Increased expression of programmed cell death protein 1 on NK cells inhibits NK-cell-mediated anti-tumor function and indicates poor prognosis in digestive cancers - PubMed (original) (raw)
Increased expression of programmed cell death protein 1 on NK cells inhibits NK-cell-mediated anti-tumor function and indicates poor prognosis in digestive cancers
Y Liu et al. Oncogene. 2017.
Abstract
Abnormal expression of activating/inhibitory receptors leads to natural killer (NK) cells dysfunction in tumor. Here we show that programmed cell death protein 1 (PD-1), a well-known immune checkpoint of T cells, is highly expressed on peripheral and tumor-infiltrating NK cells from patients with digestive cancers including esophageal, liver, colorectal, gastric and biliary cancer. The increased PD-1 expression on NK cells indicates poorer survival in esophageal and liver cancers. Blocking PD-1/PD-L1 signaling markedly enhances cytokines production and degranulation and suppresses apoptosis of NK cells in vitro. PD-1/PD-L1 exerts inhibitory effect through repressing the activation of PI3K/AKT signaling in NK cells. More importantly, a PD-1 blocking antibody was found to significantly suppress the growth of xenografts in nude mice, and this inhibition of tumor growth was completely abrogated by NK depletion. These findings strongly suggested that PD-1 is an inhibitory regulator of NK cells in digestive cancers. PD-1 blockade might be an efficient strategy in NK cell-based tumor immunotherapy.
Conflict of interest statement
The authors declare no conflict of interest.
Figures
Figure 1
PD-1 expression is upregulated on peripheral NK cells isolated from digestive cancer patients and indicates poor prognosis. (a) Lymphocytes were gated according to forward scatter and side scatter. CD3 and CD56 staining was used to identify NK cells. (b) Graph comparing PD-1 expression on NK cells from patients with digestive cancers and healthy individuals. (c) CD3− cells from a representative patient was further gated into two NK-cell subsets on the basis of CD56 and CD16 expression (left). Plots showing PD-1 expression on CD16+CD56dim(right) or CD16-CD56bright(middle) NK cells. (d) Graph comparing PD-1 expression on CD3−CD16-CD56bright(left) and CD3−CD16+CD56dim(right) NK cells in patients with digestive cancers and healthy individuals. (e) Graphs comparing PD-1 expression on CD3−CD56+ NK cells in ESCC patients with different degrees of pathological differentiation (left) or different T (middle) or N stages (right). PD-1 expression level on NK cells is represented as the percentage of PD-1+ NK cells. Cumulative data are shown as mean±s.e.m., analyzed by Student’s _t_-test. The data were normally distributed and had constant variances. *_P_⩽0.05, **_P_⩽0.01, ***_P_⩽0.001.
Figure 2
PD-1 is expressed on tumor-infiltrating NK cells. (a) Immunofluorescence staining was performed in an HCC tissue microarray (_n_=33). Representative immunofluorescence images showing co-localization of PD-1 on CD56+ NK cells. The clinical characteristics of these patients are listed in Supplementary Table S2. (b) Histograms showing PD-1 expression on CD3−CD56+ NK cells isolated from paratumor and tumor tissues from a representative patient (upper and middle) and pooled data (low). (c) Dot plots showing PD-1 expression on CD16-CD56bright and CD16+CD56dim NK cells infiltrated in tumor tissues. (d) Dot plots showing the percentage of CD49a+ NK cells in tumor tissues (left). Histograms showing high PD-1 expression on CD49a+ tumor-infiltrating NK cells (right). Cumulative data are shown as mean±s.e.m., analyzed by Student’s _t_-test. The data were normally distributed and had constant variances. *_P_⩽0.05, **_P_⩽0.01, ***_P_⩽0.001.
Figure 3
PD-1/PD-L1 blockade increased IFN-γ production and CD107a expression in NK cells. (a and b) Plots from a representative ESCC patient (a) and pooled data showing the induced surface expression of PD-1 on peripheral NK cells after PMA and ionomycin (b left panel) or IL-12 and IL-15 (b right panel) stimulation. Pooled data from ESCC patients showing dynamically increasing PD-1 expression after PMA and ionomycin stimulation (b middle panel). (c) NK cells from ESCC patients were stimulated with PMA and ionomycin for 5 h, after which CD107a degranulation(upper panel) was compared between PD-1+ and PD-1− NK cells. Expression of informative surface markers, NKG2A (middle panel) and NKG2D (low panel), was also compared between PD-1+ and PD-1− NK cells. (d–f) After pre-incubation with 2 μg/ml (d) or 8 μg/ml (e) PD-L1 blocking antibodies (aPD-L1) or 2 μg/ml PD-1 blocking antibodies (aPD-1) (f) and matched IgG, NK cells isolated from ESCC patients were stimulated with PMA and ionomycin for 5 h. IFN-γ secretion and CD107a degranulation of NK cells were compared between two groups. Representative flow cytometry analyses from one patient are located in the left and middle panels. Pooled data are located in the right panel. Cumulative data are shown as mean±s.e.m., analyzed by Student’s _t_-test. The data were normally distributed and had constant variances. *_P_⩽0.05, **_P_⩽0.01, ***_P_⩽0.001.
Figure 4
PD-1/PD-L1 signaling induces apoptosis of PD-1+ NK. (a) Representative dot plots (left panel) depicting the spontaneous apoptosis (annexin-V staining) of PD-1+ and PD-1− NK cells from one ESCC patient and pooled data (right panel). (b) NK cells from ESCC patients were cultured without or with plate-bound anti-PD-1 (aPD-1) (8 μg/ml), and apoptosis (annexin-V staining) of PD-1− and PD-1+ NK cells was compared between the two groups. PMA and ionomycin were applied for 3 h to stimulate activation-induced apoptosis. (c) After pre-incubation with 8 μg/ml IgG or PD-L1 blocking antibodies (aPD-L1), NK cells isolated from ESCC patients were stimulated with PMA and ionomycin and apoptosis of PD-1− and PD-1+ NK cells were compared. Cumulative data are shown as mean±s.e.m., analyzed by Student’s _t_-test. The data were normally distributed and had constant variances. *_P_⩽0.05, **_P_⩽0.01, ***_P_⩽0.001.
Figure 5
PD-1/PD-L1 blockade enhances the pAKT signaling pathway in NK92 and NKL cells. (a and b) After pre-incubation with 2 μg/ml IgG or matched PD-1/PD-L1 blocking antibodies (aPD-1 or aPD-L1), NK92 or NKL cells were stimulated with PMA and ionomycin (a) or EC9706 target cells for 1 h, and western blot analysis was used to determine AKT phosphorylation. The results in a and b are representative of three independent experiments.
Figure 6
Administration of PD-1 blocking antibodies inhibits tumor growth through enhancing NK-cell activity in nude mice. (a–d) The EC9706 tumor-bearing nude mouse models were established and treated with anti-PD-1 blocking antibodies (aPD-1) or matched IgG. (a) Images of tumors at the time of killing from each group are shown. (b) Tumor size was monitored every 3~4 days. (c) Flow cytometric plots showing CD69 expression on CD3−DX5+ NK cells from spleen and tumor tissues. (d) Summarized data showing CD69 expression on NK cells from spleen. (e–f) NK cells were depleted as previously described in the methods and the tumor-bearing mouse models were then established. Anti-PD-1 antibodies (aPD-1) or IgG was administered as before. (e) Flow cytometric plots showing the successful depletion of NK cells. (f) Images of tumors at the time of killing (left panel) and tumor weight (right panel) from each group are shown. The size of the tumors was also monitored (middle panel). Cumulative data are shown as mean±s.e.m., analyzed by Student’s _t_-test. The data were normally distributed and had constant variances. *_P_⩽0.05, **_P_⩽0.01, ***_P_⩽0.001.
Figure 7
Graphic model as discussed in the text. PD-1 expression on NK cells is enhanced in digestive cancer patients, and is further induced when stimulated. Ligation of PD-1 on NK cells with PD-L1 on tumor cells inhibits activation PI3K/AKT signaling of NK cells and thus suppresses NK–cell-mediated anti-tumor activity. PD-1/PD-L1 blockade treatment rescues NK-cell function and may benefit digestive cancer patients.
References
- Torre LA, Bray F, Siegel RL, Ferlay J, Lortet-Tieulent J, Jemal A. Global cancer statistics, 2012. CA Cancer J Clin 2015; 65: 87–108. - PubMed
- Giannelli G, Rani B, Dituri F, Cao Y, Palasciano G. Moving towards personalised therapy in patients with hepatocellular carcinoma: the role of the microenvironment. Gut 2014; 63: 1668–1676. - PubMed
- Zhang L, Song X, Mohri Y, Qiao L. Role of inflammation and tumor microenvironment in the development of gastrointestinal cancers: what induced pluripotent stem cells can do? Curr Stem Cell Res Ther 2015; 10: 245–250. - PubMed
Publication types
MeSH terms
Substances
LinkOut - more resources
Full Text Sources
Other Literature Sources
Molecular Biology Databases
Research Materials