The sulfhydryl groups of actin

Effect of self-association on the structural organization of partially folded proteins: inactivated actin

The propensity to associate or aggregate is one of the characteristic properties of many nonnative proteins. The aggregation of proteins is responsible for a number of human diseases and is a significant problem in biotechnology. Despite this, little is currently known about the effect of self-association on the structural properties and conformational stability of partially folded protein molecules. G-actin is shown to form equilibrium unfolding intermediate in the vicinity of 1.5 M guanidinium chloride (GdmCl). Refolding from the GdmCl unfolded state is terminated at the stage of formation of the same intermediate state. An analogous form, known as inactivated actin, can be obtained by heat treatment, or at moderate urea concentration, or by the release of Ca(2+). In all cases actin forms specific associates comprising partially folded protein molecules. The structural properties and conformational stability of inactivated actin were studied over a wide range of protein concentrations, and it was established that the process of self-association is rather specific. We have also shown that inactivated actin, being denatured, is characterized by a relatively rigid microenvironment of aromatic residues and exhibits a considerable limitation in the internal mobility of tryptophans. This means that specific self-association can play an important structure-forming role for the partially folded protein molecules.

Effect of myofibrillar muscle proteins on the in vitro bioavailability of non-haem iron

It is well established that the bioavailability of non-haem iron from foods is enhanced by the presence of meat. However, the nature of the promoter in meat has not yet been characterised. The present study was designed to compare the effects of the myofibrillar protein fractions on the bioavailability of non-haem iron in an attempt to identify the 'meat factor'. Rabbit skeletal muscle was fractionated and whole muscle, myofibrillar protein, myosin and actin were isolated. Myosin was subjected to selective proteolysis with chymotrypsin and the heavy meromyosin, light meromyosin, rod region and head region were prepared. Protein fractions (1 g) were incorporated into 100 g semi-synthetic liquid meal and the in vitro dialysability of iron was determined. Egg albumin was used as a reference protein. When compared with egg albumin, all protein fractions significantly enhanced iron dialysability, except for light meromyosin which was inhibitory. Myosin had a greater enhancing effect than actin and, within myosin, the enhancing effect was greatest for the heavy meromyosin fraction. The enhancement appeared to coincide with the known distribution of cysteine residues in the myofibrillar proteins. The presence of the sulphydryl blocking agent, N-ethylmaleimide (NEM), in meals containing myosin reduced iron dialysability in a dose-related manner, but NEM had only a small effect in meals containing actin. Meanwhile, incorporation of cysteine into meals containing actin increased iron dialysability. The present results suggest that the enhancement of non-haem iron dialysability by meat is associated with myosin, in particular, with the heavy meromyosin region. Peptide fractions rich in cysteine residues, probably constitutes the 'meat factor'.

Spectroscopic and functional characterization of an environmentally sensitive fluorescent actin conjugate

Rabbit skeletal muscle F-actin has been selectively labeled at a cysteine residue with the environmentally sensitive fluorophore 6-acryloyl-2-(dimethylamino)naphthalene. The fluorescent actin conjugate behaves similarly to native actin with respect to the polymerization kinetics, critical monomer concentration, and ability to form F-actin paracrystals. Upon polymerization to F-actin, the absorption of the actin conjugate is red-shifted, whereas the fluorescence emission is blue-shifted 740 wavenumbers and is accompanied by a decrease in the fluorescence bandwidth of 470 wavenumbers. These large shifts in the spectral properties of 6-propionyl-2-(dimethylamino)naphthalene (Prodan) in actin provide a simple method for obtaining a spectral discrimination between the G- and F-actin populations during the polymerization reaction. Steady-state fluorescence techniques were used to study the environment of the fluorophore in the monomeric and polymeric forms of actin. Fluorescence emission spectral analysis and quenching and polarization studies of G-actin-Prodan indicated that the fluorophore lies immobile on the protein surface but with one of its faces in full contact with the solvent. In F-actin, the fluorophore has a limited exposure to the solvent and is located in a dielectric environment similar to those seen for Prodan in polar, aprotic solvents or buried within a protein matrix [Macgregor, R. B., Jr., & Weber, G. (1986) Nature (London) 318, 70-73]. Additionally, our results demonstrate that the Prodan molecule conjugated to F-actin is completely immobile during its fluorescence lifetime, exhibits an increase in the resonance energy transfer (RET) from tryptophan residues compared to that observed in G-actin, and shows evidence of homologous RET within the polymer.

Fluorescence energy transfer between probes on actin and probes on myosin

The structural relationship between F-actin filaments and the biologically active fragments of myosin (either as myosin subfragment-1 or heavy meromyosin) has been investigated using the technique of fluorescence energy transfer. Donor and acceptor probes were used to obtain the following inter- and intramolecular distances. Energy transfer was measured: (1) from the SH1 groups of the myosin 'heads' to the nucleotide sites of F-actin (in the absence of free nucleotide); (2) from the SH1 groups of myosin to multiple probes on the surface of the actin filament; (3) from the nucleotide-binding sites of F-actin to the ATPase sites of myosin; (4) from the ATPase sites of myosin to the nucleotide-binding sites of F-actin; (5) from the SH1 sites of myosin to the nucleotide-binding sites of F-actin; and (6) from the Cys-373 residues of F-actin to the nucleotide binding sites of F-actin. We observed very little energy transfer between the probes on actin and the probes on myosin (10% or less) and we observed a large transfer between the actin Cys-373 and the actin nucleotide. These data strongly suggest that both the SH1 moiety and the ATPase site of myosin are located more than 6 nm from the actin sites. When these distances are combined with similar measurements by other authors and inserted into the most recent three-dimensional reconstruction of electron micrographs of the acto-subfragment-1 complex, it is apparent that the SH1 and the ATPase sites on myosin are not located adjacent to actin and are most probably located in the half of the myosin head that is distal from actin in the actomyosin complex.

Studies on the antigenic sites of actin: a comparative study of the immunogenic crossreactivity of invertebrate actins

The structural homologies of invertebrate actins with cytoplasmic vertebrate actins have recently been substantiated by comparative sequence analyses. This suggests that cytoplasmic actin is the ancestral precursor of smooth and striated muscle actin in vertebrates. We have raised antibodies in rabbits against a number of invertebrate muscle actins and have characterized the antisera by means of the highly sensitive ELISA method, which allows quantitation of nanomolar amounts of actin. Despite the fact that the invertebrate actins examined are very similar in primary structure, our results indicate that antibodies raised against them clearly distinguish between only a few amino-acid substitutions, and that the immunoreactivities quantitatively reflect the genetic divergence of this ubiquitous conservative protein. Examination of several proteolytic fragments of scallop actin for immunoreactivity with the homologous antiserum suggests that the major antigenic sites of actin are located within the amino terminal region of the molecule, while a carboxy terminal fragment comprising residues 69-372 exhibits very weak crossreactivity. Immunoadsorption experiments further indicate that species-specific antibodies are directed to antigenic determinants in the N-terminal region. This finding is supported by an examination of the effects of chemical modifications to Tyr, His, Arg, and Cys residues on the immunoreactivity of actin. Interaction with DNAase I markedly decreases the immunoreactivity of actin. This is consonant with the finding that the amino terminal peptide comprising residues 1-207 inhibits DNAase I, whilst a tryptic fragment fails to bind to the enzyme. The interaction is abolished by EDTA and the removal of the tightly bound cation is accompanied by a conformational change, shown by shifts in circular dichroic spectra. The possible involvement of the amino terminal peptide of actin in cation binding is discussed.

Change of reactivity of lysine residues upon actin polymerization

The reactivity of lysine residues of actin was measured by a surface labeling method--limited reductive methylation. After labeling, actin was subjected to CNBr and enzymatic cleavage, and all lysines were obtained either singly in a peptide or as a free residue. The specific activity of each lysine was taken as the measure of its reactivity. In actin denatured in 8 M urea, the reactivity of each lysine residue is approximately equal whereas those in G-actin fall into three categories: Lys-61 and Lys-113 are the most reactive ones; Lys-18, -213, -215, -314, and -358 are hardly reactive; the remainder, including Lys-50, -68, -84, -118, -191, -237, -283, -290, -325, -327, -335, and -372, are moderately reactive. The least reactive ones are probably buried in the native G-actin and all the others are most likely on the surface. Upon actin polymerization the reactivities of Lys-61, -68, -113, and -283 are significantly reduced while that of Lys-335 is strikingly enhanced. The decrease in reactivity could be readily explained if these residues were located in the monomer-monomer contact area although a polymerization-induced conformational change cannot be excluded. Such a conformational change may be invoked to explain the increase in the reactivity of Lys-335. Alternatively, the latter may be interacting with the bound ATP of G-actin, and the increased reactivity might be directly attributable to the loss of gamma-P for ATP accompanying polymerization.

Study of actin and its interactions with heavy meromyosin and the regulatory proteins by the pulse fluorimetry in polarized light of a fluorescent probe attached to an actin cysteine

The decay of anisotropy of the N-iodoacetyl-N'-(5-sulfo-1-naphthyl)-ethylenediamine fluorescence attached to cysteine-373 of actin can be characterized by two correlation times theta1 and theta2. theta1 has a value of several nanoseconds and is thought to represent some local protein motion. theta2 is of the order of several hundreds of nanoseconds. Its value increases with actin concentration. It represents an average of the G and F actin correlation times. When actin interacts with heavy meromyosin, theta2 increases and becomes infinite at a molar ratio of one heavy meromyosin molecule per four actin protomers. It is concluded that a definite complex is then formed between F actin and heavy meromyosin. In the same time, G actin concentration becomes equal to zero. Finally, when F actin forms a complex with the regulatory proteins tropomyosin and troponin, the value of theta2 is greater in the absence than in the presence of Ca2+. This result indicates that micromolar concentrations of Ca2+ induces a conformation change of the complex of F actin with the regulatory proteins.

Fluorescence energy transfer between epsilon-ATP at the nucleotide binding site and N-(4-dimethylamino-3,5-dinitrophenyl)-maleimide at Cys-373 of G-actin

The method of fluorescence energy transfer is used to measure the distance from the nucleotide binding site to Cys-373 of G-actin. The fluorescent ATP analogue 1-N6-ethenoadenosine 5'-triphosphate was used as donor and N-(4-dimethylamino-3,5-dinitrophenyl)-maleimide was used as acceptor. From the measurements of the efficiency of fluorescence energy transfer by both static and time resolved fluorometries, the distance between nucleotide binding site and Cys-373 residue of G-actin was calculated to be about 30 A.

Actin--membrane interactions: association of G-actin with the red cell membrane

Chemically tritiated actin from rabbit skeletal muscle was used to investigate the association of G-actin with the red cell membrane. The tritiated actin was shown to be identical to unmodified actin in its ability to polymerize and to activate heavy meromyosin ATPase. Using sealed and unsealed red cell ghosts we have shown that G-actin binds to the cytoplasmic but not the extracellular membrane surface of ghosts. Inside-out vesicles which have been stripped of endogenous actin and spectrin by low-ionic-strength incubation bind little G-actin. However, when a crude spectrin extract containing primarily spectrin, actin, and band 4.1 is added back to stripped vesicles, subsequent binding of G-actin can be increased up to 40-fold. Further, this crude spectrin extract can compete for and abolish G-actin binding to unsealed ghosts. Actin binding to ghosts increases linearly with added G-actin and requires the presence of magnesium. In addition, actin binding is inhibited by cytochalasin B and DNAase I. Negative staining reveals an abundance of actin filaments formed when G-actin is added to reconstituted inside-out vesicles but none when it is added to unreconstituted vesicles. These observations indicate that added G-actin binds to the red cell membrane via filament formation nucleated by some membrane component at the cytoplasmic surface.

Fluorescence study of N-(3-pyrene)maleimide conjugated to rabbit skeletal F-actin and plasmodium actin polymers

A fluorescent probe N-(3-pyrene)maleimide was conjugated to rabbit skeletal F-actin at the site of most reactive sulfhydryl group (Cys-373). Its fluorescence anisotropy decay showed a single correlation time of 560 ns at 25 degrees C, which is in a very good agreement with the correlation time of the dansyl-L-cysteine group conjugated to the same site of F-actin reported very recently [Wahl, Ph., Mihashi, K, and Auchet, J-C. (1975) FEBS Lett. 8, 164-167]. Actin from plasmodia of myxomycates, Physarum polycepharum, was also conjugated with N-(3-pyrene) maleimide and the fluorescence anisotropy was compared with rabbit skeletal F-actin using the classical steady excitation method. It was found that the internal mobility of the magnesium polymer of plasmodium actin is remarkably larger than both plasmodium F-actin and rabbit skeletal F-actin.

Nanosecond pulse fluorometry in polarized light of dansyl-L-cysteine linked to a unique SH group of F-actin; the influence of regulatory proteins and myosin moiety

The order of magnitude of the correlation time, which characterizes the dansyl cysteine residue linked to F-actin is ten times greater than the correlation time of the G-actin monomer [1]. Still it is much smaller than the correlation times of the F-actin polymer as a whole. The dansyl chromophore reveals that the C terminal end of the actin peptide chain, is mobile. As Ebashi and his co-workers have shown (13), Ca2+ triggers muscular contraction by acting on F-actin through the mediation of the regulatory proteins troponin and tropomyosin. By using spin label technique, Tonomura et al. [14] found that Ca2+ induces a conformational change on the troponin, tropomyosin actin complex. The quasi elastic scattering of laser light measurement of Fujime and Ishiwata [15] showed that troponin-tropomyosin F-actin has a rotational correlation time in the millisecond range which characterizes the flexibility of this complex; Ca2+ induces an increase of this flexibility. The present pulse fluorometry study shows an increase of mobility of the fluorescent probe induced by Ca2+. It seems difficult to correlate the results of the two kinds of measurements as long as we do not know the exact nature of the fluorescent kinetics unit.

Conformational change and cooperativity in actin filaments free of tropomyosin

The decrease in amplitude of the electron spin resonance spectrum of the cysteine-bound spin-label, 3-(maleimidomethyl)-2,2,5,5-tetramethyl-1-pyrrolidinoxyl, brought about by the magnetic interaction with tightly bound manganous ion, was used as a probe of conformational change in actin on binding myosin. The magnitude of this "spin--spin" interaction first decreased then increased on increasing saturation of the actin filament with heavy meromyosin subfragment-1. That the "spin--spin" interaction occurred between spins of adjacent monomers was demonstrated by the observation that the change in magnitude of the "spin--spin" interaction was maintained on binding of heavy meromyosin subfragment-1 to copolymers in which actin monomers containing both manganous ion and spin label were diluted 7-fold with native actin monomers. These data provide evidence for a conformational change in actin on interacting with heavy meromyosin subfragment-1. Further, the fact that not only the magnitude but also the sense of the change in the "spin--spin" interaction is a function of increasing saturation with heavy meromyosin subfragment-1 indicates that the monomers of the actin filament are capable of cooperative interaction in the absence of tropomyosin.

Biological activity and the 3-methylhistidine content of actin and myosin

1. The 3-methylhistidine content of myosin varies according to muscle type. It is highest in myosin from white skeletal muscle and lower values are obtained from myosin of red skeletal and smooth muscle. 2. The 3-methylhistidine content of actin was similar in all of the types of muscle from which it was isolated. 3. The 3-methylhistidine of rabbit actin is localized in a single tryptic peptide that was readily modified during fractionation procedures. 4. Photo-oxidation studies indicated that the 3-methylhistidine residues are not essential for adeonsine triphosphatase and actin-combining activities of myosin. 5. During photooxidation G-actin lost completely the ability to polymerize to the F form before all the 3-methylhistidine was destroyed.

Chemical studies on the cysteine and terminal peptides in tryptic digests of actin

1. On exhaustive digestion of carboxymethylated actin in 6m-urea solutions with carboxypeptidase A, 1 mole of phenylalanine was liberated/43000g. of protein. At a lower urea concentration and in the absence of urea, carboxymethyl-cysteine (CMCys) was also liberated. 2. Three cysteine-containing peptides were identified by the study of peptide ;maps' of tryptic digests of actin treated with thiol reagents. 3. The three peptides, each containing one residue of CMCys, were isolated from tryptic digests of carboxymethylated actin by ion-exchange chromatography. 4. One of these peptides was possibly the N-terminal peptide and contained about 17-18 residues; another was CMCys-Asp-Ile-Asp-Ile-Arg; the other, CMCys-Phe, was the C-terminal tryptic peptide. 5. The chemical evidence suggests that the actin molecule consists of a single polypeptide chain of molecular weight about 44000.