Development of a Loop-mediated isothermal amplification (LAMP) technique for specific and early detection of Mycobacterium leprae in clinical samples

Leprosy, a progressive, mutilating and highly stigmatized disease caused by Mycobacterium leprae (ML), continues to prevail in the developing world. This is due to the absence of rapid, specific and sensitive diagnostic tools for its early detection since the disease gets notified only with the advent of physical scarring in patients. This study reports the development of a Loop-mediated isothermal amplification (LAMP) technique for fast, sensitive and specific amplification of 16S rRNA gene of ML DNA for early detection of leprosy in resource-limited areas. Various parameters were optimized to obtain robust and reliable amplification of ML DNA. Blind clinical validation studies were performed which showed that this technique had complete concurrence with conventional techniques. Total absence of amplification of negative control DNA confirmed the specificity of this test. Various visual detection methods viz. colorimetric, turbidity differentiation and bridge flocculation were standardized to establish easy-to-read and rapid diagnosis. This technique eliminates the lack of accuracy and sensitivity in skin smear tests in patients and the requirement for expensive lab equipments and trained technicians. The technique holds promise for further expansion and has the potential to cater to the unmet needs of society for a cheap, highly-sensitive and robust rapid diagnosis of ML.

www.nature.com/scientificreports/ up-converting phosphor (UCP)-LFA). PCR techniques and nucleic acid amplification techniques have proven to be great tools for diagnosing ML with high sensitivity and specificity. Recently, Beissner et al. 9 developed a combined RT-qPCR method which simultaneously uses two PCR assays on different genes for increased efficiency. Cabral et al. 10 detected the nasal carriage of ML in the household contacts of infected persons by using PCR. Similarly, several PCR and immunoassay based techniques have been explored to develop early and rapid diagnostic tools [11][12][13][14][15][16][17] . However, the use of these methods for rapid diagnosis of leprosy is limited due to drawbacks such as (i) lack of specificity, (ii) false negatives due to low bacterial load, (iii) false positives due to co-circulating antibodies from other infections and (iv) requirement of expensive equipments and trained labor. Even the newest detection techniques fail to detect 60% of paucibacillary ML infections 3 . As leprosy is endemic in areas where resources are poor, it is even more imperative to develop a test which, besides being sensitive and specific, is economical, rapid and user friendly. Isothermal amplification methods are favored over traditional PCR methods for developing cheap, effective and rapid diagnostics as the requirement of expensive laboratory equipments like thermocyclers is eliminated, without compromising efficiency or sensitivity. Hence, isothermal amplification techniques have better scope and potential to be used in rapid diagnostics in developing economies. Over several isothermal-based techniques, the loop-mediated isothermal amplification technique (LAMP) has many applications in the field of point-of-care (POC) testing. It utilizes four to six primers to identify specific regions of the template which can be amplified at constant temperature, using DNA polymerase with high strand displacement activity. Therefore, DNA amplification can be completed in a single step by incubating the mixture of primers with polymerase and template in a reaction buffer using a simple heat block or water bath 18 . It can typically produce results with high yield and specificity in less than 30 min at a single temperature. These features make the technique rapid and portable which further add to the purpose of on-field applications using simple devices. LAMP technique is known to be well-suited for several molecular diagnostic applications ranging from the laboratory and POC screening of infectious agents for various diseases, food testing, environmental testing, etc. [19][20][21] . Diagnostic kits and techniques using LAMP have been developed for rapid detection of several difficult pathogenic agents for highly infectious diseases like tuberculosis 22 , leishmaniasis 23 , dengue 24 , HIV/AIDS 25 , zika virus disease 26 , salmonellosis 27 , etc. More importantly, the advent of LAMP has led to improved diagnosis of neglected tropical diseases all over the world [28][29][30][31] .
The laboratory detection method for any nucleic acid amplification technique is generally post-reaction agarose gel electrophoresis. However, for on-field, simple, rapid and user-friendly endpoint detection, various amplicon detection methods have been explored by researchers that can be combined with LAMP technique to produce immediate and easy to read results. Esmatabadi et al. 32 has reviewed a variety of detection techniques such as colorimetry, turbidometry, hybridization probes, lateral flow dipsticks, ELISA, functionalized gold nanoparticles, etc. that have potential to be combined with LAMP method for easy detection of amplification products. The objective of the present study was to make the ML detection technique rapid, easy to read as well as affordable. Therefore, different detection techniques like colorimetry, turbidity evaluation and bridge flocculation methods have been used for detection of LAMP amplification products.
Since, ML is uncultivable in vitro and possesses a highly reduced genome that has a high degree of identity with MTB genome, specific LAMP technique for ML detection has not been forthcoming, despite its innately unique benefits as a diagnostic tool for disease detection in poor and remote areas of the world. To our knowledge, current study is the first report on presenting LAMP as an efficient diagnostic tool for ML.

Materials.
QIAamp Blood mini kit (Qiagen, India) was used for DNA extraction from clinical blood samples. Bacillus stearothermophillus DNA Polymerase 3.0 (New England Biolabs, UK) and primers (Integrated DNA Technologies, USA) were used for the nucleic acid amplification. Escherichia coli (strain MG1655, JH-Institute of Molecular Medicine, Jamia Hamdard, New Delhi) was used as double negative control. MTB DNA (strain H37RV, National Institute of Immunology, New Delhi) was used as negative control. Ethidium bromide (Sigma-Aldrich, USA), propidium iodide (Merck, USA) and SPRI (solid phase reversible immobilization) magnetic beads (Canvax, Spain) were used for detection of amplification products. All other chemicals, unless otherwise specified, were from Sigma-Aldrich (USA). DNA Extraction. ML DNA was extracted from SSS samples of leprosy-positive subjects using the Proteinase K method 33 . Concentration of the extracted DNA was measured using microvolume spectrophotometer (NanoDrop 3300). Blood samples from patients were obtained in EDTA tubes and 300 μl of blood samples was used for genomic DNA extraction. DNA of E.coli was extracted using phenol-chloroform method as described by F. He with slight modifications 34 . LAMP target selection and primer designing. LAMP assay utilizes three sets of primers viz. forward and backward outer primers (F3, B3), inner primers (FIP, BIP) and loop primers (LF, LB). These promote formation of 'hairpin-like' loop structures at the start of the reaction, making way for high degree of self-priming  www.nature.com/scientificreports/ DNA polymerase I and 2 μl template DNA. The reaction was incubated at 66 °C for 1 h and amplicons were initially analyzed by agarose gel electrophoresis using ethidium bromide (EtBr) staining. The method was optimized for reaction conditions such as reaction duration, temperature and heat source for the most efficient amplification conditions. LAMP sensitivity and specificity. LAMP was performed using serial dilutions of template DNA to evaluate the sensitivity. The specificity of LAMP primers was determined by using MTB and E. coli DNA as negative controls to nullify any chance of cross reactivity.
Comparison with conventional method. To determine the rapidity and efficiency of LAMP as a nucleic acid amplification technique for ML detection, a comparison was made with conventional PCR method using protocol and PCR primers described by Donoghue et al. and Truman et al. 38,45 . In brief, primers for ML-specific RLEP gene (RF: 5′ TGC ATG TCA TGG CCT TGA GG 3′, RR: 5′ CAC CGA TAC CAG CGG CAG AA 3′) along with PCR reagents (reaction buffer, Taq polymerase, dNTPs) and samples were mixed and incubated in a thermocycler (Bio Rad, USA) with cycling conditions as follows: initial denaturation at 95° C for 10 min followed by one cycle of 94° C for 2 min, 58° C for 2 min and 72° C for 2 min. This was followed by 40 cycles of 94° C for 30 s, 60° C for 30 s and 72° C for 45 s. Final extension was done at 72° C for 10 min. The amplicons were visualized by 2% agarose gel electrophoresis.
Validation using clinical samples. Blind studies were done using random blood samples from eighteen ML and MTB infected patients to validate the LAMP method of detection. Of these, four were females and fourteen were males. Samples from patients were taken before the start of any therapy. Diagnosis was made based on their symptoms, microscopic detection and SSS test. Children and pregnant ladies were excluded from the study. DNA from blood samples were isolated using DNA extraction kit and LAMP was performed along with controls. Results were analyzed using EtBr-based gel electrophoresis.
Statistical analysis. Based on observations and outcomes, the sensitivity, specificity, positive predicted value (PPV) and negative predicted value (NPV) of clinical studies were calculated to determine the efficiency of LAMP method for ML detection.

Visual detection methods. Rapidity of a technique includes fast and convenient assessment of results.
User-friendly visual detection methods for qualitative determination of DNA amplification methods were used apart from time-intensive conventional gel electrophoresis.
Colorimetric detection. DNA intercalating dye propidium iodide was used to differentiate between positive and negative reaction samples 1 μl of 1 mg/ml stock solution of the dye was used for each reaction mixture after amplification and incubated at room temperature for 1 min before visualizing under UV transilluminator.
Visual detection of turbidity. Visual as well as spectroscopic detection of turbidity which is caused by release of free pyrophosphates during DNA multiplication was performed using increasing concentrations of magnesium ions in solution, as described by Esmatabadi et al. 32 .
Bridge flocculation assay. This method employs use of SPRI (solid phase reversible immobilization) beads where amplicons, if present, bind with the nanosized-magnetic beads and form a pellet, in contrast to the dispersed beads in the absence of amplified products 46 . The procedure was followed as described by Benjamin et al. 47 with slight modification. 5 μl of amplified product was mixed and incubated with 1.  Figure 1 shows the location of the 16S rRNA-S LAMP primers in the 16S rRNA genomic sequence.
Optimization of LAMP parameters. Optimization of the LAMP method was done by varying the reaction conditions to identify a set of conditions which would provide robust amplification signals, as described below. www.nature.com/scientificreports/ (a) Reaction temperature: LAMP reactions were performed at different temperatures (65° to 70 °C) to determine the optimum temperature necessary for the amplification. As shown in Fig. 2(a), beyond 66 °C amplification efficiency was greatly reduced, followed by no amplification at all at 70 °C. The temperature associated with highest amplification signal i.e. 66 °C was chosen for our method.   to identify the feasibility of the assay in poor resource settings. Figure 2(d) shows the comparison of heat sources depicting similar amplification results in both cases. (e) Specificity determination The main aim of our study was to develop a LAMP detection method specific to ML to avoid cross-detection of MTB, a common and persistent technical obstacle. Our LAMP primers did not show any amplification signal with MTB DNA (Fig. 2e). Even at concentrations of 20 ng/μl of template MTB DNA, no amplification was observed, confirming the high degree of specificity of our ML-specific primers. Cross reactivity for primers was tested using E. coli as double negative control in each study.
Standardization of intensity vs. concentration. The intensity of amplification was detected by spectrophotometrically measuring absorbance of tenfold serial dilutions of amplified DNA at 400 nm. A calibration curve (Fig. 3) was obtained showing increasing amplification intensity with rise in DNA concentration. The R 2 value of the curve was determined to be 0.995. Linear curve fitting was done to fit the amplification intensity response on increasing the input concentration. The equation for the best fit line was x = [(y − 0.018)/0.473] where x represents the input DNA concentration and y represents the absorbance at 400 nm, indicating the intensity of amplification.

Evaluation of LAMP technique in clinical samples.
To validate the technique in clinical samples, blind studies were performed using random blood samples from patients diagnosed with either ML or MTB. Blood samples were selected for evaluation of this assay instead of skin smears so as to make this technique less painful and more convenient. Eighteen clinical samples were assessed with the LAMP technique and the results were cross-validated with the conventional detection method, as described in Materials and Methods. DNA extracted from blood samples, along with positive and negative controls, were screened for LAMP amplification (Fig. 4). Around twelve samples were found to be positive and six negative for ML by LAMP method. The same samples were tested by conventional method and the observation details along with results are given in Table 2.
The results show complete diagnostic concurrence between LAMP technique and conventional technique, with 100% clinical specificity. The PPV and NPV too was 100% in LAMP as compared with conventional method.
Comparison of LAMP with regular PCR. So far, PCR has been seen as the only nucleic acid amplification technique available for molecular diagnosis of ML. We compared the efficiency and sensitivity of our LAMP technique with traditional PCR method. Figure 5 shows that LAMP gave similar results as regular PCR and, in fact, better visual bands in 60 min using a simple water bath. In contrast, PCR took more than two and a half hours using a thermocycler for detectable bands. This implies that LAMP is more efficient with regards to rapidity, resource utilization, and ease of detection as compared to conventional PCR method. In recent years, several PCR methods with varying sensitivities and specificities using different target sequences have been published. A side-by-side comparison was made ( Table 3) between some of those published studies with our LAMP technique of detection. It can be noted that LAMP has the quickest turnaround time to amplify the bacterial DNA in clinical blood samples and has eliminated the risk of false positives or false negatives. In Table 3, many of the references have used biopsy samples which are painful to collect and take 2 h www.nature.com/scientificreports/ or more for target amplification. Moreover, they were unable to achieve high sensitivity and specificity in their studies. The method by Beissner et al. reported comparable sensitivity and specificity but it used a combination of two techniques which made the procedure more complex and cost-and time-intensive. Therefore, it can be concluded that the LAMP method has the potential to be developed for clinical use for its rapidity, portability, high sensitivity and specificity.   www.nature.com/scientificreports/

Methods for detection of LAMP amplicons.
(a) Colorimetric detection: Apart from conventional gel electrophoresis for detecting DNA amplification, other detection methods for visual identification of amplified products were also used in order to enhance the ease and cost of operation. Propidium iodide was used to qualitatively identify positive LAMP amplification reactions without gel electrophoresis. This fluorescent dye intercalates between DNA strands and gives fluorescence under UV wavelength. Figure 6 shows propidium iodide dye detection of LAMP products where positive sample with amplified product fluoresced bright pink while negative samples appeared orange in color. (b) Turbidity visualization: During the amplification reaction, pyrophosphate ions are produced which, in the presence of magnesium ions, form magnesium pyrophosphate that precipitates in solution. The resulting turbidity can be visually detected. Difference in visual turbidity was observed between positive samples and negative controls (Fig. 7), constituting another basis of visual detection of amplification results. (c) Bridge flocculation method: This method is generally used to purify DNA amplicons by using SPRI magnetic beads. But here, it is re-purposed as a detection technique to determine the presence of LAMP amplification products. In the presence of magnetic field, the beads aggregate and amplified DNA get entangled with the beads forming floccules which prevent their re-dispersion after removal of the magnetic field. In the absence of amplification, beads re-disperse in the solution after removal of the magnetic field. Figure 8(a) shows the visual difference between positive and negative LAMP reactions using SPRI beads. Sensitivity of this detection method was also determined by using different dilutions of DNA for LAMP amplification and performing bridge flocculation assay for detection of products. Figure 8(b) shows visible flocculation with as low as 5 pg/μl template concentration. This shows the potential of this detection assay in combination with LAMP technique for easy and rapid point-of-care detection of ML DNA.

Discussion
Early and accurate detection of ML is crucial not only for proper treatment regimen but also to lower the burden and transmission in endemic areas, which in almost all cases are in socioeconomically-deprived and accesslimited regions of the world. Furthermore, specificity of ML detection is often compromised due to the presence of homologous sequences in other Mycobacterium species, most notably MTB. We report the development and standardization of a rapid, sensitive and economical method of detecting ML DNA using an optimized LAMP technique. Primers were designed that specifically amplified target regions of 16S rRNA gene of ML these primers could differentiate ML from closely-related MTB samples, forming the basis of a sensitive and specific diagnostic Additionally, this technique can be combined with rapid detection methods like colorimetry, turbidometry and bridge flocculation assay for naked-eye visualization of positive reaction products, further making this diagnostic technique for point-of-care use easily readable and user-friendly. In resource-poor settings where rapid, inexpensive and on-site testing is required to detect the potential source of infection and transmission, this method could be utilized for improved and simplified monitoring of potential leprosy patients at early stages of infection. Besides early diagnosis of infection, this technique may prove to be useful in monitoring the reduction in bacterial load during the course of multi-drug therapy in infected patients.  www.nature.com/scientificreports/ Also, this technique could be applied for transmission studies to investigate human-to-human transmission in household and community contacts. Currently, leprosy is preliminarily diagnosed through observation of physical manifestations and slit skin smear test. Apart from being painful, skin biopsy-based tests are lengthy, lack sensitivity, require an equipped lab and trained personnel and often result in false positives or false negatives. Sophisticated molecular diagnostic techniques like quantitative real-time PCR are not suitable alternatives for simple, cheap and reliable on-field testing tools. Unlike most other pathogens, detection of ML is difficult due to the fact that the bacillus cannot be cultured in vitro. The LAMP technique for ML detection using blood samples and readily available, simple instruments alongside quick visible detection methods we have described in this study, has the scope to be further developed as a point-of-care testing unit in field settings that can complement or even replace traditional diagnostic methods.