FUT2 Secretor Status Influences Susceptibility to VP4 Strain-Specific Rotavirus Infections in South African Children

Round 1

Reviewer 1 Report

This manuscript presents the hypothesis that FUT2 mutation leading to non-section of HBGA in the gut may confer resistance to the rotavirus vaccine strains, leading to a lower rate of vaccine efficacy in non-secretors. A single experiment is presented; genotyping of FUT2 for rotavirus cases and controls using qPCR with sanger sequencing for confirmation as follow up on about 10% of subjects.

This manuscript is well written and scientifically sound, but it fails to present a substantial amount of new information. It proposes a link to vaccine efficacy, which although interesting, was not explored experimentally (authors state serum was not available to determine vaccine status). The main conclusion that non-secretors are of higher prevalence in African populations and are resistant to rotavirus P[8] and P[4] infection, leading to the more frequent observation of P[6] infection in these individuals, seems to confirm previously published data (see references 10-12, 16, 17).

A novel a mutation associated with erroneous results from qPCR, detected by sanger sequencing, is reported but there is not follow up to identify better qPCR primers or to expand the Sanger sequencing to determine the prevalence of the FUT2-G514R mutation.

The statistical analysis in the methods section needs to address which statistical tests were specifically used.

There is not enough new information (data) in this manuscript to constitute an article. It may be acceptable as a short communication. If so, a more concise and targeted introduction and conclusion would be better suited.
Sheet 2 of the supplemental material is a blank calendar page that can be deleted from the file.

Please ensure that the BankIt accession numbers are added (line 242) before submitting a final copy of the manuscript.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This study by MacDonald et al., investigates the role of secretor status in susceptibility to rotavirus infection in a genotype dependent manner. Although this has been investigated in previous studies, host genetic susceptibility to rotavirus is a relatively new field with several studies reporting discrepant results, and as such the study contributes to advance the field. In general, it is a clear and well written manuscript.

There are however, some issues that should be addressed.

The authors discuss, both in the introduction and discussion sections, that the frequency of non-secretors in different populations might influence the circulation and epidemiology of rotavirus genotypes, and could be a reason for high prevalence of P[6] in South Africa, or elsewhere in Africa. However, also Europe and North America have a non-secretor prevalence of 20-25%, but almost no P[6] infections. Several studies have shown P[6] susceptibility to mainly be driven by the Lewis-negative pheno/genotype, independent of secretor status, and the Lewis-negative geno/phenotype is generally higher in sub-Saharan Africa. This should be discussed, particularly as the Lewis genotype was not investigated. Likewise, P[8] infections has also in some studies been associated with both secretor and Lewis positivity, which could also be discussed.

Methodology.

How were the 10% samples selected for Sanger sequencing and what were the genotype distribution (as per real-time PCR?)

The authors found discrepant results between real-time PCR and Sanger sequencing in 5 samples , all that had been classified as non-secretors using the real-time PCR assay, corresponding to 10% of the samples that were tested with both assays. However, as I understood the discrepancies were only found in the non-secretor sub-group. Thus, how many non-secretors (as per real-time PCR) were Sanger sequenced?, and what is the associated error rate in this sub-group? (5/x). This could influence interpretation of the result. If high, authors should consider sequencing of more non-secretors (as per real-time PCR), particularly since quite many non-secretors were infected with P[8] and P[4] compared to other studies.

Conclusions:

“Non-secretors……. appeared natural resistant…” 

The word “resistant” seems too strong, since many non-secretors were also infected.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Previous studies showed that children of secretor phenotype of HBGA, controlled by the human FUT2 gene, are more likely to be infected by rotavirus and develop diarrhea. The authors of this manuscript examined the association between the rotavirus infection-related hospitalizations and HBGA phenotypes in children from South Africa. The results further confirmed that children of the secretor phenotype of HBGA are more susceptible to the infection of rotavirus P[4] and P[8]. The experiments were carefully designed and carried out. The results provided additional knowledge in broader populations and wider geographical areas.

The TaqMan SNP genotyping assay was used to determine the presence of FUT2 G428A. However, the DNA sequencing result showed that there is about 10% misclassification of secretors and nonsecretors. Line 157 to 160, the authors claimed that the misclassification is potentially due to a mutation noted ~50 bp upstream of the SNP position, and hypothesized that misclassification by real-time PCR was influenced by sub-optimal primer binding.  Based on the mechanism of the assay, probes are the key to the typing of SNP. Therefore, both claims don’t make sense. Please analyze the primers and probes sequence carefully and revise the discussion accordingly.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf