High‐throughput microfluidic real‐time PCR for the simultaneous detection of selected vector‐borne pathogens in dogs in Bosnia and Herzegovina

Abstract A scarcity of information on the occurrence of zoonotic vector‐borne pathogens (VBPs), alongside a lack of human and animal health authorities’ awareness of pre‐existing data, augment the risk of VBP infection for local people and limit our ability to establish control programs. This holds especially true in low‐middle income countries such as Bosnia and Herzegovina (BiH). This dearth of information on zoonotic VBPs is bolstered by the inability of previously used diagnostic tests, including conventional molecular diagnostic methods, to detect the full spectrum of relevant pathogens. Considering this, we set out to apply a microfluidic qPCR assay capable of detecting 43 bacterial and protozoan pathogens from blood to accrue critical baseline data for VBPs occurrence in BiH. A total of 408 dogs were tested of which half were infected with at least one VBP of zoonotic or veterinary importance. Leishmania infantum was found in 18% of dogs, reaching a prevalence as high as 38% in urbanized areas of Sarajevo. These data highlight substantially higher levels of L. infantum prevalence when compared to that previously reported using conventional methods using the same samples. Additionally, this high‐throughput microfluidic qPCR assay was able to detect pathogens rarely or never reported in canines in BiH, including Anaplasma phagocytophilum (3%), Anaplasma platys (0.2%), haemotropic Mycoplasma (1%) and Hepatozoon canis (26%). Our report of the endemicity of important zoonotic pathogens and those of clinical significance to dogs emphasizes the need for urgent implementation of surveillance and control for VBPs in BiH, targeting both animal and human infections within the country.


INTRODUCTION
Canine vector-borne pathogens (VBPs) encompass a broad diversity of disease-causing organisms, that include viruses, bacteria, protozoa, and helminths predominantly unified by their ability to be transmitted via blood-feeding arthropods, such as ticks, fleas, sandflies, and mosquitoes. Some VBPs generate severe disease solely in dogs, whilst others use canines as natural reservoir hosts of zoonotic agents thereby posing a risk to cohabiting humans as well (Otranto et al., 2017). To date, there has been very little information exploring the diversity and prevalence of canine VBPs in Bosnia and Herzegovina (BiH), with the few studies conducted thus far focusing on a limited number of key pathogen species, typically those of zoonotic significance Omeragić et al., 2018;Colella et al., 2019;Omeragić et al., 2021;Maksimović et al., 2022). These limited published studies have recorded high prevalence of canine VBP infection, exacerbated by a growing population of stray dogs since the end of the Bosnian war, that facilitates transmission and maintenance of VBPs in BiH Katica et al., 2019;Katica et al., 2020;Omeragić et al., 2021).
Prior investigations into canine VBPs in BiH have identified numerous zoonotic pathogens, in particular, a high prevalence of Leishmania spp. in dogs, alongside sporadic case reports of human infections Colella et al., 2019;Obwaller et al., 2018;Vaselek, 2021).
The tick species Ixodes ricinus and Dermacentor reticulatus are the most common ectoparasites infesting dogs in BiH (Krčmar et al., 2014;Omeragić, 2011). As a consequence, there have been reports of the bacterial VBP Borrelia burgdorferi sensu lato, which includes the aetiological agents of Lyme disease that can be harboured by canids (Arapović et al., 2014;Dautović-Krkić et al., 2008), as well as the highly pathogenic, but not zoonotic, Babesia canis (Ćoralić et al., 2018).
Other canine VBPs that are prevalent in this country include the Oriental eyeworm Thelazia callipaeda, a zoonotic nematode parasite implicated in occasional human infections in Eastern Europe (Colella et al., 2016;Hodžić et al., 2014;Paradžik et al., 2016), and the apicomplexan pathogen Hepatozoon canis (Hodžić et al., 2015). However, there have been no reports to date of bacterial VBPs infecting Bosnian canines, likely due to a dearth of research given that evidence of such pathogens has been found in neighboring countries (Huber et al., 2017;Laušević et al., 2019). In addition, rickettsial pathogens have been molecularly detected in Bosnian ticks, including the zoonotic bacterial species Anaplasma phagocytophilum (Hodžić et al., 2017), the causative agent of potentially lethal human granulocytic anaplasmosis.  (Gondard et al., 2018;Michelet et al., 2014). More recently, advances in microfluidic technology have allowed for the development of microfluidic qPCR methods that employ microchips printed in chamber arrays as large as a 96 × 96, allowing for the simultaneous running of up to 9216 singleplex nanoliter quantity reactions (Gondard et al., 2020). The large number of reactions possible at any one time permits testing for numerous pathogens from multiple samples concurrently, reducing the amount of time, labour, and reagents used when compared to conventional qPCR, whilst also being able to easily identify coinfections. For example, microfluidic qPCR has been used to detect water-borne and tick-transmitted pathogens, permitting testing for as many as 37 organisms at the same time in various studies that have used it to unearth zoonoses as well as rare VBPs of veterinary importance (Gondard et al., 2020(Gondard et al., , 2018Michelet et al., 2014;Sprong et al., 2019).
Given the significant paucity of research exploring VBPs and the potential zoonotic risk posed by some of them, we set out to use a microfluidic qPCR to test for 43 bacterial and protozoan pathogens and accrue baseline data for their occurrence in BiH. This is particularly timely given that economic and societal upheaval in BiH following the Bosnian war have had wide-reaching impacts and may have potentially made the country vulnerable to an increase in neglected diseases such as those caused by VBPs.

High-throughput microfluidic real-time PCR
The BioMark real-time PCR system (Fluidigm, USA) was used for Briefly, amplifications were performed using 6-carboxyfluorescein (FAM)-and black hole quencher (BHQ1)-labelled TaqMan probes with PerfeCTa qPCR ToughMix, Low ROX (QuantaBio) following a previously described protocol (Michelet et al., 2014). Two kinds of controls per chip were used for experimental validation: a negative water control to exclude contamination and internal control, to check for the presence of qPCR reaction inhibitors comprising Escherichia coli strain TA B L E 1 Target organism, gene target, and primer sequences used in this study to confirm vector-borne pathogens detected by microfluidic qPCR

RESULTS
Positive controls were correctly identified in each cycle using their corresponding set of primers and probes. Of the 408 dogs sampled and tested, 204 (50%; CI: 45-55%) were found positive for at least one VBP across all field sites investigated ( Figure 2, Table 2

Leishmania infantum is a causative agent of VL and CL in humans with
prior research having an estimated annual incidence of 2-3 cases per 100,000 in BiH (Alvar et al., 2012), alongside occasional case reports (Gvozdenović & Miladinović, 1959;Obwaller et al., 2018). Our identification of L. infantum in dogs in BiH is made even more pertinent given that when national institutes and ministries in the country were The identification of dogs infected with A. phagocytophilum is a key finding given that dogs act as sentinel hosts for this important zoonotic pathogen, the aetiology of granulocytic anaplasmosis, which is a potentially lethal disease in humans (Carrade et al., 2009;Ganta, 2021).

Anaplasma phagocytophilum infects canines as accidental hosts in which
the disease is usually self-limiting, exhibited through clinical signs such as temporary lethargy, fever, inappetence, and lameness (Carrade et al., 2009). This pathogen has been only recently reported from dogs in BiH ticks in 2017 (Hodžić et al., 2017), other than being highly prevalent in ticks from Serbia, which shares an eastern border with BiH (Milutinović et al., 2008). This may be due to the high circulation of I. ricinus in the investigated area as well as to the role of other wild canids (e.g., foxes) as reservoirs for this pathogen species (Sgroi et al., 2021).
The apicomplexan species H. canis and the highly pathogenic B. canis were identified at a high prevalence in sampled dogs from BiH. Babesia canis is the most common species detected in dogs across Europe, causing anorexia and lethargy in its initial stage, that can later progress to clinical signs including fever, thrombocytopenia, and anaemia and can prove fatal Furlanello et al., 2005). This Babesia species has been previously identified in as many as 82.5% of dogs in BiH (Ćoralić et al., 2018) as well as in the neighbouring countries of Croatia and Serbia (Brkljačić et al., 2010;Davitkov et al., 2015). This study firstly reports H. canis in dogs in BiH, with this species comprising the most prevalent VBP detected in our cohort of sampled dogs, and the most common coinfection alongside L. infantum. Hepatozoon canis is transmitted to dogs by the ingestion of the tick vector as opposed to ectoparasite blood feeding and often causes a subclinical infection, although in some scenarios, for example in immunosuppressed hosts, it can generate disease (Baneth et al., 2003;Hodžić et al., 2015;Huggins et al., 2019). This VBP has been previously identified in 38.6% of red foxes from BiH (Hodžić et al., 2015) and less commonly by molecular means in dogs sampled in Serbia and Croatia (Gabrielli et al., 2015;Vojta et al., 2009 haemocanis in Bosnian dogs, as well as A. platys, is also novel data that extends the known endemic geographical range of these canine VBPs. Although none of these pathogen species is associated with significant lethality in dogs, A. platys is the aetiological agent of infectious canine cyclic thrombocytopaenia, a disease known to exacerbate clinical signs in dogs when identified in the context of VBP coinfections (Gaunt et al., 2010;Huggins et al., 2022). In addition, haemotropic Mycoplasma spp. can also produce mild disease in canines, however, such disease may be magnified in scenarios where the host is immunocompromized, potentially causing severe anaemia, leukopaenia, and cachexia (Compton et al., 2012). Similarly, co-infection with A. platys, Bartonella henselae and Mycoplasma haematoparvum seems to worsen clinical presentations in infected human patients (Maggi et al., 2013).
The high-throughput microfluidic qPCR assay applied herein is capable of detecting a large panel of agents and was instrumental for the identification of VBPs previously unreported in Bosnian canines, including species that may have been missed by conventional molecular methods. Typically, when utilizing cPCR or qPCR assays to conduct epidemiological surveys, prior data on pathogen prevalence in the region must be used to inform the selection of diagnostic targets, as each may only be able to detect one or a few pathogens. Therefore, in the context of regions with scarce previous epidemiological data selection of such conventional diagnostics may be suboptimal and pathogens TA B L E 3 Number of positive dogs and pathogen prevalence with 95% confidence intervals found for vector-borne pathogen coinfections in 408 dogs from ten different field sites in Bosnia & Herzegovina, as detected by microfluidic qPCR
Overall, the microfluidic qPCR method used in this study may be highly suited to large-scale epidemiological surveys of canine VBPs, particularly in countries with scarce prior data in conjunction with large free-roaming dog populations that may facilitate easy and unbridled VBP transmission (Katica et al., 2020;Omeragić et al., 2021). The large number of samples that can be screened concurrently for numerous VBP species, makes microfluidic qPCR techniques amenable to surveys that rely on a high-throughput diagnostic capability (Gondard et al., 2020;Michelet et al., 2014). Nonetheless, cross-validation might be needed for the characterization of some VBPs to a species level that is, through the use of conventional PCR and Sanger sequencing, which rules the microfluidic method less suited to diagnosis in a clinical setting, for which more specific and bespoke molecular or serological tests may be available.
Here our report of endemicity of the highly pathogenic and zoonotic pathogens L. infantum and A. phagocytophilum alongside a lack of awareness by the relevant animal and human health authorities regarding the occurrence of these pathogens in BiH, emphasizes the urgent need for allocation of resources to implement surveillance and control, targeting both animal and human infections within the country.

CONFLICT OF INTEREST
The authors declare no conflict of interest.

ETHICS STATEMENT
The authors confirm that the ethical policies of the journal, as noted on the journal's author guidelines page, have been adhered to. The study was performed in accordance with the Veterinary law of Bosnia

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.