Abstract
During 2010, 15 adult ticks, identified as Amblyomma cajennense, were collected from horses in Cahuita and Turrialba districts, whereas 7 fleas, identified as Ctenocephalides felis, were collected from a dog in San Jose city, Costa Rica. In the laboratory, three A. cajennense specimens, two from Cahuita and one from Turrialba, were individually processed for rickettsial isolation in cell culture, as was a pool of seven fleas. Rickettsiae were successfully isolated and established in Vero cell culture from the three ticks and from a pool of seven fleas in C6/36 cell culture. The three tick isolates were genotypically identified as Rickettsia amblyommii, and the flea isolate was identified as Rickettsia felis through DNA sequencing of portions of the rickettsial genes gltA, ompA, and ompB of each isolate. In addition, other seven ticks were shown to contain rickettsial DNA. Polymerase chain reaction products of at least two of these ticks were sequenced and also showed to correspond to R. amblyommii. Overall, 66.7% (10/15) of the A. cajennense adult ticks were found to be infected with rickettsiae. This is the first report of a successful isolation in cell culture of R. amblyommii and R. felis from Central America.
Introduction
Different strains of Rickettsia amblyommii, an SFG agent, have been reported infecting Amblyomma ticks in the United States, Panama, French Guiana, Brazil, and Argentina (Labruna 2009, Bermudez et al. 2010, Jiang et al. 2010). The role of R. amblyommii as a human pathogen has not been shown; however, a recent retrospective study in the United States demonstrated that sera from patients previously classified as probable Rocky Mountain spotted fever (RMSF) cases (presumably caused by Rickettsia rickettsii) demonstrated a fourfold or greater rise in IgG class antibody titers between paired acute and convalescent sera to R. amblyommii antigens, but not to R. rickettsii antigens. These patients were from areas where known vectors of R. rickettsii were rare or absent but, at the same time, with high abundance of R. amblyommii-infected Amblyomma americanum ticks (Apperson et al. 2008). These authors suggest that these spotted fever cases were caused by R. amblyommii rather than R. rickettsii.
Rickettsia felis, a transitional group agent, is considered an emerging pathogen that is possibly vectored by the cat flea Ctenocephalides felis (Pérez-Osorio et al. 2008). Human infection may result in mild disease that includes fever, myalgia, and rash, but severe disease has been also reported (Zavala-Castro et al. 2009). In the American continent, R. felis has been reported infecting fleas in the United States, Mexico, West Indies, Panama, Peru, Brazil, Argentina, Chile, and Uruguay (Pérez-Osorio et al. 2008). There is yet no evidence of human disease caused by R. felis in any of the Central American countries.
The SFG agent R. rickettsii is the only rickettsial agent that has been reported infecting ticks in Costa Rica, where it causes severe cases of RMSF in humans (Hun et al. 2008). The present study reports the first isolation in cell culture, coupled with molecular characterization, of R. amblyommii and R. felis from arthropod vectors in Central America.
Materials and Methods
Ticks and fleas were collected directly from domestic animals in areas of Costa Rica (Table 1) where RMSF cases have been reported. Specimens were processed individually (ticks) or in a pool (fleas) of seven specimens by isolation of rickettsiae in cell culture and polymerase chain reaction (PCR) analysis. Isolation of rickettsiae from the arthropod samples was performed according to previous protocols (Labruna et al. 2004), with some modifications. Briefly, ticks were surface sterilized, macerated, and resuspended in 1 mL of brain heart infusion. For the tick samples, confluent monolayers of Vero cells in flasks were inoculated with 500 μL of each sample and then centrifuged for 1 h at 700 g. The monolayers were washed and fed with 5 mL of minimal essential medium supplemented 5% with bovine calf serum (Hyclone) and antibiotics (penicillin, streptomycin, gentamicin, and amphotericin B) and incubated at 28°C. The medium was replaced by a medium without antibiotics every 3 days and checked by Gimenez staining and by immunofluorescence assay using anti-R. rickettsii human serum and goat anti-human IgG conjugate. Rickettsia-infected monolayers were passed to new confluent monolayers of Vero cells when infection levels reached 90%–100% cells. These same procedures were performed for isolation of rickettsiae from fleas, except that C6/36 cells were used; this later cell line was fed with RPMI medium supplemented 2.5% with bovine fetal serum (Sigma-Aldrich), without antibiotics, and incubated at 28°C.
Determined by DNA sequencing of the polymerase chain reaction products directly generated from two A. cajennense ticks, three cell culture rickettsial isolates from A. cajennense, and one cell culture isolate from C. felis.
Includes three ticks that were processed for rickettsial isolation.
This same pool was processed for rickettsial isolation.
Infected cells of the second rickettsial passage were subjected to DNA extraction using the NucleoSpin Tissue kit (Machery-Nagel), according to manufacturer's instructions. This same kit was also used to extract DNA from ticks that were not processed for rickettsial isolation. Extracted DNA samples were tested by PCR for three rickettsial genes: citrate synthase (gltA), using primers CS-78 and CS-323 (Labruna et al. 2004); 190-kDa outer membrane protein gene (ompA), using primers Rr190-70 and Rr190.701R (Roux et al. 1996); and 135-kDa outer membrane protein gene (ompB), using primers 120-M59 and 120–807 (Roux and Raoult 2000). In all reactions, DNA of a strain of R. rickettsii from Costa Rica and a strain of R. amblyommii from Brazil were used as positive controls and water was used as negative controls. PCR products were purified using QIAquick gel extraction kit (Qiagen) and sequenced in the Genetic Analyzer 3130 (Applied Biosystems/Hitachi). Edited sequences were analyzed using Basic Local Alignment Search Tool (BLAST).
Results
Adult ticks, identified as Amblyomma cajennense, were collected from horses in Cahuita and Turrialba. Fleas, identified as Ctenocephalides felis, were collected from a dog in San Jose. In the laboratory, rickettsiae were successfully isolated and established in Vero cell culture from three A. cajennense specimens, two from Cahuita and one from Turrialba. These three isolates were designated as AcCR(9-CC-1), AcCR(9-CC-3-1), and AcCR(11-TC-1-1). PCR products of the expected size were obtained for the three rickettsial genes from the three rickettsial isolates. A 350-nt region of the gltA gene was sequenced from the three tick isolates, and it showed to be identical to each other, 100% (332/332) identical to the corresponding sequence of R. amblyommii (GenBank accession no. M582435) infecting A. cajennense in Panama, and 99.7% (349/350) identical to R. amblyommii (AY375163) infecting A. cajennense in Brazil. A 512-nt region of the ompA gene was sequenced from the three tick isolates, and it showed to be identical to each other, 100% (507/507) identical to the corresponding sequence of R. amblyommii (HM582436) infecting A. cajennense in Panama, and 100% (512/512) identical to R. amblyommii (EF689731) from the United States. A 713-nt region of the ompB gene was sequenced from the three tick isolates, and it showed to be identical to each other, 99.9% (712/713) identical to the corresponding sequence of R. amblyommii (AY375164) infecting A. cajennense in Brazil, and 99.7% (711/713) identical to R. amblyommii (FJ455415) infecting A. americanum in the United States.
Rickettsiae were successfully isolated and established in C6/36 cell culture from the flea pool. This isolate was designated as strain CfCR(SJ). PCR products of the expected size were obtained from this isolate for the three rickettsial genes (gltA, ompA, and ompB), and after DNA sequencing, they all showed to be 100% identical to corresponding sequences in the genome of the type strain (URRWXCal) of R. felis (CP000053).
Besides the 3 ticks processed for rickettsial isolation, other 12 A. cajennense ticks from Cahuita were tested by PCR for rickettsial infection. Seven ticks yielded expected amplicons for the gltA, ompA, and ompB PCR protocols. DNA sequencing of amplicons were generated from two ticks, resulting in sequences identical to those generated from the R. amblyommii isolates in cell culture. Overall, 66.7% (10/15) of the A. cajennense adult ticks were found to be infected with rickettsiae.
Aliquots of R. amblyommii-infected Vero cells or R. felis-infected C6/36 cells were used to prepare antigens on glass slides and tested by immunofluorescence assay. Anti-R. rickettsii human serum reacted with R. amblyommii antigens, but no visible reaction was observed with R. felis antigen. However, R. felis forms were observed by Gimenez staining. The rickettsial isolates of the present study have been deposited in the Research Center of Tropical Diseases (CIET) of the University of Costa Rica, where they are available upon request. Partial sequences (gltA, ompA, ompB) generated in this study were deposited into GenBank and assigned nucleotide accession nos. JF694089–JF694091 for R. amblyommii and JF694092–JF694094 for R. felis.
Discussion
This is the first report of a successful isolation in cell culture of R. amblyommii and R. felis from Central America. Although a small number of A. cajennense specimens were tested, we found the majority of them to be infected by rickettsiae, suggesting that, under natural conditions, R. amblyommii infection rates are high among this tick species in Costa Rica. In fact, recent studies in Panama also reported high frequencies of R. amblyommii-infected A. cajennense ticks (Bermudez et al. 2010), similar to previous studies with Amblyomma spp. in Brazil, Argentina, and the United Sates (Labruna 2009, Jiang et al. 2010). On the basis of recent serological evidence suggesting that R. amblyommii can induce spotted fever in humans (Apperson et al. 2008), and because A. cajennense ticks are very aggressive to humans (Labruna 2009), physicians and researchers should consider the possibility of this Rickettsia to cause spotted fever in humans and other vertebrates in Costa Rica.
R. felis is a recognized human pathogen. In Latin America, human cases of spotted fever associated with R. felis have been reported only in Mexico and Brazil (Pérez-Osorio et al. 2008). In the present study, anti-R. rickettsii human serum did not react with the new isolate of R. felis. A previous study reported no cross-reaction between R. felis and R. rickettsii (Tsai et al. 2008). The number of spotted fever cases in Latin America may increase as long as diagnostic laboratories enhance their capability of detection of anti-SFG rickettsiae antibodies, through the use antigens derived from new local isolates.
Footnotes
Acknowledgments
The authors thank Francisco Vega and Carlos Vargas (University of Costa Rica) and Leonardo J. Richtzenhain and Sheila Oliveira de Souza (University of São Paulo) for providing excellent support and technical assistance. The authors are also grateful to Olger Calderon, Ivan Coronado, Adrian Avendaño, Luis Vargas, Carlos Mata, and Jussara Ortiz for providing support in the collection and processing of ticks. This research was supported in part by Netropica (grant no. 9-N-2008) and the Universidad de Costa Rica (project no. 803-A8-127).
Disclosure Statement
The authors declare that no conflicts of interest exist.
