Stromal Hedgehog signalling is downregulated in colon cancer and its restoration restrains tumour growth

A role for Hedgehog (Hh) signalling in the development of colorectal cancer (CRC) has been proposed. In CRC and other solid tumours, Hh ligands are upregulated; however, a specific Hh antagonist provided no benefit in a clinical trial. Here we use Hh reporter mice to show that downstream Hh activity is unexpectedly diminished in a mouse model of colitis-associated colon cancer, and that downstream Hh signalling is restricted to the stroma. Functionally, stroma-specific Hh activation in mice markedly reduces the tumour load and blocks progression of advanced neoplasms, partly via the modulation of BMP signalling and restriction of the colonic stem cell signature. By contrast, attenuated Hh signalling accelerates colonic tumourigenesis. In human CRC, downstream Hh activity is similarly reduced and canonical Hh signalling remains predominantly paracrine. Our results suggest that diminished downstream Hh signalling enhances CRC development, and that stromal Hh activation can act as a colonic tumour suppressor.

tests, fdr-adjusted). f) Weight curves during AOM/DSS treatment of mice treated with vismodegib or vehicle (related to figure 2f). Weight was different only on day 9 (p=0.023; t-test without multiple-test correction). g) Survival curves, related to (f). Differences were not significant (log-rank p=0.12)

Supplementary Figure 5 Differences between localised and systemic activation of Col1a2CreER
Comparison of Cre-mediated recombination between inducible Col1a2CreER;R26-LSL-tdTomato reporter mice injected with corn oil (a), injected intraperitoneally (i.p.) with tamoxifen (b) or given 4OH-tamoxifen transrectally (t.r.) (c) at 10 weeks of age (related to Figure 3e). The systemic administration of 5 mg Tam led to recombination in many organs, including colon, kidney, lung, and liver. In the i.p.-treated animals, Tomato + cells were associated with kidney glomeruli, widely distributed in the lung, and located predominantly around the major bile ducts in the liver, while upon t.r. administration, recombination is most frequent in the colon; note that i.p. administration is associated with higher recombination frequency in the muscularis externa (oblique longitudinal cells in b, magnified insets). Animals were sacrificed 7 days after Tam treatment. Images representative of n=7 i.p.-treated animals, n=7 i.r.-treated mice, and n=4 oiltreated controls. Images were acquired with a Zeiss SteREO Discovery.V20 microscope with a 0.63X lens at minimum zoom (resulting in 4.7x magnification), taken with an AxioCam MR3 camera and Zeiss Zen 2012 software v 1.1.1.0.

Supplementary Figure 7
Homozygously floxed Ptch1  Col1 mice a) Representative ISH images showing Lgr5 and Gli1 expression in a) a control mouse (lacking Col1a2CreER allele) and b) a Ptch1  Col1 mouse 7d after 5 mg Tam i.p. (related to Figure 5). Arrows in left panel magnification in (b) show dots indicating Lgr5 expression. Scale bars 50/20µm. c) Relative weight of mice during two cycles of DSS; n=7 Ptch1  Col1 mice and n=9 controls (lacking Cre recombinase). Differences were significant on days 39 (p=0.037), 41 (p=0.011), and 43 (p=0.004; all t-tests without correction for multiple testing). d) Survival curves; log-rank p=0.0002. One Ptch1  Col1 mouse was found dead on day 3 after Tam treatment (t.r.); one Ptch1  Col1 mouse was sacrificed due to self-inflicted wounds (day 42, not counted as experiment-related event). The other mice were sacrificed following ethical guidelines. e) H&E staining of a colon section from a Ptch1  Col1 mouse sacrificed on day 16 (1 st cycle of DSS-colitis) due to severe weight loss and bloody diarrhoea. Note the almost complete destruction of the epithelial lining (magnification) as well as stromal proliferation (asterisks, related to panels f and g). Scale bars: 200/50 µm. f) Macroscopic appearance of a large, cystic stromal tumour (white dashed line) in a 6 month old female Ptch1  Col1 mouse, diagnosed 8 days after application of 4OH-Tam t.r. g) RNA ISH and hematoxylin staining of the tumour in (d) showing high Gli1 expression in the stromal tumour.

Supplementary Figure 8
Correlation of GLI1 expression with driver mutations and stromal gene signatures a) Correlation of GLI1 mRNA levels with the presence of KRAS, BRAF, and APC mutations; a weak correlation was observed between APC mutations and low GLI1 expression (p-values indicated above the panels). NB: For this analysis, gene expression data were extracted from the TCGA COAD Illumina RNA-seq dataset and correlated to the Illumina mutation data set, as the mutation data corresponding to the Agilent data used for all other TCGA analysis was limited to 53 samples. b) Correlation of GLI1 to stromal gene expression signatures as defined by the ESTIMATE algorithm. A strong positive correlation was observed for GLI1 and tumour stroma content ("ESTIMATE", left panel, score combines "stromal" and "immune" cell markers). This robust association appeared to be dominated by a very strong correlation between GLI1 and non-immune cell markers (middle panel, "stromal" gene expression), while genes thought to be expressed predominantly in immune cells contributed to a lesser extent (right panel, "immune" gene expression). Correlation coefficients and p-values are indicated above the panels.

Supplementary Figure 9 Invasive tumours in mice treated with AOM/DSS
Representative examples of invasive tumours from wt C57BL/6 mice at the endpoint of the AOM/DSS protocol. In a), the muscularis mucosae appears discontinuous (arrows in magnified image, right panel), while in b), malignant cells are found below the muscularis mucosae, together with desmoplastic stroma and mucus lakes, strongly indicating invasiveness. Scale bars 200/50 µm (a), 500/100 µm (b).