Introduction
Raptors are an ecological guild of predatory birds because they broadly share similar trophic roles, despite belonging to distinct evolutionary lineages and differing in prey types, hunting modes, activity periods and habitat use (Movalli et al., Reference Movalli, Krone, Osborn and Pain2018; McClure et al., Reference McClure, Schulwitz, Anderson, Robinson, Mojica, Therrien, Oleyar and Johnson2019). Currently, raptors include species within the orders Accipitriformes, Cathartiformes, Falconiformes, Cariamiformes and Strigiformes, which evolved from terrestrial lineages and predominantly maintain carnivorous lifestyles (McClure et al., Reference McClure, Schulwitz, Anderson, Robinson, Mojica, Therrien, Oleyar and Johnson2019). As long-lived apex predators occupying upper trophic levels, they play key ecological roles, including regulating prey populations and removing carrion, thereby providing essential ecosystem services (Movalli et al., Reference Movalli, Krone, Osborn and Pain2018). Their predatory habits, longevity and, in some taxa, nocturnal activity expose them to a wide diversity of parasites via consumption of infected prey or through vector-mediated transmission, particularly when host activity patterns overlap with peak periods of hematophagous insect activity. For example, owls (Strigiformes) are active during periods of high mosquito and biting midge activity, which may increase their risk of infection (Valkiūnas, Reference Valkiūnas2004). Among blood parasites infecting raptors, haemosporidians (Phylum Apicomplexa; Order Haemosporida) are of relevance. Parasites of the genera Plasmodium, Haemoproteus and Leucocytozoon are distributed globally and occur in numerous avian families. Transmission occurs actively through various dipteran vectors, including mosquitoes (Culicidae), louse flies (Hippoboscidae), biting midges (Ceratopogonidae) and blackflies (Simuliidae) (Valkiūnas, Reference Valkiūnas2004).
A synthesis of global molecular lineage records reveals that reports of haemosporidians in raptors are concentrated primarily in Strigiformes, Accipitriformes and Falconiformes, with a total of 786 documented records (Supplementary Table 1). To date, out of approximately 557 raptor species, only 82 species are known to be susceptible to haemosporidian infection based on molecular screening; of the 786 total global parasite records, 95.9% (n = 754) are represented by lineages deposited in GenBank (Benson et al., Reference Benson, Cavanaugh, Clark, Karsch-Mizrachi, Lipman, Ostell and Sayers2013) or MalAvi (Bensch et al., Reference Bensch, Hellgren and Pérez-Tris2009) databases. Infections by Leucocytozoon (38.5%) and Haemoproteus (35.0%) predominate, whereas Plasmodium accounts for 25.6% of records (MalAvi database, accessed 23 March 2026). Geographically, studies in Asia report the highest number of lineages (29.8%), followed by Europe (29.2%) and North America (28.9%), while Africa (4.1%) and South America (7.6%) remain comparatively underrepresented. Most of these molecular records are derived from Strigiformes and Accipitriformes, which also correspond to the highest number of lineage reports in the databases. These patterns suggest that complex host–parasite–vector interactions are probably shaped by specific ecological and evolutionary processes that remain insufficiently understood (Harl et al., Reference Harl, Himmel, Valkiūnas, Ilgūnas, Nedorost, Matt, Kübber-Heiss, Alic, Konicek and Weissenböck2022, Reference Harl, Himmel, Pacheco and Weissenböck2026; Ilgūnas et al., Reference Ilgūnas, Himmel, Harl, Dagys, Valkiūnas and Weissenböck2022; Bell et al., Reference Bell, Driscoll, Achatz, Martens and Vaughan2025; Pacheco et al., Reference Pacheco, Miller, Harl, Ferreira, Cepeda, Valkiūnas, Beckerman, Oswald, Mateus-Pinilla and Escalante2025).
Despite these global patterns, raptors remain understudied compared to other avian groups, such as passerines. This disparity is partly methodological, as standard mist-net sampling is typically conducted at heights below 3 m and rarely captures high-flying raptors. Additional constraints include the legal protection status of these birds and their low population densities (Donázar et al., Reference Donázar, Cortés-Avizanda, Fargallo, Margalida, Moleón, Morales-Reyes, Moreno-Opo, Pérez-García, Sánchez-Zapata, Zuberogoitia and Serrano2016; Sarasola et al., Reference Sarasola, Grande, Bechard, Sarasola, Grande and Negro2018; Bukauskaitė et al., Reference Bukauskaitė, Dementavičius, Rumbutis and Treinys2024).
Colombia, located in northwestern South America, is one of the world’s most biodiverse countries due to its complex geological, climatic and hydrographic systems. With 1966 recorded bird species, this country harbours the highest avian diversity globally (Echeverry-Galvis et al., Reference Echeverry-Galvis, Acevedo-Charry, Avendaño, Gómez, Stiles, Estela and Cuervo2022). It ranks second worldwide in total raptor richness (104 species) and first in diurnal raptor diversity (76 species) (Cruz et al., Reference Cruz, Santulli-Sanzo and Ceballos2021). Nevertheless, only 3 haemosporidian lineages have been reported in Colombian raptors to date (MEGCHO01, PSDIS01 and ASICLA01) (Alvarez-Londoño et al., Reference Alvarez-Londoño, Cardona-Romero, Martínez-Sánchez, Ossa-López, Pérez-Cárdenas, Gonzalez, Rivera-Páez and Castaño-Villa2022; Lotta-Arévalo et al., Reference Lotta-Arévalo, González, Gamboa-Suárez, Pacheco, Escalante, Moreno, Rodríguez-Fandíño, Cuervo and Matta2023), underscoring a substantial knowledge gap in a region of exceptional biodiversity. Thus, the objective of this study was to determine the occurrence of Haemoproteus and Leucocytozoon parasites in raptors from Colombia, to characterize their species using morphological and molecular data and to establish their phylogenetic relationships.
Materials and methods
Samples and study areas
Blood smears and whole blood samples of 202 individuals, representing 30 species, 5 families and 3 orders of raptors deposited in the Biological Collection GERPH, were analysed in this study. Of these, 42 individuals were captured in the wild across 7 Colombian departments (Figure 1; Table 1; Supplementary Table 2). The remaining 160 individuals were opportunistically sampled at 3 wildlife rehabilitation centres (WRCs): the Wildlife Rescue and Rehabilitation Unit (URRAS) in Bogotá, Cundinamarca Department (138 individuals); the CARDER-APAP temporary shelter in La Florida, Pereira, Risaralda Department (17 individuals) and Los Ocarros Biopark in Villavicencio, Meta Department (5 individuals) (Table 1; Supplementary Table 2). Samples were collected from the birds immediately upon admission, prior to any clinical treatment or contact with other animals.
Political division of the Republic of Colombia. Departments highlighted in olive green indicate the regions where the raptors included in this study were sampled.
Figure 1 Long description
Colombia highlights sampled areas in olive green, including the departments of Narino, Cauca, Tolima, Boyacá, Risaralda, Caquetá, Cundinamarca, Meta and Casanare. The map includes a scale bar indicating distances in kilometers and a compass rose for orientation. An inset map of South America shows Colombia′s location, marked with a red box. Surrounding countries such as Panama, Venezuela, Ecuador, Peru and Brazil are labeled. The map also features elevation shading, indicating mountainous regions primarily in the western and central parts of Colombia.
Occurrence of Haemoproteus and Leucocytozoon species across different raptor families. CytB: partial mitochondrial cytochrome b gene sequence. Mt: near-complete mtDNA genome
Table 1 Long description
Rows list raptor groups and species, with columns for parasite species and occurrence plus CytB and Mt sequence identifiers. Under Hawks and Eagles, Broad winged Hawk had Haemoproteus sp. at 25 percent with 2 of 8, and L. toddi at 37.5 percent with 3 of 8, with CytB listed as L. toddi PZ559900 to PZ559901 and Mt shown as a dash. Swainsons Hawk had Haemoproteus sp. at 100 percent with 1 of 1, with dashes for Leucocytozoon, CytB, and Mt. Mississippi Kite had Haemoproteus sp. at 50 percent with 1 of 2, with dashes for Leucocytozoon, CytB, and Mt. Osprey had Leucocytozoon sp. at 50 percent with 1 of 2, CytB PANHAL02, and Mt shown as a dash, with a dash for Haemoproteus. Roadside Hawk had H. multivacuolatus like at 4.5 percent with 1 of 22, with dashes for Leucocytozoon, CytB, and Mt. Under Falcons, Yellow headed Caracara had H. trarotraro like at 14.2 percent with 1 of 7, with dashes for Leucocytozoon, CytB, and Mt. Under Owls, Striped Owl had H. tatilalaorum at 29.4 percent with 5 of 17, CytB PZ559886 to PZ559889, and Mt PZ575010. Stygian Owl had H. tatilalaorum at 8.3 percent with 2 of 24, CytB PZ559890, and Mt PZ575011. Tropical Screech Owl appears twice, once with H. huizhyanus at 6.9 percent with 3 of 43, CytB PZ559891 to PZ559893, and Mt PZ575012 to PZ575013, and once with H. homohuizhyanus at 16.2 percent with 7 of 43, CytB PZ559894 to PZ559898, and Mt PZ575014 to PZ575015.
Blood was obtained via venipuncture of the tibiotarsal, basilic or jugular veins. Sampling timelines varied: at WRC-URRAS, sample collection took place from January 2022 to April 2025; at WRC-CARDER-APAP, samples were collected in 2 phases (December 2022 and January 2024) and at ‘Bioparque Los Ocarros’, samples were taken in 2018. Free-ranging individuals were sampled during several biological expeditions conducted between 1998 and 2022. Birds were captured using mist nets, and blood samples were collected from the brachial vein. Sampling localities encompassed several major biogeographic units of Colombia. In the Orinoquia region, sites were located within the Orinoco and Amazon helobiomes. In the Andean region, sampling sites included high, middle and low-altitude Andean orobiomes, as well as the Upper Magdalena alternating hygric or subxerophytic zonobiome. Detailed information on sampling localities, including specific biomes, elevation, geographic coordinates and ecological characteristics, is provided in Supplementary Table 3.
Microscopic analysis
After blood collection, 3 blood smears were prepared, fixed in absolute methanol for 5 min and then stained with 4% Giemsa for 45 min. The blood cells were examined using an Olympus BX43 light microscope equipped with an Olympus DP27 digital camera. Screening was performed at 400× magnification for 20 min, followed by a more detailed examination at 100× oil-immersion magnification (1000× total magnification) for an additional 20 min per slide, according to González et al. (Reference González, Matta, Ellis, Miller, Ricklefs and Gutiérrez2014) and Valkiūnas et al. (Reference Valkiūnas, Ilgūnas, Bukauskaitė, Chagas, Bernotienė, Himmel, Harl, Weissenböck and Iezhova2019). Digital images were captured and processed using cellSens Standard 1.13 software (Olympus, Tokyo, Japan). Parasite identification was based on the morphometric features and taxonomic keys described in Valkiūnas (Reference Valkiūnas2004) and Valkiūnas and Iezhova (Reference Valkiūnas and Iezhova2018, Reference Valkiūnas and Iezhova2022). Morphometric parameters were measured from at least 100 images of the parasite stages using ImageJ software (Schneider et al., Reference Schneider, Rasband and Eliceiri2012). Parasitaemia was calculated as the number of infected cells per 10 000 erythrocytes.
Literature survey
A targeted literature survey was conducted to evaluate published records of haemosporidian infections in birds of prey. Searches were performed in Google Scholar and Web of Science using the following query: ‘Haemosporida OR Haemoproteus OR Leucocytozoon OR Plasmodium AND Strigiformes OR Accipitriformes OR Falconiformes’. In addition, the MalAvi database (Bensch et al., Reference Bensch, Hellgren and Pérez-Tris2009; February 2026 update) was searched for reported cytochrome b gene (cytb) lineages associated with these host taxa. For each record, data were systematically extracted regarding host species, parasite genus, geographic origin and the availability of corresponding molecular sequences in GenBank (Supplementary Table 1).
DNA extraction and partial cytb gene amplification
All microscopy-positive samples were subjected to partial amplification of cytb gene using standard haemosporidian nested polymerase chain reaction (PCR) protocols, including those described by Hellgren et al. (Reference Hellgren, Waldenström and Bensch2004), Pérez-Rodríguez et al. (Reference Pérez-Rodríguez, de la Puente, Onrubia and Pérez-Tris2013) and Pacheco et al. (Reference Pacheco, Escalante, Garner, Bradley and Aguilar2011, Reference Pacheco, Matta, Valkiūnas, Parker, Mello, Stanley, Lentino, Garcia-Amado, Cranfield, Kosakovsky Pond and Escalante2018). PCR screening was performed on approximately 80% of the total samples, ensuring the molecular dataset included at least 1 representative from each host species and sampling locality whenever samples were available. During initial analyses, some recovered sequences yielded ambiguous or unexpected lineage assignments, which could have led to unreliable parasite–host associations. Therefore, the molecular dataset was carefully curated and validated before including the results in the present study. Different amplification outcomes were observed among these protocols; some samples yielded amplicons across multiple protocols, whereas others amplified exclusively with a single primer set. Conversely, several microscopy-positive samples (N = 4) failed to produce any visible amplification products despite multiple testing attempts using all tested protocols. The resulting partial cytb gene sequences were submitted to GenBank under accession numbers PZ559886–PZ559901.
Haemosporidian mitochondrial genome amplification
To amplify the haemosporidian mitochondrial genomes, 6 positive owl samples, including Asio clamator (n = 1), Asio stygius (n = 1) and Megascops choliba (n = 4), were selected based on their infection with the new Haemoproteus species described here. Genomic DNA was extracted from whole blood using the DNeasy Blood & Tissue Kit (Qiagen, GmbH, Hilden, Germany) according to the manufacturer’s protocol. Then, the near-complete mitochondrial genome was amplified using a nested PCR protocol with TaKaRa LA Taq Polymerase (TaKaRa Mirus Bio Inc.) following the manufacturer’s directions (Pacheco et al., Reference Pacheco, Matta, Valkiūnas, Parker, Mello, Stanley, Lentino, Garcia-Amado, Cranfield, Kosakovsky Pond and Escalante2018). Briefly, primary PCRs were performed in 50 µL reactions using 2 µL of total DNA with the forward outer primer AE170 (5′-GAGGATTCTCTCCACACTTCAATTCGTACTTC-3′) and the reverse outer primer AE171 (5′-CAGGAAAATWATAGACCGAACCTTGG ACTC-3′). For the nested PCR, 1 µL of the primary product was used with the forward inner primer AE176 (5′-TTTCATCCTTAAATCTCGTAAC-3′) and the reverse inner primer AE136 (5′-GACCGAACCTTGGACTCTT-3′). In all runs, negative (distilled water) and positive (Plasmodium (Haemamoeba) lutzi) controls were included. Cycling conditions for both PCRs consisted of an initial denaturation at 94 °C for 1 min, followed by 30 cycles of 30 s at 94 °C and 7 min at 67 °C, with a final extension of 10 min at 72 °C. PCR products (50 µL) were excised from 1% agarose gels (bands of ∼6 kb), purified using the QIAquick Gel Extraction Kit (Qiagen) and cloned into the pGEM-T Easy Vector system (Promega, Madison, WI, USA) following the manufacturer’s directions. Both strands of 3 clones per sample were sequenced by Genewiz (Azenta Life Sciences, Burlington, MA, USA). The resulting mtDNA sequences were submitted to GenBank under accession numbers PZ575010–PZ575015.
Phylogenetic analysis and haplotype networks
Three separate alignments were generated using MAFFT v.7 (Katoh and Standley, Reference Katoh and Standley2013), followed by manual editing. The first alignment comprised 82 haemosporidian partial cytb gene sequences (467 bp excluding gaps) from the genera Haemoproteus and Leucocytozoon (Akiba) caulleryi (outgroup) available from GenBank and the MalAvi database (Bensch et al., Reference Bensch, Hellgren and Pérez-Tris2009). The second alignment focused on Leucocytozoon and comprised 56 partial cytb sequences (450 bp excluding gaps). This alignment included the morphospecies from MalAvi and specific lineages isolated from raptors, following the subclade divisions proposed by Harl et al. (Reference Harl, Himmel, Valkiūnas, Ilgūnas, Nedorost, Matt, Kübber-Heiss, Alic, Konicek and Weissenböck2022, Reference Harl, Himmel, Pacheco and Weissenböck2026). For this analysis, sequences from the Haemoproteus nisi group, L. (Akiba) caulleryi and Haemoproteus (Haemoproteus) spp. were selected as outgroups. The third alignment utilized near-complete parasite mtDNA genomes. This dataset included 146 sequences (5082 bp excluding gaps) representing 5 genera (Plasmodium, Hepatocystis, Haemocystidium, Haemoproteus and Leucocytozoon), including the new sequences generated in this study. In this phylogeny, Leucocytozoon (Leucocytozoon) spp. were used as the outgroup.
Next, phylogenetic hypotheses were estimated using the Bayesian method executed in MrBayes v3.2.7 with the default priors (Ronquist and Huelsenbeck, Reference Ronquist and Huelsenbeck2003). Based on the best-fit substitution model determined by ModelFinder in IQ-TREE (Nguyen et al., Reference Nguyen, Schmidt, von Haeseler and Minh2015; Kalyaanamoorthy et al., Reference Kalyaanamoorthy, Minh, Wong, von Haeseler and Jermiin2017), a general time-reversible model with gamma-distributed substitution rates and a proportion of invariant sites (GTR + Γ + I) was employed. Posterior probabilities (PPs) were estimated using 2 independent runs of 6 × 106 Markov Chain Monte Carlo steps, sampling every 500 generations. Convergence was confirmed when the potential scale reduction factors were between 1.00 and 1.02, and the average standard deviation of the PP was below 0.01 (Ronquist and Huelsenbeck, Reference Ronquist and Huelsenbeck2003). A 25% ‘burn-in’ of the sample was discarded from each run. Phylogenetic trees were visualized using FigTree v1.4.4 (http://tree.bio.ed.ac.uk/software/figtree/). The resulting trees display avian parasite species names, GenBank accession numbers, lineage names and PPs for each node.
To analyse the genetic relationships among the new Haemoproteus species using a haplotype network, a fourth alignment was generated in MAFFT v.7 (Katoh and Standley, Reference Katoh and Standley2013). This dataset included 126 cytb lineage sequences from owls available in the MalAvi database that shared a recent common ancestor with the Haemoproteus syrnii lineage hSTAL2, which was identified as the sister group to the new species in the phylogenetic analyses of the cytb gene and mtDNA genomes. Using this alignment, we estimated a median-joining network in PopArt 1.7 based on single-nucleotide variation (Leigh and Bryant, Reference Leigh and Bryant2015). In the analysis, transversions were weighted equally to transitions, and the epsilon parameter was set to zero. We applied 1 round of star contraction to collapse star-like structures in the network into single nodes. Additionally, the average evolutionary distance over sequence pairs of Haemoproteus spp. found in owls were estimated using the Tamura–Nei substitution model implemented in MEGA v7.0.14 (Kumar et al., Reference Kumar, Stecher and Tamura2016).
Statistical analysis of parasite morphometry
To evaluate morphometric differentiation among the new species of Haemoproteus sp. described in this study, a multivariate analysis was performed using morphometric variables obtained from macrogametocytes and microgametocytes. Variables included gametocyte area, infected erythrocyte area and nucleus displacement ratio (NDR). Prior to analysis, data were standardized to remove scale effects among variables. Differences between these parasite species were assessed using a permutational multivariate analysis of variance (PERMANOVA) based on Euclidean distances with 999 permutations. Statistical analyses were performed using R for Windows, version 4.5.1 (R Core Team, 2025), through RStudio version 2026.05.0+218
Ethics statement
The methodology used in this study was approved by the Bioethics Committee of the Faculty of Veterinary Medicine and Animal Science, Universidad Nacional de Colombia (Act No. 07 of 2023). In addition, the animal study protocol was approved by the Ethics Committee of the Faculty of Transversal Basic Sciences, UNITROPICO University, through Evaluation Act No. 001 dated 16 November 2021.
Results
Microscopic analysis
Twenty-five of the 202 examined individuals (12.4%), representing birds of 3 raptor orders, were infected with haemosporidian parasites (Table 1). Of these, 21 individuals originated from WRCs, while 4 were free-ranging birds from the Amazon–Orinoco peinobiome (N = 2) and the Orinoco–Amazon helobiome (N = 2). Overall, 23 individuals (11.3%) were positive for Haemoproteus spp. and 3 (1.4%) for Leucocytozoon spp. Leucocytozoon parasites were detected in 2 migratory raptors: Buteo platypterus (Broad-winged Hawk) and Pandion haliaetus (Osprey) sampled at the WRC-URRAS in Bogotá. However, Leucocytozoon sp. gametocytes were microscopically observed only in B. platypterus (Figure 2).
Haemosporidian parasites found in Accipitriformes raptors. (A, B, D and E) Haemoproteus parasites. (C and F–J) Leucocytozoon parasites. (A and B) Haemoproteus sp. gametocyte infecting Ictinia mississippiensis. (A) Macrogametocyte. (B) Microgametocyte. (D and E) Haemoproteus sp. gametocytes infecting Buteo platypterus. (D) Macrogametocyte. (E) Microgametocyte. (C and F–J) Gametocytes of L. toddi infecting B. platypterus. (C, F, G and J) Macrogametocyte. (H and I) Microgametocyte. Giemsa-stained thin blood films. Scale bar = 10 μm.
Within the family Strigidae, Haemoproteus infections were detected in A. clamator (Striped Owl; 29.4%, 5/17), A. stygius (Stygian Owl; 8.3%, 2/24) and M. choliba (Tropical Screech Owl; 23.2%, 10/43). Among Falconidae, Daptrius chimachima (Yellow-headed Caracara) showed an infection frequency of 14.2% (1/7) for Haemoproteus sp. Within Accipitridae, B. platypterus exhibited an infection frequency of 25% (2/8) for Haemoproteus sp. and 37.5% (3/8) for Leucocytozoon sp. Notably, 1 individual showed a triple coinfection involving Haemoproteus, Leucocytozoon and microfilariae parasites. The single examined Buteo swainsoni (Swainson’s Hawk) was infected with Haemoproteus sp., while only 1 out of 2 Ictinia mississippiensis sampled here (Mississippi Kite) was infected with Haemoproteus sp. Lastly, in Rupornis magnirostris (Roadside Hawk), the frequency of Haemoproteus infection was 4.5% (1/22). All Haemoproteus infections in free-ranging individuals were detected in the Orinoquia region, including both the peinobiome and the Orinoco–Amazon helobiome.
Phylogenetic analysis and haplotype networks
The phylogenetic reconstruction based on the cytb gene fragments (Figure 3) placed all Haemoproteus sequences obtained in this study within an independent and strongly supported clade (PP: 0.99). This clade comprises all lineages of H. syrnii and other lineages reported from various owl species. However, the relationships among these lineages remained unresolved due to the lack of informative sites in the partial cytb gene. Interestingly, all lineages reported for non-passerine birds formed a monophyletic group, except for those belonging to the Haemoproteus nisi group. This pattern was also confirmed when phylogenetic relationships were estimated from mtDNA genomes (Figure 4), though, in this case, the relationships were resolved with higher support (PP > 0.82). Specifically, the Haemoproteus lineages identified in this study shared a common ancestor with H. syrnii hSTAL2 (reported in Strix aluco from Europe), which corresponds to the redescription of the species by Karadjian et al. (Reference Karadjian, Puech, Duval, Chavatte, Snounou and Landau2013). They also shared a common ancestor with the hSTVAR01 lineage, the H. syrnii lineage reported in Bubo virginianus and Strix varia from the USA (Pacheco et al., Reference Pacheco, Miller, Harl, Ferreira, Cepeda, Valkiūnas, Beckerman, Oswald, Mateus-Pinilla and Escalante2025). It is noteworthy that H. syrnii lineages OTSCO05 (from Otus scops) and CULKIB01 (from Strix aluco) were not part of this monophyletic group in either analysis (Figures 3 and 4).
Bayesian phylogenetic hypothesis of haemosporidian parasites infecting owls sampled in Colombia. The phylogenetic tree was computed based on 82 partial parasite cytb gene sequences (467 bp excluding gaps) from the genera Haemoproteus and Leucocytozoon (Akiba) caulleryi. The values above branches are posterior probabilities. Species found in this study are shown in green, and new species are indicated with I, II and III. GenBank accession numbers and lineage names, as deposited in the MalAvi database, are provided in parentheses for the sequences used in the analysis.
Figure 3 Long description
The phylogenetic tree is vertically oriented, depicting the evolutionary relationships of haemosporidian parasites infecting owls. The tree is based on partial gene sequences from the genera Haemoproteus, Leucocytozoon and L. Akiba caulleryi. Branches are labeled with posterior probabilities, indicating the statistical support for each clade. Species found in the study are highlighted in green, with new species marked as I, II and III. GenBank accession numbers and lineage names are provided in parentheses. The tree shows a strongly supported clade with a posterior probability of 0.99, comprising Haemoproteus syrnii lineages and other owl species lineages. The relationships among these lineages are unresolved due to limited informative sites. Non-passerine bird lineages form a monophyletic group, except for the Haemoproteus nisi group. The tree includes illustrations of owls and labels for Strigiformes and non-passerine birds. The tree structure includes multiple branching points, with branches splitting and merging, representing speciation events. The posterior probabilities above branches range from 0.36 to 1.0, indicating varying levels of support for the evolutionary relationships depicted.
Bayesian phylogenetic hypothesis of haemosporidian parasites infecting owls sampled in Colombia. The phylogenetic tree was computed based on 146 near-complete parasite mtDNA genomes (5082 bp excluding gaps) representing 5 genera (Plasmodium, Hepatocystis, Haemocystidium, Haemoproteus and Leucocytozoon). The values above branches are posterior probabilities. Species found in this study are shown in green, and new species are indicated with I, II and III. GenBank accession numbers and lineage names, as deposited in the MalAvi database, are provided in parentheses for the sequences used in the analysis. In this phylogeny, Leucocytozoon (Leucocytozoon) spp. were used as the outgroup.
Figure 4 Long description
Phylogenetic tree depicting the evolutionary relationships of haemosporidian parasites infecting owls sampled in Colombia. The tree is computed based on 146 near-complete parasite mitochondrial DNA genomes, representing five genera. The tree is oriented vertically, with branches extending from top to bottom. Posterior probabilities are shown above the branches. Species found in the study are highlighted in green and new species are indicated with Roman numerals I, II and III. GenBank accession numbers and lineage names are provided in parentheses. The tree includes Plasmodium spp. from various classes, including Reptilia, Mammalia and Aves, as well as Haemocystidium spp. and Haemoproteus spp. The Haemoproteus syrnii lineages are shown with specific identifiers, such as PV849544, STAL2 and others. The tree also includes Haemoproteus noctuae, Haemoproteus majoris and Haemoproteus belopolskyi, among others. Leucocytozoon spp. are used as the outgroup. The diagram visually represents the monophyletic groupings and evolutionary relationships among the lineages, with specific clades and branches labeled accordingly.
Phylogenetic analyses (Figures 3 and 4) and the average evolutionary distance (Table 2) between sequence pairs of Haemoproteus spp. found in our owl samples suggest that these lineages may represent 3 new species. This finding was further supported by the haplotype network results (Figure 5). Furthermore, based on the estimated genetic distances and haplotype networks, none of the lineages reported from Brazil as H. syrnii (e.g., hATHCUN01, hMEGCHO03 and hASICLA02; Barino et al., Reference Barino, Rossi, de Oliveira, Reis Junior, D’Agosto and Dias2021; hASICLA01 is also listed in the MalAvi database) shared a recent common ancestor with the H. syrnii hSTAL2 or hSTVAR01. Interestingly, out of the 6 Haemoproteus lineages found in the owls sampled in this study, 2 lineages from A. clamator (N = 4) and A. stygius (N = 1) were identical to the lineage ASICLA02 previously reported for A. clamator in Brazil (Barino et al., Reference Barino, Rossi, de Oliveira, Reis Junior, D’Agosto and Dias2021). On the other hand, 3 lineages identified here in M. choliba were closely related to the lineage MEGCHO03, also reported in M. choliba from Brazil (Figures 3–5).
Minimum spanning network for Haemoproteus syrnii group and closely related parasites based on partial cytb gene lineages (N = 126, 464 bp excluding gaps) identified from owls. Circles are drawn proportional to the frequency at which lineages were detected. Lines between circles have the number of mutations that separate 2 lineages. Haemoproteus species found in this study are labelled in green. Lineages reported as H. syrnii in the literature are also indicated.
Figure 5 Long description
The minimum spanning network illustrates genetic relationships among Haemoproteus lineages found in owls. Circles represent different lineages, with sizes proportional to the frequency of detection. Lines between circles indicate the number of mutations separating two lineages. The network is divided into two main sections. The upper section includes various lineages such as MEGCHO03, CAMRUB01 and PSDIS01, with labels like UR259 and CA1776. Some lineages are marked in green, indicating species found in the study, while others are marked in yellow, representing lineages reported in Brazil. The lower section focuses on Haemoproteus syrnii, with lineages like STAL2, STVAR01 and STAL4. The circles in this section are connected by lines, showing the mutation distances between them. The diagram includes images of owls and parasite illustrations, emphasizing the host species. The network visually represents the genetic distances and evolutionary relationships among the lineages, highlighting the connections and differences between them.
Regarding Leucocytozoon parasites, 3 new lineages were detected in our samples: 2 in B. platypterus and 1 in P. haliaetus (Figure 6). Due to the limited availability of blood samples, we were only able to amplify the partial cytb gene for phylogenetic analysis. One of the lineages found in B. platypterus appeared closely related to the lineages within the Leucocytozoon toddi subclade L301, as proposed by Harl et al. (Reference Harl, Himmel, Valkiūnas, Ilgūnas, Nedorost, Matt, Kübber-Heiss, Alic, Konicek and Weissenböck2022, Reference Harl, Himmel, Pacheco and Weissenböck2026). The second lineage detected in B. platypterus clustered within the L. toddi subclade L201 (Harl et al., Reference Harl, Himmel, Valkiūnas, Ilgūnas, Nedorost, Matt, Kübber-Heiss, Alic, Konicek and Weissenböck2022, Reference Harl, Himmel, Pacheco and Weissenböck2026). In contrast, the lineage identified in P. haliaetus shares a common ancestor with a recently reported lineage from the same host in Europe (PV839588, isolate AH2344; Harl et al., Reference Harl, Himmel, Pacheco and Weissenböck2026) and falls outside of the Leucocytozoon toddi group (Figure 6). Their genetic distance was 0.023 ± 0.007.
Bayesian phylogenetic hypothesis of Leucocytozoon species infecting raptors sampled in Colombia. The phylogenetic tree was computed based on 56 partial parasite cytb gene sequences (450 bp excluding gaps). The values above branches are posterior probabilities. Species found in this study are shown in orange. Sequences from the Haemoproteus nisi group, Leucocytozoon (Akiba) caulleryi and Haemoproteus (Haemoproteus) spp. were selected as outgroups. GenBank accession numbers and lineage names, as deposited in the MalAvi database, are provided in parentheses for the sequences used in the analysis.
Figure 6 Long description
The phylogenetic tree is vertically oriented, depicting the evolutionary relationships among Leucocytozoon species infecting raptors sampled in Colombia. The tree is based on 56 partial parasite gene sequences, with branch values representing posterior probabilities. Species found in the study are highlighted in orange. The tree begins with the Leucocytozoon toddi species group, branching into various Leucocytozoon species, each labeled with specific GenBank accession numbers and lineage names. The tree includes images of raptors and parasite cells. The branches show divergence points, with some species clustering closely together, indicating shared ancestry. The tree also includes outgroup sequences from Leucocytozoon Akiba caulleryi and Haemoproteus species, which are used for comparison. The tree visually represents genetic relationships and evolutionary distances among the species, with some branches showing new lineages detected in Buteo platypterus and Pandion haliaetus.
Description of novel owl Haemoproteus species
Based on combined morphological and molecular evidence, 3 novel species are proposed and described in this study.
Haemoproteus (Parahaemoproteus) tatilalaorum sp. nov. (Figure 7; Table 3, Supplementary Figure 1)
Taxonomic summary
Type host: Striped Owl Asio clamator (Vieillot, 1807) (Strigiformes, Strigidae), code in biological collection accession number GERPH V321.
Estimates of pairwise genetic distance using partial cytochrome b gene (cytb) sequences among Haemoproteus syrnii and closely related parasites found in owls from Colombia
Table 2 Long description
Pairwise genetic distance values with standard error are listed for 16 parasite lineages, with lineages numbered 1 through 16 across both rows and columns. Values are shown for each lineage pair, with blank diagonal cells where a lineage would be compared to itself. The largest value shown is 0.052 for Haemoproteus sp. ATHCUN01, lineage 5 compared with H. syrni KJ451480 OTSCO05, lineage 3. Several pairs have a genetic distance of 0.000, including lineage 6 Haemoproteus sp. MEGCHO03 with lineage 9 H. homohuizhyanus CA1574 MEGCHO03 and with lineage 10 H. homohuizhyanus CA1776 MEGCHO03, lineage 7 Haemoproteus sp. ASICLA02 with lineage 15 H. tatilalaorum V323 ASICLA02 and with lineage 16 H. tatilalaorum UR207 ASICLA02, lineage 13 H. huizhyanus UR293 PSDIS01 with lineage 14 H. huizhyanus UR52 PSDIS01, and lineage 15 with lineage 16. Among H. syrnii entries, lineage 1 H. syrnii KF279523 STAL2 and lineage 2 H. syrnii PV948504 STVAR01 differ by 0.003 in the upper triangle and 0.004 in the lower triangle. Distances between lineage 1 and lineages 11 through 16 are each 0.005 in the upper triangle. Distances between lineage 6 and lineage 8 H. homohuizhyanus UR259 MEGCHO01 are 0.002, and between lineage 8 and lineage 9 and between lineage 8 and lineage 10 are each 0.002.
Type locality: Villavicencio-Meta, Acacías-Meta, Colombia.
Occurrence: 24.4% (5/17) in A. clamator; 8.3 % (2/24) in Stygian Owl A. stygius.
Site of infection: Mature erythrocytes; no other data.
Distribution: Villavicencio, Meta and Bogotá, Cundinamarca, Colombia.
Type specimens: The hapantotype (accession number GERPH V321, Asio clamator, parasitemia intensity 1.26 %, lineage ASICLA02, Villavicencio, Meta Los Ocarros Biopark in Villavicencio, Meta Department, collected by Pablo Felipe Cruz, 11 January 2018, Coinfection with microfilariae occurs (parasitaemia intensity 0.5 %).
Parahapantotypes: Asio clamator, accession number GERPH V322, parasitemia intensity 1.05%, GERPH V323 parasitaemia intensity 0.77% and GERPH V324 parasitaemia intensity 1.09%. Co-infection with Trypanosoma sp. (parasitaemia 0.07%). All preparations were sampled at the Los Ocarros Biopark in Villavicencio, Meta Department, by Pablo Felipe Cruz on 11 January 2018. GERPH UR115 Asio clamator, parasitaemia intensity 0.7%, sampled at the Wildlife Rescue and Rehabilitation Unit (URRAS) in Bogotá, Cundinamarca Department, by Brayan Gamboa on 13 June 2023. A. stygius, accession number GERPH UR207, parasitaemia intensity 0.08%. GERPH UR342, parasitaemia intensity 0.98%, sampled at the Wildlife Rescue and Rehabilitation Unit (URRAS) in Bogotá, Cundinamarca Department, by Brayan Gamboa on 27 September 2024. All type preparations were deposited in the Biological Collection GERPH at Universidad Nacional de Colombia, Sede Bogotá.
Reference DNA sequences: partial cytb gene: PZ559887–PZ559890; near-complete mtDNA genome: PZ575010 (GERPH V321) and PZ575011 (GERPH UR207).
Etymology: The species name tatilalaorum is dedicated to the children of the NEM author, Tomas Roque Matta and Laura Roque Matta, whose nicknames are Tati and Lala. From an early age, they have accompanied the NEM author on numerous field expeditions and continue to contribute by helping to prepare conference materials and proofreading English manuscripts. This dedication also acknowledges the many moments in academic life when they were unable to spend quality time together.
Description
Young gametocytes (Figure 7A–C) were observed in mature red blood cells; they assume lateral and subpolar positions within infected erythrocytes; the outline is mainly even or slightly ameboid, with small thread-like outgrowths that never exceed the parasite’s total length (Figure 7B,C). In early forms, a large vacuole-like space was visible in 10% of gametocytes (Figure 7A–C). Growing gametocytes usually do not contact the erythrocyte envelope or nuclei; their outline is even, though it might be slightly irregular in advanced forms. Pigment granules are tiny, few and brown.
Gametocytes of Haemoproteus tatilalaourum sp. Nov. (lineage ASICLA02) from the blood of its type host, the Striped owl (Asio clamator): (A–C) young gametocytes, (D–M) macrogametocytes and (N–T) microgametocytes. Immature gametocytes can contain vacuoles (A) and may have small thread-like extensions that never exceed the total length of the parasite (B), or have slim elongations reminiscent of fingers (C). Gametocytes are halteridial. Immature stages can touch nuclei and envelope of the erythrocytes (D–H) but also can be some forms that do not touch the erythrocytes nuclei (N). Microgametocytes have a similar conformation to macrogametocytes (N–T). Open arrowhead: vacuole. Arrowhead: space between the erythrocyte nucleus and the parasite cell membrane. Mature gametocytes grow up to the Poles of infected erythrocytes; the parasites contain centrally or subcentrally located nuclei (arrow). Heavy arrow: hemozoin granules (J). All images are from the hapantotype. Giemsa-stained thin blood films. Scale bar = 10 μm.
Macrogametocytes (Figure 7D–M, Table 3, Supplementary Figure 1) growing gametocytes are halteridial; they adhere to the erythrocyte envelope and frequently touch the host erythrocyte nuclei (Figure 7D,I,K). Some growing gametocytes can exhibit a characteristic J-shaped form (Figure 7D). The parasite nucleus is generally positioned sub-terminally as they extend along them. In some instances, the growing parasites did not touch the erythrocyte envelope and/or nuclei (Figure 7G,N). Advanced gametocytes are also halteridial (Figure 7K–M), with prominent parasite nuclei located primarily in central or slightly subcentral positions (Figure 7G–I, K). Fully grown gametocytes were broadly halteridial (Figure 7K–M), slightly enclosing the host nuclei with their ends and filling the erythrocytes up to the poles (Figure 7K–M). Pigment granules ranged from small (<0.5 µm) to medium (0.5–1 µm) in size; they were predominantly round, brown and scattered throughout the cytoplasm with no tendency to cluster. The outline of the gametocytes varied from even (Figure 7E,I–M) to slightly irregular or ameboid (Figure 7D,F,G). J-shaped forms were observed in 10% of gametocytes (Figure 7D,H), occasionally possessing cytoplasmic prolongations that embrace 1 end of the erythrocyte nucleus. The cytoplasm appeared homogeneous or finely granular in appearance; volutin granules were not observed; a few small vacuoles were occasionally visible (Figure 7K–M). Fully grown gametocytes slightly enlarged the infected erythrocytes and displaced their nuclei laterally (Table 3).
Microgametocytes (Figure 7N–T, Table 3) were similar to macrogametocytes in their general configuration. Gametocyte nuclei were markedly diffuse, centrally located and irregular in form. The cytoplasm appeared pale-stained and pinkish. Vacuoles occurred occasionally (Figure 7N,P). Volutin granules were absent.
Remarks: Two species of Haemoproteus have been described in owls (Strigiformes): Haemoproteus noctuae (Celli and Sanfelice, 1891) and H. syrnii (Mayer, 1910) (Valkiūnas and Iezhova, Reference Valkiūnas and Iezhova2022; Pacheco et al., Reference Pacheco, Miller, Harl, Ferreira, Cepeda, Valkiūnas, Beckerman, Oswald, Mateus-Pinilla and Escalante2025). Haemoproteus tatilalaourum sp. nov. can be readily distinguished from both of these. Fully grown gametocytes of H. noctuae are strictly circumnuclear, and prominent ‘pea-like’ volutin granules overfill gametocytes of H. syrnii. Neither of these characters is a feature of H. tatilalaourum sp. nov.
Haemoproteus (Parahaemoproteus) huizhyanus sp. nov. (Figure 8; Table 3)
Taxonomic summary
Type host: Tropical Screech-Owl M. choliba (Strigiformes, Strigidae) Vieillot, 1817.
Type locality: Fómeque, Cundinamarca, Colombia.
Occurrence: 6,9% (3/43 individuals) in M. choliba.
Site of infection: Mature erythrocytes; no other data.
Distribution: Fómeque, Cundinamarca, Colombia, Bogotá, Cundinamarca, Colombia, Pereira, Risaralda, Colombia.
Type specimens: Hapantotype (1 preparation with accession numbers GERPH: UR052-I, intensity of parasitaemia is 1.38%, lineage PSDIS01, blooded by Brayan Gamboa from Unidad de Rescate y Rehabilitación de Animales Silvestres URRAS 4.64 N, −74.08 W, Bogotá D.C., Universidad Nacional de Colombia on 20 February 2022) was deposited in the biological collection of the Grupo de Estudio Relación Parásito Hospedero (GERPH) at the Department of Biology, Universidad Nacional de Colombia, Sede Bogotá. Three hapantotypes are available under the codes GERPHUR052-II and UR052-III.
Parahapantotypes: Blood smears of 1 individuals of M. choliba (accession number UR293), with parasitaemia intensities of 0.78%, respectively, collected by Brayan Gamboa from the Wildlife Rescue and Rehabilitation Unit (URRAS; 4.64° N, −74.08° W, Bogotá D.C., Universidad Nacional de Colombia) on 24 April 2024. Blood smears of 1 individual of M. choliba (accession number HPP08, parasitaemia intensity 1.45%) were collected by Brayan Gamboa from Hogar de Paso CARDER-APAP (4.77° N, −75.61° W, Pereira, Colombia, December 2022). All samples are deposited in the biological collection of the GERPH at the Department of Biology, Universidad Nacional de Colombia, Bogotá.
Reference DNA sequences: Partial cytb gene: PZ559891–PZ559893; near-complete mtDNA genome: PZ575012 (GERPH UR52) and PZ575013 (GERPH UR293).
Morphological measurements of the haemosporidian parasites observed in this study. Measurements are given in micrometres (µm) and correspond to the dimensions of parasite stages observed in blood smears
Table 3 Long description
Measurements in micrometres are listed for six taxa: Haemoproteus huizhyanus lineage PSDIS01, Haemoproteus homohuizhyanus lineage CAMRUB01, Haemoproteus tatilalaourum lineage ASICLA02, Haemoproteus multivacuolatus from Rupornis magnirostris, Haemoproteus trarotraro from Daptrius chimachima, and Haemoproteus sp. from Buteo platypterus. For each taxon, rows report uninfected erythrocyte length, width, and area, plus uninfected erythrocyte nucleus length, width, and area, each with sample size. Matching rows report infected erythrocyte length, width, and area, and infected erythrocyte nucleus length, width, and area, also with sample size. Macrogametocyte measurements include length, width, area, NDR, number of pigment granules, and gametocyte nucleus length, width, and area. Microgametocyte measurements include infected erythrocyte length, width, and area, infected erythrocyte nucleus length, width, and area, gametocyte length, width, and area, NDR, number of pigment granules, and gametocyte nucleus length, width, and area. Uninfected erythrocyte mean lengths range from 12.6 in Haemoproteus huizhyanus to 14.58 in Haemoproteus multivacuolatus, while infected erythrocyte mean lengths range from 13.32 in Haemoproteus trarotraro to 15.44 in Haemoproteus multivacuolatus. Uninfected erythrocyte mean areas range from 64.21 in Haemoproteus trarotraro to 100.19 in Haemoproteus multivacuolatus, while infected erythrocyte mean areas range from 72.19 in Haemoproteus trarotraro to 115.46 in Haemoproteus multivacuolatus. Macrogametocyte mean lengths range from 9.90 in Haemoproteus homohuizhyanus to 17.94 in Haemoproteus sp., and microgametocyte mean lengths range from 9.80 in Haemoproteus homohuizhyanus to 21.55 in Haemoproteus sp.
Etymology: The word huizhy is from the Muisca indigenous language; huizhy means ‘to liberate’, ‘to save’, ‘to give life’, or ‘to be alive’. The name honours the hard work of WRC worldwide, which rescue, treat, rehabilitate and release wild animals. Approximately 80% of the samples analysed in this study originated from WRCs.
Description
Young gametocytes (Figure 8A–C,K; Table 3) developed in mature erythrocytes. The earliest forms were variable in both form and outline (Figure 8). They were found anywhere in the erythrocytes, including lateral, subpolar and polar positions, and typically lay free in the cytoplasm, without contacting the erythrocyte nuclei or envelope (Figure 8B). A prominent, dense, reddish spot of unclear origin was observed in most young gametocytes, a characteristic feature of this parasite species' development (Figure 8A,B); this spot of volutin-like matter disintegrates and disappears in advanced gametocytes. Approximately 31% of young gametocytes assumed a comma-like shape (Figure 8K). Nuclear material was dispersed throughout the early gametocytes (Figure 8A,B). As parasites grow, gametocytes extend along the erythrocyte nuclei and often lie free in the cytoplasm, not touching either the erythrocyte envelope or the nuclei. The cytoplasm appeared granular and vacuolated. In both early and advanced growing gametocytes, a medium-sized (1.035 ± 0.08 (0.29–0.89)) vacuole was frequently observed in the cytoplasm (Figure 8A,C,L). Pigment granules are small, appearing either scattered in the cytoplasm or grouped, including grouping towards a single pole in some advanced gametocytes (Figure 8D).
Gametocytes of Haemoproteus huizhyanus sp. nov. (lineage PSDIS01) from the blood of its type host, the Tropical screech owl (Megascops choliba): (A–C) young gametocytes, (D–J) macrogametocytes and (L–T) microgametocytes. In growing and late gametocytes, a medium-sized vacuole can be observed in the cytoplasm (A and C). The distribution of hemozoin granules can follow 2 patterns. In the first pattern, medium-sized granules are observed bilaterally relative to the parasite nucleus (E, F, I and J). In the second pattern, granules tend to cluster at a single pole of the parasite (D, O, R, S and T). Cytoplasmic extensions resembling ‘fingers’ are evident at a single pole of the parasite in 84% of late microgametocytes (N, O, P and Q). Open arrowhead: reddish spot. Arrow: nucleus of the parasite. Arrow outline: vacuole. Arrowhead: finger-like cytoplasmic projections of the gametocyte. Heavy arrow: clumps of hemozoin granules at 1 pole of the parasite. Asterisk flower: clumps of hemozoin granules on either side of the nucleus; star: volutin granules. All images are from the hapantotype. Giemsa-stained thin blood films. Scale bar = 10 μm.
Macrogametocytes (Figure 8D–J) possessed a markedly heterogeneous granular cytoplasm with numerous tiny vacuoles scattered throughout. Volutin granules were not seen. Gametocytes grow along the nuclei of erythrocytes, which are displaced laterally (Figure 8F–J). Growing gametocytes are halteridial; they frequently did not adhere to either the host nucleus or the erythrocyte envelope (Figure 8E). Forms with markedly irregular or wavy outlines were common among growing gametocytes (Figure 8F), although gametocytes with an entire outline were also observed (Figure 8D,H). Growing gametocytes adhered to the erythrocyte envelope in 61% of cases and remained non-adherent in the remaining 39%. As the parasite grows, gametocytes adhere to the erythrocyte envelope and subsequently to the host nucleus (Figure 8F–J). Fully grown gametocytes were halteridial, closely appressed to both the erythrocyte envelope and nucleus, and enclosed the erythrocyte nucleus with their ends to fill the erythrocyte poles; however, they never completely encircled the erythrocyte nuclei (Figure 8I,J). Gametocytes markedly enclosing the erythrocyte nuclei without complete encirclement (Figure 8J) represented 2.1% of macrogametocytes. Growing gametocytes often exhibited irregular or amoeboid outlines with serrated edges (Figure 8E,F), while fully developed gametocytes predominantly possessed smooth outlines (Figure 8H,I). The parasite nucleus was compact; it occupied a predominantly central or near-central position. Pigment granules were mainly medium-sized (0.5–1 µm), generally roundish or oval. The distribution of pigment granules in the cytoplasm follows 2 main patterns: first, discrete medium-sized granules located bilaterally to the parasite nucleus (65.2% of cases; Figure 8E–G, I, J, black flower); and second, pigment granules tend to cluster at a single pole of gametocytes (34.2% of gametocytes; Figure 8D). Infected erythrocytes were slightly enlarged, and their nuclei were displaced laterally (Table 3).
Microgametocytes (Figure 8K–T, Table 3). General configuration was similar to that of the macrogametocytes, exhibiting the sexual dimorphism typical of haemosporidians, which is characterized by a pale-staining cytoplasm and diffuse nuclei. The cytoplasm appeared heterogeneous, containing numerous tiny vacuoles. Volutin granules were absent. The distribution of pigment granules followed 2 patterns: first, granules were positioned bilaterally adjacent to the parasite nucleus (59.1% of microgametocytes; Figure 8M); and second, the granules tended to cluster at a single pole of the parasite (40.9% of microgametocytes; Figure 8O,R–T). The hemozoin granule clusters occasionally take the form of a ‘bow tie’ (Figure 8S,T). Ameboid cytoplasm outgrowths resembling ‘fingers’ were evident at a single end (84%) or both gametocyte ends (16%) of fully grown microgametocytes (Figure 8N–Q, arrowhead). These ameboid outgrowths were observed more frequently in microgametocytes than in macrogametocytes.
Remarks: Two species of Haemoproteus have been described in Strigiformes: H. syrnii (Mayer, 1910) and H. noctuae (Celli and Sanfelice, 1891). Gametocytes of H. syrnii possess conspicuous, large ‘pea-like’ volutin granules, which are particularly visible in microgametocytes (Valkiūnas and Iezhova, Reference Valkiūnas and Iezhova2022). Fully grown gametocytes of H. noctuae are strictly circumnuclear; they completely encircle the host cell nucleus of infected erythrocytes (Valkiūnas and Iezhova, Reference Valkiūnas and Iezhova2022). None of these characters is features of H. huizhyanus sp. nov. While the gametocytes of H. huizhy sp. nov. share some similarities with H. tatilalaorum sp. nov., they can be readily distinguished. The cytoplasm of both macrogametocytes and microgametocytes of H. huizhyanus sp. nov. is markedly vacuolated (Figure 8) and heterogeneous in appearance; furthermore, the pigment granules are predominantly located in clumps (Figure 8). Neither of these features is characteristic of H. tatilalaorum sp. nov. (Figure 7). Additionally, fully grown gametocytes of H. huizhyanus sp. nov. possess fewer pigment granules compared to H. tatilalaorum sp. nov. (Table 3). Additionally, while pigment granules are generally scattered throughout the cytoplasm in H. tatilalaorum sp. nov., they tend to aggregate in H. huizhyanus sp. nov. (compare Figure 7 with Figure 8).
Haemoproteus (Parahaemoproteus) homohuizhyanus sp. nov. (Figure 9, Table 3)
Taxonomic summary
Type host: Tropical Screech-Owl M. choliba (Strigiformes, Strigidae) Vieillot, 1817.
Type locality: Viotá, Cundinamarca, Colombia.
Occurrence: 16.2% (7/43 individuals) in M. choliba.
Site of infection: Mature erythrocytes; no other data.
Distribution: Viotá, Cundinamarca, Colombia, Bogotá, Cundinamarca, Colombia, Simijaca, Cundinamarca, Colombia, Macheta, Cundinamarca, Colombia, Nocaima, Cundinamarca, Colombia, Yopal, Casanare, Colombia, Paz de Ariporo, Casanare, Colombia.
Type specimens: Hapantotype (1 preparation with accession numbers GERPH: UR175-I, intensity of parasitaemia is 4.14%, lineages CAMRUB01, blooded by Brayan Gamboa from Unidad de Rescate y Rehabilitación de Animales Silvestres URRAS 4.64 N, −74.08 W, Bogotá D.C., Universidad Nacional de Colombia on 17 July 2023) was deposited in the biological collection of the GERPH at the Department of Biology, Universidad Nacional de Colombia, Sede Bogotá. Hapantotype preparations are available under the codes UR175-II and UR175-III.
Parahapantotypes: blood smears of 5 individuals of M. choliba (accession numbers GERPHUR135, UR149, UR175, UR254 and UR259), with parasitaemia intensities of 1.4%, 22.03%, 4.14%, 0.3 % and 1.3%, respectively, collected by Brayan Gamboa from the Wildlife Rescue and Rehabilitation Unit (URRAS; 4.64° N, −74.08° W, Bogotá D.C., Universidad Nacional de Colombia) between January 2022 and April 2025. Blood smears of 1 individual of M. choliba (accession number CA1574, parasitaemia intensity 3.65 %) were collected by Karly Ojeda from Yopal, Casanare Department, Finca Buenas Tardes (5.1299167, −72.1261111, 350 m a.s.l., 01 March 2024). Blood smears of 1 individual of M. choliba (accession number CA1776, parasitaemia intensity 1.14 %) were collected by Astrid Cárdenas from Paz de Ariporo, Casanare Department, Finca La Carpeta (5.9558, −70.93236, 133 m a.s.l., 17 June 2024). All samples were deposited in the biological collection of the GERPH at the Department of Biology, Universidad Nacional de Colombia, Bogotá.
Reference DNA sequences: partial cytb gene: PZ559894–PZ559898; near-complete mtDNA genome: PZ575014 (GERPH UR175; lineage CAMRUB01) and PZ575015 (GERPH UR254; lineage MEGCHO02).
Etymology: The species name homohuizhyanus refers to its morphological similarity to Haemoproteus huizhyanus sp. nov.
Description
Young gametocytes (Figure 9A–D) develop in mature erythrocytes. Growing gametocytes (Figure 9A,B) are variable in both form and outline. Their outline is predominantly irregular (Figure 9B–F). Young gametocytes were found anywhere in the erythrocytes, including lateral, subpolar and polar positions, and typically lay free in the cytoplasm, without contacting the erythrocyte nuclei and envelope (Figure 9B–D). Nuclei are prominent and located centrally in gametocytes. As parasites grow, gametocytes usually do not touch the nuclei of erythrocytes (Figure 9B,C). In both early and advanced growing gametocytes, a medium-sized (1.039 ± 0.05, on average 0.25–0.75 µm) vacuole was frequently observed in the cytoplasm (15%) (Figure 9C). Pigment granules were clumped at a pole in 65% of gametocytes (Figure 9A–D), or could be scattered in the cytoplasm (Figure 9D).
Gametocytes of Haemoproteus homohuizhyanus sp. nov. (lineage CAMRUB01) from the blood of its type host, the Tropical screech owl (Megascops choliba): (A–D) young gametocytes, (E–J) macrogametocytes and (K–P) microgametocytes. In growing and late gametocytes, a medium-sized vacuole can be observed in the cytoplasm (C). Pigment granules tend to cluster at a single spot located on 1 end of gametocytes (D–J). Open arrowhead: vacuole. Arrow: nucleus of the parasite. Arrowhead: space between the gametocyte and the erythrocyte nucleus. Heavy arrow: clumps of hemozoin granules at 1 pole of the parasite. All images are from the hapantotype. Giemsa-stained thin blood films. Scale bar = 10 μm.
Macrogametocytes (Figure 9E–J; Table 3) are predominantly irregular in outline (Figure 9E,F,H), but forms with even outline were also present (Figure 9G,I); the cytoplasm is markedly heterogeneous in appearance, possesses numerous tiny scattered vacuoles. Volutin granules are absent. Growing and fully grown gametocytes are halteridial; they frequently do not adhere to either the host nuclei or the erythrocyte envelope (Figure 9E–I) and displace the nuclei laterally (Figure 9H,I). Growing gametocytes non-adhered to the nuclei of erythrocytes represent 70% of cases. The parasite nucleus is compact and occupies predominantly central or sub-central positions. Pigment granules were mainly of medium size (0.6–1 µm), generally roundish or oval. The distribution of pigment granules in the cytoplasm follows 2 main patterns: first, the most common pattern is the pigment granules clustering at a single spot located on 1 pole of gametocytes (70.8%) (Figure 9E–J); in the remaining gametocytes, we observed a discrete medium-sized granule located bilaterally to the parasite nucleus (Figure 9I). Infected erythrocytes were slightly enlarged, and their nuclei were slightly displaced laterally (Table 3).
Microgametocytes (Figure 9K–P, Table 3). General configuration was like that of the macrogametocytes, exhibiting the typical haemosporidians sexual dimorphism of the pale-staining cytoplasm and diffuse nuclei. The cytoplasm appeared homogeneous with numerous tiny vacuoles. Ameboid outgrowths resembling ‘fingers’ were absent. The distribution of pigment granules followed 2 patterns: first, granules were positioned bilaterally adjacent to the parasite nucleus (12.8% of microgametocytes) (Figure 9L); and second, the granules clustered at a single pole of the parasite (87.2% of microgametocytes; Figure 9K–P).
Morphometric analysis: A PERMANOVA based on morphometric variables revealed significant differences between H. huizhyanus sp. nov. (UR052) and H. homohuizhyanus sp. nov. (UR135 and UR175) (Pseudo-F = 15.29, P = 0.001). Haemoproteus huizhyanus exhibited the largest overall morphometric measurements, including larger gametocytes and infected erythrocytes, whereas H. homohuizhyanus showed consistently smaller dimensions.
Remarks: Gametocytes of H. homohuizhyanus sp. nov. share some similarities in size and halteridial shape with H. huizhyanus sp. nov. and H. tatilalaorum sp. nov., but they can be readily distinguished. In H. homohuizhyanus sp. nov., the pigment granules are predominantly arranged in a single clump, which is situated at 1 end of the gametocytes (Figure 9). This feature is absent in H. huizhyanus sp. nov. and H. tatilalaorum sp. nov., nor is it characteristic of any other haemoproreids parasitizing owls (Valkiūnas and Iezhova, Reference Valkiūnas and Iezhova2022). Additionally, fully grown gametocytes of H. homohuizhyanus sp. nov. possess fewer pigment granules than those of H. tatilalaorum sp. nov. (Table 3). Haemoproteus homohuizhyanus sp. nov. differs from H. huizhyanus sp. nov. by a combination of morphological and molecular characters (Figures 8 and 9). Early gametocytes of H. homohuizhyanus lack the prominent dense reddish spot observed in H. huizhyanus (Figure 8A,B). In addition, fully grown gametocytes of H. homohuizhyanus often do not contact the erythrocyte nuclei, and microgametocytes of this parasite lack the finger-like cytoplasmic projections, which are present at the ends of the microgametocytes in H. huizhyanus. A further distinguishing feature is the distribution pattern of hemozoin granules. In H. homohuizhyanus, approximately 80% of gametocytes exhibit pattern 1: characterized by hemozoin granules aggregated as a compact clump toward a single pole of the gametocyte. In contrast, H. huizhyanus predominantly exhibits pattern 2: in which 62.15% of gametocytes show hemozoin granules clustered bilaterally around the parasite nucleus. Phylogenetic analyses based on the cytb gene revealed a genetic divergence ranging from 0.8% to 1.5% (0.008–0.015 substitutions/site) between both taxa (Table 2).
Discussion
Across the Neotropics, the diversity, host specificity and taxonomic identity of Haemosporida parasites in birds of prey remain poorly investigated (Barino et al., Reference Barino, Rossi, de Oliveira, Reis Junior, D’Agosto and Dias2021; Lotta-Arévalo et al., Reference Lotta-Arévalo, González, Gamboa-Suárez, Pacheco, Escalante, Moreno, Rodríguez-Fandíño, Cuervo and Matta2023; Vieira et al., Reference Vieira, Pacheco, Escalante and Braga2026). This study represents the first integrative taxonomic investigation to characterize Haemoproteus and Leucocytozoon parasites infecting diurnal and nocturnal raptors in Colombia. By leveraging samples from WRCs and applying an integrative framework that incorporates morphology and molecular evidence, 3 novel Haemoproteus species are described infecting owls, expanding the known diversity of haemosporidians infecting Neotropical raptors. Until recently, only 2 species or morphospecies groups within the subgenus Haemoproteus (Parahaemoproteus), H. syrnii and H. noctuae, have been formally recognized as infecting owls. Previous molecular studies in Brazil suggested cryptic diversity within H. syrnii (Barino et al., Reference Barino, Rossi, de Oliveira, Reis Junior, D’Agosto and Dias2021); however, the morphology of these gametocytes does not coincide with the original description of H. syrnii due to the lack of volutin granules, as noted by Ilgūnas et al. (Reference Ilgūnas, Himmel, Harl, Dagys, Valkiūnas and Weissenböck2022). Furthermore, the genetic divergence among H. syrnii-like lineages infecting strigiform hosts (Pacheco et al., Reference Pacheco, Miller, Harl, Ferreira, Cepeda, Valkiūnas, Beckerman, Oswald, Mateus-Pinilla and Escalante2025) strongly suggested the presence of distinct and undescribed species circulating in owls from Brazil.
Consistent with these observations, integrative analysis revealed that H. (P.) homohuizhyanus sp. nov. infecting M. choliba exhibited morphological and genetic similarities to the lineage MEGCHO03 reported from the same host in Brazil (Barino et al., Reference Barino, Rossi, de Oliveira, Reis Junior, D’Agosto and Dias2021) (Figures 3–5). However, it showed sufficient genetic and morphological divergence from H. syrnii to warrant its recognition as a distinct species. Similarly, the newly described species H. tatilalaourum corresponds to the lineage ASICLA02 previously reported in A. clamator in Brazil (Barino et al., Reference Barino, Rossi, de Oliveira, Reis Junior, D’Agosto and Dias2021) (Figures 3–5) and was also detected in A. stygius in our study (Supplementary Figure 1). These data suggest high host specificity and a widespread distribution of these novel Haemoproteus species across South America. While the genus Asio is cosmopolitan and shows significant parasite diversity across European and Asian populations, including a high prevalence (69.3%) recently reported in A. otus (Stanković et al., Reference Stanković, Rajković, Raković, Skorić, Chakarov, Pibaque and Marzal2026), the lineages we identified in American Asio species are not shared with those reported for A. otus. Notably, although A. clamator has been documented as a host for Plasmodium (Novyella) paranucleophilum in Brazil (Tostes et al., Reference Tostes, Dias, Martinele, Senra, D’Agosto and Massard2017), this parasite was not detected in our current sampling. Interestingly, the CAMRUB01 lineage of H. homohuizhyanus sp. nov. was previously detected in the Red-necked Woodpecker (Campephilus rubricollis) in Brazil (Fecchio et al., Reference Fecchio, Pinheiro, Felix, Faria, Pinho, Lacorte, Braga, Farias, Aleixo, Tkach, Collins, Bell and Weckstein2018b). In addition, the PSDI501 lineage of H. huizhyanus sp. nov. was previously reported from M. choliba in Colombia (Lotta-Arévalo et al., Reference Lotta-Arévalo, González, Gamboa-Suárez, Pacheco, Escalante, Moreno, Rodríguez-Fandíño, Cuervo and Matta2023).
Haemosporidian infection was evaluated at 3 WRCs in Colombia, located across different altitudinal gradients. Although there was a disparity in the number of individuals sampled across centres, Haemoproteus species were the most frequently detected parasites in both the highlands and the lowlands. Among the raptors sampled in Colombia, strigiform birds exhibited the highest occurrence of Haemoproteus infection. This pattern is consistent with previous findings indicating that nocturnal birds of prey are more frequently infected with haemosporidian parasites than diurnal species (Gao et al., Reference Gao, Zhou, Yang, Dong, Huang and Deng2021; Stanković et al., Reference Stanković, Rajković, Raković, Skorić, Chakarov, Pibaque and Marzal2026). This trend has been documented across the Neotropics, where owls consistently show elevated infection rates and parasite diversity (Barino et al., Reference Barino, Rossi, de Oliveira, Reis Junior, D’Agosto and Dias2021; Morel et al., Reference Morel, Webster, Prusch, Anicet, Marsicano, Trainini, Stocker, Giani, Bandarra, da Rocha, Zitelli, Umeno, Souza, Dall’Agnol and Reck2021). Among the species examined in this study, M. choliba showed the highest infection frequency, corroborating reports from Brazil (64.2%), where this host exhibits a high occurrence of infections and supports multiple parasite lineages (Barino et al., Reference Barino, Rossi, de Oliveira, Reis Junior, D’Agosto and Dias2021). The ecological traits of this host species, including its broad geographic distribution, habitat plasticity and potential exposure to diverse vector communities, are likely to contribute to its elevated susceptibility and its role as an important host in Haemosporida parasite assemblages.
Haemoproteus huizhyanus sp. nov. and H. homohuizhyanus sp. nov. appear to have a wide geographic and altitudinal distribution, since both parasite species were recorded in M. choliba from both highland and lowland regions (Supplementary Table 2). Taken together, these results suggest that certain haemosporidian lineages can persist across broad altitudinal gradients while maintaining host specificity (Supplementary Table 2). Occurrence, in a host, however, does not inform about the place of infection. Our findings indicate that although H. huizhyanus sp. nov., H. homohuizhyanus sp. nov. and H. tatilalaorum sp. nov. are widely distributed in the Neotropics, they do not appear to be generalist parasites. Instead, their occurrence seems closely associated with the broad geographic distribution of their owl hosts, including A. stygius, A. clamator and M. choliba, all of which are widely distributed across Colombia and other regions of South America (eBird, 2025a, b). It is also important to highlight the remarkable genetic diversity detected among Haemoproteus parasites infecting M. choliba. Although it is still premature to speculate on the mechanisms underlying this diversity, it is noteworthy that both H. huizhyanus and H. homohuizhyanus circulate sympatrically within a single locality in Bogotá, Colombia. This pattern may reflect complex ecological or vector-related transmission dynamics. Future studies focusing on the parasites’ life cycles, vector identity, host specificity and transmission ecology will be essential to better elucidate the processes driving this unexpected diversity and coexistence. Finally, it is worth noting that all phylogenetic analyses placed the lineages of all 3 novel species within the Parahaemoproteus clade. This result suggests that these parasites could be transmitted by Culicoides biting midges, which should be prioritized as the primary candidate definitive hosts in future vector competence research.
Regarding the other Haemoproteus species detected in other raptor species sampled in this study, the Haemoproteus morphotype observed in D. chimachima resembles the recently described species Haemoproteus trarotraro infecting D. chimachima and Caracara plancus in Brazil (Vieira et al., Reference Vieira, Pacheco, Escalante and Braga2026) (Supplementary Figure 2). This newly described species shares several morphological characteristics with the morphotype detected in the present study (Figure 2). Similarly, the Haemoproteus morphotype identified in R. magnirostris exhibits morphological similarities to Haemoproteus multivacuolatus, originally described from Buteo in Europe (Harl et al., Reference Harl, Fauchois, Puech, Gey, Ariey, Izac, Weissenböck, Chakarov, Iezhova, Valkiūnas and Duval2024). Unfortunately, cytb gene sequences could not be obtained for these infections, preventing molecular confirmation of their exact taxonomic identity (Supplementary Figure 3). The Haemoproteus morphotypes detected in Buteo swainsoni and Ictinia mississippiensis could not be assigned to a species because parasitaemia was extremely low, preventing a detailed morphological characterization. In addition, repeated attempts to amplify the parasite cytb gene were unsuccessful.
From a wildlife rehabilitation and conservation perspective, preventing the translocation of parasites beyond their natural distribution ranges is essential for maintaining local host–parasite associations. The widespread occurrence of H. huizhyanus sp. nov. and H. homohuizhyanus sp. nov. across multiple regions suggests that M. choliba individuals may be candidates for release across a broader geographic range without substantially disrupting existing local parasite dynamics. Conversely, the haemosporidian parasites infecting D. chimachima, A. clamator and R. magnirostris appear to be distributed largely within lowland habitats. Therefore, rehabilitation programs should prioritize releasing these species within their native ecological zones to minimize the risk of parasite introduction and preserve natural parasite biogeographic patterns. Rehabilitation protocols should incorporate diagnostic screening for haemosporidians in raptors prior to release or relocation to prevent the accidental introduction of lineages into naive environments where they might infect susceptible resident and/or endemic host populations. Because raptors are among the most threatened avian groups globally, active collaboration between WRCs and specialized diagnostic laboratories is a vital conservation strategy for monitoring wildlife health and preventing the spread of disease.
On the other hand, Leucocytozoon infections were infrequent and restricted to migratory species. Buteo platypterus was the only migratory raptor infected with a Leucocytozoon toddi-like parasite, with an occurrence of 37.5% (3/8). These infections were likely acquired in the Nearctic region, where Buteo species are commonly infected (Bell et al., Reference Bell, Driscoll, Achatz, Martens and Vaughan2025). In Colombia, the transmission of Leucocytozoon parasites in resident birds has been typically restricted to high-altitude environments (>2400 m a.s.l.) (Lotta et al., Reference Lotta, Valkiūnas, Pacheco, Escalante, Hernández and Matta2019). The L. toddi-species group in diurnal raptors comprises multiple clades with genus-level host specificity, including Buteo and Accipiter, and encompasses several morphotypes, such as L. toddi, L. mathisi and L. buteonis (Harl et al., Reference Harl, Himmel, Valkiūnas, Ilgūnas, Nedorost, Matt, Kübber-Heiss, Alic, Konicek and Weissenböck2022). Notably, 2 distinct Leucocytozoon lineages were detected in B. platypterus in this study, each belonging to different clades and associated with distinct morphotypes (Figures 2 and 6). Lineage BUTPLA01 (sample ID: UR142) clustered near L. buteonis and exhibited a divergent morphology, whereas lineage BUTPLA02 (sample ID: UR234) corresponded to the typical morphology of L. toddi (Figures 2 and 6). This finding further supports the hypothesis that L. toddi represents a complex of cryptic species. These results are consistent with recent mitogenomic studies indicating that the L. toddi lineage forms a distinct evolutionary group within the genus, potentially representing a subgenus specialized in Accipitriformes raptors (Harl et al., Reference Harl, Himmel, Pacheco and Weissenböck2026). In the Neotropics, infection by L. toddi has previously been reported in D. chimachima in Chile; however, this record lacks both molecular characterization and detailed morphological analysis (Forrester et al., Reference Forrester, Foster and Morrison2001). Additionally, a new Leucocytozoon lineage was detected in P. haliaetus (PANHAL02); however, no gametocytes infecting erythrocytes were observed in the blood smears examined. Nevertheless, this lineage is closely related to the previous records of Leucocytozoon infection in this host species (AH2344: PANHAL01) reported in Austria and deposited in the MalAvi Database. Future studies should investigate the morphology, diversity and phylogenetic relationships of this new clade of Leucocytozoon spp. infecting P. haliaetus and other Neotropical raptors to clarify their phylogenetic relationships (Figure 6).
Importantly, most haemosporidian infections were detected immediately upon host admission to the WRCs, confirming that they were acquired in wild populations rather than during captivity. Given that approximately 80% of the samples originated from these institutions, our findings underscore their value as platforms for epidemiological surveillance. The discovery of new parasite species in these birds highlights the critical role WRCs play in monitoring raptor health. By providing access to avian species that are traditionally difficult to sample in the wild, these centres facilitate a passive surveillance system for haemosporidians and other microorganisms as well as a broader understanding of parasitic diversity, distribution and host-parasite dynamics.
Conclusions
In the Neotropics, including Colombia, studies on haemosporidians in raptors remain scarce and fragmented. While molecular investigations have revealed substantial cryptic diversity, our results highlight the necessity of combining morphological characterization with genetic data within an integrative taxonomic framework. Observed patterns of infection suggest structured host–parasite associations rather than generalized transmission. Notably, several lineages exhibited broad altitudinal distribution while maintaining marked host specificity. Strigiformes showed the highest occurrence of infection, reinforcing global trends of elevated haemosporidian prevalence in nocturnal raptors. In contrast, Leucocytozoon infections were rare and restricted to migratory hosts such as B. platypterus and P. haliaetus, suggesting that further sampling is needed to clarify the ecological factors influencing Leucocytozoon occurrence and transmission dynamics in Neotropical raptors. Collectively, these findings demonstrate that WRC in Colombia represent critical sentinel systems for uncovering hidden parasite diversity in Neotropical wildlife, emphasizing the essential role of integrative taxonomy in resolving the true diversity, host specificity and geographic distribution of avian haemosporidians.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0031182026103023.
Acknowledgements
The authors would like to express their gratitude to Ingrid Lotta, Alejandra Quiroz, Juanita Cardenas, David Pinto and all members of Grupo de Estudio Relacion Parasito Hospedero (GERPH). Special thanks to Juan Guerrero, Margarita Jara, Andres Hermann, Maria Jaramillo, Karen Aya, Claudia Brieva, Carlos Moreno, Daniela Ospina, Sebastian Bustos, Laura Jiménez, Carol Garcia, Paula Ramirez and Angie Tellez, for all their efforts and dedication at URRAS. We are deeply grateful to Carolina Hincapié, Katherin Suárez, Sara Cardona, Daniela Garcia, Carlos Bernal, Gina, Danilbe Loaiza, Lina Rozo, Mónica Londoño, Juan Pablo Arias, Lizeth Holguin, Maria Pinillla and Felipe Arias for their invaluable support and dedication at the CARDER–APAP Wildlife Rehabilitation Centre. The authors also thank Roberto Júnio Pedroso Dias (Laboratório de Protozoologia, Protozoology Lab, UFJF, Brazil) for providing us with access to the H. syrnii sequences reported for Brazilian owls (Barino et al. 2021). We would also like to thank the environmental authorities Secretaria Distrital de Ambiente de Bogota (SDA), Corporacion Autonoma Regional de Risaralda (CARDER), Corporación Autónoma Regional del Guavio (CORPOGUAVIO) and Corporacion Autonoma Regional de Cundinamarca (CAR).
Author contributions
Conceptualization: B.A.G.S., N.E.M. and M.A.P.; Methodology: B.A.G.S., N.E.M., M.A.P., M.F.M.F. and G.V.; Software: B.A.G.S., G.A.F.R. and M.A.P.; Validation: B.A.G.S., N.E.M., M.A.P. and G.V.; Formal analysis: B.A.G.S., N.E.M., M.A.P., M.F.M.F. and G.A.F.R.; Investigation: B.A.G.S., N.E.M., M.A.P., M.F.M.F. and G.A.F.R.; Resources: B.A.G.S., N.E.M., M.A.P., A.A.E., L.J.C.H. and P.F.C.O.; Data curation: B.A.G.S., N.E.M., M.A.P., M.F.M.F., G.A.F.R., G.V. and M.J.V.; Writing – original draft: B.A.G.S., N.E.M., M.A.P., G.V. and G.A.F.R.; Writing – review and editing: B.A.G.S., N.E.M., M.A.P., G.A.F.R., M.F.M.F., G.V. and A.A.E.; Visualization: B.A.G.S., M.A.P., G.A.F.R., M.F.M.F. and M.J.V.; Supervision: N.E.M. and L.J.C.H.; Project administration: N.E.M. and LJCH; Funding acquisition: N.E.M., L.J.C.H., M.A.P. and A.A.E.
Financial support
“Researchfor the advancement of knowledge and development of regional capacities in the surveillance of potentially pathogenic agents of risk for the population of Casanare” with BPIN 2021000100338 was financed by the Sistema General de regalías between 2022 and 2025. L.J.C.H. was funded by Universidad Internacional del Trópico Americano, and N.E.M. was funded by Universidad Nacional de Colombia, Sede Bogotá. M.A.P. and A.A.E. are supported by the US National Science Foundation (Grant No. NSF-DEB 2146653).
Competing interests
The authors declare that there are no conflicts of interest.
Ethical standards
The methodology used in this study was approved by the Bioethics Committee of the Faculty of Veterinary Medicine and Animal Science, Universidad Nacional de Colombia (Act No. 07 of 2023). In addition, the animal study protocol was approved by the Ethics Committee of the Faculty of Transversal Basic Sciences, UNITROPICO University, through Evaluation Act No. 001 dated 16 November 2021.
Open access
