We have used isotope-edited nuclear magnetic resonance spectroscopy, binding studies, and ATPase activity assays to investigate the interaction with F-actin of the 10 kDa C-terminal 658C fragment of chicken gizzard caldesmon and two site-directed mutants of this fragment. Simultaneous dual-sited contacts with F-actin are observed for the segments of the 658C sequence flanking tryptophan residues 692 and 722. Competition experiments showed that both 658C contacts with actin are displaced by substoichiometric concentrations of the short inhibitory region of troponin-I indicative of different binding sites on actin for these regions of troponin-I and caldesmon. Substitution of caldesmon serine-702 by aspartic acid within the spacer region linking the two actin contacts of 658C led to weaker binding but with retention of equivalent affinity for each interaction site. Differential binding affinity of the two sites was achieved by replacement of the sequence Glu691-Trp-Leu-Thr-Lys-Thr696 by Pro-Gly-His-Tyr-Asn-Asn. Consistent with these data, the concentration of this Cg1 mutant required to achieve 50% inhibition of actin-tropomyosin-activated myosin ATPase was 4-fold greater than found for the 658C fragment. Although calmodulin binding to Cg1 was observed, calmodulin proved ineffective in relieving the inhibition induced by the binding of this mutant to actin. These results are discussed in light of the actin contacts which are involved in the inhibitory activity possessed by different regions of the C-terminus of caldesmon.

*H-NMR and 31P-NMR spectroscopy were employed to assess the electrostatic consequences of phosphorylation of single and multiple tyrosine residues in peptides derived from the core and tail autophosphorylation regions of the human insulin receptor tyrosine-kinase domain. In both peptides, phosphorylation was accompanied by changes in the resonances from basic side-chains ; those from acidic residues were unaffected. Tyrosine phosphorylation caused increases of up to one in the pK, values of histidine residues situated up to eight residues away in the primary sequence. Titration curve analysis by Hill plots suggested some cooperativity of histidine and phosphate ionizations. Behaviour closely analogous to that of the insulin receptor tail peptide was observed during changes in phosphorylation of the intact insulin receptor kinase domain, suggesting that the electrostatic dissemination effects seen for the isolated peptide are retained by the peptide sequence in the context of the much larger protein. Similar changes in the behaviour of basic residues were also observed upon tyrosine phosphorylation of a cdc2-derived peptide, suggesting that this potential of phosphorylation events to propagate directed structural changes may find a widespread utility in the activation of protein kinases and in the transduction of phosphorylation-based signalling.Reversible protein phosphorylation is the most common cellular regulation mechanism. Phosphorylation is associated with a stereochemical influence on the equilibrium distribution of conformational states of the protein, which results in a net change in the average protein structure. Two mechanisms of phosphoregulation have so far been recognised: allosteric regulation in phosphorylase, where phosphorylation induces a domain rotation (reviewed by Johnson, 1992), and, as exemplified by isocitrate dehydrogenase, direct control mediated principally by electrostatic effects at the active site, thereby modulating substrate binding (Hurley et al., 1989).The recently determined structure of a phosphoregulated kinase (ERK2) (Zhang et al., 1994) suggests that a combination of these two paradigm mechanisms underlies the regulation of catalytic activity common to many kinases, a large number of which require multisite phosphorylation to trigger kinase activity. One such example is the human insulin receptor Correspondence to B. A. Levine,

Using NMR spectroscopy to visualise tyrosine phosphorylation kinetics in real time, we have investigated the sequence‐dependent determinants of the selectivity of the human insulin receptor protein‐tyrosine kinase for different tyrosine residues. The peptides used encompass the multipletyrosine‐containing autophosphorylation site sequences from the insulin receptor kinase core domain (Tyr1158, Tyr162 and Tyr1163) and from its specific C‐terminal tail domain (Tyr1328 and Tyr1334). Comparison of the phosphorylation kinetics with those found for the tyrosine residues on a peptide comprising the regulatory tyrosine phosphorylation site of cdc2 points to the role of the primary sequence context of the phosphate acceptor. The particularly deleterious influence of a basic residue immediately C‐terminal to the tyrosine is discussed in relation to the autophosphorylation properties of the regulatory loop regions of the insulin and epidermal growth factor receptor kinases. The data further suggest that receptor tyrosine kinase active sites and their substrate targets act in concert to ensure that specific downstream effects are activated.

No abstract

No abstract