Molecular detection and genetic diversity of Rickettsia spp. in pet dogs and their infesting ticks in Harbin, northeastern China

Background Pet dogs are important companion animals that share the environment within households, and play an important role in local community life. In addition, pet dogs also are reservoirs of zoonotic agents, including Rickettsia spp., thus increasing the risk of rickettsial infections in humans. It’s meaningful to investigate the epidemiology of rickettsial agents in pet dogs, and make contribute to the surveillance of rickettsioses in human in China. Results In this study, a total of 496 pet dogs’ blood samples and 343 ticks infested in pet dogs were collected, and the presence and prevalence of Rickettsia were determined by amplifying the partial gltA and 17-kDa genes, with an overall positive rate of 8.1 % in blood samples and 14.0 % in tick samples. In addition, the rrs, gltA, groEL, and ompA genes of rickettsial were also recovered to determine the species of Rickettsia detected furtherly. Sequencing blast and phylogenetic analyses revealed the presence of three human pathogenic Rickettsia species (Rickettsia raoultii, Candidatus Rickettsia tarasevichiae and Rickettsia felis) in samples associated with pet dogs. Moreover, all the sequences of Rickettsia that we obtained presented close relationship with others available in GenBank, and Rickettsia raoultii was the most predominant Rickettsia species infected in pet dogs’ blood samples or in tick samples. Conclusions This study provides the molecular epidemiology data about the Rickettsia spp. infection associated with pet dogs in urban areas of Harbin city. Three rickettisae species pathogenic to humans were identified from pet dogs’ blood and the infested ticks in urban areas of Harbin city. Considering the intimate relationship between human and pets, these results indicate the potential transmission risk of human rickettisal infections from pet dogs through ectoparasites, and also highlighting that more attention should be paid to rickettsial infection in pet dogs and the infested ticks from the “One health” perspective.


Background
Rickettsioses, which caused by Rickettsia spp., are important emerging vector-borne diseases in humans [1], and some have been reported to infect dogs [2,3]. Genus Rickettsia (family Rickettsiaceae, order Rickettsiales) are a large group of Gram-negative obligate intracellular prokaryotic microbes [4]. They are widely distributed throughout the world, and maintained and transmitted by vector arthropods such as ticks, fleas, mites, lice and mosquitos [5]. In recent years, an increasing number of novel Rickettsia species have been described, and human diseases caused by known or novel Rickettsia species have been continuously reported, such as Rickettsia sibirica subsp. sibirica, R. raoultii, R. subsp. XY99 and Candidatus Rickettsia tarasevichiae in China [6][7][8][9], and R. monacensis in Europe and South Korea [10,11].
As in many other countries, dog has become a bonded family member. The dog population in China was estimated to be between 150 and 200 million in 2012 according to the records of the Chinese Center for Disease Control and Prevention [23], and the number of pet dogs is likely to have increased with the economic development and the improved living standards in urban populations. Dogs are important companion animals that sharing the environment within households even in the modern and urbanized society. Regardless the benefits of having pet dogs, dogs are reservoirs of zoonotic agents, serving as a nutrition source to many arthropods that also feed on humans, increasing the risk of zoonotic infections [24,25]. Thus, dogs gain increasing public attention as they are considered as sentinels for infections by agents transmitted by vectors and may contribute to the spread of vectorborne diseases, including rickettsioses [26,27]. However, to the best of our knowledge, only a few reports about the Rickettsia spp. infection in pets were available in China [18,28,29]. Currently, the status of Rickettsia spp. infection in pets remains poorly studied.
Harbin city, the capital city of Heilongjiang province, is the largest city located in northeastern of China. In this area, human rickettsiosis cases have been frequently reported in the recent years [6,7,30,31]. Pet dogs play an important role in local community life. However, there is no epidemiological data about the Rickettsia spp. infection in pet dogs in urban areas of Harbin. Given the close bond that exists between humans and pet dogs, it is therefore prudent to better understand the public health risks that may be associated with the humananimal bond in local communities. To this end, we sought to obtain the information regarding Rickettsia species infecting pet dogs, investigate the epidemiology of rickettsial agents in pet dogs in urban areas of Harbin to better assess the risk of rickettsial infection in the local populations.

Samples collection
During May to December of 2019, a total of 496 blood samples were collected from pet dogs, which were sent for vaccination, or for general inspection, or for veterinary assistance with some disease in local animal hospitals located in Songbei, Daoli, Nangang, and Xiangfang districts of Harbin city (Fig. 1). The detailed information, including gender, breed, age and number of these pet dogs are described in Table 1. The age of these pet dogs was between 5 months and 8 years. Across all the sampling sites, the percentage of the female and male pet dogs was 57.

Genetic and phylogenetic analysis
The nearly complete rrs gene, gltA gene, groEL gene, as well as the partial ompA gene of rickettsial were also Sequence blast analysis showed that the nucleotide similarity of the rrs, gltA, groEL, and ompA genes that we obtained and the corresponding sequences of R. raoultii available in GenBank were 99.7 %-100 %, 99.6 %-100 %, 99.4 %-100 %, and 99.6 %-100 %, respectively. In the phylogenetic trees based on rrs, gltA, groEL and ompA gene, which were reconstructed based on GTR + Γ + I model using the maximum-likelihood (ML) implemented in PhyML v3.0, the representative sequences of R. raoultii recovered in blood and tick samples were clutered together with the corresponding sequences of R. raoultii available in GenBank, and shared close genetic relationship with strains of R. raoultii isolate Tomsk, R. raoultii strain IM6, and R. raoultii strain Khabarovsk (Fig. 2).
The sequences of Ca. R. tarasevichiae obtained in this study were closely related to each other, with 99.8-100 % nucleotide similarity. The representative sequences of rrs, gltA, and ompA gene of Ca. R. tarasevichiae that we obtained (Dog-145 and Tick-11) shared 99.7 %, 100 % and 99.4 % of nucleotide similarity with the corresponding sequences of Ca. R. tarasevichiae strains available in GenBank, respectively; while the groEL gene of Ca. R. tarasevichiae that we obtained shared 100 % nucleotide similarity with the Ca. R. tarasevichiae strain Mulan-11, which was identified in ticks collected in the remote areas of Harbin in our previous study (in press). Consistent with the results of genetic comparison, Dog-145 and Tick-11 were clustered together with other Ca. R. tarasevichiae strains available in GenBank in all four phylogenetic trees (Fig. 2).
In the sequence blast analysis, the R. felis detected in this study presented the highest nucleotide similarity (100 %, 99.8 %, 100 %, and 100 %) with R. felis URRWX-Cal2 for rrs, gltA, groEL, and ompA genes, respectively. In the phylogenetic trees, the representative sequence of R. felis that we identified was clustered with R. felis isolates, and showed the closet genetic relationship with R. felis URRWXCal2.

Discussion
Investigation of neglected zoonotic pathogens in domestic pets and vectors infecting them is important in the prevention and control of zoonotic diseases [32]. In the present study, blood samples of pet dogs, and tick samples infested in these pet dogs, were collected and screened for rickettsial agents in urban areas of Harbin city, northeastern China. Three rickettsiae species (R. raoultii, Ca. R. tarasevichiae and R. felis) and two rickettsiae species (R. raoultii and Ca. R. tarasevichiae), were identified from pet dogs' blood samples and the infested tick samples, respectively. Additionally, the overall positive rate of Rickettsia infection in ticks (14.0 %) was significantly higher than that in pet dogs' blood samples (8.1 %). Moreover, the overall positive rate of Rickettsia infection shown no correlation with gender or breed of pet dogs, but significantly associated with age (positive rate in juveniles were significantly higher than that in adults), which might because the underdeveloped immune system of juveniles caused the slowly recovery from Rickettsia infection; thus, the examined juvenile dogs were in acute phase of infection at the time of blood collection, and they were easier to identified. It might also because the juvenile individuals are more active than adult individuals, more frequently to contact with outdoors, and also have more chance to interact with other dogs. More importantly, R. raoultii, Ca. R. tarasevichiae and R. felis, which was identified in the present study, have been confirmed as the causative agents of rickettsioses in humans [7,[33][34][35][36][37]. Given that pet dogs are considered as human companions and share the household environment with humans, and they have been identified as a notable mammalian reservoir hosts for many Rickettsia species and may play an important role as source of human infection, it is meaningful and necessary that actively surveillance of rickettsial bacteria carried by pet dogs for better assess the risk of rickettsial infection in the local populations.
Rickettsia raoultii was first identified in ticks collected from Russia in 1999 [38], and it has been detected in at least 12 tick species collected from a large geographical  [14,15,19,30,39], which suggest the wide geographical and host species distribution of R. raoultii in China. In this study, the positive rate of R. raoultii detected in blood samples (5.6 %) or in tick samples (11.1 %) were significantly higher than other species of Rickettsia identified in the present study, suggested that R. raoultii was the most predominate Rickettsia species infecting pet dogs in this area. In addition, R. raoultii was identified in seven species of pet dogs and three species of ticks, and no significantly difference was observed among the different breed of pet dogs. Although only one R. raoultii positive sample was detected from D. silvarum, the positive rate of R. raoultii detected in D. silvarum was significantly higher than that in H. concinna or in I. persulcatus. Our results are consistent with previous reports that Dermacentor spp., was the major vector for R. raoultii [14,40]. Moreover, the positive rate of R. raoultii detected in I. persulcatus was 9.8 %, which was lower than that in H. concinna, or in D. silvarum. However, to our knowledge, this is the first report about R. raoultii identified from I. persulcatus in Harbin, although I. persulcatus have been confirmed as the predominant tick species in Heilongjiang province [14]. Candidatus Rickettsia tarasevichiae is an emerging tick-borne rickettsiae species initially found in the ticks of I. persulcatus in Russia [41], it's also has been confirmed pathogenic to human, and the first human case was identified in China in 2013 [7,42]. In the present study, Ca. R. tarasevichiae was exclusively detected in I. persulcatus out of three tick species collected in this study, which are consistent with previous reports presented that Ca. R. tarasevichiae was closely associated with I. persulcatus [14,43]. Moreover, I. persulcatus is a common human parasite in northeastern of China, and it is also known to transmit a variety of other human pathogens, such as Borrelia burgdorferi, Babesia spp., Anaplasma phagocytophilum [22]. These suggest that I. persulcatus could contribute greatly to the spread of Ca. R. tarasevichiae in China, and also raised the possibility of co-infection of humans by Ca. R. tarasevichiae and other tick-borne pathogens. Additionally, we did not amplify DNA of Ca. R. tarasevichiae from H. concinna collected in this study, although it has been confirmed to be the vector for Ca. R. tarasevichiae in northeastern China previously [14,22].
Rickettsia felis also is an emerging vector-borne rickettsiae species distributed throughout the world [44][45][46][47], and the biological vector is the cat flea, Ctenocephalides felis [48], although it has also been found in other arthropods (ticks, mites, booklice and mosquitoes). It has been confirmed as the causative agent of flea-borne spotted fever (FBSF) in humans, and posing a global emerging threat to human health [49,50]. Meanwhile, domestic dogs are considered mammalian reservoir hosts for R. felis [48]. In the present study, R. felis was detected from blood samples of pet dogs with an overall prevalence of 2.0 %. Previous study has showed a high seroprevalence (47 %), and a low PCR-based positive rate (0.8 %) of R. felis infection in dogs from China [18]. The positive rate of R. felis infection in dogs based on PCR detection was significantly lower than that determined by serology assay, might attribute to the fact that there is at least some serological cross reactivity between R. felis and other Rickettsia spp. present in China [51]. Additionally, 10.3 % of the ticks collected from dogs in previous study were PCR positive, while no positive sample was determined in tick samples in this study. The incongruent results might because the tick species collected in previous study was Rhipicephalus sanguineus, which also have been confirmed as the vector for R. felis [52], no reports about the R. felis infection in the tick species collected in present study are available.

Conclusions
In summary, three rickettisae species pathogenic to humans (R. raoultii, Ca. R. tarasevichiae, and R. felis) were identified from pet dogs' blood and the infested ticks in urban areas of Harbin city. In the light of the co-existence of humans and pet dogs, these results indicate the potential transmission risk of human rickettisal infections from pet dogs through ectoparasites, and also remind the public that reduce the contact of pet dogs with outdoors, and periodically remove the ectoparasites, will help to prevent the rickettisal infection of pet dogs and humans. From the "One health" perspective, these results also suggest that sharing information between public and animal health on ectoparasites of companion animals and the disease agents they carry, could contribute to surveillance of rickettsioses in human in China.

Collection of blood and tick samples
During May to December of 2019, blood and tick samples were randomly collected from pet dogs, which were sent for vaccination, or for general inspection, or for veterinary assistance with some disease in animal hospitals located in urban areas (Songbei, Daoli, Nangang, and Xiangfang districts) of Harbin city, Heilongjiang province of China. The blood samples were collected into vacutainer tubes containing ethylenediaminetetraacetic acid (EDTA) anticoagulant, and the adult ticks were collected from the animal body surface by hand searching. Species identification of ticks were firstly determined according to their morphology as previously described [53,54], then further confirmed by amplifying, sequencing and analyzing the 16 S ribosomal RNA (rrs) gene of ticks [55]. The detailed information of pet dogs was recorded by veterinarians before blood sample collection.

DNA extraction andRickettsiaedetection
Total DNA was extracted from 200 µL of whole blood samples by using the Blood DNA Extraction Kit (OMEGA, USA) according to the manufacturer's instructions. The ticks were soaked in 70 % ethanol for sterilisation and then washed three times with double distilled water (ddH 2 O). The whole body of each tick was homogenized, and the total DNA was extracted from individual tick by using a Tissue DNA Extraction Kit (Omega, USA) according to the manufacturer's protocol. The extracted DNA was eluted in 80 µL autoclaved double distilled water (ddH 2 O) and stored at -20˚C.
Using the extracted DNA as templates, the rickettsial DNA was detected by nested-PCR with the primers targeting a 720-bp citrate synthase encoding gene (gltA) and a 450-bp 17 kilodalton antigen encoding gene (17-kDa) segments as previously described [56,57]. During the detection of Rickettsiae, a DNA sample of R. japonica and ddH 2 O were used as positive and negative control, respectively. To prevent contamination, the template isolation, PCR mix preparation, template DNA addition, and agarose gel electrophoresis were performed in separate rooms, and dedicated pipets and tips with filter elements inside were used.

Amplification of rickettsialrrs, gltA, groEL, andompAgenes
Only both the gltA and 17-kDa gene positive total DNA samples were selected to amplify the nearly entire rrs gene (1300 bp), gltA gene (1200 bp), heat shock protein-60 encoding gene (groEL, 1500 bp) as well as the partial outer membrane protein A encoding gene (ompA, 530 bp) of rickettsial to determine the species of Rickettsia detected. All the primer sequences and PCR parameters used in this study were shown in Table 3 detailed.

Sequencing of the PCR products
The PCR products were examined in agarose gel using an Image Analyzer (Bio-Rad, USA) after electrophoresis. PCR products less than 800 bp were sequenced directly, and the PCR products more than 800 bp were firstly purified using a Gel Extraction Kit (Qiagen, USA) according to the manufacturers' protocols, then cloned into pMD19-T plasmid (TaKaRa, China) and transformed into E. coli JM109 competent cells. For each amplicon, the positive inserts were confirmed by PCR, and three positive clones were sequenced with the universal M13 forward and reverse primers in cloning vector. All the sequencing procedures were performed by Sangon Biotechnology Company in China.

Sequence comparison and phylogenetic analysis
The obtained nucleotide sequences from the target genes of rickettsiae were edited and assembled using SeqMan program (DNASTAR, Madison, WI). Multiple nucleotide sequence alignment was performed using  [58]. The nucleotide identities were calculated using the DNASTAR Lasergene 12 Core Suite [59]. All the sequences obtained in this study were compared with the corresponding sequences available in GenBank using BLAST software, and the sequence data have been submitted to GenBank databases under accession numbers MT003286-MT003297 and MT019627-MT019662.
The best-fit nucleotide substitution models for phylogenetic analysis based on the different genes of rickettsiae were determined using jModel Test [60]. Phylogenetic trees were reconstructed using Maximum likelihood (ML) method implemented in PhyML v3.0 software [61]. The reliability of branches in the inferred trees was evaluated by bootstrap analysis of 1000 replicates, and the values over 70 % were considered as significant differences for presentation. All phylogenetic trees were mid-point rooted for purposes of clarity only.

Statistical data analysis
Statistical analysis of the obtained data was performed using Statistical Package for Social Sciences Version 21.0 software (SPSS, Chicago, IL, USA). Chi-square test or Fisher's exact test was used for calculating the P-value to determine the differences of the positive rate of rickettsiae. Statistical significance was defined as P < 0.05.