All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
I have checked all the revisions and can confirm the authors have addressed the reviewers' comments. This manuscript is now ready for publication.
Please address all the reviewers' suggestions for revision. In particular, it would be great if older references in the Introduction can be replaced with newer ones as suggested by Reviewer 2, and consider performing imputation to infer additional SNPs that can provide additional information, and consider including an X chromosome study as suggested by Reviewer 3.
Further, consider adding a sentence as a potential limitation - that this work only found new variants and not existing known ones - which is a concern for reproducibility of findings, and speculate why this happened.
I have checked the primers in Primer Blast and can confirm the primers and targets are correct.
**PeerJ Staff Note:** It is PeerJ policy that additional references suggested during the peer-review process should only be included if the authors agree that they are relevant and useful.
The manuscript "Genome-wide association study identifes GAK and KLF12 associated with curve severity of adolescent idiopathic scoliosis" by Lai et al. describes a genome-wide association study performed to look at curve severity in AIS patients.
The current manuscript describes a re-analysis of a subset of data from the full 2015 AIS GWAS by Zhui et al., PMID 26394188. This study included 323 patients with severe curves (> 40°) and 297 with mild curves (< 30°). The top GWAS hits were followed up in anindependent cohort of 634 severe and 546 mild cases.
The current study is a very interesting one, and looking at something that is very relevant to optimising treatment of AIS. The manuscript is well-written, in professional English.
Having read this manuscript and also the 2015 paper by Zhui et al., I feel confident in eg. the QC of the genotype data and the robustness of the baseline GWAS dataset. The study cites the relevant papers, and uses standard bioinfo-tools such as Plink and databases such as ENCODE.
Minor comment:
The link for the summarised GWAS data is written as 10.6084/m9.figshare.30022192 in the manuscript but should be changed to https://doi.org/10.6084/m9.figshare.30022192 to make it easier to access directly.
Minor comment:
Here, the authors compared severe vs mild curves, it would be interesting to see how the data would look if the curve was treated as a quantitative trait. (This is just a suggestion).
Minor comment:
The RNA expresson follow-up of candidate genes is a nice touch, and seems to show very clearly that slow vs fast-twitch muscle fibers could be an interesting thing to dig into deeper.
Not being able to correlate the expression of a neuronally expressed gene with genotype, in a muscle tissue, should probably be taken as a sign of robustness instead of a failure. Looking at GAK in a neuronal tissue would indeed be very interesting.
The methods and study design used are appropriate to the data and very much standard ones. While building on a previous GWAS, this current study stands on its own as the phenotype in question is clear and highly relevant.
Minor comment:
It might be useful to add a short comment on how these two new candidate genes (GAK should still be considered as interesting, despite the RNA data) may connect to previously described AIS candidates, to put the new data into context.
Minor comment:
It would also be useful to mention how the candidate genes/SNPs looked in the full GWAS study of AIS vs controls.
Some references a quite old, I would suggest to change them with newer ones.
There are new studies reporting on natural history and risk factors published after 2020. I think it would be interesting to add then more than the older one cited.
The main point is the patient selection. We miss too many relevant information. Ideally, for the purpose of the papers, the authors should have selected only untreated patients. But it's my understanding that patients were treated. So, if this is the case, you cannot comment of risk of progression without considering the impact of the treatment and it's quality (brace, exercise, compliance, dosage, etc).
Moreover, a sample of scoliotic patients with mild curves (less than 20 or better less than 15°) should serve as a control group. It's well known from the "twins study" that it's not just genetics, there is much more. So we need a control group
not possible until more data are provided
The manuscript is clearly written and has included some references to existing publications. However, there is a very extensive literature at this point on idiopathic scoliosis that is inadequately cited. There are fewer manuscripts that pursue analysis of severity of scoliosis and therefore the manuscript is making an important contribution. I found this recent review article helpful as I framed my review (and it is open source)
Genetics and pathogenesis of scoliosis
Petrosyan, Edgar et al.
North American Spine Society Journal (NASSJ), Volume 20, 100556
The study has an adequate design. The sample size is somewhat limited but the authors were able to study a unique resource of spinal samples from surgeries they have performed to evaluate the effect of variants that they identified on expression of genes. With less dense SNP arrays such as the Affymetrix 6.0 array that they used, the standard in the field is to use imputation to infer additional SNPs that can provide additional information about the SNPs in a region that may be relevant to disease severity. The authors can either use the Michigan imputation server to perform imputation, download the 1000 genomes data resource, or use imputation using reference panels that have been established for Chinese populations such as ChinaMap or Westlake BioBank for Chinese (WBBC). I expect that results from imputation will not dramatically change findings but should provide a more complete assessment of regional variants influencing progression of scoliosis. QC steps are a little liberal in retaining SNPs, but acceptable. The X chromosome needs to be included in the analysis if at all possible, especially since the study is only conducted in women and therefore there is less of a need to employ specialized approaches to analysis, which occur when studies are performed on males and females jointly.
The results seem reasonable and the validation analysis helps to support findings as do functional studies that were conducted. That said, the lack of citation to existing literature limits the interpretation. It would be useful to provide a supplementary table showing what results were found for any currently known genes that influence severity of scoliosis (for example). The study only found new variants and did not replicate any known variants, which raises a bit of concern about the reproducibility of its findings.
The authors focus on identifying variants associated with progression to a severe curvature, which needs more invasive management in AIS patients. This addresses the need to examine broader populations (outside of Europeans). In addition, most other GWAS studies have focused on susceptibility to AIS rather than curve progression.
Overall, the report is well written and described.
Report ethnicity information? I can’t seem to find where that is clearly stated.
Reporting is sound, with only the replicated variants further discussed.
The manuscript could be improved with some additional figures. I suggest including a genome browser view for these two variants to display their location within the gene body, and consider adding a conservation track and epigenetic data (provided as a table) to visually represent this information. Including a motif image that shows the base changed with the variant would also be helpful to illustrate the impact of these changes clearly.
Nicely demonstrated KLF12 expression to Cobb angle (negative correlation), which led to a new hypothesis that this is linked to muscle fibre. This was then tested by correlating markers of slow/fast-twitch fibres to KLF12 expression.
Rt-qPCR – should include either a citation for using GAPDH as the reference gene (ie previously used and shown to be reliable, consistent expression in paraspinal muscle) or a second reference gene can be used and normalise gene expression to both.
The discussion could be improved by noting that other components of the spine, such as the nervous system and intervertebral discs (IVD), can contribute to scoliosis progression. Other hypotheses should be noted, as muscle is composed of many cell types. Suggest finding additional support by exploring some of the available human skeletal scRNAseq muscle atlases – I had a quick look, and KLF12 is expressed in multiple cell clusters, including fibroblasts and immune cell types. The same is true for GAK, which peaks in macrophages and B/T/NK cells.
Minor
Gene names should be written out in full in the abstract
Figure 1 -the text is a bit hard to read
Missing spacing between the end of the sentence and the citations.
Methods: Missing citations for the software packages used
Lines 222, 223 – remove the paragraph break
There are a few errors in the reference list, mostly additional spaces
Line 271 scoliosis during pubertal grow th. Spine 31:1933-1942.
Line 219 idiopathic s coliosis.
Line 317 Progression in Adolescent Idiopathic Scoliosi s.
Line 329 Idiopathic Scol iosis.
Line 349 characterization in a l arge heterogeneous cohort
Line 407 human lung fibroblasts i s associate
Line 418 identifies novel susceptible loci and hi ghlights Wnt/
Line 414 new susceptibility loci for a dolescent idiopathic scoliosis
This is the original reference showing that KLF12 (AP-2rep) binds to AP-2alpha
Imhof A, Schuierer M, Werner O, Moser M, Roth C, Bauer R, Buettner R. Transcriptional regulation of the AP-2alpha promoter by BTEB-1 and AP-2rep, a novel wt-1/egr-related zinc finger repressor. Mol Cell Biol. 1999 Jan;19(1):194-204. doi: 10.1128/MCB.19.1.194.
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.