Mycobacterium leprae, and the closely related Mycobacterium lepromatosis, are the causative agents of leprosy (1, 2). Both species have tropism for skin, mucous membranes, and Schwann cells in the peripheral nerves (3). When left untreated, leprosy can cause irreversible damage to the peripheral nerves. However, with timely implementation of leprosy multidrug therapy (MDT), physical disfigurement and other disabilities can be prevented and/or minimized.

Monitoring drug efficacy and developing improved treatment regimens is difficult as proof of M. leprae viability is an elusive detection factor. Although numerous attempts have been made, this organism has never been cultured on axenic medium and does not grow outside of host tissues. Moreover, M. leprae has an extremely long division time, requiring 12-14 days for a single replication cycle, so cultivation experiments can take months to complete (4). To compound matters further, M. leprae bacilli possess a dense lipid cell wall (5), and it can take months to years for the immune system to clear dead bacilli from tissues following effective drug therapy. Because of these unique characteristics, specialized techniques are required to ascertain the viability of this organism.

Over the past several decades, a variety of assays have been developed as indicators of M. leprae viability (reviewed in (6)). One of the initial tests used on clinical specimens was the highly subjective morphological index, or ratio of solid staining to beaded bacteria in a sample. Subsequently, various viability stains and metabolic assays were developed that were quite accurate. These assays, however, require high numbers of freshly prepared and purified bacilli, and the numbers of M. leprae bacilli in clinical samples are generally too low for their use. The gold standard for determining M. leprae viability is the mouse footpad (MFP) assay, in which the M. leprae bacilli in question are inoculated into the footpads of mice, and bacterial multiplication is measured over several months. While fairly sensitive, this procedure takes up to one year of culture before data acquisition and can handicap timely clinical applications. Furthermore, few clinics have access to laboratories that can conduct MFP culture and most cannot easily transfer fresh biopsies to those centers within the requisite 48-72 hours necessary for optimum viability retention. Therefore, a sensitive, field-feasible assay that can rapidly and accurately detect viable M. leprae in tissues would have immediate clinical applications in strategically placed referral laboratories involved in drug resistance monitoring, relapse detection and validation, or new drug evaluation.

Our laboratory previously developed a molecular viability assay (MVA) for M. leprae that is based on the expression of the mycobacterial esxA and hsp18 transcripts (7, 8). Since expression of these transcripts is normalized to the number of M. leprae present in the sample and compared to a standard curve, an absolute viability measurement is obtained that eliminates the need for a paired control. Additionally, the MVA can determine M. leprae viability in fixed tissue samples in less than one week and with smaller sampling requirements. It has proven useful for experimental studies of leprosy immunology and new drug evaluation as these transcripts are highly expressed by viable M. leprae, but expression rapidly decreases upon immunological (7–9) or drug-mediated (7, 10) killing of the bacilli. Detection of these transcripts has also shown promise for leprosy transmission studies (11).

The purpose of the current study was to evaluate the performance of the MVA on clinical biopsy specimens. These samples were from three distinct populations, and included patients who were newly diagnosed, relapsed, and currently undergoing treatment with MDT.

Leprosy is primarily a clinical diagnosis that is based on signs and symptoms such as loss of sensation in a pale or reddish lesion, peripheral nerve enlargement with loss of sensation or muscle weakness, and/or the presence of acid-fast bacilli in a SSS (17–19). Histopathological examination of lesion biopsies and bacteriological analysis of SSS are used for support and confirmation as they correlate well with clinical diagnosis (20–25). Histopathology enables the confirmation of acid-fast bacilli in a nerve and the classification of disease in the Ridley-Jopling spectrum (3). The bacterial load in a patient is generally estimated by counting the number of acid-fast bacilli present in SSS obtained from 3 - 6 sites and calculating the average BI. The BI is a semi-logarithmic scale that ranges from 0 (no bacilli in 100 oil immersion fields) to 6 (1000 bacilli in 1 oil immersion field) (13). Although it is acknowledged to have limited sensitivity, the SSS is often times the primary metric used to guide treatment and monitor patient responses to drug therapy. Currently, the presence of bacilli in a SSS classifies a patient as MB, regardless of the number of lesions (17).

PCR tests have also been investigated by numerous researchers for use as diagnostic tools (26). The repetitive sequence, RLEP (27), is a highly sensitive and specific marker for M. leprae, and is superior to many other markers (28–30). Moreover, most PCR tests correlate well with clinical diagnosis, SSS, and histopathology (31–34). As PCR tests become more standardized (35, 36) and affordable, this tool will likely become routine for the diagnosis of leprosy, especially for paucibacillary cases. It is important to note, however, that RLEP PCR only detects M. leprae DNA and does not differentiate between live and dead bacilli.

In the study presented here, RLEP PCR detected M. leprae DNA in 97% (426/439) of the biopsies overall, including 80.3% (53/66) from cases with an average SSS BI < 1. This underscores the exquisite sensitivity of RLEP PCR for use as a diagnostic tool. However, the primary purpose of running RLEP PCR in this study was to enumerate M. leprae in the biopsy specimen sent for viability testing in the MVA. A tissue sample must meet a specific threshold of bacilli present based on RLEP molecular enumeration on the DNA fraction (2 x 10⁴ M. leprae per specimen) in order to normalize the RNA fraction for reverse transcription (8). Since PCR is not routinely used in most laboratories to diagnose leprosy or to estimate bacterial load in a patient, we first determined if there was a correlation between the average SSS BI calculated from samples collected from 3 - 6 index sites with the number of M. leprae recovered from the leprosy lesion biopsy sent for MVA. Our data showed a significant positive correlation between these two variables in all three cohorts. More importantly, we found that an average SSS BI of ≥ 2 is predictive of recovering sufficient M. leprae to run the MVA. Ultimately, it is important that a viability assay be sensitive so that it can be applicable to as many patients as possible, including those with low BI. On the other hand, for MVA testing one should avoid taking a biopsy from a patient that would not likely yield enough M. leprae for the assay. Therefore, this value will serve as a guide to predict the likelihood of a successful viability determination and reduce the number of unnecessary biopsies performed.

An ideal molecular viability indicator should be highly expressed by viable M. leprae but rapidly degraded once the bacteria are no longer viable. Several gene transcripts have been evaluated for expression in M. leprae-infected tissues as determinants of bacterial viability, including hsp18, hsp71, 16S, SodA, and esxA (7, 29, 37–44). Transcriptomics analyses of M. leprae are also contributing to the search for better transcripts (45, 46). We currently use the expression of hsp18 and esxA transcripts for viability assessment. hsp18 belongs to the small heat-shock family of proteins and encodes the M. leprae-specific 18kd protein. This protein is expressed by M. leprae in macrophages and functions in intracellular survival and growth (47). It is a molecular chaperone and has been shown to bind ATP, prevent thermal inactivation of enzymes, and offer protection from aggregation and killing (48). esxA codes for ESAT-6, the early secreted antigenic target-6. It is a member of the type VII ESX secretion and virulence system in mycobacteria (49). It may play a role in cell adherence, invasion, phagosomal rupture (50) and translocation of the mycobacteria from the phagosome to the cytosol (51, 52). Both hsp18 and esxA transcripts are highly expressed by viable M. leprae in the mouse model (7, 8, 45, 46). Interestingly, in the current study we saw some variance among the different cohorts in the expression of these transcripts. Overall, hsp18 was expressed by 72.7% ± 5.5% of the new cases across the three cohorts. In contrast, esxA was expressed by nearly 60% of the CSC-Philippines new cases but < 40% of the AHRI-Ethiopia and AH-Nepal cases. The differential expression of esxA among these diverse population groups is intriguing and worthy of further study.

When new leprosy cases were examined for M. leprae viability in the lesion biopsies, there was a MVA-positivity rate of 76.1 ± 1.5% ( Figures 2D , 4D , and 5D ). The precision of this data across the three diverse population groups is striking and supports the robustness of the MVA. The current gold standard for determining M. leprae viability in biopsy material is the MFP assay. Prior studies across decades of investigation have shown a recovery of viable M. leprae in the MFP assay of 48% - 87% when inoculated with MB biopsy-derived M. leprae (53–59). Our current MVA data falls toward the upper end of this range. This lends support to the value of the MVA for human biopsies. It must be acknowledged that 23.9 ± 1.5% of biopsies from untreated patients, even some with a high number of bacilli, were MVA(-). These results could be due to the age of the lesions. In nude MFP, M. leprae show a sigmoidal growth curve and eventually reach a stationary phase. Although bacterial numbers increase over several months, a higher percentage of metabolically active M. leprae are obtained early in infection compared to those derived from older, larger footpads (60, 61), emphasizing that a high number of M. leprae does not necessarily indicate viability. As with the MFP assay, it is important to take steps to optimize detection of viable M. leprae in the MVA, such as selecting an active lesion, taking the biopsy sample from the leading edge of the lesion, and immediately fixing the samples for preservation of nucleic acids.

Regarding the MFP data from AH-Nepal patients in the current study, it was somewhat serendipitous that these data were available. Comparing the MVA data to this MFP data, especially when the biopsy samples were collected at different times and likely from different lesions, is an extremely high standard to set for the MVA. Nevertheless, there was a 68.5% agreement between the MVA and MFP assay in new patients and a 74.1% agreement in relapse patients. While the correlation between the MVA and the MFP assay did not reach significance in the new patients (p = 0.077), it was highly significant among relapse patients (p = 0.018). A possible explanation for this difference may be that a relapse case is often diagnosed earlier than a new case of leprosy as a former patient will recognize the symptoms and will have been encouraged to seek diagnosis and treatment without delay if symptoms return (59). A new lesion in a relapse case likely contains metabolically active viable M. leprae that can be detected by both assays. Interestingly, for both new cases and relapse cases the main discrepancy between the two assays was that the MVA detected viable M. leprae in more specimens than the MFP assay, i.e MVA(+) and MFP(-). This would suggest that the MVA is more sensitive than the MFP assay. Alternatively, one could argue that the difference may result from “false positives” in the MVA. We have evidence (unpublished data) that hsp18 and esxA transcripts can be detected in M. leprae-infected mouse footpads 1 day after drug treatment, but detection is negligible within 4 weeks. This indicates that once bacilli are killed the transcripts are degraded in < 1 month, but there may be a short window of low-level, non-degraded transcript detection. Likewise, samples that were MVA(-) and MFP(+) contained M. leprae populations with an extremely low percent viability. Theoretically, one viable M. leprae can grow up in the MFP (62) and this may have been missed by the MVA, which can detect ~30 viable bacilli in a tissue specimen (8). Lastly, some biopsies from untreated patients were negative in both assays, despite the extensive experience of the respective laboratories in biopsy collection, MFP assay, and MVA. Running both the MVA and the MFP assay on the same biopsy specimen, while difficult, could resolve these issues.

Finally, our data show that the MVA can be used to monitor killing of M. leprae following drug treatment in patients. The long-term tissue persistence of dead M. leprae in tissues after completion of MDT can be troubling for physicians, raising suspicions of non-compliance or drug resistance. In this study, the duration of MDT treatment had no effect on either the average SSS BI or the number of bacilli in the biopsy. This data corroborates the extremely slow clearance of bacilli from the lesions and emphasizes that decreasing bacterial counts is a tediously prolonged and inadequate measure of drug efficacy. However, there was a significant decrease in M. leprae viability, as measured by the MVA, with duration of treatment. Although the number of cases presented here was small, this data supports the need for further studies on the use of a viability assay to monitor drug efficacy and perhaps even enabling the shortening of therapy or development of more personalized therapy, especially in cases involving adverse drug effects.

In summary, we have utilized a MVA that ascertains M. leprae viability by measuring the normalized mRNA expression levels of the bacterial genes, esxA and hsp18, and applied this assay to 3 cohorts of clinical cases. Despite the diversity of these cohorts, there was excellent agreement in the detection of viable M. leprae from biopsy specimens. Nevertheless, one may question if there is a need for a M. leprae viability assay in clinical medicine as new cases of leprosy will continue to be diagnosed clinically and treated with the recommended leprosy therapy. The use of a rapid molecular viability assay could be warranted, however, under certain circumstances. First, treatment efficacy could be verified following drug therapy, especially as new shorter drug regimens may be approved. In addition, detection of viable M. leprae following completion of therapy would justify drug susceptibility testing and could contribute to transmission studies. Second, such an assay would likely be useful to confirm relapse cases. Moreover, as molecular epidemiological (63) and immunological (64, 65) assays become more discerning, determination of M. leprae viability together with these assays may help differentiate relapse from reinfection and relapse from reaction. Finally, a rapid molecular viability test would greatly aid clinical trials for new leprosy drugs by replacing the time-intensive requirement for MFP passage of biopsy material from treated patients. Our MVA is specific for M. leprae, sensitive to an average SSS BI of ≥ 2, can be performed in < 4 days, is field-friendly with regard to fixation, storage, and transport of tissue specimens, and is the only assay validated with known viable M. leprae bacteria.

All authors listed have made a substantial, direct, and intellectual contribution to the work and approved it for publication.

This work was supported by funding from the Leprosy Research Initiative (LRI) and the Turing Foundation under LRI Grant No. 703.15.43.

The authors wish to thank Dr Mahesh Shah, Mr Kapil Neupane, Ms Jeni Maharjan, and Mr. Nischal Pokhrel.

Author SL was employed by company IHRC, Inc.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

The views expressed in this article are solely the opinions of the authors and do not necessarily reflect the official policies of the U. S. Department of Health and Human Services or the Health Resources and Services Administration, nor does mention of the department or agency names imply endorsement by the U.S. Government.