Identification of the Isoform-specific Interactions between the Tail and the Head of Class V Myosin*

Vertebrates have three isoforms of class V myosin (Myo5), Myo5a, Myo5b, and Myo5c, which are involved in transport of multiple cargoes. It is well established that the motor functions of Myo5a and Myo5b are regulated by a tail inhibition mechanism. Here we found that the motor function of Myo5c was also inhibited by its globular tail domain (GTD), and this inhibition was abolished by high Ca2+, indicating that the tail inhibition mechanism is conserved in vertebrate Myo5. Interestingly, we found that Myo5a-GTD and Myo5c-GTD were not interchangeable in terms of inhibition of motor function, indicating isoform-specific interactions between the GTD and the head of Myo5. To identify the isoform-specific interactions, we produced a number of Myo5 chimeras by swapping the corresponding regions of Myo5a and Myo5c. We found that Myo5a-GTD, with its H11-H12 loop being substituted with that of Myo5c, was able to inhibit the ATPase activity of Myo5c and that Myo5a-GTD was able to inhibit the ATPase activity of Myo5c-S1 and Myo5c-HMM only when their IQ1 motif was substituted with that of Myo5a. Those results indicate that the H11-H12 loop in the GTD and the IQ1 motif in the head dictate the isoform-specific interactions between the GTD and head of Myo5. Because the IQ1 motif is wrapped by calmodulin, whose conformation is influenced by the sequence of the IQ1 motif, we proposed that the calmodulin bound to the IQ1 motif interacts with the H11-H12 loop of the GTD in the inhibited state of Myo5.

Class V myosin (Myo5), a motor protein that transports cellular cargoes toward the plus end of actin filaments, is distributed widely from yeast to mammals. Vertebrates contain three isoforms of Myo5, Myo5a, Myo5b, and Myo5c (1-3), which are expressed in different tissues and exhibit distinct cargo specificities. Vertebrate Myo5a is so far the best characterized motor protein among the myosin superfamily (4,5).
The three vertebrate Myo5 isoforms share a similar structural organization. Myo5a heavy chain contains the N-terminal motor domain, followed by the neck region and the tail domain. The motor domain contains the ATP-and actin-binding site and is capable of converting energy from ATP hydrolysis into mechanical work. The neck region consists of six IQ motifs with the consensus sequence of IQXXXRGXXXR, the binding sites for calmodulin (CaM) 2 or CaM-like light chains. The neck region functions as a lever arm to amplify the small conformational change in the motor domain into large movement. The tail domain consists of two distinct regions. The proximal ϳ500 amino acids of the tail form a series of coiled coils separated by several flexible regions, and the distal ϳ400 amino acids form a globular tail domain (GTD). The tail domain plays three distinct roles. First, the coiled coil of the tail enables Myo5a to form a homodimer. Second, the GTD is an inhibitory domain of Myo5a motor function (6 -8). Finally, the whole tail, including the proximal tail and the GTD, mediates Myo5a binding to specific membrane-bound organelles such as melanosomes through scaffold proteins such as melanophilin (9 -11).
The GTD plays a central role in regulating the motor functions of Myo5. So far, the crystal structures of the GTDs of the five types of Myo5, including two types of yeast Myo5 (i.e. Myo2p and Myo4p) and three types of vertebrate Myo5 (i.e. Myo5a, 5b, and 5c), have been solved (12)(13)(14)(15)(16)(17). The structure of GTD is mainly composed of ␣ helices connected by short and long loops (Fig. 1, bottom panel). Despite the low sequence similarity among the GTD of Myo5 paralogs, their overall tertiary structures are very similar. The structural core of the GTD is composed of 12 ␣ helices and is artificially divided into subdomain I and II (12,15). The GTD interacts with different cargo-binding proteins through distinct binding sites located on the surface of both subdomain I and II (12, 14, 15, 18 -20). On the other hand, bioinformatic analysis and biochemical characterization suggest that the head-binding site is located in subdomain II (7,12,22).
The GTD interacts with the head and inhibits its motor function, and the inhibition is reversed by high Ca 2ϩ or cargo-binding proteins (6,8,(23)(24)(25). Because tail inhibition could help the cell avoid futile hydrolysis of ATP, it is tempting to speculate that tail inhibition is conserved among all members of Myo5. Indeed, we have demonstrated that mammalian Myo5b and Drosophila Myo5 are regulated by the GTD (26,27). However, tail inhibition may not exist in all types of Myo5. For example, sedimentation analyses of yeast Myo2p and Myo4p showed that both of them are in a state similar to the extended conformation of mammalian Myo5a, suggesting the absence of head-tail interaction (28). Moreover, the structure of the GTD of Myo4p reveals that surface residues involved in autoinhibition of Myo5 are neither conserved in Myo4p nor substituted by nearby residues with similar chemical properties (13).
Of the three isoforms of vertebrate Myo5, Myo5c is the most ancient type, sharing 60 -70% sequence similarity with Myo5a and Myo5b (29). The motor activities of Myo5c, including actin-gliding activity and actin-activated ATPase activity, are much lower than those of Myo5a and Myo5b (30,31). Unlike Myo5a and Myo5b, Myo5c is a non-processive motor (30,31). Recently, it was shown that two dimers of Myo5c molecules on a DNA scaffold enabled processive steps along actin filaments (32). Although the GTD of Myo5c has a similar overall structure as Myo5a and Myo5b (14 -17), it is still an open question whether Myo5c is regulated by a tail inhibition mechanism.
In this study, we demonstrated that, similar to those of vertebrate Myo5a and Myo5b, the motor function of Myo5c was inhibited by its GTD in a Ca 2ϩ -dependent manner. Interestingly, we found that Myo5a-GTD and Myo5c-GTD were not interchangeable in terms of inhibiting motor function, indicating the isoform-specific interactions between the GTD and the head of Myo5. To identify those isoform-specific interactions, we produced a number of GTD chimeras and Myo5-S1o and -HMM chimeras by swapping the corresponding regions of Myo5a and Myo5c. We found that Myo5a-GTD, with its H11-H12 loop being substituted with that of Myo5c, was able to inhibit the ATPase activity of Myo5c and that Myo5a-GTD was able to inhibit the ATPase activity of Myo5c-S1 only when its IQ1 motif was substituted with that of Myo5a. These results indicate that the H11-H12 loop in the GTD and IQ1 motif in the head dictate the isoform-specific interactions between the GTD and the head of Myo5.

Experimental Procedures
Materials-Restriction enzymes and modifying enzymes were purchased from New England Biolabs (Beverly, MA) unless indicated otherwise. Actin was prepared from rabbit skeletal muscle acetone powder according to Spudich and Watt (33). Glutathione-Sepharose 4 Fast Flow was purchased from GE Healthcare. HRP-anti-FLAG M2 antibody, anti-FLAG M2 affinity gel, phosphoenol pyruvate, 2,4-dinitrophenylhydrazine, and pyruvate kinase were from Sigma. FLAG peptide (DYKDDDDK) was synthesized by Augct Co. (Beijing, China). Oligonucleotides were synthesized by Invitrogen.
Expression and Purification of Myo5-GTD-The cDNA of mouse Myo5a-GTD (amino acids 1444 -1853) was subcloned into the pGEX4T2 expression vector (Amersham Biosciences) as described previously (7). The cDNA of human Myo5c-GTD (amino acids 1351-1742) was amplified by reverse transcription PCR from human kidney total RNA (BD Biosciences) and cloned into the pGEX4T2 vector using the restriction sites of BamH1 and Xho1. The cDNAs of the GTD chimeras of mouse Myo5a and human Myo5c were created by overlapping PCR or QuikChange PCR. The GTD was expressed as an N-terminal GST fusion protein in Escherichia coli BL21(DE3) cells and purified by glutathione-Sepharose affinity chromatography as described previously (7). The protein concentrations of the GTD were measured by absorbance at 280 nm using the following molar extinction coefficients: 80,120 liters mol Ϫ1 cm Ϫ1 , Myo5a-GTD(c281-392); 69,100 liters mol Ϫ1 cm Ϫ1 , Myo5a-GTD(c1-280); 73,270 liters mol Ϫ1 cm Ϫ1 , Myo5a-GTD(c322-392) and Myo5a-GTD(c340 -392); 77,230 liters mol Ϫ1 cm Ϫ1 , Myo5c-GTD; and 71,990 liters mol Ϫ1 cm Ϫ1 , Myo5a-GTD and the remaining GTD chimeras. Although the GTD used throughout this study is a dimeric molecule as a result of fusion with an N-terminal GST tag, the molar extinction coefficients refer to mole of polypeptide.
ATPase Assay-The ATPase activity of Myo5 was measured in a plate-based ATP regeneration system as described previously (7). For EGTA conditions, the ATPase assay was con-

Results
The ATPase Activity of Myo5c Is Inhibited by its GTD in a Ca 2ϩ -dependent Manner-We reported previously that the ATPase activities of Myo5a-HMM and Myo5a-S1 are inhibited by Myo5a-GTD in a Ca 2ϩ -dependent manner (6,37). To study whether this regulation mechanism is valid for Myo5c, we examined the inhibition of Myo5c-HMM and Myo5c-S1 by Myo5c-GTD. In this study, Myo5a-HMM contains the motor domain, six IQ motifs, and the first three predicted coiled coils, and Myo5c-HMM contains the motor domain, six IQ motifs, and the first predicted coiled coil. Both Myo5a-S1 and Myo5c-S1 contain the motor domain and the first IQ motif.
We measured the actin-activated ATPase activity (hereafter referred to as ATPase activity) of Myo5c-HMM in the presence of Myo5c-GTD under EGTA conditions or high Ca 2ϩ conditions (pCa4). Similar to that of Myo5a-HMM, the ATPase activity of Myo5c-HMM was strongly inhibited by Myo5c-GTD under EGTA conditions, and the inhibition was abolished under pCa4 conditions (Fig. 2, A and B). The apparent affinity between Myo5c-HMM and Myo5c-GTD was 102 Ϯ 17 nM, similar to that of Myo5a (57 Ϯ 2.3 nM). These results indicate that the tail inhibition mechanism is valid for the regulation of Myo5c motor activity.
We showed previously that Myo5a-GTD inhibits not only the ATPase activity of Myo5a-HMM but also that of Myo5a-S1 containing the motor domain and the IQ1 motif (34). Similarly, we found that Myo5c-GTD inhibits the ATPase activity of Myo5c-S1 in a Ca 2ϩ -dependent manner (Fig. 2C), indicating that Myo5c-S1 contains the essential components for interacting with Myo5c-GTD.
Myo5c-GTD and Myo5a-GTD Are Not Interchangeable in Terms of Inhibiting the Motor Function-To test the possibility that Myo5c-GTD is interchangeable with Myo5a-GTD in terms of inhibiting motor functions, we examined the inhibitory effect of Myo5c-GTD on Myo5a-HMM ATPase activity and that of Myo5a-GTD on Myo5c-HMM ATPase activity. Myo5a-HMM ATPase activity was inhibited in a dose-dependent manner by Myo5c-GTD, although the inhibition was not as effective as Myo5a-GTD, and the inhibition by Myo5c-GTD was abolished by high Ca 2ϩ (Fig. 3A). On the other hand, Myo5c-HMM ATPase activity was not inhibited by Myo5a-GTD (Fig. 3B). Consistent with the ATPase assay results, a FLAG pulldown assay showed that Myo5c-HMM directly binds to Myo5c-GTD but not to Myo5a-GTD (Fig. 3C). On the other hand, Myo5a-HMM binds to both Myo5a-GTD and Myo5c-GTD, although the interaction with the latter was weaker than with the former (Fig. 3C).
Identification of the Isoform-specific Region in the GTD of Myo5-Because Myo5c-HMM ATPase activity was inhibited by Myo5c-GTD but not by Myo5a-GTD and the overall structures of Myo5a-GTD and Myo5c-GTD are very similar to each other, we speculated that a specific region in Myo5c-GTD dictates the isoform-specific interaction with Myo5c-HMM. To identify this region in the GTD, we produced a number of GTD chimeras by swapping the corresponding region of Myo5a-GTD with that of Myo5c-GTD and analyzed their inhibitions on Myo5c-HMM ATPase activity.
We found that substitution of the C-terminal one-third of Myo5a-GTD (i.e. amino acids 281-392) with that of Myo5c-GTD enabled the GTD chimera to inhibit the ATPase activity of Myo5c-HMM (Fig. 4A), indicating that residues 281-392 of Myo5c-GTD are responsible for the isoform-specific interaction with Myo5c-HMM. On the other hand, substitution of the N-terminal two-thirds of Myo5a-GTD (i.e. amino acids 1-280) with that of Myo5c-GTD did not restore the inhibitory activity The ATPase activity was performed in the presence of 50 mM NaCl (for S1) or 100 mM NaCl (for HMM) under EGTA or pCa4 conditions. The data obtained under EGTA conditions were fit to a hyperbola, defining the affinity of the GTD to Myo5-HMM or Myo5-S1, which was 57 Ϯ 2.3 nM of Myo5a-GTD to Myo5a-HMM, 102 Ϯ 17 nM of Myo5c-GTD to Myo5c-HMM, or 700 Ϯ 16 nM of Myo5c-GTD to Myo5c-S1. The data are mean Ϯ S.D. from three independent assays. ( Fig. 4B). In other words, substitution of the C-terminal onethird of Myo5c-GTD with that of Myo5a-GTD abolished the inhibitory activity of Myo5c-GTD. These results indicate that a specific feature in the C-terminal one-third of Myo5c-GTD is responsible for the isoform-specific interactions with Myo5c-HMM.
Comparing with Myo5a-GTD, Myo5c-GTD contains three amino acid substitutions in residues 327-331, including a conservative substitution of E328D and the two non-conservative substitutions N327D and E331K (Fig. 1, bottom panel). We found that the N327D/E331K double mutation of Myo5a-GTD was sufficient to restore the inhibitory activity on Myo5c-HMM ATPase activity (Fig. 5D). On the other hand, single amino acid mutation of Myo5a-GTD, i.e. N327D, E328D, and E331K, could not notably restore the inhibitory activity on Myo5c-HMM (Fig. 5, A-C). These results indicate that both FIGURE 4. Identification of the critical region in the GTD for the isoform-specific interactions between the GTD and head of Myo5. The GTD chimeras were created by substituting the corresponding region in Myo5a-GTD with that of Myo5c-GTD. A-I, the ATPase activity of Myo5c-HMM in the presence of the GTD chimera. The data were fit to a hyperbola, defining the K d of the GTD chimera to Myo5c-HMM, which is indicated in the corresponding panel. The ATPase assays were performed in the presence of 100 mM NaCl under EGTA conditions. The data are mean Ϯ S.D. from three independent assays.
Asp 327 and Lys 331 are critical for the isoform-specific GTDhead interaction.
To further investigate the role of the three Myo5c-GTDspecfic residues in 327-331 on the GTD-head interaction of Myo5c, we introduced three single mutations in Myo5c-GTD, i.e. D327N, D328E, and K331E. D327N and K331E markedly decreased the inhibitory activity of Myo5c-GTD on Myo5c-HMM ATPase activity (Fig. 6, A and C), whereas D328E did not greatly affect the inhibitory activity (Fig. 6B). Moreover, we found that D327N/K331E double mutations in Myo5c-GTD strongly dampened the inhibition of Myo5c-HMM ATPase activity by Myo5c-GTD (Fig. 6D).
Identification of the Isoform-specific Region in the Head of Myo5-Similar to the effect on the ATPase activity of Myo5-HMM, Myo5a-GTD inhibited the ATPase activity of Myo5a-S1 but not that of Myo5c-S1 (Fig. 7A). The inhibitory state of Myo5 is formed by the binding of the GTD to the head. Because the overall structures of Myo5a-S1 and Myo5c-S1 are similar to each other, we hypothesized that a specific region in Myo5a-S1 dictates the isoform-specific interaction with Myo5a-GTD. To identify such a region in Myo5a-S1, we created two Myo5c-S1 chimeras, Myo5c(5aConv)-S1 and Myo5c(5aIQ1)-S1, by substituting Gln 684 -Gln 730 and Leu 753 -Arg 780 of Myo5c-S1 with the corresponding regions of Myo5a (Fig. 1, top panel). Note that, on the basis of the redefinition of the converter proposed by Houdusse et al. (38), the helix preceding the IQ1 of Myo5 should be included within the converter. Thus, Myo5c (5aConv)-S1 contains the entire converter of Myo5a except for Ala 763 . Myo5c(5aIQ1)-S1 contains the entire IQ1 of Myo5a and four additional residues, i.e. Ala 763 at the N terminus and the three residues of IQ2 (Cys 788 , Met 789 , and Gln 790 ) at the C terminus.
The ATPase activity of Myo5c(5aIQ1)-S1, but not Myo5c(5aConv)-S1, was inhibited by Myo5a-GTD (Fig. 7, B and C), indicating that IQ1 plays a key role in the isoformspecific GTD-head interaction. On the other hand, the inhibition of Myo5c(5aIQ1)-S1 by Myo5a-GTD was not as effective as that of Myo5a-S1, suggesting that other parts of Myo5a-S1 might also participate in the isoform-specific GTD-head interaction. To further investigate the role of IQ1 in isoform-specific interaction, we analyzed the effects of substitutions of the converter and IQ1, respectively, in Myo5c-HMM. Two Myo5c-HMM chimeras, Myo5c(5aConv)-HMM and Myo5c(5aIQ1)-HMM, were created by substituting Gln 684 -Gln 730 and Leu 753 -Arg 780 of Myo5c-HMM with the corresponding regions of Myo5a. Similar to the effects on Myo5c-S1 chimeras, substitution of the IQ1, but not the converter, partially restored the inhibition of Myo5c-HMM by Myo5a-GTD (Fig. 7D).

Discussion
In this study, we demonstrated that the ATPase activity of Myo5c is inhibited by its GTD and that this inhibition is abolished by a high level of Ca 2ϩ , which is similar to Myo5a and Myo5b. These results indicate that tail inhibition is a common mechanism among all three types of vertebrate Myo5. Because tail inhibition is likely a general regulatory mechanism for Myo5, the tail-head interaction would warrant a co-evolution of the tail (particularly the GTD) and the head of Myo5. This consideration is consistent with the sequence alignment of myosin heavy chain, which shows that the phylogenetic trees, on the basis of the head and the tail, respectively, are either identical or very similar to each other, suggesting that the head and tail have co-evolved and are likely to be functionally interdependent (39).
Although the three-dimensional structures of GTDs of Myo5a and Myo5c are very similar to each other, they are not interchangeable in terms of inhibiting the motor function of the head, indicating the isoform-specific interaction between the GTD and the head of Myo5. By swapping the corresponding portions of Myo5a-GTD with those of Myo5c-GTD and analyzing their inhibition on Myo5c-HMM ATPase activity, we identified the H11-H12 loop of Myo5c-GTD as the critical region dictating the isoform-specific interactions with the head of Myo5c. Using same strategy, i.e. domain swapping, we identified the IQ1 motif of Myo5a as the critical region dictating the isoform-specific interaction with Myo5a-GTD.
Because the high-resolution structure of the folded Myo5 is not available, the precise interaction between the GTD and the head of Myo5 is not known. We previously proposed a model of GTD-head interaction by manually docking yeast Myo2p-GTD onto the head of Myo5a. Recently, a similar model was independently proposed by Velvarska and Niessing (17) and by Nascimento et al. (16). Here we present the model of GTD-head interaction (Fig. 8A) by manually docking Myo5a-GTD onto the head of Myo5a with restriction described previously (7). The overall structure of our model is very similar to that proposed by Velvarska and Niessing (17) and by Nascimento et al. (16), although these structural models were created using different approaches. In our model of GTD-head interaction, the H11-H12 loop (corresponding to loop II in the model proposed by Velvarska and Niessing (17)) is in close proximity to the converter and/or the CaM bound to IQ1. Our finding that the H11-H12 loop and the IQ1 dictate the isoform-specific interaction between the GTD and the head of Myo5 suggests a direct interaction between the H11-H12 loop and the CaM bound to the IQ1 motif.
On the basis of the redefinition of the converter (38,40), the helix preceding the IQ1 (Gly 753 -Asp 764 of Myo5a and Gly 743 -Asp 754 of Myo5c) is an integral part of the converter. Myo5a and Myo5c are highly conserved in this region, except for one residue, i.e. Ala 763 in Myo5a and Leu 753 in Myo5c (Fig. 1, top  panel). The two converter-swapped constructs, Myo5c (5aConv)-S1 and -HMM, contain the entire converter of Myo5a except for Ala 763 (which is leucine in Myo5c). In other words, both Myo5c(5aConv)-S1 and -HMM contain the converter of Myo5a with the A763L mutation. However, it is unlikely that the A763L mutation affects the conformations of the converter and its interaction with the GTD. In the crystal structure of Myo5a, the side chain of Ala 763 has no interaction with the rest of the converter. Moreover, in our model of GTDhead interaction, Ala 763 is at least 10 Å away from the GTD (Fig.  8A). This distance is too far for the GTD to interact with alanine or even bulky residues, such as leucine, in this position.
CaM is a highly conserved eukaryotic protein that binds to a number of diverse target proteins. Structural analyses show that the conformation of CaM is influenced by the target sequence (21,36). There are 10 different amino acid residues between the IQ1 of mouse Myo5a and human Myo5c (Fig. 1,  top panel). Among them, six residues (Ala 769 , Ile 771 , Arg 772 , Ile 773 , Thr 776 , and Ile 777 ) interact with the C-lobe of CaM, and three residues (Arg 772 , Leu 782 , and Tyr 786 ) interact with the N-lobe of CaM in the crystal structure of CaM in complex with IQ1 of mouse Myo5a (Fig. 8B). Thus, CaM, particularly the C-lobe, should have a different conformation when bound to the IQ1 of Myo5a or to the IQ1 of Myo5c. Previously, our biochemical analyses have shown that the C-lobe of CaM bound to IQ1 of Myo5a is critical for the interaction with the GTD (34). We therefore proposed that the H11-H12 loop of the GTD interacts with the C-lobe of CaM bound to the IQ1 motif (Fig.  8A).
It is intriguing that the motor function of Myo5a was inhibited by Myo5c-GTD, whereas the motor function of Myo5c was not inhibited by Myo5a-GTD. The binding of GTD to the head Myo5a-S1 Myo5c-S1 FIGURE 7. Identification of the critical region in the head of Myo5a for the isoform-specific interactions between the GTD and head of Myo5a. A, relative ATPase activity of Myo5a-S1 and Myo5c-S1 in the presence of Myo5a-GTD. The ATPase activity of Myo5a-S1 and Myo5c-S1 in the absence of Myo5a-GTD was 9.87 Ϯ 0.24 s Ϫ1 head Ϫ1 and 2.68 Ϯ 0.27 s Ϫ1 head Ϫ1 , respectively. B, the ATPase activity of Myo5c(5aConv)-S1 in the presence of Myo5a-GTD. C, the ATPase activity of Myo5c(5aIQ1)-S1 in the presence of Myo5a-GTD. D, the ATPase activity of Myo5c(5aConv)-HMM and Myo5c(5aIQ1)-HMM in the presence of Myo5a-GTD. The ATPase activities were measured in the presence of 50 mM NaCl (for S1) or 100 mM NaCl (for HMM) under EGTA conditions. The data are mean Ϯ S.D. from three independent assays.
of Myo5 likely comprises multiple interactions, including the conserved interactions, such as the ionic interaction between the conserved acidic residue Asp 136 in the motor domain and the conserved basic residues Lys 245 and Lys 318 (corresponding to Lys 1706 and Lys 1779 of Myo5a and Lys 1595 and Lys 1668 of Myo5c) in the GTD, and the isoform-specific interactions, such as the interaction between the H11-H12 loop and the CaM bound to IQ1. It is possible that the conserved GTD-head interactions are preserved between Myo5c-GTD and the head of Myo5a but absent between Myo5a-GTD and the head of Myo5c.
We proposed that Myo5c-GTD is able to bind to the head of Myo5a through the conserved GTD-head interactions despite the absence of isoform-specific interactions between Myo5c-GTD and Myo5a-HMM. On the other hand, it is possible that the unique structure in the CaM on the IQ1 of Myo5c and/or that in Myo5a-GTD not only eliminates the isoform-specific interactions but also precludes the conserved interactions between Myo5a-GTD and the head of Myo5c, thus preventing the binding of Myo5a-GTD to the head of Myo5c.
Because IQ1 swapping in Myo5c-S1 did not completely restore the inhibition by Myo5a-GTD, it is possible that other portions of the head also participate in the isoform-specific interaction. In addition to the converter and the C-lobe of CaM bound to IQ1, other portions of the head, including loop 65-77 and the loop of Asp 136 , might also interact with the GTD (Fig. 8A). Those interactions might also be isoformspecific. In the case of IQ1 swapping in Myo5c-HMM, the inhibition of Myo5c(5aIQ1)-HMM by Myo5a-GTD was much weaker than that of Myo5a-HMM. This result is consistent with our previous finding that the distal portion of the first coiled coil of Myo5a is essential for potent inhibition by Myo5a-GTD (6). Because of the lack of a high-resolution structure of the inhibited Myo5a and Myo5c, the precise interactions between the GTD and the head of Myo5 remain to be determined.  Velvarska and Niessing (17). The residue numbers refer to mouse Myo5a. B, ribbon drawing of CaM (N-lobe, orange; C-lobe, green) in complex with IQ1 of Myo5a (PDB code 2IX7). The non-identical residues of IQ1 between mouse Myo5a and human Myo5c are shown as sticks.