On 2015 Aug 12, Víctor M. Baizabal-Aguirre commented:

The PLKO1-GSK3beta1 and PLKO1-GSK3beta2 vectors used in our work were successfully used in a previous report by Yoeli-Lerner et al. (2009). As in our study, these authors were also able to fully silence the GSK3beta isoform (PMID: 19258413, see Figures 1A and 3A). Most of the studies on GSK3-dependent regulation of beta-catenin have used unspecific inhibitors that affect both GSK3 isoforms. Therefore, the contribution of GSK3alpha and GSK3beta to the regulation of beta-catenin is an issue that remains open. In this regard, Yu et al. (2003) reported that specific gene silencing of GSK3alpha or beta by siRNA expression vectors induces the stabilization of beta-catenin in P19 mouse embryonic carcinoma cells (PMID: 12597911, see Figure 5). As to the effect of GSK3 silencing on beta-catenin, results published in 2009 by Mamaghani et al., demonstrated that GSK3alpha or beta inhibition by siRNA increased the stabilization of beta-catenin in pancreatic cancer cells (PMID: 19405981, see Figure 3A). In contrast, Ryu et al., (2012) reported that specific GSK3alpha inhibition by siRNA decreased beta-catenin levels in human gastric cancer cells (PMID: 22328534, see Figure 3E). These findings indicate that complete removal of GSK3alpha or GSK3beta, as in our work, affect the relative abundance of beta-catenin and that GSK3alpha and GSK3beta alter in different ways the stabilization of beta–catenin, depending on the type of cell.

On 2015 Aug 04, Jim Woodgett commented:

Figure 3 appears to be missing panel B (beta-catenin blot).

The RNAi knockdowns here are remarkably efficient and it looks from the legend as though only one siRNA was used (although in the methods 3 alpha sequences and 2 beta sequences are listed). Figure 5 shows essentially complete removal. Blowing up the figure reveals some image artifacts. Typically, even complete inhibition of either GSK-3alpha or GSK-3beta has no effect on beta-catenin as the other isoform compensates fully (Axin is present at far lower concentrations that either isoform of GSK-3 and is the limiting factor in beta-catenin phosphorylation).

On 2015 Aug 04, Jim Woodgett commented:

Figure 3 appears to be missing panel B (beta-catenin blot).

The RNAi knockdowns here are remarkably efficient and it looks from the legend as though only one siRNA was used (although in the methods 3 alpha sequences and 2 beta sequences are listed). Figure 5 shows essentially complete removal. Blowing up the figure reveals some image artifacts. Typically, even complete inhibition of either GSK-3alpha or GSK-3beta has no effect on beta-catenin as the other isoform compensates fully (Axin is present at far lower concentrations that either isoform of GSK-3 and is the limiting factor in beta-catenin phosphorylation).

On 2015 Aug 12, Víctor M. Baizabal-Aguirre commented:

The PLKO1-GSK3beta1 and PLKO1-GSK3beta2 vectors used in our work were successfully used in a previous report by Yoeli-Lerner et al. (2009). As in our study, these authors were also able to fully silence the GSK3beta isoform (PMID: 19258413, see Figures 1A and 3A). Most of the studies on GSK3-dependent regulation of beta-catenin have used unspecific inhibitors that affect both GSK3 isoforms. Therefore, the contribution of GSK3alpha and GSK3beta to the regulation of beta-catenin is an issue that remains open. In this regard, Yu et al. (2003) reported that specific gene silencing of GSK3alpha or beta by siRNA expression vectors induces the stabilization of beta-catenin in P19 mouse embryonic carcinoma cells (PMID: 12597911, see Figure 5). As to the effect of GSK3 silencing on beta-catenin, results published in 2009 by Mamaghani et al., demonstrated that GSK3alpha or beta inhibition by siRNA increased the stabilization of beta-catenin in pancreatic cancer cells (PMID: 19405981, see Figure 3A). In contrast, Ryu et al., (2012) reported that specific GSK3alpha inhibition by siRNA decreased beta-catenin levels in human gastric cancer cells (PMID: 22328534, see Figure 3E). These findings indicate that complete removal of GSK3alpha or GSK3beta, as in our work, affect the relative abundance of beta-catenin and that GSK3alpha and GSK3beta alter in different ways the stabilization of beta–catenin, depending on the type of cell.