Bile Acids—A Peek Into Their History and Signaling

Abstract

Bile acids wear many hats, including those of an emulsifier to facilitate nutrient absorption, a cholesterol metabolite, and a signaling molecule in various tissues modulating itching to metabolism and cellular functions. Bile acids are synthesized in the liver but exhibit wide-ranging effects indicating their ability to mediate organ-organ crosstalk. So, how does a steroid metabolite orchestrate such diverse functions? Despite the inherent chemical similarity, the side chain decorations alter the chemistry and biology of the different bile acid species and their preferences to bind downstream receptors distinctly. Identification of new modifications in bile acids is burgeoning, and some of it is associated with the microbiota within the intestine. Here, we provide a brief overview of the history and the various receptors that mediate bile acid signaling in addition to its crosstalk with the gut microbiota.

Bile acids, the body's natural detergent, are fundamentally known for their role in lipid digestion, but over the last two decades, their function as signaling molecules has been uncovered. There has been a major resurgence of interest in investigating bile acids, with obeticholic acid (OCA), one of its synthetic derivatives, being used in clinics for the treatment of nonalcoholic steatohepatitis (NASH) and due to its interactions with the gut microbiota. Several reviews have elegantly and extensively discussed bile acid synthesis, chemistry, transport, metabolic roles, and its crosstalk with the gut microbiota (1–5). Therefore, we highlight historical and major developments focusing on the different receptors and their potency to specific bile acids, novel amino acid conjugations, and interactions with the gut microbiota.

Bile Acids—A Historical Perspective

Although the bile acid structure, function, and application have been investigated for more than 150 years, new signaling pathways and their varied roles in metabolism and overall physiology are still being uncovered. The study of bile acids began with Strecker isolating the primary bile acid, cholic acid (CA), in 1848 through combustion analysis, marking the discovery of the first bile acid (6). This discovery was followed by the identification of several distinct bile acids from different species (7). In the early 1900s, bile acids were mainly used as tonics and laxatives to increase bile flow in cholestatic patients despite the lack of well-controlled efficacy studies (2, 6). The field began to gather momentum in the 1930s when cortisone was commercialized to treat rheumatoid arthritis because deoxycholic acid (DCA) could be used to synthesize cortisone (7, 8). The development of gas chromatography and mass spectroscopy to analyze bile acid levels and composition further accelerated the field (9, 10). Of note, patients with hypercholesterolemia were fed CA in 1965. It was found that the CA diet suppressed bile acid and cholesterol synthesis (11, 12). The prospect of therapeutic benefits of bile acids and advances in analytical methods led to the large-scale synthesis of conjugated bile acids. Delineating the kinetics and intermediates of bile acid synthesis was arduous and took 4 decades to refine (13–17).

The next breakthrough came in 1972 when chenodeoxycholic acid (CDCA) was shown to promote gallstone dissolution, a considerable advancement in treating hepatobiliary diseases (12, 18). Unfortunately, the hepatotoxicity of CDCA feeding was a major concern that led to the exploration of ursodeoxycholic acid (UDCA) as an alternative for gallstone dissolution. UDCA not only displayed higher efficacy but also minimal hepatotoxicity (18–22). In fact, the beneficial effects of UDCA were subsequently harnessed to mitigate the disease progression of primary biliary cirrhosis and alleviate the symptoms of intrahepatic cholestasis of pregnancy (23–25). Clinical use for UDCA has since expanded to also treat gallstones after bariatric surgery, low phospholipid–associated cholelithiasis, cystic fibrosis–associated liver injury, and hepatic injury following stem cell transplantation (26–30). While UDCA remains a drug of choice for treating cholestatic liver disease, its susceptibility to side chain amidation and lack of effectiveness against cholangiopathies led to the exploration of 24-norursodeoxycholic acid (norUDCA) as a therapeutic alternative. The shortened side chains protect norUDCA from taurine and glycine amidation and allows for cholehepatic shunting (31–33). As norUDCA promotes a bicarbonate-rich bile flow, and bile acid detoxification (34), it is currently in phase 3 clinical trials as a therapeutic for treating primary sclerosing cholangitis (35–39). In addition to treating cholestatic liver diseases, norUDCA and UDCA may alleviate NASH/nonalcoholic fatty liver disease, but, further studies are warranted to confirm their efficacy (35, 40).

The therapeutic potential of bile acids underscored their importance in health and disease and was instrumental in pioneering their large-scale medical synthesis. A summary of the major advancements and discoveries in the field is represented in Fig. 1.

Figure 1.

Timeline of historical bile acid research and innovation.

Bile Acid Structure

In 1927, a four-ring structure with a varying side chain of carbon atoms was first proposed for bile acids (41, 42). X-ray diffraction studies on vitamin D, cholesterol, and ergosterol showed that the ring structures of these molecules were similar to that of bile acids (43). This overlap in structure between cholesterol and bile acids led researchers to speculate cholesterol is a precursor of CA, but the inability of dietary cholesterol supplements to increase bile acid levels questioned this hypothesis. Radioactive labeling of cholesterol eventually led biochemists to observe deuterium incorporation into CA (44), thus solidifying that cholesterol was the true precursor for CA.

All bile acids share a saturated tetracyclic hydrocarbon perhydrocyclopentanophenanthrene system (steroid nucleus), which consists of three 6-carbon rings and one 5-carbon ring and are typically in the cis-conformation. Distinct side chains, stereochemistry, and hydroxyl groups gave rise to the varied bile acids. All primary bile acids synthesized de novo in the liver are identifiable with a 7α-hydroxyl group. The hydroxyl groups (positions 3, 7, and/or 12 in mammals) and the side chain (positions C18 and C19) of bile acids provide unique characteristics to the different bile acid species. Thus, bile acids possess both a hydrophilic (α) and hydrophobic (β) side (45), and their hydrophobic index varies distinctly (46). Their amphipathic nature enables micelle formation and the solubilization of lipids and cholesterol in the intestinal tract. Consequently, overall bile acid pool composition and concentration of individual bile acid species affect biological outcomes.

Primary and Secondary Bile Acids

The primary bile acids synthesized in human hepatocytes are CA and CDCA, while murine hepatocytes synthesize muricholic acid (MCA), CA, and CDCA. At least 17 liver enzymes are known to be involved in this process, which is divided into 2 distinct pathways termed the “classical” and “alternative” bile acid synthesis pathways (47). The classical, or neutral pathway, is responsible for producing approximately 90% or approximately 75% of total bile acids in humans and mice, respectively. The process is initiated with the 7α-hydroxylation of cholesterol by P450 cholesterol 7α-hydroxylase (CYP7A1), which is the rate-limiting step (47–49). The ratio between CA and CDCA in the bile acid pool is determined by microsomal sterol 12α-hydroxylase (CYP8B1). At basal conditions, the liver produces almost equal amounts of CA and CDCA (47, 50). The alternative, or acidic pathway, accounts for approximately 9% or approximately 25% of the bile acid pool in humans and mice, respectively (51, 52). This begins with the C27-hydroxylation of cholesterol by mitochondrial 27-hydroxylase (CYP27A1). The intermediates are then converted by the oxysterol 7α-hydroxylase (CYP7B1) into CDCA in humans and MCA in mice (47, 50). CYP2C70 was identified to synthesize MCA, a 6-hydroxylated bile acid that is highly present in mice (53). Of note, the deletion of Cyp2c70 in mice exhibited human-like bile acid metabolism (54), underscoring the role of MCA in contributing to the species differences. Bile acids are then conjugated with either glycine or taurine, which makes them less toxic and helps with their solubility (55). In humans, the number of glycine conjugates is three times higher than taurine conjugates (56, 57) whereas in mice, approximately 95% are taurine conjugated (58). Thus, when using rodent models to understand bile acid homeostasis, species-specific differences must be carefully considered and several reviews and reports have covered this aspect (59–62).

Once bile acids are synthesized, conjugated, and secreted into the intestine, approximately 95% are reabsorbed in the terminal ileum and recirculated back to the liver. A small percentage of the bile salts (∼5%) are deconjugated by bacterial bile salt hydrolases and then absorbed or metabolized to secondary bile acids by bacterial 7α-dehydroxylase. CA and CDCA in the intestine are converted to DCA and LCA, respectively. Notably, LCA is more toxic of the two and is passively absorbed at a lower rate. The processes that detoxify LCA differ between humans and mice. Human LCA can be secreted into the bile after being sulfated and N-acylamidated in the liver, while murine LCA is eliminated through the feces after it is hydroxylated at the C-6 and/or C-7 positions. LCA can also be sulfated and eliminated through urination. In addition, LCA can be converted to hyocholic acid (HCA) and UDCA in humans, whereas it can be 7α-hydroxylated to CDCA and then converted to UDCA, α-, β-, and ω-MCA, HCA, and hyodeoxycholic acid (HDCA) in mice (50, 63, 64). The differences in secondary bile acid clearance and metabolism can contribute toward humans having a more hydrophobic bile acid profile in contrast to a more hydrophilic profile in mice (61, 65, 66).

Uncommon and Novel Bile Acids and Influence of the Gut Microbiota

The gut microbiome is responsible for deconjugating and converting primary to secondary bile acids. In fact, microbiota-mediated bile acid modifications and their effect on health have been reported (67–70). Conversely, bile acids can inhibit the growth of certain bacterial strains. Thus, gut microbial diversity and bile acid composition influence each other, and their interactions can be altered when bile acid homeostasis is disrupted or upon the onset of metabolic diseases (71–76).

Unusual 7-oxygenated bile acids (3β-hydroxy-Δ5), NPCBA1 (3β-hydroxy,7β-N-acetylglucosaminyl-5-cholenoic acid) and NPCBA2 (3β,5α,6β-trihydroxycholanoyl-glycine) have been observed in Niemann-Pick C (NPC) disease, which is a cholesterol and fat transportation disorder. The concentrations of 7-oxygenated bile acids are elevated in these patients, which may contribute to cholestasis, and at the same time, can also potentially serve as biomarkers (77, 78) 3β-hydroxy(iso)-bile acids are another prominent class of bile acids synthesized by Ruminnococcus gnavus (79). Diet can influence the gut microbiota and subsequently regulate the levels of 3β-hydroxy bile acids and possibly other bile acids (80, 81).

Although amino acid–conjugated bile acids have been identified, and the synthesis of taurine and glycine conjugated bile acids well recognized in mice and humans (82–85), their physiological relevance is yet to be fully understood. A recent global mapping study in germ-free and specific-pathogen free (SPF) mice revealed that the gut microbiota could conjugate CA with either phenylalanine (phenylalanocholic acid), tyrosine (tyrosocholic acid), or leucine (leucocholic acid). These conjugations are novel and need to be actively explored (86). Intriguingly, taurine-conjugated bile acids were found to be abundant in the germ-free mice, while secondary bile acids were elevated along with higher Shannon diversity index in the cecum and colon of the SPF mice. Thus, the gut microbiome plays an inherent role in modulating bile acid conjugation and metabolism. Interestingly, amino acid–conjugated bile acid concentrations are also increased in the small intestine and have been demonstrated to be agonists for farnesoid X receptor (FXR) in vitro. Further, cultured human gut microorganisms can produce these amino acid–conjugated bile acids and were elevated in human patients with inflammatory bowel disease or cystic fibrosis as measured by their spectral signatures (86).

Apart from diseased states, age-associated changes in bile acid metabolism and its composition have been extensively studied in rats (87), mice (88), and humans (89). Aging has been shown to affect the gut microbiota and the subsequent bile acid profile (90). In fact, interactions between bile acids and gut microbiome diversity have been observed in newborn mice, wherein taurocholic acid (TCA) and β-tauromurocholic acid (TβMCA) accelerate the maturation of bacteria (91). Sardinian centenarians displayed unique gut microbiota signatures distinct from younger and older individuals (92). Similarly, Japanese centenarians harbored gut microbiome that produced unique isoforms of LCA (iso-, 3-oxo-, 3-oxoallo-, and isoallo-), and the novel isoallo-LCA can inhibit the growth of gram-positive bacteria (93, 94). These findings delineate a link between specific bile acid isoforms and aging in such a way that some of them strongly correlate with health while others with disease.

Bile Acid–Activated Nuclear Receptors

Nuclear receptors (NRs) are a widely expressed superfamily of ligand-activated transcription factors that exhibit a broad range of physiological and pathological roles (95–98). NRs share structural features with a conserved N-terminal DNA-binding domain, and a C-terminal ligand-binding domain (50, 99). Bile acids act as an endogenous ligand for certain NRs with predominant actions through FXR (100–102). Other NRs, such as pregnane X receptor (PXR) (103), vitamin D receptor (VDR) (104), and liver X receptor (LXR) (105) have been shown to be activated by secondary bile acids. In addition, each bile acid preferentially binds and activates their target receptors. This data from rodents and humans has been compiled in Table 1. Once bile acid binds the ligand-binding domain, the activation function region facilitates interaction and recruitment of coregulators. Thus, bile acids can modulate the transcriptional control of genes. Conversely, these NRs act as sensors to modulate bile acid concentrations primarily within the hepatocytes and enterocytes and mediate the metabolic consequences of bile acid signaling. NRs can also be found in other bile acid-related tissues as seen in Fig. 2.

Table 1.

Natural bile acids as ligands and activators of nuclear receptors and G protein–coupled receptors

Nuclear receptors	Model	Method	Bile acids and concentrations	Potency	Reference
Fxr	Cell-free	Ligand-binding assay—human FXR	CDCA (4.5 μM)	LCA > TLCA > CDCA > GLCA > GCDCA∼TCDCA > DCA > GDCA∼TDCA > CA∼GCA∼CA; binding affinity EC50	(106)
			GCDCA (10 μM)
			TCDCA (10 μM)
			CA (>1000 μM)
			GCA (>1000 μM)
			TCA (>1000 μM)
			LCA (3.8 μM)
			GLCA (4.7 μM)
			TLCA (3.8 μM)
			DCA (100 μM)
			GDCA (≥500 μM)
			TDCA (≥500 μM)
	CV-1 cells	Transfected with murine or human FXR and CAT assay was performed after bile acid treatment	CDCA, CA, LCA, DCA (100 μM)	CDCA > DCA > LCA > CA	(106)
	CV-1 cells	Coexpressed human IBAT and FXR; bile acid treatment	CDCA, GCDCA, TCDCA, GCA, TCA, LCA, GLCA, TLCA, GDCA, TDCA (3 μM)	TCDCA ≈ GCDCA ≈ TLCA ≈ GLCA > TDCA ≈ GDCA > CDCA > TCA ≈ GCA; conjugated bile acids need a uptake transporter and exhibit strong affinity for FXR	(106)
	CV-1 cells	Cotransfected with rat FXR and luciferease reporter gene; bile acid treatment	CDCA, TCDCA, GCDCA, CA, TCA, GCA, DCA, TDCA, LCA (50 μM)	CDCA > LCA > DCA > CA∼GCDCA∼GCA∼TCDCA∼CA∼TDCA	(107)
	HepG2 cells	Transfected with human FXR and luciferease reporter gene; bile acid treatment	EC50: CDCA (10 μM) and DCA (10 μM)	CDCA > DCA > GDCA ≥ CA ≈ TCDCA	(107)
	CV-1 cells	Transfected with CMX-β-gal, EcRE ×6 and FXR with RXR or mutant RXR	CDCA, DCA, LCA, CA, UDCA (100 μM)	CDCA > DCA > LCA > UDCA > CA	(108)
	CV-1 cells	Transfected with CMX-β-gal, EcRE ×6 and FXR with RXR and liver bile acid transporter	CDCA, GCDCA, TCDCA, CA, GCA, TCA, LCA, GLCA, TLCA, DCA, GDCA, TDCA (100 μM)	DCA > TDCA > TCA > GDCA > GCA > CA > CDCA ≥ GCDCA > TCDCA > GLCA > LCA∼TLCA	(108)
	Cell-free	Scintillation proximity binding assay	CDCA (17 ± 3 μM)	LCA > CDCA > TCDCA > GCDCA > DCA > UDCA > CA > TCA > GCA, binding affinity IC50	(109)
			DCA (131 ± 8 μM)
			CA (586 ± 64 μM)
			UDCA (185 ± 26 μM)
			LCA (3 ± 0.5 μM)
			GCDCA (32 ± 4 μM)
			TCDCA (19 ± 0 μM)
			GCA (800 ± 0 μM)
			TCA (733 ± 0 μM)
	HepG2 cells	Transfected with pcDNA3.1-hFXR, pcDNA3.1-hRXR, pGL3-enhancer-hBSEP-promoter-Luc, pCMV-lacZ; and treated with various concentrations of bile acids	CDCA (0-75 μM), CA (0-600 μM), DCA (0-100 μM), UDCA (0-400 μM)	CDCA > DCA > CA > UDCA	(109)
	C57BL/6 mice	Diet regimen for 7 d	0.01-0.3%, w/w of CA-, CDCA-, DCA-, or LCA-supplemented diet	CA and DCA > CDCA and LCA (Diet)	(110)
Shp (orphan)	HepG2 cells	Transfected with human FXR, RXR and FXREPLTPx4-tk-luc reporter plasmid; and treated with various concentrations of bile acids and precursors	5β-cholestane-3α,7α,12α,26-tetrol (26-OH-THC) (20 μM)	DHCA∼CDCA > 26-OH-DHC > 25-OH-THC > THCA > 26-OH-THC > CA	(111)
			3α,7α,12α-trihydroxy-5 β-cholestanoic acid (THCA) (50 μM)
			CA (50 μM)
			β-cholestane-3α,7α,26-triol (26-OH-DHC) (20 μM)
			3α,7α-dihydroxy-5β-cholestanoic acid (DHCA) (30 μM)
			CDCA (50 μM)
			5β-cholestane-3α,7α,12α,25-tetrol (25-OH-THC) (20 μM)
	Human primary hepatocyte cultures	Treated with various concentrations of bile acids; RT-PCR of FXR downstream targets was used for analysis	CA, CDCA, DCA, LCA, UDCA (10, 30, 100 μM)	CDCA (100 μM) > CDCA (30 μM) > CA (100 μM) > DCA (30uM) ≈ DCA (100 μM) > CA (30 μM) > LCA (30 μM) > CDCA (10 μM) > LCA (100 μM) > LCA (10 μM) > CA (10 μM) ≈ DCA (10 μM) ≈ UDCA (10, 30, 100 μM)	(112)
Lxr	CV-1 cells	Transfected with human LXR, RXR, and LXRESREBP-1cx4-tk-luc reporter; and treated with various concentration of bile acids	26-OH-THC, 3α,7α,12α-trihydroxy-5β-cholest-24-enoic acid (Δ24-THCA), 26-OH-DHC, DHCA, CDCA (20, 50 μM)	FXR-activating bile acids and precursors do not activate LXR	(111)
	HEK293 cells	Transfected with pGL3/UREluc reporter gene, pSG5/hRXRα, CMX/hLXRα, and pSG5/hGrip1; treated cells with 5β-cholanoic acid methyl esters (CAME)	Hyodeoxycholic acid (HDCA) and tauro-hyodeoxycholic acid (THDCA)	ED50: HDCA∼17 μM; THDCA∼3 μM	(105)
Vdr	HEK293 cells	Transfected with GAL4-SRC-1, VP16-VDR, and GAL4-responsive luciferase reporter; activation assay done in the presence or absence of IBAT	CDCA, LCA, TLCA, GLCA, 6-keto-LCA, 3-keto-LCA (30 μM)	LCA > 3-keto-LCA > GLCA > 6-keto-LCA > CDCA > TLCA; EC50: LCA (8 μM) and 3-keto-LCA (3 μM)	(104)
	C57BL/6 mice	VDR(+/+) and VDR(−/−) mice treated with LCA	0.3 or 0.8 mmol/kg	LCA induced Cyp24a1 as effective as 1α, 25-dihydroxyvitamin D3 [1,25(OH)2D3] in small intestine	(113)
Tgr5	HEK293 cells	Transfected with hTGR5 and treated with various bile acids to induce cAMP production	LCA, DCA, CDCA, CA, UDCA	EC50: LCA (35 nM) > DCA (575 nM) > CDCA (4004 nM) > CA (>10 μM); UDCA induced cAMP production	(114)
	NCI-H716 cells	Transfected with hTGR5 and treated with various bile acids to induce cAMP production	LCA, DCA, CDCA, CA, UDCA	LCA > DCA > CDCA > CA	(114)
	CHO cells	Transfected with TGR5-GFP and treated with various bile acids to induce cAMP production	TLCA, LCA, DCA, CDCA, CA	EC50: TLCA (0.33 μM) > LCA (0.53 μM) > DCA (1.01 μM) > CDCA (4.43 μM) > CA (7.72 μM)	(115)
S1pr	HEK293 cells	GFP-S1PR1/2 wereexpressed and then treated with bile acid to assess ERK1/2 activation	TCA (5 and 50 μM)	TCA (5 μM) ≈ TCA (50 μM); only S1PR2 was active	(116)
	Primary rat hepatocytes	Treated with various conjugated bile acids against a S1PR antagonist; Western blot analysis. The read out used was ERK1/2 and AKT expression.	TCA, TDCA, TUDCA, GCA, GDCA (50 μM)	GDCA > GCA∼TUDCA∼TDCA > TCA	(116)
Mrgprx4	HEK293 cells	Murine ortholog and other related human MRGPRs were transfected individually in cells stably expressing Gα15, and then treated with various concentrations of bile acids	DCA, CDCA, CA, UDCA, TDCA, TCDCA, TCA	EC50: UDCA (4.93 μM) > DCA (6.19 μM) > TDCA (18.9 μM) > TCDCA (30.1 μM) > CDCA (46.2 μM) >TCA (178.8 μM) > CA (430.5 μM)	(117)
	129svj/C57BL6J mice	Mice expressing humanized MRGPRX4 were injected with various bile acids to assess scratching bouts/pruritus	50 μL injections: DCA (1 mM), TDCA (1 mM), UDCA (2 mM), CDCA (2 mM)	DCA > TDCA∼UDCA∼CDCA	(117)
	HEK293T cells	Stable cell lines expressing MRGPRX4 were treated with various concentrations of bile acids and assessed using the TGFα shedding and FLIPR assay	DCA, CDCA, CA, TDCA, GDCA, TLCA, LCA, TCDCA, GCDCA, TCA, GCA	TGFα shedding EC50: DCA (2.7 μM) > TLCA (6.1 μM) > CDCA (15.0 μM) ≈ CA 15.2 μM) > TDCA (28.0 μM) ≈ GDCA (28.3 μM) > LCA (88.0 μM) >TCDCA (>100 μM) ≈ GDCA (100 μM)	(97)
				FLIPR EC50: DCA (2.6 μM) > CA (6.9 μM) > LCA (9.3 μM) > TDCA (14.1 μM) ≈ GDCA (14.4 μM) > CDCA (15.9 μM) > TLCA (>50 μM) ≈ GCDCA (>50 μM) ≈ TCA (>50 μM) ≈ GCA (>50 μM) > TCDCA (>100 μM)
	Rat dorsal root ganglion	Cultured rat DRG electrically transfected with MRGPRX4; treated with bile acids to assess Ca2+ signal	DCA and CA (10 μM)	DCA > CA	(97)
	HEK293T cells	Transfected with MRGPRX4 and treated with mixture of serum bile acids from human patients with liver disease to assess Ca2+ signal	GCA, TCA, GCDCA, TCDCA abundance highest among 12 bile acids measured in diseased patients	Ca2+ signal increased from diseased serum mixture; DCA found to not contribute to itchiness	(97)
Pxr	CV-1 cells	Transfected with expression plasmids for mouse PXR and reporter plasmid (Cyp3a23)2-tk-CAT followed by treatment with bile acids	CDCA, CA, DCA, LCA, 3-keto-LCA (100 μM)	3-keto-LCA > LCA > DCA > CDCA∼CA	(118)
	CV-1 cells	Transfected with expression plasmids for human PXR and reporter plasmid (Cyp3a23)2-tk-CAT and then treated with bile acids	CDCA, CA, DCA, LCA, 3-keto-LCA (100 μM)	3-keto-LCA > LCA > DCA > CDCA∼CA	(118)
	129/Sv mice	PXR-/- micetreated with i.p. (LCA or corn oil) 2×/d for 4 d	LCA (0.125 mg/g)	LCA induced hepatic Cyp3a11 and Oatp2 in WT, not Tg mice	(118)
	CV-1 cells	Transfected with mouse PXR-response reporter tk-3A4-Luc and expression vector for PXR	CDCA, DCA, LCA (100 μM)	DCA > LCA > CDCA	(103)
	CV-1 cells	Transfected with human PXR-response reporter tk-3A4-Luc and expression vector for PXR	CDCA, DCA, LCA (100 μM)	DCA > LCA > CDCA	(103)
	129/Sv mice	PXR+/+ and PXR-/- and transgenic mice expressing SXR controlled by the albumin promoter (Alb-VPSXR), were treated by gavage 2×/d for 4 d	LCA (8 mg/d)	LCA showed histological changes in twice as many mice compared to WT. Alb-VPSXR showed sustained SXR can protect against LCA-induced hepatotoxicity	(103)
	FABP-VP-hPXR Tg mice	Transgenic mice expressing human PXR under the control of FABP promoter; mice treated with LCA enema 2×/d for 5 wk	LCA (5 mM)	LCA activated PXR in transgenic mice	(119)

Abbreviations: Akt, protein kinase B; cAMP, 3′,5′-cyclic adenosine 5′-monophosphate; CA, cholic acid; CDCA, chenodeoxycholic acid; DCA, deoxycholic acid; DHCA, dihydroxycholestanoic acid; EC₅₀, half maximal effective concentration; FABP, fatty acid binding protein; FXR, farnesoid X receptor; GCA, glycocholic acid; GCDCA, glycochenocholicdeoxycholic acid; GDCA, glycodeoxycholic acid; GLCA, glycolithocholic acid; HDCA, hyodeoxycholic acid; i.p., intraperitoneally; LCA, lithocholic acid; LXR, liver X receptor; MRGPRX4, mas-related G-protein coupled receptor X4; PXR, pregnane X receptor; RXR, retinoid X receptor; SHP, small heterodimer partner; S1PR, sphingosine-1-phosphate receptor; SXR, steroid and xenobiotic receptor; TGR5, Takeda G-protein coupled receptor 5; TCA, taurocholic acid; TCDCA, taurochenodeoxycholic acid; TDCA, taurodeoxycholic acid; THCA, trihydroxycholestanoic acid; THDCA, taurohyodeoxycholic acid; TLCA, taurolithocholic acid; TGFα, transforming growth factor α; UDCA, ursodeoxycholic acid; VDR, vitamin D receptor; WT, wild-type.

Figure 2.

Messenger RNA (mRNA) expression of human and murine nuclear receptors (NRs) and G protein–coupled receptors (GPCRs) in tissues involved in bile acid metabolism. Relative mRNA expression of NRs and GPCRs are denoted by their font size to signify how abundant they are in those tissues. The compiled mRNA expression data were obtained from the papers cited next to each nuclear receptor in paranthesis. Human MRGPXR4 and similar murine orthologs Mrgpra1 and Mrgprb2 genes are found in trigeminal ganglia and dorsal root ganglia (97,117,120,121). Human FXRα (95, 122, 123), SHP (124, 125), LXRα/β (95,126–128), VDR (95, 129, 130), TGR5 (115, 131), S1PR2 (132, 133), and PXR (95,134–143). Murine Fxrα (96), Shp (96, 144), Lxrα/β (145, 146), Vdr (96, 129), Tgr5 (147), S1pr2 (148, 149), and Pxr (96, 135, 150). *Asterisk denotes contradictory data for mRNA expression in that tissue. Created with BioRender.com.

Farnesoid X Receptor

FXR was discovered in 1995 as a retinoid X receptor (RXR)-interacting partner primarily expressed in the liver, intestine, kidney, and adrenal gland (100). It has since been well recognized as the endogenous bile acid receptor. Of note, in vitro studies have demonstrated that CDCA, DCA, and LCA can activate FXR with CDCA being the most potent (107, 108). Intriguingly, it has been reported that mice fed CA and DCA exhibited stronger FXR activation than CDCA and LCA (110). On the other hand, FXR orchestrates bile acid homeostasis as evident from Fxr knockout (FxrKO) mice, which exhibited altered gene expression of Cyp7a1, Cyp8b1, and Cyp27a1, ileal bile acid binding protein (Ibabp), small heterodimer partner (Shp), peroxisome proliferator-activated receptor-α (Pparα), and Lxr. In addition, FxrKO mice accumulated more serum bile acid on a CA-fed diet, highlighting its importance in bile acid regulation (151).

Over the last 2 decades, the crucial role of FXR in regulating bile acid synthesis and transportation has been elucidated. Notably, FXR displays tissue-specific functions (152). FXR orchestrates the negative feedback for suppressing bile acid synthesis by inducing another NR, SHP, which inhibits the Cyp7a1 gene expression in the liver (153, 154). FXR/SHP also coordinates the repression of ileal apical sodium dependent bile acid transporter (ASBT) expression to prevent bile acid reabsorption (155, 156). Furthermore, FXR induces the intestinal transcription of fibroblast growth factor 15 (FGF15) to regulate bile acid levels and other metabolic processes. The ileal FXR-FGF15 axis is one of the mediators of the gut-liver crosstalk such that FGF15 circulates to the liver, where it acts via β-Klotho and fibroblast growth factor receptor 4 (FGFR4) and along with hepatic SHP (157, 158) suppresses bile acid synthesis. Moreover, FGF15 also stimulates gallbladder filling such that Fgf15KO mice were unable to fill their gallbladder even after fasting (159, 160).

In addition to the liver and the intestine, the kidney enables the filtering of bile acids by either reabsorbing it in the renal proximal tubule by an Na⁺-dependent mechanism or excreting it in urine (161, 162). FXR is highly expressed primarily in the tubular cells of the kidneys, and its activation allows for metabolic waste clearance through organic cations transporter (163), maintenance of glutathione homeostasis during obesity and protection against ischemia-reperfusion damage (164). These data suggest the importance of FXR in regulating kidney function.

Fxr mutations or deficiencies affect humans and mice differently. Although constitutive and conditional knockout of Fxr is tolerated and nonlethal in mice, it can alter many biological processes, such as inflammatory response, tumorigenesis (165, 166), intestinal barrier function (167), gut microbiome composition, lipid (71), and bile acid metabolism (152). Interestingly, the loss of FXR in obese mice models attenuated weight gain and improved glucose homeostasis (168), indicating that murine FXR deficiency may be beneficial in these conditions. However, in humans, mutant Fxr variants are linked to progressive familial intrahepatic cholestasis, which can rapidly progress to end-stage liver disease in infants and lead to neonatal death (169–171).

Nevertheless, the metabolic roles of FXR makes it an attractive therapeutic target. There have been several clinical trials with synthetic and semi-synthetic FXR agonists to evaluate its efficacy for type 2 diabetes, primary biliary cholangitis, and NASH (172, 173). The drawback of these agonists is the onset of pruritus (itchy skin), which occurs in a dose-dependent manner. This side effect is possibly due to the activation of Takeda G protein-coupled receptor (GPCR) 5 (TGR5) (174), which has been linked to pruritus and increased gallbladder volume (175–177). Other potential side effects of FXR agonists include the worsening of high-density lipoprotein cholesterol and low-density lipoprotein cholesterol ratio by OCA (178) and increase of hepatotoxicity by MET-409 (179).

The majority of therapeutic studies focus on FXR agonists and not its antagonists (173). But FXR antagonism has been reported to be beneficial in cholestasis (180), hypercholesterolemia (181), type 2 diabetes, cancer, and so on (173, 182). Intriguingly, intestinal FXR agonism has also been shown to be therapeutically beneficial (183, 184). For example, tempol treatment increased TβMCA, a natural antagonist of FXR, resulting in altered gut microbiome composition, reduced weight gain, and attenuated obesity in response to high-fat diet (185). While fexaramine, an intestine-specific FXR agonist, increased adipose tissue browning and reduced diet-induced weight gain, improved hepatic glucose metabolism, and decreased systemic inflammation (186). Taken together, these studies show modulating FXR can be beneficial to combat metabolic diseases.

Small heterodimer partner

Classically, SHP is considered a downstream signaling molecule of FXR. SHP is involved in regulating bile acid levels so that its constitutive or liver- or intestine-specific deletion results in altered bile acid homeostasis (187–190). There is no evidence that bile acids act as ligands for SHP, but they indirectly regulate SHP function (100). SHP is an orphan NR that lacks the conserved DNA-binding domain but it exerts its function through protein-protein interaction by forming heterodimers with other proteins to regulate biological processes (191). Although Shp transcript levels are reduced in FxrKO mice, it is not absent. Of note, Fxr and Shp DoubleKO (DKO) mice exhibited early-onset juvenile cholestasis, liver injury, and dysregulated bile acid homeostasis in contrast to the individual Fxr or ShpKOs (192), suggesting coordinate and FXR-independent roles for SHP. Several proteins can bind to the promoter region of the Shp gene to regulate its expression (188). Although FXR is known to stimulate Fgf15 transcript levels (157), we observed that loss of intestinal Shp expression was sufficient to inhibit the acute CA-mediated increase in ileal Fgf15 messenger RNA levels (188). Interestingly, 3-D imaging of CA-fed intestine-specific ShpKO (IShpKO) mice displayed lower mucin 2 (MUC2)-positive cells, which suggests there is a reduced number of goblet cells in the ileum (188). Further, IShpKO mice, despite maintaining their suppression in ASBT levels, accumulated ileal bile acids when challenged with a CA diet (188). Intriguingly, organic solute transporter β (OSTβ) protein levels were elevated in the IShpKO ileum, even under basal conditions, highlighting the importance of SHP in regulating ileal bile acid transport.

Moreover, transcriptomic analysis in global ShpKO mice revealed that SHP and FGF15 are essential for the repression of NPC1-like intracellular cholesterol transporter 1 (NPC1L1) (188). This suggests that intestinal SHP and FGF15 may coordinate to lower cholesterol transport and possibly regulate intestinal lipid and bile acid transporters independent of FXR signaling.

Liver X Receptor

LXR signaling regulates cholesterol homeostasis with oxysterols being its endogenous ligand. The two different isoforms of LXR exhibit differential expression profiles with LXRα expressed in the intestine and the liver while LXRβ is ubiquitously expressed (100). LXR activation is known to regulate reverse cholesterol transport pathways, apoptosis, and macrophage activity (193–196) in addition to regulating the Cyp7a1 gene, specifically in rodents (197). Further, LXRα activation is also associated with increasing sulphation of bile acids to promote their elimination, and hence may contribute toward modulating bile acid homeostasis (198). Additionally, LXRα can also be activated in vitro by a secondary bile acid, HDCA (105, 199–201). Although in vivo mouse and human data are lacking, acute gavage of HDCA in zebrafish larvae showed a decrease in transportation of absorbed lipids (202). Mice challenged with an HDCA diet had a significantly lower Srebp1c, Acc, Scd1, and Fasn mRNA expression, all of which are direct targets of LXR (203, 204). However, mice treated with an HDCA derivative showed no change in hepatic lipid levels nor any change in Lxrα transcript (205). These results reveal the need for more in vivo studies to concretely prove HDCA as an activator of LXRα.

Vitamin D Receptor

Although 1α,25-dihydroxyvitamin D₃ (25D) is the bona fide ligand of VDR, the secondary bile acid, LCA can also activate it (104). Depending on the ligand, VDR is linked to either calcium homeostasis with 25D or with bile acids homeostasis and metabolite clearance with LCA (206). Additionally, VDR activation both by 25D and LCA has been shown to regulate CYP3A expression alongside SULTs and ATP-binding cassette transporters in humans, rats, and mice (104, 06–13). In obesity, type 2 diabetes, and bariatric surgery, there has been increased focus on VDR and LCA because of their ability to modulate cholic acid-7-sulfate synthesis, which is linked to beneficial metabolic outcomes (214). Of note, VDR activation has also been shown to protect against liver fibrosis (215). Due to its high expression in the small and large intestine, the role of VDR in modulating gut microbiota and bile acid modification is a growing area of research (209). Along this line, recent studies revealed that VDR expression is increased in the duodenum of patients with irritable bowel syndrome. Notably, gut dysbiosis and overgrowth of intestinal Clostridia cluster XIV bacteria, responsible for LCA production from CA and CDCA, can be associated with irritable bowel syndrome onset (216).

Pregnane X Receptor

Like VDR, PXR can be activated by the secondary bile acid LCA (118). The hydrophobic ligand pocket of PXR makes it a versatile NR that can bind a myriad of ligands (217). On ligand binding and activation, PXR heterodimerizes with RXRα and transcriptionally regulates phase I (Cyp1A, Cyp2b, Cyp2c, and Cyp3a families) II (Sult family, Ugt1a1), and III (Mdr1, Mrp2, and Oatp2) genes (118, 218, 219). These metabolic pathways are also useful in modulating bile acid levels. For example, PXR ligands (medicinal herb [Schisandra sphenanthera (WZ)], rifampicin, pregnenolone-16a-carbonitrile [PCN]) induce the transcription of CYP3A, which promotes hydroxylation and subsequent detoxification of bilirubin and LCA (66, 103, 19–21). PXR exhibits differential mechanisms and preferences of ligands between humans and mice (222) Human PXR is activated by rifampicin, whereas murine PXR is activated by PCN (222–226). Recent advances revealed that defects in regeneration and bile duct degeneration seen on deletion of YAP/TAZ in biliary epithelial cells (Yap/Taz^BEC-KO) could be mimicked by PXR activation. In fact, pharmacological activation of PXR in wild-type (WT) mice recapitulated the regenerative defects observed in Yap/Taz^BEC-KO that was attributed to PXR activation secondary to bile acid excess, (227). These studies highlight the role for PXR signaling in maintaining bile acid homeostasis (227, 228).

Bile Acid–Activated G Protein–Coupled Receptors

Bile acids not only autoregulate their concentrations but also control the homeostasis of glucose and lipids and thus overall energy metabolism (50). One way of coordinating such diverse regulation is by signaling through GPCRs. GPCRs are 7-transmembrane–containing receptors that convert to a transient high-affinity state on ligand binding. Classically, the activated G protein receptor exchanges GDP for GTP, which then allows for the dissociation of the Gα subunits and Gβγ dimers, allowing for downstream activation of effectors, such as the release of Gα_s to increase 3′,5′-cyclic adenosine 5′-monophosphate (cAMP) levels (229). TGR5 (230, 231), sphingosine-1-phosphate 2 (S1PR2) (232, 233), and recently mas-related GPCR X4 (MRGPRX4) (97, 117) are GPCRs that are activated by bile acids. Similar to NRs, bile acids have different binding affinity and activation strength depending on which bile acid can interact with these GPCRs (see Table 1). TGR5 and S1PR2 can be found in many bile acid–related tissues, while human MRGPRX4 and its mouse orthologs MRGPRA1 and MRGPRB2 are mainly found in sensory neurons and specialized immune cells (234) (Fig. 2).

Takeda G Protein–Coupled Receptor 5

The G protein–coupled bile acid receptor (GPBAR1) or TGR5 was discovered in 2002 to be activated by both conjugated and unconjugated primary bile acids such as TCA, UDCA, and TLCA albeit varying in efficacy (115, 5–7). Upon bile acid binding and activation, TGR5 can modulate many cellular functions by increasing cAMP concentrations to modulate glucose homeostasis through glucagon-like peptide 1 secretion (238) and promote energy metabolism (239). Typically, TGR5 couples with stimulatory G α protein (Gα_s) but in cholangiocytes, TGR5 selectively couples to either G α protein (Gα_i) or Gα_s based on membrane location to modulate cell proliferation to varying degrees based on the bound bile acid (114, 115, 210).

The Tgr5KO mice are relatively normal and healthy under basal conditions (240) but they do exhibit reduced expression of Cyp7b1 and alteration in the composition of the bile acid pool (241–243). Further, Tgr5KO mice are more susceptible to cholestatic liver injury such as chronic bile duct ligation (CBDL) and display reduced regenerative ability in response to partial hepatectomy (244–246). However, when challenged with a CA-enriched diet, Tgr5KO mice displayed an increased mitogenic response both in hepatocytes and other nonparenchymal cells (246). Consistent with its proliferative role, TGR5 is highly overexpressed in cholangiocarcinoma (CCA) and numerous CCA cell lines (244, 247). Studies on TGR5's role in bile acid homeostasis, cell proliferation, and overall health in a tissue-specific manner needs further evaluation.

Sphingosine-1-Phosphate Receptor 2

S1PR2 is another GPCR that interacts with either Gα_i or Gα_s in a cell type–specific manner. Unlike TGR5, S1PR2 is activated primarily by conjugated bile acids and is mainly expressed in hepatocytes. Nevertheless, S1PR2 and TGR5 are both highly expressed in intestinal epithelial cells (248, 249). There are overlapping and distinct signaling pathways that are altered by these two GPCRs. For instance, their activation by conjugated bile acids can result in the regulation of epidermal growth factor, nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB), insulin receptor (IR), extracellular signal-regulated kinase 1/2 (ERK 1/2), protein kinase B (AKT), and sphingosine-1-phosphate (S1P) (231, 233, 244, 0–3). Similar to TGR5, in CCA cell lines, S1PR2 overexpression led to increased proliferation, migration, and invasiveness (250).

The conflicting nature of S1pr2KO and Tgr5KO phenotypes suggests that despite GPCRs’ ability to be activated by various bile acids, the function of these receptors is quite different. For example, unlike Tgr5KO mice being susceptible, S1pr2KO mice exhibit protection from cholestatic liver injury indicated by lower serum and hepatic bile acid accumulation on CBDL as opposed to WT mice (194). Conversely, S1pr2KO mice were more susceptible to lipid and cholesterol accumulation in the liver when challenged with a high-fat diet (191). Further studies on S1PR2 will be essential in delineating its role in hepatic injury recovery and proliferation.

Mas-Related G Protein–Coupled Receptor X4

Pruritus, or itch during cholestasis, is common and is attributed to bile acids and bilirubin (254). Until recently, TGR5 was implicated in mediating acute cholestatic itch (175) and further studies showed that this receptor is deactivated in chronic cholestasis (255). These results indicated that other neuronally expressed GPCRs may also contribute to cholestatic itch. Accordingly, in 2019, MRGPRX4 was discovered as a novel receptor both for bile acid and bilirubin independently by two laboratories (97, 117) Meixiong et al (117) reported that MRGPRX4 is present in human sensory neurons and in dorsal root ganglion, is activated by pathological levels of bile acids, and demonstrated that humanized MRGPRX4 mice scratch more when injected with various bile acids (DCA, TDCA, UDCA, CDCA). Yu and colleagues (97) showed that bile acids can elicit an increase in intracellular Ca²⁺ and action potential in neuronal cells and determined DCA to be the most potent bile acid to activate MRGPRX4.

Additionally, bilirubin was also identified to promote pruritis via mouse MRGPRA1 and human MRGPRX4 (120). Notably, nateglinide, an MRGPRX4 agonist, was sufficient to induce itchiness in humans confirming this novel receptor can contribute to cholestatic pruritis (97). Conversely, the hydrophilic bile acid UDCA is typically used to alleviate cholestatic symptoms in patients, but does not always reduce itchiness (256). The underlying mechansim may be due to the alteration in composition and concentration of bile acids present in the body (97, 254, 257, 258). Although UDCA has been prescribed to alleviate pruritus in cholestatic pregnancy, there are contradictory data that question its benefit and hence UDCA usage must be reexamined (259).

Diet and Gut Microbiota and Their Influence in Bile Acid Composition

Diet directly influences the gut microbiota and bile acid pool. Microbiota can catabolize nutrients for energy and thrive or adjust depending on the diet, while the content of the diet (such as high- fat vs high- carbohydrates) may alter bile acid levels and gut microbiome profile (260–264). In addition, gut bacteria can modify and generate secondary bile acids that can circulate in the bloodstream to regulate glucose and lipid metabolism (265).

Studies in rats, guinea pigs, and dogs showed that more dietary protein and fat increased total bile acid excretion; in contrast, carbohydrates lowered total fecal bile acid excretion. Addition of soluble fiber will further enhanced bile acid excretion (261). Consistent with these studies, vegans displayed a lower amount of total fecal bile acid than people who eat meat, suggesting that a higher intake of animal protein and fats may lead to higher total fecal bile acid excretion (260). Thus, content of the diet plays a crucial role in regulating bile acid levels.

Diet-induced obesity is a causal factor that predisposes one to various metabolic diseases. Interestingly, metabolically unhealthy patients with a high body mass index exhibited a significantly increased percentage of 12-OH bile acids (DCA, GDCA, TLCA, TDCA, TCA) in contrast to non–12-OH bile acids (HCA, HDCA, GHDCA, UDCA, CDCA, GUDCA). Further, it was demonstrated that lower non–12-OH bile acid percentages correlated with obesity and were associated with a low abundance of Clostridium scindens. Reintroducing C.scindens increased non–12-OH bile acid concentrations in these mouse livers, demonstrating that dysregulation in bile acid levels and gut microbiome profile is linked to obesity (266). The Western-style high-fat high-sugar diet (HFHSD) can also lead to gut dysbiosis and metabolic diseases. Intriguingly, Cyp2a12/Cyp2c70 double KO (CDKO) mice, which have human-like bile acid composition, do not exhibit any significant difference in gut microbiome at the basal level. Intriguingly, HFHSD alleviated the liver injury usch as bile duct proliferation and lymphocyte infilteration seen in CDKO mice. HFHSD-induced increase in BA pool and fecal BA excretion in the WT but not in the CDKO animals; however, both genotypes showed similar gut dysbiosis. These results suggest that the diet rather than the bile acid pool change leads to gut dysbiosis (263).

Concluding Perspectives

Bile acids have piqued generations of researchers over the last 2 centuries, be it their structure, synthesis, breakdown, transport, or signaling. Despite the enormous understanding of bile acid biology, many lingering questions and gaps remain to be addressed. Concentrations of individual bile acids within the bile pool are essential; still, it is cumbersome to tease apart their hierarchy since a change in one of their concentrations alters the others. Nevertheless, enzyme kinetics and monitoring the bile acid pool over time can provide some insights. How does the bile acid pool change translate to the downstream receptor signaling via NRs and GPCRs? For instance, does excess CDCA and LCA result in simultaneous activation of FXR, VDR, and TGR5? How does LCA prioritize between the receptors since it is a potent agonist both for TGR5 and VDR? Is it tissue milieu, cell-specific, or simply the vicinity and availability of the receptors? Different bile acids display varied receptor binding affinities. But if the concentration of bile acids is not limiting, is the physiological outcome a combination of mechanisms orchestrated by multiple receptors, or is there a hierarchy? How does this change in diseased states, for instance, in cholestasis or in fatty liver disease? Tissue-specific receptor KOs will help delineate if other receptors compensate for the loss and how the other receptors respond when challenged with different bile acids. This will be a labor-intensive endeavor as different bile acids must be tested. Bile acid pool composition must be constantly monitored since the loss of these receptors alters the pool, thus making it a catch-22 situation. But finding such redundancy is pertinent for understanding bile acid signaling. For example, FXR is typically investigated as the primary target of bile acids, but its absence can be compensated for by TGR5 and vice versa (267), indicating crosstalk between the NRs and GPCRs. Therefore, careful consideration of the bile acid dose and potential receptor crosstalk needs to be included as a variable in the experimental designs and interpretation of the data. Finally, with all the new bile acids being identified, teasing out gut microbiota and bile acid interactions may quickly lead to a chicken-and-egg scenario. Although germ-free mice and SPF have been helpful, they inherently have an altered bile acid pool. Despite centuries of work, our understanding of the diverse roles of bile acids feels like the tip of the iceberg. However, one thing for sure is that these cholesterol metabolites have graduated from natural detergents to signaling molecules.

“All we know is still infinitely less than all that remains unknown.”—William Harvey

Financial Support

This work was supported by the National Institute of Diabetes and Digestive and Kidney Diseases (grant No. R01 DK113080 to S.A.), the American Cancer Society (No. ACS132640-RSG132315 to S.A.), and the Cancer Center at Illinois (CCIL planning grant 6741 and minority supplement for R.P.H.S.).

Abbreviations

25D

1α,25-dihydroxyvitamin D₃

ASBT

apical sodium dependent bile acid transporter

cAMP

3′,5′-cyclic adenosine 5′-monophosphate

CA

cholic acid

CCA

cholangiocarcinoma

CDCA

chenodeoxycholic acid

CYP7A1

P450 cholesterol 7α-hydroxylase

DCA

deoxycholic acid

FABP

fatty acid binding protein

FGF15

fibroblast growth factor 15

FXR

farnesoid X receptor

GPCR

G-protein coupled receptor

HCA

hyocholic acid

HDCA

hyodeoxycholic acid

HFHSD

high-fat high-sugar diet

KO

knockout

LCA

lithocholic acid

LXR

liver X receptor

MCA

muricholic acid

MRGPRX4

Mas-related G protein–coupled receptor X4

NASH

nonalcoholic steatohepatitis

norUDCA

24-norursodeoxycholic acid

NPC

Niemann-Pick C

NR

nuclear receptor

OCA

obeticholic acid

PXR

pregnane X receptor

RXR

retinoid X receptor

S1PR2

sphingosine-1-phosphate receptor 2

SHP

small heterodimer partner

SPF

specific-pathogen free

TCA

taurocholic acid

TGR5

Takeda G protein–coupled receptor 5

UDCA

ursodeoxycholic acid

VDR

vitamin D receptor

WT

wild-type

Disclosure

The authors have nothing to disclose.

Data Availability

Data sharing is not applicable to this article as no data sets were generated or analyzed during the present study.