ties and characteristics of these BBB enzymes rela .... specific activity of the membrane-bound cerebral ... aminopeptidases than for the soluble enzymes.
Journal of Cerebral Blood Flow and Metabolism
7:801-805
©
1987 Raven Press, Ltd., New York
Characteristics of Aminopeptidase Activity from Bovine Brain Microvessel Endothelium
Anna Baranczyk-Kuzma and *Kenneth L. Audus Department of Biochemistry, Institute of Biopharmacy, Warsaw Medical School, Warsaw, Poland, and *Department of Pharmaceutical Chemistry, School of Pharmacy, University of Kansas, Lawrence, Kansas, U.S.A.
Summary: Blood-brain barrier (BBB) aminopeptidase activity was investigated using an in vitro model con sisting of primary cultures of brain microvessel endothe lium. Using two different substrates, both membrane bound and soluble aminopeptidases were found to be as sociated with brain endothelium. That the enzyme activity was aminopeptidase activity was confirmed with the competitive inhibition of substrate degradation by typical aminopeptidase inhibitors puromycin and bes tatin. The aminopeptidase activity was also competitively
inhibited by enkephalin, met-enkephalin, and leu-en kephalin. Results from parallel experiments with cerebral gray matter and kidney confirm assay conditions. This report supports previous suggestions that aminopepti dases of the enzymatic BBB may play a role in regulating levels of circulating neuropeptides in the cerebrovascula ture. Key Words: Aminopeptidase-Bestatin-B1ood brain barrier-Brain endothelium-Enkephalins-Pu romycin.
The blood-brain barrier (BBB) forms both a per meability and an enzymatic barrier that restricts the entry of circulating neurotransmitter or neuromod ulator molecules into the CNS (Rapoport, 1976). Although evidence suggests some neurotransmitter or neuromodulator peptides may cross the BBB, the properties that favor BBB penetration are not well understood (Banks and Kastin, 1985; Par dridge, 1986). Similarly, though some peptidase ac tivity at the BBB has been demonstrated (Brecher et aI. , 1978; Pardridge and Mietus, 1981), the activi ties and characteristics of these BBB enzymes rela tive to peptidases from other tissues are not known. Moreover, the role BBB peptidases may play in regulating passage of peptides across the BBB and cerebrovascular levels of circulating peptides is also not clear. The present study was initiated to evaluate BBB peptidase activity responsible for hydrolysis of the amino-terminal peptide bonds of polypeptides. These peptidases, referred to as aminopeptidases,
are involved, in part, in the degradation of circu lating neuropeptides, such as the enkephalins (Par dridge and Mietus, 1981; Hui et aI. , 1983), somato statin (Pardridge et al. , 1985) and substance P (Pal mieri and Ward, 1983). The activity of the BBB aminopeptidases was determined in an in vitro model system comprised of primary cultures of brain microvessel endothelial cells. Primary cul tures of brain microvessel endothelial cells have previously been shown to represent a useful model for studying, at the biochemical level, the enzy matic BBB (Audus and Borchardt, 1986a; Bar anczyk-Kuzma et aI. , 1986). To provide a compar ison of aminopeptidase activity associated with the BBB model to other tissues, aminopeptidase ac tivity of kidney and cerebral gray matter was also determined.
MATERIALS AND METHODS Chemicals Tyrosine-l3-naphthylamide (tyr-l3-naphthylamide), 13naphthylamine, L-Ieucine-p-nitroanilide (L-Ieu-p-nitroani lide), puromycin, bestatin, enkephalin, methionine5-en kephalin (met-enkephalin), and leucine5-enkephalin (leu en kephalin) were purchased from Sigma Chemical Co., St. Louis, MO, U.S.A. All other reagents were of the highest grade commercially available.
Received April 1987, accepted June 1987. Address correspondence and reprint requests to Dr. K. L. Audus at Department of Pharmaceutical Chemistry, School of Pharmacy, Malott Hall, University of Kansas, Lawrence, KS, U.S.A. Abbreviations used: BBB, blood-brain barrier; PBS, phos phate buffered saline.
801
802
A. BARANCZYK-KUZMA AND K. L. AUDUS
Isolation and culture of microvessel endothelium Bovine brain microvessel endothelial cells were iso lated from the superficial gray matter scraped from entire cerebral cortices and grown to monolayers (-10 days) in primary culture as described by Audus and Borchardt (1986) and Baranczyk-Kuzma et al. (1986). Extensive characterization of these monolayers has been described previously (Audus and Borchardt, 1986a; Audus and Borchardt, 1986h; Baranczyk-Kuzma et aI. , 1986; Rim et al. , 1986).
Enzyme extraction Bovine brain and kidney were obtained from a local slaughterhouse within 30 min of death and immediately transported to the laboratory on ice. Aminopeptidase were extracted from kidneys and from superficial gray matter scraped from entire cerebral cortices by the method of Hui et al. (1983). Briefly, 1:3 (weight/volume) homogenates of brain or kidney tissues were prepared in 0.32 M sucrose and centrifuged at 800 g for 60 min. The supernatant was centrifuged at 30,000 g for 20 min, and the pellet of cell debris discarded. Following the second centrifugation step, the supernatant was removed and used for assaying soluble aminopeptidase activity. The membrane pellet was resuspended in 50 mM Tris HCI (pH 7.5), and twice washed by centrifugation at 30,000 g for 20 min. The membrane pellet was then solubilized by sus pension and incubation in 20 ml of I % Triton X-IOO (weight/volume) in 50 mM Tris HCI (pH 7.5) at 37°C for 45 min. The solubilized membrane enzyme was obtained by centrifugation of the suspension at 30,000 g for 10 min and removal of resulting supernatant, which contained the solubilized membrane aminopeptidase. All steps of the preparation were carried out at 4°C. Ten-day old microvessel endothelial cell monolayers grown in several 100-mm plastic culture dishes were washed with 0.01 M isotonic phosphate buffered saline [(PBS) pH 7.4], scraped from the dishes with a rubber policeman, and collected by centrifugation at 1,000 g for 10 min. After freezing overnight, the cells were thawed quickly and aminopeptidase extraction carried out as de scribed.
Enzyme assays The hydrolysis of tyr-J3-naphthylamide was assayed by the method of Greenberg (1962). The reaction mixture consisted of 50 I-lM tyr-J3-naphthylamide in 50 mM TES buffer (pH 7.0) and 10-100 I-li of enzyme (10-100 I-lg of protein) in a total volume of 2 ml. The reaction was car ried out at 23°C for 5 to 10 min. J3-naphthylamide was used as a standard for the measurements, and the boiled enzyme was used as the blank for the assay (Hui et aI., 1983). Fluorescence emission was observed at 410 nm with excitation of 335 nm on a Perkin-Elmer spectro fluorometer. The hydrolysis of L-Ieu-p-nitroanilide was assayed by the method of Wacker et al. (1971). The reaction mixture consisted of 1.66 mM L-Ieu-p-nitroanilide in 50 mM TES buffer (pH 7.0) and 50-100 I-li of enzyme (10-100 I-lg pro tein) in a total volume of 1 ml. Blanks consisted of reac tion mixtures with substrate or enzyme omitted. Absor bance was followed at 405 nm with a Gilford spectropho tometer and the rate of hydrolysis of L-Ieu-p-nitroanilide calculated using the molar absorbance coefficient of 9,620 for p-nitroaniline. Kinetic parameters, Km and Ki, were determined from
J Cereb Blood Flow Metabol, Vol. 7, No.6, 1987
double-reciprocal plots. Ki values were determined from a fixed substrate concentration. Data points in figures and tables represent the mean of at least triplicate samples plus or minus the standard deviation. For data points where the standard deviation is now shown; stan dard deviations were generally 10-15% of the mean. Tissue protein was determined by the method of Lowry et al. (1951).
RESULTS
The time-dependency of aminopeptidase activity from bovine kidney, cerebral gray matter, and brain microvessel endothelium was linear under the con ditions of the enzyme assay with both L-Ieu-p-ni troanilide and tyr-f3-naphthylamide as substrates (data not shown). The membrane-bound kidney aminopeptidases show higher specific activity with both substrates than the soluble forms (Table 1). In contrast, the specific activity of the membrane-bound cerebral gray matter aminopeptidases is lower than the sol uble enzymes. Both forms of aminopeptidases from brain microvessel endothelium show similar spe cific activities with L-leu-p-nitroanilide, whereas with tyr-f3-naphthylamide as a substrate, the mem brane-bound aminopeptidases have a higher ac tivity and parallels results from the kidney. The ap parent affinity (Km) of both membrane-bound and soluble aminopeptidases for tyr-f3-naphthylamide is similar in all tissues (Table l). However, the ap parent Km values with L-Ieu-p-nitroanilide as a sub strate are generally higher for the membrane-bound aminopeptidases than for the soluble enzymes. Both membrane-bound and soluble aminopepti dases from all tissues assayed were inhibited by a general aminopeptidase inhibitor, bestatin, and pu romycin, an inhibitor of aminopeptidases that are arylamidases (Hui et aI. , 1983). In our reference brain and kidney tissues, the inhibition by bestatin and puromycin was time- and concentration-depen dent further confirming that assay conditions were specific for aminopeptidases. Representative re sults demonstrating the time- and concentration-de pendent inhibition of aminopeptidases by bestatin in the reference tissues are given in Figure 1. All enzymes were also inhibited by endogenous pep tides such as enkephalin, met-enkephalin, and leu enkephalin in a concentration-dependent manner. All compounds tested all displayed competitive type of inhibition. Results graphically presented in Figure 2 are representative of data obtained for all assayed aminopeptidases. Puromycin and bestatin were more effective in hibitors of the aminopeptidases than the enkeph alins (Table 2). Generally, the soluble forms of
803
AMINOPEPTIDASES OF BRAIN ENDOTHELIUM TABLE 1.
Comparison of bovine aminopeptidase activity and affinity for different substrates Tyr-f3-naphthylamide
L-Leu-p-nitroanilide Km (fLM)
Specific activity (nmol/mg/min)
Enzyme Cultured brain microvessel endothelium Soluble Membrane
1.1 1.0
Kidney
9.4 122.0
Soluble Membrane Cerebral gray matter Soluble Membrane
31.3 1.7
± ±
± ±
± ±
Spectrometer activity (fLmol/mglmin)
0.3 0.4
180 250
0.8 6.9
0.6 2.8
70 190
1.53 26.3
5.0 0.4
40 2940
36.6 0.41
± ±
± ±
± ±
Km
(fLM)
0.1 1.4
25 25
0.04 0.9
29 21
0.8 0.03
16 22
Aminopeptidase activity was assayed as described in Methods with 1.66 mM L-Ieu-p-nitroanilide and 50 fLm tyr-f3-naphthylamide.
aminopeptidase from kidney and cerebral gray matter are more sensitive to puromycin than mem brane-bound enzymes. By comparison, bestatin is the more effective inhibitor for kidney membrane bound and cerebral gray matter soluble aminopep tidases. Bestatin is also a very good inhibitor for both membrane-bound and soluble aminopepti dases from brain microvessel endothelium. In general, the different enkephalins were more BOVINE BRAIN AMINOPEPTIDASE
Soluble Form -Inhibitor
Membrane -Inhibitor
100
80
60 0.5
effective inhibitors of the membrane-bound amino peptidases from kidney and brain microvessel en dothelium than of the enzyme from gray matter (Table 2). For both soluble and membrane-bound aminopeptidases from kidney and gray matter, leu enkephalin was found to be the best inhibitor as compared to the other enkephalins. In contrast, met-enkephalin is a better inhibitor of both forms of aminopeptidases from brain microvessel endothe lium. Similar results were obtained if L-leu-p-ni troanilide is used as a substrate (data not shown) to generate data given in Table 2, with the exception that the assay was not as sensitive (i. e. , higher con centrations of inhibitors were required to inhibit substrate hydrolysis) to inhibition by any of the compounds listed.
11M
40 1.25 >-
20
20
;;
t
« Ol c:
0 0
1
2
4
5
6
7
2.5
0
1
2
3
4 5
6
11M 0.1
11M
7
0.08
BOVINE KIDNEY AMINOPEPTIDASE
'c 'OJ
Soluble Form
Membrane
E QJ 100 a:
"0 1f'.
3
11M
-Inhibitor
-
In h i b i t o r
80
60
40 10 11M
-.04
20
-.02
o
0.02
0.04
0.06
0.08
0.1
1/[Tyr-p-naphthylamide]; IIM-1
o "--.L-.L-L-L....J
0
1
2
3
4
5
o
1
2
3
4
5
Minutes FIG. 1. Inhibition of aminopeptidases from bovine cerebral gray matter and kidney by bestatin. Assay of aminopeptidase activity was performed as described in the Methods section with 50 fLM tyr-f3-naphthylamide as a substrate in the pres ence of indicated concentrations of inhibitor.
FIG. 2. Inhibition of membrane-bound aminopeptidase from bovine brain microvessel endothelium. The assay of enzyme activity was performed as described in the Methods section with tyr-f3-naphthylamide a substrate with or without indi cated inhibitors. Fifteen micromoles per liter met-enkeph alin, open circles; 0.05 fLM puromycin, open boxes; 0.75 fLM
bestatin, filled triangles; 5.0 fLM leu-enkephalin, filled boxes; 5.0 fLM enkephalin, open triangles; without inhibitor, filled circles.
J Cereb Blood Flow Metabol. Vol. 7, No.6, 1987
804
A. BARANCZYK-KUZMA AND K. L. AUDUS TABLE 2. Effect of peptidase inhibitors and enkephalins on tyr-�-naphthylamide degradation by aminopeptidase
Inhibitor
Soluble Kj (!Lm)
Cultured brain micro vessel endothelium 0.20 ± 0.02 Puromycin Bestatin 1.20 ± 0.01 50.01 ± 0.01 Enkephalin Met-enkephalin 20.8 ± 2.7 Leu-enkephalin 30.0 ± 0.2
0.06 0.8 26.8 7.5 16.2
±
0.02 0.1 ± 4.6 ± 1.4 ± 1.6 ±
Kidney Puromycin Bestatin Enkephalin Met-enkephalin Leu-enkephalin
0.09 5.3 89.3 74.4 17.5
0.Q2 0.6 ± 3.5 ± 3.4 ± 1.7
7.1 1.4 4 1. 1 1 33.3 5.70
±
Cerebral gray matter Puromycin Bestatin Enkephalin Met-enkephalin Leu-enkephalin
0.08 0.52 21.0 15.0 5.5
±
0.Q2 0.03 ± 4.1 ± 5.0 ± 1.1
6.70 1.2 51.2 35.3 9. 1
±
± ±
±
2.0 0.2 ± 0.01 ± 4.7 ± 0.02 ±
0.03 0.2 ± 8.4 ± 1.2 ± 1.9 ±
Aminopeptidase activity was assayed as described in the Methods section with tyr-l3-naphthylamide as a substrate.
DISCUSSION The aminopeptidase activity associated with brain (Hersh, 1981; Hersh and McKelvy, 1981; Hui et aI. , 1983), kidney (Wacker et aI. , 1971; Wacker et aI. , 1976; Lin and Van Wart, 1982), and the periph eral vasculature (Ward, 1984; Palmieri et ai., 1985) is relatively well characterized. In contrast, the amino peptidase activity associated with the BBB has been demonstrated only qualitatively in isolated brain microvessels (Pardridge and Mietus, 1981; Pardridge et aI. , 1985). Using two different amino peptidase substrates in this study, the presence of membrane-associated and soluble fraction amino peptidases in an in vitro model of the BBB was demonstrated. The in vitro BBB model used in this study is comprised of primary cultures of brain microvessel endothelial cell monolayers. The monolayers have previously been shown to retain biochemical, mor phological, and permeability properties character istic of the BBB in vivo (Audus and Borchardt, 1986a; Audus and Borchardt, 1986b; Baranczyk Kuzma et aI. , 1986; Rim et aI. , 1986). The use of primary cultured brain microvessel endothelium as an in vitro BBB model was considered more appro priate for these biochemical studies than isolated brain microvessels since the latter in vitro model is known to have a primary defect in energy metabo lism (Pardridge, 1986). The kinetic parameters, Km and Kj, and specific activities of the "crude" preparations of aminopepJ Cereb Blood Flow Metabol, Vol. 7, No.6, 1987
tidases from kidney and cerebral gray matter used in this study are similar to those of previous studies with "purified" preparations of the enzymes (Hersh, 1981; Hersh and McKelvy, 1981; Wacker et aI. , 1971). A significant aminopeptidase activity in brain microvessel endothelium observed in this study is also consistent with the previous findings of Pardridge and Mietus (1981), and Pardridge et ai. (1985). Results here indicate a generally lower overall aminopeptidase activity associated with brain mi crovessel endothelium when compared with cere bral gray matter. An exception is the membrane bound aminopeptidase activity of brain microvessel endothelium determined with tyr-l3-naphthylamide as substrate. Interestingly, membrane-bound ami nopeptidases from brain microvessel endothelium also have a much higher apparent affinity for the inhibitors puromycin, enkephalin, and met-enkeph alin than cerebral gray matter. The characteristics of this membrane-bound aminopeptidase are analo gous to those of a "membrane aminoenkepha linase" recently described by Giros et ai. (1986) in brain slices. Though confirmatory experiments are necessary, these observations suggest that a mem brane-bound aminopeptidase from brain tissue might exist which is specifically associated with brain microvessel endothelium. Pardridge and Mietus (1981) provided the original evidence that rapid degradation of leu-enkephalin occurs as a result of interaction with aminopepti dases of isolated brain microvessels. In that study, the disappearance of leu-enkephalin and a corre sponding increase in the levels of free tyrosine was followed as a function of time with high-perfor mance liquid chromatographic methods. Specific activities and kinetic parameters for the suspected aminopeptidase involved were not quantitated, however. The results of this study showing that the enkephalins competitively inhibit both a soluble and a membrane-bound aminopeptidase from the BBB are in agreement with the work of Pardridge and Mietus (1981). The results also show that the enkephalins have a higher affinity for the mem brane-bound aminopeptidase. The existence of sig nificant membrane-bound aminopeptidase activity in brain microvessel endothelium found in this study supports the theory that membrane-bound aminopeptidases comprise a physiological mecha nism by which the vasculature regulates circulating levels of peptides and proteins (Palmieri et aI. , 1985). In the individual tissues, differences exist be tween the kinetic parameters for aminopeptidases measured with the two substrates. This finding may
AMINOPEPTIDASES OF BRAIN ENDOTHELIUM
suggest that either the two substrates are hydro lyzed by different aminopeptidases, or that the aminopeptidases present exhibit a preference for a particular substrate. To verify that aminopeptidases assayed with one substrate (e.g., tyr-l3-naphthyla mide) might be different from aminopeptidases as sayed with a second substrate (e.g., L-Ieu-p-ni troanilide) in the same tissue, requires further study. The relatively well-characterized aminopeptidase activity associated with brain and kidney tissue was used to confirm assay conditions for initial verifica tion of aminopeptidase activity in cultured brain microvessel endothelium. As shown and discussed above, results of our studies with kidney and brain tissue were consistent with previous reports. Since these studies confirm the presence of aminopepti dases in this in vitro model, future efforts will be directed at providing a more extensive biochemical characterization of aminopeptidases associated with brain microvessel endothelium. To conclude, the present study demonstrates the presence of both membrane-bound and soluble forms of aminopeptidases in primary cultures of brain microvessel endothelium that are competi tively inhibited by typical aminopeptidase inhib itors, bestatin, and puromycin. Three representa tive circulating neuropeptides enkephalin, met-en kephalin, and leu-enkephalin were also found to competitively inhibit the aminopeptidases. These results provide the basis for further investigation of the biochemical and physiological role for BBB aminopeptidases in regulating peptide penetration of the BBB and cerebrovascular levels of blood borne peptides and proteins. Acknowledgment: This work was supported by a grant from The Upjohn Company, Kalamazoo, Michigan.
REFERENCES Audus KL, Borchardt RT (l986a) Characterization of an in vitro blood-brain barrier model system for studying drug trans port and metabolism. Pharm Res 3:81-87 Audus KL, Borchardt RT (l986b) Characteristics of the large neutral amino acid transport system of bovine brain micro vessel endothelial cell monolayers. J Neurochem 47:484488 Banks WA, Kastin AJ (1985) Permeability of the blood-brain barrier to neuropeptides: T he case for penetration. Psy choneuroendocrinology 10:385-399
805
Baranczyk-Kuzma A, Audus KL, Borchardt RT (1986) Cate cholamine-metabolizing enzymes of bovine brain micro vessel endothelial cell monolayers. J Neurochem 46: 19561960 Brecher P, Tercyak A, Gavras H, Chobanian AV (1978) Peptidyl dipeptidase in rabbit brain microvessels. Biochim Biophys Acta 526:537-546 Giros B, Gros C, Solhonne B, Schwartz J-C (1986) Characteriza tion of aminopeptidases responsible for inactivating endoge nous (met5)enkephalin in brain slices using peptidase inhib itors and anti-aminopeptidase M antibodies. Mol Pharmacol 29:281-287 Greenberg LJ ( 1962) Fluorometric measurement of alkaline phosphatase and aminopeptidase activities in the order of 10-14 mole. Biochem Biophys Res Commun 9:430-435 Hersh LB ( 1981) Solubilization and characterization of two rat brain membrane-bound aminopeptidases active on met-en kephalin. Biochemistry 20:2345-2350 Hersh LB, McKelvy JF ( 1981) An aminopeptidase from bovine brain which catalyzes the hydrolysis of enkephalin. J Neu rochem 36:171-178 Hui K-S, Wang Y -J, Lajtha A (1983) Purification and character ization of an en kephalin aminopeptidase from rat brain membranes. Biochemistry 22: 1062-1067 Lin SH, Van Wart HE (1982) Effect of cryosolvents and subzero temperatures on the hydrolysis of L-Ieucine-p-nitroanilide by porcine kidney leucine aminopeptidase. Biochemistry 21: 5528-5533 Lowry OH, Rosebrough NJ, Farr AL, Randall RJ ( 1951) Protein measurement with Folin phenol reagent. J Bioi Chem 193:265-274 Palmieri FE, Petrelli JJ, Ward PE (1985) Vascular, plasma mem brane aminopeptidase M. Biochem Pharmacol 34:23092317 Palmieri FE, Ward PE (1983) Mesentery vascular metabolism of Substance P. Biochim Biophys Acta 755:522-525 Pardridge W M ( 1986) Receptor-mediated peptide transport through the blood-brain barrier. Endocrine Rev 7:314-330 Pardridge W M, Eisenberg J, Yamada T (1985) Rapid sequestra tion and degradation of somatostatin analogues by isolated brain microvessels. J Neurochem 44: 1 178-1184 Pardridge W M, Mietus LJ ( 1981) Enkephalin and blood-brain barrier: Studies of binding and degradation in isolated brain microvessels. Endocrinology 109: 1138- 1143 Rapoport SI (1976) Sites and functions of the blood-brain bar rier. In: Blood-Brain Barrier in Physiology and Medicine (Rapoport Sl, ed), New York, Raven Press, pp 43-86 Rim S, Audus KL, Borchardt RT (1986) Relationship of octanoll buffer and octanollwater partition coefficients to transcel lular diffusion across brain microvessel endothelial cell monolayers. Int J Pharm 32:79-84 Wacker H, Lehky P, Fischer EH, Stein EA ( 1971) Physical and chemical characterization of pig kidney particulate amino peptidase. Helv Chim Acta 54:473-485 Wacker H, Lehky P, Vanderhaeghe F, Stein EA (1976) On the subunit structure of particulate aminopeptidase from pig kidney. Biochim Biophys Acta 429:546-554 Ward PE ( 1984) Immunoelectrophoretic analysis of vascular membrane-bound angiotensin I converting enzyme, amino peptidase M, and dipeptidyl(amino)peptidase IV. Biochem PharmacoI33:3183-3193
J Cereb Blood Flow Metabol, Vol. 7, No. 6, 1987
