Year: 2018
Type: article
Abstract: To the Editor: MicroRNAs (miRNAs)1 are small noncoding RNAs that regulate the expression of target genes, thereby contributing to disease development. The discovery of their presence in the bloodstream has stimulated extensive characterization of their potential as disease biomarkers. Reverse transcription followed by quantitative polymerase chain reaction (RT-qPCR) is the most widely used technique to quantify the expression of miRNAs, both in body fluids and tissue samples. In an ongoing study, we aimed to address the biomarker potential of circulating miRNAs for thoracic aortic aneurysm (TAA) using RT-qPCR. Surprisingly, a complete inhibition of RT-qPCR prevented the measurement of miRNAs in serum samples from many patients. We first hypothesized that this inhibition might be caused by exogenous heparin, which is well known to interfere with RT-qPCR assessment of miRNAs (1). However, the patients were not receiving heparin therapy and blood samples were not collected in tubes containing heparin. In addition, TAA is not known to modify endogenous heparin level, in contrast with hemorrhagic shock, infection, hypothermia, or organ transplantation (2). To check whether endogenous heparin was responsible for the RT-qPCR inhibition, we treated 5 serum samples from TAA patients with heparinase I and observed a recovery of the RT-qPCR signal for cel-miR-39 and miR-451a (not shown). The rationale for measuring these 2 miRNAs is that the former is the most widely used spike-in control and miR-451a is highly expressed in blood. Of note, the average concentration of heparin in the 5 samples was mean (SD) 0.17 (0.05) IU/mL (COATEST heparin assay; Chromogenix), which is within the physiological range of concentrations of heparin reported in the blood of healthy individuals: 0.10–0.24 IU/mL (3). Together, these elements suggested that not only exogenous but also endogenous heparin is able to inhibit RT-qPCR and prevent detection of miRNAs. However, we cannot exclude the possibility that other factors present in blood (e.g., bile salt) may also contribute to this inhibition. To further test this possibility, we used plasma samples collected in citrate tubes from 3 healthy donors. These samples were depleted of endogenous heparin by an incubation at room temperature, with heparin being rapidly degraded and neutralized when not kept on ice. The depletion was checked by measuring heparin levels in the samples, which were below 0.05 IU/mL (detection limit of the COATEST heparin assay; Chromogenix). Then, increasing concentrations of exogenous heparin (from 0.002 IU/mL to 2 IU/mL; B. Braun) were added to the 3 heparin-depleted plasma samples. Cel-miR-39 was spiked-in, RNA was extracted and treated with heparinase, and expression levels of cel-miR-39 and miR-451a were assessed by RT-qPCR. As shown in Fig. 1, a concentration of 0.02 IU/mL of heparin partially inhibited the detection of cel-miR-39 and did not affect the detection of miR-451a. This observation is consistent with previous data showing that spike-in controls are more sensitive to heparin than endogenous miRNAs, casting some doubts about the reliability of normalizing miRNA measurements with spike-in controls (1). Concentrations as low as 0.1 IU/mL prevented the detection of both miRNAs, an effect fully reverted by heparinase treatment for cel-miR-39 and only partly for miR-451a (Fig. 1). Similar results were obtained in serum samples and for other miRNAs (e.g., miR-21–5p and miR-126–3p). These data show that physiological concentrations of endogenous heparin may prevent RT-qPCR assessment of circulating miRNAs. 200 μL of plasma samples collected in citrate tubes from 3 healthy donors were depleted of endogenous heparin before addition of exogenous heparin. RNA was extracted as described (4) and treated with 0.89 U heparinase I or vehicle for 1 h at 25 °C (5) before RT-qPCR for cel-miR-39 and miR-451a using the miScript system (Qiagen) (4). Expression levels were calculated using the 2-Cq formula. Gray zone: healthy range of endogenous levels of heparin. Values are mean (SD). * P < 0.05 vs no heparin; # P < 0.05 vs heparinase vehicle (2-tailed t-test). We report here important limitations of the use of RT-qPCR for the assessment of circulating levels of miRNAs. Heparin is widely used in the hospital for its anticoagulant activity, both in patients and in blood collection tubes for plasma preparation. Inhibition of RT-qPCR by high levels of exogenous heparin was widely known, and we now show that even low levels of endogenous heparin (within physiological range) interfere with the measurement of miRNAs by RT-qPCR. As shown for miR-451a (Fig. 1), heparinase treatment of RNA samples may not be able to completely prevent the inhibition of RT-qPCR by heparin. Thus, heparinase treatment of RNA samples may add a supplementary bias in the quantification of miRNAs and should not be recommended. Instead, in addition to avoiding the use of heparin for blood collection, we recommend that heparin concentration should be measured before miRNA measurement by RT-qPCR and only samples with <0.02 IU/mL of heparin should be processed. Alternatively, technologies without an amplification step, such as those using fluorescence probes, may be favored. Finally, our recommendation can be extrapolated to any other quantification of gene expression levels by RT-qPCR in samples potentially containing even low levels of heparin. microRNA reverse transcription-quantitative polymerase chain reaction thoracic aortic aneurysm.
Source: Clinical Chemistry
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: bronze
Awards: G4465755651, G7990581908, G9000001686
OpenAlex ID: W2791582313