Protein Engineering, Vol. 13, No. 6, 385-389,
June 2000
© 2000 Oxford University Press
Communications |
Chemically prepared hevein domains: effect of C-terminal truncation and the mutagenesis of aromatic residues on the affinity for chitin
Biomolecules Department, National Institute of Bioscience and Human Technology, 1–1 Higashi, Tsukuba, Ibaraki 305-8566, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResults and discussionReferences |
|---|
Chemically prepared hevein domains (HDs), N-terminal domain of an antifungal protein from Nicotiana tabacum (CBP20-N) and an antimicrobial peptide from Amaranthus caudatus (Ac-AMP2), were examined for their affinity for chitin, a ß-1,4-linked polymer of N-acetylglucosamine. An intact binding domain, CBP20-N, showed a higher affinity than a C-terminal truncated domain, Ac-AMP2. The formation of a pyroglutamate residue from N-terminal Gln of CBP20-N increased the affinity. The single replacement of any aromatic residue of Ac-AMP2 with Ala resulted in a significant reduction in affinity, suggesting the importance of the complete set of three aromatic residues in the ligand binding site. The mutations of Phe18 of Ac-AMP2 to the residues with larger aromatic rings, i.e. Trp, ß-(1-naphthyl)alanine or ß-(2-naphthyl)alanine, enhanced the affinity, whereas the mutation of Tyr20 to Trp reduced the affinity. The affinity of an HD for chitin might be improved by adjusting the size and substituent group of stacking aromatic rings.
Keywords: Chemical synthesis/chitin-binding activity/hevein domain/non-natural amino acid residue/site-specific mutagenesis
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Hevein domain (HD) is a structural motif of 30–43 amino acids widely found in many plant proteins such as N-acetylglucosamine residue (GlcNAc) specific lectins, small antimicrobial proteins, wound-induced proteins and class I chitinases (Wright et al., 1991
; Raikhel et al., 1993; Beintema, 1994). These proteins have an inhibitory effect on fungi or show toxicity to weevils, and are therefore assumed to play an important role in plant defense (Chrispeels and Raikhel, 1991; Peumans and Van Damme, 1995). As a characteristic of the structure, HDs commonly contain one Ser and three aromatic amino acid residues in addition to eight or six disulfide-bridged Cys at conserved position (Figure 1a). The three-dimensional structures of several HDs, namely the crystal structures of wheat germ agglutinin (WGA) (Wright, 1987, 1989; Harata et al., 1995), hevein from rubber latex (Rodriguez-Romeo et al., 1991) and a stinging nettle agglutinin (UDA) (Harata and Muraki, 2000), have been determined. The solution structures of hevein (Andersen et al., 1993) and an antimicrobial protein from Amaranthus caudatus (Ac-AMP2) (Martins et al., 1996) have also been elucidated. The interaction mode between HD and GlcNAc oligomers has been characterized with WGA (Wright, 1980, 1984; Bains et al., 1992), UDA (Hom et al., 1995; Lee et al., 1998; Harata and Muraki, 2000), hevein (Ascensio et al., 1995); Garcia-Hernández et al., 1997) and Ac-AMP2 (Verheyden et al., 1995) using X-ray, NMR and microcalorimetry methods. The role of aromatic amino acids in carbohydrate binding of HD has been investigated by using the laser photo-induced dynamic nuclear polarization method of NMR (Siebert et al., 1997). However, to date, there have been few investgations of the carbohydrate recognition function of HD by site-specific mutagenesis except for a study on two WGA mutants (Nagahora et al., 1995).
|
Recently, we reported the chemical synthesis of the N-terminal domain of a wound-induced gene product from potato (WIN2-N) (Stanford et al., 1989
) by Fmoc peptide synthesis and subsequent oxidative refolding (Muraki et al., 1998). The synthetic WIN2-N showed a binding behavior very similar to that of UDA toward chitin, a water-insoluble ß-1,4-linked polymer of GlcNAc. In the present study, we applied this methodology to two other HDs, N-terminal domain of an antifungal protein from Nicotiana tabacum, CBP20-N [an intact domain (43 amino acids)] (Ponstein et al., 1994) and Ac-AMP2 [a C-terminal truncated domain (30 amino acids) (Broekaert et al., 1992). Site-specific mutagenesis of the three aromatic residues (Phe18, Tyr20 and Tyr27) of Ac-AMP2 were performed to probe the role of these residues in the expression of affinity for chitin. Chemical synthesis is a powerful method to introduce a non-natural amino acid residue into the designated position of proteins (Muir and Kent, 1993; Morii et al., 1999). In this study, 12 mutants of Ac-AMP2 including five mutants with a non-natural amino acid residue at position 18 were prepared. Here, as a first step to elucidating the structure and function relationships of HD by protein engineering, we examined their affinity for chitin using a chromatographic method and interpreted the results by referring to the three-dimensional structure of the UDA–(GlcNAc)3 complex (Figure 1b
). |
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Preparation of the sample
Solid-phase peptide synthesis, cleavage from resin, deprotection, separation by reversed-phase HPLC (RP-HPLC) and oxidative refolding of the reduced peptide were performed as described previously (Muraki et al., 1998). The samples were prepared as peptide amides. Molecular weights (MW) of the purified samples were measured with MALDI-TOF/MS Voyager (PerSeptive Biosystems). 
-Cyano-4-hydroxycinnamic acid was used as the matrix, UDA-isolectin VI(UDA VI) (Does et al., 1999), bovine insulin chain A (oxidized, purity 96%) and WIN2-N was obtained as described (Muraki et al., 1998). All chemical reagents were of the purest grade available. A polyacrylamide gradient gel (15–25%), SDS–Tris–Tricine buffer and peptide MW markers were purchased from Daiich Kagaku Yakuhin.
Assay of chitin binding activity
A chitin bead slurry was purchased from New England Biolabs (Beverly, MA). The affinity for chitin was evaluated basically according to the methods of Oita et al. (1996) and Nielsen et al. (1997) as described (Muraki et al., 1998). A 50 µg amount of sample as determined with a BCA protein assay kit (Pierce) was used for each assay. The amount of unbound sample after each elution step was determined from the peak area in RP-HPLC analysis. The amount of sample bound to chitin was calculated by subtracting the amount of unbound sample from the total amount of sample. Three replicate assays were performed with each sample at room temperature (25 ± 2°C).
|
Results and discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Chemical synthesis of CBP20-N and Ac-AMP2
The amino acid sequences of UDA VI (hinge region excluded), WIN2-N, CPB20-N and Ac-AMP2 were aligned in Fig. 1a. The RP-HPLC profiles of CBP20-N after oxidative refolding (Fig. 2a) resembled that of WIN2-N (Muraki et al., 1998) as expected from the striking similarity (40/43 identical) in amino acid sequence. Specifically, a pair of peaks (peaks 1 and 2) of the refolded product were observed after the oxidation of reduced peptide. Judging from the MW of peak 1 (4439) and peak 2 (4421), peaks 1 and 2 were identified as CBP20-N with an N-terminal glutamine residue ([Gln1]-CBP20-N) (calculated MW: 4443) and CBP20-N with an N-terminal pyroglutamate residue ([Pgl1]-CBP20-N) (calculated MW: 4426), respectively. The partial conversion of Gln1 into Pgl1 catalyzed by weak acids including acetic acid (AcOH) and the resulting similar separation of peaks in the RP-HPLC profile have been reported (Dimarchi et al., 1982, Or
owska et al., 1987). Since we used pure N-Fmoc-N
-tritylglutamine in the last coupling step, the formation of Pgl1 should be ascribed to the deprotection and purification steps using 10–50% AcOH.
|
On the other hand, the RP-HPLC profile of Ac-AMP2 possessing Val as its N-terminal residue showed a single plausible peak of the refolded product (Figure 2b
). The RP-HPLC profiles of all mutants in this study were essentially the same as that of wild-type Ac-AMP2. Also, the major peak after oxidation always eluted at a lower retention time than that before oxidation (Figures 2a and b). The non-reduced and disulfide bond reduced Ac-AMP2 samples including wild- type and Ala mutants at positions 18, 20 and 27 (Figure 2c) and non-natural admino acid residue mutants at position 18 (Figure 2d) were subjected to an SDS–PAGE analysis. In all these instances, the non-reduced sample was more or less retarded compared with the corresponding reduced sample as in the case of another C-terminal truncated HD, IWF4 (Nielsen et al., 1997), indicating the formation of disulfide bridges. Although further investigations may be necessary, the above results suggested that the mutations introduced in this study did not basically change the mode of the oxidative folding of Ac-AMP2. The measured MW (calculated MW in parentheses) of refolded products were native (Phe18/Tyr20/Tyr27) Ac-AMP2, 3185 (3184); Tyr18 mutant, 3201 (3200); 18-ß-(p-nitrophenyl)alanine (pNO2-Phe18) mutant, 3223 (3229); 18-ß-(pentafluorophenyl)alanine (F5-Phe18) mutant, 3270 (3274); Trp18 mutant, 3224 (3223); 18-ß-(1-naphthyl)alanine (1-Nal18) mutant, 3234 (3234); 18-ß-2-naphthyl)alanine (2-Nal18) mutant, 3237 (3234); Ala18 mutant, 3113 (3108); 18-ß-(cyclohexyl)alanine (Cha18) mutant, 3188 (3190); Phe20 mutant, 3172 (3168); Trp20 mutant, 3207 (3207); Ala20 mutant, 3097 (3092); and Ala27 mutant, 3095 (3092). All purified samples were homogeneous as judged by both MALDI-TOF/MS and RP-HPLC analysis.
Mutational effect on the affinity for chitin
Recently, we have determined the X-ray structure of UDA VI–(GlcNAc)3 complex at 1.9 Å resolution (Harata and Muraki, 2000). The structure of the ligand-binding region of the N-terminal domain of UDA VI (UDA VI-N) is shown in Figure 1b. As expected from the amino acid sequence homology, the overall three-dimensional structures of all HDs revealed so far are very similar to each other (Martins et al., 1996; Asensio et al., 1998). The spatial arrangement of the side chains of three aromatic residues which are involved in protein–carbohydrate interaction was also well conserved (Martins et al., 1996). Therefore, it is reaonable to discuss the mutational effect on the affinity for chitin in this study by referring to the X-ray structure of the UDA VI–(GlcNAc)3 complex.
The affinity of HD samples for chitin was examined in 10 mM Tris–HCl (pH 8.0) and 0.5 M AcOH (pH 
2.5). In Figure 3, the results of the assay are summarized. UDA VI (sample a) and insulin A-chain (sample b) were used as the positive and the negative control of the experiment, respectively. The behavior of intact HDs, WIN2-N (sample c) and [Pgl1]-CBP20-N (sample d) was similar to that of UDA VI under either pH conditions, whereas the affinity of a C-terminal truncated HD, Ac-AMP2 (sample f), was much less than that of UDA VI under acidic conditions. In the X-ray structure of the UDA VI-(GlcNAc)3 complex, several possible hydrogen bonds were observed between the main chain of the C-terminal region (residues 36–43) of the first domain and that of the core region including the ligand binding site (not shown). Therefore, the C-terminal region in intact HDs may contribute to strengthening the affinity by stabilizing the binding conformation of the domain. The stronger affinity of [Pgl1]-CBP20-N than Ac-AMP2 for chitin in this study was consistent with the larger association constant of hevein (Asension et al., 1995) than Ac-AMP2 (Verheyden et al., 1995) with (GlcNAc)3, which were determined by NMR experiments.
|
The [Gln1]-BP20-N (sample e) exhibited a weaker affinity than [Pgl1]-CBP20-N (sample d). The increase in affinity by the formation of Pgl1 suggested the involvement of Pgl1 of CBP20-N in the recognition of chitin. Supportingly, a hydrogen bond between the N atom of Pgl1 and the O6 atom of non-reducing end GlcNAc residue (NAG A) was deduced in the UDA VI–(GlcNAc)3 complex (Figure 1b
). The N-terminal Gln is often found in the primary sequences of HDs (Wright, 1991). The advantage in the recognition of chitin after the conversion into Pgl1 may partly account for this tendency.
The involvement of three conserved aromatic residues in the ligand binding of Ac-AMP2 has been suggested by NMR studies (Verheyden et al., 1995). Our previous study demonstrated that the replacement of Tyr73 in WGA which corresponds sterically to Tyr27 in Ac-AMP2 by Phe did not have much effect on the affinity for (GlcNAc)3 (Nagahora et al., 1995). In the present study, the site-specific mutant concerning Phe18, Tyr20 and Tyr27 of Ac-AMP2 were examined. At pH 8.0, all Ac-AMP2 samples possessing three aromatic residues at position 18, 20 and 27 (samples f, g, h, i, j, k, l, o and p) were completely bound to chitin, whereas a significant amount of unbound sample was observed with the mutants with replacement of any one aromatic residue of Ac-AMP2 by Ala (samples m, q and r). This suggested that the complete set of the three aromatic residues in the ligand binding site was important for the full expression of the affinity for chitin. As judged from the percentage of the remaining samples at pH 8.0, the extent of reduction of the affinity by the mutation to Ala decreased in the order Tyr20
Tyr27>Phe18.
The stacking interaction between the aromatic side-chain group of protein and the apolar face of the carbohydrate moiety frequently occurs in the ligand recognition of carbohydrate binding proteins (Vyas, 1991; Elgavish and Shaanan, 1997). The reduction in affinity for another water-insoluble ß-1,4-linked polysaccharide, crystalline cellulose, by the mutation of Tyr492 in the ligand binding site to Ala has been reported with the cellulose binding domain of Trichoderma reesei cellobiohydrase I (Reinikainen et al., 1992). In the UDA VI–(GlcNAc)3 complex, Trp21 and Trp23, which correspond to Phe18 and Tyr20 in Ac-AMP2, stacked with the apolar face of NAG C and NAG B in parallel, respectively (Figure 1b). Phe18 and Tyr20 in Ac-AMP2 were first replaced with two other natural aromatic amino acid residues. Tyr18 mutant (sample g) and more significantly Trp18 mutant (sample j) showed an enhanced affinity compared with wild-type Ac-AMP2 (sample f) under acidic conditions, whereas Phe20 mutant (sample o) and Trp20 mutant (sample p) exhibited similar and lower affinity, respectively.
The mutants containing a non-natural amino acid residue at position 18 were then further examined to probe another possibility of the affinity enhancement for chitin. Cha18 mutant (sample n) exhibited less affinity than Ala18 mutant (sample m), indicating that the side-chain benzene ring of Phe18 was important not simply as hydrophobic six-membered carbon-atom ring but as a planar aromatic ring. The pNO2-Phe18 mutant (sample h) and F5-Phe18 mutant (sample i) displayed similar and reduced affinity compared with wild-type Ac-AMP2, showing the weak and detrimental effect of these electron-withdrawing substituent groups, respectively. In contrast, the replacement of Phe18 with 1-Nal18 or 2-Nal18 enhanced the affinity. The increased extent by the replacement with 2-Nal18 (sample 1) was larger than that with 1-Nal18 (sample k) and comparable to that with Trp18 mutant (sample j).
Our recent study suggested that CH–
interactions (Nishio et al., 1995) involving a stacking aromatic residue, Tyr63, played an important role of the synergism between apolar and polar interactions in the recognition of a carbohydrate ligand by human lysozyme (Muraki et al., 2000). Although the parallelism between the side chain of Trp21 and the ring of NAG C is not perfect in the UDA VI–(GlcNAc)3 complex, six possible CH– interactions less than 3.2 Å, specifically H5 (in NAG C)–CD1 (in Trp21), 3.0; H5–CD2, 2.9; H5–NE1, 3.2; H5–CE2, 3.1; H5–CG, 2.8; and H61–CE3, 3.2 Å, were observed by generating hydrogen atoms on the NAG C residue using the `h-build' protocol in X-PLOR (Brünger, 1992). The difference in the size or the substituent group of stacking aromatic rings should affect the strength of CH– interactions by increasing or decreasing the ability of the aromatic ring as a hydrogen-bond acceptor. Therefore, the results of the present study may suggest that such an interaction is a major determinant of the affinity of Ac-AMP2 for chitin again. Although the detailed mechanism of the enhancement of affinity caused by the mutation of Phe18 to the residue with a larger aromatic ring remains to be clarified, the present study demonstrated the possibility of improving the binding strength of an HD to chitin by protein engineering. Further investigations to reveal the structure and function relationships of HD are in progress.
|
Notes |
|---|
1 To whom correspondence should be addressed. E-mail: muraki{at}nibh.go.jp
|
Acknowledgments |
|---|
We thank Mr Hiroshi Hashimoto (Hokkaido System Science) for his assistance with the MALDI-TOF/MS analysis. This work was supported by a grant from the Agency of Industrial Science and Technology MITI, Japan.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Andersen,N.H., Cao,B., Rodriguez-Romero,A. and Arreguin,B. (1993) Biochemistry, 32, 1407–1422.[Medline]
Asensio,J.L., Cañada,F.J., Bruix,M., Rodriguez-Romero,A. and Jimenez-Barbero,J. (1995) Eur. J. Biochem., 230, 621–633.[Web of Science][Medline]
Asensio,J.L., Cañada,F.J., Bruix,M., Gonzalez,C., Khiar,N., Rodriguez-Romero,A. and Jimenez-Barbero,J. (1998) Glycobiology, 8, 569–577.
Bains,G., Lee,R.T., Lee,Y.C. and Freire,E. (1992) Biochemistry, 31, 12624–12628.[Medline]
Beintema,J.J. (1994) FEBS Lett., 350, 159–163.[Web of Science][Medline]
Broekaert,W.F. et al. (1992) Biochemistry, 31, 4308–4314.[Medline]
Brünger,A.T. (1992) X-PLOR version 3.1 Manual, Yale University Press, New Haven, CT, pp. 112–114.
Chrispeels,M.J. and Raikhel,N.V. (1991) Plant Cell, 3, 1–9.
Dimarchi,R., Tam,J.P., Kent,S.B.H. and Merrifield,R.B. (1982) Int. J. Pept. Protein Res., 19, 88–93.[Medline]
Does,M.P., Ng,D.K., Dekker,H.L., Peumans,W.H., Houterman,P.M., Van Damme,E.J.M. and Cornelissen,B.J.C. (1999) Plant Mol. Biol., 39, 335–347.[Medline]
Elgavish,S. and Shaanan,B. (1997) Trends Biochem. Sci., 22, 462–467.[Web of Science][Medline]
Garcia-Hernández,E., Zubillaga,R.A., Rojo-Dominguez,A., Rodriguez-Romero,A. and Hernández-Arana,A. (1997) Proteins, 29, 467–477.[Web of Science][Medline]
Harata,K. and Muraki,M. (2000) J. Mol. Biol., 297, 673–681.[Medline]
Harata,K., Nagahora,H. and Jigami,Y. (1995) Acta Crystallogr., D51, 1013–1019.
Hom,K., Gochin,M., Peumans,W.J. and Shine,N. (1995) FEBS Lett., 361, 157–161.[Medline]
Lee,R.T., Gabius,H.-J. and Lee,Y.C. (1998) Glycoconj J., 15, 649–655.[Medline]
Martins,J.C., Maes,D., Loris,R., Pepermans,H.A.M., Wyns,L., Willem,R. and Verheyden,P. (1996) J. Mol. Biol., 258, 322–333.[Medline]
Mori,H., Uedaira,H., Ogata,K., Ishii,S. and Sarai,A. (1999) J. Mol. Biol., 292, 909–920.[Web of Science][Medline]
Muir,T.W. and Kent,S.B.H. (1993) Curr. Opin. Biotechnol., 4, 420–427.[Medline]
Muraki,M., Morii,H. and Harata,K. (1998) Protein Pept. Lett., 5, 193–198.
Muraki,M., Harata,K., Sugita,N. and Sato,K.-I. (2000) Biochemistry, 39, 292–299.[Medline]
Nagahora,H., Harata,K., Muraki,M. and Jigami,Y. (1995) Eur. J. Biochem., 233, 27–34.[Medline]
Nielsen,K.K., Nielsen,J.E., Madrid,S.M. and Mikkelsen,J.D. (1997) Plant Physiol., 113, 83–91.[Abstract]
Nishio,M., Umezawa,Y., Hirota,M. and Takeuchi,Y. (1995) Tetrahedron, 51, 8665–8701.
Oita,S., Horita,M. and Yanagi,S.O. (1996) Biosci. Biotechnol. Biochem., 60, 481–483.
Orowska,A., Wittkowska,E. and Izdebski,J. (1987) Int. J. Pept. Protein Res., 30, 141–144.
Peumans,W.J. and Van Damme,E.J.M. (1995) Histochem. J., 27, 253–271.[Web of Science][Medline]
Ponstein,A.S., Bres-Vloemans,S.,A., Sela-Buurlage,M.B., van den Elzen,P.J.M., Melchers,L.S. and Cornelissen,B.J.C. (1994) Plant Physiol., 104, 109–118.[Abstract]
Raikhel,N.V., Lee,H.-I. and Broekaert (1993) Annu. Rev., Plant Physiol. Plant Mol. Biol., 44, 591–615.[Web of Science]
Reinikainen,T., Ruohonen,L., Nevanen,T., Laaksonen,L., Kraulis,P., Jones,T.A., Knowles,J.K.C. and Teeri,T.T. (1992) Proteins, 14, 475–482.[Web of Science][Medline]
Rodriguez-Romero,A., Ravichandran,K.G. and Soriano-Garcia,M. (1991) FEBS Lett., 291, 307–309.[Medline]
Roussel,A., Fontecilla-Camps,J.-C. and Cambillau,C. (1990) J. Mol. Graphics, 8, 86–88.[Web of Science][Medline]
Siebert,H.-C. et al. (1997) Proteins, 28, 268–284.[Web of Science][Medline]
Stanford,A., Bevan,M. and Northcote,D. (1989) Mol. Gen. Genet., 215, 200–208.[Web of Science][Medline]
Verheyden,P., Pletnickx,J., Maes,D., Pepermans,H.A.M., Wyns,L., Willem,R. and Martins,J.C. (1995) FEBS Lett., 370, 245–249.[Medline]
Vyas,N.K. (1991) Curr. Opin. Struct. Biol., 1, 732–740.
Wright,C.S. (1980) J. Mol. Biol., 141, 267–291.[Web of Science][Medline]
Wright,C.S. (1984) J. Mol. Biol., 178, 91–104.[Web of Science][Medline]
Wright,C.S. (1987) J. Mol. Biol., 194, 501–529.[Web of Science][Medline]
Wright,C.S. (1989) J. Mol. Biol., 209, 475–487.[Medline]
Wright,H.T., Sandrasegaram,G. and Wright,C.S. (1991) J. Mol. Evol., 33, 283–294.[Web of Science][Medline]
Received November 11, 1999; revised February 3, 2000; accepted February 29, 2000.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Veronese, M. T. Ruiz, M. A. Coca, A. Hernandez-Lopez, H. Lee, J. I. Ibeas, B. Damsz, J. M. Pardo, P. M. Hasegawa, R. A. Bressan, et al. In Defense against Pathogens. Both Plant Sentinels and Foot Soldiers Need to Know the Enemy Plant Physiology, April 1, 2003; 131(4): 1580 - 1590. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||