AUTHOR=Baumann Volker , Athanasiou Angelos-Theodoros , Faridani Omid R. , Schwerdtfeger Andreas R. , Wallner Bernard , Steinborn Ralf TITLE=Identification of extremely GC-rich micro RNAs for RT-qPCR data normalization in human plasma JOURNAL=Frontiers in Genetics VOLUME=Volume 13 - 2022 YEAR=2023 URL=https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2022.1058668 DOI=10.3389/fgene.2022.1058668 ISSN=1664-8021 ABSTRACT=Here we focussed on the contribution from the extremely GC-rich portion of the human miRNome to a panel of highly stable miRNAs to improve normalization of reverse transcription-quantitative PCR (RT-qPCR) data produced for plasma. For high-throughput selection of stable sequences, microarray technology was preferred over small-RNA sequencing (sRNA-seq) since the latter underrepresented miRNAs with a guanine-cytosine (GC) content of at least 75% (p = 0.0002, n = 2). miRNA abundances measured on the microarray were ranked for consistency and uniformity using nine normalization approaches. The eleven most stable sequences included miRNAs of moderate, but also extreme GC content (45 to 65%: miR-320d, miR-425-5p, miR-185-5p, miR-486-5p; 80 to 95%: miR-1915-3p, miR-3656-5p, miR-3665-5p, miR-3960-5p, miR-4488-5p, miR-4497 and miR-4787-5p). In contrast, the seven extremely GC-rich miRNAs were not found in the two plasma miRNomes screened by sRNA-seq. Stem-loop RT-qPCR was employed to verify stability of the selected miRNAs in comparison with two normalizers formerly recommended for this experimental condition (miR-93-5p and miR-126-3p). Templates were isolated from plasma obtained from 32 young or middle-aged healthy male adults of Caucasian decent (age range: 18 – 55 years). The coefficient of variation (CV) was used as a measure of inter-individual variance of miRNA abundance. We determined low to very low CV values (range: 0.6 to 8.2%) with miR-3665 and miR-1915-3p outperforming in stability (0.6 and 2.4 %, respectively). The eight most stable sequences represented a mixture of miRNAs with moderate or extreme GC content (moderate: miR-425-5p, miR-320d, miR-93-5p and miR-185-5p; extreme: miR-1915-3p, miR-3665, miR-4787-5p and miR-4497, respectively). Likewise, a GC-moderate together with an extreme miRNA composed the best-performing duo normalization factor (NF) for the condition of human plasma (miR-320d and miR-4787-5p). In summary, we extended the repertoire of high-quality miRNA normalizers for RT-qPCR analysis of human plasma by adding extremely GC-rich sequences. This will help to increase accuracy of miRNA expression data quantified by qPCR or digital PCR, thus raise the potential that miRNAs become markers for psychological stress reactions or early and precise diagnosis of clinical phenotypes. The novel miRNAs might also be useful for orthologous contexts considering their conservation in related animal genomes.