MWCNTs used in this study were sourced from Mitsui & Company (Mitsui-7, XNRI 1, lot #05072001K28, Tokyo, Japan). MWCNTs were imaged using transmission electron microscopy (TEM), as described in our previous publication (Lim et al. 2023). Stock samples prepared as described below were further diluted in double distilled H₂O to 1:300 for most of the samples. One drop of the diluted samples was placed onto a formvar-coated 200 mesh copper grid and allowed to dry overnight. The samples were imaged using a JEOL 1400 TEM (Tokyo, Japan) at 80 kV. For MWCNTs, length measurements were taken from the longest straight distance between two points. The width measurement was the distance perpendicular to the structural walls of the CNTs.

The MWCNTs have a log-normal length distribution with a mean length of 4.46 μm (95% confidence interval: 4.08–4.88 μm) and a normal width distribution with a mean of 58.5 nm (95% confidence interval: 56.0–61.0 nm). The MWCNTs are rod-like fibers (Figure 1(A)) with a surface area of 26 m²/g, containing low levels of trace elements (0.78% for all metals and 0.32% for iron) (Porter et al. 2010). The level of lipopolysaccharides in the MWCNTs was determined to be <0.01 mg/mL or <0.1 EU/mL using the Pierce LAL chromogenic endotoxin quantification kit (Thermo Fisher Scientific, Pittsburgh, PA).

For animal treatment, MWCNTs were prepared as described previously (Porter et al. 2010, Lim et al. 2020). The dispersion medium (DM) consisted of 0.9% saline, 0.6 mg/mL mouse serum albumin (Millipore Sigma, Burlington, MA, USA), and 0.01 mg/mL 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (Millipore Sigma). MWCNTs were suspended in DM following a two-step dispersion procedure involving sonication. DM was also used as the vehicle control for MWCNTs in animal experiments. For cell culture treatment, MWCNTs were suspended in Dulbecco’s Modified Eagle’s Medium (DMEM, Thermo Fisher Scientific) containing 1% fetal bovine serum (FBS, Thermo Fisher Scientific) using vortexing and sonication. The DMEM with 1% FBS served as the vehicle control for MWCNTs in cell culture.

Adult male C57BL/6J mice were obtained from the Jackson Laboratories (Bar Harbor, ME, USA) and housed in an accredited, specific pathogen-free, and environmentally controlled facility at the National Institute for Occupational Safety and Health. The mice were kept in polycarbonate ventilated cages with HEPA-filtered air and maintained on a 12-hour fluorescent lighting schedule. They were provided with Harlan Teklad Rodent Diet 7913 (Indianapolis, IN, USA) and tap water ad libitum. All animal experiments were conducted in accordance with humane care guidelines and were approved by the Institutional Animal Care and Use Committee.

Mice were treated with a single dose of MWCNTs (40 μg/mouse, adjusted to body weight) or vehicle control (dispersion medium, DM, 50 μL/mouse, adjusted to body weight) via oropharyngeal aspiration. Euthanasia was performed using an intraperitoneal injection of a sodium pentobarbital euthanasia solution (100–300 mg/kg body weight; Zoetis, Florham Park, NJ, USA), followed by exsanguination once the mouse became unresponsive to toe pinch.

Oropharyngeal aspiration is a noninvasive route of administration to deliver a specific dose of respirable materials, such as nanoparticles, into animal lungs to result in an even distribution of the materials in the lungs (Rao et al. 2003). MWCNTs at 40 μg in 50 μL DM administered by aspiration induced rapid and pronounced acute inflammation in mouse lungs, characterized by significant neutrophil infiltration and the secretion of proinflammatory cytokines (Porter et al. 2010, Dong et al. 2015, Lim et al. 2020). At nearly equivalent lung burdens of the MWCNTs, lung inflammation on day 1 post-exposure was similar between a single dose exposure by aspiration at a dose of 80 μg/mouse and by aerosol inhalation at 5 mg/m³, 5 hours/day, 12 days (Mercer et al. 2013). This inhalation paradigm results in a lung burden in mice equal to a predicted human lung burden on an equivalent alveolar surface area basis for a person performing light work at 7 μg/m³ for 13 years. These MWCNTs at 40 or 80 μg were shown to produce comparable pulmonary effects (Porter et al. 2010). Thus, a dose of the MWCNTs at 40 or 80 μg by aspiration in mice would produce an initial lung burden in humans comparable to that by inhalation.

There has been a paucity of data on human exposure to engineered nano materials. It has been estimated that this dose of 40 μg MWCNT exposure in mice approximates human deposition for a person performing light work for 11 years (Lim et al. 2020). This estimation assumes that a worker performs light work in an environment with MWCNT aerosol of 10.6 mg/m³ (Erdely et al. 2013) and has a minute ventilation of 20 L/min (Galer et al. 1992), a deposition fraction of 30% (Phalen 2008), and an alveolar epithelium surface area of 102 m² (Stone et al. 1992), which gives an estimated human exposure per year of 7.3 mg/m² alveolar epithelium.

After euthanasia, mouse lungs were collected from four mice each group for analysis. The right lung lobes were minced, frozen in liquid nitrogen, and stored at −80 °C for lipidomics and protein analyses. The left lungs were formalin-fixed, paraffin-embedded, and sectioned to 5 microns. Lung sections were stained using hematoxylin-eosin (H&E) staining or Masson’s trichrome staining (MTS) following standard procedures.

For immunohistochemistry, lung sections were deparaffinized with three washes of xylene (5 minutes each), rehydrated with two washes of 100% ethanol (10 minutes each), two washes of 95% ethanol (10 minutes each), and one wash of 70% ethanol (5 minutes), followed by two washes with deionized water (5 minutes each). To unmask antigens, sections were heated in a 10 mM sodium citrate buffer (pH 6.0) at sub-boiling temperature (10 minutes) in a microwave oven, followed by cooling on a benchtop for 30 minutes. The slides were then incubated with a BLOXALL endogenous peroxidase and alkaline phosphatase blocking solution (Vector Laboratories, Burlingame, CA, USA) for 10 minutes. Immunohistochemistry was performed using the ImmPRESS polymer detection system (Vector Laboratories) with ImmPACT NovaRED as a peroxidase substrate (red in color). Nuclei were counterstained with hematoxylin OS (blue violet in color, Vector Laboratories). Primary antibodies used included anti-Mac2 for macrophage detection (Cedarlane, Burlington, Ontario, Canada) and anti-neutrophil for neutrophil detection (Abcam, Cambridge, MA, USA). Images were captured using an Olympus Provis AX-70 microscope (Olympus, Center Valley, PA, USA). Positively stained cells were counted from five images taken at 40x magnification from one section, and lung sections from four mice in each group were examined.

J774A.1 murine monocyte/macrophage cell line was obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). Cells were cultured in DMEM with 10% FBS in a cell culture incubator at 37 °C with 5% CO₂. J774A.1 cells are known to avidly phagocytose nanoparticles and secrete significant amounts of proinflammatory cytokines, such as interleukin (IL)-1β and tumor necrosis factor (TNF)-α, in response to nanoparticle exposure, mirroring the behavior of mouse alveolar macrophages (Lim et al. 2020, Lim et al. 2023, Lim, Gu, and Ma 2025).

Cells were plated on a 24-well plate overnight, followed by treatment with vehicle control or MWCNTs at a concentration of 2 μg/mL for 24 hours. At this dose, MWCNTs induce prominent proinflammatory responses in J774A.1 macrophage cells (Lim et al. 2020, Lim et al. 2023). After treatment, the cells were stained with the LipidTox Red Phospholipidosis reagent (LTRP, Thermo Fisher Scientific) for an additional 24 hours under the same culture conditions before fixation with 3.7% formaldehyde. The nuclei were counterstained using 4′,6-diamidino-2-phenylindole (DAPI, Thermo Fisher Scientific). Images were captured using the Agilent BioTek Cytation 1 Cell Imaging Multi-Mode Reader (Model CYT1FAV, Agilent, San Diego, CA, USA), utilizing phase contrast for optical morphology, a DAPI filter for DAPI fluorescence (EX 360 nm, EM 460 nm), and a TRITC filter for LTRP fluorescence (EX 595 nm, EM 615 nm). Scale bars were set at 200 μm. Images were analyzed for relative fluorescence intensity (RFU) using the Gen5 software (Agilent). Data were presented as mean ± SD from four separate cell populations.

All chemicals used for UPLC–MS/MS analysis were of LCMS grade and obtained from Fisher Scientific (Pittsburgh, PA, USA). Methanol and acetonitrile were >99.9% pure, and acetic acid was 99.9% pure. Ultrapure water was prepared using a Millipore LC pak polisher cartridge with a Millipak Express Filter MPGP04001 (Millipore Sigma).

A total of 36 target standards and five deuterium-labeled internal standards were utilized for the calibration and quantification of LMs, as described previously (Ma et al. 2025). The standards were sourced from Cayman Chemicals (Ann Arbor, MI, USA) and had purities of ≥98%. Each sample was spiked with 100 pg/mL of the internal standards. Calibration standards for the target LMs were prepared across two ranges to accommodate varying levels of endogenous LMs in samples. The linear calibration range was from 0.66 pg/μL to 675 pg/μL for most LMs, while EPA, AA, and DHA had a range of 0.82 pg/μL to 13,500 pg/μL. The limit of detection (LOD) and limit of quantitation (LOQ) were calculated at 3× and 10× the signal-to-noise ratio, respectively.

The 36 target standards included various hydroxyeicosatetraenoic acids (HETEs), resolvins, lipoxins (LXs), prostaglandins (PGs), protectins, and thromboxane (TX). The specific compounds analyzed were: 5-hydroxyeicosapentaenoic acid (5-HEPE), 5-hydroxyeicosatetraenoic acid (5-HETE), 12-hydroxyeicosapentaenoic acid (12-HEPE), 12-hydroxyeicosatetraenoic acid (12-HETE), 14S-hydroxyhexaenoic acid (14S-HDHA), 15-hydroxyeicosapentaenoic acid (15-HEPE), 15-hydroxyeicosatetraenoic acid (15-HETE), 15R-lipoxin A4 (15R-LXA4), 17-hydroxyhexaenoic acid (17-HDHA), 17R-resolvin D1 (17R-RvD1), 17R-resolvin D4 (17R-RvD4), 18-hydroxyeicosapentaenoic acid (18-HEPE), arachidonic acid (AA), docosahexaenoic acid (DHA), eicosapentaenoic acid (EPA), leukotriene B4 (LTB4), lipoxin A4 (LXA4), lipoxin A5 (LXA5), lipoxin B4 (LXB4), maresin 1 (MaR1), maresin 2 (MaR2), prostaglandin D2 (PGD2), prostaglandin E2 (PGE2), prostaglandin F2α (PGF2α), prostaglandin H2 (PGH2), protectin D1 (PD1), protectin DX (PDX), resolvin D1 (RvD1), resolvin D2 (RvD2), resolvin D3 (RvD3), resolvin D4 (RvD4), resolvin D5 (RvD5), resolvin E1 (RvE1), resolvin E2 (RvE2), resolvin E4 (RvE4), and thromboxane B2 (TXB2).

The five deuterium-labeled internal standards included: 5S-hydroxy-6E,8Z,11Z,14Z-eicosatetraenoic-5,6,8, 9,11,12,14,15-d8 acid (5S-HETE-d8), 5S,12R-dihydroxy-6Z, 8E,10E,14Z-eicosatetraenoic-6,7,14,15-d4 acid (LTB4-d4), 5S,6R,15S-trihydroxyicosa-7E,9E,11Z,13E-tetraenoic-19,19,20,20,20-d5 acid (LXA4-d5), 9-oxo-11α,15S-dihydroxy-prosta-5Z,13E-dien-1-oic-3,3,4,4-d4 acid (PGE2-d4), and 7S,16R,17S-trihydroxy-4Z,8E,10Z,12E,14E,19Z-21,21′,22,22,2 2-d5 docosahexaenoic acid (RvD2-d5).

Mouse lung tissue samples (four mice each group) were homogenized in ice-cold methanol (375 μL, LCMS grade, Fisher Scientific) spiked with 25 μL of the internal standards mixture using a Dounce homogenizer. The homogenates were incubated at −20 °C for 30 minutes, followed by centrifugation at 751 × g for 10 minutes to obtain supernatants. The supernatants were then acidified to pH 3.5.

Lipids were extracted using solid-phase extraction with a C18 spin column (BioPureSPN Midi PROTO 300 C18, 17–170 mg, P/N HEM S18V) obtained from the Nest Group (Ipswich, MA, USA). Briefly, sample extracts were evaporated under a gentle stream of nitrogen using a TurboVap (Zymark, Hopkinton, USA) and reconstituted in 250 μL of a Milli-Q water:methanol mixture (80:20, v/v) with vortexing for 15 seconds. The reconstituted sample was added to a pre-conditioned C18 spin column, followed by pulse-spin at up to 1500 × g to remove the flow-through. The C18 column was washed with two volumes of 250 μL Milli-Q water:methanol (80:20, v/v). Lipids were then eluted by adding 250 μL of acetonitrile:Milli-Q water (80:20, v/v) on top of the C18 sorbent, followed by pulse-spin centrifugation past 1500 × g. The eluent was collected in a 1.5 mL centrifuge tube and evaporated to dryness using the TurboVap under a gentle nitrogen stream at 47 °C. The dried eluent was reconstituted in 45 μL of a mobile phase solvent mixture (A: B, 1:1, v/v) for immediate LC-MS/MS analysis, along with freshly prepared calibration internal standards.

A liquid chromatography-tandem quadrupole mass spectrometer (Waters ACQUITY Xevo TQ-XS UPLC-MS, Waters, Milford, MA, USA) was utilized for lipidomic analysis (Ma et al. 2025). The Acquity Premier UPLC system was operated with an ACQUITY Premier HSS T3 column (1.8 μm, 2.1 × 100 mm, Product number 186009471, Waters) using a 10 μL injection volume. The column and sample tray temperatures were maintained at 50 °C and 4 °C, respectively, throughout the chromatographic runs. The mobile phase gradient consisted of solvent A (95% Milli-Q ultrapure water, 5% methanol, and 0.01% acetic acid with 10 mM ammonium acetate) and solvent B (95% methanol, 5% Milli-Q ultrapure water, and 0.01% acetic acid with 10 mM ammonium acetate). The flow rate was set to 200 μL/min. The gradient began with 100% solvent A and 0% solvent B, gradually increasing solvent B to 70% over 17 minutes. Solvent B was then increased to 85% over three minutes (from 17 to 20 minutes) and to 100% at 25 minutes, where it was held for 2 minutes. For re-equilibration, solvent B was reduced to 0% from 27 to 28 minutes and held for an additional 2.5 minutes. The total running time for each cycle was 30.5 minutes.

The tandem quadrupole mass spectrometer (Waters Xevo TQ-XS) was operated with an electrospray ionization (ESI) source in negative-ionization mode for the analysis of target analytes. The instrument was set with an ESI capillary voltage of 1.5 kV, and the source and desolvation temperatures were maintained at 150 °C and 250 °C, respectively. High-purity nitrogen (Peak Scientific Instruments, Westford, MA, USA) was used with desolvation and cone gas flow rates of 600 and 150 L/hour, respectively. The collision cell entrance and exit potentials were both set to 30 V. The system operated in multiple reaction monitoring (MRM) mode, with 4 to 5 optimized transitions for each compound, except for 12-HEPE, which had 2 reliable transitions optimized and used. The cone voltage and collision energy were individually optimized for each compound as described for a previous study (Ma et al. 2025).

Lung tissues treated as indicated above were extracted in a lysis buffer (10 mM Tris, pH 7.4, 1% SDS) with a 1x proteinase inhibitor cocktail (Thermo Fisher Scientific) using a tissue homogenizer (Omni International, GA, USA) following sonication for 10 seconds. The supernatant was collected and combined for each treated group (3 samples/each treated group). Tissue proteins (20 μg each sample) were fractionated on a 8 or 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 hour at room temperature to block nonspecific binding, before incubation with primary antibodies. The primary antibodies used were mouse anti-CD68 (1:200, Novus Biologicals, Centennial, CO, USA), mouse anti-CD206 (1:250, Abcam), or mouse anti-β-actin (1:4,000, Millipore Sigma) antibodies. After incubation with horseradish peroxidase-conjugated goat anti-mouse or goat anti-rabbit IgG (both from Jackson ImmunoResearch laboratories, West Grove, PA; 1:5,000), immunoreactive bands were developed with Enhanced chemiluminescence (Thermo Fisher Scientific) on X-ray films. Film images were scanned using HP scanjet (Hewlett-Packard, Palo Alto, CA, USA) and were used to quantify each band with ImageJ software (NIH) with normalization to β-actin level.

Proinflammatory type 1 (IL-1β, TNF-α) or type 2 cytokines (IL-4, IL-13) were detected in lung tissue extracts collected from mouse lungs exposed to DM or MWCNTs for the indicated dose and days using ELISA (three mice each group). All ELISA kits were purchased from MyBioSource (San Diego, CA, USA) and measurement was performed by following the manufacturer’s protocol using a microplate reader equipped with the GEN5 software (Agilent).

Statistical comparisons between two means were performed using Welch’s t-test, which accounts for possible unequal variances. Comparisons of multiple means were perfomed using mixed-model one-way analyses of vairance (AVOVA), followed by posthoc comparisons between two means. For LC–MS/MS analysis, means ± SD obtained from the lungs of four animals in each group were calculated. For cell culture analysis, means ± SD from four separate cell populations were used. Statistical analysis was conducted using the BioRender software (Science Suite Inc. DBA BioRender, Toronto, Ontario, Canada). Significance was set at p < 0.05 and indicated as follows: ns (non-significance), *p < 0.05; **p < 0.01; ***p < 0.001; and ****p < 0.0001.

The Mitsui-7 MWCNTs administered at a single dose induced acute inflammation in mouse lungs that exhibited time-dependent polarization and resolution. Acute inflammation was most prominent at 1-day post-exposure, characterized by polarization of macrophages into M1 cells (Figure 1(B)) and production of high levels of proinflammatory cytokines TNF-α and IL-1β (Figures 1(C and D)), consistent with it being M1- and neutrophilic type 1 inflammation (Dong et al. 2015; Lim et al. 2020; Ma et al. 2025).

At 7-days post-exposure, lung inflammation was marked by polarization of macrophages into M2 cells (Figure 1(B)) and production of high levels of pro-resolving cytokines IL-4 and IL-13 (Figures 1(E and F)). Lung lesions at 7-days post-exposure were further examined by immunohistochemistry. In vehicle-exposed lungs, the alveolar structure was well-preserved, with a low but detectable level of neutrophils present in both alveolar sacs and septa (Figure 2(A)). MWCNTs significantly increased the cellularity in the lung, particularly in inflammatory foci where MWCNTs were visible as black deposits (Figure 2(B)). MWCNT exposure resulted in a doubling of neutrophil numbers, although this increase did not reach statistical significance (Figure 2(E)). In contrast, the number of macrophages in the alveolar sacs increased dramatically, with MWCNTs causing a 7.7-fold increase in macrophage numbers in both the alveolar sacs and interstitial space (Figures 2(C, D and F)). Many macrophages formed clusters and contained engulfed MWCNT particles.

These results indicate a phenotypic switch from type 1 to type 2 inflammation at 7-days post-exposure. In view of the critical role of type 2 responses in inflammation resolution and tissue repair, we posited that MWCNTs activate the production of SPMs to promote the resolution of inflammation at 7 days post-exposure where a transition from acute inflammation to fibrotic progression was suggested (Dong and Ma 2018b; Ma 2020).

Given that a significant portion of macrophages in the lungs of MWCNT-exposed mice contained engulfed MWCNTs, we investigated the impact of MWCNT phagocytosis on the accumulation of phospholipids, which can serve as precursors for ILMs and SPMs in macrophages. J774A.1 murine macrophages were exposed to either vehicle control or MWCNTs at a concentration of 2 μg/mL for 24 hours. Following treatment, cells were stained with the LipidTox Red Phospholipidosis reagent (LTRP) to visualize intracellular phospholipids.

As shown in Figure 3, vehicle-exposed cells exhibited low levels of LTRP fluorescence, indicating minimal phospholipid accumulation (Figures 3(A-C)). In contrast, MWCNT-treated cells displayed a substantial increase in LTRP fluorescence, which did not overlap with DAPI fluorescence, suggesting significant accumulation of phospholipids in the cytoplasm (Figures 3(D-F)). Quantitative analysis revealed increases in mean area per cell, mean intensity per cell, and mean intensity per DAPI-positive cell in MWCNT-exposed macrophages, all with statistical significance (Figures 3(G-I)). The DAPI fluorescence levels remained comparable between vehicle- and MWCNT-exposed cells (Figures 3(J-K)). These findings support the notion that MWCNTs stimulate the release of phospholipids from membranes in macrophages, potentially serving as precursors for the synthesis of lipid mediators.

To identify and quantify ILMs and SPMs, we performed targeted lipidomics analysis of 36 LMs using UPLC–MS/MS. The analysis revealed that MWCNTs significantly elevated the levels of several inflammatory lipid mediators at 7-days post-exposure. As illustrated in Figure 4(A), the major pathways involved in the production of ILMs from arachidonic acid via COX-dependent synthesis were activated. At this time point, AA levels were detected in both vehicle- and MWCNT-exposed lungs without a significant difference (Figure 4(B)). However, the levels of three major prostaglandins—PGH2, PGE2, and PGD2—were significantly increased in MWCNT-exposed lungs compared to vehicle controls (Figures 4(C-E)). Additionally, thromboxane B2 (TXB2), a major TX, was also significantly elevated by MWCNT exposure (Figure 4(F)). Prostaglandin F2α (PGF2α) was detected in vehicle-treated lungs and showed an increase in MWCNT-exposed lungs, although this increase did not reach statistical significance (Figure 4(G)).

The analysis of leukotrienes and lipoxins produced from arachidonic acid via the Alox-mediated pathway revealed significant findings. As shown in Figure 5(A), LTs and LXs are formed from AA via separate Alox pathways. MWCNTs significantly elevated the levels of LTB4, a major chemoattractant for leukocytes, compared to control, indicating its critical role in the recruitment of polymorphonuclear leukocytes during inflammation (Figure 5(B)). Additionally, 5-HETE, which is produced from the same precursor as LTB4, was detected in vehicle-exposed lungs and showed an increase in MWCNT-exposed lungs, although this increase did not reach a statistical significance (Figure 5(C)).

Conversely, LXB4, a major SPM that inhibits leukocyte recruitment, is produced via Alox15. LXB4 was detected at high levels in vehicle-exposed lungs and was significantly elevated by MWCNT exposure, showing a 9.3-fold increase (Figure 5(D)). Furthermore, 15-HETE, produced from the same precursor as LXB4, was also significantly elevated by MWCNTs (Figure 5(E)). In contrast, 12-HETE was detected in both vehicle- and MWCNT-exposed lungs, but the means did not differ significantly (Figure 5(F)).

The lipidomics analysis also assessed the production of SPMs from docosahexaenoic acid. The pathways and major metabolites of SPMs from DHA were shown in Figure 6(A). DHA levels were detected at high levels in mouse lungs, with MWCNT exposure leading to an increase that did not reach statistical significance (Figure 6(B)). Notably, three resolvins (RvD2, RvD3, and RvD5) were found at picogram levels (Figures 6(C-E)), with RvD5 showing a statistically significant increase in MWCNT-exposed lungs compared to controls. Additionally, protectin D1 (PD1), protectin DX (PDX), and 17-hydroxydocosahexaenoic acid (HDHA), all produced from the same precursor as the RvDs, were elevated by MWCNTs, with the increases for PDX and 17-HDHA reaching statistical significance (Figures 6(F-H)). Maresins (MaRs), including MaR1 and 14S-HDHA, which are produced from 14S-hydroperoxydocosahexaenoic acid (HPDHA) via the Alox12-dependent pathway, were also significantly increased by MWCNT exposure (Figures 6(I and J)).

The pathways and enzymes involved in the production of resolvin Es and lipoxins from eicosapentaenoic acid were examined (Figure 7(A)). EPA levels were detected at high levels in vehicle-exposed lungs, with MWCNT exposure leading to an increase that did not reach statistical significance (Figure 7(B)). However, both RvE2 and 18-hydroxyeicosapentaenoic acid (HEPE) were significantly increased by MWCNTs (Figures 7(C and D)). Additionally, RvE4 and 15-HEPE, produced by Alox15 and Alox5, were detected and showed increases in MWCNT-exposed lungs, with the increase for 15-HEPE reaching a statistical significance (Figures 7(E and F)). Lastly, both 5-HEPE and 12-HEPE were detected and increased by MWCNTs, although these increases did not reach a statistical significance (Figures 7(G and H)).