Bartonella henselae and Rickettsia felis Detected in Cat Fleas ( Ctenocephalides felis ) Derived from Eastern Austrian Cats

Abstract

Cats and cat fleas (Ctenocephalides felis) are vectors of the zoonotic bacterial pathogens Bartonella henselae and Rickettsia felis, which are the causative agents of “cat scratch disease” and “cat flea typhus,” respectively. In the surroundings of Vienna (Austria), we identified 11 (10.5%; n = 105) B. henselae-positive fleas originating from 8 cats (20.5%; n = 39). One flea was positive for R. felis. There should be high levels of awareness among veterinarians and animal keepers as to the handling of cats, especially if free roaming, stray, or feral.

Introduction

B artonella henselae, the causative agent of “cat scratch disease” (CSD), aka cat-scratch fever, was divided into two main genotypes based on 16S rRNA gene (Chomel et al. 2006): type 1 (Houston strain), which is more frequently found in human patients and is linked to a higher pathogenicity, and type 2 (Marseille strain), which appears to be more abundant in cats and is not associated with invasive tropism to human liver and spleen tissues. Type 2 is supposed to be predominant in cats in the western United States, western Europe, and Australia, whereas type 1 is more prevalent in Asian cats, notwithstanding that the proportion of the types may vary in the different countries (Chomel et al. 2006). Although this classification into the two types is reasonable and thus currently used (Huwyler et al. 2017), other authors, for example, Bouchouicha et al. (2009), state that both types can be subdivided into several lineages if more discriminatory typing techniques such as MLST and MLVA are employed. However, cats and cat fleas (Ctenocephalides felis) represent reservoirs, where cats remain bacteremic for several months up to years (Huwyler et al. 2017). Cat fleas are responsible for increasing the feline reservoirs due to accumulation, proliferation, and dispersal of the bacteria (Gutiérrez et al. 2015). The transmission of the bacteria among cats and other mammals, including humans, routes through inoculation into wounds, which takes place during scratching with contaminated cat flea feces on their claws (Chomel et al. 2006).

In Austria, data from 1995 indicated a high risk for humans to contract B. henselae based on the finding of 33% seroprevalence in cats (Skerget et al. 2003). This is in accordance with a study on outpatients revealing 23% seropositivity, interestingly with no significant difference between pet owners and non-pet owners (Skerget et al. 2003). Even higher seroprevalence (65%) was found in Viennese zoo workers (Juncker-Voss et al. 2004). On the contrary, another more recent study on blood donors revealed 6% seropositivity for B. henselae (Müller et al. 2016).

Rickettsia felis is the causative agent of flea-borne spotted fever aka cat flea typhus in humans. Reports of patients with R. felis infection exist from America, Asia, Africa, the Pacific, and Europe (Pérez-Osorio et al. 2008). Knowledge of this bacterium is scarce, and many aspects of the ecology and epidemiology remain to be elucidated.

In well-protected cats, which regularly receive ectoparasitic treatment or are kept indoors, the contact with cat fleas harboring pathogens are extremely rare and the potential risk of pathogen transmission is considerably low. In contrast, feral, stray, and free-roaming cats scarcely receiving veterinary care are known to show significantly higher seroprevalence against B. henselae (Gerhold and Jessup 2013). Hence, in this study, we concentrated on this cat group and asked veterinarians participating in cat de-sexing programs or caring for stray and ownerless cats to sample fleas from those animals. The study aimed to obtain data for the occurrence of B. henselae and R. felis in a set of cat fleas derived from free-roaming or ownerless cats originating from Austria, to have a first glance at the actual risk outgoing from particularly this group of animals.

Materials and Methods

Fleas were sampled from cats during routine diagnostics from June to December 2016 by veterinarians in Vienna and Korneuburg/Langenzersdorf and stored in provided tubes with 75% ethanol at room temperature. The flea determination was done at the Institute of Parasitology, Vetmeduni, Vienna. Only cat fleas (C. felis) were found and further processed on an individual base to extract DNA. Fleas were air dried to let ethanol evaporate and placed in a 1.5 mL tube together with 50 μL of diethyl pyrocarbonate treated water and 3 mm steel beads, and homogenized by using a TissueLyser II (Qiagen, Hilden, Germany) for 2 min at 30 Hz. Thereafter, the supernatant was further processed with peqGOLD TriFast Isolation System (Peqlab, Germany) according to the manufacturer's instructions.

For each flea, several PCRs for the detection of Bartonella spp. and Rickettsia spp. targeting different genes were performed by using published protocols and primers at the concentration of 10 pmol/μL (Table 1). Positive B. henselae samples were further analyzed in a seminested approach with the primers 16SF and 16SR and Bhen16SS and 16SR (Table 1), and all products were sent for sequencing (Microsynth, Austria). The obtained sequences were uploaded to GenBank^® (B. henselae: ITS: MF374385; gltA: MF374384; 16S: MF374383; R. felis: ompA: MF374382; and gltA: MF374381).

Table 1.

Pathogens, Target Gene, Product Size, Annealing Temperature, Primer Pair, and References for the PCRs Used in this Study

Pathogen	Target gene	Product size (bp)	Annealing (°C)	Primer	Ref.
Bartonella spp.	gltA	379	54	BhCS.781p: 5′-GGGGACCAGCTCATGGTGG-3′	Norman et al. (1995)
Bartonella spp.	gltA	379	54	BhCS.1137n: 5′-AATGCAAAAAGAACAGTAAACA-3′	Norman et al. (1995)
Bartonella spp.	16S–23S ITS	500–800	66	BA325s: 5′-CTTCAGATGATGATCCCAAGCCTTYTGGCG-3′	Diniz et al. (2007)
Bartonella spp.	16S–23S ITS	500–800	66	BA1100as: 5′-GAACCGACGACCCCCTGCTTGCAAAGCA-3′	Diniz et al. (2007)
Bartonella henselae	I 16S rRNA	521	60	16SF: 5′-AGAGTTTGATCCTGGCTCAG-3′	Cited in Tapp et al. (2001)
Bartonella henselae	I 16S rRNA	521	60	16SR: 5′-CTTTACGCCCAGTAAATCCG-3′	Cited in Tapp et al. (2001)
	II 16S rRNA	477	66	Bhen16SSF: 5′-CACATGCAAGTCGAGCACACTCTTTTAGAG-3′	Tapp et al. (2001)
16SR: 5′-CTTTACGCCCAGTAAATCCG-3′	II 16S rRNA	477	66		Tapp et al. (2001)
Rickettsia spp.	gltA	381	50	RpCS.877p: 5′-GGGGGCCTGCTCACGGCGG-3′	Cited in Hodžić et al. (2016)
Rickettsia spp.	gltA	381	50	RpCS.1258n: 5′-ATTGCAAAAAGTACAGTGAACA-3′	Cited in Hodžić et al. (2016)
Rickettsia spp.	OmpA	∼632	47	Rr190–70: 5′-ATGGCGAATATTTCTCCAAAA-3′	Cited in Hodžić et al. (2016)
Rickettsia spp.	OmpA	∼632	47	Rr190–701: 5′-GTTCCGTTAATGGCAGCATCT-3′	Cited in Hodžić et al. (2016)

Results

In total, 105 cat fleas (C. felis) originating from 39 cats from eastern Austria were investigated. The collection comprised samples from the area Korneuburg/Langenzersdorf with 39 fleas from 26 cats, and the area Vienna with 66 fleas originating from 13 cats. From each cat, on average 2.7 fleas (max.19) were taken and one tube was used per cat.

Eight cats (20.5% [95% CI: 8.2–32.8%]) harbored at least 1 B. henselae-positive cat flea and another cat (2.6% [95% CI: 0–7.5%]) was carrying 1 R. felis-positive flea from the 105 investigated (0.95% [95% CI: 0–2.85%]). From the 105 fleas, 11 (10.5% [95% CI: 4.7–16.3%]) were found positive for B. henselae by using ITS and gltA primer and further processed with the 16S primer.

According to the “old” classification into two subtypes based on 16S rRNA polymorphism, five B. henselae samples were identical to type 2. The other six positive cat fleas delivered no Bartonella sp. sequence by using 16S primer (Huwyler et al. 2017), and therefore could not be assigned to a distinct type.

Discussion

It is rather intriguing that only five samples clearly could be linked to a distinct type (type 2), whereas the others did not reveal the other type (type 1). This investigated locus should be conserved and the other two gene loci (ITS, gltA) clearly delivered B. henselae sequences without any doubt. A double infection of different types could be possible, which would deliver the sequence of either one type, but not hamper the amplification at all. So there are two possible explanations for not obtaining a distinct sequence for six sequences. Most probably, the samples differ in the primer region of the particular 16S fragment and therefore delivered no useful amplification. By using another primer pair (data not shown), especially this region could be amplified, but no coincidence either to type 1 or 2 could be made. In addition, for three of the unknown six, the amount in the fleas was much too low, so that even after the nested step, no conclusive sequence could be obtained. Anyhow, these results clearly indicate the existence of more genotypes or lineages as already stated (Bouchouicha et al. 2009), and highlight the need for a standardized protocol and classification.

The finding of R. felis is of great interest and this pathogen should be observed in the near future, but no clear prediction can be made on accurate occurrence and distribution by now.

Conclusion

Free-roaming and stray cats from Austria may carry cat fleas harboring zoonotic pathogens, but feral cats pose less risk to human health than pet cats as they usually have minimal human contact. Nevertheless, in particular, those animals are difficult to handle and may show aggressive behavior if handled against their will. Therefore, individuals dealing with those animals such as veterinarians and animal keepers in animal shelters and animal welfare programs are under the risk of contracting CSD or cat flea typhus.

Footnotes

Acknowledgments

We would like to thank Dr. Föttinger and Dr. Gaisberger for collecting the fleas and Walpurga Wille-Piazzai (Institute of Parasitology, University of Veterinary Medicine Vienna, Austria) for her technical assistance. This study was conducted under the frame of EurNegVec COST Action TD1303.

Author Disclosure Statement

No competing financial interests exist.

References

Bouchouicha

, Durand

, Monteil

, Chomel

, et al. Molecular epidemiology of feline and human Bartonella henselae isolates. Emerg Infect Dis, 2009; 15:813–816.

Chomel

, Boulouis

, Maruyama

, Breitschwerdt

. Bartonella spp. in pets and effect on human health. Emerg Infect Dis, 2006; 12:389–394.

Diniz

, Maggi

, Schwartz

, Cadenas

, et al. Canine bartonellosis: Serological and molecular prevalence in Brazil and evidence of co-infection with Bartonella henselae and Bartonella vinsonii subsp. berkhoffii . Vet Res, 2007; 38:697–710.

Gerhold

, Jessup

. Zoonotic diseases associated with free-roaming cats. Zoo Public Health, 2013; 60:189–195.

Gutiérrez

, Nachum-Biala

, Harrus

. The relations between the presence and bacterial loads of Bartonella species in the cat and cat flea (Ctenocephalides felis), under natural conditions. Appl Environ Microbiol, 2015; 81:5613–5621.

Hodžić

, Fuehrer

H-P

, Duscher

. First molecular evidence of zoonotic bacteria in ticks in Bosnia and Herzegovina. Transbound Emerg Dis, 2016; 64:1313–1316.

Huwyler

, Heiniger

, Chomel

, Kim

, et al. Dynamics of co-infection with Bartonella henselae genotypes I and II in naturally infected cats: Implications for feline vaccine development. Microb Ecol, 2017; 74:474–484.

Juncker-Voss

, Prosl

, Lussy

, Enzenberg

, et al. Screening for antibodies against zoonotic agents among employees of the Zoological Garden of Vienna, Schönbrunn, Austria [Untersuchungen auf Antikörper gegen Zoonoseerreger bei Angestellten des Wieder Tiergartens Schönbrunn]. Berl Münch Tierärztl Wochenschr, 2004; 177:404–409.

Müller

, Reiter

, Schötta

, Stockinger

, et al. Detection of Bartonella spp. in Ixodes ricinus ticks and Bartonella seroprevalence in human populations. Ticks Tick Borne Dis, 2016; 7:763–767.

10.

Norman

, Regnery

, Jameson

, Greene

, et al. Differentiation of Bartonella-like isolates at the species level by PCR-restriction fragment length polymorphism in the citrate synthase gene. J Clin Microbiol, 1995; 33:1797–1803.

11.

Pérez-Osorio

, Zavala-Velázquez

, León

JJA

, Zavala-Castro

. Rickettsia felis as emergent global threat for humans. Emerg Infect Dis, 2008; 14:1019–1023.

12.

Skerget

, Wenisch

, Daxboeck

, Krause

et al. Cat or dog Ownership and seroprevalence of ehrlichiosis, Q fever, and cat-scratch disease. Emerg Infect Dis, 2003; 9:1337–1340.

13.

Tapp

, Roy

, Corstvet

, Wilson

. Differential detection of Bartonella species and strains in cat scratch disease diagnostics by polymerase chain reaction amplification of 16S ribosomal RNA gene. J Vet Diagn Invest, 2001; 13:219–229.