We have described the expression of specific iodothyronine deiodinase mRNAs (using quantitative RT-PCR) and activities in normal human placentas throughout gestation and compared our findings to those in placentas from pregnancies affected by intrauterine growth restriction (IUGR). The predominant deiodinase expressed in placenta was type III (D3); type II (D2) was also present. In general terms, the activities of the enzymes D2 and D3 (and mRNAs encoding these enzymes) were higher earlier in gestation (<28 wk) than at term and displayed an inverse relationship with the duration of gestation (P < 0.05). Comparison of the relative expressions of mRNAs encoding D2 and D3 as well as their activities in placentas associated with IUGR (early and late gestational groups) with findings from normal placentas of similar gestational ages revealed no significant differences. Immunolocalization of D2 and D3 in syncytiotrophoblast (including syncytial sprouts) and cytotrophoblast of human placentas was demonstrated at both early and late gestation. Treatment of primary cultures of term cytotrophoblast cells in vitro with increasing doses of T(3) (1, 10, and 100 nM) resulted in increased expression of mRNAs encoding both D2 and D3 at 100-nM concentrations (P < 0.01) compared with control. Experiments with JEG-3 choriocarcinoma cells demonstrated a similar effect on D3 mRNA at 10 and 100 nM T(3) (P < 0.01). The demonstrated changes in iodothyronine deiodinase expression in the placenta across pregnancy are likely to contribute to regulation of the thyroid hormone supply to the developing fetus. The lack of difference in deiodinase expression in normal placentas and those found in IUGR argues against placental deiodinases being responsible for the hypothyroxemia in circulating fetal thyroid hormones observed in this condition.

Prostate cancer is a major cause of male cancer death. In vitro and in vivo data support a role for 1a,25 Dihydroxyvitamin D 3 (1a,25(OH) 2 D 3 ) in regulating the growth and di erentiation of the normal prostate gland yet prostate cancer cells appear signi®cantly less sensitive to this action. Vitamin D 3 receptor (VDR) content or mutational status do not correlate clearly with the antiproliferative e ects of 1a,25(OH) 2 D 3 and therefore it is unclear why prostate cancer cell lines are signi®cantly less sensitive to this action. We hypothesized that the antiproliferative responses of prostate cancer cells to 1a,25(OH) 2 D 3 are suppressed by a process involving histone deacetylation. Sodium butyrate (NaB) and trichostatin A (TSA) are inhibitors of histone deacetylase (HDAC) activity. Low doses of NaB or TSA (300 mM and 15 nM respectively), which alone were relatively inactive, synergized with 1a,25(OH) 2 D 3 in liquid and semi-solid agar to inhibit the growth of LNCaP, PC-3 and DU-145 prostate cancer cells. Still greater synergy was observed between vitamin D 3 hexa¯uoride analogs and either NaB or TSA. The mechanism appeared to involve neither the cyclindependent kinase inhibitor, p21 (waf1/cip1) nor cell cycle arrest, but rather induction of apoptosis. These data suggest that cells dysregulate the normal pro-apoptotic signals of 1a,25(OH) 2 D 3 during prostate cancer development by a mechanism involving histone deacetylation. Combination therapy with potent vitamin D 3 analogs and clinically approved HDAC inhibitors may overcome this lesion and improve the treatment of both androgendependent and independent prostate cancer. Oncogene (2001) 20, 1860 ± 1872.

11beta-Hydroxysteroid dehydrogenase type 2 (11beta-HSD2) inactivates cortisol to cortisone. In the placenta 11beta-HSD2 activity is thought to protect the fetus from the deleterious effects of maternal glucocorticoids. Patients with apparent mineralocorticoid excess owing to mutations in the 11beta-HSD2 gene invariably have reduced birth weight, and we have recently shown reduced placental 11beta-HSD2 activity in pregnancies complicated by intrauterine growth restriction. This is reflected in the literature by evidence of hypercortisolemia in the fetal circulation of small babies. In this study we have determined the levels of placental 11beta-HSD2 mRNA expression across normal gestation (n = 86 placentae) and in pregnancies complicated by intrauterine growth restriction (n = 19) and evaluated the underlying mechanism for any aberrant 11beta-HSD2 mRNA expression in intrauterine growth restriction. 11beta-HSD2 mRNA expression increased more than 50-fold across gestation, peaking at term. Placental 11beta-HSD2 mRNA levels were significantly decreased in intrauterine growth restriction pregnancies when compared with gestationally matched, appropriately grown placentae [e.g. at term DeltaCt (11beta-hydroxysteroid dehydrogenase type 2/18S) 12.8 +/- 0.8 (mean +/- SE) vs. 10.2 +/- 0.2, respectively, P < 0.001]. These differences were not attributable to changes in trophoblast mass in intrauterine growth restriction placentae, as assessed by parallel analyses of cytokeratin-8 mRNA expression. No mutations were found in the 11beta-HSD2 gene in the intrauterine growth restriction cohort, and imprinting analysis revealed that the 11beta-HSD2 gene was not imprinted. Although the underlying cause is unknown, 11beta-HSD2 gene expression is reduced in intrauterine growth restriction pregnancies. These data highlight the important role of 11beta-HSD2 in regulating fetal growth, a known factor in determining fetal morbidity but also the subsequent development of cardiovascular disease in adulthood.

Twenty-four castrated male lambs initially maintained on a photoperiod of 12 h light:12 h dark were allocated to a factorial experiment with two daylengths (8 h L: 16 h D or 16 h L: 8 h D) and two levels of feeding (restricted or ad libitum). Blood samples were taken every 4 h for 24 h during the introductory period and after 24, 51 and 79 days of treatment. There were highly significant positive effects of daylength and level of feeding on serum prolactin: mean concentrations increased from a mean of 38 plus or minus 1 ng/ml during the introductory period until at day 79 they were: 8L: 16D (restricted diet), 81 ng/ml; (food ad libitum), 167 ng/ml; 16L:8D (restricted diet), 262 ng/ml; (food ad libitum), 262 ng/ml (S.E. of treatment mean plus or minus 4). Long daylength and feeding ad libitum also significantly increased growth rate.

SUMMARY1. Blood samples were taken every 20 min (for at least 28 hr) from five castrate male and two anoestrus female ad libitum fed sheep. Analysis for plasma growth hormone (GH) showed that two of the males, on two occasions, had regular, although individually specific, patterns of GH secretion (peaks 35-5 hr). The other animals all had irregular patterns of GH release.2. Throughout the experiments, meal sizes and frequency were recorded and it was found that out of eighty spontaneous meals of at least 50 g, 57 (71 %) occurred in the hour after GH peaks, which accounted for 50 % of the total time. Furthermore, on twenty out of twenty-four occasions GH levels were found to be falling before the 'expected' feeding time when fresh food was offered and the animals normally consumed a large meal. The removal of the food from three of the males for 10 hr during an experiment prompted an increase both in the size and frequency of the GH peaks. After re-feeding, GH levels immediately fell and remained low for 1-2 hr. 3. We believe that these results show an association between GH secretion and meal feeding in sheep, and that GH secretion quickly responds to fasting. As GH levels fell before, or in the early stages of meals, this indicates a neural reflex in the inhibition of GH before a meal.