Hepatic lipids promote liver metastasis

Abstract

Obesity predisposes to cancer and a virtual universality of nonalcoholic fatty liver disease (NAFLD). However, the impact of hepatic steatosis on liver metastasis is enigmatic. We find that while control mice were relatively resistant to hepatic metastasis, those which were lipodystrophic or obese, with NAFLD, had a dramatic increase in breast cancer and melanoma liver metastases. NAFLD promotes liver metastasis by reciprocal activation initiated by tumor-induced triglyceride lipolysis in juxtaposed hepatocytes. The lipolytic products are transferred to cancer cells via fatty acid transporter protein 1, where they are metabolized by mitochondrial oxidation to promote tumor growth. The histology of human liver metastasis indicated the same occurs in humans. Furthermore, comparison of isolates of normal and fatty liver established that steatotic lipids had enhanced tumor-stimulating capacity. Normalization of glucose metabolism by metformin did not reduce steatosis-induced metastasis, establishing the process is not mediated by the metabolic syndrome. Alternatively, eradication of NAFLD in lipodystrophic mice by adipose tissue transplantation reduced breast cancer metastasis to that of control mice, indicating the steatosis-induced predisposition is reversible.

Keywords: Cancer; Hepatology; Obesity; Oncology.

Figure 1. Lipodystrophic mice are predisposed to hepatic metastasis.

(A–D) Two-month-old FF and Con mice were injected with Bo1 cells via left ventricular chamber, and 12 days later tumor burden was analyzed. (A) In vivo BLI image; (B) ex vivo image of liver, bone, lung, and kidney; (C) gross appearance of Con and FF liver; and (D) quantification of tumor burden in liver, bone, lung, and kidney of Con and FF mice. n = 8–13. ROI, region of interest. (E–H) Two-month-old PPAR ADQ and Con mice were injected with Bo1 cells via left ventricular chamber, and 12 days later tumor burden was analyzed. (E) In vivo BLI image; (F) ex vivo image of liver, bone, lung, and kidney; (G) gross appearance of Con and FF liver 12 days after tumor injection; and (H) quantification of tumor burden in liver, bone, lung, and kidney of Con and FF mice. n = 7. Data are presented as mean ± SD. **P < 0.01, ***P < 0.001 as determined by unpaired 2-tailed t test.

Figure 2. Obese mice are predisposed to hepatic metastasis.

(A and B) Two-month-old leptin^–/– (ob/ob) and control mice were injected with Bo1 cells via left ventricular chamber, and 12 days later, tumor burden was analyzed. (A) Ex vivo image of liver, bone, lung, and kidney. (B) Quantification of tumor burden in liver, bone, lung, and kidney. n = 9–11. (C and D) WT mice were fed chow or HFD for 3 months, after which Bo1 cells were injected intracardiacally. (C) Ex vivo image of liver, bone, lung, and kidney and (D) quantification of tumor burden in liver, bone, lung, and kidney 12 days later. n = 8–13. (E and F) FF mice were injected intracardiacally with Bo1 cells 6 weeks after fat transplantation. (E) In vivo BLI image and (F) quantification of tumor burden in liver 12 days later. n = 8–9. (G) Following 2 months of metformin feeding, FF mice were injected with Bo1 cells. BLI quantification of liver tumor burden 12 days after injection. n = 4–5. Data are presented as mean ± SD. **P < 0.01, ***P < 0.001 as determined by unpaired 2-tailed t test (B and D) or 1-way ANOVA test with analysis of variance with Holm-Šidák multiple-comparisons test (F and G).

Figure 3. Steatotic hepatocytes transfer lipids to tumor.

(A) H&E staining of FF steatotic liver before (- tumor) and 12 days after Bo1 intracardiac injection (+ tumor). Note the disappearance of steatosis juxtaposed to tumor. Scale bar: 400 μm. T, tumor. (B) Oil red O staining of FF liver 8 days after intracardiac injection of Bo1 cells. Arrow indicates disappearance of lipid staining in hepatocytes juxtaposed to tumor. Scale bar: 1 mm. (C) H&E staining of liver of an NAFLD patient with breast cancer metastasis demonstrating areas distal (left panel) and proximal (right panel) to tumor. Scale bar: 1 mm. (D) Ultrastructural image of Bo1 cells in the liver of control or FF mice 10 days after intracardiac injection. Yellow asterisks indicate lipid droplets. Scale bar: 2 μm. (E) The number of lipid droplets inside a single Bo1 cell in the liver of control and FF mice 10 days after intracardiac injection. n = 4–5. (F) Flow cytometric analysis of frequency of mCherry-labeled Bo1- cells containing BODIPY derived from HepG2 cells after 0, 3, 6, 12, or 24 hours of coculture. n = 3. (G) Confocal Z-stack imaging of mCherry-labeled Bo1^– cell after 3 hours of coculture with BODIPY-treated HepG2 cells. Ten sections of a confocal Z-stack were obtained through the height of the cell. The optical section in the middle of cell is presented. The side image on the merged panel is the cross section, and the top image is the longitudinal section of the Bo1^– cell present in the middle. The image identifies fluorescence-labeled BODIPY within the tumor cell. Scale bar: 10 μm. Data are presented as mean ± SD. ****P < 0.0001 as determined by unpaired 2-tailed t test (E).

Figure 4. Tumor promotes lipolysis of steatotic liver.

(A) TG in control or FF liver explants cultured with or without Bo1 cells. n = 6. (B) Expression of ATGL and HSL in control or FF liver explants cultured with Bo1 cells. n = 4. (C) FFA in control or FF liver explants cultured with or without Bo1 cells. n = 6. (D) PET scan of ¹¹C-palmitate uptake by control or FF liver 10 days after Bo1 intracardiac injection. (E) ¹¹C-palmitate uptake quantification. n = 4. SUV, standardized uptake value. (F) FATP1 mRNA expression in cancer cells exposed to control or steatotic liver–conditioned medium. n = 3. Data are presented as mean ± SD. *P < 0.05, ***P < 0.001 as determined by unpaired 2-tailed t test (B, E, and F) or 2-way ANOVA test with analysis of variance with Holm-Šidák multiple-comparisons test (A and C).

Figure 5. Steatosis promotes tumor growth.

(A) Morphology of Bo1 cells cultured with Con or steatotic liver of FF mice in serum-free medium; left panel shows the normal Bo1 cells cultured with medium with 10% serum. Scale bar: 100 μm. (B and C) Bo1 cells were injected directly into livers of 2-month-old FF and control mice. Tumor burden was analyzed 8 days later: (B) in vivo BLI image and (C) quantification of tumor burden in liver. n = 3–5. (D) PCNA immunostaining of liver of control or FF mice 12 days after intracardiac injection of Bo1 cells. n = 3. Data are presented as mean ± SD. *P < 0.05, as determined by unpaired 2-tailed t test (C).

Figure 6. Steatosis promotes metastatic tumor growth.

(A) BrdU incorporation and (B) AKT and cyclin D1 immunoblot of Bo1 cells treated with lipids (100 μg/mL) from control or FF liver. n = 5. (C) Seahorse oxygen consumption rate (OCR) of Bo1 cells cocultured with control or fatty liver. n = 9. (D) Carnitine palmitoyltransferase 1 (CPT1) mRNA expression in Bo1 cells cocultured with control or FF liver. n = 4–5. (E) BLI analysis of FF liver 12 days after intracardiac injection of CRISPR control or CRISPR-CPT1–knockout Bo1 cells. n = 4–7. (F) Bo1 cells were cultured with control or FF liver explants in Transwell system and migrating cell number was analyzed. n = 4. (G and H) Two-month-old FF and control mice were injected intracardiacally with Bo1 cells. Tumor burden was analyzed by BLI 24 hours later: (G) ex vivo image of liver and (H) quantification of tumor burden in liver. n = 6. Data are presented as mean ± SD. *P < 0.05, ***P < 0.001 as determined by unpaired 2-tailed t test (A, C, D, F, and H) or 1-way (E) ANOVA test with analysis of variance with Holm-Šidák multiple-comparisons test.