Direct Involvement of Protein Myristoylation in Myristoylated Alanine-rich C Kinase Substrate (MARCKS)-Calmodulin Interaction*

MARCKS, a major in vivo substrate of protein kinase C, interacts with plasma membranes in a phosphoryla-tion-, myristoylation-, and calmodulin-dependent manner. Although we have previously observed that myristoylated and non-myristoylated MARCKS proteins behave differently during calmodulin-agarose chromatography, the role of protein myristoylation in the MARCKS-calmodulin interaction remained to be eluci-dated. Here we demonstrate that the myristoyl moiety together with the N-terminal protein domain is directly involved in the MARCKS-calmodulin interaction. Both myristoylated and non-myristoylated recombinant MARCKS bound to calmodulin-agarose at low ionic strengths, but only the former retained the affinity at high ionic strengths. A quantitative analysis obtained with dansyl (5-dimethylaminonaphthalene-1-sulfonyl)-calmodulin showed that myristoylated MARCKS has an affinity higher than the non-myristoylated protein. Fur-thermore, a synthetic peptide based on the N-terminal sequence was found to bind calmodulin only when it was myristoylated. Only the N-terminal peptide but not the canonical calmodulin-binding domain showed the ionic strength-independent calmodulin binding. A mu-tation study suggested that the importance of the posi-tive charge in the N-terminal protein domain in the binding. Since discovery a covalent-bound myristoyl group chemistry, purchased from Research Genetics. The peptides were purified by reversed-phase high-perfor-mance liquid chromatography (HPLC) using a C18 column (Waters, (cid:2) Bondasphere 5 (cid:2) C18–300Å, 1.9 (cid:2) 15 cm). They were judged to be of greater than 95% purity by analytical HPLC and electrospray mass spectrometry (14, 25). Peptide concentration was determined by quantitative amino acid analysis. The Escherichia coli strain BL21(DE3)pLysS was obtained from Stratagene. The plasmid pBB131NMT a gift from Dr. J. Expression and Purification of Non-myr MARCKS and Myr MARCKS— The E. coli strain BL21(DE3)pLysS was transformed with the plasmid pET3d containing the human MARCKS gene cloned by PCR based on the published sequence (26). For the construction, ex- pression and purification of non-myr MARCKS, the published procedure was followed exactly as described previously (27). The cells con- taining the plasmid were selected with 100 (cid:2) g/ml ampicillin. A frozen stock of transformed colonies was used to inoculate LB media contain- ing 100 (cid:2) g/ml ampicillin, and the cells were grown overnight 37 °C. 500 ml of LB medium containing 100 (cid:2) g/ml ampicillin was inoculated with the overnight culture. The cells were grown to an OD 600 of (cid:3) 1–1.2, then further grown in the presence of 0.4 m M isopropyl-1-thio- (cid:1) -galactopyl- anoside for 5 h. The cells were collected by centrifugation for 20 min at 8000 rpm at 4 °C (Hitachi 9-2 rotor) and kept frozen at (cid:4) 20 °C. Non-myr MARCKS was purified by a procedure as described previously (27). For myr-MARCKS, the E. coli strain BL21(DE3)pLysS

Since the discovery of a covalent-bound myristoyl group in the catalytic subunit of cAMP-dependent protein kinase (1), many proteins have been found to be modified not only with myristoyl group but also with a variety of fatty acids (2)(3)(4). Interestingly, the vast majority of acylated proteins are either proteins involved in cellular signaling or those of viral origin. Protein acylation is often essential for the proper functioning of these proteins (5), although in the mechanism by which the modification exerts the effect is largely unknown. Of various kinds of acylation, myristoylation has been implied in the re-versible membrane association due to its intermediate hydrophobicity (3,6). Studies from our own and other laboratories (7,8) have established that such a mechanism is, in fact, operative in the phosphorylation-dependent interaction of MARCKS 1 (myristoylated alanine-rich C kinase substrate) with membranes. In the case of recoverin, the binding of Ca 2ϩ induces a drastic conformational change of the protein, and the myristoyl group hidden inside the protein will protrude from the protein and can interact with membrane (9). The modification has also been shown to affect the protein stability of cAMP-dependent protein kinase (10). However, the involvement of the protein myristoylation in protein-protein interaction has never been clearly demonstrated, although the issue has been the subject of extensive studies (11)(12)(13). Recently, we have shown that the modification is directly involved in the interaction of a brainspecific protein kinase C substrate, CAP-23/NAP-22, with calmodulin (14,15). Furthermore, the phenomena is observed in the binding HIV-1 Nef with calmodulin (16). The acyl chain interacts with specifically with the hydrophobic pocket of calmodulin. Since the basic domain adjacent to the myristoyl group was found to be important for the CAP-23/NAP-22-calmodulin and HIV-1 Nef-calmodulin interaction, the interplay of the myristoyl group and the basic domain seems to function in a manner analogous to that found in the myristoylationmediated protein-membrane interaction.
MARCKS, a major in vivo substrate of protein kinase C as well as of proline-directed protein kinases such as mitogenactivated protein kinase (17), contains a highly conserved Nterminal domain with an N-terminal myristoyl group and a basic phosphorylation domain in the middle of the molecule that is at the same time the binding sites for calmodulin (18,19). The phosphorylation of the latter domain results in the disruption of the calmodulin binding of the domain (20,21). We have previously shown that MARCKS binds to phospholipid membrane through both the N-terminal myristoyl moiety and the phosphorylation domain of basic amphiphilic nature (7). Protein kinase C-dependent phosphorylation introduces negative charges into the positively charged effector domain, consequently neutralizing its interaction with acidic phospholipids and promoting its release from the membrane (7,8). We have also discovered a demyristoylase activity of MARCKS and shown that the modification is dynamically regulated (22,23). During the course of these studies, we found that the presence of non-myristoylated pool of MARCKS (24). Interestingly, the non-myristoylated MARCKS protein behaved differently from the myristoylated counterpart during purification procedures. While the latter bound tightly to calmodulin-agarose column in the presence of Ca 2ϩ , the former was eluted from the column with high ionic strength even in the presence of Ca 2ϩ (24).
In this study, we have produced recombinant non-myristoylated (non-myr) and myristoylated (myr) MARCKS proteins and studied their binding properties to calmodulin in detail. Non-myr MARCKS was dissociated from calmodulin with high ionic strength in the presence of Ca 2ϩ , whereas myr MARCKS bound to calmodulin even at high ionic strength. Furthermore, we found that a myristoylated peptide based on the N-terminal MARCKS sequence bound to calmodulin only when it was myristoylated. The results obtained suggested that the N-terminal myristoylated domain is directed involved in the MARCKS-calmodulin interaction.

EXPERIMENTAL PROCEDURES
Materials-Dansyl-calmodulin and calmodulin-agarose were obtained from Sigma. Source 15Q ion exchange column was from Pharmacia Corp. Tryptone and yeast extracts were from Difco, while ampicillin, kanamycin, and isopropyl-1-thio-␤-galactopylanoside were from Wako Pure Chemical Industries. A 25-amino acid peptide corresponding to the phosphorylation domain of MARCKS (KKKKKRFSFKKS-FKLSGFSFKKNKK) was synthesized with a standard t-Boc chemistry using an Applied Biosystems peptide synthesizer 430A. A myristoylated peptide based on the N-terminal sequence of MARCKS (myr-GAQFSK-TAAK) and its variants (myr-GAQASKTAAK, myr-GAQKSKTAAK, myr-GAQDSKTAAK, myr-GAQFSDTAAK), and a non-myristoylated peptide (GAQFSKTAAKGEATAER), synthesized using standard Fmoc (N-(9fluorenyl)methoxycarbonyl) chemistry, were purchased from Research Genetics. The peptides were purified by reversed-phase high-performance liquid chromatography (HPLC) using a C18 column (Waters, Bondasphere 5 C18 -300Å, 1.9 ϫ 15 cm). They were judged to be of greater than 95% purity by analytical HPLC and electrospray mass spectrometry (14,25). Peptide concentration was determined by quantitative amino acid analysis. The Escherichia coli strain BL21(DE3)pLysS was obtained from Stratagene. The plasmid pBB131NMT was a gift from Dr. J. Gordon (Washington University).
Expression and Purification of Non-myr MARCKS and Myr MARCKS-The E. coli strain BL21(DE3)pLysS was transformed with the plasmid pET3d containing the human MARCKS gene cloned by PCR based on the published sequence (26). For the construction, expression and purification of non-myr MARCKS, the published procedure was followed exactly as described previously (27). The cells containing the plasmid were selected with 100 g/ml ampicillin. A frozen stock of transformed colonies was used to inoculate LB media containing 100 g/ml ampicillin, and the cells were grown overnight 37°C. 500 ml of LB medium containing 100 g/ml ampicillin was inoculated with the overnight culture. The cells were grown to an OD 600 of ϳ1-1.2, then further grown in the presence of 0.4 mM isopropyl-1-thio-␤-galactopylanoside for 5 h. The cells were collected by centrifugation for 20 min at 8000 rpm at 4°C (Hitachi 9-2 rotor) and kept frozen at Ϫ20°C. Non-myr MARCKS was purified by a procedure as described previously (27). For myr-MARCKS, the E. coli strain BL21(DE3)pLysS was transformed with the plasmid pBB131NMT, which contained the gene coding for yeast N-myristoyltransferase (NMT) (28). The strain BL21(DE3)pLysS-pBBNMT was then transformed with the plasmid pET3d vector containing the MARCKS gene. The bacterial culture was performed as described above except that 25 g/ml kanamycin was also included in the media. Coexpression of MARCKS and NMT was induced by adding 0.4 mM isopropyl-1-thio-␤-galactopylanoside to the log-phage culture. The cells were collected by centrifugation, resuspended in 5 volumes of ice-cold buffer (10 mM Tris-HCl (pH 7.5) containing 0.1% Triton X-100, 10 mM ditiothreitol, 1 mM EDTA, 0.2 mM phenylmethylsulfonyl fluoride) and sonicated three times for 5 min with a probe-type sonicator (Branson Sonifier 250). After centrifuging for 25 min at 18,000 rpm at 4°C, the supernatants were collected and then placed in a boiling water bath for 20 min and immediately cooled to 0°C. The precipitated proteins were removed by centrifugation for 20 min at 15,000 rpm at 4°C. After adding 7 mM CaCl 2 to the supernatant, the solution was loaded on a calmodulin-agarose column equilibrated with 40 mM Tris-HCl buffer (pH 7.5) containing 0.2 M NaCl, 0.1 mM ditiothreitol, and 7 mM CaCl 2 . The column was washed with 50 ml of the same buffer containing 0.5 M NaCl, 0.1 mM ditiothreitol, 7 mM CaCl 2 and then with 50 ml of the buffer containing 0.1 mM ditiothreitol, 0.1 mM CaCl 2 . MARCKS was eluted with the same buffer containing 0.6 M NaCl, 1 mM ditiothreitol, and 2 mM EGTA. Fractions containing myr MARCKS were pooled and concentrated/buffer exchanged with 20 mM Tris-HCl buffer (pH 7.5) using Centricon 10 concentrator (Amicon). The purified proteins were kept frozen at Ϫ80°C until use. We have prepared three separated MARCKS proteins. The purification and characterization of MARCKS proteins are stable in each preparation. Furthermore, we checked the reproducibility of the experiments and separate preparations did not affect the results of the several experiments. Molecular masses of the recombinant non-myr MARCKS and myr MARCKS proteins were determined to be 31,738 and 31,947 Da by electrospray mass spectrometry. The difference, 209 Da, corresponds very well to the mass difference of 210 Da expected for myristoylation. Although the calculated molecular mass of non-myr MARCKS based on the published cDNA sequence (26,29) would be 31,623.8, several errors exist in the cDNA sequence, and the observed mass represents an intact correct protein. 2 We concluded that the difference between myr and non-myr MARCKS resided solely in the N-terminal myristoylation. The stoichiometry of myristoylation was 1 for the myristoylated MARCKS protein. Protein concentration was determined by quantitative amino acid analysis or by densitometric analysis of the Coomassie Bluestained SDS gels.
Fluorescence Measurements-Binding of MARCKS or that of MARCKS peptides to dansyl-calmodulin was analyzed with a Jasco FP-777 spectrofluorometer using 1 ϫ 1-cm quartz cuvette as described previously (30,31). With the excitation wavelength set at 340 nm, emission spectra of dansyl-calmodulin in the presence or absence of MARCKS/MARCKS peptide were recorded in 20 mM Tris-HCl buffer (pH 7.5) containing 0.1 M NaCl, 0.5 mM CaCl 2 . The fluorescence emission at 490 nm was used to calculate dissociation constant of calmodulin-MARCKS complexes by a direct fit of the data to the mass equation using a nonlinear least square method (30).
Binding of MARCKS to Calmodulin-Agarose-Binding of recombinant MARCKS proteins to calmodulin was assessed using calmodulinagarose in 20 mM Tris-HCl buffer (pH 7.5) containing 0.5 mM CaCl 2 and indicated concentrations of NaCl. After centrifugation in a tabletop centrifuge, the supernatants were removed (unbound fraction). The calmodulin-agarose was washed three times with the same buffer, and the bound proteins were eluted with the SDS sample buffer containing 1% SDS (bound fraction). Both unbound and bound fractions were analyzed by SDS-gel electrophoresis. The bands were quantified by scanning densitometry (Amersham Biosciences PDSI).

Binding of Myr and Non-myr MARCKS to Calmodulin-We
have previously observed that non-myr MARCKS purified from soluble fractions of bovine brain does not bind to the calmodulin column as efficiently as the myr MARCKS (24). The nonmyr MARCKS can be eluted with high ionic strength even in the presence of Ca 2ϩ , whereas myr MARCKS is still bound to the column and can be eluted only in the absence of Ca 2ϩ . These observations suggest that the N-terminal myristoylation of MARCKS modulates the interaction of MARCKS with calmodulin, which is effected by the calmodulin-binding domain in the middle of the MARCKS molecule (20,24). To elucidate the underlying mechanism, we have produced recombinant human non-myr MARCKS and myr MARCKS proteins and studied the binding of the two forms to calmodulin in detail.
First, the effects of ionic strength on the binding were assessed by calmodulin-agarose assay as described under "Experimental Procedures." Both myr and non-myr MARCKS were mixed with calmodulin-agarose, and the bound protein was analyzed by SDS-gel electrophoresis. The extent of the binding of non-myr MARCKS to calmodulin-agarose gradually decreased with increasing concentrations of NaCl, whereas that of myr MARCKS was almost unaffected (Fig. 1). Even in the presence of 1 M NaCl, myr MARCKS did not lose its affinity to calmodulin-agarose, while the binding of non-myr MARCKS to calmodulin was almost abolished. When agarose beads were instead of calmodulin-agarose as a control, neither myr nor non-myr MARCKS showed significant binding to the agarose, demonstrating that we were dealing with specific interactions with calmodulin (data not shown). These results suggest that the myristoylation indeed affects the binding of MARCKS to calmodulin, and myr and non-myr MARCKS have different calmodulin-binding modes, especially in terms of the sensitivity to ionic strength.
Next, fluorescence change of dansyl-calmodulin upon binding of target protein/peptides (30 -32) was used to analyze the binding quantitatively. The addition of 150 nM myr or non-myr MARCKS to 50 nM dansyl-calmodulin in a buffer containing 0.1 M NaCl induced similar shifts in the maxima of fluorescence emission spectra of dansyl-calmodulin from 510 to 490 nm and about 2-fold increases in the intensity, suggesting that both forms bound tightly to calmodulin (Fig. 2a). Although the two emission spectra obtained with myr and non-myr MARCKS were very similar, there was a clear difference in the peak maxima, and the difference was observed repeatedly. This phenomenon was definitely observed when both the concentration of dansyl-calmodulin and that of MARCKS were increased (data not shown). This may reflect a difference in the conformation of the calmodulin complexes. To obtain the dissociation constants, a fixed concentration of dansyl-calmodulin (50 nM) was titrated with either myr MARCKS or non-myr MARCKS, and the changes in the emission spectra at 490 nm were recorded (Fig. 2b). The dissociation constants obtained in 0.1 M NaCl were 4.5 Ϯ 0.4 nM (n ϭ 3) and 12.7 Ϯ 3.7 nM (n ϭ 3) for my MARCKS and non-myr MARCKS, respectively. The binding of myr MARCKS to calmodulin, therefore, is slightly but significantly tighter than the non-myr protein. Similarly, the dissociation constants were determined to be 130 and 470 nM at 0.25 M NaCl for myr MARCKS and non-myr MARCKS, respectively. Unfortunately, the fluorescence change became very small above 0.5 M NaCl, and no quantitative determination of the dissociation constants was possible. However, very little, if any, change in the fluorescence spectra was observed with non-myr MARCKS at NaCl concentrations above 0.75 M, while the addition of myr MARCKS still produced an increase in the fluorescence intensity and a shift in the peak maximum even at 1 M NaCl. Thus, the results obtained in solution were similar to those obtained with the calmodulin-agarose assay; myr MARCKS retained the affinity to calmodulin at high ionic strengths, while non-myr MARCKS lost the affinity.
Effects of Ionic Strength on the Calmodulin-Effector Domain Interaction-Since the involvement of the basic amphiphilic phosphorylation domain (residues 150 -175 in bovine sequence) in the MARCKS-calmodulin interaction has been well demonstrated (20), it is of interest to examine the effects of ionic strength on the interaction of the domain with calmodulin. The fluorescence assay using dansyl-calmodulin gave a dissociation constant of 3.8 nM at 0.1 M NaCl, which was comparable with the values reported previously (20). When the titration was carried out at 0.5 M NaCl, a dissociation constant of 203 nM was obtained. Therefore, the binding of the basic phosphorylation domain to calmodulin is highly sensitive to ionic strength. Clearly, the sensitivity to ionic strength observed with non-myr MARCKS-calmodulin interaction reflects the characteristics of the calmodulin binding of the effector domain. In contrast, myr MARCKS retained the affinity to calmodulin even at high NaCl concentrations. These results suggest the presence of a second calmodulin-binding site in myr MARCKS that is insensitive to ionic strength.
Interaction of N-terminal Myristoylation Peptide with Calmodulin-Since the only difference between the myr MARCKS and non-myr MARCKS is the presence of a myristoyl group at the N terminus, either the myristoyl moiety is directly involved in the interaction with calmodulin, or the modification induces a conformational change of the protein that may expose a new binding site. To assess the former possibility, a myristoylated peptide based on the N-terminal sequence (myr-GAQFSK-TAAK) was synthesized, and the binding of the peptide to dansyl-calmodulin was examined. As shown in Fig. 3, the addition of the myristoylated peptide to dansyl-calmodulin at 0.1 M NaCl caused a drastic increase in the intensity and a shift of the peak maximum of the emission spectra similar to those observed with the myr and non-myr MARCKS proteins (see Fig. 2). On the other hand, no significant change in the fluorescence spectra was observed when a non-myristoylated Nterminal peptide (GAQFSKTAAKGEATAER) was added to calmodulin (Fig. 3). Furthermore, the addition of myristic acid alone or that of the mixture of myristic acid and non-myristoylated peptide to dansyl-calmodulin did not affect the fluorescence spectra significantly (data not shown). These results not only indicate that the myristoyl moiety is directly involved in the peptide-calmodulin interaction but also suggest that the N-terminal peptide part is also important for the interaction. The dissociation constant was calculated from the titration data to be 2.0 M, a value that was lower than that obtained for the basic effector domain (3.8 nM), but similar to those observed with low-affinity calmodulin-binding proteins such as GAP-43 (33). Interestingly, the binding of the myrstoylated N-terminal peptide to calmodulin was not affected by ionic strength. The dissociation constant obtained at 0.5 M NaCl was 1.5 M, which was almost identical to that observed at 0.1 M NaCl.
To study the role of the N-terminal protein moiety in the myristoylated peptide-calmodulin interaction, a series of myristoylated peptides were synthesized, and their binding characteristics to calmodulin were analyzed by fluorescence measurements. Since both hydrophobic and basic amino acids play important roles in the calmodulin-target-protein interactions (34,35), we have replaced either the Phe 4 or Lys 6 in the original peptide with various amino acids (Fig. 4). When Phe 4 was replaced with an Ala or an Asp, no significant change in the fluorescence spectrum was observed, suggesting that the hydrophobic amino acid is important for the calmodulin interaction. Interestingly, when the same residue was replaced with a Lys, an increase in intensity and a shift of the peak maximum of the emission spectra similar to those observed with the original sequence were observed. The dissociation constant obtained at 0.1 M NaCl was 1.4 M, which was almost identical to that of the wild type peptide. When Lys 6 was replaced with an Asp, no detectable change in the fluorescence spectrum was observed. Therefore, both hydrophobic and basic residues are involved in the interaction of the myristoylated peptides with calmodulin, and the presence of positively charges seems to be important. Taken together, these results suggest that the myristoyl moiety, together with the N-terminal protein domain, constitutes the ionic strength-insensitive second calmodulinbinding site in myr MARCKS. DISCUSSION Studies on a variety of myristoylated proteins suggest that myristoyl group may have a variety of roles (36). Its most obvious function is that in the association of the protein to phospholipid membranes. The intermediate hydrophobicity of the modification is ideal for the reversible membrane interaction (37). We have previously shown that the myristoylation is directly involved in the phosphorylation-dependent MARCKSphospholipid interaction (7). So called myristoyl switch mechanisms are also found in some myristoylated proteins, such as recoverin (38) or ADP-ribosylation factor (39), which has been implicated in the reversible membrane interaction. On the other hand, the involvement of the modification in the proteinprotein interactions remained unsolved despite intensive studies. In the case of the catalytic subunit of cAMP-dependent protein kinase, the presence of myristoylation seems to stabilize the protein, suggesting that the myristoyl group interacts with the protein moiety (10). Recent reports on the threedimensional structures of myristoylated and non-myristoylated recoverin showed that the myristoyl moiety interacts with the recoverin protein, and the binding of Ca 2ϩ to recoverin causes a drastic conformational change; the myristoyl moiety protrudes from the protein core, which can interact directly with membrane phospholipids (40). However, the involvement of the modification in the interaction of recoverin with other proteins has been questioned (13). The presence of a myristoyl receptor has also been much speculated, but the putative receptor for p60 src has been shown not to be physiological (41,42), and the possibility of a membrane receptor for MARCKS has also been dismissed (43). The results presented in the present study, therefore, are the direct demonstration of the involvement of protein myristoylation in the protein-protein interaction. Since we have recently demonstrated that the myristoyl group of CAP-23/NAP-22 and HIV-1 Nef are essential for the interaction between the proteins and calmodulin (14 -16), we propose a new hypothesis that the protein myristoylation plays an important role in protein-calmodulin interactions.
Since MARCKS has a canonical calmodulin-binding domain in the middle of the molecule, how the second binding site consisting of the myristoyl moiety and the N-terminal domain interacts with calmodulin is an interesting question. While the dissociation constant observed with the synthetic peptide is relatively low, the myr MARCKS showed an affinity to calmodulin about 3-fold higher than that of the non-myr MARCKS. If the two sites interact with calmodulin independently, and if there is not physical interaction between the sites, one should observe two independent binding reactions with two distinct affinities. On the other hand, if there is an interaction between the two site, one would expect one binding reaction with a single dissociation constant that is a product of two independent dissociation constants. Since the presence of the second binding site with low affinity affected the overall affinity of myr MARCKS, we can assume that the N-terminal domain and the effector domain bind to calmodulin cooperatively. The elucidation of the detailed mechanism would require the determination of the three-dimensional structure. However, it should be noted in this context that protein domains other than canonical calmodulin-binding domains in other calmodulin-target proteins have been shown to be involved in the calmodulin interaction (44). The canonical calmodulin-binding domain, therefore, is clearly not the sole interaction site of the calmodulintarget proteins with calmodulin.
We have previously demonstrated the presence of demyristoylase activity in the cytoplasmic fraction of synaptosomes (22,23). We also found that calmodulin inhibited the demyristoylation reaction in a Ca 2ϩ -independent manner. This observation was rather puzzling at that time, since the N-terminal domain where the cleavage occurs and the calmodulin-binding domain are about 150 amino acids apart, and MARCKS has been shown to assume an elongated overall structure (31,(45)(46)(47). The direct interaction of the N-terminal domain with calmodulin as demonstrated in the present study solved the puzzle easily; the direct interaction of calmodulin with the MARCKS N-terminal domain could affect the susceptibility of the N-terminal myristoyl moiety to the demylystoylase. Since the demyristoylation of MARCKS is regulated by calmoduin, and since the demyristoylation affects in turn the MARCKScalmodulin interaction, the N-terminal myristoyl domain and the basic effector domain are functionally interdependent.
Schleiff et al. (48) have reported the lack of effects by protein myristoylation on the properties of MARCKS-related protein (MRP or also called F52) in solution. Using myr MRP and non-myr MRP, they have investigated the secondary structure of MRP, phosphorylation by protein kinase C, binding to calmodulin, and the inhibitory effect of divalent cations. According to their results, none of these properties were significantly altered by myristoylation. They have pointed out, however, that myristoylation might modulate the mechanism of recognition of MRP by calmodulin and consequently alter the kinetics of complex formation. They have recently demonstrated that the myristoylated N-terminal domains of MRP interact with calmodulin in the complex, whereas the non-myristoylated N-terminal domains do not (49). We have repeated the calmodulin binding experiments with MRP and observed a difference in sensitivity to ionic strength similar to that observed with MARCKS. 3 Therefore, it is reasonable to assume that the myristoyl moiety of MRP also interacts with calmodulin. Thus, the myristoylated N-terminal domains of MARCKS family proteins play important roles both in their interaction with membrane phospholipids and in the protein-protein interaction with calmodulin.