- Email: [email protected]
Characterization of Cleavage Enzymes for Sterol Regulatory Element Binding Protein in Hamster Liver Microsomes
Biochemical and Biophysical Research Communications 258, 542–547 (1999) Article ID bbrc.1999.0679, available online at http://www.idealibrary.com on
Characterization of Cleavage Enzymes for Sterol Regulatory Element Binding Protein in Hamster Liver Microsomes Mineko Yamaguchi,* ,1 Heigoro Shirai,* ,1 Ryuichiro Sato,† Yoshiki Kawabe,‡ Rie Fukuda,* Tatsuhiko Kodama,* and Takao Hamakubo* ,2 *Department of Molecular Biology and Medicine, Research Center for Advanced Science and Technology, University of Tokyo, Meguro, Tokyo 153-8904, Japan; †Department of Biochemistry and Molecular Biology, Graduate School of Pharmaceutical Sciences, Osaka University, Suita, Osaka 565-0871, Japan; and ‡Fuji-Gotemba Research Laboratories, Chugai Pharmaceutical Co., Ltd., Gotemba, Shizuoka 412-8513, Japan
Received March 30, 1999
Sterol regulatory element binding proteins (SREBP-1 and SREBP-2) are the key transcription factors for the regulation of the cellular cholesterol level.To identify proteolytic enzymes for SREBPs, a fluorogenic peptide substrate, MOCAc-GRSVLSFK(Dnp)rr-NH 2, was synthesized according to the proposed cleavage site of human SREBP-2. In microsome fractions from hamster liver, we found a peptidase activity inhibitable by the synthetic inhibitor Ac-GRSVL-aldehyde with an IC 50 of 40 nM. This peptidase separated into three peaks of approximately 400 kDa, 60 kDa, and 30 kDa (Mp400, Mp60 and Mp30 respectively) upon gel permeation chromatography. Mp30 was purified to apparent homogeneity with an M r of 32 kDa. The partial amino acid sequence of Mp30 possessed homology to cathepsin B (EC 3.4.22.1). A 109 kDa protein band on SDS-PAGE which corresponded to Mp400 exhibited homology to neprilysin (EC 3.4.24.11) in partial amino acid sequence. These findings suggest several degradative pathways for SREBP in liver microsome membranes. © 1999 Academic Press Key Words: SREBP; protease; endoplasmic reticulum; cholesterol; cathepsin B; neprilysin.
Sterol regulatory element binding proteins (SREBP-1 and SREBP-2) play a major role in the transcriptional control of sterol biosynthetic enzymes, including 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase, HMG-CoA synthase and low density lipoprotein (LDL) receptor (1). SREBPs are synthesized as 1
These authors contributed equally to this work. To whom correspondence and reprint requests should be addressed at Research Center for Advanced Science and Technology, University of Tokyo, 4-6-1 Komaba, Meguro-ku, Tokyo, 153-8904, Japan. Fax: (181) (3) 3481-4552. E-mail: [email protected]. 2
0006-291X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
precursors in the endoplasmic reticulum (ER) membrane and when the intracellular sterol level decreases, are cleaved by proteases inside the ER lumen at the sequence termed “site 1” (2). SREBPs cleaved at site 1 are successively cleaved by a second protease named site 2 protease (3), which recognizes the amino acid sequence within the transmembrane domain. The mature forms of SREBPs released from the ER membrane translocate to the nucleus and bind to sterolregulatory elements (SREs) in the promoter region of genes encoding enzymes related to cholesterol biosynthesis and the LDL receptor (4, 5). In the course of purification of the SREBP processing enzymes, two apoptosis-related cysteine proteases have been identified as being able to cleave and activate SREBPs (6, 7). To characterize the cleavage activity in the ER, we exploited an in vitro assay system using quenched fluorogenic peptide substrate (8) corresponding to eight amino acid residues of the site 1 sequence in the ER lumen portion of human SREBP-2. In microsome fractions of the hamster liver, we found cleavage activity which could be inhibited by a sequence specific inhibitor, acetyl-Gly-Arg-Ser-Val-Leualdehyde. When the hamster liver microsome proteins were solubilized and applied to gel permeation chromatography, the activity was resolvable into three different fractions. Here we characterize these peptidases, which cleave the sequence at a point corresponding to site 1 of human SREBP-2. MATERIALS AND METHODS Peptides and reagents. An internally quenched fluorogenic peptide substrate, (7-methoxycoumarin-4-yl)acetyl-Gly-Arg-Ser-Val-LeuSer-Phe-Lys-2,4-dinitrophenyl-Arg-Arg-amide (MOCAc-GRSVLSFK(Dnp)rrNH 2), the reference compound for fluorescence intensity, MOCAc-Pro-Leu-Gly, and a peptide inhibitor, acetyl-Gly-Arg-Ser-ValLeu-aldehyde (Ac-GRSVL-CHO), were synthesized by the Peptide In-
542
Vol. 258, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
stitute (Osaka, Japan) and shown by HPLC to be greater than 98% pure. These sequences were deduced from the proposed cleavage site of the endoplasmic reticulum lumen portion of human SREBP-2 (2). Other reagents were obtained from Sigma except as otherwise stated. Assay of cleavage activity. Reaction mixtures, which contained 50 mM Hepes pH 7.4, 1 mg peptide substrate, and enzyme solution, were monitored continuously in a spectrofluorometer at an excitation wavelength of 328 nm and an emission wavelength of 393 nm. One unit of enzyme activity was tentatively determined as 1 pmol of MOC-peptide (fluorescence intensity was determined from a standard peptide, MOCAc-Pro-Leu-Gly) liberated per min at room temperature. Preparation of liver microsome fraction. Under Nembutal anesthesia, male Golden Syrian hamsters aged 9 to 10 weeks were perfused from the portal vein with 50 ml of ice-cold homogenizing buffer (10 mM Hepes, containing 1 mM MgCl 2, 10 mM KCl, 1 mM DTT, 0.25 M sucrose, pH 7.5). The livers were excised and homogenized with ninefold volume of ice-cold homogenizing buffer, and then centrifuged at 10,000g for 20 min. The supernatant was centrifuged at 105,000g for 1 hour. The supernatant after 105,000g centrifugation was used as the cytosol fraction, and the pellet was resuspended with buffer A (50 mM Tris-HCl, 1 mM EDTA, 1 mM DTT, 10% glycerol, pH 7.5) and used as the microsome fraction. Purification procedures. To solubilize the membrane peptidases, a final 0.5% (w/v) MEGA9 (Dojindo, Japan) or 0.5% (w/v) lubrol (Lubrol PX, Sigma) was added to the microsome fractions. After 1 hour mixing with rotator at 4°C, the solubilized fractions were separated from membranes by centrifugation at 105,000g for 1 hour. For the purification of Mp30, about 100 ml of the MEGA9 solubilized fraction derived from five hamster livers was loaded on to an anion-exchange column (Q-sepharose HiLoad 26/10, Amersham Pharmacia Biotech), and after washing with buffer A, containing 0.5% MEGA9, the column was eluted with a linear gradient to 0.5 M NaCl. One fractional peak was collected and applied to a superdex 200 (HiLoad 26/60, Amersham Pharmacia Biotech) equilibrated with buffer A containing 0.2% MEGA9, and fraction 21, which corresponded to Mp30, was collected and stored at 280°C. The same fractions from 8 batches were combined and applied to a Mono Q (HR 5/5, Amersham Pharmacia Biotech) column equilibrated with buffer A containing 0.2% MEGA9. The active fractions, which eluted at around 0.12 M NaCl with a linear gradient, were collected and dialyzed against 10 mM phosphate buffer containing 10% glycerol and 0.25% MEGA9, pH 6.8. The dialyzed fractions were applied to a hydroxyapatite column (SH-0410F, Pentax) with a linear gradient elution from 10 mM to 400 mM phosphate buffer containing 10% glycerol and 0.25% MEGA9, pH 6.8. Partial amino acid sequence analysis was performed by APRO Science Incorporation (Tokushima, Japan). For the purification of Mp400, microsome fractions from 5 hamster livers were solubilized with buffer A containing 0.5% (w/v) lubrol, and then loaded on to Q-sepharose equilibrated with buffer A containing 0.5% lubrol. The peptidases were eluted with a linear gradient to 0.5 M NaCl. Then collected fractions were applied to 5 ml Blue sepharose (HiTrap Blue, Amersham Pharmacia Biotech) equilibrated with buffer A containing 0.25% lubrol and eluted with a 0-2 M NaCl linear gradient. The active fractions around 1 M NaCl were collected and applied to a superdex 200 equilibrated with buffer A supplemented with 0.25% lubrol and 0.2 M NaCl. The active fractions from four batches were pooled and applied to a 1 ml concanavalin A affinity column (Con A Sepharose, Amersham Pharmacia Biotech) equilibrated with buffer B (20 mM Tris, 1 mM DTT, 5% glycerol, 0.1% lubrol, pH 7.4) containing 0.5 M NaCl. After washing with ten column volume of buffer B, containing 0.5 M NaCl, the column was loaded with buffer B supplemented with 0.5 M NaCl and 0.1 M a-methyl-D-mannoside. The column was allowed to stand for 1 hour at 4°C, and then the peptidase was eluted with the same buffer. After dialysis against buffer B, the aliquot was applied to a Mono Q
FIG. 1. (A) Dose-dependent cleavage of peptide substrate by microsome fractions after freeze-thaw. The peptide substrate was incubated with 10 ml of the microsome fraction (open circles), 20 ml of the microsome fraction (filled circles), 10 ng trypsin (open squares), or without enzyme (open triangles). The amount of liberated peptide was calculated from the increase of fluorescence intensity divided by the reference compound. (B) Inhibition by peptide inhibitor, AcGRSVL-CHO, of cleavage activity in freeze-thawed liver microsome fraction.
column equilibrated with buffer B, and eluted with a 0 – 0.5 M NaCl linear gradient.
RESULTS Peptidase Activity in Hamster Liver Microsome Fractions Peptidase activity was found in both the cytosol and the microsome fractions prepared from hamster liver. The microsome fraction exhibited peptidase activity after freeze-thaw (Fig. 1A), and this activity was inhib-
543
Vol. 258, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. The elution profile of cleavage activities from hamster liver microsome on superdex 200 gel permeation chromatography. The activities were separated into three major peaks named Mp400 (a), Mp60 (b), and Mp30 (c).
itable by a peptide inhibitor, Ac-GRSVL-CHO, with an IC 50 of 40 nM (Fig. 1B), whereas peptidase in cytosol or trypsin was not inhibited by the peptide inhibitor even at 10 mM (data not shown). Since the proposed site 1 protease cleavage occurs between the leucine 522 and serine 523 bond (2), the peptidase activity in the freezethawed microsome membrane was thought to be specific for the SREBP-2 ER loop peptide. Sixty percent of the total activity was recovered by solubilization with 0.5% MEGA9, and this solubilized activity proved to be stable for several weeks when stored at 280°C. Solubilization and Purification of the Peptidases from Microsome Membrane After centrifugation at 105,000g for 1 hour, the supernatant was collected and applied to a Q-sepharose anion exchange column. There was no significant activity in the flow through fractions, and the total recovery of the activity from the column was around 50%. The active fractions were widely distributed in a range from 50 mM to 0.5 M NaCl with one clear peak at fraction 14, which eluted at around 0.12 M NaCl (data not shown). When fraction 14 was applied to gel permeation chromatography (superdex 200) the activity separated into three major peaks at approximately 400 kDa, 60 kDa, and 30 kDa (Fig. 2), named Mp400, Mp60 and Mp30, respectively. Among these, Mp30 was inhibited 90% with 1 mM of the peptide inhibitor, AcGRSVL-CHO, but Mp400 and Mp60 exhibited no significant inhibition at the same dose. On the other hand, Mp60 exhibited a two fold increase in activity when 100 mM MnCl 2 was added, whereas neither Mp30 nor Mp400 showed any increase of activity to in response to
any divalent cation tested (data not shown). Thus Mp60 was considered to be a different peptidase from Mp30 and Mp400. Mp30 was further purified by a Mono Q column and a 32 kDa band on SDS-PAGE corresponded to the peptidase activities. After hydroxyapatite column chromatography, Mp30 was purified to apparent homogeneity on SDS-PAGE with a M r of 32 kDa (Fig. 3A). Since the Mp400 activities were much more abundant when the microsome was solubilized with 0.5% lubrol compared with 0.5% MEGA9, we performed the purification of Mp400 using lubrol solubilization. Solubilized fractions were applied to an anion exchange column, Q-sepharose. The characteristic feature of Mp400 was that it required ATP after separation by column chromatography. After the anion exchange column, the cleavage activities were increased about five fold when 1 mM ATP was added to the reaction mixture. After a second chromatography with a bluesepharose column, the cleavage activities were found to be measurable only when the reaction mixture contained ATP or its analogues, such as ATP-g-S or dATP (Table 1). Mp400, which requires ATP or its analogues, was not inhibited significantly by the peptide inhibitor even at 10 mM. After the Mono Q column, the peptidase activity corresponded well with the 109 kDa protein band on SDS-PAGE. This protein band was excised from the gel and the partial amino acid sequences were determined. Inhibitor Spectrum of Mp30 and Mp400 Peptidases Using the gel permeation chromatography fractions, the effects of various protease inhibitors were tested
544
Vol. 258, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. (A) SDS-PAGE of purified Mp30. Lane a, molecular size marker; lane b, active fraction from hydroxyapatite column. (B) Comparison of amino acid sequence of 32 kDa protein from hamster with primary structure of mouse cathepsin B precursor (17). The determined partial sequences are underlined. The different residues between hamster and mouse are marked with asterisk (see text). The precursor protein is processed at the peptide bond between Asp 79 and Leu 80 to generate a single chain cathepsin B. Hatched portions indicate the light and heavy chains of the two chain form of cathepsin B.
(Table 2). Mp30 was susceptible to thiol protease inhibitors such as leupeptin or acetyl-leucyl-leucylnorleucinal (ALLN). Mp400 was resistant to serine or thiol protease inhibitors and was strongly inhibited by phosphoramidon or thiorphan (Table 2). Amino Acid Sequence Analysis From the purified Mp30 fraction, the amino acid sequences of three fragments of the 32 kDa protein were determined to be LPETFDARK, GLVSGGLYNS, and EIMAEIYK. These sequences are identical to
TABLE 1
Effect of Various Reagents on Cleavage Activity of Mp400 Reagents ATP ADP dATP ATP-g-S cAMP Actyl CoA
Concentration 2 2 2 1 2 1
mM mM mM mM mM mM
Cleavage activity (%) 100 43 175 96 10 59
Note. The activity with 2 mM ATP was taken as 100%.
mouse cathepsin B except for glutamate 88 to lysine, and 172 valine to leucine (Fig. 3B). Also, the determined amino acid sequences of the peptides derived from the 109 kDa protein band were DGDLVDWWTQ, LLPGIDLNHK, and LLPDIYGWPV. These sequences are identical to the internal sequence of mouse neprilysin. DISCUSSION We have developed an in vitro assay system for SREBP-2 ER loop cleaving enzymes, using the peptide substrate designed for the amino acid sequence around the site 1 protease cleavage site. In hamster liver microsomes, we found a peptidase which was inhibitable by Ac-GRSVL-CHO. Subsequently, from the MEGA 9 solubilized fraction of these microsomes, we purified 32 kDa single chain cathepsin B. Cathepsin B is believed to be synthesized in a precursor form (339 amino acids in the mouse), and processed to the 32 kDa single chain form in the ER by autocatalysis and/or by certain processing enzymes (9). During the sorting from the endoplasmic reticulum to the lysosome, single chain cathepsin B is processed further to generate a two chain enzyme (10). It is known that active cathepsin B dis-
545
Vol. 258, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 2
Effect of Reagents on the MOC-Ac-GRSVLSFK(Dnp)rrNH 2 Cleaving Activities from Gel Permeation Chromatography Remaining activity (%) Reagents Pefabloc Leupeptin E-64-c EDTA Lactacystin ALLN
Ac-GRSVL-CHO Phosphoramidon Thiorphan
Concentration
Mp400
Mp30
2 mM 200 mg/ml 100 mg/ml 10 mM 10 mM 52 mM 5.2 mM 0.52 mM 0.052 mM 1 mM 100 mg/ml 2 mM
106 124 91 22 113 132 N.D. N.D. N.D. 87 6 3
100 27 N.D. N.D. N.D. 34 40 54 92 2 N.D. N.D.
Note. The activities without reagents were taken as 100%. The activities of Mp400 were measured under the presence of 1 mM ATP. N.D., Not determined.
tributes not only to the lysosome but also to the ER and nuclear membranes (11). Since we did not find any corresponding band to the two chain form on SDSPAGE, the dominant form in our microsome fraction was taken to be the 32 kDa single chain Cathepsin B (Fig. 3). Cathepsin B is postulated to be involved in antigen presentation, amyloid precursor protein cleavage in Alzheimer’s disease, metastasis of malignant cells, and the degradation of foreign or intracellular proteins (12). The SREBPs are also known to undergo rapid degradation, which is inhibitable by ALLN (4). From the result that Mp30 could be inhibited by ALLN (Table 1), it is possible that cathepsin B is involved in the degradation of SREBPs. Another peptide substrate cleaving activity fragment found in the microsome fraction had a large molecular weight of around 400 kDa on gel permeation chromatography. Mp400 peptidase required ATP or its analogues for this activity. Since ATP-g-S or dATP also activated the cleavage to a similar extent as ATP, the activity was considered not to be energy dependent. Since the requirement of ATP appeared after column chromatography, it was thought that some molecule acting as a cofactor might have been separated. In partially purified Mp400, we did find a 109 kDa protein on SDS-PAGE which corresponded to this activity and the partial amino acid sequence turned out to be identical to the transmembrane metallo-endopeptidase, neprilysin. Neprilysin, neutral endopeptidase-24.11 (EC 3.4.24.11), also known as enkephalinase or CD10/ common acute lymphoblastic leukemia-associated antigen (CD10/CALLA), cleaves peptide bonds on the amino side of hydrophobic amino acid residues and is sensitive to phosphoramidon (13). It is not known whether neprilysin requires ATP or similar com-
pounds. Since the result of the inhibitor spectrum (Table 2) of Mp400 was compatible to that of neprilysin, we have concluded that the major active component of Mp400 should be attributed to neprilysin and that ATP may be required for the dissociation of the inhibitory components. As neprilysin is considered to play a role in degrading peptide hormones (13, 14), it may also degrade SREBPs on the membrane. Recently, Sakai et al. identified membrane-bound subtilisin like site 1 protease (S1P) by using an in vivo assay system for the SREBP-2 ER loop. S1P restored site 1 cleavage in a mutant CHO cell line lacking S1P (15). One reason that we did not find a significant activity fragment sensitive to serine protease inhibitors may be due to the requirement of SCAP (SREBP Cleavage-activating Protein) for the maximum activity of S1P (16). In conclusion, we have characterized SREBP ER loop cleavage activities in hamster liver microsomes. This activity, which could be inhibited by a sequence specific inhibitor, was found to be cathepsin B in single chain form. There were two other non-specific peptidases, one with a molecular weight of 60 kDa, and another, Mp400, which was associated with neprilysin activity. The susceptibility of the ER loop to these enzymes suggests that SREBPs undergo breakdown upon interaction with them. ACKNOWLEDGMENTS This work was supported by a grant from the “Research for the Future” Program from the Japan society for the Promotion of Science and by the Ministry of Education, Grant-in-Aid for Scientific Research (B), 08457258. The authors are grateful to Dr. K. Boru for reviewing the manuscript.
REFERENCES 1. Brown, M. S., and Goldstein, J. L. (1997) Cell 89, 331–340. 2. Duncan, E. A., Brown, M. S., Goldstein, J. L., and Sakai, J. (1997) J. Biol.Chem. 272, 12778 –1285. 3. Rawson, R. B., Zelenski, N. G., Nijhawan, D., Ye, J., Sakai, J., Hasan, M. T., Chang, T. Y., Brown, M. S., and Goldstein, J. L. (1997) Mol. Cell 1, 47–57. 4. Wang, X., Sato, R., Brown, M. S., Hua, X., and Goldstein, J. L. (1994) Cell 77, 53– 62. 5. Smith, J. R., Osborne, T. F., Goldstein, J. L., and Brown, M. S. (1990) J. Biol. Chem. 265, 2306 –2310. 6. Hua, X., Sakai, J., Brown, M. S., and Goldstein, J. L. (1996) J. Biol. Chem. 271, 10379 –10384. 7. Pai, J-T., Brown, M. S., and Goldstein, J. L. (1996) Proc. Natl. Acad. Sci. USA 93, 5437–5442. 8. Knight, C. G., Willenbrock, F., and Murphy, G. (1992) FEBS Lett. 296, 263. 9. Podobnik, M., Kuhelj, R., Turk, V., and Turk, D. (1997) J. Mol. Biol. 271, 774 –788. 10. Kominami., E., Tsukahara, T., Hara, K., and Katunuma, N. (1988) FEBS Lett. 231 225–228. 11. Spiess, E., Bru¨ning, A., Gack, S., Ulbricht, B., Spring, H., Trefz, G., and Ebert, W. (1994) J. Histochem. Cytochem. 42, 917–929.
546
Vol. 258, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
12. Mort, J. S., and Buttle, D. J. (1997) Int. J. Biochem. Cell Biol. 29, 715–720. 13. Turner, A., and Tanzawa, K. (1997) FASEB J. 11, 355–364. 14. Kugiyama, K., Sugiyama, S., Matsumura, T., Ohta, Y., Doi, H., and Yasue, H. (1996) Arterioscler. Thromb. Vasc. Biol. 16, 1080 –1087. 15. Sakai, J., Rawson, R. B., Espenshade, P. J., Cheng, D., Seeg-
miller, A. C., Goldstein, J. L., and Brown, M. S. (1998) Mol. Cell 2, 505–514. 16. Sakai, J. Nohturfft, A. Cheng, D. Ho, Y. K. Brown, M. S., and Goldstein, J. L. (1997) J. Biol. Chem. 272, 20213–20221. 17. Chan, S. J., San Segundo, B., McCormick, M. B., and Steiner, D. F. (1986) Proc. Natl. Acad. Sci. USA 83, 7721–7725.
547
Characterization of Cleavage Enzymes for Sterol Regulatory Element Binding Protein in Hamster Liver Microsomes Mineko Yamaguchi,* ,1 Heigoro Shirai,* ,1 Ryuichiro Sato,† Yoshiki Kawabe,‡ Rie Fukuda,* Tatsuhiko Kodama,* and Takao Hamakubo* ,2 *Department of Molecular Biology and Medicine, Research Center for Advanced Science and Technology, University of Tokyo, Meguro, Tokyo 153-8904, Japan; †Department of Biochemistry and Molecular Biology, Graduate School of Pharmaceutical Sciences, Osaka University, Suita, Osaka 565-0871, Japan; and ‡Fuji-Gotemba Research Laboratories, Chugai Pharmaceutical Co., Ltd., Gotemba, Shizuoka 412-8513, Japan
Received March 30, 1999
Sterol regulatory element binding proteins (SREBP-1 and SREBP-2) are the key transcription factors for the regulation of the cellular cholesterol level.To identify proteolytic enzymes for SREBPs, a fluorogenic peptide substrate, MOCAc-GRSVLSFK(Dnp)rr-NH 2, was synthesized according to the proposed cleavage site of human SREBP-2. In microsome fractions from hamster liver, we found a peptidase activity inhibitable by the synthetic inhibitor Ac-GRSVL-aldehyde with an IC 50 of 40 nM. This peptidase separated into three peaks of approximately 400 kDa, 60 kDa, and 30 kDa (Mp400, Mp60 and Mp30 respectively) upon gel permeation chromatography. Mp30 was purified to apparent homogeneity with an M r of 32 kDa. The partial amino acid sequence of Mp30 possessed homology to cathepsin B (EC 3.4.22.1). A 109 kDa protein band on SDS-PAGE which corresponded to Mp400 exhibited homology to neprilysin (EC 3.4.24.11) in partial amino acid sequence. These findings suggest several degradative pathways for SREBP in liver microsome membranes. © 1999 Academic Press Key Words: SREBP; protease; endoplasmic reticulum; cholesterol; cathepsin B; neprilysin.
Sterol regulatory element binding proteins (SREBP-1 and SREBP-2) play a major role in the transcriptional control of sterol biosynthetic enzymes, including 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase, HMG-CoA synthase and low density lipoprotein (LDL) receptor (1). SREBPs are synthesized as 1
These authors contributed equally to this work. To whom correspondence and reprint requests should be addressed at Research Center for Advanced Science and Technology, University of Tokyo, 4-6-1 Komaba, Meguro-ku, Tokyo, 153-8904, Japan. Fax: (181) (3) 3481-4552. E-mail: [email protected]. 2
0006-291X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
precursors in the endoplasmic reticulum (ER) membrane and when the intracellular sterol level decreases, are cleaved by proteases inside the ER lumen at the sequence termed “site 1” (2). SREBPs cleaved at site 1 are successively cleaved by a second protease named site 2 protease (3), which recognizes the amino acid sequence within the transmembrane domain. The mature forms of SREBPs released from the ER membrane translocate to the nucleus and bind to sterolregulatory elements (SREs) in the promoter region of genes encoding enzymes related to cholesterol biosynthesis and the LDL receptor (4, 5). In the course of purification of the SREBP processing enzymes, two apoptosis-related cysteine proteases have been identified as being able to cleave and activate SREBPs (6, 7). To characterize the cleavage activity in the ER, we exploited an in vitro assay system using quenched fluorogenic peptide substrate (8) corresponding to eight amino acid residues of the site 1 sequence in the ER lumen portion of human SREBP-2. In microsome fractions of the hamster liver, we found cleavage activity which could be inhibited by a sequence specific inhibitor, acetyl-Gly-Arg-Ser-Val-Leualdehyde. When the hamster liver microsome proteins were solubilized and applied to gel permeation chromatography, the activity was resolvable into three different fractions. Here we characterize these peptidases, which cleave the sequence at a point corresponding to site 1 of human SREBP-2. MATERIALS AND METHODS Peptides and reagents. An internally quenched fluorogenic peptide substrate, (7-methoxycoumarin-4-yl)acetyl-Gly-Arg-Ser-Val-LeuSer-Phe-Lys-2,4-dinitrophenyl-Arg-Arg-amide (MOCAc-GRSVLSFK(Dnp)rrNH 2), the reference compound for fluorescence intensity, MOCAc-Pro-Leu-Gly, and a peptide inhibitor, acetyl-Gly-Arg-Ser-ValLeu-aldehyde (Ac-GRSVL-CHO), were synthesized by the Peptide In-
542
Vol. 258, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
stitute (Osaka, Japan) and shown by HPLC to be greater than 98% pure. These sequences were deduced from the proposed cleavage site of the endoplasmic reticulum lumen portion of human SREBP-2 (2). Other reagents were obtained from Sigma except as otherwise stated. Assay of cleavage activity. Reaction mixtures, which contained 50 mM Hepes pH 7.4, 1 mg peptide substrate, and enzyme solution, were monitored continuously in a spectrofluorometer at an excitation wavelength of 328 nm and an emission wavelength of 393 nm. One unit of enzyme activity was tentatively determined as 1 pmol of MOC-peptide (fluorescence intensity was determined from a standard peptide, MOCAc-Pro-Leu-Gly) liberated per min at room temperature. Preparation of liver microsome fraction. Under Nembutal anesthesia, male Golden Syrian hamsters aged 9 to 10 weeks were perfused from the portal vein with 50 ml of ice-cold homogenizing buffer (10 mM Hepes, containing 1 mM MgCl 2, 10 mM KCl, 1 mM DTT, 0.25 M sucrose, pH 7.5). The livers were excised and homogenized with ninefold volume of ice-cold homogenizing buffer, and then centrifuged at 10,000g for 20 min. The supernatant was centrifuged at 105,000g for 1 hour. The supernatant after 105,000g centrifugation was used as the cytosol fraction, and the pellet was resuspended with buffer A (50 mM Tris-HCl, 1 mM EDTA, 1 mM DTT, 10% glycerol, pH 7.5) and used as the microsome fraction. Purification procedures. To solubilize the membrane peptidases, a final 0.5% (w/v) MEGA9 (Dojindo, Japan) or 0.5% (w/v) lubrol (Lubrol PX, Sigma) was added to the microsome fractions. After 1 hour mixing with rotator at 4°C, the solubilized fractions were separated from membranes by centrifugation at 105,000g for 1 hour. For the purification of Mp30, about 100 ml of the MEGA9 solubilized fraction derived from five hamster livers was loaded on to an anion-exchange column (Q-sepharose HiLoad 26/10, Amersham Pharmacia Biotech), and after washing with buffer A, containing 0.5% MEGA9, the column was eluted with a linear gradient to 0.5 M NaCl. One fractional peak was collected and applied to a superdex 200 (HiLoad 26/60, Amersham Pharmacia Biotech) equilibrated with buffer A containing 0.2% MEGA9, and fraction 21, which corresponded to Mp30, was collected and stored at 280°C. The same fractions from 8 batches were combined and applied to a Mono Q (HR 5/5, Amersham Pharmacia Biotech) column equilibrated with buffer A containing 0.2% MEGA9. The active fractions, which eluted at around 0.12 M NaCl with a linear gradient, were collected and dialyzed against 10 mM phosphate buffer containing 10% glycerol and 0.25% MEGA9, pH 6.8. The dialyzed fractions were applied to a hydroxyapatite column (SH-0410F, Pentax) with a linear gradient elution from 10 mM to 400 mM phosphate buffer containing 10% glycerol and 0.25% MEGA9, pH 6.8. Partial amino acid sequence analysis was performed by APRO Science Incorporation (Tokushima, Japan). For the purification of Mp400, microsome fractions from 5 hamster livers were solubilized with buffer A containing 0.5% (w/v) lubrol, and then loaded on to Q-sepharose equilibrated with buffer A containing 0.5% lubrol. The peptidases were eluted with a linear gradient to 0.5 M NaCl. Then collected fractions were applied to 5 ml Blue sepharose (HiTrap Blue, Amersham Pharmacia Biotech) equilibrated with buffer A containing 0.25% lubrol and eluted with a 0-2 M NaCl linear gradient. The active fractions around 1 M NaCl were collected and applied to a superdex 200 equilibrated with buffer A supplemented with 0.25% lubrol and 0.2 M NaCl. The active fractions from four batches were pooled and applied to a 1 ml concanavalin A affinity column (Con A Sepharose, Amersham Pharmacia Biotech) equilibrated with buffer B (20 mM Tris, 1 mM DTT, 5% glycerol, 0.1% lubrol, pH 7.4) containing 0.5 M NaCl. After washing with ten column volume of buffer B, containing 0.5 M NaCl, the column was loaded with buffer B supplemented with 0.5 M NaCl and 0.1 M a-methyl-D-mannoside. The column was allowed to stand for 1 hour at 4°C, and then the peptidase was eluted with the same buffer. After dialysis against buffer B, the aliquot was applied to a Mono Q
FIG. 1. (A) Dose-dependent cleavage of peptide substrate by microsome fractions after freeze-thaw. The peptide substrate was incubated with 10 ml of the microsome fraction (open circles), 20 ml of the microsome fraction (filled circles), 10 ng trypsin (open squares), or without enzyme (open triangles). The amount of liberated peptide was calculated from the increase of fluorescence intensity divided by the reference compound. (B) Inhibition by peptide inhibitor, AcGRSVL-CHO, of cleavage activity in freeze-thawed liver microsome fraction.
column equilibrated with buffer B, and eluted with a 0 – 0.5 M NaCl linear gradient.
RESULTS Peptidase Activity in Hamster Liver Microsome Fractions Peptidase activity was found in both the cytosol and the microsome fractions prepared from hamster liver. The microsome fraction exhibited peptidase activity after freeze-thaw (Fig. 1A), and this activity was inhib-
543
Vol. 258, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. The elution profile of cleavage activities from hamster liver microsome on superdex 200 gel permeation chromatography. The activities were separated into three major peaks named Mp400 (a), Mp60 (b), and Mp30 (c).
itable by a peptide inhibitor, Ac-GRSVL-CHO, with an IC 50 of 40 nM (Fig. 1B), whereas peptidase in cytosol or trypsin was not inhibited by the peptide inhibitor even at 10 mM (data not shown). Since the proposed site 1 protease cleavage occurs between the leucine 522 and serine 523 bond (2), the peptidase activity in the freezethawed microsome membrane was thought to be specific for the SREBP-2 ER loop peptide. Sixty percent of the total activity was recovered by solubilization with 0.5% MEGA9, and this solubilized activity proved to be stable for several weeks when stored at 280°C. Solubilization and Purification of the Peptidases from Microsome Membrane After centrifugation at 105,000g for 1 hour, the supernatant was collected and applied to a Q-sepharose anion exchange column. There was no significant activity in the flow through fractions, and the total recovery of the activity from the column was around 50%. The active fractions were widely distributed in a range from 50 mM to 0.5 M NaCl with one clear peak at fraction 14, which eluted at around 0.12 M NaCl (data not shown). When fraction 14 was applied to gel permeation chromatography (superdex 200) the activity separated into three major peaks at approximately 400 kDa, 60 kDa, and 30 kDa (Fig. 2), named Mp400, Mp60 and Mp30, respectively. Among these, Mp30 was inhibited 90% with 1 mM of the peptide inhibitor, AcGRSVL-CHO, but Mp400 and Mp60 exhibited no significant inhibition at the same dose. On the other hand, Mp60 exhibited a two fold increase in activity when 100 mM MnCl 2 was added, whereas neither Mp30 nor Mp400 showed any increase of activity to in response to
any divalent cation tested (data not shown). Thus Mp60 was considered to be a different peptidase from Mp30 and Mp400. Mp30 was further purified by a Mono Q column and a 32 kDa band on SDS-PAGE corresponded to the peptidase activities. After hydroxyapatite column chromatography, Mp30 was purified to apparent homogeneity on SDS-PAGE with a M r of 32 kDa (Fig. 3A). Since the Mp400 activities were much more abundant when the microsome was solubilized with 0.5% lubrol compared with 0.5% MEGA9, we performed the purification of Mp400 using lubrol solubilization. Solubilized fractions were applied to an anion exchange column, Q-sepharose. The characteristic feature of Mp400 was that it required ATP after separation by column chromatography. After the anion exchange column, the cleavage activities were increased about five fold when 1 mM ATP was added to the reaction mixture. After a second chromatography with a bluesepharose column, the cleavage activities were found to be measurable only when the reaction mixture contained ATP or its analogues, such as ATP-g-S or dATP (Table 1). Mp400, which requires ATP or its analogues, was not inhibited significantly by the peptide inhibitor even at 10 mM. After the Mono Q column, the peptidase activity corresponded well with the 109 kDa protein band on SDS-PAGE. This protein band was excised from the gel and the partial amino acid sequences were determined. Inhibitor Spectrum of Mp30 and Mp400 Peptidases Using the gel permeation chromatography fractions, the effects of various protease inhibitors were tested
544
Vol. 258, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. (A) SDS-PAGE of purified Mp30. Lane a, molecular size marker; lane b, active fraction from hydroxyapatite column. (B) Comparison of amino acid sequence of 32 kDa protein from hamster with primary structure of mouse cathepsin B precursor (17). The determined partial sequences are underlined. The different residues between hamster and mouse are marked with asterisk (see text). The precursor protein is processed at the peptide bond between Asp 79 and Leu 80 to generate a single chain cathepsin B. Hatched portions indicate the light and heavy chains of the two chain form of cathepsin B.
(Table 2). Mp30 was susceptible to thiol protease inhibitors such as leupeptin or acetyl-leucyl-leucylnorleucinal (ALLN). Mp400 was resistant to serine or thiol protease inhibitors and was strongly inhibited by phosphoramidon or thiorphan (Table 2). Amino Acid Sequence Analysis From the purified Mp30 fraction, the amino acid sequences of three fragments of the 32 kDa protein were determined to be LPETFDARK, GLVSGGLYNS, and EIMAEIYK. These sequences are identical to
TABLE 1
Effect of Various Reagents on Cleavage Activity of Mp400 Reagents ATP ADP dATP ATP-g-S cAMP Actyl CoA
Concentration 2 2 2 1 2 1
mM mM mM mM mM mM
Cleavage activity (%) 100 43 175 96 10 59
Note. The activity with 2 mM ATP was taken as 100%.
mouse cathepsin B except for glutamate 88 to lysine, and 172 valine to leucine (Fig. 3B). Also, the determined amino acid sequences of the peptides derived from the 109 kDa protein band were DGDLVDWWTQ, LLPGIDLNHK, and LLPDIYGWPV. These sequences are identical to the internal sequence of mouse neprilysin. DISCUSSION We have developed an in vitro assay system for SREBP-2 ER loop cleaving enzymes, using the peptide substrate designed for the amino acid sequence around the site 1 protease cleavage site. In hamster liver microsomes, we found a peptidase which was inhibitable by Ac-GRSVL-CHO. Subsequently, from the MEGA 9 solubilized fraction of these microsomes, we purified 32 kDa single chain cathepsin B. Cathepsin B is believed to be synthesized in a precursor form (339 amino acids in the mouse), and processed to the 32 kDa single chain form in the ER by autocatalysis and/or by certain processing enzymes (9). During the sorting from the endoplasmic reticulum to the lysosome, single chain cathepsin B is processed further to generate a two chain enzyme (10). It is known that active cathepsin B dis-
545
Vol. 258, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 2
Effect of Reagents on the MOC-Ac-GRSVLSFK(Dnp)rrNH 2 Cleaving Activities from Gel Permeation Chromatography Remaining activity (%) Reagents Pefabloc Leupeptin E-64-c EDTA Lactacystin ALLN
Ac-GRSVL-CHO Phosphoramidon Thiorphan
Concentration
Mp400
Mp30
2 mM 200 mg/ml 100 mg/ml 10 mM 10 mM 52 mM 5.2 mM 0.52 mM 0.052 mM 1 mM 100 mg/ml 2 mM
106 124 91 22 113 132 N.D. N.D. N.D. 87 6 3
100 27 N.D. N.D. N.D. 34 40 54 92 2 N.D. N.D.
Note. The activities without reagents were taken as 100%. The activities of Mp400 were measured under the presence of 1 mM ATP. N.D., Not determined.
tributes not only to the lysosome but also to the ER and nuclear membranes (11). Since we did not find any corresponding band to the two chain form on SDSPAGE, the dominant form in our microsome fraction was taken to be the 32 kDa single chain Cathepsin B (Fig. 3). Cathepsin B is postulated to be involved in antigen presentation, amyloid precursor protein cleavage in Alzheimer’s disease, metastasis of malignant cells, and the degradation of foreign or intracellular proteins (12). The SREBPs are also known to undergo rapid degradation, which is inhibitable by ALLN (4). From the result that Mp30 could be inhibited by ALLN (Table 1), it is possible that cathepsin B is involved in the degradation of SREBPs. Another peptide substrate cleaving activity fragment found in the microsome fraction had a large molecular weight of around 400 kDa on gel permeation chromatography. Mp400 peptidase required ATP or its analogues for this activity. Since ATP-g-S or dATP also activated the cleavage to a similar extent as ATP, the activity was considered not to be energy dependent. Since the requirement of ATP appeared after column chromatography, it was thought that some molecule acting as a cofactor might have been separated. In partially purified Mp400, we did find a 109 kDa protein on SDS-PAGE which corresponded to this activity and the partial amino acid sequence turned out to be identical to the transmembrane metallo-endopeptidase, neprilysin. Neprilysin, neutral endopeptidase-24.11 (EC 3.4.24.11), also known as enkephalinase or CD10/ common acute lymphoblastic leukemia-associated antigen (CD10/CALLA), cleaves peptide bonds on the amino side of hydrophobic amino acid residues and is sensitive to phosphoramidon (13). It is not known whether neprilysin requires ATP or similar com-
pounds. Since the result of the inhibitor spectrum (Table 2) of Mp400 was compatible to that of neprilysin, we have concluded that the major active component of Mp400 should be attributed to neprilysin and that ATP may be required for the dissociation of the inhibitory components. As neprilysin is considered to play a role in degrading peptide hormones (13, 14), it may also degrade SREBPs on the membrane. Recently, Sakai et al. identified membrane-bound subtilisin like site 1 protease (S1P) by using an in vivo assay system for the SREBP-2 ER loop. S1P restored site 1 cleavage in a mutant CHO cell line lacking S1P (15). One reason that we did not find a significant activity fragment sensitive to serine protease inhibitors may be due to the requirement of SCAP (SREBP Cleavage-activating Protein) for the maximum activity of S1P (16). In conclusion, we have characterized SREBP ER loop cleavage activities in hamster liver microsomes. This activity, which could be inhibited by a sequence specific inhibitor, was found to be cathepsin B in single chain form. There were two other non-specific peptidases, one with a molecular weight of 60 kDa, and another, Mp400, which was associated with neprilysin activity. The susceptibility of the ER loop to these enzymes suggests that SREBPs undergo breakdown upon interaction with them. ACKNOWLEDGMENTS This work was supported by a grant from the “Research for the Future” Program from the Japan society for the Promotion of Science and by the Ministry of Education, Grant-in-Aid for Scientific Research (B), 08457258. The authors are grateful to Dr. K. Boru for reviewing the manuscript.
REFERENCES 1. Brown, M. S., and Goldstein, J. L. (1997) Cell 89, 331–340. 2. Duncan, E. A., Brown, M. S., Goldstein, J. L., and Sakai, J. (1997) J. Biol.Chem. 272, 12778 –1285. 3. Rawson, R. B., Zelenski, N. G., Nijhawan, D., Ye, J., Sakai, J., Hasan, M. T., Chang, T. Y., Brown, M. S., and Goldstein, J. L. (1997) Mol. Cell 1, 47–57. 4. Wang, X., Sato, R., Brown, M. S., Hua, X., and Goldstein, J. L. (1994) Cell 77, 53– 62. 5. Smith, J. R., Osborne, T. F., Goldstein, J. L., and Brown, M. S. (1990) J. Biol. Chem. 265, 2306 –2310. 6. Hua, X., Sakai, J., Brown, M. S., and Goldstein, J. L. (1996) J. Biol. Chem. 271, 10379 –10384. 7. Pai, J-T., Brown, M. S., and Goldstein, J. L. (1996) Proc. Natl. Acad. Sci. USA 93, 5437–5442. 8. Knight, C. G., Willenbrock, F., and Murphy, G. (1992) FEBS Lett. 296, 263. 9. Podobnik, M., Kuhelj, R., Turk, V., and Turk, D. (1997) J. Mol. Biol. 271, 774 –788. 10. Kominami., E., Tsukahara, T., Hara, K., and Katunuma, N. (1988) FEBS Lett. 231 225–228. 11. Spiess, E., Bru¨ning, A., Gack, S., Ulbricht, B., Spring, H., Trefz, G., and Ebert, W. (1994) J. Histochem. Cytochem. 42, 917–929.
546
Vol. 258, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
12. Mort, J. S., and Buttle, D. J. (1997) Int. J. Biochem. Cell Biol. 29, 715–720. 13. Turner, A., and Tanzawa, K. (1997) FASEB J. 11, 355–364. 14. Kugiyama, K., Sugiyama, S., Matsumura, T., Ohta, Y., Doi, H., and Yasue, H. (1996) Arterioscler. Thromb. Vasc. Biol. 16, 1080 –1087. 15. Sakai, J., Rawson, R. B., Espenshade, P. J., Cheng, D., Seeg-
miller, A. C., Goldstein, J. L., and Brown, M. S. (1998) Mol. Cell 2, 505–514. 16. Sakai, J. Nohturfft, A. Cheng, D. Ho, Y. K. Brown, M. S., and Goldstein, J. L. (1997) J. Biol. Chem. 272, 20213–20221. 17. Chan, S. J., San Segundo, B., McCormick, M. B., and Steiner, D. F. (1986) Proc. Natl. Acad. Sci. USA 83, 7721–7725.
547