Cooperative NF-κB and Notch1 signaling promotes macrophage-mediated MenaINV expression in breast cancer

Abstract

Metastasis is a multistep process that leads to the formation of clinically detectable tumor foci at distant organs and frequently to patient demise. Only a subpopulation of breast cancer cells within the primary tumor can disseminate systemically and cause metastasis. To disseminate, cancer cells must express MenaINV, an isoform of the actin regulatory protein Mena, encoded by the ENAH gene, that endows tumor cells with transendothelial migration activity, allowing them to enter and exit the blood circulation. We have previously demonstrated that MenaINV mRNA and protein expression is induced in cancer cells by macrophage contact. In this study, we discovered the precise mechanism by which macrophages induce MenaINV expression in tumor cells. We examined the promoter of the human and mouse ENAH gene and discovered a conserved NF-κB transcription factor binding site. Using live imaging of an NF-κB activity reporter and staining of fixed tissues from mouse and human breast cancer, we further determined that for maximal induction of MenaINV in cancer cells, NF-κB needs to cooperate with the Notch1 signaling pathway. Mechanistically, Notch1 signaling does not directly increase MenaINV expression, but it enhances and sustains NF-κB signaling through retention of p65, an NF-κB transcription factor, in the nucleus of tumor cells, leading to increased MenaINV expression. In mice, these signals are augmented following chemotherapy treatment and abrogated upon macrophage depletion. Targeting Notch1 signaling in vivo decreased NF-κB signaling activation and MenaINV expression in the primary tumor and decreased metastasis. Altogether, these data uncover mechanistic targets for blocking MenaINV induction that should be explored clinically to decrease cancer cell dissemination and improve survival of patients with metastatic disease.

Keywords: Breast cancer; MenaINV; NF-κB; Notch1; TMEM doorways.

Fig. 1

Macrophage-mediated induction of MenaINV expression via Notch and NF-κB cooperation. A MenaINV mRNA expression in MDA-MB-231 (231) cells co-cultured with or without BAC2.1F macrophages (Mac) and with or without 10 ng/ml TNFα for 4 h. B MenaINV mRNA expression in 231 cells co-cultured with or without Macs, NF-κB inhibitor (DHMEQ), or Notch/γ secretase inhibitor (DAPT) for 4 h. C Model of potential Notch1 and NF-κB signaling crosstalk leading to enhanced transcriptional activity at the ENAH (Mena) promoter. The released Notch intracellular domain (NICD—shaded in gray) can bind to the transcription factors in the NF-κB signaling pathway and prevent their nuclear export, allowing for enhanced and sustained transcriptional activation of target genes and alternative splicing. The bars in A and B represent average fold change MenaINV mRNA compared to control (231 cells), ± S.D. The data were analyzed using a one-way ANOVA with Tukey’s multiple comparisons test.*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, n.s. = not significant. All experiments were repeated at least three times

Fig. 2

Notch1 enhances NF-κB signaling by sustaining p65 nuclear localization. A Stills from movies at 0, 17, and 240 min of MDA-MB-231/GFP-p65 cells treatment with vehicle, or 10 ng/ml human TNFα, or 80 µm Jagged1, or 10 ng/ml TNFα and 80 µm Jagged1. In all treatment groups with TNFα, the cells were treated for an initial 10 min, and then TNFα was washed out and replaced with minimal media, or with Jagged1 supplemented media. Cells were imaged live for 240 min using an EPI fluorescence microscope for the duration of the treatment, with one image captured every 2.5 min. Scale bar = 10 μm. B Quantification of normalized GFP-p65 nuclear localization over time from experiment in A. Each timepoint shows an average from 45 cells per treatment, from three independent experiments. C Western blot showing the amount of p65 in the cytoplasmic and nuclear fractions of wild type MDA-MB-231 cells treated for 30 min (upper blots) or 4 h (lower blots) with vehicle, or 10 ng/ml TNFα, or 80 μm Jagged1, or 10 ng/ml TNFα and 80 μm Jagged1 (TNFα + Jagged1). In all treatment groups with TNFα, the cells were treated for an initial 10 min, and then TNFα was washed out and replaced with minimal media, or with Jagged1 supplemented media. The experiment was repeated three times and representative western blots are shown. D Quantification of western blots in C where the nuclear p65 signal was normalized to the lamin A/C signal. The graph shows the fold change in nuclear p65 signal for each treatment relative to the control treatment at both time points. Fold changes were averaged from three independent experiments. E MenaINV mRNA expression in wild type MDA-MB-231 (231) cells treated as in C for 1 or 4 h. The experiments were repeated three times. Bars in E show average fold change MenaINV mRNA expression compared to Control at 1 or 4 h. Data in D and E were analyzed using a one-way ANOVA with Tukey’s multiple comparisons test. *p < 0.05, ****p < 0.0001, n.s. = not significant

Fig. 3

MenaINV expression in tumor cells induced by macrophages depends partially on TNFα-mediated NF-κB signaling and Notch1 Jagged1 signaling. A MenaINV mRNA expression in MDA-MB-231 (231) cells co-cultured with or without BAC2.1F macrophages (Mac) and with or without C87 (TNFα inhibitor) or SAHM1 (MAML1 inhibitor) for 4 h. B MenaINV mRNA expression in 231 cells co-cultured with or without Macs, TNFα inhibitor (C87), or Jag 1 or Jag2 blocking antibodies for 4 h. Bars in A and B represent average fold change of MenaINV mRNA expression compared to control cells (231). C MenaINV mRNA expression in 231 cells co-cultured with sgControl (WT) or Jagged1 knockout BAC1.2F5 macrophages (Jag1 KO Macs). Bars in (A–C) represent average fold change of MenaINV mRNA expression compared to control cells (A and B: 231; C: 231 + WT Macs). Data were analyzed using a one-way ANOVA with Tukey’s multiple comparisons test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, n.s. = not significant. All experiments were repeated at least three times

Fig. 4

Macrophage depletion decreases NF-κB signaling and MenaINV expression in a PDX model in vivo. A Experimental design for macrophage depletion in patient-derived xenograft (PDX) HT17 model in SCID mice. i.p. = intraperitoneal. Red arrows indicate treatment days. B Immunofluorescence co-staining of HT17 tumors xenografted in SCID mice treated as outlined in A for the macrophage marker, Iba1 (white), C p65 (red), MenaINV (green) and nuclei (blue-DAPI). Scale bars = 20 μm. D Quantification of average fold change in p65 expression from mice in A. E Quantification of average fold change in p65 nuclear localization in PDX HT17 tumors from mice treated as outlined in A. Only p65 co-localized with the nuclear DAPI signal was quantified. F Quantification of average fold change MenaINV expression from PDX HT17 tumors treated as outlined in A. Data in D–F were analyzed using a student’s t-test. **p < 0.01, ***p < 0.001. Eight mice were treated per group. Each dot represents the average measurement from an individual mouse

Fig. 5

Inhibition of Notch1 signaling in vivo decreases activation of NF-κB signaling in MDA-MB-231 orthotropic injection model. A Schematic of DAPT treatment of SCID mice bearing orthotopically injected MDA-MB-231 tumor cells. Seven weeks post tumor cell injection mice were treated with 10 mg/kg DAPT or vehicle (corn oil) by i.p. every day for 14 days. Red arrows represent treatment days. B Immunofluorescence staining of primary tumor tissues sections for DAPI (nuclear stain, blue), p65 (red) and MenaINV (green). White dotted circles indicate nuclei in the DAPI and p65 channels. Yellow arrow heads denote nuclei with p65 positive stain (active NF-κB signaling), and white arrowheads indicate nuclei without p65 positive staining (inactive NF-κB signaling). Scale bar = 10 μm. C Quantification of p65 localization (%cytoplasmic/nuclear) in tumor tissue from B. D Quantification of average fold change in MenaINV expression compared to control mice from (B). Data in C and D were analyzed using a student’s t-test. *p < 0.05, **p < 0.01. Four mice were used per treatment. Each dot represents the average measurement for an individual mouse

Fig. 6

Chemotherapy treatment enhances NF-κB activation and MenaINV expression through macrophage recruitment in patient xenograft model. A Experimental design of chemotherapy and clodronate treatments in patient-derived xenograft (PDX) HT17 tumors in SCID mice. i.p. = intraperitoneal, i.v. = intravenous. B Immunofluorescence staining of primary breast tumor tissues from mice treated as outlined in A with DAPI (nuclear stain, blue), and antibodies recognizing p65 (red), and MenaINV (green). Blue and orange outlined sections are expanded to the right and demonstrate examples of what is quantified as primarily cytoplasmic (blue) or nuclear (orange) localization of p65 in HT17 tumor tissue. Scale bars = 10 μm C Quantification of average fold change in p65 nuclear localization in treated primary tumors from A stained for p65 and DAPI. Only p65 which co-localized with the nuclear DAPI signal was quantified. D Quantification of average fold change in MenaINV expression in treated primary tumors from A. E Quantification of the percentage of MenaINV-hi-expressing tumor cells which also co-expressed p65 (regardless of cellular compartment localization), in primary tumor cells from treatments in A. F Quantification of the localization (% cytoplasmic/nuclear) of p65 in MenaINV-hi-expressing tumor cells from primary tumor cells treated in A. G Quantification of average fold change MenaINV expression associated with nuclear p65 staining of primary tumors from A stained for MenaINV. Data in C, D, and G were analyzed using a one-way ANOVA with Tukey’s multiple comparisons test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, n.s. = not significant. Eight mice were treated per group and each dot represents the average measurement for an individual mouse

Fig. 7

MenaINV expression in cancer cells is induced by macrophage-mediated cooperative NF-κB and Notch1 signaling. Juxtacrine and paracrine signaling between macrophages and tumor cells activate Notch1 and NF-κB pathways which cooperate to induce MenaINV expression in cancer cells. A Notch1 signaling alone does not induce MenaINV expression in tumor cells. B NF-κB signaling, activated by TNFα binding to the TNFR1 receptor, causes nuclear translocation of the transcription factor p65 and a 1.5-fold increase in MenaINV expression. C Notch1 and NF-κB signaling crosstalk to increase MenaINV expression further to 2.5-fold. Notch1 intracellular domain (NICD) enhances nuclear retention of NF-κB transcription factor p65 leading to sustained NF-κB signaling and induction of MenaINV expression. This mechanism of MenaINV induction is present in vivo and it explains previously observed increase in MenaINV expression upon in chemotherapy treatment [19]. This detailed understanding of MenaINV induction in clinically relevant scenarios is needed for future development of combination therapies to improve survival of patients with breast cancer. Figure created with BioRender.com