FOOD AND DRUG ADMINISTRATION

CENTER FOR BIOLOGICS EVALUATION AND RESEARCH

“This transcripts has not been edited or corrected, but appears as received from the commercial transcribing service. Accordingly the Food and Drug Administration makes no representation to its accuracy………..”

BIOLOGICAL RESPONSE MODIFIERS ADVISORY COMMITTEE

MEETING #35 - TELECONFERENCE

OPEN SESSION

Monday, June 9, 2003

1:30 p.m.

National Institutes of Health

Building 29A, Room 1A09

Rockville, Maryland

ATTENDANCE

Gail Dapolito, Executive Secretary

Rosanna L. Harvey, Committee Management Specialist

Committee Members:

Mahendra Rao, M.D., Ph.D., Chair

Jonathan Allan, D.V.M.

Bruce R. Blazar, M.D.

David Harlan, M.D.

Katherine High, M.D.

Joanne Kurtzberg, M.D.

Richard C. Mulligan, Ph.D.

Anastasios A. Tsiatis, Ph.D.

FDA Participants:

Steven Bauer, Ph.D.

Kathryn M. Carbone, M.D.

Suzanne Epstein, Ph.D.

William Freas, Advisory Committee Office

Deborah Hursh, Ph.D.

Malcolm Moos, M.D., Ph.D.

Philip Noguchi, M.D.

Raj K. Puri, M.D., Ph.D.

A G E N D A

Welcome and Introductions

Mahendra Rao, M.D., Ph.D., Acting Chair 3

Conflict of Interest Meeting Statement

Gail Dapolito, Executive Secretary 7

FDA Introductory Remarks

Philip Noguchi, M.D. 8

Laboratory Presentations

Laboratory of Stem Cell Biology

Steven Bauer, Ph.D. 9

Laboratory of Immunology and Developmental Biology

Deborah Hursh, Ph.D. 18

Malcolm Moos, M.D., Ph.D. 27

Suzanne Epstein, Ph.D. 31

Open Public Hearing 37

Adjourn 40

P R O C E E D I N G S

DR. RAO: Hi to everyone. And everyone is here, except Dr. Harlan, right, who needs to be away for a couple of minutes?

MS. DAPOLITO: Except I'm not sure if Dr. Mulligan and Dr. Blazar have come on yet.

DR. RAO: I don't think we really need introductions, right? But perhaps we may want to just briefly go around and just make sure we know who's there.

MS. DAPOLITO: On our end?

DR. RAO: Yes.

MS. DAPOLITO: Absolutely. I'll start with me. Gail Dapolito, the executive secretary. We'll go around our table here.

DR. EPSTEIN: Okay. Suzanne Epstein, Division of Cellular and Gene Therapies.

DR. NOGUCHI: Phil Noguchi, Office of Cellular, Tissue, and Gene Therapies.

DR. PURI: Raj Puri, Division of Cellular and Gene Therapies.

DR. CARBONE: Kathy Carbone, Office of Director.

MR. FREAS: Bill Freas from the Advisory Committee Office.

DR. BAUER: Steve Bauer, Division of Cell and Gene Therapies.

DR. MOOS: Malcolm Moos, Cellular and Gene Therapies.

DR. HURSH: Deb Hursh, Cellular and Gene Therapies.

MS. HARVEY: Rosanna Harvey, BRM Advisory Committee.

MS. DAPOLITO: Can everyone hear us okay on this end? Okay. And we have one member of the public, sort of, a guest student from our Office of Special Health Issues here today.

MR. FREAS: Can we do a roll call of who's on the line?

MS. DAPOLITO: Yes, Dr. Rao will do that. Okay, Dr. Rao, do you want to call the roll anyway?

DR. RAO: I'll do that.

MS. DAPOLITO: Okay.

DR. RAO: Maybe the easiest way will be to just do it in turn. Does everybody have the BRMAC participant sheet in front of them? Well, then why don't I call out that list, and then we can just go down that as a roll call?

So I'm here. I'm Mahendra Rao, and I'll be acting as the chair for this meeting. Is Dr. Allan there?

DR. ALLAN: Yes, I am.

DR. RAO: Dr. Blazar?

MS. DAPOLITO: I think he's not on the line.

DR. RAO: Dr. High? I think I heard her sign on.

DR. HIGH: Yes. I'm here.

DR. RAO: Okay. Dr. Kurtzberg?

DR. KURTZBERG: I'm here.

DR. RAO: Alison Lawton?

MS. DAPOLITO: No, Alison won't be joining us.

DR. RAO: Dr. Mulligan?

MS. DAPOLITO: He's not on yet.

DR. RAO: Dr. Tsiatis?

DR. TSIATIS: Yes, I'm here.

DR. RAO: Dr. Harlan?

DR. HARLAN: I'm here.

DR. RAO: And is the consumer rep, Alice Wolfson, here?

MS. DAPOLITO: No, she won't be joining us.

DR. RAO: I guess we have a roll, Gail?

MS. DAPOLITO: Okay, thanks.

DR. RAO: You need to make a statement?

MS. DAPOLITO: Yes, I'll read the conflict of interest meeting statement. The following announcement addresses conflict of interest issues associated with this meeting of the Biological Response Modifiers Advisory Committee on June 9, 2003.

Based on the agenda made available, it has been determined that the committee discussions present no potential for a conflict of interest. In the event that the discussions involve specific products or firms not on the agenda for which members have a financial interest, the members are aware of the need to exclude themselves from the discussion. Their exclusion will be noted for the public record.

With respect to all other meeting participants, we ask in the interest of fairness that you address any current or previous financial involvement with any firm whose products you wish to comment upon.

I also have a couple of just administrative items. The teleconference is being recorded by a transcriber. So if the committee members would identify yourselves before you make a comment, that would be a great help.

It doesn't look like we have had any requests to address the committee during the open public hearing. So we won't use that portion of the meeting, unless someone walks in between now and then from outside. So we'll go directly to the closed session after we hear from the speakers--the FDA speakers.

We'll just take a minute, though, in between to clear the room on our end before the committee goes into the closed session. And then we'll turn it back over to Dr. High and Dr. Rao and Dr. Harlan.

Thanks, Dr. Rao. I'm done, Dr. Rao.

DR. RAO: Did Dr. Noguchi need to make any introductory statement or any remarks?

DR. NOGUCHI: Yes, I'll just take a minute here. You can tell Malcolm and I seem to have the same hoarseness. But this Office of Cellular, Tissue, and Gene Therapies has just been formed officially in October of 2002. And actually, the site visit that we're going to be reviewing was the first official site visit of the office of the division under that as part of the new office.

I just want to, first, thank all the members of the site committee team, who actually put together the report, spent a lot of time and effort with this, and really would be pleased to now present the--in brief the scientists who were reviewed at that time.

So I really don't want to take very much more time than that because there's a lot to cover. And it's not really our show, it's the scientists' show that we're reviewing. Thank you very much.

DR. RAO: Do we want Dr. Bauer to go first?

DR. NOGUCHI: Yes, that would be fine. He's--

DR. BAUER: I'm here, and I'm ready. Shall I go ahead? Okay.

I have been acting lab chief of the newly--well, a new position for me, the Laboratory of Stem Cell Biology, since October 2002. And prior to that time, I was a member of the Laboratory of Immunology and Developmental Biology. So most of the time covered by the site visit, I was a member of that laboratory.

In addition to the research program report, the site visit package that I think you folks have a copy of contains a summary of my lab resources, including staff, budget, other funding, and other CBER research support infrastructure. There is also a little bit of information on my role in regulatory activities, including review of INDs in the area of gene therapy, cellular therapy, xenotransplantation products, as well as development and presentation of policy in these areas, and public outreach at scientific and regulatory meetings.

And finally, there's a list showing--or a diagram showing where the lab is in the Division of Cell and Gene Therapies and a list of other personnel in that lab.

My research program examines cell-cell interactions that govern development of lymphoid lineage cells and play a role in transformation of these cells. And I'm going to briefly describe two projects that investigate these two types of cell-cell interactions.

The first project is on the role of a molecule called dlk in stroma pre-B interactions. One of the challenges facing the field of stem cell biology is to determine what molecules and signals are exchanged between self-renewing stem cells and other cells in their microenvironment in which the stem cells proliferate and differentiate.

And to study this process, we use a model system consisting of normal, self-renewing, precursor B cells that grow on a stromal layer of mesenchymal origin. And in the presence of the growth factor, IL-7, these precursor B cells self-renew and remain undifferentiated.

However, you can get them to differentiate by removing IL-7, at which point the precursor B cells turn into mature B cells and then undergo apoptosis. The precursor B cells can be administered to immunodeficient mice and can reconstitute part of the B-cell lineage. So this model mimics the clinical use of a variety of stem cells in which there is in vitro propagation before administration.

Now our previous work has shown that this molecule I mentioned earlier, dlk, is expressed on stromal cells and that it influences the response of pre-B cells to IL-7. And the key observation was that downregulation of dlk on stroma allows precursor B cells to self-renew without IL-7 and to forgo differentiation and apoptosis normally caused by removal of IL-7. And that observation has led to a lot of subsequent work in my laboratory, both in in vitro systems and in a knockout animal.

What we have seen is that adaptation of pre-B cells on these stroma with low amounts of dlk on their cell surface involves loss of the requirement for IL-7. And to summarize a lot of work, the major finding we've seen to explain that is that there is a change in Notch-3 receptor expression on the pre-B cells under these conditions. So, basically, it increases.

We've also further examined the role of dlk using a knockout mouse model. So dlk has been knocked out, and we're characterizing lymphoid cell populations in those cells and studying functional perturbations, including differences in antibody responses. And we're also deriving knockout--or a variety of cell lines from the knockout animals to further characterize the role of dlk in those cell lines.

Overall, the knockout mice have some developmental defects, including growth retardation, fatty change in the liver or liver steatosis, obesity, and they do have some perturbations in B-cell development, which mimics our results from the in vitro culture system.

The B-cell changes that we've seen include the following. There are increased B cells in the bone marrow. There are immature B cells in the spleen. There are increases in marginal zone B cells. And functionally, there are changes in serum immunoglobulin G levels, and there is a decreased secondary antibody response to T-dependent antigens. So both development and function of B cells have been changed in this knockout animal.

So in the future we continue to look at the role of dlk and how it might affect the Notch pathway in a variety of cells through the use of microarray analysis, RT-PCR, et cetera. And we intend to develop more stem cells from these knockout mice to look at potential roles for dlk and its modulation of Notch signaling in a variety of different important tissue types that could be of clinical significance, including bone marrow derived, multi-potent adult progenitor cells, hematopoietic stem cells, neuronal stem cells, pancreatic stem cells, and perhaps others as well.

And of course, using the knockout animals, we will continue to characterize the lymphoid lineage cell function and subpopulations in the knockout animals.

And finally, we can use these mice to cross to other interesting knockout strains, such as IL-7 knockout, or Notch knockout, or knockouts of other members of the Notch/Delta family that might represent a sensitized genetic background that can further elucidate the role of dlk in those systems.

Now I'll go on to a second project, which was to look at the effects of TGF-beta on precursor B cells and its role in pre-B cell leukemagenesis.

DR. RAO: Steve, can I interrupt you for a second here?

DR. BAUER: Absolutely.

DR. RAO: I just wanted to poll the committee and ask them if they'd like to ask questions on this project before Steve moves on to a second because, given we're doing this by phone, it may be hard to keep track of all of these things.

Does anybody on the committee have any questions? Maybe they can identify themselves first and then--

[No response.]

DR. RAO: No one? I have one quick question, Steve, and that's have you considered deriving ES cell lines from this in occult mice, studying one cell line? I mean, studying somatic stem cells, you can also study ES cells?

DR. BAUER: Well, we do--we have the knockout line that was used to, you know, make the knockout animals, and we have considered doing in vitro differentiation studies using those lines.

We also would potentially be able to do that kind of study in human embryonic stem cells or stem cells of a variety of different tissues from human cell lines that are commercially available or available through other sources.

DR. RAO: And do you have access to presenilin knockout mice?

DR. BAUER: I could probably get--through some of the people here on the NIH campus, I could consider getting those animals as well.

DR. RAO: Thank you, Steve.

DR. BAUER: Okay. All right. So I'll briefly go into the second project. I'll start by saying just a little background. TGF-beta has been shown to influence B-cell development, and loss of TGF-beta sensitivity has been associated with B-cell tumorigenesis in several different B-lineage tumors.

So we've been looking at the effects of TGF-beta on these normal mouse precursor B cells, which we are able to culture, and also its potential role in mouse pre-B cell leukemagenesis.

There's previously been little data available that shows exactly what stages of normal early B-cell development in mouse could be susceptible to TGF-beta and very little data about the role of TGF-beta in tumorigenesis in early B cells in mouse.

So just to summarize our results, we've shown that these normal precursor B cells that we grow in culture express the TGF-beta receptor and are sensitive to inhibition of IL-7 induced proliferation. TGF-beta can inhibit pre-B cell proliferation by inducing G1 cell cycle arrest. And normal pre-B cell lines, as well as their malignant progeny--those are cells that we've transformed in vitro by putting in dual oncogene-containing viruses--they express the TGF-beta receptor as well. However, they've lost sensitivity to TGF-beta.

Now the retroviruses we use all have the myc oncogene in them, and we have derived pre-B cell lines from myc oncogene expressing transgenic animals. And they are not sensitive--or they remain sensitive to TGF-beta, even when there are tumors.

So all of this argues that there are multiple pathways for transformation of pre-B cells, and our ability to use the system and to introduce oncogenes, either singly or in combination, could provide a very nice methodological approach to the dissection of molecular pathway of TGF-beta signal transduction and how these pathways are abrogated during transformation.

And of course, the Division of Cell and Gene Therapies is responsible for oversight of safe and effective gene therapies, including approaches that involve vectors that insert into cellular DNA. And the prospect of cellular transformation subsequent to insertional mutagenesis is a safety concern for gene therapy. So understanding of molecular mechanisms of transformation is an important pursuit in the FDA.

And I think with that, I'll stop my presentation and welcome any more questions.

DR. RAO: Thank you, Steve. Are there any questions?

DR. HIGH: Steve, this is Kathy High. I just wanted to ask one question about is that--is the description of the dlk knockout mice published yet?

DR. BAUER: No, it's not. We're working on that.

DR. HIGH: Okay. Okay, just want to make sure I hadn't missed it.

DR. BAUER: Yes, but there was another group who had made one. And we see very similar things in terms of some of the effects on fat metabolism, obesity, and so on. But we're the only ones that I know of looking at the effects on the development of the immune system.

DR. HIGH: Okay.

DR. RAO: If there are no more questions, then we can move on.

DR. HURSH: Okay. So this is Deborah Hursh. I'm from the Laboratory of Immunology and Developmental Biology. And like Dr. Bauer said, my regulatory contributions are in the book that--the site visit book that you all should have. My regulatory contributions are in the areas of cell therapy, gene therapy, and assisted reproduction.

And now, what I'd like to do is just give a brief overview of my research program. My research program studies the role of intercellular communication pathways in pattern formation and morphogenesis. The model system is the fruitfly Drosophila melanogaster. The major focus of the laboratory is on the bone morphogenesis protein, or BMP, signaling pathway, particularly the Drosophila homolog of BMP 2 and 4, which is the product of the decapentaplegic, or dpp, gene.

BMP 2 and dpp are essentially identical. And dpp and BMP 4 are 90 percent homologous. The BMP pathway, like other extracellular signaling pathways, is highly conserved among all metazoans. So data derived from Drosophila are applicable to mammals.

This research program is significant to CBER's mission for several reasons. Currently, both TGF-betas and BMPs are being studied directly for clinical use and are part of the manufacture of cellular products.

In addition, the entire area of signal transduction is of critical importance, as the development of cell therapies and engineered tissue products will require in-depth knowledge and manipulation of extracellular signaling pathways. The behavior of stem cells in particular is governed by signal transduction, including the BMP, Hedgehog, Wnt, and Notch pathways.

My research program is taking an integrated genetic and molecular biological approach to study the action of the Drosophila BMP, dpp, in a specific event in development--the formation of the adult head. Our goal is to understand the biochemical pathways that a TGF-beta protein utilizes in the construction of a complex integrated structure like the fruitfly head.

This approach is based on my isolation of a specific class of dpp mutations whose effect is limited to the adult head. These recessive mutations reside in the 5' cis regulatory region of the gene and cause a syndrome of alterations of the adult head that include eye reduction, sensory structure eliminations, and duplications.

The research program is divided into three areas. The first is a genetic and molecular analysis of head capsule mutations, where our aim is to identify cis-acting sequences and trans-acting factors required for dpp's role in head capsule formation. The second area is to use the dpp head capsule mutations to carry out a genome-wide genetic screen to identify other participants in BMP-mediated adult had formation, and the third area is to understand the role of dpp in the formation of the adult head.

I will briefly discuss the first two projects. Project 1. I isolated mutations in dpp that specifically altered the adult head of the fly. These mutations reside in 5' cis regulatory DNA and do not alter the protein coding region of the dpp protein. Our starting hypothesis was that these mutations caused their mutant phenotype by disrupting dpp expression in a critical region of the head primordia. In flies, this structure is the eye-antennal imaginal disk.

We used these mutations to map the exact location of head determining cis-regulatory material within the dpp gene's 75 kb of cis regulatory DNA. Using these data, we have created a series of beta-galactosidase reporter constructs from DNA in the region identified by our mutations and introduced them into flies transgenically.

These constructs all express in the primordia of the adult head, the eye-antennal disk, thus we have identified a 10-kb region that contributes to dpp expression in the head. The most critical of our mutations was a 17 bp deficiency. Within this deficiency are sites predicted to bind several homeodomain transcription factors of the HOX class and also the TALE class.

The HOX proteins deformed and labial and the TALE class proteins extradenticle and homothorax are expressed in the eye-antennal disk. These two groups of homeodomain proteins are also thought to work coordinately together, further supporting our observation that dpp's head expression may, indeed, be regulated by homeodomain transcription factors.

We tested the sufficiency of the 17 bp mutation to alter dpp expression in the eye-antennal disk by creating a beta-galactosidase reporter construct from mutant DNA. These constructs fail to express in the eye-antennal disk, while an equivalent construct from wild-type flies expresses as expected in the eye-antennal disk. These data support the idea that the head capsule phenotype is caused by loss of dpp expression in the head primordia and that homeodomain transcription factors are implicated in the regulation of dpp in this developmental event.

We are currently in the process of identifying the exact homeodomain transcription factors that interact with this DNA and assessing their contributions to dpp expression in the head.

And if there are any questions on Project 1, I'd be happy to take them now.

[No response.]

DR. HURSH: Okay. Project 2. My initial interest in the head capsule phenotype stemmed from the observation that the number of flies within a class that actually displayed the mutant phenotype, which geneticists call penetrance, correlated with the level of dpp protein function. When head capsule mutations are put in transheterozygous combinations with a class of dpp mutations that reduce but do not eliminate the activity of the dpp protein by altering the protein coding region, the penetrance of the head capsule phenotype increases with the severity of the activity-reducing allele.

These data indicated that the head capsule phenotype was sensitive to BMP pathway signaling activity. I reasoned that the head capsule defect would serve as a sensitized background to recover mutations in genes that acted with dpp in head development.

The sensitized dpp background, in concert with a second unlinked mutation, would reduce the level of the TGF-beta/BMP signaling to a level where a mutant phenotype would be revealed. Such mutations could then be recovered and identified. This type of dominant enhancer screen has been used with great facility to identify other members of genetic pathways.

We have carried out several genetic screens using this idea to search for genes that act with dpp in this pathway to form the adult fruitfly head. This approach has identified approximately 20 second site mutations that interact with dpp to create head capsule defects. These genes encode transcription factors from various classes, participants in signal transduction pathways, and a variety of other known proteins, including cyclin A and type 4 collagen. Several proteins of unknown function have also been identified.

We have focused to date on one very strong interaction with the product of the Drosophila odd-paired gene. Odd-paired encodes a zinc finger transcription factor of the conserved "zinc finger of the cerebellum," or ZIC family. This family is involved in neural, specifically brain development in frogs, zebra, fish, and mice.

We have analyzed the post embryonic requirements for odd-paired and find it is expressed in the eye-antennal disk and is required for proper head development. We are focusing on understanding how dpp and odd-paired interact in head development. And the current--the future focus of my lab will be identifying and understanding the interactions of dpp with these genes we have identified.

And I'll be happy to take questions now. Thank you.

DR. RAO: Deborah, this is Mahendra Rao.

DR. HURSH: Yes?

DR. RAO: You had good success with doing these sort of dominant enhancer or suppresser screens, right?

DR. HURSH: Yes.

DR. RAO: You do it the reverse way, too? Then looking for dominant, you look for things that make the phenotype work. Did you also look for things that skip the phenotype?

DR. HURSH: No. That would be a dominant suppresser screen, and we haven't quite figured out a way to do that yet. That has been done with great facility in flies, but my phenotype is very subtle, and it's easier to make it worse than it is to make it better. To do a good dominant suppresser screen, you need to have a very, very strong phenotype that you can make better and that it's easy to see that it's been made better.

And this has been done with, for example, mutations that severely alter the eye, and then they get--they look better. I think it would be hard to do this kind of thing with the type of phenotype I'm working with. Though if I could do it, I would because it's a very powerful screen.

DR. RAO: My second question was, you know, you mentioned right in the beginning that you were going to do a genome-wide sort of scan to look at candidate binding sites in the upstream region?

DR. HURSH: Well, not a genome-wide scan. We actually were very lucky in that one of our mutations is a 17 bp deficiency.

DR. RAO: Right.

DR. HURSH: And within that 17 bps, there are HOX sites and two different TALE protein sites. So that really brought our attention to those particular proteins as perhaps interacting with this DNA. And so, within a 600 base region around that 17 bp deficiency, there are also multiple other sites that would also bind to HOX and TALE proteins. That's really focused our attention on this.

But we're not going any further to look for that, just within this very small region.

DR. RAO: Okay. Anyone else?

[No response.]

DR. RAO: Thank you, Deborah.

DR. MOOS: Are you ready? This is Malcolm Moos. With respect to regulatory endeavors, what I do is very similar to what Dr. Hursh and Dr. Bauer do. Also I think it's fair to mention, I've been involved with certain miscellaneous activities that Dr. Noguchi may wish to share with you in closed session and some projects from the center director's office involved with comparability, impurities, and specifications, the ICH document.

Our primary goal in our research endeavors has been to study critical processes controlling cell fate that will be of likely importance in the design of future improved cell and tissue-based products. Primary among these are the bone morphogenetic proteins, or BMPs, and the Wnts, and the factors that elaborate with them in these processes, including many potential unknown factors.

To do this, we've used xenopus laevis and, more recently, xenopus tropicalis, which are vertebrate model systems with several advantages. A few of these are the capability to do both gain and loss of function experiments relatively quickly and easily; the multiple readouts, which include morphology and histology; analysis of gene and protein level; analysis of gene and protein distribution, including the ability to look at the entire organism globally, both for gene expression level and distribution and protein expression and distribution.

In addition, xenopus is uniquely suited for analysis of the proteon in that it's possible to generate very large amounts of material for biochemical studies.

This approach has allowed the identification of a number of novel molecular entities, which have included the cartilage-derived morphogenetic proteins, the CDMPs 1, 2, and 3, which are TGF-beta family members involved in skeletal patterning; anti-dorsalizing morphogenetic protein, which is important in specification of the body axis; and FRZB, which is the first of the class of secreted inhibitors of the Wnt growth factor oncogenes.

In the interest of time, in case there are some questions, I'm going to discuss further studies with just one of these, CDMP 1.

Our original functional studies revealed an apparent very stringent requirement for proteolytic processing of this molecule, and it was immediately apparent that a potential explanation for this was that it shared a consensus proteolysis site with one other TGF-beta family member, known as Vg1. This protein is also known to have a stringent requirement for proteolysis.

We, at the site visit, presented data suggesting that a point mutagenesis relieves this stringency by allowing bioactivity to be released and that this effect could be mimicked by co-administering a combination of two proteolytic enzymes belonging to the subtilisin-like proprotein convertase family, or SPC family. And indeed, with the combination that we tried, we could recover an activity that was comparable to the point mutant.

Since the site visit, we have confirmed this data and tested all other known SPCs and found that the combination of furin and SPC-6 was, indeed, optimal and essentially matched what could be obtained with the point mutant.

Of particular interest to us and also to members of the review committee were analogous experiments on the Vg1 protein. We have done the biological experiments and have confirmed that a point mutagenesis of Vg1 indeed releases its activity, and it appears on the basis--and this is just one preliminary experiment--that the synergistic effect of furin and SPC-6 may also obtain.

But that needs to be repeated, and we probably need to tweak our bioassay a little bit. It needs to be a slightly different bioassay than is needed for CDMP 1.

We've also used an oversight model to confirm the cleavage of these proteins biochemically. And again, preliminary data suggests that both the point mutant and the co-administration of SPCs with the wild type causes detectable cleavage of these molecules that you can't see with the wild type.

What's left, to complete the picture, is to perform the immunoblot analyses on the Vg1 protein and also to localize the SPCs by hybridization in situ, both in the limb, to analyze the situation with CDMP 1, where we feel the patterning is--where we feel its influence on patterning is critical, and in the gastrula with Vg1 because this would resolve an important outstanding question in the specification of the vertebrate body axis.

We think we could have this done so that a manuscript can be submitted by July. I'm happy to field questions on this project or anything else that's in the book that I didn't talk about today.

[No response.]

DR. MOOS: Do we still have you?

MS. DAPOLITO: Yes.

DR. NOGUCHI: Okay. Shall we move on to Dr. Epstein?

DR. RAO: I think so.

DR. EPSTEIN: Okay. This is Suzanne Epstein, and an overview of the Laboratory of Immunology and Developmental Biology was in the book. So I won't review that now. In terms of regulatory activities, that's summarized also, and I mainly review gene therapy INDs, master files for vector production, and then serve in a policy working group.

Since the site visit, I was asked to and have agreed to serve as informal unofficial science advocate for OCTGT. To review my research program, first of all, its relevance to these regulatory responsibilities.

Viruses and plasmids have potential uses in human health care as vaccines and also as gene therapy vectors. Development of safe and effective products in both these categories requires an understanding of the immunity they induce and its consequences. Can you hear me?

Okay. They're moving the microphone. The vaccine must contain the right antigens and induce the right immune effectors with adequate potency if it is to prevent disease. And in gene therapy, immune responses to viral vector components or to transgene products are usually undesirable. They can be a limiting factor, blocking efficacy upon repeat administration and, in some cases, causing pathology due to cross-reactions on self components.

My lab studies immune responses in the influenza system. Influenza is still a major public health problem, and the current approach to vaccination is not optimal. Also the risk of a pandemic due to spread of a new subtype into humans makes it important that we develop new means of immunization.

In addition, our findings are relevant to the design of vaccines for other viruses with unpredictable strain variation--that would include HIV, for example--or unpredictable outbreaks, which would include families of potential bioterrorism agents, and also to the consequences of gene therapy.

Thus, the results can inform regulatory decisions about safety and efficacy of viral vaccines and gene therapy vectors, and the choice of endpoints to monitor in preclinical and clinical studies.

Now our studies involve inducing immunity with live virus and also plasmid DNA constructs expressing individual influenza virus proteins in the mouse model, which has a number of advantages. The report you received described four project areas. In one, we studied mechanisms of heterosubtypic immunity, which is broad cross-immunity to influenza that protects against infection with strains of viral subtypes differing from the immunizing subtype.

We have studied the roles of T cells and the antibodies in cross-protection, whether induced by virus exposure or DNA vaccination. We showed that T cells provide some cross-protection against challenge in knockout mice lacking all antibodies and B cells.

We also studied knockout mice lacking IgA, selectively, and showed that they have heterosubtypic immunity. However, we saw some impairment in their ability to clear primary infection and in their responses to challenge under some conditions. The role of IgA at mucosal surfaces will be further investigated in this model.

In the next project, we showed that DNA vaccination with conserved components nucleoprotein and matrix of an H1N1 influenza A virus could control infection with H5N1 chicken flu viruses from the lethal outbreak in 1997 in Hong Kong.

The vaccination protected against lethal challenge with a moderately virulent strain and protected partially against challenge even with the most extremely virulent strain. In a situation with a strain-matched vaccine not available, DNA vaccination with conserved components like these might provide partial defense against a pandemic virus.

These findings are also relevant to other families of viruses in which unexpected emerging strains must be controlled, even if imperfectly, before matched vaccines become available. So that would include by analogy agents like hanta viruses and arena viruses.

Another project addressed the question of whether humans have heterosubtypic immunity to influenza as seen in the animal models. I examined archival records from the Cleveland family study before and during the 1957 pandemic. Many children who had culture-proven influenza in earlier study years became infected again in 1957, but almost no adults did.

While some further analysis is still pending, the greater resistance of adults after prior infection suggests accumulated immunity and suggests that human heterosubtypic immunity may have been sufficient to alter susceptibility to the 1957 Asian influenza virus.

The last project described involves approaches to improving vaccination with conserved components of influenza. We are studying M2, or matrix 2, which is already known to provide broad cross-protection, and PA, or acidic polymerase, which we identified by screening of genes by DNA vaccination. That was a screen of all the genes of influenza A. PA DNA, followed by PA vaccinia boosting, protected against lethal challenge in BALB/c strain of mice.

It induced a CD8 T cell response that was readily detectable in B6 but not BALB/c mice by intracytoplasmic staining for interferon production. But in contrast, it protected against lethal challenge in BALB/c mice and not B6 mice. So the readily detectable staining response does not correlate with protection.

We have now made new constructs with murine optimized codon usage to enhance the potency of this vaccination, and they will be used to further investigate these findings.

Since the time of the site visit, we have succeeded in mapping a T cell epitopes in BALB/c mice of the PA antigen using ELISPOT analysis instead of intracytoplasmic staining for interferon, and we are now investigating the functional role of these peptides.

In order to alter immunodominance patterns and enhance the contribution of various antigens to protective immunity, we plan additional studies of DNA prime-viral boost regimens using both adenovirus and vaccinia recombinants. We're now making some adenoconstructs.

Simultaneous versus sequential immunization with different influenza antigens will be tested. And then the most effective prime and boost regimens, as demonstrated with our usual lab challenge strains, will be subjected to challenge with model pandemic systems H5 and H9 in collaboration with Terrence Tumpey of USDA.

Overall, we will be continuing to explore protective immunity against influenza infection and the contribution of conserved components to it. I'm happy to answer any questions, either about the research program or the Lab of Immunology and Developmental Biology.

DR. RAO: I have a quick question on your project. You know, you mentioned the difference in the immune response or susceptibility of children versus adults?

DR. EPSTEIN: Yes.

DR. RAO: Any speculation other than accumulating immunity as to why this could be true?

DR. EPSTEIN: Well, one, of course, might speculate at first that children behave differently, for example, putting things in their mouths and so on. But if you look at the data for 1950, '51, and '53, the susceptibility of children and adults was fairly similar. That was within the era of the H1N1 subtype.

When H2N2 arrived for the first time, so that all people were naive serologically--they didn't have neutralizing antibodies to that hemagglutin or neuraminidase--suddenly, the spike of difference between children and adults occurs. So it's something unique to that situation. It's not a general behavioral matter.

Also in that family study, all the adults were living in households with young children. So it's not exposure--an exposure issue that the children are exposed at school and adults are not. Frankly, I think immunity makes the most sense.

DR. RAO: It could be a really nice handle into the systems. It might be nice.

DR. EPSTEIN: Oh, absolutely. And I don't know how to duplicate that body of data except to, if there were a surveillance study going on in the future at the time of another pandemic.

What's retrievable from those archives is limited, and I welcome your ideas, if you have any. But it will not be possible to prove that that's what occurred, and we, of course, can't go back and do serological or T cell assays now. It was not considered at the time of the study.

DR. HIGH: Suzanne, this is Kathy. What's available from that study is only data, no samples?

DR. EPSTEIN: That's correct. There was a serum archive sent to Yale by Case Western. But I corresponded with Yale University, and at the time a certain person retired, it was all thrown away. It had mostly been used up by his studies. What was left was dregs. But when he retired, he cleaned out the freezer.

It wouldn't have been probably enough to use. But that's not as tragic as it might seem. What I'm interested in would not be present in serum. The serum would have shown neutralizing antibody titers, and the publications from the '50s have an extensive analysis of the antibodies by HAI and complement fixation.

That's not what mediates broad cross-protection, as we've shown in our other studies, or probably not. If it is antibody mediated at all, it's probably mucosal IgA, but probably not serum antibody. So the T cell responses could not have been tested from that anyway.

DR. HIGH: Right, because they were just serum samples.

DR. EPSTEIN: Right. What's left is case report forms and charts of various kinds. I have to return to Cleveland this summer to track down--to try to track down which individuals may have received an experimental vaccine late in the observation period and that was not very effective, but to try and see if that changes anything.

DR. RAO: Any further questions?

[No response.]

DR. RAO: Thank you, Susan.

MS. DAPOLITO: Dr. Rao, this is Gail. Could you give us a couple minutes to get things sorted out on our end? Clear the room, okay?

[Whereupon, at 2:15 p.m., the open session was adjourned, to reconvene in closed session, this same day.]

- - -