Editorial: Is Tissue Mass Regulated by Vascular Endothelial Cells? Prostate as the First Evidence

There is now considerable direct evidence that tumor growth is angiogenesis dependent, and that tumor mass is under the tight control of microvascular endothelium (1). New capillary vessels recruited by tumors carry nutrients and oxygen into a tumor and remove catabolites and carbon dioxide from it. This perfusion effect is however, reinforced by a potent paracrine effect. Microvascular endothelial cells release into their extracellular matrix and into the tumor itself a variety of growth and survival factors (2, 3). At least twenty such endothelial-derived paracrine factors are known (3), but those that have been studied in most detail include: basic fibroblast growth factor (bFGF), platelet-derived growth factor (PDGF), insulin-like growth factor-1 (IGF-1), heparin-binding epithelial growth factor and interleukin-6. Whether organ size or normal tissue mass is under the control of vascular endothelium has remained an open question. In this issue, Franck-Lissbrant et al. (4) show that growth of the rat prostate gland is regulated by vascular endothelial cells, which themselves are apparently responding to angiogenic activity elaborated by prostate epithelium under testosterone stimulation. The elegant experiments in this report demonstrate that testosterone stimulation in castrated rats causes the rapid onset of a wave of endothelial cell proliferation and vessel growth which is increased 3-fold in the first day and normalizes by 2 days. This burst of endothelial cell growth precedes by several days regrowth of glandular epithelium and subsequent enlargement of prostate mass. In unpublished data from the authors’ laboratory (Häggström et al.), castration decreased and testosterone increased vascular endothelial growth factor (VEGF) mRNA expression in the prostate. This is a seminal paper because it: 1) reports the first compelling evidence that normal tissue mass (including organ size) is tightly regulated by vascular endothelium; and 2) demonstrates a molecular pathway by which steroid hormones can mediate production of angiogenic peptides. This connection of male reproductive steroid hormones to angiogenic proteins may also hold for female reproductive steroids. For example, estradiol treatment of human endometrial cells increased expression of VEGF mRNA by 3-fold (5). Franck-Lissbrant et al. (4) also found that macrophages increased in the prostate after castration and returned to intact values after testosterone treatment. Macrophages could be a source of an angiogenesis inhibitor, angiostatin (6), which can be liberated from plasminogen by macrophage-derived metalloelastase (7). Therefore, as the authors suggest, the growth or regression of microvasculature in the prostate may be the result of a net balance of positive and negative regulators of angiogenesis, as has been demonstrated in tumors (8). Macrophages are also a source of an angiogenic stimulator, VEGF. In fact, VEGF secretion by macrophages in human endometriosis is increased in response to ovarian steroids (9). This steroid regulation of angiogenic proteins represents another similarity between the male and female reproductive systems. A study in our own laboratory (10) of hormone-responsive human prostate carcinoma confirms the results of Franck-Lissbrant et al. (4) in normal prostate. Androgen deprivation of human prostate carcinoma cells (LnCap) in vitro led to a marked decrease in VEGF mRNA expression. In addition, androgen withdrawal inhibited the hypoxic induction of VEGF. In mice bearing human prostate carcinomas, castration resulted in a rapid decrease in VEGF expression within 24 h and a marked reduction in neovascularization by 3 days. Tumor regression did not begin until 8 days, demonstrating that inhibition of vascular endothelial growth precedes reduction in tumor mass. It remains to be determined whether escape of prostate carcinoma to androgen-independence is associated with overexpression of VEGF by the tumor cells. When angiogenesis is blocked in tumors by specific inhibitors of endothelial proliferation such as angiostatin (11) or endostatin (12), there is an increase in tumor cell apoptosis, with no change in the elevated rate of tumor cell proliferation, suggesting that tumor cells may become apoptotic following withdrawal of endothelial-derived growth and survival factors. It has been shown in certain tumors that one endothelial cell supports 5 to 50 tumors cells (13). If normal cells are similarly dependent upon endothelial-derived paracrine factors, the ratio of endothelial cells to normal parenchymal cells is likely to be lower than for tumor cells. Nevertheless, the regulation of tissue mass or organ size by vascular endothelial cells may be based upon mechanisms which also operate in tumors. The report by Franck-Lissbrant et al. (4) has broad implications for our understanding of the role of vascular endothelium in embryonic development and wound repair as well as in physiological involution or hypertrophy of normal organs. If the mechanism of endothelial control of prostate size as reported here can be generalized to other tissues, it may lead to the discovery of endothelial-derived growth or survival factors that are critical for different organs, for example, liver and kidney (Fig. 1). Diagram to illustrate a proposed general mechanism by which normal tissue mass may be regulated by the paracrine effect of microvascular endothelial cells (the perfusion effect of neovascularization is not shown). A similar principle could apply to ovary, liver, or other organs. Diagram to illustrate a proposed general mechanism by which normal tissue mass may be regulated by the paracrine effect of microvascular endothelial cells (the perfusion effect of neovascularization is not shown). A similar principle could apply to ovary, liver, or other organs.