Characterization of nitric oxide synthase in the rat parotid gland

Characterization of nitric oxide synthase in the rat parotid gland

Archives of Oral Biology 45 (2000) 531±536 www.elsevier.com/locate/archoralbio Characterization of nitric oxide synthase in the rat parotid gland Yu...

142KB Sizes 1 Downloads 167 Views

Archives of Oral Biology 45 (2000) 531±536

www.elsevier.com/locate/archoralbio

Characterization of nitric oxide synthase in the rat parotid gland Yuka Mitsui, Shunsuke Furuyama* Department of Physiology, Nihon University School of Dentistry at Matsudo, Matsudo, Chiba 271-8587, Japan Accepted 15 February 2000

Abstract Nitric oxide (NO) acts as an inter- and intracellular signalling molecule of various cells such as vascular endothelium, macrophages, and neurones. NO is produced by nitric oxide synthase (NOS) from L-arginine. Here the characteristics of NOS in the rat parotid gland were investigated. Approximately 74% of total activity of NOS was present in the cytosolic fraction. For full activation of the NOS in the cytosolic fraction, tetrahydroxybiopterin, NADPH, Ca2+ and calmodulin were needed as cofactors, because the activity was clearly reduced in the absence of tetrahydroxybiopterin, NADPH, or Ca2+, or in the absence of calmodulin and presence of tri¯uoperazine, a calmodulin antagonist, in the reaction mixture. The partially puri®ed NOS activity was completely abolished in the absence of calmodulin or Ca2+, and activated by them in a dose-dependent manner; EC50 for calmodulin and Ca2+ were 10 and 340 nM, respectively. The Km for L-arginine was 1.57 mM. Immunoblot analysis revealed that a 165kDa protein band in the rat parotid gland cytosolic fraction cross-reacted with a rabbit polyclonal antibody against human brain NOS. These results suggest that NOS of the rat parotid gland is a neuronal isoform and that its activity is regulated by physiological concentrations of calmodulin and Ca2+. 7 2000 Elsevier Science Ltd. All rights reserved. Keywords: Nitric oxide; Nitric oxide synthase; Ca2+; Calmodulin; Parotid gland

1. Introduction Nitric oxide is well known as a mediator of bloodvessel relaxation. Recent evidence indicates that it has Abbreviations: [Ca2+]i, intracellular Ca2+ concentration; DTT, dithiothreitol; H4BP, (6R)-5,6,7,8-tetrahydro-L-biopterin; NOLA, N G-nitro-L-arginine; PMSF, phenylmethylsulfonyl ¯uoride; W-7, N-(6-aminohexyl)-5-chloro-1-naphthalene sulfonamide hydrochloride. * Corresponding author. Tel.: +81-47-360-9326; fax: +8147-360-9327. E-mail address: [email protected] (S. Furuyama).

a much broader spectrum of functions and acts as an intra- and intercellular messenger in the central and peripheral nervous system in addition to the nonspeci®c cytotoxic activity of macrophages. Thus, nitric oxide proves to be a highly versatile signalling molecule regulating a variety of diverse cellular functions (Moncada et al., 1991; Bredt and Snyder, 1994). Nitric oxide is synthesized from L-arginine by three isoforms of nitric oxide synthase (EC 1.14.13.39), which have been characterized by molecular cloning and sequencing: n, e and i isoforms (Nathan and Xie, 1994). The n and e isoforms are constitutively expressed and their activities are regulated by shortterm elevations of intracellular Ca2+ concentration in

0003-9969/00/$ - see front matter 7 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 0 3 - 9 9 6 9 ( 0 0 ) 0 0 0 2 9 - 7

532

Y. Mitsui, S. Furuyama / Archives of Oral Biology 45 (2000) 531±536

response to hormone or neurotransmitter stimulation (Bredt and Snyder, 1990, 1994; Pollock et al., 1991). On the other hand, the i isoform is inducibly expressed by lipopolysaccharides and cytokines such as interleukin-1b, tumour necrosis factor-a and interferon-g, and its activity is not regulated by Ca2+ (Stuehr et al., 1991). We have previously reported the presence of nitric oxide synthase in mammalian salivary glands (Mitsui et al., 1997). In the feline parotid gland, 3-morpholinosydnonimine, a nitric oxide donor, slightly reduced vascular resistance, whereas the addition of NOLA, an inhibitor of nitric oxide synthase, resulted in an increase in vascular resistance, suggesting that basal production of nitric oxide in these organs is sucient to modulate vascular resistance (Lohinai et al., 1996). In the rat parotid gland, NOLA increased secretory volume and amylase secretion in vivo, suggesting that the L-arginine/nitric oxide pathway has a modulatory e€ect on salivary ¯uid and amylase secretion (Lohinai et al., 1997). However, the characteristics of nitric oxide synthase in the rat parotid gland have been little studied, and we have now investigated them.

at 105,000 g for 40 min. A portion (1 ml) of the supernatant fraction was passed through a Shephadex G-25 column (1  20 cm) equilibrated with the homogenizing bu€er to remove low molecular-weight e€ectors for nitric oxide synthase.

2. Materials and methods

2.3. Nitric oxide synthase assay

2.1. Materials

Enzyme activity was measured by monitoring the conversion of L-[2,3-3H]arginine to L-[2,3-3H]citrulline, essentially based on the method of Bredt and Snyder (1990). For the assay in crude samples (homogenate, supernatant, and Sephadex G 25-eluted fractions), the reaction mixture contained 33.3 mM HEPES (pH 7.4), 2 mM L-arginine with L-[2,3-3H]arginine (®nal concentration 2.2 GBq/mmol), 0.67 mM NADPH, 10 mM H4BP, 0.67 mM EDTA, 1.67 mM CaCl2, 0.67 mM DTT and 400 nM calmodulin in a ®nal volume of 150 ml. When the nitric oxide synthase partially puri®ed by 2',5 '-ADP Sepharose was used, NADPH was omitted, because enough NADPH was already contained in the elute (®nal concentration 4.7 mM). Free Ca2+ concentrations in the reaction mixture were adjusted by using Ca2+/EGTA bu€er and measuring them with a CAF 110 spectrophotometer (Nihon Bunkou, Japan) using fura-2 (Grynkiewicz et al., 1985). The assay for nitric oxide synthase was started by the addition of L-arginine with L-[2,3-3H]arginine. After incubation for 20 min at 308C, the reaction was terminated with 600 ml of 100 mM HEPES solution (pH 5.5) containing 10 mM EDTA and 1 mM L-citrulline (buffer B). Then, 500 ml of the mixture was applied to 1 ml of cation-exchange columns of Dowex AG50W-X8 (Na+ form; mesh size 100±200) equilibrated with buffer B. L-[2,3-3H]citrulline was eluted with 1.5 ml water, and radioactivity was measured by liquid scintillation counter.

L-[2,3-

3

H]arginine (1335.7 GBq/mmol) was obtained from DuPont/NEM Research Products (MA, USA). H4BP was from Research Biochemicals Inc. (MA, USA). Dowex AG50W-X8 (Na+ form) was from BioRad Laboratories (CA, USA). Shephadex G-25 and 2 ',5 '-ADP Sepharose were purchased from Amersham Pharmacia Biotech (Uppsala, Sweden). Centricon-30 was purchased from Amicon (MA, USA). Calmodulin and DTT were purchased from Boehringer Mannheim (IN, USA). W-7 was from Seikagaku Corporation (Tokyo, Japan). Anti-human nitric oxide synthase (n isoform) was from Transduction Laboratories (KY, USA). All other reagents were obtained from Sigma (MO, USA) and Wako (Osaka, Japan). 2.2. Enzyme preparation 2.2.1. Sephadex G-25 fraction Male Sprague±Dawley rats (300±350 g) were anaesthetized by intraperitoneal injection of sodium pentobarbital (50 mg/kg body wt). The parotid glands were removed, minced and homogenized with 3 vol of 20 mM HEPES bu€er (pH 7.2) containing 320 mM sucrose, 0.5 mM EDTA, 0.1 mM PMSF, 1 mM DTT (homogenizing bu€er) in a Potter±Elvehjem type homogenizer with a Te¯on pestle. The homogenate was passed through three layers of gauze, and centrifuged

2.2.2. 2 ',5 '-ADP Sepharose fraction Parotid glands removed and minced as described above were homogenized with 3 vol of 10 mM Tris/ HCl bu€er (pH 7.4) containing 1 mM EDTA, 1 mM PMSF and 10 mM DTT. After centrifugation at 105,000 g for 40 min, the supernatant fraction was loaded on to 0.5 ml of 2',5 '-ADP Sepharose column equilibrated with 10 mM Tris/HCl bu€er (pH 7.4) containing 1 mM PMSF and 10 mM DTT (bu€er A). The ¯ow-rate was adjusted to 1.6 ml/min. The column was washed with 20 ml of bu€er A containing 0.5 M NaCl followed by 20 ml of bu€er A. Then, nitric oxide synthase was eluted with 10 ml of bu€er A containing 10 mM NADPH. The fractions containing the synthase activity were designated as partially puri®ed enzyme and used for further characterization. All the procedures were carried out at 48C.

Y. Mitsui, S. Furuyama / Archives of Oral Biology 45 (2000) 531±536

2.4. Immunoblot analysis The partially puri®ed enzyme fractions from the 2 ',5 '-ADP Sepharose column were concentrated by ultra®ltration (Centricon-30), and processed through 6% sodium dodecyl sulphate±polyacrylamide gel electrophoresis. The proteins on the gel were transferred to nitrocellulose membrane at 13 V for overnight in a chilled transfer bu€er. The membrane was blocked with 1% bovine serum albumin, and subsequently incubated at a room temperature for 2 h with the polyclonal antibody raised against a peptide (1095±1289) of human brain nitric oxide synthase. Immunoreactivity was determined by ECL chemiluminescence reaction with goat anti-rabbit IgG±horseradish peroxidase conjugate as secondary antibody. Rat brain 105,000 g supernatant was used as a control. 2.5. Protein determination Protein concentrations were determined according to Bradford's (1976) method using bovine serum albumin as a standard. 3. Results 3.1. Requirement of cofactors In the rat parotid gland homogenate, nitric oxide synthase activity was 41.66 2 6.84 pmol citrulline/min per g of wet tissue. When the gland homogenate was centrifuged at 105,000 g for 40 min, 74.2 2 2.3% (n = 5) of the total nitric oxide synthase activities were detected in the supernatant fraction, indicating that the synthase is dominantly located in the cytosol. To examine the requirement of cofactors, the supernatant fraction was passed through a Sephadex G-25 column

533

to remove low molecularweight e€ectors for nitric oxide synthase (Table 1). When NADPH was removed from the reaction mixture, the synthase activity was completely abolished. In the absence of H4BP, the activity was reduced by 20.5%. These ®ndings indicate that NADPH and H4BP are necessary for the full activation of nitric oxide synthase in the rat parotid gland. When Ca2+ was removed from the reaction mixture, the activity was reduced by 20.0%. The removal of calmodulin did not clearly a€ect the activity. We suspected that endogenous calmodulin contained in the crude fraction was enough to activate the nitric oxide synthase. As expected, when the enzyme was assayed in the absence of calmodulin and the presence of the calmodulin antagonist tri¯uoperazine (100 mM), the activity was completely suppressed. Another calmodulin antagonist, W-7 (200 mM), mimicked the e€ect of tri¯uoperazine (data not shown). These results suggest that Ca2+ and calmodulin are necessary for the activation of nitric oxide synthase in the rat parotid gland. 3.2. Characterization The parotid nitric oxide synthase was partially puri®ed on a 2',5 '-ADP Sepharose anity column, which is very useful for puri®cation of NADPH-dependent enzymes such as isoforms of nitric oxide synthase (Nathan and Hibbs, 1991). The rat parotid nitric oxide synthase was puri®ed 402-fold from a 105,000 g supernatant with a speci®c activity of 251 pmol citrulline/ min per mg. The partially puri®ed enzyme was used for further characterization.

Table 1 E€ects of cofactors, Ca2+ and calmodulin on nitric oxide synthase (NOS) activity of the rat parotid glanda Condition

NOS activity (%)

Control NADPH (ÿ) H4BP (ÿ) CaCl2 (ÿ) Calmodulin (ÿ) Calmodulin (ÿ) Tri¯uoperazine (+)

100 1.120.6 20.523.3 20.022.3 95.524.4 3.321.8

a Activities of NOS in the Sephadex G-25 elute were assayed in the absence of cofactors and the presence of a calmodulin antagonist, tri¯uoperazine (100 mM). The activity in the complete system is indicated as 100%. Results are mean 2SEM in ®ve to six experiments.

Fig. 1. Determination of Km for L-arginine by Lineweaver± Burk plot. Double reciprocal plots of the initial reaction velocity (1/V) vs L-arginine concentration (1/S) in the presence (*) or absence (w) of the N G-nitro-L-arginine methylester, a nitric oxide synthase inhibitor.

534

Y. Mitsui, S. Furuyama / Archives of Oral Biology 45 (2000) 531±536

3.2.1. Km for arginine To determine Km for arginine, the partially puri®ed enzyme activity from the 2 ',5'-ADP Sepharose column was assayed with increasing concentrations of L-arginine (Fig. 1). The apparent Km for L-arginine at 308C and pH 7.4 was 1.57 mM. In the presence of 2 mM N G-nitro-L-arginine methylester (L-NAME), a competitive inhibitor against L-arginine, the Km for L-arginine rose to 3.22 mM without in¯uence on the Vmax (Fig. 1). 3.2.2. Calmodulin requirement The calmodulin requirement of the partially puri®ed enzyme was examined as shown in Fig 2. In the presence of 2.5 mM Ca2+, calmodulin activated the nitric oxide synthase in a dose-dependent manner, and a half-maximal stimulation of its activity was obtained at approx. 10 nM. No activity was detected in the absence of calmodulin.

Fig. 3. Ca2+-dependency of rat parotid gland nitric oxide synthase (NOS). Partially puri®ed rat parotid gland NOS was assayed at varying concentrations of Ca2+ in the presence of 400 nM calmodulin.

3.2.3. Ca2+ requirement The Ca2+ requirement of the partially puri®ed parotid nitric oxide synthase was examined (Fig. 3). In the presence of 400 nM calmodulin, Ca2+ activated the enzyme in a dose-dependent manner, and a half-maximal stimulation of its activity was obtained at approx. 340 nM. In the absence of Ca2+, the activity was completely abolished even in the presence of 400 nM calmodulin. This suggests that the nitric oxide synthase activity is dependent on Ca2+ concentration in the presence of calmodulin.

nitric oxide synthase are Ca2+- and calmodulin-dependent, and the neuronal isoform is dominantly localized in the cytosolic fraction. To con®rm the isoform, immunoblot analyses were performed. A protein of approx. 165 kDa in the cytosolic fraction of the rat parotid gland cross-reacted with a polyclonal antiserum generated against nitric oxide synthase from human brain (Fig. 4), supporting the likelihood that the nitric oxide synthase in that fraction is a neuronal isoform, although its molecular mass was slightly larger than that in rat brain.

3.3. Immunoblot analysis Constitutive, neuronal and endothelial isoforms of

Fig. 2. Calmodulin-dependency of rat parotid gland nitric oxide synthase (NOS). Partially puri®ed rat parotid gland NOS was assayed at varying concentrations of calmodulin in the presence of 2.5 mM Ca2+.

Fig. 4. Immunoblotting of rat parotid gland nitric oxide synthase (NOS) with anti-nNOS antibody. Lane 1, rat brain 105,000 g supernatant (40 mg) as a positive control; lane 2, partially puri®ed rat parotid gland NOS (40 mg).

Y. Mitsui, S. Furuyama / Archives of Oral Biology 45 (2000) 531±536

4. Discussion We demonstrate the enzymatic characterization of nitric oxide synthase in the soluble fraction of the rat parotid gland. NADPH and H4BP were necessary for full activation of the enzyme, as reported for other isoforms (Pollock et al., 1991). The Km for L-arginine was 1.57 mM, which is very close to that in rat brain, 1.5 mM (Bredt and Snyder, 1990). The nitric oxide synthase of the rat parotid was a Ca2+- and calmodulin-dependent. Its half-maximal stimulation (EC50) by calmodulin was the same as that in rat brain, 10 nM, and the EC50 for Ca2+ was slightly higher than that in rat brain, 200 nM (Bredt and Snyder, 1990). These results suggest that the rat parotid enzyme has characteristics similar to those of neuronal nitric oxide synthase. However, the molecular mass of the rat parotid enzyme was slightly higher than that of rat brain as revealed by immunoblot. Recently, an alternatively spliced isoform of the neuronal enzyme, nitric oxide synthase-m, the molecular mass of which is slightly larger than that in brain, has been demonstrated in rat skeletal and heart muscles (Silvagno et al., 1996). Therefore, the nitric oxide synthase in the rat parotid gland is possibly an isoform of the neuronal type, though this remains to be elucidated. The activity of the nitric oxide synthase of the rat parotid was regulated by Ca2+. In rat parotid acinar cells, changes of [Ca2+]i following stimulation by secretagogues have been demonstrated. Takemura (1985) reported that [Ca2+]i in the resting condition was 162 nM and was increased by about 450 nM by the stimulation of muscarinic receptors in quin-2-loaded cells. Merritt and Rink (1987) reported that the resting [Ca2+]i was 231 nM and that [Ca2+]i was increased by 1286 nM and 890 nM by the activation of substance P and muscarinic receptors, respectively, in fura-2-loaded cells. Ambudkar and Baum (1988) have reported that the resting concentration was between 140 and 180 nM, and was increased to 700 nM when fura-2-loaded parotid acinar cells were stimulated by adrenaline (epinephrine). These results, together with those in Fig. 3, suggest that nitric oxide synthase in the rat parotid gland is actually regulated by the change of [Ca2+]i induced by the activation of muscarinic and substance P receptors. Calmodulin is required for the Ca2+-dependent activation of nitric oxide synthase in the rat parotid. The concentrations are well below those found to stimulate adenylate cyclase of the parotid cell membrane (Piascik et al., 1986) and induce a signi®cant increase in the binding of secretory granules to plasma membranes (Watkins and Cooperstein, 1997). In the rat parotid, the concentration of calmodulin is probably high, because nitric oxide synthase activity in the elute through the Sephadex G-25 column was less a€ected

535

by removal of calmodulin in the reaction mixture. The enzyme activity was clearly reduced in the presence of calmodulin inhibitors such as tri¯uoperazine and W-7. These ®ndings imply that the endogenous calmodulin concentration is high enough for the activation of nitric oxide synthase. Therefore, Ca2+ probably acts as a regulator of nitric oxide synthase in the presence of calmodulin in the rat parotid. What is the physiological role of nitric oxide in the rat parotid gland? Lohinai et al. (1996) report that a nitric oxide donor reduced vascular resistance and a nitric oxide synthase inhibitor increased vascular resistance, and suggested that basal production of nitric oxide in the parotid gland is sucient to modulate vascular resistance. In addition, nitric oxide activates soluble guanylyl cyclase (Chinkers and Garbers, 1991; Koesling et al., 1991). In the parotid glands, muscarinic cholinergic stimulation results in an increase in intracellular cGMP in rat (Butcher et al., 1976; Harper and Brooker, 1977), mouse (Watson et al., 1982) and rabbit (WoÈjcik et al., 1975). Guanylyl cyclase activity is not directly regulated by calcium (Waldman and Murad, 1987). Therefore, it is most likely that the formation of cGMP induced by muscarinic receptor activation is mediated by guanylyl cyclase activation by nitric oxide generated by the synthase. Our previous study on rabbit parotid acinar cells (Michikawa et al., 1998) demonstrated that the muscarinic receptor agonist methacholine stimulates cGMP accumulation via the generation of nitric oxide induced by Ca2+ mobilization. In the rat and mouse parotid glands, the activation of Ca2+-mobilizing receptors such as muscarinic, substance P and a-adrenergic receptors stimulate K+ release and amylase secretion (Butcher et al., 1976; Harper and Brooker, 1977; Watson et al., 1982; WoÈjcik et al., 1975). These ®ndings imply that the activation of nitric oxide synthase is involved in the release of K+ and amylase secretion induced by the receptor stimulation. On the other hand, Lohinai et al. (1997) reported that an inhibitor of nitric oxide synthase increased secretory volume and amylase secretion in the rat parotid gland in vivo. This inconsistency should be resolved by further studies in the near future.

Acknowledgements This study was supported in part by a Suzuki Memorial Grant of Nihon University School of Dentistry at Matsudo in 1999 and Research for the Frontier Science Grant (the Ministry of Education, Science, Sports and Culture).

536

Y. Mitsui, S. Furuyama / Archives of Oral Biology 45 (2000) 531±536

References Ambudkar, I.S., Baum, B.J., 1988. ATP-dependent calcium transport in rat parotid basolateral membrane vesicles is modulated by membrane potential. Journal of Membrane Biology 102, 59±69. Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry 72, 248±254. Bredt, D.S., Snyder, S.H., 1990. Isolation of nitric oxide synthetase, a calmodulin-requiring enzyme. Proceedings of the National Academy of Sciences of the United States of America 87, 682±685. Bredt, D.S., Snyder, S.H., 1994. Nitric oxide: a physiologic messenger molecule. Annual Review of Biochemistry 63, 175±195. Butcher, F.R., McBride, P.A., Rudich, L., 1976. Cholinergic regulation of cyclic nucleotide levels, amylase release, and K+ e‚ux from rat parotid glands. Molecular and Cellular Endocrinology 5, 243±254. Chinkers, M., Garbers, D.L., 1991. Signal transduction by guanylyl cyclases. Annual Review of Biochemistry 60, 553±575. Grynkiewicz, G., Poenie, M., Tsien, R.Y., 1985. A new generation of Ca2+ indicators with greatly improved ¯uorescence properties. Journal of Biological Chemistry 260, 3440±3450. Harper, J.F., Brooker, G., 1977. Refractoriness to muscarinic and adrenergic agonists in the rat parotid: responses of adenosine and guanosine cyclic 3 ',5'-monophosphates. Molecular Pharmacology 13, 1048±1059. Koesling, D., BoÈhme, E., Schultz, G., 1991. Guanylyl cyclases, a growing family of signal-transducing enzymes. The FASEB Journal 5, 2785±2791. Lohinai, Z., Balla, I., Marczis, J., Vass, Z., KovaÂch, A.G.B., 1996. The e€ect of a nitric oxide donor and an inhibitor of nitric oxide synthase on blood ¯ow and vascular resistance in feline submandibular, parotid and pancreatic glands. Archives of Oral Biology 41, 699±704. Lohinai, Z., Burghardt, B., Zelles, T., Varga, G., 1997. The e€ect of L-arginine/nitric oxide pathway on salivary amylase secretion in conscious rats. Journal of Physiology (Paris) 91, 217±221. Merritt, J.E., Rink, T.J., 1987. The e€ects of substance P and carbachol on inositol tris- and tetrakisphosphate formation and cytosolic free calcium in rat parotid acinar cells. Journal of Biological Chemistry 262, 14,912±14,916. Michikawa, H., Mitsui, Y., Fujita-Yoshigaki, J., HaraYokoyama, M., Furuyama, S., Sugiya, H., 1998. cGMP production is coupled to Ca2+-dependent nitric oxide gen-

eration in rabbit parotid acinar cells. Cell Calcium 23, 405±412. Mitsui, Y., Yasuda, N., Furuyama, S., Sugiya, H., 1997. Nitric oxide synthase activities in mammalian parotid and submandibular salivary glands. Archives of Oral Biology 42, 621±624. Moncada, S., Palmer, R.M.J., Higgs, E.A., 1991. Nitric oxide: physiology, pathophysiology, and pharmacology. Pharmacological Reviews 43, 109±142. Nathan, C.F., Hibbs Jr, J.B., 1991. Role of nitric oxide synthesis in macrophage antimicrobial activity. Current Opinion in Immunology 3, 65±70. Nathan, C., Xie, Q.W., 1994. Nitric oxide synthases: roles, tools, and controls. Cell 78, 915±918. Piascik, M.T., Babich, M., Jacobson, K.L., Watson, E.L., 1986. Calmodulin activation and calcium regulation of parotid gland adenylate cyclase. American Journal of Physiology 250, C642±C645. Pollock, J.S., FoÈrstermann, U., Mitchell, J.A., Warner, T.D., Schmidt, H.H.W., Nakane, M., Murad, F., 1991. Puri®cation and characterization of particulate endothelium-derived relaxing factor synthase from cultured and native bovine aortic endothelial cells. Proceedings of the National Academy of Sciences of the United States of America 88, 10,480±10,484. Silvagno, F., Xia, H., Bredt, D.S., 1996. Neuronal nitric-oxide synthase-m, an alternatively spliced isoform expressed in di€erentiated skeletal muscle. Journal of Biological Chemistry 271, 11,204±11,208. Stuehr, D.J., Cho, H.J., Kwon, N.S., Weise, M.F., Nathan, C.F., 1991. Puri®cation and characterization of the cytokine-induced macrophage nitric oxide synthase: an FADand FMN-containing ¯avoprotein. Proceedings of the National Academy of Sciences of the United States of America 88, 7773±7777. Takemura, H., 1985. Changes in cytosolic free calcium concentration in isolated rat parotid cells by cholinergic and b-adrenergic agonists. Biochemical and Biophysiccal Research Communications 131, 1048±1055. Waldman, S.A., Murad, F., 1987. Cyclic GMP synthesis and function. Pharmacological Reviews 39, 163±196. Watkins, D.T., Cooperstein, S.J., 1997. E€ects of calcium and calmodulin on the binding of rat parotid secretion granules to the plasma membrane. Journal of Dental Research 76, 744±753. Watson, E.L., Jacobson, K.L., Dowd, F., 1982. Does cyclic GMP mediate amylase release from mouse parotid acini? Life Sciences 31, 2053±2060. Wojcik, J.D., Grand, R.J., Kimberg, D.V., 1975. Amylase secretion by rabbit parotid gland. Role of cyclic AMP and cyclic GMP. Biochimica et Biophysica Acta 411, 250±262.