Panduratin A Induces Autophagy Through AMPK Activation Independent of mTOR Inhibition and Restricts Mycobacterium tuberculosis in Host Macrophages

ABSTRACT

Tuberculosis (TB), caused by Mycobacterium tuberculosis (Mtb), remains a major global health burden, especially with the increasing prevalence of drug‐resistant strains. There is an urgent need for new therapeutics that act via alternative mechanisms. Autophagy, a vital cell‐autonomous defense process, allows macrophages to degrade intracellular pathogens such as Mtb and has gained attention as a potential target for host‐directed therapy. In this study, we conducted a high‐content imaging screen of herb‐derived compounds to identify autophagy inducers in RAW264.7 macrophages. Panduratin A (NPA), a natural compound from Boesenbergia rotunda , was found to potently induce autophagy. NPA promoted autophagic vacuole formation in a dose‐dependent fashion at low micromolar levels. Its autophagy‐inducing effect was validated using RFP‐GFP‐LC3 dual fluorescence assays and immunoblotting in the presence of bafilomycin A1. Further mechanistic analysis revealed that NPA activates autophagy through AMPK activation, independent of mTOR inhibition. Importantly, NPA significantly promoted intracellular Mtb clearance and increased colocalization of Mtb with autophagosomes and lysosomes, in a manner dependent on Beclin‐1. These findings highlight NPA as a potent enhancer of macrophage antimicrobial responses via autophagy, supporting its potential as a candidate for host‐directed adjunctive therapy against TB.

Panduratin A, a flavonoid from Boesenbergia rotunda , was identified as a potential autophagy inducer in macrophages, where it appeared to activate AMPK independently of mTOR. This activity enhanced colocalization of Mycobacterium tuberculosis with autophagosomes and lysosomes, leading to reduced intracellular bacterial survival. These findings suggest panduratin A as a promising host‐directed candidate for adjunctive tuberculosis therapy.

1. Introduction

Tuberculosis (TB) accounts for approximately 10 million cases and 1.2 million deaths worldwide each year (WHO ²⁰²³). The emergence of multidrug‐resistant and extensively drug‐resistant Mycobacterium tuberculosis (Mtb), the causative agent of TB, has further complicated disease control efforts (Alena et al. 2012). While drug‐susceptible TB can typically be treated with a six‐month regimen comprising isoniazid, rifampicin, ethambutol, and pyrazinamide, drug‐resistant TB requires prolonged, more toxic treatments with significantly lower success rates (Alena et al. 2012). In addition, the Bacillus Calmette–Guérin (BCG) vaccine provides limited protection, preventing severe TB in children but failing to protect adults against pulmonary TB (Azis et al. 2012). Consequently, there is an urgent need for more effective drugs. Mtb infection is initiated when infectious droplet nuclei containing Mtb are inhaled and deposited in the lungs, followed by the engulfment of Mtb by the alveolar macrophages. Within these cells, Mtb evades immune clearance by disrupting phagolysosome formation (Donald et al. 2021). However, recent studies demonstrate that stimulating host autophagy enhances lysosomal delivery of intracellular Mtb, promoting bacterial clearance (Gutierrez et al. 2004; Singh et al. ²⁰⁰⁶; Delgado et al. ²⁰⁰⁸; Ponpuak et al. ²⁰¹⁰; Haque et al. ²⁰¹⁵).

Autophagy is a tightly regulated degradative process through which eukaryotic cells eliminate aggregated proteins, damaged organelles, and intracellular pathogens, thereby maintaining cellular homeostasis (Deretic 2021; Klionsky, Petroni, et al. ²⁰²¹). This pathway involves the coordinated action of multiple proteins and organelles (Parzych and Klionsky 2014). The ULK1 complex, comprising ULK1, ATG13, FIP200, and ATG101, acts as a key initiator of autophagy (Zou et al. 2022; Wong et al. ²⁰¹³). Its activity is modulated by nutrient and energy sensors; under nutrient‐rich conditions, mTOR suppresses ULK1 activation (Wong et al. 2013; Alers et al. ²⁰¹²). In contrast, nutrient deprivation inhibits mTOR and activates AMPK, leading to ULK1 activation (Dossou and Basu 2019; Kim et al. ²⁰¹¹). Activated ULK1 subsequently phosphorylates the class III PI3K complex (Beclin‐1, VPS34, VPS15, ATG14L), initiating autophagosome formation (Wong et al. 2013; Ohashi ²⁰²¹). The PI3K complex generates phosphatidylinositol 3‐phosphate (PI3P), which recruits downstream ATG proteins to the autophagosome assembly site (Wirth et al. 2013; Zhao et al. ²⁰²²). Autophagosome elongation and closure are then driven by two ubiquitin‐like conjugation systems (Nakatogawa 2013): ATG12–ATG5 conjugation (via ATG7 and ATG10) and lipidation of LC3‐I to LC3‐II (via ATG12–ATG5–ATG16L1) (Ariosa and Klionsky 2016; Wang et al. ²⁰²⁰; Iriondo et al. ²⁰²³). Subsequently, mature autophagosomes fuse with lysosomes to form autolysosomes, where cargo degradation occurs by lysosomal hydrolases in an acidic environment (Ke 2024). In the context of Mtb infection, autophagy induction, via starvation, cytokines, drugs, or pathogen‐associated molecular patterns, promotes Mtb sequestration in autophagosomes, delivery to lysosomes, and intracellular killing (Gutierrez et al. 2004; Singh et al. ²⁰⁰⁶; Delgado et al. ²⁰⁰⁸; Ponpuak et al. ²⁰¹⁰; Haque et al. ²⁰¹⁵), positioning autophagy as a promising therapeutic target in TB treatment.

Flavonoids derived from plants are well recognized for their diverse biological activities. Panduratin A (NPA), a flavonoid isolated from the rhizome of Boesenbergia rotunda (commonly known as fingerroot ginger), has been reported to exhibit strong antioxidant, anticancer, antibacterial, and antiviral properties (Cheah et al. 2013; Cheenpracha et al. ²⁰⁰⁶; Eiamart et al. ²⁰²⁵; Kanjanasirirat et al. ²⁰²⁰; Lai et al. ²⁰¹⁸; Linn et al. ²⁰²⁴; Parida et al. ²⁰¹⁷; Rukayadi et al. ²⁰⁰⁹; Salama et al. ²⁰¹⁸; Thadtapong et al. ²⁰²⁴; Worakajit et al. ²⁰²²). With regard to its antimicrobial potential, NPA displays significant antibacterial activity against a range of pathogenic bacteria, including clinical Staphylococcus strains, Enterococcus isolates, and Acinetobacter baumannii , with low minimum inhibitory concentrations (MICs) (Rukayadi et al. 2009, 2010; Thadtapong et al. ²⁰²⁴). Beyond its antibacterial effects, NPA demonstrates antiviral activity, notably against SARS‐CoV‐2, showing potent inhibitory efficacy in various cell lines, including Vero E6 and human airway epithelial cells (Calu‐3), during both the pre‐entry and post‐infection stages (Kanjanasirirat et al. 2020; Linn et al. ²⁰²⁴). It has also exhibited strong anti‐SARS‐CoV‐2 efficacy in human induced pluripotent stem cell‐derived cardiomyocytes at concentrations well below its cytotoxic threshold, and reduced lung inflammation in SARS‐CoV‐2‐infected hamsters (Linn et al. 2024; Kongratanapasert et al. ²⁰²³). Furthermore, NPA has demonstrated inhibitory activity against dengue virus, specifically by targeting the viral NS3 protease (Parida et al. 2017), and it also inhibits HIV‐1 protease (Cheenpracha et al. 2006).

In the present study, we investigated a panel of natural compounds isolated from medicinal herbs to identify agents capable of stimulating autophagy in RAW264.7 macrophages, with the goal of discovering candidates suitable for host‐directed adjunctive therapy against Mtb. Among the compounds screened, NPA emerged as a potent autophagy inducer, promoting the formation of autophagic vacuoles in a concentration‐dependent manner at low micromolar levels. This activity was confirmed through the RFP‐GFP‐LC3 dual fluorescence assay and Western blotting in the presence of bafilomycin A1. Mechanistic studies further revealed that NPA appeared to trigger autophagy via AMPK activation, independently of mTOR pathway suppression. Notably, NPA treatment led to a marked reduction in intracellular Mtb burden in a dose‐responsive fashion, and this activity was dependent on Beclin‐1. Additionally, it enhanced the colocalisation of Mtb with autophagosomes and lysosomes in a Beclin‐1‐dependent manner, indicating that its antimycobacterial effects are mediated through autophagy. Together, these results underscore the therapeutic potential of NPA as a host‐directed agent for TB treatment.

2. Materials and Methods

2.1. Compounds and Reagents

A library of 127 compounds isolated from herbs was obtained from the Excellent Center for Drug Discovery, Faculty of Science, Mahidol University. All compounds were maintained as 50 mM stock solutions in dimethyl sulfoxide (DMSO; Sigma‐Aldrich) and stored at −20°C. Working solutions were freshly prepared in complete medium on compound plates prior to use. For the negative control, cells were treated with an equivalent concentration of DMSO (Sigma‐Aldrich). Bafilomycin A1 (Baf; LC Laboratories) was used as a positive control to inhibit autophagosome–lysosome fusion. Earle's Balanced Salt Solution (EBSS; Sigma‐Aldrich) was used to induce autophagy through starvation. The plasmid encoding RFP‐GFP‐LC3B used in this study has been described previously (Tornee et al. 2005). Scrambled control siRNAs and all siRNAs were from Dharmacon.

2.2. Cell and Mycobacterial Culture

Murine RAW264.7 macrophages (ATCC, USA) were maintained at 37°C in a humidified 5% CO₂ incubator in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% (v/v) fetal bovine serum (FBS; Hyclone) and 4 mM L‐glutamine. The mCherry‐expressing Mtb H37Rv reference strain (Laopanupong et al. 2021; Tunganuntarat et al. ²⁰²³) was cultured in Middlebrook 7H9 broth (Difco) supplemented with 100 μg/mL hygromycin B, 0.2% glycerol, 0.05% Tween 80, and 10% (v/v) OADC enrichment (BD Biosciences). For infection assays, a single‐cell suspension of mid‐log phase Mtb was prepared by homogenization and applied to macrophages at a multiplicity of infection (MOI) of 1:10 (Haque et al. 2015; Ponpuak et al. ²⁰⁰⁹). Following 15 min of incubation at 37°C with 5% CO₂, cells were washed with complete DMEM three times to remove extracellular bacteria. Fresh complete DMEM was then added, and cells were incubated for an additional hour. To induce autophagy (starvation condition), cells were washed three times with PBS and subsequently incubated in EBSS.

2.3. Screening of Herb‐Derived Compounds

To assess the autophagy‐inducing potential of compounds isolated from herbs, RAW264.7 macrophages (3 × 10⁴ cells per well) were seeded into 96‐well black plates and incubated at 37°C with 5% CO₂ for 18 h. Cells were then treated for 4 h with complete DMEM containing 50 μM of each test compound, DMSO (negative control), EBSS (positive control for autophagy induction), or EBSS supplemented with Baf (to block autophagosome–lysosome fusion). After treatment, cells were fixed with 4% paraformaldehyde (PFA) for 15 min and blocked for 1 h using a blocking buffer composed of PBS containing 3% bovine serum albumin (BSA) and 0.1% saponin. Autophagosomes were labelled with rabbit anti‐LC3 antibody (RI Technologies; 1:500 dilution in blocking buffer) and incubated overnight at 4°C. Cells were then stained with Alexa488‐conjugated goat anti‐rabbit secondary antibody (Thermo Scientific; 1:400 dilution) for 2 h at room temperature in the dark. Nuclear staining was performed with Hoechst (1:500 in PBS) for 15 min. LC3⁺ puncta were quantified using the Operetta high‐content imaging system (PerkinElmer).

2.4. NPA Isolation

The purification of NPA was carried out as previously described (Kanjanasirirat et al. 2020). Briefly, dried Boesenbergia rotunda rhizomes (2.5 kg) were macerated with ethanol (15 L) at room temperature for 7 days, and this process was repeated four times. The combined extracts were filtered, evaporated, and yielded a crude ethanol extract (535.6 g). The extract was subjected to preliminary separation by flash column chromatography using gradient elution with ethyl acetate (EtOAc)‐hexanes, followed by methanol (MeOH)‐EtOAc mixtures, and finally pure MeOH, yielding fractions A1–A5. Fraction A3 (327.4 g) was further chromatographed on silica gel (900 g, Merck Art. No. 7734, 10 × 20 cm) across four rounds, using a gradient from EtOAc‐hexanes (1:4) to pure MeOH (20%, 30%, 50%, and 100% EtOAc‐hexanes). Collected fractions were monitored by TLC, pooled accordingly, and evaporated to obtain fractions B1–B7. Fraction B3 (36.9 g) was subjected to silica gel chromatography (450 g, 6 × 20 cm), eluted with MeOH‐CH₂Cl₂ (1:9) and then MeOH, producing fractions C1–C3. Fraction C2 (18.0 g) underwent further silica gel chromatography, eluted with CH₂Cl₂ followed by MeOH, yielding fractions D1–D4. Fraction D2 (16.3 g) crystallized as a solid, and recrystallization afforded pure NPA (2.2 g).

2.5. RFP‐GFP‐LC3 Puncta Assay

The autophagy‐inducing activity of NPA was assessed by quantifying RFP‐GFP‐LC3⁺ puncta in RAW264.7 macrophages using high‐content imaging, as previously described (Kimura et al. 2007). Briefly, RAW264.7 cells were transfected with 5 μg of plasmid encoding RFP‐GFP‐LC3 using the Amaxa Nucleofector device (program D‐032) and Nucleofector Solution V (Lonza). Transfected cells were transferred to fresh flasks containing complete DMEM, and the medium was replaced 6 h later. After 24 h, 3 × 10⁴ transfected cells were seeded into each well of 96‐well black plates and incubated overnight at 37°C in 5% CO₂. At 48 h post‐transfection, cells were treated for 4 h with complete DMEM containing 25 μM NPA, DMSO (negative control), or EBSS, with or without Baf. Cells were then fixed with 4% PFA for 15 min, and nuclei were stained with Hoechst (1:500 dilution in PBS) for 15 min at room temperature. Fluorescent autophagic puncta, distinguishing autolysosomes (RFP⁺GFP⁻‐LC3) and autophagosomes (RFP⁺GFP⁺‐LC3), were quantified by high‐content imaging analysis.

2.6. Western Blot Analysis

RAW264.7 macrophages (1.5 × 10⁶ cells per well) were seeded into 6‐well plates and incubated overnight at 37°C in a humidified 5% CO₂ atmosphere. Cells were subsequently treated for 2 h with complete DMEM containing 25 μM NPA, DMSO (vehicle control), or EBSS (autophagy induction control), with or without Baf. After two washes with cold PBS, cells were lysed in ice‐cold lysis buffer (20 mM Tris–HCl, pH 7.5; 150 mM NaCl; 25 mM sodium deoxycholate; 4 mM EDTA; 1 mM PMSF; 1% Triton X‐100; 1% SDS; and protease/phosphatase inhibitor cocktail). Lysates were incubated on ice for 20 min and centrifuged at 13,000 rpm for 20 min at 4°C. Protein concentrations were determined using a BCA assay. Equal amounts of protein were separated by 15% SDS‐polyacrylamide gel electrophoresis at 70 V for 2 h, followed by transfer onto PVDF membranes (Bio‐Rad). Membranes were blocked with 5% skimmed milk for 1 h and incubated overnight at 4°C with primary antibodies against p62 (Progen, 1:3000), LC3 (MBL International, 1:2000), phosphor‐mTOR (Cell Signaling, 1:1000), mTOR (Cell Signaling, 1:1000), phosphor‐4E‐BP1 (Cell Signaling, 1:1000), 4E‐BP1 (Cell Signaling, 1:1000), phosphor‐AMPKα (Cell Signaling, 1:500), AMPKα (Cell Signaling, 1:1000), and Actin (Abcam, 1:10,000). After five washes with TBST, membranes were incubated with HRP‐conjugated secondary antibodies for 1 h at room temperature. Chemiluminescent signals were detected using ECL (Thermo Scientific) and quantified using ImageJ software.

2.7. TFEB Nuclear Translocation Analysis

RAW264.7 macrophages were seeded at a density of 3 × 10⁴ cells per well in 96‐well black plates and incubated overnight at 37°C with 5% CO₂. Cells were subsequently treated with complete DMEM supplemented with 25 μM NPA, DMSO, or EBSS for 4 or 24 h. Following treatment, cells were fixed with 4% paraformaldehyde for 15 min at room temperature and then blocked for 1 h in blocking buffer comprising PBS with 3% BSA and 0.1% saponin. Cells were incubated overnight at 4°C with a rabbit anti‐TFEB antibody (Cell Signaling Technology) diluted 1:100 in blocking buffer. The next day, Alexa 488‐conjugated goat anti‐rabbit secondary antibody (Thermo Fisher Scientific), diluted 1:400 in blocking buffer, was added and incubated for 2 h at room temperature in the dark. Nuclei were stained with Hoechst (1:500 in PBS) for 15 min at room temperature. Nuclear translocation of TFEB was quantified using high‐content imaging with the Operetta system (PerkinElmer).

2.8. Mycobacterial Survival Analysis

RAW264.7 macrophages (2.5 × 10⁴ cells per well) were seeded into 96‐well black plates and incubated overnight at 37°C in a 5% CO₂ atmosphere. Cells were then infected with mCherry‐expressing Mtb H37Rv at an MOI of 10 for 15 min, followed by a 1‐h incubation (“chase”) as previously described (Laopanupong et al. 2021; Tunganuntarat et al. ²⁰²³; Ponpuak et al. ²⁰⁰⁹). After the chase period, cells were washed three times with PBS and subsequently treated with DMSO (negative control), EBSS (autophagy induction control), or varying concentrations of NPA, with or without Baf, for 24 h. Following treatment, cells were fixed with 4% PFA for 15 min, and nuclei were stained with Hoechst (1:500 dilution in PBS) for 15 min at room temperature. High‐content imaging was then performed to quantify the number of intracellular Mtb per cell. Mycobacterial survival was expressed as a percentage relative to the DMSO‐treated control group, which was set at 100%.

2.9. Colocalization of Mtb With LC3 or Cathepsin D

RAW264.7 macrophages (2.5 × 10⁴ cells per well) were seeded into 96‐well black plates and incubated overnight at 37°C with 5% CO₂. Cells were then infected with mCherry‐expressing Mtb H37Rv at a multiplicity of infection (MOI) of 10 for 15 min. After infection, uninternalized bacteria were removed by three washes with complete medium, followed by a 1‐h chase. Cells were subsequently treated for 2 h with complete DMEM containing 25 μM NPA, DMSO (negative control), or EBSS (autophagy induction control), with or without Baf. To assess Mtb‐LC3 and Mtb‐Cathepsin D colocalization, cells were fixed with 4% PFA for 15 min and blocked for 1 h in blocking buffer (PBS supplemented with 3% BSA and 0.1% saponin). Cells were then incubated overnight at 4°C with either rabbit anti‐LC3 antibody (RI Technologies, 1:500) or goat anti‐Cathepsin D antibody (R&D Systems, 1:50) diluted in blocking buffer. After washing, Alexa 488‐conjugated secondary antibodies (Thermo Scientific; 1:400 dilution) were added and incubated for 2 h at room temperature in the dark. Following three PBS washes, nuclei were stained with Hoechst (1:500 in PBS) for 15 min at room temperature. High‐content imaging was used to quantify the percentage of Mtb colocalization with LC3 or Cathepsin D.

2.10. High‐Content Imaging Analysis Pipeline

High‐content imaging was performed using an Operetta high‐content imaging system (PerkinElmer) equipped with a 20× air objective. Images were processed using basic flat‐field correction. For image analysis, the Columbus Image Data Storage and Analysis System (Revvity, formerly PerkinElmer) was employed. The workflow included: (i) nuclei identification (Hoechst 33342 channel) to define the total host cell population, (ii) cytoplasm segmentation based on the Hoechst signal, (iii) spot detection (Alexa 568 channel) to identify intracellular Mtb, (iv) infected‐cell classification by filtering host cells containing ≥ 1 bacterial spot, (v) quantification of bacterial intensity (Alexa 568 channel) and LC3 or Cathepsin D puncta intensity (Alexa 488 channel) at the single‐bacterium level, and (vi) colocalization analysis by filtering bacterial spots with green signal intensity ≥ 6000, representing LC3‐ or Cathepsin D‐positive Mtb. Output measures included the total number of host cells, infected cells, and bacterial objects, as well as bacteria per host cell, bacteria per infected cell, and LC3‐ or Cathepsin D‐positive Mtb colocalization per infected cell.

2.11. Statistical Analysis

All data were analyzed using Prism software version 9.0 (GraphPad) and are presented as mean ± SEM. Unless stated otherwise, results were derived from at least three independent experiments. Statistical comparisons were performed using one‐way ANOVA followed by Bonferroni post hoc correction. A p‐value of < 0.05 was considered statistically significant (*p < 0.05, **p < 0.01, ***p < 0.001).

3. Results

3.1. Identification of Autophagy Inducers From Herb‐Derived Compounds

As autophagy induction has previously been shown to promote intracellular Mtb elimination, we first sought to identify compounds isolated from herbs that could induce autophagy in RAW264.7 macrophages. To this end, high‐content imaging was employed to quantify LC3⁺ puncta per cell. As noted previously, LC3‐II localizes to autophagosomal and autolysosomal membranes and serves as a marker for autophagy activity (Klionsky, Abdel‐Aziz, et al. ²⁰²¹). Briefly, RAW264.7 macrophages were treated for 4 h with complete medium containing 50 μM of each compound, DMSO (negative control), or EBSS (positive control for autophagy induction via starvation). Cells were subsequently fixed, stained with an anti‐LC3 antibody to visualize autophagic vacuoles, and counterstained for nuclei. LC3⁺ puncta were then quantified by high‐content image analysis. Compounds were classified as autophagy‐inducing candidates if they increased the number of LC3⁺ puncta per cell by at least three SEM above the mean value of the DMSO‐treated controls (Figure 1A). The top ten candidates identified were E201, N1098, S1101, E145, E184, N1121, E198, E175, E2246, and E197 (Figure 1B). Among these, N1098 (panduratin A, NPA) isolated from Boesenbergia rotunda and previously reported to possess anti‐bacterial and anti‐viral activities, was selected for further investigation. Of note, NPA treatment did not affect cell viability, as determined by both the number of cells remaining attached to the plate surface after treatment and MTS assay results following exposure to different NPA concentrations (Figure S1).

FIGURE 1.

Panduratin A (NPA) is an autophagy inducer in RAW264.7 macrophages. (A, B) RAW264.7 macrophages were treated with complete media containing DMSO (negative control), EBSS (starvation, positive control), EBSS + bafilomycin A1 (Baf; autophagic flux inhibitor control), or individual herb‐derived compounds (50 μM) for 4 h. Cells were fixed, stained with anti‐LC3 antibody and analysed by high‐content imaging to quantify autophagic vacuoles. The dashed line indicates a threshold set at 3 SEM above the mean LC3⁺ puncta number observed in DMSO‐treated controls. Circles highlight the top ten candidate compounds demonstrating the greatest increase in LC3⁺ puncta per cell. Representative images of cells treated with the top ten candidate compounds showing increased LC3⁺ puncta formation are shown in (B). Scale bar = 5 μm. (C–E) Assessment of NPA‐induced autophagy by immunoblot analysis. RAW264.7 macrophages were treated with varying concentrations of NPA or DMSO with or without Baf, for 2 h. Cells were then lysed and subjected to western blot analysis. Representative immunoblot images from three independent experiments are shown (C). Densitometric analyses of p62/Actin and LC3‐II/Actin ratios were performed using ImageJ (D, E). Data are presented as mean ± SEM from three independent experiments. Statistical significance was determined using one‐way ANOVA followed by Bonferroni's multiple comparison test (**p < 0.01, ***p < 0.001, ****p < 0.0001), relative to the DMSO control, which was normalized to 1.0.

3.2. NPA Is a Genuine Autophagy Inducer

Since an increase in LC3⁺ puncta can result either from genuine autophagy induction or from inhibition of autophagic flux (impaired autophagosome‐lysosome fusion) (Klionsky, Abdel‐Aziz, et al. ²⁰²¹), we further validated the autophagy‐inducing activity of NPA in RAW264.7 macrophages and determined its EC₅₀ value using immunoblotting. In western blot analysis, a concurrent increase in LC3‐II/Actin levels and decrease in p62/Actin levels indicate true autophagy induction, distinguishing it from autophagic flux blockade (Klionsky, Abdel‐Aziz, et al. ²⁰²¹). As p62 serves as an autophagy substrate, its reduction reflects enhanced autophagic activity. To investigate this, RAW264.7 cells were treated with varying concentrations of NPA or DMSO with or without Baf, an inhibitor of V‐ATPase that raises lysosomal pH and impairs autophagic flux (Klionsky, Abdel‐Aziz, et al. ²⁰²¹). Our results demonstrated that NPA dose‐dependently increased LC3‐II/Actin ratios while decreasing p62/Actin levels, confirming its role as a true autophagy inducer (Figure 1C–E). Significant decreases in p62/Actin levels were observed at 12.5 μM and 50 μM NPA compared to the DMSO group (Figure 1D), while LC3‐II/Actin levels were significantly elevated at the same concentrations compared to the DMSO‐treated cells (Figure 1E). Based on the dose–response curve for LC3‐II/Actin levels, the EC₅₀ value for NPA was calculated to be 12.55 ± 1.17 μM.

To further confirm that NPA acts as a true autophagy inducer in RAW264.7 macrophages, we performed an RFP‐GFP‐LC3 puncta assay (Kimura et al. 2007). This assay distinguishes autophagosomes (RFP⁺GFP⁺‐LC3 puncta) from autolysosomes (RFP⁺GFP⁻‐LC3 puncta), as the acidic environment of autolysosomes quenches GFP fluorescence (Kimura et al. 2007). This distinction enables quantification of autophagosomes and autolysosomes per cell. An increase in both populations indicates autophagy induction, whereas an increase in autophagosomes without a corresponding increase in autolysosomes suggests autophagic flux inhibition (Kimura et al. 2007). To assess this, RAW264.7 cells were transfected with a plasmid encoding RFP‐GFP‐LC3, followed by treatment with complete medium containing NPA, DMSO (negative control), or EBSS (positive control for autophagy induction) with or without Baf for 4 h. Cells were then fixed and analyzed using high‐content imaging to quantify autophagosomes and autolysosomes. NPA treatment significantly increased both autophagosome and autolysosome numbers compared to the DMSO‐treated controls (Figure 2A–D), consistent with autophagy induction. Similar effects were observed with EBSS treatment, further validating the assay (Figure 2A–D). As expected, Baf treatment reduced autolysosome numbers across DMSO‐, NPA‐, and EBSS‐treated groups, confirming normal autophagic flux and that the fluorescent signals reflected autophagic structures rather than LC3 aggregation (Figure 2C,D). Additionally, immunoblotting analysis showed that NPA significantly decreased p62/Actin and increased LC3‐II/Actin levels compared to the DMSO‐treated cells (Figure 2E–G). In the presence of Baf, both LC3‐II and p62 levels were further elevated, indicating proper autophagic flux and successful inhibition of autolysosomal degradation (Figure 2E–G). Interestingly, LC3‐II/Actin levels were approximately twice as high in NPA‐treated cells as in EBSS‐treated controls, and nearly equivalent to those observed in cells treated with DMSO + Baf or EBSS + Baf, suggesting that NPA rapidly and robustly induces autophagic flux (Figure 2E–G). Together, these findings confirm that NPA is a bona fide autophagy inducer in RAW264.7 macrophages.

FIGURE 2.

NPA is a genuine autophagy inducer in RAW264.7 macrophages. (A–D) RAW264.7 macrophages expressing RFP‐GFP‐LC3 were treated with DMSO, EBSS, or NPA, with or without Baf for 4 h. High‐content imaging was performed to quantify the number of RFP⁺GFP⁺‐LC3 puncta (autophagosomes) and RFP⁺GFP⁻‐LC3 puncta (autolysosomes) per cell. Data represent mean ± SEM from three independent experiments. Statistical significance was determined using one‐way ANOVA followed by Bonferroni's multiple comparison test (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001), relative to the respective controls. Representative images are shown in (D). Scale bar = 5 μm. (E–G) Immunoblotting confirmed the autophagy‐inducing activity of NPA. RAW264.7 cells were treated with NPA, DMSO, or EBSS, with or without Baf, for 2 h. Cells were lysed and subjected to western blot analysis. Representative western blot images from three independent experiments are shown (E). Densitometric analysis of p62/Actin and LC3‐II/Actin ratios was performed using ImageJ (F, G). Data are presented as mean ± SEM from four independent experiments; statistical significance was assessed by one‐way ANOVA with Bonferroni's post hoc test (*p < 0.05, **p < 0.01, ****p < 0.0001), relative to the respective controls.

3.3. NPA Induces Autophagy via AMPK Activation Independently of mTOR Inhibition

To elucidate the signaling mechanism underlying NPA‐induced autophagy in RAW264.7 macrophages, immunoblot analyses were performed to evaluate the phosphorylation status of mTOR and its downstream target 4E‐BP1. Phosphorylation of 4E‐BP1 at Thr37/Thr46 serves as a marker of mTOR activity, while phosphorylation of mTOR at Ser2448 is commonly associated with mTOR activation (Bohm et al. 2021). Concurrently, phosphorylation of AMPK at Thr172 was assessed as an indicator of AMPK activation (Stein et al. 2000). As shown in Figure 3A, NPA treatment resulted in decreased p62 levels and increased LC3‐II accumulation, confirming autophagy induction. Notably, NPA markedly increased phospho‐AMPK levels without affecting phospho‐4E‐BP1 and phospho‐mTOR levels, suggesting an activation of AMPK in the absence of mTOR inhibition. In addition, we examined nuclear translocation of TFEB, a key transcription factor regulating autophagy‐related gene expression. Note that under nutrient‐rich conditions, mTOR is active and phosphorylates TFEB, which retains it in the cytoplasm (Roczniak‐Ferguson et al. ²⁰¹²). When mTOR is inhibited during starvation, TFEB phosphorylation is reduced, allowing TFEB to translocate to the nucleus, where it activates genes involved in autophagy (Roczniak‐Ferguson et al. ²⁰¹²). NPA treatment did not alter TFEB nuclear localization (Figure 3B,C). As expected, nutrient starvation (used as a positive control) led to increased AMPK phosphorylation, decreased phosphorylation of mTOR and 4E‐BP1, and enhanced TFEB nuclear translocation (Figure 3A–C). Collectively, these findings suggest that NPA induces autophagy in RAW264.7 macrophages via AMPK activation, independent of mTOR signaling inhibition.

FIGURE 3.

NPA induces autophagy via AMPK signaling. (A) RAW264.7 cells were treated with DMSO (negative control), NPA, or EBSS (positive control) for 4 h. Cell lysates were collected and analysed by immunoblotting to assess phosphorylation levels of mTOR (Ser2448), 4E‐BP1 (Thr37/46), and AMPK (Thr172). Expression of LC3‐II, p62, and Actin (used as a loading control) was also evaluated. Representative immunoblot images from three independent experiments, cropped from the same membrane, are shown. (B, C) Raw264.7 macrophages were treated with DMSO, NPA, or EBSS for 2, 4, and 24 h. Cells were fixed and stained with anti‐TFEB antibody and Hoechst. Levels of TFEB nuclear translocation were then determined by high‐content imaging. Data were mean ± SEM from at least three independent experiments; *p < 0.05, all relative to the DMSO‐treated control set to 100%, was determined by one‐way ANOVA with Tukey's multiple comparison test. Representative images are shown in (C). Bar 5 μm.

3.4. NPA Restricts Intracellular Mtb

Given that autophagy is a key cell‐autonomous immune mechanism by which macrophages eliminate intracellular Mtb, and that NPA is an autophagy inducer, we investigated whether NPA could restrict Mtb survival within host macrophages. RAW264.7 macrophages were infected with mCherry‐expressing Mtb H37Rv, then treated with complete medium containing varying concentrations of NPA, DMSO (negative control), or EBSS (positive control for autophagy induction) for 24 h. The number of intracellular Mtb per cell was quantified using high‐content imaging analysis (Figure 4A–B). Results demonstrated that NPA restricted intracellular Mtb in a dose‐dependent manner. Significant decreases in Mtb burden were observed at NPA concentrations of 25 and 50 μM compared to the DMSO‐treated controls. Consistent with previous studies (Gutierrez et al. 2004; Singh et al. ²⁰⁰⁶; Delgado et al. ²⁰⁰⁸; Ponpuak et al. ²⁰¹⁰; Chauhan et al. ²⁰¹⁶; Pilli et al. ²⁰¹²), EBSS treatment also promoted intracellular Mtb clearance. Since NPA has been reported to exhibit antibacterial activity, we next examined whether it possessed direct antimycobacterial effects. Mtb was treated with varying concentrations of NPA for 24 h, followed by CFU enumeration. The results showed that NPA did not display direct antimycobacterial activity (Figure S2). Overall, these findings indicate that NPA effectively enhances the elimination of intracellular Mtb in macrophages through host‐mediated mechanisms rather than direct mycobactericidal activity.

FIGURE 4.

NPA exhibits anti‐Mtb activity in RAW264.7 macrophages. (A, B) RAW264.7 macrophages were infected with mCherry‐expressing Mtb H37Rv at a multiplicity of infection (MOI) of 10 for 15 min, followed by a 1‐h chase period. Infected cells were then treated with varying concentrations of NPA, DMSO (negative control), or EBSS (positive control) for 24 h. After fixation, high‐content imaging was performed to assess intracellular Mtb survival. The percentage of intracellular Mtb survival was calculated relative to the DMSO‐treated control group, which was set at 100%. Data are presented as mean ± SEM from at least three independent experiments. Statistical significance was determined by one‐way ANOVA followed by Bonferroni's multiple comparison test (*p < 0.05, ****p < 0.0001). Representative images are shown in (B). Scale bar = 5 μm.

3.5. NPA Eliminates Intracellular Mtb Through Autophagy Induction

To confirm that NPA restricts intracellular Mtb through autophagy induction, we assessed its anti‐Mtb activity in RAW264.7 macrophages either proficient or deficient in Beclin‐1, a key protein essential for autophagosome formation (Liang et al. 2001). RAW264.7 cells were transfected with either scrambled siRNAs or siRNAs targeting Beclin‐1. Successful Beclin‐1 knockdown was verified by western blot analysis 48 h post‐transfection (Figure 5A–B). In addition, Beclin‐1 knockdown did not affect host cell viability, as determined by MTS assay (Figure S3). These cells were subsequently infected with mCherry‐expressing Mtb H37Rv and treated with complete medium containing NPA, DMSO (negative control), or EBSS (positive control for autophagy induction) for 24 h. Intracellular Mtb burden was quantified by high‐content imaging. In cells transfected with scrambled siRNAs, both NPA and EBSS treatments significantly reduced intracellular Mtb levels compared to the DMSO‐treated controls, consistent with results observed in wild‐type macrophages (Figure 5C,D). However, in Beclin‐1‐deficient macrophages, neither NPA nor EBSS treatment led to a significant reduction in intracellular Mtb compared to the DMSO‐treated group (Figure 5C,D). These findings demonstrate that NPA restricts intracellular Mtb via its ability to induce autophagy. To further support this conclusion, we treated Mtb‐infected macrophages with NPA in the presence or absence of Baf for 24 h and quantified intracellular bacterial survival, using DMSO as the negative control and EBSS as the positive control. Similar to EBSS‐treated cells, NPA treatment significantly decreased intracellular Mtb survival compared with DMSO‐treated controls, and this effect was reversed upon Baf addition (Figure S4A,B). Altogether, these findings demonstrate that NPA enhances Mtb clearance in host cells through autophagy induction.

FIGURE 5.

Beclin‐1 is essential for the anti‐Mtb activity of NPA. (A, B) RAW264.7 macrophages were transfected with either Beclin‐1‐specific siRNAs (siBeclin‐1) or scrambled control siRNAs (Scb). At 48 h after transfection, cells were harvested and analyzed by western blot. Data represent mean ± SEM from three independent experiments; statistical significance was determined by Student's t‐test (****p < 0.0001) relative to the scrambled siRNA control normalized to 1.0. A representative western blot image is shown in (A). (C, D) Beclin‐1‐proficient and ‐deficient RAW264.7 macrophages were infected with mCherry‐expressing Mtb H37Rv at an MOI of 10 for 1 h. Infected cells were then treated with NPA, DMSO, or EBSS for 24 h. Intracellular Mtb survival was quantified using high‐content imaging. Data are presented as mean ± SEM from at least three independent experiments, with statistical analysis performed using one‐way ANOVA followed by Bonferroni's multiple comparison test (****p < 0.0001; ns, non‐significant). Representative images are shown in (D). Scale bar = 5 μm.

3.6. NPA Induces Mtb‐LC3 Colocalization Which Is Dependent on Beclin‐1

To further confirm that NPA eliminates intracellular Mtb through autophagy induction, we investigated whether NPA enhances Mtb‐LC3 colocalization in host macrophages. RAW264.7 macrophages were infected with mCherry‐expressing Mtb H37Rv and subsequently treated with complete medium containing NPA, DMSO (negative control), or EBSS (autophagy induction control) in the presence or absence of Baf for 4 h. Autophagic vacuoles were stained with anti‐LC3 antibody, and Mtb‐LC3 colocalization was quantified by high‐content imaging (Figure 6A–B). NPA treatment significantly increased Mtb‐LC3 colocalization compared to the DMSO controls, and colocalization further increased upon Baf addition (Figure 6A–B). A similar pattern was observed in EBSS‐treated cells, supporting that NPA enhances the sequestration of Mtb into autophagosomes. To assess whether the aforementioned activity of NPA depends on autophagy, RAW264.7 cells were transfected with scrambled siRNAs or Beclin‐1‐targeting siRNAs for 48 h prior to infection with mCherry‐expressing Mtb H37Rv. Cells were then treated with NPA, DMSO, or EBSS, followed by LC3 staining. In scrambled siRNA‐transfected macrophages, NPA and EBSS significantly promoted Mtb‐LC3 colocalization compared to the DMSO treatment (Figure 6C,D). In contrast, Beclin‐1‐deficient cells failed to show enhanced colocalization following NPA or EBSS treatment (Figure 6C,D). Collectively, these findings demonstrate that NPA promotes intracellular Mtb clearance by inducing autophagy and increasing the recruitment of Mtb into autophagosomes within host cells.

FIGURE 6.

NPA promotes Mtb‐LC3 colocalisation through autophagy induction. (A, B) RAW264.7 macrophages were infected with mCherry‐expressing Mtb H37Rv at MOI of 10 for 15 min, followed by a 1‐h chase. Cells were then treated with NPA, DMSO (negative control), or EBSS (positive control) with or without Baf for 4 h. Following treatment, cells were stained with anti‐LC3 antibody and analyzed by high‐content imaging. The percentage of Mtb‐LC3 colocalization was quantified and compared across conditions. Data represent mean ± SEM from at least three independent experiments, and statistical significance was assessed using one‐way ANOVA with Bonferroni's multiple comparison test (*p < 0.05, ***p < 0.001, ****p < 0.0001). Representative images are shown in (B). Scale bar = 5 μm. (C, D) For Beclin‐1 dependency studies, RAW264.7 macrophages were transfected with either scrambled control siRNAs (Scb) or Beclin‐1‐targeting siRNAs (siBeclin‐1) for 48 h. Cells were then infected with mCherry‐expressing Mtb H37Rv and treated with NPA, DMSO, or EBSS for 4 h. After fixation and LC3 staining, Mtb‐LC3 colocalization was quantified by high‐content imaging. Data are shown as mean ± SEM from at least three independent experiments. Statistical analysis was performed using one‐way ANOVA with Bonferroni's multiple comparison test (**p < 0.01, ****p < 0.0001, ns = non‐significant). Representative images are presented in (D). Scale bar = 5 μm.

3.7. Beclin‐1 Is Required for NPA‐Induced Mtb Delivery to Lysosomes

We next examined whether NPA treatment enhances the colocalization of Mtb with Cathepsin D, a lysosomal marker, in host macrophages. RAW264.7 cells were infected with mCherry‐expressing Mtb H37Rv and subsequently treated with complete medium containing NPA, DMSO (negative control), or EBSS (positive control) in the presence or absence of Baf for 4 h. Following fixation, cells were stained with an anti‐Cathepsin D antibody and analyzed by high‐content imaging. Results showed that both NPA and EBSS significantly increased Mtb‐Cathepsin D colocalization compared to the DMSO‐treated controls (Figure 7A,B). As anticipated, Baf treatment reversed the increase in colocalization observed with NPA and EBSS, consistent with inhibition of lysosomal fusion (Figure 7A,B). These findings support the conclusion that NPA promotes autophagy, leading to enhanced delivery of Mtb to lysosomes in infected macrophages. Additionally, we assessed the effect of Beclin‐1 knockdown on Mtb‐Cathepsin D colocalization. RAW264.7 macrophages were transfected with either scrambled or Beclin‐1‐targeting siRNAs before infection with mCherry‐expressing Mtb H37Rv and treatment with NPA, DMSO, or EBSS for 4 h. High‐content imaging analysis revealed that, similar to wild‐type cells, NPA and EBSS significantly increased Mtb‐Cathepsin D colocalization in scrambled siRNA‐transfected macrophages compared to the DMSO treatment (Figure 7C,D). However, in Beclin‐1‐deficient cells, this increase was abolished, indicating impaired autophagy‐mediated Mtb trafficking (Figure 7C,D). Together, these findings demonstrate that NPA promotes the autophagy‐dependent delivery of Mtb to lysosomes in host macrophages.

FIGURE 7.

NPA promotes Mtb delivery to lysosomes via autophagy induction. (A, B) RAW264.7 macrophages were infected with mCherry‐expressing Mtb H37Rv at MOI of 10 for 15 min, followed by a 1‐h chase. Cells were then treated with NPA, DMSO (negative control), or EBSS (positive control) with or without Baf for 4 h. After fixation, lysosomes were stained using an anti‐cathepsin D antibody, and Mtb‐cathepsin D colocalisation was assessed by high‐content imaging. Data are presented as mean ± SEM from at least three independent experiments; statistical significance was determined by one‐way ANOVA with Bonferroni's multiple comparison test (**p < 0.01, ****p < 0.0001). Representative images are shown in (B). Scale bar = 5 μm. (C, D) RAW264.7 macrophages were transfected with scrambled control siRNAs (Scb) or Beclin‐1‐targeting siRNAs (siBeclin‐1) prior to infection with mCherry‐expressing Mtb. Following infection, cells were treated with NPA, DMSO, or EBSS for 4 h, fixed, and stained for cathepsin D as described above. Mtb‐cathepsin D colocalisation was analysed using high‐content imaging. Data represent mean ± SEM from at least three independent experiments; statistical comparisons were performed using one‐way ANOVA with Bonferroni's post hoc test (*p < 0.05, **p < 0.01, ****p < 0.0001, ns = non‐significant). Representative images are shown in (D). Scale bar = 5 μm.

4. Discussion

TB is an airborne infectious disease caused by Mtb. Pulmonary TB is transmitted via coughing, sneezing, or prolonged exposure to infected individuals, posing significant risks to immunocompromised persons, children, and the elderly (Tornee et al. 2005; Turner and Bothamley ²⁰¹⁵). Despite ongoing global control efforts, TB remains the leading cause of death from a single infectious agent, particularly in South Asia and Africa, with the emergence of drug‐resistant strains further complicating treatment (Falzon et al. 2023). The standard regimen for drug‐susceptible TB involves a six‐month course of isoniazid, rifampicin, pyrazinamide, and ethambutol. In contrast, drug‐resistant TB requires longer, more toxic treatment protocols that are often associated with lower success rates (Alena et al. 2012). Autophagy is a crucial cellular process that enables host cells to degrade intracellular pathogens, including Mtb, through lysosomal trafficking. Augmenting autophagy has been proposed as a promising strategy to enhance mycobacterial clearance (Gutierrez et al. 2004; Singh et al. ²⁰⁰⁶; Delgado et al. ²⁰⁰⁸; Ponpuak et al. ²⁰¹⁰; Haque et al. ²⁰¹⁵), making it a potential target for host‐directed adjunctive therapy.

In this study, we screened 127 natural compounds derived from herbs to assess their autophagy‐inducing potential, using high‐content imaging to quantify LC3⁺ puncta formation in RAW264.7 macrophages (Figure 1). NPA, a bioactive compound isolated from Boesenbergia rotunda , exhibited strong autophagy‐inducing activity and was selected for further investigation. A series of complementary assays confirmed that NPA is a bona fide autophagy inducer (Figure 2). Mechanistic studies suggested that NPA activates autophagy via AMPK phosphorylation, independently of mTOR inhibition (Figure 3). Notably, NPA also showed potent antimycobacterial activity in an autophagy‐dependent manner (Figures 4, 5, 6), indicating that it promotes intracellular Mtb clearance by enhancing host autophagic responses.

Previously, NPA has been reported to possess strong antioxidant, anticancer, antibacterial, and antiviral properties (Cheah et al. 2013; Cheenpracha et al. ²⁰⁰⁶; Eiamart et al. ²⁰²⁵; Kanjanasirirat et al. ²⁰²⁰; Lai et al. ²⁰¹⁸; Linn et al. ²⁰²⁴; Parida et al. ²⁰¹⁷; Rukayadi et al. ^{2009, 2010}; Salama et al. ²⁰¹⁸; Thadtapong et al. ²⁰²⁴; Worakajit et al. ²⁰²²). With respect to its antimicrobial potential, NPA demonstrates notable antibacterial activity against a broad spectrum of pathogenic bacteria, including Staphylococcus spp., Enterococcus spp., and Acinetobacter baumannii , with low MICs (Rukayadi et al. 2009, 2010; Thadtapong et al. ²⁰²⁴). It also shows effective antibacterial activity against Prevotella intermedia , Porphyromonas gingivalis , and Cutibacterium acnes (Rukayadi et al. 2009, 2010; Hwang et al. ²⁰⁰⁴). In addition to its antibacterial effects, NPA exhibits broad‐spectrum antiviral activity. It has been shown to inhibit SARS‐CoV‐2 replication in multiple cell lines (Kanjanasirirat et al. 2020; Linn et al. ²⁰²⁴) and reduce viral burden and lung inflammation in infected hamsters (Linn et al. 2024; Kongratanapasert et al. ²⁰²³). Furthermore, NPA targets the NS2B/NS3 protease complex of dengue virus (Parida et al. 2017) and inhibits HIV‐1 protease activity (Cheenpracha et al. 2006). In this study, we demonstrated that NPA significantly reduces intracellular Mtb in macrophages through the induction of autophagy, highlighting its ability to modulate host defense mechanisms in addition to directly targeting microbial factors. Taken together, these findings emphasize the broad therapeutic potential of NPA and support its further investigation as a candidate for clinical applications.

Notably, previous studies have reported contrasting effects of NPA on autophagy regulation, suggesting context‐dependent activity. A study in TNF‐α‐treated L6 rat skeletal muscle cells demonstrated that NPA suppresses autophagy by downregulating the transcription of autophagy‐related genes, as part of its protective role against muscle atrophy (Sa et al. 2017). On the other hand, in A375 melanoma cells, NPA was shown to induce autophagy (Lai et al. 2018). This was associated with activation of the AMPK pathway as well as inhibition of mTOR signaling, as evidenced by increased phosphorylation of AMPK and decreased phosphorylation of S6 protein, a downstream target of mTOR (Lai et al. 2018). This autophagy‐inducing activity of NPA in A375 cells was observed to be cytoprotective, as inhibition of autophagy enhanced the cytotoxic and pro‐apoptotic effects of NPA. In contrast, our findings in RAW264.7 macrophages indicate that NPA induces autophagy through AMPK activation without affecting mTOR activity. Specifically, we observed no significant changes in the phosphorylation levels of mTOR or its substrate 4E‐BP1 following NPA treatment. Additionally, we assessed the nuclear translocation of TFEB, another mTOR‐regulated substrate, and found no evidence of NPA‐induced TFEB nuclear accumulation in RAW264.7 cells. These discrepancies may reflect cell‐type‐specific responses to NPA and/or differences in the downstream targets used to assess mTOR pathway activity. Collectively, these findings suggest that the autophagy‐modulatory effects of NPA are highly dependent on the cellular context.

In summary, our study showed that NPA promotes autophagy, thereby enhancing the clearance of intracellular Mtb in host macrophages. Elucidating the underlying molecular mechanisms of NPA‐induced autophagy may aid in the development of innovative adjunctive strategies to bolster host immune defenses against TB.

Author Contributions

Thomanai Lamtha: investigation, formal analysis, writing – original draft, writing – review and editing. Olabisi Flora Davies‐Bolorunduro: investigation, formal analysis, writing – review and editing. Sureeporn Phlaetita: investigation, formal analysis, writing – review and editing. Chernkhwan Kaofai: investigation, formal analysis, writing – review and editing. Phongthon Kanjanasirirat: investigation, formal analysis, writing – review and editing. Tanawadee Khumpanied: investigation, writing – review and editing, formal analysis. Napason Chabang: investigation, formal analysis, writing – review and editing. Bamroong Munyoo: investigation, formal analysis, writing – review and editing. Patoomratana Tuchinda: supervision, formal analysis, writing – review and editing. Suparerk Borwornpinyo: formal analysis, supervision, writing – review and editing. Supawan Jamnongsong: investigation, formal analysis, writing – review and editing. Somponnat Sampattavanich: supervision, formal analysis, writing – review and editing. Prasit Palittapongarnpim: formal analysis, supervision, writing – review and editing. Marisa Ponpuak: conceptualization, writing – original draft, writing – review and editing, formal analysis, supervision, funding acquisition.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: mmi70025‐sup‐0001‐FigureS1‐S4.pdf.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.