Sustained anti-beta-adrenergic effect of melatonin in guinea pig heart papillary muscle.

OBJECTIVE: Anti-ss-adrenergic actions of several substances influence heart function significantly. The anti-ss-adrenergic effect of melatonin was investigated, with special attention to protein kinase C (PKC) and nitric oxide (NO). DESIGN: Guinea pig papillary muscles were exposed to melatonin (500 pM) for 15 min and 20 min washout. Contractile force was measured during a bolus of isoproterenol (300 nM) given before melatonin, at the end of melatonin-exposure and after washout. In separate experiments blockers of PKC, NO-synthase (NOS) and melatonin receptors were added, or forskolin (10 microM) substituted for isoproterenol. RESULTS: Melatonin significantly reduced the increase in contractile force in response to isoproterenol, both when present and after melatonin-washout. The reduction was unaffected by inhibition of PKC, while inhibition of melatonin receptors or NOS seemed to abolish the effect. Melatonin induced a sustained but not acute reduction of contractile force response with forskolin stimulation. This was abolished by NOS-inhibition. CONCLUSION: Receptor-mediated immediate and sustained anti-ss-adrenergic effects of melatonin were demonstrated in contractile function. A role for NO in the response was indicated, while a role for PKC was not verified.

Antinucleosome autoantibodies bind directly to cell lines in vitro and via the FcgammaRIIB receptor to B lymphocytes in vivo: a role for immune complexes in interactions between antinucleosome IgG2a and B cells of BXSB lupus mice.

The initial novel observation of this study was that most B cells of male BXSB lupus mice bear surface IgG2a(b) of extrinsic origin. To define the surface antigen, we here examine three (NZBxBXSB)F1-derived IgG2a(b) monoclonal antibodies (mAbs) selected for binding to cell surfaces. Surprisingly, all three mAbs bound the nucleosome (nuc) particle, the fundamental unit of chromatin and an early target of autoimmunity in systemic lupus erythematosus. Their tentative dissociation constant (K(d)) for soluble nuc particles was approximately 7 x 10(-10) m. The mAbs bound more weakly to both H2A-H2B-DNA and H3-H4-DNA complexes, and in immunoblot they stained all four core histones. The mAbs detected a surface antigen on all cell lines examined, present on viable cells. When stripped of nuc, and in the presence of DNase I, their binding to cell lines improved. Heparin displaced the antigen from the cell surface. In vivo, the three mAbs stained B cells of several BALB/c mice clearly stronger than the isotype control; this differential staining was significantly reduced in FcgammaRIIB-deficient mice. The results indicate that the three mAbs recognize (a) planted antigen on viable cultured cells and (b) soluble autoantigen in vivo, leading to immune complexes that bind to FcgammaRIIB. Further experiments demonstrated that antinuc IgG2a could be eluted from splenocytes of a male BXSB lupus mouse. Hence, at least part of the extrinsic IgG2a(b) found on BXSB B cells may represent FcgammaRIIB-bound nuc-IgG2a(b) complexes.

Topology and dynamics of the 10 kDa C-terminal domain of DnaK in solution.

Hsp70 molecular chaperones contain three distinct structural domains, a 44 kDa N-terminal ATPase domain, a 17 kDa peptide-binding domain, and a 10 kDa C-terminal domain. The ATPase and peptide binding domains are conserved in sequence and are functionally well characterized. The function of the 10 kDa variable C-terminal domain is less well understood. We have characterized the secondary structure and dynamics of the C-terminal domain from the Escherichia coli Hsp70, DnaK, in solution by high-resolution NMR. The domain was shown to be comprised of a rigid structure consisting of four helices and a flexible C-terminal subdomain of approximately 33 amino acids. The mobility of the flexible region is maintained in the context of the full-length protein and does not appear to be modulated by the nucleotide state. The flexibility of this region appears to be a conserved feature of Hsp70 architecture and may have important functional implications. We also developed a method to analyze 15N nuclear spin relaxation data, which allows us to extract amide bond vector directions relative to a unique diffusion axis. The extracted angles and rotational correlation times indicate that the helices form an elongated, bundle-like structure in solution.

Entropic effects of disulphide bonds on protein stability.

To measure the thermodynamic consequences of the reduction in the number of polypeptide-chain conformations that accompanies protein folding, we developed a method called loop permutation analysis. In this approach, the stabilizing contributions of three engineered disulphide bonds were compared in extended and circularly permutated mutants of phage T4 lysozyme. The observed differences in disulphide contributions, although qualitatively consistent with theoretical estimates, were not solely proportional to the differences in loop length. These findings suggest that in addition to the length of the chain, the polypeptide sequence may influence the energetic consequences of conformational restrictions.

Circular permutation of T4 lysozyme.

To examine the relationship between polypeptide chain synthesis and protein folding, we have constructed a circularly permuted variant of phage T4 lysozyme. The permuted protein begins at residue 37 of the wild-type sequence and ends at residue 36. The normal chain termini are joined by a six-residue linker, Ser-Gly4-Ala. The permuted lysozyme folds efficiently and cleaves bacterial cell walls with normal specific activity. As judged by circular dichroism, UV absorbance, fluorescence, and nuclear magnetic resonance spectroscopy, the permutation causes little change in the structure of the protein. Reversible denaturation experiments show that the permutation reduces the stability of T4 lysozyme only 0.8-1.1 kcal/mol. These results demonstrate that a protein with two domains can be permuted with little change in activity, structure, and stability. The order of chain synthesis, the sequential arrangement of secondary structures, and the position of chain termini with respect to domain boundaries do not determine the protein fold.