Targeting eIF4A triggers an interferon response to synergize with chemotherapy and suppress triple-negative breast cancer

Abstract

Protein synthesis is frequently dysregulated in cancer and selective inhibition of mRNA translation represents an attractive cancer therapy. Here, we show that therapeutically targeting the RNA helicase eIF4A with zotatifin, the first-in-class eIF4A inhibitor, exerts pleiotropic effects on both tumor cells and the tumor immune microenvironment in a diverse cohort of syngeneic triple-negative breast cancer (TNBC) mouse models. Zotatifin not only suppresses tumor cell proliferation but also directly repolarizes macrophages toward an M1-like phenotype and inhibits neutrophil infiltration, which sensitizes tumors to immune checkpoint blockade. Mechanistic studies revealed that zotatifin reprograms the tumor translational landscape, inhibits the translation of Sox4 and Fgfr1, and induces an interferon (IFN) response uniformly across models. The induction of an IFN response is partially due to the inhibition of Sox4 translation by zotatifin. A similar induction of IFN-stimulated genes was observed in breast cancer patient biopsies following zotatifin treatment. Surprisingly, zotatifin significantly synergizes with carboplatin to trigger DNA damage and an even heightened IFN response, resulting in T cell-dependent tumor suppression. These studies identified a vulnerability of eIF4A in TNBC, potential pharmacodynamic biomarkers for zotatifin, and provide a rationale for new combination regimens consisting of zotatifin and chemotherapy or immunotherapy as treatments for TNBC.

Keywords: Breast cancer; Immunotherapy; Oncology; Translation.

Figure 1. Zotatifin monotherapy inhibits tumor growth in a cohort of Trp53-null preclinical models.

(A) Scheme depicting the generation of syngeneic Trp53-null preclinical models. Donor mammary epithelium from Trp53-null BALB/c mice was transplanted in situ into cleared mammary fat pads of wild-type recipient hosts, resulting in the derivation of heterogeneous, Trp53-null, transplantable mammary tumors representative of the different intrinsic molecular subtypes of human breast cancer. (B) Top: Representative images of H&E staining and IHC staining of F4/80 and S100A8 in Trp53-null models used in this study. Scale bars: 50 μm. Bottom: Quantification of IHC staining. Three to 6 representative ×20 images were analyzed for each tumor. (C) Schematic of animal treatment. Freshly dissociated tumor cells were injected into the fourth mammary fat pad of BALB/c mice. Mice were randomized and treatment was initiated when tumors reach a volume of 70–150 mm³. Mice were treated with either vehicle or zotatifin every 3 days until ethical endpoint. (D) Tumor growth curves of BALB/c mice treated with either vehicle or zotatifin. n = 6 for 2225L-LM2 and n = 5 for all other models in each treatment arm. Data are presented as mean ± SEM and were analyzed using 2-way ANOVA with Bonferroni’s multiple-comparison test. (E) Left: Representative images of IHC staining of BrdU in ethical endpoint 2153L tumor tissues. Scale bar: 50 μm. Right: Quantification of IHC staining. Five representative ×20 images were analyzed for each tumor. n = 5 biological replicates. Data are presented as mean ± SEM and were analyzed using 2-tailed, unpaired Student’s t test. (F) Left: Representative images of cell cycle distribution of 2153L cells that were cultured in complete medium and treated with vehicle or 40 nM zotatifin for 48 hours. Right: Quantification of the cell cycle phases from 3 independent experiments. Data are presented as mean ± SD and were analyzed using 2-tailed, unpaired Student’s t test.

Figure 2. Zotatifin monotherapy alters the tumor immune microenvironment and sensitizes tumors to immune checkpoint blockade.

(A) t-Distributed stochastic neighbor embedding (t-SNE) projection of tumor-infiltrating immune cells in 2153L tumors that were treated for 7 days and analyzed using mass cytometry. Data from 3 biological replicates of each group were concatenated before t-SNE and FlowSOM analysis. Equal cell numbers are shown for each group. (B) Quantification of major immune cell populations in mass cytometry analysis. n = 3 per group. (C) Left: Representative images of IHC staining of S100A8 in 2153L tumors treated with vehicle or zotatifin. Scale bars: 50 μm. Right: Quantification of staining. Multiple ×20 images were analyzed for each tumor. n = 5 biological replicates per group. (D) Luminex array detection of Cxcl5 levels in 2153L tumor lysates. n = 4 biological replicates per group. (E and F) Flow cytometry analysis of tumor-infiltrating neutrophils (E) and CD206 median fluorescence intensity (MFI) in tumor-infiltrating macrophages (F) in ethical endpoint tumors. n ≥ 4 per group in all models. In B–F, data are presented as mean ± SEM and were analyzed using 2-tailed, unpaired Student’s t test. (G) Flow cytometry analysis of iNOS expression in BMDMs that were treated with vehicle or zotatifin in the presence of LPS and IFN-γ. (H) Flow cytometry analysis of Arg1 expression in BMDMs that were treated with vehicle or zotatifin in the presence of IL-4 and IL-13. (I) Immunoblotting of TAMs that were separated from untreated tumors and treated with vehicle or zotatifin for 24 hours in vitro. See complete unedited blots in the supplemental material. (J) Growth curves of 2153L tumors treated with indicated drugs. n = 5 per group. Data are presented as mean ± SEM and were analyzed using 2-way ANOVA with Tukey’s multiple-comparison test. (K) Kaplan-Meier survival curves of 2153L tumor–bearing mice from treatment start time. The log-rank test was used to test for the significant differences of curves. n = 5 per group.

Figure 3. Zotatifin remodels the proteomic landscape of 2153L tumors.

(A) Scheme of sample collection strategy for mass spectrometry. Freshly dissociated 2153L tumor cells were transplanted into the fourth mammary fat pad of BALB/c mice and were allowed to grow until palpable. Mice then were randomized and treated with either vehicle or zotatifin for 2 doses spanning 3 days. Tumor tissues were collected 3 hours after the second injection. (B) Volcano plot showing relative fold change (log₂) in protein abundance versus −log₁₀(P values) from 2153L tumors treated with zotatifin compared with vehicle. Proteins that demonstrate a significant change in expression (FDR q < 0.05) are colored, with decreased expression on the left in blue and increased expression on the right in red. Genes that are critically involved in cell proliferation, stem cell signaling, and IFN response pathways are labeled. n = 4 biological replicates per arm. Statistical significance was determined by 2-tailed, unpaired moderated t test. (C) GSEA of mass spectrometry results was performed with the MSigDB hallmarks data set and is summarized as the normalized enrichment score (NES) in vehicle- or zotatifin-treated 2153L tumor tissues. Top pathways that have a family-wise error rate (FWER) P < 0.25 are displayed. FWER P values for each pathway are denoted by color. (D) GSEA enrichment plots for Hallmark E2F targets and G₂M checkpoint signatures that are downregulated in zotatifin-treated tumors compared with vehicle. (E) GSEA enrichment plots for Hallmark IFN-α response and IFN-γ response signatures that are upregulated in zotatifin-treated tumors compared with vehicle.

Figure 4. Zotatifin inhibits the translation of Sox4 and Fgfr1 mRNAs.

(A) Immunoblotting analysis of tumors that were treated with vehicle or zotatifin in vivo. n = 5 biological replicates per group. (B and C) qPCR analysis for Sox4 (B) and Fgfr1 (C) mRNA expression in tumors that were treated with vehicle or zotatifin in vivo. Data are presented as mean ± SEM and were analyzed using 2-tailed, unpaired Student’s t test. n = 5 biological replicates per group. (D) Immunoblotting analysis of 2153L cells that were treated with different concentrations of zotatifin for 6 hours in vitro. (E) Immunoblotting analysis of 2153L cells that were treated with 40 nM zotatifin for different time periods. (F) Immunoblotting analysis of BT549 cells that were treated with 40 nM zotatifin for different time periods. *Denotes a nonspecific band. In D–F, data are representative of 3 independent experiments. (G) Immunoblotting analysis of HAP1 cells that were treated with 40 nM zotatifin in vitro. Data are representative of 2 independent experiments. (H) Illustration for polysome profiling analysis. (I and J) Polysome profiling of 2153L cells that were treated with vehicle or 40 nM zotatifin for 2 hours. (I) Representative polysome profiles from 3 biological replicates. (J) Distribution of Sox4 and Fgfr1 mRNAs across the different fractions. Data are presented as mean ± SEM of 3 biological replicates. See complete unedited blots in the supplemental material.

Figure 5. Inhibition of Sox4 translation contributes to zotatifin-induced IFN response.

(A) qPCR analysis of tumors that were treated with vehicle or zotatifin in vivo. The mean mRNA levels of the vehicle groups were set as 1 and fold changes were calculated for each gene. n = 5 biological replicates per group. (B) qPCR analysis of 8 paired biopsies from pretreatment (black) and on-zotatifin-treatment (red) ER⁺ breast cancer patients. The mRNA levels of pretreatment samples were set as 1 and fold changes were calculated for each paired sample. (C) qPCR analysis of HAP1 cells that were treated with 40 nM zotatifin for 6 hours. Data are representative of 2 independent experiments and are presented as mean ± SD of technical triplicates. (D) qPCR analysis of 2153L cells that were transfected with negative control siRNA with or without zotatifin treatment, or Sox4 siRNAs without zotatifin treatment for 48 hours. (E) qPCR analysis of zotatifin-induced gene fold changes in 2153L cells that were transfected with negative control siRNA or Sox4 siRNAs in the presence of vehicle or zotatifin. In D and E, representative data from 3 biological replicates are shown, and data are presented as mean ± SD of technical duplicates.

Figure 6. Zotatifin synergizes with carboplatin to suppress tumor progression.

(A) Kaplan-Meier survival curves of 2153L tumor–bearing mice from treatment start time. (B) Survival regression analysis of 2153L tumor–bearing mice. Survival data were fitted using a parametric survival regression model with a log-normal distribution. The top table reports all possible pairwise comparisons using linear contrasts that are adjusted for multiple comparisons using Holm’s method. The bottom table tests for overall main effects and interactions. In A and B, data from 3 to 5 independent experimental batches are integrated. n = 24 for vehicle, n = 22 for zotatifin, n = 13 for carboplatin, and n = 24 for zotatifin + carboplatin. (C and D) Kaplan-Meier survival curves of 2225L-LM2 (C) or 2208L (D) tumor–bearing mice from treatment start time. Mice were randomized and treatment was initiated when 2225L-LM2 tumors reached approximately 130 mm³ volume or 2208L tumors reached approximately 210 mm³ volume. n = 5 biological replicates per group. (E) Left: Representative images of IF staining of γH2A.X (in red) in 2153L tumors that were treated with indicated drugs for 3 days. Scale bars: 50 μm. Right, quantification of IF staining. At least 2 representative views were analyzed for each tumor and at least 3 tumors for each treatment group were analyzed. Data are presented as mean ± SEM and were analyzed using 2-tailed, unpaired Student’s t test. (F) Kaplan-Meier survival curves of 2153L tumor–bearing mice treated with indicated drugs. n ≥ 4 biological replicates per group. (G) Kaplan-Meier survival curves of 2225L-LM2 tumor–bearing mice treated with indicated drugs. n ≥ 3 biological replicates per group. (H) Kaplan-Meier survival curves of 2208L tumor–bearing mice treated with indicated drugs. n = 4 biological replicates per group. In C, D, and F–H, the log-rank test (2-tailed) was used to test for the significant differences of curves between groups.

Figure 7. Zotatifin synergizes with carboplatin to induce an IFN response and promote T cell–dependent tumor inhibition.

(A) Scheme of sample collection strategy for mass spectrometry. Tumor-bearing mice were randomized and treated with indicated therapies for 3 days. Tumor tissues were collected 3 hours after the second injection of zotatifin. n = 4 per group. (B and C) GSEA enrichment plot for Hallmark IFN-α response signature that is upregulated in combination therapy treated tumors compared with zotatifin monotherapy (B) or carboplatin monotherapy (C). (D) Heatmap for Hallmark IFN-α response signature proteins from tumor tissues treated with indicated therapies. (E–G) Mass cytometry analysis of tumor-infiltrating immune cells of 2153L tumors that were treated with indicated therapies for 7 days. (E) The uniform manifold approximation and projection (UMAP) plot overlaid with color-coded clusters. Data from 5 biological replicates of each group were concatenated before UMAP and FlowSOM clustering analysis. Equal numbers of events are shown for each group and major cell types are marked. (F) UMAP plot overlaid with the expression of Bst2 or granzyme B. (G) Quantification of major lymphoid populations from 2153L tumors in mass cytometry analysis. n = 5 biological replicates per group. (H) Outline of treatment design. Freshly dissociated 2153L tumor cells were transplanted into nude mice or BALB/c mice in parallel. Treatment was initiated when tumors reached 100 mm³. (I) Growth curves of 2153L tumors in nude mice treated with indicated drugs. n = 5 biological replicates per group. (J) Growth curves of 2153L tumors in BALB/c mice treated with indicated drugs. n = 3 biological replicates for monotherapy groups and n = 10 biological replicates for the combination treatment group. In I and J, data are presented as mean ± SEM and were analyzed using 2-way ANOVA with Bonferroni’s multiple-comparison test.

Figure 8. Working model of zotatifin.

Inhibition of eIF4A by zotatifin suppresses the translation of Sox4 and Fgfr1, induces an IFN response, shifts the tumor immune landscape, and ultimately enhances the response to immune checkpoint blockade or chemotherapy.