Main

NO is an important messenger molecule implicated in multiple physiologic and pathophysiologic events in both the peripheral and central nervous system(1, 2). NOS (EC 1.14.13.39) catalyzes the oxidation of L-arginine to NO plus citrulline(3). Although the calcium-dependent NOS which are normally present in the brain and other tissues have been described as constitutive, there is evidence for the regulation of their expression by estrogen(4). Proposed roles for NO within the CNS include synaptic plasticity and memory via long-term potentiation in the hippocampus and depression in the cerebellum(5–7), control of blood flow by neural activity(8, 9), and the establishment and activity-dependent refinement of axonal projections during the later stages of development(10). Proposed pathophysiologic roles for NO in the CNS include brain dysfunction associated with inflammation(11), the mediation of glutamate toxicity in cerebral ischemia(1, 12, 13), and epilepsy(14–17). Each of these roles could be important in the fetus and newborn, but there is no published data on the development of NOS activity in the brain before birth and little on its activity in the neonate. Quantitative data on the appearance of functional NOS may be particularly important in understanding the potential role of NO in brain ischemia. Qualitative immunohistochemical data on the rat brain show that there is a redistribution of NOS protein expression within the brain during the neonatal period, but do not show whether the overall activity changes during this period(18), whereas NOS activity measurements suggest that the development of functional NOS in this species is predominantly postnatal(19).

In the present study we have determined the ontogeny of functional NOS by measuring its activity in fetal and neonatal forebrain and cerebellum from two species with very different developmental profiles: the guinea pig and the rat. We also provide evidence for the involvement of the estrogen-estrogen receptor system in the ontogeny of NOS.

METHODS

Pregnant Duncan Hartley guinea pigs with a known date of conception (term 63 d average) and Wistar rats (term 21 d average) were obtained from a commercial breeder (Charles River). Fetal and neonatal brains were rapidly removed under general anesthesia (pentobarbitone 120 mg/kg intraperitonally for the mother or 30 mg intraperitonally for the neonate), divided into sections of forebrain and cerebellum, freeze-clamped in liquid nitrogen, and stored at -70°C until studied. Whole guinea pig fetal brain was used at 31 d (0.49 gestation) because of the small size.

Estrogen receptor blockade. To test the role of estrogen in changing NOS activity during gestation, guinea pigs at full term (either at 58 or 61 d of gestation) received four doses of the estrogen-receptor partial agonist tamoxifen (250 μg intraperitonally twice a day), and the fetal brains were removed 6 h after the last injection (on d 60). Neonatal brains were removed within 16 h of birth. Animals of the same gestational age were used for controls. Although tamoxifen can have both agonist and antagonist activities in different tissues, it acts predominantly as an antagonist in the brain(20). Tamoxifen treatments longer than 2 d were not attempted because of the risk of premature termination of pregnancy.

NOS activity. The frozen tissue was homogenized (with a Ystral homogenizer) at 0°C in 5 volumes of buffer containing 320 mM sucrose, 50 mM Tris, 1 mM EDTA, 1 mM DL-DTT, 100 μg/mL phenylmethylsulfonyl fluoride, 10 μg/mL leupeptin, 10 μg/mL soybean trypsin inhibitor, and 2 μg/mL aprotinin brought to a pH 7.0 at 20°C with HCl. The crude homogenate was centrifuged at 0°C at 15,000 × g for 20 min, and the pellet was discarded. NOS activity was determined in the postmitochondrial supernatant within 1 h of preparation by measuring in duplicate the conversion of L-[U-14C]arginine to L[U-14C]citrulline as previously described in detail(21). Total activity was determined by calculating the difference between the [14C]citrulline produced from control incubations and the incubations containing both 1 mM EGTA to bind calcium and 1 mM Nω-monomethyl L-arginine to inhibit NO synthase(21). The activity of calcium-independent NO synthase was determined by calculating the difference between the incubations containing 1 mM EGTA alone and the incubations containing both 1 mM EGTA and 1 mM Nω-monomethyl L-arginine. Calcium-dependent activity was calculated by subtracting calcium independent activity from total activity. Measurements are presented as picomoles of citrulline/min/mg of protein. The protein was measured in an aliquot of the crude homogenate using bicinchoninic acid(22). Intra- and interassay variations were each less than 8%.

Chemicals and statistical analyses. L-[U-14C]Arginine was obtained from Amersham Corp.; all other chemicals were from Sigma Chemical Co. The results are presented as the mean ± SEM. The day of copulation was considered to be the first day of pregnancy. Statistical significance was assessed by the t test. A p < 0.05 was considered to indicate either a significant correlation or difference among means.

RESULTS

Guinea pig . Cerebellum. Whole brain NOS activity was at the lower limit of detection at 0.49 gestation (1.9 ± 0.9 pmol/min/mg of protein). Thereafter, the calcium-dependent NOS activity increased rapidly, reaching the activity present in the adult by 0.62 gestation (24 d before delivery). NOS continued to increase until d 6 after birth, reaching 250% of the adult activity, and declining thereafter (Fig. 1).

Figure 1
figure 1

NO synthase activity in the forebrain (closed squares) and cerebellum (open circles) of the guinea pig fetus and neonate. NO synthase activity was measured by the conversion of L-arginine to L-citrulline and is shown as mean ± SEM (3-8 animals/group).

Full size image

Forebrain. The calcium-dependent NOS activity increased rapidly after 0.49 gestation, reaching the activity present in the adult by 0.95 (5 d before delivery). In contrast to the cerebellum, NOS activity increased between 0.95 gestation and 1 d of age. Thereafter, activity plateaued at 140% of the adult level until d 6 after birth. By d 10, NOS activity had declined to the adult level (Fig. 1). Calcium-independent NOS activity was observed in both tissues, but on average it was <5% of the total activity and was unaltered during gestation.

Estrogen receptor blockade. Two days of treatment with tamoxifen in the guinea pig caused significant reductions in forebrain and cerebellum calcium-dependent NOS activities in the fetus and in the neonate (Fig. 2).

Figure 2
figure 2

Effect of tamoxifen treatment on NO synthase activity in the forebrain and cerebellum of the guinea pig fetus (60 d) and neonate (1 d). Pregnant animals received four doses of tamoxifen (250 μg intraperitonally). The data are shown as means ± SEM (3-8 animals/group) *p < 0.05 (t test).

Full size image

Rat. NOS activity was low at 0.8 gestation (forebrain 16± 1.1, cerebellum 9 ± 1.3 pmol/min/mg of protein,Fig. 3). By 12 d (forebrain) or 20-30 d (cerebellum) the activity was similar to that of the adult. Thereafter, NOS increased to 130-150% of the adult level by 30 d of age. Calcium-independent NOS activity was low and was unaltered in both tissues.

Figure 3
figure 3

NO synthase activity in the forebrain (closed squares) and cerebellum (open circles) of the rat fetus and neonate. NO synthase activity was measured by the conversion of L-arginine to L-citrulline and are shown as means ± SEM (3-8 animals/group).

Full size image

DISCUSSION

We have observed a maturational process which begins after 0.49 gestation in the guinea pig but which does not begin until after 0.8 gestation in the rat. Adult NOS activity was reached by 0.62 gestation in the cerebellum and 0.95 in the forebrain in the guinea pig but was not reached until later(approximately 10-30 d after birth) in the rat.

Little is known about the regulation of the expression of the constitutives NOS (neuronal and endothelial)(3). Pregnancy is characterized by a high concentration of circulating estrogen and this steroid is known to increase the transcription of a number of enzymes. We have recently shown that estrogen induces neuronal and endothelial NOS in a range of tissues, including the brain(4). Estradiol crosses both the blood-brain barrier and the placenta, resulting in increased NOS activity in maternal tissues early in pregnancy, reaching a maximum value by 0.3 of gestation(23). Therefore maternal estrogen could alter fetal brain NOS activity if either placental transport of estrogen or the number of fetal estrogen receptors increased with advancing gestation.

The quantity of estrogen receptors within the fetal guinea pig brain increases with advancing gestation and then decreases after birth(24). We plotted the concentration of cytosolic estradiol-specific receptors in the fetal guinea pig whole brain(24) against the NOS activity measured in the current study The two curves are almost superimposable (Fig. 4), so that, as the concentration of available receptor sites increases, so does the NOS activity after a short delay. Thus, it is reasonable to hypothesize that estrogen enhances the synthesis of NOS in the fetal guinea pig brain. A prediction which would follow from this hypothesis is that the treatment of pregnant animals with tamoxifen (which acts in the brain as an estrogen antagonist)(20) at the end of gestation should reduce NOS activity in the fetal brains, and our results show that this does occur. Thus the increase in calcium-dependent NOS activity in the fetal guinea pig brain during pregnancy is, at least in part, mediated by estrogen. These results suggest that, despite the presence of the highly specificα-feto/neonatal estradiol-binding plasma protein and the related fetal/neonatal estradiol-binding protein in brain cytosol(25), sufficient free estradiol must be present in the fetus to exert biologic effects.

Figure 4
figure 4

NO synthase activity (closed squares) of the guinea pig fetus and neonate forebrain plotted against the number of estradiol cytosolic receptors (open circles) as reported by Pasqualiniet al.(24).

Full size image

In the rat brain, the estrogen receptor concentration increases at a later age than in the guinea pig brain. Whereas the most rapid increase in estrogen receptors occurs between 0.5 and 0.8 of gestation in the guinea pig(24), it occurs at around term in the rat(26, 27). Thus in both species the estrogen receptor concentration rises immediately (4-6 d) before the rise in NOS activity in the brain. Furthermore, in early neonatal rat brain, the estrogen receptors are most abundant in the cerebral cortex(26). In the present study, we have found approximately 2-fold more NOS activity in the forebrain than in the cerebellum at this stage (postnatal d 3,Fig. 3). Our data on the ontogeny of NOS activity in the rat forebrain is consistent with that of Matsumoto et al.(19) and with recent reports that the development of NADPH diaphorase-positive neurones (presumed to reflect NOS-containing neurones) and NOS immunostaining occur predominantly during the first 2 wk of postnatal life(18, 28, 29).

The increase in NOS may play several roles in the fetus and neonate. It has been shown that estrogen can be a growth factor to central nervous cells(30–33). The increase in NOS activity in rat and guinea pig correlates with the development of the estrogen receptor population, as we have seen above, and precedes the development of synaptic structures in these two species. In the guinea pig, studies using a range of methods show that the most rapid increase in the number of synaptic junctions occurs during the last days of gestation and the 1st wk of life(34–36). In contrast to the guinea pig, the number of synapses in various regions of the rat brain does not increase markedly until 2 and 3 wk postnatally(34–38). The predominantly prenatal and postnatal synaptogenesis, in the guinea pig and rat, respectively, is reflected in the general neurologic maturity of these two species at birth. The guinea pig is born essentially functional and with its eyes open, whereas the rat is born with limited mobility and with its eyes closed. Our studies, demonstrating that the increase in NOS activity in brain occurs approximately 10 d before the rapid phase of synapse formation in both guinea pigs and rats, support the hypothesis that NO formation may be important in the physiology and development of the synapse(10). Very recent data showing that NOS appears postnatally in the mouse brain and again preceding the rapid phase of synaptogenesis(39) are also consistent with this hypothesis. In the present study we have not demonstrated that the brain NOS activity measured can be attributed to neuronal NOS, although studies of adult brain NOS in several species and of developing brain NOS in the mouse make this likely(39).

NO is involved in the regulation of cerebral blood flow(9, 40), increasing local cerebral blood flow in response to neuronal activation by dilating resistance vessels. Although the normal fetal environment is not characterized by profound hypoxemia, it is one of relative hypoxemia which intensifies as gestational age advances. It is attractive to hypothesize that the elevation of NOS above adult levels may be a mechanism to cope with the demand of brain regions which are metabolically or physiologically highly active, providing a neuroprotective effect by enhancing blood flow. It is known that cerebral vascular resistance decreases and blood flow increases with advancing gestational age. The increased production of NO around blood vessels in the brain could be responsible for the decreased resistance and increased volume of flow.

Another possible role of NO that our findings might impact upon relates to memory. The newborn must form a large number of memories critical for survival. We observed high levels of NOS activity in both the forebrain and cerebellum during the first days of life in the guinea pig and at 3 wk of age in the rat. A growing body of evidence suggests that NO may act as a retrograde messenger for long-term potentiation and thus memory formation(5–7). The high levels of NOS activity that we have seen may therefore play a role in memory formation. It is not clear how the ontogeny of NOS will relate to the pathophysiologic roles suggested for NO, such as ischemic damage or epilepsy. The data presented will clearly provide the basis of studies to elucidate these roles, for example by examining the susceptibility of animals of different ages to ischemic brain damage.

In conclusion NOS activity increases in both the forebrain and cerebellum predominantly prenatally in the guinea pig and predominantly postnatally in the rat. The sequence observed in both species, of increased estrogen receptors followed by increased NOS activity followed by synaptogenesis, is consistent with the hypothesis that estrogen may play a crucial role in the ontogeny of NOS and suggests that NO formation may play an important role in synaptogenesis. The high levels of NOS activity may also have important functions in promoting cerebral blood flow and memory formation during the periods of most rapid brain development.