Bacterial Peptides Turn Tumor Cells into Immunotherapy Targets

Protein fragments from bacteria invading a tumor can be presented on the cancer cell surface and recognized by the immune system. This discovery represents a new target for cancer immunotherapies.

Human tumors are colonized by microorganisms collectively called the tumor microbiota, which can influence the tumor microenvironment. This may change how the body's immune system responds to the tumor and alter treatment responses. But does the immune system recognize intratumoral bacteria themselves?

Peptides Presented to Immune Cells

In their Nature paper, Kalaora and colleagues showed that bacterial protein fragments called peptides can be presented to the immune system on tumor cells and recognized by immune cells called T cells. This discovery could be exploited for the benefit of cancer immunotherapies.

Molecules called tumor antigens allow the immune system to distinguish tumor cells from healthy cells. Each cell contains antigen-processing machinery that enables antigen-derived peptides to be presented to the immune system by specialized molecules on the cell surface called human leukocyte antigens (HLAs). Peptides presented by HLAs and recognized by immune cells are called epitopes.

Are Personalized Immunotherapies Possible?

Tumor antigens fall into two main categories:

• tumor-associated antigens and
• tumor-specific antigens

Tumor-associated antigens are expressed in normal tissues as well as tumors and therefore do not readily activate immune responses. If immune responses increase, however, there is a risk of harmful autoimmune reactions against normal tissues expressing the antigen. Nevertheless, because tumor-associated antigens are generally found in many people with cancer and many tumor types, they may be good targets for broadly applicable immunotherapies.

By contrast, tumor-specific antigens are expressed only in tumor cells and are therefore ideal targets for enhancing a specific immune attack against the antigens. One subtype, neoantigens, arises from tumor-specific gene mutations, so neoantigens are generally specific to the tumor and patient. Targeting neoantigen-derived peptides (called neopeptides) requires entirely personalized immunotherapies.

Melanoma is a skin cancer with three known classes of tumor-associated antigens, and its cells generally carry many genetic mutations, resulting in a high likelihood of neoantigens. It has therefore been at the forefront of tumor-antigen discovery and cancer-immunotherapy development. It was thus appropriate for Kalaora and colleagues to use melanoma samples to identify another potential class of tumor antigen.

The researchers began investigating bacterial composition in 9 people with 17 melanoma metastases. They found highly similar bacterial compositions in different metastases from the same individual and sometimes in samples from different people, consistent with an earlier study reporting tumor microbiota specific to different cancer types. This finding suggests that particular bacterial species are specific to melanoma. The researchers also confirmed that these bacteria were present in melanoma cells rather than in the extracellular microenvironment surrounding the tumor.

Kalaora and colleagues went on to investigate whether peptides from these intracellular bacteria were presented to the immune system in the same way as other intracellular antigens. To do so, they used a mass-spectrometry-based approach called immunopeptidomics that allows direct detection of peptides presented by HLAs. They found nearly 300 peptides from 33 bacterial species in the same samples. Several peptides were found to be more numerous in a tumor from the same person and in tumors from different people.

The researchers next asked whether bacterial peptides were actually presented by melanoma cells rather than by immune cells called antigen-presented cells (APCs), which detect and take up pathogens and present them to other immune cells. They used an immune-cell marker protein to separate cells from two melanoma samples into APCs and tumor cells.

Immunopeptidomics revealed that both groups presented bacterial peptides. A subset of peptides was presented by both APCs and tumor cells, suggesting that the same peptide could initiate an immune response through presentation on APCs and serve as a target for immune attack on tumor cells. The researchers then showed that T cells isolated from melanomas (which recognize peptides presented by HLAs) responded to the identified bacterial peptides, including some shared across tumors and individuals.

Conclusion

Taken together, the results of Kalaora and colleagues point to the possibility that tumor-displayed bacterial peptides constitute a previously unidentified class of tumor antigen. Several questions remain, however. To be genuine tumor antigens, the identified bacterial species must not invade nontumor tissues, and their peptides must not be presented on HLAs on nontumor cells. If such presentation is detected, the peptides cannot qualify as immunotherapy targets.

In addition, bacterial peptides appear quite abundant, at least compared with the number of identified melanoma neopeptides, so why does the body not mount an effective immune response against melanomas? Further studies of tumor-displayed bacterial peptides, combined with patient information, will be needed to clarify the peptides' potential clinical role. Such data help researchers select appropriate bacterial targets for cancer-immunotherapy approaches.

Figure: A pathway for antitumor immunity triggered by bacteria. Kalaora and colleagues reported that certain bacteria can invade cells from a type of tumor called melanoma. Peptide fragments from bacteria-derived molecules are presented on tumor cell surfaces by proteins called human leukocyte antigens (HLAs). Immune cells called T cells recognize these presented peptides and become activated. Bacterial peptides may therefore be a previously unidentified class of tumor antigen, a type of molecule that allows T cells to distinguish tumor cells from normal cells.

In conclusion, the bacterial peptides identified by Kalaora and colleagues may be attractive targets for immunotherapy. Because bacterial peptides are "nonself," it should be relatively easy to elicit strong immune responses against them, and if they can be shown not to be presented on any normal tissue, there would be no concern about autoimmunity. Thus, tumor-displayed peptides could serve as tumor-specific antigens shared between people—a rare and useful combination for therapeutics. Until now, this has only been seen in virally induced tumors, where epitopes can be derived from viral cancer-causing proteins. Recent data suggest that bacteria invading tumors may be a widespread phenomenon. The work of Kalaora and colleagues may therefore form the basis for identifying shared tumor-specific antigens across many tumor types.