Portal de Periódicos da CAPES

Genetic Cholestasis Syndromes

1999; Lippincott Williams & Wilkins; Volume: 28; Issue: 2 Linguagem: Inglês

10.1097/00005176-199902000-00005

1536-4801

Benjamin L. Shneider,

Liver Diseases and Immunity

It has been recognized for many years that some forms of cholestatic liver disease have an increased prevalence in selected families and are apparently inherited diseases. In addition, many children without similarly affected family members have chronic cholestatic diseases, the exact cause of which defies characterization. These children may also have specific genetic abnormalities as the basis for their condition. As an aggregate these "genetic" cholestasis syndromes are not uncommon in pediatric hepatology and cause significant morbidity and mortality. The approach to the analysis of children with genetic cholestasis syndromes has included careful clinical characterization, gene-linkage analysis, and molecular investigations of the basic mechanisms involved in bile formation. This review summarizes many of the recent breakthroughs that have permitted the beginning of a new chapter in the investigation of genetic cholestasis. Recognition in the late 1950s and 1960s of familial cases of cholestatic liver disease was one of the first clues that there might be a primary genetic basis for these diseases (1-8). New descriptions of familial cholestasis continue to be reported into the present era (9-11). One of the critical issues in analyzing children with familial cholestasis is whether the clinical heterogeneity of their liver diseases and related morbidity indicates multiple causes or varying expression of a complex gene. Clinicians have developed a complex and confusing terminology that is associated with equally complicated associated clinical criteria. This has resulted in a bewildering array (Table 1) of diagnostic terms (syndromic bile duct paucity, nonsyndromic bile duct paucity, Alagille syndrome, Byler's disease, Byler's syndrome, progressive familial intrahepatic cholestasis, intrahepatic biliary atresia, Aagenes syndrome, and others [12]). These diagnostic terms have been developed in part in an attempt to bring clinical order to a complex set of clinical syndromes.TABLE 1: Nomenclature and the genetic basis of cholestasisThese clinical descriptions suggest important clinical characteristics of children with genetic cholestasis and appear to indicate that there are multiple causes of genetic cholestasis. For example, serum levels of γ-glutamyl transpeptidase (γ-GTP) seem to differentiate two forms of intrahepatic cholestasis, with a poorer prognosis in those children with normal γ-GTP levels (13). The presence and severity of pruritus may also distinguish different forms of genetic cholestasis, with an absence of pruritus being characteristic of abnormalities in bile acid synthesis (14). Associated malformations and syndromic features have been very important in the analysis of genetic cholestasis, with Alagille syndrome being one of the most important examples. ALAGILLE SYNDROME As early as 1965, features of what is now known as Alagille syndrome were described (5); more definitive characterization was detailed in subsequent reports (15,16). The classical features of Alagille syndrome include peculiar facies, chronic cholestasis (with intrahepatic bile duct paucity), posterior embryotoxon, butterflylike vertebral arch defects, and cardiovascular malformations (17). This very distinctive syndrome, coupled with the high prevalence of the disease within families, made it a natural choice for analysis at the level of the human genome. The initial studies demonstrating an interstitial deletion of the short arm of chromosome 20 in children with Alagille syndrome suggested two distinct possibilities (18,19). One was that Alagille syndrome was a contiguous gene syndrome, which was the consequence of the deletion of multiple genes, thereby generating the multiple features of the syndrome. The rarity of gross deletions in children with Alagille syndrome made this unlikely. The more likely possibility was that the deletion disrupted an important gene involved in normal development. The deletion served to identify the locus of the gene of interest to chromosome 20. The finding of a translocation associated with Alagille syndrome further defined the region on chromosome 20 that was involved (20) and expedited the ultimate identification of the genetic abnormality in Alagille syndrome (21,22). Contigs of bacterial and yeast artificial chromosomes and P1 clones were used to identify candidate genes in the critical region of chromosome 20 implicated in Alagille syndrome. Two separate groups identified a number of mutations in a human gene referred to as JAG1. This gene encodes a transmembrane protein that is presumed to be the ligand for a transmembrane receptor protein called NOTCH1. The interaction between JAG1 and NOTCH1 is thought to play a crucial role in early cell fate determinations (23). Thus it has the potential to account for many of the phenotypic features observed in Alagille syndrome. Apparently JAG1 plays an important role in bile duct development, and thus further analysis of its physiology and pathophysiology is of critical importance. That there is no consistent genotype-phenotype association between defects in JAG1 and clinical features of Alagille syndrome highlights how much more there is to learn about this disease and the role that this gene plays in its development (24). The investigative ability to document defects in JAG1 will permit clarification of the basis of cholestatic liver disease in children who have atypical or incomplete syndromes. This diagnosis may become relevant in children who do not have apparent liver disease, but instead are primarily affected by cardiac problems (25). BYLER'S DISEASE Byler's disease is another example of a relatively homogenous clinical entity that has been successfully explored with modern genetic techniques. This disease was first described in the Amish community in direct descendants of Jacob Byler and Nancy Kauffman (26). The clinical features of the disease usually include progressive and sometimes episodic intrahepatic cholestasis, low serum γ-GTP levels, characteristic transmission electron microscopic appearance of canalicular bile, and pruritus (27). Other less consistently reported findings include unexplained diarrhea that does not necessarily respond to liver transplantation, pancreatic insufficiency, elevated sweat chloride concentration, and wheezing (28-30). A broader term of Byler's syndrome or progressive familial intrahepatic cholestasis has been used to describe children of non-Amish origin who have a similar disease (27,30). Early investigations of bile salt metabolism in these patients suggested an abnormality in canalicular excretion of bile salts (31). This disease seems to be responsive to partial biliary diversion, implying abnormalities in the regulation of the enterohepatic circulation of toxic bile salts (32,33). The relatively homogenous clinical phenotype of Byler's disease coupled with its high incidence in an endogamous community made it a particularly good candidate for analysis by genome screening by searching for shared segments (34). This approach assumes that those affected with a given inherited condition are related to a common ancestor with a founder mutation. Analysis of genetic haplotypes for informative markers across the human genome allows for identification of a specific chromosomal region that has a high probability of being responsible for the disease in question. This approach recently lead to the identification of 18q21-q22 as a potential locus for Byler's disease (35). Of interest, this turns out to be the same locus that was identified for benign recurrent intrahepatic cholestasis (34,36). A narrowed region of chromosome 18 was analyzed for candidate genes by screening a human liver cDNA library. A P-type adenosine triphosphatase (ATPase), referred to as FIC1, was identified and found to be mutated in a number of patients with progressive familial intrahepatic cholestasis and in patients with benign recurrent intrahepatic cholestasis (37). Mutations in the patients with benign recurrent intrahepatic cholestasis occur in regions of the FIC1 protein that are thought to be less critical (i.e. less highly conserved) to the function of FIC1. The exact function of FIC1 is unknown at this time, although related genes appear to play a role in the transfer of aminophospholipids from the outer to the inner hemileaflet of the phospholipid bilayer (38). Also of interest, abnormalities in MDR 3, a gene with phosphatidylcholine transferase activity, have been recently determined to cause intrahepatic cholestasis (see later discussion). FIC1 transcripts are expressed in a wide range of tissues including liver, intestine, and pancreas. Thus Byler's disease may be a systemic disorder, and this may explain some of the clinical problems that persist even after successful liver transplantation. Future genetic analysis of children who apparently have Byler's syndrome (PFIC1) will clarify whether they have the same genetic defect that affected the descendants of Jacob Byler. CANALICULAR BILE ACID TRANSPORT DEFECTS Genetic analysis of groups of patients with progressive familial intrahepatic cholestasis has excluded the FIC1 locus in certain populations including children from Saudi Arabia, Sweden, and Eastern Greenland (9,11). Therefore, at least one alternative cause of genetic cholestasis exists. Many clinicians and investigators have been convinced that some forms of pediatric cholestasis are the result of inherited abnormalities in canalicular excretion of bile salts (39,40). An alternative to genetic analysis of these patients has been to try to identify the protein or proteins involved in canalicular excretion of bile salts and then to search for defects in those proteins in selected cases of pediatric cholestasis. This approach relies on a relatively complete understanding of the mechanisms involved in the vectorial transport of bile acids from blood to the bile (Fig. 1). Sodium-dependent and -independent transport processes have been identified at the basolateral surface of the hepatocyte and are the primary mechanisms responsible for the hepatic extraction of bile salts from the systemic circulation. At least three proteins (sodium taurocholate cotransporting polypeptide [41], organic anion transporting polypeptide [42], and microsomal epoxide hydrolase [43]) have been shown, by various techniques, to be involved in the process of uptake of bile acids at the hepatic sinusoidal membrane. Once within the hepatocyte, bile salts appear to bind to specific cytosolic proteins, including dihydrodiol dehydrogenase in the human liver (44). These intracellular binding proteins presumably help prevent reflux of bile acids across the sinusoidal membrane, sequester them to reduce their detergent properties, and may play a role in intracellular transport of bile salts. There is controversy about the mechanisms involved in the movement of bile salts from the basolateral to the canalicular membrane, with passive diffusion and microtubule-dependent vesicular pathways proposed. Once at the canalicular membrane, potential-driven and adenosine triphosphate (ATP)-dependent processes appear to be involved in the excretion of bile salts into the canaliculus. In the rat, the 100 kDa ecto-ATPase protein has been found to be involved in potential-driven canalicular excretion of bile salts (45). Given the high gradient in bile salt concentrations that exists between the canalicular lumen and the inside of the hepatocyte, most investigators believe that the ATP-dependent transport processes are physiologically the most important in bile salt excretion.FIG. 1: Diagram of the pathways of vectorial translocation of bile acids and other components of bile. Bile acids (BA) are transported from the systemic and portal circulation into the hepatocyte by one of at least three proteins, sodium taurocholate cotransporting polypeptide (NTCP), organic anion transporting polypeptide (OATP), and microsomal epoxide hydrolase (MEH). Once within the hepatocyte, they bind to intracellular binding proteins and may passively diffuse through the cytosol or are transported through the Golgi/endoplasmic reticulum network through a microtubular-dependent process. Finally, at the canaliculus at least four proteins appear to be involved in canalicular excretion of bile: canalicular multispecific organic anion transporter (cMOAT), sister gene of P-glycoprotein (SPGP), multidrug resistance proteins (MDR), and EctoATPase. SPGP and EctoATPase have been shown to be involved directly in bile acid excretion. In contrast, MDR 3 is involved in canalicular excretion of phospholipids (PL), whereas cMOAT transports conjugated bilirubin (BIL). Molecular defects in SPGP, cMOAT, and MDR 3 lead to the distinct clinical diseases that are discussed in this review.Cloning of transport proteins involved in efflux is much harder than identifying those proteins involved in uptake, and thus, characterization of the ATP-dependent canalicular bile acid transporter has been difficult. It was not until recently that it was appreciated that canalicular transport of bile salts was an ATP-dependent process (46,47). Adenosine triphosphate-dependent transport processes are often mediated by ATP binding cassette (ABC) proteins. This fact was used to clone a rat canalicular ATP-dependent bile acid transport protein (48). Endogenous efflux mechanisms made assessment of the function of this gene product in oocytes problematic (49). Therefore, ATP-dependent bile acid transport studies were performed in membrane vesicles prepared from insect cell lines that were transduced with this newly cloned gene. The cloned gene is identical to a liver-specific ABC protein referred to as the sister gene of P-glycoprotein (SPGP), which had been previously predicted to be a canalicular bile acid transporter (50). Genome-wide screening of six consanguineous Middle Eastern families has localized another form of genetic cholestasis, which some have referred to as PFIC 2 (51), to 2q24 (52). Children with PFIC 2 appear to have a more progressive disease that leads to end-stage liver disease more quickly than children with FIC1 abnormalities. Genetic analysis of 41 patients from 35 families using a variety of informative markers pinpointed the region of abnormality in these patients to 870 kb of chromosome 2. Given the bile acid-transporting properties of SPGP, its gene was a logical candidate to be tested (48). Ten different mutations have been reported in SPGP in this patient population (53). Although none of the mutations have been shown in vitro to result in bile acid-transport abnormalities, the evidence is compelling that this is the cause of cholestasis in children with PFIC 2. Absence of the SPGP protein in patients with PFIC 2 has also been observed (54). This is a wonderful example in which investigators approaching a problem from different directions (genetics and physiology) have arrived at the same predicted answer. MULTIDRUG RESISTANCE PROTEINS Some of the advances in our understanding of genetic cholestasis have stemmed from initially unrelated endeavors, with the most notable recent example being analysis of the multidrug resistance gene MDR3. The multidrug resistance genes have been the subject of intense investigation by oncologist because of their ability to impart resistance to chemotherapeutic agents. The function of one of these multidrug resistance proteins, known as mdr2 in mice and MDR3 in humans, had eluded identification for many years. Knockout mice were created in an effort to understand the function of mdr2, and unexpectedly, the major manifestation in these mice was nonsuppurative inflammatory cholangitis (55,56). Physiologic analysis of these knockout mice revealed the absence of phospholipids in their bile and suggested that the mdr2 protein was involved in phospholipid transport. Studies of normal rat canalicular plasma membrane vesicles revealed the presence of ATP-dependent phosphatidylcholine flippase activity, which is presumed to represent the exact function of the mdr2 protein (57). Ultrarapid cryofixation ultrastructural studies of mouse liver have convincingly shown that the mdr2 protein is crucial for phosphatidylcholine flipping from the inner to outer hemileaflet at the canaliculus. This ultimately results in vesicular-mediated excretion of phospholipid from the liver into bile (58). Interruption of this process can lead to impaired phospholipid excretion into bile. This yields bile with a preponderance of monomeric nonmicellar bile salts. The detergent properties of the nonmicellarbile salts are presumed to cause the canalicular injury seen in the mdr2-/- mice. The finding of a cholangitic picture in the mdr2 knockout mice prompted a search for a defect in the human homologue MDR3 in children with familial cholestasis. Children with relatively progressive cholestasis characterized by elevated γ-GTP levels have been found to have missense mutations in the MDR3 gene (Fig. 1). This is associated with an absence of immunohistochemically detectable MDR3 protein (59,60). In selected patients in whom it has been measured, biliary phospholipid levels are extremely low, similar to that observed in the knockout mice. Thus, investigations originally directed at understanding a potential cancer resistance gene have lead to the discovery of a new category of genetic cholestatic liver disease. Recent discoveries in two other areas are important to note briefly in the context of genetic forms of cholestasis. The first is the role of inborn errors of bile acid synthesis in progressive cholestasis. A variety of inborn errors of bile salt synthesis have been described, and they often have cholestasis as one of their major clinical features (61). Systematic analysis of urinary excretion of bile acid metabolites by fast atom-bombardment ionization mass spectrometry and gas chromatograph-mass spectrometry in 30 children with progressive intrahepatic cholestasis uncovered the existence of 3β-hydroxy-C27-steroid dehydrogenase-isomerase deficiency in five children (14). In addition, four of the siblings of these five children were ultimately found to have the same deficiency. The clinical features of these children are somewhat distinct in that they have cholestasis without significant pruritus and they have low serum levels of primary bile acids, γ-GTP, and cholesterol. Examination of a liver biopsy specimen reveals canalicular cholestasis with lobular or perivenous fibrosis. These children seem to have very favorable responses to ursodeoxycholic acid therapy, despite its inability to reduce primary bile acid synthesis. Another major recent advance related to genetic cholestasis is the identification of a defect in the multispecific organic anion transporter-canalicular multidrug resistance protein, cMOAT/MRP2, as the cause of Dubin-Johnson syndrome (Fig. 1). Dubin-Johnson syndrome is a disease characterized primarily by impaired canalicular excretion of conjugated bilirubin, which correlates well with the function of cMOAT/MRP2(62,63). As with MDR2, these discoveries were initially stimulated by observations in a naturally occurring animal model of Dubin-Johnson syndrome, the TR-rat (64,65). CLINICAL APPROACH TO PEDIATRIC CHOLESTASIS In light of the recent advances described in this review, the diagnostic evaluation of the child with cholestasis has become even more complex for the pediatric gastroenterologist. An algorithmic approach to these new potential diagnoses is presented in Figure 2. This approach may be viewed by some as investigative, but it is only through such studies that correct diagnoses and potential treatments can be initiated. In addition, this type of comprehensive analysis is essential in advancing our understanding of pediatric cholestasis. The initial approach to the child with cholestasis is well known, has been documented in many sources, and should proceed as usual (66). If no clear cause for the cholestasis can be found, the following approach is suggested: Alagille syndrome may be suspected in children with cholestasis and syndromic features. Unfortunately, it is not clear exactly how many of the features of this syndrome must be present to be diagnostic. Many of the characteristics such as the facial features, posterior embryotoxon, and peripheral pulmonic stenosis are individually relatively common. Therefore, diagnostic uncertainty is common in many cholestatic children with an incomplete syndrome. Ascertainment of a correct diagnosis of Alagille syndrome may be important in providing proper genetic counseling, in assessing the advisability of portoenterostomy as a therapeutic option, and in the process of evaluation of potential living-related liver transplant donors. Thus analysis for mutations in Jag1 should be pursued in patients with cholestasis and incomplete syndromes.FIG. 2: Algorithm for the approach to pediatric cholestasis.If syndromic features are absent and there is no significant pruritus-quantitative analysis of serum, bile acid concentrations should be undertaken. Significant elevations in serum bile salts should prompt further assessments for the pruritic patient described below. Cholestasis in the setting of normal serum bile acids should be followed by analysis of urine bile acid metabolites by fast atom-bombardment mass spectrometry and potentially follow-up using gas chromatography-mass spectrometry. This analysis will permit identification of specific inborn errors in bile salt biosynthesis and appropriate disease-specific therapies (e.g., primary bile acid therapy). In those patients who have pruritus and/or significant hypercholanemia, serum levels of γ-GTP can be useful in deciding in which direction to proceed. Elevated γ-GTP is seen in children with defects in MDR3, and the finding of low levels of biliary phospholipid would be highly suggestive of this diagnosis. These patients are often resistant to medical management, and liver transplantation should be considered early in their course. Finally, in those patients with normal γ-GTP levels, the possibility of a defect in the FIC1 gene should be considered. Screening of these patients by electron microscopic analysis of a liver biopsy specimen can be very helpful, because these patients have characteristic-appearing bile (27). It is important that ursodeoxycholic acid therapy be discontinued before the biopsy, because it can alter the characteristic features of the bile. Ultimately, genetic analysis for defects in FIC1 should be entertained. The responsiveness of Byler's disease to partial biliary diversion justifies careful investigation to document this disease. Finally, in those patients with normal γ-GTP levels and normal-appearing bile, the possibility of a defect in SPGP should be considered. It is not known how well these patients will respond to biliary diversion or to ursodeoxycholic acid therapy. Liver transplantation is assumed to be curative, given the hepatic-specific expression of SPGP. This algorithm clearly must be recognized as a work in progress; not every child fits into this approach. Over time, the true ranged of clinical phenotypes of the diseases described here will become apparent. In addition, new genetic defects that cause pediatric cholestatic liver disease are sure to be identified. NOMENCLATURE The recent advances in understanding the molecular basis of pediatric cholestatic liver diseases have not helped resolve the confusing nomenclature that exists in this field. The "current" terminology will most likely undergo significant changes as more molecular defects are described and as we develop a true understanding of the phenotypic expression of these genetic abnormalities. PFIC 1 (progressive familial intrahepatic cholestasis 1) should be used to describe children with the disease previously referred to as Byler's disease or syndrome and may now be limited to children with defects in the FIC1 gene. The designation PFIC 2 has been suggested to describe children who appear to have a defect in the sister gene of P-glycoprotein. (Recently investigators in this field have decided to rename this protein/gene the bile salt export pump, BSEP). Some investigators have used PFIC 3 to refer to children with an abnormality in the MDR3 gene. The specific bile acid biosynthesis enzyme deficiency can be used to describe children with inborn errors in primary bile acid synthesis. The diagnosis of Alagille syndrome has been reserved for patients who manifest at least three of five major clinical criteria. Future diagnoses may be made in patients with different presentations and may be based on genetic analyses (25). THE FUTURE: UNRESOLVED QUESTIONS The future for clinical and research investigations into pediatric cholestatic liver disease is full of great potential with many important questions to be addressed. The complexity of the mechanisms of bile flow clearly suggests that many more genetic abnormalities have yet to be identified (67). Certainly a number of idiopathic cholestatic disorders remain unexplained. Diseases with novel features can be the result of disruption of other gene products involved in the formation of bile and/or development of the intrahepatic and extrahepatic biliary systems. For instance, Aagenes syndrome appears to be a distinct form of intrahepatic cholestasis that is associated with lymphatic problems of unclear cause (68). Structural and developmentally important genes that may be involved in bile and lymphatic duct development may be affected in this disorder. Another example is an unusual form of cholestasis that has recently been described in the Amish community and consists of markedly elevated serum bile acid levels without significant hepatocellular injury (69). This isolated hypercholanemia has been hypothesized to be the result of an abnormality in hepatic clearance of bile salts, although the molecular basis of the disorder remains elusive (70). One of the most important areas of investigation is the determination of the prevalence of these various genetic abnormalities in children with liver disease. Comprehensive analysis of large groups of children with cholestasis will allow for more precise understanding of genotype-phenotype associations and will provide important insights into the structure and function of these important components of the biliary secretory apparatus. The exact mechanisms by which the Jagged and Notch genes control cell fate determination should be explored. In addition, remarkable differences in the clinical manifestations seen in people with identical genetic defects in Jagged 1 remains unexplained and suggests the existence of modifying genes that have yet to be identified. The physiologic function of the FIC1 protein remains unknown. Development of a clear explanation of how defects in FIC1 lead to the clinical constellation seen in Byler's disease and benign recurrent intrahepatic cholestasis is a high priority. Our understanding of the pathophysiologic mechanisms involved in the development of cholestasis in MDR3 defects has been greatly enhanced by the existence of a knockout mouse model. Despite these advances, the prevalence of MDR3 defects in children with liver disease is unknown (71). Finally, the recent discovery of the role of SPGP in canalicular bile acid excretion will permit detailed analysis of the regulation of this important gene product in a variety of physiologic and pathophysiologic states.