The calpain—calpastatin system in vascular smooth muscle

Vascular smooth muscle contains large amounts of the Ca²⁺-dependent protease calpain II. In this study, we compared bovine aortic muscle (muscle phenotype) to cultured bovine aortic cells of smooth muscle origin (modulated phenotype) with respect to major constituents of the calpain—calpastatin system. Bovine aortic muscle contained only calpain II by activity measurements, Western blot of tissue extracts and Northern blot of poly(A)⁺ RNA. On the other hand, using the same methodologies, both calpains I and II as well as the 110 kDa inhibitor protein, calpastatin, were identified in cultured bovine aortic cells of smooth muscle origin. We conclude that the phenotypic state of smooth muscle cells is associated with differential expression of major components of the calpain—calpastatin system. Moreover, bovine aortic muscle is the only tissue identified to date that contains calpain II exclusively.     

Molecular cloning and characterization of functional domains of a human testis-specific isoform of calpastatin

Human serum containing sperm-agglutinating antibodies was used to screen a testis cDNA expression library to identify the cognate antigens that may be responsible for this biological effect. The longest positive phage clone (1.9 kb) was sequenced and found to be a testis-specific isoform of calpastatin (tCAST). The testis-specific segment of tCAST is encoded by a single exon within intron 14 of the calpastatin gene. A unique protein isoform is produced that differs in domain structure from the somatic calpastatins (sCAST). Human sCAST most commonly has an N-terminal domain L plus the four functional calpain inhibitory domains. Human tCAST consists of a 40-amino-acid N-terminal T domain plus a part of domain II and all of domains III and IV from the somatic isoform. Our data show that the T domain can target cytosolic localization and membrane association of tCAST, whereas domain I of sCAST exhibits a nuclear localization function. Calpastatin is the endogenous inhibitor of calpain. The calpain/calpastatin system is involved in membrane fusion events for several cell types, and calpain has been localized to the sperm acrosome. We detected tCAST in human sperm and testes extracts by Western blotting with specific antisera. These observations suggest that tCAST may modulate calpain in the calcium-mediated acrosome reaction that is required for fertilization.     

Expression of calpastatin isoforms in muscle and functionality of multiple calpastatin promoters

Calpastatin is the specific endogenous inhibitor of calpain proteinase that is encoded by a single gene. Transient transfection assays in both a non-fusing skeletal muscle and non-muscle cell-line demonstrated that the putative porcine calpastatin promoter regions 5' to exons 1xa, 1xb, and 1u were functional. Both real-time quantitative and semi-quantitative RT-PCR on porcine skeletal muscle total RNA indicated that steady-state expression of Type I and III mRNAs containing exons 1xa and 1u, respectively, was at equivalent levels whilst the expression of Type II mRNA containing exon 1xb was significantly less (p<0.001). Immunoprobing of Western blotted muscle extracts with an antibody raised against a peptide sequence encoded by exon 1xa indicated that Type I protein was expressed and that there was significantly more Type I protein in cardiac than skeletal muscle (p<0.001). The results suggest that the expression of the single calpastatin gene was differentially controlled at several levels.     

Decreased expression of high-molecular-weight calmodulin-binding protein and its correlation with apoptosis in ischemia-reperfused rat heart

A cardiac high-molecular-weight calmodulin-binding protein (HMWCaMBP) was previously identified as a homologue of the calpain inhibitor, calpastatin. In the present study, we investigated the expression of HMWCaMBP and calpains in rat heart after ischemia and reperfusion. Western blot analysis of normal rat heart extract with a polyclonal antibody raised against bovine HMWCaMBP indicated a prominent immunoreactive band of 140kDa. Both the expression and the activity of HMWCaMBP were decreased by ischemia reperfusion. Immunohistochemical studies showed strong-to-moderate HMWCaMBP immunoreactivity in normal heart and poor immunoreactivity in ischemia-reperfused heart muscle. However, the expression of micro-calpain and m-calpain in ischemia-reperfused heart was increased as compared to normal heart. The calpain inhibitory activity of ischemia-reperfused heart tissues was significantly lower as compared to normal heart tissues. The pre-ischemic and post-ischemic perfusion of hearts with a cell-permeable calpain inhibitor suppressed the increase in calpain expression but increased the HMWCaMBP expression. In-vitro HMWCaMBP was proteolyzed by micro-calpain and m-calpain. We also measured apoptosis in normal and ischemia-reperfused tissues. An increase in the number of apoptotic bodies was observed with increased duration of ischemia and reperfusion. Bcl-2 expression did not change in any of the groups, whereas Bax expression increased with ischemia-reperfusion and correlated well with the degree of apoptosis. Our findings suggest that HMWCaMBP may sequester calpains from its substrates in the normal myocardium, but it is susceptible to proteolysis by calpains during ischemia-reperfusion. Thus, decreased expression of HMWCaMBP may play an important role in myocardial injury.     

Distribution of calpains and calpastatin in human blood cells

The occurrence and molecular sizes of calpains and calpastatin in the lysates of human erythrocytes, platelets, lymphocytes/monocytes, and polymorphonuclear cells were studied by immunoelectrophoretic blot analysis. The basic uniformity among these cells of the 85-kDa and 83-kDa heavy subunits of low- and high-Ca2+-requiring calpains I and II, respectively, and of the 29-kDa light subunit was confirmed. Molecular diversity of calpastatin species, ranging from 70 kDa to 107 kDa, among different blood cells was also shown. The obtained data are consistent with those known for other animal tissues, thus settling hitherto uncertain or rather controversial issues on the distribution of calpains and calpastatin in human blood cells.     

Extracellular appearance of calpain and calpastatin in the synovial fluid of the knee joint

Extracellular location of calpain and calpastatin was demonstrated in the cell-free synovial fluid obtained from the knee joint of healthy adult humans and several patients with rheumatoid arthritis. Calpains I and II and a few molecular species of calpastatin were identified by chromatographies on DEAE-cellulose and on Ultrogel AcA 34 columns as well as by immunoelectrophoretic blot analysis. Calpains I and II in the synovial fluid of the patients increased 6.7 times and 3.5 times, respectively, compared with those of the control subjects. With the patients, shortening of the heavy subunits of calpains was noted. Calpastatin also increased in the patients, but it showed rather extensive fragmentation.     

Molecular diversity of calpastatin in mammalian organs

The crude homogenates of various human and porcine organs were subjected to immunoelectrophoretic blot analysis using affinity-purified anti-calpastatin antibody which specifically reacts with human erythrocyte 70 kDa calpastatin. Multiple immuno-reactive bands were revealed which ranged from 100 to 50 kDa. The results indicated the diversity of monomeric calpastatin molecules. The band patterns were different from one organ to the other. Among them, lung, heart and skeletal muscle were characterized by the predominance of 90-100 kDa calpastatin, having a common antigenicity to erythrocyte 70 kDa calpastatin. Such molecular diversity of calpastatins was also substantiated by enzymatic and chromatographic analyses.     

Calpain and calpastatin in porcine retina. Identification and action on microtubule-associated proteins.

Two forms of Ca2+-dependent cysteine proteinase (calpain, EC 3.4.22.17) and their specific endogenous inhibitor (calpastatin) were partially purified from porcine retina: calpain I (low-Ca2+-requiring form) was half-maximally activated at 8 microM-Ca2+, and calpain II (high-Ca2+-requiring form) at 250 microM-Ca2+. Both calpain I and calpain II were inhibited by calpastatin. Calpain I from porcine retina was shown to be composed of 83 000- and 29 000-Mr subunits, and calpain II of 80 000- and 29 000-Mr subunits, by the use of monospecific antibodies. Calpains I and II were both found to hydrolyse microtubule-associated proteins 1 and 2 rapidly.     

Multiple forms of calpastatin in pig brain

Pig brain was found to contain two calpain-specific, heat-stable inhibitory fractions which could be separated by DEAE-cellulose chromatography. CS-0.1, which was eluted from the column at 0.1 M NaCl, was identified as an ordinary, well-known calpastatin. CS-0.2, eluted at 0.2 M NaCl, was different from CS-0.1 in that it inhibited calpain 1 more strongly than calpain II and that it did not cross-react with anti-calpastatin antibodies. Partial purification indicated that CS-0.2 contained inhibitor proteins smaller than ordinary calpastatin, but whether they are the products derived from CS-0.1 or entirely different genetic products has not yet been determined.     

A novel N-terminal domain directs membrane localization of mouse testis-specific calpastatin

Multiple isoforms of calpastatin have been identified with unique N-terminal regions followed by identical calpain inhibitory domains (II-IV). In many instances the isoforms are cell-type specific, although the precise functional differences among these N-terminal regions are largely unknown. Here we report a germ cell-specific isoform of calpastatin (tCAST) that consists of a novel N-terminal peptide of 40 amino acids (domain T) followed by domains II to IV of somatic calpastatin (sCAST). Domain T is responsible for membrane association of tCAST through a protein modification by myristylation. Mutation of the myristylation site eliminates membrane targeting. Unlike most of the isoforms of calpastatin that are generated through alternative RNA splicing or post-translational proteolysis, the testis-specific isoform is transcribed from an intronic promoter in haploid germ cells of the testis. The intronic promoter directs specific expression of a reporter transgene in developing germ cells of the mouse testis.     

Developmental changes of calpain and calpastatin in rabbit brain

A major part of the Ca-activated proteolytic activity in the soluble fraction from rabbit brain could be due to the activity of the neutral thiol-proteases calpain I and II. The activity of calpains exceeded that of the endogenous inhibitor, calpastatin, at all developmental stages studied. The level of calpains increased rapidly from the prenatal stage to reach a peak 10–20 days postnatally. From this period the level of calpains decreased slowly to reach the adult levels. The level of calpastatin increased steadily from the prenatal stage to old age.     

Binding of calpain fragments to calpastatin

Their cDNA-derived amino acid sequences predict that the 80-kDa subunits of the micromolar and millimolar Ca(2+)-requiring forms of the Ca(2+)-dependent proteinase (mu- and m-calpain, respectively) each consist of four domains and that the 28-kDa subunit common to both mu- and m-calpain consists of two domains. The calpains were allowed to autolyze to completion, and the autolysis products were separated and were characterized by using gel permeation chromatography, calpastatin affinity chromatography, and sequence analysis. Three major fragments were obtained after autolysis of either calpain. The largest fragment (34 kDa for mu-calpain, 35 kDa for m-calpain) contains all of domain II, the catalytic domain, plus part of domain I of the 80-kDa subunit of mu- or m-calpain. This fragment does not bind to calpastatin, a competitive inhibitor of the calpains, and has no proteolytic activity in either the absence or presence of Ca2+. The second major fragment (21 kDa for mu-calpain and 22 kDa for m-calpain) contains domain IV, the calmodulin-like domain, plus approximately 50 amino acids from domain III of the 80-kDa subunit of mu- or m-calpain. The third major fragment (18 kDa) contains domain VI, the calmodulin-like domain of the 28-kDa subunit. The second and third major fragments bind to a calpastatin affinity column in the presence of Ca2+ and are eluted with EDTA. The second and third fragments are noncovalently bound, so the 80- and 28-kDa subunits of the intact calpain molecules are noncovalently bound at domains IV and VI. After separation in 1 M NaSCN, the 28-kDa subunit binds completely to calpastatin, approximately 30-40% of the 80-kDa subunit of mu-calpain binds to calpastatin, and the 80-kDa subunit of m-calpain does not bind to calpastatin in the presence of 1 mM Ca2+.     

Calpain,calpastatin activities and ratios during myocardial ischemia-reperfusion

The purpose of this study was to test the hypothesis that myocardial ischemia-reperfusion (I/R) is accompanied by an early burst in calpain activity, resulting in decreased calpastatin activity and an increased calpain/calpastatin ratio, thereby promoting increased protein release. To determine the possibility of a calpain burst impacting cardiac calpastatin inhibitory activity, rat hearts were subjected (Langendorff) to either 45 or 60 min of ischemia followed by 30 min of reperfusion with and without pre-administration (s.c.) of a cysteine protease inhibitor (E-64c). Myocardial function, calpain activities (casein release assay), calpastatin inhibitory activity and release of CK, LDH, cTnI and cTnT were determined (n = 8 for all groups). As expected no detectable changes in calpain activities were observed following I/R with and without E-64c (p > 0.05). Both I/R conditions reduced calpastatin activity (p < 0.05) while E-64c pre-treatment was without affect, implicating a non-proteolytic event underlying the calpastatin changes. A similar result was noted for calpain–calpastatin ratios and the release of all marker proteins (p < 0.05). In regard to cardiac function, E-64c resulted in transient improvements (15 min) for left ventricular developed pressure (LVDP) and rate of pressure development (p < 0.05). E-64c had no effect on end diastolic pressure (LVEDP) or coronary pressure (CP) during I/R. These findings demonstrate that restricting the putative early burst in calpain activity, suggested for I/R, by pre-treatment of rats with E-64c does not prevent downegulation of calpastatin inhibitory activity and/or protein release despite a transient improvement in cardiac function. It is concluded that increases in calpain isoform activities are not a primary feature of I/R changes, although the role of calpastatin downregulation remains to be elucidated.     

Identification of μ-, m-calpains and calpastatin and capture of μ-calpain activation in endothelial cells

The presence of the calpain-calpastatin system in human umbilical vein endothelial cells (HUVEC) was investigated by means of ion exchange chromatography, Western blot analysis, and Northern blot analysis. On DEAE anion exchange chromatography, calpain and calpastatin activities were eluted at approximately 0.30 M and 0.15-0.25 M NaCl, respectively. For half-maximal activity, the protease required 800 μM Ca²⁺, comparable to the Ca²⁺ requirement of m-calpain. By Western blot analysis, the large subunit of μ-calpain (80 kDa) was found to be eluted with calpastatin (110 kDa). Both the large subunit of m-calpain (80 kDa) and calpastatin were detected in the respective active fractions. By Northern blot analysis, mRNAs for large subunits of μ- and m-calpains were detected in single bands, each corresponding to approximately 3.5 Kb. Calpastatin mRNA was observed in two bands corresponding to approximately 3.8 and 2.6 Kb. Furthermore, the activation of μ-calpain in HUVEC by a calcium ionophore was examined, using an antibody specifically recognizing an autolytic intermediate form of μ-calpain large subunit (78 kDa). Both talin and filamin of HUVEC were proteolyzed in a calcium-dependent manner, and the reactions were inhibited by calpeptin, a cell-permeable calpain specific inhibitor. Proteolysis of the cytoskeleton was preceded by the appearance of the autolytic intermediate form of μ-calpain, while the fully autolyzed postautolysis form of μ-calpain (76 kDa) remained below detectable levels at all time points examined. These results indicate that the calpain-calpastatin system is present in human endothelial cells and that μ-calpain may be involved in endothelial cell function mediated by Ca²⁺ via the limited proteolysis of various proteins. J. Cell. Biochem. 66:197-209, 1997. © 1997 Wiley-Liss, Inc.     

Association of the calpain/calpastatin network with subcellular organelles

The calcium-activated cysteine protease calpain is intimately involved in modulating cell adhesion and migration. The two ubiquitous isoforms of this protease, calpain I and II, are considered to be cytosolic proteins that can translocate to both focal complexes/adhesions or the plasma membrane. Using confocal microscopy and isopycnic density centrifugation, the results demonstrate that calpain I and II, the 30kDa regulatory subunit, and calpastatin associate with the endoplasmic reticulum and Golgi apparatus. Confocal microscopy reveals that calpain II colocalizes with the subcellular proteins calnexin and Rab6 in cells bound to laminin. To further verify this association, cell lysates prepared from laminin stimulated and unstimulated cells were subjected to isopycnic density centrifugation. The results reveal an increased association of calpain I, II, calpastatin, and the 30kDa regulatory subunit with the endoplasmic reticulum and Golgi apparatus as evidenced by their position in the gradient relative to calnexin, Rab6, caveolin, and beta1 integrin after laminin stimulation. This correlates with the accumulation of inducible calpain activity at the endoplasmic reticulum-Golgi apparatus interface. Further experiments established that calpain II colocalizes with phosphatidylinositol 4,5-bisphosphate. Finally, calpain II associates with membrane lipid rafts. These results provide new insights into how the calpain/calpastatin network is spatially and temporally regulated in cells binding to the extracellular matrix.     

Distributional and developmental variations of multiple forms of calpastatin in mouse brain

DEAE-cellulose chromatography of mouse brain extract demonstrated the occurrence of two calpastatin fractions, CS-0.1 and CS-0.2, with distinctly higher content of the latter. CS-0.1 emerged from the column at 0.1 M NaCl, inhibited calpain II more strongly than calpain I, and identified also immunologically with hitherto known calpastatin. CS-0.2 emerged at 0.2 M NaCl, inhibited calpain I more strongly than calpain II, and did not crossreact with anti-calpastatin antibody used. Fairly consistent amounts of CS-0.2 and calpain II were found in the brain of mice from 10 days to 10 weeks after birth, while CS-0.1 became measurable only after 4-week growth. In adult mice, CS-0.1 was highest in specific activity in brainstem, lower in cerebellum, and not detectable in cerebral hemisphere. Physiological significance of multiple forms of calpastatin and their variations found is not known.