Store-operated Ca²⁺ entry is not required for fertilization-induced Ca²⁺ signaling in mouse eggs

Abstract

Repetitive oscillations in cytoplasmic Ca²⁺ due to periodic Ca²⁺ release from the endoplasmic reticulum (ER) drive mammalian embryo development following fertilization. Influx of extracellular Ca²⁺ to support the refilling of ER stores is required for sustained Ca²⁺ oscillations, but the mechanisms underlying this Ca²⁺ influx are controversial. Although store-operated Ca²⁺ entry (SOCE) is an appealing candidate mechanism, several groups have arrived at contradictory conclusions regarding the importance of SOCE in oocytes and eggs. To definitively address this question, Ca²⁺ influx was assessed in oocytes and eggs lacking the major components of SOCE, the ER Ca²⁺ sensor STIM proteins, and the plasma membrane Ca²⁺ channel ORAI1. We generated oocyte-specific conditional knockout (cKO) mice for Stim1 and Stim2, and also generated Stim1/2 double cKO mice. Females lacking one or both STIM proteins were fertile and their ovulated eggs displayed normal patterns of Ca²⁺ oscillations following fertilization. In addition, no impairment was observed in ER Ca²⁺ stores or Ca²⁺ influx following store depletion. Similar studies were performed on eggs from mice globally lacking ORAI1; no abnormalities were observed. Furthermore, spontaneous Ca²⁺ influx was normal in oocytes from Stim1/2 cKO and ORAI1-null mice. Finally, we tested if TRPM7-like channels could support spontaneous Ca²⁺ influx, and found that it was largely prevented by NS8593, a TRPM7-specific inhibitor. Fertilization-induced Ca²⁺ oscillations were also impaired by NS8593. Combined, these data robustly show that SOCE is not required to support appropriate Ca²⁺ signaling in mouse oocytes and eggs, and that TRPM7-like channels may contribute to Ca²⁺ influx that was previously attributed to SOCE.

Graphical Abstract

1. INTRODUCTION

A large increase in the level of cytoplasmic Ca²⁺ is the key trigger for activation of development at fertilization in all animal species. In preparation for fertilization, mammalian germinal vesicle (GV)-intact, prophase I-arrested oocytes take up Ca²⁺ from the extracellular space and suppress Ca²⁺ leak from the endoplasmic reticulum (ER), resulting in a significant increase in ER Ca²⁺ stores as they resume meiosis and become arrested at metaphase II (MII eggs) [1–3]. In addition, they increase expression of the inositol 1,4,5-trisphosphate (IP₃) receptor and alter ER localization such that it becomes enriched in the cortical region below the plasma membrane [4–6]. Together, these and other maturation-associated changes result in MII eggs that efficiently release Ca²⁺ in response to the fertilizing sperm. The initial Ca²⁺ release event at fertilization is induced by the activity of the sperm-specific phospholipase C zeta (PLCζ), which enters the egg’s cytoplasm following sperm-egg plasma membrane fusion [7, 8]. PLCζ catalyzes hydrolysis of phosphatidylinositol 4,5-bisphosphate to generate diacylglycerol and IP₃; IP₃ then activates the IP₃ receptor to cause Ca²⁺ release from ER stores.

In mammals, periodic Ca²⁺ release from ER stores, known as Ca²⁺ oscillations, continue for several hours after fertilization. The persistence of these oscillations appears to be a critical driver of the success of embryo development [9]. In the absence of extracellular Ca²⁺, reductions in intraluminal ER Ca²⁺ levels that occur with each release event lead to cessation of the oscillations over time [10–12]. Hence, it is apparent that Ca²⁺ influx from the extracellular space, followed by accumulation into the ER, must occur during both oocyte maturation and following fertilization to support effective initial Ca²⁺ release, continued Ca²⁺ oscillations and optimal embryo development.

A common mechanism to replenish depleted ER Ca²⁺ stores via influx from extracellular sources is known as store-operated Ca²⁺ entry (SOCE) [13, 14]. The main molecular mediators of SOCE are STIM proteins, which sense ER Ca²⁺ levels, and ORAI proteins, which serve as plasma membrane Ca²⁺ channels. In mammals there are two STIM proteins, STIM1 and STIM2, and three ORAI proteins, ORAI1, ORAI2, and ORAI3. In response to a reduction in ER Ca²⁺, STIM proteins oligomerize and undergo redistribution within the ER to regions closely apposed to the plasma membrane (PM), known as ER-PM junctions. There, the STIM proteins interact directly with ORAI channels, inducing Ca²⁺ influx. This Ca²⁺ is subsequently pumped back into the ER by the action of sarco-endoplasmic reticulum Ca²⁺ ATPases (SERCA).

SOCE would be a logical mechanism to mediate Ca²⁺ entry following fertilization. Indeed, SOCE has been reported in both oocytes and MII eggs, though it appears more robust in oocytes [1, 15]. However, there is conflicting data in the literature regarding whether or not SOCE is an important physiological mediator of Ca²⁺ influx following fertilization. We and others demonstrated previously that chemical inhibitors of SOCE did not prevent Ca²⁺ influx following fertilization in the mouse [16, 17]. In contrast, several studies in both mouse and porcine oocytes suggest, instead, that SOCE is essential for this process [18–21].

To definitively determine whether SOCE is required to support Ca²⁺ influx at fertilization in the mouse, we generated and analyzed conditional knockout mouse lines in which the oocytes lacked the major SOCE components. Here we show that there is no requirement for either STIM1, STIM2, or ORAI1 to support Ca²⁺ influx during oocyte maturation or to support persistent Ca²⁺ oscillations following fertilization. Experiments to determine what alternative channels mediate Ca²⁺ influx in mouse oocytes uncovered compelling evidence that spontaneous Ca²⁺ influx in oocytes and post-fertilization Ca²⁺ influx, which were previously attributed to SOCE, are instead mediated by either TRPM7, the melastatin-related transient receptor potential channel, or a TRPM7-like channel.

2. MATERIALS AND METHODS

2.1. Mice

Stim1^flox/flox and Stim2^flox/flox mice [22] were crossed with a Zp3-cre transgenic line (Jackson Laboratory, Bar Harbor, ME, USA, Stock No. 003651) [23] to generate oocyte-specific conditional knockout (cKO) mice for Stim1 and Stim2. These mice were further intercrossed to obtain Stim1-Stim2 double cKO mice. All mice for these crosses were maintained on a predominantly C57Bl/6J genetic background, and the Zp3-cre transgene was bred through the male to avoid germline transmission of excised alleles. Orai1 full knockout mice [24] outbred with Institute of Cancer Research (ICR) strain mice [25] were also used. CF-1 strain female mice were obtained from Charles River (Wilmington, MA, USA). B6SJLF1 males for sperm collection and C57Bl/6J male breeders were from Jackson Laboratory. For analysis of fertility, littermate Stim1/2 cKO and Zp3-cre-negative Stim1^flox/flox;Stim2^flox/flox control females were paired with adult C57Bl/6J males at 7–12 weeks of age and were housed together as breeding pairs for 6 months. Pups were counted on the day of birth and sexed at weaning 21 days later. All mice were maintained under a standard 12 h light/dark cycle and provided food and water ad libitum. All studies and procedures were performed under an animal study protocol approved by the National Institutes of Health Animal Care and Use Committee in accordance with the Guide for the Care and Use of Laboratory Animals.

2.2. Gamete Collection

For collection of GV stage oocytes, female mice 6–15 weeks of age were injected with 5 IU pregnant mare serum gonadotropin (PMSG, Calbiochem) and sacrificed by CO₂ asphyxiation 42–48 h later. Ovaries were placed in Leibovitz L-15 medium (L-15, Life Technologies, Grand Island, NY, USA) containing 1% calf serum (Atlanta Biologicals, Norcross, GA, USA) and 10 μM milrinone, and antral follicles were punctured using 27 G needles. Cumulus cells were removed by gentle pipetting through a narrow bore capillary. For collection of metaphase II (MII)-arrested eggs, PMSG-primed mice were injected with 5 IU human chorionic gonadotropin (hCG) 46–48 h after PMSG injection and sacrificed 13–14 h later. Cumulus-oocyte masses were removed from oviducts and cumulus cells were removed by brief treatment with 0.03% hyaluronidase in L-15. Prior to imaging or collection, oocytes and eggs were held in Minimal Essential Medium Alpha (aMEM, Life Technologies) containing 5% calf serum or potassium simplex optimized medium (KSOM) with amino acids (EMD Millipore, Billerica, MA, USA) at 37° C/5% CO₂ in drops of pre-equilibrated medium under light mineral oil.

2.3. Immunoblotting

Immunoblot analysis was carried out essentially as previously described [26]. LF PVDF membranes (BioRad, Atlanta, GA, USA) were blocked in 5% non-fat dry milk in Tris-buffered saline with 0.1% Tween-20 (TBS-T). Primary antibodies directed against STIM1 and STIM2 were obtained from Cell Signaling (Danvers, MA, USA; AntibodyRegistry: AB_2271287 and AB_2198021) and were diluted 1:250 in blocking solution. Peroxidase-conjugated anti-rabbit secondary antibody from Jackson ImmunoResearch (West Grove, PA, USA; #711-035-152) was used at 1:10,000 dilution, and SuperSignal West Femto Chemiluminescent Substrate (ThermoFisher Scientific, Grand Island, NY, USA) was used for detection.

2.4. Real-time PCR

Quantitative real time PCR was performed essentially as previously described [27]. Samples were spiked with EGFP cRNA as an internal reference, as described in [16], and RNA was isolated from three groups of 30 GV oocytes each from Stim2^flox/flox (control) and Stim2^flox/flox;Zp3-cre⁺ (cKO) mice using an Arcturus PicoPure RNA Isolation Kit (Life Technologies). Reverse transcription and amplification were performed as previously described [26]. Primer sequences for detection of Stim2 were as follows: [Exon3–4 Forward: GGATTTGTGGAAACAGTGGAA; Reverse: GGGAGTGTTGTTCCCTTCAC], [Exon4–5 Forward: TGTGAAGGGAACAACACTCC; Reverse: GTCAGAGGCCCAAACAGAAC].

2.5. In vitro fertilization (IVF) and Ca²⁺ Imaging

Imaging of Ca²⁺ flux following fertilization was performed essentially as previously described [27]. Briefly, zona pellucida-free fura-2 AM (Life Technologies)-loaded eggs were adhered to Cell Tak (EMD Millipore)-treated glass-bottom dishes, and 510 nm fluorescence was imaged with excitation at 340 nm and 380 nm as 5 × 10⁴ capacitated sperm/mL was added to the imaging drop. Change in F340/F380 ratio was measured 8 times per minute and is used to reflect changes in intracellular Ca²⁺ ([Ca²⁺]_i). For all experiments using genetic mouse models, control and KO eggs were placed side-by-side in the same media drop and were imaged simultaneously. For experiments using inhibitor treatment, eggs were pre-incubated in 20 uM NS8593 for 15–30 min prior to sperm addition; parallel control experiments were performed in the same manner immediately before or after inhibitor studies using gametes from the same mice and vehicle (DMSO) treatment. Analysis of Ca²⁺ oscillation parameters was performed as previously described [27].

2.6. ER Ca²⁺ Store, Store-operated Ca²⁺ Entry, and Spontaneous Ca²⁺ Influx Assays

Measurement of Ca²⁺ release following addition of thapsigargin, reflecting ER Ca²⁺ store content, was performed essentially as described [27]. Briefly, zona pellucida-intact oocytes or eggs were adhered to glass-bottom dishes in 1.85 mL Ca²⁺-, Mg²⁺-, and bovine serum albumin (BSA)-free CZB medium [28] containing 2 mM EGTA. Solutions of thapsigargin, CaCl₂, or MgCl₂ diluted in 150 μL medium were added drop-wise to dishes during imaging. Cells were imaged for 3–5 min in Ca²⁺-free medium before adding thapsigargin at a final concentration of 10 μM; 35–40 min later, 5 mM CaCl₂ was added and imaging continued for at least 15 min. For spontaneous Ca²⁺ influx assays, Ca²⁺ influx was measured for GV oocytes without prior depletion of ER Ca²⁺ using thapsigargin [1]. GV oocytes were placed in Ca²⁺-, Mg²⁺-, and BSA-free CZB and imaged as CaCl₂ was added to a final concentration of 2 mM or 5 mM. Ten μM milrinone was included in medium used for GV oocytes to maintain meiotic arrest. Area under the curve for 10 minutes following each addition was calculated for individual oocytes or eggs by computing trapezoidal area for each curve following subtraction of the average baseline measurement for 2–3 min prior to the response.

2.7. Additional Chemicals and Reagents

If not otherwise specified, all chemicals, reagents, and inhibitors were obtained from Sigma Aldrich (St. Louis, MO, USA).

2.8. Statistical Analysis

GraphPad Prism, version 7.0a was used to perform all statistical tests. Student’s t-test was performed for normally distributed data, as determined by D’Agostino & Pearson omnibus normality test. For datasets lacking normal distribution, the Mann-Whitney U-test was used. Contingency data was analyzed using Fisher’s exact test. Asterisks indicate p < 0.05 and ns indicates non-significant differences in all graphs, and all error bars show SEM.

3. RESULTS

3.1. STIM1 and STIM2 are dispensable for Ca²⁺ influx in mouse oocytes and eggs

Oocyte-specific conditional knockout (cKO) mice for Stim1 and Stim2 were generated by crossing lines carrying loxP-flanked (flox) alleles with a Zp3-cre transgenic line [22, 23]. Eggs were collected from superovulated females for immunoblot analysis to confirm loss of STIM1 and STIM2 proteins. STIM1 was detected in a lysate of 200 eggs from Stim1^flox/flox (control) mice but was absent from eggs of Stim1^flox/flox;Zp3-cre⁺ (cKO) mice, confirming that Zp3-cre causes efficient loss of STIM1 protein in Stim1 cKO eggs (Fig. 1A). Although a specific band for STIM2 was observed at approximately 115 kD in a lysate of control neonatal lung tissue and was absent from Stim2 knockout tissue, we were unable to detect STIM2 protein by immunoblot using 260 eggs (Fig. 1B). However, using real time PCR, we found that Stim2 transcript levels were greatly reduced in Stim2 cKO eggs, with transcripts containing the loxP-flanked exon 3 reduced by 68-fold, becoming nearly undetectable (Fig. 1C). We conclude that efficient loss of both STIM1 and STIM2 was achieved using this method.

Figure 1. STIM1 and STIM2 expression in control and cKO eggs.

(A) Representative immunoblot of STIM1 protein in Stim1 cKO and control (con) eggs. 200 eggs per lane; blot is one of three independent replicates. Arrowhead indicates STIM1 protein band; asterisk indicates non-specific band. (B) Representative immunoblot of STIM2 protein in control and Stim2 cKO eggs; 260 eggs/lane. Lung protein extract (5 μg) from either control or Stim2^−/− postnatal day 1 pups was used for positive and negative controls. Arrowhead indicates STIM2 band. (C) Real time qPCR of Stim2 mRNA in control and Stim2 cKO eggs. Schematic shows Stim2 exons (blue), loxP sites (red triangles), and primer pairs (arrows). Graphs show expression relative to that in control eggs for the indicated primer pair.

To determine whether STIM1 is necessary for normal Ca²⁺ signaling following fertilization, we performed live-cell ratiometric Ca²⁺ imaging on control eggs from Stim1^flox/flox females alongside cKO eggs lacking STIM1 from Stim1^flox/flox;Zp3-cre⁺ females during in vitro fertilization. Representative traces for 60 minutes of imaging are shown, along with analysis of data from 5 replicate experiments (Fig. 2A–E). None of the parameters measured, including first transient length, maximum amplitude, oscillation frequency, and persistence of oscillations to 60 minutes, differed between control and Stim1 cKO eggs. Of note, the oscillation frequency we observe is slightly faster than that reported in some previous studies (E.g., [16, 29]), but is consistent with our previous data using the same method [27]. We attribute this variance primarily to differences in media composition, use of intracytoplasmic sperm injection vs. IVF, and genetic background of mice used in these studies. Based on our analysis of Ca²⁺ oscillation parameters, STIM1 is not required for appropriate Ca²⁺ signaling following fertilization.

Figure 2. Ca²⁺ oscillation patterns following IVF and ER Ca²⁺ stores in eggs lacking STIM1.

Ratiometric Ca²⁺ imaging of control and Stim1 cKO eggs was performed during IVF. (A) Representative traces are shown. (B–E) Analysis of Ca²⁺ oscillation patterns, n=30 eggs per genotype over 5 replicate experiments. (B) Length of first transient. (C) Amplitude of first transient. (D) Oscillation frequency. (E) Persistence of oscillations to the end of the 60 min imaging period. (F–J) Measurements of ER Ca²⁺ stores and Ca²⁺ influx 35–40 min after thapsigargin (Tg)-induced ER store depletion in Stim1 cKO and control eggs. (F) Representative traces (gray) and mean curve (red) are shown for one experiment. Timing of Tg and Ca²⁺ addition indicated in gray and black bars. (G–J) Analysis of Ca²⁺ stores and influx, n=24–33 eggs per genotype over 4 replicate experiments. (G,H) Graphs of ER Ca²⁺ stores indicated by area under the curve (G) and peak amplitude (H). * p < 0.05, unpaired t test. (I,J) Graphs of Ca²⁺ entry following store depletion indicated by area under the curve (I) and peak amplitude (J).

We next investigated whether loss of STIM1 impacts Ca²⁺ stores or Ca²⁺ influx following store depletion. Ca²⁺ imaging was performed on control and Stim1 cKO eggs placed side-by-side in Ca²⁺-free medium, and thapsigargin was added to inhibit SERCA-mediated refilling of Ca²⁺ stores and allow leak of Ca²⁺ from the ER (Fig. 2F). Area under the curve and peak amplitude were measured as indicators of ER Ca²⁺ content. Peak amplitude was similar and area under the curve was slightly higher for Stim1 cKO eggs compared to control (Fig. 2G,H), indicating that absence of STIM1 did not impair accumulation of ER Ca²⁺ in eggs. Thapsigargin treatment was continued for at least 30 min to deplete ER stores, then 5 mM CaCl₂ was added and the increase in [Ca²⁺]_i was measured for 10 minutes as an indication of Ca²⁺ influx following store depletion. Neither peak amplitude nor area under the curve was significantly different for Stim1 cKO eggs compared to controls (Fig. 2I,J). Overall, these results indicate that STIM1 is dispensable for normal Ca²⁺ oscillations and homeostasis in mouse eggs.

Whereas STIM1 is activated by substantial release of Ca²⁺ from ER stores, STIM2 can be activated by alterations in ER Ca²⁺ closer to the resting level [30, 31]. We therefore also sought to determine whether STIM2 was necessary for Ca²⁺ signaling in mouse eggs. As for the Stim1 cKO model, we performed Ca²⁺ imaging during in vitro fertilization for control and Stim2 cKO eggs. Representative traces and summary analyses are shown in Fig. 3A–E. Ca²⁺ oscillations of Stim2 cKO eggs appeared largely normal. Oscillation frequency and amplitude were unaltered, and only very subtle decreases were observed in length of the first Ca²⁺ transient and in the proportion of eggs with oscillations persisting throughout the 60 min imaging period. There was no significant difference in either ER Ca²⁺ store content or Ca²⁺ influx following store depletion (Fig. 3F–I). Based on these data, loss of STIM2 did not impair Ca²⁺ store accumulation or influx and had at most a very minor impact on fertilization-induced oscillations.

Figure 3. Ca²⁺ oscillation patterns following IVF and ER Ca²⁺ stores in eggs lacking STIM2.

Ratiometric Ca²⁺ imaging of control and Stim2 cKO eggs was performed during IVF. (A) Representative traces are shown. (B–E) Analysis of Ca²⁺ oscillation patterns, n=36–45 eggs per genotype over 6 replicate experiments. (B) Length of first transient. * p < 0.05, Mann Whitney. (C) Amplitude of first transient. (D) Oscillation frequency. (E) Persistence of oscillations to the end of the 60 min imaging period. * p < 0.05, Fisher’s exact test. (F–I) Measurements of ER Ca²⁺ stores and Ca²⁺ influx 35–40 min after thapsigargin (Tg)-induced ER store depletion in Stim2 cKO and control eggs, n=23 eggs per genotype over 4 replicate experiments. (F,G) Graphs of ER Ca²⁺ stores indicated by area under the curve (F) and peak amplitude (G). (H,I) Graphs of Ca²⁺ entry following store depletion indicated by area under the curve (I) and peak amplitude (J).

Although loss of STIM1 and STIM2 individually had very little impact on oocyte physiology and Ca²⁺ signaling, we also wanted to address the possibility of compensation between STIM1 and STIM2. Stim1 and Stim2 single cKO lines were intercrossed to produce a Stim1-Stim2 oocyte-specific double conditional knockout line (Stim1/2 cKO). Fertilization-induced Ca²⁺ oscillations of Stim1/2 cKO eggs were no different from controls (Fig. 4A–E) and ER Ca²⁺ stores measured by thapsigargin-mediated Ca²⁺ release were essentially identical to controls (Fig. 4F,G). Likewise, Ca²⁺ influx following store depletion was not significantly altered in Stim1/2 cKO eggs (Fig. 4H,I). It was previously shown that Ca²⁺ influx following store depletion is greater in GV oocytes than MII eggs [1]; therefore, we also assayed ER Ca²⁺ stores and Ca²⁺ influx in Stim1/2 cKO GV oocytes. Following thapsigargin addition, similar levels of Ca²⁺ release were observed in control and Stim1/2 cKO GV oocytes (Fig. 4J,K), and Ca²⁺ influx following store depletion was also unaltered (Fig. 4L,M). Taken together, these data indicate that in mouse oocytes and eggs neither STIM1 nor STIM2 is required for Ca²⁺ influx following ER store depletion.

Figure 4. Analysis of MII eggs and GV oocytes lacking both STIM1 and STIM2.

Ratiometric Ca²⁺ imaging of control and Stim1/2 cKO eggs was performed during IVF. (A) Representative traces are shown. (B–E) Analysis of Ca²⁺ oscillation patterns, n=28–31 eggs per genotype over 4 replicate experiments. (B) Length of first transient. (C) Amplitude of first transient. (D) Oscillation frequency. (E) Persistence of oscillations to the end of the 60 min imaging period. (F–I) Measurements of ER Ca²⁺ stores and Ca²⁺ influx 35–40 min after thapsigargin (Tg)-induced ER store depletion in Stim1/2 cKO and control eggs, n=23 eggs per genotype over 4 replicate experiments. (F,G) Graphs of ER Ca²⁺ stores indicated by area under the curve (F) and peak amplitude (G). * p < 0.05, Mann Whitney. (H,I) Graphs of Ca²⁺ entry following store depletion indicated by area under the curve (H) and peak amplitude (I). (J–M) Measurements of ER Ca²⁺ stores and Ca²⁺ influx 35–40 min after thapsigargin (Tg)-induced ER store depletion in Stim1/2 cKO and control GV oocytes, n=23–24 oocytes per genotype over 4 replicate experiments. (J,K) Graphs of ER Ca²⁺ stores indicated by area under the curve (J) and peak amplitude (K). (L,M) Graphs of Ca²⁺ entry following store depletion indicated by area under the curve (L) and peak amplitude (M). (N–P) Fertility of female Stim1/2 cKO mice. (N) Average litter size. (O) Days to first litter. (P) Total number of live pups.

To determine whether oocyte-specific loss of STIM1 and STIM2 had any impact on fertility, we conducted a formal breeding study for Stim1/2 double cKO females. Stim1/2 cKO females were fertile, and fertility parameters including average litter size, days to first litter, and total live pups per breeding pair were indistinguishable from those of littermate controls (Fig. 4N–P). One Stim1/2 cKO female did not produce pups during the study and was omitted from analysis because male infertility could not be excluded. We also housed several Stim1 and Stim2 single cKO females with males to confirm their fertility; all produced pups (4/4 females for each genotype), but a detailed breeding study was not carried out because this was done with the Stim1/2 cKO. Altogether we conclude that STIM1 and STIM2 are dispensable for normal fertility in female mice.

3.2 ORAI1 is dispensable for Ca²⁺ signaling in oocytes and eggs

Female Orai1 knockout mice are fertile [32]; however, Ca²⁺ signaling following fertilization has not been studied in eggs lacking ORAI1. As for Stim1 and Stim2 cKO mice, we performed Ca²⁺ imaging during in vitro fertilization and found no significant difference in the pattern of Ca²⁺ oscillations between Orai1 KO and wild-type control eggs (Fig. 5A–E). Similarly, there was no difference in ER Ca²⁺ stores (Fig. 5F,G) or Ca²⁺ influx following store depletion (Fig. 5H,I) for Orai1 KO MII eggs compared to controls. Likewise, ER Ca²⁺ stores in Orai1 KO GV oocytes were not significantly different from controls (Fig. 5J,K). Ca²⁺ influx following store depletion was increased in Orai1 KO eggs (Fig. 5L,M), in contrast to the expected decrease in influx that would be anticipated if ORAI1 channels were critical for mediating this influx. Thus, ORAI1 channels are also dispensable for normal Ca²⁺ signaling in oocytes and eggs.

Figure 5. Analysis of MII eggs and GV oocytes lacking ORAI1.

Ratiometric Ca²⁺ imaging of control and Orai1 KO eggs was performed during IVF. (A) Representative traces are shown. (BE) Analysis of Ca²⁺ oscillation patterns, n=20–24 eggs per genotype over 4 replicate experiments. (B) Length of first transient. (C) Amplitude of first transient. (D) Oscillation frequency. (E) Persistence of oscillations to the end of the 60 min imaging period. (F–I) Measurements of ER Ca²⁺ stores and Ca²⁺ influx 35–40 min after thapsigargin (Tg)-induced ER store depletion in Orai1 KO and control eggs, n=18–25 eggs per genotype over 3–4 replicate experiments. (F,G) Graphs of ER Ca²⁺ stores indicated by area under the curve (F) and peak amplitude (G). (H,I) Graphs of Ca²⁺ entry following store depletion indicated by area under the curve (H) and peak amplitude (I). (J–M) Measurements of ER Ca²⁺ stores and Ca²⁺ influx 40 min after thapsigargin (Tg)-induced ER store depletion in Orai1 KO and control GV oocytes, n=22–23 oocytes per genotype over 4 replicate experiments. (J,K) Graphs of ER Ca²⁺ stores indicated by area under the curve (J) and peak amplitude (K). (L,M) Graphs of Ca²⁺ entry following store depletion indicated by area under the curve (L) and peak amplitude (M). * p < 0.05, unpaired t test.

3.3. TRPM7-like channels, not SOCE components, support spontaneous Ca²⁺ influx in mouse oocytes

When placed in nominal Ca²⁺-free medium, GV oocytes respond to addition of 2–5 mM Ca²⁺ by showing a rise in [Ca²⁺]_i, reflecting measurable spontaneous Ca²⁺ influx, while MII eggs do not show such a response [1]. To investigate whether SOCE mediators STIM1, STIM2, or ORAI1 contribute to this spontaneous Ca²⁺ influx in GV oocytes, we placed Stim1/2 cKO and control oocytes in Ca²⁺-free medium and added 2 mM CaCl₂ during Ca²⁺ imaging. Both control and Stim1/2 cKO oocytes displayed increased [Ca²⁺]_i in response to Ca²⁺ addition (Fig. 6A), and no significant difference was observed in their response when peak amplitude and area under the curve for 10 min following Ca²⁺ addition were measured (Fig. 6B,C). Orai1 KO oocytes, as compared to wild-type controls, also had no difference in peak amplitude or area under the curve following Ca²⁺ addition (Fig. 6D,E). Based on these results, spontaneous Ca²⁺ influx in GV stage oocytes does not require STIM1, STIM2, or ORAI1.

Figure 6. Spontaneous Ca²⁺ influx in mouse GV oocytes and post-fertilization Ca²⁺ influx in mouse eggs are mediated by TRPM7-like channels.

(A–E) Ratiometric Ca²⁺ imaging of control and Stim1/2 cKO or Orai1 KO GV oocytes treated with 2 mM Ca²⁺ 3 min after placement into Ca²⁺-free medium. (A) Representative traces (gray) and mean curve (red) are shown for one experiment. Timing of Ca²⁺ addition indicated by black bars. (B–E) Analysis of spontaneous Ca²⁺ influx in control and Stim1/2 cKO or Orai1 KO oocytes, n=18–26 oocytes per genotype over 3–4 replicate experiments. (B) Stim1/2 cKO, area under the curve. (C) Stim1/2 cKO, peak amplitude. (D) Orai1 KO, area under the curve. (E) Orai1 KO, peak amplitude. (F–I) Effect of TRPM7 channel inhibition on spontaneous Ca²⁺ influx in control GV oocytes. (F,G) Ratiometric Ca²⁺ imaging of control oocytes treated with either vehicle (DMSO) or 20 μM NS8593 and then tested for spontaneous Ca²⁺ influx by addition of 5 mM Ca²⁺. 10 mM Mg²⁺ was added 18 min following Ca²⁺ addition. (H,I) Graphs of spontaneous Ca²⁺ entry, n=39–46 oocytes per group over 4 replicate experiments. (H) Area under the curve. (I) Peak amplitude. * p < 0.05, Mann Whitney. (J–N) Ratiometric Ca²⁺ imaging of wild-type CF-1 eggs during IVF in medium containing vehicle (DMSO) or 20 μM NS8593. J) Representative traces are shown. (K–N) Analysis of Ca²⁺ oscillation patterns, n=24–28 eggs per genotype over 3 replicate experiments. (K) Length of first transient. * p < 0.05, Mann Whitney. (L) Amplitude of first transient. * p < 0.05, Mann Whitney. (M) Oscillation frequency. * p < 0.05, unpaired t test. (N) Persistence of oscillations to the end of the 60 min imaging period. * p < 0.05, Fisher’s exact test.

We next addressed the question of what SOCE-independent mechanisms of Ca²⁺ influx are at work in mouse oocytes. We recently demonstrated that Ca_V3.2 channels contribute to the increase in ER Ca²⁺ stores during oocyte maturation and support Ca²⁺ oscillations following fertilization; however, it is evident that additional channels also support Ca²⁺ influx in oocytes and eggs [27]. We hypothesized that transient receptor potential (TRP) channels were likely candidates for this role because these cation conducting channels are expressed in all types of cells and can contribute to a wide array of cellular responses (reviewed in [33]). TRPV3 current has been demonstrated in mouse eggs, but this current was not detected in GV oocytes [34]; therefore, TRPV3 is unlikely to mediate spontaneous Ca²⁺ influx in GV oocytes. Mice lacking all seven TRPC channels have been generated and shown to be viable and fertile [35]. In a limited set of experiments, we also observed normal Ca²⁺ oscillations in eggs from Trpc hepta-KO mice (not shown), indicating that TRPC channels are unlikely to contribute significantly to Ca²⁺ influx. Through surveying publicly available microarray data ([36], GEO Accession GDS3295), we found that in mouse oocytes, Trpm7 was highly expressed relative to other Trp channels, and there is electrophysiological evidence for the presence of functional TRPM7-like channels in oocytes [37]. To test whether TRPM7-like channels support spontaneous Ca²⁺ influx in mouse oocytes, we used two different TRPM7 inhibitors, Mg²⁺ and NS8593, which was initially characterized as a small conductance Ca²⁺ activated K⁺ channel inhibitor [38–40]. GV oocytes from wild-type CF-1 females were placed in Ca²⁺- and Mg²⁺-free medium, and 5 mM CaCl₂ was added, causing an increase in [Ca²⁺]_i (Fig. 6F). After 15 minutes, 10 mM MgCl₂ was added to the medium; this treatment caused a substantial drop in [Ca²⁺]_i (Fig. 6F). In parallel experiments, GV oocytes were placed in Ca²⁺- and Mg²⁺-free medium containing 20 μM NS8593 for 15 min prior to CaCl₂ addition. In NS8593-treated oocytes, Ca²⁺ addition caused a substantially lower increase in [Ca²⁺]_i relative to controls, and subsequent Mg²⁺ addition did not further decrease [Ca²⁺]_i (Fig. 6G). Both area under the curve and peak amplitude following Ca²⁺ addition were significantly decreased in the presence of NS8593 (Fig. 6H,I). These results indicate that TRPM7-like channels contribute significantly to spontaneous Ca²⁺ influx in GV oocytes.

3.4. TRPM7-like channels contribute to Ca²⁺ influx following fertilization in mouse eggs

To test whether TRPM7-like channels also contribute to Ca²⁺ influx necessary to sustain post-fertilization Ca²⁺ oscillations, wild-type CF-1 eggs were treated with 20 μM NS8593 and fertilized in vitro (Fig. 6J). Compared to controls, NS8593-treated eggs had impaired Ca²⁺ oscillations, with significantly shorter first Ca²⁺ transients, reduced peak amplitude, lower frequency, and reduced persistence (Fig. 6K–N). These results suggest that TRPM7 or TRPM7-like channels also contribute to Ca²⁺ influx needed to maintain robust Ca²⁺ oscillations following fertilization.

4. DISCUSSION

A rise in intracellular Ca²⁺ is a universal activator of embryo development, and Ca²⁺ influx is required to sustain Ca²⁺ oscillations following mammalian fertilization [10, 11, 41]; however, the question of how this influx occurs is not fully resolved. We have recently shown that Ca_V3.2 channels contribute to Ca²⁺ influx in mouse oocytes and eggs, but additional Ca²⁺ influx mechanisms must also be functional [27]. Because Ca²⁺ content of the ER decreases and Ca²⁺ influx increases with each Ca²⁺ oscillation [12, 42], SOCE is a plausible candidate mechanism for mediating Ca²⁺ influx following fertilization. In this study we show that three SOCE mediators expressed in oocytes, STIM1, STIM2, and ORAI1, are dispensable for proper Ca²⁺ signaling and homeostasis in mouse oocytes and eggs. Our results are consistent with previous data showing that pharmacological inhibition of SOCE or expression of dominant negative SOCE components does not inhibit fertilization-induced oscillations in mouse eggs [16, 17].

Despite evidence to the contrary, there is a persistent view that SOCE is the primary mechanism of oocyte Ca²⁺influx. The idea that SOCE is critical in oocyte physiology has even appeared in review literature [43–45]. This may not be surprising because SOCE participation in fertilization seems logical, and because roles for SOCE in mouse and pig oocytes have been suggested through experiments using overexpression, knockdown, and inhibitor strategies [18–21]. However, our current data show that critical mediators of SOCE are expendable for fertilization and Ca²⁺ signaling in mouse eggs. We show that the only two STIM proteins characterized to date, STIM1 and STIM2, are both dispensable in mouse oocytes and eggs. We also show that loss of ORAI1 does not impair Ca²⁺ signaling in mouse oocytes and eggs. Based on microarray data, Orai1 is by far the most highly expressed transcript of the three Orai genes in mouse oocytes [36]; however, ORAI1 protein is internalized during meiosis so it is unlikely to function as a Ca²⁺ influx channel in eggs [15]. Results of prior studies supporting the actions of SOCE in mammalian eggs could have arisen in part from off-target effects of siRNA or morpholino oligonucleotides; however, it is likely that important differences between species also exist. For example, porcine oocytes may regulate ER Ca²⁺ differently than mouse oocytes and could therefore rely more heavily on SOCE. This interpretation is supported by experiments showing that the same SOCE inhibitors that halt Ca²⁺ oscillations in porcine eggs do not impair oscillations in mouse eggs [16, 17, 21]. Mouse eggs also have so-called “thapsigargin-insensitive” Ca²⁺ stores, capable of release by IP₃ or Ca²⁺ ionophore but not thapsigargin, whereas these stores are absent in porcine eggs [11, 21].

SOCE supports Ca²⁺ oscillations in somatic cells [46]; therefore, it seems somewhat unusual that eggs use alternate means to refill ER stores following fertilization. Meiotic cell cycle arrest is unique to oocytes, and Ca²⁺ is a critical signal that regulates meiosis, prompting the MII-arrested egg to reenter the cell cycle and switch to a mitotic program. Therefore, it is logical that mature eggs must control Ca²⁺ signaling very tightly and have multiple mechanisms to ensure that meiotic arrest is robust [47]. Perhaps down-regulating SOCE at this time serves as an additional means of insulating the egg from extraneous Ca²⁺ signals. SOCE is downregulated in Xenopus oocytes during meiosis [48, 49], and although we show it is not reliant on SOCE components, Ca²⁺ influx following store depletion decreases during oocyte maturation in mice [1, 15]. This finding is also consistent with the down-regulation of SOCE that occurs as somatic cells enter mitotic M phase, which may help ensure that cell stage-specific needs for ion fluxes are met [50, 51].

Given that SOCE is dispensable for appropriate Ca²⁺ signaling in mouse oocytes, the question of what pathways support Ca²⁺ influx remains. The data reported here and that of others suggest that TRPM7-like channels contribute, at least in part, to Ca²⁺ influx in oocytes. Functional expression of TRPM7-like channels was recently shown in mouse oocytes [37]. Similar to our observation that TRPM7-like channels support spontaneous Ca²⁺ influx in GV oocytes, NS8593 reduces oocyte Ca²⁺ oscillations that occur in response to raising extracellular Ca²⁺ from 2 mM to 5mM [37]. Spontaneous influx in GV oocytes was previously attributed to SOCE because this response could be inhibited by 2-APB [1], but 2-APB is a notoriously non-specific inhibitor [33]. In addition to inhibiting Ca²⁺ influx mediated by SOCE [52], 2-APB also acts on a number of other channels and pathways, including acting as a TRPM7 inhibitor at concentrations below 1 mM [39]. Therefore, our hypothesis that spontaneous Ca²⁺ influx is mediated by TRPM7 is further supported by past experiments showing that 2-APB blocks Ca²⁺ influx in GV oocytes [1]. Furthermore, we show that inhibition of TRPM7-like channels during IVF impairs Ca²⁺ oscillations, suggesting that these channels also contribute to Ca²⁺ influx following fertilization. This conclusion is further supported by recent data demonstrating that NS8593 effectively blocks TRPM7-like current in mouse eggs [37]. TRPM7 is a promising candidate for mediating SOCE-like responses in mouse oocytes and eggs; however, as we have seen with SOCE, knowledge that can be gained using solely pharmacological inhibition is inherently limited. Therefore, a genetic approach will be needed to definitively determine the contribution of TRPM7.

We postulate that TRPM7 contributes to the Ca²⁺ influx that had previously been attributed to SOCE, but TRPM7 is unlikely to act alone in supporting Ca²⁺ influx in oocytes and eggs. We previously demonstrated that Ca_V3.2 contributes to Ca²⁺ influx to support ER store accrual and robust Ca²⁺ oscillations, but females lacking Ca_V3.2 are merely subfertile, indicating that this channel does not function alone either [27]. TRPV3 channels are present in mouse eggs and contribute to Sr²⁺-mediated parthenogenesis; however, Trpv3^−/− females are fertile and their eggs have normal Ca²⁺ oscillations following fertilization, indicating that these channels do not have a required physiological function in eggs [34]. It will be interesting to begin investigating the effects of knocking out these channels in combination to overcome inherent limitations of studies using pharmacological inhibitors and ultimately determine the mechanisms of Ca²⁺ influx that support egg activation.

We show that genetic knockout of SOCE components causes no measurable disruption of Ca²⁺ signaling in mouse oocytes and eggs; however, these cells display several behaviors reminiscent of SOCE, and explanations for these SOCE-like effects are still undetermined. Introduction of extracellular Ca²⁺ causes measurable Ca²⁺ influx in eggs that have depleted ER stores, but not in those with intact stores [1, 11, 15]. Membrane permeability to Ca²⁺ appears to rise following fertilization, and influx spikes occur concomitant with reduction in ER Ca²⁺ during initial Ca²⁺ oscillations [11, 42]. Overexpression of SOCE components in oocytes and eggs enhances the magnitude of SOCE-like responses and impacts Ca²⁺ oscillations and influx, indicating that SOCE components can function in eggs when expressed at supra-physiological levels [1, 15]. Furthermore, while some effects of 2-APB on oocyte and egg responses can be explained by modulation of TRPM7 or TRPV3 channels, it is possible that this inhibitor also impacts SOCE-like responses in other ways in these cells [1, 15, 34, 37]. Thus, it is apparent that the status of ER Ca²⁺ stores influences Ca²⁺ influx regulation in mouse oocytes and eggs, but this effect is not mediated by STIM-ORAI interaction. Studies to define the pathways connecting ER Ca²⁺ stores to channels other than ORAI and TRPC in oocytes and eggs and may lead to discovery of a yet undescribed cellular mechanism linking Ca²⁺ store regulation and influx in these specialized cells.

Highlights.

• SOCE is not required for Ca²⁺ influx in mouse oocytes and eggs

• Eggs lacking STIM1, STIM2, or ORAI1 have normal Ca²⁺ oscillations after fertilization

• Oocyte spontaneous Ca²⁺ influx does not require STIM1, STIM2, or ORAI1

• Trpm7-like channels support Ca²⁺ influx in mouse oocytes and eggs