Gastrokine-1, an anti-amyloidogenic protein secreted by the stomach, regulates diet-induced obesity

Abstract

Obesity and its sequelae have a major impact on human health. The stomach contributes to obesity in ways that extend beyond its role in digestion, including through effects on the microbiome. Gastrokine-1 (GKN1) is an anti-amyloidogenic protein abundantly and specifically secreted into the stomach lumen. We examined whether GKN1 plays a role in the development of obesity and regulation of the gut microbiome. Gkn1^-/- mice were resistant to diet-induced obesity and hepatic steatosis (high fat diet (HFD) fat mass (g) = 10.4 ± 3.0 (WT) versus 2.9 ± 2.3 (Gkn1^-/-) p < 0.005; HFD liver mass (g) = 1.3 ± 0.11 (WT) versus 1.1 ± 0.07 (Gkn1^-/-) p < 0.05). Gkn1^-/- mice also exhibited increased expression of the lipid-regulating hormone ANGPTL4 in the small bowel. The microbiome of Gkn1^-/- mice exhibited reduced populations of microbes implicated in obesity, namely Firmicutes of the class Erysipelotrichia. Altered metabolism consistent with use of fat as an energy source was evident in Gkn1^-/- mice during the sleep period. GKN1 may contribute to the effects of the stomach on the microbiome and obesity. Inhibition of GKN1 may be a means to prevent obesity.

Figure 1

GKN1 regulates age-associated accumulation of body fat. Male WT and GKN1^−/− mice were maintained on NCD ad libitum and assessed at 6, 12 and 24 weeks of age for (a) body weight and (b) fat mass (by qMRI) or (c) fat mass as a percent of body weight. (d) Perigonadal fat pads were excised and weighed. (E&F) lean mass was also assessed by qMRI. *p < 0.05, **p < 0.01, ***p < 0.001 n = 4.

Figure 2

GKN1 regulates Angptl4 expression in the intestine. WT and GKN1^−/− mice were assessed for gastric mucosal expression of (a) ghrelin and intestinal gene expression of (b) CCK, (c) GIP mRNA measured by qPCR (n = 2–4). Serum levels of (d) GLP-1 or (e) leptin by ELISA (n > 4). (f) Intestinal expression of ANGPTL4 mRNA measured using qPCR (n = 2–4). The qPCR data was normalized to GAPDH. Immunolocalization of Gkn1 in the gastric mucosa compared to (g) ghrelin or (h) leptin in WT vs. Gkn1^−/− mice (scale bar is 75 μM). *p < 0.05.

Figure 3

GKN1 regulates the small bowel microbiome. (A-D) Phylogenetic analysis of small bowel microbial populations in WT and GKN1^−/− mice at 6 and 12 weeks of age on NCD. (a,c) Relative phyla abundance in the small bowel of mice showing significantly reduced abundance of Firmicutes in 12-week, but not 6-week-old GKN1^−/− mice (bolded numbers are p < 0.01 n = 7). (b,d) Relative class abundance in the small bowel of mice showing decreased abundance of Erysipelotrichia in 12-week but not 6-week-old mice (bolded numbers are p < 0.01 n = 7).

Figure 4

GKN1 regulates fat accumulation in response to high fat diet. Male WT and GKN1^−/− mice were individually caged and then switched from NCD to HFD and monitored for 8 weeks for (a) body weight and (b) fat mass (by qMRI) (n = 12–15). Livers were excised and assessed for (c) total weight, (d) triglyceride content and (e) presence of steatosis by H&E histology or presence of lipid by (f) staining with oil red O (scale bar is 200 μM). Expression levels of gene for (g) gluconeogenesis (PepcK, G6P, FBP, PC) or (h) inflammation (IL1b, IL6) were assessed in liver tissue from HFD treated mice. (n = 3–4) *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 5

GKN1^−/− do not exhibit signs of cancer or inflammation. (a) Male WT and GKN1^−/− were imaged using PET scans. Highlighted areas are as follows: BAT – brown adipose tissue, H – heart, B – bladder. (b) Histology of aged (6 month old) antrum and fundus in WT and GKN1^−/− mice. (c) Histology of the small intestine, cecum and colon of WT and GKN1^−/− mice. (d) Flow cytometry analysis of small intestine and spleen cells gates on CD44⁺ and CD4⁺ (n = 4–7).

Figure 6

GKN1^−/− mice show no lean inducing phenotypes. WT and GKN1^−/− were monitored for (a) daily food intake (b) fat absorption (c) digestive efficiency (d) Blood glucose levels (e) insulin (f) body temperature (G,h) activity and (i) weight when raised under thermoneutral conditions (30 °C). *p < 0.05 (n = 2–19).

Figure 7

GKN1 regulates metabolism. Male WT and GKN1^−/− mice were individually housed in metabolic cages and monitored for CO₂ production (VCO₂) and O₂ consumption (VO₂) for 48 h with 12 hr cycles of light and dark and RER was calculated as VCO₂/VO₂. (a) Mice maintained on NCD show decreased RER during the light cycle (resting period) and increased RER during the dark cycle (active period) with (b) a lower average RER in GKN1^−/− mice (black line), compared to WT mice (blue line) during the light cycle. (c) Mice maintained on HFD showing a cyclical but blunted pattern of RER with (d) a lower average RER in GKN1^−/− mice (black line) compared to WT mice (blue line) during both light and dark cycles. The total body energy expenditure was not significantly different between WT and Gkn1^−/− mice on either the (e,f) NCD or the (g,h) HFD. In (a),(C),(e) and (g) the lines are the mean of 4 mice with error bars removed for clarity. *p < 0.05, **p < 0.01 n = 4.