Investigating the role of HMGB1 in the regulation of human airway smooth muscle cell function

Leonarda Di Candia1, Edith Gomez2, John Challiss2, Christopher Brightling1, Ruth Saunders1. 1Institute for Lung Health, Department of Infection, Immunity and Inflammation, University of Leicester, Leicester, UK, 2Department of Cell Physiology and Pharmacology, University of Leicester, Leicester, UK

Introduction: Levels of high mobility group box 1 (HMGB1) protein, which signals through the receptor for advanced glycosylation end-products (RAGE) to promote inflammation and tissue repair [1], are increased in the airways of asthmatic subjects [2]. We hypothesized that the HMGB1/RAGE axis contributes to the dysregulation of airway smooth muscle cell (ASMC) function in asthma [3].

Aim: To determine RAGE and HMGB1 expression in cultured human ASMCs, HMGB1 levels in sputum samples and the effects of HMGB1 stimulation on hASMC function.

Methods: hASMCs were isolated from bronchial biopsies and large airway tissue from subjects undergoing surgery. Sputum samples were obtained by sputum induction. The procedures were approved by the Leicestershire Ethics Committee. All subjects gave written informed consent. hASMCs were characterized for α-smooth muscle actin expression. RAGE and HMGB1 expression were characterized by PCR and flow cytometry, sputum HMGB1 levels by ELISA, NF-κB subunit p65 nuclear translocation by immunofluorescence and hASMC migration using the Oris™ cell migration kit (Platypus Technologies).

Results: hASMCs expressed full-length and soluble RAGE mRNA isoforms, and cell-surface RAGE protein (9.1 ± 1.9% of hASMCs, mean ± SEM, n=11 in controls vs 14.1 ± 2.7% in asthmatic subjects, n=11; p=0.139); HMGB1 mRNA (0.78 ± 0.09 relative to 18S rRNA, n=7 in controls vs 0.59 ± 0.07 in asthmatic subjects, n=9; p=0.103), and HMGB1 protein (31.5 ± 5.5%, n=10 in controls vs 16 ± 2.1% in asthmatic subjects, n=11; p=0.022). A trend towards an increase in HMGB1 concentration in sputum supernatants from severe asthma vs non-asthmatics and mild/moderate asthma was seen (177 ng/mL, (87-457 ng/mL); median (25-75% percentiles) in controls; 227 ng/mL (113-426 ng/mL) in mild/moderate asthma; 553 ng/mL (380-997 ng/mL) in severe asthma; p=0.064). HMGB1 exists in two forms with differential cell functions: a reduced form with chemotactic activity, and a form containing a disulphide bond, which causes cytokine release [4]. 3S-HMGB1, a mutated form of HMGB1 resistant to inactivation by oxidation, did not induce increased hASMC migration after 24 h incubation at 3, 10 or 30 ng/mL (n=7, p=0.155; n=5, p=0.88; n=5, p=0.35, respectively). In addition, recombinant human HMGB1 (rhHMGB1, 30, 300, 1000 ng/mL) failed to evoke nuclear translocation of p65 after 5-120 min, while TNFα (20 ng/mL) caused a clear nuclear translocation of p65 in hASMCs (n=3).

Conclusions: hASMCs express RAGE, with no difference seen between asthmatic and non-asthmatic subjects. HMGB1 expression is reduced in hASMCs from asthmatic subjects, although the functional consequences of this remain unclear. Sputum data suggest that hASMCs may be exposed to more HMGB1 in the asthmatic airways. rhHMGB1 or 3S-HMGB1 (both reduced forms) did not induce NF-κB activation or migration of hASMCs. We are currently investigating the different HMGB1 forms present in sputum and the functional effects of the HMGB1 form that induces cytokine release on hASMCs.

[1] Klune JR, et al. (2008) Mol Med 14, 476-484.

[2] Watanabe T, et al. (2011) Respir Med 105, 519-525.

[3] Koziol-White CJ, et al. (2011) Expert Rev Respir Med 5, 767-777.

[4] Venereau E, et al. (2012) J Exp Med 209, 1519-1528.