Main

NPY, a 36-amino acid peptide and the most abundant neuropeptide in the CNS(1), inhibits seizures(2), mediates anxiolysis(3,4), inhibits sexual behavior(5,6), causes hyperphagia and obesity(7), modulates vasomotor stability(8), and regulates hormonal secretion and metabolic control(9). In contrast, the 42-amino acid CRF is a stress hormone, which causes anxiety, anorexia, and has endocrine control over corticosteroid hormonal secretion(10–13). The net balance between these two neuropeptides with opposing effects determines the ultimate biologic action resulting in the phenotypic outcome(14). The metabolic and hormonal factors which regulate this balance between brain NPY and CRF concentrations have been identified in adult animals(13–15). In contrast, although NPY and CRF have been identified in fetal brain(16,17), the factors which regulate fetal and postnatal brain NPY and CRF remain unknown.

The adaptive responses of hunger, hyperphagia, and a reduction in energy expenditure are hallmarks of diabetes mellitus with hyperglycemia and hypoinsulinemia, and starvation with hypoglycemia and hypoinsulinemia. These responses are partly mediated by the enhanced adult rat NPY synthesis and concentrations without a concomitant change in CRF levels(14,15). On the other hand, the adipocytic release of leptin, which increases with obesity(18), provides a feedback control on adult rat brain NPY synthesis, by signaling the fat mass to the brain and decreasing food intake and body weight(19). Although no similar investigations exist in the fetus and newborn, we have recently observed that isolated hyperglycemia in the absence of changes in circulating insulin levels as seen in the fetus of a diabetic rat pregnancy and isolated acute fetal hyperinsulinism separately suppressed fetal brain NPY synthesis and concentrations(20).

Uteroplacental insufficiency, which causes asymmetrical IUGR is associated with fetal hypoxia, acidosis, hypoglycemia, hypoinsulinemia, altered amino acid and mitochondrial metabolism(21–24), along with altered IGF-I, IGF-II, and IGF-binding protein concentrations(25). This intrauterine nutritionally restrictive environment leads to a diminution in the postnatal body weight(26). We hypothesized that uteroplacental insufficiency alters the brain (hypothalamic) neuropeptide balance during the suckling phase, which in turn may predict the postsuckling feeding behavior. To test this hypothesis we used the previously characterized model of uteroplacental insufficiency(27) and examined the effect upon fetal and postnatal brain NPY and CRF synthesis, peptide concentrations and distribution, along with circulating glucose, insulin, corticosterone, and leptin levels.

METHODS

Animals. Sprague-Dawley pregnant rats were individually housed under 12-h light and dark cycles with free access to standard rat chow and water. The care and use of animals followed the guidelines established by the National Institutes of Health and was approved by the Institutional Committees for Animal Care and Use at Magee-Womens Research Institute and Northwestern University. Surgical methods were followed as previously described(27). On d 19 of gestation (term = 21.5 d), the maternal rats were anesthetized intraperitoneally with xylazine (8 mg/kg) and ketamine (40 mg/kg), and both uterine arteries were ligated (IUGR). SHAM animals that underwent the identical anesthetic and surgical procedures except for ligation served as controls. Rats recovered within 2 h and had ad libitum access to food and water. In a subset of these animals on d 20, mothers were killed by cervical dislocation or pentobarbital and the fetuses were delivered immediately. Fetuses were either used for immunohistochemical experiments or decapitated and the brains were removed and snap frozen in liquid nitrogen. Another subset of maternal rats was allowed to deliver spontaneously, the litter size was randomly reduced to six at birth, the progeny followed postnatally, and either used for immunohistochemical analysis or the brains were harvested and examined at d 4, 14, and 21.

Plasma assays. Pooled fetal and postnatal jugular vein blood was obtained, and the plasma was separated and stored frozen at -20°C until further analyses. Plasma glucose concentrations were measured by the glucose oxidase method using an automated glucose analyzer (BeckmanII; Beckman Instruments Inc., Fullerton, CA) and plasma insulin, corticosterone, and leptin levels were assessed by the double antibody RIA using species-specific antibodies and standards(25,28,29).

RNA extraction. Poly(A)+ enriched RNA was extracted from SHAM and IUGR animal brains using the Miniribosep mRNA extraction kit(Collaborative Biomedical Products, Bedford, MA). RNA was quantitated spectrophotometrically, and the purity of the samples was assessed as a ratio of 1.8-2.0 at 260/280 nM wavelength.

Northern blot analysis. Ten micrograms of poly(A)+-enriched RNA was loaded on 1.2% agarose-2.2 M formaldehyde slab gels and run at 22 V for 16 h. The gels were stained with ethidium bromide and visualized under UV light to confirm the integrity of RNA samples. The RNA from the gels were transferred to Nytran membranes (Micron Separations, Inc., Westboro, MA) and cross-linked by UV light in a stratalinker at 1200 × 100 µJ for 45 s (Stratalinker 1800, Stratagene, La Jolla, CA). The filters were prehybridized for 2 h at 42°C in 50% formamide, 0.1% polyvinylpyrrolidone, 0.1% BSA, 0.1% Ficoll, 75 mM sodium chloride, 50 mM sodium monobasic phosphate, 1.25 mM EDTA, 0.2% SDS, and 200 µg of denatured salmon sperm DNA. The filters were subjected to hybridization with 1 × 106 cpm/mL of 32P-randomly primed 511-bp rat preproNPY cDNA(30) for 24 h. The filters were washed at room temperature in 2 × SSC (1 × SSC = 0.15 M sodium chloride, 0.015 M sodium citrate, pH 7.0) and 0.1% SDS. The final wash was performed in 0.1% SSC and 0.1% SDS for 30 min at 45°C. The filters were subsequently exposed to x-ray film at -70°C for differing lengths of time until optimal resolution was achieved. The same filters were stripped and reprobed under the same conditions with a 32P-randomly primed 1.2 kb rat CRF cDNA(31). The interlane loading variability was corrected by rehybridizing the filters with an 1800-bp rat 18 S ribosomal RNA cDNA(32). The autoradiograms were subjected to densitometry, the NPY and CRF mRNA optical densities were expressed as a ratio to the 18 S rRNA OD, and the ratio was calculated as a percent of either the SHAM 21-d-old values (for the developmental studies) or the age-matched SHAM control (IUGR studies)(20,33).

RIA. The IUGR and SHAM rat brains were homogenized in four volumes of 0.1 N HCl and sonicated for 10 s (Sonic dismembrator, Fisher Scientific, Pittsburgh, PA) using 10 W output power. The extracts were centrifuged, separated into aliquots, and freeze dried. The extracts were reconstituted in 0.05 M Tris-HCl buffer containing 0.1% BSA (pH 7.8) and assayed in triplicate. Differing concentrations of the porcine NPY peptide(Peninsula Laboratory, Belmont, CA) were used for generation of the standard curve. Polyclonal rabbit anti-rat NPY antibody(34) was used in a final dilution of 1:1 000 000, and the Bolton-Hunter125 I-labeled NPY (New England Nuclear Corp., Boston, MA) was added after a 24-h incubation. Forty eight hours later 0.2 mL of 0.01% goat anti-rabbit serum (Calbiochem, La Jolla, CA) was added as the secondary antibody, and the assay mix was incubated overnight at 4°C with 0.2 mL of 19% polyethylene glycol to precipitate the antigen-antibody complex. The antigen-antibody complex was separated by centrifugation at 5000 ×g for 15 min, and the radioactivity in the pellet was assessed. The assay has a sensitivity of 15 pg/mL and the anti-rat NPY antibody exhibited no cross-reactivity with the intestinal polypeptide, cholecystokinin, motilin, or the pancreatic polypeptide, with less than 0.1% cross-reactivity with the peptide YY(19).

The CRF RIA was carried out in duplicate using the RIA kit (Peninsula Laboratories, Inc., Belmont, CA) with the rabbit anti-CRF antibody (human, rat). The sensitivity of the assay is 12 pg/mL. The anti-CRF antibody exhibited 100% cross-reactivity with rat and human CRF and no cross-reactivity with LH-releasing hormone, ACTH,[Arg8]-vasopressin, or brain natriuretic peptide-45 (rat), with less than 0.01% cross-reactivity with sauvagine and ovine CRF.

Immunohistochemistry. The 20-d fetal, 4-, 14-, and 21-d postnatal IUGR (n = 3 at each age) and SHAM (n = 3 at each age) rats were anesthetized by a combination of ketamine (40 mg/kg) and xylazine (8 mg/kg), and their brains perfused first with normal saline followed by a 4% solution of paraformaldehyde in PBS through a catheter inserted into the aorta via the left ventricle. The brains were then removed from the cranium and transferred into a 30% sucrose solution for cryopreservation. Serial rostrocaudal floating microtome coronal brain sections were obtained (35 µm at all ages, 45 µm for the fetal stage) and placed in 10 mM PBS, pH 7.2, for immunohistochemical analysis, which was performed as described previously(35). The sections were incubated with rabbit anti-rat NPY (1:8000) (Peninsula Laboratories) or rabbit anti-rat CRF (1:400) (Peninsula Laboratories), and anti-ovine CRF (1:3000) (INCSTAR Corp., Stillwater, MN) IgGs for 48 h at 4°C. Primary antibody incubations were followed by successive incubations at room temperature with biotinylated goat anti-rabbit IgG (1:600) and avidin-biotin-peroxidase complex (1:200) (Vector Laboratories, Burlingame, CA), each for 60 min. Immunolabeling was produced with 3′,3′-diaminobenzadine and enhanced with 2.5% nickel sulfate in 0.175 mM sodium acetate. The sections were mounted in rostrocaudal sequence on charged ProbeOn™Plus glass slides (Fisher Scientific, Pittsburgh, PA) and coverslipped under Histomount (National Diagnostics, Atlanta, GA). PBS buffer alone or buffer containing appropriate dilutions of the normal rabbit serum and lacking the primary antibodies were used as appropriate controls.

Data analysis. All results are presented as a mean ± SEM. Percentages were transformed before analysis(36). Intergroup differences were detected by comparing the IUGR to the age-matched SHAM group by t test. Developmental changes were analyzed by the one-way ANOVA. and intergroup differences were validated by the Newman Kuel's test.

RESULTS

Body weights of the IUGR and SHAM animals are depicted in Table 1. The IUGR animals of 20-d fetal stage showed a decline in body weight when compared with the SHAM. This decline in body weight persisted through 4-, 14-, and 21-d postnatal ages when compared with their age-matched SHAM. In contrast, the brain weights remained unchanged at all ages in the IUGR group versus the age matched SHAM(Table 1). These findings demonstrate the presence of asymmetrical IUGR similar to previous reports(21–27).

Table 1 Total body and brain weights
Full size table

The plasma insulin, glucose, corticosterone, and total leptin levels of the IUGR and SHAM animals are shown in Table 2. At fetal d 20 the mean plasma glucose and insulin concentrations were significantly lower in the IUGR group when compared with the SHAM group. The insulin and glucose concentrations during the postnatal stages were no different in the two groups. Fetal corticosterone levels could not be measured in these animals due to limited amounts of pooled plasma samples that were available at this age. We have previously measured fetal corticosterone levels in the IUGR and SHAM groups and found no difference(26). Corticosterone levels were not statistically different in the two groups at all postnatal stages examined. However, at d 4 a trend toward a decline followed by an increase at d 14 and 21 was observed in the IUGR when compared with the SHAM progeny. Similarly, we were able to assess total leptin concentrations in two pooled fetal plasma samples. Total leptin concentrations were no different between the 20-d fetal and 4-d old samples, and between the 14-d and the 21-d old samples. However, a doubling between the 4-d old and the 14-d old samples (Newman-Kuel's test = p < 0.05) was noted. The leptin levels in the IUGR progeny were no different when compared with the the age-matched SHAM group at all ages examined.

Table 2 Plasma glucose, insulin, corticosterone, and total leptin levels
Full size table

The brain NPY (0.8 kb) and CRF (1.4 kb) mRNA levels in the IUGR and SHAM animals analyzed by Northern blot analysis demonstrated a developmental increase in the expression of both these neuropeptides (Figs. 1 and 2) with additional differences at the various ages between the IUGR and the age-matched SHAM groups. Quantitation of the mRNA levels demonstrated an increase in NPY and CRF mRNA concentrations with age, the maximal amounts being noted at postnatal d 14 (Fig. 3). Comparison between the two experimental groups revealed a 75% increase in NPY mRNA levels in the IUGR fetal brain when compared with that of the SHAM(Fig. 3). This fetal increase in the NPY mRNA concentrations persisted into postnatal d 4, was no different at d 14, and was 40% higher in the IUGR at postnatal d 21. In contrast, the fetal brain CRF mRNA levels were no different between the fetal IUGR and SHAM groups. The CRF mRNA concentrations at postnatal d 4 in the IUGR was 2.5-fold higher than the age-matched SHAM levels, with no difference at d 14 and 21(Fig. 3).

Figure 1
figure 1

Brain NPY mRNA. Representative Northern blots demonstrating the 0.8-kb NPY (top panel) mRNA and the 1.8-kb 18 S rRNA (bottom panel) in the IUGR (I) and SHAM(S) groups at 20-d fetal (F), 4-d postnatal (P), 14-d P, and 21-d P ages.

Full size image
Figure 2
figure 2

Brain CRF mRNA. Representative Northern blots demonstrating the 1.4-kb CRF (top panel) mRNA and the 1.8-kb 18 S rRNA (bottom panel) in the IUGR (I) and the SHAM(S) groups at 20-d fetal (F), 4-d postnatal (P), 14-d P, and 21-d P ages.

Full size image
Figure 3
figure 3

Quantitation of brain NPY and CRF mRNA during development in the IUGR and SHAM groups. Densitometric quantitation of NPY (left panel) and CRF (right panel) mRNAs standardized to the 18 S rRNA at 20-d fetal (F), 4-d postnatal (P), 14-d P, and 21-d P ages, and expressed as a percent of the corresponding 21-d P SHAM group's mean mRNA amount. NPY mRNA in the 20-d F IUGR (n = 13) vs SHAM (n = 13) groups, 4-d P (n = 7vs 7), 14-d P (n = 7 vs 6), and 21-d P(n = 4 vs 4). CRF mRNA in the 20-d F IUGR (n = 17) vs SHAM (n = 17), 4-d P (n = 7 vs 6), 14-d P (n = 5 vs 4), and 21-d P (n = 4vs 4). ANOVA (p) = 0.0001 for both the age-dependent changes in NPY and CRF mRNA levels in the SHAM group. o = p < 0.05 when compared by Newman Kuel's test to the corresponding 21-d P mRNA concentrations; × represents the two groups that are different from each other by Newman Kuel's test (p < 0.05). t test =*p < 0.01 when the IUGR group is compared with the corresponding age-matched SHAM group.

Full size image

Developmental changes in the brain NPY and CRF peptide levels paralleled that of the corresponding mRNA amounts. An increase in brain NPY peptide concentrations was noted with peak amounts in the postnatal 21-d-old animals(Fig. 4). Similarly, brain CRF peptide concentrations increased from that in the fetus to peak amounts in those 14 to 21 d old(Fig. 4). Assessment of the differences between the two experimental groups demonstrated an increase in the NPY peptide levels of the 4- and 21-d postnatal IUGR animals with no difference at the fetal and the postnatal 14-d stages. The brain CRF peptide concentrations were no different between the two groups in the fetus, with a 2-fold increase in the postnatal 4-d IUGR animals when compared with the age-matched SHAM group. Subsequently no difference between the two groups was noted at 14 and 21 d(Fig. 4).

Figure 4
figure 4

Quantitation of the brain NPY and CRF peptide concentrations. RIA NPY peptide concentrations (left panel) in nanograms/mg of protein (left scale) and CRF peptide levels (right panel) in picograms/mg of protein (right scale) are shown in the IUGR and SHAM groups at the 20-d fetal(F), 4-d postnatal (P), 14-d P, and 21-d P ages. ANOVA(p) = 0.0001 for both the age-dependent changes in NPY and CRF peptide levels in the SHAM group. o = p < 0.05 when compared by Newman Kuel's test to the corresponding 21-d P peptide concentrations. *p < 0.05 when compared by the t test to the age-matched SHAM group. NPY peptide in the 20-d F IUGR(n = 4) vs SHAM (n = 4), 4-d P (n= 4 vs 4), 14-d P (n = 4 vs 4), and 21-d P(n = 3 vs 3). CRF peptide in the 20-d F (n = 8 vs 8), 4-d P (n = 7 vs 7), 14-d P(n = 8 vs 8) and 21-d P (n = 5 vs 7).

Full size image

Immunohistochemical analysis demonstrated the presence of NPY in cortical neurons, hippocampus, thalamus, and hypothalamus as reported previously(17). However, age-dependent changes and IUGR-induced alterations in NPY immunoreactivity were most visible in the hypothalamic region, particularly in the paraventricular and parvicellular regions(Fig. 5,A-L). At all ages, except at d 14 (not shown), IUGR demonstrated an increase in the hypothalamic NPY immunoreactivity, whereas the other regions demonstrated no major difference. Cortical neuronal NPY immunoreactivity at all ages demonstrated no difference between the IUGR and SHAM groups (Fig. 5,M and N). In contrast to NPY immunoreactivity, a more widespread distribution of CRF immunoreactive neurons was noted, making it difficult to discern the immunoreaction under low power microscopy. In addition, a lower intensity of neuronal CRF immunoreactivity was noted when compared with that of NPY (Fig. 5,M-P). This observation paralleled the RIA results where NPY was noted in nanograms/mg, whereas CRF was present in picograms/mg (Fig. 4). The CRF immunoreactive neurons appeared scattered both in the hypothalamus and cortex (Fig. 5,O and P). Again CRF immunoreactivity in neurons was higher only in the d-4-old IUGR brain sections when compared with SHAM (Fig. 5,O and P).

Figure 5
figure 5

Immunohistochemical analysis-NPY and CRF studies. Representative photomicrographs demonstrating coronal brain sections of the 20-d fetus (A-D), 4-d (E-H), and 21-d(I-L) postnatal rats from the IUGR (C, D; G, H; K, L) and SHAM (A, B; E, F; I, J) groups. Whole brain sections (A, C; E, F; I, J) and the hypothalamic regions (B, D; G, H; K, L) are seen. Fourteen-day-old cortical neurons demonstrating NPY (M, N) and 4d old supraoptic nuclear neurons showing CRF (O, P) immunoreactivity in the IUGR (N, P) and SHAM (M, O) groups. lhn, lateral habenular nucleus; ic, internal capsule, v3h, third ventricle; v3e, third ventricle epithalamic;pa, hypothalamic paraventricular nucleus; pva, anterior thalamic paraventricular nucleus; cpu, caudate putamen (striatum); long arrows = immunoreactive neuronal cell bodies (M-P).

Full size image

DISCUSSION

Our present investigation is the first attempt at demonstrating fetal and postnatal changes in brain NPY and CRF mRNA and protein concentrations secondary to uteroplacental insufficiency. Maternal uterine artery ligation in the rat causes asymmetrical IUGR due to nutritive restriction(21–27). This in utero restrictive environment is evidenced by the presence of fetal hypoxia, acidosis, hypoglycemia, hypoinsulinemia, altered amino acid, and mitochondrial metabolism(21–24), along with diminished circulating insulin-like growth factors and increased IGF-binding proteins(25). Despite removal of this restrictive environment postnatally and a normalcy in circulating glucose and insulin concentrations, the IUGR progeny continues to be growth-retarded during the suckling phase.

The developmental changes in brain NPY and CRF mRNA and peptide levels in the SHAM group mimicked those reported previously in unmanipulated animals(16,37). In both cases, an increase in mRNA and peptide levels was noted with advancing age, with peak amounts noted either at 14 or 21 d. This developmental pattern was altered postnatally in the IUGR progeny. Fetal NPY mRNA levels were comparable to the 21-d-old concentrations, whereas the CRF mRNA levels increased prematurely at the 4-d postnatal age. The NPY and CRF peptide levels increased at 4 d instead of the normal 21 or 14 d, respectively. These intergroup differences in brain NPY and CRF mRNA and peptide levels disappeared at 14 d, only to demonstrate an increase in the NPY mRNA and peptide levels alone at the time of weaning from a high fat milk diet(21 d).

Immunolocalization studies demonstrated that the IUGR-induced changes in NPY and CRF immunoreactivity were restricted predominantly to the hypothalamus and surrounding regions. CRF immunoreactivity was noted mainly in neuronal cell bodies with some staining in the cellular processes. NPY immunoreactivity was in the cortical neuronal soma and processes to the extent that it allowed tracing of neuronal networks. In the hypothalamic region, although CRF-positive neurons were sparsely distributed, the NPY immunoreactive neurons were easier to discern along with the neuronal networks. The NPY immunoreactivity in the hypothalamic paraventricular and parvicellular regions appeared limited to cellular processes rather than to the cells themselves. In the adult the paraventricular region of the hypothalamus is responsible for the hyperphagic action of NPY(7,14,38). Therefore, it is intriguing that the intergroup differences in the NPY immunoreactivity between the IUGR and SHAM groups were noted mainly in the hypothalamic paraventricular region and not the other anatomical regions.

The increase in brain CRF levels at postnatal d 4 in the IUGR animals was associated with a trend toward a decline in circulating corticosterone levels at this age. However, when no difference in brain CRF levels was noted on postnatal d 14 and 21, a trend toward an increase in circulating corticosterone concentrations was observed in the IUGR progeny when compared with the age-matched SHAM progeny. An increase in circulating corticosterone concentrations can theoretically inhibit central CRF levels through a feedback mechanism, thereby suppressing an expected increase in CRF synthesis and concentrations(10). Given that corticosterone levels were not statistically different, this CRF increase at only d 4 may reflect the stress hyporesponsive period that begins at d 5 and lasts through d 10(37), along with the marked developmental increase in brain CRF that occurs at d 14 and 21 in the normal postnatal rat. Thus, uteroplacental insufficiency, although not eliciting a stress-response in utero, may be capable of doing so only on d 4, with d 5 through 21 being resistant for separate reasons as stated above.

The in utero nutritively restrictive environment although causing an increase in NPY mRNA levels did not increase the corresponding peptide levels until the 4-d postnatal stage. At this stage, however, a simultaneous increase in both the NPY and CRF peptide levels did not appear to alter the postnatal body weight gain pattern of the IUGR progeny that was set in utero when compared with the age-matched SHAM group. The biologic effect of an increase in the NPY and CRF peptides would depend upon the peptide with the more dominant change. At 4 d, CRF peptide levels doubled, whereas NPY peptide levels increased only by 40%, suggesting that the CRF action could dominate. In keeping with these changes, the IUGR progeny at this stage remained growth-retarded when compared with the SHAM group. In contrast at 21 d, NPY peptide levels increased, whereas CRF peptide levels remained unchanged. This increase in the hypothalamic paraventricular NPY concentrations should lead toward the development of hyperphagia and obesity. However, the IUGR animals remained smaller than the SHAM group.

One can argue that the postnatal (4 and 21 d) elevation in hypothalamic NPY peptide levels with the diminution in body weight is reminiscent of the adult starved state, where hypoglycemia with hypoinsulinemia was associated with an increase in brain NPY concentrations(15,38). In the present study, although the presence of a nutritively restrictive environment may form the stimulus for brain NPY synthesis, there is no change in postnatal circulating glucose or insulin concentrations. Besides, the IUGR progeny had equal access to the mother as did the SHAM group. Previously in this rat model of uteroplacental insufficiency no differences in food intake between the IUGR and SHAM progeny at 21-d postnatal age was observed(23). In keeping with this previous study, our present investigation did not detect any change in the rate of body weight gain in the IUGR versus the SHAM groups. The IUGR progeny therefore continued to remain smaller than the age-matched controls, reflecting the in utero restrictive environment rather than postnatal starvation. In support of this concept, postnatal circulating leptin concentrations were noted to be no different between the IUGR and the SHAM progeny. This observation is of significance because, in the adult, circulating leptin concentrations increase in proportion to the fat mass. Thus with obesity an increase is noted and with starvation a decline(18). Leptin in turn is known to suppress hypothalamic NPY synthesis(19), thereby causing a diminution in food intake.

Thus, the elevation in the hypothalamic NPY synthesis and concentrations at the 21-d stage is not due to postnatal starvation because food intake(23), postnatal growth rate, and circulating leptin concentrations remained constant in both groups. Instead, this postnatal change in brain NPY concentrations reflects the immediate (4-d postnatal) and long-term (21 d) effects of an in utero fetal deprivation state, which manifests fully only during the suckling phase. The biologic effect of increased NPY concentrations at 21 d may manifest at a later stage of development as well. Thus, the IUGR-induced increase in the hypothalamic NPY levels, with no change in CRF concentrations, may serve as the initiator of hyperphagia during the postsuckling phase. This change may herald the onset of IUGR-associated adult-onset obesity and insulin resistance as previously reported(39,40).

In summary, uteroplacental insufficiency leading to fetal growth restriction resulted in a premature increase in brain NPY and CRF synthesis and concentrations. During the early postnatal phase of development (4 d), both NPY and CRF peptide concentrations increased; the net balance of these perturbations did not correct the postnatal growth diminution. At the stage of weaning, hypothalamic NPY peptide levels increased with no change in CRF concentrations, perhaps tilting the balance toward the development of hyperphagia in the postsuckling phase.