Dual Role of TNF and LTα in Carcinogenesis as Implicated by Studies in Mice

Abstract

Tumor necrosis factor (TNF) and lymphotoxin alpha (LTα) are two related cytokines from the TNF superfamily, yet they mediate their functions in soluble and membrane-bound forms via overlapping, as well as distinct, molecular pathways. Their genes are encoded within the major histocompatibility complex class III cluster in close proximity to each other. TNF is involved in host defense, maintenance of lymphoid tissues, regulation of cell death and survival, and antiviral and antibacterial responses. LTα, known for some time as TNFβ, has pleiotropic functions including control of lymphoid tissue development and homeostasis cross talk between lymphocytes and their environment, as well as lymphoid tissue neogenesis with formation of lymphoid follicles outside the lymph nodes. Along with their homeostatic functions, deregulation of these two cytokines may be associated with initiation and progression of chronic inflammation, autoimmunity, and tumorigenesis. In this review, we summarize the current state of knowledge concerning TNF/LTα functions in tumor promotion and suppression, with the focus on the recently uncovered significance of host-microbiota interplay in cancer development that may explain some earlier controversial results.

Keywords: LTβR; TNFR2; cancer; lymphotoxin alpha; microbiota; mouse models; tumor necrosis factor.

Figure 1

Ligands and receptors of the tumor necrosis factor (TNF)/lymphotoxin (LT) axis. Tumor necrosis factor (TNF) exists in either soluble (sTNF) or trans-membrane (tmTNF) form and inte-racts with its two receptors: TNFR1 (TNFp55) and TNFR2 (TNFp75). Another ligand for TNF receptors (TNFRs) is lymphotoxin α homotrimer (LTα3) that, in addition, binds herpes virus entry mediator (HVEM). HVEM, lymphotoxin β receptor (LTβR), and decoy receptor 3 (DcR3) are receptors for LIGHT, while LTβR also interacts with LTαβ heterotrimer, a membrane-bound form of lymphotoxin.

Figure 2

TNF-deficient mice with unperturbed LTα expression [3] are partially protected from DMBA/TPA skin carcinogenesis, while genetic LTα ablation sensitizes mice to this type of skin cancer. TNF-deficient mice were compared to heterozygous TNF^+/− co-housed littermate control mice. LTα-deficient mice were compared to heterozygous LTα^+/− co-housed littermate control mice. All animals aged 6–8 weeks were provided by the Unique Scientific Unit “Biomodel”, Branch of Shemyakin–Ovchinnikov Institute of Bioorganic Chemistry of the Russian Academy of Sciences (BIBCh, RAS), Pushchino, Moscow Region, Russia. The skin carcinogenesis was induced as previously described [66]. Briefly, a single application of 25 mcg of DMBA in 100 mcl of acetone was administrated on day 0 to the shaved back skin area. TPA was administered at 4 mcg per 100 mcl acetone to the same area three times a week for 20 weeks. After that, mice were further monitored for an additional 20 weeks and papilloma formation and size were measured on a weekly basis. (A). Percent of papilloma-free TNF-deficient and littermate control mice throughout the experiment. (B). Number of papillomas per TNF^+/− and TNF^−/− mouse throughout the experiment. (C). Percent of papilloma-free LTα-deficient and littermate control mice. (D). Number of papillomas per LTα^+/− and LTα^−/− mouse throughout the experiment. (E). Total papilloma volume per LTα^+/− and LTα^−/− mouse.

Figure 3

LTαβ–LTβR interaction protects mice from the development of DMBA/TPA chemically induced skin cancer. Mice [71,72,73] and the protocol for chemically induced skin carcinogenesis were previously described [66]. (A). Day of the first papilloma’s onset. * p < 0.05. (B). Number of papillomas per mouse at week 20 of the experiment. * p < 0.05. (C). Number of papillomas per mouse at week 30 of the experiment. * p < 0.05. (D). Heat map representing global variations in cytokine/chemokine responses (log₁₀) in mice of different genotypes at week 20 of the experiment, as determined by multiplex analysis. Cytokine and chemokine levels in mouse sera were measured simultaneously using a multiplex microbead-based immunoassay, MILLIPLEX MAP Mouse Cytokine/Chemokine Magnetic Bead Panel-Premixed 32 Plex (MCYTMAG-70K-PX32, Merck) according to the manufacturer’s protocol.

Figure 4

Relative expression of Tnf, Il-23, Cxcl1, S100a8, S100a9, Il-13, Il-6, Cathelecidin, and Defensin b in the LTα-deficient and control mouse skin two hours after the final phorbol myristate acetate (PMA) application. * p < 0.05. Acute skin inflammation was initiated in LTα-deficient and control mouse skin by administration of 4 mcg PMA in 100 mcl acetone on the shaved back skin area. Starting from day 2, the same amount of PMA treatment was followed by 25 mcg of Aldara cream application to the same area 10 min later with 4–6 cycles of PMA/Aldara every 2 days. Skin biopsies were collected, homogenized in TRK Lysis Buffer, and RNA extracted using an E.Z.N.A.^® Total RNA Kit (according to the manufacturer’s instructions). mRNA was transcribed into cDNA using a standard protocol and RevertAid First Strand cDNA Synthesis kit reagents followed by quantitative real-time PCR (see Table S2).

Figure 5

Genetic TNF ablation does not affect tumor incidence in the model of azoxymethane (AOM)/dextran sodium sulfate (DSS)-induced colorectal cancer following long-term co-housing with littermate control mice. TNF-deficient and heterozygous TNF^+/− littermate control mice received a single intraperitoneal (i.p.) injection of AOM (Sigma-Aldrich, Darmstadt, Germany) at a dose of 12 mg/kg body weight. A week after AOM injection, mice were given three cycles of 2% DSS in drinking water (Thermo Fisher) for five consecutive days followed by a two-week interval on drinking water without DSS. Body weight was measured throughout the experiment. After the last water cycle, mice were sacrificed and examined for polyp formation and colon length. (A). Percentage of initial body weight during the experiment. (B). Tumor number. N.S.: not significant. (C). Tumor load. N.S.: not significant. (D). Average tumor size. N.S.: not significant.