http://orcid.org/0000-0001-8164-3523
Figures
Abstract
Leptospirosis is a zoonosis with a worldwide distribution, caused by pathogenic spirochetes of the genus Leptospira. The classification and identification of leptospires can be conducted by both genotyping and serotyping which are time-consuming and established in few reference laboratories. This study used matrix assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) as rapid and accurate tool for the identification of leptospires. The whole cell protein spectra of 116 Leptospira isolates including 15 references Leptospira spp. (pathogenic, n = 8; intermediate, n = 2; non-pathogenic, n = 5) and 101 Leptospira spp. clinical isolates was created as an in-house MALDI-TOF MS database. Ninety-seven clinical isolates from Thailand and Laos was validated with these protein spectra and revealed 98.9% correct identification when compared with 16S rRNA gene sequences method. Moreover, MALDI-TOF MS could identify spiked leptospires whole cell in urine. Biomarkers for differentiation of leptospires phylogeny and specific protein spectra for most found Leptospira spp. in this area (L. interrogans, L. kirschneri, L. borgpetersenii) based on MALDI-MS algorithm were demonstrated.
Author summary
Leptospirosis is a world-wide zoonosis caused by pathogenic members of the genus Leptospira. Severity of infection ranges from mild features to multi-organ failure and death, although the reason for this is unclear. Clinical symptoms are similar to several other febrile illnesses such as dengue, scrub typhus, malaria, viral hepatitis. Identification of Leptospira spp. is generally based on genotyping and serotyping. This study used matrix assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) for rapid and accurate identification of Leptospira whole cell. The methods generated protein spectra of Leptospira isolates and created large protein database which can be used to compare and identify other Leptospira isolates. The machine learning based on the computational mathematic analysis of this database provide the classification of Leptospira phylogeny and biomarkers for species identification.
Citation: Sonthayanon P, Jaresitthikunchai J, Mangmee S, Thiangtrongjit T, Wuthiekanun V, Amornchai P, et al. (2019) Whole cell matrix assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) for identification of Leptospira spp. in Thailand and Lao PDR. PLoS Negl Trop Dis 13(4): e0007232. https://doi.org/10.1371/journal.pntd.0007232
Editor: Tao Lin, Baylor College of Medicine, UNITED STATES
Received: November 15, 2018; Accepted: February 11, 2019; Published: April 10, 2019
Copyright: © 2019 Sonthayanon et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data are available upon request to the Mahidol Oxford Tropical Medicine Research Unit Data Access Committee (http://www.tropmedres.ac/data-sharing for researchers and following the Mahidol Oxford Tropical Medicine Research Unit data access policy (http://www.tropmedres.ac/_asset/file/data-sharing-policy-v1-1.pdf). Queries and applications for datasets should be directed to Rita Chanviriyavuth (rita@tropmedres.ac).
Funding: This research was supported by Thailand Research Fund (Grant no. RSA5580003) and Wellcome Trust (Grant no. 089275/Z/09/Z). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Leptospirosis is a major public health problem in the developing countries. More than 1 million cases occur worldwide per year, including about 58,900 deaths [1]. In Thailand, prior to 1996, reported cases numbered approximately 200 cases per year, mainly in the central and southern regions. A dramatic increase occurred from 358 cases in 1996 to 14,285 cases in 2000, followed by a continual decline to 2,868 cases in 2005. Reported cases has stabilized at around 2,800–5,500 cases per year during the last 10 years [2]. Although the trend over time is consistent with a sustained outbreak, one of the problems associated with the interpretation of reported cases is that reporting can be done on the basis of a clinical diagnosis alone. Accuracy of clinical diagnosis varied from 0% to 50% between the provinces and was highest during the rainy season [3]. The current laboratory tests available for the diagnosis of leptospirosis are serological testing and molecular-based diagnosis method [4–7]. Matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS) is a novel method that is increasingly used for identification and classification of microorganisms from different genera, species and strains including Leptospira spp. [8–10]. The method detects the mass-to-charge ratio of biomolecules from whole cell bacteria and provides spectra within minutes. The protein mass spectra obtained can be added in the MALDI-TOF MS database and used as reference for identification of unknown microorganisms. However, none of known Leptospira mass spectra database are commercially available. Here, we employed MALDI-TOF MS to reveal the whole cell protein mass spectra of Leptospira spp. including pathogenic, intermediate and saprophytic groups. A large Leptospira spp. mass spectra database was established for identification of Leptospira spp. from clinical isolates in Thailand and Laos.
Material and methods
Bacterial strains
Fifteen references Leptospira spp. including pathogenic, intermediate and non-pathogenic groups (Table 1); 101 of Leptospira spp. isolates and one isolate of Turneriella parva, a spirochete closely related to Leptospira, were included in this study (Table 2). Leptospira were cultured in Ellinghausen-McCullough-Johnson-Harris liquid medium containing Leptospira Medium Base EMJH (Difco, Becton Dickinson, USA), Leptospira Enrichment EMJH (Difco) with 3% normal rabbit serum (Invitrogen, USA) at 30°C. All Leptospira from human clinical samples were isolated and cultured as described [11]. Serovar identification was performed using cross-agglutination absorption test (CAAT) at WHO/FAO/OIE Collaborating Center for Reference and Research on Leptospirosis, Brisbane, Queensland, Australia. All clinical samples were identified at specie level based on sequencing of 16S rRNA gene as described [12, 13].
Ethics statement
The studies were approved by the National Ethics Committee for Health Research, Ministry of Public Health of Laos (No 25/NECHR), the Oxford Tropical Ethics Committee, UK and the Ethical Committee of Faculty of Tropical Medicine, Mahidol University, Thailand (Approval no. MUTM-EXMPT 2015–004). Urine sample from healthy donor was provided by the project investigator (Approval no. MUTM2018-054-01).
Sample preparation for MALDI-TOF Mass spectrometry
Leptospira was cultured for a week to obtain 108 colony forming unit (CFU) per ml as previously described [14]. Three milliliters of culture were centrifuged at 13,000 g for 2 min at room temperature and cell pellet was washed with 70% ethanol and centrifuged at 13,000 g for 2 min. The pellet was mixed with 50 μl sinapinic acid matrix solution (Bruker Daltonics, Germany) containing 10 mg of sinapinic acid in 1 ml of 2.5% trifluoroacetic acid (Sigma-Aldrich, USA) and 50% acetonitrile (Sigma-Aldrich, USA). After being well mixed by pipetting and centrifuged at 13,000 g for 1 min, then 2 μl of cell-matrix suspension were added to the ground steel MALDI target plate and dried at room temperature. Each isolates were spotted 24 dots onto the MALDI target plate to test technical replication. Mass spectra were collected using a Ultraflex II MALDI-TOF/TOF mass spectrometer (Bruker Daltonic, USA) operated with FlexControl software in linear positive mode, i.e. using a mass range of 2,000 to 20,000 Daltons. The instrument was externally calibrated with E. coli strain DH5α ribosomal proteins as recommended protocol.
Protein spectra database of Leptospira spp.
For the 15 reference Leptospiral strains, each individual mass spectrum was analyzed and adjusted for smoothness and baseline using FlexAnalysis software 3.0 (Bruker Daltonics, USA). For each database entry, at least 20 individually measured mass spectra were imported into the MALDI Biotyper 2.0 software. A reference protein main spectra profile (MSP) was calculated and created by following the manufacturer’s recommendations for Ultraflex II measurement and the MALDI Biotyper 2.0 software package. The MSP of 101 Leptospira isolates were acquired and added to MALDI Biotyper database for Leptospira species identification. The 97 Leptospira clinical isolates were blindly identified using Biotyper based-on MSP pattern matching. MSP of each isolates were compared to 101 reference mass spectra in the database and calculated a score value between 0 and 3 reflecting the similarity between isolate and reference spectrum. The results were recorded as identification score and displayed 10-best matching in reference pattern in ranking table. Identification scores ≥ 2.0 were regarded as accepted for a reliable identification at species level and score ≥ 1.7 and ≤ 2.0 for identification at genus level. Scores < 1.7 were considered unreliable as recommended by manufactural.
The reproducibility of MSP was performed. Three reference isolates including L. interrogans, L. borgpetersenii, and L. biflexa were cultured 3 times, MALDI-TOF MS analysed and interpreted using identification score. The MSP reproducibility was also evaluated on different media and different sub-culture times using low passage L. wolffii from 3 human isolates (strain H8, H9, H16) which had been kept in liquid nitrogen for 14 years. Those isolates were grown in 2 different media (EMJH with 3% normal rabbit serum, EMJH without normal rabbit serum) and 10 sub-culture times. L. wolffii from each media and culture time were subjected to MALDI-TOF MS after protein extraction. The identification scores were analyzed using ANOVA and student t-tests.
In order to test the detection limit of MALDI-TOF MS method, L. interrogans MD08 were cultured to reach 108 CFU/ml. After protein extraction, the mixture was diluted to 10-folded serial dilutions to 10 CFU/ml and applied to MALDI-TOF MS.
MALDI-TOF MS analysis
The statistical analysis was performed using ClinProTools software version 2.2 (Bruker Daltonics, Germany). Twenty spectra of individual isolates were analyzed and visualized for spectral comparison and automatically displayed in a three-dimensional plot based on the Principal Component Analysis (PCA) to discriminate among the analyzed strains.
To discover discriminating peaks or biomarkers specifically for analysis of leptospires phylogeny (n = 51), Leptospira species (n = 40), and leptospires serotype (n = 43), four technical replicate spectra of each isolate for training set, as listed in Table 3, were used to generate mathematical models with 3 different algorithms (Genetic Algorithm (GA), Supervised Neural Network (SNN), Quick Classifier (QC)) for classification. The generated classification models were compared based on the calculated Recognition Capacity (RC) and Cross Validation (CV) which indicated the accuracy and reliability of the model. A model with the highest RC and CV was chosen for further peak statistical evaluation. Intensity of peaks presented in the model were evaluated by statistical tests. Peak intensity with p-values < 0.05 indicated statistically significant difference. Interesting informative peaks were assessed the capability to discrimination based on Area under the Receiver Operating Characteristic (ROC) curve (AUC) values.
Detection of spiked Leptospira spp. in urine
In order to mimic the detection of leptospires in urine, 4 Leptospira spp. including 2 from pathogenic group (L. interrogans and L. borgpetersenii), one from the intermediate group (L. wolffii) and one from the non-pathogenic group (L. biflexa) were cultured to 108 CFU/ml. Three milliliters of culture were centrifuged at 3,000 g for 10 min, pellet was added with 3 ml urine from healthy donor. After centrifugation at 8,000 g for 10 min., pellet underwent protein extraction. The protein spectra from spiked Leptospira in urine were analyzed and compared to the protein spectra acquired from Leptospira culture. Urine without spiked Leptospira was used as negative control.
Results
Identification of protein mass fingerprinting of Leptospires groups
Fifteen references Leptospira spp. (included 8 pathogenic, 2 intermediate and 5 non-pathogenic spp.) generated by MALDI-TOF MS revealed protein profile at mass to charge ratio (m/z) ranged from 2–20 kDa. These protein profiles showed specificity representing individual species (Fig 1). The discrimination of Leptospira spectra was demonstrated by differentiating 3 different groups in the MSP dendrogram which related to their phylogeny (pathogenic, intermediate and non-pathogenic groups) (Fig 2). This result indicated that phylogeny of Leptospira could be clustered based on their protein profile using MALDI-TOF MS.
Protein spectra of pathogenic group (red spectra), intermediate group (blue spectra) and non-pathogenic leptospires (green spectra) are presented. X and Y axis represent mass to charge (m/z) ratio and ion intensity, respectively.
The clustering based on MSP of pathogenic, intermediate and non-pathogenic group were determined.
The reproducibility of MSP from MALDI-TOF MS demonstrated by identification score of 3 Leptospira spp. (L. interrogans, L. borgpetersenii and L. biflexa) were higher than 2.0 after triplicate tested. The identification score of low passage L. wolffii strain H8, H9 and H16 revealed no statistically significant differences in two different growth media (p-value >0.05). According to sub-culture time, identification score of less than 6 sub-culture times (1–5 times) and that of more than 6 sub-culture times (6–10 times) have no statistically differences (p-value > 0.05). These results indicated the high reproducibility of Leptospira protein spectra generated by MALDI-TOF MS. The detection limit of MALDI-TOF MS method was demonstrated in 10-fold serial dilution of L. interrogans (MD08) protein spectra at 106 CFU/ml (Fig 3).
X-axis represent Spectra at mass to charge ratio of 2000–20000, Y axis is peak intensity at arbitrary unit (a.u.).
The new protein spectra database of Leptospira spp. was created based on 116 Leptospira isolates of 15 references and 101 Leptospira isolates (including 100 pathogenic and 6 non-pathogenic group) as listed in Table 2. To validate this database, 97 Leptospira human isolates from Thailand and Laos were blindly analyzed based on profile matching to this Leptospira protein mass spectra database. The identification score of these isolates are shown in Table 4. The clinical isolates from Laos and Thailand were identified as L. interrogans (n = 84), L. borgpetersenii (n = 4), L. weilii (n = 7) and L. kirschneri (n = 1) with scores of 2.0–2.6. One isolates from Laos (No. UI14721) has score equal to 1.903 for L. interrogans best match and the second match (with score 1.702) was identified as the same species. The identification score evaluation of MALDI-TOF MS database demonstrated that 96 out of 97 (98.9%) correct identification at specie level.
Biomarker identification for Leptospira spp and serovars
Protein mass spectra in mass range of 3,000 to 15,000 Da were analyzed by mathematical algorithms to define the protein peaks that could represent as biomarkers for leptospires. Discriminating peaks for pathogenic and intermediate leptospires were discovered by GA models. Eight peaks specifically presented in for pathogenic leptospires (Table 5) with 99.71% and 99.12% of RC and CV, respectively. The intermediate leptospires showed and 8 specific peaks (Table 6) with 100% of RC and CV. Peaks at m/z of 7006.70 for pathogenic and 6786.51 for intermediate leptospires demonstrated highest intensity and AUC value among peak list of the models. These indicated the best uniqueness and specificity of the peaks to pathogenic and intermediate leptospires, respectively.
Ten specific peaks to discriminate among L. interrogans, L. borgpetersenii, and L. kirschneri were reported by the GA model with 99.2% and 96.1% of RC and CV, respectively. Each species of Leptospira showed a unique combination of peaks (Table 7). Four peaks at m/z of 3605.24, 5529.89, 8747.26, and 11260.81 were common peaks, which were observed in three Leptospira species (L. interrogans, L. borgpetersenii, and L. kirschneri), but with differences in their relative intensities. Species-specific peaks were also detected. The presence of peaks at 6422.61 Da for L. borgpetersenii and peaks at 8084.81 Da for L. kirschneri was a unique characteristic of those two Leptospira species. Three peaks (6422.61 Da, 6924.67 Da, and 8084.81 Da) were not detected in L. interrogans. In addition, among the absent 3 peaks, lack of 6924.67 Da was specific to L. interrogans.
Using the GA model, L. interrogans serovar-specific peaks were identified with 98.1% and 82.0% of RC and CV, respectively. The peak list was reported in Table 8. Regarding to the GA classification model which demonstrated as the best capacity for serovar discrimination, 19 peaks were picked to discriminate the 8 serovars of L. interrogans including Autumnalis, Bataviae, Conicola, Grippotyphosa, Medanensis, Pomona, Pyrogenes and Sejroe. Each serovar was characterized by the unique combination of peaks in both presence or absence and relative intensity level (Table 8). Seven out of the 8 serovars presented 17–18 peaks from all 19 peaks included in the model. The serovar Pomona showed all of the peaks. The serovar of Autumnalis was characterized by the absence of peak at m/z of 11325.41. The other serovars were demonstrated the 17 peaks as report in Table 8. Among the serovars with absence of 2 peaks, the serovar-specific pattern could be characterized by specific lacking of the peaks to all serovars but Autumnalis and Pomona. Particularly, the serovar Bataviae, Canicola, and Pyrogenes were characterized by the absence of the peak at 8619.54 Da, 10293.62 Da, and 5638.80 Da, respectively
Spiked Leptospira spp. in urine
The protein spectra of 4 spiked Leptospira spp. (L. interrogans, L. borgpetersenii, L. wolffii, L.biflexa) demonstrated profile matching in both spiked urine and media (Fig 4). The small protein peak at m/z among 2,000–4,000 for example at m/z 3,442 were detected in protein profile of urine alone and in leptospires spiked urine. These peaks might be other ionized proteins in donor urine because they could not be detected in media. The identification score was high in both spiked leptospires in urine (range of 1.96–2.15) and in media (range of 2.0–2.5) while that of urine alone was unreliable (below 0.5). Using principle component analysis (PCA), 4 spiked Leptospira spp. in urine were located in the same cluster where each species represented; pathogenic group (L. interrogans, L. borgpetersenii), intermediate (L. wolffii) and non-pathogenic group (L. biflexa) as shown in Fig 5.
The protein spectra of L. interrogans, L. borgpetersenii, L. wolffii and L. biflexa in culture media (a), spiked in urine (b). Urine without any leptospires used as control (c).
Each dot displayed a protein spectrum of each sample. Color dot represent L. interrogans (Li), L. borgpetersenii (Lb), L. wolffii (Lw) and L. biflexa (Lx) in media and urine. The clustering of pathogenic, intermediate and non-pathogenic group is demonstrated in dash line.
Discussion
Whole cell MALDI-TOF MS of Leptospira was used to generate unique protein spectra or protein fingerprinting of 15 Leptospira spp. and clustered them according to their phylogeny. This study created the largest number of Leptospira protein database (116 protein spectra) that has been reported. The large protein database has strengthened the accuracy of Leptospira spp. MALDI-TOF MS identification with 98.9% correct identification of 96 Leptospira clinical isolates from different times and geographical areas of Thailand and Laos. One isolate with identification score below 2 possibly occurred because the protein spectra had low intensity. It was also possible that detection of proteins with relatively high molecular weight are more difficult to detect than protein in the low molecular weight range since this study used sinapinic acid matrix which ionized intact protein at high molecular weight [15].
MALDI-TOF MS method is simple, rapid and reliable for species identification but the method required bacterial culture which is of low sensitivity, laborious and time consuming. New culture agar called Leptospira Vanaporn Wuthiekanun (LVW) for Leptospira could reduce the isolation time but they still required a week for culture [14, 16]. Urine is a non-invasive, in comparison with serum or blood, which might be of interest for diagnosis of leptospirosis. Our results using spiked 108 cfu/ml leptospires in urine are suggesting that MALDI-TOF MS method may be able to identify Leptospira in urine. However, there are limitations. We lacked urine from leptospirosis patients and further investigation on detection limit of Leptospira in urine is required. The lowest amount of Leptospira which could generate the protein spectra by MALDI-TOF MS was 106 cfu/ml of culture media which is comparable to two reports [8, 9]. However, the number of Leptospira in mammal urine is varied. A meta-analysis on the quantity of Leptospira in urine of infected animals and human in Ecuador has reported that rats excreted amount of Leptospira in urine with median of 5.7 x 106 cells/ml while large mammals excreted and shed 5.1 × 108 to 1.3 × 109 cells Leptospira per day [17]. Moreover, protein spectra of human and animal urine need identification in order to clearly define disease-associated biomarker. The standardize method for detection of Leptospira in human or animal urine needed to be optimized and validated.
The major benefit from large Leptospira protein spectra database was that the database can be used as training set for machine learning and making a decision based on the computational mathematic analysis which demonstrated good classification in phylogeny. The specific peak at m/z of 6786.51 and 7006.70 were consistently found in intermediate and pathogenic Leptorspira respectively. These 2 specific peaks could be used as biomarker to differentiate between those 2 groups. Moreover, three Leptospira spp. (L. interrogans, L. kirschneri, and L. borgpetersenii) which are most found species from rodent in Thailand, Laos and Cambodia [18] could present the differentiating peaks among each other. There were few studies [8–10] reporting the difficulty to differentiate between L. interrogans and L. kirschneri using MALDI-TOF MS. Our data revealed that two peak at m/z of 6924.67 and 8084.81 were presented in L. kirschneri but absent in L. interrogans.
Serotyping is useful in one health epidemiological studies of Leptospirosis. The attempt of serotype differentiation by MALDI-TOF MS was examined and indicated up to 19 peaks among 8 serotypes (serovars) of L. interrogans. However, the low number of individual serovars used as training set here may reduce the result accuracy. In addition, serovar diversity of Leptospira is based upon variation of lipopolysaccharide (LPS) at the bacterial outer membrane [19]. LPS structure consists of a lipid and a polysaccharide of O-antigen, outer core and inner core. Those components may not be ionized using MALDI-TOF MS settings adapted for bacterial protein detection. Besides, Leptospira isolates in Thailand has been reported as unpredicted serovars by microscopic agglutination test (MAT) [20].
In conclusion, our whole cell MALDI-TOF MS method can cluster Leptospira spp. according to their phylogeny and identify the large scale cultured leptospiral strains at the species level based on their protein profile with highly reproducibility. The biomarkers based on protein profile could be used for identification the common pathogenic species L. interrogans, L. kirschneri, and L. borgpetersenii of human isolates in endemic areas like Thailand and Laos PDR. Therefore, whole cell MALDI-TOF MS method is a simple, faster and reliable tool for confirmation of leptospiral species.
Acknowledgments
We thank Prof. Lee Smythe, WHO/FAO/OIE Collaborating Centre for Reference & Research on Leptospirosis, Western Pacific Region, Communicable Disease Unit, Queensland Health Forensic and Scientific Services, Brisbane, Australia for serovar identification; Prof. Wanpen Chaikumpa, Faculty of Medicine Siriraj Hospital, Mahidol University for providing low passage L. wolffii; Dr. Janjira thaipadungpanit for 16S rRNA data, Sayan Langla, Anisone Chanthongthip from MORU, Bangkok and LOMWRU, Vientiane for leptospire isolates. We are very grateful to patients, doctors, nurses and the staff of Udon Thani hospital and the Infectious Disease Ward of Mahosot Hospital of the LOMWRU. We thank Ass. Prof. Bounthaphany Bounxouei, the Director of Mahosot Hospital, Ass. Prof. Chanphomma Vongsamphan, the Director of Department of Health Care, Ministry of Health, and Prof. Bounkong Syhavong, Minister of Health, Laos PDR for their very kind help, collaboration and administrative support.
References
- 1. Costa F, Hagan JE, Calcagno J, Kane M, Torgerson P, Martinez-Silveira MS, et al. Global Morbidity and Mortality of Leptospirosis: A Systematic Review. PLoS Neglected Tropical Diseases. 2015;9(9):e0003898. PMC4574773. pmid:26379143
- 2.
Hinjoy S, Choomkasien P. Summaries of selected notifiable diseases. In: Akarasewi P, editor. Annual Epidemiological Surveillance Report 2011. Bangkok: Bureau of Epidemiology, Department of Disease Control, Ministry of Public Health of Thailand; 2011. p. 60–4.
- 3. Wuthiekanun V, Sirisukkarn N, Daengsupa P, Sakaraserane P, Sangkakam A, Chierakul W, et al. Clinical diagnosis and geographic distribution of leptospirosis, Thailand. Emerg Infect Dis. 2007;13(1):124–6. Epub 2007/03/21. pmid:17370525; PubMed Central PMCID: PMCPMC2725830.
- 4. Smythe LD, Smith IL, Smith GA, Dohnt MF, Symonds ML, Barnett LJ, et al. A quantitative PCR (TaqMan) assay for pathogenic Leptospira spp. BMC Infect Dis. 2002;2:13. pmid:12100734.
- 5. Agampodi SB, Peacock SJ, Thevanesam V, Nugegoda DB, Smythe L, Thaipadungpanit J, et al. Leptospirosis outbreak in Sri Lanka in 2008: lessons for assessing the global burden of disease. Am J Trop Med Hyg. 2011;85(3):471–8. pmid:21896807; PubMed Central PMCID: PMC3163869.
- 6. Sonthayanon P, Chierakul W, Wuthiekanun V, Thaipadungpanit J, Kalambaheti T, Boonsilp S, et al. Accuracy of loop-mediated isothermal amplification for diagnosis of human leptospirosis in Thailand. Am J Trop Med Hyg. 2011;84(4):614–20. pmid:21460019; PubMed Central PMCID: PMC3062458.
- 7. Esteves LM, Bulhões SM, Branco CC, Carreira T, Vieira ML, Gomes-Solecki M, et al. Diagnosis of Human Leptospirosis in a Clinical Setting: Real-Time PCR High Resolution Melting Analysis for Detection of Leptospira at the Onset of Disease. Scientific reports. 2018;8:9213. PMC6003994. pmid:29907838
- 8. Calderaro A, Piccolo G, Gorrini C, Montecchini S, Buttrini M, Rossi S, et al. Leptospira species and serovars identified by MALDI-TOF mass spectrometry after database implementation. BMC research notes. 2014;7:330. pmid:24890024; PubMed Central PMCID: PMC4048046.
- 9. Rettinger A, Krupka I, Grunwald K, Dyachenko V, Fingerle V, Konrad R, et al. Leptospira spp. strain identification by MALDI TOF MS is an equivalent tool to 16S rRNA gene sequencing and multi locus sequence typing (MLST). BMC microbiology. 2012;12:185. pmid:22925589; PubMed Central PMCID: PMC3460781.
- 10. Djelouadji Z, Roux V, Raoult D, Kodjo A, Drancourt M. Rapid MALDI-TOF mass spectrometry identification of Leptospira organisms: a reply. Vet Microbiol. 2012;159(3–4):544. pmid:22572479.
- 11. Wuthiekanun V, Chierakul W, Limmathurotsakul D, Smythe LD, Symonds ML, Dohnt MF, et al. Optimization of culture of Leptospira from humans with leptospirosis. J Clin Microbiol. 2007;45(4):1363–5. Epub 2007/02/16. pmid:17301285; PubMed Central PMCID: PMCPMC1865830.
- 12. Boonsilp S, Thaipadungpanit J, Amornchai P, Wuthiekanun V, Chierakul W, Limmathurotsakul D, et al. Molecular detection and speciation of pathogenic Leptospira spp. in blood from patients with culture-negative leptospirosis. BMC Infect Dis. 2011;11:338. pmid:22151687; PubMed Central PMCID: PMC3297668.
- 13. Thaipadungpanit J, Wuthiekanun V, Chierakul W, Smythe LD, Petkanchanapong W, Limpaiboon R, et al. A dominant clone of Leptospira interrogans associated with an outbreak of human leptospirosis in Thailand. PLoS neglected tropical diseases. 2007;1(1):e56. Epub 2007/11/09. pmid:17989782; PubMed Central PMCID: PMCPMC2041815.
- 14. Wuthiekanun V, Amornchai P, Paris DH, Langla S, Thaipadunpanit J, Chierakul W, et al. Rapid isolation and susceptibility testing of Leptospira spp. using a new solid medium, LVW agar. Antimicrobial agents and chemotherapy. 2013;57(1):297–302. pmid:23114772; PubMed Central PMCID: PMC3535913.
- 15. Liu Z, Schey KL. Optimization of a MALDI TOF-TOF mass spectrometer for intact protein analysis. J Am Soc Mass Spectrom. 2005;16(4):482–90. pmid:15792717
- 16. Wuthiekanun V, Amornchai P, Langla S, White NJ, Day NP, Limmathurotsakul D, et al. Antimicrobial Disk Susceptibility Testing of Leptospira spp. Using Leptospira Vanaporn Wuthiekanun (LVW) Agar. Am J Trop Med Hyg. 2015;93(2):241–3. pmid:26055750; PubMed Central PMCID: PMC4530741.
- 17. Barragan V, Nieto N, Keim P, Pearson T. Meta-analysis to estimate the load of Leptospira excreted in urine: beyond rats as important sources of transmission in low-income rural communities. BMC research notes. 2017;10:71. PMC5273803. pmid:28129788
- 18. Cosson JF, Picardeau M, Mielcarek M, Tatard C, Chaval Y, Suputtamongkol Y, et al. Epidemiology of Leptospira transmitted by rodents in southeast Asia. PLoS Negl Trop Dis. 2014;8(6):e2902. pmid:24901706; PubMed Central PMCID: PMC4046967.
- 19. Bulach DM, Kalambaheti T, de la Pena-Moctezuma A, Adler B. Lipopolysaccharide biosynthesis in Leptospira. J Mol Microbiol Biotechnol. 2000;2(4):375–80. Epub 2000/11/15. pmid:11075908.
- 20. Smythe LD, Wuthiekanun V, Chierakul W, Suputtamongkol Y, Tiengrim S, Dohnt MF, et al. The microscopic agglutination test (MAT) is an unreliable predictor of infecting Leptospira serovar in Thailand. Am J Trop Med Hyg. 2009;81(4):695–7. pmid:19815889.