The authors have declared that no competing interests exist.
Conceived and designed the experiments: JYK JWP. Performed the experiments: JYK JHS. Analyzed the data: JYK. Contributed reagents/materials/analysis tools: JHL. Wrote the paper: JYK JWP.
Obesity is a known risk factor for allergic asthma. It has been recognized as a key player in the pathogenesis of several inflammatory disorders via activation of macrophages, which is also vital to the development of allergic asthma. We investigated the mechanism of obesity-related asthma and whether treating obesity through exercise or diet ameliorates the severity of asthma in the obesity-related asthma model. We generated diet-induced obesity (DIO) in C57BL/6 mice by high-fat-feeding and ovalbumin-induced asthma (lean-OVA or DIO-OVA). The DIO-OVA mice were then treated with tumor necrosis factor (TNF)-α neutralizing antibody as a TNF-α blockade or a Cl2MDP-containing liposome to induce an alveolar macrophage deficiency. To treat obesity, the DIO-OVA mice were under dietary restrictions or exercised. The pathophysiological and immunological responses were analyzed. Airway hyperresponsiveness (AHR), serum IgE and TNF-α levels in the lung tissue increased in the DIO-OVA mice compared to the lean-OVA mice. Both the TNF-α blockade and depletion of alveolar macrophages in the DIO-OVA mice decreased AHR compared to the DIO-OVA mice. Treating obesity by exercise or through dietary means also reduced pulmonary TNF-α levels and AHR in the DIO-OVA mice. These results suggest that restoring normal body weight is an appropriate strategy for reducing TNF-α levels, and controlling inflammation may help improve asthma severity and control in obesity-related asthma.
Obesity is a metabolic disease and a major risk factor for several noncommunicable diseases, such as diabetes, and cardiovascular diseases. The World Health Organization estimates that more than 1.4 billion adults are overweight, and of these overweight adults, over 200 million men and nearly 300 million women are obese [
The mechanisms of action between obesity and asthma are not fully understood. Clinical studies showed that subjects with obesity-related asthma usually have noneosinophilic asthma, unexplained by Th2 immune responses [
To treat allergic symptoms in obesity-related asthma, several investigators have suggested that weight reduction by diet, exercise, or bariatric surgery might prevent the development of asthma, or at least decrease asthma-related symptoms, and improve asthma-specific quality of life, as measured by questionnaire or degree of health care utilization [
In this study, we investigated the mechanism of obesity-related asthma and whether treating obesity through weight reduction affects the pathogenesis of the obesity-related asthma model.
Female C57BL/6 mice (4-weeks old) were purchased from Japan-SLC (Hamamatsu, Japan) and were randomly allocated to experimental groups. Total of three independent experiments were performed and each experimental data was obtained from five mice per group. According to the retrospective statistical calculation for this study (
To make diet-induced obesity (DIO) mice, the 4-week-old C57BL/6 mice were fed a high fat diet (HFD) for 16 weeks. The HFD (D12492; Research Diets, Inc., New Brunswick, NJ) contained 60% kcal from fat. The lean mice, as a control, were fed a normal chow diet (D12450B; Research Diets, Inc.) containing 10% kcal from fat (
(a) Scheme of this study. C57BL/6 mice fed HFD for 16 weeks and some of the DIO mice underwent OVA sensitization and challenge (DIO-OVA). Some of the DIO-OVA mice were treated with TNF-α neutralizing antibody for TNF-α blockade or a Cl2MDP-containing liposome for alveolar macrophage depletion. For the treatment of obesity, the DIO-OVA mice performed voluntary exercise (DIO-OVA-Ex) or underwent dietary restriction (DIO-OVA-N) after 12 weeks of HFD feeding. (b) Body weight and (c) blood glucose tolerance was measured at the end of 16 weeks after HFD feeding. *, Statistical significance to lean mice (p<0.05); #, Statistical significance to DIO mice.
A glucose tolerance test (GTT) was performed 16 weeks after HFD feeding. Mice were fasted overnight and were intraperitoneally injected with glucose (2 g/kg body weight; Sigma-Aldrich, St. Louis, MO). Blood from the tail vein was collected at 0, 15, 30, 60, and 120 min after glucose injection. Blood glucose levels were measured with a blood glucose meter (One Touch Ultra; Life Scan, Inc., Milpitas, CA).
After 12 weeks of HFD feeding, some of the DIO mice were housed individually and performed voluntary exercise using a Mouse Igloo Fast-Trac (Bio-Serv, Frenchtown, NJ) for 4 weeks (
To make an established asthma model in lean or DIO mice, some of the mice underwent intraperitoneal sensitization with a mixture of OVA (100 μg per mouse; Sigma-Aldrich) and Imject Alum (100 μL per mouse; Thermo Scientific, Rockford, IL) at 12 and 14 weeks, followed by an intranasal challenge 3 times with the same antigen (OVA, 10μg per mouse) via intranasal route at 16 weeks. Two days after last OVA challenge, we sacrificed the asthmatic mice for analysis.
Liposomes encapsulated with dichloromethylene diphosphonic acid disodium salt (Cl2MDP, Sigma-Aldrich) were prepared as previously described [
To block TNF-α, goat anti-mouse TNF-α polyclonal antibody (10 μg/mouse; R&D Systems, Inc., Minneapolis, MN) or goat IgG (10 μg/mouse, R&D Systems) was injected via intravein route 12 hours before the OVA challenge.
Lung functions were measured as previously described [
To collect bronchoalveolar lavage (BAL) fluid, the lungs were lavaged with 1 mL Hank’s balanced salt solution via the tracheostomy tube. BAL cells were counted with a hemocytometer, smeared by cytocentrifugation (Cytospin3, Thermo, Billerica, MA) at 1000 rpm for 3 min, and then stained with a Hemacolor Staining Kit (Merck, Darmstadt, Germany). BAL cells from each group were counted and classified as macrophages, lymphocytes, neutrophils, or eosinophils. To minimize the effects of subjective bias in the classification of the BAL cells, blind outcome assessment was used.
For protein extraction, lung tissues were homogenized in 20 mL/g tissue protein extraction reagent (Thermo Fisher Scientific Inc., Rockford, IL) using a tissue homogenizer (Biospec Products, Bartlesville, OK). Homogenates were incubated at 4°C for 30 min and then centrifuged at 1000 × g for 10 min. Supernatants were collected, passed through a 0.45-micron filter (Gelman Sciences, Ann Arbor, MI), and then stored at -70°C for assessment of cytokine levels. The measured cytokine levels were normalized to lung tissue weight and expressed as ng per mL per lung tissue.
TNF-α, and total and specific IgE levels from lung homogenates or blood sera were measured by means of enzyme-linked immunosorbent assay (ELISA) with commercially available materials. Briefly, TNF-α and total IgE levels were assayed with Mouse TNF-α DuoSet (R&D Systems) and BD OptEIA Mouse IgE ELISA Set (BD Pharmingen, San Diego, CA), respectively. Specific IgE levels were measured by modified-sandwich ELISA [
Periodic Acid-Schiff (PAS) staining was performed in the formalin-fixed/paraffin-embedded lung tissues. Tissue sections were examined with an Olympus BX40 microscope in conjunction with an Olympus U-TV0.63XC digital camera (Olympus Corp., Melvile, NY). Images were acquired using DP Controller and Manager software (Olympus Corp.).
For assessments of cell phenotypes (CD11bIntF4/80High cells) in the lung tissue, multicolor-flow cytometric analysis was performed (LSRII; BD Biosciences). The data were analyzed using FACSDiva (BD Biosciences) or FlowJo ver.7.6.2 (Three Star, Ashland, OR) and expressed as a percentage value.
The data are expressed as mean±standard error (n = 5). Statistical analyses were performed using SPSS ver. 12.0 (Chicago, IL). Groups in the GTT and the methacholine challenge tests were compared using two-way ANOVA with Tukey post hoc analysis, and the others were compared using one-way ANOVA with Bonferroni post hoc analysis. All differences were considered significant at
During the experimental period, the DIO mice were fed a HFD and their body weights significantly increased by 75% more than the lean mice at 16 weeks (
To determine whether dietary obesity affects baseline lung function in a mouse model, airway and lung parenchymal mechanics were measured. RL was significantly greater in DIO mice than in lean mice, and the increased RL of the DIO mice was decreased in the DIO-N mice with statistical significance (
Lean | 0.64±0.10 | 0.043±0.007 | 0.31±0.06 | 4.57±0.71 | 19.58±1.84 |
DIO | 1.00±0.09* | 0.032±0.007 | 0.67±0.24 | 7.66±2.43 | 25.65±7.31 |
DIO-Ex | 0.79±0.17 | 0.048±0.003 | 0.43±0.11 | 4.25±0.34 | 17.16±1.35 |
DIO-N | 0.63±0.06# | 0.05±0.001 | 0.33±0.07 | 4.18±0.14 | 17.94±1.70 |
Values are means±SD; RL, pulmonary resistance; Cdyn, dynamic compliance; Raw, airway resistance; G, tissue damping; H, tissue elastance; DIO, diet induced obesity; Ex, free-wheel exercise; diet, normal chow diet; *, P<0.05 compared with control group; #, P<0.05 compared with DIO group. All data are representative of three independent experiments.
To determine whether obesity exacerbates asthma, dietary-induced obese mice were subjected to a protocol of OVA sensitization and challenges (DIO-OVA mice;
(a) AHR, (b) inflammatory cell infiltrations in the BAL fluids, (c) total IgE and (d) OVA-IgE levels in the sera were measured in the asthma model (lean-OVA) and obesity-related asthma model (DIO-OVA). *, Statistical significance to their control group (lean or DIO; p<0.05); #, Statistical significance between lean-OVA and DIO-OVA (p<0.05). Error bars indicated mean±SEM of five mice per group. All data are representative of three independent experiments.
As a notable feature of the asthma model, increased total and OVA-specific IgE levels in blood were revealed in both lean-OVA and DIO-OVA mice. However, there were no significant differences in OVA-specific IgE between the lean-OVA and DIO-OVA mice. Interestingly, specific IgE was not elevated, but total IgE levels in the DIO mice were significantly elevated compared with the lean mice (
As a representative pathologic change in the lung of the asthma model, goblet cell hyperplasia in the peri-bronchiolar area was measured, but we found no significant differences between the lean-OVA and DIO-OVA mice (
Previously, we proposed that TNF-α plays a key role in the allergic asthma model and alveolar macrophages can produce large amounts of TNF-α in the lung environment [
We found significantly higher TNF-α levels in the lung homogenates of DIO-OVA mice compared with the lean-OVA mice (
TNF-α levels in (a) the bronchoalveolar lavage fluids and (b) the blood sera were measured in the asthma models. The solid lines indicate statistical significance between each group (p<0.05).
To confirm the role of TNF-α in the obesity-related asthma model, TNF-α was blocked 12 hours before the first OVA challenge (
(a) Both lean-OVA and DIO-OVA mice were treated with TNF-α blockade antibody or Cl2MDP for depletion of TNF-α or alveolar macrophages, respectively, and TNF-α levels in the lung homogenates were measured. (b) MCh AHR was measured in the TNF-α or alveolar macrophage depleted DIO-OVA. *, Statistical significance to lean mice (p<0.05); #, Statistical significance to DIO mice.
To examine the function of alveolar macrophages in the obesity-related asthma model, lean-OVA and DIO-OVA mice were treated with Cl2MDP-containing liposomes intranasally 12 hours before the first OVA challenge (
In this part, we found more decreased AHR in the TNF-α or alveolar macrophage depleted DIO-OVA mice compared with the TNF-α or alveolar macrophage depleted lean-OVA mice (
In this study, we identified that pre-existing obesity exacerbates subsequent asthmatic lesions via alveolar macrophages and their TNF-α production. For that reason, we investigated whether treating obesity ameliorates the severity of asthma. Both exercise and normal diet dramatically reduced TNF-α secretion from the lung homogenates of DIO-OVA mice (
DIO mice performed voluntary exercise or consumed a normal chow diet to treat obesity. TNF-α levels in the lung homogenates were measured in the weight-reduced, obesity-related asthma mice. The solid lines indicate statistical significance between each group (p<0.05). Error bars indicated mean±SEM of five mice per group. All data are representative of three independent experiments.
DIO mice performed voluntary exercise or consumed a normal chow diet to treat obesity. (a) Airway hyperresponsiveness and (b) inflammatory cell infiltration in the bronchoalveolar lavage fluid were measured in the weight-reduced, obesity-related asthma mice. *, Statistical significance to lean mice (p<0.05); #, Statistical significance to DIO mice.
Obesity is a disease of affluence and the prevalence of obesity has increased. Asthma, another non-communicable disease, also has increased. Some studies suggest that obesity plays a substantial role in the development, control, and severity of asthma [
Many studies have supported an association between obesity and asthma, but the particular mechanisms for this linkage have not been fully investigated. As one possible cue, obesity is associated with a state of low-grade systemic inflammation and leads to an increased release of proinflammatory molecules that play a crucial role in the pathogenesis of obesity-associated complications [
To clarify the relationship between obesity and asthma, adipose-derived secreted factors have been studied. Leptin, a typical adipokine, regulates feeding behavior through the central nervous system. It is produced in adipocytes and then secreted to the blood stream. Leptin levels in the blood positively correlate with adipose mass in human subjects, as well as in mice. Adiponectin has also been identified as an adipocyte-specific adipokine and its expression is found to be lower in obesity. In this study, leptin and adiponectin levels in both lung homogenates and blood sera were not noticeably different between the obese and obesity-treated mice. However, we found that TNF-α levels were greater in the lungs and blood of DIO-OVA mice. In clinical and epidemiological studies, TNF-α levels are greater in the adipose tissue and plasma of obese individuals, and weight control in these individuals is associated with a decrease in TNF-α expression [
Previously, we identified that TNF-α plays an important role in the etiology of asthma and alveolar macrophages can produce large amounts of TNF-α in the lung environment [
In many studies, anti-TNF-α therapy has been used as a strategy for the treatment of asthma [
Various types of immune cells that produce TNF-α exist in the lung environment and alveolar macrophages are one of these cell types. Activation of alveolar macrophages results in airway inflammation and pulmonary fibrosis [
Judging from the relationship between obesity and asthma, it may be possible that an appropriate treatment of obesity leads to recovery of asthma lesions in the obesity-related asthma model. To investigate our last question, we proceeded to treating obesity.
Caloric restriction based on a low calorie intake, and additional energy expenditure by physical exercise, are typical treatments for obesity [
Restoration of body weight certainly has a therapeutic effect on asthmatic diseases. In systematic reviews on obesity and asthma, several investigators have noted improvements in asthma outcome and control after weight loss in overweight or obese subjects [
There are several missing points in this study that should be elucidated through further studies. Particularly warranted are investigations into which factors stimulate TNF-α-producing alveolar macrophages in the obesity-related asthma and which mechanisms are involved in the relationship between treatments of obesity and improved asthmatic lesions.
In summary, we identified that inflammatory response is a possible mechanism that can explain the relationship between obesity and asthma, and control of the inflammatory response through the TNF-α pathway is a means for treating obesity-related asthma. We also showed that restoring normal body weight reduces pulmonary TNF-α levels and then improves typical symptoms of asthma in obesity-related asthma mice, which suggests that restoring normal body weight is an appropriate treatment strategy for obesity-related asthma.
(PDF)
Lung tissues were stained with PAS. All data are representative of three independent experiments.
(TIF)
Lean-OVA mice was treated with TNF-α blockade antibody or Cl2MDP for depletion of TNF-α or alveolar macrophages, respectively. (a) Airway hyperresponsiveness and (b) alveolar macrophage levels in the lung tissue were measured in the TNF-α or alveolar macrophage depleted DIO-OVA. *, Statistical significance to lean mice (p<0.05); #, Statistical significance to DIO mice.
(TIF)
DIO-OVA and Lean-OVA mice was treated with TNF-α blockade antibody or Cl2MDP for depletion of TNF-α or alveolar macrophages, respectively. Inflammatory cell infiltrations in the BAL fluid were measured in the TNF-α or alveolar macrophage depleted mice. Error bars indicated mean±SEM of five mice per group. All data are representative of three independent experiments.
(TIF)
DIO mice performed voluntary exercise or diet restriction for the treatment of obesity. (a) Leptin and (b) adiponectin levels of the lung homogenates and blood sera were measured in the weight-reduced obese asthma mice.
(TIF)