Mycobacterium tuberculosis (H37Rv strain; a virulent laboratory strain of M. tuberculosis, American Type Culture Collection) was cultured in Middlebrook 7H9 broth media (Difco) supplemented with 10% albumin-dextrose-catalase (ADC) (Becton, Dickinson), 0.4% glycerol, and 0.05% Tween 80 and the cultures were grown to mid-log phase (OD_600nm 0.4–0.7).

The work plans for the animal experiments were examined and approved by the Indian Institute of Science, Bangalore’s Institute Animals Ethical Committee (IAEC), and the Institute Biosafety Committee (IBSC). The experiment was carried out in compliance with the guidelines set forth by the Committee for the Purpose of Control and Supervision of Experiments on Animals, or CPCSEA. A required number of 4-to 6-week-old female BALB/C mice were procured from the Central Animal Facility at IISc, Bangalore. Before being infected, the animals were allowed to acclimatize for two weeks at the ABSL-3 laboratory.

The toxicity of PRK was assessed in 4-6 weeks-old female BALB/C mice. The mice were divided into two groups (n = 5), one for PRK treatment and the other for PBS treatment as a control. PRK was orally administered @ 40mg/kg/b/wt, weekly at 5 doses for up to 4 weeks, and during this period mice were monitored for health and clinical signs, and weight was taken at weekly intervals for 4 weeks. On the 4^th week of post-treatment 500ul of blood was collected from all the animals through the retro-orbital puncture. Before this, the animals were given an inhalation anesthetic treatment consisting of isoflurane (2–3%). Serum was collected from blood and subsequently, blood serum tests were performed to measure the levels of blood urea nitrogen(BUN), serum glutamic pyruvic transaminase (SGPT), and alkaline phosphatase (ALP) level. All animals of both groups were humanely sacrificed by the cervical dislocation method to collect the lungs and liver for histopathological evaluation.

Chronic model- For the chronic model of infection, BALB/c mice were infected via aerosol through a Madison chamber aerosol generation instrument calibrated to deliver 100 CFU. After aerosol infection mice were housed for establishment of infection for 4 weeks, post which they were randomly grouped (n=6) into “H37Rv Control” and “PRK treated”, and treatment was initiated (PRK; 5 doses weekly@ 40mg/kg.b.wt) through oral route. 8 weeks post-infection, the mice were humanely sacrificed by cervical dislocation method to assess the bacterial load in the lungs.

To assess the bactericidal activity of PRK, we established acute model of infection, in an acute model of Infection, 4-6 weeks female BALB/C mice were infected with 500 bacilli (≈ 500 CFU) of H37Rv, and after one week of post-infection, (≈ 10⁴ CFU), mice were randomly grouped as mentioned in chronic model and treatment started as in chronic models for four weeks (PRK; 5 doses weekly@ 40mg/kg.b.wt through oral route). After that, the mice were humanely sacrificed by the cervical dislocation method to access the bacterial load in the lungs.

On the day of infection, the Mid-log phase (OD_600nm 0.4–0.7) grown bacterial cultures (H37Rv) were passed through different gauges of needles from lower to higher (23G, 26G, 30G) for making a single cell suspension and turbidity of the culture was measured. Animals were infected through aerosol means using the Madison Aerosol Chamber for Infection. Infected animals (n=5) were sacrificed by the cervical dislocation method, post-24 hours of infection to assess the infection dose. The remaining animals were, humanely sacrificed by cervical dislocation method at the appropriate time point of the study plan. Lungs and spleens were harvested and weighed for assessing CFU. One part of the right upper lobe of the lungs was fixed in 10% neutral buffered formalin for histopathology, the remaining tissues were homogenized in 2ml of PBS and homogenate of lungs in appropriate dilution were plated on 7H11 agar media supplemented with OADC (10% oleic acid, albumin, dextrose, and catalase)and PANTA (polymyxin B, amphotericin B, nalidixic acid, trimethoprim, and azlocillin)for enumeration of CFU. All experiments were conducted in the BSL-3 laboratory of IISC. Bangalore.

After sacrificing animals, the right upper lobe of the lung tissues was infused with 10% neutral buffered formalin and preserved until processed for histopathological assessment. At the time of tissue processing, the tissues were embedded in paraffin and sections of 4µm thickness from formalin-fixed and paraffin-embedded tissues were cut onto glass slides and stained with hematoxylin and eosin (H&E) for histopathological evaluation and imaging. The number of granulomas and granuloma areas with specific pathology was assessed.

Granuloma scoring was performed by observing and manually counting individual granulomas. we developed a scientific method for scoring the formation of granuloma with specific pathological markers (27). Granulomas with necrosis were given a score of 5, granulomas with no necrosis were given a score of 2.5, and granulomas with fibrosis were given a score of 1. These scores were summed up to provide the total granuloma scores ( Table 1 ).

The histopathological evaluation was done by the histopathology scoring system for the lung tissues, we developed a scientific method using Mitchison’s virulence scoring system with some modification, considering the number and areas of granulomas, necrosis, alveolar consolidation along with the type of infiltration of immune cells. The histopathology scores were graded as 0-3, (Severe pathology -3; Moderate pathology -2; Minor/minimum pathology -1; No pathology -0) (8, 28).

The lungs were removed aseptically, minced, and digested in a 2% FBS/DMEM solution with 0.2 mg/ml liberase and 0.1 mg/ml of DNase (Roche). The mixture was then incubated for 30 minutes at 37°C while being shaken. Subsequently, the pre-digested lung samples underwent processing in a Miltenyi Biotec mild MACS dissociator, following the usual lung dissociation technique. After the lung material was processed, it was put through a 70 µm cell strainer. The filtered samples were then centrifuged at 1600 rpm for 5 minutes at 4°C, and the supernatant was discarded. Next, the pellet containing the red blood cells (RBCs) was lysed with freshly made RBC lysis buffer and allowed to sit at room temperature (RT) for 5 to 10 minutes. To counteract the lysis buffer’s effects, 10 ml of PBS was added to each sample. The lung cell suspensions were then treated for five minutes on ice with Fc block (BD Bioscience), after which the cells were counted and stained for cell surface markers with a cocktail of antibodies and incubated in the dark at room temperature for 30 minutes. For the intracellular antigens, BD Biosciences’ fix and perm buffer were used to fix and permeabilize the cells. The samples were then treated with antibodies in accordance with the BD bioscience protocol. Samples were acquired on a BD FACSAria™ Fusion flow cytometer (BD Biosciences, San Jose, CA). The antibodies used in the present investigation are listed in Supplementary Table 1 .

The Flow cytometry data were analyzed using BD FACSDiva™ version 8.0.1 and FlowJo 10.8.0 software (BD Biosciences). The gating sequence included the following order: live (SSC-A⁺ FSC-A⁺), FSC-H vs FSC-A to exclude doublets (singlet gate), SSC-A⁺CD45⁺ (leucocytes), and so on.

Statistical analyses were conducted using GraphPad Prism software (version 8.0), and values were presented as mean ± SD. Data information: Statistical significance between experimental groups was determined by a two-tailed, unpaired Student’s t-test (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, and P > 0.5; n.s. not significant).

To evaluate the toxicity effect of PRK in BALB/c mice, PRK was given orally at a dose of 40 mg/kg/b/wt, five times a week for a maximum of four weeks ( Figure 1A ). Throughout this time, the mice’s health and clinical signs were tracked, and their weight was recorded every week for four weeks. Throughout the whole experimental period, all the cohorts’ animals remained healthy and alive. There were no appreciable variations in body weights between the PBS and PRK groups ( Figure 1B ). During the entire experiment no clinical symptoms such as weight loss or piloerection were seen. Blood urea nitrogen (BUN), serum alkaline phosphatase (ALP), and serum glutamic pyruvic transaminase (SGPT) levels were measured by blood chemistry analysis. Compared to the PBS-treated groups, the enzymatic levels of SGPT and ALP, which serve as indices of liver function, showed insignificant changes in the PRK-treated group ( Figures 1C, D ). Their absolute values (for all animals combined) were mainly contained within the published reference limits (29) suggesting that PRK has no hepatotoxic effects. Insignificant alterations in BUN levels, which are indicators of kidney function, were also observed in all animals in the PRK-treated cohorts compared to the PBS-treated cohort ( Figure 1E ). These findings suggest that PRK did not have any nephrotoxic consequences. Additionally, we found no histological abnormalities in the liver or lungs of the PRK-treated groups ( Figure 1F ), suggesting that PRK is a safe drug for treatment.

After establishing that PRK has no adverse effect on the health of mice. We performed the infection and treatment experiment to check the efficacy of PRK in chronic as well as acute Mtb infection models of mice. For the chronic model of infection, 4-6 weeks female BALB/c mice were infected via aerosol through a Madison chamber aerosol generation instrument calibrated to deliver 100 CFU, then mice were housed for 4 weeks of incubation for the establishment of infection, and after 4 weeks of post-infection, the mice were randomly grouped (n=6) as H37Rv Control group and PRK Treatment group, (PRK; 5 doses weekly@ 40mg/kg.b.wt through oral route), and on the 8th week of post-infection (p.i.), the mice were humanely sacrificed for accessing bacterial load in the lungs. For the acute model of infection, 4 -6 weeks female BALB/C mice were infected with 500 bacilli (≈ 500 CFU) of H37Rv through the aerosolization in the Madison aerosol chamber, and after one week of post-infection, mice were randomly grouped as H37Rv infected control and PRK treated group (n=6), and treatment started as in chronic models for four weeks ( Figure 2A ). After 4 weeks of treatment in both infection models, the mice were humanely sacrificed to assess bacterial load in the lungs and spleens. In the acute model, the differences in bacterial load between H37Rv control and PRK treated were 1 log, whereas the differences in bacterial load in the chronic model between H37Rv control and PRK treated were 0.8 log in the lungs of mice ( Figures 2C, E ). Similarly, the bacterial load between H37Rv control and PRK-treated groups of spleens in the acute infection model was 1.1 logs, whereas the differences in bacterial load in the chronic model between H37Rv control and PRK-treated groups were 0.4 logs. ( Figures 2D, F ). The gross pathology ( Figure 2B ) of H37Rv infected control and PRK treated groups in acute and chronic models of infection and treatment, respectively in the lungs were found consistent with the CFU count. We observed severe pathology with multiple small and large granulomas, alveolar consolidation, and infiltration of immune cells in the H37Rv infected control of the acute infection model of lungs ( Figure 2G ). Similarly, we observed severe to moderate pathology in the H37Rv-infected chronic model of mice lungs ( Figure 2H ). In PRK treatment groups, in both acute and chronic models of infection, we observed minimum pathology with large areas with normal alveolar space ( Figures 2G, H ). Granuloma scores and histopathology scores were observed to be consistent with bacterial infection and PRK treatment in the acute H37Rv control and acute PRK treated groups ( Figure 2I, J ) as well as the chronic H37Rv control and chronic PRK treated groups ( Figures 2K, L ), respectively. Thus, PRK works better in the acute model of bacterial infection than in the chronic model of infection in reducing the bacterial burden. Taken together, these results indicate the bactericidal properties of PRK. The results of acute and chronic infection model are now summarized in Table 2 .

There are two primary types of macrophages found in the lung, the tissue-resident alveolar macrophages (AMs; CD11c⁺, Siglec F⁺), and the interstitial macrophages (IMs; CD11b⁺, Ly6C⁺, and MHC-II⁺), which are derived from monocytes (16, 17, 19, 30). Nevertheless, it has been previously noted that the immune cell population most vulnerable to Mtb infection is constituted by permissive monocytes, dendritic cells, and alveolar macrophages (18, 19, 31, 32). Additionally, Huang et al. (19) reported that AMs’ anti-inflammatory properties foster an atmosphere that is favorable for Mtb replication and propagation. On the other hand, IMs promote inflammation and regulate the spread and replication of Mtb. In the present study, we found that in the acute model of mice infection and treatment, PRK-treated groups had lower populations of double positive, SSC-A⁺CD11c⁺ macrophage phenotypes ( Figures 3A, B ), which are representative of resident alveolar macrophages (AMs), double positive F480⁺Ly6c⁺ monocytes ( Figure 3C ), and dendritic cells (DCs) ( Figure 3D ). Because the alveolar macrophage populations rely on the metabolism of fatty acids for energy, they could change into inflammatory foamy macrophages, whose surroundings are favorable for Mtb replication and spread (19). The decline in the bacterial load in PRK treatment groups is apparently related to a reduction in the alveolar macrophage population. Similar patterns were also seen in the chronic model of mouse infection, where the populations of macrophages that correspond to alveolar macrophages (AMs); SSC-A⁺CD11c⁺ ( Figures 4A, B ), CD64⁺CD11c⁺ ( Figure 4C ), CD11c^high ( Figure 4E ), and SiglecF^high ( Figure 4D ) were reduced in a somewhat non-significant manner in the cohort that received PRK treatment. It was earlier reported that macrophages generated from monocytes with CD11c^high expression are extremely receptive to Mtb infection (33) (Figure 3A). Taken together, our findings are also consistent with a prior study.

The expression of phenotypic markers CD11c and CD11b on macrophages after Mtb infection is important in characterizing interstitial macrophages (IMs) (32). Here, we report that the CD11b⁺ ( Figures 5A, B ) and CD11b⁺Ly6G⁺ ( Figure 5C ) macrophage population corresponding to the interstitial macrophage (IMs) and neutrophils, significantly increased in the PRK-treated cohort as compared to the H37Rv infected control in the acute model of infection and treatment. Since interstitial macrophages (IMs) follow glycolytically active pathways, their enrichment is not supportive of the growth of Mtb (19). Interestingly, we were not able to observe a similar pattern of interstitial macrophage (CD11b⁺) upregulation in chronic models of infection and treatment, but the CD11b⁺ population significantly reduced in chronic infection and treatment models ( Figure 5D ). Nevertheless, despite the decline in IMs (CD11b⁺) population, we saw a noteworthy decrease in the bacterial load both in the lungs and the spleens in the chronic model of infection and treatment ( Figures 2E, F ). Furthermore, Huang et al. (19) reported that depletion of AMs reduced bacterial burden (Figure 5A), whereas depletion of IMs increased bacterial burden. On the other hand, we note that post-infection lifetime appears to be a determining factor for both IMs populations and bacterial load, rather than an arithmetic proportion under all settings. IMs population and bacterial burden in the acute model of infection and treatment were shown to be in arithmetic proportion in our experiment; however, in the chronic model of infection and treatment, IMs were not elevated, despite a reduction in bacterial burden. These findings lead us to the conclusion that PRK alters the metabolic transition of host macrophages from fatty acid metabolism to glycolytic metabolism at the log phage of the bacterial infection in the host.

The expression of the TREM-2 marker on Ly6c⁺ macrophages determine their activity in a proportional manner (34, 35). Furthermore, Mtb-infected mice showed monocytosis at 2- and 4 weeks post-infection, as evidenced by a rise in Ly6C^high, CD11b⁺, and CD115⁺ cells in the blood (19). Normally, in the acute phase of tissue damage, Ly6C^high inflammatory monocyte precursors move to wounded areas and cause inflammation. Nonetheless, monocyte-derived macrophages aid in the resolution of inflammation and tissue healing if the sources of tissue injury are removed (36, 37) Consistent with the above, we observed an upregulation of Ly6C⁺ and Ly6C^high inflammatory monocytes after 5 and 8 weeks of post-infection in H37Rv control groups of acute and chronic infection model, respectively; into contrast, we saw a notable decrease in Ly6C⁺ and Ly6C^high inflammatory monocytes in the groups that received PRK treatment in both acute and chronic model of infection and treatment ( Figures 6A–C ).

Microbial infection results in tissue damage and acute inflammation by the activation and assembly of multiple immune cell types at the injury site. Additionally, following the elimination of the tissue injury’s source, the activated and assembled immune cells coordinate the reduction of inflammation, tissue repair, and regeneration (30, 35, 38–40). The primary movers in tissue regeneration and repair following microbial infection are tissue-resident macrophages and inflammatory monocytes. They achieve this by phenotypically altering their recognition and activity during the resolution phases of inflammation and injury (14, 35). In mice, the expression of the proteins Ym1 (chitinase 3-like 3) and arginase-1 (Arg-1) consistently indicates the presence of tissue repair-associated macrophage phenotypes that are variable in activation (15, 37, 38, 41–43). Here, we report that, in the chronic model of infection and therapy, PRK-treated mice showed an increase in the Ym1+Ym2 ( Figures 7A, B ) and arginase-1 (Arg-1) ( Figure 7C ) positive population of macrophage in comparison to the H37Rv infected group. Since Arginase -1 (Arg-1) and Ym1+Ym2 positive macrophages are associated with the healing of lung tissues, our observation of their upregulation highlights a salutary role of PRK not only in reducing the tubercular burden but also in the subsequent healing of the tissue pathology. Given that the repair of lung tissues is linked to Arginase-1 (Arg-1) and Ym1+Ym2 positive macrophages, our observation of their upregulation underscores a protective function of PRK in both lowering the tubercular burden and promoting the healing of tissue disease that follows. The results of immune response in acute and chronic infection model are now summarized in Table 3 . ( Supplementary material; Supplementary Figure 1 ).