Insertion Sequence 1 of Muscle-specific Calpain, p94, Acts as an Internal Propeptide
2004; Elsevier BV; Volume: 279; Issue: 26 Linguagem: Inglês
10.1074/jbc.m313290200
ISSN1083-351X
AutoresBeatríz García Díaz, Tudor Moldoveanu, Michael J. Kuiper, Robert L. Campbell, Peter L. Davies,
Tópico(s)Hereditary Neurological Disorders
ResumoThe physiological role of the skeletal muscle-specific calpain 3, p94, is presently unknown, but defects in its gene cause limb girdle muscular dystrophy type 2A. This calcium-dependent cysteine protease resembles the large subunit of m-calpain but with three unique additional sequences: an N-terminal region (NS), and two insertions (IS1 and IS2). The latter two insertions have been linked to the chronic instability of the whole enzyme both in vivo and in vitro. We have shown previously that the core of p94 comprising NS, domains I and II, and IS1 is stable as a recombinant protein in the absence of Ca2+ and undergoes autolysis in its presence. Here we show that p94I-II cannot hydrolyze an exogenous substrate before autolysis but is increasingly able to do so when autolysis proceeds for several hours. This gain in activity is caused by cleavage of IS1 during autolysis because a deletion mutant lacking the NS region (p94I-II ΔNS) shows the same activation profile. Similarly, the calpain inhibitors E-64 and leupeptin have almost no inhibitory effect on substrate hydrolysis by p94I-II soon after calcium addition but cause complete inhibition when autolysis progresses for several hours. As autolysis proceeds, there is release of the internal IS1 peptide, but the two portions of the core remain tightly associated. Modeling of p94I-II suggests that IS1 contains an amphipathic α-helix flanked by extended loops. The latter are the targets of autolysis and limited digestion by exogenous proteases. The presence and location of the α-helix in recombinant IS1 were confirmed by circular dichroism and by the introduction of a L286P helix-disrupting mutation. Within p94I-II, L286P caused premature autoproteolysis of the enzyme. IS1 is an elaboration of a loop in domain II near the active site, and it acts as an internal autoinhibitory propeptide, blocking the active site of p94 from substrates and inhibitors. The physiological role of the skeletal muscle-specific calpain 3, p94, is presently unknown, but defects in its gene cause limb girdle muscular dystrophy type 2A. This calcium-dependent cysteine protease resembles the large subunit of m-calpain but with three unique additional sequences: an N-terminal region (NS), and two insertions (IS1 and IS2). The latter two insertions have been linked to the chronic instability of the whole enzyme both in vivo and in vitro. We have shown previously that the core of p94 comprising NS, domains I and II, and IS1 is stable as a recombinant protein in the absence of Ca2+ and undergoes autolysis in its presence. Here we show that p94I-II cannot hydrolyze an exogenous substrate before autolysis but is increasingly able to do so when autolysis proceeds for several hours. This gain in activity is caused by cleavage of IS1 during autolysis because a deletion mutant lacking the NS region (p94I-II ΔNS) shows the same activation profile. Similarly, the calpain inhibitors E-64 and leupeptin have almost no inhibitory effect on substrate hydrolysis by p94I-II soon after calcium addition but cause complete inhibition when autolysis progresses for several hours. As autolysis proceeds, there is release of the internal IS1 peptide, but the two portions of the core remain tightly associated. Modeling of p94I-II suggests that IS1 contains an amphipathic α-helix flanked by extended loops. The latter are the targets of autolysis and limited digestion by exogenous proteases. The presence and location of the α-helix in recombinant IS1 were confirmed by circular dichroism and by the introduction of a L286P helix-disrupting mutation. Within p94I-II, L286P caused premature autoproteolysis of the enzyme. IS1 is an elaboration of a loop in domain II near the active site, and it acts as an internal autoinhibitory propeptide, blocking the active site of p94 from substrates and inhibitors. Calpains comprise a large family of cytosolic Ca2+-dependent cysteine proteinases with homologues present in mammals, insects, nematodes, and yeast (1Sorimachi H. Suzuki K. J. Biochem. (Tokyo). 2001; 129: 653-664Crossref PubMed Scopus (246) Google Scholar, 2Goll D.E. Thompson V.F. Li H. Wei W. Cong J. Physiol. Rev. 2003; 83: 731-801Crossref PubMed Scopus (2378) Google Scholar). Two well-studied members of this family are the mammalian heterodimeric μ- and m-calpains, which are composed of a large catalytic subunit (80 kDa) and a small regulatory subunit (28 kDa). The domain structure of the large subunit was redefined by x-ray crystallography (3Hosfield C.M. Elce J.S. Davies P.L. Jia Z. EMBO J. 1999; 18: 6880-6889Crossref PubMed Scopus (290) Google Scholar, 4Strobl S. Fernandez-Catalan C. Braun M. Huber R. Masumoto H. Nakagawa K. Irie A. Sorimachi H. Bourenkow G. Bartunik H. Suzuki K. Bode W. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 588-592Crossref PubMed Scopus (316) Google Scholar). Following the classification of Hosfield et al. (3Hosfield C.M. Elce J.S. Davies P.L. Jia Z. EMBO J. 1999; 18: 6880-6889Crossref PubMed Scopus (290) Google Scholar), the large subunit has four structural domains (I-IV). The first two (I and II) make up the papain-like catalytic core common to all calpains (1Sorimachi H. Suzuki K. J. Biochem. (Tokyo). 2001; 129: 653-664Crossref PubMed Scopus (246) Google Scholar), whereas domains III and IV are the C2-like and penta-EF-hand domains, respectively. The small subunit contains two domains, V and VI, of which the latter is also a penta-EF-hand domain that forms extensive interactions with the homologous penta-EF-hand domain IV in the large subunit. Several more of the ∼14 human calpain isoforms are composed of two subunits and share the same domain structure as μ- and m-calpain. Other calpain homologues do not seem to form heterodimers, and some also contain different domains such as SOH (small optic lobe homologous), PBH (PalB homologous), and T (tra-3 specific) as additions or substitutions. Some members of the calpain family appear to be ubiquitously expressed, whereas others are only found in specific tissues. However, these distinctions may not be as clear-cut as once thought (5Farkas A. Tompa P. Friedrich P. Biol. Chem. 2003; 384: 945-949Crossref PubMed Scopus (26) Google Scholar). Before the discovery of many new calpain genes by way of the Human Genome Project and other sequencing projects, molecular cloning experiments had identified a third member of the calpain superfamily, calpain 3 (p94), whose mRNA was predominantly expressed in skeletal muscle (6Sorimachi H. Imajoh-Ohmi S. Emori Y. Kawasaki H. Ohno S. Minami Y. Suzuki K. J. Biol. Chem. 1989; 264: 20106-20111Abstract Full Text PDF PubMed Google Scholar). Based on sequence homology to μ- and m-calpain (54% and 51% identity, respectively), p94 has a domain organization similar to that of their large subunits (6Sorimachi H. Imajoh-Ohmi S. Emori Y. Kawasaki H. Ohno S. Minami Y. Suzuki K. J. Biol. Chem. 1989; 264: 20106-20111Abstract Full Text PDF PubMed Google Scholar). In addition, p94 contains three unique sequences: an N-terminal extension (NS) of 47 amino acids located where the anchor peptide would be in m-calpain, and two insertion sequences. The first insertion sequence (IS1) is 48 amino acids long and interrupts domain II after Asp267. This insertion is encoded by a single exon (exon 6). The second insertion of 77 residues is encoded by two exons (exons 15 and 16) and located between domains III and IV (IS2). Although p94 has been modeled to contain all four domains of the large subunit (7Jia Z. Petrounevitch V. Wong A. Moldoveanu T. Davies P.L. Elce J.S. Beckmann J.S. Biophys. J. 2001; 80: 2590-2596Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar), it does not seem to form a heterodimer (8Kinbara K. Ishiura S. Tomioka S. Sorimachi H. Jeong S.Y. Amano S. Kawasaki H. Kolmerer B. Kimura S. Labeit S. Suzuki K. Biochem. J. 1998; 335: 589-596Crossref PubMed Scopus (99) Google Scholar). Consistent with this theory, protein-protein interaction analysis using the yeast two-hybrid system did not show an interaction between p94 and the 28-kDa small subunit (9Sorimachi H. Kinbara K. Kimura S. Takahashi M. Ishiura S. Sasagawa N. Sorimachi N. Shimada H. Tagawa K. Maruyama K. Suzuki K. J. Biol. Chem. 1995; 270: 31158-31162Abstract Full Text Full Text PDF PubMed Scopus (296) Google Scholar). The importance of p94 in muscle physiology was confirmed by the discovery that mutations in the calpain 3 gene (capn3) lead to limb girdle muscular dystrophy type 2A (10Richard I. Broux O. Allamand V. Fougerousse F. Chiannilkulchai N. Bourg N. Brenguier L. Devaud C. Pasturaud P. Roudaut C. Hillaire D. Passes Bueno M.R. Zatz M. Tischfield J.A. Fardeau M. Jackson C.E. Cohen D. Beckmann J.S. Cell. 1995; 81: 27-40Abstract Full Text PDF PubMed Scopus (868) Google Scholar). Numerous efforts have been made to isolate p94 from muscle or to produce a recombinant form of the enzyme in Escherichia coli (11Sorimachi H. Ishiura S. Suzuki K. J. Biol. Chem. 1993; 268: 19476-19482Abstract Full Text PDF PubMed Google Scholar, 12Federici C. Eshdat Y. Richard I. Bertin B. Guillaume J.L. Hattab M. Beckmann J.S. Strosberg A.D. Camoin L. Arch. Biochem. Biophys. 1999; 363: 237-245Crossref PubMed Scopus (12) Google Scholar), mammalian cells (11Sorimachi H. Ishiura S. Suzuki K. J. Biol. Chem. 1993; 268: 19476-19482Abstract Full Text PDF PubMed Google Scholar), or insect cells (12Federici C. Eshdat Y. Richard I. Bertin B. Guillaume J.L. Hattab M. Beckmann J.S. Strosberg A.D. Camoin L. Arch. Biochem. Biophys. 1999; 363: 237-245Crossref PubMed Scopus (12) Google Scholar, 13Branca D. Gugliucci A. Bano D. Brini M. Carafoli E. Eur. J. Biochem. 1999; 265: 839-846Crossref PubMed Scopus (54) Google Scholar). These attempts failed to produce large amounts of p94 due to its rapid autolytic degradation, which has hindered a thorough biochemical characterization of the enzyme. As a result, the roles of the three p94-specific sequences and their contribution to the unique properties and physiological function of calpain 3 have remained speculative. IS1 and IS2 are each thought to contribute to the lack of stability of p94. Splicing variants of p94 lacking these novel insertions have been found in different eye tissues such as lens (Lp82) (14Ma H. Fukiage C. Azuma M. Shearer T.R. Investig. Ophthalmol. Vis. Sci. 1998; 39: 454-461PubMed Google Scholar), retina (Rt88) (15Azuma M. Fukiage C. Higashine M. Nakajima T. Ma H. Shearer T.R. Curr. Eye Res. 2000; 21: 710-720Crossref PubMed Scopus (41) Google Scholar), and cornea (Cn94) (16Nakajima T. Fukiage C. Azuma M. Ma H. Shearer T.R. Biochim. Biophys. Acta. 2001; 1519: 55-64Crossref PubMed Scopus (33) Google Scholar); in skeletal muscle (17Herasse M. Ono Y. Fougerousse F. Kimura E. Stockholm D. Beley C. Montarras D. Pinset C. Sorimachi H. Suzuki K. Beckmann J.S. Richard I. Mol. Cell. Biol. 1999; 19: 4047-4055Crossref PubMed Scopus (109) Google Scholar); and in peripheral blood mononuclear cells (18De Tullio R. Stifanese R. Salamino F. Pontremoli S. Melloni E. Biochem. J. 2003; 375: 689-696Crossref PubMed Scopus (42) Google Scholar). In contrast to p94, Lp82, the lens-specific calpain isoform that lacks NS, IS1, and IS2, has been detected as a full-length 82-kDa protein band in young rodent lenses, and the corresponding enzymatic activity band has been detected on casein zymograms (19Ma H. Shih M. Hata I. Fukiage C. Azuma M. Shearer T.R. Exp. Eye Res. 1998; 67: 221-229Crossref PubMed Scopus (43) Google Scholar). Furthermore, a recombinant construct of Lp82 was stably produced in insect cells, purified, and characterized (20Fukiage C. Nakajima E. Ma H. Azuma M. Shearer T.R. J. Biol. Chem. 2002; 277: 20678-20685Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). Very recently, an alternatively spliced variant of mouse skeletal muscle p94 that lacks IS1 and IS2 (p94:exon 6-15-16-, p94Δ) has been successfully expressed in mammalian cells and shown to be proteolytically active (21Ono Y. Kakinuma K. Torii F. Irie A. Nakagawa K. Labeit S. Abe K. Suzuki K. Sorimachi H. J. Biol. Chem. 2004; 279: 2761-2771Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar). Introduction of IS1 into Lp82 destabilized the protein and caused a loss of enzymatic activity, whereas substitution of the shorter N-terminal sequence (AX1) present in Lp82 by NS or introduction of IS2 did not affect stability (22Ma H. Shih M. Fukiage C. Azuma M. Duncan M.K. Reed N.A. Richard I. Beckmann J.S. Shearer T.R. Investig. Ophthalmol. Vis. Sci. 2000; 41: 4232-4239PubMed Google Scholar). These results strongly suggest that the stability and abundance of enzymatically active Lp82 in rodent lens are due to the lack of IS1 and may explain why IS1-containing isoforms of calpains such as p94 and Rt88 (15Azuma M. Fukiage C. Higashine M. Nakajima T. Ma H. Shearer T.R. Curr. Eye Res. 2000; 21: 710-720Crossref PubMed Scopus (41) Google Scholar) are unstable (23Sorimachi H. Toyama-Sorimachi N. Saido T.C. Kawasaki H. Sugita H. Miyasaka M. Arahata K. Ishiura S. Suzuki K. J. Biol. Chem. 1993; 268: 10593-10605Abstract Full Text PDF PubMed Google Scholar). Rapid autolysis of p94 during purification from rabbit muscle fractions was not prevented by the calpain inhibitors E-64 (trans-epoxysuccinyl-l-leucylamido(4-guanidino)butane) and leupeptin (8Kinbara K. Ishiura S. Tomioka S. Sorimachi H. Jeong S.Y. Amano S. Kawasaki H. Kolmerer B. Kimura S. Labeit S. Suzuki K. Biochem. J. 1998; 335: 589-596Crossref PubMed Scopus (99) Google Scholar), whereas autolysis of purified recombinant Lp82 was totally inhibited by E-64 or EGTA (20Fukiage C. Nakajima E. Ma H. Azuma M. Shearer T.R. J. Biol. Chem. 2002; 277: 20678-20685Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). Moreover, a recently reported p94 isoform present in human peripheral blood mononuclear cells that lacks IS1 and the lysine-rich IS2 sequence was also inhibited by leupeptin and calpain inhibitors I and II (18De Tullio R. Stifanese R. Salamino F. Pontremoli S. Melloni E. Biochem. J. 2003; 375: 689-696Crossref PubMed Scopus (42) Google Scholar). In contrast to μ- and m-calpain, p94 and its splice variants have been shown to be insensitive to inhibition by calpastatin (8Kinbara K. Ishiura S. Tomioka S. Sorimachi H. Jeong S.Y. Amano S. Kawasaki H. Kolmerer B. Kimura S. Labeit S. Suzuki K. Biochem. J. 1998; 335: 589-596Crossref PubMed Scopus (99) Google Scholar, 18De Tullio R. Stifanese R. Salamino F. Pontremoli S. Melloni E. Biochem. J. 2003; 375: 689-696Crossref PubMed Scopus (42) Google Scholar, 20Fukiage C. Nakajima E. Ma H. Azuma M. Shearer T.R. J. Biol. Chem. 2002; 277: 20678-20685Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). The basis for this insensitivity is not known, but it could be due in part to the absence of the small subunit, which is one of the defined calpastatin binding sites present in m-calpain (24Todd B. Moore D. Deivanayagam C.C. Lin G.D. Chattopadhyay D. Maki M. Wang K.K. Narayana S.V. J. Mol. Biol. 2003; 328: 131-146Crossref PubMed Scopus (85) Google Scholar). Unlike whole p94, the proteolytic core of p94 containing NS and IS1 can be produced in relatively large amounts in E. coli, and the recombinant protein is fairly stable during purification (25Rey M.A. Davies P.L. FEBS Lett. 2002; 532: 401-406Crossref PubMed Scopus (39) Google Scholar). Using this truncated form of the enzyme, we have previously shown that the core of p94 is stable in EDTA but undergoes autolysis in the presence of Ca2+. NS and IS1 are rapidly cleaved during this process at specific N-terminal sites by a strictly intramolecular reaction (Fig. 1). Here we have studied the effect of NS and IS1 on p94 enzymatic properties. Our results show that IS1 acts as an internal propeptide. Autoproteolysis serves to remove IS1 (and NS), making the active site available for hydrolysis of exogenous substrates and accessible to inhibitors. Cloning, Mutagenesis, Expression, and Purification of Recombinant Proteins—The protease core of human p94 (p94I-II) and its active site knockout mutant, p94I-II C129S, were prepared as described previously (25Rey M.A. Davies P.L. FEBS Lett. 2002; 532: 401-406Crossref PubMed Scopus (39) Google Scholar). These constructs contain NS and IS1 and are effectively a C-terminal truncation of p94 after Asp419 (Fig. 1). The subsequent deletion of the NS region was achieved by PCR using expand high-fidelity DNA polymerase (Roche Applied Science) with the p94I-II DNA construct as a template. A new 5′ primer (5′-GATATACATATGATCATCAGCCGCAATTTTCCTAT-3′) deleted the first 45 amino acids, i.e. 1 residue C-terminal to the second (2a) autolysis site in NS, and inserted an initiating methionine (Fig. 1). The amplified product was ligated into the pET24a vector between the NdeI and XhoI sites to encode p94I-II ΔNS with a C-terminal His6 tag (MI46ISRN—CNLTAD419LEH6). The DNA sequences of positive clones were verified. p94I-II ΔNS was expressed in E. coli BL21(DE3) under kanamycin selection. This deletion construct was purified according to the protocol described previously for p94I-II (25Rey M.A. Davies P.L. FEBS Lett. 2002; 532: 401-406Crossref PubMed Scopus (39) Google Scholar), but with size-exclusion chromatography on Sephacryl S-200 included before the final Q-Sepharose fast protein liquid chromatography. The 48-amino acid IS1 sequence was amplified by PCR and ligated into the pET28a(+) vector between the NdeI and XhoI sites. The recombinant IS1 was expressed as a 68-residue His-tagged polypeptide in E. coli BL21(DE3) at room temperature under kanamycin selection. It was purified from the supernatant fraction by metal affinity chromatography on a Ni-NTA 1The abbreviations used are: Ni-NTA, nickel-nitrilotriacetic acid; SLY, succinyl-leucine-tyrosine; MCA, aminomethyl coumarin; CAPS, 3-(cyclohexylamino)propanesulfonic acid; HPLC, high pressure liquid chromatography; TFE, trifluoroethanol.-agarose column followed by reversed-phase HPLC on a C18 semi-preparative column. The L286P helix-breaking mutation was introduced into p94I-II, its active site knockout mutant (p94I-II C129S), and IS1 by site-directed mutagenesis using the single-stranded DNA method of Kunkel et al. (26Kunkel T.A. Bebenek K. McClary J. Methods Enzymol. 1991; 204: 125-139Crossref PubMed Scopus (633) Google Scholar). The mutants were expressed and purified using the same procedure as the wild-type proteins. Enzyme Assays—Autoproteolytic activity was measured by incubating either p94I-II or its deletion construct p94I-II ΔNS in 50 mm HEPES-NaOH (pH 7.6), 10 mm Ca2+ at room temperature in a final volume of 200-500 μl at a protein concentration of 0.5 mg/ml. Aliquots were removed at different times, and the reaction was stopped by the addition of 3× SDS sample buffer, 187.5 mm Tris-HCl (pH 6.8), 6% SDS, 30% glycerol, and 0.03% bromphenol blue. The progress of autolysis was analyzed by 10% SDS-PAGE and Coomassie Blue staining. The control assay was done with 10 mm EDTA in place of Ca2+. The activity of p94I-II and p94I-II ΔNS against a 0.5 mm solution of the fluorogenic peptide substrate succinyl-Leu-Tyr (SLY)-aminomethyl coumarin (MCA) was assayed in 50 mm HEPES-NaOH (pH 7.6), 200 mm NaCl, 1 mm dithiothreitol, and 10 mm Ca2+ at room temperature. MCA release was monitored over time (excitation at 360 nm and emission at 460 nm). To examine the effect of autolysis on the rate of SLY-MCA hydrolysis, 5 μm p94I-II or p94I-II ΔNS was incubated in 50 mm HEPES-NaOH (pH 7.6), 200 mm NaCl, 1 mm dithiothreitol, and 10 mm Ca2+. Aliquots were removed at different autolysis times and assayed for hydrolysis of SLY-MCA (0.5 mm). The initial rate of MCA release was calculated for each autolysis time point, and the degree of autolysis was analyzed by SDS-PAGE. The effect of two calpain inhibitors, leupeptin and E-64, on the hydrolysis of SLY-MCA by p94I-II and p94I-II ΔNS was tested at different autolysis times. MCA release was monitored for 20-30 min before the addition of a 40-fold molar excess of inhibitor. Binding of p94I-II Autolytic Fragments to a Nickel Chelation Matrix—Autolysis fragments produced after 24 h of incubation of p94I-II in 50 mm HEPES-NaOH (pH 7.6), 200 mm NaCl, 1 mm dithiothreitol, and 10 mm Ca2+ at room temperature were loaded on a Ni-NTA column (Qiagen) previously equilibrated in 25 mm Tris-HCl (pH 7.6), 100 mm NaCl, 5 mm imidazole, and 2% glycerol (N-buffer). Unbound material was collected in the flow-through fractions, and two wash steps with N-buffer were performed before bound material was eluted with 250 mm imidazole in N-buffer (pH 8.5). Fractions from the column were analyzed by SDS-PAGE on a 10% gel. Modeling of p94I-II ΔNS—Secondary structure predictions were performed with the web-based servers for PredictProtein (27Rost B. Proteins. 1997; : 192-197Crossref PubMed Scopus (26) Google Scholar, 28Rost B. Liu J. Nucleic Acids Res. 2003; 31: 3300-3304Crossref PubMed Scopus (180) Google Scholar), nnpredict (29Kneller D.G. Cohen F.E. Langridge R. J. Mol. Biol. 1990; 214: 171-182Crossref PubMed Scopus (647) Google Scholar), and 3D-PSSM (30Kelley L.A. MacCallum R.M. Sternberg M.J. J. Mol. Biol. 2000; 299: 499-520Crossref PubMed Scopus (1121) Google Scholar). The propensity for globularity and disorder was also predicted using the GlobPlot server. The sequence of p94I-II was modeled onto the structure of the rat μI-II protease core (31Moldoveanu T. Hosfield C.M. Lim D. Elce J.S. Jia Z. Davies P.L. Cell. 2002; 108: 649-660Abstract Full Text Full Text PDF PubMed Scopus (275) Google Scholar) using the Sybyl software program (Tripos, Inc.). Homologous regions were restrained to the rat μI-II structure. The structure of IS1 was manually built into place and subjected to energy minimization and molecular dynamics. CD Spectroscopy—CD spectra were recorded on an Olis RSM 1000 CD spectrophotometer using a 1-mm path length quartz cell. Data collection (260-180 nm) was done at 4 °C in 10 mm sodium phosphate buffer (pH 7.6), with and without 10% and 50% trifluoroethanol (TFE). The protein concentration of HPLC-pure IS1 polypeptide and its L286P mutant was 0.25 mg/ml. CD spectra were averages of at least 10 scans corrected for buffer background signal. Deconvolution of the spectra was done using the CDNN software (32Bohm G. Muhr R. Jaenicke R. Protein Eng. 1992; 5: 191-195Crossref PubMed Scopus (1022) Google Scholar). The results are expressed as mean residue molar ellipticity [Θ] (degrees cm2 dmol-1). Limited Proteolysis—A time course of limited proteolytic digestion of the inactive mutant of p94I-II by trypsin or chymotrypsin was performed at room temperature in 50 mm HEPES-NaOH (pH 7.6), 1 mm Ca2+. The inactive mutant was used as substrate to avoid the complication of autolytic processing. The final concentration of p94I-II C129S was 1 mg/ml, whereas the concentrations of trypsin and chymotrypsin were 1 and 100 μg/ml, respectively. Aliquots were removed at different time points, and the proteolytic fragments were analyzed by 10% SDS-PAGE and Coomassie Blue staining. Protein fragments were transferred to polyvinylidene difluoride membranes and immunoblotted with an anti-IS1 antibody (RP2-CP3; Triple Point Biologics, Portland, OR) to identify the IS1-linked fragments. N-terminal Sequencing and Mass Analysis—The protein fragments generated by tryptic and chymotryptic proteolysis were electroblotted onto a polyvinylidene difluoride membrane using 10 mm CAPS-NaOH (pH 11.0), 10% methanol as transfer buffer and visualized using Coomassie Blue R-250 (0.1% dye in 50% methanol). After destaining, the membrane was washed several times with distilled water. Bands corresponding to the main proteolytically resistant products were cut out for N-terminal sequencing (Alberta Peptide Institute, Edmonton, Alberta, Canada). Quadrupole time-of-flight liquid chromatography-mass spectrometry spectra of p94I-II fragments after 24 h of autolysis were acquired at the Protein Function Discovery Facility, Queen's University. p94I-II Gains Activity during Autolysis—Previously, this laboratory showed that the protease core of p94 (p94I-II) can autolyze in the presence of Ca2+ and that the first cuts made (near the N-terminal ends of NS and IS1) were strictly intramolecular (25Rey M.A. Davies P.L. FEBS Lett. 2002; 532: 401-406Crossref PubMed Scopus (39) Google Scholar). It was not possible to say whether the subsequent cuts, made near the C-terminal ends of NS and IS1, were intra- or intermolecular. We noted, however, that the combination of early and late cuts had the potential to remove NS and IS1, and speculated that either or both of these sequences might block access to the active site of p94 and prevent it from cleaving exogenous substrates. To test this hypothesis, we incubated p94I-II with the small synthetic calpain substrate SLY-MCA. At the point when Ca2+ was added, there was no hydrolysis of SLY-MCA by p94I-II (Fig. 2A). Cleavage was detected soon after, however, and gradually increased in rate over 25 min. All activity was then abruptly stopped by the addition of 10 mm EDTA. Mg2+ did not substitute for Ca2+ in priming the reaction, nor did it interfere with the priming when Ca2+ was subsequently added. The initial rate of SLY-MCA hydrolysis catalyzed by p94I-II was very weak in comparison with that of the proteolytic core of the μ-calpain standard (μI-II), even when assayed at a 10- or 20-fold higher concentration (Fig. 2B). However, the enzymatic activity of p94I-II improved considerably when it was allowed to autoproteolyze before the addition of exogenous substrate (Fig. 3A, inset). The plot of initial reaction rates versus autolysis times showed that the beneficial effect of autolytic preincubation on enzymatic activity was maximal between 10 and 24 h of autolysis (Fig. 3A). SDS-PAGE showed a correlation between the gain in SLY-MCA hydrolyzing activity and the extent of p94I-II autoproteolysis. The generation of fragments of ∼30 and 17 kDa accompanied the early, rapid increase in enzymatic activity in the first hours of autolysis (Fig. 3B). The 30-kDa fragment is bounded by cleavages at 1a and 1b near the N termini of NS and IS1, respectively (Fig. 1A), and the 17-kDa fragment is the C-terminal portion of p94I-II from 1b onward (25Rey M.A. Davies P.L. FEBS Lett. 2002; 532: 401-406Crossref PubMed Scopus (39) Google Scholar). Further increase in enzymatic activity coincided with the generation of two new cleavage products at ∼26 and 13 kDa, which correspond to protein fragments generated after autolytic cleavage at or near the C termini of NS and IS1 (2a and 2b), respectively. Cleavages at positions 1a and 2a and at positions 1b and 2b have the potential to release polypeptides corresponding to major portions of NS and IS1, respectively (Fig. 1A). The generation of these fragments was confirmed by mass spectrometry. Liquid chromatography-mass spectrometry analysis of the 24 h autolysis sample showed prominent peaks of masses 4500.96 and 2829.28 Da consistent with the theoretical masses of the IS1 peptide from 1b to 2b (4502.97 Da) and the NS peptide from 1a to one amino acid before 2a (2830.07 Da) (data not shown). Autolysis at the two internal sites (1b and 2b) that release the IS1 peptide did not, however, result in dissociation of the 26- and 13-kDa fragments of the enzyme. This was demonstrated by chromatography of the 24 h autolysis products on a Ni-NTA resin. The material that bound to the column and was eluted by imidazole was indistinguishable from the 24 h autolysis products in the number and stoichiometry of fragments detected by SDS-PAGE (Fig. 4A). In particular, the 26-kDa fragment corresponding to the region between NS and IS1 lacks a His tag and could only be retained on the column if this fragment remained tightly bound to the His-tagged C-terminal fragment (13 kDa). Analysis of the three-dimensional structure of the μ-calpain protease core (31Moldoveanu T. Hosfield C.M. Lim D. Elce J.S. Jia Z. Davies P.L. Cell. 2002; 108: 649-660Abstract Full Text Full Text PDF PubMed Scopus (275) Google Scholar) shows that the equivalent regions of μI-II (shown in red and yellow in Fig. 4B) lie on either side of the loop formed by residues Asn253 to Arg267 (shown in green in Fig. 4B) and display extensive interactions within the overall protein fold. These interactions are composed of a large interdomain surface and the insertion of β-strand 14 between β-strands 7 and 8 to form a β-sheet (Fig. 4B). According to model building, these interactions are largely conserved in p94I-II (7Jia Z. Petrounevitch V. Wong A. Moldoveanu T. Davies P.L. Elce J.S. Beckmann J.S. Biophys. J. 2001; 80: 2590-2596Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar). p94I-II Active Site Becomes More Accessible to Inhibitors after Extensive Autolysis—Another demonstration of the increasing accessibility of the active site during autolysis was provided by inhibitor studies. A 40-fold molar excess of the calpain active site inhibitors, E-64 and leupeptin, had relatively little inhibitory effect on SLY-MCA hydrolysis by p94I-II during the early stages of Ca2+ addition (Fig. 5, A and B, traces 0.5), whereas hydrolysis of this substrate was completely (E-64) or largely (leupeptin) inhibited when the same concentration of inhibitors was added to a p94I-II sample that had been in the presence of Ca2+ for 24 h (Fig. 5, A and B, traces 24). An intermediate degree of inhibition was observed after the enzyme was incubated with Ca2+ for 2, 4, and 10 h (Fig. 5, A and B, traces 2, 4, and 10), suggesting that the active site of p94I-II becomes increasingly available to small inhibitors as autolysis progresses. Fig. 5C shows the plots of activation (circles) and inhibition by leupeptin (triangles) and E-64 (squares) as a function of time of autolysis. There is a consistent reciprocal trend between p94 activation and inhibition. SDS-PAGE analysis of autolysis of p94I-II in the
Referência(s)