Dose-related cardiomyopathy is a major side effect following doxorubicin (Dox). To investigate whether exercise (Ex)-induced vasculogenesis plays a role in reducing Dox-induced cardiotoxicity, GFP⁺ bone marrow (BM) cells from GFP transgenic mice were transplanted into wild-type mice. Transplanted mice were treated with Dox, Ex, Dox+Ex, or control. We found Dox therapy resulted in decreased systolic and diastolic blood flow, decreased ejection fraction and fractional shortening, and decreased vascular endothelial cells and pericytes. These abnormalities were not seen in Dox+Ex hearts. Heart tissues from control-, Ex-, or Dox-treated mice showed a small number of GFP⁺ cells. By contrast, the Dox+Ex-treated hearts had a significant increase in GFP⁺ cells. Further analyses demonstrated these GFP⁺ BM cells had differentiated into vascular endothelial cells (GFP⁺CD31⁺) and pericytes (GFP⁺NG2⁺). Decreased cardiomyocytes were also seen in Dox-treated but not Dox+Ex-treated hearts. Ex induced an increase in GFP⁺c-Kit⁺ cells. However, these c-Kit⁺ BM stem cells had not differentiated into cardiomyocytes. Dox therapy induced phosphorylation of MST1/2, LATS1, and YAP; a decrease in total YAP; and cleavage of caspase-3 and PARP in the heart tissues. Dox+Ex prevented these effects. Our data demonstrated Dox-induced cardiotoxicity is mediated by vascular damage resulting in decreased cardiac blood flow and through activation of Hippo-YAP signaling resulting in cardiomyocyte apoptosis. Furthermore, Ex inhibited these effects by promoting migration of BM stem cells into the heart to repair the cardiac vessels damaged by Dox and through inhibiting Dox-induced Hippo-YAP signaling-mediated apoptosis. These data support the concept of using exercise as an intervention to decrease Dox-induced cardiotoxicity.

Ultrasonography (US) is useful for visual detection of edematous tissues to assess subcutaneous echogenicity. However, visualization of subcutaneous echogenicity is interpreted differently among operators because the evaluation is subjective and individual operators have unique knowledge. This study objectively assessed leg edema using US with a gel pad including fat for normalization of echogenicity in subcutaneous tissue. Five younger adults and four elderly people with leg edema were recruited. We compared assessments of US and limb circumference before and after the intervention of vibration to decrease edema in younger adults, and edema prior to going to sleep and reduced edema in the early morning in elderly people. These assessments were performed twice in elderly people by three operators and reliability, interrater differences, and bias were assessed. For US assessment, echogenicity in subcutaneous tissue was normalized to that of the gel pad by dividing the mean echogenicity of subcutaneous tissue by the mean echogenicity of the gel pad. In younger adults, the normalized subcutaneous echogenicity before the intervention was significantly higher than that after the intervention. In elderly people, echogenicity indicating edema was significantly higher than that after edema reduction. Edema was detected with accuracy rates of 76.9% in younger adults and 75.0% in elderly people. Meanwhile, limb circumference could be used to detect edema in 50.0% of healthy adults and 87.8% of elderly people. The intra-reliability was excellent (intraclass correlation coefficient > 0.9, p < 0.01), and the inter-reliability was good (intraclass correlation coefficient > 0.7, p < 0.01) for normalized subcutaneous echogenicity. Bland-Altman plots revealed that inter-rater differences and systematic bias were small. Normalized subcutaneous echogenicity with the pad can sensitively and objectively assess leg edema with high reliability. Therefore, this method has the potential to become a new gold standard for objective assessment of leg edema in clinical practice.