Variance component analysis of circulating miR-122 in serum from healthy human volunteers.
Author(s): Vogt J, Sheinson D, Katavolos P, Irimagawa H, Tseng M, Alatsis KR, Proctor WR
Publication: PLoS One, 2019, Vol. 14, Page e0220406
PubMed ID: 31348817 PubMed Review Paper? No
Purpose of Paper
This paper investigated inter and intra-donor variability in microRNA (miRNA, miR)-122 levels and determined the contribution of donor gender and race, serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels, and normalization method to the variability. Variability in miR-122 levels was also compared with that of other miRNAs.
Conclusion of Paper
miR-122 levels were found to be highly variable and much more variable than the other miRNAs in the panel. Variability of miR-122 levels and that of the next five most variable miRNAs was not affected by normalization method (mean of five miRNAs versus cel-miR-39). Half the variability in miR-122 and most other miRNAs in the panel was attributed to inter-donor variability and the remaining was intra-donor variability. miR-122 levels were comparable in males and females but were 3 to 4-fold more variable in serum from subjects who identified as non-Caucasian than those that identified as Caucasian (50% of subjects). The only other miRNA that demonstrated a 3-fold difference in variability between serum from subjects who identified as non-Caucasian and those that identified as Caucasian was miR-483-5p. miR-122 levels were modestly correlated with ALT and AST. While 80% of the variability in ALT was attributed to inter individual variability, only 50% of the variability in AST was inter-individual variability.
Studies
-
Study Purpose
The purpose of this study was to investigate inter and intra-donor variability in miR-122 levels and to determine the contribution of donor gender and race, serum ALT and AST levels, and normalization method to the variability. Variability in miR-122 levels was also compared with that of other miRNAs. Whole blood was collected six times from 40 healthy fasted employees (19 females and 23 males aged 29-62 years). All blood was collected into serum separator tubes between 7 and 10 AM at one or two week intervals (six specimens over 6-10 weeks). Serum was obtained by centrifugation at 2000 x g for 10 min at 4˚C. Serum was then aliquoted into RNAse-free tubes, frozen on dry ice, and then stored at -80˚C for up to 6 months. Serum aliquots were sent to a commercial diagnostics laboratory for standard clinical chemistry analysis including AST and ALT using an autoanalyzer to verify that all patients were healthy. RNA was extracted using the miRNeasy Serum/Plasma Kit with a QiaCube and frozen at -80 ˚C until analysis. Levels of 80 miRNAs including miR-122, cel-miR-39, and 80 other miRNAs were quantified by TaqMan real-time RT-PCR using Fluidigm 96x96 Dynamic Array Integrated Fluidic chips with preamplification.
Summary of Findings:
The MiRA-norm algorithm identified miR-let7b, miR-17-5p, miR-19b-5p, miR-20a-5p, and miR-20b as the most appropriate normalizers; therefore, the mean of these five miRNAs was used for normalization. There was a high degree of variability with 117-fold differences between the lower and upper bounds when normalized using the mean of the five miRNAs and 111-fold differences when normalized to cel-miR-39. Half the variability was attributed to inter-donor variability and the remaining was intra-donor variability, regardless of normalization method. This was also true for most of the miRNAs in the panel. The remaining 36 miRNAs measured in at least 95% of specimens varied only 2 to 45-fold when normalized to the mean of the five miRNAs and 6 to 62-fold when normalized to cel-miR-39. Importantly, the next five most variable miRNAs (miR-133a, miR29a, miR-375, miR-335-5p, and miR-10a) are most variable regardless of the normalization method.
miR-122 levels were comparable in males and females but were 3 to 4-fold more variable in serum from subjects who identified as non-Caucasian (identified as 18% Latino/Hispanic, 23% Asian, 2% Middle Eastern, 2% Black, 2% Native Peoples, and 2% Other) than those that identified as Caucasian (50% of subjects). The only other miRNA that demonstrated a 3-fold difference in variability between serum from subjects who identified as non-Caucasian and those that identified as Caucasian was miR-483-5p. miR-122 levels were modestly correlated with ALT and AST when normalized to the mean of the five miRNAs (r=0.544, and r=0.454 respectively) or cel-miR-39 (r=0.516 and r=0.472). While 80% of the variability in ALT was attributed to inter individual variability, only 50% of the variability in AST was inter-individual variability.
Biospecimens
Preservative Types
- Frozen
Diagnoses:
- Normal
Platform:
Analyte Technology Platform Protein Clinical chemistry/auto analyzer RNA Real-time qRT-PCR RNA Low density array Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Preaquisition Patient race Caucasian
Non-Caucasian (Identified (Latino/Hispanic, Asian, Middle Eastern, Black, Native Peoples, and Other)
Biospecimen Acquisition Time of biospecimen collection Collected at 1-2 week intervals over 6-10 weeks
Preaquisition Patient gender Male
Female
Real-time qRT-PCR Specific Data handling Normalized to mean of miR-let7b, miR-17-5p, miR-19b-5p, miR-20a-5p, and miR-20b
Normalized to cel-miR-39
Real-time qRT-PCR Specific Targeted nucleic acid miR-122
80 miRNA