When lightning strikes twice: The plot thickens for a dual role of the anion exchanger 2 (AE2/SLC4A2) in the pathogenesis and treatment of primary biliary cirrhosis☆
Article Outline
Ae2a,b-deficient mice develop antimitochondrial antibodies and other features resembling primary biliary cirrhosis. Salas JT, Banales JM, Sarvide S, Recalde S, Ferrer A, Uriarte I, Oude Elferink RP, Prieto J, Medina JF.
Background/aims
Cl(−)/HCO(3)(−) anion exchanger 2 (AE2) is involved in intracellular pH (pH(i)) regulation and transepithelial acid–base transport, including secretin-stimulated biliary bicarbonate excretion. AE2 gene expression was found to be reduced in liver biopsy specimens and blood mononuclear cells from patients with primary biliary cirrhosis (PBC), a disease characterized by chronic nonsuppurative cholangitis associated with antimitochondrial antibodies (AMA) and other autoimmune phenomena. In mice with widespread Ae2 gene disruption, we previously reported altered spermiogenesis and reduced gastric acid secretion. We now describe the hepatobiliary and immunologic changes observed in these Ae2(a,b)-deficient mice.
Methods
In this murine model, splenocyte pH(i) and T-cell populations were studied by flow cytometry. CD3-stimulated cytokine secretion was estimated using cytokine arrays. AMA were evaluated by immunoblotting and proteomics. Hepatobiliary changes were assessed by immunohistopathology, flow cytometry, and serum biochemistry. Cholangiocyte gene expression was analyzed by real-time polymerase chain reaction.
Results
Ae2(a,b)(−/−) mice exhibit splenomegaly, elevated pH(i) in splenocytes, increased production of interleukin-12p70 and interferon gamma, expanded CD8(+) T-cell population, and under represented CD4(+)FoxP3(+)/regulatory T cells. Most Ae2(a,b)(−/−) mice tested positively for AMA, showing increased serum levels of immunoglobulin M and G, and liver-specific alkaline phosphatase. About one third of Ae2(a,b)(−/−) mice had extensive portal inflammation with CD8(+) and CD4(+) T lymphocytes surrounding damaged bile ducts. Cholangiocytes isolated from Ae2(a,b)(−/−) mice showed gene expression changes compatible with oxidative stress and increased antigen presentation.
Conclusions
Ae2 deficiency alters pH(i) homeostasis in immunocytes and gene expression profile in cholangiocytes, leading to immunologic and hepatobiliary changes that resemble PBC.
[Abstract reproduced by permission of Gastroenterology 2008;134:1482–1493]
The last decade represents an enormously productive era in hepatobiliary transport physiology with the identification and characterization of the key transport systems for bile secretion by hepatocytes and bile duct epithelial cells [1], [2]. Major progress has been made in understanding their role in cholestatic liver diseases showing that inborn and acquired defects may contribute to cholestasis [3] and that adaptive overexpression in response to cholestasis may help the liver to eliminate or limit accumulation of toxic biliary constituents.
The group around Jesus Prieto and Juan F. Medina has a longstanding track record in investigating the potential pathophysiological and therapeutic role of AE2, which mediates biliary bicarbonate secretion via Cl−/HCO3− exchange, in primary biliary cirrhosis (PBC) [4], [5], [6], [7], [8], [9]. After the initial observation that AE2 mRNA levels are reduced in livers and – notably – lymphocytes of PBC patients [4] and that UDCA restores hepatic AE2 expression levels and function [5], they expanded their findings to salivary glands in Sjögren’s syndrome (frequently associated with PBC), providing the molecular basis for PBC as “dry gland syndrome” [6]. The functional importance of their findings was elegantly underlined by demonstrating impaired AE2 activity with reduced HCO3− secretion in cholangiocytes isolated from PBC patients [7]. Of note, the existing data altogether suggest that the observed alterations in AE2 function and expression may not be merely secondary to inflammation and cholestasis. In addition, they recently showed that the combination of ursodeoxycholic acid and glucocorticoids activates the AE2 promoter in human hepatocytes via an interaction between hepatocyte nuclear factor 1 and the glucocorticoid receptor, thus providing for the first time a molecular explanation for the beneficial effects of combination therapy with these compounds in PBC [8].
In the current paper by Salas et al. they now demonstrate that 
mice show immunologic, and, at least in part, biochemical and hepatobiliary morphological features resembling those found in PBC patients [9]. The animals showed immunologic abnormalities such a significantly reduced CD4+/CD8+ ratio as a consequence of CD8 expansion and a reduced number of natural regulatory T cells (T regs) which are critical for immune tolerance to self-antigens. Moreover, most mice spontaneously developed increased serum IgG, IgM, and importantly AMA equivalent to those in PBC patients, since these antibodies were tested against a recombinant mouse PDC-E2 peptide with the PBC autoepitope. 
mice had significantly elevated ALP serum levels which, however, did not increase over time. Applying real-time PCR analysis on isolated cholangiocytes revealed altered expression of genes involved in oxidative stress and proteasome-mediated protein degradation. The authors concluded that Ae2 deficiency in mice alters intracellular pH homeostasis in immunocytes and cholangiocytes consequently leading to characteristic immunologic and hepatobiliary changes also observed in PBC patients. This paper may open a new chapter in understanding the pathogenesis of cholangiopathies by probably linking defects in bile formation with autoimmunity by one common molecular mechanism.
One stunning finding is the wide variation of the liver phenotype in 
mice indicating that their liver disease may primarily be caused by immunologic mechanisms and not the transporter defect of bile duct epithelial cells with subsequent alterations of intracellular pH which could contribute to oxidative stress. Alternatively, decreased hepatic Ae2 function could result in impaired bile flow and contribute to the cholestatic liver disease in 
mice. However, only 2 out of 11 animals showed intense portal inflammation, 2 moderate, 6 mild portal inflammation, and 1 animal no inflammation at all favouring immunologic mechanisms since a general defect in bile formation should not lead to such variations. It is also worthwhile to mention that serum ALP levels did not further increase over time indicating some stabilization of the liver phenotype and the lack of a progressive type of disease. It will be interesting to see which mechanisms contribute to this finding (e.g. stimulation of alternative transport routes, adaptive metabolic changes). Moreover, no female gender predominance as seen in PBC in humans was noted.
The hunt for appropriate PBC models has been problematic because of poor face validity using highly artificial experimental models which reduplicated only parts of the immunologic abnormalities and lacked progressive liver disease. Interestingly enough, a common feature of all recently emerged “spontaneous” autoimmune biliary disease mouse models (“PBC models”) not requiring previous manipulations for breakdown of immunotolerance to pyruvate dehydrogenase (PDC)-E2 protein including NOD.c3c4 and NOD.c3c4-derived mice, IL-2Rα−/− mice, dominant negative TGF-β receptor II mice and – as of now – 
mice appears to be a relative reduction of circulating naturally T regs; thus, disturbed T reg function or relatively reduced number may play a key role in the pathogenesis of autoimmune diseases through loss of self tolerance (see excellent recent review by Chuang et al. [10]).
The authors of the current paper on 
mice are wise enough to clearly state the potential limitations of their mouse model and that extrapolations to PBC should be made with great caution. Several important questions emerge when the findings of this interesting model are integrated into a possible PBC model: Will these mice develop ductopenia and biliary fibrosis/cirrhosis over time? Could lack of disease progression be related to the more hydrophilic bile acid pole in mice? While wild-type mice are relatively resistant to cholic acid feeding depending on their strain, “humanizing the bile” through cholic acid could lead to progressive liver disease in 
mice. What are the reasons for the considerable inter-individual variations of the liver phenotype and is this model suitable for treatment studies? Which mechanisms are responsible for the obviously self-stabilisation of the liver phenotype in these mice? Do 
mice have normal bile formation and composition? What is the role of genetic AE2 variants for disease susceptibility, progression and/or response to therapy in (a subgroup of) PBC patients?
As it may be typical for a great step forward in science, the work by Prieto and Medina not only provides some important answers, but also leads us to many novel and exciting questions which will hopefully be resolved in the near future. Current and future treatment strategies for PBC patients and perhaps for cholangiopathies in general, should consider therapeutic stimulation of AE2 expression and function in their concepts.
References
- . Molecular pathogenesis of cholestasis. N Engl J Med. 1998;339:1217–1227
- . Enterohepatic bile salt transporters in normal physiology and liver disease. Gastroenterology. 2004;126:322–342
- . Hepatocanalicular transport defects: pathophysiologic mechanisms of rare diseases. Gastroenterology. 2006;130:908–925
- . Abnormal expression of anion exchanger genes in primary biliary cirrhosis. Gastroenterology. 1993;105:572–578
- . Decreased anion exchanger 2 immunoreactivity in the liver of patients with primary biliary cirrhosis. Hepatology. 1997;25:12–17
- Anion exchanger immunoreactivity in human salivary glands in health and Sjögren’s syndrome. Am J Pathol. 1995;146:1422–1432
- Defective regulation of cholangiocyte Cl−/HCO3(−) and Na+/H+ exchanger activities in primary biliary cirrhosis. Hepatology. 2002;35:1513–1521
- . Combination of ursodeoxycholic acid and glucocorticoids upregulates the AE2 alternate promoter in human liver cells. J Clin Invest. 2008;118:695–709
- Ae2a,b-deficient mice develop antimitochondrial antibodies and other features resembling primary biliary cirrhosis. Gastroenterology. 2008;134:1482–1493
- . Animal models of primary biliary cirrhosis. Clin Liver Dis. 2008;12:333–347
☆ The authors declare that they do not have anything to disclose regarding funding from industries or conflict of interest with respect to this manuscript.
PII: S0168-8278(08)00803-9
doi:10.1016/j.jhep.2008.12.006
© 2009 European Association for the Study of the Liver. Published by Elsevier Inc. All rights reserved.
