Statistics
The data are expressed as the means ± SEM. We performed the experimental procedures and treatment and data analyses under blinded conditions. To reduce the baseline variability between independent experiments, we normalized the quantitative analyses of immunoblots and mRNA expression. The data were normalized as the fold change over the mean of the control. We compared 2 groups by Student’s t test; 3 or more different groups were evaluated by one-way analysis of variance (ANOVA), And if F reached significance two post hoc tests were used: the Dunnett post hoc test when comparing each group with control, or the Sidak post hoc test whenever a multiple group comparison was necessary. For in vivo studies, all groups were initially designed to contain 8 or 6 mice. We used nonparametric methods (Kruskal–Wallis test followed by Dunn’s post hoc tests when F reached significance). P ˂ 0.05 was considered significant. We performed statistical analyses of the data using SPSS v18.0 software.