An article in this issue by Fujita et al1 describes markedly increased concentrations of bFGF in the pericardial fluid of human patients who were operated on because of unstable angina. The authors should be congratulated for this highly original approach, which is bound to shed light on the molecular mechanisms of collateral growth in humans. That growth factors must play a role in collateral vessel development has been evident for many years. However, it is still unclear which of the numerous factors is the decisive one. Furthermore, it is unclear whether this role is defined by a single factor or by a regulated sequence of a cascade of growth factors and which cell type is responsible for growth factor production, not to mention that a multitude of growth factor receptors and binding proteins complicate the analysis. Collateral vessel growth may be an equation with a two-digit number of variables (FGF-1, FGF-2, VEGF, IGF-1 plus six IGF binding proteins, and four FGF and two VEGF receptors=16 variables) that may not all be relevant at the same time, but many do play a role, as we now know.
Moreover, we have to realize that collateral vessels are a heterogeneous group that consists mainly of two classes: those of capillary size or slightly larger and that are devoid of vascular smooth muscle and those that are much larger and have a muscular coat.2 3 The latter usually are located epicardially, whereas the capillary collaterals have a predilection for the subendocardium (where they form a plexus), but they also are observed in all layers of the ventricular muscle. Capillary collaterals develop by sprouting (angiogenesis). Muscular collateral arteries develop in situ from preexisting arterioles, and they recapitulate the embryonic program of vasculogenesis.3 Angiogenesis and vasculogenesis differ in their molecular mechanisms (which we understand only in part), and accordingly their requirement of growth factors is different. Experiments in the porcine heart have shown that the response to regional ischemia is generally capillary sprouting, which becomes so intense that the minimal (anatomic) resistance of the receiving vascular bed is halved. VEGF certainly plays a role in this type of angiogenesis, although it is notoriously difficult to demonstrate the peptide and its mRNA in convincing amounts and in plausible spatial relations to the sites of vascular growth.
The results of Fujita et al1 are a case in point: although they used an accepted and sensitive method, they were unable to detect a difference in the pericardial VEGF concentration between the two groups studied.
Recapitulated vasculogenesis, the process of muscular artery growth, is an even more complicated process because the preexisting artery must be completely remodeled, its elastic skeleton and the old smooth muscle must be completely replaced, and the mitotic growth of endothelium and smooth muscle has to occur. It is here that FGF may come into play, because it is a general mitogen, in contrast to VEGF, which is specific to endothelial cells.
Fujita et al discuss the idea that the ischemic cardiac myocyte is the main producer of angiogenic factors, particularly FGF-2. Indeed, the conventional view is still inspired by the tight association between ischemia and vascular growth. However, this correlation may be only spurious and not causal.4 A closer look at the localization of muscular collaterals on one hand and the occurrence of ischemia on the other reveals that they do not share a common space: muscular collaterals develop epicardially, whereas ischemia exists, if only transiently, subendocardially. Collateral vessels and ischemia do not even share a common time constraint; they continue to grow in size and wall thickness long after ischemia has vanished. The most glaring dissociation between collateral growth and ischemia occurs in the peripheral circulation: in femoral artery occlusion, the lower leg becomes ischemic but collaterals develop in the upper leg. It is imaginable that the drastically increased blood flow velocity produced by stenosis and occlusion initiates the remodeling and arterial growth in preformed collaterals by means of invading monocytes, the angiogenic cell type par excellence.5
A better and probably causal relationship between ischemia and vascular growth exists for capillary sprouting. Here, the link between ischemia and vascular growth is established by the expression of VEGF: ischemia leads to ATP catabolism and adenosine formation, which in turn increases the stability of the VEGF mRNA so that new peptide can be synthesized without new transcription.6 Our group at Max-Planck-Institute3 and others7 had shown that VEGF mRNA is induced by acute myocardial ischemia. The increase is substantial and demonstrable by conventional Northern blot hybridization. We were surprised not only by the rapidity (≈20 minutes) but also by the transient nature of the response. No such convincing association exists for the FGF family between ischemia and growth factor expression. Acute ischemia does not lead to rapid increases in FGF expression, but chronic ischemia does through the activation of macrophages.
We have found it somewhat difficult to understand the presence of a potent growth factor in an essentially nonproliferative situation (FGF-1 is constitutively expressed in the normal heart), and we have come to the conclusion that the FGFs may have functions other than or in addition to their functions as mitogens: FGFs probably are trophic factors with a survival value for terminally differentiated cells such as neurons8 and cardiac myocytes. FGFs are potent neurotrophic factors, and their main functions in the heart may be very similar. The highly interesting findings by Fujita et al1 therefore may allow an additional or alternative interpretation: chronically ischemic myocardium triggers the expression of trophic factors with a survival value.
A question that Fujita et al appropriately address is how a peptide without a secretory sequence can leave an undamaged cell and finally arrive in the pericardial fluid. Maciag and colleagues9 recently showed that heat-shock proteins also act as chaperones for FGF, facilitating its secretion. It is of note that ischemia leads to the expression of heat-shock proteins.10 Fujita et al1 show that the self-defense mechanisms of the heart are not perfectly developed: the patients studied each needed a bypass operation. This defective adaptation should lead the way to the development of therapeutic strategies that use pathways that are already outlined by nature.
Selected Abbreviations and Acronyms
|bFGF||=||basic fibroblast growth factor|
|FGF||=||fibroblast growth factor|
|IGF||=||insulin-like growth factor|
|VEGF||=||vascular endothelial growth factor|
The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association.
- Copyright © 1996 by American Heart Association
Fujita M, Ikemoto M, Kishishita M, Otani H, Nohara R, Tanaka T, Tamaki S, Yamazato A, Sasayama S. Elevated basic fibroblast growth factor in pericardial fluid of patients with unstable angina. Circulation.. 1996;94:610-613.
Schaper W, Schaper J. In: Collateral Circulation: Heart, Brain, Kidney, Limbs. Dordrecht, Netherlands: Kluwer Academic Publishers; 1993.
Schaper J, Koenig R, Franz D, Schaper W. The endothelial surface of growing coronary collateral arteries: intimal margination and diapedesis of monocytes: a combined SEM and TEM study. Virchows Arch A (Pathol Anat).. 1976;370:193-205.
Shweiki D, Itin A, Neufeld G, Gitay-Goren H, Ke`shet E. Patterns of expression of vascular endothelial growth factor (VEGF) and VEGF receptors in mice suggest a role in hormonally regulated angiogenesis. J Clin Invest.. 1993;91:2235-2243.
Jackson A, Tarantini F, Gamble S, Friedman S, Maciag T. The release of fibroblast growth factor-1 from NIH 3T3 cells in response to temperature involves the function of cysteine residues. J Biol Chem.. 1995;270:33-36.
Andres J, Sharma HS, Kno¨ll R, Stahl J, Sassen LMA, Verdouw PD, Schaper W. Expression of heat shock proteins in the normal and stunned porcine myocardium. Cardiovasc Res.. 1993;27:1421-1429.