Rickettsia felis in Aedes albopictus Mosquitoes, Libreville, Gabon

To the Editor: Rickettsia felis, an emerging pathogen first identified in the cat flea (1), has been detected in other fleas, ticks, mites, and booklice (2). R. felis can be cultured in mosquito cell lines derived from Anopheles gambiae and Aedes albopictus (Asian tiger) mosquitoes (2), so its compatibility with mosquitoes in nature can be suspected. In sub-Saharan Africa, R. felis bacteremia in humans is common¸ especially during the rainy season, when mosquitoes proliferate. We tested anthropophilic mosquitoes for the presence of R. felis DNA (3–5). 
 
During December 2008–January 2010, we randomly selected female Ae. albopictus and Ae. aegypti mosquitoes (96 each) from specimens obtained by human-landing collections from 4 sites in Libreville, Gabon (6). Specimens were collected during the rainy season (mid-January–end of May and end of September–mid-December); no parity data were available. 
 
We extracted 192 DNA samples from homogenate (abdomen, wings, legs) of each nonengorged, host-seeking, adult mosquito by using the BioRobot 8000 (QIAGEN S.A.S., Courtaboeuf, France) and QIAamp Media MDx Kit (QIAGEN) according to the manufacturer’s instructions. Samples were screened by quantitative real-time PCR (qPCR) targeting the biotin synthase (bioB) gene (4). Positive results were confirmed by qPCR-based molecular detection targeting the orfB gene, which codes for a transposition helper protein. This qPCR used a set of primers not previously used in our laboratory (R_fel.OrfB_F: 5′-CCCTTTTCGTAACGCTTTGCT-3′ and R_fel.OrfB_R: 5′-GGGCTAAACCAGGGAAACCT-3′) and the probe R_fel.OrfB_P: 6-FAM-TGTTCCGGTTTTAACGGCAGATACCCA-TAMRA. Specificity of the qPCR was tested in silico and on the 31 Rickettsia spp. from our laboratory. The final qPCR reaction mixture contained extracted DNA (5 μL) and mix (15 μL) that contained master mix (10 μL) from the QuantiTect Probe PCR Kit (QIAGEN, Hilden, Germany), each primer (0.5 μL, 20 pmol), probe (0.5 μL, 62.5 nmol), and RNase-free water (3.5 μL). Amplification and sequence detection were performed in a CFX96 Touch thermocycler (Bio-Rad, Marnes-la-Coquette, France) as follows:15 min at 95°C followed by 40 cycles of 1 s at 95°C, 40 s at 60°C, and 40 s at 45°C. 
 
Test results for all Ae. aegypti homogenates were negative for R. felis DNA. Of the 96 Ae. albopictus specimens, 3 (3.1%) had positive test results for the R. felis species–specific real-time qPCR and the confirmatory qPCR, with mean cycle thresholds ± SDs of 37.34 ± 1.7 (bioB gene; mean copies/mosquito 5 × 102 [minimum 1.2 × 102, maximum 1.4 × 103]) and 33.64 ± 1.4 (orfB gene; mean copies/mosquito 5 × 102 [minimum 1.5 × 102, maximum 1 × 103). One of the 3 samples was collected in January and 2 in March. The samples came from 3 different districts of Libreville (Akebe Poteau, Alibandeng, Camp des Boys) and were tested by nested PCR targeting the citrate synthase (gltA) gene (7). Rickettsia montanensis DNA was used as a positive control. Sequencing was performed as described (7), and ChromasPro version 1.34 (Technelysium Pty Ltd., Tewantin, Queensland, Australia) was used to analyze sequence data. Sequences of the bioB (120/120) and gltA (1,230/1,230) amplicons at the nucleotide level were 100% homologous to sequences for R. felis URRWXCal2 (GenBank accession no. {"type":"entrez-nucleotide","attrs":{"text":"CP000053","term_id":"67003925","term_text":"CP000053"}}CP000053). The gltA fragment sequence was deposited in GenBank (accession no. {"type":"entrez-nucleotide","attrs":{"text":"JQ674484","term_id":"386870494","term_text":"JQ674484"}}JQ674484). Mosquitoes were considered positive for R. felis when the qPCR result was 35 cycle thresholds for both genes. 
 
Contamination is a critical problem for the PCR-based identification of microbes. However, the validity of the data we report is based on strict laboratory procedures and controls that are commonly used in the World Health Organization Reference Center for Rickettsial Diseases, including rigorous positive and negative controls to validate the test. Each positive qPCR result was confirmed by another R. felis–specific qPCR (orfB) not previously used in our laboratory (to avoid contamination with other amplicons). 
 
Ae. albopictus mosquitos are native to Southeast Asia, colonizing rural and peri-urban sites. In Gabon, Ae. albopictus was the vector for outbreaks of chikungunya and dengue virus infections (6). Our study indicates that mosquitoes can carry R. felis, and the prevalence and load (1.8% –70% and 1.3 × 103–1.6 × 107, respectively) detected in mosquitoes in this study are consistent with the low-end range of those detected in cat fleas, the confirmed biological vector and reservoir (8,9). 
 
We investigated the presence of Rickettsia spp. in mosquitoes neglected as possible vectors of rickettsial diseases (2). Other Aedes spp. and other genera of mosquitoes should be tested. The role of mosquitoes as Rickettsia spp. vectors remains to be demonstrated in additional studies that use the Mitchell criteria. These studies should include the use of cell culture to isolate or detect R. felis in salivary glands of specimens from wild-caught mosquitoes, PCR, immunofluorescence, and the fluorescence in situ hybridization technique; demonstration of infection of a mosquito after experimental feeding on a bacteremic host or bacterial suspension; and demonstration of the transmission of bacteria to a vertebrate through the bite of a mosquito (10).

HAT data indicate seasonality of this disease; incidence is higher during January, February, and March (p = 0.04, by Mann-Whitney test). Seasonality of HAT incidence has been noted elsewhere and linked to seasonal infl uences on tsetse habitat suitability. We propose that seasonality of cattle trading may also play a role because cattle purchases increase before the Christmas season, which promote pathogen spread and increased transmission. This fi nding is consistent with research highlighting the role of livestock markets in the spread of T. b. rhodesiense in central Uganda and would further support a body of literature suggesting, as espoused by the SOS initiative, that control of animal reservoirs of the disease is a critical component of intervention measures (2,(7)(8)(9). Implementation and enforcement of regulations for treatment of cattle before sale at markets would also contribute to limiting spread (9,10); Interventions in districts in central Uganda in which convergence is predicted have been slow and incomplete. If convergence has occurred, this fi nding indicates that a specifi c region in Africa has had concurrent infection with both causes of HAT, with implications for prevention, treatment, and control. Since 2000, Uganda has had continued northward spread of T. b. rhodesiense infections, reducing the distance with TbG to <100 km, which we believe is a conservative estimate. Reinstatement of active surveillance of HAT and support for central data collection in Uganda are long overdo and warranted immediately.

Rickettsia felis in Aedes albopictus
Mosquitoes, Libreville, Gabon To the Editor: Rickettsia felis, an emerging pathogen fi rst identifi ed in the cat fl ea (1), has been detected in other fl eas, ticks, mites, and booklice (2). R. felis can be cultured in mosquito cell lines derived from Anopheles gambiae and Aedes albopictus (Asian tiger) mosquitoes (2), so its compatibility with mosquitoes in nature can be suspected. In sub-Saharan Africa, R. felis bacteremia in humans is common¸ especially during the rainy season, when mosquitoes proliferate. We tested anthropophilic mosquitoes for the presence of R. felis DNA (3)(4)(5).
During December 2008-January 2010, we randomly selected female Ae. albopictus and Ae. aegypti mosquitoes (96 each) from specimens obtained by human-landing collections from 4 sites in Libreville, Gabon (6). Specimens were collected during the rainy season (mid-January-end of May and end of September-mid-December); no parity data were available.
Mosquitoes were considered positive for R. felis when the qPCR result was <35 cycle thresholds for 1 target gene and the additional DNA sequence was successfully amplifi ed. No sample in this study was positive for only 1 target gene or had a qPCR threshold >35 cycle thresholds for both genes.
Contamination is a critical problem for the PCR-based identifi cation of microbes. However, the validity of the data we report is based on strict laboratory procedures and controls that are commonly used in the World Health Organization Reference Center for Rickettsial Diseases, including rigorous positive and negative controls to validate the test. Each positive qPCR result was confi rmed by another R. felis-specifi c qPCR (orfB) not previously used in our laboratory (to avoid contamination with other amplicons).
Ae. albopictus mosquitos are native to Southeast Asia, colonizing rural and peri-urban sites. In Gabon, Ae. albopictus was the vector for outbreaks of chikungunya and dengue virus infections (6). Our study indicates that mosquitoes can carry R. felis, and the prevalence and load (1.8%-70% and 1.3 × 10 3 -1.6 × 10 7 , respectively) detected in mosquitoes in this study are consistent with the low-end range of those detected in cat fl eas, the confi rmed biological vector and reservoir (8,9).
We investigated the presence of Rickettsia spp. in mosquitoes neglected as possible vectors of rickettsial diseases (2 On the basis of the ampliconspecifi c melting temperature and DNA bands representing the specifi c size of 249-bp after gel electrophoresis, results of qPCR showed 100 (4.76%) infected I. ricinus ticks (Table). Positive results did not vary by developmental tick stages; 4.84% (18/372) adult ticks (5.08% [9/177] female and 4.62% [9/195] male), 4.71% (80/1,698) nymphs, and 6.67% (2/30) larvae were infected (Table). Because Bartonella spp. do not seem to be transmitted transovarially (6), it is likely that larvae had interrupted blood meals and thus did not take up enough blood to develop into the nymphal stage.
Seasonal changes in Bartonella spp. infection rates resulted in a higher peak in May (38/300 [12.67%]) than in the other months (Table)