Effect of lysine methylation and other ATPase modulators on the active site of myosin subfragment 1.

Many and diverse modifications of the myosin subfragment 1 (S-1) increase (modulate) its ATPase activity, including interaction of this particle with actin; a recent addition to these modifications is the extensive lysine modification of S-1 that seems prerequisite to crystallizing it for structure analysis. In this study we first established kinetically the ATPase modulations induced by various treatments of the myosin S-1 enzyme, and we also measured two properties of the S-1 active site--the affinity with which the site binds (a fluorescent analog of) the enzymatic nucleotide product and the access that a fluorescence quencher has to the bound ADP product--in an effort to get at the mechanism of modulation. Modulations achieved by substituting Ca2+ for the normal Mg2+ cocatalyst or by substituting Cl- for the normal carboxylate anion seem due to the product being held more loosely by the modulated enzyme. In other illustrative modulations (lysine methylation, or alkylation of Cys-707, or transition from neutral pH to pH 9.2) nucleotide product affinity and access to quencher do change, but not in a pattern explained simply by a lifting of product inhibition. Lysine methylation results in weaker binding of nucleotide product.

On how a myosin tryptophan may be perturbed.

A well-known indication that a nucleotide has bound to myosin is the enhancement of the fluorescence of a specific tryptophan in the "subfragment 1" segment of the protein. Empirically the effect has been enormously useful in myosin enzymology. But beyond an early suggestion that it arises from a purine-tryptophan charge-transfer complex, the mechanism of the effect has not been considered. Here we consider the alternative that it arises from an ionizable group (either another residue or the phosphate of the nucleotide) whose proximity to the tryptophan is altered by substrate binding. We study this possibility by studying the interaction of an ionizable residue and tryptophan when both are incorporated in a diketopiperazine structure. The geometry of the situation is inferred from molecular mechanics simulations. Unexpectedly, the best explanation seems to be that the field of the imposed charge, acting across space, affects events in the excited state of the indole.

The region in myosin S-1 that may be involved in energy transduction.

Newly-reported structural information about certain proximities between points on bound nucleotide and points on the heavy chain of myosin S-1 are incorporated into a previously-reported [Botts, J. Thomason, J.F. & Morales, M.F. Proc. Nat. Acad. Sci. USA, 86, 2204-2208 (1989)] structure of S-1. The resulting, enhanced structure is then used to identify some functionalities (e.g., the ATP-perturbable tryptophans), and to explain certain observations (e.g., some concerning the role of bound Mg2+ in the spectral response of TNBS-labelled Lys-83, and some concerning the response of the S-1 CD signal to nucleotide binding and to temperature change). These considerations lead to the suggestion that a strand of the 50 kDa "domain" (residues 510 to 540), and a strand of the 20 kDa 'domain' (residues 697-719) are involved in transmitting the effects of nucleotide binding and hydrolysis to the loop (constituted from the same "domain") that reaches a major (S-1)-actin interface.

A search for protein structural changes accompanying the contractile interaction.

It appears that small movements (detected hitherto only by fluorescence resonance energy transfer measurements and crosslinking studies) in a region of the myosin S-1 particle may mediate chemomechanical energy transduction in the contractile system. Here we find under conditions of high precision at 10 degrees C and 20 degrees C that ATP binding to S-1 causes small (0.4%) changes in CD signal, delta epsilon 222, as do temperature changes in the regime below 16 degrees C. ATP binding perturbs tryptophan residues that we now think are in the mobile region, and we find here that temperature affects tryptophan fluorescence in much the same way that it affects the CD signal, so we believe that the CD signal reports transduction-related movements in S-1. If S-1 is exposed to the range 16-30 degrees C, CD signal falls with temperature; ATP counteracts this fall. Analysis of vacuum-UV CD spectra yields 42% alpha-helix, 9% antiparallel beta-sheet, 7% parallel beta-sheet, 14% beta-turns, and 29% other structures.

Cross-helix separation of tropomyosin molecules in acto-tropomyosin as determined by neutron scattering.

The cross-helix separation of Tm molecules in acto-tropomyosin has been determined using neutron scattering. Deuterated Dictyostelium discoideum actin was density matched in a 93% D2O buffer so that effectively only the protonated tropomyosin was "visible" to neutrons. Analysis of the solution scattering pattern in the region of the first oscillation yielded a value for the cross-helix separation of 7.9 +/- 0.3 nm. The implications of this value for the mechanism of the regulation of muscle contraction are discussed in light of recent results by others.

Photoactive retinal pigments in haloalkaliphilic bacteria.

Light-induced fast transient absorbance changes were detected by time-resolved spectroscopy in 38 of 51 haloalkaliphilic isolates from alkaline salt lakes in Kenya and the Wadi Natrun in Egypt. They indicate the presence of two retinal pigments, Pf and Ps, which undergo cyclic photoreactions with half-times of 2 ms and 500 ms respectively. Pf absorbs maximally near 580 nm and Ps near 500 nm. The pigments differ in their sensitivity to hydroxylamine and detergent bleaching and the photoreactions of Pf are strongly dependent on chloride concentration. Of the 38 pigment-containing strains, 29 possess both Pf and Ps, 9 possess only Ps. Inhibition of retinal synthesis with nicotine blocks pigment formation and addition of retinal restores it. Hydroxylamine-bleached pigments can be reconstituted with retinal or retinal analogues. Their similarity to the retinal pigments of Halobacterium halobium strongly suggests that they are also rhodopsin-like retinyledene proteins. Pf in all properties tested is almost identical to halorhodopsin, the light-driven chloride pump of H. halobium, and may serve the same function in the haloalkaliphiles. Ps has photocycle kinetics similar to sensory rhodopsin and a far-blue-shifted long-lived photocycle intermediate, but its ground state absorption maximum is near 500 nm instead of 587 nm. We have not found a bacteriorhodopsin-like pigment in the haloalkaliphiles.

Coupling between the bacteriorhodopsin photocycle and the protonmotive force in Halobacterium halobium cell envelope vesicles. III. Time-resolved increase in the transmembrane electric potential and modeling of the associated ion fluxes.

Bacteriorhodopsin functions as an electrogenic, light-driven proton pump in Halobacterium halobium. In cell envelope vesicles, its photocycle kinetics can be correlated with membrane potential. The initial decay rate of the M photocycle intermediate(s) decreases with increasing membrane potential, allowing the construction of a calibration curve. The laser (592.5 nm) was flashed at various time delays following the start of background illumination (592 +/- 25 nm) and transient absorbance changes at 418 nm monitored in cell envelope vesicles. The vesicles were loaded with and suspended in either 3 M NaCl or 3 M KCl buffered with 50 mM HEPES at pH 7.5 and the membrane permeability to protons modified by pretreatment with N,N'-dicyclohexylcarbodiimide. In each case the membrane potential rose with a halftime of approximately 75 ms. The steady-state potential achieved depends on the cation present and the proton permeability of the membrane, i.e., higher potentials are developed in dicyclohexylcarbodiimide treated vesicles or in NaCl media as compared with KCl media. The results are modeled using an irreversible thermodynamics formulation, which assumes a constant driving reaction affinity (Ach) and a variable reaction rate (Jr) for the proton-pumping cycle of bacteriorhodopsin. Additionally, the model includes a voltage-gated, electrogenic Na+/H+ antiporter that is active when vesicles are suspended in NaCl. Estimates for the linear phenomenological coefficients describing the overall proton-pumping cycle (Lr = 3.5 X 10(-11)/mol2/J X g X s), passive cation permeabilities (LHu = 2 X 10(-10), LKu = 2.2 X 10(-10), LNau = 1 X 10(-11)), and the Na+/H+ exchange via the antiporter (Lex = 5 X 10(-11)) have been obtained.