Thank you, Mr. Chairman. Ladies and gentlemen, it is a privilege to talk to you about some aspects of the cell surface using immunology as an example. Immunologists, I think, can be divided into two kinds.

Those who study molecules and those who study cells and their relationship, at least until recently, can perhaps best be exemplified by a story of two young and quite bright boys who were rivals at school all their life and who finally grew up, one to become an admiral, the other to become a bishop.

And continuing their competition, they found themselves one day in a railroad station at a conveniently haughty distance, eyeing each other coldly, each in his appropriate uniform or regalia. Finally, the bishop walked up to the admiral and said, pardon me, conductor.

Is this the way to Lindau? And the admiral looked over the bishop and said, yes, madam, but in your condition should you go. Well, there is now emerging, I believe, a third kind of immunologist who uses the lymphocyte as a model for general cell

biological properties. And the antibody, the product of that lymphocyte, as a tool, which I hope to persuade you, is the most fundamental one and one which we will see more of in the future of cell biology. It is, in fact, to a consideration of this approach that I wish to direct my remarks today.

I want especially to consider some relationships of events at the cell surface to growth control and to cell division and to see how these processes bear upon events at the cell surface, as suggested, in fact, by cell biologists for a long period of time.

They have, through anecdotal observations, noted that there must be some kind of relationship between cell interaction, cell motion, and cell division. But relatively little evidence is accumulated to indicate what the molecular basis of this relationship might be. Now, in fact, there are two main ways

in which one might look upon the mechanism of cell division and its control of mammalian cells, both normal and neoplastic. The first is to consider the whole thing as a problem in gene expression in which sources and sinks of activating substances depend upon regulation of that gene expression.

But without denying the validity of that view, there is another way of looking at the problem. This is to consider growth control as a reflection of the workings of a macromolecular system or an assembly associated with the cell membrane and already in one or another control state. It's to a consideration of this hypothesis

of a surface modulating assembly, an hypothesis, I believe, which is not generally held, that I want to direct most of my remarks today. Some time ago, my colleagues and I advanced such an hypothesis to account for lymphocyte mitogenesis. And there has been, since then, some evidence to support that view.

Today, what I plan to do is discuss work by my colleagues John Wong and Ichiro Yahara on lymphocytes but also on other cells. I shall not mention that work in great detail that I believe supports the view that an appropriate macromolecular assembly for growth and control exists and that also suggests something about the nature of its constituent elements.

I hope you will forgive me, in fact, for discussing so incomplete an account of a rapidly developing and new subject. But I hope you will finally agree with me that it has some excitement and interest for a number of fundamental problems of cell biology. Now, I really want to make two points.

First of all, there is an assembly controlling the mobility and the distribution of cell surface receptors. And second, some of the elements of that assembly seem to be involved in signaling and particularly in control of nuclear responses to particular surface events during the initiation

of mitosis or cell division. Now, as I mentioned, although the work I shall discuss is on lymphocytes, other cells have been studied in this regard by other workers with similar results. And in fact, I should tell you that the main approach I shall take will be first to describe the use of a plant protein

such as a lectin, concanavalin A, as a tool to examine this whole concept and then to consider the kinetics of lymphocyte responses as a basis for understanding whether there are surface interactions involved in the control of cell division. The proper beginning of my subject, I believe,

is in fact cell division as expressed in clonal selection. May I have the first slide, please? Now, on this slide, you see diagrammed the typical response of a repertoire of lymphocytes that have arisen during embryonic development, each to express on its surface

a singular antibody molecule indicated by the Arabic numbers on each cell. An antigen polling or going through these antigen binding cells may bind to the surface receptor antibody with a sufficient avidity or affinity in order to stimulate maturation and division

of that particular clone of cells as indicated on the bottom of the slide. Now, you will notice that although there are several different responses to a given antigen indicated by the black dot binding at the surface of this assembly of cells,

the fact is that a very small proportion of these cells will respond to any single antigen. And therefore, this process of maturation and mitosis is very difficult to study using any particular antigen, even a very pure one. For this reason, people have turned their attention to other kinds of molecules.

Although I should emphasize, may I have the next slide, that it is the antibody molecule shown here in a schematic version consisting of symmetrical structure, two light and two heavy polypeptide chains, the collaboration of which leads to a molecule having two antigen binding sites,

which confers specificity on the entire immune response. For the purposes of my remarks, however, we are more interested in the nature of the signal that occurs when the antigen binds or when any other mitogenic molecule binds to the cell surface using the lymphocyte as an example. And for that reason, my colleagues and I

have turned our attention to a molecule known as conkinavalin A, a lectin, a molecule which is taken from the plant known as the jack bean. This molecule has the singular property, may I have the next slide, of binding to certain carbohydrates. Could I have the lights down, please?

And this is a representation of its three-dimensional structure as solved by my colleagues, Becca and Ricky. And the main point I wish to make in showing you this slide is that this molecule consists of four identical subunits. You can see two of them, one the red subunit and the other the blue subunit

related by a rotational axis. And finally, an identical pair related to those two to form a tetrameric molecule. Each subunit has one binding site for a particular sugar and therefore the entire structure is tetravalent. And I emphasize this because this is important

for my subsequent remarks. Now these molecules, these lectins, or proteins from plants, although of unknown function in the plant, have the curious property of stimulating cell division in lymphocyte populations. And I'm going to devote a good deal of my remarks to that subject, but before I turn to it, let me point out to you that a particular lectin

has a restricted specificity for certain sugars, carbohydrate molecules, which may be associated with proteins on the cell surface. Indeed, it has been known for some time that the cell surface has a coat of glycoproteins consisting of a variety of sugars linked in covalent linkage,

hooked to proteins also by covalent linkage. The attachment of concanavalin A to the sugar moieties of one of these or several of these glycoproteins is the initial and cardinal event in stimulating the commitment of that cell to undergo mitosis. And the very great usefulness of this kind of molecule

and may I have the next slide, not only resides in its specificity for sugars, as you can see here, this particular molecule is specific in that it can bind glucose and mannoseal residues, but cannot bind galactoseal residues. But also, in the fact that these molecules

will stimulate a great number of different lymphocytes and thus we can study a very large proportion of the lymphocyte population in terms of events at the cell surface as well as the cell division itself. Now I should turn now to a consideration of the kind of assay of cell division

or commitment to cell division that I shall call mitogenesis. This consists in fact of two separable phenomena. The first, a maturation event akin to that induced by antigen, which is called blast transformation. May I have the next slide please? And the second, the induction of DNA synthesis.

And here on this slide, particularly in the bottom frame, you see the stages in blast transformation and induction of DNA synthesis that occur in a lymphocyte that has been placed in a tissue culture dish and exposed to binding of concanavalin A to its surface.

On the left you see the small lymphocyte characteristic of your immune system and capable of circulating in your body either as a thymus-derived lymphocyte, a T lymphocyte, or a bone marrow-derived lymphocyte or B lymphocyte. In this case you are seeing a T lymphocyte which has bound concanavalin A

and some 20 to 30 hours later has enlarged, as you see in the middle cell, to undergo blast transformation which includes RNA synthesis and proteome synthesis. And finally, the replication of DNA, which you notice as an accumulation of a large number of thymidine grains

in autoradiography over the nucleus. So this cell has undergone a very large change in its structure, finally to prepare for cell division. And indeed, this provides us with one of the assays of the activity of a molecule like concanavalin A. If one examines this process of commitment to mitosis

for all of these events precede cell division by a very large number of hours, between 24 and 48 hours, then in fact one sees a very characteristic dose-response curve of the lectin or mitogenic substance. May I have the next slide please? And here is a characteristic dose-response curve

to concanavalin A. On the abscissa you see the dose of concanavalin A in the tissue culture dish containing either human peripheral lymphocytes or mouse splenic lymphocytes, perhaps 10 to the seventh cells for each dish, in terms of the micrograms per milliliter

of the lectin or mitogen. And on the ordinate, you see an indication in terms of counts per minute of the uptake of radioactive thymidine, signaling the initiation of DNA synthesis. Well now, this curve poses two problems. First of all, what is the initial event

that commits the cell to the onset of the ascending limb, which as you see, has an optimal response at about three to five micrograms per milliliter of the stimulating mitogen? And second of all, what accounts for the decrement or the decreasing limb in which the cell is inhibited

from the process of maturation and DNA synthesis? This feature of the curve, which has not really received particularly much attention, is in fact an important feature of this response and of an indication of the control systems at work, for it does not represent killing of the cell

and is entirely a reversible phenomenon. Therefore, how do we account for the stimulation limb and for the inhibition limb? Well, let me proceed directly to the hypothesis that I intend to put forth to you, and then I shall proceed from there to events at the cell surface.

But before I do that, let me say that there is adequate evidence to indicate that the action of this lectin, concanavalin A, is at the surface of the lymphocyte. This is indicated by two main types of observations, the first being that if the concanavalin A is quantitatively removed by competitive inhibition

using a sugar capable of binding to its binding site, for example, alpha-methylmanicide, then all of this process of commitment to stimulation for division stops. Furthermore, covalent linkage of concanavalin A to solid beads of various types

is also capable of stimulating the cell to undergo commitment to cell division. Well therefore, what we have to account for is the early events at the cell surface and how they relate to the initial signaling that commits the cell. And also, we have to describe the various components

of control that are involved in these complex processes. The hypothesis I wish to put forth to you is this. May I have the next slide? That there is, in fact, just under the cell surface, a complex assembly of proteins dedicated to the control and regulation

of signaling at the cell surface. And without going into a great number of details at this stage of my remarks, I would like to point out some of the components. Here you see a segment of the bilayer membrane, the lipid bilayer membrane. I want you to consider that not only to be a model for the lymphocyte surface,

but a general kind of picture of a number of mammalian cells. And the lipid molecules are indicated by those little circles with tails on them. Inserted into this layer are particles, presumably protean known as intramembranous particles

and marked IMP here. And finally, there is some evidence, in fact, that cell surface molecules, such as these glycoproteans with the triangles indicating the carbohydrates, penetrate entirely through the fluid bilayer membrane to put portions of themselves into the cytoplasm itself.

And I intend to present evidence that indicates that certain proteins related to muscle, in fact, known as microfilaments, marked MF, and other proteins known as microtubular proteins, made up of certain proteins known as tubulin, are involved in the regulation and control

of the distribution of cell surface movement, and also that some of these elements, particularly the microtubules, are involved in the regulation of the commitment to mitosis. Well now, let me proceed by showing you an assay, may I have the next slide, based on the observations of Taylor and Raff and their associates.

As you have probably heard, there has been a quiet change in our views of the cell surface over the last several years. One of the key findings has been that of Hardin-McConnell, that the lipid bilayer is, in fact, in a fluid state. And Edidin and his associates have shown that cell surface receptors

intercalated in such a lipid bilayer can, in fact, move or diffuse within the two-dimensional plane of the cell membrane. They do not stand fixed, just as you saw in those pictures of the F. pyli of E. coli that Dr. Ochoa showed, but in fact, they are capable of diffusing in the plane of the membrane.

Now Taylor and his associates made a signal observation. They took bivalent antibody, of the kind that I showed you in my earlier slide, capable of binding and cross-linking cell surface receptors. Specific for one of them, let us take the immunoglobulin molecule itself on a mouse lymphocyte, for example, to which one is adding antibody made in a rabbit

against that immunoglobulin receptor. Taylor and his associates labeled this anti-antibody with fluorescein so they could observe its distribution in the fluorescence microscope on the surface of the cell. And what they observed is shown on this slide.

At the first place, they showed a diffuse distribution of fluorescence. And shortly afterwards, the fluorescence was gathered into a series of patches, as you see in B and below. And finally, if the cell were capable of active metabolism, these patches were gathered actively into a cap at one pole of the cell.

Now this ipso facto implies that receptors are mobile. And through a series of elegant studies, it has been shown that patches are the result of a diffusional nucleation process, which does not depend on metabolism, but as I mentioned, caps. The gathering of patches into one pole of the cell,

thereby denuding the cell surface of that specific kind of receptor, are formed by an active process. What does this have to do with my proposal? Let me, in the first place, show you a picture of caps and patches. And here we will have to have the lights out. Could I have the next slide, please?

On the lower right or left, you see the diffuse picture of staining of mouse splenic lymphocytes by anti-immunoglobulin that has been labeled with fluorescein so that it becomes visible. Within a minute or so, on your upper right in the part of the slide mark B,

you will see that these diffuse distributions have been altered into an accumulation of patches. And between five and 20 minutes later, under the circumstances of this experiment, all of those patches were gathered, as you can see, into caps in the part of the slide marked A.

If we added a metabolic antagonist, such as dinitrophenol or sodium azide, the process would stop at patches because metabolism was inhibited. Now, the observation that Dr. Yohara and I made was the following, that if concanavalin A, which is unlabeled, is first added to these lymphocytes

so that it can bind to the glycoproteins at their surface, then both patch formation and cap formation are inhibited. This, in fact, can be shown by electron microscopy to take place also at the level of the individual receptors.

And indeed, it is a reversible process. So concanavalin A restricts the mobility of the receptors in such a way that neither patch formation nor cap formation can occur. And it does that at doses so low as to represent less than 10% of the occupancy of the glycoproteins at the surface of the cell.

May I have the next slide, please? Here on this slide is indicated the fact that this entire process of the inhibition of patch formation by freezing the cell in the diffuse distribution is a reversible process. And the experiment by which this is carried out is a simple one.

One simply adds alpha methyl mannicide in sufficient concentration to compete with the binding of the concanavalin A at the cell surface, remove the concanavalin A, and then those cells will then enter the cycle, undergo patch formation and cap formation. Well, not only does the concanavalin A

inhibit the motion thereby of immunoglobulin receptors, but as far as we can tell, it inhibits the motion of the receptors of a very great different number of types, and not only on the lymphocyte, but also, for example, on the fibroblast. May I have the next slide, please? I've listed on this next slide some of the surface receptors,

the mobility of which seems to be inhibited by binding of concanavalin A. They include, as I have shown, immunoglobulin receptors on B lymphocytes, certain alloantigens or surface protean molecules that are characteristic of T lymphocytes known as theta antigens,

various lectin receptors of different specificity and therefore different kinds of glycoproteins, and finally, histocompatibility antigens on both lymphocytes and fibroblasts, and now the list is much longer. Indeed, we have found no exception to the fact that binding of concanavalin A to the cell surface in rather minute amounts

is capable of restricting the mobility of cell surface receptors. Well, how can we account for this reversible restriction of the mobility of receptors by concanavalin A when the concanavalin A is not, in fact, binding to those receptors? Well, in order to account for that, let me ask the question,

what structure in the cell might be responsible? There are, in fact, known to be certain proteans under the cell surface, distributed in particular ways in mammalian cells, and it has long been suspected that these proteans have not only to do with the morphology of the cell, but also with its motion. These include, and this list is not exhaustive,

proteans such as microfilamentous proteans, which are found just under the surface and are related to muscle actin. Now recently, it has been shown that some proteans are related to myosin, and finally, there is a set of proteans underlying the cell surface 100 angstroms below,

known as microtubules, which are capable of very rapid assembly in the cytoplasm, and which are ubiquitous in the cell. It is, in fact, these structures that I wish to direct your attention to, for I believe the evidence supports their role in this modulation of cell surface receptor mobility.

May I have the next slide, please? Here, in fact, are some electron micrographs of lymphocytes structures, looking at the surface on the top, and then making a section, and the dark structure at the bottom of each slide is, in fact, the nucleus. Now, you will notice, I hope it is visible,

that there are long tubular assemblies about halfway down from the surface, over here, and circled over there across section through such an assembly, which is known as a microtubule, about 240 angstroms in diameter. And on the right, I don't believe it's very visible

under these lighting conditions, there is a microvillous, a projection of the lymphocyte, under which there's a considerable network of microfilament, or actin-like structures. We have recently demonstrated that if we attach a lymphocyte to a stable surface,

and disrupt it in such a way that we can take a look at the inner lamella of the cell membrane, then antibodies directed against the muscle protein actin will, in fact, stain the microfilamentous structures, and I believe on the next slide, you will see a picture of such an experiment.

Yes, on the left is a phase contrast micrograph of the ghosts of lymphocytes that have been disrupted mechanically and by fluid disruption in such a way as to leave the inner portion of the lamella visible. Actin antibodies made in the rabbit against smooth muscle, actin and labeled with fluorescein,

light up the inner portion of this membrane in the way that I've shown, and this is confirmed in the electron microscope. These appear to be the microfilamentous structures just under the inner lamella of the cell membrane. Now, in fact, may I have the next slide? There is a way in which both the actin-like structures

and the microtubules may be dissociated. We have observed, and De Petris has observed, that if the drug cytocolazin B, which is known to disrupt microfilaments, is applied to the lymphocytes in culture, then capping is prevented,

presumably because this actin-like or muscle-like molecule is no longer capable of exerting its action on the patches. In turn, we have also observed that a molecule known to disrupt microtubules called colchicine, and shown here, capable of inhibiting the assembly from tubulin of the microtubular structures of the cell

will in fact reverse the modulation or restriction of cell receptor mobility induced by concanavalin A. And colchicine neither binds to the concanavalin A receptors nor to concanavalin A itself. And on the next slide, I believe is a summary of the experiments which indicate

that a whole variety of drugs that are specific for the disruption of microtubules are capable of in fact reversing the inhibition of cell surface receptor mobility induced by the binding of concanavalin A to the cell. Let me make that clear. The concanavalin A remains bound to these cells,

but the addition of colchicine, colcimid, vinblastine, vincristine, or a variety of other drugs related and capable of dissociating the microtubules reverses the process of inhibition. The use of drugs which are related to colchicine but are known not to bind to microtubules,

such as the ultraviolet inactivated product, lumicultricine, fails to reverse the inhibition of receptor mobility. In fact, one could explain this entire process perhaps by some other route. It has been tempting, for example, to say that perhaps the whole restriction

of cell surface receptors does not have anything to do with these structures of microfilamentous type or of the microtubular type, but instead represents simply the induction of a network on the cell surface of concanavalin A molecules which restrict movement or a phase transition in the lipid bilayer membrane.

Well, the latter has been ruled out by examination using spin labeling techniques. Indeed, the fluidity of the membrane is unchanged, and I wish now to show you that, in fact, the restrictive or modulation effect of mitogenic lectins such as concanavalin A

can be induced as a propagated phenomenon with no free concanavalin A present. In order to do this, I have to show you a technique that we invented. May I have the next slide, please? Called fiber fractionation. This technique was, in fact, the result of a great deal of effort and failure

in attempting to fractionate cells upon chromatographic columns. It is, in fact, the result of an afternoon of violin playing in which I thought that perhaps we might bind cells to a horsehair of violin bow. And when I came to the laboratory at that particular moment, my colleagues

were quite certain that I had turned around the bend. But in fact, one of them suggested that instead of a horsehair, we use nylon fibers, and the principle of this technique, which is an effective technique for both examining cells in situ and fractionating them, is shown on this slide. One derivatizes the fiber by covalent means.

I won't go into the technical details. And then shakes the cells onto a dish consisting of a framework of these nylon fibers in such a way that they collide with the specific molecules to which the fibers are linked. And then cells bind to the fibers, specifically, the rest may be washed away,

and you may recover the cells. But the purpose I wish to use this for now is the fact that it allows you to give a cell an address. May I have the next slide? And therefore, to examine the local application to a great variety of cells of a particular agent. And I would direct your attention to the last portion of this slide, Mark D,

in which you see blown up under the fluorescence microscope, a concanavalin A-derivitized fiber on the bottom, and a single lymphocyte which is bound to that fiber, and to which fluorescein-labeled anti-immunoglobulin has been attached. And you can see that the immunoglobulin receptors,

as revealed by this probe, are diffusely distributed all over the cell perimeter. No free concanavalin A is present. The concanavalin A binding the cell touches the cell only at the locus where the cell touches the fiber. And if this experiment is repeated with an irrelevant antigen, for example,

that against a haptene such as dinitrophenol, one does not, in fact, see this restriction of mobility. One sees patches and caps. If one now takes this restricted cell and adds colchicine to the preparation, then one sees patches, and ultimately, as you can see over here, that cell which is bound to the fiber also caps.

Therefore, the local application of concanavalin A to a small portion of the cell surface is sufficient to induce the restriction. This restriction, in turn, therefore must be a propagated one, and also it can be reversed by colchicine. This effectively, I believe, rules out any hypothesis that says

that the concanavalin A simply acts as a fence. Well, given all of this, what can we make of the actual signal that says propagate some kind of change within the cell that will alter the receptor distribution and mobility?

And here, I want just briefly to describe one experiment. May I have the next slide, please? Here, what we have done is taken concanavalin A, knowing its structure, and have derivatized it by adding succinyl groups to its side chains in such a way as to increase its net charge and disrupt the molecule into two dimers,

each of which has, therefore, only two sugar binding sites instead of four. This dimeric succinyl concanavalin A does not, when it binds to the same set of receptors, restrict either patch or cap formation, nor does it induce patch or cap formation.

May I have the next slide? This is indicated on the next slide, which shows that the derivatization of the molecule to reduce its valence from four to two is sufficient to remove the entire effect, despite the fact that the specificity of the dimeric molecule is identical with that of the tetramer.

And now, if one adds divalent antibodies to the dimeric concanavalin A that is already bound to the lymphocyte, its properties are converted to those of the tetramer. And to us, this indicates that the adequate signal is most likely cross-linking of some subset of the cell surface receptors.

Now, in fact, what I have said is that this restriction or modulation of cell surface receptor mobility, which we have also observed in fibroblasts, is a locally inducible, propagated phenomenon, which can be reversed by the addition of colchicine

and therefore by the disruption of the submembranous microtubular structures. And now I would like to turn my attention to the question that I raised initially in this lecture, and that is, what has all of this got to do with the signaling for mitogenesis, or the commitment of the cell to mitogenesis?

To do that, I must show you, in fact, an hypothesis for how this whole process works. May I have the next slide? Our hypothesis is, as I indicated earlier, that under the membrane, there are a set of structures related to muscle as well as a set of structures that are connected with tubulin,

and that the receptors can exist either in an anchored state, the A state, or in a free state, capable of diffusing on the cell surface. And that the attachment of a multivalent molecule, such as concanavalin A, cross-links some subset of these receptors, changing a set of equilibria amongst the microfilaments,

the through and through receptors, and the microtubules in such a way as to induce the polymerization of the microtubules, and therefore propagation of this restriction of cell surface receptor mobility, attaching other receptors to that whole assembly. Now as I mentioned before,

this assembly and its components undoubtedly have something to do with cell movement, as a number of other workers have shown. But I want to ask the question, what is the relationship of all of this to the commitment of a cell to divide? Now let me show you again, therefore, the cell surface, may I have the next slide,

the binding of both concanavalin A and the dimeric derivative that I just showed you to the cell surface. In the first case, as I've already shown you, the curve consists of a stimulation and an inhibition, which is now on this slide much compressed, and is shown in the dark circles. If one, however, uses the dimeric derivative,

although it is capable of the stimulatory limb, the inhibition is no longer observed over several logs of the dose. And this indicates that the dimeric derivative is perfectly capable of stimulating the cell, but it is not capable of inhibiting the cell. You will remember that only the tetramer

is capable of modulation or restriction of cell surface receptor mobility, and we suspect, in fact, that the initial signal for commitment of the cell is blocked by the modulation of cell surface receptor mobility. Now that is not proven, but all of the dose relationships and the specificities are consistent with that hypothesis.

What about the stimulatory limb, finally, the most important part of mitosis and the signal to divide? Let me show you a kinetic assay that we have developed to examine whether it might be that the microtubules are involved in the regulatory signal.

The protocol for this is shown on the next slide. The idea is simply to stimulate the cells at the optimal dosage of the mitogenic plant proteome, and then to separate tissue culture dishes, add the inhibitory sugar, removing the lectin from the cell surface, and thereby removing the signal to divide.

And after an appropriate period of time, adding labeled radioactive thymidine in order to measure both the incorporation of that thymidine into DNA, as well as the appearance of autoradiographic granules of synthesized DNA over the nucleus, and blast transformation.

And when we did that experiment, we were surprised to find that, in fact, the response of the cells is not graded within each cell but that cells are recruited as a function of time, up to 20 hours in terms of their commitment to divide. May I have the next slide? Here is, in fact, the result. On the top, one sees that the average number

of grains of incorporated thymidine is independent of the time of exposure for all of the cells, but that for three different assays, the incorporation of the thymidine into DNA and DNA synthesis, indicating commitment to division, blast transformation as indicated by the morphology, and radioactively labeled blasts,

the time dependence of the cells was all the same. The system became independent of removal of the mitogenic substance only after about 20 hours. And this result is consistent only with the idea, may I have the next slide, that each cell is recruited in time,

that the rate of DNA synthesis is independent of the length of exposure, that increasing numbers of cells become committed, and therefore, that each cell has its own clock, if you will, and that finally, cellular commitment to mitosis is an all-or-none phenomenon. Well, why am I telling you all of this? It is to prepare you, in fact, for an experiment in which we repeat the entire protocol,

but do not remove the canary. We simply add colchicine in order to disrupt the microtubular structures and ask what the kinetics of commitment is in that case. Could I have the next slide, please? Well, in fact, as you can see in this slide, which is on the top, a repeat of the earlier experiment,

the inhibition by colchicine follows very similar, if not identical, kinetics to that inhibition which results from the removal of concanavalin A. And you'll remember that I mentioned to you before that colchicine does not bind to concanavalin A, nor does it bind to its receptors. This inhibition, which is paralleled

by the disappearance of microtubular structures, cannot be induced by the use of the analogous compound, which is inactive, lumi-colchicine. And now, therefore, we have, as I show on the next slide, a resume of this effect that indicates that a component of the surface modulating assembly seems to be involved in the regulation

of commitment to mitosis. This colchicine inhibits the mitogenic stimulation much before any blast appears or DNA synthesis or mitotic figure appears. It does not act by killing the cells or blocking cell division, because we have tried it on cells that have been brought to the G1S boundary

of the mitotic cycle, and there it does not inhibit the process. The related drugs, vinblastine and vincristine, inhibit. It can be reversed if the colchicine is removed. And these kinetics that I show you are, in fact, consistent with the notion that the inhibition is very close to the time of commitment.

And therefore, we have the idea, may I have the last slide, please? We have the idea that somehow there is, in fact, a macromolecular regulatory assembly of the signal of a cell to divide. And some experiments done by other workers, notably Temin on fibroblasts, show consistent results,

which, in fact, would lead me to believe that it is not simply a matter of transport of ions, but a matter of transport of ions controlled by large macromolecular regulatory assemblies that are responsible for the key notion of initiation of cell division. Now, I don't want to leave you the impression

that we understand this process very, very deeply. Much remains to be done, in fact, to isolate causally the various events. And indeed, in this slide, what I have tried to do is show you some of these key events. The first event is probably a cross-linking of the cell surface receptors

that leads to a change in the transport of ions such as calcium. And indeed, it has been shown very early that a stimulated lymphocyte will receive a pulse of calcium ion through its membrane. Secondly, there appears to be some kind of feedback relationship between this calcium input,

probably the result of the formation of channels through aggregation of these receptors, and microtubular arrays, which have been shown by other workers to be influenced both by calcium ion and by cyclic nucleotides. And finally, the assembly process of modulation appears to cancel this entire process

of calcium transport. Freezing, in other words, by mitogenic lectins through a propagated assembly is sufficient to block the entire assembly, and therefore, to block commitment of the cell to mitosis. Well, ladies and gentlemen, it has been my intention in this lecture to draw your attention to the existence

of several new phenomena at the cell surface. These provide some support for the view that the cell surface receptors do not merely float around, but they are actually capable of modulation and control. And the perturbation experiments that I've outlined point to the involvement of microfilament and microtubular structures in this function.

But of course, proof of our hypothesis that there is such a surface modulating assembly must come, and can only come, from direct structural evidence of the interactions that are postulated, an interaction between a through and through receptor, and microfilaments, and possibly myosin, and thereafter, an interaction

of the actin-like microfilaments with those of the microtubules, and the demonstration that this is a reversible phenomenon. The variable response of individual resting cells with recruitment to a mitogenic signal, I think is a fascinating observation, a new and somewhat unexpected property that I have not emphasized in this lecture,

but one which I believe supports the notion of a stochastic or statistical control of early events in the commitment of a cell to divide. And perhaps there is a relationship between this kind of response and the state of the various components of this surface modulating assembly. Finally, as I've discussed here,

there is now some evidence to suggest that microtubular function is involved, not only in modulation, but also in regulation of the early commitment of a lymphocyte to divide. And I believe that it is particularly intriguing to consider that it is through such an assembly that the coordination of cell surface receptor mobility,

global cellular motion, and the coordination of growth control is mediated. And while due caution, I think, must be exercised in the interpretation of the drug effects that I've shown, the specificity seems remarkable. And I think we can look forward, in fact, to an exciting time extending these kinds of experiments

to the immunological responses, but even more exciting in extending them to various differentiated cells, both normal and neoplastic. And of course, it may be that in the not too distant future someone will find out what the true activator in the cytoplasm is that says to the nucleus,

go, be committed, and divide. And when that day comes, I believe it'll be a very happy day indeed. Thank you.