Pulmonary inflammation induced by low-dose particulate matter exposure in mice

Abstract

Air pollution is a ubiquitous problem and comprises gaseous and particulate matter (PM). Epidemiological studies have clearly shown that exposure to PM is associated with impaired lung function and the development of lung diseases, such as chronic obstructive pulmonary disease and asthma. To understand the mechanisms involved, animal models are often used. However, the majority of such models represent high levels of exposure and are not representative of the exposure levels in less polluted countries, such as Australia. Therefore, in this study, we aimed to determine whether low dose PM₁₀ exposure has any detrimental effect on the lungs. Mice were intranasally exposed to saline or traffic-related PM₁₀ (1μg or 5μg/day) for 3 wk. Bronchoalveolar lavage (BAL) and lung tissue were analyzed. PM₁₀ at 1 μg did not significantly affect inflammatory and mitochondrial markers. At 5 μg, PM₁₀ exposure increased lymphocytes and macrophages in BAL fluid. Increased NACHT, LRR and PYD domains-containing protein 3 (NLRP3) and IL-1β production occurred following PM₁₀ exposure. PM₁₀ (5 μg) exposure reduced mitochondrial antioxidant manganese superoxide (antioxidant defense system) and mitochondrial fusion marker (OPA-1), while it increased fission marker (Drp-1). Autophagy marker light-chain 3 microtubule-associated protein (LC3)-II and phosphorylated-AMPK were reduced, and apoptosis marker (caspase 3) was increased. No significant change of remodeling markers was observed. In conclusion, a subchronic low-level exposure to PM can have an adverse effect on lung health, which should be taken into consideration for the planning of roads and residential buildings.

INTRODUCTION

The World Health Organization (WHO) air quality model demonstrates that ambient air pollution annually causes 4.2 million deaths, and 91% of the world’s population lives in places where air quality exceeds the limits of WHO guidelines. Air pollution causes 1.8 million deaths from lung diseases (1). Forty three percent of chronic obstructive pulmonary diseases (COPD) and 29% of lung cancer deaths are attributable to air pollution (2). PM is the sum of all particles suspended in the air, which includes both organic and inorganic particles, such as dust, pollens, and vehicle emissions. Respirable PM is thought to be the most detrimental to human health. PM sized equal or below 10 μm (PM₁₀) is capable of entering the lungs, while PM sized equal or below 2.5 μm (PM_2.5) can reach the distal lung segments including alveoli (17).

In adults, every 5 μg/m³ increment of PM exposure is associated with a 39% to 56% increased risk of developing COPD (13). In developed countries such as the UK, traffic-related air pollution (TRAP) accounts for 13% of total PM (4). In Sydney Australia, the levels of TRAP are among the lowest in the world, accounting for 14% of total PM (5), which is often assumed to be safe. However, a study on 65,000 children in Canada found that children exposed to TRAP, even in urban areas with low levels of pollution, had a 25% increased risk of developing asthma by the age of 5 yr.

PM is a strong oxidant, with its oxidant capacity regulated by antioxidants, such as manganese superoxide dismutase (16). However, in humans, even short-term exposure of PM₁₀ increased circulating levels of IL-1β, IL-6, and TNF-α (28). PM₁₀ contains ~10¹⁶ free radicals/g, which can increase oxidative stress in human macrophages and lung epithelial cells (8, 29). ROS can induce inflammatory responses via the activation of the nucleotide-binding domain and leucine-rich repeat protein 3(NLRP3) inflammasome, which, in turn, cleaves pro-IL-1β into IL-1β. Interestingly, Hirota et al. (14) have shown that PM activates the NLRP3 inflammasome, resulting in increased IL-1β in bronchial epithelial cells.

Mitochondria can be damaged by both oxidative stress and the activation of NLRP3 inflammasome, resulting in reduced capacity to produce ATP. Mitophagy is a quality control process where fission removes damaged mitochondria fragments, and fusion merges healthy mitochondrial fragments to generate new mitochondria (7), which has been shown to ameliorate inflammatory disorders (23). The impact in low-level PM exposure on mitophagy markers has not been reported.

TRAP contains both gaseous and PM components. While the gaseous components are equally toxic as PM, gases dissipate quicker in air than the PMs, which can remain airborne for long periods of time. However, most PM/TRAP exposure models used very high PM exposure regimens [e.g., 50 to 200 μg (11, 21)], which are not relevant to the PM/TRAP levels in countries with low levels of air pollution. We hypothesized that exposure to low levels of PM would be detrimental for lung health. Our objective was to establish an environmentally relevant model of TRAP-related PM exposure and to characterize pulmonary changes, including inflammasome activation (NLRP3 and IL-1β), IL-6 production, mitochondrial fission, and fusion markers [Optic atrophy (Opa)-1 and dynamin-related protein (Drp)-1], autophagy markers, and fibrotic markers [fibronectin, collagen III, and transforming growth factor β1 (TGFβ1)].

MATERIALS AND METHODS

PM collection.

Twenty-four-hour integrated PM₁₀ were collected through a 47-mm Teflon (Pall Life Sciences, Ann Arbor, MI) and prefired (800°C, 3 h) 47-mm quartz-fiber filters (Whatman, Clifton, NJ) from a busy roadside in Hong Kong (114,000 vehicles per day) with URG PM samplers (URG-2000-30EH) in the summer (June 24th to July 11th, 2017), with a flow rate of 8 l/min at each channel. Filter preparation (e.g., equilibrated for 24 h at 25°C and relative humidity of 40% before and after sampling) and gravimetric analysis were conducted in a high-efficiency particulate absorption clean room (ISO 14644, Class 7) at The Hong Kong Polytechnic University. All filters were stored at −20°C and in the dark before the analysis. PM was extracted in 90% ethanol with 5 min of sonication, followed by freeze drying overnight.

PM analysis.

Energy-dispersive X-ray fluorescence spectrometry (PANalytical Epsilon 5) was used to determine concentrations of Al, Si, Ca, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Ba, and Pb. Each sample was analyzed for 30 min. Thin-film standards were used for calibration (MicroMatter, Arlington, TX) (34). All reported chemical concentrations were corrected for field blanks, and duplicated samples were analyzed for quality assurance.

Ion chromatography (IC) for water-soluble inorganic ion analysis. One quarter of the filter was extracted with 10 mL of distilled deionized water, and the extract underwent IC (Dionex DX-600) analysis (IonPac CS12A and AS14A columns). Six species were analyzed as previously described (36). Analysis of organic carbon and elemental carbon was performed by the thermal optical reflectance technique on a thermal/optical carbon analyzer (DRI model 2001, Atmoslytic, Calabasas, CA), as described in Pathak et al. (22).

In vivo PM exposure.

Animal experiments were approved by the Animal Care and Ethics Committee at the University of Technology Sydney (ACEC no. ETH16-0886). Male Balb/c mice (6 wk, Animal Resources Centre, Perth, Australia) were housed at 20 ± 2°C and maintained on a 12:12-h light-dark cycle (lights on at 0600) with ad libitum access to standard laboratory chow and water. After the acclimatization period, mice were assigned to three groups (n = 10), which were exposed to either particulate matter with 1 µg PM₁₀ or 5 μg PM₁₀ or saline as control (SHAM). In urban Sydney, the average PM₁₀ levels are 17 μg/m³, equating to a daily human exposure of 181 μg (3). On the basis of the breathing volumes, mice should be exposed to around 5 μg/day to reflect air pollution levels in Sydney. Mice were exposed intranasally by instillation of 40 µl of saline or saline resuspended PM₁₀ daily for 3 wk.

At the end point, the animals were euthanized via cardiac puncture after deep anesthesia (3% isoflurane). Lungs were perfused with PBS to obtain bronchoalveolar lavage (BAL) fluid. Lungs were then harvested, snap frozen, and stored at −80°C for protein analysis. Anthropometry measurements were done following dissection and measurement on a microbalance.

BAL analysis.

The BAL cells evaluated by Diff-Quik staining (Polyscience, Taipei, Taiwan). Differential cell counts were performed for macrophages, lymphocytes, eosinophils, and neutrophils.

Western blot analysis.

Lung tissue homogenates (20 μg) were analyzed using standard techniques, as described previously (9). Antibodies were purchased from Cell Signaling Technology (Danvers, MA): IL-1β and IL-6 (1:1,000); caspase 3, p-Akt, Akt, AMPK, light chain 3 microtuble-associated protein (LC) 3A/B-I/II (1:2,000); from Novus Biotechnology, Centennial, CO: Drp-1, Opa-1 (1:2,000) and collagen III (1:1,000); from Millipore, Burlington, MA: MnSOD (1:2,000); from Sigma-Aldrich, USA fibronectin (St. Louis, MO; 1:2,000); and R&D Systems: TGF-β1 (1:500).

Mitochondrial DNA copy number.

mtDNA was measured using quantitative PCR on DNA, as we have previously described (25, 26).

Statistical methods.

The data conformed to the normal distribution and differences between groups were analyzed using one-way ANOVA followed by a Bonferroni post hoc tests. P < 0.05 was considered significant.

RESULTS

PM characterization.

The main components of the PM were organic carbons. Sulfate, elemental carbon, chloride, and nitrate were the other components in abundance in the PM sample. Traces of other substances such as titanium, manganese, lead, chromium, and nickel were also detected, see Table 1.

Table 1.

Chemical characteristics of PM₁₀

	Component Quantity, μg/m3
PM10 mass	22.61 ± 1.26
Organic carbon	4.19 ± 0.20
Sulfate	4.00 ± 0.34
Elemental carbon	3.26 ± 0.17
Chloride	2.52 ± 0.41
Nitrate	1.92 ± 0.13
Iron	0.85 ± 0.04
Calcium	0.43 ± 0.03
Silicon	0.35 ± 0.02
Aluminium	0.17 ± 0.02
Ammonium	0.16 ± 0.03
Barium	0.08 ± 0.003
Zinc	0.08 ± 0.01
Copper	0.04 ± 0.03
Titanium	0.02 ± 0.004
Manganese	0.02 ± 0.002
Lead	0.02 ± 0.002
Vanadium	0.01 ± 0.002
Chromium	0.01 ± 0.001
Nickel	0.01 ± 0.001

Values are means ± SE. Data show different components inside the traffic-related air pollutants (n = 10). PM₁₀, particulate matter of a size equal or less than 10 μm.

Anthropometry markers.

Weight gain was used as a generic indicator of health status. As shown in Table 2, body weight was not affected by PM exposure (Table 2). However, PM₁₀ (5 μg)-exposed animals had significantly more retroperitoneal fat mass compared with the Sham group (P < 0.05). There were no significant changes in liver or muscle weights.

Table 2.

Effects of PM₁₀ exposure on anthropometry markers

	Sham	PM10 (1 μg)	PM10 (5 μg)
Body weight	22.39 ± 0.31	22.26 ± 0.36	22.13 ± 0.37
Liver, g	1.26 ± 0.045	1.21 ± 0.037	1.15 ± 0.037
Liver %	5.62 ± 0.0015	5.47 ± 0.0011	5.21 ± 0.0015
Muscle, g	0.073 ± 0.0024	0.075 ± 0.0023	0.072 ± 0.0032
Muscle %	0.33 ± 0.00013	0.34 ± 0.00011	0.33 ± 0.00019
Retroperitoneal fat weight, g	0.077 ± 0.0037	0.109 ± 0.014	0.12 ± 0.012*
Retroperitoneal fat %	0.34 ± 0.00016	0.50 ± 0.00064	0.55 ± 0.00052*
Glucose, mM	9.13 ± 1.14	9.6 ± 1.07	9.27 ± 1.1

Values are means ± SE; n = 10. Data were analyzed by one-way ANOVA followed by Bonferroni post hoc test. PM₁₀, particulate matter.

^*P < 0.05, compared with Sham.

Bronchoalveolar cell count.

PM₁₀ (5 μg) exposure increased leukocyte counts in bronchoalveolar lavage (BAL) fluid [P < 0.01, PM₁₀ (5 μg) vs. Sham; Fig. 1A], as well as lymphocytes and macrophages (both P < 0.01 vs. Sham; Fig. 1, A and B). There were no neutrophils or eosinophils observed.

Fig. 1.

A–C: leukocyte counts in bronchoalveolar lavage fluid. D–I: lung protein levels of NLRP3 (NACHT, LRR and PYD domains-containing protein) (D), IL-1β (E), IL-6 (F), fibronectin (G), TGF-β₁ (H), and collagen III (I) in Sham, particulate matter (PM)₁₀ (1 μg), and PM₁₀ (5 μg) groups. Results are expressed as means ± SE; n = 8–10 (one-way ANOVA followed by Bonferroni post hoc test). *P < 0.05, **P < 0.01, compared with Sham. #P < 0.05, ##P < 0.01, compared with PM₁₀ (1 μg).

Lung inflammation.

NLRP3 and IL-1β were increased in the PM₁₀ (5 μg) group compared with the Sham group (P < 0.05, Fig. 1, D and E), but not IL-6 (Fig. 1F).

Markers of matrix remodeling.

Protein levels of fibronectin, TGF-β₁ and collagen III were not changed in any of the PM groups compared with the Sham group (Fig. 1, G–I).

Mitochondrial antioxidant, mitophagy markers and mitochondrial DNA copy number.

PM₁₀ (5 μg) exposure significantly increased mitochondrial fission protein Drp-1 (P < 0.05, PM₁₀ (5 μg) vs. Sham; Fig. 2A) and reduced mitochondrial fusion protein OPA-1 and the antioxidant MnSOD levels [both P < 0.05, PM₁₀ (5 μg) vs. Sham, Fig. 2, B and C]. Mitochondrial DNA copy number was not changed between Sham and PM₁₀ (5 μg) (Fig. 2D).

Fig. 2.

Lung mitochondrial protein levels of Drp-1 (A), Opa-1(B), MnSOD (C), mitochondrial DNA copy number (D), lung protein levels of LC3A/B-II (E), LC3A/B-II to I ratio (F), caspase-3 (G), p-Akt (H), Akt (I), p-Akt/Akt ratio (J), p-AMPK (K), AMPK (L) and p-AMPK to AMPK ratio (M) in Sham, PM₁₀ (1 μg) and PM₁₀ (5 μg) groups. Results are expressed as means ± SE; n = 5–8 (one-way ANOVA with Bonferroni tests). *P < 0.05 compared with Sham. **P < 0.01, compared with Sham. #P < 0.05, compared with PM₁₀ (1 μg). Akt, protein kinase 3; AMPK, 5′-adenosine monophosphate-activated protein kinase; Drp-1, dynamin-related protein 1; LC3A/B, light-chain 3 microtubule-associated protein A/B; MnSOD, manganese superoxide dismutase; Opa-1, optic atrophy 1; PM, particulate matter.

Autophagy and apoptosis.

Autophagy marker LC3A/B-II, LC3A/B-II to I ratio were reduced in PM₁₀ (5 μg) compared with Sham (P < 0.05, Fig. 2, E and F). Apoptotic marker caspase 3 was increased in the PM₁₀ (5 μg) group compared with the Sham group (P < 0.05, Fig. 2G). The upstream marker of autophagy, p-AMPK, and p-AMPK to AMPK ratio were reduced by the exposure to PM₁₀ (5 μg) compared with the Sham exposure (P < 0.05 vs. Sham, Fig. 2, K and M). Akt and AMPK protein levels were increased in the PM₁₀ (5 μg) group compared with the Sham group (P < 0.05 vs. Sham, Fig. 2, I and L), but there were no changes in p-Akt protein levels and p-Akt to Akt ratio by PM₁₀ exposure (Fig. 2, I and J).

DISCUSSION

We found that the exposure to low levels of traffic-related PM₁₀ induced marked pulmonary activation of the NLRP3 inflammasome and inflammation, reduced mitochondrial antioxidants, and impaired mitophagy capacity.

PM₁₀ exposure for 3 wk did not affect the overall well-being of the mice reflected by body weight, suggesting low toxicity. However, fat mass was increased following the exposure to 5 μg of PM₁₀, consistent with other human and mouse studies (27, 31).

We found increased lymphocytes and macrophages, which has also been observed with high-dose PM exposure (8). However, PM₁₀ (5 µg) did not induce eosinophilic or neutrophilic inflammation. Increased IL-1β was accompanied by NLRP3 inflammasome activation as expected. Zheng et al. (37) also found that 3 wk of exposure to 50 μg of PM_2.5 daily increased IL-1β and TGF-β₁ levels in BAL. Inflammasome activation has been observed in asthma and COPD, as well as during pulmonary inflammation (10, 18, 35), suggesting that continuous exposure to even a low level of PM may increase the susceptibility to these conditions.

Mitochondrial dysfunction is associated with various pulmonary diseases. COPD patients have mitochondrial fragmentation through an increase in Drp-1. In vitro prolonged cigarette smoke exposure increased mitochondrial fission (6, 15). Damaged mitochondria increase oxidative stress, which can consume the antioxidative MnSOD. Our study shows that 5 μg of PM reduced MnSOD, suggesting reduced antioxidant capacity. Mitochondrial DNA copy number was unaffected, suggesting mitochondrial biogenesis was not changed by PM in this model. The reduction in LC3A/B-II protein in the PM₁₀ (5 µg) group indicates that there was reduced capacity of autophagy, which can increase apoptosis. This was confirmed by the increased protein levels of caspase 3 in our study.

Activated AMPK was reduced by PM₁₀ exposure. AMPK is a stress sensor that is crucial for maintaining intracellular homeostasis during oxidative stress and importantly, AMPK-deficient mice have increased progression of COPD (19). AMPK typically suppresses Akt, but we found no change in Akt levels, suggesting dysregulation of AMPK/Akt signaling. In our study, we found PM reduced AMPK activation with reduced autophagy; however, in vitro studies have found PM increases AMPK and autophagy. We postulate that such differences are related to the 10–20 times higher dose of PM used in vitro, which induce cell death, in addition to activating AMPK and autophagy (20, 30, 32). The in vitro response is consistent with the notion that autophagy generally acts to keep cells alive and is upregulated in response to stress (for a review, see Ref. 12). Differences may also occur as a result of PM processing for in vitro studies, in which steam sterilization to remove LPS may also remove other PM components. Interestingly LPS inhibits AMPK activation (33).

Inflammasome activation by asbestos or crystalline silica is strongly associated with the development of lung fibrosis (24). However, in this study, exposure to a low level of PM did not induce fibrosis. The negative findings are most likely attributable to the low-PM dose and the short duration of this study.

This study has several limitations. PM₁₀ composition varies by generation source, and as such, future studies need to compare different types of PM. We did not assess endotoxin levels in PM, which are likely to influence the proinflammatory capacity of the PM. The lung tissues were not fixed to assess any histological changes or mitochondrial morphology, which needs to be addressed in future studies.

In conclusion, this study shows that the exposure to low levels of roadside PM has detrimental effects on lung health. As such, people living alongside major traffic corridors need to be aware of the potential adverse effects on their respiratory health. Our results also have implications for government agencies responsible for urban planning.

GRANTS

This study was funded by SPHERE, RCG of the Hong Kong Special Administrative Region (CRF/C5004-15E), the Strategic Focus Area scheme of The Research Institute for Sustainable Urban Development at The Hong Kong Polytechnic University (1-BBW9). Y. Chan, H. Chen, and B. Oliver are supported by fellowships (iCare, 81750110554 from the National Natural Science Foundation of China and APP1110368 from the National Health and Medical Research Council Australia). B. Wang is supported by the China Scholarship Council.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Y.L.C., H.C., K.F.H., and B.G.G.O. conceived and designed research; Y.L.C., B.W., and H.C. performed experiments; Y.L.C., H.C., K.F.H., J.C., G.H., and B.G.G.O. analyzed data; Y.L.C., S.S., and B.G.G.O. interpreted results of experiments; Y.L.C. and B.W. prepared figures; Y.L.C. and B.G.G.O. drafted manuscript; Y.L.C., B.W., H.C., B.J., C.H., P.S.T., S.S., and B.G.G.O. edited and revised manuscript; Y.L.C., H.C., B.J., P.S.T., and B.G.G.O. approved final version of manuscript.