Biology Reference
In-Depth Information
the cells that have accomplished EMT is also closely associated with their enhanced
invasiveness in vitro and in vivo, hence evoking the functional contribution of
MMPs known to assist matrix invasion (Gilles et al.
2005
).
Among the aforementioned transcriptional factors, Snail induces expression
of MMPs capable of basement membrane (BM) degradation (Thiery et al.
2009
).
Thus, MMP-13 (collagenase 3) expression was shown to be upregulated and
maintained in the NBT-II bladder carcinoma cells exposed to FGF-1, a potent
inducer of EMT (Billottet et al.
2008
). In agreement with the suggested induction
of MMPs during EMT, soluble E-cadherin fragments induced expression of several
MMPs, including MMP-2 (gelatinase A), MMP-9 (gelatinase B), and MMP-14
(MT1-MMP), in lung carcinoma cells and promoted their invasion into chick heart
and into collagen type I gels (Nawrocki-Raby et al.
2003a
). Conversely, over-
expression of E-cadherin in highly invasive bronchial tumor cells, which would
cause dampening of EMT, resulted in decrease of
-catenin transcriptional activity,
decreased invasion, and a concomitant decrease of MMP-1 (interstitial collagenase),
MMP-3 (stromelysin 1), MMP-9, and MT1-MMP, at both mRNA and protein levels
(Nawrocki-Raby et al.
2003b
). Recently, MT1-MMP and MT2-MMP (MMP-15)
were demonstrated to cooperatively function as direct-acting, proinvasive factors
in Snail1-triggered breast carcinoma MCF-7 cells employed in the chick embryo
CAM model (Ota et al.
2009
). The expression and proteolytic activity of these two
membrane-anchored MMPs were required for the EMT invasion program induced
by Snail1, while a number of secreted MMPs, including MMP-1, MMP-2, MMP-3,
MMP-7 (matrilysin), MMP-9, and MMP-13, were dispensable and indicated to be
irrelevant in this experimental model system. However, there are also documented
cases where cancer cells, e.g., generated from spontaneously developed murine
mammary tumors, had undergone an EMT, but were still noninvasive and nonmeta-
static despite the expression of vimentin and MT1-MMP (Tester et al.
2000
).
Therefore, EMT-induced expression of MMPs might be insufficient to assure the
metastatic potential of cancer cells in vivo.
Alternative to the notion that MMPs are induced after the onset of EMT, some
MMPs were shown to function as stabilizers or even triggers of EMT. However, the
evidence for a
proteolytic
role of MMPs in the
induction
of EMT which would
facilitate metastasis in vivo remains rather circumstantial. Thus, expression of
MMP-3 in the mammary epithelial cells in transgenic mice resulted in the forma-
tion of invasive mesenchymal-like tumors (Sternlicht et al.
1999
). On the other
hand, MMP-3 null animals demonstrate an increased sensitivity to the development
of squamous cell carcinoma and, conversely, overexpression of MMP-3 specifically
targeted to keratinocytes reduces tumor multiplicity in transgenic mice (McCawley
et al.
2008
). This apparent discrepancy highlights contrasting roles of the same
tumor MMP depending on the experimental setting or specifics of mammary gland
tumorigenesis (Almholt et al.
2007
) or skin carcinogenesis (Martin and Matrisian
2007
). In vitro, treatment of SCp2 mouse mammary epithelial cells with exogenous
MMP-3 caused a loss of intact E-cadherin, increased motility and invasiveness,
downregulation of epithelial markers, and upregulation of mesenchymal markers,
including Snail1 (Przybylo and Radisky
2007
). Furthermore, proteolytic disruption
b
