immune recognition of proteins: conclusions, dilemmas and enigmas

112 BioEssays Vol. 6 , No. 3

REVIEW ARTICLES

23 ROTHBLUM, L. I., REDDY, R. & CASSIDY, tion unit from mouse. Nucl. Acids Res. 9, B. (1982). Transcription initiation site of rat 1559-1 569. ribosomal DNA. Nucl. Acids Res. 10, 25 KUEHN, M. & ARNHEIM, N. (1983). 1345-1362. Nucleotide sequence of the genetically labile 24 BACH, R., GRUMMT, I. & ALLET, B. repeated elements 5’ to the origin of mouse (1981). The nucleotide sequence of the rRNA transcription. Nucl. Acids Res. 11, initiation region of the ribosomal transcrip- 21 1-224.

R O N A L D H . R E E D E R , P A U L L A B H A R T A N D B R I A N McSTAY are at the Fred Hutchinson Cancer Research Centre, 1124 Columbia St., Seattle, WA 98104, USA.

Immune Recognition of Proteins: Conclusions, Dilemmas and John A. Smith and George D. Rose

Enigmas

Summary

The immune system distinguishes between two types of antigenic sites: one of these binds to immunoglobulins (ZgGs) (i.e. antibodies), while the other binds to receptor molecules on T lymphocytes (i.e. the T-cell receptors (TcRs)). The latter interaction occurs only when the antigen is presented in association with a self-transplantation antigen, a so-called MHC-restriction element. This article discusses what is known about the structure of antigenic sites and their molecular interactions with antibodies, MHC-restriction elements, and T-lym- phocyte receptors.

I nt roduct ion

A successful immune response against an invading organism (e.g. a parasite, virus or bacterium) is dependent on interactions between antigenic sites of surface proteins of the organism and IgGs and between the same or different antigenic sites derived from the proteins and TcRs. Cytotoxic (‘killer’) T lymphocytes (CTLs) are also known to provide additional protective immunity against viruses but will not be discussed, because little is known about how CTLs recognize and interact with antigenic sites. The interaction of antigenic sites with Ig-binding sites is understood more completely than the interaction of antigenic sites with TcRs. However, it seems certain that the binding sites of Igs and TcRs will be fundamentally the same, so that differences in recognition will be due to the structure of the antigenic site itself or the structure presented to a binding site.’ The

anti-peptide antibody studies of Lerner and co-workers,2 though related, are outside the scope of the present review.

Antibody-Antigenic Site Recognition

Immunogenicity is defined as the func- tional property of an antigenic site to elicit an immune response (antibody or T-cell), when introduced into a host (for reviews see refs. 3-5). Antigenicity is defined as the physical property of an antigenic site to bind to an antibody or to cellular receptors. The former refers to processing and subsequent stimulation of the immune system, while the latter refers to molecular recognition and binding. These two properties need not coincide. In either case, the antigenic site is functionally defined by its ability to bind specifically to a complementary Ig-binding site. Antigenic sites are also called antigenic determinants or epitopes. Each such site is coextensive with a region of the protein surface.

Antigenic sites are subclassified as ‘continuous’ (i.e. comprised of a con- tiguous sequence of amino acids, defined by the primary structure of the protein) and ‘discontinuous’ (i.e. comprised of a spatial cluster of amino acids whose juxtaposition is a consequence of the secondary and tertiary structure of the protein).‘j However, these distinc- tions are somewhat artificial, since the protein surface is most likely a continu- um of overlapping antigenic sites. In- deed, X-ray diffraction studies of a protein antigen (1ysozyme)-antibody complex reveal an antigenic site that contains at least 7 amino acids contrib-

uted by discrete linear segments and clustered together due to folding of the protein chain (Fig. l).’ These residues interact with a shallow groove on the antibody, the combining site, by a ‘lock and key’ mechanism without inducing structural changes in the conformation of the protein antigen.’ The binding forces include hydrogen bonds, hydrophobic interactions and van der Waals interactions. The dimensions of an antigenic site are delimited by the region of space that can be spanned by the antibody’s combining site. Different antibodies directed against a similar region of the protein surface will likely have differing combining orientations, as well as differing degrees of interaction with pertinent residues.

There appear to be no distinctive structural features (e.g. reverse turns, P-sheets or a-helices) that are charac- teristic of antigen sites. In principle, the entire accessible surface of a protein could elicit antibodies directed against all possible sites. However, the sites that successfully induce an immune response are dependent on additional levels of immune regulation involving: immuno- logical tolerance (e.g. a diminished immune response develops in an animal whose own proteins bear epitopes similar to epitopes of the foreign antigen); Ig gene repertoire; major histocompatibility complex of transplantation antigens (cell surface proteins involved in cellular recognition) ; regulation of B lymphocytes (cells bearing surface immunoglobulin) by helper and suppressor T lymphocytes; and complex regulatory mechanisms (involving idiotypic network^).^

BioEssays Vol. 6, No. 3 11 3

REVIEW ARTICLES

Predictive Methods to Localize Antigenic Sites Recognized by Antibodies

During the past two decades most of our knowledge concerning protein antigenicity has been gleaned from studying X-ray-elucidated globular proteins (e.g. myoglobin, cytochrome, lysozyme and staphylococcal nuclease) (for review see ref. 4). However, the complete identification of the antigenic sites for even these well-characterized models is incomplete, although our ability to predict antigenic sites in proteins of known structure has been increasing.

There are three predictive methods that utilize single crystal X-ray data. The first of these involves probing the protein surface with a sphere of radius lOA (approx. size of an antibody- binding domain); enumerating the residues that can establish physical contact with the sphere; and ascribing the putative antigenic sites to these accessible regions.s In essence, the method simply scans the protein surface and identifies the protuberant regions. If the probe sphere were smaller (e.g. the size of a water molecule), the resultant accessible surface would be large^.^ In fact, it is this solvent-accessible surface (probe radius 1.8 A) that is what most immunologists refer to as the ‘entire’ accessible surface that can be recognized by the immune system. Clearly, not all accessible regions will be equally immunogenic. The second method uses molecular cartography to plot a contour map (analogous to the terrain maps used by hikers) of the protein in order to quantitate its surface topography.1° Again, it is the most protuberant regions that are predicted to be antigenic. The third method assumes correlation between the location of antigenic sites and atomic mobility and uses Debye-Waller temperature factors (i.e. a measure of atomic disorder derived from X-ray refinement) to identify regions of greatest mobility.”. l2 Any of these methods could be used to predict either continuous or discontinuous sites; none predicts perfectly.

Prediction of antigenic sites in proteins of unknown structure is uncertain, because there is no reliable way to map the amino acid sequence into a defined chainfold. The only available method is based upon a hydrophobicity profile, a plot of residue hydrophobicity against position in the linear sequence.13 In such a profile, local maxima in hydrophobicity are predicted to be buried within the molecule, while local minima

Fig. 1. The antigenic residues of an antigen-antibody complex. The structure of the complex between hen egg-white lysozyme and the Fab of a monoclonal anti-lysozyme antibody has been elucidated by X-ray crystallography.8 In the illustration, the lysozyme a-carbon backbone is displayed as a white ribbon. All atoms within antigenic residues (namely residues 13, 14, 19, 21, 22, 24, 117, 119, I21 and 1251, together with their van der Waals surfaces, are shown as clusters of white dots. A 30 A x 30 d planar grid is superimposed upon the protein: each grid square is I da. It can be seen that the antigenic site is coextensive with a broadpatch of the protein surface. Designedsynthetic peptides often elicit polyclonal antibodies that cross-react with the protein, but such antibodies rarely confer protective immunity. This enigma may arise because typical linear synthetic peptides are not large enough to mimic a ‘complete’ antigenic site adequately.

are predicted to be exposed.14 Since antibodies bind to regions of the exposed surface, the method can be used to predict antigenic sites.15 How- ever, the correlation between surface exposure and local minima in hydrophobicity does not imply that antigenic sites are always coincident with hydro- philic surface regions, since hydrophobic and aromatic residues do occur in antigenic sites. In addition, the correlation of reverse turns (i.e. chain reversals occurring within a stretch of four consecutive residues; the surface of

globular proteins has been likened to a ‘turn-cage’) and antigenic sites is also not absolute.la In spite of these caveats, these methods are the best available and are often used to predict the location of protein epitopes.

In summary, what is desired is an accurate method for predicting which surface regions (i.e. ‘discontinuous’ antigenic sites) will elicit antibodies leading to protective immunity. There is currently no available method that can do this reliably. Further, although designed synthetic peptides often elicit


REVIEW ARTICLES

polyclonal antibodies that cross-react with the protein, they infrequently confer protective immunity. Apparently such peptides fail to adopt confor- mations identical to the ‘native’ protein or else they fail to mimic a ‘complete’ antigenic site adequately.

Experimental Methods to Localize Antigenic Sites Recognized by Antibodies

There are four experimental methods that can be used to identify antigenic sites (for review see ref. 5). Three of these allow identification of ‘discontinuous’ antigenic sites; the fourth is limited to identification of ‘continuous’ antigenic sites.

The first method involves a fine specificity analysis of evolutionarily related proteins (for review see ref. 4). This method relies on antibodies from one species (e.g. rabbit) directed against antigenic sites in a protein from another species (e.g. chicken egg-white lysozyme) being able to discriminate among subtly altered antigenic sites of proteins from still other species (e.g. pheasant egg-white lysozyme). Although this method has been widely used for lysozyme, cytochrome c and myoglobin, the method is biased toward detecting epitopes in highly evolutionarily vari- able and ‘ non-tolerized’ (defined above) surface regions. In spite of these limitations, much that is known about the localization of antigenic sites in globular proteins is based on this method.

The second method uses a monoclonal antibody directed at a ‘discontinuous’ epitope to block by steric hindrance the binding of another monoclonal antibody (for review see ref. 5). This method may be used without knowing the primary structure of a protein; it yields limited information about the specificity of individual antibodies directed against either overlapping or distinct epitopes.

The third method detects, albeit incompletely, discontinuous antigenic sites. The method relies on the fact that monoclonal antibodies binding to an antigenic site can inhibit cleavage by proteolytic enzymes at normally sus- ceptible cleavage sites.” However, such an experiment can be interpreted only if the Ig-combining site is not digested by the protease, the antigen is readily digested, and there are available methods to separate and identify the cleavage products. This last requirement is a significant limitation to the application

of this method to high molecular- weight proteins.

In contrast to the above methods, synthetic peptides or peptides derived from chemical cleavage and/or enzymic digestion of a protein can be used to mimic ‘continuous’ antigenic sites.18* l9 At best such linear epitopes can account only partially for a ‘discontinuous’ epitope, and this is a major limitation of this approach. However, provided that the protein fragments can be isolated and that the peptide sequence of each reactive peptide can be determined unequivocally, the use of protein fragments, usually isolated by high-per- formance liquid chromatography, can be an inexpensive (i.e. in comparison with peptide synthesis) approach to rapidly assess the distribution of antigenic sites of a protein, provided the protein is available in sufficient quanti- ties. However, this approach is not a panacea, since it may fail to detect certain antigenic sites. For example, if trypsin is used to cleave the protein chain at lysyl or argininyl residues and an antigenic site contains such residues, the antigenic site may be ablated, and hence will no longer react with anti- protein antibodies. This and other artefacts can also be introduced by an injudicious choice of synthetic peptides (e.g. failure to bind to ‘incomplete’ antigenic sites, because the N- or C- terninus of the synthetic peptide does not include the amino acid residues localized to a boundary of an antigenic site; also non-specific binding to short, lysine-containing peptides can occur).

Three synthetic methods are available for rapid screening of ‘continuous’ antigenic sites: (1) solid-phase peptide synthesis and radioimmunoassay assay using the same peptide-linked solid support;20 (2) multiple, simultaneous syntheses, based on synthetic approach (1) with detection using an enzyme- linked immunosorbent assay (a colori- metic assay resulting from enzymic conversion of a substrate by an enzyme linked to a second antibody which reacts with the antibody recognizing a antigenic site);z1 and (3) small-scale, simultaneous, solid-phase synthesis of large numbers of synthetic peptides, cleavage of individual peptides from the resin, and purification.22 Two methods (methods 1 and 2) are rapid screening methods but, because the synthetic peptides are used without purification, interpretation of the data is frequently equivocal. The interpretation of the binding data from peptides prepared by method 3 has the same limitations as

that for peptides isolated from protein fragments, described above. On a posi- tive note, peptide synthesis is uniquely able to provide synthetic analogues for the evaluation of the fine specificity of antibody recognition and is extremely suited for elucidating the molecular recognition of antigenic sites by T- lymphocytes (see below).

T-cell Receptor/Antigenic Site Recognition

There are two types of T-cell receptor: the a-,8 heterodimer typez3 (where a and /3 are M , 50000 and 40000 dalton protein subunits, respectively) and the y-S heterodimer type (where y and S are M, 55000 and 40000 dalton protein subunits, re~pectively).~~ In contrast to antibody recognition, antigenic sites recognized by T-cell receptor must be presented in the context of Ia molecules (so-called class I1 molecules), which are present within the surface membranes of B-cells, macrophages, dendritic cells, epithelial Langerhans cells, thymic epithelial, as well as some T - ~ e l l s . ~ ~ Presumably, different Ia molecules interact with different epitopes of multisite protein antigens, although it is unclear why some epitopes are preferentially recognized by certain Ia-presenting cells and others are not. Further, it is unclear how a potentially large number of foreign proteins are able to form unique complexes with a very limited number of Ia molecules. In any event, T-cell response is dependent upon subsequent recognition of a combination of the antigenic peptide and the Ia molecule.

Unlike antibody recognition, T-cell recognition cannot distinguish between ‘native’ and ‘denatured’ proteins.26 This observation is explained by the apparent requirement for the proteolytic cleavage of intact proteins to protein fragments within phagocytic cells.25 It is these protein fragments in association with Ia molecules that are presented to the T-cell receptor. Syn- thetic peptides or protein fragments prepared in the laboratory can be substituted for cellularly processed protein fragments. An evaluation of the binding of these analogues to Ia molecules is shedding light on the binding specificity of these molecules.27 The minimal size of an antigenic site recognized by T-cells varies between 7 and 17 amino acid residues. The fine specificity of this recognition is exquisitely sensi- tive (e.g. the exclusion of a single methy- lene group (i.e. a CH,) leads to the loss of recognition).

BioEssays Vol. 6, No. 3 11 5

REVIEW ARTICLES

It is clear that T-cells and B-cells recognize different antigenic sites.28 Further, the T-cells recognize a limited number of antigenic sites, while antibodies are directed at multiple sites probably involving the entire protein surface. It is also evident that different classes of T-cells (e.g. suppressor and helper) recognize different, non-overlapping e p i t o p e ~ . ~ ~ There is no experi- mentally proven, structural basis for the recognition of antigenic sites by T-cells, but this topic is being intensively studied.

Predictive and Experimental Methods to Localize Antigenic Sites Recognized by T-cells

No predictive method has proved to be useful for localizing the T-cell recognition sites of proteins.

There are two experimental ap- proaches that are used frequently. The first approach relies on the use of evolutionarily related proteins with subtle differences in primary structure, and assessment of the effects of these subtle changes on the stimulation of the immune response (measured by the incorporation of radiolabelled deoxy- thymidine into the DNA of prolifera- ting T-cells) after stimulation of other specific or polyclonal T-cell clones (or T-cell hybridomas). The second approach, which bypasses the cellular processing events required to generate fragments that interact with Ia molecules, utilizes protein fragments or synthetic peptides to trigger a proliferative T-cell response. Although there is apparently no distinction between ‘continuous’ and ‘discontinuous’ epitopes as recognized by T-cells, the caveats that apply to the use of synthetic peptides or protein fragments for determining antibody recognition sites (see above) also apply to peptides used to map T-cell epitopes. The preparation of nested sets of synthetic peptides and specific analogues by solid phase methods is now being used to unravel the intricacies of the binding specificities of Ia molecules and will be useful for elucidating the structures of the Ia molecule-peptidic epitope complex that is in turn presented to the T-cell receptor. However, the later experiments await the isolation of purified T-cell receptors and major histocompatibility complex-restricting elements.

Epilogue

There is a limited understanding of the molecule recognition d antigenic sites by both antibodies and T-cell receptors. The information that has been accu- mulated is still insufficient to allow the rational design of synthetic immuno- suppressive agents and vaccines that can selectively regulate immunity. A complete understanding of the nature of antigenic sites will revolutionize our ability to manipulate the immune response and will augment our understanding of macromolecular recognition.

REFERENCES 1 NOVOTNY, J., TONEGAWA, S., SAITO, H., KRANZ, D. M. & EISEN, H. N. (1986). Secondary, tertiary, and quaternary structure of T-cell-specific immunoglobulin-like polypeptide chains. Proc. Natl. Acad. Sci.

2 LERNER, R. A. (1982). Tapping the im- munological repertoire to produce antibodies of predetermined specificity. Nature

3 BERZOFSKY, J. A. (1985). Intrinsic and extrinsic factors in protein antigenic structure. Science 229, 932-940. 4 BENJAMIN, D. C., BERZOFSKY, J. A., EAST, I. J., GURD, F. N. R., HANNUM, C., LEACH, S. J., MARGOLIASH, E., MICHAEL, J. G., MILLER, A., PRAGER, E. M., REICHLIN, M., SERCARZ, E. E., SMITH-GILL, S. J., TODD, P. E. & WILSON, A. C. (1984). The antigenic structure of proteins: a reappraisal. Annu. Rev. Immunol. 2, 67-101. 5 JEMERSON, R. & PATERSON, Y. (1986). Mapping antigenic sites on proteins: impli- cation for the design of synthetic vaccines. Bio Techniques 4, 18-3 1. 6 ATASSI, M. Z. & SMITH, J. A. (1978). A proposal for the nomenclature of antigenic sites in peptides and proteins. Immuno- chemistry 15, 609-610. 7 AMIT, A. G., MARIUZZA, R. A., F’HILLIPS, S. E. V. & POLJAK, R. J. (1985). Three- dimensional structure of an antigen-antibody complex at 6 8, resolution. Nature 313, 156-158. 8 NOVOTNY, J., HANDSCHUMACHER, M., HABER, E., BRUCCOLERI, R. E., CARLSON, W. B., FANNING, D. W., SMITH, J. A. & ROSE, G. D. (1986). Antigenic determinants in proteins coincide with surface regions accessible to large probes (antibody do- mains). Proc. Natl. Acad. Sci., U.S.A. 83, 226230. 9 RICHARDS, F. M. (1977). Areas, volumes, packing and protein structure. Annu. Rev. Biophys. Bioeng. 6, 151-176. 10 FANNING, D. W., SMITH, J. A. & ROSE, G. D. (1986). Molecular cartography of globular proteins with application to antigenic sites. Biopolymers 25, 863-883. 11 WESTHOF, E., ALTSCHUH, D., MORAS, D., BLOOMER, A. C., MONDRAGON, A.,

USA 83, 742-746.

299, 592-596.

KLUG, A. & VAN REGENMORTEL, M. H. V. (1984). Correlation between segmental mobility and the location of antigenic determinants in protein. Nature 311, 123-126. 12 TAINER, J. A., GETZOFF, E. D., ALEX- ANDER, H., HOUGHTEN, R. A., OLSON, A. J., LERNER, R.A. & HENDRICKSON, W.A. (1984). The reactivity of anti-peptide antibodies is a function of the atomic mobility sites in a protein. Nature 312, 127-134 13 ROSE, G. D. (1978). Prediction of chain turns in globular proteins on a hydrophobic basis. Nature 272, 586-590. 14 KYTE, J. & DOOLITTLE, R. F. (1982). A simple method for displaying the hydrophobic character of a protein. J. Molec. Biol. 157, 105-132. 15 HOPP, T.P. & WOODS, K. R. (1981). Prediction of antigenic determinants from amino acid sequences. Proc. Natl. Acad. Sci. USA 78, 38243828. 16 ROSE, G. D., GIERASCH, L. M. & SMITH, J. A. (1985). Turns in peptides and proteins. Adv. Protein Chem. 37, 1-109. 17 JEMMERSON, R. & PATERSON, Y. (1985). Identification of antigenic determinants in proteins by analysis of peptides from proteolysed antigen-antibody complexes. In Peptides: Structure and Function (Proceed- ings of the Ninth American Peptide Sym- posium) (ed. V. J. Hruby, C. M. Deber & K. D. Kopple), pp. 67-70. Pierce Chemical Company, Rockford, IL. 18 ANDERER, F. A. (1963). Preparation and isolation of an artificial antigen immunolo- gically related to tobacco mosaic virus. Biochim. Biophys. Acta 71, 246248. 19 ATASSI, M. Z. (1975). Antigenic structure of myoglobin: the complete immuno- chemical anatomy of a protein and conclusions relating to antigenic structures of proteins. Immunochem. 12, 423438. 20 SMITH, J. A,, HURRELL, J. G. R. & LEACH, S. J. (1977). A novel method for delineating antigenic determinants : peptide synthesis and radioimmunoassay using the same solid support. Immunochernistry 14,

21 GEYSEN, H. M., MELOEN, R. H. & BAR- TELING, S. J. (1984). Use of peptide synthesis to probe viral antigens for epitopes to a resolution of a single amino acid. Proc. Natl. Acad. Sci. USA 81, 39984022. 22 HOUGHTEN, R. A. (1985). General method for the rapid solid-phase synthesis of large numbers of peptides: specificity of antigen-antibody interaction at the level of individual amino acids. Proc. Natl Acad. Sci. USA 82, 5131-5135. 23 HASKINS, K., KUBO, R., WHITE, J., PIGEON, M., KLAPPER, J. & MARRACK, P. J. (1983). The major histocompatibility complex-restricted antigen receptor on T-cells. I. Isolation with a monoclonal antibody. J. Exp. Med. 157, 1149-1 169. 24 BRENNER, M. B., MCLEAN, J., DIA- LYNAS, D. P., STROMINGER, J. L., SMITH, J. A., OWEN, F. L., SEIDMAN, J. G., IP, S., ROSEN, F. & KRANGEL, M. S. (1986). Identification of a putative second T cell receptor. Nature 322, 145-149.

565-568.


REVIEW ARTICLES

25 UNANUE, E. R. (1984). Antigen-presenting function of the macrophage. Annu. Rev. Immunol. 2, 395428. 26 GELL, P. G. H. & BENACERRAF (1959). Studies on hypersensitivity. 11. Delayed hypersensitivity to denatured proteins in guinea pigs. Immunology 2, 64-70. 27 BABBITT, B. P., ALLEN, P. M., MAT- SUEDA, G., HABER, E. & UNANUE, E. R. (1985). Binding of immunogenic peptides to Ia histocompatibility molecules. Nature 317, 359-361.

28 MAIZELS, R. M., CLARKE, J. A., HARVEY, M. A., MILLER, A. & SERCARZ, E. E. (1980). Epitope specificity of the T-cell proliferative response to lysozyrne: T proliferative cells react predominantly to different regions from those recognized by B cells. Eur. J . Immunol. 10, 509-51 5 . 29 GAMMON, G., DUNN, K., SHASTRI, N., OKI, A., WILBUR, S. & SERCARZ, E. E. (1986). Neonatal T-cell tolerance to minimal immunogenic peptides is caused by clonal inactivation. Nature 319, 413415.

JOHN A. SMITH is at the Departments of Molecular Biology and Pathology, Massachusetts General Hospital, Fruit Street, Boston, M A 02114, and the Department of Pathology, Harvard Medical School, Boston, MA 02115.

GEORGE D. ROSE is at the Department of Biological Chemistry, Pennsylvania State University Hershey Medical Center, Hershey, PA 17033.

Overcoming Random Diffusion in Polarized Cells - Corralling the Drunken -

Beggar David E. Wolf

Summary

Cells are capable of overcoming the randomizing effect of lateral diffusion in order to regionally differentiate their surfaces. Such local structural specializ- ations are of major signijicance to cellular function. In some cases, they may be explained by dtfusion rates that are insuficient to completely randomize surface gradients over biologically rel- evant times scales. However, in other cases, absolute and permanent regionalizations are also observed. Mechanisti- cally, the problem is analogous to equili- brium across a dialysis bag: either an absolute barrier exists or the chemical potential between two adjacent regions must be equal. The interactive nature of the system, where localizations of one component lead to localization of others, are also considered here. (And, as you might like to try the thing yourself, some winter day, I will tell you how the Dodo managed it.)

First it marked out a racecourse, in a sort of circle (‘the exact shape doesn’t matter’, it said), and there all the party were placed along the course, here and there. There was no ‘One, two, three, and away!’ but they began running when they liked, and left off when they liked, so that it was not easy to know when the race was over. However, when they had been running half an hour or so, and were quite dry again, the Dodo suddenly called out, ‘The race is over!’ and they a11 crowded round it, panting, and asking, ‘But who has won?’

‘A Caucus-Race’ from Alice’s Adven- tures in Wonderland, by Charles L. Dodgson (Lewis Carroll)

Abbreviations used : fluorescence recovery after photobleaching (FPR); percentage of molecules free to diffuse (% R); diffusion coefficient (D).

The Fluid Mosaic Model

Since the experiments of Frye and Edidin’ demonstrating the ability of surface antigens to intermix upon heterokaryon fusion, it has been clear that, in general, the components of the cell’s plasma membrane are free to diffuse within the plane of the membrane. This experiment was quintessen- tial to the development of the ‘Fluid Mosaic Model’,2 which envisages the lipids of the membrane as being organ- ized into a lipid bilayer in which membrane proteins are situated, some spanning the bilayer. Some membrane proteins are thought to be anchored on the cytoplasmic side to cytoskeletal fibers or on the exoplasmic side to elements of the glycocalyx. The lipid bilayer was postulated to be a quasi- two-dimensional fluid in which the lipids and proteins move freely. The term fluid implies that there is a bulk or overall property of the membrane called its ‘fluidity’ or ‘viscosity’ which con- trols the rotational and lateral diffusion of a large object such as a membrane protein.

Diffusion is a Random Process

The defining feature of diffusion is that it is a random process. Diffusing molecules are like the participants in Alice’s ‘ Caucus-Race’ running at random rates in random directions. So that even if they all started at the same point they would soon be randomly distributed about the racecourse. The sheer chaos and lack of direction in diffusion is often illustrated by analogy to the mindless reelings of a drunken beggar. The beggar begins under a lamp

post. In his drunken stupor he takes random steps in random directions. Because he is as likely to go in one direction as its opposite, the beggar on average remains under the lamp post. However, the mean squared distance of the beggar from the lamp post is proportional to the number of steps, or the time, since the beggar first left the lamp post. These two points may seem contradictory. However, consider a crowd of beggars all independently involved in this drunken chaotic dance. The crowd is always centered about the lamp post (the average position) but the width of the crowd (the root mean squared distance) is an ever-expanding circle.

Overcoming Random Diffusion During Cellular Differentiation

If membrane molecules were completely free to diffuse in the plane of cell plasma membranes one would expect them to be randomly distributed. However, during a number of important processes of cellular differentiation, such as sperm maturation and capacitation,3-6 early embryogene~is,~ erythropoiesis: tight junction formation in e~ithelia,~ and myotube development,1° cells demon- strate their ability to overcome the randomizing effect of diffusion and to localize certain membrane components to specific regions of the cell surface. The ubiquity of surface regionalizations during cellular differentiation suggests a close relationship between these two phenomena, and demonstrates the need to understand how cells restrain the free diffusion of their surface components if one hopes to understand the processes of differentiaton.

immune recognition of proteins: conclusions, dilemmas and enigmas

Documents