Main

Abnormal fetal and placental growth has been associated with the subsequent development of elevated blood pressure in a number of rat models(1–5). In humans recent epidemilogic studies indicate that low birth weight and placentomegaly are associated with a significantly increased risk of hypertension and related diseases in later life(6–8). It has been suggested that a poor fetal environment may cause physiologic, metabolic, endocrine, or structural changes during critical stages of development, which predispose the individual to the subsequent development of hypertension. It has been postulated that this predisposition may be initiated in fetal life, may persist during growth and differentiation, and continue into adult life(9, 10).

The SHR is an extensively studied paradigm developed by selectively breeding the Wistar strain of rats for high blood pressure(11). Blood pressure becomes elevated during the postnatal period in the SHR(12). One report provides evidence for raised umbilical blood pressure even before birth(13). Cardiac hypertrophy has been reported in the 1st wk after birth, and there is evidence of increased heart weight to body weight ratios as early as d 19 of gestation(12–14). Blood pressure and cardiomegaly continue to increase as the animal matures, and by 10 mo of age there is a 68% increase in left ventricular weight with a greater than 30% higher blood pressure in SHR compared with the WKY(14). The SHR is also insulin-resistant(15, 16), a state often associated with essential hypertension(17).

There is evidence that hypertension in the SHR may be initiated in the perinatal period. SHR neonates fostered to WKY dams have lower blood pressures in later life(18), suggesting that hypertension could be programmed in early postnatal life. Because WKY neonates fostered to SHR dams do not develop hypertension, it appears that the SHR neonate is susceptible to the maternal environment(18). Perinatal angiotensin-converting enzyme inhibitor treatment can completely prevent the development of hypertension in the SHR(19), whereas postnatal angiotensin-converting enzyme inhibitor treatment can reduce but not prevent the development of hypertension in the SHR(20). This suggests that fetal as well as postnatal influences affect the development of hypertension in the SHR.

Our current studies are designed to characterize perinatal growth in SHR. Our hypothesis is that, if metabolic and endocrine changes occur in the perinatal period, these may contribute to the later onset of hypertension and related disease in this strain. We have therefore investigated growth and selected aspects of endocrinology and metabolism in the perinatal SHR.

METHODS

Animals. These studies were carried out using a protocol approved by the University of Auckland Animal Ethical Committee. SHR and WKY rats were bred from colonies maintained by the University of Auckland Animal Resources Unit. These colonies were obtained from Flinders University, Adelaide, Australia, which received F40 SHR and F14 WKY from the National Institutes of Health, Washington, DC, in 1976. Both strains have since been maintained as inbred colonies.

Blood pressures were measured in 10 SHR and 10 WKY virgin females aged between 80 and 90 d. Blood pressure was recorded by tail cuff according to the manufacturers' instructions (blood pressure analyzer IITC, Life Science, Woodland Hills, CA). Rats were restrained in a clear plastic tube in a warm room (25-28°C). After acclimatizing for 10-15 min the cuff was placed on the tail and inflated to 240 mm Hg. Pulses were recorded during deflation at a rate of 3 mm Hg/s, and reappearance of a pulse was used to determine systolic blood pressure. The systolic blood pressure recorded for each animal was the average of three to six readings.

Experimental protocols. Date-mated female rats, age 80-90 d, were fed ad libitum. The presence of sperm in vaginal lavage on the morning after mating indicated d 1 of pregnancy. Pregnant SHR and WKY dams were killed on d 16, 18, 20, and 22 of gestation (n = 5 litters, SHR also on d 23 of gestation) by decapitation after halothane anesthesia. In both strains at each gestational age, fetuses and placentas were dissected, and the weights were recorded. Fetal livers were frozen in liquid nitrogen and stored at -80°C until RNA extraction. In addition, offspring were killed on d 1, 5, and 15 postnatally (n = 20-25, from five separate litters). The morning after birth was defined as postnatal d 1 in both strains.

To determine whether there were any differences in fetal, placental, or neonatal water content, a second study was carried out on fetal d 20, 22, and 23 and on postnatal d 1 and 5. Fetal crown to rump and postnatal nose to anus measurements were made, and the tissues were collected into preweighed containers, weighed, and frozen. For dry weights, tissues were freezedried to a constant weight over 3 d. Trunk blood was collected into heparinized tubes and pooled from four to seven fetuses or neonates per litter. Blood glucose and lactate were measured using a glucose-lactate analyzer (model 2300, Yellow Springs Instrument Co., Yellow Springs, OH). In a separate group of animals, fetal and maternal hematocrit was determined using the standard microcapillary centrifugation technique.

IGF-I and IGF-II RIA. Blood plasma was acidified to disassociate IGF-I and IGF-II from their binding proteins, separated by HPLC, and assayed by RIA as described previously(21).

IGFBPs. IGFBPs in fetal and neonatal serum were estimated by ligand blot in two separate analyses(22). Briefly, dilutions of serum equivalent to 1 μL were separated by polyacrylamide gel(12%) electrophoresis under nonreducing conditions and blotted onto nitrocellulose membranes (Schleicher & Schuell, Keene, NH). The IGFBPs on the blot were renatured with 3% Tween 20, blocked with BSA (1%), and incubated with 125I-IGF-II at 4°C for 16-24 h. Unbound 125I-IGF-II was removed by washing with buffer. The membrane was placed on x-ray film (Kodak XAR) at -80°C for 24-72 h, and the film developed. The intensity of the autoradiographic signal in each of the approximately 30-kD bands (identified as IGFBP-2 by immunoblotting with polyclonal rabbit anti-bovine IGFBP-2 antibody, Upstate Biotechnology Inc., Lake Placid, NY) was determined using a densitometer (Molecular Dynamics, Sunnyvale, CA).

RNA extraction and Northern analysis. The rat IGFBP-2 cDNA used in these studies was as described by Brown et al.(23). RNA was extracted from pooled frozen fetal tissues using Trizol reagent (Life Technologies, Inc., Gaithersburg, MD) according to the manufacturer's instructions. Aliquots of total RNA (10 μg) were size-separated by electrophoresis on 1% agarose gels containing 0.6 M formaldehyde, blotted onto nylon membrane (Hybond N+. Amersham International, Bucks, UK) by capillary action, and baked at 80°C for 2 h to fix RNA to the membrane. Hybridization conditions were as described previously(24). Membranes were washed at medium stringency (30 min in 2 × SSC at room temperature and 30 min in 0.5× SSC, 0.1% SDS at 60°C) and exposed to x-ray film at -80°C. Northern blots were performed twice.

Statistics. Where multiple members of the same litter were included together in an analysis, overall differences between the two strains were analyzed using ANOVA with litter nested within strain. Statistical differences at individual time points were analyzed using the unpairedt test on litter means, with n = number of litters. All data are expressed as mean ± SD. Statistical significance was assumed at a value of p < 0.05.

RESULTS

SHR and WKY rat colony characteristics. The strain differences in blood pressure between SHR and WKY females was confirmed in our colony before breeding. Maternal systolic blood pressure was significantly higher in SHR females (165 ± 4 mm Hg, n = 10) than the mean pressure in WKY females (116 ± 2 mm Hg, n = 10, p < 0.001).

SHR dams had a longer gestation time than did WKY dams. SHR dams delivered on the morning of the 23rd d after mating, whereas WKY dams delivered 20 h earlier on the afternoon of d 22. There were significantly fewer fetuses in SHR litters (9.0 ± 0.6, n = 25) compared with WKY (11.1± 0.5, n = 24, p < 0.007). However, this difference was not significant at any of the individual time points.

Fetal and neonatal growth characteristics. There was an overall difference in fetal and neonatal body weight between SHR and WKY during the study period (d 16 of gestation to d 15 after birth, p < 0.001.Fig. 1). There was also an overall difference between SHR and WKY placental weight during late gestation (d 16 to birth. p< 0.003, Fig. 2). However, the differences were not seen at all of the time points. During late gestation SHR fetuses were significantly lighter than WKY fetuses at d 16, 18, and 20 of gestation(p < 0.01). By fetal d 22 there was no significant difference in fetal weight between SHR and WKY. On fetal d 23 the SHR were significantly heavier than either strain on fetal d 22 or on neonatal d 1 (p < 0.05). On neonatal d 1 there was no significant difference in weight between the two strains, but by d 5 and 15 the SHR again had lower weights(p < 0.05) in comparison to WKY.

Figure 1
figure 1

(a) Fetal weight from d 16 until the day of birth. Data are mean ± SD for five litters. (b) Body weight on postnatal d 1, 5, and 15. Data are mean ± SD for 20-25 neonates from five litters. ANOVA indicated a significant strain difference between SHR and WKY. At each time point, data were analyzed using unpaired t tests(*p < 0.05 and **p < 0.01). Error bars are plotted but are not visible on many of the columns.

Figure 2
figure 2

Placental weight in the SHR and WKY from fetal d 16 until the morning of birth. Data are mean ± SD for five litters. ANOVA indicated a significant strain difference (p < 0.005). At each time point data were analyzed using unpaired t test(*p < 0.01).

Body length was significantly different between the strains at fetal d 20 and postnatal d 1 (Table 1). At fetal d 20 the SHR were shorter and, unlike the WKY, there was no significant correlation between fetal weight and length (r = 0.30, p < 0.15). The d 1 newborn SHR were longer than WKY (11%, p < 0.02) and again, there was no correlation between weight and length in d 1 newborn SHR. At all other time points in SHR, weight and length were positively correlated. At all time points in WKY, there were significant correlations between fetal weight and length. On postnatal d 1 the ponderal index was significantly lower in the SHR than in WKY (Table 1). There were no significant differences in the ponderal index at other ages.

Table 1 Body length and ponderal index in SHR and WKY

Between fetal d 20 and postnatal d 5, wet and dry tissue weights were assessed. Fetal and neonatal wet and dry weights remained in the same proportion throughout this period except in the d 1 newborn SHR which had a lower water content (p < 0.02, data not shown).

Over the time period investigated, organ weight as a proportion of body weight was significantly different overall in the SHR heart, lung, liver, kidney, and brain (Table 2). Heart and kidney weights were proportionately heavier on d 20, and by d 22 heart, kidney, and lung were proportionately heavier. On the day of birth in the SHR (d 23) and WKY (d 22) there were no significant differences in organ weights, as a percentage of body weight, except in the SHR brain, which was significantly lighter(p < 0.001). Postnatally there was considerable variation in organ weight although heart and kidney were, for the most part, proportionately heavier (Table 2).

Table 2 Organ weights in the SHR and WKY

Placental growth characteristics. On fetal d 16 and 18, SHR placental wet weight was not significantly different from WKY. On fetal d 20, 22, and 23 the wet placental weights in SHR were heavier than those of WKY on d 20 and 22 by 16-30% (Fig. 2, p < 0.005). Placental dry weights were recorded on d 20, 22, and 23 and were all in similar proportion to the wet weights and significantly heavier at d 20, 22, and 23 in SHR (p < 0.01. data not shown).

Metabolic indices. Blood glucose levels in the SHR on fetal d 20 and 22 were lower than in WKY fetuses (Table 3,p < 0.05). On fetal d 23, SHR glucose levels had risen significantly compared with those in SHR at fetal d 22 (p < 0.01) and were not different from blood glucose in the WKY at fetal d 22. After birth glucose levels were not different between the strains. Fetal blood lactate levels were increased in the SHR compared with the WKY on d 20 and 22(p < 0.05). In postnatal animals (d 1 and 5) blood lactate fell to approximately a fifth of the prenatal levels (Table 3).

Table 3 Fetal and neonatal blood glucose and lactate levels

Fetal blood hematocrit was decreased in SHR compared with WKY at d 20 of gestation ((Fig. 3, p < 0.001). Maternal hematocrit was significantly elevated in the SHR dams (Fig. 3, p < 0.05).

Figure 3
figure 3

Fetal and maternal hematocrit in the SHR and WKY. The data were analyzed by unpaired t test (*p < 0.05 and **p < 0.001; n = 7 in SHR, n = 10 in WKY).

IGF-I, IGF-II, and IGFBPs in blood and liver. There was no difference in IGF-I concentrations in blood between the strains or on different days of gestation (Table 4). Fetal plasma IGF-II levels were significantly lower in the SHR compared with the WKY at fetal d 22(p < 0.01). However, on comparing the last day of gestation in each strain, the IGF-II in fetal plasma had risen to similar levels.

Table 4 Fetal plasma IGF-1 and IGF-2 levels

After separation of blood plasma by gel electrophoresis, the major band with IGFBP activity detected and estimated by ligand binding of labeled IGF-II was of molecular mass 28-32 kD and was identified as IGFBP-2 by Western immunoblot. IGFBP-2 was decreased by approximately 60% in the SHR compared with the WKY on fetal d 20 and 22 (p < 0.01,Fig. 4). On d 1 after birth there was no difference in IGFBP-2 between SHR and WKY. However, on d 5 after birth IGFBP-2 was decreased by 20% in the SHR (p < 0.05).

Figure 4
figure 4

(a) IGFBP-2 activity in fetal d 20 and 22 and postnatal d 1 and 5 plasma in the SHR and WKY. All WKY data are standardized to 1. The data were analyzed by unpaired t test (*p< 0.05 and **p < 0.01, n ≥ 4 in each group).(b) Ligand blots showing IGFBP-2 activity in SHR and WKY plasma. The band at approximately 30 kD was confirmed to be IGFBP-2 by immunoblotting with a specific antibody.

On prolonged exposure of the ligand blots to x-ray film, bands in the expected positions for IGFBP-3 and IGFBP-4 were visible in fetal and neonatal plasma. These were not quantifiable but did not appear to be different between the strains. The expression of IGFBP-2 mRNA in the fetal liver was not different at fetal d 16 and 18, but on d 20 and 22 of gestation it was decreased in SHR compared with WKY (p < 0.01.Fig. 5).

Figure 5
figure 5

Relative levels of IGFBP-2 mRNA in fetal livers of SHR and WKY. Values are adjusted for loading using methylene blue staining of 28 S ribosomal RNA. Each age group was analyzed separately, and all WKY data are standardized to 1. The data were analyzed by an unpaired t test(*p < 0.01. n = 5 in all groups).

DISCUSSION

The SHR has been used as a model for human essential hypertension for many years(11). In this report we show that fetal and placental growth, metabolism, and IGF systems are disturbed in the SHR in late gestation and the early perinatal period. This may be a unique model as growth disturbance arises without nutritional, surgical, or endocrine intervention. The growth retardation observed in SHR, maximal at d 18 of gestation (19%), is not as severe as in many other models, where corticosteriod treatment(25), chronic undernutrition(26, 27), temporary starvation(28), and hypoxia(29) may induce IUGR of up to 30% before term. Several of these IUGR animal models develop hypertension in later life(1–5, 30).

Gestation in the SHR was approximately 20 h longer than in the WKY. A 23-d gestation in the SHR has been reported previously by Kerr et al.(31). In a previous report we demonstrated that the SHR fetus was growth-retarded at d 20 of gestation compared with the control WKY strain and had a relatively large placenta(32). In the present study we found that the ontogeny of growth retardation is complex. The SHR fetus is small in relation to WKY from d 16 to 20 of gestation and then grows rapidly from d 20 catching up to the WKY on d 22. This indicates that there are two phases of growth in the SHR fetus, the first phase up to d 20, when growth in SHR fetus is slower than in the WKY, and a second phase from d 20 to 23 when fetal growth is faster in the SHR. The difference in gestation length between SHR and other Wistar strains, including WKY, is critical in biologic terms because many maturational changes occur at this time in late gestation. The extension of this phase may change the synchrony between successive stages of differentiation.

There are several other models in which intrauterine growth disturbance is associated with elevated blood pressure in later life, in which fetal and placental growth have been reported. In the studies of Woodall et al.(26) and Crowe et al.(1), both fetal and placental weight were decreased, whereas in that of Langley-Evans et al.(2) fetal and placental weight were increased. This suggests that in the rat there is no common aberrant pattern of fetal and placental growth that is associated with elevated blood pressure in later life.

Because dry weights remained in constant proportion to wet weights throughout most of the study period, the rapid growth appears to be due to an increase in tissue mass and not due to mechanisms related to tissue edema. An exception is in the SHR newborn, which appear to be under hydrated by 8.5%. The report of Erkadius et al.(33) suggested impaired fluid and electrolyte reabsorption in the SHR before birth. This may contribute to the dehydration, low correlation between weight and length, and reduced ponderal index in the d 1 SHR neonate. The low ponderal index in the d 1 SHR neonate is of interest as, in human infants, low ponderal index has been associated with the development of hypertension and insulin resistance in the adult(8, 34). Postnatally, at d 5 and 15, the SHR again became growth-retarded compared with WKY. These alterations in postnatal development may reflect differences in the growth patterns between the strains(35). Alternatively the slow postnatal growth may be caused by poor nutrition due to inadequate nursing by the SHR dams(18, 36), which have been shown to produce milk that has a reduced protein content and altered electrolyte concentration(37).

The relative increase in body length between d 20 and 22 is consistent with the catch-up growth phase in terms of total weight. However, the proportionately heavier heart, kidney, and lung in the fetal SHR indicates that this growth is disproportionate. Abnormal kidney development may relate to the altered renal function previously demonstrated in fetal SHR and may influence the subsequent development of hypertension(33).

There was also disproportionate growth of the SHR placenta, which became significant between fetal d 18 and 20 and increased to term. Placental hypertrophy in SHR occurs before the fetal catch-up growth phase. The mechanisms responsible are not known but the increase in placental size must require nutritional resources, possibly at the expense of the fetus. In normal sheep pregnancy the uteroplacental unit may consume up to two-thirds of the glucose delivered from the maternal circulation, passing only one-third to the fetus(38). During the period of placental overgrowth in the SHR the placenta's nutrient demand may be even greater. Nevertheless, when the placenta has achieved greater mass, and possibly increased capillary-villous surface area, it may allow an increase in nutrient transfer to the fetus and facilitate the catch-up growth phase. Erkadius et al.(33) have reported a less severe (≤7%) placentomegaly in the SHR that occurred only after d 21. Placentomegaly has been observed in rats subjected to certain levels of protein undernutrition(27) and in human and rat pregnancies complicated by diabetes(39, 40).

The mechanisms controlling the fetoplacental growth spurt in the SHR during late gestation are not obvious. There is strong evidence that maternal hypertension in SHR lessens in late gestation(41, 42). This may possibly allow more efficient uteroplacental function, thus providing the fetus with the substrates for increased growth. There is also evidence that the transfer of nutrients, principally glucose, amino acids, and oxygen from maternal to fetal blood may not only determine the supply of substrates for growth but may also influence the fetal endocrine system(43). The analyses of blood glucose reported here show that SHR fetuses are hypoglycemic in late gestation compared with WKY. This is in contrast to maternal SHR glucose levels that are higher than in WKY dams(32). This observation supports the hypothesis that there is a poor capacity of the SHR uteroplacental unit for transferring glucose to the fetus, despite the more favourable concentration gradient from maternal to fetal blood. Low fetal glucose in the SHR may also reflect altered fetal and placental metabolism. The high levels of lactate in SHR fetuses may be caused by hypoxemia(44). In the fetal sheep there is evidence that increased fetal lactate may result from increased glucose metabolism by the placenta(38). These differences in glucose and lactate persist in the SHR during late gestation despite placentomegaly and decreased maternal blood pressure(41, 42).

The low fetal hematocrit in the SHR suggests long-term fetal undernutrition or iron deficiency. Low hematocrit may result in poor oxygen delivery to the tissues and may contribute to the increased lactate and IUGR seen in the SHR at this age(45). It is interesting that, although hematocrit is elevated in the adult SHR, fetal hematocrit is low(46).

IGF-I and IGF-II are important fetal growth factors(47, 48). Plasma IGF-II rose on the last day of gestation in both strains, suggesting an association with parturition. A rise in IGF-II in late gestation has not previously been reported. The rise in IGF-II levels occurred after the period of catch-up growth between d 20 and 22 in the SHR. Unlike IGF-II, plasma IGF-I levels were not significantly different between SHR and WKY and did not increase on either d 22 or 23 of gestation. In many models of IUGR, IGF-1 is decreased(26, 49). However, at fetal d 20, when the SHR is still growth-retarded, both IGF-I and IGF-II levels were normal. As fetal d 20 is immediately before the onset of catch-up growth in the SHR, any decrease in the IGFs that had been present may already have normalized.

IGFBPs modulate the effects of IGFs, IGFBP-2 is the major IGFBP in fetal rat blood and amniotic fluid, and binds both IGF-I and IGF-II(50). IGFBP-2 gene expression in the fetal liver was decreased in SHR compared with WKY at d 20 of gestation, a time when placental growth accelerates but before fetal catch-up growth occurs. Similarly, plasma IGFBP-2 levels were decreased in the late gestation SHR, but recovered immediately after birth. The low level of IGFBP-2 gene expression in the fetal liver may be the cause of the reduced levels of IGFBP-2 seen in plasma. The lower IGFBP-2 activity occurs at a time when placental growth may be altering the availability of nutrients to the fetus. Thus, the differences in production and activity of IGFBP-2 may be related to the metabolic and nutritional differences observed between the SHR and WKY. Although IGFBP-2 is lower in the fetal SHR, it remains by far the most abundant IGFBP compared with IGFBP-1, IGFBP-3, and IGFBP-4, IGFBP-2 remains low during the subsequent increase in fetal growth, suggesting that low IGFBP-2 may be associated with rapid growth in this model. This contrasts with chronic undernutrition and chronic hypoxia in the pregnant rat where the growth rate in late gestation remains low and IGFBP-2 in plasma is increased(26, 29). A lower total binding capacity for IGF-I and -II in the extracellular environment may allow the IGFs to be more available to interact with cell surface receptors and to promote growth.

The SHR is an established genetic model of adult hypertension, but there is strong evidence that the perinatal environment is a determinant of this hypertension. Cross-fostering the SHR neonate changes their hypertensive development(18, 36), and treatment with angiotensin-converting enzyme inhibitors during pregnancy and suckling prevents the development of hypertension(19). It therefore seems likely that one or more of the genetic influences causing adult hypertension may be activated or programmed early in life. The findings reported here indicate substantial disturbances in fetal, neonatal, and placental growth, as well as in metabolic and endocrine systems in the SHR. Similar disturbances in perinatal growth are associated with the development of hypertension in the human(7) and rat(1–5). This raises the possibility of common underlying mechanisms linking early growth disturbance with the development of adult hypertension. We suggest that further investigation into the causes of the perinatal growth disturbance in SHR may uncover such a mechanism and provide a suitable model for investigating this association in the human.