Comparison of PCR methods for detection of Leishmania siamensis infection

Background Leishmania siamensis, a newly identified species, has been reported as a causative agent of leishmaniasis in Thailand. This organism has been identified and genetically characterized using PCR techniques based on several target genes. However, the sensitivities and specificities of these methods for the diagnosis of L. siamensis infection have never been evaluated. Methods To evaluate the sensitivities and specificities of PCR methods to detect L. siamensis infection, PCR for different genetic markers, i.e., the small subunit ribosomal RNA region (SSU-rRNA), the internal transcribed spacer 1 region (ITS1), cysteine protease B (cpb), cytochrome b (cyt b), heat shock protein 70 (hsp70), the spliced leader mini-exon, and the triose-phosphate isomerase (tim) genes were compared. Results Both the ITS1-PCR and the SSU rRNA-PCR could detect promastigote of L. siamensis at concentrations as low as 0.05 parasites/μl or the DNA concentration at 2.3 pg/μl. Though the ITS1-PCR method only recognized 8 samples as positive, all of these could be assessed as true positive according to microscopic diagnosis and/or amplifying the results of the PCR and their sequencing, while other methods also produced false positive results. Compared with the ITS1-PCR method, the PCR amplified SSU-rRNA and cpb gene showed 100% sensitivity for the detection of L. siamensis in clinical specimens. The PCR amplified mini-exon and hsp70 gene also gave a high sensitivity of 87.5%. In contrast, the PCR methods for cyt b and tim gene showed low sensitivity. The PCR methods for cyt b, mini-exon and tim gene showed 100% specificity compared with the ITS1-PCR. Conclusion As a result, the ITS1-PCR method is a suitable target for PCR-based detection of L. siamensis infection in clinical specimens due to its high sensitivity and specificity. The results of this study can be used for molecular diagnosis as well as in epidemiological studies of L. siamensis in affected areas.


Background
Leishmaniasis, one of the most important vector-borne diseases, has been reported worldwide with an estimated 500,000 new cases and more than 50,000 deaths per year [1]. Leishmania siamensis, a species recently reported in Thailand, causes both cutaneous and visceral leishmaniasis especially in patients with immunocompromised conditions such as HIV/AIDS [2][3][4][5][6]. In addition, leishmaniasis caused by L. siamensis was also reported in animals [7][8][9].
Conventional diagnostic methods based on microscopic examination as well as parasite cultures from clinical specimens such as buffy coat, bone marrow and spleen aspirates show a limited sensitivity [10,11]. Serological diagnostic methods also give a high sensitivity; however cross-reactivity could occur with other kinetoplastid parasites. These methods also fail to distinguish between past and present infection [12]. Accordingly, PCR methods have been widely used to detect Leishmania infection in clinical specimens with a high sensitivity and specificity [13]. Several PCR approaches have been published based on different target genes, i.e., the cysteine protease B (cpb) [14], the cytochrome b (cyt b) [15], the internal transcribed spacer 1 (ITS1) region of the small subunit ribosomal RNA (SSU-rRNA) gene [16], the glucose-6-phosphate dehydrogenase (G6PD) [17], the heat shock protein 70 (hsp70) [18], the spliced leader mini-exon [19], the SSU-rRNA gene [20] and the triose-phosphate isomerase (tim) genes [21]. Of these, the ITS1 region of the SSU-rRNA gene has been one of the common genetic markers used to detect Leishmania species in the Old World [13,16,[22][23][24].
L. siamensis have been identified and genetically characterized using several target genes [2][3][4][5][6]. However, diagnosis of L. siamensis infection using PCR methods based on different genetic markers has never been evaluated. Thus, this study aimed to compare the sensitivity and specificity of different PCR methods, i.e., the cpb, the cyt b, the hsp70, the spliced leader mini-exon, the SSU-rRNA, the tim and the ITS1-PCR method.

Leishmania parasite
Promastigotes of L. siamensis (MHOM/TH/2010/TR), harvested from axenic culture in Schneider's Drosophila medium, supplemented with 20% fetal bovine serum (FBS), were used in this study [25]. Promastigotes were centrifuged at 2,000 g for 15 min, washed three times with phosphate buffered saline (PBS), and resuspended in 1 ml of PBS. Numbers of parasites were counted and calculated using a hemocytometer. DNA was extracted using Illustra tissue & cells genomic Prep Mini Spin Kit (GE Healthcare, UK). DNA extraction from promastigotes of L. donovani strain MHOM/SD/68/1S (Biomedical Research, Amsterdam, The Netherlands) was also performed and used as a reference strain in this study. All extracted DNA samples were stored at −20°C until used. Table 1 shows different primer pairs used to detect L. siamensis in this study, which amplified different target genes, i.e., the cpb, the cyt b, the ITS1 region of the SSU-rRNA, the hsp70, the spliced leader mini-exon, the SSU-rRNA and the tim gene. To amplify the cyt b [15], the ITS1 region of SSU-rRNA [16], the mini-exon [19] and the SSU-rRNA gene [20], the previously described primer pairs were used. For the hsp70 gene, the HSP70senKS primer was designed and replaced the HSP70sen of the reported primer set (HSP70sen/HSP70ant) [18] to increase the sensitivity to detect L. siamensis. In addition, new primer pairs for the cpb (LH-CPBEF-F2/LH-CPBEF-R) and tim (TIM-F/TIM-R) gene were designed and used in this study.

PCR assay
All PCR methods were performed at a final volume of 50 μl, containing 25 pmol of each primer, 0.2 mM dNTP, 2 mM MgCl 2 , 1x PCR buffer, 1.5 U of Taq DNA polymerase and 1 μl of DNA template, using the optimal annealing temperature (Table 1). Extracted DNA from normal human blood was used as the internal control and distilled water was used as the negative control. Expected amplicons of PCR products were visualized using electrophoresis at 100 V on 1.5% agarose gel mixed with SYBR® Safe DNA gel stain (Invitrogen, USA) in 1X Tris Borate EDTA (TBE) buffer.

Detection limit of PCR methods
Detection limit of all PCR methods was performed to detect L. siamensis. Ten-fold serial dilutions were performed to obtain samples containing 500, 50, 5, 0.5, 0.05, and 0.005 parasites/μl. DNA was then extracted using Illustra tissue & cells genomic Prep Mini Spin Kit (GE Healthcare, UK). Further PCR steps were performed using 1 μl of DNA template, containing the equivalent of the above dilutions. Another set of DNA templates was also used. DNA concentration was measured using a Nano Drop spectrophotometer (ND-1000 model, Thermo Scientific, USA). Ten-fold serial dilutions of the extracted DNA were then performed to obtain the DNA concentration at 23,000, 2,300, 230, 23, 2.3, and 0.23 pg/μl. One microliter of each concentration was used as DNA template for PCR amplification. The most sensitive method was defined as the method that could amplify the DNA extracted from the lowest number of promastigotes or the lowest DNA concentration of L. siamensis. In addition, DNA samples of Trypanosoma brucei, T. cruzi, T. evansi, Trichomonas vaginalis, Giardia intestinalis, Plasmodium falciparum, and Entamoeba histolytica were also test for the specificity of each PCR method.
Sensitivities and specificities of PCR methods to detect L. siamensis All 7 PCR methods were then evaluated for their sensitivities and specificities to detect L. siamensis in 42 clinical specimens. These clinical samples were sent to the Department of Parasitology, Phramongkutklao College of Medicine for microscopic and PCR diagnosis of leishmaniasis. Most samples were collected from the patients who presented with fever, hepatosplenomegaly and pancytopenia at tertiary hospitals in the South, the endemic areas of leishmaniasis in Thailand. Extracted DNA of 42 clinical samples from patients including buffy coat, bone marrow and tissue biopsy (one skin biopsy) were used. Of these, eight samples were assessed as true positive according to the results of the PCR amplifying the ITS1 region of the SSU-rRNA gene and their sequencing. Thirty four samples were considered as true negative because ITS1-PCR showed negative result. Of eight positive samples, three samples from bone marrow aspiration and one sample of skin biopsy were microscopically positive for amastigotes and ITS1-PCR positive for Leishmania. The other four positive samples were diagnosed by ITS1-PCR amplification using buffy coat samples. All eight samples were confirmed as L. siamensis infection by DNA sequencing of the ITS1 region of the SSU-rRNA gene. Sensitivities and specificities of six PCR methods to detect L. siamensis were determined comparing to the ITS1-PCR method.

DNA sequencing
All positive PCR products from any PCR assays were confirmed for L. siamensis sequence identity. Bidirectional DNA sequencing was performed by the 1st Base Pte., Ltd., Singapore. Nucleotide sequences were validated using the Bioedit Program version 7.0.1.

Statistical analysis
Sensitivities, specificities and positive and negative predictive values were calculated using two-by-two tables and StataCorp. 2009, Stata Statistical Software: Release 11. College Station, TX: StataCorp LP.

Ethics statement
This study was approved by the Ethics Committee of the Royal Thai Army Medical Department, Thailand. No information on the patients was presented in this study.

Cross-amplification of PCR methods using DNA of other parasites
We determined the cross-amplification of these seven PCR methods using the DNA of other parasites. PCR amplifications of the mini-exon and the tim genes amplified no other parasites' DNA ( Table 3). The PCR method for the cyt b gene showed cross-amplification with T. brucei DNA. Both PCR methods targeting the SSU-rRNA and hsp70 genes could amplify the DNA of T. brucei, and T. evansi. The PCR method for the cpb gene amplified all the DNA of flagellates used in this study. The L5.8S/LITSR primers amplifying the ITS1 region could not amplify DNA of other parasites at 348 bp. A band at 600 bp appeared when amplifying T. evansi DNA. However, in contrast to the cross reactions produced by the other PCR methods, the positive band of T. evansi could easily be differentiated from that of L. siamensis by their distinctive sizes.
Sensitivities and specificities of PCR methods to detect L. siamensis in clinical samples Table 4  Both PCR method for the mini-exon and the hsp70 genes showed the sensitivity of 87.5%. PCR method for the cyt b gene showed the lowest sensitivity for the detection of L. siamensis. The specificity of 100% was also observed for PCR amplified the cyt b, the mini-exon and the tim gene.

Discussion
The present study showed that the PCR amplifying the ITS1 region of SSU-rRNA gene (L5.8S/LITSR) was the most accurate method to detect L. siamensis. Though only 8 samples were recognized as positive, all of these could be assessed as true positive according to microscopic diagnosis and/or amplifying the results of the PCR and their sequencing, while other methods also produced false positive results. To date, there is still no gold standard technique for the diagnosis of leishmaniasis. Thus, the sensitivities and the specificities of other six PCR methods were determined by comparing to the ITS1-PCR method. Our result was similar to a recent study that tested the ITS1-PCR method in L. donovani [26]. This method could detect the promatigote of L. siamensis as low as 0.05 parasites/μl or a DNA concentration at 2.3 pg/μl. For clinical samples, 100% of L. siamensispositive samples (eight samples) were identified by this method. Using the ITS1 region of the SSU-rRNA gene as the target for PCR showed a high sensitivity because approximately 20-400 copies of the SSU-rRNA gene per Leishmania genome exist [27][28][29]. The ITS1-PCR method can also differentiate L. siamensis from L. donovani in a single-tube PCR. The DNA of L. siamensis was detected at 348 bp whereas that of L. donovani was shown at 319 bp. Additionally, this primer pair could not amplify the DNA of other protozoa parasites, except T. evansi, which was detected at 600 bp. A study in Brazil also showed a cross-amplification of the ITS1-PCR method in T. cruzi showing the amplicon at 700 bp [30]. However, L. siamensis and T. evansi can be clearly discriminated by the size of their amplicons. The ITS1-PCR method also gave a negative result when human DNA was used as the internal control (data not shown). The detection limit of S4/S12, the primer pair targeting the SSU-rRNA gene was compatible with the ITS1-PCR method. Although the SSU-rRNA PCR showed 100% sensitivity to detect L. siamensis in clinical samples, false positive results could be detected in six samples. DNA sequencing of these PCR products indicated that this primer pair could amplify human DNA. In addition, the S4/ S12 primers could also amplify the DNA of other flagellate protozoa, i.e., T. brucei and T. evansi and gave the amplicon at 563 bp. Thus, the SSU-rRNA PCR method to diagnose Leishmania infection in clinical specimens should be used with caution and the positive PCR products have to *Ten-fold serial dilutions were performed to obtain samples containing 500, 50, 5, 0.5, 0.05 and 0.005 promastigotes/μl before DNA extraction.
be confirmed by sequence analysis. Another disadvantage of the SSU-rRNA PCR method is that it cannot differentiate among the different species of Leishmania [13,25,30,31]. The Fme/Rme primer set, amplifying the mini-exon gene [19], could detect L. donovani at the DNA concentration as low as 10 pg/μl. The limit of this primer set to detect L. siamensis was in the same range as previously reported in L. donovani [32]. The mini-exon gene contains 100-200 copies in the nuclear Leishmania genome, consisting of transcribed conserved and non-transcribed variable regions [32]. These regions differ among Leishmania species. Thus, it could be used to differentiate Leishmania species. For clinical specimens, this primer set showed the sensitivity of 87.5% for the detection of L. siamensis. Although a previous study showed a crossspecies amplification in T. cruzi, T. brucei, and Crithidia fasciculata [33], no amplification in a wide range of parasites and also human DNA was found in this study.
Montalvo et al. [34] reported a primer set specific to the hsp70 to detect Leishmania. This primer set showed different detection limits in different Leishmania subgenera. In the present study, the HSP70sen/HSP70ant primer set could amplify the DNA of L. siamensis at the lowest concentration of 230 pg/μl. Therefore, the HSP70senKS primer was designed and replaced the HSP70sen to increase the sensitivity showing the detection limit at 23 pg/ μl. However, this newly designed as well as the previously described primer pairs [34] could detect other hemoflagellates. False positive result might also limit the use of HSP70 PCR. DNA sequencing of 7 false positive PCR products showed that the HSP70 PCR could amplify human DNA as well.
The cpbEF-frd/cpbEF-rev primer set has been used to amplify the cpb gene of Leishmania species, both L. donovani and L. infantum [35,36]. However, this primer set could not amplify the DNA of L. siamensis. Therefore, a new primer set (LH-CPBEF-F2/LH-CPBEF-R) was designed based on the cpb sequence data of all Leishmania species reported in GenBank. This new primer set amplified L. siamensis DNA and showed 100% sensitivity, equivalent to the ITS1 and SSU-rRNA PCR methods. Unfortunately, LH-CPBEF-F2/LH-CPBEF-R primer set could amplify the DNA of other flagellate parasites including T. brucei, T. cruzi, T. evansi, T. vaginalis and G. intestinalis. Among all tested PCR methods, the LH-CPBEF-F2/LH-CPBEF-R primer set gave the lowest specificity to detect L. siamensis in clinical specimens.
To date, the information on tim gene are limited because only a small number of sequences of Leishmania species were reported in GenBank. In the present study, a new set of primers amplifying the tim gene has been designed which could amplify Leishmania at the concentration as low as 0.5 parasites/μl or a DNA concentration at 23 pg/μl. However, the sensitivity of tim-PCR method was still lower compared with the ITS1-PCR method. The L.cyt-S/L.cyt-R primer set for the PCR method amplifying the cytb gene also showed a low sensitivity at 37.5%. Thus, these two PCR methods might not be suitable for the screening of L. siamensis infection.
The limitation of this study was that the number of patients/clinical samples used in this study was rather small and the 95%-CIs were therefore extremely wide. Increased accuracy of these tests may need a larger sample size. However, the acquisition of sufficient positive  samples can be a troublesome labor and maybe the acquisition of higher number samples of this newly recognized parasite species was not possible.

Conclusion
Among these tested PCR methods, the ITS1-based method is the most sensitive and suitable to diagnose leishmaniasis caused by L. siamensis. This PCR method could differentiate L. siamensis and L. donovani by their amplicon sizes in a single-tube PCR. Due to its high sensitivity, this method would be useful for the epidemiological survey of L. siamensis infection in Thailand.