SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

There have been several descriptions of mouse models that manifest select immunological and clinical features of autoimmune cholangitis with similarities to primary biliary cirrhosis in humans. Some of these models require immunization with complete Freund's adjuvant, whereas others suggest that a decreased frequency of T regulatory cells (Tregs) facilitates spontaneous disease. We hypothesized that antimitochondrial antibodies (AMAs) and development of autoimmune cholangitis would be found in mice genetically deficient in components essential for the development and homeostasis of forkhead box 3 (Foxp3)+ Tregs. Therefore, we examined Scurfy (Sf) mice, animals that have a mutation in the gene encoding the Foxp3 transcription factor that results in a complete abolition of Foxp3+ Tregs. At 3 to 4 weeks of age, 100% of animals exhibit high-titer serum AMA of all isotypes. Furthermore, mice have moderate to severe lymphocytic infiltrates surrounding portal areas with evidence of biliary duct damage, and dramatic elevation of cytokines in serum and messenger RNAs encoding cytokines in liver tissue, including tumor necrosis factor α, interferon-γ, interleukin (IL)-6, IL-12, and IL-23. Conclusion: The lack of functional Foxp3 is a major predisposing feature for loss of tolerance that leads to autoimmune cholangitis. These findings reflect on the importance of regulatory T cells in other murine models as well as in patients with primary biliary cirrhosis. (HEPATOLOGY 2008.)

Considerable research has described important regulatory mechanisms that provide immune homeostasis by limiting excessive immune responses and preventing loss of tolerance.1, 2 Naturally occurring regulatory T cells (Tregs) specifically express the transcription factor known as forkhead box 3 (Foxp3), a member of the forkhead/winged-helix family of transcription factors that is essential for the development, maintenance, and function of Tregs. Importantly, Foxp3 expression is sufficient to confer suppressive activity to conventional non-Tregs and critical for T cell receptor (TCR)+ T cells to differentiate to Tregs in the thymus.3 Thus, functional deficiency of Tregs, an abnormality in Foxp3 expression and/or mutations in the Foxp3 gene locus results in increased susceptibility to several autoimmune diseases. Humans that have mutations in the Foxp3 gene develop IPEX (immune dysregulation, polyendocrinopathy and enteropathy, X-linked syndrome), a severe autoimmune disease, resulting in fatal lymphoproliferation at a very early age. A significantly reduced suppressive function of Tregs has been found in patients with polyglandular syndrome type II (APS-II),4 rheumatoid arthritis,5 type I diabetes,6 myasthenia graves,7, 8 and autoimmune hepatitis.9 Patients with multiple sclerosis not only lose functional suppression10 but also exhibit decreases in Foxp3 messenger RNA (mRNA) and protein expressions.11

Primary biliary cirrhosis (PBC) is a progressive autoimmune liver disease characterized by immunomediated destruction of intrahepatic small bile ducts.12 It has been suggested that the loss of Tregs plays a key role in susceptibility to PBC.13 For example, we have demonstrated a significant decrease in the frequency of Tregs in patients with PBC as well as daughters and sisters of PBC patients.14 We also reported a male child who developed liver dysfunction in association with serum anti–pyruvate dehydrogenase-E2 (PDC-E2) antibodies at the age of five, who had a complete deficiency of the alpha subunit of the interleukin (IL)-2 receptor (IL-2Rα, CD25) in peripheral blood lymphocytes.15 CD25 is a critical component for functional Treg development and expansion. Mice deficient in IL-2 receptor also demonstrated PBC-like lesions in the liver.16 In agreement with our findings, in Ae2-deficient mice that developed features resembling PBC, the proportion of CD4+CD25+FoxP3+ Tregs was reduced.17 Based on the above, we hypothesized that the genetic components essential for the development and homeostasis of Foxp3+ Tregs are responsible for the loss of self-tolerance in PBC. Hence, we took advantage of Scurfy (Sf) mice, which have a Foxp3 gene mutation that results in a deficiency of functional Tregs,18, 19 and examined levels of autoantibodies, cytokines, and liver histology. We report that Sf mice have serological, histological, and cytokine features characteristic of autoimmune cholangitis similar to patients with PBC.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Mice.

Heterozygous female B6.Cg-foxp3sf/X/J mice obtained from Jackson Laboratories (Bar Harbor, ME) were bred with male C57BL/6J mice to generate hemizygous foxp3sf/Y Sf mice. All animals were maintained in individually ventilated cages under specific pathogen-free conditions in our animal facility. Mice with the Foxp3 mutation were identified by polymerase chain reaction genotyping.19, 20 Due to the fact that Sf phenotype occurs in hemizygous males and XO females,21 hemizygous male mice with mutated Foxp3 gene (foxp3sf/Y) were used herein. Male mice with wild-type Foxp3 gene (foxp3X/Y) were used as controls. Because Foxp3-deficient male mice have 100% mortality by 4 weeks of age, all experiments were performed at 3 to 4 weeks of age. All animal experiments were approved by the Animal Care Committees of the Universities of California and Virginia.

Experimental Protocol.

Mice were left unmanipulated until 3 to 4 weeks of age. At that time all animals were sacrificed; 4 to 12 mice were used in each group. Sera were obtained for serum antibody and cytokine tests. Livers were used for isolating RNA to determine cytokine mRNA levels and for histological studies. Liver and spleens were used for flow cytometric analysis of lymphocytes. Other tissues have been previously studied by other groups.20, 22–25

Serum Antimitochondrial Antibodies.

The level of antimitochondrial antibodies (AMAs) was measured via enzyme-linked immunosorbent assay (ELISA), using recombinant PDC-E2 as the test antigen following previously described procedures.16, 26

Cytokine Analysis.

Concentrations of tumor necrosis factor (TNF-α), interferon-γ (IFN-γ), IL-6, IL-2, IL-4, IL-6, and IL-10 in sera were measured with the mouse inflammatory Cytometric Bead Array kit or the mouse Th1/Th2 cytokine Cytometric Bead Array kit (BD Biosciences, San Jose, CA). Serum IL-12p40 was measured using a mouse IL-12/IL-23 p40 allele-specific DuoSet ELISA development kit (R&D Systems, Minneapolis, MN). Serum IL-18 was measured using a mouse IL-18 ELISA kit (Medical & Biological Laboratories, Japan).

Total RNA was extracted using the QIAGEN RNeasy Mini Kit (Qiagen, Valencia, CA). One microgram of total RNA was reverse-transcribed and quantified on an ABI Prism 7900HT Sequence Detection System. Amplification was performed for 40 cycles in a total volume of 20 μL and products detected using SYBR Green. The relative expression level of each target gene was determined by normalizing its mRNA level to an internal control gene glyceraldehyde 3-phosphate dehydrogenase (Table 1).

Table 1. Real-Time Polymerase Chain Reaction Primers
IDTargetForwardReverse
    
1GADPHCATGGCCTTCCGTGTTCCTACCTGCTTCACCACCTTCTTGAT
2TNF-αAAGCCTGTAGCCCACGTCGTAAGGTACAACCCATCGGCTGG
3IFN-γTAGCCAAGACTGTGATTGCGGAGACATCTCCTCCCATCAGCAG
4IL-6TCCATCCAGTTGCCTTCTTGTTCCACGATTTCCCAGAGAAC
5IL-12p40GGAAGCACGGCAGCAGAATAAACTTGAGGGAGAAGTAGGAATGG
6IL-18GCCATGTCAGAAGACTCTTGCGTCGTACAGTGAAGTCGGCCAAAGTTGTC
7IL-10ATGCTGCCTGCTCTTACTGACTGCCCAAGTAACCCTTAAAGTCCTGC
8TGF-βAACAATTCCTGGCGTTACCTTCTGCCGTACAACTCCAGTGA
9IL-17AGCAAGAGATCCTGGTCCTGAAGCATCTTCTCGACCCTGAA
10IL-23CTTCTCCGTTCCAAGATCCTTCGGGCACTAAGGGCTCAGTCAGA
11T-betTGCCCGAACTACAGTCACGAACAGTGACCTCGCCTGGTGAAATG
12ROR-γtCCCACTGACCTTGAACCACTAGGAGGGTGTTGGTGAGATG

Flow Cytometry.

Mononuclear cells were isolated from spleen and liver tissues using density gradient centrifugation with Accu-Paque (density, 1.086). Anti-mouse CD16/32 (clone 93, Biolegend) were used for FcR blocking. Mononuclear cells were stained with fluorochrome-conjugate antibodies, including Alexa Fluor 750–conjugated anti–TCR-β (clone H57-597, eBiosciences), Alexa Fluor 647–conjugated anti-CD19 (clone eBio1 D3, eBiosciences), PerCP-conjugated anti-CD4 (clone RM4-5, Biolegend), fluorescein isothiocyanate–conjugated anti-CD8a (clone 53-6.7, Biolegend), and phycoerythrin–conjugated anti-NK1.1 (clone PK136; BD PharMingen, San Diego, CA). Stained cells were analyzed using a FACScan flow cytometer to allow for five-color analysis. Data were analyzed with CELLQUEST software (BD Bioscience).

Histochemistry of Liver.

Sections of liver tissue were immediately fixed with 10% buffered formalin at room temperature for 2 days, then embedded in paraffin and cut into 5-μm slices for staining with hematoxylin-eosin or for immunostaining. For phenotypic analysis of the lymphocytic filtrates, rat anti-mouse CD4 (1:30 dilution, clone L3T4; eBioscience, San Diego, CA), anti-mouse CD8 (1:30 dilution, Ly-2, clone 53.6.7, eBioscience) and anti-mouse IFN-γ (1:30 dilution, clone R4-6A2; Biolegend, San Diego, CA) antibodies were used for immunostaining as described in our previous publication.27

Statistical Analysis.

The data are expressed as the mean ± standard error of the mean. Two-sample comparisons were performed with two-sided unpaired t test. A P value of <0.05 was considered statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Sf Mice Develop PDC-E2 Specific AMAs.

Sf mice display significantly elevated immunoglobulin (Ig) G, IgA, and IgM reactivity against PDC-E2 than their littermate controls (IgG and IgM, P < 0.001; IgA, P < 0.01) (Fig. 1).

thumbnail image

Figure 1. Sf mice develop serum antibodies against mitochondrial antigens (AMA). Bars denote the mean of optical density (OD) values. B6, littermate controls. **P < 0.01; ***P < 0.001.

Download figure to PowerPoint

Increased Circulating Inflammatory Cytokines in Sf Mice.

The levels of inflammatory cytokines TNF-α, IFN-γ, IL-6, IL-12p40, and IL-18 were significantly elevated in the sera of Sf mice compared with their littermate controls (TNF-α, P < 0.001; IFN-γ, P < 0.05; IL-6, IL-12p40, and IL-18, P < 0.01) (Fig. 2). The mean serum concentration of IL-10 in Sf mice was higher than the control mice (27.0 ± 12.1 pg/mL versus 2.4 ± 2.4 pg/mL), although the difference was not statistically significant (P > 0.05).

thumbnail image

Figure 2. Concentration of inflammatory cytokines in sera of Sf mice (n = 10). *P < 0.05; **P < 0.01; ***P < 0.001.

Download figure to PowerPoint

Up-regulated Hepatic mRNA Levels for Inflammatory Cytokines and Related Transcription Factors in Sf Mice.

The inflammatory cytokines TNF-α, IFN-γ, and IL-6 were up-regulated 10- to 20-fold in Sf compared with control mice (IFN-γ, P < 0.05; TNF-α and IL-6, P < 0.001) (Fig. 2). The T helper 1 (Th1)-associated cytokine IL-12p40 gene was also significantly higher in the Sf mice than controls (P < 0.001). In contrast to the increased level of IL-18 in the serum of Sf mice than controls, such a difference was not seen in liver (data not shown). Transforming growth factor β (TGF-β) mRNA was expressed six-fold higher in Sf versus control mice (P < 0.01), while IL-17A and its key transcription factor retinoic acid-related orphan receptor γt were both elevated at similar levels in the livers of Sf mice (p < 0.01). A similar increase was also observed in Th1-specific T box transcription factor (T-bet) in Sf compared with control mice (P < 0.001). Interestingly, IL-23 mRNA was presented at an extremely high level in Sf livers, which was approximately 200- to 300-fold greater than that of control mice (P < 0.001, Fig. 3).

thumbnail image

Figure 3. Relative levels of mRNA for cytokines and related transcription factors in liver (Sf mice, n = 4; B6 littermate controls, n = 5). *P < 0.05; **P <0.01; ***P < 0.001.

Download figure to PowerPoint

Sf Mice Exhibit PBC-Like Portal Inflammation and Bile Duct Damage.

In Sf mice, clusters of lymphoplasma cells aggregated in the parenchyma and most portal tracts in association with bile duct damage (Fig. 4A-D). Necroinflammatory changes to various degrees were also observed in hepatic parenchyma (Fig. 4E,F). In the hepatic parenchyma of Sf mice, neutrophils, eosinophils, and histiocytes were also seen (Fig. 4G,H). Entrapped hepatocytes and interface hepatitis were observed in cellularly enlarged portal tracts, suggesting that portal inflammation progressed rapidly. Interlobular bile ducts were surrounded by inflammatory cells and showed degenerative changes (Fig. 4C-E). In several bile ducts, epithelial destruction with intraepithelial lymphoid infiltration was observed (Fig. 5A,B). These findings were similar to the chronic nonsuppurative destructive cholangitis aspects of human PBC. To highlight the shape of interlobular bile ducts among inflammatory cells, immunohistochemical staining with pan-keratin cocktails antibodies was performed (Fig. 5C,D). Scattered cytokeratin positive cells were observed in the portal infiltrates and direct bile duct destruction in 11 out of 12 liver tissues from Sf mice. Moreover, bile duct loss was identified in several portal tracts of Sf mice. Bile ducts disappeared in portal tracts where only proliferative bile ductules remain (Fig. 5D), similar to that observed in human PBC.

thumbnail image

Figure 4. Histological examinations of livers from Sf mice (hematoxylin-eosin staining). (A,B) Distinct inflammatory change (blue arrow) in most portal tracts and various degrees of necroinflammatory change in hepatic parenchyma (red arrow). (C-F) Lymphoplasmacytic infiltration in enlarged portal tracts. Various degrees of degenerative change were detected in (A-C) interlobular bile ducts and (D) large bile ducts surrounded by inflammatory lymphoid aggregates showing focal necrosis in hepatic parenchyma (red arrow). (G) Extramedullary hematopoiesis containing an erythroblastic island (red arrow). (H) Erythrophagocytosis by an activated Kupffer cell (red arrow). BD, bile duct.

Download figure to PowerPoint

thumbnail image

Figure 5. Small bile duct damage and bile duct loss in Sf mice. (A,B) Irregular interlobular bile ducts surrounded by inflammatory cells. Immunostaining with pan-keratin cocktail antibodies revealed (C) irregular and thick interlobular bile ducts and (D) disappearing interlobular bile ducts as well as proliferating bile ductules in the periphery of portal tract (blue arrows).

Download figure to PowerPoint

Predominance of CD8+ T Cells in Hepatic Lymphocytic Infiltrate Populations of Sf Mice.

To investigate the phenotypes of liver infiltrating lymphocytes, liver sections were stained for CD4, CD8, or IFN-γ. As shown in Fig. 6A, CD4+ cells distributed diffusely in portal tracts, and a few CD4+ cells were found around the periductal area. Aggregated lymphocytes in the hepatic parenchyma were also positive for CD4 staining. In contrast, CD8+ cells concentrated near bile ducts and interface areas, while the IFN-γ+ cells also aggregated near interlobular bile ducts.

thumbnail image

Figure 6. Phenotypic characterization of intrahepatic lymphocytes. (A) Immunohistological staining of liver sections from Sf mice with antibodies against CD4, CD8, or IFN-γ. (B) Flow cytometric analysis of CD4+ and CD8+ T cell populations in the lymphocytes isolated from liver and spleen. Displayed in the dot plots are cells gated on lymphocyte population (upper panels) or NK1.1TCR-β+ T cell population (lower panels). Numbers in the dot plots are percentages of NK1.1TCR-β+ T cells out of the gated lymphocyte population (upper panels) or percentages of cells in the specific quadrants out of the gated T cell population (lower panels).

Download figure to PowerPoint

The frequency of CD4+ and CD8+ T cells in the liver infiltrates and the spleens were further assessed by flow cytometry (Fig. 6B). In both the liver and the spleen, the percentage of TCRβ+ T cells in total lymphocytes was substantially higher in the Sf mice than controls. The liver infiltrating T cells from the B6 normal controls were predominantly CD4+ T cells; in contrast, those derived from the Sf mice were predominantly CD8+ T cells. A similar prevalence of CD8+ T cells was also observed in the splenocytes of Sf mice.

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Although the etiology of PBC remains enigmatic, there have been several studies that suggest a role for environmental factors in initiating loss of tolerance superimposed on a genetic predisposition.13, 28–33 Previously, we reported that the frequency of CD4+CD25+ Tregs was reduced in PBC patients and their first-degree relatives.14 In addition, we demonstrated that PBC-like liver lesions developed in a CD25-deficient child15 and in CD25-deficient mice,16 suggesting a role of Treg deficiency in the induction of PBC. In the present study, we took advantage of the Sf mouse model with a mutated Foxp3 gene and performed a systematic analyses with serological, histological, flow cytometric, and molecular approaches for the consequence of disrupting this key functional protein of Tregs. We demonstrate that at a very early stage of life, Sf mice spontaneously develop characteristics of human PBC (Table 2).

Table 2. Human PBC Comparisons with Sf Mice
 Human PBCSf Mice
  1. −, none; +, mild; ++, moderate; +++, severe.

   
Serum autoantibodies  
 AMA90%-95%100%
Liver histology  
 Portal lymphoid infiltrates+++++ to +++
 Biliary ductal destruction+ to ++++ to ++
 Granuloma+ to ++
 Interface hepatitis++
Liver immunohistochemistry  
 CD4 cell++
 CD8 cell++++
 B cell++

Evidence indicates a genetic predisposition to PBC. First, family members of a PBC patient have up to a 100-fold higher risk of developing PBC. Second, the concordance rate of PBC in genetically identical twin sets is 0.63.34 Third, a higher frequency of monosomy X in women with PBC suggests that specific X chromosome-linked genes for immunoregulatory proteins are critical for PBC.13, 35 The disorder in immune regulation of the Sf mouse is caused by a recessive mutation of the Foxp3 gene on the X chromosome. Several other mouse models with defined or undefined mutations resulting in spontaneously developed PBC-like liver diseases have been reported, including the role of xenobiotics.27, 36–39 Characterization of the common phenotypes of these models will be critical to elucidate individual mechanisms that lead to autoimmune cholangitis.

The specificity of pathological changes localized to the bile ducts, the presence of lymphoid infiltration in the portal tracts, and the readily detectable expression of major histocompatibility complex antigens on the biliary epithelium suggest that autoantigen-specific T cell responses are directed against biliary epithelial cells. Our laboratory has accumulated substantial data suggesting that the destruction of biliary cells is mediated by liver-infiltrating autoreactive T cells, especially cytotoxic CD8 T cells highly enriched in PBC liver.40, 41 In another mouse model of PBC, CD8 T cells comprised a majority of lymphocytic infiltrates in diseased liver and were sufficient for inducing bile duct destruction after adoptive transfer.42 Consistent with these previous studies, we demonstrate in the present study that CD8+ cells are the predominant T cell subset in liver infiltrating lymphocytes, which aggregate near bile ducts and interface areas at high frequencies. Taken together, our data suggest that the lack of Foxp3 protein results in abnormal Treg function, which is responsible for the loss of tolerance in the liver leading to autoreactive CD8 T cell–mediated bile duct damage.

Although AMA is detected in 90% to 95% of PBC patients and is considered a humoral hallmark of PBC, the pathological role of AMA has not been established. Similar to other previously described murine models for PBC with different underlying gene mutations,36 in Sf mice the PBC-like liver disease is again associated with the presence of serum AMAs. It is possible that AMA is an active player in the pathological process of bile duct destruction. A potential mechanism is the formation of AMA-autoantigen immune complexes, which are efficiently presented by DCs or biliary epithelial cells to activate autoreactive cytotoxic CD8 T cells.40, 43 Alternatively, AMA could be a mere marker for liver damage resulting from diverse etiologies ranging from autoimmune diseases to viral infection. We note that elevated serum IgM has already been reported in the original description of Sf mice and therefore sera immunoglobulin levels were not studied herein.18

Increased expression levels of IFN-γ, TNF-α, and IL-6 have been demonstrated in the livers of PBC patients.44–46 In the present study, we observed elevated hepatic expression of cytokine genes including TNF-α, IFN-γ, IL-6, IL-12, TGF-β, IL-17, and IL-23, indicating the involvement of IFN-γ-secreting Th1 and IL-17–secreting Th17 cells in liver damage. Previous studies have linked abnormal expression of IL-12 with expansion of naïve T cells and autoreactive T cells leading to the breakdown of self-tolerance and exacerbation of autoimmune pathology.47, 48 IL-12 is primarily produced by antigen-presenting cells such as activated dendritic cells, monocytes, and macrophages. Increased expression of IL-12 was found in serum, cerebrospinal fluid, and plaques of multiple sclerosis patients49–52 and in synovial tissue of rheumatoid arthritis patients.53 In mice, high levels of IL-12 were associated with a number of experimental autoimmune disease models.54, 55 We also observed similar results in previous studies of murine models for PBC, including IL-2Rα–deficient mice16 and TGF-β receptor II dominant-negative mice.56 Here we report that in Sf mice, the PBC-like serological and pathological alteration is associated with elevated levels of serum IL-12 and in particular elevated levels of IL-12p40 mRNA transcripts in the liver. It will be important to further investigate whether the pathogenic process of bile duct destruction is directly induced by IL-12 or by specific immune cell subsets such as autoreactive CD8 T cells that are activated by IL-12.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We thank Willy M. Hsu for AMA detection, Yuki Moritoki for valuable suggestions, and Thomas P. Kenny for technical support. We are grateful to Nikki Phipps for manuscript preparation.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References