U6 as a microRNA normalizer in serum of patients with hepatocellular carcinoma

To the Editor,

Selection of an appropriate internal control is required to identify true changes in microRNA expression and has a crucial impact on the interpretation of data. Shen et al. studied serum miR-574-3p as a diagnostic biomarker for hepatocellular carcinoma (HCC). ¹ They found that its expression was significantly higher in patients with HCC compared with cirrhosis patients and healthy controls. They normalized cycle threshold (Ct) values using U6 as an internal control. Although U6 is a suitable normalizer for use in tissues, it is not a reliable internal control in serum, as shown in the studies described below.

In up to eight cycles of quantitative polymerase chain reaction analysis from serum of 44 healthy donors, Benz et al. found high interindividual variability of serum U6 levels compared with SV40 spiked-in RNA. ² They made a similar finding in the serum of patients with liver fibrosis, and U6 concentrations were lower in the serum of patients with liver fibrosis compared with healthy controls. ² Xiang et al. evaluated the expression of U6 in 30 healthy controls and 80 patients with different kinds of cancer, and also demonstrated large fluctuations in U6 concentrations; ΔCt – the difference between the highest and the lowest expression level – was 3.29 for U6. ³ Tang et al. used four statistical methods for selecting reference genes (GeNorm, Normfinder, BestKeeper and the comparative ΔCt method) and found that U6 was the least stable reference gene in patients with HCC and cirrhosis. ⁴ Finally, U6 expression is susceptible to freeze–thaw cycles. ³

We conclude that U6 is not suitable for serum microRNA normalization. In the absence of a reliable endogenous microRNA that can be used as a reference gene in serum, recent studies have used exogenous cel-miR-39, derived from Caenorhabditis elegans, for microRNA normalization. ⁵