modification of bovine pancreatic ribonuclease a with 6-chloropurine riboside

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 246, No. 2, May 1, pp. 681-689,1986

Modification of Bovine Pancreatic Ribonuclease A with 6-Chloropurine Riboside’

JULI ALONSO,* M. VICTORIA NOGUtiS,* AND CLAUDI M. CUCHILLO*jt

*Depa&ment de Bioquimica, Facultat de CiGncies, and tIn&itut de Biologia Fonamentd “Vicent Villar Palo&,” Universitat Authnna de Barcehn, Spain

Received July 81985, and in revised form December 18,1985

The chemical modification of bovine pancreatic ribonuclease A by 6-chloropurine riboside was studied to obtain information about the role of the purine nucleoside moiety of the ribonucleic acid in the enzyme-substrate interaction. The residues involved in the reaction were identified, after performic acid oxidation and trypsin digestion, by reverse-phase HPLC peptide mapping. The labeled peptides were detected by following the absorbance at 254 nm, and amino acid analyses of these peptides showed that the reaction had taken place with the amino groups of Lys-1, -37, -41, and -91. The specificity of the reaction was unaffected by changing the 1igand:protein molar ratio. Partial separation of the reaction products was accomplished by means of chromatography on CM- Sepharose: four labeled fractions corresponding to mono- and bisubstituted derivatives were found. One of the monosubstituted fractions (fraction E) contained a homogeneous protein &ith the nucleoside bound to the a-amino group of Lys-1 whereas the other (fraction D) was a mixture of derivatives labeled in the t-amino group of Lys-1, -37, -41, and -91. Kinetic studies of these two monosubstituted fractions were performed with cytidine 2’,3’-phosphate and ribonucleic acid as substrates. These derivatives showed a noncompetitive inhibition-like behavior with respect to RNase A. Results support the existence of several RNase A regions with affinity for purine nucleosides. o 1986 Academic

Press, Inc.

The interaction between ribonuclease A (RNase A, EC 3.1.27.5) and its high molec- ular weight substrate RNA has been studied in several ways (1, 2). The use of substrate analog ligands that can react covalently with the enzyme has allowed postulation of the existence of several binding subsites. One of these analogs is the halogenated ribonucleotide 6-chloropurine riboside 5’-monophosphate (c~~RMP)~ which was used with the aim of locating

i This work was supported by Grant 1716/82 from the Comision Asesora de Investigation Cientifica y Tecnica of the Ministerio de Education y Ciencia, Spain, and by the Fundacio M. F. de Roviralta, Bar- celona, Spain.

* Abbreviations used: clsRMP, 6-chloropurine riboside S-monophosphate; TPCK, L-1-tosylamido-2- phenylethyl chloromethyl ketone; TFA, trifluoroacetic acid.

the purine binding subsite( Reaction of this compound with the enzyme is highly specific as only one major derivative is found. The nucleotide reacted with the (Y- amino group of Lys-1 by means of a nu- cleophilic attack from the protein to the carbon in the 6 position of the base (3). However, it was found that the very high specificity of the reaction was due mainly to the interaction of the phosphate group of the nucleotide with a phosphate-binding subsite in the enzyme which was called p2 (3). In the present work the study of the role of the purine nucleoside moiety was approached by using the nucleoside 6-chloropurine riboside. In this case the reaction with RNase A is less specific as several derivatives were separated on a CM-Sephar- ose column. It was found that the reaction products were mono- and bisubstituted de-

681 0003-9861/86 $3.06 Copyright 0 1986 by Aeademie Preaa, Inc.

All rights of reproduction in any form reserved.

682 ALONSO, NOGUfiS, AND CUCHILLO

rivatives and that the reaction sites were the c-amino groups of some lysine residues and the a-amino group of Lys-1. The first step in the study of the derivatives was the identification of the residues involved in the reaction and the partial characterization of one of the derivatives obtained. Re- verse-phase HPLC was used in this work as the results found in previous studies in which electrophoretic. and thin-layer chromatography techniques were used for the separation of tryptic digest products of oxidized RNase A and its derivatives were not clear. The identification and complete purification of different derivatives can be of great value in order to obtain information about the involvement of specific regions of the enzyme in the interaction with the substrate.

MATERIALS AND METHODS

Materials Bovine pancreatic ribonuclease (twice crystallized) was from Biozyme (Blaenavon, U. K.). TPCK (L-1-tosylamido-2-phenylethyl chloromethyl ketone)-treated trypsin was from Worthington Bio- chemical Corporation (Freehold, N. J.). RNA from yeast (type XI), cytidine 2’(3’)-phosphate (mixture of isomers), 6-chloropurine riboside, Trizma base, and dialysis sacks 250-‘7U were obtained from Sigma Chemical Company (Saint Louis, MO.). Trifluoroacetic acid (TFA) was from Fluka (Buchs, Switzerland). Constant-boiling-point 6 N HCl was purchased from Pierce Chemical Company (Rockland, Ill.). HPLC grade acetonitrile and HCl were products from Far- mitalia Carlo Erba (Milan, Italy). (NH&COI, NaCl, sodium acetate, formic acid, H20*,2-mercaptoethanol, and phenol were obtained from Merck (Darmstadt, GFR). Twice distilled water filtered through a Nor- ganic cartridge and a 0.22~pm membrane filter (Mil- lipore Corp., Bedford, Mass.) was used in the HPLC experiments.

A reverse-phase HPLC column, PBondapak C1s (300 X 4 mm), and a precolumn stationary phase, Bondapak Cls Corasil, were purchased from Waters (Milford, Mass.).

CM-Sepharose CL-6B resin was obtained from Pharmacia Fine Chemicals (Uppsala, Sweden).

Cytidine 2’,3’-phosphate was synthesized by the method of Szer and Shugar (4).

The Bio-Rad protein assay (Bio-Rad Laboratories, Munich, GFR) was used in the determination of the molar absorbance coefficient.

Apparatus. All HPLC experiments were carried out with a modular Waters HPLC consisting of two pumps (Model 6000 A controlled by the Automated Gradient

Controller Model 680). Samples were detected by monitoring the effluent at 214 nm and either 254 or 280 nm with an Absorbance Detector Model 441 and a Variable Wavelength Data Model 450. A Micropro- cessor Data Module M730 and a Universal Liquid Chromatography Injector U6K were also used. Injec- tions were made with a Hamilton precision syringe (802 RNE, Reno, Nev.). An Eppendorf 3414 centrifuge (Eppendorf Geratebau, Hamburg, GFR) was used, routinely, to remove any particulate material before injecting all samples into the HPLC system.

RNasepur$cation Bovine pancreatic ribonuclease was purified by the method of Taborsky (5) to obtain the RNase A fraction. The method was modified in that CM-Sepharose CL-6B was used instead of the original CM-cellulose.

Reaction of RNase A with &chlcrropurine riboside. 6-Chloropurine riboside (1 g, 3.487 mmol) was dissolved in 50 ml of Tris/HCl, pH 7.4 (containing 0.1 M NaCl), at 40°C and 275 mg (0.02 mmol) of RNase A was added. The pH was checked and adjusted, when necessary, with HCI. The mixture was kept at 40°C for 48 h. Separation and purification of the different fractions were carried out by the method described by Pa&s et al. (3) but using CM-Sepharose CL-6B instead of CM-cellulose. The molar absorbance coefficient at 270 nm of the different fractions was determined by means of the Bio-Rad protein assay (6), using RNase A as reference. No interference from the nucleoside was found.

To determine the influence of the nucleoside concentration on the reaction, analytical-scale experiments with different RNase A:nucleoside molar ratios were carried out with 10 mg (0.731 pmol) of RNase A and the following amounts of 6-chloropurine riboside dissolved in 5 ml: 36 mg (0.127 mmol), 21 mg (74.6 pmol), 2.1 mg (7.46 pmol), 0.9 mg (3.14 pmol), and 0.2 mg (0.704 pmol).

Oxidation and tryptic digestion The different samples (2 mg) were oxidized with performic acid by the procedure of Hirs (7) and dissolved in 0.3 ml of 0.1 M

ammonium carbonate, pH 8.5, and 10 ~1 of a 1 mg/ml trypsin solution in 1 mM HCl was added. The mixture was incubated at 37°C for 2 h and an additional 10 ~1 of enzyme solution was added. After 4 h the mixture was lyophilized.

HPLC separation of tryptic peptides. Peptide maps of RNase A and derivatives were prepared by injecting 0.1-0.5 mg of the different digested samples dissolved in solvent A (water + 0.05% TFA). Elution was carried out with an initial 5-min wash and an 80-min linear gradient from 100% solvent A to 60% solvent A plus 40% solvent B [acetonitrile:water (80~20, v/v) + 0.04% TFA] at a flow rate of 1 ml/min. Routinely, the column was run at a pressure between 7600 and 9650 kPa (1100-1400 psi). At the end of each day, the column and the chromatographic system were washed with acetonitrile for 30 min.

RIBONUCLEASE A DERIVATIVES 683

Amino acid analysis. Amino acid analyses were carried out with a Rank Hilger Chromaspek J 180 ion-exchange chromatograph (London, U. K.) provided with a 350 X 3-mm column. The effluent was monitored by fluorescence. About 15 nmol of the different peptides was dissolved in 50 ~1 of 6 N constant-boiling- point HCl (with 0.05% each of 2-mercaptoethanol and phenol added), hydrolyzed for 48 h in ‘uocuo at 105”C, and lyophilized. The different hydrolyzates were in- jected into the amino acid analyzer and the elution was carried out by using citrate-borate buffers of in- creasing pH (between 2 and 11.6) according to the method of Murren et al (8). The fluorescent amino acid derivatives were obtained by the method of Drescher et al (9). Norleucine was added as internal standard to the samples.

Kinetic assays. Either cytidine 2’,3’-phosphate or RNA was used as substrate. In the case of cytidine 2’,3’-phosphate the substrate concentration range was from 0.102 to 1.387 mM. The activity was measured by recording the increase in absorbance at 296 nm. The kinetic parameters for RNA were determined using a concentration range from 0.2 to 4 mg ml-‘. The decrease in absorbance at 300 nm was measured. All assays were carried out in 0.1 M sodium acetate/HCl buffer, pH 5.0, containing 0.1 M NaCl at 25°C.

The kinetic parameters were calculated by means of the Lineweaver-Burk plot and the values were sta- tistically adjusted by means of the least-squares method.

RESULTS AND DISCUSSION

RNase A trgptic peptide map. Separation by reverse-phase high-performance liquid chromatography of the tryptic peptides obtained from performate-oxidized samples of RNase A is shown in Fig. 1. In this chromatogram there appear 15 peaks which correspond to the 13 peptides pre- dicted for trypsin action plus two peptides arising from the unspecific partial cleavage of the Tyr-‘76~Ser-77 bond. The stability of both the Lys-1-Glu-2 and the Lys-41-Pro- 42 bonds was confirmed and the general order of elution of the different peptides is in agreement with that reported by Black- burn and Gavilanes (10,ll) although in our system all peaks are completely resolved and the separation is achieved in a signif- icantly shorter time.

The use of performic acid is satisfactory for the oxidation of RNase A because this protein does not contain tryptophan. It should be noted that the four methionine residues are transformed to methionine sulfone and the eight cysteine residues to cysteic acid (7). Recently, McWherter et al.

l-

‘P

FIG. 1. Separation of a tryptic digest of RNase A by reverse-phase HPLC. See text for experimental conditions. Peak identification and sequences are as follows: Peak 4, Ser-32-Arg-33; Peak 6, Asp- 3%Arg-39; Peak 11, Glu-86-Lys-91; Peak 13, Thr-99-Lys-104; Peak 1, Lys-1-Lys-7; Peak 9, Asn-62- Lys-66; Peak 5, Asn-34-Lys-37; Peak 2, Phe-8-Arg-1O; Peak 3, Gln-11-Lys-31; Peak lob, Ser-77-Arg- 85; Peak 12, Tyr-92-Lys-98; Peak lOa, Asn-67-Tyr-76; Peak 10, Asn-67-Arg-85; Peaks 7-8, Cys-40- Lys-61; and Peak 14, His-105-Val-124

684 ALONSO, NOGUfiS, AND CUCHILLO

(12) published an HPLC separation of a tryptic digest of carboxymethylated RNase A. The experimental conditions for the separation are very similar although the order of elution is somewhat different, the changes being due to the different hydrophobicity of cysteic acid and methionine sulfone present in the performic acid-oxidized sample with respect to methionine and carboxymethylated cysteine resulting from iodoacetate treatment. The peptides with cysteic acid and/or methionine sulfone are more hydrophilic than those with carboxymethylated cysteine and/or methionine. It is for this reason that in the elution pattern described by McWherter et al. (12) these peptides are more retarded.

Identification of the residues involved in the reaction between RNase A and 6-chloropurine r&side. The residues involved in the reaction between RNase A and 6-chloropurine riboside were characterized by means of the HPLC tryptic peptide map of the unfractionated mixture of reaction products.

In the HPLC chromatogram the labeled peptides were detected by following the absorbance at 254 nm characteristic of nu- cleic acid derivatives. The retention time of the labeled peptides is higher than that of the normal peptides because the nucleoside increases the hydrophobicity of the

peptide. Figure 2 shows the peptide map of the reaction products where there appear six main peaks, absent from the RNase A peptide map, with a high 254-nm absorbance. Peak I corresponds to unreacted nucleoside that was not eliminated in the dialysis treatment.

The sequences of the peptides corresponding to the remaining peaks can be deduced from the amino acid composition. They show a very low recovery of lysine residues, a fact that is in agreement with the results of Pares et al. (3) and that points to the amino groups of lysine residues as the site of reaction. The nucleoside-modified lysines appear in the amino acid chromatogram as a peak in the region between glycine and alanine. The results are also in agreement with the known resistance of peptide bonds with modified lysine residues to trypsin action (13).

Amino acid analysis of peak II showed that this peak corresponds to peptide 1 (Lys-1-Lys-7). The reaction with the nucleoside took place in Lys-1 because the bond between Lys-7 and Phe-8 was cleaved. Peak III contains peptide 5-6 (Asn-3CArg- 39), and, therefore, the labeled amino acid residue is Lys-37. Amino acid analysis of peak IV showed that peptide 11-12 (Glu- 86-Lys-98) is labeled in Lys-91. Peaks V and VI both correspond to peptides modi-

FIG. 2. Separation by reverse-phase HPLC of a tryptic digest of the reaction products between RNase A and 6-chloropurine riboside. See text for experimental conditions.


TABLE I

PERCENTAGE OF EACH LABELED RESIDUE IN THE REACTION BETWEEN RNase A AND

6-CHLOROPURINE RIBOSIDE

Labeled residue Extent of labeling”

(%I

Lys-1 42 Lys-41 34 Lys-37 9 Lys-91 8 Other 7

a As percentage of the total amount of labeled derivatives.

lied at residue Lys-41. Amino acid composition shows that peak V is Cys-40-Phe-46 and is produced by an unspecific cleavage of the peptide chain whereas peak VI corresponds to peptide ‘7-8 (Cys-40-Lys-61). We do not know the reason for this unspecific cleavage when Lys-41 is modified, but Bello et al. (14) found the peptide Cys-40- Asn-44 in a tryptic digest of RNase A modified at Lys-41 with the reagent 2-carboxy- 4,6-dinitrochlorobenzene. Bello et al. (14) also observed an increase of other unspecific cleavages in Lys-41-modified RNase A derivatives.

2

To obtain a high degree of specificity in a site-directed reaction it is necessary to use a low 1igand:protein molar ratio (14). For this reason, the effect of the ligand: protein ratio (175:1, lOO:l, lO:l, 4:1, and 1:l) on the relative amounts of each modified residue was studied. It was found that (i) the labeled peptides are the same at all molar ratios, (ii) the relative amounts of each labeled peptide (Table I) are indepen- dent of the molar ratio, and (iii) the extent of the reaction decreases markedly when the 1igand:protein molar ratio decreases. It can be concluded that the specificity of the reaction is essentially unaffected by the molar ratio of the reagents and thus it was possible to use a high 1igand:protein molar ratio to obtain the highest yield of derivatives.

Separation and identification of derivatives of the reaction between RNase A and 6-chloropurine r&side. Figure 3 shows the chromatographic separation of the products from a large-scale reaction between RNase A (0.02 mmol) and the nucleoside (3.487 mmol) on a CM-Sepharose column. On the basis of spectral analyses the following tentative identification can be made. Fraction A is unreacted nucleoside not eliminated by the dialysis treatment, fractions B-E are nucleoside-containing deriv-

FIG. 3. CM-Sepharose CL-6B chromatography of the products of the reaction between RNase A and 6-chloropurine riboside. Column, 1.6 X 40 cm. Flow rate, 25 ml/h. Elution was carried out in 5 mM Tris-HCl, pH 8.0, with a linear NaCl gradient (O-O.15 M). (-) Absorbance at 278 nm, (---) NaCl concentration.

686 ALONSO, NOGUtiS, AND CUCHILLO

atives of the enzyme, and fraction F is unreacted native RNase A. The RNase A recovery is usually around 40% .These results are similar to those described by Pares et al. in the reaction with the nucleotide (3). The spectral properties of fractions B-E, both direct and difference, are very similar to those of a 6-aminoalkyl purine riboside derivative, a structure that should have been found if the reaction had taken place with the amino group of a lysine residue. The molar absorbance coefficients at 270 nm of fractions B, C, D, and E are 30,920, 28,885,20,500, and 21,900 M-l cm-‘, respec- tively. This indicates that fractions B and C are bisubstituted derivatives whereas fractions D and E are monosubstituted ones. Fractions B-E were rechromato- graphed and submitted to performic acid

II 8.5-

? * N l.25- z 1

TllE (nln)

!

,0.4

?

f N

0.2 ;

f I

3

0

FIG. 4. Separation by reverse-phase HPLC of a tryptic digest of fraction E (Panel I) and fraction D (Panel II) obtained as shown in Fig. 3. See text for experimental conditions.

I I I 1 I I 1 I

IO 20 30 40 50 60 70 80 TIIE WI)

II

IO 20 10 50 8U 70 80 TllE (mln)

FIG. 5. Separation by reverse-phase HPLC of a tryptic digest of fraction C (Panel I) and fraction B (Panel II) obtained as shown in Fig. 3. See text for experimental conditions.

oxidation and tryptic digestion in order to determine the corresponding modified residues. Figures 4 and 5 show the reverse- phase HPLC peptide maps.

In Fig. 4, panel I, we can see the tryptic peptide map obtained from fraction E. The most important fact with respect to the RNase A peptide map is the presence of a new peak with a high 254-nm absorbance and the disappearance of the peak corresponding to peptide 1 (Lys-1-Lys-7). Amino acid analysis of the substituted peptide showed that it corresponded to peak II of Fig. 2; therefore the label was on Lys-1. It will be seen in the discussion about fraction D that one of its components is also a monosubstituted derivative on Lys-1. It can be accepted that the hydrophobicities of both a-NH2 and c-NH2 substituted peptides


must be very likely the same and thus these peptides will have the same retention time on HPLC chromatography (peak II). How- ever, as fraction D is less basic than fraction E according to the elution pattern from the CM-Sepharose chromatography, it is reasonable to think that in fraction D a lysine amino group with a pK higher than that of fraction E has been substituted. It can be tentatively concluded that in fraction D the substitution takes place in the c-NH2 of Lys-1 (pK 10.5) and that in fraction E the substitution takes place in the a-NH2 (pK 7.8). It can be seen then that fraction E is an analog of derivative II obtained by reaction with 6-chloropurine riboside 5’-monophosphate (3, 15).

The peptide map corresponding to fraction D shown in Fig. 4, panel II, is more complex. In particular there can be observed five peaks with absorbance at 254 nm as well as all the peaks observed in the RNase A profile at 214 nm. Amino acid analyses indicate that the peaks with absorbance at 254 nm correspond to peaks II, III, IV, V, and VI of Fig. 2. Peak II corresponds to peptide Lys-1-Lys-7 modified at Lys-1. As commented above peak II of fraction D probably corresponds to the e- NH2 substituted derivative of Lys-1. Since peaks V and VI both correspond to derivatives substituted at the Lys-41 position it can be concluded that fraction D is a mixture of four monoderivatives labeled in the t-amino groups of lysines 1, 37 (peak III), 41, and 91 (peak IV).

Figure 5, panel I, shows the tryptic peptide map obtained from fraction C. We can see four peaks that absorb at 254 nm, the area of the first peak (peak II) being equal to the sum of the other three peaks, and the disappearance (at 214 nm) of the peak corresponding to peptide 1 (Lys-1-Lys-7). The amino acid analyses demonstrated that the labeled peaks correspond to peaks II, III, V, and VI of Fig. 2. These results together with the spectral data indicate that fraction C is a mixture of two bisubstituted derivatives labeled both in the LY- amino group of Lys-1 and in the c-amino group of either Lys-37 or Lys-41. The as- signment to the cu-NHz of Lys-1 as the site of substitution, as well as that of c-NH2 of

Lys-1 for fraction B, is based on the same reasoning used in the case of fractions D and E.

The chromatographic profile of the peptide map corresponding to fraction B, shown in Fig. 5, panel II, is complex. The main feature is its great similarity with fraction D (Fig. 4, panel II). Amino acid analyses showed that the labeled residues are the same as those of fraction D. There- fore, fraction B is likely a mixture of bisubstituted derivatives labeled in several t-amino groups of the enzyme.

Kinetic studies. Table II shows the kinetic parameters of RNase A and fractions D and E of Fig. 3. By using either cytidine 2’,3’-phosphate or RNA as substrate (at pH 5.0) the derivatives show a noncompetitive inhibition-like pattern with respect to RNase A. The enzymatic activity is more affected in fraction D, a fact that can be explained by the role of Lys-41 in the mechanism of catalysis (1). The apparent Km is not affected by the bound nucleoside and thus it can be suggested that none of the different labeled positions interfere with the pyrimidine nucleotide binding site.

The comparison of the kinetic properties of fraction E and derivative II described by Pares et al. (3) suggests a role for the phosphate group in the interaction between the RNA and the modified enzyme as the nucleotide-modified derivative has a larger Km for RNA with respect to native RNase A than the nucleoside-modified derivatives.

GENERAL DISCUSSION

The existence of a binding site in RNase A that prefers purines has long been known (16, 17). Richards and Wyckoff (1) found that the nucleoside moiety of 5’-AMP was bound in a region (&Rz) different from that corresponding to 3’-CMP (B,Ri). Pares et al (3) tried to covalently label the & region by reaction of RNase A with the nucleotide 6-chloropurine riboside 5’-monophosphate. However, at the high 1igand:enzyme molar ratio used and probably because no ade- quate nucleophile is present in the Bz site, the nucleotide reacted in a site different from that expected, mainly due to the ex-

688 ALONSO, NOGUES, AND CUCHILLO

TABLE II

COMPARATIVE KINETIC PROPERTIES OF RNase A AND SOME DERIVATIVES AT pH 5.0

Enzyme Substrate Km b-0 kat (mini) Relative V/[E,]

RNase A 2,3’Cp’ 0.36 + 0.02 61.3 f 1.9 Derivative E 2,3’Cp 0.32 f 0.02 43.7 5 1.5 Derivative D 2,3’Cp 0.31 * 0.03 24.9 f 1.0

(mg ml-‘)

RNase A RNA 0.68 + 0.04 100 Derivative E RNA 0.63 + 0.03 96.2 I? 1.8 Derivative D RNA 0.54 + 0.03 59.3 -+ 1.6

Note. See text for experimental conditions. In the experiments with RNA, relative values of V/[Eo] are given. For each set of parallel experiments they are equal to lOo[( V/[E,])derivative]/[( V/[Eo])native]. The values of native RNase A are taken as 100.

a 2’,3’Cp = cytidine 2,3’-phosphate.

istence of a secondary phosphate binding site (pz). The directionality of the reaction was likely due to the presence of a positive charge in p2 (probably Lys-7 or Arg-10) which could interact specifically with the phosphate group of the halogenated nucleotide. However, because of the strong interaction between the phosphate and the enzyme the question of the role of the purine base (or nucleoside) moiety in such interaction remained without answer. The affinity of a nucleoside toward the enzyme is much lower than that of the corresponding nucleotide (whatever the position of the phosphate) and thus it is not surprising to see more derivatives in the reaction between RNase A and the halogenated nucleoside than with the halogenated nucleotide. Nevertheless, it can be seen that from the 11 amino groups found in RNase A only 5 (a-amino of Lys-1 and c-amino of Lys-1, Lys-37, Lys-41, and Lys-91) react to some extent with the nucleoside. The extent of labeling of the different reacting residues (Table I) indicates that Lys-1 and Lys-41 are the positions with a higher degree of reaction, while the other reacting residues (Lys-3’7 and Lys-91) have a lower degree of reaction. Although the higher reactivity of a-amino of Lys-1 and c-NH2 of Lys-41 could also be accounted for in terms of their lower pK, values, it is clear that there is some specificity in the reaction. In fact no

significant reaction is seen with Lys-7, -31, -61, -66, -98, and -104 even though they are exposed to the solvent (1) and have pK, values similar to those of Lys-37 and Lys- 91 (18). In addition, Blackburn and Gavi- lanes (11) observed a partial protection of these lysines in the study of the effect of poly(A) on the amidination of lysine residues. Although it could be argued that the reaction takes place under rather drastic conditions it should be noted that they are the same as those used in the affinity labeling by the corresponding nucleotide (3).

The kinetic study of derivative E is par- ticularly interesting, since it is the nucleoside derivative homologous to the nucleotide derivative, called derivative II, that has been extensively studied (3). On the one hand it has the same Km values, at pH 5.0, as RNase A both with RNA and with cytidine 2’,3’-phosphate as substrate. On the other hand it has a lower kcat with cytidine 2’,3’-phosphate but not with RNA. It can be imagined that the presence of the label in the binding site B3R3 [see Pares et aL (3) for nomenclature] does not prevent the binding of either RNA or cytidine 2’,3’- phosphate. However, in the case of the substrate RNA this binding is prevented to some extent in derivative II, where the phosphate group of a covalently bound nucleotide in Lys-1 is present in p2 (3). There is no obvious explanation for the decrease


in the kcat values found with the low mo- lecular weight substrate cytidine 2’,3’- phosphate, although it could be accounted for in terms of an unfavorable conforma- tional change of the enzyme brought about by the ligand (either nucleoside or nucleotide). This conformation change could be responsible for the unspecific trypsin cleavage observed at Phe-46 (see above).

3.

4.

zymes (Boyer, P. D., ed.), 3rd ed., Vol. 15, pp. 317-433, Academic Press, New York.

PAR!%, X., LLORENS, R., AR&, C., AND CUCHILLO, C. M. (1980) Eur. J. Biochem 105,571-579.

SZER, W., AND SHUGAR, D. (1963) Biochem Prep. 10,139-144.

5. 6. 7. 8.

To further clarify the role of the nucleoside moiety, experiments are in progress on the purification and characterization of the derivatives labeled in other positions (peaks B-D) as well as on the possible in- hibitory effect of natural purine nucleosides. In conclusion, we have developed an HPLC technique that allows the separation and identification of the tryptic peptides of RNase A and its derivatives resulting from the reaction with 6-chloropurine riboside. Localization and identification of the labeled residues suggest that Lys-1, Lys-41, Lys-37, and Lys-91 may correspond to areas of nucleoside binding in RNase A.

9.

10.

11.

12.

13.

14.

15.

TABORSKY, G. (1959) J. Biol C&m. 234,2652-2656. BRADFORD, M. (1976) And B&hem 72,248-254. HIRS, C. H. W. (1956) J. BioL Ch 219,611-621. MURREN, C., STELLING, D., AND FELSTEAD, G. (1975)

J. Chromatogr. 115,236-239. DRESCHER, M. J., MEDINA, J. E., AND DRESCHER,

D. G. (1981) AnaL Biochem 116,280-286. BLACKBURN, P., AND GAVILANES, J. G. (1980) J.

BioL Chem 255,10959-10965. BLACKBURN, P., AND GAVILANES, J. G. (1982) J.

BioL Chm 257,316-321. MCWHERTER, C. A., THANNHAUSER, T. W., FRED-

RICKSON, R. A., ZAGOTPA, M. T., AND SCHERAGA, H. A. (1984) Anal. Biochxm. 141,523-537.

SMYTH, D. G. (1967) in Methods in Enzymology (Hirs, C. H. W., and Timasheff, S. N., eds.), Vol. 11, pp. 214-231, Academic Press, New York.

BELLO, J., IIJIMA, H., AND KARTHA, G. (1979) Int. J. Peptide Protein Res. 14,199-212.

PAR!&, X., PUIGDOM~~NECH, P., AND CUCHILLO, C. M. (1980) Int J. Peptide Protein Res. 16,241- 244.

REFERENCES 16.

1. RICHARDS, F. M., AND WYCKOFF, H. W. (1971) in The Enzymes (Boyer, P. D., ed.), 3rd ed., Vol. 4, pp. 64’7-806, Academic Press, New York.

2. BLACKBURN, P., AND MOORE, S. (1982) in The En-

17.

18.

DEAVIN, A., MATHIAS, A. P., AND RABIN, B. R. (1966) Nature &n&n) 211,252-255.

DEAVIN, A., MATHIAS, A. P., AND RABIN, B. R. (1966) Bimhem+ J. 101,14-16.

GOLDFARB, A. R., AND MARTIN, P. D. (1976) Bioorg. Chew. 5,137-145.

modification of bovine pancreatic ribonuclease a with 6-chloropurine riboside

Documents