An inducible transgenic mouse model for immune mediated hepatitis showing clearance of antigen expressing hepatocytes by CD8+ T cells

Abstract

The liver has the ability to prime immune responses against neo antigens provided upon infections. However, T cell immunity in liver is uniquely modulated by the complex tolerogenic property of this organ that has to also cope with foreign agents such as endotoxins or food antigens. In this respect, the nature of intrahepatic T cell responses remains to be fully characterized. To gain deeper insight into the mechanisms that regulate the CD8+ T cell responses in the liver, we established a novel OVA_X_CreER(T2) mouse model. Upon tamoxifen administration OVA antigen expression is observed in a fraction of hepatocytes, resulting in a mosaic expression pattern. To elucidate the cross-talk of CD8+ T cells with antigen-expressing hepatocytes, we adoptively transferred K(b)/OVA257-264-specific OT-I T cells to OVA_X_CreER(T2) mice or generated triple transgenic OVA_X CreER(T2)_X_OT-I mice. OT-I T cells become activated in OVA_X_CreER(T2) mice and induce an acute and transient hepatitis accompanied by liver damage. In OVA_X_CreER(T2)_X_OT-I mice, OVA induction triggers an OT-I T cell mediated, fulminant hepatitis resulting in 50% mortality. Surviving mice manifest a long lasting hepatitis, and recover after 9 weeks. In these experimental settings, recovery from hepatitis correlates with a complete loss of OVA expression indicating efficient clearance of the antigen-expressing hepatocytes. Moreover, a relapse of hepatitis can be induced upon re-induction of cured OVA_X_CreER(T2)_X_OT-I mice indicating absence of tolerogenic mechanisms. This pathogen-free, conditional mouse model has the advantage of tamoxifen inducible tissue specific antigen expression that reflects the heterogeneity of viral antigen expression and enables the study of intrahepatic immune responses to both de novo and persistent antigen. It allows following the course of intrahepatic immune responses: initiation, the acute phase and antigen clearance.

Figure 1. Schematic presentation of antigen induction and its expression levels in the mouse models.

A. Schematic presentation of the antigen expression system in the mouse models. The OVA cassette is integrated in antisense orientation (OFF-state ‘1’, red) in the ROSA26 locus. The cassette is activated through an Albumin driven expression of CreER^T2 in hepatocytes by a single application of TAM. Upon binding of TAM, the CreER^T2 fusion protein translocates to the nucleus and can invert the OVA cassette (ON-state, green). In the presence of TAM, CreER^T2 continuously flips the OVA cassette and the antigen is stochastically ‘ON’ or ‘OFF’ (‘2’, see also 16). Upon TAM clearance the recombination status becomes fixed (‘3’). Assuming that the activated Cre recombinase is active in all hepatocytes, 50% of cells carry the cassette in the ON-state. The binding sites of primers 1a and 4a used to amplify the OVA cassette in the ON-state are indicated. B. OVA mRNA expression levels in the liver of OVA_X_CreER^T2 mice. OVA_X_CreER^T2 were treated with a single dose of TAM. 9 days later, the liver was isolated and RNA was subjected to qRT-PCR using the indicated primers. After TAM clearance 50% of all hepatocytes remain in ON-state. The mRNA levels in the non-treated OVA_X_CreER^T2 mice are 5 fold lower than those from the induced state or the standard. This corresponds to approximately 10% of hepatocytes that express the OVA antigen. OVA_X_Cre mice were used as a standard. Due to the constitutive Cre activity in these mice 50% of all cassettes are supposed to be in the ON-state. The OVA expression levels are related to albumin. The bars reflect samples from 4–8 animals that were individually analyzed.

Figure 2. Antigen-driven liver infiltrations and fate of adoptively transferred OT-I T cells.

Adoptive transfer of OT-I CD8+ cells to the OVA_X_CreER^T2 mice either pre-treated with TAM (day -30) or untreated was performed. For analysis mice were sacrificed at day +3 post adoptive transfer. A. Accumulation of NPCs and OT-I CD8+ T cells in the liver at day +3 post adoptive transfer (n=6-10). B. T cells infiltrating the liver tissue post adoptive transfer. T cells were visualized by indirect immunohistochemical staining using anti-CD3 antibody (brown). Data from representative mice are shown. C. Proliferation of OT-I CD8+ T cells quantified by CFSE dilution. Cells were re-isolated on day +3 post transfer. Histograms from representative mice are shown. D.T cell receptor down-regulation on OT-I CD8+ T cells as determined by staining with the PE labelled OVA antigen specific tetramers (n=4). E. Liver damage assessed by ALT activity measured in blood plasma of OVA_X_CreER^T2 mice. ALT activity below 40 U/L is considered as physiological, dashed line (n=3-10), # - not done. F. Elimination of OVA antigen expressing hepatocytes. ‘ON-state’ OVA mRNA expression in liver samples quantified by qRT-PCR (n=5-8). As control, 4 single transgenic OVA or CreER^T2 animals were treated with TAM and adoptively transferred with OT-I cells.

Figure 3. Induction of hepatocyte-specific antigen expression in OVA X CreER^T2 X OT-I mice.

A. Survival of OVA_X_CreER^T2_X_OT-I mice upon treatment with TAM (n=48). B. Determination of liver damage assessed by ALT activity. 3-16 animals were analysed per group, statistical significance calculated for groups n≥6. C. Histology analysis of liver tissue sections 2 days post TAM application. Sections were stained for T cells (anti-CD3) and apoptotic cells (anti-Caspase 3, brown). D. Accumulation of NPCs and OT-I CD8 T cells in the liver at days 7 and 81 post TAM application (n=3-4).

Figure 4. Elimination and re-induction of OVA expressing hepatocytes in OVA_X_OT-1 mice.

OVA_X_CreER^T2_X_OT-I mice were induced with TAM and monitored for OVA antigen expression (qRT-PCR, A.) and liver damage (ALT levels, B.). One experimental group was re-induced with TAM at day 80 (n=3).

Figure 5. Expression of exhaustion markers in induced OVA X CreERT2 X OT-I mice.

A. PD-1 and B. LAG-3 expression on hepatic OT-I CD8 T cells assessed by FACS analysis prior and post TAM (n=3-4).