Janus phenomenon: the interrelated tradeoffs inherent in therapies designed to enhance collateral formation and those designed to inhibit atherogenesis.

Enormous advances have occurred in the understanding of the molecular and cellular mechanisms responsible for both collateral development (collaterogenesis*) and atherogenesis. This advancement has been accompanied by parallel advances in therapies designed to enhance collaterals and to inhibit the development and progression of atherosclerosis. Our laboratory has been interested in both areas, and during our work, we noticed a consistent tradeoff: Whatever intervention enhanced collaterals increased atherogenesis and visa versa. We refer to this tradeoff as the Janus phenomenon.† This observation, if correct, is of more than passing interest given the many antiatherosclerotic interventions used in patients and the proangiogenic interventions being tested that may soon be used therapeutically. Therefore, if the Janus phenomenon is real, its clinical implications would be of critical importance. Equally as important, the concept might provide mechanistic insights into both atherogenesis and collaterogenesis. Consequently, we began to track phenomena compatible with the Janus concept. Table 1 lists a few molecules for which reasonable evidence exists relating to both their atherosclerotic and collaterogenic effects and the actions of which are compatible with the Janus phenomenon.1–18 Complementary to the information contained in Table 1 are the opposite concordances displayed by different mouse strains regarding collaterogenic potential versus atherogenic susceptibility. C57BL/6 mice are susceptible to atherosclerosis and are exuberant collateral formers. Conversely, Balb/C mice are resistant to atherosclerosis and have an inferior collaterogenic capacity,19,20 suggesting that genetic determinants conveying susceptibility to atherogenesis will convey an enhanced capacity for collaterogenesis and visa versa. Therefore, the Janus phenomenon, expressing concordance between atherogenic and angiogenic proclivities, appears also to apply to genetic predispositions. (However, just the opposite propensities in the neointimal response to acute vascular injury have been observed between C57Bl/6 and BalbC mice— BalbC C57BL/6.21 It is therefore uncertain whether the Janus phenomenon holds for genetic propensities to develop restenosis.) Such concordances indicate that shared mechanisms contribute to collaterogenesis and atherogenesis, that stimulating these mechanisms will activate both processes, and conversely that inhibiting these mechanisms will inhibit both processes. The concept (the Figure) therefore lends itself to posing and testing new mechanistic hypotheses. Applying what we know about molecular mechanisms of atherogenesis to collaterogenesis might provide mechanistic insights about collaterogenesis that would otherwise go unsuspected and visa versa. Also, determining the molecular basis for the different capacities of C57BL/6 and Balb/C mice to develop collaterals will probably shed light on molecular mechanisms contributing to atherogenesis.