Can Turner syndrome teach us about the pathogenesis of chronic cholestasis?



The mechanisms that cause the female predominance of primary biliary cirrhosis (PBC) are uncertain, but the X chromosome includes genes involved in immunological tolerance. We assessed the rate of X monosomy in peripheral white blood cells from 100 women with PBC, 50 with chronic hepatitis C, and 50 healthy controls, by fluorescence in-situ hybridisation. Frequency of X monosomy increased with age in all groups, but was significantly higher in women with PBC than in controls (p<0.0001); age-adjusted back-transformed mean frequencies were 0.050 (95% CI 0.046-0.055) in women with PBC, 0.032 (0.028-0.036) in those with chronic hepatitis C, and 0.028 (0.025-0.032) in controls. We suggest that haploinsufficiency for specific X-linked genes leads to female susceptibility to PBC.

Invernizzi P, Miozzo M, Battezzati PM, Bianchi I, Grati FR, Simoni G, Selmi C, Watnik M, Gershwin ME, Podda M. Frequency of monosomy X in women with primary biliary cirrhosis. Lancet 2004; 363: 533535. (Reprinted with permission of Elsevier Ltd.)


In the majority of autoimmune mediated diseases, females are more commonly affected than males. Primary biliary cirrhosis (PBC) along with Hashimoto thyroiditis, systemic lupus erythrematosus (SLE) and Sjogren syndrome represent typical examples of this phenomenon with the prevalence of female to male estimated to be 8-10 to 1. PBC most commonly affects middle-aged women. Progressive destruction of the interlobular bile ducts causing cholestasis to promote the development of fibrosis is the histological hallmark of this condition.1 Antimitochondrial antibodies (AMA) are found in 90%-95% of patients and the major mitochondrial antigen recognized by AMA is the lipoyl domain of the E2 subunit of mitochondrial PDC.

The reasons why autoimmune disease predominantly affects females remain to be elucidated. Pregnancy, being an obvious immunological challenge, has been suggested to have an impact on subsequent immune/autoimmune responses.2 In several animal models of experimentally induced autoimmune diseases, estrogens have been found to aggravate and androgens to suppress disease activity, thus it has been suggested that hormonal differences between males and females may contribute to the female preponderance to autoimmune disease.3 In fact, high levels of estrogen were found to play a role in favouring secretion of Th2 cytokines4 and this phenomenon was linked to the beneficial effect of pregnancy on multiple sclerosis or autoimmune hepatitis.5 However the role of hormones in the development of either SLE or other human autoimmune disorders has not been confirmed, thus hormones, although they may play a contributory role, are probably not a prime cause for autoimmunity.6 Even in children, autoimmune diseases show a female preponderance so other, nonhormonal risk factors need to be considered. It has been recently postulated that “inadequate tolerization” during an inactivation of one X chromosome in early embryonic phases may promote the predisposition of females to develop SLE.7 The potential role of this phenomenon has not been studied in PBC. Rather, the female predisposition to PBC has been linked to its purported microbial pathogenesis, as replication of betaretroviruses, pathogens recently associated with PBC8 are known to be regulated by a progesterone-responsive glucocorticoid regulatory element in the viral promoter region.9

Conditions associated with X-chromosome monosomy or structural abnormalities of this chromosome such as Turner syndrome (TS) or premature ovarian failure are often associated with autoimmune disorders leading Invernizzi et al. to analyze the frequency of X-chromosome monosomy in females with PBC.10 Using fluorescence in situ hybridization in different subpopulations of peripheral white cells, the authors showed that X-chromosome monosomy occurred significantly more commonly in patients with PBC than in age-matched healthy controls and also age-matched female patients with chronic hepatitis C. This phenomenon was more pronounced in cells involved in adaptive immune reactions such as T and B cells than in other leukocyte subpopulations. Rate of X-chromosome monosomy was positively correlated with age but not bilirubin level or stage of the disease. The presence of liver disease alone had no effect on the rate of X-chromosome monosomy as it was no different when healthy controls and patients with chronic hepatitis C were compared. The authors suggest that acquired deficiency of genes localized to the X chromosome may promote altered immune regulation which in turn enhances the risk of PBC. They also postulate that the cause of the increased presence of X-chromosome monosomy could be a consequence of increased mitotic activity, induced by exposure to various genotoxins.

The finding by Invernizzi et al. of an increased frequency of X chromosome monosomy in PBC begs the obvious question: how often do we see PBC in patients with Turner Syndrome (TS), a condition associated with complete or partial X-chromosome monosomy. TS manifests itself in phenotypic females and its classical clinical characteristics include short stature and features secondary to gonadal dysgenesis.11 The prevalence of TS is estimated to be 1:2500 births (this number may probably be challenged as there are reports suggesting that prenatal diagnosis of TS now prompts abortion in 71%-100% of affected pregnancies12).

Does PBC occur in TS? There are at least 3 case reports which describe PBC in TS.13–15 In our centre, which is a provincial referral centre for PBC, we also have one patient with TS and PBC. Simple calculation shows that if the prevalence of TS is 1:2500, there should be around 5000 patients with TS in our province. Even if we assume that we see all patients with PBC from our province and this is the only case of TS with PBC in Ontario, the calculated prevalence of PBC in TS would be 10 times higher than in general population of Ontario.16 As life expectancy in TS is significantly lower than that for the general population, many individuals with TS may not survive the age at which PBC manifests itself clinically thus this figure may be much higher than tenfold.

Are there any similarities between TS and PBC? First of all, TS is associated with chronic cholestatic liver disease of unknown etiology. Progression to cirrhosis is described and the relative risk of developing liver cirrhosis in patients with TS is 5.7 times higher than that of general population.17 As is the case for PBC, liver disease associated with TS increases with age and 80% of TS patients over the age of 35 are reported to have abnormal liver enzymes, predominantly the GGT.18 An association between age and liver test abnormalities was also seen in another study in TS patients where subjects affected with anicteric cholestasis were significantly older (mean age 33 years) than those with normal liver biochemistry (mean age 25 years).19 Several other reports and case studies also indicate that age related and persistent yet asymptomatic, cholestasis is observed in TS.20–24 To date no cause has been ascribed to this phenomenon and as AMA have not been reported in majority of theses studies it remains unknown whether this could be PBC. Of interest, in study by Larizza, et al.20 2 patients were found to have anti-LKM antibodies detected with indirect immunofluorescence, a method frequently associated with misinterpretation of AMA as LKM.25 In the same study by Larizza et al., 21% of patients with TS were found to have positive ANA or SMA (again AMA were not tested for).20 This ratio of ANA and SMA is similar to that observed in AMA positive PBC.26 It has also been reported that pruritus may be the presenting symptom in TS with anicteric cholestasis and histological features of bile duct proliferation may be found on liver biopsy in these patients.22, 27

Both TS and PBC are strongly associated with autoimmune thyroiditis. Antithyroid antibodies are found in approximately 40% of patients with TS and occur significantly more often in those with abnormal liver biochemistry (50% vs. 28%).20 Osteopenia, which commonly complicates PBC is also associated with TS where the risk of osteoporosis is 10 times higher than in general population.17 Although estrogen deficiency seems to be the most likely risk factor, bone mass improves but does not normalize on hormonal therapy in patients with TS.28 Thus the presence of an unknown “intrinsic bone defect” has been postulated to explain decreased bone mass in these patients.11 It can be speculated that chronic cholestasis may partially contribute to abnormal bone density in TS. There are many known but also unknown risk factors for the metabolic bone disease associated with chronic cholestasis. One such risk factor which also has a similar prevalence in both PBC and TS is celiac disease. A recent, large multicentre Italian study showed that 6.4% of TS patients have celiac disease, a figure very similar to that reported in PBC.29

The etiology of the chronic cholestasis reported to occur in TS has not been well characterized. At least four other potential factors/conditions have been linked with liver abnormalities in these patients: (a) exogenous estrogen therapy; (b) nonalcocholic steatohepatitis (NASH); (c) vascular abnormalities; (d) primary sclerosing cholangitis. Some studies have shown an increase of liver enzymes related to treatment with conjugated estrogens30, 31 but this was not confirmed by more recent work.19, 32, 33 In fact, in one study a significant improvement of liver biochemistry was observed in patients with TS treated with E2 valerate.33 Therefore, it is unlikely that hormone supplementation alone is responsible for the chronic cholestasis seen in TS.11 Histological features of NASH are reported in some patients with TS and as in general population were associated with hyperlipidemia and central obesity.32 Hepatic vascular abnormalities such as NRH or FNH found in 8 of 27 patients with TS were recently reported in this journal.32 The risk of IBD in TS is estimated to be between 2.6% and 3% and is significantly higher that in general population.34 Interestingly, bile duct alterations on liver biopsy, including periductular fibrosis (which is typical for PSC) were observed in 77% of patients with TS undergoing liver biopsy for work up of abnormal liver tests.32 There are at least two other reports in the literature on PSC occurring in patients with TS.22, 35

The reports described above suggest different causes for liver disease in TS, however chronic cholestasis seems to be the most prevalent abnormality. Also if patients with TS and chronic cholestasis, are treated with ursodeoxycholic acid (used in doses recommended for PBC) significant improvement of their liver biochemistry is observed.14, 22, 24

The paper by Invernizzi et al points to a possible relationship between X-chromosome monosomy and PBC. The prevalence of PBC among individuals with TS has never been investigated but many studies in patients with TS indicate a high prevalence of chronic and age-related cholestasis, often associated with other autoimmune diseases. If the cause is PBC or its variant, namely AMA negative PBC, this finding would lend further support to the role of aneuploidy in the pathogenesis of PBC. Certainly, a potential contribution of X chromosome related genes in the development of PBC requires further investigation.