Diverse novel phleboviruses in sandflies from the Panama Canal area, Central Panama

The genus Phlebovirus (order Bunyavirales , family Phenuiviridae ) comprises 57 viruses that are grouped into nine species-complexes. Sandfly-transmitted phleboviruses are found in Europe, Africa and the Americas and are responsible for febrile illness and infections of the nervous system in humans. The aim of this study was to assess the genetic diversity of sandfly-transmitted phleboviruses in connected and isolated forest habitats throughout the Panama Canal area in Central Panama. In total, we collected 13 807 sandflies comprising eight phlebotomine species. We detected several strains pertaining to five previously unknown viruses showing maximum pairwise identities of 45–78 % to the RNA-dependent RNA polymerase genes of phleboviruses. Entire coding regions were directly sequenced from infected sandflies as virus isolation in cell culture was not successful. The viruses were tentatively named La Gloria virus (LAGV), Mona Grita virus (MOGV), Peña Blanca virus (PEBV), Tico virus (TICV) and Tres Almendras virus (TRAV). Inferred phylogenies and p -distance-based analyses revealed that PEBV groups with the Bujaru phlebovirus species-complex, TRAV with the Candiru phlebovirus species-complex and MOGV belongs to the proposed Icoarci phlebovirus species-complex, whereas LAGV and TICV seem to be distant members of the Bujaru phlebovirus species-complex. No specific vector or habitat association was found for any of the five viruses. Relative abundance of sandflies was similar over habitat types. Our study shows that blood-feeding insects originating from remote and biodiverse habitats harbour multiple previously unknown phleboviruses. These viruses should be included in future surveillance studies to assess their geographic distribution and to elucidate if these viruses cause symptoms of disease in animals or humans.


InTRoduCTIon
Neotropical sandflies are a diverse group of insects in the order Diptera, family Psychodidae, subfamily Phlebotominae, which has undergone notable taxonomic revision recently [1].Historically, phlebotomine sandflies have been recognized as vectors of various disease-causing pathogens to humans [2].In Panama, anthropophilic species of at least six species of phlebotomine sandflies have been implicated as vectors of Leishmania parasites, the causing agent of American Cutaneous Leishmaniasis (ACL) [3][4][5], and sandflies of some of these taxa are also suspected of transmitting phleboviruses to a broad range of animals and humans [6][7][8][9].Sandflies of the six species Psychodopygus panamensis, Nyssomyia trapidoi, Lutzomyia gomezi, Nyssomyia ylephiletor, Lutzomyia sanguinaria and Psychodopygus thula are widespread across forested areas of Panama, feeding opportunistically on animals of various species depending on habitat quality, host availability and biomass [10].
Sandflies almost exclusively transmit viruses of the genus Phlebovirus (family Phenuiviridae, order Bunyavirales).According to the International Committee on Taxonomy of Viruses (ICTV) members of the genus Phlebovirus are classified into nine virus species that comprise 57 viruses [11].In addition, more than 15 putative phlebovirus species have been described that are waiting to be classified by the ICTV.The classified sandfly-transmitted viruses of the New World belong to the species-complexes Punta Toro, Candiru, Chilibre and Frijoles phlebovirus.Infections with sandfly-borne phleboviruses can cause unspecific symptoms in humans often misdiagnosed as dengue fever, malaria or influenza [7,12].Clinical symptoms are ranging from high fever, severe headache, muscle ache and aseptic meningitis to mild or severe meningoencephalitis [13].A recent study in Panama has shown that Punta Toro virus (PTV) causes febrile illness in humans with symptoms similar to infections with dengue viruses [12].Between 1964 and 2009 various members of the PTV species-complex were isolated from phlebotomine sandflies, humans and sentinel hamsters in Panama and Colombia [7].In addition, Chagres virus (CHGV) has been reported as being responsible for outbreaks of febrile illness in humans [14].Other phleboviruses detected in Panamanian sandflies in the 1950s are Aguacate, Cacao, Frijoles and Nique viruses [6,[15][16][17].More recent contributions to the expansion of the genetic diversity of phleboviruses of the New World have been provided by sequencing of viruses isolated from sandflies (Phlebotominae spp.), spiny rats (Proechimys sp.) and the Southern two-toed sloth (Choloepus brasiliensis) that was collected in Brazil between 1962 and 1985 [18], and 2014-2015 [19].It is unknown if these viruses are pathogenic for humans.The highest diversity of phleboviruses in Panama was detected in phlebotomine sandflies and human samples from the Central Panama region and the Darien province [7,20].Both regions show increasing levels of habitat fragmentation due to land-use change for cattle and crop farming [21,22].Despite their impact on human and veterinary health, phleboviruses have been insufficiently studied in Panama [7,12].
The aim of this study was to investigate the genetic diversity of phleboviruses in sandflies of forested areas of the Panama Canal area.The lowland tropical rainforest of the Panama Canal area harbours a large diversity of phlebotomine sandfly species of the New World.This region has a long history of arbovirus outbreaks.For example, during the construction of the Panama Canal at the end of the nineteenth century tens of thousands of people died from yellow fever [23].

Sandfly collection
Fieldwork was conducted in the Panama Canal area, Central Panama, between 2013 and 2014.Sandflies were collected in landscapes differing in their level of anthropogenic disturbance, (i) in continuous tropical lowland forest at the shore of Gatún Lake (continuous forests), (ii) in small forest fragments (1.5-51 ha) surrounded by agricultural fields (forest fragments) and (iii) in isolated forested islands (5.2-17.5 ha) (forested islands) that are remnants of the original continuous forest and emerged after flooding the Panama Canal in 1914.In each of the three landscape types, five replicate sites were selected resulting in 15 sampling sites in total [24].
Adult sandflies were collected using CDC (U.S. Centers for Disease Control) miniature light traps (John W. Hock Company, Gainesville, FL, USA).The following sampling design was used for each sampling site.In total, nine traps were placed along two parallel transects while two traps were placed in the canopy at approximately 10 m high.Traps operated at night for 12 h.Sandfly specimens were collected after sunrise, identified using female taxonomic keys, and stored in liquid nitrogen until further processing [25,26].

Phlebotomine sandfly community metrics
Phlebotomine sandfly species diversity was assessed at each landscape type using Shannon and Simpson Alpha diversity indexes [27,28].Species relative abundance was measured and compared among landscape types using one-way Kruskal-Wallis nonparametric test as implemented in the PAST3 program (available at https:// folk.uio.no/ ohammer/ past/).

RnA-extraction and PCR screening
Sandflies were pooled according to species, collection site and date by individuals of ten.Specimens containing blood were processed individually.Pools were homogenized in 500 µl L-15 medium (Thermo Fisher Scientific, Waltham, USA) using ceramic beads and a SpeedMill Plus (Analytik Jena, Germany).The suspensions were cleared by centrifugation at 2500 r.p.m. for 10 min.RNA was extracted from 50 µl supernatant using the QIAamp Viral RNA Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions.cDNA was synthesized using the SuperScript III RT System (Invitrogen, Karlsruhe, Germany) and random hexamer primers according to the instructions from the manufacturing company.Sandfly pools were tested for phleboviruses in a PCR assay using Platinum Taq DNA Polymerase (Thermo Fisher Scientific, Waltham, USA) and first round primers F 5'-TCAA RAAG AMNC AACA TGGTGG, R 5'-TATGCCYTG-CATCATYCCWG followed by nested round primers Fn 5'-GGAC TTAG AGAG ATYT AYGT NTTGG and Rn 5'-ACATGRTGACCYTGRTTCCA. Thermocycler conditions were as follows: 10 cycles of 95 °C for 3 min, 94 °C for 15 s, 55 °C for 20 s and 72 °C for 30 s, then 30 cycles from 95 °C to 15 s, 50 °C to 20 s, 72 °C to 30 s and a final extension step of 72 °C for 5 min.PCR products were visualized on 2 % agarose gels and positive samples were Sanger sequenced (Seqlab, Göttingen, Germany).Obtained sequences were analysed using Geneious R9 [29] and compared to GenBank database ( www.ncbi.nlm.nih.gov/ genbank/).

Genome sequencing
RNA derived from phlebovirus-positive sandfly pools was used to sequence full genomes on an Illumina MiSeq nextgeneration sequencing (NGS) platform.Library preparation and Illumina MiSeq sequencing was carried out by using the SuperScript One-Cycle cDNA Kit, the Nextera XT DNA Library Preparation Kit, and V3 chemistry (2×300 bp) according to the manufacturers' instructions.The individual samples were normalized, pooled and sequenced on an Illumina MiSeq instrument, with a designated yield of ~25 million paired-end reads.After demultiplexing the obtained raw reads were subjected to two filtering steps using Lighter v1.1.0[30] for correcting sequencing errors and bwa v0.7.17 [31] to subtract host-originated background information using the mitochondrial reference genome of Phlebotomus papatasi as no genome of a phlebotomine sandfly species closer related to the species used in this study is available.The remaining reads were de novo assembled into contigs with the genome assembler software Spades v3.11.1 [32].An initial blastx search against the Reference Viral Database (RVDB) [33] was performed, followed by a second blastx search against two custom databases containing only the S and the M segment sequences of Phenuiviridae.Reference mapping of NGS reads to the respective PCR fragment was performed using Geneious mapper [29].Genome ends were amplified by conventional semi-nested RT-PCR (reverse transcription PCR) using Platinum Taq DNA Polymerase (Thermo Fisher Scientific, Waltham, USA), segment-specific primers (Table 1) and a primer containing the terminal conserved nucleotides of phleboviruses fused to an anchor sequence PTN 5'-GACC ATCT AGCG ACCT CCAC ACACAAAG.Products were subsequently Sanger sequenced.

Phylogenetic analysis
Translational alignments of N, Gn, Gc and L genes were performed using the MAFFT E-INS-I algorithm [34].Modeltest was performed as implemented in mega7 to identify the most suitable evolutionary model [35].Phylogenies were inferred using PhyML with the WAG substitution model as implemented in Geneious R9, with 1000 bootstrap replicates [36].

Sandfly species composition and diversity
In total, 13 692 phlebotomine sandflies were collected in continuous tropical lowland forest (continuous forests), in fragmented forest surrounded by agricultural fields (forest fragments) and in isolated forest on islands in the Panama Canal area, Central Panama (forested islands).Eight different phlebotomine sandfly species were identified, with Psychodopygus panamensis as the predominating species (41.6 %) (Table 2).Due to damage of specimens, 28.7 % of the sandflies could not be identified to species level and were summarized as Phlebotominae spp..

Virus detection and genome analyses
Sandflies were tested for phleboviruses by generic RT-PCR.Sequences with 45-78 % nt identity to the RNA-dependent RNA polymerase (RdRp) gene of phleboviruses were identified in 12 pools (Table 3).The sequences formed five groups with intra-group pairwise nt identities ranging from 96-100 % and inter-group pairwise nt identities ranging from 67-74 %, suggesting the detection of several strains pertaining to five phlebovirus species.Attempts to isolate the viruses in vertebrate (VeroE6/7) and sandfly (LL-5) cells were not successful.
Complete coding regions of one strain from each putative species were obtained by NGS and semi-nested RT-PCR from homogenates of infected sandflies.The viruses were tentatively named after the geographic location where the sandflies have been collected, namely La Gloria virus (LAGV), Mona Grita virus (MOGV), Peña Blanca virus (PEBV), Tico virus (TICV) and Tres Almendras virus (TRAV).
All viruses had a tripartite genome organization similar to phleboviruses, comprising a large (L), middle (M) and small (S) segment (Fig. 1).The transduced L segment ORFs of LAGV, MOGV, PEBV, TICV and TRAV showed the highest similarities of 68-91 % to phlebovirus RdRp proteins suggesting the identification of five previously unknown viruses.Bunyaviruses, including phleboviruses, carry the conserved palm motifs Pre-A and A through E on the L segment.These motifs were detected in all five viruses (Fig. 2a).The conserved N-terminal motifs H...D...PD...ExT…K (cation-binding residues and catalytic lysine) of the endonuclease domain of phleboviruses were present as well (Fig. 2b).This domain is able to bind and to cleave RNA for transcription processes and is involved in the cap-snatching process typical for negativesense RNA viral polymerases like in bunya-and orthomyxoviruses [35,38].A second bunyavirus typical region, the motif G (RY), was present between aa 668 and aa 675 in all protein sequences of the novel viruses [39].The conserved arginine is positioned in the polymerase active site to interact with the priming nucleotide triphosphate.
The M segments of LAGV, MOGV, PEBV, TICV and TRAV encode one single ORF of 3992-4277 nt in length with the highest similarity (40-86 %) to glycoprotein precursor genes (GPC) of phleboviruses (Fig. 1).The translated GPC is predicted to be posttranslationally cleaved into the glycoproteins Gn and Gc, and the nonstructural protein NSm.Protein sequence searches using the pfam database revealed a similar predicted molecular weight of 59-60 kDa for both Gn and Gc.
The predicted molecular weights of the NSm proteins were more variable, ranging from 33 to 44 kDa.
The S segments of the five putative novel viruses LAGV, MOGV, PEBV, TICV and TRAV contained two ORFs (Fig. 1).The ORF closer to the 3′-end showed highest similarities of 64-81 % to the nucleocapsid gene (N) of phleboviruses.
The second ORF is encoded in ambisense, which is typical for phleboviruses, and most likely encodes the nonstructural protein NSs.

Phylogenetic analyses and classification
The novel viruses consistently fall into three main clades in the phylogenetic trees of RdRp, N, Gc and Gn proteins.TRAV branches as a sister taxon to Nique virus (NIQV) within the Candiru phlebovirus species-complex (Fig. 3).MOGV is part of the Frijoles phlebovirus species-complex as it is placed as a sister taxon to Joá virus (JOAV).PEBV is a novel lineage in basal phylogenetic relationship to the Bujaru phlebovirus species-complex, which is defined by Bujaru and Munguba viruses.TICV and LAGV share a common ancestor and are placed as two deep rooting lineages in basal position to the Bujaru phlebovirus species-complex.
In order to further analyse if the novel viruses may pertain to Phlebovirus species-complexes, the range of intra-species pairwise distances (p-distance) was analysed based on    To further test if LAGV and TICV group with other described phleboviruses, we extended our analysis and included the sequences of more than 15 putative novel phlebovirus species that have not yet been officially classified by the ICTV.These viruses are suggested to form 11 novel phlebovirus speciescomplexes.Among these, the closely related CHGV and Uriurana virus (URIV) have been suggested to establish distinct virus species-complexes as they do not show serological cross reactivity [14,18].Our analyses compare CHGV and URIV to established and suggested species complexes, and reveal p-distances of 19.5 and 27.1 % in the RdRp and N proteins, respectively.These values represent the lowest values among all established and suggested species complexes and may define a novel cut-off for species-complex demarcation (shown as black, dashed line in Fig. 4).
In a next step, we sought to identify genetic-based species demarcation criteria.As the intra-species genetic diversity of phleboviruses is unknown, we used Rift Valley fever virus (RVFV) as a surrogate since it represents the only established phlebovirus species in a phylogenetic solitary position, from which numerous strains have been sequenced.We compared the pairwise distances among all available RVFV strains in their RdRp genes.The analysis of 154 RVFV strains reveals maximum pairwise nucleotide and aa intraspecies genetic distances of 5 and 2 %, respectively.Next, all available Diptera-associated phlebovirus protein sequences, which were ≤5 % different from the closest related sequence on aa level, were selected as they are likely to present distinct species.Comparing the pairwise distances among the selected sequences reveals that all known Diptera-associated phleboviruses form 48 distinct species when using a conservative cut-off value of 5 % aa distance as species demarcation criterion.These species may be grouped into 28 species-complexes when applying a cut-off value of 20 % as species-complex demarcation criterion (Fig. 5).

Virus-vector associations
Virus infection rates were very low and prevented comprehensive analyses of virus prevalence patterns in the different types of landscape.In general, virus abundance was mainly linked to a geographic location.For example, two of the five viruses have been exclusively detected in sandflies collected on forested islands, MOGV in Nyssomyia trapidoi and TRAV in Psychodopygus panamensis, although both vector species were found in high numbers in all landscape types (Table 3).Furthermore, TICV was detected in one pool of Phlebotominae spp.collected in a forest fragment, and PEBV was found in three pools of sandflies collected in continuous forest containing specimens of Psychodopygus panamensis, Nyssomyia trapidoi and Phlebotominae spp.. LAGV was detected in six pools also consisting of Psychodopygus panamensis, Nyssomyia trapidoi and Phlebotominae spp.originating from continuous forest and forest fragments.LAGV was also detected in a pool of male Phlebotominae spp.sandflies suggesting a vertical transmission.The results from the correspondence analysis indicate that phlebotomine sandfly species composition and habitat type are not significantly associated (χ 2 (2)=5.527,p=0.063, df=16).

dISCuSSIon
Insects originating from biodiverse tropical forests have been shown to be infected with genetically highly diverse viruses.While the majority of viruses is found in mosquitoes [17,37,40,41] other blood-feeding arthropods such as sandflies are largely unexplored for viral infections.Here, we collected sandflies in remote forest areas of the Panama Canal area and screened them for infections with phleboviruses.We detected and genetically characterized five previously unknown phleboviruses found in Psychodopygus panamensis and Nyssomyia trapidoi species.Although a comparatively large number of sandflies was analysed (n=13 807), which have been collected within 1 year, we did not detect any of the sandfly-associated viruses known to occur in Panama.The latter viruses are mainly found in urban areas and their prevalence may be low in forested areas with comparatively little landscape modification.
Continuous forests showed the highest abundance of sandflies overall, which may be due to the extensive presence of leaflitter with decomposing organic matter for larval feeding, plus abundant burrows and crevices that serve as optimal breeding and resting sites for adult phlebotomine sandflies in more humid and undisrupted forest environments [25].We did not find differences in phlebotomine sandfly species richness among landscape types, suggesting that phlebotomine sandflies are probably more resilient to forest fragmentation than other blood-feeding dipterans [42].However, this could also be an artefact of our sampling strategy as CDC light traps tend to over select for certain species.Hence, community metrics might be better estimated using a combination of sampling strategies instead of using a single trapping method.Bichromomyia olmeca sp. was the only taxon exclusively found in continuous forest.The absence of Bichromomyia olmeca sp.from other landscape types is in line with findings by Fairchild and Theodor [43].Our results, in terms of phlebotomine sandfly species richness and overall species abundance, are very similar to those reported by Azpurua and colleagues who also investigated sandfly community structure and pathogen transmission in the Panama Canal area using CDC light traps [5].However, like Azpurua et al., we only surveyed for phlebotomine sandflies at night, therefore diurnal species were not part of our analysis.Further studies will have to corroborate whether diurnal species fit the community outcomes presented here.
PEBV, TRAV and LAGV were detected in Psychodopygus panamensis, which is a known vector of NIQV in Panama [17].NIQV belongs to the Candiru phlebovirus speciescomplex and is the closest relative to TRAV.Some members of the Candiru phlebovirus species-complex have been shown to infect humans [17].Whether TRAV can also infect humans and cause symptoms of disease needs further investigation.MOGV groups as a sister taxon to JOAV within the Frijoles phlebovirus species-complex and was detected in Nyssomyia trapidoi.JOAV was isolated from an unidentified species of phlebotomine sandflies in Brazil [18].The vector of TICV could not be not identified.
Due to the huge expansion of sequence information from phleboviruses, which were not isolated in cell culture, there is an urgent need to establish genetic-based classification criteria.Until recently, the classification of phleboviruses into species-complexes was only based on the presence or absence of serological cross-reactivity.Distinct species were differentiated by fourfold difference in neutralization tests, which requires the availability of reference viruses and antisera, as well as high laboratory safety standards.To circumvent this, haemagglutination inhibition (HI) and complement fixation (CI) tests have been widely used to test for serological crossreactivity, leading to the establishment of species-complexes as these tests cannot differentiate between serologically closely related viruses.In 2015, the ICTV has started an extensive reorganization of the bunyavirus taxonomy, including the genus Phlebovirus.The genomes of nearly all phlebovirus isolates that had only been characterized by serology were sequenced and classified into distinct viruses (designated member 'virus' by the ICTV), which were assigned to speciescomplexes (designated 'species' by the ICTV).This classification largely reflects the previous serology-based classification [11].Although, the genus Phlebovirus currently contains nine species-complexes, which comprise 57 recognized viruses, no genetic-based species demarcation criteria have been defined yet.In addition, the grouping of viruses into species-complexes seems to be a rather rough classification, as for example the Salehabad phlebovirus species-complex is extremely diverse and includes viruses with distances of up to 37 % within their RdRp proteins.Like in orthonairo-and orthohantaviruses, the members of the genus Phlebovirus are not classified in species but assigned to species-complexes.Our analyses of the intraspecies genetic diversity of 154 RVFV strains yielded a maximum distance of 2 % distance in RdRp protein sequences.Thus, a species demarcation criterion of 5 % aa distance between RdRp protein sequences would represent a rather conservative cut-off value.Such a criterion would be in line with the recently discovered novel sandfly-borne Ntepes virus (NPV), that groups with other members of the so far unclassified 'Karimabad phlebovirus' species-complex [44].NPV is at least 7 % distant in the RdRp protein from other phleboviruses (Fig. 5) and did not react with related viruses in neutralization assays suggesting that it defines a distinct species [45].Furthermore, distinct species in a related bunyavirus genus, the genus Orthobunyavirus, shall be differentiated by at least 4 % distance in their RdRp proteins [38].Taken together, these data suggest that cut-off values of 4-5 % aa distance within RdRp proteins are appropriate in a biological sense as species demarcation criteria.
Our phylogenetic analyses differed from the one presented by Nunes-Neto and colleagues [18].In their analyses, Uriurana virus groups together with Tapara virus and the authors propose to classify both viruses into a single species-complex.In contrast, our phylogenetic analysis places Uriurana virus as a sister taxon to Chagres virus (Fig. 3), a topology which has also been found by others [19].Chagres virus has been isolated from a human in 1960 from Panama [14] and Uriurana virus was isolated in 1985 from Phlebotominae sp.collected in Brazil [18].Despite their long history, both viruses are not officially classified by the ICTV.According to their genetic distance of 20 % in their RdRp protein sequences (Fig. 5), Uriurana virus and Chagres virus seem to be two distinct species of a single species-complex, tentatively named 'Chagres phlebovirus' species-complex.
The detection and genomic characterization of five novel viruses identified in Panamanian phlebotomine sandflies species show that the taxonomic diversity of phleboviruses is larger than what has been reported before in the Panama Canal area.In this study, Psychodopygus panamensis and Nyssomyia trapidoi were the most prevalent species showing the highest overall proportions regardless of landscape type.The fact that both species have been previously incriminated as vectors of human pathogens in Panama, combined with their considerable abundance, is very suggestive of their involvement in the transmission of LAGV, MOGV, PEBV, TICV and TRAV.However, future studies will have to investigate if these sandfly species are competent vectors, as well as if these viruses infect mammals.Our findings underline the importance to extend surveillance to blood-feeding arthropods other than mosquitoes, including hard ticks, biting-midges and sandflies.

Fig. 1 .
Fig. 1.Schematic genome organization of phleboviruses sequenced in this study.Dark lines represent genome segments, arrows represent coding sequences and boxes represent translated proteins.

Fig. 2 .
Fig.2.Phleboviral polymerase motifs and endonuclease domain.Sequence alignments of highly conserved motifs within the phleboviral RdRp protein of species-complex prototype viruses and viruses sequenced in this study.Residues conserved throughout all taxa are marked with an asterisk.Highly conserved endonuclease residues are highlighted in red.

Fig. 3 .
Fig. 3. Phylogenetic analyses of phleboviruses.Maximum-likelihood phylogenies of selected phleboviruses (>5 % aa distance in RdRp proteins) inferred from translated RdRp, N, Gn and Gc genes.Coloured branches represent established (bold type) and suggested species-complexes.Viruses described in this study are shown in red.

Fig. 4 .
Fig. 4. Mean pairwise distance among established and suggested phlebovirus species-complexes and phleboviruses sequenced in this study.Mean aa p-distance comparison of RdRp and N protein among established (bold, asterisk) and suggested Phlebovirus speciescomplexes and viruses sequenced in this study (red).Whiskers represent the data range with the whisker top representing the highest and the whisker bottom the lowest values.Dashed lines represent cut-off value based on projected serological cross reactivity to genetic level using established (grey) and including suggested species-complexes (black).

Fig. 5 .
Fig. 5. Distance matrix of phleboviruses.Distance matrix of the RdRp protein of selected phleboviruses (<95 % aa identity, RdRp) and viruses sequenced in this study (bold).ICTV classified species-complexes are illustrated by black frames.Shades of green to yellow to red are shown in accordance to the percentage of aa distance (green=66 %; yellow=31 %; dark red=4 % aa distance).Putative speciescomplexes applying ≤20 % aa distance as the cut-off value are numbered by yellow boxes and framed in red.

Table 2 .
Virus names, designated strains, locality of sandfly specimen collection, sex and species of viruspositive sandflies Abbreviation of localities are as follows: forest fragment (F), continuous forest (C), forested island (I).
Marklewitz et al., Journal of General Virology 2019;100:938-949 22. Condit R, Robinson WD, Ibáñez R, Aguilar S, Sanjur A et al.The status of the Panama canal watershed and its biodiversity at the beginning of the 21st century: long-term ecological studies reveal a diverse flora and fauna near the Panama canal, harbored within a corridor of forest stretching from the Caribbean to the Pacific, but deforestation, land degradation, erosion, and overhunting remain threats.Bioscience 2001;51:389-398.