by ecancer reporter Clare Sansom

There is accumulating evidence to suggest that the complex innate and acquired immune responses to growing tumours can be controlled by the tumour micro-environment.

Once the mechanisms through which this occurs has been elucidated, it should be possible to develop or modify therapies to take advantage of them.

Dendritic cells (DCs), which act as “critical sensors” of innate immune responses, are known to infiltrate tumours and to suppress the activation of the innate immune system’s anti-tumour response.

Masahisa Jinushi of Hokkaido University, Sapporo, Japan, and his co-workers have now shown that these cells suppress immune responses mediated by nucleic acids through an interaction between the receptor TIM-3 and an alarmin, HMGB1.

Jinushi and colleagues first showed using several tumour-bearing mouse models that the receptor TIM-3 was highly expressed on tumour cells and on tumour-infiltrating dendritic cells and CD8⁺ T cells, but not on similar immune cells not associated with tumours in the same mice.

Dendritic cells isolated from human cancer patients also showed high TIM-3 expression, and some immature dendritic cells derived from bone marrow could be induced to express TIM-3 by incubation with tumour cells or tumour cell supernatants.

The researchers next investigated the role that TIM-3 plays in dendritic cell function by comparing the expression patterns of TIM-3 expressing (TIM-3⁺) bone marrow dendritic cells (BMDCs) with their TIM-3^- counterparts. Cells expressing TIM-3 had much lower expression of a number of key cytokines than those that did not.

These cytokines are expressed in response to signals from toll-like receptors (TLRs), some of which act as recognition sensors for nucleic acids. This indicated that TIM-3 might be involved in the regulation of innate immune responses mediated by nucleic acids.

To test this hypothesis, the researchers transfected TIM-3^- mouse embryonic fibroblasts with vector expressing TIM-3 or control vector, and found that cells transfected with TIM-3 expressed cytokines less strongly in response to DNA from a variety of sources including pathogens. TIM-3 also suppressed the activity of transcription factors including NF-kB in response to DNA. 

These transcription factors were also found to be suppressed in tumour-infiltrating dendritic cells isolated from tumour-bearing mice and human cancer patients, indicating that this effect also occurs in vivo.

These results suggested that TIM-3 based suppression of innate immune responses to DNA might have the effect of undermining the effect of cancer vaccines based on DNA. To explore the mechanism of this response further, Jinushi and co-workers generated bone marrow chimaeras that could be selectively depleted of TIM-3⁺ dendritic cells.

They found that DNA had a synergistic effect with TIM-3 directed monoclonal antibodies in inhibiting tumour growth, and further experiments suggested that blocking TIM-3 with antibodies might trigger the anti-tumour response mechanism in dendritic cells.

Some of the immune functions of TIM-3 are known to be mediated through interaction with a lectin, galectin-9. However, tumour-associated dendritic cells were found to express less galectin-9 than other dendritic cells, and antibodies to this protein did not enhance the anti-tumour effect of DNA in TIM-3⁺ cells. This suggested that TIM-3 in tumour-infiltrating dendritic cells must suppress the anti-tumour response through a different mechanism. 

Further analysis implicated a conserved nuclear protein, the alarmin HMGB1, as the TIM-3 ligand involved in this interaction. This protein is also known to bind DNA, and Jinushi and co-workers showed through biotin labelling of the interaction partners that TIM-3 competes with nucleic acids for binding to one domain of HMGB1, termed the A-box domain. Tumours were also found to express significantly more HMGB1 than normal tissue.

Finally, the researchers showed that TIM-3 inhibits the trafficking of DNA into endosomes, which is a key early event in innate immune signalling, through its interaction with HMGB1. This inhibits the response of the innate immune system to, for example, the release of nucleic acids by dying tumour cells that have been treated with cytotoxic agents. Thus, the action of TIM-3 expressed by tumour-infiltrating dendritic cells can diminish the effects of chemotherapy as well as of DNA-containing cancer vaccines. This further suggests that this protein might itself be a useful target for anti-cancer drugs. 

Reference 

 Chiba, S., Baghdadi, M., Akiba, H. and 9 others (2012). Tumor-infiltrating DCs suppress nucleic acid–mediated innate immune responses through interactions between the receptor TIM-3 and the alarmin HMGB1. Nature Immunology, published online ahead of print 29 July 2012. doi:10.1038/ni.2376