Coxiella burnetii as the causative agent of the zoonosis Q fever is distributed worldwide, except New Zealand. This infectious disease may have a significant impact on animal welfare, human health, and economies (1, 2). Coxiella burnetii is a gram-negative bacterium that replicates as an obligate intracellular pathogen under acidic and microaerophilic conditions in phagolysosome-like compartments of cells, predominantly macrophages (3). Infection of hosts mainly occurs due to inhalation of contaminated dust and aerosols (4). During infection, C. burnetii shows a tropism for the reproductive tissue and the mammary gland. Therefore, C. burnetii is primarily shed during parturition and via the milk (5). The considered main reservoirs for human infections are domestic ruminants, whereas other species, such as dogs or horses, can play a minor role as carriers (6). In their birth products, large amounts of infectious agents can be detected (7, 8). As the infective dose by inhalation is <10 bacteria, farmers, and veterinarians are especially at high risk of acquiring this disease through contact with infected animals and their products (9, 10). Coxiella burnetii develops spore-like forms, which are resistant to environmental stressors such as desiccation or sunlight. Therefore, the bacteria survive under adverse conditions over long periods in the soil or other dry substances (11).

An infection in ruminants, also termed coxiellosis, is often asymptomatic and not noticed until human Q fever cases occur (12). Decrease in fertility and increase of abortion and stillbirth are common indications of an ongoing Q fever disease in herds or flocks of ruminants (13, 14).

Outbreaks of Q fever, as seen in the Netherlands in 2007–2010, clearly show the huge impact of this infectious disease not only on agricultural economics but also on human health (15, 16). Acute human Q fever infection presents mainly flu-like symptoms with possible complications concerning the lung and liver. Although 60% of acute cases remain asymptomatic, in 1–5% of cases, chronic Q fever may develop (17, 18). This often affects the heart valves as endocarditis. Antibiotic treatment is mandatory over months and, in some cases, up to several years (17–19). Besides the long and difficult treatment, the disease can also be fatal in acute and chronic cases. Although the mortality of acute Q fever was assessed to be 1% in a study on hospitalized patients, the mortality in chronic cases has been reported as 13% and even up to 38% during the Dutch outbreak mentioned earlier (1, 18, 20).

The bacterium is known as an infectious agent since the 1930s when it caused an outbreak of query fever among abattoir workers. In the years after its first isolation from a Dermacentor andersonii tick (21), early experimental studies showed horizontal transmission of the agent from ticks to mammalian hosts and suggested that several tick species, e.g., Haemaphysalis humerosa, Haemaphysalis bispinosa, and Rhipicephalus sanguineus, may be capable of transmitting C. burnetii (22–25). Although transstadial transmission was shown for most of the examined tick species, the transovarial transmission was reported rarely (25, 26). Later, it was demonstrated that Hyalomma dromedarii infected by inoculation excreted C. burnetii with their saliva (27). Duron et al. summarized the vector competence data for seven hard tick species (28). Hence, since its first detection, C. burnetii has been discussed as a ruminant-associated tick-borne bacterium.

During outbreaks of Q fever, ticks are regularly tested for C. burnetii. Until today, no C. burnetii-positive ticks associated with outbreaks have been documented. Human infection after tick exposure has been reported, but due to the predominant infection route by inhalation, the source of infection remains unclear (29, 30).

Besides pathogen transmission, ticks are host to many endosymbionts and harbor a diverse microbiome (31). Although C. burnetii was the only known member of the genus Coxiella for a long time, closely related bacteria were found in recent studies, referred to as Coxiella-like endosymbionts (CLEs) (32, 33). This heterogenic group of bacteria with different genome sizes and gene content shows up to 97% genome identity with C. burnetii (33–37). Genetically, CLEs are classified into four clades (A–D), and C. burnetii belongs to clade A, which otherwise contains CLE associated with soft ticks (38).

Coxiella-like endosymbionts were detected in many different tick species, and for some species, an obligatory mutualism was proven (39–41). As obligate hematophagous arthropods, ticks are dependent on external vitamin B synthesis, likely supplied by the harbored bacteria (40, 42). Furthermore, a positive effect on fecundity was shown for R. sanguineus. Treatment with ofloxacin resulted in a significantly lower egg mass, hatching rate, and viability of larvae in this tick species (39). Similar results were shown for Haemaphysalis longicornis, Rhipicephalus microplus, and Rhipicephalus haemaphysaloides, which were treated with tetracycline and kanamycin, respectively, indicating the influence of CLE on reproduction (43–45).

CLEs seem to be apathogenic, but single studies showed pathogenic potential in particular cases. Fatal infection with CLE was occasionally observed in different bird species (46–48). Furthermore, a skin-associated inflammation caused by the tick-borne bacterium Candidatus Coxiella massiliensis was described (49).

The polymerase chain reaction was established in the 1980s and was subsequently used as the predominant detection method. Until then, ticks were primarily tested for the occurrence of C. burnetii and related pathogens with staining methods and animal infection experiments (50–52). Because C. burnetii is difficult to isolate and cultivate as an obligate intracellular organism, molecular diagnostic using PCR is the method of choice for its detection. Various protocols target several plasmids and chromosomal genes, for example, IS1111, icd, com1, sodB, and GroEL/htpB (53–55). The target gene IS1111 is a transposase-like insertion sequence, and the amount of copies from 7 to 110 per genome varies between isolates. Thus, the use for quantification is limited, but due to the higher number of targets per bacterium, this signal amplification leads to a higher sensitivity compared with single-copy targets (53). In contrast, icd, encoding for the isocitrate dehydrogenase gene, and com1, encoding for the C. burnetii outer membrane protein 1, are single-copy genes allowing quantification (53, 54). The commonly used target genes for C. burnetii cross-react with CLE and may lead to misidentification (28, 56). In a study by Duron, roughly one-third of CLE-positive ticks were positive for the IS1111 element (57). Furthermore, IS1111 was shown to be the most unspecific marker for detection of C. burnetii in CLE-positive ticks, followed by GroEL/htpAB (for chaperone heat shock protein), whereas icd was not amplified in the samples (56). However, the icd sequence of a novel CLE derived from Carios capensis soft ticks was >90% similar to C. burnetii (58). As a consequence, usage of a target gene of low specificity could lead to an overestimation of the role of ticks as host and vector of C. burnetii. To distinguish between C. burnetii and CLE, there is no specific method available, owing to the fact that the group of CLE is very heterogeneous. Therefore, the most reliable method is the sequencing of PCR-positive samples using highly conserved genes such as rrs or groEL (56, 59).

Ticks (Ixodida) are obligate blood-feeding ectoparasites with a global distribution. A total of 67 tick species are reported in Europe and Northern Africa (60). Out of these, only a small number, i.e., 15–17 hard tick species, depending on their confirmed taxonomic status as a distinct species or subspecies, have been found to harbor either C. burnetii or CLE (28, 61). When considering a particular tick species to serve as a vector for a given pathogen, certain aspects of the tick's life cycle should be considered in addition to the sole finding of molecular traces or even morphological structures within an examined tick. This information will help to understand further whether this tick species has vector competence, defined as the capability to transmit a given pathogen horizontally or vertically, which is an essential aspect of the vector capacity. The latter is more relevant for determining the infection risk, as it includes parameters such as longevity, feeding behavior (duration, frequency, and preferred blood source), population density, and frequency of host contact encounters. In this regard, ticks, which are feeding on different hosts in every life stage (three-host-ticks) with a large geographical range, high population densities, and promiscuous feeding behavior are relevant, when it comes to pathogen transmission. The most common hard tick of Europe, the castor bean tick Ixodes ricinus, is fulfilling these criteria, making it the most important vector tick species for bacterial and viral tick-borne diseases, e.g., Borrelia burgdorferi s.l. or tick-borne encephalitis virus, in Europe (62). Such vector-borne agents, some with a wide range of host species, may have their reservoir in wildlife, leading to a sylvatic transmission cycle between arthropods and wild animals. In addition, transmission can also occur in an urban or domestic cycle, including pets, livestock, or humans. Although I. ricinus is ubiquitous in Europe, other relevant tick species are not, and only for some geo-referenced distribution maps are available, e.g., for both Dermacentor species (63). However, the geographical distribution of a particular tick species is not static but rather subject to permanent habitat changes due to increased movement of humans, animals, and goods and, very importantly, the increasing ambient temperatures as major drivers. The high frequency of recent introductions of Hyalomma spp. ticks into temperate Europe is impressive sign of a very menacing trend of potent vector tick species moving into new areas (64, 65).

Recent prevalence studies using molecular methods aim at determining the percentage of ticks carrying C. burnetii among other diverse tick-borne pathogens. This allows surveillance of C. burnetii in ticks and benefits the identification of a potential risk of acquiring a Q fever infection by a tick-bite for humans or domestic animals.

Different methods exist for the sampling of ticks. As most European hard ticks seek hosts by questing on the vegetation and naturally intend to be wiped up by a passing host, the mainly used collection method is flagging or dragging. Using this method, a cotton blanket is dragged or flagged over the vegetation, considering that both methods favor different tick species (66). Ticks are stripped off and are hindered in their motion by the weave. The success of this method depends on the texture of the blanket and the structure of the vegetation (67). Activity peaks of most questing ticks are in spring and autumn. However, other studies observed a trend to a unimodal peak with the highest activity in the spring months (68, 69). Ticks can also be removed directly from their host. This method is used to examine the burden of pathogens associated with these animals and for the sampling of hunting ticks, such as Hyalomma spp., which are less likely to be collected by flagging (70).

This review gives a comprehensive overview of the present literature on C. burnetii in ticks and discusses the vector competence of European tick species for the Q fever agent. To this end, data of prevalence studies conducted in Europe concerning the occurrence of C. burnetii in ticks were systematically compiled and analyzed regarding the spatial occurrence, molecular detection method used, and the frequency of detection per tick species, as well as the detection of CLE. Furthermore, experimental approaches to investigate the vector competence of tick species for C. burnetii were collected. The determination of the vector competence of ticks is important to assess the role of ticks as a reservoir in Q fever epidemiology.

Studies about C. burnetii in ticks were classified into two different categories. One group investigates the prevalence of ticks for C. burnetii in Europe. The other group discusses the vector competence of ticks based on experimental approaches on ticks infected under laboratory conditions. For identification of studies, the databases Pubmed and Web of Science were systematically searched for articles in English, published between 2000 and 2020. Search terms were “coxiella tick,” “coxiella-like,” “coxiella vector competence.” The search was completed on October 27, 2020. Additional studies were included using Google Scholar. References were sorted by the use of Endnote X7 (Clarivate Analytics, Philadelphia, USA). Only studies using DNA detection methods for Coxiella spp. on hard ticks obtained in Europe were included as prevalence studies. The definition of Europe used in this study refers to a geographical demarcation as von Strahlenberg's, including all countries of which the major part is situated between the Urals and the Bosphorus. Regions were defined according to the United Nations geoscheme for Europe in Southern Europe, Eastern Europe, Northern Europe, and Western Europe (71).

Further criteria for eligibility as prevalence studies were stated information of the number of tested ticks, the number of Coxiella-positive ticks, or amount and description of tick pools and at a minimum description of the tick genus or species name.

Data were extracted regarding information about the country in which the ticks were obtained. Furthermore, the number of tested and Coxiella-positive ticks and tick pools and the collection and detection method were included in the analysis. All publications were read, and data were extracted by two persons independently.

The proportion of positive ticks was calculated as the ratio of positive ticks to the total number of tested ticks. Prevalence was defined as the proportion of positive ticks multiplied by 100. When ticks were tested in pools, the proportion was estimated as minimum infection rate (MIR) using the formula

Studies were evaluated, and confidence interval was calculated using Excel 2016 (Microsoft Corporation, Redmond, USA). For statistical analysis, the Kruskal–Wallis test and Mann–Whitney U test for data lacking normal distribution were performed using SPSS V.22 (IBM, Armonk, USA), and results were considered significant in the case of p < 0.05.

A total of 590 studies were found using the mentioned search terms. Additionally, two studies using the platform Google Scholar were included. Based on the title and abstract, 502 studies were excluded. Main reasons for exclusion were non-European studies (n = 272), no hard ticks tested (n = 120), or reviews (n = 44); 66 studies were excluded for other reasons (Figure 1: PRISMA analysis). After full-text analysis, 16 further studies were removed. Finally, 72 publications in which the prevalence of C. burnetii in ticks collected in European countries were examined, compiled, and evaluated. Additionally, two vector competence studies were included.

The evaluation of European studies shows a significantly higher prevalence of Coxiella spp. in ticks in Mediterranean countries. This is likely to be driven by the abundance of different tick species in these countries, i.e., Hyalomma spp. and Rhipicephalus spp. In the context of global warming, the geographical distribution of these tick species changes and thus the epidemiology of Q fever in currently temperate Europe. Based on the large number of studies, which failed to detect C. burnetii DNA in tick samples, ticks carrying C. burnetii seem to be restricted to certain regions. Positive results have to be interpreted carefully because no distinction between C. burnetii and CLE was made in most of these studies. Planning of a prevalence study on ticks should, therefore, in particular, focus on the choice of detection methods for specific results. Methods should aim to differentiate C. burnetii and CLE using sequencing or at least to use different target genes in positive samples.

To address the questions on the role ticks play as a reservoir in Q fever transmission, vector competence studies using different and relevant tick species should be performed.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

SK and KM-S designed the study, SK conducted the literature search, SK, GM, SU, MP, and KM-S performed the data extraction, and SK analyzed the data and drafted the manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.