Báo cáo Y học: Domain V of m-calpain shows the potential to form an oblique-orientated a-helix, which may modulate the enzyme’s activity via interactions with anionic lipid potx

Domain V of m-calpain shows the potential to form an oblique-orientated a-helix, which may modulate the enzyme’s activity via interactions with anionic lipid Klaus Brandenburg 1 , Frederick Harris 2 , Sarah Dennison 2 , Ulrich Seydel 1 and David Phoenix 2 1 Division of Biophysics, Forschunginstitute Borstel, Germany; 2 Department of Forensic and Investigative Science, University of Central Lancashire, Preston, UK The activity of m-calpain, a heterodimeric, Ca 2+ -dependent cysteine protease appears to be modulated by membrane interactions involving oblique-orientated a-helix formation by a segment, GTAMRILGGVI, in the protein’s smaller subunit. Here, graphical and hydrophobic moment-based analyses predicted that this segment may form an a-helix with strong structural resemblance to the influenza virus peptide, HA2, a known oblique-orientated a-helix former. Fourier transform infrared spectroscopy showed that a peptide homologue of the GTAMRILGGVI segment, VP1, adopted low levels of a-helical structure ( 20%) in the presence of zwitterionic lipid and induced a minor decrease (3 °C) in the gel to liquid-crystalline phase transition tem- perature, T C , of the hydrocarbon chains of zwitterionic membranes, suggesting interaction with the lipid headgroup region. In contrast, VP1 adopted high levels of a-helical structure (65%) in the presence of anionic lipid, induced a large increase (10 °C) in the T C of anionic membranes, and showed high levels of anionic lipid monolayer penetration (DSP ¼ 5.5 mNÆm )1 ), suggesting deep levels of membrane penetration. VP1 showed strong haemolytic ability (LD 50 ¼ 1.45 m M ), but in the presence of ionic agents, this ability, and that of VP1 to penetrate anionic lipid mono- layers, was greatly reduced. In combination, our results suggest that m-calpain domain V may penetrate membranes via the adoption of an oblique-orientated a-helix and elec- trostatic interactions. We speculate that these interactions may involve snorkelling by an arginine residue located in the polar face of this a-helix. Keywords: domain V; hydrophobicity gradient; m-calpain; membrane; oblique-orientated a-helix. Calpains are a growing family [1] of structurally related intracellular Ca 2+ -dependent cysteine proteases [2,3], with calpain 10 the most recently characterized member [4]. The physiological functions of calpains are not fully understood but they are believed to play important roles in such processes as cytoskeletal remodelling, cell differentiation, apoptosis and signal transduction [5–7]. Calpains are also of medical importance, having been implicated in a number of pathological conditions including: cataract formation [8], type 2 diabetes [9], muscular dystrophy, rheumatoid arth- ritis, ischaemic tissue damage, and neurodegenerative con- ditions such as Alzheimer’s and Parkinson’s disease [10,11]. The major calpains are l-calpain (calpain 1) and m-calpain (calpain 2), which are ubiquitous in mammalian cells [10]. These enzymes are heterodimeric and possess larger, 80-kDa, subunits, which show high levels of homo- logy, and smaller, 30-kDa, subunits, which show lower levels of homology [2,3]. Originally based on sequence comparisons [12], these calpains were assigned a domainal organization, with the larger subunit divided into domains I to IV and the smaller subunit divided into domains V and VI. Domain II possesses the active site and is a papain-like cysteine protease domain, and domain IV contains a calmodulin-like Ca 2+ -binding domain with multiple EF-hand motifs. The smaller subunit is divided into domain VI, which also possesses a calmodulin-like Ca 2+ -binding domain, and domain V [2,3,13]. The recently solved crystal structures of calcium-free rat m-calpain [14] and human m-calpain [15] confirmed that this domain structure is essentially correct. l-Calpain and m-calpain possess similar substrate spe- cificity and show an absolute requirement for Ca 2+ for activation, although they differ in the level of this require- ment: 5–50 l M and 250–1000 l M , respectively [2,3]. These observations suggested that other factors may be involved in the activation of m-calpain because, clearly, the millimolar levels of Ca 2+ needed to activate the enzyme in vivo far exceed normal intracellular levels. Calpains are known to exist in a membrane-associated form [16,17], and it has been shown that the presence of membrane lipid mixtures can lower the Ca 2+ requirement for m-calpain activation to near physiological levels [18]. On the basis of these and other results [19–21], it has been suggested that lipid or membrane interaction may modulate the activity of the enzyme [3]. Consistent with this suggestion, it has been shown that m-calpain domain III folds into an antiparallel b-sandwich, which is structurally related to C2 domains [14,15,22]. These domains bind phospholipid in a Ca 2+ -dependent manner and are believed to be responsible for orchestrating the Correspondence to D. A. Phoenix, Department of Forensic and Investigative Science, University of Central Lancashire, Preston PR1 2HE, UK. Fax: + 1772 894981, Tel.: + 1772 894381, E-mail: daphoenix@uclan.ac.uk Abbreviations:Myr 2 PtdCho, dimyristoylphosphatidylcholine; Myr 2 PtdEtn, dimyristoylphosphatidylethanolamine; Myr 2 PtdSer, dimyristoylphosphatidylserine; FTIR, Fourier transform infrared; SUV, small unilamellar vesicle. (Received 20 May 2002, accepted 2 September 2002) Eur. J. Biochem. 269, 5414–5422 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.03225.x membrane–Ca 2+ regulation of enzyme activity of a number of proteins [23]. A number of studies have suggested that interaction of m-calpain domain V with lipid or membranes may also be involved in the modulation of the enzyme’s activity. Earlier investigations showed that a C-terminal segmentofdomainV,G 17 TAMRILGG, is required for lipid interaction of the enzyme [24]. More recent investiga- tions, using peptides homologous to various regions of m-calpain’s domain V, showed that, although the presence of the TAMRIL sequence was required for m-calpain–lipid interaction, the presence of a polyglycine sequence was also necessary for such interaction [25]. The recently determined crystal structures of m-calpain did not include domain V [15,16] and no three-dimensional structure for this domain is currently available. Further- more, the primary structure of domain V shows no apparent homology with that of any other known protein [22]. However, a recent theoretical analysis of domain V derived from four different mammalian m-calpains showed that each possessed a common segment, GTAMRILGGVI, which was a candidate for formation of a lipid-interactive oblique-orientated a-helix [26]. These a-helices show a highly specialized structure/function relationship and pene- trate membranes at a shallow angle because of a hydro- phobicity gradient along the a-helical long axis [27,28]. It was suggested by Daman et al.[26]thattheformationof such an a-helix may feature in the membrane interactions of m-calpain domain V. Here, to investigate this suggestion, we have undertaken theoretical analysis and studied the lipid– membrane interactions of a peptide homologue of the GTAMRILGGVI segment using haemolytic analysis, monolayer studies, Fourier transform infrared (FTIR) conformational analysis and FTIR lipid-phase transition analysis. Our results are discussed in relation to the influenza viral fusion peptide, HA2 [29,30], a known oblique-orientated a-helix former [31–35]. MATERIALS AND METHODS Reagents The peptide VP1 was supplied by Pepsyn, University of Liverpool, UK, produced by solid-state synthesis and purified by HPLC to a purity of greater than 99%. The peptide was stored as a stock solution (10 m M ) in 10% (v/v) ethanol at 4 °C. Packed human red blood cells were supplied by the Royal Preston Hospital, UK. The phos- pholipids dimyristoylphosphatidylcholine (Myr 2 PtdCho), dimyristoylphosphatidylethanolamine (Myr 2 PtdEtn) and dimyristoylphosphatidylserine (Myr 2 PtdSer) and all sol- vents, which were of spectroscopic grade, were purchased from Sigma (UK). For FTIR spectroscopy, deuterated Myr 2 PtdSer, purchased from Avanti, was used. Theoretical analyses of candidate oblique-orientated a-helix-forming segments The sequences of the influenza viral fusion peptide, HA2, a known oblique-orientated a-helix former [31–35], and that of the putative oblique-orientated a-helix of m-calpain domain V (Table 1) were analysed by conventional hydro- phobic moment methods [36]. The hydrophobicity of successive amino acids in these sequences are treated as vectors and summed in two dimensions, assuming an amino-acid side chain periodicity of 100 °. The resultant of this summation, the hydrophobic moment, l H ,providesa measure of a-helix amphiphilicity. Our analysis used a moving window of 11 residues, and, for each sequence under investigation (Table 1), the window with the highest hydrophobic moment was identified (Table 1). For these windows, the mean hydrophobic moment, Æl H æ,andthe corresponding mean hydrophobicity, ÆH 0 æ (Table 1), were computed using the normalized consensus hydrophobicity scale of Eisenberg et al. [37] and plotted on the hydrophobic moment plot diagram of Eisenberg et al. [38] as modified by Harris et al.[32](Fig.1). WINGEN / WINPEG software [39] was used to perform hydropathy plot analysis (Fig. 2) of the GTAMRILGGVI sequence using the hydrophobicity scale of Kyte & Dolittle [40] and a seven-residue window. The software was also used to represent both this sequence and that of HA2 as two-dimensional axial projections assuming an angular periodicity of 100 ° (Fig. 3). Haemolytic assay of VP1 Haemolytic assay was conducted as described by Harris & Phoenix [41]. Essentially, packed red blood cells were washed three times in Tris-buffered sucrose (0.25 M sucrose, 10 m M Tris/HCl, pH 7.5) and resuspended in the same medium to give an initial blood cell concentra- tion of  0.05%. For haemolytic assay, this concentration was adjusted such that incubation with 0.1% (v/v) Triton X-100 for 1 h produced a supernatant with A 416 ¼ 1.0, and this was taken as 100% haemolysis. Aliquots (1 mL) of blood cells at assay concentration were then used to solubilize various amounts of stock peptide solution, each of which had been added to a test-tube and dried under nitrogen gas. The resulting mixtures were incu- bated at room temperature with gentle shaking. After 1 h, the suspensions were centrifuged at low speed (1500 g, 15 min, 25 °C), and the A 416 of the supernatants Table 1. Hydrophobic moment analysis of protein structure. Primary structure of the putative oblique-orientated a-helix-forming segment identified in m-calpain, domain V [26] and that of the influenza peptide, HA2, a known oblique-orientated a-helix former, obtained from Peuvot et al. [47]. Values of Æl H æ and ÆH 0 æ for each sequence were determined by the method of Eisenberg et al.[36]. Source protein Segment sequence Æl H æÆH 0 æ + m-calpain, domain V GTAMRILGGVI 0.46 0.47 )) ) HA2 fusion peptide GLFGAIAGFIENGWEGMIDG 0.65 0.23 Ó FEBS 2002 Lipid-interactive a-helical structure in m-calpain (Eur. J. Biochem. 269) 5415 determined. Similar experiments were also performed except that Tris-buffered sucrose was replaced with Tris- buffered saline (100 m M NaCl, 10 m M Tris/HCl, pH 7.5). In all cases, levels of haemolysis were determined as the percentage haemolysis relative to that of Triton X-100 and the results recorded (Fig. 4). Basal lysis was less than 3% in all cases. Preparation of phospholipid small unilamellar vesicles (SUVs) SUVs were prepared by the method of Keller et al.[42]. Essentially, lipid/chloroform mixtures were dried with nitrogen gas and hydrated with aqueous Hepes at pH 7.5 to give final phospholipid concentrations of 50 m M .The resulting cloudy suspensions were sonicated at 4 °Cwitha Soniprep 150 sonicator (amplitude 10 lm) until clear suspensions resulted (30 cycles of 30 s), which were then centrifuged (15 min, 3000 g,4°C). FTIR conformational analyses of VP1 To give a final peptide concentration of 1 m M ,VP1was solubilized in 50 m M aqueous Hepes (pH 7.5) or suspen- sions of SUVs, which were formed from Myr 2 PtdSer, Myr 2 PtdCho or Myr 2 PtdEtn, prepared as described above. Samples of solubilized peptide were spread on a CaF 2 crystal, and the free excess water was evaporated at room temperature. The single-band components of the VP1 amide I vibrational band (predominantly C¼O stretch) was monitored using an FTIR Ô5-DXÕ spectro- meter (Nicolet Instruments, Madison, WI, USA), and, for each sample, absorbance spectra were produced (Fig. 5). These spectra were analysed, and, for those with strong absorption bands, the band parameters (peak position, band width, and intensity) were evaluated with the original spectra, if necessary after the subtraction of strong water bands. In the case of spectra with weak absorption bands, resolution-enhancement techniques such as Fourier self-deconvolution [43] were applied after baseline subtraction with the parameters: bandwidth, 22– 28 cm )1 ; resolution-enhancement factor, 1.2–1.4; Gauss/ Lorentz ratio, 0.55. In the case of overlapping bands, curve fitting was applied using a modified version of the CURFIT procedure written by D. Moffat (National Research Council, Ottowa, Ontario, Canada). An esti- mation of the number of band components was obtained from deconvolution of the spectra; curve fitting was then applied within the original spectra after the subtraction of baselines resulting from neighbouring bands. Similar to the deconvolution technique, the bandshapes of the single components are superpositions of Gaussian and Lorentzian bandshapes. Best fits were obtained by assuming a Gauss fraction of 0.55–0.6. The CURFIT procedure measures the peak areas of single band components and, after statistical evaluation, determines the relative percentages of primary structure involved in secondary-structure formation. For VP1, relative levels of a-helical structure (1650–1655 cm )1 )andb-sheet struc- tures (1625–1640 cm )1 ) were computed and are shown in Table 2. FTIR analysis of phospholipid phase transition properties Using FTIR spectroscopy, the effects of VP1 on the phase- transition properties of phospholipid were investigated. To give a final peptide concentration of 1 m M ,VP1was solubilized in suspensions of SUVs formed from Myr 2 Ptd- Ser, Myr 2 PtdCho or Myr 2 PtdEtn, prepared as described above. As controls, SUVs formed from Myr 2 PtdSer, Myr 2 PtdCho or Myr 2 PtdEtn alone were prepared as described above. These samples were then subjected to automatic temperature scans with a heating rate of 3 °CÆ(5 min) )1 and within the temperature range 0–60 °C. For every 3 °C interval, 50 interferograms were accumu- lated, apodized, Fourier transformed, and converted into Fig. 1. Hydrophobic moment plot analysis of protein segments. Acon- ventional hydrophobic moment plot diagram of Eisenberg et al. [38] is shown with an overlaid grey region delineating candidate oblique- orientated a-helices [32]. The sequences shown in Table 1 were plotted on the diagram according to their Æl H æ and corresponding ÆH 0 æ values (Table 1). The data point representing the m-calpain domain V seg- ment, GTAMRILGGVI (1), can be seen to lie in the grey region, proximal to that representing the HA2 peptide (2), indicating that the segment may be a candidate for oblique-orientated a-helix formation. Fig. 2. Hydropathy plot analysis of protein segments. Hydropathy plot analysis of the m-calpain domain V segment, GTAMRILGGVI, was performed using a seven-residue window and the software of Hennig [39]. It can be seen that hydrophobicty progressively increases along the length of the segment with a maximal value centred on the C-terminal glycine, residue 8. 5416 K. Brandenburg et al.(Eur. J. Biochem. 269) Ó FEBS 2002 absorbance spectra [44] (Fig. 6). These spectra monitored changes in the b fi a acyl chain melting behaviour of phospholipids, with these changes determined as shifts in the peak position of the symmetric stretching vibration of the methylene groups, m s (CH 2 ), which is known to be a sensitive marker of lipid order. The peak position of m s (CH 2 ) lies at 2850 cm )1 in the gel phase and shifts at a lipid-specific temperature T c to 2852.0–2852.5 cm )1 in the liquid-crystal- line state. For deuterated Myr 2 PtdSer, the values for the peak position of m s (CD 2 ) are at 2089–2093 cm )1 , respectively. Monolayer studies on VP1 All monolayer equipment was supplied by NIMA, Coven- try, UK. Surface tension was monitiored by the Wilhelmy plate method using a microbalance [45]. Studies were conducted using a 5 · 15 cm Teflon trough containing 80 mL buffer subphase (10 m M Tris/HCl, pH 7.5). The trough was equipped with moveable barriers, which responded to the microbalance and could be adjusted to maintain monolayers at either constant surface pressure or constant surface area. Monolayers were formed by spread- ing pure phospholipids (10 m M ) in chloroform, compressed to give a surface pressure of 30 mNÆm )1 ,andthen maintained at constant area. Stock VP1 was added to the subphase via a reservoir extending into the subphase, which was contiuously stirred by a magnetic bar, and VP1– monolayer interactions were recorded as changes in mono- layer surface pressure (Fig. 7). RESULTS Theoretical analyses of candidate oblique-orientated a-helix-forming segments The sequences of the influenza viral fusion peptide, HA2, and m-calpain domain V were analysed. The resulting values of Æl H æ and ÆH 0 æ (Table 1) were plotted on the hydrophobic moment plot diagram (Fig. 1). The data points representing these sequences lie proximal in the area delineating candidate oblique-orientated a-helix-forming segments, indicating that m-calpain domain V may contain an oblique-orientated a-helix comparable to that of HA2. Consistent with these results, hydropathy plot analysis of the GTAMRILGGVI sequence showed a progressive increase in hydrophobicity in moving from the N-terminus to the C-terminus (Fig. 2), suggesting the ability to form an a-helix with an asymmetric distribution of hydrophobicty along the a-helical axis. Furthermore, when the sequences of HA2 and m-calpain domain V were modelled as a-helices [39], each formed an amphiphilic a-helix with similar structural properties (Fig. 3). Each a-helix possesses a glycine-rich hydrophilic face and a wide hydrophobic face, which includes the bulky amino-acid residues tryptophan, phenylalanine, leucine and isoleucine. Fig. 3. Two-dimensional axial projections of protein sequences. Primary structures of (A) the influenza peptide, HA2, a known oblique-orientated a-helix former and (B) the putative oblique-orientated a-helix-forming segment identified in m-calpain, domain V (Table 1), represented as two- dimensional axial projections using the software of Hennig [39]. Annotated numbers represent the relative locations of amino-acid residues within protein primary structure, and hydrophobic residues are circled. It can be seen that each a-helix possesses a glycine-rich polar face and a wide hydrophobic face rich in bulky amino-acid residues. In the case of the GTAMRILGGVI segment, these residues, isoleucine (6) and leucine (7 and 11), can be seen to be localized in the C-terminal region of the a-helix. Fig. 4. Haemolytic analysis of VP1. Haemolytic curve of VP1, deter- mined by the method of Harris & Phoenix [41]. The peptide was incubated with either human erythrocytes (r) or these erythrocytes in thepresenceof100m M NaCl (j). Percentage haemolysis was deter- mined (n ¼ 3) and plotted as a function of VP1 concentration. It can be seen that at a concentration of 2.4 m M , VP1 showed 100% lysis of erythrocytes (LD 50 ¼ 1.45 m M ), but, in the presence of 100 m M NaCl, this ability was reduced by 60%. Ó FEBS 2002 Lipid-interactive a-helical structure in m-calpain (Eur. J. Biochem. 269) 5417 Haemolytic assay of VP1 It can be seen from Fig. 4 that VP1 is strongly haemolytic with a sigmoidal relationship between VP1 concentration and percentage haemolysis. The peptide achieved 100% lysis of erthrocyte membranes at 2.4 m M (LD 50 ¼ 1.45 m M ), but in the presence of 100 m M NaCl, this ability of VP1 was reduced the order of 60% (LD 50 ¼ 1.85 m M ). FTIR conformational analysis of VP1 FTIR spectroscopy was used for conformational analysis of VP1, either in aqueous solution or in the presence of SUVs formed from Myr 2 PtdSer, Myr 2 PtdCho or Myr 2 PtdEtn (Fig. 5). In each case, the relative percentages of a-helical secondary structure (1650–1655 cm )1 )andb-sheet secon- dary structure (1625–1640 cm )1 ) were computed and are shown in Table 2. In aqueous solution, VP1 showed no evidence of a-helical structure and was primarily formed from b-sheet structures (> 90%) (data not shown). In the presence of Myr 2 PtdEtn and Myr 2 PtdCho, VP1 showed some evidence of a-helical structure ( 20%) but remained predominantly formed from b-sheet structures (48% and 61%, respectively) (Fig. 5A,B). In contrast, VP1 showed high levels of a-helical structure (65%) in the presence of Myr 2 PtdSer and reduced levels of b-sheet structures (32%) (Fig. 5C). FTIR analysis of phospholipid phase-transition properties Using FTIR spectroscopy, absorbance spectra representing the effects of VP1 on the phase-transition temperature and membrane fluidity of membranes formed from Myr 2 PtdEtn, Myr 2 PtdCho or Myr 2 PtdSer were derived as a function of temperature (Fig. 6). Control experiments recorded the lipid phase-transition temperature, T c ,ofMyr 2 PtdCho mem- branes as 27 °C, Myr 2 PtdEtn membranes as 55 °C, and Myr 2 PtdSer membranes as 37 °C (Figs 6A-6C). The pres- ence of VP1 had a minor effect on the T c and membrane fluidity of both Myr 2 PtdCho and Myr 2 PtdEtn membranes, with T c being recorded as 24 °Cand52°C, respectively, and accompanied in each case by a minor increase in membrane fluidity (Fig. 6A,B). In contrast, the presence of VP1 led to a large decrease in the fluidity of Myr 2 PtdSer membrane, accompanied by a large increase in the T c of the membranes, with T c being recorded as 47 °C (Fig. 6C). Monolayer studies on VP1 The interactions of VP1 with Myr 2 PtdSer monolayers were studied as described above. Myr 2 PtdSer was found to form Fig. 5. FTIR conformational analysis of VP1 in the presence of lipid. (A)–(C) Spectra representing FTIR conformational analyses of VP1 in the presence of lipid with annotated numbers indicating band peak absorbances. For each spectrum, the relative percentages of a-helical secondary structure and b-sheet secondary structure were computed, as described in Materials and Methods, and are shown in Table 2. (A) and (B) show that, in the presence of Myr 2 PtdEtn and Myr 2 PtdCho, the major secondary structural contribution to VP1 came from b-sheet structures (1627 cm )1 and 1628 cm )1 , respectively). In contrast, (C) shows that, in the presence of Myr 2 PtdSer, a-helical structure made the major contribution to VP1 secondary structure (1650 cm )1 ). Table 2. VP1 secondary-structural contributions. Relative levels of a-helical structure and b-sheet structure determined in VP1, a peptide homologue of the putative oblique-orientated a-helix-forming segment identified in m-calpain domain V (Table 1). The peptide was either in aqueous solution (–) or in the presence of lipid. Conformational ana- lysis of VP1 was performed using FTIR spectroscopy, and the resulting spectra (Fig. 5) were used to determine relative levels of secondary structure as described in Materials and methods. Lipid % a-helix % b-sheet ––92 Myr 2 PtdSer 65 32 Myr 2 PtdEtn 18 61 Myr 2 PtdCho 19 48 5418 K. Brandenburg et al.(Eur. J. Biochem. 269) Ó FEBS 2002 stable monolayers at a surface pressure of 30 mNÆm )1 , which was taken to represent that of naturally occurring membranes. At a final subphase concentration of 20 l M , VP1 showed maximal levels of Myr 2 PtdSer monolayer penentration, which led to a change in monolayer surface pressure of 5.5 mNÆm )1 (Fig. 7). This ability was reduced to negligible levels in the presence of 100 m M NaCl (data not shown). DISCUSSION There is evidence to suggest that the enzymatic activity of m-calpain is modulated by the membrane interaction of a segment, GTAMRILGGVI, located in domain V of the protein’s smaller subunit [24,25]. It has been predicted that this segment may interact with membrane via the formation of an oblique-orientated a-helix [26], a class of a-helices [29– 35] that penetrate membranes at a shallow angle because of a hydrophobicity gradient along the a-helical long axis [27,28]. Here we have used theoretical techniques to examine the structural characteristics of the putative domain V a-helix. In addition, the ability of the GTAMRILGGV segment to adopt a-helical structure and to interact with membranes has been investigated using VP1, a peptide homologue of this segment, in conjunction with haemolytic analysis, monolayer studies, and FTIR spectroscopy. Our results are discussed in relation to HA2, a viral fusion peptide known to penetrate membranes via oblique-orien- tated a-helix formation [29–35]. The sequences of the GTAMRILGGVI segment and HA2 were analysed, and data points representing their a-helices were found to lay proximal in the region of the hydrophobic moment plot diagram delineating candidate oblique-orientated a-helices (Fig. 1). This observation sug- gests that the GTAMRILGGVI segment may form such an a-helix and therefore may possess a hydrophobicity gradi- ent. Consistent with this suggestion, hydropathy plot analysis showed the GTAMRILGGVI segment to become progressively more hydrophobic in moving from the Fig. 6. FTIR phase-transition analysis of lipid in the presence of VP1. Spectra representing FTIR phase-transition analysis of lipid in the presence of VP1. In the absence of the peptide, the phase-transition temperature (T c )ofMyr 2 PtdEtn was recorded as 55 °C(j;A),of Myr 2 PtdCho as 27 °C(j; B) and of Myr 2 PtdSeras37°C(j;C).The presence of VP1 led to a minor increase in the fluidity of both Myr 2 PtdEtn and Myr 2 PtdCho membranes, which in each case was accompanied by a minor decrease in T c ,withT c recorded as 52 °Cfor Myr 2 PtdEtn membranes (h;A)and24°CforMyr 2 PtdCho mem- branes (h; B). In contrast, the presence of VP1 led to a large decrease in the fluidity of Myr 2 PtdSer membranes, accompanied by a large increase in the T c of the membranes (h;C)withT c being recorded as 47 °C. Fig. 7. Monolayer interactions of VP1. Time course for interactions of VP1 with Myr 2 PtdSer monolayers at a surface pressure of 30 mNÆm )1 , taken to represent that of naturally occurring membranes. VP1 (final concentration 20 l M ) was introduced into the monolayer subphase at time zero and after 100 s showed rapid penetration of Myr 2 PtdSer monolayers. Maximal levels of penentration were reached after 600 s, with a concomitant change in monolayer surface pressure of 5.5 mNÆm )1 . Ó FEBS 2002 Lipid-interactive a-helical structure in m-calpain (Eur. J. Biochem. 269) 5419 N-terminus to the C-terminus (Fig. 2), and graphical analysis showed the GTAMRILGGVI a-helix to possess a number of structural resemblances to the HA2 a-helix (Fig. 3). It can be seen that each a-helix shown in Fig. 3 possesses a glycine-rich polar face, which studies on HA2 and a number of other oblique-orientated a-helices have shown to be critical for maintaining their hydrophobicity gradients [27,28,46]. It can also be seen from Fig. 3 that each a-helix possesses a wide hydrophobic face rich in bulky amino-acid residues. In the case of the GTAMRILGGVI segment (Fig. 3), leucine and isoleucine can be seen to be preponderant in the C-terminal region of the a-helix. This localization of strongly hydrophobic amino-acid residues is structurally consistent with the higher levels of hydropho- bicity predicted for the C-terminal region of the segment (Fig. 2) and possession of a hydrophobicity gradient [47]. Furthermore, when these wide hydrophobic faces, rich in bulky residues, are combined with narrow polar faces, rich in glycine residues (Fig. 3), a-helices are given an effective wedge shape, which appears to assist HA2 and other oblique-orientated a-helix-forming peptides, to destabilize membranes in the promotion of their biological activity [46,48]. It is clear from our theoretical analyses that the GTAMRILGGVI segment has the potential to form an a-helix with strong structural similarities to the oblique- orientated a-helix formed by HA2 and other membrane- interactive peptides. Consistent with these observations, FTIR spectroscopic analysis showed that VP1 was able to adopt a-helical structure in the presence of lipid membranes (Fig. 5) and to affect the lipid-phase transition properties of these membranes (Fig. 6). In addition, haemolytic analysis showed that the peptide is able to lyse erythrocyte membranes (Fig. 4), and monolayer studies showed that it is able to penetrate lipid monolayers mimetic of naturally occurring membranes (Fig. 7). In combination, these results clearly show that the GTAMRILGGVI segment is able to form a membrane-interactive a-helix. In aqueous solution, VP1 showed no evidence of a-helical structure (data not shown) but was found to adopt such structure in the presence of Myr 2 PtdCho, Myr 2 PtdEtn and Myr 2 PtdSer (Fig. 6). This suggests that VP1 requires the amphiphilic environment of a lipid interface to adopt a-helical structure, a requirement also observed for HA2 [49]. In the presence of Myr 2 PtdEtn and Myr 2 PtdCho membranes, VP1 adopted relatively low levels of a-helical structure ( 20%; Fig. 5A,B) and induced only minor decreases in the T c of these membranes ( 3 °C), accom- panied by minor increases in liquid-crystalline phase fluidity (Fig. 6,AB). Such changes in lipid-phase transition proper- ties are consistent with VP1 binding to the headgroup regions of Myr 2 PtdCho and Myr 2 PtdEtn membranes and suggest that the peptide penetrates the surface regions of these membranes. In contrast, VP1 adopted high levels of a-helical structure in the presence of Myr 2 PtdSer (65%; Fig. 5C) and induced a 10 °CriseintheT c of these membranes accompanied by a large decrease in its liquid- crystalline phase fluidity (Fig. 6C). Such changes are consistent with VP1 penetration of the Myr 2 PtdSer mem- brane hydrophobic core region and suggest that the peptide shows high levels of interaction with these membranes. Strongly supporting this suggestion, VP1 was found to show high levels of Myr 2 PtdSer monolayer penetration (Fig. 7), inducing surface pressure changes of 5.5 mNÆm )1 .Com- parable levels of monolayer penetration have been observed for the C-terminal a-helices of Escherichia coli penicillin- binding proteins 5 and 6 [50], which function as membrane anchors for these proteins [51]. Interestingly, a recent study has suggested that these anchors may form oblique- orientated a-helices (D. Phoenix & F. Harris, unpublished work). Taken in combination, these results suggest that VP1 has a low affinity for zwitterionic lipid but a high affinity for Myr 2 PtdSer and may have a requirement for either this specific lipid, or anionic lipid in general, to achieve higher levels of membrane penetration. These results contrast with HA2, which shows an affinity for zwitterionic lipid [52,53] and requires the low pH of the endosomal membrane surface for both oblique-orientated a-helix formation and enhanced levels of membrane penetration via the deproto- nation of its negatively charged residues [35,54]. VP1 was found to be strongly haemolytic, but, in the presence of 100 m M NaCl, this ability was reduced by the order of 60% (Fig. 4) and the ability of VP1 to interact with Myr 2 PtdSer monolayers was reduced to negligible levels (data not shown). In combination, these results suggest a strong electrostatic contribution to VP1–membrane inter- action and further support a VP1 requirement for anionic lipid. Indeed, the relatively high LD 50 (1.45 m M ) shown by VP1 for haemolytic action could reflect the fact that erythrocyte membranes possess an asymmetric distribution of anionic lipids with the extracytoplasmic leaflet depleted of such lipids [55]. It is interesting to note that the HA2 a-helix is also strongly haemolytic and that the mutation of polar face glycine residues, essential for maintaining the HA2 a-helix hydrophobicity gradient, results in the loss of haemolytic and fusogenic ability [49,56]. In conclusion, based on structural similarities to HA2, we have suggested that the segment, GTAMRILGGVI, of m-calpain domain V forms an a-helix, which possesses a hydrophobicity gradient and penetrates membranes in an oblique orientation. We speculate that glycine residues in the polar face of this a-helix could play an important role in facilitating this mechanism of membrane penetration. This a-helix has a preference for anionic lipid, which leads to higher levels of a-helicity and membrane penetration via electrostatic interactions. These results are consistent with those of previous authors, which have shown that m-calpain activity is modulated by the presence of anionic lipid [25]. To satisfy a VP1 requirement for anionic lipid, it seems probable that the peptide’s single positively charged amino- acid residue, arginine (Table 1; Fig. 3) would engage in charge–charge interaction with negatively charged mem- brane lipid. Furthermore, to achieve the deeper levels of membrane penetration indicated for Myr 2 PtdSer mem- branes, it is possible that these interactions may involve the snorkelling mechanism that features in the membrane interactions of a number of lipid-interactive a-helices [28,57]. According to this mechanism, the VP1 arginine residue would extend or snorkel its long hydrophobic alkyl chain, facilitating penetration of the membrane hydropho- bic core region yet still allowing its positively charged moiety to interact with negatively charged moieties in the mem- brane lipid headgroup region. We speculate that utilization of this mechanism by the m-calpain domain V a-helix could result in enhanced levels of membrane interaction by domain V, which would support work indicating lipid 5420 K. Brandenburg et al.(Eur. J. Biochem. 269) Ó FEBS 2002 involvement in the reduction of Ca 2+ levels necessary for the efficient activation of m-calpain. REFERENCES 1. Sorimachi, H. & Suzuki, K. (2001) The structure of calpain. J. Biol. Chem. 129, 653–664. 2.Sorimachi,H.,Ishiura,S.&Suzuki,K.(1997)Structureand physiological function of calpains. Biochem. J. 328, 721–732. 3. Johnson, G.V.W. & Guttmann, R.P. (1997) Calpains: intact and active? Bioessays 19, 1011–1018. 4. Ma, H., Fukiage, C., Kim, Y.H., Duncan, M.K., Reed, N.A., Shih, M., Azuma, M. & Shearer, T.R. (2001) Characterization and expression of calpain 10. A novel ubiquitous calpain with nuclear localization. J. Biol. Chem. 276, 28525–28531. 5. Sato, K. & Kawashima, S. (2001) Calpain function in the mod- ulation of signal transduction molecules. Biol. Chem. 382, 743–751. 6. Calafoli, E. & Molinari, M. (1998) Calpain: a protease in search of afunction?Biochem. Biophys. Res. Commun. 247, 193–203. 7. Ono, Y., Sorimachi, H. & Suzuki, K. (1998) Structure and phy- siology of calpain, an enigmatic protease. Biochem. Biophys. Res. Commun. 245, 289–294. 8. Azuma, M., Fukiage, C., David, L. & Shearer, T.R. (1997) Acti- vation of calpain in lens: a review and proposed mechanism. Lens. Exp. Eye Res. 64, 529–538. 9. Horikawa, H. et al. (2000) Genetic variation in the gene encoding calpain-10 is associated with type 2 diabetes mellitus. Nat. Genet. 26, 163–175. 10. Huang, Y.H. & Wang, K.K.W. (2001) The calpain family and human disease. Trends Mol. Med. 7, 355–362. 11. Wang, K.K.W. (2000) Calpain and caspase: Can you tell the dif- ference. Trends Neurosci. 23, 20–26. 12. Ohno,S.,Emori,Y.,Imajoh,S.,Kawasaki,H.,Kisaragi,M.& Suzuki, K. (1984) Evolutionary origin of a calcium-dependent protease by fusion of genes for a thiol protease and a calcium- binding protein? Nature (London) 312, 566–570. 13. Croall, D.E. & DeMartino, G.N. (1991) Calcium-activated neu- tral protease (calpain) system: structure, function, and regulation. Physiol. Rev. 71, 813–847. 14. Hosfield, C.M., Elce, J.S., Davies, P.L. & Jia, Z.C. (1999) Crystal structure of calpain reveals the structural basis for Ca 2+ dependent protease activity and a novel mode of enzyme activa- tion. EMBO J. 18, 6880–6889. 15. Strobl, S., Fernandez-Catalan, C., Braun, M., Huber, R., Masu- moto, H., Nakagawa, K., Irie, A., Sorimachi, H., Bourenkow, G., Bartunik,H.,Suzuki,K.&Bode,W.(2000)Thecrystalstructure of calcium-free human m-calpain suggests an electrostatic switch mechanism for activation by calcium. Proc. Natl. Acad. Sci. USA 97, 588–592. 16. Molinari, M. & Carafoli, E. (1997) Calpain: a cytosolic proteinase active at the membranes. J. Membr. Biol. 156, 1–8. 17. Kawasaki, H. & Kawashima, S. (1996) Regulation of the calpain- calpastatin system by membranes (review). Mol. Membr. Biol. 13, 217–224. 18. Chakrabarti, A.K., Dasgupta, S., Gadsden, R.H., Hogan, E.L. & Banik, N. (1996) Regulation of brain m-calpain Ca 2+ sensitivity by mixtures of membrane lipids: activation at intracellular Ca 2+ level. J. Neurosci. Res. 44, 374–380. 19. Saido, T.C., Shibata, M., Takenawa, T., Murofushi, H. & Suzuki, K. (1992) Positive regulation of mu-calpain action by polypho- sphoinositides. J. Biol. Chem. 267, 24585–24590. 20. Melloni, E. & Pontremoli, S. (1989) The calpains. Trends Neurosci. 12, 438–444. 21. Imajoh, S., Kawasaki, H. & Suzuki, K. (1986) The amino-terminal hydrophobic region of the small subunit of calcium-activated neutral protease (CANP) is essential for its activation by phos- phatidylinositol. J. Biochem. (Tokyo) 99, 1281–1284. 22. Tompa, P., Emori, Y., Sorimachi, H., Suzuki, K. & Friedrich, P. (2001) Domain III of calpain is a Ca 2+ -regulated phospholipid- binding domain. Biochem. Biophys. Res. Commun. 280, 1333– 1339. 23. Rizo, J. & Su ¨ dhof, T.C. (1998) C2-domains, structure and func- tion of a universal Ca 2+ -binding domain. J. Biol. Chem. 273, 15879–15882. 24. Crawford, C., Brown, N.R. & Willis, A.C. (1990) Investigation of the structural basis of interaction of calpain II with phospholipid and with carbohydrate. Biochem. J. 265, 575–579. 25. Arthur, J.S.C. & Crawford, C. (1996) Investigation of the inter- action of m-calpain with phospholipids: calpain–phospholipid interactions. Biochim. Biophys. Acta. 1293, 201–206. 26. Daman,O.A.,Biswas,S.,Harris,F.,Wallace,J.&Phoenix,D.A. (2001) Theoretical investigation into the lipid interaction of m-calpain. Mol. Cell. Biochem. 223, 159–163. 27. Decout, A., Labeur, C., Vanloo, B., Goethals, M., Vandekerck- hove, J., Brasseur, R. & Rosseneu, M. (1999) Contribution of the hydrophobicity gradient to the secondary structure and activity of fusogenic peptides. Mol. Membr. Biol. 16, 37–246. 28. Phoenix, D.A., Harris, F., Daman, O.A. & Wallace, J. (2002) The prediction of amphiphilic a-helices. Curr. Protein Peptide Sci. in press. 29. Bentz, J. & Mittal, A. (2000) Deployment of membrane fusion protein domains during fusion. Cell Biol. Int. 24, 819–838. 30. Luneberg, J., Martin, I., Nussler, F., Ruysschaert, J M. & Herr- mann, A. (1995) Structure and topology of the influenza virus fusion peptide in lipid bilayers. J. Biol. Chem. 270, 27606–27614. 31. Brasseur, R. (2000) Tilted peptides: a motif for membrane desta- bilisation (Hypothesis). Mol. Membr. Biol. 17, 31–40. 32. Harris, F., Wallace, J. & Phoenix, D.A. (2000) Use of the hydrophobic moment plot to aid the identification of oblique orientated a-helices. Mol. Membr. Biol. 17, 201–207. 33. Chang, D.K., Cheng, S.F., Trivedi, V.D., Yang, S.H. (2000) The amino-terminal region of the fusion peptide of influenza virus hemagglutinin, HA2, inserts into sodium dodecyl sulfate micelle with residues 16–18 at the aqueous boundary at acidic pH: oligomerization and the conformational flexibility. J. Biol. Chem. 275, 19150–19158. 34. Macosko, J.C., Kim, C.H. & Shin, Y.K. (1997) The membrane topology of the fusion peptide region of influenza hemagglutinin determined by spin-labeling EPR. J. Mol. Biol. 267, 1139–1148. 35. Han, X., Bushweller, J.H., Cafiso, D.S. & Tamm, L.K. (2001) Membrane structure and fusion-triggering conformational change of the fusion domain from influenza hemagluttinin. Nat. Struct. Biol. 8, 715–720. 36. Eisenberg, D., Weiss, R.M. & Terwilliger, T.C. (1982) The helical hydrophobic moment: a measure of the amphiphilicity of a-helix. Nature (London) 299, 371–374. 37. Eisenberg, D., Weiss, R.M., Terwilliger, T.C. & Wilcox, W. (1982) Hydrophobic moment and protein structure. Faraday Symp. Chem. Soc. 17, 109–120. 38. Eisenberg, D., Schwarz, E., Komaromy, M. & Wall, R. (1984) Analysis of membrane and surface protein sequences with the hydrophobic moment plot. J. Mol. Biol. 179, 125–142. 39. Hennig, L. (1999) Wingen/Winpeg: user friendly software for the analysis of amino acid sequences. Biotechniques 26, 1170–1172. 40. Kyte, J. & Dolittle, R.F. (1982) A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157, 105–132. 41. Harris, F. & Phoenix, D.A. (1997) An investigation into the ability of C-terminal homologues of the Escherichia coli low molecular mass penicillin-binding proteins 4, 5 and 6 to undergo membrane interaction. Biochemie 79, 171–174. 42. Keller, R.C., Killian, J.A. & De Kruijff, B. (1992) Anionic phos- pholipids are essential for alpha-helix formation of the signal peptide of prePhoE upon interaction with phospholipid vesicles. Biochemistry 31, 1672–1677. Ó FEBS 2002 Lipid-interactive a-helical structure in m-calpain (Eur. J. Biochem. 269) 5421 43. Kauppinen, J.K., Moffat, D.J., Mantsch, H.H. & Cameron, D.G. (1981) Fourier self-deconvolution: a method for resolving intrinsically overlapped bands. Appl. Spectrosc. 35, 271–276. 44. Brandenburg, K., Kusomoto, S. & Seydel, U. (1997) Conforma- tional studies of synthetic lipid A analogues and partial structures by infrared spectroscopy. Biochim. Biophys. Acta 1329, 183–201. 45. Demel, R.A. (1974) Model membrane monolayers: description of use and interaction. Methods Enzymol. 32, 539–545. 46. Fujii, G. (1999) To fuse or not to fuse: the effects of electrostatic interactions, hydrophobic forces and structural amphiphilicity on protein-mediated membrane destabilisation. Advanced Drug Delivery Review 38, 257–277. 47. Peuvot, J., Schank, A., Lins, L. & Brasseur, R. (1999) Are the fusion processes involved in birth, life and death of the cell depending on tilted insertion of peptides into membranes? J. Theor. Biol. 198, 173–181. 48. White, J.M. (1990) Viral and cellular membrane fusion proteins. Annu. Rev. Physiol. 52, 675–697. 49. Plank, C., Zauner, W. & Wagner, E. (1999) Application of membrane-active peptides for drug and gene delivery across cel- lular membranes. Advanced Drug Delivery Review 34, 21–35. 50. Harris, F., Demel, R.A., Phoenix, D.A. & De Kruijff, B. (1998) An investigation into the lipid interactions of peptides corresponding to the C-terminal anchoring domains of Escherichia coli penicillin- binding proteins 4, 5 and 6. Biochim. Biophys. Acta 1415, 10–22. 51. Phoenix, D.A. & Harris, F. (1998) Amphiphilic a-helices and lipid interactions. In Protein Targeting and Translocation (Phoenix, D.A., ed.), pp. 19–36. Portland Press, London. 52. Zhou, Z., Macosko, J.C., Hughes, D.W., Sayer, B.G., Hawes, J. & Epand, R.M. (2000) N-15 NMR study of the ionization properties of the influenza virus fusion peptide in zwitterionic phospholipid dispersions. Biophys. J. 78, 2418–2425. 53. Duzgunes, N. & Shavnin, S.A. (1992) Membrane destabilization by N-terminal peptides of viral envelope proteins. J. Membr. Biol. 128, 71–80. 54. Stegmann, T. & Helenius, A. (1993) Influenza virus fusion: from models toward a mechanism. In Viral Fusion Mechanisms (Bentz, J., ed.). CRC Press, Boca Raton, FL. 55. Op den Kamp, J.A.F. (1979) Lipid asymmetry in membranes. Annu. Rev. Biochem. 48, 47–71. 56. Tamm, L.K. & Han, X. (2000) Viral fusion peptides: a tool to disrupt and connect biological membranes. Biosci. Rep. 29,501– 518. 57. Segrest, J.P., Venkatachalapathi, Y.V., Srinivas, S.K., Gupta, K.B., De Loof, H. & Anatharamaiah, G.M. (1992) Role of basic amino acid residues in the amphipathic helix: the snorkel hypothesis. In Molecular Conformation and Biological Interactions (Balaram, P. & Ramaseshan, S., eds), pp. 597–635. Indian Academy of Sciences, Bangalore. 5422 K. Brandenburg et al.(Eur. J. Biochem. 269) Ó FEBS 2002 . Domain V of m-calpain shows the potential to form an oblique-orientated a-helix, which may modulate the enzyme’s activity via interactions with anionic. previous authors, which have shown that m-calpain activity is modulated by the presence of anionic lipid [25]. To satisfy a VP1 requirement for anionic lipid,