INTRODUCTION

TOP ABSTRACT INTRODUCTION STRUCTURAL ALTERATIONS IN AIRWAY FUNCTIONAL ALTERATIONS IN AIRWAY FUNCTIONAL ALTERATIONS IN AIRWAY REFERENCES

Considering that an abnormality of airway smooth muscle is thought to be the basis of the airway hyperresponsiveness that characterizes asthma, there are surprisingly few data describing its structure and function in human asthma. Here we intend to summarize the structural and functional alterations of airway smooth muscle that have been identified in asthmatic subjects. We will initially review the descriptive and quantitative studies that describe the amount and distribution of airway smooth muscle in asthma, then summarize the limited number of studies in which the function of asthmatic airway smooth muscle has been examined in vitro, and finally examine some of the data from animal models of "asthma." Space does not permit a thorough review of the studies that suggest abnormal in vivo airway smooth muscle behavior; however, we will touch briefly on the altered airway response to deep inspiration in asthma. This diminished response to applied stress may be related to altered smooth muscle function and could be a fundamental contributor to increased airway narrowing.

	STRUCTURAL ALTERATIONS IN AIRWAY SMOOTH MUSCLE IN ASTHMA

TOP ABSTRACT INTRODUCTION STRUCTURAL ALTERATIONS IN AIRWAY FUNCTIONAL ALTERATIONS IN AIRWAY FUNCTIONAL ALTERATIONS IN AIRWAY REFERENCES

Smooth Muscle Layer Thickness in Asthmatic Airways

The first systematic study of the amount of airway smooth muscle in asthmatic subjects was conducted by Huber and Koessler (1), two pathologists working in Chicago in 1922. These investigators examined the airways of five subjects who died of asthma and compared them to the airways of four subjects who died suddenly of nonpulmonary conditions. The investigators recognized that airway wall and smooth muscle layer thickness would appear to be increased in asthmatic subjects if the lumenal diameter were to be used as a marker of the airway size. This is because the muscle would be shortened and thickened and the lumen would be narrowed. They therefore used the airway external diameter as their "yardstick." Figure 1 shows their plot of airway smooth muscle thickness versus airway external diameter; clearly the asthmatic subjects have a marked increase in the amount of muscle in all sizes of airways. In this very sophisticated study, the investigators recognized all the potential causes of excessive airway narrowing in asthma, and a statement from their introduction would surely not be out of place even today:

View larger version (39K): [in this window] [in a new window]   Figure 1.   Smooth muscle layer thickness in subjects with asthma (n = 5) and subjects without asthma (n = 4). (Adapted from Figure 17 of Reference 1 by permission.)

The theories that have been formulated to explain the attacks of paroxysmal dyspnea are agreed today that the difficulty of respiration is due to a stenosis of the bronchi, but whether this narrowing is due chiefly or exclusively to a spasm of the bronchial smooth muscle system, to swelling and exudation of the bronchial mucosa, to a true obturation stenosis by the secretion from the bronchial glands, or to a combination of two or more of these conditions seems still to be an unsettled question (1).

Since that time there have been a number of studies in which it has been shown that the airway smooth muscle layer is markedly thickened in patients who die of asthma (2, 3). However, in most of these studies the artifact related to airway narrowing causing apparent thickening was not considered. Even the use of airway external diameter, as employed by Huber and Koessler, is not adequate to control for this error. This is because during airway narrowing, external diameter will decrease, albeit slightly less than internal diameter. Thus, in these studies the degree of thickening could have been overestimated since the airways of asthmatic subjects are often contracted and narrowed when examined at the time of postmortem. The demonstration by James and colleagues (4, 5) that the airway basement membrane perimeter is relatively constant after smooth muscle contraction or changes in lung volume has allowed a number of investigators to examine the relationship between airway smooth muscle area and a more robust estimate of airway size. The slope and intercept of this relationship can be constructed and the pooled data from different groups of subjects can be compared using valid techniques for combining data (6). Table 1 shows a summary of these data from subjects with fatal asthma, patients with chronic obstructive pulmonary disease (COPD), and control subjects (7), which confirms that patients who have fatal asthma show a marked increase in smooth muscle layer thickness.

There are fewer data on patients who have had asthma but who died for other reasons or had a lobectomy. Although some of these data suggest that the amount of airway smooth muscle in these subjects is intermediate between the subjects with fatal asthma and the control or normal subjects (9), others have shown no significant increase in airway smooth muscle amount, at least in some categories of airway size (13). A recent study reported by Thomson and coworkers (14) provides additional support for the observation that airway smooth muscle mass is not invariably increased in patients with asthma. They also suggested that the increase in airway smooth muscle area that has been reported by others could have been overestimated. These investigators measured the airway smooth muscle area in the large central airways of five asthmatic subjects whose airways were obtained at the time of surgical resection of the lung or at autopsy, and showed no significant difference compared with a matched control group. They used 1.5-µm sections of plastic-embedded tissue, sectioned the muscle perpendicular to the long axis of the smooth muscle bundles, and discriminated between smooth muscle cells and their surrounding matrix. They reasoned that the plane of section usually used (parallel to the muscle bundles), combined with the use of thick sections and a failure to distinguish between smooth muscle cells and their associated extracellular matrix, could explain an overestimation of smooth muscle area in other studies. Although they studied only large central cartilaginous airways, and most reports of an increase in smooth muscle area in asthma have involved examination of peripheral airways, their results are particularly significant because they measured increased forced generation by these same smooth muscle preparations in comparison to airway preparations of similar sizes from control subjects.

There is evidence that in subjects with increased muscle mass, the increase is due both to hypertrophy of existing airway smooth muscle cells, as well as hyperplasia. Ebina and associates (10, 15) have reported two patterns of airway smooth muscle hypertrophy and hyperplasia. In subjects with type 1 asthma, airway smooth muscle mass was increased only in central bronchi, where hyperplasia predominated. In subjects with type 2 asthma, there was increased muscle throughout the tracheobronchial tree, and the increased muscle was characterized by hyperplasia as well as hypertrophy, especially in peripheral airways.

Possible Consequences of Smooth Muscle Layer Thickening in the Airways

An increase in airway smooth muscle mass in asthma could have a simple geometric effect to increase airway narrowing, much like the effect of thickening of the submucosal region of the airway wall, and can narrow the airways as well as amplify the effect of smooth muscle shortening (6). However, an increase in smooth muscle mass, if associated with a parallel and concomitant increase in force-generating ability of the muscle, will have the additional effect of allowing the airway smooth muscle to shorten excessively against the elastic loads provided by the lung parenchyma, parallel elastic elements, and mucosal folding.

Although one might expect that an increase in airway smooth muscle mass would be accompanied by an increase in force generation, this is not necessarily the case. Pulmonary vascular smooth muscle proliferation induced in rabbits by hyperoxia produces an increase in smooth muscle mass but a decrease in the ability to generate maximal stress (16). In vitro, when airway smooth muscle is stimulated to proliferate, the muscle differentiates from a contractile to a more synthetic phenotype, concomitant with decreased expression of -smooth muscle actin and increased -actin and nonmuscle myosin expression (17). Similar de-differentiation of vascular smooth muscle occurs in vascular remodeling associated with atherosclerosis (18), and it is possible that chronic stimulation by cytokines and growth factors in the inflamed airway walls of asthmatic subjects could result in proliferation and de-differentiation of the airway smooth muscle, making it less contractile.

Lambert and colleagues (19) have attempted to determine how important an increase in smooth muscle mass could be as a contributor to exaggerated airway narrowing. They developed a computer model of the tracheobronchial tree in which airway smooth muscle shortening occurred in a dose-response fashion. Airway smooth muscle shortening was allowed to proceed until the force generated by the muscle was equal and opposite to the force occurring in the surrounding lung parenchyma. They assumed that the contractile function and the length-tension relationship of the asthmatic airway smooth muscle was the same as normal, but that the force-generating capacity was increased in proportion to the increase in muscle mass. Using the model they compared the importance of airway mucosal thickening and adventitial thickening to increased smooth muscle mass. Although all three of these structural changes, which occur in asthma, contributed to increased maximal airway narrowing, the analysis showed that, of these factors, the increase in airway smooth muscle thickness was the most important abnormality. The increased submucosal and adventitial thickness could increase the maximal airway narrowing two- to 10-fold, whereas the increased muscle thickness has the potential to increase maximal airway narrowing by two orders of magnitude. In this analysis, a number of assumptions were made that remain to be tested. The most important of these was that the increased airway smooth muscle mass that occurs in asthma is accompanied by a parallel increase in the ability of the muscle to generate force (i.e., maximal airway smooth muscle stress remains constant). However, as described above, there is reason to believe that when airway smooth muscle proliferates, there could be a decrease in the contractility of the muscle. The other major assumption in the model was that the only external load impeding airway smooth muscle contraction was the series elastance related to lung elastic recoil. More recent studies show that the folding of the airway mucosal membrane that occurs when airways narrow in response to airway smooth muscle contraction may provide a considerable load, especially if there is thickening of the airway wall or changes in its mechanical properties secondary to connective tissue deposition (20).

	FUNCTIONAL ALTERATIONS IN AIRWAY SMOOTH MUSCLE IN ASTHMA

TOP ABSTRACT INTRODUCTION STRUCTURAL ALTERATIONS IN AIRWAY FUNCTIONAL ALTERATIONS IN AIRWAY FUNCTIONAL ALTERATIONS IN AIRWAY REFERENCES

Few studies have assessed the functional properties of airway smooth muscle from asthmatic subjects, and even fewer have addressed methodologic constraints in interpreting the data. From the standpoint of smooth muscle alterations, excessive airway narrowing could be brought about by: (1) an increased amount of smooth muscle with normal contractile properties; (2) altered smooth muscle cellular characteristics allowing for increased contractility of individual cells; (3) less compliant intercellular components connecting myocytes; and (4) decreased stiffness of the parallel elastic elements, either within myocytes or within connective tissue elements between and surrounding myocytes. Keeping these considerations in mind, we shall evaluate the data available and suggest potential mechanisms whereby these changes may occur in asthma.

Mechanical Changes in Asthmatic Airway Smooth Muscle

Force generation. If one makes the assumption that there is more airway smooth muscle in asthma, this muscle should also generate more force simply due to the increased muscle mass. Thus, it is surprising to find conflicting data in studies evaluating this parameter when the force has not been normalized by the cross-sectional area. In fact, data from only three laboratories demonstrate an increase in force generation of asthmatic versus control airway smooth muscle preparations, with four of the reports dealing with surgically resected specimens (23- 26) and one with autopsy material (27). These results differ from a number of other reports in which no increased force generation was found (28). There are significant methodologic limitations when evaluating all studies, since none of the studies except one (26) determined the amount of muscle within the preparation analyzed. Perhaps a greater limitation was the failure to determine the optimal length for maximal force generation in the asthmatic tissue compared with control tissue. As dealt with in detail elsewhere (32), this could account for the failure to appreciate a real difference in force, whereas it would mean that the studies showing increased force generation could have underestimated these changes. This limitation relates to the hypothesis that airway preparations of subjects with asthma have decreased tissue elastance. Placing a standard load on the tissue that is appropriate for nonasthmatic airways would stretch the muscle to lengths exceeding optimal lengths, thereby producing less force. In studies evaluating the cross-sectional airway smooth muscle in asthmatic versus control preparations (14, 26), the force per cross-sectional area (stress) was significantly greater for asthmatic subjects (Figure 2A). This finding suggests that an increased amount in smooth muscle alone cannot account for the force produced, and raises the possibilities of altered contractility of myocytes or changes in myocyte interactions or parallel elastic elements placing a load on the muscle.

View larger version (17K): [in this window] [in a new window]   Figure 2.   Mechanical responses of lobar airway smooth muscle preparations from subjects with asthma (n = 2) compared with those of subjects without asthma (n = 5). Mean and standard errors are calculated based on Table 2 in Reference 14. (A) Force generation normalized to cross-sectional area of smooth muscle. (B) Shortening expressed as percent of the starting length.

Isotonic shortening. In the only studies evaluating shortening (14, 26), asthmatic preparations produce a two- to threefold increase in shortening compared with controls (Figure 2B). These changes were striking and are consistent with the hypothesis that increased airway smooth muscle shortening is a major determinant of the excessive lumenal narrowing seen in asthma. Such increased shortening could be due to any of the factors mentioned for increased force generation. Velocities of shortening have not been evaluated in asthmatic airway smooth muscle but one would anticipate that these would parallel the degree of shortening if any of the proposed mechanisms are involved.

Passive length-tension characteristics. Limited data are available, but our findings from two asthmatic subjects suggest that large airway preparations have decreased tissue elastance compared with control subjects (26). Thus, less force is produced upon passive stretch of the tissue, and the passive resting force at the optimal length for isometric contraction was less for the two asthmatic preparations evaluated. This finding, coupled with the increased force and shortening of asthmatic smooth muscle (Figure 2), led us to hypothesize that decreased airway tissue elastance allows increased shortening because of the lower load required to be overcome for shortening to occur (26).

Potential Factors Increasing Airway Smooth Muscle Shortening in Asthma

Effects of allergic serum. A number of reports regarding the effects of passive sensitization with sera obtained from allergic subjects have suggested that such treatment enhances contractile responses. However, there are again conflicting data, and some findings are difficult to explain in light of the enhanced in vivo responses to contractile agonists seen in asthmatic subjects. For example, the two original studies using human airways (33, 34) reached the opposite conclusions. Roberts and Thomson (33) found no effect of passive sensitization on histamine-induced contractions. Black and coworkers (34) found an increase in response to histamine but a decrease in response to carbachol. These studies evaluated force generation, whereas one by Mitchell and associates (35) showed increases in both amount of shortening and velocity of shortening without any increase in force generation upon passive sensitization with a high immunoglobin E (IgE) titer serum. Whether changes seen are related to IgE itself having an effect on myocytes (no evidence for IgE receptors) or other cell types, such as mast cells within the preparation, remains a debate. It is possible that other cytokines within the serum are having effects, and the roles for interleukin (IL)-1 and tumor necrosis factor alpha (TNF-) have been suggested, since these decrease -adrenoceptor relaxation of airway smooth muscle of rabbits (36).

Cytokines. With the myriad of cytokines demonstrated to be produced in the inflammatory process of asthma, it is tempting to suggest that one or more of these are intricately involved in altering smooth muscle responses. Although lacking direct evidence, a role for TNF- and IL-1 is appealing, with a number of possible effects. These could include increases in airway smooth muscle proliferation (37), although in vitro studies demonstrating such a response suggested changes to a noncontractile phenotype (17). This phenotypic change would decrease, rather than increase, smooth muscle shortening. In addition, TNF- via the generation of nitric oxide decreases contractility of isolated cardiac myocytes (38). However, both TNF- and IL-1 can lead to the secondary production of collagenases (39, 40), which would decrease the load on the muscle and allow greater shortening.

Matrix-degrading enzymes. The prominence of extracellular matrix within airway smooth muscle preparations (14) raises the distinct possibility that inflammatory mediators stimulate the release of enzymes capable of altering collagen, elastin, and proteoglycans. Bramley and associates (41) demonstrated that treatment of human airway smooth muscle preparations with collagenase enhanced force generation and shortening of the preparations (Figure 3). Meiss (42) made similar observations with rabbit trachea regarding the effects of collagenase and has demonstrated that the restrictive radial expansion of muscle cells limits force generation and shortening (43). These data are intriguing in that they suggest that human airway smooth muscle is constrained in its ability to shorten not only by the loads that parallel elastic elements may place upon the muscle but also by their effects on preventing the radial expansion of muscle cells as they shorten. We anticipate that numerous different enzymes, altering extracellular matrix elements, may enhance airway smooth muscle shortening. Studies evaluating smooth muscle in its normal local environment may provide more relevant answers than those of isolated cells when we consider the changes in asthma.

View larger version (12K): [in this window] [in a new window]   Figure 3.   Effects of collagenase upon human airway smooth muscle mechanics. Force and shortening for each preparation were determined before and 2 h after the addition of collagenase, which enhanced both mechanical properties. (Reprinted from Reference 41 by permission.)

	FUNCTIONAL ALTERATIONS IN AIRWAY SMOOTH MUSCLE IN ANIMAL MODELS OF ASTHMA

TOP ABSTRACT INTRODUCTION STRUCTURAL ALTERATIONS IN AIRWAY FUNCTIONAL ALTERATIONS IN AIRWAY FUNCTIONAL ALTERATIONS IN AIRWAY REFERENCES

Since the emphasis of this review is on human airway smooth muscle, inclusion of a discussion on animal models of asthma is to serve the purpose of better understanding the mechanisms underlying some alterations observed in human asthmatic airways. Only a small selection of the relevant literature is reviewed, and the focus is on describing analytical methods that can be applied to human airway smooth muscle.

Force-Velocity Relationships

In the limited number of mechanical assessments of human asthmatic airway smooth muscle, no force-velocity relationship of the muscle has been determined. Passive sensitization of human bronchial smooth muscle showed an increase in both shortening velocity and capacity (35). Measurements of airway smooth muscle strips from two patients with asthma showed an increased amount of shortening (Figure 2). In airway smooth muscles from allergen-sensitized animals, both shortening velocity and capacity (but not isometric force) have been shown to increase compared with the nonsensitized littermate control animals (44). The increase in shortening velocity in sensitized canine trachealis is partially attributed to an increase in the quantity and total activity of the myosin light chain kinase found in the same preparation (47). Table 2 summarizes maximum shortening velocity, maximum amount of shortening, and force-velocity constants from five ragweed-sensitized dogs and their littermate controls (taken from tabulated values in Reference 48).

Shortening velocity and airway narrowing. In a static mechanical equilibrium, smooth muscle length in the airways (and presumably the airway diameter) is determined by the force generated by the muscle cells and the elastic loads against which the muscle cells shorten. The rate at which the muscle arrives at its final length is not as important as the final length itself in this static model. There are, however, conditions under which shortening velocity becomes important in determining the amount of shortening. Smooth muscle is known to possess two distinct phases of contraction, the initial fast shortening phase and the sustained phase, characterized by a reduced shortening velocity (49). The amount of shortening is likely determined by the shortening accomplished during the initial phase, before the "latch" phase sets in. In this scenario, an increased shortening velocity would translate directly into increased amount of shortening.

It has been recognized recently that airway smooth muscle must have been stretched during tidal breathing, certainly during deep inspiration (50). The mean diameter of the airways in this dynamic event of tidal breathing becomes critically dependent on airway smooth muscle shortening velocity during the period of exhalation. An increase in shortening velocity would lead, at least theoretically, to excessive airway narrowing. In this dynamic model, shortening velocity, but not shortening capacity, becomes the determinant factor controlling airway caliber.

Length, Passive Tension (Compression), and Velocity Relations

Our limited data indicate that passive tension associated with optimal isometric tension generation is less in airway preparations from asthmatic subjects compared with those from subjects without asthma (26). The significance of a reduced stiffness of the parallel elastic component in terms of airway narrowing is twofold. First, the resistive load to shortening is reduced, allowing the muscle to shorten to a greater extent. Second, the reduced load to shortening would result in an increase in shortening velocity, and hence in airway narrowing during tidal breathing. In light of the dynamic equilibrium observed during tidal breathing, the second mechanism may be more critical in producing excessive airway narrowing. It is important then to add a velocity component to the length- tension relationship in order to understand the intricate relationship among these parameters and how they could affect the degree of airway narrowing.

Compliance of resistive load and airway narrowing. A passive length-tension curve obtained from stretching a relaxed airway preparation provides us with information about the compliance of the parallel elastic component of the preparation in tensile mode. The length-tension behavior of the passive component could be very different in compressive mode, that is, during active shortening. So far there is no direct method for measuring compliance of a resistive load encountered by the contractile apparatus in muscle cells during an active shortening. Here we describe an indirect method for estimating the compliance of internal resistive load of a muscle preparation during active shortening. Limitations associated with this method of analysis are also discussed.

It has been found that the relationship between muscle length and zero-load shortening velocity can be described by a parabolic function (55) of the form:

(1)

where V_o is zero-load velocity as a function of muscle length (L), V_max is the zero-load velocity at optimal muscle length (L_max), and L_min is the minimum muscle length or maximally contracted length under zero load. Force-velocity relationships are usually characterized by the Hill equation (56):

(2)

where V is shortening velocity as a function of load (P), and a and b are Hill constants.

Figure 4 illustrates idealized length-velocity and force-velocity curves. The velocity variation described by the length-velocity curve is due entirely to factors intrinsic to the muscle preparation that are sensitive to length change, because no external load is involved. The velocity change described by the force- velocity curve, on the other hand, is due entirely to variations in external load, because the velocities are measured at optimal muscle length where no compression of the internal resistive load is involved. Assuming that shortening velocity of the contractile apparatus will be the same if the load against which it shortens is the same, regardless of it being from an internal or an external source, we can obtain an estimate of the magnitude of an internal load associated with a particular length. This is accomplished by equating isovelocity points in both the length-velocity and force-velocity curves (Figure 4A and 4B) and obtaining the corresponding muscle length from the length- velocity curve and external load from the force-velocity curve. The external load is assumed to be equal to the internal resistive load. Plotting the load versus length (Figure 4C) we obtain, graphically, a curve that describes the relationship between muscle length (at L < L_max) and passive internal resistive load. The same relationship can be obtained mathematically as well, by equating the right sides of Equation 1 and Equation 2 and rearranging the terms:

(3)

View larger version (16K): [in this window] [in a new window]   Figure 4.   Illustration of the method for obtaining force (or stress) versus shortened length (or compression) curve of the parallel elastic component of muscle. (A) Length-velocity curve described by Equation 1. (B) Force-velocity curve described by Equation 2. (C ) Stress-compression curve of the internal resistive load (or parallel elastic component) described by Equation 3.

Although conventionally we regard the passive load as originating from the parallel elastic component of the muscle preparation, it should be noted that the anatomical origins of these loads are not clear and that some of the loads may not be "parallel" but rather "radial" in relationship to the contractile apparatus geometrically (43).

Validity of this type of analysis is limited by the simplifying assumption that the ability of the muscle to generate force is constant throughout the length range, which is not true over a large length range. However, smooth muscle is known to be able to generate optimal force over a much greater length range than skeletal muscle; a threefold length range over which isometric force is constant has been found in canine trachealis when the muscle is allowed to adapt to length change (57). During tidal breathing, the variation of airway smooth muscle length is probably within a range over which the relationship described by Equation 3 is accurate. Because airway resistance is related to the fourth power of the airway diameter, a slight change in muscle length due to variation in resistive-load compliance will result in drastic change in airway resistance.

Alteration in the compliance of resistive load in sensitized canine airway preparation. Table 2 lists values of force-velocity parameters and L_min for airway preparations from sensitized dogs and their littermate controls. These are all the parameters needed to construct a stress-compression curve according to Equation 3. Stress is the resistive load (P) normalized to the cross-sectional area of the preparation, and compression is the shortened length (L) normalized to initial muscle length. Figure 5 shows the stress-compression curves for sensitized and control preparations, plotted according to Equation 3 and Table 2. Not surprisingly, the sensitized tissue is shown to be more compliant, a cause for the increased shortening velocity and capacity, according to our analysis. It should be pointed out that the airway preparations used to obtain the results shown in Figure 5 are relatively devoid of connective tissue; smooth muscle cells occupy approximately 80% of the cross-sectional area of the preparation. The increased compliance in the sensitized tissue therefore is likely to reflect changes in the smooth muscle cells themselves. The changes allow the muscle to shorten to a greater extent and could result in excessive airway narrowing. The increased velocity due to a more compliant resistive internal load is also likely to cause more airway narrowing, because during the exhalation part of the breathing cycle the sensitized muscle is likely to shorten more and reduce the mean airway diameter.

View larger version (12K): [in this window] [in a new window]   Figure 5.   Stress-compression curves plotted from mean values for sensitized and control canine airway preparations listed in Table 2. The curves are described by Equation 3.