Isolation of Rickettsia rhipicephali and Rickettsia bellii from Haemaphysalis juxtakochi Ticks in the State of São Paulo, Brazil

Abstract

In the present study, attempts to isolate Rickettsia in cell culture were performed individually in seven specimens of Haemaphysalis juxtakochi ticks collected in the state of São Paulo (southeastern Brazil). Rickettsia was successfully isolated by the shell vial technique and established in Vero cell culture from six ticks (six isolates). DNA extracted from infected cells of these isolates was tested by PCR and DNA sequencing, using genus-specific Rickettsia primers targeting the genes gltA, htrA, ompA, and ompB. After the generated sequences were compared with available sequences in GenBank, five out of the six isolates were identified as Rickettsia bellii (isolates HJ#1, HJ#2, HJ#3, HJ#4, and HJ#7). The sixth isolate (HJ#5) was closest to Rickettsia sp. strain R300, previously detected in H. juxtakochi in northern Brazil, and to Rickettsia rhipicephali, isolated from ticks in the United States. Following recent gene sequence-based criteria proposed for the identification of Rickettsia isolates, both isolate HJ#5 and strain R300 were identified as South American strains of R. rhipicephali, which was confirmed in this continent for the first time. Isolation of R. bellii from H. juxtakochi ticks, added to eight other tick species that have been reported to be infected with this bacterium in Brazil, indicates that R. bellii is indeed the most frequent Rickettsia species infecting ticks in Brazil. Currently, the role of both R. rhipicephali and R. bellii as human pathogens is regarded as unknown.

Rickettsia spp. are intracellular bacteria in obligatory association with eukaryotic cells. They are associated primarily with invertebrates (ticks, mites, insects, and leeches), but some species are also capable of infecting and causing disease in humans, animals, or plants (9, 21, 30). Rickettsia spp. have been classically divided into two groups based on their antigenic, morphological, and ecologic patterns: the typhus group, associated primarily with lice and fleas, and the spotted fever group, associated mostly with ticks (45). The use of molecular tools during the last 15 years has led to the discovery of different Rickettsia species and genotypes, suggesting a revision of the classical classification of Rickettsia spp. into groups (37, 38, 40, 41, 42).

In Latin America, several Rickettsia species (belonging mostly to the spotted fever group) have been reported as infecting ticks. Rickettsia rickettsii, the etiological agent of the most severe spotted fever in the world, has been reported in Amblyomma cajennense ticks in Brazil, Colombia, Panama, and Mexico (6, 10, 15, 31), Amblyomma aureolatum ticks in Brazil (35), and Rhipicephalus sanguineus ticks in Mexico (7). More recently, Rickettsia parkeri, the etiological agent of a rickettsiosis with skin lesions and lymphadenitis, was reported in Amblyomma triste ticks in Uruguay (44); an closely related to R. parkeri (strain COOPERI) was isolated from Amblyomma dubitatum (Amblyomma cooperi) in Brazil (22). Rickettsia massiliae (a recognized human pathogen in Europe) was detected in R. sanguineus ticks from Argentina (8). Rickettsia bellii (a species of unknown pathogenicity to humans) was reported as infecting A. aureolatum, A. dubitatum, Amblyomma humerale, Amblyomma rotundatum, Amblyomma oblongoguttatum, Amblyomma scalpturatum, Amblyomma ovale, and Ixodes loricatus in Brazil (19, 22, 23, 35) and Amblyomma neumanni in Argentina (27). Rickettsia amblyommii (also of unknown pathogenicity) was reported in A. cajennense and Amblyomma coelebs in Brazil (23) and A. neumanni in Argentina (27). “Candidatus Rickettsia andeanae” was reported in Amblyomma maculatum and Ixodes boliviensis ticks in Peru (3). In addition, Rickettsia prowazekii (the agent of epidemic typhus) was reported in A. cajennense or Amblyomma imitator ticks in Mexico (28). Interestingly, all of these Rickettsia species (except for “Candidatus Rickettsia andeanae”) have been isolated in the United States, from the same tick species and/or from tick species different from those in Latin America (12, 23, 28, 34, 45).

A recent molecular study reported a Rickettsia sp. strain (named strain R300) in a single Haemaphysalis juxtakochi tick specimen collected in the Amazon forest in the state of Rondônia, northern Brazil (25). Phylogenetic analyses inferred by the rickettsial genes gltA (citrate synthase), htrA (17-kDa protein), ompA (190-kDa protein), and ompB (135-kDa protein) showed that strain R300 was closest to Rickettsia rhipicephali from North America. These results suggested that strain R300 was a South American strain of R. rhipicephali (25). In the present study, we tested a sample of H. juxtakochi ticks collected in the state of São Paulo (southeastern Brazil), attempting to isolate rickettsiae in cell culture, especially a possible strain of R. rhipicephali.

MATERIALS AND METHODS

On 12 March and 9 December 2005, free-living H. juxtakochi ticks were collected on the vegetation in an Atlantic rain forest area in Intervales State Park, Ribeirão Grande Municipality, state of São Paulo (24°18′S, 48°24′W). In addition, on 10 January 2005, H. juxtakochi ticks were collected while feeding on a brocket deer (Mazama gouazoubira) captured in Jaraguá Park, São Paulo Municipality, SP (23°32′S, 46°38′W).

Ticks were brought alive to the laboratory, where they were frozen at −80°C until processed by the shell vial technique for isolation of rickettsiae in cell culture, as previously described (20) with some modifications (22). Briefly, ticks were individually thawed and triturated in brain heart infusion broth, which was inoculated into shell vials previously seeded with approximately 200,000 Vero cells. After being centrifuged for 1 hour at 700 × g at 22°C, the monolayer was incubated at 28°C with minimal essential medium containing 5% bovine calf serum and antibiotics (1% penicillin and streptomycin). After 3 days, the medium was switched to antibiotic-free medium and the aspirated medium was checked by Gimenez staining for the presence of Rickettsia-like organisms (14). If the result was positive, the monolayer of the shell vial was harvested and inoculated into a 25-cm² flask containing a monolayer of confluent uninfected Vero cells. The cells of the 25-cm² flask were checked by Gimenez staining until more than 90% of the cells were infected, after which they were harvested and inoculated into 150-cm² flasks of Vero cells. A Rickettsia isolate was considered established in the laboratory after three passages through 150-cm² flasks, each reaching more than 90% of infected cells. Part of the infected cells from one of these passages was harvested and used for the production of Rickettsia antigen slides, which were used in immunofluorescence assays (IFA) as previously described (17). For this purpose, a human serum previously shown to be reactive to R. rickettsii antigen (endpoint titer, 4,096) was used at a 1:64 dilution with a fluorescein isothiocyanate-labeled anti-human immunoglobulin G secondary antibody (Biomanguinhos, Rio de Janeiro, Brazil).

Cell passages of isolates were genotypically identified by PCR amplification and sequencing of the product of the remnants of the original infected tick and the resultant infected cells. For this purpose, by boiling (100°C for 20 min) as previously described (18), DNA was extracted from the remnants of each original individual tick and from infected cell passages, which were tested by PCR amplification using primers CS-78 (forward) and CS-323 (reverse), which amplify a 401-bp fragment of the gltA gene of all known Rickettsia species (22). DNA of infected cells yielding an expected PCR product by this protocol was further tested by other PCR protocols in order to perform a genetic characterization of the Rickettsia isolate. These protocols adopted primers CS-239 and CS-1069, which amplify an 834-bp fragment of the gltA gene (24), primers 17k-5 and 17k-3, which amplify a 549-bp fragment of the htrA gene (24), primers Rr190.70p and Rr190.602n, which amplify a 530-bp fragment of the rickettsial ompA gene (36), and primers 120-M59 and 120-807, which amplify an 865-bp fragment of the rickettsial ompB gene (38). For each set of reactions, a negative control (5 μl of water) and positive control (5 μl of DNA extracted from Amblyomma cajennense ticks experimentally infected with R. parkeri [39]) were included.

All PCR products of the expected size obtained in the present study were purified using ExoSap (USB) and sequenced in an automatic sequencer (ABI Prism 310 genetic analyzer; Applied Biosystems/Perkin-Elmer, CA) according to the manufacturer's protocol. The partial sequences obtained were submitted for BLAST analysis (1) to determine similarities to other Rickettsia species.

Nucleotide sequence accession numbers.

The GenBank nucleotide sequence accession numbers for the sequences generated in this study are DQ865204 and DQ865205 for HJ#7 partial sequences of the gltA and htrA genes, respectively, and DQ865206, DQ865207, DQ865208, and DQ865209 for HJ#5 partial sequences of the gltA, htrA, ompA, and ompB genes, respectively.

RESULTS

A total of five adult free-living ticks were collected on the vegetation in Ribeirão Grande, and two adult ticks were collected from a brocket deer in São Paulo. After each of the seven ticks was individually processed by the shell vial technique, rickettsial DNA was successfully amplified (primers CS-78 and CS-323) from six individual tick remnants (only one tick from Ribeirão Grande was PCR negative). The six PCR-positive tick DNA samples yielded expected PCR products, which, after sequenced, showed the following results: two São Paulo ticks (HJ#1 and HJ#2) and three Ribeirão Grande ticks (HJ#3, HJ#4, and HJ#7) yielded 350-bp gltA DNA sequences identical to each other and 100% (350/350) identical to a corresponding sequence of R. bellii in GenBank (U59716); one Ribeirão Grande tick (HJ#5) yielded a 350-bp sequence 100% (350/350) identical to the corresponding sequence of Rickettsia sp. strain R300 in GenBank (AY472038).

No Rickettsia species were successfully isolated in Vero cell culture from the single PCR-negative tick. Nevertheless, Rickettsia was successfully isolated and established in cell culture from all six PCR-positive ticks. DNA of infected cells of five isolates (HJ#1, HJ#2, HJ#3, HJ#4, and HJ#7) yielded expected PCR products when tested by both gltA primer pairs, but no product was visualized when ompA or ompB primers were used. DNA sequences obtained from both gltA primer pairs on each isolate were aligned, and the resultant sequences (1,080 bp from each isolate) were identical to each other and 100% (1,080/1,080) identical to the corresponding sequence of R. bellii (strain 369-C from the United States) in GenBank (U59716). In addition, a 491-bp fragment of the htrA gene of isolate HJ#7 was obtained, being 99.8% (490/491) identical to the corresponding sequence of the Brazilian isolate Ac25 of R. bellii in GenBank (AY362702).

DNA of infected cells of isolate HJ#5 yielded expected PCR products when tested by all primer pairs. Fragments of 1,092, 490, 401, and 790 bp of the gltA, htrA, ompA, and ompB genes, respectively, were obtained. Their highest identities with the corresponding sequences in GenBank are shown in Table 1. The gltA, htrA, ompA, and ompB fragments were 99.9, 100, 100, and 99.7% identical, respectively, to the corresponding sequences of the Rickettsia sp. strain R300, previously reported in H. juxtakochi in Brazil.

TABLE 1.

Closest relative sequences to gene fragments of a Rickettsia sp. strain isolated from the H. juxtakochi tick from Ribeirão Grande, state of São Paulo, Brazil^a

Gene and GenBank accession no.	Species (strain)	% Similarity (bp)
gltA
AY472038	Rickettsia sp. (R300)	99.9 (1,091 of 1,092)
U59721	R. rhipicephali (3-7-6)	99.8 (1,090 of 1,092)
U59720	R. massiliae (Bar29)	99.5 (1,087 of 1,092)
htrA
AY472039	Rickettsia sp. (R300)	100 (490 of 490)
U11020	R. rhipicephali (3-7-6)	99.0 (484 of 489)b
AY360215	Rickettsia sp. (Aranha)	98.4 (482 of 490)
ompA
AY472040	Rickettsia sp. (R300)	100 (491 of 491)
U43803	R. rhipicephali (3-7-6)	99.0 (486 of 491)
U43792	R. massiliae (Bar29)	98.6 (484 of 491)
ompB
AY472041	Rickettsia sp. (strain R300)	99.7 (788 of 790)
AF123719	R. rhipicephali (3-7-6)	99.2 (784 of 790)
AF123714	R. massiliae (Mtu1)	98.9 (781 of 790)

^aThe gene fragments included a 1,092-bp fragment of the gltA gene, a 490-bp fragment of the htrA gene, a 491-bp fragment of the ompA gene, and a 790-bp fragment of the ompB gene of isolate HJ#5.

^bThe available sequence was shorter than 490 bp.

Morphologically, both isolates HJ#5 and HJ#7 were characterized by typical coccobacillary organisms, as demonstrated by Gimenez staining and IFA in infected Vero cells (Fig. 1). The rickettsial isolates obtained in the present study have been deposited in the Rickettsial Collection of the Laboratory of Parasitic Diseases of the Faculty of Veterinary Medicine at the University of São Paulo and in the reference collection of the UTMB Rickettsial and Ehrlichial Diseases Research Laboratories, Galveston, TX, where they are available upon request.

FIG. 1.

Vero cells infected by rickettsiae as demonstrated by Gimenez staining (A and B) and by immunofluorescence assay using human anti-Rickettsia serum and anti-human immunoglobulin G secondary antibody (C and D). (A and C) Isolate HJ#7. (B and D) Isolate HJ#5. Photographs were taken using an Olympus optical microscope with a 10× ocular and a 100× objective (A and B) and an Olympus fluorescence microscope with a 10× ocular and a 40× objective (C and D).

DISCUSSION

Five out of six isolates of rickettsiae obtained in the present study showed gltA partial sequences that were 100% identical to the corresponding sequence of R. bellii (strain 369-C from the United States). One of these isolates (HJ#7) also showed a htrA partial sequence that was 99.8% identical to the isolate Ac25 of R. bellii, isolated from the A. dubitatum (formerly A. cooperi) tick from Brazil (22). In addition, these five rickettsial isolates from H. juxtakochi ticks possessed typical morphological features of Rickettsia (Fig. 1A and C) and demonstrated strong IFA reactivity with anti-R. rickettsii human serum. Serum cross-reactivity between R. rickettsii and R. bellii has been well documented in the literature (17, 34). Thus, these five isolates (HJ#1, HJ#2, HJ#3, HJ#4, and HJ#7) from H. juxtakochi ticks were determined to be R. bellii.

Isolate HJ#5 also possessed typical morphological features of Rickettsia (Fig. 1B) and strong IFA reactivity with anti-R. rickettsii human serum (Fig. 1D). Molecular analysis of isolate HJ#5 showed it to be closest to strain R300 (previously detected by PCR in an H. juxtakochi tick in Brazil) and to R. rhipicephali (previously reported in R. sanguineus, Dermacentor andersoni, Dermacentor occidentalis, and Dermacentor variabilis in the United States [4, 5, 32, 33]). A recent report (13) of gene sequence-based criteria for the identification of new Rickettsia species proposed that in order to be classified as a new Rickettsia species, an isolate should not exhibit more than one of the following degrees of nucleotide similarity (cutoff values) with the most homologous validated species: ≥99.8 and ≥99.9% for the rrs (16S rRNA) and gltA genes, respectively, and, when amplifiable, ≥98.8, ≥99.2, and ≥99.3% for the ompA, ompB, and sca4 (“gene D”) genes, respectively. As both ompA and ompB partial sequences of HJ#5 showed degrees of similarities with R. rhipicephali (Table 1) equal to or above these cutoff values, HJ#5 should be considered a strain of R. rhipicephali. Similarly, as the gltA, ompA, and ompB partial sequences of strain R300 previously reported (25) also showed degrees of similarities with HJ#5 (Table 1) equal to or above these cutoff values, strain R300 should also be considered a strain of R. rhipicephali. Thus, the present study reports the first isolation of R. rhipicephali from outside the United States, confirming the presence of this organism in two Brazilian states: strain HJ#5 in São Paulo (present study) and strain R300 in Rondônia (22). Previous reports of R. rhipicephali in Europe and Africa were based on PCR-restriction fragment length polymorphism profiles compatible with R. rhipicephali after digestion of ompA PCR products obtained from ticks (2, 11, 16). Thus, further studies based on isolation in cell culture and/or DNA sequencing of PCR products need to be conducted to confirm the presence of R. rhipicephali outside the New World.

The present study reports six isolates of rickettsiae from six H. juxtakochi ticks from the state of São Paulo, Brazil. As only one out of seven ticks was not infected by rickettsiae, the results suggest high infection rates by Rickettsia spp. among this tick species. Unfortunately, no more H. juxtakochi specimens were available for better evaluating the Rickettsia infection among tick populations, since H. juxtakochi is a rare tick in Brazil, often collected in very low numbers either on host or on vegetation (26, 43). Interestingly, the only previous report of Rickettsia infection in H. juxtakochi was based on the evaluation of a single specimen, which was shown to contain DNA of Rickettsia sp. strain R300 (25).

Since its first isolation from R. sanguineus ticks in Mississippi (4), R. rhipicephali has never been associated with human infection, whereas laboratory experiments have shown this Rickettsia sp. to be moderately pathogenic for guinea pigs (5). Regarding R. bellii, a large inoculum of this organism was shown to be slightly pathogenic for rabbits under experimental conditions (29). Nevertheless, it must be emphasized that six of the current Rickettsia species of recognized pathogenicity for humans were firstly considered nonpathogenic or of unknown pathogenicity until their further isolation from infected human patients (30). Thus, the role of both R. rhipicephali and R. bellii as human pathogens should be regarded as unknown.

Our isolation of R. bellii from H. juxtakochi ticks, added to eight other tick species that have been reported to be infected with this bacterium in Brazil, indicates that R. bellii is indeed the most frequent Rickettsia species infecting ticks in this country. In the United States, R. bellii has been isolated from eight other tick species (34). Interestingly, previous reports of isolation of R. bellii in Brazil (22) and the United States (32, 34) have referred to bacillary to “long forms” of the organisms, in contrast to the typical coccobacillary forms observed for the spotted fever group of Rickettsia species (23, 32, 34). Contrastingly, the coccobacillary forms predominated for the H. juxtakochi isolates of R. bellii of the present study whereas no “long forms” of the organisms were observed. It is noteworthy that the cell culture conditions used in the present study were the same as those used for previous isolates of R. bellii from Brazil (22). In addition, our laboratory has maintained several other Brazilian isolates of R. bellii in Vero cell culture under the same growth conditions, always showing the typical bacillary to “long forms” of the organisms (unpublished data). The reasons for such phenotypical differences are unknown. Finally, the recent genome sequencing of a North American isolate of R. bellii showed several specific features, such as the presence of a complete set of putative conjugal DNA transfer genes most similar to homologues found in endosymbionts of amoebae and the presence of sex pilus-like cell surface appendages (29).