Mycobacterium avium subsp. hominissuis, isolated from a patient with pulmonary MAC disease, was used as the M. avium complex (MAC) bacterium, the same bacteria used in the previous report (14). The bacterium was grown to mid-log phase in Middlebrook 7H9 liquid medium (Difco/Becton Dickinson), aliquoted, and frozen at −80 °C until use. Bacterial counts in each organ were determined by plating serial dilutions of organ homogenates from individual mice on Middlebrook 7H10 agar plates and counting bacterial colonies (colony forming unit, CFU) 2 weeks after plating.

Wild-type BALB/c mice were purchased from The Jackson Laboratory Japan (Yokohama, Japan). Nrf2^−/− mice were generated as described (15) and backcrossed with BALB/c mice for nine generations. Female wild-type mice (8 to 12 weeks old [young mice] and 18 months old [old mice]) and female Nrf2-deficient mice (8 to 12 weeks old) were infected with MAC by intranasal inoculation at a dose of 1 × 10⁷ CFU in 50 μL of phosphate-buffered saline (PBS). Control mice were treated with 50 μL of PBS. After anesthesia with isoflurane, the MAC bacterial suspension or PBS was administered repeatedly intranasally, one drop at a time. Wild-type BALB/c mice and Nrf2-deficient mice were mated and bred in the specific pathogen-free (SPF) facility (Laboratory Animal Resource Center in University of Tsukuba). Female mice were raised 5–7 mice per cage. Infection was performed at P2 level infection laboratory of Laboratory Animal Resource Center in University of Tsukuba. All animal studies were approved by the Institutional Review Board in University of Tsukuba (approval number 24-269).

Lung sections were stained with hematoxylin and eosin. Ziehl-Neelsen staining was used to detect bacilli. Paraffin-embedded sections of lung tissue were immunostained with an antibody against SLPI (NOVUS: NBP1-76803) using the universal immuno-enzyme polymer method (Histofine Simple Stain; Nichirei, Tokyo, Japan). For immunofluorescence staining, lung sections were incubated with appropriately diluted primary antibodies (SLPI [NOVUS: NBP1-76803] and F4/80 [Cell Signaling TECHNOLOGY: #70076S]). Secondary antibodies were species-specific, matching the source of the primary antibodies (SLPI [Alexa Fluor 488] and F4/80 [Alexa Fluor 594]). Finally, the sections were counterstained with DAPI (VECTASHIELD MOUNTING MEDIUM). The stained sections were then observed, with a fluorescence microscope, OLYMPUS DP70 (OLYMPUS, Japan).

The lungs were lavaged with six sequential 1-mL aliquots of saline. Cells were counted using a hemocytometer, and differential cell counts were obtained by staining with Diff-Quick (Polysciences, Inc.) after cytospins.

Total RNA was extracted from lungs using the RNeasy mini kit (Qiagen). Real-time quantitative reverse transcription-PCR (RT-PCR) was performed using QuantStudio 5 (Applied Biosystems). The PCR primers used in this study are listed in Supplementary Table 1. Target gene expression levels were calculated using the ΔΔCT method and normalized against glyceraldehyde 3-phosphate dehydrogenase mRNA.

Mouse SLPI/anti-leukoproteinase recombinant (His+S) protein (aa20-131) (LifeSpan Biosciences, Inc.) was used. The SLPI protein was dissolved in PBS. The MAC suspension was diluted with 7H9 liquid. M. avium (2.0 × 10⁶ CFU/ml) was incubated with SLPI (3 μM) or PBS as a control at 37 °C for 7 days. Colonies were counted (CFU/ml) at 24 h, 3 days, and 7 days.

Nuclear extracts were prepared from lung tissue using a nuclear extraction kit (Invent Biotechnologies, Inc., Plymouth City, MN, USA), according to the manufacturer’s instructions. Ten to twenty micrograms of nuclear extracts were separated with 10% SDS-PAGE gels and transferred to polyvinylidene difluoride (PVDF) membranes. After blocking for nonspecific sites, the PVDF membranes were incubated with anti-Nrf2 antibody (D1Z9C), followed by incubation with secondary antibodies (IRDye^® 680RD donkey anti-rabbit IgG; LICOR bio). Immunoreactive bands were visualized using a near-infrared fluorescence imaging system (ODYSSEY CLx, LICOR bio). Lamin B was used as an internal control. Values were normalized to lamin B to evaluate Nrf2 expression.

R, S-sulforaphane (SFN; LKT Laboratories Inc., St. Paul, MN, USA) was used in this study. A stock solution of SFN was prepared using ethanol as the solvent and stored at −20 °C in the dark. The SFN stock solution was diluted with PBS immediately before use. SFN was injected intraperitoneally into 18-month-old mice at a dose of 5 mg/kg, 5 days/week for 4 weeks immediately after intranasal inoculation of MAC.

Total RNA was extracted from lung tissues using TRIzol with a homogenizer in 12 individual samples (3 from the lungs of the young mice treated with PBS, 3 from the lungs of the old mice treated with PBS, 3 from the lungs of the young mice infected with M. avium for 2 months, and 3 from the lungs of the old mice infected with M. avium for 2 months); 500  ng of total RNA were used for RNA-seq. Sequencing was performed with NextSeq500 (Illumina) by Tsukuba i-Laboratory LLP (Tsukuba, Ibaraki, Japan). FASTQ files were analyzed using CLC Genomics Workbench (CLC-GW, version 10.1.1; Qiagen). Reads were mapped to the mouse reference genome (mm10) and quantified for annotated genes. The Empirical Analysis of DGE tool in CLC-GW was used to detect differential expression of genes (FDR- adjusted P<0.001 and with more than 2-fold changes). The data are available under GEO series accession number GSE287501. Heat maps were generated using Morpheus (https://software.broadinstitute.org/morpheus/), and Venn diagrams were generated using Venny 2.1.0 (https://bioinfogp.cnb.csic.es/tools/venny/).

A multicenter, non-interventional, prospective, observational study was performed. The study protocol was approved by the ethics committees of Tsukuba University Hospital and other hospitals, and the study was performed in accordance with the Declaration of Helsinki (approval number R01-379). Written, informed consent was obtained from all participating patients. Between May 2020 and December 2024, 100 patients consented to participate after being screened for eligibility.

The participants were 100 patients newly diagnosed with pulmonary MAC disease according to ATS/ERS/ESCMID/IDSA criteria and scheduled to receive guideline-based therapy (GBT) at the discretion of their treating physicians. Baseline characteristics prior to GBT were obtained from patients’ medical records, including age, sex, body mass index (BMI), smoking status, comorbidities, presence of respiratory symptoms, St. George’s Respiratory Questionnaire (SGRQ) score, and laboratory test results. Laboratory data were collected, including white blood cell count, neutrophil count, lymphocyte count, monocyte count, eosinophil count, basophil count, albumin, CRP, and GPL-core IgA antibody. A GPL core IgA antibody titer ≥0.7 U/mL was considered positive. For radiological evaluation, the locations of the lesions and the number of lobes involved were analyzed. The locations of the lesions were evaluated in separate sections: upper, middle (lingula), and lower lobes, with the lingula considered a separate lobe. The number of lobes with lesions was defined as the computed tomography (CT) score. As for the BACES score, a CRP of 0.3 or higher (1 point) was used as a substitute for an elevated erythrocyte sedimentation rate (ESR) (1 point), since ESR was not measured in this study. This new score was called the BACScrp score instead of the BACES score. Sputum samples were inoculated into MGIT liquid media and Ogawa solid media.

Blood was collected from all participants before treatment using PAXgene blood RNA tubes (Qiagen). Silica membrane-based RNA isolation and purification in a spin-column format were performed using the PAXgene Blood RNA kit (Qiagen). Sequencing was performed by Tsukuba i-Laboratory LLP using NextSeq500 (Illumina) for whole blood RNA. Sequencing reads were mapped to the hg19 human reference genome and quantified using CLC Genomics Workbench version 10.1.1 (Qiagen). Differentially expressed genes were identified by filtering according to the P-values obtained by Empirical Analysis of DGE tool in CLC-GW. Genes were identified as differentially expressed if they had an adjusted (Benjamini-Hochberg FDR method for correction of multiple testing) p-value of 0.05. Data are available under the GEO series accession number GSE 288604.

To determine the effect of aging on MAC infection in mice, the survival rate, bronchoalveolar lavage fluid (BALF), organ bacterial load, and lung pathology of mice 2 months after MAC infection were examined. The survival rate after infection was significantly lower in old mice than in young mice (Figure 1A). The number of inflammatory cells in BALF was not significantly different between young and old mice 2 months after MAC infection (Figure 1B). Organ colony-forming unit (CFU) measurement showed higher mycobacterial counts in lungs, livers, and spleens of old mice than in young mice 2 months after MAC infection (Figure 1C). Lung pathology 2 months after infection was evaluated by HE staining. Both young and old infected mice showed inflammatory cell infiltration in the lungs, but granuloma formation was poor in infected old mice (Figure 1D). These results suggest that old mice are more susceptible to MAC infection than young mice.

To understand the detailed mechanisms explaining why old mice have high susceptibility to MAC infection, RNA-seq of lung tissues was performed using young and old mice with or without MAC infection. The mRNA expression levels were compared and analyzed between uninfected and infected young lungs, between uninfected and infected old lungs, and between infected young lungs and infected old lungs (FDR-adjusted P<0.001 and with more than 2-fold changes) (Supplementary Table 2; Figures 2A, B). There were 17 genes that were upregulated in both young and old mice 2 months after infection, and their expression levels were significantly different between infected young mice and infected old mice (Figures 2A, C). In addition, there was one gene that was downregulated in both young and old mice 2 months after infection, and its expression level was significantly different between infected young and infected old mice (Figures 2B, D). Of these 18 genes, secretory leukocyte protease inhibitor (SLPI), which encodes a peptide with antimicrobial and antiprotease activity, was selected for evaluation as a potential gene involved in the control of MAC infection.

To validate the RNA-seq data, SLPI mRNA expression was assessed by qPCR in lung tissues from young and old mice infected with MAC bacteria. Similar to the RNA-seq result, the expression of SLPI was induced in both mice after MAC infection, but it was significantly decreased in infected old mice compared to infected young mice (Figure 3A). To confirm the expression and localization of SLPI protein in MAC-infected lung tissues, SLPI immunostaining was performed using anti-SLPI antibody. SLPI was expressed in the bronchial epithelial cells of both young and old mice at the same level before infection. In the infected lung tissue, there was no change in the expression of SLPI protein in bronchial epithelial cells. However, SLPI expression was observed in inflammatory cells, mainly macrophages, in both mice and was higher in young mice than in old mice (Figures 3B, C).

To examine the antibacterial effect of SLPI on MAC bacteria, SLPI protein was added directly to MAC bacteria. After 24 hours of incubation, the SLPI-treated group showed a significant reduction in the amount of MAC bacteria compared with the control group, to almost zero CFU (Figure 4A). After 3 and 7 days of incubation, the SLPI-treated group showed no increase in the number of bacteria throughout the observation period, whereas the control group showed an increase in the number of MAC bacteria (Figure 4B). These results suggest that SLPI itself has an antibacterial effect on MAC bacteria.

SLPI has been reported to be regulated by Nrf2 (11–13). It has also been reported that Nrf2 activation decreases with aging (9, 10). Therefore, a relationship between Nrf2 activation and SLPI expression in the lungs of MAC-infected old mice was hypothesized, and whether nuclear translocation of Nrf2 occurs in the lung tissues after MAC infection was investigated. Western blot analysis showed that the degree of nuclear translocation of Nrf2 was higher in the infected young lungs than in the infected old lungs (Figures 5A, B). These results indicated that a significant decrease in Nrf2 activation was observed in the lungs of infected old mice compared with infected young mice.

Next, to confirm that SLPI expression in the lung is regulated by Nrf2 during MAC infection, BALB/c Nrf2-deficient mice were infected intranasally with MAC bacteria, and SLPI expression and bacterial load in the lungs 2 months after infection were examined. SLPI expression was significantly lower in the lungs of Nrf2^-/- mice than in those of wild-type mice (Figure 5C). In addition, organ CFU measurement showed increased mycobacterial counts in the lungs of Nrf2^-/- mice compared with wild-type mice 2 months after MAC infection (Figure 5D). These results suggest that SLPI expression is regulated by Nrf2 and may contribute to infection control in MAC infection.

To confirm whether activation of Nrf2 induces SLPI expression and reduces the growth of MAC bacteria, old mice were treated with SFN during MAC infection. SFN increased the expression of SLPI (Figure 5E). The number of mycobacteria in the lungs was also reduced after SFN treatment (Figure 5F). Thus, it was confirmed that SFN suppressed the growth of MAC correlating with increased lung expression of SLPI and other Nrf2-associated genes in old mice.

Experiments in old mice suggested that decreased Nrf2-regulated SLPI expression by macrophages in MAC-infected lungs is associated with the poor prognosis with age. We hypothesized that, in humans as well, age-related decreases in SLPI expression may contribute to the worsening of pulmonary MAC disease. To investigate the relationships of SLPI expression and age with disease severity in humans, a two-step cluster analysis using patient age and SLPI mRNA expression levels in whole blood cells from 100 patients with pulmonary MAC disease was performed. These 100 patients were newly diagnosed and scheduled to receive GBT at the discretion of their treating physicians. The analysis identified three clusters (Table 1): Cluster 1 (C1), younger patients (median age: 52.0 years) with moderate SLPI expression (median transcripts per million [TPM]: 14.6, 26 cases); Cluster 2 (C2), older patients (median age: 80.5 years) with high SLPI expression (median TPM: 26.7, 18 cases); and Cluster 3 (C3), predominantly older patients (median age: 70.0 years) with low SLPI expression (median TPM: 10.3, 56 cases). Compared to C1, C3 had larger pulmonary lesions on computed tomography (CT). Differentially expressed gene (DEG) analysis showed 88 DEGs between C1 and C3, 46 DEGs between C1 and C2, and 7 DEGs between C2 and C3 (FDR < 0.05) (Figures 6A-C; Supplementary Table 3). Upstream regulator analysis in IPA (Qiagen) showed reduced Nrf2 activation in C3 compared with C1, consistent with the findings from the mouse experiments (Supplementary Figure 1; Supplementary Table 4). These characteristics of C3 closely resembled those observed in MAC-infected old mice, suggesting that reduced SLPI expression may contribute to human pulmonary MAC disease in the context of aging, which is associated with a poor prognosis. Although C2 was also an older group, it was characterized by high SLPI expression (median TPM: 26.7). The CRP level in C2 was higher than in both C1 and C3, and the BACScrp score in C2 was significantly higher than in C1 (Figure 6D). Data from CIBERSORTx, a program that estimates the proportion of immune cells in each cluster based on gene expression profiles, showed a higher proportion of neutrophils in C2 (Table 2; Supplementary Figure 2). In addition, expression of neutrophil-related genes, such as DEFA3 and ELANE, was significantly higher in C2 than in C3. These findings suggest that C2 represents a group of more severe cases in which neutrophil-driven inflammation plays a significant role (Figure 6D).