|Home | About | Journals | Submit | Contact Us | Français|
The origins of our understanding of the cellular and molecular mechanisms by which signaling pathways and downstream transcription factors coordinate the specification of adrenocortical cells within the adrenal gland have arisen from studies on the role of Sf1 in steroidogenesis and adrenal development initiated 20 years ago in the laboratory of Dr. Keith Parker.
Adrenocortical stem/progenitor cells have been predicted to be undifferentiated and quiescent cells that remain at the periphery of the cortex until needed to replenish the organ, at which time they undergo proliferation and terminal differentiation. Identification of these stem/progenitor cells has only recently been explored. Recent efforts have examined signaling molecules, including Wnt, Shh, and Dax1, which may coordinate intricate lineage and signaling relationships between the adrenal capsule (stem cell niche) and underlying cortex (progenitor cell pool) to maintain organ homeostasis in the adrenal gland.
Dr. Keith Parker serendipitously became a pioneer in the molecular study of adrenal gland development and organogenesis when he cloned the gene for steroidogenic factor 1 (Sf1) as a transcriptional regulator of genes encoding steroidogenic enzymes (Luo et al., 1994). Dr. Parker’s work occurred concurrently with studies emerging from the Morohashi laboratory on the steroidogenic transcription factor Ad4BP, which turned out to be identical to Sf1 (Hatano et al., 1996). Together, these studies surprisingly found that the adrenal glands did not form in mice with a genetically modified null allele of Sf1. Studies of Sf1 for over two decades have provided the molecular framework for the emerging field of adrenocortical stem and progenitor biology. The current research in the emerging field of adrenal organogenesis and data supporting the presence of adrenocortical stem/progenitor cells was recently reviewed (Kim et al., 2009). The current summary herein highlights a working model of adrenocortical stem/progenitor cell biology and the proposed stem cell niche and includes the emerging data supporting the hypothesis.
The adrenal glands are formed from two embryologically distinct tissues. The inner adrenal medulla is derived from neural crest cells of the neuroectoderm lineage and synthesizes catecholamines essential for the “fight-or-flight” response. The cortex is derived from cells of the intermediate mesoderm and is responsible for secretion of steroid hormones. The cortex is organized into three concentric zones, each with a discrete function, the synthesis and secretion of steroid hormones: 1) zona glomerulosa, mineralocorticoids; 2) zona fasciculata, glucocorticoids, and 3) zona reticularis, sex-steroid precursors.
Adrenal gland organogenesis is orchestrated in discrete histological phases [Figure 1; reviewed in (Else and Hammer, 2005; Keegan and Hammer, 2002; Uotila, 1940)]. In the initial phase (4th week of gestation in humans, embryonic day (E) 9.0 in mice), the adrenogonadal primordium (AGP) is first distinguished and expresses the essential transcription factor SF1/Sf1 (Hatano et al., 1996; Luo et al., 1994). In the second phase (8th week gestation in humans, E10.5 in mice), the AGP separates into two distinct structures, the adrenal and the gonadal primordia. This phase is followed by migration of neural crest cells through the fetal cortex to establish the medulla and formation of a mesenchymal capsule around the fetal cortex (8th–9th week of gestation in humans, E11.5–E12.5 in mice). As encapsulation progresses, formation of the adult cortex (or definitive zone) is initiated. The human fetal zone histologically regresses at birth while the mouse fetal zone (X zone) regresses during puberty in males or at the time of first pregnancy in females.
Sf1 expression is critical for proper adrenal organogenesis and is required for steroidogenic function in both the fetal and adult cortex. As addressed by work of Zubair et al. (2006), the differential activity and regulation of Sf1 in the development and function of these two cell populations (fetal and adult cortex), is emerging as an essential mediator of adrenal gland organogenesis. These investigators identified a Fetal Adrenal Enhancer (FAdE) that directs Sf1 expression solely in the fetal cortex. During fetal adrenal development, a transcription complex containing the homeobox protein PKNOX1 (Prep1), homeobox gene 9b (Hox) and pre B-cell leukemia transcription factor 1 (Pbx1) are recruited to the FAdE upon separation of the adrenal primordium from the AGP. This complex initiates fetal zone expression of Sf1, which is later maintained through autoregulation by Sf1 itself (Zubair et al., 2006). Further studies of Sf1 regulation by FAdE suggest that control of Sf1 expression through the FAdE is abrogated at E14.5, at which time the fetal cortex begins to regress (Zubair et al., 2008). Through the use of lineage tracing studies in mouse, in which cells expressing Sf1 driven by the FAdE are followed throughout development, Zubair, et al. concluded that the adult cortex arises from cells of the fetal cortex (Figure 2B). These studies propose a switch in the transcriptional regulation of Sf1 in which the FAdE is inhibited and a presumed, yet-to-be identified, Definitive Adrenal Enhancer (DAdE) is activated to perpetuate the expression of Sf1 specifically in the adult cortex. The work from these studies allows important questions to be raised: 1) how is the FAdE inhibited at the transition to definitive cortex? 2) where is the DAdE sequence and how is it activated? 3) what are the molecular, cellular and organ-specific mechanisms that coordinate the overall fetal to adult transition? Emerging studies will address these questions to ascertain if fetal adrenal cells are indeed adrenocortical precursor cells of the definitive cortex.
The adult adrenal gland maintains organ homeostasis by replenishment of adrenocortical cells throughout life. Since the 1950s, numerous studies have provided evidence that cells from the capsule/subcapsular region of the adrenal gland are centripetally displaced inward to repopulate the adrenal cortex [reviewed in (Kim et al., 2009)]. These studies have been primarily conducted using histology and markers of proliferation. Only within the last year has genetic data emerged to support this hypothesis. Three laboratories have provided evidence that the sonic hedgehog (Shh) signaling pathway is essential for adrenal gland development and maintenance (Ching and Vilain, 2009; Huang et al., 2010; King et al., 2009). The hedgehog pathway has been found to be involved in the development of a number of vertebral organ systems and the regulation of both embryonic stem cells (ES cells) and adult tissue stem cells (Han et al., 2008; Ingham and McMahon, 2001; Xie and Abbruzzese, 2003). In this pathway, a hedgehog ligand (Shh in the adrenal) binds to the cell surface receptor Patched-1, which releases inhibition of the Smoothened receptor and allows downstream activation of Gli transcription factors.
King, et al. (2009) showed that Shh can be detected in the adrenal gland at E11.5 and is expressed primarily in the subcapsular region of the adrenal cortex. Shh expression colocalizes with Sf1 in cortical cells of the subcapsular region but not in cells throughout the remainder of the cortex, which express both Sf1 and markers of fully differentiated steroidogenic cells (i.e. Cyp11b1, Cyp11b2). Using mice in which Shh is ablated specifically in Sf1-expressing cells, experiments revealed marked adrenal hypoplasia, decreased proliferation, and a depleted capsule. Huang et al. (2010) observed that despite a decreased adrenal size in the tissue-specific Shh knockout mice, the adrenal glands maintain proper zonation and, thus, proposed that Shh does not have a role in the initiation of differentiation. Together, these data implicate the Shh pathway in proliferation and maintenance of the adrenal cortex. Lineage tracing studies of Shh expressing cells reveal that descendents of these cells express adrenocortical differentiation markers [Figure 2C, (King et al., 2009)]. While these studies provide evidence that Shh-positive, Sf1-positive cells may serve as progenitor cells for the adrenal cortex, the data do not address an origin of these progenitor cells. Evidence that the adrenal capsule could be the adrenocortical stem/progenitor cell niche was provided by studies of the downstream activator of the hedgehog pathway, Gli1 (Ching and Vilain, 2009; Huang et al., 2010; King et al., 2009). Gli1-expressing cells of the adrenal gland, specifically in the adrenal capsule, do not express Sf1. Lineage tracing experiments revealed that this subpopulation of cells is capable of giving rise to Sf1-expressing, differentiated adrenocortical cells (Figure 2C).
Together, these studies support two different hypotheses as to the cell population serving the function of adrenal stem/progenitor cells that at first analysis appear disparate: 1) Gli1-positive, Sf1-negative cells give rise to Sf1-positive, differentiated cells and 2) Shh-positive, Sf1-positive cells give rise to Sf1-positive, differentiated cells. However, one could hypothesize that Gli1-positive, Sf1-negative cells (stem cells) give rise to Shh-positive, Sf1-positive cells (transit amplifying progenitor cells) which in turn give rise to Shh-negative, Sf1-positive differentiated steroidogenic cells. If this two part hypothesis holds, it will be important to determine the exact mechanism of Shh signaling to the Gli1-positive cells and what endocrine or paracrine cues might regulate the Shh signal. It is logical to predict that the Shh-positive cells signal to the Gli1-positive cells to induce asymmetric division of stem cells to replenish progenitor cells as they are depleted during cortical maintenance through centripetal displacement. While the presence of stem and progenitor cells in the adrenal capsule/subcapsular region are being determined through the use of genetic lineage tracing studies in mice, further studies are necessary to definitively identify and characterize these cells during normal adrenal homeostasis. Specifically, it will be critical to determine if both the capsule and the underlying cortex (or subcapsular region) contain long-term retention cells and the functional significance of these subpopulations of stem/progenitor cells.
It is reasonable to consider a model that unifies the two experimental observations that cells of the Sf1-positive fetal zone and cells of the Sf1-negative capsule both give rise to Sf1-positive adult adrenocortical cells (Figure 2B,C). Zubair, et al. (2006, 2008) propose that a switch occurs in the regulation of Sf1 expression in which cells of the definitive zone arise from cells in which the FAdE is transiently active. However, these studies do not provide data suggesting that cells expressing Sf1 under control of the FAdE contribute to continued adult homeostasis. It is possible that a Sf1-negative cell serves as an intermediate during the transition in Sf1 transcriptional regulation from a cell in which the FAdE directs Sf1 expression (fetal adrenal cell) to a cell in which the DAdE directs Sf1 expression (adult adrenal cell). This Sf1-negative transition cell is predicted to reside within the adrenal capsule which lacks Sf1 expression. Indeed, the Shh and Gli1 studies have shown that cells within the adrenal capsule are capable of giving rise to cells of the adrenal cortex [Figure 2C, (Ching and Vilain, 2009; Huang et al., 2010; King et al., 2009)]. However, it remains to be determined if Shh-responsive, Gli1-positive, Sf1-negative capsular cells are derived from fetal adrenal cells (Figure 2D). It is predicted that endocrine and paracrine signals facilitate reactivation of Sf1 transcription in a Gli1-positive, Sf1-negative cell as it transitions to a subcapsular, Sf1-positive (adult) adrenocortical cell. Further studies are required to support or refute this unifying model and to explain the present data supporting two sources of adrenocortical stem/progenitor cells.
As anticipated, a multitude of signaling pathways and transcription factors are emerging as critical mediators of adrenocortical growth and differentiation, including the regulation of adrenocortical stem/progenitor cells. The transcriptional mediators, β-catenin and Dax1 have drawn particular interest due to their role in human disease and their demonstrated regulation of Sf1 activity.
The Wnt/β-catenin signaling pathway is critical for embryonic development, stem cell maintenance, and differentiation in many tissues (Blanpain et al., 2007; Logan and Nusse, 2004). In humans, β-catenin dysregulation has been observed in a subset of sporadic adrenocortical adenomas and carcinomas (Giordano et al., 2009; Giordano et al., 2003; Tissier et al., 2005). Characterization of β-catenin signaling in the mouse adrenal revealed that β-catenin expression and active signaling becomes distinctly localized to the subcapsular region of the newly formed adult adrenal cortex after the capsule has coalesced around the fetal zone (Kim et al., 2008). Involvement of β-catenin signaling in adrenocortical progenitor cells became evident in studies employing cre-lox technology where β-catenin was ablated specifically in Sf1-expressing cells of the adrenal cortex using a mouse harboring floxed Ctnnb1 alleles. In mice expressing a high level of the Sf1-cre transgene, near complete excision of Ctnnb1 alleles was achieved. Normal adrenal development was detected until E12.5 when adrenal failure became evident precisely when the definitive/adult cortex was emerging between the coalescing capsule and the fetal zone. Intriguingly, in studies using a low expressing Sf1-cre transgene where Ctnnb1 excision was incomplete and approximately half of Sf1 expressing cells continued to express β-catenin, adrenal development progressed normally. However, as the mice aged (i.e. 30 weeks) the adrenal glands displayed a progressive cortical thinning and a decreased steroidogenic capacity (Kim et al., 2008). This depletion of the cortex is hypothesized to be due to the loss of adrenocortical progenitor cells, consistent with multiple lines of data indicating a necessary role of Wnt/β-catenin signaling in the maintenance of progenitor cells in multiple systems. These data support the notion that overactivation of this pathway leads to the dysregulation of Wnt/β-catenin signaling observed in adrenocortical carcinomas (Assie et al., 2010; Berthon et al., 2010; Giordano et al., 2009; Giordano et al., 2003; Tissier et al., 2005).
DAX1 was cloned as the gene responsible for X-linked Cytomegalic Adrenal Hypoplasia Congenita. DAX1-deficient patients classically exhibit histologic adrenal hypoplasia and resultant adrenal insufficiency (Phelan and McCabe, 2001; Zanaria et al., 1994). While most patients historically present early in life with adrenal crisis, recent screening programs have revealed that the presentation of adrenal insufficiency is indeed a stochastic event even in families with two or more afflicted siblings harboring identical mutations (Peter et al., 1998). Dax1 plays an essential role in the inhibition of Sf1-mediated steroidogenesis (Lalli and Sassone-Corsi, 2003) and is specifically enriched in the subcapsular region of the adrenal cortex, similar to Shh. The combined clinical and molecular data presented above support the hypothesis that loss of adrenal function is due to a depletion of an adrenocortical progenitor reserve (Lin et al., 2006; Muscatelli et al., 1994; Niakan et al., 2006).
The regulation of Dax1 expression is predicted to be a dynamic process of activation by signaling pathways involved in progenitor biology and inhibition by pathways involved in adrenocortical differentiation and steroidogenesis. Such a dual model of regulation of a progenitor population combines paracrine and endocrine signals to provide an integrated homeostatic mechanism of organ maintenance. Indeed, Dax1 transcription in the adrenal gland is activated by Sf1 in cooperation with other transcription factors/co-activators. Specifically, paracrine Wnt-signaling directly activates Dax1 expression through a β-catenin/Sf1 complex in the developing ovary (Mizusaki et al., 2003). This is compelling as only the subcapsular adrenal cortex exhibits active canonical Wnt signaling and would be presumed to signal in a paracrine fashion to support Dax1 expression (Kim et al., 2008). Similarly, glucocorticoids synthesized in the differentiated adult cortex activate Dax1 transcription through a glucocorticoid receptor/Sf1 complex (Gummow et al., 2003). Conversely, ACTH has been shown to completely remove Sf1 complexes from the Dax1 promoter, effectively shutting off Dax1 transcription. It is predicted that without Dax1, the Sf1-positive progenitor cells can respond to ACTH to appropriately initiate steroidogenesis (Gummow et al., 2006). In support of the role of Dax1 in adrenocortical progenitor cells, studies in mouse embryonic stem cells suggest that Dax1 helps maintain pluripotency and knockdown of Dax1 in mouse embryonic stem cells induces differentiation (Kelly et al., 2010; Khalfallah et al., 2009; Niakan et al., 2006). Together, these data point to Dax1 as a critical mediator of both mouse ES cell and adrenocortical progenitor maintenance/differentiation but further studies are required to fully understand how Dax1 contributes to these processes.
β-catenin and Dax1 are just two molecules emerging as critical regulators of adrenocortical progenitor cells. As regulation of adrenal stem/progenitor cells likely require exquisite orchestration, current studies are shedding light on a number of other pathways that contribute to the regulation of both maintenance and differentiation of these cells (summarized in Table 1). IGFII, Notch, Pod1, and telomerase, among others, are currently the focus of study. The contribution of these pathways to adrenocortical homeostasis has been reviewed recently (Kim et al., 2009). Additionally, studies into post-translational modifications of Sf1, such as phosphorylation (Campbell et al., 2008; Hammer et al., 1999; Lewis et al., 2008; Sewer and Waterman, 2002; Winnay and Hammer, 2006) and SUMOylation (Chen et al., 2004; Komatsu et al., 2004; Lee et al., 2005; Yang et al., 2009), may provide essential clues into the regulation of adrenal stem/progenitor cells through investigation of this critical mediator of adrenocortical specification. As new data become available, the mechanisms by which this network of signaling cascades and transcriptional regulators control adrenocortical stem/progenitor cell maintenance and differentiation should allow for a more complete understanding of adrenocortical physiology and disease.
Since the inception of Keith Parker’s groundbreaking work, much progress has been made to define the development and the signaling pathways involved in the maintenance and differentiation of the adrenal gland. In this review, data have been highlighted which support two different models for the source of adrenal stem/progenitor cells: the fetal adrenal and the adrenal capsule. Further studies are required to support or refute a unifying model whereby proposed transient cells arise from the fetal adrenal and give rise directly or indirectly to the proposed stem/progenitor cells within the adrenal capsule. Further investigation into the signaling and transcriptional regulators of stem/progenitor cell maintenance and differentiation will provide essential clues into our understanding of adrenal cortex replenishment. In turn, the understanding of adrenal stem/progenitor cells will contribute to our knowledge and will greatly enhance our ability to treat adrenal diseases that are caused by disruption of the pathways which control adrenocortical maintenance and function. The authors dedicate this review from the XIV Adrenal Cortex Conference Keith L. Parker Memorial Symposium to Keith’s memory.
The authors thank Joanne H. Heaton for critical review and editing of this manuscript. This work was supported by National Institutes of Health (NIH) Grant DK062027 from National Institute of Diabetes and Digestive and Kidney Diseases (to G.D.H.) and Grant CA134606 from the National Cancer Institute (to G.D.H.); M.A.W was supported, by NIH Grant T32 DK07245 from the Training Program in Endocrinology at the University of Michigan.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.