1. Introduction
Parasitic filarial nematodes (Nematoda: Filarioidea) of the genus
Dirofilaria, represent a severe threat to veterinary and public health, particularly affecting dogs and cats, and in rare occasions, humans as well
[1,2,3,4]. Besides canines and felines, these cosmopolitan parasitic worms
[5] might also infect other carnivores as well, such as wolves (
Canis lupus), red foxes (
Vulpes vulpes) and golden jackals (
Canis aureus)
[6,7,8,9].
Dirofilaria immitis (Leidy 1856)
, an important mosquito-borne nematode, known as the dog heartworm, causes cardiopulmonary dirofilariasis, invading the heart and large blood vessels
[10]. The damage caused by this parasite to arteries and right cardiac chambers of infected hosts might have a fatal outcome, especially if not treated or if treatment is delayed. Another dirofilarial worm is
D. repens Railliet et Henry, 1911
, which causes subcutaneous infections in dogs and other carnivores
[11]. Both
Dirofilaria species can accidentally be transmitted to humans
[12,13,14,15]. Although humans are dead-end hosts to these filarial nematodes (as they cannot proliferate in the human body), they can still cause health issues depending on the invaded body part. The infection may manifest superficially with the adult nematodes appearing subcutaneously and subconjunctival
[16]. However, the major concerns in human populations are the benign pulmonary nodules caused by
D. immitis in human lungs, frequently mistaken for malignant lung tumors
[13,17,18,19].
Nowadays, cases of dirofilarial infections have been detected worldwide
[1]. The process of parasite transmission to hosts is very complex. Successful transmission requires the presence of competent mosquito vectors. Once a mosquito female intakes blood infected with microfilariae, in the following two or more weeks nematodes are going to molt to the infective third larval stage. The infective stage moves from the tubules via the hemocoel to the lumen of the labial sheath in the mosquito’s mouthparts
[20]. The duration of this period, measured in the body of several mosquito species (
Aedes vexans, Ae. triseratus, Ae. trivittatus, and
Anopheles quadrimaculatus), lasted 14 days and is directly temperature-dependent
[21,22,23,24]. Subsequent blood meal intake of an infected female mosquito will result in the parasite transmission to the bitten host
[10,13].
Regardless of the fact, around 70 mosquito species classified to the
Anopheles,
Aedes,
Culex,
Culiseta, and
Coquillettidia genera have been considered as potential vectors of animal and human dirofilariasis, where only a few species have been proven as competent vectors
[10,25].
Serbia is considered as an endemic country of
Dirofilaria sp. in animals and humans for many years
[10]. Several studies have been conducted targeting
Dirofilaria in reservoirs (animals) and humans
[26,27,28,29,30,31,32]. Between 2006 and 2007 the reported prevalence for
D. immitis was 7.2% in the Vojvodina and 3.2% in Branicevo regions
[26,27]. In the region of Belgrade, a few years later, the prevalence was 22.01%, with 3.97% of dogs showing co-infections with
D.
repens [10].
Despite all the knowledge about the presence of
Dirofilaria sp. in Serbia, little is known about their vectors. So far, only one publication has focused on the vectors of
Dirofilaria [33]. Kurucz et al.
[33] showed that 8.3 % of tested mosquito pools were positive for
Dirofilaria. Positive mosquitoes belonged to five mosquito species:
Aedes vexans,
Ae. caspius, Ae. sticticus, Culex pipiens and
Coquillettidia richiardii. Mosquitoes were found positive for both
D. immitis and
D. repens at several localities throughout the entire mosquito breeding season.
Therefore, the aim of the present study is to contribute to the mapping of Dirofilaria hotspots in the Vojvodina Province and Mačva region, Serbia. Detecting positive mosquitoes can provide insights into the distribution of the parasite in Serbia. This information could help us better understand how the parasite spreads and adapts to temperate environmental conditions. Analyzing the blood meals of vectors could help create a list of animal species that may be at risk due to potential Dirofilaria infections.
2. Materials and Methods
2.1. Mosquito Sampling and Vector Identification
Mosquito sampling was conducted in Vojvodina Province, Serbia (65 locations), covering an area of 21,506 km
2. In addition, eight locations belonging to the Mačva region (612 km²) were included. Sampling was carried out at 73 locations in total (
Figure 1), during the summer season of mosquito activity in 2021, starting from May till October. Due to the low number of
Aedes albopictus collected in 2021 and the high significance in filarial transmission of this invasive species, we included the samples from 2022 to increase the likelihood of parasite detection. The geo-coordinates of locations are shown in
Appendix A. This study only included adult female mosquitoes. Females were collected using CO
2 baited (dry ice) adult traps (NS2 trap type). Traps were set up in the afternoon hours and operated overnight. Mosquito samples were then kept in dry ice until transferred to the laboratory within the Centre of Excellence – One Health at the Faculty of Agriculture, University of Novi Sad, Serbia. When the samples arrived in the laboratory, mosquitoes were morphologically identified to species level, using identification key Becker et al.
[34].
All collected females per location were categorized based on the presence of blood meal in their abdomen as non-blood fed and blood fed. Females were separated in pools of up to 100 individuals per species per tube. From each mosquito trap only one pool per species was taken. Samples were conserved dry in 2 ml tubes (Eppendorf, Hamburg, Germany) and stored in the freezer on -20 ⁰C until analyzed.
Due to the regularly high number (> 200 per trap) of non-blood fed mosquitoes in traps in the majority of locations, a selection of mosquito species (aimed for further analysis) from this category was based on vector competence to transmit Dirofilaria sp. Selected mosquito species were: Aedes vexans (Meigen 1830), Aedes caspius (Pallas 1771), Aedes albopictus Skuse 1894, and Anopheles maculipennis Complex Meigen 1818.
The number of blood fed females in traps was usually very low (< 5 per trap), therefore we analyzed all captured blood fed mosquito species for the presence of Dirofilaria sp. Because of this low number of blood fed specimens, we also included mosquitoes collected in 2022.
After screening non-blood fed and blood fed mosquitoes for the presence of parasites, we analyzed the blood meal source in blood fed females to identify the putative host species. The following selection for host detection included: a) females from the positive locations belonging to the same species as the positive ones, b) females from locations in the close vicinity to the positive locations. Additionally, non-blood fed females which belonged to the same species and same locations (refers to a and b from above) were also added to try to detect the host (it was assumed that some females might have already digested a blood meal and it was not visible in the abdomen).
2.2. DNA extraction
Extractions and the molecular analysis of all samples were conducted at the Institute of Research and Development, within the Mivegec research unit, Montpellier, France.
Extraction of DNA was carried out by using the DNeasy Blood and Tissue Kit (Qiagen, Hilden, Germany) according to the manufacturers’ instructions.
For parasite detection, non-blood fed mosquitoes were pooled in tubes by up to 20 individuals for DNA extraction. Therefore, pools with the number of mosquitoes higher than 20 had to be divided. While for blood fed females we put one mosquito per tube, in order to be further analyzed (if positive) for blood meal source detection.
Positive controls of Dirofilaria repens and Dirofilaria immitis were extracted from infected dogs’ blood and were provided by dr. Ettore Napoli (University of Messina, Department of Veterinary Sciences). DNA extraction of positive controls was also done using the Dneasy Blood and Tissue Kit.
2.3. Identification of Dirofilaria sp.
Screening of mosquito pools for the presence of
Dirofilaria sp. was conducted using a conventional PCR approach based on the amplification of the COI (Cox1) gene of parasites. The COI gene was targeted using the primer pair COIintF (5′-TGATTGGTGGTTTTGGTAA-3′) and COIintR (5′-ATAAGTACGAGTATCAATATC-3′) under the modified PCR conditions described in Casiraghi et al.
[35,36], Gabrielli et al.
[37] and Tasić-Otašević et al.
[38].
Polymerase chain reaction (PCR) was performed in 25 µl volumes of mix under the following final conditions: 16.05 µl of water, Tp 10x 2.5 µl (Eurogentec, Seraing, Belgium) including 50mM MgCl2 0.75 µl (Eurogentec), 10mM dNTP 0.5 µl (Eurogentec), primer COI-int-F (10pmol/µl = 10µM) 1.5 µl, primer COI-int-R (10pmol/µl = 10µM) 1.5 µl and TAQ Platinum (5U/µl) 0.2 µl (Invitrogen, Waltham, Massachusetts, USA). Two µL of sample DNA were added to 23 µl of Master mix.
The thermal profile used was 94 °C 10 min, and then 5 cycles of 94 °C, 30 sec, 52 °C 45 sec, 72 °C 1 min, afterward 30 cycles of 94 °C 30 sec, 58 °C 45 sec, 72 °C 1 min, and the final was 72 °C 7 min. These conditions provided PCR products of 650 bp.
PCR products were separated by TAE 0.5X and 1.3% agarose gel electrophoresis (Eurogentec) stained with gelred (Biotium, San Francisco, California, USA) and sized with 4.5 µl ladder (Generuler 100 bp, Thermo Scientific, Waltham, Massachusetts, USA). The quantity used for the preparation of gel was as follows: 50 ml of TAE 0.5X, 0.65 g of agarose and 10 µl of stain gelred. The product was then migrated for 35 min at 100 V.
Samples which produced non-specific bands were further processed by sequencing (Eurofins Genomics, Germany). Results of the Blast analysis showing only the highest percent identity (98-100%) were considered in this study.
The consensus sequences were made and cleaned in BioEdit. Sequence alignment was performed using the ClustalW method. Same was done for the five samples which were positive for D. immitis aimed to validate previously detected parasites.
2.4. Identification of Blood Meal Host
Molecular identification of blood meal source species was performed following the protocol by Boessenkool et al.
[39]. The primers used were 16Smam1 (CGGTTGGGGTGACCTCGGA) and 16Smam2 (GCTGTTATCCCTAGGGTAACT). PCR was performed in a final volume of 50 µl under the following conditions: water 36 µl, Tp 10X 5 µl (Eurogentec), Mgcl2 50mM 2 µl (Eurogentec), dNTP 10mM 0.2 µl (Eurogentec), primer 16Smam1 (10pmol/µl =10µM) 0.8 µl, primer 16Smam2 (10pmol/µl = 10µM) 0.8 µl and TAQ Platinum (5U/µl) 0.2 µl (Invitrogen). We added 45 µl of Master mix + 5 µL of DNA.
Thermal profile consisted of 55 cycles with the temperatures as follows: 94 °C 2 min, 94 °C 30 sec, 60 °C 30 sec, 72 °C 30 sec, 72 °C 10 min. These conditions provided PCR products of 150 bp.
PCR products were separated by TAE 0.5X and 2% agarose gel electrophoresis stained with gelred and sized with 4.5 µl ladder. The quantity used for the preparation of gel was: 50 ml of TAE 0.5X, 1 g of agarose and 10 µl of stain gelred. The product was then migrated for 35 min at 100 V. Amplicons were sent for sequencing to Eurofins.
Regarding the results of the Blast analysis, only those with the highest percent identity (98-100%) were included in this study.
3. Results
3.1. Presence of Dirofilaria immitis and Setaria tundra in Mosquitoes
The total number of analyzed mosquitoes was 2,902, of which 2,521 were non-blood fed, and 381 were blood fed mosquitoes. For non-blood fed, analyzed specimens belonged to An. maculipennis complex, Ae. vexans, Ae. caspius and Ae. albopictus. For blood fed mosquitoes, in addition to the species mentioned, we also analyzed Aedes sticticus (Meigen 1835), Culex pipiens Linnaeus 1758, Culiseta annulata (Schrank 1776) and Coquillettidia richiardii (Ficalbi 1889) specimens.
For the filarial worms screening we analyzed: 398
An. maculipennis complex (383 non-blood fed and 15 blood fed), 1,340
Ae. vexans specimens (1,253 non-blood fed and 87 blood fed), 316
Ae. caspius (305 non-blood fed and 11 bloodfed), 8
Ae. sticticus (all blood fed), 580
Ae. albopictus (all non-blood fed), 225
Cx. pipiens (all blood fed), 7
Cs. annulata (all blood fed) and 28
Cq. richiardii (all blood fed) specimens. Out of 2,902 screened mosquitoes, the genome of filaria was found in only five mosquito pools (in total six mosquitoes, one pool consisted of two mosquitoes) (
Figure 2). After sequencing and Blast analysis, three were found corresponding to
D. immitis and two to
Setaria tundra. All positive mosquitoes were collected in the Vojvodina Province. Mosquitoes from the Mačva region were not positive for the aimed parasites.
The sequencing and the Blast Analysis confirmed the presence of Dirofilaria immitis in three samples, all of which were detected in Cx. pipiens mosquitoes. Positive Cx. pipiens were collected in three different locations: Glogonj, Svetozar Miletić and Zrenjanin. Positive mosquitoes in Glogonj and Svetozar Miletić were collected in July 2021, while in Zrenjanin Cx. pipens was positive at the end of August 2021. D. immits was present only in blood fed Cx. pipiens.
The results also showed that two out of five positive samples were positive for Setaria tundra (Issaitshikoff & Rajewskaya, 1928), a species of nematode which has not been detected before on the territory of Serbia. In this study, S. tundra was detected in two mosquito species, Ae. caspius and Ae. vexans. Aedes caspius was collected in the location named Mali Iđoš, at the end of July 2021, while Ae. vexans was collected at the location Iđoš, during mid-August of 2021. This parasite was detected in non-blood fed mosquitoes.
All five locations with positive mosquitoes are shown in
Figure 3.
3.2. Blood Meal Host Detection
Out of five positive locations for parasites, blood fed females were collected only in four. Besides these four, additional four neighboring locations were included in the analyses. In total blood fed females from eight locations were analyzed.
Out of 30 selected females, 22 were blood fed and eight were non-blood fed females. We analyzed 19 Cx. pipiens (blood fed), seven Ae. vexans (three blood fed and four non-blood fed), and four Ae. caspius (non-blood fed).
In total, 16 mosquitoes resulted in successful host detection. One mosquito was non-blood fed, and the rest of them were blood fed. Identified hosts are presented in
Figure 3. The host was not identified in any of the analyzed
Ae. caspius females.
3.3. Phylogenetic analysis of Setaria tundra
Phylogenetic analysis of S. tundra nucleotide sequnces were analysed using BLAST NCBI and MEGA v. 11.0 software [
40] to align sequences and determine phylogenetic relationships. Maximum Likelihood with the Jones-Taylor-Thornton substitution model was used as the tree construction method. Additionaly, BLAST searches were performed in GenBank (
https://www.ncbi.nlm.nih.gov), and S. tundra matches showing a high genetic affinity were downloaded and incorporated into the alignment. Bootstrap analysis of 1000 randomly generated sample trees were performed to assess the stability of the inferred phylogenies. The selected outgroup was D. immitis.
All new nucleotide sequences in this study have been deposited in GenBank NCBI with the accession numbers: PP475177 (S. tundra isolated from Ae. caspius) and PP475174 (S. tundra isolated from Ae. vexans).
The approximate 650 bp fragment of the COX-1 gene was analyzed in two isolates.
S. tundra isolated from
Ae. caspius has shown similarity with
S. tundra originally isolated from
Cq. richiardii in Austria (MF695090), while
S. tundra isolated from
Ae. vexans has shown a similarity with
S. tundra isolated from
Ae. vexans in Hungary (KM452922) (
Figure 4).
4. Discussion
This study represents contribution to the distribution of Dirofilaria spp. and the first record of S. tundra in Serbia. Also, it provides valuable insight in the species of mosquito vectors and their host preference in the temperate climate.
The previous paper published by Kurucz et al. [
33] provided the first molecular evidence of
D. immitis and
D. repens nematodes from mosquito samples in Serbia. However, out of 73 locations, the present study only confirmed
D. immitis in 3 locations in Vojvodina Province. Considering that the previous study treated a high number of
Cx. pipiens, our study presented more information on other vectors such as
Ae. vexans,
Ae. caspius,
Ae. albopictus and
An. maculipennis complex. European studies have confirmed infections by
D. immitis in the following mosquito species:
Cx. pipiens in Spain [
41], Italy [
25], and Turkey [
42];
Cx. theileri in Madeira, Portugal [
43], and on the Canary Islands, Spain [
44];
Ae. vexans in Turkey [
42,
45] and
Ae. albopictus,
Ae. caspius,
An. maculipennis, and
Cq. richiardii in Italy [
25,
46,
47].
In our study, only
D. immitis was detected in analyzed mosquitoes collected at 73 locations. Although
D. repens was earlier detected by Kurucz et al. [
33], in this research, it was not found. All three positive samples in the present study belonged to Cx.
pipiens. These mosquitoes were collected in three different locations (Svetozar Miletić, Glogonj and Zrenjanin), not close to each other (Glogonj vs Zrenjanin 53 km, Svetozar Miletić vs Zrenjanin 135 km and Glogonj vs Svetozar Miletić 172 km). Two of these locations are villages and one is an urban settlement. The study of Kurucz et al. (2016) [
33] detected these parasitic worms in mosquitoes at six locations, and also their positive locations were very distant. Bearing in mind that
Cx. pipiens is a very bad flier, it is indicative that
D. immitis is a widely spread parasite in Vojvodina Province. One location selected by Kurucz et al. [
33] (Zrenjanin) overlaps with our results, demonstrating the persistent circulation of
D. immitis in this city (from 2014 till 2021).
Our analysis of blood meal sources from mosquitoes collected in Zrenjanin and Glogonj (both locations positive for
D. immitis) demonstrated that
Cx. pipiens took the blood (at least the last blood meal) from humans, two mosquitoes being from Zrenjanin and one from Glogonj. In Zrenjanin, other blood meals were identified from various animals including a dog, raven, wild boar, sparrow and pigeon (2 times). In the location close to Zrenjanin, it was demonstrated that
Cx. pipiens was feeding on a pigeon. These findings could also represent a contribution to the understanding of the West Nile virus circulation which is very frequently detected in these locations [
48]. Interestingly, blood meals of other collected mosquitoes (two
Cx. pipiens) were cats’ blood. Earlier studies demonstrated that cats could get infected with
Dirofilaria sp., but it does not cause severe disease in them. Cats are not considered as good hosts for
Dirofilaria because the infections are cleaned by their immune system before the nematodes can become adults [
2]. It is estimated that the prevalence of feline infections in Europe is between 5 and 20% of the total canine prevalence in the same region [
5].
The first systematic studies of dirofilariasis in dogs in Serbia were initiated at the beginning of the 21st century. The study was performed in Vojvodina Province showing endemic status for
D. repens and
D. immitis infection in dogs [
27,
28]. The climatic conditions in Serbia, coupled with the long activity periods of competent vectors such as
Cx. pipiens and
Ae. albopictus (Kavran et al., unpub. data), are considered suitable for the transmission of
D. immitis and
D. repens to humans and animals for at least half of the year (sometimes even more), depending on the air temperature [
49,
50]. Findings of Savić et al. [
17] showed a prevalence of 26.30% for
D. immitis infections in dogs, with 25.72% showing microfilariae. The prevalence of
D. repens larvae was 1.45%. An earlier study showed a prevalence of 22.9% for
D. immitis, while for
D. repens it was 39.34% [
28]. Several studies conducted in Serbia demonstrated an increasing trend of
D. immitis infections and a decreasing trend of
D. repens [
17,
27,
51,
52].
Setaria tundra is a new species on the list of parasites in Serbia. In this study, S. tundra was found in two locations (Iđoš and Mali Iđoš), which are almost 93 km apart from each other.
Setaria nematodes are classified to the Filarioidea superfamily, family Onchocercidae and are parasites of different ungulates. At least four species of the genus Setaria are present in Europe:
S. equina [
53],
S. cervi [
54],
S. labiatopapillosa [
55] and
S. tundra [
56].
Setaria tundra was first described in Russia in 1928 [
57] and up to now it has been reported in many European countries [
58]. The reports from the European countries are given chronologically: Russia 1928 [
57], Austria 1969 [
59], Finland 1970 [
60], Sweden 1973 [
61], Norway 1973 [
62], Bulgaria 1973 [
63], Switzerland 1974 [
64], Germany 1975 [
65], Italy 2003 [
66], France 2006 [
67], Denmark 2011 [
68], Poland 2010 [
69], Hungary 2013 [
70], Spain 2016 [
71], Croatia 2018 [
72], Slovakia 2022 [
73].
Olos et al. [
74] hypothesized that geographical expansion of
Setaria nematodes may be indirectly related to wet and warm summers. This is because intermediate hosts are found in abundance, along with the high density of possible definitive hosts as well as wild and domesticated ungulates. These authors stated that the recent focus on
S. tundra has been due to its spreading range to the southern regions of Europe. This species of nematode has expanded its geographical range by hundreds of kilometers and is known to be a major cause of mass mortality of wild and semi domesticated reindeer in Fennoscandia, Finland [
75,
76]. In northern Europe, the reindeer (
Rangifer tarandus) is the major definitive host, yet the moose can serve as an asymptomatic carrier [
56,
77,
78], while roe deer and red deer (
Cervus elaphus) serve as the definitive hosts in central and southern Europe [
74,
76,
79]. In the review of Olos et al. [
58] it was stated that domestic species such as sheep, goats, cattle, and horses are also potentially at risk [
80,
81,
82,
83]. Over the past decade, the populations of wild ruminants and wild boars have increased across Europe [
84,
85]. This expansion is accompanied by an apparent negative relationship between their abundance in the wild and their health status [
86]. Considering that wild animals often enter cattle pastures and spread parasites to livestock, it is of great importance to maintain surveillance and control wildlife diseases [
87].
This parasite can be transmitted by several species of mosquitoes, but particularly by those of the genus
Aedes [
88,
89,
90]. Microfilariae of this parasite have been reported in
Ae. vexans,
Ae. caspius,
Cx. pipiens, Culex torretium, Aedes annulipes, Ae. sticticus, A. rossicus and
Cq. richiardii, in the following countries
: Poland [
91,
92], Hungary [
70,
93] and in Germany [
94,
95].
In the present study S. tundra was found in Vojvodina Province in two analyzed mosquito species: Ae. vexans and Ae. caspius. When the blood meals of other mosquitoes from the same location and a nearby one were analyzed, the results showed that two mosquitoes of Ae. vexans were feeding on roe deer, and one had fed on a sheep. The DNA from the blood meal of Ae. caspius was not successfully identified. It is interesting to note that, upon analyzing the locations where these mosquitoes were collected, their traps were not very near to the forests. One trap is located in the middle of a human settlement, while the other one is approximately one km away from the settlement. The second trap was actually placed between a field of sunflower and a vineyard. Bearing in mind that the tested mosquitoes contained the blood of deer, we can assume two possibilities. Either mosquitoes flew from the forest to the humans’ vicinity (Ae. vexans and Ae. caspius have good flight capacities and can fly long distances) or the deer did it.
The number of analyzed mosquitoes did not yield a high number of positive cases of either Dirofilaria or S. tundra. Therefore, we cannot determine the prevalence. According to previous studies that focused on the detection of Dirofilaria in animals and humans, the expected positivity in mosquitoes was much higher than what was demonstrated in this study. It is necessary to perform a systematic screening of mosquitoes, at least in the locations with positive animals and humans, to better understand the prevalence and behavior/preferences of the parasite and to determine potential risks for human and animal populations.
Author Contributions
Conceptualization, MK, DP, CP, AIĆ; methodology, DP and CP; formal analysis, MK, SŠ, TM; investigation, MK, SŠ, TM, DJ; data curation, MK, SŠ, TM, DJ, AV; writing—original draft preparation, MK, SŠ, TM; writing—review and editing, MK, SŠ, TM, AIĆ, DJ, AV, CP; visualization, MK, SŠ, TM, AV; supervision, DP and CP; funding acquisition, DP and CP. All authors have read and agreed to the published version of the manuscript.