Insulin, apo-transferrin, penicillin, streptomycin, fetal bovine serum (FBS), Leibowitz medium (L-15), and α-minimal essential medium (α-MEM) were obtained from Gibco BRL (Gaithersburg, MD. USA). Ryanodine, xestospongin C, thapsigargin, and follicle-stimulant hormone (FSH) were from Calbiochem (La Jolla, CA. USA). Fluo-4 AM, BODIPY TR-X Ryanodine, BODIPY-FL thapsigargin, TO-PRO-1 Iodide, brefeldin A BODIPY 558/568 conjugate isomer 1 were obtained from Molecular Probes (Eugene, OR. USA). Sodium pyruvate, 1,4-diazabicyclo 222 octane (DABCO), paraformaldehyde (PFA), glutaraldehyde, ATP, brefeldin A, bovine serum bovine (BSA), dimethyl sulfoxide (DMSO), the proteinase inhibitors: phenylmethylsulfonyl fluoride (PMSF), aprotinin, leupeptin, pepstatin, and other salts were obtained from Sigma (St. Louis, MO, USA). The Complete mini, Protease inhibitor cocktail tablets was purchased from Roche, (Germany). The NE-PER Nuclear and Cytoplasmic Extraction Reagents Kit was from Pierce (Rockford, IL. USA). The Alkaline Phosphatase (AP) conjugate substrate kit was purchased from Bio-Rad (Hercules, CA. USA). Jung tissue-freezing medium was obtained from Leica (Germany).

Goat polyclonal IgG (immunoglobin G) for isoforms 1, 2, and 3 of the IP₃R, rabbit anti-goat IgG-Texas Red (TR), rabbit anti-goat IgG AP, and rabbit polyclonal IgG α-actin (H-196) were obtained from Santa Cruz (Santa Cruz, CA. USA). Rabbit anti-goat IgG (H-L) FITC-Conjugate was purchased from ZYMED (San Francisco, CA. USA).

Cells were obtained based on a published protocol 321. Briefly, C57BL/6 NHsd female mice (all animal work was conducted using procedures reviewed and approved by our Institutional animal care, Mexican University), 40–60 days old in different stages of the estrous cycle, were killed by cervical dislocation, and the ovaries were dissected and transferred to a Petri dish with L-15 medium (supplemented with 50 μl/ml FBS, 100 U/ml penicillin, and 100 μg/ml streptomycin). Ovaries were cleaned of neighboring tissue and opened to allow visualization of follicles. Antral follicles (usually 8–10 per ovary) were manually separated from the ovaries and transferred to α-MEM (supplemented with 100 ng/ml FSH, 1 mM sodium pyruvate, 10 μg/ml apo-transferrin, 10 μg/ml insulin, 100 U/ml penicillin, and 100 μg/ml streptomycin). Each follicle was opened using fine forceps, and the granulosa cells were carefully removed and transferred to freshly prepared α-MEM, disaggregated mechanically, and placed on glass coverslips coated with poly-D-lysine. The primary culture was maintained in an incubator at 37°C and 5% CO₂ for 2–3 days.

In independent experiments, female C57BL/6 NHsd mice were anesthetized and perfused by intracardiac puncture with phosphate buffer (PBS in mM: 2 KH₂PO₄, 3 KCl, 10 Na₂HPO₄, 140 NaCl, pH adjusted to 7.4 with NaOH, containing 40 μl/ml PFA) for 8–10 min. Subsequently the mice were decapitated; and the ovaries were removed and cleaned to eliminate the adjacent tissue, then incubated in PBS-40 mg/ml PFA for 4 h at room temperature. Isolated ovaries were transferred to PBS containing 300 mg/ml sucrose and incubated for 12 h at 4°C. Finally, the ovaries were put into Jung Tissue Freezing Medium and preserved at -20°C for 12 h. Cryostat sections of 8–10 μm were used. Samples were stored at -80°C until use.

For immunochemical examination, coverslips of granulosa cells were fixed in PBS containing 40 mg/ml glutaraldehyde for 15 min at 40°C and washed three times with PBS. For cryostat sections, the slices were placed for 1 h at room temperature, transferred to 40 μl/ml PFA for 10 min, and washed three times, each for 5 min, with PBS.

Immunodetection of IP₃R types 1, 2, and 3 was achieved using the method reported in reference 22, with the following modifications: After fixation, the cells were exposed to 50 mg/ml fat-free milk in PBS for 1 h at room temperature (to block protein-binding sites) and washed three times with PBS, each for 5 min. Anti-IP₃R types 1, 2, and 3 antibodies diluted 1:200 with 50 mg/ml fat-free milk in PBS including 1 μl/ml Triton X-100 (PBST) (to block residual protein-binding sites) were added. The cells were incubated overnight at 4°C and then washed three times with PBS. In order to detect the primary antibody, cells were incubated with rabbit anti-goat IgG-TR or anti-goat IgG (H-L) FITC-Conjugate, diluted 1:1000 in PBST for 1 h at room temperature, and washed six times in PBST. The coverslips were protected with DAPCO, an aqueous mounting medium. Visualization was performed in a confocal microscope with appropriate filters. Images were obtained using a LSM510 laser scanning microscope with a plan apochromatic 63 × oil-immersion objective (numerical aperture = 1.4), and captured at a resolution of 1024 × 1024 pixels.

Subcellular fractionation of mouse ovary cells was done using the protocol reported by 23. Briefly, mice were killed by cervical dislocation, and the ovaries were dissected and gently homogenized with a silicon-coated glass homogenizer in SET buffer (containing in mM): 300 sucrose, 1 EDTA, 1, 2-mercaptoethanol, 50 Tris-HCl pH 8.0. The buffer was supplemented with the following peptidase inhibitors: 0.2 mM PMSF and 10 μg/ml each of aprotinin, leupeptin, and pepstatin or Protease inhibitor cocktail tablets. The homogenized tissue was centrifuged at 2,000 g for 10 min to remove residual tissue and heavy particles. The supernatant was recovered and then centrifuged at 105,000 g for 45 min. Microsomal precipitates were diluted in SET buffer. Membranes were kept at -80°C until use. Protein was determined according to the method of Lowry 24.

Membrane fractions were boiled for 10 min and loaded onto 60 mg/ml SDS-polyacrylamide gels. Ovarian proteins were transferred to nitrocellulose membranes using a Mini Trans Blot Semi dry transfer cell (BioRad, Hercules CA). The nitrocellulose membranes were blocked 2 h in PBS-0.5 μl/ml Tween-2 supplemented with 50 mg/ml fat-free milk. After 3 washes with 150 mM NaCl-0.5 μl/ml Tween-20 (NaCl-T), each for 10 min, the membranes were incubated overnight at 4°C with the primary antibody (Goat polyclonal IgG anti-IP₃R types 1, 2, and 3, dilution 1:250) in PBST supplemented with 1 mg/ml BSA, washed again three times with NaCl-T for 10 min and incubated for 1 h with the secondary antibody (rabbit anti goat IgG-AP, dilution 1:500) in PBST supplemented with 1 mg/ml BSA. After 3 washes with 0.1 M Tris pH 9.5, color associated with the complexes of IP₃Rs-antibodies was developed using the AP conjugate substrate kit (Bio-Rad, Hercules CA). All fractions were incubated with rabbit polyclonal IgG α-actin (H-196) (1:1000), a rabbit polyclonal antibody raised against amino acids 180–375 of α-actin of human origin. From this point the protocol mentioned in the previous section was followed.

Nuclear extracts were obtained according to NE-PER Nuclear and Cytoplasmic Extraction Reagents Kit (complemented with Protease inhibitor cocktail tablets) 13. Briefly, the isolation of cytoplasmic and nuclear fractions using the NE-PER kit maintains the integrity of the two cellular compartments before extraction. This prevents cross-contamination of proteins between the two fractions. Additionally, we performed a centrifugation (16,000 × g) to obtain nuclear membranes and nucleoplasmic fraction. The protein was calculated according to the Bradford method 25, and analyzed by Western blotting (see previous section).

Following fixation, the mouse granulosa cells were incubated with the following fluorescent probes: BODIPY TR-X Ryanodine (1 μM) to detect the ryanodine receptor 2627, BODIPY-Red thapsigargin (1 μM) to localize the thapsigargin-sensitive Ca²⁺-ATPase (SERCA) 27, TO-PRO-1 iodide (50 nM) and DAPI (1 μg/ml) to stain nuclei 28, brefeldin A BODIPY 558/568 and BODIPY FL-conjugate isomer 1 (1 μM) to localize the ER and Golgi apparatus 29. Fluorescent probes were incubated for 60–90 min at room temperature and washed 3 times with PBS for 5 min to minimize non-specific binding. To estimate non-specific binding of the fluorescent probes, cells were pre-incubated with 100 μM ryanodine, 100 μM thapsigargin, 100 μM brefeldin A according to 21. Treated cells were visualized by confocal microscopy (see Immunochemistry Section).

Ca²⁺ mobilization was determined according to 21. Briefly, mouse granulosa cells were loaded for 20–30 min at room temperature with 5 μM Fluo-4 AM in Krebs solution (KS; containing in mM): 150 NaCl, 1 KCl, 1 MgCl₂, 1.8 CaCl₂, 4 Glucose, 10 HEPES, pH adjusted to 7.4, with the addition of 5 mg/ml of BSA and 0.1 mg/ml of pluronic acid. The solution was filtered to eliminate particles. Cells were washed 3 times with KS to remove extracellular Fluo-4 AM and incubated for 10 min to complete de-esterification of the dye. The coverslips were mounted in a recording chamber and placed in a Nikon Eclipse E600 microscope. To apply the drugs, a home-made multichannel perfusion system was used. Perfusion of the solutions was continuous at 1 ml/min. Confocal images were obtained using a Nikon Plan-Flour × 20 multi-immersion objective (numerical aperture = 0.75) and captured at a resolution of 640 × 640 pixels in the scan mode (1 image/min). Photo-bleaching and photo-damage were minimized by reducing the laser power (95% attenuation). Fluorescent measurements were performed with an Argon laser (Fluo-4 AM was excited at 506 nm, and emission was collected at 526 nm).

Multiphotonic and confocal images (for control and experimental conditions) were obtained and analyzed in identical situation by LSM image and SIMPLE PCI software respectively. Immunochemical analyzes involved ≈ 20 cells, in which each cell was divided in 4–5 quadrant of similar dimensions (10 μm²) in both cytoplasmic and nuclear areas (n = 5). For analyses of Ca²⁺ dynamics experiments, the complete nuclear and cytoplasmic areas of ≈ 200 cells were considered (n = 4).

Values are expressed as means ± S.E. Significance was tested by the Student's t-test. P < 0.05 was considered significant. Graphics were made with the Origin program (Version 5.0), and the images were processed with Photoshop program.

Commercial specific antibodies, tested in numerous previous reports 2330, coupled to fluorescent moieties were used to determine the presence and intracellular distribution of IP₃Rs in primary cultures of mouse granulosa cells by confocal microscopy (Figure 1) (analyzed cells ≈ 20; n = 5). All three types of IP₃Rs were expressed at discernible levels, but in different patterns. Apparently, IP₃R types 2 and 3 were the most abundant, whereas IP₃R type 1 was detected in lesser amount, especially in the cytoplasm (Figure 1, images in the first column). Remarkably, the nuclear signal of all three IP₃R isoforms was intense and well defined. Confirmation of the nuclear localization of the three IP₃Rs isoforms was obtained by co-localization with DAPI, which is a specific nuclear marker (Nuclei and Merge columns in Figure 1). Images seen at higher magnification show an intricate cytoplasmic pattern for the three types of IP₃Rs, with the signals for type 2 and 3 extended more towards the periphery of the cell (fourth column of Figure 1). In addition, magnified images clearly display nuclear signals for the three isoforms in both the nuclear envelope and the interior of the organelle. To further confirm the nuclear localization of the IP₃R isoforms, the fifth column depicts a z-stack reconstruction (≈13 optical slices corresponding to 4 μm section) of the granulosa cells nuclei. It can be seen that the 3 types of IP₃R show intranuclear localization, being more abundant the isoform 2 and 3. Therefore, primary cultures of mouse granulosa cells co-express all three IP₃R isoforms, and unexpectedly, all of them are present within the nuclei. Low signal, barely detectable, was observed when the primary antibodies were omitted (Figure 1, insets in first column). To estimate the difference between the cytoplasmic and the nuclear IP₃R distribution, a semi-quantitative analyzes was done (histograms at the bottom of Figure 1; cells analyzed ≈ 20; n = 5). IP₃R type 1 and 2 showed an increased nuclear presence, whereas IP₃R type 3 displayed similar signals in both compartments.

Figure 2 panel A shows biochemical detection of IP₃Rs isoforms by Western blot in reductive conditions, in granulosa cells (10⁸ cells; n = 3), ovary and cerebellum (8 mice; n = 3). IP₃R type 1 was detected in granulosa cells, ovary and cerebellum with a expected high molecular weight band (≈230–250 kDa 223132. In cerebellum, a second band (≈200 kDa) was also observed. IP₃R type 2 was also seen in granulosa cells in a similar high molecular weight band (≈230–250 kDa), but in addition, three smaller bands were also detected (≈120, 100 and 90 kDa). The high molecular weight and smaller bands were also observed in ovary and cerebellum. The low molecular weight variants of IP₃R type 2 might correspond to regulated proteolytic activity or to alternative mRNA processing 313233343536. IP₃R type 3 in granulosa cells was detected as a high molecular weight band (≈230–250 kDa), but also in a band with lower molecular weight (≈130 kDa). IP₃R type 3 was also present in ovary and cerebellum. At the right side of Figure 2 panel A, note that the signal corresponding to α-actin was detected in ovary tissue and not in granulosa cells treated with FSH in vitro, since the cytoskeleton of these cells does not contain this protein 37. Taken together these results, it is shown that granulosa cells express the 3 IP₃R isoforms, and in our experimental conditions granulosa cells are luteinized, and then they do not express α-actin.

Figure 2 panel B shows the presence of the three IP₃R isoforms in isolated nuclei of granulosa cells (10⁸ cells; n = 3), in two fractions: nuclear membranes (NM) and nucleoplasm (NP). For all three isoforms, a high molecular weight band (≈230–250 kDa) corresponding to the complete form of IP₃R was observed in the NM fraction. However the band of IP₃R type 1 was clearly fainter than the other two types. Two smaller variants of IP₃R types 2 and 3 were also detected in this fraction. The NP fraction showed no signal for the complete IP₃R type 1; in contrast a weak band corresponding to IP₃R type 2 whereas the band for type 3 was stronger. Low molecular weight variant was clearly detected for types 2 and 3.

To examine the functionality of the nuclear Ca²⁺-handling proteins detected in the mouse granulosa cells, ATP-induced Ca²⁺ mobilization was assessed by confocal microscopy (analyzed cells = 800; n = 4). Application of ATP (50 μM) to the assay promoted Ca²⁺ mobilization by activation of purinergic G-protein-coupled receptor P2Y2 in mouse granulosa cells 321. The Fluo-4 associated signal was higher in the nuclei than in the cytoplasm, even in basal, non-stimulated conditions (Figure 3, first block, and column α). However, this fact could be due to nuclear milieu influence over the fluorescent signal, and not necessarily represent an elevated Ca²⁺ level within the nucleus. Upon ATP stimulation, the Ca²⁺ transient elicited in the nucleus was higher than the one recorded in the cytoplasm (Figure 3, first block, and column β). After the transient, intracellular Ca²⁺ returned to basal levels, but the nuclear fluorescent signal remained higher than in the cytosol (Figure 3, first block, and column γ). Treatment with the IP₃R non-competitive inhibitor xestospongin C (5 μM for 15 min) prevented the ATP-induced Ca²⁺ mobilization in both compartments, without changes in the basal Ca²⁺ (Figure 3, second block, columns α, β, and γ). Interestingly, the ATP-induced Ca²⁺ transient in the presence of the antibiotic brefeldin A (5 μg/ml for 2 h) showed a differential response comparing the cytoplasmic and nuclear compartments. In this case, Ca²⁺ mobilization was completely suppressed in the cytoplasm, but in the nucleus the brefeldin A effect was only partial, and a net Ca²⁺ mobilization could be observed (Figure 3, third block, and column β). In addition, brefeldin A treatment promoted a decrease in the basal Fluo-4 associated-signal within the nucleus (Figure 3, third block, columns α, β and γ). These data support the notion that the nuclei in the granulosa cells, as the nuclei of others cellular systems, also have the capacity to mobilize Ca²⁺ 14.

By means of specific fluorescent probes, the subcellular localization of RyR, SERCA, and the network of endomembranes in granulosa cells were investigated (n = 5). The results are shown in Figure 4. RyR formed clusters throughout the cytosol, and its expression within the nucleus was demonstrated by the coincidence of the fluorescent ryanodine derivative with the nuclear marker DAPI. The magnified image depicts more clearly the nuclear and perinuclear localization of the RyR. The fluorescent signal associated with thapsigargin showed the presence of SERCA in the cytoplasm as thread-like structures, and as with the RyR, the nuclear localization of SERCA was also evident from the coincidence with the DAPI signal. With higher magnification, the widespread distribution of SERCA throughout the entire cytoplasm as well as its intranuclear localization can be appreciated. Additionally, mouse granulosa cells were incubated with a fluorescent derivative of brefeldin A, a fungal metabolite that binds selectively to endomembranes, namely endoplasmic reticulum and the Golgi complex 29. As with RyR and SERCA, the endomembrane marker was present in the cytoplasm, where it formed numerous clusters, and also within the nucleus. Co-localization with DAPI further supported the intranuclear localization of fluorescent brefeldin A, suggesting the presence of a membranous system within the nucleus (possibly the nucleoplasmic reticulum) in mouse granulosa cells. These characteristics are apparent in the magnified image. Taken together, these results indicate that the nuclei of granulosa cells in culture conditions express the principal Ca²⁺-handling proteins, as well as membranal structures in the nucleoplasmic compartment. These findings are similar to results reported in other cellular systems 1011121314.

To eliminate possibly that our results reflected an artifact related to the culture conditions, we tested if the three IP₃R isoforms were also observed within the granulosa cells nuclei in histological slices of mouse ovarian tissue (n = 3). Figure 5 shows the signals associated with the same isoform-specific antibodies against the IP₃R types 1, 2, and 3 that were used in the assays with granulosa cells in vitro. Consistent with the previous results, the three IP₃R isoforms were also detected in the granulosa cell nuclei in situ when these cells are still part of the ovary, which are located between the oocyte and the theca cells. The specificity of the experiments was confirmed by the very low unspecific signal seen when only the secondary antibodies were used (Figure 5, Control). These results confirm previous reports in which the three IP₃R isoforms were detected in ovary 1920.

The distribution of the three types of IP₃R was similar along the slices of ovarian tissue. The nuclear expression of the three IP₃R isoforms was evident from their co-localization with the nuclear marker TO-PRO 1 iodide. Analogous results were also obtained with the theca cells. These data indicate that mouse granulosa cells express, both in culture and in situ, all three types of IP₃Rs with a cytoplasmic and nuclear localization.

Granulosa cells, like many other cellular types, contain at least two forms of Ca²⁺ release channels: RyR and IP₃R 21. Both channels are localized in the endoplasmic reticulum membranes, but also within the nuclear structure (Figures 1 and 4). Our data show that granulosa cells express the three IP₃R isoforms (Figure 1). Detection of the IP₃R isoforms was similar in both experimental conditions tested: in primary cultures (Figures 1 and 2), and in ovary slices (Figure 5). Certainly, we do not know if the immuno-detected IP₃Rs are homo or heterotetramers. However, we have evidence that granulosa cells express low molecular weight variants of IP₃R types 2 and 3 which could result from regulated proteolysis or mRNA splicing (Figure 2) 313233343536. Hence, granulosa cells have the potential to display a great variety of Ca²⁺ signaling responses based in the molecular diversity of IP₃Rs forms.

The occurrence of at least two different isoforms of IP₃R has been reported in several cellular types, such as hepatocytes (types 1 and 2) 38, lung (types 2 and 3) 32, colonic epithelium (types 2 and 3) 39, and deep cerebellar nuclei (types 1 and 3) 40. The congregation of the three isoforms of IP₃R in a unique cellular population is less frequent. Besides the finding from this study in mouse granulosa cells, the three types of IP₃R have been reported to coexist in rat bile duct epithelial cells or cholangiocytes 41 and in bovine adrenal chromaffin cells 42.

When two or three IP₃R isoforms are expressed in the same cell, they are usually distributed in different subcellular regions: For example, IP₃R type 3 is localized in the apical section of cholangiocytes and non-pigmented epithelium cells, whereas the other isoforms are present in the rest of the cell structure, especially in the basolateral region 3843. In the case of the mouse granulosa cells, we did not observe any preferential subcellular location for the IP₃R isoforms: The signal from the three types of IP₃Rs occurs throughout most of cytoplasm, presumably along the granulosa cells endomembranes, with a less intense signal for the type 1 (Figure 1). The identity of the organelles in which the IP₃R isoforms were detected will be treated later.

In heterologous systems (Sf9 insect cells) the expression of each isoform of the IP₃Rs presents the same Ca²⁺ gating and similar ionic conductance 44. However, they differ in their sensitivity to IP₃, intracellular Ca²⁺, and ATP: type 1 shows medium IP₃-affinity, high ATP-affinity and low Ca²⁺ affinity. IP₃R type 2 has a high IP₃-affinity, a medium Ca²⁺ affinity, and is ATP independent. IP₃R type 3 shows a low IP₃-affinity, a low ATP-affinity, and a high Ca²⁺ affinity 44.

IP₃R type 3 is related to the triggering of Ca²⁺ waves in diverse tissues, such as polarized epithelia 38, as well as to act as an apoptotic mediator in different cellular systems 45. As to granulosa cells, it is not clear if they can be considered polarized, but some authors postulate a directionality in their function when, as a cellular population, the granulosa cells are surrounding the oocyte. For example, the handling of intracellular Ca²⁺ in the granulosa cells close to the theca is different from the granulosa cells close to the oocyte 46. However, at least in the ovarian slice, there was no indication of the existence of a cellular sub-population with regard to the three IP₃R isoforms. As to the role of Ca²⁺ release channels in promoting Ca²⁺ signaling, it was reported that the activity of RyR was necessary for ATP-induced Ca²⁺ mobilization in mouse granulosa cells 21. More systematic studies are needed to define the precise role of the three IP₃R isoforms during spontaneous and ligand-induced Ca²⁺ transients in granulosa cells. Although it is well documented that granulosa cells are prone to apoptosis according to hormonal and nutritional factors 47, the exact role of IP₃R type 3 or the other two isoforms has not be substantiated during this process when this cell population is luteinized.

IP₃Rs have long been known to be present within the sarco-endoplasmic reticulum membranes of many cellular types and tissues (for review see 48). However, growing evidence indicates the existence of the IP₃Rs in other intracellular organelles such as the Golgi apparatus 49, plasma membrane 5051, and nucleus 8910111213141516171852. Interestingly, in several cell types the IP₃R has been detected in the nuclei, specifically in the nuclear envelope membranes, but also within membranous reticular structures known as the nucleoplasmic reticulum 17. This intranuclear system is able to store and release Ca²⁺ in the same way as the endoplasmic reticulum does in the cytoplasm 1753. To our knowledge, the results reported here are the first to show that the nuclei of granulosa cells contain all three isoforms of the IP₃R. The intranuclear distribution of each isoform of the IP₃Rs is variable; for example, types 1 and 3 were preferentially localized in the inner nuclear membrane of skeletal muscle myocytes 54. In contrast, isoform 1 in the nuclei of bovine aortic endothelial cells, bovine adrenal glomerulosa cells, COS-7 cells 55, and ventricular myocytes 56 was not confined to the nuclear envelope, but distributed uniformly within the nucleus. Type 2 was detected forming part of the nucleoplasmic reticulum in rat hepatocytes 17. The exact role of each IP₃R isoform in the generation and kinetics of Ca²⁺ transients in the nucleus and cytoplasm of all these cells remains to be explored.

Nuclei in the granulosa cells contain the most important elements to accomplish Ca²⁺ release and uptake: besides IP₃Rs, we detected positive and specific signals for RyR, thapsigargin-sensitive SERCA, and endomembranes (Figures 1 and 4). It has been postulated that the cellular pattern of Ca²⁺ dynamics depends on the isoforms of the IP₃Rs, RyRs, and SERCAs present in the different organelles 11. Indeed, the results in Figure 3 indicate that cytoplasm and nuclei differ in their capacity for ATP-induced Ca²⁺ mobilization due to differences in the extent to which brefeldin A alters Ca²⁺ dynamics in the two compartments. Nuclear Ca²⁺ has been shown to be specifically and autonomously mobilized in a great variety of other cellular types 53. Questions arise regarding the potential functions of Ca²⁺ fluctuations within the nucleus. Several reports have shown that nuclear Ca²⁺ can control the transcriptional activity of certain genes 57, protein transport across the nuclear envelope 58, and translocation of protein kinases 1759. More experiments are needed to determine how these ion channels and metabolic pumps are coordinated to enable nuclear Ca²⁺ transients in synchronization with cytosolic Ca²⁺ dynamics in granulosa cells.

Granulosa cells are part of an ovarian complex where different follicular cell types are congregated to promote the development and maturation of the oocyte. Within this complex, granulosa cells are found in a precise location between the oocyte and the theca cells. It is a controversial issue if granulosa cells change their physiological characteristics when they are dissected and placed in culture conditions 60. However, there are numerous reports that consider granulosa cells in vitro to be a suitable experimental system. Hence, studies of granulosa cells properties in culture conditions regarding hormonal action, signal transduction, and cell differentiation are common 61.

The findings of this study indicate that the presence of the three IP₃R isoforms in the cytoplasmic and nuclear membranes is comparable in granulosa cells maintained in in vitro conditions with granulosa cells as part of the in vivo histological architecture; however, this conclusion must be confirmed. There are three potential interpretations of these observations: First, the expression of the three IP₃R isoforms is not an artifact associated with the manipulation of granulosa cells when they are put in culture. Second, it is highly probable that granulosa cells express the three IP₃R types in vivo, which would indicate a potential enriched repertoire in the intracellular Ca²⁺ dynamics of these ovarian endocrine cells. Third, granulosa cells are among the cellular types that express consistently the three isoforms of the IP₃R, and the only type known so far that expresses all of them within the nucleus.