Local nebulization of 1α,25(OH)2D3 attenuates LPS-induced acute lung inflammation

Abstract

Background: Evidence supports a critical role of vitamin D status on exacerbation in chronic obstructive pulmonary disease, indicating the need to avoid vitamin D deficiency in these patients. However, oral vitamin D supplementation is limited by the potential risk for hypercalcemia. In this study, we investigated if local delivery of vitamin D to the lungs improves vitamin D-mediated anti-inflammatory action in response to acute inflammation without inducing hypercalcemia.

Methods: We studied vitamin D sufficient (VDS) or deficient (VDD) mice in whom 1α,25(OH)₂D₃ (0.2 μg/kg) or a vehicle followed by lipopolysaccharide (LPS 25 µg) were delivered to the lung as a micro-spray.

Results: Local 1α,25(OH)₂D₃ reduced LPS-induced inflammatory cells in bronchoalveolar lavage (BAL) in VDS (absolute number of cells: - 57% and neutrophils - 51% p < 0.01) and tended to diminish LPS-increased CXCL5 BAL levels in VDS (- 40%, p = 0.05) while it had no effect on CXCL1 and CXCL2 in BAL and mRNA in lung of VDS and VDD. It also significantly attenuated the increased IL-13 in BAL and lung, especially in VDD mice (- 41 and - 75%, respectively). mRNA expression of Claudin-18 in lung was significantly lower in VDS mice with local 1α,25(OH)₂D₃ while Claudin-3, -5 and -8 mRNA levels remained unchanged. Finally, in VDD mice only, LPS reduced lung mRNA expression of adhesion junction Zona-occludens-1, in addition to increasing uric acid and total protein in BAL, which both were prevented by local 1α,25(OH)₂D₃.

Conclusion: Under normal levels of vitamin D, local 1α,25(OH)₂D₃ nebulization into the lung efficiently reduced LPS induction of inflammatory cells in BAL and slightly attenuated LPS-increase in CXCL5. In case of severe vitamin D deficiency, although local 1α,25(OH)₂D₃ nebulization failed to significantly minimize cellular inflammation in BAL at this dose, it prevented epithelial barrier leakage and damage in lung. Additional research is needed to determine the potential long-term beneficial effects of local 1α,25(OH)2D3 nebulization on lung inflammation.

Keywords: Acute lung inflammation; Local vitamin D nebulization; Vitamin D deficiency.

Fig. 1

Study design. Three-week old male mice were fed a vitamin D deficient or standard control diet during the whole study period. At the age of 8–9 weeks, mice were instilled via an aerosolizer syringe with 0.2 μg/kg 1α,25(OH)₂D₃ or a vehicle 2 h prior to a 25 μg LPS instillation and measurements were performed 24 h later

Fig. 2

Absolute number of total cells (A), neutrophils (B) and macrophages (C) in BAL of vitamin D sufficient or deficient mice treated either with vehicle (open circles) or 1α,25(OH)₂D₃ (closed circles) followed by LPS. NS: not significant, **: p < 0.01 (Kruskal–Wallis); ##: p < 0.01 (Mann–Whitney). Data are expressed as median ± IQR

Fig. 3

Chemokines A CXCL1, B CXCL2 and C CXCL5 in BAL and mRNA expression of D CXCL1, E CXCL2 and F CXCL5 in lung homogenate in vitamin D sufficient or deficient mice treated either with vehicle (open circles) or 1α,25(OH)₂D₃ (closed circles) followed by LPS. mRNA data expressed relative to RPL27. NS: not significant, φ: p = 0.05 (Mann Whitney). Data are expressed as median ± IQR

Fig. 4

Lung mRNA expression of tight junctions namely claudin (Cldn) 3 (A), 5 (B), 8 (C) and 18 (D) and adhesion junction Zona occludens-1 (ZO-1) (E) in vitamin D sufficient or deficient mice treated either with vehicle (open circles) or 1α,25(OH)₂D₃ (closed circles) followed by LPS. mRNA data expressed relative to RPL27. * and #: p < 0.05 (Kruskal–Wallis and Mann Whitney); +  + : p < 0.01 (Mann Whitney). Data are expressed as median ± IQR

Fig. 5

Protein (A) and uric acid (B) concentration in BAL fluid expressed as µg/ml and µM, respectively and serum Surfactant protein-D (SP-D) concentration expressed as pg/ml measured in vitamin D sufficient or deficient mice treated either with vehicle (open circles) or 1α,25(OH)₂D₃ (closed circles) followed by LPS. NS: not significant, ***: p < 0.001 (Kruskal–Wallis); φ: p = 0.05 (Mann Whitney); #: p < 0.05 (Mann Whitney); ###: p < 0.001 (Mann Whitney) +++ : p < 0.005 (Mann Whitney). Data are expressed as median ± IQR