|
|
||||||||
1 Department of Medicine, Meakins-Christie Laboratories, McGill University, Montréal, Québec, Canada
Correspondence and requests for reprints should be addressed to James G. Martin, B.Sc., M.B. B.Ch., M.D., Meakins-Christie Laboratories, Department of Medicine, McGill University, 3626 St. Urbain Street, Montréal, PQ, H2X 2P2 Canada. E-mail: james.martin{at}mcgill.ca
ABSTRACT
The genetic basis for airway smooth muscle properties is poorly explored. Contraction and relaxation are altered in asthmatic airway smooth muscle, but the basis for the alterations and the role that muscle-specific susceptibility genes may play is largely unexplored. Alterations in the β-adrenergic receptor, signaling pathways affecting inositol phosphate metabolism, adenylyl and guanylyl cyclase activity, and contractile proteins such as the myosin heavy chain are all suggested by experimental model systems. Significant changes in proliferative and secretory capacities of asthmatic smooth muscle are also demonstrated, but their genetic basis also requires elucidation. Certain asthma-related genes such as ADAM33, although potentially important for smooth muscle function, have been incompletely explored.
Key Words: airway responsiveness F344 rats A/J mice β-adrenergic receptors myosin isoforms
The genetic basis for asthma susceptibility has received a great deal of attention in recent years. Despite the importance of airway smooth muscle in airway responsiveness and the potential for altered properties of muscle to be at the root of excessive airway narrowing in asthma, there are few studies that have addressed the genetic basis for altered airway smooth muscle properties. The literature on "asthma genes" related to immunity, inflammation, and atopy abounds, and one can only be struck by the lack of mention of genes related to smooth muscle function. In this article we will address the areas that have received attention and the areas that we believe merit particular consideration for future studies. Genes with the potential to affect various aspects of airway smooth muscle function have been summarized in Figure 1.
|
In a recent review of the genetics of asthma (5), the only gene mentioned that might have a bearing on airway smooth muscle function was that of the β adrenergic receptor, which has interesting polymorphisms that significantly affect its function (6). The discovery of an association of ADAM33 with asthma (7) has identified another polymorphic protein of possible relevance to airway smooth muscle function. The logic underlying the major thrust to relate particular asthma phenotypic characteristics, frequently unrelated to airway smooth muscle function, to genetic constitution is understandable because the definition of phenotype is easier. Quantification of immune response or even symptoms is relatively straightforward. Quantification of the contribution of airway smooth muscle properties to asthma is problematic. The best surrogate for the contractile properties of airway smooth muscle in vivo is the responsiveness to inhaled methacholine and other agonists. However, there are several other determinants of airway responsiveness, including the lung elasticity, the coupling of airways to parenchyma and lung and airway architecture (8). The cyclical changes in lung volume and consequently in transpulmonary pressure associated with breathing potently relax pre-contracted airway smooth muscle (9). Lung elastic recoil limits airway smooth muscle shortening and it has been shown to vary among normal healthy subjects (10), and any genetically determined variability may confound the interpretation of airway responsiveness as a measure of airway smooth muscle properties. Despite these limitations, we will assume for the purposes of this review that airway responsiveness contains information reflecting smooth muscle properties (11).
RELATIONSHIP BETWEEN METHACHOLINE RESPONSIVENESS AND GENETICS: HUMAN AND ANIMAL DATA
Airway responsiveness shows a substantial degree of variability in human populations (12) and in animal studies (13–15). Studies in humans and animals have addressed the genetic basis for airway hyperresponsiveness to some extent. Methacholine responsiveness was compared in pairs of monozygotic and dizygotic twins, and the intra-pair difference in responsiveness was no different between the mono-and dizygotic pairs (16). However, other studies have suggested a genetic basis, likely polygenic, for airway responsiveness (17). Clearer evidence of a genetic influence on airway responsiveness and smooth muscle function comes from animal models. One of the earliest observations was made on the Basenji greyhound cross, which led to a series of publications indicating the tendency for this animal to have hyperresponsiveness to a number of contractile agonists (18, 19), and therefore meeting the criterion for so called nonspecific hyperresponsiveness. The Basenji greyhound cross had hyporesponsiveness of the trachealis to direct vagal stimulation, but had a post-synaptic hyperresponsivenss to cholinergic stimulation, leading the investigators to conclude that the airway smooth muscle was altered in its contractile responses (20). However, the trachealis responsiveness may have been specific to muscarinic receptor stimulation.
Strain-related differences in airway responsiveness to methacholine, histamine, or serotonin challenge have been demonstrated in several other species. Rats were among the first animals to be reported to have less intra- than inter-strain variance in airway responsiveness, and have served as a useful model of innate hyperresponsiveness. The F344 strain was consistently the most responsive of the strains screened (15, 21). The methacholine dose–response curve of normal human subjects and of various other species is very sensitive to lung volume (22, 23). An increase in lung volume significantly reduces the responsiveness. Interestingly the F344 rat also is relatively insensitive to the bronchodilator effects of lung volume (24), a feature also observed in studies of asthmatic airway mechanics (25) that likely reflects alterations of airway smooth muscle properties.
Measurements of methacholine and serotonin responses of nine inbred strains of mice have also revealed substantial differences in responses that are significantly strain dependent. The A/J strain was most responsive and the C3H-HeJ the least responsive to the contractile agonists (26). The C57/BL6 strain is also a low responder and has been more extensively studied subsequently. Gavett and Wills-Karp found that enhanced muscarinic receptor signaling was apparently responsible for differences in responsiveness between A/J and C3H mice (27). Other investigators have found that the rank order of responsiveness may vary according to agonist, and there may therefore be genetic differences specific to certain agonists. This form of hyperresponsiveness may or may not be the kind that we wish to examine in relationship to asthma. Initial conclusions that airway responsiveness was inherited as an autosomal recessive trait have given way to the notion of polygenic inheritance. De Sanctis and colleagues found that an F2 generation of animals derived from an F1 generation from A/J and C57 parents showed a normal and unimodal distribution of responsiveness compatible with a complex inheritance (28). Quantitative trait locus analysis identified three chromosomal regions associated with airway responsiveness. However, the potential genes mentioned included only genes related to inflammation and none related to smooth muscle function.
Nonspecific airway hyperresponsiveness in asymptomatic healthy subjects is associated with homozygosity for the Gly16 β2-AR allele, suggesting that polymorphisms of this receptor contribute to altered airway responsiveness in asymptomatic individuals (29). The administration of a β-adrenergic receptor antagonist has long been known to leave airway responsiveness to histamine unaffected in subjects without asthma (30). However, the hyperresponsive Basenji greyhound also shows impaired β–adrenergic relaxation that was attributable to an uncoupling of the β2 receptors and impaired adenylyl cyclase activation (31). Impaired relaxation to β-adrenergic stimulation has been well documented for airway smooth muscle from subjects with asthma (32). An impairment in relaxation can be induced by passive sensitization of airway smooth muscle and is associated with an upregulation of G
i (33), a G protein inhibitory of adenylyl cyclase. Coupling of the β-adrenergic receptor to G
i and hyperresponsiveness also occur (34). G
i is also up-regulated by exposure to cytokines such as IL-13 (reviewed in Reference 35) and may have relevance for the observed increased responsiveness of animals exposed to IL-13 (36). A number of studies have addressed the effects of various β2-adrenergic receptor genotypes on responsiveness and desensitization to β-agonists, airway hyperresponsiveness, and asthma control. The topic has been extensively reviewed recently (37). The β2-adrenoreceptor gene (ADRB2) on chromosome 5q31–32 has been shown to have a number of polymorphisms (38). Among those, Arg 16 Gly, Gln 27 Glu, and Thr 164 Ile are identified as polymorphisms related to the receptor function. These polymorphisms seem to be relevant to the kinetics of airway smooth muscle relaxation, affecting down-regulation or altering ligand affinity for the receptor (39, 40).
Different levels of β2-adrenoreceptor receptor expression independent of desensitization also appear to occur. A polymorphism in a short open-reading frame upstream of the β-adrenergic receptor coding block encodes a peptide that inhibits expression of the receptor at the translational level (41). This polymorphism is common and results in a twofold difference in β-receptor expression in human airway smooth muscle cells. It may therefore alter β-adrenergic receptor responsiveness. Properties of the receptors as characterized in culture systems have not always predicted the behavior of subjects treated with β-agonists (37). In a cohort study of over 8,000 participants, there was a poor association between polymorphisms and asthma severity and prognosis (42). However, the complex nature of asthma outcomes may have confounded the anticipated relationships.
An adenylyl cyclase type 9 polymorphism that results in substitution of Met for Ile at amino acid 772 results in decreased catalytic activity. Corticosteroid treatment increases salbutamol-stimulated adenylyl cyclase responses in the Met772 substitution compared with those expressing Ile772. These in vitro observations have been supported by human studies and are consistent with an effect of this AC9 polymorphism on β2-receptor responsiveness with concomitant inhaled corticosteroid administration (43).
AIRWAY SMOOTH MUSCLE PHENOTYPING
Airway smooth muscle properties as determinants of airway responsiveness can be assessed by a variety of techniques ranging from measuring responses to bronchoconstrictors in cellular models to intact animals. Each technique has its limitations that must be considered in the interpretation of results. The relevance of any set of experimental observations to whole animal physiology increases as the complexity of the model increases, but uncertainty with regard to the contribution of the smooth muscle also increases. The measurement of the isometric, isotonic, and auxotonic contractions of airway tissues using classical organ baths provides useful information, but there is conflicting evidence surrounding force generation by asthmatic muscle from such experiments (44, 45). It has been argued that the dynamics of airway narrowing may be more revealing than quasi-static properties. Dynamic properties of airway smooth muscle have been assessed in cultured lung explants (46, 47) and cultured airway smooth muscle cells (48). Although the study of cultured cells is open to criticism because of the loss of contractile proteins in culture, it has been possible to retain aspects of altered airway smooth muscle function that parallel airway hyprresponsiveness in vivo (49, 50). Indeed, human asthmatic airway smooth muscle cells are more contractile in three-dimensional gels than control cells (51). Single cells isolated from bronchial biopsy specimens from individuals with asthma have also shown a greater responsiveness than control cells (52).
CONTRACTILE SIGNALING AND AIRWAY HYPERRESPONSIVENESS
Rodent models have provided insights into some differences in signaling associated with innate airway hyperresponsiveness. Comparisons of the characteristics of F344 and Lewis rats have demonstrated several sites of differences in specific pathways related to both contraction and proliferation of airway smooth muscle. Fortunately, the hyperresponsiveness identified in vivo can be found in several reductionist models, ranging in complexity of organization from lung explants to single cells. Agarose-embedded lungs, when sliced and placed in culture, can be used to study airway contractions. Airway contractions in lung explants have shown that F344 airways are more responsive to methacholine, serotonin, and bradykinin than are the control Lewis strain airways (49, 53, 54). Thus far contractile signaling in the rat has been explored on single cells in culture. Single-cell calcium imaging also reveals differences in responsiveness of airway smooth muscle cells from the Fisher rat to serotonin and bradykinin compared with the normoresponsive Lewis strain (54). The enhanced calcium signals are attributable to differences in inositol phosphate metabolism (55). Inositol trisphosphate (IP-3), the intracellular mediator of calcium release from the sarcoplasmic reticulum, accumulates in greater concentrations in the cells of F344 rats after serotonin stimulation than in Lewis rats, most probably because of a reduced rate of enzymatically mediated hydrolysis by an IP-3–specific phosphatase rather than differences in the production of IP-3 by phospholipase C (55). For a given dose of methacholine, the level of phosphorylation of the myosin regulatory light chain (LC20) is greater in the F344 than in the Lewis rats (56), consistent with the enhanced agonist-induced calcium signals. While a greater level of LC20 phosphorylation will lead to greater tension development (56), its effects on rate of shortening remain to be clarified. Human asthmatic cells are also more contractile than control cells (51, 52), but the two studies do not agree on the relationship with myosin light chain kinase expression.
The F344 and Lewis rat cells have also been studied in culture using optical magnetic twisting cytometry (50). Using ferromagnetic microbeads attached to the cell surface by RGS peptide linked to integrins, it has been possible to obtain measurements of cell stiffness. Traction microscopy was used to measure contractile forces and the dynamics of cytoskeletal remodeling, an integral feature of contraction of smooth muscle, by measuring spontaneous nanoscale motion of microbeads. In response to a series of contractile agonists, Fisher ASM cells showed greater stiffening, bigger contractile forces, and faster remodeling, as indicated by the microbead movements. These changes are consistent with other characteristics of this hyperresponsive strain. The more rapid cytoskeletal rearrangement is an interesting property of the muscle and it should favor more rapid adjustment to new lengths (50). Such plasticity may be an important property, influencing the regulation of muscle tone. The molecular basis for this plasticity is unclear but is presumably linked to other genetically determined properties reflected in the hyperresponsiveness of the muscle.
These two strains of rat manifest other differences, presumably also genetically determined, that could enhance responsiveness. In addition to the differences in inositol phosphate metabolism, a portion of the difference in responsiveness between F344 and Lewis rats in vivo may be attributable to differences in sensitivity to endogenous nitric oxide production in the airways. Lewis rats in vivo become more responsive after the inhibition of nitric oxide synthase with L-NAME, whereas F344 rats are unaffected by the intervention (57). F344 airways ex vivo are resistant to the effects of relaxant effects of nitroprusside, a source of exogenous NO; tracheal tissues produce less 3,5 cyclic guanosine monophosphate when treated with nitroprusside (58). It seems likely that the epithelium is a major source of NO in the rat. However, the resistance of smooth muscle to its effects is presumably determined by genetic factors; difference in guanylyl cyclase expression or activity could be associated with changes in airway smooth muscle phenotype. Indeed, overexpression of NOS-2 in the airway epithelium of mice makes them hyporesponsive to methacholine (59), confirming the ability of endogenously produced NO to affect airway responsiveness in vivo. However, the extent to which airway smooth muscle may be a source of NO or may vary in its responsiveness to NO, and indeed its genetic basis, requires exploration.
Less information is available on mouse airway smooth muscle. In CD38 knockout mice that lack the enzyme responsible for the synthesis of cyclic ADP-ribose, which regulates Ca2+ signaling in smooth muscle cells, responsiveness to aerosolized methacholine challenge is attenuated. Intracellular Ca2+ response to agonist in smooth muscle cells obtained from CD38 knockout mice are significantly lower than those from wild-type mice (60). Interestingly, this enzyme has been linked to IL-13–induced airway hyperresponsiveness, providing a possible mechanism by which sensitivity of airway smooth muscle to Th2 inflammation could lead to more or less responsiveness in certain individuals (61). Whether there are polymorphisms affecting CD38 does not appear to have been reported. T-bet, a T helper type 1 cell–specific transcription factor has been linked to airway hyperresponsiveness in mice and to asthma in humans (62, 63). Part of the hyperresponsiveness is IL-13 dependent (64), but there is also an alteration in contractility and Ca2+ signaling of airway smooth muscle cells. There is a higher frequency of intracellular Ca2+ oscillations and greater intracellular Ca2+ stores in airway smooth muscle in cultured lung explants (65).
CONTRACTILE PROTEIN EXPRESSION AND FUNCTION
One could speculate that the expression of contractile proteins should be among the genetically determined factors affecting airway smooth muscle responsiveness. The expression and function of contractile proteins in asthmatic airways have begun to be examined. Airway smooth muscle from innately hyperresponsive subjects before the development of overt asthma is required to address the issue. Thus far the expression of contractile proteins has been examined in the F344 and Lewis rats. In particular, the expression of the various smooth muscle myosin heavy chain (SMMHC) isoforms has been investigated because of the potential for these isoforms to influence contractility of airway smooth muscle. There are four isoforms of the SMMHC that are generated by alternative splicing of a single gene. Two of these isoforms differ by the presence ([+]insert) or absence ([–]insert) of a 7 amino acid insert in the amino terminus region. The rate of shortening of the isoform with the 7 amino acid insert, the [+]insert isoform, propels actin twice as fast as the [–]insert isoform (66, 67). A greater proportion of the faster (+)insert SMMHC isoform is expressed in the tracheal tissues of the hyperresponsive F344 compared with the normoresponsive Lewis rat, whereas other contractile proteins such as h-caldesmon,
-actin, or myosin light chain-17 (LC17) isoforms do not differ (56).
The fast [+]insert SMMHC isoform has greater expression in rapidly contracting phasic muscle such as the intestine, and lower expression in slowly contracting tonic muscle such as the aorta (68). The greater expression of the [+]insert isoform in the Fisher rat trachea is consistent with the more rapid rate of shortening of airways to stimulation with contractile agonists in lung explants (53) and in muscle strips (69). This is in agreement with the fact that the [+] and [–]insert myosin isoforms do not exhibit any difference in their molecular force production. A knockout mouse that lacks the [+]insert SMMHC isoform has also been engineered in a C57BL/6 background (70) and it shows a longer time to peak airway narrowing after a challenge with methacholine but is not hyperresponsive (71). Unfortunately, the wild type-mice are relatively unresponsive to methacholine, and the inserted isoform might have greater importance in a background strain that is more responsive.
SECRETORY AND PROLIFERATIVE CAPACITY OF AIRWAY SMOOTH MUSCLE
There is increasing recognition of the potential importance of the secretory and proliferative capacity of airway smooth muscle for the eventual development of disease. Airway smooth muscle is a source of proinflammatory cytokines and chemokines. If muscle over-produced such molecules one could imagine a predisposition for disease. The examination of cellular phenotype in culture has substantial promise. The finding that fibroblasts from individuals with asthma retain characteristics in culture that distinguish them from fibroblasts from control subjects suggests that genetically determined differences are present, and are not eliminated by culture conditions (72). Studying airway smooth muscle in the same way has also revealed differences (73, 74). Faffe and colleagues examined the importance of single-nucleotide polymorphisms in the IL-4R
and vascular endothelial growth factor (VEGF) genes in an attempt to explain observed variance in the release of VEGF by ASM cells in culture (75). The results were stratified by IL-4R
and VEGF genotype and demonstrated a substantial decrease in IL-13– or IL-4–induced VEGF release in cells bearing the –460C/–152A/+405G VEGF haplotype. Such a finding is of substantial interest, as it indicates potential mechanisms by which the secretory function of airway smooth muscle may be conditioned by the cell phenotype and may help to explain variations in susceptibility to airway smooth muscle remodeling in asthma. The exploration of secretory mechanisms in isolated airway smooth muscle and the genetic mechanisms underlying them may well reveal other susceptibilities of interest.
Airway smooth muscle from F344 rats is not only more responsive to contractile stimuli, but shows a hyperresponsiveness to growth factors (76). The exploration of sensitivity to growth factors was motivated by the finding of an excess of airway smooth muscle in the trachea and lower airways of the F344 rat (77). The sensitivity to growth-promoting factors in vascular smooth muscle from spontaneously hypertensive rats has been reported (78). Indeed, vascular muscle from these rats shows hyperresponsiveness to both contractile and proliferative stimuli. The mechanisms for hyperresponsiveness to growth factors and the links to contractile hyperresponsiveness have not been unraveled. They appear to be protein kinase C dependent (79). Human asthmatic airway smooth muscle in culture also shows enhanced proliferative responses to mitogens (74) that may be attributable to the absence of an inhibitory transcription factor C/EBP-
, the influence of an altered extracellular matrix, and/or a reduction in the synthesis of factors such as PGE (reviewed in Reference 80). Such a phenomenon could have pertinence for susceptibility to airway smooth muscle remodeling, but the genetic basis for the altered properties of asthmatic airway smooth muscle needs to be elucidated. The selective growth of more proliferative cells during airway remodeling in asthma is an alternative explanation for the observed differences in properties.
OTHER GENETIC DETERMINANTS OF AIRWAY RESPONSIVENESS OF UNCERTAIN RELATIONSHIP TO AIRWAY SMOOTH MUSCLE FUNCTION
The isoforms of nitric oxide synthase (NOS) have been linked to asthma in some studies. Corresponding mouse phenotypes have been also studied (81). NOS2 knockout mice have increased AHR to intravenous methacholine challenge associated with decreased nitrate. The effects could be mediated indirectly and a variety of immune functions are altered (82). De Sanctis and coworkers have reported significant participation of neuronal NOS (nNOS, NOS1) in NO synthesis modulating airway hyperresponsiveness but inflammation using nNOS (NOS1) knockout mice, iNOS (NOS2) knockout mice and endothelial NOS (eNOS, NOS3) knockout mice, nNOS and eNOS double knockout mice, supporting other evidence of linkage of asthma with the region harboring the nNOS gene in humans (81). nNOS is described to be present in ovine airway smooth muscle (83) and in human lungs (84), but whether the increase in airway responsiveness in in vivo models is directly attributable to altered NOS gene expression by airway smooth muscle is unknown.
One of the asthma susceptibility genes, GPRA (GPR154), which is a member of G protein–coupled receptor family gene that has several splice variants, is expressed in human bronchial smooth muscle cells. Two main transcripts GPRA-A and -B are translocated to plasma membrane and are speculated to be associated with growth inhibition (85, 86). The AA homozygote of rs324981 appeared to be protective against airway hyperresponsiveness in a Chinese population (87).
Single-nucleotide polymorphisms in the estrogen receptor
gene are not associated with susceptibility to asthma and with FEV1 percent predicted in asthma. However, there is a significant association with the susceptibility to AHR, particularly in female subjects, that is strongest in the IVS1–397 C/T polymorphism, with the T allele the risk allele (88). Furthermore, there is an association with more rapid decline in lung function in asthma. Estrogen receptor-
–knockout mice have airway hyperresponsiveness to inhaled methacholine in the absence of airway inflammation (89). There is evidence of down-regulation of M2 muscarinic receptor expression and function in airway smooth muscle cells, but how this causes airway hyperresponsiveness is not clear.
ADAM33 is a candidate asthma-susceptibility gene (7). It is in the metalloproteinase family and it is expressed in human airway smooth muscle cells (90). Since ADAM33 is a cell surface protein mediating adhesion and cell signaling, its relevance may be greater for remodeling than for contractile function. Gene-deleted mice did not show any alteration in asthma phenotype, including airway hyperresponsiveness, after allergen sensitization and challenge. There was also no difference in baseline responsiveness to acetylcholine challenge. ADAM33 does not contribute to AHR and mucosal remodeling induced by OVA sensitization and challenge, as evidenced by the lack of alteration in the phenotype in ADAM33 knockout mice (91). As this study did not evaluate effects on airway smooth muscle remodeling, it is still possible that ADAM33 may be involved in allergen-driven remodeling of smooth muscle, and may require a more chronic model for its demonstration.
The cystic fibrosis conductance regulator (CFTR) gene has been linked to asthma (92). Although the focus has been on its role in epithelial function, it is expressed in other tissues, including airway smooth muscle (93). It has been shown to be involved in smooth muscle relaxation, and its deficiency may lead to impaired relaxation of the muscle.
CONCLUSIONS
To date, the genetic factors leading to alterations in airway smooth muscle properties are poorly explored. Several models of innate airway hyperresponsiveness have been studied and the molecular basis for potentially important alterations resulting from gene polymorphisms has been established. However, there is a lack of definitive studies directly implicating altered expression levels or function of key proteins in airway smooth muscle responsiveness. The mechanisms proposed from genetic studies have generally revolved around proinflammatory genes despite the obvious role of airway smooth muscle in responsiveness. The most extensively studied model from the standpoint of the biochemical pathways accounting for contractile signaling is the F344 rat. In this strain the multiplicity of alterations in airway smooth muscle properties suggests a global alteration in phenotype. The regulation of the phenotype is as yet unexplained. The secretory properties of airway smooth muscle are of increasing interest in determining the inflammatory milieu and the matrix composition of the airway wall. Other properties, such as growth responsiveness or capacity for migration, may also be programmed, and as such could be forms of susceptibility of pertinence for disease.
FOOTNOTES
Supported by the Canadian Institutes of Health Research grant number MOP 36334.
Conflict of Interest Statement: Neither author has a financial relationship with a commercial entity that has an interest in the subject of this manuscript.
(Received in original form June 11, 2007; accepted in final form July 10, 2007)
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |