First detection of Jingmen tick virus in Corsica, France and development of a real time detection system for multiple tick-associated jingmenviruses

Jingmen tick virus (JMTV) is a recently discovered segmented RNA virus, closely related to flaviviruses. It was identified for the first time in 2014, in China and subsequently in Brazil. Following this discovery, JMTV-related sequences have been identified in arthropods, vertebrates (including humans), plants, fungus and environmental samples from Asia, America, Africa, Europe and Oceania. Several studies suggest an association between these segmented flavi-like viruses, termed jingmenviruses, and febrile illness in humans. The development of rapid diagnostic assays for these viruses is therefore crucial to be prepared for a potential epidemic, for the early detection of these viruses via vector surveillance or hospital diagnosis. In this study, we designed a RT-qPCR assay to detect tick-associated jingmenviruses, validated it and tested its range and limit of detection with six tick-associated jingmenviruses using in vitro transcripts. Then we screened ticks collected in Corsica (France) from different livestock species, in order to determine the distribution of these viruses on the island. In total, 6,269 ticks from eight species were collected from 763 cattle, 538 horses, 106 sheep and 218 wild boars and grouped in 1,715 pools. We report the first detection of JMTV in Corsica, in Rhipicephalus bursa, Hyalomma marginatum and R. sanguineus ticks collected from cattle and sheep. The highest prevalence was found in the Rhipicephalus genus. The complete genome of a Corsican JMTV was obtained from a pool of Rhipicephalus bursa ticks and shares between 94.7% and 95.1% nucleotide identity with a JMTV sequence corresponding to a human patient in Kosovo and groups phylogenetically with European JMTV strains. These results show that a Mediterranean island such as Corsica could act as a sentinel zone for future epidemics.


Introduction
Jingmen tick virus (JMTV) is a segmented RNA virus which was identi ed for the rst time in 2014 in ticks from the Jingmen region of the Hubei province in China and simultaneously from the Mogiana region of Brazil 1,2 .The genome comprises four segments of positive-sense single-stranded RNA.Segments 1 and 3 encode nonstructural proteins, genetically and functionally close to the non-structural proteins NS3 and NS5 of members of the Flavivirus genus in the Flaviviridae family.Segments 2 and 4 encode putative structural proteins, which are not as closely related to avivirus structural proteins as the non-structural proteins seem to be 2 .Following this discovery, JMTV RNA was detected in arthropods (including ticks within the Rhipicephalus, Haemaphysalis, Ixodes, Dermacentor, Amblyomma, Hyalomma genera), reptiles and mammals (including cattle and humans) from all continents, alongside other related segmented avi-like virus sequences termed jingmenviruses [1][2][3][4][5][6][7] .The jingmenvirus sequences group phylogenetically into two clades: tick-associated jingmenviruses (also found in vertebrates), and sequences detected from insects (type species: Guaico Culex virus, from Culex mosquitoes), crustaceans, plants and fungi 8 .Two tick-associated jingmenviruses have been found in Europe: JMTV and Alongshan virus (ALSV).JMTV has been detected in humans from Kosovo, in eld-collected ticks from Türkiye and Romania, and in an Aedes albopictus mosquito laboratory colony in Italy, while ALSV was found in ticks from Finland, France, Germany and Switzerland 3,[9][10][11][12][13][14][15][16] .Concurrently, these are the only two jingmenviruses which have been found in humans: JMTV and ALSV-derived RNA and antibodies were detected patients with tick bites and febrile illness in China and Kosovo 5,9,17 .
To our knowledge, no analytically validated molecular assay is available for tick-associated jingmenviruses.
Such an assay would allow to shed some light on the vector competence of different tick species and help formally elucidate the transmission cycle of jingmenviruses, which would in term facilitate the prediction of virus introduction into new territories.The aim of this study was to design and validate a real-time RT-PCR assay for the detection of tick-associated jingmenviruses, and to generate epidemiological data in Corsica, a French Merditerranean island, by screening ticks collected from farmed and wild animals on the island.We chose Corsica because: i) many genera of ticks, such as Ixodes, Hyalomma, Dermacentor, Haemaphysalis, Rhipicephalus and Amblyomma, are present [23][24][25] and ii) the circulation of zoonotic diseases in Corsica is facilitated by the widespread practice of mixed livestock farming, the presence of avian migration corridors, and strong interactions between livestock, wildlife and human populations 23,26 .

Molecular detection system design
We aligned all sequences of JMTV, ALSV, YGTV available at the time of design (14th of June 2021) from the taxonomy browser of the National Center of Biotechnology Information (NCBI) using the L-INS-I algorithm implemented in MAFFT version 7. Two real-time RT-qPCR assays were designed targeting conserved regions of segments 1 and 2 based on the generated alignment (Table 1).In particular, segment 2 was selected due to its lack of homology with other virus sequences, in an effort to prevent non-speci c ampli cation.

Molecular detection system selection
To ensure that the RT-qPCR detection systems targeting segments 1 and 2 were functional, a 0.1 µg/µL DNA plasmid synthetized by Geneart (ThermoFisher) containing the regions targeted by the systems in JMTV was serially diluted 10-fold from 100 pg/µL to 0.1 pg/µL and used as template in a qPCR assay (Superscript III, ThermoFisher).

Generation of in vitro transcribed RNAs
In vitro transcripts (IVT) were generated as positive controls and to evaluate the sensitivity and range of detection of the RT-qPCR systems.Plasmids were synthetized by Geneart containing the region of segment 2 ampli ed by the gTJ-seg2 system, for six jingmenviruses: JMTV (MH133315), ALSV (MN095520), YGTV (MH688530), TAKV (LC628181), PLJV (MN095532) and GJLV (MW896894) (sequences available in Supplementary Table S1).An exogenic NotI hybridization sequence was incorporated to detect potential laboratory contamination, and an actin sequence was also included as another target to quantify the IVT (see Supplementary Table S1).The plasmids were in vitro transcribed into RNA using Megashortscript™ T7 transcription kit (Invitrogen-Thermo Fisher) with Turbo DNase to remove DNA.RNA transcripts were puri ed using MEGAclear™ Puri cation of transcription reaction kit (Invitrogen-Thermo Fisher Scienti c).A Thermo Scienti c™ NanoDrop™ was used to determine the RNA concentration in ng/µL, converted to RNA copies/µL using the molecular weight of the molecule, calculated with the AAT Bioquest calculator online.
Determining the range and limit of detection of the gTJ-seg2 system The six IVTs were serially diluted 1:3 with dilutions ranging from 10 7 RNA copies/µL to 15 RNA copies/µL, and used as templates in RT-qPCR using gTJ-seg2, with 4 to 8 replicates.All IVTs were tested with the actin system to con rm their reactivity.The cycle threshold (Ct) values obtained were graphed relative to the IVT concentration in logarithmic scale.A Ct value > 40 was considered negative.IVTs negative for all dilutions under 10 7 RNA copies/ µL were tested further, with dilutions starting from 10 10 RNA copies/µL.
SPSS Statistics software version 24 (IBM) was used to determine the lower limit of detection (LOD), de ned as the lowest concentration of RNA achieving a 95% hit rate (LOD95).GraphPad Prism 9.4.1 was used to estimate a linear regression and 95% con dence intervals between the Ct values < 40 obtained with gTJ-seg2 and the RNA copies / µL.
Reactions were set up with 12.5 µL buffer, 0.25 µL RT enzyme, 0.5 µL Rox, 800 nM of each primer and 200 nM of probe for a nal volume of 25 µL.The optimal cycling conditions were: 50°C for 10 min; 95°C for 5 min; 45 cycles of 95°C for 10 sec and Tm for 30 sec, with Tm = 57°C for segment 1, Tm = 55°C for segment 2 and Tm = 60°C for actin.Probes were labeled with FAM dye and Tamra quencher.

Sample collection
Several types of vertebrate animals were inspected for adult ticks between August 2018 and June 2020, all over Corsica (Fig. All ticks were collected and kept alive until morphological identi cation and storage.Ticks were identi ed at species level under a stereomicroscope using an identi cation key, and immediately stored at − 80°C 27 .

Sample processing and nucleic acid extraction
Ticks collected from the animals were washed once in 70% ethanol and twice in distilled water, then were divided either individually or in monospeci c pools of 2 to 10 ticks, according to developmental stage, sex, and animal of collection.They were homogenized in Minimum Essential Medium (MEM) containing 15% of fetal bovine serum, antibiotics (1% penicillin-streptomycin, 1% kanamycin), fungicide (5% amphotericin B) and 1% L-glutamine, using a TissueLyser II (Qiagen, Hilden, Germany) at 30 Hz for 3 min.Nucleic acid extractions were performed on a QIAcube HT (Qiagen) using QIAamp cador Pathogen Mini kits, according to the manufacturer's instructions.Nucleic acid extracts were eluted in 100 µL buffer and stored at -80°C until they were used as templates in RT-qPCR.The extraction quality was monitored by systematically spiking MS2 bacteriophage and quantifying it by RT-qPCR.

Complete genome and RT-qPCR amplicon sequencing
One gTJ-seg2-positive pool containing 6 male R. bursa ticks (164BOV19) was selected to obtain a complete genome using next generation sequencing (NGS).A total of 200 µL homogenate supernatant was incubated at 37°C for 7 h with 25 U of Benzonase (Novagen) and MgCl 2 .RNA extraction was performed using the Viral RNA mini kit (Qiagen) on the BioRobot EZ1-XL Advanced (Qiagen).Random two-step RT-PCR was performed using tagged random primers in a ProtoScript® II Reverse Transcriptase (New England Biolabs) reaction followed by a Platinum® Taq High Fidelity polymerase (ThermoFisher Scienti c) reaction with speci c primers 28 .These samples and selected RT-q-PCR amplicons produced as described above using gTJ-seg2 were quanti ed using the Qubit® dsDNA HS Assay Kit and a Qubit 2.0 uorometer (ThermoFisher Scienti c).All amplicons were sonicated into 200 bp fragments and libraries were built and barcoded using AB Library Builder System (ThermoFisher Scienti c).The barcoded fragments were quanti ed by RT-qPCR using the Ion Library TaqMan™ Quantitation Kit (ThermoFisher Scienti c) and pooled equimolarly.An emulsion PCR of the pooled fragments was performed and the samples were loaded on an Ion 520 chip (ThermoFisher Scienti c) using an automated Ion Chef instrument (ThermoFisher Scienti c).Sequencing was performed using the S5 Ion torrent technology (ThermoFisher Scienti c) following the manufacturer's instructions.Reads were trimmed (reads with quality score < 0.99 or length < 100pb were removed, and the rst and last 30 nucleotides were removed from all reads) and de novo contigs were generated with CLC Genomics Workbench software v.21 (Qiagen).These contigs were blasted (in house BLASTn algorithm) to determine the best reference sequence, and reads were then mapped to that sequence.Parameters for reference-based assembly consisted of match score = 1, mismatch cost = 2, length fraction = 0.5, similarity fraction = 0.8, insertion cost = 3, and deletion cost = 3.

Phylogenetic analyses
The complete nucleotide sequences of JMTV Corsica 164BOV19 four genome segments obtained in this study were aligned with published tick-associated jingmenvirus sequences using the E-INS-I algorithm implemented in MAFFT version 7 29 .Phylogenetic analyses were inferred for each genomic segment, with maximum likelihood in Mega X, with 100 bootstraps, a general time reversible model gamma distributed 30 .

Isolation attempts of virus
A total of 10 µL of homogenized tick pool 164BOV19, diluted 1:10 with MEM was injected intracerebrally into 2day-old OF1 mice (n = 15).This corresponded to a dose of 2.3x10^8 RNA copies per animal.The mice were observed for 14 days for clinical signs of disease, then euthanized by cervical dislocation under general anaesthesia and brains were collected.Nucleic acids were puri ed from the brain tissues using QIAcube HT with QIAamp 96 Virus QIAcube HT Kit, and tested for presence of JMTV with the segment 2 RT-qPCR described above.
In vivo experiments were approved by the French 'Ministère de l'Enseignement Supérieur, de la Recherche et de l'Innovation' (APAFIS#9368) and performed in accordance with the French national guidelines and the European legislation covering the use of animals for scienti c purposes.All experiments were conducted in a BSL3 laboratory.

Molecular detection system selection
In order to ensure that the two systems designed in this study to detect tick-associated jingmenviruses were functional, the plasmid containing the region targeted by both systems was diluted from 100 pg/µL to 0.1 pg/µL and used as template in qPCR assays with gTJ-seg1 or gTJ-seg2.Both systems were indeed functional, gTJ-seg2 was more sensitive than gTJ-seg1 by 3 to 7 Ct (Table 2) and was therefore selected to be used for large-scale screening of all tick pools.Range and limit of detection of the gTJ-seg2 system Tripling dilutions of in vitro transcribed RNAs corresponding to increasingly divergent jingmenviruses in the tickassociated clade were used in a RT-qPCR to determine the range of virus sequences detected by gTJ-seg2.A RT-qPCR system detecting actin was used alongside as a positive control for all IVTs.
These results are in line with the in silico sequence analysis of these viruses over the region ampli ed by the segment 2 system (Fig. 3).Indeed, JMTV, YGTV and ALSV have either none or only one mismatch when comparing their sequence with the forward primer, the probe, or the reverse primer.In contrast, PLJV contains three mismatches with the probe and reverse primer, which could explain its higher limit of detection.TAKV presents six mismatches with the reverse primer which could explain why it cannot be ampli ed by gTJ-seg2.
The divergent GJLV cannot be ampli ed either using gTJ-seg2, as it has three mismatches in the forward primer, eight mismatches over the probe and ten mismatches over the reverse primer.
The overall adult male to female ratio in the collected ticks was 1.44.In cattle, that ratio was 1.72 (1,421 females, 2,452 males).

Detection of tick-associated jingmenvirus RNA
In the 1,715 pools tested, tick-associated jingmenvirus RNA was detected in 21 tick pools collected from three cattle and in one tick pool collected from a sheep with a minimum infection rate of 1.22 (MIR).
Only one of the 107 sampled sheep had a positive tick pool.That animal had 1/2 (50%) pools positive.
The highest prevalence was found in the Rhipicephalus genus with a MIR of 0.58% in R. bursa and 0.55% in R. sanguineus followed by H. marginatum 0.19%.The MIR of positive ticks collected in cattle was 0.54%, more speci cally 0.70% for R. bursa, 1.16% for R. sanguineus and 0.39% for H. marginatum.# Whole genome sequence obtained by random RNA NGS.* Con rmed to be JMTV by NGS of RT-qPCR amplicon.† Estimated using the linear regression formula obtained for JMTV (see Fig. 2).

Sequence and phylogenetic analyses
We obtained the full genome sequence of JMTV Corsica 164BOV19 from a pool of 6 male R. bursa ticks collected from cattle in 2020 (bolded in Table 4) (Genbank accession numbers PP275067-PP275070).The segment sequences shared approximately 95% nucleotide identity (94.0-95.3%depending on the strains and segments) with other European sequences of JMTV (clade II), detected in humans from Kosovo (MH133321-MH133324), and in R. bursa from Romania (MW561147-MW561150) and Türkiye (MN486256, MN486261, MN486264, MN486267).JMTV Corsica 164BOV19 was more distantly related (84.4-90.2%depending on the strains and segments) to three other clade II strains, detected in ticks from the French Antilles and Trinidad and Tobago or in a mosquito colony from Italy.JMTV Corsica 164BOV19 is even more distantly related (77.2-82.6% depending on the strains and segments) to sequences from clade I, of Asian, South American and African origins (Table 5).We obtained partial sequences for 15 positive samples, by NGS of RT-PCR amplicons; all were 100% identical at the nucleotide level to the JMTV sequence obtained for 164BOV19.The JMTV Corsica 164BOV19 sequences were aligned with published jingmenvirus sequences and phylogenetic trees were built for each segment (Fig. 4).We found that JMTV Corsica 164BOV19 was most closely related to European and Caribbean JMTV sequences (clade II), separately from Asian, South American, and African JMTV strains, which form another clade (clade I).

Virus isolation
We attempted the isolation of JMTV in new-born mice by inoculating JMTV positive tick homogenate intracranially in two to three days old OF1 mice.No clinical signs were observed in the inoculated mice for two weeks.
While JMTV RNA was detectable in the inoculum, viral RNA was not detected in the mice brains fourteen days after intra-cranial inoculation using gTJ-seg2 in RT-qPCR, suggesting the lack of virus propagation.

Discussion
In this study, we designed and characterized two RT-qPCR systems (gTJ-seg1 and gTJ-seg2) targeting segments 1 and 2 of the three tick-associated jingmenviruses published at the time (JMTV, ALSV and YGTV) and we report evidence of circulation of JMTV in ticks collected from cattle in Corsica.
Segment 2 was chosen as a target for the detection and quanti cation of tick-associated jingmenviruses due to its relatively high homology between target species, the absence of homology with other viruses or living beings 1 and the lower limit of detection of gTJ-seg2 compared to gTJ-seg1.We showed that gTJ-seg2 can detect several jingmenvirus sequences (JMTV, YGTV, ALSV and PLJV), all in the tick-associated jingmenvirus group.While some RT-qPCR assays have been developed to detect single species of jingmenviruses, to our knowledge, gTJ-seg2 is the rst RT-qPCR assay to detect all aforementioned tick-associated jingmenviruses 31 .
Therefore, gTJ-seg2 assay was used to screen ticks, belonging to eight species, collected in Corsica, and successfully detected JMTV in 22 pools of R. bursa, H. marginatum and R. sanguineus, with the highest JMTV detection rate found in R. bursa.JMTV has previously been detected in these tick species from other regions: R. bursa from Türkiye and Romania, in H. marginatum from Türkiye and in R. sanguineus from Türkiye and China 1,10,11,32 .The varying tick numbers and different detection methods prevent direct comparisons with these studies.Taken together, published data to date and our ndings seem to indicate an association between JMTV and Rhipicephalus ticks, while JMTV can be found in other tick genera, albeit in lower prevalence 10,11 .
The RNA loads detected in Corsican ticks were higher than the limit of detection for JMTV, YGTV and ALSV by a few orders of magnitude (up to 10 7 RNA copies/µL detected vs a limit of detection around 10 2 RNA copies/uL) suggesting that YGTV and ALSV would have been detected in those ticks, had they been present.PLJV might have been detected (limit of detection around 10 4 RNA copies/µL) if present, although this sequence has only ever been detected in bat-derived samples.However, TAKV would not have been detected by our system and would require screening using a virus-speci c detection system or a newly developed generic system, based on alignments including this more recently discovered sequence.
We were not successful in isolating JMTV Corsica from new-born mice, despite inoculating each animal with > 10 8 viral RNA copies.JMTV isolation was previously reported as challenging in both new-born mice and cell culture 1,5 .Elucidating the mechanisms preventing JMTV propagation in laboratory models would allow for a better understanding of viral replication and host-restriction mechanisms and would be crucial in case they emerge as human or veterinary pathogens.
Many questions remain unanswered regarding the natural life cycle of tick-associated jingmenviruses, but there is mounting evidence that they are arthropod-borne viruses.Jingmenviruses have been detected both in male, female, engorged and non-engorged adult ticks as well as non-engorged larvae, which can be evidence for transovarian transmission 18,33 .Our cattle-harvested JMTV-positive ticks corresponded to three animals, and these animals were in contact with up to 13 positive tick pools; this may represent a sign of horizontal transmission from the cattle to the ticks, or of co-feeding transmission from tick to tick, as previously described for tick-borne aviviruses 3,34,35 .Another argument in favour of the horizontal transmission hypothesis is the presence of jingmenviruses in the salivary glands of ticks and in association with vertebrates, including humans presenting tick bites and febrile illness symptoms, monkeys, rodents, bats, pigs, sheep, cattle and tortoises 5,17 .Moreover, there appears to be a lack of correlation between jingmenvirus sequences and their respective hosts of origin.
Instead, phylogenetic clades and geographical distribution are notably associated 18,36 .This pattern is prominently evident in our phylogenetic analyses, incorporating JMTV Corsica within the Caribbean-European JMTV clade (clade II), distinct from the African-Asian-South American clade (clade I).Together, these data align with the hypothesis that jingmenviruses can be transmitted horizontally between ticks and vertebrate hosts.
In light of the strong emerging potential of tick-associated jingmenviruses, the development of a molecular detection tool is important to improve the knowledge on virus circulation by implementing surveillance on vectors and reservoirs, as well as to encourage physicians to consider jingmenvirus infection in patients with undiagnosed febrile illness.Whether the gTJ-seg2 real-time molecular assay, developed in our study, is sensitive enough to be suited for diagnostics in clinical samples from human patients remains to be investigated.The need for serological techniques must also be raised.To the best of our knowledge, we provide here the rst evidence of JMTV circulation in Corsica and more widely in the West Mediterranean region, and the rst RT-qPCR system validated for the detection of multiple tick-associated jingmenviruses.

Declarations Figures
Page 18/  Colour coded by vertebrate host.
License:   This work is licensed under a Creative Commons Attribution 4.0 International License.Read Full License Additional Declarations: No competing interests reported.
1); 763 bovine cattle from various farms were inspected from January 2019 to June 2020 at the main active slaughterhouse in Corsica, in Ponte-Leccia and; 657 horses were inspected several times between March and August 2019, and in May and June 2020, after each riding excursion in the natural environment across Corsica.Ticks were collected from 218 wild boars in the northeast of Corsica from August to December (hunting season) in 2018 and 2019.A total of 107 sheep were inspected monthly for ticks in May and in June 2020 in a farm located in Corte (Fig. 1).

Figure 1 Map
Figure 1

Figure 2 Range
Figure 2

Figure 3 Multiple
Figure 3

Table 1
Nucleotide sequences and amplicon size of two generic tick-associated jingmenvirus (gTJ) RT-qPCR systems designed on multiple sequence alignments of segments 1 and 2.

Table 2
Cycle thresholds (Ct) obtained with gTJ-seg1 and gTJ-seg2 on serial dilutions of plasmids containing both systems' targets in JMTV.

Table 3
Ticks collected from farmed and wild animals by tick species and host species.

Table 5
Percentage nucleotide identity over the open reading frames of the four genomic segments of selected published JMTV strains and JMTV Corsica (164BOV19, Genbank accession numbers PP275067-PP275070).