Role of airway smooth muscle in asthmaPossible relation to the neuroendocrine system.код для вставкиСкачать
THE ANATOMICAL RECORD 236:152-163 (1993) Role of Airway Smooth Muscle in Asthma: Possible Relation to the Neuroendocrine System NEWMAN L. STEPHENS, HE JIANG, AND ANDREW HALAYKO Luzhou Uniuersity, Peoples’ Republic China (H.J.), Plant Science, University of Manitoba, Winnipeg, Manitoba, Canada (A.H.);London, England (N.L.S.) ABSTRACT Though not yet firmly established, it appears likely that the neuroendocrine system (NES)regulates airway smooth muscle function. As it is the latter which is altered in asthma, the importance of the role of the NES in this disease is clear. The fact that transmitters from the NE cells are released from their basal aspect, and are in close proximity to the subjacent airway smooth muscle, further indicates an interaction. The question then arises as to what are the experimental desiderata for conducting studies of the ASM. These should constitute what Sergei Sorokin has called the “Koch’s postulates of airway smooth muscle research.” As human tissues from asthmatics are difficult to obtain, animal models have been developed. The requirements are that, in these animals, the allergy be IgE based, that a congenital or familial factor be operative, that a noncholinergic nonadrenergic inhibitory system be a component of the neural regulatory system, and that the antigen for immunization be of a type commonly found in human asthmatics. Ideally, evidence of clinical asthma and exercise-induced asthma and nocturnal attacks should also be present. Unfortunately, no ideal animal models exist and one cannot talk about asthmatic animals, but only of animals with allergic bronchospasm. If in vitro research is to be conducted, there are additional requirements. The tissue should be from a relevant location. The tracheal smooth muscle which has been the favorite, purely because of its convenience, is not a good model. For the early asthmatic attack, central bronchi (3-5 mm diameter) should be used. Muscle strips obtained from them should be parallelfibred and the cartilage plaques should be carefully dissected away, otherwise they contribute unwanted frictional forces when velocity is measured. Care should be taken to ensure that the epithelial cell layer is intact, as evidence indicates that it may regulate airway muscle function, though this has not been established for all the animal species used in asthma research. The isolated muscle strip should be in a steady state, particularly with respect to the functional variable under study, before definitive data are collected. Most importantly, it is shortening capacity that must be studied, as this is the in vitro analogue to in vivo narrowing of airways. Isometric force development provides information about wall stiffness and is of very little relevance to the elucidation of the mechanism of bronchospasm. Furthermore, as force is measured at the plateau of the record, it only yields data relating to latch bridges which seem to play little role in narrowing the airway. Unfortunately, isometric force is the most common measurement made, purely because it is easy to carry out, while shortening is technically more difficult. Received October 23, 1991; accepted December 10, 1991. Address reprint requests to N.L. Stephens, Prof. Dept of Physiology, Faculty of Medicine, University of Manitoba, 425, Basic Medical Sciences Building, 770 Bannatyne Avenue, Winnipeg, Manitoba R3E 357 Canada. Q 1993 WILEY-LISS, INC AIRWAY SMOOTH MUSCLE AND THE NEUROENDOCRINE SYSTEM 153 The above notwithstanding, if force is the parameter being studied, its correct normalization is a very important consideration. To make comparisons, force should be converted to stress. In tissues which are more than 5 mm in length and 5 mg in weight, the required cross-sectional area can be obtained from the weight of the blotted tissue and its optimal length. For tissues smaller than this measurement errors render the approach invalid and direct measurements must be made using high performance optics. It must be remembered too that the cross-sectional area of the tissue is only appropriate when the entire tissue consists of muscle, as for example, in the case of striated muscle. In smooth muscle, muscle content may only be 25%. For this reason, tissue stress should be converted to muscle cell stress and ultimately to myosin stress, as it is the myosin crossbridges that generate the force. Estimation of muscle stress requires quantitative morphometry while myosin stress requires morphometry of immunohistochemical micrographs. Finally, in conducting studies of shortening and velocity, the nature of the loading has to be kept in mind. Such studies are usually carried out with isotonic loads as this holds that variable constant. However, in vivo the load is more likely to be elastic or visco-elastic, hence, in vitro studies should employ similar loading. In our own studies we have attempted to pay attention to the desiderata mentioned above. In bronchial smooth muscle (central airways) from 4-month-old, ragweed pollen-sensitized dogs, we have found that maximum shortening capacity and velocity are both increased while force production is normal. These changes are typical of early disease. In addition, we have found that the compliance of the muscle’s so-called internal resistor is increased with sensitization. This could account for the increased shortening capacity of the muscle. We have also shown that the maximum velocity of the sensitized muscle is associated with increased myofibrillar ATPase activity. This results, not from a change in distribution in myosin heavy chain isozymes, but from increased phosphorylation of the 20,000 dalton myosin light chain that is due to an increased content of myosin light chain kinase. Our studies indicate that this increase is due to increased gene translation rather than transcription, as the content of messenger RNA for myosin light chain kinase is unchanged. o 1993 Wiley-Liss, Inc. Key words: Airway smooth muscle, Sensitized airway smooth muscle, Smooth muscle myosin light chain kinase In a publication devoted to consideration of the structure and function of neuroendocrine (NE) cells in health and disease, this chapter on the physiology and pathophysiology of airway smooth muscle (ASM) is assuredly a n outlier. It has been included because the importance of the NE system, whose function is still not known, is “intuitively” felt to be in its putative regulatory role in airway narrowing. If this is to be substantiated, a clear understanding of airway smooth muscle contractility is mandatory. Perhaps the most important role for impaired interaction between the NE system and ASM could be in the pathogenesis of asthma or of the effects of hypoxia on pulmonary function. When so much has been written about airway smooth muscle in recent times, it is a curious anomaly that we do not yet know the physiological function of ASM, mainly because appropriate and reliable data are not available. This has been clearly pointed out by Otis (1983). Much of the physiology of ventilation and its regional distribution can be accounted for by considering the airways as passive conducting tubes and the respiratory skeletal muscles a s the driving pump for ventilation, smooth muscle itself being not required. Speculation has, however, provided a list of ASM functions. Tracheo-bronchoconstriction,by reducing anatomical dead space, would improve alveolar ventilation. This would only be of benefit if ventilation frequency did not increase too greatly, a s the increased work of breathing would limit alveolar ventilation. Regional distribution of ventilation is also regulated by local bronchoconstriction, as is the optimization of the matching of ventilation to perfusion. Outpouring of mucus from the glands into the airways has also been ascribed to rhythmic ASM contraction. All these speculations aside, there is little doubt that, a t least, the early asthmatic response in patients is the result of increased central airway smooth muscle contraction. To what extent ASM contraction is involved in the late asthmatic response where bronchial narrowing involves peripheral airways, is the source of debate. The narrowing is felt to be due to geometrical factors wherein local oedema and inflammation narrow the lumen. 154 N.L. STEPHENS ET AL Study of Airway Smooth Muscle Function: The Nature of Loading Whether one wishes to study the role of ASM in asthma or elucidate its relations to the function of neuroepithelial bodies, it is important to identify the correct parameter to study. This statement would be trivial were it not for the fact that the major portion of in vitro research described in the literature is devoted to measurement of isometric force development (Shioya et al., 1989; Armour et al., 1985).The use of this approach has been sanctioned by custom. It is difficult to justify this, as isometric force development only provides insight into the stiffness of the tissue without providing any information about shortening, which is the in vitro counterpart to in vivo bronchoconstriction. In vivo studies, which are mainly those of specific airway conductance, are free from this criticism. Having indicated that shortening is the study parameter of choice, use of the correct loading regimen for the muscle during shortening is a sine qua non. Classically, studies of shortening are carried out under isotonic loading conditions chiefly because information is sought about intrinsic crossbridge activity; time-dependent load changes would constitute a confounding variable. However, for studies that aim to relate to in vivo function, the load that is imposed on the shortening muscle in vivo must be identified. A recent hypothesis about the pathogenesis of asthma is critically based on the nature of this loading. As ASM contracts and the airway narrows, the elastic properties of the components of the airway wall (connective tissue and cartilaginous plaques) and lung parenchyma provide additional loads that limit the extent of shortening. This is based on the idea that ASM is directly attached to the cartilaginous plaques which in turn are the sites of insertion of lung parenchyma. However, histologic data published by von Hayek (1960) show that ASM is not directly attached to the cartilage but via loose connective tissue. Furthermore, in the central airways the lung parenchyma is not directly inserted into the bronchial wall but again via loose connective tissue. Thus, the load on the ASM may not be elastic but isotonic. This is a n important point to settle if meaningful asthma research is to be undertaken. At any rate, the hypothesis states that were parenchymal compliance to increase, this would result in reduced bronchial loading and account for asthmatic bronchoconstriction. The Concept of Contractility What has been discussed above is the role of ASM shortening and though this may help explain phenomena a t whole organ level, i t will not elucidate mechanisms which usually operate a t subcellular and molecular levels. To elucidate the former, auxotonic shortening must be studied; for the latter, isotonic. Measurement of shortening, however, does not provide complete information regarding muscle contractility, even though it is paramount with respect to allergic bronchospasm. To obtain comprehensive information, all modalities of function should be measured as shown by cardiac muscle investigators (Brutsaert e t al., 1971).These consist of maximum shortening capacity (Almax),maximum shortening velocity (VJ, maximum force development (Po)and time. The importance of time stems from the fact that maximum values of the other factors can only be obtained if rates of energy liberation are a t a steady maximum. The last named varies quite considerably with time, maximally in the case of cardiac muscle. To eliminate time-dependent effects, tetanization has been used successfully. However, in the case of cardiac muscle, tetanization is not possible, so steps must be taken to ensure that only measurements taken at the moment of peak energy liberation are used in defining contractility. Special Considerations With Respect to Contractility in Airway Smooth Muscle Consideration of what mode of contraction to study and how to assess contractility, though adequate for the study of striated muscle, is not so for smooth muscle which possesses another layer of complexity. In striated muscle the assumption can be made that average crossbridge function is homogeneous throughout the course of contraction. This is not so for smooth muscle where the bridges are heterogeneous. Unlike striated muscle, in which velocity of shortening, once maximum is reached, remains unchanged, smooth muscle shows progressive drop. In canine tracheal smooth muscle (Stephens et al., 1985b) and in bronchial (Jiang and Stephens, 1990) have shown that, as in other smooth muscles (Dillon et al., 1981), velocity decreases considerably with time. This has been explained by the development of so-called latch bridges (LBR) which retard the velocity of normally cycling bridges (NBR) that commenced operating during early contraction. Alternatively it has been hypothesized that the mechanism is the uniform and progressive slowing of all the bridges. This controversy has not yet been resolved. That notwithstanding, we have reported (Stephens et al., 1985) that peak velocity develops within 2 seconds of onset of muscle shortening and is brought about by the activity of NBR. Within this period the muscle achieves 75% of its shortening; hence, in conditions where shortening is altered, as for example asthma, shortening must be studied within this 2 second period. The major regulatory biochemical mechanism involved is phosphorylation of myosin light chain by myosin light chain kinase and it is that which needs study. After 2 seconds, LBR replace NBR, at least in a functional sense. The former produce most of the force developed by the muscle and little shortening. If muscle stiffness is to be studied, then studies a t about 8-10 seconds after stimulus onset should be undertaken. The biochemical process underlying the development of LBR (though controversial) is dephosphorylation of the myosin light chain which entails study of myosin light chain phosphatase (Dillon et al., 1981). From this, one may conclude that the study to conduct and the parameter to choose depend critically on the question being asked. Smooth muscle presents yet another problem and that relates to the existence of a so-called internal resistance to shortening (IRS). This is a hypothetical intracellular structure or process that resists shortening as it becomes compressed. Its stiffness therefore dictates the magnitude of muscle shortening. Its presence has been shown in cardiac muscle by Chiu et al. (1982). We have reported its presence in airway smooth muscle and have developed a method to delineate its compres- AIRWAY SMOOTH MUSCLE AND THE NEUROENDOCRINE SYSTEM 155 sion-tension properties (Seow and Stephens, 1986). We have also reported that its compliance is increased in ASM from a dog model of asthma; this could account for the increased bronchospasm seen in such preparations. To sum up this section, to describe the change occurring in asthma one must measure the airway muscle’s maximum shortening capacity (Al,,,), but to elucidate the underlying mechanism, the internal resistance to shortening must be studied. Molecular Mechanisms of Airway Smooth Muscle (ASM) Contraction Both the physiology of normal ASM contraction and the pathophysiology of asthma are based on biochemical changes involving processes which include excitation, excitation-contraction coupling, and contraction. These involve the biochemistry of neuromuscular transmitters and junctions (Burnstock, 1970) and peptides, the behavior of the resting membrane potential (Small e t al., 1980; Kroeger and Stephens, 1975), the activity of ion channels, the role of G-proteins, inositol triphosphate, diacyl glycerol (Somlyo et al., 1987), the release of calcium and its interaction with calmodulin, and the activation of calcium-calmodulin-dependent myosin light chain kinase (Kamm and Stull, 1988) with resultant phosphorylation of the 20 kilodalton myosin light chain (MLC). The process of relaxation requires reversal of these processes with the addition of dephosphorylation of MLC by a specific phosphatase. Our own studies have revealed the following: Airway smooth muscle model Our original work was performed on canine tracheal smooth muscle (Antonissen et al., 1979) even though we knew the site of physiological regulation of ventiFig. 1 . Montage of a longitudinal section of canine tracheal smooth lation was a t the level of 3 mm diameter airways. However, at the time, we felt that the presence of cartilage muscle that has been fixed at the peak of an isometric contraction accounts for the dentrate cell outlines. The perifasicular space plaques in those bronchi, by increasing visco-elastic which above the muscle bundle contains college, a blood vessel, fibroblasts loads, would render evaluation of force-velocity-and- and a mast cell typified by its granules. The calibration bar represents shortening properties difficult. Figure 1shows tracheal 10 +m. (Taken from Stephens N.L. in Asthma: Basic Mechanisms and smooth muscle in longitudinal section. The fibres are Clinical Managment. P.J. Barnes, I.W. Rodgers and N.C. Thomson, parallel, which is a requirement for valid mechanical eds. Academic Press, New York, 1988, pp. 16.) studies. Figure 2a shows a section from a 5th generation ca- Length-tension properties of canine tracheal smooth nine bronchus. The parallel arrangement of fibres is muscle (TSM). (Fig. 3) evident a s is the fact that there is much less muscle These were elicited by conventional techniques and present (25% of tissue area) than in tracheal (75%). show that maximum stress development (1.1kg/cm2 or Finally i t must be noted that the muscle is not directly = 1 x 105N/M2)was of the same value a s seen in other attached to the cartilage but via loose connective tis- smooth and striated muscles. The curves enable idensue. Therefore, a s pointed out before, the role of carti- tification of optimal length (1, or l,,,) which is the lage as a source of loading of the shortening muscle is length a t which most studies should be conducted. The debatable. practice of applying a fixed, arbitrary load on the musFigure 2b shows a section similar to that shown cle to conduct pharmacological studies is incorrect, as above but here the cartilage has been carefully dis- the length achieved is not correspondingly fixed but sected away. The intactness of the subjacent muscle depends on the intrinsic compliance of the tissue. From fibres is evident. Studies have shown that the maxi- the curves-see horizontal line labelled Alma,-the mum isometric force (PJ, the maximum velocity of muscle’s maximum isotonic shortening capacity can be shortening (V,) and the maximum shortening capacity deduced and is a surprising 90% of 1,. This, as i t is (Al,,,) are all increased after removal of the cartilage. derived from static measurements, is essentially a n The reasons for the increase in V and Alma, are evi- overestimate. dent; that for the increased P,/mm’ is not. We feel that Maximum isotonic shortening measured dynamithe increase results from the normalization procedure, cally is about 65% which represents the true value. as the tissue cross-section has been reduced by elimiFigure 4. One of our earliest findings (Antonissen et nation of the cartilage. al., 1979) was that in ASM from ragweed pollen- and/or 156 N.L. S T E P H E N S ET AL. Fig. 2. a: Histological sections of 5th order canine from bronchi, before removal of cartilage; smooth muscle is located right beneath epithelium and does attach directly to cartilage external to it. EP1, epithelium; SM, bronchial smooth muscle; CT, cartilage. Haematox- ylin and eosin stain. x 200. (From Jiang and Stephens, J. Appl. Physiol., 69: 1990.) b Same as in a except that cartilage has been dissected away. AIRWAY SMOOTH MUSCLE AND THE NEUROENDOCRINE SYSTEM 157 the internal resistance to shortening (IRS). However, to do that the force-velocity (FV) properties of the muscle should first be dealt with. TISSUE WT. 0.015f0.002 G M 1.04* ~ ~ 0.05CM. L Force-velocity (FV) relationships LENGTH PERCENT LMAX Fig. 3. Length-tension curves from canine tracheal smooth muscle (From Stephens e t al., J. Appl. Physiol., 26:685-692, 1969.) ovalbumin-sensitized dogs, Al,, was increased. The increase, though only 10-15% of lo, accounted for a 50% increase in airway resistance. Studies to determine the causes of the increased Al,, require consideration of the factors limiting muscle shortening, viz 7( The dynamic properties of ASM, which are the really physiological ones, are best assessed by delineating the force velocity relationships of ASM (either tracheal or bronchial) (Fig. 5 ) . However, as indicated before, independent FV curves for NBR and LBR are required. These, elicited by standard quick-release techniques during the course of a n isotonic contraction are shown in Figure 6 . They reveal that the maximum velocity of shortening for NBR is about 80% greater than that for LBR; however, the pattern is reversed for Po;Al,,, for NBR is, of course, greater than that for LBR. In sensitized ASM (see test curve in Fig. 6 ) while P, remains unchanged, V, and Al,, are both increased. For the latter refer back to Fig. 4. An important point to be made here is that these are changes seen in a model of early disease. At later times, secondary changes obscure the picture. It is interesting to note also that had the hypothesis been tested isometrically, the results would have been misleading, as the absence of difference in Po would have led us to conclude the muscle was normal. This again points up the superiority of studies of shortening. In sensitized muscle, only the 2 second NBR show a n increase in Al,, and V,, the LBR remain normal. This confirms that it is critically important to study the ap- TSM SHORTENING(% I,,) .( 5t 4c %b aa 10 10 0 % Fig. 4.Plot of changes in length during isotonic shortening a t different loads for control and sensitized canine trachealis. (From Antonissen et al., J. Appl. Physiol., 1974.) 158 - N.L. STEPHENS ET AL. 0.1 0 500 0- 400 0.06 300 0.04 200 development only tell us about the development and regression of wall stiffness and are limited in scope. 3. Alterations in shortening ability depend not only on changes in crossbridge properties but also upon the stiffness of the so-called internal resistor. 4. While velocity in skeletal muscle is not a limitation to extent of shortening, it is a limitation in smooth muscle. I n the latter, most of the shortening is completed within 2 seconds by normally cycling crossbridge, so changes in V, could affect maximum shortening capacity. " Elucidation of the molecular mechanisms for changes in ASM structure and function require biochemical studies. As our data showed that shortening capacity (Alma*) and velocity (V,) were increased, the following studies were undertaken. Changes in AI,,, This is a relatively new area and much needs to be done. The internal resistor is situated either between 0 cells or between cell bundles and is likley to be made up 0 -aof collagen, elastin and other matrix substances. Alter4 8 12 16 20 0 natively, it could be the cytoskeleton of the smooth LOAD IN GRAMS(P) muscle cell. With respect to the latter, desmin, tubulin (aand 6) and a-actinin are likely candidates. We have currently set up methods to identify these by using ;ap! specific antibodies and Western blotting techniques. Figure 9 is a gel showing that vimentin and desmin Fig. 5. Mean isotonic after-loaded force-velocity curves from canine can be identified in canine tracheal smooth muscle. tracheal smooth muscle, n = 32. P = load; V = maximum velocity of shortening for a given load. Po = maximum load muscle can support Using densitometry, they can also be quantitated. Very without shortening; a and b are constants with units of force and preliminary experiments suggest there is no difference velocity, respectively. Note: the right hand ordinate applies to the between these proteins from control and sensitized linear transform of Hill's equation (From Stephens et al., J. Appl. ASM. In hypertensive blood vessels, Berner et al. Physiol., 26t685-692, 1969.) (1981) have reported considerable increase in these proteins; however, in these vessels, hypertrophy was well advanced and the protein changes are likely late propriate crossbridges in testing our hypotheses. TO changes. In our model there is no evidence of cellular study the time course of crossbridge activity, maximum morphologic change and the changes we see are early velocity of shortening by the application of so-called and therefore more likely to relate to early causes. zero load clamps should be measured at second or half second intervals throughout the course of a contracChanges in properties of normally cycling tion. The results of a n experiment are shown in Figures 7 crossbridge (NBR) We have already alluded to the role of NBR in regand 8. The maximum slopes of the slow transient confirm that the velocity of the NBR is about 3-4 times ulating Alma,. The most likely cause of the increased V, we have reported (Stephens et al., 1985) is increased faster than that of the LBR (Stephens et al., 1985a). The importance of V, in studies of asthma may well activity of actin-activated myosin M S +-ATPase activbe queried, as it is only Al,, that relates to resistance ity. in the intact airway, However, if one remembers that Figure 9 shows a gradient polyacrylamide gel loaded the bulk of shortening is achieved by NBR in the first with a crude extract of canine tracheal smooth muscle. two seconds of the contraction, a reduced or perhaps The corresponding densitometrogram is also displayed. The bands for the contractile proteins actin (A), myosin even a normal V, may not permit any increase in Al,, heavy chain (M), myosin light chain (MLC) and one of to develop because of the time limitation. To sum up what has been covered so far, the follow- the regulatory proteins, tropomyosin (T) are clearly ing points are relevant. seen. The bands for the cytoskeleton proteins filamin, desmin and vimentin are also identifiable. Figure 10 shows that myofibrillar ATPase activity is 1. If insight into human physiology and pathophysiology is sought, the choice of a n appropriate animal increased in sensitized TSM. The next step was to determine the cause of the inmodel is critical. 2. If information is sought relating to normal abnor- creased ATPase activity. To this end we identified mymal airway function, studies of airway conductance are osin heavy chain isoforms (MW 204000 dalton and all important. In vitro, this can best be assessed by 200,000) and compared their distributions in control studies of muscle shortening. Studies of isometric force and sensitized muscles, a s this could account for al0.02 100 159 AIRWAY SMOOTH MUSCLE AND T H E NEUROENDOCRINE SYSTEM \ .o F-V CURVES .4 .6 .8 1.0 1.a 1.4 LOAD( kg/trn') Fig. 6. Mean isotonic after-loaded force-velocity curves from control and test (ovalbumen-sensitized) canine tracheal smooth muscle strips. The control curve is similar to that shown in Figure 6. While maximum velocity of shortening is increased in the test muscle, the maximum force developed (shown by x axis) is not. (From Antonissen et al., J. Appl. Physiol., 46:681-687, 1974.) tered ATPase activity. The fairly typical densitometrogram suggested no difference, and statistical analysis of a series of experiments confirmed this. Another explanation was that increased phosphorylation of the regulatory 20,000 dalton myosin light chain was responsible for the increased ATPase activity. A linear relation between the two has been reported. Figure 11shows 2-dimensional polyacrylamide gels with 20,000 dalton myosin light chain bands; B is the control and A the sensitized; no difference in densitometric pattern or area was detected. The two lower gels show radioautograms taken after 32Ptreatment. Increased phosphorylation of MLC2O is evident. In smooth muscle, purified actin-activated myosin ATPase has different properties from that of striated, wherein the activity of the latter is not directly stimulated by calcium concentration and under resting conditions displays close to maximum activity. Smooth muscles under these conditions displays very little activity. Thus, to evaluate any change in smooth muscle myosin ATPase activity one has to also consider the properties of the membrane and the excitation-contraction coupling mechanism, since they regulate calcium concentration. We have initiated studies of membrane electrophys- iology including patch clamping. Preliminary results in&@e that the resting sensitized membrane is depolarized by 5 mV and that three types of Ca2 activated K + channels appear to be present with conductances of 200pS, 8OpS, and 20pS. Studies of Ca+ channels, especially the L type, are under way. Returning to phosphorylation of the light chain, calcium-calmodulin and myosin light chain kinase (MLCK) are the important regulators, with the latter being the rate limiter. Using specific antibodies obtained from chicken gizzard (supplied by Dr. Mary Pato) we have found that the content of MLCK is increased (Fig. 12). This could account for practically all the changes we have described above. To determine the cause of the increased content of MLCK, measurements of the content of MLCK-mRNA were made employing Western blot techniques. The cDNA probe for the analysis was kindly sent to us by Dr. Vince Guerrero of the University of Arizona a t Tucson, Arizona. Results revealed no increase in the content of the message. We thus concluded that increased translation of the message, rather than transcription, must be the underlying cause for increased MLCK content. Studies of stability of the message and of translation are to be conducted in the near future. + 160 N.L. STEPHENS ET AL. 1 I STIMULUS .5. 0 5 TIME 10 IN Fig. 7. Shortening in normalized units (1, = optimal muscle length) is shown; it is plotted against time. The lowest sigmoidal curve represents lightly preloaded isotonic shortening. At different times during shortening, zero load clamps were applied. Each load clamp record shows an initial fast transient stemming from the elastic recoil of the ZERO LOAD-CLAMP DURING SECs. series elastic component and a shows one whole maximum slope provides the maximum velocity of shortening (V,) for that instant in time. The progressive dimminution of V, with time is evident. (From Stephens et al., Chest 88sr2235-2295, 1985.) VELOCITIES SHORTENING 3.: A / VELOCITY CURVE 2r 0 w v) \ E E v) z I.€ 0.6 3 > I0 0.7 0 w > > z I -1 9 .8 0.8 0 f Z W c 0.9 a 0 I v) 0 1 4 TIME 1 6 IN ' 0 1.0 SEC Fig. 8. Plot of V, vs. time. The original lightly preloaded isotonic shortening is also shown (right hand ordinate). (From Stephens et al., 1985.) 161 AIRWAY SMOOTH MUSCLE AND THE NEUROENDOCRINE SYSTEM Actin I I Tropomyosin 20% 4% Fig. 9. Gradient SDS-polyacrylamidegel (4-205) with densitometrogram shown contractile, regulatory and cytoskeletal proteins. to study of the latter by investigators primarily interested and possessing expertise in the former. In such a context considerable care has to be taken in studies of mechanical properties of smooth muscle. Some of the requirements are listed below. T STSM CTSM Fig. 10. Myofibrillar ATPase activity of sensitized (STSM) and control (CTSM) canine tracheal smooth muscle. The "Koch's Postulates" of Smooth Muscle Research vis-a-vis the Neuroendocrine System We stated a t the outset that the interrelation, a t least topographically, between the neuroendocrine system and the subjacent airway smooth muscle must lead 1. As the main purpose is to obtain insight into human physiology and pathophysiology, the choice of a n appropriate in vitro model is paramount. This is because, currently, it is very difficult to obtain human specimens. The objectives must be to reproduce the human disease as closely a s possible. This dictates that the disease in the animal model be IgE based. In the guinea pig, for example, allergic bronchospasm is IgG based which renders the model inappropriate. A genetic trait or familial influences in the animal are important, a s they very probably operate in humans. Bronchospasm should develop on challenge with specific antigen as well as to exercise. 2. The correct segment of airway should be studied with respect to the hypothesis being studied. If, for example, the early asthmatic response is to be studied then the central airways (3rd to 5th generation) should be chosen. For the late response the peripheral airways (less than 1 mm in diameter) should be selected. 3. The epithelial layer should be intact. In some animals the epithelium regulates the contractile response of the smooth muscle by liberating dilators such as nitric oxide or constrictors such a s the endothelins. In these, steps must be taken to insure epithelial function is intact. 162 N.L. STEPHENS ET AL Fig. 11. Two-dimensional gels electrophoresis of crude extracts from control and sensitized canine tracheal smooth muscle (A and B in upper panels). Radioautograms (32P) of the same gels as shown above are shown in the lower panels (A and B). 4. To compare muscle contractility between strips from the same animal or in the same segmental strip between animals or between animals a t different ages, certain steps must be taken to assure that meaningful comparisons can be made. These are: i. Performance of all experiments a t optimal muscle length (lo).This ensures that maximal contractile response is being studied. It is also known that excitability o r sensitivity of a muscle is length-dependent. If 1, is not carefully monitored, horizontal shifts in doseresponse curves may not be the effect of treatment but of length difference. ii. The use of a supramaximal stimulus. This is because with time (this may amount to 4-6 hours) muscle excitability diminishes. A stimulus initially identified as maximal may prove inadequate by the end of the experiment. Electrical field stimulation with mass platinum electrodes is the best. The repeated washouts required when chemical stimulation is used, result in rapid muscle deterioration. iii. The assurance that when data capture commences, the muscle is in satisfactory steady state. This generally requires gentle handling of the tissue during preparation, adequate oxygenation (and pH adjustment) and incubation a t the appropriate temperature for one to three hours. Steady state is said to be achieved when the maximum isometric force developed (Po), the maximum rate of force development (dPidt) and the contraction time do not vary by more than 10%. iv. Utilizing animals, when comparison studies are to be carried out, that are at the same level of maturity. At different ages the excitability of the muscle membrane and the activity of enzymes (Na+-K+-ATPase, myosin heavy chain ATPase) can vary quite considerably and influence the results of studies, v. Measurement of the appropriate parameter with respect to the exact functional modality to be studied. If muscle stiffness is to be studied then isometric force versus muscle length should be measured. a) To quantitate muscle strength, maximal force developed (Po)should be measured. However, for comparisons to be made between muscles,'normalization is necessary. The most convenient is normalization to tissue stress, i.e., force developed per unit tissue crosssectional area. This is satisfactory for skeletal muscle where more than 90% of the tissue cross-section is made up of muscle. In smooth muscle, however, the cross-sectional content of muscle cells, per unit area of the whole tissue, varies from 75% (tracheal smooth muscle) to 25% (bronchial smooth muscle). The normalization must therefore be made with respect to crosssectional area of muscle cell and not to that of the tissue, i.e., in lieu of tissue stress we need muscle cell stress. Even this does not represent the final normalization, a s not all the tissue in a muscle cell is contractile; normalization should therefore be made with respect to amount of contractile protein (actomyosin) present in the cross-section, i.e., the ultimate parameter is actomyosin stress. It must be stressed that to elucidate the muscular mechanisms that regulate distribution of ventilation or underlie the pathogenesis of allergic bronchoconstriction, it is shortening capacity that must be studied. Study of force development can yield very little information about airway resistance. b) To compare maximum shortening capacity and velocity, allowance must be made for the different lengths of muscle strips employed in different experiments. The best normalization is to relate length changes to optimal length (lo). In elucidating shortening capacity one must bear in mind that in smooth and cardiac muscle, this may be affected by maximal shortening velocity. This limitation does not hold for skeletal muscle. The smooth muscle limitation is due to the fact that 75% of the muscle's shortening develops in the first two seconds of the contraction. At this time, normally cycling bridges, with their relatively rapid velocities, are active. Changes in velocity could limit maximum shortening capacity. In studying the velocity of shortening it is, therefore, the velocity at two seconds that is important. c) Muscle power (the product of force by velocity) is normalized satisfactorily by referring the product to the weight of the muscle. ACKNOWLEDGMENTS The work described in this communication was supported by a n operating grant from the Medical Research Council of Canada and the American Council for Tobacco Research; J.H. was the recipient of a fellowship from the same council. AIRWAY SMOOTH MUSCLE AND THE NEUKOENDOCRINE SYSTEM Sensitized Control * T 163 0 TSM * T n=4 n=4 n=8 BSM n=9 Fig. 12. Myosin light chain kinase content in control and sensitized tracheal (TSM) and bronchial (BSM) smooth muscles. LITERATURE CITED Antonissen, L.A., R.W. Mitchell, E.A. Kroeger, W. Kepron, K.S. Tse, and N.L. Stephens 1979 Mechanical alterations of airway smooth muscle in a canine asthmatic model. J . Appl. Physiol., 46:681687. Armour, C.L., J.L. Black, and P.R.A. Johnson 1985 A role for inflammatory mediators in airway hyperresponsiveness. In: Mechanisms in Asthma: Pharmacology, Physiology and Management. Alan R. Liss, Inc., New York, pp. 99-108. Berner, P.F., A.V. Somlyo, and A.P. Somlyo 1981 Hypertrophy-induced increase of intermediate filaments in vascular smooth muscle. 3. Cell Biol., 88:96-101. Burnstock, G. 1970 Structure of smooth muscle and its innervation. In: Smooth Muscle. A. Brading, A.F. Jones, and T. Tomita, eds. Arnold, London, pp. 1-69. Brutsaert, D.L., V.A. Claes, and M.A. Goethals 1971 Velocity ofshortening of unloaded heart muscle and the length-tension relation. Circ. Res., 29t63-75. Chiu, Y., E.W. Ballou, and L.E. Ford 1982 Velocity transients and viscoelastic resistance to active shortening in cat oaoillarv muscle. Biophys. J., 40t121. Dillon, P.F., M.O. Aksoy, S.P. Driska, and R.A. Murphy 1981 Myosin phosphorylation and the crossbridge cycle in arterial smooth muscle. Science, 211t495-497. Jiang, H.. and N.L. Stephens 1990 Contractile Drouerties of bronchial smooth muscle with and without cartilag; J.-Appl. Physiol., 69: 120-126. Kamm, K.E., and J.T. Stull 1988 The function of myosin and myosin light chain kinase phosphorylation in smooth muscle. Am. Rev. Pharmacol. Toxicol., 25~593-670. Kroeger, E.A., and N.L. Stephens 1975 Effects of tetraethylammonium on tonic airway smooth muscle: Initiation of phasic electrical activity. Am. J. Physiol., 228:633-636. Otis, A.B. 1983 A perspective of respiratory mechanisms. J. Appl. Physiol., 54t1183. Seow, C.Y., and N.L. Stephens 1986 Force-velocity curves for smooth muscle: Analysis of internal factors reducing - velocity. Am. J . Physiol., 25:C362-C366. Shioya T., E.R. Pollack, N.M. Munoz, and A.R. Leff 1989 Distribution of airway contractile responses in major resistance airways of the dog. Am. J. Pathol., 129:102-117. Small, R.C., J.P. Boyle, R.W. Foster, and D.M. Good 1990 Airway Smooth Muscle: Electrophysiological properties and behaviour. In: Airway Smooth Muscle: Modulation of Receptors and Response. D.K. Agrawal and R.G. Townley, eds. CRC Press, Boca Raton, pp. 69-94. Somlyo, A.V., M. Bond, A.P. Somylo, and A. Scarpa 1987 Inositol triphosphate-induced calcium release and contraction in vascular smooth muscle. Proc. Natl. Acad. Sci. USA, 82t523. Stephens, N.L., C.S. Packer, and S.K. Kong 1985a Smooth muscle contractility: Effects of hypoxia. Chest, 88~223s-229s. Stephens, N.L., E.A. Kroeger, U. Kromer, and J.A. Mehta 1969 ForceI I . I velocity characteristics of respiratory airway smooth muscle. J. Appl. Physiol., 26:685-692. Stephens, N.L., G. Morgan, C.S. Packer, and S.K. Kong 1985b Smooth muscle contractility: Effects of hypoxia. Chest, 88st2235-2295. von Hayek, H. 1960 The Human Lung. Hafner Publishing Company, New York. pp. 127-226. DR. HOYT: Dr. Stephens, I wonder if I might lead you out onto the ice and ask you to make a prediction? From what we’ve seen in the last two days, I think most people would agree that neuroepithelial bodies, no matter whether they are autonomous or whether they are innervated, are sited so as to release the bulk of their granules across the basal cell membrane into the peribronchial connective tissues. And of course we’re interested in knowing what their targets are under normal circumstances. There has been mention made of adjusting ventilation and perfusion and of the possibility that there may be different functions for cells a t different positions. Can I ask you, with your knowledge of the bronchial muscle and airflow patterns, to tell us where you would site endocrine cells to adjust airflow significantly in the lungs to match perfusion? In other words, are there critical places along the bronchial tree where an effect on the muscle would be most striking and produce the desired result? DR. STEPHENS: Well, with respect to optimizing ventilation, I would put them at the 3-5 millimeter diameter airways in the human being. The other issue that you raised was the matching of ventilation and perfusion. That’s a little more difficult. That’s going to occur at the alveolar level. And that’s going to be either perhaps controlled by the neuroendocrine system, as I had mentioned earlier on in this meeting, or by a group of cells described by Kapanci (Kapanci et al., 1974), who called them interstitial cells. They have contractile filaments in them, and he figured that these made the final fine adjustments for matching ventilation to perfusion. DR. JOHNSON: This will sound extraordinarily naive, but can you make any generalizations about hy- 164 WORKSHOP ON PULMONARY NEUROENDOCRINE CELLS pertrophied airway smooth muscle in terms of its mechanical properties? DR. STEPHENS: Yes. Clinically, by the time the patient comes in there is almost certainly some hypertrophy and perhaps some hyperplasia, and what’s happening to the mechanical properties, of course, you can assess by doing a time study. We don’t have the hard information on airway smooth muscle, but we know from high blood pressure that if you start early enough the first change you see is increased shortening and increased velocity, with no change in force development. Now, if you wait until later, when hypertrophy is being developed, force development increases, so you now have increased velocity, increased shortening and increased force development. If you wait even longer, the cytoskeleton starts to hypertrophy, and one of the most striking things that occurs in high blood pressure is that the whole cell gets simply stuck to the cytoskeleton such that the actomyosin gets reduced and, a t that point, force development drops, velocity drops, and shortening also drops. So there seems to be a whole progression of these various modalities of function. GENERAL DISCUSSION: FUTURE DIRECTIONS DR. SOROKIN: We’ve saved a little time for discussion of what we think are important areas for study, given our current knowledge about pulmonary neuroendocrine cells. The National Heart, Lung, and Blood Institute is also interested to know what kinds of things we recommend for future research in this field. Dr. Polak has a number of suggestions which she will offer presently. Meanwhile, I’ll just say that for me the outstanding characteristic of this Workshop has been that people trained in very different research disciplines were able to come together and present findings that stimulated us all and enriched our understanding of the subject. Hence I believe that for future studies on the pulmonary neuroendocrine system all sorts of approaches should be encouraged, because no single discipline or approach has so far had a monopoly on significant new findings. DR. HOYT: In other words, you think that the door should not be slammed on any particular technique just because it’s two years old. Is that the idea? DR. SOROKIN: Exactly. DR. CUTZ: Well, I think what has transpired, a t least from my perspective, is that there is a consensus now that the function of these cells perhaps varies with development. In other words, they may function during intrauterine life and postnatal life, so that studies should still focus on both those areas. DR. SOROKIN: On a biphasic manner of function in these two periods. Yes. DR. AGUAYO: I think one of the important things is that we try to reach a consensus about nomenclature. I don’t believe anybody here is confused about what is a neuroendocrine cell, but in terms of making this literature accessible to other people interested in getting into the field, perhaps we could devise some kind of common nomenclature so that we all mean roughly the same thing when we’re talking about neuroepithelial bodies, single neuroendocrine cells or whatever. The same is true with respect to the roster of neuropeptides or markers that are going to be used to define what is a neuroendocrine cell. I think everybody agrees on what is the morphology and the location and the specific characteristics of these cells. But when we report and discuss our findings, we have to be very clear about what type of marker we’re using and about recognizing the limitations of such markers versus others that you may or may not want to use. It’s also very important that we start recognizing, as Dr. Battey was saying, the heterogeneous families of peptides and mediators and receptors. When we start reporting our data we should try to be a little bit more specific about what is bombesin-like immunoreactivity, for example, and not say, “This cell makes bombesin.” If you don’t have mRNA present in the cell you have to be careful about saying these cells produce the peptide, a s opposed to having peptide immunoreactivity. Attention to such small details in nomenclature and definitions should facilitate our ability to exchange information and reach consensus. DR. SCHEUERMANN: I think it is important that we continue receptor research on these cells and also make in situ hybridization a high priority. DR. HOYT: I’d like to go back to what Sergei has said about approaches. We now see that it’s possible to isolate these cells and deal with them under controlled conditions. Dr. Cutz and others clearly show us that now, at last, we’ll be able to get at some of their receptor biology and electrophysiology. That’s fine. But I wouldn’t like the powers that be to get the impression that this should be the only emphasis. As always, we have the whole animal before us, and so we still need a broad approach. These cells are not all that easy to find. They’re not all that easy to pin down. But by their particular siting, they may perform different functions, and so we have to push both aspects a t once, in vivo and in vitro. I also think we should talk together more often. DR. POLAK: Well, following what you’ve said, I’d like to say this Workshop has been really a n excellent idea on Dorothy Gail’s part. As Sergei pointed out, there hasn’t been a meeting of this kind for more than 10 years, and in fact most of the faces are new. Not new in terms of age, but new to me. I haven’t met them. In fact, some of the people said that it was very nice to come and meet the oldest-I think they thought I might be 70 and in a wheelchair. It’s absolutely necessary that we try, if possible, to meet at regular intervals. Not too frequently, of course, because there are far too many meetings. But something like every 2 years. Since the initiative of 10 years ago, we haven’t met, and lots of developments have occurred. One of the biggest diseases of this century is lack of communication. I am sure in the next two years we will see, following this meeting, a n enormous amount of data being generated. From talking to various people, I have the impression they’re all going back enthusiastically to continue their work and to answer a number of the questions raised here. I see two points that need special attention. One relates to technology and one to the questions we all want to answer. In terms of technology, over and above classical histological methods and immunocytochemistry we need to consider new means of quantifying these cells in a reliable way. The idea we had in the early days, that biochemistry can give us quantitative answers, is not WORKSHOP ON PULMONARY NEUROENDOCRINE CELLS enough. We need to know precisely where cells and molecules are in the tissues. But of course there are problems in that as well. We have very gently skimmed over molecular biological approaches to the question of hormone receptors, and as Dr. Scheuermann said, these have been little used in the lung. And so in terms of technology, we’ve got it. It’s with us, but we need to apply it. In terms of questions and answers, it is clear, as has been pointed out all along, that looking a t these cells in isolation is extremely useful because i t simplifies a number of things, but we cannot extrapolate the data completely to what happens in vivo where there are lots of nerves and where we now have regulatory factors in the endothelium. We need to relate cellular mechanisms to the structure of the tissue. So I would expect we will have meetings where the innervation as well as endothelial and vascular aspects will be thoroughly discussed. DR. HOYT: I think the innervation is absolutely critical. DR. POLAK: Yes. I think so. We can’t discuss these cells in isolation. Many times in this meeting we have asked what will happen to the innervation as well as the endocrine cells. We can already recommend to the Institute more frequent meetings and use of novel technology to study the pulmonary neuroendocrine system in its integral relations with nerves and the vasculature. The use of genetic models has been mentioned in a number of instances. These horrify me because, of course, they are complex models. But I think they will play a n important role in answering certain questions. Again, we cannot simply extrapolate from them to what happens in a genetically normal animal or human being, but these techniques can be helpful. There aren’t very many models as yet, and very little has been investigated, practically nothing. DR. HOYT: Something that has emerged over the last two days is a desire to know more about the mechanisms that induce formation of the endocrine cells and that maintain their population. This is a difficult question that will require not only work in vitro but also studies in situ, in the tissues. It is extremely important: if you can understand that, then I think you’re on the verge of understanding what happens in the lungs in many situations where there has been a pathologic challenge. DR. POLAK: I think you are saying in a very kind way that you don’t think we’ve been intellectual enough in this meeting. DR. HOYT: No. It’s just that many people are working towards this problem from different approaches, and yet we still know virtually nothing about it. I think this is the central event that links the normal development and normal functions of pulmonary neuroendocrine cells with clinical and pathological disorders. DR. POLAK: Phenomenology is useful, but it’s sort of, “Gee whiz, how exciting!” We need to put it to work, and not just in trying to understand development. I think Elizabeth-is she there? DR. McDOWELL: I’m here. DR. POLAK: She brought up something that we must not forget. She asked, “Can you look a t the airway epithelium in its integrity?” and I’m making a 165 plea that we look at the whole lung in its integrity. It’s really necessary if we want to interpret, let’s say, pulmonary hypertension and hyperplasia of endocrine cells, to know what happens with the oncogene expression in the early phases, and-again I’ll come back to novel technology-what is the receptor, is it up or down regulated, et cetera. We need, a s Dr. Hoyt has said, to try to interpret the mechanisms, which is the most difficult thing to do. But I think it goes two ways. If you can understand what’s happening in pathology you then begin to unravel what could happen in normal circumstances. DR. HOYT Yes. I think they’re absolutely intertwined. DR. POLAK: The way this Workshop has been organized is very good in that we have a mixture of people with different knowledge. I mean, who would know what Frank Cuttitta knows? He’s made a fantastic contribution; physicians and pathologists are defining relevant clinical entities, etcetera, etcetera. So I think we need this multidisciplinary approach, but let’s not wait 10 years for another meeting. I, myself, I’m in a hurry. (Laughter.) DR. CUTZ: Well, I’d like to come back to normal functions. It’s important to decide whether there’s enough evidence to say, for example, that the neuroepithelial body is a chemoreceptor. Or to say that i t has some function regulating growth and differentiation. Or that it functions only during development and then gets switched off. Or that i t can be switched off and switched on again later. If you can decide this is a chemoreceptor in the same class as the carotid body, then you can study i t along those same lines. Now, how do you prove something is chemoreceptor? You can use classical physiology or you can bypass it looking a t the membrane properties, and this is why you need isolated cells. It helps you get around the difficulty that this is a widely dispersed system. You can’t simply remove i t for study as you can the pituitary or the carotid body, but if you are able to characterize the membrane properties of the cells in isolation, you can compare them with those of better known systems. DR. SCHEUERMANN: The functional histological approach is very important also. Perhaps we are thinking about a n electrophysiological approach to the nerve endings near these cells as one of our aims? DR. SOROKIN: I think there’s a clear case for studying these cells in isolation to get the sort of information Dr. Cutz has mentioned. Since we are dealing with neuroepithelial bodies, however, it’s equally clear that to get the full picture we have to understand how they’re related to the nerves. It seems that patterns may differ from one species to another, perhaps along Cartesian coordinates of afferent and efferent neurites interfaced with neuroendocrine cells. There are still huge gaps in our detailed knowledge that must be repaired if we are fully to understand the role of these cells in chemoreception. DR. POLAK: Yes. DR. LINNOILA: Well, I certainly agree with all of this. But in addition to methods I think it would be useful to note currently the functions that are postulated and half-proven, those for which there is some kind of intellectual basis, or those actively being 166 WORKSHOP ON PULMONARY NEUROENDOCKINE CELLS worked on. We also mentioned that it would be helpful for other researchers if we tabulate all the peptides, amines, and the like, and tabulate the functions as well. It might be useful conceptually to put the fetal and newborn functions in one list, adult functions in another, and maybe then the pathologies. Just create lists of functions. One would be the chemoreceptor function, but the question is whether we would put chemoreception in all the lists. DR. POLAK: I think you are very optimistic. Ideally one should define them broadly as chemoreception and then growth and development. Otherwise you will list control of smooth muscle, control of endothelium, vascular and airway smooth muscle, epithelium, blood gas levels. The whole lot. Then, as you were saying, you have to think which one is real and point a n accusing finger a t the others. The main divisions which appear to be the most solid ones are growth and development and tissue repair and chemoreceptor function. The nitpicking of whether they are vasodilators or vasoconstrictors and whether endothelin is-it’s impossible right now! DR. LINNOILA: Yes. It may be impossible. But if we also want t o be of service to the outside community, then listing some of these functions is useful. And if we can attach a reference, or abstract, or something of that kind, it would be even better. I agree, you know, that it may be a n impossible task to list all these functions. DR. SCHEUERMANN: But all the functions are not known momently. DR. CUTZ: Perhaps those are hypotheses? DR. HOYT: Some we’ve begun to test, some have not been tested, and some have been tested a little bit and found to be either too difficult to deal with right now or perhaps-DR. POLAK: Well today a t lunch time some of us were discussing the paper published on the effect of bombesin on human bronchial epithelium. There haven’t been many papers on the subject, and when people have tried to reproduce these results they’ve encountered difficulties. But it was such a predictable answer. It fit so nicely with the theory that people do quote it. I mean, i t may be true but i t needs [confirmation]. As you said, at the moment the functions need a little more work, whereas we can start building up a table on peptides in the different species. DR. HOYT: I think Ernest is right. We should perhaps, a t this stage, not become hung up because we don’t find a given peptide in all species, or because we have evidence for a function in one species but not yet in another. The first priority ought to be to figure out what these cells do. To me it’s obvious they’re going to do different things a t different times of life. It’s not a n accident that they appear soon after the airway is laid down, a very long time before birth, and they’re well recognized in adult life where their morphological and physiological circumstances are quite different. DR. SUNDAY: I was going to say that it’s just by the nature of the organ that we’re probably several decades behind fields like hematology or even pituitary cell biology, where you can get normal and abnormal tissue a t any time you wish. As Dr. Polak was saying, a lot of this is sort of nonintellectual; I think she means nonmolecular, in that a lot of this(Laughter.) DR. POLAK: What I meant was, we could go on and on in a complacent way, saying, “Oh, gee whiz, this is fascinating. They twitch on the left and they twitch on the right.” And remember electron microscopy-well you don’t, you weren’t born-but electron microscopy, just describing and describing and describing. DR. SUNDAY: But the descriptive work is extremely important and will lay the foundation. It’s not completed. It would be very helpful if we could make a chart with the different peptides in different species, potential roles, and, as Dr. Cutz says, antibody sources-which are good antibodies, which are bad antibodies. It would be like the codification of human leukocytic antigens. Common reagents can be very important in trying to understand what’s going on. DR. POLAK: But I’m sure if we tried to determine the possible roles of pulmonary endocrine cells we’ll conclude that we need to do a basic, systematic, histological analysis of a species. DR. SUNDAY: Well, you know, now a lot of the molecular work is going back to morphology. DR. POLAK: Exactly. DR. SUNDAY: But I say, do the descriptive work and then perhaps in two years we’ll have more ideas, not just about molecules that are involved in pathways, but cell to cell interactions, like what goes on between the neuroendocrine cells and the nerve endings and the epithelial cells. DR. POLAK: Exactly. DR. SUNDAY: And across muscle to fibroblasts. DR. HOYT: What about gap junctions? DR. SCHEUERMANN: The membrane biology of these cells? But I think also our attention must be, should be, focused on the relationship of these neuroendocrine cells to the autonomic nervous system. It should be included in this large field of research. DR. HOYT: Well, the source of the innervation is still a n open question. Allusions have been made to possible sensory feedback loops between the endocrine cells and the intrinsic nervous system in the lung. Their innervation shouldn’t simply be viewed as vagal afferents with local axon reflexes. Take the heart. I remember seeing a poster some years ago a t a meeting in Montreal. There was a picture of a muscle spindle in the myocardium, and the fiber from that muscle spindle very definitely synapsed on a cardiac ganglion cell. That’s a good example of local sensory feedback, not upstairs but right a t the local level. It isn’t just a matter of what the endings are like and how they relate to the endocrine cells; you have to have the rest of the wiring diagram. DR. SCHEUERMANN: But perhaps we must also focus our attention on the relationship between these neuroendocrine cells and the chromaffin cells in the ganglia located in the connective tissue. I think it is necessary. DR. CUTZ: Obviously when you talk about a receptor you have to know its innervation. That’s fundamental. Since you know the innervation of the carotid body you can manipulate it. I would agree 100 percent that the innervation of the endocrine cells has to be worked out; the species differences have to be worked out. There are some wonderful neuroanatomical methods which Dr. Polak uses, retrograde tracing and these kinds of things, which could be easily- WUKKSHUI‘ VN Y U L M O N A K I DR. HOYT: Hallmark preparations. DR. CUTZ: Right. And I don’t think this has been done in the lung, and it would help you to understand the connections. By no means do you get all the answers looking a t the cells. It’s the whole receptor organ. DR. POLAK: If you do as you suggest-retrograde tracing, which is so important and very little investigated-and use confocal microscopy, how do we get Dorothy to fund all that? DR. GAIL: I’ll try. I need your ideas. DR. POLAK: No problem with ideas. The cash is missing. DR. SUNDAY: Well i t seems that animal models would also be very welcome. Dr. Polak described the emphysema model, but there may be many models that nobody has analyzed. There are a lot of inbred strains of mice, and maybe some of these have abnormalities in the neuroendocrine cells, or there are a lot of transgenic mice with various homozygous deletions. DR. GOSNEY: The thing that strikes me sitting here these past two days is that there is no real justification for a revolution. I think that we’re all working away a t different aspects of the problem, most of us making some progress and each contributing in a different way. I think it would be a shame if anybody felt that what they were contributing was somehow redundant in the major task of sorting all this out. DR. POLAK: Speak for yourself. (Laughter.) DR. GOSNEY: No. I mean I feel, personally, speaking for the other people with a pathological bent, the amateur scientists amongst us, it’s a question of more of the same but with greater care, perhaps using some of the new technology, more in situ hybridization of receptors and so on and so forth. My own feeling is that I should be going back, for instance in these pulmonary hypertension cases, looking a t them more carefully, reexamining them, doing things that have been suggested, teasing them out more and more. But I don’t think there is any justification for a huge revolution. Let’s not put all our eggs in one basket. I think we all need to carry on. DR. POLAK: I think the important thing is, as you said, to keep your identity. It’ll be foolish if you become a cell biologist and, you knowDR. GOSNEY: I couldn’t do it. (Laughter.) DR. POLAK: I think we all play a very key role. DR. GOSNEY: I agree. I think one of the interesting things has been how many differentDR. P0LAK:-Aspects. DR. GOSNEY: Yes. DR. SUNDAY: The multidisciplinary approach is very important. DR. POLAK: The problem is how do we cure the jet lag? That’s the big problem. (Laughter.) DR. GAIL: Sunshine, isn’t it? DR. SOROKIN: In a way we have solved the problem. If Dr. Polaks suggestion to meet more frequently is taken up, then the multidisciplinary approach makes a lot of sense. When we sit down and talk together it’s much clearer what each has to contribute. DR. POLAK: As I said before, I think the lack of communication is a big disease. DR. GAIL: Thank you all very, very much. DR. POLAK: Thank you. DR. GAIL: Thank you, Sergei. (Applause.1 (Whereupon, a t 3:lO p.m., on September 6, 1991, the workshop concluded.) WORKSHOP CONSENSUS: CURRENT STATE OF KNOWLEDGE A consensus of the Workshop was that pulmonary neuroendocrine cells are biphasic in function, having: (1) a “mitogenic” role i n controlling the growth o f the airway epithelium, particularly during the morphogenesis of the bronchial tree i n prenatal life, and (2) a chemoreceptiveleffector role that predominates in postnatal life and appears directed to several target cells, especially airway smooth muscle. New evidence tends to confirm their role as chemoreceptors for oxygen; their mitogenic role may be reasserted postnatally as the population undergoes hypertrophy in regions o f the lung severely damaged by certain lung diseases (e.g., bronchopulmonary dysplasia, cystic fibrosis, and among a damage-susceptible subgroup o f cigarette smokers), accompanied by developing fibrosis, epithelial metaplasia, and airway and vascular smooth muscle hypertrophy. These functions are associated with secretion of regulatory peptides that the cells synthesize (bombesinlGRP, calcitonin, CGRP, etc.) and store i n cytoplasmic granules, although the spectrum o f peptides synthesized is not the same in all species. Many pulmonary neuroendocrine cells are elaborately innervated by visceral sensory and autonomic fibers which play a role i n integrating chemosensory responses; work on endocrine cellenriched cultures, however, indicates that i n the absence o f innervation, cells will degranulate in response to hypoxia, and during lung formation noninnervated cells appear responsible for growth regulation. Apart from some 10 regulatory peptides known to be associated with these cells in different species, pulmonary endocrine cells likely contain, or can synthesize, a number of additional, hitherto unrecognized ones, O n the one hand, electron microscopic immunocytochemistry indicates that no regulatory peptides have yet been found for certain morphological subtypes of these cells existing in the lungs; o n the other, neuroendocrine cells are provided with specific enzymes of the peptidyl glycine alpha-amidating monooxygenase ( P A M ) group, which are capable o f making biologically active peptides from larger prohormones or molecules like chromogranins, present in the secretory granules. Once released from the cells, these substances gain ready access to lung tissues and act mainly at specific local sites, although a select subunit of pulmonary endocrine cells may reach more distant targets because of especially favorable access to the circulation. Receptors for some of the peptides secreted by these cells have been found i n the lungs, but much of the critical work i n this area has been done o n cultures ofpulmonary small-cell carcinoma cells, and the picture in the intact lung is yet far from clear. From evidence of bombesiniGRP receptors presented at the Workshop, however, it appears likely that pulmonary cells possess both high- and low -affinity receptors for a given peptide, and that control over the effects of the neuroendocrine cell peptides also occurs at the receptor level. 168 WORKSHOP ON PULMONARY NEUROENDOCRINE CELLS RECOMMENDATIONS 1. More work is needed to characterize the “mitogenic” and “chemoreceptive” roles of pulmonary neuroendocrine cells. What are the normal times of onset and duration of these functions, do they overlap, and are they carried out by the same cells or by distinct subpopulations? 2. Investigation should be encouraged into factors or mechanisms regulating pulmonary neuroendocrine cell population size in normal and diseased lungs. What causes their hyperplasia after certain types of lung injury, and does this cause renewed growth of the lungs, or is it only a consequence of the repair process? 3. New openings in research should be followed up. As presented a t the Workshop, these included: (1) a role for pulmonary neuroendocrine cells in regulation of lung development, ( 2 ) improved ways to isolate the cells for experimental study, and (3) experimental studies on the response of cells to hypoxia (and other) stimuli, the mechanism of secretion, and stimulus-secretion coupling, now made more feasible by improvements in technology. Some of these studies can be performed on neuroendocrine cell-enriched cultures or lung organ cultures, but related matters can also be investigated in intact lungs through application of newer microscopic techniques and analytical methods (immunocytochemistry, mRNA probes, receptor localization, and cell blot assays for secretion), enhanced by confocal microscopy. 4.Further studies are needed to identify new bioactive peptides in pulmonary neuroendocrine cells and the mechanism of their synthesis, since the full range of peptides that these cells are capable of producing is not fully appreciated. Furthermore, it is not known to what extent the cells normally are capable of shifting production from a prevalence of one type of peptide (e.g., for regulating growth) to another (e.g., for influencing smooth muscle contraction), or whether autocrine mechanisms may be involved in effecting the change. 5. Comprehensive studies are needed on the relationship among pulmonary endocrine cells, the central and peripheral nervous system, and likely local target cells like bronchial epithelium or smooth muscle, in order to understand how all of these elements interact to promote regulation of the lungs. These should include investigations into the “wiring” pattern of neuroepithelial bodies and the nature of synaptic connections present and neurotransmitters within them, as well as the location of receptor proteins for transmitters and/or endocrine cell peptides on target cells and on the endocrine cells themselves. Current work has not yet opened up this area, but this can be anticipated for the near future. 6. In relation to the preceding, specific information is needed about the molecular biology of peptide receptors, together with their binding affinities, in the intact lungs of human beings and laboratory animals. 7. Because of its diffuse distribution and activity predominantly a t small focal sites along airway branches, the pulmonary neuroendocrine system has been very difficult to study. In consequence, much of what is known about its function necessarily has been obtained by indirect, idiosyncratic, time-consuming, and occasionally ingenious procedures rather than by more conventional physiological research methods. The presentations and discussions at the Workshop made i t clear to the participants that our current understanding about these cells is based on contributions of investigators using very diverse approaches. Continued use of a variety of quantitative methods, microscopic techniques, and molecular biologic methods to relate structural features of pulmonary neuroendocrine cells to their biological functions should be encouraged. 8. In view of the currently increased interest and pace of work in this field, Workshop participants recommend that future meetings of a similar kind be held a t approximately two year intervals.