First Report of Bartonella SP. In Sand Flies (Diptera: Psychodidae: Phlebotominae) From Southern Mexico

The genus Bartonella encompasses more than 33 species of Gram-negative, facultative intracellular and hemotropic bacteria, which can cause chronic intraerythrocytic infections in their mammalian hosts, especially rodents and humans (Chomel et al. 2009). Several blood-feeding arthropods are involved in the transmission of these bacteria, such as fleas, sand flies, sucking lice, Acari, and other arthropods (Kosoy et al. 2012, Kamani et al. 2013, Battisti et al. 2015, Regier et al. 2016).

Some Bartonella species are associated with an increasing number of emerging and re-emerging diseases in the human population (Chomel et al. 2003, Regier et al. 2016). In particular, B. bacilliformis, which causes an endemic anthropozoonosis in Peru, has also been reported in outbreaks in southeastern Ecuador and southern Colombia. Additionally, some unconfirmed cases have been reported in Thailand, Bolivia, Chile, and Guatemala (Patiño-Camargo 1939, Hambuch et al. 2004, Sanchez Clemente et al. 2012, Maroli et al. 2013). This pathogen is transmitted by sand fly species such as Pintomyia verrucarum (Townsend), Lutzomyia noguchii (Shannon), and Lu. peruensis (Shannon) in the localities of Huarochiri, Lima, and Urubamba, Cusco, Peru (Noguchi et al. 1929, Hertig 1942, Montoya et al. 1998, Ellis et al. 1999, Villaseca et al. 1999).

In Mexico, few studies have been carried out for the detection of Bartonella in arthropods (Sánchez-Montes et al. 2016, López-Pérez et al. 2017, Moskaluk et al. 2018). Despite 54 species of sand flies having been recorded in Mexico, no studies have been conducted to detect the presence of Bartonella in sand flies (Ibáñez-Bernal 2005, Ibáñez-Bernal et al. 2015). Thus, the aim of this study was to detect a possible natural infection of sand flies with Bartonella in a natural reserve in Los Tuxtlas, Veracruz, Mexico.

Sampling was conducted at the Estación de Biología Tropical, Los Tuxtlas, located in the municipality of San Andrés Tuxtla in Veracruz, Mexico (18°34′N, 95°04′W). Sand flies were sampled during 3 consecutive nights between April and May 2016. Sampling was carried out between 18:00 and 23:00 h, using 5 automatic light traps (model 512; John W. Hock Co., Gainesville, FL, USA). Traps were hung from trees or branches approximately 1 m above the ground. The collected material was separated, fixed, and stored in 70% ethanol. The head and the last segments of the abdomen of the female and male sand flies were mounted on slides using Canada balsam medium for morphological identification, following the procedures outlined by Ibáñez-Bernal (2005). For molecular analysis, we separated the thorax, legs, and abdomen individually in a microtube with 70% ethanol. Identification was carried out according to the classification proposed by Galati (2003).

Genomic deoxyribonucleic acid (DNA) was extracted from the abdomen of sand fly females using Chelex-100 (García-González et al. 2004). For Bartonella detection, we amplified an ∼379-base-pair (bp) fragment of the citrate synthase gene (gltA) using the primers BhCS871.p and BhCS1137.n (Norman et al. 1995). A reaction mixture was prepared in a final volume of 25 μl, with 12.5 μl of GoTaq® Green Master Mix (2X Promega Corporation, Madison, WI, USA), 1 μl of each primer (100 ng each), 10 μl of DNA (∼50 ng), and 0.5 μl of nuclease-free water. We included Bartonella DNA taken from a flea as a positive control, whereas the negative control consisted of ultrapure water devoid of DNA. We followed the polymerase chain reaction (PCR) conditions specified by Kamani et al. (2013). The amplification products were subjected to electrophoresis in 2% agarose gel, and positive samples were sequenced at Laboratorio de Secuenciación Genómica de la Biodiversidad y de la Salud, Instituto de Biología, Universidad Nacional Autónoma de México, Mexico.

Each sequence was compared with sequences available in the National Center for Biotechnology Information database using BLASTn (Basic Local Alignment Search Tool), as a preliminary confirmation of Bartonella amplification. The DNA sequences were aligned with other Bartonella sequences in GenBank using CLUSTAL W in MEGA 6.0. Furthermore, a neighbor joining analysis was carried out using the Tamura 3-parameter distance model with 10,000 iterations. The sequences obtained were deposited in GenBank under accessions numbers MN325838 and MN325839.

In total, 29 phlebotomine sand fly specimens, belonging to 6 species, were collected: Bichromomyia olmeca olmeca (Vargas and Díaz-Nájera), Brumptomyia mesai (Sherlock), Lu. cruciata (Coquillett), Psathyromyia carpenteri (Fairchild and Hertig), Pa. undulata (Fairchild and Hertig), and Psychodopygus panamensis (Shannon). Five specimens could not be identified due to the damage inflicted during the slide mounting process and were therefore recorded as Lutzomyia sp. The most abundant species were Lu. cruciata, with 7 specimens, and Pa. carpenteri, with 5 specimens, whereas the least abundant were Pa. undulata, with 2 specimens, and Br. mesai with 1 specimen. The proportion of females (75.86%) was higher than that of males (20.68%).

All female specimens were tested for Bartonella DNA, and 2 specimens of Lutzomyia sp. tested positive, representing a prevalence of 8.69% (2/23). Recovered sequences showed a 100% similarity between them. The obtained consensus sequences had 307 bp, 99% coverage, and 96% similarities with sequences of Bartonella sp. from Thailand and China (FJ668633 and FJ589054). Additionally, the neighbor joining analysis (Fig. 1) grouped our sequences close to B. grahamii, B. elizabethae, and B. rattimassiliensis. For more than 80 years, extensive research on the epidemiology of B. bacilliformis has been carried out in South America, yet only 1 study detected the presence of B. grahamii in Lu. peruensis from Peru (Ellis et al. 1999). In this study, we show, for the first time, the presence of Bartonella in phlebotomine sand flies outside the endemic area of Verruga peruana in South America. Our work also represents the 1st record of natural infection by Bartonella sp. in sand flies from Mexico. Surveillance studies of Bartonella in Mexico are scattered and restricted to few localities in northern (Janos Biosphere Reserve), central (San Luis Potosi), and southeastern Mexico (Yucatan Peninsula), showing 14 new genetic linages (Rubio et al. 2014, Sánchez-Montes et al. 2016). The prevalence of Bartonella sp. in this study was 8.69% (2/23); this was higher compared to the other studies, which reported a prevalence less than 2% (Ellis et al. 1999, Villaseca et al. 1999).

The recovered sequences exhibited a close relationship with several Bartonella species previously detected in rodents and their ectoparasites, such as B. grahamii and B. elizabethae, which are emerging pathogens that can cause endocarditis in dogs and humans (Bitam et al. 2012, Regier et al. 2016). However, little is known about the potential vectors of this genus, their geographic distribution, or their genetic diversity (Kosoy et al. 2012). We cannot confirm whether sand flies are vectors of this bacterium in this area, since the arthropod could feed on blood meal from an infected host and not necessarily transmit the microorganism. It is possible that other lineages of Bartonella (pathogenic or nonpathogenic) can circulate in the fauna from Veracruz, yet the range of vectors involved in the transmission of Bartonella species remains to be established (Maroli et al. 2013, Battisti et al. 2015). Since an ample diversity of Bartonella species has been recorded in Mexico, further studies are needed to establish the role of sand fly species as potential vectors in order to understand the complex ecological networks in the life cycle of this bacterium.

We thank Giovana De La Cruz Vásquez for her support in collecting bibliographic information and Laura Márquez-Valdelamar for the sequencing of PCR products. Yokomi Nisei Lozano Sardaneta is a doctoral student from Programa de Doctorado en Ciencias Biomédicas, Universidad Nacional Autónoma de México (UNAM) and received a fellowship 444617 from CONACYT. This work was supported by Programa de Apoyo a Proyectos de Investigación e Innovación Tecnológica (PAPIIT) IN 211418.