After publication of this article [1], concerns were raised about results presented in Figs 3 and 7. Specifically:

• The ‘-‘panel of Fig 3A appears to partially overlap with the 20 mg/kg panel of Fig 3B.
• The β-Tubulin panel in Fig 7F appears similar to the Lamin B panel in Fig 7G.

The corresponding author stated that the 20 mg/kg CK panel in Fig 3B is incorrect, and provided a revised version of Fig 3 including the correct image from the original study along with the underlying images for Fig 3 in S1 File.

The corresponding author stated that the β-Tubulin panel in Fig 7F and the Lamin B panel in Fig 7G were duplicated in error and that it is not known which one is incorrect as the underlying images for these figures are no longer available. They provided data in support of the results reported in Figs 7F and 7G from later repeat experiments which had two independent replicates; the repeat western blot image files and densitometry data are provided in S2-S4 Files.

Members of the PLOS One Editorial Board reviewed S1-S4 Files and stated that the updated 20 mg/kg CK panel in Fig 3B and the repeat data for Figs 7F and G appear to accurately support the conclusion that BB and CX inhibit LPS-induced NF-κB, but noted the limitation of a small sample size for the repeat western blot experiment.

In addition, there is an error in the last sentence of the second paragraph of the Results subsection titled CK Inhibited Activation of NF-κB Pathway in Macrophages. The correct sentence is: Moreover, western blotting results indicated that CK inhibited IκB phosphorylation in a concentration-dependent manner.

During editorial follow-up, members of the PLOS One Editorial Board noted that the study design of [1] does not test causality as analysis in cells or organisms with targeted deletions in the pathway needed to abrogate the effect are not included in [1], and, as such, statements regarding causal relationships between NF-κB inhibition and the beneficial effects of compound K on DSS-induced colitis are speculative. Examples of statements that are unsupported by the study design include ‘by Suppressing NF-κB Activation’ in the title, abstract, and introduction of [1], “through inhibiting NF-κB pathway activation, thus leading to...” in the Results section, and “through modulating NF-κB signalling pathways activation that dominate the production of cytokines” in the Discussion section. Members of the PLOS One Editorial Board stated that while the data demonstrate that CK treatment is associated with both improved colitis outcomes and suppression of NF-κB activation, they do not include mechanistic experiments that would directly establish causality. PLOS also has concerns about the small sample sizes reported for the in vivo experiments.

The corresponding author stated that the individual-level underlying data and statistical analysis for Figs 1-2 and 4-7 are no longer available.

The PLOS One Editors issue this Expression of Concern to inform readers of the above issues, which call into question the reliability of the article’s conclusions.