Submucosal microinfusion of endothelin and adrenaline mobilizes ECL-cell histamine in rat stomach, and causes mucosal damage: a microdialysis study

Abstract

Rat stomach ECL cells release histamine in response to gastrin. Submucosal microinfusion of endothelin or adrenaline, known to cause vasoconstriction and gastric lesions, mobilized striking amounts of histamine. While the histamine response to gastrin is sustainable for hours, that to endothelin and adrenaline was characteristically short-lasting (1-2 h). The aims of this study were to identify the cellular source of histamine mobilized by endothelin and adrenaline, and examine the differences between the histamine-mobilizing effects of gastrin, and of endothelin and adrenaline. Endothelin, adrenaline or gastrin were administered by submucosal microinfusion. Gastric histamine mobilization was monitored by microdialysis. Local pretreatment with the H1-receptor antagonist mepyramine and the H2-receptor antagonist ranitidine did not prevent endothelin- or adrenaline-induced mucosal damage. Submucosal microinfusion of histamine did not cause damage. Acid blockade by ranitidine or omeprazole prevented the damage, suggesting that acid back diffusion contributes. Gastrin raised histidine decarboxylase (HDC) activity close to the probe, without affecting the histamine concentration. Endothelin and adrenaline lowered histamine by 50-70%, without activating HDC. Histamine mobilization declined upon repeated administration. Endothelin reduced the number of histamine-immunoreactive ECL cells locally, and reduced the number of secretory vesicles. Thus, unlike gastrin, endothelin (and adrenaline) is capable of exhausting ECL-cell histamine. Microinfusion of alpha-fluoromethylhistidine (known to deplete ECL cells but not mast cells of histamine) reduced the histamine-mobilizing effect of endothelin by 80%, while 1-week pretreatment with omeprazole enhanced it, supporting the involvement of ECL cells. Somatostatin or the prostanoid misoprostol inhibited gastrin-, but not endothelin-stimulated histamine release, suggesting that endothelin and gastrin mobilize histamine via different mechanisms. While gastrin effectively mobilized histamine from ECL cells in primary culture, endothelin had no effect, and adrenaline, a modest effect. Hence, the striking effects of endothelin and adrenaline on ECL cells in situ are probably indirect, possibly a consequence of ischemia.

Figure 1

Gastric mucosal damage caused by microinfusion for 3 h of adrenaline (1 mmol l⁻¹) (a) or endothelin (10 μmol l⁻¹) (b) into the submucosa of the acid-producing part of the rat stomach via microdialysis probes. Local administration of mepyramine and ranitidine, 1 h prior to and during the local administration of adrenaline failed to prevent the damage (c). Daily dosing with omeprazole, (400 μmol kg⁻¹) for 1 week prevented the gastric mucosal damage caused by local administration of endothelin (d). The black dotted line shows the position of the microdialysis probe in the submucosa.

Figure 2

Time course of gastric histamine mobilization in response to gastrin (0.1 mmol l⁻¹ (a)), endothelin (30 μmol l⁻¹ (b)) and adrenaline (1 mmol l⁻¹ (c)). Stimulation started at time 0 and lasted for 3 h (indicated by the horizontal line). The compounds were administered via microdialysis probes in the gastric submucosa. Mean±s.e.m., n=6–9.

Figure 3

Concentration–response curves for the histamine response to gastrin (a, b), endothelin (c, d) and adrenaline (e, f). The results are from experiments such as those shown in Figure 2, and are expressed as either peak response (a, c, e) or integrated response (b, d, f) (see Methods). The compounds were administered via microdialysis probes in the gastric submucosa. The results shown include data from an earlier study (Norlén et al., 2001). Mean±s.e.m., n=6–9.

Figure 4

Histamine concentration (a) and HDC activity (b) in small whole-wall specimens collected along the microdialysis probes in the stomachs of fasted rats challenged for 3 h with 0.9% NaCl (basal), gastrin (0.1 mmol l⁻¹), endothelin (10 μmol l⁻¹) or adrenaline (1 mmol l⁻¹) administered via the probes. Mean±s.e.m., n=6–8. (**) P<0.01, (***) P<0.005.

Figure 5

Microdialysate histamine concentration in response to repeated submucosal microinfusion of endothelin (10 μmol l⁻¹) 3 (a) or 48 (b) h after the first administration. In another series of experiments, adrenaline (1 mmol l⁻¹) was given twice, 5 (c) or 24 (d) h after the first administration. After 24 h, the response was greater than that after 5 h (P<0.05). The agents were administered via the microdialysis probes as indicated. Mean±s.e.m., n=4–5.

Figure 6

Histamine immunofluorescence in ECL cells in the gastric mucosa of rats exposed to endothelin (10 μmol l⁻¹ for 3 h) via a microdialysis probe. Tissue specimens were collected close to the probe (a), and from the contralateral side of the stomach (b). Immunofluorescent ECL cells were greatly reduced in number near the probe compared to the contralateral side of the stomach. The probe is indicated by asterisk. Bar=100 μm.

Figure 7

Electron micrographs of ECL cells in the gastric mucosa of rats exposed to submucosal microinfusion of endothelin (10 μmol l⁻¹ for 3 h) via a microdialysis probe. An ECL cell close to the probe (a) and a control cell (b) from the contralateral side of the stomach. Typical secretory vesicles (examples indicated by arrows) were greatly reduced in number in ECL cells near the probe compared to the contralateral side of the stomach. Bar=200 μm.

Figure 8

Endothelin-evoked histamine mobilization after continuous subcutaneous infusion of α-FMH (3 nmol kg⁻¹ h⁻¹) for 20 h prior to microdialysis. Comparison with vehicle-treated rats. Endothelin was administered via microdialysis probes for 3 h, as indicated by the horizontal line. Pretreatment with α-FMH greatly suppressed the histamine response to endothelin. Mean±s.e.m., n=4–5.

Figure 9

Time course of the histamine response to gastrin (0.1 mmol l⁻¹), endothelin (10 μmol l⁻¹) and adrenaline (1 mmol l⁻¹) in freely fed (a) and omeprazole-treated rats (b). The substances were administered via microdialysis probes, as indicated by the horizontal line. Note the enhanced histamine response to endothelin and adrenaline following pretreatment with omeprazole. Mean±s.e.m., n=4–8.

Figure 10

Effect of somatostatin (SST) (a, b) or misoprostol (c, d) on gastrin-induced histamine mobilization (a, c) and on endothelin-induced histamine mobilization (b, d). Gastrin (3 nmol kg⁻¹ h⁻¹) was given by intravenous infusion for 4 h as indicated. SST (0.1 mmol l⁻¹), misoprostol (0.1 mmol l⁻¹) and endothelin (10 μmol l⁻¹) were administered via microdialysis probes as indicated. Mean±s.e.m., n=3–7.

Figure 11

Effect of increasing concentrations of gastrin, endothelin and adrenaline on histamine secretion from isolated ECL cells in primary culture. Gastrin has a powerful effect, adrenaline is moderately active, and endothelin is without any effect. The control is baseline before adding stimuli. Mean±s.e.m., n=5 (n is the number of independent cell preparations).