Phlebotomine sand flies (Diptera: Psychodidae: Phlebotominae) are the vectors of protozoan parasites of the genus Leishmania and of several viruses mostly belonging to the genus Phlebovirus of the family Bunyaviridae [1–3]. Zoonotic visceral leishmaniosis caused by L. infantum is the most dangerous form of Leishmania infection, being lethal when untreated. Domestic dogs are the main reservoir hosts for human infection. Canine leishmaniosis (CanL) is endemic in Portugal, with a nation-wide overall seroprevalence of 6.3%, being higher than 12% in some districts [4] while human visceral leishmaniosis is considered a hypoendemic disease [5].

Sand fly-borne phleboviruses pathogenic to humans such as Toscana virus (TOSV), sand fly fever Naples (SFNV) and sand fly fever Sicilian (SFSV) viruses are endemic in the Mediterranean region [1, 2]. Although most of human infections are either asymptomatic or influenza-like syndromes, TOSV has been considered emergent, as outbreaks and sporadic cases of acute meningitis or meningo-encephalitis due to this virus have been reported in southern Europe [1, 6]. On the other hand, SFNV and SFSV also frequently cause epidemics of febrile illness during the summer [1]. In Portugal, human cases of TOSV meningitis have been reported [7–10] along with the detection of antibodies to TOSV in blood donors, professionals with outdoor activity, environmental health care technicians, bird banders, hunters, individuals seropositive to Leishmania and individuals with symptoms and laboratory diagnostic request for vector-borne viruses [10, 11]. The presence of SFSV antibodies has also been detected in human sera [12]. In addition, Alcube, Arrabida and Massilia viruses have been isolated from phlebotomine sand flies collected in the southwest of Portugal [13, 14]. Although no animal reservoir hosts of these viruses have been identified yet, recent data showed that domestic animals are frequently exposed to them [15–20]. Seroprevalence rates up to 6.8% for TOSV and up to 50.8% for SFSV were recently described in dogs from Portugal [21].

As phlebotomine sand flies are the common vectors of Leishmania and of several phleboviruses, demonstration of an epidemiological link between human leishmaniosis and TOSV infections in southeast France was not unexpected [22]; this finding reinforces the importance of better understanding the interactions among sand fly-borne phleboviruses, Leishmania parasites, vectors and vertebrate hosts in order to ascertain their impact on public and animal health. Thus, the aim of the present study was to investigate a possible association between exposure to Leishmania, TOSV and SFSV in sheltered dogs from Lisboa and Setúbal districts, south of Portugal, where the presence of the protozoan [23] and both phleboviruses [21] was previously reported.

In the context of climate change and globalization, phlebotomine sand fly-borne diseases caused by Leishmania protozoan parasites and viral agents, among which the most important are grouped into the genus Phlebovirus, are emerging threats to public and animal health worldwide. In fact, in the Mediterranean region the geographical ranges of several agents of these diseases and their frequency are expanding [1, 3, 5, 30, 31]. Some of the assumptions that domestic dogs fulfil as reservoir hosts of L. infantum is the close relationship with humans together with the exposure to the bites of the same sand fly vectors [32]. Therefore, a similar presumption can be made regarding the role of dogs as reservoir, amplifying hosts or sentinels for human infections caused by sand fly-borne viruses. Thus, the present study was carried out to investigate a possible association between TOSV, SFSV and L. infantum exposure in sheltered dogs and to evaluate the main implications for public and animal health.

The sera samples examined in the present study were not collected to determine seroprevalence values but with the aim to investigate the possible relationship between exposure to the protozoan and to both phleboviruses. However, the percentage of dogs with antibodies to L. infantum (13.1%): (i) was higher than the one previously obtained in the same region but in dogs attending veterinarian clinics randomly selected (6%; [4]) or apparently clinically healthy (5.9%; [24]); (ii) was lower than the one obtained in dogs with clinical signs compatible with a canine vector borne disease (27.2%; [24]) or in stray dogs (28.6%; [27]) confirming the endemicity of CanL in Lisboa and Setúbal districts. The seroprevalence to TOSV (6.2%) observed here was in the same order of magnitude as those recently reported in dogs in Algeria (4.3%; [20]), in Cyprus (8.4%; [18]), in France (3.9%; [19]), in Greece (4.4%; [18]) and in Tunisia (6.8%; [16]) reinforcing that dogs can harbour TOSV and their role as sentinels should be investigated. Similarly, the seroprevalence to SFSV (56.3%) obtained in the present study reinforces previous results from Cyprus (60.2%; [18]), Greece (71.9%; [18]) and Tunisia (50.8%; [16]) that dogs are frequently exposed to the virus.

The present data revealed that sheltered dogs are a blood source for phlebotomine sand flies probably due to the lack of regular access to prophylaxis against ectoparasites (in the present study only 19.7% of the dogs were treated with an insecticide). In fact, the risk to possess SFSV neutralising antibodies was 1.6 times higher in dogs not using insecticides in comparison to the ones that use it, which is in agreement with the data obtained by Cortes et al. [4] where dogs not protected against insects had 1.5 times higher risk of Leishmania infection. Gender predisposition to Leishmania infection has been a field of discussion, with some studies reporting a higher prevalence in male dogs, others in female dogs and others not finding any differences among them [33]. With the exception of the work by Alwassouf et al. [18] in Greece where seroprevalence of SFSV was higher in female dogs, and the present study, where seroprevalence to SFSV was higher in male dogs, no studies of gender predisposition to sand fly-borne phlebovirus have been performed.

On the other hand, the higher seroprevalence to L. infantum in pure breed dogs in comparison with mongrels observed in the present study might be related to a certain level of resistance to disease development by the latter [33] as previously observed in a national CanL survey [4]. Whether or not all dog breeds are identically susceptible to SFSV infection is not known, but the rate of SFSV neutralising antibodies was significantly lower in crossed breed dogs compared with pure breed. As increasing of antibody levels has been associated with clinical progression of disease, it is not surprising that we found that antibodies to L. infantum were significantly higher in dogs presenting clinical signs compatible with leishmaniosis [34]. Seroprevalence to L. infantum (albeit not statistically significant) increased with age probably associated to a cumulative exposure of older animals to the protozoan [33]. Furthermore, the increased contact with the vectors was likely the reason for the age trend observed for TOSV and SFSV antibodies in previous canine surveys [16, 19]. Although in the present study this trend was not observed, the seroprevalence to both viruses was significantly higher in dogs older than 7 years than in those aged 1 to 7 years.

Co-infection with Leishmania spp. and different sand fly-borne phleboviruses in the insect on one hand [35, 36] and in dogs on the other hand [17] have been reported. In addition, the detection of L. infantum and TOSV and Massilia virus in phlebotomine sand flies [13, 37], humans [22] or dogs [38] sharing the same environmental area has also been described. Thus, we evaluated if the seroprevalence to one of the tested pathogens was significantly associated with seroprevalence to the others. An association between the presence of antibodies to L. infantum and SFSV was observed as 51 dogs presented IgG against both microorganisms. Since there is no cross-reactivity between Leishmania and SFSV, dogs exposed to the protozoan are at greater risk of being exposed to SFSV (and vice versa). The capacity of SFSV to cause disease in dogs is currently unknown, nevertheless it is probably not pathogenic to them as no association between the presence of clinical signs and the exposure to this phlebovirus was observed. However, and since it can cause disease in humans, it will be important to determine if dogs may act as a potential reservoir or amplifying hosts for this pathogen. The only human data regarding the circulation of SFSV in Portugal were obtained in the seventies [12] with a haemagglutination technique prone to cross-react with phleboviruses from other serocomplexes [39]. Therefore, and as our results imply the presence of either SFSV or an SFSV-related virus in Portugal, it would be important to pursue serological studies in humans based on neutralization-assays to determine their level of exposure to SFSV. In addition, attempts to isolate SFSV or another virus genetically related to SFSV from phlebotomine sand flies and vertebrate hosts should be made as no strain has been isolated yet in Portugal.

In the present study no association between the presence of antibodies to L. infantum and TOSV was detected (only six dogs were seropositive for both pathogens), not corroborating the epidemiological relationship between human [22] and canine [38] leishmaniosis and TOSV infection previously suggested. Nevertheless, the seroprevalence rate obtained in dogs in the present study, together with the seroprevalence detected in humans (ranging from 2% in control population to 4.2% in population with central nervous system disease) [10] and the reports of human cases of TOSV meningitis [9, 10] demonstrate the circulation of the virus or TOSV-like viruses in Portugal.

Our data demonstrate the co-circulation of Leishmania and both phleboviruses in the same region, thus the detection of neutralizing antibodies to both TOSV and SFSV in 25 dogs and the presence of antibodies to the three microorganisms in one dog was not unexpected and is congruent with entomological data regarding the presence of some of the sand fly species responsible or putative responsible for their circulation [1, 3, 36, 39], namely Phlebotomus ariasi, Phlebotomus perniciosus, Phlebotomus sergenti and Sergentomyia minuta [40]. Given the low probability of co-infections by different pathogens in phlebotomine sand flies, as observed in nature [35, 36], it can be speculated that the mechanism leading to a double and even a triple serological reactivity in vertebrate hosts is more likely to occur due to successive bites by phlebotomine sand flies, of the same or of different species, infected by a single pathogen [17, 22].

This study demonstrates the exposure to and co-circulation of phlebotomine sand fly-borne phleboviruses belonging to two distinct genetic and antigenic groups and Leishmania infantum in sheltered dogs, reinforcing the need to implement efficient prophylactic measures to prevent sand fly-borne infections among vertebrate hosts including humans. The results also indicate that dogs are good sentinels for assessing human exposure to TOSV and SFSV and since both viruses are proven human pathogens, further studies must be performed to elucidate if dogs play a role as reservoir or amplifying hosts in the natural cycle of these viruses. It is also a matter of priority to address the public and animal health impacts of these phleboviruses in Portugal via serological and virological studies on both phlebotomine sand flies and vertebrate hosts, especially on humans.