Midostaurin Modulates Tumor Microenvironment and Enhances Efficacy of Anti-PD-1 against Colon Cancer

Abstract

Immunotherapy modulating the tumor microenvironment (TME) immune function has a promising effect on various types of cancers, but it remains as a limited efficacy in colon cancer. Midostaurin (PKC412) has been used in the clinical treatment of fms-like tyrosine kinase 3 (FLT3)-mutant acute myeloid leukemia and has demonstrated immunomodulatory activity. We aimed to evaluate the effect of midostaurin on the modulation of TME and the efficacy of anti-programmed cell death protein 1 (PD-1) against colon cancer. Midostaurin inhibited the growth of murine CT26 and human HCT116 and SW480 cells with multinucleation and micronuclei formation in morphology examination. The cell cycle arrested in the G2/M phase and the formation of the polyploid phase was noted. The formation of cytosolic DNA, including double-strand and single-strand DNA, was increased. Midostaurin increased mRNA expressions of cGAS, IRF3, and IFNAR1 in colorectal adenocarcinoma cells and mouse spleen macrophages. The protein expressions of Trex-1, c-KIT, and Flt3, but not PKCα/β/γ and VEGFR1, were down-regulated in midostaurin-treated colorectal adenocarcinoma cells and macrophages. Trex-1 protein expression was abrogated after FLT3L activation. In vivo, the combination of midostaurin and anti-PD-1 exhibited the greatest growth inhibition on a CT26-implanted tumor without major toxicity. TME analysis demonstrated that midostaurin alone decreased Treg cells and increased neutrophils and inflammatory monocytes. NKG2D⁺ and PD-1 were suppressed and M1 macrophage was increased after combination therapy. When combined with anti-PD-1, STING and INFβ protein expression was elevated in the tumor. The oral administration of midostaurin may have the potential to enhance anti-PD-1 efficacy, accompanied by the modulation of cytosolic DNA-sensing signaling and tumor microenvironment.

Keywords: anti-programmed cell death protein 1 (PD-1); cGAS-STING signaling; colon cancer; midostaurin (PKC412); tumor microenvironment.

Figure 1

Cell viability of mouse and human colorectal adenocarcinoma after treatment with midostaurin. Mouse colorectal adenocarcinoma CT26 cells (a), human colorectal carcinoma HCT116 (b), and SW480 cells (c) were treated with different concentrations of midostaurin (PKC412) for 24, 48, and 72 h. Cell viability was measured by MTT reduction assay. Data from four separate experiments were expressed as mean ± standard error of the mean (SEM). Significant differences between control cells and cells treated with PKC412 are indicated by * p < 0.05, ** p < 0.01, and *** p < 0.001.

Figure 2

Cell-cycle analysis of colorectal adenocarcinoma cells treated with midostaurin (PKC412). Mouse and human colorectal adenocarcinoma cells were treated with midostaurin at 0 (a,d,g), 5 (b,e,h), and 20 μM (c,f,i) for 24 h. After treatment, cells were fixed and stained with propidium iodide. Cell-cycle distribution was acquisitioned by flow cytometry. Representative DNA histograms of colorectal cells treated with PKC412 were shown and the expression percentage of each cell-cycle phase was indicated in the panels. Data from four separate experiments were expressed as mean ± SEM.

Figure 3

Morphology examination of colorectal carcinoma cells treated with midostaurin (PKC412). After treatment, morphologies were stained with Liu’s stain and photographed under light microscopy in 1000-fold magnification. The representative morphology of CT26 (a), HCT116 (b), and SW480 (c) cells was shown after staining. For micronucleic detection, CT26 cells were washed with PBS and fixed in 4% paraformaldehyde. The slides were subsequently stained with 1 μg/mL of DAPI for 10 min. The morphology and fluorescence expression were visualized and photographed under ImageXpress^® Micro 4 microscope in 200-fold magnification (d). Arrow (red) indicates the micronucleic formation. The proportion of micronucleic induced by PKC412 was counted (e). Data from three separate experiments were expressed as mean ± SEM. Significant differences between control cells and cells treated with PKC412 are indicated by * p < 0.05 and ** p < 0.01.

Figure 4

Cytosolic DNA expression profiles after midostaurin (PKC412) was administered in colorectal adenocarcinoma cells. CT26 (a,b), HCT116 (c,d), and SW480 (e,f) cells were grown in monolayer and treated with midostaurin for 24 h. After treatment, cells were detached, harvested, and the cell pellet was added with lysis solution to disrupt the cells. The samples were then centrifuged at 1200 rpm for 5 min and the supernatant containing cytosolic DNA was collected. The double-strand DNA (dsDNA) (a,c,e) and single-strand DNA (ssDNA) (b,d,f) were measured at 260 nm wavelength. Data from three separate experiments were expressed as mean ± SEM. Significant differences between control cells and cells treated with PKC412 are indicated by * p < 0.05 and ** p < 0.01.

Figure 5

mRNA expression levels in CT26 (a), HCT116 (b), and SW480 (c) cells after treatment with midostaurin (PKC412). After treatment, RNA samples were isolated, followed by reverse transcription of RNA using primers of DNA-sensing specific genes. Data from four independent experiments were expressed as mean ± SEM. Significant differences between control cells and cells treated with PKC412 are indicated by * p < 0.05 and *** p < 0.001.

Figure 6

Effect of midostaurin (PKC412) on expression of putative target proteins in CT26 (a), HCT116 (b), and SW480 (c). Cells were treated with PKC412 for 24 h and then cells were lysed and harvested. Equal amounts of proteins were subjected to immunoblotting to detect midostaurin targets and Trex-1 proteins. Representative blots of midostaurin target proteins were shown. The densitometric analysis of each band was shown in the bottom panel. Original Western Blot images can be found in Figure S1.

Figure 7

Effect of Trex-1 protein in midostaurin (PKC412)/FLT3L-treated colorectal adenocarcinoma cells. CT26 (a), HCT116 (b), and SW480 (c) cells were treated with PKC412 or/with FLT3L for 24 h and then cells were lysed and harvested. Equal amounts of proteins were subjected to immunoblotting to detect Flt3 and Trex-1 proteins. Representative blots of target proteins were shown. The densitometric analysis of each band was shown in the bottom panel. Original Western Blot images can be found in Figure S1.

Figure 8

Effect of DNA-sensing signaling and molecular target proteins in midostaurin (PKC412)-treated spleen macrophages. Macrophages were treated with PKC412 for 2 or 24 h and then cells were lysed and harvested. RNA samples were isolated followed by reverse transcription of RNA using primers of DNA-sensing specific genes (a). Equal amounts of proteins were subjected to immunoblotting to detect midostaurin targets and Trex-1 proteins. Representative blots of midostaurin target proteins were shown (b). The densitometric analysis of each band was shown in the bottom panel. Data from three mice spleen macrophages were expressed as mean ± SEM. Significant differences between control cells and cells treated with PKC412 are indicated by * p < 0.05 and *** p < 0.001. Original Western Blot images can be found in Figure S1.

Figure 9

Therapeutic effect and toxicity in a CT26 syngeneic animal model treated with midostaurin (PKC412), anti-PD-1, and combination of midostaurin and anti-PD-1. BALB/c mice were implanted with CT26 colon cancer cells for tumor growth. After seven days, mice were administered with PKC412 100 mg/kg, anti-PD-1 antibody (200 μg), or a combination of PKC412 with anti-PD-1 antibody. Tumor volume was recorded using electronic caliper (a). Biological toxicities were examined by white blood cell counts (b), body weight (c), renal function with creatinine (d), and liver function with alanine transaminase (ALT) (e). Data from six mice of each group were expressed as mean ± SEM. Significant differences between the control group and the drug-treated group are indicated by *** p < 0.001. Significant differences between the monotherapy group and the combination group are indicated by ++ p < 0.01.

Figure 10

Expression profiles of circulating immune cells in a CT26 syngeneic animal model after monotherapy or combination of midostaurin (PKC412) and anti-PD-1. At the end of the experiment, mice were sacrificed and tumor and spleen specimens were harvested for flow cytometry analysis immune cell expression. The gating strategy of tumor and spleen specimens for flow cytometry analysis was shown (a,b). Quantification of immune cell profiles in tumor (c) and spleen (d) on day 39 after tumor cells implanted in syngeneic BALB/c mice. Data from six mice of each group were expressed as mean ± SEM. Significant differences between control group and PKC412-treated group, or anti-PD-1-treated group and combination-treated group are indicated by * p < 0.05.

Figure 11

Expression levels of STING and IFNβ in a CT26 syngeneic animal model after monotherapy or combination of midostaurin (PKC412) and anti-PD-1. At the end of the experiment, mice were sacrificed and tumor specimens were harvested for immunohistochemistry staining STING (a) and IFNβ (b) protein expressions. The score of STING and IFNβ protein expressions level in tumor (c) after tumor cells implanted in syngeneic BALB/c mice. Data from six mice of each group were expressed as mean ± SEM. Significant differences between control group and PKC412-treated group, or anti-PD-1-treated group and combination-treated group are indicated by * p < 0.05, ** p < 0.01.