Natural killer cell homing and trafficking in tissues and tumors: from biology to application

Abstract

Natural killer (NK) cells, a subgroup of innate lymphoid cells, act as the first line of defense against cancer. Although some evidence shows that NK cells can develop in secondary lymphoid tissues, NK cells develop mainly in the bone marrow (BM) and egress into the blood circulation when they mature. They then migrate to and settle down in peripheral tissues, though some special subsets home back into the BM or secondary lymphoid organs. Owing to its success in allogeneic adoptive transfer for cancer treatment and its "off-the-shelf" potential, NK cell-based immunotherapy is attracting increasing attention in the treatment of various cancers. However, insufficient infiltration of adoptively transferred NK cells limits clinical utility, especially for solid tumors. Expansion of NK cells or engineered chimeric antigen receptor (CAR) NK cells ex vivo prior to adoptive transfer by using various cytokines alters the profiles of chemokine receptors, which affects the infiltration of transferred NK cells into tumor tissue. Several factors control NK cell trafficking and homing, including cell-intrinsic factors (e.g., transcriptional factors), cell-extrinsic factors (e.g., integrins, selectins, chemokines and their corresponding receptors, signals induced by cytokines, sphingosine-1-phosphate (S1P), etc.), and the cellular microenvironment. Here, we summarize the profiles and mechanisms of NK cell homing and trafficking at steady state and during tumor development, aiming to improve NK cell-based cancer immunotherapy.

Fig. 1

The development of immunotherapy based on NK cells. Applications of various NK cell strategies in preclinical and clinical trials are listed. Since the discovery of NK cells in 1970s, NK cells have been used in clinical treatment. The combination of IL-2 and killer cells in the treatment of non-Hodgkin’s lymphoma (NHL), metastatic RCC (mRCC), and metastatic melanoma (mM), starting from 1987. Antibodies enhancing NK cell effector function were produced starting from 1992, including NK bi-specific monoclonal antibody targeting ovarian cancer cells in 1992, Trastuzumab (anti-Her2) in the treatment of breast cancer and gastric cancer, in 1998, anti-KIR antibody preclinical test in 2014, and tri-specific NK cell engagers preclinical test in 2016. Clinical trials using ex vivo expanded peripheral blood mononuclear cell (PBMC)-derived NK cells for non-small cell lung cancer (NSCLC) were started in 2010, while iPSC-derived NK (iPSC-NK) cells were used for AML treatment in 2017, and for multiple myeloma (MM) in 2019. CD4 zeta chimeric receptors were firstly introduced into NK cells and showed highly effective at killing target cells expressing HIV gp120 in vitro in 1995. Recently, a variety of clinical phase I or II trials were started, including Phase1/2 CD19-chimeric antigen NK cells for CD19-positive lymphoid malignancies in 2017 (NCT03056339), Phase1/2 CD33-CAR NK cells for AML treatment in 2021, and 5T4-CAR NK cells for treating advanced solid tumors (NCT05194709) in 2022. Preclinical trial for CAR iPSC-NK cells for ovarian cancer were started in 2018, while clinical trials of CAR iPSC-NK cells for treating NHL or chronic lymphoblastic leukemia (CLL) in 2020. Created with BioRender.com

Fig. 2

Schematic representation of the cellular intermediates in both murine and human NK cell development. Surface antigens help distinguish the intermediate populations in NK development. a Linear path (left to right) in the development of murine NK cells from HSC to mature CD11b⁺CD27⁻DX5⁺ NK cells. b Linear path (left to right) in the development of human NK cells from HSC to mature CD56^dim NK cells. + (expression), − (no expression), hi (high expression), low (low expression). Created with BioRender.com

Fig. 3

NK cell trafficking and homing at steady state. NK cells develop from HSCs in the BM parenchyma and gradually reach blood and peripheral tissues as they mature. CXCR4 is highly expressed in immature NK cells, including NKPs and iNKs, making NK cell remains in or homes back to the BM parenchyma. When CXCR4 expression is downregulated and the expression of CXCR3, CXCR6, S1P5 and CX3CR1 is upregulated, NK cells gradually egress from the BM and enter peripheral tissues via circulating blood. Created with BioRender.com

Fig. 4

NK cells in the TME. NK cells extravasate from the blood, traverse the ECM and tumor stroma, and reach the tumor bed when they are recruited by integrin, chemokine receptor, and selectin. In the TME, NK cells eliminate tumor via degranulation, ADCC, or FASL/TRAIL-induced apoptosis. NK cells can also secrete cytokines or chemokines to recruit other immune cells and upregulate the anti-tumor response. However, NK cell responsiveness is often hindered by repressive factors secreted by tumor cells or other cells or by direct cell–cell interaction. In addition, NK cells can be educated by tumor cells, for example, to secrete pro-angiogenic factors to promote tumor angiogenesis. This education can switch anti-tumor immunity to pro-tumor immunity in the TME. Created with BioRender.com

Fig. 5

Factors affecting NK cell trafficking and homing. Cell intrinsic factors (e.g., transcription factors), extrinsic factors (e.g., cytokines, hypoxia, low pH), and cell–cell interactions affect NK cell trafficking and homing in both healthy and tumor tissues. During NK cell development, some transcription factors cell-intrinsically regulate the expression of chemokine receptors, selectins, and integrins, etc., thus helping to regulate NK cell homing and trafficking in a cell-intrinsic manner. Cytokines and immune suppressive factors from the TME (e.g., hypoxia, low pH, and adenosine) can also sabotage the recruitment and homing of NK cells in specific tissues by affecting the expression of chemokine receptors, selectins, and integrins on NK cells or their corresponding ligands in the TME. DCs, B cells, and CD4⁺ cells also affect the migration of NK cells via the chemokine–chemokine receptor axis. Created with BioRender.com

Fig. 6

Strategies to improve the infiltration, activation, and survival of CAR-NK cells in the TME. Expressing a CAR on NK cells can alter the profiles of chemokine receptors, selectins, and cytokines, boosting the infiltration of NK cells into the TME and modulating the TME’s immune microenvironment. Co-expression and administration of cytokine(s) and chemokine(s) or a corresponding receptor with a CAR have multiple effects on NK cell-based immunotherapy, including on infiltration, activation, and survival of NK cells in the TME. In addition, a switch receptor strategy was used to equip a CAR with an extracellular domain targeting a suppressive antigen in the TME and an intracellular domain with an activation signal. This strategy (to produce TN-CAR and 4/7 ICR CAR, for example) converts the TME’s suppressive signal into an activation signal, enhancing NK cell effector functions. 4/7 ICR-CAR: IL4R-IL-7R-CAR. Created with BioRender.com