Main

IUGR caused by uteroplacental insufficiency is a frequently occurring complication of pregnancy that has both immediate and long-term effects. Growth-retarded fetuses are at risk for intrauterine mortality and neonatal morbidities(1). Among the more serious of the long-term consequences is the failure of these infants to achieve their full genetic growth potential into adulthood(24). The cellular mechanisms for somatic growth retardation are not known. During intrauterine life a limitation of maternally derived metabolic fuels along with altered gas exchange due to reduced uteroplacental blood flow is believed to retard fetal growth(58). The mechanisms responsible for retarding postnatal growth are less clear.

Uterine artery ligation in the pregnant Sprague-Dawley rat is a well characterized model of IUGR that mimics uteroplacental insufficiency in the human(57, 911). Growth-retarded fetal rats are hypoxic, acidotic, and have reduced plasma concentrations of glucose, amino acids, and insulin(57, 911). These fetuses have altered hepatic energy and redox states. Fetal hepatic ATP/ADP and cytosolic NAD+/NADH ratios are increased, whereas the mitochondrial NAD+/NADH ratio is decreased(6). Although it is likely that changes in metabolic fuel availability and hepatic oxygenation trigger these changes, the molecular mechanisms underlying these alterations are unknown.

We used a RT-PCR differential display technique to compare fetal hepatic mRNA populations of IUGR and normal fetuses(1216). In this technique, mRNA is reverse transcribed, and partial cDNA sequences are synthesized, then amplified by PCR using random primers. cDNA segments that differ between the two populations are then isolated and sequenced. This sensitive and highly specific method allows simultaneous comparisons of multiple RNA samples from different cell types.

Using this technique, we isolated and sequenced 18 different partial cDNA segments from normal and IUGR fetal and neonatal liver. One sequence contained a segment of the gene encoding NAD dehydrogenase (ubiquinone) chain 4L(ND-4L). This finding, in light of our previous observation of altered hepatic redox and energy states in the IUGR fetus, suggested that altered expression of genes involved in the generation of ATP and NADH may be associated with IUGR. To test this hypothesis, we used semiquantitative RT-PCR to measure mRNA levels encoding other proteins that maintain energy balance and redox state, as well as normal mitochondrial function. Our data indicate that maternal uterine artery ligation alters these critical metabolic variables by directly affecting gene expression.

METHODS

Animals. Time-dated, pregnant Sprague-Dawley rats(Indianapolis, IN) were individually housed under standard conditions and allowed free access to rat chow and water. On d 19 of gestation (term is 21.5 d), pregnant female rats (n = 16) were anesthetized with intraperitoneal xylazine (8 mg/kg) and ketamine (40 mg/kg), and both uterine arteries were ligated. Sixteen pregnant sham-operated (laparotomy using xylazine and ketamine anesthesia) maternal rats were used as controls. Rats recovered within 2 h and had ad libitum access to food and water. On d 20, eight animals in each group were killed. The fetuses were immediately delivered, decapitated, and the livers removed and stored in liquid nitrogen for later RNA isolation. The remaining eight pregnant rats in each group were allowed to continue to term and to deliver. Litter size was reduced to six pups, and all pups were allowed free access to maternal milk. On d 4 of neonatal life, the pups were killed by decapitation, and the livers were removed and stored in liquid nitrogen. These protocols were approved by the Animal Care and Use Committee of Children's Memorial Institute for Education and Research.

mRNA differential display. RNA was prepared using the acid guanidinium thiocyanate-phenol-chloroform method(17) and quantitated using UV absorbance at 260 nm. cDNA was prepared from 2 μg of RNA using an oligo(dT)11GC primer and Superscript RT (Life Technologies, Inc., Gaithersburg, MD) as suggested by the manufacturer. After incubation at 42°C for 1 h, the reaction mixture was heated at 65°C for 5 min, diluted 1:100 with water, and stored at -20°C. The cDNA was amplified using the PCR and oligo(dT)11GC and the octamers, 5′-GTAGCTCC-3′, 5′-AGTCCTGG-3′, and 5′-TCCAGAAC-3′, in separate reactions. Each PCR was carried out in a reaction volume of 20 μL containing cDNA, 0.5 pmol of each primer, 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 1.5 mM MgCl2, 10 μmol dNTP, 1 μCi of α-35S-dATP (Amersham Corp., Arlington Heights, IL), and 0.3 unit of Taq DNA polymerase using a DNA Thermal Cycler (Perkin-Elmer, Norwalk, CT). The amplification consisted of initial denaturation at 94°C for 5 min and 40 cycles of denaturation at 94°C for 1 min, annealing at 40°C for 1 min, and extension at 72°C for 1 min, and a final extension at 72°C for 10 min. The PCR products were separated on a 5% polyacrylamide sequencing gel, which was dried and exposed to autoradiographic film for 48-72 h. PCR products that were unique or whose relative abundance differed between cDNAs from IUGR and control liver RNA were eluted from the dried gels and reamplified using conditions identical to those described above. The reamplified PCR products were cloned into pGEM-3Z (Promega Biotech Inc., Madison, WI) and sequenced using Sequenase Version 2.0 (Amersham). The sequences of the PCR products were compared with those in the nonredundant nucleotide and peptide database at the National Center for Biotechnology Information using the Blast network service(18).

Quantitation of mRNA levels using RT-PCR. cDNA was prepared from 1 μg of RNA using an oligo(dT)12-18 primer and Superscript RT(Life Technologies, Inc.). After incubation at 42°C for 1 h, the reaction mixture was heated at 65°C for 5 min, diluted 1:100 with water, and stored at -20°C. Amplification of specific mRNA was replicated 8-10 times in a volume of 50 μL using 10 μL of the diluted RT reaction product, 2-10 pmol of each primer (primers listed in Table 1), 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 1.5 mM MgCl2, 40 μmol of dNTP, 0.2μCi of [α-32P]dCTP, and 0.5 unit of Taq DNA polymerase. PCR consisted of an initial denaturation at 94°C for 5 min and 25-30 cycles of denaturation at 94°C for 1 min, annealing at 55-60°C for 1 min, extension at 72°C for 1 min, and a final extension at 72°C for 10 min. The PCR products were separated on a nondenaturing 5% acrylamidegel, and the radioactivity was incorporated into each specific fragment determined directly using a PhosphorImager and ImageQuant software(Molecular Dynamics, Sunnyvale, CA). The abundance of each transcript was quantified relative to that of the control which has been arbitrarily defined as unity. Statistical analyses were performed using an unpaired t test(19).

Table 1 Sequences of PCR primers

RESULTS

Growth. The body weight of growth-retarded fetuses was significantly reduced (3.05 ± 0.20 g versus 3.75 ± 0.18 g in controls, p < 0.05). At postnatal d 4, body weight of growth-retarded pups remained significantly lower (8.34 ± 0.39 gversus 10.01 ± 0.32 g in controls, p < 0.01). Liver weight was also significantly decreased in IUGR fetal rats compared with controls (0.17 ± 0.02 g versus 0.25 ± 0.02 g,p < 0.01). However, there was no significant difference in liver weight between neonatal d 4 IUGR and control pups (0.32 ± 0.044 gversus 0.328 ± 0.014 g).

mRNA differential display. Three different studies were performed using one of 64 possible oligo(dT)11 primers. These revealed multiple differences in liver mRNA expression between IUGR and control fetuses(Fig. 1). Eighteen of these differentially expressed products were cloned and sequenced. Of these, 11 products were increased and 7 were decreased in IUGR fetal liver. In each of the three studies, sequences that were increased in the IUGR fetus showed sequence homology to a portion of the ND-4L mRNA(20). The remaining products were not homologous with any known sequence.

Figure 1
figure 1

Representative autoradiograph showing the effect of maternal bilateral uterine artery ligation on hepatic gene expression using mRNA differential display. cDNA was amplified with an oligo(dT)11GC primer and 5′-GTAGCTCC-3′ octomer. The arrows identify bands that were isolated, cloned, and sequenced. The 205-bp fragment derived was ND-4L mRNA.

Semiquantitative RT-PCR. The finding of markedly elevated levels of ND-4L in liver tissue of IUGR fetal rats suggested that IUGR may be associated with altered expression of genes regulating energy metabolism. We used a semiquantitative RT-PCR-based procedure to measure levels of mRNA of five other genes involved in maintaining energy balance and redox state as well as normal mitochondrial function: ANT-2(21), G6PD(22), MMD(23), OTC(24), and PFK-2(25). As shown inFigure 2, each of these enzymes is important for ATP or NADH production. Although OTC is not directly involved in energy production, its activity changes in response to alterations in substrate supply (amino acids); as transplacental amino acid availability is limited under conditions of uteroplacental insufficiency, fetal hepatic OTC levels may be altered.

Figure 2
figure 2

Enzymes and metabolic pathways studied in IUGR fetuses and neonates.

The RT-PCR procedure confirmed the increase in ND-4L seen with mRNA differential display. The level of ND-4L liver mRNA was increased 10-fold in IUGR fetuses and 3-fold in IUGR neonates (p < 0.01)(Figs. 3 and4). Similarly, the levels of ANT-2, G6PD, and MMD were elevated 10-30-fold in liver of IUGR fetuses (p< 0.01) (Figs. 3 and4). Levels were also increased, albeit to a lesser degree, in IUGR neonatal rats as mRNA levels were 3-16 times greater than controls (Figs. 3 and4). By contrast, abundance of hepatic PFK-2 mRNA was not altered in IUGR fetuses. Interestingly, PFK-2 mRNA abundance was significantly increased in IUGR neonates and were 14 times greater than that of control (p < 0.01) (Figs. 3 and4). Hepatic OTC mRNA levels were reduced 65% in IUGR fetuses (p < 0.01)(Figs. 3 and4). OTC levels were increased 7-fold in IUGR neonates (p < 0.01) (Figs. 3 and4).

Figure 3
figure 3

Representative PhosphorImager results showing the effect of maternal bilateral uterine artery ligation on hepatic mRNA levels in IUGR and control (Con) fetuses (upper row) and neonates (bottom row). Semiquantitative RT-PCR was carried out as described in“Methods.” The PCR products, separated by electrophoresis on a 5% polyacrylamide gel, were scanned using a PhosphorImager and analyzed with ImageQuant software (Molecular Dynamics, Sunnyvale, CA).

Figure 4
figure 4

Effect of maternal bilateral uterine artery ligation on hepatic mRNA levels. The relative mRNA levels in the fetal liver 24 h after ligation and in the neonate 4 d after birth are compared with those of age matched controls which have been arbitrarily been assigned a value of one. The RT-PCR results were quantified using a PhosphorImager and ImageQuant software(Molecular Dynamics). Values represent ± SE for 8-10 independent RT-PCR reactions. *p < 0.01 vs control.

DISCUSSION

Uteroplacental insufficiency in the rat causes fetal hypoxia and acidosis and limits glucose and amino acid availability. As a consequence, the IUGR fetal rat has reduced hepatic ATP levels and increased cytosolic and decreased mitochondrial NAD+/NADH ratios. This study indicates that these metabolic and physiologic changes are associated with alterations in hepatic mRNA expression. The increase in ND-4L mRNA, which encodes an important protein of oxidative phosphorylation, suggested that IUGR may be associated with alterations in mRNA encoding other proteins involved in the generation of ATP, NADH, and other aspects of mitochondrial function (Fig. 2). Thus, we examined the levels of ANT-2, G6PD, MMD, OTC, and PFK-2 mRNA in liver of IUGR and control fetal and neonatal rats. These enzymes are encoded by nuclear genes as opposed to ND-4L which is encoded by the H-strand of the mitochondrial genome.

ANT-2, G6PD, MMD, OTC, and PFK-2 have diverse functions. ANT-2 is an inner mitochondrial membrane protein involved in the transport of ATP and ADP across the mitochondrial inner membrane and thus plays a central role in mitochondrial and cytosolic ATP metabolism. MMD, an NAD-linked dehydrogenase located in the mitochondria, plays an important role in the metabolism of three-carbon moieties by catalyzing the dehydrogenation of L-malate to L-oxaloacetate. MMD is also directly involved in energy production through the reduction of NAD+ to NADH. OTC is also localized within the mitochondria and is a component of the urea cycle. Its activity is modulated by amino acid supply. G6PD and PFK-2 are cytosolic enzymes. G6PD is an important regulatory enzyme in the pentose phosphate shunt(Fig. 2), a pathway that generates NADPH, and is especially important in tissues such as the liver which are involved in the synthesis of fatty acids. PFK-2 is a key regulatory enzyme of glycolysis and gluconeogenesis.

The increased levels of hepatic ND-4L, MMD, and ANT-2 mRNA in the growth retarded fetus represents a response by the hepatocyte to events associated with maternal bilateral uterine artery ligation and subsequent growth retardation. The decrease in oxygen, glucose, and amino acids to the fetal hepatocyte decreases ATP availability and the NAD+/NADH ratio(6). It is likely that decreased ATP levels induce the expression of ND-4L and ANT2, whereas the decrease in the mitochondrial NAD+/NADH ratio induces the expression of MMD. Interestingly, the decrease in energy production in the IUGR fetus is transitory as both energy and redox states return to normal after birth(6). The induction of ND-4L, MMD, and ANT-2 gene expression may contribute to this recovery.

The observation that NADH dehydrogenase, malate dehydrogenase, and ANT2 mRNA levels remain increased in the IUGR neonate is particularly intriguing because many metabolic variables that are altered in the growth retarded fetus, i.e. hypoxia, acidosis, and hypoglycemia, return to normal in the neonate. Based upon measures of oxygen consumption and indirect calorimetry, the metabolism of the growth retarded neonate is not completely normal. For example, IUGR neonates are hypermetabolic(2628). The persistent elevation of ND-4L, MMD, and ANT-2 mRNA suggests that the hypermetabolic state results in part from these changes in neonatal liver.

PFK-2 mRNA levels were normal in the IUGR fetus, suggesting that the many metabolic and physiologic factors altered in the IUGR fetus do not affect PFK-2 gene expression. In contrast, PFK-2 mRNA levels in the IUGR neonate were markedly increased. This elevation probably stimulates gluconeogenesis to a greater extent than in the normal newborn. The preferential stimulation of gluconeogenesis at the expense of glycolysis by increasing levels of PFK-2 may be a consequence of the limited hepatic glycogen stores of the IUGR newborn.

The increase in G6PD in IUGR liver may represent a means for the hepatocyte to increase the amount of glucose shunted through the pentose-phosphate pathway, which in turn increases production of the antioxidant NADPH. Oxidative stress has been shown to induce G6PD in rat hepatocytes(29), and the increase in G6PD mRNA in the IUGR fetus may be a response to such stress and reflect a need to increase production of the antioxidant NADPH. G6PD mRNA levels were also increased in the neonatal rat.

Finally, we speculate that the decrease in OTC mRNA levels may be secondary to a decrease in amino acid availability and ATP production in the growth-retarded fetus. As amino acid availability and energy production return to normal at birth, OTC mRNA levels also increase to meet the metabolic demands of the growth-retarded neonate.

Little is known of the specific molecular mechanisms involved in transcriptional regulation or mRNA stability for the genes encoding these enzymes. It is likely that both mechanisms are responsible for the changes observed in gene expression in SGA liver. Further studies are required to elucidate these mechanisms.

Uteroplacental insufficiency alters a number of metabolic and physiologic variables in the fetus which force adjustments by the hepatocyte to ensure survival during intrauterine and neonatal life. mRNA differential display and RT-PCR identified changes in the gene expression of a number of hepatic enzymes involved in numerous metabolic functions. These findings underscore the profound effect of uteroplacental insufficiency upon hepatic gene expression. mRNA differential display and RT-PCR techniques are well suited for studying the changing pattern of gene expression in fetal and neonatal tissue, as they require only small amounts of RNA and are sensitive and specific. However, they are not a direct measurement of protein levels or enzymatic activity, both of which may be regulated independently of gene expression. Nonetheless these techniques have allowed examination at the molecular level of effects of intrauterine hypoxia and substrate deprivation on fetal and neonatal growth and development.