*Pediatric Gastroenterology and Liver Unit
†Pediatric Surgery and Transplant Unit
‡Human Genetics Centre, Cliniques Saint-Luc, Université Catholique de Louvain, Brussels, Belgium
§Pediatric Hepatology, Bicêtre Hospital, Assistance Publique-Hôpitaux de Paris, Paris, France
||Genetics Department, Necker Enfants Malades Hospital, Paris, France.
Address correspondence and reprint requests to Professor Etienne M. Sokal, MD, PhD, Department of Paediatrics, Université Catholique de Louvain, 10, Av Hippocrate—1200 Brussels, Belgium (e-mail: firstname.lastname@example.org).
Received 6 December, 2010
Accepted 13 February, 2011
The authors report no conflicts of interest.
Tight junctions between hepatocytes and cholangiocytes play a fundamental role in separating bile flow from plasma. Secondary alterations of tight junctions are well described in many cholestatic disorders inducing an increase in paracellular permeability and subsequent liver damage (1). The core proteins of tight junctions are composed of strands of claudins and occludin. Claudin-1 is a member of the claudin family expressed in liver and skin and mapping on chromosome 3q27-q28. The CLDN1 gene homozygous mutation was shown to cause a rare autosomal recessive syndrome associating neonatal ichthyosis to sclerosing cholangitis (NISCH syndrome) and first described in 2 unrelated Moroccan families (2,3). Only 8 patients, from 4 different families, have been described as affected by this syndrome so far (2–5). They presented with ichthyosis and neonatal cholestatic jaundice. Cholestasis varied from a transient remitting form to more severe fibrogenic conditions (2–5). Two patients required liver transplantation (3,5). One of them showed regression of skin lesions and alopecia after the transplant (3). We describe here 4 new patients from an inbred family of Moroccan origins, all presenting with the clinical features of NISCH syndrome and a mutation of the CLDN1 gene.
In all patients, other causes of neonatal cholestasis were excluded (6,7). Table 1 summarizes characteristics of patients described here and of previously reported patients (2–5).
Patient 1, a girl, was born from consanguineous healthy parents (Fig. 1A V:1). At birth she showed generalized lamellar desquamation and mild jaundice. Acholic stools, dark urine, itching, and hepatomegaly were noted in the first weeks of life. At 1 month of age, serum total bilirubin was 9.1 mg/dL, conjugated bilirubin 8.3 mg/dL, alanine aminotransferase (ALT) 142 U/L (normal range <63 U/L), and gamma-glutamyl transferase (GGT) 75 U/L (normal value for age <230 U/L). The abdominal ultrasound showed no particularity, but the gallbladder was not visible. A hepatobiliary scintigraphy with 99mTc-IDA showed a markedly delayed bile flow. A liver biopsy revealed a panlobular cholestasis, without bile duct proliferation, and no evidence of acute hepatitis nor bile duct paucity. Despite lack of evidence of biliary atresia, the patient underwent a portocholecystostomy at the age of 6 weeks. No cholangiography was performed. There was no significant change in jaundice or itching following the surgery. The patient was then referred for further evaluation and liver transplantation. At the age of 4 months, the patient had marked jaundice, hepatomegaly, refractory itching, dry skin, moderate ichthyosis, and rickets. Her height was at −1.3 standard deviations. She had no clinical or ultrasonographic signs of portal hypertension, and liver ultrasonography showed a heterogeneous pattern. Direct bilirubin was 5.6 mg/dL, ALT 446 U/L, GGT 849 U/L, and serum bile acids 264 μmol/L (normal range <8 μmol/L). In view of the persistent cholestasis, the child underwent split liver transplantation from cadaveric donor at age 1 year. Liver histology performed on the explanted organ showed extensive portal fibrosis with interlobular septa and moderate inflammatory infiltrate with important bile ductular proliferation, bile plugs, and absence of interlobular bile ducts. The posttransplant course was uneventful, with normalization of hepatic function and disappearance of cholestasis. Itching and skin lesions improved moderately but did not disappear, although serum bile acid levels had normalized. Enamel dysplasia and hypodontia were noted. The diagnosis of NISCH syndrome was established after the patient's younger sister was referred with the same clinical picture. Genetic testing revealed a homozygous deletion in exon 1 (c.200-201delTT) in the CLDN1 gene as previously described (3). Five years after transplantation, serum liver function tests and liver ultrasonography are normal. Itching remained significant and was related to ichthyosis.
A second sister was born prematurely for an unknown reason, and died a few days after delivery from unspecified causes (Fig. 1A V:2). No investigations were performed to evaluate possible liver disease in this child.
The second affected sister (Fig. 1A V:3) was admitted to the hospital at the age of 2 weeks for bleeding from the umbilicus, 4 days after the umbilical stub fell. At physical examination, the patient showed scleral icterus, extremely dry skin with lamellar desquamation, and parietal and occipital alopecia. Blood tests showed a markedly elevated international normalized ratio (INR [>7.0]) and cholestasis (total serum bilirubin 6.4 mg/dL, conjugated bilirubin 2.8 mg/dL), with normal serum ALT and GGT. After vitamin K intravenous injection, INR normalized and bleeding resolved. Liver ultrasonography was unremarkable. A hepatobiliary scintigraphy with 99mTc-IDA revealed a mild reduction of radionuclide liver uptake and absence of its excretion. A liver biopsy was performed and showed normal interlobular bile ducts, normal portal spaces, and regular lobular architecture with intracellular and canalicular cholestasis. Skin biopsy showed orthokeratosis, hyperkeratosis, and granular layer hyperplasia. Microscopic examination of the hair revealed trichoptilosis and focal area of trichorrhexis nodosa. No intracytoplasmic vacuoles were seen in blood cell smears. Genetic testing revealed the same homozygous CLDN1 gene mutation as in her elder sister, confirming the diagnosis of NISCH syndrome. The patient was treated with 20 mg/kg/day of ursodeoxycholic acid (UDCA) and ADEK vitamin supplementation. Biochemical cholestasis and physical signs resolved and at 4-year follow-up she complained only of mild pruritus and dry skin, which were treated by moisturizing topical therapy.
The third affected sister (Fig. 1A V:4) showed ichthyosis without jaundice at birth. Since the third day of life, she received UDCA (20 mg/kg/day). At 3 months of age she presented with cholestatic jaundice (serum ALT 495 U/L, serum conjugated bilirubin 5.4 mg/dL, and normal serum GGT). Liver ultrasonography was unremarkable. Moderate itching appeared along with parietal and occipital alopecia. Genetic analysis confirmed the presence of the known homozygous CLDN1 gene deletion. At 9 months of age, serum bilirubin and ALT normalized, whereas GGT rose to 474 U/L. Mild biochemical signs of cytolysis and cholestasis persisted at 3-year follow-up under UDCA therapy (serum ALT 74 U/L and GGT 139 U/L, normal serum bilirubin) in the absence of hepatomegaly and signs of portal hypertension. Alopecia and ichthyosis were still present and the child complained of moderate itching.
As with her elder sisters, the fourth affected baby (Fig. 1A V:5) had dry skin and mild lamellar desquamation at birth. At 5 days of age, the physical examination was unremarkable, but biochemical analysis showed a mild cholestasis (conjugated serum bilirubin 3.1 mg/dL, normal serum ALT and GGT). Treatment with UDCA (10 mg/kg/day) and liposoluble vitamin supplementation was started. At 1 month of age, she had hypocholic stools, and the dose of UDCA was increased to 30 mg/kg/day, with subsequent normalization of stools and serum bilirubin. Liver ultrasound showed a marked parenchymal heterogeneity with surface nodularity in the absence of portal hypertension. The same homozygous CLDN1 gene mutation as her elder sisters was found. At 6-month follow-up, alopecia and ichthyosis were evident (Fig. 1C-D), and biochemical signs of biliary involvement were still present (serum ALT 60 U/L, GGT 299 U/L, normal total serum bilirubin). No liver biopsy has been performed. The parents did not allow her to undergo magnetic resonance cholangiopancreatography.
The parents had a double consanguineous relationship and they were both phenotypically normal. The mother was shown to be heterozygous for the described mutation in the CLDN1 gene, whereas the father refused genetic analysis (Fig. 1A, IV:1 and IV:2).
NISCH syndrome is a rare disorder characterized by alopecia, ichthyosis, and progressive cholestasis from birth. It was first described in 2002 by Baala et al (2) who analysed 4 patients belonging to 2 different Moroccan families from the Rif region. NISCH was assigned initially to chromosome 3q27-q28. A homozygous 2-bp deletion in exon 1 of the CLDN1 gene (c.200-201delTT), leading to a frameshift and a subsequent stop codon, was identified as the cause of the syndrome (3). A second homozygous frameshift mutation in exon 2 of the CLDN1 gene was reported in a Swiss child with NISCH syndrome characteristics (4).
We report here a new family of 4 daughters affected by NISCH syndrome, originating from the same Moroccan region as 3 of the published families. The mutation identified also was the same as the one previously described in these families (3,5). Therefore, a founder effect can be supposed. The children were born full term after uneventful pregnancy and presented with mild ichthyosis from birth, which was initially misinterpreted as simple dry skin. They had neonatal jaundice, of various severities, and developed clinical and biochemical cholestasis within the first few weeks of life. Even though initially not always high, GGT levels were invariably elevated within the first months of life and they were a marker of biliary disease during follow-up. Overall, despite the same homozygous mutation in the CLDN1 gene was found, the liver disease appeared with variable severity among the affected members of the same family, ranging from regression of cholestasis to extremely early liver fibrosis requiring liver transplantation (Table 1) (2–5).
All previously published patients presented with ichthyosis and hypotrichosis. One baby was born with a collodion membrane. Alopecia was usually frontal, parietal, or occipital, with sparse eyelashes and eyebrows. Neonatal jaundice, pruritus, pale stools, and biochemical cholestasis were present in all of the described patients but 1, along with hepatomegaly and portal hypertension in some cases. Cholangiographic features of intrahepatic biliary tree anomalies were found in all 5 patients who underwent imaging studies. At liver biopsy the typical “onion-skin” appearance of periductal fibrosis was never seen, as expected considering the age of the patients (8,9). Additional features included enamel dysplasia, oligodontia, hypodontia, and vacuoled eosinophils. Like the 4 girls described here, these 8 patients showed different degrees of liver disease, ranging from absence of liver involvement or transient neonatal cholestasis to persistent cholestasis with cirrhosis (Table 1) (2–5). The reason for such phenotypic variability is not known.
In NISCH syndrome it is believed that the primary defect of claudin-1 in hepatocytes and cholangiocytes leads to loss of cell polarity and to paracellular leakage of bile constituents (ie, bile acids) that may induce cholestasis by decreasing bile flow and fibrosis, respectively. It is likely that UDCA, a choleretic and cytoprotective bile acid, led to some improvement of cholestasis in treated cases as reported in some patients with isolated neonatal sclerosing cholangitis (10). The use of UDCA was supported by its well-known effects on biochemical values and histological progression of primary sclerosing cholangitis (11). Even though GGT remained high in 3 of the 4 patients described, serum bilirubin normalized under UDCA by the ninth month of life in all of the children except for 1 (patient 1), who started therapy when the disease was far too progressed. Nevertheless, it is difficult to state whether improvement was secondary to UDCA administration or to the natural progression (or regression) of the disease.
Patient 1, who had been mismanaged undergoing corrective biliary surgery, which probably worsened the disease course, received liver transplantation at the age of 1 year. Despite normalization of cholestasis and normal serum bile acids levels, liver transplantation did not lead to significant improvement of ichthyosis, contrary to previous observation (2,3). This may be expected, because claudin-1 is also a key component of tight junctions between keratinocytes. Furuse et al (12) showed that claudin-1 knockout mice presented a severe impairment of epidermal barrier function and died within 24 hours after birth because of dehydration. Ichthyosis found in all of the described patients is secondary to this defect of epidermal barrier. Therefore, it cannot be corrected by liver transplantation, which is only indicated in case of end-stage liver disease and not for the extrahepatic, tissue-bound, manifestations of the syndrome (13,14).
Nagtzaam et al (5) suggested that bile duct paucity could be a primary phenomenon in NISCH syndrome because their patient bile ducts were not replaced by scar tissue. Ductopenia is a well-described feature of histopathological pattern found in sclerosing cholangitis, and it is typically found in stage 3 (according to Ludwig classification) (15,16). In children, it has been described in up to 60% of patients (8,9,17). The absence of interlobular bile duct in patient 1 was found on histopathological examination of the explanted organ after liver transplantation and was accompanied by bridging fibrosis (Fig. 1B). These findings were consistent with a stage 3 sclerosing cholangitis and were not present at the liver biopsy performed at presentation. Fibrosis and minimal bile duct proliferation were associated with ductopenia in 1 of the patients described by Hadj-Rabia et al (patient A3, Table 1) (3). Even though the possibility of a direct role of tight junction disease in bile duct paucity is fascinating and deserves further investigation, portal inflammatory infiltrate, ductular proliferation, portal fibrosis, and periportal inflammation, found by Nagtzaam et al, are compatible with sclerosing cholangitis, and it is difficult to affirm that ductopenia was a primary phenomenon.
One more patient with thickened skin, cholestasis, alopecia, and hypothyroxinemia was recently described by Shah and Bhatnagar (18) and suspected of having NISCH syndrome. The magnetic resonance cholangiopancreatography was unremarkable and he had no signs of sclerosing cholangitis at liver biopsy. No clear signs of ichthyosis were evident and no skin biopsy has been performed. The alopecia was localized and appeared after the resolution of a neonatal ulcer. Because this patient did not undergo genetic testing for CLDN1 mutations and the clinical picture was not typical, no definite diagnosis of NISCH syndrome can be posed, and this case was not included in Table 1.
Claudin-1 has been found to be necessary to hepatitis C virus (HCV) entry in permissive cells. Evans et al (19) have shown that even a moderate downregulation of CLDN1 expression inhibited HCV infection, thus explaining HCV tissue tropism by means of low-level expression of CLDN1 in extrahepatic tissues. Recently, monoclonal anti-claudin-1 antibodies proved to inhibit HCV infection in primary human hepatocytes (20). Claudin-1 is therefore being considered an interesting candidate target for anti-HCV therapy because its blockage could be complementary to HCV protease and polymerase inhibition. NISCH syndrome is the only known human condition involving a deficit of claudin-1. The existent murine model does not live long enough to allow liver disease to develop. For such a reason, NISCH syndrome phenotypic and pathophysiological characterization is of major interest to understand and preview possible adverse effects of a future anti-HCV treatment consisting of claudin-1 inhibition.
These reported cases confirm the existence of NISCH syndrome and its intra- and interfamilial phenotypic variability. Liver disease severity proved to be unpredictable. Early UDCA therapy may be of benefit and may slow disease progression. Liver transplantation does not lead to the resolution of extrahepatic manifestations and should be reserved for progressive liver disease. Finally, even though it is an extremely rare condition, a better definition of NISCH syndrome features could be of general interest as a model of claudin-1 deficiency.
1. Trauner M, Meier PJ, Boyer JL. Molecular pathogenesis of cholestasis. N Engl J Med 1998; 339:1217–1227.
2. Baala L, Hadj-Rabia S, Hamel-Teillac D, et al. Homozygosity mapping of a locus for a novel syndromic ichthyosis to chromosome 3q27-q28. J Invest Dermatol 2002; 119:70–76.
3. Hadj-Rabia S, Baala L, Vabres P, et al. Claudin-1 gene mutations in neonatal sclerosing cholangitis associated with ichthyosis: a tight junction disease. Gastroenterology 2004; 127:1386–90.
4. Feldmeyer L, Huber M, Fellmann F, et al. Confirmation of the origin of NISCH syndrome. Hum Mutat 2006; 27:408–410.
5. Nagtzaam IF, van Geel M, Driessen A, et al. Bile duct paucity is part of the neonatal ichthyosis-sclerosing cholangitis phenotype. Br J Dermatol 2010; 163:205–207.
6. Balistreri WF. Neonatal cholestasis. J Pediatr 1985; 106:171–184.
7. Roberts EA. Neonatal hepatitis syndrome. Semin Neonatol 2003; 8:357–374.
8. Debray D, Pariente D, Urvoas E, et al. Sclerosing cholangitis in children. J Pediatr 1994; 124:49–56.
9. Gregorio G, Portmann B, Karani J, et al. Autoimmune hepatitis/sclerosing cholangitis overlap syndrome in childhood: a 16-year prospective study. Hepatology 2001; 33:544–553.
10. Davit-Spraul A, Fabre M, Branchereau S, et al. ATP8B1 and ABCB11 analysis in 62 children with normal gamma-glutamyl transferase progressive familial intrahepatic cholestasis (PFIC): phenotypic differences between PFIC1 and PFIC2 and natural history. Hepatology 2010; 51:1645–1655.
11. Mitchell SA, Bansi DS, Hunt N, et al. A preliminary trial of high-dose ursodeoxycholic acid in primary sclerosing cholangitis. Gatroenterology 2001; 121:900–907.
12. Furuse M, Hata M, Furuse K, et al. Claudin-based tight junctions are crucial for the mammalian epidermal barrier: a lesson from claudin-1-deficient mice. J Cell Biol 2002; 156:1099–1111.
13. Sokal EM, Goldstein D, Ciocca M, et al. End-stage liver disease and liver transplant: current situation and key issues. J Pediatr Gastroenterol Nutr 2008; 47:239–246.
14. Sokal EM. Liver transplantation for inborn errors of liver metabolism. J Inherit Metab Dis 2006; 29:426–430.
15. Ludwig J. Surgical pathology of the syndrome of primary sclerosing cholangitis. Am J Surg Pathol 1989; 13:43–49.
16. European Association for the Study of the Liver. EASL clinical practice guidelines: management of cholestatic liver diseases. J Hepatol 2009;51:237–67.
17. Feldstein AE, Perrault J, El-Youssif M, et al. Primary sclerosing cholangitis in children: a long-term follow-up study. Hepatology 2003; 38:210–217.
18. Shah I, Bhatnagar S. NISCH syndrome with hypothyroxinemia. Ann Hepatol 2010; 9:299–301.
19. Evans M, von Hahn T, Tscherne DM, et al. Bieniasz PD Rice CM. Claudin-1 is a hepatitis C virus co-receptor required for a late step in entry. Nature 2007; 446:801–805.
20. Fofana I, Krieger SE, Grunert F, et al. Monoclonal anti-claudin 1 antibodies prevent hepatitis C virus infection of primary human hepatocytes. Gastroenterology 2010; 139:953–964.