Fetal glucocorticoid synthesis is required for development of fetal adrenal medulla and hypothalamus feedback suppression

Abstract

During pregnancy, fetal glucocorticoid is derived from both maternal supply and fetal secretion. We have created mice with a disruption of the Cyp11a1 gene resulting in loss of fetal steroid secretion but preserving the maternal supply. Cyp11a1null embryos have appreciable although lower amounts of circulating corticosterone, the major mouse glucocorticoid, suggesting that transplacental corticosterone is a major source of corticosterone in fetal circulation. These embryos thus provide a means to examine the effect of fetal glucocorticoids. The adrenal in Cyp11a1 null embryos was disorganized with abnormal mitochondria and oil accumulation. The adrenal medullary cells did not express phenylethanolamine N-methyltransferase and synthesized no epinephrine. Cyp11a1 null embryos had decreased diencephalon Hsd11b1, increased diencephalon Crh, and increased pituitary Pomc expression, leading to higher adrenocorticotropin level in the plasma. These data indicate blunted feedback suppression despite reasonable amounts of circulating corticosterone. Thus, the corticosterone synthesized in situ by the fetus is required for negative feedback suppression of the hypothalamus-pituitary-adrenal axis and for catecholamine synthesis in adrenal medulla.

Fig. 1.

Defective hormone secretion of CYP11A1 null embryos. A, Diagrams of the Cyp11a1 alleles showing the insertion of the neo gene or of the loxP site in the first exon in E1-Neo and E1-loxP, respectively. B, Western blot analysis of E18.5 adrenal CYP11A1 from the wild-type (WT) (+/+), heterozygous (+/−), or homozygous (−/−) Cyp11a1 mutant mice. C, Plasma levels of pregnenolone (P5) at E18.5 measured by RIA are shown as mean ± sem [*, P = 0.002 vs. control littermates; WT, n = 5; knockout (KO), n = 4]. D, Tissue levels of corticosterone (CORT.) in E18.5 adrenal measure by RIA are shown as mean ± sem (*, P = 0.03 vs. control littermates; WT, n = 4; KO, n = 5). E, Plasma levels of corticosterone at E18.5 (WT, n = 15; KO, n = 10) and E19 (WT, n = 11; KO, n = 10) measured by RIA are shown as mean ± sem (*, P = 0.001; #, P = 0.02). F, Transmission electron micrographs of mitochondria in adrenocortical cells at E17.5.

Fig. 2.

Cyp11a1 null adrenal cells are disorganized and accumulate oil. A–F, H&E staining was performed on adrenal sections. G–J, Immunohistochemistry for PCNA was performed on adrenal sections. K–N, Oil Red O staining was performed on adrenal sections. Scale bars, 50 μm. WT, Wild type; KO, knockout.

Fig. 3.

Cyp11a1 null embryos fail to express medullary PNMT and synthesize epinephrine. Immunohistochemistry for (A–D) TH (brown staining) and (E–H) PNMT (brown staining) on adrenal sections. I, Amount of epinephrine (Epi.) and norepinephrine (Norepi.) in E18.5 adrenal measured by HPLC are shown as mean ± sd [wild type (WT), n = 5; knockout (KO), n = 6]. J and K, Chromaffin reaction was performed on E18.5 adrenals. N.D., Undetected. Scale bars, 50 μm.

Fig. 4.

Unsuppressed HPA axis in Cyp11a1 null embryos. A, Plasma levels of ACTH at E18.5 measured by RIA are shown as mean ± sem. B, Relative RNA levels of Pomc and Crh at E18.5 measured by real-time RT-PCR are shown as mean ± sem [pituitary Pomc: *, P < 0.01 vs. control littermates; wild type (WT), n = 4; knockout (KO), n = 4; diencephalon Crh: *, P = 0.04 vs. control littermates; WT, n = 6; KO, n = 7; and diencephalon Pomc: **, P < 0.01 vs. control littermates; WT, n = 6; KO, n = 7]. C, RNA levels of genes related to HPA axis activity at E18.5 measured by real-time RT-PCR are shown as mean ± sem (diencephalon: WT, n = 6; KO, n = 6; pituitary: WT, n = 4; KO, n = 4). D, Relative RNA levels of Hsd11b1 in E18.5 diencephalon measured by real-time RT-PCR are shown as mean ± sem (*, P < 0.001 vs. control littermates; WT, n = 5; KO, n = 8). E, Tissue levels of corticosterone (CORT.) in E18.5 diencephalon measured by RIA are shown as mean ± sem (*, P = 0.01 vs. control littermates; WT, n = 4; KO, n = 5).