CSCs have been shown to promote tumorigenesis in numerous cancer types 
, and recent work has begun to define a role for CSCs in cancer metastasis as well 
. Despite the established therapeutic importance of targeting metastasis and a growing understanding of the role CSCs play in metastasis, to date no functional molecular regulators have been identified that promote aggressive, metastatic BCSC behavior. Here we identify RhoC as a potential promoter; one that is both necessary and sufficient for SUM149 and MCF-10A BCSC metastasis.
By approaching RhoC expression from two distinct angles–its necessity for metastasis of a breast cancer cell line (SUM149), and its sufficiency to induce metastasis of a non-metastatic mammary epithelial cell line (MCF-10A)–we were able to clearly elucidate a putative role of RhoC in BCSC metastasis. It is important to note that we used an orthotopic xenograft system rather than an intracardiac or tail vein injection assay to measure metastasis. As emphasized in a recent publication 
, the orthotopic xenograft model of spontaneous breast cancer metastasis is a more physiologically relevant model of breast cancer metastasis, which more accurately recapitulates the microenvironmental obstacles metastatic cells encounter in human patients.
We have shown that RhoC expression and BCSC marker expression are intimately linked in multiple ways (, ). Taken together, these data reveal a close association between RhoC and SUM149/MCF-10A BCSCs; one in which RhoC expression determines SUM149/MCF-10A BCSC metastatic potential and may also contribute to BCSC frequency within the cell lines. Furthermore, we demonstrated that this relationship between RhoC and the BCSC marker ALDH1 persists in a heterogeneous patient population, suggesting that RhoC may indeed be linked to BCSCs beyond the SUM149 and MCF-10A cell lines. This evidence supports the theory that a larger BCSC population–and thus higher RhoC expression–may confer a worse prognosis 
We were surprised to find that a significant number of mice developed pleural metastases independent of primary tumor formation ( and ). We initially hypothesized that these metastases may have resulted from improper injection into the mammary gland; however, identical metastases were observed in both tumor-bearing and tumor-free mice, which strongly suggests that the metastases in tumor-free mice did not result from improper injection.
As an alternative explanation, we propose that high RhoC expression–either as a consequence of inherent tumor cell biology (SUM149) or genetic modification (MCF-10A G14V), and amplified by ALDH (+) status–may, in some cases, cause a sublimation of cell behavior from non-tumorigenic directly to metastatic. Such a phenomenon is observed clinically and is defined as cancer of unknown primary site, or CUP 
. Given that RhoC primarily mediates motility, invasion, and angiogenesis 
–all metastasis-associated properties–it stands to reason that RhoC may be capable of driving metastatic progression independent from primary tumor formation.
Interestingly, several mice injected with non-CSC ALDH (−) MCF-10A G14V still developed lung metastases (). That these cells were metastatic further supports RhoC sufficiency to induce metastasis. Although ALDH (−) MCF-10A G14V cells had lower RhoC expression, RhoC was not completely eliminated ()–as is to be expected from a cell line forcibly overexpressing a transgene. Furthermore, the residual RhoC is predominately constitutively active RhoC, thus amplifying the effects of even low expression levels. Therefore, the fact that several ALDH (−) MCF-10A G14V mice developed metastases is not surprising and supports the hypothesis that RhoC may be able to promote metastasis independent of BCSC status–although, under normal conditions, RhoC expression remains closely associated with the BCSC population.
In agreement with this assertion, we also observed metastasis in one mouse injected with ALDH (+) SUM149 shRhoC cells. As we previously observed, RhoC is preferentially expressed by the ALDH (+) SUM149 population (), and although we achieved significant RhoC knockdown, RhoC expression was not completely eliminated from SUM149 shRhoC cells (). Accordingly, the remaining RhoC was primarily confined to the ALDH (+) population (). Again, though correlative, this evidence supports the hypothesis that RhoC expression is both necessary and sufficient for SUM149 BCSC metastasis and is intimately linked to the BCSC population.
These observations raise important questions about the true meaning of CSC identity. Labeling a cell a CSC may indicate that it has a specific collection of features (i.e. unlimited replication potential, increased metastatic potential, and others), but ultimately these features are a product of the genetics of the CSC. Extending from this assumption, one can reason that targeting the specific molecular cogs driving the CSC machinery–rather than focusing on incidental markers that delineate CSCs–may have therapeutic potential. Eliminating CSCs may ultimately be necessary to cure certain cancers, but disrupting the molecular CSC machinery may be able to manage the CSC population in the interim.
The work presented here provides strong rationale for therapeutically targeting RhoC. RhoC was previously shown to be essential for metastasis 
and is overexpressed in many different cancers 
–in particular IBC, which presently lacks effective therapies 
–yet this is the first work relating RhoC to BCSCs. To this end, our lab has designed a small molecule RhoC inhibitor that has shown good in vitro
and in vivo
efficacy with no apparent toxicity (unpublished data). As therapies targeting CSCs emerge 
, it will be important to address which CSC population is being targeted–the tumorigenic or the metastatic population–in order to effectively combat the disease.