Introduction
Ectoparasites are often medically important arthropod vectors that play a significant role in disease transmission in many countries and environments. Many of these diseases are zoonotic, being transmitted between animals and humans via a variety of transmission mechanisms associated with arthropod vectors [1]. Two of the most important groups of ectoparasites are ticks and fleas, which are commonly found in companion animals, wildlife and the environment [2,3]. Rickettsiae are a group of zoonotic bacteria that are commonly reported in ectoparasites, including ticks and fleas, and cause a variety of infectious diseases in many countries, especially in South and Southeast Asia regions such as India, Thailand, Myanmar, Cambodia and Laos [4–7].
In Lao P.D.R. (Laos), several studies have described rickettsial infection. In patients admitted to Mahosot Hospital, Vientiane, 27% had evidence of rickettsial infection including 14.8% with Orientia tsutsugamushi (causative agent of scrub typhus), 9.6% with Rickettsia typhi (murine typhus) and 2.6% with spotted fever group rickettsiosis (including R. felis, R. helvetica and R. conorii) [8]. Rickettsial infections have also been reported from patients in the north and south of Laos, with 7% of non-malarial febrile patients with scrub typhus infections, and less than 1% for those with murine typhus, R. felis or undetermined Rickettsia spp. [9]. Slightly lower prevalences were reported by Mayxay et al. in the south of the country, with 2.6% of non-malarial febrile patients with scrub typhus, 0.9% with Spotted fever group rickettsioses (SFGRs) and 0.4% with murine typhus infection [10].
Ticks and fleas are hematophagous ectoparasites of vertebrates, such as dogs, cats, rats and cows [2,3]. Because of their hematophagous life-style, a number of species of ticks and fleas have also implicated as vectors of various rickettsioses, including SFGRs such as the flea-borne spotted fever or cat-flea typhus (Rickettsia felis), R. japonica, R. tamurae, and R. asembonensis [11–15]. This group of pathogens can infect animals and humans through the bite of an infected ectoparasite, or contamination of a bite site or injury by the fecal material of an infected vector, or a crushed vector itself, depending on the type and species of vectors. Infection in humans is usually non-specific, generally causing high fever, headache, myalgia, and rash [12].
Given the importance of ectoparasites such as ticks and fleas in the transmission of vector-borne zoonoses that can cause febrile illness, it is essential to investigate their diversity and associated pathogens [7,12]. In Laos, particularly in urban areas like Vientiane Capital, there are very few comprehensive reports on ectoparasite diversity and pathogen prevalence, despite close human–animal interactions, and the country’s location within a region recognized as a hotspot for emerging infectious diseases. Several studies in Laos have identified ectoparasites carrying clinically relevant pathogens. Rickettsia spp. were identified in 5.7% of tick pools collected directly from the environment and companion animals in Khamouane Province (central Laos), including R. tamurae, R. japonica and other Rickettsia spp. isolates, and a further 2.3% likely positive for Rickettsia spp. Additionally, other relevant non-rickettsial pathogens were also identified, including Borrelia spp. in 1.6% pools, Ehrlichia spp. in 1.6%, Coxiella spp. in 1% and Anaplasma spp. in 0.3% pools [16]. Whilst this study looked at free-ranging vectors, two studies from northern Laos have looked at vectors collected from domesticated dogs. Kernif et al., collected fleas from domesticated dogs in Luang Namtha Province, and detected Rickettsia felis in 76.6% of fleas and Bartonella spp. in 3.3% [17]. Calvani et al., identified Rickettsiae in fleas collected from both cats and dogs, with Rickettsia spp. (including R. felis) detected in 100% and Bartonella spp. in 33.3% of flea samples [18]. A recent study reported the presence of two members of the family Anaplasmataceae, Candidatus Anaplasma pangolinii and an Ehrlichia spp., in 43.7% of pangolin ticks collected from pangolins in Vietnam and Laos [19]. Nguyen et al, collected vectors from dogs in the urban area of Vientiane capital. Over 86% of flea pools were positive for Rickettsia spp. and 100% were positive for Anaplasmataceae species (likely Wolbachia spp.). In ticks, 14.3% were positive for Anaplasma spp. (two pools confirmed as A. platys), 6.7% were positive for Rickettsia spp. and all of lice pools (two) were positive for Rickettsia spp. Of the Rickettsia spp. positives, 46% were identified as R. felis [20]. The previous studies highlight the prevalence of clinically relevant pathogens identified in ectoparasites found on companion animals and in the environment in Laos, which may pose a possible danger of disease exposure to pets and people. Whilst the majority of studies have been done in rural regions of the country, with a small number in urban areas, there is no information on actual pathogen exposure to humans in urban areas with a high proportion of companion animals (dogs and cats) and a high population density in the country [20].
Identifying arthropod vectors is crucial for comprehending epidemiology, monitoring epidemics, and planning vector control programs. However, there is a challenge in morphologically identifying the species of a large number of individuals, which is a time-intensive process, as well as limitation in availability of entomological expertise and dichotomous keys. In addition, the engorged/unengorged status and immature life stages (particularly of ticks) may provide difficulties during morphological identification [21]. Molecular identification is essential to efficiently describe biodiversity of ectoparasites and has been used as an alternative to overcome the limitations of morphological identification. The mitochondrial cytochrome c oxidase subunit 1 (COI), a fragment of about 500–800 base pairs from mitochondrial DNA, is commonly used as the standard marker for DNA barcoding [21,22]. In Laos, few studies describe the genetic identification and divergence among ectoparasites, indicating the potential benefits of implementing molecular tools to support morphological identifications [18]. However, molecular identification is laborious, expensive, time-consuming and difficult to apply for species for which sequences are not available [21]. Previous studies have been shown the potential of matrix-assisted laser deionization time-of-flight mass spectrometry (MALDI-TOF MS) to allow high throughput identification of vectors, using small amounts of tissue, for a relatively low cost (when used as high-throughput). Studies have shown a concordance rate of 99.6% in tick identification using MALDI-TOF MS in comparison to morphological and molecular methods, and suggested a potential application of MALDI-TOF MS for species identification on ticks obtained from domestic animals and cattle [21].
In this study, arthropod ectoparasites were collected from dogs and cats in Vientiane capital. Identification of the vectors was done using morphological and molecular methods, along with molecular methods to determine any pathogen carriage. In addition, the present work evaluated the potential of the MALDI-TOF (BioMerieux, France) mass spectrophotometer to identify the species of ticks found on dogs in Laos.
Result
Throughout the enrolment period, a total of 282 dogs and 25 cats were recruited and examined for the presence of vectors (S1 Data). In total, 3,771 arthropods, including ticks, fleas and lice, were collected from 151 of dogs and four of the cats. Of the total animal enrolment, 69.3% (104/150) of dogs and all cats were healthy, 30.6% (46/150) of dogs were indicated to have an illness during the enrolment period, and 28.6% (43/150) reported an illness in the last 7 days (Table 1). A total of 3,653 ticks were removed from 151 dogs with the median of 9 per dog (inter quartile range or IQR 4.0-22.7, range 1–330), and five ticks were removed from a cat. Of 105 fleas, 82 were collected from 12 dogs (78.1%), with a median number of 2 per dog (IQR 2.0 -7.0, range 1–31) and 23 from four cats (21.9%), with a median of 5 per cat (IQR 3.5-7.3, range 2–11). Eight lice were obtained from two dogs (median 4/dog, IQR 2.5-5.5, range 1–7) (Table 1). Whilst the majority of dogs had only one type of vector, ten dogs (6.6%) were reported with both ticks and fleas, one dog with both ticks and lice, and one dog was reported with all three of these arthropods. In addition, one cat was reported with both ticks and fleas.
Species identification of ectoparasites
Morphological identification.
All ticks collected from dogs and cats were morphologically identified as Rhipicephalus sanguineus s.l.; the percentage of developmental stages were composed of 35.5% adult males, 36.7% adult females, 22.1% nymphs, and 5.7% larvae. The male to female adult tick ratio was 1:1.03. Ticks were classified by blood feeding status during specimen collection as blood fed (including partially blood-fed and fully engorged individuals) or unfed status. Of 3,653 ticks collected from dogs, 23.8% were blood fed adult females, 12.9% were unfed adult females, 15.7% were blood fed nymphs, 6.4% were unfed nymphs, 5.2% were blood fed larvae, and 0.5% were unfed larvae. Feeding status of adult males could not be determined (Table 2). From the one cat identified with 5 ticks, the number of developmental stages was composed of 1 adult male, 1 adult unfed female, 2 unfed nymphs, and 1 fed nymph.
From dogs, two species of fleas were morphologically identified from 82 specimens: Ctenocephalides felis felis (45.1%) and Ctenocephalides felis orientis (54.9%). The proportion of C. f. felis were 13.4% males, and 31.7% females. The male to female flea ratio was 1:2.36. The percentages of C. f. orientis were 15.9% males and 39.0% females, with a 1:2.46 male to female ratio (Table 3). The fleas identified on cats included 30.4% (7/23) of male C. f. felis and 69.6% (16/23) of female C. f. felis, with 1: 21.28 male to female ratio (Table 3). All of the lice found on dogs were identified as Heterodoxus spiniger, and comprised of 75% of male H. spiniger and 25% of female H. spiniger.
Molecular identification.
40 DNA samples consisting of 33 pooled tick samples, five pooled flea samples, and two pooled lice samples, were examined for the COI gene by PCR. All samples provided a positive result with expected band at 650 base pairs. A subset of PCR products were sent for sequencing to confirm morphological identification. Pooled tick samples showed 100% similarity to Rh. sanguineus s.l. from Colombia (KT906182), Angola (MF425994), Vietnam (PP389595) and China (OQ704652, OQ704663, OQ704678, JQ737084), confirming morphological identification. Specimens of C. f. felis had 97-99.8% similarity to C. f. felis from India (KX467335) and Thailand (OQ291320), whilst C. f. orientis specimen had 99.8% similarity to C. f. orientis from Thailand (OQ291337). A louse specimen was 97.9-99.5% identical to H. spiniger from South East Asia regions (MT027225) and Saint Kitts and Nevis (OQ779697) (S1 Table). The representative sequences of each ectoparasite species found on dogs and cats were submitted to the GenBank data base under accession numbers PV341327-PV341341 for ticks and PV341342-PV341345 for fleas (see S2 Table).
Phylogenetic tree of the partial COI gene of the representative ectoparasites collected from dogs were analyzed using MEGA11, with the comparison of the sequence obtained from the NCBI GenBank, rooted by Drosophila melanogaster (MZ631766). The representative COI gene sequences of ectoparasites in this study, including Rh. sanguineus s.l. (P004-T1), C. f. felis (P008-F1), C. f. orientis (P018-F1), and H. spiniger (P031-L1), showed a different genetic clade compared to the sequence obtained from GenBank (Fig 1). C. f. felis was more closely related to C. f. felis obtained from dogs in Malaysia (KY800498) and C. f. felis from stray dogs in Thailand (MH523417) than C. f. felis collected from red fox in Australia (JN008917). C. f. orientis sharing clade with C. f. orientis collected from dog in Thailand (MH523412) and Malaysia (KY800499, KR827040). Rh. sanguineus s.l. collected from dog in this study and Rh. sanguineus collected from dogs in Malaysia (MH481878) and Thailand (MZ401443) were more closely related to each other. H. spiniger was genetically closer to H. spiniger collected from a dog in East- and Southeast Asia region (MT027225) (i.e., Vietnam, Philippines, Thailand, China, Taiwan).
Phylogenetic were construct by maximum likelihood analysis using MEGA11, with 1,000 number of Bootstrap replicates. Percent homology of ectoparasite indicates as the number in the branch points.
https://doi.org/10.1371/journal.pntd.0013625.g001
MALDI-TOF MS evaluation.
Of the 79 tick leg samples analyzed by MALDI-TOF MS, good quality spectra were obtained from 14 tick leg samples, either preserved at -20°C (26.09%) or preserved in 70% EtOH (73.91%). Protein analysis was conducted in the range of 2–20 kDa. Key peaks of mass-to-charge ratio (m/z) were identified at around, 4,023 (4,023–4,024), 4,088 (4,088-4,089), 6,152 (6,151–6,154), 8,047 (8,045–8,049), 11,492 (11,477–11,588) and 12,292 (12,217–12,292) Daltons (Da). Spectra profiles were similar to spectra published by other studies [21,30–33], with some limited variation. Published Rh. sanguineus profiles indicated key identifying peaks at 4,020, 6,082, 8,042, 11,500, and 12,207 Da (S1 Fig).
Pathogen detection
PCR screening for bacterial pathogens was conducted on 296 pools of ticks from 142 dogs, 3 pools of ticks from a cat, 15 pools of fleas from 12 dogs, 4 pools of fleas from 4 cats, and 2 pools of lice from 2 dogs (Table 4). Among pooled samples from dogs, 13.8% (41/296) of tick pools were positive for Anaplasmataceae, and 1.7% (5/296) were positive for Rickettsia spp. Of the Anaplasmataceae, 21.9% were identified as Ehrlichia canis, with 97.6-100% similarity (accession no MN922610, KT357374), 39.0% as Anaplasma platys (96.7-100% identity, accession no CP046391, KT359590, MN922608, KU586028, LC126863) and two pools (4.9%) with 91.9-94.2% similarity to R. asembonensis (accession no MN003387, MK923744). In addition, Rickettsia spp. were detected in 11.1% of C. f. felis and 83.3% of C. f. orientis pooled samples. One C. f. felis pooled sample was 100% match to R. felis (accession no MK509750.1). Rickettsia spp. detected in all of C. f. orientis pooled samples were 99.8-100% similarity to R. asembonensis (accession no MK923744). One of the 2 lice pools was positive for R. asembonensis with 91.7% identity (accession no MK923744) (Table 4). Relating the positive pools to host animals, 22.5% of dogs with ticks (32/142) had at least one tick pool positive for Anaplasmataceae (E. canis, A. platys), and 3.5% positive for Rickettsia spp. Lice from one dog was confirmed positive for R. asembonensis (Table 5).
Among pooled samples from cats, 50% of fleas (C. f. felis) were positive for Rickettsia spp., with one sample confirmed as R. felis by sequencing (accession no MK509750). There were no positive result from ticks found on cats. In addition, 100% of C. f. felis pooled samples collected from dogs and cats, and 83.33% of C. f. orientis pooled samples collected from dogs were positive for Anaplasmataceae. Sequencing these as Wolbachia spp. in 50% and 33.3% of C. f. felis (collected from cats and dogs, respectively), and 16.6% of positive C. f. orientis collected from dogs (accession no CP051156) (Table 4). A total of five dogs had C. f. orientis positive for R. asembonensis, one of them also had fleas positive for Wolbachia spp., whilst 11.1% (1/9) dogs with Wolbachia-positive C. f. felis pools were also positive for R. felis. Of four cats identified with C. f. felis, one had fleas positive for R. felis (Table 5).
Rickettsia spp. in ticks was detected in 4% (1/25) of unfed nymphs, 2.9% (2/69) of fed nymphs and 2.2% (2/93) of fed adult females. Anaplasmataceae were found in 3.8% (1/26) of fed larvae, 4% (1/25) of unfed nymphs, 10.2% (7/69) of fed nymphs, 17.7% (14/79) of unfed adult females and 19.4% (18/93) of fed adult females. There were no pathogens detected in unfed larvae, Fig 2. The distribution of pathogens found in fleas and lice collected from dogs and cats are described in Figs 3 and 4, respectively. Where possible, estimated prevalence rates in individual vectors was calculated along with 95% confidence intervals, MinIR and MaxIR. The minimum possible infection rate (MinIR) and the maximum infection rate (MaxIR) was calculated to provide a possible range of pathogen infection rates in ectoparasites. Infection rates for Anaplasmataceae in ticks from dogs (E. canis and A. platys) infection rate was estimated at 43.06 per 1,000 (95%CI 31.62-57.33; MinIR = 4%, MaxIR = 19.2%). Rickettsia spp. in fleas from dogs was estimated at 127.13 per 1,000 (95%CI 59.79-247.24; MinIR = 9.6%, MaxIR = 56.2%), whilst Rickettsia spp. in C. f. felis collected from cats was estimated at 102.9 per 1,000 (95%CI 22.49-323.39; MinIR = 10.0%, MaxIR = 45.0%) (S3 Table).
Discussion
The presence of the causative agents of a number of clinically important in ectoparasites collected from animals and environment continues to pose a significant concern to animal and human health [16–18,20]. These pathogens have been reported in various regions of Laos, but mainly in the rural areas. This study confirms the distribution of vectors and the prevalence of pathogens in ectoparasites collected from dogs and cats in Vientiane, the capital city of Laos, a relatively densely populated urban area. The ectoparasite species collected from dogs and cats in this study are consistent with the diversity in species collected previously in Vientiane [20] and as described from dogs and cats in Northern Laos [18]. High prevalences of pathogens were seen, particularly in fleas collected from both dogs and cats. Rickettsia spp. (R. asembonensis, R. felis) were detected in 44.44% (8/18) of pooled fleas, collected from 12 dogs and four cats, with a possible infection rate in fleas ranging from 9.6% to 56.2%, based on possible minimum and maximum infection rates. Rickettsia spp. (R. asembonensis, R. felis) was detected in 100% (5/5) of C. f. orientis and 11% (1/9) of C. f. felis from dogs, with a possible minimum and maximum infection rate between 15.4% to 100% and 2.9% to 5.9%, respectively, and in 50% (2/4) of C. f. felis from cats, with 10% to 45% of minimum and maximum infection rates. Whilst the results suggests that there is a possibility of risk exposure to pathogens among companion animals and their owners, caution must be taken when basing interpretation on species of companion animal. The limited numbers of cats in this study being taken to veterinary surgeries may induce some bias in positivity data.
All ticks were identified as Rh. sanguineus s.l., a brown dog tick commonly found on dogs and previously described in the region [17,20]. As confirmed by sequencing, it is likely that these ticks are Rh. sanguineus s.l. tropical lineage (recently renamed as Rh. linnaei [37]). Rh. sanguineus is recorded as being highly suited to living in urban as well as rural areas, and is particularly adapted to living within human habitation, being active throughout the year in tropical and subtropical regions [38], and therefore its exclusive presence in this study is not unexpected. The fleas found on cats were identified as C. f. felis, while fleas found on dogs were both C. f. felis and C. f. orientis, corresponding to previous studies [18,20]. Whilst one study reports a higher incidence of C. f. orientis in dogs [17], this current study supports the results of by Calvani et al. [18], which indicated a higher proportion of C. f. felis compared to C. f. orientis among dogs and cats in the region using similar molecular identification (COI gene). There was no C. canis identified in this study, supporting previous reports [2,18,20,39]. The large number of ectoparasites collected, and therefore needing morphological identification, proved highly time-consuming and challenging at times. Therefore, identification at genetic level remains important to confirm and overcome the limitation of morphological identification. Correct morphological identification was confirmed by molecular identification through PCR and sequencing of the COI gene. Whilst there was a risk that pools might have accidentally contained different species of vectors, successful sequencing of PCR products indicated that this was not the case. Whilst molecular identification proved successful, it is still time consuming and resource intensive, particularly if ectoparasites cannot be pooled. This study described the initial information on the utilisation of the Maldi-TOF MS approach for use in species identification of ticks in Laos. The spectra for Rh. sanguineus s.l. obtained here showed similarity to key identification peaks obtained in previous studies [21,30–33]. Whilst only one species of tick was identified in this study and therefore trialed on MALDI-TOF, other studies have successfully differentiated between different species of ticks (such as Rh. bursa and D. marginatus [31]), and Rh. sanguineus infected with R. conorii [32]. Given that a proportion of the samples produced good peaks (14/79), further optimization on sample preparation is required to evaluate the use of MALDI-TOF in the Lao context, as this requires a high concentration of protein containing in the samples and appropriate condition for sample preservation. Inherent biological variations, such as feeding status and the presence of potential pathogens, may also influence the results [33]. Samples that were identified using all three methods gave matching species identifications, indicating that MALDI-TOF MS could be reliably used for high-throughput ectoparasite identification, once further optimization has been done. With no commercially available spectra database, immediate application of MALDI-TOF MS is limited until a region-specific database is built.
Rickettsia spp., a known genus of pathogens causing a variety of human rickettsiosis in Laos [8–10], was detected in ectoparasites collected from dogs and cats. R. felis was detected in 25% and 11% of fleas collected from cats and dogs, respectively. R. felis, is a known agent causing rickettsiosis [11] and has previously been reported in patients in Laos [8,9] as well as in ectoparasites including tick and fleas in rural and urban areas [17,18,20]. The R. felis-like organism R. asembonensis was found in 100% C. f. orientis, 50% H. spiniger, and in 1.4% ticks collected from dogs. R. asembonensis is seen as an agent with potential for human and animal infection, but with uncertain human pathogenicity at present [40] and is commonly found in fleas around the world, including South East Asia and neighboring country, such as Thailand [39,41,42]. In Laos, R. asembonensis was previously reported in 27.3% of C. felis collected from dogs [20]. This study reports R. asembonensis in two tick pools collected from dogs. There is little to no information regarding R. asembonensis in ticks, and may be due to ticks consuming an infected blood meal. Additional research is needed to confirm the possible implication of ticks in the R. asembonensis transmission cycle. Wolbachia spp. detected in all flea samples collected from dogs and cats is a known bacterial endosymbiont commonly found in fleas [20,43] but with no clinical implications for humans. However, its presence is noteworthy in a One Health context, as Wolbachia can influence vector biology and pathogen interactions, and has been considered in vector control strategies [44].
Pathogen detections were presented in each life stage and feeding status of tick (except for unfed larvae). Previous study reported pathogens (Rickettsia spp., Ehrlichia spp, Borrelia spp., Anaplasma spp., Coxiella spp.) in each life stage of tick collected from the environment and companion animals in the south of Laos [16]. Both Rickettsiae and Anaplasmataceae have been found to be transovarially and transtadially transmitted in ticks, therefore the presence of these pathogens in unfed ticks suggests that there is a possibility that pathogens could be transmitted to hosts (including humans) when feeding [3,45–47]. However, this should be further investigated to confirm the efficacy of pathogen transmission among ticks, and other vectors, and the risk of exposure for humans. There is a risk that pools will contain multiple pathogens which may be overlooked when relying on PCR identification. Using broad-range PCRs and sequencing of PCR products from pools in this study, indicated that this was not the case. In addition, pooled extraction of samples may result in inaccurate assessment of pathogen prevalence in assessing the risk of pathogens in individual vectors. Despite this, the minimum and maximum infection rates (IR) calculated in this study suggests there is potential for a high level of risk of exposure of humans to Rickettsia spp. from ectoparasite infested dogs and cats, especially in fleas, and highlights the need for raising awareness of adequate vector control with pets.
The detection of pathogens in ectoparasites collected from cats and dogs suggests a potential risk for subclinical or clinical infections in these animals. This highlights the importance of routine veterinary monitoring and ectoparasite control. Cats and dogs may act as reservoirs, and ectoparasites such as fleas and ticks serve as vectors. Consequently, humans in close contact with pets may be at risk of acquiring infections. These findings emphasize the importance of regular ectoparasite control, routine grooming, and vaccination in pets, along with public awareness campaigns, to reduce the risk of zoonotic transmission.
