CD8 T Cell Hyperfunction and Reduced Tumour Control in Murine Models of Advanced Liver Disease

Abstract

Immune dysfunction in liver disease contributes to significant morbidities, depending on liver damage severity and aetiology. We previously reported long-lasting generalized CD8 T cell hyperfunction in chronic HCV infection with advanced fibrosis, yet its separation from viral and fibrosis-driven effects, as well as clinical outcomes of advanced fibrosis, remains unclear. In a murine model of carbon tetrachloride-induced progressive liver fibrosis, advanced fibrosis was observed by 12 weeks, with pathologies similar to those of human chronic HCV infection. Blood-circulating CD8 T cells showed IFN-γ and granzyme B (GrB) hyperfunction in response to anti-CD3/28 stimulation, as well as impaired responses to ectopic tumour challenge and anti-PD-1/CTLA-4 immunotherapy. Hyperfunction and impaired tumour responses were retained despite liver insult cessation. In a 45% HFD model, which induced steatosis and minimal fibrosis, IFN-γ and GrB hyperfunction was also observed in blood-circulating CD8 T cells. This study highlights a prolonged systemic CD8 T cell dysfunction acquired during progressive liver disease, associated with impaired antitumour and immunotherapy responses. These mirror the bulk CD8 T cell dysfunction observed in advanced liver diseases in humans, suggesting that these models could be valuable for future mechanistic studies aimed at identifying targets to help improve clinical outcomes in chronic liver disease.

Keywords: CD8 T cells; immunotherapy; liver fibrosis; steatosis; tumour response.

FIGURE 1

Advanced liver fibrosis is induced in C57BL/6 mice by prolonged CCl₄ exposure. (A) Mice were administered CCl₄ or olive oil (vehicle control) for ≈12 weeks, and liver tissue and blood were collected to assess liver disease severity and CD8 T cell function. (B) Representative Masson's trichrome (top) and hematoxylin & eosin (bottom) staining of mouse liver tissues at 7 and 12 weeks into CCl₄ exposure, where progressive liver fibrosis and inflammation development was observed and respective pathologist‐determined scores are indicated. Yellow arrows highlight areas of fibrosis (top) or lobular inflammation (bottom). Scale bars: 200 µm. Summary of (C) liver fibrosis and (D) inflammation severity induced by ≈12 weeks of CCl₄ exposure across three independent experiments are shown. Fibrosis scores were determined using the Metavir scoring system (F0 no fibrosis, F0‐1 minimal, F2‐3 intermediate, F3‐4 advanced fibrosis, F4 cirrhosis) while inflammation was scored across stages (I0‐I3). Study group sizes may range from 3–10, depending on the experiment. Comparisons by unpaired Student's t‐test, *p ≤ 0.05, **p ≤ 0.01.

FIGURE 2

Mice with CCl₄‐induced advanced liver fibrosis exhibit CD8 T cell hyperfunction. CD8 T cell function was assessed after anti‐CD3/28 stimulation of PBMCs from mice with advanced liver fibrosis induced by ≈12 weeks of CCl₄ exposure, compared with oil control mice. Data from several independent experiments are shown in the figure panels. Hyperfunction in bulk CD8 T cells was observed for both (A) GrB and (B) IFN‐γ expression. The remaining figure panels display results from CD8 T cells stimulated with anti‐CD3/CD‐28 antibodies, comparing Oil controls to CCl₄‐treated mice: (C) Increased proportions of poly‐functional (GrB⁺IFN‐γ⁺, GrB⁺CD107a⁺, GrB⁺IFN‐γ⁺CD107a⁺) CD8 T cells were also observed. This hyperfunction was seen in (D) T_N, T_E/EM, and T_CM subsets for GrB expression, and (E) T_E/EM subset for IFN‐γ expression. (F) The expression of PD‐1 on CD8 T cells remained unchanged. (G) Increased plasma levels of inflammatory cytokines were observed in mice with CCl₄‐induced advanced fibrosis. Comparisons by unpaired Student's t‐test, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001.

FIGURE 3

Advanced liver fibrosis regresses after the cessation of CCl₄ exposure. (A) At peak CCl₄‐induced liver fibrosis, mice underwent either (1) continued (Cont.) CCl₄ administration, or (2) stopped (Stop) CCl₄ treatment. (B) Representative Masson's trichrome (top) and hematoxylin & eosin (bottom) staining of mouse liver tissues after continued or ceased CCl₄ treatment. Yellow arrows highlight areas of fibrosis. Scale bars: 200 µm. (C) Summary of liver fibrosis severity after 16 weeks of continuous CCl₄ exposure, or after 12 weeks of CCl₄ exposure followed by 4 weeks of exposure cessation (Metavir scoring: F0 no fibrosis, F0‐1 minimal, F2‐3 intermediate, F3‐4 advanced fibrosis, F4 cirrhosis). Multiple comparisons by 1‐way ANOVA with Dunnett's post‐test, **p ≤ 0.01.

FIGURE 4

Bulk CD8 T cell hyperfunction is not fully reversed despite regression of advanced liver fibrosis. CD8 T cell function was assessed in mice after 12 weeks of CCl₄ exposure, followed by 4 weeks of either CCl₄ cessation or continuation. (A) The proportion of CD107a⁺ CD8 T cells is significantly higher in CCl₄‐treated mice with both continued or stopped exposure, with (B) hyperfunction more prominently observed in male mice. By contrast, (C) GrB hyperfunction in CD8 T cells was not sustained after CCl₄ cessation. M: Males, F: Females. (D) Both GrB hyperfunction and functional regression were more prominently observed in female mice compared with male mice. (E) IFN‐γ expression, while not statistically significant between groups, followed a similar increased and persistent function pattern as CD107a, including when stratified by sex (panel F). Multiple comparisons by 1‐way ANOVA with Dunnett's post‐test, *p ≤ 0.05, **p ≤ 0.01.

FIGURE 5

Ectopic tumour growth is exacerbated and response to checkpoint inhibitor therapy is delayed in mice with CCl4‐induced advanced fibrosis. (A) Mice with CCl₄‐induced advanced fibrosis were administered MC38 tumour cells, followed by anti‐PD‐1 and anti‐CTLA‐4 immunotherapy after tumour growth. (B) A low dose of MC38 cells (1×10⁵) resulted in CCl₄‐treated mice responding significantly poorer to tumour growth compared with oil controls. (C) Higher doses of MC38 cells (1 × 10⁶) lead to comparable tumour growth in CCl₄‐treated and control mice, where CCl₄‐treated mice exhibited delayed and weakened responses to tumour growth under anti‐PD‐1 and anti‐CTLA‐4 immunotherapy. (D) At peak liver fibrosis, mice were put on continued or ceased CCl₄ treatment for an additional 4 weeks, prior to MC38 tumour challenge (1 × 10⁵ cells). (E) Impaired tumour responses observed in CCl₄‐treated mice were not fully reversed after fibrosis regression. Multiple comparisons by 2‐way ANOVA with Šídák's post‐test, *p≤0.05, **p≤0.01.

FIGURE 6

High‐fat diet induces CD8 T cell hyperfunction associated with liver steatosis and ballooning in the absence of significant fibrosis. (A) C57BL/6 mice were placed on a 45% HFD for 14 weeks prior to blood collection and liver collection. (B) Both female (left) and male (right) mice under HFD significantly gained weight compared with chow‐fed controls. (C) Both female (left) and male (right) mice significantly gained percent fat mass when assessed at ≈11 weeks of HFD compared with chow‐fed mice. (D) Mice fed HFD for 14 weeks developed steatotic liver disease with fibrosis (F), steatosis (S), ballooning (B), and inflammation (I). (E) Summarized steatotic liver disease scores across all mice, with significantly increased steatosis and ballooning induced by HFD. (F) Bulk CD8 T cell hyperfunction in HFD‐fed mice compared with controls, as observed by increased IFN‐γ (left) and GrB (right) expression. (G) GrB hyperfunction was predominantly observed in T_E/EM (left) and T_CM (right) subsets in mice fed HFD. Body weight comparisons by two‐way ANOVA with Šídák's post‐test; fat mass, disease scores, and T cell function comparisons by unpaired Student's t‐test; *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001.