Establishment of micro droplet digital polymerase 2 chain reaction and real-time fluorescence 3 quantitative polymerase chain reaction 4 technologies for detecting Zika virus 5

1 Guangdong Provincial Key Laboratory of Tropical Disease Research, School of Public 8 Health, Southern Medical University, No. 1023, South Shatai Road, Guangzhou, 9 Guangdong Province 510515, 10 China;yuanhuiapple@163.com(Y.H.);962698900@qq.com(Z.M.W.);baobaofei666666@outlook 11 .com(Z.R.Q.);zhuli89@126.com(L.Z.);819348382@qq.com(X.J.L.);1575424395@qq.com(M.H.F.) 12 ;1019765518@qq.com(S.Y.F.);xweiwei74@126.com(W.W.X.);wuqh@smu.edu.cn(Q.H.W.) 13


Introduction
Zika virus (ZIKA) is a single-stranded RNA virus belonging to the family Flaviviride genus Flavivirus.It was first isolated from a forest in Uganda in 1947 [1] and encodes three structural proteins (capsid protein C, precursor protein pr M, and envelope protein E) and seven nonstructural proteins (NS1, NS2A, NS2B, NS3, NS4A, NS4B, and NS5) [2].
The NS5 gene is the largest gene in the ZIKA genome and encodes NS5 protein, which has the highest relative molecular weight of all proteins encoded by ZIKA.NS5 contains an RNA-dependent RNA polymerase active site at the N terminal and harbors a methylation site, which is associated with the viral cap reaction of the viral RNA.The ZIKA genome has only one open reading frame (ORF).NS5 is located at the 3′ end of the ORF and is a nucleic acid detection target commonly used by the flavivirus [3].Currently, several laboratories have established nucleic acid detection technologies for ZIKA; the main targets of detection are the E and NS5 genes [4].The E gene has undergone several adaptive changes, whereas the NS5 gene is relativelystable.Therefore, the conserved sequence of NS5 gene may be more reliable as a detection target.
ZIKA is transmitted through mosquito bites, particularly those from Aedes aegypti and A. albopictus [5], and can also spread between independent mosquitoes [6].More importantly, the virus can be transmitted from mother to fetus via amniotic fluid and fetal tissues [7] and can be transmitted through sexual activity [8].In humans, ZIKA infection is characterized by mild fever (37.8-38.5 ), 3oint pain (particularly pain in the hands and feet), myalgia, headache, ℃ orbital pain, con3unctivitis, and skin rash and can also cause microcephaly and Guillain-Barré syndrome [9].However, only 20% of infected individuals exhibit symptoms [10],and ZIKA infection can be a benign, self-limiting disease [11].As a result, the incidence of ZIKA infection may be underestimated in epidemics or returning travelers.For the detection of ZIKA, Giada Rossini etal [12] showed that the blood test period is longer after onset of the disease than urine or whole blood tests.ZIKA can be detected in whole blood in about 3-26 days and in the plasma in 3-10 days [13].
Accordingly, in this study, we aimed to establish micro-droplet digital polymerase chain reaction (ddPCR) and real-time fluorescent quantification PCR (qPCR) protocols for the detection of Zika virus based on the NS5 gene.

Sample processing
Asian Zika virus Z16006 strain was provided by the Institute of Microbiology in the Center for Disease Control and Prevention of Guangdong Province, China (GenBank no. KU955589.1).Four serotypes of Dengue virus (DV1 Hawaii strain, DV2 NGC strain, DV3 H87 strain, and DV4 H241 strain) were preserved in our laboratory.The above viruses were all amplified and cultured in C6/36 cells.After observing the cytopathic effect, the cells were repeatedly frozen and thawed three times, centrifuged and frozen at -80 .Clinically positive ℃ ZIKA blood samples were obtained from the Centerfor Disease Control and Prevention of Jiangmen,Guangdong Province, China.

Instruments and reagents
The micro-ddPCR instrument, amplification premixed reagent, Droplet PCR Supermix, and micro-drop reagent were all from Bio-Rad (Hercules, CA, USA).Reverse transcription was performed using a TAKARA Prime Script RT reagent Kit (Perfect Real Time kit; TaKaRa, Shiga, Japan), Bestar qPCR Master Mix (TaqMan Probe) (DBI Bioscience, Germany), and a QuantStudio 6 Flex Real-Time PCR System(ThermoFIsher, USA ).

Design of primers and probes
The sequences of 30 strains of ZIKA were downloaded from GenBank and compared.After identifying conserved sequences, primers and probes were designed using Primer 5.0with the Z16006 strain as the reference strain.The primers and probes were designed using Oligo 7.

RNA extraction and reverse transcription
Samples were brought to room temperature before use.Viral RNA was extracted using the QIAamp® (Qiagen, Hilden, Germany), with a final RNA volume of 50 μL.The RNA was separated into aliquots and preserved at -80 .℃ For reverse transcription, the components were prepared on ice, as follows: 2 μL PrimeScript RT Enzyme Mix I, 0.5 μL Oligo dT Primer, 2 μL Random 6-mers, 4 μL total RNA, and 1 μL RNase-free dH2O.The reaction conditions were as follows: reverse transcription at

ddPCR
For ddPCR, the 20μL reaction system contained 10 μL ddPCR Supermix (no dUTP), 6 μL primer probe premix (initial concentration of 10 μM upstream primer and 0.4 μL downstream primer, probe 0.2 μL, deionized water 5 μL), and 4 μL nucleic acid extract.After mixing, 20 μL of the sample reaction system was added to the middle of a DGB cartridge.Next, 70 μL oil was added to the bottom row of each lane to avoid the formation of bubbles, and the wells were covered.The reaction system and droplet-forming cartridge were placed in a droplet generator and sub3ected to micro-droplet treatment.
Droplets were generated in the top row of the wells, and the suction volume was ad3usted to 40 μL.Samples were then slowly transferred to 96-well plates, and a preheated PX1 heat sealing device was used as a sealing film (with the red line up) at 180°C for 5 s.The PCR conditions were as follows: predenaturation at 95°C for 10 min, temperature change rate of 2°C/s; and 40 cycles of 94°C for 30 s and 55°C for 1 min, at 2°C/1°C/s.The 96-well plate containing the PCR-amplified products was then placed on a QX200 micro-drop reader and analyzed using Quanta Soft software.

Design and evaluation of primers and probes
We downloaded 30 popular ZIKA strains, compared them, and found conserved sequences.
Primers were designed according to the NS5 gene, keeping the GC content and Tm values as similar as possible.The length of the target fragment was within 100 bp without hairpin structures to avoid the formation of stable dimers and mismatches at the C-terminus.The designed primers are shown in Table 1.

Table 1. Primers and probe for ZIKA detection
Primer/probe Sequence

Evaluation of the sensitivity of detection
Ten-fold dilutions of ZIKA plasmid standards (3.2 × 10 10 copies/μL) were performed.
Detection by qPCR showed that the standard curve R 2 value was 0.999, and the amplification efficiency was 92.203%.The relationship between the Ct value and the copy number was as follows: Y = -3.524X+ 38.045, based on the standard curve (Figure 1).The initial concentration of 7.8 × 10 6 copies/μL was gradient diluted 5-fold from the cell culture ZIKA solution, and qPCR was used to test each sample in order to evaluate the detection method.The results are shown in Figure 2; a detection limit of about 100 copies/μL was obtained.Next, we used ddPCR to detect ZIKA from cell cultures, as described above for qPCR.As shown in Figure 3A, qPCR detected ZIKA with clear differences among sample concentrations.In Figure 3B, the threshold was set to 1000, with positive nucleic acid droplets showing values above 1000 and negative nucleic acid droplets showing values below 1000.As indicated in Figure 3C, almost all samples generated more than 10,000 droplets, suggesting that ddPCR may be more sensitive, with a detection limit of 1 copy/μL.

Evaluation of the accuracy of qPCR and ddPCR
Next, we evaluated the accuracy of the assays using four Dengue virus serotypes.All of qPCR results were negative.Figure 4 shows the results of ddPCR.As shown in Figure 4A, nucleic acid concentrations detected by the absolute quantification method were all zero.
Moreover, only negative micro-droplets were observed (Figure 4B), and histogram analysis showed that the total number of droplets was more than 10,000, of which none were positive (Figure 4C).

qPCR detection of Dengue virus samples in clinically positive blood samples
The lowest Ct value of clinically positive blood samples in qPCR analysis was 38.868 after 5-fold dilutions; this exceeded the detection range of qPCR (Ct values of 15-35; Figure 5).DdPCR analysis showed that the sample concentration was 14.2 copies/μL (Figure 6).
Figure 6A shows the scatter plot of the events, and Figure 6B shows the concentrations based on absolute quantification.Finally, the histogram of the event number showed that the total

Discussion
Approximately 20% of individuals infected with ZIKA may develop symptoms similar to other flaviviruses [11], presenting a great challenge to clinical diagnosis.There are two types of clinical diagnostic tests for ZIKA, serological and molecular detection.IgM specific neutralizing antibodies and antibody capture enzyme-linked immunosorbent assays can be used for qualitative detection of ZIKA IgM antibodies, but may crossreact with other flaviviruses, making the results difficult to explain.Therefore, molecular biology diagnosis has become a commonly used means of laboratory testing.
In molecular biology, traditional PCR can only roughly detect the amount of amplification after the reaction has ended and cannot be used to quantitatively detect nucleic acids in the sample.Real-time fluorescence qPCR technology utilizes changes in the fluorescence intensity of chemical substances in the reaction system to realize quantitative detection of nucleic acids.Xu et [1]al [14] used one-step SYBR Green real-time PCR for the detection of ZIKA with a detection limit of at least 1.0 PFU/mL (1 PFU is approximately equal to 2 × 10 5 RNA genome copies).However, this method allows the simultaneous detection of both specific and nonspecific PCR products and therefore produces false positives.Calvert et al. [15] used reverse transcriptase loop-mediated isothermal amplification to detect RNA from ZIKA as low as 1.2 copies/μL; however, they observed a very high false-negative rate.Our laboratory uses a probe-based real-time PCR method with a detection limit of approximately 100 copies/μL.Micro-ddPCR detects nucleic acid molecules without relying on external standard curves based on the Poisson distribution principle.With the advantage of absolute quantitative, this method permits better accuracy at low concentrations without the need for preparing standard curves [16].
In this study, we designed specific primers and probes based on the NS5 gene of ZIKA and established a method for ZIKA nucleic acid detection using ddPCR and qPCR.Our results showed that good specificity could be obtained by designing primers and probes based on the NS5 gene and that these primers/probes could be used for detection the differential ZIKA and four serotypes of Dengue virus.Moreover, our findings demonstrated that ddPCR had good sensitivity but was inaccurate for samples with high concentrations.Thus, ddPCR may be more suitable for low viral loads.Thus, ZIKA, as well as four serotypes of Dengue virus, can be detected by analyzing the NS5 gene at concentrations of about 1-10 5 copies/μL.
In summary, in clinical samples with low concentrations of ZIKA, such as those at 1-3 days after infection, micro-droplet digital ddPCR may show the best diagnostic accuracy and sensitivity, whereas in routine analyses laboratory of viral nucleic acid detection, including analysis of clinical samples at more than 3 days after infection, fluorescence qPCR can be used.

Figure 1 .
Figure 1.The standard curve for fluorescent quantitative PCR.The concentration of each

Figure 2 .
Figure 2. Results of 5-fold dilutions of ZIKA cell culture by qPCR.

Figure 3 .
Figure 3. Results of ddPCR for 5-fold dilutions of cell culture.

Figure 4B .
Figure 4B.Scatter plot of the total number of droplets (from left to right: DV4, DV3, DV2,

Figure 4C .
Figure 4C.Total number of micro-droplets, as shown in a histogram (positive on left and total

Figure 6A .
Figure 6A.Scatter plot showing the event numbers.