Ticks are vectors of important pathogens of humans and animals and serve as indicators of infection in nature . The geographical distribution and habitats of several generalist tick species have expanded in the recent years. Major drivers for this trend include land use, climate changes and globalization [31, 32]. On the other hand, for certain tick species which are co-distributed with their endangered hosts, like the case of H. aegyptium, the trend is a decreasing geographical range . However, in general, a decrease in the availability of natural host populations could lead to host-switching behaviour . As H. aegyptium is reported to alternatively feed on various other hosts, mainly during their pre-imaginal stages, evaluation of its zoonotic pathogen burden is of particular interest.
Regarding their role in the ecology of zoonotic infectious diseases, tortoises and their ticks have received significantly less attention compared to mammals and birds. Among small mammals, hedgehogs (Erinaceus spp.) are important mainly in synanthropic environments as reservoir hosts for important human pathogens like A. phagocytophilum, Babesia spp.  or B. burgdorferi s.l. . As H. aegyptium occasionally feeds on hedgehogs and it can potentially attack humans , the evaluation of this species as a carrier host for zoonotic pathogens is important.
For B. burgdorferi s.l., the main vectors are ticks of genus Ixodes and the reservoir hosts, mostly small mammals . Although in this survey there were no H. aegyptium positive for the Lyme disease agent, some other studies reported that this tick can feed on reservoir hosts of Borrelia lusitaniae. The role of reptiles in the ecology of B. lusitaniae was shown in the past by several authors [38, 39]. Borrelia burgdorferi s.l. is one of the most extensively studied tick-borne pathogens in the world. Hence, there were numerous experimental trials for assessing the vectorial capacity of various ticks. So far, experimental data suggest that only ticks of genus Ixodes are competent vectors for the Lyme disease spirochetes . However, Kar et al.  found B. burgdorferi s.l. in two out of 28 pools of H. aegyptium collected from Testudo graeca in Turkey. In the present study no ticks were positive for this pathogen, supporting the hypothesis that ticks other than Ixodes spp. should not be assumed to serve as bridging vectors for B. burgdorferi s.l. or to play any role in the maintenance of these spirochetes in natural cycles .
Anaplasma phagocytophilum, the agent of human granulocytic anaplasmosis, is vectored in Europe by Ixodes ricinus and can infect a wide range of domestic and wild vertebrate hosts, including rodents, horses, dogs and humans . This is the first report of A. phagocytophilum in H. aegyptium ticks. The relatively high prevalence of A. phagocytophilum in this study (18.8%) in fully engorged H. aegyptium collected from tortoises can be theoretically caused by two factors: (1) the infection was acquired by ticks during an earlier developmental stage feeding on competent reservoir hosts or (2) tortoises are competent reservoir hosts. The second hypothesis is less probable, as several studies showed that reptiles are not competent reservoir hosts for this pathogen [42, 43]. Moreover, surveillance of other ticks parasitic on reptiles yielded negative results . Prevalence of Anaplasma spp. in engorged Hyalomma lusitanicum and H. marginatum collected on domestic mammals in Sicily was much lower (~1%) . All these data (improbable reservoir role of tortoises and relatively high prevalence), suggest that H. aegyptium is able to transstadially pass in A. phagocytophilum. However, the probable lack of transovarial transmission of A. phagocytophilum in ticks  confers little eco-epidemiological importance to this vector-pathogen association.
The genus Ehrlichia includes five species , but only E. canis is found in Europe. This study reports the presence of E. canis for the first time in H. aegyptium, with a noteworthy prevalence. The only recognized vector for E. canis is Rhipicephalus sanguineus. In Romania, the distribution range of H. aegyptium overlaps with the distribution of R. sanguineus. Although R. sanguineus feeds mainly on dogs , in Romania, it has been found also on hedgehogs (E. roumanicus) . This is indicative of a possible cross transmission of E. canis from R. sanguineus to H. aegyptium using hedgehogs as bridging hosts.
The Q fever agent, C. burnetii was present only in Măcin Mountains with a relatively high prevalence in ticks (37.9%). An interesting aspect in this area was that all tortoises harbouring C. burnetii-infected ticks had also A. phagocytophilum-infected ticks and 72.7% of these tortoises had ticks infected with E. canis, too. Additionally, out of the 11 tortoises infested with C. burnetii-infected ticks, 10 (90.9%) had co-infected ticks. Q fever affects a wide range of domestic and free living mammals, birds, reptiles, fish, and arthropods, as well as humans [49, 50]. The etiological agent of Q fever, C. burnetii, has been identified in over 40 tick species [50, 51]. Hyalomma aegyptium was shown to have an unquestionable potential in the epidemiology of Q fever natural foci . Ticks transmit C. burnetii vertically (transstadially and transovarially) and horizontally (by biting, via saliva)  but also through their faeces . Sharing of pastures by tortoises and domestic ruminants was suggested by Široký et al.  to have an important role in the natural cycle, especially if considering that tortoises maintain natural foci of Q fever by hosting long-term infected ticks .
We found a marked dissimilarity between the occurrences of individual pathogens in ticks among the different locations: Anaplasma phagocytophilum and E. canis being found in all three locations while C. burnetii was only present in one. This variance may be caused by the different habitat associations and its host-fauna composition. The two forested sample sites have high occurrence rates of small mammals, especially hedgehogs, while the third location is primarily used by small domestic ruminants . Coxiella burnetii is commonly reported in sheep and goats , hence its occurrence is more likely in the later habitat. This is consistent with the present findings. Moreover, the local agricultural practice (i.e. high turnover rate of domestic herds on extensive used pastures) in this region provides chances for a continuous presence of this pathogen in the environment .