Kelly K. Hunt, MD: Induction of Apoptosis in Human Breast Carcinoma By Overexpression of a Truncated and Mutated Retinoblastoma Gene Product: Implications for Clinical Gene Therapy Trials

My interest in this area of research actually started because of my clinical work. While I was doing research as a resident, as many of us do in training, I was never exactly sure what I was going to do with that research training or whether I wanted to pursue it. But in taking care of cancer patients and doing what seemed to be good surgery and then watching them receive good adjuvant therapies but still recurring and many developing metastatic disease and dying, encouraged me to look at other ways of treating our patients. And so in studying carcinoma of the breast which is one of my areas of interest in Surgical Oncology I began to look at some of the molecular mechanisms for breast cancer development and treatment, not only in familial cases but also sporadic breast cancer cases. And it's exciting that, as surgeons, we may actually be able to know what some of these lesions are at the genetic molecular levels and possibly repair these lesions, excise them, or replace the defective elements much as we do on a much larger scale now surgically.

[ Slide 01 ]   Introduction

I want to just talk briefly about one of the areas of gene therapy that we're working on in my laboratory. And we're studying both tumor suppressor genes and cell cycle regulatory genes, and using them in gene therapy protocols, hoping to apply these in the clinical setting for our patients who fail current standard treatment strategies. Two of the genes that we are particularly interested in are the retinoblastoma tumor suppressor gene and the E2F-1 transcription factor. These genes interact with each other. The retinoblastoma tumor suppressor gene is a well described tumor suppressor gene. It plays a role not only in regulation of the cell cycle, but also in apoptosis or programmed cell death. And the E2F-1 gene is a transcription factor which we've shown that using this transcription factor and overexpressing it in breast carcinoma cells can lead to apoptotic cell death.

[ Slide 02 ]   pRb and E2F-1

And the interaction of these two genes is shown here in that when the retinoblastoma protein pRb is bound to E2F-1, this results in cell cycle arrest in G1 phase. However, when the retinoblastoma gene is phosphorylated, this releases E2F-1, which then promotes S phase genes and promotion of the cell cycle.

[ Slide 03 ]   p53, pRb and E2F-1 in Cell Cycle and Aprptosis

Now there are many other cell cycle regulatory proteins that interact with these two characters, and here are just a few of those. This is just a very abbreviated cartoon modified from Dr. Kasten's review in Cell Death and Differentiation recently. But we think that when the retinoblastoma protein is phosphorylated and releases E2F-1 that this... the S phase genes are turned on and this results in inappropriate cell proliferation signals which can then result in apoptosis. Now many people believe that this results occurs in a p53 dependent manner, but in fact we've shown that this does not require p53. And this is quite important since p53 is one of the main genes that is mutated in many different types of cancers.

[ Slide 04 ]   Purpose of Study

Now in this particular study that I'm going to describe to you today what we wanted to do was understand what some of the interactions were between pRb and E2F-1 in the regulation of apoptosis in breast carcinoma cells. And we wanted to evaluate what effect the overexpression of pRb and a truncated form of pRb might have on breast cancer cell growth and apoptosis. Now in general, the retinoblastoma tumor suppressor gene is known to arrest cells in G1 phase and when it's rapidly phosphorylated it can easily be inactivated in the cell. One of my collaborations with some other investigators at MD Anderson created adenoviral vectors with these genes. We created a truncated and mutant form of this retinoblastoma gene and wanted to use this to see if we could create more or less a superactive form of this gene.

[ Slide 05 ]   Methods

Some of the studies that we've done are looking at cell growth and apoptotic assays and also looking at expression of many different proteins, both apoptotic proteins and also other cell cycle regulatory proteins. We've also begun to look at the level of transcription in many of these settings.

[ Slide 06 ]   Genetic Background of Human Breast Carcinoma Cell Lines

Now these are three of the cell lines that we used for this particular study that I'm going to describe for you, all different breast carcinoma cell lines. And I show you this information here because knowing the genetic background of these different cell lines becomes very important in their responses to these different genes that we introduce. And this is very helpful to us in order to determine how effective these gene therapy trials might be in a clinical setting.

[ Slide 07 ]   Recombinant Adenovirus

These genes were introduced using recombinant adenovirus vectors in most cases, and we used a type 5 adenovirus. We have multiple different vectors, but in this study we used luciferase as our control vector, a full-length retinoblastoma gene inserted into the adenovirus, and also our truncated and mutant superactive form of the retinoblastoma gene.

[ Slide 08 ]   56kDa Protein is Overexpressed in Cells Treated with Ad5Rb56m

This shows that we can overexpress the protein, the 56 kiloDalton protein product of this truncated and mutant form of Rb here, is highly overexpressed in all three of these cell lines and in fact easily expressed in most any cell line that we tested.

[ Slide 09 ]   Growth Curve

One of the first things we did was to look at cell growth and we expected it to inhibit cell growth and indeed in the 468 and SKBr3 cell lines the truncated mutant form of Rb causes significant reduction in cell growth, whereas in control and luciferase treated cells there was no change at all in cell growth. The MCF7 cell line however did not have any inhibition of cell growth using these proteins.

[ Slide 10 ]   Morphologic Change Indicates Apoptotic Cell Death

[ Slide 11 ]   % Apoptotic Cells (FACS)

[ Slide 12 ]   PARP Cleavage Indicates Apoptosis

We saw PARP cleavage as our previous speaker described and this was consistent and confirmed apoptotic cell death.

[ Slide 13 ]   Apoptotic Mediator Bcl-2 is Upregulated in Cells Treated with Ad5Rb56m

We began to look at many apoptotic mediators in this process and I'll just show you a brief overview of this, and that one very interesting thing we found is that the anti-apoptotic mediator BCL2 was markedly up-regulated in our cells treated with the truncated mutant form of Rb that were actually undergoing significant apoptotic cell death. And there is now evidence to support that this occurs for a number of reasons. One may be stress-related cell interactions but the other is that there does seem to be some cleavage products of this anti-apoptotic mediator that actually promote apoptosis within the cells. Now the MCF7 cell line has up-regulation of BCL2 even at baseline, which may be one reason why it does not react with the truncated mutant Rb.

[ Slide 14 ]   Caspase 3 is Cleaved in the Cells Treated with Ad5Rb56m

We did see caspase-3 cleavage in these two cell lines here and MCF7 does not have caspase-3 which in fact is most likely the reason that this cell does not undergo apoptotic cell death when we treat it with our truncated and mutant form of Rb.

[ Slide 15 ]   E2F-1 is Overexpressed in Breast Cancer Cell Lines Following Infection with Ad5Rb56m

Now as I showed you earlier, the full-length Rb actually inhibits the function of E2F-1 in the normal cell. And in fact we were surprised to find that when we truncated and mutated Rb, it becomes superactive, it causes apoptotic cell death, and in fact it seems to be up regulating E2F-1 within these cells. And this again may be one reason why we're seeing apoptotic cell death.

[ Slide 16 ]   In Situ TUNEL Staining

We've subsequently taken these in vitro studies into an animal model where we inject breast carcinoma cells into the mammary fat pad of nude mice and then once we have established tumors growing in this mammary fat pad, we then treat the tumors with our truncated and mutant form of Rb. And using TUNEL staining you can see here that we have significant apoptotic cell death within these treated tumors, and in fact see a significant percentage of regression of tumors in this established model.

[ Slide 17 ]   cDNA Expression Assay

We're now taking this a step further in the lab and looking at cDNA expression arrays with both our Rb and our E2F-1 models and looking at the other genes which are important in this entire process.

[ Slide 18 ]   Summary

So I'll just summarize here by saying that in this one little vignette from our lab we found that using knowledge that we have about tumor suppressor genes and molecular events that go on within the cell cycle we're able to overexpress these genes and treat the cells and actually result in apoptotic cell death both in vitro and in vivo. It's important to know what other genes are involved and we're exploring that because we feel that we can then take this to the clinical setting, analyze the patient's tumor for what different genetic defects may be there, and then actually specifically treat a patient based on the genetic profile of that tumor.

[ Slide 19 ]   Acknowledgements

And I'd like to just thank all of the people in my laboratory and the collaborators that I've worked with at MD Anderson and again I appreciate your attention and the opportunity to participate today.