18
CHD4, leads to plasmin-dependent disruption of vascular integrity in both bloodderived and lymphatic EC-lined vessels during vascular development [22, 59].
Of great interest is that the MMP-1 and MMP-10 regression phenomena are
strongly abrogated in our in vitro model when pericytes are added along with the
ECs [87]. Pericyte recruitment to the tubes occurs, and they become much more
resistant to pro-regressive stimuli. Thus, our model mimics that observed in vivo
where EC tubes without pericytes are much more susceptible to regression [5, 13,
14]. Interestingly, tumor vessels have associated pericytes, but the interactions are
abnormal. Despite these abnormalities the tumor vessels persist, which in this case
may relate to the fact that many aggressive tumors overproduce TIMP-1 [101],
which can interfere with the MMP-1 and MMP-10-dependent regression system
[33]. We also observed that EC-pericyte interactions upregulate the production of
EC TIMP-2 and pericyte TIMP-3 which together control how EC-pericyte interactions protect against pro-regressive MMP-1 and MMP-10, and also ADAM-15 [87].
Pericytes are a rich source of TIMP-3 [87] which is interesting because of its ECMbinding ability (i.e., ability to bind cell surfaces and basement membrane matrices)
and its dual ability to inhibit soluble and membrane MMPs, as well as membrane
ADAM proteinases [7]. In addition to interfering with pro-regressive stimuli,
TIMP-3 and TIMP-2 block MT1-MMP, to interfere with further EC tube formation.
Thus, these TIMPs contribute to vascular tube stabilization by inhibiting both vascular tube regression phenomena and further vascular tube morphogenesis.
Additional support for this their latter influence, both TIMP-2 and TIMP-3 are
antagonists of VEGFR2 [80, 90], an important EC signaling receptor. Interestingly,
we have described the concept that VEGF does not directly stimulate vascular tube
morphogenesis but that it primes ECs so that they respond in a more robust manner
to a distinct set of downstream growth factors (see later on). Using this reasoning,
TIMP-2 and TIMP-3 might be capable of suppressing EC priming and, thus,
enhance EC tube stability in this manner.
1.3.11 Critical Functional Role for EC-Generated Vascular
Guidance Tunnels During Blood Vessel Assembly in 3D
Matrices
As shown in Fig. 1.2, ECs utilize a lumen signaling complex to form tubes in 3D
matrices while at the same time generating networks of vascular guidance tunnel
spaces. ECs are able to migrate through these spaces in an MMP-independent manner. We have also shown that ECs can regrow within these spaces following tube
collapse. EC-lined tubes were treated with thrombin which reversibly causes tube
collapse, leaving rounded up ECs within tunnel spaces [95]. After inhibition of
thrombin with the thrombin inhibitor hirudin, the ECs regrow within tunnels to reassemble the collapsed tube [95]. Thus, pre-existing tunnel spaces are matrix conduits
that allow for rearrangement of tubes and migration of ECs, and, thus, they are used
G. E. Davis
CHD4, leads to plasmin-dependent disruption of vascular integrity in both bloodderived and lymphatic EC-lined vessels during vascular development [22, 59].
Of great interest is that the MMP-1 and MMP-10 regression phenomena are
strongly abrogated in our in vitro model when pericytes are added along with the
ECs [87]. Pericyte recruitment to the tubes occurs, and they become much more
resistant to pro-regressive stimuli. Thus, our model mimics that observed in vivo
where EC tubes without pericytes are much more susceptible to regression [5, 13,
14]. Interestingly, tumor vessels have associated pericytes, but the interactions are
abnormal. Despite these abnormalities the tumor vessels persist, which in this case
may relate to the fact that many aggressive tumors overproduce TIMP-1 [101],
which can interfere with the MMP-1 and MMP-10-dependent regression system
[33]. We also observed that EC-pericyte interactions upregulate the production of
EC TIMP-2 and pericyte TIMP-3 which together control how EC-pericyte interactions protect against pro-regressive MMP-1 and MMP-10, and also ADAM-15 [87].
Pericytes are a rich source of TIMP-3 [87] which is interesting because of its ECMbinding ability (i.e., ability to bind cell surfaces and basement membrane matrices)
and its dual ability to inhibit soluble and membrane MMPs, as well as membrane
ADAM proteinases [7]. In addition to interfering with pro-regressive stimuli,
TIMP-3 and TIMP-2 block MT1-MMP, to interfere with further EC tube formation.
Thus, these TIMPs contribute to vascular tube stabilization by inhibiting both vascular tube regression phenomena and further vascular tube morphogenesis.
Additional support for this their latter influence, both TIMP-2 and TIMP-3 are
antagonists of VEGFR2 [80, 90], an important EC signaling receptor. Interestingly,
we have described the concept that VEGF does not directly stimulate vascular tube
morphogenesis but that it primes ECs so that they respond in a more robust manner
to a distinct set of downstream growth factors (see later on). Using this reasoning,
TIMP-2 and TIMP-3 might be capable of suppressing EC priming and, thus,
enhance EC tube stability in this manner.
1.3.11 Critical Functional Role for EC-Generated Vascular
Guidance Tunnels During Blood Vessel Assembly in 3D
Matrices
As shown in Fig. 1.2, ECs utilize a lumen signaling complex to form tubes in 3D
matrices while at the same time generating networks of vascular guidance tunnel
spaces. ECs are able to migrate through these spaces in an MMP-independent manner. We have also shown that ECs can regrow within these spaces following tube
collapse. EC-lined tubes were treated with thrombin which reversibly causes tube
collapse, leaving rounded up ECs within tunnel spaces [95]. After inhibition of
thrombin with the thrombin inhibitor hirudin, the ECs regrow within tunnels to reassemble the collapsed tube [95]. Thus, pre-existing tunnel spaces are matrix conduits
that allow for rearrangement of tubes and migration of ECs, and, thus, they are used
G. E. Davis