A Bitter Taste in Your Heart

The human genome contains ∼29 bitter taste receptors (T2Rs), which are responsible for detecting thousands of bitter ligands, including toxic and aversive compounds. This sentinel function varies between individuals and is underpinned by naturally occurring T2R polymorphisms, which have also been associated with disease. Recent studies have reported the expression of T2Rs and their downstream signaling components within non-gustatory tissues, including the heart. Though the precise role of T2Rs in the heart remains unclear, evidence points toward a role in cardiac contractility and overall vascular tone. In this review, we summarize the extra-oral expression of T2Rs, focusing on evidence for expression in heart; we speculate on the range of potential ligands that may activate them; we define the possible signaling pathways they activate; and we argue that their discovery in heart predicts an, as yet, unappreciated cardiac physiology.

TAS2R/T2Rs (gene and protein) were first discovered within type II taste receptor cells in the tongue and act as sentinels in protecting against the ingestion of potentially toxic substances Lu et al., 2017). Since these pioneering studies, T2R expression has been reported in a multitude of extra-oral tissues, including the gut, lungs, brain, and heart (Shah et al., 2009;Foster et al., 2013;Garcia-Esparcia et al., 2013), but their complete function(s) in physiology and pathophysiology remain to be defined. In Table 1, we have summarized the location, expression profile and proposed function for the T2R family across a range of human tissues and cells. In regard to function, we would offer a note of caution that a number of studies (listed in Table 1) have proposed functions based on stimulation with various bitter compounds in the micromolar to millimolar range where the selectivity and specificity toward T2Rs may reasonably be questioned. Despite this, the expression of T2Rs within the cardiovascular system, particularly the heart and vasculature, has gained significant interest in recent years. Following our initial discovery of TAS2Rs within the heart (Foster et al., 2013), a number of subsequent studies have focused on the vasculature (Lund et al., 2013;Manson et al., 2014;Upadhyaya et al., 2014;Chen et al., 2017). An unambiguous definition of their function has, however, lagged behind the capacity to demonstrate their expression.

MODEL SYSTEMS FOR EXPRESSING T2Rs AND DEFINING THEIR FUNCTION
In attempting to define the function of, and to identify ligands for, the T2Rs, researchers have established heterologous expression systems in human cells (e.g., HEK293 or HEK293T) . However, the use of these cells for understanding the underlying mechanisms and signaling pathways within cardiovascular tissues/cells has obvious limitations. Firstly, due to the insufficient cell surface targeting of T2Rs in heterologous cells , chimeric T2Rs encompassing the amino terminus of the rat somatostatin receptor subtype 3 are often used to improve expression and functionality (Bufe et al., 2002;Behrens et al., 2006). Furthermore, a chimeric G protein consisting of the Gα 16 and 44 amino acids of gustducin attached to the carboxyl terminus is widely used in calcium mobilization assays (Liu et al., 2003;Ueda et al., 2003). Gα 16 has been coined the 'universal adaptor' due to its ability to interact with numerous GPCRs and provides a robust readout for receptor activation, including for T2Rs (Ueda et al., 2003). While these artificial heterologous systems have proven useful in identifying ligands for orphan receptors  and interrogating the structure-function aspects of T2Rs , the field is now moving toward more relevant cellular models with endogenous receptors and signaling partners (Freund et al., 2018). Studies using the aforementioned heterologous expression system have demonstrated that the majority of T2Rs form oligomers, both homodimers and heterodimers . However, unlike the situation for umami/sweet taste sensation (requiring dimerization of T1R1/T1R2 and T1R1/T1R3), T2R homodimers did not appear to alter the pharmacology of the receptors, nor do they have obvious influence on protein expression or membrane localization . In contrast, Kim et al. (2016) used immuno-fluorescent microscopy to show that the co-expression of the adrenergic (ADRβ 2 ) receptor with T2R14 resulted in a ∼3-fold increase in cell-surface expression of T2R14. Co-immunoprecipitation and biomolecular fluorescence complementation experiments confirmed that the increase of cell-surface expression was attributed to the formation of T2R14:ADRβ 2 heterodimers. These complexes may be particularly important in heart where the actions of adrenergic receptors are well described. Interestingly, coimmunoprecipitation and co-internalization of ADRβ 2 :M71 OR (mouse 71 olfactory receptor) was observed in response to their specific ligands (Hague et al., 2004). These seminal observations in heterologous systems need to be confirmed and extended with endogenous models to clarify our understanding of how T2Rs function in vivo and to define their potential modulation of (or by) established GPCRs.
Another important issue in considering model expression systems for studying T2Rs is the requirement for appropriate accessory proteins and correct post-translational processing. It is now well-established that chemosensory receptors [e.g., odorant (McClintock et al., 1997) and pheromone (Loconto et al., 2003) receptors] rely on endogenous proteins in order to be targeted to the cell-surface. A study by Behrens et al. (2006) demonstrated that certain members of the receptor-transporting protein (RTP) and receptor expression enhancing protein (REEP) families enhance cell-surface localization and functionality of certain TAS2Rs, likely through protein-protein interactions. Furthermore, it was shown that varying combinations of these proteins are expressed endogenously within tissues (circumvallate papillae and testis) that express TAS2R genes. Interestingly, the human heart differentially expresses REEP 1, 2, 3, 5, and 6 across heart regions (Doll et al., 2017), suggesting that efficient cell-surface TAS2R expression may also be region specific. Nonetheless, these trafficking proteins do not universally promote T2R functionality, for instance T2R14 showed no increase in capacity to mobilize calcium when coexpressed with either RTP or REEP (Behrens et al., 2006). There is accumulating evidence that the degree of T2R membrane insertion is dependent on the specific tissue. As T2Rs are detected in a myriad of tissues, multiple endogenous mechanisms may contribute to their appropriate expression and localization. As for many GPCRs, N-glycosylation of T2Rs is important for cell-surface localization- Reichling et al. (2008) reported that glycosylation of the second extracellular loop is essential for the recruitment (via association with the cellular chaperone calnexin) and insertion of TAS2Rs in the cell membrane; moreover, the function of non-glycosylated TAS2R16 could be rescued when co-expressed with RTP3 and RTP4.

THE CARDIAC GPCR REPERTOIRE INCLUDES T2Rs
The human heart expresses over 200 different GPCRs (Wang et al., 2018), some of which are critical for regulating cardiac morphology and function (Capote et al., 2015). Intriguingly, the gene transcripts for more than half of the TAS2R family were detected in both left ventricle and right atria (Foster et al., 2013) ranging in abundance between that observed for two classically important cardiac GPCRs -the angiotensin II type 1 receptor and β 1 -adrenergic receptor (ADRβ 1 ). It is notable that the expression of TAS2R14 was equivalent to that of ADRβ 1 in the left ventricle. These findings are supported by publicly available Illumina Human BodyMap 2.0 project RNA-seq dataset (Flegel et al., 2013), which showed widespread TAS2R expression in human tissues and highest expression of TAS2R14 in heart. It is important, however, to note that T2Rs are not uniformly detected by all techniques, with TAS2R9, TAS2R39, and TAS2R45 not detected in the Illumina RNA-seq data set, but detected by qPCR (Foster et al., 2013). These differences could reflect individual variations, noting the body map is from one patient or the more specific nature of RNA-seq over qPCR. Interestingly, the expression of TAS2Rs are differentially regulated with age in mice (Foster et al., 2013), but not with sex or in heart failure (Foster et al., 2015a). Furthermore, analysis of the publicly available GTEx LDACC and BioGPS Human Cell Type and Tissue Gene Expression Profiles RNA-seq datasets, highlight the expression of GNAT3 (the taste receptor specific G protein, Gα Gustducin ) in a variety of human tissues, including the heart.
We previously investigated the factors contributing to cardiac TAS2R gene expression in silico (Foster et al., 2015a). Similar to rodent Tas2rs, there was no evidence of enrichment for particular transcription factor binding sites in the proximal promoter regions of the human TAS2R genes. However, we observed thatTAS2R14 (the most abundantly expressed) had the strongest evidence of regulatory activity in its promoter region, i.e., active methylation marks overlapping with the DNase I hypersensitivity cluster. On this basis, although we cannot rule out the presence of specific transcription factors that regulate TAS2R gene expression, we reason that the proximal regulatory regions for some, but not all, TAS2R genes might show a basal level of transcriptional activity. This, combined with their multigene cluster expression profiles could facilitate preferential transcription of the specific TAS2Rs (Foster et al., 2015a).
The heart is made up of 2-3 billion cardiomyocytes and yet these cells constitute less than a third of all heart tissue (Tirziu et al., 2010). The remaining, more than two thirds of the heart consists of smooth muscle, fibroblasts, other connective tissue cells, endothelial cells, sinoatrial cells, atrioventricular cells, Purkinje cells, pluripotent cardiac stem cells, mast cells, and other immune system-related cells (Tirziu et al., 2010). We have demonstrated that certain Tas2rs (rodent) were expressed within both cardiomyocytes and fibroblasts, as well as their downstream signaling effectors (Gnat3, Plcβ2, Trpm5) (Foster et al., 2013). These data suggest that specific cells within the heart may express varying populations of TAS2R, similar to that seen in other systems ( Table 1). As technology advances, including single cell sequencing and proteomics (Uhlen et al., 2015), the topography of T2Rs within the heart will provide insight into how these receptors function within this system.

SIGNALING AND FUNCTION OF T2Rs WITHIN THE CARDIOVASCULAR SYSTEM
The binding of bitter ligands to T2Rs results in a conformation change in the receptor allowing it to interact with Gα Gustducin and Gβ 1/3 γ 13 (Huang et al., 1999), which then activate subsequent downstream pathways (Yan et al., 2001). Knockout (KO) studies have provided conclusive evidence supporting these signaling pathways. Mice lacking either PLCβ2 or TRPM5 exhibited diminished or ablated taste responses to bitter compounds (Zhang et al., 2003). Furthermore, Gα Gustducin KO mice had increased levels of cAMP, compared to the wild-type mice as well as displaying severely impaired responses to the tested compounds (Clapp et al., 2008). As with Gαi-family G proteins, Gα Gustducin can decrease the levels of cAMP, via the activation of phosphodiesterases, which has been observed in response to two bitter compounds, denatonium and strychnine (Yan et al., 2001). Finally, mice that were genetically modified to express novel human T2Rs demonstrated a strong aversive response to ligands that was not evident in wild-type mice (Mueller et al., 2005).
With the discovery of TAS2R expression in cardiac tissue (Foster et al., 2013), defining the signaling transduction pathway is of particular interest, yet there is limited evidence for the presence of all the classical taste signaling components in heart. The expression of Gα Gustducin has been shown in human heart tissue, and is particularly enriched in cardiomyocytes (BioGPS Human Cell Type and Tissue Gene Expression Profiles RNAseq datasets). However, studies have not observed Gγ 13 (Huang et al., 1999) or TRPM5 (Demir et al., 2014). TRPM4 is present within human heart tissues (Guinamard et al., 2004;Demir et al., 2014), however, both TRPM4 and TRMP5 are considered necessary for taste signal transduction (Dutta Banik et al., 2018). Hence, alternative signal transduction pathways that could mediate the effects of taste receptors in the cardiovascular system should be considered.
A study by Ueda et al. (2003) demonstrated that T2R16 could couple to a chimeric G protein consisting of the N-terminus of Gα 16 and the last 44 amino acids of either Gα Gustducin , Gα t2 or Gα i2 . Furthermore, the expression of all three of these Gα subunits have been identified in taste receptor cells, with the frequency of Gα i2 being higher than that of Gα Gustducin (Ueda et al., 2003). Figure 1 shows a comparison of Gα Gustducin and Gα i2 , highlighting the highly conserved amino acid residues and the region known to interact with TAS2Rs. Substitution of glycine 352 for proline in Gα Gustducin disrupts T2R interaction with Gα Gustducin , although its coupling to the Gβγ and effector molecules was preserved (Ruiz-Avila et al., 2001). This suggests that the extreme C terminus of both Gα Gustducin and Gα i2 are capable of, and necessary for T2R:G protein coupling and transduction. Importantly, in human airway smooth muscle (HASM), the reported expression of Gα i2 expression was 100fold higher than that of Gα Gustducin and T2R14 was shown to couple to all Gα i proteins, particularly Gα i2 . The use of pertussis toxin was able to abrogate the T2R mediated relaxation in HASM , consistent with previous studies, where T2Rs has been shown to couple with inhibitory signaling pathways (Ozeck et al., 2004). The actions of T2R may also include other inhibitory type processes, such as described by Zhang et al. (2013) in airway smooth muscle cells . Taken together these observations suggest that depending on the level of G protein expression and the strength of the subsequent signal, T2Rs likely couple and signal in a cell/tissue specific manner, which may include (or not) Gα Gustducin .
Indeed, T2R signaling within cardiac cells might reasonably reflect those described for the respiratory system and vascular systems (summarized in Figure 2). The heart is known to express specific varying combinations of Gα (including Gα i2 ), Gβγ and various signaling effector molecules (Doll et al., 2017). In a series of experiments in Langendorff-perfused mouse hearts, we observed dose-dependent negative inotropic effects in response to bitter ligands (Foster et al., 2014). A ∼40% decrease in FIGURE 1 | Comparison of 2D snake representations of Gα Gustducin and Gα i2 [constructed via GPCRdb (Isberg et al., 2016)]. Amino acids are labeled following the Ballesteros-Weinstein numbering system. Conserved amino acids are colored green. The final 44 amino acids are known to be necessary for T2R coupling and signaling (Ueda et al., 2003) and are highlighted by black boxes. Arrow points to the amino acid (Glycine 352 ) in Gα Gustducin that disrupts T2R coupling when mutated to proline (Ruiz-Avila et al., 2001). left ventricular developed pressure and an increase in aortic pressure in response to sodium thiocynate were shown to be Gα i -dependent. Some alterations in cardiovascular physiology were not attributed to G proteins (not blocked by pertussis toxin and gallein), however, it was shown that rodents express GNAT3 (Gα Gustducin ) in their cardiomyocytes (Foster et al., 2013). This further supports the premise that T2Rs can signal through various G proteins. While there is no clear consensus on the precise mechanism, there is an agreement that bitter ligands mediate contractile responses in the vasculature. One study demonstrated a transient drop in blood pressure upon intravenous injection of denatonium benzoate into rats (Lund et al., 2013). Additionally, Manson et al. (2014) attributed the endothelium-independent relaxation of precontracted human pulmonary arteries to the application of bitter ligands for T2Rs (3, 4, 10, and 14). In contrast, denatonium benzoate has been shown to enhance the tone of endothelium-denuded rat aorta rings, which was attributed to specific Tas2r activation (Tas2r40,108,126,135,137,143) via Gα Gustducin (Liu et al., 2020). Whether the actions of T2Rs in cardiomyocytes have a direct effect on the force and strength of contraction of individual myocytes remains to be determined. Equally, there is a possibility that these receptors may be expressed in other cell populations, including the specific cells of the conduction system (SA node, AV node, Purkinje Fibers).

NATURALLY OCCURRING POLYMORPHISMS AND DISEASE
GPCRs and their respective ligands have profound homeostatic and regulatory effects on the cardiovascular system. Not surprisingly, mutations and modifications of cardiovascular GPCRs, G proteins and their regulatory proteins are linked to dysfunction and disease (Foster et al., 2015b). T2Rs are one of the most heterogenous and unique families of GPCRs and are now considered as a separate group of receptors (Di Pizio and Niv, 2015). According to the HGNC database, there are 39 genetically diverse and highly polymorphic TAS2R single exon genes that encode for 29 functional T2Rs (and 10 non-coding pseudogenes) in humans (Devillier et al., 2015). This is in contrast to the majority of literature that cite the existence of only 25 functional T2Rs Lossow et al., 2016). On average, TAS2R genes contain four single nucleotide polymorphisms (SNPs) of which the vast majority are non-synonyms mutations that encode amino acid substitutions (Kim et al., 2005). Table 2 outlines all of the non-synonymous SNPs present within the population and their penetrance. These TAS2R genes are located on chromosomes 5, 7, and 12 Foster et al., 2015a), with dense clustering on chromosomes 7 and 12. The close proximity is thought to underpin the enormous variation and diversification of the T2R repertoire within humans.
The importance of uncovering the primary function of T2Rs in the heart is supported by the critical role they play within the respiratory system. TAS2R38 is expressed in all aspects of the upper and lower respiratory tracts including sinonasal epithelial cells, bronchial epithelial cells, bronchial smooth muscle, and pulmonary vasculature smooth muscle (Shah et al., 2009;Grassin-Delyle et al., 2013;Upadhyaya et al., 2014;Devillier et al., 2015). Application of phenylthiocarbamide (PTC) or two quorum sensing molecules (C4HSL and C12HSL) secreted by Pseudomonas aeruginosa were shown to increase mucociliary clearance, bronchodilation, and production of bactericidal levels nitric oxide in explanted human tissue samples and primary airway-liquid interface cultures (Lee et al., 2012). This supports the recent finding that T2Rs play a role in innate immunity, as quorum sensing molecules serve to communicate between bacterial populations, allowing them to establish themselves during infection (Lee et al., 2014). Bitter taste receptors, particularly TAS2R38, are a unique and diverse family of GPCRs due to the number of their naturally occurring genetic variants (Kim et al., 2005). Compared to the functional (PAV) haplotype, individuals with the non-functional (AVI) haplotype were shown to be more susceptible to respiratory infections as the receptor was unable to detect the compounds and respond appropriately (Lee et al., 2012). A similar result was seen with regard to oral innate immunity (Gil et al., 2015). TAS2R38 PAV/PAV mRNA was upregulated ∼4.3-fold in response to Streptococcus mutans bacteria (over the unstimulated control) whereas the AVI/AVI was only ∼1.2-fold. Furthermore, the level of hBD-2 (antimicrobial peptide) induced was highest in those with the PAV/PAV genotype (Gil et al., 2015). On this basis, the authors concluded that a person's T2R38 genotype determines oral innate immunity.
Natural polymorphisms are no longer thought only to account for differences in oral bitter taste perception . It is now recognized that these polymorphisms also influence other important aspects of our physiology including alcohol dependence, eating behavior, longevity, glucose homeostasis and regulation of thyroid hormones (Dotson, 2008;Hayes et al., 2011;Campa et al., 2012;Clark et al., 2015). There are 132 naturally occurring non-synonymous polymorphisms for cardiac-expressed T2Rs and it is clear that the majority of these remain uncharacterized (Table 3). One polymorphism that is of particular interest is T2R50-rs1376251, as debate remains in the literature over its potential association with myocardial infarction and coronary heart disease (Shiffman et al., 2008;Tepper et al., 2008;Yan et al., 2009;Koch et al., 2011;Ivanova et al., 2017;Tsygankova et al., 2017). There are also polymorphisms outside of the taste receptor coding region, or those that result in synonymous mutations that have been associated with changes in physiology. Of note, T2R14 rs3741843 has been associated with decreased sperm motility (Gentiluomo et al., 2017). Individuals that were homozygous carriers for the (G) allele, encoding arginine (R -AGG), showed a decreased sperm progressive motility compared to heterozygotes and homozygotes for the (A) allele, which encodes arginine (R -AGA). The authors rationalized using in silico analysis that T2R14 regulates the expression of T2R43. Furthermore, an upstream mutation of TAS2R3 rs11763979 can regulate the expression of WEE2 antisense RNA one (WEE2-AS1), which increases the expression of WEE2 within the testis. WEE2 is a protein tyrosine kinase involved in the regulation of cell cycle progression (Nakanishi et al., 2000). Overexpression of WEE2 in the testis was hypothesized to increase the number of abnormal sperm cells (Gentiluomo et al., 2017). Despite recent progress, it is unclear the full extent to which polymorphisms can influence T2R physiology, although it is clear investigation into their effects is warranted.

POTENTIAL CARDIOVASCULAR T2R LIGANDS
T2Rs are unique as they lack most of the conserved motifs of the class A GPCR family (Lagerstrom and Schioth, 2008). The intracellular loops -regions necessary for signal transduction and feedback modulation (Moreira, 2014), were shown to be more conserved across T2Rs than the extracellular loops that are generally implicated in receptor binding (Meyerhof, 2005). Using T2R14 as an example, Nowak et al. (2018) demonstrated that in vitro mutagenesis of 19 receptor mutants (all within the binding pocket) retained the ability to bind at least one of the 7 tested agonists while some improved signaling compared to the wild type. These results are consistent with previous literature that ligands bind within the transmembrane and extracellular domain regions Upadhyaya et al., 2015). Interestingly, of the highly expressed cardiac T2Rs, T2R10, T2R14, and T2R46 were shown to bind a wide array of ligands, which is considered disproportional in comparison to the others . Over 75% of the list of ligands in Table 3 were shown to activate these three broadly tuned T2Rs.
Universally, researchers have used chemicals that 'taste bitter' to test for potential ligands. However, if heart tissue expresses over half of the T2Rs family, a major question arises -what is the source of ligands for these T2Rs within the cardiovascular system? We would argue there are four major sources: (1) bitter compounds in food, (2) endogenously  (Shiffman et al., 2005(Shiffman et al., , 2008Tsygankova et al., 2017), coronary heart disease (Yan et al., 2009) Italicized polymorphisms represent those that have less than 300 alleles detected in the sample population (* denotes stop codon).
The post-prandial concentration of bitter compounds in the blood increases. One perhaps common example of this is caffeine, which reportedly modulates calcium signaling via interaction with the ryanodine receptor (Kong et al., 2008). Interestingly, caffeine also activates T2R10, -14, and -46 Cappelletti et al., 2018) at concentrations that occur in blood post-prandially and which are equivalent to levels that modulate the ryanodine receptor (Kong et al., 2008). In the gut, caffeine activation of T2R has been linked to gastric secretion (Liszt et al., 2017). Caffeine may also act as a stimulate for the central nervous system via the antagonism of adenosine receptors (Fisone et al., 2004). Hence in considering the homeostatic consequence of bitter compounds (such as caffeine) one must also accept that at high concentrations they are interacting with multiple receptor systems. We would anticipate that many bitter compounds in food would have actions on both T2Rs and other targets.
Another interesting possibility is that the body produces endogenous factors that could activate T2Rs. Currently, alanine, pantothenic acid (vitamin B5), steroids (androsterone and progesterone) and taurocholic acid (primary bile acid) have all been identified as ligands for specific receptors (Ji et al., 2014;Lossow et al., 2016). Potentially, the cardiotonic steroids may be ligands for cardiac-expressed T2Rs, although ouabain has already been shown not to be an agonist in vitro , despite being able to augment calcium transients in arterial smooth muscle (Arnon et al., 2000). The other members of this family could also be investigated as potential ligands for cardiac T2Rs. Whether hormones/factors produced by other tissues, or indeed paracrine factors released from cardiac cells, can bind and activate cardiac T2Rs remains to be determined, but is an area of intense interest.
A more provocative idea is that colonizing bacteria, in complex organisms, could produce bitter compounds, including metabolic by-products and other signaling molecules that alter our physiology via T2Rs. A recent study showed commensal bacteria are enable to synthesize GPCR ligands that mimic human signaling molecules (Cohen et al., 2017). Broad screening for bacterial metabolites that activate GPCRs (Chen et al., 2019;Colosimo et al., 2019) have identified numerous candidates, but unfortunately these screens have not included the taste receptors. Interestingly, an olfactory receptor (Olfr78) has been reported to respond to short chain fatty acids produced by gut bacteria (Pluznick et al., 2013). Olfr78 KO mice had elevated blood pressure when treated with antibiotics. As for the T2Rs, T2R38 although its expression is low in the heart, was shown to be broadly tuned for seven bacterial metabolites (Verbeurgt et al., 2017). It is also worth noting that during infections bacterial toxins could be 'bitter' and interact with T2Rs once they reached a certain concentration in the blood. One example is quorum sensing molecules -when they reach a certain concentration, bacteria produce a biofilm in order to evade and survive the host immune defense system (Davies et al., 1998). Therefore, it is plausible that T2Rs may alter cardiovascular physiology in response to systemic infections such as sepsis where dramatic cardiovascular changes are observed, e.g., decreased myocardial contractility, vasodilation, endothelial injury and increased heart rate (Singer et al., 2016).
Finally, it is important to address the possibility that off-target activation of T2Rs may play a role beyond normal physiology and mediate unexpected responses to therapeutic drugs, many of which are bitter. Indeed, the possibility that T2Rs act as the mediators of off-target drug effects due to the prevalence of their expression throughout the body has been discussed previously (Clark et al., 2012). There are numerous drugs/chemicals that have specific, detrimental cardiovascular effects and, moreover, these chemicals have been shown to activate specific T2Rs at concentrations to those that elicit these adverse effects.

FUTURE DIRECTIONS
The continuous, proper functioning of the heart is fundamental to life. The discovery of T2Rs expressed in cardiac cells predicts important (but yet to be appreciated) roles in heart physiology, as well as its response to external challenges (e.g., diet, metabolic changes, infections, and drugs). Research and knowledge regarding the physiology of T2Rs within the human heart is challenging, primarily due to the constraints of readily acquiring suitable human heart tissue samples. Furthermore, the lack of homology between rodent Tas2rs and human T2Rs (Foster et al., 2013), limits the utility of gene modified animal models to directly inform human physiology. Additionally, the 29 T2Rs (and their many variants) have been historically difficult to heterologously express on the cell membrane of model cells, and this has impeded further investigation of their signaling properties.
It is important to note, that researchers have ectopically expressed human T2Rs in mice and this has provided strong confirmation that a given ligand (tastants) can activate a specific human T2R (Mueller et al., 2005). Perhaps future experiments might extend this approach to develop transgenic mice expressing human T2Rs in a cell-specific context. Stimulation of these receptors with ligands that selectively bind and activate only human T2Rs could provide important insights into the physiological role(s) of T2Rs in human tissues.
Another critical objective will be to develop appropriate cardiac models that express endogenous receptors and recapitulate cardiac physiology. One major advance in cardiovascular research has been the development of induced pluripotent stem cell-derived human cardiomyocytes (Hudson et al., 2012;Soong et al., 2012) and human cardiac organoids (Nugraha et al., 2019). These models will offer the unique opportunity to modulate T2R expression in cardiomyocytes and to thereby investigate bitter ligand-driven changes in cardiac gene transcription, as well as to define alterations in cardiac contractility and function.
Finally, the ultimate goal will be to attribute T2R-mediated expression, activation and signaling to definitive changes in human cardiovascular function in vivo. In order for this to succeed, the following challenges need to be resolved -  (TC) or effective concentration (EC50) in vitro concentration], reported effects in the cardiovascular system and corresponding dose/serum (* = based on 5.5 L of blood in human body, or without First Pass Effect of liver).

∼3500-8000
Strychnine Pesticide 10, 46 0.43 ± 0.02 (EC50) Cardiac arrest, bradycardia, ECG changes (Heiser et al., 1992;Wood et al., 2002;Ponraj et al., 2017) 3.8 mg/L ∼11 ClassificationActivates T2R (Jaggupilli et al., 2016) In vitro (µM) -TC/EC50 Jaggupilli et al. (2016) Reported effects on cardiovascular system Reported doses Frontiers in Physiology | www.frontiersin.org the promiscuity of bitter receptor-ligand interactions, the elucidation of tissue-specific T2R signaling, as well as the lack of definitive research tools (e.g., selective antibodies to T2Rs and specific receptor antagonists). We anticipate that studies focused on examining the functionality (or lack thereof) for the various highly penetrant, cardiac-expressed T2R polymorphisms may provide the means for unambiguously attributing T2R activation to a specific physiological outcome. Analogous to the advances made with non-functional T2R38 variants (T2R38AVI) in the lung, we predict that non-functional, cardiac-expressed T2Rs can be identified and these will prove to be critical in providing the necessary controls for investigating explanted cardiac tissues.

AUTHOR CONTRIBUTIONS
CB, SF, and WT designed the scope and structure of the review. CB collated data. CB, SF, and WT wrote and edited the final manuscript. CB produced the figures and tables in consultation with SF and WT.