Author manuscript, published in "The Journal of Biological Chemistry 2005;280(13):12833-9" DOI: 10.1074/jbc.M412521200 TRANSDUCTION OF THE SCORPION TOXIN MAUROCALCINE INTO CELLS – EVIDENCE THAT THE TOXIN CROSSES THE PLASMA MEMBRANE

Maurocalcine (MCa) is a 33-amino-acid residue peptide toxin isolated from the scorpion Scorpio maurus palmatus. External application of MCa to cultured myotubes is known to produce Ca2+ release from intracellular stores. MCa binds directly to the skeletal muscle isoform of the ryanodine receptor, an intracellular channel target of the endoplasmic reticulum, and induces long lasting channel openings in a mode of smaller conductance. Here we investigated the way MCa proceeds to cross biological membranes to reach its target. A biotinylated derivative of MCa was produced (MCab) and complexed with a fluorescent indicator (streptavidine-cyanine 3) to follow the cell penetration of the toxin. The toxin complex efficiently penetrated into various cell types without requiring metabolic energy (low temperature) or implicating an endocytosis mechanism. MCa appeared to share the same features as the so-called cell-penetrating peptides. Our results provide evidence that MCa has the ability to act as a molecular carrier and to cross cell membranes in a rapid manner (1–2 min), making this toxin the first demonstrated example of a scorpion toxin that translocates into cells.

Maurocalcine (MCa) 1 is a 33-mer toxin isolated from the venom of the scorpion Scorpio maurus palmatus. Since its initial isolation (1), MCa has been successfully produced by chemical synthesis, which allows the characterization of its biological activity and the solution of its three-dimensional structure (2). MCa has been shown to be one of the most potent effectors of the skeletal muscle ryanodine receptor type 1 (RyR1), an intracellular calcium channel target. The toxin stimulates the binding of [ 3 H]ryanodine onto RyR1 present in sarcoplasmic reticulum (SR) vesicles (3,4). It induces strong modifications in the gating of RyR1 channels reconstituted in lipid bilayers that are characterized by the appearance of long lasting subconductance states (4,5). MCa application to purified SR vesicles produces Ca 2ϩ release, which is consistent with these biophysical effects (3,4). Using a calcium-imaging approach, it was also shown that the addition of MCa to the extracellular medium of cultured myotubes induces Ca 2ϩ release from the SR into the cytoplasm (4). Taken together, these data are coherent with a direct effect of MCa on RyR1. These observations were confirmed with the identification of the binding site of MCa on RyR1, which indicates a cytoplasmic localization (6).
Two pieces of evidence suggest that MCa should be able to cross the plasma membrane to reach its intracellular target. First, MCa has biological activity consistent with the direct activation of RyR1 when added to the extracellular medium. Second, a structural analysis of MCa reveals a stretch of positively charged amino acid residues that is reminiscent of the protein transduction domains (PTD) found in proteins, such as Tat from the HIV-1 virus (7,8), the Drosophila homeotic transcription factor ANTP (encoded by the antennapedia gene) (9), and the herpes simplex virus type 1 (HSV-1) VP22 transcription vector (10). Synthetic peptides corresponding to these domains are called cell-penetrating peptides (CPPs) and have in common the inherent trait of containing many basic residues (arginine and lysine), often oriented toward the same face of the molecule (Fig. 1B). Because of this structural feature, CPPs possess the interesting ability to cross biological membranes in a receptor-or transporter-independent manner (11)(12)(13). These peptides seem to target the lipid bilayer directly, using complementary charges, and penetrate the cell through a mechanism called translocation, which remains unclear. Three possible models have been put forth in the literature to explain peptide penetration: 1) the inverted mycelle model (14), with its membrane-destabilization model variant (11), 2) the carpet model (15), and 3) the pore formation model (16). Regardless of the precise mode of entry, CPPs possess unique features worth mentioning: (i) delivery in 100% of the cells (17), (ii) delivery in all cell types both in vitro and in vivo (18,19), (iii) the ability to cross the blood-brain barrier (18), and (iv) the potential to act as carriers and to translocate large compounds, such as proteins and oligonucleotides (20).
The current study investigated the ability of MCa to penetrate into cells. We showed that a biotinylated version of MCa effectively and rapidly enters various cell types and behaves as a good carrier to allow the penetration of larger compounds. As for other CPPs, the mechanism of cell penetration is not linked to an endocytotic pathway. We concluded that MCa represents a new class of disulfide-linked CPP and is the first scorpion toxin shown to enter cells.

Chemical Synthesis of MCa and Biotinylated
Mca-N-␣-Fmoc-L-amino acids, 4-hydroxymethylphenyloxy resin, and reagents used for peptide synthesis were obtained from PerkinElmer Life Sciences. N-␣-Fmoc-L-Lys(Biotin)-OH was purchased from Neosystem Groupe Societe Nationale des Poudres et Explosifs. The MCa and biotinylated MCa (MCa b ) were obtained by solid phase peptide synthesis (21), using an automated peptide synthesizer (Model 433A, Applied Biosystems Inc.). Peptide chains were assembled stepwise on 0.25 meq of hydroxymethylphenyloxy resin (1% cross-linked; 0.89 meq of amino group/g) using 1 mmol of N-␣-Fmoc amino acid derivatives. The side chain-protecting groups were as follows: trityl for Cys and Asn; tert-butyl for Ser, Thr, Glu, and Asp; pentamethylchroman for Arg; and tert-butyloxycarbonyl or Biotin for Lys. N-␣-Amino groups were deprotected by treatment with 18 and 20% (v/v) piperidine/N-methylpyrrolidone for 3 and 8 min, respectively. The Fmoc-amino acid derivatives were coupled (20 min) the same as their hydroxybenzotriazole-active esters in N-methylpyrrolidone (4-fold excess). To obtain MCa, an aliquot of peptide resin was removed at its corresponding cycle. After peptide chain assembly, the peptide resin (ϳ0.9 g) was treated between 2 and 3 h at room temperature in constant shaking with a mixture of trifluoroacetic acid:H 2 O/ thioanisol/ethanedithiol (88:5/5/2, v/v) in the presence of crystalline phenol (2.25 g). The peptide mixture was then filtered, and the filtrate was precipitated by adding cold tert-butylmethyl ether. The crude peptide was pelleted by centrifugation (3,000 ϫ g for 10 min), and the supernatant was discarded. The reduced peptide was then dissolved in 200 mM Tris-HCl buffer, pH 8.3, at a final concentration of 2.5 mM and stirred under air to allow oxidation/folding (between 50 and 72 h, room temperature). The target products, MCa and MCa b , were purified to homogeneity, first by reversed-phase high pressure liquid chromatography (PerkinElmer Life Sciences, C18 Aquapore ODS, 20 m, 250 ϫ 10 mm) by means of a 60-min linear gradient of 0.08% (v/v) trifluoroacetic acid/0 -30% acetonitrile in 0.1% (v/v) trifluoroacetic acid/ H 2 O at a flow rate of 6 ml/min ( ϭ 230 nm). A second step of purification of MCa and MCa b was achieved by ion exchange chromatography on a carboxymethyl cellulose matrix using 10 mM (buffer A) and 500 mM (buffer B) sodium phosphate buffers, pH 9.0 (60-min linear gradient from 0 to 60% buffer B at a flow rate of 1 ml/min). The homogeneity and identity of MCa or MCa b were assessed by the following: (i) analytical C18 reversed-phase high pressure liquid chromatography (Merck, C18 Li-Chrospher, 5 m, 4 ϫ 200 mm) using a 60-min linear gradient of 0.08% (v/v) trifluoroacetic acid/0 -60% acetonitrile in 0.1% (v/v) trifluoroacetic acid/H 2 O at a flow rate of 1 ml/min; (ii) amino acid analysis after acidolysis (6 N HCl/2% (w/v) phenol, for 20 h at 118°C in N 2 atmosphere); and (iii) mass determination by matrix-assisted laser desorption ionization time-of-flight mass spectrometry.
Formation of the MCa b ⅐Streptavidin-Cyanine 3 Complex-Soluble streptavidin-cyanine 3 (Strept-Cy3, Amersham Biosciences) was mixed with four molar equivalents of MCa b (1 mM) for 2 h at room temperature in phosphate-buffered saline.
Preparation of Heavy SR Vesicles-Heavy SR vesicles were prepared following the method of Kim et al. (22) modified as described previously (23). Protein concentration was measured by the Biuret method.
Cell Cultures-L6 cells (rat myogenic L6 cells) were obtained from the European Collection of Animal Cell Cultures ECACC (clone C5) and cultured in Dulbecco's modified Eagle's medium supplemented with 15% fetal bovine serum (Invitrogen) and 1% penicillin/streptomycin (Invitrogen). Differentiation of L6 cells was induced by a shift to differentiation medium (Dulbecco's modified Eagle's medium ϩ 5% horse serum) when they reached confluence, as described previously (24).
CA1 Hippocampal Neurons-Hippocampi from neonatal mice (between days 1 and 2 postpartum) were dissected, cleaned of meninges, and placed in Hanks' balanced salt solution (Invitrogen). They were transferred into a dissociation medium containing Hanks' balanced salt solution, 1% penicillin/streptomycin (Invitrogen), 2,000 units/ml DNase I and 1% (w/v) trypsin/EDTA and incubated at 37°C for 7 min. After sedimentation, the supernatant was removed, and the tissue was washed with a Hanks' balanced salt solution containing 1% penicillin/ streptomycin. The tissue was gently triturated in Hanks' balanced salt solution containing 2,000 units/ml DNase I, 10% fetal bovine serum, and 1% penicillin/streptomycin with a plastic pipette until a homogeneous suspension was obtained. After centrifugation at 100 ϫ g for 1 min, the cell pellet was resuspended in Neurobasal/B27 medium (Invitrogen) containing 0.5 mM L-glutamine and 1% penicillin/streptomycin. The cell cultures were seeded at a density of 10 5 cells/cm 2 on culture dishes coated previously with 20 g/ml poly(L-lysine) at 37°C for 2 h. After 2 days, 3 M cytosine arabinoside was added to the cultures to control the proliferation of non-neuronal cells, and 24 h later, half of the medium was replaced. The culture medium was then subsequently changed every other day.
HEK293 Cells-Human embryonic kidney cells (HEK293, Invitrogen) were passaged by 1% trypsin treatment prior to confluence and maintained in Dulbecco's modified Eagle's medium containing 10% (v/v) fetal bovine serum and 1% penicillin/streptomycin (Invitrogen) and were placed into an incubator and 5% CO 2 . The culture medium was changed every other day. Ca 2ϩ Release Measurements-Ca 2ϩ release from heavy SR vesicles was measured using the Ca 2ϩ -sensitive dye antipyrylazo III. The absorbance was monitored at 710 nm by a diode array spectrophotometer (MOS-200 optical system, Biologic, Claix, France). Heavy SR vesicles (50 g) were actively loaded with Ca 2ϩ at 37°C in 2 ml of a buffer containing 100 mM KCl, 7.5 mM sodium pyrophosphate, 20 mM MOPS, pH 7.0, supplemented with 250 M antipyrylazo III, 1 mM ATP/MgCl 2 , 5 mM phosphocreatine, and 12 g/ml creatine phosphokinase (25). Ca 2ϩ loading was started by sequential additions of 50 and 20 M CaCl 2 . In these loading conditions, no calcium-induced calcium release interfered with the observations. At the end of each experiment, Ca 2ϩ remaining in the vesicles was determined by the addition of 4 M Ca 2ϩ ionophore A23187 (Sigma), and the absorbance signal was calibrated by two consecutive additions of 20 M CaCl 2 .

Imaging of MCa b ⅐Strept-Cy3 Translocation by Confocal Microscopy
Imaging on Fixed Cells-Cell cultures were incubated with 100 nM (final concentration) of the MCa b ⅐Strept-Cy3 complex in the cell culture medium in the dark and at room temperature, unless otherwise stated, for 30 min or 1 h. After three washes with phosphate-buffered saline, the cells were fixed at room temperature and in the dark by 3% paraformaldehyde for 20 min, washed with phosphate-buffered saline, and mounted with fluorescence antifading medium (Dako) for observation with the confocal microscope. All conditions were compared under identical settings. The samples were analyzed by confocal laser scanning microscopy using a Leica TCS-SP2 operating system. Cy3 fluorescence was excited using the 543-nm line of a helium-neon laser, and the fluorescence emission was collected from 554 to 625 nm.
Imaging on Living Cells-Living cells were incubated at room temperature in fresh culture medium on the stage of an upright compound microscope (Eclipse 600 FN, Nikon) equipped with a water immersion 40ϫ objective (numerical aperture 0.8) and a confocal head (PCM 2000, Nikon) or with the Leica TCS-SP2 microscope with a 100ϫ objective under "XYZt" mode. MCa b ⅐Strept-Cy3 penetration kinetics were followed immediately after injection of this complex at 100 nM into the culture medium. Cy3 fluorescence was excited with the 488-nm wavelength line of an argon laser. Emitted light was filtered by a 595 Ϯ 35nm filter. Images were acquired and analyzed with the EZ2000 software (Nikon). Relative fluorescence quantitative analysis was realized on the stack using the devoted Leica software.

Synthesis of Biotinylated MCa and Complex Formation with
Cy3-labeled Streptavidine-We have previously identified MCa as a potent activator of the skeletal muscle RyR1 calcium channel (3,4). As RyR1 is an intracellular target, these observations suggested that MCa may modulate RyR1 by one of two mechanisms: (i) indirectly through binding onto a plasma membrane receptor (implies no cell penetration) or (ii) directly through binding onto RyR1 itself (implies cell penetration). With the recent identification of a RyR1 binding site for MCa (6) and earlier evidence that MCa alters the gating of purified RyR1 channels reconstituted in lipid bilayers, we favor the hypothesis that MCa can reach its intracellular target directly by crossing the plasma membrane. To investigate this question, we synthesized a biotinylated version of MCa (MCa b ) to create a fluorescent complex MCa b coupled to Cy3-labeled streptavidine (MCa b ⅐Strept-Cy3). We first tested whether biotinylation and complex formation with Strept-Cy3 impairs the functional effects of MCa on RyR1. Fig. 2 demonstrates that the MCa b ⅐Strept-Cy3 complex retains the ability to stimulate [ 3 H]ryanodine binding on SR vesicles (panel A) as well as to induce Ca 2ϩ release from SR vesicles (panel B). The slower kinetics of Ca 2ϩ release induced by the MCa b ⅐Strept-Cy3 complex, as compared with MCa alone, is indicative of a slightly reduced efficiency (Fig. 2B). This difference probably stems from a reduced accessibility of MCa b ⅐Strept-Cy3 to the active MCa binding site on RyR1 because of the bulkiness of Strept-Cy3 (molecular mass of streptavidine of 60,000 Da compared with 4,108 Da for MCa b ). However, these data indicate that the formation of a complex with Strept-Cy3 does not alter MCa structure and regulatory function, making the MCa b ⅐Strept-Cy3 complex a valid fluorescent indicator in detecting the ability of MCa to penetrate into cells.
Cell Penetration of MCa b ⅐Strept-Cy3 Complex-Using the MCa b ⅐Strept-Cy3 complex, we investigated the ability of MCa to transduce various cell types (Fig. 3). A primary cell culture of hippocampal CA1 neurons and two cell lines, HEK293 and L6 cells (before and after differentiation), were incubated for 30 min (room temperature) with 100 nM MCa b complexed to Strept-Cy3. The cells were then fixed and the fluorescence observed by confocal microscopy. The data demonstrated that all cells were labeled, with a strong and uniform staining at the plasma membrane and the cytoplasm, whereas the nucleus is weakly labeled (Fig. 3). In contrast, cells incubated in the presence of 100 nM non-biotinylated MCa and equivalent concentrations of Strept-Cy3 do not display any labeling, demonstrating that Strept-Cy3 alone is not able to transduce into cells. The association of Strept-Cy3 with MCa b is thus required for the fluorescent complex to enter cells, demonstrating the active function of MCa in this process. These data thus indicate that MCa not only acts as a CPP, but similar to many other CPPs, has the ability to cargo large molecular mass proteins (streptavidine is 14.6-fold larger than MCa).

Kinetics of Cell Penetration by the MCa b ⅐Strept-Cy3
Complex-To further characterize the transduction properties of MCa, we analyzed the time course of the fluorescence labeling of living non-differentiated L6 cells by MCa b ⅐Strept-Cy3 com-plex. To achieve this study, we used time-lapse confocal microscopy in a way that preserved cell integrity and viability over at least a 1-h period. Cell labeling was evident for periods of time as short as 3 min after the addition of 100 nM MCa b ⅐Strept-Cy3 complex in the extracellular medium (Fig. 4A). We defined three regions of interest (ROI) corresponding to the plasma membrane (ROI-1), the cytoplasm (ROI-2), and the nucleus (ROI-3). The positioning of the different ROIs was facilitated by the analysis of the transmitted light image of the cell under observation (Fig. 4B). We then analyzed the evolution of the fluorescence intensity as a function of time in the different ROIs with a Leica confocal microscope under the XYZt mode (Fig. 4C). At the start of application of the MCa b ⅐Strept-Cy3 complex at 100 nM in the bath, the fluorescence intensity increased in all cell compartments, although at different rates. The evolution was fastest and greatest for the plasma membrane (ROI-1) and reached a peak in 10 min. A slower, but nevertheless important, increase in fluorescence was also observed for the cytoplasmic compartment (ROI-2), whereas a much smaller, but above background, fluorescence increase was observed for the nucleus (ROI-3). The rate and the relative intensity at which fluorescence increased in each compartment was consistent with the direction of progression of the MCa b ⅐Strept-Cy3 complex within the cell, i.e. from the extracellular space to the plasma membrane, from the plasma membrane to the cytoplasm, and then from the cytoplasm to the nucleus. Of note, the fluorescence signal recorded in ROI-1 corresponded to the fluorescence associated to the plasma membrane plus the cytoplasmic area close to the membrane. Because the signal was normalized, it underestimated the plasma membrane-associated fluorescence. In contrast, the relative fluorescence intensity of the cytoplasm should be more accurate. Two conclusions should be drawn from these data. First, the passage of the MCa b ⅐Strept-Cy3 complex from the cytoplasm to the nucleus is very weak, suggesting that this transition is far less favored than the two other ones. Second, the fluorescence associated to cell compartments is far greater than the fluorescence from the bath, demonstrating that cells act as MCa "concentrators." We also followed the evolution of the fluorescence once the MCa b ⅐Strept-Cy3 complex was washed out of the bath, i.e. after 12 min of MCa b ⅐Strept-Cy3 application and 5 min of wash (Fig. 4C). We observed that the fluorescence intensity of ROI-1 decreased, whereas that of ROI-2 increased. ROI-3 fluorescence intensity remained constant. The fluorescence increase of the cytoplasmic compartment must thus be related to a continued transfer of the complex from the plasma membrane into the cytoplasm. This plasma membrane/cytoplasm transfer is probably the rate-limiting step in MCa entry. The decrease in fluorescence intensity of the plasma membrane must thus result both from a leakage of the fluorescent complex to the extracellular medium and from its transfer to the cytoplasm. Longer periods of observations were precluded by the appearance of an inhomogeneous distribution of the fluorescent dye in the cytoplasm, maybe linked to degradation pathways (data not shown). However, we conclude from these data that complexes of MCa b ⅐Strept-Cy3 that penetrate into cells appear to stably remain inside the cell, suggesting that MCa has a favored direction of plasma membrane crossing.
Endocytosis Is Not Required for Translocation of the MCa b ⅐Strept-Cy3 Complex-Several cell mechanisms are known that may allow the entry of MCa. Transduction mechanisms are postulated to be energy-independent and do not require a specific receptor for translocation across the plasma membrane, whereas endocytosis and pinocytosis are energy-dependent. Although a contribution of an endocytotic process is possible, two reasons suggest that it is not the primary mode of entry of MCa. First, the distribution of the MCa b ⅐Strept-Cy3 complex in the cytoplasm is quite uniform. Second, the normal intracellular target of MCa (RyR1) requires a direct release of MCa into the cytoplasm, which is hardly compatible with endocytosis. We, nevertheless, investigated the contribution of energy-dependent processes in the entry of the MCa b ⅐Strept-Cy3 complex (Fig. 5). We first tested the effect of decreasing the temperature on the entry of MCa b ⅐Strept-Cy3 complex (Fig.   FIG. 2. The  5B). The data unambiguously demonstrated that the entry of the complex still occurs at 4°C. At this temperature, the complex similarly labels the plasma membrane and the cytoplasm.
Also, we tested the effect of specific inhibitors of pinocytosis and endocytosis. Neither 3 mM amiloride (Fig. 5C) nor 50 M nystatin (Fig. 5D) had any effect on the cell entry of the com- plex or its relative distribution in the plasma membrane and in the cytoplasm, again confirming that the main pathway for entry of MCa into cells does not involve either endocytosis or pinocytosis.

DISCUSSION
In this study, we demonstrated that MCa has the ability to translocate into cells by crossing the plasma membrane. This process is rapid, because translocation appears evident even for incubation times as short as 1 or 2 min. It reaches saturation in the plasma membrane around 10 min, although longer times were required for entry into the cytoplasm. This is consistent with the kinetics of entry of other CPPs with a peak of entry of 15 min for the Tat⅐p27 Kip1 (26) and Tat⅐␤-galactosidase complexes (18).
In our experiments, we coupled MCa b to streptavidine, which is of significantly higher mass than MCa itself, demonstrating that MCa can also carry large molecules into cells, similar to other CPPs (27,28). The fact that MCa b is so efficient with such a large cargo suggests that uncoupled MCa should be even more efficient when added to the extracellular medium of cells. Indeed, it was observed that Ca 2ϩ release from myotubes alone occurred as fast as 3 s after the addition of MCa, suggesting a much faster translocation in the absence of a cargo molecule. The demonstration that uncoupled MCa has an even faster kinetics of cell entry awaits the synthesis of a fluorescent derivative of MCa. Attempts to produce such a derivative have been unsuccessful so far. In our experiments, we used a concentration of MCa b (100 nM) that is in the range for which other CPPs demonstrate efficient translocation (usually between 25 to 200 nM (17). We show that translocation of MCa occurs at 4°C, as well as in the presence of inhibitors of pinocytosis and endocytosis, demonstrating that the process is energy-independent, similar to other CPPs (17,29,30). Although the current mechanism of MCa translocation is unknown, we can only propose that it may occur along one of the mechanisms suggested for other CPPs. As with other CPPs, we found that maurocalcine has the ability to effectively translocate in differ-ent cell types, suggesting that the process is insensitive to the nature of the biological membrane or to the types of receptors expressed on its surface. It is thought that CPPs interact with the plasma membrane before translocation through their positively charged surface, presumably with negative charges present at the extracellular surface of the lipid bilayer (31). Because MCa also possesses such a positively charged surface, we can assume that it interacts with the lipid bilayer in a manner akin to other CPPs. Nevertheless, it should be mentioned that endocytosis may represent an additional contributing factor to MCa cell entry for longer periods of incubation, although we did not investigate this question. Referencing the literature, we noticed that all CPPs are small peptides. In Tat, the PTD encompasses the basic amino acid residues 47-57 (17,32,33), whereas in VP22 and ANTP, the PTD goes from residues 267 to 300 (29) and 43 to 58 (14), respectively. Similarly, the putative PTD sequence of maurocalcine should be shorter than 33 amino acid residues. The sequence of MCa that triggers translocation is unknown. However, because of its similarities with other CPPs, the basic amino acid-rich sequence is the principal suspect. One intriguing observation is that part of the same basic sequence involved in translocation also appears to mediate the interaction with RyR1. For instance, mutation of Arg 24 of MCa, which belongs to the putative CPP sequence, results in a complete loss of interaction with RyR1 (4). Interestingly, part of the PTD of HIV-1 Tat, the Gly-Arg-Lys-Lys-Arg-Arg sequence, is also a potential nuclear localization sequence (12). In addition, Tat, VP22, and ANTP are all proteins implicated in transcriptional regulation, and the PTDs are the domains that make contact with nucleic acids. This seems to indicate that PTD may intrinsically possess multiple functions, such as protein translocation (all CPPs), protein targeting (Tat), and protein (MCa) or nucleic acid (Tat, VP22 and ANTP) interaction.
Overall, we thus have several arguments to believe that MCa behaves as a CPP. Similar to recognized CPP sequences, the following appears to be evident: 1) MCa is a small peptide, 2) it is positively charged, 3) it enters many cell types, 4) it enters in an efficient manner and at low concentration, 5) the translocation is a fast process that is energy-independent, and 6) it can carry a cargo molecule. It possesses yet two other essential properties that may not be shared with other CPPs, which are: 1) its apparent capacity to enter cells against its concentration gradient and 2) it enters the cell far more rapidly than its exit. The reasons for these properties are unknown, but we tentatively assume that the membrane potential may work as a driving force for cell penetration and maintenance. Because MCa is structured by disulfide bridges, it should be classified as belonging to a new class of CPP. Also, the disulfide linkage of MCa, which makes it more rigid than other CPPs, implies that the transduction mechanism at the basis of MCa cell penetration does not rely on extensive peptide unfolding. Because of the growing importance of CPPs in delivering functionally important molecules into cells, the identification of MCa will be useful to further understand the molecular basis of cell penetration of CPPs. In addition, MCa may constitute a leading compound for the design of new and efficient analogues to deliver active drugs inside cells that would otherwise have limited or no bioavailability. A new CPP may be of interest for biotechnological and pharmaceutical companies.