Abrogation of TNFα production during cancer immunotherapy is crucial for suppressing side effects due to the systemic expression of IL-12

Abstract

For more than a decade, the cytokine interleukin-12 (IL-12) has been utilized, either alone or in combination with other drugs, as a treatment for cancer. The numerous anti-tumor properties of IL-12 still generate interest in the clinical use of this cytokine, even though it has demonstrated toxicity when administrated systemically. As an approach to overcome this toxicity, numerous laboratories have attempted to induce IL-12 expression at the site of the tumor. However for tumors that are difficult to remove surgically or for the treatment of disseminated metastases, systemic expression of this cytokine still remains as the most efficient method of administration. Nevertheless, finding alternative approaches for the use of IL-12 in the treatment of cancer and unraveling the basis of IL-12-side effects remain a challenge. In the present work we demonstrate that systemic expression of IL-12 through hydrodynamic injection of IL-12 cDNA is able to induce different types of liver lesions associated with a toxic pathology. However we report here that hepatic toxicity is diminished and survival of mice enhanced in the absence of tumor necrosis factor alpha (TNFα). This observation is in contrast to several murine models and clinical trials that postulate interferon gamma (IFNγ) as the main cytokine responsible for IL-12 toxicity. Moreover, our work demonstrates that when IL-12 cDNA is co-injected with IL-18 cDNA or when mice are pre-treated with a low dose of IL-12 cDNA prior to receiving a high dose of IL-12 cDNA, systemic levels of TNFα are almost completely abrogated, resulting in improved survival and less hepatic damage. Importantly, abrogation of TNFα signaling does not affect the strong anti-tumor activity of IL-12. Thus, neutralizing TNFα with antagonists already approved for human use offers a promising approach to abrogate IL-12 side effects during the use of this cytokine for the treatment of cancer.

Figure 1. Relative absolute cell numbers of splenic cell subpopulations in mice hydrodynamically injected with cytokine cDNAs.

C57BL/6 mice were hydrodynamically injected with control, IL-12 and/or IL-18 cDNAs. Up to 50 days post-injection, animals were sacrificed and single cells suspensions of splenic cells were obtained. The percentage of the different subsets was obtained by surface lineage staining with combination of the following Abs: NK1.1, CD3, CD11b, CD19, CD4 and CD8 and analyzed by flow cytometry. Relative cell numbers of the different cell populations were calculated by dividing the absolute cell numbers obtained from the cytokine-treated mice by the absolute cell numbers obtained from control mice. Data is expressed as the mean of a pool of 4 different experiments (total n = 12 mice per group). *12+18 and 12 vs control p<0.05, ^#12 vs 12+18 p<0.05.

Figure 2. inflammatory profile of Mφ following IL-12 cDNA treatment.

(A) RT-PCR analysis of iNOS gene expression in splenocytes obtained from mice injected with the designated cDNAs, 12+18 and 12 vs control p<0.05. (B) Western blot analysis of splenocyte protein extracts obtained from mice injected with the designated cDNAs, 12 vs 12+18, p<0.05. (C) Survival monitored in B6 (WT) or iNOS KO mice in the days post-IL-12 or IL-12+IL-18-cDNA injection. (D) Arginase activity measured determination of urea levels in 1×10⁶ splenocytes. The data in A, B and D are expressed as the mean of 2 different experiments with 4–5 mice per group. Survival curve was built as a pool of 2 different experiments with 5 mice per group. *12+18 vs 12 p<0.05.

Figure 3. Splenocytes produce Th1 and Th2 cytokines after IL-12+IL-18 cDNA treatment.

Splenocytes from IL-12+IL-18 mice were obtained and stained with anti-CD4, anti-CD8 and anti-CD11b/c fluorocrome labeled Abs and then sorted to a purity of 97–99% using a MoFlo sorter (Dako Cytomation). Sorted cells were then cultured in the presence of anti-CD3 coated Ab (2 µg/ml) and LPS (1 µg/ml) for 72 h. Supernatants were then harvested and (A) TNFα, (B) IFNγ and (C) IL-10 expression were evaluated by ELISA. Data are expressed as the mean of 2 different experiments with 4-5 mice per group. *Stimulated vs non-stimulated, p<0.001.

Figure 4. TNFα plays a critical role in mediating IL-12 systemic adverse effects.

(A) Survival was monitored in WT, TNFα KO and TNFRI KO mice in the days post-IL-12 cDNA treatment. *WT vs TNFα KO and TNFRI KO, p<0.01. (B) Mice from control, IL-12 or IL-12+IL-18-cDNA treated groups were bled and TNFα sera levels were determined in the days post-treatment. (C) Splenocytes from control, IL-12 or IL-12+IL-18 cDNA treated mice were obtained and cultured in the presence of media (NS), anti-CD3 coated Ab or LPS for 72 h. Supernatants were then harvested and TNFα expression was evaluated by ELISA. *IL-12 vs IL-12+IL-18, p<0.05. (D) Sera from B6 mice treated with control or IL-12 cDNA and from TNFRI KO mice treated with IL-12 cDNA were obtained in the days post-treatment. ALT levels were determined using a specific colorimetric kit. The data in A, B and C are expressed as the mean of 2 different experiments with 4 mice per group. The data in D are expressed as the mean of the pool of 3 different experiments (WT n = 12, TNFR1 KO n = 9). *WT IL-12 cDNA vs TNFRI KO IL-12 cDNA, p<0.05.

Figure 5. Histopathological alterations in hepatic tissue after systemic expression of IL-12 or IL-12+IL-18.

Livers obtained from mice in the different groups were obtained on day 14-16 post-cDNA treatment. Sections of 6 µm were obtained and (A–G) hematoxilin/eosin (H/E) or (H–I) Periodic acid-Schiff (PAS) stains were applied to the sections. (A) Hepatocytes and sinusoids from a control mouse, 100×. (B) Ballooning hepatocytes and absence of hepatic sinusoids in an IL-12-cDNA treated mouse, 100×. (C) Focus of ectopic hematopoiesis in an IL-12-cDNA treated mouse, 100×. (D) Megakaryocyte in an IL-12-cDNA treated mouse, 100×. (E) Mallory and (F) Councilman bodies in an IL-12-cDNA treated mouse, 100×. (G) Focus of necrosis; 10×. Glycogen deposits obtained in a mouse from (H) the control group or (I) the IL-12-cDNA group, 20×. Liver sections from (J) WT or (K) TNFR1 KO mice 15 days after IL-12 cDNA treatment. Pictures are representative of 2 different experiments with 4–5 mice per group.

Figure 6. IL-12+IL-18 co-expression and low amounts of IL-12 cDNA reduce IL-12-side effects by decreasing TNFα expression.

B6 mice were treated with IL-12 cDNA alone (black circle), IL-12+IL-18 cDNA (open circle) or low IL-12 cDNA (0,5 µg) 2 days prior to administration of a high amount of IL-12 cDNA (5 µg) (black square). (A) Survival was monitored and (B) IFNγ, (C) TNFα and (C) IL-10 sera expression were evaluated by ELISA in the days post-treatment. The data are expressed as the mean of 2 different experiments with 5 mice per group. IL-12+IL-18 or IL-12 low cDNA + IL-12 high cDNA vs IL-12, ***p<0.01 or ^#p<0.05. IL-12+IL-18 vs the rest of the groups, *p<0.05.

Figure 7. Abrogation of TNFα signaling does not affect IL-12 antitumor activity.

For subcutaneous tumor growth, 1×10⁶ B16 cells were injected s.c. in C57BL/6 (WT) and TNFR1 KO mice (8–10 weeks old). Between days 10–14, when tumors had reached a size of approximately 5 mm, mice were hydrodynamically injected with IL-12 cDNA. In the days post cDNA injection, tumor growth was measured with a caliper (upper panel). For liver metastases analysis, intrasplenic injection was performed on day 0 with 0.5×10⁶ B16 melanoma cells and 3 days later with IL-12 cDNA via hydrodynamic shear. Liver examination was performed on day 14 post treatment (lower panel). IL-12 vs control p values are plotted in the figure. The data in the upper panel are expressed as the mean of 2 different experiments with 4 mice per group. Pictures of the lower panel are representative of 2 different experiments with 4–5 mice per group.