Real-time quantitative PCR and droplet digital PCR for plant miRNAs in mammalian blood provide little evidence for general uptake of dietary miRNAs: limited evidence for general uptake of dietary plant xenomiRs

Abstract

Evidence that exogenous dietary miRNAs enter the bloodstream and tissues of ingesting animals has been accompanied by an indication that at least one plant miRNA, miR168, participates in "cross-kingdom" regulation of a mammalian transcript. If confirmed, these findings would support investigation of miRNA-based dietary interventions in disease. Here, blood was obtained pre- and post-prandially (1, 4, 12 h) from pigtailed macaques that received a miRNA-rich plant-based substance. Plant and endogenous miRNAs were measured by RT-qPCR. Although low-level amplification was observed for some plant miRNA assays, amplification was variable and possibly non-specific, as suggested by droplet digital PCR. A consistent response to dietary intake was not observed. While our results do not support general and consistent uptake of dietary plant miRNAs, additional studies are needed to establish whether or not plant or animal xenomiRs are transferred across the gut in sufficient quantity to regulate endogenous genes.

Keywords: RT-qPCR; diet; digital PCR; exosome; extracellular vesicle; microRNA; nutrition; plant; xenomiR.

Figures

Figure 1. miRNA abundance and amplification efficiency in a plant-based dietary substance and in mammalian plasma. (A) Relative abundance of several highly conserved plant miRNAs in a soy- and fruit substance. RNA was purified from 100 µl “Silk” using the mirVana total RNA isolation protocol (Ambion/Life Technologies). Hydrolysis probe-based RT-qPCR measurements were made from up to three independent RNA isolations. Fold abundance was calculated relative to miR160. Similar results were obtained at different input RNA dilutions (not shown). Error bars represent standard deviation. (B and C) RT-qPCR assay linearity and amplification efficiency was assessed (shown here: miR160) using 2-fold serial dilutions of cDNA prepared with mature miRNA-specific stem-loop primers and plant RNA from the dietary substance (B) or RNA from pre- or post-prandial plasma samples (C). For plant RNA, (B) starting dilution (#1) corresponded to cDNA from approximately 6.3 nanoliters of the original substance. Final dilution (#6) was from about 200 picoliters original volume. Amplification occurred in all wells, and amplification efficiency was approximately 100%. Error bars are standard deviation of triplicate measurements (less than half a cycle for all dilutions). For plasma, dilutions corresponded to material from 2 µl plasma (#1) down to 60 nl plasma (#6). Standard deviation corresponded to half a cycle or greater for 8 of 12 samples, and several wells within technical replicate groups did not amplify. Amplification efficiencies were 94% (plasma 1) and 65% (plasma 2).

Figure 2. Plant miRNAs in circulation before and following intake of a miRNA-rich food source. (A) Apparent abundance in plasma of miR166 relative to miR160, as assessed by RT-qPCR with RNA samples purified by mirVana technique, did not match relative abundance in the food source at any time point pre- or post-prandial (compare with Fig. 1A) in either of two macaque subjects designated “67X” and “HL26.” (B and C) Plasma miR160 and miR166 Cq values and standard deviations by time point and subject (67X and HL26). Missing standard deviation bars are due to insufficient numbers of amplifying wells. Note that, to the extent it could be estimated, direction of apparent regulation often differed between the two subjects.

Figure 3. RT-qPCR results for RNA isolated with a high-performance biofluids-specific protocol (Exiqon). Cq and standard deviation of RT-qPCR results for five plant miRNAs and one exogenous spike-in RNA are shown for plasma from subjects 67X (left) and HL26 (right) obtained at the pre-prandial (0) time point and at 1, 4 and 12 h post-prandial. Asterisk indicates a time point for which no reactions amplified for that miRNA. In the bottom panels, highly consistent recovery is shown for a miRNA (cel-miR-39) that was spiked in to the RNA isolation reactions to assess variability in the isolation technique that could have affected results.

Figure 4. Droplet digital PCR indicates specific amplification of plant miRNA from plant source but not from pre- or postprandial animal plasma. (A and B) Intensity plots of DX100 QuantaSoft results for miR167 and miR168: for RNA from the dietary plant material (left), positive droplets clustered at high and relatively consistent intensity (specific amplification). Droplets from four hour postprandial plasma samples (right) had low and varying intensity, especially for miR168, consistent with inefficient or spurious amplification. Triplicate reactions are shown. Note that the intensity scale is not necessarily equivalent between different plots. (C) miR160 showed the highest level of amplification from plasma for miRNAs we examined by RT-qPCR (see Fig. 2), but by ddPCR with pre- and post-prandial plasma samples, there was a wide range of droplet intensities. From plant material, droplet intensities clustered consistently as for the miRNAs in (A and B) (not shown). The horizontal line is an example of where a threshold might be drawn. (D) Low-level non-specific amplification for plant miRs166, 167 and 168 from one no-template control reaction each (water). (E) Copy number counts and confidence interval minimum and maximum for miR160 per microliter of reaction volume over a 1:4 dilution series of material corresponding to cDNA from input into a reverse transcription reaction that corresponded to RNA obtained from 6.25 nl (dilution 1), 1.6 nl (dilution 2) and 400 pl (dilution 3) of the “Silk” plant-based substance. (F) Counts and confidence intervals for undiluted material from plasma RNA isolations for the pre-intake and 4 h postprandial time points for miR160 for subjects 67X and HL26.