Proteolysis of IGFBPs by cathepsin D in vitro and in cathepsin D-deficient mice

~ )

Pergamon

Progress in Growth Factor Research, Vol. 6. Nos. 24, pp. 265-271, 1995 Copyright © 1996 Elsevier Science Ltd. All rights reserved Printed in Great Britain. 0955-2235/95 $29.00 + .00

0955-2235(95)00005-4

PROTEOLYSIS OF IGFBPs BY CATHEPSIN D IN VITRO AND IN CATHEPSIN D-DEFICIENT MICE Thomas Braulke, *t Max Claussen, t Paul Saftig,t Martin Wendland, t Klaus Neifer, t Bernhard Schmidt, t Jt~rgen Zapf, t~ Kurt von Figurat and Christoph Peterst tlnstitute of BiochemistryII, University of G6ttingen, Gosslerstr. 12D, D-37073 G6ttingen, Germany :~Department of Medicine, University Hospital, CH 8091 Ztirich, Switzerland

Affinity-purified lysosomal protease cathepsin D cleaved recombinant human IGFBP-1 to -5 in fragments of defined sizes, while IGFBP-6 was not degraded. To assess the role of cathepsin D for proteolytic processing of lGFBP in vivo, serum from cathepsin D-deficient mice and conditioned media from cathepsin D-deficient fibroblasts and organ explants were analyzed. No differences for the pattern and level of IGFBPs were detected. When conditioned media from fibroblasts were incubated at acid pH, proteolysis of IGFBP-1 and -4 was observed only in media derived from cathepsin D-expressing cells. Additional experiments showed that the proteolysis of IGFBP-4 is mediated by cathepsm D and not by a protease activated by cathepsin D. The IGFBP-4 degrading activities in media from organ explants from cathepsin D-deficient mice were found to be sensitive to inhibitors of aspartyl and cysteine proteases. The data indicate that different classes of acid pH-dependent proteases can contribute to the regulation of IGFBP-4 abundance.

Keywords: IGFBP, proteolysis, cathepsin D, gene-targeting, organ explants. INTRODUCTION Six I G F binding proteins (IGFBPs) have been proposed to be involved in control and coordination of biological actions of IGFs. In the plasma the IGFBPs prolong the half-lives and regulate the transcapillary transport of IGFs, while in the extracellular fluid IGFBPs mediate cell type-specificity and modulate I G F interation with their cellular receptors. The abundance of IGFBPs in the extracellular fluid, however, is determined by (i) tissue and cell-specific transcriptional activity; (ii) the presence of I G F B P binding sites on the cell surface and in the extracellular matrix;

*Correspondenceto: T. Braulke. Acknowledgements--We thank Drs J. Cox (Amgen, Boulder, CO) and A. Sommer and C. Maack (Celtrix, Santa Clara, CA) for kindly providing IGFBP-I and -3. This work was supported by the Deutsche Forschungsgemeinschaft (SFB 402/A6 and Fi 353/10-1) and the Fonds der Chemischen Industrie. 265

T. Braulke et al.

266

and (iii) posttranslational modifications that alter IGFBP affinity [for review see 1, 2]. Regulation on transcriptional level is the best characterized, whereas binding sites for distinct IGFBPs (e.g. proteoglycans and integrins) as well as the characterization of proteases responsible for proteolytic IGFBP processing, have begun to unravel only recently. Thus, IGFBP proteolytic activities have been found in human pregnancy serum and seminal plasma [3-5], as well as in conditioned medium of cultured cells [6, 7]. Additionally, acid-activated IGFBP proteases were identified in media of human and mouse fibroblasts and different carcinoma cell lines [8-10]. One of the proteases involved was determined to be the aspartic proteinase cathepsin D. CATHEPSIN D Cathepsin D is the major aspartic protease of lysosomes and endosomes, while other mammalian members of this gene family such as renin, pepsin, cathepsin E and chymosin (45-49% sequence identity to cathepsin D) are all secretory proteins [11]. The enzymes are synthesized as inactive prepro-forms containing an N-terminal signal sequence that ensures the translocation into the lumen of the endoplasmic reticulum. In contrast to other aspartic proteases, cathepsin D acquires two to three phosphomannosyl residues in its two high mannose-type oligosaccharides that function as lysosomal targeting signals for two mannose 6-phosphate receptors. The receptor-enzyme complexes are transported via clathrin-coated vesicles to an acidic prelysosomal compartment, where the complexes dissociate and proteolytic processing of the 53-kDa cathepsin D precursor to a 47-kDa intermediate is initiated [12]. The final proteolytic maturation, resulting in the formation of the active 31- and 14kDa non-covalently linked polypeptide chains, is completed in lysosomes. Several studies also have described the mannose 6-phosphate receptor-independent lysosomal targeting of cathepsin D, which may involve complex formation with prosaposin forms [13]. Procathepsin D is capable of acid-dependent autoactivation in vitro removing 26 residues to yield an active form, designated pseudocathepsin D, that is not a processing intermediate in vivo [14, 15]. Depending on the cell type studied, the portion of newly synthesized cathepsin D that fails lysosomal sorting and is secreted varies from 2 to 66% [16]. It is proposed that the extent to which the secreted pro-cathepsin D can be activated parallels the degradation of basement membrane, malignant progression in human breast carcinoma and formation of metastasis [17]. In addition, cathepsin D has been reported to be involved in antigen processing [18]. IGFBP PROTEOLYSIS BY CATHEPSIN D I N V I T R O

To determine the IGFBP specificity of mouse cathepsin D, recombinant human ~25I-labeled IGFBP-1 to -6 were incubated with the protease for 20 h at pH 4.0 and the products were analyzed by SDS-PAGE and autoradiography. With the exception of IGFBP-6, all IGFBPs were degraded by cathepsin D. At neutral pH no proteolysis of IGFBP was observed. The main proteolytic fragments formed by cathepsin D were an approximately 12-kDa IGFBP-1, 23- and 18-kDa IGFBP-2,

Cathepsin D-mediated IGFBP Proteolysis

267

15- and 12-kDa non-glycosylated IGFBP-3 and 10-kDa IGFBP-4. However, in several experiments intermediate forms were observed whose appearance seemed to depend on variations in IGFBP : cathepsin D ratio, pH or incubation time. In addition, when glycosylated and non-glycosylated IGFBP-3 were incubated with cathepsin D under identical conditions, the proteolysis of the non-glycosylated IGFBP-3 was almost completed within 20 h, whereas a higher percentage of two glycosylated IGFBP-3 intermediate forms remained, suggesting a protective role of oligosaccharides against proteolysis. The incubation of IGFBP-5 with cathepsin D resulted in the formation of several fragments of 14-23 kDa (Fig. 1). I N VIVO ACTIVITIES OF CATHEPSIN D

To assess the role of cathepsin D in IGFBP degradation in vivo, IGFBP levels were studied in serum and conditioned media from tissue explants as well as IGFBP degrading activities in the conditioned media of cathepsin D-deficient cells or organ explants. Disruption of the cathepsin D gene neither results in embryonic lethality nor in phenotypic or behaviour abnormalities during the first two postnatal weeks. Following the progressive development of an atrophy of the ileal mucosa and loss of lymphoid cells in the third week, cathepsin D-deficient mice die 25-27 days after birth [19]. The level and pattern of IGFBPs in serum from 12-, 16- and 23day-old homozygous mutant animals were similar to those of wild-type mice. In conditioned media from fibroblasts or explants of lung, liver, spleen, kidney or

[1251 ]- IGFBP- 4

[ 1251 ]- IGFBP- 5

BP-4 I BP-5

pH 4.0 Cathepsin D Cath - D mice

-

+

-

+/+-/-

_

-

pH 7.4 +

.

.

.

+/+-/-

.

+/+

+/+

FIGURE 1. IGFBP-4 and -5 proteolytic activity of cathepsin D and in fibroblast-conditioned media: 125Ilabeled rhlGFBP-4 and -5 (each 40,000 c.p.m.) were incubated in the absence (-) or presence (+) of 0.8 ~g m-cathepsin D or in conditioned media from wild-type (+/+) and cathepsin D-deficient (-/-) fibroblasts for 20 h at 37°C at the indicated pH. The reaction products were separated by SDS--PAGE and visualized by autoradiography.

T. Braulke et al.

268

thymus tissue of 12-23-day-old mice, organ-specific IGFBP patterns were found, but no differences between media derived from control and cathepsin D-deficient mice were detected. During cell-free incubation of conditioned media from both wild-type and cathepsin D-deficient fibroblasts with recombinant human 125I-labeled IGFBPs for 20 h at 37°C at pH 7.4, only proteolysis of IGFBP-5 was noted, yielding 22- and 15-kDa fragments (Fig. 1). When IGFBPs were incubated with conditioned media at pH 4.0 no proteolysis of IGFBP-2 and -6 was observed, while IGFBP-3 (not shown) and -5 (Fig. 1) were degraded whether the media were derived from wild-type or cathepsin D-deficient fibroblasts. However, the proteolysis of IGFBP-1 and -4 was observed only in media from cathepsin D-expression cells. Cleavage of IGFBP-1 resulted in the formation of a 12-kDa fragment (not shown). The same fragment was also produced at acid pH by conditioned media from human embryonic skin fibroblasts [20]. Acid pH-dependent IGFBP-4 proteolysis yielded 17- and 10-kDa peptide fragments (Fig. 1). When conditioned media from tissue explants, e.g. liver or thymus, were incubated with [125I]IGFBP-4 at pH 4.0, 17- and 10-kDa fragments were formed irrespective of the expression of cathepsin D in the tissues (Fig. 2). While the IGFBP-4 protease activity in these media was sensitive to inhibitors of aspartyl and cysteine proteases (i.e. 200/./M pepstatin and 5 ]./M E64, respectively), the proteolysis of IGFBP-4 by fibroblast derived medium was inhibited only by pepstatin. These results indicate that different classes of proteases contribute to acid pH-dependent IGFBP-4 proteolysis in cell-free media from several tissue explants.

Cathepsin D Conditioned medium Cath-D mice

+

Fibroblasts +'+l

"/"

Liver +/+ I "/"

Thymus ~-/+

-/-

FIGURE 2. IGFBP-4 protease activity in conditioned medium from fibroblasts and tissue explants of wildtype and cathepsin D-deficient mouse: ]251-1abeledIGFBP-4 was incubated for 20 h at 37°C at pH 4.0 in the absence or presence of 48-h conditioned medium from fibroblasts, liver or thymus explants of wild-type (+/+) or cathepsin D-deficient mouse (-/-). For comparison the p2Sl]IGFBP-4 was incubated at pH 4.0 with 0.5 pg purified m-cathepsin D. Samples were analyzed by SDS-PAGE and autoradiography.

Cathepsin D-mediated IGFBP Proteolysis CaIh-D mice

+I+

Cathepsin D P e p s t a t i n during Preincubation

269

-I-

-I-

-I-

-I-

-

-

+

+

+

-

-

-

+

-

-I

I-

-

+

+

_

+

-

-

......

Preinculoetlon 2h at

pH

1

2

4.0

3

pH

4

5

7.4

6

pH

7

4.0

8

FIGURE 3. Cathepsin D-mediated proteolysis of IGFBP-4 in acidified conditioned media from fibroblasts IzSI-labeled IGFBP-4 was incubated for 20 h at 37°C at the indicated pH in the absence (lane 8) or presence of 48-h conditioned medium from fibroblasts of wild type (+/+, lane 1) or cathepsin D-deficient (-/-, lane 2) mouse. Lanes 3-7 represent reaction products after a 2-h preincubation of medium from cathepsin D-deficient fibroblasts at the indicated pH with 1 ~g m-cathepsin D in the absence (lanes 3, 5 and 7) or presence of 100/JIM pepstatin (lanes 4 and 6). After addition of pepstatin to samples 3-6 (final concentration 200/IM) the incubation was continued for 20 h followed by analysis by SDS-PAGE and autoradiography.

To differentiate whether IGFBP-4 proteolysis in fibroblast media is catalyzed directly by cathepsin D or whether cathepsin D is required for activation of other protease(s), the following experiment was carried out. Media from cathepsin D-deficient fibroblasts that lack IGFBP-4 degrading activity were preincubated for 2 h at 37°C at pH 4.0 in the absence or presence of cathepsin D. The cathepsin D-containing mixture was split and one half received pepstatin (200 pM final concentration) to inhibit cathepsin D activity. All samples were then incubated for 16 h in the presence of p25I] IGFBP-4. Proteolysis of IGFBP-4 was observed only in the sample that was substituted with cathepsin D and did not receive pepstatin. This clearly demonstrates that IGFBP-4 is degraded directly by cathepsin D and not by a protease activated by cathepsin D. CONCLUSION The formation of IGFBP fragments with reduced affinities for IGFs by IGFBP proteases may play an important role in regulation of bioavailability of IGFs. In various body fluids and media of cultured cells IGFBP proteases have been detected, including plasmin, matrix metalloproteinases and cathepsin D. While the function of the aspartic proteinase cathepsin D in lysosomal degradation of intracellular proteins is well established, its role in degradation of extracellular matrix and soluble proteins under physiological conditions is unclear. Cathepsin D meets

270

T. Braulke et al.

some criteria for being a physiological IGFBP protease: cell type dependent variations in the amount of secreted pro-cathepsin D and in vitro cleavage of four IGFBPs (-1 to -4) by purified cathepsin D has been demonstrated, yielding proteolytic fragments resembling those present in serum or media from cultured cells. In addition, conditioned media derived from fibroblasts of cathepsin D gene-disrupted mice lost their ability to cleave IGFBP-4 at acidic pH. However, neither differences in the amount and pattern of IGFBPs accumulated in 48-h conditioned medium of fibroblasts or organ explants, nor in the serum, were observed between wild-type and cathepsin Ddeficient mice. Furthermore, studies on IGFBP-4 proteolysis by media of cathepsin Ddeficient organ explants demonstrated that different classes of acid pH-dependent proteases, especially cysteine proteinases, can degrade IGFBP and therefore may compensate for the loss of cathepsin D. Whether the expression of cathepsin D in mouse leads to an accumulation of proteolytic IGFBP fragments with IGF-independent biological functions and the role of cathepsin D in tissue-specific proteolysis processes in response to metabolic changes remain to be investigated. REFERENCES 1. Jones JI, Clemmons DR. Insulin-like growth factors and their binding proteins: biological actions. Endocrine Rev. 1995; 16: 3-34. 2. Zapf J. Physiological role of the insulin-like growth factor binding proteins. Eur J Endocrinol. 1995; 132: 645-654. 3. Hossenlopp P, Segovia B, Lassarre C, Roghani M, Bredon M, Binoux M. Evidence of enzymic degradation of insulin-like growth factor-binding proteins in the 150 K complex during pregnancy. J Clin Endocrinol Metab. 1990; 71: 797-805. 4. Claussen M, Zapf J, Braulke T. Proteolysis of insulin-like growth factor binding protein-5 by pregnancy serum and amniotic fluid. Endocrinology 1994; 134: 1964-1966. 5. Cohen P, Graves HCB, Peehl DM, Kamarei M, Guidice LC, Rosenfeld RG. Prostate-specific antigen (PSA) is an insulin-like growth factor binding protein-3 protease found in seminal plasma. J Clin Endocrinol Metab. 1992; 75: 1046-1053. 6. Conover CA, Kiefer MC, Zapf J. Posttranslational regulation of insulin-like growth factor binding protein-4 in normal and transformed human fibroblasts. J Clin Invest. 1993; 91:1129-1137. 7. Fielder PJ, Pham H, Adashi EY, Rosenfeld RG. Insulin-like growth factors (IGF) block FSHinduced proteolysis of IGF-binding protein-5 (BP-5) in cultured rat granulosa cells. Endocrinology 1993; 133: 415-418. 8. Conover CA, De Leon DD. Acid-activated insulin-like growth factor-binding protein-3 proteolysis in normal and transformed cells. J Biol Chem. 1994; 269: 7076-7080. 9. Conover CA, Perry JE, Tindall DJ. Endogenous cathepsin D-mediated hydrolysis of insulin-like growth factor-binding proteins in cultured human prostatic carcinoma cells. J Clin Endocrinol Metab. 1995; 80: 987-993. 10. Claussen M, Buergisser D, Schuller AGP, Matzner U, Braulke T. Regulation of insulin-like growth factor (IGF)-binding protein-6 and mannose 6-phosphate/IGF II receptor expression in IGF lIoverexpressing NIH 3T3 cells. Mol Endocrinol. 1995; 9: 902-912. 11. Tang J, Wong RNS. Evolution in the structure and function of aspartic proteases. J Cell Biochem. 1987; 33: 135-145. 12. Hasilik A. The early and late processing of lysosomal enzymes: proteolysis and compartmentation. Experientia 1992; 48: 130-151. 13. Zhu Y, Conner GE. Intermolecular association of lysosomal protein precursors during biosynthesis. J Biol Chem. 1994; 269: 3846-3851. 14. Hasilik A, von Figura K, Conzelmann E, Nehrk6rri H, Sandhoff K. Eur J Biochem. 1982; 125: 317-321. 15. Richo GR, Conner GE. Structural requirements of procathepsin D activation and maturation. J Biol Chem. 1994; 269: 14,806-14,812.

Cathepsin D-mediated IGFBP Proteolysis

271

16. Capony F, Braulke T, Rougeot C, Roux S, Montcourrier, Rochefort H. Specific mannose-6-phosphate receptor-independent sorting of pro-cathepsin D in breast cancer cells. Exp Cell Res. 1994; 215: 154-160. 17. Rochefort H. Biological and clinical significance of cathepsin D in breast cancer. Semin Cancer Biol. 1990; 1: 153-160. 18. Marie MA, Taylor MD, Blum JS. Endosomal aspartic proteinases are required for invariant-chain processing. Proc Natl Acad Sci USA. 1994; 91: 2171-2175. 19. Saftig P, Hetman M, Schmahl W, Weber K, Heine L, Mossmann H, Ktister A, Hess B, Evers M, von Figura K, Peters C. Mice deficient for the lysosomal proteinase cathepsin D exhibit progressive atrophy of the intestinal mucosa and profound destruction of lymphoid cells. EMBO J. 1995; 14: 3599-3608. 20. Braulke T, G/Stz W, Claussen M. lmmunohistochemical localisation of insulin-like growth factor binding protein-l, -3 and -4 in human fetal tissues and their analysis in media from fetal tissue explants. Growth Regul. (In press).