Promoting vessel stabilization : toward a safe mode of therapeutic angiogenesis

Blood vessel growth is a complex process that requires the coordinated interactions of many signaling pathways and different cell types. However, despite this molecular complexity, the whole cascade of events can be set in motion by a single factor, VEGF, that is the most important regulator of both physiological and pathological angiogenesis and its over-expressed has been investigated in several ischemic tissues with the aim to restore blood flow supply. However, VEGF gene delivery presents two main limitations that make it challenging to achieve therapeutic angiogenesis in a clinical setting:
1) VEGF can induce normal or aberrant angiogenesis depending strictly on its expression level in the microenvironment around each producing cell in vivo and not on its total dose. Therefore, itÕs not sufficient to control the total amount of delivered VEGF-expressing vector, but it is necessary to control the distribution of expression levels from each transduced cell in vivo
2) In addition, new induced vessels require at least 4 weeks of expression in order to become stable and independent from further VEGF signaling.
The use of short-term expressing gene therapy vectors, such as adenoviruses, is desirable both for the intrinsic safety of limited duration of expression and for the convenience of an off-the-shelf treatment. However, traditional short-term expressing vectors cannot fulfill the requirements of either dose control and or sufficient duration of expression necessary in order to induce safe and efficient angiogenesis
Myoblast-mediated gene transfer provides a highly controlled delivery system that allows the selection and the in vitro characterization of myoblasts expressing specific VEGF levels and ensures a sustained expression over time.
Using this tool, we previously found that co-expression of PDGF-BB, which controls pericyte recruitment and vascular maturation, ensures the induction of only normal and functional capillaries despite the expression of high and heterogeneous VEGF levels. Here, we aimed to establish whether PDGF-BB co-expression could also accelerate the stabilization of VEGF-induced vessels and, therefore, provide a safe and efficient strategy to achieve therapeutic angiogenesis within the short duration of expression afforded by adenoviruses.
We assesed the stabilization time-course of vessels induced by myoblasts expressing heterogeneous VEGF levels, in the absence or presence of PDGF-BB, after 2 or 3 weeks of expression. We found that 50% of capillaries induced by VEGF and PDGF-BB co-expression were VEGF-independent already after 2 weeks and 90% after 3 weeks, compared to none after 2 weeks and only 30% after 3 weeks with VEGF alone. These data suggest that VEGF and PDGF co-expression from a single bicistronic construct represents a convenient strategy to induce persistent vessels within a short time of expression, with clear implications for therapeutic applications.
Next we defined, for the first time, whether VEGF dose regulates the time-course of vascular stabilization and how PDGF-BB affects stabilization at different VEGF doses. Interestingly, we found that VEGF negatively regulates vascular stabilization in a dose-dependent fashion. In fact, normal capillaries induced by low VEGF levels stabilized faster than similarly normal vessels induced by higher doses at every time point considered. Aberrant vascular structures induced by very high VEGF levels never stabilized. Surprisingly, we found that PDGF-BB accelerated vascular stability only at medium and high VEGF levels, while it did not have significant effects on vessels induced by low VEGF doses.
Mechanistically, we found that vascular stabilization is not simply dependent on the presence of pericytes, because normal capillaries induced by different doses of VEGF alone or with PDGF-BB were equally pericyte-covered but stabilized with different time-courses. Therefore, we investigated the regulation of several factors involved in vascular maturation in the different conditions. Intriguingly, we observed a correlation between the trend of expression of TGF-?1 and Sema3A and the trend of vascular stabilization at different VEGF and PDGF-BB doses.
Sema3A has been recently described to promote vascular stabilization indirectly through the recruitment of a subset of neuropilin-1+ (NP1+)-CD11b+ bone marrow-derived cells. These cells are thought to exert their stabilizing effect by expressing pro-maturation factors, such as TGF-?1 and PDGF-BB.
On the basis of results obtained, we tested whether VEGF and PDGF-BB co-expression from an adenoviral vector might induce safe and persistent angiogenesis in the short frame of expression allowed by this vector. Indeed, we found that PDFG-BB co-expression prevented the formation of aberrant vessels induced by Ad-VEGF and yielded only normal capillaries, when the two factors were expressed from a single bicistronic vector. Furthermore, in an immunocompetent model, in which adenoviral vectors are cleared within 10-14 days, we found that a proportion of vessels induced by VEGF and PDGF-BB co-expression had successfully stabilized and persisted after the cessation of VEGF expression. Based on this evidence, we conclude that VEGF and PDGF-BB co-expression provides a clinically relevant gene therapy approach for therapeutic angiogenesis.