Steroidogenesis in the human brain: trends on sexual dimorphism

Local production and metabolism of steroids in the Central Nervous System: Neurosteroidogenesis, is believed to be a crucial process in normal brain development and function. Increasing number of studies tend to confirm the importance of this process, through the number of physiopathological conditions in which neurosteroids seem to play a key role. Although in recent years isolated efforts have sought to establish the expression of several key steroidogenic enzymes in specific brain structures, to date, no systematic study has been undertaken to understand the intricate pathways of neurosteroidogenesis in the brain. Such efforts have so far been hindered technically by low enzymatic gene expression and hormones production. The difficult access to human encephalic tissue is also a major drawback for this kind of studies. The aim of our work was to quantify by the means of Real-Time quantitative PCR (qPCR) a set of 63 genes of interest corresponding to a broad selection of steroidogenic enzymes, implicated in de novo biosynthesis from cholesterol, so well as key transformation steps for bioactive neurosteroids (Cytochrome P450 family, Aldo-Keto reductase family, hydroxysteroid dehydrogenase family, hormone receptors, GABA receptors...). qPCRs have been performed on RNA extracted from fresh frozen material (116 samples, 9 tissues from 24 autopsies in 7 age groups paired by sex), from non-dement controls obtained from the Netherlands Brain Bank, for a total of more than 20,000 reactions. Human brain tissues harvested for this study included cerebellum, caudate nucleus, medial frontal gyrus, superior frontal gyrus, superior occipital gyrus, superior parietal gyrus, cingulate gyrus, thalamus (pulvinar) and white matter. Overall, we investigated the controversial question of which steroids are directly produced in the human brain and which are produced in peripheral endocrine organs like the adrenal gland and subsequently modified in the brain tissue, we analyzed the expression of key enzymes involved in corticosteroid and sex steroids formation. In contrast to the adrenal gland, that served as positive control, CYP17A1 [conversion of C21 steroids into C19 steroids], SULT2A1 (“DHEA sulfotransferase”), CYP11β1 (11β-hydroxylase) and CYP11β2 (aldosterone synthase) mRNAs expressions were not detected in the human brain (with the exception of the cerebellum). Furthermore, strong mRNA expression of STS (steroid sulfatase) has been confirmed in the areas examined in the study. We therefore conclude that local peripheral transformation, and not de novo synthesis, could be the main source of aldosterone, cortisol and DHEA in the human brain. We reported the first large scale undertaking of human brain cartography which suggested a tissue-specific sexual dimorphism in gene expression of some neurosteroidogenic enzymes, such as P450 aromatase, P450scc (Cholesterol side-chain cleavage enzyme), STS (steroid sulfatase), 3β-HSD [transformation of pregnenolone and DHEA into progesterone and androstenedione respectively], AR (androgen receptor), ESR (estrogen receptor), CYP21A2 [transformation of progesterone and 17-α hydroxyprogesterone to 11-deoxycorticosterone and 11-deoxycortisol], and many GABA receptors. Neuroactive steroids are endogenous neuromodulators. They have potent effects on neurotransmission mediated by γ-aminobutyric acid type A (GABAA) receptors. In this work, statistical analysis revealed significant sex differences of mRNA expression in human thalamus. The expression of STS, 3β-HSD2, CYP19A1, HSD11β1, AR, GR, and GABRA4 was higher in women, while CYP21A2, HSD17β3, HSD11β2, PGR and GABRδ were more expressed in men. We therefore hypothesize that two sex-dependant pathways inhibit the neurotransmission via interaction with the GABAA receptor to modulate the flow of visceral information to the thalamus. Women appear to preferentially modulate GABRA4 through synthesis of DHEA and estrogen, while the formation of TH-PROG, TH-DOC and their precursors was the men tendency with GABRδ modulation. THPROG and THDOC are potent modulators of the GABAA receptor. GABA mediates most of the inhibitory neurotransmission in the mammalian brain. Both THDOC and THPROG have significant sedative effects in vivo. THDOC is a metabolite of the mineralocorticoid DOC and is responsible for the sedative and anti-seizure activity of DOC in animal models. DOC can be metabolised from progesterone, and CYP21A2 mediates this conversion. CYP21A2 mRNA was detected in all samples studied. In the human brain, we found a higher gene expression of CYP21A2 in men than women (with the exception of cerebellum and caudate nucleus). In absence of CYP11B2 which converts DOC in corticosterone, we can conclude that THDOC is the principal DOC metabolite. In men brain, the tendency seems to be the formation of progesterone metabolites which act as potent modulators of GABAR. Recently the Purkinje cell, an important cerebellar neuron, has been identified as a major site for neurosteroid formation in vertebrates. The cerebellum contains more than half the neurons in the brain. Interestingly, gene expression profile of the cerebellum seems to be unusual compared to the other brain specimens analyzed. Indeed, this is the only tissue that expresses genes of de novo synthesis like CYP17A1, SULT2A1 or CYP11B2. Furthermore, we observed the strongest mRNA expression of key genes: 3β-HSD2 and GABRδ (δ-containing GABAR are the most sensitive to modulation by steroids). These data suggest that the cerebellum could have a crucial role in the steroidogenesis in human brain. To conclude, this is the first time, to our knowledge, that a comprehensive quantitative survey of steroidogenic gene transcription in the human CNS has been performed, using a common methodological approach in the same set of individual samples, permitting direct comparison of brain tissue and sex. Our gene expression results demonstrated for the first time a sexual dimorphism on steroid synthesis in the human brain. Neurosteroids can have immediate and sex specific effects on selected neuronal pathways. Our work tends to show that neurosteroidogenesis is an ubiquitous process in the CNS, and is not limited to specific structures. The difference in the enzymatic set in the different regions studied suggests a complex interplay among them. This generalized process is not incompatible with the existence of specialized structures with more predominant roles like the cerebellum.

La produzione locale ed il metabolismo degli steroidi nel Sistema Nervoso Centrale, la Neurosteroidigenesi, si pensa possa giocare un ruolo chiave nello sviluppo e nel corretto funzionamento del cervello. Un numero in crescita di studi sembra confermare l’importanza di questo processo svelando diverse condizioni fisio-patologiche dove i neurosteroidi sembrano un punto cardine. Sebbene in tempi recenti ci siano stati degli sforzi isolati di confermare diversi enzimi chiave della steroidogenesi in specifiche strutture cerebrali, ad oggi nessuno studio sistematico è stato effettuato per comprendere le complesse pathways della steroidogenesi cerebrale. La difficoltà di questi studi è stata resa maggiore sino ad ora dalla bassa espressione genica degli enzimi e degli ormoni coinvolti, e dalla difficoltà nel reperire tessuto umano encefalico. Lo scopo della nostra ricerca è stato di quantificare tramite la Real-Time quantitative PCR (qPCR) un set di 63 geni di specifico interesse nella steroidogenesi, dalla biosintesi de-novo a partire dal colesterolo, ai passaggi metabolici chiave per formare i neurosteroidi bioattivi (Cytochrome P450 family, Aldo-Keto reductase family, hydroxysteroid dehydrogenase family, hormone receptors, GABA receptors...). Le PCR quantitative sono state eseguite su RNA estratto dal materiale umano congelato (116 campioni, 9 tessuti da 24 autopsie in 7 gruppi d’età ed accoppiati per i diversi sessi) da controlli “non dementi” ottenuti dalla Netherlands Brain Bank, per un totale oltre le 20000 reazioni chimiche. Il tessuto umano raccolto per lo studio includeva: cervelletto, nucleo caudato, giro frontale mediale, giro frontale superiore, giro superiore occipitale, giro superiore parietale, giro cingolato, talamo (pulvinar) e materia bianca. Globalmente parlando, abbiamo esaminato la domanda controversa di quali steroidi sono prodotti direttamente nel cervello umano e quali siano prodotti negli organi periferici endocrini, come il surrene, e poi successivamente modificati nel tessuto cerebrale. Per questo fine abbiamo analizzato l’espressione di enzimi chiave coinvolti nella formazione di corticosteroidi e degli ormoni sessuali. In maniera opposta al controllo positivo surrenalico, le espressioni degli mRNA CYP17A1 (converte gli steroidi C21 in C19), SULT2A1 (DHEA sulfotransferasi), CYP11β1 (11β-idrossilasi) e CYP11β2 (aldosterone sintetasi) non sono stati trovati nel cervello umano (con l’eccezione del cervelletto). In aggiunta grossi livelli di espressione sono stati riscontrati per STS (sulfatasi steroidea) enzima chiave per attivare i solfati steroidei non biologicamente attivi nelle aree esaminate nello studio. Possiamo quindi ipotizzare che la trasformazione periferica e non la sintesi de-novo siano la fonte primaria di aldosterone, cortisolo e DHEA nel cervello umano. Riportiamo inoltre la prima cartografia su larga scala della steroidogenesi steroidea cerebrale che sembra suggerire un dimorfismo sessuale tessuto-specifico negli enzimi neurosteroidogenetici come l’aromatasi, il P450scc (Cholesterol side-chain cleavage enzyme), STS (steroid sulfatase), 3β-HSD [trasformazione del pregnenolone e DHEA in progesterone ed androstenedione rispettivamente], AR (androgen receptor), ESR (estrogen receptor), CYP21A2 (trasformazione del progesterone e 17-α idrossiprogesterone in 11-deossicorticosterone and 11-deossicortisolo), e molti recettori GABA. Concludendo, questa è la prima volta che una ricerca quantitativa e comprensiva sulla trascrizione genetica della steroidogenesi nel cervello umano sia stata portata a termine, usando un approccio metodologico di largo uso, negli stessi “set” di campioni individuali, rendendo così possibile un confronto diretto tra tessuto cerebrale nei due sessi. I risultati della nostra espressione genica hanno mostrato per la prima volta un dimorfismo sessuale nella sintesi steroidea nel cervello umano. I neurosteroidi posso quindi avere effetti immediati e sesso-specifici su alcune pathways neuronali. Il nostro lavoro sembra indicare che la neurosteroidogenesi sia un processo ubiquitario nel sistema nervoso centrale e non limitato a strutture specifiche. La differenza riscontrata nei set enzimatici nelle diverse regioni cerebrali indica un’interazione molto complessa tra di esse. Questo processo generalizzato è differente in strutture specifiche con ruoli importanti come il cervelletto.