MALARIA DIAGNOSIS BY LOOP-MEDIATED ISOTHERMAL AMPLIFICATION (LAMP) IN THAILAND

Abstract

The loop-mediated isothermal amplification method (LAMP) is a recently developed molecular technique that amplifies nucleic acid under isothermal conditions. For malaria diagnosis, 150 blood samples from consecutive febrile malaria patients, and healthy subjects were screened in Thailand. Each sample was diagnosed by LAMP, microscopy and nested polymerase chain reaction (nPCR), using nPCR as the gold standard. Malaria LAMP was performed using Plasmodiumgenus and Plasmodium falciparum specific assays in parallel. For the genus Plasmodium, microscopy showed a sensitivity and specificity of 100%, while LAMP presented 99% of sensitivity and 93% of specificity. For P. falciparum, microscopy had a sensitivity of 95%, and LAMP of 90%, regarding the specificity; and microscopy presented 93% and LAMP 97% of specificity. The results of the genus-specific LAMP technique were highly consistent with those of nPCR and the sensitivity of P. falciparum detection was only marginally lower.

INTRODUCTION

Malaria is a major cause of morbidity and mortality, leading to approximately 198 million cases and 584,000 deaths in 20131. The disease is endemic in a broad band around the equator, placing approximately half of the world's population at risk of infection1.

Several strategies are currently being used to strengthen malaria control and to optimally lead to malaria elimination. Such reasoning is based on an accurate and prompt diagnosis, measuring the impact of any intervention, as well as being a prerequisite for an effective treatment with anti-malarial drugs2, especially for the potentially fatal cases of Plasmodium falciparum infections3. Due to the WHO recommendation of using microscopy and malaria rapid diagnosis tests (RDT) in all suspected malaria cases, they are widely applied for instant parasitological confirmation, especially in regional clinics in endemic areas4. However, to obtain acceptable accuracy of the microscopic results, lengthy training sessions and experience are essential.

The molecular diagnostic tools applying DNA amplification have the advantage of distinguishing between similar appearing species and real-time amplification techniques are even able to quantitate the amount of parasites, providing a rapid processing of the samples5 ^, 6. Nested PCR is considered the most sensitive and specific tool for malaria diagnosis7. Several studies have shown that the molecular methods detect up to eight times more Plasmodium spp. infections than microscopy, and up to one third of these are mixed infections8 ^, 9. Therefore, the interpretation of malaria epidemiology has been affected by molecular tools, for instance by revealing large reservoirs of asymptomatic infections10, by detecting a shift in age distribution of Plasmodium spp. infections11, by facilitating the automation and standardisation along with the ability to differentiate species and detect drug resistance12 ^, 13. However, disadvantages of nested PCR are the high costs of sophisticated equipment such as the thermal cycler, the time-consuming procedure which delays the release of results to the physician and the need for well-trained laboratory staff. Due to limited economic resources, this diagnosis is not applicable in many endemic areas14.

The loop-mediated isothermal amplification (LAMP) is a recently developed molecular technique for nucleic acid amplification and it was designed to overcome two disadvantages of PCR by being simpler and faster, while still providing a high level of accuracy. Briefly, a set of four specifically designed primers to recognise six distinct regions of the target DNA as well as a Bst polymerase are used for auto-cycling strand-displacement DNA synthesis. The amplification is conducted under isothermal conditions. Therefore, the cost of the technique can be reduced by minimally using a water bath or a heat block15. The method can also be less expensive if the assay is conducted with heat-treated blood samples instead of purified DNA16. LAMP accumulates approximately 109 copies of target DNA within a time frame of less than an hour17 ^, 18. Due to the relative stability of the LAMP reagents at 25 °C and 37 °C, it holds the potential for field application in tropical countries19. In an area of low endemicity, a portable LAMP-based assay called RealAmp (Real-time Fluorescence Loop-Mediated Isothermal Amplification) has been validated for the detection of submicroscopic infections20.

Therefore, it is considered a promising candidate to be used as a diagnostic tool in field studies and in regional clinics, especially in endemic areas where costs must be reduced.

The aim of the present study was to apply the LAMP assay for malaria diagnosis ofP. falciparum LAMP results were compared to the standard detection methods, microscopy and nested PCR in order to evaluate the sensitivity and specificity.

MATERIALS AND METHODS

Patients and biological material collection. The collection of samples and their analysis was performed from May to July 2011 at the Regional Medical Sciences Center 3, Department of Medical Science, Ministry of Public Health Medical Science, Chonburi, Thailand. DNA was provided by the DNA bank of the Science Center. The samples were previously obtained from various clinics of the mainly central Thailand provinces in May and June 2011, they were received during this period, randomly selected and included in our study. One hundred and thirty two of the samples were obtained from patients following hospital admission, who were either diagnosed with malaria due to a positive microscopy (118 samples) or by signs/ symptoms suggestive of malaria, as observed from the clinical history and geographical origin, but they later proved to be negative by microscopy (14 samples). EDTA-blood samples were taken before patients were treated with anti-malarials. In addition, 18 malaria free blood samples were obtained from the blood bank of Chonburi, drawn from clinically healthy subjects. Only these samples were not analysed by microscopy in the current study.

This study was performed using a protocol approved for medical research on human subjects, Department of Medical Sciences, Ministry of Public Health, Thailand. Informed consent was obtained from all the human adult participants or from the parents or legal guardians of minors.

Microscopy. For the analysis of the samples by microscopy, the hospital's staff collected EDTA-blood samples from the patient's forearm vein upon admission. A drop of this blood sample was placed upon a glass slide and both thin and thick films were prepared21. Following this, they were incubated with the Wright stain for 3 min and finally rinsed with water. The hospital's experienced laboratory staff examined the slides under a light microscope using the 100x oil immersion objective, screening more than 100 fields per slide at a pace of approximately 6 min/slide. The microscopy results were compared to the molecular methods (nPCR and LAMP).

DNA preparation. Template DNA for nested PCR and LAMP assays was extracted from 200 µL of EDTA whole blood using the QIAamp DNA Blood Mini Kit (QIAGEN GmbH, Hilden, Germany), according to the manufacturer protocol. The resulting 200 µL aliquots of template genomic DNA was stored at -20 °C.

Nested PCR assay based on 18S rRNA gene for Plasmodiumspecies. All 150 samples were tested for the genusPlasmodium, P. falciparum and P. vivax by nested PCR. The species-specific nucleotide sequences of the 18S rRNA genes were applied as described previously by Kimura et al.22, with slight modifications. For the outer PCR, 1 µL of template DNA was added to a 21 µL PCR mixture that consisted of 0.88 µM of each universal primer (P1 forward and P2 reverse primer) (10 mM each), 0.44 µM deoxynucleotide triphosphate (10 mM each), 1.5 µM MgCl2 (25 mM), 2.2 µM 10x ImmoBuffer (100 mM NaCl, 20 mM Tris-HCl, pH 7.5; 0.1 mM EDTA, 2 mM DTT, 50% Glycerol, stabilizers), 0.25 U IMMUNLASE DNA Polymerase (5 U/µL) and 14.85 µL of distilled water. The dNTP, MgCl₂, ImmoBuffer and IMMUNLASE DNA polymerase were obtained from Bioline USA Inc. (Taunton, Ma, USA). The DNA amplification was carried out under the following conditions: 94 °C for 10 min and then 35 cycles at 92 °C for 30 s., 60 °C for 1.5 min, and 72 °C for 1 min, followed by a final extension at 72 °C for 5 min. The outer PCR product was diluted 40-fold in sterile water. One microlitre of this solution was used in the second amplification. The inner PCR was performed at 94°C for 10 min and then 20 cycles at 92 °C for 30 s, 60 °C for 1.5 min, 72 °C for 1 min, followed by a final extension at 72 °C for 5 min with the P1 forward primer in combination with each species-specific reverse primer. The amplified products were analysed in 2.5% agarose gels stained with ethidium bromide. The expected band sizes were approximately 160 bp for the first PCR product and approximately 110 bp for the second. The DNA ladder was from Promega (Madison, WI, USA). The WHO considers nPCR to be the most sensitive and specific diagnostic tool in malaria diagnosis7 ^, 23, for this reason it was used as a reference standard for comparison in this study.

LAMP assay based on 18S rRNA for the detection of the Plasmodium genus and P. falciparumspecies. All 150 collected blood samples were tested by a primer set forPlasmodium DNA and a species-specific P. falciparum. One sample positive for P. falciparumnested PCR served as a positive control. A sample of human DNA from a healthy person, obtained from the blood bank of Chonburi served as a negative control. The LAMP assays as performed in this study, were firstly reported by Poon et al.16, and Han et al.18 (Table 1). Few modifications had to be applied as follows: in the LAMP reaction for genus Plasmodium, 1 µL of template DNA was added to a 19 µL of LAMP mixture, containing 6.3 µL of distilled water, 1.3 µL of primer mix, 10.4 µL of 2x Reaction LAMP buffer and 1 µL of Bst-DNA-polymerase (BioLabs Inc., Ipswich, MA, USA). For P. falciparum amplification slight modifications had to be applied: 1 µL of Betain was added, the 2x reaction LAMP buffer was adjusted and distilled water was reduced to 4.74 µL. The reactions were incubated at 65 °C for 60 min; the amplification was performed in a thermal cycler. All tubes were analysed by gel electrophoresis to assess the presence of the LAMP DNA product. Therefore 5 µL of each labelled LAMP product were mixed with 1 µL of loading buffer (RBC BioScience Corp., Taiwan) and added to each lane of 2.0% agarose gel. The gel electrophoresis was run for 50 min at 100 V in 1x Tris-Acetate-EDTA (TAE) buffer and afterwards stained with ethidium bromide (Promega, Madison, WI, USA). The molecular tests were performed by a well-trained staff closely supervised by a medical technologist with long lasting experience. This laboratory personnel has been blinded from the results of the nested PCR while performing the LAMP assays to avoid bias.

Table 1. Primer sets used for amplification of 18S rRNA genes by LAMP.

Species	Primer	Sequence (5'-- 3')
genus Plasmodium	F3	GTATCAATCGAGTTTCTGACC
	B3c	CTTGTCACTACCTCTCTTCT
	FIP (F1c-F2)	TCGAACTCTAATTCCCCGTTACCTATCAGCTTTTGATGTTAGGGT
	BIP (B1-B2c)	CGGAGAGGGAGCCTGAGAAATAGAATTGGGTAATTTACGCG
	LPF	CGTCATAGCCATGTTAGGCC
	LPB	AGCTACCACATCTAAGGAAGGCAG
P. falciparum	F3	TGTAATTGGAATGATAGGAATTTA
	B3c	GAAAACCTTATTTTGAACAAAGC
	FIP (F1c-F2)	AGCTGGAATTACCGCGGCTGGGTTCCTAGAGAAACAATTGG
	BIP (B1-B2c)	TGTTGCAGTTAAAACGTTCGTAGCCCAAACCAGTTTAAATGAAAC
	LPF	GCACCAGACTTGCCCT
	LPB	TTGAATATTAAAGAA

Statistical methods. GraphPad Software 2005-2009 (GraphPad Software, Inc., La Jolla, California, USA), an online statistical calculation program, was used to calculate test performances and acceptability evaluation indices using nested PCR results as the standard24. The 95% confidence interval for sensitivity and specificity was calculated by the modified Wald method25.

RESULTS

Overall, the prevalence of Plasmodium species in the investigated samples. Out of 150 collected samples, 132 samples were examined by all three tests. The remaining 18 control samples were examined only by nPCR and LAMP. APlasmodium infection was not found in any of these samples.Table 2 shows the results of the prevalence of genus Plasmodium, and of P. falciparum. In the 132 samples, nPCR detected 117 ( 88.6%), microscopy 118 (89.4%), and LAMP 116 ( 87.9%) positive samples for the genusPlasmodium. Microscopy showed one false-positive and LAMP showed one false-negative.

Table 2. Comparison of nPCR, microscopy, and LAMP for the Plasmodium genus detection and species identification*.

Parasite(s) detected by each method (no. of samples) †
nested PCR	Microscopy	LAMP
genus Plasmodium(117)	genus Plasmodium(118)	genus Plasmodium (116)
		negative (1)
P. falciparum(77)	P. falciparum(72)	P. falciparum (69)
		negative (4)
	P. vivax (5)‡	P. falciparum (4)
negative (15) ¶ negative (18)°	negative(15)	negative(15) negative (18)

*n PCR=nested polymerase chain reaction; LAMP = Loop-mediated isothermal amplification. † Each row shows results obtained from identical blood samples. Microscopy + LAMP results that were not concordant are shown in bold. ‡ Four of these samples were not detected due to mixed infections. ¶ Samples of patients with fever of unknown origin. Samples provided by the blood bank

Regarding P. falciparum, nPCR detected 77 positive samples (58.3%). Among these there were four samples with mixed infection, also containing P. vivax. Microscopy also detected 77 P. falciparumpositive samples (58.3%), including two false-positive and two false-negative results. The false-negative samples were found to be of mixed infection by nPCR, but were only diagnosed positive for P. vivax by microscopy. Similarly, for the other four samples of mixed infection, P. vivax was also dominant by microscopy, and diagnosed as P. falciparum only. LAMP also detected 77 positive samples (58.3%), including, however, four false positive and four false negative results.

Concerning P. vivax, nPCR detected 46 positive samples (34.8%), including two samples that also contained P. falciparum. Microscopy detected 42 (31.8%) positive results, including two samples that also containedP. falciparum that were not detected and one false positive result. All of the 15 samples of patients with fever of unknown origin were tested negative by all diagnostic tools.

Comparison of sensitivity and specificity, of microscopy, nested PCR and LAMP is shown in Table 3: the 117 nPCR-positive samples for the genus Plasmodium were also detected by microscopy (sensitivity of 100%; 95% CI: 96.1-100%), 116 were detected by LAMP (sensitivity of 99%; 95% CI: 94.8-99.9%). All 15 nPCR-negative samples for the genusPlasmodium were also negative by the LAMP assay (specificity of 100%; 95% CI: 76.1-100%), by microscopy 14 were negative (specificity of 93%; 95% CI: 68.2-100%).

Table 3. Sensitivity, specificity of microscopy and LAMP for genus Plasmodium and P. falciparumdetection. Nested PCR results were used as the reference (gold standard) for comparison.

	Method	Sensitivity	95% CI Sensitivity	Specificity	95% CI Specificity
Genus Plasmodium	Microscopy	100%	96.-1-100%	93%	68.2-100%
	LAMP	99%	94.8-99.9%	100%	76.1-100%
P. falciparumspecies	Microscopy	95%	87.0-98.4%	97%	87.0-99.7%
	LAMP	90%	80.6-94.9%	93%	82.3-97.6%

Regarding the 77 nPCR-positive samples for P. falciparum, 72 were positive by microscopy (sensitivity of 95%; 95% CI: 87.0-98.4%), and 69 by the LAMP assay (sensitivity of 90%; 95% CI: 80.6-94.9%). Among the 55 nPCR-negative samples for P. falciparum, LAMP was also negative in 51 (specificity of 93%; 95% CI: 82.3-97.6%), and microscopy in 53 nPCR-negatives (specificity of 97%; 95% CI: 87.0-99.7%).

In all 6 samples containing mixed infections, microscopy detected only one Plasmodium species.

DISCUSSION

The risk of malaria transmission is considered to be low in most central parts of Thailand, but remains to be a burden in rural, forest areas at the border of Myanmar and Cambodia. Multidrug-resistant P. falciparum, in particular, is constraining malaria control programs. In order to provide an accurate treatment, clinical suspected malaria cases need to be confirmed by adequate laboratory diagnosis. In this study we compared a basic LAMP assay for malaria diagnosis of P. falciparum infections, to microscopy and to nested PCR as the gold standard.

Since the initial description of the LAMP method, genus and species-specific LAMP assays for the malaria parasites that infect humans have been developed18 ^, 26. Several studies currently working on the establishment of routine diagnosis are highly favouring this method27 ^, 28 ^, 29. The LAMP technique aims at combining an accuracy close to PCR with basic reagents, low technical requirements, and accomplishment by minimally-trained health workers. A rapid turnaround time produces results in about one hour30. Furthermore, recent techniques such as a variety of non-instrumented nucleic acid amplification (NINA) heater configurations have been developed for assays such as LAMP, even providing independence from electricity and/or instrumentation31 ^, 32.

The genus-specific LAMP assay in our study provided highly consistent results, concurring with those of a previous study using the same primer set33. However, the species-specificPlasmodium detection performed marginally less accurately, presenting lower sensitivity and specificity than microscopy. This test yielded lower results in comparison with those of a previous study that used the same LAMP assay for malaria diagnosis in Northern Thailand and had detected falciparum malaria parasites in 48 out of 48 nPCR positive samples (100% sensitivity)34. In that study, a real-time coupled to a turbidimetric assay was used, possibly improving the accuracy.

The study held by Patel et al.20, evaluated the RealAMP system in low transmission areas of India and Thailand. The study group from India presented a similar study group size (141 patients) compared to our study. The test presented 95% of sensitivity and 100% of specificity compared to nPCR, respectively. For low-density asymptomatic infections in the Thai study group, this system showed an explicit higher sensitivity than microscopy. In the RealAMP protocol, DNA is extracted from dried blood spots, real-time detection is carried out with fluorescence, thereby simplifying and shortening the time of analysis.

A possible explanation for the four cases of false-positive P. falciparum results via LAMP should be contamination (carry over) due to the high sensitivity of the LAMP assay. Despite preparing the reagent in a separated location and working according to standards to minimize the risk of contamination, it might have occurred when the post-LAMP microtubes were opened. To reduce the risk of contamination a LAMP combined with DNA filter paper (FTA card) for the diagnosis of P. falciparum has been developed, using a melting curve analysis instead of gel electrophoresis to visualize specific amplicons35. Furthermore, several methods of visual detection of the post-LAMP product in closed tubes without the need of opening them have been employed33 ^, 36 ^, 37 ^, 38. In order to increase sensitivity, the use of mitochondrial targets for LAMP-based detection of any Plasmodium genus parasite and of P. falciparumin particular, as recommended by Polley et al.39, could be applied instead of the 18S rRNA Plasmodiumgene. It has been claimed that mitochondrial primer sets can detect as few as 5 parasites per µL, opposed to around 100 parasites per µL as applied in this study. Incorporated into a LAMP kit the time to release a final result can be reduced, still providing similar accuracy to nPCR40. The detection of low parasitaemia favours the LAMP method as a point-of-care test for the treatment and follow-up of patients41. But taking in account that the LAMP technique has been proven to be very sensitive in other studies34, it should also be considered that some of the false positive results by LAMP might have been real positives.

The application of the molecular tests in this study had some limitations, resulting firstly from the fact that the DNA for both PCR and LAMP was extracted with a commercial kit and secondly because both techniques were coupled to electrophoresis. This resulted in a high workload, the need for well-trained staff and the requirement of special reagents as well as equipments. The costs were not lowered. The turnaround time remained high because the genus-specific primers were firstly tested and afterwards each sample was screened with the species-specific primers. Therefore, the disadvantages of the PCR method could not be solved by the LAMP method in the present study, but these limitations can be eliminated in future investigations42.

Due to the fact that microscopy is the standard method of malaria diagnosis, it has been applied for comparison in this study. However, microscopy has the disadvantage of a poor performance in low parasitaemia cases, as well as when the patients have already been treated or have taken anti-malarial drugs. Patients previously diagnosed with malaria and treated with anti-malarials within a week before admission, were excluded from this study. In this study, the examiner did not diagnose by microscopy the six mixed infections identified by nested PCR, and has also detected two false-positive results for P. falciparum and one false-positive result for P. vivax (data not shown ). These are common problems associated with microscopy. Both, the non-detected mixed infections, as well as the over-reporting of positive findings, lead to an either insufficient treatment or over-prescription of antimalarial drugs, thereby delaying the differential diagnosis of other febrile illnesses33. In addition, microscopy yielded results similar to those of nPCR. However, the results obtained from the various hospitals have probably limited the comparison with other methods due to the existence of differences in skill, concentration, and motivation of the microscopists. Therefore, the validity of such work, should be optimized in clinical trials in which microscopy will be strictly standardised. However, this effort is rarely found in most settings of malaria diagnosis.

The LAMP assay is a promising methodology for molecular diagnosis and molecular screening of malaria.The significant utility of the LAMP assay for thePlasmodium genus-specific detection developed by Han et al.18, has been confirmed. Hence, it could be used as a diagnostic tool for malaria infection instead of a standard PCR detection. In conclusion, implementation of innovative research is necessary to ensure a more accurate molecular diagnosis of falciparum malaria in a resource limited setting.

CONCLUSIONS

Any method with a potential to improve the malaria diagnosis will be beneficial to human health and it is worthy and valuable to work in the field. The development of simple and effective molecular methods with the characteristics of LAMP for malaria diagnosis and Plasmodium infections in humans and mosquito vectors will be useful. Implementation of innovative research is necessary to ensure a more accurate molecular diagnosis of Plasmodium malaria in a resource limited setting.