Novel Genotypes of Nidicolous Argas Ticks and Their Associated Microorganisms From Spain

The knowledge of the distribution, richness and epidemiological importance of soft ticks of the genus Argas is incomplete. In Spain, five Argas species have been recorded, including three ornitophilic nidicolous ticks, but their associated microorganisms remain unknown. This study aimed to investigate ticks from bird nests and their microorganisms. Ticks were collected extensively from natural cavities and nest-boxes used by European rollers (Coracias garrulus) and little owls (Athene noctua) in Southeastern and Central Spain. Ticks were morphologically and genetically identified and corresponding DNA/RNA tick extracts were analyzed [individually (n = 150) or pooled (n = 43)] using specific PCR assays for bacteria (Anaplasmataceae, Bartonella, Borrelia, Coxiella/Rickettsiella, and Rickettsia spp.), viruses (Flaviviruses, Orthonairoviruses, and Phenuiviruses), and protozoa (Babesia/Theileria spp.). Six Argas genotypes were identified, of which only those of Argas reflexus (n = 8) were identified to the species level. Two other genotypes were closely related to each other and to Argas vulgaris (n = 83) and Argas polonicus (n = 33), respectively. These two species have not been previously reported from Western Europe. Two additional genotypes (n = 4) clustered with Argas persicus, previously reported in Spain. The remaining genotype (n = 22) showed low sequence identity with any Argas species, being most similar to the African Argas africolumbae. The microbiological screening revealed infection with a rickettsial strain belonging to Rickettsia fournieri and Candidatus Rickettsia vini group in 74.7% of ticks, mainly comprising ticks genetically related to A. vulgaris and A. polonicus. Other tick endosymbionts belonging to Coxiella, Francisella and Rickettsiella species were detected in ten, one and one tick pools, respectively. In addition, one Babesia genotype, closely related to avian Babesia species, was found in one tick pool. Lastly, Anaplasmataceae, Bartonella, Borrelia, and viruses were not detected. In conclusion, five novel Argas genotypes and their associated microorganisms with unproven pathogenicity are reported for Spain. The re-use of nests between and within years by different bird species appears to be ideal for the transmission of tick-borne microorganisms in cavity-nesting birds of semiarid areas. Further work should be performed to clarify the taxonomy and the potential role of soft Argas ticks and their microorganisms in the epidemiology of zoonoses.

The knowledge of the distribution, richness and epidemiological importance of soft ticks of the genus Argas is incomplete. In Spain, five Argas species have been recorded, including three ornitophilic nidicolous ticks, but their associated microorganisms remain unknown. This study aimed to investigate ticks from bird nests and their microorganisms. Ticks were collected extensively from natural cavities and nest-boxes used by European rollers (Coracias garrulus) and little owls (Athene noctua) in Southeastern and Central Spain. Ticks were morphologically and genetically identified and corresponding DNA/RNA tick extracts were analyzed [individually (n = 150) or pooled (n = 43)] using specific PCR assays for bacteria (Anaplasmataceae, Bartonella, Borrelia, Coxiella/Rickettsiella, and Rickettsia spp.), viruses (Flaviviruses, Orthonairoviruses, and Phenuiviruses), and protozoa (Babesia/Theileria spp.). Six Argas genotypes were identified, of which only those of Argas reflexus (n = 8) were identified to the species level. Two other genotypes were closely related to each other and to Argas vulgaris (n = 83) and Argas polonicus (n = 33), respectively. These two species have not been previously reported from Western Europe. Two additional genotypes (n = 4) clustered with Argas persicus, previously reported in Spain. The remaining genotype (n = 22) showed low sequence identity with any Argas species, being most similar to the African Argas africolumbae. The microbiological screening revealed infection with a rickettsial strain belonging to Rickettsia fournieri and Candidatus Rickettsia vini group in 74.7% of ticks, mainly comprising ticks genetically related to A. vulgaris and A. polonicus. Other tick endosymbionts belonging to Coxiella, Francisella and Rickettsiella species were detected in ten, one and one tick pools, respectively. In addition, one Babesia genotype, closely related to avian Babesia species, was found in one tick pool. Lastly, Anaplasmataceae, Bartonella, Borrelia, and viruses were not detected. In conclusion, five novel Argas genotypes and their associated microorganisms with unproven pathogenicity are reported for Spain. The re-use of nests between and within years by different bird species appears to be ideal for the transmission of tick-borne microorganisms in cavity-nesting birds of semiarid areas. Further work should be performed to clarify the taxonomy and the potential role of soft Argas ticks and their microorganisms in the epidemiology of zoonoses.
The lack of information on the natural history and distribution of various Argas species, their incorrect or incomplete taxonomic description, and the fact that some species share morphological features but have not been molecularly examined, are responsible for the poor knowledge of Argas ticks in Spain. Moreover, their role in the epidemiology of tickborne microorganisms has not been studied in this country. Here, we aim at describing soft ticks from natural and artificial nests occupied by different cavity-nesting birds in Spain and the prevalence of selected tick-borne microorganisms.

Study Area and Study System
The main study area (∼50 km 2 ) lies in the Desert of Tabernas (Almería province, SE Spain, 37.08 • N, 2.35 • W). The landscape mostly consists of open shrubland with olive and almond groves interspersed among numerous dry riverbeds-ramblas. The climate in this area is semiarid Mediterranean with a marked water deficit during long, hot summer months. The mean annual rainfall is ∼ 230 mm, with high inter-and intra-annual variability (18). Tick samples also were collected in Segovia and Guadalajara provinces (both in the interior of the Iberian Peninsula), whose climate is Mediterranean with some continental characteristics.
In the main study area in Almería, natural cavities in sandstone cliffs, seminatural cavities in stone bridges and abandoned farmhouses and nest boxes provide nest sites for cavity-nesting birds, namely the European roller (Coracias garrulus, hereafter roller), the little owl (Athene noctua) and the rock/feral pigeon (Columba livia, hereafter pigeon). In this study, we sampled ticks in cavities occupied by rollers and little owls. The roller is a migratory bird that arrives at its breeding grounds in the study area during the second fortnight of April whereas the little owl is a resident bird. Both species rear a single brood per year (19). In contrast to these species, the pigeon is a resident bird that breeds at any time of the year in our study area and does not use nest boxes. Other species breeding in natural and seminatural cavities mainly include jackdaws (Corvus monedula), and common kestrels (Falco tinnunculus), whereas Scops owls (Otus scops), spotless starlings (Sturnus unicolor), and house sparrows (Passer domesticus) can breed in nest boxes.
Given nest-site limitation in the study area, both intraand interspecific competition for suitable nesting holes occur and individual cavities can be re-used by different species both within and between years. This is frequently the case in Almería, so that many samples were collected from natural and seminatural cavities of rollers and little owls previously used by pigeons. The samples from Segovia were collected from a natural tree hole occupied by rollers but excavated by the Iberian green woodpecker (Picus sharpei), whereas the samples from Guadalajara were taken from rollers breeding in nest boxes.

Tick Collection and Preservation
In the framework of a long-term project of cavity-nesting birds in the Desert of Tabernas, cavities and nest boxes have been routinely inspected during each breeding season since 2005 and both nestlings and nest material periodically examined for ectoparasites. Ticks were collected from cavities occupied by breeding rollers and little owls during 2009, 2012, 2015, and 2018-2020. Additionally, four tick individuals were obtained from roller nests in Central Spain (Guadalajara and Segovia) in 2004 ( Table 1). Ticks collected until 2018 were preserved in ethanol, while ticks obtained in 2019 and 2020 were kept fresh upon delivery to the Centre of Rickettsiosis and Arthropod-borne Diseases (CRETAV). Before frozen at −80 • C until later analysis, fresh ticks were identified and a single leg of each specimen was dissected.

Tick Identification
The taxonomic identification of the ticks was carried out using morphological keys (20)(21)(22)(23). Tick individuals were surfacesterilized and DNA was individually extracted from a single leg of each tick specimen using incubations with ammonium hydroxide (24). The obtained DNA templates were used for genetic characterization by the amplification and sequencing of the 16S rRNA fragment gene (25). Two other mitochondrial genes, 12S rRNA, and cytochrome oxidase subunit I (COI), were also used in analyses (Supplementary Table 1).

Microbial Screening
Ticks were pooled (from 1 to 7 specimens; whole larvae and body halves for the other life stages) according to tick species or genotype, origin and date of collection and, when possible, tick developmental stage. DNA extracts from pools of ticks preserved in ethanol were obtained using the Qiagen DNA DNeasy blood and tissue kit (Qiagen, Hilden, Germany), following the manufacturer's recommendations. Ticks of each pool that were preserved frozen were homogenized in 600 µL of culture medium with antibiotics [Dulbecco's Modified Eagle Medium (DMEM) with 100 units/mL penicillin and 100 µg/mL streptomycin, Sigma]. Four hundred µL of the homogenate were used for nucleic acid extraction (DNA and RNA) and the remaining 200 µL were preserved at −80 • C for future analysis. The DNA and RNA were extracted using the DNeasy blood and tissue kit and RNeasy Mini kit (Qiagen, Hilden, Germany), respectively, following the manufacturer's recommendations. The quality of nucleic acid extraction was checked using the 16S rRNA PCR assay (25). Positive samples were subjected to microbial screening using specific PCR assays for the analysis of (i) bacteria: Anaplasmataceae family, Bartonella spp., Borrelia spp., Coxiella spp., Rickettsiella spp., and spotted fever group (SFG) Rickettsia spp., (ii) viruses: Flaviviridae, Orthonairoviridae and Phenuiviridae families, and (iii) protozoans: Babesia and Theileria spp. Negative and positive controls (DNA or cDNA extracts of Anaplasma phagocytophilum, Bartonella henselae, Borrelia spielmanii, Borrelia miyamotoi, Coxiella-like of Rhipicephalus bursa, Rickettsia amblyommatis, Crimean-Congo haemorrhagic fever virus, Japanese encephalitis virus, Uukuniemi uukuvirus virus, and Babesia sp. from Rhipicephalus microplus) were included in all the PCR assays performed. Primers and PCR conditions are described in the Supplementary Table 1. The SFG Rickettsia was screened in tick legs for all ticks. In addition, pools formed by specimens that gave negative results were also tested for Rickettsia spp. Moreover, all the pools were screened for the presence of the remaining bacteria and protozoans. Lastly, the viral screening was performed on tick pools comprising specimens of fresh/frozen ticks ( Table 1).

Prevalence of Infection
The prevalence of infection (PI) was estimated by: PI = (No. of positive ticks/total No. of ticks analyzed) × 100%. When microorganisms were amplified from pools of more than one tick, prevalence was calculated assuming that each positive pool contained only one positive tick. This estimate, known as minimum infectious rate (MIR), is calculated as follow: MIR = (No. of positive pools/total No. of individual ticks analyzed) × 100%.

Analysis of Nucleotide Sequences
Nucleotide sequences were analyzed using the BLAST search (https://blast.ncbi.nlm.nih.gov/Blast.cgi), and the resulting sequences were submitted to GeneBank. The Clustal Omega online software (https://www.ebi.ac.uk/Tools/msa/clustalo/) was used for multiple sequence alignment. Phylogenetic analyses were conducted with MEGA X (http://www.megasoftware. net) using the maximum likelihood method including all sites. The nucleotide substitution model was selected according to the Akaike information criterion implemented in MEGAX. Confidence values for individual branches of the resulting trees were determined by bootstrap analysis with 500 replicates.

Tick Identification
One hundred and sixty-three ticks, mainly nymphs, were collected from bird nests in Almería (n = 159), Guadalajara (n = 2), and Segovia (n = 2). Arthropods were obtained from nest material in cavities occupied by little owl (n = 129) and roller (n = 34) or, in few cases, from nestlings of these species ( Table 1). All the specimens were morphologically identified as Argas spp. and 150 specimens were further studied molecularly. Examination of morphological characters enabled the identification of eight nymphs as A. reflexus, but the morphological identification of the remaining ticks could not be accurately performed with available keys. The A. reflexus identification was corroborated molecularly based on 16S rRNA fragment gene ( Table 2). The molecular identification was not conclusive for the remaining 142 tick samples, which were grouped based on the 16S rRNA results into five different genotypes, designated as Argas spp. EEZA-CRETAV1-5 (Tables 1, 2). Based on 16S rRNA gene analyses, Argas sp. EEZA-CRETAV1 (n = 83) and Argas sp. EEZA-CRETAV2 (n = 33) were closest to A. vulgaris and A. polonicus, respectively ( Table 2). In turn, the Argas sp. EEZA-CRETAV3 genotype (n = 22) shared the highest identity (<92.2%) with Argas africolumbae. Lastly, Argas sp. EEZA-CRETAV4 (n = 2) and Argas sp. EEZA-CRETAV5 genotypes (n = 2) shared highest identities with A. persicus ( Table 2). The phylogeny inferred from 16S rRNA analysis corroborates the BLAST results (Figure 1). Phylogenetic analyses based on 12S rRNA and COI fragment genes could not be performed because of the lack of homologue sequences for the majority of Argas spp. Argas spp. EEZA-CRETAV1-3 specimens were collected in Almería, whereas those of Argas spp. EEZA-CRETAV4-5 were obtained in Segovia and Guadalajara. Also, some of the ticks of Argas spp. EEZA-CRETAV1-2 genotypes were collected from the same nests (Table 1).

Bacterial Screening
All the DNA extracts (individual samples and pools) gave positive results for the tick-16S rRNA PCR assay and, consequently, were screened for bacteria ( Table 3). Amplicons for ompA gene were obtained from 112 DNA extracts from tick legs (PI = 74.7%). Specifically, Rickettsia was amplified from 81 and 31 samples belonging to Argas sp. EEZA-CRETAV1 (PI = 97.6%) and Argas sp. EEZA-CRETAV2 (PI = 93.9%) samples, respectively. DNA extracts corresponding to individual tickleg samples from A. reflexus and Argas sp. EEZA-CRETAV3-5 specimens were negative, but body-halve extracts of these specimens were further analyzed using pooled samples. The nucleotide extracts from these pools were negative for all A. reflexus specimens, but yielded positive results for two out of six pools of Argas sp. EEZA-CRETAV3 specimens (MIR = 9.1%) (comprising a nymph collected from a roller nest in Almería in 2015 and larvae that were attached to a roller nestling in Almería in 2020), and all the pools of Argas sp. EEZA-CRETAV4-5 specimens (MIR = 100% for the two pools) ( Table 3).
All the ompA gene sequences obtained were identical and showed the highest identity with Rickettsia fournieri ( Table 4). Selected Rickettsia-positive samples were further genetically characterized by the amplification of five more rickettsial fragment genes (26). Nucleotide sequences for the respective genes were identical and showed highest identities with R. fournieri and Candidatus Rickettsia vini ( Table 4). The phylogenetic tree based on the concatenated fragment genes of the Rickettsia strain detected, designated as Rickettsia sp. EEZA-CRETAV, corroborated the close relation with both R. fournieri and Ca. R. vini (Figure 2).
A total of 27 nucleotide sequences were obtained using the rpoB PCR assay selected for the Coxiella/Rickettsiella detection but highest identities with validated bacterial species reached < 85% for 14 samples. The genetic analysis of the amplicon obtained from an Argas sp. EEZA-CRETAV1 nymph, collected in Almería in 2012, showed the highest identity with Francisella persicus (PI = 0.7% and PI = 1.2% for Argas sp. EEZA-CRETAV1) (Tables 3, 5). The novel Francisella strain molecularly described in this study was designated as Francisella sp. EEZA-CRETAV. Coxiella-like strains were successfully amplified from 10 pools (MIR = 6.7%). Two out of three A. reflexus pools (with ticks collected in a little owl nest in Almería in 2018) showed infection with a Coxiella strain previously detected in this tick species (MIR = 25%) ( Table 5) Table 3). Nucleotide sequences corresponding to rpoB an groEL genes of these two novel strains shared 95.3 and 97% identity, respectively, and reached highest identities with Coxiella strains detected in Ornithodoros ticks (Table 5). Moreover, pools integrated over a single specimen were also submitted for groEL analysis and a nymph belonging to Argas sp. EEZA-CRETAV2, collected in Almería in 2018, showed infection with Rickettsiella sp. (PI = 0.7% and PI for Argas sp. EEZA-CRETAV2 = 3%; Table 3). The corresponding amplicon, designated as Rickettsiella EEZA-CRETAV, showed highest identities with Rickettsiella species amplified from Ornithodoros normandi ( Table 5).
All the tick pools were examined by PCR assays for the presence of Anaplasmataceae, Bartonella, and Borrelia species and all gave negative results.

Sequences Submission to a Public Database
The nucleotide sequences of ticks and microorganisms detected in this study (n = 42) were deposited in the GenBank database under accession numbers showed in Table 6.

DISCUSSION
Soft ticks are important vectors of microbial agents of animal and human diseases. Despite this, the knowledge of argasid tick species and their associated microorganisms is generally scarce. Herein, the occurrence of five novel Argas spp. genotypes, in addition to A. reflexus, collected in the Iberian Peninsula (Spain) in nests occupied by little owls and European rollers is reported. Moreover, we detected the presence of tickborne microorganisms belonging to genera Rickettsia, Coxiella, Francisella, Rickettsiella, and Babesia in Argas ticks. In contrast, Argas sp. Frontiers in Veterinary Science | www.frontiersin.org

Tick Identification
Of the six Argas genotypes detected in this study, only one could be identified to the species level, namely, A. reflexus. The morphological identification of ticks is challenging, even for experts (27), while the molecular approach appears to be an accurate tool for tick identification (25,27). Nevertheless, increased effort for molecular characterization of more Argas species is needed for a reliable taxonomic inference based on molecular tools. In order to confirm if the genotypes identified in this study represent validated or potentially novel tick species, a further morphological analysis including unfed larva specimens, should be performed along with rigorous molecular analyses of Argas ticks from diverse geographical locations.
Argas reflexus specimens, known as the pigeon tick, have been collected in two bird nests in Almería (Southern Spain), one occupied by little owl (n = 7) and one by roller (n = 1). This species occurs in Spain, and it is a well-known ectoparasite of little owls, whereas information regarding roller infestations is scarce (6,20,28). Argas sp. EEZA-CRETAV4 and Argas sp. EEZA-CRETAV5, genotypes amplified in Central Spain, clustered molecularly with A. persicus from China (29), but in a different branch than other A. persicus specimens (Figure 1). Nevertheless, the broad genetic divergence of this group (29,30), also revealed in the phylogeny inferred herein (Figure 1), suggests that cryptic species could occur in this taxon known as fowl tick. This worldwide distributed tick has been previously reported in Spain and is known to infest wild birds also in other countries (6,28). In addition, ticks of three more Argas genotypes have been identified, all of them in Almería. The Argas sp. EEZA-CRETAV1-2 genotypes clustered together and appear to be closely related with A. vulgaris and A. polonicus, respectively (Figure 1). Neither of these two tick species have been previously reported from western Europe and their occurrence is only documented in a few eastern European countries (20,21,31). These genotypes have been found in the nest material of roller and little owls, occurring in the same nests in several cases ( Table 1). In the study area, there is a high competition among cavity-nesting birds for suitable cavities and the same cavity can be successively used by little owls, pigeons and rollers (32). This fact could explain the infestation of both bird species by the same nidicolous tick taxa, i.e., A. reflexus and Argas sp. EEZA-CRETAV1-2. In contrast, Argas sp. EEZA-CRETAV3, closely related to A. africolumbae, has been found only in nests occupied by rollers. The roller is a long-distant migrant species (trans-Saharan migrant) and the Spanish populations overwinter in different African regions (33). It is worth mentioning that A. africolumbae is an ornithophilic tick that occurs in Africa: South Africa, Kenya, Tanzania and Burkina Faso (34,35). It is wellknown that birds can serve as dispersers of ticks and tick-borne microorganisms, even though this information pertains mainly to hard ticks (36,37). Some studies suggest that the role of birds as dispersers of soft ticks is less important, due to the biology of these ticks (nidicolous behavior and shorter blood-feeding time), but the role of migratory birds as reservoirs or amplifiers of tickborne microorganisms associated with soft ticks remains to be better investigated (38,39).

Tick-Borne Microorganisms
The microbiological screening of ticks is important to identify the local risks of emergence of tick-borne diseases. Argas species, including A. reflexus and A. persicus, have been described as important pests and vectors of diseases in poultry and wild birds, being responsible for high economic losses (7,11,40). These tick species have also been recorded biting humans and causing anaphylaxis episodes (8,10). Although humans are accidental hosts of Argas ticks and the ticks carry numerous microorganisms, human pathogens among them, the role of these ticks as vectors of human infectious agents has not been proven.
The most prevalent microorganisms amplified in this study belong to the Rickettsia genus (α-Proteobacteria; Rickettsiaceae). The high prevalence of Rickettsia spp. found in ticks of Argas sp. EEZA-CRETAV1-2 specimens suggests that the bacterium is a common endosymbiont of ticks of the two genotypes. The detection of the newly described Rickettsia strain in tick-body samples, but not in tick-leg samples of ticks of other genotypes (Argas sp. EEZA-CRETAV3-5), suggests that this Rickettsia species may not be a true intracellular endosymbiont and its presence in the former genotypes could be acquired through feeding on infected hosts or by cofeeding. The phylogenetic analysis of Rickettsia sp. EEZA-CRETAV reveals its close relation with R. fournieri and Ca. R. vini, both Rickettsia species associated with ornithophilic nidicolous ticks (41,42). While R. fournieri  has been described only once from A. lagenoplastis in Australia (41), Ca. R. vini has been detected in several European countries in Ixodes spp. (36,(42)(43)(44)(45). The single study performed suggests that Ca. R. vini is not pathogenic (43). Nevertheless, the two Rickettsia taxa are closely related to other Rickettsia species that are recognized as human pathogens, specifically Rickettsia japonica and Rickettsia heilongjiangensis (46,47). Thus, the epidemiology and pathological potential of Rickettsia strains such as Rickettsia sp. EEZA-CRETAV, in addition to R. fournieri and Ca. R. vini, should be further investigated. Likewise, the isolation of Rickettsia sp. EEZA-CRETAV is necessary to gain an insight into the epidemiological importance of this strain. In addition to the Rickettsia taxon, this study has revealed for the first time different proteobacterial tick endosymbionts in Argas spp. from Spain, namely, Coxiella, Rickettsiella (Gamma-proteobacterium; Coxiellaceae) and Francisella (Gamma-proteobacterium; Francisellaceae) species. The detected species, commonly known as Rickettsiella-like, Coxiella-like, and Francisella-like, are intracellular obligatory endosymbionts important for tick survival. They play some role in B vitamins biosynthesis and their presence may interfere with the transmission (positively or negatively) of other microorganisms, including tick-borne pathogens (48). They are related to species responsible for important human diseases. For instance, Coxiella burnetii and Francisella tularensis cause Q fever and tularemia, respectively (49,50). Coxiella-like species have been implicated in human and avian diseases (51,52). Three different Coxiella strains have been successfully detected in this study. Of them, the Coxiella sp. from A. reflexus was homologous to the species previously detected in the same tick species, but the remaining two strains, designated as Coxiella sp. EEZA-CRETAV1-2, differ from the scarcely-known Coxiella-like strains of Argas species (14,(53)(54)(55). To date, Rickettsiella has been mainly reported in hard ticks (Ixodes spp. and Haemaphysalis spp.) and soft ticks belonging to the genus Ornithodoros (14). The presence of Rickettsiella in Argas ticks has been suggested for the bat tick A. transgariepinus from Namibia, but the available 16S rRNA sequences shared low identities with the published Rickettsiella sequences (17). In the present study, a Rickettsiella sp. similar to that of O. normandi from Tunisia (55) was detected in an Argas sp. EEZA-CRETAV2 specimen. Lastly, a strain closely related to F. persicus was found in an Argas sp. EEZA-CRETAV1 specimen. Francisella persicus, formerly Wolbachia persica, is an endosymbiont of Argas arboreus (previously referred to as A. persicus) (56). This bacterium has not been identified as an animal or human pathogen, but the analysis of its genome shows that this species conserves an important number of potentially functional virulence-associated genes, suggesting that it could be pathogenic to mammals (57).
Bacteria of Anaplasmataceae family, Bartonella, and Borrelia spp. have not been detected in the ticks analyzed. Argas spp. are recognized vectors of B. anserine, the agent of the avian  The GenBank accession number of the sequences used in this analysis is shown in brackets after Babesia taxon name. The species found in this study is marked with diamond and the species detected in birds with asterisk. Taxon names are followed by GenBank accession numbers and collection location where available. Plasmodium falciparium is used as outgroup. spirochetosis, a worldwide distributed disease of veterinary importance that has not been reported from Spain (11,40). The lack of virus detection in this study was unexpected, because diverse viruses are readily detected in Argas ticks (7). Scarce tick-borne viruses have been described from Spain and all but Meaban-like virus, a flavivirus found in Ornithodoros maritimus (38), are associated to ixodid ticks. Some of these viruses have a great relevance for human health, e.g., Crimean-Congo hemorrhagic fever virus, whose arrival in infected ticks has been suggested to take place through migratory birds (37,58). This fact highlights the importance of studying viruses in soft ticks associated to birds in Spain, an important area in the migratory routes of many avian species, because what it is not sought, it is not found (59). Two apicomplexan parasites have been found in this work, Babesia sp. and A. bambarooniae, though the latter species is not known as a tick-borne microorganism. In turn, the apicomplexan piroplasms Babesia (Aconoidasida; Babesiidae) are mainly vectored by ixodid ticks, though argasid ticks also were suggested as potential vectors (60). Babesia sp. EEZA-CRETAV has been amplified from an Argas sp. EEZA-CRETAV3 tick associated with rollers and the presence of the protozoan in the blood of rollers cannot be rejected. Sixteen Babesia species responsible for avian piroplasmosis, in addition to several strains that are not fully identified, are known (61). Of these, Babesia frugilegica, Babesia shortti and Babesia benneti have been reported from Spain (62)(63)(64). Babesia sp. EEZA-CRETAV is closely related to some of the scarcely genetically characterized Babesia species, mainly to B. ardeae (Figure 3). This species has been detected in Asia and its pathogenicity is unknown (61). This strain is also close to the human-pathogenic B. ducani that has been identified in North America, the United Kingdom and Australia (65). To our knowledge, Babesia spp. have not been identified in owls.

CONCLUSIONS
The sedentary lifestyle of soft ticks could imply a limited role of these ticks in the circulation of infectious agents (66). However, as indicated by this study, the high re-use of cavities within and between years by different bird species could importantly enhance the spread of microorganisms associated with soft nidicolous ticks, such as Argas ticks.
This study highlights the richness of nidicolous Argas ticks associated with cavity-nesting birds in a semi-desert area in Western Europe, and suggests that the diversity of this genus in Spain might be underestimated. Moreover, this work provides the first report of Rickettsia sp., Coxiella spp., Rickettsiella sp., Francisella sp. and Babesia sp., from soft ticks in Spain, and A. bambarooniae from Ixodida.
Further research should be carried out to confirm if the new genotypes of ticks and their microorganisms represent novel taxa and, additionally, to understand their role in the epidemiology of zoonoses using the One Health approach.

DATA AVAILABILITY STATEMENT
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/ Supplementary Material.