Examination of the role of tyrosine-174 in the catalytic mechanism of the Arabidopsis thaliana chitinase: comparison of variant chitinases generated by site-directed mutagenesis and expressed in insect cells using baculovirus vectors.

Using the catalytic mechanism of lysozyme as a paradigm for the mechanism of other enzymes that catalyze the hydrolysis of beta-1,4-glycosidic linkages, including chitinase, we have examined the effect of chemical modification with 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide (EDC) on the reaction catalyzed by Zea mays chitinase. Inactivation with EDC did not result in derivatization of essential carboxylic acid residues, but resulted in the selective modification of a single essential tyrosine residue (Verburg, J. G., Smith, C. E., Lisek, C. A., and Huynh, Q. K., 1991, J. Biol. Chem. 267, 3886-3893). Here, we examine the role of the homologous tyrosine residue in the catalytic mechanism of the Arabidopsis thaliana chitinase. Tyrosine-174 of the Arabidopsis chitinase was replaced, with phenylalanine, alanine, histidine, and methionine by site-directed mutagenesis, and the variant chitinases were expressed in insect cells using baculovirus transfer vectors. A comparison of the reaction catalyzed by each of the variant enzymes indicates that substitution of another amino acid for Tyr-174 alters, but does not eliminate, enzymatic activity. Estimates of the specific activities of the variant chitinases reveal that substitution of His for Tyr-174 has a minimal effect on catalysis, the specific activities of the Phe and Met variants are approximately equivalent to each other, but are 60% the specific activity of wild-type Arabidopsis chitinase, and the specific activity of the Ala variant is only 40% that of wild-type. The observation that the Arabidopsis chitinase is tolerant to mutagenesis at this position suggests that Tyr-174 does not participate directly in catalysis.

Identification of an essential tyrosine residue in the catalytic site of a chitinase isolated from Zea mays that is selectively modified during inactivation with 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide.

Chitinase isolated from Zea mays seeds is inactivated by 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) in the absence of exogenous nucleophiles. Oligomers of N-acetylglucosamine,N,N',N",N"'-tetra-N-acetylchitotetraose (GlcNAc4), and to a lesser extent, N,N',N"-tri-N-acetylchitotriose (GlcNAc3) and N,N'-di-N-acetylchitobiose (GlcNAc2) provide partial protection against inactivation by the reagent. An examination of the concentration dependence of the protection afforded by GlcNAc4 revealed direct competition between the substrate analog and the reagent for the same binding sites on the enzyme. Isolation and Edman degradation of a "new" tryptic fragment, observed after inactivation of chitinase with EDC, revealed the sequence G-P-L-Q-I-S-W-N-*-N-Y-G-P-A-G-R, where the asterisk represents a cycle in which no amino acid was detected, presumably as a consequence of derivatization with EDC. In basic chitinases from dicotyledonous plants such as Arabidopsis thaliana, Phaseolis vulgaris (bean), Nicotiana tabacum (tobacco), and Solanum tuberosum (potato), as well as in the chitinase isolated from the monocotyledonous plant Hordeum vulgare (barley), this position is invariably occupied by a tyrosine. However, in the Oryza sativa (rice) basic chitinase, this position is occupied by a phenylalanine. The following additional evidence supports identification of this residue as tyrosine in Z. mays chitinase. (a) Inactivation of chitinase with EDC is reversible by treatment with hydroxylamine. (b) Liquid secondary ion mass spectrometric analysis of the isolated derivatized peptide revealed the presence of a molecular ion with a mass to charge ratio consistent with the peptide containing a derivatized tyrosine residue. These results provide evidence for an essential tyrosine residue at or near the catalytic site of chitinase that is selectively modified during inactivation with EDC.

Purification and Characterization of an Antifungal Chitinase from Arabidopsis thaliana.

Plants exhibit an altered pattern of protein synthesis in response to pathogen invasion and abiotic stress. One of these ;pathogenesis-related' proteins has been identified as chitinase, which is capable of inhibiting fungal growth in vitro. This observation has led to the suggestion that the in vivo role of chitinases is to protect plants against fungal invasion. Here, we report the purification and characterization of a basic chitinase from Arabidopsis thaliana (L.) Heynh. Columbia wild type. The purified enzyme has a molecular mass of approximately 32 kilodaltons as determined by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate, and an apparent pl of approximately 8.7 as determined by isoelectric focusing. The purified protein is an effective inhibitor of the growth of Trichoderma reesei in vitro but does not affect the growth of several other fungi. Amino acid composition analysis of the intact protein as well as amino acid composition analysis and automatic Edman degradation of isolated tryptic fragments of the enzyme indicate that it may be identical to the product of a chitinase gene isolated from an Arabidopsis genomic library (Samac DA, Hironaka CM, Yallaly PE, Shah DM [1990] Plant Physiol 93: 907-914).

Tyrosine alpha 244 is derivatized when the bovine heart mitochondrial F1-ATPase is inactivated with 5'-p-fluorosulfonylbenzoylethenoadenosine.

The bovine heart mitochondrial F1-ATPase (MF1) is inactivated by 5'-p-fluorosulfonylbenzoylethenoadenosine (FSB epsilon A) with pseudo-first order kinetics. The dependence of the rate of inactivation on the concentration of FSB epsilon A revealed an apparent Kd of 0.25 mM. ATP and ADP, and to a lesser extent, ITP and IDP provide partial protection against inactivation by the reagent. Isolation and sequence analysis of major radioactive fragments in peptic or cyanogen bromide digests of MF1 inactivated with [3H]FSB epsilon A indicate that modification of Tyr-alpha 244 is associated with the loss of activity observed. Assessment of the amount of Tyr-alpha 244 derivatized with [3H]FSB epsilon A at specific points during inactivation of the ATPase indicates that maximal inactivation is achieved on modification of this residue in slightly greater than one copy of the alpha subunit. The following characteristics of inactivation of MF1 by FSB epsilon A have also been determined. (a) The rate of inactivation of ITPase activity by FSB epsilon A is 1.4 times greater than that observed for inactivation of ATPase activity under identical conditions. (b) After maximally inactivating the capacity of MF1 to hydrolyze saturating ATP with FSB epsilon A, the modified enzyme retained its capacity to hydrolyze substoichiometric ATP. (c) Inactivation of the ATPase by FSB epsilon A is accelerated by Pi. In each of the above characteristics, MF1 modified by FSB epsilon A resembles enzyme inactivated with 5'-p-fluorosulfonylbenzoyladenosine (FSBA) more than it does enzyme inactivated with 5'-p-fluorosulfonylbenzoylinosine (FSBI). Furthermore, prior inactivation of MF1 with FSBA completely prevents labeling of Tyr-alpha 244 with [3H]FSB epsilon A, whereas prior inactivation of the enzyme with FSBI does not. Since a single catalytic site is modified when FSBI inactivates MF1 whereas three noncatalytic sites are modified when it is maximally inactivated with FSBA, it is concluded that FSB epsilon A also modifies noncatalytic sites.

Selectivity of modification when latent and activated forms of the chloroplast F1-ATPase are inactivated by 7-chloro-4-nitrobenzofurazan.

The characteristics and specificity of inactivation of the chloroplast F1-ATPase (CF1) with 7-chloro-4-nitrobenzofurazan (Nbf-Cl) have been investigated. Inactivation of the octylglucoside-dependent Mg2+-ATPase activity of latent CF1 by Nbf-Cl can be correlated with the formation of about 1.2 mol of Nbf-O-Tyr per mole of enzyme. Following inactivation of CF1 with [14C]Nbf-Cl, polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate revealed that the majority of the radioactive reagent incorporated is present in the beta subunit. Treatment of the enzyme with [14C]Nbf-Cl following dithiothreitol heat activation, led to similar labeling of the beta subunit and substantial incorporation of 14C into the gamma subunit. On complete inactivation, about 4 mol of Nbf-S-Cys is formed per mole of dithiothreitol-heat-activated CF1. Incorporation of 14C into the gamma subunit is prevented by prior treatment of the latent CF1 or of the dithiothreitol-heat-activated CF1 with iodoacetamide. Following incubation of the dithiothreitol-heat-activated CF1 with iodoacetamide, complete inactivation of the octylglucoside-dependent Mg2+-ATPase activity by Nbf-Cl can be correlated with the formation of about 1.2 mol of Nbf-O-Tyr per mole of enzyme. After stabilization of the [14C]Nbf-O-Tyr derivative by treatment with sodium dithionite, a labeled peptide was purified. Automatic Edman degradation of this peptide revealed the sequence V-X-V-P-A-D-(D). The majority of the radioactivity was cleaved in the second cycle, the position occupied in CF1 by Tyr-beta-328, which is homologous to Tyr-beta-311, the residue reactive with Nbf-Cl in the beef heart mitochondrial F1-ATPase. When CF1, modified at Tyr-beta-328 with Nbf-Cl, is incubated at pH 9.0, the Nbf-O-Tyr adduct is hydrolyzed, leading to concomitant recovery of the ATPase activity. In double labeling experiments, two-dimensional isoelectric focusing in the presence of urea followed by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate indicates that 2-azido-ADP, covalently bound at the tight ADP binding site, and the tyrosine modified by [14C]Nbf-Cl are located in different beta subunits.

Localization of sites modified during inactivation of the bovine heart mitochondrial F1-ATPase by quinacrine mustard using [3H]aniline as a probe.

The aziridinium of purified quinacrine mustard at 50 microM inactivates the bovine heart mitochondrial F1-ATPase with a pseudo-first order rate constant of 0.07 min-1 at pH 7.0 and 23 degrees C. An apparent Kd of 27 microM for the enzyme-reagent complex was estimated from the dependence of the rate of inactivation on the concentration of quinacrine mustard. The pH inactivation profile revealed that deprotonation of a group with a pKa of about 6.7 is necessary for inactivation. The amount of reagent incorporated into the protein increased linearly with the extent of inactivation. Complete inactivation was estimated to occur when 3 mol of reagent were incorporated/mol of F1. Enzyme, in which steady state ATPase was inactivated by 98% by quinacrine mustard, hydrolyzed substoichiometric ATP with zero order kinetics suggesting that residual activity is catalyzed by F1 in which at least one beta subunit is modified. By exploiting the reactivity of the aziridinium of covalently attached reagent with [3H] aniline, sites modified by quinacrine mustard were labeled with 3H. Isolation of radioactive cyanogen bromide peptides derived from F1 inactivated with the reagent in the presence of [3H]aniline which were identified by sequence analysis and sequence analyses of radioactive tryptic fragments arising from them have revealed the following. About two thirds of the radioactivity incorporated into the enzyme during inactivation is apparently esterified to one or more of the carboxylic acid side chains in a CNBr-tryptic fragment of the beta subunit with the sequence: 394DELSEEDK401. The remainder of the radioactivity is associated with at least two sites within the cyanogen bromide peptide containing residues 293-358 of the beta subunit. From these results it is concluded that inactivation of F1 by the aziridinium of quinacrine mustard is due, at least in part, to modification of one or more of the carboxylic acid side chains in the DELSEED segment of the beta subunit and possibly also to modification of unspecified amino acid side chains between residues 302-356 of the beta subunit.

Evidence for functional heterogeneity among the catalytic sites of the bovine heart mitochondrial F1-ATPase.

The characteristics of ATP hydrolysis at a single catalytic site of the bovine heart F1-ATPase (MF1) as originally described by Grubmeyer et al. (Grubmeyer, C., Cross, R.L., and Penefsky, H.S. (1982) J. Biol. Chem. 257, 12092-12100) were compared with those of various chemically modified preparations of MF1 in which the steady state activity was severely attenuated. Although it was not necessary to age our preparations of native MF1 in the presence of 2 mM Pi to observe the same characteristics of single site catalysis, such aging did shift the equilibrium of bound substrate and bound products at the single catalytic site in favor of ATP. After loading a single catalytic site on the enzyme with substoichiometric [alpha,gamma-32P]ATP, the addition of 5-20 microM ATP or ADP was effective in promoting both the hydrolysis of bound [alpha,gamma-32P]ATP and release of radioactive products. Under these conditions, the 5-20 microM ATP added as promoter was hydrolyzed at a rate commensurate with the turnover rate of the enzyme, whereas the promoted hydrolysis of the [alpha,gamma-32P]ATP, preloaded at a single catalytic site, was considerably slower. Therefore, the high affinity, single catalytic site loaded first does not directly contribute to steady state ATP hydrolysis. That the single, high affinity catalytic site is not a "normal" catalytic site is supported by the properties of enzyme modified by 5'-p-fluorosulfonylbenzoyladenosine which exhibits only slightly altered characteristics of single site catalysis and promoted single site catalysis, despite exhibiting severely attenuated steady state turnover. Other modified forms of the enzyme in which the steady state activity was severely attenuated by derivatization with 5'-p-fluorosulfonylbenzoylinosine, 7-chloro-4-nitrobenzofurazan, or 1,5-difluoro-2,4-dinitrobenzene also bound substoichiometric ATP at a single catalytic site. However, the characteristics of single site hydrolysis by these modified forms of the enzyme differed considerably from those of native MF1.

The use of dithionite reduction to identify the essential tyrosine residue in the F1-ATPase from the thermophilic bacterium, PS3, that reacts with 7-chloro-4-nitrobenzofurazan.

When the F1-ATPase from the thermophilic bacterium, PS3, was inactivated by greater than 90% with 7-chloro-4-nitro[14C]benzofurazan ([14C]Nbf-Cl) at pH 7.4, 1.4 mol of [14C]Nbf were incorporated per mol of enzyme. After pepsin digestion of the labeled enzyme at pH 3.0, a single, major peak of radioactivity was resolved by reversed-phase high-performance liquid chromatography under acidic conditions were peptidyl Nbf-O-tyrosine is stable. This radioactive peak, designated RP-1, eluted with a retention time of 95 min. When the material in RP-1 was subjected to reversed-phase high-performance liquid chromatography under the same conditions after treatment with sodium dithionite, a single, major peak of radioactivity, designated RP-2, was resolved with a retention time of 52 min. Automatic Edman degradation of this material revealed that it has the amino acid sequence I-Y*-V-P-A-D-(D), where Y* presumably represents peptidyl [14C]Nbf-O-tyrosine. These results provide the basis for a facile method to purify peptides containing [14C]Nbf-O-tyrosine in which the labeled residues can be identified by amino acid sequence analysis using the Edman degradation.