Extended Data Fig. 9: RAP2 deletion selectively promotes MCF7 malignancy at low stiffness in vivo.
a, Western blot showing LOX overexpression in NIH3T3 fibroblasts. Ectopic LOX expression promotes cross-linking of ECM proteins and thus increases matrix stiffness28. Image representative of two independent experiments with similar results. b, RAP2 is involved in the xenograft growth enhancement caused by LOX-overexpressing fibroblasts. Xenografts were generated by co-injection of 0.4 × 106 fibroblasts (control and Flag–LOX-overexpressing) and 2.0 × 106 MCF7 cells. The xenografts were removed on day 6 and examined for collagen deposition and crosslinking and localization of YAP and TAZ. LOX overexpression led to a woven-like structure of collagen in the xenografts based on Sirius Red staining (red colour for collagen) and Masson staining (blue colour for collagen), and increased nuclear YAP and TAZ. Images representative of three biologically independent experiments with similar results. Scale bars, 50 µm. c, LOX-induced tumour growth requires RAP2. Xenografts were generated by co-injection of 0.4 × 106 NIH3T3 cells and 2.0 × 106 MCF7 cells. The growth of the xenografts with different combinations of NIH3T3 and MCF7 cells is shown as mean ± s.e.m. Deletion of RAP2A, RAP2B and RAP2C promoted MCF7 tumour growth and masked the enhancement induced by co-injected LOX-expressing fibroblasts. **two-way ANOVA test (wild-type MCF7 + wild-type NIH3T3 cells versus wild-type MCF7 +LOX-overexpressing NIH3T3 cells), n = 6 biologically independent samples, P = 0.0027. ***two-way ANOVA test (sgRAP2 MCF7 + wild-type NIH3T3 cells versus wild-type MCF7 + wild-type NIH3T3 cells), n = 6 biologically independent samples, P = 0.002. d, Tumour weights (mean ± s.e.m.). *two-tailed t-test, n = 6 biologically independent xenografts, P = 0.014 (wild-type MCF7 + wild-type NIH3T3 cells versus wild-type MCF7 + LOX-overexpressing NIH3T3 cells) or 0.029 (sgRAP2 MCF7 + wild-type NIH3T3 cells versus wild-type MCF7 + wild-type NIH3T3 cells). e, The elasticity or stiffness of the ‘soft’ and ‘stiff’ semi-synthetic hyaluronan-derived gels was measured by atomic force microscopy. Results presented as mean ± s.e.m. The measurements were made more than 55 times for each stiffness at each time. f, RAP2 inhibits expression of the YAP and TAZ target genes CTGF and CYR61 at low stiffness but not at high stiffness. Quantitative real-time PCR analyses of CTGF and CYR61 in wild-type and sgRAP2 MCF7 cells (as in Extended Data Fig. 1g) cultured in vitro for 48 h in soft or stiff hyaluronan gel. Relative mRNA levels presented as mean ± s.e.m. n = 3 biologically independent samples, two-tailed t-test. For CTGF, *sgRAP2 versus wild-type at 0.4 kPa, P = 0.020; **0.4 kPa versus 9.0 kPa for wild-type, P = 0.0032. For CYR61, *sgRAP2 versus wild-type, P = 0.025; **0.4 kPa versus 9.0 kPa for wild-type, P = 0.0033. g, Xenograft tumours. For wild-type cells grown at 0.4 kPa, eight independent xenografts were initially generated in nude mice. However, owing to animal deaths, only six xenografts were recovered. For wild-type cells grown at 9 kPa, nine independent xenografts were generated. For sgRAP2 cells grown at both 0.4 kPa and 9 kPa, eight independent xenografts were generated. h, RAP2 deletion preferentially promotes MCF7 malignancy at low stiffness. MCF7 xenografts stained with haematoxylin and eosin. Wild-type cells embedded in 0.4 kPa matrix produced mostly tubular and some cribriform structures whereas those embedded in 9 kPa matrix produced mostly solid nests, as well as more marked cellular pleomorphism and nuclear atypia. sgRAP2 cells showed more malignant architecture and morphology at 0.4 ka, while at high stiffness (9 kPa) sgRAP2 and wild-type cells exhibited similar morphology. Images representative of three biologically independent experiments with similar results. Scale bars, 50 µm. i, Immunohistochemistry for YAP and TAZ in xenografts. RAP2 deletion increased nuclear YAP and TAZ. Images representative of three biologically independent experiments with similar results. Scale bars, 50 µm.
