Mouse models
The intestinal epithelium-specific Villin-creERT2 (ref. 61) mouse line was intercrossed with various conditional tumour driver models: Kras(lsl)G12D/WT (KrasG12D)62, Pik3ca(lsl)H1047R/WT (Pik3caH1047R)63, ApcΔ14/WT (Apchet)64, Trp53Δ2-10/Δ2-10 (Trp53null)65, Trp53Δ2-10/WT (Trp53het), PtenΔ5/Δ5 (Ptennull)66, Fbxw7Δ5/Δ5 (Fbxw7null)67, R26-Notch1ic/WT (N1-ICDhet)68, Arid1aΔ8/Δ8 (Arid1anull)69, and R26-CAG-Brainbow2.1/Confetti (R26(lsl)Confetti/WT)70,71. Lgr5-DTR-eGFP72 mice were used to study the effects of ENU mutagenesis on Lgr5-positive cells. Mice were from mixed backgrounds with high C57Bl/6J inheritance. Genotyping was done by Transnetyx.
Animal husbandry
All experiments used male and female mice of at least 6 and 8 weeks of age, respectively. Mice were housed under controlled conditions (temperature 21 ± 2 °C, humidity 55 ± 10%, under a 12 h–12 h light–dark cycle) in individually ventilated cages in a specific-pathogen-free facility (tested according to the recommendations for health monitoring by the Federation of European Laboratory Animal Science Associations). Food and water were provided ad libitum. None of the mice were involved in any procedure before the study. For survival curve generation, mice were aged until they showed clinical signs of tumour burden (anaemia, hunching and loss of body condition). All of the animal experiments were performed in accordance with the guidelines of the UK Home Office under the authority of project licence PD5F099BE, approved by the Animal Welfare and Ethical Review Body at the CRUK Cambridge Institute, University of Cambridge.
Field induction and ENU mutagenesis
Complete intestinal field induction was achieved by a single intraperitoneal injection of 4 mg (3 mg for mice <20 g in weight) of tamoxifen (Merck T5648) dissolved in ethanol and sunflower oil (Merck S5007) at a 1:9 ratio (20 mg ml−1 stock solution). A lower tamoxifen dose (0.15 mg) was used for KrasG12D mice in the sequencing cohort to reduce tumour multiplicity and therefore random tumour collisions. We resuspended ENU (Merck N3385) in 5 ml 95% ethanol/45 ml phosphate/citrate buffer at a final concentration of 20 mg ml−1 (ref. 73) and injected it intraperitoneally at 200 mg per kg.
Mouse dissection
Survival data and most tumour counts were generated from mice culled at the humane end point. Timed culls before this end point were used to assess tumour initiation kinetics in the control cohort and to allow tumour counts in high tumour density models (KrasG12D cohort). The pancreas, mesenteric lymph nodes, liver and lung were dissected and placed in formalin solution overnight at room temperature then exchanged to 70% ethanol for histological assessment of metastasis. Other organs were also collected and processed if relevant pathology was observed during dissection. Intestines were dissected, longitudinally opened, whole-mounted and fixed in 4% PFA overnight at 4 °C in an orbital shaker. Tumours were counted and either excised and processed for genomic DNA (gDNA) extraction or Swiss-rolled for histopathological assessment.
Intestinal tumour count and microscopy
Tumours in whole-mounted intestines were counted under a dissection microscope. Proximal SI was defined as the first 16–20 cm, comprising SI10 and SI20. Distal SI was defined by the remaining length, comprising SI30, SI40 and, sometimes, SI50. The colon location comprised the caecum and proximal and distal colon. Apchet (SI + colon) and KrasG12D (colon) whole-mounts were sliced into 2–3 cm pieces, optically cleared using the CUBIC1a protocol74 for 7 days at 37 °C with solution change every other day, stained with DAPI (1:1,000; Thermo Fisher Scientific, D1306), and refractive-index-matched in Rapiclear 1.52 (Sunjin Lab, 152002) for 2 days before mounting in 1 mm iSpacers (Sunjin Lab). Representative tissue pieces for each region were imaged throughout the whole thickness in a Leica SP5 TCS confocal microscope (×10/0.4 NA objective, 8–11 µm z-step, ×1.5 optical zoom). Tumours were manually counted using Fiji image analysis software75 and the total tumour count was extrapolated.
... continue reading