Alteration of the nasal responses to influenza virus by tobacco smoke

Abstract

Purpose of review: The purpose of this review is to highlight recent data regarding the impact of exposure to tobacco smoke on influenza virus infection. This is timely because of the continuing pattern for influenza to cause epidemics and pandemics.

Recent findings: Experimental animal studies suggest tobacco smoke increases severity of respiratory disease with influenza. The interaction is complex and dependent on dose and chronicity of both virus and smoke exposure. Smoke-induced oxidant stress and suppression of innate immunity are mechanistic factors leading to worse disease. Experiments using human respiratory cells show that tobacco smoke increases viral replication through mechanisms including suppression of antiviral pathways and altered cytokine patterns in cell types with central roles in mucosal innate immunity, such as epithelium, dendritic cells, and natural killer cells. Studies also suggest a role for antioxidant strategies in reducing risk. Human volunteer studies using live attenuated influenza virus as a model appear to corroborate many of these findings.

Summary: Exposure to tobacco smoke remains extremely prevalent worldwide. Although avoidance of exposure is a primary goal, it is important to understand the mechanisms underlying increased infection risk with tobacco smoke and other pollutant exposures, so that novel preventive or treatment strategies can be developed.

Fig. 1

Expression of IRF7 in NECs from smokers and nonsmokers. The NECs were infected with influenza A Bangkok/2/79 or left uninfected, and analyzed for IRF7 expression 24 hours after infection. (A) Total RNA was analyzed for IRF7 mRNA and normalized to β-actin mRNA concentrations (n = 16 smokers; n = 14 nonsmokers). (B) Whole cell lysates were analyzed for IRF7 and IRF3 protein levels by Western blotting. Membranes were stripped and analyzed for β-actin to assure equal loading. Representative immunoblots are shown. (C) Densitometric analysis of IRF7 protein levels (n = 6 smokers; n = 6 nonsmokers). *Significantly different from noninfected cells (P < 0.05). ^# Significantly different from smokers (P < 0.05). Previously published in reference [16].

Fig. 2

Total RNA from nonsmoker and smoker NEC/mono-DC co-cultures was collected 24 h post influenza infection. qRT-PCR (normalized to β-actin and expressed as fold induction over non-infected control) was performed for (A) IP-10 in NEC, (B) IP-10 in mono-DC, (C) RANTES inNEC, and (D) RANTES in mono-DC. Data are expressed as mean ± SEM. Nonsmoker n=5, smoker n=5. #p<0.05, ##p<0.01, ###p<0.001 vs non-infected control, *p<0.05, ***p<0.001 for nonsmoker vs. smoker. Previously published in reference [28].

Fig. 3

Time course of influenza virus measured by qRT-PCR as ratio of influenza type B HA RNA to β-actin mRNA in NLF cells and cytokines in NLF, after inoculation with LAIV in healthy nonsmoking subjects (controls). Cytokine data are shown as fold change (△) from day 0 baseline, to illustrate differential responses among the cytokines. All data points are shown as median for study day. IP-10, IL-6, and IFNγ but not other cytokines showed statistically significant increases (days 2-4) compared with day 0. Previously published in reference [35].

Fig. 4

(A) Mixed-effects model curves for influenza B virus quantity (hemagglutinin RNA normalized to β-actin RNA) and (B) for IL-6 in NLF from Controls (blue circles and solid line), SHS-exposed nonsmokers (red circles and broken line), and active Smokers (black circles and dotted line). In the linear mixed model fitting, the response variable is the log-transformed fold change from the baseline. Covariates include time, time², group and their interactions. Not previously published.

Fig. 5

Cytotoxic NK cell percentages and activity were analyzed in NLF of nonsmokers and smokers after LAIV inoculation. A) CD56⁺CD16⁺ cytotoxic NK cell percentages of total NK cells were decreased in the NLF of smokers following LAIV. Kruskal-Wallis p=0.09, *p<0.05 nonsmoker vs smoker posttest. Nonsmokers n=12 (■, solid line), smokers n=9 (▲, dashed line). B) Granzyme B activity was decreased in NLF of smokers following LAIV inoculation. Kruskal-Wallis p<0.01, *p<0.05 nonsmoker vs smoker posttest. Nonsmokers n=13 (■, solid line), smokers n=13 (▲, dashed line). Previously published in reference [38].