Biology Reference
In-Depth Information
Dissolution of tight cell-cell junctions in the developing carcinoma is accom-
panied by the proteolytic degradation of the BM, allowing for initiation of tumor
cell invasion into the surrounding stroma. The major components of the BM include
laminins, type IV collagen, nidogen and proteoglycans (Kalluri
2003
; Rowe and
Weiss
2008
), most of which constitute multiple proteolytic targets of many MMPs
in vitro, including MMP-2 and MMP-9, the two type IV collagenases. The func-
tional contribution of MMPs to the proteolysis of epithelial BM in vivo is somewhat
restricted and sometimes is attributed to only membrane-anchored MMPs, in
particular MT1-MMP, overexpressed in various cell types, including fibrosarcoma
cells, carcinoma cells, fibroblasts, and bone marrow mesenchymal cells (Sabeh
et al.
2004
; Hotary et al.
2006
; Li et al.
2008
; Ota et al.
2009
; Rowe et al.
2009
;Lu
et al.
2010
). However, the putative degradation and remodeling of epithelial BMs
in vivo by MT1-MMP is conceptually compromised by the absence of in vivo and
even in vitro biochemical data which would demonstrate and characterize that the
specific BM components, including type IV collagen, are indeed cleaved by MT1-
MMP in situ. Nevertheless, in an ex vivo model employing epithelial BMs recov-
ered from the rat or mouse mesentery, it was the three MT-MMPs, i.e., MT1-MMP,
MT2-MMP, and MT3-MMP (MMP-16), which conferred the transfected cells with
the ability to perforate BMs (Hotary et al.
2006
). Furthermore, in the chick embryo
CAM model, invasion in vivo of various cells through the BM underlying the
chorionic ectoderm was demonstrated to solely depend on the proteolytic activity of
MT1-MMP and MT2-MMP, but not on the activity of MT3-MMP or a number of
secreted MMPs produced by the tested cells, including MMP-1, MMP-2, MMP-3,
MMP-7, MMP-9, and MMP-13 (Sabeh et al.
2004
; Hotary et al.
2006
; Li et al.
2008
; Ota et al.
2009
; Rowe et al.
2009
; Lu et al.
2010
). In particular, MT1-siRNA
targeting confirmed that expression of MT1-MMP in human HT-1080 fibrosarcoma
cells was critical for CAM invasion (Sabeh et al.
2004
). However in the same CAM
model system, the role of MT1-MMP was not demonstrated to be essential for the
ability of a highly disseminating variant of HT-1080 fibrosarcoma to penetrate the
undamaged chorionic ectoderm and epithelial BM and invade the CAM mesoderm
(Deryugina et al.
2005
). Moreover, the specific targeting (by both siRNA and
neutralizing antibody approaches) of several HT-1080-derived MMPs, such as
MMP-1, MMP-2, and MMP-9, actually facilitated HT-1080 fibrosarcoma cells to
breach the CAM and disseminate to other tissues (Deryugina et al.
2005
; Partridge
et al.
2007
), apparently indicating that some of the tumor-derived MMPs can have
protective roles in tumor progression.
Consistent with the above suggestion, data obtained in several mammalian model
systems demonstrate that exclusive expression of MT1-MMP in tumor cells might
actually be insufficient for tumor growth and proliferation in vivo. Thus, the
transplants from colon carcinoma cells, generated from MT1-MMP-deficient mice
and transfected with inducible MT1-MMP, grew slowly in MMP-2-deficient mice
despite the presence of induced MT1-MMP. However, transplantation of the same
tumor cells into MMP-2-expressing recipients or transfection of tumor cells with
MMP-2 or coimplantation of tumor cells with MMP-2-expressing fibroblasts all
resulted in a rapid growth of transplants uponMT1-MMP induction and concomitant