> Departments > Teams > Ghyselinck/Mark Group

[Research Interest]

[Recent results]

[Projects]

[References]

Research Interest

Vitamin A (retinol) and its active derivatives (collectively referred to as retinoids) exert a wide variety of effects on vertebrate embryonic body shaping and organogenesis, tissue homeostasis, cell proliferation, differentiation and apoptosis. The natural active metabolite of vitamin A, retinoic acid (RA), exerts its pleiotropic effect through activating two classes of nuclear receptors, RAR (α, β and γ isotypes) and RXR (α, β and γ isotypes) that function, in the form of RAR/RXR heterodimers, as ligand-dependent transcriptional regulators by binding to regulatory regions located in target genes. The key way for the organism to control both the sites and the timing of vitamin A actions lies not only at the RAR and RXR levels but also in the modulation of the cellular machinery that leads to the formation of RA from its hormonally inactive precursor, retinol. This machinery comprises enzymes that catalyze storage of vitamin A or synthesis and degradation of RA, as well as cellular binding proteins for retinoids. Storage of vitamin A is essentially dependent upon lecithin:retinol acyl transferase (LRAT) and cellular retinol binding proteins (CRBP1 to CRBP3). Synthesis of RA is catalysed by four isotypes of retinaldehyde dehydrogenases (RALDH1 to RALDH4), while degradation of RA is catalyzed by three specific isotypes of cytochrome P450 hydroxylases (CYP26A1, CYP26B1 and CYP26C1), each displaying a specific pattern of expression, both during development and in adult tissues. In addition, a specific carrier (RBP) transports vitamin A in the blood to ensure proper distribution to peripheral tissues. Our goal is to understand, through combinations of genetic and pharmacological approaches, the respective contributions of these actors to the RA-signalling pathway (Figure 1).

Figure 1. Overview of vitamin A metabolism and of retinoic acid signal transduction in cells. Retinol (vitamin A) is transported in blood bound to its carrier protein RBP. Vitamin A is internalized by the RBP membrane receptor STRA6. Inside cell, vitamin A is either stored in the form of retinyl esters or oxidized into retinoic acid through a two-step process. Retinoic acid is then either hydroxylated into inactive forms or transferred into the nucleus, where it binds to its nuclear receptors. The latter control the expression of target genes where they bound to regulatory sequences called RARE. Legend: ADH1, 2 and 7, alcohol dehydrogenases types 1, 2 and 7, respectively; CRABP1 and CRABP2, cellular retinoic acid binding protein types 1 and 2, respectively; CRBP1, cellular retinol binding protein type 1; CYP26, P450 cytochrome hydroxylases A1, B1 and C1 isotypes; Esters, retinyl esters; LRAT, lecithin retinol acyl transferase; OH-RA, hydroxylated retinoic acid; Ral, retinaldehyde; RALDH1-4, retinaldehyde dehydrogenases types 1 to 4; RA, retinoic acid; RAR, retinoic acid receptor (α, β and γ isotypes); RBP, retinol binding protein; REH, retinylester hydrolase; RoDH1, 2 and 4, retinol dehydrogenases types 1, 2 and 4, respectively; Rol, retinol (vitamin A); RXR, rexinoid receptors (α, β and γ isotypes); RARE, retinoic acid response element; SDR1, short chain reductase type 1; STRA6, membrane receptor for RBP-Rol complexes. The putative ligand activating RXR is schematized using a green square (see Publication 1).

The systematic gene knockout approach in vivo that we have performed over the last fifteen years has demonstrated that (i) RAR/RXRα heterodimers, in which transcriptional activation function of both partners is often required to mediate the physiological effects of RA, are involved at multiple stages of early morphogenesis and organogenesis; (ii) in many instances functional redundancy in single null mutants allows, possibly artefactually, the remaining RAR or RXR to compensate for the missing one (Publication 1). However, that compound null mutants die at birth or earlier during embryonic development has precluded analysis of RAR functions at late developmental stages and in adult tissues. Additionally, due to the pleiotropic effects of RA, the germline mutations as yet generated did not allow to distinguish cell-autonomous from non-cell autonomous functions of RA. Deciphering the precise functions of RAR during development and adulthood requires therefore spatially and temporally-controlled somatic mutations. To this end, we make use of the conditional mutagenesis approach (Cre/loxP; Figure 2) that permit to inactivate genes, singly or in combination, at any given time, and in a given cell type (Publication 2). This innovative method, combined with biochemical, morphological, and pharmacological approaches enables us to characterize the circuits controlled by retinoids under basal conditions, and to define how alterations in these signalling pathways contribute to diseases. During the last years, we have been studying the role of retinoids during embryogenesis and in adult tissues, notably in epidermis and in testis. Now, we focus our primary attention on the functions exerted by retinoids in stem cell self-renewal, using the seminiferous epithelium as a model.

Recent Results

Role of retinoic acid signal transduction in male seminiferous epithelium

The seminiferous epithelium is a major target of vitamin A and represents a remarkable paradigm for cellular biologists to investigate the pleiotropic actions of RA in vivo, as it integrates the problematic of stem cell renewal, cell proliferation, switch from mitotic to meiotic cell division, and programmed cell death. The seminiferous epithelium of the testis, surrounded by peritubular myoid cells, is composed of germ cells and of somatic, supporting, cells called Sertoli cells. This epithelium undergoes an 8.5 days cycle divided into 12 stages (I to XII). Each stage, corresponding to a precise association of germ cells, is most likely related to a specific functional state of the epithelium, which is regulated by complex signalling networks involving endocrine, paracrine and autocrine factors, amongst which vitamin A. Acting as the operational pivot of spermatogenesis, Sertoli cells cyclically express numerous genes, thereby providing specialized micro-environments, or niche(s). These niches direct the fate of spermatogonia stem cells either towards self-renewal or, alternatively, arrest of mitotic division and entry into meiosis. Outside their niches, spermatogonia can become multipotent and therefore represent promising tools for cellular therapy, if the risk of transformation associated with acquisition of multipotency is diverted. Our studies, which make use of a combination of innovative genetic, pharmacological and molecular approaches, are aimed at deciphering the molecular mechanisms that underlie the capabilities of RA to promote spermatogonia differentiation and, beyond, differentiation of normal and cancerous stem cells, in vivo.

In 2004, we have shown that mice expressing RXRβ impaired in its transcriptional activation function AF2 (Rxrb^af2o mutation) do not display the spermiation defects observed in Rxrb-null mutants, indicating that the role played by RXRβ in spermatid release is ligand-independent. We have also provided genetic and molecular evidences that cholesterol homeostasis in Sertoli cells depends upon the coactivator TIF2 and RXRβ/LXRβ heterodimers, in which RXRβ AF2 is transcriptionally active, to control expression of ABCA1 transporter mediating cholesterol efflux from Sertoli cells (Figure 6). Most importantly, our results indicate that RXRβ may be activated by a ligand distinct from 9-cis retinoic acid (Publication 3). On the other hand, we have also demonstrated, through analyses of compound mutant mice lacking two members of the coactivator family, that SRC1 can partially compensate for the loss of TIF2 in Sertoli cells. Interestingly, the abnormal features displayed by all these mutants, including spermatid maturation defects, increase in Sertoli cell lipid stores, loss of immature germ cells, and formation of giant multinucleated spermatids, are commonly detected in testes of elderly men, suggesting that deficiencies in molecular pathways involving TIF2, SRC1, RXRβ and/or LXRβ in Sertoli cells could participate in testicular senescence (Publication 4).

Later, as a first step before investigating the role of RA in mouse testis by genetic means, we have analyzed in details the effects of vitamin A deficiency (VAD) on spermatogenesis (Publication 5), as well as the distribution patterns of the different actors of the machinery that controls RA availability, and of all isotypes of RAR and RXR transducing the RA signal (Publication 6). Our data indicate that (i) CYP26 enzymes may generate, in peritubular myoid cells, a catabolic barrier that prevents circulating RA and RA synthesized by Leydig cells to enter the seminiferous epithelium; (ii) compartmentalization of RA synthesis within this epithelium may modulate, through paracrine mechanisms, the coupling between spermatogonia proliferation and their differentiation; (iii) spermatogonia proliferation may involve, independently of RXR, two distinct RAR-mediated signalling pathways, the first one operating through RARα in Sertoli cells (Figure 7) and the second one operating through RARγ in spermatogonia. Our data also suggest that the involvement of RA in testis development starts when primary spermatogonia first appear (Publication 6). We have also established precisely, using Rbp4-null mice as models, the kinetics of the spermatogenetic alterations occurring during the course of VAD. We have shown that the VAD-induced testicular degeneration arises through the normal maturation of germ cells in a context of spermatogonia differentiation arrest. Thus, RA appears dispensable for the transition of premeiotic to meiotic spermatocytes, for meiosis and for spermiogenesis. In contrast, RA plays critical roles in controlling spermatogonia differentiation, spermatid adhesion to Sertoli cells and spermiation. In addition, our data suggest, as expected from the expression patterns (Publication 6), that the VAD-induced arrest of spermatogonia differentiation results from the simultaneous block in RA-dependent events mediated by RARγ in spermatogonia and by RARα in Sertoli cells (Publication 5).

It was known for a decade that a germline null mutation of Rara results in a testicular degeneration comprising some, but not all, abnormal features observed upon dietary VAD, whereas a germline inactivation of either Rarb or Rarg do not cause primary testicular defects. We have started our somatic mutagenesis program by inactivating Rara in Sertoli cells (Publication 7). Our results indicate that a selective ablation of the RARα gene in these cells (Rara^Ser-/- mutation) yields testis defects identical to those generated upon inactivation of RARα in the whole organism, indicating that all RARα functions in reproduction are Sertoli cell-autonomous. The absence of RARα abolishes the normal cyclical expression of numerous proteins in Sertoli cells. It also induces testis degeneration features, as well as a delay in Stra8 expression, two hallmarks of RA deficiency. Thus, our data demonstrate that RARα is a master regulator of the Sertoli cell cycle and plays important paracrine functions in spermatogonia differentiation and in survival of germ cells. However, ablation of all three RAR in Sertoli cells does not induce an arrest in spermatogonia differentiation and does not abolish Stra8 expression, as does VAD. These phenotypic differences point to a cooperation between distinct RA dependent pathways mediated by RARα in Sertoli cells and by RARγ in spermatogonia during spermatogenesis, as expected from our previous studies (Publications 5 and publication 6). Additionally and most importantly, ablation of all three RXR (α, β and γ isotypes) in Sertoli cells does not recapitulate the phenotype induced by ablation of RARα (Figure 8), providing thereby the first evidence that RARα exerts functions in vivo independently of its heterodimerization with a RXR (Publication 7). On the other hand we have recently found that mice lacking RXRβ only in Sertoli cells (i.e., Rxrb^Ser-/- mutants) display the same defects as those carrying a germline mutation of RXRβ, indicating that RXRβ acts cell autonomously on cholesterol homeostasis in Sertoli cells. As Sertoli cells lacking all 3 RXR display the same abnormalities as those lacking only RXRβ. (Publication 7), our study shows thereby that RXRβ is the sole RXR heterodimerizing with LXRβ in these cells (Manuscript in preparation).

Altogether, our initial studies on the role of RA signal transduction in male seminiferous epithelium therefore provide evidence that (i) the arrest in spermatogonia differentiation induced by VAD results from impairments in 2 distinct RA-dependent events mediated by RARγ in spermatogonia and by RARα in Sertoli cells, both events being paracrine in nature; (ii) neither event involves the canonical functional units of the RA signalling pathway, namely RAR/RXR heterodimers; and (iii) spermatogonia are “protected” from circulating RA by peritubular myoid cells which, through CYP26 expression, form a catabolic barrier.

1. Mark M, Ghyselinck NB, Chambon P. (2006). Functions of retinoid nuclear receptors: lessons from genetic and pharmacological dissections of the retinoic acid signalling pathway during mouse embryogenesis. Annu. Rev. Pharmacol. Toxicol. 46:451-480.

2. Schnutgen F, Ghyselinck NB. (2007). Adopting the good reFLEXes when generating conditional alterations in the mouse genome. Transgenic Res. 16:405-413.

3. Mascrez B, Ghyselinck NB, Watanabe M, Annicotte JS, Chambon P, Auwerx J, Mark M. (2004). Ligand-dependent contribution of RXRβ to cholesterol homeostasis in Sertoli cells. EMBO Rep. 5:285-290.

4. Mark M, Yoshida-Komiya H, Gehin M, Liao L, Tsai MJ, O'Malley BW, Chambon P, Xu J. (2004). Partially redundant functions of SRC-1 and TIF2 in postnatal survival and male reproduction. Proc. Natl. Acad. Sci. USA 101:4453-4458.

5. Ghyselinck NB, Vernet N, Dennefeld C, Giese N, Nau H, Chambon P, Viville S, Mark M. (2006). Retinoids and spermatogenesis: lessons from mutant mice lacking the plasma Retinol Binding Protein. Dev. Dyn. 235:1608-1622.

6. Vernet N, Dennefeld C, Rochette-Egly C, Oulad-Abdelghani M, Chambon P, Ghyselinck NB, Mark M. (2006). Retinoic acid metabolism and signalling pathways in the adult and developing mouse testis. Endocrinology 147:96-110.

7. Vernet N, Dennefeld C, Guillou F, Chambon P, Ghyselinck NB, Mark M. (2006). Prepubertal testis development relies on retinoic acid but not rexinoid receptors in Sertoli cells. EMBO J. 25:5816-5825.

webmaster@igbmc.u-strasbg.fr

© IGBMC 13/01/2010