Asthma and chronic obstructive pulmonary disease (COPD) comprise chronic inflammatory disorders characterized by airway hyperresponsiveness and airflow obstruction that can fluctuate over time. They have an increasing economic burden, not to mention their associated disabilities and fatal outcomes. Despite of outnumbered research, we still have not completely understood the big picture of their natural history. Quite a lot of evidence arising from epidemiological, clinical, pathological, and molecular studies, have only provided a short view of the pathobiological mechanisms generating those diseases. The development of better functional assessment techniques along with a wider availability of new biomarkers allowed to recognize that inflammatory airway diseases, especially asthma, involve multiple subphenotypes that differ in clinical severity, histopathology, response to therapy, and long-term outcome. In consequence, heterogeneous groups have been identified, which are likely originated from a unique genetic background/ environment combination.

Severe phenotypes of airway diseases that course with airway hyperreactivity, implicate airflow obstruction that is either irreversible or only partially reversible. Their severe nature apparently is due to longstanding inflammation in the airways. In this setting, injury cyclicity, age, genetic factors, and previous tissue history induce structural changes; a phenomenon commonly coined as ‘airway remodeling’. Airway remodeling is assumed to result in severe phenotypes. However, several clinical and animal studies indicate that the relationship between inflammation, remodeling, and hyperresponsiveness is complex, and still not completely understood. Considering that airway hyperreactivity results in an abnormal airway tone, smooth muscle thickening is thought as the main substrate of abnormal airway mechanics. This has been deeply explored; hence, considerable data is available. In this review, we gathered the last experimental findings in this field to formulate a model of airway smooth muscle (ASM) remodeling that fits into the natural history of common airway diseases, mainly focusing on molecular mechanisms.

I. Overview of airway smooth muscle remodeling

Growing evidence supports various pathophysiological mechanisms of airway diseases, including structural changes seen on severe asthma, COPD, chronic bronchitis, cystic fibrosis, and bronchiectasis^(1-4). Thus, airway remodeling has been defined by modifications in the composition, amount, and organization of local cells, including epithelium, glands, blood vessels, extracellular matrix (ECM), and smooth muscle (see Fig. 1). 

Figure 1. Histopathology of small airways in OVA-sensitized Rats. A bronchiole from OVA-sensitized rat(A) in comparison to a normal bronchiole from saline-nebulized rat(B). Tissue shows remodeling features, such as: epithelial hyperplasia (arrowhead), ASM thickening (asterisk), lymphocytic/eosinophilic inflammation (arrows), and luminal exudate (circle), compare with the normal airway. Lung samples were extracted from rats after a protocol previously described(66). Magnification 200X, Hematoxylin-eosin staining.

Interaction of genetic and environmental factors evolve into assorted outcomes after injury. Many models⁽⁵⁾ and clinical studies have shown that symptoms and functional findings are caused by three interconnected factors: 1) chronic inflammation, 2) airway hyperresponsiveness (AHR), and 3) tissue remodeling. Notwithstanding, a major concern in this field is that there is no clear chronologic and quantitative relationships. A current perspective considers that unbalanced immune responses to external factors, such as allergens in asthma, sets up a harmful microenvironment of cyclic injury and repair, which leads to abnormal structure and function^{(6, 7)}. Furthermore, a recent study suggests that chronic mechanical stress resulting from bronchoconstriction per se may also lead to remodeling without inflammation⁽⁸⁾. Despite of airway remodeling could be disorder-specific, the airway structure may play a common role to airway narrowing and airflow limitation carrying out poorly reversible airway obstruction.

Evidence of ASM thickening in Asthma and COPD

Asthma and COPD are well-differentiated clinical entities with some overlapping syndromes in the middle^{(3, 9)}. They share some features, especially at pathological level, including: epithelial hyperplasia and dysfunction, subepithelial fibrosis, increased myofibroblasts, increased vascularization, abnormal neurite branching, and dense ASM layers^{(2, 7, 10)}; highlighting that a greater basement membrane thickness as well as smooth muscle hyperplasia are commonest seen on asthma^{(10, 11)}. Their deepness frequently wedges with clinical expression and severity^{(12, 13)}. A study that evaluated bronchial wall thickness by high resolution computed tomography in mild-to-moderate asthma and COPD revealed the airway diameter and thickness were similar⁽¹⁴⁾, but asthma still has received more attention respect to ASM remodeling. Increased ASM mass could be attributable to hyperplasia, hypertrophy or both. Under physiological conditions, ASM located in the central and peripheral airways are bands that wrap up around the airways in a helical pattern. Its thickness, relative to the diameter of the airway lumen, increases towards the periphery, but in absolute terms, the amount is less in the peripheral airways⁽¹⁵⁾. A morphometric study indicates that the bronchial smooth muscle mass of patients suffering of fatal asthma was twice than non-asthmatics⁽¹⁶⁾. A major concern of this proposal is the ECM volume was not measured. To solve it, a recent study showed ASM hypertrophy in the large airways in both nonfatal and fatal asthma, but hyperplasia was only seen in large and small airways in fatal cases. Both groups were associated with an absolute increase in ECM⁽¹⁷⁾. Some degree of airway wall thickening was regularly detected in asthmatics of all severities with predominance in severe cases⁽¹⁸⁾. The occurrence of remodeling does not seem to depend on the inflammatory response subtype; since, airway structure does not differ between asthmatics with eosinophilia and those without⁽¹⁹⁾. Contrasting results debate the importance of ASM remodeling because it is not always found in asthma, hence, no differences in averageof smooth musclecell cross-sectional area⁽²⁰⁾.

There is less evidence supporting ASM thickening in COPD than asthma. Obliteration and fibrosis of the alveolar wall, mucous gland hypertrophy, and goblet cell hyperplasia are well-known pathological features of COPD⁽²¹⁾. Increased ASM thickness has been found as compared to control, but lesser than asthmatics⁽¹⁵⁾. Functional implications have been shown, as it correlates with the airway obstruction degree⁽²²⁾. Additionally, ASM mass and adventitia increased together by 50% in severe COPD affecting the small airway physiology⁽²³⁾. Biopsy studies from large airways reported no increase in ASM; moreover, smooth muscle protein isoforms were not increased, but there was a slight increment in myosin light chain kinase (MLCK) without changing the myosin light chain phosphorylation⁽²⁴⁾. Conflicting results showed that remodeling may occur in the central airways by greater ECM protein deposition and increased ASM⁽²⁵⁾.

This data points out that asthma and COPD could progress with variable degree of ASM remodeling, but no direct evidence has been obtained supporting reversibility. A murine model of asthma suggests that after allergen cessation, the goblet hyperplasia and collagen deposition resolved first and then lymphocytic infiltration along with ASM thickening⁽²⁶⁾. This brings up an open question whether in human diseases a complete removal of the tissue hazard can be accompanied by spontaneous resolution of airway remodeling.

Inflammatory Microenvironment Orchestrates ASM Remodeling

Airway remodeling is associated with longstanding inflammation. Interleukin (IL)-1β, IL-6, and Tumor Necrosis Factor-α (TNF-α) as well as growth factors such as Platelet-Derived Growth Factor (PDGF), Epidermal Growth Factor (EGF), Insulin-like Growth Factor (IGF), and Transforming Growth Factor-β (TGF-β), have pleiotropic effects; however, specific immune responses portray distinctive pathologic features^{(7, 27)}. T cell reactions cover a wide spectrum of divergent cytokine networks. In asthma, for example, clinical phenotypes match inflammatory profiles: I-type hypersensitivity reaction (IgE-dependent), Th2 predominant inflammation, and non-Th2 associated response⁽²⁸⁾. However, a common feature seen in all cases is the ASM thickening (see Fig. 2).

Figure 2. Crosstalk between ASM with cellular and non-cellular components of the airways during inflammation. (See the text for explanation).

Eosinophils are the most prominent inflammatory cells in the airways of asthmatics⁽²¹⁾. In the course of hypersensitivity reactions, eosinophils localize in close relation to ASM. For instance, small airways contain eosinophils in their outer portion (between ASM and alveolar attachment), whereas in the large airways they are present predominantly in their inner portion (between ASM and basement membrane)⁽²⁹⁾. This eosinophil-ASM relationship can enhance cell proliferation by cysteinyl leukotriene secretion⁽³⁰⁾. Eosinophil homing in the airways depends on Th2 cytokines and eotaxin. Adhesion to ASM is mediated by cell adhesion molecules (ICAM-1, VCAM-1) that are constitutively expressed and also upregulated. ASM cells (ASMC)-derived cytokines could promote eosinophil differentiation, perpetuating the burden of eosinophils into ASM bundles⁽³¹⁾.

In a similar way, mastocyte infiltration is prominent in ASM bundles⁽³²⁾. Mastocyte migration and adhesion potentiate tissue remodeling because histamine, tryptase, activin A, sphingosine 1-phosphate (S1P), β- hexosaminidase, and TNF-α, stimulate many ASMC functions⁽³³⁾. Despite these mediators can stimulate cell proliferation, mast cell seems not to be relevant for neither proliferation or survival⁽³⁴⁾. On the other hand, mastocyte can induce the thymic stromal lymphopoietin (TSLP) in ASM that is highly determinant of Th2 polarization⁽³⁵⁾. Mast cell placement, proliferation and survival into the ASM could occur through allergen-independent mechanisms⁽³⁶⁾.

Neutrophilic inflammation also occurs in severe asthma and COPD. Even though, clinical phenotypes that course with predominant eosinophilic inflammation lead to ASM thickening in both large and small airways, the neutrophil infiltration is almost restricted to concurrent small airway remodeling⁽³⁷⁾. A histopathological study of children with fatal untreated respiratory syncytial virus (RSV) infection showed vascular leakage and neutrophil recruitment into the submuscular layer, smooth muscle, and airway epithelium, resembling fatal cases of obstructive diseases⁽³⁸⁾. IL-8 has been implicated, and it seems to be secreted by Th17 cells⁽³⁹⁾. Nowadays, it is known that Th9, Th22, and Th25 cells also modulate countless aspects of airway immunity⁽⁴⁰⁾. Nonetheless, the most significant cytokine expression in asthma includes IL-4, IL-9, IL-13, eotaxin and RANTES. This profile correspond to an upregulated Th2 reaction⁽⁴¹⁾. CD4⁺ T cells transfer from ovalbumin (OVA)-sensitized rats to non-sensitized rats (adoptive lymphocyte transfer), showed that an specific subset of Th2 cells drove airway remodeling in non-sensitized rats after few OVA challenges⁽⁴²⁾. However, evidence from animal models and humans indicate that Th2 hypothesis is an incomplete explanation for asthma pathogenesis, as allergic and nonallergic types are pathologically indistinguishable⁽⁴³⁾. It has also been reported that airway epithelial cells in asthmatics upregulate the EGF receptor (EGFR) expression, a receptor tyrosine kinase (RTK), even in absence of significant eosinophilic inflammation⁽⁴⁴⁾. A recent study demonstrated subepithelial fibrosis in severe asthmatics without evidence of Th2 inflammation⁽⁴⁵⁾. Depletion of CD4⁺ cells, previous to chronic OVA challenge, significantly reduced peribronchial inflammation but did not completely reverse ASM thickening⁽⁴⁶⁾. Although Th2 cytokines have pro-remodeling actions in vitro, controversial results have been found as IL-5 and IL-13 do not increase ASMC proliferation, but they induce phenotypic switching^{(47, 48)}; and IL-4 inhibits ASMC replication⁽⁴⁹⁾. These studies suggest that remodeling can also occur independently of Th2 inflammation and other factors are needed.

COPD is also accompanied by airway inflammation that is different from asthma, but ASM remodeling still occur⁽²¹⁾. Chronic inflammation induced by chronic cigarette smoking consists of neutrophil, macrophage, B cell, and CD8⁺ T cell recruitment, and it worsens as disease severity increases⁽⁵⁰⁾. T‐lymphocytes and macrophages are the predominant cells, being CD8⁺ T‐lymphocyte infiltration the most remarkable feature in both large and small airways, and there is also absence of significant eosinophilic inflammation⁽⁵¹⁾. Although, Foxp3⁺ regulatory T cells play a role in fibrogenesis, there is no predominant T CD4⁺ subset. This supports the concept that cyclic events of cytokine and growth factor surges could be the main drivers regardless of the etiology and immune polarization.

An intricate network underlies the ASM and its surroundings, not only immune cells but also neural parasympathetic endings, mesenchymal cells, ECM, and epithelium (see Fig.2). For example, ASM activation by proinflammatory cytokines and substance P can induce the brain-derived neurotrophic factor (BDNF) expression for spatial coordination of neuronal branching⁽⁵²⁾, and vascular endothelial growth factor (VEGF) for control of angiogenesis to assure adequate perfusion, the latter have an important repercussion on vascular leakage and vasogenic edema during fatal asthma⁽⁵³⁾. Neuronal development also would coordinate spatial distribution of ASM, because substance P induces both migration and proliferation⁽⁵⁴⁾. Nevertheless, the airway epithelium could have a greater contribution due to its plasticity and inflammatory properties. Dysfunctional epithelial cells release growth factors, as well as, acetylcholine (ACh) and leukotrienes that could contribute to ASM growth, ECM deposition, and angiogenesis⁽⁵⁵⁾. Moreover, the epithelium is an important source of nitric oxide (NO) in the airways, which has relaxing and other anti-remodeling effects. Physiological NO is produced by constitutively expressed neuronal and endothelial NO synthase (n-,e-NOS)⁽⁵⁶⁾. However, cytokines increase inducible NOS (iNOS) and arginase expression. A greater iNOS/ arginase activity decreases L-arginine bioavailability, which generates an uncoupled iNOS that not only synthases NO, but it also produces superoxide and peroxynitrite. These molecules are capable of causing cellular toxicity and promoting AHR⁽⁵⁷⁾. Functional consequences of increased arginase are reinforced by L-arginine transport blockage with eosinophil-derived polycations. L-Ornithine, a product of urea cycle, is a precursor of polyamines and L-proline, both involved in cell proliferation, collagen synthesis and chromatin remodeling⁽⁵⁸⁾. This exemplifies how noncontractile ASM functions are modulated by many conditions; therefore, the commonest experimental approaches based on univariate analysis can under- or overestimate their contribution on smooth muscle processes.

ASMC are multifunctional

The relevance of ASMCs in pulmonary diseases has been recognized since the last century. The consensus until a few years ago was to consider them just as effectors. However, far from their abilities to contract and relax, ASMCs proliferate, migrate, secrete chemokines/cytokines, and express surface receptors for cell adhesion and leukocyte activation, having a crucial role in airway dysfunction⁽⁵⁹⁾. A concept of plasticity emerged when those functions were associated with specific circumstances and required wide adjustment in gene expression^{(60, 61)}. ASM hypertrophy and/or hyperplasia involve not only outer cell influences, but also ASMC reactions with paracrine/autocrine properties^{(7, 62)}. Quite a lot of molecules could coordinate this loop, such as: growth factors, cytokines, chemokines, ECM molecules, G protein-coupled receptor (GPCR) agonists, natriuretic peptides (NPs), NO, and others^(63-66).

Several in vitro synthetic functions have been shown. Also, ASM in mild asthmatics has constitutive staining for RANTES⁽⁶⁷⁾. Further cytokines secreted by ASM include IL-1β and IL-6 family cytokines, such as leukemia inhibitory factor (LIF) and IL-11⁽⁶⁸⁾. These have deeper effects on recruitment, proliferation, and differentiation of eosinophils, mastocytes, T cells, and B cells, establishing a bidirectional regulatory network. Mainly, a CD4⁺ T cell- myocyte crosstalk through direct contact has shown to be determinant of ASM remodeling⁽⁴²⁾. Airway homing of T cells is CCL5 or RANTES-guided, which is released by ASMCs. Likewise, strong adhesion between these two cell types has also been described⁽⁶⁹⁾. Remarkably, even though ASMCs are not usually thought as antigen-presenting cells, evidence supports the expression of major histocompatibility complex class (MHC) II molecules making them capable of antigen presentation. Moreover, ASMCs express the cell adhesion molecules (CAMs)/costimulatory molecules, CD40, CD40L, CD80, CD86, ICAM-1 (CD54), VCAM-1 and LFA-1 (CD11a/CD18)⁽⁷⁰⁾. The CD44-dependent T cell adhesion to ASMCs is not only significant to exchange inflammatory signals, but also to induce ASM hyperplasia through RTK activation⁽⁷¹⁾. This cooperative signaling mediates proasthmatic-like changes in ASM responsiveness, and denotes a potential mechanism of remodeling.

In the airway, a net of collagenous and noncollagenous proteins influences cellular behaviors. ECM components include collagens, fibronectin, members of the matrix metalloproteinase (MMP) family, as well as their inhibitors (TIMP)⁽⁵⁹⁾. After serum stimulation, ASMCs were found to generate elastin, laminin-β1,-2, and -γ1, thrombospondin, collagen-I-V, and decorin⁽⁷²⁾. In addition to promoting ECM deposition, ASMCs are capable to affect its degradation. Human ASMCs release progelatinase A (MMP-2 precursor) and, after TNF-α stimulation, gelatinase B (MMP-9)⁽⁷³⁾. MMP production suggests that ASM contributes to ECM turnover, and subsequently the airway remodeling, because inhibition of the autocrine-derived MMP-2 has antiproliferative effects on ASMC culture⁽⁷⁴⁾. Therefore, ECM degradation could be essential for ASM phenotypic modulation, being degradation of the pericellular collagen fibrils a requirement to allow cell division⁽⁷⁵⁾. Serum levels of TIMP-1 and MMP-9 are raised in both asthma and COPD, supporting a straight relationship between clinical expression and tissue remodeling. The MMP-9/TIMP-1 ratio and periostin levels could be consider biomarkers of active disease⁽⁷⁶⁾. Cyclic inflammation/ repair simultaneously occur to cyclic ECM degradation and deposition, which could be a critical phase in ASM thickening.

Crosstalk between ASM Remodeling and Hyperresponsiveness

Airway narrowing and abnormal muscle relaxation are the hallmarks of asthma, COPD, and bronchitis. Multiple mechanisms have been proposed to explain the AHR, like increased vagal tone, cytokine-potentiated increment of free intracellular calcium, increased MLCK activity, and activation of the procontractile Rho kinase pathway⁽⁷⁷⁾. All of them have in common that could hasten the shortening velocity. Therefore, even though remodeling can be triggered by hypersensitivity reactions, infections, environmental pollutants, and developmental abnormalities, AHR could be just generated by unbalanced responses to contractile vs relaxing factors⁽⁷⁸⁾. The role of ASM remodeling as a substrate of AHR was uncertain because functional abnormalities can be seen without changes in the bronchial smooth muscle mass⁽⁷⁹⁾. However, increasingly data supports a role in severe AHR phenotypes, and irreversible or partially reversible airflow obstruction⁽⁴³⁾. The structure determines both passive tone and active responses to agonist stimulation. ASM remodeling involves phenotypic changes that enhance its thickening, and during this process a decline in force induced by repetitive length changes is seen, but then it rapidly adapts and recovers its ability to generate force. In this way, higher passive stiffness could contribute to increased AHR by attenuating the extent of ASM length fluctuations during tidal breathing, i.e., ASMCs adapt by assuming a shorter resting length while retaining its ability to generate force⁽²⁷⁾. For that reason, after induced bronchoconstriction, deep inspiration causes airways of asthmatic individuals to dilate transiently.

Expression of immunomodulatory molecules by ASMCs can delay inflammation resolution and lead to aberrant healing, which is a potential mechanism of AHR⁽⁷⁸⁾. The change in the ASMC population compromises an increase of synthetic properties, which can modulate the contractile mass. Particularly, if it is considered that the whole ASM is coupled by gap junctions, and the calcium dynamic differs between ASMC phenotypes. Propagation of wave-like calcium currents from modulated ASMC to contractile ASM would hypothetically affect not only contractile functions but also noncontractile activities, as discussed in following sections. Other noncontractile elements, including excessive ECM content, may lead to nonreversible airway obstruction by reducing airway distensibility⁽⁸⁰⁾. Whether increased ASM supports abnormal reactions to agonists or makes the airway stiffer, it definitely provides an exceptional substrate for AHR in a framework of progression and severity, at least for asthma. In fatal asthma, airway wall thickness is increased around 50-230%, while in nonfatal asthma it ranges from 25 to 150%, most studies pointing out hyperplasia over hypertrophy as the predominant mechanism^{(11-13, 43)}.