pelC encodes a cytoplasmic lyase, although no intracellular pectin-degrading enzymes have been detected in fungi heretofore. The function of an intracellular pectate lyase could be to degrade galacturonate oligomers produced by extracellular pectic enzymes and transported into the cytoplasm (14). It has been known that pectin-degrading enzymes could also activate host plant defense systems by producing elicitors of plant defense responses (2, 6, 15, 27). Degradation of such oligomeric elicitor molecules by the pathogen could reduce the defense response of the host and thus provide an advantage to the invading pathogen. ACKNOWLEDGMENTS We thank Debra Gamble for her assistance in preparing the manuscript and Dr. Clifford Beall for his help in analyzing the sequence data.

REFERENCES 1. Ballance, D. J. (1991) in Molecular Industrial Mycology: Systems and Applications for Filamentous Fungi. Mycology Series (Leong, S. A., and Berka, R. M., Eds.), Vol. 8, pp. 1–17, Dekker, New York. 2. Barras, F., Gijsegem, F. V., and Chatterjee, A. K. (1994) Annu. Rev. Phytopathol. 32, 201–234. 3. Bauer, W. D., Bateman, D. F., and Whalen, C. H. (1977) Phytopathology 67, 862–868. 4. Bussink, H. J. D., Buxton, F. P., Fraaye, B. A., deGraaff, L. H., and Visser, J. (1992) Eur. J. Biochem. 208, 83–90. 5. Chothia, C., and Lesk, A. M. (1986) EMBO J. 5, 823–826. 6. Collmer, A., and Keen, N. T. (1986) Annu. Rev. Phytopathol. 24, 383–409. 7. Collmer, A., Ried, J. L., and Mount, M. S. (1988) Methods Enzymol. 161, 329–335. 8. Crawford, M. S., and Kolattukudy, P. E. (1987) Arch. Biochem. Biophys. 258, 196–205. 9. Gonza´lez-Candelas, L., and Kolattukudy, P. E. (1992) J. Bacteriol. 174, 6343–6349. 10. Gurr, S. J., Unkles, S. E., and Kinghorn, J. R. (1987) in Gene Structure in Eukaryotic Microbes (Kinghorn, J. R., Eds.), pp. 93–139, IRL Press, Oxford. 11. Hahn, M. G., Bucheli, P., Cervone, F., Doares, S. H., O’Neill, R. A., Darvill, A., and Albersheim, P. (1989) in Plant–Microbe Interactions. Molecular and Genetic Perspectives (Kosuge, T., and Nester, E. W., Eds.), Vol. 3, pp. 131–181, McGraw-Hill, New York. 12. Hankin, L., and Kolattukudy, P. E. (1968) J. Gen. Microbiol. 51, 457–463. 13. Harmsen, J. A. M., Someren, M. A. K., and Visser, J. (1990) Curr. Genet. 18, 161–166. 14. Hinton, J. C. D., Sidebotham, J. M., Gill, D. R., and Salmond, G. P. C. (1989) Mol. Microbiol. 3, 1785–1795.

AP: Archives

´ LEZ-CANDELAS, AND KOLATTUKUDY GUO, GONZA

360

15. Jin, D. F., and West, C. A. (1984) Plant Physiol. 74, 989–992. 16. Jones, T. M., Anderson, A. J., and Albersheim, P. (1972) Physiol. Plant Pathol. 2, 153–166. 17. Ka¨mper, J. T., Ka¨mper, U., Rogers, L. M., and Kolattukudy, P. E. (1994) J. Biol. Chem. 269, 9195–9204. 18. Kolattukudy, P. E., Podila, G. K., Sherf, B. A., Bajar, M. A., and Mohan, R. (1990) in Advances in Molecular Genetics of Plant– Microbe Interactions (Hennecke, H., and Verma, D. P. S., Eds.), Vol. 1, pp. 242–249, Kluwer Academic, Dordrecht. 19. Kolattukudy, P. E., Sebastian, J., Ettinger, W. F., and Crawford, M. S. (1987) in Molecular Genetics of Plant–Microbe Interactions (Verma, D. P. S., and Brisson, N., Eds.), pp. 43–50, Nijhoff, Boston. 20. Kozak, M. (1987) Nucleic Acids Res. 15, 8125–8148. 21. Laemmli, U. K. (1970) Nature 227, 680–685. 22. Mankarios, A. T., and Friend, J. (1980) Physiol. Plant Pathol. 17, 93–104.

/ m4345f9164

10-16-95 12:11:00

arcal

23. McNeil, M., Darvill, A. G., Fry, S. C., and Albersheim, P. (1984) Annu. Rev. Biochem. 53, 625–663. 24. Pyee, J., Yu, H., and Kolattukudy, P. E. (1994) Arch. Biochem. Biophys. 311, 460–468. 25. Reymond, P., Dele´age, G., Rascle, C., and Fe`vre, M. (1994) Gene 146, 233–237. 26. Riou, C., Freyssinet, G., and Fe`vre, M. (1991) Appl. Environ. Microbiol. 57, 1478–1484. 27. Ryan, C. L., and Farmer, E. E. (1991) Annu. Rev. Plant Physiol. Mol. Biol. 42, 651–674. 28. Scott-Craig, J. S., Panaccione, D. G., Cervone, F., and Walton, J. D. (1990) Plant Cell 2, 1191–1200. 29. Templeton, M. D., Sharrock, K. R., Bowen, J. K., Crowhurst, R. N., and Rikkerink, E. H. A. (1994) Gene 142, 141–146. 30. Trollinger, D., Berry, S., Belser, W., and Keen, N. T. (1989) Mol. Plant–Microbe Interact. 2, 17–25. 31. Wattad, C., Dinoor, A., and Prusky, D. (1994) Mol. Plant–Microbe Interact. 7, 293–297.

AP: Archives