Bacterial and protozoal agents of canine vector-borne diseases in the blood of domestic and stray dogs from southern Portugal

This is the most comprehensive study carried out in dogs from southern Portugal on
the prevalence of infection with CVBD agents as it included domestic and stray animals
with and without clinical signs compatible with a vector-borne disease. DNA from these
pathogens taken all together was less frequently detected in dogs (6.7%; p??0.001) than in cats (29.9%; 194/649) from the same region 26]. Furthermore, only one (0.1%) dog was found co-infected (with two pathogens), whereas
29 (4.5%) cats were positive to two agents and four (0.6%) cats to three agents 26].

In this study A. platys has been molecularly confirmed to infect dogs from the south of the country, corroborating
previous detection of this bacterium in dogs 11],23] and in R. sanguineus32] from the same region. The prevalence of positivity to Anaplasma/Ehrlichia in this work (1.9%) was lower than the 4.0% obtained in Spain 33] and the 3.7-6.0% in Italy 34], which might be related with the targeted population. In fact, in the works of Tabar
et al. 33] and Trotta et al. 34], all the positive dogs were sick animals with clinical signs compatible with vector-borne
diseases and admitted for medical treatment, while in the present work most of the
enrolled animals were apparently healthy. Interestingly, in our study most of the
animals harbouring Anaplasma/Ehrlichia DNA were from Faro, overlapping the Algarve region, the southern most district of
continental Portugal, which seem to follow the trend revealed by Cardoso et al. 3] that the prevalence of antibodies against Anaplasma spp. and E. canis in dogs from southern Portugal was significantly higher than in dogs from the northern
and central regions of the country.

In the present work, B. burgdorferi s.l. DNA was amplified from 0.8% of the screened animals, providing the first molecular
evidence of naturally occurring B. burgdorferi s.l. infection in dogs from Portugal. The exposure of dogs to these spirochetes was
previously demonstrated by specific serology in the Algarve 35] and in the Alentejo and Lisbon regions 3]. Furthermore, B. burgdorferi s.l. genospecies, Borrelia lusitaniae was isolated from humans 36]-38] and DNA of B. burgdorferi s.l. was detected in ticks 32],39] and cats from the south of the country 26]. Nevertheless, information on the clinical signs associated with Borrelia infections in dogs and their role as sentinels is still limited 6].

H. canis was the most prevalent pathogen detected in all the assessed dogs, with a significantly
higher prevalence in animals living in the Algarve. In fact, H. canis has recently been identified in dogs 23], in R. sanguineus collected from dogs living in this region 32], and also in foxes from the south 40], showing that the protozoan is widespread in this area of the country. Although in
this study only three out of the 31 infected dogs presented clinical signs, subclinical
infections should not be neglected as they may progress to a severe disease and warrant
treatment 41]. Concurrent infections of H. canis with other canine pathogens are common 21]; however, in the present work only one animal apparently healthy was co-infected
with A. platys and the protozoan. Although this individual dog had no clinical signs of a CVBD,
co-infections may potentiate disease pathogenesis, altering clinical manifestations
associated with single infections 42].

The overall prevalence of L. infantum infection in the present study (1.1%) was much lower (p??0.001) than the 34.9% obtained in 152 dogs from Lisbon 43]. The lower detection of Leishmania DNA might be due to the: (i) dynamics of infection over time, which may depend on
the abundance and distribution of the proven vector species in conjunction with the
number of infected vertebrate hosts 44], and (ii) insufficient data regarding the duration of parasitaemia in infected dogs.
In fact, and taking into account a seroprevalence of 18.2% recently obtained in 170
dogs from the Algarve region 45], PCR with blood should be used to complement serological results and not only by
itself to detect Leishmania infection, as it can lead to false negative results, especially in subclinically
infected dogs 46].

PCR-positivity to one or more genera/complex of CVBD agents was found to be associated
with domestic dogs, with animals living in the Algarve and with an outdoor or mixed
(i.e. with outdoor access) housing. In fact, most of the domestic dogs harbouring
DNA of the studied pathogens lived in rural areas from the Algarve region and used
to spend most of their time exclusively outdoors, thus increasing their exposure to
arthropod vectors and the agents they might transmit.

The role of domestic dogs as reservoirs of Bartonella spp. is less clear than for cats, and the former are probably accidental hosts. Nevertheless,
they are excellent sentinels for human infections because a similar disease spectrum
develops in dogs 47]. Serologic and molecular evidence of Bartonella henselae and Bartonella clarridgeiae exist for cats from the south of Portugal 26],48]. Thus, the non-detection of Bartonella DNA in the present study might be related with differences in immune responses, host
preference of particular vectors or innate resistance in dogs to these bacteria. On
the other hand, the definitive diagnosis of Bartonella infection is challenging due to the fastidious nature and intracellular tropism of
these bacteria for erythrocytes and endothelial cells 49]. According to Perez et al. 50], enrichment culture and subculture, followed by PCR amplification, enhances molecular
diagnostic sensitivity in dogs. Thus, it is possible that the PCR done directly from
blood samples might have missed some positive cases; nevertheless, the prevalence
of infection at the population level, if any, must be very low.

Albeit the detection of B. canis, B. vogeli and the B. microti-like piroplasm has already been reported in dogs from the north of Portugal 13],19],20],22] and B. vogeli in dogs from the south of the country 23], in the present study none of the screened animals harboured piroplasmid DNA. The
non-detection of B. canis could somehow be expected as its vector, D. reticulatus, is more abundant in the north of the country. However, the absence of B. vogeli and B. microti-like DNA is more difficult to explain, since both have already been detected in southern
Portugal, the former in cats 26], dogs 23] and ticks 32], and the latter in foxes 51]. According to a recent questionnaire-based survey on the distribution of canine babesiosis
in western Europe, the annual incidence of this parasitosis in southern Spain, which
is geographically close to the area surveyed in this study, was estimated to be 0.0-0.7%
52]. Furthermore, a 58% prevalence of antibodies anti-Babesia spp. was reported among 331 dogs from kennels/shelters in southern Portugal 53]. The absence of Babesia spp. infection in the present study might be related with differences in the genetic
background/immune system or between vector-dog interactions. Further studies are needed
to better understand the epidemiological importance of these findings.