CREB regulates the expression of type 1 inositol 1,4,5-trisphosphate receptors

Inositol 1,4,5-trisphosphate (IP₃) receptor (IP₃R)-mediated redistribution of Ca²⁺ from the endoplasmic reticulum to the cytosol and other organelles occurs upon binding of IP₃ (and the co-agonist Ca²⁺). IP₃ is formed in response to a number of hormones, growth factors or neurotransmitters that activate phospholipase C isoforms (Berridge et al., 2000; Bootman et al., 2002; Clapham, 2007). The subsequent increase in intracellular Ca²⁺ regulates a myriad of physiological processes including secretion, contraction, gene transcription, proliferation, apoptosis, oxidative phosphorylation, learning and memory (Berridge et al., 2000; Clapham, 2007; Decuypere et al., 2011; Foskett et al., 2007). Humans have three types of IP₃Rs: type 1 (IP₃R1), type 2 (IP₃R2) and type 3 (IP₃R3), which are encoded by ITPR1, ITPR2 and ITPR3, respectively (Berridge, 1993; Bezprozvanny, 2005; Blondel et al., 1993; Furuichi et al., 1989; Südhof et al., 1991). Each isoform has a molecular mass of ∼300 kDa, they share 60-70% sequence similarity, and are assembled as homo- and hetero-tetrameric Ca²⁺ channels that predominantly localize to the endoplasmic reticulum (Mignery et al., 1989; Monkawa et al., 1995; Wojcikiewicz and He, 1995). IP₃R subtypes differ in their binding affinities for IP₃ (Iwai et al., 2007; Newton et al., 1994). Their activity is differentially modulated by various post-translational modifications, including phosphorylation by protein kinase A (PKA) (Masuda et al., 2010), PKC, CaMKII (Ferris et al., 1991b), AKT (Khan et al., 2006), ubiquitylation (Alzayady and Wojcikiewicz, 2005), glycosylation (Rengifo et al., 2007), redox state of cells (Joseph et al., 2018), binding to ATP (Bezprozvanny and Ehrlich, 1993) and interaction with other proteins (Prole and Taylor, 2016).

IP₃Rs are ubiquitously expressed; however, their expression is regulated in a tissue- and development-specific manner (De Smedt et al., 1997; Newton et al., 1994). For example, IP₃R1 is predominantly expressed in the central nervous system while IP₃R2 and IP₃R3 are enriched in the peripheral tissues (Alzayady et al., 2013; Furuichi et al., 1989; Wojcikiewicz, 1995). IP₃R1 knockout mice predominately die in utero or very early during post-natal development as a result of severe ataxia and seizures (Matsumoto et al., 1996; van de Leemput et al., 2007). Individual IP₃R2 or IP₃R3 knockout mice, by contrast, develop normally with subtle phenotypes; however, the dual IP₃R2–IP₃R3 knockout animal displays severe defects in salivary, pancreatic and lacrimal exocrine secretion, and die shortly after weaning due to their inability to assimilate macronutrients (Futatsugi et al., 2005). Moreover, perturbations in IP₃R signaling or mutations in IP₃Rs (Foskett, 2010; Terry et al., 2020) are associated with several pathological conditions, including Alzheimer disease (Cheung et al., 2010), Huntington disease (Tang et al., 2005), schizophrenia (Tsuboi et al., 2015), amyotrophic lateral sclerosis (van Es et al., 2007), spinocerebellar ataxia (Ando et al., 2018; Ogura et al., 2001), Gillespie syndrome (Gerber et al., 2016), anhidrosis (Klar et al., 2014), diabetes (Roach et al., 2006) and cancer (Ueasilamongkol et al., 2020).

Previous reports also indicate that the number of IP₃Rs can be acutely regulated. For example, Ca²⁺ influx through the L-type Ca²⁺-channels or NMDA receptors augments expression of IP₃R1 in neuronal cultures (Genazzani et al., 1999; Graef et al., 1999). Similarly, the heteromeric transcription factor activator protein 1 (AP-1) and nuclear factor of activated T-cells 4 (NFATc4) augments IP₃R1 expression in mouse cerebral cortical neurons (Mizuno et al., 2012, 2015). The box-I, which is located at position −334 to −318 bp relative to the transcription initiation site, in the cis-promoter region of Itpr1 was shown to be required for expression of IP₃R1 in mouse neural crest-derived neuroblastoma (N2a) cells (Konishi et al., 1997). Moreover, 12-O-tetradecanoylphorbol-13-acetate (TPA) or serum were shown to activate, whereas 1,25-dihydroxyvitamin D3 or 17β-estradiol were shown to inhibit ITPR1 promoter activity (Kirkwood et al., 1997). Likewise, TGF-β is known to increase the phosphorylation of IP₃R1 while inhibiting its expression in mesangial cells (Sharma et al., 1997). TGF-β is also known to inhibit the expression of IP₃R1 and IP₃R3 in pre-glomerular afferent arteriolar smooth muscle cells (Pacher et al., 2008). We have previously reported that TNF-α augments the activity and expression of IP₃R1 via the transcription factor SP1 (Park et al., 2008, 2009). Nuclear factor erythroid 2-related factor 2 (NFE2L2) inhibits the expression of IP₃R3 in both rat and human cholangiocytes by binding to musculo-aponeurotic fibrosarcoma recognition element (MARE) in the ITPR3 promoter (Weerachayaphorn et al., 2015). At the post-transcriptional level, microRNAs miR-133a and miR-506 inhibit the expression of IP₃R2 and IP₃R3, respectively (Ananthanarayanan et al., 2015; Drawnel et al., 2012). The expression of IP₃Rs is also induced by exposure to methamphetamine or cocaine (Kurokawa et al., 2012). However, additional molecular factors regulating the expression of IP₃Rs have so far remained largely underexplored.

Although several reports have clearly established the short-term/acute effects of raising intracellular levels of cAMP upon both the activity and phosphorylation status of IP₃Rs (Ferris et al., 1991a; Patel et al., 1999; Soulsby and Wojcikiewicz, 2005; Wagner et al., 2008, 2004, 2003), any long-term consequences of PKA activity have not been defined. In our present study, we examined the long-term effects of chronic stimulation of PKA on the expression of IP₃Rs in HEK-293 cells. Our results indicate that the endogenous protein levels of IP₃R1 are augmented in response to increasing PKA activity by exposure to forskolin, whereas the protein levels of IP₃R2 and IP₃R3 remain unaffected. We also show that this effect is mediated through the transcription factor cyclic AMP-response element-binding protein (CREB). Consistent with this observation, ectopic overexpression of constitutively active CREB in HEK-293 cells augmented protein levels of IP₃R1 and ITPR-interacting domain containing 2 (ITPRID2, also known as and hereafter referred to as KRAP), a binding partner of IP₃R that is important for agonist-induced Ca²⁺ release, localization and function (Fujimoto et al., 2011b; Thillaiappan et al., 2021). This transcriptional regulation of IP₃R1 levels had functional consequences, as dominant-negative CREB transfection significantly diminished agonist-induced Ca²⁺ release and, notably, ‘Ca²⁺ puffs’ – the elementary Ca²⁺ signals generated by IP₃Rs. Overexpression of VP16-CREB – a constitutively active form of CREB – increased the Ca²⁺ puff activity. Taken together, our results indicate a crucial role for CREB in governing the expression of ITPR1 through its endogenous promoter and, subsequently, the underlying agonist-evoked Ca²⁺ puffs by IP₃R1 in HEK-293 cells.

Previous results from our laboratory and other groups have conclusively documented that all the three IP₃Rs are phosphorylated by PKA (Wagner et al., 2008; Wojcikiewicz and Luo, 1998). The sites of phosphorylation and the acute functional consequences of phosphorylation regarding IP₃R activity have also been extensively studied (Soulsby et al., 2004). Nevertheless, the long-term effects of PKA activation on the endogenous protein levels of IP₃Rs, i.e. the stimulus–transcriptional coupling, have not been fully determined.

To investigate whether PKA activation has any long-term effects on the endogenous protein levels of IP₃Rs, we treated HEK-293 cells with 10 μM forskolin for 12 h to raise intracellular cAMP and activate PKA, and then determined IP₃R protein levels by western blotting. We found that the endogenous protein levels of IP₃R1 significantly increased following treatment with 10 μM forskolin (Fig. 1A,B) but that the endogenous protein levels of IP₃R2 (Fig. 1C,D) and IP₃R3 (Fig. 1E,F) remained unaltered. Moreover, treatment with 10 μM forskolin augmented endogenous IP₃R1 protein levels in a time-dependent manner (Fig. S1). Consistently, the IP₃R1 remained phosphorylated following 12 h treatment with 10 μM forskolin (Fig. S2). These results indicated that long-term treatment with forskolin augments both endogenous levels of IP₃R1 protein and the extent to which the receptor remains phosphorylated in HEK-293 cells.

Forskolin activates adenylyl cyclase, resulting in increased levels of intracellular cAMP (Insel and Ostrom, 2003). cAMP binds to the regulatory subunits of PKA, with the important consequence that free catalytic subunits translocate into the nucleus, where they phosphorylate and activate the transcription factor CREB. Once phosphorylated, CREB recruits the transcriptional machinery required for the transcription of target genes (Altarejos and Montminy, 2011). Since forskolin treatment augmented the endogenous protein level of IP₃R1, we employed several in silico transcription factor-binding prediction tools, including Alibaba2.1, TFBIND, Consite and PROMO, which all rely on different algorithms to predict CREB-binding sites within the 1 kb proximal promoter domains of IP₃Rs. Interestingly, three out of four tools predicted at least one CREB-binding site within the proximal promoter domain of all the three IP₃Rs (Fig. 2A). In addition, these tools also predicted binding sites for other transcription factors, such as SP1, EGR1 and NF-KB amongst others. This concurs with a previous report by our group wherein we demonstrated that SP1 is involved in regulating the gene expression and protein synthesis of IP₃R1 (Park et al., 2009). To functionally validate our in silico predictions, we performed luciferase assays using reporter constructs comprising the ITPR1 or ITPR2 promoter. Co-transfection of the 1 kb proximal ITPR1 promoter-reporter construct with VP16-CREB expression plasmid resulted in a dose-dependent increase in promoter activity (Fig. 2B). VP16-CREB binds the cAMP response element (CRE) through the DNA-binding domain of CREB and transcriptional activation is facilitated by transcriptional activation domain of the herpes virus VP16 protein (Riccio et al., 1999). In contrast, the ITPR1 promoter activity diminished upon co-transfection with a dominant-negative form of CREB (i.e. KCREB) (Fig. 2C) or upon treatment with 20 µM of the PKA inhibitor H89 (Fig. 2D). KCREB plasmid, encoding a dominant-negative form of CREB, was transfected into HEK-293 cells to block endogenous CREB (Walton et al., 1992). The reduction in ITPR1 promoter activity following co-transfection of KCREB or exposure to H89 suggested that activation of PKA under basal conditions plays an important role in regulating the transcription of ITPR1 in these cells, which is consistent with the observation that basal PKA activity is relatively high in HEK-293 cells (Mignen et al., 2005). A similar increase in the promoter activity was observed when the ITPR2 promoter-reporter construct was co-transfected with the VP16-CREB expression plasmid (Fig. 2E) that previously had been shown to be regulated by CREB (Kruglov et al., 2017). These results indicate a potential role for CREB in governing expression of both ITPR1 and ITPR2.

To determine whether there is any correlation between the transcript levels of IP₃Rs and CREB, we next mined transcriptomics data for CREB1, ITPR1, ITPR2 and ITPR3 from GTEx database, a publicly available database. The transcript levels of ITPR1, ITPR2 and ITPR3 displayed significant positive correlations when compared with those of CREB1 in 54 different tissue types (Fig. 2F–H; Table S1). Taken together with the predictions of in silico transcription factors and our in vitro promoter-reporter assays, these results indicate a role for CREB in regulating the expression of IP₃Rs across various human tissues.

To dissociate the acute effects of IP₃R1 phosphorylation and to demonstrate the involvement of CREB in regulating the expression of IP₃Rs, increasing amounts of the constitutively active VP16-CREB-expression plasmid was transfected into HEK-293 cells followed by western blotting to determine the endogenous protein levels of the three IP₃Rs subtypes. Interestingly, transient transfection of constitutively active VP16-CREB augmented the endogenous protein levels of IP₃R1 (Fig. 3A,B), whereas levels of IP₃R2 and IP₃R3 remained unaltered (Fig. S3A,B). To substantiate the role of CREB regarding the levels of IP₃R1, the dominant-negative KCREB plasmid was transfected into HEK-293 cells to inhibit endogenous CREB. In this case, western blotting revealed that KCREB transfection resulted in a concentration-dependent decrease in the endogenous IP₃R1 protein levels within HEK-293 cells (Fig. 3C,D). These results indicated that, under basal conditions, CREB regulates the levels of IP₃R1 protein and that IP₃R1 is a bona fide target of CREB in HEK-293 cells.

Since treatment of cells with forskolin activates PKA, we assessed the role PKA has in regulating the expression of IP₃R1 through CREB. For this, a PKA inhibitor (PKI) plasmid (Day et al., 1989) was transiently transfected into HEK-293 cells followed by western blotting to probe for endogenous IP₃R1 levels. As expected, transfection of the PKI plasmid resulted in a decrease in endogenous IP₃R1 protein levels within HEK-293 cells (Fig. 3E,F), which further emphasizes a role of PKA in regulating IP₃R1 under basal conditions. Our results indicate that the PKA–CREB axis is involved in regulating the expression of IP₃R1 in HEK-293 cells.

Does PKA also play a similar role in regulating the expression of IP₃R1 in other cell lines? To address this, we treated human neuroblastoma SH-SY5Y and HeLa cells with the PKA inhibitor H89 for 12 h. Compared with untreated controls, the endogenous levels of IP₃R1 protein diminished in both SH-SY5Y and HeLa cells upon treatment with H89 (Fig. S4A,C). These results indicate a role for PKA in governing IP₃R1 expression across various cell lines.

To determine the involvement of CREB in mediating the effect of PKA activation above basal activity, HEK-293 cells were transiently transfected with either pcDNA as a control or KCREB, followed by treatment with 10 μM forskolin and western blotting analysis. Transfection of KCREB, i.e. expression of dominant-negative CREB, resulted in decreased IP₃R1 levels, whereas treatment with 10 μM forskolin enhanced levels further (Fig. 3G,H). Forskolin-induced endogenous IP₃R1 levels were modestly diminished upon blocking endogenous CREB (Fig. 3G,H). These results further highlight a role of CREB in forskolin-induced IP₃R1 expression.

We next investigated the impact of CREB on agonist-induced Ca²⁺ release from HEK-293 cells that only express IP₃R1 at endogenous levels. These cells were engineered using CRISPR-Cas9 technology through deletion of ITPR2 and ITPR3 normally expressed in HEK-293 cells (Alzayady et al., 2016). These cells – hereafter referred to as hR1 endo cells – were utilized in all subsequent experiments to prevent any contribution of IP₃R2 and IP₃R3 to Ca²⁺ release by using a heterotetrameric population of IP₃Rs. Transfection of hR1 endo cells with either pcDNA or KCREB, and Ca²⁺ release in response to the muscarinic receptor agonist carbachol (CCh) was monitored by using a Flexstation plate-reader assay platform with microfluidics. The increase in cytosolic Ca²⁺ in response to increasing concentrations of CCh (0.1–100 μM) was then determined. Notably, following stimulation with sub-maximal concentration of CCh (10 µm), Ca²⁺ increase in KCREB-transfected cells was significantly less compared to cells transfected with pcDNA (Fig. 4A). These data demonstrated that KCREB transfection of hR1 endo cells results in decreased IP₃R1 expression, consistent with decreased release of Ca²⁺ in response to agonist stimulation.

In contrast, HEK-293 cells that only express IP₃R3 – hereafter referred to as hR3 endo cells (Alzayady et al., 2016) – and were transfected with equal amounts of either pcDNA or KCREB showed no significant difference in agonist-induced Ca²⁺ release (Fig. 4B). Indeed, these results are in agreement with our observations that the endogenous protein levels of IP₃R2 and IP₃R3 remain unaltered upon overexpressing CREB (Fig. S3A,B), suggesting that the effect of CREB is specific to only the IP₃R1 subtype in HEK-293 cells.

Given that the agonist-induced Ca²⁺ release is inhibited in a population-based assay with, presumably, non-uniform transfection efficiency, we next performed single-cell imaging using an epifluorescence microscope under which transfected cells could be optically identified. For single-cell imaging, hR1 endo cells were co-transfected with mCherry-3HA (herein referred to as mCherry) plasmid and either pcDNA or KCREB. Following 36 h of transfection, cells were loaded with 2 μM Fura-2/AM to monitor changes in cytosolic Ca²⁺ in response to increasing doses of CCh. In response to stimulation with CCh at varying concentrations, mCherry-positive hR1 endo cells transfected with KCREB displayed much lower changes in fluorescence ratio (Fig. 5B,E) compared to mCherry-positive hR1 endo cells transfected with pcDNA (Fig. 5A,D). Consistent with this, the average increases in amplitude and the peak amplitude changes tended to be lower in cells transfected with KCREB compared to those transfected with pcDNA (Fig. 5C,F,H). Moreover, the percentage of KCREB-transfected cells responding to different concentrations of CCh at a specific threshold (F/F₀>0.02, where F/F₀ represents the change of fluorescence intensity relative to basal fluorescence) were diminished when compared to pcDNA transfected cells (Fig. 6G). Overall, these results provide mechanistic insights of CREB having a key role in the regulation of IP₃R1 levels and agonist-evoked Ca²⁺ release.

To further investigate the effect of CREB on IP₃R1 protein levels, we used total internal reflection fluorescence (TIRF) microscopy to measure elementary Ca²⁺ signals, i.e. Ca²⁺ puffs, evoked following the uncaging of a cell-permeable form of caged IP₃ (ci-IP₃) in hR1 endo cells. Paradoxically, these Ca²⁺ signals represent a small majority of IP₃R clusters that are active upon stimulation and are known as ‘licensed’ IP₃R (Thillaiappan et al., 2017). For this purpose, hR1 endo cells co-transfected with mCherry and pcDNA, VP16-CREB or KCREB plasmids were loaded with Cal-520 AM and caged IP₃(ci-IP₃/PM) followed by EGTA-AM, and Ca²⁺ puffs were recorded after photolysis of ci-IP₃ for 40 s at a rate of ∼97 frames per second. TIRF measurements of Ca²⁺ puffs following ci-IP₃ uncaging were designed to facilitate analysis of the fundamental properties of small IP₃R1 clusters, which are less influenced by other processes. In turn, this might influence the characteristics of the Ca²⁺ signal observed in global measurements of whole cells. Ca²⁺ puffs were detected throughout the 40 s recording period following photolysis of ci-IP₃ in hR1 endo cells co-transfected with mCherry and pcDNA, VP16-CREB or KCREB plasmids (Fig. 6A–C). We specifically imaged and analyzed only mCherry-positive cells (Movies 1, 2, and 3). Interestingly, the number of Ca²⁺-puff sites per cell were significantly decreased in hR1 endo cells transfected with KCREB plasmid (65, n=9 cells) compared to those in cells transfected with pcDNA (281, n=9 cells) or VP16-CREB (368, n=9 cells) (Fig. 6E). Although the number of Ca²⁺-puff sites per cell was slightly higher in cells transfected with CREB compared to those transfected with pcDNA, this increase was not statistically significant (Fig. 6E). Furthermore, similar to the number of Ca²⁺-puff sites per cell, the number of Ca²⁺ puffs per cell remained significantly diminished in KCREB-transfected cells (109, n=9 cells) compared to those in pcDNA-transfected cells (490, n=9 cells) (Fig. 6D). In contrast, the number of Ca²⁺ puffs per cell were significantly elevated in VP16-CREB-transfected cells (1483, n=9 cells) compared to those in pcDNA-transfected cells (Fig. 6D). The amplitudes (Fig. 6F) and the mean time of a rise (r) and fall (f) in Ca²⁺-puff fluorescence (Fig. 6G) induced by photolysis of ci-IP₃ remained similar in hR1 endo cells transfected with pcDNA, KCREB or VP16-CREB plasmid, indicating that the fundamental biophysical properties of IP₃R1 clusters remained unaltered in these cells. These data further substantiate the involvement of CREB in governing the expression of IP₃R1, which is destined to be active following agonist stimulation.

Based on the previous reports that KRAP is required for proper localization of IP₃Rs and in defining their ability to respond to IP₃ (Fujimoto et al., 2011a; Thillaiappan et al., 2021), we performed in silico analysis to investigate potential CREB binding sites within the 1 kb proximal promoter region of ITPRID2 (encoding KRAP). Interestingly, by using various transcription factor prediction tools, our in silico analysis revealed putative CREB-binding sites in the proximal promoter domain of ITPRID2 (Table S2). Furthermore, we observed an increase in endogenous protein levels of KRAP following treatment with 10 μM forskolin (Fig. 7A,B). To functionally validate the involvement of CREB in regulating ITPRID2 expression, we transfected HEK-293 cells with VP16-CREB or KCREB plasmids followed by western blotting to determine the KRAP protein levels. Overexpression of CREB in HEK-293 cells augmented the endogenous levels of KRAP (Fig. 7C,D), whereas blockage of endogenous CREB by using the KCREB-expression plasmid diminished endogenous KRAP levels in HEK-293 cells (Fig. 7E,F). Moreover, endogenous KRAP levels were reduced upon transfection of the PKI-expression plasmid (Fig. 7G,H). Consistently, endogenous KRAP levels were also diminished in SH-SY5Y and HeLa cells treated with H89 (Fig. S4B,D). These results suggest a role for CREB/PKA in governing the expression of ITPRID2 under basal and forskolin-induced conditions thus, contributing to the proper localization and appropriate activity of IP₃Rs.

In total, our results provide molecular insights to the crucial role of CREB in modulating the expression of both ITPR1 and ITPRID2 in HEK-293 cells. We speculate that aberrations in the levels of any of these three proteins might be an underlying cause for various pathological conditions associated with dysregulated cAMP and/or Ca²⁺ signaling.

IP₃Rs are ubiquitously expressed and, often, more than one subtype is present in any given cell type. The functional channel is assembled as a tetramer and can be homo or heterotetrameric (Foskett et al., 2007). The acute channel activity of the IP₃R is regulated by a host of regulatory factors, including IP₃, Ca²⁺, ATP and protein-binding partners (Ferris et al., 1990; Foskett and Daniel Mak, 2010; Prole and Taylor, 2016). In addition, IP₃Rs can be phosphorylated by several kinases. In particular, an early observation was that – in response to elevated cAMP levels – IP₃Rs are phosphorylated by PKA, thereby representing a major focus of crosstalk between canonical signaling pathways (Yule et al., 2010). Although the acute effect on channel activity by regulatory factors, including those that phosphorylate PKA, has been extensively studied, there are relatively few reports of more-chronic regulatory events that have the potential to control IP₃R expression per se. For example, previous reports from our laboratory and other groups have established the roles of SP1 (Park et al., 2008, 2009), AP-1 and NFAT (Mizuno et al., 2012, 2015) in regulating ITPR1 transcription. Similarly, the gene expression and protein synthesis of IP₃R1 are also governed by several cellular factors including TNF-α (Park et al., 2009) and TGF-β (Pacher et al., 2008; Sharma et al., 1997). However, the roles of additional factors influencing the expression of IP₃R1 remain obscure. As an important consequence of PKA activation in many cell types is to phosphorylate the transcription factor CREB – which leads to its nuclear translocation and the subsequent expression of target genes. In this current study, we investigated whether IP₃Rs are targets of transcriptional regulation by CREB and whether modulating CREB activity has any functional consequences on agonist-induced Ca²⁺ release via IP₃Rs.

We observed a significant increase in the endogenous IP₃R1 protein levels upon treatment of HEK-293 cells with 10 µM forskolin for 12 h. Protein expression of IP₃R2 and IP₃R3 subtypes in these cells, however, remained unaltered (Fig. 1). These observations are supported by evidence from the CREB target gene database, wherein the binding of CREB to the ITPR1 promoter was augmented following treatment with forskolin for 4 h in HEK-293 T cells (binding ratio 1.7) (Zhang et al., 2005). Contrary to our observations in HEK-293 cells, a recent report provided conclusive evidence for involvement of CREB in augmenting the expression of IP₃R2 levels upon fasting or treatment of hepatocytes with forskolin (Kruglov et al., 2017). This is consistent with our observation that the reporter activities of both ITPR1 and ITPR2 are augmented upon overexpression of VP16-CREB in HEK-293 cells (Fig. 2). Because IP₃R2 protein levels were unaltered in HEK-293 cells, we speculate that IP₃Rs are differentially regulated by CREB, perhaps, due to differences in epigenetic modifications in a cell type-specific or tissue-specific manner.

Next, our in silico predictions using various tools regarding transcription factor binding revealed putative binding sites for CREB within the 1 kb proximal promoter regions of all three IP₃R subtype genes. In agreement with this, our promoter-reporter assays suggest a potential role for CREB in governing the expression of ITPR1 and ITPR2. Moreover, RNAseq data mined using the GTEx portal revealed a significant positive correlation between transcript levels of ITPR2 and ITPR3, and CREB across various human tissues (n=54), underscoring a potential role for CREB in regulating the gene expression of all the three IP₃R subtypes (Fig. 2). However, overexpression of CREB in HEK-293 cells resulted in a dose-dependent increase in the reporter activity and endogenous IP₃R1 protein levels, and was without any effect on the protein levels of IP₃R2 and IP₃R3 subtypes (Fig. S3). By contrast, blocking CREB resulted in a decrease in IP₃R1 levels in HEK-293 cells (Fig. 3), suggesting a key role for CREB in regulating the expression of ITPR1 even under basal conditions. Interestingly, the ITPR1 promoter harbors a TATA box, whereas promoters of both ITPR2 and ITPR3 lack a typical TATA box (Konishi et al., 1997). Initial screening showed that CREB binds to proximal TATA-box-containing promoter regions (Zhang et al., 2005); however, later investigations provided evidence that CREB can bind to distal promoter regions as well (Everett et al., 2013). This idea is further supported by the fact that ITPR2 (which lacks a typical TATA box) can be regulated by CREB (Kruglov et al., 2017) (Fig. 2E). It is established that CREB is not recruited to promoters that are methylated (Zhang et al., 2005) and, thus, it is possible that the ITPR2 and ITPR3 promoters are methylated in HEK-293 cells, therefore, leading to differentially regulated gene expression of IP₃R subtypes.

CREB is a widely expressed transcription factor. During fasting, phosphorylation of CREB at Ser133 is promoted through glucagon via PKA. In turn, phosphorylated CREB augments the expression of gluconeogenic genes, including pyruvate carboxylase (PC), phosphoenolpyruvate carboxykinase 1 (PCK1) and glucose-6-phosphatase (G6PC), peroxisome proliferator-activated receptor-gamma coactivator 1 alpha (PPARGC1A) (Koo et al., 2005). Of note, glucagon also phosphorylates IP₃R1 via PKA, which results in increased cytosolic Ca²⁺ levels. This increase in cytosolic Ca²⁺ promotes CRTC2 dephosphorylation and nuclear localization of CREB-regulated transcription coactivator 2 (CRTC2), in turn, enhancing the expression of gluconeogenic genes (Wang et al., 2012). Furthermore, gluconeogenesis is promoted through glucagon via IP₃R1 in a transcriptional-independent manner (Perry et al., 2020). Strikingly, D2D transgenic mice, which are characterized by disruption of the Itpr1 loci as well as Itpr1 heterozygous mutant mice [also known as opisthotonos (opt/+) mice], exhibit glucose intolerance, thereby supporting a role for IP₃R1 in glucose homeostasis (Ye et al., 2011). We observed a decrease in IP₃R1 protein levels following PKI-mediated inhibition of PKA, the basal activity of which is reported to be high in HEK-293 cells (Mignen et al., 2005), as well as upon H89 treatment of SH-SY5Y and HeLa cells, both of which predominantly express the IP₃R1 subtype (Thillaiappan et al., 2021; Van Acker et al., 2002). Moreover, IP₃R1 protein levels slightly diminished in response to forskolin after blocking CREB with its dominant-negative form KCREB (Fig. 3G,H). Our results indicate that PKA, in addition to its role in phosphorylating IP₃R1, also governs the expression of IP₃R1 in various cell lines (Fig. S4A,C). This evidence suggests that Ca²⁺ release through IP₃R1 may act in a positive feedback loop via CREB to regulate ITPR1 expression.

Does blocking the function of CREB have any effect on agonist-induced Ca²⁺ release from IP₃R1 receptors? Indeed, transient transfection of KCREB in a population-based assay revealed a modest decrease in agonist-induced Ca²⁺ release from hR1 endo cells (Fig. 4A). However, such decrease in agonist-induced Ca²⁺ release was not observed in KCREB-transfected hR3 endo cells, reinforcing the notion that CREB does not regulate expression of ITPR3 in HEK-293 cells (Fig. 4). We verified these results by performing single-cell imaging in hR1 endo cells transfected with the KCREB plasmid (Fig. 5).

Small but statistically significant changes were observed at lower concentration of CCh (10 µM) although no statistically significant changes were seen at higher agonist concentration. Agonist-induced Ca²⁺ release has been described within a hierarchy of Ca²⁺ release events. Ca²⁺ release from individual IP₃Rs results in blips; increased levels of IP₃ together with positive feedback through Ca²⁺-induced Ca²⁺ release (CICR) (Foskett et al., 2007) can recruit additional IP₃Rs to generate larger events – termed Ca²⁺ puffs – that arise from a small cluster of IP₃Rs (Bootman et al., 1997; Smith and Parker, 2009; Thillaiappan et al., 2017). At greater stimulus strengths, additional IP₃R clusters are engaged through this positive feedback and Ca²⁺ signals globalize. The relatively small changes in agonist-stimulated whole-cell Ca²⁺ signals observed both in population-based and single-cell assays might be explained by the significant positive feedback due to CICR, which is measured in whole-cell assays. Additionally, at greater stimulus strengths, significant Ca²⁺ influx occurs; again, potentially, obscuring any effects of increased IP₃R levels. Notwithstanding the effects on global Ca²⁺ signals, the consequence of modulating CREB activity is – as discussed below – more significant regarding the fundamental building blocks of Ca²⁺ signals.

Research in the past decade has revealed that stimulation of all three IP₃Rs can evoke Ca²⁺ puffs of, essentially, identical properties (Lock et al., 2018; Mataragka and Taylor, 2018). Each cell type expresses several IP₃Rs, most of which are mobile and diffuse freely in the endoplasmic reticulum. In response to various stimuli or upon uncaging ci-IP₃, Ca²⁺ puffs arise repeatedly from a small fraction of localized and immobile IP₃Rs. These clusters of active IP₃Rs have been termed licensed IP₃Rs (Thillaiappan et al., 2017). KRAP associates with the N-terminal residues of IP₃Rs and co-immunoprecipitates with IP₃Rs both in vivo as well as in various cell lines, including HEK-293 and HeLa cells (Fujimoto et al., 2011a). Further, siRNA-mediated knockdown of KRAP diminished IP₃R-mediated Ca²⁺ release and immobilization/licensing of IP₃Rs (Fujimoto et al., 2011b; Thillaiappan et al., 2021). Thus, evidence indicates that KRAP plays a crucial role in both immobilizing IP₃R to the actin cytoskeleton and defining the activity of the licensed IP₃Rs. We, therefore, asked whether regulation of KRAP through CREB, in addition to that of IP₃R1 levels, plays a role in augmenting Ca²⁺ puff activity by, potentially, increasing the availability of this important binding partner. We observed an increase in endogenous KRAP levels upon treatment with forskolin. Similarly, the KRAP levels were regulated following overexpression of VP16-CREB, or by blocking CREB or PKA (Fig. 7 and Fig. S4B,D). In addition to regulating the expression of IP₃R1, these findings suggest that CREB/PKA activation plays a vital role in licensing of IP₃R1 through modulating the expression of KRAP. Interestingly, we observed a significant decrease in the number of Ca²⁺ puffs and Ca²⁺-puff sites following uncaging ci-IP₃/PM upon blocking CREB in hR1 endo cells compared to control cells. This is in agreement with reduced Ca²⁺ puffs and fewer Ca²⁺-puff sites upon siRNA-mediated knockdown of KRAP in HeLa cells (Thillaiappan et al., 2021). In contrast, the number of Ca²⁺ puffs significantly increased upon overexpression of VP16-CREB (Fig. 6), similar to those upon overexpression of KRAP in HeLa cells. We did not observe a significant increase in the number of Ca²⁺-puff sites upon overexpression of VP16-CREB; however, the number of Ca²⁺-puff sites significantly increased upon overexpression of KRAP in HeLa cells. These differences might be attributed to the use of different cell lines or the extent of increase in endogenous KRAP caused by overexpression of VP16-CREB (∼0.5-times compared with endogenous KRAP levels in HEK-293 cells) versus overexpression of KRAP (∼10-times compared with endogenous KRAP levels in HeLa cells) (Thillaiappan et al., 2021). The observed differences might also arise due to variations between HEK-293 and HeLa cells regarding the percentage of IP₃Rs and KRAP colocalizing within the TIRF field close to the plasma membrane. Nevertheless, our results underscore the functional consequences that the modulation of CREB activity, and thus IP₃R1 and KRAP, has for Ca²⁺ puffs in hR1 endo cells.

In conclusion, by using HEK-293 cells as a model cell line, our study indicates a crucial role for CREB in governing not only the expression of IP₃R1 under basal and forskolin-induced conditions but also its proper localization by regulating the levels of KRAP and, thus, the licensing of IP₃R1s destined to respond to various cues (Fig. 8). We speculate that dysregulation of CREB-mediated regulation of IP₃R1 has implications in a number of pathophysiological conditions that remain to be further investigated.

VP16-CREB expression plasmid, a constitutively active form of CREB, was kindly provided by David Ginty (Harvard Medical School, Boston, MA). KCREB plasmid, a dominant-negative form of CREB, was kindly provided by Richard Goodman (Vollum Institute, Portland, OR). ITPR2 promoter-reporter construct was kindly provided by Michael Nathanson (Yale University, New Haven, CT). pRL-TK plasmid was kindly provided by Cesare Orlandi (University of Rochester, Rochester, NY). pRSV-PKI-v2, expressing the PKA inhibitor PKI, was obtained from Addgene (#45066). pcDNA 3.1+ was from Invitrogen, USA. mCherry-3HA plasmid was generated in our laboratory. The ITPR1 promoter-reporter construct was generated by cloning the PCR amplified 1 kb proximal promoter region of ITPR1 (NM_001099952) between XhoI and HindIII sites in pGL3-Basic Vector (Promega, USA).

The sequences for upstream 1 kb proximal promoter regions of ITPR1 (NM_001099952), ITPR2 (NM_002223), ITPR3 (NM_002224), and KRAP (NM_006751) were retrieved from UCSC genome browser. Predictions for putative transcription factor-binding sites in the proximal promoter domains were performed using Alibaba2.1, TFBIND, Consite and PROMO tools. For correlation analysis between the transcript levels of ITPRs and CREB across various tissues, RNAseq data were mined using the GTEx portal.

HeLa, HEK-293, hR1 endo and hR3 endo cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin and 100 μg/ml streptomycin in 10 cm² cell culture dishes. SH-SY5Y cells were cultured in 1:1 mixture of Eagle's minimum essential medium (EMEM) and F12 medium supplemented with 10% FBS, 100 U/ml penicillin and 100 μg/ml streptomycin in 10 cm² cell culture dishes. All cell lines were obtained from ATCC. For transient transfections, cells were seeded in appropriate cell culture dishes. Following 24 h seeding, cells were transfected with indicated amounts of pcDNA, VP16-CREB, KCREB or PKI plasmids using Lipofectamine 2000 Transfection Reagent (Thermo Fisher Scientific, USA) according to the manufacturer's instructions. pcDNA was used as a balancing plasmid to transfect equal amounts of DNA in each well. Similarly, hR1 or hR3 endo cells were transfected with pcDNA or KCREB prior to imaging. In certain instances, cells were co-transfected with mCherry plasmid at a 1:10 ratio (mCherry:plasmid of interest). Cells were left in an incubator for 36 h post transfection before proceeding to western blotting or Ca²⁺ imaging.

HEK-293 cells cultured in 12-well plates were co-transfected with 250 ng of ITPR1 or ITPR2 promoter-reporter construct, indicated amounts of VP16-CREB or KCREB, 100 ng pRL-TK plasmid (as internal control, for normalization). pcDNA was used as a balancing plasmid to transfect equal amount of DNA in each well. Following 24–36 h transfection, cells were harvested in 1 ml PBS supplemented with 0.5 mM MgCl₂ and 0.1% glucose and centrifuged at 500 g for 5 min. The supernatant was discarded and cells were resuspended in 500 µl fresh PBS supplemented with 0.5 mM MgCl₂ and 0.1% glucose. Subsequently, 50 µl of the resuspended cells was added to each well of a white 96-well plate. To each well, 50 µl firefly luciferase substrate resuspended in lysis buffer (Promega Bright-Glo E2610) was added and incubated at room temperature in the dark for 15 min. Following 15 min incubation, luminescence was read using a plate reader. Freshly prepared Renilla buffer [1.1 M NaCl, 2.2 mM EDTA, 0.44 mg/ml BSA, 0.22 M KH₂PO₄ pH 5.1, 3 µM e-Coelenterazine (CAT#355-500, NanoLight Technology)] was added (50 µl/well), incubated at room temperature for 5 min before recording luminescence by using a plate reader. Firefly luciferase values were normalized to Renilla luciferase from respective wells and graphs are presented as percentage over control. Each experiment was performed in triplicates and repeated at least three times. To determine the effect of the PKA inhibitor H89 on ITPR1 promoter activity, HEK-293 cells were co-transfected with 250 ng of ITPR1 and 100 ng pRL-TK plasmid. Following 24 h transfection, cells were treated with 20 µM H89 for 12 h.

For western blotting, total protein was isolated from HEK-293, SH-SY5Y, HeLa control cells or cells treated with 10 μM forskolin or 20 μM H89 for 12 h using membrane-bound extraction buffer [10 mM Tris-HCl, 10 mM NaCl, 1 mM EGTA, 1 mM EDTA, 1 mMNaF, 20 mM Na₄P₂O₇, 2 mM Na₃VO₄, 1% Triton X-100 (v/v), 0.5% sodium deoxycholate (w/v), and 10% glycerol] supplemented with protease inhibitors. Similarly, total protein was isolated from cells transfected with pcDNA, VP16-CREB, KCREB or PKI plasmids 36 h post transfection. To determine the involvement of CREB in mediating the effects of forskolin, cells were treated with forskolin 24 h after transfection with pcDNA or KCREB and left for an additional 12 h before protein isolation. Briefly, for protein isolation, following addition of the appropriate amount of lysis buffer, cells were harvested in 1.5 ml tubes and placed on ice for 30 min. To disrupt the pellet, tubes were vortexed for 10 s every 10 min and returned on ice. Following 30 min incubation on ice, cells lysates were centrifuged at 16,000 g at 4°C for 10 min. The supernatant was transferred to newly labeled tubes. Protein concentration in the lysates was estimated using Dc protein assay kit (Bio-Rad). Lysates (30 μg) were then subjected to SDS-PAGE and transferred to a nitrocellulose membrane. Membranes were incubated with indicated primary antibodies and appropriate secondary antibodies before imaging with an Odyssey infrared imaging system (LICOR Biosciences). Band intensities were quantified using Image Studio Lite Ver 5.2 and presented as ratios of IP₃Rs/KRAP to GAPDH. The IP₃R1 antibody (#ARC154) was from Antibody Research Corporation and used at 1:1000 dilution. IP₃R2 antibody against amino acids 320–338 (NT2) was generated by Pocono Rabbit Farms and Laboratories, and used at 1:1000 dilution. IP₃R3 antibody (#610313) was from BD Transduction Laboratories and used at 1:1000 dilution. Phosphorylated IP₃ receptor (Ser1756) antibody (#3760) was from Cell Signaling Technology and used at 1:1000 dilution. Antibody for KRAP (#14157-1-AP) was from Proteintech and used at 1:1000 dilution. GAPDH (#AM4300), secondary goat anti-rabbit (SA535571) and secondary goat anti-mouse (SA535521) antibodies were from Invitrogen and used at 1:10,000 dilution.

hR1 or hR3 endo cells in 10 cm² cell culture dishes were transfected with 5 μg pcDNA or KCREB plasmid. Following 36 h transfection, the cells were loaded with 4 μM Fura-2/AM in cell culture medium and left in the dark for 1 hour. The cells were subsequently washed with imaging buffer (10 mM HEPES, 1.26 mM Ca²⁺, 137 mM NaCl, 4.7 mM KCl, 5.5 mM glucose, 1 mM Na₂HPO₄, 0.56 mM MgCl₂, at pH 7.4). Equal numbers (300,000 cells/well) of cells transfected with pcDNA and KCREB were seeded into each well of a black-walled 96-well plate. The cells were centrifuged at 200 g for 2 min to coat the bottom of the wells and left in an incubator for another 30 min before imaging. Fura-2/AM imaging was carried out by alternately exciting loaded cells at 340 nm and 380 nm; emission was monitored at 510 nm using FlexStation 3 (Molecular Devices). Peak response to various concentrations of CCh (0.1–100 μM) was determined using SoftMax Pro Microplate Data Acquisition and Analysis software as described previously (Terry et al., 2020). Data from at least three individual plates after curve fitting using a logistic dose-response equation in GraphPad Prism 8 are presented.

hR1 endo cells seeded on 15 mm glass coverslips in 35-mm dishes were co-transfected with mCherry plasmid (to optically identify transfected cells) and either pcDNA or KCREB. Following 36 h transfection, cells were washed with imaging buffer. Glass coverslips were attached to a Warner perfusion chamber using vacuum grease. The cells were incubated with 2 μM Fura-2/AM for 25 min in the dark at room temperature for loading. Cells were then perfused with imaging buffer and stimulated with indicated concentrations of CCh. Ca²⁺ imaging was performed using an inverted epifluorescence Nikon microscope equipped with a 40× oil immersion objective. Fura-2/AM imaging was carried out by alternately exciting loaded cells at 340 nm and 380 nm; emission was monitored at 505 nm. Images were captured every second with an exposure of 20 ms and 4×4 binning, using a digital camera driven by TILL Photonics software as previously described (Terry et al., 2020). Image acquisition was performed using TILLvisION software and data were exported to Microsoft excel. Experiments were repeated at least three times and only cells positive for mCherry were analyzed.

hR1 endo cells grown on 15-mm glass coverslips coated with poly-D-lysine (100 μg/ml) in a 35-mm dish for 24 h were co-transfected with 100 ng mCherry and 1 μg of pcDNA, VP16-CREB or KCREB plasmids. Following 36 h transfection and prior to imaging, cells were washed three times with imaging buffer. The cells were subsequently incubated with the Ca²⁺ dye Cal520-AM (5 µM; AAT Bioquest #21130) and caged IP3 (ci-IP₃/PM; 1 μM, Tocris #6210) in imaging buffer with 0.01% BSA in the dark at room temperature. After 1 h incubation, 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 medium 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, coverslips were mounted in a chamber and imaged using an Olympus IX81 inverted total internal reflection fluorescence microscope (TIRFM) equipped with oil-immersion PLAPO OTIRFM 60× objective lens/1.45 numerical aperture. Olympus CellSens Dimensions 2.3 (Build 189987) software was used for imaging. Cells positive for mCherry were identified by using a 561 nm laser. Subsequently, cells were illuminated by using a 488 nm laser to excite Cal-520 and 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 a final field of 86.7 µm×86.7 µm (400×400 pixels, one pixel=216 nm) at a rate of ∼97 frames/second (binning 2×2) by directly streaming into RAM. To photorelease 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 Ca²⁺ puff activity in the absence of loaded ci-IP₃. Images were exported as vsi files. Images, 5 s before and 40 s after flash photolysis of ci-IP₃, were captured as described previously (Arige et al., 2021; Emrich et al., 2021).

The vsi files were converted to tiff files using Fiji software and further processed using FLIKA, a Python programming-based tool for image processing (Ellefsen et al., 2014). From each recording, ∼300 frames (∼3 s) were averaged before photolysis of ci-IP₃, to obtain a ratio image stack (F/F₀, with  F=fluorescence, F₀=basal fluorescence) and standard deviation for each pixel for ≤30 s following photolysis. The image stack was Gaussian filtered, and pixels that exceeded the critical value of 1.0 (for our analysis) were located. The ‘Detect-puffs’ plug-in was used to detect the number of clusters (Ca²⁺-puff sites), number of events (number of puffs), amplitudes and durations of localized Ca²⁺ signals from individual cells. All Ca²⁺ puffs, identified automatically by the algorithm, were manually confirmed before analysis. The results from FLIKA analysis were saved as excel files and graphs were plotted using GraphPad Prism 8.

All statistical tests were conducted using GraphPad Prism 8 and data are presented as the mean±s.e.m. Statistical significance was determined by Student's t-test (unpaired, two-tailed) or one-way ANOVA with Tukey's multiple comparisons post-test. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 with respect to control condition. ns, not significant.

The authors thank David Ginty (Harvard Medical School, USA) for providing the VP16-CREB expression plasmid construct. The authors also thank Richard Goodman (Vollum Institute, USA) for providing the KCREB plasmid construct. The authors thank Michael Nathanson (Yale University, USA) for providing the ITPR2 promoter-reporter construct. The authors thank Cesare Orlandi (University of Rochester, USA) for technical advice with luciferase assays and for providing pRL-TK plasmid. The authors thank all the members of the lab, especially Kai-Ting Huang and Amanda M. Wahl, for their valuable suggestions.

The peer review history is available online at https://journals.biologists.com/jcs/article-lookup/doi/10.1242/jcs.258875.