[Frontiers in Bioscience 14, 4242-4256, January 1, 2009]

[Frontiers in Bioscience 14, 4242-4256, January 1, 2009]

Liver diseases related to MDR3 (ABCB4) gene deficiency

Emmanuel Gonzales^1,2, Anne Davit-Spraul³, Christiane Baussan³, Catherine Buffet⁴, Michele Maurice⁵, Emmanuel Jacquemin^1,2

1Pediatric Hepatology and National Reference Centre for Biliary Atresia, Bicêtre Hospital, University of Paris - South 11, AP-HP, Paris, France, ²INSERM, UMR-S757, University of Paris - South 11, Orsay, France, ³Biochemistry, Bicetre Hospital, University of Paris - South 11, AP-HP, Paris, France, ⁴Hepatology, Bicetre Hospital, University of Paris - South 11, AP-HP, Paris, France, ⁵INSERM, U538, University Pierre and Marie Curie, Paris, France

TABLE OF CONTENTS

1. Abstract 2 .Introduction
3. The mdr2 knockout mouse : a model of liver pathology deficient in biliary phospholipid secretion
4. The spectrum of liver diseases related to MDR3 deficiency: 4.1.Progressive familial intrahepatic cholestasis type 3 (PFIC3)
5. Genotype - phenotype correlation among liver diseases related to MDR3 deficiency
6. Treatment of MDR3 deficiency (mainly PFIC3): present and future
7. References

1. ABSTRACT

Class III multidrug resistance P-glycoproteins, mdr2 in mice and MDR3 in human, are canalicular phospholipid translocators involved in biliary phospholipid (phosphatidylcholine) excretion.The role of a MDR3 (ABCB4) gene defect in liver disease has been initially proven in a subtype of progressive familial intrahepatic cholestasis called PFIC3, a severe pediatric liver disease that may require liver transplantation.Several MDR3 mutations have been identified in children with PFIC3 and are associated to low level of phospholipids in bile leading to high biliary cholesterol saturation index.MDR3 mutations are associated to loss of canalicular MDR3 protein and /or to loss of protein function.There is evidence that biallelic or monoallelic MDR3 defect causes or predisposes to 6 human liver diseases (PFIC3, adult biliary cirrhosis, low phospholipid associated cholelithiasis syndrome, transient neonatal cholestasis, intrahepatic cholestasis of pregnancy, drug induced cholestasis).Some patients with MDR3 deficiency may benefit from ursodeoxycholic acid therapy and could be good candidates to a targeted pharmacological approach and/or to cell therapy in the future.

2. INTRODUCTION

The role of a defect of the multidrug resistance 3 gene (MDR3) in liver disease has been initially suspected in a subtype of progressive familial intrahepatic cholestasis (PFIC) (1).In a general sense, PFIC is an heterogeneous group of autosomal recessive liver disorders of childhood in which cholestasis of hepatocellular origin often presents in the neonatal period or the first year of life and leads to death from liver failure at ages ranging from infancy to adolescence (2-7).Recent molecular and genetic studies have allowed the identification of genes responsible for three types of PFIC and have shown that PFIC was related to mutations in hepatocellular transport system genes involved in bile formation (8-12).These findings now provide specific tools for the precise diagnosis of PFIC.The first type of PFIC, PFIC1, is caused by mutation of the ATP8B1 gene (13, 14).This gene codes for a P-type ATPase (FIC1) which is expressed in several organs but whose function is not precisely known.The second type of PFIC, PFIC2, is caused by mutation of the ABCB11 gene that codes for the ATP-dependent canalicular bile salt export pump (BSEP) (15-18).Despite cholestasis, patients with PFIC1 and PFIC2 have normal serum gamma-glutamyltransferase (GGT) activity (4, 12, 16, 19, 20).Liver histology is characterized by the absence of a true ductular proliferation with only periportal biliary metaplasia of hepatocytes (19, 21).PFIC1 and PFIC2 will be considered in details in other chapters of this issue.The third type of PFIC, PFIC3, is caused by mutation of the MDR3 (ABCB4) gene (1, 22-29).ABCB4 codes for a P-glycoprotein of class III multidrug resistance, involved in biliary phospholipid excretion.PFIC3, can be distinguished from the other types by a high serum GGT activity and liver histology which shows ductular proliferation and inflammatory infiltrate in the early stages despite patency of intra and extrahepatic bile ducts (1, 22-29).The histological pattern occurring in this disorder is very similar to the hepatic injury observed in mice with a homozygous disruption of the mdr2 gene (mdr2 -/- mice), the murine orthologue of MDR3 (30).Mdr2 in mice and MDR3 in human, are phospholipid translocators involved in biliary phospholipid (phosphatidylcholine) excretion and are predominantly, if not exclusively, expressed in the canalicular membrane of the hepatocyte (26, 28, 31-34).An abnormal expression of the MDR3 gene (low amount of liver mRNA) and low biliary phospholipid concentration have been found in PFIC3 patients (1, 29).MDR3 P-glycoprotein belongs to the family of ATP binding cassette (ABC) transporters and the MDR3 gene is localized on chromosome 7q21.This review article will consider the pathophysiology of genetic mdr2/MDR3 deficiency and the different pediatric and adult cholestatic diseases which are now known to be caused by a genetic MDR3 defect.So far, there is evidence that biallelic or monoallelic MDR3 defect causes or predisposes to 6 human liver diseases (PFIC3, adult biliary cirrhosis, low phospholipid associated cholelithiasis syndrome, intrahepatic cholestasis of pregnancy, transient neonatal cholestasis, drug induced cholestasis) (26, 28).Finally, the therapeutic options, including drug therapy that may target MDR3 expression and function, will be considered.

3. THE Mdr2 KNOCKOUT MOUSE : A MODEL OF LIVER PATHOLOGY DEFICIENT IN BILIARY PHOSPHOLIPID SECRETION

Mdr2 (-/-) mice suffer from liver disease that starts at a few weeks of age and progresses throughout life (30, 35).The most histologic striking feature, besides hepatocyte necrosis and dilated canaliculi, is the presence of a cholangiopathy represented by portal tract inflammation and a severe ductular proliferation which progress through the first 3 months of age.At the age of 4 to 6 months, the mdr2 (-/-) mice start to develop nodules in the liver parenchyma, which histologically resemble the picture of chemically induced hepatocarcinogenesis.These nodules develop in hepatocellular carcinoma, and in mice older than 1 year metastases were observed in the lungs.The mechanism of tumor formation is unclear.Hepatocyte replication as a consequence of cell damage could be involved in tumor formation but this does not hold for cholangiocytes which proliferate even stronger.The absence of mdr2 P-glycoprotein could lead to accumulation of carcinogenic compounds in the hepatocytes.This could explain that tumors seem to be derived from hepatocytes rather than from cholangiocytes.An important observation from this model is that the bile of mdr2 (-/-) mice is almost devoided of phosphatidylcholine whereas bile salt secretion is normal.This suggested that mdr2 P-glycoprotein was involved in the biliary secretion of phosphatidylcholine.In this model, the absence of mdr2 P-glycoprotein function clearly has deleterious effects on the bile canaliculi and the biliary cells.The cholangiopathy is probably caused by the cytotoxicity of bile salts in absence of biliary phospholipids.The cholangiopathy observed in the mdr2 (-/-) mice may be relatively mild because of the relatively hydrophilic bile salt composition in this animal.Increasing the hydrophobicity of the bile salt pool by cholate feeding leads to a more severe liver pathology, whereas further decreasing its hydrophobicity by ursodeoxycholate acid (UDCA) feeding improves liver histology (36).Upon cholate feeding, bile cholesterol crystals have been observed in mdr2 (-/-) mice.Because the human bile salt pool is much more hydrophobic than in mice, it was expected that a defect of biliary phospholipid secretion in human could have much more dramatic consequences than in the mouse.Nevertheless, mdr2 (-/-) mice develop sclerosing cholangitis, while sclerosing cholangitis like cholangiopathy has not been described in PFIC3 patients (29, 37, 38).Currently, there is no evidence for a role of MDR3 variants as modifier gene in sclerosing cholangitis in adults (39-43).The production of this mouse model with a specific defect in biliary phospholipid secretion has made possible the identification of the analogous inherited human liver disease called PFIC3 and due to MDR3 deficiency (1, 22, 29).

4. THE SPECTRUM OF LIVER DISEASES RELATED TO MDR3 DEFICIENCY P ALIGN="JUSTIFY">4.1.Progressive familial intrahepatic cholestasis type 3 (PFIC3)

4.1.1.PFIC3 phenotype

The disease transmission is autosomal recessive.Parents of affected patients are frequently related and a similar liver disease is often observed in siblings.Age at first symptoms ranges from 1 month to 20.5 years of age.First symptoms in most cases consist of the presence of jaundice, discoloured stools, hepatomegaly, splenomegaly or pruritus.Clinical signs of cholestasis are noted within the first year of life in about one third of patients and rarely in the first month (neonatal) period.Gastrointestinal bleeding due to portal hypertension and cirrhosis is the presenting symptom in adolescent or young adult patients.Initial serum liver tests (mean) show: elevated alanine aminotransaminase activity (5xN), conjugated bilirubin concentration (2xN), alkaline phosphatase activity (2xN), GGT activity (13xN), total bile acid concentration (25xN), and normal cholesterol concentration and prothrombin time.Evolution is characterized by chronic icteric or anicteric cholestasis, portal hypertension and liver failure.In half of the patient, liver transplantation is required at a mean age of 7.5 years.No liver tumor has been reported.Liver histology obtained at time of diagnosis shows portal fibrosis and ductular proliferation with mixed inflammatory infiltrate.In a few instances cholestasis is present in the lobule and there is giant transformation of hepatocytes.Cytokeratin immunostaining confirms the strong ductular proliferation within the portal tract (22).At a later stage, there is extensive portal fibrosis and a typical picture of biliary cirrhosis.Interlobular bile ducts are seen in most portal tracts, and there is neither periductal fibrosis nor biliary epithelium injury.In a few instances cholestasis is present in the lobule and in some ductules containing bile plugs.Cholangiography is normal and ultrasonography of the liver shows normal bile ducts in all patients (1, 22, 29).These results allow to distinguish PFIC3 patients from those with sclerosing cholangitis on the basis of histological and cholangiographic data.They can also be distinguished from patients with the other types of PFIC (PFIC1 and PFIC2) in that they present very rarely with cholestatic jaundice at the neonatal period, but rather in late infancy, childhood or in young adulthood (4, 9, 10, 16, 18, 29).Patients with PFIC3 have a persistent high serum GGT activity, moderately raised concentrations of serum primary bile salts and a mild pruritus. PFIC3 carries a higher risk of portal hypertension and gastrointestinal bleeding and ends in liver failure at a later age but as patients with PFIC1 and PFIC2 phenotypes, most of them have cholesterol level within the normal range.This may be explained by the inability of PFIC3 patients to generate lipoprotein X, the formation of which is mediated by class III P-glycoproteins (44).Indeed, we found no lipoprotein X in the serum of patients harboring homozygous nonsense MDR3 mutations (29).A schematic approach to the diagnosis of PFIC3 is proposed in Figure 1.This combined clinical, biochemical, histological and radiological approach associated to biliary phospholipid dosage, and liver MDR3 immunostaining should help to select PFIC3 candidates in whom a molecular diagnosis of MDR3 deficiency could be proposed.Children with PFIC3 are at risk to develop intrahepatic and extrahepatic cholesterol cholelithiasis and drug induced cholestasis (DIC) (26).Girls under UDCA therapy who reach adulthood with their native liver are at risk to develop severe intrahepatic cholestasis of pregnancy and must not stop UDCA during pregnancy (24, personal communication, Emmanuel Jacquemin, Nathalie Ganne-Carrié, Hôpital Jean Verdier, Bondy, France).

4.1.2.Biliary lipids

In PFIC3 patients, the biliary phospholipid level is dramatically decreased (1-15 % of total biliary lipids; N = 19-24%) despite the presence of normal concentration of bile salts in bile.Such finding is in favour of MDR3 deficiency.Biliary bile salt to phospholipid and cholesterol to phospholipid ratios are approximately 5 fold higher than in wild type bile.The residual percentage of biliary phospholipids seems directly related to the severity of MDR3 mutation and consequently to residual activity of MDR3 P-glycoprotein.Patients with "severe" mutation (nonsense, frameshift) have percentage of biliary phospholipids < 2%, while patients with missense mutations have percentage of biliary phospholipids ³ 2%.In our experience, the threshold that predicts a positive answer to UDCA therapy, is represented by a percentage of biliary hospholipids of 7%.The normal concentration of biliary primary bile salts distinguish PFIC3 patients from those with PFIC1 and PFIC2 (6, 7, 16).

4.1.3. Liver MDR3 immunostaining

Patients with MDR3 mutations have different extent of MDR3 P-glycoprotein canalicular immunostaining.Complete absence of canalicular staining is observed in patients with mutations leading to a truncated protein and in a few patients with missense mutations.A faint or normal MDR3 canalicular staining is only observed in patients with missense mutations (22, 25, 29).This means that a faint canalicular staining also suggests the existence of a MDR3 defect and that a normal canalicular staining does not exclude the presence of a MDR3 dysfunction.The combination of abnormal MDR3 canalicular immunostaining and low percentage of biliary phospholipids is highly suggestive for MDR3 deficiency.

4.1.4.MDR3 mutations

In our experience, MDR3 sequence analysis in 50 PFIC3 patients revealed around 30 different MDR3 mutations (29, personal communication Emmanuel Jacquemin) (Figure 2).Mutations were characterized on both alleles in most cases.In one third of cases, mutations gave rise to a truncated protein.When tested, no MDR3 P-glycoprotein could be detected by immunostaining in the livers of these patients.The absence of MDR3 protein can be explained in two ways.The truncated protein may be broken down very rapidly after synthesis giving rise to extremely low steady state levels of the protein.More likely, the premature stop codon may lead to instability and decay of MDR3 mRNA (45).This latter explanation is supported by the near absence of MDR3 mRNA by northern blotting of livers of several patients (1).The two third remaining patients had missense mutations.Some of them were found in the highly conserved aminoacids sequences of the Walker A and B motifs which are involved in ATP-binding (46).Such aminoacid changes in the Walker A or B motif are generally not compatible with ATPase activity and transport processes (46-49).Other missense mutations, were located in transmembrane domains and near the first Walker motifs.Site-directed mutagenetic analysis of P-glycoprotein has shown that mutations in transmembrane domains are important for substrate specificity and that mutations localized near Walker motifs may disrupt the function of the transporter (46, 50-52).Alternatively, missense mutations might result in intracellular misprocessing of MDR3 as shown for other ABC transporters (53-56).Indeed, such missense mutations were associated with a decreased level of MDR3 canalicular protein (29).Whatever the mechanism involved, the low level of biliary phospholipids found in patients with missense mutations demonstrates the MDR3 functional defect (29).Interestingly, we found evidence that some affected children had a missense mutation representing probably a polymorphism (R652G) (29).It may be that such aminoacid polymorphism has mild consequences, explaining the favourable outcome of patients with this mutation under UDCA therapy, or that in certain circumstances such as pregnancy, it leads to clinical symptoms (18, 26, 57-59).In human liver, this polymorphism is associated with low expression of MDR3.It has been shown for MDR1 P-glycoprotein that aminoacid polymorphism may affect the protein function (60).Ideally, it would be desirable to have a functional mean to distinguish disease mutations from normal variants (49, 56, 61, 62).In very rare patients (< 10%) with a PFIC3 phenotype, only one mutated allele or no mutation was identified.This can be explained by mutations that may map in regulatory sequences of the gene.A gene involved in MDR3 transcription (i.e.FXR) or in protein trafficking could also be involved (63).It is also possible that other genes to be discovered and involved in bile formation may be responsible for PFIC3 phenotype.Furthermore, it may be hypothesized that combined heterozygous mutations for 2 genes (i.e.MDR3 and BSEP) lead to PFIC3 like phenotype.An interesting possibility is also that in heterozygous state, the mutated protein may have a dominant negative effect on MDR3 expression/function (62).Heterozygosity of parents for the MDR3 defects found in affected patients confirmed the recessive inheritance of the disease.This understanding has already allowed prenatal diagnosis (64).Other teams have subsequently reported on identification of MDR3 mutations in PFIC3 patients (23, 27).The use of a resequensing chip dedicated to genetic cholestasis could facilitate identification of MDR3 mutation (65).

4.1.5.Genotype - phenotype correlation

In our experience, compared to children having a MDR3 mutation leading to a truncated protein, children with a MDR3 missense mutation have a less severe disease, with an onset later in life and a slower progression which could be favorably modified by chronic administration of UDCA in about 50% of cases (66).One can hypothesize that these differences are related to a residual transport activity in case of missense mutation.The fact that none of the patients with truncated protein responded to UDCA treatment is in line with this hypothesis (22).Response to UDCA in patients with missense mutation may be the result of residual transport activity, leading to residual phospholipid concentration in bile combined with enrichment of the bile salt pool with UDCA reducing bile salt toxicity below a critical threshold (24, 36).Alternatively, the effect of UDCA might be related to up-regulation of MDR3 P-glycoprotein expression since it has been shown that UDCA up-regulates mdr2 P-glycoprotein expression in primary hepatocytes (67).

4.1.6.Mechanism of liver pathology

Findings in PFIC3 patients confirm the functional homology between the mouse and human genes and further suggests that biliary phospholipid excretion is limited by the amount of mdr2 or MDR3 P-glycoproteins present at the canalicular membrane of the hepatocyte (22, 25, 29, 30, 33, 49, 68).The mechanism of liver damage in PFIC3 patients is likely related to the absence of biliary phospholipids (29).Injury to bile canaliculi and biliary epithelium results probably from continuous exposure to hydrophobic bile salts, the detergent effects of which are no longer countered by phospholipids leading to cholangitis (Figure 3) (26).In addition, the stability of mixed micelles in bile is determined by a three-phase system, in which a proper proportion of bile salts and phospholipids are necessary to maintain solubility of cholesterol.The absence of phospholipids would be expected to destabilize micelles and promote lithogenicity of bile with crystallization of cholesterol, which could favour small bile duct obstruction (Figure 4).These cholangiopathy mechanisms fit well with the histologic findings such as ductular proliferation and ductules containing bile plugs (22, 26, 29).Thus, PFIC3 represents an important example of hepatocellular (canalicular) transport defect that leads to the development of cholangiopathy.While mdr2 (-/-) mice develop sclerosing cholangitis like disease, this type of cholangiopathy has not been reported in PFIC3 patients, and there is no evidence for a role of MDR3 genetic variation in the pathogenesis of primary sclerosing cholangitis (28, 29, 37, 39, 42, 43).

4.2.Intrahepatic cholestasis of pregnancy

Intrahepatic cholestasis of pregnancy (ICP) is characterized by the occurrence of cholestasis during pregnancy in women with an otherwise normal medical history (69).ICP causes fetal distress, spontaneous premature delivery and unexplained third trimester intra uterine death.The classical maternal feature is generalized pruritus, becoming more severe with advancing gestation and abnormal serum liver tests.Maternal serum total bile salt concentration is raised compared to normal pregnancy and this is thought to be due to abnormal biliary transport across the canalicular membrane of hepatocyte.Usually serum GGT activity is within the normal range but in a subgroup of women it is increased (18, 53, 59, 69-76).Liver tests return to normal and pruritus disappears after delivery.Familial cases of ICP have been reported as well as cholestasis induced by oral contraceptive pill in non pregnant women who suffered previously of ICP or belonging to a family with a history of ICP (69, 77).This suggests that a genetic predisposition may exist in some cases of ICP.The link between PFIC3 and ICP has been established when it has been found, within the families of several different children with PFIC3, each child having a distinct nonsense or missense homozygous MDR3 mutation, that heterozygous women had experienced typical recurrent episodes of ICP (22, 29, 59).These familial observations of ICP with slightly elevated serum GGT activity provide arguments for a genetic basis of ICP and may explain the aspect of a dominant transmission trait reported previously (77).It is likely that the heterozygous state for a MDR3 gene defect represents a genetic predisposition in these families, since cholestasis was not present in every pregnancy in these women (18, 59).Associated non genetic factors, such as female sex hormones and metabolites, could modify MDR3 heterozygous state expressivity by decreasing normal allele expression (59).Indeed, a sterol responsive element exists in the mdr2 gene promotor and could be involved in transcriptional control of mdr2 expression (63, 67, 78, 79).Alternatively, hormones could directly interact with MDR3 P-glycoprotein impairing its function (59, 80).Such events could favour the transient decompensation of the heterozygous state for a MDR3 gene defect during pregnancy leading to ICP (18, 53, 59, 69-75b).As for PFIC3, cholestasis would result from the toxicity of bile in which detergent bile salts are not inactivated by phospholipids.While heterozygous mdr2 (+/-) mice, with a maximal phospholipid secretion of 60% of controls do not develop liver disease, the appearance of liver injury in a heterozygous patient could be expected because in humans the bile salt pool is much more hydrophobic than in mice (30, 33, 35, 40, 81).This justifies to search for a MDR3 gene mutation in ICP, particularly if serum GGT activity is high.Since the initial publication, numerous cases of ICP linked to a heterozygous MDR3 defect, mostly with high serum GGT activity, have been reported in women with no known family history of PFIC, confirming that a MDR3 defect represents a genetic predisposition to develop ICP (18, 53, 59, 69-75b).Furthermore, benign recurrent intrahepatic cholestasis (BRIC) and ICP have been reported in a single kindred suggesting that both cholestatic syndromes may be inter-related (82).This observation is very interesting because if usually in BRIC serum GGT activity is within the normal range, in a subgroup of women it is increased (83).All in all these findings suggest that in case of BRIC and/or ICP episodes with high serum GGT activity, a MDR3 defect should be considered (18, 53, 59, 70-75b).According to current classification of diseases related to FIC1 or BSEP genes, ICP and BRIC related to MDR3 deficiency should be reported under the terms ICP3 and BRIC3 (9, 11).It goes without saying that women who develop ICP due MDR3 deficiency should received UDCA therapy during the period of ICP, since UDCA has been proven efficient to reduce maternal and fetal complications of ICP (84, 85).

4.3.Cholesterol gallstone disease

Intrahepatic biliary lithiasis or gallbladder lithiasis have been found in children with PFIC3 and nonsense or missense MDR3 mutations (29).Gallstones were also identified in some of their parents.Biliary lithiasis could be related to an increased biliary cholesterol to phospholipid ratio and suggests that, also in man, cholesterol and phospholipid biliary secretion can be partly uncoupled.This observation provides a mechanistic basis for gallstone formation (Figure 4).The absence or low level of phospholipids in bile would be expected to destabilize micelles and promote lithogenicity of bile with crystallization of cholesterol (81).This is very likely since cholesterol saturation index is abnormally increased in bile of PFIC3 patients (personal communication, Emmanuel Jacquemin).Indeed, many studies have reported ethnic and familial clusters of cholesterol gallstones suggesting genetic predisposition.Several studies performed in adult patients with symptomatic intrahepatic sludge, biliary microlithiasis and/or gallbladder cholesterol stones, have shown further evidence that MDR3 mutations represent a genetic predisposition for cholesterol gallstone disease.This entity is reported under the term "Low Phospholipid Associated Cholelithiasis (LPAC)" syndrome (86-89).Mutations are heterozygous frame-shift, non sense or missense mutations or homozygous missense mutations and are mainly localized in important presumed protein domains.Also in these studies were biliary cholesterol to phospholipid ratio and cholesterol saturation index abnormally elevated.Of importance was also the observation that the symptomatology (mild cholestasis, biliary pain, pancreatitis or cholangitis) recurred after cholecystectomy and was dramatically enhanced during pregnancy or after starting oral contraception.Again, it is likely that the genetic predisposition represented by the MDR3 defect is decompensated by female sex hormones.Further recurrence is prevented by long term oral administration of UDCA.Cholecystectomy is only indicated in the case of symptomatic gallstones but not when only sludge is present in the gallbladder, because it usually disappears under UDCA therapy.Biliary drainage or partial hepatectomy may be indicated in case of symptomatic non cystic intrahepatic bile duct dilatations filled with gallstones (89).

4.4.Drug induced cholestasis

Several lines of evidence suggest that MDR3 deficiency could be involved in drug induced cholestasis (i.e., oral contraceptive pill induced cholestasis).First, MDR3 deficiency predisposes to ICP (see above).Secondly, cases of cholestasis induced by oral contraceptive pill in non pregnant women who suffered previously of ICP or belonging to a family with a history of ICP have been reported (77, 82).Third, in patients with cholesterol gallstone disease due to MDR3 deficiency, symptomatology was dramatically enhanced during pregnancy or after starting oral contraception (86, 87, 89).Fourth, a case of asymptomatic cirrhosis due to MDR3 deficiency in a young adult woman has been revealed after starting oral contraceptive pills (24).As we have already discussed in this review the genetic predisposition represented by nonsense or missense (including aminoacid polymorphism) MDR3 mutations could manifest under the pressure of xenobiotic intake (26, 57).A MDR3 defect should be searched in women who experienced cholestasis under oral contraception, especially if cholestasis is characterized by high serum GGT activity.It is postulated that xenobiotics that inhibit ABC proteins, and more specifically MDR P-glycoprotein, could induce cholestasis in predisposed patients with MDR3 deficiency (80).Recent data have clearly shown that some MDR3 mutations are associated to the occurrence of drug induced cholestasis and therefore MDR3 deficiency represents very likely a genetic predisposition to develop drug induced cholestasis (58).

4.5.Transient neonatal cholestasis

This spontaneous resolving form of neonatal cholestasis, results from the association of several factors, including immaturity of bile secretion and perinatal disease leading to hepatic hypoxia or ischemia (90).In 10% of the children with transient neonatal cholestasis no remarkable events are identified during the neonatal period.It is obvious that transient neonatal cholestasis preferentially appears in children who have had perinatal suffering but it could also develop in predisposed children in absence of perinatal events (90).The predisposition could be represented by a heterozygous genetic defect in any hepatocellular canalicular ATP dependent transport system involved in bile formation (8, 11, 26).This could favour the transient decompensation of bile secretion processes which are underdeveloped in neonates (90).Since most of the children with transient neonatal cholestasis have high serum GGT activity when cholestatic, MDR3 gene could be involved.Obviously, this speculative and provocative hypothesis needs to be proved in a large cohort of children, but we have already some evidence for it.Indeed, we have found (personal communication, Etienne Sokal, Cliniques St Luc, Brussels, Belgium, and Emmanuel Jacquemin) heterozygous MDR3 mutation in 3 children, without perinatal suffering, who had neonatal cholestasis with features compatible with PFIC3 (26, 29, 64, 91).In one of these children, the MDR3 P-glycoprotein function defect has been documented by low percentage of biliary phospholid and normal biliary bile salt level (29).In all children, clinical condition and liver tests normalized under UDCA therapy and remained normal after stopping UDCA.This evolution was similar to the one of transient neonatal cholestasis (90).These observations suggest that a heterozygous MDR3 missense mutation may be involved in some children with transient neonatal cholestasis.Furthermore, it is known that transient neonatal cholestasis does not develop in all neonates with perinatal distress.In this view, it might be that those neonates who develop transient neonatal cholestasis after perinatal suffering are in fact genetically predisposed (92).A genetic predisposition could precipitate the consequences of hypoxia-ischemia on ATP dependent canalicular mechanisms involved in bile secretion, leading to transient neonatal cholestasis (90).

4.6.Adult " idiopathic " biliary cirrhosis

A MDR3 defect was found in several young adults in whom a diagnosis of cirrhosis was made between the age of 13.5 and 20.5 years (24, 29).All had a missense mutation, have responded to UDCA therapy and are still alive with their native liver and under UDCA therapy at a mean age of 28 years (24, 29).Genotype-phenotype analysis has shown that MDR3 gene mutation that leads to complete absence of function is lethal within the first decade of life unless liver transplantation is performed (29).By contrast, certain missense mutations that lead to residual activity (i.e., as expected in patients mentioned above) are probably spontaneously compatible with life until adolescence or early adulthood.For this reason, we believe that some young adults, or even older, with unexplained " idiopathic " cirrhosis of biliary type may have MDR3 deficiency (26a, 26b).Indeed, a MDR3 mutation has been detected in a woman who developed cholelithiasis in adolescence, followed by cholestasis of pregnancy and finally adulthood biliary cirrhosis at 47 years (93).

5.GENOTYPE - PHENOTYPE CORRELATION AMONG LIVER DISEASES RELATED TO MDR3 DEFICIENCY

Schematically, tentative and coarse correlations between genotype, taking into account mutated allele number and mutation class, and initial phenotype can be established.Patients with PFIC3 and/or young adult biliary cirrhosis harbor biallelic mutations, compound heterozygous or homozygous, that are thought to have a serious deleterious effect on protein function in most cases (23, 24, 27, 29, 64).Other phenotypes mainly harbor heterozygous mutations, "severe" or "soft", or "soft" biallelic missense mutations (18, 24, 26, 29, 53, 58, 59, 64, 70-75b, 86-89, 93).Meanwhile, it is obvious that such correlations are difficult and fragile because a patient may develop several phenotypes during life (24, 93).

6.TREATMENT OF MDR3 DEFICIENCY (MAINLY PFIC3): PRESENT AND FUTURE

In our experience, oral administration of UDCA represents an alternative to liver transplantation in some children with PFIC3 (26, 29, 66).Indeed, around 30% and 15% of PFIC3 patients normalized or improved their liver tests under UDCA, respectively.All the 11 children previously reported and who responded to UDCA therapy are still alive with their native liver and under UDCA treatment, at ages ranging from 13 to 33 years (29).In patients with PFIC3 as in other types of cholestasis, the beneficial effect of UDCA may also be related to the modulation of biliary bile acid composition in favour of hydrophilic bile acids which might diminish cellular injury (29, 66).In the mdr2 (-/-) mouse model, feeding the noncytotoxic bile salt UDCA led to a complete replacement of the endogenous bile salt pool with UDCA and this halted the progression of the liver disease (36).It has been shown that nonresponders have a complete defect in phospholipid secretion (i.e., nonsense MDR3 mutation) and it is likely that partial UDCA replacement is insufficient to reduce the increased bile salt toxicity in phospholipid-free bile of these patients (22, 29). Patients who do respond to UDCA therapy have a partial defect (e.g., missense MDR3 mutation) and the residual phospholipid concentration in bile (threshold of 7% of total biliary lipids), combined with a partial UDCA replacement, may be sufficient to reduce the bile salt toxicity below a critical threshold (29).These correlations should help to select those PFIC3 patients who could benefit from UDCA therapy.It is our policy to consider UDCA in the initial therapeutic management of children with PFIC3.It could prevent evolution towards cirrhosis and therefore avoid, at least in the mean term, the need for liver transplantation in some children (29, 66).In case of failure with this therapy, liver transplantation still represents the treatment of choice.Furthermore, UDCA has been shown to be effective in the treatment of ICP and cholesterol gallstone disease (84-87, 89).In these conditions, UDCA should be the treatment of choice, especially if the disease causing mechanism is MDR3 deficiency.

New therapeutic options for MDR3 deficiency related diseases deserve also consideration and will certainly be tested in the future (Figure 5).First, a targeted pharmacological approach intended to induce MDR3 function is conceivable.Indeed several drugs (UDCA, statins, fibrates) are already known to up-regulate mdr2 P-glycoprotein expression in rat liver (67,78, 94, 95).MDR3 transcription seems mainly regulated by the nuclear receptor FXR (28, 63, 79, 96, 97).Drugs that could increase MDR3 expression via FXR will be good candidates for PFIC3 therapy (97-100).It is possible that the beneficial effect of UDCA in PFIC3 patients is in part related to up-regulation of MDR3 P-glycoprotein expression (97, 98).It is hypothesized that patients who have a partial MDR3 defect with residual activity (i.e., missense MDR3 mutation) might benefit from such treatments in the future (26, 28).A second pharmacological approach intended to overcome the arrest of translation induced by specific nonsense mutations (i.e., using aminoglycosides) or to retarget at the canalicular membrane a misfolded mutated protein blocked within the hepatocyte (i.e., using chaperone drugs) should also be considered (55, 56, 101-103).A third pharmacological approach, could be to induce another canalicular protein (i.e.MDR1 P-glycoprotein) that could functionally complement MDR3 P-glycoprotein (26).Experimental data exist that show that there is a kind of overlap substrate specificity between MDR1 and MDR3 proteins (80, 104).Such approach has already been tested fortuitously with success in one patient with cystic fibrosis.Indeed, a preliminary report has suggested that MDR1 could functionally complement CFTR (105, 106).While shown efficient in mdr2 (-/-) mice, the effect of a lecithin rich diet has not been tested in children with PFIC3 (107).Finally, cell and gene therapies will certainly be considered in the future.An elegant study performed in mdr2 (-/-) mice has shown that the liver disease due to mdr2 deficiency was corrected by transplantation of mdr2 (+/+) or MDR3 transgenic hepatocytes (108).It is expected that in PFIC3 normal transplanted human hepatocytes will profit from liver regeneration induced by the disease and that their proliferation will be selectively favoured by the fact that they protect themselves from bile salts.Consequently repopulation of the native liver by MDR3 expressing hepatocytes should be possible.This selective advantage should also apply to a gene therapy approach for PFIC3.When cell or gene therapy will be reasonably feasible in this liver disease, it will be probably necessary to propose it before the presence of severe liver fibrosis which might hamper liver repopulation by normal or genetically modified hepatocytes.In this view, it will be a new challenge for pediatricians and geneticians, who will have in a short time period, first, to identify children with PFIC3 and, second, to confirm and characterize (i.e.at the functional level) the MDR3 gene defect (26).Waiting for the molecular diagnosis of MDR3 deficiency, therapy with UDCA (or other drugs in the future) should be initiated in order to prevent liver damage.In case of mutation leading to protein truncation, no answer to UDCA is expected.In this situation, cell or gene therapy could be proposed.In presence of a mutation leading to residual activity (or to protein mistargeting) a positive response to UDCA (or other drugs) could be expected.In this situation the choice of therapeutic options between the pursuit of UDCA (or other drugs) and the decision to start cell/gene therapy, or a combination of drugs and cell/gene therapy, will need to be defined.The fact that liver transplantation will still be a rescue option in case of failure of these new treatments should facilitate their application to patients with PFIC3 and MDR3 deficiency.

7. REFERENCES

1.Deleuze JF, Jacquemin E, Dubuisson C, Cresteil D, Dumont M, Erlinger S, Bernard O, hadchouel M.Defect of multidrug-resistance 3 gene expression in a subtype of progressive familial intrahepatic cholestasis.Hepatology 23, 904-908 (1996)
http://dx.doi.org/10.1002/hep.510230435
PMid:8666348

2.Clayton RJ, Iber FL, Ruebner BH, Mc Kusick VA.Byler disease: fatal familial intrahepatic cholestasis in an Amish kindred.Am J Dis Child 117, 112-124 (1969)

3.Linarelli LG, Williams CN, Phillips MJ.Byler's disease: fatal intrahepatic cholestasis.J Pediatr 81, 484-492 (1972)
http://dx.doi.org/10.1016/S0022-3476(72)80174-4
PMid:5049817

4.Whitington PF, Freese DK, Alonso EM, Schwarzenberg SJ, Sharp HL.Clinical and biochemical findings in progressive familial intrahepatic cholestasis.J Pediatr Gastroenterol Nutr 18, 134-141 (1994)
http://dx.doi.org/10.1097/00005176-199402000-00003
PMid:7912266

5.Williams CN, Kaye R, Baker L, Hurwitz R, Senior JR.Progressive familial cholestatic cirrhosis and bile acid metabolism.J Pediatr 81, 493-500 (1972)
http://dx.doi.org/10.1016/S0022-3476(72)80175-6
PMid:5049818

6.Jacquemin E, Dumont M, Bernard O, Erlinger S, Hadchouel M.Evidence for defective primary bile acid secretion in children with progressive familial intrahepatic cholestasis (Byler disease).Eur J Pediatr 153, 424-428 (1994)
http://dx.doi.org/10.1007/BF01983406
PMid:8088298

7.Bull LN, Carlton VE, Stricker NL, Baharloo S, DeYoung JA, Freimer NB, Magid MS, Kahn E, Markowitz J, DiCarlo FJ, McLoughlin L, Boyle JT, Dahms BB, Faught pr, Fitzgerald JF, Piccoli DA, Witzleben CL, O'Connell NC, Setchell KD, Agostini Jr RM, Kocoshis SA, Reyes J, Knisely AS.Genetic and morphological findings in progressive familial intrahepatic cholestasis (Byler disease (PFIC-1) and Byler syndrome): Evidence for heterogeneity.Hepatology 26, 155-164 (1997)
http://dx.doi.org/10.1002/hep.510260121
PMid:9214465

8.Trauner M, Meier PJ, Boyer JL.Molecular pathogenesis of cholestasis.N Engl J Med 339, 1217-1227 (1998)
http://dx.doi.org/10.1056/NEJM199810223391707
PMid:9780343

9.Bezerra JA, Balistreri WF: Intrahepatic cholestasis: order out of chaos.Gastroenterology 117, 1496-1498 (1999)
http://dx.doi.org/10.1016/S0016-5085(99)70302-1
PMid:10579993

10.Jacquemin E, Hadchouel M: Genetic basis of progressive familial intrahepatic cholestasis.J Hepatol 31, 377-381 (1999)
http://dx.doi.org/10.1016/S0168-8278(99)80240-2
PMid:10453956

11.Pauli-Magnus C, Stieger B, Meier Y, Kullak-Ulbick GA, Meier PJ.Enterohepatic transport of bile salts and genetics of cholestasis.J Hepatol 43, 342-357 (2005)
http://dx.doi.org/10.1016/j.jhep.2005.03.017
PMid:15975683

12.Baussan C, Cresteil D, Gonzales E, Raynaud N, Dumont M, Bernard O, Hadchouel M, Jacquemin E.Genetic cholestatic liver diseases : The exemple of progressive familial intrahepatic cholestasis and related disorders.Acta Gastro-Enterologica Belgica LXVII, 179-183 (2004)

13.Bull LN, van Eijk MJT, Pawlikowska L, DeYoung JA, Juijn JA, Liao M, Klomp LWJ, Lomri N, Berger R, Scharschmidt BR, Knisely AS, Houwen RHJ, Freimer NB.A gene encoding a P-type ATPase mutated in two forms of hereditary cholestasis.Nat Genet 18, 219-224 (1998)
http://dx.doi.org/10.1038/ng0398-219
PMid:9500542

14.Klomp LWJ, Vargas JC, Van Mil SWC, Pawlikowska L, Strautnieks SS, Van Eijk MJT, Juijn JA, Pabón-Peña C, Smith LB, DeYoung JA, Byrne JA, Gombert J, Van Der Brugge G, Berger R, Jankowska J, Villa E, Knisely AS, Thompson RJ, Freimer NB, Houwen RHJ, Bull LN.Characterization of mutations in ATP8B1 Associated with hereditary cholestasis.Hepatology 40, 27-38 (2004)
http://dx.doi.org/10.1002/hep.20285
PMid:15239083

15.Strautnieks SS, Bull LN, Knisely AS, Kocoshis SA, Dahl N, Arnell H, Sokal E, Dahan K, Childs S, Ling V, Tanner MS, Kagalwalla AF, Németh A, Pawlowska J, Baker A, Mieli-Vergani G, Freimer NB, Gardiner RM, Thompson RJ.A gene encoding a liver-specific ABC transporter is mutated in progressive familial intrahepatic cholestasis.Nat Genet 20, 233-238 (1998)
http://dx.doi.org/10.1038/3034
PMid:9806540

16.Jansen PLM, Strautnieks SS, Jacquemin E, Hadchouel M, Sokal EM, Hooiveld GJEJ, Koning JH, de Jager-Krikken A, Kuipers F, Stellard F, Bijleveld CMA, Gouw A, van Goor H, Thompson RJ, Muller M.Hepatocanalicular bile salt export pump deficiency in patients with progressive familial intrahepatic cholestasis.Gastroenterology 117, 1370-1379 (1999)
http://dx.doi.org/10.1016/S0016-5085(99)70287-8
PMid:10579978

17.Van Mil SWC, Van Der Woerd WL, Van Der Brugge G, Strum E, Jansen PLM, Bull LN, Van Den Berg IET, Berger R, Houwen RHJ, Klomp LWJ.Benign recurrent intrahepatic cholestasis type 2 is caused by mutations in ABCB11.Gastroenterology 127, 379-384 (2004)
http://dx.doi.org/10.1053/j.gastro.2004.04.065
PMid:15300568

18.Pauli-Magnus C, Lang T, Meier Y, Zodan-Marin T, Jung D, Breymann C, Zimmermann R, Kenngott S, Beuers U, Reichel C, Kerb R, Penger A, Meier PJ, Kullak-Ulbick GA.Sequence analysis of bile salt export pump (ABCB11) and multidrug resistance p-glycoprotein 3 (ABCB4, MDR3) in patients with intrahepatic cholestasis of pregnancy.Pharmacogenetics 14, 91-102 (2004)
http://dx.doi.org/10.1097/00008571-200402000-00003
PMid:15077010

19.Maggiore G, Bernard O, Riely C, Hadchouel M, Lemonnier A, Alagille D.Normal serum gamma-glutamyl-transpeptidase activity identifies groups of infants with idiopathic cholestasis with poor prognosis.J Pediatr 111, 251-252 (1987)
http://dx.doi.org/10.1016/S0022-3476(87)80079-3
PMid:2886574

20.Maggiore G, Bernard O, Hadchouel M, Lemonnier A, Alagille D.Diagnostic value of serum g-glutamyl transpeptidase activity in liver diseases in children.J Pediatr Gastroenterol Nutr 12, 21-26 (1991)
http://dx.doi.org/10.1097/00005176-199101000-00005
PMid:1676410

21.Alonso EM, Snover DC, Montag A, Freese DK, Whitington PF.Histologic pathology of the liver in progressive familial intrahepatic cholestasis.J Pediatr Gastroenterol Nutr 18, 128-133 (1994)
http://dx.doi.org/10.1097/00005176-199402000-00002
PMid:8014759

22.de Vree JML, Jacquemin E, Sturm E, Cresteil D, Bosma PJ, Aten J, Deleuze JF, Desrochers M, Burdelski M, Bernard O, Oude Elferink RPJ, Hadchouel M.Mutations in the MDR3 gene cause progressive familial intrahepatic cholestasis.Proc Natl Acad Sci USA 95, 282-287 (1998)
http://dx.doi.org/10.1073/pnas.95.1.282

23.Degiorgio D, Colombo C, Seia M, Porcaro L, Costantino L, Zazzeron L, Bordo D, Coviello DA.Molecular characterization and structural implications of 25 new ABCB4 mutations in progressive familial intrahepatic cholestasis type 3 (PFIC3).Eur J Hum Genet, e29 (2007)

24.Ganne-Carrié N, Baussan C, Grando V, Gaudelus J, Cresteil D, Jacquemin E.Progressive familial inrahepatic cholestasis type 3 revealed by oral contraceptive pills.J Hepatol 38, 693-694 (2003)
http://dx.doi.org/10.1016/S0168-8278(03)00049-7
PMid:12713886

25.Keitel V, Burdelski M, Warskulat U, Kühlkamp T, Keppler D, Häussinger D, Kubitz R.Expression and localization of hepatobiliary transport proteins in progressive familial intrahepatic cholestasis.Hepatology 41, 1160-1172 (2005)
http://dx.doi.org/10.1002/hep.20682
PMid:15841457

26a.Jacquemin E.Role of multidrug resistance 3 deficiency in pediatric an adult liver disease : One gene for three diseases.Semin Liver Dis 21, 551-562 (2001)
http://dx.doi.org/10.1055/s-2001-19033
PMid:11745043

26b. Ziol M, Barbu V, Rosmorduc O, Frassati-Biaggi A, Barget N, Hermelin B, Scheffer GL, Bennouna S, Trinchet JC, Beaugrand M, Ganne-Carrié N. ABCB4 heterozygous gene mutations associated with fibrosing cholestatic liver disease in adults. Gastroenterology.135,131-141 (2008)
http://dx.doi.org/10.1053/j.gastro.2008.03.044
PMid:18482588

27.Chen HL, Chang PS, Hsu HC, Lee JH, Ni YH, Hsu HY, Jeng YM, Chang MH.Progressive familial intrahepatic cholestasis with high γ-glutamyltranspeptidase levels in Taiwanese infants : Role of MDR3 gene defect ? Pediatr Res 50, 50-55 (2001)
http://dx.doi.org/10.1203/00006450-200107000-00011
PMid:11420418

28.Trauner M, Fickert P, Wagner M.MDR3 (ABCB4) defects: a paradigm for the genetics of adult cholestatic syndromes.Semin Liver Dis 27, 77-98 (2007)
http://dx.doi.org/10.1055/s-2006-960172

29.Jacquemin E, de Vree JLM, Cresteil D, Sokal EM, Sturm E, Dumont M, Scheffer GL, Paul M, Burdelski M, PJ Bosma, Bernard O, Michelle Hadchouel M, Oude Elferink RPJ.The wide spectrum of multidrug resistance 3 deficiency : from neonatal cholestasis to cirrhosis of adulthood.Gastroenterology 120, 1448-1458 (2001)
http://dx.doi.org/10.1053/gast.2001.23984
PMid:11313315

30.Smit JJ, Shinkel AH, Oude Elferink RP, Groen AK, Wagennar E, van Deemter L, Mol CA, Ottenhoff R, van der Lugt NM, van Room MA.Homozygous disruption of the murine mdr2 P-glycoprotein gene leads to a complete absence of phospholipid from bile and to liver disease.Cell 75, 451-462 (1993)
http://dx.doi.org/10.1016/0092-8674(93)90380-9
PMid:8106172

31.Reutz S, Gros P.Phosphatidylcholine translocase: a physiological role for the mdr2 gene.Cell 77, 1071-1081 (1994)
http://dx.doi.org/10.1016/0092-8674(94)90446-4
PMid:7912658

32.Smit JJ, Shinkel AH, Mol CA, Majoor D, Mooi WJ, Jongsma AP, Lincke CR, Borst P.Tissue distribution of the human MDR3 P-glycoprotein.Lab Invest 71, 638-649 (1994)

33.Smith AJ, de Vree JML, Ottenhoff R, Elferink RP, Schinkel AH, Borst P.Hepatocyte-specific expression of the human MDR3 P-glycoprotein gene restores the biliary phosphatidycholine excretion absent in mdr2 (-/-) mice.Hepatology 28, 530-536 (1998)
http://dx.doi.org/10.1002/hep.510280234
PMid:9696021

34.Oude Elferink RPJ, Paulusma CC.Function and pathophysiological importance of ABCB4 (MDR3 P-glycoprotein).Pflugers Arch.453, 601-10 (2007)
http://dx.doi.org/10.1007/s00424-006-0062-9
PMid:16622704

35.Mauad TH, van Nieuwkerk CM, Dingemans KP, Smit JJ, Schinkel AH, Notenboom RG, van den Bergh Weerman MA, Verkruisen RP, Groen AK, Oude Elferink RPJ, van der Valk MA, Borst P, Offerhaus GJ.Mice with homozygous disruption of the mdr2 P-glycoprotein gene.A novel animal model for studies of nonsuppurative inflammatory cholangitis and hepatocarcinogenesis.Am J Pathol 145, 1237-1245 (1994)

36.van Nieuwkerk CJM, Oude Elferink RPJ, Groen AK, Ottenhoff R, Tytgat GN, Dingemans KP, Van Den Bergh Weerman MA, and Offerhaus GJ.Effects of ursodeoxycholate and cholate feeding on liver disease in FVB mice with a disrupted mdr2 p-glycoprotein gene.Gastroenterology 111, 165-171 (1996)
http://dx.doi.org/10.1053/gast.1996.v111.pm8698195
PMid:8698195

37.Fickert P, Fuschsbichler A, Wagner M, Zollner G, Kaser A, Tilg H, Krause R, Lammert F, Langner C, Zatloukal K, Marschall HU, Denk H, Trauner M.Regurgitation of bile acids from leaky bile ducts causes sclerosing cholangites in MDR2 (ABCB4) knockout mice.Gastroenterology 127, 261-274 (2004)
http://dx.doi.org/10.1053/j.gastro.2004.04.009
PMid:15236191

38.Fickert P, Zollner G, Fuchsbichler A, Stumptner C, Weiglein AH, Lammert F, Marschall HU, Tsybrovskyy O, Zatloukal K, Denk H, Trauner M.Ursodesoycholic acid aggravates bile infarcts in bile duct-ligated and MDR2 knockout mice via disruption of cholangioles.Gastroenterology 123, 1238-1251 (2002)
http://dx.doi.org/10.1053/gast.2002.35948
PMid:12360485

39.Pauli-Magnus C, Reinhold K, Fattinger K, Lang T, Anwald B, Kullac-Ulbick GA, Beuers U, Meier PJ.BSEP and MDR3 haplotype structure in healthy Caucasians, primary biliary cirrhosis and primary sclerosing cholangitis.Hepatology 39, 779-791 (2004)
http://dx.doi.org/10.1002/hep.20159
PMid:14999697

40.Hoekstra H, Porte RJ, Tian Y, Jochum W, Stieger B, Moritz W, Slooff MJH, Graf R, Clavien PA.Bile salt toxicity aggravates cold ischemic injury of bile ducts after liver transplantation in MDR2 +/- mice.Hepatology 43, 1022-1031 (2006)
http://dx.doi.org/10.1002/hep.21169
PMid:16628673

41.Strautnieks S, Lopes A, Underhill J, Gerred S, Nibbering K, Knisely A, Portman B, Bomford A, Heaton N, Mieli-Vergani G, Oude Elferink R, O'Grady J, Wali S, Thompson R.Critical residues in the multidrug resistance 3 protein gene associated with adult onset of cholangiopathy and intrahepatic cholestasis of pregnancy.Hepatology 36, 415A (2002)

42.Rosmorduc O, Hermelin B, Boelle PY, Poupon RE, Poupon R, Chazouilleres O.ABCB4 gene mutations and primary sclerosing cholangitis.Gastroenterology 126, 1220-1222 (2004)
http://dx.doi.org/10.1053/j.gastro.2004.02.042
PMid:15057773

43.Trauner M, Fickert P, Lammert F, Marschall HU.Correspondence.Gastroenterology 126, 1222-1223 (2004)
http://dx.doi.org/10.1053/j.gastro.2004.02.043

44.Elferink RP, Ottenhoff R, van Marle J, Frijters CM, Smith AJ, Groen KA.Class III P-glycoproteins mediate the formation of lipoprotein X in the mouse.J Clin Invest 102, 1749-1757 (1998)
http://dx.doi.org/10.1172/JCI3597
PMid:9802889 PMCid:509123

45.Hamosh A, Rosenstein BJ, Cutting GR.CFTR nonsense mutations G542X and W1282X associated with severe reduction of CFTR mRNA in nasal epithelial cells.Hum Mol Genet 1, 542-544 (1992)
http://dx.doi.org/10.1093/hmg/1.7.542
PMid:1284888

46.Gottesman MM, Hrycyna CA, Schoenlein PV, Germann UA, Pastan I.Genetic analysis of the multidrug transporter.Annu Rev Genet 29, 607-649 (1995)
http://dx.doi.org/10.1146/annurev.ge.29.120195.003135
PMid:8825488

47.Beaudet L, Urbatsch IL, Gros P.Mutations in the nucleotide-binding sites of P-glycoprotein that affect substrate specificity modulate substrate-induced adenosine triphosphatase activity.Biochemistry 37, 9073-9082 (1998)
http://dx.doi.org/10.1021/bi972656j
PMid:9636053

48.Urbatsch IL, Beaudet L, Carrier I, Gros P.Mutations in either nucleotide-binding site of P-glycoprotein (Mdr3) prevent vanadate trapping of nucleotide at both sites.Biochemistry 37, 4592-4602 (1998)
http://dx.doi.org/10.1021/bi9728001
PMid:9521779

49.Morita SY, Kobayashi A, Takanezawa Y, Kioka N, Handa T, Arai H, Matsuo M, Ueda K.Bile salt-dependant efflux of cellular phospholipids mediated by ATP binding cassette protein B4.Hepatology 46, 188-199 (2007)
http://dx.doi.org/10.1002/hep.21591
PMid:17523162

50.Loo TW, Clarke DM.Mutations to amino acids located in predicted transmembrane segment 6 (TM6) modulate the activity and substrate specificity of human P-glycoprotein.Biochemistry 33, 14049-14057 (1994)
http://dx.doi.org/10.1021/bi00251a013
PMid:7947814

101.Wilschanski MN, Yahav Y, Yaccov Y, Blau H, Bentur L, Rivlin J, Aviram M, Bdolah-Abram T, Bebok Z, Shushi L, Kerem B, Kerem E.Gentamicin-induced correction of CFTR function in patients with cystic fibrosis and CFTR stop mutations.N Engl J Med 349, 1433-1441 (2003)
http://dx.doi.org/10.1056/NEJMoa022170
PMid:14534336

102.Egan ME, Pearson M, Weiner SA, Rajendran V, Rubin D, Glockner-Pagel J, Canny S, Du K, Lukacs GL, Caplan MJ.Cucurmin, a major constituent of turmeric corrects cystic fibrosis defects.Science 304, 600-602 (2004)
http://dx.doi.org/10.1126/science.1093941
PMid:15105504

103.Ananthanarayanan M.Therapy for hepatocyte transporter trafficking mutations: the time is now.Hepatology 45, 1340-1342 (2007)
http://dx.doi.org/10.1002/hep.21780
PMid:17538962

104.van Helvoort A, Smith JA, Sprong H, Fritzsche I, Schinkel AH, Piet Borst P, van Meer G.MDR1 P-glycoprotein is a lipid translocase of broad specificity, while MDR3 P-glycoprotein specifically translocates phosphatidylcholine.Cell 87, 507-517 (1996)
http://dx.doi.org/10.1016/S0092-8674(00)81370-7
PMid:8898203

105.Lallemand JY, Stoven V, Annereau JP, Boucher J, Blanquet S, Barthe J, Lenoir G.Induction by antitumoral drugs of proteins that functionally complement CFTR : a novel therapy for cystic fibrosis ? Lancet 350, 711-712 (1997)
http://dx.doi.org/10.1016/S0140-6736(05)63510-6

106.Altschuler EL.Azithromycin, the multidrug-resistant protein and cystic fibrosis.Lancet 351, 1286 (1998)
http://dx.doi.org/10.1016/S0140-6736(05)79350-8

107.LamireauT, Bouchard G, Yousef IM, Clouzeau-Girard H, Rosenbaum J, Desmouliere A, Tuchweber B.Dietary lecithin protects against cholestatics liver disease in cholic acid-fed ABCB4-deficient mice.Pediatr Res 6, 185-190 (2007)

108.de Vree JML, Ottenhoff R, Bosma PJ, Smith AJ, Aten J, Oude Elferink RPJ.Correction of liver disease by hepatocyte transplantation in a mouse model of progressive familial intrahepatic cholestasis.Gastroenterology 119, 1720-1730 (2000)
http://dx.doi.org/10.1053/gast.2000.20222
PMid:11113093

Abbreviations: mdr2, multidrug resistance 2; MDR3, multidrug resistance 3; PFIC, progressive familial intrahepatic cholestasis ; BSEP, bile salt export pump ; GGT, gamma-glutamyltransferase; UDCA, ursodeoxycholic acid ; BRIC, benign recurrent intrahepatic cholestasis ; ICP, intrahepatic cholestasis of pregnancy ; TNC, transient neonatal cholestasis ; LPAC, low phospholipid associated cholelithiasis ; DIC, drug induced cholestasis

Key Words: Multidrug resistance 3, ABCB4, Progressive Familial Intrahepatic Cholestasis Type 3, Cholesterol Gallstone, Cholestasis Of Pregnancy, Biliary Cirrhosis, LPAC, Transient Neonatal Cholestasis, Drug Induced Cholestasis, Ursodeoxycholic Acid, Review

Send correspondence to: Emmanuel Jacquemin, Service d'Hepatologie Pediatrique, Departement de Pediatrie, Centre Hospitalier Universitaire de Bicetre, 78 rue du General Leclerc, 94275 Le Kremlin-Bicetre Cedex, France, Tel: 33-1-45-21-31-64, Fax: 33-1-45-21-28-16, E-mail: emmanuel.jacquemin@bct.aphp.fr