Mucus and mucosa

Ciba Foundation symposium 109

1984

Pitman London

Mucus and mucosa

The Ciba Foundation is ;in international scientific and educational charity. I t wits established in 1947 by the Swiss chemical and pharmaccutical company of CIBA Limited. -now CIBA-GEIGY Limited. The Foundation operates independcntly in London under English trust law.

The Ciba Foundation cxists t o promote international cooperation in hiological. medical and chemical research. I t organizes about eight international multidisciplin topics that wem ready for disc .ion by ;I small group of' research discussions arc published in the ha Foundation bymposium series. Thc t.oundution also holds many shorter meetings ( n o t published). organized by the Foundation itsell o r by outside scicntific organizations. The staff always welcome suggestions for future meetings.

The Foundation's housc at JI Portland Place. London W1N J R N , provides lacilitics lor a11 the meetings. Its library. open seven days a week to any gr;iduate i n sciencc or medicine, also provides information on scicntilic meetings throughout the world and answers general enquiries o n biomedical and chemical suhjects. Scientists from any part o f the world may stay in the house during working visits to London.

Mucus and mucosa

Ciba Foundation symposium 109

1984

Pitman London

0 Ciba Foundation 1984

ISBN 0 212 79783 9

Published in November 1984 by Pitman Publishing Ltd. 128 Long Acre, London WC2E 9AN. U K Distributed in North America by Ciba Pharmaceutical Company (Medical Education Division). P.O. Box 12832, Newark. NJ 07101. USA

Suggested series entry f o r library catalogues: Ciba Foundation symposia

Ciba Foundation symposium 109 ix + 246 pages. 40 figures, 23 tables

British Library Cataloguing in Publication Data

Mucus and mucosa.-(Ciba Foundation symposium; 109) I . Mucuh I . Nugent. Jonathan 111. Serieh 61 l’.Ol81 OP215

11. O‘Connor. Maeve

Printed in Great Britain at The Pitman Press. Bath

Con tents

Symposium on Mucus arid mucosa, held at the Ciba Foundation, London, 28 February-I March 1984

Editors: Jonathan Nugent (Organizer) arid Maeve O’Connor

A. Silberberg Introduction 1

C. B. Basbaum trachea 4 Discussion 1.5

Regulation of secretion from serous and mucous cells in the

M. R. Neutra, T. L. Phillips and T. E. Phillips Regulation of intestinal 20 goblet cells in situ, in mucosal explants and in the isolated epithelium

Discussion 29

S. J. Coles, K. R. Bhaskar, D. D. O’Sullivan, K. H. Neil1 and L. M. Reid Airway mucus: composition and regulation of its secretion by neuropeptides in vitro 40 Discussion 54

J. Forstner, N. Roomi, R. Fahim, G. Gall, M. Perdue and G. Forstner Acute and chronic models for hypersecretion of intestinal mucin 61 Discussion 68

T. F. Boat, P. W. Cheng, T. D. Klinger, C. M. Liedtke and B. Tandler Proteinases release mucin from airways goblet cells Discussion 84

72

General Discussion Regulation of mucus secretion 89

G. Flemstrom and A. Garner Some characteristics of duodenal epithelium 94 Discussion 105

J. H. Widdicombe Fluid transport across airway epithelia 109 Discussion 117

V

vi CONTENTS

J. R. Clamp and J. M. Creeth Some non-mucin components of mucus and their possible biological roles 121 Discussion 13 1

A. Allen, D. A. Hutton, J. P. Pearson and L. A. Sellers Mucus glycoprotein 137 structure, gel formation and gastrointestinal mucus function

Discussion 15 1

I. Carlstedt and J. K. Sheehan Macromolecular properties and polymeric structure of mucus glycoproteins 157 Discussion 166

M. Elstein and G. M. Fawcett Effects of the anti-oestrogens, clomiphene and tamoxifen, on the cervical factor in female infertility 173

E. N. Chantler and P. R. Scudder Terminal glycosylation in human cervical mucin 180 Discussion of the two preceding papers 188

M. Litt Comparative studies of mucus and mucin physicochemistry 196 Discussion 206

P. Verdugo Hydration kinetics of exocytosed mucins in cultured secretory cells of the rabbit trachea: a new model Discussion 222

212

Final general discussion Mucus swelling, secretion and effects on cilia 226 Epithelial preparations 229

A. Silberberg Closing remarks 235

Index of contributors 237

Subject index 238

Participants

A. Allen Department of Physiological Sciences. University of Newcastle upon Tyne, Medical School, Newcastle upon Tyne NE1 7RU, UK

C. B. Basbaum Department of Anatomy and Cardiovascular Research Institute, 1315-M, University of California-San Francisco, San Francisco, CA 94143, USA

J. Bilski (Ciha Foundafion Bursar) Institute of Physiology, Nicolaus Coper- nicus Academy of Medicine, 31-531 Krakow, Grzegorzecka 16, Poland

T. F. Boat Department of Pediatrics, University of North Carolina at Chapel Hill, Medical School, Burnett-Womack Clinical Sciences Bldg 229H, Chapel Hill, North Carolina 27514, USA

I. Carlstedt Department of Physiological Chemistry 2, University of Lund, P 0 Box 750, S-220 07 Lund 7, Sweden

E. Chantler Department of Obstetrics & Gynaecology, University of Man- Chester, University Hospital of South Manchester, West Didsbury, Manchester M20 8LR, UK

J. R. Clamp Department of Medicine, University of Bristol, Bristol Royal Infirmary, Bristol BS2 8HW, UK

S. Coles Department of Pathology, Harvard Medical School, Children’s Hospital Medical Center, 300 Longwood Avenue, Boston, MA 02115, USA (Present address: Medical Department, Abbott Laboratories, Queen- borough, Kent ME11 5EL, UK)

J. M. Creeth Department of Medicine, University of Bristol, Bristol Royal Infirmary, Bristol BS2 8HW, UK

M. Elstein Department of Obstetrics & Gynaecology, University of Man- Chester, University Hospital of South Manchester, West Didsbury, Manchester M20 8LR, UK

vii

PARTICIPANTS ...

Vl l l

G. Flemstrom Department of Physiology 8( Medical Biophysics, University of Uppsala, Biomedical Center, P 0 Box 572. S-751 23 Uppsala. Sweden

G. Forstner Department of Paediatrics, Physiology, University of Toronto, Hospital for Sick Children, Research Institute, 555 University Avenue, Toronto, Ontario M5G 1x8, Canada

J. Forstner Division of Biochemistry, Hospital for Sick Children, Research Institute, 555 University Avenue, Toronto, Ontario M5G 1x8. Canada

A. Garner Bioscience Department, Pharmaceuticals Division, ICI plc, Mereside, Alderley Park, Macclesfield, Cheshire SKlO 4TG, UK

P. W. Kent Nuffield Department of Clinical Biochemistry, University of Oxford, Radcliffe Infirmary, Oxford OX2 6HE, UK

M. Litt Department of Bioengineering, School of Engineering and Applied Science, University of Pennsylvania, 285 Towne/D3, 220 South 33rd Street, Philadelphia, PA 19104, USA

C. Marriott Department of Pharmacy, Brighton Polytechnic, Moulse- coomb, Brighton, Sussex BN2 4GJ, UK

J. A. Nadel Cardiovascular Research Institute, 1315-M, University of California-San Francisco, San Francisco, CA 94143, USA

M. Neutra Department of Anatomy, Harvard Medical School, 25 Shattuck Street, Boston, MA 02115, USA

L. M. Reid Department of Pathology, Harvard Medical School, Children's Hospital Medical Center, 300 Longwood Avenue, Boston, MA 021 15, USA

P. S. Richardson Department of Physiology, St George's Hospital Medical School, Cranmer Terrace, Tooting, London SW17 ORE, UK

P. Roussel Department of Biochemistry, Protein Unit, INSERM, (No. 16), Place de Verdun, 59045 Lille Cedex, France

A. Silberberg (Chairman) Department of Polymer Research, The Weiz- mann Institute of Science, IL-76100 Rehovot, Israel

PARTICIPANTS ix

P. Verdugo Center for Bioengineering, WD-12, University of Washington School of Medicine, Seattle, Washington 98195, USA

J. G. Widdicombe Department of Physiology, St George’s Hospital Medical School, Cranmer Terrace, Tooting, London SW17 ORE, UK

J. H. Widdicombe Department of Physiology and Cardiovascular Research Institute, 1315-M, University of California-San Francisco, San Francisco, CA 94143, USA

Introduction

A . SILBERBERG

Department of’ Polymer Keseurch. The Weizrriurin Iti.siiiuie of Science, 76/00 Rehovor. 1,srueI

It is an honour and pleasure to have been asked to preside at this meeting. Let me emphasize a few points. Firstly, in general terms, what we would like

to achieve here is the generation of ideas - ideas that will not necessarily be fully expressed in the printed proceedings, but that each of us can take back and attribute to this symposium. However, if we have a good and fruitful discussion, it may help to inspire those who later read the book.

Secondly, during the last few months, I have asked most of the participants for their views on what the main topics of our discussion should be - not so much their own personal interests, but what they felt could be usefully discus- sed by this group. There was a strong consensus that the link between the structure and function of mucus should be discussed. Knowing the one would help us to understand the other, or to generate ideas about the other. So this, I hope, will be one very important focus of our discussions.

It is intriguing, in this connection, that most of the papers do not address function directly, but only indirectly. Structure is more kindly dealt with on the whole. Here, of course, there is agreement. There is little doubt today that mucus is a loose gel involving a network composed of a typical, repeating glycoprotein moiety. These units are held together by two kinds of cross-links. One kind of link strings the glycoprotein moieties into a chain, and these may be very long chains - a polymerization type of interaction. The other kind of link ties the polymerized strands together into the network, i.e. establishes the gel-forming cross-links. We know very little about either of these links, howev- er. They may, indeed, be of many kinds. In particular, they may or may not be covalent. They may also involve other macromolecules. But certainly for the gel to possess the functional properties it has, such bonds must have life-times of minutes or seconds at least. These are long times in relation to the life-time of a normal secondary bond, so we must expect that if the cross-links are secon- dary, they must be endowed with very special properties. For example, they

1984 Mucus and mucosu. Pitman, London (Ciha Foundulion symposium 109) p 1-3

1

2 SILBEKBERG

may be sterically so hemmed in that separation of the strands is difficult to achieve in a short time. Alternatively, a number of secondary bonds may be involved simultaneously i n a link, so that bond breakage can be achieved only by removing them all at the same time, effectively extending the strength and thus the lifetime of the bond.

The nature of the bonds represents the core of the problem of characterizing the overall structure. As we have no good way of getting at thc bonds, at least not at present, our approach has had to be indirect. That is, we have been looking at the overall structure and chemistry of the glycoprotein molecules and have tried to infer how interactions can be produced, of what kind these interactions could be, and under what conditions they could arise.

At the same time, we have been looking at how the glycoprotein moieties are synthesized and secreted. We have, in fact, more easily been able to look at this aspect of secretion than at the overall mechanical and physical-chemical properties of the product as it is made. The suspicion has been raised, indeed. that the secretory product that we can collect and study is not necessarily in the form in which it was synthesized, stored and secreted by thc cell. There are factors influencing the product, between its synthesis by the cell and its final appearance in the region of application, that regulate the form in which the mucus is ultimately utilized. These purposes definitely vary.

Why then, indeed, do epithelia whose functions are clearly very different secrete and utilize a glycoprotein material of very similar basic chemistry and similar primary structure? The clue to that probably lies in the varied effects that can be achieved by modest modifications of such weak-gel-forming macro- molecules as mucus glycoproteins. Few systems are required t o do such diverse things in the biological environment as mucus. There is seemingly a world of difference between clearance from the lung and filtering out of low mobility sperm in the cervix, between clearance from the middle ear and the control of gastrointestinal pH. A priori, these functions are not linked at all . and it seems strange that the same material, with modest structural modifications, should have been selected to do these different things. Gels, however, have many more features to them than the rheological, viscoelastic aspects that we tend to see immediately. One of their strong features is their ability to convey mecha- nical messages at long distances. They are macroscopically cross-linked sys- tems that can extend their influence over a much greater distance than a macromolecular dimension. Gels have the capacity to filter or separate mole- cules, to immobilize molecules and to control their movement in a manner quite different from that associated with an open, unimpeded fluid space. Gels are thus very useful for manipulating and regulating events on the molecular, macromolecular and colloidal scale. Mucus, as a gel of the appropriate con- sistency, is consequently endowed with the properties required to fulfil a cell-protective mission along paths which are now, at least in part, fairly clear to

INTRODUCTION 3

us. Mucus, as a gel of altered character, can correspondingly be associated with a failure of function, i.e. with cellular malfunction and with disease.

The functions that we have s o far identified in various epithelia may be only part of the story. These epithelia probably fulfil other roles that we have not yet fully recognized and analysed. It is thus important to consider all the aspects of organ and epithelial function and to aim at a full understanding of what is happening physiologically and biochemically in the epithelium. It is too much to hope that when we have worked out the biophysics and biochemistry of mucus all the other questions and their answers will automatically fall into our

Several other topics were also stressed t o m e in my preliminary discussions. One was that mucus contains many substances in addition to mucus glycopro- teins and that these should now be looked at and their function investigated. The role of mucus as a gel is clearly of importance here. As a gel, mucus can bind these components or hold them in place and release them when and where they are needed. By the same token, mucus can also sequester molecules and remove them from the environment. The significance of materials contained in mucus may thus, in part, lie with what they cannot do , i.e. are prevented from doing, under these circumstances.

The great importance of trying to understand what happens to mucus and mucosa when a tissue becomes diseased was also pointed out to me. How is mucus structure affected? How is mucus production affected? And to what extent are the mucus glycoproteins modified'?

Finally, in my current list, is the question of water management. Mucus is probably synthesized and stored in granules at concentrations of the glycopro- tein which are quite different from the final concentrations. Water has thus to be added to the secretion, and the amount added may indeed be the method of regulation which determines whether the mucus will function properly at its final site of operation. If so. how is that water supplied? How is its amount controlled? And what parameter is 'measured' by the system to determine the concentration, particularly in tissues where the concentration of mucus seems to vary from period to period and where other factors in the environment exert their influence?

W e may not be able to reach conclusions. but by concentrating on these topics and, obviously, o n the other issues as well. I hope we will achieve a clearer understanding and definition of several outstanding questions.

laps.

Regulation of secretion from serous and mucous cells in the trachea

C. B. BASBAUM

Department of Anatomy und Curdiovusculur Research Institute. 1315-M, University of Culiforniu. San Francisco, CA 94/43. USA

Ahstrcict. The physical properties of mucus and the efficiency of tracheal mucociliary clearance depend on maintenance of a balanced interaction among several epithelial cell types. Some of these cell types are specialized to perform ion and water transport, others to perform synthesis and secretion of macromolecules. Our studies have been aimed specifically at identifying the neural mechanisms regulating macromoleculc secretion from two of these cell types, i.e. serous and mucous gland cells. Because these cells occur as part of a complex epithelium. i t is diHicult to monitor the properties and functions o f each cell type individually. We have therefore relied principally on morphological methods, which can potentially focus on a single cell type within a heterogeneous tissue. Such studies, however. depend on the availability of visible markers (enzyme-labelled anti- bodies, radioligands, etc.), and many important aspects of gland cell function cannot be assessed morphologically. Two alternative approaches are therefore k i n g developed: the isolation and segregation of gland cells according to type, and the production of monoclonal antibodies that recognize secretory products of individual cell types. These methods allow scrous and mucous cells to be studied by biochemical as well as morphological methods.

l0X-1 M u c u s ntid t n u m s u . Pittnan. London (Cihu Foundurion symposiirtn 100) p 4-10

Three morphologically distinct exocrine cells can be identified in histological sections of the trachea. The first two, serous and mucous gland cells, occur in mixed acini in the submucosa (Fig. 1) and discharge into common ducts which lead their products to the epithelial surface. At the surface, the secretion is mixed with that of a third cell type, the goblet cell. Hydration of the secreted material is thought to occur via osmotic mechanisms that chiefly reside in ciliated epithelial cells. Mucus, therefore, is a heterogeneous substance, derived from several cell types. The control mechanisms regulating the function of each cell have not been thoroughly characterized. However, the differential responsiveness of the cells to a variety of neuro- humoral substances has been documented. For example, serous cell secretion is stimulated more potently by a-adrenergic and cholinergic agonists than by

REGULATION OF SECRETION IN TRACHEA 5

FIG. 1. Serous (ser) and mucous (muc) cells occur in mixed acini of tracheal submucosal glands. m, myoepithelial cells, Ferret trachea. Scale bar. 10pm.

P-adrenergic agonists (Basbaum et al1981, Tom-Moy et al 1983). Mucous cell secretion is stimulated more potently by P-adrenergic and cholinergic agonists than by a-adrenergic agonists (Gashi et al 1984) and goblet cell secretion does not appear to be stimulated by either adrenergic or cholinergic drugs (Sherman et al 1981). In view of these cellular differences. it is not surprising that the composition (Ueki et al 1980, Quinton 1979, Boat & Cheng 1978) and physical properties (Leikauf et al 1984) of mucus vary according to the particular stimulus applied. Possibly, differences between mucus from normal and diseased airways (Boat et al 1977) reflect a shift in the profile of naturally occurring stimuli, receptor mechanisms and ultimately the cell types activated in the two conditions.

In order for differences in cellular ‘responsiveness’ to be translated into differences in the chemistry of the secretion, it has to be assumed that the cells discharge different molecular constituents when stimulated. Evidence has been obtained indicating differences in the composition of secretory products formed by the various cell types. Histochemical results (Jones 1978, Lamb &

6 BASBAUM

Reid 1969,1970) have shown that goblet and mucous gland cells contain large amounts of alcian blue-positive material (i.e. glycoproteins containing acidic residues) whereas serous cells are chiefly alcian blue-negative, containing mainly neutral glycoproteins. The use of lectin histochemistry has revealed differences in the exposed sugar residues characteristic of each cell type (Mazzuca et al 1982). Furthermore, biochemical differences in glycoprotein composition have been demonstrated by comparing [3H]glucosamine-labelled glycoconjugates derived from the goblet cell-containing epithelium with those derived from the gland-containing submucosa of cat trachea (Sherman et al 1981). The epithelial secretion was found to contain more sialic acid relative to galactose and N-acetylglucosamine and appeared to be less highly sul- phated than the submucosal secretion. Mucus composition must therefore vary with the biochemical profile of those cells activated by a particular stimulus.

Stimuli with the capacity to activate mucus secretion have been only partially identified. Among the most effective secretagogues are neurotrans- mitter-like drugs that mimic the action of the autonomic nervous system. When we began work in this area, neural regulatory mechanisms had not yet been studied on a cell-specific basis. In our first experiments we did a quantitative survey of the innervation pattern associated with adrenergic and cholinergic axon varicosities vis-u-vis serous and mucous cells (Murlas et al 1980). Considering only those varicosities containing five or more synaptic vesicles, we found that cholinergic outnumbered adrenergic varicosities by approximately 9 : 1, with serous cells receiving a considerably denser innerva- tion by both axon types than mucous cells received. We found no evidence, however, for selective innervation of either cell type by either axon type and we therefore speculated that selective neural regulation of the activity of each cell might occur as a function of the distribution of adrenergic and cholinergic receptors on the two cell types.

To test this hypothesis, we used radioligand binding to tracheal rings to study receptor localization. We localized adrenergic and cholinergic receptors under the light microscope, using autoradiography (Barnes & Basbaum 1983, Basbaum et al 1984). When we normalized values for autoradiographic grain density by taking into account section thickness, specific activity of the ligand and exposure time, we obtained figures for the number of binding sites per receptor type per cell type. The glands as a whole contained about .5 muscarinic, 3.5 a-adrenergic and 24 p-adrenergic binding sites/pm2 cell surface. When we analysed the two cell types separately, we found that serous cells contained significantly more (Y receptors than mucous cells did, and that mucous cells contained significantly more p receptors than serous cells did. Serous and mucous gland cells contained equal numbers of muscarinic receptors (5/pm2) and these were confined to the basolateral membranes facing interstitial axons (Fig. 2).

REGULATION OF SECRETION IN TRACHEA 7

FIG. 2. Distribution of muscarinic receptors as determined by autoradiography with the covalently binding muscarinic antagonist ["H]propylbenzylilcholine mustard used as a radioli- gand. SER, serous; MUC, mucous. (a) N o atropine; (b) atropine (10-4M). Ferret trachea. Scale bar, 25wm. (From Basbaum et al 19x4.)

8 BASBAUM

The receptor localization studies indicated that all three receptor types occur on serous and mucous cells. To determine which receptors mediate macromolecule release from each cell type, we did morphometric studies to measure secretion-related changes provoked by activation of specific recep- tors.

In these experiments, we incubated tracheal rings from the ferret with adrenergic and cholinergic drugs, then prepared the tissue for electron microscopy and morphometric analysis. By this means, we established that the volume density of secretory granules (Vvsg) for serous cells in controls (0.30 k 0.02 [mean f SE], n = 4) was significantly reduced by lo-' M-phenyl- ephrine (0.19 f 0.03, tz = 4) and lo-' M-methacholine (0.17 k 0.01, n = 4) but not by M-terbutaline(0.27 +- 0.04, n = 4) (Basbaum et a1 1981). The presence of antagonists in the medium (phentolamine, atropine or proprano- 101, each at 10F4M) prevented the reduction in Vvsg. In a later study, we found that serous cells (as monitored by lysozyme release) showed a small but significant response to isoproterenol (isoprenaline) (Tom-Moy et a1 1983).

Unlike serous cells, mucous cells showed no reduction in Vvsg in response to adrenergic and cholinergic drugs. Despite this, mucous cells showed obvious changes in their morphology in response to @-adrenergic and cholinergic agonists. In a later study (Gashi et a1 1984) we therefore measured two parameters in addition to Vvsg: volume density of the mucous cells (Vvmc) and surface density of mucous cell apical membrane (Svam). Although electron microscopy and morphometric analysis again showed no change in Vvsg for mucous cells in response to adrenergic and cholinergic drugs, Vvmc significantly decreased in response to lo-' M-bethanechol (control = 0.34 k 0.03; bethanechol = 0.16 + 0.02, mean k SD, n = 4. P<O.01) and M-isoproterenol (0.23 k 0.01) but not lo-' M-phenyl- ephrine (0.34 f 0.02). Similarly, Svam increased in response to bethanechol (control = 0.15 f 0.02; bethanechol = 0.33 k 0.04, IZ = 4, P<0.01) and isoproterenol (0.36 f 0.03) but not phenylephrine (0.19 k 0.04).

We believe that decreases in Vvsg and Vvmc and increases in Svam reflect cell degranulation with expansion of cell apical membrane. Presumably because mucous cells lose cytoplasm during degrada t ion (Specian & Neutra 1980) the parameter Vvsg does not reflect secretion. With the parameters Vvmc and Svam, however, it is clear that mucous cells secrete in response to bethanechol and isoproterenol but not phenylephrine. The appearance of serous and mucous cells after stimulation with adrenergic and cholinergic drugs is shown in Fig. 3 .

Thus, morphometric methods indicate that serous and mucous cells differ in their response to certain drugs. We have so far analysed secretion-related structural changes and receptor distribution for three receptors in a single species. Since species differences may occur and the spectrum of possible

REGULATION OF SECRETION IN TRACHEA

FIG. 3. Morphology of serous (SER) and mucous (MUC) cells seen in different incubation conditions. (a) 30-min stimulation with lo-' M-isoproterenol; (b) 30-min stimulation with M-phenylephrine. Ferret trachea. Scale bars: ( a ) 20pn1. ( b ) 1 0 p m . (From Basbaum et a1 1981.)

9

10 BASBAUM

secretagogues is wide, it seemed useful to develop improved screening procedures for serous and mucous cells, i.e. procedures not involving production of large numbers of micrographs and cell-by-cell analysis.

Most previous studies of gland regulation have used short-term organ culture of tracheal explants radiolabelled with mucin precursors. Measure- ment of gland secretion evoked by pharmacological stimulation has typically been accomplished by monitoring output of non-dialysable or trichloroacetic acid-precipitable radiolabel. However, no radiolabelled precursor has been identified that is incorporated selectively by serous, mucous or goblet cclls (Lamb & Reid 1969, 1970, Meyrick & Reid 1975). Therefore, radiolabel secreted in response to a given stimulus represents the collective output of several cell types. Each cell type might be differently affected by a given drug, but differences cannot be resolved by using radiolabels that are incorporated and released from a heterogeneous cell population. Similarly ambiguous is the identity of the specific cells responsible for secretion of the many distinct molecular constituents of mucus. Although mucosal secretions have been separated from submucosal secretions and analysed independently (Sher- man et al 19Sl), these samples themselves represent mixed contributions from several cell types. What is needed is an experimental approach that permits study of the regulation and contents of each secretory cell indepen- dently.

One such approach is the isolation and purification of serous and mucous cells. Once obtained, purified suspensions of single cell types could be radiolabelled and stimulated and the output of radioactivity referred back unambiguously to the responding cell. In addition, cells and supernatant could be analysed biochemically to characterize cell-specific secretions.

A second approach is the production of antibodies directed against the contents of each cell type. These could be used in immunoassays to monitor the activity of each cell independently, with cells still in situ. In addition, antibodies could be used via affinity chromatography to purify and character- ize cell-specific antigens. Thus, a single antibody could serve as a probe for both the activity and the biochemical contents of each cell type.

The capacity to monitor activity of a given cell type biochemically clearly offers significant advantages over morphological methods in terms of speed, ease of quantification, and breadth of questions asked. The type of questions that remain to be answered include: what are the roles of intracellular mediators (e.g. Ca?+ and cyclic nucleotides) in the regulation of macro- molecule secretion from serous and mucous cells? which substances in addition to sympathetic and parasympathetic agonists are effective stimuli for secretion? which intracellular proteins are modified (e.g. phosphorylated) as part of the secretory process? and what are the materials secreted by each cell type? Despite the difficulties inherent in removing gland cells from the

REGULATION OF SECRETION IN TRACHEA 11

collagenous matrix binding them to tracheal cartilage, a preparation of disaggregated gland cells from cat trachea has recently been described (Culp et al 1983). A major disadvantage of the preparation described in the cat, however, is the relatively low yield of viable cells and the loss or damage of P-adrenergic receptors incurred during disaggregation. We have examined the possibility of using larger species for such a preparation (Finkbeiner 6i Basbaum 1983) and have found that the tracheas of the sheep and the cow provide an abundant source of viable gland cells (Fig. 4). We obtain the cells after dissecting soft tissue from tracheal cartilage and subjecting it to enzymic and mechanical disaggregation. The cells in our preparations avidly accumu- late %. Rather than doing extensive studies on the mixed cell preparation, however, we are attempting to raise monoclonal antibodies which will distinguish between antigens present on serous and mucous cell surfaces. By immobilization of such antibodies on a solid support it should be possible to separate serous from mucous cells and study their secretory mechanisms and products in either cell suspension or culture.

To monitor activity of the individual cell types in vivo or in tissue explants, we have generated many cell-specific anti-secretion monoclonal antibodies. We have done this by immunizing mice with crude secretions from perfusates of sheep tracheal pieces stimulated in vivo with adrenergic and cholinergic agonists (Basbaum et a1 1983).

Because antibodies can be obtained from clones of single B lymphocytes taken from the mouse spleen, the antibodies show a very narrow specificity. We assayed a number of monoclonal antibodies by immunofluorescence and found that, of 337 hybridomas screened, 100 produced antibodies recognizing goblet cell granules, 64 recognized cell granules, and in three the antigen was confined to the ciliated apical surface of the epithelium. Many tracheal goblet cell antibodies (such as that shown in Fig. 5 ) were strongly cross-reactive with intestinal goblet cells and a subpopulation of submandibular gland cells, but not with cells of Brunner’s glands or the ciliated cell apical membrane. We have also produced antibodies that distinguish between products of serous and mucous cells. Serous cell antibodies (such as that shown in Fig. 6) were not cross-reactive with tracheal mucous, goblet, Brunner’s or submandibular cells, or with the ciliated cell apical membrane. Antibodies directed against the apical membrane of ciliated cells did not cross-react with gland or goblet cells or the apical membrane of epithelial cells in the duodenum. Moreover, many of the antibodies showed that cellular specificity was conserved across species (cat, ferret, sheep).

We are now attempting to characterize the cell-specific antigens. For preliminary characterization we are using an enzyme-linked immunosorbent assay (ELISA). We have evidence indicating that antibody 4C4. shown by immunofluorescence to recognize a component of serous cell granules, is

12 BASBAUM

FIG. 4. Sections through cell pellcts consisting of disaggregated submucosal gland cells from sheep trachea. ( a ) Scrous cell: ( b ) mucous cell. Magnilications: ( a ) 13000 X , ( b ) 11 000 X .

REGULATION OF SECRETION I N TRACHEA 13

FIG. 5 . Immunofluorescent staining obtained using a monoclonal antibody directed against goblet cell secretory granules. E , epithelium; SG. submucosal glands. Cat trachea. Scale bar. 100pm.

directed against the oligosaccharide portion of a material that elutes in the void volume of an A15M column.

In summary, we have demonstrated thus far that serous and mucous cells differ in their responsiveness to autonomomimetic drugs and in the im- munochemical profiles of their secretory products. In addition. we have established the feasibility of producing, from a complex secretion, antibodies which are strictly specific for the products present in a single cell type. The antibodies can be used not only in chemical studies to purify and characterize cell-specific products, but also in physiological studies (using ELISAs) as

FIG. 6 . Immunofluorescent staining obtained using a monoclonal antibody directed against serous cell secretory granules. E. epithelium: SG, submucosal glands. Cat trachea. Scale bar. 100,um.

14 BASBAUM

markers for secretion by individual cell types. As such, they can be used to measure secretion from cells in situ. The outcome of future studies will reveal whether it is similarly possible to obtain monoclonal antibodies which distinguish between the membranes of serous and mucous cells or between apical and basolateral membranes of these cells (Quaroni 1983). Cell surface antibodies may help to provide information about the biologically active molecules (pumps, channels, receptors, etc.) which mediate functions crucial to and characteristic of the several cell types engaged in secretion of tracheal mucus.

Acknowledgements

This work was supported by National Institutes of Health PPG HL-24136 and by grants from the Cystic Fibrosis Foundation and the Strohel Foundation o f the American Lung Association of San Francisco. I wish to thank my colleagues, Drs C. Murlas. M. Tom-Moy. A . Gashi. P. Barnes, J . H . Widdicombe, M. Grillo. W. Finkbeiner. J . Mann and J. Nadel, who collaborated in these studies.

REFERENCES

Barnes PJ. Bashaurn C B 1983 Mapping of adrenergic receptors in the trachca by autoradiogra- phy. Exp Lung Res 5:183-192

Basbaum CB. Ueki I. Brezina L. Nadel JA 1981 Tracheal submucosal gland serous cells stimulated in vitro with adrenergic and cholincrgic agonists: a morphornctric study. Cell Tissue Res 220:481-498

Basbaum CB. Finkbeiner WE, Mann J E 1983 Monoclonal antibodies directed against cell- specific granules of tracheal secretory cells. J Cell Biol 97:78a

Basbaum CB, Grillo MA, Widdicombe J H 1984 Muscarinic receptors: evidence for a nonuniform distribution in tracheal smooth muscle and exocrine glands. J Neurosci 4:509-521

Boat TF, Cheng PW 1978 Alteration of glycoproteins released from human tracheal explants by methacholine stimulation. Am Rev Respir Dis 177:288 (abstr.)

Boat TF, Cheng P , Wood R E 1977 Tracheobronchial mucus secretion in vivo and in vitro by epithelial tissues from cystic fibrosis and control subjects. Mod Probl Paediatr 19: 141-152

Culp DJ. Penney DP. Marin MG 1983 A technique for the isolation of submucosal gland cells from cat trachea. J Appl Physiol 5s: IU35-104l

Finkbeiner WE, Basbaum CB 1983 Evaluation of sheep and dog tracheas as a source of isolated submucosal gland cells. Physiologist 26:A125

Gashi A. Nadel JA, Basbaum CB 1984 Morphometric studies of tracheal gland mucous cell stimulated with autonomomimetic drugs. Clin Res 32:429A

Jones R 1978 The glycoproteins of secretory cells in airway epithelium. In: Respiratory tract mucus. Elsevier/Excerpta Medica, Amsterdam (Ciba Found Symp 54) p 175-188

Lamb D. Reid L 1969 Histochemical types of acidic glycoprotein produced by mucous cells of the tracheobronchial glands in man. J Pathol 98:213-229