IP3 Receptor Types 2 and 3 Mediate Exocrine Secretion Underlying Energy Metabolism

See allHide authors and affiliations

Science  30 Sep 2005:
Vol. 309, Issue 5744, pp. 2232-2234
DOI: 10.1126/science.1114110


Type 2 and type 3 inositol 1,4,5-trisphosphate receptors (IP3R2 and IP3R3) are intracellular calcium-release channels whose physiological roles are unknown. We show exocrine dysfunction in IP3R2 and IP3R3 double knock-out mice, which caused difficulties in nutrient digestion. Severely impaired calcium signaling in acinar cells of the salivary glands and the pancreas in the double mutants ascribed the secretion deficits to a lack of intracellular calcium release. Despite a normal caloric intake, the double mutants were hypoglycemic and lean. These results reveal IP3R2 and IP3R3 as key molecules in exocrine physiology underlying energy metabolism and animal growth.

Inositol 1,4,5-trisphosphate receptors (IP3Rs) are intracellular Ca2+ release channels located on the endoplasmic reticulum (ER) that mediate Ca2+ mobilization from the ER to the cytoplasm in response to the binding of a second messenger, inositol 1,4,5-trisphosphate (IP3) (1). IP3-induced Ca2+ release is triggered by various external stimuli, and most non-excitable cells use this mechanism as the primary Ca2+ signaling pathway. IP3Rs are therefore thought to have important physiological roles in various cell types and tissues (2). Three subtypes of IP3Rs, derived from three distinct genes, have been identified in mammals (3). Type 1 IP3R (IP3R1) is predominantly expressed in brain tissue and plays a critical role in the regulation of motor and learning systems (47). The other two subtypes, type 2 and 3 IP3Rs (IP3R2 and IP3R3), are expressed in various tissues and cell lines (811); however, the importance of these subtypes in vivo has been difficult to assess because of their co-expression in tissues and the lack of selective inhibitors. In this study, we examined mice lacking both IP3R2 and IP3R3 and observed defects in the digestive system resulting from the lack of Ca2+ signaling in exocrine tissues. In such exocrine tissues, secretagogue-induced increases in intracellular Ca2+ concentration ([Ca2+]i) trigger the secretion of enzymes or water by acting on the Ca2+-dependent exocytotic machinery or ion channels, respectively (1216). A crucial physiological role of IP3Rs in exocrine Ca2+ signaling was demonstrated (15, 17); however, the relative importance of the three different IP3R subtypes has been unclear.

We generated mice lacking either IP3R2 or IP3R3 by disrupting the corresponding genes within their first coding exons (figs. S1A and S1B). The single-gene mutants were viable and showed no distinct abnormalities in appearance, at least for several months after birth. Mutant mice lacking both of these IP3R subtypes were also viable during the embryonic period. Immunoblot analysis of the submandibular glands and the pancreas, where IP3R2 and IP3R3 are expressed (fig. S1C), showed that expression of IP3R2 and IP3R3 was abolished in the mutants (Fig. 1A). At birth, the appearance of double homozygotes was indistinguishable from that of nonhomozygous littermates, but double homozygotes had gained less body weight after birth. After the weaning period, around postnatal day 20 (P20), the homozygotes began losing weight and died within the 4th week of age (Fig. 1B). We suspected that an incapability of the double mutants to eat dry food after weaning might have caused body weight loss and eventual death. Indeed, double mutants did not consume dry food at all. When the double mutants were fed wet mash food beginning at P20, they consumed this type of food and survived thereafter. Body weight increases of the double mutants, however, were still smaller than those of nondouble mutant littermates equally fed with wet mash food (Fig. 1C). Interestingly, despite their reduced body weights, the double mutants consumed no less wet mash food than did the control mice (Fig. 2A and fig. S2A). The double mutants also took as much milk as did control mice when they were fed milk instead of wet mash food after weaning (fig. S2B). Thus, the caloric intake of the IP3R2–/–-IP3R3–/– double mutants appeared to be slightly greater than that of the control mice. In addition, the amount of feces produced by adult mice fed wet mash food was higher in the double mutants (Fig. 2B). The total amount of proteins and lipids in the feces were higher in the double mutants (Fig. 2C and fig. S2C). Furthermore, blood glucose concentrations were significantly lower in the double mutants (86.1 ± 5.3 mg/dl, n = 11) than those in control mice (156.1 ± 6.5 mg/dl, n = 14). Altogether, these results suggest that digestive system dysfunction causes the malnutrition phenotype of the double mutants. Actually, when the double mutants were fed a predigested diet containing glucose and amino acids for a week, they gained weight (1.7 ± 0.7 g, n = 8), whereas those fed wet mash food did not (–0.5 ± 0.3 g, n = 5).

Fig. 1.

Postnatal growth of mice lacking IP3R2 and IP3R3. (A) Western blot analysis of IP3R subtypes in the SMGs and the pancreas. Lane 1, wild type (IP3R2+/+-IP3R3+/+); 2, IP3R3 knock-out (IP3R2+/+-IP3R3–/–); 3, IP3R2 knock-out (IP3R2–/–-IP3R3+/+); and 4, IP3R2/IP3R3 double knock-out (IP3R2–/–-IP3R3–/–). Protein samples (50 μg) were probed with subtype-specific antibodies. (B and C) Growth curve of male mice fed (B) dry or (C) wet mash food. Each point is the average body weight (± SEM) of six animals. Control: nondouble mutants.

Fig. 2.

Feeding behavior and fecal amount of IP3R2–/–-IP3R3–/– mice. Error bars indicate SEM. (A) Wet mash food intake of the double mutant (KO) (n = 4) and the nondouble mutant (control) (n = 6) mice. Food intake was monitored for 10 days beginning at 5 weeks of age and normalized to their (body weights)0.75. (B) Daily stool mass of adult mice normalized to their (body weights)0.75 (n = 12). (C) Protein content in the feces of adult (n = 8 per group) and 2 weeks old (n = 7 per group) mice.

Because the lethal double mutant pheno-type was partially rescued by macerating the food with water, we hypothesized that the double mutants might be deficient in saliva production. We therefore examined saliva secretion in adult mice stimulated by subcutaneous (s.c.) injection of pilocarpine and isoproterenol, which mimic cholinergic and β-adrenergic stimulation, respectively. Pilocarpine-stimulated salivation was impaired only in the double knock-outs (Fig. 3A). Mice tested immediately after the weaning period also showed similar defects in salivation as above. Electrolyte concentrations of saliva were also altered in the double mutants (fig. S3A). Immunoblot analysis showed that in the submandibular gland (SMG), the largest salivary gland in the mouse, the expression of aquaporin 5 (AQP5), a water channel responsible for saliva secretion (18, 19), was not altered in the double mutants (fig. S3B). This indicates that the normal water route was intact. Cholinergically induced amylase secretion was also greatly impaired in the double knock-outs [6.4 ± 0.4 units (n = 6), compared with 46.2 ± 3.2 units amylase activity in control animals (n = 6), in the total volume of saliva produced in 30 min after pilocarpine stimulation (10 mg/kg)]. These results demonstrate a deficiency in saliva secretion in the double mutants and suggest that it may lead to starvation. In some cases, however, animals with salivation defects survive with standard dry food pellets by drinking more water (20, 21). Because IP3R double mutants did not show frequent access to water during eating, a learning impairment may be present in the double mutants that prevents them from adapting to an alteration in feeding behavior.

Fig. 3.

Saliva secretion and calcium signaling. (A) Saliva secretion of adult mice (2 to 3 months old and 4 or 5 mice per group) stimulated by the indicated agonists for 30 min. 1, wild-type; 2, IP3R2 knock-out; 3, IP3R3 knock-out; and 4, IP3R2-IP3R3 double knock-out. Error bars indicate SEM. (B) [Ca2+]i changes induced by muscarinic receptor stimulation in SMG cells of the indicated genotype. Fura-2-loaded SMG cells were stimulated with CCh at the time point depicted by the arrow.

Because salivation in salivary gland acinar cells is triggered by increased [Ca2+]i (14), we measured the [Ca2+]i in SMG cells isolated from wild-type, IP3R2–/–, IP3R3–/–, or IP3R2–/–-IP3R3–/– mice incubated with the Ca2+-sensing dye, fura-2. The effect of carbachol (CCh), a cholinergic stimulant that increases [Ca2+]i, was greatly reduced in IP3R2–/–-IP3R3–/– SMG cells and was moderately reduced in single knock-out IP3R2–/– or IP3R3–/– mice (Fig. 3B and fig. S3, C and D). IP3 production in response to CCh stimulation and capacitative Ca2+ entry were normal, and the Ca2+ stores were not emptied in IP3R2–/–-IP3R3–/– SMG cells (fig. S4).

Histological analysis of the pancreatic tissues, another major exocrine gland, of the double mutants also revealed abnormalities. Pancreatic acinar cells of the IP3R2–/–-IP3R3–/– double mutants were highly eosinophilic (fig. S5A) and showed abnormal accumulation of zymogen granules (Fig. 4A). These results suggest that unreleased zymogen granules accumulate in the cytoplasm of acinar cells. To evaluate the function of the exocrine pancreas, we stimulated dissociated pancreatic cells with CCh for 30 min and measured the pancreatic amylase secreted. Although muscarinic receptor–mediated amylase secretion from wild-type, IP3R2–/–, and IP3R3–/– cells was observed, there was almost no CCh-induced amylase secretion from IP3R2–/–-IP3R3–/– pancreatic cells (Fig. 4B). CCh-induced secretion of lipase and trypsinogen was also abolished exclusively in the double mutants (fig. S5, B and C). Normal amounts of amylase were present in the pancreata of the double mutants (fig. S6A), and ionomycin, a Ca2+ ionophore, induced the release of amylase (Fig. 4B). These results indicate that a process mediated by IP3R2 and IP3R3 serves as a key step in triggering secretion. This deficiency in exocrine pancreas function may also contribute to the smaller body size of IP3R2–/–-IP3R3–/– double mutants, which eat no less food than control animals (Fig. 2A). Incomplete digestion of ingested food, resulting from exocrine secretion deficits, would lead to reduced absorption, resulting in an undernourished phenotype.

Fig. 4.

Exocrine pancreas function and calcium signaling. (A) Electron micrographs of the pancreata from wild-type (WT) and IP3R2–/–-IP3R3–/– double knock-out mice. (B) Amylase release from dissociated pancreatic acinar cells stimulated by CCh for 30 min. Release is represented as a percentage of the initial intracellular content of amylase (amylase activity in unstimulated cells). The value of each point is the average ± SEM of each genotype (n from 6 to 8). Lane 1, WT; 2, IP3R2 knock-out; 3, IP3R3 knock-out; and 4, IP3R2/IP3R3 double knock-out. (C) [Ca2+]i changes induced by ACh receptor or CCK8 receptor stimulation in pancreatic cells. Fura-2-loaded pancreatic cells were stimulated with ACh or CCK8 at the time points depicted by the arrows.

We examined Ca2+ signaling induced by pancreatic exocrine secretagogues in fura-2-loaded, enzymatically dispersed pancreatic cells. No [Ca2+]i increase was induced by acetylcholine (ACh) or cholecystokinin octapeptide (CCK8) in IP3R2–/–-IP3R3–/– pancreatic cells (Fig. 4C and fig. S6B and fig. S6C). We confirmed these results with Ca2+ measurements in individual acinar cells (fig. S6D). Either IP3R2 or IP3R3 could cause Ca2+ waves and oscillations initiating at apical poles (fig. S7), consistent with apical localization of these receptors (2225). Because the intracellular signal transduction pathway to produce IP3 was functional and the sarco/endoplasmic reticulum Ca2+-adenosine triphosphatase (ATPase)-sensitive Ca2+ stores were not emptied in these cells (fig. S8), we presume that lack of both IP3R2 and IP3R3 causes the dysfunction in pancreatic Ca2+ signaling. These results demonstrate that IP3R2 and IP3R3 are the major Ca2+ release channels responsible for secretagogue-induced Ca2+ signaling in pancreatic acinar cells and subsequent digestive enzyme secretion. The characteristics of the IP3R2–/–-IP3R3–/– double mutants represent some symptoms of human diseases. Thus, increasing or decreasing activity of IP3R2 and IP3R3 could potentially be of therapeutic benefit.

Supporting Online Material

Materials and Methods

Figs. S1 to S8


References and Notes

Stay Connected to Science

Navigate This Article