MWCNTs were obtained from Mitsui & Company (Mitsui-7, XNRI 1, lot #-05072001K28). Carbon black (CB) was purchased commercially (Printex 90, Degussa Engineered Carbons, L.P.). CB are amorphous, carbonaceous particulates and therefore are used as a nonfiber, carbon-based particle material control for MWCNTs. MWCNTs and CB were dispersed in Dulbecco’s Modified Eagle Medium with 1% fetal bovine serum (FBS; both from Thermo Fisher Scientific) at a concentration of 2 mg/mL by vortex and sonication as described previously (Hindman and Ma, 2019; Lim et al, 2023). Properties of the MWCNTs and CB have been characterized and reported previously (Porter et al, 2010; Lim et al, 2020, 2023). The levels of LPS in the MWCNT and CB preparations were determined to be <0.1 EU/mL (<0.01 mg/mL) by using the Pierce LAL chromogenic endotoxin quantification kit (Thermo Scientific). For cell treatment, stock solutions were diluted with the culture media, and the dilution preparation for treatment was sonicated and vortexed immediately before use. The control medium (Dulbecco’s modified Eagle’s medium plus 1% FBS) was prepared and used to establish a baseline response. Some cells were kept untreated and used as a cell background control. All cells were treated in duplicates, and all treatments were conducted for 3 times.

The human acute monocytic cell line THP-1 was acquired from American Type Culture Collection and maintained in the Roswell Park Memorial Institute-1640 medium containing Glutamax and HEPES and supplemented with 1× antibiotic-antimycotic and 10% heat-inactivated FBS (Thermo Fisher Scientific). Cells were maintained at 37 °C in a humidified 5% CO₂ incubator. Cells were passaged at a cell density of up to 5.0 × 10⁵ cells/mL every 3–4 days. Cells from passage 4–10 were used in this experiment.

To differentiate cells into adherent, macrophage-like cells, THP-1 cells were plated at a cell density of 5.0 × 10⁵ cells/mL and incubated for 3 days in the presence of 40 nM phorbol-12-myristate-13-acetate (Sigma-Aldrich). Polarization into M1 macrophages was induced by incubation of differentiated THP-1 cells with IFN-γ at 20 ng/mL plus LPS at 100 ng/mL (both from Sigma-Aldrich) for 2 days. Polarization into M2 cells was induced by incubation with IL-4 (Sigma-Aldrich) at 20 ng/mL for 2 days. Polarized M1 and M2 cells were washed with warm, sterile PBS and were exposed to MWCNTs or CB for 1 or 3 days at concentrations indicated in figures. Inhibition of the NLRP3 inflammasome pathway and pyroptosis was carried out by using the caspase-1 inhibitor AC-YVAD-CMK at 25 and 50 μg/mL, the NLRP3 inflammasome inhibitor MCC-950 at 0.5 and 1 μM, and the cathepsin B inhibitor CA-074 methyl ester (CA-074 Me) at 50 and 100 μM (All from Sigma-Aldrich) for 6 hours before the cells were exposed to MWCNTs or CB.

For experiments on the NLRP3 inflammasome, differentiated macrophages were primed with LPS at 100 ng/mL for 24 hours, followed by adenosine 5′-triphosphate (ATP) at 5 mM (both from Sigma-Aldrich) for an additional hour. As a positive control, nigericin at 10 μM (Cayman Chemicals) was applied to induce the NLRP3 inflammasome-mediated IL-1β release. After treatment, the culture medium was collected and centrifuged at 4 °C to remove particles, cells, and cell debris. The cell-free medium was used for the detection of cytokines by ELISA. After washing, the cells were collected and used for further analysis of cells and proteins.

Polarized M1 or M2 macrophages were incubated with an Fc block solution (1× PBS containing rat serum and 25 μg/mL unlabeled anti-CD16/CD32 antibody 2.4G2) for 5 minutes at room temperature. The cells were stained with the FITC-conjugated mouse anti-human CD86 [clone 2331 (FUN-1), #560958] or CD206 (clone 19.2, #551135) antibodies or its isotype-matched control mouse IgG antibody (all from BD Biosciences) for 25 minutes at 4 °C in the dark. The cells were washed with 1 mL of FACs buffer (1× PBS containing 2 mM EDTA and 5% FBS). The cells were fixed using the Cytofix buffer (BD Biosciences) for 10 minutes at room temperature, followed by washing with 1 mL of FACs buffer. Cells were resuspended in the FACs buffer, and flow cytometry was carried out on a BD LSR II flow cytometer (BD Biosciences). Data were analyzed using FlowJo software (V10.6).

Cell culture media were collected from untreated cells (M0), or cells treated with IFN-γ+LPS for M1 cells or IL-4 for M2 cells. Cell-free culture supernatants were prepared by centrifugation at 800 × g, 4 °C to remove particles and cell debris. Inflammatory cytokines, that is, IL-1β, tumor necrosis factor-α (TNF-α), IL-10, and IL-13, were detected by ELISA.

For experiments on IL-1β release, M1 or M2 macrophages were untreated or treated with the control medium, MWCNTs at 2.5 μg or 10 μg/mL, CB at 2.5 μg or 10 μg/mL, or LPS followed by ATP (LPS+ATP), as described above under cell culture and treatment. For inhibition of the NLRP3 inflammasome pathway, cells were treated with AC-YVAD-CMK at 50 μg/mL, MCC-950 at 1 μM, or CA-074 Me at 100 μM for 6 hours before the treatment with MWCNTs at 2.5 or 10 μg/mL. IL-1β and TNF-α were detected by ELISA in cell-free culture supernatants. All ELISA kits were from MyBioSource, and measurement was performed by following the manufacturer’s protocols.

Release of lactate dehydrogenase (LDH) from macrophages into the culture medium indicates cell damage/cell death and was determined using the CyQUANT LDH Cytotoxicity Assay (Thermo Fisher Scientific, C20300). Briefly, cells were seeded in 96-well plates at a density of 5 × 10⁴ cells/well in the Roswell Park Memorial Institute-1640 medium. Following treatment, the cells were incubated with the LDH reaction mixture. Absorbance at 490 and 680 nm was measured, and LDH release was quantified using a BioTek synergy H1 multimode microplate reader (Agilent Technologies). As a control for interference from the absorption of light by particles, MWCNTs or CB were kept in the cell-free medium during LDH measurement. For calculation of the LDH activity, the 680 nm absorbance value (background) was subtracted from the 490 nm absorbance, and then, the % of cytotoxicity was calculated following the manufacturer’s instructions. The experiments were repeated at least 3 times with duplicate measures performed at each treatment concentration.

M1 cells were untreated or treated with MWCNTs at 2.5 or 10 μg/mL, CB at 10 μg/mL, LPS at 100 ng/mL, or nigericin at 10 μM for 24 hours. For experiments on the NLRP3 inflammasome, cells were treated with MWCNTs at 10 μg/mL or LPS at 100 ng/mL for 24 hours, followed by 1 hour of incubation with 5 mM ATP. Cells were collected, centrifuged, and lysed for immediate use. Caspase-1 activity was measured using a colorimetric Caspase-1 activity assay kit (R&D Systems). Samples containing no cell lysate, or a substrate were included as negative controls.

Macrophages were treated as indicated and were lysed in a lysis buffer (10 mM Tris, pH 7.4, 1% SDS) containing a proteinase inhibitor cocktail (Thermo Fisher Scientific). Cell lysates were collected and sonicated for 10 seconds. The supernatant was collected, and protein concentration was determined using a Bicinchoninic acid protein assay kit (Thermo Fisher Scientific). Lysate proteins of 10–20 μg each were resolved on a 10% SDS-PAGE gel and were transferred onto a nitrocellulose membrane. The membrane was incubated with 5% nonfat dry milk in Tris-buffered saline containing 0.05% Tween 20 for 1 hour at room temperature to block nonspecific binding before the incubation with primary antibodies. The primary antibodies used include rabbit anti-GSDMD (1:1000, Abcam, ab219800) for the detection of the full-length GSDMD, rabbit anti-GSDMD-N-term (1:1000, Abcam, ab215203) for detection of GSDMD-N, or mouse anti-β-actin (1:4000, Sigma-Aldrich, A5441) antibodies for loading control. After incubation with either the horseradish peroxidase-conjugated goat anti-mouse (1:5000, Jackson ImmunoResearch laboratories, 115-035-146) or the goat anti-rabbit IgG (1:5000, Jackson ImmunoResearch Laboratories, 111-035-144) antibody, immunoreactive bands were visualized using the Enhanced chemiluminescence substrates (Thermo Fisher Scientific). Band signals were captured onto X-ray film by exposure for 30 seconds, and the film was developed using a film processor (Konica Minolta). Scanned images were used to quantify band intensities using ImageJ software (NIH, USA), and each band was normalized to β-actin.

The expression and localization of cleaved GSDMD, NLRP3, or ASC were examined using immunofluorescence staining and fluorescent microscopy. THP-1 cells were seeded on chamber glass culture slides and differentiated and polarized into M1 cells as described above. The cells were untreated or treated with MWCNTs at 10 μg/mL, LPS plus ATP, or nigericin at 10 μM. The cells were fixed in 1% paraformaldehyde (w/v in 1× PBS), permeabilized in 0.5% Triton X-100 (v/v in 1X PBS), blocked in 5% normal goat serum (v/v in 1× PBS; all from Sigma-Aldrich), and stained with the rabbit anti-GSDMD-N-term (1:100, Abcam, ab215203), rabbit anti-NLRP3 (1:100, Thermo Fisher Scientific, PA520838), or mouse anti-ASC (1:50, Santa Cruz Biotechnology, sc-514414) antibodies in blocking buffer for overnight at 4 °C. Cells were then incubated with 1:1000 dilutions of Alexa 488 goat anti-rabbit or Alexa 555 goat anti-mouse IgG antibodies (Fisher Scientific) for 1 hour before being mounted in a photo-bleach resistant mounting medium containing the 4′,6-diamidino-2-phenylindole counterstain (Vector Labs). Cells were examined under a BX63 automated fluorescence microscope, and images were captured using a 63× oil objective. For colocalization of GSDMD-N-term and ASC, the same cells showing GSDND-N staining and ASC speck staining were counted using the ImageJ software. The percent of colocalization of GSDND-N and ASC was calculated from more than 100 cells for each treatment group. For ASC speck formation, cells showing ASC speck staining were counted and used for the calculation of percent positive cells from more than 100 cells for each treatment group.

Release of cathepsin B from lysosomes is a maker of lysosomal damage and was measured using the Magic Red fluorescent cathepsin B assay kit (Immunochemistry Technologies, 937). Briefly, cells were grown on a 96-well black-walled microplate (Fisher Scientific 07-000-135) at a cell density of 2.5 × 10⁵ cells/mL and were treated as described above. After washing with 1× PBS (37 °C), the cells were stained with the Magic Red substrate reconstituted in DMSO. After incubation for 1 hour at 37 °C, fluorescence at excitation/emission of λ = 592/628 nm was measured using a Synergy H1 microplate reader (BioTek). The cathepsin B activity was calculated and expressed as fold changes for treated groups over controls.

Dependent measurements were analyzed using mixed-model 1-way analyses of variance, with each analysis incorporating the experiment as a random factor. For each experiment, the 2 replicate samples were averaged to give a single value for each treatment combination for each of the 3 independent experiments prepared on different days. For some variables, data were log-transformed to reduce heterogeneous variance and meet the assumptions of the analysis. Post-hoc comparisons were carried out using Fishers least significant difference test. All differences were considered significant at *, P < .05; **, P < .01; ***, P < .001; compared with untreated control or MWCNT-treated group. All analyses were carried out using the JMP version 16 software (SAS Institute).

Polarized macrophages have distinct markers, metabolic and transcriptional profiles, and functions, which enable them to serve as major effector cells in inflammatory and immune responses under physiological and disease conditions (Wynn and Barron, 2010; Ma, 2020). Although fibrogenic MWCNTs have been shown to stimulate M1 and M2 polarization in animal lungs, few studies examined the effect of MWCNTs on polarized macrophages and their fate. Here, we took an in vitro approach to examine the interplay between inflammation and cell death of polarized macrophages upon exposure to MWCNTs.

Macrophages were differentiated from the human monocytic THP-1 cells with phorbol-12-myristate-13-acetate, followed by polarization to M1 or M2 cells with an M1 inducer LPS+ATP or M2 inducer IL-4, respectively. Polarized macrophages exhibited phenotypes of M1 or M2 cells as revealed by their cell surface markers and production of cytokines (Fig. 1). Expression of CD86 was evident after 2 days in M1, but not M2, macrophages, whereas expression of CD206 was markedly increased in M2, but not M1, macrophages (Fig. 1A). Production and secretion of type 1 and type 2 cytokines by polarized macrophages was examined after polarization was induced for 2 days. Cell-free media were assayed for production of IL-1β, TNF-α, IL-10, and IL-13 by ELISA. IL-1β was produced at 78.42 pg/mL by M1 macrophages, but only 18.37 pg/mL from M2 cells. Similar to IL-1β, TNF-α was detected at 27.71 pg/mL from M1 cells and only 2.37 pg/mL from M2 cells (Fig. 1B). On the contrary, IL-10 was ata high level (106.84 pg/mL) from M2 cells, but only at a low level (27.92 pg/mL) from M1 cells. IL-13 also showed 5.23-fold higher production and secretion from M2 than M1 cells (ie, 16.80 pg/mL from M2 versus 3.21 pg/mL from M1 cells; Fig. 1B).

We then evaluated the cellular damage to M1 and M2 cells from MWCNTs by measuring LDH release (Fig. 1C). The cells were exposed to increasing concentrations of MWCNTs as indicated for 1 or 3 days. MWCNTs triggered dose-dependent cell damage/cell death, which were significant at 50 μg/mL and 100 μg/mL of MWCNTs in M1 and, to a lesser extent, M2, macrophages (Fig. 1C). As a comparison, the amorphous carbon material control CB did not cause apparent cell damage in M1 and M2 macrophages, except there was a slight trend for CB to cause an increase in LDH release when tested at 100 μg/mL for 3 days in M1 macrophages (Fig. 1C). The results indicate that MWCNTs, but not CB, caused cell damage and cell death in both M1 and M2 cells, and M1 macrophages are more sensitive to the cytotoxic effect of MWCNTs than M2 cells.

Given the prominent inflammatory effect of MWCNTs in the early phase of the pulmonary response, we examined the inflammatory status of the polarized cells in the presence of MWCNTs. We first compared the secretion of proinflammatory IL-1β from M1 and M2 macrophages exposed to MWCNTs or CB (Fig. 2). MWCNTs at 2.5 or 10 μg/mL induced IL-1β secretion from M1 macrophages with an increase of 2.3-fold or 3.9-fold, respectively, compared with either the untreated control or the control medium-treated cells. MWCNTs at 10 μg/mL slightly increased IL-1β secretion from M2 macrophages, which was not statistically significant. CB did not increase IL-1β secretion from either M1 or M2 macrophages. The NLRP3 inflammasome activator LPS plus ATP was used as a positive control. As expected, LPS plus ATP induced a high level of IL-1β from M1, but not M2, cells.

Production of IL-1β by macrophages requires the activation of caspase-1, which was examined in Fig. 2B. LPS at 100 ng/mL plus ATP at 5 mM (LPS+ATP) or nigericin at 10 μM was used as positive control, which caused large increases in the caspase-1 activity (31.83-fold for LPS+ATP and 36.22-fold for nigericin over control) in M1 cells, but much smaller increases (5.49-fold for LPS+ATP and 9.34-fold for nigericin) in M2 cells (Fig. 2B). LPS alone increased the caspase-1 activity in M1 macrophages by 17.02-fold. Treatment with MWCNTs at 2.5 or 10 μg/mL induced the caspase-1 activity by 7.33-fold and 14.96-fold over control, respectively. Interestingly, ATP was found to augment the increase in caspase-1 activity by MWCNTs at 10 mg/mL, that is, a 25.48-fold increase by MWCNTs plus ATP, compared with the 14.96-fold increase by MWCNTs alone. CB at 10 μg/mL did not increase caspase-1 activity. In M2 macrophages, caspase-1 activation by MWCNTs was much lower than that in M1 macrophages; MWCNTs at 10 μg/mL induced caspase-1 activation by 4.35-fold increase, compared with untreated control. ATP slightly further increased the activity. LPS alone did not significantly increase caspase-1 activity in M2 macrophages.

Activation of caspase-1 indicates the activation of the NLRP3 inflammasome, which may require the release of the lysosome enzyme cathepsin B upon phagocytosis of MWCNTs by M1 macrophages (Ma and Lim, 2024). To examine the lysosomal integrity and release of lysosomal cathepsin B into the cytoplasm in MWCNT-exposed cells, we tested cathepsin B activity using a cathepsin B-specific substrate (Fig. 3A). MWCNTs at 2.5 or 10 mg/mL increased the release of cathepsin B at 24 hours post-exposure by 2.26-fold and 5.47-fold, respectively. No cathepsin B release was noted in cells exposed to CB at 10 μg/mL. LPS and LPS+ATP, but not nigericin (10 mM), caused significant increases in cathepsin B release. These results support the role of the lysosomal release of cathepsin B for NLRP3 activation by MWCNTs.

Activated NLRP3 assembles with the ASC protein through homotypic binding and directs the prion-like, directional oligomerization of ASC into filamentous bundles leading to the formation of ASC specks in the cytoplasm. For this reason, formation of ASC specks is often considered a marker of NLRP3 inflammasome activation in cells, although the ASC speck does not always colocalize with the NLRP3 inflammasome inside a cell. M1 macrophages in the absence of MWCNTs showed a diffused staining of ASC. MWCNTs at 10 μg/mL caused the formation of intensely stained ASC specks with a perinuclear localization (Fig. 3B). As expected, both positive controls, LPS+ATP and nigericin, induced marked increases in the formation of ASC specks in M1 macrophages.

Taken together, these findings presented a path of NLRP3 inflammasomal activation in M1 macrophages by MWCNTs, starting with the phagocytosis of CNTs, followed by lysosomal damage and release of lysosomal cathepsin B, which triggers the assembly of the NLRP3 inflammasome and formation of ASC specks, caspase-1 activation, and the maturation and secretion of IL-1β from M1 macrophages.

One consequence of NLRP3 activation is the activation of pyroptosis through the cleavage of GSDMD and pore formation by the cleavage product, GSDMD-N, in the plasma membrane of the cell (Shi et al, 2015). Formation of the pores enables the secretion of mature IL-1β and IL-18 from the cell, as well as disruption of the osmotic gradient across the plasma membrane leading to the osmotic rupture, and pyroptotic death of the cell and release of the engulfed particles, inflammatory mediators, and other intracellular contents to amplify the inflammatory response. To examine GSDMD activation, M1 cells were treated with MWCNTs at 2.5 or 10 μg/mL, CB at 2.5 or 10 μg/mL, LPS at 100 ng/mL, LPS (100 ng/mL) + ATP (5 mM), or nigericin at 10 μM for 1 day. GSDMD and GSDMD-N were analyzed by immunoblotting using specific antibodies (Fig. 4A). MWCNTs at 2.5 or 10 μg/mL reduced the protein levels of GSDMD by 17.1% and 58.62%, respectively, compared with untreated control. LPS reduced the GSDMD level by approximately 21%. Addition of ATP to LPS further reduced the level by 18.15% (ie, 39.14% compared with untreated control). Nigericin strongly reduced the GSDMD level by 66.7%, compared with untreated control. CB did not affect the GSDMD protein level.

GSDMD-N was detected as a 31 kDa fragment. MWCNTs at 2.5 or 10 μg/mL increased the level of GSDMD-N by 423.5% or 557.3%, respectively, compared with untreated control. LPS increased the level by 398.2%, and the addition of ATP further increased the level by 170% (ie, 568.2%, compared with untreated control). Nigericin increased the level of GSDMD-M by 551.5%, compared with untreated control. CB slightly increased the N-terminal fragment level by 144.2% at 2.5 μg/mL and 134.9% at 10 μg/mL, compared with the untreated control, which was statistically not significant.

To validate if GSDMD cleavage and NLRP3 inflammasome activation are associated with each other in M1 macrophages, we examined the distribution of GSDMD-N in M1 cells using immunofluorescence imaging (Fig. 4B). Untreated cells exhibited no detectable GSDMD-N. MWCNTs at 10 μg/mL induced a diffuse staining pattern of GSDMD-N. Inflammasome activators, LPS+ATP or nigericin, also induced a diffuse expression of GSDMD-N in M1 cells. In both cases, GSDMD-N colocalized with the ASC specks in significant numbers of cells as shown by the quantification of cells with positive colocalization. These results indicate that MWCNTs induced pyroptosis of M1 macrophages via the cleavage of GSDMD through the NLRP3 inflammasome-activated caspase-1.

To substantiate the activation of the NLRP3 pathway and its role in M1 cell pyroptosis, various inhibitors were employed. As shown in Fig. 5A, the MWCNT-induced release of IL-1β into the medium is nearly completely inhibited by preincubation of the cells with specific caspase-1 inhibitor AC-YVAD-CMK at 50 μg/mL and by NLRP3 inflammasome inhibitor MCC-950 at 1 μM. Cathepsin B inhibitor CA-074 Me at 100 μM also blocked IL-1β production and secretion. However, the secretion of TNF-α, another major proinflammatory cytokine secreted by M1 macrophages in response to MWCNTs, was only slightly decreased by the inhibitors (Fig. 5B), indicating that induction of TNF-α in M1 macrophages by MWCNTs does not require the NLRP3 inflammasome.

Specific inhibitors were used to inhibit MWCNT-induced GSDMD cleavage in M1 macrophages (Fig. 5C). Treatment with caspase-1 inhibitor AC-YVAD-CMK at 25 μg/mL prior to MWCNTs partially blocked GSDMD cleavage and GSDMD-N generation. At 50 μg/mL, the inhibitor completely blocked the cleavage of GSDMD cleavage and production of GSDMD-N induced by MWCNTs. The NLRP3 inhibitor MCC-950 at 0.5 or 1 μM, and cathepsin B inhibitor CA-074 Me at 50 or 100 μM, inhibited the generation of GSDMD-N significantly in M1 macrophages.

Taken together, the results from the inhibition experiment support the notion that once phagocytosed by M1 macrophages, MWCNTs trigger lysosomal instability to release cathepsin B into the cytoplasm where a cascade of events was initiated, leading to the NLRP3 inflammasome activation, caspase-1 activation, cleavage of pro-IL-1β and GSDMD, maturation and secretion of IL-1β, and, ultimately, pyroptotic death of the cell.