Development of a Singleplex Real-Time Reverse Transcriptase PCR Assay for Pan-Dengue Virus Detection and Quantification

Abstract

Dengue virus (DENV) infection is a significant global health problem. There are no specific therapeutics or widely available vaccines. Early diagnosis is critical for patient management. Viral RNA detection by multiplex RT-PCR using multiple pairs of primers/probes allowing the simultaneous detection of all four DENV serotypes is commonly used. However, increasing the number of primers in the RT-PCR reaction reduces the sensitivity of detection due to the increased possibility of primer dimer formation. Here, a one tube, singleplex real-time RT-PCR specific to DENV 3'-UTR was developed for the detection and quantification of pan-DENV with no cross reactivity to other flaviviruses. The sensitivity of DENV detection was as high as 96.9% in clinical specimens collected at the first day of hospitalization. Our assay provided equivalent PCR efficiency and RNA quantification among each DENV serotype. The assay's performance was comparable with previously established real-time RT-PCR targeting coding sequences. Using both assays on the same specimens, our results indicate the presence of defective virus particles in the circulation of patients infected with all serotypes. Dual regions targeting RT-PCR enhanced the sensitivity of viral genome detection especially during the late acute phase when viremia rapidly decline and an incomplete viral genome was clinically evident.

Keywords: DENV 3′-UTR detection; defective virus particles; dengue viral load quantification; dengue virus detection; dual regions detection; incomplete viral genome; one tube; singleplex real-time RT-PCR.

Figure 1

Performance of 3′-UTR primers/probes. Quantification cycle (Cq) of 3′-UTR region (a–e) was determined using serial 10-fold diluted (10¹ to 10⁶ RNA copies) in vitro transcript DENV1-4RNA as template. Coefficient of determination (R²) and PCR amplification efficiency (E) for DENV1 (a), DENV2 (b), DENV3 (c), and DENV4 (d) were analyzed from 8 independent experiments. The variation of Cq values at each RNA concentration among serotypes was analyzed by Bonferroni’s Multiple Comparison Test (e). RNAs extracted from cultured supernatants of Japanese encephalitis virus (JEV, Nakayama strain), yellow fever virus (YFV, 17D strain), zika virus (ZIKV, ZV0127 strain), DENV1 Hawaii, DENV2 16681, DENV3 H87, and DENV4 H241 were used as RNA templates to verify the specificity of our 3′-UTR primers/probes (f). Primers specific to E gene of JEV, YFV, or ZIKV were used to confirm the existence of RNA templates of each virus type. The sizes of PCR products for DENV1, DENV2, DENV3, DENV4, JEV, YFV, and ZIKV were 185, 187, 184, 189, 333, 306, and 365 base pairs, respectively. No RNA template (Neg) was used as a negative control. The PCR product was run in 2% agarose gel electrophoresis and was stained with gel red before visualization under UV light (f).

Figure 2

Correlation of DENV genome levels quantified by RT-PCR specific to 3′-UTR and coding sequence. Quantification cycle (Cq) values or DENV genome levels (Log copies/mL) in various types of samples quantified by RT-PCR using the two types of probe/primer regions were compared and analyzed for correlation coefficient (R) and p values of linear regression. (a) A correlation plot showing Cq values from quantification of 144 samples of in vitro transcribed DENV1-4 RNA (ranging from 10¹–10⁶ copies/mL) from 6 independent experiments. (b) A correlation plot showing DENV genome levels (Log copies/mL) in DENV1-4 infected cell cultured supernatants (2–200,000 ffu/mL) from 15 independent experiments. (c) A correlation plot showing DENV genome levels (Log copies/mL) measured in plasma of 161 DENV infected patients collected since the first day of hospitalization to the day to defervescence (499 samples in total).

Figure 3

Rate of DENV genome detection by real-time RT-PCR specific to only 3′-UTR or coding sequence. (a) Viral RNA levels in plasma of 161 DENV infected patients collected since the first day of hospitalization to the day to defervescence quantified by RT-PCR specific to 3′-UTR (gray circles) or coding sequence (white circles) were re-analyzed according to “Day to defervescence”. The number of samples are labeled on the top of each group. Detection rate of DENV detected by only 3′-UTR (gray bar) or only coding sequence (white bar) were analyzed according to “Day to defervescence” (b) or DENV serotypes (c).

Figure 4

Efficiency of DENV genome detection. Data from Figure 2c were re-analyzed according to “Day to defervescence”. Detection rates of 3′-UTR assay (black bar) or coding sequence assay (white bar) or detected by either assay (gray bar) were analyzed according to “Day to defervescence”. Proportions of DENV detection rates among groups were analyzed by McNemar’s test. Asterisks (** and ***) represent McNemar’s p values < 0.01 and <0.001, respectively.