Cathepsin L Inactivates cul-Proteinase Inhibitor by Cleavage in the

THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1986 by The American Society of Biological Chemists, Inc.

Val ,261,No. 31, Issue of November 5, PP

Cathepsin L Inactivates cul-Proteinase Inhibitor by Cleavage in the Reactive Site Region* (Received for publication, June 10,1986)

David A. Johnson$& AlanJ. BarrettlI, and RobertW. MasonlI From the 3 Department of Biochemistry, Quillen-Dishner College of Medicine, East Tennessee State University, Johnson City. Tennessee 37614 and the liDeDartment of Biochemistry, Strangeways Research Laboratory, Cambridge GB1 4RN, Great Britain

The lysosomal cysteine proteinases cathepsinL and catalytically cleave and inactivate human alPI (Johnson and cathepsinB were examinedfortheir effect onthe Travis, 1977). As part of that study, the effect of human neutrophil elastase inhibitory activity ofhuman el- cathepsin B on alPI was investigated also; but in contrast to proteinase inhibitor (aIPI).Human cathepsin L cata- papain, equal molar amounts of cathepsin Bwere required for lytically inactivated human alPI by cleavage of the inactivation of alPI. bonds G 1 ~ ~ ~ ~and - AMet35s-Sers5e l a ~ ~ ~ (the serine proIn the present report, we examine the effects of purified teinase inhibitorysite). Cathepsin B did not inactivate cathepsins B and L on the ability of alPI to inhibit human alPI, even when equimolar amounts of enzyme were neutrophil elastase. These studiesshow that cathepsin L may employed. Cathepsin L is the first human proteinase shown tocatalytically inactivate oIPI. These findings, contribute both directly and indirectly to elastin degradation. in conjunction with other reports, suggest that aIPI Additional evidence is presented for a proteinase-sensitive contains a proteolyticallysensitive region encompass- region in a,PI. ing residues350-358. Taken togetherwith the discovEXPERIMENTALPROCEDURES ery of theelastinolytic activity of cathepsinL (Mason, Materials-Human alPI was purified as previously described R. W., Johnson, D. A., Barrett, A. J., and Chapman, H. A. (1986) Biochem. J. 233, 925-927), the present (Travis and Johnson, 1981). Cathepsins B and L were purified from findings emphasize the possible importance of cathep- human liver according to the procedures of Barrett and Kirschke (1981) and Mason et al. (1985),respectively. Humanneutrophil sin L inthepathologicalproteolysisof elastin and elastase was isolated from sputum essentially as described by MarB todam et al. (1979).Papain (Type 111, twice crystallized) and the diminish the role that can be attributed to cathepsin in such processes. were from substrate methoxysuccinyl-Ala-Ala-Pro-Val-p-nitroanilide Sigma. Z-Phe-Arg-NHMec and Z-Arg-Arg-NHMec were purchased from Enzyme Systems Products, Livermore, CA. E-64 was kindly provided by Dr. K. Hanada,TaishoPharmaceutical Co., Saitma, Japan. Blood plasma al-proteinase inhibitor (alPI1) inhibits sevAssay of Enzymes and Inhibitors-Human a,PI was assayed by eral serine proteinases including elastase, trypsin, and chy- measuring its inhibition of human neutrophil elastase activity on motrypsin. Individuals genetically deficient in alPI are pre- methoxysuccinyl-Ala-Ala-Pro-Val-p-nitroanilide(Castillo et al., disposed to emphysema (Ericksson and Laurell, 1963), and 1979). Neutrophil elastase and alPI were determined to be 95 and the primary function of the protein is probably the inhibition 92% active, respectively (Smith andJohnson, 1985).Cathepsin Bwas assayed with Z-Arg-Arg-NHMec (Barrett and Kirschke, 1981), and of human neutrophil elastase (Travis andSalvesen, 1983). L was assayed with Z-Phe-Arg-NHMec (Mason et al., cathepsin Bronchoalveolar lavage fluids have been found to contain 1985); molar concentrations of both enzymes were determined by lysosomal cysteine proteinase activity (Orlowski et al., 1981; titration with E-64 (Barrett andKirschke, 1981).Enzyme and inhibBurnettand Stockley, 1985), anda major portion of the itor concentrations given in the text refer to active proteins. elastinolytic activity of cultured humanalveolar macrophages Inactivation Studies-The inactivation of a1PI was measured by assaying for elastase inhibitory activity following incubation with has been attributed to such proteinases (Chapman and Stone, L. The inhibitor (1nmol) was incubated with cathepsin 1984a, 1984b). Cathepsins B and L, two well-recognized ly- cathepsin B or sosomal cysteine proteinases, were recently examined for B (1 nmol) or cathepsin L (2.5pmol) in 0.1 M sodium acetate buffer, pH 5.5,1 mM EDTA, 4 mM dithioerythritol in a total volume of 500 elastinolytic activity (Mason et al., 1986).At pH 5.5, cathepsin p1 a t 25 “C. Atvarious times, 50-p1 samples were removed and mixed L hydrolyzed elastin at a rate comparable to porcine pan- with 150 r l of 0.2 M Tris-HC1, pH 8.5,which contained 1 FM E-64. creatic elastase at pH 8.0, but cathepsin B had little activity Both alkaline pH and E-64 inactivate the cathepsins. An identically incubated and treated sample of alPI served as thecontrol. Samples against this substrate. Papain (a plant cysteine proteinase) has been shown to and controls were assayed at the same time for neutrophil elastase inhibitory activity. pH Optimum Determination-The pH optimum for cathepsin L * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby inactivation of alPI was determined by incubating alPI with cathepmarked “aduertisement” in accordance with 18 U.S.C. Section 1734 sin L for 30 min a t a 2001 molar ratio of inhibitor to enzyme in a buffer containg 50 mM citrate, 50 mM MES, 50 mM BisTris, 1 mM solely to indicate this fact. 8 Supported by The Health Effects Institute, Cambridge, MA. To EDTA, and 4 mM dithioerythritol adjusted to the desired pH with NaOH. Because alPI is inactivated at acid pH, control samples were whom correspondence should be addressed. The abbreviations used are: alPI, al-proteinase inhibitor; E-64, incubated in the same buffer as the samples with cathepsin L. The inactivation of alPI ateach pH was calculated relative to the appro~-3-carboxy-trans-2,3-epoxypropionylleucylamido-(4-guanidino)butane; Z, benzyloxycarbonyl; NHMec, 7-(4-methyl)coumarylamide; priate control. Reactions were stopped by dilution with an equal PTH, phenylthiohydantoin; MES, 2-(N-morpholino)ethanesulfonic volume of 0.2 M Tris-HC1, pH 8.5,1 p M E-64. acid; BisTris, bis(2-hydroxyethyl)iminotris(hydroxymethyl)methane; SDS-Polyacrylamide Gel Electrophoresis-Samples for electrophoresis containing 50 pg of alPI were incubated for 1 h in the pH 5.5 SDS, sodium dodecyl sulfate.

14748

Cathepsin L Inactivates cY1-ProteinaseInhibitor buffer, followed by precipitation of the proteins with an equal volume of cold 20% (w/v) trichloroacetic acid. After centrifugation, the protein precipitates were washed once with 10% (w/v) trichloroacetic acid and twice with acetone. The dried samples were taken up in 50 p l of SDS sample buffer containing 2 mM dithioerythritol and boiled for 5 min prior to electrophoresis in a 21% (w/v) polyacrylamide gel according to Thomas and Kornberg (1975). Gels were stained with a solution of Coomassie Brilliant Blue containing formaldehyde as described by Steck et al. (1980). Sequence Analysis-The sample for NH,-terminal sequence analysis was prepared by incubating 2 nmol of alPI (104 pg) with 5 pmol of cathepsin L, as for the preparation of electrophoresis samples. The trichloroacetic acid-precipitated and acetone-washed sample was dissolved in 50 p l of 50% (v/v) acetic acid and sequenced with an Applied Biosystems gas-phase sequenator Model 470A (Hewick et al., 1981). PTH-derivatives were identified by high performance liquid chromatography (Hunkapillar and Hood, 1983). RESULTSANDDISCUSSION

The incubation of aIPIwith equal molar amounts of cathepsin B did not result in any measurable decrease in elastase-inhibiting activity (Fig. 1). The cathepsin B remained completely active against 2-Arg-Arg-NHMec during the incubation period at pH 5.5. Some cleavageof aIPIwas detected by SDS-polyacrylamide gel electrophoresis of the reaction products (Fig. 2), but this may have been caused by a very small amount of cathepsin L (or some other proteinase) in the cathepsin B preparation. Alternatively, cathepsin B may have cleaved the minor fraction of inactive aIPIpresent or cathepsin B may slowly cleaveaIPIwithout inactivation. The density of the COOH-terminal peptide band in lane 2 is greater than would be expected based on the minor amount of alPIcleaved, but cathepsin B contains a5-kDa light chain (Barrett andKirschke, 1981) that would co-migrate with the aIPICOOH-terminal fragment. These results conflict with the previously reported stoichiometric inactivation of aIPIby cathepsin B (Johnson and Travis, 1977). Cathepsin L was neither a well-recognized enzyme in 1977 nor had it been identified in humans. Consequently, the possibility of cathepsin L contamination in the cathepsin B preparation used in the earlier study was not considered.

1

2

.,,. . _ _ _

14749

3

i

a,-PI

4 /

.

_

'

a,-

5

-Std

kDa 68

c .50

Cat B+

C-ter-e peptide

21 13 6.5

-3

FIG.2. SDS-polyacrylamide gel electrophoresis alp1 of and cysteine proteinase reaction mixtures. Cathepsins B and L and papain were incubated with a l P I for 1 h a t pH 5.5 prior to precipitation, denaturation, reduction, and electrophoresis in a 0.75-mm thick 21% gel. Each sample contained 10 pg of alPI, and theinhibitor:enzyme molar ratios are given in parentheses. Lane I, alPI control; lane 2, alPI and cathepsin B (1:l); lane 3, alPI and papain (200:l); lane 4, alPI andcathepsin L (2009); lane 5,alPI control.

-

,

ZERO INHIB. ACT.

h lW1

ai

80

k

z 0 I-

60

< 2 I-

2 z

40

8 20

0 0

4 4.5 5

5.5 6 6.5 7 7.5 pH

FIG.3. Effect of pH on cathepsin L inactivation of alPI. After incubating alPI with cathepsin L a t a 200:l molar ratio of inhibitor to enzyme at the indicated pH, samples were adjusted to pH 8.0 and made 0.5 p M in E-64 to inhibit cathepsin L. Neutrophil elastaseinhibitory activites were determined, and the data were corrected for pH-dependent losses in the activity of identically treated alPI without cathepsin L.

0

1

1

1

I

I

1

10

20

30

40

50

60

I

MINUTES

FIG.1. Effect of cathepsins

B and L on alp1 inhibitory

activity. Equal molar amounts of cathepsin B and alPI were incubated a t pH 5.5, followed by assay for alPI inhibition of neutrophil Cathepsin L and alPI were incubated at an elastase (W). inhibitor:enzyme ratio of 400:l (mol/mol) a t pH 5.5. Residual neutrophil elastase inhibitory activity is plotted relative to the zero time control ( C - . ) .

When aIPIwas incubated with cathepsin L at a 4001molar ratio of aIPIto cathepsin L, the elastase inhibitory activity of aIPIdecreased to 11%of the control in 1 h (Fig. 1).Thus, cathepsin L catalytically inactivated aIPIin a manner similar to thatof papain (Johnson and Travis, 1977). Electrophoretic analysis (Fig. 2) showed that cathepsin L had releaseda small peptide from aIPI.A sample of aIPIinactivated by papain produced an identical electrophoretic pattern. The pH optimum for the inactivation of aIPIby cathepsin L was in the range pH 5.0-5.5 (Fig. 3), which is consistent with the reported pH optima for cathepsin L activity and stability (Mason et al., 1985). These results indicate that cathepsin L inactivation of alPIwould only occur inan acid environment. The control aIPIlost 75% of its activity after 30 min at pH 4.0, 50% at pH 4.5, 15% at pH 5.0,8% at pH 5.5, and less than 5% at pH 6.0 or above. NH2-terminalsequence analysis of the inactive alp1 yielded three different residues at cycles 1 and 4-7. The PTH-derivatives and their quantity obtained at each cycle are given in

Cathepsin L Inactivates al-Proteinase Inhibitor

14750

Table 1. The predominant amino acid at each cycle corresponded to the sequence of the normal amino terminus of alPI, but two other sequences were also found in almost equal amounts. Only 2 residues were found on cycle 2; and since the yield of PTH-Ile did not decrease, both of the newly formed amino termini musthave isoleucine at cycle 2. Similarly, only PTH-Pro was found on cycle 3, and the high yield indicated that proline must be the third residue in all three sequences. Based on the known sequence of alPI (Carrell et al., 1982), there can be little doubt that cathepsin L cleaved alPI in two places. In addition to cleaving on the carboxyl side of Met358, the serine proteinase inhibitor site,cleavage also occurred on the carboxyl side of G ~ u ~ ~ ~ . Kargel et al. (1980) examined the specificity of cathepsin L on the B-chain of oxidized insulin and found preferential cleavage at positions where Pz and P3 residues, according to the nomenclature of Schechter and Berger (1967), were hydrophobic. The best synthetic substrate for cathepsin L is ZPhe-Arg-NHMec, although Z-Phe-Met-NHMec was also hydrolyzed (Mason et al., 1985). Thus, thereaction of cathepsin L with alPI is generally consistent with the known specificity of the enzyme. Elastase, trypsin, and chymotrypsin, which are inhibited by alPI, all react with the inhibitor at theMet358-Ser359 bond (Johnson and Travis,1978). Reaction with chymotrypsin has TABLE I Amino acid sequence of aIPIinactivated by cathepsin L NHz-terminal amino acids of atPI which had been inactivated by cathepsin L were determined as described under "Experimental Procedures." Up to 3 different residues were found at each cycle.By comparison with the known sequence of alPI (Carrell et al., 1982), it was possible to align the amino acids with three portions of alPI. Yield of PTH-derivative acid at each cycle is given in picomoles. Inactivated alPI Cycle no.

1

2

3

4

5

6

alPI residue no. 1 2 3 4 5 6 Amino acid Glu Asp Pro GlnAla Asp Gly 110 120 pmol 290 210 315" 250 530

52

47

alPI residue no. 354 355 356 357 358 359 360 361 Amino acid Glu Ala Ile Pro Met pmol 69 86 160" 160

Ser

alPI residue no. 358 359 360 361 362 363 364 365 Amino acid Met Ser Ile 65 nmol 154 140

Glu Val

~

Ile

7

7

Pro

Lys

~~~

350

355

1 I1

380

385

A-G-A-M-F-I-E-A-I-P-M-S-I-P-P-E-V-K-F INACTIVATION

I

I

Acknowledgment-We thank Dr. Mark Lively of the Bowman Gray School of Medicine, Oncology Research Center, Wake Forrest University for the amino acid sequence determinations. REFERENCES

Pro Pro

Total recovery of PTH-derivative.

INHIBITION

been shown to result in cleavage of this bond, and the x-ray crystallographic structure of this modified alPI showed that a dramatic structural change must have occurred upon modification because Met35aand S e P 9 were located at opposite ends of the molecule (Loebermann et al., 1984). alPI can now be seen to contain an amino acid sequence encompassing residues 350-358 that is very susceptible to proteolysis by a number of enzymes (Fig. 4). Papain (Johnson and Travis, 1977) anda metalloproteinase from Serratia marcescem (Virca et al., 1982) both catalytically inactivate alPI by cleavage at theinhibitory site. Two other proteinases that catalytically cleave andinactivate a,PI are Crotalus adamanteus snake venom protease 11, which hydrolyzes the Ala350-Met351 peptide bond (Kress et al., 1979), and Pseudom o n m aeruginosa elastase, which cleaves the P r ~ ~ ~ l - M e t ~ ~ ' bond (Morihara et al., 1984). The present report shows that a human enzyme can catalytically cleave alPI in theproteinase-sensitive region, resulting in inactivation of this inhibitor. Although the physiological significance of alPI inactivation by cathepsin L has yet to be determined, it could play a role in the intracellular turnover of alPI since cathepsin L is a lysosomal enzyme. Evidence that theenzyme may also be active outside the cell was given by Etherington (1980), who reported the local extracellular pH around macrophages to be 5.0, and by the observation of Chapman and Stone (1984b) that 60% of the elastinolytic activity of live human macrophages was due to cysteine proteinases. A generally acidic environment in the lung, which would favor cathepsin L activity, is suggested by the reports that fetal lamb alveolar fluids have a pH of 6.3 (Adamson et al., 1969) andthatthecat alveolar surface is at pH 6.6 (Scarpelli, 1977). Cathepsin L inactivation of alPI also would allow neutrophil elastaseto attackelastin. Proteolytically inactivated alPI was recently reported to be chemotactic for neutrophils (Banda et al., 1986), whichwould add to the elastinolytic burden. Thus, cathepsin L not only seems capable of degrading elastin (Mason et al., 1986), but may also contribute to neutrophil elastase activity by inactivation of alPI.

FIG.4. Proteinase cleavagesin the a,PI reactive site region. Elastase,trypsin, chymotrypsin, and otherserine proteinases are inhibited by alPI by reaction at Metz8. Several other metallo and cysteine proteinases inactivatealPI by cleavage at theindicated sites.

Adamson, T. M., Boyd, R. D. H., Platt, H. S. & Strang, L. B. (1969) J.Phy~iol.(Lond.) 204, 159-168 Banda. M. J.. Griffin, G. L.. Clark, E. J. & Senior, R. M. (1986) Fed. Proc; 45, 636 Barrett. A. J. & Kirschke. H. (1981) Methods Enzvmol. 80.535-561 Burnett, D. & Stockley, R. A. (1985) Clin. Sci. 68; 469-474 Carrell, R. W., Jeppson, J. O., Laurell, C. B., Brennan, S. O., Owen, M. C., Vaughan, L. & Boswell, D. R. (1982) Nature 296, 329-334 Castillo, M. J., Nakajima, K., Zimmerman, M. & Powers, J. C. (1979) Anal. Biochem. 99,53-64 Chapman, H. A., Jr. & Stone, 0.L. (1984a)J. Clin. Znuest. 74, 16931700 Chapman, H. A., Jr. & Stone, 0.L. (1984b) Biochem. J. 222, 721728 Ericksson, S. & Laurell, C. B. (1963) Scand. J. Clin. Lab. Inuest. 1 5 , 132-140 Etherington, D. J. (1980) inProtein Degradation in Health and Disease (Evered. D. & Whelan, J., ed) pp. 87-103, Excerpta Medica, Amsterdam Hewick, R.M., Hunkapiller, M. W., Hood, L. E. & Dreyer, W. J. (1981) J.Biol. Chem. 256, 7990-7997 Hunkapiller, M. W. & Hood, L. E. (1983) Methods Enzymol. 91,486493 Johnson, D. & Travis, J. (1977) Biochem. J. 1 6 3 , 639-641 Johnson, D. & Travis, J. (1978) J. Bwl. Chem. 253, 7142-7144 Kargel, H. J., Dettmer, R., Etzold, G., Kirschke, H., Bohley, P. &

Cathepsin L Inactivates al-Proteinase Inhibitor Langner, J. (1980) FEBS Lett. 114,257-260 Kress, L. F., Kurecki, T., Chan, S. K. & Laskowski, M., Sr (1979) J. Biol. Chem. 254,5317-5320 Loebermann, H., Tokuoka, R., Deisenhoffer, J. & Huber, R. (1984) J. Mol. Biol. 177,531-556 Martodam, R. R., Baugh, R. J., Twumasi, D. Y. & Liener, I. E. (1979) Prep. Biochem. 9,15-31 Mason, R. W., Green, G. D. J. & Barrett, A. J. (1985) Biochem. J. 226,233-241 Mason, R. W., Johnson, D. A., Barrett, A. J. & Chapman, H. A. (1986) Biochem. J. 233,925-927 Morihara, K., Tsuzuki, H., Harada, M. & Iwata, T. (1984) J. Biochem. (Tokyo) 95,795-804

14751

Orlowski, M., Orlowski, J., Lesser, M. & Kilburn, K. H. (1981) J. Lab. Clin. Med. 97,467-476 Scarpelli, E. M. (1977) Znt. Anesthesiol. Clin. 1 5 , 19-60 Schechter, I. and Berger, I. (1967) Biochem. Biophys. Res. Commun. 2 7 , 157-162 Smith, C. E.& Johnson, D. A. (1985) Biochem. J. 225,463-472 Steck, G., Leuthard, P. & Burk, R. R. (1980) Anal. Biochem. 1 0 7 , 21-24 Thomas. J. 0. & Kornbere. ", R. D. 11975) . , Proc. Natl. Acad. Sci. U.S. A. 72; 2626 Travis. J. & Johnson. D. (1981) Methods Enzvmol. 80. 754-765 Travis; J. & Salvesen; G. S . (1983) Annu. Reu: Biocheh. 52,655-709 Virca, G . D., Lyerly, D., Kreger, A. & Travis, J. (1982) Biochim. Biophys. Acta 704,267-271