The BET Protein Inhibitor JQ1 Decreases Hypoxia and Improves the Therapeutic Benefit of Anti-PD-1 in a High-Risk Neuroblastoma Mouse Model

Abstract

Anti-programmed death 1 (PD-1) is a revolutionary treatment for many cancers. The response to anti-PD-1 relies on several properties of tumor and immune cells, including the expression of PD-L1 and PD-1. Despite the impressive clinical benefit achieved with anti-PD-1 in several cancers in adults, the use of this therapy for high-risk neuroblastoma remains modest. Here, we evaluated the therapeutic benefit of anti-PD-1 in combination with JQ1 in a highly relevant TH-MYCN neuroblastoma transgenic mouse model. JQ1 is a small molecule inhibitor of the extra-terminal domain (BET) family of bromodomain proteins, competitively binding to bromodomains. Using several neuroblastoma cell lines in vitro, we showed that JQ1 inhibited hypoxia-dependent induction of HIF-1α and decreased the expression of the well-known HIF-1α downstream target gene CA9. Using MRI relaxometry performed on TH-MYCN tumor-bearing mice, we showed that JQ1 decreases R2* in tumors, a parameter associated with intra-tumor hypoxia in pre-clinical settings. Decreasing hypoxia by JQ1 was associated with improved blood vessel quality and integrity, as revealed by CD31 and αSMA staining on tumor sections. By analyzing the immune landscape of TH-MYCN tumors in mice, we found that JQ1 had no major impact on infiltrating immune cells into the tumor microenvironment but significantly increased the percentage of CD8⁺ PD-1⁺, conventional CD4⁺ PD-1⁺, and Treg PD-1⁺ cells. While anti-PD-1 monotherapy did not affect TH-MYCN tumor growth, we showed that combinatorial therapy associating JQ1 significantly decreased the tumor volume and improved the therapeutic benefit of anti-PD-1. This study provided the pre-clinical proof of concept needed to establish a new combination immunotherapy approach that may create tremendous enthusiasm for treating high-risk childhood neuroblastoma.

Keywords: JQ1; PD-1/PD-L1; combinatorial therapy; epigenetic drugs; hypoxia; immune checkpoint inhibitors; immunotherapy; neuroblastoma; tumor vasculature.

Figure 1

Effect of JQ1 on the hypoxia status of NB cells in vitro. (A) Western blot test showing the protein expression of HIF-1α and CA9 in NHO2A, CHP-134, and SIMA cells cultured under normoxia (N) or hypoxia and treated with JQ1 at the indicated concentration. Actin was used as a loading control. The quantification of band intensity corresponding to HIF-1α and CA9 in treated hypoxic cells is reported compared to untreated hypoxic cells and normoxic cells, respectively. (B) RT-qPCR measurement of CA9 mRNA in cells described in (A). Bars represent means from three independent experiments ± SEM. Statistically significant difference was calculated by unpaired two-tailed Student’s t-test (* p < 0.05, ** p <0.01; and *** p < 0.001).

Figure 2

Effect of JQ1 on the hypoxia status and blood vessels in TH-MYCN tumor-bearing mice. (A) Experimental design of JQ1 treatment of TH-MYCN tumor-bearing mice showing the schedule and dosing. Following the development of homozygote TH-MYCN tumors, the tumor volume and R2* values were assessed by an MRI in all mice. Mice were randomly assigned to a vehicle-treated group (n = 6) or JQ1-treated group (n = 9). Mice were treated twice with either the vehicle or JQ1 (50 mg/kg) by IP on days 1 and 2. On day 3 post-treatment, an MRI was performed to determine the tumor volume and R2* values. Tumors were harvested for subsequent experiments. (B) The volume of TH-MYCN tumors treated with either the vehicle (Veh) or JQ1 as described in (A) on day 3. Each dot represents one tumor. Results are shown as mean ± SEM (error bars). Statistically significant differences were calculated compared to the Veh-treated tumors using an unpaired two-tailed Student’s t-test (ns: not significant). R2* values in TH-MYCN tumors before and after treatment on day 3 with the vehicle (C) and JQ1 (D) according to the experimental design in (A). The R2* values in TH-MYCN tumors after the treatment with either the vehicle (Veh) or JQ1 are reported in (E). The R2* ratio of after to before treatment with the vehicle (Veh) or JQ1 is reported in (F). Each dot represents one tumor. Results are shown as mean ± SEM (error bars). Statistically significant differences are calculated using a Mann–Whitney test for (C) and (D) or an unpaired two-tailed Student’s t-test for (E) and (F) (* p < 0.05, ** p < 0.01; and *** p < 0.001; ns: non-significant). (G) Western-blot showing the protein expression of CAIX in three different (1, 2 and 3) TH-MYCN tumors treated with either the vehicle or JQ1 according to the experimental schedule described in (A). Actin was used as a loading control. (H) Staining of vehicle- or JQ1-treated TH-MYCN tumors described in (A) with H&E (upper panels), CD31 (middle panels), or aSMA lower panels. Enlarged images of the zones delineated with black boxes (in CD31 and aSMA stained tumors) are shown (Scale bars 100 or 50 µm).

Figure 2

Effect of JQ1 on the hypoxia status and blood vessels in TH-MYCN tumor-bearing mice. (A) Experimental design of JQ1 treatment of TH-MYCN tumor-bearing mice showing the schedule and dosing. Following the development of homozygote TH-MYCN tumors, the tumor volume and R2* values were assessed by an MRI in all mice. Mice were randomly assigned to a vehicle-treated group (n = 6) or JQ1-treated group (n = 9). Mice were treated twice with either the vehicle or JQ1 (50 mg/kg) by IP on days 1 and 2. On day 3 post-treatment, an MRI was performed to determine the tumor volume and R2* values. Tumors were harvested for subsequent experiments. (B) The volume of TH-MYCN tumors treated with either the vehicle (Veh) or JQ1 as described in (A) on day 3. Each dot represents one tumor. Results are shown as mean ± SEM (error bars). Statistically significant differences were calculated compared to the Veh-treated tumors using an unpaired two-tailed Student’s t-test (ns: not significant). R2* values in TH-MYCN tumors before and after treatment on day 3 with the vehicle (C) and JQ1 (D) according to the experimental design in (A). The R2* values in TH-MYCN tumors after the treatment with either the vehicle (Veh) or JQ1 are reported in (E). The R2* ratio of after to before treatment with the vehicle (Veh) or JQ1 is reported in (F). Each dot represents one tumor. Results are shown as mean ± SEM (error bars). Statistically significant differences are calculated using a Mann–Whitney test for (C) and (D) or an unpaired two-tailed Student’s t-test for (E) and (F) (* p < 0.05, ** p < 0.01; and *** p < 0.001; ns: non-significant). (G) Western-blot showing the protein expression of CAIX in three different (1, 2 and 3) TH-MYCN tumors treated with either the vehicle or JQ1 according to the experimental schedule described in (A). Actin was used as a loading control. (H) Staining of vehicle- or JQ1-treated TH-MYCN tumors described in (A) with H&E (upper panels), CD31 (middle panels), or aSMA lower panels. Enlarged images of the zones delineated with black boxes (in CD31 and aSMA stained tumors) are shown (Scale bars 100 or 50 µm).

Figure 3

Immune phenotyping of JQ1-treated TH-MYCN tumors. (A) Experimental design of JQ1 treatment of TH-MYCN tumor-bearing mice showing the schedule and dosing. Following the development of homozygote TH-MYCN tumors, the tumor volume was assessed by MRI in all mice. Mice were randomly assigned to a vehicle-treated group (n = 6) or a JQ1-treated group (n = 5). Mice were treated three times with vehicle or JQ1 (25 mg/kg) by IP on days 1, 2, and 3. On day 4 post-treatment, MRI was performed to determine the tumor volume, and tumors were harvested for immune phenotyping experiments. (B–D) Flow cytometry quantification of total (B), CD69⁺ (C), or PD-1⁺ (D) CD8⁺ T cells, conventional (Conv.) CD4⁺ T cells and Tregs infiltrating vehicle-treated or JQ-1-treated TH-MYCN tumors at day 4. The defined subpopulations were gated and quantified in live CD45⁺ cells. Each dot represents one tumor. The data are reported as the average of six or five mice per group. Results are shown as mean ± SEM (error bars). Statistically significant differences (indicated by asterisks) are calculated compared to vehicle-treated tumors using an unpaired two-tailed Student’s t-test (ns: not significant *: p < 0.05).

Figure 3

Immune phenotyping of JQ1-treated TH-MYCN tumors. (A) Experimental design of JQ1 treatment of TH-MYCN tumor-bearing mice showing the schedule and dosing. Following the development of homozygote TH-MYCN tumors, the tumor volume was assessed by MRI in all mice. Mice were randomly assigned to a vehicle-treated group (n = 6) or a JQ1-treated group (n = 5). Mice were treated three times with vehicle or JQ1 (25 mg/kg) by IP on days 1, 2, and 3. On day 4 post-treatment, MRI was performed to determine the tumor volume, and tumors were harvested for immune phenotyping experiments. (B–D) Flow cytometry quantification of total (B), CD69⁺ (C), or PD-1⁺ (D) CD8⁺ T cells, conventional (Conv.) CD4⁺ T cells and Tregs infiltrating vehicle-treated or JQ-1-treated TH-MYCN tumors at day 4. The defined subpopulations were gated and quantified in live CD45⁺ cells. Each dot represents one tumor. The data are reported as the average of six or five mice per group. Results are shown as mean ± SEM (error bars). Statistically significant differences (indicated by asterisks) are calculated compared to vehicle-treated tumors using an unpaired two-tailed Student’s t-test (ns: not significant *: p < 0.05).

Figure 4

Impact of combining JQ1 and PD-1 on TH-MYCN tumor volume and mice survival. (A) Experimental design of JQ1 and anti-PD-1 treatment of TH-MYCN tumor-bearing mice showing the schedule and dosing. Following the development of homozygote TH-MYCN tumors, the tumor volume was assessed by MRI in all mice. Mice were randomly assigned into four groups: vehicle- and isotope-treated group (n = 9), vehicle and anti-PD-1-treated group (n = 8), JQ1 and isotype-treated group (n = 8), and JQ1 and anti-PD-1-treated group (n = 8). Mice were treated daily from days 1 to 7 with either vehicle (group 1 and 2) or JQ1 (group 3 and 4) and on days 1, 4, 7, 10 and 13 with either isotype (group 1 and 3) or anti-PD-1 (group 2 and 4). After day 15, only the treatment with isotype or anti-PD-1 was continued using the same schedule until the end of experiment (mice euthanasia). JQ1 (25 mg/kg/day) and anti-PD-1 (10 mg/kg/day) were administered by IP. MRI was performed on days 1, 4, 8 and 15 post-treatment and on a regular basis to determine the tumor volume. (B) Representative images of TH-MYCN tumor-bearing mice on days 1, 4, 8 and 15 for the groups described in (A). On day 15, no images for groups 1 and 2 are provided as animals died. Abdominal tumor masses are delineated in red. (C) Volumes of TH-MYCN tumors on days 1, 4 and 8, and in mice treated with JQ1 and isotype or JQ1 and anti-PD-1. Each dot represents one tumor. Results are shown as mean ± SEM (error bars). Statistically significant differences are calculated compared to the control group (JQ1 and isotype) using an unpaired two-tailed Student’s t-test (ns: not significant *: p < 0.05). (D) Mice survival curves were generated from tumor-bearing mice treated with JQ1 and isotype or JQ1 and anti-PD-1. Lack of survival was defined as death or tumor size > 2000 mm³. The probability of survival was defined using GraphPad Prism, and p-values were calculated using the log-rank (Mantel-Cox) test (* p ≤ 0.05).