Introduction

Leishmaniasis is one of the most serious neglected infectious diseases, with 0.7 to 1 million new cases each year and between 20,000 and 30,000 deaths [1,2]. As a vector-borne disease, leishmaniasis poses a substantial threat to millions of individuals in endemic regions across Asia, South and Central America, Africa, and the Mediterranean basin [3]. Leishmaniasis results from an infection with protozoan parasites of the genus Leishmania that are transmitted to humans (and other mammals) through the bite of infected sandflies and manifests in two main clinical forms: cutaneous (or tegumentary) and visceral leishmaniasis, each presenting unique challenges in terms of diagnosis, treatment, and control [4,5]. In the Americas, tegumentary leishmaniasis is distributed in 18 endemic countries and comprises a spectrum of clinical conditions ranging from localized skin ulcers to severe disfiguring mucosal lesions [5]. This clinical pleomorphism is largely dependent on the immune responses of the human host as well as on factors intrinsic to the different infecting Leishmania species [6]. Diverse Leishmania species from the Viannia and Leishmania subgenera circulate in Central and South American countries, with the former being responsible for the majority of leishmaniasis cases [4]. While efforts toward the development of effective prophylactic and therapeutic vaccines against human leishmaniasis are ongoing, early and accurate diagnosis paired with timely chemotherapy remain vital in combating these diseases [7].

Current diagnostic tools for leishmaniasis include microscopy, culture, immunological and molecular methods [8]. Among the nucleic acid amplification techniques (NAATs), the polymerase chain reaction (PCR) and its variants provide the most sensitive and specific tools for the detection and identification of Leishmania species [9,10]. While accessible in research and reference laboratories, PCR-based methods may not be readily available in rural endemic areas since they require infrastructure, resources, and technical expertise [11]. This has led to (near) point-of-care (PoC) diagnostic tests becoming a noteworthy concept, as they would allow easier, faster and less expensive alternatives that need only minimal laboratory setup [12]. Advancements towards simplifying molecular diagnostic tools gave rise to isothermal NAATs, such as loop-mediated isothermal amplification (LAMP) and recombinase polymerase amplification (RPA), which hold potential to decentralize molecular diagnostic testing [13]. Among these, LAMP has gained prominence for its simplicity and efficacy in detecting nucleic acid sequences with high specificity. LAMP operates at a constant temperature (60–65°C), so that it can be performed with a low-cost dry bath incubator [14,15]. The LAMP reaction uses 4–6 primers that target specific DNA sequences, along with the Bst DNA polymerase, an enzyme that displays a strong strand displacement activity [14]. LAMP amplifies the target DNA producing a concatenated DNA product with stem-loop structures, yielding up to 10⁹ copies in less than an hour [14]. The results can be easily visualized, e.g., on the basis of the turbidity caused by reaction by-products [16], or with the use of sequence-specific probes [17,18]. Due to its user-friendly nature and rapidity, LAMP holds great promise as a tool for the diagnosis of infectious diseases in resource-limited settings [19]. LAMP assays have been developed for genus-, complex-, and species-specific detection of Leishmania, demonstrating high sensitivity and specificity with good diagnostic performance in both human cutaneous leishmaniasis (CL) and visceral leishmaniasis (VL) (reviewed in [20,21]). The Loopamp Leishmania Detection Kit (Eiken Chemical, Japan) has also been evaluated for CL and VL diagnosis, showing diagnostic performance comparable to PCR-based methods [22].

Clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated proteins (Cas) technology has recently been harnessed as a tool for molecular detection [23], based on the discovery that once Cas12a and Cas13a enzymes bound to CRISPR RNA (crRNA) recognize specific nucleic acid targets, they become activated, thereby catalyzing cleavage of both the targets (cis-cleavage) as well as non-target nucleic acids (trans-cleavage) [24,25]. This collateral trans-cleavage activity can be applied for nucleic acid detection by coupling it with a reporter molecule. In fluorescence-based assays, the reporter molecule consists of a DNA or RNA oligomer carrying a fluorophore and a quencher. Upon target recognition, reporter cleavage by the activated Cas enzyme generates a detectable fluorescence signal, indicating the presence of the target nucleic acid sequence in a sample. Alternatively, visualization of the Cas detection reaction can be achieved by a lateral flow strip readout using a reporter molecule labeled with FAM or digoxin and biotin [26,27]. CRISPR-based assay platforms reported thus far enable a highly sensitive detection of nucleic acids in the attomolar range when combined with a target preamplification step using either isothermal NAATs (e.g., DETECTR [24] and SHERLOCK [25]) or PCR-based NAATs (e.g., HOLMES [28]). Notwithstanding that this two-step CRISPR-based assay format increases assay complexity and poses the risk of cross-contamination during sample transfer to the CRISPR reaction, it is a convenient approach at the initial development stage to demonstrate the potential for further assay development [29]. More recent efforts have focused on amplification-free detection of nucleic acid targets with CRISPR, and the proof-of-concept was demonstrated for direct detection of pathogen RNA and DNA with quantitative ability [30–33].

The combination of CRISPR-Cas and LAMP technologies has advanced the progress of nucleic acid-based PoC diagnostic test development. Coupling LAMP reactions to downstream CRISPR-based detection improves specificity [19,23], since LAMP is prone to non-specific amplification [34]. Integrated LAMP-CRISPR assays have been developed for the detection of several pathogens, including viruses (e.g., SARS-CoV-2 [26,35–37], human papillomavirus [38]), bacteria (e.g., Mycobacterium tuberculosis [39], Klebsiella pneumoniae [40]), and fungi (e.g., Verticillium dahliae [41]). CRISPR-based assays have also been developed for the detection of parasites of medical and veterinary importance, such as the protozoa Plasmodium spp. [29,42], Toxoplasma gondii [43], Leishmania spp. [44–47], Trypanosoma brucei [48,49], and Trypanosoma cruzi [50]; and the trematode Schistosoma species [51,52], typically in combination with RPA preamplification. Applications of LAMP-CRISPR for parasite detection are increasing, as illustrated by the detection of Schistosoma infection in host samples [53], the in silico design of LAMP primers and Cas12a crRNA sequences for drug resistance genotyping of P. falciparum [54], the detection of T. gondii in environmental samples [55], and the diagnosis of acute congenital T. cruzi infection in infants [56].

At present, leishmaniasis diagnosis in endemic areas is impaired mainly by the lack of infrastructure, equipment, trained human resources, and low access (if any) to effective diagnostic tools. New molecular diagnostic tools that are accurate, easy to use, accessible and affordable for use in resource-limited settings are required. In previous work, we developed CRISPR-Cas12a-based assays coupled with PCR preamplification for detecting Leishmania spp. in clinical samples [45]. Here, we adapted these assays, which target the Leishmania 18S ribosomal RNA gene (18S rDNA) and minicircle kinetoplast DNA (kDNA), by replacing PCR with LAMP in the preamplification step, to facilitate the road towards the development of a nucleic acid-based near-PoC diagnostic tool for leishmaniasis. We optimized assay conditions for a fluorescence signal readout in a plate reader and for visual readout on lateral flow strips. We then evaluated the diagnostic performance of the LAMP-CRISPR assays against a kDNA real-time PCR assay using stored clinical samples from the Cusco region, where Leishmania species of the Viannia subgenus circulate [57,58].

Methods

DNA samples from reference strains of Leishmania and other microbial pathogens

Genomic DNA (gDNA) extracted from cultured promastigotes of reference strains of Leishmania spp. was retrieved from the DNA biobank of the Leishmania research group, Molecular Epidemiology Unit, Instituto de Medicina Tropical Alexander von Humboldt (IMTAvH)-UPCH in Lima, Peru, and from the collection hosted by the Institute for Scientific Research and High Technology Services (INDICASAT-AIP) in Panama. gDNA from cultured laboratory reference strains of Trypanosoma cruzi, Plasmodium falciparum, and Mycobacterium tuberculosis was provided by the Infectious Diseases Research Laboratory at UPCH and INDICASAT-AIP. Strain designations and sources are listed in Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Reference strains used in this work.

https://doi.org/10.1371/journal.pntd.0013456.t001

At the IMTAvH and UPCH, DNA was isolated using the High Pure PCR Template Preparation Kit (Roche, Mannheim, Germany). At the INDICASAT-AIP, the Wizard Genomic DNA Purification Kit (Promega, Madison, WI, USA) was used. DNA was isolated according to the manufacturer’s protocol, suspended in elution buffer or TE buffer and stored at -20°C until use.

Results

Design and optimization of LAMP-CRISPR assays

We developed two assays for the detection of Leishmania spp., targeting the multicopy genetic markers kDNA minicircles and 18S rDNA. These assays combine LAMP with CRISPR-LbCas12a-mediated DNA target recognition and ssDNA reporter cleavage, with results assessed either by fluorescence or lateral flow strip readout (Fig 1). The two readouts are enabled by distinct dual-labeled ssDNA reporters: one labeled with a fluorophore-quencher pair for real-time monitoring of the fluorescence signal on a plate reader, and the other labeled with biotin and FAM for visual detection on paper-based lateral flow strips [26].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Schematic representation of the workflow of LAMP-CRISPR assays for detection of Leishmania spp. in clinical samples.

Sample preparation is performed from skin lesion specimens taken from patients with suspected CL. DNA samples are subjected to LAMP amplification for 30 min (kDNA) or 50 min (18S rDNA), followed by Cas12a-based reactions with either fluorescence or lateral flow strip readout. Figure created in BioRender. Upc, C. (2026) https://BioRender.com/rpoj3uw, with permission to sublicense under CC-BY 4.0.

https://doi.org/10.1371/journal.pntd.0013456.g001

We designed LAMP primers flanking reported crRNA target recognition sites [45] within Leishmania kDNA minicircles (S1 Fig) and 18S rDNA (S2 Fig). According to in silico analyses, the kDNA primer set targets a region most conserved within the L. (Viannia) subgenus and less conserved within the L. (Leishmania) subgenus (S1 Fig and S1 File). In silico analysis of the selected 18S primer set confirmed the expected conserved nature of the targeted genomic region at the Leishmania genus level, whereas sequence variations were more frequent within the Trypanosoma genus (S2 Fig and S2 File).

A schematic depicting the mechanism of LAMP amplification, with the resulting LAMP amplicon containing the Cas12a crRNA target sequence is shown in Fig 2. We optimized the LAMP assay reaction conditions using a Bst 2.0 DNA polymerase and tested the incubation time required for optimal amplification of the target DNA. The CRISPR-Cas12a assay was used as the readout. For kDNA, the LAMP reaction was evaluated at 4 incubation times (10, 20, 30, and 40 min) (S3A-S3D Fig). While 10 min were sufficient to amplify the target DNA sequence present at a high quantity (2 × 10² GE) (S3A Fig), at least 20 min were required for the amplification of low abundant target DNA (2 × 10^-1 GE) (S3B Fig). These results were corroborated by agarose gel electrophoresis (S3F Fig). We chose an incubation time of 30 min as optimal for the kDNA LAMP reaction (S3C and S3E Fig) for further experiments. For 18S rDNA, we tested the LAMP reaction at 5 incubation times (10, 30, 40, 50, and 60 min) (S3G-S3K Fig). At least 40 min were required to amplify the target DNA (S3I Fig). In both assays, extending the incubation time of the LAMP reaction by 20 min for each target resulted in amplification in the NTC controls (S3D and S3K Fig), indicative of contamination. CRISPR-Cas12a-based detection of LAMP amplicons could effectively discriminate between specific and non-specific amplification. The latter was observed, for instance, at the incubation times of 10 min and 40 min in the NTC reaction by agarose gel electrophoresis (S3M Fig) but did not result in false positives in the Cas12a assay (S3G and S3I Fig). On the basis of these results, we selected an incubation time of 50 min for the 18S LAMP reaction (S3J and S3L Fig) for further experiments.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Schematic representation of the LAMP reaction, highlighting the location of the crRNA target sequence in the target DNA.

(A-E) Steps in the LAMP reaction to amplify the target DNA. The eight annealing locations of the six LAMP primers are shown. The primers were designed to target unique sequences within a conserved region of Leishmania kDNA minicircles (S1 Fig) or nuclear 18S rDNA (S2 Fig). The crRNA target sequence and protospacer adjacent motif (PAM) site in the target DNA are shown. For simplicity, the locations of both crRNA target sites are depicted in the same graph. This figure was created de novo by the authors based on the mechanism described in Notomi et al. [14] and Parida et al. [69]. Created in BioRender. Upc, C. (2026) https://BioRender.com/wgofq3n, with permission to sublicense under CC-BY 4.0.

https://doi.org/10.1371/journal.pntd.0013456.g002

The CRISPR-Cas12a assay is based on purified recombinant LbCas12a [67]. For the fluorescence-based assay, we used the optimized conditions [45,65,66] and a quenched fluorescent ssDNA reporter [24]. With respect to the LFA readout, we verified the expected signal patterns of the lateral flow strips (S4A Fig) by analyzing one positive control and one negative control, confirming the switch from C-line to T-line upon reporter cleavage in the former (S4B-S4F Fig). The optimized conditions for the LFA readout were as follows: reporter probe at a final concentration of 100 nM in our in-house 1X Cas12a reaction buffer (which performed similarly to the HybriDetect universal assay buffer); incubation of the Cas12a reaction at 37°C for 60 min; and addition of 2% PEG-8000 to enhance detection (S4B-S4F Fig).

Analytical validation of LAMP-CRISPR assays

We then assessed the analytical sensitivity of each primer set - crRNA combination using serial dilutions of L. braziliensis M2904 gDNA. The kDNA LAMP-CRISPR assay displayed positive L. braziliensis detection down to 2 × 10^-1 GE per reaction, both with a fluorescence readout (Fig 3A) and an LFA readout (Fig 3B). Visualization of the kDNA LAMP products on an agarose gel confirmed the presence of ladder-like bands indicative of a successful amplification (Fig 3C), consistent with the Cas12a-based detection (Fig 3A and 3B). We found a similar analytical sensitivity for the 18S LAMP-CRISPR assay, as detection of at least 2 × 10^-1 GE per reaction was achieved with both readouts (Fig 3D and 3E). The analysis of the 18S LAMP products by agarose gel electrophoresis (Fig 3F) was consistent with the Cas12a assay results (Fig 3D and 3E). The NTC tested in parallel in each assay showed a background signal in the Cas12a fluorescence-based assay (Fig 3A and 3D, see the two lowest GE input amounts whose fluorescence ratio is 1 given that the denominator is the NTC signal). Likewise, the NTC did not exhibit a band (or the band observed was faint) at the test line in the lateral flow strips (Fig 3B and 3E). This ruled out the possibility of cross-contamination and non-specific amplification.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Analytical sensitivity of kDNA and 18S LAMP-CRISPR assays.

Serial dilutions of L. braziliensis M2904 gDNA encompassing 2 × 10⁴ to 2 × 10^-3 genome equivalents (GE) per reaction were subjected to LAMP amplification, followed by Cas12a-based reactions for the kDNA (A-C) or 18S rDNA (D-F) target. (A, D) Normalized fluorescence signals from Cas12a reactions (i.e., fluorescence signals taken at t = 25 min in the test sample relative to the NTC). Data are represented as mean ± SD (n = 3 independent amplification and detection runs). The X-axis has a logarithmic scale. A fluorescence ratio ≥ 2 is considered detected. (B, E) Cas12a reactions were performed using a biotin-FAM reporter molecule and visualized on lateral flow strips. The arrow indicates the direction of flow. Two independent amplification and detection runs (r1 and r2) were analyzed. Photos taken by the authors. (C, F) LAMP reaction products (5 µL) visualized by 2% agarose gel electrophoresis ran at 100 V for 1 h using SYBR Gold staining. L, GeneRuler 100 bp (C) or 100 bp Plus (F) DNA ladder. NC (negative control): NTC (no-template control).

https://doi.org/10.1371/journal.pntd.0013456.g003

Specificity testing (S5 Fig) was performed using gDNA samples listed in Table 1. The kDNA LAMP-CRISPR assay consistently detected representative strains of the L. (Viannia) species present in our sample set (10 of 10; Fig 4A). Additionally, the assay detected 2 out of 6 tested strains belonging to the L. (Leishmania) subgenus (Fig 4A). The 18S LAMP-CRISPR assay showed pan-Leishmania detection ability (Fig 4A). In both assays, no cross-reactivity was observed with related trypanosomatid (T. cruzi) strains and other protozoan (P. falciparum) and bacterial (M. tuberculosis) pathogens tested (Fig 4A). Also, the negative control with human gDNA as template was not detected (Fig 4A). The analytical specificity was also assessed with an LFA readout. Three L. (Viannia) strains used as positive controls were detected (i.e., showing a strong T-line and a very weak or missing C-line) by both LAMP-CRISPR assays (Fig 4B, strips # 3–5). The negative controls (human gDNA and NTC) had a negative test result (Fig 4B, strips # 1 and 2). There was no cross-reactivity to other microbial pathogens tested (Fig 4B, strips # 6–10, which show a strong C-line and a faint (or absent) T-line, similar to the negative controls).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Analytical specificity of kDNA and 18S LAMP-CRISPR assays.

The specificity of the LAMP-CRISPR assays was evaluated using gDNA samples listed in Table 1. (A) Heat maps depicting fluorescence-based detection data. The color scale represents normalized fluorescence signals from Cas12a reactions (n = 2 independent assay repeats). A fluorescence ratio ≥ 2 is considered detected. (B) Cas12a reactions with a biotin-FAM reporter were visualized on lateral flow strips (arrow indicates the direction of flow). Top panel: kDNA assay; bottom panel: 18S assay. Negative controls: HC, human gDNA (strip # 1); NC, no-template control (NTC) (strip # 2). Positive controls: Lbra, L. braziliensis M2904 (strip # 3); Lguy, L. guyanensis M4147 (strip # 4); Lper, L. peruviana LCA08 (strip # 5). Non-related microbes: Mt, M. tuberculosis PA155-18C (strip # 6); Pf, P. falciparum HB3 (strip # 7). T. cruzi (Tc) strains: Tulahuen, Sylvio X10, and VT-1 (strips # 8-10). Photos taken by the authors.

https://doi.org/10.1371/journal.pntd.0013456.g004

Performance evaluation of LAMP-CRISPR assays with clinical samples

Positive patient samples exhibited robust fluorescence curves in the Cas12a assay, indicating the presence of Leishmania kDNA (S6A Fig) or 18S rDNA (S6B Fig) molecules. These results were consistent with the respective positive LAMP reactions visualized on agarose gels (S6C and S6D Fig). We noted, however, that agarose gel analysis of LAMP products can give ambiguous results. For instance, ladder-like bands were observed in samples S72 and S09 amplified by the 18S LAMP assay (S6D Fig), but the banding pattern differed from true positive reactions. This was resolved by the Cas12a assay on these amplicons, confirming that S72 and S09 were negative for 18S rDNA (S6B Fig).

We next evaluated the performance of the LAMP-CRISPR assays with a fluorescence readout using gDNA from 90 clinical samples of skin lesions, compared with a reference kDNA qPCR assay. Cutoff values derived from negative samples were 1.46 for the kDNA assay and 1.30 for the 18S assay (S1 Data). The kDNA LAMP-CRISPR assay detected 50 of 55 qPCR-positive samples (90.9%) (Figs 5 and 6A, Table 2), while the 18S LAMP-CRISPR assay detected 40 of 55 (72.7%) (Figs 5 and 6C, Table 2). Both assays achieved 100% specificity (Table 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Diagnostic performance of LAMP-CRISPR assays with fluorescence readout compared to kDNA qPCR for Leishmania DNA detection.

https://doi.org/10.1371/journal.pntd.0013456.t002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. Performance of kDNA and 18S LAMP-CRISPR assays on clinical samples.

Heat maps showing fluorescence-based detection data from tested clinical samples (n = 90). The color scale represents normalized fluorescence signals from Cas12a reactions. Samples are coded 1-90 (‘Code_UPC’, S1 Data). PC, positive control (L. braziliensis M2904 gDNA, 10³ – 10⁴ genome equivalents). NC, no-template control. Cutoff values were 1.46 (kDNA) and 1.30 (18S). Cq values from the reference kDNA qPCR assay are shown (U = undetermined, indicating that no Cq values were detected within 35 cycles of amplification; X, not applicable).

https://doi.org/10.1371/journal.pntd.0013456.g005

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. Performance of kDNA and 18S LAMP-CRISPR assays across parasite load levels in clinical samples.

(A, C) Scatter plots showing normalized fluorescence signals from Cas12a reactions for Leishmania kDNA (A) and 18S rDNA (C) in clinical samples versus Cq values from kDNA qPCR (data in Fig 5 and S1 Data). (B, D) Scatter plots showing normalized fluorescence signals from Cas12a reactions for kDNA (B) and 18S rDNA (D) versus the estimated parasite load in quantifiable samples (n = 55). The parasite load is expressed as the number of Leishmania parasites per 10⁶ human cells. Horizontal dotted lines indicate LAMP-CRISPR assay cutoff values: 1.46 (kDNA) and 1.30 (18S).

https://doi.org/10.1371/journal.pntd.0013456.g006

The five qPCR-positive samples not detected by the kDNA LAMP-CRISPR assay (codes 3, 5, 6, 28, 71) harbored low parasite loads (≤20 parasites per 10⁶ human cells), corresponding to 10^-2 to 10^-3 parasite GE per reaction (Fig 6B, S1 Data). Similarly, the 15 qPCR-positive samples not detected by the 18S LAMP-CRISPR assay contained ≤150 parasites per 10⁶ human cells, corresponding to 10^-1 to 10^-3 parasite GE per reaction (Fig 6D, S1 Data). The complete dataset of this study is provided in S1 Data.

ROC curve analysis demonstrated good diagnostic performance of the LAMP-CRISPR assays in distinguishing Leishmania-positive from -negative clinical samples, with AUC values of 0.973 for the kDNA assay (Fig 7A) and 0.876 for the 18S assay (Fig 7B).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 7. ROC curve analysis of kDNA and 18S LAMP-CRISPR assays in clinical samples.

ROC curves were generated from fluorescence ratio values of the kDNA (A) and 18S (B) LAMP-CRISPR assays in 90 clinical samples, using kDNA qPCR as the reference test. The area under the curve (AUC) with 95% CIs (calculated using the Wilson/Brown method) is shown. The diagonal dotted line corresponds to the null hypothesis with AUC of 0.5, in which the accuracy of a diagnostic test is no different from random chance.

https://doi.org/10.1371/journal.pntd.0013456.g007

We also assessed the applicability of the LAMP-CRISPR assays with an LFA readout on gDNA extracted from 18 specimens encompassing a wide range of parasite load levels as determined by kDNA qPCR. Lateral flow testing results were consistent across two assay repeats per sample (n = 8; S7 Fig), indicating robust assay reproducibility. Of the 18 samples, 16 (88.9%) tested positive by the kDNA LAMP-CRISPR assay, covering the three parasite load categories: low (4 of 6 tested), intermediate (6 of 6), and high (6 of 6) (Fig 8). These LFA results were 100% concordant with the fluorescence-based readout. The 18S LAMP-CRISPR assay with LFA readout detected Leishmania DNA in 13 of 18 samples (72.2%), which had parasite load levels defined as: low (2 of 6), intermediate (5 of 6), and high (6 of 6) (Fig 8). One sample (code 12) showed discordant results between LFA and fluorescence-based readouts in the 18S assay, associated with a low parasite load (56.6 parasites per 10⁶ human cells). Two samples from non-leishmaniasis patients (kDNA qPCR-negative; samples 09 and C7) tested negative by both LAMP-CRISPR assays. These results are shown in Figs 8, S7 and S1 Data.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 8. Pilot testing of kDNA and 18S LAMP-CRISPR assays on clinical samples using a lateral flow readout.

A subset of clinical samples with varying parasite load levels was tested with LAMP-CRISPR assays using an LFA readout. Visualization on strips was achieved through the cleavage of the biotin-FAM reporter. C, control line; T, test line. The fluorescence (RFU) readout (+, detected; -, not detected) as well as Cq values of the kDNA qPCR are shown below each strip. Controls included DNA samples from two non-leishmaniasis patients (qPCR-negative; samples 09 and C7), a positive control (PC, 2 × 10² GE from L. braziliensis M2904 gDNA), a human negative control (HC, 20 ng of PBMC gDNA), and a no-template control (NTC). Photos taken by the authors.

https://doi.org/10.1371/journal.pntd.0013456.g008

Discussion

The pressing need for new diagnostic tools has been identified as a priority in the World Health Organization (WHO) 2030 roadmap to support the control and elimination of neglected tropical diseases [70]. Leishmaniasis is among the most neglected, causing a large negative impact on people living in rural endemic areas who are often economically disadvantaged and lack access to an early, correct diagnosis, leading to delayed treatment. PCR-based molecular diagnostics provide the most sensitive and specific tools for Leishmania detection and are pivotal for early disease diagnosis, discrimination of Leishmania species, detection of asymptomatic infections, and epidemiological surveillance [71]. Beyond their use in centralized laboratories, the progress in the automation and simplification of the workflow from sample preparation to target amplification and detection steps has made molecular diagnostics more suitable and accessible for resource-limited settings in the form of near-PoC tests [72]. Promising formats include the Leishmania OligoC-TesT (a PCR-oligochromatographic test) [73,74] and a PCR-based method applied directly on EDTA blood for VL diagnosis, using a portable miniPCR device with lateral flow detection [75]. Recent advances also involve isothermal NAATs, such as LAMP and RPA, combined with lateral flow or other portable devices [76–80]. Among the most promising approaches is the integration of isothermal NAATs with CRISPR technology, which has fostered the development of a new generation of molecular diagnostics with the potential to advance toward near-PoC applications in low-resource settings [81].

In this study, we aimed to develop LAMP-coupled CRISPR-Cas12a assays for the molecular detection of Leishmania spp. We designed new LAMP primer sets that flank reported crRNA target sites [45] within two commonly used multicopy regions of the Leishmania genome: the 18S rDNA and kDNA minicircles. We harnessed the CRISPR-Cas12a system as the detection method to avoid possible false positives caused by non-specific amplification in the LAMP reaction, aiming to enhance assay specificity. The assay can be completed within 1.5 h, with results detectable either through a fluorescence-based readout (compatible with basic molecular biology equipment such as a blue-light transilluminator) or visually on portable lateral flow strips. We report the analytical validation and initial performance evaluation of these assays in clinical samples under controlled conditions in a research laboratory setting. Building on our previous proof-of-concept PCR-CRISPR assays [45], the work presented here lays the groundwork for future efforts aimed at making CRISPR-based diagnostics accessible in a portable testing kit format for the diagnosis of leishmaniasis in rural endemic areas.

Here, in keeping with our in silico primer and crRNA design within conserved target genomic regions, the analytical validation demonstrated the ability of the 18S LAMP-CRISPR assay to achieve pan-Leishmania detection, whereas the kDNA LAMP-CRISPR assay reliably detected species belonging to the L. (Viannia) subgenus. The kDNA assay also detected certain strains from the analyzed species of the L. (Leishmania) subgenus, specifically L. mexicana from Central America and L. donovani from the Old World. These results reflect both the similarity and variations in mitochondrial kDNA minicircle sequences between the two subgenera (see S1 Fig). Furthermore, our assays showed no cross-reactivity against other pathogens tested, which co-circulate in the same geographic regions and can thus coinfect the same human host, such as T. cruzi (a closely related trypanosomatid), P. falciparum, and M. tuberculosis, or cause cutaneous lesions similar to CL (M. tuberculosis) [82]. Our LAMP-CRISPR assays achieved high analytical sensitivity on serially diluted gDNA from the L. braziliensis M2904 strain, with a detection level equivalent to 0.2 parasites per reaction. This detection sensitivity (below one parasite GE) compares well with that of the reference kDNA qPCR assay (5 × 10^-3 GE per reaction; [60]), which is clinically relevant.

To explore the performance of our LAMP-CRISPR assays to detect Leishmania DNA in clinical samples, we evaluated a blinded panel of human skin lesion samples (n = 90) from patients enrolled in the Cusco region of Peru, where CL is caused by L. (Viannia) species. Both assays with a fluorescence readout attained good diagnostic performance, with sensitivity rates of 90.9% for kDNA and 72.7% for 18S rDNA, and a specificity of 100%, compared to the reference kDNA qPCR assay. Discrepancies between the two methods corresponded to samples with low parasite loads (S1 Data). Overall, LAMP-CRISPR assay results for the subgroup of samples tested with both fluorescence-based and lateral flow strip readouts were concordant. The 18S assay showed a somewhat reduced sensitivity when applied to clinical samples. This is likely related to differences in target copy number: Leishmania parasites present an estimated 10–170 copies of the nuclear 18S rDNA region per genome [83,84], while they harbor about 10,000 copies of extrachromosomal kDNA minicircles per cell [85].

Several studies have developed LAMP assays for detecting Leishmania infections in diagnosing human VL, CL, and post-kala-azar dermal leishmaniasis (PKDL), as well as in domestic animal reservoirs (dogs) and sand flies. They target multicopy genomic regions, such as kDNA minicircles [86–88], the 18S rRNA gene [89–91], the ITS1 region in the rDNA array [92], cysteine proteinase B genes (cpb) [93], and the heat shock protein 70 gene (HSP70) [94]. These assays proved to be highly sensitive for the detection of VL- and CL-causing Leishmania species (at the genus, species complex, or species level, depending on the target used). To date, few studies have evaluated the diagnostic performance of LAMP assays for New World CL. Nzelu et al. developed a colorimetric LAMP assay using malachite green that targets the Leishmania 18S rRNA gene [90]. This assay could reliably amplify Leishmania DNA from patients’ skin lesion material spotted onto FTA cards [91]. In an evaluation of 122 clinical samples from CL patients in Peru, the FTA-LAMP detected 71/122 (58.2%) samples, performing comparably to nested PCR [91]. León et al. [95] assessed the same LAMP assay [90] on direct smears (n = 50) from suspected CL patients in Colombia, reporting a sensitivity of 100% (95% CI: 98.7-100%) and a specificity of 90.9% (95% CI: 69.4-100%) relative to qPCR. More recently, in Brazil, Soares et al. [94] developed a LAMP assay targeting the Leishmania HSP70 gene and evaluated its diagnostic accuracy in 100 skin biopsy samples (60 CL cases and 40 non-cases) against qPCR. The LAMP-Leish/HSP70 assay achieved a sensitivity of 86.7% (95% CI: 75.4-94.1%), a specificity of 100% (95% CI: 91.2-100%), and an overall accuracy of 92% (95% CI: 84.8-96.5%). Taken together, these studies underscore the potential of LAMP as a diagnostic and surveillance tool in endemic regions.

With genus-specific targets, because of their high sequence conservation among the Trypanosomatidae, some nonspecific results were observed. Adams et al. [89] reported a LAMP assay based on 18S rDNA that was unable to discriminate between infections with Leishmania, T. brucei, and T. cruzi. Since the diseases caused by these related pathogens differ in clinical presentation, differential diagnosis can be made on clinical symptoms and confirmed using species-specific LAMP assays [89]. In this study, when we tested our 18S LAMP primer set - crRNA combination, no cross-reactivity was observed with DNA from three T. cruzi strains belonging to distinct genetic lineages (discrete typing units, DTUs): TcI (Sylvio X10 strain), TcIV (VT-1 strain), and TcVI (Tulahuen LacZ clone C4 strain). These results demonstrate the high analytical specificity of the LAMP-CRISPR assay for Leishmania spp. We also noted that integrating LAMP amplification with downstream CRISPR-based detection improved discrimination between positive and negative amplification results, as shown in previous studies [37,56]. This outcome reflects the two layers of specificity intrinsic to LAMP-coupled CRISPR assays: first, the use of 4–6 LAMP primers that selectively target 6–8 distinct regions in the target DNA, and second, the crRNA-guided Cas12a-mediated target sequence recognition [23]. An optimization to improve LAMP assay specificity involves incorporating a probe-specific detection readout [18], as demonstrated by Ruang-Areerate et al. [96] in detecting asymptomatic Leishmania infections in HIV-infected patients.

The LAMP technology is commercially available as the Loopamp Leishmania Detection Kit (Eiken Chemical Co., Ltd., Japan), which does not require cold-chain storage. This kit targets kDNA minicircles and 18S rDNA in the same reaction to enable pan-Leishmania detection based on the LAMP assay developed by Adams et al. [97]. The diagnostic performance evaluation of this kit in different endemic areas around the world has demonstrated its high sensitivity and specificity for the diagnosis of VL [22,77,97–99], CL [22,97,100,101], and PKDL [102], underscoring its potential for deployment in field settings. So far, two studies have evaluated the efficacy of the Loopamp assay for CL diagnosis in the New World. Adams et al. [97] conducted a prospective cohort study on 105 clinically suspected CL patients in Colombia. Using lesion swab samples, the Loopamp assay achieved a sensitivity of 95% (95% CI: 87.2-98.5%) and a specificity of 86% (95% CI: 67.3-95.9%) compared with a composite reference standard of microscopy and/or culture. In Suriname, Schallig et al. [101] assessed 93 suspected CL cases. The assay had a sensitivity of 84.8% (95% CI: 75.0-91.9%) compared to microscopy and 91.4% (95% CI: 83.0-96.5%) compared to PCR. Specificity was moderate against microscopy (42.9%, 95% CI: 17.7-71.1%) but substantially higher when compared with PCR (91.7%, 95% CI: 61.5-99.8%).

Undoubtedly, the DNA extraction method can impact LAMP assay performance when testing clinical samples. As shown in the study by Ghosh et al. [102], a lower efficiency of LAMP to diagnose PKDL was observed when the LAMP reaction was performed using crude DNA obtained by a simple in-house boil and spin protocol compared to DNA extracted by a silica column-based (Qiagen) kit (67.2% vs. 89.7% sensitivity, respectively). Further, a recent study found a compromised sensitivity (48.4%) of the Loopamp kit for CL diagnosis in Ethiopia, possibly due to a primer mismatch with the L. aethiopica 18S rDNA target [103]. This finding points out the need to extensively validate existing LAMP assays in diverse clinical contexts and epidemiological scenarios in different geographical areas to confirm their effectiveness in detecting local parasite variants that are circulating. The same applies to our new LAMP-CRISPR assays.

Combining highly sensitive isothermal amplification with the highly specific CRISPR-Cas12a system has been applied to detect pathogenic trypanosomatids. Wiggins et al. [46] developed proof-of-concept RPA-CRISPR assays that target kDNA maxicircles for Leishmania spp. detection, and later adapted them into a multichambered, electricity-free device for diagnostic applications in low-resource settings [104]. Yang et al. [47] developed a one-tube RPA-CRISPR assay that targets the Leishmania Kmp11 gene. The assay can be completed within 50 min, with the fluorescent signal of reaction tubes observed visually under blue light illumination. The assay yielded a sensitivity of 97.9% and a specificity of 100% on simulated samples. Ortiz-Rodríguez et al. [50] developed a PCR/RPA-CRISPR platform for T. cruzi detection. Of three nuclear target genes evaluated, cytochrome B (Cytb), the 18S rRNA gene, and histone H2A, only Cytb resulted in a workable assay since no mutations were found in the chosen target site during testing of T. cruzi DNA extracted from the intestine of triatomine vectors. Furthermore, the authors designed a portable prototype fluorescence reader (TropD-Detector) suitable for low-resource settings. Gutierrez Guarnizo et al. [56] designed a LAMP-CRISPR assay targeting the nuclear HSP70 gene for T. cruzi detection in acute Chagas disease. When evaluated on samples from 100 infants born to Chagas-positive mothers in Bolivia, the assay achieved a sensitivity of 77.27% and a specificity of 100% compared to a TaqMan-based qPCR assay targeting satellite DNA.

Our study shows that LAMP-coupled CRISPR-Cas12a assays are an effective alternative for detecting Leishmania infections in various non-invasive and invasive specimens (S1 Data) obtained from human skin lesions. While previous kDNA LAMP primers targeted either the L. donovani complex [97] or L. amazonensis [88], our kDNA LAMP-CRISPR assay (specifically the crRNA design) exploited a kDNA minicircle region conserved in New World L. (Viannia) species ( [45] and S1 Fig). In addition, our pan-Leishmania 18S LAMP-CRISPR assay is expected to detect species from both the New and Old World (S2 Fig). The performance of these CRISPR-based assays, as demonstrated herein, supports future studies to validate them on a broader scale across diverse geographical regions. Currently, our proof-of-concept workflow requires controlled laboratory conditions. Future work should consider adapting these assays to meet the ASSURED/REASSURED criteria for a diagnostic test suitable for adoption across different healthcare levels, which require assays to be affordable, sensitive, specific, user-friendly, robust, rapid, as well as equipment-free (or operating with minimal equipment), deliverable, with real-time connectivity, and workable on easily collected specimens [105,106]. A major challenge for the implementation of molecular diagnostics in resource-limited settings is sample preparation [107]. Therefore, optimizing rapid DNA extraction protocols, such as boil-spin [102] and SpeedXtract (a magnetic bead-based method) [108], combined with a streamlined CRISPR assay workflow, will be key for advancing toward a near-PoC nucleic acid-based diagnostic assay for CL and other forms of leishmaniasis. Both LAMP and CRISPR reactions are compatible with freeze-drying, which enhances reagent stability for easy transport and storage at room temperature and may reduce contamination risks [109]. This approach could enable a pre-designed kit format for portable assay setup in resource-limited settings, though its impact on assay performance must be evaluated. Incorporating these refinements alongside user-friendly devices, such as LFA strips or a blue-light transilluminator, would facilitate adoption in minimally-equipped laboratories in endemic regions, and even in a mobile suitcase laboratory. The latter approach has been explored with an RPA assay in field studies conducted in Asia, showing promising performance for VL, PKDL, and CL diagnosis [108,110,111]. Another challenge for implementing these assays in field settings is ensuring continuous training and competency of personnel, particularly given the high turnover among staff. To maintain reliable performance, training programs should not only cover routine test procedures but also emphasize quality control and troubleshooting skills [112]. These training needs can be greatly reduced if the reaction setup and signal readout are integrated into a self-contained, user-friendly device [81]. Another critical factor is the cost of CRISPR-based assay development and validation. The estimated per-sample cost for our laboratory LAMP-CRISPR assays was USD 13.4 with an LFA readout and USD 11.2 with a fluorescence readout, compared to USD 18.3 for kDNA qPCR (see S2 Table). The largest cost component (USD 3.2 per reaction) derives from commercial enzymes in the LAMP master mix kit, which could be reduced through scaling. One alternative is large-scale local production of recombinant enzymes for LAMP and CRISPR reactions; however, this would require additional investment in lot-to-lot analytical validation to ensure consistent reagent performance and to meet regulatory standards for in vitro diagnostic applications. The cost of the Milenia HybriDetect LFA kit (USD 2.7 per test) may be brought down through high-volume procurement. In this context, a cost-effective strategy may involve sourcing reagents and consumables in large quantities through supply chains embedded within the healthcare system. Ensuring sustained availability and affordability in low- and middle-income countries will require the integration of CRISPR-based tools and other molecular diagnostics into national health policies.

After further optimization, streamlining, and validation, we envisage that our assays may become valuable tools for providing rapid testing results directly in affected regions, reducing delays associated with testing in centralized laboratories and supporting real-time epidemiological surveillance. Looking ahead, integrating our assays with lab-on-chip systems, such as the paper-based LAMP-CRISPR integrated diagnostic platform (PLACID) [37], could significantly enhance field applicability. PLACID uses freeze-dried reagents and paper microfluidics to perform amplification and detection without fluid transfer, with fluorescence readout via smartphone. This platform showed robust performance for bacterial and viral targets without false positives [37]. These refinements represent important considerations for future assay development. Indeed, our current two-step LAMP-CRISPR open assay format carries an inherent risk of cross-contamination during sample transfer from tube-based (LAMP) to plate-based (CRISPR) reactions, as reported in other studies [109].

In conclusion, we developed novel LAMP-CRISPR assays for the molecular detection of Leishmania, targeting two multicopy genomic regions: 18S rDNA for pan-Leishmania detection and a kDNA minicircle region conserved among New World L. (Viannia) species. By integrating LAMP with CRISPR-Cas12a technology, the assays enable accurate detection, with results readout via fluorescence or lateral flow strips, the latter aimed at facilitating a simpler workflow. Analytical validation and initial performance testing on clinical samples demonstrated high sensitivity and specificity, underscoring the assays’ potential as effective alternatives for detecting Leishmania infections. With further optimization and validation, these proof-of-concept assays hold promise as next-generation molecular tools with great potential to improve the diagnosis and surveillance of leishmaniasis in endemic regions, thereby supporting One Health strategies for disease control.

Supporting information

Acknowledgments

We thank Pohl Milón (Universidad Peruana de Ciencias Aplicadas) for inspiring discussions; Yomara Romero (Universidad Peruana de Ciencias Aplicadas) for assistance with sequence alignments and advice on their presentation; Edith Málaga and Manuela Verástegui (Universidad Peruana Cayetano Heredia) for providing genomic DNA from the Trypanosoma cruzi strains tested here; and the Office of Research at the Universidad Peruana de Ciencias Aplicadas for the support provided to carry out this research work.