Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The . gov means it’s official VSports app下载. Federal government websites often end in . gov or . mil. Before sharing sensitive information, make sure you’re on a federal government site. .

Https

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely V体育官网. .

. 2017 Jan 28;109(6):djw283.
doi: 10.1093/jnci/djw283. Print 2017 Jan.

The MLL1-H3K4me3 Axis-Mediated PD-L1 Expression and Pancreatic Cancer Immune Evasion

Chunwan Lu  1 , Amy V Paschall  1 , Huidong Shi  1 , Natasha Savage  1 , Jennifer L Waller  1 , Maria E Sabbatini  1 , Nicholas H Oberlies  1 , Cedric Pearce  1 , Kebin Liu  1
Affiliations

"VSports" The MLL1-H3K4me3 Axis-Mediated PD-L1 Expression and Pancreatic Cancer Immune Evasion

Chunwan Lu et al. J Natl Cancer Inst. .

V体育官网 - Abstract

Background: Pancreatic cancer is one of the cancers where anti-PD-L1/PD-1 immunotherapy has been unsuccessful. What confers pancreatic cancer resistance to checkpoint immunotherapy is unknown VSports手机版. The aim of this study is to elucidate the underlying mechanism of PD-L1 expression regulation in the context of pancreatic cancer immune evasion. .

Methods: Pancreatic cancer mouse models and human specimens were used to determine PD-L1 and PD-1 expression and cancer immune evasion. Histone methyltransferase inhibitors, RNAi, and overexpression were used to elucidate the underlying molecular mechanism of PD-L1 expression regulation. All statistical tests were two-sided. V体育安卓版.

Results: PD-L1 is expressed in 60% to 90% of tumor cells in human pancreatic carcinomas and in nine of 10 human pancreatic cancer cell lines V体育ios版. PD-1 is expressed in 51. 2% to 52. 1% of pancreatic tumor-infiltrating cytotoxic T lymphocytes (CTLs). Tumors grow statistically significantly faster in FasL-deficient mice than in wild-type mice (P = . 03-. 001) and when CTLs are neutralized (P = . 03-<. 001). H3K4 trimethylation (H3K4me3) is enriched in the cd274 promoter in pancreatic tumor cells. MLL1 directly binds to the cd274 promoter to catalyze H3K4me3 to activate PD-L1 transcription in tumor cells. Inhibition or silencing of MLL1 decreases the H3K4me3 level in the cd274 promoter and PD-L1 expression in tumor cells. Accordingly, inhibition of MLL1 in combination with anti-PD-L1 or anti-PD-1 antibody immunotherapy effectively suppresses pancreatic tumor growth in a FasL- and CTL-dependent manner. .

Conclusions: The Fas-FasL/CTLs and the MLL1-H3K4me3-PD-L1 axis play contrasting roles in pancreatic cancer immune surveillance and evasion. Targeting the MLL1-H3K4me3 axis is an effective approach to enhance the efficacy of checkpoint immunotherapy against pancreatic cancer VSports最新版本. .

PubMed Disclaimer

Figures

Figure 1.
Figure 1.
PD-L1 and PD-1 expression profiles in pancreatic cancer in vivo. A) PANC02-H7 cells were surgically transplanted to pancreas of C57BL/6 mice to establish orthotopic pancreatic tumor (yellow arrow). B) PD-L1 expression level on cultured pancreatic cancer cells (in vitro) and the tumor (in vivo) as shown in (A). The PD-L1 protein level (mean fluorescent intensity [MFI]) was analyzed by two-sided t test and presented at the right. Column: mean; bar: SD. C) Cell suspensions were prepared from pancreatic tumor tissues, spleen, LN, and peripheral blood from five tumor-bearing mice (n = 5) and stained with 7-AAD, CD8-, and PD-1-specific antibodies. Stained cells were analyzed by flow cytometry. D) Quantification of CD8+PD-1+ CTLs in the indicated tissues. 7-AAD+ dead cells were gated out. CD8+ cells were gated and analyzed for PD-1+ cells. E) UN-KC-6141 cells were surgically transplanted to the pancreas of C57BL/6 mice to establish orthotopic pancreatic tumor (yellow arrow). F) PD-L1 expression levels in cultured UN-KC-6141 cells (in vitro) and orthotopic UN-KC-6141 tumor (in vivo). The PD-L1 protein level (MFI) was analyzed by two-sided t test and presented at the right. Column: mean; bar: SD. G) Analysis of CD8+PD-1+ cells in the UN-KC-6141 tumor model in the indicated tissues. H) Quantification of CD8+PD-1+ CTLs in the indicated tissues. 7-AAD = 7-aminoactinomycin D; LN = lymph node; MFI = mean fluorescent intensity.
Figure 2.
Figure 2.
Function of MLL1 in H3K4me3 regulation at the cd274 promoter and PD-L1 expression in tumor cells. A) The human CD274 (left panel) and mouse cd274 (right panel) gene promoter structures. B) Chromatin immunoprecipitation (ChIP) analysis of H3K4me3 levels in the CD274 promoter region in MiaPaca-2 cells. C) ChIP analysis of H3K4me3 levels in the cd274 promoter region in PANC02-H7 (left panel) and UN-KC-6141 (right panel) cells. D) ChIP analysis of H3K4me3 levels in the cd274 promoter region in normal mouse pancreatic tissues (left panel) and the orthotopic PANC02-H7 (middle panel) and UN-KC-6141 (right panel) tumor tissues, respectively. E) ChIP DNA fragments prepared as in (D) from normal mouse pancreas (n = 2: N1 and N2), the orthotopic PANC02-H7 tumors (n = 2: H1 and H2) and orthotopic UN-KC-6141 tumors (n = 2:U1 and U2) were cloned to construct DNA library and sequenced with a high-throughout sequencing system. The normalized sequence reads of H3K4me3 were mapped to cd274 gene region (-4000 of the 5’ regulatory region and the entire cd274 gene) on mouse chromosome 19 (left panel). The -4000 to + 1000 region relative to cd274 transcription start site (+1) is shown at the right panel. F) C57BL/6 mouse pancreas (n = 3) and tumor tissues from PNAC02-H7 tumor-bearing (n = 3) and UN-KC-6141 tumor-bearing (n = 3) mice were analyzed by real-time polymerase chain reaction (top panel) for MLL1 and PD-L1 transcript levels and by immunoblotting (bottom panel) for MLL1 protein level. G) ChIP analysis of MLL1 protein association with the cd274 promoter chromatin in PANC02-H7 cells (in vitro) and in orthotopic PANC02-H7 tumors (in vivo). H) MiaPaCa-2 cells were transiently transfected with scramble- and MLL1-specific siRNA, respectively, for 24 hours and analyzed for MLL1 mRNA level by quantitative polymerase chain reaction (qPCR), H3K4me3 level at the CD274 promoter region by ChIP (ChIP 4 as in (B)), and PD-L1 mRNA level by qPCR as indicated. I) PANC02-H7 cells were transiently transfected with scramble- and MLL1-specific siRNAs, respectively, for 24 hours and analyzed for MLL1 mRNA level by qPCR, H3K4me3 level in the cd274 promoter region by ChIP (ChIP 4 as in (C)), and PD-L1 mRNA level by qPCR as indicated. All statistical t tests were two-sided. ChIP = chromatin immunoprecipitation.
Figure 3.
Figure 3.
MLL1 and PD-L1 protein levels in human pancreatic carcinoma cells. A) Immunohistochemical staining of human tissues. A1a and b) Human tonsil tissues were stained with anti-PD-L1 antibody. Shown are low (A1a) and high magnification (A1b) images of the stained tonsil tissue. A1a) The yellow arrows indicate PD-L1+ cells, the black arrow points to crypts, and green arrows point to lymphoid cells. A1b) The yellow arrows indicate PD-L1+ epithelial cells surrounding the crypt. A1c) Negative staining without anti-PD-L1 antibody. A2a–c) Human adrenal tumor tissues were stained with anti-human PD-L1 antibody. Shown are low (A2a) and high magnification (A2b and c) images of the stained adrenal tumor tissue. A2a) The yellow arrow indicates tumor cells, and the black arrow indicates tumor cells and leukocytes aggregate area. A2b) The yellow arrow indicates PD-L1+ tumor cells. A2c) The yellow arrow indicates PD-L1+ tumor cells, and the black arrows point to PD-L1+ monocytes. B1a and b and C1a and b) Two human pancreatic adenocarcinoma specimens were stained with anti-MLL1 antibody. Shown are low (B1a and C1a) and high magnification (B1b and C1b) images of the stained tumor tissues. Red arrows indicate MLL1+ tumor cells. B2a and b and C2a and b) The two human pancreatic adenocarcinoma specimens as shown in (B1 and C1) were stained with anti-PD-L1 antibody. Shown are low (B2a and C2a) and high magnification (B2b and C2b) images of the stained tumor tissues. Yellow arrows indicate PD-L1+ tumor cells. Pink arrows in (B2b and C2b) indicate PD-L1 protein staining in both cell membrane and cytoplasm. MLL1- and PD-L1-specific staining is indicated by brown color, and nuclei were counterstained with hematoxylin in blue. B) Ten human pancreatic cancer cell lines were stained with IgG isotype control and PD-L1-specific monoclonal antibody, respectively, and analyzed for PD-L1 protein level by flow cytometry. C) Quantification of PD-L1 protein levels on human pancreatic cancer cell surface. The mean fluorescent intensity of PD-L1 protein as analyzed in (B) were quantified and presented. Column: mean; bar: SD. Scale bar (redline at the bottom left) = 50 μm.
Figure 4.
Figure 4.
Function of MLL1 in regulation of PD-L1 expression through H3K4me3. A) PANC02-H7 cells were treated with verticillin A at the indicated doses for two days and analyzed by chromatin immunoprecipitation (ChIP) for H3K4me3 level in the ChIP 4 region of the cd274 promoter region. B) The untreated and verticillin A–treated (50 nM) PANC02-H7 cells as shown in (A) were analyzed for PD-L1 mRNA levels by quantitative polymerase chain reaction (qPCR). C) The untreated and verticillin A–treated (50 nM) PANC02-H7 cells as shown in (A) were stained with PD-L1-specific monoclonal antibody (MAb) and analyzed by flow cytometry. PD-L1 protein mean fluorescent intensity (MFI) is presented at the right. D) PANC02-H7 tumor-bearing mice (n = 5) were treated with verticillin A at 0.5 mg/kg body weight once every two days for 10 days, and the tumor tissues were analyzed by ChIP for H3K4me3 level as in (A). The H3K4me3 level is normalized to input. E) The orthotopic PANC02-H7 tumor tissues from untreated and verticillin A–treated mice (n = 5) were analyzed for PD-L1 mRNA levels by qPCR. F) Tumor cells from untreated and verticillin A–treated mice were stained with PD-L1-specific MAb and analyzed by flow cytometry. PD-L1 protein MFI is presented at the right. All statistical t tests were two-sided. ChIP = chromatin immunoprecipitation; MFI = mean fluorescent intensity; qPCR = quantitative polymerase chain reaction.
Figure 5.
Figure 5.
Function of MLL1 in regulation of H3K4me3 in the CD274 promoter and PD-L1 expression in human pancreatic cancer cells. A) Chaetocin is a potent MLL1 inhibitor. Chaetocin was tested in a 10-dose IC50 mode with threefold serial dilutions using 3H-S-adenosyl-methionine as substrate with MLL1 recombinant protein complex. The enzyme activity was then analyzed and plotted against chaetocin concentrations. IC50 was calculated using the Graphpad Prism program. B) PANC10.05 cells were treated with chaetocin (50 nM) for two days and analyzed with H3K4me3-specific monoclonal antibody (MAb) by chromatin immunoprecipitation (ChIP). The ChIP DNA was amplified using ChIP primer 4. The relative H3K4me3 level was normalized to input DNA and analyzed by two-sided t test. C) PANC10.05 cells were treated with chaetocin at the indicated doses for two days, stained with PD-L1-specific MAb, and analyzed by flow cytometry. The PD-L1 protein mean fluorescent intensity is presented at the right panel. Column: mean; bar: SD. D) H3K9me3 is not enriched in the CD274 promoter region in human pancreatic cancer cells. PANC10.05 cells were analyzed by ChIP with H3K9me3-specific MAb and polymerase chain reaction primers. All statistical t tests were two-sided. ChIP = chromatin immunoprecipitation; mAb = monoclonal antibody.
Figure 6.
Figure 6.
Role of the Fas-FasL system in suppression of pancreatic tumor growth in vivo. A) PANC02-H7 cells were surgically transplanted to the pancreas of wild-type (WT) C57BL/6 (n = 5) and faslgld (n = 5) mice. Shown are the established pancreatic tumors 15 days after tumor transplant. The tumor volume and tumor weight were analyzed by two-sided t test and are presented in the right panels. B) The cultured PANC02-H7 cells and PANC02-H7 tumor tissues from WT mice were analyzed for cell surface Fas protein level by flow cytometry. C) Cells were prepared from the indicated tissues from tumor-bearing WT mice as shown in (A), stained with 7-AAD, CD8-, and FasL-specific monoclonal antibodies (MAbs), and analyzed by flow cytometry. The 7-AAD-CD8+ cells were gated and quantified for FasL+ cells. D) Quantification of % CD8+FasL+ cells as shown in (C). E) UN-KC-6141 cells were surgically transplanted to the pancreas of WT C57BL/6 (n = 5) and faslgld (n = 5) mice. Shown are the established tumors 15 days after tumor transplant. The tumor volume and tumor weight were analyzed by two-sided t test and are presented in the right panels. F) The cultured UN-KC-6141 cells and UN-KC-6141 tumor tissues from WT mice were analyzed for cell surface Fas protein level by flow cytometry. G) Cells were prepared from the indicated tissues from tumor-bearing WT mice as shown in (E), stained with 7-AAD, CD8-, and FasL-specific MAbs and analyzed by flow cytometry. The 7-AAD-CD8+ cells were gated and quantified for FasL+ cells. H) Quantification of % CD8+FasL+ cells as shown in (G). LN = lymph node; WT = wild-type.
Figure 7.
Figure 7.
Pharmacological inhibition of H3K4me3 and efficacy of anti-PD-L1 immunotherapy aganist pancreatic tumor growth in vivo. A) PANC02-H7 cells were surgically transplanted to wild-type (WT) mice. Five days later, tumor-bearing mice were randomly divided into four groups and treated with verticillin A (0.5 mg/kg body weight) and anti-PD-L1 (200 μg/mouse), either alone or in combination (n = 5) once every two days for 10 days. Shown are tumors of the different treated groups. Tumor volume and tumor weight were analyzed by two-sided t tests and are presented in the right panels. B) UN-KC-6141 cells were surgically transplanted to WT mice. Five days later, tumor-bearing mice were treated as in (A). Shown are tumors of the different treated groups. Tumor volume and tumor weight were analyzed by two-sided t tests and are presented in the right panels. C) The four groups of PANC02-H7 tumors as shown in (A) were sectioned and stained for Ki67 (left panel) and TUNEL (right panel). Shown are representative images of one of five tumors: a) control; b) verticillin A; c) anti-PD-L1 monoclonal antibody (MAb); d) verticillin A and anti-PD-L1 MAb. Scale bars in (C) are 20 μm, unless noted otherwise. D) Ki67+ (top panel) and TUNEL+ (bottom panel) cells in tumor tissues as shown in (C) were analyzed by two-sided t tests. Column: mean; bar: SD.
Figure 8.
Figure 8.
Roles of FasL and T cells in enhancement of verticillin A and anti-PD-L1 immunotherapy. A) PANC02-H7 cells were surgically transplanted to faslgld mice. Five days later, tumor-bearing mice were randomly divided into two groups and treated with either solvent or verticillin A (0.5 mg/kg body weight) and anti-PD-L1 monoclonal antibody (MAb; 200 μg/mouse) once every two days for 10 days. Tumor volume and weight were analyzed by two-sided t tests (right panels). B) Comparison of tumor growth inhibition efficacy by combined verticillin A and anti-PD-L1 therapy in wild-type (WT) and faslgld mice. Tumor volume and weight in WT (as shown in Figure 6A) and faslgld (as shown in Figure 7A) mice treated with verticillin A and anti-PD-L1 MAb were compared with those of untreated mice and presented as % of control tumors. The difference between WT and Faslgld mice was analyzed by two-sided t test. Column: mean; bar: SD. C) PANC02-H7 cells were surgically transplanted to WT mice. Five days later, tumor-bearing mice were randomly divided into four groups and treated with solvent (control, n = 5), anti-CD8 MAb (clone 2.43, 200 μg/mouse, n = 5), verticillin A (0.5 mg/kg body weight, n = 5) and anti-PD-L1 MAb (200 μg/mouse, n = 5), and verticillin A plus anti-PD-L1 plus anti-CD8 MAbs (n = 5) once every two days for 10 days. Tumor volume and tumor weight were analyzed by two-sided t tests and are presented in the right panels. D) UN-KC-6141 cells were surgically transplanted to WT mice. Five days later, tumor-bearing mice were treated as in (C). Tumor volume and tumor weight were analyzed by two-side t tests and are presented in the right panels. E and F) Tumor tissues from both PANC02-H7 (E) and UN-KC-6141 (F) tumor-bearing mice as shown in Figure 7, A and B, were collected after the indicated treatments and analyzed for CD8 and IFNγ mRNA levels by quantitative polymerase chain reaction. The differences between control and treatment group were analyzed by two-sided t tests. Column: mean; bar: SD. WT = wild-type.
See this image and copyright information in PMC

V体育官网 - Comment in

References

    1. Sun L, St Leger AJ, Yu CR, et al. Interferon regulator factor 8 (IRF8) limits ocular pathology during HSV-1 infection by restraining the activation and expansion of CD8+ T Cells. PLoS One. 2016;11(5):e0155420. - PMC - PubMed
    1. Munir S, Andersen GH, Met O, et al. HLA-restricted CTL that are specific for the immune checkpoint ligand PD-L1 occur with high frequency in cancer patients. Cancer Res. 2013;73(6):1764–1776. - PubMed
    1. Hayes JB, Sircy LM, Heusinkveld LE, et al. Modulation of macrophage inflammatory nuclear factor kappaB (NF-kappaB) signaling by intracellular cryptococcus neoformans. J Biol Chem. 2016;291(30):15614–15627. - PMC - PubMed
    1. Hirano F, Kaneko K, Tamura H, et al. Blockade of B7-H1 and PD-1 by monoclonal antibodies potentiates cancer therapeutic immunity. Cancer Res. 2005;65(3):1089–1096. - VSports在线直播 - PubMed
    1. Peng J, Hamanishi J, Matsumura N, et al. Chemotherapy induces programmed cell death-ligand 1 overexpression via the nuclear factor-kappaB to foster an immunosuppressive tumor microenvironment in ovarian cancer. Cancer Res. 2015;75(23):5034–5045. - "V体育安卓版" PubMed

MeSH terms