Development of Real-Time Molecular Assays for the Detection of Wesselsbron Virus in Africa

Wesselsbron is a neglected, mosquito-borne zoonotic disease endemic to Africa. The virus is mainly transmitted by the mosquitoes of the Aedes genus and primarily affects domestic livestock species with teratogenic effects but can jump to humans. Although no major outbreak or fatal case in humans has been reported as yet worldwide, a total of 31 acute human cases of Wesselsbron infection have been previously described since its first isolation in 1955. However, most of these cases were reported from Sub-Saharan Africa where resources are limited and a lack of diagnostic means exists. We describe here two molecular diagnostic tools suitable for Wesselsbron virus detection. The newly established reverse transcription-quantitative polymerase chain reaction and reverse-transcription-recombinase polymerase amplification assays are highly specific and repeatable, and exhibit good agreement with the reference assay on the samples tested. The validation on clinical and veterinary samples shows that they can be accurately used for Wesselsbron virus detection in public health activities and the veterinary field. Considering the increasing extension of Aedes species worldwide, these new assays could be useful not only in laboratory studies for Wesselsbron virus, but also in routine surveillance activities for zoonotic arboviruses and could be applied in well-equipped central laboratories or in remote areas in Africa, regarding the reverse-transcription-recombinase polymerase amplification assay.


Introduction
Wesselsbron virus (WSLV) is a mosquito-borne flavivirus first isolated in 1955 from the blood of a febrile man and from a dead lamb during an outbreak in the South African town of Wesselsbron [1,2]. WSLV is responsible for the Wesselsbron disease (WSL), a neglected disease associated with teratogenic effects in lambs, and abortion and mortality in pregnant ewes as the Rift Valley Fever virus. However, only a less severe fever was reported in adult WSLV-positive goats, cattle and pigs [3,4]. In addition, WSLV has also been associated with neurological damage in horses in South Africa [5] and, more recently, the virus has been isolated from rodent's brain tissues in Senegal [6]. However, human cases were commonly characterized by a short period of fever, arthralgia and myalgia [6,7].
Thus, development of reliable and specific assays is needed for detection and surveillance of WSLV not only in countries where the virus circulation has been previously reported [7,19], but also in geographical regions where it may emerge due to the presence of appropriate mosquito vectors [8][9][10][11].
Herein, sensitive fluorescent probe-based reverse transcription-quantitative polymerase chain reaction (RT-qPCR) and reverse-transcription-recombinase polymerase amplification (RT-RPA) assays have been developed and evaluated for rapid and specific detection of WSLV. The new methods were further evaluated with clinical and veterinary samples and are suitable for use in routine laboratory diagnosis of WSLV infection in both humans and animals and for field diagnosis in limited-resource settings during entomological or veterinary surveillance.

Ethical Statement
Mosquito pools used in this study were collected in the frame of the national integrated surveillance program for arbovirus in Senegal while archived clinical specimens were provided by the WHO Collaborating Centre for Arboviruses and Hemorrhagic Fevers in Institute Pasteur in Dakar, accredited for routine diagnostic, surveillance and animal research, according to IACUC anima [26]. The Senegalese national ethical committee approved the protocol as a less than minimal risk research, and written consent forms were not required. All viral isolations in suckling mice were performed in accordance with the ARRIVE guidelines [27].

Virus Stocks and Samples Collection
A selection of twenty-eight WSLV stocks previously prepared by inoculating Aedes albopictus continuous cell lines (C6-36) for 4 days, followed by a specific immunofluorescence assay (IFA) as previously described [28], were obtained from the collection of the WHO collaborating centre for arboviruses and viral hemorrhagic fevers (CRORA) at the Institute Pasteur of Dakar (IPD) in Senegal. Filtrated virus stocks were passed one time by intracerebral inoculation of newborn Swiss suckling mice (1-2 days old) in the animal laboratory at the IPD. Brain tissues from febrile mice were homogenized in L-15 medium (Gibco BRL, Grand Island, NY, USA) and tested by RT-qPCR as previously described [29]. In addition, eleven additional virus stocks representing seven other flaviviruses important in the public health context in Africa were also used for specificity assessment of the new assays (Table A1). In addition, a total of 50 pan-flavivirus negative mosquito pools and 20 pan-flavivirus negative human sera, collected in the field in Senegal from January to December 2019, were used for assessment of the diagnostic specificity. WSLV positive sera from two patients and brain tissues from a black rat previously collected from the field in Senegal [6] were also tested for assessment of clinical and veterinary sensitivity using both assays.

Primers Design
Multiple alignments of WSLV nucleotide sequences available online (www.ncbi.nlm. nih.gov/genbank/; last accessed on 11 August 2021) were carried out by using Muscle algorithm [30] within Unipro UGENE software [31]. Both primers and inverse-sense TaqMan probes were designed on the NS3 protein using Primer3web ® software (version 4.0.0, Whitehead Institute for Biomedical Research, Cambridge, MA, USA). According to RPA guidelines from TwistDx (Cambridge, UK), two forward primers, two reverse primers and one inverse-sense fluorescent exo probe were designed manually based on the conserved region of the envelope protein (E). To avoid non-specific cross-reactions with other flaviviruses, primers and probes were validated by BLAST analysis on NCBI (https://blast.ncbi.nlm.nih.gov//; last accessed on 15 August 2021) and oligonucleotides were produced by TIB Mol-Biol (Berlin, Germany). Four primer combinations were tested to select the RPA primers and probe set exhibiting the highest analytical sensitivity with 10 3 molecules of the in vitro RNA standard.
Plasmids were generated at GenExpress (Berlin, Germany) by inserting the ligated E, NS3 and NS5 amplicons into pCRII (Life Technologies, GmbH, Darmstadt, Germany) and the in vitro RNA standard was synthetized with T7 RNA-polymerase by TIBMolBiol (Berlin, Germany) at a concentration of 10 8 RNA molecules/reaction, according to the manufacturer's recommendations. The combination with the highest and earliest start of the exponential amplification curve was selected and tested with 10-fold serial dilutions (from 10 3 to 10 RNA molecules/reaction) of the in vitro RNA standard. The RT-qPCR primers were also tested using the same in vitro RNA standard.

Samples Preparation and RNA Extraction
All virus stocks, negative sera and negative mosquito pools analyzed in this study derived from collection of the WHO Collaborating Centre for Arboviruses and Hemorrhagic fevers (CRORA) in Senegal at Institute Pasteur of Dakar (IPD). Extraction of viral RNA from 100 µL of virus stocks or ten-fold serial dilutions of virus-spiked L15-medium, serum and CSF sample was performed with the QIAamp viral RNA mini kit (Qiagen, Heiden, Germany) according to the manufacturer's instructions. Viral RNA was eluted in a final volume of 60 µL and frozen at −80 • C prior to downstream applications.

Real-Time RT-qPCR Conditions
Real-time RT-qPCR was performed in duplicates using the qScript One-step qRT-PCR Kit (Quanta Biosciences, Gaithersburg, MD, USA) in a final volume of 25 µL following the previously established protocol [32] and the reaction was carried out on a 7500 Fast Real-Time system cycler (Applied Biosystems, Foster City, CA, USA).

RPA Assay Conditions
RT-RPA amplifications were achieved in a final volume of 50 µL by adding 0.2 µL of the SuperScript II Reverse Transcriptase enzyme (Invitrogen) to the TwistAmp exo kit (TwistDx, Cambridge, UK) and reduction of the volume of water in the master mix as previously described [33]. For each reaction, 45 µL of master mix was prepared in each tube lid and 5 µL of viral RNA was subsequently added to each lid tube. Then, the lids were closed carefully and the master mix was centrifuged into the rehydrated reaction pellet containing a dried enzyme using a mini spin centrifuge. The reaction was performed at 42 • C for 15 min in the Twista Tubescanner device (TwistDx, Cambridge, UK) connected to a computer for real-time monitoring of fluorescence as previously described [34].

Specificity Assessment
In order to evaluate the analytical specificity of the newly developed WSLV RT-qPCR and RT-RPA, RNA samples from 28 WSLV positive isolates and 11 isolates of other flaviviruses were tested in duplicates using both assays. Positive and negative controls containing positive RNA and nuclease-free water, respectively, were included in each run. In addition, a total of 50 pan-flavivirus negative mosquito pools and 20 pan-flavivirus negative human sera were also analyzed in duplicates using both assays for determination of the diagnostic specificity. A previously described pan-flavivirus RT-qPCR assay was used as a reference test [29]. RNA standard (10 8 -1 RNA molecules/reaction). A linear regression analysis and a semilog regression analysis were performed for the RT-qPCR assay and the RT-RPA assay, respectively. The probit regression analysis was also performed using data of eight runs from both assays to determine the limit of detection at 95% probability. The graphs were plotted using PRISM (GraphPad Software Inc., San Diego, CA, USA).

Sensitivity in Human Serum and L-15 Medium
Ten-fold serial dilutions of two virus stocks with known titer (pfu/mL) were prepared in L-15 medium and human serum (Sigma-Aldrich, Saint-Louis, MO, USA) and analyzed in triplicates using both assays to determine diagnostic sensitivity of the newly developed assays. Regression curves were obtained representing the pfu/reaction versus the threshold cycle (Ct) and threshold time (Tt) values for the RT-qPCR and the RT-RPA, respectively. The lowest titer with amplification was considered as the analytical limit of detection (LOD). The amplification efficiency was calculated for both assays from the slope of the pfu/reaction regression lines (E = 10 1/slope -1).
In addition, extracted RNA from a virus stock with a titer was analyzed eight times in the same run and in eight different runs to determine intra-assay and inter-assay coefficients of variation (CV).

Sensitivity on Clinical and Veterinary Samples
Extracted RNA from two WSLV positive human sera and brain tissues from one rodent isolated in Senegal were tested in duplicates for confirmation of the reliability of the newly established assays. The pan-flavivirus RT-qPCR assay was used as a reference test [29].

Assessment of the Impact of Genetic Diversity on Assays
In silico analysis of both assays was performed again sequences of WSLV isolates from South Africa and Senegal obtained from GenBank (www.ncbi.nlm.nih.gov/genbank/; last accessed on 15 August 2021), using the MAFFT alignment algorithm implemented in the Unipro UGENE software [31].

Statistical Analysis
The limit of detection of the WSLV assays was calculated by performing a probit regression analysis on the data set of eight RPA assays using STATISTICA software (StatSoft, Hamburg, Germany) in order to determine the number of RNA molecules/reaction at 95% of probability. The correlation between regression curves obtained from serial dilutions in L-15 medium and human serum was determined for each assay using the Pearson correlation test where a coefficient of 1 represents a good correlation. A kappa test was used to compare detection performances of the newly established assays and the pan-flavivirus RT-qPCR assay [29], where the Cohen's kappa coefficient (k) represents a measure of the agreement between assays with a 95% confidence interval and a p < 0.05 is considered to be statistically significant. In addition, positive and negative predictive values (PPV and NPV), diagnostic sensitivity and specificity were calculated using standard formulas [35]. The PPVs and NPVs were compared between the RT-qPCR and the RT-RPA assays using Fisher's exact test; considering a p < 0.05 as statistically significant.

Primers Selection
In silico analysis of the primers and probes designed herein using BLAST exhibited no possible cross-reactivity with none-WSLV sequences on NCBI. The RT-qPCR primers (WS-BFOW/WSBREV) detected well the dilution 10 3 RNA molecules/reaction of the in vitro RNA standard. For the RT-RPA assay, the four combinations of the RPA primers were also screened with the 10 3 RNA molecules/reaction of the in vitro transcribed RNA standard. The combination with the highest and earliest start of the exponential amplification curve was selected and tested with 10-fold serial dilutions (from 10 3 to 10 RNA Microorganisms 2022, 10, 550 5 of 13 molecules/reaction) of the in vitro RNA standard. Fortunately, the primer pair (RF2/RR2) enabled detection down to 10 2 RNA molecules/reaction of the in vitro RNA standard. Therefore, it was selected for further assay validation (Table 1).

Analytical Specificity
All 28 WSLV positive isolates were detected using both assays while amplification was not observed for any other none-WSLV strain, resulting then in an analytical specificity of 100% for the newly established assays ( Table 2).

Analytical Sensitivity
The analytical sensitivity of the newly established Wesselsbron virus RT-qPCR and RT-RPA assays was determined with Ct and Tt data values from eight sets of ten-fold dilutions of the in vitro RNA standard ranging from 10 8 to 1 molecules/reaction. The WSLV RT-qPCR assay detected the in vitro RNA standard with the concentration from 10 8 to 10 molecules/reaction in all eight RPA runs and the concentration of 1 molecule/reaction in five runs (lmtest p = 3.427 × 10 −13 ) ( Figure 1A). However, the RT-RPA assay produced positive results until dilution of 100 molecules/reaction in eight replicates, while no amplification was observed in the tube containing ten and one molecule/reaction. Ten min is the maximum time needed to amplify as low as 100 RNA molecules by the RT-RPA assay ( Figure 1B). With these data sets, a probit regression analysis was performed and revealed a detection limit at 95% probability of 4 and 130 RNA molecules/reaction for the newly established RT-qPCR and RT-RPA assays, respectively ( Figure 1C,D). action in five runs (lmtest p = 3.427 x 10 -13 ) ( Figure 1A). However, the RT-RPA assay produced positive results until dilution of 100 molecules/reaction in eight replicates, while no amplification was observed in the tube containing ten and one molecule/reaction. Ten min is the maximum time needed to amplify as low as 100 RNA molecules by the RT-RPA assay ( Figure 1B). With these data sets, a probit regression analysis was performed and revealed a detection limit at 95% probability of 4 and 130 RNA molecules/reaction for the newly established RT-qPCR and RT-RPA assays, respectively ( Figure 1C,D). However, the RT-RPA assay produced positive results until 100 molecules/reaction between 3 and 10 min. The probit regression analysis was performed using data of eight RT-qPCR assay runs (C) and eight RPA assay runs (D). The graphs were plotted using PRISM (GraphPad Software Inc., San Diego, CA, USA) and the limit of detection at 95% probability is depicted by the red triangle. The limit of detection at 95% probability is of 4 and 130 RNA molecules/reaction for the RT-qPCR and RT-RPA assays, respectively.

Diagnostic Performances
All the 50 mosquito pools and 20 human sera which tested negative for the pan-flavivirus RT-qPCR, also tested negative for both new WSLV assays. However, all the 31 positive samples including 28 WSLV viral stocks, 2 human sera and 1 rodent brain tissues, were detected using both diagnostic tools in a mean Tt value of 43.31 ± 0.18 min (ranging from 31.99 to 51.12), 37.74 ± 0.16 min (ranging from 31.92 to 46.73) and 4.55 ± 0.25 min (ranging from 3.21 to 5.78) for the pan-flavivirus RT-qPCR, the WSLV RT-qPCR and the WSLV RT-RPA, respectively (Table 2). Using the standard formulas [26], NPVs and PPVs However, the RT-RPA assay produced positive results until 100 molecules/reaction between 3 and 10 min. The probit regression analysis was performed using data of eight RT-qPCR assay runs (C) and eight RPA assay runs (D). The graphs were plotted using PRISM (GraphPad Software Inc., San Diego, CA, USA) and the limit of detection at 95% probability is depicted by the red triangle. The limit of detection at 95% probability is of 4 and 130 RNA molecules/reaction for the RT-qPCR and RT-RPA assays, respectively.

Diagnostic Performances
All the 50 mosquito pools and 20 human sera which tested negative for the panflavivirus RT-qPCR, also tested negative for both new WSLV assays. However, all the 31 positive samples including 28 WSLV viral stocks, 2 human sera and 1 rodent brain tissues, were detected using both diagnostic tools in a mean Tt value of 43.31 ± 0.18 min (ranging from 31.99 to 51.12), 37.74 ± 0.16 min (ranging from 31.92 to 46.73) and 4.55 ± 0.25 min (ranging from 3.21 to 5.78) for the pan-flavivirus RT-qPCR, the WSLV RT-qPCR and the WSLV RT-RPA, respectively (Table 2). Using the standard formulas [26], NPVs and PPVs of 1 and an accuracy of 100% (95% CI; 96.41-100%) were determined for both newly established RT-qPCR and RT-RPA assays (Fisher's exact test p < 0.0001). In addition, a diagnostic specificity of 100% (95% CI; 94.87-100%) and a diagnostic sensitivity of 100% (95% CI; 88.78-100%) were determined for both WSLV RT-qPCR and RT-RPA assays. Both new assays also showed good agreement with the real-time pan-flavivirus RT-qPCR [26] used as reference test (Cohen's Kappa test, k = 1 ± 0.09 (95% CI; 0.80-1.20); p < 0.05), indicating that the newly established WSLV assays are accurate and give 100% concordance to results obtained with the real-time pan-flavivirus RT-qPCR [29] on the same samples (p < 0.0001). In addition, intra-run and inter-run CV of 0.009 and 0.007, respectively, were found for the newly established RT-qPCR assay while the RT-RPA assay showed intra-run and inter-run CV of 0.002 and 0.001, respectively, indicating that these assays are highly repeatable.

Sensitivity in Human Serum and L-15 Medium
Ten-fold dilutions of two WSLV stocks prepared in human serum and L-15 medium were analyzed in triplicates using both the WSLV-specific RT-qPCR and RT-RPA assays. The new RT-qPCR assay yielded a sensitivity down to 100 pfu/reaction in both human serum and L-15 medium within a mean Tt of 47.63 ± 4.82 min (40.12-55.34) and 44.39 ± 4.57 min (37.15-52.74), respectively, while the new RT-RPA assay exhibited an LOD of 100 pfu/reaction in both human serum and L-15 medium within a mean Tt of 6.81 ± 0.12 min and 7.08 ± 0.52 min, respectively. Efficiencies of 132% and 133% were shown by the new RT-qPCR assay for dilutions in human serum and dilutions in L-15 medium, respectively, while the new RT-RPA assay exhibited efficiencies of 101% and 100%, respectively (Figure 2A,B). Pearson's correlation coefficients of 0.9939 (p = 5.493 × 10 −5 ) and 0.9967 (p = 1.606 × 10 −5 ) were also determined between regression lines from dilutions in human serum and L-15 medium tested with the new RT-qPCR and RT-RPA assays, respectively. Both assays detected until 100 pfu, corresponding to 100 RNA molecules calculated from Ct values and the equation obtained from the linear regression analysis of eight RT-qPCR data sets of ten-fold dilutions of the molecular standard RNA ( Figure 2C). run and inter-run CV of 0.002 and 0.001, respectively, indicating that these assays are highly repeatable.

Sensitivity in Human Serum and L-15 Medium
Ten-fold dilutions of two WSLV stocks prepared in human serum and L-15 medium were analyzed in triplicates using both the WSLV-specific RT-qPCR and RT-RPA assays. The new RT-qPCR assay yielded a sensitivity down to 100 pfu/reaction in both human serum and L-15 medium within a mean Tt of 47.63 ± 4.82 min (40.12-55.34) and 44.39 ± 4.57 min (37.15-52.74), respectively, while the new RT-RPA assay exhibited an LOD of 100 pfu/reaction in both human serum and L-15 medium within a mean Tt of 6.81 ± 0.12 min and 7.08 ± 0.52 min, respectively. Efficiencies of 132% and 133% were shown by the new RT-qPCR assay for dilutions in human serum and dilutions in L-15 medium, respectively, while the new RT-RPA assay exhibited efficiencies of 101% and 100%, respectively ( Figure  2A,B). Pearson's correlation coefficients of 0.9939 (p = 5.493 x 10 -05 ) and 0.9967 (p = 1.606 x 10 -05 ) were also determined between regression lines from dilutions in human serum and L-15 medium tested with the new RT-qPCR and RT-RPA assays, respectively. Both assays detected until 100 pfu, corresponding to 100 RNA molecules calculated from Ct values and the equation obtained from the linear regression analysis of eight RT-qPCR data sets of ten-fold dilutions of the molecular standard RNA ( Figure 2C).  . Dilutions were tested in triplicates using both assays. Both assays detected until 100 pfu, corresponding to 100 RNA molecules using the equation obtained in Figure 1A (C).

Sensitivity in Clinical and Veterinary Samples
All collected samples were screened in duplicates with both assays. As the reference pan-flavivirus RT-qPCR method, all three positive specimens tested positive with both WSLV assays. The WSLV-specific RT-qPCR detected the RNA in both human sera and rodent's brain tissues in a mean Tt of 44.51 ± 0.51 min (43.78-45.88), while the RT-RPA assay gave amplification with these samples in a mean time of 7.46 ± 0.15 min (7.30-7.60); indicating that these assays are highly sensitive on clinical and veterinary samples and six times faster than the RT-qPCR (Table 3).

In Silico Analysis of New Primers and Probes Sequences
Available coding-complete sequences from South Africa (SA) and Senegal (SN) enabled in silico evaluation of the newly developed WSLV-specific assays using the BLAST program (https://blast.ncbi.nlm.nih.gov/, accessed on 1 November 2021). A total of 7 and 6 mismatches between the targeted WSLV gene sequence and the currently available sequences have been identified for the new RT-qPCR and RT-RPA assays, respectively. The forward and reverse primers of the RT-qPCR assay (WSBFOW and WSBREV) are highly similar to the target region of the sequences from Senegal, while they show a dissimilarity of 0% and 5% against the isolates from South Africa. The RT-qPCR assay's probe (WS-BPROBE) reveals a dissimilarity of 5% to sequences from Senegal, while it is more distant to the isolate EU707555_SAH177_SA_1955 from South Africa with a dissimilarity of 15% (Table A2; Figure A1). The forward and reverse primers of the RT-RPA assay (RF2 and RR2) were also highly similar to the sequences from Senegal, while the reverse primer reveals dissimilarities of 3%, 3% and 6% to the isolates from South Africa. The probe of the RT-RPA assay (exoProbe) shows a dissimilarity of 2% to the Senegalese sequences. However, it is distant from the isolates from South Africa with dissimilarities of 2%, 2% and 8% (Table A2; Figure A1). Nevertheless, these dissimilarities did not omit detection of the aligned sequences from Senegal by the new WSLV assays (Table 3).

Discussion
WSL is a neglected, mosquito-borne infection reported in Africa [6,14] with symptoms close to those caused by Rift Valley Fever infection in animals [14]. In addition, the geographic distribution of WSL remains sparsely determined [6,7,14,19]. Rare data are currently available from Africa [13] where resources, infrastructure and diagnostic capacities are limited. In addition, the clinical diagnosis of WSLV infection is difficult as it is at least always confused with the clinical presentation of Rift Valley Fever disease. Therefore, we developed two real-time molecular assays for rapid detection of WSLV. These assays are based on conserved regions on NS3 and E Wesselsbron virus proteins.
The newly established assays are highly specific for WSLV detection and showed good agreement with the pan-flavivirus RT-qPCR [29]. In addition, our WSLV assays showed excellent diagnostic performances, which indicates that they are highly sensitive, similar to the pan-flavivirus RT-qPCR [29]. The newly established RT-RPA assay is much faster than the pan-flavivirus RT-qPCR [29] and needs less than 6 min for WSLV detection in all isolates that tested positive. Both protocols require extraction and thus actual time saved could be a lot less significant. However, the RT-RPA assay could be used with rapid extraction methods which had previously shown best performances in extraction of viral RNA [36] and were applied as the first-line assay in low-resource settings.
The recorded inter-assay and intra-assay variability below 1% also ensures high repeatability for the two new assays. A total of 7 and 6 mismatches between the targeted WSLV gene sequence and the currently available sequences were identified for the new RT-qPCR and RT-RPA assays, respectively. However these mismatches are mostly bases belonging to the same nucleotide category and are not present at positions critical for binding of the oligonucleotides [37]. However, the performances of the new WSLV-specific assays could be further assessed on isolates from South Africa.
Our WSLV RT-qPCR assay could be applied in well-equipped central laboratories for routine WSLV diagnosis as complementary to an existing method [38] while the RT-RPA assay showed more advantages in remote areas where highly equipped laboratories are not available, since it uses cold-chain independent reagents and is stable under different environmental conditions including temperatures above 30 • C [39]. However, our RT-RPA assay is about 50 times less sensitive than the RT-qPCR. Thus, in order to prevent risks of misdiagnosis for patients, the RT-qPCR assay could also be applied as a confirmation method on samples that were tested negative by RT-RPA assay. Our WSLV-specific assays appear then as appropriate tools for the point-of-need detection of acute WSLV cases as they are highly sensitive and cover the currently known diversity of WSLV in West Africa [6].
Although no fatal human case has been identified yet, these new tools could be used not only in a context of other investigations in areas where the virus circulation has been previously reported [7,19], but also in active entomological, veterinary and clinical surveillance studies ruled out in regions where the Aedes vectors have been previously identified [8][9][10][11]40], in order to determine its viral pathogenesis, identify its potential reservoirs and estimate the real public and veterinary health impact of this disease, at least in Africa.  Acknowledgments: We acknowledge colleagues in the virology department at Institut Pasteur de Dakar, Senegal for sharing supportive information necessary for establishment and accomplishment of this study.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.   Figure A1. In silico analysis of (A) the newly established RT-qPCR assay and (B) the RT-RPA assay against sequences of Wesselsbron virus isolates from South Africa (SA) and Senegal (SN), respectively. The alignment was performed using the MAFFT algorithm implemented in the Unipro UGENE software [31]. The in silico analysis revealed more signature erosion in the RT-qPCR primers when compared to sequences from South Africa. Figure A1. In silico analysis of (A) the newly established RT-qPCR assay and (B) the RT-RPA assay against sequences of Wesselsbron virus isolates from South Africa (SA) and Senegal (SN), respectively. The alignment was performed using the MAFFT algorithm implemented in the Unipro UGENE software [31]. The in silico analysis revealed more signature erosion in the RT-qPCR primers when compared to sequences from South Africa.