Natural Infection by Toscana Virus and Other Phleboviruses in Wild Sand Flies From Spain: Prevalence, Genetic Characterization and Tripartite Interactions With Leishmania infantum and Wolbachia

ABSTRACT

Introduction

Toscana virus (TOSV) and Leishmania species are sand fly‐borne pathogens that pose a significant public health threat and an increasing risk of exposure in Europe. The objective of this study was to identify the phleboviruses circulating in an area endemic for sand fly‐borne pathogens and to estimate infection prevalence in wild sand fly populations. We also aimed to analyse co‐infections with L. infantum and the symbiotic bacterium Wolbachia.

Methods

The study was conducted in southern Spain. Sand flies were collected using CDC light traps, morphologically identified and DNA/RNA was extracted individually or in pools. A generic RT‐nested PCR was used to detect phleboviruses, followed by sequencing and phylogenetic analysis. Leishmania infantum loads were quantified by qPCR, and Wolbachia was detected by PCR.

Results

We found that 2/443 individual sand flies (0.45%) and 8/64 pools (12.5%) were infected with phleboviruses, with an estimated prevalence of 1.9%. We confirm the circulation of TOSV‐lineage B, together with Massilia‐like virus and Alcube virus, in three sand fly species, and report novel strains of these phleboviruses. In addition, two Phlebotomus perniciosus individuals were co‐infected with TOSV‐lineage B and Wolbachia wPrn. We also detected L. infantum in 27.6% (60/217) of individuals, 66.7% (40/60) of which were co‐infected with Wolbachia. Mean L. infantum loads [log‐(parasite load +1)] in specimens co‐infected with Wolbachia were not significantly different from those in uninfected specimens (p = 0.4).

Conclusions

This study provides further evidence for the continued circulation of TOSV and other phleboviruses in sand fly populations in southern Spain, with an estimated prevalence higher than that reported in the same region over a decade ago. Our study provides the first evidence of co‐infection with TOSV‐lineage B and Wolbachia in P. perniciosus, as well as a high proportion of specimens co‐infected with Wolbachia and L. infantum. Understanding these co‐infection dynamics is essential to elucidate the potential role of Wolbachia and the impact of Leishmania–phlebovirus co‐infection on transmission.

Impacts

• Novel Toscana virus, Massilia‐like virus and Alcube virus strains circulate in three sand fly species in southern Spain.

• This study provides the first evidence of TOSV‐lineage B and Wolbachia co‐infection in Phlebotomus perniciosus.

• Co‐infection between Wolbachia wPrn strain and L. infantum was detected in 66.7% of the P. perniciosus specimens analysed.

1. Introduction

Pathogens transmitted to humans by phlebotomine sand flies (Diptera, Psychodidae) remain neglected, even though infections caused by protozoa of the genus Leishmania (Kinetoplastida, Trypanosomatidae) and viruses from genus Phlebovirus (Hareavirales, Phenuiviridae) have important implications for human health. Both infections are emerging or re‐emerging in Europe (Moriconi et al. 2017; Vaselek 2021). Several environmental factors—such as climate change, increasing urbanization, population movements and the colonization of urban areas by wildlife—are driving changes in the epidemiology of sand fly‐borne infections by altering the accessible ranges of vectors, reservoirs and pathogens (Molina et al. 2012; Vadmal et al. 2023; Barroso et al. 2024; Martín‐Sánchez et al. 2024). Among the emerging challenges related to sand fly‐borne pathogens, particular attention should be given to the interactions between Leishmania and sand fly‐borne viruses, as well as those involving sand fly endosymbionts (Torres‐Llamas et al. 2025). These interactions could have implications for vector competence and transmission dynamics (Torres‐Llamas et al. 2025; Stejskalova et al. 2026). In Spain, Leishmania infantum is responsible for both canine leishmaniasis (CanL) and the three clinical forms of human leishmaniasis: visceral (VL), cutaneous (CL) and mucosal (ML), with dogs being the primary domestic reservoir and leporids the main sylvatic reservoir for the parasite (Molina et al. 2012; Martín‐Sánchez et al. 2020, 2024; Morales‐Yuste et al. 2022; Fernández‐Arévalo et al. 2024). Nevertheless, knowledge about infections caused by phleboviruses is still limited.

Phleboviruses are single‐stranded RNA viruses whose genome is segmented into three parts of different sizes, small (S), medium (M) and large (L) (Calzolari et al. 2021). Toscana virus (TOSV) is among the most pathogenic phleboviruses infecting humans and is one of the principal causes of viral meningitis in the Mediterranean region, along with enteroviruses and herpesviruses (Ayhan and Charrel 2020; Hong et al. 2024). Over the past 38 years, 1381 confirmed cases of TOSV infection have been reported (reviewed by Ayhan et al. 2025), with most cases diagnosed in Italy and Spain. Although mortality related to TOSV infections is low and appears to be associated with age and other comorbidities, a large proportion of cases present neurological manifestations such as meningitis, meningoencephalitis and other symptoms affecting the central nervous system (Ayhan et al. 2025). However, due to the lack of specific symptoms, pathognomonic features and established diagnostic protocols, it is likely that many cases remain undiagnosed or unreported (García San Miguel et al. 2021; Hong et al. 2024; Ayhan et al. 2025).

The geographical expansion of sand flies in Europe and surrounding regions poses a growing threat of exposure to TOSV, placing millions of people at risk of infection (Hong et al. 2024; Wang et al. 2025). Despite the detection of TOSV in several vertebrate animals—both serologically and, in some cases, molecularly—their role as potential reservoirs remains uncertain (Ayhan et al. 2020; Keskek Turk et al. 2024; Xhekaj et al. 2025). There are currently three genetic lineages of Toscana virus circulating in Europe, TOSV‐lineage A, B and C (Ayhan et al. 2017, 2020). While lineage A has been detected in France, Italy, Türkiye, Tunisia and Algeria, lineage B has been reported in Spain, Portugal, France, Morocco and Türkiye. In addition, lineage C has only been identified in Croatia and Greece. Notably, the co‐circulation of multiple lineages has been documented in France and Türkiye (lineages A and B) and in Croatia (lineages B and C) (reviewed by Ayhan and Charrel 2020). Recent studies suggest that TOSV‐lineage A may be more pathogenic than lineage B, due to its higher replication rate, increased virion production and greater neuroinvasive capacity (Roy et al. 2025; Thiesson et al. 2025). To date, only TOSV‐lineage B has been detected in sand flies on the Iberian Peninsula, and several viral strains have been characterized (Sanbonmatsu‐Gámez et al. 2005; Amaro et al. 2024).

The circulation of several phleboviruses other than TOSV has also been detected in Spain. Massilia virus (MASV), a species within the Sandfly fever Naples complex, was first identified in southeastern France, and has since been found in sand flies in both Portugal and Spain (Charrel et al. 2009; Amaro et al. 2015, 2021; Davó et al. 2020). Another Massilia‐like virus—tentatively named Granada virus (GRV)—was detected in sand flies in the Granada Province (southern Spain). This putative new virus shares high nucleotide homology with L and S segments of MASV (Collao et al. 2010). However, based on current International Committee on Taxonomy of Viruses (ICTV) criteria for defining new phlebovirus species, GRV should be classified within the MASV species (Davó et al. 2020; Amaro et al. 2022).

Although both phleboviruses and Leishmania have been found to circulate within the same geographic area, potential interactions between these sand fly‐borne pathogens remain to be elucidated (Amaro et al. 2023). The transmission of pathogens by insect vectors may be modulated not only by interactions between co‐infecting pathogens themselves, but also by their interactions with other microorganisms, including the endosymbiotic bacterium Wolbachia pipientis (Wong et al. 2020). Wolbachia can interfere with the establishment of both viral and parasitic infections in insect hosts. For instance, Wolbachia has been shown to protect insects against RNA viruses (Moreira et al. 2009; Pimentel et al. 2025), and when introduced into Anopheles mosquitoes, can induce refractoriness to Plasmodium infection (Bian et al. 2013). Despite these reported properties, the effects of Wolbachia on sand flies and the pathogens they transmit remain poorly understood (Yilmaz and Kasap 2025) and are likely influenced by both the sand fly species or lineage and the infecting Wolbachia strain (Stejskalova et al. 2025; Torres‐Llamas et al. 2025). Only a few experimental studies have investigated the potential interactions between Wolbachia (or its surface protein WSP) and L. infantum, and none has assessed a Wolbachia strain naturally associated with sand flies (da Silva Gonçalves et al. 2019; Varotto‐Boccazzi et al. 2020; Latrofa et al. 2023).

Thus, potential associations among the development of Leishmania parasites, phleboviruses and Wolbachia warrant further investigation, particularly regarding epidemiological links and the conditional capacity of these pathogens to interact in ways that may enhance transmission or virulence. To address these gaps, field surveys are needed to assess all three components within the same sand fly populations. The primary objective of this study was to identify the phleboviruses circulating in an area endemic for sand fly‐borne pathogens, such as TOSV and L. infantum, and to estimate the prevalence of infection in wild sand fly populations. We also focused on characterizing viral strains based on the L and S segments and establishing their phylogenetic relationships. Ultimately, we aimed to analyse tripartite co‐infections involving TOSV, L. infantum and the endosymbiotic bacterium Wolbachia.

2. Methods

2.1. Study Area

The study was conducted in Granada Province, located in southern Spain (Andalusia; centroid coordinates: 37°10′ N, 3°35′ W; Figure 1). The province spans a marked altitudinal gradient ranging from 0 to 3479 m above sea level. Five sampling sites, all of which fall within the mesomediterranean bioclimatic stage (Rivas Martínez et al. 1987), were selected in the peri‐urban area of Granada city (695–885 m a.s.l.) based on previous records of established sand fly populations. The coordinates, elevation, dominant vegetation and fauna of each site are summarized in Table 1.

FIGURE 1.

Geographical distribution of sand fly collection sites in southern Spain.

TABLE 1.

Characteristics of the sampling sites and distribution of the sand flies collected by location, species, sex and processing method (pooled or individual).

Collection sites and features	Species	No. of specimens	No. of pools	Individuals
Mantas: At an elevation of 885 m (37°12′34.5″N 3°34′9.5″W) the predominant vegetation consists of almond, fig and olive trees and dry straw. The animals present include sheep, goats, peacocks, cats and chickens.	P. perniciosus	156M; 235F	23 (5F; 18M)	72M; 181F
	P. ariasi	36F	3 (3F)	—
	P. papatasi	4M; 4F	—	4M; 4F
	P. sergenti	3M; 2F	—	3M; 2F
	S. minuta	35F	1 (1F)	22F
Cañas: At an elevation of 780 m (37°11′42.4″N 3°35′34.4″W), the predominant vegetation consists of fig and olive trees, along with shrubs. The animals present include dogs, cats and rabbits.	P. perniciosus	81M; 53F	22 (5F; 17M)	—
	P. ariasi	22M; 26F	4 (2F; 2M)	—
	P. papatasi	5F	1 (1F)	—
	P. sergenti	12F	1 (1F)	—
	S. minuta	18M; 27F	9 (3F; 6M)	—
Gójar: At an elevation of 760 m (37°6′40.0″N 3°36′22.3″W), the predominant vegetation consists of cypresses and firewood. The animals present include dogs and cats	P. perniciosus	8F	—	8F
Zubia 1: At an elevation of 700 m (37°8′6.8″N 3°35′23.0″W), the predominant vegetation consists of almond and lemon trees, and dry straw. The animals present are dogs, cats, horses, mules, donkeys, goats and sheep.	P. perniciosus	9F	—	9F
	S. minuta	5F	—	5F
Zubia 2: At an elevation of 695 m (37°7′44.9″N 3°35′33.6″W), the predominant vegetation consists of almond and olive trees, and dry straw. The animals present include dogs, mules, pigs, mice, ducks, geese, chickens, turkeys, sheep, goats and rabbits.	P. perniciosus	22M; 19F	—	22M; 19F
	P. papatasi	26M; 18F	—	26M; 18F
	P. sergenti	12M; 17F	—	12M; 17F
	S. minuta	7M; 12F	—	7M; 12F
Total	Pp (n = 583; 66.7%)	874 (351M; 523F)	64 (21F; 43M)	443 (146M; 297F)
	Sm (n = 104; 11.9%)
	Pa (n = 84; 9.6%)
	Ppa (n = 57; 6.5%)
	Ps (n = 46; 5.3%)

Abbreviations: F, female; M, male; Pa, P. ariasi ; Pp, P. perniciosus; Ppa, P. papatasi; Ps, P. sergenti; Sm, S. minuta .

2.2. Sand Flies Collection and Species Identification

Phlebotomine sand flies were collected in 2019 and 2022 using battery‐operated CDC miniature light traps without CO₂ bait (J. W. Hock Company, Gainesville, FL, USA). At each site, two traps were deployed for one night per month from July to September (6 trap‐nights per site and year), operating for approximately 12 h from dusk to dawn (19:00–20:00 to 07:00–08:00). Traps were placed close to animal shelters, preferentially indoors. When indoor access was not feasible, traps were deployed outdoors in the immediate peridomestic surroundings. All sampling sites were located approximately 1–3 km from the nearest residential area. After collection, the specimens were promptly transferred to the laboratory and preserved in RNAprotect Tissue Reagent (QIAGEN, Germany) at −80°C to maintain RNA integrity. For morphological identification, the genitalia were dissected on ice and mounted, whereas the remainder of the body was used for nucleic acid extraction. Identification was performed using taxonomic keys, with particular attention to the spermathecae in females and the external genitalia in males, especially the aedeagus (Sáez et al. 2018). Following identification, specimens were either processed individually or grouped into pools comprising 3–17 individuals of the same species, sex, location and date.

2.3. Sand Fly Nucleic Acid Extraction

Genomic DNA and RNA were extracted from the remaining sand fly tissues after morphological identification. A commercial MasterPure Complete DNA and RNA Purification Kit (cat. no. MC85200) was used according to the manufacturer's instructions. Sand flies were processed individually or in pools and then placed in a sterile 1.5 mL Eppendorf tube and kept in liquid nitrogen for a few seconds to facilitate the mechanical tissue disruption with a pestle. The nucleic acids were resuspended in 20 μL of nuclease‐free water and stored at −80°C until use.

2.4. Detection of Phleboviruses RNA

A generic RT‐nested PCR adapted from Sánchez‐Seco et al. (2003) was used to detect TOSV and other phleboviruses. First, reverse transcription was performed from 8 μL of RNA using SensiFAST cDNA Synthesis Kit (Cat. No. BIO‐65054, Meridian bioscience, UK), following the manufacturer's instructions. The resulting cDNA was used as a template for the first‐round PCR targeting a conserved region of the L segment with primers NPhlebo1+/NPhlebo1–. The first‐round product was then used as a template for two parallel nested reactions: (i) a nested‐PCR with primers NPhlebo2+/NPhlebo2–, capable of detecting a broad range of phlebovirus species; and (ii) a TOSV‐specific nested‐PCR in which NPhlebo2– was replaced by the TOSV‐specific antisense primer ATos2−. Amplicons from the generic nested PCR were sequenced using the same primers (NPhlebo2+/NPhlebo2−). Additionally, the S segment was amplified using primers TosN123/TosN829 in the first PCR and TosN234/TosN794 in the nested‐PCR, with the latter also used for sequencing. Primer sequences and amplification conditions are provided in Supporting Information. RNA extracted from a TOSV strain isolated elsewhere was used as a positive control (Sánchez‐Seco et al. 2003). A negative control containing all reagents and nuclease‐free water was included in each PCR run.

2.5. Sequencing and Phylogenetic Analysis

Amplified PCR products were eluted from agarose gel using Real Clean Spin Kit (Zymoclean Gel DNA Recovery Kit, cat. no. D4007), according to the manufacturer's protocol. Direct cycle Sanger sequencing of the PCR‐amplified products was performed bidirectionally using an ABI 3130XL automated sequencer. The resulting sequences were visualized and manually curated with Chromas version 2.6.6, then aligned using ClustalW as implemented in MEGA11. The best nucleotide substitution model was selected using jModelTest version 2.1.10 and MEGA11. Phylogenetic relationships among phlebovirus consensus sequences from both L and S segments, as well as additional sequences available in GenBank, were inferred using MEGA11. Genetic distances were calculated using the Kimura 2‐parameter (K2P) model. Three phylogenetic methods were applied: Neighbour‐Joining (NJ), Maximum Likelihood (ML) and Maximum Parsimony (MP). The NJ and ML trees were inferred under the best‐fitting substitution model for both segments (K2P+G), whereas the MP tree was inferred using the max‐mini branch‐and‐bound search method. Model selection was based on the lowest Akaike Information Criterion (AIC) score. All positions containing gaps or missing data were excluded from the analyses using the complete deletion option. The robustness of internal branches was assessed by bootstrap analysis with 1000 replicates, and values below 50% were hidden. Uukuniemi virus sequences for both L (Accession: M33551) and S (Accession: KM114249) segments were used as an outgroup.

2.6. Determination of Leishmania infantum Loads by qPCR

GRANALEISH Multiplex qPCR (University of Granada, Spain, Trademark number 3667362/5) (Gijón‐Robles et al. 2018) was used for identification of L. infantum and quantification of the parasite load. Primers and the TaqMan probe were provided by the manufacturer. The thermal cycling profile used was as follows: an initial denaturation at 95°C for 10 min, followed by 40 cycles of 30 s at 95°C and 60 s at 60°C. A standard curve was constructed from serial dilutions of DNA extracted from cultured L. infantum strain MHOM/ES/14/DP581 representing concentrations equivalent to 1000–0.009 promastigotes/μL. The standard curve showed a coefficient of determination (R ²) of 0.975 and a PCR efficiency of 99.5%. Parasite numbers in each qPCR reaction were calculated by interpolating the Ct values against the standard curve. DNA extracted from the cultured L. infantum reference strain MHOM/ES/14/DP581 was used as a positive control. A no‐template control containing all reagents and nuclease‐free water was included in each PCR run. Additionally, negative extraction controls were used to monitor potential cross‐contamination.

2.7. Detection of Wolbachia DNA and Sequencing

The presence of Wolbachia DNA in sand flies was assessed using conventional PCR. Primers 81F (5′‐TGG TCC AAT AAG TGA TGA AGA AAC‐3′) and 691R: (5′‐AAA AAT TAA ACG CTA CTC CA‐3′) were used to amplify a fragment of 590–632 base pairs (bp) of the Wolbachia wsp gene. The variability in amplicon size reflects natural sequence length polymorphism in the wsp gene among Wolbachia strains (Zhou et al. 1998). The reagent quantities and PCR final volume used in this study follow the protocol described by Torres‐Llamas et al. (2025). The amplification conditions were: 2 min denaturation at 94°C followed by 35 cycles of denaturation at 94°C for 30 s, annealing at 55°C for 30 s and extension at 72°C for 1 min, with a final extension at 72°C for 10 min. Sand fly specimens that tested positive or negative for Wolbachia in a previous study (Torres‐Llamas et al. 2025) were used as positive and negative controls, respectively. Positive specimens were sequenced to determine the Wolbachia strain.

2.8. Data Analysis

Sand flies were processed and phleboviruses detected using two approaches: individual sampling and pooled sampling. For the individual samples (Method 1), the number of positive specimens was recorded, and the observed prevalence was calculated as the proportion of infected individuals out of the total tested. Infection prevalence was also estimated from pooled samples of different sizes (Method 2). The infection rate in pooled samples was obtained using Maximum Likelihood Estimation (MLE). This approach defines the infection rate as the proportion of infected individuals that maximizes the likelihood of observing the number of positive and negative pools of the recorded sizes. The method assumes a binomial distribution of infected individuals within each pool and identifies the prevalence value that best explains the observed pooling results (Khongpetch et al. 2025). These analyses were performed in R v. 4.5.2.

For the analysis of interactions within the vector only individualized female P. perniciosus sand flies were used (n = 217). To investigate potential associations between L. infantum and Wolbachia infections, a logistic regression model was fitted with L. infantum infection status (presence/absence) as the dependent variable, and Wolbachia presence and sampling site as independent variables. Sampling site was included as a covariate to account for potential geographic confounding, as prevalence of L. infantum differed among sites. Given the uneven distribution of specimens across sampling sites, they were grouped for the regression analysis into Mantas (the main site, n = 181) and all remaining sites combined (n = 36) to obtain more stable model estimates and avoid complete separation. The effect of Wolbachia presence on L. infantum parasite load was assessed by logistic regression using a log‐transformation of parasite load [log (parasite load +1)] for variable normalization. All tests were conducted using IBM SPSS Statistics v. 21.0 for Windows (IBM Corp., Armonk, NY, USA).

3. Results

3.1. Phleboviruses Detection in Sand Flies

The presence of phleboviruses was assessed in 874 sand fly specimens collected from five locations in southern Spain, 351 males (40.2%) and 523 females (59.8%), belonging to two genera—Phlebotomus and Sergentomyia—and five species (Table 1). A total of 443 specimens were individually analysed (Table 1), of which 205 were captured in 2019 and 238 in 2022. Phlebovirus infection was detected in only two individually analysed female P. perniciosus specimens (2/443), yielding an observed prevalence of 0.45% (95% CI: 0.05%–1.62%). Both specimens were collected at the same sampling site (Mantas), albeit in different years—one in 2019 and the other in 2022. Only one of the two genomic segments analysed per specimen was successfully amplified: the L segment in one case and the S segment in the other (Table 2).

TABLE 2.

Sand fly individuals and pools positive for phleboviruses and co‐detection of L. infantum and Wolbachia.

ID	Location	Pool size	Sand fly species	Sex	Phlebovirus detection (L/S)	GenBank nucleotide homology (countries)	L. infantum detection	Wolbachia detection
I 128	Mantas	−	P. perniciosus	Female	TOSV (S)	96%–98% (Spain, Portugal, France, Austria, Türkiye)	−	+
I 183	Mantas	−	P. perniciosus	Female	TOSV (L)	93%–98% (Spain, France, Morocco, Türkiye, Tunisia)	−	+
Pool 10	Cañas	9	P. perniciosus	Female	TOSV (L)	95%–99% (Spain, France, Morocco, Türkiye, Tunisia)	+	+
Pool 13	Mantas	3	P. perniciosus	Male	MASV (L)	95%–98% (Spain, Portugal, France)	NA	−
Pool 19	Mantas	3	P. perniciosus	Male	ACBV (L)	94%–95% (Spain and Portugal)	NA	+
					MASV (S)	90%–98% (Spain, Portugal, France)
Pool 24	Mantas	3	P. perniciosus	Male	TOSV (L)	95%–99% (Spain, France, Morocco, Türkiye, Tunisia)	NA	−
					TOSV (S)	97%–99% (Spain, Portugal, France, Austria, Türkiye)
Pool 26	Mantas	3	P. perniciosus	Male	MASV (S)	96%–98% (Spain, Portugal, France)	NA	−
Pool 44	Cañas	15	P. perniciosus	Male	TOSV (L)	95%–98% (Spain, France, Tunisia, Türkiye)	NA	+
Pool 54	Cañas	11	P. ariasi	Male	TOSV (L)	95%–99% (Spain, France, Morocco, Türkiye, Tunisia)	NA	+
					TOSV (S)	97%–99% (Spain, Portugal, France, Austria, Türkiye)
Pool 64	Cañas	3	S. minuta	Male	TOSV (L)	95%–99% (Spain, France, Morocco, Türkiye, Tunisia)	NA	−

Note: Parentheses indicate amplified RNA segment.

Abbreviation: NA, not applicable.

The remaining 431 specimens, all captured in 2019, were grouped in 64 pools containing between 3 and 17 individuals each. Of these 64 pools, 21 were composed of females and 43 of males (Table 1). In total, eight pools (12.5%) tested positive for phlebovirus infection. Among the positive pools, six were from P. perniciosus, one from P. ariasi , and one from S. minuta (Table 2). The MLE‐based infection rate for all sand fly species was 1.9% (95% CI: 0.9%–3.6%). Three out of eight positive pools showed amplification of both genomic segments, four pools were positive exclusively for the L segment, and one pool for the S segment (Table 2).

3.2. Identification and Characterization of Phleboviruses Species

Following amplification of either the L or S segment, or both, Sanger sequencing enabled the identification of three phleboviruses circulating in the study area: Toscana virus (TOSV), Massilia‐like virus (MASV) and Alcube virus (ACBV).

3.2.1. Toscana Virus (TOSV)

Toscana virus was detected in two P. perniciosus individuals, three P. perniciosus pools, one Phlebotomus ariasi pool and one Sergentomyia minuta pool, in one or both genomic segments (Table 2). Sequences obtained from pools were identical within each genomic segment, whereas both individual sand flies–derived sequences showed a small number of nucleotide polymorphisms for each genomic segment. BLAST analyses revealed 93%–99% nucleotide homology for both genomic segments with TOSV strains previously reported in several countries (Table 2). When all sequences were aligned, including those previously reported from the same study area (Granada) and other regions of Spain, 14 to 38 polymorphic sites were identified depending on the genomic segment analysed. Nucleotide polymorphisms among the sequences obtained in this study were all synonymous at the amino acid level. Compared with reference sequences from GenBank, a single amino acid substitution was identified in the L segment. This substitution was not shared with previously reported sequences from Granada but was consistent with those reported from other regions of Spain (Table 3). Genetic distances between TOSV‐lineage A and B ranged from 0.005 to 0.072 for the L segment and TOSV‐lineage A, B and C ranged from 0.013 to 0.195 for the S segment, with the greatest divergence observed relative to lineage C strains.

TABLE 3.

Genetic characterization of phleboviruses circulating in the study area compared with Spanish reference strains from different years.

Phlebovirus	ID samples	Nucleotide polymorphisms	Amino acid sequence (residue/change)	Strain reference (Spain)
TOSV (S)	I128	38	—	Granada (2004, 2017), Córdoba (2015, 2017, 2018, 2019)
	Pools 24 and 54	33	—	Granada (2004, 2017), Córdoba (2015, 2017, 2018, 2019)
TOSV (L)	I183, Pools 10, 24, 44, 54 and 64	14	30/threonine to alanine1	Granada (2004, 2017), Córdoba (2015, 20171, 2018, 2019), Madrid (2012)
MASV (S)	Pool 19 and 26	21	162/aspartic acid to asparagine	Granada (2004)
MASV (L)	Pool 13	25	9/alanine to serine2, 36/isoleucine to valine3	Granada (2004)2, Barcelona (2001–2004), Cabrera Island (2012)3
ACBV (L)	Pool 19	9	—	Formentera Island (2012)

Note: Nucleotide polymorphisms represent the total number of variable positions identified in a multiple sequence alignment including our sequences and all the reference strains listed. Superscript numbers (^1–3) indicate the reference strain(s) relative to which each amino acid change was detected in our sequences. —: no amino‐acid changes detected.

Abbreviations: ACBV, Alcube virus; I, individually processed specimen; L, large genomic segment; MASV, Massilia virus; S, small genomic segment; TOSV, Toscana virus.

3.2.2. Massilia‐Like Virus (MASV)

Massilia‐like virus was detected in three P. perniciosus pools through the amplification of either the S or L segment. Sequence homology revealed 90%–98% nucleotide homology with previously reported MASV sequences from Spain, France and Portugal (Table 2). Twenty‐one polymorphisms were detected when the S‐segment sequences were aligned with one previously reported in Granada in 2004 (then referred to as Granada virus), and only a single amino‐acid change was observed (Table 3). Similarly, 25 polymorphic sites were identified when the L segment sequence obtained in this study was aligned with those previously reported from Granada, Barcelona and Cabrera Island (Balearic Archipelago, Spain). These polymorphisms resulted in two amino‐acid changes (Table 3). Genetic distances relative to GenBank sequences ranged from 0.024 to 0.037 for the L segment and from 0.017 to 0.047 for the S segment.

3.2.3. Alcube Virus (ACBV)

Alcube virus sequence was detected in a P. perniciosus pool (L segment), showing 94%–95% nucleotide homology with strains from Portugal and Formentera Island (Spain), (Table 2). When compared with the Formentera strain, nine polymorphisms were identified in the amplified segment (Table 3). All observed polymorphisms were synonymous at the amino‐acid level, and genetic distances ranged from 0.047 to 0.053 when compared with the Formentera Island and Portugal strains, respectively.

3.3. Phylogenetic Analysis of Phleboviruses

Eight new sequences have been deposited in GenBank under the following accessions: PX982977‐PX982984. Both the L and S segments of the phleboviruses detected in this study, along with other phlebovirus strains and species available in GenBank, were used to construct a phylogenetic tree. The three methods employed to infer phylogenetic relationships—Neighbour‐Joining (NJ), Maximum Likelihood (ML) and Maximum Parsimony (MP)—produced similar topologies. Neighbour‐Joining trees for both S (Figure 2a) and L (Figure 2b) segments are shown. All sequences corresponding to TOSV were clustered within lineage B, which belongs to the Sandfly fever Naples complex. Similarly, the MASV sequences were clustered with other reported sequences of this virus detected in several countries, such as France and Portugal, which belong to the same complex. In contrast, the sequence belonging to Alcube virus was clustered with other sequences reported in Spain and Portugal, which belong to the Salehabad complex.

FIGURE 2.

Neighbour‐Joining trees illustrating the phylogenetic relationships of the phlebovirus strains detected in this study (indicated with a circle) for the S segment (a) and the L segment (b), and their distribution within the different phlebovirus complexes.

3.4. Multi‐Microbe Co‐Infections in Phlebotomus perniciosus

Potential interactions among Leishmania, TOSV and Wolbachia were analysed in 217 individual female specimens of the species P. perniciosus. The prevalence of infection was 27.6% (60/217) for L. infantum, 0.9% (2/217) for TOSV and 64.1% (139/217) for Wolbachia wPrn strain, as determined by sequencing. Co‐infection with L. infantum and Wolbachia was detected in 40 specimens, representing 66.7% of L. infantum‐infected individuals. Logistic regression revealed no significant association between L. infantum and Wolbachia infection after adjusting for sampling site (OR = 1.15, 95% CI: 0.61–2.18, p = 0.663). Sampling site was the only significant predictor of L. infantum infection, with lower odds of infection at sites other than Mantas (OR = 0.28, 95% CI: 0.09–0.83, p = 0.022). No co‐infection between L. infantum and TOSV was observed. Both TOSV‐positive specimens were also positive for Wolbachia; however, given the extremely small number of phlebovirus‐positive sand flies (n = 2), these observations are purely descriptive and do not allow an assessment of a potential association between TOSV and Wolbachia. Among the 60 L. infantum‐infected sand flies, parasite load ranged from 0.01 to 675 parasites per specimen. Wolbachia‐uninfected females had a mean [log‐(parasite load +1)] of 0.826 (95% CI: 0.405–1.246), whereas Wolbachia‐infected females had a mean value of 0.650 (95% CI: 0.429–0.870). Logistic regression revealed no significant association between the two variables (OR = 0.743, 95% CI: 0.373–1.482, p = 0.4).

4. Discussion

Although TOSV is an emerging pathogen in the Mediterranean Basin with the potential to spread to other parts of Europe, it continues to be neglected (Laroche et al. 2024). Monitoring sand fly‐borne pathogens is essential for implementing control measures, which remain the most reliable strategy for preventing vector‐borne diseases (Wilson et al. 2020). In this study, TOSV and other phleboviruses were detected circulating in three sand fly species. The higher estimated prevalence obtained through pooled sampling may suggest that this method is more effective in detecting low‐level infections. Entomological surveys have been conducted to assess phlebovirus infection in sand fly populations across countries neighbouring Spain, including Italy, Portugal, France and Morocco (Charrel et al. 2007; Es‐Sette et al. 2012, 2015; Calzolari et al. 2021; Amaro et al. 2023, 2024). Reported phlebovirus infection rates vary by region, country and sampling year. In Spain, Sanbonmatsu‐Gámez et al. (2005) reported a pool‐based prevalence of 0.05% in the Granada Province (southern Spain), which is considerably lower than the MLE‐based prevalence of 1.9% estimated in this study. In contrast, Remoli et al. (2016) detected phleboviruses in 6 out of 50 pools analysed (12.0%) in the Community of Madrid whereas Collao et al. (2010) identified 12 positive pools among the 103 tested (11.6%) from Granada and Ibiza Island (Spain), including sequences corresponding to MASV. Despite the similar proportion of positive pools reported in these studies and in the present study (8/64; 12.5%), direct comparisons should be interpreted with caution, as they are affected by pool size and the proportion of positive pools does not directly reflect individual infection prevalence.

Overall, pooled sampling proves useful for surveillance in low‐prevalence settings, whereas individual testing remains essential for confirmation and co‐infection studies (Huemer et al. 2017). Since most studies rely on pooled samples, extrapolating our findings from individually analysed specimens remains challenging. Nevertheless, a study conducted in France individually screened 427 specimens and identified five phlebovirus‐infected individuals, although none tested positive for TOSV (Peyrefitte et al. 2013). The resulting infection rate was 1.2% (5/427), compared to the 0.45% (2/443) observed in our study. These findings underscore the need for further efforts to assess individual infection status, which is critical for elucidating co‐infection dynamics at the specimen level. In the present study, TOSV was detected in one individual specimen in each sampling year (2019: 1/205 individuals; 2022: 1/238 individuals). However, because sampling was limited to two non‐consecutive years, these data do not allow any inference of temporal trends or the stability of phlebovirus prevalence in the study area. We also compared these results with those reported by Sanbonmatsu‐Gámez et al. (2005), who sampled the same site (Mantas, Table 1) more than a decade earlier. This comparison revealed an increase in prevalence over time, as well as several silent polymorphisms among the sequences. The sequences detected for each genomic segment in the pools were identical to each other for each phlebovirus, but showed several polymorphic differences compared with individual specimens. In addition, we report new sequences of MASV and ACBV that differ from those previously described by other authors in earlier years (Collao et al. 2010; Amaro et al. 2015; Davó et al. 2020), although all of them were also synonymous at the amino acid level. This finding underscores the importance of molecular surveillance, as mutations in RNA viruses may lead to increased pathogenicity (Bakhshandeh et al. 2021).

Phlebotomus perniciosus and Phlebotomus perfiliewi are thought to be the main vectors of TOSV in the Mediterranean Basin (Hong et al. 2024). Moreover, viral RNA has been detected in other species such as Sergentomyia minuta, Phlebotomus longicuspis and Phlebotomus sergenti, suggesting their role as potential vectors (Charrel et al. 2006; Es‐Sette et al. 2014, 2015). Recent findings indicate that sand fly susceptibility to TOSV infection is influenced by both viral lineage and sand fly species. Notably, Phlebotomus tobbi and, to a lesser extent, P. sergenti were found to be susceptible to experimental infection with TOSV‐lineage B and therefore capable of transmitting the virus (Jancarova et al. 2025). Here, we report the natural infection of TOSV‐lineage B in P. perniciosus (6 pools +2 individuals), P. ariasi (1 pool) and S. minuta (1 pool). These species of sand flies are thought to be either vectors of TOSV or contributors to maintaining the virus and its transmission (Ayhan et al. 2020; Rocha et al. 2025). To our knowledge, this is the first report of P. ariasi being naturally infected by TOSV and the second record of S. minuta infected by the same virus—representing the first time this species has been found infected in Spain. Phylogenetic analyses show that the circulating strains of TOSV based on L and S segments in southern Spain belong to lineage B, which is consistent with previous reports from the same area (Sanbonmatsu‐Gámez et al. 2005).

Although considered a neglected pathogen, TOSV continues to be the main cause of meningitis during the summer months and its seroprevalence in the Mediterranean region highlights the need to strengthen epidemiological surveillance (Martinez et al. 2022; Rocha et al. 2025). Furthermore, while the animals that may act as potential reservoirs for TOSV remain unknown, several seroepidemiological studies have detected neutralizing antibodies against the virus in various mammalian species (Ayhan et al. 2020; Xhekaj et al. 2025). In southern Spain, anti‐TOSV antibodies were detected in domestic animals (dogs, cats, sheep, goats, cows, horses and pigs), with a substantial proportion (36.2%) testing positive (Navarro‐Marí et al. 2011). In that same study, TOSV RNA was also detected in a goat from a locality considered to be hyperendemic for the virus. The widespread distribution of TOSV vectors on the Iberian Peninsula, together with the presence of these animals in an area where livestock farming is a major economic activity, highlights the need for further epidemiological studies involving both humans and animals to better elucidate this issue.

Massilia‐like virus has also been detected in humans from southern Spain through IgG detection (Collao et al. 2010; Navarro‐Marí et al. 2013), although its impact on human health remains unclear. Here, we detect MASV in three sand fly pools (Table 2). However, the L segment of one of them exhibited 94%–95% homology with the ACBV, which belongs to the Salehabad complex (Amaro et al. 2015). This species was first identified in the Arrábida region of Portugal and has also been detected in a pool of P. perniciosus from Formentera Island (Balearic Archipelago, Spain) (Amaro et al. 2015; Davó et al. 2020). Interspecies phleboviruses resulting from genetic reassortments have been previously reported, which may lead to increased pathogenicity or virulence (Harris et al. 2025). Because these sequences were obtained from a pool, it is not possible to determine whether the L and S segments originated from a single infected specimen—potentially having undergone genetic reassortment—or from two distinct sand fly individuals infected within the same pool. Confirmation of reassortment requires sequencing of individually processed specimens and should be a priority for future studies. In most positive samples of this study, only one of the two genomic segments (L or S) was successfully amplified. This may reflect a low viral load in naturally infected sand flies and the variable performance of generic degenerate primers across phlebovirus species and genomic segments. Comparative assessments have shown that these primer sets, including the nested primers used here, yield weak or absent amplification for certain phlebovirus strains owing to primer–template mismatches (Polat et al. 2024). In addition, RNA secondary structures can reduce the accessibility of the target molecule and therefore the sensitivity of RT‐PCR assays (Weidmann et al. 2008). While pooled sampling may increase the sensitivity of detecting infected specimens, analysis of individual sand flies remains essential for confirming reassortment events. Similarly, further research is warranted to assess the impact of MASV and ACBV on human health.

Interactions between phleboviruses and Leishmania, as well as with the endosymbiont Wolbachia, remain unclear. The circulation of TOSV and Leishmania in the same area has previously been documented, and in some cases, both pathogens have been detected within the same pool of sand flies (Ergunay et al. 2014; Calzolari et al. 2021). Indeed, the co‐circulation of MASV and L. infantum has also been reported (Faucher et al. 2014). Here, we found no co‐infection between Phlebovirus and Leishmania in any of the female P. perniciosus individuals analysed. The experimental evidence on Phlebovirus–Leishmania interactions is still limited. In vertebrate hosts and cell cultures, phleboviral co‐infection generally enhances Leishmania infection (Rossi et al. 2017; Rath et al. 2019). In contrast, the only experimental study conducted in sand flies found competitive interactions, with L. infantum reducing TOSV infection rates in P. tobbi, which may explain why co‐infected sand flies are rarely detected in the field (Stejskalova et al. 2026). These interactions may involve the vector's immune responses and midgut microbiome. Accordingly, our finding should be interpreted with caution, as the absence of TOSV–Leishmania co‐infection is compatible with a potential negative interaction; however, this interpretation is limited by the very small number of TOSV‐positive specimens (n = 2).

The relatively high prevalence of L. infantum infection detected in P. perniciosus in this study (27.6%) underscores the ongoing public health relevance of leishmaniasis in Spain. Notably, this figure is consistent with the L. infantum infection rate of 24.0% recently reported for P. perniciosus collected in and around the dwellings of autochthonous human leishmaniasis cases in areas of Granada and Jaén provinces, where incidence rates of 15.1 and 8.3 human cases/100,000 inhabitants/year were reported, respectively (Martín‐Sánchez et al. 2026). Furthermore, the public health importance of L. infantum has been highlighted by other outbreaks. The one that occurred in the Madrid region (2009–2012) remains the largest documented outbreak of human leishmaniasis in Western Europe, with an incidence of 22.2 cases/100,000 inhabitants/year (Arce et al. 2013). Another outbreak was reported in Córdoba (southern Spain) between 2017 and 2021, resulting in a cumulative incidence of 10.3 cases/100,000 inhabitants (Barbero‐Moyano et al. 2025). In all instances, P. perniciosus was identified as the main vector. These findings illustrate the continuing capacity of L. infantum to cause significant human disease in Spain and underscore the importance of detecting high infection rates in vector populations as part of entomological surveillance.

The simultaneous presence of Wolbachia and Leishmania within the same sand fly vector has been reported in neotropical phlebotomine species (Rosário et al. 2022). In this study, Wolbachia–Leishmania co‐infection was detected in 40 P. perniciosus specimens, representing 66.7% of L. infantum‐infected individuals of this sand fly species. However, this high proportion may simply reflect the high background prevalence of Wolbachia rather than a specific biological interaction between the two microorganisms. This high Wolbachia prevalence is consistent with previous reports for P. perniciosus in the same region, where infection rates above 60% have been documented (Torres‐Llamas et al. 2025). The Wolbachia strain detected in P. perniciosus in this study corresponded to wPrn, with 100% homology to the strain previously identified in this species in Spain, Italy and Morocco (Torres‐Llamas et al. 2025). When L. infantum parasite loads were compared, no significant difference was observed between Wolbachia‐infected and Wolbachia‐uninfected females (p = 0.4). Although different Wolbachia strains have been proposed to interact with Leishmania spp., either enhancing or inhibiting parasite development (Torres‐Llamas et al. 2025), the role of wPrn in L. infantum infection remains unclear. This question can only be addressed through broader epidemiological and experimental studies, ideally incorporating quantification of Wolbachia density. Furthermore, the two TOSV‐infected individuals also harboured Wolbachia, showing that co‐infection with phleboviruses and wPrn can occur in P. perniciosus.

Although Wolbachia can inhibit the replication of several RNA viruses in other insects, such antiviral effects likely depend on the strain and the host (Teixeira et al. 2008; Moreira et al. 2009). Moreover, given the limited number of TOSV‐positive individuals, no mechanistic conclusions can be drawn regarding the interaction between wPrn and TOSV. Further studies including larger numbers of phlebovirus‐positive specimens, ideally combined with Wolbachia quantification, are required to assess whether wPrn modulates phlebovirus infection in P. perniciosus. Finally, no triple infection was detected in individual P. perniciosus specimens, although it was observed in one of the analysed pools (Table 2). However, it could not be confirmed whether the pathogens were present in the same individual.

Some limitations of this study should be noted, such as the small number of specimens analysed, the absence of infectivity assays, and the lack of data on vertebrate hosts, which can be mainly addressed through One Health approaches.

5. Conclusions

We confirm the circulation of TOSV‐lineage B, along with other phleboviruses, in southern Spain. While the estimated prevalence has increased when compared with data from more than a decade ago in that region, further longitudinal studies are needed to monitor the circulation of TOSV and estimate its prevalence over time. Despite differences in infection prevalence between individually analysed specimens and pools, both approaches provide valuable insights in epidemiological contexts. We also report new strains of TOSV‐lineage B, MASV and ACBV. Although we detected several polymorphisms across the three viral species, their amino acid sequences remain relatively conserved. Our study provides the first evidence of co‐infection with TOSV‐lineage B and Wolbachia in P. perniciosus, as well as a high percentage of specimens co‐infected with Wolbachia and L. infantum. Understanding these co‐infection dynamics is essential to elucidate the potential role Wolbachia may play in modulating other pathogens in sand flies, as well as for assessing the potential impact of Leishmania–Phlebovirus co‐infection on transmission to humans and other animals.

Author Contributions

Andrés Torres‐Llamas: writing – review and editing, writing – original draft, visualization, validation, methodology, formal analysis, data curation, conceptualization. Antonio Sampedro: writing – review and editing, visualization, validation, resources, investigation, formal analysis. Victoriano Díaz‐Sáez: visualization, validation, resources, methodology, investigation, formal analysis. Manuel Morales‐Yuste: writing – review and editing, visualization, validation, software, investigation, data curation. Javier Rodríguez‐Granger: visualization, validation, resources, investigation, formal analysis, data curation. Irene Pedrosa‐Corral: visualization, validation, resources, investigation, formal analysis. José María Navarro‐Marí: writing – review and editing, visualization, validation, resources, investigation, formal analysis. Francisco Morillas‐Márquez: writing – review and editing, visualization, validation, methodology, investigation, data curation. Joaquina Martín‐Sánchez: writing – review and editing, writing – original draft, visualization, validation, supervision, project administration, funding acquisition, formal analysis, data curation, conceptualization.

Funding

This study was funded by the Spanish Ministry of Science, Innovation and Universities, MICIU/AEI/10.13039/501100011033 and FEDER, EU through project PID2022‐142230NB‐I00. Andrés Torres‐Llamas was financed by a predoctoral contract FPI of the Spanish Ministry of Science, Innovation and Universities and FSE+ (grant number PREP2022‐000570). Funding for open access charge: Universidad de Granada/CBUA.

Ethics Statement

This study did not involve humans or animals and therefore did not require approval by an ethics committee.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting Information: S1. Primers used in the L segment.

Data Availability Statement

Data is provided within the manuscript or Supporting Information. Eight sequences have been deposited in GenBank under the following accessions: PX982977–PX982984.