Wild type (Wt), Prom1 knockout mice (Prom1^-/-), and Apc mutant mice (Apc^-/+) ranging from 1 day to 5 months old were used. Wt and 129-Apc^tm1.Δ716 mice were bred in our animal facility (27). Prom1^-/- mice were provided by Dr. Kyuson Yun (un-published). To generate double mutant Apc^-/+Prom1^-/- mice, 129-Apc^-/+ were crossed with 129-Prom1^−/− mice. All animals were maintained on a 129SvEv background (>10 generations). Pups were genotyped by PCR to determine Apc and Prom1 gene status. The wild-type Prom1 allele generated a 204-bp product, and the mutant allele formed a 329-bp product. Mice were fed an AIN-76A diet and water ad libitum, exposed to 12-h-light/12-h-dark cycles, and maintained under specific pathogen free conditions without pinworms, Helicobacter spp, or Citrobacter rodentium. We used age- and sex-matched littermate controls. All animal experiments were performed in accordance with protocols approved by the Animal Care and Use Committee of the Johns Hopkins Medical Institutions.

At the age of 4 months, Apc^-/+Prom1^-/- and Apc^-/+Prom1^+/+ female mice were euthanized by cervical dislocation. The intestines were opened, stained with methylene blue (Figure 7A), and the number of adenomas was counted as previously described (27).

Aberrant crypt foci (ACF) were induced using AOM and DSS. Six-week-old female Wt and Prom1^-/- mice (N = 5 per group) fed an AIN-76A diet and water ad libitum were treated with a single dose of AOM (2.5, 5, 7.5, or 10 mg/kg in saline, i.p.). Mice were assessed every 12 h for 5 days. Another group of Wt and Prom1^-/- mice (N = 10 per group) were treated with a single dose of AOM (5 mg/kg in saline, i.p.). Seven days after injection with AOM, the mice were given normal drinking water or drinking water containing 2% DSS for a period of 7 days followed by normal drinking water for 3 weeks. ACF in surviving mice were evaluated as previously described with minor modifications (32). Intestinal lesions were scored for the presence of inflammation, mucosal damage, and proliferation. Grade 0, normal intestine; Grade 1, mild mixed infiltration of inflammatory cells; Grade 2, moderate infiltration of mixed inflammatory cells and edema within the mucosal layer and mild epithelial crowding; Grade 3, moderate to severe inflammation, edema, mucosal disruption, and crypt proliferation (epithelial crowding, deep crypts, and thicken mucosa); and Grade 4, severe transmural inflammation, edema, mucosal disruption, and severe proliferation (thickened mucosa, elongated and branched crypts, loss of goblet cells).

At the age of 6 weeks, blood was taken from both Wt and Prom1^-/- mice. Both serum and blood were sent to Antech Diagnostics for analysis. Serum was analyzed for liver, kidney, and pancreas functions as well as muscle changes. The blood exam included total red blood cells (RBC), total white blood cells (WBC), and differential WBC counts. Thrombin, partial thrombin, hemoglobin, packed cell volume, mean corpuscular hemoglobulin (MCH), and mean corpuscular hemoglobulin concentration (MCHC) were also measured.

The adult Wt, Prom1^−/−, and Apc^-/+ mice were euthanized by cervical dislocation. The intestines were opened longitudinally, washed in PBS, and fixed in 10% buffered formalin. Then, the intestines were rolled and submitted for embedding. Five-micrometer-thick sections were prepared, and sections were deparaffinized in xylene and rehydrated through graded alcohols. Slides were transferred to a jar containing unmasking solution (Vector Laboratories, H-3300), boiled for 10 min, and left in the same solution at room temperature for 20 min. All slides were then incubated with 10% blocking serum (Vector Laboratories), in PBS for 30 min at room temperature. The slides were incubated with primary antibody (anti-Prom1, MAB4310, or with anti-Ki67, AB9260, Millipore) diluted 1:100 for 60 min at room temperature. After three washes with 0.1% Tween 20 in PBS, sections were incubated for 30 min with fluorescence (for Prom1) or non-fluorescence (for Ki67) biotinylated secondary antibody IgG (Vector laboratories) diluted 1:500 in blocking solution. Slides were washed three times in PBS for 3 min each, and rinsed three times with distilled water. Cover slips were mounted with crystal mount (Biomeda, M02). The number of positive cells within the crypts and adenomas was quantified in 10 fields at 400× magnification for Prom1 and at 200× magnification for Ki67. Photographs of histological sections were taken using a Nikon digital camera.

Western blot analyses were performed on SDS-PAGE gels under denaturing conditions. Total proteins (30 μg/ml) from both Wt and Prom1^-/- mice from intestinal crypt preparations were separated on 12 or 15% gels and electrotransferred onto nitrocellulose membranes (BioRad, 162-0115). Non-specific protein binding was blocked using 10% Blotto–0.1% Tween followed by incubation with anti-Prom1 primary antibody (Millipore, MAB4310) diluted in the blocking solution overnight at 4°C. After washing with PBS, membranes were incubated with horseradish peroxidase-conjugated anti-mouse secondary antibodies and developed using the Immobilon Western Kit (Millipore, WBKLS0100).

Real-time PCR was performed to determine Prom1 expression on the crypt cells from both Wt and Prom1^-/-. Total RNA of mouse intestinal tissues were extracted using the RNAeasy Kit (Invitrogen), and cDNA was synthesized using random hexamers. The Taqman primers for the Prom1 and K19 genes were from Applied Biosystems. We calculated relative gene expression by the ΔCT method.

Data are presented as the mean and SEM or as a percentage with SD (GraphPad, PRISM software, San Diego, CA, USA). The Mann–Whitney test was used to compare body weights, and P values were determined using the Student two-tailed t-test unless otherwise indicated. Tumor data were analyzed using the Chi square (χ²) test and other differences using the Student’s t-test. Kaplan–Meier survival curves were generated, and difference in survival was determined using the log-rank test. Data were considered to be significantly relevant at p < 0.05 and are presented as mean ± SEM.

Immunofluorescence staining for the normal epithelial cell crypts in both the small and the large intestines revealed that anti-PROM1 antibody marked rare intestinal crypt cells (Figure 1A). In the Apc^-/+ mice, that anti-PROM1 antibody showed intense staining in the majority of the epithelial cells within the adenomas (Figures 1B,C). The isotype matched control was negative (data not shown). Quantification showed an expansion of PROM1 positive cells in the adenomas, P < 0.001 (Figure 1D). Western blot analysis showed that mucosal cells from Apc^-/+ mice have an enhanced expression of PROM1 in comparison to normal crypts from wild-type mice (Figure 2A). Real-time PCR was performed on the crypt cells using K19 or Gadph (data not shown for Gadph) as controls. Results demonstrated significant increases (four- to six-fold) in Prom1 expression in the crypt epithelial cells in the Apc^-/+ mice compared to the crypts from wild-type mice (Figure 2B). These results indicate that Prom1 expression and PROM1 positive cells are increased in the mucosa of Apc^-/+ mice.

To determine the role of Prom1 in intestinal crypt, we analyzed Prom1^-/- mice. Null mice of both sexes were phenotypically normal, and crossing heterozygous male and female mice resulted in Wt, Het, and null offspring at the expected Mendelian ratios (Figure 3). Body weight analysis revealed mature obesity in Prom1^-/- mice compared to Wt mice (Figure 4A). Complete histopathological analysis revealed that there is a moderate degree of germinal arrest in the testes of Prom1^-/- mice compared to wild-type mice. There was multifocal seminiferous tubule atrophy and degeneration and arrest of spermatogenesis within the testes in the Prom1^-/- mice at 4 months of age (Figure 4C). Serum chemistry revealed a significant increase in fasting blood glucose level in Prom1^-/- mice. The mean blood glucose level was 254 mg/dl in the Prom1^-/- mice compared to 158 mg/dl in the wild-type mice (Figure 4B). The knockout did not affect hematology. There were no changes in blood scores in the Prom1^-/- mice compared to wild-type mice (data not shown).

To determine whether Prom1 has a role in maintaining intestinal epithelium homeostasis and injury response, we exposed control and Prom1^-/- to AOM and DSS. Prom1^-/- mice were more sensitive to AOM than Wt mice. After injection with a single dose of AOM (5 mg/kg), 50% of the mice died. All the Prom1^-/- mice died when given a single injection of AOM at 7.5 or 10 mg/kg (Figure 5A). Remarkably, an injection of AOM followed by oral exposure to DSS resulted in increased intestinal inflammation in Prom1^-/- mice, relative to Wt mice (Figures 5C,D). Inflammatory infiltrate consists predominantly of macrophages, neutrophils, and lesser numbers of lymphocytes and rare plasma cells. In addition to increase in inflammation, Prom1^-/- mice showed severe dysplastic crypts and abnormal crypt proliferation. Hyperplastic crypts in the knockout mice showed markedly reduced number of goblet cells, elongated nuclei with prominent nucleoli, and abnormal crypt lumen. The wild-type mice exhibited mild crypt hyperplasia with minimal dysplasia. Peripheral complete blood counts revealed significant leukocytosis, neutrophilia, and monocytosis in Prom1 knockout mice compared with the wild-type mice (Figure 5B). The proliferation index of hyperplastic crypts measured by nuclear expression of Ki67 showed a significant positive correlation to the lack of Prom1, p < 0.01 (Figure 6).

To determine whether loss of Prom1 promotes intestinal tumorigenesis, we crossed Prom1^-/- mice to Apc^-/+ mice. All sizes of adenomas increased in the double mutant mice (Apc^-/+Prom1^-/-) (Figure 7A), especially small adenomas (<2 mm in diameter) (Figure 7B). Double mutant mice exhibited the highest number of adenomas (mean, 33) compared with Apc^−/+ mice (mean, 27) along their entire small and large intestines at ~120 days of age. Average adenoma count was 23 in the small intestine and 10 in the large intestine of double mutant mice (Apc^-/+Prom1^-/-) compared to 17 and 10 in single mutant mice (Apc^-/+Prom^+/+), respectively. The number of adenomas increased by 18% in Apc^-/+Prom1^-/- mice compared to those with Apc^-/+Prom^+/+. This escalation was highly significant as mediated by a p value of <0.05 using a two-tailed Student’s t-test.