Biology Reference
In-Depth Information
The requirement of MT1-MMP in angiogenesis in vivo is more difficult to
validate since MT1-MMP knockout mice are not compromised in vasculogenesis
during embryogenesis and do not live long enough to allow for studying tumor-
induced angiogenesis (Holmbeck et al.
1999
). The lack of MT1-MMP in null mice
results in severe impaired vascular invasion of cartilage during early postnatal
development (Zhou et al.
2000
). In addition, MT1-MMP-deficient mice do not
exhibit angiogenic response to FGF-2 in a corneal model (Zhou et al.
2000
). The
abnormal vessel wall morphology in the brain of MT1-MMP-deficient mice is
associated with a significant reduction in vessel coverage with mural cells (i.e.,
pericytes and vascular smooth muscle cells), rather than with direct abnormalities
in the endothelial cells (Lehti et al.
2005
). Although MT1-MMP appears to regulate
mural cell
coverage
of brain microvasculature, pericyte
recruitment
to the sites of
tumor neovascularization requires the proteolytic activity of MMP-9. Thus, the
angiogenic vasculature in orthotopic or s.c. human neuroblastoma xenografts in
MMP-9 knockout mice displays fewer and smaller blood vessels and significant
decrease in pericytes along microvessels, indicating that stroma-derived MMP-9
contributes to tumor angiogenesis and vessel maturation (Chantrain et al.
2004
).
These architectural defects in tumor vasculature observed in the MMP-9 defic-
ient microenvironment could be rescued by transplantation of bone marrow from
wild-type donors, and therefore it is stromal MMP-9 produced by bone marrow-
derived cells (BMDCs), namely infiltrating leukocytes, that critically contributes
to the formation of a mature vasculature with efficient pericyte coverage (Jodele
et al.
2005
). Furthermore, wild-type bone marrow derived myelomonocytic cells
contribute to tumor growth and vasculogenesis specifically by supplying their
MMP-9 to the MMP-9-deficient microenvironment of knockout recipients (Ahn
and Brown
2008
).
7.8
Intravasation: MMP-Dependent Proteolysis of
the Basement Membrane at the Abluminal Surface
The intravasation step is the least studied in the metastatic cascade, possibly because
there are basically no available in vitro models and only a few in vivo models for this
complex step. Although there is evidence that tumor cells might shed passively into
the blood or lymphatic vessels (Wong and Hynes
2006
; Bockhorn et al.
2007
), in
general intravasation is viewed as an active process whereby tumor cells escaping
from the primary tumor enter the circulation. To accomplish this, intravasating
cells must first breach the BM of blood or lymphatic vessels and then traverse the
endothelial cell layer. It is assumed that breach points in the vascular BM are
proteolytically created by MMPs or other proteases produced by tumor cells or
tumor-activated stromal cells, inflammatory cells, and angiogenic endothelial
cells. The basic components constituting vascular BM include laminins, type IV
and XVIII collagens, and nidogen-1 (Kalluri
2003
; Davis and Senger
2005
;
Rowe and Weiss
2008
), relative representation of which is different in the BM of
