Collagenase L was obtained from Nitta Gelatin (Osaka, Japan). Pilocarpine was purchased from Nacalai Tesque (Kyoto, Japan). Cevimeline was a gift from Nippon Kayaku (Tokyo, Japan). Fura-2 AM was obtained from Dojindo (Kumamoto, Japan). Sodium chloride, potassium chloride, sodium bicarbonate, magnesium chloride, calcium chloride dihydrate, D-(+)-glucose, HEPES, bovine serum albumin, carbachol, 3’-O-(4-benzoylbenzoyl) adenosine 5’-triphosphoric acid (BzATP), propranolol, atropine, A23187, and all other chemicals were purchased from Sigma Japan (Tokyo, Japan).

All experiments were approved by the Animal Use Committee of Kyushu Dental College. C57BL/6J mice, purchased from Kyushu Animal Laboratory, were acclimated to a 12-h light/dark cycle and were fed ad libitum. Experiments were performed in mice at 8 - 10 weeks of age.

An ex vivo vascular perfusion technique was used to accurately control thermal conditions and eliminate neurological input for analysis of glandular fluid secretion. The surgical procedure has been previously described [12,13]. The submandibular glands were dissected, and the main artery was cannulated with a 31-gauge bluntend cannula and perfused at a flow rate of 1 ml/min. The perfusion solution consisted of 120 mM NaCl, 4.3 mM KCl, 25 mM NaHCO₃, 1.0 mM MgCl₂, 1.0 mM CaCl₂, 5 mM glucose, and 10 mM HEPES equilibrated with 95% O₂/5% CO₂ and stored at room temperature. A custommade water-jacket heating system was used to control the temperature of the solutions and glands. Fluid secretion was stimulated by the addition of a submaximal concentration of the muscarinic agonist carbachol (CCh, 0.3 µM). The secreted saliva was collected in a pre-calibrated glass capillary tube.

Intracellular Ca²⁺ ([Ca²⁺]_i) was measured as previously described [13,14]. The submandibular glands were removed from anesthetized mice, placed in physiological saline (120 mM NaCl, 4.3 mM KCl, 25 mM NaHCO₃, 1.0 mM MgCl₂, 1.0 mM CaCl₂, 5 mM glucose, and 10 mM HEPES supplemented with 1 mg/ml bovine serum albumin), minced approximately 100 times with fine scissors, and digested with 520 U/ml collagenase L for 20 min with continuous shaking. Dissociated cells were incubated with 1 μM Fura-2 AM to measure [Ca²⁺]_i. A custom-made water-jacket heating system was used to control the temperature of the cells and perfusion solutions. Fluorescence was detected under a microscope equipped with a fluorescence analysis system (Argus/ Aquacosmos; Hamamatsu Photonics, Hamamatsu, Japan), with excitation at 340 and 380 nm and emission at 510 nm. The [Ca²⁺]_i was calculated as the ratio of the emissions following excitation at 380 and 340 nm.

The data are shown as means ± standard error (S.E.). Student’s t-test was used to compare group means. To detect statistically significant differences in multiple comparisons, one-way analysis of variance (ANOVA) followed by Tukey’s post-hoc test was performed using SPSS v.14.0J (SPSS, Chicago, IL). Results were deemed statistically significant at P ≤ 0.05.

We compared fluid secretion in response to the clinically used muscarinic agonists pilocarpine and cevimeline, and the experimentally used agonist CCh. The flow rate data are summarized in Figures 1(a), (c) and (e) (Figure 1). At all temperatures tested, all three muscarinic agonists induced fluid secretion. At 25˚C and 30˚C, secretion was high for the first 30 s, decreased slightly over the next 30 s, and then remained constant. At 37˚C and 42˚C, there was no decrease at 30 - 60 s.

The total secretion over 10 min is shown in Figures 1(b), (d) and (f). Initially, 10 µM pilocarpine induced 175.1 ± 17.3 μL of saliva ex vivo during the 10-min stimulation at 37˚C; this was 5.5-fold the amount of saliva produced at 25˚C. The increase between 37˚C and 42˚C was not significant (Figure 1(b)). In contrast, 30 µM cevimeline produced 118.0 ± 4.3 μL of saliva at 37˚C, which was 9.8-fold the amount of saliva produced at 25˚C, and all cevimeline-induced increases were significant at all temperatures tested (Figure 1(d)). Treatment with 0.3 µM CCh induced the production of 148.4 ± 3.2 μL of saliva ex vivo during the 10-min stimulation at 37˚C, which was 2.3-fold the amount of saliva produced at 25˚C; the increase between 37˚C and 42˚C was not significant (Figure 1(f)).

Only small amounts of saliva were collected in response to lower concentrations (~3 µM) of pilocarpine and cevimeline at 25˚C, leading us to investigate the response to a series of CCh concentrations at 25˚C and 37˚C.

Although the dose-response curve was shifted upward at the higher temperature, the CCh concentration producing the peak flow was the same for both curves (Figure 2). These results suggest that muscarinic receptors are not responsible for the temperature dependence of the increased secretion ex vivo. We analyzed the ionic composition of the collected saliva and found no significant

changes among the tested conditions.

To identify the effect of the P2X7 receptor on the tem-

perature-dependent increases, we tested the effect of the P2X7 agonist 3’-O-(4-benzoylbenzoyl) adenosine 5’-triphosphoric acid (BzATP) at 0.5 mM in the presence of both the muscarinic antagonist atropine (0.5 µM) and the β-antagonist propranolol (20 µM). The initial peak flow rate resulted in a 1.5-fold increase between 25˚C and 37˚C, but the difference in the total amount of saliva secretion was not statistically significant (Figure 3). We also tested A23187 (1 µM), a calcium ionophore, to bypass muscarinic-dependent signaling. A23187 produced almost identical amounts of saliva secretion at 25˚C and 37˚C (Figure 4), indicating that Ca²⁺ influx mechanisms are not involved in temperature-dependent fluid secretion ex vivo.

The measurements of [Ca²⁺]_i during carbachol stimulation revealed significant differences in the fluorescence ratios at the peak and the plateau level between 25˚C and 37˚C (Figure 5). This result, combined with the results in Figure 1(c), suggests that the muscarinic-stimulated increase in saliva secretion may be attributable to an intracellular Ca²⁺ handling mechanism.