Small extracellular vesicles from failing heart accelerate tumour growth

Abstract

Abstract Background Heart failure (HF) is associated with an increased incidence of cancer independently of shared risk factors. The mechanism behind this link is not entirely clear. Purpose We aimed to test the hypothesis that cardiac extracellular vesicles (cEVs) contribute to the link between HF and cancer. In addition, we aimed to determine if heart failure (HF) therapy will reduce the tumour-promoting effects of HF. Methods We used a mouse model of heterotopic lung cancer tumours and post-myocardial infarction (MI) HF. To determine the role of cEVs in tumour growth, we focused on cardiac mesenchymal stromal cells (cMSCs). First, we isolated cMSCs from mice hearts 10d after MI or sham MI and purified cMSC-EVs from the conditioned medium using size exclusion chromatography. Then, we analyzed cMSC-EV characteristics, cargo, and neoplastic effects on lung cancer cells. We also used cMSC-EV transfer to assess EV biodistribution and its effect on tumour growth. Next, we tested the effects of EV depletion on tumour growth. Finally, we used spironolactone to determine if HF therapy would attenuate the harmful interaction between HF and cancer. Results Mice with HF developed larger lung tumours than sham-MI mice during 28 days of follow-up. Failing hearts, particularly cMSCs, produced twice more EVs with neoplastic properties than non-failing hearts (Fig-A). Proteomics analysis revealed unique protein profiles of cMSC-EVs from failing hearts (Fig-B). cMSC-EVs from the failing hearts harbored higher quantities of tumour-promoting cytokines such as periostin, osteopontin, vascular endothelial growth factor, interleukin-6 and tumour necrosis factor-alpha; and tumorigenic microRNAs (miR) such as miR-221 and mir-21. Transfer of labeled cMSC-EVs from the failing heart into the systemic circulation of tumour-bearing mice targeted tumour cells and accelerated tumour growth (Fig-C). Moreover, EV depletion reduced the tumour-promoting effects of HF, and adoptive transfer of cMSC-EVs from failing hearts accelerated tumour growth in mice during systemic EV depletion (Fig-D). Finally, we used spironolactone to determine if HF therapy would inhibit the tumour trophic effects of HF. Spironolactone attenuated LV dilatation and improved global longitudinal strain following MI. Subsequently, spironolactone treatment reduced the number of EVs secreted by cMSCs from failing hearts by 26% and reduced the accelerated tumour growth after MI and HF (Fig E). Notably, spironolactone treatment did not reduce the volume of the tumours in sham-MI mice, indicating that the anti-tumour effects of spironolactone were related to its effects on the faling heart. Conclusions Our results provide new insight into the link between heart failure and cancer. We show, for the first time, that cardiac extracellular vesicles promote tumour growth independently of other secreted factors. Moreover, heart failure therapy reduces the tumour-promoting effects of heart failure.