Result 1: MRCKa knockout mice is partially embryonic lethal
Prior to carrying out studies on the role of MRCKa in cancer, I characterized the first MRCKa knockout mice generated by our lab (Figure 1a-b). Crosses of heterozygous mice revealed that about 50% of the ko mice were lost during embryonic development (Figure 1c). This suggests a critical role for MRCKa in embryogenesis. However, the surviving ko pups lived to adulthood with no adverse effects on growth and development, for example in body weight (Figure 1d). Based on our findings, further investigations can be carried out to investigate the role of MRCKa in development.
Result 2: MRCKa is not essential for breast tumor formation and metastasis in vivo
To fulfil objective 1, which was to test for breast cancer development and metastasis in MRCKa knockout mice, I utilized a MMTV-PyMT MRCKa ko spontaneous breast cancer mouse model to investigate tumor development and lung micrometastasis. My data indicate that loss of MRCKa did not affect survival (Figure 2a), weeks tumor-free (Figure 2b), volume (Figure 2c) and tissue invasion (Figure 2d).
In addition, I observed similar numbers of lung metastasis (Figure 3a-b). Investigating breast cancer invasion in cancer organoids also showed comparable invasion of both control and ko tumors (Figure 3c). These data suggest that inhibition of MRCKa alone is not sufficient to inhibit metastasis in breast cancer.
Result 3: MRCKa and MRCKb functions redundantly in signaling
MRCKa and its related kinase MRCKb share >80% homology at the kinase domains. To investigate whether the lack of contribution of MRCKa in cancer metastasis in vivo was due to MRCKb compensation, I used a CRISPR/Cas9 strategy to delete either or both proteins in MDA-MB-231 triple negative breast cancer cells. (Figure 4a). Knockout of MRCKa alone resulted increase in MRCKb expression (Figure 4b-c). Interestingly, this correlated with a trend towards increased phosphorylated myosin light chain (p-MLC) and cofilin (p-cofilin) (Figure 4d –f) and F-actin (Figure 4g-h). These data indicate a complex crosstalk between MRCKa and MRCKb.
Result 4: MRCKa and MRCKb redundantly regulate MDA-MB-231 cell migration and invasion in vitro
As an extension to the above finding, I investigated whether MRCKa and MRCKb contribute to 3D cell migration. In a mixed-spheroid 3D collagen matrix invasion assay, where the MRCK single and dko cells were mixed with cancer-associated fibroblasts (CAFs) as a spheroid for analysis, MRCK dko spheroids invaded significantly less than in control (Figure 5a-b). Also, MRCKa, but not MRCKb ko spheroids showed defective 3D migration. In summary, I showed that MRCKa and MRCKb function redundantly to promote invasion.
Result 5: MRCKa and MRCKb cooperate with ROCK to promote 3D migration in 4T1 cells
To check whether my results were cell type-specific, the experiments were repeated in 4T1 triple negative breast cancer cells (Figure 6a), which utilize a collective migration strategy instead of the mesenchymal, single-cell migration in MDA-MB-231. Importantly, my results show that neither dko nor single MRCK ko spheroids showed significant reduction in invasion in 3D collagen matrix (Figure 6b). I further investigated whether ROCK kinase, another closely related class of Rho GTPase effector, contributes to 4T1 cell migration. Surprisingly in the presence of a ROCK inhibitor Y-27632, 3D collagen matrix invasion in control 4T1 spheroids was instead increased (Figure 6b). This was observed to be due to a change in migration strategy, from the more cohesive collective migration, to a mechanism resembling single-cell migration (Figure 6c). As this increase in invasion was prevented by MRCKb and dko, my results suggest that MRCK and ROCK signalling contribute to the 3D invasion in 4T1.
