On 2015 Nov 19, Kenneth Witwer commented:

Shu et al have conducted a fascinating analysis of miRNA sequences. It is unclear, however, how this analysis relates to dietary, circulating, and "transportable" categories of RNA that are hypothesized to be absorbed from the diet in functional form. Although the stated intention was "to heavily rely on experimental data to identify features that can differentiate secreted miRNAs from the rest," the data in question are either unreliable (the circulating miRNA data is questionable, based on only one biological and technical measurement), incompletely described (assignment of "exosomal" status based on the two cited databases is unclear) or missing (the pivotal milk uptake experiment--unless I'm missing something). Thus, the practical validity of the sequence analysis cannot be assessed.

Endogenous miRNAs are classified by Shu et al as circulating or not based on a list from Weber JA, 2010. This preliminary publication reported only one qPCR threshold cycle measurement for each of several hundred miRNAs using only one sample of pooled plasma. Other issues, such as a lack of correlation with results of other studies and a failure to detect abundant plasma miRNAs, such as miR-16 and miR-223, were previously noted in Watson AK, 2012. Thus, the "circulating" classification made by the authors is not supported by reliable data. Perhaps the authors might wish to revisit their study with a more comprehensive ranking of plasma miRNAs supported by reliable public sequencing and microarray data.

Which miRNAs are packaged into extracellular vesicles is an even more complicated question than simple presence in circulation. Since the majority of miRNAs in circulation appear to be in free protein complexes, not EVs, contaminants of EV preparations have strong potential to skew experimental results. It would be helpful if the authors could clarify how they used the EVpedia and ExoCarta databases to identify EV-packaged miRNAs and how this information (presumably including abundance ranks?) was used in the study.

Also unclear was where to obtain the sequencing data from the described milk feeding experiments. Although all data were said to be found on a university website, I could not find the sequencing data there or elsewhere. A public link to these data and further clarification of how they were used to validate the findings would be very helpful, as well as consistent with journal guidelines. Perhaps I missed this link?

I would note that the evidence in support of the dietary miRNA transfer hypothesis described as "unambiguous" consists of a study by the authors. The results of this study have not been confirmed. Alternative hypotheses (Witwer KW, 2014) were omitted, as well as published evidence that contradicts the hypothesis, most strikingly a recent study (Title AC, 2015) in which no miRNA uptake was observed from milk in miR-200c and miR-375 knockout mouse pups.

In conclusion, the sequencing analysis looks quite interesting, but the underlying assumptions are debatable at best.

On 2015 Nov 19, Kenneth Witwer commented:

Shu et al have conducted a fascinating analysis of miRNA sequences. It is unclear, however, how this analysis relates to dietary, circulating, and "transportable" categories of RNA that are hypothesized to be absorbed from the diet in functional form. Although the stated intention was "to heavily rely on experimental data to identify features that can differentiate secreted miRNAs from the rest," the data in question are either unreliable (the circulating miRNA data is questionable, based on only one biological and technical measurement), incompletely described (assignment of "exosomal" status based on the two cited databases is unclear) or missing (the pivotal milk uptake experiment--unless I'm missing something). Thus, the practical validity of the sequence analysis cannot be assessed.

Endogenous miRNAs are classified by Shu et al as circulating or not based on a list from Weber JA, 2010. This preliminary publication reported only one qPCR threshold cycle measurement for each of several hundred miRNAs using only one sample of pooled plasma. Other issues, such as a lack of correlation with results of other studies and a failure to detect abundant plasma miRNAs, such as miR-16 and miR-223, were previously noted in Watson AK, 2012. Thus, the "circulating" classification made by the authors is not supported by reliable data. Perhaps the authors might wish to revisit their study with a more comprehensive ranking of plasma miRNAs supported by reliable public sequencing and microarray data.

Which miRNAs are packaged into extracellular vesicles is an even more complicated question than simple presence in circulation. Since the majority of miRNAs in circulation appear to be in free protein complexes, not EVs, contaminants of EV preparations have strong potential to skew experimental results. It would be helpful if the authors could clarify how they used the EVpedia and ExoCarta databases to identify EV-packaged miRNAs and how this information (presumably including abundance ranks?) was used in the study.

Also unclear was where to obtain the sequencing data from the described milk feeding experiments. Although all data were said to be found on a university website, I could not find the sequencing data there or elsewhere. A public link to these data and further clarification of how they were used to validate the findings would be very helpful, as well as consistent with journal guidelines. Perhaps I missed this link?

I would note that the evidence in support of the dietary miRNA transfer hypothesis described as "unambiguous" consists of a study by the authors. The results of this study have not been confirmed. Alternative hypotheses (Witwer KW, 2014) were omitted, as well as published evidence that contradicts the hypothesis, most strikingly a recent study (Title AC, 2015) in which no miRNA uptake was observed from milk in miR-200c and miR-375 knockout mouse pups.

In conclusion, the sequencing analysis looks quite interesting, but the underlying assumptions are debatable at best.