Earlier 'H-NMR experiments on the myosin subfragment-1 (Sl) light chain isoenzymes from rabbit fast muscle, containing either the A1 or the A2 alkali light chains [Sl(Al) or Sl(A2)], have shown that the 41-residue N-terminal extension of Al, rich in proline, alanine and lysine residues, is freely mobile in solution but that this mobility is constrained in the acto-Sl(A1) complex [Prince et al. (1981) Eur. J. Biochem. 122,213-2191. It is now established that this N-terminal region of the Al-light chain interacts directly with the C-terminal region of actin in the acto-Sl(A1) complex. This was shown by covalently labelling the Cys-374 residue of actin with a spin-label and observing the enhanced relaxation this paramagnetic centre induced in the 'H-NMR spectrum of Sl(A1). In particular, the signal arising from the -N+(CH& protons of a-N-trimethylalanine (Me,Ala) were monitored as this residue is uniquely sited at the N-terminus of the A1 light chain [Henry et al. (1982) FEBS Lett. 144, 11 -151. Experiments using complexes of actin with either the N-terminal37-residue peptide of Al, Sl(A1) or heavy meromyosin indicate that the N-terminal region of A1 is binding in a similar manner to actin in each case, with the N-terminal Me3Ala residue within 1.5 nm of the spin label introduced to Cys-374 of actin.A similar strategy was adopted to show that the Me3Ala residue can also be found close (< 1.5 nm) to the fast-reacting SH1 thiol group on the S1 heavy chain. These data, together with published work, have been used to suggest a possible organisation for the polypeptide chains in the myosin head.The mechanism of muscle contraction is generally considered to consist of a cyclic series of detachments and attachments of the myosin heads (the cross-bridges) to actin in response to ATP hydrolysis resulting in displacement of two filament types relative to one another [l -31. Although the kinetic events surrounding the mechanism are well understood [4, 51, the molecular events responsible for the forcegenerating step remain elusive, despite intense study in many laboratories. The currently held theories involve a reorientation of the myosin head relative to the actin filament during the course of ATP hydrolysis [l, 61. Implicit in this model is the existence of (at least) two differently attached states one corresponding to the rigor actomyosin complex and one corresponding to some actin . myosin. nucleotide complex in

High-resolution proton N M R spectrocopy has been used to study the solution structures of the subfragment 1 (SI) isoenzymes (containing either the A1 or A2 light chains) from rabbit skeletal muscle myosin and to investigate their interaction with actin. Superimposed upon broad components, the narrow signals of the S1 spectra are unexpectedly sharp, indicating that domains of varying sidechain mobility occur in the conformation adopted in solution. These observations are in agreement with previous studies of the mixed isoenzymes [Highsmith et al. (1 979) Biochemistry, 18, 4238 -42431. Peptide amide exchange studies show also that the S 1 structure accommodates fluctuations of sufficient amplitude to allow most of the peptide groups to come into contact with the solvent on the time scale-of the 'H-NMR experiment. The overall impression is that S1 is a molecule possessing backbone motility as well as domains of different sidechain mobility.Close comparison of the Sl(A1) and Sl(A2) spectra indicate that the N-terminal41 residues of the A1 light chain, rich in lysine, proline and alanine, display a high degree of segmental mobility. The difference spectrum [SI(AI)-Sl(A2)] obtained closely resembles the spectral simulation of the 41-residue segment. Upon addition of actin, many of the narrow S1 resonances decrease in intensity or progressively disappear altogether, indicative of intermediateslow exchange conditions consistent with the recognised high affinity between the two proteins. These changes are interpreted as an overall modulation in the observed and hence more mobile regions of S1 as has been suggested in earlier H-NMR studies referred to above. In particular, the differences noted between S l(A 1 ) and S l(A2) have now largely disappeared in their complexes with actin indicating a marked reduction in the segmental mobility of the N-terminal region of the light chain in S l(A 1). Together with other affinity chromatography results [Winstanley and Trayer (1 979) Biochem. Soc. Trans. 7, 703 -7041, this is good evidence for a direct interaction between this area of S l(A 1) and actin.The mechanism of muscle contraction, as originally proposed by Huxley [I] and modified by several groups of authors [2, 31, consists of a cycle of cross-bridge detachments and attachments, which are the basis of the 'sliding filament' theory, The myosin heads form cross-bridges with actin, which are then detached by ATP. The hydrolysis of ATP produces some conformational change in myosin which is restored as mechanical energy when the heads rebind actin. There is now ample evidence that this theory is correct. There are, however, considerable differences of opinion with regard to the intermolecular and intramolecular interactions involved in the actomyosin complex, and several models have been proposed, as reviewed by Taylor [4]. X-ray diffraction and electron microscopy indicate that the cross-bridge or part of it rotates during contraction [5] and the existence of a 'swivel' and a 'hinge' have been postulated in myosin [6 -1 I...

There are two isoforms (A1 and A2) of the myosin essential light chain (ELC) and consequently two isoenzymes of myosin subfragment 1 (S1), S1(A1) and S1(A2). The two isoenzymes differ in their kinetic properties with S1(A1) having a lower apparent K m for actin and a slower turnover of MgATP (k cat ) than S1(A2). The two forms of the ELC differ only at their N-termini where A1 has an additional 40-odd amino acids that are not present in A2.The human atrial ELC (an A1-type ELC) was overexpressed in Escherichia coli and purified by ammonium sulphate fractionation and ion-exchange chromatography. The recombinant ELC had actinactivated MgATPase kinetics similar to those for rabbit skeletal S1(A1) under the same conditions. Deletion of the first 45 amino acid residues resulted in an ELC similar to the rabbit skeletal A2 isoform and, when hybridised into S1, in S1(A2)-like kinetic properties. Results obtained with an ELC mutant that lacks the first 11 residues were intermediate between these two extremes but tending towards the S1(A2)-like phenotype.The wild-type ELC (both hybridised into S1 or free in solution) could be cross-linked to F-actin, whereas the deletion mutant lacking the first 45 amino acids could not. The deletion mutant lacking the first 11 amino acids cross-linked only poorly under the same conditions, consistent with the MgATPase data. We therefore conclude that these N-terminal eleven amino acids predominantly encode an actinbinding site which modulates the kinetics of the myosin motor. Furthermore, while free A1-type ELC cross-linked to both polymeric F-actin and the monomeric G-actin:DNase-I complex, the same ELC in S1(A1) could only cross-link to F-actin. This suggests that the light chain binds to a different actin monomer than the heavy chain.

Methods of synthesizing a series of chemically-defined AMP, ADP, ATP, adenylyl imidodiphosphate and pyrophosphate derivatives suitable for affinity chromatography are extensively described. Each derivative has a single primary amino group at the end of a hexamethylene ;spacer' chain for attachment to CNBr-activated agarose. The synthesis of the derivative where the ;spacer' arm is attached directly to the 8 position of the adenine ring to produce 8-(6-aminohexyl)amino-AMP involves the direct bromination of AMP in the 8 position followed by displacement of the halogen by 1,6-diaminohexane. This monophosphate derivative can then be converted into the corresponding di- or triphosphate forms by direct phosphate condensation with carbonyl di-imidazole. A second series of adenosine phosphate derivatives with the phosphate moieties unsubstituted has been similarly prepared from N(6)-(6-aminohexyl)-AMP (Guilford et al., 1972). A third type of ligand has been synthesized by condensing the phosphoryl imidazolide of AMP with 6-aminohex-1-yl phosphate. This compound, P(1)-(6-aminohex-1-yl) P(2)-(5'-adenosyl) pyrophosphate, has an unsubstituted adenine ring. The synthesis of a fourth type of ligand, 6-aminohex-1-yl pyrophosphate, was done by heating 6-aminohexan-1-ol with crystalline pyrophosphoric acid under reduced pressure. The structures of the synthesized compounds were confirmed by chemical, electrophoretic and chromatographic methods and by u.v. spectrometry. The general applicability of the synthetic methods used is discussed in relation to the preparation of other affinity adsorbents. Examples are given where these derivatives have been successful in reversibly binding dehydrogenases, kinases and myosin and its proteolytic subfragments. The partial purification of rat liver glucokinase on an ADP derivative is shown.

Fluorescence resonance energy transfer measurements have been made between Cys-374 on actin and Cys-177 on the alkali light chain of myosin subfragment 1 (S1) using several pairs of donor-acceptor chromophores. The labeled light chain was exchanged into subfragment 1 and the resulting fluorescently labeled subfragment 1 isolated by ion-exchange chromatography on SP-Trisacryl. The efficiency of energy transfer was measured by steady-state fluorescence in a strong binding complex of acto-S1 and found to represent a spatial separation between the two probes of 5.6-6.3 nm. The same measurements were then made with weak binding acto-S1 complexes generated in two ways. First, actin was complexed with p-phenylenedimaleimide-S1, a stable analogue of S1-adenosine 5'-triphosphate (ATP), obtained by cross-linking the SH1 and SH2 heavy-chain thiols of subfragment 1 [Greene, L. E., Chalovich, J. M., & Eisenberg, E. (1986) Biochemistry 25, 704-709]. Large increases in transfer efficiency indicated that the two probes had moved closer together by some 3 nm. Second, weak binding complexes were formed between subfragment 1 and actin in the presence of the regulatory proteins troponin and tropomyosin, the absence of calcium, and the presence of ATP [Chalovich, J. M., & Eisenberg, E. (1982) J. Biol. Chem. 257, 2432-2437]. The measured efficiency of energy transfer again indicated that the distance between the two labeled sites had moved closer by about 3 nm. These data support the idea that there is a considerable difference in the structure of the acto-S1 complex between the weakly and strongly bound states.