Development of a Diagnostic Marker for Phlebotomus papatasi to Initiate a Potential Vector Surveillance Program in North America

Phlebotomus papatasi, an Old World sand fly species, is primarily responsible for the transmission of leishmaniasis, a highly infectious and potentially lethal disease. International travel, especially military rotations, between domestic locations and P. papatasi-prevalent regions in the Middle East poses an imminent threat to the public health of US citizens. Because of its small size and cryptic morphology, identification of P. papatasi is challenging and labor-intensive. Here, we developed a ribosomal DNA-polymerase chain reaction (PCR)-based diagnostic assay that is capable of detecting P. papatasi genomic DNA from mixed samples containing multiple sand flies native to the Americas. Serial dilution of P. papatasi samples demonstrated that this diagnostic assay could detect one P. papatasi from up to 255 non-target sand flies. Due to its simplicity, sensitivity and specificity, this rapid identification tool is suited for a long-term surveillance program to screen for the presence of P. papatasi in the continental United States and to reveal geographical regions potentially vulnerable to sand fly-borne diseases.


Introduction
Leishmaniasis, a vector-borne disease caused by protozoans, is an often-neglected illness endemic to a total of 98 primarily tropical and subtropical countries. It is estimated that around 2 million new cases of leishmaniasis occur each year, the majority of which occur in South America, East Africa and the Middle East [1]. Of the primary forms of leishmaniasis, the most important are its cutaneous and visceral forms. Cutaneous leishmaniasis (CL) is the most common form of the disease and can cause severe skin lesions and permanent scarring. Visceral leishmaniasis (VL) is responsible for the majority of leishmaniasis-linked deaths and can damage the immune system and lead to deadly complications if untreated [2]. Leishmaniasis is caused by trypanosomes of the genus Leishmania and is transferred to humans through the bite of an infected female sand fly. Sand flies are a group of morphologically challenging-to-distinguish species that includes the major vectors of leishmaniasis. Specifically, 98 sand fly species are known or suspected to act as vectors of leishmaniasis, all belonging to the genera Phlebotomus and Lutzomyia (found in the Old World and New World respectively) [3]. They are small, rarely exceeding a length of 3.5 mm, and noiseless, rendering their attacks on hosts largely undetectable [4]. In addition, symptoms of leishmaniasis normally develop 2 to 8 months after being bitten, obscuring the link between the bite of a sand fly and the onset of disease. Considering the threat posed by leishmaniasis, it is important to establish vector surveillance programs for Phlebotomine sand flies in regions where they are suspected to exist or become established.

Sand Fly Collection and Storage
Lutzomyia shannoni and Lutzomyia vexator specimens were collected from field sites during 2008 at the Fort Campbell Army Installation near Clarksville, TN and the University of Kentucky's Western Research and Education Center in Princeton, KY using standard Center for Disease Control (CDC) light traps (Model 512, John W. Hock, Gainesville, FL, USA). Phlebotomus papatasi specimens originating from Israel, Jordan, North Sinai and Turkey were obtained from lab colonies maintained by the Walter Reed Army Institute of Research in Maryland. Specimens of Lutzomyia longipalpis originating from Brazil were also obtained from a lab colony maintained by Kansas State University. Lutzomyia longipalpis was chosen as a third representative species that is native to the Americas, although its native range is confined to Central and South America.
Sand flies were stored in 95% ethanol upon removal from −20 • C storage. Individual specimens were temporarily removed from ethanol for dissection of the heads and last 2-3 abdominal segments. Dissected body parts were placed in separate wells of 0.25 mL PCR strip tubes along with approximately 0.2 mL of a lactic acid-phenol based commercial clearing solution (Bioquip Inc. Rancho Dominguez, CA, USA) for subsequent taxonomic identification. The remainder of each specimen was individually stored in centrifuge tubes with 95% ethanol and labeled with specimen accession numbers. Specimen vouchers of field collected material were retained in the collection of the Public Health Entomology Laboratory at the University of Kentucky.

Taxonomic Identification
The head and last 2-3 abdominal segments of each specimen were cleared and processed using a modification of the methods presented in Reference [33] with commercial clearing solution used as a substitute for boiling sodium hydroxide. The fly fragments were temporarily mounted on glass microscope slides for viewing at 20× magnification under a compound light microscope and then identified to species using a morphological key [34].

Genomic DNA Extraction and Sample Preparation
Remaining portions of specimens were individually dried in a rotary evaporator to remove ethanol. A single 2.5 mm glass bead was added to each tube along with 75 µL of PCR nanopure water. Tubes were placed in a Mini beadbeater (BioSpec Products Inc., Bartlesville, OK, USA) for 1.5 min of grinding. DNA slurries were mixed with 180 µL ATL lysis buffer (Qiagen Inc., Hilden, Germany) and 20 µL Proteinase K (Qiagen Inc.) and incubated overnight on a dry heating block at 56 • C. The standard DNAeasy Tissue Kit (Qiagen Inc.) extraction protocol was followed from this point on, ending with two final elutions in 100 µL of buffer AE. The DNA concentration of each sample was determined using a NanoDrop ND-1000 Spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA).
Individual sand fly DNA samples were sorted according to species identification. Artificially mixed samples of L. shannoni, L. vexator and L. longipalpis were made by combining individual samples of the same concentration from the three species with a 1:1:1 ratio. Sand fly DNA mixtures were prepared by adding one P. papatasi specimen to each of these mixed samples. Individual P. papatasi samples from Israel, Jordan, North Sinai and Turkey were subjected to a serial dilution with ddH 2 0 or sand fly DNA mixtures described above.

Primer Design
The complete CDS of the mammalian-like lipase (accession number: AY179968) and salivary apyrase (accession number: AF261768) Phlebotomus papatasi mRNA sequences were obtained from NCBI. Both were subjected to Blastn analysis and searched against all insect nucleotide entries. Of the top 100 Blastn results, sequences belonging to the Phlebotomine sand flies, including the query sequence, were retained for multiple sequence alignment with MUltiple Sequence Comparison by Log-Expectation (MUSCLE) [35]. Three primer sets were designed for each mRNA within the regions with the lowest homogeneity. Table 1 shows the sequence for the primer sets along with the expected length of amplified products.

PCR Amplification
Sand fly DNA samples were amplified using the designed primer sets and iQ™ SYBR ® Green Supermix. Each reaction contained 10.0 µL of enzyme supermix, 1.0 µL of template DNA, 1.0 µL of forward primer and reverse primer each and 7.0 µL of ddH 2 0. PCR was performed in accordance with the supermix manufacturer's protocol. PCR was carried out in a thermal cycler with the following cycling conditions: initial denaturing at 94 • C for 3 min; 40 cycles of 15 s denaturing at 94 • C, 30 s annealing at 55 • C and 30 s extension at 72 • C; and a final extension at 72 • C for 5 min.

In Silico Analysis of the Specificity of the Selected Diagnostic Marker
After being selected as the diagnostic marker, the specificity of P. papatasi salivary apyrase (APY) mRNA primer set 2 was analyzed bioinformatically. First, a nucleotide search was conducted on NCBI with the items "salivary" and "apyrase" and the results were restricted to Psychodidae. All returned sequences were aligned using MUSCLE [35] and edited by Mega7 [36]. Another MUSCLE alignment was conducted between the APY primer set 2 product and aligned sequences from the previous step using EMBL-EBI (European Bioinformatics Institute), to generate a percent identity matrix. Pair-wise comparison between the APY primer set 2 product and returned sand fly APY sequences was carried out following the resultant matrix. Second, the APY primer set 2 product was subjected to Blastn search against nucleotide entries in all organisms. Program selection was optimized to "Somewhat similar sequences (blastn)." Returned sequences were compared by query coverage and percent identity. Information regarding the sources of the returned sequences was summarized in Tables S1 and S2.

Results
In order to differentiate the exotic sand fly species P. papatasi from native L. shannoni and L. vexator, our candidate primer sets should be able to only amplify PCR products from P. papatasi genomic DNA. Figure 1 displays the band patterns for amplifications using different sand fly DNA and different primer sets. For reactions using the three mammalian-like lipase (LIP) primer sets, bands corresponding to P. papatasi can be visualized at approximately 130, 130 and 180 bp respectively. However, bands corresponding to non-target species were also observed in reactions using LIP primer sets 1 and 3. Reactions with salivary apyrase (APY) primer sets yielded bands at approximately 170, Insects 2018, 9, 162 5 of 12 230 and 170 bp for P. papatasi. Among them, amplified products of APY primer set 2 exhibited the most intense band for P. papatasi without additional bands for other species, such as the band for L. shannoni found in primer set 1. Therefore, this primer set was selected as the diagnostic marker for the subsequent sensitivity testing.
Insects 2018, 9, x FOR PEER REVIEW 5 of 12 set 2 exhibited the most intense band for P. papatasi without additional bands for other species, such as the band for L. shannoni found in primer set 1. Therefore, this primer set was selected as the diagnostic marker for the subsequent sensitivity testing.  Table 1. Some bands of very low intensity were visible after running the gel but are not visible in photos. Notably, bands pertaining to all three non-target species appear when LIP primer set 3 is used.
When APY primer set 2 was compared with all sand fly APY transcripts, 135 results were returned. After MUSCLE alignment and editing, 98 sequences remained. Most of the sequences ranged from 40% to 70% percent identity with the primer set 2 product, as shown in Figure 2. 53 sequences' identities were between 40% and 50%, 24 were between 50% and 60% and 12 were between 60% and 70%. There were only 4 sequences with identities higher than 80%, all of them belonging to either P. papatasi or P. duboscqi. Information on the identities of the apyrase transcripts with the greatest sequence identity can be found in Table S1.  Table 1. Some bands of very low intensity were visible after running the gel but are not visible in photos. Notably, bands pertaining to all three non-target species appear when LIP primer set 3 is used.
When APY primer set 2 was compared with all sand fly APY transcripts, 135 results were returned. After MUSCLE alignment and editing, 98 sequences remained. Most of the sequences ranged from 40% to 70% percent identity with the primer set 2 product, as shown in Figure 2. 53 sequences' identities were between 40% and 50%, 24 were between 50% and 60% and 12 were between 60% and 70%. There were only 4 sequences with identities higher than 80%, all of them belonging to either P. papatasi or Insects 2018, 9, 162 6 of 12 P. duboscqi. Information on the identities of the apyrase transcripts with the greatest sequence identity can be found in Table S1. returned from the Blastn search across 4 kingdoms: Animalia, Plantae, Fungi and Bacteria. 23 of the sequences were from animals, of which 12 sequences belonged to insects and half of the insects' sequences belonged to sand flies. Among all returned results, only 4 of them had a more than 50% query coverage: 2 from P. papatasi (100%) and 2 from P. duboscqi (99%) with over 87% percent identity, which indicated a significant match with the APY primer set 2 product.
To test the sensitivity of the primer set, we investigated the effect of sample dilution on the amplification of PCR products. Figure 3 illustrates the visibility of PCR products amplified from P. papatasi DNA samples diluted with either ddH20 or artificial DNA mixes of the three non-target native species (L. shannoni, L. vexator and L. longipalpis). The Blastn results with the highest sequence identity are shown in Table S2. 33 results were returned from the Blastn search across 4 kingdoms: Animalia, Plantae, Fungi and Bacteria. 23 of the sequences were from animals, of which 12 sequences belonged to insects and half of the insects' sequences belonged to sand flies. Among all returned results, only 4 of them had a more than 50% query coverage: 2 from P. papatasi (100%) and 2 from P. duboscqi (99%) with over 87% percent identity, which indicated a significant match with the APY primer set 2 product.
To test the sensitivity of the primer set, we investigated the effect of sample dilution on the amplification of PCR products. Figure 3 illustrates the visibility of PCR products amplified from P. papatasi DNA samples diluted with either ddH 2 0 or artificial DNA mixes of the three non-target native species (L. shannoni, L. vexator and L. longipalpis).
Insects 2018, 9, x FOR PEER REVIEW 7 of 12 Figure 3. Sensitivity of PCR-based diagnostic assay. To examine the sensitivity of this diagnostic assay, P. papatasi samples were diluted into ddH20 (A) or sand-fly DNA mixtures (B). In addition, the number of PCR cycles also contributed to the sensitivity of this rapid detection method. Dilution factor based on serial dilution ranged from 0-to 256-fold.
After 50 cycles, in both dilution strategies, bands of amplified product were visible up to the 256th-fold dilution level. Figure 4 illustrates the visibility of PCR products amplified from diluted P. papatasi DNA samples originating from Israel, Jordan, North Sinai and Turkey via APY primer set 2. Bands were visible at the 256th-fold dilution level in all but the Turkey sample, where bands were visible up to the 16th-fold dilution.

Discussion
Previously, members of our lab developed and tested a PCR-restriction fragment length polymorphism (RFLP)-based assay for the differentiation of sand fly species using the mitochondrial cytochrome oxidase 1 (CO1) gene [37]. Here, we demonstrate the use of a similar PCR-based assay as an effective form of surveillance in the detection of non-endemic P. papatasi sand flies. PCR-RFLP has After 50 cycles, in both dilution strategies, bands of amplified product were visible up to the 256th-fold dilution level. Figure 4 illustrates the visibility of PCR products amplified from diluted P. papatasi DNA samples originating from Israel, Jordan, North Sinai and Turkey via APY primer set 2. Bands were visible at the 256th-fold dilution level in all but the Turkey sample, where bands were visible up to the 16th-fold dilution. After 50 cycles, in both dilution strategies, bands of amplified product were visible up to the 256th-fold dilution level. Figure 4 illustrates the visibility of PCR products amplified from diluted P. papatasi DNA samples originating from Israel, Jordan, North Sinai and Turkey via APY primer set 2. Bands were visible at the 256th-fold dilution level in all but the Turkey sample, where bands were visible up to the 16th-fold dilution.

Discussion
Previously, members of our lab developed and tested a PCR-restriction fragment length polymorphism (RFLP)-based assay for the differentiation of sand fly species using the mitochondrial cytochrome oxidase 1 (CO1) gene [37]. Here, we demonstrate the use of a similar PCR-based assay as an effective form of surveillance in the detection of non-endemic P. papatasi sand flies. PCR-RFLP has been used extensively in regions where leishmaniasis is endemic as a means of identification for both

Discussion
Previously, members of our lab developed and tested a PCR-restriction fragment length polymorphism (RFLP)-based assay for the differentiation of sand fly species using the mitochondrial cytochrome oxidase 1 (CO1) gene [37]. Here, we demonstrate the use of a similar PCR-based assay as an effective form of surveillance in the detection of non-endemic P. papatasi sand flies. PCR-RFLP has been used extensively in regions where leishmaniasis is endemic as a means of identification for both sand fly [38][39][40][41][42] and Leishmania [27][28][29] species. In addition to PCR-RFLP, DNA barcoding [43][44][45][46][47] and matrix-assisted laser desorption/ionization time of flight mass spectrometry [48][49][50] have also been used to differentiate sand fly species. In comparison to other methods of identification, the PCR basis of our assay removes the need for DNA sequencing and requires only rudimentary equipment to perform. In addition, it is tuned specifically towards detection of P. papatasi sand flies, which allows the assay to produce completely unambiguous results. Finally, our assay can detect P. paptasi DNA even when it is homogenized with other insect samples, making it ideal as a means of monitoring P. papatasi presence using light trap captures.
Previous studies using P. papatasi revealed that mammalian-like lipase (LIP) is a major protein involved in the secretions of female reproductive accessory glands [51]. Considering that the anatomy of internal reproductive organs is used for sand fly species identification, differences in morphology may reflect on the mRNA sequence of the LIP protein. Research on salivary apyrase (APY) in P. papatasi also indicates that this protein family is highly diverse in hematophagous arthropods [52]. For both proteins, their corresponding mRNAs can be potentially utilized for species identification. Blast searches against insect nucleotide entries generally resulted in low homogeneity with other species, with the only exception being that P. papatasi and P. duboscqi from Mali share 90% identity with the APY mRNA. This similarity has been verified previously [53]. Phlebotomus duboscqi's native range overlaps with that of P. papatasi and it is also capable of carrying Leishmania [3,54]. Thus, although it is likely that our surveillance system yields a positive result with P. duboscqi DNA, this exception is acceptable due to the risk posed by this species.
Multiple sequence alignment for mammalian-like lipase (LIP) and salivary apyrase (APY) of P. papatasi and related sand fly species revealed conserved domains among closely related sand fly species. As denoted in Figure 4, primer sets were preferably selected at disparate regions in the alignment. If the conserved region was overwhelmingly long as in the case of salivary apyrase, high-scoring segment pair (HSP) regions between P. papatasi and non-target species would be used as the secondary selection criteria. This design strategy is intended to improve the probability of obtaining species-specific primer sets at the bioinformatics level.
Candidate primer sets were evaluated on DNA samples of four sand fly species to ensure that a sufficient level of specificity was achieved to distinguish P. papatasi from non-target species. APY primer set 2 was able to amplify DNA samples of P. papatasi but no other species, indicating a strong and specific binding of the primer set to P. papatasi DNA. Other primer sets yielded bands exhibiting various degrees of anomaly including weak intensity, the presence of multiple bands and band deformation. These anomalies were probably due to unspecific binding to DNA samples of non-target species. The length of amplified products was generally in accordance with the expected values predicted from mRNA sequences, indicating the absence of introns and splicing variants that could affect primer positions.
Subsequently, we investigated the sensitivity of APY primer set 2 by diluting P. papatasi samples with ddH 2 0 and DNA mixes of non-target sand flies. Phlebotomus papatasi is a sand fly species which poses an invasion threat to the US. At the initial stage of a potential invasion, the number of P. papatasi that are present in CDC traps would be extremely limited, which poses a great challenge to its detection from a pool of insect DNA mixtures. Nevertheless, APY primer set 2 was able to detect and amplify P. papatasi originating from most regions up to the ninth dilution level, indicating an equivalent identification power of detecting one P. papatasi from 255 native sand fly individuals. Due to possible contamination of the Turkey DNA sample, bands above the fifth dilution level were not readily visible, which is equivalent to detecting one P. papatasi from 15 native sand fly individuals. Compared to the clear gel background of the ddH 2 0 dilution group, some barely visible bands at 300 bp were present in the DNA mix dilution group. While the intensity of the target band at around 140bp was not greatly affected, the presence of irreverent bands illustrates the influence of non-target DNA on P. papatasi detection.

Conclusions
In conclusion, our PCR-based assay can successfully detect the presence of P. papatasi genomic DNA from DNA mixtures consisting of native sand fly species. Our next goal is to apply our assay to captures from standard CDC light traps acquired from Spindletop Farm, a research farm located near the University of Kentucky, and Fort Campbell, Kentucky as a means of field validation. If successful, development of a mobile field kit is also a possibility, to allow technicians to analyze trap captures on-site. Furthermore, it may also be possible to adapt our assay for use in detecting exotic sand fly species in other regions at risk of invasion, including Europe and the Middle East. The most effective time window for any surveillance strategy is prior to or during the initial stage of invasion when introduced populations of non-native species have yet to become established. Using readily available PCR technologies, our detection strategy achieves significant levels of sensitivity and specificity while theoretically reducing the time, labor, cost and expertise required by traditional surveillance strategies based on either morphological traits, molecular features, or both.
Funding: This research was funded by a start-up fund and a gift fund to X.Z. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.