Missense mutations in inositol 1,4,5-trisphosphate receptor type 3 result in leaky Ca²⁺ channels and activation of store-operated Ca²⁺ entry

Summary

Mutations in all subtypes of the inositol 1,4,5-trisphosphate receptor Ca²⁺ release channel are associated with human diseases. In this report, we investigated the functionality of three neuropathy-associated missense mutations in IP₃R3 (V615M, T1424M, and R2524C). The mutants only exhibited function when highly over-expressed compared to endogenous hIP₃R3. All variants resulted in elevated basal cytosolic Ca²⁺ levels, decreased endoplasmic reticulum Ca²⁺ store content, and constitutive store-operated Ca²⁺ entry in the absence of any stimuli, consistent with a leaky IP₃R channel pore. These variants differed in channel function; when stably over-expressed the R2524C mutant was essentially dead, V615M was poorly functional, and T1424M exhibited activity greater than that of the corresponding wild-type following threshold stimulation. These results demonstrate that a common feature of these mutations is decreased IP₃R3 function. In addition, these mutations exhibit a novel phenotype manifested as a constitutively open channel, which inappropriately gates SOCE in the absence of stimulation.

Graphical abstract

Highlights

• •IP₃R3 variants associated with human disease exhibited decreased channel activity
• •IP₃R3 variants resulted in constitutively active “leaky” channels
• •Leaky channels resulted in depleted ER Ca²⁺ stores and SOCE

Molecular biology; Molecular neuroscience; Cell biology

Introduction

Mutations in the inositol 1,4,5-trisphosphate (IP₃) receptors (IP₃Rs) are increasingly associated with human disease.^1,2 IP₃Rs are calcium (Ca²⁺) channels predominantly located on the endoplasmic reticulum (ER) membrane, that mediate Ca²⁺ release from the ER Ca²⁺ store upon binding of IP₃.^3,4 IP₃ is produced as the result of a signal transduction cascade initiated by agonist binding to G protein-coupled receptors or receptor tyrosine kinases at the plasma membrane (PM).⁵ Both receptor classes result in the activation of phospholipase C isozymes, which cleave phosphatidylinositol 4,5-bisphosphate into membrane-bound diacylglycerol and freely diffusible IP₃. The spatial and temporal aspects of the resulting Ca²⁺ release into the cytosol allow for the interaction of Ca²⁺ with various effectors leading to a diverse set of cell functions including apoptosis, gene transcription, fertilization, proliferation, growth and development, metabolism, fluid secretion, and exocytosis.^{5,6,7,8,9,10,11,12,13,14,15} A secondary effect of the Ca²⁺ release is the depletion of the ER Ca²⁺ store, which when sensed by the Stromal Interaction Molecule (STIM) activates Ca²⁺ entry into the cytoplasm through the Orai channels resident in the PM—a process known as store-operated calcium entry (SOCE). In order to maintain homeostasis, excess Ca²⁺ is removed from the cytosol through extrusion from the cell via the plasma membrane Ca²⁺ ATPase (PMCA), through reuptake into the ER via the sarcoplasmic ER Ca²⁺ ATPase (SERCA) or sequestration by mitochondria through the mitochondrial Ca²⁺ uniporter (MCU).⁹

Three IP₃R genes in the human genome (ITPR1, ITPR2, ITPR3) produce monomeric (∼300 kDa) isoforms of the receptor (IP₃R1, IP₃R2, and IP₃R3), which share ∼60-70% sequence homology.^{5,16,17,18,19,20} A single subunit consists of a highly conserved N-terminal suppressor domain (SD; 1-226 (Sequence used here is human IP₃R3 (NP_002215.2) and domains are based on previously identified domains and structures in rat IP₃R1 (NP_001007236))²¹ and ligand-binding domain (LBD; 227-578)²² and a highly conserved C-terminal channel pore and cytosolic tail (2201-2671).^23,24,25 These termini are separated by a large, non-conserved regulatory and coupling domain (579-2200)²⁶ where regulation by Ca²⁺, ATP, PKA phosphorylation, and other small molecules and binding partners occurs.^{27,28,29,30,31,32} As a result, IP₃ binding to the N-terminal LBD, which is about 70 Å from the channel pore induces long-range confirmational changes in the cytosolic domain enabling pore opening.^27,30

The three isoforms of the receptor are obligate tetramers and can assemble as both homo- or hetero-tetrameric complexes.^33,34 While IP₃Rs are ubiquitously expressed, one isoform or a combination of isoforms are preferentially expressed in various tissues and cell types.^{35,36,37,38,39} Thus, the stoichiometry of the IP₃R tetramers is likely dependent upon the distribution and endogenous expression level of each isoform.⁴⁰ Previous studies have established that IP₃R1 is highly expressed in the nervous system and is essential for the normal development of the nervous system. Mice engineered to lack native IP₃R1 often die prior to birth, or if born, suffer from severe ataxia and seizures prior to early postnatal death.^41,42,43 In contrast, IP₃R2 and IP₃R3 are commonly co-expressed in the periphery such as in secretory cells of the digestive system and exocrine system though, they can also be expressed in isolation, for example, IP₃R2 in the sweat glands. As a result, IP₃R2 or IP₃R3 single knock-out mice develop subtle phenotypes.^44,45 However, IP₃R2 and IP₃R3 compound knockout mice exhibit severe salivary and pancreatic insufficiencies resulting in death shortly after weaning due to the inability to assimilate macronutrients from dry food.⁴⁴

Mutations in all three IP₃R isoforms, located in all the major functional domains, are associated with several human diseases.^1,2 The majority of the mutations described thus far, are found in the IP₃R1 isoform and mainly impact nervous system function consistent with the IP₃R1 null mouse models.^{46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74} In contrast, fewer mutations have been identified in the IP₃R2 and IP₃R3 isoforms and are associated with diseases including Sézary Syndrome, familial isolated primary hyperparathyroidism (FIHP),⁷⁵ anhidrosis,⁴⁵ head and neck squamous cell carcinoma,⁷⁶ neuropathy,^77,78,79,80 and immunopathy.⁸⁰ Since most reports present case studies, the functional consequences of particular mutations on the overall structure and function of the receptors remain poorly understood. Of those published, most previous studies indicate mutations in IP₃Rs are associated with either diminished or complete loss of channel activity,^2,80,81 while a few mutations are associated with increased channel activity.^82,83

Recently, there has been an increase in the number of IP₃R3 missense mutations reported (Figure 1). Most of these mutations are in the regulatory and coupling domain and associated with peripheral neuropathies.^77,78,79,80 Two studies investigated the functional consequences of these mutations on Ca²⁺ signaling in patient-derived cells; however, the consequences of these mutations on IP₃R3 function were difficult to discern likely due to a contribution and compensation from endogenous wild-type IP₃R1 and IP₃R2 isoforms present in these cells.^77,80 To unambiguously probe any alteration in human IP₃R3 function, in this study we generated individual stable cell lines expressing human IP₃R3 (hIP₃R3) V615M, T1424M, R2524C mutations associated with neuropathy in a HEK-3KO background. This cell line was previously generated in our laboratory by CRISPR-Cas technology to lack all the three native IP₃Rs.⁸⁴ Interestingly, the hR V615M, T1424M, and R2524C stable cells lines all exhibited elevated basal intracellular Ca²⁺ concentration ([Ca²⁺]_i), reduced ER Ca²⁺ content and constitutive SOCE activation under basal conditions. These data are consistent with a “leaky” channel phenotype. However, all three mutations exhibited drastically different Ca²⁺ release profiles in response to agonist stimulation. The V615M was poorly functional, R2524C was non-functional, and T1424M exhibited increased function at low agonist concentrations. Overall, we report the functional consequences of three neuropathy-associated IP₃R3 mutations including the first reports of IP₃R disease-associated mutations that result in leaky channels.

Figure 1.

Linear-schematic of disease-associated mutations in the four main functional domains of the IP₃R3

Disease-associated mutations in IP₃R3 have been identified in the suppressor domain, regulatory and coupling domain, and C-terminal channel domain. These diseases include head and neck squamous cell carcinoma (red), neuropathy (green), immunopathy (blue), and immunopathy and neuropathy (purple). One family with mutations in IP₃R3 exhibited multiple diseases including chronic fatigue syndrome, small fiber neuropathy, and IgG3 deficiency (pink).

Results

While the functional effects of disease-associated mutations in IP₃R1 and IP₃R2 have been previously investigated,^45,61,80,81 little work has been done to characterize mutations in the IP₃R3 isoform. Several novel mutations in IP₃R3 (Figure 1) have been identified in the past decade with most associated with hereditary motor and sensory neuropathies (HMSN) such as Charcot-Marie-Tooth (CMT), which is characterized by progressive distal motor and sensory impairments (Table S1).^{77,78,79,80,85} In this study, we characterized the functional consequences of three mutations in hIP₃R3 located at different locations within the regulatory and coupling domain or channel domain, in order to gain mechanistic insight into how these mutations alter channel function and ultimately contribute to the disease phenotype.

The V615M mutation results in a poorly functional and “leaky” channel

Valine 615 is located in the regulatory and coupling domain of the IP₃R3 armadillo repeat domain (ARM) 1 (Figure 2A), distal to the LBD.^22,86 The hydrophobic/non-polar Val615 residue is replaced by a larger, hydrophobic/non-polar, sulfur-containing Met (V615M). Individuals from three generations of a Finnish family express this variant on a single IP₃R3 allele (Table S1). The patients were diagnosed with demyelinating neuropathy (CMT) inherited in an autosomal dominant manner, and presented with progressive muscle weakness, wasting in the legs and sometimes hands, as well as intermediately reduced nerve conduction values (NCV).⁷⁷ The Val615 residue, is not conserved among hIP₃R isoforms, but is evolutionarily well conserved between IP₃R3 in different species (Figure S1A). We mutated the codon encoding the Val615 residue in hIP₃R3 cDNA to Met and expressed the variant in isolation in HEK-3KO cells (hR3 V615M) (Figure 2B).⁸⁴ The hIP₃R3 V615M mutation localized properly to the ER membrane (Figure S1B). Our approach was to generate several stable cell lines expressing hR3 V615M homo-tetramers at varying levels (Figure 2C) allowing for comparison of channel function with wild-type (WT) cell lines of a similar expression level. For example, HEK293 IP₃R1/2 double knockout cells expressing endogeous levels of IP₃R3 were previously generated through CRISPR/Cas9 gene editing to only express endogenous hIP₃R3 (Endo. hR3),⁸⁷ while high IP₃R3 expressing HEK-3KO cells were generated by stably over-expressing hIP₃R3 cDNA (Exo. hR3). In Fura2/AM loaded hR3 V615M cells, agonist-induced Ca²⁺ release was utilized as a measurement of channel function. Prior to any agonist addition, all three V615M mutant cell lines exhibited significantly increased basal 340/380 ratios when compared to WT hR3 expressing cells and HEK-3KOs, indicative of elevated basal [Ca²⁺]_i (Figure 2D and 2E). The mean basal [Ca²⁺]_i corelated with V615M hIP₃R3 protein level, such that the lowest expressing hR3 V615M cell line (hR3 V615M #503), with the expression between that of the Endo. hR3 and Exo. hR3, exhibited only a small increase compared to the null and WT cell lines. In contrast, the hR3 V615M #524 cell line, with expression slightly greater than that of the Exo. hR3, and the V615M #506 cell line, with expression several-fold greater than the Exo. hR3 cell line exhibited a much larger increase in basal [Ca²⁺]_i. In response to stimulation with increasing concentrations of carbachol (CCh), a muscarinic receptor agonist, which results in an increase in intracellular IP₃, all three hR3 V615M cell lines exhibited a significant attenuation of the initial Ca²⁺ release (Figure 2F) and percentage of responding cells (Figure S1C) compared to the Endo. hR3 and Exo. hR3 cell lines. Of note, those hR3 V615M cells with larger basal [Ca²⁺]_i were those that exhibited decreased function, while the small percentage of individual hR3 V615M cells with larger changes in Ca²⁺ release were those with lower basal [Ca²⁺]_i (Figure S1D).

Figure 2.

hR3 V615M is a poorly functional Ca²⁺ channel with elevated basal cytosolic [Ca²⁺]

(A) Chimera (PDB: 6DR0) was used to visualize Val615 (yellow) in ARM1 of the regulatory and coupling domain (red) of IP₃R3.

(B) Cell lines with varying expression of human IP₃R3 harboring the V615M mutation (hR3 V615M) were generated in IP₃R-null HEK-3KO cells and western blotted alongside WT cell lines, either endogenously expressed hIP₃R3 (Endo. hR3) and exogenously expressed hIP₃R3 (Exo. hR3).

(C) Quantification of expression of hIP₃R3 expression with respect to GAPDH in HEK-3KO (blue), Endo. hR3 (purple), Exo. hR3 (green), and hR3 V615M (pink, orange, red) cell lines. Colored lines represent the mean of at least n = 3 experiments, and error bars represent SEM. Averages were normalized to that of the Endo. hR3 cell line.

(D) Representative traces of Ca²⁺ signals in fura-2 loaded cells in response to the addition of indicated concentrations of CCh.

(E) Scatterplot summarizing the basal Ca²⁺ (average of the initial 20,340/380 ratio points in Ca²⁺-containing media) from experiments similar to those in (D).

(F) Scatterplot summarizing change in amplitude (Peak 340/380 ratio – Basal 340/380 ratio (E)) of cell lines in response to CCh in single-cell imaging experiments similar to those in (D).

(G) Dose-response curve showing the change in amplitude of Ca²⁺ signals of Fura-2/AM loaded cell lines when treated with increasing [CCh] using a FlexStation3 96-well plate reader.

(H) Representative traces of Ca²⁺ puffs recorded by TIRFM in Exo. hR3 or hR3 V615M. Uncaging of ci-IP₃ at the arrowhead.

(I) Number of Ca²⁺ puff sites per cell in Exo. hR3 (green) and hR3 V615M (red) cell lines loaded with Cal-520 in TIRFM experiments and treated with1 μM ci-IP₃.

(J) Number of Ca²⁺ puff per cell in Exo. hR3 (green) and hR3 V615M (red) cell lines loaded with Cal-520 in TIRFM microscopy experiments and treated with 1 μM ci-IP₃. All data are mean ± SEM of at least three (N = 3) independent experiments. ^###p < 0.001 when compared to HEK-3KO; ^tttP < 0.001, ^ttP < 0.01 when compared to Endo. hR3; ∗∗∗p < 0.001 when compared to Exo. hR3; and ∗∗∗(red)p < 0.001 when compared to other stably expressed hR3 V615M cell lines; one-way ANOVA with Tukey’s test performed in E, F and two-tailed t-test performed in (I andJ).

In population-based Ca²⁺ imaging assays performed on a FlexStation3 with micro-fluidics, all three hR3 V615M cells lines similarly showed significantly diminished Ca²⁺ release (Figure 2G) together with the area under the curve (AUC) (Figure S1E) in response to increasing [CCh]. Likewise, in Total Internal Reflection Microscopy (TIRFM) experiments, measuring the Ca²⁺ release from small clusters of IP₃R3, the highest expressing hR3 V615M cell line exhibited significantly decreased numbers of Ca²⁺ puff sites (Figure 2H and 2I) and Ca²⁺ puff per cell (Figure 2H and 2J) as compared to the Exo. hR3 cell line (Video S1; Video S2). These results further support the single-cell Ca²⁺ imaging data showing that the hR3 V615M Ca²⁺ channel is poorly functional.

Next, we performed experiments to investigate the mechanism underlying the raised basal [Ca²⁺]_i in the V615M hIP₃R3 cells. In contrast to WT hR3 expressing cells or HEK-3KOs, removal of extracellular Ca²⁺ resulted in a significant decrease in basal cytosolic [Ca²⁺]_i in the hR3 V615M cell lines (Figures 3A and 3B) to levels similar to the IP₃R null and WT cell lines. A direct correlation was observed between the basal [Ca²⁺]_i and the magnitude of change in amplitude seen following the removal of extracellular Ca²⁺ (Figure S2A). These data indicate that Ca²⁺ influx across the PM contributed to the elevated basal [Ca²⁺]_i in hR3 V615M expressing cells. Thus, we hypothesized that the elevated cytosolic Ca²⁺ was derived, at least in part, from Ca²⁺ leaking from the ER store through the mutant channel and subsequently gating SOCE. First, to evaluate the ER Ca²⁺ store content, the SERCA pump was inhibited by exposure to 30 μM cyclopiazonic acid (CPA), in the absence of extracellular Ca² (Figure 3A). Consistent with decreased ER store content, the CPA-induced Ca²⁺ leak in the hR3 V615M cell lines was less (amplitude in Figures 3A and S2B and AUC in Figure S2C) when compared to the WT hR3 cell lines with similar IP₃R3 expression. Additionally, these changes in amplitude (Figure 3C) and AUC (Figure S2D) were clearly correlated with the basal [Ca²⁺]_i, as individual cells with elevated basal 340/380 ratios exhibited smaller changes in amplitude and AUC. Overall, these data suggest that the elevated basal [Ca²⁺ ]_i in hR3V615M expressing cells results from ER Ca²⁺ leak and subsequent Ca²⁺ influx across the PM.

Figure 3.

hR3 V615M cell lines exhibited depleted ER [Ca²⁺] and SOCE in the absence of agonist stimulation

(A) Representative traces of changes in cytosolic [Ca²⁺] following removal of extracellular Ca²⁺ in HEK-3KO (blue), Endo. hR3 (purple), Exo. hR3 (green), and hR3 V615M (pink, orange, red) cell lines loaded with Fura-2/AM. Cells were subsequently treated with 30 μM CPA allowing the measurement of the ER Ca²⁺ store content.

(B) Scatterplot summarizing the change in the basal 340/380 ratio following removal of extracellular Ca²⁺ in experiments similar to those in A. Colored lines represent the mean of at least n = 3 experiments, and error bars represent SEM.

(C) Scatterplot summarizing the correlation between an elevated basal 340/380 Ca²⁺ ratio (Figure 2E) and the change in 340/380 ratio following treatment with 30 μM CPA (maximum CPA-induced amplitude – basal 340/380 ratio following removal of extracellular Ca²⁺) in experiments similar to those in (A).

(D) Representative traces of Ca²⁺ signals in fura-2 loaded cells in response to the addition of 10 μM GSK-7975a.

(E) Scatterplot summarizing the change in 340/380 ratio following the addition of GSK-7975a (average of 20,340/380 ratio points prior to GSK-7975a addition – average of 20,340/380 ratio points following 200 s of GSK-7975a addition) from experiments similar to those in (D).

(F) Scatterplot summarizing the correlation between an elevated basal 340/380 Ca²⁺ ratio (Figure 2E) and the change in 340/380 ratio following treatment with 10 μM GSK-7975a (E) in experiments similar to those in (D). All data are mean ± SEM of at least three (N = 3) independent experiments. ^###p < 0.001 when compared to HEK-3KO; ^tttP < 0.001, ^ttP < 0.01 when compared to Endo. hR3; ∗∗∗p < 0.001, ∗∗p < 0.01 when compared to Exo. hR3; and ∗∗∗(red)p < 0.001 when compared to other stably expressed hR3 V615M cell lines; one-way ANOVA with Tukey’s test performed in (B and E).

Depletion of the ER Ca²⁺ store triggers SOCE through the interaction of STIM and Orai proteins.⁸⁸ To investigate the contribution of Ca²⁺ influx via Orai-based channels to the elevated [Ca²⁺]_i in the hR3 V615M mutants, we utilized the Orai inhibitor GSK-7975a.⁸⁹ In HEK293 cells, GSK-7975a inhibited 67.4% of CCh-induced Ca²⁺ influx (Figure S3A and S3B) and 50.8% of CPA-induced Ca²⁺ influx (Figure S3C and S3D). The addition of GSK-7975a resulted in a significant decrease in the basal [Ca²⁺]_i in the hR3 V615M cell lines, which was not observed in the WT hR3 and HEK-3KO cell lines (Figures 3D, 3E, and S3E). The elevated basal [Ca²⁺]_i levels correlated with a greater effect of GSK-7975a to reduce the elevated basal [Ca²⁺]_i levels (Figure 3F). In total, these results suggest that the mutation of V615M in hIP₃R3 results in a poorly functional, “leaky” channel. Ca²⁺ leak through the mutant channel leads to the partial depletion of ER Ca²⁺ stores thereby activating SOCE via Orai-based channels.

The T1424M mutation results in a “leaky” channel that is functional

The T1424M variant was identified as one of the two potential candidate gene mutations in a mother and her two daughters diagnosed with late-onset lower limb muscle weakness and sensory loss (Table S1, Phenotypes of relevant IP₃R3 mutations associated with neuropathy, related to Figure 1).⁷⁹ In these subjects, the hydrophilic/polar Thr1424 harboring a hydroxyl group is exchanged for a larger, hydrophobic/non-polar Met (T1424M) containing a sulfur group. Again, this mutation is present only on one ITPR3 allele and thus the effect of the mutation appears to be dominantly inherited. The T1424M mutant is located within the regulatory and coupling domain of IP₃R3 in the armadillo repeat 2 domain (ARM2) (Figure 4A). The T1424 residue is conserved both among all IP₃R isoforms, as well as across multiple species in hIP₃R3 (Figure S4A). Multiple cell lines stably expressing hR3 T1424M in HEK-3KO were generated with the hR3 T1424M #203 cell line expressing a low level of the IP₃R3 similar to that of the Endo. hR3, while the hR3 T1424M #209 and #204 cell lines expressed intermediate amounts of IP₃R3 between that of the Endo. hR3 and Exo. hR3 (Figures 4B and 4C). Homo-tetramers of hR3 T1424M were properly localized to the ER membrane (Figure S4B).

Figure 4.

hR3 T1424M increased Ca²⁺ channel function with elevated basal cytosolic [Ca²⁺]

(A) Chimera (PDB: 6DR0) was used to visualize Thr1424 (yellow) in the regulatory and coupling domain (red) of IP₃R3.

(B) Cell lines containing varying expression of human IP₃R3 harboring the T1424M mutation (hR3 T1424M) were generated in IP₃R-null HEK-3KO cells and western blotted alongside WT cell lines; Endo. hR3 and Exo. hR3.

(C) Quantification of hIP₃R expression with respect to GAPDH of HEK-3KO (blue), Endo. hR3 (purple), Exo. hR3 (green), and hR3 T1424M (pink, orange, red) cell lines. Colored lines represent the mean of at least n = 3 experiments, and error bars represent SEM. Averages were normalized to that of the Endo. hR3 cell line.

(D) Representative traces of Ca²⁺ signals in cells expressing the indicated variant in response to the addition of increasing [CCh].

(E) Scatterplot summarizing the basal Ca²⁺ (average of the initial 20,340/380 ratio points in Ca²⁺-containing media) from experiments similar to those in (D).

(F) Scatterplot summarizing change in amplitude (Peak 340/380 ratio – Basal 340/380 ratio (E)) of cell lines in response to increasing [CCh] in single-cell imaging experiments similar to those in (D).

(G) Dose-response curve showing the change in amplitude of Ca²⁺ signals when treated with increasing [CCh] using a FlexStation3 96-well plate reader.

(H) Representative traces of Ca²⁺ puffs in the absence of stimulation in Exo. hR3 and hR3 T1424M expressing cells.

(I) Number of Ca²⁺ puff sites and Ca²⁺ puffs per cell in Exo. hR3 (green) and hR3 T1424M #204 (red) cell lines loaded with Cal-520 in TIRFM microscopy experiments without photo release of ci-IP₃.

(J) Number of Ca²⁺ puff sites and Ca²⁺ puffs per cell in Exo. hR3 (green) and hR3 T1424M (red) cell lines loaded with Cal-520 in TIRFM microscopy experiments following photo-release of 0.1 μM ci-IP₃. All data are mean ± SEM of at least three (N = 3) independent experiments. Control HEK-3KO (blue), Endo. hR3 (purple), and Exo. hR3 (green) data in E, F repeated from Figure 2 above. ^###p < 0.001, ^##p < 0.01 when compared to HEK-3KO; ^tttP < 0.001, ^ttP < 0.01, ^tP < 0.05 when compared to Endo. hR3; ∗∗∗p < 0.001 when compared to Exo. hR3; and ∗∗∗(red)p < 0.001, ∗∗(red)p < 0.01 when compared to other stably expressed hR3 T1424M cell lines; one-way ANOVA with Tukey’s test performed in (E and F) and two-tailed t-test performed in (I and J).

While the low IP₃R3 expressing hR3 T1424M line (#203) did not exhibit an increased basal [Ca²⁺]_i (Figures 4D and 4E), the intermediate level expressing hR3 T1424M (#209) and hR3 T1424M (#204) cell lines both exhibited significantly increased basal [Ca²⁺]_i (Figures 4D and 4E). Ca²⁺ release in the low expressing hR3 T1424M cell line was refractory in response to any tested [CCh]. In contrast, the Endo. hR3 cell line, with the most similar expressions exhibited robust responses to [CCh] greater than 3 μM (Figures 4F and S4C). Notably, the higher expressing hR3 T1424M cell lines, both exhibited significant Ca²⁺ release and number of responding cells following stimulation with lower [CCh] (Figures 4F and S4C). Once again, individual hR3 T1424M cells with elevated basal [Ca²⁺]_i exhibited smaller changes in CCh-induced Ca²⁺ release than cells with more normal basal [Ca²⁺]_i (Figure S4D). Remarkably, despite only intermediate levels of IP₃R3 expression in the hR3 T1424M #209 and #204 cell lines the amplitude of Ca²⁺ release was greater in these lines than the WT hR3 cell lines at lower [CCh] indicating that the mutant channel is perhaps in a pre-activated confirmation (Figure 4F). Nevertheless, at higher [CCh] greater Ca²⁺ release was observed in the WT hR3 lines reinforcing the importance of this conserved residue for proper gating of the channel. In population-based FlexStation3 assays, the functionality of the higher expressing hR3 T1424M cell lines was confirmed and exhibited a shift in sensitivity at lower [CCh] (Figures 4G and S4E).

Notably, in TIRFM imaging experiments, spontaneous Ca²⁺ puffs were readily evident in the hR3 T1424M #204 cell line but were uncommon in Exo. hR3 expressing cells (Figure 4H) indicating that the mutant channel is active, even under basal conditions in absence of agonist stimulation (Figure 4, Video S3, Video S4). Additionally, upon stimulation through the uncaging of ci-IP₃, the number of Ca²⁺ puffs and Ca²⁺ puff sites per cell (Figure 4J) were significantly higher in the hR3 T1424M #204 cell line as compared to the Exo. hR3 cell line and signal tended to globalize in the hR3 T1424M expressing cells (Video S5 and Video S6).

Despite differences in functionality, the elevated basal [Ca²⁺]_i of the higher expressing hR3 T1424M cell lines again suggests that the T1424M mutation results in a leaky IP₃R3 similar to the V615M mutation. In the higher expressing hR3 T1424M cell removal of extracellular Ca²⁺ resulted in a significantly larger reduction in [Ca²⁺]_i than in the HEK-3KO and WT hR3 cell lines (Figures 5A and 5B). Similarly, the magnitude of the elevated basal [Ca²⁺]_i of individual hR3 T1424M mutant cells correlated with larger changes in the [Ca²⁺]_i following the removal of extracellular Ca²⁺(Figure S5A). SERCA pump inhibition resulted in a significantly smaller Ca²⁺ leak from the ER Ca²⁺ store in the hR3 T1424M mutant cell lines with elevated basal [Ca²⁺]_i levels (Figures 5A, S5B, and S5C) and the degree of inhibition was again correlated with the magnitude of the elevated basal [Ca²⁺ ] in individual cells (Figures 5C and S5D). Similar to V615M expressing cell lines, the Orai inhibitor GSK-7975a effectively decreased the elevated basal [Ca²⁺]_i (Figures 5D-5F and S5E) of the higher expressing hR3 T1424M mutant cell lines. Overall, these data demonstrate that expression of hR3 T1424M at levels greater than that of Endo. hR3 led to a leaky channel that results in the inappropriate activation of SOCE in the absence of stimulation similar to the hR3 V615M mutation. However, unlike the V615M mutation, which resulted in poorly functional IP₃R3 channels, the T1424M mutation, when over-expressed, leads to a variant with increased channel function at threshold levels of stimulation when compared to WT hIP₃R3 lines with comparable expression.

Figure 5.

hR3 T1424M cell lines exhibited depleted ER [Ca²⁺] and SOCE in the absence of agonist stimulation

(A) Representative traces of changes in cytosolic [Ca²⁺] following removal of extracellular Ca²⁺ in HEK-3KO (blue), Endo. hR3 (purple), Exo. hR3 (green), and hR3 T1424M (pink, orange, red) cell line. Cells were subsequently treated with 30 μM CPA allowing the measurement of the ER Ca²⁺ store content.

(B) Scatterplot summarizing the change in the basal 340/380 ratio following removal of extracellular Ca²⁺ in experiments similar to those in (A). Colored lines represent the mean of at least n = 3 experiments, and error bars represent SEM.

(C) Scatterplot summarizing the correlation between an elevated basal 340/380 Ca²⁺ ratio (Figure 2E) and the change in 340/380 ratio following treatment with 30 μM CPA (maximum CPA-induced amplitude – basal 340/380 ratio following removal of extracellular Ca²⁺) in experiments similar to those in (A).

(D) Representative traces of Ca²⁺ signals in response to the addition of 10 μM GSK-7975a when loaded with Fura-2/AM.

(E) Scatterplot summarizing the change in 340/380 ratio following the addition of GSK-7975a (average of 20,340/380 ratio points prior to GSK-7975a addition – average of 20,340/380 ratio points following 200 s of GSK-7975a addition) from experiments similar to those in (D).

(F) Scatterplot summarizing the correlation between an elevated basal 340/380 Ca²⁺ ratio (Figure 2E) and the change in 340/380 ratio following treatment with 10 μM GSK-7975a (E) in experiments similar to those in (D). All data are mean ± SEM of at least three (N = 3) independent experiments. Control HEK-3KO (blue), Endo. hR3 (purple), and Exo. hR3 (green) data in (B, C, and E), and F repeated from Figure 3 above. ^###p < 0.001 when compared to HEK-3KO; ^tttP < 0.001 when compared to Endo. hR3; ∗∗∗p < 0.001 when compared to Exo. hR3; and ∗∗∗(red)p < 0.001 when compared to other stably expressed hR3 T1424M cell lines; one-way ANOVA with Tukey’s test performed in (B and E).

The R2524C mutation forms a functionally dead but leaky channel

The R2524 residue is located in the channel domain at the junction of the 6^th transmembrane (TM) and the linker (LNK) domain (Figure 6A) and is highly conserved in hIP₃R subtypes and across different species (Figure S6A). The R2524C mutation was reported as a de novo mutation in two unrelated individuals (Table S1, Phenotypes of relevant IP₃R3 mutations associated with neuropathy, related to Figure 1). One individual, harboring solely the R2524C mutant on one ITPR3 allele, exhibited a phenotype associated with neuropathy.⁷⁷ The second individual, who also expressed an additional mutation (R1850Q) on their second ITPR3 allele, also presented with immunopathy symptoms resulting in a diagnosis of severe combined immunodeficiency disease (Table S1).⁸⁰ In these individuals, the positively charged Arg2524 residue is mutated to an uncharged, smaller Cys with a reactive thiol (-SH) group. The R1850Q mutation is present in 5% of the population as a polymorphism and when present alone in a parent of this individual did not result in an obvious phenotype and was previously characterized as poorly functional only when expressed at endogenous levels.⁸⁰ We therefore mutated the Arg2524 residue to Cys in WT hIP₃R3 (hR3 R2524C) and stably expressed the mutant in IP₃R-null HEK-3KO cells. As before, we generated several stable cell lines of hR3 R2524C homo-tetramers with varying expression levels (Figure 6B). The hR3 R2524C mutant was expressed at similar levels to Endo. hR3 in the hR3 R2524C #607 line (Figures 6B and C). hR3 R2524C #612 and #616 exhibited expression levels similar to and greater than Exo. hR3, respectively. The hR3 R2524C mutant appeared properly localized to ER membranes (Figure S6B). The basal [Ca²⁺ ]_i levels of the hR3 R2524C mutant cell lines were significantly increased in the absence of stimulation and the extent of the increase correlated with the relative expression level of the mutant (Figures 6D and 6E). However, in contrast to the V615M and T1424M mutants, channel function in the hR3 R2524C mutant in response to stimulation was completely absent (Figures 6F and S6C), irrespective of the basal [Ca²⁺]_i (Figure S6D) when measured using single cell or population-based⁸⁰ paradigms.

Figure 6.

hR3 R2524C exhibited absent Ca²⁺ channel function and an elevated basal cytosolic [Ca²⁺]

(A) Chimera (PDB: 6DR0) was used to visualize Arg2524 (yellow) at the junction of the 6^th TM (purple) and LNK domain (green) in the channel pore near the negatively charged Asp2518 of neighboring IP₃R3 monomers (blue).

(B) Cell lines with varying expression of human IP₃R3 harboring the R2524C mutation (hR3 R2524C) were generated in IP₃R-null HEK-3KO cells and western blotted alongside WT cell lines – Endo. hR3 and Exo. hR3.

(C) Quantification of expression of hIP₃R3 with respect to GAPDH, in HEK-3KO (blue), Endo. hR3 (purple), Exo. hR3 (green), and hR3 R2524C (pink, orange, red) cell lines. Colored lines represent the mean of at least n = 3 experiments, and error bars represent SEM. Averages were normalized to that of the Endo. hR3 cell line.

(D) Representative traces of Ca²⁺ signals from the indicated cell lines in response to the addition of increasing [CCh].

(E) Scatterplot summarizing the basal Ca²⁺ (average of the initial 20,340/380 ratio points in Ca²⁺-containing media) from experiments similar to those in (D).

(F) Scatterplot summarizing change in amplitude (Peak 340/380 ratio – Basal 340/380 ratio (E)) of cell lines in response to increasing [CCh] in single-cell imaging experiments similar to those in (D). All data are mean ± SEM of at least three (N = 3) independent experiments. Control HEK-3KO (blue), Endo. hR3 (purple), and Exo. hR3 (green) data in (E and F) repeated from Figures 2 and 4 above. ^###p < 0.001 when compared to HEK-3KO; ^tttP < 0.001, ^ttP < 0.01 when compared to Endo. hR3; ∗∗∗p < 0.001 when compared to Exo. hR3; and ∗∗∗(red)p < 0.001 when compared to other stably expressed hR3 R2524C cell lines; one-way ANOVA with Tukey’s test performed in (E and F).

The removal of extracellular Ca²⁺ again resulted in a decrease in basal [Ca²⁺]_i in the high R2524C expressing cells (Figures 7A, 7B, and S7A). Similarly, the CPA-induced Ca²⁺ leak from the ER was also decreased in the higher expressing hR3 R2524C cell lines indicating reduced ER Ca²⁺ store content in these cell lines (Figures 7A, 7C, and S7B-S7D). Addition of GSK-7975a resulted in a significant decrease in the elevated basal [Ca²⁺]_i of the higher expressing hR3 R2524C cell lines in the absence of stimulation (Figures 7D-7F and S7E), again suggesting the inappropriate activation of SOCE. Overall, these results indicate that the R2524C mutant is a completely non- functional Ca²⁺ channel, which again exhibits a leaky phenotype similar to the V615M and T1424M mutants.

Figure 7.

hR3 R2524C cell lines exhibited depleted ER [Ca²⁺] and SOCE in the absence of agonist stimulation

(A) Representative traces of changes in cytosolic [Ca²⁺] following removal of extracellular Ca²⁺ in HEK-3KO (blue), Endo. hR3 (purple), Exo. hR3 (green), and hR3 R2524C (pink, orange, red) cell lines. Cells were subsequently treated with 30 μM CPA allowing the measurement of the ER Ca²⁺ store content.

(B) Scatterplot summarizing the change in the basal 340/380 ratio following removal of extracellular Ca²⁺ in experiments similar to those in (A). Colored lines represent the mean of at least n = 3 experiments, and error bars represent SEM.

(C) Scatterplot summarizing the correlation between an elevated basal 340/380 Ca²⁺ ratio (Figure 2E) and the change in 340/380 ratio following treatment with 30 μM CPA (maximum CPA-induced amplitude – basal 340/380 ratio following removal of extracellular Ca²⁺) in experiments similar to those in (A).

(D) Representative traces of Ca²⁺ signals in the indicated cell lines in response to the addition of 10 μM GSK-7975a.

(E) Scatterplot summarizing the change in 340/380 ratio following the addition of GSK-7975a (average of 20,340/380 ratio points prior to GSK-7975a addition – average of 20,340/380 ratio points following 200 s of GSK-7975a addition) from experiments similar to those in (D).

(F) Scatterplot summarizing the correlation between an elevated basal 340/380 Ca²⁺ ratio (Figure 2E) and the change in 340/380 ratio following treatment with 10 μM GSK-7975a (E) in experiments similar to those in (D). All data are mean ± SEM of at least three (N = 3) independent experiments. Control HEK-3KO (blue), Endo. hR3 (purple), and Exo. hR3 (green) data in (B, C, and E), and F repeated from Figures 3 and 5 above. ^###p < 0.001 when compared to HEK-3KO; ^tttP < 0.001 when compared to Endo. hR3; ∗∗∗p < 0.001 when compared to Exo. hR3; and ∗∗∗(red)p < 0.001 when compared to other stably expressed hR3 R2524CM cell lines; one-way ANOVA with Tukey’s test performed in (B and E).

Discussion

While disease-associated mutations have been identified in all three human isoforms of the IP₃R, the functional consequences of mutations in IP₃R3 are yet to be determined. Previous reports using patient-derived fibroblasts with neuropathy-associated IP₃R3 mutations exhibited drastically disparate mRNA levels and protein expression of all three IP₃R isoforms.^77,80 While the patient’s fibroblasts exhibited a trend toward decreased Ca²⁺ signaling, the presence and potential compensation by endogenous IP₃R1 and IP₃R2 isoforms may have confounded elucidating the full, functional effect of these mutant IP₃R3. As a result, no consensus could be reached regarding any change in channel function observed in fibroblasts harboring the V615M or compound heterozygous R1850Q/R2524C mutation.^77,80 Our functional studies were designed to study the mutations as homotetramers on an unambiguously null background. We demonstrate that all mutations tested exhibited markedly reduced or absent function when expressed at levels close to endogenous hR3 levels in IP₃R-null HEK-3KO cells. Notably, our studies have identified a novel molecular and cellular phenotype associated with a subset of the variants tested. Over-expression of hR3 V615M, T1424M, and R2524C led to an elevated basal [Ca²⁺]_i, decreased ER store Ca²⁺ content, and subsequent inappropriate activation of SOCE through ORAI channels. This phenotype is consistent with these mutations resulting in a constitutively leaky IP₃R3 channel pore, revealed as over-expression results in Ca²⁺ leak through the receptor, effective coupling to the SOCE machinery, and subsequent Ca²⁺ influx across the PM that is not matched by Ca²⁺ clearance mechanisms resulting in an elevated basal [Ca²⁺]_i. In combination with reduced IP₃-stimulated activity, it is possible that even when expressed at endogenous levels, that the leaky phenotype of these variants contributes to the etiology of the disease in affected individuals through local, inappropriate activation of effectors. In this scenario, a blocker of the IP₃R3 pore may ameliorate disease, if other IP₃R subtypes present can compensate for the decreased mutant IP₃R3 channel activity.

Over-expression also revealed that hR3 V615M and T1424M retained, to some limited extent, their ability to be gated by elevated IP₃, while hR3 R2524C was refractory to stimulation and thus appears to form a dead channel. Only following expression at levels similar to, or greater than that of Exo. hR3 was channel function measurable. We have previously shown that increased expression of IP₃R mutants in HEK-3KO cells compared to IP₃R-null, chicken lymphocyte DT40-3KO cells facilitated measurable changes in the channel function when not previously evident.⁸⁰ These observations indicate that over-expression of channels allows for the visualization of changes in channel gating resulting in measurable cytosolic Ca²⁺ signals. This may result from increased clustering of IP₃R channels on the ER membrane possibly allowing for increased amplification of signals through Ca²⁺-induced Ca²⁺ release and interactions with licensing proteins such as KRAP.⁹⁰ Among the partially functional channels, different degrees of functionality were evident. Detectable Ca²⁺ release was only evident in a small percentage of hR3 V615M cells at expression levels above or equal to Exo. hR3 and at high [CCh]. Remarkably, when over-expressed, the T1424M hR3 mutation exhibited increased function when compared to WT hIP₃R3 at similar expression levels such that at lower concentrations of agonist, larger evoked responses were observed. The efficacy of Ca²⁺ release through this mutant was however somewhat reduced because the magnitude of Ca²⁺ release achieved at high [CCh] was diminished. A remaining question is why this mutant uniquely exhibits increased activity compared to WT only when over-expressed? As stated, over-expression might reasonably lead to receptor clusters of increased size with enhanced function. In turn, these channels appear to be inherently poorly functioning, but gating competent channels with a leaky pore and a possibility is that this Ca²⁺ leak sensitizes these channels to low levels of IP₃, ultimately resulting in augmented Ca²⁺ signals at low [CCh]. The observation that significant Ca²⁺ puff activity was observed in cells expressing this variant even without an elevation in IP₃ is consistent with this idea.

Structural changes underlying alterations in channel function of IP₃R1 and IP₃R2 have been proposed for mutations in the SD, LBD, and channel pore.^2,61,63,91 However, the mechanisms underlying the changes in channel function in the regulatory and coupling domain, where two of the neuropathy-associated IP₃R3 mutations are located, are harder to identify. As the human V615M and T1424M variants are all properly localized and would not be predicted to interfere with ligand binding or directly alter the channel pore structure, the functional effects of these mutations are likely mediated by the disruption of channel modulation by regulatory input or by channel gating mechanisms. Notably, in all mutations the amino acid substitution results in substitution to either a methionine residue or cysteine residue, both containing sulfur moieties. Methionine residues have a unique combination of size, flexibility, and chemistry to act as an effective latch for sulfur-aromatic interactions. Sulfur-aromatic interactions occur in the majority of native protein structures, yet little is known about their roles in ion channels and how this substitution in variants might alter function. Of note, both methionine and cysteine residues are subject to modification by reactive oxygen species (ROS) which have been shown to alter IP₃R activity.^92,93 Future work may involve investigating whether these mutant residues are modified by ROS.⁹⁴

In contrast to the uncertain mechanisms underlying changes in channel function in the neuropathy-associated IP₃R3 mutations in the regulatory and coupling domain, the R2524C mutation in IP₃R3 at the junction of the 6^th TM and the LNK domain likely directly alters the channel pore, similar to the G2498S mutation in IP₃R2.⁸⁰ Based on recently published structures of the IP₃R1^27,30 and IP₃R3 pores,³² the positively charged Arg2524 is proposed to interact with Asp2518 of the neighboring subunit. As a result, mutation of the positively charged Arg2524 to an uncharged Cys would likely alter the structure of the channel pore by hindering its interaction with Asp2518. This may result in a re-arrangement of Arg2524 and Asp2518 that in the absence of ligand binding weakens the hydrophobic interactions that hold the TM region together and thus results in a non-functional, open IP₃R3 channel, depletion of the ER, and activation of SOCE culminating in an increased basal [Ca²⁺]_i.

Given that the location of individual mutations is widely spread through different domains of the protein, together with the dissimilar effects on IP₃-gated activity, it is not clear whether a common mechanism is responsible for the leaky channel phenotypes and decreased activity observed. It seems likely that the structural disruption imparted by the amino acid substitution alters both long-distance IP₃-induced conformational changes from the LBD to the channel domain to different degrees, but also results in structural changes to the channel domain which dilate the pore without IP₃ binding. In the case of V615M and T1424M, in contrast to R2425C, it appears that the pore can be further expanded by the ligation of the LBD. Further high-resolution structural studies will be necessary to define how each mutation results in the altered channel structure.

Since all the mutations are present on a single ITPR3 allele, a question is whether the disease phenotype results from the variants acting in a dominant negative fashion when incorporated into IP₃R heterotetramers. Our previous study investigating mutants associated with ataxia in IP₃R1 or anhidrosis in IP₃R2⁸⁰ suggests that this is dependent on the region of the IP₃R harboring the mutant, but nevertheless is not necessarily intuitively predictable. For example, incorporation of a mutation in critical residues important for IP₃ binding within the LBD in a single monomer results in an inactive tetramer, consistent with the requirement for all LBDs to be occupied with IP₃ before channel opening.⁸⁴ Since IP₃R subtypes readily form heterotetrameric assemblies, this dominant negative scenario likely leads to disabling a large proportion of the complement of IP₃R expressed in a particular cell. Clearly, the extent of the disruption of activity by mutations in other regions such as the channel domain depends on how radical the changes in properties of the substituted amino acid are and how inter-subunit interactions are disrupted. For example, somewhat unexpectedly, while a mutation in the selectivity filter of the pore-forming region of the IP₃R2 resulted in a dead channel when expressed as a homotetramer, IP₃R2 retained significant activity when only two of the monomers harbored the variant. This concatenated receptor approach is powerful because the subunit components of the tetramer can be precisely defined as the cell machinery faithfully only assembles tetrameric IP₃R structures without re-arrangement of the linked subunits or forming higher order structures^84,95,96,97 that occurs in other systems.⁹⁸ We envision a similar concatenated IP₃R3 approach, with defined numbers of mutant monomers within the tetramer will be required to define whether these variants act as dominant-negative subunits to disable function.

Disruption of signaling in Purkinje cells of the cerebellum, which abundantly and almost exclusively express IP₃R1 is central to understanding the disturbances in gait characteristic of SCA, pontocerebellar ataxia, and Gillespie Syndrome-associated mutations in IP₃R1.^2,41,99 We have identified here two distinct molecular effects of neuropathy-associated mutations on IP₃R3 function, however, the impacted cell type, which underpins the pathological phenotype remains to be established. The pathophysiology of peripheral neuropathies is generally associated with abnormalities in genes that regulate myelin assembly and axonal transport resulting in demyelination and axonopathy. Mutations resulting in the dysregulation of myelin assembly, cytoskeletal structure, endosomal sorting and cell signaling, proteasome, and protein aggregation, and the mitochondria may result in CMT.^100,101,102 In addition to roles in several of the processes that may result in CMT, IP₃Rs have also been identified in Schwann cells, with the dominant isoform IP₃R3 expressed in distinct regions from the other isoforms in dense patches in the perinodal region in close proximity to connexin 32.^103,104 It is therefore possible that the unique location or predominant expression of IP₃R3 in Swann cells underlies the etiology of disease. The presence of IP₃R3 in this region and its role in the other processes that result in CMT, suggest that further work looking at the downstream effects of these IP₃R3 mutations on mitochondrial function, cell structure, propagation of Ca²⁺ signaling in Schwann cells may be necessary to identify the specific pathophysiological mechanism.

Limitations of this study

The activity of IP₃R3 variants in this study was monitored by assessing changes in [Ca²⁺]_i which is an indirect, global readout of the cellular Ca²⁺ signaling machinery. Further experiments with more direct measurements of IP₃R activity, for example using “on nucleus” patch clamp will be necessary to define the biophysical mechanism whereby these disease-associated mutants alter function. Similarly, electrophysiological measurements of SOCE in cells expressing these mutants will define how efficiently the mutants activate Ca²⁺ influx.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Purified Mouse Anti-IP3R-3	BD Transduction	Cat. #610313
GAPDH Monoclonal Antibody (6C5)	Invitrogen	Cat. #AM4300
Goat anti-mouse Dylight™ 800CW secondary antibody	Invitrogen	Cat. #SA5-35521
Chemicals, peptides, and recombinant proteins
Alexa Fluor 488	Invitrogen	Cat. #A28175
Fura-2/AM	Invitrogen	Cat. # F1221
Cyclopiazonic Acid (CPA)	TOCRIS	Cat. # 1235 CAS #18172-33-3
CRAC Channel Inhibitor IV, GSK-7975a	Sigma Aldrich	Cat. # 53453100001 CAS #1253186-56-9
Cal520-AM	AAT Bioquest	Cat. #21130
ci-IP3/PM	TOCRIS	Cat. #6210 CAS #1009832-82-9
EGTA-AM	Invitrogen	Cat. #E1219
Dulbecco’s Modified Eagle Medium (DMEM)	Gibco	Cat. #11995073
Fetal Bovine Serum	Gibco	Cat. #16-000-044
Penicillin/Streptomycin	Gibco/Life Technologies	Cat. #15140122
Geneticin Sulfate (G418)	Gibco/Life Technologies	Cat. #11811031
Pfu Ultra II Hotstart 2X Master Mix	Agilent	Cat. #600850
Pierce™ Protease Inhibitor Tablets, EDTA-free	Thermo Scientific	Cat. #A32965
Paraformaldehyde (PFA)	Sigma Aldrich	CAS#30525-89-4
Bovine Serum Albumin (BSA)	Fisher	Cat. #BP9700100
poly-D-lysine	Sigma Aldrich	CAS#25988-63-0
Carbamoylcholine chloride (Carbachol)	Sigma Aldrich	Cat. #C4382 CAS #51-83-2
Critical commercial assays
Dc protein assay kit	Bio-Rad	Cat. #5000112
Experimental models: Cell lines
IP3R Null HEK-293 Cell Line (HEK-3KO)	Kerafast	Cat. # EUR030
HEK293 endogenous hIP3R3	This Paper	N/A
HEK-3KO exogenous hIP3R3	This Paper	N/A
HEK-3KO exogenous hIP3R3 V615M	This Paper	N/A
HEK-3KO exogenous hIP3R3 T1424M	This Paper	N/A
HEK-3KO exogenous hIP3R3 R2524C	This Paper	N/A
Oligonucleotides
hIP3R3 T1424M forward primer: CTGCTAC GTAGACATGGAGGTGGAGATG	This Paper	N/A
hIP3R3 T1424M reverse primer: CTGCTACG TAGACATGGAGGTGGAGATG	This Paper	N/A
Recombinant DNA
hIP3R3	Dr Suresh Joseph	N/A
hIP3R3 V615M	Julika Neumann (KU Leuven)	N/A
hIP3R3 T1424M	This Paper	N/A
hIP3R3 R2524C	Julika Neumann (KU Leuven)	N/A
Software and algorithms
TILLvisION	TILL Photonics	N/A
SoftMax© Pro Microplate Data Acquisition and Analysis software	Molecular Devices	https://www.moleculardevices.com/products/microplate-readers/acquisition-and-analysis-software/softmax-pro-software#gref
Olympus CellSens Dimensions 2.3 (Build 189987)	Olympus	https://www.olympus-lifescience.com/en/software/cellsens/
Image Studio Lite Quantification Software	LICOR Biosciences	N/A
Fiji	Schindelin et al. 2012105	https://imagej.net/software/fiji/
FLIKA	Ellefsen et al. 2014106	https://flika-org.github.io/
GraphPad Prism	GraphPad software	https://www.graphpad.com/scientific-software/prism/
Other
FlexStation3	Molecular Devices	https://www.moleculardevices.com/products/microplate-readers/multi-mode-readers/flexstation-3-reader#gref
Amaxa Nucleofector	Lonza Laboratories	Cat. #AAB-1001
Odyssey infrared imaging system	LICOR Biosciences	https://www.licor.com/bio/imaging-systems
Olympus Fluoview 1000 microscope	Olympus	https://www.olympus-lifescience.com/en/laser-scanning/
Olympus IX81 inverted Total Internal Reflection Fluorescence Microscope (TIRFM)	Olympus	https://www.olympus-lifescience.com/en/microscopes/inverted/
Oil-immersion PLAPO TIRFM 60x objective lens/1.45 numerical aperture	Olympus	https://www.olympus-lifescience.com/en/landing/objectives/research/tirf/
Hamamatsu ORCA-Fusion CMOS camera	Hamamatsu	https://www.hamamatsu.com/us/en/product/cameras/cmos-cameras.html
Inverted epifluorescence Nikon microscope	Nikon Instruments	https://www.microscope.healthcare.nikon.com/products/inverted-microscopes

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, David Yule (David_Yule@urmc.rochester.edu).

Materials availability

Plasmids (hIP₃R3 V615M, hIP₃R3 T1424M, and hIP₃R3 R2524C) and cell lines (Endo. hR3, Exo. hR3, and IP₃R3 mutants) generated in this study are available upon request.

Experimental model and subject details

Cell culture

All cell lines used were derived from female, human embryonic kidney epithelial cells (HEK293) originally obtained from ATCC and their identity was authenticated by genomic sequencing prior to gene editing. HEK-3KO cells, HEK293 cells engineered through CRISPR/Cas9 gene editing for the deletion of the three endogenous IP₃R isoforms^80,107 and HEK293 cells modified by CRISPR/Cas9 to only express endogenous hIP₃R3 (Endo. hR3). Gene editing was confirmed by genomic sequencing.⁸⁴ The cells were grown at 37°C with 5% CO₂ in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum, 100 U/mL penicillin, 100 μg/mL streptomycin (Gibco/Life Technologies). HEK-3KO cells stably expressing mutant IP₃R3 (hR3 V615M, hR3 T1424M, and h/r3 R2524C) were grown at 37°C with 5% CO₂ in DMEM supplemented with 10% fetal bovine serum, 100 U/mL penicillin, 100 μg/mL streptomycin, and 1.5–2 mg/mL Geneticin sulfate (G418; Gibco/Life Technologies).

Method

Generation of IP₃R mutants

A two-step QuikChange mutagenesis protocol was used to introduce amino acid substitutions into cDNA encoding the human IP₃R3 (hIP₃R3; NM_002224.4) in pDNA3.1. Mutagenesis and all DNA modifications were carried out using Pfu Ultra II Hotstart 2X Master Mix (Agilent). Mutagenesis primers for hIP₃R3 T1424M forward (CTGCTACGTAGACATGGAGGTGGAGATG), hIP₃R3 T1424M reverse (CATCTCCACCTCCATGTCTACGTAGCAG) were synthesized by Integrated DNA Technologies (IDT). In addition to encoding the specified mutations, primers also silently introduced a restriction site for verification purposes. The coding regions for all constructs were confirmed by sequencing. hIP₃R3 V615M and hIP₃R3 R2524C mutant cDNAs were obtained from Julika Neumann (KU Leuven).

Generation of stable HEK cell lines

Transfection of HEK-3KO cells to exogenously express desired hIP₃R3 WT or mutant constructs stably was performed as previously described.^80,107 In brief, 5 million cells were pelleted, washed once with Phosphate Buffered Saline (PBS), and resuspended in a transfection reagent (362.88 mM ATP-disodium salt, 590.26 mM MgCl₂ 6.H₂O, 146.97 mM KH₂PO₄, 23.81mM NaHCO₃, and 3.7 mM glucose at pH 7.4). 4-6 μg of DNA was mixed with the resuspended cells and electroporated using the Amaxa cell nucleofector (Lonza Laboratories) program Q-001. Cells were allowed to recover for 48 h before passage into new 10 cm plates containing DMEM media supplemented with 1.5–2 mg/mL G418. Following 7 days of selection, cell colonies were picked and transferred to 24-well plates containing DMEM media supplemented with 1.5–2 mg/mL G418. Wells that exhibited growth were expanded and those expressing the desired constructs were confirmed by western blotting.

Cell lysis and SDS-PAGE analyses

HEK-3KO cells, Endo. hR1 cells, and HEK-3KO cells stably expressing IP₃R constructs were harvested by centrifugation (200 × g for 5 min), washed once with PBS, and solubilized in membrane-bound extraction lysis buffer containing 10 mm Tris-HCl, 10 mm NaCl, 1 mm EGTA, 1 mm EDTA, 1 mm NaF, 20 mm Na₄P₂O₇, 2 mm Na₃ VO₄, 1% Triton X-100 (v/v), 0.5% sodium deoxycholate (w/v), and 10% glycerol supplemented with a cocktail of protease inhibitors (Pierce). Lysates were incubated for a minimum of 30 min on ice and cleared by centrifugation (16,000 × g for 10 min) at 4°C. Protein concentrations in cleared lysates were determined using D_c protein assay kit (Bio-Rad) and 4x SDS gel loading buffer was subsequently added to 5 μg of lysate. Proteins were resolved using 8% SDS-PAGE and transferred to a nitrocellulose membrane (Pall Corporation). Membranes were probed with a mouse monoclonal antibody against IP₃R3 (R3; #610313, BD Transduction) at a 1:1000 dilution and GAPDH (AM4300, Invitrogen) at a 1:75,000 dilution, followed by the goat anti-mouse Dylight™ 800CW secondary antibodies at a 1:10,000 dilution (SA535521; Invitrogen). Membranes were imaged with an Odyssey infrared imaging system (LICOR Biosciences).

Immunocytochemistry and confocal microscopy

HEK-3KO cells stably expressing WT and mutant IP₃R3 constructs were plated on poly-d-lysine coated coverslips. At roughly 50% confluent, cells were fixed using 4% PFA at room temperature for 10 min. Subsequently, coverslips were washed with PBS, and cells were blocked in 10% Bovine Serum Albumin (BSA) for 1 h. Following blocking, cells were incubated in primary antibody against IP₃R3 (BD Transduction) overnight at 4°C. The following day, the primary antibody was removed, and coverslips were washed 3 times with PBS for 10 min with gentle rocking. Subsequently, the anti-mouse secondary antibody conjugated to Alexa Fluor 488 (Invitrogen) was incubated for 1 h at room temperature with gentle rocking. After incubation, coverslips were washed with PBS and mounted on slides. After allowing slides to dry, coverslips were sealed onto slides and imaged using confocal microscopy using an Olympus Fluoview 1000 microscope.

Single-cell Ca²⁺ imaging

Single-cell Ca²⁺ imaging was performed in intact cells as described previously.^80,107 Glass coverslips were plated with HEK-3KO, Endo. hR3, Exo. hR3, or HEK-3KO cells stably expressing IP₃R mutant constructs at least 18 h prior to imaging experiments. Subsequently, the glass coverslips were mounted onto a Warner chamber and the cells were loaded with 2 μM Fura-2/AM (Invitrogen) in Ca²⁺-containing imaging buffer (Ca²⁺ IB: 10 mM HEPES, 1.26 mM Ca²⁺, 137 mM NaCl, 4.7 mM KCl, 5.5 mM glucose, 1 mM Na₂HPO₄, 0.56 mM MgCl_2, at pH 7.4) at room temperature for 25 min. Following loading, cells were perfused with Ca²⁺ IB which provided a basal 340/380 ratio of the [Ca²⁺]. To obtain measurements of agonist-induced Ca²⁺ release into the cytoplasm cells were stimulated with 0.3–100 μM, CCh. To interrogate the leakiness of the expressed IP₃R, perfusion was switched from Ca²⁺ IB and Ca²⁺-free imaging buffer (Ca²⁺-free IB; 10 mM HEPES, 137 mM NaCl, 4.7 mM KCl, 5.5 mM glucose, 1 mM Na₂HPO₄, 0.56 mM MgCl_2, 1 mM EGTA at pH 7.4) and in the absence of extracellular Ca²⁺, cells were treated with 30 μM CPA (TOCRIS) to measure ER Ca²⁺ content. Finally, 10 μM GSK-7975a (Sigma Aldrich) was used to probe the contribution of Orai to cytosolic [Ca²⁺].

Ca²⁺ imaging was performed using an inverted epifluorescence Nikon microscope with a 40x oil immersion objective. Cells were alternately excited at 340 and 380 nm, and emission was monitored at 505 nm. Images were captured every second with an exposure of 15 ms and 4 × 4 binning using a digital camera driven by TILL Photonics software. Image acquisition was performed using TILLvisION software and data was exported to Microsoft Excel where means were calculated.

Population-based Ca²⁺ imaging

Population-based fluorescence imaging was carried out using a FlexStation3 (Molecular Devices), a high-throughput benchtop optical system utilizing micro-fluidics. HEK-3KO cells or HEK-3KO cells stably expressing IP₃R constructs were loaded with 4 μM Fura-2/AM in complete DMEM media. After 1 h, cells were harvested and subsequently washed, resuspended in Ca²⁺ IB, counted, and dispensed into a black-walled flat-bottom 96-well plate (∼300,000 cells/well). The plate was centrifuged (200 × g for 2 min) and placed at 37°C for 30 min prior to commencing the assay. Ca²⁺ IB and varying concentrations of CCh (0.1–100 μM CCh) were added to induce Ca²⁺ release through the IP₃R. Excitation of cells loaded with Fura-2/AM alternated between 340 and 380 nm and emission 510 nm. The total experiment time was set for 200 s, with readings taken every 4 s (∼51 readings/run). Data from SoftMax© Pro Microplate Data Acquisition and Analysis software was exported to Microsoft Excel where appropriate ratios, normalizations, and means were calculated.

Detection of Ca²⁺ puffs using TIRF microscopy

HEK-3KO cells exogenously expressing WT or mutant hIP₃R3 were cultured on 15-mm glass coverslips coated with poly-D-lysine (100 μg/mL) in a 35- mm dish for 36 h. Prior to imaging, the cells were washed three times with imaging buffer. The cells were subsequently incubated with Cal520-AM (5 μM; AAT Bioquest #21130) and ci-IP₃/PM (0.1 or 1 μM, Tocris #6210) in imaging buffer with 0.01% BSA in dark at room temperature. After 1-h incubation, the cells were washed three times with imaging buffer and incubated in imaging buffer containing EGTA-AM (5 μM, Invitrogen #E1219). After 45 min incubation, the media was replaced with fresh imaging buffer and incubated for additional 30 min at room temperature to allow for de-esterification of loaded reagents. Following loading, the coverslip was mounted in a chamber and imaged using an Olympus IX81 inverted Total Internal Reflection Fluorescence Microscope (TIRFM) equipped with oil-immersion PLAPO TIRFM 60x objective lens/1.45 numerical aperture. Olympus CellSens Dimensions 2.3 (Build 189,987) software was used for imaging. The cells were illuminated using a 488 nm laser to excite Cal-520 and the emitted fluorescence was collected through a band-pass filter by a Hamamatsu ORCA-Fusion CMOS camera. The angle of the excitation beam was adjusted to achieve TIRF with a penetration depth of ∼140 nm. Images were captured from equal areas using a 4 X 4 or 2 X 2 pixel binning (433.333 or 216 nm/pixel) at a rate of ∼166 or ∼50 frames/second by directly streaming into RAM. To photo-release IP₃, UV light from a laser was introduced to uniformly illuminate the field of view. Both the intensity of the UV flash and the duration (1 s) for uncaging IP₃ were optimized to prevent spontaneous puffs in the absence of loaded ci-IP₃. Images were exported as vsi files.

Quantification and statistical analysis

Quantification of SDS-PAGE

Membranes were quantified using Image Studio Lite (LICOR Biosciences), while mean of at least 3 experiments (n = 3) and SEM were calculated using GraphPad Prism (GraphPad).

Analysis of single-cell Ca²⁺ imaging

In Microsoft Excel mean basal [Ca²⁺], change in 340/380 amplitude, and percentage of responding cells for each experiment (multiple cells) was calculated. Area under the curve (AUC) was calculated using GraphPad Prism as increases above baseline that are greater than 10% of the distance from the minimum to the maximum Y using the 20 data points prior to agonist addition as the baseline. The mean and SEM of at least n = 3 experiments for each cell line was calculated in GraphPad Prism and statistical analysis was performed using one-way ANOVA with Tukey’s test.

Analysis of population-based Ca²⁺ imaging

In Microsoft Excel, peak amplitude of CCh-induced Ca²⁺ release was calculated by normalizing the 340/380 ratio to the average of the first 5 340/380 ratio data points. Area under the curve (AUC) was calculated using GraphPad Prism as increases above baseline that are greater than 10% of the distance from the minimum to the maximum Y using the five data points prior to agonist addition as the baseline. Means and SEM from at least 3 individual plates and nonlinear curve fits were calculated in GraphPad Prism. Statistical analysis was performed in GraphPad Prism using one-way ANOVA with Tukey’s test.

Analysis of Ca²⁺ puffs using TIRF microscopy

Images, 5 s prior and 30–60 s after flash photolysis of ci-IP₃, were captured, as described previously.^108,109 The vsi files were converted to tiff files using Fiji¹⁰⁵ and further processed using FLIKA, a Python programming-based tool for image processing.¹⁰⁶ From each recording, ∼100 frames (∼2 s) before photolysis of ci-IP₃ were averaged to obtain a ratio image stack (F/F₀) and standard deviation for each pixel for recording up to 30 s following photolysis. The image stack was Gaussian-filtered, and pixels that exceeded a critical value (1.0 for our analysis) were located. The ‘Detect-puffs’ plug-in was utilized to detect the number of clusters (puff sites), number of events (number of puffs), amplitudes and durations of localized Ca²⁺ signals from individual cells. All the puffs identified automatically by the algorithm were manually confirmed before analysis. The results from FLIKA were saved to Microsoft Excel and graphs were plotted using GraphPad Prism 8 where the mean and SEM were calculated for at least 3 indiviudal experiments. Statistical analysis was performed in GraphPad Prism using a two tailed t-test.

Published: December 22, 2022

Data and code availability

• •All data reported in this paper will be shared by the lead contact upon request.
• •This paper does not report original code
• •Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.