Leptin Alters Adrenal Responsiveness by Decreasing Expression of ACTH-R, StAR, and P450c21 in Hypoxemic Fetal Sheep

Abstract

The late gestation increase in adrenal responsiveness to adrenocorticotropin (ACTH) is dependent upon the upregulation of the ACTH receptor (ACTH-R), steroidogenic acute regulatory protein (StAR), and steroidogenic enzymes in the fetal adrenal. Long-term hypoxia decreases the expression of these and adrenal responsiveness to ACTH in vivo. Leptin, an adipocyte-derived hormone which attenuates the peripartum increase in fetal plasma cortisol is elevated in hypoxic fetuses. Therefore, we hypothesized that increases in plasma leptin will inhibit the expression of the ACTH-R, StAR, and steroidogenic enzymes and attenuate adrenal responsiveness in hypoxic fetuses. Spontaneously hypoxemic fetal sheep (132 days of gestation, PO₂ ∼15 mm Hg) were infused with recombinant human leptin (n = 8) or saline (n = 7) for 96 hours. An ACTH challenge was performed at 72 hours of infusion to assess adrenal responsiveness. Plasma cortisol and ACTH were measured daily and adrenals were collected after 96 hours infusion for messenger RNA (mRNA) and protein expression measurement. Plasma cortisol concentrations were lower in leptin- compared with saline-infused fetuses (14.8 ± 3.2 vs 42.3 ± 9.6 ng/mL, P < .05), as was the cortisol:ACTH ratio (0.9 ± 0.074 vs 46 ± 1.49, P < .05). Increases in cortisol concentrations were blunted in the leptin-treated group after ACTH_1-24 challenge (F = 12.2, P < .0001). Adrenal ACTH-R, StAR, and P450c21 expression levels were reduced in leptin-treated fetuses (P < .05), whereas the expression of Ob-Ra and Ob-Rb leptin receptor isoforms remained unchanged. Our results indicate that leptin blunts adrenal responsiveness in the late gestation hypoxemic fetus, and this effect appears mediated by decreased adrenal ACTH-R, StAR, and P450c21 expression.

Introduction

In fetal sheep, the prepartum increase in circulating cortisol is required for the differentiation and maturation of key fetal organs such as the fetal lung, liver, kidney, and brain and for the normal timing of parturition and the successful transition to extrauterine life.¹ This increase is dependent upon an increase in fetal adrenal responsiveness to adrenocorticotropin (ACTH). Many factors are presumably involved in regulating this increase in responsiveness; leptin may be one of these factors.

Leptin is well known to influence food intake and energy expenditure,² and increasing evidence suggests that leptin also influences hypothalamic–pituitary–adrenal (HPA) axis activity.³ For instance, chronic administration of leptin to ob/ob mice decreases plasma corticosterone levels.⁴ In addition, leptin inhibits the release of corticotropin releasing hormone (CRH) from the hypothalamus in vitro and blunts plasma ACTH and corticosterone responses to restraint the in vivo stress in adult rodents.⁵

Gestation-related changes in plasma leptin and effects on the HPA axis have also been reported. Specifically, plasma leptin concentrations have been found to increase in the late gestation fetus⁶ along with the expression of leptin in perirenal adipocytes from fetal sheep.⁷ Furthermore, it has been shown that the late gestation increases in plasma ACTH and cortisol in the fetal sheep are suppressed by intracerebroventricular or intravenous infusions of leptin.^6,8 The mechanisms underlying the suppressive effects of leptin have not been established.

Among the elements important for the prepartum increase in fetal adrenal steroid production are increases in the expression of the ACTH receptor (ACTH-R) and steroidogenic acute regulatory protein (StAR). The ACTH-R is a 7-transmembrane-spanning G-protein-coupled receptor located in the zona fasciculata of the adrenal gland, whereas StAR transports the steroid precursor cholesterol from the outer to the inner mitochondrial membrane for further processing. We and others have reported that responsiveness of the fetal sheep adrenal gland to ACTH and the expression of adrenal ACTH-R, StAR, P450scc (cytochrome p450 side-chain cleavage), and P450c21 (steroid 21-hydroxylase) increase in late pregnancy, resulting in increased glucocorticoid production. We have also shown that if the peripartum increases in ACTH-R and StAR expression are blocked, adrenal responsiveness is inhibited.^9–11

Evidence of regulation of fetal endocrine responses by hypoxemia has been obtained in an ovine mild long-term hypoxemia model (MLTH) among others. Mild long-term hypoxemia model fetuses display relatively normal¹² or slightly elevated¹³ basal ACTH and normal basal cortisol levels but have a blunted cortisol response to an ACTH challenge¹²; this is likely the result of the lower expression of ACTH-R, P450scc, and P450c17 (17α-hydroxylase/17,20-lyase) steroidogenic enzymes observed in MLTH fetuses.¹⁴ Chronic hypoxemia also alters the leptin system with a reported increase in plasma leptin and the adrenal leptin receptor Ob-Rb messenger RNA (mRNA) in MLTH.¹⁵

Given the upregulation of adrenal leptin receptors by hypoxemia and the importance of ACTH-R, StAR, and steroidogenic enzymes in mediating adrenal responses, we hypothesized that increased leptin levels would reduce ACTH-R, StAR, P450scc, and P450c21 expression and suppress adrenal responses to ACTH in the late gestation mildly hypoxemic fetal sheep. To test this hypothesis, we infused late gestation, mildly hypoxemic fetal sheep with leptin and examined the effects of leptin infusion on (1) adrenal responsiveness by giving an ACTH challenge; (2) basal levels of plasma cortisol and ACTH concentrations; and (3) ACTH-R, StAR, P450scc, and P450c21 expression. We also measured leptin receptor Ob-Ra and Ob-Rb isoform expression in the fetal adrenal to determine whether leptin downregulates its receptor. Our results suggest that the mechanism by which leptin impairs fetal adrenal responsiveness to ACTH is related to a suppression of the ACTH-R, StAR, and P450c21 expression.

Materials and Methods

Animals and Surgery

A total of 15 pregnant ewes (obtained from local breeders) were used in this study. Sheep were housed in straw-lined pens with ad libitum access to food and water. Ewes were fasted 24 hours prior to surgery which was performed under general anesthesia. Aseptic techniques were used to insert fetal femoral arterial and venous and amniotic catheters. Bolus gentamicin (80 mg; Abbott Laboratories, North Chicago, Illinois) and ampicillin (500 mg; American Pharmaceutical Partners, Schaumburg, Illinois) in isotonic saline were administered to the ewe for 2 days after surgery. All procedures used in this study were approved by the Animal Care and Use Committee at Wake Forest University School of Medicine.

Leptin Infusion

Approximately 5 days after surgery (at 132 days of gestational age), fetuses were intravenously infused with recombinant human leptin (R&D Systems, Abingdon, UK) in saline at a dose intended to deliver approximately 20 µg/kg per hour over 96 hours. This dose was selected based on results from a previous study showing that 20 µg/h leptin affects ovarian steroidogenesis in sheep.¹⁶ The use of human leptin is supported by evidence of its bioactivity in sheep. Human and ovine leptin are highly conserved and no differences have been reported in their biological activity.^17,18 Human leptin has been utilized in studies of ovarian function in sheep,^16,19 and we have recently shown that human leptin inhibits directly adrenocortical steroidogenesis in ovine adrenocortical cells from adult sheep.²⁰

Control fetuses were infused with saline only. On the first day leptin infusion started at 0900 hours, blood samples were collected before infusion started, at 1000, and 1400 hours. On each subsequent day, the blood samples were collected daily at 0900 hours for measurement of plasma ACTH, cortisol, and leptin concentrations. Blood gases and pH (ABL5 blood gas analyzer; Radiometer, Copenhagen, Denmark) were monitored daily during the experiment.

At 72 hours of infusion, after the daily blood sample, a bolus of ACTH was administered (ACTH_1-24, 0.5 mL, 100 ng/kg body weight; Cortrosyn, Organon Inc, West Orange, New Jersey) to assess adrenal responsiveness, and blood samples were taken at 15 minutes, 30 minutes, 1 hour, 6 hours, and 24 hours thereafter. The dose of 100 ng/kg has been demonstrated to maximally elevate cortisol concentrations in fetuses of a similar gestational age.²¹

After 96 hours (24 hours after ACTH bolus) of continuous leptin or saline infusion, ewes and fetuses were killed with an overdose of intravenous sodium pentobarbitone (85 mg/kg), and fetal adrenals were removed and snap frozen for later ACTH-R, StAR, and leptin receptor mRNA and protein analysis.

Adrenocorticotropin and Cortisol Radioimmunoassay

Fetal plasma immunoreactive ACTH concentrations were measured using a radioimmunoassay kit (MP Biomedical LLC, Orangeburg, New York) according to the manufacturer’s instructions. The sensitivity of measurement was 5.7 pg/mL. The intra- and inter-assay coefficients of variation were 7.1% and 9.5%, respectively. According to the manufacturer, the cross-reactivity of the rabbit anti-human ACTH_1-39 is 100% with ACTH_1-24, <1% with β-endorphin, α-melanocyte-stimulating hormone, α-lipotrophin, and β-lipotrophin. The sensitivity of the assay was 9 pg/mL. The intra-assay coefficient of variation was < 7.4%.

Plasma cortisol concentrations were measured using the ImmuChen ¹²⁵I Cortisol Radioimmunoassay Kit (MP Biomedicals) according to the manufacturer’s instructions. The sensitivity of measurement was 0.7 ng/mL. The intra- and inter-assay coefficients of variation were 6.8% and 9.5%, respectively.

Leptin Enzyme-Linked Immunosorbent Assay

Plasma leptin concentrations were measured with a commercial enzyme-linked immunosorbent assay ([ELISA] kit; Multispecies Leptin Assay, Linco Research, St Charles, Missouri) using recombinant ovine leptin standards. Calibration curves made using recombinant ovine leptin standards and the recombinant human leptin standards (provided with the kit) were similar. This kit has previously been reported to detect ovine leptin,¹⁵ and no differences have been observed in biological activity between ovine and human leptin.^22,23 The sensitivity of measurement was 1 ng/mL of recombinant ovine standard. The intra- and interassay coefficients of variation were 8.7% and 8.1%, respectively.

Quantitative Real-Time Reverse Transcriptase–Polymerase Chain Reaction

The relative abundance of ACTH-R, StAR, P450scc, P450c21, and Ob-Ra and Ob-Rb mRNA transcripts in fetal adrenal cortex were measured by quantitative real-time reverse transcriptase–polymerase chain reaction (RT-PCR) using TaqMan PCR. Primers of PCR were designed from the sheep sequences with Genbank accession numbers: AF116874 (MC2R gene for ACTH-R), AF290202 (STAR gene for StAR), S65754 (CYP11A1 gene for P450scc), M92837 (CYP21D gene for P450c21), AY278244 (Ob-Ra gene for leptin receptor short form), and U62124 (Ob-Rb gene for leptin receptor long form), respectively. Glyceraldehyde 3-phosphate dehydrogenase (Genbank accession number U94889) was used as control. Total RNA of 1 μg was reverse transcribed in a 20 μL reaction mixture using an ABI High Capacity complementary DNA (cDNA) Archive kit, according to the manufacturer’s instructions (Applied Biosystems, Foster City, California). The reaction contained 1× RT buffer, 100 μmol/L of each deoxynucleoside triphosphate, 1× random primer, and 100 U of reverse transcriptase. The reaction was carried out at 25°C for 10 minutes then at 37°C for 2 hours. Control reactions were those in which the RT enzyme or the target RNA was omitted from the reaction.

Taqman PCR was performed on the cDNA samples using an ABI PRISM 7500 Sequence Detection System (Applied Biosystems). For each gene tested (see Table 1), PCR was carried out in multiplex mode, with every 20 μL reaction containing 2 μL of cDNA reaction, 1× Taqman universal PCR master mix, 250 nmol/L of a gene-specific primer, 250 nmol/L of FAM (fluorescein amidite)-labeled fluorogenic Taqman probe, and 2.5 U of TaqMan enzymes. The thermal cycling conditions were as follows: 50°C for 2 minutes, 95°C for 10 minutes, 40 cycles at 95°C for 15 seconds, and 60°C for 1 minute. An increase in fluorescence was obtained at the annealing and extension step at 60°C.

Table 1.

Primers and Probes Used in Real-Time PCR Assays^a

Gene	Forward Primer	Reverse Primer	Probe Sequence
ACTH-R	GTATGAAAACATCAACAGTACAGCAAGAA	AAAACTCCGACAATGGATACTGTGA	FAM-CTGCTGTGATTTTGCC-MGB
StAR	GCCCCACCTGCATGGT	GAGTTTGGTCCTTGAGGGACTTC	FAM-CCGCCCCCTGGCTG-MGB
P450scc	GCAGAGATACCCTGAAAGTGACTT	GGCATAGATGGCCACTTGCA	FAM-TTCCTGCCAAGACACTGT-MGB
P450c21	CGGTGGCCTTCCTACTTCAC	TCTCGATCCAACTCCTCCTGAA	FAM-CACCCCGAGATTCAGT-MGB
Ob-Ra	CCCCGAGGAAAGTTCACCTATG	TGGTGGCACTCTTGCTCATT	FAM-ACGCAGTGTACTGCTGC-MGB
Ob-Rb	GGAGACAGCCCTCTGTTAAATATGC	TGAGCTGTTTATAAGCCCTTGCT	FAM-CCTCCTCGGCTTCACC-MGB
GAPDH	GCATCGTGGAGGGACTTATGAC	GGGCCATCCACAGTCTTCTG	FAM-ACGCCATCACTGCCACCT-MGB

Abbreviations: ACTH-R, adrenocorticotropic hormone receptor; PCR, polymerase chain reaction; StAR, steroidogenic acute regulatory protein; P450scc, cytochrome P450, family 11, subfamily A, polypeptide 1; P450c21, cytochrome P450 21-hydroxylase; Ob-Ra, leptin receptor short form; Ob-Rb, leptin receptor long form; GAPDH, glyceraldehyde 3-phosphate dehydrogenase.

^a Nucleotide sequences are 5’-3’.

The relative level of expression of each gene in the samples was determined using the relative 2^ΔΔCt expression method as previously described.²⁴ After the linear range of amplification (threshold cycle, Ct) was determined for the genes of interest, it was normalized against an endogenous GAPDH control and then against the untreated control sample, which served as the calibrator. The value of the relative level of expression for the gene of interest represents 2 independent reactions performed in triplicate.

Isolation of Adrenal Membrane and Mitochondrial Protein

To isolate cell membranes, adrenal cortex tissue was homogenized in cold 50 mmol/L Tris HCL (pH 7.5), 0.1 mmol/L EDTA, 0.5 mmol/L DTT, and protease inhibitor cocktail (1:200 diluted; Sigma, St Louis, Missouri). Homogenates were centrifuged for 8 minutes at 2000g at 4°C. The resultant supernatant was centrifuged for 20 minutes at 10 000g at 4°C. Finally the supernatant was centrifuged for a further 1 hour at 100 000g at 4°C to pellet the cellular membrane.

To isolate mitochondria, adrenal cortex tissue was first homogenized in cold sucrose buffer containing 50 mmol/L Tris HCL, 0.25 mol/L sucrose, 0.1 mmol/L EDTA (pH 7.4), and protease inhibitor cocktail (1:200 diluted: P8340, Sigma). Adrenal homogenates were centrifuged at 600g for 30 minutes at 4°C. The resultant supernatant was centrifuged at 9000g for 30 minutes at 4°C to pellet the mitochondria.

Adrenocorticotropin  Receptor, StAR, and Ob-R Western Blot Analysis

Protein samples (50 μg of either cellular membranes for ACTH-R and Ob-R or mitochondrial for StAR) were loaded and separated on 12% SDS-polyacrylamide gel. Proteins were transferred onto a polyvinylidene fluoride membrane and blocked with 5% nonfat dry milk in 0.05% Tween 20 in Tris-buffered saline for 60 minutes at room temperature before incubation with the rabbit polyclonal antibodies against ACTH-R, StAR, or Ob-R (Santa Cruz Biotechnology, Santa Cruz, California), immunoreactive proteins were visualized using the enhanced chemiluminescence method as described by the manufacturer and exposed to Amersham Hyperfilm ECL.

Statistical Analysis

All numerical data are presented as mean ± standard error of the mean. Differences between means were evaluated using the unpaired Student t test or analysis of variance (ANOVA) with Newman-Keuls multiple test, as appropriate. Significance was set at the .05 level.

Results

Fetal body weights at postmortem were 3956 ± 30 g (n = 8) in the leptin-infused fetuses and 3816 ± 40 g (n = 7) in the control fetuses. There was no difference in total adrenal weights between leptin-infused (0.46 ± 0.04 g) and control (0.49 ± 0.04 g) fetuses. Fetal blood gases were similar between the groups at each time point. Before infusion, the pH, PCO₂, and PO₂ values were, respectively, 7.37 ± 0.03, 44.2 ± 1.4 mm Hg, and 16.4 ± 1.1 mm Hg in leptin-infused fetuses and 7.32 ± 0.04, 46.6 ± 3.0 mm Hg, and 14.1 ± 1.9 mm Hg in control fetuses. After 96 hours of infusion, the pH, PCO₂, and PO₂ values were, respectively, 7.37 ± 0.02, 48 ± 0.5 mm Hg, and 15.4 ± 1.07 mm Hg in leptin-infused fetuses and 7.4 ± 0.02, 47.9 ± 0.7 mm Hg, and 14.0 ± 1.1 mm Hg in control fetuses.

Plasma Leptin Concentrations

There were significant effects of treatment (F = 159.1 P < .0001), time (F = 11.11, P < .0001), and interaction (F = 11.68, P < .0001) for plasma leptin concentrations (Figure 1). Plasma leptin concentrations were significantly higher than preinfusion concentrations by the second day of leptin infusion (P < .05) vs control fetuses and remained significantly elevated for the duration of the experiment.

Figure 1.

Plasma Leptin levels in fetuses receiving an infusion of vehicle (^, n = 7) or Leptin (•, n = 8) for 96 hours. Asterisks denote significant differences (P < .05) between saline- and leptin-infused groups.

Plasma ACTH and Cortisol Concentrations

For plasma ACTH concentrations, there were no within- or between-group differences during the 96 hours of infusion period (Figure 2A). In contrast, for plasma cortisol concentrations, there were effects of time (F = 5.21, P < .001) and a time and treatment interaction (F = 4.56, P < .003) between the groups (Figure 2B). The ratio of plasma cortisol:ACTH concentration was significantly lower in leptin-infused group compared with saline-infused group; there were effects of treatment (F = 11.1, P < .001) and time (F = 2.44, P = .05; Figure 2C).

Figure 2.

Plasma ACTH (A), Cortisol (B), and ratio of Cortisol:ACTH (C) in fetuses receiving an infusion of vehicle (^, n = 7) or leptin (•, n = 8) for 96 hours. Asterisks denote significant differences (P < .05) between saline- and leptin-infused groups. ACTH indicates adrenocorticotropin.

Adrenal Response to ACTH_1-24

There was a tendency for plasma ACTH concentrations to increase in leptin-infused or control fetuses after the ACTH_1-24 challenge, but the change was not significant (F = 1.9, P = .1; Figure 3A). Cortisol concentrations increased significantly (F = 12.2, P < .0001). This increase was greater in the control group than in the leptin infusion group (F = 12.4, P = .0008; Figure 3B). The ratio of cortisol:ACTH concentration was significantly higher (F = 14.9, P = .0003) in saline-infused group at 30 minutes after ACTH challenge compared with leptin-infused group (Figure 3C).

Figure 3.

Plasma ACTH (A), Cortisol (B), and ratio of Cortisol:ACTH (C) before and after ACTH_1-24 challenge in fetuses receiving an infusion of vehicle (^, n = 7) or leptin (•, n = 8) for 96 hours. Asterisks denote significant differences (P < .05) between saline- and leptin-infused groups.

Leptin Receptor Expression

There were no statistically significant differences in leptin receptor short (Ob-Ra) or long (Ob-Rb) isoforms in both mRNA (Figure 4 A, B) and protein levels (Figure 4 C, D) in adrenal glands of control and leptin-infused animals.

Figure 4.

Effect of leptin infusion on leptin receptor short (Ob-Ra) or long (Ob-Rb) isoforms in both mRNA (A, B) and protein levels (C, D) in adrenal glands of saline- and leptin-infused animals (P > .05). ACTH indicates adrenocorticotropin; mRNA, messenger RNA.

P450scc and P450c21 mRNA Expression

P450c21 mRNA expression levels were significantly lower in leptin-infused fetuses compared with control fetuses (P < .05; Figure 5) while P450scc tended to be reduced (P > .05).

Figure 5.

Effect of leptin infusion on P450c21 (A) and P450scc (B) mRNA expression. P450c21 mRNA was significantly reduced in the Leptin-infused fetuses (▪, n = 7) compared with saline-infused (□, n = 8). Asterisks denote significant differences (P < .05) between saline- and leptin-infused groups. ACTH indicates adrenocorticotropin; mRNA, messenger RNA.

ACTH-R and StAR Protein Expression

Adrenal cortical ACTH-R (Figure 6) and StAR (Figure 7) mRNA and protein expression levels were significantly lower in leptin-infused fetuses compared with control fetuses (P < .05).

Figure 6.

Effect of Leptin infusion on ACTH-R mRNA (A) and protein (B) expression. ACTH-R mRNA and protein were significantly reduced in the Leptin-infused fetuses (▪, n = 7) compared with saline-infused (□, n = 8). Asterisks denote significant differences (P < .05) between saline- and leptin-infused groups. ACTH indicates adrenocorticotropin; mRNA, messenger RNA.

Figure 7.

Effect of leptin infusion on StAR mRNA (A) and protein (B) expression. StAR mRNA and protein were significantly reduced in the leptin-infused fetuses (▪, n = 7) compared with saline-infused (□, n = 8). Asterisks denote significant differences (P < .05) between saline- and leptin-infused groups. StAR indicates steroidogenic acute regulatory protein; mRNA, messenger RNA.

Discussion

This study was designed to determine whether increasing plasma leptin levels in the spontaneously hypoxemic fetus attenuates adrenal responsiveness to ACTH and the mechanisms potentially responsible for any attenuation.

We found that leptin, infused into the hypoxemic late gestation sheep fetus blunts adrenal cortisol responsiveness to ACTH and inhibits adrenal expression of ACTH-R, StAR, and P450c21. These results suggest that the leptin-related inhibition of the fetal cortisol response to ACTH is a consequence of decreased ACTH-R, StAR, and steroidogenic enzyme P450c21 expression in the fetal adrenal gland.

With an average PO₂ of 15 mm Hg, animals in both of our experimental groups were spontaneously hypoxemic. The effects of hypoxemia on the HPA axis, and more specifically on adrenocortical responses, have been shown dependent on the duration and intensity of the insult; fetal hypoxemia does not change^25,12 or activates the HPA axis.^26–33 In our study, the levels of ACTH and cortisol prior to any manipulation were not different between groups and were similar to the levels we and others have reported previously in normoxic fetuses.^34,13 Thus, the mild, spontaneous hypoxemia in our animals was insufficient to induce activation of baseline cortisol secretion. However, the response to ACTH in the mildly hypoxic, saline-infused animals in our present study is substantially less than the responses to ACTH in similar aged normoxic fetuses we reported previously³⁴ (23 ± 3 ng/mL present study vs 45 ± 6 ng/mL earlier study; P < .01), consistent with the observations in MLTH fetuses.¹² It has been suggested that this loss of responsiveness is related to an increase in fetal plasma leptin and leptin receptors in the adrenal.¹⁵ The spontaneous hypoxemia in our animals induces a moderate increase in plasma leptin (∼1.5 ng/mL) compared to values observed in normoxic animals¹⁵ which agree with the above suggestion.

The effects of MLTH on adrenal responsiveness appear to involve alterations in adrenocortical steroidogenic capacity including reduced expression of the ACTH receptor and the steroidogenic enzymes P450c17 and P450scc with no detectable changes in P450c21 or StAR expression.¹⁴ In our study, leptin infusion downregulated the expression of ACTH-R. In addition, StAR mRNA and protein and also P450c21 mRNA were reduced which suggests that supraphysiological plasma leptin concentrations affect some different points in the adrenal glucocorticoid biosynthetic pathway than those influenced by the smaller increases produced by hypoxemia.¹⁴ Overall these findings suggest a diminished responsiveness to ACTH (decreased ACTH receptor expression) coupled with a decreased capacity of the hypoxemic leptin-treated fetal adrenal to synthesize cortisol (lower StAR and P450c21). Reduced expression of ACTH-R and StAR protein would induce a reduction in the cholesterol levels available to the mitochondrial P450scc enzyme, whereas reduced expression of P450c21 would interfere with the steps from 17α-hydroxyprogesterone to 11-deoxycortisol and progesterone to 11-deoxycorticosterone essential for cortisol and aldosterone production. Acting in concert, these changes could easily attenuate cortisol responses to ACTH.

The other result of the leptin induced changes in the adrenal was an inhibition in the peripartum increase in fetal plasma cortisol. This has been reported previously in normoxic fetuses when the plasma leptin levels were raised to concentrations similar to what our infusions achieved⁶; however, it contrasts with an earlier study in which lower doses of leptin did not alter fetal plasma cortisol levels.³⁵ Our protocol of infusion produced steady-state plasma leptin concentrations higher than those reported by Forhead et al.³⁶ These levels of plasma leptin may be necessary to suppress basal cortisol concentrations.

It appears that leptin can modulate HPA axis activity at both central and peripheral levels. Leptin receptors have been identified in the hypothalamus^36–39 and leptin has also been implicated in the modulation of neurotransmitter levels. Diabetic rats treated with leptin display normalization in serum corticosterone and a concomitant reduction in norepinephrine (NE) levels in the PVN (paraventricular nucleus).⁴⁰ The stimulation of CRH transcription by direct administration of NE into the PVN⁴¹ and the dramatic suppression of the stress response in animals with lesions of the noradrenergic innervations to the hypothalamus⁴² together with the reduction in NE levels caused by leptin treatment suggest a NE-mediated central role for leptin in the HPA axis modulation. The ACTH levels in turn are known to regulate adrenal steroidogenesis. Specifically, previous in vitro and in vivo studies showed that ACTH positively regulates adrenal responsiveness, upregulating ACTH-R and StAR mRNA expression in the fetal sheep adrenal.^9,43–48

In the present study, we did not detect a significant increase in immunoreactive ACTH 15 minutes after a bolus injection. Since the antibody used recognized the ACTH amino acid sequence 5 to 18 on the human ACTH molecule, it is plausible that binding to additional ACTH peptides makes it difficult to discriminate the increase in ACTH after ACTH_1-24 injection. Perhaps a more likely explanation relates to the half-life of the peptide. The half-life reported for injected ACTH in fetal sheep is around 1 minute,⁴⁹ probably making any significant increase in ACTH difficult to detect in our first sample taken about 15 half-lives after the injection.

Leptin exerts its function by binding to the leptin receptor (Ob-R) and several splice variants of Ob-R (Ob-Ra through Ob-Rf) have been detected in mouse, human, and rat; these isoforms differ in the length of their cytoplasmic domain with the long Ob-Rb isoform displaying full signaling capacity.⁵⁰ Downregulation of the leptin receptor has been observed after incubation of rat adrenal slices with leptin.⁵¹ Considering that our protocol of infusion produced plasma concentrations of leptin well above the normal physiological range, we examined the effects of leptin infusion on the expression of the short (Ob-Ra ∼90-100 kDa) and long (Ob-Rb ∼120-130 kDa) isoforms of the leptin receptor. Our results indicating no differences in leptin receptor mRNA and protein expression between control and leptin-treated animals suggested that the leptin transduction pathway, at least at the level of receptor expression, is not altered after 96 hours of leptin infusion. Thus, the effect of leptin does not represent withdrawal of a leptin-induced function in the adrenal because of receptor downregulation. The mechanisms by which leptin regulates its receptor in adrenal glands are unclear and remain to be determined.

Evidence of the mechanisms by which leptin modifies gene expression obtained in the human adrenocortical NCI-H295 tumor cell line indicates that leptin inhibits P450scc expression through activation of a cyclic adenosine monophosphate (cAMP)-degrading pathway that down-regulates CYP11A1 promoter activity,⁵² how these mechanisms are operating in vivo in the whole animal and the role of the leptin receptor in the ovine adrenal⁵⁴ warrant further investigation.

In terms of enzymatic activity effects modulating adrenal steroidogenesis, the increased expression of endothelial nitric oxide synthase in adrenal tissue from MLTH fetuses and its colocalization with P450c17 suggest a role for nitric oxide (NO) in the regulation of adrenal steroidogenesis.⁵⁵ This was recently confirmed with the reported NO-dependent inhibition of ACTH-induced cortisol production in vitro in adrenocortical cells from MLTH fetuses.⁵⁶ It has been recognized that diatomic gases such as CO and NO competitively interact with the oxygen-binding site of heme-containing cytochrome enzymes, including the steroidogenic rate-limiting enzymes P450scc,⁵⁷ aldosterone synthase P450 (P450aldo),⁵⁸ and P450c17.⁵⁹ Due to the multistep nature of the conversion of cholesterol to pregnenolone by P450scc (involving three distinct chemical reactions), and the 2 activities of P450c17 (hydroxylase and lyase),⁶⁰ these chemical steps in steroidogenesis may be greatly affected by NO-heme interaction, as previously proposed.⁶¹ It is conceivable that leptin acts via NO to diminish adrenal responses in both of our experimental groups. Further experiments are needed to understand the interplay between hypoxemia-induced NO and leptin on the regulation of fetal endocrine responses.

Perspectives and Significance

Our results suggest a role for leptin in modulating adrenal function in the hypoxic fetus. We demonstrated that an increase in circulating leptin concentration in the late gestation hypoxemic sheep fetus reduces adrenal responsiveness to ACTH, and these effects most probably result from direct downregulation of adrenal ACTH-R, StAR, and P450c21 expression. How leptin mediates its effects on gene expression and on enzymatic activities of the adrenal steroidogenic pathways remains to be investigated. Leptin is becoming an important factor in the intrauterine hormonal environment able to influence the development of obesity and its comorbidities,⁶² and the study of the interactions of leptin and the fetal HPA axis will clarify the possible role/roles of leptin in determining early life events that may have long-term metabolic consequences.