Development and validation of a multiplex qPCR assay for detection and relative quantification of HPV16 and HPV18 E6 and E7 oncogenes

Human papillomaviruses (HPV) play a key role in promoting human anogenital cancers. Current high-risk HPV screening or diagnosis tests involve cytological or molecular techniques mostly based on qualitative HPV DNA detection. Here, we describe the development of a rapid quantitative polymerase chain reaction (qPCR) detection test of HPV16 and HPV18 oncogenes (E6 and E7) normalized on human gene encoding GAPDH. Optimized qPCR parameters were defined, and analytical specificities were validated. The limit of detection was 101 for all genes tested. Assay performances were evaluated on clinical samples (n = 96). Concordance between the Xpert HPV assay and the triplex assay developed here was 93.44% for HPV16 and 73.58% for HPV18. HPV co-infections were detected in 15 samples. The systems developed in the present study can be used in complement to traditional HPV tests for specifically validating the presence of HPV16 and/or HPV18. It can also be used for the follow-up of patients with confirmed infection and at risk of developing lesions, through the quantification of E6 and E7 oncogene expression (mRNA) normalized on the GAPDH expression levels.

www.nature.com/scientificreports/ (Rb) by E7 [12][13][14] . In contrast, LR-HPV oncogenes are less able to interfere with p53 and Rb functions than E6/E7 proteins from HR-types 12,15 . Nowadays, HR-HPV screening or diagnosis mostly involve cytological techniques that have lowest specificity and sensitivity than molecular techniques 16,17 but detection of HPV DNA is becoming the first line strategy for cervical cancer in women older than 30 years [18][19][20] . Commercially available molecular tests are mainly based on the detection of HPV DNA such as Hybrid Capture 2 assay (QIAGEN) 21 , Cervista HPV HR 22 or HPV 16/18 23 (HOLOGIC) Care HPV (QIAGEN) 24 and Cobas HPV (ROCHE DIAGNOSTICS) 25 , are mainly based on the detection of HPV DNA. These allow detecting the presence of the virus but not evaluating its oncogenic activity. In cervical infections HPV load and its variations have been shown to predict the persistence and progression of HPV infections and the severity of the lesions 26,27 . In 2010, a study showed that mRNA expression levels of E6 and E7 oncogenes were correlated with the severity of cervical lesions 28 , but so far, only few HPV tests target viral mRNAs. They are qualitative (e.g. HPV-Proofer (PRETECT)) 29 and do not discriminate among HPV types (Aptima HPV Assay (HOLOGIC)) 30 .
The main aim of the present study was to develop a rapid quantitative detection test of HPV16 and HPV18 infections by measuring E6 and E7 oncogene DNA levels normalized by the cellular GAPDH gene level. Multiplex detection of these three genes in a single assay enables reduction in cost, time and labor as compared to separate detection methods. This technique allows for an extra support of traditional HPV tests to validate and quantify specifically the presence of HPV16 and/or HPV18 infection. Once persistent infection is confirmed, it could be used for a follow-up of patients at risk of cervical cancer, by correlating the activation of oncogenes to the severity of the lesions.

Materials and methods
Primer and probe design. Target sequences of E6 and E7 genes were obtained using the NCBI reference genomes of HPV16 (accession number: NC_001526.4) and HPV18 (accession number: X05015.1). The primers and probes were designed to amplify and detect specifically E6 and E7 oncogenes of HPV16 and HPV18 using PrimerQuest Tool software with default parameters. The specificity of the primers and probes was checked using the online NCBI BLASTn tool against the human genome (https ://blast .ncbi.nlm.nih.gov/). Human GAPDH primers were obtained from previous study 31 .
The hydrolysis probes of E6, E7 and GAPDH were labeled on the 5′-end with FAM, VIC and CY5 fluorescent dyes, respectively, and on the 3′-end with a TAMRA compatible quencher for E6 and E7 and BHQ (Black Hole Quencher) for GAPDH.
The sequences of primers and probes, nucleotide positions on HPV16 or HPV18 reference genomes and amplicon lengths are presented in Table 1. Primers and probes were manufactured by EUROGENTEC (Angers, France). Amplicons were produced by standard PCR in a Mastercycler nexus gradient (EPPENDORF) according to the following program: 10 min at 95 °C; then 40 cycles of 30 s at 95 °C, 30 s at 60 °C, and 20 s at 72 °C; and a final extension step of 7 min at 72 °C. PCR products were analyzed by electrophoresis on a 2% agarose gel and purified by QIAquick PCR purification (QIAGEN) before being cloned into a PGEM-T easy vector (PRO-MEGA). The ligation reaction was conducted overnight at 4 °C according to the manufacturer recommendations. JM109 competent cells (PROMEGA) were transformed by heat shock, before being spread with beads onto Luria-Bertani (LB) (BD DIFCO) agar plates containing 100 µg/ml ampicillin (SIGMA), 0.5 mM isopropyl-βd-thiogalactopyranoside (EUROMEDEX) and 80 µg/ml X-gal (EUROMEDEX). Cells were incubated at 37 °C for 24 h or during the weekend at room temperature. Transformants were analyzed by PCR using SP6 (5′-TAT TTA GGT GAC ACT ATA G-3′) and T7 primers (5′-TAA TAC GAC TCA CTA TAG GG-3′) according to the following program: 10 min at 95 °C; then 40 cycles of 30 s at 95 °C, 30 s at 55 °C, and 20 s at 72 °C; and a final extension step of 7 min at 72 °C. Positive E6 and E7 clones for each HPV genotype and GAPDH clones were placed into 7 mL of LB broth containing 100 µg/mL of ampicillin then incubated overnight at 37 °C under shaking (~ 200 rpm). Plasmid DNA was extracted using the Smart Pure Plasmid Kit (EUROGENTEC). Purified plasmids were linearized by digestion with the NcoI restriction enzyme during 15 min at 37 °C according to the manufacturer's instructions. Digestion products were run on a 1% agarose gel and the band was purified using the QIAquick Gel Extraction (QIAGEN). The plasmid concentration was quantified by Nanodrop 2000 (THERMO SCIENTIFIC) for HPV16 and by Multiskan GO (THERMO SCIENTIFIC) for HPV18. The copy numbers (copies/µL) were determined with the following formula: Plasmid copy numbers were 1.59 × 10 10 copies/µL (GAPDH), 1.44 × 10 10 copies/µL (HPV16, E6), 9.9 × 10 9 copies/µL (HPV16, E7), 1.54 × 10 10 copies/µL (HPV18, E6) and 1.58 × 10 10 copies/µL (HPV18, E7). Serial dilutions (from 10 7 to 10 0 ) of the control plasmids were aliquoted and stored at − 20 °C until further use. All steps of qPCR development were run with 2 µL of plasmid solution. Specificity of the primers. The specificities of the primers-probe couples were tested on each plasmid produced as mentioned above at the concentration of 10 5 copies in a multiplex qPCR by using the optimized parameters on a CFX C1000 Touch (BIORAD). Each run consisted in three steps: 2 min at 50 °C, 5 min at 95 °C, then 40 cycles of 10 s at 95 °C and 20 s at 60 °C.
Efficiency and sensitivity of the PCR systems. PCR efficiency and sensitivity of each primers-probe couple with the optimized conditions were determined using a quantitative standard curve (serial dilution of the positive control plasmid from 10 7 to 10 0 copies). All assays were performed in biological and technical triplicates on a CFX C1000 Touch (BIORAD) with the following program: 2 min at 50 °C; 5 min at 95 °C; 40 cycles of 10 s at 95 °C and 20 s at 60 °C. Assays were validated with a slope of regression curve between − 3.9 and − 2.9 cycles/ log 10 , a R 2 > 0.98 and an efficiency range between 80 and 120% 32 . The limit of detection (LOD) of the assays was defined as the lowest concentration (target gene copy number per reaction) for which ≥ 95% of test runs gave positive results.
Multiplex reproducibility. Intra-and inter-assay reproducibility was evaluated based on the cycle threshold (Ct), and the coefficient of variation (CV) of both the biological and the technical replicates, using a concentration of 10 5 copies of each target gene.
Triplex validation and application on clinical samples. As reference for HPV detection, we used the diagnostic results obtained at the IHU Méditerranée Infection laboratory using the Xpert HPV Assay (CEP-HEID) from 96 clinical materials (67 cervico-vaginal smears, four vaginal swabs, one cervix, 19 anal smears, five rectal swabs) collected in liquid-based media (PreservCyt, HOLOGIC) (Supplementary Table 1). The Xpert HPV Assay is a real-time polymerase chain reaction assay using disposable cartridges that is able to detect 14 For testing the multiplex systems developed in the present work, DNA extraction from 1 mL of each of the 96 original clinical samples was performed using the BioRobot EZ1 instrument (QIAGEN) or High Pure Nucleic Acid Large Volume extraction kits (ROCHE) according to the manufacturer recommendations. qPCR on DNA samples (5 µL) were run on a CFX C1000 Touch (BIORAD) according to the optimized parameters determined in this work. qPCR products with Ct values > 38 and PCR products visible on a 2% agarose gel using universal HPV primers were verified by Sanger sequencing using Big Dye Terminator v1.1 Cycle Sequencing Kit (LIFE TECHNOLOGIES) according to manufacturer recommendations on a 3500XL Genetic Analyzer capillary sequencer (THERMOFISHER).
Using the standard curve, the copy number of each HPV18/16 oncogenes and GAPDH was determined. The HPV viral load was calculated according to the following formula 26 : Ethics statement. Clinical samples used in this study have been collected in the setting of routine standard clinical management and informed consent was obtained from all subjects. The study was approved by the ethical committee of the University Hospital Institute Méditerranée Infection (No. 2020-03). The study was performed according to the good clinical practices recommended by the Declaration of Helsinki (and its amendments) and all methods were carried out in accordance with relevant guidelines and regulations.
Statistical analyses. All statistical analyses were performed using the Prism 7 software (GRAPHPAD).
Two-way ANOVA and Tukey's multiple comparison test were performed to analyze the differences between the Ct and RFU values (for primer and probe concentrations) between conditions, with H0 rejected for a p-value < 0.05. Concordance and Cohen's kappa value 33 were calculated to determine the level of agreement of HPV16 and HPV18 detection results between the multiplex systems developed in this study and the Xpert HPV Assay. Kappa values of < 0.20, 0.21-0.40, 0.41-0.60, 0.61-0.80 and 0.81-1.00 were considered poor, fair, moderate, good and very good agreement, respectively.

Results
Optimization of the multiplex qPCR parameters. To optimize the fluorescence rate of each gene in the multiplex system, four primer concentrations (400 nM, 600 nM, 800 nM, 1000 nM) and four probe concentrations (50 nM, 100 nM, 150 nM, 200 nM) were tested. The optimal concentration for the two multiplex qPCRs (HPV16 or HPV18) was determined as the highest relative fluorescence unit (RFU) for the lowest Ct value (Table 2a, b). Results showed that best RFU means were obtained with 200 nM of probe (p ≤ 0.05) and that there were no significant differences between Ct values and primer concentrations independently of probe concentration (p > 0.05). The lowest concentration of primers (400 nM) was then selected as optimal condition.
A temperature gradient (54 °C, 56.4 °C, 57.8 °C and 60 °C) was tested to select the best annealing temperature for the lowest Ct value (Table 3a,b). For the two multiplex qPCRs, Ct values were not significantly different among all the temperatures tested (p > 0.05). To be as specific as possible, we thus selected the annealing temperature of 60 °C as optimal temperature. Multiplex qPCR specificity. To determine the multiplex specificity for the two HPVs, each plasmid gene (E6, E7, GAPDH) separately or pooled were run in qPCR reaction mixes with the pool of primers and probes for HPV16 or HPV18 using the optimal qPCR parameters previously determined. qPCR resulted in a single curve for each specific plasmid run in the HPV16 multiplex mix. This indicated that this triplex was able to detect all the three genes without cross-amplification and non-specific background fluorescence (Fig. 1a). For HPV18, non-specific background amplification of E6 and E7 and cross-amplification between HPV16 and HPV18 were observed after a Ct of 38 (Fig. 1b).
Limits of detection (LOD) were determined as the last dilution of plasmids for which ≥ 95% of the intra-and inter-triplicates gave a positive result. The mean frequency of positives for the multiplex qPCR was 100% at 10 2 plasmid copies/µL, and 97.88% and 97.34% at 10 1 plasmid copies/µL for HPV16 and HPV18, respectively (Table 4).
Assay reproducibility. The coefficient of variations for the biological (intra-assay) triplicates were all less than 8% and those for the technical (inter-assay) triplicates were all less than 5% (Tables 2 and 3). www.nature.com/scientificreports/ were HPV18-negative were expected to be HPV45 and were verified by Sanger sequencing after conventional PCR using universal HPV primers 34 . The results confirmed HPV45 in seven cases, but also detected HPV53 in three cases. Co-infection with HPV16 was detected for three samples (no. 58, 64 and 90). All negative samples (n = 8) were found negative, except one (sample no. 5), for which HPV16 and HPV18 coinfections were detected with late Ct values for E7 gene, 38.02 and 36.67, respectively. HPV16 was also detected in one of the other HR-HPV samples (no. 24).
Considering the negative controls (negative samples and other HR-HPVs) and the positive samples, the concordance between the Xpert HPV Assay and the triplex qPCRs developed in this study was 93.44% for HPV16 with a Cohen's kappa value of 0.82 (almost perfect agreement between the two tests) and 73.58% for HPV18 with a Cohen's kappa value of 0.48 (moderate agreement).

Discussion
In most of cases (80-90%), HPV infections are transient as the result of viral clearance by the host immune system whereas viral persistence is necessary to induce HPV related-diseases (warts, cancers…) 35,36 . Between 60 and 90% of worldwide HPV cancers cases are due to HPV16 and/or HPV18 6 and infection by one of those HPV types has a predictive value of cervical intraepithelial neoplasia grade 3 lesion 37 . The present study describes the development and validation of two real-time PCR triplexes for fast detection and relative quantification of E6 and E7 DNA oncogenes for HPV16 and HPV18. Amplification efficiencies of each gene of the two multiplex systems ranged between 80 and 120% with good standard curve coefficient of correlation (R 2 > 0.99). The difference between the PCR efficiencies obtained for each target gene of the multiplex system never exceeded 15%, which further demonstrated the absence of competition between the different targets 32,38 . The low coefficient of variation for both intra-(≤ 7.4%) and inter-(≤ 3.7%) assays showed a good reproducibility of each steps of the multiplex development. The triplex qPCR assays allowed the detection of up to 10 1 plasmid copies/µL for all genes tested.
Currently, diagnosis of HPV infection often uses commercially available tests that target the complete genome, L1 gene or E6/E7 genes of a panel of HPV types (Hybrid Capture 2, Cervista HPV HR or HPV 16/18, Cobas Figure 2. Sensibility and linearity of HPV16 and HPV18 triplexes. Standard curves were generated by amplification of serial dilutions from 10 7 to 10 -2 of E6, E7 and GAPDH plasmid pool for HPV16 (a-c) and HPV18 (d-f) using multiplex qPCR. Each dilution was run in biological and technical triplicate. R 2 represent the coefficient of determination and the PCR efficiency (E) of each target was calculated using the slope of each standard curve with the formula E = (10 -1/slope − 1)*100. Table 4. Positive frequency for multiplex with plasmid serial dilution (10 7 -10 0 copies/µL). Positive frequency was calculated according to the ratio between the numbers of positive amplification detection and the number of expected amplification curve for each dilution of HPV16 (a) and HPV18 (b).  www.nature.com/scientificreports/ HPV or Xpert HPV) 39 . Some of these qualitative HPV tests uses specific technologies such as hybrid capture with signal amplification by chemiluminescence (Hybrid Capture 2, Care HPV) or hybridization by FRET probe (Cervista HPV HR). Other HPV tests are easier to use but need specific additional laboratory equipment (Cobas HPV, Xpert HPV, Papillocheck). The two multiplex systems developed in this study allow detecting DNA of two of the most prevalent HR-HPVs and to quantify E6 and E7 oncogenes DNA levels compare to GAPDH housekeeping gene within the same amplification reaction. Moreover, in house probe-based real-time qPCR is a fast, easy to use, specific and reproducible method that only requires a minimal amount of DNA, does not need specific, assay-dedicated laboratory instrument, and can allow testing in the same run multiple sample DNA extracts on 96-or 384-well plates.
The multiplexes developed in this study confirmed the clinical diagnosis of the Xpert HPV assay with a concordance of 93.44% for HPV16 and 73.58% for HPV18. Lower concordance and Cohen's kappa value were observed for HPV18 as the Xpert HPV assay provided a positive signal for both HPV18 and HPV45 whereas our system only targeted HPV18. Using HPV DNA quantification, we noted for some patients that oncogenes E6 and E7 have a Ct value lower than the GAPDH gene. This could be a signature of an active virus replication and transient infection. In contrast, patients with Ct for oncogenes equal to or higher than the GAPDH gene may reflect persistence under an episomal state or HPV integration to the human genome. In this work, 15 samples with HPV18 and HPV16 co-infections were detected. Detection of HPV co-infections is of particular interest as a combination of HPV16/18 co-infection has been associated with grade 2 and 3 cervical intraepithelial neoplasia (CIN) 40 . Viral loads per millions of human cells were determined for HPV16 and HPV18 as these could be a marker of CIN2 and be a predictor of lesion progression 27,41,42 . The value measured in this study were within the limits of the standard curves and showed large variations across patients and samples.
In 2010, a study showed that mRNA levels of E6 and E7 oncogenes are correlated with the severity of cervical intraepithelial neoplasia 28 . Quantification of mRNA could thus be a novel marker of HPV infection and cervical intraepithelial neoplasia progression 43 . So far, only few HPV detection test target mRNAs. These tests are qualitative (HPV-Proofer) and do not discriminate HPV types (Aptima HPV Assay). In 2018, a multiplex reverse transcription real-time PCR for E6/E7 mRNA detection of 14 h-HPV without discrimination among types was developed 44 . This study confirmed the association of E6/E7 mRNA increased levels with the severity of the cervical lesions. Another study showed that the detection of both E6 and E7 mRNAs showed significant association with the occurrence of upgraded abnormal cytology 45 . In this study, we could simultaneously detect and quantify E6 and E7 genes within the same amplification reaction. The triplex qPCR developed in this study could then be used to assess E6/E7 oncogene mRNA levels for a personalized follow-up of patients with confirmed HPV18 and/or HPV16 persistent infections, in order to survey women at risk of developing cervical carcinoma.
To conclude, the multiplex PCR systems developed in this study could be used as extra support of traditional HPV tests to validate specifically the presence of and quantify per million cells HPV16 and/or HPV18 DNA. It may also be used for research in second intention for the follow-up of patients with confirmed infection who are at risk of developing lesions by quantifying expression levels of E6 and E7 oncogenes normalized on the GAPDH expression. This could be easily realized simultaneously on many samples (96-384 wells qPCR plate) and could be developed in the future for other less frequent HR-HPVs. www.nature.com/scientificreports/ Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.