Gene, 117 (1992) 145-150 © 1992 Elsevier Science Publishers B.V. All rights reserved. 0378-1119/92/$05.00

145

GENE 06537

Cloning and expression in Escherichia coli of a xylanase-encoding gene from the yeast Cryptococcus albidus (Recombinant DNA; eDNA; intron; exon; polymerase chain reaction; glycosylation)

Rolf Morosoli, Serge Durand and Alain Moreau Centre de Recherche en Microbiologie Appliqude, lnstitut Armand-Frappier. UniversitE du QuEbec, ViUede Lavai. QuEbec. H7N 4Z3 (Canada)

Received by: J.K.C. Knowles: 19 March 1991; Revised/Accepted: 28 January/24 February 1992; Received at publishers: 1 April 1992

SUMMARY

In the yeast, Cryptococcus albidus, a comparison between the sequence of the xylanase (XLN)-encoding chromosomal gene (XLN) and the cDNA sequence reveals the presence of seven introns, ranging in length from 51 to 69 bp. One of their 5' splice site sequences is similar to the consensus sequence for yeast, while the other six resemble the consensus sequence for higher eukaryotes. Their 3'end splice site sequences are representative of the conserved sequence found in eukaryotes. Their putative branching point sequences are different from the well-known conserved sequence, 5'-TACTAAC, observed in yeast, but again resemble the mammzlian one. The eDNA encodingXLN is expressed by Escherichia coil, under the control of the lacZ promoter. The gene product remains inside the cell and has a molecular size of 40 kDa, which matches the size of the nonglycosylated protein. When compared to the glycosylated enzyme, the nongiycosylated XLN from E. coli shows twofold less affinity for substrate and its Vmaxis 100-fold lower. Moreover, the nonglycosylated XLN only acts on large xylan polymers and very slightly on xylohexaose.

INTRODUCTION

The yeast strain C. aibidus secretes a 48 kDa XLN (Morosoli et al., 1986). This enzyme is highly glycosylated, but when cells are grown in presence of Tu, a known inhibitor of protein glycosylation, an active XLN having a molecular size of 40 kDa is secreted in the medium (Morosoli et al., 1988). Since Tu inhibits only the N-(Asn) glycosyla-

Correspondence to: Dr. R. Morosoli, Institut Armand-Frappier, 531 Blvd, des Prairies, Ville de Laval, Qu6bec, H7N 4Z3 (Canada), Tel. (514)687-5010; Fax (514)686-5501.

Abbreviations: bp, base pair(s); C.. Cryptococcus; cDNA, DNA complementary to RNA; ds, double strand(ed); IU, international unit(s); kb,

kilobase(s) or 1000 bp; N, any nucleoside; nt, nucleotide(s);oligo, oligodeoxyribonucleotide;PAGE, polyacrylamide-geielectrophoresis;PCR, polymerase chain reaction; R, A or G; re, recombinant; SDS, sodium dodecyl sulfate;Tu, tunicamycin;W, A or T; XLN, xylanase;XLN, gene encoding XLN; Y, C or T.

tion, the enzyme could still contain some O-glycosylation (Tanner and Lehle, 1987). In order to study the specific role of the carbohydrate moiety in enzymatic activity, it appeared suitable to prepare a carbohydrate-free enzyme by cloning the X L N gene in E. coil, a host which does not glycosylate foreign proteins. Several XLN-encoding genes from prokaryotes have been cloned in E. coil and in few cases, XLNs have been secreted in the culture supernatants (Morosoli etal., 1990). Production of enzyme by this method was sufficient to test the biochemical properties of the cloned enzyme. The genomic X L N gene from C. albidus had already been sequenced (Boucher et al., 1988) and the low-resolution S 1-mapping, at this time, had revealed the presence of introns. In order to express the X L N gene in E. coil, a full-length eDNA had to be synthesized from the mRNA encoding the X L N gene. The aim of the present study was to prepare eDNA, to sequence it, and to express it in E. coil, as to determine the properties of the nonglycosylated reXLN.

146

a

EXPERIMENTAL AND DISCUSSION

(a) lntrons Isolation of full-length c D N A clones having an insert corresponding to the X L N m R N A had previously been unsuccessful (Morosoli and Durand, 1988). For this reason, a specific X L N sequence was amplified by the PCR method, using at the 5' end an oligo (30 nt) corresponding to a part of the X L N sequence and at the 3' end an oligo with a poly(T) tail which annealed to the poly(A) region of the eDNA. The amplified sequence (1020 bp) was resolved on agarose gel (Fig. 1), and the fragment was restricted by Kpnl, by Sstl and by BamHI, producing three smaller fragments of 380 bp, 290 bp and 350 bp, respectively. These restriction fragments were ligated to vectors M l3mpl8/ mpl9, and their sequences were determined. Comparison of the genomic with the e DNA sequences indicated the presence of seven introns, sequences of which are given in Fig. 2. There is little homology between those introns and their size is, by order of appearance: 54 bp, 51 bp, 69 bp, 60 bp, 60 bp, 51 bp, and 57 bp (Fig. 3). Exon size between introns is never larger than 150 bp, which explains the difficulties encountered in the S l-mapping experiments carried-out earlier on the chromosomal X L N (Boucher et al., 1988). Exon sequences obtained with the e D N A exactly match those deduced from the genomic XLN, indicating that no error was introduced during e D N A synthesis by the PCR method. Four of the 5'intron-exon junctions (GTRAGT) are similar to the consensus sequence for mammalian introns, two others ( GT R AGY) are close to the same sequence, and the last one ( G T A T G T ) is comparable to 5' end junctions in yeast introns according to Padgett et al. (1986), The 3'intron-exon junctions, YAG, are the same as the consensus sequence found in cukaryotic genes. The putative branching points of introns 2, 5 and 6 are similar to the consensus sequence for mammalian introns (CTRAY), Intron 3 shows the sequence T A C T T A A C which is close to the strictly conserved sequence found in yeast, except that it contains an additional T in the middle of the sequence. It was not possible to find a consensus sequence for the introns 1, 4 and 7. Translation through the introns could not proceed because of the presence of either in-frame stop codons (introns 1, 2, 3, 4, 6) or frame-shift sequences (introns 5, 7). The C residue in position 1646 represents the 3' end of the XLN mRNA, where the poly(A) tail was added.

kb

b

c kb

2.0-1.6-

I

m

0.5-

Fig. !. Agarose gel (0.7%) electrophoresis of eDNA fragments amplified by PCR. Lanes: a, l-kb DNA ladder; b, amplified eDNA of the specific XLN sequence; e, amplified XLN sequence for expression in E. coil Methodv. eDNA synthesiswas carried out with enriched mRNA fractions prepared by sucrose gradient centrifugation as previouslydescribed (Morosoli, 1985). The fraction containing XLN mRNA (20 #g) was used to produce ds eDNA according to the procedures of Gubler and Hoffman (1983). The specific XLN sequence was amplified by the PCR method (Frohman, 1990) using oligo probes (50 pmol) and total eDNA (50 ng). At the 5' end, we used the exact nt sequence 5'-ACCGAGGGTAACTTTGATTTCCACCGGTAC, which is located at nt positions 241.271 in the genomic nt sequence,just in front of the first intron. At the 3' end, the poly(T)-tail-eontainingoligo 5'-GGGTCTAGAGCTCGAAGTTTT. TTTTTTTTTTTTT was used because the exact terminal 3' end of the mRNA was not known. In order to facilitate further cloning experiments, sequences for tile restriction sites Xbal, Xhol and Sstl were added. Reaction conditions for amplificationwere: denaturation 95 °C, I mix; primer annealing 52°C, 1 mix: primer extension 72°C, 3 mix; cycling, 30 cycles. The amplified DNA fragment was further analyzed by agarose gel electrophoresis and purified by electroelution. Construction of a complete structural XLN from the eDNA, for expression in E. coil, was also carried out by the PCR method, We used pIAF113 as template and at the 5' end the oligo 5'-GGATCCTCTAGAGGAGGAAAAAATTATGC. TCTCTCTTCAACCAcontaining the restriction site Xbal, a translational initiation site and 15 nt of the XLN gene from the ATG start codon. At the 3' end was the oligo 5'--GTATCTGCAGTTCTGCCATCTTCTT, constituted of 15 nt from the chromosomal XLN gene to which was added a sequence for the restriction site Pstl. Those 15 nt are from positions 1520-1534, located 40 nt downstream from the stop codon.

(b) Construction and cloning of the XLN gene The seven introns are located between the restriction sites Kpnl and HindIIl on the X L N structural gene. An X L N gene without introns was thus constructed by ligating two fragments isolated from the chromosomal XLN: SstlKpnl (1.2 kb) containing the promoter region of X L N and

Hindlll-PstI (0.75 kb) containing the 3' end of X L N to a KpnI-HindIII (0.8-kb) fragment, isolated from the e D N A prepared by the P e R method and containing a part of the structural XYL. The new construction was inserted into phagemid vector pTZ 18 digested by Sstl + PstI and cloned

147 IntEons

10

20

30

I

I

I

40

50

I

60 nt

I

I

1

GTgagcctggagtcataagtgacaaagcagtcagacttaccctacttgctccAG

2

GTaagtggtccaagaaagtaaagcgatacgacgactqatcatgggctctAG

3

GTgagtagcctcggccgcgaccgaatcgcttcgaata=ttaacgatggttttectttcttttcctccAG

4

GTaagccacgcgattcattctcgaaatcctcaacgttgatcagtcacctttttgctatAG

5

GTatgttcaagaatgcgaagagacgatcgtccggactgctqatccgaaccgcatcttcAG

6

GTaagtgcttcgcaatcacgaaagatcagaggctaacacgttcgagtgcAG

7

GTaagttgattgcggtctcctgcacattctcaggatgcttattgcaatactgcgtAG

C. a2bidus

Mammalian

GT

ASA 6 T 5 G G2T 1 C 2

A AC G T A G T ............ CT A G GT

T Saccharomyces cerevisiae

C4 AG T3

C AG T

C

G T A G T ............ TACTAAC-

C

AG

T

Fig. 2. Nucleotide sequence of seven introns in the XLN gene from C. albidus and putative consensus sequences for correct splicing. The capital letters represent the conserved nt at the 5' and 3' intron-exon junctions. The underlined sequence in introns 2, 5 and 6, resemble the consensus sequence for the branching point in mammalian introns and the bold-type sequence in intron 3 is almost the strictly conserved branching point sequence found in Saccharomyces cerevisiae (Padgett et al., 1986). The sequence is available from EMBL accession No. X12596.

I 8

b

I

aa

/

u

0

u

\ see

I

1oo~ Fig. 3. Location of the introns on the XLN gene from C. albidus. (a) Map of the 3.15-kb genomic clone (single line) containing the XLN gene (hatched box), showing restriction sites; (b) refined map (1646 bp) derived from the nt sequence. Numbers refer to nt position: 1 is start codon; 1480, stop codon; 1646, site of poly(A)-tailing of mRNA. The other numbers indicate the 5'- and 3'-junctions for the seven introns represented by boxes. E. coli JM 103 was used for the sequencing experiments (Messing, 1983).

148 in E. coli. However, the resulting clone (IAF113) did not synthesize XLN, and the X L N promoter was thought to be responsible for this. The structural XLN gene was therefore placed near the iacZ promoter present in phagemid pTZ 18. This construction was achieved by PCR, using at the 5' end an oligo containing the restriction site XbaI, a translational initiation sequence and an additional 15 nt of the X L N gene downstream from the ATG start codon. At the 3' end, the oligo contained 15 nt from the X L N chromosomal sequence located 40 nt downstream from the stop codon and a Pstl restriction site. After amplification, using pIAFII3 as template, a l.l-kb fragment was obtained. This fragment was digested with Xbal + Pstl, ligated to phagemid pTZI8 restricted with the same enzymes and cloned in E. co/i. The new clone (IAFll4) expressed its XLN gene under the control of the lacZ promoter.

(c) Expression of the XLN gene in Escherichia coli E. coil exponential-phase cells containing plasmid pIAF114 were tested for XLN production. While none was found in the culture supernatant, XLN activity (0.047 IU per mg of proteins) was measured after cell disruption. Supernatant and cell extract were then subjected to SDSPAGE, followed by Western transfer on nitrocellulose membrane. The presence of XLN polypeptides was verified by anti-XLN antibodies coupled to [125I]protein A

(Fig. 4A). As expected, no X L N was detected in the culture supematant, even after loading samples equivalent to 10 ml of culture. The size of XLN detected in the cell extract was 40 kDa, which corresponds to the size of the under-glycosylated XLN secreted by Cryptococcus albidus after Tu treatment (Morosoli et al., 1988). To compare its properties with those of the glycosylated enzyme, a certain amount of reXLN was isolated from the cytoplasmic fraction of clone IAF114 by immunoailinity column chromatography. Yield from 2 liters of a 6-h culture was 450/~g, which represented about 0.9% of the total proteins. The enzyme preparation was almost pure, as revealed by S D S PAGE followed by Coomassie blue staining (Fig. 4B). The enzyme kinetics was determined using xylan as substrate. The values for the Michaelis constant (Kin), which is the concentration of substrate at which halfthe maximum velocity (Vm,,x)is reached, were measured. The Km values were 6.6 mg/ml and 3.2 mg/ml for the nonglycosylated and glycosylated enzymes, respectively. In the same order, V,nax values were 2 #tool/rag and 180 #tool/rag. A most interesting observation was made when comparing hydrolysis patterns. The nonglycosylated enzyme degraded large xylan polymers but was not able to achieve their complete hydrolysis into small xylooligosaccharides as did the glycosylated enzyme (Fig. 5A). In addition, xylohexaose (contaminated by xyloheptaose) was only slightly degraded by

kDa

B a kDa 92 -,,0,,~,,~,

68"-

6a-I

A

a

b

¢

d

4a4340-



f

b

¢

43-~,

Q 26

"'= ",' ~4%:

26Fig. 4, PAGE of the nonglycosylated reXLN from clone IAF114. (Panel A) Analysis of culture supernatant and intraccllular protein content of clone IAF114. Total cellular extracts were prepared by breaking the cells from a 2 liter culture with a French press (Whittle et al., 1982). XLN activity was determined as previously described (Morosoli et al., 1986), and proteins were analysed by 0,1% SDS-9% PAGE (Laemmli, 1970), followed by Western blotting (Towbin et al., 1979) and immunodetection using anti-XLN antibodies and (r~'~I)protein A. Lanes: a, purified glycosylated XLN (100 ng); b, purified under-glycosylatcd XLN (100 ng); e, control, total cell extract from E. coli (100/~g), d, total cell extract from clone IAF114 (100 pg): e, control, supernatant of E. coli (100 pg): f, supernatant from clone IAFi 14 (100 sg), (Panel B) Purification of cytoplasmic reXLN. The enzyme Was purified by immunoaffinity using a Zetattinity TM column where 5 mg of immunopurified anti-XLN antibodies had been immobilized according to the manufacturer's instructions (CUNO, Inc., Meriden, CT). The cytoplasmic fraction, at a concentration of I mg/ml of proteins in buffer B (0. l M Na,phosphatc/ 0.25 M NaCl pH "(,6) was recirculated through the column for 1 h at 4~C at a flow rate of I ml/min. The column was washed with 200 ml of the above buffer B, and eluted ~ith 0.2 M glycine. HCI, pH 2.3. The fractions containing the purified reXLN were immediately neutralized with 2 M rris.HCI, pH 8, pooled, dialyzed with l mM Tris.HCI, pH 7.5 and kept at -20°C after lyophilisation. The purified protein was analyzed on 0. 1% SDS-9% PAGE and stained with Coomassie brillant blue. Lanes: a, molecular size standards; b, [~lycosylatedXLN (5 pg); e, immunopurified nonglycosylated reXLN (5/~g).

149

A

xs

B

a

that disulfide bridges are not essential to enzymatic activity and that it is the glycosidic moiety of the XLN which plays an important role in enzyme specificity by allowing the proper enzyme conformation to permit a complete hydrolysis of xylan to xylose, xylobiose and xylotriose.

(d) Conclusions

j

b

x,

Xa

X~

y x "

×4 Xe ¢

1(4

(1) The finding of seven small introns in the X L N gene of C. albidus is rather unusual, since the number of introns in yeasts is usually two to five (Guilbert et ai., 1986). However, in fungi, up to eight introns have been found in the ligninase-encoding gene of Phanerochaete chrysosporium (Smith et al., 1988) and in the amylase-encoding gene of Aspergillus oo,zae (Gines et al., 1989). There is nevertheless no sequence homology between them and the introns of C. albidus. (2) Expression of the X L N gene of C. albidus by E. coil is limited and there is no secretion of the gene product in the culture medium, which is similar to the results reported for r ther cloned xylanase genes. (3) Substrate specificity of the nonglycosylated XLN is changed as compared to the glycosylated XLN, giving hydrolysis of only large xylan polymers, and it seems that the glycosidic moiety of the XLN, involved in its conformation is responsible for the complete hydrolysis of xylan.

,,J Fig, 5, Hydrolysispatterns of xylan and hexaxylosideobtained with glycosylated XLN and nonglycosylatedreXLN. The samples were analyzed by high-performance liquid chromatography (Kluepfd et al., 1990). (A) Enzymes (0.1 IU) were incubated for 4 h, at 25°C, in a l-ml solution of 1% oat spelts xylan. Curve a, pure xylan as control; curve b, with glycosylated XLN, xylan is completely degraded to small xylooligosaccharides (xylotetraose X4, xyiotriose X3, xylobiose X2); curve e, with nonglyeosylated reXLN, xylan is partially degraded to large xylan polymers. (B) Enzymes(0.1 IU) were incubated for 4 h, at 25°C, in a 1-ml solution of 1% hexaxyloside. Curve a, hexaxylosideas control (slightlycontaminated with hexaheptaose XT);curve b, glycosylatedXLN degrades hexaxyloside X6 almost completely into smaller xylooligosaccharides:x 5 is xylopentaose; curve c, nonglycosylatedreXLN shows almost no degradation of X6. the nonglycosylated enzyme (Fig. 5B), which confirmed the previous observation. This meant a change in enzyme specificity, which could be due, among other reasons, to the absence of glycosylation and to conformation changes in the recombinant enzyme. The XLN contains five Cys residues, which can form disulfide bonds contributing to the protein structure and it is known that E. coil does not make proper disulfide bridges when synthesizing heterologous proteins, which could account for the above results. But after incubation of the glycosylated, the under-glycosylated (resulting from Tu treatment) and the nonglycosylated enzymes in presence of 100 mM dithiothreitol for 24 h at 25 ° C, there was no loss of activity. Then, it can be assumed

ACKNOWLEDGEMENTS The authors wish to thank the Natural Sciences and Engineering Research Council of Canada for its financial support.

REFERENCES Boucher, F., Morosoli, R. and Durand, S.: Complete nucleotide sequence of the xylanase gene from the yeast Cryptococcus albidus. Nucleic Acids Res. 16 (1988) 9874. Frohman, M.A.: RACE: Rapid amplification of cDNA ends. In: Innis, M.A., Gelfand, D.H., Sninsky, J.J. and White, TJ. (Eds.), PCR Protocols, a Guide to Methods and Applications. Academic Press, San Diego, CA, 1990, pp. 28-38. Gilbert, M.F., Marchionni, M. and McKnight, G.: On the antiquity of introns. Cell 46 (1986) 151-154. Gines, M.J., Dove, MJ. and Seligy, V.L.: Aspergiilus or)'zae has two nearly identical Taka-amylase genes, each containing eight introns. Gone 79 (1989) 107-117. Gubler, U. and Hoffman, B.J.: A simple and very efficient method for generating cDNA libraries. Gone 25 (1983) 263-269. Klucpfcl, D., Vats-Metha, S., Aumont, F., Shareck, F. and Morosoli, R.: Purification and characterization of a new xylanase (xylanase B) produced by Streptomyces iividans 66. Biochem. J. 267 (1990) 4550. Laemmli, UK: Cleavageof structural proteins during the assembly of the head of the bacteriophage T4. Nature 227 (1970) 680-685.

150 Messing, J.: New M 13 vectors for cloning. Methods Enzymol. 101 (1983) 20-78. Morosoli, R.: Molecular expression of xylanase gene in Cryptococcus albidus. Biochim. Biophys. Acta 826 (1985) 202-207. Morosoli, R. and Durand, S.: Molecular cloning of mRNA sequences encoding xylanase from Cryptococcus aibidus. FEMS Microbiol. Letters 51 (1988) 217-224. Morosoli, R., Roy, C. and Yaguchi, M.: Isolation and partial primary sequence of a xylanase from the yeast Cryptococcus albidus. Biochim. Biophys. Acta 870 (1986) 473-478. Morosoli, R., Lecher, P. and Durand, S.: Effect of tunicamycin on xylanase secretion in the yeast Cryptococcus albidus. Arch. Biochem. Biophys. 265 (1988) 183-189. Morosoli, R., Shareck, F., Moreau, A. and Kluepfel, D.: Expression of xylanase genes in $treptomyces lividans. 6th International Symposium on Genetics of Industrial Microorganisms, Strasbourg, France, 1990, pp. 935-946.

Padgett, R.A., Grabowski, P.J., Konarska, M.M., Seiler, S. and Sharp, P.A.: Splicing of messenger RNA precursors. Annu. Rev. Biochem. 55 (1986) 1119-1150. Smith, T.L., Schlach, H., GaskeU, J., Covert, S. and Cullen, D.: Nucleotide sequence of a ligninase gene from Phanerochaete cho,sosporilon. Nucleic Acids Res. 16 (1988) 1219. Tanner, W. and Lehle, L.: Protein glycosylation in yeast. Biochim. Biophys. Acta 906 (1987) 81-89. Towbin, H., Staehelin, T. and Gordon, J.: Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Prec. Natl. Acad. Sci. USA 77 (1979)43504354. Whittle, D.J., Kilburn, D.G., Warren R.AJ. and Miller Jr., R.C.: Molecular cloning of a Celhdomonas~ni cellulase gene in Escherichia coli, Gene 17 (1982) 139-154.

Cloning and expression in Escherichia coli of a xylanase-encoding gene from the yeast Cryptococcus albidus.

In the yeast, Cryptococcus albidus, a comparison between the sequence of the xylanase (XLN)-encoding chromosomal gene (XLN) and the cDNA sequence reve...
601KB Sizes 0 Downloads 0 Views