Desmoglein 2 Functions as a Receptor for Fatty Acid Binding Protein 4 in Breast Cancer Epithelial Cells

Abstract

Fatty acid binding protein 4 (FABP4) is a secreted adipokine linked to obesity and progression of a variety of cancers. Obesity increases extracellular FABP4 (eFABP4) levels in animal models and in obese breast cancer patients compared with lean healthy controls. Using MCF-7 and T47D breast cancer epithelial cells, we show herein that eFABP4 stimulates cellular proliferation in a time and concentration dependent manner while the non-fatty acid-binding mutant, R126Q, failed to potentiate growth. When E0771 murine breast cancer cells were injected into mice, FABP4 null animals exhibited delayed tumor growth and enhanced survival compared with injections into control C57Bl/6J animals. eFABP4 treatment of MCF-7 cells resulted in a significant increase in phosphorylation of extracellular signal-regulated kinase 1/2 (pERK), transcriptional activation of nuclear factor E2-related factor 2 (NRF2) and corresponding gene targets ALDH1A1, CYP1A1, HMOX1, SOD1 and decreased oxidative stress, while R126Q treatment did not show any effects. Proximity-labeling employing an APEX2-FABP4 fusion protein revealed several proteins functioning in desmosomes as eFABP4 receptor candidates including desmoglein (DSG), desmocollin, junction plankoglobin, desomoplankin, and cytokeratins. AlphaFold modeling predicted an interaction between eFABP4, and the extracellular cadherin repeats of DSG2 and pull-down and immunoprecipitation assays confirmed complex formation that was potentiated by oleic acid. Silencing of DSG2 in MCF-7 cells attenuated eFABP4 effects on cellular proliferation, pERK levels, and ALDH1A1 expression compared with controls.

Implications: These results suggest desmosomal proteins, and in particular desmoglein 2, may function as receptors of eFABP4 and provide new insight into the development and progression of obesity-associated cancers.

Figure 1.. Proliferative effects of eFABP4 on breast cancer cell culture.

A) Effect of FABP4 or mutant R126Q on MCF-7 proliferation. B) Effect of varying concentrations and time of eFABP4 on MCF-7 proliferation. C) Effect of FABP4 or mutant R126Q on T47D proliferation. D) Growth of E0771 cells in response to varying concentrations of eFABP4. E) Effect of FABP1 compared to FABP4 and R126Q on MCF-7 cell growth. Results are shown for one experiment that was repeated 3 times. *, p < 0.05.

Figure 2.. FABP4 null (AKO) mice injected with E0771 murine breast cancer showed decreased tumor volume and enhanced survival.

A) Tumor volume over time. B) Survival curves of FABP4 −/− null (AKO) mice vs. wild type (WT) mice. N = 20, 10 per group. *, p < 0.05.

Figure 3.. FABP4 mediated breast cancer growth was dependent on the ERK/NRF2 pathway.

A) MCF-7 cells were treated with 100 ng/mL FABP4 or R126Q for 5 minutes and pERK levels were evaluated using western blot. B) Quantification of pERK level in A). Bands of pERK were normalized to the level of tERK in the same samples. C) MCF-7 cells were treated with 100 ng/mL FABP4 or 100 ng/mL R126Q for 48 hours and the NRF2 activity was evaluated using the antioxidant response element (ARE) luciferase reporter gene system. D-G) mRNA expression of ALDH1A1, CYP1A1, HMOX1, and SOD1 were evaluated using qRT-PCR (normalized to TBP) following treatment of MCF-7 cells with 100 ng/mL FABP4 or 100 ng/mL R126Q for 48 hours. H) Levels of H₂O₂ were measured following treatment with 100 ng/mL FABP4 for 48 hours in MCF-7 cells. I) MCF-7 proliferation after chemical inhibition of NRF2 with 20 nM brusatol. Results are shown for one experiment that was repeated 3 times. *, p < 0.05. ns, nonsignificant.

Figure 4.. Recombinant protein APEX2-FABP4 was highly functional.

A) Recombinant proteins APEX2-FABP4, FABP4 and APEX2 expressed in E. coli were analyzed by SDS-PAGE stained with Coomassie blue. B) Fatty acid binding affinity of recombinant proteins was assessed with ANS binding assay. (C) Peroxidase activity of recombinant proteins were detected with luminol-based enhanced-chemiluminescence of HRP substrate.

Figure 5.. Top FABP4 receptor candidates identified by proximal labeling.

MCF-7 cells were treated with biotin-SS-phenol and APEX2-FABP4 or APEX2 followed by H₂O₂ to initiate biotin labeling. The biotinylated proteins were enriched using streptavidin magnetic and analyzed by mass spectrometry. The top FABP4 receptor candidates reported in mass spectrometry was normalized to APEX2. Bars in grey mean the output was “0” in APEX2 but put in “1” instead for normalization calculation. Asterisks showing the candidates are components of desmosome complex.

Figure 6.. FABP4 interacted with ECs of DSG2.

A) AlphaFold predicted the interaction between FABP4 and each EC (1 through 4) of DSG2 individually. Top panel shows ribbon cartoons of the complex colored by AlphaFold’s pLDDT parameter. Each residue in the sequence is color-coded based on the model confidence score. Model confidence from high to low: magenta (100), white (50), cyan (0). B) PAE distance plots show the distance error map. Blue is 0 angstroms error, yellow is 30 angstroms error. C) Docking of F57 (red) of FABP4 (grey) and DSG2 EC3–4 (magenta). The native structures of FABP4 (grey), and DSG2 EC3–4(magenta) are represented as ribbons. D) MCF-7 cell lysates were incubated with DSG2 antibody bound Dynabeads as described in Methods 1μM FABP4 (pre-incubated with 10 μM oleate) was added to the protein-beads complex for 2 hr. and complexed proteins captured by centrifugation. Proteins in the FABP4-DSG2 complex were eluted and evaluated by immunoblotting. E) GST-tagged FABP4 or R126Q was immobilized to glutathione resin in the presence or absence of 20 μM oleate followed by incubation with 0.2 μM His-DSG2 EC3–4. Immunoblotting was performed to analyze the eluted proteins after incubation. F) Quantification of the results from panel E where the level of DSG2 EC3–4 was normalized to GST-FABP4 or GST-R126Q.

Figure 7.. Loss of DSG2 abolished FABP4 mediated cell proliferation through ERK/NRF2 pathway in MCF-7 cells.

A) Two DSG2 knockdown (DSG2 siRNA#1 and DSG2 siRNA#2) MCF-7 cell lines were made. B) mRNA level of DSG2 expression in two DSG2 knockdown cells lines. C) Cells with stable knockdown of DSG2 were treated were treated with 200 ng/mL FABP4 for 48 hours and cell numbers were detected using CyQUANT Cell Proliferation Assay Kit. D) DSG2 knockdown cells were treated with 200 ng/mL FABP4 for 5 minutes and pERK level was evaluated using western blot. E) Quantification of pERK level in D). F) Cells with stable knockdown of DSG2 were treated with 200 ng/mL FABP4 for 48 hours and mRNA expression level of ALDH1A1 was evaluated using qRT-PCR. **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, nonsignificant.