Scap is required for sterol synthesis and crypt growth in intestinal mucosa

Abstract

SREBP cleavage-activating protein (Scap) is an endoplasmic reticulum membrane protein required for cleavage and activation of sterol regulatory element-binding proteins (SREBPs), which activate the transcription of genes in sterol and fatty acid biosynthesis. Liver-specific loss of Scap is well tolerated; hepatic synthesis of sterols and fatty acids is reduced, but mice are otherwise healthy. To determine whether Scap loss is tolerated in the intestine, we generated a mouse model (Vil-Scap(-)) in which tamoxifen-inducible Cre-ER(T2), a fusion protein of Cre recombinase with a mutated ligand binding domain of the human estrogen receptor, ablates Scap in intestinal mucosa. After 4 days of tamoxifen, Vil-Scap(-) mice succumb with a severe enteropathy and near-complete collapse of intestinal mucosa. Organoids grown ex vivo from intestinal crypts of Vil-Scap(-) mice are readily killed when Scap is deleted by 4-hydroxytamoxifen. Death is prevented when culture medium is supplemented with cholesterol and oleate. These data show that, unlike the liver, the intestine requires Scap to sustain tissue integrity by maintaining the high levels of lipid synthesis necessary for proliferation of intestinal crypts.

Keywords: Niemann-Pick C1-like 1 protein; SREBP cleavage-activating protein; cholesterol/biosynthesis; fatty acid/synthesis; gene expression; nuclear receptors/ sterol regulatory element-binding protein; organoid, intestine.

Fig. 1.

Time course of inducible disruption of Scap in intestinal epithelia. A: Scap^flox/flox; Villin-Cre-ER^T2 transgenic (Vil-Scap⁻) mice (female, 14–17 weeks of age, four mice per time point) were orally gavaged with tamoxifen, up to a maximum of four daily doses, as shown in supplementary Fig. 1. Small intestines were harvested without tamoxifen dosage (0 days of tamoxifen exposure), or at 1, 2, 2.5, 3, or 4 days after the initial dose of tamoxifen. Total RNA was prepared from isolated IECs, and Scap mRNA was measured by QPCR using cyclophilin as the invariant control. The mean mRNA value at day 0 is arbitrarily defined as 1.0. Filled circles indicate the value of individual mice, and the black line indicates the mean. B: Scap^flox and Vil-Scap⁻ mice (female, 14–15 weeks of age, four per group) were administered three doses of tamoxifen, and then tissues were collected at day 2.5 after the initial dose. Total RNA was isolated from liver and from IECs from proximal small intestine (Prox. SI), distal small intestine (Dist. SI), and colon. Scap mRNA was measured by QPCR as described in A. The fold expression in Vil-Scap⁻ tissues is expressed as the ratio of Scap mRNA values of Vil-Scap⁻ to Scap^flox mice, which were arbitrarily defined as 1.0. Bars indicate the mean of four mice per tissue.

Fig. 2.

Disruption of intestinal Scap rapidly produces a severe enteropathy. A: Scap^flox and Vil-Scap⁻ mice (male, 10–12 weeks of age, 10 mice per group) were administered tamoxifen once daily for four doses as indicated by black arrows. Mice were euthanized and dissected on day 4 after the initial dose of tamoxifen and body weights measured. B, C: Gross appearance of viscera in situ and dissected gastrointestinal tracts from Scap^flox and Vil-Scap⁻ are shown. A green arrow indicates a fluid-filled small intestine. Yellow arrows indicate a dilated, food-filled stomach and shrunken, fluid-filled appendix. Statistical significance was assessed by the two-tailed Student’s t-test (A). * P < 0.05; ** P < 0.001.

Fig. 3.

Destruction of villus and crypt structures in Vil-Scap⁻ mice. Representative histologic sections of mid-small intestine from 8-week-old male Vil-Scap⁻ mice that were administered tamoxifen as described in supplementary Fig. 1. Intestines were harvested on day 0 (without tamoxifen administration) or on day 3 or 4 after the initial dose of tamoxifen. Magnification, 20×.

Fig. 4.

Diminished SREBP processing in Scap-deficient small intestine. Scap^flox and Vil-Scap⁻ mice (male, 14–18 weeks of age, six mice per group) were administered three doses of corn oil vehicle or tamoxifen, and tissues were collected at day 2.5 after the initial dose of tamoxifen (supplementary Fig. 1). A: Immunoblot analysis. IECs were fractionated, and equal amounts of protein from each mouse were pooled; aliquots of the pooled membranes (75 μg) and nuclear extract (50 μg) were subjected to SDS-PAGE and immunoblot analysis. The precursor and nuclear forms of SREBPs are denoted as P and N, respectively. Immunoblots of CREB and calnexin were used as loading controls for the nuclear extract and membrane fractions, respectively. B: Relative mRNA levels in tamoxifen-dosed control and Vil-Scap⁻ mice (the same mice as in lanes 2 and 3 in A). Total RNA from IECs was separately isolated; equal amounts of RNA from each mouse were pooled and subjected to QPCR. Each value represents the mean ± SEM of data from six Vil-Scap⁻ mice relative to that of Scap^flox mice, which was arbitrarily defined as 1.0. * P < 0.01 denotes the level of statistical significance (two-tailed Student’s t-test) between Scap^flox and Vil-Scap⁻ mice. Tg, transgenic.

Fig. 5.

Decreased sterol synthesis in Scap-deficient small intestine. Scap^flox and Vil-Scap⁻ mice (male, 11–12 weeks of age, seven per group) were administered tamoxifen as described in Fig. 4. On day 2.5, mice were injected intraperitoneally with ³H-labeled water (50 mCi in 0.2 ml of saline). One hour later, tissues were removed and digitonin-precipitable sterols (A) or fatty acids (B) were isolated and subjected to scintillation counting. The small intestine was divided into two segments of equal length, proximal (Prox. S. Int.) and distal (Dist. S. Int.). Each bar represents mean ± SEM of data from seven mice. * P < 0.05, ** P < 0.01 denotes the level of statistical significance (two-tailed Student’s t-test) between Scap^flox and Vil-Scap⁻ mice.

Fig. 6.

Scap-deficient small intestinal organoids are nonviable unless cultured with cholesterol and oleate. Intestinal crypts from three Vil-Scap⁻ mice were isolated, embedded in matrigel, and cultured in independent experiments. Data from the crypts of one mouse are shown in A and from two mice in B. Organoids were grown for 4 days, at which time the wells were supplemented with 10 μM 4-OHT, 50 μM MβCD-cholesterol, and/or 100 μM oleate as indicated. Organoids not receiving MβCD-cholesterol or 4-OHT received an equal concentration of vehicle (MβCD and/or ethanol, respectively). A: Representative organoids imaged under bright-field microscopy (magnification, 10×) on the seventh day in culture. Bar indicates 200 μm. B: Cell viability was measured by ATP luminescence on the eighth day in culture. Filled circles indicate the values from two independent experiments; gray bars indicate the mean. The mean value of organoids not receiving 4-OHT or lipid supplements is arbitrarily defined as 1.0.

Fig. 7.

Time-lapse imaging of Scap-deficient organoids. Intestinal crypts from a Vil-Scap⁻ mouse were isolated, matrigel-embedded, and grown for 4 days, when lipids and 4-OHT were added as described in Fig. 6. Single organoids were visualized under bright-field microscopy (magnification, 20×) and photographed every 20 min for the following 96 h. Time-lapse videos can be seen in supplementary Videos 1–4. Still images taken every 24 h from the time-lapse sequences are shown. Bar indicates 200 μm.

Fig. 8.

Cholesterol suppresses SREBP processing and obviates nuclear SREBP requirement in Scap-deficient organoids. Intestinal crypts from three Vil-Scap⁻ mice were separately isolated, embedded in matrigel, and grown for 5 days, at which time half of the wells were supplemented with 50 μM MβCD-cholesterol and 100 μM oleate. 4-OHT (10 μM) was then added, and organoids were harvested 1.5 or 3 days later. Wells not receiving 4-OHT instead received ethanol vehicle and are indicated (0 days 4-OHT). Organoids were recovered from matrigel on the eighth day in culture. A: Immunoblot analysis. Organoids from five wells per mouse were pooled, and whole-cell extracts prepared. Aliquots of pooled protein (45 μg) were subjected to SDS-PAGE and immunoblot analysis. The precursor and nuclear forms of SREBPs are denoted as P and N, respectively. Immunoblot of calnexin was used as a loading control. B: Cell viability as estimated by ATP luminescence. C: Relative mRNA levels. Organoids from five wells per mouse were pooled, and total RNA was isolated and subjected to QPCR using cyclophilin as the invariant control. In B and C, filled circles indicate the value of organoids from each individual mouse, and bars indicate the mean value. Values from organoids receiving neither 4-OHT nor MβCD-cholesterol plus oleate were arbitrarily defined as 1.0. QPCR data from organoids that received 4-OHT for 3 days without lipid supplementation are not shown, as no RNA was recovered.