Over the past 2 decades, multiple epidemiologic studies demonstrated that individuals born with low birth weight (LBW) are predisposed to cardiovascular disease and hypertension,1,2 but the cellular and molecular mechanisms underlying this association remained poorly understood. In this issue of Blood, Ligi and colleagues demonstrate that endothelial progenitors from LBW infants exhibit striking reductions in their angiogenic properties, compared with those from infants born with normal weight, thus providing for the first time a potential mechanistic link between LBW and adult hypertension (see figure).3
The mechanisms that mediate the fetal programming of hypertension are likely numerous. Studies so far have demonstrated changes in the kidneys (reduction of nephron number), the neuroendocrine system (up-regulation of the hypothalamic-pituitary-adrenal axis), and the vascular tree (vascular dysfunction and reduced density of arterioles and capillaries).4 In their study, Ligi and collaborators focused on the origins of the vascular abnormalities, and evaluated the angiogenic properties of endothelial colony-forming cells (ECFCs) present in the cord blood from preterm LBW infants, compared with those of full-term infants with normal birth weight.3 ECFCs are a type of endothelial progenitor characterized by their ability to form endothelial cell colonies in vitro and to incorporate into new vessels in immunodeficient mice. When cultured, ECFC colonies derived from LBW infants appeared significantly later and were present in smaller numbers than those derived from full-term infants. In addition, Ligi et al applied a battery of in vitro assays to demonstrate that LBW ECFCs have a decreased ability to form tubular and capillary structures and to migrate. In vivo, unlike full-term controls, LBW-ECFCs were also unable to form robust capillary networks in Matrigel plugs injected in nu/nu mice, thus confirming their reduced angiogenic potential. These findings provide the first evidence of a relationship between birth weight and the angiogenic properties of ECFCs, and a potential mechanism for the microvascular rarefaction, arteriolar narrowing, and decreased branching previously described in animal models of intrauterine growth restriction5 and in former preterm infants with developmentally programmed hypertension.6,7
Using gene expression profiling, Ligi et al also found that, at the molecular level, the angiogenic defects of LBW-ECFCs were associated with increased expression levels of anti-angiogenic genes, particularly thrombospondin 1, compared with full-term controls. The elevated levels of thrombospondin 1 were confirmed at the protein level. Perhaps most interestingly, these investigators demonstrated that silencing of thrombospondin with siRNA partially restored the in vitro angiogenic properties of LBW-ECFCs.
LBW, defined as weighing < 2500 g at birth, can result from either intrauterine growth restriction in full-term neonates, or from premature birth. While both causes of LBW have been epidemiologically associated with adult hypertension, these are different problems from a pathophysiologic perspective: Intrauterine growth restriction reflects an adverse intrauterine environment with restricted nutrients and poor fetal oxygenation, which leads to inadequate fetal growth. Low birth weight because of preterm delivery results from ending the pregnancy before term (38 weeks), at a time when the fetus is still growing. The focus of the study by Ligi et al was on infants with LBW because of preterm delivery, although 5 of the 25 preterm LBW infants studied were also smaller than expected for their gestational age, reflecting intrauterine growth restriction. Interestingly, these investigators found a strong inverse correlation between birth weight and the ECFCs' angiogenic defects, with infants weighing < 1500 g being the most severely affected. This was consistent with previous epidemiologic studies showing a correlation between the risk of hypertension in young adults and the degree of immaturity at birth.2
Overall, the findings of Ligi et al significantly advance our understanding of developmentally programmed hypertension, but also generate several new questions and areas for future research. Specifically, it is unclear what causes the angiostatic profile of LBW ECFCs, and how long it persists. Furthermore, a causal association between the angiogenic properties of cord blood LBW-ECFCs and the abnormal microvascular pattern seen in adults with developmentally programmed hypertension remains to be demonstrated (see figure). When addressing these questions, it is important to keep in mind that—in LBW because of preterm delivery—the abnormal vascular development results from a premature nonphysiologic transition from the intrauterine to the extrauterine environment, which occurs before the fetus is fully developed. The significance of this environmental change as a key pathologic event is evidenced by the fact that, if the same organism was allowed to fully mature in utero, blood vessels would develop normally and there would be no increased risk for hypertension. Among many other differences, the normal intrauterine environment is markedly hypoxic compared with the extrauterine environment. Throughout gestation, the fetus progressively prepares for the transition to the relatively oxygen-rich extrauterine environment, as exemplified by the substantial increase in the concentration of antioxidant enzymes during the last weeks of gestation.8 If delivery occurs prematurely (particularly before 32 weeks), however, this preparation is not completed, leaving the fetus highly susceptible to environmental factors such as oxidative injury.9 How exactly this environmental change affects the angiogenic properties of preterm ECFCs and leads to abnormal vasculature remains to be determined. One possibility is that the angiostatic profile observed in LBW-ECFCs is the result of a premature exposure to the extrauterine milieu, for which these progenitors are not prepared. If so, we would expect LBW-ECFCs to exhibit normal angiogenic properties if examined in the hypoxic intrauterine environment. Alternatively, it is possible that the reduced angiogenic properties of LBW-ECFCs reflect a physiologic “transitional” status that, because of the premature delivery and/or interventions in the early postnatal period, persists for an abnormal period of time or over the life course of the individual. To address this possibility, it would be interesting to evaluate the angiogenic properties of ECFCs from former preterm LBW infants at different stages of life. Finally, from the clinical perspective, the finding that down-regulating thrombospondin significantly increased the angiogenic properties of LBW-ECFCs opens the door to potential novel therapeutic interventions aimed at preventing developmentally programmed hypertension, and perhaps other diseases of preterm infants characterized by abnormal angiogenesis.
Conflict-of-interest disclosure: The author declares no competing financial interests. ■