The important role of Perforin in protecting against Mycobacterium avium infection in mice

Abstract

Mycobacterium avium is a drug-resistant bacterium that causes refractory respiratory infection. Perforin protects hosts against viral infection and tumors by inducing apoptosis. It is also thought to be important in innate immunity against intracellular pathogen infection, although its role remains controversial. The role of perforin in M. avium infection prevention was examined using perforin-deficient mice. Clinically-isolated strains of M. avium were used, along with perforin-deficient and wild-type C57Bl/6 mice. M. avium (1 × 10⁷ CFU/mouse) was administered intratracheally. Mice were euthanized 7, 21, and 60 days after infection. Lung homogenates were inoculated onto Middlebrook 7H10 agar plates for colony counting. Bronchoalveolar lavage was also performed. Lung tissue sections were stained with hematoxylin-eosin and the Ziehl–Neelsen method. Terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling using lung histology, and a cell death detection kit using lavage fluid, were used to detect apoptosis in the lungs. Perforin-deficient mice demonstrated severe pulmonary involvement after M. avium infection compared with wild-type mice. Colony counts indicated increased M. avium in the lungs of perforin-deficient mice. More apoptotic cells were detected in the lungs of wild-type mice compared with perforin-deficient mice. Histone-complexed DNA fragments were more prevalent in lavage fluids from wild-type mice compared with perforin-deficient mice. These results confirmed that perforin plays an important role in protection against M. avium infection in mice through the induction of apoptosis in infected cells. Perforin might therefore be a therapeutic target for drug-resistant pulmonary M. avium infection.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-16943-0.

Introduction

Mycobacterium avium, also known as Mycobacterium avium-intracellulare complex (MAC), causes difficult-to-treat respiratory infections^1–3. Currently, there are few effective medications able to eradicate MAC². In previous studies, although almost 80% of infections responded to a clarithromycin-containing regimen^4–6 half of the patients experienced relapse within three years^7,8. The incidence rate of pulmonary MAC disease is increasing⁹ indicating that new medication is required to control this disease. However, the drug resistance of MAC is a major obstacle. As well as developing pharmaceuticals, we should seek other therapies to control pulmonary MAC disease aside from antimicrobial agents, such as modification of the host defense system.

Apoptosis, the process of programmed cell death, is currently an area of extensive research because of its relationship to many diseases. Apoptosis plays an important role in the immune response to infection. Cytotoxic T lymphocytes (CTLs) and natural killer (NK) cells have cytoplasmic granules that contain proteins, perforin, and granzymes. Perforin molecules insert themselves into the plasma membrane of target cells when the CTL or NK cells bind to their target. Then, granzymes, which are serine proteases, enter the cells and cleave the precursors of caspases, activating these enzymes. Finally, perforin/granzymes cause the cell to self-destruct via apoptosis^10–12. In this manner, perforin induces apoptosis of cancer cells through different pathways, such as Fas/Fas ligands or tumor necrosis factor (TNF)-death domains¹⁰ thereby protecting the host. Besides tumor immunity, perforin is required for protection against virus proliferation, as it also induces apoptosis in virus-infected cells^13,14. Perforin knock-out (PKO) mice have been reported to be susceptible to virus infection, as well as intracellularly proliferating bacteria¹⁵.

Regarding mycobacterial infection, several reports have shown that M. avium induces apoptosis in vitro^16,17. Moreover, TNF-alpha receptor 1 (TNFR1)-KO mice are susceptible to M. avium infection¹⁸ since TNFR1 possesses a death domain that induces apoptosis. By contrast, in vivo, apoptosis plays a limited role in Mycobacterium tuberculosis infection^19,20. We have recently discovered that apoptosis contributes to the host defense system against M. avium infection²¹; however, knowledge surrounding the role of perforin as well as the role of apoptosis in M. avium infection remains limited. In this study, we investigated the role of perforin as well as apoptosis in M. avium infection using PKO mice.

Results

To investigate the role of perforin in M. avium infection, the expression of perforin was first determined. Immunohistochemistry predominantly detected perforin in infiltrating mononuclear cells in the lung tissues of wild-type (WT) mice after M. avium inoculation (Fig. 1). Preliminary immunoblot examination revealed that perforin was expressed after M. avium administration (Supplementary data) and returned to normal levels at day 60 after administration. Inflammation likely contributed to the expression of perforin. M. avium infection provoked more severe pathological changes in PKO mice compared with WT mice (Fig. 2). Inflammatory cells, mainly lymphocytes, infiltrated the bronchi and adjacent areas in PKO mice, whereas little inflammatory cell infiltration was detected in WT mice. Both clinical isolates of M. avium led to similar results. Ziehl–Neelsen (Z-N) staining also demonstrated proliferation of acid-fast bacilli with pathological involvement in the lungs of PKO mice (data not shown). Morphometric analyses using lung histology at day 60 after M. avium infection also demonstrated more severe pathologic involvement in PKO mice than in WT mice. In both WT and PKO mice, no mortalities occurred after M. avium administration. Bacterial growth in the lungs on day 21 after inoculation was investigated and higher growth of M. avium was observed in the lungs of PKO mice compared with WT mice (Fig. 3A). The pathological changes observed with hematoxylin-eosin (H-E) staining and bacterial growth were correlated. Then, we investigated M. avium proliferation within macrophages. There was no difference between M. avium growth in vitro in macrophages from PKO mice and those from WT mice at days 1 and 2 after inoculation (Fig. 3B). These results suggested that perforin contributed to a mechanism other than bacterial killing or phagocytosis. Perforin is considered to be a key player in apoptosis, so we next investigated apoptosis in M. avium-infected PKO mice. We observed more TUNEL-positive cells (apoptotic cells) in the lungs of WT mice compared with PKO mice at 21 days after M. avium inoculation. TUNEL-positive cells were considered to be macrophages and lymphocytes, based on lung histology (Fig. 4). A cell death detection assay on bronchoalveolar lavage (BAL) fluids on day 21 after M. avium inoculation was also performed. WT mice had higher amounts of histone-complexed DNA fragments than PKO mice (Fig. 5A), which was considered to be another indicator of apoptosis. High levels of cell death correlated with low bacterial growth and minor pathological changes. BAL fluid analysis indicated that alveolar macrophages were increased; lymphocytes and neutrophils levels also tended to be elevated, but not significantly (Fig. 5B). Alveolar macrophages might be considered to be the source of histone-complexed DNA fragments.

Fig. 1.

Perforin expression after M. avium infection. The results of immunohistochemistry of perforin proteins in the lung tissues of mice. Perforin was detected predominantly in infiltrating mononuclear cells in the lung tissues obtained 21 days after M. avium infection (B), but not in sterile saline-treated mice (A). Original magnification: 250×.

Fig. 2.

Lung histology of mice after M. avium administration. Lung histology was obtained 21 days after M. avium administration. (A) Lung section from WT mice administered M. avium and (B) Lung section from PKO mice administered M. avium. Note the marked inflammatory cell infiltrate and granulomatous inflammation in the lung section from PKO mice (B). All panels are H-E stained and viewed at the same magnification (scale bar = 500 μm). Original magnification: 40×. (C) The effect of perforin deficiency (PKO, black diamond) on the pathological grade of M. avium infection in mice. Each square corresponds to the data from one mouse. The results were compared with WT (white diamond) mice treated with M. avium. * indicates a significant difference (p < 0.05, p = 0.011).

Fig. 3.

Bacterial growth of M. avium in the lungs and macrophages. (A) Bacterial growth in the lungs was measured on day 21 after M. avium inoculation. Each group consisted of five mice. * indicates a significant difference (p < 0.05, p = 0.016). (B) Macrophages were obtained from either PKO or WT mice and infected with M. avium at a multiplicity of infection = 10. Bacterial growth was counted at days 1 and 2 post-infection and is shown by a log scale in comparison with day 0. Each group consisted of five mice. There was no significant difference between PKO and WT mice.

Fig. 4.

Effects of perforin deficiency on terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) staining. No positive signals for TUNEL were detected in normal lung parenchyma. (A) TUNEL-positive cells in lung tissues were investigated at 21 days after M. avium infection. The TUNEL-positive signals (arrowheads) were decreased in PKO mice. Original magnification: 250×. (B) The number of TUNEL-positive cells in M. avium infection in mice. Data are shown as the mean obtained from five mice. ** indicates significant differences (p < 0.01, p = 0.0079).

Fig. 5.

Effects of perforin deficiency on histone-complexed DNA fragments in BAL fluids after M. avium infection. (A) Histone-complexed DNA fragments, as an indicator of apoptosis, were measured using BAL fluids. Higher titers were observed in WT mice infected with M. avium compared with PKO mice. Data are shown as the mean obtained from five mice. * indicates differences (p < 0.05, p = 0.028). ** indicates significant differences (p < 0.01, p = 0.0097). (B) Cell counts in BAL fluids at 21 days after M. avium infection. Macrophage accumulation was observed in PKO mice. PKO, black bar; WT, white bar. * indicates a significant difference (p < 0.05, p = 0.0478). N.S. indicates no significant difference.

Discussion

Perforin plays an important role in cell-injury mechanisms involving CTLs^22,23. It induces apoptosis via different pathways, such as by interacting with Fas/Fas ligands or TNF-death domains¹⁰. It plays a critical role in protecting hosts by inducing apoptosis in cancer cells. Besides tumor immunity, perforin has been reported to be associated with various disorders. Perforin and granzyme B are expressed in lymphocytic infiltrates in patients with primary Sjogren’s syndrome²⁴ and perforin is associated with human acute respiratory distress syndrome²⁵; in addition, it contributes to muscle fiber injury in polymyositis and dermatomyositis²⁶. Recently, a perforin gene mutation was shown to play a critical role in hemophagocytic lymphohistiocytosis²⁷. Perforin is required for protection against virus proliferation by inducing apoptosis in virus-infected cells^13,14; however, knowledge about its role in bacterial infection is lacking. In this study, we investigated the role of perforin in M. avium infection. Perforin protein levels were upregulated after M. avium infection. Immunohistochemistry of the lung after M. avium infection revealed the presence of perforin in areas of inflammation. In previous research, perforin was shown to be upregulated by lung injury²⁸ indicating that it might play a role in the early phase of inflammation. We also examined whether perforin deficiency affected M. avium infection in mice. PKO mice demonstrated attenuated M. avium infection based on lung histology and bacterial growth, and severe pulmonary involvement after M. avium inoculation compared with WT mice. Lung histology, examined by Z-N staining and colony counts, indicated increased M. avium in the lungs of PKO mice. These data suggest that perforin plays a pivotal role in innate immunity against M. avium infection. However, several studies have indicated that perforin was not important for M. tuberculosis infection. For example, Laochumroonvorapong et al.. reported that the absence of either perforin or Fas receptor gene function did not modify the course of experimental mycobacterial infection in mice²⁹; Cooper et al.. also indicated that M. tuberculosis infection did not express perforin and does not appear to influence the course of infection¹⁹. The present results indicate that there is a difference in the protective immunity mechanisms involved in M. tuberculosis and those of M. avium infection.

In vitro studies using macrophages revealed no difference in M. avium proliferation. Therefore, we suspected that neither phagocytosis nor bacterial killing contributed to attenuation of M. avium in PKO mice. Perforin and granzyme B are known to induce apoptosis^10,11,22. Perforin is a cytotoxic pore-forming protein, but is unable to induce DNA fragmentation individually³⁰. Granzyme B is a serine protease released by CTLs and NK cells that triggers a series of biochemical events that lead to apoptosis³¹. Granzyme B remains inactive without the signal from perforin to initiate apoptosis³². These two cytotoxic granules together produce the characteristic features of T-cell-mediated apoptosis³³. In the present study, therefore, we investigated the effect of perforin deletion on apoptosis. More apoptotic cells were detected in the lungs of WT mice compared with PKO mice as shown by TUNEL staining. Additionally, larger quantities of histone-complexed DNA, an indicator of apoptosis, were detected in BAL fluids from WT mice compared with PKO mice. These data indicated that apoptosis induced by perforin plays a role in protective immunity against M. avium.

As previously described, perforin was expressed in CTL, CD8 + T cells. Evidence has already been reported that CD8 + T cells play an important role against M. avium and M. abscessus infection^34,35. Inversely, M. avium induced the apoptosis of CD8 + cells³⁶. In this study, we confirmed that perforin deficiency directly led to M. avium susceptibility in vivo.

Apoptosis is known to be induced by TNFR1. Studies have indicated that TNFR1-deficient mice are susceptible to M. avium infection^18,37. Macrophage apoptosis is reported to contribute to host defense mechanisms against mycobacterial infection^16,17,38. Non-infected macrophages phagocytose infected macrophages to eliminate M. avium and prevent the spread of the pathogen³⁹. Bermudez et al.. reported that M. avium-infected macrophages demonstrated higher levels of apoptosis than uninfected macrophages³⁴. Apoptosis could prevent the spread of mycobacterial infection by sequestering pathogens within apoptotic bodies⁴⁰. Lu et al.. demonstrated that human NK cells can kill Mycobacterium kansasii or M. tuberculosis using perforin and granulysin⁴¹. Mycobacteria induce enhanced expression of the cytolytic proteins via activation of the NK group 2D/natural cytotoxic receptor cell-surface receptors and intracellular signaling pathways involving extracellular signal-regulated kinases, c-Jun amino-terminal kinases, and p38 mitogen-activated protein kinases. We also previously reported that Fas/FasL KO mice were susceptible to M. avium infection²¹. Taken together, this evidence suggests a crucial role for apoptosis in macrophages during protective immunity against M. avium infection. However, it is difficult for the use of an apoptosis inducer for pulmonary MAC disease in actual clinical practice. As an alternative, we considered herbal medicines, which are widely used clinically in Japan, and some of these herbal medicines have been shown to have the potential to induce apoptosis⁴².

The intracellular bacterial growth in macrophages from PKO mice was similar to that in WT mice. The phagocytosis ability was also similar between PKO and WT mice as reflected by the bacterial numbers in the M. avium lysate immediately after inoculation. We think that attachment to the dish may compromise the phagocytosis function for apoptotic cells; hence, macrophage phagocytosis ability in vitro may differ from that in vivo. The precise mechanism involved requires further study.

There were several limitations in this research. First, we only investigated the relationship between perforin and M. avium infection, and whether apoptosis directly affects the immune response against M. avium infection requires clarification. Second, only female mice were used. Sex differences, such as hormone levels, may affect the results and male mice should be included in future studies. Third, phagocytosis ability was not sufficiently evaluated in this study, and comprehensive in vivo studies are required in the future. Fourth, histone-complexed DNA fragments are no longer considered direct indicators of apoptosis but are a surrogate for cell damage⁴³. The possibility of lung tissue damage should be ruled out in future studies. Lastly, the role of apoptosis in PKO mice was one possibility, but other mechanisms should still be investigated. For example, other immune factors, such as cytokine profiles, could be analyzed along with perforin deficiency as potential compensatory mechanisms.

In conclusion, the data in this study show that perforin plays an important role in protective immunity against M. avium infection in mice through the induction of apoptosis in infected cells. Verification of the role of perforin in humans is needed in future studies. However, despite the limitations of this study, the data indicate the possibility that perforin may be a new therapeutic target for drug-resistant pulmonary M. avium infection.

Methods

Bacterial strains

In the present study, we used two strains of M. avium, which were isolated from our hospital⁴⁴. We cultured the bacteria using either liquid culture medium or solid agar plates. As liquid culture medium, Middlebrook 7H9 broth with Middlebrook ADC enrichment (Becton, Dickinson and Company, Sparks, MD, USA) was used. Bacteria were grown at 37 °C with shaking. As solid culture medium, Middlebrook 7H10 agar plates with Middlebrook OADC enrichment (Becton, Dickinson and Company) were used. Bacteria were grown at 37 °C for 14 days before colonies were counted⁴⁵.

Mice

Eight-week-old female PKO mice (of 25–30 g bodyweight) were kindly provided by Dr H. Hengartner at the University of Miami, USA³³. Eight-week-old female wild-type C57BL/6 mice (WT) were used as controls. Mice were housed in an environmentally controlled room and provided with sterile food and water.

Animal model of M. avium infection

M. avium (1 × 10⁷ CFU/mice) in 50 µl of sterile saline was intratracheally administered through tracheotomy under tribromoethanol anesthesia^46–48. As a control, 50 µl of sterile saline without bacteria was injected. At days 7, 21, and 60 after M. avium infection, mice were euthanized by cervical dislocation under tribromoethanol anesthesia. The lungs were dissected and homogenized with sterilized stainless steel mesh. The lung homogenates were inoculated onto Middlebrook 7H10 agar plates for colony counting. Experiments were repeated several times.

This study was approved by the Institutional Animal Care and Use Committee (IACUC) of Kyushu University (11–77, 17-010-0) and IACUC of Fukuoka University (1302628, 1805006) and followed the Guidelines for Animal Experimentation from Fukuoka University. The IACUC is charged with protecting the safety and welfare of animals used in research at or in conjunction with Fukuoka University.

Lung histology and morphometry

The lungs were fixed with 10% formalin for 24 h, embedded in paraffin, and 4-µm sliced sections were stained using the hematoxylin and eosin (H-E) and Ziehl–Neelsen (Z-N) methods. Morphological evaluation of lung sections was completed as previously described, with some modifications⁴⁶. Briefly, 30 randomly chosen regions per tissue slide were assessed and scored on a scale of zero to five. Grade 0 was a normal lung, Grade 3 demonstrated lung injury with definite damage to the lung structure, and Grade 5 was defined as total obliteration of the field.

Immunohistochemistry

Immunohistochemistry was performed as previously described⁴⁹. Briefly, endogenous peroxidase activity was quenched by incubating tissue sections with 3% H₂O₂ for 30 min. After rinsing with phosphate-buffered saline (PBS), slides were incubated with normal control serum. Then, the slides were reacted with a 1:100 dilution of goat anti-perforin polyclonal antibody (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA) at 4 °C overnight. As a control, non-specific goat IgG was used. After incubation with a secondary antibody conjugated with horseradish peroxidase, slides were treated with diaminobenzidine.

Bronchoalveolar lavage

Bronchoalveolar lavage (BAL) was performed 21 days after M. avium infection following a previously described method^44,47. The lungs were lavaged five times with 1 ml PBS through a tracheal intubation tube. The number of cells in the lavage fluid was counted with a hemocytometer. Fixed cells were stained with modified Wright’s stain (DiffQuik; American Scientific Products, McGas Park, IL, USA). Differential counts of BAL fluid were performed on 200 cells.

Macrophage isolation and infection by M. avium

Bacterial clearance by macrophages in vitro was examined using a previously described method⁵⁰ with some modifications. Briefly, after peritoneal injection with 2 ml of 3% thioglycollate medium (Becton, Dickinson and Company, Franklin Lakes, NJ, USA), ascites was obtained. The cells in the ascites were mostly macrophages. The macrophages were cultured in RPMI1640 medium without antibiotics. A multiplicity of infection of 10 M. avium was added to the culture medium. After incubation, macrophages were collected, washed twice in PBS, then lysed in 1.0 ml of sterilized water. The lysed medium was inoculated onto Middlebrook 7H10 agar plates. After two weeks of incubation at 37 °C, bacteria were enumerated.

Apoptosis detection

Terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) using lung histology was used to detect apoptosis in the lungs. Lung specimens were first stained using the In Situ Cell Death Detection Kit (Boehringer Mannheim, Indianapolis, IN, USA) according to the manufacturer’s protocol⁵¹. Cell death detection was performed on BAL fluids adjusted to 1000 cells/20 µl using ELISA PLUS (Roche Diagnostics GmbH, Penzberg, Germany) according to the manufacturer’s protocol. Cell death detection ELISA PLUS can determine cytoplasmic histone-associated DNA fragments after induced cell death^21,52.

Western blot analysis

Western blotting was performed as previously described²¹. Lung homogenates were analyzed by immunoblotting using a 1:2000 dilution of goat anti-perforin polyclonal antibody. Lung homogenates were resolved under reducing conditions by SDS-PAGE, then electroblotted onto a polyvinylindene fluoride membrane at 100 V for 1 h in transfer buffer of 25 mM Tris (pH 8.3), 192 mM glycine, and 20% vol/vol methanol. The blots were then blocked overnight in Tris-buffered saline (20 mM Tris, pH 7.6) with 0.1% Tween and 1% bovine serum albumin. After blocking, the blots were incubated for 1 h with an antibody, followed by incubation with the horseradish peroxidase-labeled secondary antibody. Finally, the blots were reacted with a chemiluminescent detection substrate using the ECL Western Blotting Substrate kit (Thermo Fisher Scientific, Tokyo, Japan).

Statistical analysis

The data were expressed as the mean ± standard error (SE) and were analyzed by the Mann–Whitney U test or the Student’s t-test using StatView 5.0 (SAS Institute Inc., Cary, NC, USA). A value of p < 0.05 was considered to indicate a significant difference.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

MAC

Mycobacterium avium-intracellulare complex

CTL

Cytotoxic T lymphocytes

PKO

Perforin knock-out

TNF

Tumor necrosis factor

TNFR1

TNF-alpha receptor 1

WT

Wild-type

IACUC

Institutional Animal Care and Use Committee

H-E

Hematoxylin and eosin

Z-N

Ziehl–Neelsen

PBS

Phosphate-buffered saline

BAL

Bronchoalveolar lavage

TUNEL

Terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling

SE

Standard error

NK

Natural killer

Author contributions

M.F., T.I., Y.S., and H.O. conceived and carried out all experiments and prepared the manuscript, while R.O. and T.M. performed the animal studies.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

This study was approved by the IACUC of Kyushu University (11–77, 17-010-0) and IACUC of Fukuoka University (1302628, 1805006) and followed the Guidelines for Animal Experimentation from Fukuoka University. The IACUC are charged with protecting the safety and welfare of animals used in research at or in conjunction with Fukuoka University. Informed consent was not applicable. The study is reported in accordance with ARRIVE guidelines (https://arriveguidelines.org).

Consent for publication

All authors read the manuscript and agreed with submission. They guarantee that the manuscript has not been published elsewhere and is not being considered for publication elsewhere.