Crystal structures of calpain-E64 and -leupeptin inhibitor complexes reveal mobile loops gating the active site

The endogenous calpain inhibitor, calpastatin, modulates some patho-physiological aspects of calpain signaling. Excess calpain can escape this inhibition and as well, many calpain isoforms and autolytically generated protease core fragments are not inhibited by calpastatin. There is a need, therefore, to develop specific, cell-permeable calpain inhibitors to block uncontrolled proteolysis and prevent tissue damage during brain and heart ischemia, spinal-cord injury and Alzheimer's diseases. Here, we report the first high-resolution crystal structures of rat mu-calpain protease core complexed with two traditional, low molecular mass inhibitors, leupeptin and E64. These structures show that access to a slightly deeper, but otherwise papain-like active site is gated by two flexible loops. These loops are divergent among the calpain isoforms giving a potential structural basis for substrate/inhibitor selectivity over other papain-like cysteine proteases and between members of the calpain family.     

Effect of Ca2+ on binding of the calpains to calpastatin

Autolyzed mu-calpain, unautolyzed mu-calpain, autolyzed m-calpain, and unautolyzed m-calpain (mu-calpain is the micromolar Ca2+-requiring proteinase, m-calpain is the millimolar Ca2+-requiring proteinase) were passed through a calpastatin-affinity column at different free Ca2+ concentrations, and binding of the calpains to calpastatin was compared with proteolytic activity of that calpain at each Ca2+ concentration. Unautolyzed m-calpain, autolyzed m-calpain, and autolyzed mu-calpain required less Ca2+ for half-maximal binding to calpastatin than for half-maximal activity. Unautolyzed mu-calpain, however, required slightly more Ca2+ for half-maximal binding to calpastatin than for half-maximal activity. Half-maximal binding of oxidatively inactivated mu- or m-calpain to calpastatin required approximately the same Ca2+ concentrations as half-maximal binding of unautolyzed mu- or m-calpain, respectively, to calpastatin. Binding of unautolyzed m-calpain and autolyzed mu-calpain to calpastatin occurred over a wide range of Ca2+ concentrations, and it seems likely that two or more Ca2+-binding sites with different Ca2+-binding constants are involved in binding of the calpains to calpastatin. Proteolytic activity occurs at different Ca2+ concentrations than calpastatin binding, suggesting a second set of Ca2+-binding sites associated with proteolytic activity. Third and fourth sets of Ca2+-binding sites may be involved in autolysis and in binding to phosphatidylinositol or cell membranes; these four Ca2+-dependent properties of the calpains may require the eight potential Ca2+-binding sites that amino acid sequences predict are present in the calpain molecules.     

Activation of calpain by Ca2+: roles of the large subunit N-terminal and domain III-IV linker peptides

The calpains are a family of cysteine proteases with closely related amino acid sequences, but a wide range of Ca(2+) requirements (K(d)). For m-calpain, K(d) is approximately 325microM, for mu-calpain it is approximately 50microM, and for calpain 3 it is not strictly known but may be approximately 0.1microM. On the basis of previous structure determination of m-calpain we postulated that two regions of the calpain large subunits, the N-terminal peptide (residues 1-20) and a domain III-IV linker peptide (residues 514-530 in m-calpain) were important in defining K(d). The mutations Lys10Thr in the N-terminal peptide, and Glu517Pro in the domain linker peptide, reduced K(d) of m-calpain by 30% and 42%, respectively, revealing that these two regions are functionally important. The increased Ca(2+)-sensitivity of these mutants demonstrate that the Lys10-Asp148 salt link and the short beta-sheet interaction involving Glu517 are factors contributing to the high K(d) of m-calpain. Though these two regions are physically remote from the active site and Ca(2+)-binding site, they play significant roles in regulating the response of calpain to Ca(2+). Differences in these interactions in mu-calpain and in calpain 3 are also consistent with their progressively lower K(d) values.     

Genistein induces Ca2+ -mediated, calpain/caspase-12-dependent apoptosis in breast cancer cells

Genistein, a soy-derived isoflavone, has been suggested for breast cancer prevention; however, use of soy products for this purpose remains controversial. Genistein has been reported to regulate growth of tumor cells, although the involved molecular mechanisms are not defined. Here we report that genistein induces apoptosis in breast cancer cells via activation of the Ca2+ -dependent proapoptotic proteases, mu-calpain, and caspase-12. The treatment of MCF-7 breast cancer cells with genistein induced a sustained increase in concentration of intracellular Ca2+ resulting from depletion of the endoplasmic reticulum Ca2+ stores. This increase in Ca2+ was associated with activation of mu-calpain and caspase-12, as evaluated with the calpain and caspase-12 substrates and antibodies to active (cleaved) forms of the enzymes. Selective inhibition of Ca2+ binding sites of mu-calpain, forced increase of the cytosolic Ca2+ buffering capacity, and caspase inhibition decreased apoptotic indices in the genistein-treated cells. Our results suggest that Ca2+ -dependent proteases are potential targets for genistein in breast cancer cells and that the cellular Ca2+ regulatory activity of genistein underlies its apoptotic mechanism.     

Mitochondrial localization of mu-calpain

Calcium-dependent cysteine proteases, calpains, have physiological roles in cell motility and differentiation but also play a pathological role following insult or disease. The ubiquitous calpains are widely considered to be cytosolic enzymes, although there has been speculation of a mitochondrial calpain. Within a highly enriched fraction of mitochondria obtained from rat cortex and SH-SY5Y human neuroblastoma cells, immunoblotting demonstrated enrichment of the 80kDa mu-calpain large subunit and 28kDa small subunit. In rat cortex, antibodies against domains II and III of the large mu-calpain subunit also detected a 40kDa fragment, similar to the autolytic fragment generated following incubation of human erythrocyte mu-calpain with Ca(2+). Mitochondrial proteins including apoptosis inducing factor and mitochondrial Bax are calpain substrates, but the mechanism by which calpains gain access to these proteins is uncertain. Mitochondrial localization of mu-calpain places the enzyme in proximity to its mitochondrial substrates and to Ca(2+) released from mitochondrial stores.     

Peptide-biphenyl hybrids as calpain inhibitors

Calpain is a cysteine protease that is activated by Ca2+. The over-activation of calpain, which occurs on increasing Ca2+ concentration, causes a variety of diseases. This paper reports experimental results on the inhibition of calpain I (mu-calpain) by peptide-biphenyl hybrids. We have found that some peptide-biphenyl hybrids, with aromatic amino acids in the peptide chains, inhibit calpain with IC50 values in the nanomolar range. Since the peptide-biphenyl hybrids reported in the present paper do not possess a reactive electrophilic functionality, we hypothesize that they interfere with the activation of calpain by Ca2+, and present experimental and computational results on the binding of peptide-biphenyl hybrids to Ca2+.     

Autolysis of mu- and m-calpain from bovine skeletal muscle

The rate of autolysis of mu- and m-calpain from bovine skeletal muscle was measured by using densitometry of SDS polyacrylamide gels and determining the rate of disappearance of the 28 and 80 kDa subunits of the native, unautolyzed calpain molecules. Rate of autolysis of both the 28 and 80 kDa subunits of mu-calpain decreased when mu-calpain concentration decreased and when beta-casein, a good substrate for the calpains, was present. Hence, autolysis of both mu-calpain subunits is an intermolecular process at pH 7.5, 0 or 25.0 degrees C, and low ionic strength. The 78 kDa subunit formed in the first step of autolysis of m-calpain was not resolved from the 80 kDa subunit of the native, unautolyzed m-calpain by our densitometer, so autolysis of m-calpain was measured by determining rate of disappearance of the 28 kDa subunit and the 78/80 kDa complex. At Ca2+ concentrations of 1000 microM or higher, neither the m-calpain concentration nor the presence of beta-casein affected the rate of autolysis of m-calpain. Hence, m-calpain autolysis is intramolecular at Ca2+ concentrations of 1000 microM or higher and pH 7.5. At Ca2+ concentrations of 350 microM or less, the rate of m-calpain autolysis decreased with decreasing m-calpain concentration and in the presence of beta-casein. Thus, m-calpain autolysis is an intermolecular process at Ca2+ concentrations of 350 microM or less. If calpain autolysis is an intermolecular process, autolysis of a membrane-bound calpain would require selective participation of a second, cytosolic calpain, making it an inefficient process. By incubating the calpains at Ca2+ concentrations below those required for half-maximal activity, it is possible to show that unautolyzed calpains degrade a beta-casein substrate, proving that unautolyzed calpains are active proteases.     

Chicken skeletal muscle has three Ca2+-dependent proteinases

Chicken breast muscle has three Ca2+-dependent proteinases, two requiring millimolar Ca2+ (m-calpain and high m-calpain) and one requiring micromolar Ca2+ (mu-calpain). High m-calpain co-purifies with mu-calpain through successive DEAE-cellulose (steep gradient), phenyl-Sepharose, octylamine agarose, and Sephacryl S-300 columns, but elutes after mu-calpain when using a shallow KCl gradient to elute a DEAE-cellulose column. The mu- and m-calpains have 80 and 28 kDa polypeptides and are analogous to the mu- and m-calpains that have been purified from bovine, porcine and rabbit skeletal muscle. High m-calpain, which seems to be a new Ca2+-dependent proteinase, is still heterogeneous after the DEAE-cellulose column eluted with a shallow KCl gradient. Additional purification through two successive HPLC-DEAE columns and one HPLC-SW-4000 gel permeation column produces a fraction having six major polypeptides and 6-8 minor polypeptides on SDS-PAGE. A 74-76 kDa polypeptide in this fraction reacts in Western blots with monospecific, polyclonal anti-calpain antibodies that react with both the 80 kDa and the 28 kDa polypeptides of mu- or m-calpain. High m-calpain also is related to mu- and m-calpain in that it causes the same limited digestion of skeletal muscle myofibrils, has a similar pH optimum near pH 7.9-8.4, requires Ca2+ for activity, and reacts with the calpain inhibitor, calpastatin, and a variety of serine and cysteine proteinase inhibitors in a manner identical to mu- and m-calpain. High m-calpain differs from mu- and m-calpain in its elution off DEAE-cellulose columns and its requirement of 3800 microM Ca2+ for one-half maximal activity compared with 5.35 microM Ca2+ for mu-calpain and 420 microM Ca2+ for m-calpain. The physiological significance of high m-calpain in unclear. The presence of mu-calpain in chicken breast muscle suggests that all skeletal muscles contain both mu- and m-calpain, although the relative proportions of these two proteinases may vary in different species.     

Cystatins as calpain inhibitors: engineered chicken cystatin- and stefin B-kininogen domain 2 hybrids support a cystatin-like mode of interaction with the catalytic subunit of mu-calpain

Within the cystatin superfamily, only kininogen domain 2 (KD2) is able to inhibit mu- and m-calpain. In an attempt to elucidate the structural requirements of cystatins for calpain inhibition, we constructed recombinant hybrids of human stefin B (an intracellular family 1 cystatin) with KD2 and deltaL110 deletion mutants of chicken cystatin-KD2 hybrids. Substitution of the N-terminal contact region of stefin B by the corresponding KD2 sequence resulted in a calpain inhibitor of Ki = 188 nM. Deletion of L110, which forms a beta-bulge in family 1 and 2 cystatins but is lacking in KD2, improved inhibition of mu-calpain 4- to 8-fold. All engineered cystatins were temporary inhibitors of calpain due to slow substrate-like cleavage of a single peptide bond corresponding to Gly9-Ala10 in chicken cystatin. Biomolecular interaction analysis revealed that, unlike calpastatin, the cystatin-type inhibitors do not bind to the calmodulin-like domain of the small subunit of calpain, and their interaction with the mu-calpain heterodimer is completely prevented by a synthetic peptide comprising subdomain B of calpastatin domain 1. Based on these results we propose that (i) cystatin-type calpain inhibitors interact with the active site of the catalytic domain of calpain in a similar cystatin-like mode as with papain and (ii) the potential for calpain inhibition is due to specific subsites within the papain-binding regions of the general cystatin fold.     

Effect of mu-calpain on m-calpain

The free Ca(2+) concentrations required for half-maximal proteolytic activity of m-calpain are in the range of 400-800 microM and are much higher than the 50-500 nM free Ca(2+) concentrations that exist in living cells. Consequently, a number of studies have attempted to find mechanisms that would lower the Ca(2+) concentration required for proteolytic activity of m-calpain. Although autolysis lowers the Ca(2+) concentration required for proteolytic activity of m-calpain, 90-400 microM Ca(2+) is required for a half-maximal rate of autolysis of m-calpain, even in the presence of phospholipid. It has been suggested that mu-calpain, which has a lower Ca(2+) requirement than m-calpain, might proteolyze m-calpain and reduce its Ca(2+) requirement to a level that would allow it to be active at physiological Ca(2+) concentrations. We have incubated m-calpain with mu-calpain for 60 min at a ratio of 1:50 mu-calpain:m-calpain, in the presence of 50 microM free Ca(2+); this Ca(2+) concentration is high enough for more than half-maximal activity of mu-calpain, but does not activate m-calpain. Under these conditions, mu-calpain caused no detectable proteolytic degradation of the m-calpain polypeptide and did not change the Ca(2+) concentration required for proteolytic activity of m-calpain. mu-Calpain also did not degrade the m-calpain polypeptide at 1000 microM Ca(2+), which is a Ca(2+) concentration high enough to completely activate m-calpain. It seems unlikely that mu-calpain could act as an "activator" of m-calpain in living cells. Because m-calpain rapidly degrades itself (autolyzes) at 1000 microM Ca(2+) and because the subsite specificities of mu- and m-calpain are very similar if not identical, failure of mu-calpain to rapidly degrade m-calpain at 1000 microM Ca(2+) suggests a unique role of autolysis in calpain function.     

Ritonavir does not inhibit calpain in vitro

Ritonavir, an inhibitor of HIV-1 protease, has been reported to also inhibit the Ca2+-dependent cysteine protease, calpain. We have investigated these claims with an in vitro study of the effect of ritonavir on the m-calpain and mu-calpain isoforms. Ritonavir failed to block either autolytic or hydrolytic calpain activity, but remained fully capable of inhibiting the HIV-1 protease. Any calpain-related effects of ritonavir in cells must, therefore, arise by a mechanism other than direct inhibition of calpains.     

Identification of calpain cleavage sites in the G1 cyclin-dependent kinase inhibitor p19(INK4d)

Calpains are a large family of Ca2+-dependent cysteine proteases that are ubiquitously distributed across most cell types and vertebrate species. Calpains play a role in cell differentiation, apoptosis, cytoskeletal remodeling, signal transduction and the cell cycle. The cell cycle proteins cyclin D1 and p21(KIP1), for example, have been shown to be affected by calpains. However, the rules that govern calpain cleavage specificity are poorly understood. We report here studies on the pattern of mu-calpain proteolysis of the p19(INK4d) protein, a cyclin-dependent kinase 4/6 inhibitor that negatively regulates the mammalian cell cycle. Our data show new characteristics of calpain action: mu-calpain cleaves p19(INK4d) immediately after the first and second ankyrin repeats that are structurally less stable compared to the other repeats. This is in contrast to features observed so far in the specificity of calpains for their substrates. These results imply that calpain may be involved in the cell cycle by regulating the cell cycle regulatory protein turnover through CDK inhibitors and cyclins.     

Mu-calpain is functionally required for alpha-processing of Alzheimer's beta-amyloid precursor protein

Alzheimer's beta-amyloid precursor protein (APP) is normally processed by an unidentified alpha-secretase. A unique feature of this protease is its high sensitivity to phorbol esters, yet the mechanism involved is unclear. We have previously reported that phorbol 12,13-dibutyrate (PDBu) activates calpain, a Ca2+-dependent protease, and PDBu-induced release of APPs (secreted APP) is sensitive to calpain inhibitors, suggesting that calpain is involved in APP alpha-processing. In the present study, we found that PDBu markedly promoted the expression of both mu- and m-calpains in cultured fibroblasts. Dose-response and time course studies revealed that mu-calpain was more sensitive to PDBu than m-calpain and the temporal course of the mu-calpain change coincides better with that of APPs release. Moreover, the stimulatory effect of PDBu on mu-calpain was selectively blocked by mu-calpain-specific siRNA (small interference RNA) and the blockage was accompanied by a concomitant decrease in APPs release. In contrast, m-calpain siRNA did not affect APPs release significantly. Measurement of amyloid beta protein (Abeta) release in the mu-calpain siRNA-treated cells indicated that Abeta40 and Abeta42 levels inversely changed in relation to APPs, and the changes in Abeta42 were more prominent than in Abeta40. Together, these data suggest that calpain, particularly mu-calpain, is a potential candidate for alpha-secretase in the regulated APP alpha-processing, and that changes in this protease can affect the outcome of the overall APP processing.     

The calpain system in human placenta

The calpain system is involved in a number of human pathologies ranging from the muscular dystrophies to Alzheimer's disease. It is important, therefore, to be able to obtain and to characterize both mu-calpain and m-calpain from human tissue. Although human mu-calpain can be conveniently obtained from either erythrocytes or platelets, no readily available source of human m-calpain has been described. Human placenta extracts contain both mu-calpain and m-calpain in nearly equal proportions and in significant quantities (3-4 mg mu-calpain and 4-5 mg m-calpain/1000 g placenta tissue). Placenta also contains calpastatin that elutes off ion-exchange columns over a wide range of KCl concentrations completely masking the mu-calpain activity eluting off these columns and even partly overlapping m-calpain elution. Placenta mu-calpain requires 50-70 microM Ca2+ and placenta m-calpain requires 450-460 microM Ca2+ for half-maximal proteolytic activity. Western analysis of washed placenta tissue shows that placenta contains both mu- and m-calpain, although some of the mu-calpain in whole placenta extracts likely originates from the erythrocytes that are abundant in the highly vascularized placenta. Placenta calpastatin could not be purified with conventional methods. The most prominent form of calpastatin in Western analyses of placenta obtained as soon as possible after birth was approximately 48-51 kDa; partly purified preparations of placenta calpastatin also contained 48-51 and 70 kDa polypeptides. Human placenta extracts likely contain two different calpastatin isoforms, a 48-51 kDa "placenta calpastatin" and a 70 kDa erythrocyte calpastatin.     

Synthesis and biological evaluation of chromone carboxamides as calpain inhibitors

Excessive calpain activations contribute to serious cellular damage and have been found in many pathological conditions. Novel chromone carboxamides derived from ketoamides were prepared and evaluated for mu-calpain inhibition. Among synthesized, compound 2i was the most potent calpain inhibitor with an IC(50) value of 0.24 +/- 0.11 microM comparable to the activity of peptide aldehyde calpain inhibitor MDL 28,170. Furthermore, compound 2i showed higher selectivity for mu-calpain over two related cysteine proteases cathepsin B and cathepsin L, suggesting the chromone ring as a good scaffold for selective mu-calpain inhibitors.     

Binding of reversible spin-labeled inhibitors with an butyrylcholinesterase active center

The interaction of spin-labeled metacyn, procaine, carbolin and bivalent cations (Ca2+, Co2+, Ni2+) with butyrylcholinesterase (BChE) was studied by ESR and enzyme kinetic methods. The effect of pH, ionic strength and organic solvent was analysed. Spin-labeled metacyn binds at the anionic site of BChE active centre. This complex is stabilized both with coulombic and hydrophobic interactions, ionizing group of active centre with pK 6-7 also affects the binding. Spin-labeled procaine appeared to be enzyme competitive inhibitor (Ki = 4 X 10(-5) M) and is located, most probably, at the same site. Activating effect of Ca2+ ions on BChE was confirmed. Simultaneous application of spin labels and paramagnetic ions demonstrates that cations Co2+ and Ni2+ bind with BChE in the close vicinity of spin-labeled inhibitor site. Paramagnetic cations are located more closely to the cationic part of the inhibitor molecule than to the hydrophobic one, and can be displaced by surplus of Ca2+ ions. The experimental data testify the model of anionic centre which consists of bivalent metal ions and aminoalcyl cationic group subsites and is located in a hydrophobic pocket of the enzyme surface.     

N terminus of calpain 1 is a mitochondrial targeting sequence

The ubiquitous m- and mu-calpains are thought to be localized in the cytosolic compartment, as is their endogenous inhibitor calpastatin. Previously, mu-calpain was found to be enriched in mitochondrial fractions isolated from rat cerebral cortex and SH-SY5Y neuroblastoma cells, but the submitochondrial localization of mu-calpain was not determined. In the present study, submitochondrial fractionation and digitonin permeabilization studies indicated that both calpain 1 and calpain small subunit 1, which together form mu-calpain, are present in the mitochondrial intermembrane space. The N terminus of calpain 1 contains an amphipathic alpha-helical domain, and is distinct from the N terminus of calpain 2. Calpain 1, but not calpain 2, was imported into mitochondria. Removal of the N-terminal 22 amino acids of calpain 1 blocked the mitochondrial calpain import, while addition of this N-terminal region to calpain 2 or green fluorescent protein enabled mitochondrial import. The N terminus of calpain 1 was not processed following mitochondrial import, but was removed by autolysis following calpain activation. Calpain small subunit 1 was not directly imported into mitochondria, but was imported in the presence of calpain 1. The presence of a mitochondrial targeting sequence in the N-terminal region of calpain 1 is consistent with the localization of mu-calpain to the mitochondrial intermembrane space and provides new insight into the possible functions of this cysteine protease.     

Calpain mutants with increased Ca2+ sensitivity and implications for the role of the C(2)-like domain

The ubiquitous calpain isoforms (mu- and m-calpain) are Ca(2+)-dependent cysteine proteases that require surprisingly high Ca(2+) concentrations for activation in vitro ( approximately 50 and approximately 300 microm, respectively). The molecular basis of such a high requirement for Ca(2+) in vitro is not known. In this study, we substantially reduced the concentration of Ca(2+) required for the activation of m-calpain in vitro through the specific disruption of interdomain interactions by structure-guided site-directed mutagenesis. Several interdomain electrostatic interactions involving lysine residues in domain II and acidic residues in the C(2)-like domain III were disrupted, and the effects of these mutations on activity and Ca(2+) sensitivity were analyzed. The mutation to serine of Glu-504, a residue that is conserved in both mu- and m-calpain and interacts most notably with Lys-234, reduced the in vitro Ca(2+) requirement for activity by almost 50%. The mutation of Lys-234 to serine or glutamic acid resulted in a similar reduction. These are the first reported cases in which point mutations have been able to reduce the Ca(2+) requirement of calpain. The structures of the mutants in the absence of Ca(2+) were shown by x-ray crystallography to be unchanged from the wild type, demonstrating that the increase in Ca(2+) sensitivity was not attributable to conformational change prior to activation. The conservation of sequence between mu-calpain, m-calpain, and calpain 3 in this region suggests that the results can be extended to all of these isoforms. Whereas the primary Ca(2+) binding is assumed to occur at EF-hands in domains IV and VI, these results show that domain II-domain III salt bridges are important in the process of the Ca(2+)-induced activation of calpain and that they influence the overall Ca(2+) requirement of the enzyme.     

Identification of two calpastatin forms in rat skeletal muscle and their susceptibility to digestion by homologous calpains

Two forms of calpastatin, differing in their specificity for the homologous calpain isozymes I and II, have been separated from rat skeletal muscle extracts and purified to homogeneity. Calpastatin I, the first form to elute in chromatography on DE32, is more effective against calpain I, while calpastatin II is more effective as an inhibitor of calpain II. Based on their molecular mass (approximately 105 kDa) both calpastatin forms belong to the high molecular mass class found in muscles of other animal species (Murachi, T., 1989, Biochem. Int. 18, 263-294). For calpain I, which is active with low (mu-M) concentrations of Ca2+, maximum inhibition with either calpastatin form was observed over a wide range of Ca2+ concentrations. With calpain II, which requires high (mM) concentrations of Ca2+ for activity, maximum inhibition required Ca2+ concentrations above 1 mM. Both calpastatin forms were found to be highly sensitive to degradation by calpain II, but almost completely resistant to degradation by calpain I. Degradation of calpastatin by calpain II is competitively inhibited by the addition of a calpain substrate. Isovaleryl carnitine (IVC), an intermediate product of L-leucine catabolism, previously demonstrated to be a potent and specific activator of rat skeletal muscle calpain II (Pontremoli, S., Melloni, E., Viotti, P. L., Michetti, M., Di Lisa, F., and Siliprandi, N., 1990. Biochem. Biophys. Res. Commun. 167, 373-380) greatly enhances the rate of degradation of calpastatins by calpain II. IVC, which decreases the Ca2+ requirement for maximal calpain II activity, also decreases the concentration of Ca2+ required for digestion of the inhibitor. For calpain II, regulation by either calpastatins may occur only in the presence of high [Ca2+].     

Mu-calpain binds to lipid bilayers via the exposed hydrophobic surface of its Ca2+-activated conformation

Mu- and m-calpain are cysteine proteases requiring micro- and millimolar Ca2+ concentrations for their activation in vitro. Among other mechanisms, interaction of calpains with membrane phospholipids has been proposed to facilitate their activation by nanomolar [Ca2+] in living cells. Here the interaction of non-autolysing, C115A active-site mutated heterodimeric human mu-calpain with phospholipid bilayers was studied in vitro using protein-to-lipid fluorescence resonance energy transfer and surface plasmon resonance. Binding to liposomes was Ca2+-dependent, but not selective for specific phospholipid head groups. [Ca2+]0.5 for association with lipid bilayers was not lower than that required for the exposure of hydrophobic surface (detected by TNS fluorescence) or for enzyme activity in the absence of lipids. Deletion of domain V reduced the lipid affinity of the isolated small subunit (600-fold) and of the heterodimer (10- to 15-fold), thus confirming the proposed role of domain V for membrane binding. Unexpectedly, mutations in the acidic loop of the 'C2-like' domain III, a putative Ca2+ and phospholipid-binding site, did not affect lipid affinity. Taken together, these results support the hypothesis that in vitro membrane binding of mu-calpain is due to the exposed hydrophobic surface of the active conformation and does not reduce the Ca2+ requirement for activation.