Liver transplantation: An “in vivo” model for the pathophysiology of hemochromatosis?


  • See Article on Page 1655

Hereditary hemochromatosis is a disorder characterized by iron overload in multiple parenchymal organs due to excessive absorption of iron from the gastrointestinal tract. There have been major advances in our understanding of the clinical features, epidemiology, genetics, cell biology, and pathophysiology of this disease over the past several years. The identification of the HFE gene in 1996 allowed, for the first time, specific genotypic diagnosis in probands, and permitted large-scale studies of the relationship between genotype and phenotype.1 The identification of specific iron transporters in the small intestine such as divalent metal transporter 1 (DMT1), ferroportin (FP) along with Dcytb and hephaestin have elucidated the mechanism of iron entry into the enterocyte and subsequent transport across the duodenum into the circulation; increased iron absorption in patients with hemochromatosis is associated with increased expression of DMT1 and FP.2–5 These studies have provided pathophysiological correlation to previous studies documenting increased iron absorption among humans with hereditary hemochromatosis.6 The mechanistic link between the C282Y mutation in the HFE gene and increased iron absorption in hemochromatosis has been more elusive. The initial studies of Feder and colleagues, using a model of HeLa cells, suggested that HFE-associated may have an inhibitory effect on the binding of transferrin receptor to its ligand transferrin, and that the mutant HFE-associated protein may abrogate this inhibition, leading to increased tissue iron overload.7 However, subsequent studies have proposed that the C282Y mutation may be associated with failure to internalize transferrin into crypt cells, leading to a state of “iron deficiency.”8 In support of this hypothesis are the observations that the physiology of hemochromatosis is very similar to iron deficiency, with decreased ferritin messenger RNA and protein content in absorptive cells, and increased expression of transferrin receptor 1 (TfR1).5, 9 Furthermore, the cell biology of the mutant HFE protein is characterized by defective folding and failure to translocate across the cell membrane, and an inability to associate with the transferrin receptor complex and beta2 microglobulin, steps considered necessary for maintenance of normal iron homeostasis.10


DMT1, divalent metal transporter; FP, ferroportin; Dcytb, duodenal cytochrome b; TfR1, transferrin receptor 1; HLA, human leukocyte antigen; IfR2, transferrin receptor 2.

The role of the liver in hemochromatosis has remained controversial. Bomford and colleagues noted in 1989 that expression of transferrin receptor in hemochromatosis appeared appropriate for body iron stores, arguing against a primary hepatic defect.11 The recent discovery of hepcidin, a novel circulating antimicrobial peptide produced in the liver, that is capable of inhibiting iron absorption in the duodenum has generated much enthusiasm for this protein as the “missing link” between the liver and intestine in hemochromatosis.12 Hepcidin expression is inversely related to body iron stores in normal subjects and in patients with hemochromatosis, although the hepcidin response appears to be attenuated in the setting of hemochromatosis.13, 14 These data suggest that iron hyperabsorption in hemochromatosis may be related to inappropriately low hepatic hepcidin expression. Loss of function mutations in HAMP, the gene encoding hepcidin, are associated with a severe form of juvenile hemochromatosis and may serve as modifiers increasing the severity of expression of hemochromatosis in C282Y homozygotes.15 Similarly, mutations in transferrin receptor 2 (TfR2), which is localized to the liver, may be associated with a type of primary iron overload.9, 15 These data provide evidence that the liver may play an important and independent role in the clinical expression of HFE-associated hemochromatosis and that gene-gene interactions may play a role in the expressivity of this condition.16, 17

The liver transplantation model provides an excellent opportunity to dissect the relative contributions of the liver and the intestine to iron overload in hemochromatosis. Placement of an HFE wild-type liver allograft into a C282Y homozygous recipient provides an opportunity to study the isolated effect of the C282Y mutation in the intestine. Will recurrent iron overload occur in such a situation? Similarly, inadvertent transplantation of a liver from a C282Y homozyote into a wild-type recipient allows us to examine the question of whether spontaneous iron mobilization would occur in such patients.

There is limited literature examining the utility and outcome after liver transplantation for known or suspected hemochromatosis. The first report from the University of Pittsburgh suggested excellent survival in a small cohort of patients with the human leukocyte antigen (HLA) A3B7 haplotype, most of whom had been iron depleted prior to liver transplantation.18 Subsequent studies of patients with significant hepatic hemosiderosis and presumed hemochromatosis suggested poor 1- and 5-year survival, with most deaths related to sepsis or cardiac causes.19, 20 However, it became apparent that only a small minority of patients with hepatic hemosiderosis in the setting of end-stage liver disease have the homozygous C282Y mutation.21, 22 Preliminary studies showed that in C282Y homozygotes, posttransplant survival appears clearly lower than expected; infections and cardiac complications (in particular, arrhythmias and heart failure) account for most of the posttransplant morbidity.23 It remains unclear whether patients with hepatic iron overload in the absence of the C282Y mutation are at increased risk for premature mortality; some studies have found this to be the case, whereas others have not.24, 25

In this issue of HEPATOLOGY, Crawford and colleagues address two unresolved questions with regard to hemochromatosis and liver transplantation—namely, posttransplant outcomes and recurrent iron overload.26 They retrospectively reviewed more than 3,000 liver transplantations performed at several centers in Australia, New Zealand, and the United Kingdom between January, 1982 and June, 2001. Twenty-two patients with a diagnosis of hemochromatosis were identified. Seventeen were homozygous for the C282Y mutation, and 5 others were presumed to have hereditary hemochromatosis based on presence of systemic iron overload with a family history of iron overload or HLA A3B7 haplotype. Additional details of the pattern and degree of iron overload and whether these 5 patients underwent HFE mutation analysis were not provided. All patients were male, and approximately half had a history of heavy alcohol consumption (>60 gm per day for more than 10 years). Hepatocellular carcinoma was present in 8 of 22 (36%), 4 of whom were outside of current United Network for Organ Sharing criteria. Survival rates at 1, 3, and 5 years were 72%, 62%, and 55%, respectively. Recurrent hepatocellular carcinoma, arrhythmias and heart failure were the cause of death in most patients. There was no evidence of recurrent iron overload in 10 of 11 surviving patients with a median follow-up of 4 years; no patients had an elevation in serum transferrin-iron saturation. One patient developed hyperferritinemia associated with stainable iron on biopsy.

Crawford and colleagues simultaneously identified an additional 12 donor livers from a database at the University Hospital in Birmingham, United Kingdom, with increased stainable iron (>1+) in the “time zero” biopsy obtained after reperfusion of the allograft; HFE mutation analysis, performed in 5 of 12 subjects, revealed that 2 donors were C282Y homozygotes and 2 were C282Y heterozygotes. Iron stores in the liver became normal in 4 patients over a median 4-year follow-up period. However, liver iron was only mildly increased in all cases, and only 1 donor-recipient pair had HFE genotyping (both were wild-type). Two donor organs from C282Y homozygotes transplanted into 1 wild-type recipient and 1 C282Y heterozygote, respectively, demonstrated persistent iron overload 3 years or longer after liver transplantation, although there was a decrease in hepatic iron concentration in both patients. Crawford and colleagues conclude that patients with HFE-associated hemochromatosis have a disappointing outcome after liver transplantation. In addition, they suggest that the pattern of hepatic iron storage after liver transplantation in these donor-recipient pairs points to a central and essential role for the liver in contributing to the pathophysiology of iron overload in this disease.

The authors deserve credit for taking on this difficult area of clinical investigation and for their perseverance in attempting to understand the pathophysiology of hemochromatosis using the human liver transplant model. As would be expected in a retrospective study design, the unavoidable lack of complete data regarding HFE mutation status, prior gastrointestinal bleeding, blood transfusions, alcohol consumption, body iron stores, and the unknown effects of immunosuppression on iron absorption make interpretation of their results somewhat problematic. In addition, the posttransplant follow-up period was relatively short, and HFE mutation analysis was only available in a small number of donor-recipient pairs. Previous anecdotal reports from the liver transplant population have both supported and challenged the concept that the liver plays an essential role in the pathophysiology of iron overload associated with hemochromatosis.27–31 However, most of these studies were conducted prior to the era of HFE gene testing. Recent studies conducted after HFE gene testing became available have not convincingly shown evidence of hepatic iron reaccumulation after liver transplantation among C282Y homozygotes.32

The recent discovery of several proteins involved in iron metabolism such as TfR2 and hepcidin that are selectively expressed in the liver support the concept that expression of hemochromatosis requires abnormalities in iron handling in both the liver and the small intestine (Fig. 1). Lending additional support to the concept that the liver plays an important role in the aberrant iron metabolism of hemochromatosis are recent data suggesting that the mutant HFE protein may alter handling of iron by macrophages, implying that replacement with a wild-type liver might in fact return hepatic iron metabolism to normal and may lead to a reduction iron absorption after liver transplantation.33 In fact, a previous study demonstrated that wild-type HFE protein normalized iron accumulation in macrophages from patients with hemochromatosis.34 Even more convincingly, a recent study showed that reconstitution of hepcidin production by cross-breeding HFE−/− knockout mice with mice overexpressing hepcidin prevented iron accumulation, suggesting a central role for hepatic hepcidin production, rather than HFE action at the crypt, as the primary source of iron hyperabsorption in HFE-associated hemochromatosis.35 Frazer and Anderson suggested that hepcidin production by the hepatocyte may be regulated by an interaction between hepatic TfR2, HFE, and TfR1.36 Additional research examining this hypothesis is awaited.

Figure 1.

Body iron stores are regulated at the level of intestinal absorption. Non-heme iron is absorbed primarily in the duodenum and involves a coordinated process; iron is reduced from the ferric to ferrous form via a ferric reductase (Dcytb) and then transported across the brush border of the enterocyte via a mucosal iron transporter (DMT1) and subsequently across the basolateral side of the enterocyte by ferroportin 1 (FP1) in conjunction with hephaestin, a ferroxidase that oxidizes iron back into the ferrous form for entry into the circulation. The crypt cell has been proposed as an “iron-sensing” cell, wherein expression of the iron transporters may be regulated in response to body iron stores. Hepcidin, a circulating peptide secreted by the liver, has an inhibitory effect on iron absorption and has been suggested as the downstream signal from the liver and the bone marrow to the duodenum that regulates iron absorption. Hepcidin expression is inappropriately low in patients with HFE-associated hemochromatosis.

In conclusion, the interesting observations by Crawford and colleagues of the pattern of iron accumulation in the liver of both donors and recipients with the homozygous C282Y mutation and hepatic iron overload is an important addition to the available information about the relative contributions of the liver and intestine to the pathophysiology of iron overload in hereditary hemochromatosis. Long-term follow-up studies in such donor-recipient pairs will undoubtedly provide clinical correlates to transgenic knockin and knockout mouse models and will inform us about the key determinants of disease expression. In the meantime, increased awareness of this condition among physicians and knowledge and use of appropriate diagnostic tests to screen for hemochromatosis in at-risk persons should be emphasized, given the availability of safe and effective therapy in the form of phlebotomy.


The author is grateful to Anne Tiller for help with graphics in Figure 1.