Figure 1.
NK cells express all class I PI3K isoforms but isoform-selective inhibitors have little effect on cytotoxicity.
(A) Immunoblots were prepared using equivalent protein amounts from lysates of purified NK cells, T cells, B cells or two solid tumor cell lines. Blots were probed with PI3K isoform-specific antibodies and for β-actin as a loading control. (B) RMA-S, YAC-1, and EL4H60 cells were labeled with calcein AM and co-cultured with NK cells at the indicated E:T ratios in the presence of 1 µM indicated inhibitors (TGX-221 was 0.5 µM) for 2 h. (C) RMA-S and YAC-1 cells were labeled with calcein AM and co-cultured with NK cells at 10∶1 E:T ratio in the presence of 1 µM indicated inhibitors for 2 h. (D) NK cells were treated with a higher concentration of MLN1117 (2 µM) and incubated with calcein AM-labeled YAC-1 cells at 10∶1 E:T ratio. Percentage specific lysis was calculated from pentaplicate cultures as follows: ((mean experimental release - mean spontaneous release)/(mean maximum release -mean spontaneous release)) x 100. The values are presented as the means ± SEM from three independent experiments. *p<0.05 and **p<0.01, as determined by the unpaired Student's t test using Sigma Plot 10 software and as compared to the vehicle control group.
Table 1.
Selectivity and effective concentrations of inhibitors used in this study (reported in the following references: 11, 26, 31, 34, 38, 40).
Figure 2.
A pan-PI3K inhibitor, but not isoform-selective inhibitors, suppresses ADCC.
CCRF-CEM, human acute T lymphocytic leukemia cell line was labeled with calcein AM, incubated with mouse anti-human CD4 mAb, and then co-cultured with NK cells at 10∶1 E:T ratio in the presence of indicated inhibitors for 2 h. Culture supernatants were collected and calcein fluorescence was measured and analyzed as in Figure 1. The values are presented as the means ± SEM from three independent experiments. *p<0.05 and **p<0.01, as determined by the unpaired Student's t test using Sigma Plot 10 software and as compared to the vehicle control group.
Figure 3.
Isoform-selective inhibitors have little effect on cytotoxicity of human NK cells.
(A) K562 cells were labeled with 51Cr and co-cultured with human primary NK cells at the indicated E:T ratios in the presence of 1 µM indicated inhibitors (TGX-221, GDC-0941, and ZSTK474 were 0.5 µM) for 2 h. Specific 51Cr release was measured as described [46]. (B) K562 cells were labeled with calcein AM and co-cultured with human NKL cells at 10∶1 E:T ratio in the presence of 1 µM indicated inhibitors (TGX-221 was 0.5 µM) for 2 h. Culture supernatants were collected and calcein fluorescence was measured and analyzed as in Figure 1. The data are expressed as the means ± SEM of three independent experiments. An asterisk denotes any response that is significantly different from the vehicle control group as determined by the unpaired Student's t test using Sigma Plot 10 software (*, p<0.05, **, p<0.01, and ***, p<0.001).
Figure 4.
Isoform-selective inhibitors partially reduce AKT phosphorylation in anti-NKG2D-stimulated NK cells.
NK cells from C57BL/6 mice were purified from spleen and expanded for 7–8 days in IL-2. (A) YAC-1 cells were labeled with calcein AM and co-cultured with NK cells at the indicated E:T ratios in the presence of AKTi, a specific inhibitor of AKT1 and AKT2, for 2 h. Culture supernatants were collected and calcein fluorescence was measured. Percentage specific lysis was calculated from pentaplicate cultures as follows: ((mean experimental release - mean spontaneous release)/(mean maximum release - mean spontaneous release)) x 100. Data presented are means ± SEM from three independent experiments. An asterisk denotes any response that is significantly different from the vehicle control group as determined by the unpaired Student's t test using Sigma Plot 10 software (*, p<0.05, **, p<0.01, and ***, p<0.001). (B) NK cells were stimulated with plate-bound anti-NKG2D mAb for the indicated times. Whole-cell lysates were probed for pAKT (S473) and reprobed for total AKT. (C) NK cells were pretreated with 1 µM of the indicated inhibitors (TGX-221 was 0.5 µM) before anti-NKG2D stimulation for 20 minutes. Similar results were obtained in two additional experiments, though p110δ inhibitor IC87114 caused less reduction in pAKT than GDC-0941 in other experiments.
Figure 5.
Selective p110α inhibitors have lesser effect on IFN-γ production in anti-NKG2D- or anti-NK1.1-stimulated NK cells.
IL-2-expanded mouse NK cells were stimulated with plate-bound anti-NKG2D mAb (A–B) or anti-NK1.1 mAb (C) in the presence of 1 µM indicated inhibitors (TGX-221 was 0.5 µM) (A) or 250 nM, 500 nM, and 1 µM indicated inhibitors (B–C). After 18 h stimulation, the supernatants were collected and IFN-γ secretion was determined by ELISA (A, left). After 6 h stimulation, total RNA was isolated and IFN-γ mRNA expression was determined by real-time RT-PCR using SYBR Green (A, right). The relative transcript quantities were calculated as described in Materials and Methods section and normalized using the unstimulated group defined as 1. After 18 h stimulation, the NK cells were harvested and the intracellular IFN-γ level was determined by flow cytometry (B–C). Brefeldin A was added for the last 4 h before cell harvest and IFN-γ production was measured in CD3−NK1.1+ NK cells by intracellular staining. The data are expressed as the means ± SEM of three independent experiments. Statistical analysis was performed with one-way ANOVA using Prism 6 (GraphPad Software, Inc.) to compare the differences between vehicle and each inhibitor-treated group. *, p<0.05, **, p<0.01, and ***, p<0.001. (D) NK cells were pre-treated with vehicle (0.1% DMSO) or 1 µM GDC-0941 for 15 min, followed by stimulation with plate-bound anti-NKG2D mAb in the absence or presence of IL-12 and IL-18 for 18 h. The intracellular IFN-γ level was determined as in B–C.
Figure 6.
Selective p110α inhibitors have lesser effect on cytokine/chemokine production in NK cells.
NK cells from C57BL/6 mice were purified from spleen and expanded for 7–8 days in IL-2. (A–B) The NK cells were pre-treated with vehicle (0.1% DMSO or 0.1% ethanol) or 1 µM indicated inhibitors (TGX-221 was 0.5 µM) for 15 min, followed by stimulation with plate-bound anti-NKG2D mAb in the absence (A) or presence (B) of IL-12 and IL-18 for 18 h. The supernatants were collected and analyzed by Multiplex assay. The data are expressed as the means ± SEM of four independent NK cell preparations that were stimulated and analyzed concurrently. Statistical analysis was performed with one-way ANOVA using Prism 6 (GraphPad Software, Inc.) to compare the differences between vehicle and each inhibitor-treated group. *, p<0.05, **, p<0.01, and ***, p<0.001.
Figure 7.
Effects of PI3K inhibitors on additional cytokines and chemokines produced by NK cells.
Data from additional bioplex analytes are shown as in Figures 5 and 6. Displayed are analytes present at concentrations greater than 1 ng/ml in at least one condition, based on the Bioplex product information showing a limit of detection between 0.1–1 ng/ml.
Figure 8.
PI3K inhibitors have minimal effect on the development of NK cells in vivo.
Wild-type C57BL/6 mice were orally dosed with vehicle, the p110α selective inhibitor MLN1117 (60 mg/kg), or the pan-PI3K inhibitor GDC-0941 (70 mg/kg) for 7 days (n = 4). Spleen and bone marrow (BM) were isolated and analyzed for NK cell development by flow cytometry. (A) Splenocytes were stained with CD21 and CD23 for marginal zone (MZ) B cell population. Among the B220+ population, CD21highCD23low cells were gated for MZ B cell population. (B, C) Percentage of B/T cells and Th/Tc cells in spleen were determined using surface staining with anti-B220/anti-Thy1.2 mAbs and anti-CD4/anti-CD8 mAbs, respectively. (D) NK cells as specified by CD3−NK1.1+ were gated and analyzed in single-cell suspension of spleen and BM, respectively. (E) Spleen- or BM-derived CD3−NK1.1+ NK cells were analyzed for NK subset based on the expression level of CD11b and CD27. (F) Spleen- or BM-derived CD3−NCR1+ NK cells were analyzed for maturation marker, Ly49A, LY49C/I, Ly49D, Ly49G2, and Ly49H. (G) Spleen- or BM-derived CD3−NCR1+ NK cells were analyzed for the expression of CD122, NKG2D, and NKG2A/C/E. The data are expressed as the means ± SEM. Statistical analysis was performed with the unpaired Student's t test using Sigma Plot 10 software to compare the differences between vehicle and each inhibitor-treated group. *, p<0.05, **, p<0.01, and ***, p<0.001.