MicroRNAs (miRNAs) are small, non-coding RNAs that are differentially expressed throughout the body, including the endometrium and chorion during gestation. Although they are produced by individual tissues, many of these miRNAs are excreted into circulation, either within exosomes or attached to protein carrier molecules. Circulating miRNAs are surprisingly stable, with an average half-life of 5 days. These characteristics suggest that circulating miRNAs have the potential to be powerful biomarkers following the characterisation of the normal population of miRNAs during pregnancy. Additionally, evaluation techniques must be considered, as some miRNAs are extremely sensitive to varying collection and analysis techniques. To evaluate the population of miRNAs in both pregnant and nonpregnant mares, whole blood was collected from mares at 10 months of gestation (n = 3), as well as at 9 days of diestrus (n = 3 for microarray; n = 4 for qPCR). Blood was collected using PAXgene blood miRNA kits (Qiagen, Valencia, CA, USA), and then stored at –20°C until use. RNA extraction was performed following the manufacturer’s instructions. Two different analysis platforms were compared. Quantitative RT-PCR (Lightcycler 480; Roche, Indianapolis, IN, USA) was used to evaluate 346 distinct miRNAs, whereas microarray (LC Sciences, Houston, TX, USA) was used to evaluate 275 miRNAs. In total, 222 of these miRNAs were analysed on both platforms for comparison purposes. Because of the large number of miRNAs evaluated, the robust response screening analysis method (JMP ver. 11.1; SAS Institute Inc., Cary, NC, USA) was used, and multiple comparisons were based upon a false discovery rate of P < 0.1. Microarray counts were normalized by log10, whereas qPCR cycle threshold (Ct) counts were normalized by delta Ct (Ct – overall average Ct). Correlation was evaluated by the restricted maximum likelihood method (REML). Between pregnant and nonpregnant animals, qPCR analysis found 93 transcripts to be differentially expressed, whereas microarray analysis identified 15 transcripts. Of the transcripts evaluated on both platforms, 46% of transcripts identified as significant by microarray were also identified by qPCR analysis. The samples analysed by qPCR were highly correlated within themselves (r = 0.9655), as were those analysed by microarray (r = 0.8568). The microarray and qPCR platforms showed some degree of correlation between the platforms (r = –0.6980); this correlation was negative because of the different measurement methods between microarray and qPCR. These results indicate there is a correlation between the qPCR and microarray platforms; however, it may be unrealistic to expect consistently complementary results between the platforms. Overall, qRT-PCR provided more consistent results and appears to be a more powerful analysis tool. Additionally, these data suggest that there is a distinct population of circulating miRNAs of mares at 10 months gestation compared with nonpregnant individuals. These differences should be further explored to characterise the normal miRNA populations of healthy mares throughout gestation, as well as during various disease states.
Read full abstract