Metallothionein-mediated antioxidant defense system and its response to exercise training are impaired in human type 2 diabetes. Diabetes 2005;54:3089–3094

In November 2005, our group published the above-mentioned article. At the root of the paper was a series of immunohistochemical stains for metallothionein (MTI1II), which one of the co-authors, Milena Penkowa (former professor at the University of Copenhagen), had performed. Recently, the integrity of Milena Penkowa’s entire scientific work has been under investigation by an independent international panel set up by the University of Copenhagen. Their findings put into serious question the validity of Dr. Penkowa’s contribution to several papers, including this one. In order to support the efforts of the panel, we attempted to replicate Dr. Penkowa’s results presented in the article using other techniques (Western blots as opposed to the original immunohistochemical stains). A researcher outside of the original author group (Andreas Bergdahl, PhD, MSc, Concordia University, Montreal, Canada) kindly agreed to perform the protein analysis using a commercially available antibody, which would enable us to support the data obtained by Dr. Penkowa. However, we were not able to replicate the data presented in the article. The Western blot analysis revealed that there was an overall effect of diabetes on skeletal muscle metallothionein (MT) levels but that MT was higher in the diabetic subjects versus the control subjects. There was no effect of exercise. This is contrary to the data reported in the article. As there is serious question to the integrity of Dr. Penkowa’s contribution to this work and our inability to experimentally support the original findings, we, the authors, have requested that the article be withdrawn. We regret any inconvenience to the readers of Diabetes.