The MWCNTs used in this study were obtained from Mitsui & Company (Mitsui-7, XNRI 1, lot # - 0507 2001K28, Tokyo, Japan). Carbon black (CB) was purchased commercially (Printex 90, Degussa Engineered Carbons, L.P., Parsippany, NJ, USA) and was used as a non-fiber, carbon-based particle control for MWCNTs. Some properties of the MWCNTs and CB have been described previously (Lim et al. 2020, 2023; Porter et al. 2010).

For cell treatment, MWCNTs and CB were dispersed in Dulbecco’s Modified Eagle Medium (DMEM) with 1% fetal bovine serum (FBS) (both from Thermo Fisher Scientific, Waltham, MA, USA) at a concentration of 2 mg/mL by vortex and sonication as described previously (Hindman and Ma 2018; Lim et al. 2023). Stock solutions were further diluted with the culture media and sonicated immediately before use. DMEM with 1% FBS was used as a control media. The level of lipopolysaccharides (LPS) in the MWCNTs or CB was determined to be < 0.1 EU/mL (< 0.01 μg/mL) using the Pierce LAL chromogenic endotoxin quantification kit (Thermo Scientific, Pittsburgh, PA).

The J774A.1 murine macrophage cell line was purchased from the American Type Culture Collection (TIB-67, ATCC, Manassas, VA, USA). The cells were grown in DMEM with 10% FBS and 1X Antibiotic–Antimycotic (Thermo Fisher Scientific). MWCNTs at 2.5 or 10 μg/mL or CB at 2.5 or 10 μg/mL were used to treat cells followed by continued culture of the cells in an incubator with 5% CO₂ and 37°C till harvest. The DMEM plus 1% FBS media were used to establish a negative control and baseline levels. Some cells were kept as untreated to serve as a cell control in the absence of treatment. Interferon-γ (IFN-γ, Sigma-Aldrich, St. Louis, MO, USA) at 20 ng/mL plus LPS (Sigma-Aldrich) at 100 ng/mL was used as a positive control for induction of Alox5 mRNA and Alox5 protein. All cells were examined in duplicates and all treatment experiments were conducted 3 times. After treatment with a single dose of MWCNTs or CB, the cells were collected for further analysis.

Human whole peripheral blood with EDTA as an anticoagulant was purchased from STEMCELL Technologies (70508, Vancouver, BC, Canada). Human monocytes were isolated from the blood using the EasySep Magnet for column-free immunomagnetic separation and the EasySep direct human monocyte isolation kit according to the manufacturer’s protocol (STEMCELL Technologies).

Isolated human monocytes were plated in 6-well tissue culture plates (Fisher Scientific) at 2.5 × 10⁶ cells/well in the ImmunoCult-SF macrophage medium supplemented with M-CSF at 50 ng/mL (both from STEMCELL Technologies). The cells were cultured for 4 days to induce their differentiation into macrophages (human monocyte-derived macrophages or HMDMs). HMDMs were left untreated or were treated with the control medium, MWCNTs at 2.5 or 10 μg/mL, CB at 2.5 or 10 μg/mL, or IFN-γ plus LPS (positive control). After 24 h of incubation, the cells were collected, and total RNA was prepared for quantification of ALOX5 mRNA. Cell-free supernatants of the culture media were collected and used for quantification of LMs by ELISA.

Total RNA was isolated using the RNeasy Mini Kit (Qiagen, Germantown, MD, USA) per the manufacturer’s protocol. For reverse transcription, 1 μg of total RNA was reversely transcribed using a high-capacity RNA-to-cDNA kit (Thermo Fisher Scientific) at 37°C for 1 h. Real-time qPCR was performed for 35 cycles with the SYBR Green 1 PCR Master Mix on a 7500 real-time PCR machine (Thermo Fisher Scientific) using specific primers for the mouse Alox5 (Qiagen, PPM28755C) or β-actin (Qiagen, PPM02945B), and human ALOX5 (Qiagen, PPH02590G-200) or ACTB (β-actin) (Qiagen, PPH00073G-200). Reactions were run in triplicates for each sample and a dissociation curve was generated. Threshold cycles (Ct) for Alox5 amplification were normalized to the housekeeping gene β-actin (ΔCt). Each experimental sample was referred to its control (ΔΔCt). Relative expression values were calculated as 2^−ΔΔCt and expressed as fold change in comparison to untreated control.

Luciferase reporter plasmids carrying various lengths of the promoter region of the mouse Alox5 gene were generated by using PCR and restriction endonuclease digestion as follows. Briefly, the 2-kb promoter sequence upstream of the transcription start site (−2000 bp, 5′-upstream regulatory region or 5′-URR) of mouse Alox5 was amplified from the C57BL/6 J mouse genomic DNA by using PCR with the forward primer murine Alox5–2000-F and the reverse primer murine Alox5-URR-R (Table 1). The amplified DNA fragment was inserted into the pGL3-Basic vector (Promega, Madison, WI, USA) at the XhoI and HindIII restriction sites to generate the reporter plasmid pGL3–2,000-Luc. Other reporter constructs corresponding to specific deletions in the promoter region (−1500, −1000, −500, and −250) were generated from pGL3–2000-Luc by PCR using corresponding forward primers and Alox5-URR-R as a reverse primer (Table 1). All constructs were confirmed by DNA sequencing analysis. pCMV-sport-β-gal (Thermo Fisher Scientific) was used as a positive control for transfection efficiency in mammalian cells. Cells were co-transfected with 500 ng of a reporter plasmid and 100 ng of the β-gal plasmid using Lipofectamine^™ LTX Reagent with PLUS Reagent (Thermo Fisher Scientific) following the manufacturer’s instruction. Cells were harvested 24 h after transfection and the luciferase activity was measured by using a GloMax 20/20 luminometer (Promega) following the manufacturer’s instruction. Luciferase activities were normalized to the β-gal intensity of the same samples to correct for variations in transfection efficiency among samples.

ChIP assays were performed using a Magnify chromatin immunoprecipitation kit (Thermo Fisher Scientific, 49–2024) according to the manufacturer’s instruction. Cells were treated as described above. Intracellular proteins were crosslinked to DNA using 1% formaldehyde for 10 min at room temperature. Crosslinking was stopped by adding a 10X glycine solution. After two washes with cold 1X PBS, the cells were harvested in the RIPA buffer (150 mM NaCl, 1% IGEPAL CA-630, 0.5% deoxycholate, 0.1% SDS, and 50 mM Tris–HCl at pH 8). Nuclei were obtained by centrifugation, followed by treatment with a nucleolytic buffer to release chromosomal DNA. The chromosomal DNA was sonicated using the Covaris S2 ultrasonicator (Woburn, MA, USA) to shear the chromatin to an average size of ~ 200–500 bp. A fraction (10%) of the supernatants was reserved. For immunoprecipitation, the samples were centrifuged at 20,000×g at 4 °C for 5 min. One mg of the protein extract underwent pre-clearance with 30 μL of Protein A-Agarose/Salmon Sperm DNA (50% Slurry) for 30 min. After centrifugation to remove the agarose, the supernatant fractions were incubated with five μg of anti-PU.1 rabbit monoclonal antibody (Abcam, ab227835, Waltham, MA, USA) or the same amount of normal rabbit IgG for 4 h at 4°C, followed by an additional incubation with Dynabeads Protein A/G for 1 h at 4°C. The beads were washed with 1X PBS and the immunocomplexes were collected and eluted twice, each with 1% SDS/0.1 M NaHCO₃ for 10 min at 65°C. Crosslinks were reversed by adding the reverse crosslinking buffer provided with the kit followed by incubation at 65°C for 4 h. Immunoprecipitated DNAs were purified by treatment with RNase A and proteinase K. Purified DNA was evaluated and analyzed by conventional PCR using the following primer sets specific to the murine Alox5 gene promoter: mouse Alox5 promoter P1 (−229/−72; forward primer, 5′-GGCAAGAAGCCATAAAGATGG-3′ and reverse primer, 5′-CATCCTTCTCTCCTGTGGC-3′), P2 (−1024/−863; forward primer, 5′-GAGCAGGGTATTATCGAAGG-3′ and reverse primer, 5′-TGGGCAGCATTAATTTGTCTG-3′), P3 (−1245/−1109; forward primer, 5′-GTAGCAGTCAAAGAGAGGC-3′ and reverse primer, 5′-TCCACAGATCCTTGGCAC-3′)], and P4 (−1788/−1640; forward primer, 5′-TGCGGAATGTCTGACTGTG-3′ and reverse primer, 5′-GGACAAAGCAGACCTGGC-3′). Amplified DNA fragments were separated in 1.5% agarose gels.

After one day of plating, J774A.1 cells were transduced with mouse PU.1-specific or scrambled control shRNA lentiviral particles (5 × 10³ viral particles/μl) (Santa Cruz Biotechnology, Dallas, TX, USA) at a multiplicity of infection (m.o.i.) of 4 to ensure efficient infection. Cells transduced with control shRNA with scrambled sequence were included as negative controls. Some cells were left un-transduced to serve as a control. Cells were incubated overnight as recommended by the manufacturer. The media were replaced by a complete growth medium, and the cells were cultured for an additional day. A portion of the cells were lysed for isolation of the total RNA (for RT-qPCR) and protein (for immunoblotting). The remaining cells were treated with MWCNTs or CB and subsequently lysed to obtain RNA or protein as described above. The cell-free culture medium was collected for enzyme-linked immunosorbent assay (ELISA) assays.

J774A.1 cells were treated as indicated and were lysed in a lysis buffer (10 mM Tris, pH 7.4, 1% SDS) with a 1X proteinase inhibitor cocktail (Thermo Fisher Scientific) at the end of the experiment. Cell lysates were collected and sonicated for 10 s. The supernatant was collected, and the 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 transferred onto a nitrocellulose membrane. The membrane was incubated with 5% nonfat dry milk in Tris-buffered saline with 0.05% Tween 20 for 1 h at room temperature to block non-specific binding. Primary antibodies used were rabbit anti-Alox5 (1:1000, Abcam, ab169755), rabbit anti-PU.1 (1:1000, Abcam, ab227835) or mouse anti-β-actin (1:4000, Sigma-Aldrich, A5441) antibodies. After incubation with a second antibody, either horseradish peroxidase-conjugated goat anti-rabbit IgG (1:5000, Jackson ImmunoResearch laboratories, 111-035-144) or goat anti-mouse (1:5000, Jackson ImmunoResearch laboratories, 115-035-146, West Grove, PA, USA), immunoreactive bands were visualized with the enhanced chemiluminescence substrates (Thermo Fisher Scientific). Band signals were captured onto an X-ray film by exposure for 30 s and the film was developed using a film processor (Konica Minolta, Wayne, NJ, USA). Scanned images were used to quantify band intensities using the ImageJ software (1.52a, NIH, USA). Each band was normalized to β-actin and the relative amount of the PU.1 protein was calculated.

Proinflammatory LMs like LTB4 and PGE2 were detected in cell-free culture supernatants collected from cells treated with the control medium or MWCNTs after lentiviral particle transduction. Both ELISA kits were from MyBioSource (MBS268428 for LTB4, MBS266212 for PGE2, San Diego, CA, USA) and measurement was performed following the manufacturer’s protocol using a microplate reader equipped with GEN5 software (Agilent Technologies, Santa Clare, CA, USA).

Dependent measures were analyzed using mixed-model one-way analyses of variance (ANOVA), with each analysis incorporating experiment as a random factor. For each experiment, the 2 replicate samples were averaged to result in a single value for each treatment combination for each of the 3 independent experiments. 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 Fisher’s least significant difference test. All differences were considered significant at *p < 0.05; **p < 0.01; ***p < 0.001; compared to untreated control; #p < 0.01, compared to control-treated group. All analyses were carried out using JMP version 16 software (SAS Institute, Cary, NC, USA).

We have previously found that the pulmonary proinflammatory response to nanoparticles by macrophages is dependent on elevated production of proinflammatory lipid mediators in mouse lungs and in vitro, such as the Alox5-dependent production of the leukocyte chemotactic factor LTB4 (Lim et al. 2020, 2023). To elucidate the molecular underpinnings for the regulation of inflammation by MWCNTs, we focused on analyzing the mechanism by which MWCNTs induce Alox5 mRNA in J774A.1 macrophages.

Figure 1 reveals that the basal levels of Alox5 mRNA and protein were low in macrophages as determined by RT-qPCR and immunoblotting (Fig. 1A and B, untreated). The vehicle control medium did not affect the Alox5 mRNA and protein levels at 1-day post-exposure. As expected, the M1 inducer (IFN-γ + LPS, positive control) increased the Alox5 mRNA by 6.5-fold and protein by 4.1-fold at 1-day post-exposure compared to untreated control. Treatment with MWCNTs at a concentration of 2.5 μg/mL increased the expression of Alox5 mRNA by 2.6-fold and protein by 2.3-fold at 1-day post-exposure. At 10 μg/mL, MWCNTs increased the Alox5 mRNA expression by 4.7-fold and protein by 3.3-fold. CB is a paracrystalline carbon material that has the hexagonal carbon structure of CNTs but lacks the ordered, three-dimensional structure of CNTs and therefore, was used as an amorphous, carbonaceous material control for MWCNTs. Treatment with CB at a concentration of 2.5 or 10 μg/mL did not induce the Alox5 mRNA or protein significantly at 1-day post-exposure, compared to untreated control.

Induction and activation of Alox5 stimulate the production of LTB4 to mediate the acute proinflammatory response. We therefore measured the production of LTB4 (Fig. 1C). MWCNTs at 2.5 μg/mL or 10 μg/mL increased the LTB4 production to 36.98 pg/mL (6.2-fold increase over untreated control) and to 51.98 pg/mL (8.8-fold increase), respectively (Fig. 1C). M1 inducers increased LTB4 production to 83.37 pg/mL (14.0-fold increase). No apparent increase in the LTB4 level was observed in the culture media of the control medium- or CB-treated macrophages.

To examine if MWCNTs induce ALOX5 mRNA in human macrophages, human peripheral blood monocytes were differentiated into adherent macrophages following stimulation with M-CSF for 4 days. Differentiated HMDMs were left untreated or were treated with the control medium, MWCNTs at 2.5 or 10 μg/mL, CB at 2.5 or 10 μg/mL, or IFN-γ plus LPS. After 24 h of incubation, the ALOX5 mRNA levels were examined by RT-qPCR (Fig. 1D). Like the murine macrophages, untreated human macrophages had low levels of basal expression of ALOX5 mRNA. The vehicle control medium did not affect the ALOX5 mRNA expression. Treatment with MWCNTs increased the expression of ALOX5 mRNA by 3.2-fold at 2.5 μg/mL or 7.0-fold at 10 μg/mL at 1-day post-exposure, respectively. Treatment with CB at concentrations of 2.5 or 10 μg/mL for one day did not induce the ALOX5 mRNA, compared to untreated control. As expected, the M1 inducer (IFN-γ + LPS) increased the ALOX5 mRNA by 10.4-fold at 1-day post-exposure, compared to untreated control.

We also measured the production and induction of LTB4 in HMDMs by ELISA assay (Fig. 1E). MWCNTs at 2.5 μg/mL or 10 μg/mL increased the LTB4 production to 44.73 pg/mL (3.4-fold increase over untreated control) or to 118.17 pg/mL (8.9-fold increase), respectively. M1 inducers increased LTB4 production to 170.28 pg/mL (12.9-fold increase), whereas no apparent increase in the LTB4 level was observed in the culture media of the control medium- or CB-treated macrophages (Fig. 1E). These results indicate that the MWCNT-induced Alox5 expressions in murine macrophages and human macrophages were comparable to each other.

Induction of Alox5 mRNA by MWCNTs suggests a mechanism of transcriptional control of Alox5 mRNA expression by MWCNTs. We therefore examined the transcriptional activation of the Alox5 gene promoter. The 2 kilobase sequence upstream of the transcription start site of the murine Alox5 gene was obtained from the genomic DNA of a C57BL/6 J mouse by PCR and was used to construct a luciferase report plasmid (pGL3–2000-Luc) and its 5′-deletion mutants as shown in Fig. 2. The reporter plasmids were transfected to J774A.1 cells together with the pCMV-sport-β-gal plasmid (control for transfection efficiency). The Alox5 promoter activity was determined using the luciferase reporter assay expressed as RLU after normalization with β-gal intensity. The luciferase plasmid without the Alox5 promoter (pGL3-basic-Luc) showed a detectable, low-level luciferase activity. The pGL3–2000-Luc containing the 2 kb promoter sequence had a promoter activity of larger than 11.2-fold over the background (pGL3-basic-Luc). Deletion of the 500 bp at the 5′-end (pGL3–1500-Luc) reduced the activity of the 2 kb promoter by 34.2%. Further deletions at −1000, −500, and −250 bp steadily increased the activities, with the deletion at −250 bp (i.e., pGL3–250-Luc) producing the highest promoter activity of 15.9-fold over the background, indicating a strong proximal promoter for the basal expression of Alox5.

Next, we examined if MWCNTs induce transactivation of the Alox5 promoter. Un-transfected and untreated macrophages were used to provide a baseline activity. Macrophages transfected with the pGL3–250-Luc plus pCMV-sport-β-gal were used to test each treatment. Macrophages transfected with pGL3-basic-Luc plus pCMV-sport-β-gal were used as a control for each treatment. The cells were treated with the control medium, MWCNTs at 10 μg/mL, CB at 10 μg/mL, or the M1 inducer (IFN-γ at 20 ng/mL plus LPS at 100 ng/mL). Luciferase activities were determined at 1-day post-exposure (Fig. 3). Macrophages transfected with pGL3-basic-Luc had low levels of luciferase activities comparable to the baseline activity with or without the treatments. Transfection with the pGL3–250-Luc plasmid produced an activity of 18.5-fold over the pGL3-basic-Luc activity treated with the control medium. MWCNTs at 10 μg/mL further increased the activity to 4.2-fold over the pGL3–250-Luc treated with the control medium, whereas CB at 10 μg/mL had only a marginal increase in luciferase activity over the -pGL3–250-Luc with the control medium. As a positive control, treatment with IFN-γ + LPS produced the highest activity with a more than 5.5-fold increase in activity compared to the control medium.

Induced transactivation of the Alox5 promoter by MWCNTs confirmed the transcriptional activation of Alox5 expression by the nanotubes. We therefore investigated specific transcription factors that mediate the induction of the gene. Inspection of the 2 kb upstream regulatory region of mouse Alox5 revealed a TATAA-less promoter with several putative enhancer sequences for specific transcription factors. Among the sequences, four purine-rich (PU) box-like enhancers were identified at positions of −137, −937, −1166, and −1711. The PU box-like enhancer sequence resembles the binding sequence (xxAGGAAxx) for the transcription factor PU.1, an E26 transformation-specific (ETS)-domain-containing transcription factor required for myeloid and B-lymphoid cell development (Klemsz et al. 1990; McKercher et al. 1996; Pham et al. 2013; Scott et al. 1994). Hence, the sequences were named as putative PU.1 binding sites 1, 2, 3, and 4, respectively (Fig. 4). To examine if PU.1 indeed binds to these enhancers in the genomic Alox5 promoter, we performed a ChIP assay using an anti-mouse PU.1 antibody. The results revealed that PU.1 binds to each of the PU-Box-like enhancers in macrophages constitutively. Moreover, the binding of PU.1 at each enhancer was markedly increased by treatment with MWCNTs at the concentration of 10 μg/mL. The results identified PU.1 as a transcription factor that likely mediates the induced transactivation of the Alox5 promoter by MWCNTs by binding to the four PU-box-like enhancers within the 2 kb upstream regulatory region of the mouse Alox5 promoter.

To establish the role of PU.1 in the basal and induced expression of Alox5, the PU.1 gene was knocked down in macrophages using a PU.1-specific shRNA. Figure 5 showed that the transduction of PU.1-specific shRNA into macrophages reduced the Alox5 mRNA level by 76.5% compared to cells without the transduction and cells transduced with a control shRNA (Fig. 5A). Immunoblotting revealed that macrophages transduced with PU.1-specific shRNA had a 73.9% decrease of the Alox5 protein, compared to cells without the transduction and cells transduced with the control shRNA (Fig. 5B).

We then examined if the PU.1 knockdown affected the MWCNT-induced expression of Alox5. Macrophages transduced with PU.1 shRNA were treated with MWCNTs at 10 μg/mL or with the M1 inducer for 1 day. Induction of Alox5 mRNA was apparent in cells transduced with the control shRNA with a 4.9-fold increase by MWCNT at 10 μg/mL and a 6.3-fold increase by the M1 inducer over untreated control, respectively. However, no induction of Alox5 mRNA was observed in cells transduced with the PU.1 shRNA by either MWCNTs or the M1 inducer. CB at 10 μg/mL did not induce Alox5 mRNA in either control shRNA- or PU.1 shRNA-transduced cells (Fig. 5C).

To examine the functional impact of PU.1 knockdown, we analyzed the production of proinflammatory LMs LTB4 and PGE2 (Fig. 6). Transduction of the PU.1 shRNA effectively reduced LTB4 (2.3 pg/mL) production by 63.7% and PGE2 (6.1 pg/mL) production by 58.2%, compared to control shRNA-transduced cells (LTB4, 6.3 pg/mL and PGE2, 14.5 pg/mL, respectively) (Fig. 6A and B). Treatment of PU.1 shRNA-transduced cells with MWCNTs at 10 μg/mL for an additional 1 day slightly increased the production of LTB4 (8.0 pg/mL) and PGE2 (35.6 pg/mL). However, cells transduced with control shRNA showed a drastic increase of LTB4 (52.5 pg/mL) and PGE2 (222.4 pg/mL) production by 15.7-fold and 8.3-fold, respectively, upon treatment with MWCNTs at 10 μg/mL (Fig. 6A and B). These results indicate that PU.1 is a key transcription factor mediating the transactivation of the Alox5 promoter and induction of Alox5 mRNA and Alox5 protein by MWCNTs, which account for the increased production of proinflammatory LMs LTB4 and PGE2.