Twentieth Annual Meeting February 24–27, 1976 Olympic Hotel, Seattle, Washington

1976 Biophysical Journal  
The MgATPase activity of rabbit skeletal and bovine cardiac myosin is activated by micromolar concentrations of calcium. This activation is not abolished nor shifted to higher concentrations of calcium when the free magnesium concentration is increased to 1 iM. To date it has not been possible to link this response directly to the L2 light chain. In the presence of pure F-actin the MgATPase activity of both myosins is activated maximally at very low concentrations of free magnesium. Increasing
more » ... he free calcium concentration from 10-7 to 5 x 10-5 M results in a 20 to 30 % inhibition at 2 x 10-5 M (see also R.D. Bremel and A. Weber, Biochimica et Biophysica Acta 376 (1975) 366). Increasing the magnesium concentration to 1 mM inhibits the actin-activated MgATPase activity and the specific response at 20 pM calcium. DTNB-treated skeletal myosin containing 1 mole L2/mole and various subfragment 1 populations containing 0.5 moles L2/mole (papain digestion) or 0.0 moles L2/mole (chymotryptic digestion with and without an (NH4)2S04 precipitation step) were analyzed for the above phenomena. The magnesium inhibition of the actin-activated MgATPase requires the full complement of the L2 light chain. The calcium-dependent inhibition at 2 x 10-5 M free calcium which is observed at low concentrations of free magnesium is reduced substantially in all L2 deficient proteins. All preparations possessed comparable K+/EDTA-ATPase activities and were capable of forming a calcium-sensitive complex with actin in the presence of troponin and tropomyosin. (S.M.P. is a Fellow of the New Sedimentation velocity, analytical gel filtration and diffusion measurements were used to estimate the geometry of the light chains of Aequipecten irradians, Mercenaria mercenaria, Spisula solidissima, and Loligo pealei. The frictional coefficient measured by these techniques was compared to the calculated value for an equivalent hydrated sphere of the same molecular weight. The data of I. D. Kuntz and W.Kauzmann(Adv. Prot. Chem., 28, 239-345(1974) were used to estimate hydration from the amino acid composition. The frictional ratio was calculated and axial ratios, a:b:c, were estimated from both Perrin's equation and numerical solutions for the general ellipsoid of E. W. Small and I.Isenberg (personal communication) and gave a length ranging from 100-140A depending on the ellipsoidal model chosen. The regulatory light chain appears to be nearly as long as the Sl portion of myosin and possibly forms multiple interacting links with it. The asymmetric geometry may play an important role in the regulatory process. Circular dichroic spectra of the light chains were analyzed according to the method of N. Greenfield and G. D. Fasman (Biochem., 8, 4108 (1969) to estimate the amounts of alpha, beta, and random structure. The amount of alpha-helix varied from 19-47% and the amount of beta structure, from 8-30% depending on the species from which the light chain was obtained. All species had about 50% random structure. to rabbit skeletal muscle troponin-C induces a large conformational chane as detected by proton magnetic resonance spectroscopy. Successive additions of Ca to the metal-free protein leads to four distinguishable transitions or sets of transitions which may correspond to the four binding sites of troponin-C. Spectra of the native form of troponin-C are indicative of a tightly folded protein conformation with little conformational change under different pH conditions. In contrast an overall conformational change of the Ca2+_free protein to a more2compact form occurs with increasing pH.2 This conformation differs from that of the Ca +-bound species. The pH titration of the Ca -free form appears to involve at least a two proton cooperativity in that the effective titration curve for histidine -125 shows a transition which is not that for one proton binding at a single site. Several spectral assignments have been made to help the identification of the four calcium binding sites and thereby elucidate the nature of the conformational changes induced by Ca+2. 2+ Previous studies of the spectral effects of Ca binding to troponin C (TnC) have employed Ca-EDTA or Ca-EGTA buffer systems. To determine stoichiometry, we have performed direct Ca2+ and Tb3+ titrations on metal-free TnC in the absence and presence of 6M urea. Tiirations of native TnC show that fluorescence and CD changes require two moles of Ca2+ or Tb + per mole TnC. The fluorescence of bound Tb3+ (Xe-280 nm) is 300-fold higher than the free ion due to energy transfer from Tyr. CD and difference absorption spectral changes produced by Tb3+ are similar to those reported for Ca2+. CD changes at 222 nm and in the 260 nm region indicate increased Ca-helix and increased asymmetry of Phe respectively. Difference spectra suggest that Phe residues are shifted to a less polar milieu. Tb3+ binds more strongly than Ca2+ to the two high affinity sites from competitive binding studies. In 6M urea the spectral changes are complete by the binding of one mole of Ca2+ per mole of TnC, in contrast to the two required in the native state. Tb3+-induced changes in 6M urea, however, saturate at a ratio of 2:1, although the sites appear to have slightly different affinities. Tb3+ preferentially displaces the single bound Ca +. Studies of Tb3+ binding to the cyanogen bromide peptide, CB9, which includes one Ca2+-binding site containing Tyr 109, show fluorescence and CD changes qualitatively similar to whole TnC at a mole ratio of 1. These results present evidence that this site is responsible for the reported spectral changes. Recent X-ray diffraction studies of striated muscle by several laboratories have provided evidence for an appreciable spatial displacement of tropomyosin in response to a signal from troponin during calcium ion (Ca++) activation. In view of the distances involved a reasonable model for the transmition of such a signal would be a delocalization of a Ca++ induced conformation change in one or both of the regulatory proteins. However, previous circular dichroic studies of cardiac regulatory proteins in this laboratory have failed to find evidence for such a conformation change for the in vitro interaction of Ca++ and troponin (Biophys. J. 15, 36a(1975). To determine if perhaps tropomyosin was essential to the induction of such a conformation change by Ca++ similar studies have now been done with the troponin-tropomyosin complex. We find that (1) the complexing of the two regulatory proteins does not result in conformation changes (as determined by circular dichroism) of either protein and (2) the conformational response of the complex to Ca++ interaction is very similar to the one previously observed for the troponin alone. In view of these results we suggest the following possibilities: (1) actin and the ordered structure of the thin filament may be essential to the induction and propagation of the proposed conformation change, (2) the Ca++ regulatory mechanisms of cardiac and striated muscle may be different, and (3) a delocalized conformation model for the propagation of Ca" induced signal may not be valid (Supported by Central Ohio Heart Association grant #73-41). Ca2+-binding to two sites of TnC in the native state, or to one in 6M urea, produces the same increase in a helical regions, viz. &Ee]222 = 5,200 deg cm2 dmole-I mean residue weight. Phe sidechain circular dichroic (CD) and difference spectra, and the change in reactivity of the single cys-98 residue suggest that one of the segments undergoing such coil-helix transition involves residues 94-103 of the amino acid sequence (Nagy et al., Biophys. J. 15, 35a, 1975; Potter, ibid, 36a; Collins et al., FEBS Lett. 36, 268, 1973). On Ca2+ binding a chromophoric marker, 3-nitro-4-hydroxyacetophenone, attached to cys-98 shows a red shift (indicating transfer to a more hydrophobic environment) and becomes optically active, in accord with the above suggestion. A cyano en bromide fragment, CB-9 (Collins et al., l.c.) including residues 84-134, contains Ca binding site III with 3 phe, 1 tyr and the cys residue. It binds Ca2+ with K-105M-1 as judged by the Ca2+ dependence of the tyr fluorescence. On Ca2+ binding the cys-98 reactivity with Nbs decreases. Ca2+ binding produces L[1]222 = 4800 suggesting that about 13% of the fragment of CB-9 (about 7 residues) is involved. They include phe (red shift in difference spectrum, increase in phe CD bands), and tyr (fluorescence increases). The chromophoric marker attached to cys-98 in CB-9 also shows a red shift and becomes oTtically active on Ca2+-binding. Three conclusions emerged from these studies: a) the Ca2 -binding site is preserved in the fragmented CB-9; 2) shifts in the environment of the residues (chemical reactivity, absorption, and CD spectra) occur within a limited domain; 3) conformational changes in one region affect other regions of TnC. (Supported by grants from NIH, NSF and MDAA) 71a Ca2+-activated isometric tension was recorded from functionally skinned rabbit skeletal and cardiac and scallop muscle fibers (sarcolemma mechanically disrupted) which were immersed into test solutions containing 1 mM Mg2+, 1 x 10-7M H+, 70 mM K , 7 mM EGTA, 2 mM MgATP, 1 x 10-6 -2.5 x 10-4 M Ca2+, 5 x 10-5 M of appropriate rabbit protein (troponin, TNI, TNB, TNC, or TNT), imidazole propionate (used to adjust the ionic strength to .15), and propionate as the major anion. Troponin, TNC and TNT do not affect Ca2+-activated tension. TNI and TNB will reversibly inhibit the Ca2+-activated tension in white, red and cardiac muscle. The time course of phosphorylated vs. non-phosphorylated TNI inhibition is exponential and ten times faster in white than in red or cardiac rabbit muscle. When inhibition is complete, it cannot be reversed unless the fiber is soaked in a solution containing TNC. None of the troponin subunits affect the Ca2+-activated tension relationship in mechanically skinned fibers of the scallop which is consistent with the known myosin Ca2+ control of these muscles. In contrast, rabbit troponin increases the Ca2+-sensitivity of the scallop fibers ten-fold such that the relationship between % Ca2+-activated tension and % Ca2+ binding to troponin is the same for identical Ca2+ test solutions. These data indicate that the troponin Ca2+ control of actomyosin can override the myosin Ca2+ control of the scallop. Papillary muscles and trabeculae from rat ventricle have been "chemically skinned" by soaking the tissues in 1OmM EGTA overnight. Such preparations reproducibly begin to develop tension at pCa of 7.1 and achieve maximum tension at pCa of 5. 0 in the presence of a standard contraction solution (140mM KCI, 4mM MgATP). The supernatant from a suspension of rat heart ventricle which has been homogenized in relaxing solution with 3mM EGTA and 4mM MgATP contains a cardiotonic substance. The activity remains in the supernatant even after centrifugation at 89, 000 g for 1 hr. Skinned fibers suspended in the supernatant buffered at a pCa of 9. 0 produce as niuch as 30-40% of the tension produced in pCa of 5. 0 and relax almost completely when the supernatant is replaced by a standard relaxing solution. The active component of the homogenate can be destroyed by heat. It passes through dialysis tubing and retains considerable activity even when the homogenized tissue has been diluted 2500X as long as ATP is present but not in its absence. These observations are consistent with enzymatic synthesis from ATP of a compound which activates the contractile proteins at high pCa. Single skinned cardiac cells were obtained by homogenization and micro-dissection of rat ventricle. The free [Ca2+] was buffered with 4.0mM total EGTA and the pH with Imidazole. Schwarzenbach' s constant was used for Ca-BGTA. Curves of tension as a function of the pCa were obtained at pH 6.6, 7.0 and 7.4. The pMg was maintained at 3.5 and the pMgATP at 2.5. Temperature was 220C. Increase of pH did not modify the maximum tension but produced a shift to the left of the tension-pCa curve which started at a higher pCa for a higher pH. The pCa necessary for 50% of maximum tension was pCa 5.70 at pH 6.6, pCa 6.05 at pH 7.0 and pCa 6.35 at pH 7.4. Qualitatively similar results were obtained on skinned cells of frog cardiac muscle (ventricle) and frog skeletal muscle (semi-tendinosus). However, the effects of the same variations of pH were smaller in skeletal muscle as compared to cardiac muscle. Phasic contractions by Ca2+-triggered release of Ca2+ from the sarcoplasmic reticulum (SR) were induced in skinned cardiac cells from the rat ventricle by a slight decrease of pCa from 7.70 to 7.40 in the presence of 0.05mM total EGTA. The increase of amplitude of the phasic contractions when pH was increased from 6.6 to 7.4 was accounted for by the effect of pH on the sensitivity of the myofilaments to Ca2+. Contractions were induced by caffeine or the ionophore A 23187 at a constant pH of 7.0 after the contractions triggered by Ca2+ at variable pH. The amplitude and duration of these contractions were not significantly modified by variations of the pH at which the Ca2+-triggered release of Ca2+ was obtained. A sudden increase of pH from 6.6 to 7.4 at a constant pCa of 7.70 with 0.50mM total EGTA did not induce a phasic contraction. In conclusion the increase of pH from 6.6 to 7.4 neither induces a Ca24+ -release from the SR nor modifies the Ca2+-triggered release. Single frog semitendinosus muscle fibers were skinned by the method of Natori, 1954. Tension was measured with a photodiode transducer similar to that used by Hellam and Podolsky, 1969. Ca2+ concentration was buffered with 7mM EGTA, and pH and ionic strength (.17) were controlled with MES or imidazole. All solutions contained 2mM MgATP2-, 90mM K+, 1mM Mg2+, 15mM creatine phosphate, 15 units per ml c eatine phosphokinase; and the major anion was propionate. For each fiber the maximum Ca +-activated tension was measured at different pHs and normalized to the maximum Ca2+-activated tension developed in the same fiber at pH 7.0. Normalized maximum tensions (NMT) for each pH were averaged. At a given pH the ratio of submaximal Ca2+-activated tension to maximal Ca2+-activated tension was used as a measure of the percent Ca2+ activation (%A). All ratios at a specified pCa and pH were averaged. The pCa corresponding to 50% activation and mean NMT values are shown in the table below. All changes in NMT or %A were reversible. The shift in pCa required for 50% activation at each pH can be explained by a single competitive site for CaZ+ and H+ with a pKm for Ca2+ equal to 5.73 M and pH 7.5 7.0 6.5 6.0 5.5 5.0 a pKi for H+ equal to 7.03 M. The decline pCa for 50% A 5.6 5.4 5.0 4.6 4.2
doi:10.1016/s0006-3495(76)85676-7 fatcat:q2qfvlueibahpdvcbyuwef3ygm