Cardiac Dysfunction Promotes Cancer Progression via Multiple Secreted Factors

Abstract

Heart failure and cancer are the leading cause of deaths worldwide. While heart failure and cancer have been considered separate diseases, it is becoming evident that they are highly connected and affect each other's outcomes. Recent studies using experimental mouse models have suggested that heart failure promotes tumor progression. The mouse models used involve major irreversible surgery. Here, we induced heart hypertrophy via expression of activating transcription factor 3 (ATF3) in cardiomyocytes, followed by cancer cells' implantation. Tumors developing in ATF3-transgenic mice grew larger and displayed a more highly metastatic phenotype compared with tumors in wild-type mice. To address whether ATF3 expression or the cardiac outcome are necessary for tumor progression, ATF3 expression was turned off after cardiac hypertrophy development followed by cancer cell implantation. The tumor promotion phenotype and the enhancement of metastatic properties were preserved, suggesting that the failing heart per se is sufficient to promote tumor progression. Serum derived from ATF3-transgenic mice enhanced cancer cell proliferation and increased cancer cell metastatic properties in vitro. Using a cytokine array panel, multiple factors responsible for promoting tumor cell proliferation and the metastatic phenotype were identified. Interestingly, the failing heart and the tumor separately and simultaneously contributed to higher levels of these factors in the serum as well as other tissues and organs. These data suggest the existence of intimate cross-talk between the hypertrophied heart and the tumor that is mediated by secreted factors, leading to cancer promotion and disease deterioration.

Significance: This work highlights the importance of early diagnosis and treatment of heart failure prior to reaching the irreversible stage that can exacerbate cancer progression.

Figure 1.

Cardiac ATF3-transgenic promotes PyMT tumor progression. A, Schematic experimental timeline for polyoma middle T (PyMT) cancer cell model with female mice control (cont.; n = 13) or ATF3 transgenic (ATF3 tg.; n = 10). Doxycycline (Dox) was provided until weaning. Control and ATF3-transgenic were orthotopically implanted into the mammary fat pad with PyMT cells (10⁵ cells per mouse). B, Tumor volume was monitored over time. C, Tumor weight at the endpoint. D, Representative images of tumor sections stained for Masson's trichrome. Scale bar, 200 µm. E, Quantification of the percent of the interstitial fibrosis. F, Tumor mRNA levels of the Collagen1α and TGFβ3 genes were measured by qRT-PCR, normalized to GAPDH housekeeping gene. Data are presented as relative expression compared with control determined as 1. Each dot represents one mouse. Data are presented as mean ±SD. Two-way repeated measures ANOVA followed by Bonferroni posttests (B), Student t test (C and E) or one-way repeated measures ANOVA followed by Tukey posttests. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 2.

Cardiac ATF3-transgenic promotes cancer cell proliferation. A, Representative image of PyMT tumor sections of control and ATF3-transgenic stained with anti-Ki67 (proliferating cells; red) and DAPI (nuclei; blue). Scale bar, 50 µm. B, Quantification of the number of proliferating cells in tumor section field. Each dot represents the mean of five fields in tumor section derived from a single mouse (n = 4 mice each) C, PyMT cells were cultured for 48 hours in the absence or supplemented with 10%: FBS, mouse blood serum drawn from either control or ATF3-transgenic. Proliferation was measured by Luminescent Cell Viability Assay kit using serum derived from at least three mice per group (n = 4 wells per treatment). D, Tumor sections derived from either ATF3-transgenic or control mice (n = 3 mice per group shown). Scale bar, 200 μm. E and F, Migration (E) and invasion (F) of PyMT cells assessed by Boyden chamber assays in the presence of fibronectin or Matrigel, respectively. Representative images are shown. Scale bar, 200 µm. Cell coverage was quantified from the images (n = 5–8 fields/group). Data are presented as mean ± SD. Student t test (B) or one-way repeated measures ANOVA followed by Tukey posttests (C, E, and F). *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 3.

Cardiac dysfunction enhances cancer cell colonization to the lungs in ATF3-transgenic mice model. A, Schematic experimental timeline for ATF3-transgenic and control mice followed by PyMT cell pulmonary experimental metastasis assay by tail vein injection (TV Inj.) of 2 × 10⁶ cells per mouse; Control (Cont.; n = 3) or ATF3-transgenic (ATF3 tg.; n = 4) female mice. B, Representative image of lung sections stained with H&E. Scale bar, 1,000 µm. C, Number of metastatic lesions in the lungs. D, Average area of the metastatic lesions. Data are presented as mean ±SD. Student t test. *, P < 0.05.

Figure 4.

Tumor promotion phenotype is independent on continuous ATF3-transgene expression. A, Schematic experimental timeline for PyMT cell model with control (cont.; n = 5) or ATF3-transgenic (ATF3 tg.; n = 7) female mice. Doxycycline (Dox) was removed at 3 weeks of age and was added back at 7 weeks of age. PyMT cells were injected one week later. Mice were sacrificed at humane endpoint. B, Tumor volume was monitored over time. C, Tumor weight at the endpoint. Each dot represents one mouse. D, PyMT cells were cultured for 48 hours in serum-free medium in the absence or supplemented with 10%: FBS, mouse blood serum drawn from either control or ATF3-transgenic. Proliferation was measured by Luminescent Cell Viability Assay using serum from at least three mice per group (n = 4 wells per treatment). Data are presented as mean ±SD. Two-way repeated measures ANOVA followed by Bonferroni posttests (B), Student t test (C), or one-way repeated measures ANOVA followed by Tukey posttests (D). ***, P < 0.001.

Figure 5.

Tumor progression phenotype is mediated by multiple secreted factors derived from the heart, tumor, and other tissues. A, Serum from either ATF3-transgenic or control mice was used to probe proteome cytokine array. For each protein, serum levels are presented as fold change relative to control serum. B, Serum levels as in A, obtained by ELISA for ceruloplasmin (CP) and CTGF and fibronectin (FN). C and D, Heart (C) and tumor (D) mRNA levels of the indicated genes were measured by qRT-PCR. Color code for the secreted factors from the heart (red), tumor (black), tumor and heart simultaneously (light blue), and other tissues and organs (green). Data are presented as mean ±SD relative expression compared with control, which was determined as 1. One-way ANOVA followed by Tukey posttests or multiple Student t tests. *, P < 0.05; **, P < 0.01; ***, P < 0.001.