ABSTRACT
Introduction: An inadequate wound healing following myocardial infarction (MI) is one of the main etiologies of heart failure (HF) development. Interventions aiming at improving this process may contribute to preserving cardiac function after MI. Our group, as well as others, have demonstrated the crucial role of Wnt/frizzled signaling in post-MI remodeling. In this overview, we provide the results of different studies aimed at confirming an initial study from our group, in which we observed beneficial effects of administration of a peptide fragment of Wnt5a, UM206, on infarct healing in a mouse MI model. Methods: Mice were subjected to permanent left coronary artery ligation, and treated with saline (control) or UM206, administered via osmotic minipumps. Cardiac function was assessed by echocardiography and hemodynamic measurements, while infarct size and myofibroblast content were characterized by (immuno)histochemistry. Results: In total, we performed seven follow-up studies, but we were unable to reproduce the beneficial effects of UM206 on infarct healing in most of them. Variations in dose and timing of UM206 administration, its manufacturer and the genetic background of the mice could not restore the phenotype. An in-depth analysis of the datasets revealed that the absence of effect of UM206 coincided with a lack of adverse cardiac remodeling and HF development in all experimental groups, irrespective of the treatment. Discussion: Irreproducibility of experimental observations is a major issue in biomedical sciences. It can arise from a relatively low number of experimental observations in the original study, a faulty hypothesis or a variation in the experimental model that cannot be controlled. In this case, the lack of adverse cardiac remodeling and lung weight increases in the follow-up studies point out to altered experimental conditions as the most likely explanation.
