Immune cells called regulatory T cells (Treg cells) are a subset of T cells that selectively dampen immune responses. They do this by suppressing the activation of inflammation-promoting T cells, and also by secreting anti-inflammatory factors1. Such blunting of an immune response is valuable because it prevents the immune system from turning on a person’s own body — a type of malfunction that occurs in autoimmune disease. However, Treg cells can benefit tumours by suppressing cancer-attacking immune cells, such as CD8 T cells (also known as killer T cells). Writing in Nature, Lim et al.2 identify a metabolic dependency of Treg cells in the tumour microenvironment, a finding that reveals how the cells operate there.
Immunotherapy is used in the clinic to overcome a tumour’s evasion of killer T cells. The approach can include treatments such as antibodies that target Treg cells3. Although such therapy boosts anticancer immune responses, it can have a negative effect on Treg cells elsewhere in the body that help to keep the immune system in balance. As a result, people receiving such treatments often develop autoimmune disease4. A major unmet need is therefore immunotherapy that targets only the ‘bad’ Treg cells in the tumour vicinity while leaving the beneficial Treg cells untouched.
To find a way to single out the unwanted Treg cells, Lim and colleagues used mice that had a type of tumour called melanoma. They compared the gene-expression profiles of Treg cells extracted from the tumour vicinity with those taken from elsewhere in the animal’s body. Only tumour-associated Treg cells expressed genes whose expression is controlled by a group of transcription factors called sterol regulatory element binding proteins (SREBPs). These proteins drive the expression of genes encoding enzymes that produce lipids5, such as fatty acids and cholesterol (Fig. 1), which are needed for processes including cellular signalling and the construction of cell membranes.
To test whether this lipid-producing transcriptional signature is functionally important, Lim and colleagues used genetically engineered mice in which the SREBP-mediated gene-expression pathway was switched off specifically in Treg cells. The authors monitored the growth of tumour cells transplanted beneath the animals’ skin, and found that this interruption of SREBP resulted in much better antitumour immune responses in two forms of cancer than occurred in animals that had functional SREBPs.
Mice that did not receive tumour transplants but lacked SREBP-mediated gene expression showed no signs of autoimmune disease. This indicates that Treg cells outside the tumour environment were functioning normally without the need for SREBP-mediated gene expression. Even when these animals were manipulated to develop an autoimmune brain disease similar to human multiple sclerosis, they had the same level of disease severity as did mice with normal Treg cells. That result demonstrates that SREBP-mediated gene expression is needed for Treg cells in the tumour environment but can be dispensable for other Treg cells.
Why is SREBP-mediated lipid production needed for tumour Treg cells? Cancers extract lipids from their surroundings and use these molecules to fuel their energy and growth6. In theory, a scarcity of lipids around tumours might mean that tumour Treg cells must make their own lipids. But there is more to this requirement for SREBPs than just to satisfy the cell proliferation and energy needs of Treg cells.
Lim and colleagues identify two key roles for SREBPs (Fig. 1a, b). First, they show that tumour Treg cells need SREBPs to generate fatty-acid synthase, an enzyme involved in fatty-acid synthesis. If this enzyme is missing, tumour Treg cells do not become fully mature, losing effectiveness and showing a diminished ability to blunt immune responses compared with Treg cells that have this enzyme.
Second, Lim et al. demonstrate that, for Treg cells to carry out their usual anti-inflammatory role in the tumour environment, they rely on what is called the mevalonate pathway (Fig. 1b). This SREBP-dependent pathway produces cholesterol, as well as other molecules, including geranylgeranyl pyrophosphate (GGPP). GGPP becomes bound to proteins through a process called prenylation. The addition of GGPP changes the target protein’s chemical properties, in much the same way that other types of protein modification, such as phosphorylation and acetylation, alter the modified protein.
Lim and co-workers provide evidence linking GGPP production through the mevalonate pathway to the expression of a gene that encodes an immunosuppressive protein called PD-1. The prenylated protein that is presumably required for PD-1 expression is unknown; however, the authors demonstrate that, without GGPP, tumour Treg cells did not upregulate the gene encoding PD-1. They show that PD-1 is required to ‘stabilize’ tumour Treg cells: treatment of the tumour-bearing mice with an antibody that blocks PD-1 function leads to the expression of genes not normally associated with Treg cells, such as a gene that encodes the pro-inflammatory protein interferon-γ (Fig. 1c). Treg cells that produce interferon-γ can’t shield a tumour from attack by the immune system7.
The fact that a Treg-cell population found in the context of cancer is metabolically vulnerable is a profound revelation. It might point the way towards the development of less toxic immunotherapies that selectively target damaging Treg cells. With hundreds of clinical trials currently under way that are examining how anticancer immune responses might be boosted, attempts to destabilize tumour Treg cells by targeting the pathways highlighted by Lim and colleagues will undoubtedly be of interest.
Drugs that specifically inhibit the mevalonate pathway are already in clinical use to combat cardiovascular disorders. For example, statins are a class of cholesterol-lowering drug that has been used by millions of people since the 1980s. Indeed, mortality is lower in people with tumours who are taking statins — a finding observed for cancers that include multiple myeloma8, oesophageal cancer9 and pancreatic cancer10. The idea of interrupting the mevalonate pathway as a way of treating cancer is gaining support because it has been observed that, compared with normal cells, some tumour cells have an increased demand for molecules generated downstream of this pathway11. It is fascinating to speculate that Treg cells might have contributed to these earlier clinical observations. Perhaps inhibitors of the mevalonate pathway or inhibitors of GGPP-mediated prenylation will play a part in future anticancer therapies.
The key role of fatty-acid synthase in tumour Treg-cell function is an interesting discovery, given that other research12 indicates that inhibition of the enzyme acetyl-CoA carboxylase 1 (which functions one step upstream of fatty-acid synthase in the same pathway) boosts the formation and function of Treg cells in the same autoimmune brain disease mouse model as that used by Lim and colleagues. These findings suggest that the effects of disrupting Treg-cell function by interrupting SREBP-dependent fatty-acid synthesis is context dependent. Outside the tumour environment, disrupting fatty-acid synthase had no effect, whereas inhibiting acetyl-CoA carboxylase 1 actually conferred benefits on Treg-cell function12.
Lim and colleagues’ study has implications beyond the realm of cancer. A rare autoinflammatory disease called mevalonate kinase deficiency is caused by a mutation in the gene encoding the enzyme mevalonate kinase, which acts in the mevalonate pathway. The disease is thought to be driven by defective protein prenylation, but the lack of a clear mechanistic understanding of the underlying cause has hampered efforts to develop an effective treatment13. Lim and colleagues’ evidence raises the question of whether PD-1 or Treg cells might be linked to this disease. The possibility warrants further investigation.
The research by Lim et al. reinforces the need to understand the relationship between metabolic pathways and the regulation of immune-system function. As this work shows, such insights could be vital in efforts to treat cancer.
Nature 591, 204-206 (2021)
Competing Financial Interests
U.H.B. anticipates employment at Janssen later this year.