Natural killer cells suppress cancer metastasis by eliminating circulating cancer cells

Abstract

Despite significant advances in cancer treatment, the metastatic spread of malignant cells to distant organs remains a major cause of cancer-related deaths. Natural killer (NK) cells play a crucial role in controlling tumor metastasis; however, the dynamics of NK cell-mediated clearance of metastatic tumors are not entirely understood. Herein, we demonstrate the cooperative role of NK and T cells in the surveillance of melanoma metastasis. We found that NK cells effectively limited the pulmonary seeding of B16 melanoma cells, while T cells played a primary role in restricting metastatic foci growth in the lungs. Although the metastatic foci in the lungs at the endpoint were largely devoid of NK cells, they played a prominent role in promoting T cell recruitment into the metastatic foci. Our data suggested that the most productive interaction between NK cells and metastatic cancer cells occurred when cancer cells were in circulation. Modifying the route of administration so that intravenously injected melanoma cells bypass the first liver passage resulted in significantly more melanoma metastasis to the lung. This finding indicated the liver as a prominent site where NK cells cleared melanoma cells to regulate their seeding in the lungs. Consistent with this notion, the liver and the lungs of the tumor-bearing mice showed dominance of NK and T cell activation, respectively. Thus, NK cells and T cells control pulmonary metastasis of melanoma cells by distinct mechanisms where NK cells play a critical function in shaping T cell-mediated in situ control of lung-seeded cancer cells. A precise understanding of the cooperative role of NK and T cells in controlling tumor metastasis will enable the development of the next generation of cancer immunotherapies.

Keywords: NK cells; T cells; cancer immunology; helper function; metastatic melanoma.

Figure 1

Distinct roles of NK and T cells in controlling the pulmonary metastasis of B16 melanoma. (A) Quantification of WT and B2m^-/- B16 melanoma metastatic foci in the lung following tail vein IV melanoma cell injection and treatment with control IgG, anti-NK1.1, anti-CD4/CD8, anti-NK1.1/CD4/CD8 antibodies. n = 9-15 mice per group. (B) Quantification of the area of WT and B2m^-/- B16 melanoma metastatic foci in the lungs following tail vein IV melanoma cell injection and treatment with control IgG, anti-NK1.1, anti-CD4/CD8, anti-NK1.1/CD4/CD8 antibodies. n = 27-87 macroscopic metastatic foci measured per group. (C) Representative macroscopic images of lung metastases of WT and B2m_-/- B16 melanoma at day 14 post-tail vein IV injection of melanoma cells. Antibody treatments are listed below the lung images. (D) Representative IF images of large and small B2m^-/- B16 melanoma lung metastases in Ncr1^iCre,ROSA^mT-mG reporter mice at day 14 post-tail vein IV melanoma cell injection. The presence of NKp46-GFP+ NK cells (green) in the lung is highlighted with white arrow. Dash lines outline the melanoma metastatic foci in the lung. (E) Quantification of B2m^-/- B16 melanoma metastatic foci in the lung following tail vein (T.V.) and retro-orbital (R.O.) IV melanoma cell injection and treatment with control IgG and anti-NK1.1 antibody. n = 10-15 mice per group. Note that the data for T.V. groups are also shown in (A). (F) Representative macroscopic images of lung metastases of B2m^-/- B16 melanoma at day 14 post-retro-orbital IV melanoma cell injection. Antibody treatments are listed below lung images. (A, B, E) Graphs show mean + SD, one-way ANOVA, ns: not significant, * p < 0.05, *** p < 0.001, (C, F) Scale bars = 1 cm, (D) Scale bars = 100 µm.

Figure 2

NK cells control T cell recruitment to melanoma foci in the lungs. (A, B) Representative images (A) and quantification (B) of CD4⁺ T cells (CD4/CD3 stained) and CD8⁺ T cells (CD8/CD3 stained) in WT B16 melanoma lung metastases. (C, D) Representative images (C) and quantification (D) of CD4⁺ T cells (CD4/CD3 stained) and CD8⁺ T cells (CD8/CD3 stained) in B2m^-/- B16 melanoma lung metastases. Antibody treatments are listed above IF images and below the quantification graphs. (A, C) Dash lines outline the melanoma metastatic foci in the lung where T cells are highlighted with white arrows. Scale bars = 100 μm. (B, D) T cells were counted in 35-140 metastatic foci between 0.01 to 0.2 mm² per group. Mann-Whitney U test, ns, not significant, *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 3

NK cells are enriched in the liver while T cells are enriched in the lung in response to metastatic melanoma. (A) Quantification of total NK1.1⁺ NKp46⁺ NK cells in peripheral blood (PB), liver and lungs at day 14 after WT B16 (WT) and B2m^-/- B16 (B2m^-/-) melanoma tail vein IV injection in comparison with untreated mice (Ф). (B) Quantification of lung-infiltrating IV^neg NK cells at day 14 after WT B16 (WT) and B2m^-/- B16 (B2m^-/-) melanoma tail vein IV injection in comparison with untreated mice (Ф). (C) Quantification of total CD4⁺ T cells in PB, liver and lungs at day 14 after WT B16 (WT) and B2m^-/- B16 (B2m^-/-) melanoma tail vein IV injection in comparison with untreated mice (Ф). (D) Quantification of lung-infiltrating IV^neg CD4⁺ T cells at day 14 after WT B16 (WT) and B2m^-/- B16 (B2m^-/-) melanoma tail vein IV injection in comparison with untreated mice (Ф). (E) Quantification of total CD8⁺ T cells in PB, liver and lungs at day 14 after WT B16 (WT) and B2m^-/- B16 (B2m^-/-) melanoma tail vein IV injection in comparison with untreated mice (Ф). (F) Quantification of lung-infiltrating IV^neg CD8⁺ T cells at day 14 after WT B16 (WT) and B2m^-/- B16 (B2m^-/-) melanoma tail vein IV injection in comparison with untreated mice (Ф). n = 4-8 mice per group. Graphs show mean + SD. Mann-Whitney U test, ns, not significant, *p < 0.05, **p < 0.01.

Figure 4

Characterization of lung-infiltrating NK cells in response to metastatic melanoma. (A) Quantification of granzyme B⁺/perforin⁺ circulating (IV^pos) and lung-infiltrating (IV^neg) NK cells at day 14 after WT B16 (WT) and B2m^-/- B16 (B2m^-/-) melanoma tail vein IV injection in comparison with untreated mice (Ф). (B) Quantification of conventional (CD49b⁺) and tissue-resident (CD49a⁺) lung-infiltrating (IV^neg) NK cells at day 14 after WT B16 (WT) and B2m^-/- B16 (B2m^-/-) melanoma tail vein IV injection in comparison with untreated mice (Ф). (C) Quantification of maturation status defined by CD11b and CD27 expression in lung-infiltrating (IV^neg) NK cells at day 14 after WT B16 (WT) and B2m^-/- B16 (B2m^-/-) melanoma tail vein IV injection in comparison with untreated mice (Ф). (D) Quantification of TNFα⁺ and IFNγ⁺ lung-infiltrating (IV^neg) NK cells at day 14 after WT B16 (WT) and B2m^-/- B16 (B2m^-/-) melanoma tail vein IV injection in comparison with untreated mice (Ф). n = 4-8 mice per group. Graphs show mean + SD. Mann-Whitney U test, ns, not significant, *p < 0.05, **p < 0.01.

Figure 5

Characterization of lung-infiltrating T cells in response to metastatic melanoma. (A–C) Quantification of PD1⁺ (A), TNFα⁺ (B), and IFNγ⁺ (C) lung-infiltrating (IV^neg) CD4⁺ and CD8⁺ T cells at day 14 after WT B16 (WT) and B2m^-/- B16 (B2m^-/-) melanoma tail vein IV injection in comparison with untreated mice (Ф). n = 4-8 mice per group. Graphs show mean + SD. Mann-Whitney U test, ns, not significant, *p < 0.05, **p < 0.01.