Evidence of Brucella melitensis DNA in the Microbiome of Ctenocephalides felis from Pet Cats in Greece

Introduction

Brucella melitensis is the most prevalent Brucella species encountered in the Mediterranean region and responsible for the majority of human and animal brucellosis cases. Sheep and goats are the natural reservoirs of B. melitensis but a wide range of wild animal species may contribute to the propagation of the pathogen (Zheludkov and Tsirelson 2010). The cat involvement in the epidemiology of brucellosis is considered insignificant (Zheludkov and Tsirelson 2010); however, Wareth et al. (2017) described a cat infected with B. abortus possibly by cattle actively disseminating the pathogen in the environment through uterine discharge.

Fleas are not considered to participate in the natural transmission of B. melitensis. Nevertheless, experimental studies have demonstrated that the common cat fleas Ctenocephalides felis have the potential to harbor B. melitensis and transmit it to healthy animals under laboratory conditions (Tovar 1947).

Hereby we present our findings of detection of B. melitensis DNA in the microbiota of fleas collected from cats in Attica region, Greece.

Materials and Methods

During the period 2016–2017, a network of collaborating small-animal clinics in the region of Attica, Greece, was formed to collect fleas from pet animals and study their microbiome. Fleas were identified at the genus and species level and as male or female using taxonomical keys (Pratt and Stojanovich 1966) and were grouped in pools, stratified per animal host, and insect genus, species, and sex. A predetermined sample size of 100 flea-pools, collected from an equal number of parasitized animals and including only female C. felis, was selected as the most representative from the aspect of geographical coverage, animal host species, ownership status, sex, and age. Removal of superficial flora on flea exoskeleton was accomplished by a series of washings and mechanical stirrings in 70% ethanol, according to a previously published protocol (Andrews 2013). DNA extraction was performed in each flea-pool, using the QIAamp DNA Mini Kit (Qiagen, Hilden, Germany). 16S Next Generation Sequencing (16S NGS) was performed using the Ion 16S Metagenomics kit Thermo Fisher Scientific, Waltham, MA on an Ion Torrent Personal Genome Machine (PGM) (Thermo Fisher Scientific). QIIME ver. 2 (Caporaso et al. 2010) was utilized through the Ion Reporter v5.10 software (Thermo Fisher Scientific) for processing 16S metagenomic data against curated MicroSEQ v2013.1 and GreenGenes v13.5 16S rRNA reference databases. The nucleotide alignments had percentage identity cutoffs of 97% and 99% for identifying genus and species, respectively. The nucleotide sequences were additionally cross examined with the BLASTn tool. A genus detection protocol (Baily et al. 1992) and two species identification and differentiation protocols, the Bruce-Ladder PCR (Lopez-Goni et al. 2008) and a Real-Time TaqMan PCR assay (Hinić et al. 2008), were utilized for further supporting the Brucella spp. identification; a positive and a negative control were included in each PCR run.

Results

A total of 423 flea-pools, containing 409, 4, and 10 C. felis, Ctenocephalides canis, and Pulex irritans, respectively, were collected from 321 animals, that is, 225 cats, 95 dogs, and 1 rabbit. Among the 100 selected flea-pools for 16S NGS analysis, 67 were derived from cats (59 ownerless, 8 owned), with a median of 3 insects per flea-pool (IQR: 2–6), and 33 from dogs (12 ownerless, 20 owned, 1 unknown), with a median of 2 insects (IQR: 1–3) per flea-pool, respectively. The majority of the animal hosts resided in Central Athens (37%) and in East Attica (30%) (Fig. 1).

OPEN IN VIEWER

FIG. 1.

Flea-pools examined with 16S Metagenomics in Attica, Greece. Intensity of sampling per Regional Unit is illustrated with circles; dark circles represent flea-pools sampled from cats; light circles represent flea-pools sampled from dogs. Numbers represent the total flea-pools examined per Regional Unit. Triangles depict residence locations of animal hosts (cats) with flea-pools (designated A, B, C, D, and E) bearing Brucella melitensis DNA.

A total of five pools (designated A, B, C, D, and E), comprising one, one, one, three and seven fleas, respectively, were found positive for Brucella spp. DNA by the 16S NGS platform, with a total of 24, 226, 59, 47, and 32 Operational Taxonomic Units from each pool, respectively. These sequences, comprising 228 or 229 bp each, are depicted in the Supplementary Data. BLASTn search (in the highly similar option) revealed 100% homology (no gaps) with various Brucella spp. ribosomal 16S sequences. No species identification was possible using the 16S NGS platform or BLASTn tool.

All pools were positive using the genus conventional PCR for Brucella spp. DNA (Baily et al. 1992). The Bruce-Ladder PCR detected Brucella DNA, but species identification was not possible, as the fragments of 1682 and 1071 bp were not amplified, possibly due to the limited amount of DNA in the specimens. The TaqMan real-time assay, targeting the BMEII0466 gene, was positive for B. melitensis (Hinić et al. 2008).

The cat hosts of A, B, C, D, and E pools were 12 months, 4 years, 1 year, 8 months, and 2 months old, female, the first one with an owner and the remaining four stray, respectively. Two of the cats, corresponding to pools A and B, resided in semiurban areas, whereas the rest lived in an urban environment (in the town of Athens) (Fig. 1). Background epizootiological information, or any data regarding the clinical presentation or the habitat conditions of the animals, was not available.

Discussion

16S Metagenomics has widened our perception of the bacterial diversity in ecosystems, identifying or differentiating in depth bacterial species. In this study, while using this technique to study the microbiome of C. felis, the most cosmopolitan flea species from pet animals, in Attica region, Greece, we detected naturally occurring DNA of B. melitensis. Although there is a limited number of reports (Tovar 1947) suggesting experimental infection of C. felis, to the best of our knowledge Brucella spp. has never been detected in wild-type insects of this species.

Cat fleas are opportunistic obligatory hematophagous arthropods; in the absence of their normal host they can prey on any available warm-blooded animal, including humans. The fact that the fleas were collected from cats living in urban and semiurban areas, stray or with an owner, increases the level of concern as to the implications for public health.

Background information for the cat hosts could not be retrieved retrospectively. Previous reports suggest that rodents are susceptible to infection with B. melitensis (Zheludkov and Tsirelson 2010, Ayyal et al. 2019). An experimental study demonstrated that B. melitensis may persist for several weeks inside the erythrocytes of mice, suggesting that the pathogen could be transmitted from murine species to blood-sucking arthropods (Vitry et al. 2014).

In our study, the interpretation of the molecular outcomes was hindered due to lack of pathogen isolation from fleas and possibly serology tests in the cat hosts.

In conclusion, our findings are tentative; only culture could affirm the presence of viable bacteria in flea microbiome, and even in this case, the vectorial capacity would require further verification through carefully designed studies. Nevertheless, we conclude that the current findings raise public health concerns, and appropriate studies are required to elucidate the potential role of cat fleas as vectors of B. melitensis, the ecological pathways that may lead to their infection and the risk for public health.