THE JOURNAL OF BIOLOGICAL CHEMISTRY 7 1986 hy The American Society of Biological Chemists, Inc.

Vol. 261, No. 20, Issue of July 15, pp. 9368-9374,1986 Printed in U. S. A.

Comparative Studies of Two Cathepsin B Isozymes from Porcine Spleen ISOLATION, POLYPEPTIDE CHAIN ARRANGEMENTS, AND ENZYME SPECIFICITY*

(Received for publication, October 25, 1985)

Takayuki Takahashi, Satoshi YonezawaS, Abdul H. Dehdarani, and Jordan Tang From the Laboratory of Protein Studies, Oklahoma Medical Research Foundation, and the Department of Biochemistry and Molecular Biology, University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma 73104

Our previous studies on carbohydrate structures of purified porcine spleen cathepsin B indicated that there are two cathepsin B isozymes, each containing a different carbohydrate (Takahashi, T., Schmidt, P. G., and Tang, J. (1984) J. Biol. Chem. 259, 6059-6062). We have now isolated these two enzymes and carried out a comparative study on their structures and en- zymic properties. The major isozyme (CB-I) is a two- chain enzyme (Mr = 28,000) with a light chain (Mr = 5,000) and a heavy chain (Mr = 23,000), whereas the minor enzyme (CB-11) is a single chain enzyme (Mr = 27,000). The NH2-terminal amino acid residues of CB- I were leucine and valine for the light and heavy chain, respectively. However, the NHderminal residue of CB-I1 was not available for automated Edman degra- dation. In addition, peptide mapping experiments in- dicated a difference in the primary structure of these two proteins. Despite such structural differences, they are similar in many enzymic properties. CB-I was more catalytically efficient than CB-I1 toward synthetic sub- strates, except for the substrate benzoyl-L-arginine @- naphthylamide for which the relative catalytic effi- ciency is reversed. Both isozymes degraded glucagon by a dipeptidyl carboxypeptidase activity. Under the same conditions, CB-I was 4-5 times more efficient than CB-11. The results indicate that the cathepsin B isozymes are two separate gene products, but they are similar in enzymic properties.

Cathepsin B is an intracellular proteolytic enzyme which belongs to the group of closely related thiol proteinases in- cluding cathepsins H and L (1, 2 ) . Cathepsin B is generally known as an endopeptidase which has a specificity €or peptide bonds at the carboxyl side of 2 basic residues in synthetic substrates, such as benzyloxycarbonylarginylarginine p- naphthylamide (2,3). The enzyme is also a dipeptidyl carbox- ypeptidase which releases dipeptides sequentially from the carboxyl termini of polypeptides (4-6).

Cathepsin B is generally assumed to be located in the lysosomes because of its acidic pH optimum and of the studies by Shibko and Tapple (7) and Bouma and Gruber (8) who demonstrated that in sucrose density centrifugation the ca-

*This study was supported by Research Grants GM35424 and AM01107 from the National Institutes of Health. The costs of pub- lication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertise- ment” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$ Present address: Dept. of Zoology, Faculty of Science, Hokkaido University, Sapporo 060, Japan.

thepsin B activity was associated with lysosomes. The phys- iological role of this enzyme is thought to be that of the degradation of tissue proteins within the lysosomes. In addi- tion, a cathepsin B or cathepsin B-like enzyme has been postulated to be involved in the in vivo proteolytic processing of protein and hormone precursors (9-13).

Recently, we have purified cathepsin B from porcine spleens and found that this enzyme contains two novel carbohydrate structures drastically different from those found on other lysosomal hydrolases (14, 15). Also, the amino acid sequences near the glycosylation site in the isozymes are homologous but not identical. Thus, the structural evidence would suggest that these are distinct cathepsin B isozymes. Since these isozymes have not been separated from each other, it is not known whether the various enzymic activities attributed to cathepsin B per se are identifiable with the particular iso- zymes. In order to gain insight into the structure and function of these isozymes and their physiological roles, it was inter- esting to separate them and compare their enzymic properties. This paper reports the results of such studies.

EXPERIMENTAL PROCEDURES AND RESULTS’

DISCUSSION

In the present study, we have purified two cathepsin B isozymes from porcine spleens with an activity ratio of 20:l (isozyme CB-1:CB-11) using as substrate Cbz-Ala-Arg-Arg- 4MeOpNA.’ Table I summarizes the results from our com- parative study on the two cathepsin B isozymes. These data suggest that the isozymes are structurally different in many aspects.

The two enzymes were observed to be slightly different in size as determined by gel filtration on a Sephadex G-75 column. The molecular weight of isozyme CB-I was 27,000

Portions of this paper (including “Experimental Procedures,” “Results,” Figs. Sl-S7, and Tables SI-SIII) are presented in mini- print at the end of this paper. Miniprint is easily read with the aid of

the Journal of Biological Chemistry, 9650 Rockville Pike, Bethesda, a standard magnifying glass. Full size photocopies are available from

MD 20814. Request Document No. 85M-3551, cite the authors, and include a check or money order for $8.80 per set of photocopies. Full size photocopies are also included in the microfilm edition of the Journal that is available from Waverly Press.

The abbreviations used are: Cbz-, Na-benzyloxycarbonyl-; MeOpNA, methoxy-6-naphthylamide; ConA, concanavalin A BANA, benzoyl-L-arginine P-naphthylamide; Bz-, benzoyl-; pNA, p-nitroan- ilide; Boc-, butyloxycarbonyl-; ONP, p-nitrophenyl ester; SDS, so- dium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; PTH, phenylthiohydantoin; DTT, dithiothreitol; TPCK, L-l-tos- ylamido-2-phenylethyl chloromethyl ketone; TLCK, N“-p-tosyl-L- lysine chloromethyl ketone; PMSF, phenylmethanesulfonyl fluoride; HPLC, high performance liquid chromatography.

9368

Cathepsin B Isozymes 9369

TABLE I Comparison of molecular properties of cathepsin B isozymes CB-I and CB-II -

CB-I CB-I1 Ref.

No. of polypeptide chains 2

Molecular weight Light chain Heavy chain

28,000 5,000

23,000

Binding to ConA No

Carbohydrate structure Asn-linked

GlcNAc-Asn (31.N

Amino acid composition Similar

NHz-terminal residue Leu in light chain, Val in heavy chain

Specific activity (units/mg) for Cbz-Ala-Arg-Arg-4MeO(3NA 10.8 N-t-Boc-Gln-ONP 8.1

pH optimum for Cbz-Ala-Arg- 6.1

Response to inhibitors Similar

Glucagon degradation by dipeptidyl Faster (4.5 times)

Arg-4MeOpNA

carboxypeptidase action

Aldolase inactivation Faster (2.5 times)

1

27,000,” 33,OOOb

Fig. S2

Fig. S3

Yes Fig. S1

Asn-linked Ref. 14

Man-ManalcNAc-GlcNac-Asn Ref. 14 a 1 6 8 1 4 (31.4 61 N

1011.6 Fuc

Table SI

Not detected (blocked?)

4.0 2.4

6.1

Slower

Tables SI1 and SI11

Fig. S6

Slower Fig. S7 a Value obtained in sodium dodecyl sulfate-polyacrylamide gel electrophoresis.

Value obtained in Sephadex G-75 column chromatography.

and that of isozyme CB-I1 was 33,000. However, for both isozymes, nearly the same values (27,000-28,000) were ob- tained in sodium dodecyl sulfate-polyacrylamide gel electro- phoresis. The discrepancy is the apparent molecular weight of isozyme CB-I1 is probably due to the presence of a larger oligosaccharide unit in its structure (14). We tentatively as- signed the protein molecular weights of the two isozymes to be 27,000-28,000.

Isozyme CB-I consists of two polypeptide chains, a light chain ( M , = 5,000) and a heavy chain (AI, = 22,000). This two-chain cathepsin B structure is similar to the enzyme found in rat (34) and human liver (35). On the other hand, isozyme CB-I1 is apparently a single-chain enzyme. Unex- pectedly, neither the native nor denatured protein produced an NHyterminal amino acid sequence by automated Edman degradation. This suggests that the enzyme contains a NH2- terminal residue whose a-amino group is not available for the automated Edman sequence analysis. At present, it is not clear whether this is a result of a natural blocking group (36) or the presence of terminal pyroglutamic acid formed during the isolation of the enzyme (37).

The amino acid compositions calculated based on these values showed apparent similarity in the enzymes. The pep- tide mapping of tryptic and a-chymotryptic digests, however, revealed differences in the primary structure of these proteins. Recently, we have reported that there is a single amino acid replacement 6 residues away from the carboxyl side of the glycosylated Asn residue in both enzymes (14). Additional difference in the primary structure of the two enzymes is highly probable, as suggested by the results of peptide map- ping experiments. These structural differences appear to ex- clude the possibility that isozyme CB-I1 is a direct precursor of CB-I. I t seems more likely that the two cathepsin B

isozymes are the products of separate genes. Since mammalian tissues are known to contain several

cysteine proteinases, such as cathepsins H and L with similar and somewhat overlapping enzymic properties (1, 2), the specificities of the isozymes have been thoroughly examined. These results, as summarized in Tables J and 11, clearly indicate that both are cathepsin B isozymes. The major sup- porting evidence is as follows: (a) both enzymes have similar specificities toward synthetic substrates, and they only differ from each other quantitatively; (b) the enzymes inactivate rabbit muscle aldolase; and (c ) the enzymes degrade glucagon by a dipeptidyl carboxypeptidase activity.

Despite the noticeable difference in their structures, the two enzymes exhibit as a whole similar enzymic properties. In fact, virtually no difference was observed in their stability, pH activity profile, and inhibition by appropriate inhibitors. The kinetic study (Table 11) with different synthetic sub- strates, however, displayed subtle but significant differences in the enzymes. The study shows that, toward the oligopeptide substrates tested, the catalytic efficiency of CB-I is consist- ently higher than that of CB-11. In contrast, the efficiency is reversed in the case of substrate BANA. The differences in the enzymic action are not surprising since the isozymes are structurally distinguishable. The higher catalytic efficiency of CB-I persists for the hydrolysis of peptide and protein sub- strates. The enzyme inactivates intact aldolase and degrades glucagon approximately 3-4 times faster than CB-11.

The question of whether the isozymes are derived separately from genetically different animals has been examined. We have found that the two cathepsin B isozymes are present in a single porcine spleen with an activity ratio corresponding to that found in the purified enzyme preparation? This ob-

T. Takahashi, A. H. Dehdarani, and J. Tang, unpublished results.

9370 Cathepsin B Isozymes TABLE I1

Kinetic parameters of CB-I and CB-IZ as measured on synthetic peptide substrates CB-I

~

Substrate cb-i1

V- KI" LtIKm v,, K , L? K..t/K" BANA 6.3 2.9 2.9 1.0 3.0 0.83 1.4 Cbz-Ala-Arg-Arg-4MeOpNA

1.7 14 0.13 6.5 50 3.9 0.066 1.8 27

Cbz-Val-Lys-Lys-Arg-4MeOBNA 250 0.20 120 600 290 0.91 130 140 Bz-Pro-Phe-Arg-pNA 7.6 0.43 3.5 8.1 4.0 1.0 1.8 1.8 Bz-Phe-Val-Arg-pNA 3.6 0.070 1.7 24 1.7 0.13 0.76 5.8 N-t-Boc-Gln-ONP 16 0.072 7.5 100 3.8 0.072 1.7 24

a Calculated assuming that 28 mg of protein represents 1 pmol of enzyme. Calculated assuming that 27 mg of protein represents 1 pmol of enzyme

servation indicates that both cathepsin B isozyme genes are present in an individual animal. The physiological signifi- cance for occurrence of the two enzymes is not clear at this time since the individual biological function of the isozymes is not known.

It is interesting to compare our present results with those of Olstein and Liener (33) who, examined in a comparative manner, mouse liver cathepsin B and an analogous thiol proteinase from a transplantable tumor induced by methyl- cholanthrene. CB-I and CB-I1 in this study apparently resem- ble the mouse liver and the tumor enzyme, respectively. In particular, the correlation of specificity toward synthetic sub- strates is remarkable. It is well-established that isozyme com- position patterns change depending on the degree of differ- entiation of tissues. A number of such changes have been reproted for liver enzymes (38). A striking feature of these changes is the appearance of either the fetal liver or whole, early embryo isozymes as the predominant or sole forms in poorly differentiated tissues. These fetal enzymes are known to become predominant in the tumor tissue. These consider- ations lead us to speculate that CB-I may arise from a gene that is expressed in the well-differentiated cells, whereas CB- I1 represents the fetal gene expression taking place in poorly differentiated cells. Indeed, spleen is histologically a complex organ consisting of many different cell types with varying degrees of differentiation.

Acknowledgments-We would like to thank Dr. J. A. Hartsuck for helpful suggestions in preparation of the manuscript. We also wish to thank Azar Fesmire for her able technical assistance.

1.

2.

3. 4.

5.

6. 7. 8.

REFERENCES

Barrett, A. J. (1977) in Proteinase in Mammalian Cells and Tissues (Barrett, A. J., ed) pp. 181-208, Elsevier/North-Hol- land Biomedical Press, Amsterdam

Barrett, A. J., and Kirschke, H. (1981) Methods Enzymol. 80,

McDonald, J. K., and Ellis, S. (1975) Life Sci. 17, 1269-1276 Aronson, N. N., Jr., and Barrett, A. J. (1978) Biochem. J. 171,

Nakai, N., Wada, K., Kobashi, K., and Hase, J. (1978) Biochem.

Bond, J. S., and Barrett, A. J. (1980) Biochem. J. 189, 17-25 Shibko, S., and Tapple, A. (1965) Biochem. J. 95, 731-741 Bouma, J. M. W., and Gruber, M. (1966) Biochim. Bbphys. Acta

535-561

759-765

Bwphys. Res. Commun. 83,881-885

113,350-358

9. Ansorge, S., Kirschke, H., and Freidrick, K. (1977) Acta Biol.

10. Quinn, P. S., and Judah, J. D. (1978) Biochem. J. 172, 301-309 11. Docherty, K., Carroll, R. J., and Steiner, D. F. (1982) Proc. Natl.

12. Loh, Y. P., and Gainer, H. (1982) Proc. Natl. Acad. Sci. U. S. A.

13. Fletcher, D. J., Quigley, J. P., Bauer, G. E., and Noe, B. D. (1981)

14. Takahashi, T., Schmidt, P. G., and Tang, J. (1984) J. Bwl. Chem.

15. Takahashi, T., Dehdarani, A., Schmidt, P. G., and Tang, J. (1984)

16. Schwartz, W. N., and Barrett, A. J. (1980) Bwchem. J. 191,487-

17. Erlanger, B. F., Kokowski, N., and Cohen, W. (1961) Arch.

18. Csoma, C., and Polgar, L. (1984) Bwchem. J. 222, 769-776 19. Lineweaver, H., and Burk, D. (1934) J. Am. Chem. SOC. 56,658-

20. Dixon, M. (1953) Biochem. J. 55, 170-171 21. Takahashi, T., and Tang, J. (1983) J. Bwl. Chem. 268, 6435-

22. Weber, K., and Osborn, M. (1969) J. Biol. Chem. 244,4406-4412 23. Laemmli, U. K. (1970) Nature 227,680-685 24. Andrews, P. (1969) J. Biol. Chem. 244,4406-4412 25. Spackman, D. H., Stein, W. H., and Moore, S. (1958) Anal. Chem.

26. Brauer, A. W., Margolies, M. N., and Haber, E. (1975) Bwchem-

27. Hirs, C. H. W. (1967) Methuds Enzymol. 11, 199-203 28. Carraway, K. L., and Koshland, D. E., Jr. (1972) Methods En-

29. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J.

30. Liu, T.-Y., and Chang, Y. H. (1971) J. Bwl. Chem. 246, 2842-

31. Kornfeld, R., Keller, J., Baenziger, J., and Kornfeld, S. (1971) J.

32. Olstein, A. D., andLiener, I. E. (1983) J. Bwl. Chem. 258,11049-

33. Knight, C. G. (1980) Biochem. J. 189,447-453 34. Takio, K., Towatari, T., Katunuma, N., Teller, D. C., and Titani,

K. (1983) Proc. Natl. Acad. Sci. U. S. A. 80,3666-3670 35. Ritonja, A., Popovic, T., Wiedenmann, K., and Machleidt, W.

36. Tsunasawa, S., and Sakiyama, F. (1984) Methods Enzymol. 106, 165-170

37. Sanger, F., and Thompson, E. 0. P. (1953) Biochem. J. 53, 366- 374

38. Weinhouse, S., Shatton, J. B., and Morris, H. P. (1976) in Cancer Enzymology (Schultz, J., and Ahmad, F., eds) pp. 41-61, Aca- demic Press, New York

Med. Ger. 36,1723-1727

Acad. Sei. U. S. A. 79,4613-4617

79,108-112

J. Cell Bwl. 90, 312-322

259,6059-6062

J. Biol. Chem. 259,9874-9882

497

Biochem. Biophys. 96,271-277

663

6443

30,1190-1206

istv 14,3029-3035

zym01. 27, 616-623

(1951) J. Bioi. Chem. 193,265-275

2848

Bwl. Chem. 246,3259-3268

11056

(1985) FEBS Lett. 181, 169-172

Cathepsin B Isozymes 9371

I8olation, Polypeptide Chain Arrangemnts, and En=- Specificity

0Y

ufut of HI M vztivitv and st~bility.

9372 Cathepsin B Isozymes

A

a h

FRACTION NUMBER

A

0

27K- 23K-

SK-

FRACTION NUMBER

flmr. 52. Chromatopraphv of reduced and car&wwthqlated CB-I and CB-I1 on i Sephsdex C-75 column.

? 0

X

r

3 f

1 I

,o 1.2 1.4 1.6 1

Ve/Vo

Cathepsin B Isozymes 9373

CB-l I

0 10 20 30 4c MINUTES

Flzur. s4. Tryptic peptide MPPIW of CB-I and CB-11 on a XPLC column.

CB- I

L 10 20 30

MINUTES

MINUTES MINUTES

< >e < >d < )C

< >b < >. 1 20 25 a9 HI.-S.r -.........- Gln-A.p-Ph.-V.l-G1"-T~P-L.~-R.t-~."-Th~-COOH

<-><-><-><-> 4 3 2 1

TABLt SI

9374 Cathepsin B Isozymes

TABLE SI1 Effect# of Proteinaaa Inblbitora on CB-I and CB-I1

TABLE SI11 Inhibition Conatants of esveraible Inhibitors

Inhibitors Final Concentration Enzme Activity (X)

Inhibitor of

CB-I CB-I1

NaAaO fodos2etic Acid N-sthylmaleimids Chloroquine Trypsin Inhibitor Trasylol PMSF

TLCK TPCK

Antipain Leupeptin

Elastatinal

1

10

0.05 0.05

1 md.1 0.5 md.1 I 0.005 0.005 0.01 0.002

0.01

16.6 1.6

02.8 38.0

111.9 95.9 56.8 55.6

2.1

2.5 19.0

1.1

27.8

104.5 61.6 5.1

118.2 108.6 83.3 81.3

1.6 3.0

31.9 3.0

8 - 100 /NONE -

w v)

-I a B -I a

B

0 I I I I I I I

0 1 2 3 4

INCUBATION I h 1

I

CB-I CB-I I A B C D E F G H ,

P ’ ”.

92K- 66K- 45K-

31 K-

21 K- 14K-

F i v e 51. Aldolase insctivstion by CB-I and CB-11.

A. Aldolase was incubated with L S O P ~ S CB-I and CB-I1 s.p.rat.1y o r without the s n r m a. demeribed In the Vlethods’. Aldolase activity w u warnured by taking aliquots of the incubation mixture a t different times. The Initial aldol... ectivity Of the mixture warn 3 unltslml under the assay condition^ described in ‘Uorthlngton nanual’. No cathepain B activity 10.. in the mixture wan asan after 6 hour Incubation.

B. Aliquot. of reaction mixtures in A *.re electrophoroaed in 12X polyacryluid. gel in SDS. The mixture. which had basn incubated for 0 h (A,E). 2 h (B.P), 4 h (C.6). and 6 h (D.H). we1 applied.

Inhibitors CB-I CB-I1 tn ) t n ~

Lsup.pti” 9.6 x 10” 4.8 x 10”

Antipain 5.1 x lo4 6.4 x

Elaetatinal 6.0 x 1.2 x 10-4

or premence of inhibitors. Inhibition conetante were obtained with Dixon plots.

Enzyme activity woe meamwed a* in Table SI1 in the absence