Research ArticleHormone Signaling

Functional coupling of GABAA/B receptors and the channel TRPV4 mediates rapid progesterone signaling in the oviduct

See allHide authors and affiliations

Science Signaling  14 Aug 2018:
Vol. 11, Issue 543, eaam6558
DOI: 10.1126/scisignal.aam6558

How progesterone stimulates cilia

The transport of eggs along the ciliated epithelium of the oviduct depends on the ciliary beat frequency (CBF), a process that requires intracellular Ca2+ signaling. The hormone progesterone is thought to accelerate egg transport along the oviduct. Because any rapid effects of progesterone on CBF would be incompatible with its function as a transcriptional regulator, Jung et al. investigated how progesterone affected Ca2+ signaling in mouse ciliated oviduct cells. They found that progesterone stimulated increased intracellular Ca2+ concentrations in a process that required the nonselective cationic channel TRPV4 and the stepwise activation of GABAA and GABAB receptors, which physically associated with each other in response to progesterone and GABAergic agonists. Together, these data implicate progesterone and GABA receptor agonists in the rapid acceleration of cilial activity in the oviduct.


The molecular mechanism by which progesterone (P4) modulates the transport of ova and embryos along the oviduct is not fully resolved. We report a rapid response to P4 and agonists of γ-aminobutyric acid receptors A and B (GABAA/B) in the mouse oviduct that was characterized by oscillatory Ca2+ signals and increased ciliary beat frequency (CBF). Pharmacological manipulation, genetic ablation, and siRNA-mediated knockdown in oviductal cells, as well as overexpression experiments in HEK 293T cells, confirmed the participation of the cationic channel TRPV4, different subunits of GABAA (α1 to α3, β2, and β3), and GABAB1 in P4-induced responses. TRPV4-mediated Ca2+ entry in close proximity to the inositol trisphosphate receptor was required to initiate and maintain Ca2+ oscillations after P4 binding to GABAA and transactivation of Gi/o protein–coupled GABAB receptors. Coimmunoprecipitation experiments and imaging of native tissue and HEK 293T cells demonstrated the close association of GABAA and GABAB1 receptors and the activation of Gi/o proteins in response to P4 and GABA receptor agonists, confirming a molecular mechanism in which P4 and GABAergic agonists cooperatively stimulate cilial beating.


Progesterone (P4) is present at high concentrations (up to 9 μM) in the ovarian secretions into the fimbriae and peritoneum in humans (1), as well as in the hamster oviduct, where it reaches a concentration of 500 nM (2). In the oviduct, P4 affects muscle and epithelial cells, as well as sperm (3). Early reports showed that the administration of progestogens accelerates the transport of eggs from the oviduct to the uterus in several animal species (46). In the ciliated epithelium of the oviduct, the transport of gametes and embryos relies on the maintenance of the appropriate velocity of mucociliary transport (7), which, in turn, is driven by ciliary beat frequency (CBF) (8). However, a number of studies showed that oviductal (911) and tracheal (12) CBF is reduced by long-term administration of P4 (the first effects occur 20 to 30 min after hormone addition and are maintained for 24 to 36 hours in the presence of hormone), as well as during the periods of the ovarian cycle with increased P4 amounts (13), which contradicts the reported P4-mediated acceleration of egg transport in the oviduct. There is also controversy about the involvement of the classical P4 receptor in CBF reduction, with studies both supporting (11) and questioning (9) its participation.

Steroid hormones exert their long-term actions through receptors that function as transcription factors (14). Rapid, nongenomic actions of steroid hormones have also been reported (1517). In addition to the first description of the rapid anesthetic effect of P4 due to the modulation of the ligand-gated γ-aminobutyric acid receptor A (GABAA) chloride channel (1820), other nongenomic mechanisms of action of P4 have been described (21). Rapid generation of Ca2+ signals has been found in diverse tissues (2225). Although both Ca2+ release from the endoplasmic reticulum (ER) and Ca2+ entry through plasma membrane channels have been reported in response to P4, the molecular players participating in such responses remain largely unknown, except for examples such as the steroid-activated CatSper (26, 27) and TRPM3 (28) cationic channels.

Considering that intracellular Ca2+ is a key regulator of the CBF (8, 29), we sought to identify whether P4 regulated CBF through the modulation of Ca2+ signaling in oviductal ciliated cells. We found that P4 increased the intracellular concentration of Ca2+ through a mechanism that required both the TRPV4 channel, which produces Ca2+ signals linked to CBF modulation (3033), and GABA receptors, which are the main inhibitory receptors in the mammalian central nervous system (34). GABAergic signaling involving the ligand-gated GABAA chloride channel and the metabotropic Gi/o protein–coupled GABAB receptor has been found in epithelial tissues (3537). The GABAA channel is a pentameric complex formed by the combination of two or three different subunits (38, 39), whereas the metabotropic GABAB receptor is a heterodimer composed of the agonist-binding GABAB1 and the Gi/o protein–coupled GABAB2 subunits (40). Steroids bind to GABAA at the α subunit, β subunits, or both (the αβ interface), although other subunits may also participate (19, 41, 42). No interaction of steroids with the GABAB receptor has been documented to date (43). Here, we report a close functional interaction between the cationic channel TRPV4 and GABAA/B receptors that promoted rapid responses to P4 in oviductal ciliated cells.


The production of rapid, P4-induced intracellular Ca2+ signals in oviductal ciliated cells requires TRPV4

Monitoring the intracellular Ca2+ concentration in Fura-2–loaded ciliated oviductal cells from mice showed that treatment with vehicle alone (Fig. 1A) did not affect the intracellular Ca2+ concentration, whereas rapid oscillatory Ca2+ signals were generated in response to 100 nM P4 in about 25% of the ciliated cells (Fig. 1B). We also examined the responses of ciliated oviductal cells to 10 nM and 10 μM P4 (fig. S1, A and B). No Ca2+ signals were stimulated in response to the less active metabolite of P4, 20α-hydroxyprogesterone (20αDHP4) (fig. S1C). Pretreatment of oviductal cells with the phospholipase C (PLC) inhibitor U73122 prevented the generation of P4-induced Ca2+ signals (Fig. 1C), suggesting a need for the generation of inositol 1,4,5-trisphosphate (IP3) and the release of Ca2+ from the ER in the P4-mediated response. This hypothesis was further confirmed by depleting the ER Ca2+ stores with thapsigargin, a sarco-ER Ca2+ pump inhibitor (Fig. 1D), and using the IP3 receptor (IP3R) inhibitor xestospongin C (Fig. 1E). Both actions prevented the generation of P4-induced Ca2+ signals.

Fig. 1 P4 increases the intracellular Ca2+ concentration in mouse ciliated oviductal cells.

(A to F) Representative intracellular Ca2+ signals obtained from different primary cultures of Trpv4+/+ (A to E) or Trpv4−/− (F) mouse ciliated oviductal cells stimulated with vehicle (A), 100 nM P4 (B), 2 μM U73122 and 100 nM P4 (C), 1 μM thapsigargin and 100 nM P4 (D), or 1 μM xestospongin C and 100 nM P4 (E). Trpv4−/− ciliated cells were stimulated with P4 (F). (G) Average percentage of Trpv4+/+ and Trpv4−/− mouse ciliated oviductal cells that showed increased intracellular Ca2+ concentrations in response to the indicated treatments. The number of coverslips analyzed for each condition is indicated in each bar. Total number of cells per condition, >100. Data are means ± SEM. *P < 0.05, **P < 0.01, and ***P < 0.001 when comparing vehicle-treated cells with any other condition [as determined by one-way analysis of variance (ANOVA) followed by a Bonferroni post hoc test] and when comparing Trpv4+/+ cells to Trpv4−/− cells (as determined by Student’s t test).

In addition to playing a role in mechanosensitive and osmosensitive signaling in ciliated cells (44), the TRPV4 cationic channel interacts with and shows reciprocal modulation by the IP3R (31, 32), suggesting that it may participate in the generation of P4-mediated Ca2+ signals. Therefore, we tested whether pharmacological or genetic ablation of TRPV4 affected this response. P4 did not elicit Ca2+ signals in Trpv4−/− cells or in Trpv4+/+ cells treated with the TRPV4 antagonist HC067047 (Fig. 1, F and G). Direct activation of TRPV4 by P4 was absent in isolated ciliated oviductal cells and in TRPV4-overexpressing human embryonic kidney (HEK) 293T cells (fig. S2, A and B), suggesting that basal TRPV4 activity is sufficient to promote P4-induced Ca2+ signals.

Although classical and plasma membrane P4 receptors are present in the ciliated epithelia of the oviduct (45, 46), we found that their treatment with the P4 receptor antagonist RU-486 and 1(S),2(R)-d-erythro-2-(N-myristoylamino)-1-phenyl-1-propanolan (d-eMAPP) (47), respectively, failed to inhibit the P4-induced Ca2+ response (Fig. 1G). Finally, pretreatment with pertussis toxin (PTX), which inhibits the activation of Gi/o proteins, prevented P4-induced Ca2+ signals (Fig. 1G). Together, these experiments suggested a rapid, nongenomic effect of P4 that was independent of the activity of classical and plasma membrane P4 receptors but required the involvement of Gi/o proteins, the generation of IP3, the release of Ca2+ from the ER, and the participation of the TRPV4 channel.

Progesterone uses GABA receptor signaling to stimulate Ca2+ responses in oviductal ciliated cells

In searching for the signaling pathway activated by P4, we focused on the GABAergic system, because it is capable of binding to progestogens (19), is linked to Gi/o proteins (40), is found in the oviduct (35, 36), and can modulate ciliary activity in Paramecium (48). Considering that steroid binding to GABAA uses different combinations of subunits (49), we first examined which GABAA-encoding mRNAs were found in the mouse oviduct by reverse transcription polymerase chain reaction (RT-PCR) analysis. We found that the genes encoding GABAA α1 to α4, β1 to β3, γ3, and δ subunits, as well as GABAB1 were expressed in the oviduct (fig. S3). Note that the gene encoding GABAB2 was not expressed. GABAB2 is proposed to promote the cell surface expression of GABAB1 and to couple to Gi/o proteins (40), although there is evidence of atypical GABAB1 trafficking to the plasma membrane and activation of Gi/o pathways in the absence of GABAB2 (5052).

To check the functionality of GABA receptors, we used specific agonists of GABAB (baclofen) and GABAA (muscimol) and measured intracellular Ca2+ concentrations in primary cultures of mouse oviductal ciliated cells. Similar to P4, baclofen elicited oscillatory Ca2+ signals in cultures obtained from Trpv4+/+ mice (Fig. 2A). These responses were absent in cultures obtained from Trpv4−/− mice and challenged with baclofen (Fig. 2B) and in Trpv4+/+ cells treated with P4 in the presence of CGP35348, a specific antagonist of GABAB (Fig. 2C). Muscimol also generated oscillatory Ca2+ signals in Trpv4+/+ cells (Fig. 2D) but not in Trpv4−/− cells (Fig. 2E). The GABAA antagonist bicuculline prevented P4-induced responses in Trpv4+/+ cells (Fig. 2F). Both of the GABA receptor antagonists were equally effective on cells previously activated by P4 (fig. S4). The responses to baclofen and muscimol were absent in cultures obtained from Trpv4+/+ mice and treated with the TRPV4 antagonist HC067047 (Fig. 2, G and H). These experiments demonstrated that GABA receptor agonists and P4 shared a common mechanism for generating Ca2+ signals that required TRPV4. Given that inhibitors of both GABAA and GABAB prevented P4 responses but steroids have only been shown to bind to GABAA receptors (43), we hypothesized that the sequence of events leading to P4-mediated Ca2+ oscillations was initiated by the interaction of P4 with GABAA followed by the transactivation of GABAB.

Fig. 2 GABAergic activity in mouse ciliated oviductal cells.

(A to F) Representative intracellular Ca2+ signals obtained from different primary cultures of Trpv4+/+ (A, C, D, and F) and Trpv4−/− (B and E) mouse ciliated oviductal cells stimulated with 20 μM baclofen (A and B), 100 μM CGP35348 and 100 nM P4 (C), 10 μM muscimol (D and E), or 10 μM bicuculline and 100 nM P4 (F). (G to J) Average percentage of Trpv4+/+ and Trpv4−/− mouse ciliated oviductal cells that showed increased intracellular Ca2+ concentration in response to the indicated treatments. The number of coverslips analyzed for each condition is indicated in each bar. Total number of cells per condition, >90. Data are means ± SEM. *P < 0.05 and ***P < 0.001 when comparing baclofen-treated cells (G) or muscimol-treated cells (H) with any other condition (as determined by one-way ANOVA followed by a Bonferroni post hoc test); for (I) and (J), a Student’s t test was used.

To test whether activation of GABAA was upstream of GABAB, we performed experiments combining specific agonists and antagonists of these receptors. The Ca2+ response to the GABAA agonist muscimol was prevented in the presence of the GABAB antagonist CGP35348 (Fig. 2I), whereas the response to the GABAB agonist baclofen was insensitive to the GABAA antagonist bicuculline (Fig. 2J), which suggests that the activation of GABAA was upstream of GABAB activation. This sequence of events fits with the hypothesis that P4 binds to GABAA (20) and uses GABAA to activate GABAB and stimulate the subsequent release of Ca2+ from the ER through a Gi/o-PLC-IP3 pathway. We were unable to measure GABAA chloride currents in ciliated oviductal cells in response to GABA (fig. S2D), suggesting that the GABAA subunits present in these cells may form a complex that is capable of binding to steroids and stimulating Ca2+ signals but that they are unable to generate a functional GABAA chloride channel.

GABAA and GABAB1 physically interact with each other and become activated

To further investigate the molecular mechanism used by P4 to stimulate TRPV4-dependent Ca2+ responses, we combined native tissue and heterologous expression systems to directly test two underlying premises of our model: first, the interaction between the GABAA and GABAB1 subunits, which was previously reported in heterologous expression systems (50), and second, the activation of Gi/o proteins in response to P4 receptor and GABA receptor agonists. The presence of GABAA and GABAB1 subunits in ciliated cells was examined by confocal immunofluorescence microscopy of mouse oviduct slices using antibodies against GABAAα2, GABAB1, acetylated α-tubulin (to detect cilia), and TRPV4. We found that GABAA/B receptors were localized to ciliated epithelial cells that also had TRPV4 (Fig. 3A) and acetylated α-tubulin (fig. S5A). The specificity of the antibody against GABAAα2 was examined with a GABAAα2 immunogenic peptide in oviduct sections (fig. S5A) and oviduct lysates analyzed by Western blotting (fig. S5B). The specificity of the antibody against TRPV4 was examined in oviduct sections obtained from Trpv4+/+ and Trpv4−/− mice (fig. S5C). Next, we evaluated the physical interaction between GABAA and GABAB1 receptor subunits in native oviductal tissue and heterologous expression systems. First, we tested for the presence of GABAAα2, GABAB1, and TRPV4 in oviduct lysates (fig. S5D). Immunoprecipitates obtained from mouse oviduct lysates with antibodies against GABAB1, TRPV4, or GABAAα2 were analyzed by Western blotting for the presence of GABAB1 (Fig. 3B). We found that GABAB1 coimmunoprecipitated with GABAAα2 but not with TRPV4. Similar results were obtained when we analyzed the proximity of GABAAα1-YFP (yellow fluorescent protein) and GABAB1-CFP (cyan fluorescent protein) using the fluorescence resonance energy transfer (FRET) technique, as previously described (53, 54). The FRET signal was statistically significantly increased with the coexpression of GABAAα1-YFP and GABAB1a-CFP in HEK 293T cells, but not when GABAAα1-YFP or GABAB1a-CFP was coexpressed with unfused CFP or YFP, respectively (Fig. 3C and fig. S6). A study previously showed a direct interaction between GABAB1 and the TRPV1 channel (55), one that modifies channel desensitization through a noncanonical GABAergic mechanism: It does not require the presence of the GABAB2 subunit, involves a direct interaction between TRPV1 and GABAB1, and is independent of signaling downstream of GABAB. Our coimmunoprecipitation experiments performed with mouse oviductal tissue did not provide evidence of a similar physical interaction between TRPV4 and GABAB subunits. To further confirm this observation, we expressed YFP- or CFP-tagged GABAB1, GABAB2, and TRPV4 proteins in HEK 293T cells and analyzed any potential interactions by FRET. We observed a statistically significant increase in FRET signal when TRPV4-YFP was coexpressed with TRPV4-CFP and when GABAB1-CFP was coexpressed with GABAB2-YFP (Fig. 3D), indicating the formation of the respective TRPV4 (53) and GABAB (56) multimeric complexes, but not when TRPV4-CFP or GABAB1-YFP was coexpressed either with or without GABAB2 (Fig. 3D).

Fig. 3 Interaction and activation of GABAA/B receptors.

(A) Top: Confocal immunofluorescence images of TRPV4, GABAB1, and GABAAα2 in tissue sections of mouse oviduct. Bottom: Corresponding differential interference contrast images. Scale bar, 50 μm. (B) Coimmunoprecipitation of GABAB1 and GABAAα2. Immunoprecipitates (IP) obtained from mouse oviduct with antibodies specific for GABAB1, TRPV4, and GABAAα2 were analyzed by Western blotting (IB) with an anti-GABAB1 antibody. Data are representative of two independent experiments. (C) Quantification of the FRET efficiency measured in HEK 293T cells expressing GABAB1-CFP and YFP (n = 6 experiments), GABAAα1-YFP and CFP (n = 5), or GABAB1-CFP and GABAAα2-YFP (n = 6). Data are means ± SEM. *P < 0.05 when comparing all pairwise conditions (as determined by Kruskal-Wallis one-way ANOVA followed by Dunn’s method for multiple comparisons). (D) Study of the proximity between GABAB and TRPV4 proteins expressed in HEK 293T cells. FRET efficiencies are expressed as the increase in the FRET donor CFP after bleaching the FRET acceptor YFP. FRET efficiency was determined at the plasma membrane of HEK 293T coexpressing soluble CFP and TRPV4-YFP, CFP- and YFP-fused TRPV4, GABAB1-CFP and GABAB2-YFP, TRPV4-CFP and GABAB1-YFP, and GABAB2 or TRPV4-CFP and GABAB1-YFP. ***P < 0.001 when comparing soluble CFP and TRPV4-YFP to any other condition, as determined by one-way ANOVA followed by Bonferroni post hoc test. (E) Baclofen- and P4-induced changes in the BRET ratio were determined in HEK 293T cells expressing GABAB1–2 receptors, Gαi/o-RLuc, Gγ2-Venus, and Gβ2 or in HEK 293T cells expressing the α2-AR, Gαi/o-RLuc, Gγ2-Venus, and Gβ2. The numbers of experiments (each performed in triplicate) are shown for each condition. *P < 0.05 and **P < 0.01 compared to vehicle, as determined by one-way ANOVA followed by a Bonferroni post hoc test. NS, not significant; NA, not applicable.

We then used bioluminescence resonance energy transfer (BRET) to study the conformational rearrangements of G protein subunits during their activation. A study previously showed that after the binding of agonists to the GABAB receptor, the BRET ratio decreases during G protein activation because of the conformational rearrangements of the Gαi/o-RLuc and Gγ2-Venus subunits (57). Consistent with the activation of G proteins by P4 receptor and GABAB agonists, both P4 and baclofen, but not the vehicle, induced changes in the BRET ratios in HEK 293T cells overexpressing GABAB1 and GABAB2 receptors, Gαi/o-RLuc, Gγ2-Venus, and Gβ2 (Fig. 3E). No substantial effects of baclofen or P4 were observed in HEK 293T cells overexpressing the α2-adrenergic receptor (α2-AR), Gαi/o-RLuc, Gγ2-Venus, and Gβ2. We used noradrenaline as a positive control to stimulate decreases in BRET after activation of the α2-AR (Fig. 3E).

P4-mediated Ca2+ signals can be reconstituted in HEK 293 cells

We next tested whether P4-stimulated increases in intracellular Ca2+ concentration could be reconstituted in cell systems commonly used for heterologous expression (Fig. 4 and fig. S7). HEK 293T and HEK 293WSS1 cells, stably overexpressing α1, β2, and γ2 GABAA subunits (58), were used. First, we detected by RT-PCR the expression of GABA subunits in both cell lines (fig. S7, A to D). HEK 293WSS1 cells expressed no GABAB subunits, whereas HEK 293T cells expressed GABAB2 and a truncated, inactive form (GABAB1e) typically found in the peripheral nervous system and other tissues (fig. S7D) (59). In HEK 293T cells transfected to express GABAB receptors or TRPV4 alone, P4 (Fig. 4, A to C) or GABA receptor agonists (fig. S7E) did not stimulate a substantial increase in intracellular Ca2+ concentration. A marked increase in Ca2+ concentration in response to GABA receptor agonists (fig. S7E) or P4 (Fig. 4D) occurred when TRPV4 was coexpressed with GABAB1 and GABAB2, a response that was prevented in the presence of the TRPV4 inhibitor HC067047 (Fig. 4E). We calculated the mean percentages of HEK 293T cells overexpressing different combinations of GABAB receptor subunits and the TRPV4 channel that responded to P4 (Fig. 4G). A statistically significant increase in the percentage of HEK 293T cells responding to P4 with changes in intracellular Ca2+ concentration only occurred when TRPV4 was coexpressed with GABAB1 in the presence or absence of GABAB2 (Fig. 4G). Pharmacological manipulations in HEK 293T cells overexpressing TRPV4 and GABA receptors (Fig. 4H) provided similar results to those obtained from mouse oviduct cultures: The response to P4 was prevented in cells treated with U73122 (PLC inhibitor), HC067047 (TRPV4 inhibitor), CGP35348 (GABAB inhibitor), or bicuculline (GABAA inhibitor) or when extracellular Ca2+ was removed.

Fig. 4 Ca2+ signals induced by P4 and GABA receptor agonists in heterologous expression systems.

(A to F) Representative intracellular Ca2+ signals obtained from HEK 293T cells transfected with plasmids encoding the indicated constructs and treated with the indicated compounds. (G and H) Average percentage of HEK 293T cells responding with increases in intracellular Ca2+ concentration in response to 100 nM P4 when overexpressing different combinations of TRPV4 and GABA receptors. Drugs were used at the following concentrations: 10 μM HC067047, 2 μM U73122, 100 μM CGP35348, and 10 μM bicuculline. The number of coverslips used for each condition is shown for each bar. Data are means ± SEM. *P < 0.05 and ***P < 0.001 when comparing P4-treated nontransfected cells with any other condition, as determined by one-way ANOVA followed by a Bonferroni post hoc test.

Similar to HEK 293T cells, HEK 293WSS1 cells, stably overexpressing GABAA subunits, generated Ca2+ signals in response to muscimol, P4, and baclofen when both GABAB and TRPV4 were transiently overexpressed (fig. S7F). Moreover, a statistically significantly increased percentage of cells responded to P4 when only the GABAB1 subunit was coexpressed with TRPV4 in HEK 293WSS1 cells, which do not express endogenous GABAB subunits (fig. S7F). Together, these results suggest that the endogenous amounts of GABAA present in HEK 293T cells are sufficient to maintain the response to P4. However, similar to the native tissue, HEK 293T cells were not capable of producing GABAA-mediated chloride currents (fig. S2C), which were only recorded in HEK 293WSS1 cells (fig. S2E). To further confirm the participation of GABAA subunits in the response of transfected HEK 293T cells to P4, we knocked down various GABAA subunits in HEK 293T cells, both individually and in combination (Fig. 5). HEK 293T cells transiently overexpressing GABAB1, GABAB2, and TRPV4 were transfected with a control small interfering RNA (siRNA) or with siRNAs individually targeting genes encoding GABAAα1 to GABAAα3, GABAAβ2, and GABAAβ3, or a pool or siRNAs targeting genes encoding all five GABAA subunits. Quantitative PCR analysis showed that the mRNAs of genes encoding GABAA subunits were decreased in abundance by 50 to 70% when the cells were treated with the corresponding siRNAs (Fig. 5A). Changes in intracellular Ca2+ flux in HEK 293T cells were recorded in response to P4, and the area under the curve for each individual cell trace was integrated (Fig. 5B). A consistent reduction in the P4-induced Ca2+ signal was observed in HEK 293T cells transfected with pooled siRNAs targeting all five GABAA subunits, as well as with siRNA targeting GABAAα2 alone (Fig. 5B). There was also a trend toward a reduced Ca2+ response in cells treated with siRNA targeting GABAAα3, but this did not reach statistical significance. Consistent with the view that multiple combinations of different GABAA subunits support the interaction with steroids, we observed that the largest decrease in the response of the cells to P4 was obtained using a pool of siRNAs targeting GABAA subunits to simultaneously reduce the abundances of all GABAA subunits that may mediate the interaction with P4.

Fig. 5 Knockdown of GABAA subunits reduces the P4-induced changes in intracellular Ca2+ concentration.

(A) Total RNA was extracted from HEK 293T cells transfected with control and pooled GABRA/GABRB siRNAs and analyzed for the efficiency of knockdown of GABRA1 to GABRA3, GABRB2, and GABRAB3 by quantitative real-time PCR analysis of mRNA amounts. GABAA values were normalized to values of the housekeeping gene encoding β-actin. Knockdown of GABAA subunits was represented as the relative value compared to cells transfected with control siRNA. Data are means ± SEM of three independent experiments. *P < 0.05 and **P < 0.01 using Mann-Whitney (one-tailed) test to compare each target siRNA with its corresponding control siRNA. RQ, relative quantification. (B) Mean intracellular Ca2+ concentration increases in HEK 293T cells transfected with the indicated control and GABRA/GABRB siRNAs. Ca2+ signals were integrated for a period of 15 min after the addition of 100 nM P4 to HEK 293T cells, which were also transfected to express TRPV4, GABAB1, and GABAB2. The number of cells analyzed for each condition is shown for each bar. *P < 0.05 and **P < 0.01 when comparing control siRNA–transfected cells with any other condition, as determined by one-way ANOVA followed by a Bonferroni post hoc test. a.u., arbitrary units.

The interaction of TRPV4 with IP3R is involved in the response of cells to P4

There is a close interaction between TRPV4 and IP3R, as described previously for ciliated epithelial cells (31) and heterologous expression systems (60), which we reasoned might be relevant to the generation of P4-induced Ca2+ signals through a mechanism that may involve Ca2+-induced Ca2+ release. Under conditions of low activation of the PLC-IP3 pathway (usually associated with oscillatory Ca2+ signals), IP3 binds to the IP3R and releases little stored Ca2+, a response that is magnified by the sensitizing effect of Ca2+ entry through plasma membrane channels situated in close proximity to the IP3R (31, 61). To test this hypothesis, we used a TRPV4 mutant (TRPV4ΔCaM) that lacks the IP3R-binding site located between amino acid residues 812 and 831 and therefore is unable to physically interact with the IP3R (31, 60). We found that there was no statistically significant Ca2+ response to P4 when GABAB receptors were coexpressed with TRPV4ΔCaM (Fig. 4, F and G).

P4 has short- and long-term effects on CBF

Ca2+ signals, which are mostly oscillatory and generated at the base of the cilia, are linked to increased CBF in different ciliated cells (8, 29, 62, 63). To test this association in our experimental model, we measured intracellular Ca2+ concentrations and CBF in Trpv4+/+ and Trpv4−/− oviductal cells using high-speed digital video microscopy (32). Trpv4+/+ cells presented greater basal CBF, which was consistent with the larger percentage of Trpv4+/+ cells presenting spontaneous Ca2+ oscillations (Fig. 6A). We also demonstrated the TRPV4-dependent increase in CBF obtained after the addition of the TRPV4 agonist GSK1016790A (64). CBF increased in response to GSK1016790A only in Trpv4+/+ ciliated oviductal cells (fig. S8). Our observation that basal CBF was reduced in Trpv4−/− oviductal cells differs from our previous results showing identical basal CBF in Trpv4+/+ and Trpv4−/− tracheal cells (32) and suggests a different role for the integration of TRPV4 in the physiological control of CBF in the airways and the oviduct.

Fig. 6 Basal and stimulated CBF of mouse ciliated oviductal cells.

(A) Left: Mean basal CBF measured in Trpv4+/+ (white bars) and Trpv4−/− cells (hatched bars). Right: Percentage of Trpv4+/+ (white bars) and Trpv4−/− cells (hatched bars) presenting spontaneous Ca2+ oscillations. (B) Time course of the mean changes in CBF (percentage of control) of Trpv4+/+ cells exposed to 100 nM P4 or vehicle. (C and D) Mean CBF response (percentage of control) measured 5 to 20 min after the addition of the indicated drugs (at the same concentrations used in previous figures, R5020 was used at 10 nM) in Trpv4+/+ (filled bars) and Trpv4−/− cells (hatched bars). (E) Mean basal CBF after 12 hours of incubation with vehicle [dimethyl sulfoxide (DMSO)] or 100 nM P4 with or without 10 μM RU-486. Data are means ± SEM of three to five experiments (as indicated in the bars). *P < 0.05, **P < 0.01, and ***P < 0.001 when comparing Trpv4+/+ and Trpv4−/− cells, as determined by Student’s t test in (A), (C), and (D), or when comparing the vehicle-treated condition with any other condition, as determined by one-way ANOVA followed by Bonferroni post hoc test in (B) and (C).

Considering the link between CBF and Ca2+ flux and that P4 induces Ca2+ signals in the oviductal ciliated cells, an increase in CBF would be expected in response to the hormone. We found that cells obtained from Trpv4+/+ mice, unlike those obtained from Trpv4−/− mice, rapidly exhibited increased CBF after the addition of P4 (Fig. 6, B and C). This response was potentiated using warmed solutions (Fig. 6B), which were previously shown to increase the activity of TRPV4 (54, 65), and prevented by incubation with the PLC inhibitor U73122 but was insensitive to the P4 receptor antagonist RU-486 (Fig. 6C). Furthermore, and consistent with previous reports (11), CBF decreased in response to the classical progesterone receptor (PR) agonist R5020 (Fig. 6C), which does not interact with GABAA receptors at this low concentration (66). CBF did not change after the addition of 10 nM estradiol (fig. S9). Consistent with the action of GABAA and GABAB agonists on Ca2+ signaling, muscimol and baclofen increased CBF, but only in Trpv4+/+ oviductal cells (Fig. 6D). In addition, the increase in CBF induced by baclofen was prevented in the presence of the GABAB antagonist CGP35348 (Fig. 6D). The change in CBF in response to P4 and its sensitivity to RU-486 showed a different pattern when the CBF was measured 12 hours after the addition of the hormone (Fig. 6E). Under these experimental conditions, P4 reduced CBF and this effect was reversed by the addition of RU-486, supporting previous reports demonstrating that P4 reduces CBF in a PR-dependent manner at times ≥20 min (11).


Here, we sought to determine whether P4 regulated CBF through the modulation of Ca2+ signaling in ciliated oviductal cells. The combination of experiments with native mouse oviductal ciliated cells and with heterologous expression of GABAA, GABAB, and TRPV4 reported the following observations: (i) the contribution of GABAA, GABAB1, and TRPV4 to the P4-stimulated Ca2+ responses; (ii) the physical interaction between the GABAA and GABAB1 subunits; and (iii) the GABA receptor–dependent activation of G proteins by P4. Heterologous expression in HEK 293 cells also enabled us to add a further element to this molecular mechanism: the need for TRPV4 to be in close proximity to the IP3R. Ca2+ entry is required for continuous ER-dependent Ca2+ oscillations, and TRPV4 plays a role in such process (31, 32). Therefore, basal TRPV4-mediated Ca2+ entry in close proximity to the IP3R seems to be required to initiate and maintain the Ca2+ signals triggered by P4, a process that was lost when the TRPV4-IP3R interaction was prevented in cells expressing the mutant TRPV4ΔCaM, which lacks the IP3R-binding site.

Our findings suggest two patterns of action of P4 regarding the control of CBF in mouse ciliated oviductal cells (Fig. 7): a classical, P4 receptor–dependent, long-term effect, which is characterized by a decrease in CBF, and a short-term effect, which is characterized by the increase in intracellular Ca2+ concentration and CBF that is mimicked by the addition of GABA receptor agonists. The data also highlight a previously unknown molecular mechanism for the rapid effect of P4 in mouse ciliated oviductal cells that involves the orchestrated action of GABA receptors and the TRPV4 channel to generate Ca2+ signals and accelerate CBF. Note that female GabaB1−/− mice exhibit fertility problems that are characterized by a reduced percentage of pregnancies within the first 30 days of mating and a lengthened interval between the exposure of female mouse to a male mouse and the delivery of the first litter, although this difference did not attain statistical significance when compared with wild-type mice (67). A detailed analysis of the different fertility parameters of Trpv4−/− mice may be worth exploring in future studies.

Fig. 7 Model of short- and long-term modulation of CBF by progesterone.

Progesterone uses two signaling pathways to modulate CBF in ciliated oviductal cells. On the one hand, progesterone generates a rapid (seconds to minutes) response (blue arrows), which is initiated by its interaction with GABAA and is followed by the transactivation of GABAB and the subsequent release of Ca2+ from the ER through a Gi/o-PLC-IP3 pathway. Ca2+ entry through the channel TRPV4, which is facilitated by the close proximity of TRPV4 and the IP3R, is required for the progesterone-induced, ER-dependent Ca2+ oscillations and the subsequent increase in CBF. GABA receptor agonists (baclofen and muscimol) mimic the progesterone effect. On the other hand, long-term (hours) exposure to progesterone decreases CBF through a classical PR (purple arrows). The molecular details of this signaling pathway are unknown at present. PIP2, phosphatidylinositol 4,5-bisphosphate.

As previously reported (11), the long-term P4-mediated reduction in CBF required the participation of classical P4 receptors. The molecular mechanism of the P4-mediated reduction in CBF has not yet been delineated but may involve inhibition of Trpv4 expression by classical P4 receptors (68). However, and at odds with the report by Bylander et al. (11), rather than observing a reduction in CBF at short times (<20 min) after P4 administration, we observed an increase. The apparent contradiction between these two studies may be related to the age of the animals used: Immature mice were used in the other study, whereas we used adult mice in the present study. How age may affect the CBF response of mouse ciliated oviductal cells to P4 has not yet been studied but may be linked to changes in the abundances of GABA receptors that occur during maturation and aging (69, 70).

The physical interaction and cross-activation of GABAB by GABAA adds to the list of described interactions between receptors belonging to different families, including GABAA, purinergic P2X, and dopamine D1 and D5 receptors (71, 72). However, unlike these examples, the interaction we described between GABAA and GABAB receptors is destined because (i) it did not require a functional ionotropic GABAA receptor and (ii) it involved only GABAB1, and not GABAB2, subunits. The stoichiometry of the GABAA/GABAB complex and the question of whether other plasma membrane Ca2+-permeable channels can substitute for TRPV4 in this functional complex remain to be investigated.



All chemicals were purchased from Sigma-Aldrich except for fura-2 AM, which was obtained from Molecular Probes, Invitrogen. The isotonic bathing solutions used for imaging experiments contained 140 mM NaCl, 5 mM KCl, 1.2 mM CaCl2, 0.5 mM MgCl2, 5 mM glucose, and 10 mM Hepes (pH 7.4) and were 300 mosmol/liter. Ca2+-free extracellular solutions were obtained by replacing CaCl2 with MgCl2 and adding 0.5 mM EGTA. Primers used for RT-PCR analysis are listed in tables S1 and S2. Whole-brain extracts were used as positive controls. CBF measurements were performed in an isotonic extracellular solution. For electrophysiology bathing solutions, CaCl2 was removed and 1 mM MgCl2 and 1 mM EGTA were added.


Adult (10- to 14-week-old) Trpv4+/+ and Trpv4−/− littermate mice generated on a C57Bl6/J background were used (73). Primary cultures were prepared as previously described (31). Animals were maintained, and the experiments were performed according to the guidelines issued by the Institutional Ethics Committees of the Institut Municipal d’Investigació Mèdica/Universitat Pompeu Fabra. HEK 293T and HEK 293WSS1 cells were transfected with different combinations of the following complementary DNAs (cDNAs): HA-GABAB1a-YFP-pRK6 or myc-GABAB1a-pCI; HA-GABAB2-YFP-pCI or HA-GABAB2-pCI, TRPV4-pCDNA3.

Measurement of intracellular Ca2+ concentration

Cytosolic Ca2+ signals were determined at room temperature (~24°C, unless otherwise indicated) in ciliated cells loaded with 4.5 μM fura-2 AM for 45 min, as previously described in detail (31). Changes in the intracellular Ca2+ concentration are presented as the ratio of emitted fluorescence at 510 nm after dual excitation at 340 and 380 nm relative to the ratio measured before stimulation (ratio, 340:380)

Measurement of CBF

The CBF of cultured ciliated cells was detected and quantified with high-speed digital imaging, as previously described (32). Phase-contrast images (512 × 512 pixels) were collected at 130 frames s−1 with a high-speed charge-coupled device camera using a frame grabber and recording software from Video Savant (IO Industries). The CBF was determined from the variation in the light intensity of the image that resulted from the repetitive motion of cilia. Video recordings of beating cilia for a duration of 2 s were analyzed using Video Savant software, and the frequency of each ciliary beat cycle was determined from the period of each gray-intensity waveform cycle.


Mouse oviducts were fixed overnight at 4°C with 4% paraformaldehyde, cryopreserved with 30% sucrose for 24 hours at 4°C, and embedded into optimal cutting temperature (OCT) for sectioning with a cryostat (10-μm sections). The rabbit polyclonal anti-TRPV4 antibody was used at a final concentration of 6.4 μg/ml together with GABA-specific antibodies (rabbit polyclonal anti-GABAAα2 from Alomone Labs and rabbit polyclonal R-300 GABAB1 from Santa Cruz Biotechnology) at a 1:100 dilution. The immunogenic peptide for the GABAAα2 antibody was provided by Alomone Labs and used at an antibody-peptide ratio of 1:1. For immunodetection, we used an Alexa Fluor 488–conjugated goat anti-rabbit immunoglobulin G (IgG) (Molecular Probes) and an Alexa Fluor 555–conjugated goat anti-mouse IgG (Molecular Probes) diluted at 1:750 in the same solution used with the primary antibodies. Images were captured with an inverted Leica SP2 confocal microscope, using an HCX PL APO 40× 1.25 Oil Ph3 CS objective, LCS Leica Confocal software, and an argon (488 nm, JDS Uniphase Corporation) laser. Original images were not further processed except for adjustments of brightness, contrast, and color balance. For Western blotting analysis, proteins were resolved on a precast 4 to 12% polyacrylamide gel NuPAGE (Invitrogen) and transferred to nitrocellulose membranes using a dry blotting system (iBlot, Invitrogen) according to standard protocols. Detection of immunoreactive bands was performed with SuperSignal West chemiluminescent substrate (Pierce).

Immunoprecipitation assay

Mouse oviducts were removed, homogenized, and resuspended in radioimmunoprecipitation assay buffer (25 mM tris-HCl, 150 mM NaCl, 1% NP-40, 1% sodium deoxycholate, 0.1% SDS) containing 10% protease inhibitor cocktail (Roche). Insoluble material was removed by centrifugation at 12,000g for 20 min at 4°C. Soluble proteins (about 500 μg per condition) were incubated overnight with protein G magnetic Dynabeads (Thermo Fisher) coated with antibodies against GABAAα2 (rabbit polyclonal from Alomone Labs), GABAB1 (rabbit polyclonal R-300 from Santa Cruz Biotechnology), and TRPV4 (rabbit polyclonal) and incubated overnight at 4°C. Immunoprecipitates were then washed four times with lysis buffer, resuspended in Laemmli sample buffer with 5% β-mercaptoethanol, and boiled for 7 min. The samples were subjected to SDS–polyacrylamide gel electrophoresis using an 8% gel and transferred to nitrocellulose membranes using a dry blotting system (iBlot, Invitrogen). Membranes were blocked for 1 hour in tris-buffered saline, 0.1% Tween 20 containing 5% skim milk. Membranes were then incubated with antibody against GABAB1 at a 1:400 dilution. A donkey anti-rabbit secondary antibody (GE Healthcare) was used at a 1:2000 dilution.

FRET and BRET experiments

FRET and BRET experiments were performed as previously described (53, 74). FRET between CFP and YFP in cells expressing the corresponding constructs was determined by donor recovery after acceptor bleaching; if FRET occurs, photobleaching of the acceptor (YFP) yields a substantial increase in the fluorescence of the donor (CFP). BRET measurements were performed in HEK 293T cells transiently transfected with plasmids encoding GABAB1 and GABAB2, pcDNA3-Gαi/o1-RLuc8, pcDNA3.1-Gγ2-Venus, and pcDNA3.1-FLAG-Gβ2 or with plasmid encoding the α2-AR, Gαi/o-RLuc, Gγ2-Venus, and Gβ2. BRET signals between Gαi/o-RLuc and Gγ2-Venus in the presence of 5 mM coelenterazine h (NanoLight Technology) were measured on a POLARstar OPTIMA plate reader (BMG Labtech) after receptor activation with baclofen, P4, or noradrenaline, as appropriate.

RT-PCR analysis

Total RNA from HEK 293T cells was isolated with the NucleoSpin RNA Isolation Kit (Macherey-Nagel). cDNAs were synthesized with the SuperScript First-Strand Synthesis System for RT-PCR (Invitrogen). Real-time PCR was performed using primers specific for the human GABA receptors. Actb was used as an internal control for the quantification of gene expression. Real-time PCR was performed with the SYBR Green PCR Master Mix (Applied Biosystems) in a 7900 HT Sequence Detection System (Applied Biosystems). Relative mRNA abundance in triplicate samples was calculated using the ΔΔCT method.

Statistical analysis

All data are presented as means ± SEM. Statistical analysis was performed using GraphPad InStat or SigmaPlot software, applying either Student’s paired or unpaired t tests and one-way ANOVA followed by the Bonferroni post hoc test, as appropriate, when comparing data that followed normal distributions or Mann-Whitney’s unpaired t test and nonparametric ANOVA (Kruskal-Wallis) followed by Dunn’s post hoc test when data did not fit a normal distribution. The criterion for a statistically significant difference was P < 0.05.


Fig. S1. Effects of P4 and 20αDHP4 on the intracellular Ca2+ concentration in mouse ciliated oviductal cells.

Fig. S2. Cationic and anionic currents in HEK 293 cells and mouse ciliated oviductal cells.

Fig. S3. Detection of GABAA and GABAB subunit mRNAs in the mouse oviduct by RT-PCR analysis.

Fig. S4. Effect of GABA receptor antagonists on P4-stimulated Ca2+ signals in mouse ciliated oviductal cells.

Fig. S5. Immunodetection of GABAAα2, GABAB1, and TRPV4 in the mouse oviduct.

Fig. S6. HEK 293T cell dFRAP analysis.

Fig. S7. Expression and activity of GABAA and GABAB subunits in HEK 293 cells.

Fig. S8. CBF measurements in mouse oviductal cells.

Fig. S9. Analysis of the effect of E2 on CBF.

Table S1. Primers used to detect mouse GABA receptor subunits by RT-PCR.

Table S2. Primers used to detect human GABA receptor subunits by RT-PCR.


Acknowledgments: We thank W. Liedtke (Duke University) for the Trpv4−/− mice, B. Bettler (University of Basel, Switzerland) for the GABAB cDNAs, and I. Sarto-Jackson (Konrad Lorenz Institute, Klosterneuburg, Austria) for the GABAA cDNAs. Funding: This work was supported by the Spanish Ministry of Economy and Competitiveness (SAF2015-69762-R, SAF2014-55700-P, and PCIN-2013-019-C03-03), Fondo de Investigación Sanitaria (PIE14/00034), FEDER (Fondos Europeos de Desarrollo Económico y Regional) Funds, and Animal Welfare Trust (SBO-140028). Author contributions: C.J., J.M.F.-F., and M.A.V. designed the research; C.J., V.F.-D., C.P., A.G.-E., and J.M.F.-F. performed the research; C.J., V.F.-D., A.G.-E., F.C., and M.A.V. analyzed the data; M.A.V. wrote the manuscript; and all authors collaborated in editing the manuscript. Competing interests: The authors declare that they have no competing interests. Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper or the Supplementary Materials.
View Abstract

Navigate This Article