- Email: [email protected]
Studies on the subsite specificity of the rat brain puromycin-sensitive aminopeptidase
ARCHIVES
OF BIOCHEMISTRY
Vol. 276, No. 2, February
AND
BIOPHYSICS
1, pp. 305-309,199O
Studies on the Subsite Specificity of the Rat Brain Puromycin-Sensitive Aminopeptidase’ Gary D. Johnson and Louis B. Hersh’ Department
of Biochemistry,
University
of Texas Southwestern
Medical Center, Dallas, Texas 75235
Received April 17,1989, and in revised form August 18,1989
The specificity of the puromycin-sensitive aminopeptidase from rat brain was examined. Using L-alanyl/3-naphthylamide as substrate V,,,,, of the reaction was shown to be pH independent over the range of 5.5-9.0, while K, exhibited a pK, of 7.7. This latter value corresponds to the pKa of the amino group of the substrate. Using X-Ala and X-Leu to examine the specificity of the Pl site it was found that Arg and Lys exhibit the highest affinity, followed by Met, Val, Leu, Trp, and Phe, which bind -5- to ZO-fold less well. Although K,,, varied more than ZO-fold within this series, less variation. Significantly V max showed considerably weaker binding was observed with a Pl Gly, Ala, Ser, or Pro with no binding detectable with a Pl Glu. The presence of a P’l Leu compared to P’l Ala results in an approximate lo-fold decrease in Km with little change P’l residues was examin V,,,. The effect of varying ined with the series Leu-X. In this case basic and hydrophobic amino acids, with the exception of Val, all exhibit nearly the same K,. The binding of Arg-Arg and Lys-Lys showed the same K, as obtained for ArgLeu or Lys-Leu, respectively. When Leu-Ser-Phe was compared to Leu-Ser the P’2 residue led to a loo-fold decrease in Km and slightly less than a 5-fold increase the addition of a P’2 Met to Leuin V,,,. In contrast Trp results in only a 3-fold decrease in K,,, and a 3-fold increase in V,,, . The results indicate a preference for a basic or hydrophobic residue in the Pl and P’l sites and indicate subsite-subsite interactions which prio lsso Academic press, h. marily affect binding.
The arylamidases represent a subclass of aminopeptidases which are characterized by their ability to hydro’ This work was supported in part by National Institute on Drug Abuse Grant DA02243 and Welch Foundation Grant 1391. ’ TO whom correspondence should be addressed at Department of Biochemistry, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas, TX 75235 0003.9861/90
$3.00
Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
lyze amino acid naphthylamides in addition to peptide substrates (1). Among the mammalian representatives of this group of metallo exopeptidases are a glutamate/ aspartate specific enzyme, aminopeptidase A, EC 3.4.11.7 (2-4), a lysine/arginine specific enzyme, EC 3.4.11.6 (5), which is probably the same enzyme as aminopeptidase MI described by Hersh (6), and a group of aminopeptidases exhibiting a broad substrate specificity which include leucine aminopeptidase, EC 3.4.11.1 (7), aminopeptidase M, EC 3.4.11.2 (8), and a more recently described puromycin-sensitive aminopeptidase, EC 3.4.11.- (5,9-14). These enzymes have received renewed attention because of their possible involvement in the degradation of a variety of physiologically active peptides including the enkephalins (15, 16), dynorphins (17), angiotensins I and II (l&19), and oxytocin (20), to name but a few. Of the group of aminopeptidases exhibiting a broad substrate specificity, the puromycin-sensitive aminopeptidase is the least well characterized. This enzyme can be distinguished from other aminopeptidases on the basis of its sensitivity to puromycin as an inhibitor (K, N 1 FM) (21-23) whereas other aminopeptidases such as aminopeptidase M are inhibited by puromycin in the range of 100-1000 PM (24,25). This puromycin-sensitive aminopeptidase has been described as a brain enkephalin-degrading aminopeptidase and has been referred to as “aminoenkephalinase” (26) or “the enkephalin-degrading aminopeptidase” (27). However, it has recently been shown to exhibit a broad tissue distribution, being the major arylamidase in brain, heart, and skeletal muscle (28). In addition cytosolic and membrane-associated forms of the enzyme have been described. The membrane enzyme form appears to be identical to the cytosolic enzyme, but the nature of its association with membranes is unclear (S. Dyer et al., manuscript in preparation). The puromycin-sensitive aminopeptidase has been purified from a variety of sources; however, its specificity has only been examined with amino acid-fl-naphthyl305
306
JOHNSON
AND
HERSH
amides or nitroanilides as substrates (6, 11-13, 29) and amino acid-hydroxamates as inhibitors (27). Although peptides have been tested as substrates or inhibitors for the enzyme (14, 19, 26, 30) a systematic series has not been examined in terms of both binding and turnover. We present here a study of the specificity of the enzyme using dipeptides as substrates and inhibitors. MATERIALS
AND
5
METHODS
Determination of leucine and alan.ine. Leucine and alanine were determined by a modification of the method of Williamson et al. (31) in which the amino acid is converted to the corresponding keto acid with the concomitant reduction of NAD to NADH. Reaction mixtures contained 167 mM Tris buffer, pH 10.2, 0.5 mM EDTA, 0.8 mg/ml NAD, 20 munits of Bacillus sp. leucine dehydrogenase or 25 munit of Bacillus subtilis alanine dehydrogenase, and the unknown sample in a total volume of 0.3 ml. Samples were incubated for 1 h at 37°C and then the NADH formed was measured using an Aminco-Bowman spectrofluorometer at an excitation wavelength of 340 nm and an emission wavelength of 460 nm. Standard curves were included with each kinetic determination. The dehydrogenases were obtained from Sigma Chemical Co. Determination of K,,, or K,. K, was determined by using dipeptides or tripeptides as alternate substrate inhibitors of L-Ala-fl-naphthylamide hydrolysis. Reaction mixtures contained 100 mM bis-Tris-propane buffer, pH 7.5,10 FM L-Ala-P-naphthylamide (K,,, = 100 KM), and variable concentrations ofpeptide in a final volume of 0.25 ml. At least six inhibitor concentrations were utilized over the range of 0.3 to 3.0 times the K,,,. The rate of fl-naphthylamine appearance was measured at 37°C with an Aminco-Bowman spectrofluorometer connected to a strip chart recorder. The reaction was monitored at an excitation wavelength of 335 nm and an emission wavelength of 410 nm. Free pnaphthylamine was used to quantitate the reaction. Data were analyzed by using the Fortran computer programs of Cleland (32). pH dependence of V,,,,, and K,,. The pH dependence of k,,, and K,, was determined using L-alanine-P-naphthylamide as a variable substrate and 0.1 M bis-Tris-propane buffer over the pH range of 5.5 to 9.0. The reaction was monitored by following the liberation of free @naphthylamine as described above. The fluorescence of free /%naphthylamine was found to be pH independent over the pH range of 5.5 to 9.0. Determination of V,,,. Kinetic runs were conducted at 37°C by incubating enzyme with the dipeptide substrate at 10 times its K,,, in 50 mM bis-Tris-propane buffer, pH 7.5. At various times an aliquot was removed, the reaction stopped by boiling, and alanine or leucine release determined with the appropriate amino acid dehydrogenase as described. In all cases the reaction was shown to be linear with respect to both time and enzyme concentration. Doubling the substrate concentration did not significantly increase or decrease the observed rate, indicating saturation of the enzyme and the absence of substrate inhibition. Preparation of the puromycin-sensitive aminopeptidase. The cytosolic form of the puromycin-sensitive aminopeptidase was purified from rat brain as previously described (29). Briefly this procedure involves acid precipitation of contaminating proteins from a high speed supernatant of a tissue homogenate, ammonium sulfate precipitation (40-70s ammonium sulfate) of the reneutralized supernatant, molecular-sieve chromatography on a column of AcA 34, and ion-exchange chromatography on a FPLC mono Q column. The purified enzyme exhibited one major and several minor bands upon analysis by SDS PAGE.” The major band reacted with an antisera to the enzyme as judged by Western blot analysis.
,’ Abbreviations used: SDS-PAGE, sodium dodecyl acrylamide gel electrophoresis; @-NA, P-naphthylamide.
sulfate-poly-
5
6
7
8
9
10
PH FIG. 1. pH dependence of aminopeptidase
hydrolysis of ~-Ala+ naphthylamide. Reaction mixtures contained 0.1 M bis-Tris-propane buffer at the indicated pH, L-Ala-fl-naphthylamide (varied from 25 to 200 PM at pH’s 5.5 to 7.5 and 40 to 800 PM at pH’s 8.0 to 9.0), and enzyme in a final volume of 200 ~1. The reaction was monitored by following the increase in free P-naphthylamine fluorescence as described under Materials and Methods.
RESULTS
Prior to studies on the subsite specificity of the puromycin-sensitive aminopeptidase we measured the pH dependence of V,,,,, and Km using L-alanine-p-naphthylamide as substrate. As shown in Fig. 1, V,,, appears pH independent over the range of 5.5 to 9.0; however, K,,, and thus V,,,,,/K,,, exhibit a kinetic pK,, of 7.7. Potentiometric titration of L-alanine-P-naphthylamide under the same conditions revealed that the pK, of its free amino group is 7.7. Thus the reactive form of the substrate contains a protonated amino group, and the enzyme itself does not exhibit any ionizations which can be observed kinetically over the pH range studied. We next examined the effect of varying the N-terminal amino acid (Pl residue) on V,,, and Km using two series of dipeptides: X-Ala and X-Leu. In order to measure dipeptide hydrolysis we made use of the alanine and leucine dehydrogenase reactions in which the free amino acid liberated from the dipeptide is oxidized to the corresponding keto acid in an NAD-dependent reaction. The NADH thus formed is measured fluorometrically. The dehydrogenase reactions require rather alkaline pH conditions (pH optima around lo), therefore the reaction was conducted as a discontinuous assay: first generation of free amino acid by the aminopeptidase and then the determination of the free amino acid with the dehydrolarge genase. Since these assays require relatively amounts of the dehydrogenases we chose to measure the Km for each dipeptide by using it as an alternate substrate inhibitor of L-alanine-P-naphthylamide hydroly-
SPECIFICITY
OF RAT
BRAIN
PUROMYCIN-SENSITIVE TABLE
Comparison
of X-Ala
and X-Leu
I
Dipeptides
as Substrates
and Inhibitors X-Leu
X-Ala
X Ax LYS Met Leu Val Trp Phe Ala Ser Pro GUY
Glu
(:;, 0.06 0.17 0.74 0.81 1.25 1.58 2.09 14.7 18.1 >lOO >50
-t
V lmdr (units/mg)
SE 0.002 0.02 0.07 0.07 0.09 0.09 0.19 2.1 1.9
3.40 3.29 2.51 2.22 0.32 1.54 5.78
at 10
V ll,RX (units/mg)
K Vmx /Km
(mM)
56.7 19.4 3.4 2.7 0.3 1.0 2.8
SE
t
0.006 0.013 0.273 0.062 0.22 0.17 0.43 8.9 1.5 13.5
No inhibition
307
AMINOPEPTIDASE
mM
0.0008 0.0002 0.063 0.009 0.02 0.005 0.09 1.0 0.2 No inhibition 3.4 No inhibition
Vm /Km
1.75 4.03 1.68 4.65” 1.00 3.24 8.26
291.7 310.0 6.2 75.0 4.5 19.1 19.2
0.46
0.3
at 5 mM at 10 mM
Note. K,, was determined using the peptide as an inhibitor of L-Ala-P-NA hydrolysis. The reaction was measured at pH 7.5 in 100 mM bisTris-propane buffer with 10 pM l.-Ala-@-NA as substrate. V,,,, was determined by measuring the rate of product formation at a concentration of substrate 10X K,,. One unit corresponds to the formation of 1 pmol of product per minute. ” The observed rate was divided in half since two leucines were generated.
sis. Preliminary experiments demonstrated that the dipeptides were competitive inhibitors. Using this strategy the K,,, for each dipeptide was measured and then V,,, was determined by following the liberation of free alanine or free leucine at a saturating substrate concentration. The data are summarized in Table I. With both series of dipeptides the highest V,,,,/K, values are produced with arginine and lysine as the Nterminal amino acid. Amino acids containing hydrophobic side chains all exhibit similar Km and V,,,, values with the exception of valine, which exhibits a decreased V,,,,. Alanine, serine, and glycine have considerably lower Km values than do the hydrophobic amino acids. The effect of substituents in the P’l position of a dipeptide on hydrolysis of a Pl alanine was determined, with the data presented in Table II. This subsite also favors an arginine residue but binds hydrophobic residues, except valine, with a similar affinity. By comparing the data in Table I to that in Table II it can be seen that the affinities for alanine, serine, and glycine are considerably higher in the S’l site than in the Sl site. Again the predominant effect is one on Km rather than on V,,,. For example, a P’l glycine which binds some 60-fold weaker than arginine or tryptophan reacts =1.5- and e3.5fold slower than these amino acids, respectively. Since basic amino acids appeared to be the preferential amino acids for both the Sl and S’l subsites the K, for Lys-Lys and Arg-Arg was determined. As shown in Table III the K, values for these dipeptides are the same as observed with a P’l leucine. In order to determine whether the active site extends to a P’2 residue we examined Leu-Ser-Phe and Leu-
Trp-Met as substrates. As shown in Table III, a P’2 phenylalanine residue reduces K,,, by a factor of over 100 compared to Leu-Ser, while a P’2 methionine reduced Km by only threefold relative to Leu-Trp. For these substrates V,,, was increased five- and threefold, respectively. DISCUSSION
Previous studies on the specificity of the puromycinsensitive aminopeptidase have been conducted with TABLE
II
Comparison of Leu-X Dipeptides at Substrates and Inhibitors
X Arg Met Leu Val Tw Phe Ala Pro Asn GUY Tyr Ser
K,,,
(mM)
0.03 0.04 0.06 0.37 0.04 0.03 0.81 >lOO 1.77 1.77 0.03 0.28
k
SE
V m.ax (units/mg)
Vrnax/Km
0.0076 0.009 0.009 0.050 0.004 0.002 0.07
2.77 8.55 4.65” 7.07 6.59 3.06 0.73*
126 214 75 19 169 4 1
0.14 0.08 0.0008 0.07
3.56 1.94
2 1
0.73
3
Note. K,, and V,,,, were determined as described in Table I. ” The observed rate was divided in half since two leucines were generated. ‘I The same rate was observed whether alanine formation or leucine formation was measured.
308
JOHNSON TABLE
III
Comparison of Dibasic Dipeptides and Tripeptides as Inhibitors and Substrates Dipeptide
&orK,(mM)
Lys-Lys
Arg-Arg Leu-Ser-Phe Leu-Trp-Met Note.
Ser-Trp
The K, for Lys-Lys and Leu-Trp-Met
0.012 0.007 0.002 0.014
f
SE
V max
0.0016 0.0009 0.0005 0.004
3.42 18.00
and Arg-Arg and the K, and V,,,,, for Leuwere determined as described in Table I.
amino acid+-naphthylamides as substrates and thus only the specificity of the Sl site could be investigated. In addition the influence of P’l and P’2 residues on substrate binding and catalysis could not be determined. These problems have been circumvented by the use of di- and tripeptides as substrates and alternate substrate inhibitors of the enzyme. The data in Tables I and II show that there is subsite specificity for both the Sl and s’l subsites. Arginine and lysine show the highest affinity as the Pl amino acid. Hydrophobic amino acids all exhibit similar affinities, which are somewhat lower than Arg and Lys. This is similar to the results obtained with amino acid-/I-naphthylamides (29); however, the K,,, values for the amino acid-P-naphthylamides are significantly higher than those for dipeptides. The amino acids with small side chains, Ala, Ser, and Gly, exhibit a low affinity for the enzyme. Although the binding affinities vary over more than 2 orders of magnitude there is a considerably smaller variation in V,,,. Thus in terms of the Sl subsite the specificity of the enzyme is manifested almost solely in K,,,. The finding that arginine and lysine bind preferentially to this subsite suggests either that an anionic enzyme residue may be in the vicinity of this subsite or that there is an interaction with a backbone carbonyl. As shown in Fig. 1, the pH dependency of the reaction indicates that only the protonated form of the substrate is a reactant. Thus an anionic residue appears to be involved in binding the free amino group of the substrate. The possibility might be considered that this anionic group interacts with both the positive charge on the amino acid side chain and the amino terminus of the substrate. This is further strengthened by the observation that glutamate does not bind at this site. In addition to the ionic interaction at the Sl subsite it is apparent from Table I that at this binding site hydrophobic interactions can exist. It is also evident when comparing Leu and Ala that the amino acid in the s’l subsite can influence the strength of binding at the Sl subsite. This effect results in a 5- to lo-fold decrease in K,,, for the amino acid in the S’l subsite. Thus there appears to be a synergistic effect
AND
HERSH
between subsites which affects substrate binding. Based on the data in Table II it would appear that the S’l subsite shows considerably less variation in affinity for different amino acids than does the Sl subsite. This subsite does not show a preference for cationic side chains and thus probably utilizes hydrophobic interactions as the predominant binding interaction. However, this subsite can accommodate a positively charged side chain. The data in Table III show that the aminopeptidase active site extends at least to a S’2 subsite. It can be seen that the effect of the P’2 residue on substrate affinity and V max depends on the affinity of the parent dipeptide. Leu-Ser binds with a relatively weak affinity, Ki = 250 PM. Placement of a Phe in the S’2 subsite reduces Ki lOOfold to 2 PM. In contrast Leu-Trp binds with relatively high affinity, K, = 30 PM, and the addition of a Met residue reduces K, only 3-fold to 10 FM. This again indicates synergism between subsites which serves to maximize substrate binding. With both tripeptides V,,, was increased approximately 3-fold. Since arginine is the preferred amino acid in the Sl subsite and shows a high affinity for the $71 subsite it might be anticipated that placing arginine in both of these subsites would facilitate binding. However, when Arg-Arg is compared to Arg-Leu no difference in Km is observed. This is consistent with the hypothesis that subsite-subsite interactions maximize substrate binding. Hui et al. studied Arg-X and Lys-X (X = Arg, Phe, or Tyr) dipeptides as inhibitors of enkephalin hydrolysis and reported essentially no differences in their IC,, values, which were all approximately 10 PM (30). They reported that extending the dipeptide to a tripeptide in no case decreased K,,, and in several cases actually increased K,,,. In a separate study using enkephalin-related peptides as inhibitors of enkephalin hydrolysis Hui et al. (26) observed effects on I& among different P’4 residues and by extending the inhibitor to include a P’5 residue. However, in most cases the effects observed were small and unfortunately the effects on catalysis were not assessed. These results do suggest the presence of an extended active site beyond the S’2 position. However, it has been established that there is a size limitation for substrates for the enzyme which is presumably somewhere between 8 and 13 amino acid residues (33). REFERENCES 1. Patterson, E. K., Hsiao, S.-H., and Keppel, A. (1963) J. Biol. Chem. 238,3611-3620. 2. Danielson, E. M., Noren, O., Sjostrom, H., Ingram, J., and Kenny, A. J. (1980) Biochem. J. 189,591-603. 3. Tobe, H., Kojima, F., Aoyagi, T., and Umezawa, H. (1980) Biochim. Biophys. Acta 613,459-468. 4. Glenner, G. G., McMillan, P. J., and Folk, J. E. (1962) Nature (London) 194,867. 5. Hopsu, V. K., Makinen, K. K., and Glenner, G. G. (1966) Arch. Biochem. Biophys. 114,557-566.
SPECIFICITY
OF RAT
BRAIN
PUROMYCIN-SENSITIVE
6. Hersh, L. B. (1981) Biochemistry 20,2345-2350. 7. Smith, E. L., and Spackman, D. H. (1955) J. Biol. Chem. 212, 271-299. 8. Wachsmuth, D., Fritze, I., and Pfleiderer, G. (1966) Biochemistry t&169-174. 9. Brecker, A. S., and Suszkiw, J. B. (1969) Biochem. J. 112, 335-
342. 10. Suszkiw, J. B., and Brecher,
A. S. (1970) Biochemistry
9, 4008-
4017. 11. Hayashi, M., and Oshima, K. (1977) J. Biochem. 81,631-639. 12. Schnebli, H. P., Phillipps, M. A., and Barclay, R. K. (1979) Biochim. Biophys. Acta 669,89-98. 13. Hersh, L. B., and McKelvy, J. F. (1981) J. Neurochem. 36, 171178. 14. Hui, K. S., Wang, Y.-J., and Lajtha, A. (1983) Biochemistry 22, 1062-1067. 15. Hambrook, J. M., Morgan, B. A., Rance, M. J., and Smith, C. F. C. (1976) Nature (London) 262,782-783. 16. Meek, J. L., Yang, H. Y. T., and Costa, E. (1977) Neuropharmacol-
ogy16,151-154. 17. Ulrich, C., and Hersh, L. B. (1985) Peptides 6,475-482. 18. Bausback, H. H., Churchill, L., and Ward, P. E. (1988) Biochem. Pharmacol. 37,155-160. 19. Hayaishi, M. (1978) J. Biochem. 84,1363-1372.
AMINOPEPTIDASE
20. Ryden, G. (1966) Acta O&et.
309
Gynecol. Scar& 45, Suppl. 3,1-105.
21. Ellis, S., and Perry, M. (1964) Biochem. Biophys.
Res. Commun.
X,502-505. 22. Marks, N., Datta, R. K., and Lajtha, A. (1968) J. Biol. Chem. 243, 2882-2889. 23. Hersh, L. B. (1981) J. Neurochem. 36,1594-1596. 24. Gros, C., Giros, B., and Schwartz, J.-C. (1985) Biochemistry 24, 2179-2185. 25. Hersh, L. B., AboukHair, N., and Watson, S. (1987) Peptides 8,
523-532. 26. Hui, K.-S., Hui, M., Banay-Schwartz,
M., DeGuzman, T., Ling, N., and Lajtha, A. (1983) Peptides 4,639-646. 27. Coletti-Previero, M.-A., de Paulet, C., Mattras, H., and Previero, A. (1982) Biochem. Biophys. Res. Commun. 107,465-469. 28. McClellen, S., Dyer, S., Rodriguez, G., and Hersh, L. B. (1988) J. Neurochem. 61,1552-1559. 29. Hersh, L. B. (1985) J. Neurochem. 44,1427-1435. 30. Hui, K.-S., Hui, M. P.-P., Ling, N., and Lajtha, A. (1985) Life Sci.
36,2309-2315. 31. Williamson, D. H., Lopez-Vieira, O., and Walker, B. (1967) Biochem.J. 104,497-502. 32. Cleland, W. W. (1979) in Methods in Enzymology (Purich, D. L., Ed.), Vol. 63, pp. 103-137, Academic Press, San Diego. 33. Berg, M. J., and Marks, N. (1984) J. Neurosci. Res. 11,313-321.
OF BIOCHEMISTRY
Vol. 276, No. 2, February
AND
BIOPHYSICS
1, pp. 305-309,199O
Studies on the Subsite Specificity of the Rat Brain Puromycin-Sensitive Aminopeptidase’ Gary D. Johnson and Louis B. Hersh’ Department
of Biochemistry,
University
of Texas Southwestern
Medical Center, Dallas, Texas 75235
Received April 17,1989, and in revised form August 18,1989
The specificity of the puromycin-sensitive aminopeptidase from rat brain was examined. Using L-alanyl/3-naphthylamide as substrate V,,,,, of the reaction was shown to be pH independent over the range of 5.5-9.0, while K, exhibited a pK, of 7.7. This latter value corresponds to the pKa of the amino group of the substrate. Using X-Ala and X-Leu to examine the specificity of the Pl site it was found that Arg and Lys exhibit the highest affinity, followed by Met, Val, Leu, Trp, and Phe, which bind -5- to ZO-fold less well. Although K,,, varied more than ZO-fold within this series, less variation. Significantly V max showed considerably weaker binding was observed with a Pl Gly, Ala, Ser, or Pro with no binding detectable with a Pl Glu. The presence of a P’l Leu compared to P’l Ala results in an approximate lo-fold decrease in Km with little change P’l residues was examin V,,,. The effect of varying ined with the series Leu-X. In this case basic and hydrophobic amino acids, with the exception of Val, all exhibit nearly the same K,. The binding of Arg-Arg and Lys-Lys showed the same K, as obtained for ArgLeu or Lys-Leu, respectively. When Leu-Ser-Phe was compared to Leu-Ser the P’2 residue led to a loo-fold decrease in Km and slightly less than a 5-fold increase the addition of a P’2 Met to Leuin V,,,. In contrast Trp results in only a 3-fold decrease in K,,, and a 3-fold increase in V,,, . The results indicate a preference for a basic or hydrophobic residue in the Pl and P’l sites and indicate subsite-subsite interactions which prio lsso Academic press, h. marily affect binding.
The arylamidases represent a subclass of aminopeptidases which are characterized by their ability to hydro’ This work was supported in part by National Institute on Drug Abuse Grant DA02243 and Welch Foundation Grant 1391. ’ TO whom correspondence should be addressed at Department of Biochemistry, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas, TX 75235 0003.9861/90
$3.00
Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
lyze amino acid naphthylamides in addition to peptide substrates (1). Among the mammalian representatives of this group of metallo exopeptidases are a glutamate/ aspartate specific enzyme, aminopeptidase A, EC 3.4.11.7 (2-4), a lysine/arginine specific enzyme, EC 3.4.11.6 (5), which is probably the same enzyme as aminopeptidase MI described by Hersh (6), and a group of aminopeptidases exhibiting a broad substrate specificity which include leucine aminopeptidase, EC 3.4.11.1 (7), aminopeptidase M, EC 3.4.11.2 (8), and a more recently described puromycin-sensitive aminopeptidase, EC 3.4.11.- (5,9-14). These enzymes have received renewed attention because of their possible involvement in the degradation of a variety of physiologically active peptides including the enkephalins (15, 16), dynorphins (17), angiotensins I and II (l&19), and oxytocin (20), to name but a few. Of the group of aminopeptidases exhibiting a broad substrate specificity, the puromycin-sensitive aminopeptidase is the least well characterized. This enzyme can be distinguished from other aminopeptidases on the basis of its sensitivity to puromycin as an inhibitor (K, N 1 FM) (21-23) whereas other aminopeptidases such as aminopeptidase M are inhibited by puromycin in the range of 100-1000 PM (24,25). This puromycin-sensitive aminopeptidase has been described as a brain enkephalin-degrading aminopeptidase and has been referred to as “aminoenkephalinase” (26) or “the enkephalin-degrading aminopeptidase” (27). However, it has recently been shown to exhibit a broad tissue distribution, being the major arylamidase in brain, heart, and skeletal muscle (28). In addition cytosolic and membrane-associated forms of the enzyme have been described. The membrane enzyme form appears to be identical to the cytosolic enzyme, but the nature of its association with membranes is unclear (S. Dyer et al., manuscript in preparation). The puromycin-sensitive aminopeptidase has been purified from a variety of sources; however, its specificity has only been examined with amino acid-fl-naphthyl305
306
JOHNSON
AND
HERSH
amides or nitroanilides as substrates (6, 11-13, 29) and amino acid-hydroxamates as inhibitors (27). Although peptides have been tested as substrates or inhibitors for the enzyme (14, 19, 26, 30) a systematic series has not been examined in terms of both binding and turnover. We present here a study of the specificity of the enzyme using dipeptides as substrates and inhibitors. MATERIALS
AND
5
METHODS
Determination of leucine and alan.ine. Leucine and alanine were determined by a modification of the method of Williamson et al. (31) in which the amino acid is converted to the corresponding keto acid with the concomitant reduction of NAD to NADH. Reaction mixtures contained 167 mM Tris buffer, pH 10.2, 0.5 mM EDTA, 0.8 mg/ml NAD, 20 munits of Bacillus sp. leucine dehydrogenase or 25 munit of Bacillus subtilis alanine dehydrogenase, and the unknown sample in a total volume of 0.3 ml. Samples were incubated for 1 h at 37°C and then the NADH formed was measured using an Aminco-Bowman spectrofluorometer at an excitation wavelength of 340 nm and an emission wavelength of 460 nm. Standard curves were included with each kinetic determination. The dehydrogenases were obtained from Sigma Chemical Co. Determination of K,,, or K,. K, was determined by using dipeptides or tripeptides as alternate substrate inhibitors of L-Ala-fl-naphthylamide hydrolysis. Reaction mixtures contained 100 mM bis-Tris-propane buffer, pH 7.5,10 FM L-Ala-P-naphthylamide (K,,, = 100 KM), and variable concentrations ofpeptide in a final volume of 0.25 ml. At least six inhibitor concentrations were utilized over the range of 0.3 to 3.0 times the K,,,. The rate of fl-naphthylamine appearance was measured at 37°C with an Aminco-Bowman spectrofluorometer connected to a strip chart recorder. The reaction was monitored at an excitation wavelength of 335 nm and an emission wavelength of 410 nm. Free pnaphthylamine was used to quantitate the reaction. Data were analyzed by using the Fortran computer programs of Cleland (32). pH dependence of V,,,,, and K,,. The pH dependence of k,,, and K,, was determined using L-alanine-P-naphthylamide as a variable substrate and 0.1 M bis-Tris-propane buffer over the pH range of 5.5 to 9.0. The reaction was monitored by following the liberation of free @naphthylamine as described above. The fluorescence of free /%naphthylamine was found to be pH independent over the pH range of 5.5 to 9.0. Determination of V,,,. Kinetic runs were conducted at 37°C by incubating enzyme with the dipeptide substrate at 10 times its K,,, in 50 mM bis-Tris-propane buffer, pH 7.5. At various times an aliquot was removed, the reaction stopped by boiling, and alanine or leucine release determined with the appropriate amino acid dehydrogenase as described. In all cases the reaction was shown to be linear with respect to both time and enzyme concentration. Doubling the substrate concentration did not significantly increase or decrease the observed rate, indicating saturation of the enzyme and the absence of substrate inhibition. Preparation of the puromycin-sensitive aminopeptidase. The cytosolic form of the puromycin-sensitive aminopeptidase was purified from rat brain as previously described (29). Briefly this procedure involves acid precipitation of contaminating proteins from a high speed supernatant of a tissue homogenate, ammonium sulfate precipitation (40-70s ammonium sulfate) of the reneutralized supernatant, molecular-sieve chromatography on a column of AcA 34, and ion-exchange chromatography on a FPLC mono Q column. The purified enzyme exhibited one major and several minor bands upon analysis by SDS PAGE.” The major band reacted with an antisera to the enzyme as judged by Western blot analysis.
,’ Abbreviations used: SDS-PAGE, sodium dodecyl acrylamide gel electrophoresis; @-NA, P-naphthylamide.
sulfate-poly-
5
6
7
8
9
10
PH FIG. 1. pH dependence of aminopeptidase
hydrolysis of ~-Ala+ naphthylamide. Reaction mixtures contained 0.1 M bis-Tris-propane buffer at the indicated pH, L-Ala-fl-naphthylamide (varied from 25 to 200 PM at pH’s 5.5 to 7.5 and 40 to 800 PM at pH’s 8.0 to 9.0), and enzyme in a final volume of 200 ~1. The reaction was monitored by following the increase in free P-naphthylamine fluorescence as described under Materials and Methods.
RESULTS
Prior to studies on the subsite specificity of the puromycin-sensitive aminopeptidase we measured the pH dependence of V,,,,, and Km using L-alanine-p-naphthylamide as substrate. As shown in Fig. 1, V,,, appears pH independent over the range of 5.5 to 9.0; however, K,,, and thus V,,,,,/K,,, exhibit a kinetic pK,, of 7.7. Potentiometric titration of L-alanine-P-naphthylamide under the same conditions revealed that the pK, of its free amino group is 7.7. Thus the reactive form of the substrate contains a protonated amino group, and the enzyme itself does not exhibit any ionizations which can be observed kinetically over the pH range studied. We next examined the effect of varying the N-terminal amino acid (Pl residue) on V,,, and Km using two series of dipeptides: X-Ala and X-Leu. In order to measure dipeptide hydrolysis we made use of the alanine and leucine dehydrogenase reactions in which the free amino acid liberated from the dipeptide is oxidized to the corresponding keto acid in an NAD-dependent reaction. The NADH thus formed is measured fluorometrically. The dehydrogenase reactions require rather alkaline pH conditions (pH optima around lo), therefore the reaction was conducted as a discontinuous assay: first generation of free amino acid by the aminopeptidase and then the determination of the free amino acid with the dehydrolarge genase. Since these assays require relatively amounts of the dehydrogenases we chose to measure the Km for each dipeptide by using it as an alternate substrate inhibitor of L-alanine-P-naphthylamide hydroly-
SPECIFICITY
OF RAT
BRAIN
PUROMYCIN-SENSITIVE TABLE
Comparison
of X-Ala
and X-Leu
I
Dipeptides
as Substrates
and Inhibitors X-Leu
X-Ala
X Ax LYS Met Leu Val Trp Phe Ala Ser Pro GUY
Glu
(:;, 0.06 0.17 0.74 0.81 1.25 1.58 2.09 14.7 18.1 >lOO >50
-t
V lmdr (units/mg)
SE 0.002 0.02 0.07 0.07 0.09 0.09 0.19 2.1 1.9
3.40 3.29 2.51 2.22 0.32 1.54 5.78
at 10
V ll,RX (units/mg)
K Vmx /Km
(mM)
56.7 19.4 3.4 2.7 0.3 1.0 2.8
SE
t
0.006 0.013 0.273 0.062 0.22 0.17 0.43 8.9 1.5 13.5
No inhibition
307
AMINOPEPTIDASE
mM
0.0008 0.0002 0.063 0.009 0.02 0.005 0.09 1.0 0.2 No inhibition 3.4 No inhibition
Vm /Km
1.75 4.03 1.68 4.65” 1.00 3.24 8.26
291.7 310.0 6.2 75.0 4.5 19.1 19.2
0.46
0.3
at 5 mM at 10 mM
Note. K,, was determined using the peptide as an inhibitor of L-Ala-P-NA hydrolysis. The reaction was measured at pH 7.5 in 100 mM bisTris-propane buffer with 10 pM l.-Ala-@-NA as substrate. V,,,, was determined by measuring the rate of product formation at a concentration of substrate 10X K,,. One unit corresponds to the formation of 1 pmol of product per minute. ” The observed rate was divided in half since two leucines were generated.
sis. Preliminary experiments demonstrated that the dipeptides were competitive inhibitors. Using this strategy the K,,, for each dipeptide was measured and then V,,, was determined by following the liberation of free alanine or free leucine at a saturating substrate concentration. The data are summarized in Table I. With both series of dipeptides the highest V,,,,/K, values are produced with arginine and lysine as the Nterminal amino acid. Amino acids containing hydrophobic side chains all exhibit similar Km and V,,,, values with the exception of valine, which exhibits a decreased V,,,,. Alanine, serine, and glycine have considerably lower Km values than do the hydrophobic amino acids. The effect of substituents in the P’l position of a dipeptide on hydrolysis of a Pl alanine was determined, with the data presented in Table II. This subsite also favors an arginine residue but binds hydrophobic residues, except valine, with a similar affinity. By comparing the data in Table I to that in Table II it can be seen that the affinities for alanine, serine, and glycine are considerably higher in the S’l site than in the Sl site. Again the predominant effect is one on Km rather than on V,,,. For example, a P’l glycine which binds some 60-fold weaker than arginine or tryptophan reacts =1.5- and e3.5fold slower than these amino acids, respectively. Since basic amino acids appeared to be the preferential amino acids for both the Sl and S’l subsites the K, for Lys-Lys and Arg-Arg was determined. As shown in Table III the K, values for these dipeptides are the same as observed with a P’l leucine. In order to determine whether the active site extends to a P’2 residue we examined Leu-Ser-Phe and Leu-
Trp-Met as substrates. As shown in Table III, a P’2 phenylalanine residue reduces K,,, by a factor of over 100 compared to Leu-Ser, while a P’2 methionine reduced Km by only threefold relative to Leu-Trp. For these substrates V,,, was increased five- and threefold, respectively. DISCUSSION
Previous studies on the specificity of the puromycinsensitive aminopeptidase have been conducted with TABLE
II
Comparison of Leu-X Dipeptides at Substrates and Inhibitors
X Arg Met Leu Val Tw Phe Ala Pro Asn GUY Tyr Ser
K,,,
(mM)
0.03 0.04 0.06 0.37 0.04 0.03 0.81 >lOO 1.77 1.77 0.03 0.28
k
SE
V m.ax (units/mg)
Vrnax/Km
0.0076 0.009 0.009 0.050 0.004 0.002 0.07
2.77 8.55 4.65” 7.07 6.59 3.06 0.73*
126 214 75 19 169 4 1
0.14 0.08 0.0008 0.07
3.56 1.94
2 1
0.73
3
Note. K,, and V,,,, were determined as described in Table I. ” The observed rate was divided in half since two leucines were generated. ‘I The same rate was observed whether alanine formation or leucine formation was measured.
308
JOHNSON TABLE
III
Comparison of Dibasic Dipeptides and Tripeptides as Inhibitors and Substrates Dipeptide
&orK,(mM)
Lys-Lys
Arg-Arg Leu-Ser-Phe Leu-Trp-Met Note.
Ser-Trp
The K, for Lys-Lys and Leu-Trp-Met
0.012 0.007 0.002 0.014
f
SE
V max
0.0016 0.0009 0.0005 0.004
3.42 18.00
and Arg-Arg and the K, and V,,,,, for Leuwere determined as described in Table I.
amino acid+-naphthylamides as substrates and thus only the specificity of the Sl site could be investigated. In addition the influence of P’l and P’2 residues on substrate binding and catalysis could not be determined. These problems have been circumvented by the use of di- and tripeptides as substrates and alternate substrate inhibitors of the enzyme. The data in Tables I and II show that there is subsite specificity for both the Sl and s’l subsites. Arginine and lysine show the highest affinity as the Pl amino acid. Hydrophobic amino acids all exhibit similar affinities, which are somewhat lower than Arg and Lys. This is similar to the results obtained with amino acid-/I-naphthylamides (29); however, the K,,, values for the amino acid-P-naphthylamides are significantly higher than those for dipeptides. The amino acids with small side chains, Ala, Ser, and Gly, exhibit a low affinity for the enzyme. Although the binding affinities vary over more than 2 orders of magnitude there is a considerably smaller variation in V,,,. Thus in terms of the Sl subsite the specificity of the enzyme is manifested almost solely in K,,,. The finding that arginine and lysine bind preferentially to this subsite suggests either that an anionic enzyme residue may be in the vicinity of this subsite or that there is an interaction with a backbone carbonyl. As shown in Fig. 1, the pH dependency of the reaction indicates that only the protonated form of the substrate is a reactant. Thus an anionic residue appears to be involved in binding the free amino group of the substrate. The possibility might be considered that this anionic group interacts with both the positive charge on the amino acid side chain and the amino terminus of the substrate. This is further strengthened by the observation that glutamate does not bind at this site. In addition to the ionic interaction at the Sl subsite it is apparent from Table I that at this binding site hydrophobic interactions can exist. It is also evident when comparing Leu and Ala that the amino acid in the s’l subsite can influence the strength of binding at the Sl subsite. This effect results in a 5- to lo-fold decrease in K,,, for the amino acid in the S’l subsite. Thus there appears to be a synergistic effect
AND
HERSH
between subsites which affects substrate binding. Based on the data in Table II it would appear that the S’l subsite shows considerably less variation in affinity for different amino acids than does the Sl subsite. This subsite does not show a preference for cationic side chains and thus probably utilizes hydrophobic interactions as the predominant binding interaction. However, this subsite can accommodate a positively charged side chain. The data in Table III show that the aminopeptidase active site extends at least to a S’2 subsite. It can be seen that the effect of the P’2 residue on substrate affinity and V max depends on the affinity of the parent dipeptide. Leu-Ser binds with a relatively weak affinity, Ki = 250 PM. Placement of a Phe in the S’2 subsite reduces Ki lOOfold to 2 PM. In contrast Leu-Trp binds with relatively high affinity, K, = 30 PM, and the addition of a Met residue reduces K, only 3-fold to 10 FM. This again indicates synergism between subsites which serves to maximize substrate binding. With both tripeptides V,,, was increased approximately 3-fold. Since arginine is the preferred amino acid in the Sl subsite and shows a high affinity for the $71 subsite it might be anticipated that placing arginine in both of these subsites would facilitate binding. However, when Arg-Arg is compared to Arg-Leu no difference in Km is observed. This is consistent with the hypothesis that subsite-subsite interactions maximize substrate binding. Hui et al. studied Arg-X and Lys-X (X = Arg, Phe, or Tyr) dipeptides as inhibitors of enkephalin hydrolysis and reported essentially no differences in their IC,, values, which were all approximately 10 PM (30). They reported that extending the dipeptide to a tripeptide in no case decreased K,,, and in several cases actually increased K,,,. In a separate study using enkephalin-related peptides as inhibitors of enkephalin hydrolysis Hui et al. (26) observed effects on I& among different P’4 residues and by extending the inhibitor to include a P’5 residue. However, in most cases the effects observed were small and unfortunately the effects on catalysis were not assessed. These results do suggest the presence of an extended active site beyond the S’2 position. However, it has been established that there is a size limitation for substrates for the enzyme which is presumably somewhere between 8 and 13 amino acid residues (33). REFERENCES 1. Patterson, E. K., Hsiao, S.-H., and Keppel, A. (1963) J. Biol. Chem. 238,3611-3620. 2. Danielson, E. M., Noren, O., Sjostrom, H., Ingram, J., and Kenny, A. J. (1980) Biochem. J. 189,591-603. 3. Tobe, H., Kojima, F., Aoyagi, T., and Umezawa, H. (1980) Biochim. Biophys. Acta 613,459-468. 4. Glenner, G. G., McMillan, P. J., and Folk, J. E. (1962) Nature (London) 194,867. 5. Hopsu, V. K., Makinen, K. K., and Glenner, G. G. (1966) Arch. Biochem. Biophys. 114,557-566.
SPECIFICITY
OF RAT
BRAIN
PUROMYCIN-SENSITIVE
6. Hersh, L. B. (1981) Biochemistry 20,2345-2350. 7. Smith, E. L., and Spackman, D. H. (1955) J. Biol. Chem. 212, 271-299. 8. Wachsmuth, D., Fritze, I., and Pfleiderer, G. (1966) Biochemistry t&169-174. 9. Brecker, A. S., and Suszkiw, J. B. (1969) Biochem. J. 112, 335-
342. 10. Suszkiw, J. B., and Brecher,
A. S. (1970) Biochemistry
9, 4008-
4017. 11. Hayashi, M., and Oshima, K. (1977) J. Biochem. 81,631-639. 12. Schnebli, H. P., Phillipps, M. A., and Barclay, R. K. (1979) Biochim. Biophys. Acta 669,89-98. 13. Hersh, L. B., and McKelvy, J. F. (1981) J. Neurochem. 36, 171178. 14. Hui, K. S., Wang, Y.-J., and Lajtha, A. (1983) Biochemistry 22, 1062-1067. 15. Hambrook, J. M., Morgan, B. A., Rance, M. J., and Smith, C. F. C. (1976) Nature (London) 262,782-783. 16. Meek, J. L., Yang, H. Y. T., and Costa, E. (1977) Neuropharmacol-
ogy16,151-154. 17. Ulrich, C., and Hersh, L. B. (1985) Peptides 6,475-482. 18. Bausback, H. H., Churchill, L., and Ward, P. E. (1988) Biochem. Pharmacol. 37,155-160. 19. Hayaishi, M. (1978) J. Biochem. 84,1363-1372.
AMINOPEPTIDASE
20. Ryden, G. (1966) Acta O&et.
309
Gynecol. Scar& 45, Suppl. 3,1-105.
21. Ellis, S., and Perry, M. (1964) Biochem. Biophys.
Res. Commun.
X,502-505. 22. Marks, N., Datta, R. K., and Lajtha, A. (1968) J. Biol. Chem. 243, 2882-2889. 23. Hersh, L. B. (1981) J. Neurochem. 36,1594-1596. 24. Gros, C., Giros, B., and Schwartz, J.-C. (1985) Biochemistry 24, 2179-2185. 25. Hersh, L. B., AboukHair, N., and Watson, S. (1987) Peptides 8,
523-532. 26. Hui, K.-S., Hui, M., Banay-Schwartz,
M., DeGuzman, T., Ling, N., and Lajtha, A. (1983) Peptides 4,639-646. 27. Coletti-Previero, M.-A., de Paulet, C., Mattras, H., and Previero, A. (1982) Biochem. Biophys. Res. Commun. 107,465-469. 28. McClellen, S., Dyer, S., Rodriguez, G., and Hersh, L. B. (1988) J. Neurochem. 61,1552-1559. 29. Hersh, L. B. (1985) J. Neurochem. 44,1427-1435. 30. Hui, K.-S., Hui, M. P.-P., Ling, N., and Lajtha, A. (1985) Life Sci.
36,2309-2315. 31. Williamson, D. H., Lopez-Vieira, O., and Walker, B. (1967) Biochem.J. 104,497-502. 32. Cleland, W. W. (1979) in Methods in Enzymology (Purich, D. L., Ed.), Vol. 63, pp. 103-137, Academic Press, San Diego. 33. Berg, M. J., and Marks, N. (1984) J. Neurosci. Res. 11,313-321.