Main

Angiogenin, a polypeptide with a molecular size of 14 kD, is a potent inducer of vascular growth(1). It was initially isolated as a tumor cell-secreted product but subsequently found to be a normal constituent of human plasma(2). Expression of angiogenin mRNA is detected in a wide variety of normal cells as epithelial cells, fibroblasts, and peripheral blood cells(3). Angiogenin belongs to the pancreatic RNase superfamily; however, it has only a 35% amino acid sequence identity(4) and an overall 68% homology with pancreatic RNase(2). Indeed, angiogenin, although reported an efficient tRNase in cells(5), is generally characterized as a poor RNase(5) being inactive toward the more conventional substrates of RNase(6–9). Shapiro and Vallee(10) and Shapiro et al.(11) have found that angiogenin's ribonucleolytic activity is a requisite for its angiogenic activity. Certainly, angiogenin is one of the most potent inducers of neovascularization(1) when compared with other angiogenic polypeptides, such as acidic and basic FGF, TGF-α and -β(12), and tumor necrosis factor-α(13, 14). Nevertheless, angiogenin alone has no known effect on endothelial cell proliferation(15), although it possibly interacts with cell-surface receptors and the extracellular matrix of endothelial cells(16). Although the detailed molecular mechanism by which angiogenin elicits new blood vessels has not been elucidated, results from various studies indicate that this process requires minimally the ribonucleolytic action of angiogenin, its binding to cellular receptors, translocation of the protein to the nucleolus of endothelial cells, activation of endothelial cell-associated proteases, and consequent promotion of cell invasiveness by angiogenin(17). Last, it has been reported that angiogenin stimulates umbilical vein endothelial cells to secrete prostacyclin and to increase intracellular levels of 1,2-diacylglycerol and inositol triphosphate by activation of phospholipase A2(18) and C(19), respectively.

Shapiro and Vallee(20) studied a human PRI that binds extremely tightly to angiogenin, abolishes all its properties(ribonucleolytic, angiogenic, activating endothelial cell phospholipase), regulates it in vivo, and may have important mechanistic, physiologic, and pharmacologic implications.

Considering its properties it has been suggested(21, 22) that angiogenin may play a role in the vascular development of the fetus and in the neovascularization that accompanies diseases of the neonatal period (retinopathy of prematurity and hemangiomas)(12) and wound healing.

Up to now all existing studies on angiogenin developmental pattern apply only to animals(22), and angiogenin levels in the perinatal period of humans have not been explored. Nevertheless, this period is a very crucial one, as the transition from intra- to extrauterine life, implying the elimination of the placenta and consequently the decrease of PRI activity could influence angiogenin serum concentrations(20, 22). On the other hand, eventual developmental errors in the placenta-leading to spurious angiogenesis(23)-or other diseases or abnormalities in the newborn could be reflected in abnormal circulating angiogenin levels. Therefore, we undertook this study aiming at determining a baseline for angiogenin serum concentrations in the healthy mother, fetus, and infant 1 and 4 d postpartum. We assumed that this protein with its angiogenic properties on the one hand and its inhibition by PRI on the other may in the future provide a diagnostic index(24) analogous, e.g. to the use ofα-fetoprotein. We further investigated whether birth weight and sex could affect angiogenin levels in the early neonatal life of healthy full-term infants, as levels of other angiogenic factors are correlated with fetal size(FGF)(25) and interfere with sex (FGF, TGF)(26, 27). Also, we studied whether perinatal stress, reflected in the mode of delivery, could play a role in angiogenin serum concentrations.

The study was approved by the Ethics Committee of our Teaching Hospital.

METHODS

Subjects. Thirty healthy appropriate for gestational age full-term infants born from 30 healthy nonsmoking mothers with uncomplicated single pregnancies were included in this study after informed consent of their parents. Thirteen were boys and 17 were girls. Mean (±SD) birth weight was 3376 ± 334 g (range 2800-4000 g), mean gestational age was 39.5± 1.5 wk (range 37.5-41.5 wk). Seventeen infants were born by vaginal delivery and 13 by cesarean section because of a previous one. Apgar scores ranged from 8 to 10 at birth and 9 to 10 at 5 min. Placentas were in all cases normal in appearance and weight.

Procedure. One milliliter of blood was drawn from a peripheral vein of all 30 infants on d 1 and 4 of life. In 10 cases blood was also drawn from the mother before delivery as well as from the umbilical vein at delivery. Blood was collected in pyrogen-free tubes, and serum was immediately separated by centrifugation after clotting and was kept frozen at -20 °C until assay. The analysis was performed in prediluted (1:200) serum samples by an enzyme immunoassay, using a commercially available kit (angiogenin human ELISA system RPN 2161, Amersham International plc). The sensitivity, intra- and interassay coefficients of variation of the assay were, respectively, 0.6μg/L, 2.8 and 9.0%.

Data analysis. Data on angiogenin serum concentrations showed a normal distribution (Kolmogorov-Smirnov-test). Therefore, parametric methods,t test, and paired t test were used in the statistical analysis as appropriate. Pearson's correlation coefficients were also calculated.

RESULTS

Angiogenin concentrations (μg/L) in maternal serum, umbilical cord serum, neonatal d 1 serum, and neonatal d 4 serum were, respectively (mean± SD; range): 225.7 ± 49.6; (145-314), 119.0 ± 34.2;(70-180), 166.4 ± 44.9; (104-310), 240.8 ± 52.6; (156-400). The paired t test for the 10 maternal, umbilical cord, and neonatal d 1 and 4 serum samples showed that angiogenin serum concentrations were significantly higher in maternal serum compared with those in umbilical cord serum (n = 10, p < 0.0002) as well as in neonatal d 1(n = 10, p < 0.01) but not in neonatal d 4 serum(n = 10, NS). Also, they were significantly lower in umbilical cord serum compared with those in neonatal d 1 and 4 serum (n = 10,p < 0.0002 and p < 0.0003, respectively). The paired t test for the 30 neonatal d 1 and 4 serum samples showed that there was also a statistically significant increase of angiogenin serum concentrations from d 1 to 4 (n = 30, p < 10-7). The application of the t test showed that no statistically significant differences were found in the respective umbilical cord and neonatal d 1 and 4 angiogenin serum concentrations between boys and girls, as well as between infants born by vaginal delivery and those born by cesarean section. A statistically significant correlation was found between angiogenin serum concentrations in umbilical cord and neonatal d 1 samples (r = 0.84, n = 10, p < 0.002) as well as between those in d 1 and 4 (r = 0.37, n = 30, p < 0.04), but not between those in umbilical cord and neonatal d 4 samples at the conventional level (r = 0.60, n = 10, p < 0.08) or between those in maternal serum and umbilical cord serum. Last, umbilical cord and d 1 and 4 angiogenin serum concentrations were not correlated with birth weight.

DISCUSSION

Plasma or serum angiogenin levels have been measured in normal volunteers and were found to range between 60 and 150 μg/L(24). A later study from the same group reports angiogenin to be present in human plasma at a concentration of 400 μg/L(2). Furthermore, the accompanying leaflet of the kit applied in this study gives plasma and serum angiogenin concentrations as ranging between 156 and 476μg/L. Our results for angiogenin concentrations in healthy mothers are within these limits.

The finding of significantly lower angiogenin serum concentrations in the umbilical cord and d 1 neonates than in maternal serum could seem strange, as in the developing fetus vascular growth is rapid and factors promoting angiogenesis are expected to be found in abundant quantity(22). Nevertheless, other studies have shown that the developmental pattern of angiogenin gene expression in rat liver was also unanticipated(22). Thus, angiogenin mRNA levels in rat liver were found low or undetectable in the developing fetus, increased in the neonate, and maximal in the adult(22), although in the latter new vascular growth is slow, as shown by endothelial cell turnover times of approximately 100-1000 d in adult mouse organs(28). Moreover, densitometric analysis revealed that angiogenin mRNA content in the rat liver increases by a factor of 10 between d 19 of gestation and postnatal d 2 and further during growth to adulthood to twice that found in the neonate(22). These data lead to the assumption that angiogenin is not directly responsible for fetal vascular growth(5, 22), possibly leaving this role to other factors, such as acidic and basic FGF, TGF-α, tumor necrosis factor-α, and prostaglandins E1 and E2(12, 29). The presence of PRI, which binds to angiogenin and abolishes its activity, offers hereto a plausible explanation. On the other hand, elimination of the placenta and consequent decrease of PRI activity after birth could justify the rapid increase of angiogenin serum concentrations, which reached in our study very soon to maternal levels. Thus, angiogenin serum concentrations were significantly higher in neonatal d 1 samples compared with umbilical cord samples (p < 0.0002) and neonatal d 4 samples, which did not differ from maternal samples, compared with neonatal d 1 samples (p < 10-7). In this respect, it could be hypothesized that errors in placental development or placental dysfunction, implying abnormal PRI secretion, might lead to different results, and thus, measurement of this protein could be of diagnostic value(24).

Perinatal stress, reflected by the mode of delivery, does not seem to affect angiogenin levels, presumably due to inhibition of its activity through existing PRI. Therefore, the documented lack of differences in respective umbilical cord and neonatal d 1 and 4 angiogenin serum concentrations of infants born by vaginal delivery and those born by cesarean section is evident. Consequently, angiogenin being inhibited during labor cannot be implicated in its process. In addition, for the same reason, angiogenin serum concentrations seem not to be dependent on sex or fetal size, as has been reported for other angiogenic factors(25–27). The lack of significant correlation between birth weight and angiogenin values in umbilical cord and neonatal d 1 and 4 samples may rely on the fact that the birth weights in this study reflected normal placental function and were representative of those of healthy full-term neonates. Last, it appears that, because maternal levels were nearly twice as high as fetal and no correlation existed between angiogenin values in maternal serum and umbilical cord serum, levels measured in umbilical cord serum reflect fetal synthesis and not maternal transfer.

In the meanwhile nothing is known about the regulation of angiogenin synthesis at the tissue or cell level(30). Nevertheless, its presence in normal blood suggests that it might be involved in endothelium homeostasis(16), and its widespread expression in different human cells implies a biologic function not only related to angiogenesis(30). Possibly, in the future, angiogenin could serve as a diagnostic index, depending on varying plasma or serum levels in normal and pathologic states, particularly those involving blood vessel proliferation(30). The latter could also apply to the neonate, especially to the preterm infant.

In conclusion, a rapid increase of angiogenin serum concentrations to maternal levels takes place in the first four postnatal days in healthy full-term neonates, possibly due to the elimination of the placenta and the expected decrease of PRI activity.