ARCHIVES

Vol.

OF BIOCHEMISTRY

AND

290, No. 1, October,

BIOPHYSICS

pp. 229-232,

1991

An Elastolytic Proteinase from Rabbit Leukocytes: Purification and Partial Characterization’ Concetta

Gardi,

Institute

Received

Paola

of General Pathology

February

Calzoni,

Eleonora

and *Institute

1, 1991, and in revised

form

Cavarra,

Adriana

of General Physiology,

March

Academic

Press,

Inc.

Granulocyte elastase has been isolated and characterized in a variety of mammalian species, including human (l-5). Although the role played by this enzyme in physiological conditions is not well understood, evidence has been reported about the importance of leukocyte elastase in the pathogenesis of various diseases (6, 7). Elastases are a heterogeneous group of enzymes able to solubilize fibrous elastin (7). Moreover, they are also able to solubilize other matrix proteins such as collagen, fibronectin, proteoglycans, laminin, and fibrin (8, 9). For this reason, elastase is generally thought to play a pivotal role in degradation processes of extracellular matrix. ’ This work was supported by funds from MURST, Rome, Italy (Ricerca Scientifica 60% and 40%). * To whom correspondence should be addressed at Institute of General Pathology, Siena University, Via Laterino 8, 53100 Siena, Italy. fax: (577) 270642. 0003-9861/91$3.00

Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.

and Giuseppe

Lungarella2

of Siena, 53100 Siena, Italy

29, 1991

A proteinase with elastolytic activity was isolated from granules of rabbit bloodstream leukocytes, and purified to apparent homogeneity by a multi-step procedure consisting of ammonium sulfate precipitation, batch fractionation on DEAE-Sephadex A-50, and finally by preparative isoelectric focusing (IEF) on Sephadex G-75 Superfine. The molecular weight of the enzyme, as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), was 28,500. This enzyme shows an isoelectric point at pH 9.0. The proteinase is active against natural elastins as well as toward Suc(Ala),-NA, Methoxy-Sue-(Ala),-Pro-Val-NA, and (to a lesser extent) against Suc-(Ala)2-Pro-Leu-NA and Boc-Ala-ONp. The inhibition profile of the isolated enzyme indicates that rabbit granulocyte elastase belongs to the group of serine proteinases. Inhibition by some natural proteinase inhibitors is also observed. Unlike other mammalian elastases, it is insensitive to elastatinal. @ 1991

Pacini,*

University

We recently demonstrated that lung collagen breakdown products (CDP) derived from porcine pancreatic elastase digestion specifically stimulate lung collagen synthesis when intravenously injected in rabbits (10,ll). This fact lead us to supposethat lung collagen metabolism could be influenced by CDP derived from collagen breakdown by homologous elastase, likely by a positive feedback control. Unfortunately, leukocyte elastase in rabbit has not been isolated and characterized as yet. Although the presence of an elastolytic activity in rabbit granulocytes was primarily reported by some authors (12, 13), subsequent studies failed to demonstrate a true elastase activity in these (14, 15) cells. These latter studies have led to the suggestion that rabbit leukocytes, unlike other mammalian species, do not contain a true elastase but only an “elastase-like” activity that hydrolyzes synthetic substrates conventionally used for elastase. Actually, it is important to establish whether a counterpart of the human enzyme really exists in rabbit leukocytes, since many experimental models of human diseases, in which elastase has been implicated, have been developed in rabbit (16, 17). In this paper, we demonstrate the presence of an elastolytic enzyme in the lysosomal fraction obtained from rabbit granulocytes. This enzyme has been purified to homogeneity and partially characterized. The availability of pure elastase from rabbit leukocytes may facilitate studies on the physiological and pathological roles of such enzymes in various conditions. MATERIALS

AND

METHODS

Mater&. The following materials were obtained from the indicated sources: Sephadex G-75 Superfine and DEAE-Sephadex A-50 from

3 Abbreviations used: IEF, isoelectric focusing; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; CDP, collagen breakdown products; PMSF, phenylmethylsulfonyl fluoride; DFP, diisopropyl fluorophosphate; al-AP, ol,-antiprotease; PGF, postgranular supernatant fraction; GE, granular leukocyte extracts; RLE, rabbit leukocyte elastase. 229

GARDI Pharmacia Fine Chemicals; succinyl-L-alanine-L-alanine-L-alanine-pnitroanilide (SAPNA), succinyl-L-alanine-L-alanine-L-proline.L-leucinep-nitroanilide (SALPNA), methoxy succinyl-L-alanine-L-alanine-rproline-L-valine-p-nitroanilide (MSAPN), t-butyloxycarbonyl-L-alanine 4-nitrophenyl ester (Boc-Ala-ONp), phenylmethylsulfonyl fluoride (PMSF), diisopropyl fluorophosphate (DFP), elastatinal, o-phenanthroline, soybean trypsin inhibitor, chicken ovoinhibitor, leupeptin, cyiantiprotease (a,-AP), elastin from bovine neck ligament, and molecular weight standards from Sigma; crz-macroglobulin, E 64 and phosphoramidon from Boehringer-Mannheim; carrier ampholytes from LKB. Other reagents were of the highest quality available and were used without further purification. Rabbit lung elastin was purified essentially according to Davis and Mackle (18) from New Zealand rabbits. Isolation of neutrophils. Neutrophils were obtained from peripheral blood of New Zealand rabbits after removal of erythrocytes by sedimentation at unit gravity through de&ran as described by Boyum (19). Purification of leukocyte elastase. Isolated leukocytes were suspended in 0.15 M NaCl and homogenized at 4’C with glass/glass Potter type homogenizer. Nuclei and cell debris were centrifuged off at SOOgfor 10 min. The resulting supernatant was again centrifuged at 16,000g for 30 min at 4°C to separate the granular fraction (pellet) from the postgranular supernatant fraction (PGF). The granular fraction was resuspended in 1 M NaCl in the presence of 0.05% Triton X-100, and stored overnight at 4’C for protease extraction. The suspension was centrifuged at 16,OOOgfor 30 min, and water added to supernatant (5:1, v/v) to restore isotonicity. The supernatant, designated “granular leukocyte extracts” (GE), was stored at -20°C and thawed just before use (Step I). Further purification of enzyme activity was achieved by two additional steps consisting of ammonium sulfate precipitation (60% of saturation) (Step II), followed by a batchwise fractionation on DEAE-Sephadex A50 (Step III) according to a procedure previously reported in detail (5, 20), and preparative IEF in Sephadex G-75 Superfine (Step IV). The step III of purification turned out to be necessary to avoid disturbances, which commonly occur during preparative IEF in a narrow pH range, removing inactive acidic fraction. Preparative IEF was performed in (110 X 230 X 3.5 mm) layer gels, with a pH range 7-10, containing 4% Sephadex G-75 Superfine and 2% carrier ampholytes (pH 779 and 9-11 in equal amounts) according to a method previously described (5, 20). After completion of the electrophoretic run single protein bands were eluted from preparative IEF gel slices. The identification of the protein with elastolytic activity was carried out on [“Hlelastin from bovine neck ligament. The homogeneity of the purified materials was checked by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (21) and analytical IEF. Fractions obtained during various steps of purification were extensively dialyzed and assayed for enzyme activity and for protein concentration (31). Enzyme assays. Elastase activity was measured using tritium-labeled elastin from bovine neck ligament prepared by modification of a NaBH, reduction procedure as previously described by Banda and Werb (23). The labeled elastin suspension prepared by us showed a specific activity of 5.8 X lo5 dpm/mg. The same procedure was used to label rabbit lung elastin (sp act 5.7 X 10’ dpm/mg). Elastase activity was measured by determining the amount of radioactivity released after incubation of enzyme preparations, at various steps of purification, with 250 pg of [“Hlelastin in 0.1 M Hepes, pH 7.5, containing 0.5 M NaCl, 0.1% Brij. Final reaction volume was 130 pl. The reactions were carried out in Eppendorf tubes at 37°C for 12-36 h. At the end of incubation, assay tubes were centrifuged for 20 min at 20,OOOgin an Eppendorf microfuge. The radioactivity released into the supernatant of the reaction mixture was used as a misure of t,he enzyme activity. The radioactivity was determined by dissolving 100 ~1 of the supernatant in 18 ml of Hionic Fluor (Packard) and counted in a Packard liquid-scintillation counter (Tricarb 2000 CA). Elastase-like activity against synthetic substrates was determined USing SAPNA (24), MSAPN (25), SALPNA (26), and Boc-Ala-ONp (27). Kinetic studies. Hydrolysis of SAPNA, SALPNA, and MSAPN at 410 nm as well as Boc-Ala-ONp at 347 nm was employed for K, studies.

ET AL. All assays were performed at 25°C. Substrates were used at a minimal range of concentration of 0.1-3 times K,,,. Purified enzymes were employed at a final concentration of 0.2, 0.4, or 0.6 Fg/ml. The parameters K, and KC,, were determined from initial rate of hydrolysis by the Lineweaver-Burk met,hod. The activity on [3H]elastin was compared with that of rat pancreatic elastase and expressed as percentage of this value. Inhibitor studies. Twelve proteinase inhibitors were used to characterize the enzyme activity. The enzyme (final concentration 1Om6M) was preincubated with each inhibitor in 0.1 M Hepes buffer, pH 7.5,0.5 M NaCl, 0.1% Brij at 37°C for 30 min before assaying. The residual enzyme activities were measured on [3H]elastin substrate. Acrylamide electrophoresis. SDS-PAGE was carried out according to Laemmli (19). For the molecular weight determination of purified enzyme, bovine serum albumin (M, 66,000), ovalbumin (M, 45,000), glyceraldehyde-3-phosphate dehydrogenase (M, 29,000), trypsinogen (M, 24,000), soybean trypsin inhibitor (M, ZO,lOO), and n-lactalbumin (M, 14,200) were used as marker proteins. Analytical IEF was performed in a LKB flatplate apparatus (LKB 2117) using 5% polyacrylamide gels, pH 7.9-10.0, containing 2% carrier ampholytes.

RESULTS

The natural substrate elastin and a variety of synthetic substrates, reported as substrates for elastase, were used to detect the presence of enzyme activity in various preparations of rabbit leukocyte extracts. The data obtained indicate that an elastolytic activity detectable on tritiated elastin is present only in GE. On the other hand, enzyme activity toward various synthetic substrates (i.e., BocAla-ONp, MSAPN) is detectable in both GE and PGF. This implies that PGF shares with GE molecule(s) with esterase activity. The purification scheme for the isolation of rabbit leukocyte elastase (RLE), the protein recoveries, and the enzyme activity on [3H]elastin are given in Table I. The purification procedure detailed above yielded an enzyme preparation with approximately 62-fold purification compared with the initial crude lysosomal extracts. The isolated protein was judged pure by the presence of only a single band on SDS-PAGE and analytical IEF. Molecular weight and isoelectric point of the purified enzyme were 28.5 kDa and 9.0, respectively. This band (Fig. 1, lane d) was absent in the SDS-PAGE profile of PGF (Fig.

TABLE I Purification

of Rabbit

Total

protein Step”

b!&

Total

activity

(dpm/h)

X lo6

Leukocyte

Elastase

Specific activity (dpm/h/mg protein)

Yield

Purification

x 10’

(B)

factor

I

1636.0

20.1

1.2

100

II

314.6

13.9

4.4

69

1

III

41.3

9.4

22.8

46.8

19

IV

4.1

3.0

74.9

15.2

62.4

3.7

’ Purification steps described under Materials and Methods. I: granule extract; II: ammonium sulfate fractionation; III: batchwise fractionation with DEAE-Sephadex A-50, IV: isoelectric focusing. The protein content and the enzyme activity toward [3H]elastin (dpm X h) were determined as described under Materials and Methods. Data are from a typical experiment out of four.

RABBIT

LEUKOCYTE

231

ELASTASE TABLE

II

Summary of Some Properties of Rabbit Leukocyte Elastase Substrate SALPNA” MSAPN” SAPNA” Boc-Ala-ONp” Proteolytic Proteolytic

Km (mM) 5.51 2.07 1.32 6.01 activity activity

i * t If

Kc,, km11

1.50 0.07 0.05 0.02

1.65 22.65 16.35 1.38

on bovine on rabbit

-+ + f f

0.01 0.21 0.18 0.07

K.tIKm

b-’

299 10,942 12,386 229

s-‘1

f 19 f 980 f 790 f 23 19 50

[3H]elastinb [“H]elastin*

’ Expressed as mean + SD of triplicate determinations. ’ The enzyme activity was compared with that of pancreatic and expressed as percentage of this value.

enzyme

DISCUSSION

of various rabbit leukocyte preparations in a FIG. 1. SDSPAGE 15% gel of 0.75 mm thickness. (a) reference protein mixture comprising (top to bottom) bovine serum albumin (M, 66,000), ovalbumin (M, 45,000), glyceraldehyde-3-phosphate dehydrogenase (M, 36,000), carbonic anhydrase (M, 29,000), trypsinogen (M, 24,000), soybean trypsin inhibitor (M, 20,100), and a-lactalbumin (M, 14,200); (b) PGF; (c) GE; (d) RLE; and (e) GE after adsorption with insoluble elastin. The slab gel was stained with Coomassie brilliant blue.

1, lane b). The adsorption of GE with insoluble elastin for 20 min at 4°C resulted in a marked reduction of this band in the GE profile (Fig. 1, lane e). In Table II a summary of kinetic parameters for the hydrolysis of the synthetic substrates by rabbit leukocyte elastase is reported. The proteolytic activity on bovine and rabbit [3H]elastin by the isolated enzyme is also given. Although some commercial substrates for human leukocyte elastase (i.e., MSAPN and SAPNA) are readily hydrolyzed by RLE, this enzyme shows a lower activity against the commonly used elastase substrate SALPNA, as well as toward Boc-Ala-ONp. In addition, the data obtained with different elastin substrates indicate that RLE hydrolyzes homologous elastin better than bovine elastin. In an attempt to define inhibition profile of the enzyme isolated from lysosomal estracts, natural and synthetic inhibitors were examined in the elastolytic assay (Table III). As can be seen, the activity of elastase toward tritiated lung elastin is completely inhibited by DFP and PMSF, potent inhibitors of serine proteinases, whereas it is unaffected by o-phenanthroline, EDTA and phosphoramidon, inhibitors of metalloproteases, and E 64, an inhibitor of cysteine proteases. In addition, the proteolytic activity is inhibited by 20% by elastatinal and 25% by leupeptin, an inhibitor of serine and cysteine proteases, such as plasmin, trypsin, and cathepsin B. Chicken ovoinhibitor, soybean trypsin inhibitor, a,-antiprotease, and a,-macroglobulin inhibit RLE by 44 to 66%.

Although several investigators have raised some doubts about the existence of a true rabbit elastase activity, the results reported in the present paper clearly demonstrate that rabbit leukocytes, like other mammalian leukocytes, do contain a serine proteinase that will degrade elastin. The enzyme was purified from lysosomal extracts to apparent electrophoretic homogeneity by preparative isoelectric focusing. Unlike other mammals, rabbits have only a single form of leukocyte elastase with a “relative mobility” corresponding to 28,500 and an isoelectric point of 9.0. This enzyme efficiently hydrolyzes bovine-insoluble elastin even if its inherent efficiency on this substrate is substantially lower than that observed on homologous elastin. TABLE

III

Effects of Inhibitors on Rabbit Leukocyte Elastase

Inhibitor PMSFb DFP’ Elastatinal Soybean trypsin inhibitor Chicken ovoinhibitor Leupeptin o-Phenanthroline Phosphoramidon E 64d EDTA n,-Antiprotease nz-Macroglobulin

Concentration 1mM ImM 100 pg/ml 100 *g/ml 100 pg/ml 10 ~Lglml ImM ImM ImM 1mM 20 fiM 12 FM

Enzyme activity ( % of control value)

a

2 0 80 48 60 75 95 100 92 100 44 66

Note. The enzyme was preincubated with each inhibitor in 0.1 M Hepes buffer, pH 7.5,0.5 M NaCl, 0.1% Brij at 37°C for 30 min before assaying. The activity was determined on [“Hlelastin and was expressed as a percentage of the control activity. ’ Values were obtained from three different measurements. * Isopropanol (4%, v/v) present in preincubation mixture. ’ Dimethyl sulfoxide (4%, v/v) present in preincubation mixture. d Ethanol (1.25%, v/v) present in preincubation mixture.

232

GARDI

In regard to its esterase activity, RLE, like other mammalian leukocyte elastases, readily hydrolyzes MSAPN and SAPNA even if to a different extent. In addition, rabbit enzyme shows a lower affinity toward Boc-AlaONp and SALPNA. The latter finding was unexpected, since SALPNA has been found to be a particularly good substrate for rat and mouse elastases and is commonly used for human leukocyte elastase (5, 26, 28). The inhibition profile observed for the enzyme isolated by us provides evidence that RLE is a serine proteinase, since its activity is completely inhibited by PMSF and DFP (potent inhibitors of serine proteinases), whereas it is insensitive to inhibitors of metalloproteinases (i.e., ophenanthroline, phosphoramidon, EDTA) or cysteine proteinases (i.e., E 64). In addition, RLE shares with other mammalian leukocyte elastases some characteristic properties of sensitivity to (Y~-AP, a*-macroglobulin, soybean trypsin inhibitor, and chicken ovoinhibitor (29,30). Another unexpected finding is the poor sensitivity of this enzyme to elastatinal, which is commonly thought to be a specific inhibitor for elastase (31). As mentioned above, previous studies carried out by two different laboratories failed to demonstrate a true rabbit leukocyte elastase activity (14,15). In our opinion, these results may be explained by different possibilities, such as the method used for determining elastolytic activity (elastin-orcein), the source of rabbit granulocytes, and the amount of elastase present in the rabbit neutrophils. After histochemical stains for elastase, rabbit leukocytes have been reported to contain an amount of staining considerably less than that found in other mammalian neutrophils (32). This fact and the well known relatively poor sensitivity of the elastolytic assays carried out on elastin substrates labeled with chromophore may explain the discrepancy between earlier observations (14,15) and our reported results. An additional cause for this discrepancy in results may be due to the use of leukocytes harvested from an inflammatory peritoneal exudate with the possibility of an uptake of plasmatic inhibitors active toward elastase but not against another esterase( s) cleaving Boc-Ala-ONp. In any case, the acidic protein with BocAla-ONp esterase activity (“elastase-like”) enzyme isolated by Cotter and Robinson (14) from rabbit leukocytes is substantially different from the cationic enzyme isolated by us for biochemical and physical properties (i.e., molecular weight and isoelectric point). This view is further supported by the marked loss of Boc-Ala-ONp esterase activity observed (during our purification procedure) after the removal of anionic proteins by DEAE-Sephadex A50 (data not shown). Although additional studies are required to further characterize the enzyme isolated by us, the availability of pure RLE is expected to facilitate studies on the possible

ET AL.

role played by this enzyme under different physiological and pathological conditions. REFERENCES 1. Baugh, R. J., and Travis, J. (1976) Biochemistry 15, 836-841. 2. Ardelt, W., Tomczak, Z., Ksiezny, S., and Dudek-Wojciechowska, G. (1976) Biochim. Biophys. Acta 445,683-693. 3. Von Fellenberg, R., Kohler, L., Gtinig, G., and Pellegrini, A. (1985) Am. J. Vet. Res. 46, 2480-2484. 4. Gardi, C., and Lungarella, G. (1986) Arch. Biochem. 63-69. 5. Gardi, C., and Lungarella, G. (1988) Biochem.

Biophys.

260,

Znt. 16, 185-191.

6. Werb, Z., Banda, M. J., McKerrow, J. H., and Sandhaus, R. A. (1982) J. Znuest. Dermatol. 79,154s-159s. 7. Bieth, J. G. (1980) in Frontiers of Matrix Biology (Robert, L., Ed.) Vol. 8, pp. 216-227, Karger, Basel. 8. Janoff, A. (1985) Annu. Reu. Med. 36, 207-216. 9. Heck, L. W., Blackburn, W. D., Irwin, M. H., and Abrahamson, D. R. (1990) Am. J. Pathol. 136, 1267-1274. 10. Pacini, A., Gardi, C., Corradeschi, F., Viti, A., Belli, C., Calzoni, P., and Lungarella, G. (1990) Res. Commun. Chem. Path&. Pharmacol. 68,89-101. 11. Gardi, C., Pacini, A., de Santi, M. M., Calzoni, P., Viti, A., Corradeschi, F., and Lungarella, G. (1990) Res. Commun. Chem. Pathol. Pharmacol. 68, 235-250. 12. Janoff, A., Rosenberg, R., and Galdston, M. (1971) Proc. Sot. Exp. Biol. Med. 136, 1054-1058. 13. Fonzi, L., and Lungarella, G. (1979) Exp. Mol. Pathol. 31,486-491. 14. Cotter, T. G., and Robinson, 615,414-425.

G. B. (1980) Biochim.

Biophys.

Acta

15. Brim, M. L., and Lowther, D. A. (1981) Ajebak 59,63-75. 16. Fonzi, L., and Lungarella, G. (1980) Lung 158, 165-171. 17. Hayashi, K., Takamizawa, K., Nakamura, T., Kato, T., and Tsushima, N. (1987) Atherosclerosis 66, 259-267. 18. Davis, P. F., and Mackle, Z. N. (1981) Anal. Biochem. 19. Boyum, A. (1968) &and.

J. Clin. Lab. Invest.

20. Gardi, C., and Lungarella, G. (1987) Arch. Biochem. 98-104. 21. Laemmli, U. K. (1970) Nature

(London)

117, 11-17.

Suppl. 21, 27-30.

227,

Biophys.

259,

680-685.

22. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265-275. 23. Banda, M. J., and Werb, Z. (1981) Biochem.

J. 193,589-605.

24. Bieth, J. G., Spiess, B., and Wermuth, C. G. (1974) Biochem. 101, 232-241.

Med.

25. Nakajima, K., Powers, J. C., Ashe, B. M., and Zimmerman, M. (1979) J. Biol. Chem. 254, 4027-4032. 26. Del Mar, E. G., Largman, C., Brodrick, J. W., Fasset, M., and Geokas, M. C. (1980) Biochemistry 19,468-472. 27. Visser, L., and Blout, E. (1972) Biochim. 260. 28. Largman, C. (1983) Biochemistry

Biophys.

Acta 268,

257-

22,3763-3770.

29. Stockley, R. A. (1983) Clin. Sci. 64,

119-126.

30. Stein, R. L., Trainor, D. A., and Wildonger, Rep. Med. Chem. 20, 237-245.

R. A. (1985) Annu.

31. Umezawa, H., and Aoyagi, T. (1983) in Medical and Biological Aspects (Katunuma, M. et al., Eds.) pp. 3-15, Japan Sci. Sot. Press, Tokyo, Springer-Verlang, Berlin. 32. International Symposium on Biochemistry, Pathology and Genetics of Pulmonary Emphysema. Section 2: Animal Models (General Discussion) Clin. Resp. Physiol. (1980) lG(Suppl.), 175-176.