MyD88 mediates in vivo effector functions of alveolar macrophages in acute lung inflammatory responses to carbon nanotube exposure

Abstract

Carbon nanotubes (CNTs) are rapidly emerging as high-priority occupational toxicants. CNT powders contain fibrous particles that aerosolize readily in places of manufacture and handling, posing an inhalation risk for workers. Studies using animal models indicate that lung exposure to CNTs causes prolonged inflammatory responses and diffuse alveolar injury. The mechanisms governing CNT-induced lung inflammation are not fully understood but have been suggested to involve alveolar macrophages (AMs). In the current study, we sought to systematically assess the effector role of AMs in vivo in the induction of lung inflammatory responses to CNT exposures and investigate their cell type-specific mechanisms. Multi-wall CNTs characterized for various physicochemical attributes were used as the CNT type. Using an AM-specific depletion and repopulation approach in a mouse model, we unambiguously demonstrated that AMs are major effector cells necessary for the in vivo elaboration of CNT-induced lung inflammation. We further investigated in vitro AM responses and identified molecular targets which proved critical to pro-inflammatory responses in this model, namely MyD88 as well as MAPKs and Ca(2+)/CamKII. We further demonstrated that MyD88 inhibition in donor AMs abrogated their capacity to reconstitute CNT-induced inflammation when adoptively transferred into AM-depleted mice. Taken together, this is the first in vivo demonstration that AMs act as critical effector cell types in CNT-induced lung inflammation and that MyD88 is required for this in vivo effector function. AMs and their cell type-specific mechanisms may therefore represent potential targets for future therapeutic intervention of CNT-related lung injury.

Keywords: Carbon nanotubes; Macrophage depletion; Molecular toxicology; Nanotoxicology; Particle toxicology.

Fig. 1. TEM characterization of CNTs used in this study

TEM micrographs were used to characterize physical dimensions of CNTs following preparation in aqueous delivery medium. A: Low magnification, B: higher magnification, C: Frequency distribution of CNT diameter. Measured diameters yielded the following statistics (in nm): mean=10.7; SD=3.1; Q1=8.6; Q3=11.9; min=4.9; max=19.9.

Fig. 2. CNT exposures caused acute neutrophilic inflammation in the lung in a mouse model

NSA mice (n=4) received 4 mg/kg CNT or vehicle at t=0 and sampled over 72 h. BAL fluid was collected at time points shown and cells were counted by type (A). Vehicle controls (not shown) performed at 24, 48, and 72 h showed no signs of inflammation in neutrophilia or cytokines. Cytokines were measured using ELISA (B) and an average of 24, 48, and 72 h vehicle controls is shown (VH) for background reference.

Fig. 3. Depletion of AMs attenuated neutrophilic inflammation of the lung in a mouse model of CNT exposure

Neutrophils were quantified in BAL fluid collected at indicated time points following exposure to CNTs (4 mg/kg). Peak neutrophilia induced by CNT exposure (at 24 h) was reduced to 49%±17 in animals which received an AM-depleting dose of LC (LC+CNT) prior to exposure. n=4, *p<0.5.

Fig. 4. Depletion of AMs attenuated levels of CNT-induced cytokines in the lung in a mouse model

TNF and IL-6 were measured in BAL fluid collected 12 h following CNT exposure. Both cytokines were decreased in animals which received an AM-depleting dose of LC prior to exposure. n=4, *p<0.5

Fig. 5. Reintroduction of AMs rescued CNT-induced neutrophilic response in the lung of a mouse model of CNT exposure

Values are expressed here as percentage of baseline neutrophilic response (CNT). Macrophage depletion here (LC+CNT) attenuated the response to 44.6±14.7% (of CNT alone). Adoptive transfer of AMs into macrophage-depleted animals (LC+AM+CNT) resulted in partial rescue of CNT-induced neutrophils (75.6±5.6% of CNT alone). Reintroduction of AMs without subsequent CNT exposure (LC+AM) had no observed effect. n=4, *p<.05

Fig. 6. Effect of reintroduction of AMs on CNT-induced pro-inflammatory cytokines in the lung of a mouse model of CNT exposure

Adoptive transfer of AMs into macrophage-depleted animals (LC+AM+CNT) resulted in partial rescue of CNT-induced TNF (44.2±14.9% in LC+CNT to 79.4±5.5% in LC+AM+CNT) as measured in BAL fluid in a manner which mirrored the effect on neutrophilia. n=4.

Fig. 7. Cytokine induction in AMs (MH-S cells) following CNT exposure is partially dependent on the MAP kinases p38 and JNK

TNF production by AMs was significantly reduced (A) by pre-incubation with 20 μM of p38 inhibitor SB203580 or JNK inhibitor SP600125. Pre-incubation with 20 μM p38 inhibitor significantly reduced IL-1ß production (B); JNK inhibition had no effect. n=4, *p<0.05.

Fig. 8. CNT-induced TNFα and IL-1ß production in AMs (MH-S cells) is dependent on calcium- and MyD88- signaling

Pretreatment with drugs chlorpromazine (CPZ, 20 μM) or inhibitory peptide pepinh-MYD (MYD, 100 μM) were used to inhibit calcium signaling and MyD88 signaling, respectively. Both drugs abrogated production of TNF (A) and IL-1ß (B). n=4, p<0.05.

Fig. 9. Inhibition of MyD88 in primary AMs transferred into AM-depleted mice led to no rescue of CNT-induced neutrophilia

Values are expressed here as percentage of baseline neutrophilic response (CNT). Adoptive transfer of AMs (10⁵/animal) treated with MyD88 inhibitory peptide (2 h pre-incubation) into macrophage-depleted animals (LC+AM(inh)+CNT) resulted in no rescue of CNT-induced neutrophils (43.1±4% of CNT alone), while normal rescue (76.4±6.5% of CNT alone) was observed after reintroduction of AMs incubated without inhibitory peptide (LC+AM(na)+CNT. n=4, *p<.05

Fig. 10. Inhibition of MyD88 in primary AMs transferred into AM-depleted mice did not rescue CNT-induced proinflammatory cytokines

Adoptive transfer of AMs incubated with MyD88 inhibitory peptide into macrophage-depleted animals (LC+AM (inh)+CNT) resulted in no rescue of CNT-induced TNFα and IL-6, while robust rescue was observed in both endpoints after reintroduction of AMs incubated without inhibitory peptide (LC+AM(na)+CNT). n=4, *p<.05

Fig. 11. Proposed role of MyD88 and MAPKs in the overall mechanism of action in AM responses to CNTs

MyD88 may link CNT exposures to inflammatory responses though IL-1-dependent feedback through IL-1R or activation of TLRs, with p38 and JNK MAPKs modulating downstream AP-1 activity. Known interactions in these pathways are shown by solid arrows. Proposed mechanisms of action by CNTs (dotted arrows) are based on the current and related studies [Girtsman et al, 2014, Hussain et al, 2014, Palomäki et al, 2011].