TY - JOUR

T1 - Survey of 800+ data sets from human tissue and body fluid reveals xenomiRs are likely artifacts

AU - Kang, Wenjing

AU - Bang-Berthelsen, Claus Heiner

AU - Holm, Anja

AU - Houben, Anna J.S.

AU - Müller, Anne Holt

AU - Thymann, Thomas

AU - Pociot, Flemming

AU - Estivill, Xavier

AU - Friedländer, Marc R.

PY - 2017

Y1 - 2017

N2 - miRNAs are small 22-nucleotide RNAs that can post-transcriptionally regulate gene expression. It has been proposed that dietary plant miRNAs can enter the human bloodstream and regulate host transcripts; however, these findings have been widely disputed. We here conduct the first comprehensive meta-study in the field, surveying the presence and abundances of cross-species miRNAs (xenomiRs) in 824 sequencing data sets from various human tissues and body fluids. We find that xenomiRs are commonly present in tissues (17%) and body fluids (69%); however, the abundances are low, comprising 0.001% of host human miRNA counts. Further, we do not detect a significant enrichment of xenomiRs in sequencing data originating from tissues and body fluids that are exposed to dietary intake (such as liver). Likewise, there is no significant depletion of xenomiRs in tissues and body fluids that are relatively separated from the main bloodstream (such as brain and cerebro-spinal fluids). Interestingly, the majority (81%) of body fluid xenomiRs stem from rodents, which are a rare human dietary contribution but common laboratory animals. Body fluid samples from the same studies tend to group together when clustered by xenomiR compositions, suggesting technical batch effects. Last, we performed carefully designed and controlled animal feeding studies, in which we detected no transfer of plant miRNAs into rat blood, or bovine milk sequences into piglet blood. In summary, our comprehensive computational and experimental results indicate that xenomiRs originate from technical artifacts rather than dietary intake.

AB - miRNAs are small 22-nucleotide RNAs that can post-transcriptionally regulate gene expression. It has been proposed that dietary plant miRNAs can enter the human bloodstream and regulate host transcripts; however, these findings have been widely disputed. We here conduct the first comprehensive meta-study in the field, surveying the presence and abundances of cross-species miRNAs (xenomiRs) in 824 sequencing data sets from various human tissues and body fluids. We find that xenomiRs are commonly present in tissues (17%) and body fluids (69%); however, the abundances are low, comprising 0.001% of host human miRNA counts. Further, we do not detect a significant enrichment of xenomiRs in sequencing data originating from tissues and body fluids that are exposed to dietary intake (such as liver). Likewise, there is no significant depletion of xenomiRs in tissues and body fluids that are relatively separated from the main bloodstream (such as brain and cerebro-spinal fluids). Interestingly, the majority (81%) of body fluid xenomiRs stem from rodents, which are a rare human dietary contribution but common laboratory animals. Body fluid samples from the same studies tend to group together when clustered by xenomiR compositions, suggesting technical batch effects. Last, we performed carefully designed and controlled animal feeding studies, in which we detected no transfer of plant miRNAs into rat blood, or bovine milk sequences into piglet blood. In summary, our comprehensive computational and experimental results indicate that xenomiRs originate from technical artifacts rather than dietary intake.

KW - Artifacts

KW - Contamination

KW - Exogenous miRNAs

KW - Meta-study

KW - MicroRNA

KW - MiRNA

KW - Next-generation sequencing

U2 - 10.1261/rna.059725.116

DO - 10.1261/rna.059725.116

M3 - Review

C2 - 28062594

AN - SCOPUS:85016240833

VL - 23

SP - 433

EP - 445

JO - RNA

JF - RNA

SN - 1355-8382

IS - 4

ER -