Rickettsia spp. in Ticks of South Luangwa Valley, Eastern Province, Zambia

Ticks are important vectors for Rickettsia spp. belonging to the Spotted Fever Group responsible for causing Rickettsiosis worldwide. Rickettsioses pose an underestimated health risk to tourists and local inhabitants. There is evidence of the presence of Rickettsia spp. in Zambia, however there is limited data. A total of 1465 ticks were collected in 20 different locations from dogs and cattle including one cat. Ticks were identified by morphological features or by sequencing of the 16S mitochondrial rRNA gene. Individual ticks were further tested for rickettsiae using a pan-Rickettsia real-time-PCR. Rickettsia species in PCR-positive ticks were identified by sequencing the 23S-5S intergenic spacer region or partial ompA gene, respectively. Seven tick species belonging to three different tick genera were found, namely: Amblyomma variegatum, Rhipicephalus appendiculatus, Rhipicephalus (Boophilus) microplus, Rhipicephalus simus, Rhipicephalus sanguineus, Rhipicephalus zambesiensis and Haemaphysalis elliptica. Out of the 1465 ticks collected, 67 (4.6%) tested positive in the pan-Rickettsia PCR. This study provides detailed data about the presence of Rickettsia species in South Luangwa Valley, Eastern Province, Zambia for the first time. High prevalence of Rickettsia africae in Amblyomma variegatum was found, which indicates the potential risk of infection in the investigated area. Furthermore, to our best knowledge, this is the first time Rickettsia massiliae, a human pathogen causing spotted fever, has been detected in Zambia.


Introduction
Rickettsioses of the Spotted Fever Group (SFG) are common in many sub-Saharan African countries [1][2][3]. Several human pathogenic Rickettsia spp. are known to occur in Zambia's neighboring states [1,3,4]. Although known vectors of rickettsiae such as Amblyomma (A.) variegatum, Rhipicephalus (Boophilus) microplus, Rhipicephalus (Rh.) simus, Rhipicephalus sanguineus, Rhipicephalus zambesiensis and Rhipicephalus appendiculatus occur in Zambia [3,[5][6][7][8], the medical importance of rickettsioses in Zambia is unclear and epidemiological data is scarce. So far, only one study has reported IgG antibodies against SFG rickettsiae in Zambia's inhabitants, thus indicating the occurrence of rickettsial infections in humans [9]. The highest prevalence of SFG Rickettsia antibodies was found in Chipata district in Eastern Province of Zambia [9]. In addition, the occurrence of anti-rickettsial antibodies in baboons and vervet monkeys in South and North Luangwa National Park Microorganisms 2023, 11, 167 2 of 11 confirmed this observation [10], while more detailed studies on the detection and identification of rickettsial species in Zambia are not available. Another study described Rickettsia (R.) africae in Amblyomma sparsum on tortoises imported to Japan from Zambian tortoise populations [11], as the only direct confirmation of the occurrence of this human pathogen in Zambia.
The aims of this study were to show the prevalence of SFG rickettsiae in ticks and to ascertain which species of SFG rickettsiae occur in South Luangwa Valley. As many SFG rickettsiae are human pathogens, these data are of great importance for travel medicine In addition, the data will add value to the knowledge of local general practitioners concerning the possible causes of febrile diseases in the local population [12].

Materials and Methods
Ethical clearance to conduct the study was obtained from ERES-CONVERGE IRB in Zambia (Ref. No. 2022-July-001).

Sampled Area and Tick Collection
Ticks were collected in the chiefdoms Jumbe and Kakumbi in the Mambwe district, South Luangwa Valley, of Zambia's Eastern Province. Tick sample collection and laboratory analysis were performed between 19 July 2022 and 16 September 2022 ( Figure 1). Collection sites were located between 500 and 800 m above sea level and spread over an area of approximately 2093 km 2 . Ticks were collected from watchdogs, cattle and one pet cat. The ticks were stored at 4 • C, transported to Germany alive and stored at −80 • C until identification and processing. The map with the collection sites was designed in ArcGIS 10.2.2 (Copyright © 1995-2014 Esri) using open source data of ©OpenStreetMap-Contributors (license CC BY-SA, [13]).
Microorganisms 2023, 11, x FOR PEER REVIEW 2 of antibodies in baboons and vervet monkeys in South and North Luangwa National Par confirmed this observation [10], while more detailed studies on the detection and ident fication of rickettsial species in Zambia are not available. Another study described Ric ettsia (R.) africae in Amblyomma sparsum on tortoises imported to Japan from Zambian to toise populations [11], as the only direct confirmation of the occurrence of this huma pathogen in Zambia. The aims of this study were to show the prevalence of SFG rickettsiae in ticks and t ascertain which species of SFG rickettsiae occur in South Luangwa Valley. As many SF rickettsiae are human pathogens, these data are of great importance for travel medicin In addition, the data will add value to the knowledge of local general practitioners con cerning the possible causes of febrile diseases in the local population [12].

Materials and Methods
Ethical clearance to conduct the study was obtained from ERES-CONVERGE IRB i Zambia (Ref. No. 2022-July-001).

Sampled Area and Tick Collection
Ticks were collected in the chiefdoms Jumbe and Kakumbi in the Mambwe distric South Luangwa Valley, of Zambia's Eastern Province. Tick sample collection and labora tory analysis were performed between 19 July 2022 and 16 September 2022 ( Figure 1 Collection sites were located between 500 and 800 m above sea level and spread over a area of approximately 2093 km 2 . Ticks were collected from watchdogs, cattle and one p cat. The ticks were stored at 4 °C, transported to Germany alive and stored at −80 °C unt identification and processing. The map with the collection sites was designed in ArcGI 10.2.2 (Copyright © 1995-2014 Esri) using open source data of ©OpenStreetMap-Contrib utors (license CC BY-SA, [13]).

Collection Sites
Ticks were collected from dogs and cattle of small farms located individually or in little villages in the chiefdom Jumbe's wards Mphomwa, Chikowa and Jumbe ( Figure 1). These farms kept pigeons, chicken and grow crops. In the eastern part of Mphomwa ward, farms were breeding cattle (Location 3, Figure 1). Farmland and natural vegetation surrounded the study site. The cattle were grazing free-ranged during day and were fenced at night.
In addition, some samples were collected from a cat and two dogs in homes with well-tended gardens at the riverside of the Luangwa in Kakumbi chiefdom in Kakumbi ward (Table 2, Figure 1). Largely, the original vegetation around the houses and gardens remained unchanged. Due to the direct proximity of the South Luangwa National Park across the river there was no land or livestock farming in this region. Different antelope species and other wildlife species were observed regularly in the immediate vicinity of the houses.

Identification of Tick Species
The ticks were identified using morphological characters according to Walker et al. [14,15] and Apanaskevich et al. [16]. For ticks, which could not be unambiguously identified by morphology, molecular identification was performed using a polymerase chain reaction (PCR) protocol targeting the mitochondrial 16S rRNA gene as described by Halos et al. [17].

Tick Preparation and Nucleic Acid Extraction
After identification, 1051 ticks were homogenized individually, 414 ticks in pools up to 7 individuals in 800 µL minimal essential medium (MEM, Life Technologies), using Lysing matrix tubes A (MP Biomedicals) in a FastPrep Instrument BIO101 (MP Biomedicals). Total nucleic acid was extracted from 200 µL homogenate using the MagNA Pure LC RNA/DNA Kit (Roche, Mannheim, Germany) in a MagNA Pure LC instrument (Roche), according to the manufacturer's instructions. The total nucleic acid was stored at −80 • C until use.

Detection of Rickettsia spp. in Tick Samples
The ticks were tested for Rickettsia spp. using a pan-Rickettsia real-time-PCR, targeting the citrate synthase gene [18]. All specimens with a CT-value above zero were rated as positive and processed in molecular identification. The molecular identification to species level was achieved by analyzing the ompA fragment IV and the 23S-5S intergenic spacer region according to previously published protocols [19,20]. (Table 1).

Sequence Analysis and Phylogenetic Classification
Obtained PCR products were purified using the QIAquick ® PCR Purification Kit 205 according to the manufacturer's instructions. For identification, the amplicons were sequenced using Sanger technique by an external contractor (GATC Biotech, Konstanz, Germany). Sequences were analyzed using BioEdit Alignment Editor (Version 7.2.5.0) [21] and compared with the gene bank of NCBI (National Centre for Biotechnology Information) [22] by BLAST (Basic Local Alignment Search Tool) [23]. Phylogenetic analysis was performed in MEGA 7 [24]. For multiple sequence alignment Clustal W in MEGA 7 was used [25]. Molecular phylogenetic analysis was performed by the Maximum Likelihood method using the Tamura-Nei model [26] and bootstrap method (1000 replications). A maximum likelihood phylogenetic tree was created for the partial gene sequences of 23S-5S rRNA and ompA fragment IV. Identical clusters and similar confidence values were obtained by the Maximum Parsimony method and the Neighbour-joining method.

Data Analysis
To perform data analysis, chi 2 -test and adjusted Pearson's coefficient of contingency were conducted in Microsoft ® Excel 2016. For all chi 2 -test analysis a p-value of < 0.05 was defined as statistically significant. For Pearson's coefficient of contingency values between 0 to 0.2 were seen as week association, between 0.2 to 0.6 average and higher than 0.6 as high association between the tested characteristics.

Results
A total of 1465 ticks were collected at 20 different locations spread over 4 wards and from different hosts: 1438 from dogs (98.2%), 24 ticks from cattle (1.6%), and three ticks from one cat (0.2%). The ticks belonged to three genera and seven species. Percentage of species distribution, prevalence of Rickettsia spp. and infection rates are listed in Table 2. In the wards Chikowa and Jumbe, only dogs were sampled. In Chikowa, none of the 40 collected ticks (8 Rh. appendiculatus, 32 Rh. sanguineus "tropical lineage" [27]) were positive in pan-Rickettsia real-time-PCR. In Jumbe, one of 764 (0.1%) specimens was positive in pan-Rickettsia PCR and was identified as R. conorii ssp. caspia [28]. In detail 5 Rh. zambesiensis, 7 Rhipicephalus simus, 26 H. elliptica, 99 Rh. appendiculatus, 626 Rh. sanguineus "tropical lineage" were negative for Rickettsia spp. and one Rh. sanguineus "tropical lineage" was positive (cf. Table 3). Table 3. List of specimens with collection site, host, tick species, stage and the result of pan-Rickettsia PCR.

Chiefdom, Ward
Host    (Table 3) In Kakumbi, neither the two A. variegatum and the one Rh. sanguineus "tropical lineage" collected from a cat nor the two females (Rh. sanguineus "tropical lineage" and Rh. appendiculatus) from two dogs were positive for Rickettsia spp. (Table 3).
From the total 1255 collected Rh. sanguineus "tropical lineage" specimens, for R. massiliae and R. conorii ssp. caspia [28] a prevalence of 0.08%, respectively, was calculated. The prevalence of R. africae in Rh. sanguineus "tropical lineage" was found to be 0.24% with a median CT-value of 35, 28 in the pan-Rickettsia real-time-PCR. From 30 A. variegatum specimens, 24 A. variegatum (80%) were found positive with a median CT-value of 26,71 for R. africae (Tables 2 and 3).

Phylogenetic Analysis
The majority of detected R. africae showed 100% identity in both analyzed genomic targets to the type strain gb|CP001612.1 R. africae ESF-5. In the ompA target the phylogenetic analysis revealed one additional, well-supported clade of slightly different R. africae variants with identities from 99.4-99.9% to the type strain R. africae ESF-5 (Figure 2A,B). The phylogenetic investigation of the hypervariable 23S-5S intergenic spacer region showed a stronger variability, resulting in three additional well supported clades with identities from 98.3-99.7% to the type strain R. africae ESF-5 (Figure 2A,B). The repeated analysis of the 23S-5S intergenic spacer region of sample 3, at location 2 (Figure 1), revealed no known sequence of a Rickettsia species but the sequence analysis of partial ompA revealed a 100% identity to R. africae ESF-5. The R. massiliae sequence of sample 1172 was 100% identical to gb|CP003319.1 R. massiliae str. AZT80 in both investigated genomic targets, respectively. Sample 582 showed the closest phylogenetic relationship to gb|AH011786.2 R. conorii ssp. caspia Chad (partial ompA 100%) and gb|U83437.1 R. conorii ssp. caspia A-167 (23S-5S intergenic spacer 100%) [28].

Statistical Analysis
The statistical analysis by chi 2 -test and adjusted Pearson's coefficient of contingency (P) revealed that the spread of tick species was strongly dependent on the host (Table 4). Additionally, the positive or negative CT-value was strongly dependent on the tick species and the host of the tick. There was a low to middle association between the positive or negative CT-value and the ward where the specimens were collected. There was also a low to middle association between ward and the tick species.

Statistical Analysis
The statistical analysis by chi 2 -test and adjusted Pearson's coefficient of contingency (P) revealed that the spread of tick species was strongly dependent on the host (Table 4). Additionally, the positive or negative CT-value was strongly dependent on the tick species and the host of the tick. There was a low to middle association between the positive or negative CT-value and the ward where the specimens were collected. There was also a low to middle association between ward and the tick species.

Discussion
Although R. africae is probably the most common rickettsial species in Africa, surprisingly only few studies are available to test genetic differences in isolates or strains of R. africae from different locations. In our study, the analysis of the partial ompA gene showed two and the 23S-5S intergenic spacer region revealed four different genetic clusters, which occurred in one single location named Mphomwa 3 (cf. Figures 1 and 2). The 23S-5S intergenic spacer region is more suitable to detect minor genetic changes due to its hypervariability. However, the reliability of these changes is lower because mutations in non-coding regions producing distinct changes will usually not cause any phenotypic changes in the bacterial organism and therefore might accumulate randomly. It has to be studied in more detail whether the observed genetic differences result in phenotypic differences, e.g., regarding pathogenicity and if they provide more insight in the evolution and dispersal of R. africae on the African continent.
Rickettsia africae is one of the most widespread rickettsial species in Africa [2,29]. Therefore, it was expected to find a high prevalence of R. africae in A. variegatum ticks, as it has been observed in other African countries [3]. Three Rh. sanguineus "tropical lineage" ticks were positive for R. africae. In this tick species R. africae is rarely found and might be only a contaminant from rickettsiaemic blood of the host. This assumption was supported by the fact that the amount of rickettsial DNA in Rh. sanguineus "tropical lineage" was low in comparison to A. variegatum. The statistical analysis also showed a strong association between the positive CT-value and A. variegatum, respectively, the negative CT-value and Rh. sanguineus "tropical lineage". In line with our own data several other studies found that the prevalence of R. africae in ticks from the genus Rhipicephalus is always lower than in the genus Amblyomma [30,31].
In the present study, R. africae was reported only in the eastern part of Mphomwa ward (Locations 1, 2, 3 in Figure 1). A likely reason for this focal occurrence is that exclusively in this region, ticks collected from cattle were obtained, being represented solely by A. variegatum with a high prevalence for R. africae. The Pearson's coefficient of contingency showed a strong association between the host and the tick species which is in line with literature saying A. variegatum is mostly found on cattle and Rh. sanguineus mostly on dogs. Usually, such a focal distribution of R. africae is unusual as this species is generally found endemically over large areas in Africa. High prevalence rates of R. africae in Amblyomma ticks are seen in pasture settings in different countries in Africa [3]. The host animal at the location is a basic characteristic of the location as it is influenced by man and not by a natural distribution. Therefore, it is not possible to test an association by Pearson for host and location. As Maina et al. suggested, cattle are not the reservoir host of R. africae [32]. However, in this study a strong association between cattle, A. variegatum and R. africae was found. If looking at the data of ticks from dogs only (Table 4), there is a low to middle association between the ward and a positive or a negative specimen. Furthermore, there is a low to middle association of the place of collection and the tick species. Accordingly, there have to be further important environmental factors affecting tick abundance and by that the positivity for rickettsiae aside from the host. It is assumed that small mammals, large antelopes, and ungulates may serve as vertebrate hosts. This relation concurs with knowledge in travel medicine that certain behavioral factors, such as game hunting and excessive contact to ground vegetation as well as environmental factors and travelling during rainy season and high humidity, increase the risk of an infection with R. africae in tourists [33]. Additionally, the natural transmission cycle of R. africae has not been elucidated, so far. Moreover, it is known that transovarial transmission and transstadial persistence may play a major role for the maintenance of rickettsiae in the vectors while the role of vertebrates is still not completely understood [34]. Therefore, our results may implicate that one or more essential factors for the natural transmission cycle of R. africae are present in the eastern region of the study area but seemingly not in other regions. However, these essential factors have yet to be identified.
Rickettsia africae is the etiologic agent of African tick-bite fever. This is usually a benign febrile disease without any complications or sequelae. Human infection might cause malaria-like symptoms, and therefore, could be under-or misdiagnosed.
In this study, R. massiliae was detected for the first time in ticks in Zambia and also in its Eastern Province. This rickettsial species is known to occur in Africa [1,3,30]. A study by Cicuttin et al. suggested the prevalence of R. massiliae in the genus Rhipicephalus to be 5-6% in North Africa [35]. Although it has been detected mainly in northern and eastern Africa, so far, R. massiliae was also recently described in South Africa [36] and in Botswana [31]. Therefore, the detection in Zambia in the tick species Rh. sanguineus "tropical lineage" is in line with published data [1,3,4,35,37]. Our data indicate a very low prevalence of R. massiliae in Zambia as it could be detected only in one Rh. sanguineus "tropical lineage" in the ward Mphomwa (Location 14 in Figure 1). Recently, it was shown that at least three genetic forms of Rh. sanguineus s. l. occur in different parts of the world, which exhibit different patterns of vector competence [27,38]. It is assumed that R. massiliae has its main vertebrate host in dogs and canids [35]. The low prevalence rate in Rh. sanguineus "tropical lineage" might indicate, that the circulating genetic form of Rh. sanguineus "tropical lineage" might be an incompetent vector for the local R. massiliae strain and therefore occur only in a very low frequency in this vector. It is known that R. massiliae is not common in rural areas [4]. Therefore, it has to be studied whether it can be found more regularly in urbanized areas in Zambia. Rickettsia massiliae was reported to cause human diseases comparable with Mediterranean spotted fever [39]. In Argentina, it was the agent of a severe form of spotted fever [37]. So far, no clinical forms of human R. massiliae infections have been described in Africa, but it may be the cause of undifferentiated febrile diseases in the region that, due to the lack of knowledge, remain unrecognized.
Finally, R. conorii ssp. caspia was detected in a single sample [28]. So far, only one other study has reported the occurrence of this Rickettsia ssp. much further north in Chad [40]. As R. conorii ssp. caspia has been described as causative agent of Astrakhan fever with the possibility of hemorrhagic manifestation in Russia [41] it must be taken into consideration as a possible cause of hemorrhagic fever manifestations in Zambia and the whole southern African region.

Conclusions
We studied a tick population collected from dogs, cattle and one cat in Zambia. In total, seven different tick species were detected. In A. variegatum and Rh. sanguineus "tropical lineage" three pathogenic rickettsiae, namely R. africae, R. massiliae and R. conorii ssp. caspia were detected. The prevalence of R. africae was localized and occurred only in an agricultural area with cattle breeding. The impact of the observed genetic variability of R. africae on pathogenic features has to be clarified in future studies. A second rickettsial species, R. massiliae, was found in one Rh. sanguineus "tropical lineage" for the first time in Zambia.