Acute cobalt-induced lung injury and the role of hypoxia-inducible factor 1alpha in modulating inflammation

Abstract

Air pollution is a critical factor in the development and exacerbation of pulmonary diseases. Ozone, automobile exhaust, cigarette smoke, and metallic dust are among the potentially harmful pollution components that are linked to disease progression. Transition metals, such as cobalt, have been identified at significant levels in air pollution. Cobalt exerts numerous biological effects, including mimicking hypoxia. Similar to hypoxia, cobalt exposure results in the stabilization of hypoxia-inducible factors (HIFs), a family of proteins that regulate the cellular response to oxygen deficit. HIFs also play an important role in innate immunity and inflammatory processes. To characterize the role of HIF1alpha, the most ubiquitously expressed HIF, in the early events during cobalt-induced lung inflammation, an inducible lung-specific HIF1alpha deletion model was employed. Control mice showed classical signs of metal-induced injury following cobalt exposure, including neutrophilic infiltration and induction of Th1 cytokines. In contrast, HIF1alpha-deficient mice exhibited pronounced eosinophil counts in bronchoalveolar lavage fluid and lung tissue complemented with Th2 cytokine induction. The timing of these results suggests that the loss of epithelial-derived HIF1alpha alters the lung's innate immune response and biases the tissue toward a Th2-mediated inflammation.

FIG. 1.

Experimental design. HIF1αΔ/Δ mice were generated through postnatal doxycycline treatment paradigm (doxycycline given from PN4 to PN42). Control (n = 36) and HIF1αΔ/Δ (n = 36) male mice randomly assigned to three different treatment groups (24, 48, and 120 h). For each time point, mice were challenged with saline (n = 6) or cobalt chloride (60 μg, n = 6) via oropharyngeal aspiration. Animals were euthanized 24 h after their first dose (24-h treatment group), second dose (48-h treatment group), or fifth dose (120-h treatment group).

FIG. 2.

BALF protein and total cell counts from control and HIF1αΔ/Δ mice. Total protein concentrations (A) and cell counts (B) were assessed from BALF of saline-treated (white bars) and cobalt (hatched bars)-treated control mice and saline-treated (checkered bars) and cobalt (black bars)-treated HIF1αΔ/Δ mice as described in Materials and Methods section. n > 5 mice per group. Data are expressed as mean ± SE. Significance is noted where applicable.

FIG. 3.

Effect of cobalt treatment on inflammatory cells recovered in BALF. Differential cell counts were performed for lymphocytes (A), macrophages (B), neutrophils (C), and eosinophils (D) from BALF of saline-treated (white bars) and cobalt (hatched bars)-treated control mice and saline-treated (checkered bars) and cobalt (black bars)-treated HIF1αΔ/Δ mice. n > 5 mice per group. Data are expressed as mean ± SE. Significance is noted where applicable.

FIG. 4.

Histopathology staining of saline- and cobalt-treated control and HIF1αΔ/Δ mice. Light photomicrographs of hematoxylin and eosin–stained lung sections taken from control (A, C, E, G) and HIF1αΔ/Δ mice (B, D, F, H) that were dosed by oropharyngeal aspiration with cobalt in saline (C, D, E, F, G, H) or saline alone (saline-vehicle controls; A and B). Mice were exposed once a day for 1 (C, D), 2 (E, F), or 5 (A, B, G, H) days. Small insert in the right lower corner of each figure is a photomicrograph at low magnification of the transverse left lung lobe (proximal aspect) taken at generation five of the intrapulmonary axial airway (AA). Solid arrows identify areas of cobalt-induced mixed inflammatory cell infiltration. In (G and H), focal areas of epithelial hyperplasia lining terminal bronchioles and/or alveolar ducts (stippled arrows) are circumscribed by areas of cellular inflammation and alveolar interstitial fibrosis (asterisks). Tb, terminal bronchiole; ad, alveolar duct; and a, alveolus.

FIG. 5.

MBP immunohistochemistry in lungs from saline- and cobalt-treated control and HIF1αΔ/Δ mice. Light photomicrographs of lung sections immunohistochemically stained for eosinophil-specific MBP (inflammatory cells with cytoplasmic red staining). Sections were taken from the left lung lobe of control mice (A, C, E, G) and HIF1αΔ/Δ mice (B, D, F, H) that were dosed by oropharyngeal aspiration with cobalt in saline (C, D, E, F, G, H) or saline alone (saline-vehicle controls; A and B). Mice were instilled once a day for 1 (C, D), 2 (E, F), or 5 (A, B, G, H) days. All tissue sections were counter stained with hematoxylin. Tb, terminal bronchiole; ad, alveolar duct; and a, alveolus. Large peribronchiolar and alveolar infiltrates of immunohistochemically positive eosinophils are present in cobalt-instilled lungs after 1, 2, and 5 days of cobalt instillation (arrows). Only a few eosinophils (individual cells or small aggregates) are present in the mixed inflammatory cell infiltrates of control mice similarly instilled with cobalt (arrows).

FIG. 6.

Neutrophil (PMN) immunohistochemistry in lungs from control and HIF1αΔ/Δ mice. Light photomicrographs of lung sections immunohistochemically stained for neutrophils (inflammatory cells with red chromagen). (A) Tissue section from the left lung lobe of a control mouse that received 2 days of cobalt aspiration. A mixed inflammatory cell infiltration containing numerous neutrophils (arrows) is present in the centriacinar region that includes the terminal bronchiole (tb), alveolar duct (ad), and surrounding alveoli (a). (B) Tissue section from the left lung lobe of a HIF1αΔ/Δ mouse that also received 2 days of cobalt aspiration. In contrast to the lung section in (A), the centriacinar inflammatory cell influx in (B) contains markedly less neutrophils.

FIG. 7.

Cytokine levels in BALF from cobalt-treated control and HIF1αΔ/Δ mice. The levels of cytokines in BALF were assessed using BD CBA Mouse Soluble Protein Flex Sets and FACSCalibur flow cytometer following exposure of saline-treated (white bars) or cobalt (hatched bars)-treated control mice and saline-treated (checkered bars) or cobalt (black bars)-treated HIF1αΔ/Δ mice. n > 5 mice per group. Outliers were removed by Grubb's test, and ANOVA was performed with Boneferroni post-test. a = significance at p < 0.05 when cobalt treated group is compared with controls within genotype. b = significance at p < 0.05 when cobalt treated groups are compared across genotypes.