From a receptor pharmacology standpoint, the mechanism of progesterone action implicates the classical PR (e.g., PR-B or its N-terminally truncated variant, PR-A). Indeed, there are neuroprotective mechanisms of progesterone that require the classical PR. For example, our laboratory has determined that the ability of progesterone to increase the expression (mRNA and protein levels) of brain-derived neurotrophic factor (BDNF), a key mediator of progesterone's protective effects, requires the classical PR (Figure ) (Jodhka et al., 2009
). Further, Cai and colleagues (2008
) have implicated the classical/intracellular PR in the protective effects of progesterone against an experimental model (middle cerebral artery occlusion) of stroke. However, evidence also exists for alternative mechanisms of action, including that which involves integral membrane PRs. For example, the effect of progesterone has been reported in the brain of PR knock-out mice (Krebs et al., 2000
), suggesting PRs other than the classical PR may mediate the effect of progesterone in the CNS. In fact, several lines of evidence now support the role of cell membrane-associated PRs in mediating the effects of progesterone on the brain (Balasubramanian et al., 2008
; Liu and Arbogast, 2009
; Tokmakov and Fukami, 2009
; Intlekofer and Petersen, 2011
). The notion that membrane PRs exist is not new, and in fact, was supported by Hans Selye's pioneer work in the 40′s that showed various steroid hormones, including progesterone, had very rapid anesthetic effects in contrast to the delayed “main” hormone actions (Selye, 1942
). Four decades later, specific, displaceable binding sites for progesterone were identified in synaptosomal membrane preparations (Towle and Sze, 1983
; Ke and Ramirez, 1990
). Further, progesterone's is quite lipophilic, having a logP
value, or octanol/water partition coefficient, of 4. This value reflects the relative solubility of a compound in an organic phase (i.e., octanol) vs. an aqueous phase (e.g., water). As such, a logP
value of 4 indicates that for every molecule of progesterone that partitions into the aqueous phase, 10,000 molecules partition into the organic phase. This further supports the idea that progesterone interacts with a plasma membrane-associated receptor.
Figure 1 Mechanism of progesterone action in the brain. This figure provides a conceptual overview of how progesterone can elicit both genomic and non-genomic effects that impact its protective effects on the brain, and exemplifies how activation of complementary (more ...)
Two types of distinct cell surface-associated proteins unrelated to classical PRs have been identified so far: membrane PRs (mPRs) and the progesterone membrane receptor component (PGMRC). The mPRs (molecular mass of approximately 40 kDa) had thought to be comprised of three subtypes, mPR α, β, and γ, which belong to the seven-transmembrane domain adiponectin Q receptor (PAQR) family (Zhu et al., 2003a
). Two new subtypes, mPRδ and mPRε, have also been characterized recently in human brain (Pang et al., 2013
). The mPRs bind to progesterone with high affinity (Kd
~5 nM) (Zhu et al., 2003a
), and mediate important physiological functions in male and female reproductive tracts, liver, neuroendocrine tissues, and the immune system as well as in breast and ovarian cancer (Sleiter et al., 2009
; Pang and Thomas, 2011
). Uniquely, recent experimental evidence supports mPRs as G-protein-coupled receptors, as supported by the observation that activation of the mPRα can result in recruitment/activation of pertussis-sensitive inhibitory proteins (Gi
) to down-regulate membrane-bound adenylyl cyclase activity in the sea trout and in humans (Thomas et al., 2007
Despite the fact that the classical PR and mPRs have overlapping regional expression (e.g., both are expressed in the hippocampus, cortex, hypothalamus, and cerebellum) (Brinton et al., 2008
; Meffre et al., 2013
), their profile of ligand specificity is not identical. For example, mPRs bind to 17α-hydroxyprogesterone and 5-dihydroprogesterone with greater affinity than to the classical PRs (Grazzini et al., 1998
; Smith et al., 2008
). In terms of cellular distribution, under non-injured conditions, the mPRα isoform was expressed principally by neuronal cells and not by oligodendrocytes or astrocytes. However, following traumatic brain injury (TBI) mPRα expression was observed in oligodendrocytes, astrocytes, and reactive microglia. This increase in mPR expression was proposed to mediate the anti-inflammatory effects of progesterone under conditions of injury (Meffre et al., 2013
). Thus, the complement of PRs expressed in the brain may be driven by the health of the tissue.
In comparison to the mPRs, the single-transmembrane protein Pgrmc1 (molecular mass 25–28 kDa) and the closely related Pgrmc2 are thought to be a part of a multi-protein complex that binds to progesterone and other steroids, as well as pharmaceutical compounds (Thomas, 2008
). Pgrmc1 was originally discovered in porcine liver and vascular smooth muscles (Falkenstein et al., 1996
; Meyer et al., 1998
), and later cloned in other species, including humans. Pgrmc1 has also been termed 25-Dx in rat and Hpr6 in human [for review, see (Cahill, 2007
)], a result of being identified in different biological systems from multiple species. Pgrmc1 has an N-terminal transmembrane domain and a putative cytoplasmic cytochrome b
5 domain ligand-binding motif. The cytoplasmic domain has target sequences for binding by SH2- and SH3-domain containing proteins as well as tyrosine kinases, implicating a potential role for Pgrmc1 as an adaptor involved in protein interactions and intracellular signal transduction. The subcellular localization of Pgrmc1 has been open to argument, since it was reported to localize in endoplasmic reticulum (Nolte et al., 2000
), Golgi apparatus (Sakamoto et al., 2004
) and nuclei (Beausoleil et al., 2004
). However, evidence supporting the cell surface localization of Pgrmc1 includes the reports by Peluso et al., (Peluso et al., 2005
) and ours (Su et al., 2012
), in which biotinylated Pgrmc1 was localized to the surface (i.e., plasma membrane) of non-permeabilized cells.
Both mPRs and Pgrmc are expressed at high levels in the brain, but their functions relevant to progesterone effect in the CNS have only just started to be revealed. For example, a recent report demonstrated that allopregnanolone and other neurosteroids bound to mPRδ and decreased starvation-induced apoptosis in in hippocampal neuronal cells at low nanomolar concentrations (Pang et al., 2013
). In addition, mPRα, mPRβ and Pgrmc1 have been implicated in progesterone-repressed gonadotrophin-releasing hormone release from hypothalamic neurons (Sleiter et al., 2009
; Bashour and Wray, 2012
). Progesterone-increased neural progenitor proliferation may also be mediated by Pgrmc1 as this effect was blocked by siRNA against Pgrmc1/2 (Liu et al., 2009
). Further, a recent study by Frye, et al., revealed that progesterone-facilitated lordosis (sexual behavior) was significantly reduced by antisense oligodeoxynucleotides (AS-ODNs) against mPRβ, or AS-ODNs against both mPRβ and mPRα, when administered into the ventral tegmental area (VTA) (Frye et al., 2013
). This data supports the potential role of mPRs in progesterone-facilitated lordosis of rats. The cell signaling pathways and associated downstream effects for progesterone-induced non-genomic actions are summarized in Table .
Receptor pharmacology and signaling pathways associated with progesterone-induced non-genomic effects.
The functions of membrane-associated PRs in the CNS are not limited to neurons. In fact, work from our laboratory supports that progesterone triggers BDNF release via Pgrmc1 signaling specifically from glia (Su et al., 2012
). Another report showed that mPRα expression was induced in oligodendrocytes, astrocytes and reactive microglia after TBI (Meffre et al., 2013
), supporting a potential role in mediating the effects of progesterone in inflammation and water homeostasis in the injured brain.
The two receptors do not always work independently. For example, Thomas and colleagues reported that activation of mPRα and -β in human myometrium leads to transactivation of PR-B (Karteris et al., 2006
), as the first evidence that cross talk between the classical PR signaling and membrane-associated PR signaling exists.
It is also worth noting that the classical PR can also mediate the effects of progesterone on signaling pathways through non-genomic/extranuclear mechanisms of progesterone. Human PR-B contains a polyproline motif in its amino-terminal domain that interacts with the SH3 domain of Src (Boonyaratanakornkit et al., 2001
). Therefore, cytoplasmic PR can mediate progesterone-induced rapid activation of c-Src and downstream Ras/Raf/ERK1/2 signaling independent of its transcriptional activity. Activation of the MAPK pathway ultimately results in the phosphorylation/activation of transcription factors such as c-Fos, c-Jun and nuclear PRs to control gene transcription. For example, progesterone was shown to inhibit aortic smooth muscle cell proliferation via Src phosphorylation that in turn, results in RhoA inactivation. The involvement of the PR was supported by the fact that this effect was blocked by RU486, a PR antagonist (Hsu et al., 2011
). PR also mediates progesterone's effects on breast cancer development and progression by activating the Src/ERK1/2 or PI3K/Akt pathways (Saitoh et al., 2005
; Fu et al., 2008
), which leads to activation of the transcription factor Elk-1 and consequent changes in gene expression (Boonyaratanakornkit et al., 2008
In addition to the well-characterized Src pathway downstream of extranuclear PR, there is evidence supporting the activation of G-protein signaling by PR in frogs (Xenopus laevis
). For example, the Xenopus
PR isoform related to the mammalian PR-B localizes to the plasma membrane of oocytes, and that activation of the PR regulates Xenopus
oocyte maturation via the Gβ γ activation of adenylyl cyclase (Guzman et al., 2005
; Evaul et al., 2007
). Interestingly, another study concluded that the Xenopus
ortholog of mPRβ mediated progesterone-induced oocyte maturation via the MAPK signaling (Josefsberg Ben-Yehoshua et al., 2007
). Whether there is a cross talk between the PR/Gβ γ pathway and the mPRβ/MAPK pathway in this system remains unclear.