Maps

Functional analyses of the chitin-binding domains and the catalytic domain of Brassica juncea chitinase BjCHI1

Description
We previously isolated a Brassica juncea cDNA encoding BjCHI1, a novel chitinase with two chitin-binding domains. Synthesis of its mRNA is induced by wounding, methyl jasmonate treatment, Aspergillus niger infection and caterpillar Pieris rapae
Categories
Published
of 14
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Functional analyses of the chitin-binding domains and the catalyticdomain of   Brassica juncea  chitinase BjCHI1 Ce Mun Tang 1 , Mee-Len Chye 1, *, Sathishkumar Ramalingam 1 , Shi-Wen Ouyang 2 ,Kai-Jun Zhao 2 , Wimal Ubhayasekera 3 and Sherry L. Mowbray 3 1 Department of Botany, The University of Hong Kong, Pokfulam Road, Hong Kong, China (*author forcorrespondence; e-mail mlchye@hkucc.hku.hk);  2 Institute of Crop Breeding and Cultivation, ChineseAcademy of Agricultural Sciences, Beijing 100081, China;  3 Department of Molecular Biology, Box 590,Biomedical Center, Swedish University of Agricultural Sciences, SE-751 24, Uppsala, Sweden Received 20 April 2004; accepted in revised form 16 September 2004 Key words:  agglutination, glycoside hydrolases, homology modeling, lectin, pathogenesis-related protein,site-directed mutagenesis Abstract We previously isolated a  Brassica juncea  cDNA encoding BjCHI1, a novel chitinase with two chitin-bindingdomains. Synthesis of its mRNA is induced by wounding, methyl jasmonate treatment,  Aspergillus niger infection and caterpillar ( Pieris rapae ) feeding, suggesting that the protein has a role in defense. In that itpossesses two chitin-binding domains, BjCHI1 resembles the precursor of   Urtica dioica  agglutinin butunlike that protein, BjCHI1 retains its chitinase catalytic domain after post-translational processing. Toexplore the properties of multi-domain BjCHI1, we have expressed recombinant BjCHI1 and two deriv-atives, which lack one (BjCHI2) or both (BjCHI3) chitin-binding domains, as secreted proteins in  Pichia pastoris . Recombinant BjCHI1 and BjCHI2 showed apparent molecular masses on SDS-PAGE larger thancalculated, and could be deglycosylated using  a -mannosidase. Recombinant BjCHI3, without the proline/threonine-rich linker region containing predicted  O -glycosylation sites, did not appear to be processed by  a -mannosidase. BjCHI1’s ability to agglutinate rabbit erythrocytes is unique among known chitinases. Bothchitin-binding domains are essential for agglutination; this property is absent in recombinant BjCHI2 andBjCHI3. To identify potential catalytic residues, we generated site-directed mutations in recombinantBjCHI3. Mutation E212A showed the largest effect, exhibiting 0% of wild-type specific activity. H211Nand R361A resulted in considerable (>91%) activity loss, implying these charged residues are alsoimportant in catalysis. E234A showed 36% retention of activity and substitution Y269D, 50%. The leastaffected mutants were E349A and D360A, with 73% and 68% retention, respectively. Like Y269, E349 andD360 are possibly involved in substrate binding rather than catalysis. Introduction Chitinases (EC 3.2.1.14) are pathogenesis-relatedproteins that catalyze the random cleavage of internal  b -1,4 glycosidic linkages in chitin, amajor component of fungal cell walls and insectexoskeletons (Boller, 1985). The majority of family19 chitinases include a chitin-binding domain(homologous with hevein), which is joined to thecatalytic domain by a linker region (Collinge  et al. ,1993). Since plant lectins consist of one or morechitin-binding domains, chitinases that contain achitin-binding domain may be considered as aclass of lectins that are structurally linked to anunrelated domain (Van Damme  et al. , 2004).Lectins in general exhibit a wide variety of struc-tures. Hevein, a 4.7-kDa lectin from rubber latex(Van Parijs  et al. , 1991), for example, possesses Plant Molecular Biology  56:  285–298, 2004.   2004  Kluwer Academic Publishers. Printed in the Netherlands. 285  one chitin-binding domain, while  Urtica dioica agglutinin (UDA; Beintema and Peumans, 1992)has two and wheat germ agglutinin, four (Wright et al. , 1991). The mature forms of hevein(Soedjanaatmadja  et al. , 1995) and UDA (Lernerand Raikhel, 1992) contain chitin-binding do-mains alone and show anti-fungal activity  in vitro (Broekaert  et al. , 1989; Van Parijs  et al. , 1991).Chitinases exhibit anti-fungal activity (Schlum-baum  et al. , 1986; Roberts and Selitrennikoff,1988; Melchers  et al. , 1994), lysing fungal tips andinhibiting growth (Mauch  et al. , 1988). Thus,transgenic plants expressing chitinases show en-hanced resistance to fungal pathogens (Broglie et al. , 1991; Lin  et al. , 1995; Grison  et al. , 1996).Furthermore, chitinases have been shown to actsynergistically with UDA (Broekaert  et al. , 1989)or  b -1,3-glucanase (Mauch  et al. , 1988; Zhu  et al. ,1994) in inhibiting fungal growth.The known chitinases can be classified intotwo major groups based on features of their cat-alytic domains, i.e. families 18 and 19 of theglycoside (earlier, glycosyl) hydrolases (Henrissatand Bairoch, 1993). The two families differ fromeach other in terms of their amino acid sequences,three-dimensional structures and hydrolyticmechanisms (Fukamizo, 2000). The enzymes havefurther been divided into a number of classesbased on particular features. Family 18 enzymesof class III (and in some instances, VI) representmostly bacterial, fungal and animal chitinases.Family 19 chitinases include almost exclusivelyplant enzymes, consisting of classes I, II, IV andV (Iseli  et al. , 1996), as well as some members of class VI (Neuhaus  et al. , 1996). With few excep-tions (e.g. O’Riordain  et al. , 2002), chitinasesfrom classes I, IV and V are synthesized with anN-terminal signal sequence preceding the chitin-binding domain (Lerner and Raikhel, 1992;Collinge  et al. , 1993). According to plant chitin-ase classification (Collinge  et al. , 1993), the ma-ture class I chitinases have an N-terminalcysteine-rich segment, which functions as a chitin-binding domain. Class II chitinases are verysimilar to the class I enzymes, but lack the chitin-binding domain. Class IV chitinases have a chitin-binding domain, but their catalytic domains areslightly smaller than those of the class I or IIenzymes. The UDA precursor is the only memberof class V ( Chia5 , according to the newer plantchitinase classification nomenclature of Neuhaus et al. , 1996). It is characterized by the presenceof two chitin-binding domains and is post-transla-tionally processed to yield UDA, which has beenseparated from the chitinase domain. Some classVI chitinases contain one chitin-binding domain( Chic1 ; Neuhaus  et al. , 1996) while others possesshalf a domain ( Chia6 ; Neuhaus  et al. , 1996). The1.3-kb cDNA we earlier isolated from  Brassica juncea  (Indian mustard) encodes BjCHI1, achitinase that shows strongest (50–60%) amino-acid sequence identity to the catalytic domains of family 19 (class I) enzymes (Zhao and Chye,1999). However, BjCHI1 is structurally distinctfrom essentially all previously reported plantchitinases by virtue of the presence of a secondchitin-binding domain. Hence, we have previouslyproposed that BjCHI1 belongs to a new class( Chia 7) since it has two chitin-binding domainsand lacks strong similarity to the UDA precursorin the chitinase domain (Zhao and Chye, 1999).More recently, Ueda  et al  . (2003) have isolated afamily 19 chitinase with two chitin-bindingdomains from the bacterium  Aeromonas . Thislarger chitinase (calculated molecular mass70.39 kDa) shares 42.2% identity to BjCHI1 inthe catalytic domain and 48.7% identity in thechitin-binding domains.Few structure-activity studies (Hart  et al. ,1995; Song and Suh, 1996) have been reported forthe family 19 plant chitinases despite much re-search on their anti-fungal properties. In order toinvestigate the properties of the unusual chitinaseBjCHI1, we here compared the agglutinationactivity of   Pichia -expressed recombinant proteinwith that of deletion derivatives lacking one orboth chitin-binding domains. To identify residuesimportant in chitin hydrolysis within its active-sitecleft, we generated mutants with single amino acidsubstitutions at conserved residues appropriate foracid–base catalysis. Materials and methods Construction of yeast expression plasmids Polymerase chain reaction (PCR)-generated DNAfragments encoding BjCHI1, BjCHI2 and BjCHI3were cloned into the  P. pastoris  expression vectorpPIC9K (Invitrogen, San Diego, CA, USA), in-frame to the pPIC9K N-terminal secretory signal286  peptide, to produce recombinant fusion proteinsthat would be secreted into the growth media.The yeast plasmid pBj106 for recombinantBjCHI1 expression was constructed by cloning a1.1-kb DNA fragment encoding amino acids18–393 of BjCHI1, excluding its N-terminal sig-nal peptide and C-terminal vacuolar targetingsequence (Zhao and Chye, 1999, GenBankaccession no. AAF02299) in vector pPIC9K (Fig-ure 1A). This 1.1-kb fragment was PCR-amplifiedusing primer pair C1/C2, and template plasmidpBj17 containing the full-length  BjCHI1  cDNA(Fung  et al. , 2002). The forward primer C1 (5 0 -CTGAATTC TCC  TCCGGTGAGCAATGCG-3 0 )has an  Eco RI site (underlined) adjacent to the S18codon (bold italics). Reverse primer C2 (5 0 -GCGACTGCGGCCGCGTTACTACCTTCATTAAACG-3 0 ) contains a  Not I site (underlined).Plasmid pBj106 produces a recombinant pre-pro-tein (473 amino acids), consisting of vector-derivedN-terminal secretory signal and C-terminal resi-dues ‘‘AAAN’’, fused to BjCHI1-derived S18-N393 between (Figure 1B). Upon cleavage of thesecretory peptide of 89 residues, a 384-amino acidrecombinant BjCHI1 protein is secreted into themedia. This protein includes two chitin-bindingdomains that are almost identical (Zhao and Chye,1999).BjCHI2 is a deletion derivative of BjCHI1 inwhich the C-terminal end of the first chitin-bindingdomain, the spacer and the N-terminal part of thesecond chitin-binding domain in BjCHI1 havebeen removed exactly in-phase; as a result there isonly a single chitin-binding domain in this con-struct. Plasmid pBj107 was constructed forexpression of recombinant BjCHI2 (Figure 1). A0.95-kb DNA fragment was PCR-amplified usingprimer pair C1/C2 and plasmid pBj28 (Fung  et al. ,2002) as template. Plasmid pBj28 was derived frompBj17 by deletion of a  Hin dIII fragment encoding Figure 1 . Domain structure and sequence of secreted  Pichia -expressed recombinant BjCHI1, BjCHI2 and BjCHI3 expressed frompPIC9K derivatives. (A) Structure of proteins in comparison to the intact BjCHI1 pre-protein (bottom). The recombinant proteins lackthe N-terminal signal peptide (SP, amino acids 1–20) and the C-terminal vacuolar targeting peptide (V, amino acids 394–400) of thepre-protein; arrows indicate cleavage sites. Instead each  Pichia -expressed protein is fused to the pPIC9K secretory signal peptide (notshown). Recombinant BjCHI1 contains S18-N393 of native BjCHI1 (Zhao and Chye, 1999) including the two chitin-binding domains(CBD, amino acids 22–61 and 73–112) and the spacer (S) between, the linker region (L, amino acids 113-145) and the catalytic domain.Recombinant BjCHI2, with residues A51-E101 removed, effectively lacks one chitin-binding domain. Recombinant BjCHI3 (G145– N393) consists only of the catalytic domain. (B) Amino acid sequence of secreted recombinant BjCHI1 in which S18–N393 of nativeBjCHI1 is fused to pPIC9K-derived residues  YVEF   and  AAAN   at the N- and C-termini, respectively. Amino acid residues arenumbered as according to the intact BjCHI1 pre-protein (Zhao and Chye, 1999). Amino acids 51–101 (lower-case) are deleted inrecombinant BjCHI2. The underlined residues (S18–S144) are deleted in BjCHI3. T and S residues shown in boldface indicate potential O -glycosylation sites, which are predominantly within the linker region (T113–S145). Amino acids selected for  in vitro  mutagenesis(H211N, E212A, E234A, Y269D, E349A, D360A and R361A) are indicated with gray boxes. Amino acids 231–242 (lower-case italics)indicate the synthetic peptide used in raising polyclonal antibodies. 287  amino acids A51–E101 (Fung  et al. , 2002) thatconstitutes a chitin-binding domain and spacer(Figure 1B). The BjCHI2 fusion pre-protein (422amino acids) expressed from plasmid pBj107consists of vector-derived residues at its N- andC-termini with BjCHI1-derived residues (S18–E50and A102–N393) between (Figure 1B). Followingcleavage of the N-terminal secretory peptide, thesecreted BjCHI2 recombinant protein consists of 333 amino acids.Plasmid pBj108 was constructed for expressionof recombinant BjCHI3, a deletion derivativelacking both chitin-binding domains (Figure 1). A0.75-kb DNA fragment was PCR-amplified usingprimer pair C3/C2 and plasmid pBj17 (Fung  et al. ,2002) as template. Forward primer C3(5 0 CTGAATTC GGG  GATCTTTCTGGCATC3 0 )contains an  Eco RI site (underlined) adjacent to theG145 codon (bold italics). The BjCHI3 fusion pre-protein of 346 amino acids consists of vector-de-rived residues at its N- and C-termini and aminoacids G145–N393 from BjCHI1 between (Fig-ure 1B). The secreted recombinant BjCHI3 proteinincludes 257 amino acids.PCR amplification was performed using 100 pgof plasmid DNA in a 100- l l reaction mixture,containing 2.5 U  Pfu  DNA polymerase (LifeTechnologies), 100  l M of each dNTP and 100 pMof primers. Reactions were incubated at 95   C for4 min, followed by 29 cycles of denaturation(95   C for 15 s), annealing (55   C for 15 s) andextension (72   C for 90 s), with a final extension at72   C for 10 min. Following PCR, the DNA wascleaved with  Eco RI and  Not I, before cloning intocorresponding sites in plasmid pPIC9K. Escherichia coli   Top10F’ (Invitrogen) cells weretransformed with the pPIC9K derivatives. PCR-derived DNA in plasmids pBj106, pBj107 andpBj108 were analyzed by DNA sequencing beforetransformation of   P. pastoris . Transformation of   P. pastorisFollowing linearization with  Sac I, each plasmidwas used in transformation of   P. pastoris  strainKM71 by electroporation (Scorer  et al.,  1994).His + transformants were selected on RDB agar(1 M sorbitol, 2% dextrose, 1.34% yeast nitrogenbase, 4  ·  10 ) 5 % biotin) and were subsequentlyscreened for G418 resistance at a final concen-tration of 1.0 mg ml ) 1 G418, according to theinstructions specified in the Multi-copy  Pichia Expression Kit (Invitrogen). PCR and Southernblot analyses were used to confirm the targetDNA had recombined into the chromosomesof recombinant yeast before use in proteinexpression. Production and purification of secreted chitinases expressed in  P. pastoris Pichia pastoris  strain KM71 was used for expres-sion of recombinant BjCHI1, BjCHI2 andBjCHI3. Each yeast strain was maintained onYPD agar (1% (w/v) yeast extract, 2% (w/v)peptone, 2% dextrose, 2% agar). Minimal med-ium, consisting of 1.34% (w/v) yeast nitrogenbase, 4  ·  10 ) 5 % (w/v) biotin and 2% (w/v)dextrose, was used in selection.To initiate protein production in  Pichia ,10 ml of BMGY medium (1% (w/v) yeastextract, 2% (w/v) peptone, 100 mM potassiumphosphate, pH 6.0, 1.34% (w/v) yeast nitrogenbase, 4  ·  10 ) 5 % (w/v) biotin, 1% (v/v) glycerol)was inoculated with a single colony and incu-bated overnight at 30   C. Subsequently, theovernight culture was added to 800 ml of BMGY medium and further incubated until theoptical density (OD) at 600 nm reached at least2.0. The cells were harvested by centrifugation at5000  ·  g  at room temperature for 5 min. Toinduce recombinant protein expression, the cellpellet was resuspended in 80 ml of BMMYmedium (1% (w/v) yeast extract, 2% (w/v) pep-tone, 100 mM potassium phosphate, pH 6.0,1.34% (w/v) yeast nitrogen base, 4  ·  10 ) 5 % (w/v) biotin, 0.5% (v/v) methanol) and grown at30   C for a further two days, during whichmethanol was added to a final concentration of 0.5% (v/v) every 24 h. Cells were then harvestedby centrifugation (5000  ·  g  at 4   C for 15 min).The  Pichia -expressed protein in the supernatantwas precipitated overnight using 65% (w/v)ammonium sulfate. The precipitated protein wasresuspended in 10 mM sodium phosphate buffer,pH 7.2, then dialyzed against the same buffer.The dialyzed sample was centrifuged at 5000  ·  g at 4   C for 15 min, filtered through a membranefilter (0.2  l M, Nalgene) and concentrated usinga Centricon concentrator (10 kDa MW cut-off;Millipore). The protein was then loaded on a gel288  filtration Superdex HiLoad 16/60 column in anFPLC system (Amersham Pharmacia Biotech-nology, Uppsala, Sweden). The column was pre-equilibrated with 10 mM sodium phosphatebuffer, pH 7.2. Fractions of 2 ml each, werecollected and analyzed by (SDS-PAGE) electro-phoresis according to the method of Laemmli(1970). The gel was stained with CoomassieBrilliant Blue to check purity. The purest frac-tions containing the desired band were pooledand stored at  ) 80   C. Extraction of total plant protein Total plant protein was extracted (Kush  et al. ,1990) from leaves of wild-type tobacco, trans-genic tobacco transformed with vector pBI121(Clontech) and transgenic tobacco expressingBjCHI1 from plasmid pBj17. Plasmid pBj17,containing a 1.3-kb  Sma I fragment of   BjCHI1 cDNA cloned in the plant transformation vectorpBI121, was used in  Agrobacterium -mediatedtransformation of tobacco (Fung  et al. , 2002).BjCHI1 is expressed from the Cauliflower Mo-saic Virus promoter 35S (CaMV 35S) promoterin transgenic tobacco (Fung  et al. , 2002). Determination of protein concentration Protein concentration was determined using themethod of Bradford (1976). Routinely, Pichia -expressed crude and FPLC-purified re-combinant proteins BjCHI1, BjCHI2 and BjCHI3,as well as mutant derivatives of BjCHI3, wereloaded on SDS-PAGE to check protein qualityafter determination of protein concentrationbefore further use. N-terminal sequencing of FPLC-purified  Pichia -expressed secreted recombinant BjCHI1 FPLC-purified recombinant BjCHI1 was electro-phoresed on a 10% SDS-polyacrylamide gel andtransferred to PVDF membrane (Invitrogen)according to the manufacturer. The membranewas stained with Coomassie Blue and destained ina solution of 50% (v/v) methanol and 10% (v/v)acetic acid. The band of apparent molecular mass53-kDa was excised and analyzed by Edman deg-radation using a HP Protein Sequencer (ModelG1000A) according to the manufacturer’sinstructions. Deglycosylation of recombinant chitinasesBjCHI1, BjCHI2 and BjCHI3 Deglycosylation of recombinant chitinases wasperformed using jack bean  a -mannosidase (Sig-ma). FPLC-purified recombinant chitinaseBjCHI1, BjCHI2 or BjCHI3 (3  l g) was incubatedfor 6 days at 30   C with 1 unit of   a -mannosidase ina 30- l l solution of 20 mM sodium acetate, pH 4.5,and 1 mM zinc sulfate (Heimo  et al. , 1997). Ineach control sample, the chitinase without  a -mannosidase was similarly treated. After incuba-tion, the samples were resolved by SDS-PAGEusing a 10% polyacrylamide gel followed byWestern blot analysis using anti-peptide antibodiesagainst BjCHI1. A sample consisting of   a -man-nosidase alone was also tested to confirm that itdid not cross-react with antibodies againstBjCHI1. Western blot analysis Proteins on SDS-PAGE gels were electrophoreti-cally transferred onto nitrocellulose membranes(Hybond ECL) for Western blot analysis followingSambrook  et al  . (1989). Polyclonal antibodiesraised in rabbits against a synthetic peptide(YKEEIDKSDPHC) corresponding to aminoacids 231–242 of BjCHI1 (Zhao and Chye, 1999)were used as the primary antibody. Detection of cross-reacting bands was performed using theAmplified Alkaline Phosphatase Goat Anti-rabbitImmun-Blot Assay Kit (Bio-Rad). Agglutination assays Agglutination assays were carried out according tothe procedure of Does  et al.  (1999) using FPLC-purified samples of   Pichia -expressed recombinantBjCHI1, BjCHI2 and BjCHI3 in a microtiter tray.Protein concentrations tested were at 0 (control),0.12, 0.25, 0.5, 2, 4, 8, 16 and 24  l g of FPLC-purified protein per 60- l l final volume. In eachprotein-containing well, trypsin-treated rabbiterythrocytes (30  l l) were added and 10 mM289
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks