HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 26, 19545-19551, June 30, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/26/19545    most recent M909020199v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chow, K. M. Articles by Hersh, L. B. Search for Related Content PubMed PubMed Citation Articles by Chow, K. M. Articles by Hersh, L. B. Social Bookmarking                      What's this?

Studies on the Subsite Specificity of Rat Nardilysin (N-Arginine Dibasic Convertase)^*

K. Martin Chow^§, Eva Csuhai^¶, Maria Aparecida Juliano, Jan St. Pyrek^**, Luiz Juliano, and Louis B. Hersh^§

From the ^§ Department of Biochemistry and ^** Mass Spectrometry Facility, University of Kentucky, Lexington, Kentucky 40563-0298, the ^¶ Department of Chemistry, Transylvania University, Lexington, Kentucky 40508, and the  Department of Biophysics, Escola Paulista de Medicina, 04034 Sao Paulo, Brazil

Received for publication, November 10, 1999, and in revised form, April 7, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The subsite specificity of rat nardilysin was investigated using fluorogenic substrates of the type 2-aminobenzoyl-GGX₁X₂RKX₃GQ-ethylenediamine-2,4-dinitrophenyl, where P₂, P₂', and P₃ residues were varied. (The nomenclature of Schechter and Berger (Schechter, I., and Berger, A. (1967) Biochem. Biophys. Res. Commun. 27, 157-162) is used where cleavage of a peptide occurs between the P₁ and P₁' residues, and adjacent residues are designated P₂, P₃, P₂', P₃', etc.) There was little effect on K_m among different residues at any of these positions. In contrast, residues at each position affected k_cat, with P₂ residues having the greatest effect. The S₃, S₂, and S₂' subsites differed in their amino acid preference. Tryptophan and serine, which produced poor substrates at the P₂ position, were among the best P₂' residues. The specificity at P₃ was generally opposite that of P₂. Residues at P₂, and to a lesser extent at P₃, influenced the cleavage site. At the P₂ position, His, Phe, Tyr, Asn, or Trp produced cleavage at the amino side of the first basic residue. In contrast, a P₂ Ile or Val produced cleavage between the dibasic pair. Other residues produced intermediate effects. The pH dependence for substrate binding showed that the enzyme prefers to bind a protonated histidine. A comparison of the effect of arginine or lysine at the P₁' or P₁ position showed that there is a tendency to cleave on the amino side of arginine and that this cleavage produces the highest k_cat values.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Nardilysin (N-arginine dibasic convertase, EC 3.4.24.61) is a 130-kDa metallopeptidase and a member of the relatively newly discovered pitrilysin family of metallopeptidases, also referred to as the inverzincins. An active site zinc binding motif, HXXEH, which is inverted relative to the more common HEXXH motif (1), characterizes this family of metallopeptidases. The geometry and mechanism of peptide cleavage at an inverted zinc-binding site has not been elucidated. Aside from nardilysin, three other members of the pitrilysin family have been identified: insulin-degrading enzyme (EC 3.4.24.56) (2, 3), protease III from E. coli (EC 3.4.24.55) (4), and a recently described human metallopeptidase (5).

Nardilysin is unique in that it contains an acidic region, which, depending on the particular species, is composed of 43 (human) to 59 (mouse) glutamate and aspartate residues within a 76-amino acid stretch. It has been suggested that this acidic domain might play a role in the regulation of nardilysin activity by forming charge-charge complexes with other cellular components (6, 7).

The initial characterization of the specificity of nardilysin led to the conclusion that the enzyme cleaves peptide substrates on the amino side of an arginine residue of paired basic residues (8-10). It has been found that although the enzyme cleaves a number of peptides between paired dibasic residues of the type Arg-Arg (dynorphin A, BAM12 residues 1-8) and Arg-Lys (-neoendorphin), with somatostatin 28 as substrate cleavage occurs on the amino side of the arginine residue in the Arg-Lys dibasic pair (8, 10).

We recently described the use of quenched fluorogenic substrates to measure nardilysin activity (11). Here we report the use of these substrates to study the effect of amino acid substitutions at the P₂, P₂', and P₃ positions on the kinetics of substrate hydrolysis as well as influencing the site of cleavage. The results of this study show that P₂ and P₂' residues have a significant effect on the rate of substrate hydrolysis. The site of cleavage is governed by both the residue in the P₂ position and a preference to cleave on the amino side of arginine. These effects can produce multiple cleavage sites in a substrate.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Peptides were synthesized as described (12) on a Shimadzu PSSM 8 automated solid-phase peptide synthesizer employing Fmoc (9-fluorenylmethoxycarbonyl) methodology. The histidine derivative, N,N-dimethylhistidine (His(Me)₂), in which both of the imidazole nitrogens are methylated, was synthesized as described previously (13). Nova Syn TGR resin (NovaBiochem) was used, and all couplings were performed with O-benzotriazol-1-yl-N,N,N',N'-tetramethyluronium tetrafluoroborate/1-hydroxybenzotriazole using N-methyl morpholine as base.

Separation of Peptides and Cleavage Products-- Peptides were purified by reverse phase chromatography on a Vydac C₄ column using a Waters HPLC¹ system with detection of the peptides by absorbance at 214 nm. The mobile phase was 0.1% trifluoroacetic acid in water/acetonitrile. Unless otherwise indicated, HPLC separations were carried out in a linear gradient from 5 to 50% acetonitrile. For the separation of the cleavage fragments of Abz-GGFHRRHGQ-EDDnp, a concave gradient from 5 to 50% acetonitrile was employed. Peak areas were used to quantitate products. In addition, products were collected, freeze-dried, and analyzed by mass spectrometry. In some cases, products were identified by mass spectrometry directly from reaction mixtures in which ammonium acetate buffer replaced phosphate buffer. When appropriate, products were identified by comparing their retention times on HPLC to authentic standards previously identified by mass spectrometry. Alternatively, peptides were identified either by matrix-assisted laser desorption ionization-time of flight mass spectrometry or by electrospray ionization-Fourier transform-ion cyclotron resonance mass spectrometry at the University of Kentucky Mass Spectrometry Facility. For matrix-assisted laser desorption ionization-mass spectrometry, peptides were prepared in an -cyano-4-hydroxycinnamic acid matrix, which was found to minimize fragmentation of blocked peptides.

Enzyme Purification-- Nardilysin was purified from rat testes as described previously (6, 14).

Fluorometric Assay of Nardilysin-- Monitoring the increase in fluorescence, which occurred upon peptide bond cleavage, was used to follow enzymatic hydrolysis of the fluorogenic peptide substrates. A Hitachi F-2000 fluorescence spectrophotometer connected to a strip chart recorder was utilized with an excitation wavelength of 319 nm and an emission wavelength of 419 nm. Reaction mixtures of 400 µl contained 50 mM potassium phosphate buffer, pH 7.0, and variable fluorogenic peptide substrate. The reaction was initiated by the addition of enzyme. The concentration of the fluorogenic peptides was determined by measuring their absorbance at 357 nm and by determining the total fluorescence change by complete hydrolysis with trypsin (11). Kinetic parameters were calculated using the computer programs of Cleland (15).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

To assess the subsite specificity of rat nardilysin, we utilized a series of fluorogenic peptides containing the sequence Abz-GGX₁X₂RKX₃GQ-EDDnp. Residues at positions X₁, X₂, and X₃ were varied to study effects at the P₃, P₂, and P₂' positions, respectively.² Previous studies showed that a P₃' residue was not required for substrate hydrolysis (10); therefore, amino acid substitutions at this position were not studied. Since the bovine adrenal medulla peptide analog Abz-GGFLRRVGQ-EDDnp has been found to be a good substrate for the enzyme (10) we used phenylalanine as the nonvaried P₃ residue, leucine as the nonvaried P₂ residue, and valine as the nonvaried P₂' residue.

The effect of varying the amino acid at the P₂ position on the kinetics of the nardilysin reaction is shown in Table I. With the exception of a P₂ aspartate, the enzyme accepts a variety of amino acids at this position. In terms of K_m, there is little variation (~3-fold) among the different amino acids at the P₂ position. On the other hand, there is an ~20-fold variation in k_cat. The amino acids at the P₂ position can be grouped on the basis of their influence on k_cat. Histidine and methionine are the preferred P₂ residues, while the hydrophobic amino acids phenylalanine, tyrosine, and leucine constitute a group that produces reasonably good substrates. The next group contains isoleucine, alanine, asparagine, and tryptophan, which are ~ to as reactive as histidine, and finally valine and serine, which exhibit ~ the reactivity of a P₂ histidine.

The range of values for the specificity constant k_cat/K_m parallels k_cat. This reflects the fact that the major difference among the P₂ residues is reflected in k_cat, with little variation in K_m. There is a slight decrease in k_cat/K_m relative to k_cat alone, seen with the hydrophobic amino acids phenylalanine, tyrosine, and leucine. This is the result of the slightly higher K_m values seen with this group of P₂ residues.

Nardilysin has been shown to cleave between paired dibasic residues as well as on the amino side of an arginine as the first basic residue in a dibasic pair (8, 10). We therefore determined the cleavage site for each of the variable P₂ residues. Reverse phase HPLC on a C₄ column was used to separate and quantitate products, while mass spectrometry was used to identify products. As noted in Table I and illustrated in Fig. 1, the P₂ residue has a significant influence on the site of cleavage. When histidine, phenylalanine, tyrosine, asparagine, or tryptophan occupy the P₂ position, cleavage occurs exclusively on the amino side of arginine, which is the first basic residue of the dibasic pair (XRK). In contrast, with isoleucine or valine at the P₂ position, cleavage occurs exclusively at the RK bond (XRK). When methionine and leucine occupied the P₂ position, 20-30% of the cleavage occurred on the amino side of the arginine residue, and 70-80% of the cleavage occurred between the dibasic pair. When alanine and serine occupied the P₂ position, the situation was reversed. In this case, 65-70% of the cleavage occurred on the amino side of the arginine, and 30-35% of the cleavage occurred between the dibasic pair. This is illustrated in Fig. 1 with a P₂ tyrosine, leucine, and alanine. Thus, there is a rather significant influence of the P₂ residue on both the rate of cleavage and the site of cleavage. However, there is no apparent correlation between the rate of cleavage and the site of cleavage.

View larger version (18K): [in this window] [in a new window]   Fig. 1.   HPLC chromatogram of the hydrolysis products generated from Abz-GGFXRKVGQ-EDDnp, where X represents Y, L, or A as variable P2 residues. 30 µM peptide containing either a P2 tyrosine (top), leucine (middle), or alanine (bottom) was reacted with rat nardilysin in 50 mM potassium phosphate buffer, pH 7. The reaction products were separated by gradient HPLC on a Vydac C4 reverse phase column. The mobile phase ranged from 5 to 50% acetonitrile in 0.1% trifluoroacetic acid in water during the course of 50 min as described under "Materials and Methods." Product peaks were identified by comparison with standards previously determined by mass spectrometry.

The effect of the P₂' residue on the kinetics of substrate hydrolysis was next examined with the results given in Table II. As with the P₂ residues, different amino acids at the P₂' position have only small effects on K_m but do affect k_cat. Again histidine was the preferred P₂' amino acid, although in this case the P₂' histidine-containing substrate was cleaved almost 2.5 times faster than the substrate containing a P₂' tryptophan, the second most reactive P₂' residue. The P₂' residues were not as easily grouped as the P₂ residues. The K_m values exhibited by the different P₂' residues varied just over a 2-fold range, while, with the exception of aspartate, k_cat varied ~8-fold. Some residues that were found to produce less reactive substrates when in the P₂ position, such as tryptophan and serine, were among the P₂' residues producing substrates exhibiting high k_cat values. Aspartate, which was not tolerated as a P₂ residue, produced a poor yet reactive substrate as a P₂' residue. As seen with the variable P₂ residues, the values of k_cat/K_m parallel k_cat.

We also examined the effect of the P₂' residue on the site of cleavage of the substrate. As noted above, the presence of a P₂ leucine residue produced 77% cleavage of the Arg-Lys bond and 23% cleavage of the Leu-Arg bond. Most of the P₂' residues had little influence on this cleavage pattern with a few notable exceptions. With a P₂' histidine, 90% of the cleavage occurred at the Arg-Lys bond, and with a P₂' tryptophan 87% of the cleavage occurred at this site.

We also examined the effect of the P₃ position on the kinetics of the reaction. As with variable residues in the P₂ and P₂' positions, with the exception of glutamate and aspartate, the residue occupying the P₃ position has little influence on the substrate K_m (Table III). The specificity at the P₃ position seemed almost the opposite of that of the P₂ position, with a P₃ serine and methionine producing the most reactive substrates and a P₃ phenylalanine and tyrosine among the least reactive substrates. The amino acids occupying the P₃ position had about the same effect as the P₂' residues on k_cat, showing a variation of ~9-fold. The specificity constant k_cat/K_m followed k_cat values as noted with the variable P₂ and P'₂ residues. A number of P₃ residues influenced the cleavage site, with serine, tyrosine, and asparagine favoring cleavage at the Arg-Lys bond and methionine, leucine, and isoleucine favoring cleavage at the Leu-Arg bond.

Further analysis of the data in Tables I and II permits an assignment of the rate-limiting step of the nardilysin reaction with the synthetic peptides to be the actual bond-breaking step. The nardilysin reaction follows Scheme 1, in which a single substrate is converted to two products. The rate-determining step of the reaction could be the bond-breaking step (k₃), the dissociation of the first product P (k₅), or the dissociation of the second product Q (k₇). We can arbitrarily assign the product derived from the C-terminal or EEDnp-containing part of the substrate as P and the product derived from the N-terminal or Abz-containing part of the substrate as Q. The data in Table I show that different k_cat values are obtained when identical products (RKVGQ-EDDnp or KVGQ-EDDnp) are formed from substrates containing variable P₂ residues. If the release of the product P (k₅ in Scheme 1) were rate-limiting, all of these substrates would exhibit the same k_cat. Similarly, as seen in Table II, substrates that release the common products Abz-GGFLR or Abz-GGFL show variable k_cat values. Therefore, release of the product Q (k₇ in Scheme 1) cannot be rate-limiting. The data are thus consistent with the cleavage step (k₃) being the rate-determining step. From the rate equation for a kinetic mechanism involving a single substrate yielding two products, k_cat is defined as k₃k₅k₇/(k₅k₇ + k₄k₇ + k₃k₇ + k₃k₅)/Et. If the bond-breaking step (k₃) is rate-limiting, k_cat reduces to k₃/Et. Similarly, K_m, which is defined as k₇(k₃k₅ + k₂k₅ + k₂k₄)/k₁(k₅k₇ + k₄k₇ + k₃k₇ + k₃k₅), becomes k₂/k_1, the true dissociation constant K_d, when k₃ is rate-limiting.

Since histidine residues appear to be the best P₂ and P₂' residues, we synthesized the potential "ideal" substrate Abz-GGFHRRHGQ-EDDnp and measured its kinetics. This substrate yielded a K_m of 2.2 µM, k_cat of 455 min¹, and a k_cat/K_m of 20.7 × 10⁷ min¹ M¹. This substrate is similar in reactivity to substrates containing either a P₂ or P₂' histidine residue. Abz-GGFHRRHGQ-EDDnp has the potential to carry two additional positive charges at acidic pH. The pK_a values of these two histidines were determined to be ~6.1 and 6.6 by electrometric titration. It was of interest to measure the pH dependence of the cleavage of Abz-GGFHRRHGQ-EDDnp compared with a substrate with nonionizable P₂ and P₂' residues, namely Abz-GGFLRRVGQ-EDDnp. As shown in Fig. 2, the K_m for hydrolysis of Abz-GGFHRRHGQ-EDDnp increased with increasing pH between pH 6.5 and 7.0 and then remained essentially unchanged. We could not go lower than pH 6.5 due to irreversible inactivation of the enzyme (7). The curve shows a pK_a of ~7 and a slope of 1.9. This pK_a reflects the ionization of the histidine in the enzyme substrate complex. The observation that the K_m for Abz-GGFHRRHGQ-EDDnp leveled off at alkaline pH to a value of ~2.5 µM provides an estimate of the K_m for the substrate when both the P₂ and P₂' histidines are in their free base form. It can be estimated from the data in Fig. 2 that the diprotonated form of the substrate bound approximately 70 times better than the free base form. In order to verify these findings, we measured the pH dependence for the hydrolysis of Abz-GGFH(Me)₂RRH(Me)₂GQ-EDDnp in which N,N-dimethylhistidine, which carries a permanent positive charge, was substituted for each of the two histidines. As shown in Fig. 2, the K_m of the hydrolysis of the peptide containing the dimethylhistidines was essentially constant over the pH range studied and exhibited approximately the same K_m as Abz-GGFHRRHGQ-EDDnp at pH 6.5. The K_m for the substrate Abz-GGFLRRVGQ-EDDnp with nonionizable P₂ and P₂' residues was essentially pH-independent over the same pH range (Fig. 2). Although not shown, the pH dependence of V_max for substrates containing ionizable and nonionizable amino acids was essentially the same.

View larger version (13K): [in this window] [in a new window]   Fig. 2.   pH dependence of the Km for the hydrolysis of Abz-GGFHRRHGQ-EDDnp (), Abz-GGFH(Me)2RRH(Me)2GQ-EDDnp (), and Abz-GGFLRRVGQ-EDDnp (). Peptide hydrolysis was conducted in 50 mM potassium phosphate buffer at the indicated pH. H(Me)2 corresponds to a derivative of histidine in which both of the imidazole nitrogens are methylated, producing a permanent positive charge on the imidazole ring.

We also examined the pH dependence for substrates containing either a P₂ or a P₂' histidine. The pH dependence of K_m with these substrates was similar to the pH dependence with the substrates containing histidine in both the P₂ and P₂' positions, except the slope of the lines decreased to 0.94 for a P₂ histidine and 0.90 for a P₂' histidine. This is shown in Fig. 3A for a P₂ histidine and in Fig. 3B for a P₂' histidine. With these substrates, the pK_a of the histidine in the enzyme-substrate complex was also ~7. It can be estimated that the protonated forms of these substrates bound ~10-fold better than the free base form. As shown in Fig. 3, the pH dependence for the parent substrate containing nonionizable P₂ and P₂' residues was pH-independent. Interestingly, although the corresponding substrates containing a single dimethylhistidine exhibited the expected pH-independent profile, the positively charged dimethylhistidine exhibited a K_m value that was intermediate between the free base form and the protonated histidine-containing peptides (Fig. 3).

View larger version (22K): [in this window] [in a new window]   Fig. 3.   pH dependence for the hydrolysis for peptides containing a P2 or P2' His. A, pH dependence of Km for the hydrolysis of Abz-GGFHRKVGQ-EDDnp (), Abz-GGFH(Me)2RKVGQ-EDDnp (), and Abz-GGFLRKVGQ-EDDnp (). Peptide hydrolysis was conducted in 50 mM potassium phosphate buffer at the indicated pH. B, pH dependence of Km for the hydrolysis of Abz-GGFLRKHGQ-EDDnp (), Abz-GGFLRKH(Me)2GQ-EDDnp (), and Abz-GGFLRKVGQ-EDDnp ().

In order to determine whether the presence of a protonated histidine at the P₂ and/or P₂' position would influence the site of cleavage, we examined the products formed at pH 6.5 and pH 8.0 by reverse phase HPLC. At pH 6.5, the substrate Abz-GGFHRRHGQ-EDDnp is cleaved both at the amino side of the dibasic pair (His-Arg bond) as well as between the dibasic pair (Arg-Arg bond), Fig. 4. Based on the area of the product peaks, it can be estimated that both sites are cleaved at approximately the same rate at pH 6.5. At pH 8.0, cleavage at the Arg-Arg bond increases to 66% cleavage. We found that the site of cleavage of the nonionizable substrate Abz-GGFLRRVGQ-EDDnp did not change between pH 6.5 and pH 8. With a permanent positive charge at both the P₂ and P₂' positions, three cleavage sites were observed: between the His(Me)₂-Arg bond, between the Arg-Arg bond, and between the Arg-His(Me)₂ bond.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   HPLC chromatogram of the hydrolysis products generated from Abz-GGFHRRHGQ-EDDnp at pH 6.5 and pH 8.0. Reaction mixtures containing 30 µM Abz-GGFHRRHGQ-EDDnp in 50 mM potassium phosphate buffer, pH 6.5 (top curve) or pH 8.0 (bottom curve) were reacted with nardilysin, and products were analyzed by gradient HPLC as described in the legend to Fig. 1. Peptides were identified as their protonated ions by electrospray ionization-Fourier transform-ion cyclotron resonance mass spectroscopy as follows: RHGQ-EDDnp, m/z 705.313 (Calc. 705.318); Abz-GGFH, m/z 535.9 (Calc. 536.225); RRHGQ-EDDnp, m/z 861.410 (Calc. 861.419); Abz-GGFHR, m/z 692.315 (Calc. 692.326).

We also examined the effect of pH on the cleavage site for the substrates containing a single histidine at either the P₂ or the P₂' site. As noted in Table I, at pH 7 when histidine occupies the P₂ position, cleavage occurs on the amino side of the first arginine residue in the dibasic pair. We found that this pattern did not change when cleavage was conducted at pH 6.5 or at pH 8.0. With a P₂' histidine, there was also no effect of pH on the site of cleavage. Thus, protonation of the histidine per se does not influence the cleavage site. With a dimethylhistidine in the P₂ position, cleavage occurred on the amino side of the first arginine residue of the dibasic pair, similar to what was observed with a His in this position. However, a dimethylhistidine in the P₂' position produced cleavage on the amino side of the dibasic pair (~20% of total cleavage), between the dibasic pair (~50% of total cleavage), and on the carboxyl side of the dibasic pair (~30% of total cleavage). This last cleavage suggests that dimethylhistidine was recognized as a basic residue.

We utilized the synthetic fluorogenic substrates to determine if N-arginine dibasic convertase is truly an arginine-specific enzyme. We thus compared the reactivity of substrates containing arginine or lysine at the P₁' or P₁ position, while at the same time looking at the influence of the P₂ residue. We used P₂ residues that appeared to favor cleavage between the dibasic pair (isoleucine), on the amino side of the dibasic pair (tyrosine) or produced a mixture of these two cleavages (leucine). As shown in Table IV, varying the dibasic pair has little effect on K_m regardless of the P₂ residue. On the other hand, there are noticeable effects of the P₁' and P₁ basic residues on k_cat, k_cat/K_m, and the cleavage site, and these are dependent on the P₂ residue. With a P₂ leucine, the enzyme exhibits the highest k_cat values with arginine at the P₁' site but exhibits a similar reactivity with either arginine or lysine at the P₁ site. As shown in Fig. 5 and summarized in Table IV, cleavage between the dibasic pair represents the major route of hydrolysis. However, cleavage on the amino side of the dibasic pair is seen with arginine occupying the P₁ position. When tyrosine occupies the P₂ position, a P₁ arginine produces the highest k_cat values. Although cleavage is favored on the amino side of the first basic residue of the dibasic pair, the presence of a P₁' arginine produces significant cleavage between the dibasic pair. When isoleucine is in the P₂ position, higher k_cat values are obtained with a P₁' arginine. However, regardless of the P₁' or P₁ basic residue, cleavage occurs exclusively between the dibasic pair.

View larger version (18K): [in this window] [in a new window]   Fig. 5.   HPLC chromatogram of the hydrolysis products generated from Abz-GGFL P1P1'VGQ-EDDnp with variable basic residues. Reaction mixtures containing 30 µM peptide in 50 mM potassium phosphate buffer, pH 7 were incubated with nardilysin and analyzed by gradient HPLC as described in the legend to Fig. 1. Carboxyl-terminal cleavage products were not well resolved in these experiments.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Nardilysin cleaves most physiological and related peptide substrates between paired basic residues (8-10) but can also cleave somatostatin 28 on the amino side of an arginine residue in the dibasic pair (8). This observation prompted us to examine the effects of varying residues in the P₃, P₂, and P₂' positions on the kinetics of cleavage as well as the on site of cleavage. The amino acid in the S₂ subsite seems to have the greatest influence on both k_cat and the site of cleavage, with little effect on K_m. Residues in the S₂' subsite generally exhibit smaller effects on k_cat, have little effect on K_m, and with a few exceptions have little influence on the site of cleavage. The same can be said for residues in the S₃ subsite, except that these have a greater influence on the cleavage site. As noted above, the kinetic K_m appears to reflect the true affinity of the enzyme for its peptide substrate, while k_cat reflects the rate of peptide bond cleavage. There is a general lack of variance of K_m with different amino acids at the P₃, P₂, and P₂' positions but a significant variance of k_cat dependent on the particular amino acid in these positions. This may be explained by the use of the binding energy derived from substrate-enzyme subsite interactions for catalysis. Thus, binding interactions at each subsite contribute to catalysis.

Within the P₂ position, the presence of the aromatic amino acids, histidine, or asparagine favors cleavage on the amino side of the arginine residue, which is the first basic residue of the dibasic pair. In contrast, the bulky aliphatic amino acids valine and isoleucine induce cleavage exclusively between the dibasic pair. Other P₂ residues produce partial effects. The small aliphatic side chains of alanine and serine cause a more favorable cleavage at the amino side of arginine, while the larger methionine and leucine tend to favor cleavage between the paired basic residues. Since the mechanism of bond cleavage is considered to be independent of the residues within the peptide substrate, the positioning of the substrate relative to the catalytic site will determine which bonds are broken. Two mechanisms can therefore be envisioned that account for changes in the bond being cleaved dependent on residues in the putative P₂ position. In the first mechanism, the residues of the substrate interact with enzyme subsites in a variable manner. For example, aromatic amino acids fit into the S₁ subsite rather than the expected S₂ subsite, causing the first basic residue of the dibasic pair to occupy the S₁' subsite. The second basic residue of the dibasic pair becomes relatively unimportant, since it becomes a P₂' residue. Different amino acids in the P₂ position of the substrate can bind in this mode to varying degrees and can also bind in a more "normal" manner whereby the basic residues in the dibasic pair occupy the S₁ and S₁' subsites. It would appear that the rate of substrate cleavage is not significantly influenced by the mode of binding. Substrates that, according to this postulate, bind with the nonbasic residue in the S₁ subsite show a wide range of k_cat values. They vary from among the highest k_cat values for His- and Phe-containing substrates to among the lowest for the Trp-containing substrate. Evidence that favors this hypothesis comes from the recent observation that nardilysin cleaves -endorphin 1-31 at a single basic residue between a Phe¹⁸-Lys¹⁹ bond within the sequence Leu¹⁷-Phe¹⁸-Lys¹⁹-Asn²⁰.³ Camargo et al. (17) noted a shift in the cleavage site of peptides by thimet oligopeptidase (endopeptidase 24.15) dependent on which residue occupied a C-terminal position. They favored a similar mechanism involving two distinct modes of substrate binding.

An alternative explanation is that the binding of aromatic amino acids, histidine, or asparagine within the enzyme subsites is the same as with any other substrate. However, these residues distort the active site in such a way as to place catalytic residues in a conformation that favors cleavage on the amino side of the first basic residue in the dibasic pair.

When His or Trp occupy the P₂' position of the substrate, cleavage between the dibasic pair is favored, and the highest k_cat values among the variable P₂' residues are observed. This would suggest that these two amino acids produce the most favorable interactions within the S₂' subsite.

The presence of a His as either the P₂ or P₂' residue produces high k_cat values and favors cleavage on the amino side of the arginine residue of the dibasic pair. Protonation of the His at either position increases the affinity more than 10-fold, with no effect on k_cat. However, the protonated His does not appear to be recognized as a "basic residue," since the cleavage site does not change between a protonated His and a free base His. This can be contrasted to the presence of N,N-dimethylhistidine as the P₂' residue in which it appears this residue was recognized as a basic residue.

We also determined the influence of arginine versus lysine as the P₁ or P₁' residue. Regardless of the P₂ residue, higher k_cat values were always obtained when cleavage occurred on the amino side of an arginine residue. This was seen regardless of whether cleavage occurred between the dibasic pair or on the amino side of the first basic residue of the dibasic pair. In terms of the actual cleavage site, there appear to be two forces at work. On the one hand, the P₂ residue influences whether cleavage occurs between the dibasic pair or on the amino side of the first basic residue of the dibasic pair. On the other hand, the enzyme exhibits a preference for cleavage on the amino side of arginine. The relative contribution of these two forces determines at which site a substrate is cleaved and can produce multiple cleavage sites within the same substrate.

In summary, the cleavage of peptide substrates by nardilysin is influenced not only by the presence of a dibasic pair but also by residues occupying adjacent subsites on the enzyme. The elucidation of a number of these effects provides the first step toward understanding the complex cleavage pattern of this and similar peptidases.

	ACKNOWLEDGEMENT

We thank Dr. W. W. Cleland for helpful discussions of the kinetic data.

	FOOTNOTES

^* This work was supported in part by National Institute on Drug Abuse Grant DA02243 (to L. B. H.) and Grant DA11987 (to E. C.). This work was also supported in Brazil by Fundação de Amparo Pesquisa do Estado de São Paulo and Conselho Nacional de Desenvolvimento Científico e Tecnológico.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

These authors contributed equally to this work.

To whom correspondence should be addressed: Dept. of Biochemistry, University of Kentucky, 800 Rose St., Lexington, KY 40563-0298. Tel.: 859-323-5549; Fax: 859-323-1727; E-mail: lhersh@pop.uky.edu.

Published, JBC Papers in Press, April 11, 2000, DOI 10.1074/jbc.M909020199

² The first arginine or lysine in the peptide sequence is designated the P₁ residue and the second arginine or lysine as the P₁' residue. In most cases this conforms to the nomenclature of Schechter and Berger (16), however in some peptides either there are multiple cleavage sites or the cleavage site is shifted.

³ O. Oakley and L. B. Hersh, manuscript in preparation.

	ABBREVIATIONS

The abbreviations used are: HPLC, high pressure liquid chromatography; Abz, 2-aminobenzoyl; EDDnp, ethylenediamine-2,4-dinitrophenyl.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				G. Fontes, A.-D. Lajoix, F. Bergeron, S. Cadel, A. Prat, T. Foulon, R. Gross, S. Dalle, D. Le-Nguyen, F. Tribillac, et al. Miniglucagon (MG)-Generating Endopeptidase, which Processes Glucagon into MG, Is Composed of N-Arginine Dibasic Convertase and Aminopeptidase B Endocrinology, February 1, 2005; 146(2): 702 - 712. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 275/26/19545    most recent M909020199v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chow, K. M. Articles by Hersh, L. B. Search for Related Content PubMed PubMed Citation Articles by Chow, K. M. Articles by Hersh, L. B. Social Bookmarking                      What's this?