Retraction: Neuroprotective Role of L-NG-Nitroarginine Methyl Ester (L-NAME) against Chronic Hypobaric Hypoxia with Crowding Stress (CHC) Induced Depression-Like Behaviour

Abstract

Full text

After this article [1] was published, the following concerns were raised:

In response to the concern for Fig 7C Control+Veh and Fig 7C Control+L-NAME, the first author stated that the Fig 7C Control+Veh panel in [1] is incorrect. An updated Fig 7C Control+Veh panel was provided by the first author along with the original data underlying Fig 7. PLOS considers this concern to be resolved.

In response to the concerns for partial overlap between Fig 7C Control+L-NAME and Fig 7D Control+L-NAME, and between Fig 10A CHC + Veh and Fig 10B CHC + Veh, the first author stated that these panels represent two different regions of the hippocampus (CA1 and CA3). They stated that overlap is visible between the panels in question as they were captured from the same sections, and due to the close proximity of these regions. They also stated that with image stack overflow, there is expected to be a certain degree of overlap between images. Regarding the effect of these partial overlaps between Figs 7C and 7D and between Figs 10A and 10B on the corresponding quantitative data in Figs 7E-G, and Figs 10D-F respectively, the first author stated that the regions selected for quantification were delineated using anatomical landmarks and standardized field-of-view coordinates using blind scoring protocols, to avoid any overlap in image analysis. They also stated that each group consisted of 20 animals (n = 20), and that for each animal, 10–15 coronal sections spanning the septal hippocampus were analyzed. They stated that the quantified values per animal were averaged across the multiple sections, and the group mean (± SEM) was then calculated across the 20 animals. Image data for all 20 subjects were not provided during post-publication discussions and the first author stated that most of the primary data underlying this article [1] are no longer available.

During editorial follow-up, a member of the PLOS One Editorial Board reviewed the above concerns and the responses received, and stated that although it is common practice in image stack processing to have a certain degree of overlap between images, the above overlap within Fig 7 and within Fig 10 in [1] comprises extensive overlapping regions of interest that includes both the CA1 and CA3 regions of the hippocampus, and that image stack processing is employed when image stitching is performed. Neither image stack processing nor image stitching are described in [1]. Given that the discrepancy between the response to the partial overlap concerns between Fig 7C Control+L-NAME and Fig 7D Control+L-NAME, and between Fig 10A CHC + Veh and Fig 10B CHC + Veh. and the methods described in [1], this issue remains unresolved.

Due to the concerns with peer review, another member of the PLOS One Editorial Board reviewed this article [1] and noted the following:

In light of the above concerns which remain unresolved in the absence of much of the underlying data and that question the reliability of the reported results, the PLOS One Editors retract this article.

SND did not respond to the final editorial decision. IB, AS, AKSG, DP, and SBS did not respond directly or could not be reached.