Multifunctional angiogenic factors: add GnRH to the list. Focus on "Gonadotropin-releasing hormone-regulated chemokine expression in human placentation".

THE HISTORICAL VIEW that angiogenesis is governed by the action of a few dominant trophic factors, such as VEGF, FGF, and transforming growth factor(TGF), has steadily eroded over the past two decades to give rise to the vision of a much more dynamic regulation in which angiogenesis is balanced among multiple angiogenic and angiostatic influences (5, 19). Equally important has been the identification of multiple varieties of molecules that influence this balance but are not primarily angiogenic factors themselves and have other critical regulatory functions. Examples of this latter class include integrins (21), nerve growth factor (17), brain-derived neurotrophic factor (9), hepatocyte growth factor (25), and many others. Not surprisingly, angiogenesis is also subject to modulation by interstitial products of inflammation, such as cytokines of both the interleukin (16) and CXC (22) families. Several types of lymphocytes also influence angiogenesis (3, 7, 22). In light of this complexity, it is important to ask how the multiple facets of angiogenesis are coordinated and what molecules are in charge? The recent study by Babwah and colleagues (1) addresses these questions, and for the angiogenic processes involved in uterine spiral artery remodeling during human placentation, the authors’ evidence reveals gonadotropin-releasing hormone (GnRH) to be a multifunctional coordinating player. Development of the human placenta begins near the end of the first trimester with the invasion of cytotrophoblasts into the uterine wall to anchor the embryonic chorion to the endometrium (12). Differentiation of the invading cytotrophoblasts yields extravillous trophoblasts (EVTs), which initiate remodeling of the uterine spiral arteries by promoting the programmed cell death of smooth muscle and endothelial cells (11). Whereas this basic sequence has been known for many years (14), its molecular regulation remains unclear. The involvement of GnRH in human placentation was recognized more than a decade ago, but its role was originally associated with regulation of matrix metalloproteinase activity (18). Similarly, involvement of lymphoid cells in placentation was proposed more than two decades ago (20), but the idea that a major subset of these cells, the uterine natural killer cells, could produce and release pro-angiogenic factors, is a much more recent idea (10) and one with uncertain physiological significance. A role for chemokines in placentation was also proposed several years ago (15), as was the idea that CXC chemokines might play a role in angiogenesis (13) but direct evidence of a role for CXC chemokines during placentation has been lacking. Reasoning that the main effects of GnRH on placentation could be detected at the mRNA level in ETVs, Babwah and colleagues treated an immortalized human placental trophoblast cell line (human HTR-8/SVneo trophoblasts) with buserelin (a synthetic GnRH-I analogue), GnRH-II, or antide (a GnRH-I competitive antagonist) for varying intervals at physiologically relevant concentrations, and then they screened for changes in gene expression using an Affymetrix microarray. This screen identified 37 genes of which the expression changed at least 1.5-fold in response to GnRH manipulation. Among these genes were the pro-angiogenic chemokines CXCL2, CXCL3, CXCL6, and CXCL8. To verify that these chemokine genes were responsive to GnRH, the HTR-8/SVneo cell line experiment was repeated with mRNA analysis performed via qPCR. In addition to verifying the microarray results, the qPCR measurements also established a timecourse for gene expression that was similar to what has been observed for GnRH-promoted matrix metalloproteinase responses (2). To be sure that these results were not a unique feature of the HTR-8/SVneo cell line, the authors used immunofluorescence to identify expression of CXCL8 within trophoblasts in sections of fixed human placenta. Other experiments revealed that GnRH treatment of cultured trophoblasts enhanced the release of CXCL8 into the culture media. Together the results clearly implicate GnRH as a regulator of angiogenic chemokines in placental trophoblasts. Given the connection between GnRH stimulation and CXC chemokine release from trophoblasts, Babwah and colleagues next explored the hypothesis that the chemokine release stimulated by GnRH helps to mediate lymphocyte recruitment during placentation. To test this possibility, the migration of fluorescently labeled Jurkat T cells, purified CD4 /CD8 T cells, and uterine natural killer cells was measured using transwell assays of chemotaxis. Conditioned media from the HTR-8/SVneo cell line and primary EVTs induced similar and potent migration responses in each of the lymphocyte populations tested. Most importantly, the GnRH receptor antagonist antide blocked the effect of GnRH on lymphocyte recruitment, thus confirming the critical role of this receptor as a mediator of GnRH actions in the placenta. In parallel, the effect of GnRH on lymphocyte recruitment was also blocked by repertaxin, an allosteric noncompetitive inhibitor of the CXCR1 and CXCR2 receptors that mediate the intracellular effects of CXCL2, CXCL3, CXCL6, and CXCL8. Interestingly, however, the ability of repertaxin to inhibit recruitment was more pronounced for uterine natural killer cells than for the Jurkat T cells, suggesting an important difference in how migration is regulated in the two cell types. Another important feature of these results was the continued slow recruitment of lymphocytes even in the presence of repertaxin. This residual recruitment suggests that the CXCR1 and CXCR2 receptors do not mediate all lymphocyte recruitment and perhaps that other chemokines (e.g., CXCL12 or CXCL16) and receptors (e.g., CXCR4 or CXCR6) are also involved (8, 24). Even so, the results from Babwah and colleagues provide convincing new Address for reprint requests and other correspondence: W. J. Pearce, Center for Perinatal Biology, Loma Linda Univ. School of Medicine, Loma Linda, CA 92354 ([email protected]). Am J Physiol Cell Physiol 297: C4–C5, 2009; doi:10.1152/ajpcell.00209.2009.