Main

Adrenarche is an event of postnatal sexual maturation in which there is an increase in the secretion of adrenal androgens, principally DHEA and DHEA sulfate, not accompanied by an increase in cortisol secretion. This event occurs only in higher primates, typically at about 6-8 y of age in humans(1–3). The mechanism of this phenomenon is not known, but it is independent of the gonadotropin-releasing hormone pulse generator activation at sexual maturation. The initial step of the steroidogenic pathway is the side chain cleavage of cholesterol, catalyzed by cytochrome P450 side chain cleavage to form pregnenolone(4), which might be converted to progesterone by 3βHSD type II, the isoform present in human adrenals and gonads. In the adrenal cortex, the subsequent 17α hydroxylation of pregnenolone or progesterone is a branch point for the formation of cortisol and adrenal androgens. 17α-hydroxylase and 17,20 lyase activities are catalyzed by microsomal cytochrome P450c17, a product of a single copy gene CYP17. The 17,20 lyase activity cleaves the c17,20 bond to convert the 21-carbon steroid 17OH pregnenolone to the corresponding 19-carbon steroid DHEA(5,6). Whether the 17,20 lyase activity of human P450c17 can convert 17OH-progesterone to androstenedione is not clearly established(7–9). In addition, secretion of cortisol and androgens is dissociated in many clinical situations(7–10). Although little is known about the mechanism of this dissociation, it has been suggested that the electron transfer system might be a clue, as the flux of reducing equivalents to P450c17 might be crucial to regulate 17,20 lyase activity(11–13). However, a decrease of 3βHSD type II mRNA with normal P450c17 mRNA and a normal ratio of 17,20 lyase to 17α-hydroxylase activities has been found in virilized adrenocortical carcinomas, suggesting little contribution of the electron transfer system to the modulation of 17,20 lyase activity(14). These authors speculated that the high androgen secretion observed in patients with adrenocortical carcinomas might be secondary to a marked reduction in 3βHSD activity. Therefore, we have hypothesized that a progressive decrease in 3βHSD activity as a function of age during the prepubertal years could be a mechanism of the increment of human adrenal DHEA production at adrenarche. Therefore, in this study we have examined the existence of maturative changes in the expression of the 3βHSD type II gene. To this end, we have studied the abundance of 3βHSD mRNA in 11 specimens of normal human adrenal tissue obtained from prepubertal and early pubertal subjects.

MATERIALS AND METHODS

Clinical Material

Adrenal glands from subjects aged 0.1 to 13 y were obtained from organ donors, patients undergoing resection of the kidney for renal neoplasms or necropsies with less than 6 h of post-mortem time (Table 1). Immediately after adrenal gland removal, all samples were stored at -190°C for mRNA analysis. The study was approved by the Research Committee of the Garrahan Pediatric Hospital.

Table 1 Patient's age, sex, and source of adrenal tissue
Full size table

For analysis of the results, subjects were divided into two groups as a function of age: group 1, less than 8 y old (n = 6; range 0.1 to 2.48 y) and group 2, equal or older than 8 y old (n = 5; range, 8 to 13 y). This division was based on the mean age of adrenarche reported in normal human subjects(2).

Preparation of RNA

Total RNA was isolated from tissues by homogenizing tissue samples in the presence of 1 mL of trizol reagent (Life Technologies, Inc.) per 50-100 mg of tissue, according to the manufacturer's instructions. Extraction was carried out in the presence of 0.2 mL of chloroform per 1 mL of trizol reagent. Samples were centrifuged at 12 000 × g for 15 min at 4°C. RNA was precipitated from the aqueous phase by mixing with 0.5 mL of isopropyl alcohol at -20°C for 2 h. The RNA pellet was washed with 1 mL of 75% ethanol, dissolved in RNAase free water, and stored at -85°C.

RNA Analysis

In most samples, 3βHSD type II mRNA was analyzed by do not and by relative RT-PCR. The size of the 3βHSD type II mRNA transcript was examined by Northern blot in three samples.

Dot blot. The method of Voutilainen et al.(15) was followed for dot blot analysis. RNA samples were denatured in 7.5% formaldehyde and 6 × SSC (0.9 M NaCl, 0.09 M sodium citrate) at 60°C for 30 min and then spotted on nitrocellulose filters (BA 85, Schleicher and Schuell). Filters were baked for 2 h at 80°C and prehybridized in buffer containing 50% formamide, 6 × SSC, 0.1% Ficoll, 0.1% BSA, 0.1% polyvinyl pyrrolidone, 100 µg/mL salmon sperm DNA, and 100 µg/mL yeast RNA for 4 h at 42°C before probe was added. Hybridization was performed using labeled human placental 3βHSD cDNA (kindly supplied by Dr. F. Labrie)(16). Probes were labeled in the presence of deoxy-[α-32P]CTP, using the technique described by Feinberg and Vogelstein(17). After hybridization, the filters underwent four 20-min washes in 0.2 × SSC, 0.1% SDS at 50°C, before autoradiography at -70°C for 24 h. Dots for all samples studied were analyzed on a single film. The uniformity of loading was tested using a labeled human βactin cDNA probe generated by PCR using the following primers: 5′ GGACCTGACTGACTACCTCATGAA 3′ and 5′ GATCCACATCTGCTGGAAGGTGG 3′ (a 524-bp amplified fragment from exon 3 to exon 5). The relative 3βHSD/βactin intensities of hybridization signals (arbitrary units) were quantified by scanning densitometry.

RT-PCR. First strand of cDNA was synthesized using Molony murine leukemia virus reverse transcriptase (Pro-mega). One to 5 µg of total RNA was incubated at 37°C for 60 min, with 200 U of enzyme, 1 × reaction buffer, 20-40 U of rRNAsin ribonuclease inhibitor, 200 ng of oligo-dt, and 10 mM of each deoxynucleotide triphosphate. The exponential phase for both 3βHSD type II and βactin PCRs was standardized using placental RT-derived cDNA at different cycles. It was found that 20 cycles corresponded to the exponential phase in the two PCRs. Relative PCR was performed in replicates using two dilutions of the RT product as template. 3βHSD and βactin cDNA were coamplified at the exponential phase (20 cycles) in the same tube, using specific primers. The βactin primer was prepared as described previously(18). Amplification of a 421-bp fragment (from bp +36 to +457) of the 3βHSD type II gene was performed using the following primers: 5′ GGGCTGGAGCTGCCTTGTGA 3′ and 5′ TCGTGGCGTTCTGGATGAT 3′ PCR was carried out in a 25-µL mixture containing 10 × reaction buffer, 2.5 µL of MgCl2 (25 mM), 2.5 µL of DMSO, 1 µL of deoxynucleotide triphosphate mixture (1.25 mM each one), 4 µL of 0.05 µM of each forward and reverse βactin primers, 0.5 µM of each forward and reverse 3βHSD type II primers, and 1.5 U of Taq polymerase. The 5′ ends of forward primers were labeled with [γ-32P]ATP by T4 polynucleotide kinase (Promega) before PCR amplification. Labeled PCR products were resolved by electrophoresis on 6% denaturing polyacrylamide gels. The expected bands were visualized by autoradiography, and radioactivity was measured in a β-scintillation counter. The 3βHSD/βactin ratio was calculated for every sample. Measurements were compared at two dilutions to confirm that values were quantitatively the correct ones as expected in the exponential phase.

Northern blot. The method of Sambrook et al.(19) was followed for Northern blot analysis. About 15 µg of RNA was denatured in a buffer containing 10% formaldehyde and 50% formamide and electrophoresed on 10% formaldehyde, 1% agarose gel. RNA from these gels was transferred to a nitrocellulose membrane, and the blot was baked and prehybridized as described above for dot blots. Autoradiography was carried out at -70°C for 5 d. As expected, a 1.7-kb band(20) was found for the 3βHSD type II transcript (Fig. 1, lower panels).

Figure 1
figure 1

Upper panels, autoradiography of dot blots after successive hybridization (and stripping) with βactin and 3βHSD type II. From a3 to d1, both for 3βHSD type II (left) and βactin (right), dots correspond to adrenal RNA from the 11 patients, arranged by age from the youngest to the oldest, from left to right in the four rows (d2, placental tissue; d3, normal adult adrenal; d4, normal peripheral leukocytes). Middle panel, 6% PAGE of cDNA-amplified products after relative RT-PCR in one subject of group 1 and in two subjects of group 2. Lanes 1-4, adrenal tissue from a 0.56-year-old subject at two dilutions in duplicate; lanes 5-7, adrenal tissue from a 9-year-old subject at two dilutions, single sample (lane 5) and duplicates (lanes 6 and 7). Lane 8, negative control (no template). Lanes 9-12, adrenal tissue from an 11-year-old subject at two dilutions, in duplicate. Lower panels, Northern blot of 3βHSD type II (left) and βactin (right) mRNA in three patients, aged 0.1, 0.56, and 1.64 y.

Full size image

To check if the expression of βactin mRNA in human prepubertal adrenal tissue is age dependent, Northern blots analysis, using 7 µg of total RNA and a labeled human βactin cDNA probe, was carried out in four samples of group 1 and in three samples of group 2. Data were normalized to ethidium bromide-stained 18 S ribosomal RNA. To this effect, the relative βactin/18 S intensities were determined using a Molecular Image System (Bio-Rad GS-505). Means (±SD) were 0.35 ± 0.03 in group 1 and 0.28 ± 0.06 AU in group 2 (p not significant).

Statistical Analysis

All 3βHSD type II m RNA values are reported as median and range. Data were analyzed using the Mann-Whitney's test. p < 0.05 was considered statistically significant.

RESULTS

Measurement of 3βHSD type II mRNA by dot blot in groups 1 and 2 is shown in Table 1 and in Figure 1 (upper panels). For a valid comparison, RNA samples from adrenal tissues obtained from the 11 subjects were studied using the same probes in the same blot. Median 3βHSD type II mRNA in group 1 was significantly higher than in group 2 (p < 0.05).

Results of relative RT-PCR for every sample is also shown in Table 1. Two patients could not be studied with this method. Similarly to what was observed by dot blot, the median of group 1 was significantly higher than that of group 2 (p < 0.05). Examples of 6% PAGE of amplified cDNAs in three subjects are shown in Figure 1 (middle panel).

DISCUSSION

The data presented here show, by two different methods, that in human adrenal tissue the 3βHSD type II mRNA transcript in the young prepubertal group is significantly higher than in the old prepubertal and early adolescent group. Even though we have determined abundance of 3βHSD type II mRNA, it is assumed that abundance reflects gene expression rather than degradation of the transcript. This finding might be related to the mechanism of adrenarche. At adrenarche, the human adrenal cortex begins to secrete significant quantities of Δ5 C19 steroids, suggestive of an altered regulation of 3βHSD type II activity relative to other enzymes of the steroidogenic pathway, particularly cytochrome P450c17. It has been suggested that the adrenal 3βHSD expression in humans may be the critical regulatory step governing production of the various C21 and C19 steroid hormones(21,22).

The regulation and function of the primate fetal adrenal gland has been a subject of considerable interest(23). The fetal zone of the human fetal adrenal has a remarkable capacity to produce very large amounts of DHEA and minimal cortisol, suggesting a major selective block of either 3βHSD type II enzyme activity or gene expression during fetal life. Indeed, it was established that the massive secretion of Δ5 C19 steroids from the fetal zone of fetal adrenal occurred because of the absence of 3βHSD type II gene expression and not inhibition of 3βHSD type II activity(24). The nature of the suppression of the 3βHSD type II gene in fetal adrenal has to be elucidated. Studies of functional zonation have established that the fetal zone of the fetal adrenal is the site of Δ5 steroid production and of repression of the expression of the 3βHSD type II gene(25).

In adult human adrenal cortex, the zona reticularis is the site of biosynthesis of DHEA and DHEA sulfate resulting from the low gene expression of 3βHSD type II mRNA. The age-related decline in adult adrenal androgen biosynthesis might be secondary to an age-related decline in the number of functional reticularis cells(26). Furthermore, in a previous study, Gell et al.(27) have reported an age-dependent decrease in immunohistochemically localized 3βHSD in adrenal zona reticularis during prepuberty and have suggested that this decrease may contribute to the increased production of DHEA and DHEA sulfate seen at adrenarche.

Adrenal steroidogenesis might be under the influence of paracrine regulators. It has been reported that the effect of TGFβ1 in cultured human fasciculata reticularis cells is not to reduce basal or ACTH-stimulated cortisol production but to decrease the production of DHEA sulfate(28). These authors postulated that a local diminution of TGFβ1 might be involved in the mechanism of adrenarche.

On the other hand, it has been suggested that adrenarche might be secondary to an increment of 17,20 lyase activity by a regulatory effect independent of 17 α-hydroxylase activity. It has been suggested that an unidentified intra-adrenal event influencing the molar abundance of redox partners(11–13) rather than the level of P450c17 mRNA, protein abundance, or P450 oxidoreductase abundance would be responsible for the regulatory shift. In addition, it has been hypothesized that the regulation of 17,20 lyase activity was mediated by posttranslational modification of P450c17 by serine threonine phosphorylation in response to a cAMP-dependent mechanism. This would increase the efficiency of electron transfer secondary to an increment in the affinity of P450c17 for P450 oxidoreductase(29).

3βHSD type II is an NAD+ membrane-bound enzyme, but its intracellular localization remains controversial because the enzyme has been located in the microsomal and in the mitochondrial fractions of adrenal tissue(30–33). In this respect, and because little is known about the anchoring of 3βHSD to the membrane, an interaction between the decrement of 3βHSD activity and an increment of 17,20 activity cannot be ruled out.

In summary, our data indicate that a decrease in the activity of 3βHSD type II enzyme, secondary to an inhibition in the expression of the 3βHSD type II gene in adrenal cells, is an important underlying mechanism in the increase of DHEA secretion characteristic of adrenarche.