Mycobacterium leprae’s Infective Capacity Is Associated with Activation of Genes Involved in PGL-I Biosynthesis in a Schwann Cells Infection Model

Peripheral nerves and Schwann cells (SCs) are privileged and protected sites for initial colonization, survival, and spread of leprosy bacillus. Mycobacterium leprae strains that survive multidrug therapy show a metabolic inactivation that subsequently induces the recurrence of typical clinical manifestations of leprosy. Furthermore, the role of the cell wall phenolic glycolipid I (PGL-I) in the M. leprae internalization in SCs and the pathogenicity of M. leprae have been extensively known. This study assessed the infectivity in SCs of recurrent and non-recurrent M. leprae and their possible correlation with the genes involved in the PGL-I biosynthesis. The initial infectivity of non-recurrent strains in SCs was greater (27%) than a recurrent strain (6.5%). In addition, as the trials progressed, the infectivity of the recurrent and non-recurrent strains increased 2.5- and 2.0-fold, respectively; however, the maximum infectivity was displayed by non-recurrent strains at 12 days post-infection. On the other hand, qRT-PCR experiments showed that the transcription of key genes involved in PGL-I biosynthesis in non-recurrent strains was higher and faster (Day 3) than observed in the recurrent strain (Day 7). Thus, the results indicate that the capacity of PGL-I production is diminished in the recurrent strain, possibly affecting the infective capacity of these strains previously subjected to multidrug therapy. The present work opens the need to address more extensive and in-depth studies of the analysis of markers in the clinical isolates that indicate a possible future recurrence.


Introduction
Leprosy is a non-traumatic peripheral neuropathy whose causal agents are Mycobacterium leprae and Mycobacterium lepromatosis [1,2]. This generalized peripheral neuropathy is a chronic infection that produces inflammatory lesions on the skin and in peripheral nerves [3], which comprises a complex spectrum of clinical and immunological manifestations that range from localized forms imperceptible to the patient-indeterminate leprosy-to disseminated forms involving peripheral nerves, deformity, and disability. The clinical manifestations depend on the specific immunological reaction of the patient against the bacillus and are classified into two polar forms: lepromatous leprosy (LL), a predominantly humoral response, and tuberculoid leprosy (LT), a predominantly cellular response [4].
M. leprae is an obligate intracellular pathogen with an exquisite tropism for Schwann cells (SCs), glial cells of the peripheral nervous system (PNS) derived from neural crest precursors. M. leprae shows remarkable plasticity and is sensitive to changes that may affect its functions; in addition to lacking antimicrobial activity, the SCs can tolerate a high bacterial load [5,6]. The protection given by the SCs to M. leprae during its multiplication

Mycobacterium leprae Purification from Biopsy Samples of Patients with Recurrent or Non-Recurrent Leprosy Events
Five multibacillary (MB) patients diagnosed with LL (4 with non-recurrent and 1 with recurrent leprosy) who had bacteriological indices (BI) ranging from 1.5+ to 5+ were included in this study. The samples of new cases presented clinical signs of the disease between 7 and 48 months before diagnosis. The recurrent case's sample came from a patient who, 40 months before, had completed MDT. The appearance of new clinical signs and the recurrence diagnosis included positive bacillary viability and negative test to drug resistance determined by mutations in the drug resistance-determining region (DRDR) for the folP1, rpo β, and gyrA genes from M. leprae [24].
Different bacillary concentrations of suspension were obtained from the fresh biopsies, variations that were expected to be found due to the intrinsic characteristics of each patient and are related to their genetic repertoire, immunological response, and the disease stage (recurrence or non-recurrence) ( Table 1).

Morphological Description of Schwann Cells Infected with M. leprae from Recurrent and Non-Recurrent Events
For a successful entry of M. leprae, a pathogen with an exquisite tropism for SCs, these cells must have a homogeneous structure that allows the entry of M. leprae and its multiplication and viability in addition to facilitating the infection of more cells in the long term. Primary human Schwann cell lines (PHSC) (ScienCell) were infected with M. leprae clinical isolates from patients with recurrent or non-recurrent leprosy and remained in culture for 3, 7, 10, and 12 days after infection ( Figure 1).
Through Zielh Neelsen staining, intracellular bacilli were confirmed at all time points evaluated, and changes in the morphology of infected versus uninfected PHSCs were observed at these time points (Figure 1). At 3 days post-infection, intra-cellular bacilli were observed that followed the direction and surface given by the cytoplasm of the cells. This evidences cell culture's purity, congruence, and richness in the well. At 7 days, a cellular alteration was observed in the PHSC, characterized by cytoplasm contraction and increased vacuolization/lipid droplets in the infected cells. Finally, at 12 days of infection, notable cell loss was observed. Most cells show bacilli at the cytoplasm level and with a characteristic contraction of the same, which is also evidenced by the decrease in the monolayer over time.

Infection Kinetics of Schwann Cells with Different Clinical Isolates of M. leprae
As the infection model allowed M. leprae entry into PHSC, the infective capacity of each clinical isolate was evaluated by determining the proportion of infected cells over time. Significant changes were observed in the proportion of infected cells between each case analyzed (non-recurrent vs. recurrent) ( Figure 2, Table 2). The infectivity of M. leprae clinical isolates from patients with non-recurrent events is always more significant; a median increase of 12.83% to 18.5% was observed between days 0 and 3 in all cases and remained constant until day 12, reaching a maximum of 52.5% infection.
The recurrent clinical isolate showed a lower capacity of infection compared with the non-recurrent clinical isolates. At Day 0 to Days 3, 7 and 12 post-infection, the percentage of cell infection was 6.5%, 8.5%, and 22%, respectively. In the case of dead bacilli (M. leprae Thai-53), cell infection reached 22%, which remained constant over time (Tables S1-S11). Figure 3 shows the average number of bacilli by the cell in each experiment at each time analyzed. In the study, clinical isolates showed differences in multiplication capacity over time. On Days 3 and 7, the clinical isolates of the experiments that used live bacilli showed a progressive increase in the number of bacilli per infected cell, indicating that from this day on, M. leprae multiplied in the model used (Table 2).  Through Zielh Neelsen staining, intracellular bacilli were confirmed at all time points evaluated, and changes in the morphology of infected versus uninfected PHSCs were observed at these time points (Figure 1). At 3 days post-infection, intra-cellular bacilli were observed that followed the direction and surface given by the cytoplasm of the cells. This evidences cell culture's purity, congruence, and richness in the well. At 7 days, a cellular alteration was observed in the PHSC, characterized by cytoplasm contraction and increased vacuolization/lipid droplets in the infected cells. Finally, at 12 days of infection, notable cell loss was observed. Most cells show bacilli at the cytoplasm level and with a characteristic As the infection model allowed M. leprae entry into PHSC, the infective capacity of each clinical isolate was evaluated by determining the proportion of infected cells over time. Significant changes were observed in the proportion of infected cells between each case analyzed (non-recurrent vs. recurrent) ( Figure 2, Table 2). The infectivity of M. leprae clinical isolates from patients with non-recurrent events is always more significant; a median increase of 12.83% to 18.5% was observed between days 0 and 3 in all cases and remained constant until day 12, reaching a maximum of 52.5% infection.   The number of bacilli per infected cell of non-recurrent clinical isolates increased at

Relationship between Multiplication Capacity and Viability of M. leprae
The increase in the number of cells infected by M. leprae over time depends on the viability of the bacillus long enough to infect new cells; Table 2 shows the average number of infected cells per well and the average number of infecting bacilli per well at each of the times evaluated. Since the number of bacilli increased at 3 days post-infection in non-recurrent events and at 7 days post-infection in recurrent events and was maintained over time, the evaluation of bacterial viability correlated with the normalized expression of the sod gene.
Regarding non-recurrent bacilli, a 10-fold increase average in the quantification of the transcription of the sod gene on Day 3 post-infection compared with Day 0 (Figure 4) was observed. In the case of recurrent clinical isolates, the sod gene transcription increased Asterisks indicate one-way analysis of variance (ANOVA) with Tukey's test for each time * = p < 0.05; ** = p < 0.01; *** = p < 0.001 and **** = p < 0.0001.

Relationship between Multiplication Capacity and Viability of M. leprae
The increase in the number of cells infected by M. leprae over time depends on the viability of the bacillus long enough to infect new cells; Table 2

Level Transcription of Genes Involved in PGL-I Synthesis during Schwann Cell Infection with M. leprae Clinical Isolates
The genes for transcriptional analyses were selected for their pivotal role in PGL-I biosynthesis [16]. Concerning the recurrent clinical isolate, the ppsC gene (involved in the synthesis of the phenolphthiocerol/phthiocerol portion (Figure 5a

Level Transcription of Genes Involved in PGL-I Synthesis during Schwann Cell Infection with M. leprae Clinical Isolates
The genes for transcriptional analyses were selected for their pivotal role in PGL-I biosynthesis [16]. Concerning the recurrent clinical isolate, the ppsC gene (involved in the synthesis of the phenolphthiocerol/phthiocerol portion (Figure 5a For the non-recurrent clinical isolates, at 2 h post-infection the ppsC gene showed an average relative transcription of 9.08, which increased 29-fold on Day 3 post-infection, with a subsequent decrease on Days 7 and 12 post-infection (p > 0.0001). When the level transcription was compared between M. leprae clinical isolates depending on their origin, non-recurrent M. leprae increased transcription 30-fold on Day 3. In contrast, recurrent M. leprae increased transcription 6-fold on Day 7 ( Figure 5c). As observed for the ppsC gene, the transcriptional level of the ML2348 (coding for glucosyltransferase), ML2346c, and ML2347 (coding for methyltransferases) genes was upregulated in non-recurrent events, which could be partially due to those genes being part of the same operon at the same locus (Figure 5a,b,d-f).
The analysis of the transcriptional behavior of the ML0126, ML0127, and ML0128 genes, located in a different locus in the M. leprae genome (Figure 5g), which are involved in the process of methylation and translocation of the rhamnosyl residues of PGL-I (Figure 5b), were intriguingly compared with genes at locus 1 (ML2346c, ML2347, ML2348). Gene transcription of ML0126 and ML0127 was activated in recurrent M. leprae, specifically ML0126 between Days 0 and 7 post-infection (Figure 5h,i). For the non-recurrent M. leprae, the ML0126 and ML0127 genes were activated on all the days analyzed, reaching their maximum peak at 3 days post-infection, which decreased over time. The ML0128 gene began with a negative transcription at 2 h post-infection, rising at 3 days and decreasing negatively at 7 days post-infection (Figure 5j). Since the function of glycosyltransferases during the synthesis of PGL-I may be assumed by the ML2348 gene (locus 1), ML0128 does not have a significant action, especially in clinical isolates from non-recurrent events. 9

Discussion
The most severe clinical sequela of leprosy is the development of neuropathies that result from an infection of SCs by M. leprae. In multibacillary leprosy, there is a high number of intracellular bacilli, producing axonal degeneration that generates not only disability but also bone loss and muscle atrophy secondary to loss of sensitivity, which is also related to complications that affect the adjacent tissues of the skin [3,28,29]. These conditions have determined that leprosy remains one of humanity's most feared and stigmatized diseases [30].
A major obstacle to laboratory research on M. leprae is associated with the difficulty of multiplying the bacillus in cell culture and its slow doubling time (∼14 days) [25]. The capacity for interaction, invasion, and subsequent multiplication of M. leprae in SCs is a process that has been investigated on numerous occasions using clinical isolates of M. leprae from animal models. One of them is the nine-banded armadillo (Dasypus novemcinctus) and mouse footpads (MFP) [31][32][33], models that allow the multiplication of the bacillus due to an inadequate immune response against infection by M. leprae. These studies included significant numbers of bacilli to investigate cellular, biochemical, and molecular aspects [13].
After the application of MDT in recent years, an increase in the number of relapses has been observed.  [34,35]. When reviewing the worldwide incidence of relapse cases in leprosy, it is clear that it is low (1.16 per 1000 py (95% CI = 0.5915-2.076)) [18,36], that it is necessary to follow 1000 leprosy cases to obtain less than two cases of relapse of the disease, and that the period of follow-up could be long and variable because of censoring and loss to follow-up. This makes it challenging to include more cases of leprosy relapses in the present study, [36,37] especially considering that Colombia has 400 new cases of leprosy per year [34]. This work is a descriptive observational study that was not intended to extrapolate national or international data. Furthermore, given the characteristics of the disease, the inclusion of at least one case of leprosy relapse provides us with essential data that deserves to be analyzed and socialized.
The present study compared the infective capacity of M. leprae clinical isolates extracted and purified from fresh tissue of patients with initial leprosy untreated to bacilli from patients with recurrent events of leprosy. Different amounts of the bacillus was isolated from the biopsy samples, depending on the clinical, genetic, and immune characteristics of patients, thus producing different BI (r = 0.919), which was estimated when the biopsies were obtained from donor patients (Table 2). However, these amounts (BI), used as infective suspensions in our experiments, are not related to the percentage of infected cells in the T0 of experiments, which may be related to the inherent characteristics of each clinical isolate ( Table 1).
The quantities of bacilli obtained allowed us to adapt a multiplicity of infection (MOI) of 10:1, unlike what occurred with previous studies [38]. The conditions used in this study allowed the cultures to be kept infected for up to 12 days, with no monolayer detaching due to dehydration or the action of catabolic products from M. leprae. When the percentage of infection of the SCs with clinical isolates of M. leprae from patients with recurrent and nonrecurrent events was evaluated, a differential behavior was found, showing that M. leprae from recurrent patients increases its infective capacity at 7 days post-infection compared to M. leprae from non-recurrent patients, which increases steadily and sustains earlier, from Day 3 post-infection. The initial interaction time between the bacillus and the cells was only 2 h. After this time, the cells were washed four times with PBS 1X to remove the bacteria in the supernatant that was not internalized by the cell, which guaranteed that subsequent bacillary load quantifications over time correspond to actual infections and the increase in the number of bacilli (Table 2).
An essential feature of mycobacteria is their ability to adapt to stress conditions, such as the presence of antibiotics, an active immune response, and permanence in the host cell.
The lag phase is characterized by growth arrest or retardation, anaerobic respiration, and tolerance to antibiotics [39]. Under physiological conditions, bacteria in a latent state are contained mainly in granulomas, cells that protect them from the action of the immune system [40]. In our studies, the results of the infective capacity among the clinical isolates evaluated show that the clinical isolates from recurrent events require a longer adaptation time to the infection environment, so we can speculate that they were in a state similar to latency, reflected by decreased metabolism [41,42].
By Day 12, it was observed that cellular infection varied between 13% and 54% ( Figure 2 and Table 2); this result shows that the SCs can reach between 80% and 100% infection with the leprosy bacillus, according to the results presented by Diaz Acosta et al. [13]. When we used samples with dead M. leprae, a 26% infection of SCs was reached, a result that also was observed by Diaz et al., in which it was established that the percentage of SCs with dead M. leprae bacilli decreased to less than 50% compared to that achieved by viable bacilli. This may be because upon the death of the bacillus, the structures of the cell wall may be altered so that the interaction with the host cell is affected [13,43].
The present work also evaluated the dynamics of the bacillary load over time, that is, the number of bacilli per infected cell ( Figure 3 and Table 2). Our results showed that the number of infective bacilli does not depend on the number of bacilli in suspension obtained from the original sample used for infection (r = 0.295) and that the viability of the bacillus is not required for the initial entry of M. leprae to the SCs but is essential for the increase of the infection or bacillary load over time, similar to the results obtained in previous studies [13,43,44]. Assuming that our infectious populations are asynchronous, as they come directly from the patient, we cannot assume that the population increase has doubled in terms of generation time (g), which is a period of 14 days. However, Tukey's analysis shows that as in Figure 3 and Table 2, the increase in population at these points in the experiment was significant, with an increase in the total population per well being achieved from 28,180 on Day 0 to 34,200 on Day 7 from the clinical isolates of recurrent events, while for clinical isolates of non-recurrent events, these data were 45,200 to 51,600, respectively. These results were confirmed with the analysis of the viability that M. leprae showed during its multiplication in the SCs, which demonstrated that the experimental model used allowed us to not only maintain viability but also stimulate the multiplication of the M. leprae clinical isolates coming directly from the patient, this being the first study to perform in vitro analysis of M. leprae from clinical samples.
For our study, bacillus viability estimated by sod gene transcription was related to the number of bacilli per cell and increased as the bacillus multiplied in the infection model, in which at least 10,000 bacteria were used at Time 0 (2 h) [45]. The count of the bacilli and the determination of the viability showed that the bacteria are capable of multiplying, even in the case of M. leprae, causing recurrent events, possibly reduced due to the previous application of antibiotic treatment in the patient. Previous studies have shown that once antibiotic treatment is withdrawn, intracellular mycobacteria resume their growth immediately. After a few days, their replication rate was identical to that of bacteria that had not been previously treated [46]. It is essential to highlight that in this case, M. leprae is the bacteria's longest generation time (∼14 days); therefore, it could require a longer adaptation time to restart its post-treatment growth.
The ability of M. leprae to bind through a specific interaction of PGL-I with the alpha-2 subunit of human laminin (LAMA2) promotes the attachment of the bacillus to the basal lamina of the SCs. This allows the internalization of the pathogen [12,17,47]. In addition, M. leprae has developed PGL-I production to evade innate immunity and establish longterm residence in the host [16]. In this study, the possible participation of genes involved in the synthesis of the trisaccharide component of PGL-I in the infective capacity of M. leprae from clinical samples was evaluated. Since it is known that intracellular pathogens such as M. leprae require the lipids of the host cell to ensure the successful colonization of the microorganism and the progression of the infection, specially PGL-I, the measurement of transcription of genes involved in the biosynthesis of this glycolipid is a reliable way to estimate the multiplication and infective capacity of M. leprae at the intracellular level [48]. This should be valid in clinical isolates of M. leprae without previous exposure to antibiotics (non-recurrent) and for microorganisms in a state of latency or recurrence.
The PGL-I biosynthetic pathway involves more than 20 enzymatic steps, including the synthesis of the lipid core common in mycobacteria. Specifically, the genes involved in the synthesis of the trisaccharide portion of PGL-I are ML0128, which codifies a rhamnosyl transferase of the second rhamnosyl residue in position 2 of the first glycosyl residue; ML0127, which codifies a methyltransferase involved in the methylation of position 2 of the second rhamnosyl residue; and ML0126, which codifies the enzyme responsible for methylation at position 3 of the first and second sugar residues. Regarding the transfer and modification of the terminal glycosyl residue, ML2348 codifies the glycosyltransferase and ML2346c, and ML2347 codifies the methyltransferases required for the modification of positions 3 and 6 [16]. The mentioned genes grouped in two regions of the M. leprae genome have orthologs involved in developing PGL and thiocerol dimicoserosate of M. tuberculosis [16,49]. On the other hand, the ppsC gene synthesizes the phenol-thiocerol portion of PGL-I. Suppose that transcription levels of ppsC (>30-fold) are activated in clinical isolates of non-recurrent events. In that case, it is strongly suggested that these clinical isolates contain higher levels of PGL-I, which have been associated with the pathogenicity of M. leprae; since this glycolipid is unique to M. leprae and is highly antigenic, it is therefore not found in other mycobacteria [12,50]. This is consistent with the increased infection kinetics of clinical isolates from the non-recurrent event observed in SCs.
We also compare the transcriptional behavior of the genes involved in the synthesis of the saccharide core of PGL-I, which are distributed into two different loci. In Locus 1, we found that ML2348 (glycosyltransferase), ML2346c, and ML2347 (methyltransferases) required structural modification of the sugars. We found a differential behavior between transcription levels of the clinical isolates of M. leprae from recurrent and non-recurrent events that were higher on Day 3 post-infection (p < 0.0001) for non-recurrent events and at Day 7 post-infection for the recurrent event (p = 0.8867) and, accordingly, with transcription levels of ML2348, ML2346c, and ML2347, indicating that the synthesis of PGL-I could increase in those times. Therefore, the interaction of M. leprae with the host improves, producing an increased number of infected cells. At Locus 2, we analyzed gene transcription ML0128 (rhamnosyl transferase), ML0126, and ML0127 (methyltransferase). In this case, we observed a different behavior from that observed in Locus 1 because the ML0128 gene showed a transcript less than 1 in recurrent M. leprae. In the case of the non-recurrent clinical isolates, the ML0128 gene only presented a transcript greater than 1 at Day 3 post-infection. The function of this glycosyltransferase could be supported by another involved glycosyltransferase (ML2348), which is found in Operon 1. The complete transcription analysis suggests that the rate of synthesis of the PGL-I, essential for the entry of M. leprae into SCs, is different in the two types of clinical isolates, explaining why the recurrent event of the disease occurs on average 7 years after the first cure post-MDT and why the causative bacillus could be found in a state of latency [18].
These differences are related to the ability of M. leprae to limit its rate of replication in extreme conditions; a recovery of the replicative state occurs when favorable conditions return [51]. In our case, the strain of M. leprae from the recurrent case required a longer adaptation time to increase its replication rate and infectivity when compared to a strain that had not been previously exposed to antibiotic treatment. Another aspect that contributes to the growth kinetics and transcription genes involved in the biosynthesis of M. leprae PGL-I is the genetic and immune background of the host cell. Previous studies found that IFN-β secretion by macrophages correlated with the multiplicity of infection by M. leprae in vitro [52]. Although the induction of IFN-I is an essential factor for controlling viral infections, it has been shown to suppress the antibacterial response, particularly with infections by intracellular bacteria [52,53].

Ethics Statement
This study was approved by the ethical committee of Hospital Universitario Centro Dermatológico Federico Lleras Acosta (HUCDFLLA) (Empresa Social del Estado/Ministerio de Salud y Protección Social, Colombia). Informed consent was obtained from all adult volunteer patients before inclusion in the study (study identifier code: 4000-16.2H), considering the ethical standards in medical research according to the Declaration of Helsinki, where it is stipulated that it is a risk-free investigation.

Collection and Processing of Clinical Material
Five target patients were classified clinically, bacteriologically, and histopathologically according to the Ridley-Jopling scale [4]. Inclusion criteria for recurrent cases previously treated with MDT (and followed up at HUCDFLLA), included negative drug-resistance evaluation and positive bacillus viability; in addition, the diagnosis of recurrence was confirmed with a positive biopsy. In the cases of non-recurrent events (new cases), these should not have received prior treatment. From these patients, five elliptical incisional biopsies were obtained with the following dimensions: 1 cm long × 5 mm wide × 7 mm deep and performed by the dermatologist responsible for Hansen's clinical at HUCDFLLA, Bogotá, Colombia.

Purification and Quantification of Mycobacterium leprae
The biopsies of the patients were transferred to a sterile petri dish, and with the help of a scalpel, the tissue was sectioned. The sectioned material was transferred to a sterile glass tissue homogenizer (Ref 7726-L, PYREX ® , Corning, Tewksbury MA, USA) and macerated, keeping the homogenizer on ice. Once the tissue had been macerated, 1 mL of cold Hank's balanced salt solution (SSBH) (Ref H4641-100ML, Sigma, Irvine, UK) was added and homogenized into a suspension and filtered through a Falcon ® 40 µm Cell Strainer (Ref 352340 BD). The cell debris that remained in the filter was taken and placed again in the homogenizer. An amount of 1 mL of cold SSBH was added, and the tissue was homogenized again, and the suspension was then filtered again and once more until reaching a volume of 2.5 mL. An amount of 500 µL of 0.5% trypsin ([0.05%] final) was added to the suspension, and it was incubated in a water bath at 37 • C for 1 h. Subsequently, 10 mL of SSBH was added to the suspension and centrifuged at 3500× g for 30 min at 4 • C. Once the supernatant was discarded and 1 mL of SSBH was added to the precipitate, the suspension was homogenized [26,31].
Bacillary quantification was performed according to the criteria established by Shepard and McRae (Shepard and McRae, 1968). Briefly, the number of acid-fast bacilli (AFB) was counted by direct examination of 20 fields at 100× (immersion oil) of each of the three circles of 1 cm in diameter, each containing 10 µL of suspension bacillary, checking along the horizontal axis of the stained smear and using a microscope with a calibrated objective. The mean number of bacilli in each of the three smears was determined and multiplied by the appropriate calibration factor to obtain a mean and standard deviation for the AFB count. Care was taken to quantify only intact and fully stained bacilli. Bacillary viability was confirmed using the BacLight LIVE/DEAD viability kit (L7007, ThermoFisher Scientific, Eugene, OR, USA), according to the manufacturer's instructions.

Human Schwann Cell Culture
Primary human Schwann cell lines (PHSC) were obtained from the human neural crest derivatives that ensheathe and myelinate axons of peripheral nerves and were purchased from ScienCell, Carlsbad, CA, USA (Cat No. 1700). The PHSC was maintained in Schwann cell medium (SCM, Cat. no. 1701, ScienCell), supplemented with Schwann cell growth supplement (SCGS, Cat. no. 1752, ScienCell) in cell culture flasks previously coated with Poly-L-lysine 2 µg/cm 2 (Cat. No. 0413, ScienCell) to promote cell adhesion, according to the suppliers' recommendations. The cells were cultivated at 37 • C with 5% CO 2 , and the culture medium was renewed every 3-4 days, with two previous washes with PBS 1X.
The cells were washed with PBS 1X twice and subjected to a monolayer separation process with a Trypsin/EDTA solution at 0.0625% (ScienCell) for 1 min. Once the cell separation was verified in the optical microscope, the cells were collected in a 50 mL tube containing Bovine Fetal Serum (FBS); 5 mL of neutralization solution (ScienCell) was added, and then the mixture was centrifuged at 3000× g for 5 min. Once centrifuged, the count and evaluation of cell viability (in a Neubauer chamber) were carried out. An amount of 10 µL of the cell suspension was taken and mixed with 90 µL of trypan blue (1:10 dilution); 10 µL of the dilution was placed in a Neubauer chamber, and the cells were observed under a light microscope and counted in the four grids for white blood cells. The number of cells/mL was calculated as the average in the four quadrants × dilution factor × 10,000. The cell suspension was then prepared in supplemented SCM medium, according to the number of cells required for the infection assay (20,000 cells per well). Cells were cultured in 24-well plates on slides at the bottom of the well and were previously treated with Poly-L-lysine.

Schwann Cell Infection with Clinical Isolates of M. leprae
Once the SCs were in a monolayer on a slide, in 24-well plates (Falcon, Franklin Lakes, NJ, USA), at a density of 20,000 cells per well, the previously quantified bacillary suspensions of each case were added in a ratio of 10 bacilli per cell (MOI = 10:1). The samples were allowed to interact at a temperature of 33 • C with 5% CO 2 . The initial interaction time between the bacillus and the cells was 2 h only. After this time, the cells were washed four times with PBS 1X to remove the bacteria in the supernatant that was not internalized by the cell. Infected cells were washed every 72 h before adding an SCM-supplemented medium to avoid dehydration and alkalinization and were kept at a temperature of 33 • C with 5% CO 2 .
Once the incubation time of SCs infected with M. leprae was completed-2 h (T0 = 2 h), 72 h (T1 = 3 days), 7 days (T2 = 7 days), 10 days (T3 = 10 days) and 12 days (T4 = 12 days) in the 24-well plates-the medium was discarded, and the cells were washed twice with 1X PBS. Cells were fixed by adding 4% paraformaldehyde and incubated at 37 • C for 10 min. Once the paraformaldehyde had been discarded, the cells were washed twice with 1X PBS (between each wash, they were incubated at room temperature for 5 min). Then, the slides were removed from the well, and the slides were stained using the Zielh Neelsen stain.
To establish the kinetics of infection, the previously fixed slides were covered with filtered carbon fuchsin, and the dye was left in contact with the monolayer for 30 min. After the fuchsin was discarded, the slides were washed with water; then, 3% acid alcohol was added for 5 min, and the slides were washed again with water. Subsequently, the slides were covered with Loeffler's methylene blue (contrast dye) for 20 min, washed with water, and dried at room temperature. Mounting was then carried out with a cytoresin around the slide, leaving the study sample in contact with a slide to preserve the previously stained sample.
The infected cells were checked and quantified by the number of bacilli per cell by counting the proportion of 200 infected SCs and the number of bacilli in each cytoplasm using the light microscope. As a control, uninfected cells and other cells infected with dead bacilli from the mouse plantar pad were used, kindly supplied by Patricia Sammarco Rosa, Ph.D., of Laboratory Animal House, Instituto Lauro de Souza Lima (ILSL), Brazil.

Viability Assays of Mycobacterium leprae Infecting SC Culture
The infected cells were collected at 0, 3, 7, 10, and 12 days post-infection and were stored at −80 • C until RNA extraction by mechanical lysis. On the day of the experiment, the cell suspension was centrifuged at 15,000× g for 30 min at 4 • C. After mechanical lysis, we used the Qiagen RNeasy kit per the manufacturer's instructions. The isolated RNA was stored at −80 • C until use. The quality (260/280 nm) RNA concentration was measured in a NanoDrop 1000TM equipment (Thermo Scientific, Wilmington, DE, USA).
The normalized sod gene was used to stablish M. leprae viability. Briefly, the extracted RNA was treated with DNAse, and cDNA was obtained by reverse transcription using M-MLV RT 200 U/µL and random hexamers. Transcription of the sod gene was assessed using primers Sod-F-CACCGTTCGGAGAGAGGTTC-and sod-R-TCAACGAGATCCACCACACC (Table 3). To verify the viability of M. leprae in this assay, a standard curve was used to extrapolate viability with the number of bacilli per infected cell. Two biological replicates for infection assays were performed. Serial dilutions of clinical isolates of M. leprae's that infected Schwann cells were pooled and lysed, and cDNA was prepared to qPCR run for the sod gene. The CT values of the sod gene were obtained from the clinical isolates of recurrence, non-recurrence, and death bacillus, and their difference were calculated. DCT = CT (recurrence or nonrecurrence) − CT (dead). These ∆CT values were plotted against the number of bacilli per well to create a semi-log regression line.

qRT-PCR of Genes Involved in PGL-I Biosynthesis
The transcription of genes involved in PGL-I biosynthesis was quantified by real-time PCR (qPCR). For this purpose, specific oligonucleotides were designed for the selected genes (Table 1) using the primer-Blast [54] and Primer3Plus tools (Table 3) [55]. In the amplification reactions, SYBR Green was used as a fluorescence generator in a T100TM Thermal Cycler (BioRad).
All samples of total RNA were reverse-transcribed using reverse transcriptase (M-MLV RT 200 U/µL) (Thermo Scientific) per manufacturer recommendations. A mix of Oligo(dT) and random primers was used as the priming agent, and samples were stored at −20 • C until use.
For the relative quantification of the cDNA by qPCR, the commercial kit ssoAdvanced Universal SYBR Green Supermix kit from BioRad ® was used, following the manufacturer's recommendations, using 10 ng of cDNA for quantification reactions. Reactions were supplemented with oligonucleotides at a final concentration of 0.25 µM (sod, ML0126, ML0127, ML2348), 0.375 µM (ML2346, ML2347), and 0.5 µM (ML0128 and ppsC). The thermal cycles included an initial denaturation at 95 • C for 5 min, 40 cycles of denaturation at 95 • C for 15 s, and an extension at 62 • C for 30 s. All experiments were performed in triplicate on two experimental replicates. The sod gene (superoxide dismutase) was used as an endogenous control. The negative expression values of the genes were obtained by converting the cycle threshold (Ct) values according to the following formula: Expression value = 2 (−∆∆Ct) DCT = CT (recurrence or non-recurrence) − CT (dead). These ∆CT values were plotted against the number of bacilli per well to create a semi-log regression line [56].

Statistical Analysis
The results were analyzed in terms of mean ± standard error of the mean (SEM) of independent experiments in triplicate. Data were analyzed by applying a one-way or twoway multi-measurement analysis of variance (ANOVA) with Tukey's test. Numerical data were analyzed using the non-parametric Mann-Whitney test. Using GraphPrims version 9 for macOS Big Sur, GraphPad Software, San Diego, CA, USA, www.graphpad.com.

Conclusions
The results of the present study of human SC infection demonstrated that the ability of M. leprae to multiply under the optimal conditions used depends on the microbiological characteristics of each clinical isolate as well as possibly on the previous exposure of the bacillus to drugs of MDT; this can induce a modification in the metabolism of the biosynthetic pathways of crucial molecules such as PGL-I, specifically, for clinical isolates from recurrent events. In this study, for the first time two different behaviors at the cellular and transcriptional levels were investigated and found between the clinical isolates causing recurrent and non-recurrent events, so it is necessary to delve into the transcriptional studies of M. leprae clinical isolates coming directly from the patient to guide their management against the possible prognosis. Therefore, the present work opens the need to address more extensive and in-depth studies of the analysis of markers in the clinical isolates that indicate a possible future recurrence. This means that patient monitoring is necessary until treatment completion and a more prolonged post-MDT clinical follow-up to guarantee that no new symptoms have appeared. Institutional Review Board Statement: This study was approved by the ethical committee, Hospital Universitario Centro Dermatológico Federico Lleras Acosta (HUCDFLLA) (Empresa Social del Estado/Ministerio de Salud y Protección Social, Colombia). Informed consent was obtained from all adult subjects before inclusion in the study (study identifier code: 4000-16.2H). This study considered the ethical standards in medical research according to the Declaration of Helsinki, where it is stipulated that it is a risk-free investigation.

Data Availability Statement:
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.