David Baltimore (Cal Tech) Part 2: Why Gene Therapy Might be a Reasonable Tool for Attacking HIV

Uploaded by ibioseminars on 01.04.2010

Hello, I'm David Baltimore,
Professor of Biology at the California Institute of Technology,
and I'm going to talk more about HIV.
In particular, I want to develop the idea that gene therapy may be a reasonable tool
for attacking HIV.
And I'm going to do that in the context of why standard approaches to HIV
have not worked, and why something as heroic as gene therapy
may, in fact, be the only way to go.
So, let me talk about the more general issue of why you would take
a sort of molecular biologist's approach or a molecular engineer's approach
to an infectious disease.
For most infectious diseases, you don't need anything like that.
Most infectious diseases are handled by standard methodologies:
drugs, antibiotics, vaccines to prevent them, but certain infectious agents --
HIV, malaria, tuberculosis, and others -- are unsolved medical problems...
are great needs today, because they have proven extremely difficult.
They will not respond to the standard methodologies.
Take vaccines, which are the right way to deal with infectious agents.
That is, prevention is better than treatment.
How do vaccines work?
They work by inducing antibodies or what are called protective T cells,
and those cells then, and the antibodies, attack a virus
if it comes into your body.
So, you're pre-prepared for a virus.
And what they're doing is taking advantage of the immune system
which would respond to that virus anyway, but would respond more slowly,
doesn't remember having seen it before, and so has to learn all about it
de novo. And so what the vaccine, really, is doing
is short-circuiting the recognition of the virus, getting rid of the learning stages
so that the body can go right in and handle it.
But, you have to have good antibodies or good T cells to do this.
And, I'm going to ignore T cells for the moment, because antibodies
seem to be the major thing that vaccines elicit.
So, if the organism itself (and that's true of tuberculosis and others)
cannot elicit good, broadly neutralizing, reproducible antibodies,
then vaccines aren't going to work.
And that's the situation.
So, how do these kinds of agents elude antibodies?
Well, what HIV does, and we'll go into this more in a little bit,
is it hides its jewels in coatings and in structures,
so that antibodies just can't find the relevant parts of the virus.
Malaria and other agents like it vary their structure continually,
so even though you get a good antibody, by the time the antibody is being made,
the organism has changed its stripes and now is no longer sensitive to that antibody.
Tuberculosis, on the other hand, sort of becomes latent.
It goes and hides itself away and takes advantage of immune deficiencies
to come back out and cause transmissible disease.
And there are other tricks that viruses and other organisms use
like direct immunoinhibition -- inhibiting the immune system
using viral protein.
Now, viruses like polio and measles and mumps lulled us into complacence
because we could easily make vaccines.
And I remember the day that the then-Secretary of Health Education Welfare
came before the nation and said we have discovered HIV,
and we should have a vaccine in no time.
Well, she was right in that we had made vaccines against lots of other agents.
She was wrong -- HIV is different than polio or measles or mumps.
And you can see the consequence of that difference here,
where we look at the spread of HIV through the world.
So, HIV almost certainly first appeared in Africa,
and these red areas are the areas where it's highly endemic,
where many people are infected with it.
That then spread over to the United States, spread to Europe,
spread ultimately to South America, and is now spreading
more and more through Asia... tremendous problems in Russia, in India.
As I said before, the numbers in China are still low, but we worry about them getting larger.
And in fact, basically, the whole world suffers from this problem.
What are we going to do about it?
How are we going to put the genie back in the box?
Well, drugs -- small molecules that target key parts of the virus.
They target the reverse transcriptase, they target the integrase,
they target the protease... all of those things that I showed you are inside the virus
are targets for these drugs.
Drugs have been very successful in the developed world.
And many people of today are living with HIV infection,
who would have died 5 years ago, 10 years ago, 20 years ago,
if they had been infected with the virus.
And those drugs aren't terrific.
They are hard to take, they have side effects. They're not perfect.
But, they've made a huge difference in the life of infected people.
But they're expensive, and so, only recently has the world come together and said,
We're going to find ways to make these drugs available in the less developed world.
So now the drugs are available in Africa and elsewhere,
but still they're relatively expensive even for those circumstances,
and they have their side effects and are not perfect.
What you really want to do is to prevent the spread of the virus,
not treat the infected people.
So, one way to do that is education.
We know how the virus is transmitted.
It's transmitted sexually, mainly. It's transmitted through contaminated blood.
So, we can give people advice. We can educate people
about how to prevent the spread of the virus.
And that has been effective.
Rates of transmission went way down in Uganda and Thailand
after very explicit, very widespread educational programs.
But, if we just look at the United States,
we have young people coming along into a gay lifestyle
in particular in the United States,
who are continuing to get infected.
And so, vigilance is something that has to be maintained continually
if education is going to be the mode of preventing spread.
So, what other response is possible?
Well, the right response is a vaccine.
If we could get people to be in an immune state to the virus,
then it wouldn't matter if they were exposed or not.
But, HIV is not controlled by our immune systems when we get an infection,
and so making a vaccine that's able to do what the ordinary virus can't do
is a really daunting problem.
And people have worked on it now for quite a while.
They've tried to make proteins that will elicit antibodies that would bind to the virus.
They've tried to stimulate T cells.
We are seeing progress. Particularly on the T cell front, we're seeing significant progress.
But, it's slow, and it is unclear that any direction that we're taking today
is going to get us to a vaccine.
Now, let me spend just a minute on why HIV is so resistant to antibody.
Because all other viruses... basically almost all other viruses
are sensitive.
So, antibodies have to attack a virus on the outside.
They have to attack these spikes on the outside of the virus
and cover them up, or trigger them, or get rid of them...
Do something that will make it impossible for the virus to bind to an infected cell.
And the spike is made of 2 components.
It's made of a head, which is called gp120. You can see that down here very well.
And it has a spike that goes into the membrane
which you can see over here very well.
And, that spike is known as gp41.
So, we have gp120 on the surface, gp41 on the spike,
and there are antibodies... known human antibodies...
monoclonal antibodies that will bind both to the head and to the spike
and prevent infection.
The problem is that those antibodies do not have a high enough affinity
to provide protection to people.
And furthermore, you can't elicit those antibodies from the human immune system
with any degree of reproducibility.
They appeared once. We have captured them.
But, we can't do it over and over again.
So, the monoclonal antibodies are not the answer.
But, let's look at how HIV actually goes into a cell.
So, here is an HIV particle, and here's one spike blown up to be very large.
It actually has three balls on it, two in the front and one behind that can't be seen.
And, so it's a trimer. It has this spike that goes into the membrane,
which is gp41, which is also a trimer.
In fact, there's one monomer each of gp41 and gp120, in complex,
that were separated by a protease.
This is the infected cell down here.
That cell has a molecule called CD4 on its surface.
Only cells that have CD4 on their surface are infected by HIV,
because the first thing that HIV does is interact with CD4
by a place on the gp120 molecule.
That interaction causes a wholesale change in the structure of this trimer.
And what it does is to generate down here a binding site for another protein,
which is called the co-receptor, CCR-5.
The virus, after CD4 is bound, now develops a CCR-5 binding site,
binds there... the fusion actually occurs there between the membrane up here on the virus
and the membrane on the cell.
You can see the extent to which the CD4 molecule changes the structure of the protein.
This is at rest. This is HIV at rest, outside of cells,
and this is the structure of one of those 3 monomers.
And you can just see there are ribbons in it. All you need to do is look at it impressionistically.
Now, CD4 binds to it.
CD4 binds over here. You can see CD4 binding,
and now look at the change in the structure of this protein.
This helix is over here. It used to be over there.
There's a wholesale reorganization.
The consequence of that can be seen best in this model.
So, here is the at-rest state.
This is the binding site for CCR-5. It's in two pieces.
It isn't actually put together. Only when CD4 binds does it come together.
This is the binding site for CD4. It's in pieces.
Only when CD4 is in the environment does it induce a change in structure
that puts together the elements of the CD4 binding site.
And then, this is perhaps seen on the previous slide,
there's also all of this stuff around here.
And what that stuff is... you can see, this is looking down on the virus,
you see it all around the outside.
That stuff is sugar. Carbohydrate.
And that prevents antibodies from binding.
So, the antibodies can't bind all around. The only place they could possibly bind
would be the CD4 binding site or the CCR-5 binding site,
and neither one of those exist in the resting state.
They're both split into pieces.
And so, making an antibody that will neutralize this virus is extremely difficult.
The only hope I think that we have is not trying to make
the human immune system do something it can't do.
But, rather, to use the intelligence of modern molecular biologists
and structural biologists, who can look at these structures and say,
Well, I see a different place. I see a different place where we might attack it.
Or, maybe we can attack it, not with a standard antibody,
but with some different kind of protein, or some modified antibody,
that can get into a crevice which a standard antibody can't get into,
because antibodies are actually quite big on the scale of these molecules.
So, what has been the response by the scientific community?
Well, either people have just bulled ahead and said,
I don't care about the arguments. We're going to see if we can induce antibodies.
And, that's pretty well been given up by now,
because it has failed in so many people's hands,
who've tried so many different ways.
And so people have moved to T cells,
which, as I said, are rarely involved in vaccines,
but maybe occasionally.
And, you can expect that there will be partial control of certain viral infections
by using T cells. And, the way you stimulate T cells
is not with proteins, it's with little peptides.
Some people have tried to use peptides. They've used DNA
to try to encode viral proteins or peptides,
and tried to make novel kinds of vaccines.
Well, this is all new technology.
Nobody's ever done it before.
And although there are hopeful results,
it's going very slowly and we really don't know,
even if we could make a good peptide-based, DNA-based, virus-based vaccine,
if that would be of any use to anybody.
So, we in my own laboratory have taken a different approach.
We've said, let's assume that you're never going to elicit an antibody
from humans that's going to be effective.
Let's assume that T cells may be good,
but they're not going to do the total job.
They're not going to provide protection that we need.
Is there another way to go about this?
Well, I said... gave you part of the answer already.
That is, that there are techniques that we have
for designing antibodies or antibody-like molecules
that should be able to attack the virus.
So, let's do that. But, if we do that, we have to give up all of our standard methodology,
because we can never elicit that from the body,
since the body has never seen, never made it, doesn't know how to do it.
So, we're going to have to direct the immune system
to make the thing that we've designed,
and that's where gene therapy comes in.
Because with gene therapy, we can put genes into the immune system.
Those genes now can encode kinds of molecules that the immune system
doesn't ordinarily make.
We've also taken a separate approach.
So, that's one approach.
And, that is to use something called RNAi.
RNAi is an interfering RNA. It directs the cell to degrade messenger RNAs.
So, let's say we targeted an RNAi to the receptor for HIV -- to CCR-5.
Now, if we could get that into a cell,
by gene therapy again, then we could protect the cell, so it couldn't
be infected by HIV.
So, these are two (and there are others),
but these are two methods that we're actually trying,
I'll tell you about the RNAi approach now,
and in the third segment of this lecture, I'll develop the ideas around the other approach.
There are a number of things that any sort of gene therapy approach
will have in common.
First of all, that the gene therapy has to be directed to blood stem cells.
Gene therapy is an old idea.
And it's been tried in many different contexts,
in particular for rare, inherited immune defects.
And it works. But, it is not standard therapy because it has side effect problems
that have held it from being adopted as a standard therapy.
So, what we want to do is get rid of some of those problems by actually using HIV itself,
which, in a funny way, is a safe virus for gene therapy.
And to bring therapeutically useful HIV derivatives into the body.
Now the stem cells are the ones that we want to infect.
These are not the kinds of embryonic stem cells
that have been discussed widely as new therapies for developmental and genetic diseases.
But rather, stem cells that solely give rise to blood,
and stem cells that are found in the bone marrow.
Now, I'm talking about using a virus in a positive way.
Using a virus that can do gene therapy.
This would be only the second case in history of viruses doing something useful.
Viruses are this enormous kingdom of agents,
but they never do anything positive for us.
They do mainly negative things.
They cause colds and they cause polio, and whatever.
But they do one positive thing,
and that is that they cause these wonderful variegations in the surface of carnations or tulips.
And in particular, in tulips, they cause something called Tulipmania,
which was a time in Holland when people were spending as much for a single tulip bulb
as they were for a house.
And that tulip bulb was actually infected by a virus,
and it was the virus causing the lovely variegations
that people were spending all this money on.
But, we, as scientists, can do something else useful with viruses.
And that is gene therapy.
And the reason for that is pretty straightforward.
I told you, retroviruses integrate their genetic material as a DNA copy
into the genetic material of cells that they infect.
But, they do not kill those cells.
So, they are natural carriers of genes.
That's in fact what a tumor virus -- a cancer inducing virus -- does.
It carries a gene into the cell. That gene now takes over the growth of that cell.
Well, we don't want to do that. We want to bring in genes which will have therapeutic value.
But that's a matter of getting into the lab,
engineering these viruses so they're no longer dangerous,
only do good things, can't do bad things,
and that is in fact what we're about.
Now, I keep talking about hematopoetic stem cells
or blood stem cells. Let me make that point explicit.
There is, in our bone marrow, something called the hematopoetic stem cell
(HSC). That cell, when it divides, either gives rise to more of itself,
or it gives rise to either the common lymphoid progenitor
or the common myeloid progenitor.
Those in turn give rise to other cells,
and those in turn give rise to all of the cells of the blood.
So, red blood cells, platelets (those little things that help you coagulate the blood),
granulocytes that fight off infection, monocytes that fight off infection,
and B cells and T cells, which are the two kinds of lymphocytes
that make... B cells make antibodies, T cells kill cells directly.
So, all of those cells actually derive from this cell.
If we can modify the genetic properties of this cell,
it will be maintained because it's self-renewing in the bone marrow,
and it will give rise to all of the blood elements.
In particular, we can sneak in this way new kinds of T cell receptors
and new kinds of antibodies, or sneak in a protective molecule like an RNAi.
So let's focus on the RNAi concept.
We're focusing on CCR-5, the co-receptor for HIV.
Why is that a good target?
It's a cellular gene, don't we need it? If you knock it out, won't we lose important function?
It happens that there is a natural population of people
who carry two mutant copies of the gene for CCR-5
and can't make any CCR-5.
So, their cells have no CCR-5 on their surface.
Those people are actually resistant to HIV infection,
because CCR-5 is an absolute requirement for HIV infection.
But, amazingly, they have virtually no other immune defects.
They may have defects in relation to one or another specific organisms,
but to a first approximation, people live a normal life without this gene.
So, if we can knock down this gene in the cells of an HIV-infected person,
those cells will now be able to grow up and replace helper T cell function.
Now, some people carry one mutant gene and one normal gene.
Those people make somewhat less than 50% of the normal amount of CCR-5,
and they develop AIDS much more slowly than normal people do.
And so, we know that even if we could just knock it down 50 or 60%,
we could actually affect the course of the disease.
If we can knock it down 95%,
we can probably have a major effect.
So, what we're trying to do is to bring into helper T cells
an interfering RNA (it's called an RNAi or siRNA or an shRNA, variously)
which is able to block the translation of the messenger RNA
for CCR-5 or cause degradation of the messenger RNA for CCR-5.
And thus, it'll act like that mutation and protect those cells and those people.
How do you do that?
This is a vector based on HIV.
But, if you knew what all these abbreviations meant,
(and I won't go into them)
what you would see is there's very little HIV left here.
All of this is stuff we've put in there.
We've left just the bones of HIV,
because those bones help the mechanics of this virus work.
And in particular, we have put in here,
this little cassette, which encodes an RNAi.
RNAi is an RNA that has two strands that are complementary to each other.
That's what these two arrows are.
They're held together in here by a loop. The loop has to be gotten rid of.
So, when this goes into a cell,
it is transcribed. You get this little hairpin of RNA.
It's got one strand here, the other strand here,
and the little loop over here.
There's a protein in our cells called Dicer that recognizes these loops and cuts them out.
So now we have just the two little strands,
about 20-odd nucleotides long.
They get piled into something called the RISC complex with a bunch of proteins,
and only one strand of this two strands ends up in RISC.
RISC with the RNA in it goes around looking for other RNAs
that have the complement of this sequence.
They find them in messenger RNAs.
They bind to those messenger RNAs
and either cause them to be degraded or cause them to stop translation.
So, we can specifically target CCR-5 and, in principle, nothing else in the cell
with this methodology. Does it work?
Yes, it does work.
I don't want to go into the details here,
but if you look over here at this black line, this is the amount of CCR-5
in normal cells. This is no CCR-5 at all.
So, you can see that a lot of cells have CCR-5.
This is if you put in a very good RNAi.
All those cells have gone away... there's just a little bit left down here.
So, when we quantitate this kind of thing,
we can see that we've gotten rid of most of the CCR-5.
And now we want to infect these cells with HIV and see if they're resistant.
And so, we go in with an HIV here that.
Sorry. We go in with an HIV here that tests whether these cells are no longer infectable.
And, this is the kind of data that you get.
I won't go through the details, but if ordinarily 50% of the cells have CCR-5,
now only 3.4% have it.
And, if ordinarily, you infect 7% of the cell, now you infect 2% of the cells.
These, in fact, were our early attempts to do this.
We can now get much higher protection than this.
And we are actually taking this into human beings, trying to do Phase I trials as we speak.
So, our conclusions are that you can make a vector based on HIV.
HIV is actually called a lentivirus, so these are called lentiviral vectors.
It can deliver an siRNA, which is specific to CCR-5,
to primary peripheral blood lymphocytes.
We've proven that.
And that was the data that I was showing you.
We haven't shown we can do this in bone marrow,
but we believe we should be able to.
We can get reductions in the range of 10-fold of CCR-5.
That's a lot more than we already know is at least partially protective.
We can show that the inhibition is quite specific. I didn't mention that.
We can get, now, roughly 5-fold reductions in the number of infected cells.
That's a lot -- it means a lot of protected cells around.
And, there's another virus that doesn't use this receptor
which is unaffected by it.
So, we have the conditions for doing well.
We've gotten better at making these siRNAs,
better about delivering them.
The only thing left is to see if we can actually do this in human beings.
And we're trying to do that.
So, what will it take to make this a real therapy?
First of all, as I said, we need to optimize the inhibition of HIV growth,
and that we've done.
We need to organize a clinical trial.
For something like this, which is an agent that's never been tried before,
this is a complicated process that requires all sorts of regulatory approvals.
But, we're in the process of forming a company that will do that,
and we hope to see progress in the near future. Thank you.