Mouse models of gallstone disease

Abstract

Purpose of review

The establishment of mouse models of gallstones, and the contribution of mouse models to genetic studies of gallstone disease, as well as the latest advances in the pathophysiology of gallstones from mouse experiments are summarized.

Recent findings

The combined uses of genomic strategies and phenotypic studies in mice have successfully led to the identification of many Lith genes, which pave the way for the discovery of human LITH genes. The physical–chemical, genetic, and molecular biological studies of gallstone disease in mice with knockout or transgene of specific target genes have provided many novel insights into the complex pathophysiological mechanisms of this very common hepatobiliary disease worldwide, showing that interactions of five primary defects play a critical role in the pathogenesis of cholesterol gallstones. Based on mouse studies, a new concept has been proposed that hepatic hypersecretion of biliary cholesterol is induced by multiple Lith genes, with insulin resistance as part of the metabolic syndrome interacting with cholelithogenic environmental factors to cause the phenotype.

Summary

The mouse model of gallstones is crucial for elucidating the physical–chemical and genetic mechanisms of cholesterol crystallization and gallstone formation, which greatly increase our understanding of the pathogenesis of this disease in humans.

INTRODUCTION

Cholesterol gallstone disease is one of the most prevalent digestive diseases and affects ~12% of American adults, leading to a considerable financial and social burden in the USA [1^▪▪,2^▪▪,3–5]. To reduce the morbidity, mortality, and costs of healthcare associated with this disease, it is important to understand the pathophysiology of gallstones and this will facilitate the development of a novel, effective, and noninvasive therapy for patients with gallstones. From 1950s to 1970s, to establish animal models of cholesterol gallstones for studying the pathogenesis of this disease, many animal species, including hamsters, mice, guinea pigs, ground squirrels, prairie dogs, rabbits, and monkeys, have been extensively investigated by feeding different types of lithogenic diets from weeks to months. Although cholesterol gallstones were successfully induced in mice after feeding a lithogenic diet in 1964 [6], mouse gallstones were largely studied until the early 1990s because the first gallstone gene, Lith1 was identified in the inbred C57L/J strain by genetic methods [7]. In this review, we summarize the establishment of mouse models of gallstones, and the contribution of mouse models to genetic studies of gallstone disease, as well as the latest advances in the pathophysiology of gallstones from mouse experiments.

ESTABLISHMENT OF MOUSE MODELS OF GALLSTONES

Tepperman et al. [6] were the first to find that the combination of cholesterol and cholic acid in the diet is required for the formation of cholesterol gallstones in mice. Under lithogenic diet feeding conditions, the extracted sterols from mouse gallstones are predominantly cholesterol, which constitutes ~94% of the pooled stone weight [6]. Because mouse bile contains large amounts of hydrophilic bile acid, tauro-β-muricholic acid [8–10], feeding 1% cholesterol alone for 2–8 months cannot induce gallstones [6,8,10]. Thus, cholic acid is essential in the mouse gallstone model because after hepatic conjugation with taurine, taurocholate replaces most tauro-β-muricholate in mouse bile [8], enhancing cholelithogenesis. In 1978, Fujihira et al. [11] observed that at 8 weeks on the lithogenic diet, the prevalence of gallstones varies from 0 to 100% in seven strains of inbred mice. Alexander and Portman [12] further found that even feeding the lithogenic diet for a longer time, from 8 to 12 weeks, strain differences in gallstone prevalence still exist in mice. These studies paved the way for the genetic analysis of gallstone disease in mice. More importantly, Wang et al. [13] found that the evolutionary sequences of cholesterol crystallization and gallstone formation in mice during the 8-week period of feeding the lithogenic diet are characterized by the initial accumulation of mucin gel, followed by appearances of liquid crystals and/or anhydrous cholesterol crystals and cholesterol monohydrate crystals, and then agglomerated cholesterol crystals, sandy stones, and true gallstones (Fig. 1). These sequences are in agreement with the results from other animal models of cholesterol gallstone formation and in humans.

FIGURE 1.

Representative photomicrographs of cholesterol crystallization and gallstone formation observed by phase contrast and polarizing light microscopy in gallbladder bile of gallstone-susceptible C57L/J mice during the 8-week period of feeding a lithogenic diet: (a) mucin gel; (b) small liquid crystals; (c) aggregated liquid crystals; (d) fused liquid crystals with Maltese-cross birefringence and focal conic textures; (e) arc-like (possible anhydrous cholesterol) crystal; (f) filamentous crystals; (g) tubular crystal fracturing at ends to produce cholesterol monohydrate crystals; (h) typical cholesterol monohydrate crystal, with 79.2° and 100.8° angles; (i) agglomerated cholesterol monohydrate crystals; (j and k) sandy stones; and (l) gallstones. Reproduced with permission from reference [13].

CONTRIBUTION OF MOUSE MODELS TO GENETIC STUDIES OF GALLSTONE DISEASE

Epidemiological and clinical investigations, as well as family and twin studies have clearly demonstrated that a complex genetic basis plays a key role in determining individual predisposition to develop gallstones in response to environmental factors [14]. However, the genetic contribution to gallstones in humans is not completely understood. Because gallstone disease could be caused by complex interactions of many environmental factors and the effects of multiple but as yet unknown genes [14–16], methods for studying genetics of gallstones, a complex genetic trait, are completely different from those that are used to discover genes for simple Mendelian defects. Furthermore, to study quantitative polygenic traits, conventional genetic mapping methods are inappropriate because they are designed for single gene traits. The inbred mouse, with its superior genetic resources [17,18], is an excellent animal model for investigating genetic determinants of gallstone disease. By using mouse backcross strategies and a powerful genetic technique, quantitative trait locus (QTL) analysis [19–24], Khanuja et al. [7] were the first to find that differences in gallstone susceptibility between gallstone-susceptible C57L/J and gallstone-resistant AKR/J strains, after fed the lithogenic diet for 12 weeks, were determined by at least two Lith genes, with Lith1 and Lith2 mapping on mouse chromosomes 2 and 19, respectively. Subsequently, the physical–chemical phenotypes of these Lith genes in gallbladder and hepatic bile, as well as in the liver, gallbladder, and small intestine of C57L/J vs. AKR/J mice were systematically investigated [13,25–27]. These studies found that gallstone disease is determined by multiple genes, which is a dominant trait. To confirm whether Lith1 or Lith2 alone could independently induce gallstones, congenic strains carrying the C57L/J alleles only in the Lith1 or the Lith2 QTL regions on an AKR/J genetic background were created [28]. Both Lith1 and Lith2 congenic strains showed gallstone formation comparable to the C57L/J parent, confirming that Lith1 and Lith2 could cause gallstones mostly through different lithogenic mechanisms [28]. Furthermore, Lith1 plays a major role in determining hepatic cholesterol hypersecretion as the biliary phenotypes in the Lith1 congenic strain are basically the same as those in the C57L/J parent. Lith2 could influence bile acid-independent bile flow. After many additional strains of inbred mice fed the lithogenic diet for 8–12 weeks were investigated for the prevalence and pathophysiology of gallstones, new gallstone QTL regions containing Lith genes were identified. As a result, a mouse gallstone gene map, containing 25 Lith genes from Lith1 to Lith25 and other candidate genes, was established, which are distributed from chromosome 1 to X in mice, as shown in Fig. 2 and Table 1 [14–16,29–31].

FIGURE 2.

The gallstone (Lith) gene map shows the quantitative trait loci (QTL) regions containing Lith genes, as well as candidate genes for cholesterol gallstone disease on mouse chromosomes. A vertical line represents one chromosome, with the centromere at the top. Genetic distances from the centromere (horizontal black lines) are indicated to the left of the chromosomes in centimorgans (cM). The locations of gallstone QTLs (Lith genes) and candidate genes are indicated by horizontal black lines with the gene symbols to the right (see Table 1 for the list of gene symbols and names). Reproduced with permission from reference [31].

Table 1.

2017 Inventory of the candidate gallstone genes in mice and humans

Gene symbol	Gene name	Mouse chromosome	cM	Human orthologue
A. Liver
(i) Lipid membrane transporters
ABCG5 and ABCG8	ATP-binding cassette, sub-family G (WHITE), member 5 and member 8	17	55.02	2p21
ABCB4 (MDR2)	ATP-binding cassette, sub-family B (MDR/TAP), member 4 (phosphatidylcholine transporter; multiple drug resistance 2)	5	3.43	7q21.1
ABCB11 (BSEP, SPGP)	ATP-binding cassette, sub-family B (MDR/TAP), member 11 (bile salt export pump; sister of P-glycoprotein)	2	39.69	2q24
ABCC2 (CMOAT, MRP2)	ATP-binding cassette, sub-family C (CFTR/MRP), member 2 (canalicular multispecific organic anion transporter; multidrug resistance-related protein 2)	19	36.67	10q24
SLC10A1 (NTCP)	solute carrier family 10 (sodium/bile acid cotransporter family), member 1 (sodium/taurocholate cotransporting polypeptide)	12	37.21	14q24.1
SLC21A1 (OATP1)	solute carrier organic anion transporter family, member 1a1 (organic anion transporting polypeptide 1)	6	73.09	n/a
SLC22A1 (OCT1, ORCT)	solute carrier family 22 (organic cation transporter), member 1 (organic cation transporter 1)	17	8.63	6q25.3
ATP8B1 (FIC1)	ATPase, class I, type 8B, member 1 (familial intrahepatic cholestasis type 1)	18	37.49	18q21.31
NPC1L1	Niemann-Pick C1-like 1	11	3.93	7p13
(ii) Lipid regulatory enzymes
CYP7A1	cytochrome P450, family 7, subfamily a, polypeptide 1 (cholesterol 7α-hydroxylase)	4	2.91	8q11-q12
CYP7B1	cytochrome P450, family 7, subfamily b, polypeptide 1 (oxysterol 7α-hydroxylase)	3	4.98	8q21.3
CYP27A1	cytochrome P450, family 27, subfamily a, polypeptide 1 (sterol 27-hydroxylase)	1	38.54	2q35
HMGCR	3-hydroxy-3-methylglutaryl-Coenzyme A reductase	13	50.65	5q13.3-q14
SOAT2 (ACAT2)	sterol O-acyltransferase 2 (acyl-coenzyme A:cholesterol acyltransferase 2)	15	57.33	12q13.13
UGT1A1	uridine diphosphate (UDP)-glucuronosyltransferase 1 family, polypeptide A1	1	44.55	2q37
(iii) Intracellular lipid transporters
CAV1	caveolin 1	6	7.73	7q31.1
CAV2	caveolin 2	6	7.72	7q31.1
FABP1	fatty acid binding protein 1, liver	6	32.14	2p11
NPC1	Niemann-Pick type C1	18	6.15	18q11.2
OSBP	oxysterol binding protein	19	8.58	11q12-q13
PCTP	phosphatidylcholine transfer protein	11	54.7	17q21-q24
SCP2	sterol carrier protein 2	4	50.2	1p32
CETP	cholesteryl ester transfer protein	n/a	n/a	16
(iv) Lipid regulatory transcription factors
NR2B1 (RXRA)	nuclear receptor subfamily 2, group B, member 1 (retinoid X receptor alpha)	2	19.38	9q34.3
NR1H3 (LXRA)	nuclear receptor subfamily 1, group H, member 3 (liver X receptor alpha)	2	50.52	11p11.2
NR1H2 (LXRB)	nuclear receptor subfamily 1, group H, member 2 (liver X receptor beta)	7	28.83	19q13.3
NR1H4 (FXR)	nuclear receptor subfamily 1, group H, member 4 (farnesoid X receptor)	10	44.98	12q23.1
NR1C1 (PPARA)	peroxisome proliferator activated receptor alpha	15	40.42	22q13.31
NR1C2 (PPARD)	peroxisome proliferator activator receptor delta	17	14.64	6p21.2
NR1C3 (PPARG)	peroxisome proliferator activated receptor gamma	6	53.41	3p25
NR0B2 (SHP)	nuclear receptor subfamily 0, group B, member 2 (small heterodimer partner)	4	66.25	1p36.1
NR1I3 (CAR)	nuclear receptor subfamily 1, group I, member 3 (constitutive androstane receptor)	1	79.21	1q23.3
SREBF1	sterol regulatory element binding transcription factor 1	11	37.81	17p11.2
SREBF2	sterol regulatory element binding transcription factor 2	15	38.49	22q13
SCAP	SREBF chaperone (SREBF cleavage activating protein)	9	59.91	3p21.31
ESR1 (ERα)	estrogen receptor 1 (alpha)	10	2.03	6q25.1
ESR2 (ERβ)	estrogen receptor 2 (beta)	12	33.52	14q23.2
GPER1 (GPR30)	G protein-coupled estrogen receptor 1	5	78.58	7p22.3
AR	androgen receptor	X	42.82	Xq12
ADRB3	adrenergic receptor, beta 3	8	15.94	8p12
(v) Lipoprotein receptors and related genes
APOB	apolipoprotein B	12	2.53	2p24-p23
APOE	apolipoprotein E	7	9.94	19q13.2
APOA5	apolipoprotein A-V	9	25.37	11q23
LDLR	low density lipoprotein receptor	9	7.87	19p13.2
LRP1	low density lipoprotein receptor-related protein 1	10	74.52	12q13.3
LRP2 (GP330)	low density lipoprotein receptor-related protein 2 (glycoprotein 330; megalin)	2	40.74	2q31.1
SCARB1 (SRB1)	scavenger receptor class B, member 1	5	64.11	12q24.31
LPL	lipoprotein lipase	8	33.88	8p22
LIPC (HPL)	hepatic lipase	9	39.57	15q21-q23
LCAT	lecithin cholesterol acyltransferase	8	53.06	16q22.1
LRPAP1	low density lipoprotein receptor-related protein associated protein 1	5	18.01	4p16.3
PLTP	phospholipid transfer protein	2	85.27	20q13.12
B. Gallbladder
(i) Hormone receptors
CCK1R (CCKAR)	cholecystokinin-1 receptor (cholecystokinin-A receptor)	5	29.52	4p15.2
ESR1 (ERα)	estrogen receptor 1 (alpha)	10	2.03	6q25.1
ESR2 (ERβ)	estrogen receptor 2 (beta)	12	33.52	14q23.2
GPER1 (GPR30)	G protein-coupled estrogen receptor 1	5	78.58	7p22.3
PGR	progesterone receptor	9	2.46	11q22-q23
(ii) Mucin
MUC1	mucin 1	3	39.02	1q21
MUC2	mucin 2	7	87.1	11p15.5
MUC3	mucin 3	5	76.22	n/a
MUC4	mucin 4	16	23.3	3q29
MUC5AC	mucin 5ac	7	87.23	11p15.5
MUC5B	mucin 5b	7	87.29	11p15.5
MUC6	mucin 6	7	87.03	11p15.5
(iii) Lipid membrane transporters
ABCG5 and ABCG8	ATP-binding cassette, sub-family G (WHITE), member 5 and member 8	17	55.02	2p21
SCARB1 (SRB1)	scavenger receptor class B, member 1	5	64.11	12q24.31
(iv) Lipid regulatory enzymes
HMGCR	3-hydroxy-3-methylglutaryl-Coenzyme A reductase	13	50.65	5q13.3-q14
SOAT2 (ACAT2)	sterol O-acyltransferase 2 (acyl-coenzyme A:cholesterol acyltransferase 2)	15	57.33	12q13.13
C. Small intestine
(i) Lipid membrane transporters
ABCG5 and ABCG8	ATP-binding cassette, sub-family G (WHITE), member 5 and member 8	17	55.02	2p21
SCARB1 (SRB1)	scavenger receptor class B, member 1	5	64.11	12q24.31
SLC10A2 (IBAT)	solute carrier family 10, member 2 (ileal bile salt transporter)	8	2.16	13q33
(ii) Lipid regulatory enzymes
HMGCR	3-hydroxy-3-methylglutaryl-Coenzyme A reductase	13	50.65	5q13.3-q14
SOAT2 (ACAT2)	sterol O-acyltransferase 2 (acyl-coenzyme A:cholesterol acyltransferase 2)	15	57.33	12q13.13
(iii) Intracellular lipid transporters
CAV1	caveolin 1	6	7.73	7q31.1
CAV2	caveolin 2	6	7.72	7q31.1
FABP6 (ILLBP)	fatty acid binding protein 6, ileal	11	25.81	5q33.3-q34
OSBP	oxysterol binding protein	19	8.58	11q12-q13
SCP2	sterol carrier protein 2	4	50.2	1p32
NPC1L1	Niemann-Pick C1-like 1	11	3.93	7p13
APOB	apolipoprotein B	12	2.53	2p24-p23
APOA1	apolipoprotein A-I	9	25.36	11q23-q24
APOA4	apolipoprotein A-IV	9	25.36	11q23
APOC3	apolipoprotein C-III	9	25.36	11q23
(iv) Lipid regulatory transcription factors
NR2B1 (RXRA)	nuclear receptor subfamily 2, group B, member 1 (retinoid X receptor alpha)	2	19.38	9q34.3
NR1H3 (LXRA)	nuclear receptor subfamily 1, group H, member 3 (liver X receptor alpha)	2	50.52	11p11.2
NR1H2 (LXRB)	nuclear receptor subfamily 1, group H, member 2 (liver X receptor beta)	7	28.83	19q13.3
NR1H4 (FXR)	nuclear receptor subfamily 1, group H, member 4 (farnesoid X receptor)	10	44.98	12q23.1
NR1C1 (PPARA)	peroxisome proliferator activated receptor alpha	15	40.42	22q13.31
NR1C2 (PPARD)	peroxisome proliferator activator receptor delta	17	14.64	6p21.2
NR1C3 (PPARG)	peroxisome proliferator activated receptor gamma	6	53.41	3p25
(v) Hormone receptors
CCK	cholecystokinin	9	72.43	3p22.1
CCK1R (CCKAR)	cholecystokinin-1 receptor (cholecystokinin-A receptor)	5	29.52	4p15.2
ESR1 (ERα)	estrogen receptor 1 (alpha)	10	2.03	6q25.1
ESR2 (ERβ)	estrogen receptor 2 (beta)	12	33.52	14q23.2
GPER1 (GPR30)	G protein-coupled estrogen receptor 1	5	78.58	7p22.3

cM, centimorgan; n/a, not available; p, short arm of the chromosome; q, long arm of the chromosome.

Map position is based on conserved homology between mouse and human genomes and assigned indirectly from localization in other species. Information on homologous regions was retrieved from the mouse/human homology databases maintained at the Jackson Laboratory (http://www.informatics.jax.org/searches/marker_form.shtml) and the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/HomoloGene). Reproduced with permission from reference [31].

These results from mouse genetic studies have led to the discovery of human LITH genes because of homologues between human and mouse chromosomes. Such a successful study is the confirmation of ABCG5/G8 as a human LITH gene based on mouse studies. The Abcg5/g8 is first identified as the mouse Lith9 by the QTL mapping methods [29,32,33], and subsequently, two major gallstone-associated variants in ABCG5/G8 (i.e., ABCG5-R50C and ABCG8-D19H) are found not only in Germans and Chileans, but also in Chinese and Indians [34–41]. Therefore, based on the mouse Lith gene map, more human LITH genes will be identified and their pathogenic mechanisms will be elucidated in the near future.

Based on extensive mouse and human studies, it was first proposed in 2010 [31] that interactions of five primary defects play a critical role in the pathogenesis of cholesterol gallstone disease (Fig. 3): genetic factors and Lith genes; hepatic cholesterol hypersecretion, leading to cholesterol-supersaturated gallbladder bile (i.e., high cholesterol saturation index); rapid phase transitions by accelerated cholesterol crystallization and solid cholesterol crystal growth; impaired gallbladder motility; and intestinal factors, including increased amounts of the absorbed cholesterol delivered from the small intestine to the liver for biliary hypersecretion, as well as changes in intestinal microbiota, warrant a systematical study to provide novel insight into gallstone pathogenesis. Obviously, rapid growth and agglomeration of solid cholesterol crystals to form microlithiasis and macroscopic stones is a consequence of cholesterol-supersaturated gallbladder bile, followed by gallbladder mucin hypersecretion and gel formation with impaired gallbladder emptying and refilling, eventually leading to the formation of biliary sludge, that is, the precursor of gallstones.

FIGURE 3.

Interactions of five primary defects promote the formation of cholesterol gallstones: (i) genetic factors and Lith genes; (ii) hepatic hypersecretion; (iii) gallbladder hypomotility; (iv) rapid phase transitions; and (v) intestinal factors. Persistent hepatic cholesterol hypersecretion is the consequence of complex genetic predispositions, leading to the formation of cholesterol-supersaturated bile and accelerating cholesterol crystallization. Impaired gallbladder motility results in the production and accumulation of excess mucin gel, promoting the formation of biliary sludge and the growth of microlithiasis. These alterations also disrupt the kinetics of the enterohepatic circulation of bile acids (intestinal factors), leading to a diminished intestinal absorption and pool size of bile acids. Increased cholesterol absorption delivers dietary and re-absorbed biliary cholesterol to the liver for secretion into bile. Abnormal intestinal microbiota may disrupt cholesterol and bile acid metabolism in the liver, intestine, and bile, as well as impair gallbladder emptying and refilling. Reproduced with permission from reference [31].

THE LATEST ADVANCES IN THE PATHOPHYSIOLOGY OF GALLSTONES FROM MOUSE EXPERIMENTS

Gallstone prevalence is markedly higher in women than in men at all ages, suggesting that estrogen plays a critical role in the pathogenesis of cholesterol cholelithiasis [42]. Although the classical estrogen receptor α (ERα), but not ERβ, in the liver plays a critical role in the formation of 17β-estradiol (E₂)-induced gallstones in female mice [43–45], the metabolic abnormalities underlying the lithogenic effect of E₂ on gallstone formation is still unclear. Gallstones are still found in ~30% of ovariectomized ERα knockout mice treated with high doses of E₂ [46]. Moreover, genetic analysis in mice reveals that the G-protein-coupled receptor 30 (GPR30), a novel estrogen receptor, is a gallstone gene named Lith18 [16,31,47,48]. This discovery led to a new question of how E₂, through GPR30, ERα, or both, influences the biliary and gallstone phenotypes because it can efficiently bind to and activate both GPR30 and ERα [49]. To distinguish the lithogenic effect of ERα from that of GPR30, the entire spectrum of cholesterol crystallization pathways and sequences during the early stage of gallstone formation are investigated in gallbladder bile of ovariectomized female wild-type, GPR30 knockout, ERα knockout, and GPR30/ERα double knockout (DKO) mice. E₂ activates GPR30 and ERα to produce liquid crystalline vs. anhydrous crystalline metastable intermediates, respectively, evolving to cholesterol monohydrate crystals from supersaturated bile [50^▪]. Moreover, cholesterol crystallization is drastically retarded in GPR30/ERα DKO mice. This indicates that GPR30 exerts a synergistic lithogenic action with ERα to enhance E₂-induced gallstone formation. Because GPR30 is localized predominantly in the endoplasmic reticulum, but not in the nucleus, of hepatocytes, GPR30 activation by E₂, likely through the epidermal growth factor receptor signaling cascade, could inhibit hepatic cholesterol 7α-hydroxylase and the classical pathway of bile acid synthesis, leading to the availability of excess cholesterol for hepatic hypersecretion and bile lithogenesis [50^▪].

Although nonalcoholic fatty liver disease (NAFLD) is an important risk factor for gallstone formation [51–53], the mechanisms of linking NAFLD with gallstone disease are still unknown. Hypoxia-inducible factor 1 (HIF1) is an important transcription factor for regulating expression of the genes involved in oxygen delivery, cellular growth, and redox homeostasis, facilitating adaptive responses to hypoxic conditions [54–57]. HIF1A is found mainly in the perivenous area of the liver, where it is physiologically hypoxic [58]. During the development of liver steatosis, lipid accumulation markedly enlarges the size of hepatocytes, thereby reducing hepatic sinusoidal perfusion and microcirculation and eventually leading to hepatic hypoxia [59]. Asai et al. [60^▪▪] found that expression of aquaporin-8 (AQP8), a water channel that is responsible for hepatic water secretion into the bile canaliculi [61–63], is markedly increased in the inducible hepatocyte-selective HIF1A knockout mice fed the lithogenic diet for 2 weeks. These changes dramatically increase bile flow by 35%, dilute biliary lipid concentrations by 36% in both gallbladder and hepatic bile, and alleviate gallbladder inflammation. As a result, cholesterol crystallization is inhibited and gallstone formation is prevented in liver-specific HIF1A knockout mice. In contrast, activation of the HIF1A pathway in diet-induced steatotic liver accelerates gallstone formation in wild-type mice. Moreover, expression of HIF1A and its downstream targets in the liver is increased in NAFLD patients with gallstones, suggesting that hepatic HIF1A may play a vital role in the formation of cholesterol gallstones in patients with NAFLD.

Bile flow is determined by both bile acid-dependent and bile acid-independent pathways, in which transhepatocyte transport of water and ionic/non-ionic solutes through the transcellular and the para-cellular routes facilitates the formation of bile [64]. Claudin 2 (Cldn2) is a paracellular channel-forming protein that is highly expressed in the hepatocyte and cholangiocyte [65–67]. Matsumoto et al. observed [68] that compared to wild-type mice, Cldn2 knockout mice display reduced bile flow rate by ~50%, which is attributable to a decrease in transepithelial electrical conductance and water permeability in the hepatobiliary system. This significantly increases total lipid concentrations in hepatic and gallbladder bile. Furthermore, osmotic gradient-driven water flow is dramatically reduced in hepatocyte bile canaliculi and bile ducts isolated from Cldn2 knockout mice. After 4 weeks on lithogenic diets, gallstones are formed in Cldn2 knockout, but not wild-type, mice. These findings indicate that claudin 2 is critical for bile formation and biliary lipid homeostasis. Dysfunctional claudin 2 enhances cholelithogensis by impairing paracellular water movement into bile, leading to concentrated bile that greatly facilitates cholesterol crystallization [69–71].

Similar to patients with sitosterolemia [72–75], hepatic cholesterol output is significantly reduced, but not completely eliminated in mice with the deletion of either Abcg5 or Abcg8 alone, or both genes [76–80]. These findings indicate that an ABCG5/G8-independent pathway could exist for regulating biliary cholesterol secretion in humans and mice. Wang et al. [81] reported that the ABCG5/G8-independent pathway accounts for 30% to 40% of hepatic cholesterol output in the lithogenic state and plays a pivotal role in regulating biliary secretion of cholesterol in response to high dietary cholesterol. In the absence of ABCG5/G8, this pathway is involved in biliary cholesterol secretion and gallstone formation. Different from ABCG5/G8, this pathway is not activated by the LXR agonist through the LXR signaling pathway. This finding indicates that the importance of both ABCG5/G8-dependent and ABCG5/G8-independent pathways in hepatic cholesterol secretion and gallstone formation.

Clinical studies have found that intestinal cholecystokinin (CCK) secretion and gallbladder emptying in response to a fatty meal are impaired before celiac patients start the gluten-free diet [82–87]. However, it has not been really appreciated whether celiac disease is associated with gallstones because few studies are performed to investigate the impact of celiac disease on gallstone formation [88]. After a mouse model with the lack of endogenous CCK is created to study the biliary characteristics of celiac disease, Wang et al. found [88–90] that CCK knockout mice display rapid cholesterol crystallization and gallstone formation with impaired gallbladder emptying and with excess amounts of intestine-derived cholesterol for biliary hypersecretion through reducing small intestinal transit time and increasing intestinal cholesterol absorption. These mouse experiments provide important clues to study the prevalence and pathogenesis of gallstones in patients with celiac disease [88]. Moreover, the potent CCK-1 receptor antagonist devazepide increases susceptibility to gallstone formation by impairing gallbladder emptying function, disrupting biliary cholesterol metabolism and enhancing intestinal cholesterol absorption in mice [91].

Although bile may contain both pro-nucleating and antinucleating agents and an imbalance in their abundance may lead to rapid cholesterol crystallization in gallbladder bile of patients with cholesterol gallstones [92–97], this concept still needs extensive studies to prove it. Mucin is the first biliary protein found to play a lithogenic role in accelerating cholesterol crystallization in animal and human studies [98–101]. Although many other biliary proteins have been proposed to be either pro-nucleating or antinucleating agents for influencing cholesterol crystallization in bile, their in vivo roles, if any, in the pathogenesis of gallstones remain largely unclear [102]. Niemann-Pick C2 (NPC2), a soluble lysosomal protein, is expressed in human and mouse liver and regulates hepatic cholesterol secretion by stimulating Abcg5/g8-mediated cholesterol transport [103]. After secreted into bile [104–106], it promotes cholesterol crystallization and gallstone formation in mice transgenic human NPC2 [107^▪]. In contrast, Npc2 knockout mice show a reduction in hepatic cholesterol secretion and gallstone formation at 2 weeks on a lithogenic diet [106]. These results imply that biliary proteins could influence gallstone formation by modulating cholesterol crystallization likely through a liquid crystalline pathways in bile [108–111].

High efficiency of intestinal cholesterol absorption and high dietary cholesterol such as the Western diet are two independent risk for gallstones in mice [112]. Further studies found that numerous factors that modulate intestinal cholesterol uptake and absorption, cholesterol esterification, lipid transport within the enterocytes, and chylomicron assembly and secretion could influence cholesterol crystallization and gallstone formation by regulatingdeliveryof the intestinal origin of cholesterol to the liver for secretion into bile [113^▪,114–119]. Moreover, the importance of intestinal microbiota and the immune system in the pathogenesis of gallstones needs to be investigated extensively in the future.

Obviously, many factors that change hepatic synthesis, transhepatocyte transport, and biliary secretion of cholesterol, phospholipids, and bile acids could influence gallstone formation by causing cholesterol hypersecretion and cholesterol-supersaturated bile [120–127]. Nuclear receptors such as liver X receptor (LXR), farnesoid X receptor (FXR), pregnane X receptor (PXR), constitutive androstane receptor (CAR), and ERα have been found to influence gallstone formation through these regulatory pathways of lipid metabolism in the liver [128–131]. More importantly, disruption of the liver-specific insulin receptor in mice increases susceptibility to gallstones by increasing hepatic cholesterol secretion through the forkhead transcription factor FoxO1 and Abcg5/g8 and leading to a lithogenic bile acid profile through reduction of enzymatic activity of cholesterol 7α-hydroxylase and bile acid synthesis of classical pathway [127]. However, the pathogenic mechanisms underlying the roles of insulin resistance in gallstone formation remain elusive.

CONCLUSION

Based on findings from the genetic studies of gallstones in mice, a novel concept regarding the genetic mechanisms of gallstone disease has been established, which has opened the door for investigating the orthologous human LITH genes and for exploring their cholelithogenic mechanisms in humans [132]. For future research, mouse studies will have to focus on the roles of nuclear receptors, intestinal microbiota, dyslipidemia, hyperinsulinemia, NAFLD, obesity, diabetes, aging, and sedentary lifestyle in the mechanisms underlying the formation of cholesterol gallstones. These results from mouse experiments will provide novel strategies for the treatment of this very prevalent hepatobiliary disease worldwide.

KEY POINTS.

• Cholesterol gallstone disease is caused by complex genetic and environmental factors. It is one of the most common and costly digestive diseases in Western countries. At least 20 million Americans (~12% of adults) have gallstones.

• Compelling evidence from pathophysiological, physical–chemical, and genetic studies shows that cholesterol gallstone disease is determined by multiple Lith genes, which is a dominant trait. However, no mode of inheritance fitting to the Mendelian pattern could be found in most cases. The overarching pathogenic factor is persistent hepatic hypersecretion of cholesterol into bile, leading to cholesterol-supersaturated gallbladder bile.

• Understanding the molecular genetics of gallstone disease in mice will push for the identification of human LITH genes and for the understanding of the pathogenic mechanisms of each of LITH genes in humans.

• Further mouse studies will investigate the roles of nuclear receptors, intestinal microbiota, dyslipidemia, hyperinsulinemia, NAFLD, obesity, diabetes, aging, and sedentary lifestyle in the formation of cholesterol gallstones.

• With such knowledge of the genetic and environmental factors involved in gallstone pathogenesis, mouse studies will provide important insight into the complex pathogenic mechanisms of gallstone disease. This will facilitate the development of a novel, effective, and noninvasive therapy for patients with gallstones.