Comparison of LAMP and PCR for molecular mass screening of sand flies for Leishmania martiniquensis infection

BACKGROUND Leishmaniasis caused by Leishmania martiniquensis infection has been reported in human and domestic animals of Martinique Island, Germany, Switzerland, USA, Myanmar and Thailand. The peculiar clinical features of disseminated cutaneous and visceral forms co-existence render the urgent need of specific diagnostic tool to identify the natural sand fly vectors for effective prevention and control strategies. Loop-mediated isothermal amplification (LAMP) of 18S rRNA gene as well as polymerase chain reaction (PCR) of minicircle kinetoplast DNA gene (PCR-mkDNA) have never been applied to detect L. martiniquensis and L. siamensis in sand fly vectors. OBJECTIVE The present study was aimed to validate malachite green-LAMP (MG-LAMP) and PCR-mkDNA techniques to detect L. martiniquensis in sand fly vectors, compared with the conventional PCR of internal transcribed spacer 1 (PCR-ITS1). METHODS We compared the validity of LAMP of 18S rRNA gene and PCR-mkDNA, to PCR-ITS1 in simulation model of L. martiniquensis infection in Sergentomyia gemmea sand flies. Attributable to the sensitivity and specificity, PCR-mkDNA was consecutively applied to detect L. martiniquensis in 380 female sand fly individuals captured in the newly identified affected region of Lamphun Province, Thailand. FINDINGS AND MAIN CONCLUSIONS Results showed that PCR-mkDNA could detect at least one promastigote per sand fly, which was 10-time superior to LAMP and PCR-ITS1. In addition, PCR-mkDNA was more specific, able to differentiate L. martiniquensis from other viscerotropic Leishmania species, such as L. siamensis, L. (L.) donovani, and L. (L.) infantum. Consecutively, mass screening of L. martiniquensis in 380 female sand fly individuals by PCR-mkDNA was implemented in a new affected area of Thailand where a patient with leishmaniasis/HIV co-infection resides; however Leishmania DNA was undetected. In conclusion, PCR-mkDNA is a promising tool for molecular mass screening of L. martiniquensis infection in outbreak areas where several species of Leishmania and sand flies co-exist.

Leishmaniasis is a vector-borne protozoan disease caused by several species of the genus Leishmania. Main clinical manifestations include cutaneous leishmaniasis (CL), mucocutaneous leishmaniasis (MCL), and visceral leishmaniasis (VL), generally associated with the Leishmania species. Leishmania promastigote develop in the gut of female sand flies, and differentiate into intracellular amastigote forms in vertebrate hosts after transmission. The spread of disease depends on the distribution of the vectors and reservoir animal hosts. Autochthonous cutaneous and VL is now considered an emerging disease in Thailand (Leelayoova et al. 2013, Chiewchanvit et al. 2015. Characterisation of Leishmania isolates in Thailand is based on sequence analysis of the internal transcribed spacer 1 (ITS1) and the large subunit of RNA polymerase II genes suggesting that two distinct species belonging to L. enriettii complex are the causative agents; L. siamensis and L. martiniquensis (Pothirat et al. 2014, Chiewchanvit et al. 2015. In contrast to L. siamensis that was isolated only in one patient from Trang Province (Leelayoova et al. 2013), L. martiniquensis is more dominant and has a wider geographical distribution, including France, Germany, Switzerland, USA, Myanmar and Thailand (Chiewchanvit et al. 2015).
L. martiniquensis was first isolated in Martinique, Caribbean, in 1995; its taxonomic position was established in 2002, and it was named as Leishmania (Leishmania) martiniquensis, Desbois, Pratlong and Dedet n. sp. in 2014(Dedet et al. 1995, Noyes et al. 2002, Desbois et al. 2014. This species was also considered in southern Thailand and Myanmar causing various clinical manifestations from asymptomatic, CL alone, VL alone, and atypically disseminated cutaneous and visceral co-existing forms (Phumee et al. 2013, Chiewchanvit et al. 2015, especially in people with human im-munodeficiency virus co-infection. Regarding to animal reservoirs and vectors, Leishmania DNA was identified in black rats (Rattus rattus) and in two species of sand flies; Sergentomyia (Neophlebotomus) gemmea and S. barraudi in southern Thailand (Kanjanopas et al. 2013, Chusri et al. 2014. Between 2011 and 2014, at least five cases of L. martiniquensis infection have occurred in northern Thailand including one case in Chiang Rai province (Phumee et al. 2013); one case in Chiang Mai province (Chiewchanvit et al. 2015) and three cases in Lamphun province (BVBD/MoPH 2013, Pothirat et al. 2014, Chiewchanvit et al. 2015. Due to the continually increasing number of cases of L. martiniquensis in Thailand, the development of a specific diagnostic tool to identify Leishmania infection in circulating sand flies in the affected areas is urgently needed.
Detecting and identifying Leishmania species in sand flies and animal reservoirs are important to predict the risk and transmission of the disease in outbreak and surrounding areas (Kato et al. 2007(Kato et al. , 2010. Molecular techniques, such as polymerase chain reaction (PCR) and PCR-restriction fragment length polymorphism (PCR-RFLP) have been applied to detect and identify Leishmania species in reservoir hosts and sand fly vectors with high sensitivity and specificity (Kato et al. 2007(Kato et al. , 2010. Due to various limitations in the microscopic detection of Leishmania in sand flies, a molecular mass screening method for Leishmania infection of sand fly individuals has been established (Kato et al. 2007(Kato et al. , 2010. This method is a powerful tool for research confirmed on prevalent sand fly species and vector-host-parasite inter-relationships (Kato et al. 2007, 2010, Tiwananthagorn et al. 2012. PCR targeting various genes, such as ITS1, small subunit 18S ribosomal RNA (18S rRNA), minicircle kinetoplast DNA (mkDNA), mitochondrial cytochrome b (cyt b), have been used to identify Leishmania infection in sand flies (Kato et al. 2010, Kanjanopas et al. 2013, Chusri et al. 2014, human patients and animal reservoirs (Leelayoova et al. 2013, Chusri et al. 2014, Hitakarun et al. 2014, Chiewchanvit et al. 2015. PCR targeting the mkDNA gene (PCR-mkDNA) has high sensitivity even when only one Leishmania parasite exists in a sample (Kato et al. 2007). PCR targeting the ITS1 gene (PCR-ITS1) showed high sensitivity to detect L. siamensis as low as 0.05 promastigotes/µL (Hitakarun et al. 2014) and is the classical technique to detect L. siamensis and L. martiniquensis in sand fly vectors (Kanjanopas et al. 2013, Chusri et al. 2014. A colorimetric malachite green based Loop-mediated isothermal amplification (MG-LAMP) assay targeting the 18S rRNA gene has been developed for the robustness and superior sensitivity for mass screening of L. mexicana and L. major infection in sand flies, with a detection sensitivity of 0.01 parasite (Nzelu et al. 2014). Recently, the LAMP assay has been developed for simple detection of L. siamensis in clinical samples with the low detection limit as 10 3 parasites/mL whole blood or 2.5 parasites/tube (Sriworarat et al. 2015). However, LAMP as well as PCR-mkDNA have never been applied to detect L. martiniquensis and L. siamensis in sand fly vectors. The present study, therefore, was aimed to validate MG-LAMP and PCR-mkDNA techniques to detect L. martiniquensis in sand fly vectors, compared with the conventional PCR-ITS1. Attributable to the sensitivity and specificity, PCR-mkDNA was consecutively applied to detect L. martiniquensis in 380 female sand fly individuals captured in the newly identified affected region of Lamphun Province.

MATERIALS AND METHODS
Parasites -Promastigotes of L. martiniquensis (MHOM/TH/2011/PG) were harvested from axenic culture in Schneider's Drosophila medium with L-glutamate (Sigma-Aldrich, USA), supplemented with 20% fetal bovine serum (Merck Millipore, Germany), 100 U/mL penicillin, 100 μg/mL streptomycin, 50 µg/mL gentamicin at 25ºC.  (Chiewchanvit et al. 2015). The sites were the patient's house and the surrounding areas at a radius of 200 m. Collections using CDC light traps were conducted for 12 h between 6:00 pm and 6:00 am both indoors (living room, kitchen), and outdoors (animal shed, crafting studio, animal burrow), bamboo plantation, as well as Doi Khurea mountain (altitude 480 m), where the patient has been working as a lumberjack. All sand flies were stored individually in absolute ethanol and kept at -20ºC until further examination.
Each unfed and blood-fed female sand fly was dissected using sterile techniques under a stereomicroscope. The head and last three abdominal segments of each sand fly were mounted on a microscopic slide in Hoyer's medium. Taxonomic identification was conducted morphologically following Lewis keys (Lewis 1978), such as morphology of cibarium and spermatheca. The remnant parts of sand flies were stored in absolute ethanol individually and kept at -20ºC until DNA was extracted.
DNA preparation -For the preparation of parasite DNA, 10,000 promastigotes of L. martiniquensis were suspended in 50 μL of DNA extraction buffer (150 mM NaCl, 10 mM Tris-HCl [pH 8.0], 10 mM EDTA, and 0.1% sodium dodecyl sulfate) in the presence of proteinase K (200 μg/mL), and serially diluted 10-fold in the same buffer. The samples, without homogenisation, were then incubated at 56ºC for 12 h, heat inactivated at 95ºC for 5 min, and 25 μL distilled water was added. The DNA samples were stored at -20ºC for further use.
To extract DNA from sand flies, a mass extraction technique (Kato et al. 2007) was implemented with a minor modification. Briefly, the ethanol-fixed sand fly specimens were placed individually in each microcentrifuge tube and lysed in 50 μL DNA extraction buffer without homogenisation. The samples were then processed and stored, as mentioned above.
DNA samples of other Leishmania species used in this study were prepared from the following reference strains, In addition, DNA samples of the local stains of Trypanosoma evansi, Leucocytozoon sabrazesi, and Plasmodium gallinaceum were used for the specificity test in this study.
LAMP and PCR assays -MG-LAMP assay targeting the Leishmania 18S rRNA gene (Nzelu et al. 2014) was validated for its sensitivity and specificity for L. martiniquensis detection. Briefly, the reaction was conducted in 15 μL of a reaction mixture consisting of 1.6 μM of each inner primer (FIP and BIP), a 0.4 μM of each outer primer (F3 and B3), 1x reaction mix (Eiken, Japan), 8 U Bst DNA polymerase (Eiken), 0.004% malachite green (MG) dye (dissolved in distilled water), and 1 μL of template DNA. The mixture was incubated at 64ºC for 60 min and then heated at 80ºC to terminate the reaction using MJ Research PTC-200 Thermal Cycler (Bio-Rad Laboratories, CA). At the end of incubation, the amplification of the target gene was confirmed based on direct visual inspection of the reaction tubes by the naked eye; a positive amplification showed as light blue, whereas in the absence of amplification, the reaction mixture became colorless. In addition, LAMP products were analysed on a 2.5% agarose gel electrophoresis.
To identify sand fly species using molecular techniques, PCR and sequencing of the gene mitochondrial cytochrome oxidase subunit I (COI) of metazoan invertebrate (LCO1490/HCO2198) were performed, with the conditions described previously (Nzelu et al. 2015). All PCR products were purified using a QIAquick PCR purification kit (QIAGEN, Germany) and subsequently sent to Applied Biosystems DNA sequencing service (Thermo Fisher Scientific, Japan) for direct sequencing. The sequences were analysed by nucleotide BLAST program (National Center for Biotechnology Information, National Library of Medicine, Bethesda, USA). The sequences were aligned by Clustal W incorporated into MEGA (Molecular Evolutionary Genetics Analysis) version 6 (Tamura et al. 2013). The nucleotide compositions and sequence divergences within and between species were calculated using the distance model Kimura 2-Parameter. A neighbor-joining tree of Kimura 2-Parameter distances with bootstrapping calculation (1,000 replicates) was created to provide the phylogenetic trees that represent the clustering pattern among different species.
Simulation method -Due to the lack of establishment and maintenance of S. gemmea colonies for experimental infection, a simulation model of L. martiniquensis infection in S. gemmea sand flies was established for the validation of MG-LAMP, PCR-mkDNA, and PCR-ITS1 assays. The S. gemmea sand flies, collected from the bamboo plantations, were previously examined for Leishmania infection using PCR-mkDNA. The bodies of each uninfected S. gemmea were separated and consecutively used for the simulation models. The 2 x 10 5 promastigotes/mL of L. martiniquensis were suspended in DNA extraction buffer. Each concentration of 10-fold serial dilutions from 10 4 to 1 promastigote was made in 50 µL of DNA extraction buffer for each fly. The crude DNA was extracted from each fly, processed and stored as previously mentioned.
Sensitivity, specificity, and field application -To determine the sensitivity of MG-LAMP, PCR-mkDNA, and PCR-ITS1 to detect L. martiniquensis in sand flies, the 10-fold serial dilutions of L. martiniquensis (MHOM/ TH/2011/PG) alone (10 4 to 1 parasite in 75 µL; equivalent to 133 to 0.013 parasites/µL), and the crude extracts of L. martiniquensis with S. gemmea DNA (equivalent to 10 4 to 1 parasite per sand fly) were used as the templates. The most sensitive method was defined as the method that could amplify the crude DNA extracted from the lowest number of promastigotes simulated in sand flies. To determine the specificity of each assay, cross amplification of other species of Leishmania and hemoparasites were also used as the template. The most specific method was defined as the method that could identify only L. martiniquensis. For field application, the most sensitive and specific amplification method was applied to detect L. martiniquensis parasites in 380 field captured female sand flies from the newly identified affected area of Tha Mae Lop Subdistrict, Mae Tha District, Lamphun Province (Chiewchanvit et al. 2015).

Sensitivity of MG-LAMP versus PCR-mkDNA -MG-
LAMP and PCR-mkDNA were successfully performed to amplify L. martiniquensis using Leishmania 18S rRNA-LAMP primers (Nzelu et al. 2014) and L.MC-1S/ L.MC-1R primers (Kato et al. 2007), respectively. Detection limit and cross-amplification of MG-LAMP and PCR-mkDNA assays were compared with PCR-ITS1 assay, which was reported as the most sensitive method for L. siamensis detection (Hitakarun et al. 2014). The sensitivities of these assays were assessed with the serial dilutions of L. martiniquensis (MHOM/TH/2011/PG) DNA alone (equivalent to 133 to 0.013 parasites/µL), and the crude extracts of L. martiniquensis mixed with S. gemmea DNA (equivalent to 10 4 to 1 parasite per fly).
PCR-mkDNA revealed a lowest detection limit at 0.013 parasites/µL or at least 1 promastigote of L. martiniquensis in one S. gemmea sample (Figs 1C, 2C). Comparable to PCR-ITS1 assay, MG-LAMP was able to detect 0.13 parasites/µL (Fig. 1B) or 10 promastigotes of L. martiniquensis in one sand fly (Fig. 2B). No amplification was detected in the negative control using DW or sand fly alone. Positive results were visually discriminated when the sample turned light blue (Figs 1A, 2A), whereas the negative control turned from green to colorless. Gel electrophoresis also showed results in agreement with the colorimetric LAMP method using DNA intercalating malachite green dye (Fig. 1A versus 1B, and 2A versus 2B).
Specificity of MG-LAMP versus PCR-mkDNA -DNA of S. gemmea, other Leishmania species, and some vectorborne protozoan parasites including T. evansi, L. sabrazesi, and P. gallinaceum were determined for cross-amplification of MG-LAMP, PCR-mkDNA, and PCR-ITS1 assays. All assays showed no cross-amplification with  S. gemmea sand fly DNA. PCR-mkDNA assay was the most specific to amplify only Leishmania DNA, no crossamplification with T. evansi, L. sabrazesi, and P. gallinaceum (Fig. 3C). Surprisingly, the L.MC-1S/ L.MC-1R primers, amplifying Leishmania mkDNA in this study, could discriminate between L. martiniquensis and L. siamensis with different PCR amplicon sizes, approximately 650 bp for L. martiniquensis and approximately 750 bp for L. siamensis (Fig. 3C). On the other hand, MG-LAMP and PCR-ITS1 assays could amplify T. evansi but no reactivity was detected other avian haemosporozoan DNA samples (Fig. 3A-B 4B). PCR-mkDNA using primer L.MC-1S/ L.MC-1R was also able to amplify other Leishmania species. Similar results of PCR-mkDNA product at approximately 650 bp was observed when the assay amplified L. martiniquensis   (Fig. 4A).
After the validity of mass screening was confirmed, the crude DNA extracts of 380 sand fly individuals were then used as a template for PCR-mkDNA to identify L. martiniquensis; however, the infection was undetected.

DISCUSSION
The present study emphasised the high sensitivity and specificity of PCR-mkDNA to detect L. martiniquensis, at least one promastigote in a sand fly, and revealed the notable ability of PCR-mkDNA assay (L.MC-1S/ L.MC-1R primers) to discriminate L. martiniquensis from other viscerotropic Leishmania species, rendering this PCR-mkDNA assay as a promising tool for molecular mass screening of an individual sand fly for L. martiniquensis infection.
Due to various limitations in the microscopic detection of Leishmania in sand flies, several PCR-based techniques have been developed. The PCR-ITS1 (L5.8S/ LITSR primers) has been reported as the most accurate method to detect L. siamensis (MHOM/TH/2010/TR), as low as 0.05 parasites/µL, and used as the reference assay to compare analytical and filed clinical sensitivity with PCR targeting the 18S rRNA, cyt b, heat shock protein 70, cysteine protease B, spliced leader mini-exon, and triose-phosphate isomerase genes (Hitakarun et al. 2014). This assay has also been employed to detect L. siamensis and L. martiniquensis DNA within the sand fly pools captured from outbreak areas in southern Thailand (Kanjanopas et al. 2013, Chusri et al. 2014. However, the present study showed that PCR-ITS1 had 10-fold less sensitive than PCR-mkDNA and could not discriminate L. martiniquensis from other Leishmania species. The PCR-mkDNA may be an attractive molecular method to apply in the epidemiological study of L. martiniquensis infection in forthcoming outbreak. Wide applicability of LAMP in the detection of parasitic protozoa such as Babesia, Plasmodium, Trypanosome, as well as Leishmania have been reported, due to its advantages, fast and simple amplification without the need of an expensive thermocycler. Recently, colorimetric malachite green based LAMP technique based on the 18S rRNA gene was developed for the detection of L. siamensis, with the detection limit of at least 2.5 in clinical samples, such as whole blood and saliva (Sriworarat et al. 2015). The colorimetric LAMP protocol in the present study, however, showed a higher sensitivity to detect L. martiniquensis in a sand fly. Approximately 20-400 copies of the ITS1 and 18S rRNA gene in individual parasites have been described, although differing somewhat among Leishmania species (Inga et al. 1998).
The PCR-mkDNA using primers L.MC-1S/ L.MC-1R can detect at least one L. (L.) major existed in a sand fly sample (Kato et al. 2007). Along the similar line, this assay also showed the highest analytical sensitivity to detect L. martiniquensis even when only one promastigote existed in a sand fly sample, possibly due to the higher copy number of approximately 10,000 copies of the mkDNA gene in individual parasites (Simpson 1986). Regarding the specificity, this PCR-mkDNA protocol was the most specific to detect only Leishmania parasites, comparing to MG-LAMP and PCR-ITS1. The attractive feature of PCR-mkDNA is the ability to differentiate L. martiniquensis  (Brewster & Barker 2002). In addition, PCR of kDNA gene represent the most reliable tool to detect L. infantum naturally infection in Lutzomyia longipalpis in endemic areas of Brazil, comparing with mini-exon and 18S rRNA genes (Freitas-Lidani et al. 2014). Lastly, the present study found that PCR-mkDNA showed no crossamplification with L. sabrazesi, and P. gallinaceum, in which the potential vector of these avian hemosporidians are ceratopogonid midges. Seblova et al. (2015) suggested that Culicoides soronensis could be potential vectors of L. enriettii, relating to L. martiniquensis and L. siamensis. Validation of this PCR-mkDNA for detection of Leishmania in biting Culicoides midges should be further evaluated for research on ceratopogonid midges as the possible vector of L. martiniquensis and L. siamensis infection.
In Thailand, a few survey studies of the distribution of sand fly species and their habitats have been conducted. Sergentomyia fly was the most predominant genus found in the country. Until now, at least 26 species of sand fly have been reported in different provinces of Thailand, but only S. barraudi had been reported in Lumphun province (Polseela et al. 2016). The present study could provide more information of sand fly populations in Lumphun province, especially in the area where the affected patient resides. At least seven species of sand flies were identified, including P. stantoni, S. gemmea, S. barraudi, S. iyengari, S. bailyi, S. perturbans, and S. indica, of which S. gemmea and S. barraudi were the predominant species. Various studies demonstrated Leishmania DNA in Sergentomyia sand flies, e.g. L. (L.) donovani in S. babu in India (Mukherjee et al. 1997), L. (L.) major DNA in S. minuta in Portugal (Campino et al. 2013), as well as L. siamensis in S. gemmea (Kanjanopas et al. 2013), and L. martiniquensis in S. gemmea and S. barraudi in Thailand (Chusri et al. 2014). Moreover, L. (L.) major was also isolated from S. garnhami and successfully cultured in NNN medium (Mutinga et al. 1994). In this study, Leishmania DNA was not detected, probably due to the very low infection rate (0.01-1%) among sand fly populations even in endemic areas (Kato et al. 2016). Further surveillance of larger populations using the present mass screening approach will provide more information about sand flies in each endemic area. The abundance of S. gemmea and S. barraudi in the local environment of the affected patient may raise awareness of public health concerns for prevention and control of leishmaniasis among policy-and decisionmakers, physicians and the general public. Further surveillance of larger populations using mass screening will provide more information.
In conclusion, the present study highlighted the potential of PCR-mkDNA method as a promising tool to detect L. martiniquensis in sand flies due to its high sensitivity and specificity. Above all, PCR-mkDNA has the valuable ability to discriminate between L. martiniquensis and other viscerotropic Leishmania species; L. siamensis, L. (L.) donovani, and L. (L.) infantum, which may encourage researchers to adopt this approach for epidemiological studies of VL in such areas where many Leishmania species are circulating. Identifying the potential vector for L. martiniquensis still remains an urgent needed. The molecular mass screening of individual sand fly for Leishmania infection by PCR-mkDNA is applicable to provide informative data on the vector and vector-Leishmania relationship in outbreak areas where several Leishmania and sand fly species co-exist and the species of potential vectors remain unknown.