Steatohepatitis Impairs T-cell-Directed Immunotherapies Against Liver Tumors in Mice - PubMed (original) (raw)

. 2021 Jan;160(1):331-345.e6.

doi: 10.1053/j.gastro.2020.09.031. Epub 2020 Oct 1.

Zachary J Brown 2, Laurence P Diggs 3, Mathias Vormehr 4, Chi Ma 1, Varun Subramanyam 1, Umberto Rosato 1, Benjamin Ruf 1, Juliane S Walz 5, John C McVey 1, Simon Wabitsch 1, Qiong Fu 1, Su Jong Yu 6, Qianfei Zhang 1, Chunwei W Lai 7, Ugur Sahin 8, Tim F Greten 9

Affiliations

Steatohepatitis Impairs T-cell-Directed Immunotherapies Against Liver Tumors in Mice

Bernd Heinrich et al. Gastroenterology. 2021 Jan.

Abstract

Background & aims: Nonalcoholic steatohepatitis causes loss of hepatic CD4+ T cells and promotes tumor growth. The liver is the most common site of distant metastases from a variety of malignancies, many of which respond to immunotherapy. We investigated the effects of steatohepatitis on the efficacy of immunotherapeutic agents against liver tumors in mice.

Methods: Steatohepatitis was induced by feeding C57BL/6NCrl or BALB/c AnNCr mice a methionine and choline-deficient diet or a choline-deficient l-amino acid-defined diet. Mice were given intrahepatic or subcutaneous injections of B16 melanoma and CT26 colon cancer cells, followed by intravenous injections of M30-RNA vaccine (M30) or intraperitoneal injections of an antibody against OX40 (aOX40) on days 3, 7, and 10 after injection of the tumor cells. We measured tumor growth and analyzed immune cells in tumor tissues by flow cytometry. Mice were given N-acetylcysteine to prevent loss of CD4+ T cells from liver.

Results: Administration of M30 and aOX40 inhibited growth of tumors from intrahepatic injections of B16 or CT26 cells in mice on regular diet. However, M30 and/or aOX40 did not slow growth of liver tumors from B16 or CT26 cells in mice with diet-induced steatohepatitis (methionine and choline-deficient diet or choline-deficient l-amino acid-defined diet). Steatohepatitis did not affect the ability of M30 to slow growth of subcutaneous B16 tumors. In mice with steatohepatitis given N-acetylcysteine, which prevents loss of CD4+ T cells, M30 and aOX40 were able slow growth of hepatic tumors. Flow cytometry analysis of liver tumors revealed reduced CD4+ T cells and effector memory cells in mice with vs without steatohepatitis.

Conclusions: Steatohepatitis reduces the abilities of immunotherapeutic agents, such as M30 and aOX40, to inhibit tumor liver growth by reducing tumor infiltration by CD4+ T cells and effector memory cells. N-acetylcysteine restores T-cell numbers in tumors and increases the ability of M30 and aOX40 to slow tumor growth in mice.

Keywords: Anti-Tumor Immune Response; Metastasis; NAFLD; NASH.

Published by Elsevier Inc.

PubMed Disclaimer

Conflict of interest statement

Conflict of interest: The authors declare no competing interest. Some of the authors are employees at BioNTech AG (Mainz, Germany) as mentioned in the affiliations. U.S. is stock owner of BioNTech AG (Mainz, Germany). U.S., M.V. are inventors on patents and patent applications, which cover parts of this article.

Figures

Figure 1:

Figure 1:. M30-RNA vaccine controls tumor growth of B16 melanoma liver tumors

A. Experimental setup: Mouse model of B6(Cg)-Tyrc-2J/J albino mice with intrahepatic injection of B16 luciferase-expressing melanoma cells and RNA vaccination (at indicated time points). B. Tumor growth of intrahepatic B16 tumors using bioluminescence imaging. Representative images of 4 mice per group are shown. 2 groups: control (n=7), M30 RNA-vaccination (n=8), both groups regular diet. C. Quantification of in vitro bioluminescence assay displayed as tumor growth curve. D. Tumor burden (same groups as in B) at d27 presented as tumor-to-liver weight-ratio. E. Representative photos of intrahepatic melanoma tumors. 2 groups: control (upper panel), M30 RNA-vaccination (lower panel), both reg diet.

Figure 2:

Figure 2:. M30-RNA vaccine loses efficacy for B16 liver tumors in mice with MCD diet-induced steatohepatitis but not in subcutaneous tumors.

A. Experimental setup: Mouse model of B6(Cg)-Tyrc-2J/J albino mice and MCD diet-induced steatohepatitis with intrahepatic injection of B16 luciferase-expressing melanoma cells and RNA vaccination (at indicated time points). B. Tumor growth of intrahepatic B16 tumors presented as bioluminescence imaging showing radiance of liver tumors over time. 4 groups: control - MCD diet, M30 -MCD diet, control – reg. diet, M30 – reg. diet. C. Quantification of in vitro bioluminescence assay displayed as tumor growth curve. Data represent D. Tumor burden at d20 presented as tumor-to-liver weight-ratio. E. Representative pictures of intrahepatic melanoma tumors). 4 groups: Control-MCD diet, M30-MCD diet, Control – reg. diet, M30 - reg. diet. F. Experimental setup (left): M30-vaccine in C57BL/6 mice fed MCD diet and given s.c. injection of B16 melanoma cells and RNA vaccination (at indicated time points). G. Subcutaneous tumor model. Tumor growth over time shown as tumor volume over time. H. Subcutaneous tumor model. Weight of subcutaneous tumors comparing groups: control RNA, M30 (both MCD diet).

Figure 3:

Figure 3:. M30-RNA vaccine also loses efficacy for B16 liver tumors in mice with a second diet-induced steatohepatitis model (CDAA diet).

A. Experimental setup: Mouse model of CDAA diet-induced steatohepatitis. B6(Cg)-Tyrc-2J/J albino mice were given CDAA diet for 22 weeks before intrahepatic injection of B16 luciferase-expressing melanoma cells and RNA vaccination (at indicated time points). B. Mouse weight of B6(Cg)-Tyrc-2J/J albino mice at day of tumor inoculation after 22 weeks of reg diet vs CDAA diet (both n=17). C. Tumor growth presented as bioluminescence imaging of intrahepatic B16 tumors showing radiance of liver tumors over time (left). 4 groups: Control RNA - CDAA diet, M30 - CDAA diet, control RNA – reg. diet, M30 – reg. diet. D. Quantification of in vitro bioluminescence assay displayed as tumor growth curve (right). Data represent E. Tumor burden at d20 presented as tumor-to-liver weight-ratio. F. Representative pictures of intrahepatic melanoma tumors. 4 groups: Control RNA - MCD diet, M30 - MCD diet, control RNA– reg. diet, M30 – reg - diet.

Figure 4:

Figure 4:. NAC administration rescues M30-RNA vaccine efficacy for B16 liver tumors in mice with steatohepatitis.

A. Experimental setup: Mouse model of B6(Cg)-Tyrc-2J/J albino mice and MCD diet-induced steatohepatitis ± NAC in the drinking water. Mice received intrahepatic injection of B16 luciferase-expressing melanoma cells and RNA vaccination at indicated time points. B. Quantification of in vitro bioluminescence assay displayed as tumor growth curve. C. Tumor growth presented as bioluminescence imaging of intrahepatic B16 tumors showing radiance of liver tumors over time. 4 groups: (all MCD diet): control RNA (n=6), M30 (n=6), control RNA + NAC (n=7), M30 + NAC (n=6). D. Tumor burden at d20 presented as tumor-to-liver weight-ratio. E. Representative pictures of intrahepatic melanoma tumors (right). Groups as in B. F. NAC-rescue of M30 vaccine efficacy is CD4+ T cell dependent. Experimental setup as in A. An additional group received anti-CD4-depleting antibodies. 3 groups: (all MCD diet): M30+H2O (n=7), M30+NAC (n=7), M30+NAC+anti-CD4 (n=6). Tumor burden at d20 presented as tumor-to-liver weight-ratio.

Figure 5:

Figure 5:. CD4+ and CD8+ T cell-dependent agonistic anti-OX40 immunotherapy loses therapeutic efficacy for CT26 liver tumors in mice with steatohepatitis but not in subcutaneous tumors.

A. Experimental setup: BALB/c AnNCr mice were fed MCD diet (+/− NAC in the drinking water) to induce steatohepatitis or regular diet. For depletion experiments, some groups received anti-CD4 or anti-CD8-depleting antibodies prior to tumor inoculation and then once weekly. On d0 mice were injected intrahepatically with CT26 murine colon carcinoma cells. Injection with aOX40 antibody or IgG-Control i.p. was performed at indicated time points. B. Tumor burden at d20 in CT26 tumor model presented as tumor-to-liver weight-ratio. Mice fed regular diet and injected with aOX40 or control antibody (n=9 IgG/regular diet; n=9 aOX40/regular diet). C. Tumor burden at d20 in CT26 tumor model presented as tumor-to-liver weight-ratio. MCD diet fed mice injected with aOX40 or control antibody (n=9 IgG/MCD diet; n=9 aOX40/MCD diet). D. Representative pictures of intrahepatic CT26 tumors are shown. E. Tumor burden at d20 in CT26 tumor model with aOX40 and NAC. Mice were fed MCD diet and given NAC in drinking water and injected intrahepatically with CT26. (All groups MCD diet. n=7 IgG/water; n=7 aOX40/water; n=10 IgG/NAC; n=11 aOX40/NAC). F. Representative pictures of intrahepatic CT26 tumors are shown. G. NAC-rescue of aOX40 is CD4+ and CD8+ T cell dependent. Experimental setup as in C. An additional group received anti-CD4 or anti-CD8-depletion antibody (starting 1 week prior to tumor inoculation, then once weekly), respectively. 4 groups: (all MCD diet): IgG/H2O (n=5) aOX40/ H2O (n=5), aOX40/NAC + anti-CD4 (n=8), aOX40/NAC + anti-CD8 (n=8). H. CT26 subcutaneous model. BALB/c AnNCr mice were fed MCD diet to induce steatohepatitis prior to s.c. injection of CT26 cells. Mice received agonistic aOX40 antibody on d3, d7, d10. Tumor growth shown as tumor volume over time. I. Weight of subcutaneous CT26 tumors (as in E) at d14 (left) comparing groups IgG control (n=5), aOX40 (n=5) (both MCD diet). J. Representative pictures of s.c. CT26 tumors after excision are shown. Two animals in the aOX40 group had no detectable tumors.

Figure 6:

Figure 6:. Steatohepatitis causes a selective loss of intrahepatic CD4+ T lymphocytes and abrogates M30 RNA-vaccine mediated changes in the liver tumor microenvironment.

A. Effect of M30-RNA vaccine (n=8) vs control RNA (n=7, both regular diet) on the frequency of intrahepatic CD4+ and CD8+ T cells in a mouse model of B16 liver tumors (experimental setup Fig. 1A). B. Effect of M30-RNA vaccine (n=8) vs control RNA (n=7, both regular diet) on the frequency of Treg cells in a mouse model of B16 liver tumors. C-E. Effect of MCD diet on M30-RNA vaccine-mediated changes of CD4+ T cell numbers presented as cells per gram tissue in the liver (C), tumor (D) and spleen (E). Experimental setup as in Fig. 2A. Four groups: control RNA - MCD diet, M30 - MCD diet, control RNA – reg. diet, M30 – reg. diet. F-H. Effect of NAC administration on M30-RNA vaccine-mediated changes of CD4+ T cell numbers in the liver (F), tumor (G) and spleen (H) in mice fed with MCD diet. Experimental setup as in Fig. 4A. 4 groups (all MCD diet): control RNA (n=6), M30 (n=6), control RNA + NAC (n=7), M30 + NAC (n=5). I-L. Effect of CDAA diet on M30-RNA vaccine-mediated changes of CD4+ T cell numbers in the liver (I), tumor (K) and spleen (L). Experimental setup as in Fig. 3A. Four groups: control RNA - CDAA diet, M30 - CDAA diet, control RNA – reg. diet, M30 - reg diet.

Figure 7:

Figure 7:. The effect of diet-induced steatohepatitis on intrahepatic and tumor-infiltrating T cell subsets in CT26 liver tumor-bearing mice receiving aOX40.

A-B. Effect of MCD diet on aOX40-mediated changes of CD4+ T cell numbers in the liver(A) and tumor(B). Experimental setup as in Fig. 5A. 4 groups: IgG control -MCD diet (n=19), aOX40 - MCD diet (n=13), IgG control – reg. diet (n=16), aOX40 – reg. diet (n=21). C-D. Effect of NAC on aOX40-mediated changes of tumor CD4+ (C) and CD8+ T cells (D) in mice fed MCD diet. Experimental setup as in Fig. 5A. (All groups MCD diet. n=7 IgG/water; n=7 aOX40/water; n=10 IgG/NAC; n=11 aOX40/NAC). E-F. Effect of MCD diet on aOX40-mediated changes of CD4+ (E) and CD8+ effector memory T cell numbers (F) in the tumor. Experimental setup as in Fig. 5A, groups as in A. Numbers of tumor-infiltrating CD4+ effector memory (CD3+CD4+CD44+CD62L−) cells (E) and CD8+ effector memory (CD3+CD4+CD44+CD62L−) cells (F) are shown as cells per gram tumor. G-H. Effect of NAC on aOX40-mediated changes of liver (G) and tumor (H) CD4+ effector memory T cells in mice fed MCD diet. Experimental setup as in Fig. 5A. (All groups MCD diet. n=7 IgG/water; n=7 aOX40/water; n=10 IgG/NAC; n=11 aOX40/NAC).

References

    1. Younossi Z, Anstee QM, Marietti M, et al. Global burden of NAFLD and NASH: trends, predictions, risk factors and prevention. Nature Reviews Gastroenterology & Hepatology 2018;15:11–20. - PubMed
    1. Anstee QM, Reeves HL, Kotsiliti E, et al. From NASH to HCC: current concepts and future challenges. Nature Reviews Gastroenterology & Hepatology 2019;16:411–428. - PubMed
    1. Estes C, Razavi H, Loomba R, et al. Modeling the epidemic of nonalcoholic fatty liver disease demonstrates an exponential increase in burden of disease. Hepatology 2018;67:123–133. - PMC - PubMed
    1. Younossi ZM, Koenig AB, Abdelatif D, et al. Global epidemiology of nonalcoholic fatty liver disease-Meta-analytic assessment of prevalence, incidence, and outcomes. Hepatology 2016;64:73–84. - PubMed
    1. Disibio G, French SW. Metastatic patterns of cancers: results from a large autopsy study. Arch Pathol Lab Med 2008;132:931–9. - PubMed

Publication types

MeSH terms

LinkOut - more resources