Evaluation of molecular detection for viral hemorrhagic infection cases of yellow fever and dengue fever in Côte d'Ivoire 2010-2012, West Africa.

Background: Viral hemorrhagic fevers are emergent and endemic in Africa and in South America. In Cote d’Ivoire, Yellow fever cases were reported yearly and the distribution of mosquitoes in the country are the main factors for high incidence of Flaviviruses. The poorly reporting of viral hemorrhagic fever cases in some regions, the lack of international interest land and the underestimation of molecular surveillance method contribute to increase the risk for public health. Objectives: To evaluate the performance of molecular diagnostic methods in national surveillance Short Communication Solange et al.; BJMMR, 5(3): 396-403, 2015; Article no.BJMMR.2015.043 397 of two emergent Flaviviruses, Yellow fever virus and Dengue virus in Cote d’Ivoire. Study Design: 63 sera from suspected cases in 2010-2011 of viral hemorrhagic fevers were analyzed to detect viral RNA of Flaviviruses and to compare the results in three different methods. Results and Conclusion: The Flavivirus RT-PCR has showed the high molecular detection by12% and 6% for real time PCR. The methods are specific and high sensitive for the screening of tick and mosquito-borne Flaviviruses in clinical samples. This study confirms the high circulation of Flaviviruses and the introduction of Dengue virus in Cote d’Ivoire. The combination of real time PCR and the Flavivirus RT-PCR contribute to ameliorate the detection panel of molecular detection in Cote d’Ivoire and was a suitable method for the National Laboratory Reference.

Since 2004, the World Health Organization (WHO) reported 414,785 cases of Yellow fever worldwide, of these, 200,000 cases were from South America and Africa [5][6]. In Côte d'Ivoire, cases of Yellow fever were reported each year and some imported cases from West Africa in Europa have been most reported [7][8][9][10]. In 2011, the national surveillance of YF in Côte d'Ivoire has reported 10 deaths and the first cases of dengue hemorrhagic fever and the cases are been confirmed in Regional Reference Laboratory [11][12]. Most cases of hemorrhagic fever are likely to be large underestimates, such that Flavivirus infections may still represent a significant risk for the public health in Africa [13].
Because there are no specific clinical symptoms for infection by a determined virus, and because different Flaviviruses could be present in the same area, a genus diagnosis by PCR would be a useful first-line diagnostic method. The clinical diagnosis of Flavivirus infections is not unambiguous due to unspecific symptoms varying from mild, febrile illness to viral hemorrhagic fever.

Conventional
Flaviviruses diagnosis recommended by the WHO is based on serology tests screening for the presence of virus-specific YF-IgM antibodies in the patient serum. The exceptional sensitivity of the PCR method allowed rapid detection and identification of Flaviviruses in mosquitoes and clinical samples [14][15][16][17][18][19][20], in which virus culture is difficult or timeconsuming and when early diagnosis is necessary for clinical treatment and has implications for vaccination and mosquito control. Most current molecular assays for Flaviviruses use highly specific primers, which may only amplify from one species, or a range of closely related species [17,[21][22][23][24][25][26][27][28]. The NS5 region is the most amplified gene for Flaviviruses detection and has the most conserved region in genome [25,27,[28][29][30][31].
In this study, the performance of molecular methods for detection of both Flaviviruses: Yellow Fever and Dengue Fever will be evaluate for clinical samples from the National Program Surveillance in Côte d'Ivoire.

Clinical Samples
63 Blood sera were collected in 2010-2011 from suspected cases of hemorrhagic fever and were stored at -20ºC before treatment. The samples were collected from National Surveillance Program for viral hemorrhagic fevers of Côte d'Ivoire and were transferring tothe National Reference Laboratory in Pasteur Institute Abidjan. The sera were collected between 1-7 days after the clinical symptoms. Epidemiological data of the patients were collected for the National Surveillance Program in CNR Arbovirus (Data not shown). Informed written consent was obtained from all patients.

RNA Extraction
200 µl of sera were used in RNA extraction protocol by magnetic extraction [32]. Briefly, after the lysis incubation, RNA was captured by silica and the solution was washed in magnetic stirring rotation. RNA was eluted and stored at -20ºC or used immediately for molecular analysis. Positive strains controls were strain IvoryC1999, strain 17D (Gift of Robert Koch Institut) and Vaccine anti-amaril 17D (Sanofi Aventis) were extracted by the same RNA extraction method. Serial dilutions of RNA from the control strains were applied and each sample was tested in triplicate for molecular tests.

Hemi-nested Flavivirus RT-PCR
The molecular detection of viral hemorrhagic fever was performed by the protocol of Scaramozzino et al. [24] with minor modifications for the detection of Flaviviruses using NS5gene.The primers were listed ( Table 1). The first RT-PCR was involved with 8 µl of extracted RNA, 0.2 µM of each primer CFD2 and MAMD, 1X Reaction puffer (Invitrogen) and 0.2 µl of Reverse Transcriptase-Taq Polymerase (In vitrogen). The second hemi-nested was performed with the primers FS778 and CFD2. The reaction has included 5µl of amplicon from the first RT-PCR, 1X Go-Taqbuffer (Promega), 0.4 mM of dNTPS, 0.5 µM of each primer FS778 and CFD2 and 1Unit Go-Taq Flexi DNA Polymerase (Promega). The PCR reactions were running by using the 9700 Thermocycleur with the same conditions described by Scaramozzino et al. [24]. 10 µl amplicon of hemi-nested Flavivirus PCR were identified by their molecular weights analyzed by electrophoresis in a 1.5% agarose gel, and the separated fragments were stained with ethidium bromide and visualized under UV light trans-illumination.

Real Time RT-PCR for Yellow Fever/
Dengue Fever 5µl of extract RNA was amplified in 20µl reaction volume containing 1X Reaction buffer (Invitrogen), 0.3 µM of each primer (Table 1), 0.6 µM Probe, 1 Unit Enzyme-mix (Invitrogen) and 0.5 µl of ROX dye (Promega). The amplification was running in 7500 Real time PCR (Applied Biosystems) by the follow conditions: Reverse Transcription: 50ºC, 2 min, initial denaturation: 95ºC, 15 min: (40 cycles) 95ºC, 15 sec, 60ºC, 1 min. Positive controls RNA wereviral RNA from the vaccine YF 17D and positive strains YF and negative controls were performed by RNase free water in each amplification run.

Sensitivity of Molecular Methods
The viral vaccine YF17D was diluted in 500 µl sterile PBS1X and the RNA was extracted by magnetic extraction (32). Serial dilutions of RNA were performed in RNase-free water from 10 -1 -10 -8 to evaluate the limit detection of viral load in hemi-nested Flavivirus PCR and in real time RT-PCR. Each 5µl of viral RNA was tested triplicate in both methods as the same above conditions.10 µl amplicon of Flavivirus PCR were identified by their molecular weights analyzed by electrophoresis in a 1.5% agarose gel, and the separated fragments were stained with ethidium bromide and visualized under UV light transillumination.

Sensitivity of Molecular Methods by
Serial Dilution of RNA  Table 2). The real time PCR has showed a good correlation of target dilution with linear increasing of the threshold cycle (Ct). The high detection limit of those methods is suitable for detect clinical sera of suspected cases of Flaviviruses.  (Table 3). The quality (storage condition) and the low quantity of the samples were poorly and affect the molecular detection because 36.5% (23/63) of samples cannot be analyzing. These results confirm the previous results for the high sensitivity of the method 1 for Flaviviruses detection.

Detection of Flaviruses by Molecular Methods in Clinical Sera
The hemi-nested Flavivirus was a good indication for general screening of Flaviviruses and been used for mosquito national surveillance. These results confirm the circulation of Flaviviruses in West Africa, and the need to implement the molecular detection in National Reference Laboratory.

DISCUSSION
Molecular assays have showed to be more sensitive than virus isolation or serological tests. This is because; serological assays were the first screening method and show the limitation by cross reactivity and their consequences. The mass campaign of Yellow Fever vaccination in Côte d'Ivoire can reduce the incidence but the mosquito control was not effective in the country and the introduction of new variants can promise the serological surveillance recommended by WHO [34]. Our approach was to evaluate the molecular Flaviviruses diagnosis methods in clinical sera and subsequently to compare with the results from International Reference Laboratory.
In this study, we have tested the molecular detection for sera suspected cases from 2010 and 2011 and compared the sensitivity and specificity. The hemi-nested Flavivirus PCR has showed 12%, while real time RT-PCR for dengue group-specific has detected 6% of the samples. The results of the Reference laboratory have higher by 20% and confirm the circulation of Flaviviruses in Côte d'Ivoire. The use of generated oligonucleotides is proven to ameliorate the detection by variable region occurs from mutations. This results shows that the NS5 gene and 3'UTR region are the most conserved region of Flaviviruses [24,25,28,35]. Scaramozzino et al. [24] have reported the detection of 30 different tick-and mosquito-borne Flaviviruses by using the hemi-nested method with generated primers for broad detection. Due to the establishment of a sentinel network for viral hemorrhagic fever surveillance, which includes blood collections from febrile cases for virus detection, it was possible to detect early the introduction of new strains of Flaviviruses into the human population. Traditional serological methods based on neutralization and fixed cell ELISA have proven effective for identifying Flaviviruses and must be confirmed by molecular methods. The using of generated primers by method detection allows the broad detection of the five distinct genotypes of YF strains in Africa, dengue viruses and new variants strains of Flaviviruses.
Our results have demonstrated similar high sensitivity of the molecular test in serial dilution of extracted viral RNA with the previous results [28,31] using the same target sequence in panflavivivirus detection RT-PCR.
In this study, the hemi-nested Flavivirus has demonstrated similar sensitivity as first described in previous studies using degenerate primers with internal consensus region of NS5 region of all Flaviviruses [24,27]. In our study, the heminested method cannot discriminate Yellow Fever virus and Dengue virus because the positive amplicon were not being sequencing because limiting of the laboratory. The high detection limit of real time PCR (2log) was significant and confirm the sensitivity by 2-10 PFU/ml corresponding of 6 genomes copies/reaction using real time RT-PCR in several studies for the detection of Yellow fever and Dengue fever [16,28,36].
The high number of not tested samples in Laboratory could be due to (i) the long transport of samples by not optimal freezing conditions from patient to the laboratory (ii) the small quantity of some sera for all available tests. The accessibility for molecular tests was necessary for rapid specific screening of hemorrhagic Flaviviruses surveillance in Côte d'Ivoire. However, the real-time PCR is also used to quantitate the viral load in blood samples, making it a useful tool to investigate the role of viremia in pathogenesis of hemorrhagic Flaviviruses [36].
The confirmation of molecular tests in the International Reference Laboratory contributes to the quality control in Institut Pasteur, Côte d'Ivoire for detection of viral hemorrhagic fevers. The integration in external quality control compendium for molecular tests for Flaviviruses was very necessary in West Africa [20,34].

CONCLUSION
The

ETHICAL APPROVAL
Not applicable.

ACKNOWLEDGMENTS
This work was supported by research grants from the Ministère de la Santé by the national WHO, the CDC and RIIP. We thank the Robert Koch Institut for providing of positive strains and effective collaboration for the establishment of molecular techniques and the Department of Epidemic Viruses in Pasteur Institute, Abidjan.