Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Oct 15;24(20):5165-5177.
doi: 10.1158/1078-0432.CCR-18-0279. Epub 2018 Jun 25.

Loss of E-cadherin Enhances IGF1-IGF1R Pathway Activation and Sensitizes Breast Cancers to Anti-IGF1R/InsR Inhibitors

Free PMC article

Loss of E-cadherin Enhances IGF1-IGF1R Pathway Activation and Sensitizes Breast Cancers to Anti-IGF1R/InsR Inhibitors

Alison M Nagle et al. Clin Cancer Res. .
Free PMC article


Purpose: Insulin-like growth factor 1 (IGF1) signaling regulates breast cancer initiation and progression and associated cancer phenotypes. We previously identified E-cadherin (CDH1) as a repressor of IGF1 signaling and in this study examined how loss of E-cadherin affects IGF1R signaling and response to anti-IGF1R/insulin receptor (InsR) therapies in breast cancer.Experimental Design: Breast cancer cell lines were used to assess how altered E-cadherin levels regulate IGF1R signaling and response to two anti-IGF1R/InsR therapies. In situ proximity ligation assay (PLA) was used to define interaction between IGF1R and E-cadherin. TCGA RNA-seq and RPPA data were used to compare IGF1R/InsR activation in estrogen receptor-positive (ER+) invasive lobular carcinoma (ILC) and invasive ductal carcinoma (IDC) tumors. ER+ ILC cell lines and xenograft tumor explant cultures were used to evaluate efficacy to IGF1R pathway inhibition in combination with endocrine therapy.Results: Diminished functional E-cadherin increased both activation of IGF1R signaling and efficacy to anti-IGF1R/InsR therapies. PLA demonstrated a direct endogenous interaction between IGF1R and E-cadherin at points of cell-cell contact. Increased expression of IGF1 ligand and levels of IGF1R/InsR phosphorylation were observed in E-cadherin-deficient ER+ ILC compared with IDC tumors. IGF1R pathway inhibitors were effective in inhibiting growth in ER+ ILC cell lines and synergized with endocrine therapy and similarly IGF1R/InsR inhibition reduced proliferation in ILC tumor explant culture.Conclusions: We provide evidence that loss of E-cadherin hyperactivates the IGF1R pathway and increases sensitivity to IGF1R/InsR targeted therapy, thus identifying the IGF1R pathway as a potential novel target in E-cadherin-deficient breast cancers. Clin Cancer Res; 24(20); 5165-77. ©2018 AACR.

Conflict of interest statement

The authors declare no potential conflicts of interest.


Figure 1:
Figure 1:. Loss or inhibition of E-cadherin (CDH1) expression enhances IGF1R/InsR signaling.
(A) MCF-7, (B) ZR75.1, and (C) T47D breast cancer cells transfected with SCR (siSCR) or CDH1 (siCDH1) siRNA were stimulated with increasing doses of IGF1 (0–100nM) for 10 min. IGF1R/InsR and Akt signaling was assessed by immunoblot. (D) MCF-7 cells were treated with 25ug/ml HECD-1 antibody for 24 hours and imaged by phase-contrast microscopy for dissociation of adherens junctions. Cells were stimulated with Vhc or 10nM IGF1 for 10 min and IGF1R and Akt signaling assessed by immunoblot. (E) MCF-7 cells were plated at sub-confluency (200k cells in 6-well) or high confluency (800k cells) and then stimulated with either Vhc or 10nM IGF1 for 10 min. IGF1R/InsR signaling was assessed by immunoblot. Representative phase-contrast microscopy images of the cell plating densities are shown. (F) MCF-7 and ZR75.1 siSCR and siCDH1 cells were serum-starved and stimulated with 10nM IGF1 for 17 hours and DNA stained with propidium iodide to measure cell cycle profile. The percent of cells in the IGF1/Vhc conditions in the S- and G2/M phases of the cell cycle for siSCR and siCDH1 are shown (representative experiment shown; n=2 or 3 each with 3 biological replicates).
Figure 2:
Figure 2:. Knockdown of E-cadherin increases sensitivity to IGF1R/InsR inhibition in breast cancer cells.
MCF-7 cells were reverse transfected with SCR or CDH1 siRNA and seeded into 96-well 2D or ULA plates and treated with IGF1R inhibitor (OSI-906 or BMS-754807) for 6 days. Conditions in the panels as follows: (A) OSI-906; 2D, (B) OSI-906; ULA, (C) BMS-754807; 2D, (D) BMS-754807; ULA. The CellTiter Glo assay was used to assess cell viability (relative luminescence). EC50 values for viability were calculated by non-linear regression and statistical differences evaluated using sum-of-squares Global f-test (p<0.05; representative experiment shown; n=3 each with 6 biological replicates).
Figure 3:
Figure 3:. Proximity ligation assay reveals interaction between IGF1R and E-cadherin and recruitment of IG1R to adherens junctions.
In situ proximity ligation assay (PLA) was used to analyze the direct interaction between IGF1R and E-cadherin in breast cancer cells. (A) MCF-7 and (B) T47D cells were plated on coverslips, fixed, and stained with IGF1R and E-cadherin antibody overnight. The Duolink (Sigma) protocol was followed and coverslips were imaged using confocal microscopy to reveal red puncta. (C) MCF-7 siCDH1 and (D) siIGF1R cells were used as negative controls for the assay to assess primary antibody specificity. MCF-7 cells were plated on coverslips and treated with either (E) Vhc or 10nM IGF1 for (F) 30 minutes, (G) 6 hours, or (H) 24 hours. PLA protocol for IGF1R and E-cadherin was followed as described above. (I) Red puncta and nuclei (stained with DAPI) were quantified and displayed as a ratio of puncta/nuclei. All puncta and nuclei in 60x images were counted. One-way ANOVA was used to determine significant difference between groups (p<0.05; one independent experiment, n=5 images per slide counted). The co-localization of IGF1R (green) and E-cadherin (red) was analyzed by immunofluorescence staining in (J) MCF-7 siSCR and (K) siCDH1 knockdown cells.
Figure 4:
Figure 4:. IGF1-IGF1R pathway is active in invasive lobular breast carcinoma with genetic loss of CDH1.
(A) SUM44PE, (B) MDA-MB-134, and (C) BCK4 ILC cells were immunostained for IGF1R (green) and E-cadherin (red) and imaged by confocal microscopy. Of note, BCK4 cells were imaged at an increased exposure compared to MM134 and SUM44PE cells. (D) CDH1 mRNA, (E) IGF1 mRNA, (F) IGF2 mRNA, and (G) pIGF1R/InsR levels in ER+ IDC compared to ER+ ILC in TCGA were plotted using RNAseq (log2 TPM+1) and RPPA (median normalized) data. The TCGA cohort includes n=417 IDC cases and n=137 ILC cases that have matched data for RNAseq and RPPA. Man-Whitney test was used to determine significant differences in expression level between the two subtypes, p<0.05. Correlation between pIGF1R/InsR and (H) IGF1 and (I) IGF2 ligand expression is plotted for IDC (left) and ILC (right).
Figure 5:
Figure 5:. IGF1R pathway inhibitors and endocrine therapy synergize to inhibit cell viability in ILC breast cancer cells.
SUM44PE ILC cells were plated into 96-well ULA plates and treated for 6 days with increasing doses of (A, B) OSI-906, (C, D) BMS-754807, or (E, F) BEZ235 in combination with increasing doses of ICI 182,780. The dose response curves and heat maps shown indicate inhibition of cell viability (CellTiter Glo). Representative experiment shown; n=3 independent experiments each with 2 biological replicates per combination of doses.
Figure 6:
Figure 6:. IGF1R/InsR inhibition reduces Ki67 staining in ILC tumor ex vivo culture.
MM134 and BCK4 xenograft tumors were harvested from immunocompromised mice, minced into 1–2mm3 tumor chunks and then plated on gelatin sponges in 12-well plate containing 1.5ml media. Media was treated with DMSO Vhc or 1uM BMS-754807 for 72 hours. Tumor pieces were harvested by FFPE and stained for Ki67 as a marker of proliferation (A-B, MM134; D-E, BCK4). Staining was quantified by counting all clearly defined nuclei in 20x images (C and F). Statistical difference was assessed using a Student’s t-test (p<0.05; n=3–6).

Similar articles

See all similar articles

Cited by 5 articles

Publication types

MeSH terms