Immunoregulatory properties of CD137-CD137 ligand axis in tumours and chronic inflammatory conditions

Abstract

CD137 (4-1BB/TNFRSF9) is a member of the TNF receptor superfamily. CD137 expression is activation-dependent, and usually reported on T cells and NK cells. Its natural ligand is CD137 ligand (CD137L), mainly found on antigen-presenting cells (APC) such as monocytes, or dendritic cells. CD137-CD137L interaction will send a stimulatory signal to activate both interacting cells, thus enhancing the immune response. Hence, it is curious to detect CD137 on tumour cells and regulatory T cells (Treg), who mainly aim to suppress the immune functions. Hence, we aim to investigate how the CD137-CD137L can confer immunosuppression which is beneficial for tumour growth, or the maintenance of a chronic inflammatory condition. Because of RNA alternative splicing, soluble isoforms of CD137 (sCD137) that lack the cytoplasmic domain can be produced and secreted to the surrounding environment.

In Paper I, we reveal that sCD137 is a marker of liver cirrhosis in patients suffering hepatitis C and alcohol-associated disease. sCD137 impairs type-1 and type-2 T cell response but maintain type-17 and regulatory T cell response. Thus, it helps maintain the chronic inflammation associated with cirrhosis. In paper II, we reveal how Tregs express more CD137 than conventional T cells (Tcon), use CD137 to form a complex with CD137L, internalize the complex and deplete CD137L on APC, in a process termed “trogocytosis”. This process does not require CD137 signaling because our modified truncated CD137 that lacks the cytoplasmic domain can still deplete CD137L. APC that have their CD137L depleted in the process are less able to co-stimulate T cells. In paper III, we study how the stimulation of CD137 on T cells can induce B7-H7 expression. B7-H7 is a ligand usually upregulated in cancers and associated with worse prognosis for a wide variety of cancers. On immune cells, B7-H7 is usually reported on APC. Our study reveals that B7-H7 can be induced on T cells to regulate T cell function and serve as a biomarker for exhausted T cells. In paper IV, using the nasopharyngeal carcinoma (NPC) model, we study why tumours are selected to express CD137. Besides using CD137 for trogocytosis to deplete CD137L from APC, tumour can also receive CD137 signaling to upregulate p38 MAPK pathway, secrete more Interleukin (IL)-8, and express LAMC2 to enhance their survival, invasion, and metastasis. IL-8 has been reported to be important to recruit other immunosuppressive cells into the tumour microenvironment, and LAMC2 is involved in the Epithelial-Mesenchymal Transformation (EMT) process. Learning from these mechanisms, in paper V, we devise a fusion protein PD-1-CD137L that masks PD-L1, a co-inhibitory molecule, with CD137L, a costimulatory molecule. In effect, this converts the tumour microenvironment from cold to hot. Not only blocking PD-L1/PD-1 pathway, but this fusion protein also presents more CD137L in the tumour microenvironment, sequesters sCD137, inhibits trogocytosis and liberates existing CD137L to activate tumour-specific T cells.

In conclusion, while CD137 has been exploited as target for immunotherapy (as seen in several clinical trials where agonistic anti-CD137 antibodies are used for cancer treatment), it can induce immunosuppression beneficial to tumour and chronic inflammation. Studying these mechanisms of action elucidate how we can improve future immunotherapy involving CD137 and similar receptors.