SLIDES & TRANSCRIPTS
Tuesday, September 14, 2000

Receptors and Signal Transduction Pathways as Targets for Therapy in Small Cell Lung Cancer
Enrique Rozengurt, PhD

DR. SAXMAN: The next speaker this morning is Dr. Enrique Rozengurt. Dr. Rozengurt, is a professor of medicine at the UCLA School of Medicine and has done extensive work in small cell lung cancer in regard to signaling pathways. He is going to speak with us this morning about receptors and signal transduction pathways as targets for therapy in small cell lung cancer.

DR. ROZENGURT: It has already been mentioned many times that one of the major characteristics of small cell is the production of multiple hormones and neuropeptides. This is not an extensive list, but it gives a feeling of the large number of molecules that are produced by these remarkable tumors. It has been known from the early eighties from the work of John Minna and Terry Moody (who is in the audience) that bombesin and GRP are produced by these tumors. Also, it has been observed that the number of other hormones like cholecystokinin, gastrin and neurotensin, vasopressin in particular are produced by these tumors in remarkable degree, and especially some of them are responsible for the paraneoplastic syndromes that we already heard about.

In addition to producing these peptides, which was of great interest for diagnosis and for prognosis in some cases, we and others have been interested in the possibility that that these cells respond to these peptides.

TOP

The work done by a number of groups, including John Minna, Paul Bunn, and our own led to the idea that there are multiple neuropeptides that actually stimulate the colony formation of small cells. I am listing here the numbers which were particularly studied by a number of groups. In some cases, like bombesin, CCK, gastrin, neurotensin and vasopressin, there is very good evidence for production of peptides by the tumor cells. I would like just to give you a few examples of the kind of effects we are observing in vitro,

TOP

taking the work for example, with gastrin. The effect of gastrin was studied in the 510 cell line. We have some colonies in semisolid medium, but the addition of 10 nanomolar gastrin resulted in a very dramatic increase in colony formation.

TOP

These effects were actually due to both an increase in the number of colonies as well as in the size of the individual colonies, indicating that we have more cells formed in colonies as well as an increased multiplication of the cells in each one of these colonies.

Gastrin is an interesting example. Gastrin binds to a receptor called the CCKB receptor, and there is a great deal of interest lately in the finding that there are many types of small cells, including tumors, in which the CCKB receptor is expressed in a considerable degree. There has been a proposal that the CCKB receptor not only could be a target, obviously for antagonist, which I will talk in detail about later, but it could also be interesting for imaging.

TOP

Other peptides like bombesin, bradykinin, vasopressin, again show increases in both numbers as well as size in colonies.

TOP

All these data and much more data produced by many people, many of which are in the room, led to this idea that small cell carcinoma is driven by multiple autocrine and paracrine sequences which are mediated by regulatory peptides.

So this concept is different from the very early concepts in which we have a prominent single loop; we are now moving to multiple loops.

TOP

In addition to the sort of regulatory neuropeptides, there are also very important loops that are mediated by tyrosine kinase receptors. I would like to emphasize those as well.

Tyrosine kinase receptors in small cell: the major loop that we have to think is stem cell growth factor and its ligand C-kit. This has been detected in about 70-75 percent of small cells and tumors, and both the ligand and the receptor are present.

The other interesting system of tyrosine kinase receptor is mediated by the hepatocyte growth factor also known as scatter factor, and its receptor C-met.

In the case of small cells, actually C-met is present in a fair number of small cell lines and I think tumors, but the ligand is present in a minority of the cases. It is produced to a large degree by fibroblasts, and therefore the hepatocyte growth factor C-met is very likely to be a paracrine loop of considerable interest. There is also a loop, which has been known for years, mediated by IGF and its receptor.

TOP

Here we have another example of stem cell factor inducing colonies, and again you can see the spontaneous colony formation. This is, again, the 510 cell line, and in the presence of 3 nanograms per ml of stem cell growth factor, there is a dramatic increase in colony formation.

TOP

The effects can be dose dependent and one can actually see the dose responses. Maximal effects are obtained at 10 nanograms. They are seen in the 510, and as I mentioned before this is seen in many other cells. Here are examples with 345 and 69s in which the SCF is inducing cell proliferation.

TOP

Hepatocyte growth factor, C-met receptor present also in the 510, 345 and 69. These are numbers of colonies induced by increasing concentrations of hepatocyte growth factor, and you can see that one can actually obtain nice dose responses in all these cells. As I mentioned before, C-met is very common, but the ligand is not produced—and then of course this is probably a paracrine system.

TOP

So, we can actually now think that the formation of colonies, growth, and by extension the growth of these tumors, is probably mediated by two major signal transduction pathways.

One is mediated by G protein coupled receptors, those that are listed on the left and has been mentioned before. I also would like to emphasize those on the right, the tyrosine kinase receptors, in particular the C-kit and C-met. There is a new element in these pathways, the new discovery of the ephrins, which also have been found to be expressed, at least in some small cells. So, in our initial work when we discovered that bombesin was a mitogen, for example, in fibroblasts many years ago and, also, some of the others, one of the most interesting features that we observed is that there was a dramatic synergistic effect between the mitogenic activity of bombesin and the co-administration of, for example ,tyrosine kinase receptor, in that case with IgF but also EgF.

So I would like to suggest that these loops that we are seeing here are very likely to interact and cross talk in many complicated ways in which probably when we are thinking about one loop over on the left of this scheme, we actually are seeing a synergistic effect between this loop and the tyrosine kinase receptor on the right.

TOP

So that gives us some ideas of possible targeting for these receptors and the ligands, and some of them were already mentioned, but we have to think, and antibodies against the ligand has been tried against GRP or receptors. In the case of receptors, probably EGF is an example. They are not relevant for small cells. They are more important for non-small cells, but there are other possibilities. The next speaker will surely allude to the issue of angiogenesis, and there are much more possibilities there.

Tyrosine kinase inhibitor is the obvious way for targeting C-kit and C-met and in fact, C-kit is a tyrosine kinase receptor with many structural features similar to the PDGF receptor. Split tyrosine kinase receptor and tyrosine kinase receptors have been developed. Tyrosine kinase inhibitors have been developed for PDgF receptor and also active against C-kit and they are entering into more interesting small cell trials as possible chemotherapy, and I think that they are particularly exciting.

I mention two other possibilities here. Neuropeptides are synthesized in these cells through a number of steps, but the final step is amidation and theoretically inhibition of amidation is a possible therapeutic intervention. Neutral peptidases have been worked on by Paul Bunn. He has shown that neutral peptidases could inhibit responses of small cells, potentially another way of tackling these autocrine loops, but the most interesting possibility for development in the future is that of antagonists, either specific receptor antagonists or what we call broad spectrum neuropeptide antagonists.

TOP

If we think in terms of a classic single autocrine loop which is, for example, the bombesin loop, it is easy to imagine that a monoclonal antibody or a specific antagonist will be actually the ideal way of targeting this loop. I think that now we understand that there are a number of reasons why this is not going to work. First because not every cell is actually expressing ligand or receptor, and second because there are multiple loops, not a single loop.

So the idea of a highly specific antagonist is unlikely to work at least in the majority of the cases. It could be that a careful selection of patients with determination of ligand and receptor could be a way to target certain loops. I would like to just leave it at that stage. We can discuss that possibility more in the breakout session,

TOP

but the interesting biologics that are emerging, and that is something that in a way we started and Paul Bunn immediately came with more work and more recently Gary Johnson has come with very interesting ideas about mechanism. I am not going to talk much about mechanism. I would like to really focus on the effects of these molecules and what they are—a broad spectrum antagonist. They are able to block the binding and the biological activity of multiple neuropeptide receptors.

TOP

The first one that we really worked with was the substance P antagonist, which is heavily substituted by D-amino acid and consequently it does not degrade so quickly as substance P.

Substance P, of course, degrades extraordinarily fast. These molecules are very stable because they have all these D amino acid substitutions. The first one was sufficient to show in the lab many years ago to be able to inhibit bombesin and vasopressin, in addition to substance P mitogenesis. Subsequently we found that substitution of the glutamine in position 5 by D-phenylalanile resulted in an increase by 5 to 10 times of the inhibitory potency. More recently, we actually made further substitutions in this position that appears to be critical and found that the D;tryptophane substitution produced a further increase in the inhibitory potency. Just to give you a feeling of what we are talking about,

TOP

I show you an experiment which was done in fibroblasts in which we measured the incorporation of thymidine to DNA. These cells have been a model system that we used for many years in my lab, partly because they respond to multiple neuropeptides. Here we have an experiment in which we stimulated cells with bombesin or GRP. Of course, they are bound to the same receptor, Vasopressin, bradykinin, vistine receptor, tyrosine kinase pathway, EGF, or we also stimulated with forskolin or phorbol dibutyrate to increase AMP or activate PKC just to bypass receptors and get directly to signaling pathways. It is very obvious here that either antagonist that has a D-phenylalanine in position five or the D-tryptophane in position five, the yellow one is a very important inhibitor of DNA synthesis induced by bombesin, GRP, vasopressin and bradykinin but did not have any effect—neither inhibitory nor enhancing effect—on the EGF, forskolin or PBD. We have more recent data to really extend all these findings with dose responses in more conditions.

TOP

Just to give you another example, this is a dose response of bombesin inducing DNA synthesis. Again, we have here the two antagonists that I mentioned before. You can see that the antagonist induces a dramatic displacement of the dose response, and at high agonist concentrations we can obtain a complete reversal of the effect.

This is interesting because it suggests that this antagonist behaves in a competitive fashion, rather than binding to another unknown receptor and producing a negative signal of some sort. I think that data of this type now is also available from other labs, particularly from Gary Johnson.

TOP

Now, moving to small cell, what we found—and now I am sort of focusing on the newest antagonist with the tryptophane in position five—is that it is a very important inhibitor of drug in liquid culture, and this is a very dramatic inhibition of the increasing cell number. It is very likely that this inhibition is a complex mixture of inhibition of growth and apoptosis taking place at the same time.

TOP

There is also, inhibition of colony formation. This is the spontaneous colony formation inhibited by compounds with the tryptophane, less potent but still quite effective the D-phenylalanine. What I like to emphasize here is that there is a difference in potency in the small cell as it was also in the fibroblast system.

TOP

We have moved on, and we have done experiments in nude mice. I am showing here an example which has essentially two different experiments plotted, and here we have given this molecule for a period of 7 days and essentially there is a very clear retardation of growth. When the treatment is suspended, there is inhibition of growth for some time, and subsequently some re-growth is observed.

TOP

So with that and much more data that, of course, I am not going to show, but many of which is published, we have been able to team up with a biotechnology company, have been able to do a large-scale production. I found that this is one of the major rate-limiting steps in all the studies, at least that I have been involved.

They have been able to produce 100 grams of this peptide to really do the sort of studies involving formulation, pharmacokinetics, and toxicology. Sufficient data to persuade an ethical committee in England and then they were moved into a Phase I clinical trial, which was largely conducted by John Smythe in the Edinburgh unit of the Imperial Cancer Research Institution that I was with before coming to UCLA This has continued, and the Phase I has been completed. This is an interesting aspect of the problem to talk about later in the breakout sessions, as well as the possible mechanisms that are involved.

TOP

Now I would like to move on from the receptors to the signaling pathway, and we know that bombesin here taken as a paradigm for the other neuropeptides binds to a seven transmembrane main receptor and the areas that are occurring, both in time and also in terms of how they were discovered. This is a rapid activation of phospholipase CB mediated by the heterotrimeic G protein GQ and that results in the formation of two major groups of second messengers. One is the inositol polyphosphates that are responsible for calcium organization, and the other is the oxyglycerol that activates PKC. I would like to emphasize that this pathway is actually important in small cell and is the major pathway that leads to the activation of the MAP kinase cascade. We already heard that ras is not mutated in these cells, and these cells do not tolerate actually ras activation. However, they do activate in response to peptides, MAP kinases, and they do that primarily through PKC.

MAP kinase, of course, is very important in the phosphorylation of transcription factors and, also, of other kinases like the P90 or RSK.

TOP

Peptides do activate MAP kinase, and we here have three different cell lines, peptides like bradykinin, neurotensin, galanin are showing increases in ERK activation,

TOP

and this activation can be blocked by either chronic treatment with phorbol ester. This is a maneuver to reduce the endogenous level of PKC, and you can see that neurotensin, for example, or galanin stimulation is blocked by these maneuvers, and the important point here is that these are mediated through PKC pathways. The same can be obtained if one measures the p90 which is the downstream target of the ERK.

TOP

Now, the reason for bringing these data because the inhibition of these pathways has very dramatic effects on colony formation, and we can see here, for example, that colony formation can be induced by phorbol esters or by peptides and completely blocked by the Parke Davis inhibitor that blocks the ERK pathway. The same happens in these other cell lines stimulated by galanin. So, this is an interesting point in which in part the cells do not tolerate an upstream element like ras that will continuously activate this pathway, but yet the pathway appears to be required for colony formation. This is a very interesting biological situation, most likely is related to the kinetics of pathway activation.

The pathway needs to be activated in a transient fashion, rather than in a continuous fashion. When it is activated in a continuous fashion by ras mutation or by other work that has gone in the literature with activated raf results in a complete inhibition of these cells.

So continuous activation of the pathway is inhibitory. Transient activation induces cellular proliferation. Here on the lower panel we have an inhibitor of PKC which is the precursor of this pathway and again producing a very strong suppression of colony formation.

TOP

So, ERKs and PKCs are undoubtedly interesting targets that can come in this particular scheme and there are potential clinical trials that come in with inhibitors with PKC. There is a very interesting study that came in colon cancer and inhibition of the ERK by the group at Parke Davis, demonstrating that in fact there is a potential biological activity with minimal toxicity to the animals.

In addition to transcription factors, the MAP kinase also activates and phosphorylates phospholipase A2 and it has already been mentioned that arachidonic acid release leading to COX-2 mediated formation of eicosonoids is an interesting area. This is an area that again will be quite important to consider as a potential target.

TOP

In some cases, another element that flows from PKC is that of the activation of the p70S6kinase. This enzyme is attracting a great deal of attention. It is a very important enzyme in the regulation of the protein synthesis and growth of the cells, the mass growth of the cells, and

TOP

it can be inhibited by the immunosuppressant rapamycin.

TOP

Rapamycin is an immunosuppressant, and the enzyme is present in two forms. One is called the alpha-2, which is p70, and the other is the p85. This is a nuclear form. It is primarily cytosolic form, but the point that I want to make is that this enzyme is present in a constitutive active form in small cells. So this is a very interesting point. This enzyme is regulated through multiple pathways. I showed you before the actual pathway coming from PKC, but tyrosine kinase receptors activate this pathway through a PIC kinase pathway. It is very likely that this pathway is reflecting autocrine loops of tyrosine kinases that are there.

Rapamycin in a dose-dependent fashion induces the dephosphorylation of the enzymes. So, this is a retardation assay. You can see the enzyme is retarded here. You add rapamycin, and it is desphosphorylated, and it comes to an accelerated migration, and that is the same for either form. In fact, the situation has become a little more complicated more recently with the cloning of a second form, a beta form of these enzymes.

TOP

The reason for introducing it is because the enzyme is actually rate limiting for growth in small cell carcinomas. We can see again that there is inhibition of growth. This is liquid culture growth, and these are dose responses the cell lines showing that rapamycin is a kind of potent inhibitor of these cells, giving really the idea that the P7 kinase is a new target for these cells.

TOP

Here we have some colony formations with rapamycin and in the 510 cell line we have data with hepatocyte growth factor showing that the tyrosine kinase pathway is, also, blocked by these agents.

TOP

So in addition to all these downstream elements coming from PCK we really know very little about the immediate targets of PKC. We have very little knowledge, not only in small cell but in fact in most cells, and we don't understand in many cases the precise mechanisms. We know that there are 11 isoforms of PKC, and very likely they have different functions, but we don't understand their downstream targets. We did clone in our lab one target that is directly activated by PKC. We call it protein kinase D. I am not going to dwell much on this, but it suffices to say that PKC can directly phosphorylate and activate protein kinase D, and we are working now on this particular pathway.

TOP

In small cells, protein kinase D is very prominently expressed, and here we have these cell lines responding to phorbol ester. This is an autophosphorylation assay. So, we are here measuring the activity of the enzyme, and within a minute of addition of phorbol ester we see a very dramatic increase in this particular enzyme. This effect is mediated by PKC.

TOP

A few years ago we found that in addition to all the second messenger pathways that are described here, bombesin GRP also induces tyrosine phosphorylation. That was a very interesting finding at the time because it was really not expected.

TOP

The targets that we identified were focal adhesion kinase (FAK).

TOP

The other protein was called paxillin. Another protein is P130 CAS. CAS stands for Crk, crack associated substrate. These molecules are all tyrosine phosphorylated, and they form molecular complexes.

For example, P130 CAS forms a complex with Crk and more recently we found that FAK can form a very strong complex with Src, endogenous Src, in the undisturbed cells, and that primarily works with fibroblasts. We also found that disruption of the actin stress fibers or focal adhesions blocked this tyrosine phosphorylation, and therefore we positioned them probably upstream in this pathway.

It was known from work from Alan Hall in London that bombesin induces very strongly actin stress fibers and focal addition assemblies in cells and that effect was mediated by Rho. Shortly after, early work from Gary Johnson showed that actually micro-injection of the G alpha 13, the heterotrimetic G protein G alpha-13, and G alpha-12 resulted in Rho dependent formation of actin stress fibers and focal additions in the cells. We subsequently showed that transient transfection of this constitutive active form of G alpha-13 was sufficient to recapitulate some or all these downstream targets and, in particular, the tyrosine phosphorylation.

So, we believe that there is a new pathway which is associated with the G alpha-12, G alpha-13, primarily mediated by Rho, and there are a number of downstream targets which collaborate in the formation of actin stress fibers like the RO kinase and other gene products that are involved. There are multiple proteins in this. This is going to be a very complicated area, but the reason I am emphasizing this area is because it has a central role in cell migration, and this is probably an area that is going to be very important for invasion and for metastasis. We know something about small cells and this pathway. There are tyrosine phosphorylated proteins in small cell, and focal addition kinase has already been found in them. There was one study in which using C3 toxin, which is a way of interfering with the activity of Rho, resulted in alterations in the adhesion of small cell carcinoma.

Specifically, what happens when C3 was added to these cells is the cells appear to each other in a mediated fashion. That is a way in which actually it limits very much the movement of these cells outside and the invasion of these cells.

Another interesting point is that transferase inhibitors have already been mentioned many times during this morning, several times at least, and there is now increasing realization that they do not act through ras, but they are very likely to act through Rho. It is very likely that it is interference of this pathway which happens with the FTIs.

TOP

So, we can put now together a more complicated scheme in which these receptors are actually acting through at least several G proteins. I have only here two, Gq and G12 families, giving rise to really a constellation of effects. One of these that I have highlighted is potential targets which could be really utilized for these tumors.

TOP

Question and Answer

DR. SAXMAN: I think there is time for a question or two.

DR. MUFSON: Do you have any evidence as to which isoforms of PKC might be specific to activation in the small cell system, because the specificity you might hope to get there with a key enzyme or where the key inhibitor would be dependent hopefully on isozyme specificity?

DR. ROZENGURT: We have not done very specifically on small cell, but we have actually looked at a large number of co-transfections with constitutive active forms of various PKCs which were either mutated or deleted in their other inhibitory region. What it comes to is that epsilon and eta are the most potent forms to activate PKD, and they are very well expressed in small cell. ETAC is expressed. ETAC has a rather narrow range of expression, but this is expressed at least in our work and work from other people in small cells. Interestingly, PKD can form a molecular complex with ETAC. The interaction between these two enzymes is very close, very intimate. So we really think that it is the novel isoforms of PKC primarily ETAC and epsilon, the ones that are going to target PKD in that particular phosphorylation cascade.

DR. MUFSON: Do you have any specific novel inhibitors that would target that?

DR. ROZENGURT: I think that there are essentially no specific, highly specific inhibitors that I am aware of. There are some which have preference for conventional PKCs. There is a very disputed inhibitor for PCK delta, but targeting selectively epsilon—what there is actually is antisense that has actually been developed and has been given to animals and has had some antitumor activity. I don't remember exactly the precise model system at this moment.

DR. BUNN: Enrique, in the NCI trial of the anti-bombesin antibody, there was one responder in the Phase I trial. In the Phase I trial from the UK of the substance P derivative, did you reach steady state levels of 20 micromola,r which are sort of the IC50? Were there, if you reached those levels, were there any responders in the Phase I trial?

DR. ROZENGURT: As my recollection is there was no selection of patients with small cell. Any patient that was available was used just to really look at pharmacokinetics, and there was no way of really evaluating activity in small cell.

There was considerable concern in many meetings in which we had at the time to really come up with some kind of end point, some kind of biological end point that will tell us that we are reaching that. In addition to the concentration that you remember extremely well that was in the micromolar range and biologically effective, there was actually a concentration that was really producing some kind of effect, and there were some surrogate measurements of the blood pressure changes. Particularly in arm, one can actually do those experiments. They reach sufficient concentration to block the effects of peptides that were inducing blood pressure changes, but that is as it stands at the moment.

I really think that one of the problems you already alluded to, the major problem of this type of how to move forward, is the availability of the molecule, how to make it really, how to have enough and to actually allow different people to work. I think that I really do hope that that will come up in the breakout sessions, and then we will really think about those possibilities.

TOP