Non-alcoholic fatty liver disease promotes breast cancer progression through upregulated hepatic fibroblast growth factor 21

Abstract

Non-alcoholic fatty liver disease (NAFLD) has been shown to influence breast cancer progression, but the underlying mechanisms remain unclear. In this study, we investigated the impact of NAFLD on breast cancer tumor growth and cell viability through the potential mediator, hepatic fibroblast growth factor 21 (FGF21). Both peritumoral and systemic administration of FGF21 promoted breast cancer tumor growth, while FGF21 knockout attenuated the tumor-promoting effects of the high-fat diet. Mechanistically, exogenous FGF21 treatment enhanced the anti-apoptotic ability of breast cancer cells through STAT3 and Akt/FoXO1 signaling pathways, and mitigated doxorubicin-induced cell death. Furthermore, we observed overexpression of FGF21 in tumor tissues from breast cancer patients, which was associated with poor prognosis. These findings suggest a novel role for FGF21 as an upregulated mediator in the context of NAFLD, promoting breast cancer development and highlighting its potential as a therapeutic target for cancer treatment.

Fig. 1. HFD induces NAFLD and promotes the development of mammary tumors in PyVT mice.

A Trial schematic for establishing HFD-fed NAFLD-breast cancer mice model. Briefly, hemizygous female MMTV-PyMT mice were bred for 3 weeks and genotyped. Mice were randomly divided into the HFD and SFD groups. Weekly measurements included body weight, food intake, and tumor volume with body composition tested before sacrifice. B, C Mice on the HFD showed accelerated body weight growth (B) and higher body fat rates (C). D Increased weights of adipose tissue were observed in both posterior and gonadal white adipose tissues. E–G Evaluation of NAFLD revealed that the HFD did not induce significant changes in liver enzymes (E) but stimulated lipid droplet accumulation in the liver tissues, as observed through H&E staining (F) and oil red O staining (G); Images were shown at ×4 magnification, scale bars of 200 μm, and insert images were at ×20 magnification. H–O For tumors, mice on the HFD showed accelerated mammary tumor growth (H) and higher tumor wet weight (I); J, K Lung metastasis was evaluated by H&E staining (J), and the number of metastasis foci was quantified (K). L, M The expression levels of PCNA were confirmed by western blot (L) and quantified (M). N, O Immunofluorescence staining of Ki67 was performed on tumor slices (N) and the percentage of Ki67-positive cells was quantified (O). Representative images were shown at ×40 magnification, and scale bars were 10 μm. n = 15 for each group; Data were expressed as mean ± SD. The difference between groups was assessed using Student’s t-test, *p < 0.05, **p < 0.01. HFD high-fat diet group, SFD standard-food diet group, PWAT posterior white adipose tissue, GWAT gonadal white adipose tissue.

Fig. 2. HFD induces NAFLD and promotes the development of mammary tumors in C57BL/6J mice.

A Trial schematic for establishing an HFD-fed NAFLD-breast cancer mice model. B, C Mice on the HFD diet showed accelerated body weight growth (B) and higher body fat rates (C). D Increased wet weights of adipose tissue were observed. E–H Evaluation of NAFLD revealed that the HFD induced higher liver wet weight (E) and aberrant expression of liver enzymes (H). Lipid droplet accumulation in the liver tissue was detected by H&E (F) and oil red O (G) staining; Images were shown at ×4 magnification, with scale bars of 200 μm, and insert images were at ×20 magnification. I–L For tumors, mice fed with HFD showed accelerated mammary tumor growth (I) and higher tumor wet weight (J). The expression levels of PCNA were confirmed by western blot (K) and quantified (L). n = 10 for each group; Data were expressed as mean ± SD. The difference between groups was assessed by Student’s t-test, *p < 0.05, **p < 0.01. HFD high-fat diet group, SFD standard-food diet group, PWAT posterior white adipose tissue, GWAT gonadal white adipose tissue.

Fig. 3. Conditioned media from hepatocytes promote cell viability of breast cancer cell lines.

A Trial schematic for establishing an in vitro NAFLD model. Briefly, primary hepatocytes were isolated from C57BL/6J mice and identified with immunofluorescence staining of KRT18. To establish the NAFLD model, hepatocytes were treated with FFAs for 24 h and then cultured with serum-free medium for an additional 12 h. The conditioned medium was collected, filtered, and diluted to 25% with fresh culture medium before being used to treat breast cancer cell lines. B The presence of lipid droplets in hepatocytes was detected by oil-red O staining. The representative images were shown at ×20 magnification, and the scale bars were 50 μm. C Cell viability was assessed using the MTS assay. The results suggested that conditioned medium from hepatocytes promoted breast cancer cell growth. n = 5 for each group; Data were expressed as mean ± SD. The difference between groups was assessed by Student’s t-test, * p < 0.05, **p < 0.01. FFAs: 1 mM palmitic acid combined with 0.25 mM oleic acid. The trial schematic was created with BioRender.com.

Fig. 4. FGF21 is found to be over-expressed in the NAFLD-breast cancer model.

A–C The significantly changed secretory proteins in the livers of HFD-fed C57BL/6J mice were investigated. A A total of 231 genes were sequenced using mRNA sequencing and found to be differentially expressed between the groups. Among these, 52 genes were identified to have the secretory signal peptides using SignalP 6.0. Their functional information was further evaluated by UniProt and GeneCards. B Finally, 25 genes with potential extracellular functions were identified as differentially expressed in NAFLD livers. C The expression levels of these 25 genes were further confirmed by RT-qPCR. The experiments were conducted with n = 3 for each group in (A–C). D, E The expression of FGF21 in the livers was evaluated by western blot (D) and quantified (E). F, G The expression of FGF21 in tumors was evaluated by western blot (F) and quantified (G). H The expression pattern of FGF21 in tumor tissues was analyzed by immunohistochemistry staining. Images were shown at ×4 magnification, scale bars 200 μm. Black arrow: peritumor area; Brown arrow: tumor area. J, K The expression levels of FGF21 were tested in immortalized hepatic cell lines MIHA and AML-12, as well as primary hepatocytes. J The mRNA expression levels of FGF21 were elevated in the conditioned media of FFAs-treated hepatocytes. K The protein levels of FGF21 were elevated in the conditioned media of FFAs-treated hepatocytes. The experiments were conducted with n = 3 for each group in (J and K). I The increased serum FGF21 levels in C57BL/6J mice were detected using an ELISA kit, with n = 6 for each group. Data were expressed as mean ± SD. The difference between groups was assessed by Student’s t-test, *p < 0.05, **p < 0.01. HFD high-fat diet group, SFD standard-food diet group, FFAs 1 mM palmitic acid combined with 0.25 mM oleic acid.

Fig. 5. Administration of recombinant FGF21 promotes breast cancer tumor growth.

A–D FGF21 peritumoral injection model. A Trial schematic. B, C The promoting effects of FGF21 on breast cancer were observed as accelerated tumor growth (B) and higher tumor weight (C). D Tumor sampling was performed 2 h after FGF21 administration was applied to confirm the in-situ enrichment of FGF21. E–I FGF21 sustained-release model. E Trial schematic. F The serum levels of FGF21 were measured by ELISA at the endpoint, and mice in the FGF21 group showed comparable serum levels to the HFD-fed mice shown in Fig. 4I. G, H Mice with FGF21 supplementation showed faster tumor growth (G) and higher tumor weight (H). I The expression levels of tumoral FGF21 were increased in the treatment group. J–O FGF21 knockout model. J Trial schematic. K, L Liver tissue exhibited over-accumulation of lipid droplets with HFD (K), stained red with oil red O (L). M, N Mice on the HFD showed comparable tumor growth curves (M) and tumor weights (N) to mice in the SFD group. O Conditioned medium from FGF21 KO hepatocytes was collected to test the effects on breast cancer cell lines 4T1 and E0771. n = 6 for each group. Data were expressed as mean ± SD. The difference between groups was assessed by Student’s t-test, *p < 0.05, **p < 0.01, ns p > 0.05. Images were shown at ×4 magnification, with scale bars 200 μm, and inset images at ×20 magnification. HFD high-fat diet group, SFD standard-food diet group, FFAs 1 mM palmitic acid combined with 0.25 mM oleic acid.

Fig. 6. Recombinant FGF21 enhances the anti-apoptotic capability of breast cancer cells through STAT3 and Akt/FoXO1 pathways.

A MDA-MB-231 cells were exposed to recombinant FGF21 at different concentrations (0, 0.5, 5, 50 ng/mL) for 24 h prior to the MTS assay. B–E The treatment with FGF21 activated FGF receptors and anti-apoptotic pathways, as indicated by the quantification of the blots. F–I The antagonistic effects of FGF21 towards doxorubicin were examined. MDA-MB-231 cells were treated with 50 ng/mL FGF21 and 1.25 μM DOX for 24 h. F Cell apoptosis was evaluated using Annexin V assay, with representative images shown (left) and quantified (right). G TUNEL assay was also performed to evaluate cell apoptosis with representative images shown (left) and quantified (right). Images were shown at ×4 magnification, scale bars 500 μm, and insert images were at ×20 magnification. The difference in late apoptosis between DOX and D + F groups was marked. H, I The expression levels of apoptosis-related proteins were determined by the caspase activity test (H) and western blot (I). J, K FGF21 was found to protect mitochondria. J Mitochondrial membrane potential was evaluated using JC-10, and the fluorescence intensity in live cells was detected by confocal microscopy. K The intra-mitochondrial cytochrome c was detected by immunofluorescence. Cytochrome c was stained with anti-cytochrome c antibody in green, the nucleus was counterstained with DAPI in blue, and the mitochondrion was stained with mitotracker in red. Images for J and K were shown at ×40 magnification, scale bars 10 μm. L, M The activation of apoptosis pathways was detected in the FGF21 sustained-release model. Graphs represent the quantification of the blots. Data were expressed as mean ± SD. The difference between groups was assessed by Student’s t-test or One-way ANOVA combined with Turkey’s test for multiple comparison tests, *p < 0.05, **p < 0.01 compared to control groups unless otherwise stated. Ctrl control group, DOX doxorubicin treatment group, D + F doxorubicin plus FGF21 treatment group, S sham group, F FGF21 treatment group.

Fig. 7. FGF21 is overexpressed in breast cancer tissue and correlates with prognosis.

IHC staining was utilized to investigate the relationship between FGF21 expression levels and clinical pathological characteristics. A Tumor samples from patients with different molecular types of breast cancer were collected and subjected to staining with anti-FGF21 antibodies. Images were shown at ×4 magnification, with scale bars 200 μm, and insert images at ×20 magnification. B, C A tissue microarray consisting of tumors from 157 TNBC patients was analyzed. B Overexpression of FGF21 was observed in subjects with recurrent and deceased patients. The difference between groups was assessed by Student’s t-test, and the mean with SD was presented. C Patients were categorized into low and high-expression groups based on FGF21 levels. The disease-free survival rate and overall survival between these groups were analyzed using the Kaplan–Meier tool, and differences were analyzed by log-rank test. PNTAT paired normal tissues adjacent to the tumor.