A uroguanylin-GUCY2C endocrine axis regulates feeding in mice

Abstract

Intestinal enteroendocrine cells are critical to central regulation of caloric consumption, since they activate hypothalamic circuits that decrease appetite and thereby restrict meal size by secreting hormones in response to nutrients in the gut. Although guanylyl cyclase and downstream cGMP are essential regulators of centrally regulated feeding behavior in invertebrates, the role of this primordial signaling mechanism in mammalian appetite regulation has eluded definition. In intestinal epithelial cells, guanylyl cyclase 2C (GUCY2C) is a transmembrane receptor that makes cGMP in response to the paracrine hormones guanylin and uroguanylin, which regulate epithelial cell dynamics along the crypt-villus axis. Here, we show that silencing of GUCY2C in mice disrupts satiation, resulting in hyperphagia and subsequent obesity and metabolic syndrome. This defined an appetite-regulating uroguanylin-GUCY2C endocrine axis, which we confirmed by showing that nutrient intake induces intestinal prouroguanylin secretion into the circulation. The prohormone signal is selectively decoded in the hypothalamus by proteolytic liberation of uroguanylin, inducing GUCY2C signaling and consequent activation of downstream anorexigenic pathways. Thus, evolutionary diversification of primitive guanylyl cyclase signaling pathways allows GUCY2C to coordinate endocrine regulation of central food acquisition pathways with paracrine control of intestinal homeostasis. Moreover, the uroguanylin-GUCY2C endocrine axis may provide a therapeutic target to control appetite, obesity, and metabolic syndrome.

Figure 1. Gucy2c^–/– mice exhibit increased body weight, reflecting excess adiposity.

(A) Growth curves of male Gucy2c^+/+ and Gucy2c^–/– mice raised on LCD (n = 20; P < 0.001). (B) Growth curves of female Gucy2c^+/+ and Gucy2c^–/– mice raised on HCD (n = 20–40; P < 0.01). (C) Adiposity index (percentage body fat) of 12-month-old mice (n = 10–13 per group). (D) Visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), and total adipose tissue of 12-month-old mice (n = 10–13 per group). (E) Quantification of lean body mass of 12-month-old mice raised on LCD or HCD (n = 10–13 per group). (F) Total body water content of mice determined by 72 hours desiccation at 90°C (n = 6). All data are mean ± SEM. ***P < 0.001.

Figure 2. GUCY2C deficiency exacerbates metabolic disease.

(A) Liver triglyceride content of 12-month-old mice raised on LCD (n = 19). (B) H&E staining of representative left liver lobe sections of 12-month-old mice raised on LCD (scale bar: 200 μM). (C) Mean fasted serum leptin concentrations of mice raised on LCD (n = 6–10 per group). (D) Mean heart size (wet weight) of 12-month-old mice raised on LCD or HCD (n = 10–13 per group). (E) Mean fasted serum leptin concentrations of 12-month-old mice raised on HCD (n = 10). (F) Correlation of fasted serum leptin level and body weight of 12-month-old mice raised on HCD (r² = 0.47, P < 0.01). The diagonal line represents a linear regression model. (G) Mean fasted serum insulin concentrations of 12-month-old mice raised on HCD (n = 10). (H) Glucose tolerance test of fasted 12-month-old mice raised on HCD injected i.p. with glucose (2.5 mg/g) (n = 6). All data are mean ± SEM. NS = P > 0.05. *P < 0.05, **P < 0.01.

Figure 3. GUCY2C-deficient mice exhibit hyperphagia and diminished satiation.

(A) Daily food consumption of 3-month-old female and male mice fed MCD or HCD. Points represent the mean of 10 daily food intake measurements (n = 10 per group). (B) Growth of female mice pair-fed HCD (2.3 g/mouse/d) (n = 12). (C) Correlation of mean daily food intake and weight gain of 4-month-old mice fed HCD during a 10-day period (r² = 0.72, P < 0.001) (n = 10 per group). The diagonal line represents a linear regression model. (D) Twenty-four–hour food intake of fasted 3- to 4-month-old female and male mice refed MCD or HCD (n = 10–20 per group). (E) Cumulative food intake of 3- to 4-month-old fasted mice refed HCD (n = 10 per group). (F) Two-hour food consumption of fasted mice gavaged with an isovolumetric bolus (300 μl) of water, olive oil, or 35% glucose (n = 5 per group). All data are mean ± SEM. Horizontal bars represent mean values. Scattergram points represent data for individual mice. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 4. GUCY2C-deficient mice do not display increased lipid absorption efficiency or decreased activity/metabolic rate.

(A) Free fatty acid content of feces of 3- to 4-month-old mice raised on HCD (n = 16). Scattergram points represent data for individual mice. (B and C) Serum triglyceride concentrations of fasted mice with or without tyloxapol (0.75 mg/g) after olive oil gavage (1.5 mg/g) (n = 5). (D) Levels of expression of metabolic genes in intestine, determined by qRT-PCR, normalized to Vil1 expression (n = 4–5 per group). (E) Daily activity patterns of Gucy2c^+/+ and Gucy2c^–/– mice (n = 6). (F) Core body temperatures of mice exposed to a 4°C environment for 24 hours (n = 6). All data are mean ± SEM. Horizontal bars represent mean values.

Figure 5. Systemic administration of GUCY2C ligand induces satiation.

(A) Cumulative food intake of fasted Gucy2c^+/+ mice orally gavaged with 1 μg ST or the inactive alanine-substituted ST analogue, TJU, and refed HCD (n = 10 per group). (B) Cumulative food intake of fasted Gucy2c^+/+ mice injected i.v. with 1 μg ST or TJU and refed HCD (n = 10 per group). (C) Two-hour food intake of fasted Gucy2c^+/+ and Gucy2c^–/– mice injected i.v. with 1 μg TJU or ST and refed HCD (n = 10 per group). (D) Cumulative food intake of fasted Gucy2c^+/+ mice injected with TJU (1 μg) or ST and refed HCD (n = 10 per group). (E) Twelve-hour food intake of nonfasted Gucy2c^+/+ mice fed HCD and injected with 1 μg TJU or ST every 3 hours (n = 10 per group). (F) Two-hour food intake of fasted Gucy2c^+/+ and Gucy2c^–/– mice injected i.v. with TJU (1 μg) or PYY (3 μg) and refed HCD (n = 10 per group). (G) Two-hour food intake of fasted Gucy2c^+/+ mice injected i.v. with TJU (1 μg), ST (1 μg), or PYY (3 μg) and refed HCD (n = 10 per group). All data are mean ± SEM. **P < 0.01, ***P < 0.001.

Figure 6. GUCY2C expression in hypothalamus.

(A) Gucy2c expression in tissues of Gucy2c^+/+ mice, normalized to β-actin (Actb) expression (n = 4–6). The inset shows a representative gel image of the Gucy2c qRT-PCR products. (B) PCR products of the coding sequence of Gucy2c from cDNA prepared from Gucy2c^+/+ mouse tissues. Int, intestine; Hyp, hypothalamus; Lu, lung; Kid, kidney; Spl, spleen; NTC, nontemplate control. (C) Representative immunoblot of intestine and hypothalamic GUCY2C. (D) Intestine and hypothalamic GUCY2C protein content determined by immunoblot analyses (n = 3). (E) Guanylyl cyclase activity of membrane preparations with or without 1 μM ST (n = 4). (F) Two-hour food intake of fasted Gucy2c^+/+ mice after administration of TJU (10 μg), ST (10 μg), or exendin-4 (1 μg) into the third ventricle and refed HCD (n = 3 per group). (G) Hypothalamic expression of appetite-regulating neuropeptides, normalized to Actb expression, in fasted Gucy2c^+/+ mice injected i.v. with 1 μg TJU or ST every 2 hours for 8 hours (n = 8). All data are mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 7. Food intake stimulates intestinal prouroguanylin secretion inducing central satiation.

(A) Cumulative food intake of fasted Gucy2c^+/+ mice injected i.v. with TJU (10 μg), prouroguanylin (10 μg), or proguanylin (10 μg) and refed HCD (n = 10 per group). (B) cGMP production of CT26-GUCY2C cells treated with PBS, prouroguanylin (5 μg), proguanylin (5 μg), hypothalamic protein (350 μg), or combinations for 30 minutes (black bars, pH 5.5; red bars, pH 8.0; dark red bar, combined data of pH 5.5/8.0) (n = 4–10 per group). (C) Food intake and serum prouroguanylin concentrations of fasted Gucy2c^+/+ mice refed HCD (n = 7). (D) Serum prouroguanylin concentrations of human volunteers fasted for 12 hours and fed a standardized test meal (Supplemental Table 2) (P < 0.01, mean postprandial [15–150 minutes] level versus fasting [0 minutes] level) (n = 9). (E) Food intake of fasted Gucy2c^+/+ mice injected with PBS (control) or prouroguanylin antiserum (100 μl, 1:50 dilution) during the 1- to 2-hour interval after refeeding (n = 10 per group). All data are mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001.