Adipokines, factors produced by adipose tissue, may be proinflammatory (such as leptin and resistin) or anti-inflammatory (such as adiponectin). Effects of these adipokines on the lungs have the potential to evoke or exacerbate asthma. This review summarizes basic mechanistic data through population-based and clinical studies addressing the potential role of adipokines in asthma. Augmenting circulating concentrations of adiponectin attenuates allergic airway inflammation and airway hyperresponsiveness in mice. Murine data is supported by human data that suggest that low serum adiponectin is associated with greater risk for asthma among women and peripubertal girls. Further, higher serum total adiponectin may be associated with lower clinical asthma severity among children and women with asthma. In contrast, exogenous administration of leptin results in augmented allergic airway hyperresponsiveness in mice. Alveolar macrophages obtained from obese asthmatics are uniquely sensitive to leptin in terms of their potential to augment inflammation. Consistent with this basic mechanistic data, epidemiologic studies demonstrate that higher serum leptin is associated with greater asthma prevalence and/or severity and that these associations may be stronger among women, postpubertal girls, and prepubertal boys. The role of adipokines in asthma is still evolving, and it is not currently known whether modulation of adipokines may be helpful in asthma prevention or treatment.
Obesity is increasingly appreciated as a risk factor for asthma and has been the subject of multiple recent reviews in the literature [
Adiponectin is an insulin sensitizing hormone that also plays a role in inflammation. Adiponectin inhibits effects of proinflammatory cytokines, such as tumor necrosis factor (TNF)-
Curiously, even though visceral adipocytes are its most important source [
Adiponectin monomers have a globular head and a collagen-like tail. However, adiponectin monomers do not circulate. Instead, adiponectin multimerizes and circulates in the blood as trimeric, hexameric, and higher order multimeric forms that have low, medium, and high molecular weights (LMW, MMW, and HMW), respectively, as shown in Figure
Schematic representation of the sexual dimorphism of the absolute concentrations of the circulating adiponectin isoforms. Compared to men, women have higher absolute concentrations of circulating total adiponectin (mean values of 11.3 versus 23.5
Adiponectin isoforms vary in efficacy. For example, HMW adiponectin is the most biologically active isoform with respect to insulin sensitivity [
Several adiponectin binding proteins have been identified, including AdipoR1, AdipoR2, T-cadherin, and calreticulin. Adiponectin can also induce effects in a receptor-independent fashion. Multiple cell types in the lung express adiponectin binding proteins, including the bronchial epithelium [
T-cadherin appears to be important for adiponectin transport into the lungs. The concentration of adiponectin in murine bronchoalveolar lavage (BAL) fluid is relatively high [
Given the presence of adiponectin and its receptors in the lung and the declines in adiponectin concentrations in obesity, it is conceivable that the loss of the anti-inflammatory effects of adiponectin in obesity contributes to asthma prevalence or severity. Below, we discuss data exploring this hypothesis. We first discuss studies performed in animals and then present data from population-based and clinical studies in humans.
In mice, obesity-related declines in adiponectin appear to contribute to the development of type 2 diabetes mellitus and atherosclerosis. Exogenous administration of adiponectin protects obese mice against these conditions, while adiponectin knockout mice are susceptible (see [
A schematic representation of the suggested role for adiponectin in allergen induced asthma in mice, based upon the work by Shore et al. [
Despite the marked reductions in allergic airways responses observed with adiponectin treatment after acute allergen challenge in sensitized mice [
Studies addressing the nature of the receptors that contribute to the protective effect of adiponectin for asthma on the lung are more limited. Williams et al. observed a marked reduction in allergen induced airway hyperresponsiveness, eosinophil recruitment to the airways, Th2 cytokine expression, and mucous cell hyperplasia in mice deficient in T-cadherin [
The cell type that is the target of effects of adiponectin that limit allergic airway responses has also not been established. In addition to epithelial cells, endothelial cells, and airway smooth muscle (see above), it is conceivable that lung macrophages are involved. Eotaxin, an important eosinophil chemotactic factor, is released from cultured bone marrow derived macrophages treated with IL-4 and TNF, and adiponectin attenuates this release [
It is interesting to note that compared to wild-type mice, T-cadherin deficient mice with high serum adiponectin also had reductions in BAL IL-17A expression after allergen challenge [
Despite the largely anti-inflammatory effects of adiponectin in the setting of allergic airways disease, adiponectin also appears capable of causing proinflammatory/proasthmatic effects in the lungs in response to other stimuli. Following high dose acute ozone exposure (2 ppm for 3 h), mice develop airway hyperreactivity and a neutrophilic inflammation characterized by increases in acute phase cytokines and chemokines. These responses are attenuated in mice deficient in adiponectin [
There are also data from human subjects suggesting a protective role for adiponectin in asthma. It is important to note that reagents necessary to perform mechanistic studies in humans are still lacking and awaiting a better understanding of adiponectin signalling processes. Hence, studies in human subjects have been mostly limited to associations between circulating or lung levels of adiponectin and clinical disease outcomes with limited data on inflammatory outcomes (Table
A tabular summary of the current evidence supporting the roles for systemic adiponectin or leptin with respect to asthma in human subpopulations.
Adiponectin-asthma association | Leptin-asthma association | |||
---|---|---|---|---|
Asthma prevalence | Asthma severity | Asthma prevalence | Asthma severity | |
Males | ||||
Prepubertal boys | Inadequately studied (#) | Beneficial effect on exercise induced bronchoconstriction and |
Harmful effect [ |
Harmful effect on clinical outcomes; peak expiratory flow rates; exercise induced bronchoconstriction [ |
Peri/postpubertal boys | No effect [ |
Beneficial effect on clinical outcomes and FEV1/FVC ratio [ |
No effect [ |
No effect [ |
Men | Unclear effect—no effect on clinical outcomes [ |
Harmful effect on clinical outcomes [ |
No effect [ |
Inadequately studied (#) |
| ||||
Females | ||||
Prepubertal girls | Inadequately studied (#) | Inadequately studied (#) | Inadequately studied (#) | Harmful effect on clinical outcomes [ |
Peri/postpubertal girls | Beneficial effect [ |
Possible beneficial effect on clinical outcomes and spirometry [ |
Harmful effect [ |
Harmful effect on clinical outcomes [ |
Premenopausal women | Beneficial effect [ |
Beneficial effect on clinical outcomes [ |
Harmful effect [ |
Inadequately studied (#) |
Postmenopausal women | Inadequately studied (#) | Beneficial effect on clinical outcomes [ |
Harmful effect [ |
Harmful effect on clinical outcomes and FEV1/FVC ratio [ |
Note 1: Inadequately studied, beneficial; and harmful associations as well as no effects or unclear effects are depicted by different symbols (#, *, †, and ‡).
The strongest evidence supporting a relationship between adiponectin and asthma comes from a US-based longitudinal cohort that showed that low serum total adiponectin concentrations (
Unlike the above-mentioned longitudinal study [
Nonlinear relationship between serum adiponectin and risk of incident asthma in a US-based longitudinal study [
Another explanation for these apparent discrepancies is that serum concentration of adiponectin may be a poor surrogate for the adiponectin that actually impacts asthma. In a preliminary study by Sood et al. on nonsmokers, sputum total adiponectin concentrations were lower among asthmatics than controls [
Although it is certainly possible that adiponectin does not modify asthma in humans, another potential explanation for discrepant findings between studies lies in the possible nonlinear effects of adiponectin that are missed by linear statistical modelling. These effects may be threshold effects (i.e., manifest only when a critical threshold in serum concentrations is reached, as suggested by Sood et al. [
In a large community-based cross-sectional study of subjects with asthma, Sood et al. showed that high serum total adiponectin was associated with more frequent active disease (including more frequent use of any asthma medication) and greater number of respiratory symptoms and asthma medications among men with asthma but beneficial effects among women with asthma, with significant sex-specific interactions [
Pediatric studies, examining mostly prepubertal and peripubertal asthmatic boys, show a consistent trend. High serum total adiponectin concentrations were associated with less severe exercise induced bronchoconstriction [
In a small study of obese women undergoing bariatric surgery, adiponectin mRNA expression in visceral abdominal (omental) adipose tissue from asthmatics was lower than that from controls, even after adjustment for BMI, and even though no differences in serum adiponectin concentrations were reported between the two groups [
Another explanation for the data from the morbidly obese asthmatics undergoing bariatric surgery is that the change in adiponectin expression was a consequence and not a cause of asthmatic bronchoconstriction. Indeed, others have shown that acute severe asthma exacerbations result in a transient decrease in serum adiponectin concentrations [
In summary, adiponectin is present in the lung, and adiponectin receptors are expressed on key lung cells. There are relatively strong data indicating that adiponectin limits allergic airways responses in lean mice, though in models of non allergic asthma, such as ozone exposure, adiponectin paradoxically promotes rather than inhibits airway hyperresponsiveness. Similarly, it is possible that proinflammatory airway effects of adiponectin dominate under certain physiologic conditions in humans, while anti-inflammatory airway effects dominate under others. For instance, in subjects with chronic obstructive pulmonary disease (COPD), an inflammatory obstructive airway disease related to asthma, proinflammatory airway effects of adiponectin appear to dominate [
The hormone leptin, which derives its name from the Greek word leptos, meaning thin, is a 16 kD protein derived from the obese (ob) gene and is expressed predominantly in adipocytes [
Leptin mediates its effect by binding to the leptin receptor (OB-R), a single membrane spanning receptor of the class I cytokine family with closest homology to gp130 of the IL-6 family of cytokine receptors. Several OB-R variants are generated by alternative splicing. The intracellular domain of each of these, except OB-Re, which lacks the transmembrane domain and is a soluble receptor, contains an interaction motif for members of the janus kinase (JAK) family of tyrosine kinases. However, only the long form, OB-Rb, contains a binding motif for members of the signal transducer and activator of transcription (STAT) family of transcription factors [
Leptin plays a key role in the regulation of appetite, metabolism, and body weight, and both leptin deficient and OB-Rb deficient mice are massively obese [
Effects of leptin on lung cells might also impact asthma. Bronchial epithelial cells are increasingly recognized as contributing to asthmatic airway inflammation. Bronchial epithelial cells express leptin receptors, and leptin causes epithelial cell proliferation and mucin protein expression [
Given the effects of leptin on the immune system and the lung, it is conceivable that obesity-related increases in leptin could initiate or worsen asthma. Data from both animal and human studies that have explored this hypothesis are discussed below.
Shore et al. [
A schematic representation of the suggested role for leptin in allergen induced asthma in mice, based upon the work by Shore et al. [
Leptin deficient (
Studies in human subjects have been mostly limited to associations between circulating or lung levels of leptin and clinical disease outcomes such as asthma prevalence or severity. Data on inflammatory outcomes is very limited. Further, since there are no longitudinal or interventional studies examining this association, the direction of causation for this association cannot be definitively established. Our understanding of the effect of leptin on asthma in humans, independent of the confounding effect of BMI, is therefore still evolving (see Table
A large cross-sectional analysis of participants of the US-based Third National Health and Nutrition Examination Survey (NHANES III) showed that women with high serum leptin concentrations (
The association between systemic leptin and asthma prevalence is more consistent in clinic-based studies of children than adults. In two case-control studies of prepubertal children by Guler et al. and Gurkan et al., serum leptin concentrations were higher in asthmatics than controls, independent of BMI [
Data in adults from clinic-based studies are less convincing. Lessard et al. showed higher sputum leptin concentrations in adult asthmatics than controls, but these findings may be confounded by the remarkably different BMI levels between the groups [
It is unclear why the leptin-asthma association is more consistent among children than adults. A possible explanation may be that asthma in children may be a more uniform phenotype, whereas in adults it is a heterogeneous collection of phenotypes and that leptin may be differentially associated with one of the various asthma phenotypes. Indeed, the asthma phenotype in children is more likely to be atopic than that in adults. Atopic status may modify the leptin-asthma association—this is suggested by the study by Guler et al. [
In a small clinical study of obese women undergoing bariatric surgery, visceral (i.e., omental) adipose tissue from asthmatics showed greater expression of leptin than controls after adjustment for BMI, even though no differences in serum concentrations were reported [
Relationship between visceral (omental) fat leptin expression and methacholine airway reactivity in morbidly obese women with asthma at the time of bariatric surgery (rho = −0.8;
There is also a recent human
Clinic-based studies of prepubertal asthmatic children, mostly boys, showed that greater serum leptin concentrations were associated with greater clinical asthma severity [
As was the case for asthma prevalence, the association between serum leptin and asthma severity is less consistent in adults than in children. A small case-control study showed that high serum leptin concentrations may discriminate women with severe asthma from mild/moderate asthma, although this association may be confounded by the remarkably different BMI values between the groups [
Even in studies demonstrating a leptin-asthma association, serum leptin does not appear to be the only intermediary factor that explains the obesity-asthma association. For example, Sood et al. [
In summary, although leptin and its receptors are expressed in human airway cells, our understanding of the relationship between leptin and asthma is still evolving. Current evidence suggests that systemic leptin or visceral fat expression of leptin may be associated with greater asthma prevalence and/or severity, particularly among prepubertal boys, peripubertal and postpubertal girls, and women. It remains to be established whether modulation of leptin, independent of BMI, may be helpful in asthma prevention or treatment.
Resistin (or “resistance to insulin”), a proinflammatory adipokine originally discovered in mice, was named for its ability to resist insulin action [
Much less data exists addressing a possible relationship between resistin and asthma than that existing for either leptin or adiponectin and asthma. Interestingly, human studies of resistin show associations opposite to those expected from
Although murine data are convincing, it is less clear whether adiponectin, leptin, or resistin plays a role in modulating asthma risk and/or severity in human subjects, although most of the human data are still limited to association studies. There are many possible explanations for the conflicting findings, as discussed above. Most of the human studies described above did not provide phenotypic characterization of asthmatics. In murine studies, there are opposing effects of adiponectin depending on whether the asthma model used was an allergic one or one related to oxidative stress. Hence, it is conceivable that adipokines may only be important for certain phenotypes of asthma. Indeed it is possible that the associations observed in certain subgroups (i.e., prepubertal boys) may relate to greater uniformity of asthma phenotypes within these populations. Finally, it is likely that adipokines are only one part of the obesity-asthma puzzle. Mechanical, developmental, hormonal, genetic, and epigenetic effects of obesity may also affect both asthma prevalence and severity in obese humans.
Adenosine monophosphate-activated protein kinase
Bronchoalveolar lavage
Body mass index
Chronic obstructive pulmonary disease
Extracellular signal-regulated kinases
Fractional excretion of nitric oxide
Maximum midexpiratory flow
Forced expiratory volume in one second
Ratio of forced expiratory volume in one second to forced vital capacity
High-density lipoprotein
Interleukin
Janus-activated kinase
Mitogen-activated protein kinase
Third National Health and Nutrition Examination Survey
Leptin receptor
Phosphatidylinositide 3-kinases
Provocative concentration of methacholine causing a 20% fall in FEV1
Platelet-derived growth factor
Peroxisome proliferator-activated receptor
Protein kinase B
Resistin-like molecule
Sphingosine-1-phosphate
Single-nucleotide polymorphism
Signal transducer and activator of transcription
Toll-like receptor
Tumor necrosis factor
Vascular endothelial growth factor.
The authors have no personal or financial support or involvement with organization(s) with financial interest in the subject matter or any other actual or potential conflict of interests.
The authors would like to acknowledge the assistance provided by Mark Schuyler, M.D., at University of New Mexico, in proof reading and critiquing this review. This work was supported by funding from the National Institutes of Health (K23 HL 094531-01 and CTSA 1ULRR031977-01 for Akshay Sood and ES-013307, HL-084044, and ES-000002 for Stephanie A. Shore).