Search

Recent analysis has shown the presence of three sequences in the C-terminal 170 amino acids of human caldesmon (domain 4) which are involved in actin binding and tropomyosin-dependent inhibition of actomyosin ATPase. Two are in domain 4b (amino acids 715-793) and one is in domain 4a (amino acids 636-714). In the present work we have compared recombinant peptides containing either domain 4a or domain 4b to address the question as to whether domain 4a alone has any inhibitory activity. We have produced three new recombinant fragments containing domain 4a: H10 [622-708], H12 [506-708] and H13 [622-726] and we have characterized their functional properties. All three fragments bound to actin and tropomyosin. Caldesmon, but not domain 4b, was able to displace the fragments H10, H12 and H13 from actin. Thus the isolated caldesmon domain 4a peptides bind to the same region on actin as in the whole molecule while domains 4a and 4b occupy different sites on the actin molecule. Unlike domain 4b, none of the domain 4a fragments inhibited the actomyosin ATPase in the absence of tropomyosin. However both domain 4a and 4b fragments displayed an inhibitory activity in the presence of tropomyosin. H13 and H12 were more potent inhibitors than H10. Ca2+-calmodulin bound to H13 and reversed the inhibitory activity of this fragment but did not bind to H10 and H12. We conclude that domain 4a can act as an independent inhibitory actin-tropomyosin binding domain, but its properties are very different from the extreme C-terminal domain 4b.

Caldesmon interaction with smooth muscle myosin and its ability to cross-link actin filaments to myosin were investigated by the use of several bacterially expressed myosin-binding fragments of caldesmon. We have confirmed the presence of two functionally different myosin-binding sites located in domains 1 and 3/4a of caldesmon. The binding of the C-terminal site is highly sensitive to ionic strength and hardly participates in acto-myosin cross-linking, while the N-terminal binding site is relatively independent of ionic strength and apparently contains two separate myosin contact regions within residues 1-28 and 29-128 of chicken gizzard caldesmon. Both these N-terminal sub-sites are involved in the interaction with myosin and are predominantly responsible for the caldesmon-mediated high-affinity cross-linking of actin and myosin filaments, without affecting the affinity of direct acto-myosin interaction. Binding of caldesmon and its fragments to myosin or rod filaments revealed affinity in the micromolar range. We determined various stoichiometries at maximal binding, which depended on the ionic strength and the concentration of Mg2+ ions. At 30 mM NaCl and 1 mM Mg2+ the maximum stoichiometry was 4 moles of caldesmon (or caldesmon fragment) per mole of myosin. At 130 mM NaCl/1 mM Mg2+, or at 30 mM NaCl/5mM Mg2+ it decreased to about two caldesmon molecules bound per myosin, while remaining 4:1 for individual caldesmon fragments, suggesting that all binding sequences on myosin were still fully capable of interaction. A further increase in the Mg2+ concentration led to a substantial decrease in both the affinity and maximum stoichiometry of caldesmon and the fragments binding to myosin. We suggest that caldesmon-myosin interaction varies according to the conformation of caldesmon in solution, that caldesmon-binding sites on myosin are not well defined and that their accessibility is determined by spatial organization and is blocked by divalent cations like Mg2+.

The interaction of intact calmodulin and its four tryptic peptides with deletion mutants of caldesmon was analysed by native gel electrophoresis, fluorescence spectroscopy and zero-length cross-linking. Deletion mutants H2 (containing calmodulin-binding sites A and B) and H9 (containing sites B and B') interacted with intact calmodulin to form complexes whose stoichiometries varied from 2:1 to 1:1. The N-terminal peptides of calmodulin (TR1C, residues 1-77, and TR2E, residues 1-90) bound H2 with higher affinity than H9. At the same time H2 was less effective than H9 in binding to the C-terminal peptides of calmodulin TR2C (residues 78-148) and TR3E (residues 107-148). The N-terminal peptides of calmodulin (TR1C and TR2E) could be cross-linked to intact caldesmon and its deletion mutants H2 and H9. The similarity in the primary structures of sites A and B' of caldesmon and our measurements of the affinities of H2 and H9 to calmodulin and its peptides strongly indicate an orientation of the protein complex where sites A and B' interact with the N-terminal domain of calmodulin, whereas site B interacts with the C-terminal domain of calmodulin. The spatial organization of contact sites in the caldesmon-calmodulin complex agrees with the earlier proposed two-dimensional model of interaction of the two proteins [Huber, El-Mezgueldi, Grabarek, Slatter, Levine and Marston (1996) Biochem. J. 316, 413-420].

The binding of Ca(2+)- and Ba(2+)-calmodulin to caldesmon and its functional consequence was investigated with three different calmodulin mutants. Two calmodulin mutants have pairs of cysteine residues substituted and oxidized to a disulphide bond in either the N- or C-terminal lobe (C41/75 and C85/112). The third mutant has phenylalanine-92 replaced by alanine (F92A). Binding measurements in the presence of Ca2+ by separation on native gels and by carbodiimide-induced cross-linking showed a lower affinity for caldesmon in all the mutants. When Ca2+ was replaced by Ba2+ the affinity of calmodulin for caldesmon was further reduced. The ability of Ca(2+)-calmodulin to release caldesmon's inhibition of the actin-tropomyosin-activated myosin ATPase was virtually abolished by mutation of phenylalanine-92 to alanine or by replacing Ba2+ for Ca2+ in native calmodulin. Both cysteine mutants retained their functional ability, but the increased concentration needed for 50% release of caldesmon inhibition reflected their decreased affinity. Ca2+ -calmodulin produced a broadening in the signals of the NMR spectrum of the 10 kDa Ca(2+)-calmodulin-binding C-terminal fragment of caldesmon arising from tryptophans -749 and -779 and caused an enhancement of maximum tryptophan fluorescence of 49% and a 16 nm blue shift of the maximum. Ca(2+)-calmodulin F92A produced a change in wavelength of 4 nm but no change in maximum, whereas Ca(2+)-calmodulin C41/75 binding produced a decrease in fluorescence with no shift of the maximum. We conclude that functional binding of Ca(2+)-calmodulin to caldesmon requires multiple interaction sites on both molecules. However, some structural modification in calmodulin does not abolish the caldesmon-related functionality. This suggests that various EF hand proteins can substitute for the calmodulin molecule.

We have produced nine recombinant fragments, H1 to H9, from a human cDNA that codes for the C-terminal 288 residues of caldesmon. The fragment H1, encompassing the 288 residues, is equivalent to domains 3 and 4 of caldesmon (amino acids 506-793 in human, 476-737 in the chicken gizzard sequence). It has been shown [Huber, Redwood, Avent, Tanner and Marston (1993) J. Muscle Res. Cell Motil. 14, 385-391] to bind to actin, Ca(2+)-calmodulin, tropomyosin and myosin. The fragments, H2 to H9, differ in length between 60 and 176 residues and cover the whole of domains 3 and 4 with many of the fragments overlapping. We have characterized the myosin and tropomyosin binding of these fragments. The binding of both tropomyosin and myosin is highly dependent on salt concentration, indicating the ionic nature of these interactions. The location of the myosin binding is an extended region encompassing the junction of domains 3/4 and domain 4a (residues 622-714, human; 566-657, chicken gizzard). Tropomyosin binds in a smaller region within domain 4a of caldesmon (residues 663-714, human; 606-657 chicken gizzard). We confirmed predictions based on sequence similarities of a tropomyosin binding site in domain 3 of caldesmon; however, this site bound to skeletal-muscle tropomyosin and had little affinity for the smooth-muscle tropomyosin isoform. None of the protein fragments H2-H9 retained the affinity of the parent fragment H1 for either myosin or tropomyosin. This indicates the need for several interaction sites scattered over an extended region to attain higher affinity. The regions interacting with caldesmon in both tropomyosin and myosin are coiled-coil structures. This is probably the reason for their shared interaction sites on caldesmon and their similar natures of binding.