1. Progesterone Receptors
Actions of steroid hormones to influence cellular function fall into two classifications: the slower classical genomic response and the rapid non-genomic response. The genomic actions of P
4 in target tissues are mediated through interactions with intracellular progesterone receptors (PGR; also called PR), ligand-activated transcription factors that belong to the nuclear receptor family
[1][2][3]. This family of transcription factors includes estrogen receptors (ER, ESR1, and ESR2), androgen receptors (AR), mineralocorticoid receptors (MR, nuclear receptor subfamily 3 group C member 2/NR3C2), and glucocorticoid receptors (GR, nuclear receptor subfamily 3 group C member 1/NR3C1). These are referred to as “classical” steroid receptors in which steroid binding leads to a long-lasting but slowly emerging response
[2]. The transcription factor PGR is expressed in all P
4-responsive organs, including the reproductive tract, mammary glands, cardiovascular system, and the oviduct
[4][5][6][7][8]. To stimulate this “classical” response, binding of P
4 to the ligand-binding domain of the PGR induces a conformational change that transforms the receptor from a static, non-DNA-binding configuration into one that activates gene transcription. This occurs by loss of associated (heat shock) proteins and dimerization of receptor moieties. The activated receptor–ligand complex can then activate the transcriptional machinery by direct action on regulatory motifs, most commonly at PGR response elements (PRE) sites, or by direct association of ligand-bound PGR with other transcription factors and coactivators
[3][9][10][11][12].
Differences among fixation methodologies can greatly affect the outcome of ER and PGR localization by immunohistochemistry (IHC). For instance, studies of OCT-embedded cryosections of oviductal fimbria and ampulla of ovariectomized macaques revealed staining for ER and PGR localized to the nuclei of epithelial, underlying stromal cells and smooth muscles. However, staining of paraffin-embedded sections produced variable cytoplasmic plus nuclear localization. Since binding assays revealed that most of the ER and PR were recovered from the cytosol, IHC on cryosections was interpreted as indicating that PGR was rapidly translocated from the cytoplasm to the nucleus regardless of the hormonal state of the animal and interacted strongly with chromatin in the nucleus. It is noteworthy to mention that ESR-2 (ERβ) expression is reported for the oviduct of several mammalian species
[13], but the role of ESR-2 during cyclic regulation in NHPs in unknown.
It is worth mentioning that PGR exists in two primary isoforms (A and B) encoded by a single gene but with different initiation sites that permit transcription of either a large or short isoform
[5][14][15][16]. The larger (PR-B) isoform contains an N-terminal fragment of 164 amino acids that is absent from the short (PR-A) isoform. Thus, PR-B exhibits three transcription-activating domains (AF-1, AF-2, and AF-3), whereas PRA contains only two (AF-1 and AF-2)
[17]. The two PR isoforms have similar steroid hormone and DNA binding activities but have distinct functions depending on the cell type and context of the target gene promoter. PRB appears to be a stronger transcription activator than PRA
[15]. Due to the structural overlap of the two PGR isoforms, assessing the localization of the two PGR isoforms in the oviduct has been challenging. One approach is to use differential immunostaining with antibodies directed against PR-B and PR-A plus PR-B as well as specific differential PCR approaches. Using this approach, researchers at the University of Edinburgh reported attenuated PR-B in human fallopian tubes during the luteal phase of the menstrual cycle and during ectopic pregnancy
[18]. However, cyclic regulation of oviductal PR-A/PR-B isoforms has not been confirmed in NHP related ones.
Rapid, nongenomic actions of P
4 are also reported for the oviduct. These are mainly attributed to so-called “membrane” receptors or “non-classical” progesterone receptors that appear to activate cellular second messenger pathways
[2][19][20][21]. Among the rapid actions of P
4 in the oviduct are the effects on ciliary beat frequency
[22][23][24] and rapid alteration to spermatozoa motility
[25]. Non-classical PRs include a family of membrane progestin receptors (mPRs) as well as the G-protein-coupled receptor (GPCR) family, which includes progesterone receptor membrane component (PGRMC), PGRMC1, and PGRMC2
[1][26]. The PGRMC family shares properties not associated with P
4, including a heme-binding domain related to some cytochromes. The mPRs were first reported in fish
[20][27][28], and subsequently, five mPR subtypes (α, β, γ, δ, and ε) were identified
[27] in a wide array of cell types in many mammalian species, including primates. The mPRs have no known homologies with GPCRs or nuclear PGRs, but are structurally related to adiponectin receptors and are classified as the progestin and adipoQ receptor (PAQR) superfamily. They display a predicted seven-transmembrane region and bind small steroid molecules, resulting in G-protein activation. However, the function of mPRs remains less clearly defined than that of the nuclear receptors
[2]. This is largely due to a lack of data on the mPR steroid binding domains
[29] and the absence of well-defined mPR modulators.
2. Cyclic Regulation of Steroid Responsiveness
It is well recognized that the oviduct is an estrogen-responsive organ that expresses ERα (ESR1) and PGR. Interestingly, ESR2 (ERβ) is also expressed in human fallopian tube ciliated cells, but the role of ERβ in NHPs remains to be determined. ER (not specific to ESR1 or ESR2) and PGR abundance have been assayed in naturally cycling NHPs as well as in NHPs treated sequentially with E
2 and P
4 to create artificial menstrual cycles. The earliest research characterizing the abundance of PGR in the primate oviduct utilized radiolabeled steroid binding on human fallopian tube
[30]. These were followed by NHP that employed steroid binding and exchange assays to estimate levels of estrogen receptor and PGR (for example, specific binding) in tissue homogenates
[31][32]. These assays often used radiolabeled R2858 (a nuclear ER ligand) and R5020 (a nuclear PGR ligand) to avoid the metabolism of estrogen and P
4. The sum of specifically bound steroids to the nuclear and cytosolic fractions from the homogenates represented an estimate of total receptor abundance. Binding of labeled R2858 and of R5020 were found to be significantly elevated in ovariectomized animals treated with E
2 (or at mid-menstrual cycle) compared to hormone-depleted animals. This technique revealed that the oviduct’s differentiation into a fully ciliated and secretory endosalpinx epithelium was accompanied by significant increases in total ER and PGR
[31]. Treatment of E
2-primed monkeys with E
2 in combination with P
4 similar to the luteal phase resulted in significantly reduced levels of ER and PGR. In the case of ER, levels were reduced below those of ovariectomized untreated animals. Thus, ER and PGR expression were dependent on E
2 action. Moreover, average ER and PGR levels were lower in animals treated with a combination of E
2 and P
4 than those observed in ovariectomized untreated monkeys. Because treatment with P
4 alone failed to stimulate either ER or PGR, it was proposed that P
4 acted to antagonize the effects of E
2 on oviductal differentiation by suppressing ER levels below the threshold required to facilitate E
2 action.
The overall relationship provided by classical binding assays appears to be more complex than was initially proposed. In concert with biochemical binding assays, cellular localization of ER and PGR by IHC on cryosections revealed that both cell and tissue type affected P4 suppression of ER and PGR. In support of binding assay results, the abundance of cells with strong nuclear staining for both ER and PGR increased in the follicular phase (and after E2 treatment) and decreased in the luteal phase (or after E2 plus P4 treatment). However, specific staining for epithelial ER and PGR were localized to the secretory epithelial cells, not the ciliated cells. This represents a paradox in that E2 and P4 strongly affect the ciliated phenotype, but staining is minimal in the ciliated cells. How can the dramatic effects of both E2 and P4 on the ciliated cells occur when the ciliated cells lack or express minimal receptors for both steroids? Moreover, PGR staining was almost completely absent in the epithelium during the luteal phase or after P4 treatment. This produces the question: How does P4 maintain its effects while suppressing its own receptor?
IHC revealed that strong ERα and PGR staining were present in stromal, smooth muscle, and secretory epithelial cells, suggesting that the effects of P4 on ciliated cells may be indirect. In the luteal phase (or after P4 treatment), ERα staining is retained in all the undifferentiated epithelial cells and in the underlying stromal cells, whereas PGR is minimal in the epithelium and retained (but noticeably less intense) in the stromal compartment. Therefore, one possibility is that the state of differentiation of the oviductal epithelium is mediated indirectly through soluble growth factors (or other unidentified mediators) secreted by ERα- and PGR-positive stromal cells. Moreover, stromal cells are separated from the epithelium by a definitive basement membrane, which could reduce the influence of soluble factors.
One potential mediator of P
4 progesterone action, particularly in PR-negative cells, is the presence of specific mPRs or other non-classical PRs reported for human, murine, bovine, and canine oviducts, as well as ovarian cancers that may be of tubal origin. Oviductal mPR (beta and gamma) have been localized to bovine, human, and mouse ciliated epithelial cells
[33] and may mediate the rapid effect of P
4 on cilia beat frequency. However, localization of mPRs to oviductal cilia does not appear to reflect expected cyclic changes in ciliated cell abundance
[33], as observed in nonhuman primates. Cyclic PGRMC1 and PGRMC2 expression and localization are reported for the macaque endometrium, but cyclic regulation in the NHP oviduct has not been extensively learned. The absence of reliable mPR/PGRMC modulators has significantly limited these pathways in NHP models. In contrast, the expression and cellular action of nuclear PGR in the mammalian oviduct have been learned extensively.
3. Progesterone Receptor Modulators
The characterization of nuclear PGR isoforms has prompted the development of synthetic compounds called progesterone receptor modulators (PRMs)
[34]. These include synthetic P
4 analogs (progestins) and P
4 antagonists (anti-progestins; PRAs) that bind to PGR and either stimulate or block PGR function
[35][36][37]. It is noteworthy that long-term treatment with P
4 and synthetic progestins reduces the abundance of ciliated cells in NHPs
[38], and short-term treatment decreases cilia beat frequency in human oviductal cultures
[39].
Mifepristone (RU486), the first well-characterized PRA, acts as a glucocorticoid receptor antagonist in the primate uterus, opposing various estrogen effects. The action of PRMs is often unique to the target organ, cell type, and sometimes the animal model examined. This has led to tissue-selective or physiologically selective PRMs
[35]. The nuclear action of PRMs on classical (genomic) action of P
4 has been most extensively evaluated because of the pharmaceutical potential of these compounds to treat gynecological disorders
[34][35][36][40].
Compared to nuclear receptors, the action of PRM on mPRα, mPRβ, and mPRγ appears less clearly defined, with ranging from the minimal binding of mPRs to synthetic PRMs, especially classical PGR antagonists such as mifepristone
[29][41], to putative or predictable actions
[42]. Moreover, much of the it on the nuclear action of PRMs have been conducted on non-primate models with strikingly different hormone profiles. In vivo assessments of mPR actions are confounded by co-expression of nuclear PGRs in many of the responsive cell types. However, differential binding of synthetic ligands offers the potential development of mPR-selective agonists and antagonists
[29].
It has treated rhesus macaques with an array of potent PRAs including mifepristone
[43][44], ZK 137 316
[45][46], ZK 230 211
[47], and CDB 2914 (Ulipristal)
[48]. However, the primary experimental goal of these studies was to evaluate the action of these compounds on the uterine endometrium. Driving these were the observation that some PRA compounds, including mifepristone, have been reported to have unexpected anti-estrogenic actions on the endometrium. However, in the oviduct, pure PRA compounds such as ZK 230-211 appear to lack antiestrogen effects and block the genomic action of P
4 [43][44]. In this condition, the estrogen action is unopposed, and the oviducts appear in a fully differentiated and ciliated secretory state.
PRMs can also provide contraception. Levonorgestrel is a contraceptive progestin that has inhibitory actions on oviductal cilia beat frequency and ovulation. Ulpristral and levonorgestrel have potential as emergency contraception, presumably by blocking ovulation. However, a secondary target may include oviductal cilia function. In vitro, progesterone can decrease human oviductal ciliary beat frequency (CBF) and muscular contractions, and the inhibitory effect of progesterone on CBF can be antagonized by mifepristone, a progesterone receptor (PR) modulator. Treatment of cycling rhesus monkeys with low-dose ZK 137-316, a compound very similar to mifepristone
[49], prevented pregnancy at low doses that allowed menstrual cycles
[50]. However, low-dose ZK 137-316 did not block ovulation and failed to alter oviductal differentiation and sperm passage but did significantly increase oviductal fluid levels of OVGP1
[51]. In contrast, ulipristal acts as a mixed agonist–antagonist compound and could disrupt gamete passage. This outcome on sperm passage may be PRA dose-dependent because other were indicated that both ulipristal and mifepristone reduce ciliary beat frequency and contractility in human oviductal explants
[52].
As indicated above, blockade of P
4 action in NHPs is not associated with well-defined tubal abnormalities. Treatments with pure PRA, including ZK137-316
[49], mifepristone, and ZK 230-211, result in a fully ciliated and secretory tubal epithelium. This is not abnormal for the proliferative phase of the cycle. However, tubal abnormalities such as ectopic pregnancy are almost nonexistent in NHPs compared to women. Reproductive tract infections occur in NHPs and appear to be affected by estrogen and P
4 action on the cervix, endometrium, and oviduct
[53]. It can be speculated that treatment with mixed-action PRM therapy could alter normal cyclic changes. However, this represents a knowledge gap and further are required to assess the impact of P
4 modulation on tubal dysfunction in NHP models.