Association of Phlebotomus guggisbergi with Leishmania major and Leishmania tropica in a complex transmission setting for cutaneous leishmaniasis in Gilgil, Nakuru county, Kenya

Abstract

Background: Phlebotomus (Larroussius) guggisbergi is among the confirmed vectors for cutaneous leishmaniasis (CL) transmission in Kenya. This scarring and stigmatizing form of leishmaniasis accounts for over one million annual cases worldwide. Most recent CL epidemics in Kenya have been reported in Gilgil, Nakuru County, where the disease has become a public health issue. However, little is known about the factors that drive its transmission. Here, we sought to determine the occurrence, distribution and host blood feeding preference of the vectors, and to identify Leishmania species and infection rates in sandflies using molecular techniques. This information could lead to a better understanding of the disease transmission and improvement of control strategies in the area.

Methodology/ principal findings: An entomological survey of sandflies using CDC light traps was conducted for one week per month in April 2016, and in June and July 2017 from five villages of Gilgil, Nakuru county; Jaica, Sogonoi, Utut, Gitare and Njeru. Sandflies were identified to species level using morphological keys and further verified by PCR analysis of cytochrome c oxidase subunit I (COI) gene. Midguts of female sandflies found to harbour Leishmania were ruptured and the isolated parasites cultured in Novy-MacNeal-Nicolle (NNN) media overlaid with Schneider's insect media to identify the species. Leishmania parasite screening and identification in 198 randomly selected Phlebotomus females and parasite cultures was done by PCR-RFLP analysis of ITS1 gene, nested kDNA-PCR and real-time PCR-HRM followed by sequencing. Bloodmeal source identification was done by real-time PCR-HRM of the vertebrate cytochrome-b gene. A total of 729 sandflies (males: n = 310; females: n = 419) were collected from Utut (36.6%), Jaica (24.3%), Sogonoi (34.4%), Njeru (4.5%), and Gitare (0.1%). These were found to consist of nine species: three Phlebotomus spp. and six Sergentomyia spp. Ph. guggisbergi was the most abundant species (75.4%, n = 550) followed by Ph. saevus sensu lato (11.3%, n = 82). Sandfly species distribution across the villages was found to be significantly different (p<0.001) with Jaica recording the highest diversity. The overall Leishmania infection rate in sandflies was estimated at 7.07% (14/198). Infection rates in Ph. guggisbergi and Ph. saevus s.l. were 9.09% (12/132) and 3.57% (2/56) respectively. L. tropica was found to be the predominant parasite in Gilgil with an overall infection rate of 6.91% (13/188) in Ph. guggisbergi (n = 11) and Ph. saevus s.l. (n = 2) sandflies. However, PCR analysis also revealed L. major infection in one Ph. guggisbergi specimen. Bloodmeal analysis in the 74 blood-fed sandflies disclosed a diverse range of vertebrate hosts in Ph. guggisbergi bloodmeals, while Ph. saevus s.l. fed mainly on humans.

Conclusions/ significance: The high infection rates of L. tropica and abundance of Ph. guggisbergi in this study confirms this sandfly as a vector of L. tropica in Kenya. Furthermore, isolation of live L. tropica parasites from Ph. saevus s.l. suggest that there are at least three potential vectors of this parasite species in Gilgil; Ph. guggisbergi, Ph. aculeatus and Ph. saevus s.l. Molecular identification of L. major infections in Ph. guggisbergi suggested this sandfly species as a potential permissive vector of L. major, which needs to be investigated further. Sandfly host preference analysis revealed the possibility of zoonotic transmissions of L. tropica in Gilgil since the main vector (Ph. guggisbergi) does not feed exclusively on humans but also other vertebrate species. Further investigations are needed to determine the potential role of these vertebrate species in L. tropica and L. major transmission in the area.

Fig 1. Map of Kenya (inset) and Gilgil Sub-county showing the location of the trapping sites.

Gilgil covers an area of approximately 1348.4 Km² and has a population of 138,448 people. Lowest temperatures are experienced in June and July while the highest temperatures occur from December to February. The map was designed using QGIS (v 3.0.3).

Fig 2. Sandfly species diversity across the sampling sites in Gilgil.

M: male; F: female.

Fig 3. Phylogenetic analysis of the sandfly mitochondrial COI sequences by Maximum Likelihood method.

The analysis was performed in MEGA v.7 based on the Kimura 2-parameter model. Bootstrap support values based on 1000 replicates are shown at the nodes. Other reference phlebotomine species belonging to different subgenera were included for comparison. The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. Sample identities are shown in red while sequences retrieved from the GenBank are indicated in black. Representative CO1 sequences for each species are available in the GenBank under accession numbers: MK169221-MK169223.

Fig 4. Leishmania parasite identification by ITS1-PCR-RFLP.

M: 100 bp ladder;1 and 2: Leishmania spp. isolated from sandflies; 3 and 4: L. major (Friedlin strain) and L. tropica (Lv357 strain) positive controls; 5: negative control. Molecular sizes of RFLP product for each species are shown; L. major (220 and 140 bp); L. tropica (200 and 60 bp).

Fig 5. Real-time PCR-HRM analysis of Leishmania infection in female sandflies.

A derivative dissociation curve of 179 bp L. tropica (Tm = 84.2°C) and 195 bp L. major (Tm = 85.8°C) ITS1-PCR amplicons. Infection was inferred by comparing the melting profiles of the samples to the controls.

Fig 6. Map of Gilgil showing the proportions of bloodmeal sources per trapping site.

The map was designed using ArcGIS Online version.

Fig 7. HRM profiles of vertebrate hosts present in sandfly bloodmeals using CYT-B.

Positive controls are shown in the legend; human, rock hyrax, goat and rabbit. The melting profiles of mixed bloodmeals varied slightly compared to those of pure bloodmeals. Bloodmeal source detection failed in two partially blood-fed sandflies (2.70%). This could be due to the degradation of DNA in the bloodmeal. The e-values of BLAST search matches to GenBank sequences, percentage identities and GenBank accessions for some of the exclusive blood-fed sandflies are provided in Table 3.