Developmental Expression of Genes Involved in ...

0013-7227/03/$15.00/0 Printed in U.S.A.

Endocrinology 144(7):2902–2911 Copyright © 2003 by The Endocrine Society doi: 10.1210/en.2002-0073

Developmental Expression of Genes Involved in Neurosteroidogenesis: 3␤-Hydroxysteroid Dehydrogenase/⌬5-⌬4 Isomerase in the Rat Brain CHRYSTELLE IBANEZ, RACHIDA GUENNOUN, PHILIPPE LIERE, BERNARD EYCHENNE, ANTOINE PIANOS, MARTINE EL-ETR, ETIENNE-EMILE BAULIEU, AND MICHAEL SCHUMACHER Institut National de la Sante´ et de la Recherche Me´dicale, Unite´ 488, Ste´roı¨des et Syste`me Nerveux, 94276 Biceˆtre, France In the central nervous system, neurosteroids, in particular progesterone, have neurotrophic and neuroprotective effects. We thus decided to study the developmental expression of 3␤-hydroxysteroid-dehydrogenase/⌬5-⌬4 isomerase (3␤HSD), an enzyme that converts pregnenolone to progesterone, in the male rat brain at 0, 7, 14, and 70 d after birth. 3␤HSD mRNA was widely distributed throughout the brain, as shown by in situ hybridization. At all ages, the same cerebral structures were labeled, but the intensity of the hybridization signal constantly decreased during postnatal development. As the hippocampus is of particular interest because of its neuronal plasticity, we chose to quantify the changes in 3␤HSD mRNA levels as well as progesterone and pregnenolone concentrations in this structure. Quantitative in situ hybridization confirmed a decrease in the expression of

S

TEROIDS ARE PRESENT in the nervous system and have been shown to exert neurotrophic and neuroprotective effects. Among them are the hormonal steroids, which are produced by the endocrine glands and easily reach the brain via the bloodstream, and the neurosteroids, which are synthesized locally in the nervous system de novo from cholesterol or from steroid precursors (1, 2) and may act via paracrine or autocrine modes of signaling. The synthesis of progesterone has been demonstrated in cultured neurons and glial cells of the central nervous system (CNS) (3, 4) and the peripheral nervous system (PNS) (5, 6). In addition to their action via the classical steroid nuclear receptors, neurosteroids have been shown to modulate the activities of membrane receptors for neurotransmitters (7), in particular the ␥-aminobutyric acid A receptor (8), the nicotinic acetylcholine receptor (9), the N-methyl-d-aspartate receptor (10), and the ␴ type 1 receptor (11, 12). Progesterone, independently of its origin (hormone or neurosteroid), has multiple actions in the nervous system. Several neuroprotective effects of progesterone have been identified in animal models of neurodegeneration (13), spinal cord trauma (14), cerebral edema (15), and after transient middle cerebral artery occlusion (16) in rats. Progesterone also plays a role in myelination by increasing the expression of genes encoding proteins involved in the ultrastructure of the myelin sheaths in both the PNS and CNS. This characAbbreviations: AD, Adult rats; CNS, central nervous system; DHP, 5␣-dihydroprogesterone; GC/MS, gas chromatography/mass spectrometry; 3␤HSD, 3␤-hydroxysteroid-dehydrogenase/⌬5-⌬4 isomerase; IS, internal standard; P, postnatal day; PNS, peripheral nervous system.

3␤HSD mRNA with progressing age, as revealed by a significant reduction in the density of silver grains per cell in the CA1 layer. This decrease was confirmed by semiquantitative RT-PCR on hippocampal samples. Concentrations of hippocampal pregnenolone and progesterone measured by gas chromatography/mass spectrometry were highest on the day of birth and lower at the other ages. Plasma concentrations of these steroids were lower than those in the hippocampus, suggesting that they may have been mostly synthesized in situ since the day of birth. These results demonstrate variations in the expression of a gene coding for an enzyme critically involved in progesterone synthesis in the hippocampus throughout postnatal development. (Endocrinology 144: 2902–2911, 2003)

teristic has been demonstrated in primary cultures of Schwann cells for two specific peripheral myelin genes, P0 and PMP22 (17), and in cultured CNS glial cells for myelin basic protein and 2⬘,3⬘-cyclic nucleotide 3⬘-phosphodiesterase (18, 19). Progesterone, synthesized by Schwann cells, has been shown to promote the formation of new myelin sheaths after cryolesion of the sciatic nerve of male mice (20). The synthesis of neurosteroids requires the presence of functional enzymes in the nervous system. Conversion of pregnenolone to progesterone is accomplished by the 3␤hydroxysteroid-dehydrogenase/⌬5-⌬4 isomerase (3␤HSD) enzyme. This enzyme is expressed at high levels in classical steroidogenic tissues, i.e. adrenal cortex, testis, ovary, and placenta (21, 22), and also in liver and skin (23). Four isoforms of rat 3␤HSD cDNAs have been characterized (21, 22, 24), and their expression is tissue specific (25). 3␤HSD mRNA expression has been determined in the adult rat brain by in situ hybridization (26), and the type I and II isoforms are the major ones in the brain (26, 27). As progesterone influences neurobiological processes, study of expression of the enzyme involved in its synthesis during rat brain development is necessary. The postnatal period is of particular interest, as many maturation processes take place in the nervous system at this time: cell proliferation, axonal growth, establishment of synaptic contacts, and myelination. The first part of this paper is devoted to a description of the cerebral structures expressing 3␤HSD mRNA, and then a more focused study on its mRNA levels and on progesterone and pregnenolone concentrations in the hippocampus is reported. The latter structure undergoes im-

2902

Ibanez et al. • 3␤HSD and Postnatal Development of the Brain

portant maturation events after birth and is of particular interest because of its implication for learning and memory processes and for neurodegenerative diseases such as Alzheimer’s disease. The hippocampal formation is a steroid-sensitive structure because of the presence of steroid receptors, in particular for progesterone (28), and of steroidsensitive receptor for neurotransmitters. Three postnatal stages were chosen for this study, 0, 7, and 14 d after birth (P0, P7, and P14), and compared with results obtained in the adult rats (AD). Materials and Methods Animals Neonatal and adult male Sprague Dawley rats were used for all experiments (Iffa Credo, L’Arbesle, France). The neonates were used on the day of birth (P0) or either 7 or 14 d after birth (P7 and P14). All adult rats (AD) used for this study were 10 wk old.

In situ hybridization Tissue processing. Animals were perfused via the left ventricle under deep pentobarbitone anesthesia (40 mg/kg, ip) with 0.9% saline, followed by 1% paraformaldehyde. The brain and adrenal gland were dissected, postfixed in the same fixative for 1 h, then cryoprotected by immersion in 15% sucrose buffer for 24 h and frozen in dry ice. Serial sections (14 ␮m) were cut in a cryostat at –20 C, mounted on siliconized slides, and stored at ⫺80 C until use. Probe preparation and labeling. A 40-base synthetic oligonucleotide, 3␤HSDI-IV (5⬘-GATGCTCCCATTGGCTGCCAGCACTGCCTTCTCGGCCATC-3⬘), which is complementary to a sequence common to the four known isoforms of this enzyme (nucleotides 690 – 651 according to the sequence of the 3␤HSD type I isoform), was labeled with [␣-35S]deoxy-ATP (1000 mCi/mmol; NEN Life Science Products, Gentilly, France) to a specific activity of 2 ⫻ 109 cpm/␮g using terminal deoxynucleotidyltransferase (Amersham International, Les Ulis, France). Hybridization procedure and signal detection. Adjacent brain sections were hybridized with the labeled probe. Hybridization was performed according to a published protocol (29) with some modifications. Briefly, 0.25 ng labeled probe in the hybridization buffer (50% deionized formamide, 10% dextran sulfate, 500 ␮g/ml denatured salmon sperm DNA, 1% Denhardt’s solution, 5% sarcosyl, 250 ␮g/ml yeast tRNA, 200 mm dithiothreitol, and 20 mm Na2PO4 in 2⫻ standard saline citrate buffer) was applied on tissue sections, and hybridization proceeded overnight at 40 C. The sections were then rinsed in several washes of standard saline citrate buffer (1⫻ and 0.1⫻), dehydrated, and placed in contact with Kodak BioMax film (Eastman Kodak, Rochester, NY) for 5 d. Then they were coated with Ilford K5 emulsion, exposed for 5– 8 wk, developed, and stained with toluidine blue. Sections from all of the different postnatal stages were submitted to the same time of exposure, which was chosen to avoid saturation of the hybridization signal. Sections from adrenal gland served as positive controls and were processed under the same conditions as the brain sections in the same experiments. Controls for specificity of the in situ hybridization reaction were performed by the addition of an excess of related and unrelated probes to the hybridization medium. In situ hybridization quantification on emulsion autoradiography. To make the data from each developmental stage comparable, we took the precaution of processing the sections from each postnatal stage together with the same probe preparation and in the same buffer preparations during the in situ hybridization protocol. The density of silver grains, expressed as the number of grains per square micron per cell, was quantified in the CA1 layer of the hippocampus in the pyramidal cells, comparing equivalent planes of section between the different stages of development. At each stage and for each animal (3 animals/stage), the number of grains was counted in 100 pyramidal cells of the CA1 layer, and the cell surface was measured using NIH imaging software (n ⫽ 300 cells/stage).

Endocrinology, July 2003, 144(7):2902–2911 2903

Semiquantitative RT-PCR Total RNA was extracted from hippocampus using TRIzol reagent (Life Technologies, Cergy-Pontoise, France) according to the manufacturer’s instructions. The concentration and purity of total RNA were determined from the OD at 260 and 280 nm. Total RNA was subjected to deoxyribonuclease I treatment (Stratagene, Paris, France) for 10 min at 37 C to remove possible residual contaminating genomic DNA. cDNA templates for PCR amplification were synthesized from 1 ␮g total RNA using Superscript II ribonuclease H reverse transcriptase (Invitrogen, Cergy-Pontoise, France) for 90 min at 42 C in the presence of random hexamer primers. Three microliters of the RT products were subjected to PCR amplification. A forward primer (5⬘-GCCTCCACCTTGATTCCAGTTGGA-3⬘) and a reverse primer (5⬘-TGTAAAATGGATCCAGCAGGAAGC-3⬘) common to rat 3␤HSDI–IV cDNA were used for amplification of a 213-bp fragment. A 386-bp fragment of rat 18S ribosomal RNA was amplified in parallel in separate reactions for normalization of the results using a forward primer (5⬘-CTACCACATCCAAGGAAGGC-3⬘) and a reverse primer (5⬘-CTCGGGCCTGCTTTGAACAC-3⬘). Each PCR contained cDNA template, 1⫻ Taq DNA polymerase buffer, 0.1 ␮m forward and reverse 3␤HSDI–IV or 18S primers, 100 ␮m of each deoxy-NTP, 0.025 ␮Ci [32P]deoxy-CTP to radiolabel the amplification products, and 1 U Taq DNA polymerase (ATGC Biotechnologie, Marne-la-Valleé, France) in a total volume of 50 ␮l. A variable number of cycles (between 20 and 40) at 94 C for 1 min, 57 C for 1 min, and 72 C for 1 min were performed during the PCR amplification to verify that both PCR products amplify linearly and to determine the optimal number of cycles allowing their detection without saturation of the signal. Using this procedure, 20 and 35 cycles were chosen for 18S and 3␤HSDI–IV, respectively. After amplification, a 20-␮l aliquot of each reaction and a 100-bp DNA ladder were separated on a 1.2% agarose gel, visualized by ethidium bromide staining, blotted, and quantified using an Instant Imager (Packard Instrument Co., Meridian, CT). The relative levels of 3␤HSDI–IV gene expression were measured by determining the ratio between the radioactive products generated from the target gene 3␤HSDI–IV and the endogenous internal standard 18S (n ⫽ 3 animals/ stage).

Gas chromatography/mass spectrometry (GC/MS) Plasma and tissue concentrations of progesterone and pregnenolone were measured according to a published protocol (30). Briefly, progesterone and pregnenolone were extracted from hippocampus in methanol after sonication, centrifugation, and addition of an internal standard (IS; 5␤-androstane-3␤-ol,17-one). The supernatant was then purified by solid phase extraction on a C18 reverse phase chromatography column, and the unconjugated steroids were purified by HPLC. The steroids were derivatized with heptafluorobutyric anhydride and analyzed by GC/MS with electronic impact ionization and in the single ion monitoring detection mode. The ions used for quantification of the derivatized steroids were selected according to their mass/charge (m/z) ratios: m/z 510 for progesterone, m/z 298 for pregnenolone, and m/z 486 for IS. Progesterone and pregnenolone concentrations in the hippocampus were calculated with a standard curve of steroids and IS peak area ratios (n ⫽ 4 animals/stage).

Statistical analysis All numerical results are expressed as the mean ⫾ sem. Comparisons of the results at different postnatal stages were carried out with the nonparametric Kruskal-Wallis test (significance level, P ⬍ 0.05) for quantitative in situ hybridization, semiquantitative RT-PCR experiments, and GC/MS analysis. The Mann-Whitney test is used to compare hippocampal vs. plasma steroid concentrations measured by GC/MS on P0 (significance level, P ⬍ 0.05).

Results 3␤HSD mRNA is largely distributed in the neonatal and adult rat brain

The expression of 3␤HSD mRNA along the rostro-caudal axis of the brain on serial coronal sections was determined

2904

Endocrinology, July 2003, 144(7):2902–2911

either in autoradiograms to show the topography of cells expressing the 3␤HSD gene (Figs. 1–3) or on emulsion-coated slides to obtain a cellular resolution in cerebral areas of particular interest. In situ hybridization control experiments with the [␣-35S]deoxy-ATP-labeled oligonucleotide 3␤HSDIIV probe showed a strong hybridization signal in adrenal gland sections, which served as a positive control (not shown), limited to the adrenal cortex. The specificity of the hybridization signal was confirmed by displacement experiments; in the adrenal gland and in all brain structures, a 500-fold excess of unlabeled homologous oligonucleotide completely abolished the hybridization signal, as shown in Fig. 2. It is interesting to note that 3␤HSD mRNA was expressed in the same cerebral structures at all ages. Indeed, if the autoradiograms obtained on P0 were compared with those of the adults, the same structures expressed 3␤HSD mRNA throughout postnatal development and in the adults (Figs. 1–3). The anatomical distribution of 3␤HSD mRNA is summarized in Table 1.

FIG. 1. Representative in situ hybridization experiment showing 3␤HSD mRNA expression on autoradiographic films during postnatal development on P0, P7, and P14 and in AD with brain coronal sections of the frontal cortex (Fr.Cx), olfactory tubercle (Tub.), piriform cortex of the olfactory tubercle (Pir.), caudate-putamen (CdP), and accumbens nucleus (Acb). The anterior commissure (ac) and corpus callosum (cc) remained unlabeled.


At the rostral level of the brain, autoradiography on x-ray films demonstrated hybridization signal for 3␤HSD in the frontal cortex and the olfactory tubercle (Fig. 1). In situ hybridization emulsion autoradiography revealed that cells of the piriform cortex in the olfactory tubercle were the most strongly labeled (not shown). The striatum (caudate-putamen), accumbens nucleus, and cerebral cortex showed a strong labeling, whereas the corpus callosum and the anterior commissure both exhibited a very weak signal (Fig. 1). On emulsion-coated slides, cells in the corpus callosum exhibited a signal that could not be considered different from the background. Also, the globuspallidus exhibited no hybridization signal, in contrast to the septum, which expressed 3␤HSD mRNA (Fig. 2). The cerebral cortex showed a laminar pattern of labeling, which was similar at all ages, and no obvious regional differences were revealed for a given age (Figs. 2 and 3). Further along the rostro-caudal axis of the brain, the hippocampus showed a strong pattern of labeling from P0, as



FIG. 2. Representative in situ hybridization experiment showing 3␤HSD mRNA expression on autoradiographic films during postnatal development on P0, P7, and P14 and in AD with brain coronal sections of the cerebral cortex (Cer.Cx.), striatum (S), septum (Sp), and thalamus (Th). The globus pallidus (GP) and corpus callosum (cc) were unlabeled. On the right are the control experiments with a 500-fold excess of unlabeled probe.

can be seen in the autoradiograms (Fig. 3). Emulsion autoradiography demonstrated that pyramidal cells of the CA1, CA2, and CA3 layers were strongly labeled (Fig. 4A) as were granule cells of the dentate gyrus. In adult animals, the CA3 region of the hippocampus exhibited the weakest signal compared with the other layers (Fig. 3). Diencephalic structures, such as thalamic and hypothalamic nuclei, expressed the 3␤HSD gene (Figs. 2 and 3). From P0 to adulthood, both granular and molecular layers of the cerebellum were intensely labeled (Fig. 3). As the cerebellum has a very specific postnatal process of maturation, it is important to specify that on P0 and P7, both the external and internal granular layers showed a positive hybridization signal. The external layer disappeared progressively with age as granule cell progenitors were migrating toward the internal layer and was almost absent on P14, but the mature internal layers still retained high levels of 3␤HSD mRNA even in the adult. Emulsion autoradiography showed that granule cells and Purkinje cells, situated just beneath the granular layer, strongly expressed 3␤HSD mRNA (not shown).

In the molecular layer, stellate and basket cells were also labeled. In the early postnatal stages, that is, P0 and P7, most of these cells were present as progenitors, and this observation suggests that both the progenitors and the differentiated cells express the mRNA encoding for 3␤HSD. It should be noted that the white matter layers remained unlabeled at all ages (Fig. 3). In situ hybridization emulsion autoradiography did not reveal cells significantly exhibiting a positive hybridization signal for 3␤HSD mRNA in the white matter. Expression of 3␤HSD mRNA decreases with age in the hippocampus

The autoradiograms showed that even though the cerebral distribution of 3␤HSD mRNA did not change during postnatal development, a striking feature was that the intensity of the hybridization signal on x-ray films decreased with age, suggesting that the level of 3␤HSD mRNA expression decreases as the animals age (Figs. 1–3). Because cell density and cell size are different at different

2906



FIG. 3. Representative in situ hybridization experiment showing 3␤HSD mRNA expression on autoradiographic films with brain coronal sections of the hippocampus and cerebellum during the postnatal development on P0, P7, and P14 and in AD. CA1 and CA3 layers of Ammon’s horn (CA1 and CA3), dentate gyrus (DG), hypothalamus (Hyp.), andcerebellar white matter (wm) are indicated.

stages, the analysis of the developmental decrease in 3␤HSD mRNA expression from autoradiograms is inaccurate. A more accurate method of quantification is to determine the density of silver grains per cell (number of grains per square micron per cell) using in situ hybridization emulsion autoradiography and to compare equivalent planes of sections between the different stages of development. This method was used on pyramidal cells of the CA1 layer of the hippocampus, which is a layer of particular interest in terms of synaptic connectivity and plasticity. The distribution of silver grain density per cell is shown in Fig. 4B for each developmental stage. On P0, the majority of cells exhibited between 24 –26 grains/␮m2/cell. This density decreased as development progressed; most cells had 16 –18 grains/␮m2/ cell on P7, 12–14 on P14, and 6 – 8 in adults. The decrease became obvious when the means of grain density per cell were compared between the different ages (Fig. 4C; ***, P ⬍ 0.0001). Moreover, the expression of 3␤HSD mRNA was

significantly higher on P0 than at all other ages (by MannWhitney test, P0 vs. P7: ***, P ⬍ 0.0001). Thus, the decrease in the hybridization signal intensity observed on x-ray films is not due to a decrease in cellular density, but corresponds to a real decrease in the expression of 3␤HSD mRNA per cell. The decrease with age in 3␤HSD mRNA expression in the hippocampus is confirmed by semiquantitative RT-PCR

We used another approach to confirm the results obtained by quantitative in situ hybridization and performed semiquantitative RT-PCR for 3␤HSD mRNA in the hippocampus from neonatal and adult animals in parallel with RT-PCR for 18S mRNA as an internal standard. Evolution of 3␤HSDI-IV mRNA expression was evaluated by determining the ratio of 3␤HSDI-IV/18S PCR products. After normalization of the results, the levels of 3␤HSD mRNA expression were com-


TABLE 1. Table summarizing the anatomical distribution of 3␤HSD mRNA Unlabeled cerebral structures

Labeled cerebral structures

Olfactory tubercle (piriform cortex) Frontal cortex Cerebral cortex Striatum (caudate-putamen) Accumbens nucleus Hippocampus layers CA1 CA2 (pyramidal neurons) CA3 Dentate gyrus (granular neurons) Thalamus Hypothalamus Cerebellum Granular layer (granule cells) Molecular layer (stellate and basket cells) Purkinje cells

Anterior commissure Corpus callosum Globus pallidus

}

Cerebellar white matter

pared between the different postnatal stages (Fig. 5A). As seen with in situ hybridization, a progressive decrease in 3␤HSD gene expression was observed as the animals got older (Fig. 5B; *, P ⬍ 0.05). Changes in hippocampal progesterone and pregnenolone concentrations during postnatal development

We then measured progesterone and pregnenolone concentrations in the hippocampus and blood plasma from neonates and adult animals using GC/MS. Once again we observed a variation with age. Hippocampal concentrations of pregnenolone and progesterone were highest on P0 (4.6 ⫾ 0.9 and 2.1 ⫾ 0.3 ng/g, respectively) and then dropped during the first postnatal week to levels that remained similar to those observed in adults (2.0 ⫾ 0.3 and 1.1 ⫾ 0.2 ng/g, respectively; Fig. 6; P ⬍ 0.05). In contrast, plasma steroid levels were much lower than those measured in the hippocampus (*, P ⬍ 0.05 on P0) and did not show any significant changes during development (Fig. 6). In the hippocampus, concentrations of pregnenolone were higher than those of progesterone at all ages. Discussion

Neurotrophic and neuroprotective effects of progesterone in the nervous system are now well documented. Moreover, as indicated initially, progesterone is considered as a neurosteroid. After removal of the peripheral steroidogenic glands, it remains present in the brain of adult male rats. We thus found it important to study the evolution of the expression of the enzyme involved in progesterone synthesis, 3␤HSD, and the evolution of progesterone and pregnenolone concentrations at 0, 7, and 14 d, and 10 wk after birth in male rat brains. Cerebral distribution of 3␤HSD mRNA

The present study represents the first systematic study describing the developmental changes of 3␤HSD mRNA within distinct brain regions. In situ hybridization experi-


ments showed that 3␤HSD mRNA is largely distributed in the brain throughout postnatal development and that the same cerebral structures are labeled at all ages. The olfactory tubercles, cerebral cortex, striatum, hippocampus, thalamus, hypothalamus and the cerebellum are positively labeled for 3␤HSD mRNA. These observations indicate that the expression of 3␤HSD messengers is region-specific. Developmental expression of 3␤HSD mRNA has previously been examined by Kohchi et al. (27) using semi-quantitative RT-PCR in larger brain regions of the rat. It was detected in the cerebrum, diencephalon, mesencephalon, and cerebellum from 3– 61 d after birth. In terms of cerebral localization, these results are in accordance with ours, because these regions of the brain contain the cerebral structures positively labeled for 3␤HSD mRNA in our in situ hybridization experiment. Other in situ hybridization studies performed in adult animals have to be mentioned. Our present data agree with observations previously obtained in our laboratory (26) and by Furukawa in 1998 (31). In contrast, other reports describe the presence of 3␤HSD messenger exclusively at the border of the fourth ventricle in the brainstem (32) or in the cerebellar white matter and corpus callosum (33). Differences in the technical procedures may explain these differences: they used a cDNA probe, instead of an oligonucleotidic probe, which penetrates less easily into the tissue. Moreover, the time of exposure used in these studies may not have been sufficiently long to detect the low amounts of 3␤HSD mRNA in other cerebral structures. 3␤HSD mRNA expression in the hippocampus

The hippocampus and, in particular, the pyramidal neurons of the CA1–CA3 layer of Ammon’s horn and the granule neurons of the dentate gyrus, express 3␤HSD mRNA during postnatal life and in the adult. From P0 to adulthood, 3␤HSD mRNA expression decreases progressively as confirmed both by in situ hybridization and RT-PCR experiments. 3␤HSD mRNA levels are the highest on P0, a period of intense cerebral maturation. These results reveal the existence of a developmental regulation of 3␤HSD mRNA expression either at the transcriptional level or at the level of mRNA stability. Other experimental paradigms could be of interest, such as the investigation of the consequences of progesterone depletion on the developmental process, for example by locally inhibiting progesterone function. As the probe used in the in situ hybridization experiments and the primers used in the RT-PCR protocol recognize the four isoforms of 3␤HSD, it is unknown whether the developmental changes in the expression of its mRNA affect all or only one of the isoforms. It has already been demonstrated, in the adult rat, that the type I and II isoforms are the major ones in the brain (26, 27). However, it is clearly impossible to decide whether both or only one of these isoforms are up-regulated on P0. We cannot even exclude the existence of a transient expression of another isomer of 3␤HSD, which would be particularly important in the early development and then disappears progressively with age.

2908


FIG. 4. A, Representative in situ hybridization emulsion autoradiography showing 3␤HSD mRNA expression in pyramidal neurons in the CA1 layer of the hippocampus at different postnatal stages: P0, P7, and P14 and AD. Scale bar, 50 ␮m. B, Distribution of silver grain density per cell on P0, P7, and P14 and in AD (n ⫽ 300 cells/ stage). C, Comparison of the means of grain density per cell between the different stages by Kruskal-Wallis test (***, P ⬍ 0.0001; KW ⫽ 641.2; df ⫽ 3).




FIG. 5. A, Representative RT-PCR experiment showing the variation in 3␤HSD mRNA expression in the hippocampus during postnatal development on P0, P7, and P14 and in AD. The negative control was water (H2O). the positive control was mRNA from testis (T). B, Semiquantitative analysis after normalization with 18S (n ⫽ 3 animals/ stage). *, P ⬍ 0.05, by Kruskal-Wallis test (KW ⫽ 7.3; df ⫽ 3).

FIG. 6. Concentrations of progesterone and pregnenolone in plasma (nanograms per milliliter) and hippocampus (nanograms per gram) measured by GC/MS during postnatal development on P0, P7, and P14 and in AD (n ⫽ 4 animals/stage). PROGp, PREGp, PROGh, and PREGh are the concentrations of progesterone and pregnenolone in the plasma and hippocampus, respectively. Levels of significance were determined by Kruskal-Wallis test (*, P ⬍ 0.05; KW ⫽ 8.8; df ⫽ 3). On P0, steroid concentrations in the hippocampus were significantly higher than in plasma (*, P ⬍ 0.05 for progesterone and pregnenolone, by Mann-Whitney test).

Progesterone and pregnenolone concentrations in the hippocampus

The measurements of progesterone and pregnenolone levels in plasma and hippocampus provide two major items of

information. First, the concentrations of progesterone and pregnenolone measured in the hippocampus were higher than in the plasma, suggesting that these steroids may be synthesized in situ. These results provide further support for the neurosteroid concept and strongly suggest an important role for the local synthesis of neurosteroids during brain maturation. This local synthesis of progesterone in the brain may be particularly important on P0 when its levels and those of its precursor, pregnenolone, are the highest. In contrast to the hippocampus, on P0, circulating levels of these two steroids remain low. The hypothesis of a local synthesis of progesterone is reinforced by the enzymatic activity of 3␤HSD, which can be significantly inhibited by trilostane, on P10 found in another cerebral structure, cerebellar slices, incubated with pregnenolone (34). However, the hypothesis of an accumulation of maternal progesterone in the newborn brains cannot be formally excluded. Circulating progesterone is elevated during pregnancy and can easily cross the placenta, and due to the lipidic composition of the brain, progesterone can accumulate, free or conjugated to a binding protein/receptor. On the other hand, it is known that in preparation for labor, progesterone blood levels drop dramatically before parturition (35). The latter researchers have also demonstrated that the plasma progesterone concentration is much lower in the fetus than in the mother, and that the decrease in progesterone concentration starts earlier in the fetus than in the mother. Therefore, there is a period of time (several days) during which the source of maternal progesterone and fetal progesterone levels are declining dramatically. Moreover, progesterone, accumulated in the brain,

2910


could also be metabolized rapidly during this period. Even though we cannot translate the level of expression of a messenger directly into an enzymatic activity, it is nevertheless interesting to note the correspondence between the maximum levels of 3␤HSD mRNA and progesterone concentration on P0, which is another argument in favor of a local synthesis of this steroid even if the accumulated maternal progesterone in newborn rat brain on P0 could also participate partially in this phenomenon. Afterward, from P7 to the adult, it can be hypothesized that the hippocampal progesterone concentration, remaining higher than that in plasma, results mainly from endogenous neurosteroidogenesis, as we can exclude, at these ages, the persistence of maternal steroids in the cerebral tissue. Second, a striking parallelism between hippocampal concentrations of progesterone and pregnenolone was observed. Their concentrations were highest on P0 and much lower at all the other ages. Moreover, concentrations of pregnenolone in the hippocampus were always about twice those of progesterone at all ages. It should be kept in mind that hippocampal concentrations of progesterone and pregnenolone may also depend on the availability of cholesterol and the activity of the P450scc enzyme, which converts cholesterol to pregnenolone. Thus, understanding the potential regulatory mechanisms that may apply to 3␤HSD mRNA expression during development is even more complex. The expression of the other enzymes of steroidogenesis and metabolism during the development of the CNS and in the adult brain must be taken into account. P450scc expression has been found during both embryogenesis (36) and postnatal development (27) where its expression has been determined in the Purkinje cells of the cerebellum (37) and also in the adult cerebellum (38). The 5␣-reductase, which allows the formation of the 5␣-reduced progesterone metabolite, 5␣-dihydroprogesterone (DHP), has also been detected in rat brain both during development and in the adult (39, 40). Moreover, there is already other in vivo evidence for neurosteroid synthesis in the developing CNS (41) and, in particular, large amounts of 5␣-reduced progestins, DHP and 3␣,5␣-tetrahydroprogesterone, during fetal development (42). Hence, the machinery for C21 neurosteroid synthesis, from cholesterol to DHP, seems to be active from the first few days of life in rat brain, as all of the enzymatic tools are present. Attention was then paid to the cell types expressing 3␤HSD mRNA. In situ hybridization experiments on autoradiograms and emulsion autoradiography indicated that it seemed to be mainly expressed in neurons (cortex, hippocampus, striatum, and cerebellum) in vivo, whereas cells in the white matter areas did not exhibit a larger signal than the background. On the other hand, type 1 5␣-reductase activity has been found to be greater in myelin membranes (43). These observations could suggest the existence of cooperation between different cellular populations in the rat CNS. The myelination process, whose initiation is dependent on the contact between axons and glial cells during postnatal development of the CNS (44, 45), is a good example. Progesterone could be synthesized by neurons and then metabolized into DHP in the myelin membranes and thus actively participate in myelin formation. Nevertheless, progesterone


synthesis has also been demonstrated in oligodendrocyte progenitors (46). In the PNS, progesterone synthesized by Schwann cells can regulate myelination in vitro (47) and remyelination of the sciatic nerve after cryolesion in vivo (20). Another in vitro study has confirmed the presence of the 3␤HSD enzyme and progesterone synthesis in neurons as well as in astrocytes and oligodendrocytes (48). We thus cannot exclude that glial cells also express 3␤HSD in vivo, perhaps in low amounts that remains undetectable by in situ hybridization, the resolution of which is not high enough. As suggested by Zwain and Yen (48), neurosteroidogenesis as a whole might require a tripartite contribution of the different cell types of the CNS (neurons, oligodendrocytes, and astrocytes). These results are part of a number of studies confirming the concept of neurosteroids, as they demonstrate the presence of the enzyme involved in progesterone synthesis that is part of an enzymatic pathway leading from cholesterol to the formation of progesterone and its 5␣-reduced metabolites. The results demonstrate that both the expression of the 3␤HSD gene and the concentrations of progesterone and pregnenolone in the hippocampus are maximum on P0, a period of intense cerebral maturation, suggesting that this enzyme could be implicated in several important neurotrophic events during postnatal development. More experiments are needed to better understand the developmental changes affecting the regulation of 3␤HSD mRNA expression. Acknowledgments We thank Dr. Hector Coirini for his helpful advice on the quantitative in situ hybridization. Monique Goue´ zou is gratefully acknowledged for her excellent technical assistance. We are grateful to Dr. Krzysztof Rajkowski for critical reading of the manuscript. Received November 25, 2002. Accepted March 10, 2003. Address all correspondence and requests for reprints to: Dr. Chrystelle Ibanez, Institut National de la Sante´ et de la Recherche Me´ dicale, Unite´ 488, Ste´ roı¨des et Syste`me Nerveux, 80 rue du Geńe´ral Leclerc, 94276 Biceˆtre, France. E-mail: [email protected]. This work was supported in part by the European Community (5e`me PCRDT, Key Action 6, Contract QLK6-CT-2000-00179), the Socie´te´ de Secours des Amis des Sciences, the Myelin Project (U.S.), and the Projet Mye´line (France).

References 1. Corpechot C, Robel P, Axelson M, Sjovall J, Baulieu EE 1981 Characterization and measurement of dehydroepiandrosterone sulfate in rat brain. Proc Natl Acad Sci USA 78:4704 – 47017 2. Corpechot C, Synguelakis M, Talha S, Axelson M, Sjovall J, Vihko R, Baulieu EE, Robel P 1983 Pregnenolone and its sulfate ester in the rat brain. Brain Res 270:119 –125 3. Jung-Testas I, Hu ZY, Baulieu EE, Robel P 1989 Neurosteroids: biosynthesis of pregnenolone and progesterone in primary cultures of rat glial cells. Endocrinology 125:2083–2091 4. Akwa Y, Sananes N, Gouezou M, Robel P, Baulieu EE, Le Goascogne C 1993 Astrocytes and neurosteroids: metabolism of pregnenolone and dehydroepiandrosterone. Regulation by cell density. J Cell Biol 121:135–143 5. Guennoun R, Schumacher M, Robert F, Delespierre B, Gouezou M, Eychenne B, Akwa Y, Robel P, Baulieu EE 1997 Neurosteroids: expression of functional 3␤-hydroxysteroid dehydrogenase by rat sensory neurons and Schwann cells. Eur J Neurosci 9:2236 –2247 6. Robert F, Guennoun R, Desarnaud F, Do-Thi A, Benmessahel Y, Baulieu EE, Schumacher M 2001 Synthesis of progesterone in Schwann cells: regulation by sensory neurons. Eur J Neurosci 13:916 –924 7. Majewska MD 1987 Steroids and brain activity. Essential dialogue between body and mind. Biochem Pharmacol 36:3781–3788 8. Majewska MD 1992 Neurosteroids: endogenous bimodal modulators of the


9. 10. 11. 12. 13. 14. 15. 16. 17.

18. 19. 20. 21.

22.

23.

24. 25.

26.

27.

GABAA receptor. Mechanism of action and physiological significance. Prog Neurobiol 38:379 –395 Valera S, Ballivet M, Bertrand D 1992 Progesterone modulates a neuronal nicotinic acetylcholine receptor. Proc Natl Acad Sci USA 89:9949 –9953 Wu FS, Gibbs TT, Farb DH 1991 Pregnenolone sulfate: a positive allosteric modulator at the N-methyl-d-aspartate receptor. Mol Pharmacol 40:333–336 Su TP, London ED, Jaffe JH 1988 Steroid binding at sigma receptors suggests a link between endocrine, nervous, and immune systems. Science 240:219 –221 Monnet FP, Mahe V, Robel P, Baulieu EE 1995 Neurosteroids, via sigma receptors, modulate the [3H]norepinephrine release evoked by N-methyl-daspartate in the rat hippocampus. Proc Natl Acad Sci USA 92:3774 –3778 Vongher JM, Frye CA 1999 Progesterone in conjunction with estradiol has neuroprotective effects in an animal model of neurodegeneration. Pharmacol Biochem Behav 64:777–785 Thomas AJ, Nockels RP, Pan HQ, Shaffrey CI, Chopp M 1999 Progesterone is neuroprotective after acute experimental spinal cord trauma in rats. Spine 24:2134 –2138 Roof RL, Duvdevani R, Heyburn JW, Stein DG 1996 Progesterone rapidly decreases brain edema: treatment delayed up to 24 hours is still effective. Exp Neurol 138:246 –251 Jiang N, Chopp M, Stein D, Feit H 1996 Progesterone is neuroprotective after transient middle cerebral artery occlusion in male rats. Brain Res 735:101–107 Desarnaud F, Do Thi AN, Brown AM, Lemke G, Suter U, Baulieu EE, Schumacher M 1998 Progesterone stimulates the activity of the promoters of peripheral myelin protein-22 and protein zero genes in Schwann cells. J Neurochem 71:1765–1768 Jung-Testas I, Schumacher M, Robel P, Baulieu EE 1996 The neurosteroid progesterone increases the expression of myelin proteins (MBP and CNPase) in rat oligodendrocytes in primary culture. Cell Mol Neurobiol 16:439 – 443 Verdi JM, Campagnoni AT 1990 Translational regulation by steroids. Identification of a steroid modulatory element in the 5⬘-untranslated region of the myelin basic protein messenger RNA. J Biol Chem 265:20314 –20320 Koenig HL, Schumacher M, Ferzaz B, Thi AN, Ressouches A, Guennoun R, Jung-Testas I, Robel P, Akwa Y, Baulieu EE 1995 Progesterone synthesis and myelin formation by Schwann cells. Science 268:1500 –1503 Lorence MC, Naville D, Graham-Lorence SE, Mack SO, Murry BA, Trant JM, Mason JI 1991 3␤-Hydroxysteroid dehydrogenase/⌬5-⌬4-isomerase expression in rat and characterization of the testis isoform. Mol Cell Endocrinol 80:21–31 Zhao HF, Labrie C, Simard J, de Launoit Y, Trudel C, Martel C, Rheaume E, Dupont E, Luu-The V, Pelletier G, Labrie F 1991 Characterization of rat 3␤-hydroxysteroid dehydrogenase/⌬5-⌬4 isomerase cDNAs and differential tissue-specific expression of the corresponding mRNAs in steroidogenic and peripheral tissues. J Biol Chem 266:583–593 Simard J, Couet J, Durocher F, Labrie Y, Sanchez R, Breton N, Turgeon C, Labrie F 1993 Structure and tissue-specific expression of a novel member of the rat 3␤-hydroxysteroid dehydrogenase/⌬5-⌬4 isomerase (3␤-HSD) family. The exclusive 3␤-HSD gene expression in the skin. J Biol Chem 268:19659 – 19668 Zhao HF, Rheaume E, Trudel C, Couet J, Labrie F, Simard J 1990 Structure and sexual dimorphic expression of a liver-specific rat 3␤-hydroxysteroid dehydrogenase/isomerase. Endocrinology 127:3237–3239 Labrie F, Simard J, Luu-The V, Pelletier G, Belanger A, Lachance Y, Zhao HF, Labrie C, Breton N, de Launoit Y, Dupont M, Rheáume E, Martel C, Couet J, Trudel C 1992 Structure and tissue-specific expression of 3␤-hydroxysteroid dehydrogenase/5-ene-4-ene isomerase genes in human and rat classical and peripheral steroidogenic tissues. J Steroid Biochem Mol Biol 41:421– 435 Guennoun R, Fiddes RJ, Gouezou M, Lombes M, Baulieu EE 1995 A key enzyme in the biosynthesis of neurosteroids, 3␤-hydroxysteroid dehydrogenase/⌬5-⌬4-isomerase (3␤-HSD), is expressed in rat brain. Brain Res Mol Brain Res 30:287–300 Kohchi C, Ukena K, Tsutsui K 1998 Age- and region-specific expressions of the messenger RNAs encoding for steroidogenic enzymes p450scc, P450c17 and 3␤-HSD in the postnatal rat brain. Brain Res 801:233–238


28. Parsons B, Rainbow TC, MacLusky NJ, McEwen BS 1982 Progestin receptor levels in rat hypothalamic and limbic nuclei. J Neurosci 2:1446 –1452 29. Bloch B, Popovici T, Le Guellec D, Normand E, Chouham S, Guitteny AF, Bohlen P 1986 In situ hybridization histochemistry for the analysis of gene expression in the endocrine and central nervous system tissues: a 3-year experience. J Neurosci Res 16:183–200 30. Liere P, Akwa Y, Weill-Engerer S, Eychenne B, Pianos A, Robel P, Sjovall J, Schumacher M, Baulieu EE 2000 Validation of an analytical procedure to measure trace amounts of neurosteroids in brain tissue by gas chromatography-mass spectrometry. J Chromatogr B Biomed Sci Appl 739:301–312 31. Furukawa A, Miyatake A, Ohnishi T, Ichikawa Y 1998 Steroidogenic acute regulatory protein (StAR) transcripts constitutively expressed in the adult rat central nervous system: colocalization of StAR, cytochrome P-450SCC (CYP XIA1), and 3␤-hydroxysteroid dehydrogenase in the rat brain. J Neurochem 71:2231–2238 32. Dupont E, Simard J, Luu-The V, Labrie F, Pelletier G 1994 Localization of 3␤-hydroxysteroid dehydrogenase in rat brain as studied by in situ hybridization. Mol Cell Neurosci 5:119 –123 33. Sanne JL, Krueger KE 1995 Expression of cytochrome P450 side-chain cleavage enzyme and 3␤-hydroxysteroid dehydrogenase in the rat central nervous system: a study by polymerase chain reaction and in situ hybridization. J Neurochem 65:528 –536 34. Ukena K, Kohchi C, Tsutsui K 1999 Expression and activity of 3␤-hydroxysteroid dehydrogenase/⌬5-⌬4-isomerase in the rat Purkinje neuron during neonatal life. Endocrinology 140:805– 813 35. Weisz J, Ward IL 1980 Plasma testosterone and progesterone titers of pregnant rats, their male and female fetuses, and neonatal offspring. Endocrinology 106:306 –316 36. Compagnone NA, Bulfone A, Rubenstein JL, Mellon SH 1995 Expression of the steroidogenic enzyme P450scc in the central and peripheral nervous systems during rodent embryogenesis. Endocrinology 136:2689 –2696 37. Ukena K, Usui M, Kohchi C, Tsutsui K 1998 Cytochrome P450 side-chain cleavage enzyme in the cerebellar Purkinje neuron and its neonatal change in rats. Endocrinology 139:137–147 38. Le Goascogne C, Robel P, Gouezou M, Sananes N, Baulieu EE, Waterman M 1987 Neurosteroids: cytochrome P-450scc in rat brain. Science 237:1212–1215 39. Celotti F, Melcangi RC, Martini L 1992 The 5␣-reductase in the brain: molecular aspects and relation to brain function. Front Neuroendocrinol 13:163–215 40. Poletti A, Negri-Cesi P, Rabuffetti M, Colciago A, Celotti F, Martini L 1998 Transient expression of the 5␣-reductase type 2 isozyme in the rat brain in late fetal and early postnatal life. Endocrinology 139:2171–2178 41. Pomata PE, Colman-Lerner AA, Baranao JL, Fiszman ML 2000 In vivo evidences of early neurosteroid synthesis in the developing rat central nervous system and placenta. Brain Res Dev Brain Res 120:83– 86 42. Kellogg CK, Frye CA 1999 Endogenous levels of 5␣-reduced progestins and androgens in fetal vs. adult rat brains. Brain Res Dev Brain Res 115:17–24 43. Poletti A, Celotti F, Rumio C, Rabuffetti M, Martini L 1997 Identification of type 1 5␣-reductase in myelin membranes of male and female rat brain. Mol Cell Endocrinol 129:181–190 44. Notterpek LM, Rome LH 1994 Functional evidence for the role of axolemma in CNS myelination. Neuron 13:473– 485 45. Barres BA, Raff MC 1999 Axonal control of oligodendrocyte development. J Cell Biol 147:1123–1128 46. Gago N, Akwa Y, Sananes N, Guennoun R, Baulieu EE, El-Etr M, Schumacher M 2001 Progesterone and the oligodendroglial lineage: stage-dependent biosynthesis and metabolism. Glia 36:295–308 47. Chan JR, Rodriguez-Waitkus PM, Ng BK, Liang P, Glaser M 2000 Progesterone synthesized by Schwann cells during myelin formation regulates neuronal gene expression. Mol Biol Cell 11:2283–2295 48. Zwain IH, Yen SS 1999 Neurosteroidogenesis in astrocytes, oligodendrocytes, and neurons of cerebral cortex of rat brain. Endocrinology 140:3843–3852