Conformational changes in the unique loops bordering the ATP binding cleft of skeletal muscle myosin mediate energy transduction

Myosin has three highly-conserved, unique loops [B (320-327), M (677-689), and N (127-136)] at the entrance of the ATP binding cleft, and we previously showed that the effects of actin are mediated by a conformational change in loop M [Maruta and Homma (1998) J. Biochem. 124, 528-533]. In the present study, loops M and N were photolabeled respectively with fluorescent probes Mant-8-N(3)-ADP and Mant-2-N(3)-ADP in order to study conformational changes in the loops related to energy transduction. The effect of actin on the conformation of loop N was examined by analyzing fluorescence polarization and acrylamide quenching; the results were then compared with those previously reported for loop M. In contrast to loop M, the fluorescence polarization and the value of K(sv) of the Mant-groups crosslinked to loop N were slightly affected by actin binding. To study conformational changes in loops M and N during the ATPase cycle, FRET was analyzed using TNP-ADP.BeFn and TNP-ADP. AlF(4)(-) as FRET acceptors of Mant fluorescence. The resultant estimated distances between loop M and the active site differed for the Mant-S1.TNP-ADP.BeFn and Mant-S1.TNP-ADP.AlF(4)(-) complexes, whereas the distances between loop N and the active site differed slightly. These findings indicate that the conformation of loop M changes during the ATPase cycle, suggesting that Loop M acts as a signal transducer mediating communication between the ATP- and actin-binding sites. Loop N, by contrast, is not significantly flexible.

Solution structure of myosin-ADP-MgFn ternary complex by fluorescent probes and small-angle synchrotron X-ray scattering

In the presence of excess amounts of fluorine, a physiological divalent cation, magnesium (Mg(2+)), forms a novel phosphate analogue, magnesium fluoride (MgFn). Park et al. [Biochim. Biophys. Acta 1430, 127-140 (1999)] previously demonstrated that MgADP. MgFn forms a complex with myosin subfragment-1 (S-1), and the S-1.ADP. MgFn ternary complex mimics a transient state in the activity cycle of ATPase. In the present study, localized conformations in the regions of highly reactive cysteine and lysine residues, Cys 707 (SH1), Cys 697 (SH2), and Lys 83 (RLR), which change their conformations markedly during ATP hydrolysis, were studied using fluorescent probes and chemical modification. The global shape of the complex was also studied using small angle X-ray solution scattering and compared it with other previously reported myosin.ADP. fluorometal ternary complexes. The results suggest that the overall conformation and localized functional regions of the complex are quite similar to those in the presence of ATP, indicating that the complex mimics the M(**).ADP.P(i) steady state.

Analysis of conformational changes at the unique loop adjacent to the ATP binding site of smooth muscle myosin using a fluorescent probe

Recent crystallographic studies have shown that smooth muscle myosin has three highly conserved unique loops, loop B (320-327), loop M (687-699), and loop N (125-134), similar to other myosins, skeletal muscle and dictyostelium myosins. We previously demonstrated that the effect of actin is mediated by a conformational change in one of the loops, loop M comprising amino acids 677 to 689 of skeletal muscle myosin [Maruta and Homma (1998) J. Biochem. 124, 528-533]. In the present study, in order to clarify the role of these smooth muscle myosin loops in energy transduction, we specifically labeled the loops with a fluorescent photoreactive ADP analogue, 3'-O-(N-methylanthraniloyl)-8-azido-ADP (Mant-8-N(3)-ADP), and then measured the fluorescent polarization. When Mant-8-N(3)-ADP was trapped by aluminium fluoride or vanadate into the ATPase site, Mant-8-N(3)-ADP was covalently incorporated into loop N (125-134). In contrast, Mant-8-N(3)-ADP trapped by beryllium fluoride was covalently incorporated into both loop M (687-699) and loop N (125-134) at an almost equimolar ratio. Actin binding to smooth muscle myosin S1 (SMO-S1) labeled at only loop N (125-134) increased the polarization due to the viscosity of actin. In contrast, S1 labeled at both loops N and M showed a much smaller increase in polarization. Our results indicate that the probe at loop M (687-699) of smooth muscle myosin moved to a less hindered region, suggesting that actin binding induces conformational changes at loop M (687-699) similar to those of the corresponding loop (677-689) in skeletal muscle myosin, as previously demonstrated in our laboratory.

Formation of the myosin.ADP.gallium fluoride complex and its solution structure by small-angle synchrotron X-ray scattering

In the presence of MgADP, a novel phosphate analogue of gallium fluoride (GaFn) forms a ternary complex with the myosin subfragment-1 (S-1), in the same way that has been previously reported with aluminum fluoride (AlF4-), beryllium fluoride (BeFn), scandium fluoride (ScFn), and vanadate (Vi), and this complex formation may mimic different states along the ATPase kinetic pathway. This novel complex has been characterized and compared with other complexes to ascertain whether it forms a transition-state analogue of myosin ATPase. The complex formed quickly, although several times slower than the BeFn complex. The half-life of the myosin.ADP.GaFn complex was about 50 h at 4 degreesC. The formation of the myosin.ADP.GaFn complex was accompanied by an increase in tryptophane fluorescence, similar to that observed upon the addition of ATP, but slightly lower than that of the M**.ADP.Pi complex. Upon addition of GaFn to acto-myosin.ADP, acto-myosin did not dissociate, and the S-1.ADP.GaFn complex was scarcely decomposed by actin, like the AlF4- and ScFn complexes but unlike the BeFn and Vi complexes. The conformations at the localized region of SH1, SH2, and RLR, which are very accessible to the binding of ATP, were studied by fluorescent labeling and chemical modification, and the results suggested that these conformations are very similar to that of the M**.ADP.Pi state. Small-angle X-ray solution scattering showed that the radius of gyration value decreases by about 3 A when S-1 forms an S-1.ADP.GaFn complex, suggesting that the shape of the complex becomes compact or rounded in shape, similar to that in the presence of ATP or complexes with other phosphate analogues, and thus mimics the myosin**.ADP.Pi state closely. The overall results may indicate that the complex mimics a somewhat different transient state from that of other complexes but has a similar global conformation along the ATPase kinetic pathway.

ADP/vanadate mediated photocleavage of myosin light chain kinase at the autoinhibitory region

The vanadate (Vi)-mediated photocleavage reaction was used to study the interaction between the regulatory segment and the catalytic site of smooth muscle myosin light chain kinase (MLCK). When MLCK was irradiated with long-wave UV (366 nm) in the presence of ADP and Vi, kinase activity was substantially decreased, and the MLCK polypeptide of 130 kDa was cleaved into several smaller fragments with apparent molecular masses of 100, 70, 60, 32, and 28 kDa. Inhibition of kinase activity and photocleavage were both competitively antagonized by the addition of ATP. Inconsistency between the observed maximum levels of UV-induced inhibition of MLCK-mediated phosphorylation (80%) and photocleavage (15-20%) suggested that the photocleavage reaction proceeds as a two-step process. Monoclonal antibodies recognizing the C-terminus of MLCK labeled the 60- and 28-kDa fragments, indicating that MLCK was cleaved at two sites, at 28 and 60 kDa from the C-terminus, within what are believed to be the autoinhibitory region and the catalytic site, respectively. Moreover, Ca2+-calmodulin (Ca2+-CaM) protected against cleavage at the site at 28 kDa from the C-terminus. Analysis of the amino acid composition of the fragment revealed that the cleavage site at 28 kDa from C-terminus occurred at Lys 799 +/- 3 amino acid residues, which is in a region where the CaM-binding and pseudosubstrate regions overlap. These results suggest that the three-dimensional structure of MLCK brings the regulatory segment into direct contact with the ATP-binding site. Moreover, the binding of Ca2+-CaM displaces the regulatory segment away from the catalytic site.

A unique loop contributing to the structure of the ATP-binding cleft of skeletal muscle myosin communicates with the actin-binding site

Actin binding to skeletal muscle myosin subfragment-1 (S1) increases the dissociation rate of reaction products from the myosin ATPase site; conversely, ATP binding facilitates dissociation of complexed acto-S1. However, details of the molecular mechanism by which the ATP- and actin-binding sites communicate with each other is still obscure. We present evidence that the effect of actin is mediated by a conformational change in the loop containing amino acids from 677 to 689 [loop M (677-689)], a segment of the 20-kDa tryptic fragment that contributes to the structure of the ATP-binding cleft. Initially, a fluorescent ADP analogue, methylanthranyloyl-8-azido-ADP (Mant-8-N3-ADP), was covalently crosslinked to loop M (Mant-S1), perhaps at Lys 681. Actin-activated Mg2+-ATP hydrolysis by Mant-S1 was accelerated approximately 6 times over that by unmodified S1, suggesting that the ATPase site is not blocked by the ADP analogue crosslinked in the loop M (677-689). Nevertheless, analysis of Mant-group fluorescence polarization and acrylamide-induced quenching showed the crosslinked probe to be entrapped within the ATP-binding cleft at a location where Mant-group rotational mobility was hindered, and where it was relatively inaccessible to the solvent. Exposing Mant-S1 to Mg2+-ATP and/or actin elicited similar decreases in fluorescence polarization, indicating increased rotational mobility of the Mant-group and movement of crosslinked Mant-8-N3-ADP to a less hindered position. Stern-Volmer quench curves showed that Mant-8-N3-ADP was translocated to a site where it was more accessible to dissolved quencher, perhaps outside the ATP-binding cleft. Since actin does not bind to the ATPase site, actin-induced translocation of Mant-8-N3-ADP crosslinked to loop M (677-689) probably results from a conformational change in loop M (677-689). These results suggest that loop M acts as a signal transducer mediating communication between the ATP- and actin-binding sites.

Conformational changes at the highly reactive cystein and lysine regions of skeletal muscle myosin induced by formation of transition state analogues

Myosin forms stable ternary complexes with Mg2+-ADP and phosphate analogues of aluminum fluoride (AlF4-), beryllium fluoride (BeFn), and scandium fluoride (ScFn). These complexes are distinct from each other and may mimic different transient states in the ATPase cycle [Maruta et al. (1993) J. Biol. Chem. 268, 7093-7100]. Regions of skeletal muscle myosin containing the highly reactive residues Cys 707 (SH1), Cys 697 (SH2), and lysine 83 (RLR) dramatically alter their local conformation when myosin hydrolyzes ATP, and these changes may reflect formation of a series of transient intermediates during ATP hydrolysis. We used the fluorescent probes 4-fluoro-7-sulfamoylbezofurazan, 2-(4'-maleimidylanilino)naphthalene-6-sulfonic acid, and trinitrobenzene-sulfonate, which bind to SH1, SH2, and RLR, respectively, to examine differences in local conformations within myosin.ADP.phosphate analogue (BeFn, Vi, AlF4-, and ScFn) complexes. It was observed that the ternary complexes had SH1 conformations similar to those seen on S-1 in the presence of ATP. In contrast, local conformations in the SH2 and RLR regions of S-1.ADP.BeFn were different from those in corresponding regions of S-1.ADP.AlF4- or ScFn. These results suggest that SH1 and SH2 move distinctly during ATP hydrolysis and that the local conformations of the SH2 and RLR regions more sensitively reflect different transient states.

Characterization of the interaction of myosin with ATP analogues having the syn conformation with respect to the adenine-ribose bond

Numerous analytical experiments have shown that, in solution, ATP analogues with bulky substitutions at the eighth position of the adenine ring predominantly assume the syn conformation with respect to the adenine-ribose bond. Two such analogues, 3'-O-(N-methylanthraniloyl)-8-azido-ATP (Mant-8-N3-ATP) and 8-Br-ATP, were synthesized and used to probe the conformation of the ATP-binding site of myosin. In the presence of these analogues, actomyosin was rapidly dissociated; Mg2+-dependent ATP hydrolysis was significantly activated by actin; and Pi bursting was observed. For skeletal myosin, however, these analogues failed to support actin translocation, and they did not significantly enhance the intrinsic tryptophan fluorescence of skeletal muscle myosin subfragment-1 (SKE S-1). These results suggest that although myosin**/ADP/Pi intermediates can be formed with these analogues, the crucial conformational changes required for cross-bridge cycling do not occur in skeletal muscle myosin. The conformations of the ATP-binding sites of skeletal and smooth-muscle myosin were compared using the ternary complexes, myosin-ADP-beryllium fluoride (BeFn) or myosin-ADP-aluminium fluoride (AIF4-). In AlF4- complexes, Mant-8-N3-ADP affinity labeled the N-terminal 29-kDa domain of smooth-muscle myosin subfragment-1 (SM S-1), as did ATP analogues having the anti conformation, whereas it labeled the C-terminal 20-kDa domain of skeletal S-1. In smooth muscle BeFn complexes, Mant-8-N3-ADP was equally likely to cross-link to the 29-kDa N-terminal and the 25-kDa C-terminal domains. These analogues induced smooth muscle actomyosin super-precipitation and increased intrinsic tryptophan fluorescence to the same degree as ATP itself. As was expected from above results, the analogues supported smooth-muscle-myosin-induced actin translocation. These results suggest that smooth-muscle myosin adopts the eight-substituted ATP analogue in the normal conformation, but skeletal muscle myosin does not. This reflects the likely differences in the structures of their respective ATPase sites.

Analysis of stress in the active site of myosin accompanied by conformational changes in transient state intermediate complexes using photoaffinity labeling and 19F-NMR spectroscopy

Myosin forms stable ternary complexes with ADP and the phosphate analogues, fluoroaluminate (Al F4-), fluoroberyllate (BeFn) or orthovanadate (Vi); these ternary complexes mimic transient intermediates in the myosin ATPase cycle. Moreover, we previously demonstrated that these complexes may mimic different myosin ATPase reaction intermediates corresponding to separate steps in the cross-bridge cycle [Maruta, S., Henry, G. D., Sykes, B. D. & Ikebe, M. (1993) J. Biol. Chem. 268, 7093-7100]. Park et al. suggested that the changing conformation of ATP during hydrolysis stresses the active site of myosin subfragment-1 (S-1) through protein-nucleotide contacts at the gamma-phosphate and nucleotide base, and the stress-induced strain in the cross-bridge may be the mechanism by which energy in ATP is transferred to the myosin structure [Park, S., Ajtai, K. & Burghardt, T. P. (1997) Biochemistry 36, 3368-3372]. In the present study, the photoactive ADP analogue, 3'-O-(N-methylanthraniloyl)-2-azido-ADP (Mant-2-N3-ADP), and the 19F-labeled ADP analogue, 2-[(trifluoromethylnitrophenyl)aminoethyl]diphosphate, were employed to examine conformational differences in protein-nucleotide contact in the ATP-binding site that may correlate with energy transduction. Mant-2-N3-ADP was trapped within the active site of skeletal and smooth muscle myosin in the presence of AlF4-, BeFn or Vi. For both skeletal and smooth muscle myosins, trapped Mant-2-N3-ADP was covalently linked to the 25-kDa N-terminal fragment of S-1 of both myosin/Mant-2-N3-ADP/AlF4- and BeFn complexes, presumably at Trp130. However, the efficiency of the incorporation was much higher for skeletal than for smooth muscle myosin suggesting that the conformations of the adenine-binding pockets of the two myosins are somewhat different. Although the amount of Mant-2-N3-ADP trapped in the presence of AlF4- and BeFn was the same for both myosins, the efficiency of photolabeling skeletal muscle myosin was approximately two times higher for BeFn complex than for AlF4- complex. The 19F-NMR spectra of the bound 2-[(trifluoromethylnitrophenyl)aminoethyl]diphosphate in the ternary complexes formed in the presence of AlF4-, BeFn or Vi showed small but distinguishable differences. Taken together, these results indicate that there is some variation in the protein-nucleotide contacts at the nucleotide base among the ternary complexes studied, and these differences mimic separate steps occurring transiently during the contractile cycle.

A case of early cancer in cystic intrahepatic duct associated with congenital choledochal cyst

Herein, we report an extremely rare case of early cancer localized in the cystic intrahepatic duct, which is associated with the choledochal cyst including both the intra- and extrahepatic duct (type IV-A cyst). A small polypoid lesion in the cyst was detected by computed tomography. Extensive resection of the choledochal cyst with lobectomy was performed on the tumor in a 23-year-old woman. Tubular adenocarcinoma localized in the wall of the cystic left hepatic duct was histologically confirmed. Awareness of the diagnostic and treatment modalities, with a focus on potential malignancy in the cystic intrahepatic duct, in cases with anomalous arrangement of the pancreatobiliary duct is necessary.

The coexistence of esophageal submucosal tumor and carcinoma

The coexistence of esophageal submucosal tumor and carcinoma has been reported only in very few cases up to now, and the relationship between them is unclear. To elucidate the incidence and the pathogenesis of their coexistence, endoscopy and endoscopic ultrasonography (EUS) were sequentially performed in 95 patients who were suspected of having a submucosal tumor of the esophagus. EUS revealed 83 submucosal tumors and 12 cases of extraluminal compression. Two cases of esophageal carcinoma were observed in the series. In one patient, intraepithelial carcinoma was present on a leiomyoma originating in the muscularis mucosae, and in the other patient, multiple superficial carcinomas were presented.

Superficial esophageal carcinoma: evaluation by endoscopic ultrasonography

OBJECTIVES

The aim of this study was to evaluate the usefulness of endoscopic ultrasonography (EUS) in the staging of superficial esophageal carcinoma (SEC).

METHODS

We examined the histopathologic findings of 28 patients with SEC which was confirmed in the postoperative histologic evaluation. The EUS results preoperatively estimated were compared with them.

RESULTS

There were nine patients with mucosal carcinoma (two intraepithelial carcinomas and seven carcinomas limited to the mucosal layer) and 19 patients with submucosal carcinoma. The mucosal carcinomas revealed no lymph node metastasis or vessel permeation, whereas the submucosal carcinomas revealed lymph node metastasis (71%) and vessel permeation (lymphatic, 58%, vascular, 21%). The accuracy rates of depth of invasion by EUS were mucosa, 67% (6/9); submucosa, 79% (15/19); and total, 75% (21/28). EUS could not detect microinvasion to the submucosa. The overall accuracy rate of EUS in the evaluation of periesophagogastric lymph node metastasis was 72%: sensitivity 58% and specificity 85%.

CONCLUSIONS

It has been clinically proved important to differentiate mucosal carcinoma from submucosal carcinoma in the staging of SEC, and EUS is recommended in differentiating between them.

Observation of multiple myosin subfragment 1-ADP-fluoroberyllate complexes by 19F NMR spectroscopy

Fluoroaluminate and fluoroberyllate are potent inhibitors of the ATPase activity of myosin. Inhibition requires the presence of ADP, and much evidence has accumulated to suggest that the tetrahedral fluoroaluminate and fluoroberyllate ions act as phosphate analogues, binding with high affinity at the active site in the position normally occupied by the terminal phosphate of ATP. Both the S1-ADP-fluoroaluminate and the S1-ADP-fluoroberyllate species are thought to resemble kinetic intermediates in the actomyosin ATPase cycle. Characterization of S1-bound fluoroaluminate by 19F NMR is straightforward; a single resonance identified as AlF4- is observed easily [Maruta, S., Henry, G.D., Sykes, B.D., & Ikebe, M (1993) J. Biol. Chem. 268, 7093-7100]. Bound fluoroberyllate, by contrast, was found to give rise to four separate peaks: a downfield pair at -80 and -83.5 ppm and an upfield pair at -101.5 and -103 ppm, suggesting the existence of four distinct types of S1-ADP-fluoroberyllate complex. The relative intensities of the bound resonances can be altered by changing rhe F:Be ratio during complex formation. Integration of a spectrum acquired in the presence of a fluorine-labeled nucleotide derivative, 3'(2')-O-(4-fluorobenzoyl)-ADP, in place of ADP yielded a bound fluoride to nucleotide ratio of 1.7-1.9 to 1, showing that the major bound fluoroberyllate species cannot be BeF3- as is usually thought. It is proposed that the bound fluoroberyllates correspond to the neutral species BeF2(H2O)2 and BeFOH(H2O)2 and the negatively charged species [BeF2OH.H2O]- and [BeF3.H2O]-, although other possibilities are discussed.

Photoaffinity labelling of smooth-muscle myosin by methylanthraniloyl-8-azido-ATP

Methylanthraniloyl-8-azido-ATP (Mant-8-N3-ATP), which binds to the 20 kDa C-terminal tryptic fragment of skeletal-muscle myosin subfragment-1 [Maruta, Miyanishi and Matsuda (1989) Eur. J. Biochem. 184, 213-221], was synthesized and used as a probe of the conformational change of smooth-muscle myosin. Mant-8-N3-ATP, like ATP, induced the formation of the 10 S conformation at low ionic strength. In the presence of vanadate, smooth-muscle myosin formed a stable complex with Mant-8-N3-ADP, and this complex showed the 10 S-->6 S transition of myosin. ATP-binding sites for 6 S (extended state) and 10 S (folded state) myosin were studied by photolabelling of myosin with Mant-8-N3-ADP. For both 6 S and 10 S myosin, Mant-8-N3-ATP was incorporated into the 29 kDa N-terminal tryptic fragment of myosin heavy chain. This is unlike the labelling of skeletal-muscle myosin, in which the 20 kDa C-terminal fragment is labelled. The labelling of 29 kDa fragment was diminished significantly by addition of ATP. These results suggest that the conformation of the ATP-binding site of smooth-muscle myosin is different from that of skeletal-muscle myosin. To examine further the possible differences in the labelling site between 6 S and 10 S myosin, the affinity-labelled 29 kDa fragment was subjected to complete proteolysis by lysylendo-peptidase. The fluorescent-labelled-peptide map suggested that the Mant-8-N3-ADP-binding sites for 6 S and 10 S myosin were identical.

Formation of the stable myosin-ADP-aluminum fluoride and myosin-ADP-beryllium fluoride complexes and their analysis using 19F NMR

The effects of aluminum fluoride and beryllium fluoride on smooth muscle myosin and its subfragments were studied. Mg(2+)-ATPase activity was inhibited in the presence of aluminum fluoride (beryllium fluoride). [3H]ADP bound to heavy meromyosin (HMM) in the presence of aluminum fluoride (beryllium fluoride) and was not dissociated after 3 days of dialysis demonstrating that [3H]ADP was trapped in HMM. These results suggest the formation of a stable HMM-ADP-fluoroaluminate (fluoroberyllate) complex. The intrinsic tryptophane fluorescence intensity was increased in the presence of ADP and aluminum fluoride (beryllium fluoride). Acto-S1 was dissociated upon the formation of S1-ADP-fluoroberyllate and actin destabilized S1-ADP-fluoroberyllate complex, while S1-ADP-fluoroaluminate failed to bind to actin. Furthermore, when S1 formed the complex with actin, nucleotide trapping did not occur in the presence of fluoraluminate. These results indicated that the myosin-ADP-fluoroberyllate complex resembles a weak binding state while myosin-ADP-fluoroaluminate complex is a distinct conformation although the binding to actin was also weak. The structure of the ternary complex was investigated using 19F NMR. The 19F NMR spectrum of the S1-ADP-fluoroaluminate complex showed a peak at -66.7 ppm which is due to the binding of fluoraluminate to S1. The peak was not observed when 5'-adenylylimidodiphosphate was substituted for ADP suggesting that aluminum fluoride plays a role as a phosphate analogue. The stoichiometry of the bound fluoride was determined to be 3.8 mol/mol S1 suggesting that the bound species is AlF-4.

Endoscopic ultrasonography for assessing the horizontal extent of invasive gastric carcinoma

Accurately determining the surgical resection line is essential for improving the prognosis of patients with gastric carcinoma. Endoscopic ultrasonography (EUS) was used to assess the horizontal extent of cancerous invasion below the mucosal layer. In 105 patients (110 lesions) with gastric carcinoma, ultrasonographic images of resected specimens were compared with histological findings. The horizontal extent of the hypoechoic region of in vitro ultrasound was compared with the horizontal extent of the cancerous invasion and fibrosis in histological sections. In cases showing echo patterns characteristic of peptic ulcerations within the tumor focus, the horizontal spread of the echo-poor region correlated well with the spread of fibrosis, but not with the cancerous lesion. In other cases with no ulceration, the horizontal spread of the echo-poor region corresponded well with the cancerous lesions. To evaluate oral invasion of the cancerous lesion below the mucosa, 16 patients with gastric carcinoma who had clips placed during endoscopy were subsequently examined by EUS. The use of EUS combined with the clips allowed for a more accurate diagnosis of the invasion, not only in the mucosa, but also into the submucosa and beyond. This combined diagnostic method was safe and useful for determining a suitable surgical resection line.

Cross-linking of the 25- and 20-kilodalton fragments of skeletal myosin subfragment 1 by a bifunctional ATP analogue

The bifunctional photoreactive ATP analogue azidonitrobenzoyl-8-azido-ATP (ANB-8-N3-ATP) was synthesized. This ATP analogue carriers photoreactive azido groups at the eighth position of the adenine ring and at the 3' position of ribose. Photolysis of this analogue in the presence of skeletal muscle alpha-chymotryptic subfragment 1 (S-1) resulted in a new 120-kDa band, while photolysis in the presence of the tryptic S-1 produced a new 45-kDa band. The 45-kDa peptide was shown to be combined with the 25-kDa N-terminal and 20-kDa C-terminal fragments since it was labeled with a monoclonal antibody specific for the N-terminal 25-kDa segment of the S-1 heavy chain, and it was also found to retain the fluorescence of (iodoacetamido)fluorescein attached specifically to the SH-1 thiol of the C-terminal 20-kDa segment. These results indicate that the 25- and 20-kDa peptides are in close contact with the ATPase active site.

A second consensus sequence of ATP-requiring proteins resides in the 21-kDa C-terminal segment of myosin subfragment 1

Previous comparisons of sequence homologies of ATP-requiring enzymes have defined three consensus sequences which appear to be involved in the binding of the nucleotide. One of these was identified in the N-terminal 27-kDa segment of the myosin heavy chain but the other two sequences have not hitherto been located in myosin. The present paper proposes that one of these other two consensus sequences is in the 21-kDa C-terminal portion of S1 and that it may contribute to the ATP binding domain.

Hyposmolar stimulation of secretion of thyrotropin, prolactin, and luteinizing hormone does not require extracellular calcium and is not inhibited by colchicine, cytochalasin B, ouabain, or tetrodotoxin

Hyposmolar stimulation of thyroid-stimulating hormone, prolactin, and luteinizing hormone secretion by dispersed perifused rat pituitary cells was not depressed by removal of Ca2+ from the perifusion medium or by 0.1 mM colchicine, 20 microM cytochalasin B, 0.1 mM ouabain, or 3 microM tetrodotoxin. The secretory response induced by medium hyposmolarity or by thyrotropin-releasing hormone was not appreciably different at 23, 37, or 43 degrees C, but was markedly reduced or abolished when the experiments were performed at 1 degree C. These data indicate that microtubules or microfilaments, transport of extracellular Ca2+ into the cytoplasm, and plasmalemma ion transport mechanisms sensitive to ouabain or tetrodotoxin are not essential components of the mechanism by which extracellular hyposmolarity induces secretion.