Cardiac inositol 1,4,5-trisphosphate receptors

Abstract

Calcium is a second messenger that regulates almost all cellular functions. In cardiomyocytes, calcium plays an integral role in many functions including muscle contraction, gene expression, and cell death. Inositol 1,4,5-trisphosphate receptors (IP₃Rs) are a family of calcium channels that are ubiquitously expressed in all tissues. In the heart, IP₃Rs have been associated with regulation of cardiomyocyte function in response to a variety of neurohormonal agonists, including those implicated in cardiac disease. Notably, IP₃R activity is thought to be essential for mediating the hypertrophic response to multiple stimuli including endothelin-1 and angiotensin II. In this review, we will explore the functional implications of IP₃R activity in the heart in health and disease.

Graphical abstract

1. Introduction

Cardiac hypertrophy is an adaptive response to a wide variety of cardiovascular diseases such as hypertension, ischemic insults, and cardiomyopathy [1–3]. Ventricular hypertrophy can occur in either the right or left ventricle and usually occurs as a compensatory mechanism to maintain normal contractility. Ventricular hypertrophy is characterized by increased ventricular wall thickness as a means to decrease wall tension and cardiac stress. Sustained pathological hypertrophic remodeling can lead to arrhythmias, heart failure, and sudden death [4].

Hormonal factors such as endothelin-1 (ET-1) are known to contribute to the development of cardiac hypertrophy. ET-1 is a potent vasoconstrictor known to increase in the circulation with age and during cardiac stress [5–8]. In the heart ET-1 regulates cardiac muscle function by modulating contractility, cardiomyocyte size, and cardiac gene expression [9, 10]. ET-1 binds and activates the G-protein coupled receptor endothelin receptor type A (ET_A) localized on the cardiomyocyte plasma membrane, leading to activation of phospholipase C (PLC) and production of inositol 1,4,5-trisphosphate (IP₃) and diacylglycerol (DAG) [11–13]. An increase in cytosolic IP₃ leads to calcium release via the IP₃R calcium channel. Increases in cytosolic calcium via the IP₃R can lead to modulation of cardiomyocyte contraction and activation of signaling pathways that modulate gene expression. Nuclear factor of activated T cells (NFAT) is a transcription factor that is well known to be one of the targets of the ET-1/IP₃R signaling cascade [14–16]. Calcium release through IP₃Rs leads to activation of calcineurin, which is a calcium-activated phosphatase. Dephosphorylation of cytosolic NFAT by calcineurin leads to the translocation of NFAT to the nucleus, where it initiates transcription of different genes that contribute to hypertrophic remodeling. The ET-1 signaling cascade has been extensively studied and it is known to play a crucial role during pathological hypertrophic remodeling of the heart [10, 14].

In the heart IP₃Rs are thought to play an important role by modulating calcium signals in response to extracellular stimulation. IP₃R expression levels have been shown to be increased in cardiac hypertrophy, failing myocardium, atrial fibrillation, ischemic dilated cardiomyopathy, and hypertension [17–19]. Cardiomyocyte IP₃R expression is less abundant compared to the main calcium release channel of the heart, the ryanodine receptor (RYR). However, during cardiac hypertrophy IP₃R expression and function is increased, and this is thought to contribute to cardiac remodeling [17, 18, 20]. To what extent IP₃Rs specifically contributes to the hypertrophic remodeling is not completely understood and somewhat controversial.

2. Regulation of Inositol 1,4,5-Trisphosphate Receptor Function

The IP₃R is a ligand-gated calcium channel that is primarily localized to the endoplasmic/sarcoplasmic reticulum (ER/SR). In some cell types, the IP₃R may also be found at the plasma membrane, Golgi apparatus, secretory vesicles, and perinuclear/nuclear membrane [11, 18, 21]. The IP₃R family is comprised of three isoforms that are encoded by three distinct genes. Each IP₃R isoform is about 300kDa and they form homo- and hetero- tetramers when co-expressed in different tissues [22–24]. The IP₃R N-terminus contains the ligand (IP₃) binding domain and the suppressor domain. The C-terminal domain has six transmembrane alpha helices that forms an ion conduction pore and a C-terminal tail that is thought to modulate gating by interacting with the ligand binding domain from an adjacent subunit [25]. The C-terminal tail is also the site of binding of multiple proteins that modulate channel function, as well as being targeted for post-translational modifications such as phosphorylation [26]. Between the N-terminal ligand binding domain and the C-terminal channel domain the IP₃R has a modulatory domain, which is an additional site of interaction for a variety of proteins and molecules that regulate channel function [27]. The modulatory domain is also a the site for post-translational modifications that are thought to further modulate IP₃-induced calcium release [28–30]. The three IP₃R isoforms share ~60% overall sequence identity, with the N- and C- terminal domains sharing the highest homology between isoforms [31, 32]. As expected, the three IP₃R isoforms also share similarities in gating and conductance characteristics consistent with their high sequence homology [33, 34]. IP₃R-1 is the main isoform expressed in the central nervous system (and in particular the cerebellum), and is critical for proper central nervous system function as evidenced by the hereditary spinocerebellar ataxias 15/16 and 29 that are caused by deletions/mutations in one allele of the ITPR1 gene [35, 36]. IP₃R-2 is expressed at high levels in the heart, pancreas, liver, and salivary glands. Similar to IP₃R-2, IP₃R-3 is also highly expressed in the pancreas and salivary glands [36]. IP₃R isoform expression is presumed to be regulated in a cell and tissue specific manner [36, 37], however the mechanisms regulating differential expression are unclear. Practically speaking, most tissues have detectable levels of all three IP₃R isoforms that can form both homo and heterotetramers [22, 24, 36, 38]. Thus, evaluating or interpreting isoform-specific function in vivo is a daunting task.

2.1. Allosteric Regulators Of IP₃-induced Calcium Release

Signal transduction pathways leading to calcium release classically originate at the plasma membrane (PM). Activation of PM receptors can lead to generation of the signaling molecules IP₃ and diacylglycerol through the action of phospholipase C (PLC). IP₃ binding to the IP₃R causes a conformational change that leads to calcium release from intracellular calcium stores. [39]. IP₃-induced calcium release (sometimes abbreviated IICR) can be allosterically regulated by calcium, ATP, and many proteins [40–47]. It has been shown using a wide variety of techniques that the different IP₃R isoforms are differentially regulated by IP₃ [31, 37]. It was later confirmed by single channel recording in planar lipid bilayers that IP₃R-2 has the highest affinity for IP₃, followed by IP₃R-1 and IP₃R-3 [33, 48]. Much of what we know about IP₃R function, including isoform-specific function, has been elucidated in the DT40 chicken B cell line in which each IP₃R isoform has been knocked out using homologous recombination either singly or in combination [48, 49]. B-cell receptor mediated calcium release in DT40 cells exclusively expressing IP₃R-2 was characterized by calcium oscillations that were more robust and longer-lasting compared to those observed in wild-type cells [48]. On response the other hand, IP₃R-1 mediated calcium release was characterized by an initial monophasic calcium that was followed by irregular calcium oscillations with smaller amplitude compared to IP₃R-2. The IP₃R-3 mediated calcium response was characterized by a single calcium release event, and in contrast to IP₃R-1 and -2, there were no additional calcium oscillations [48]. These studies depict the different functional characteristics of calcium release events mediated by each isoform, at least in the DT40 cell system in response to B cell receptor stimulation. Similar results were obtained using concatenated homo- and heterotetramers [50]. The ligand IP₃ is not the only regulator of IP₃R-mediated calcium release. It has been shown that calcium functions as a co-agonist for IP₃R calcium release at low concentrations while at higher calcium concentrations IP₃R activity is inhibited [51–55]. Therefore, calcium has the ability to bi-phasically regulate IP₃R function. These positive and negative feedback loops contribute to the complex spatio-temporal aspects of calcium release.

The crucial event that activates IP₃R calcium release is the binding of IP₃ to the IP₃R. IP₃R activation by IP₃ is known to be modulated by ATP. Binding of ATP to the IP₃R results in an increase in IP₃R open probability and this modulation is isoform specific [33, 40, 42, 56, 57]. Accordingly, each receptor has different affinities for ATP. IP₃R-2 is known to have the highest affinity, approximately 10 fold-higher compared to IP₃R-1. Interestingly, IP₃R-2 is modulated by ATP only at submaximal IP₃ concentrations [42]. IP₃R-3 is known to have a lower affinity compared to the other two isoforms [42]. ATP binds to a glycine rich region (GXGXXG) on the IP₃R that is commonly found in several ATP binding proteins and is known as a Walker motif. IP₃Rs have two of these motifs named ATP-binding motif A and B (ATPA/ATPB) [58]. It is known that the ATPA binding motif is exclusively found on IP₃R-1, while ATPB binding motif is conserved among all three isoforms. Using mutagenesis experiments, it was shown that mutation of the ATP binding motifs on IP₃R-1, ATPA and ATPB, does not abrogate ATP’s positive effect on IP₃R-1. Similarly, mutation of IP₃R-3 ATPB binding motif does not affect IP₃R-3 open probability in response to ATP. Thus, available evidence suggests that IP₃R-1 and IP₃R-3 have other still unknown ATP binding sites. In contrast, mutagenesis experiments done on IP₃R-2 show that ATPB is essential for ATP modulation of IP₃R-2 [42, 58]. Interestingly, a recent report using recombinant concatemeric IP₃R constructs has shown conclusively that at least two subunits of IP₃R-2 is sufficient to confer ATP sensitivity [59]. In contrast, other functional aspects such as IP₃ sensitivity display an intermediate phenotype. Thus, the difference in IP₃R activation/modulation by IP₃, calcium, and ATP highlights the unique characteristics of each IP₃R isoform that is further enriched by heterotetramer formation. However, the three IP₃R isoforms have been shown to be at least partially functionally redundant in vivo.

2.2. Isoform-Specific Function

It is difficult to assess the importance and role of an individual IP₃R isoform because they are co-expressed in almost all tissues [36, 37]. As noted above, in vitro studies using DT40 cells revealed differences in IP₃R function between isoforms, and this is likely mediated by distinct regulation by IP₃, calcium, and ATP. DT40 experiments with concatenated receptors also revealed unique properties of heterotetrameric channels [59]. Mutations in IP₃R channels causing human disease have been very instructive for establishing isoform-specific function in vivo. Heterozygous mutations of the ITPR1 gene cause spinocerebellar ataxia (SCA) 15 and 29 [60–64], which are progressive neurodegenerative disorders characterized by cerebellar ataxia and tremors. SCA16 is also caused by mutations in ITPR1, and it is now clear that SCA15 and SCA16 are clinically indistinguishable. As such, SCA16 is now considered a “vacant SCA” [65]. SCA15 is an autosomal dominant disorder characterized by a decrease in IP₃R-1 expression that leads to slow degeneration of Purkinje cells of the cerebellum where normally IP₃R-1 is highly expressed [61, 66]. Lower amounts of IP₃R-1 cause Purkinje cell degeneration, possibly by the dysregulation of intracellular calcium homeostasis in these cells. The related disorder SCA29 is distinguished clinically by earlier age of onset and cognitive impairment [63, 64]. At the molecular level, it is associated with a missense mutation in the regulatory domain. However, the functional consequences at the cellular level of the mutation observed in SCA29 disorder have yet to be determined [64]. Missense, nonsense, and in-frame deletion mutations in ITPR1 are also the cause of a very rare disease associated with ataxia known as Gillespie Syndrome [67, 68]. Interestingly, some Gillespie Syndrome patients have homozygous truncating mutations in the IP₃R protein resulting in what would be predicted to be a complete loss IP₃R-1 calcium release function [68]. Other mutants might function as dominant-negative subunits to inhibit channel function [68]. Thus, the loss of IP₃R-1 function appears to be compatible with life in humans. A homozygous missense mutation in ITPR2 is a rare cause of isolated anhidrosis with normal sweat glands [69]. The mutation is within the putative selectivity filter of the IP₃R pore, substituting glycine 2498 for serine. This mutation results in normal levels of IP₃R-2 expression in clear cells of the sweat gland, but loss of IP₃R-2 mediated calcium release. Thus the complete loss of IP₃R-2 calcium release function, like IP₃R-1, is compatible with life in humans. Currently, there are no known human diseases definitively linked to mutations in the IP₃R-3 protein, however single nucleotide polymorphisms in the ITPR3 gene are associated with increased risk of type 1 diabetes [70] and systemic lupus erythematosus [71].

Several mouse models have been used to explore the role of IP₃Rs in vivo. One example is the spontaneous mouse mutant opisthotonos (opt), which has an in-frame deletion of exon 43 and 44 of the itpr1 gene (IP₃R^opt). These two exons encode a portion of the modulatory domain that includes a key site that is essential for ATP binding and modulation of IP₃R-1 [72]. Opt mice display a severe phenotype characterized by seizures that can range from body tremors to tonic-clonic seizures. Homozygous opt mice show lower levels of functional IP₃R-1 expression [73]. Furthermore, using single channel recording it was shown that the IP₃R-1 opt mutant has a lower single channel conductance compared to the wild type channel [72].ATP and conductance alterations in the opt mutant may help explain the severe phenotype observed in opt mouse. Furthermore, a heterozygous knockout (IP₃R-1^−/+) mice model of the itpr1 exhibits impaired motor coordination manifested primarily on the Rota-Rod test [74]. Interestingly, this mouse model is mostly undistinguishable from WT mice. Double knockout (IP₃R-1^−/−) mice have a more severe phenotype that is mostly embryonic lethal, though a few mice survive to birth and display severe ataxia with seizures until death before weaning [35]. In both cases, IP₃R-1^−/− and IP₃R-1^−/+ mice show no significant anatomical differences in any organs including the brain, heart, and spleen when compared to wild-type littermates. Overall the clinical phenotype observed in both human and animal models with mutations on the IP₃R-1 is mainly due to high expression of this specific receptor in the cerebellum, with very important differences between the two models.

Global homozygous knockout of itpr2 in mice does not results in any obvious phenotypic abnormalities [75]. Similar to a loss of function ITPR2 mutation in patients, itpr2 double knockout mice exhibit reduced sweat production and diminished sweat gland responses to acetylcholine [69]. Homozygous knockout mice retain residual sweat secretion that was attributed to the expression of the other two IP₃R isoforms in mice [69], whereas in humans the IP₃R-2 isoform appears to be the primary isoform expressed in eccrine sweat glands.

Using double itpr2 and itpr3 knockout mice, it was found that these mice lack salivary secretion leading to low weight and eventually death caused by starvation [75]. Interestingly, itpr2 or itpr3 single mutants were indistinguishable from control showing functional redundancy of the two isoforms. Using submandibular glands of wild type, itpr2^−/−, itpr3^−/− or -3^−/−, and itpr2^−/−3^−/− mice, it was shown that double mutants exhibit impaired cholinergic and beta-adrenergic responses. The impairment of this response led to a decrease in the salivary secretion [75]. Furthermore, itpr2^−/−3^−/− mice also showed deficits in exocrine pancreas function leading to alterations in lipase, amylase, and trypsinogen secretion in pancreatic tissue. IP₃R-2 and IP₃R-3 single knockout mice were phenotypically indistinguishable from wild type mice, which would suggest that the overlapping function of these channels compensate for the lack of one IP₃R subtype, and that IP₃R-2 and IP₃R-3 may have redundant roles in vivo in mouse models [76, 77].

Double knockout of the itpr1 and itpr2 genes in mice leads to embryonic development defects of the ventricular myocardium, and as expected based upon the itpr1 knockout phenotype, the double knockouts die in utero [78]. Interestingly, the double knockout mouse model phenocopies NFAT triple knockout mice, including thin myocardial walls, poor trabeculation of the ventricles, poor ventricular cell proliferation and decreased NFAT nuclear translocation [78]. In contrast, single IP₃R mutants do not have the same physiological abnormalities that are seen in the double knockout, suggesting that during heart development IP₃R-1 and IP₃R-2 have redundant roles in regulating NFAT activation [78]. In general, both mouse studies and human diseases associated with IP₃R mutations support the notion that the different IP₃R isoforms have at least partially functional redundant roles in vivo, and the phenotypes observed in the knockout mice are due to restricted tissue-specific expression of the isoforms.

3. IP₃Rs And Calcium Homeostasis In The Heart

Calcium is a critical regulator of cardiac function. Cardiomyocyte contraction and relaxation is directly regulated by cytosolic calcium. Contraction is initiated by the activation of L-type voltage gated calcium channels (also known as the dihydropyridine receptor or Cav1.2 calcium channel). Calcium entry through L-type channels activates SR-localized ryanodine receptor (RYR) calcium channels via a process known as calcium-induced calcium release (CICR) [79]. Gating of the RYR allows SR calcium to be released into the cytoplasm where it binds to troponin C to initiate contraction. Relaxation is promoted when calcium is cleared from the cytoplasm primarily by the plasma membrane-localized sodium-calcium exchanger (NCX) and sarco/endoplasmic reticulum calcium ATPase (SERCA2), however other pumps, exchangers, and buffers can modulate relaxation [80]. In addition to the RYR, cytosolic calcium levels are modulated by IP₃Rs in the heart. IP₃R expression levels are thought to be relatively low compared to RYR, typically 50-fold less abundant than RYR2 [20, 81–83]. However, an important caveat is that these studies rely almost exclusively on the quantification of mRNA levels, not protein levels. Despite apparently low expression levels, IP₃Rs have been linked to modulation of cardiac contraction and activation of cardiac gene expression. Importantly, it has been reported that IP₃R expression and localization pattern are altered during heart failure and cardiac ventricular hypertrophy. IP₃R overexpression has been linked to cardiac remodeling in response to multiple stressors that lead to hypertrophy [84]. Overall, there is strong evidence that that IP₃Rs play an important role in the heart physiology in health and disease (Figure 1).

Figure 1. Control of cardiomyocyte function by IP₃Rs.

Activation of GPCRs coupled to PLC such as the ET_A receptor lead to the production of IP₃ and diacylglycerol. IP₃ then diffuses through the cytosol to bind and activate IP₃R channels. IP₃-mediated calcium release can modulate CICR and muscle contraction via modulation of RYR activity. Nuclear/perinuclear IP₃Rs are tough to modulate gene transcription.

IP₃R-2 is thought to be the predominant isoform found in the heart. Consequently, most of the studies that focus on the role of IP₃R in the heart have focused solely on IP₃R-2 function [83, 84]. Global genetic knockout of the IP₃R-2 in mice does not cause any significant difference in the hypertrophic response in pressure overload or dilated cardiomyopathy mouse models [85]. However, the potential involvement of IP₃R-1 and IP₃R-3 activity was not investigated further. Another group showed that global knockout of IP₃R-2 eliminates the positive inotropic effects of endothelin-1 (ET-1) in the atria and protects against arrhythmias [86]. In support of a role of IP₃R-2 in regulating cardiac hypertrophy, transgenic overexpression of the channel in the heart was sufficient to induce mild hypertrophy and exaggerated responses to ET-1 [15]. Targeted inducible overexpression of the IP₃-binding site of the IP₃R-1, also known as the IP₃“sponge”, was used to block cardiac signaling through all IP₃R channels [15, 87]. The IP₃ sponge acts by sequestering IP₃ generated by PLC-coupled agonists and preventing the activation of endogenous IP₃R channels. The IP₃ sponge inhibited hypertrophy in response to isoproterenol and angiotensin-II, but not a pressure overload model (transverse aortic constriction) [15]. Using an ischemic model (left anterior descending artery ligation), cardiac-specific deletion of the IP₃R-2 lead to improved cardiac function, reduced cell death, and reduced cardiac fibrosis [88], but did not reduce mitochondrial calcium overload and dysfunction [89]. Thus, there are numerous conflicting reports in mouse models that suggest that IP₃R-2 may (or may not) contribute to cardiac physiology and dysfunction in disease. However, there is a general consensus that the IP₃R-2 protein contributes at the very least to signaling downstream of ET-1 stimulation.

There have been few attempts to elucidate the role of IP₃R-1 and IP₃R-3 in the heart, either in vitro or in vivo, despite measureable expression levels [18, 19, 36, 81, 83, 85, 90, 91]. This may lead to erroneous conclusions regarding IP₃R function in the heart using IP₃R-2 mouse models, as several groups have reported that all three IP₃R isoforms are expressed in the heart with detectable mRNA and protein levels in mouse, rat and human ventricular cardiomyocytes (Table 1). In addition to total expression levels, the subcellular localization and precise aspects of how IP₃Rs regulate spatio-temporal aspects of calcium release in the contracting myocyte is an area of debate. In particular, how IP₃R signals are decoded to regulate processes such as transcriptional activation in the beating cardiomyocyte is unclear. IP₃Rs are known to be mainly localized at the ER/SR membranes in most cells, however in the heart it is thought to be concentrated at the nuclear and perinuclear membranes, where it is thought to play a key role in gene transcription [10, 90, 92, 93]. There is an abundant amount of evidence that suggests that the IP₃R and IP₃ producing machinery is localized in or adjacent to the nucleus of several cell types including cardiomyocytes [94–97], and this localization is thought to be important for spatially restricting calcium transients mediated by this channel to the nuclear matrix. It has been shown that IP₃ and ET-1 can trigger nuclear calcium sparks and nuclear localized calcium transients [90, 92, 93, 98, 99]. However, other studies argue against this notion and suggest that IP₃Rs are localized at the perinuclear membrane area, and IP₃-induced calcium transients from the cytosol can diffuse into the nucleus to increase nuclear calcium [10, 51, 100, 101]. Indeed, as the nuclear pore is freely permeable to calcium it is unclear why an intranuclear IP₃R channel would be required. The IP₃R has been shown to activate different transcription factors that initiate hypertrophic remodeling induced by hormonal stimulation such as ET-1 and angiotensin II (AngII) [10, 15, 102]. In turn, IP₃R calcium release activates calcium sensitive proteins such as calcineurin and CamKII leading to activation of transcription factors that are known to be localized in the cytoplasm such as NFAT or the nucleus such as HDAC5 [10, 92]. The role of NFAT in hypertrophic remodeling has been well studied [14, 15, 93, 103] and it has been shown that NFAT de-phosphorylation and nuclear translocation leads to expression of different genes and microRNAs (miRs) that modulate hypertrophic remodeling of the heart such as atrial natriuretic peptide (ANP), brain natriuretic peptide (BNP), and miR-23a [10, 103, 104]. As mentioned previously, the IP₃R nuclear/perinuclear localization is proposed as a mechanism by which IP₃Rs are able to specifically activate nuclear calcium transients without the disruption of the cytosolic calcium signals that are mediated by the RYR [10]. Experiments using calcium chelators targeted to the nucleus have shown quite definitely that nuclear calcium is required for hypertrophic gene expression [10], however this does not discriminate whether the calcium signal originated in the cytosolic compartment. Studies using a combination of nuclear and cytosolic chelators support the notion that IP₃R-dependent calcium transients originate in the cytosol and translocate to the nucleus [105]. It should be noted that most studies have relied on confocal imaging to infer “nuclear only” calcium transients, which is indirect at best. The use of genetically encoded calcium indicators targeted to the nucleus is a much more direct method, and has already been used to evaluate nuclear transients evoked by various stimuli [105]. However, additional experimental work needs to be done in order to completely understand how nuclear calcium is regulated in the healthy and diseased heart.

Table 1.

Isoform studied	Species or cell type used	Localization	Isoform (mRNA)	Isoform (protein)	Method of detection	Year Published	Comments	Ref
1	ARVM	ND	IP3R-1	IP3R-1	mRNA: In situ hybridization Protein: Immunohist ochemistry	1993		[81]
1 2 3	Adult rat heart	ND		IP3R-1 IP3R-2 IP3R-3	Protein: Western blot	1995		[36]
1	HLV myocardium (Healthy and CHF)	ND	IP3R-1		mRNA: In situ hybridization HLV myocytes Northern blot of HLV, HRV and septum	1995		[20]
1 2 3	Adult mouse heart	ND	IP3R-1 IP3R-2 IP3R-3		mRNA: RT- PCR	1997	Relative expression of each isoform: P3R-1 (6.1%) IP3R-2 (78.0%) IP3R-3 (15.9%).	[108]
1 2 3	Rat and ferret isolated ventricular myocytes	ND	IP3R-1 IP3R-2 IP3R-3		mRNA: RT-PCR Protein: Immunoblotting	1997	Relative expression of each isoform: P3R-1 (1.9%) IP3R-2 (84.6%) IP3R-3 (13.5%). At the protein level they looked at IP3R-1 expression and it wasn’t found in rat cardiac myocytes	[82]
1 2 3	3 week old rats with cardiac hypertrophy	ND	IP3R-1 IP3R-2		mRNA: in situ hybridizatio n	1998		[109]
1 2 3	ARVM and atrial cardiomyocytes	IP3R-2 localized near the sarcolemma and RYR in atrial myocytes	IP3R-1 IP3R-2 P3R-3	IP3R-1 IP3R-2	mRNA: RT-PCR Protein: Western blot	2000	At the mRNA level IP3R-2 is the most predominant isoform. At the protein level IP3R-1 and -2 are the most abundant in ARVM.	[83]
1 2	Neonatal Mouse cardiomyocytes	Striated localization SR pattern		IP3R-1 IP3R-2	Protein: Immunofluorescence	2002		[91]
2 3	Rabbit atrial and ventricular myocytes	ND		IP3R-2 IP3R-3	Protein: Western blot	2007	IP3R-2 and -3 are more abundant in atrial myocytes compared to ventricles.	[107]
1 2 3	NRVM	IP3R-2 found at the nuclear envelop IP3R-3 found at the cytosol	IP3R-1 IP3R-2 IP3R-3	IP3R-2 IP3R-3	mRNA: RT- PCR Protein: Western blot and immunofluorescence	2008	All the IP3R isoforms were detected at the mRNA level. Only IP3R-2 and -3 were detected at the protein level.	[90]
2	NRVM and ARVM	Perinuclear localization in NRVM Surrounding the nuclei and sarcoplasmic reticulum in ARVM		IP3R-2	Protein: immunofluorescence	2009		[10]
1 2 3	ASHR, normotensive WKY. HLV failing hearts	In WKY mainly nuclear and sarcoplasmic localization. In ASHR mainly localized at the SR, and co- localized with RYR2	IP3R-2	IP3R-2	mRNA: RT- PCR Protein: Immunofluor escence and Western blot	2009/2010	They looked at IP3R-1 and IP3R-3 protein expression by western blot. IP3R-1 and -3 is not found in ASHR or WKY cardiac myocytes	[17, 18]
2	AMVM	Sarcoplasmic localization		IP3R-2	Protein: Immunofluor escence and western blot	2010		[15]
1 2 3	AMVM		IP3R-1 IP3R-2 IP3R-3	IP3R-2	mRNA: RT- PCR Protein: Western blot	2013		[85]
1 2 3	HLV myocytes and MLV myocytes		IP3R-1 IP3R-2 IP3R-3	IP3R-2	mRNA: RT- PCR Protein: Western blot	2013	They used a pan-antibody to look at the protein expression of all 3 IP3R isoforms	[19]

ARVM, adult rat ventricular myocytes; HLV, Human left ventricle; CHF, Congestive heart failure; ND; Not determined; NRVM, Neonatal rat ventricular myocytes; ASHR, Adult spontaneously hypertensive rat; WKY, normotensive Wistar-Kyoto; AMVM, Adult mouse ventricular myocytes; MLV, Mice left ventricle

Calcium is directly responsible for cardiac contraction and relaxation. RYRs are the main modulators of the SR calcium release mediating excitation-contraction coupling (ECC) in cardiomyocytes. ECC is a process by which an action potential triggers a contraction via L-type calcium channel opening and subsequent RYR activation. IP₃Rs have been also shown to modulate ECC both in ventricular and atrial cardiomyocytes [83, 106]. However, the role of IP₃Rs in ventricular ECC is controversial due to lower expression levels in ventricular cardiomyocytes (at least as determined by mRNA levels) compared to other cell types such as atrial cardiomyocytes [20, 83]. Indeed, it has been suggested that IP₃R signaling does not contribute to the positive inotropic effects of ET-1 in ventricular cardiomyocytes [84]. In atrial cardiomyocytes the role of IP₃Rs has been extensively studied. It has been shown that after neurohormonal stimulation, IP₃R can modulate ECC by increasing action potential (AP) amplitude, calcium spark frequency, and by creating spontaneous calcium transients [83, 107]. In healthy ventricular cardiomyocytes, IP₃R is found in perinuclear/nuclear areas and near the RYRs at the dyadic cleft. IP₃R localization at the SR/dyadic cleft is believed to also modulate ECC in ventricular cardiomyocytes, and is characterized by increasing AP amplitude, increasing spontaneous calcium transient frequency, and decreasing resting membrane potential [18, 19, 106]. It is well established that IP₃R expression is increased in cardiac hypertrophy, heart failure, dilated cardiomyopathy, spontaneously hypertensive rats, and in transverse aortic constriction (TAC) models [17–20]. It was found that the increase in IP₃R expression was observed mainly in the dyadic cleft of salt hypertensive rats, and that IP₃Rs may contribute to ECC during disease by sensitizing nearby RYR for activation [18]. In disease states there are alterations in t-tubule morphology that compromise L-type channel/RYR coupling, and therefore an increase in IP₃Rs expression could be a way to compensate for the decreased coupling in failing heart by promoting CICR. A more recent study showed that failing human ventricular cardiomyocytes have a reduction in the resting membrane potential and AP prolongation, and suggested that IP₃Rs might modulate ECC using an alternative mechanism [19]. This report found that IP₃Rs are localized to specific domains in the cardiomyocyte that lack RYR, and IICR results in direct activation of NCX leading to AP prolongation and alterations in electric properties of cardiomyocytes [19]. Overall these studies indicate that IP₃R localization to the SR/ER leads to modulation of ECC in ventricular cardiomyocytes, providing a potential explanation for left ventricular arrhythmias.

4. Re-Evaluating IP₃R Expression And Function In The Heart

IP₃Rs are the primary regulators of intracellular calcium release in many cells. However, the dominant role of RYR in modulating calcium changes in the heart has made investigating the role of IP₃R signaling challenging. IP₃R was first shown to be expressed in cardiac myocytes over 20 years ago [81]. This seminal study was the first to show that IP₃R-1 was expressed in the heart. It was subsequently shown that IP₃R-1 is upregulated in human chronic heart failure tissue, whereas RYR2 expression was significantly reduced [20]. It was concluded that IP₃R-1 upregulation might be an alternative pathway to stabilize calcium homeostasis in end-stage heart failure [20]. Despite these early reports suggesting IP₃R-1 may have a functional role in the heart a series of studies emerged indicating that, at least at the mRNA level, IP₃R-2 predominates in both atrial and ventricular cardiomyocytes (Table 1).

More recently, the mRNA expression levels of the different IP₃R isoforms have been re-evaluated in both human and mouse tissue [19]. It was found that the relative expression level of the receptors varies between both species (human IP₃R-3>IP₃R-1>IP₃R-2 and mouse IP₃R-2> IP₃R-1> IP₃R-3). As mentioned above, a large amount of work using transgenic/knockout animals have suggested that IP₃R-2 plays at least some role in cardiac physiology [15, 86, 88], however this is quite controversial [85, 89]. This leads to the question as to whether IP₃R-1 or -3 are able to compensate for IP₃R-2 deficiency in mouse models. Unfortunately IP₃R-1/2 double knockout mice are lethal; however it is theoretically possible to make a conditional knockout of all three IP₃Rs in the heart. In addition, in vitro studies with targeted knockdown or knockout of individual isoforms in cardiomyocytes would be a useful endeavor to elucidate the potential functional redundancy of these isoforms in regulating cardiomyocyte physiology.

Highlights.

• Inositol 1,4,5-trisphosphate receptors (IP₃Rs) are ubiquitous regulators of intracellular calcium.

• The three IP₃R proteins have functionally redundant roles in vivo.

• IP₃Rs are expressed in the heart and may contribute to cardiac hypertrophy.

• There are many unanswered questions on how IP₃Rs regulate cardiomyocyte physiology in health and disease.