Fluorescently labeled endocytic cargos enable imaging of caveolar endocytosis in the intact zebrafish intestine. (A) Representative images show that the caveolar-specific cargos Alexa Fluor–albumin and BODIPY–d-LacCer are internalized from the basolateral PM of enterocytes, but not the intestinal lumen. In contrast, the cargo transported specifically by clathrin-coated vesicles, BODIPY–l-LacCer, is transported into enterocytes from both the basolateral and luminal PMs. BB, brush border; L, lateral membrane; B, basolateral membrane; I, intracellular; N, nucleus; arrowhead, intracellular puncta. (B) The mean fluorescence intensity of Alexa Fluor–albumin and BODIPY–d-LacCer on the lateral PM of enterocytes is significantly greater following basolateral injection compared to luminal injection. In contrast, the mean fluorescence intensity of BODIPY–l-LacCer on the lateral PM of enterocytes is the same following basolateral and luminal injection. Data is presented relative to lateral PM mean fluorescence intensity following luminal injection. Mean±s.e.m, n=3: nine fish per replicate, three areas of each region per fish; Student's t-test; *P<0.05.

Deletion of Cav1 from mouse intestinal epithelial cells (CAV1^IEC-KO). (A) Schematic representation of deletion of Cav1 in the intestinal epithelium. (B) PCR of genomic DNA from whole mouse jejunum shows that Cre recombination of Cav1 has occurred in CAV1^IEC-KO jejunum but not in Cav1^fl/fl WT littermates. (C) Cav1 mRNA is decreased 68% in CAV1^IEC-KO mouse jejunum as evidenced by RT-PCR (mean, Student's t-test, *P=0.01, n=10). (D) Cav2 mRNA is decreased 75% in CAV1^IEC-KO mouse jejunum as evidenced by RT-PCR (mean, Student's t-test, *P=0.01, n=10). (E,F) CAV1 protein is reduced in the jejunum of CAV1^IEC-KO mice as measured by western blot (F) and normalized to α-tubulin. (E) Data are expressed relative to Cav1^fl/fl WT CAV1 protein, n=3 western blots, five WT and five CAV1^IEC-KO mice per blot, Student's t-test; *P<0.05. (G,H) Body mass of male (G) and female (H) mice; mice were fed a low-fat (10%) or high-fat (60%) diet starting at 10 weeks (n=10-15). HFD mice had significantly higher body mass than LFD mice, but loss of intestinal epithelial cell CAV1 did not affect body mass. Mean±s.e.m, linear regression; *P<0.05.

Loss of CAV1 in the intestinal epithelia alters plasma cholesterol levels. (A) In male mice, total and esterified plasma cholesterol are elevated by HFD compared with LFD in WT, but not CAV1^IEC-KO (IKO), mice following a 4 h fast (n=5-8). (B) Conversely, for female mice, total, free, and esterified plasma cholesterol are elevated by HFD compared with LFD in 4 h fasted CAV1^IEC-KO, but not WT, mice (n=8-9). (C) Postprandial male CAV1^IEC-KO mice fed CD have greater plasma-free cholesterol mice than WT (n=6-8). Data are mean±s.e.m, two-way ANOVA, *P<0.05; groups with different brackets show an effect of diet, groups with different letters are significantly different by post hoc testing.

Lipoprotein cholesterol levels are affected by loss of CAV1 in the intestinal epithelia. (A,B) Male CAV1^IEC-KO (IKO) mice are protected from HFD-induced increase in fasted plasma LDL cholesterol (n=6-10, mean±s.e.m, two-way ANOVA, significant interaction between diet and genotype) (A), but female mice are not (n=8-9) (B). Female CAV1^IEC-KO, but not WT, mice have higher fasted plasma HDL cholesterol on HFD than LFD. Mean±s.e.m, two-way ANOVA, groups with different letters are significantly different by post hoc testing. (C) CAV1^IEC-KO mice on CD have higher postprandial plasma LDL cholesterol than WT mice. Mean±s.e.m, Student's t-test, *P<0.05, n=6-9.