From Structure and Function of Ribosomes to New Antibiotics

Uploaded by YaleUniversity on 20.01.2010

Well, it's a great pleasure to be here, and talk to you all. I have to say I've never
seen such a big audience
here. It's terrific.
So what I'm going to do is I'm going to give essentially
the talk that I gave in Stockholm. A few little add-ons I have to say, too.
And I'll start
with a--
a little bit of a history
tour of how I
got where I am.
Not a lot but some.
And I should say that my
passion, and it really is my passion,
for trying to understand
how macromolecules function in biology, and understanding the biology of the
molecular level--
started from a lecture
that I heard from
by Max Perutz in
It was a Dunham lecture at Harvard,
and he talked about
his structural studies
on hemoglobin,
studies on myoglobin.
Now they had won the Nobel Prize in Chemistry just the year before for their determining
and so Max
showed a stereo slide
of myoglobin, and the
with at least as many people as here, or maybe even more--
and it was a very large screen over his head and
we all had colored glasses on, and he went back and forth--
you know how that works.
And all of a sudden it popped into three dimensions.
And the audience
went, "Whoaaa."
I don't expect that will happen here.
But none of us had ever seen
a molecule, a protein,
in three dimensions.
And that was when I
that's what I wanted to do. I wanted to understand
biological molecules of the atomic level in three dimensions, and that's what I've been
So then I went and worked in Bill Epstein's lab.
at Harvard on the structure
of Carboxypeptidase A. Lipscomb won the prize in--
the Chemistry Prize in 1976 for his work on boron hydrides-- not exactly
protein crystallography.
And then I-- and there I got to interact with
the Watson group,
learning about ribosomes, which they were studying at that time--
and also Wally Gilbert on transcription regulation.
And then I went to Cambridge
to work on
and interacted
Francis Crick.
it was really my interactions with those individuals,
particularly at Cambridge,
that triggered my interest in
trying to understand
the simple dogma
of molecular biology.
The central dogma, as you all know, is:
is copied into DNA by DNA polymerases and we enable studied many DNA
polymerases, and continue to do so--
and then the RNA is copied into RNA-- or, DNA is copied into RNA
RNA polymerase, and that's regulated, we've been studying that.
And then the tRNA has an amino acid attached
by amino acetyl tRNA synthetase, and we've been studying that.
So we went through this
and then about
we got to the point that--
we'd reached
the stage where we need to look at the ribosome.
Now, when I was a graduate student at Harvard,
this was a figure that Jim Watson made
for a publication, and that was what was known in 1964 about ribosome structure
and function.
It was known that there was an amino
acetyl tRNA-- didn't know what the structure was--
and there was decoding
by the messenger RNA in the small subunit,
and there was a peptidyl trine
and the-- what
was called the P site.
But it wasn't known
that there was an E site, the exit site;
it wasn't known that the polypeptide went through a tunnel;
and there were no molecular
at all.
It was known that there was
a translocation of-- a number of good things known.
Well, the ribosome
is a very large RNA-containing machine. It's the largest RNA-containing
in cells.
It has three thousand nucleotides in the bacterial
ribosome. Large subunit.
34 proteins. And then the smaller subunit that is about half that size.
It's two thirds RNA by mass
and it's one quarter of the bacterial cell mass. And that's useful--
if you want to do structural studies, you need to be able to make materials, so that was very
Now Jim Lake started structural studies
in 1976, and this is using
electron microscopy,
negatively stained, and
he could get the shape of the ribosome, and you see that's--
it's pretty much right.
This is called the crown view for obvious reasons,
and then this was the shape of a small subunit, and then they sort of snuggle up to each other
when they form the 70S.
And then in
Joachim Frank
used cryo-EM methods--
single particle cryo-EM methods to get us
somewhat higher resolution
something alike
twenty four X resolution, and you see little indentations.
He positioned the tRNA's-- the A site, P site, and E site, and-- but
not exactly accurately, and
guessed where the
messenger RNA would go.
So that's where it was
in the mid-nineties,
and as I said, we had gotten to the point where we
were thinking, what should we do next?
And Ada Yonath had,
starting in 1980,
grown crystals of the ribosome,
and then got crystals of the
species that I'm going to work on the Haloarcula marismortui-- a large
But then she published
a paper in 1995--
now, not everybody in this audience is going to look at this and be able to decide whether
it's good, bad, or indifferent--
but this was a 7 x resolution electron density map
that she
thought was
moving along to understand the ribosome structure. Well, many of us
looked at that and realized
that it wasn't actually
correctly done yet.
So we wanted to get into it. Well,
the right person at the right time
came. Nenad Ban
joined my lab in 1995
and he wanted to work
on the ribosome. So I said,
"Great." And then I decided we should
with Peter Moore. I've known Peters since we were graduate students--
graduate students at Harvard--
and he's a long time ribosome expert
and he likes to catch big fish, as you can see.
And of course, the ribosome is a big fish.
And so I thought,
yes, he is the right person.
And then Poul Nissen
came from
a couple years after
Nenad Ban,
and he joined in, and it really was Ban and Nissen
who I think were the core
of solving
the structure.
So then-- you might ask, well, what was the problem?
Why was this hard?
Well the ribosome's big. But so?
And what does that mean?
So we were able to make crystals following Ada Yonath's procedure, but the crystals
initially had a lot of problems. They were very very thin, they were
not, well-- they often had
They had problems.
And then we had to improve that.
But what you can do-- I'm not going to give you an x-ray course here, you'll be be happy to know--
I give a semester course, or half-semester course, if you wish--
but what I wanna say to you is, the problem is--
you can measure the intensities of the diffraction pattern,
but you need to get something else:
it's called the phase.
I won't go into what that is. But the way you do that
is to use
a method
that was--
initiated by Max Bruce-- one of the reasons he got the Prize, and that is, you bind a heavy
to the crystal
and you collect the data again,
and you locate that heavy atom.
Never mind how you do that.
And I say the problem with the ribosome
is like the problem
of trying to measure the weight of the ship captain
with and without the boat. Now if you have a sailboat
and a pretty good-sized ship captain,
you can measure these two and measure this, and then subtract them, and you can
do it. And that's sort of like
a lysozyme-- 14,000 molecular weight,
and tungsten was 78 electrons.
if you try to do that with the Queen Mary,
this is a bigger challenge.
Because the error would be much larger and you can't measure the weight of the ship captain.
And I think this is like--
the ribosome's sort of the Queen Mary of the macromolecular assemblies.
And so
this is not a heavy enough
atom. So what one had to do
was use
what are called cluster compounds.
So this one, tungsten 18, has 18 tungsten atoms,
and with the other atoms it has 2,000 electrons instead of 78.
And the-- we use some other cluster compounds as well.
And it turns out that--
this is the most detailed crystallographic
part of the talk you're going to hear--
I won't go--
press on this for a long time--
but this is the important point and this is what got us going
and the others as well, subsequently. So--
at very low resolution,
not that you
you all know what resolution is, but at low resolution like 20 angstroms, which is down here somewhere,
the scatter
from these cluster compounds
goes more or less through the ceiling.
And that's because
the scatter goes as the square of the number of electrons,
and so the signal is
about six hundred times larger
then the signal from
just a single atom.
And here,
this is the scatter
with a hundred osmium hexamines.
So this is a very--what we did is we took
very low resolution data,
we got the signal, we measured the weight of the ship captain,
in spite of the fact it was the
Queen Mary,
located the heavy atom, and started
solving the structure.
and this is where Joachim Frank was
using cryo-EM--
I would say this is like a Moore statue--
(not Peter Moore; Henry Moore)--
with the--
he's holding a torch and there's his head there's his arm, he's kneeling down here--
so it was very hard to make an atomic interpretation of this...
So our first publication was in 1998,
in which we had a nine angstrom resolution map--
very low resolution--
but we could see
the twists and turns of the RNA, so we knew
we were on the right path,
and besides,
it looked--
the shape was the same as the twenty angstrom
of reconstruction.
And then the next year
we got to
five angstrom resolution-- in fact
Ada Yonath and Venki Ramakrishnan
followed the same
route, and they got there, too,
as well as
Harry Noller,
and then that--
we got to 2.4 angstrom resolution,
and you can't see--
certainly not in this room--
the detail here, but we could actually position every single atom.
It took us several months to do that, but we were able to do that.
And so here is the structure that we got, or a representation of the structure that
we got in
the RNA is shown with the red backbone-- sugar phosphate backbone-- and
white bases.
Very tightly packed, RNA, so
we had no idea what RNA was going to look like
in this large assembly.
Not a clue.
Proteins are in blue,
embedded around the surface.
Again, the contrast isn't very large, but there is where the substrate binds
and you'll notice there's not much blue there.
I'll come back to that
So this is--
this is the large ribosomal unit that was determined here at Yale.
This is a small subunit determined by Venki Ramakrishnan
and put together using Harry Noller's data
on the--
on the 70S.
And here are three tRNAs borrowed as well.
And we're going to spend most of our time down here
in what's called the peptidyl transferase center, where peptide bond formation occurs.
That's where the amino acids are stuck together.
And then
this hole is where the messenger RNA goes in
and Venki Ramakrishnan got a lot of information as to how in the world
the decoding goes.
If I just rotate this around, you can see the very complicated
that the RNA takes.
It's pretty closely packed
but there are little gaps.
And the protein--
the RNA has little gaps; protein fits
into those--
into those gaps.
Again, you can see the complicated shapes
Now we split the
large subunit open like--
like an apple
and this is a space-filling rendition.
In white
is RNA--
you can see, again,
that's pretty tightly packedÉ little-- little holes here and there--
and the protein is
is penetrating into the RNA to sort of
act as a little bit of a glue to hold it together.
And here's, again, the peptidyl transferase center, PT,
and this is a model built
where the polypeptide exit would be.
That's about a hundred angstroms long.
So that's the exit tunnel.
So there are a lot of proteins--
I think there're about fourteen--
where there's a globular domain,
and then just a sort of wind-y,
highly charged piece
of protein,
and that
piece of protein--
these pieces of protein sort of
fit into the RNA,
sort of hold it together.
And particularly I point out L2 and L3, which have these big
red parts,
and then if we look at where they are,
a close-up view,
here's where the active site is,
an analog--
substrate analog here.
There is the globular part
and the non-globular part just sort of threads in
It comes close to the peptidyl transferase center
but it doesn't
actually make it all the way.
Now Francis Crick,
who always predicted good things-- I mean, right things--he's amazing--
he thought in 19--
about 1968,
the following quote.
I'm not sure if I can do it in a British accent.
"It is tempting to wonder
if the primitive ribosome could have been made entirely of RNA."
And of course the reason for saying this
was that, how could you have
a protein make the first protein? Because how would you get the first protein
to make the first protein? It's a chicken and egg problem.
So it was clearly a good idea and really endorsed by
almost everybody
through the years, but there wasn't
any proof. So when we looked at what's the closest polypeptide to the peptidyl transferase
this is what we found
in 2000.
Now, it turns out there's a little bit of a disordered
chain here that we did know about at that time,
but it still can't get close enough
to get to the peptidyl transferase center.
So-- just so you don't miss the point--
the ribosome is a ribozyme. That means
the catalysis is done
by RNA.
Now of course, Sid Altman
shared the prize
twenty years ago
with Tom Cech
for also discovering the first
So, ok-- so RNA does the catalysis. How does it do it?
What's the source of the ribosome's catalytic power?
'Cause RNA wouldn't necessarily be a chemist's first choice as to what
he or she would use
to catalyze a reaction.
So how does it do it?
Well this work was done
initially by Jeff Hansen and then
most of it
by Martin Schmeing.
Jeff a postdoc, Martin a graduate student at that time.
And here's Martin
with his sailboat
on our salt marsh.
He had finished his degree but I wanted him to finish two papers, so I said, "Ok, you can
put your boat on our salt marsh and you just hang out
for another six months
and he did, and he published a paper in Nature on--
I forget, cellular and molecular biology or something.
So that was good.
I point out, by the way, that global warming is a problem.
This picture was taken five years ago
and now
the salt marsh has eroded, so this rock
is up against water.
And my geology colleagues
tell me that that's because the ocean levels are rising and the salt marshes are eroding
very fast.
Well, anyhow--
back to--
back to
peptide bond formation.
So here is the reaction that's catalyzed; it's kind of a dark slide,
I have to apologize.
There's an A site tRNA that has an amino acid on it,
and it's the alpha amino group of that amino acid that
attacks the
carbonyl carbon
of that peptidyl
And that's in the P site.
Well, we couldn't get
these tRNA's into our crystals
so what we did is we used a fragment, and biochemists have been
using a fragment reaction for many years.
And so one just uses CCA
amino acid, or CCA and a peptide analog.
Then this goes on to form what's called a tetrahedral carbon intermediate
with a negative
oxyanion here,
and then goes to product--
with the product winding up in the A site and has to translocate back to the P site.
So what we did-- what we've done--
and I'm not going to show you but a very small fraction of the work on this--
is solve many structures of intermediate products
and substrates to try and put it together in a movie, and
I will show you
a movie
of this process that Martin Schmeing made, which is always
quite popular.
So again, if you take the 50S subunit and split it
so that you can see the
exit tunnel,
here the three tRNA's borrowed from Harry Noller,
the proteins penetrating into the interior,
and there's the peptidyl transferase center.
And we're going to be thinking inside this--
this little box for a while
and addressing the problem:
how does peptide bond formation happen?
Well, initially what was done
was to find
to the
A site
with an amino acid on it, and it binds to what Harry Noller had described
earlier as the A loop.
Hydrogen bonds to
one of the base--
from one base pair.
And the CCA with a peptide
to what's called the P loop
Again, making
two hydrogen bonds
in this case.
And there's
the attacking alpha amino group-- it has to attack--
we looked at-- we said, well,
that orientation doesn't look quite right, but, eh,
it's probably just not an accurate structure.
Well then, I went to--
Sweeden, to give--
there're a lot of Nobel
symposia that
one can get invited to sometimes.
I thought it was sort of like a qualifying exam.
You're up there
giving you talk, and the Nobel committee sitting in the front row, and you're like,
"Oh, my God."
The students understand that, right?
So the question was asked, why is peptidyl tRNA not hydrolyzed
in the absence of A-site tRNA?
If the structure
is all set up
to go,
well, why can't water attack
the carbonyl carbon instead of the amino group?
And this was the same
question that was asked by Kashlan [sic]
fifty years ago about hexokinase, and why we studied hexokinase in the first place
is because he asked, well, why isn't
ATP hydrolyzed in the presence of glucose? Because, after all,
glucose is just a water molecule with a few carbons attached.
And he said, Ah! there has to be a conformational change.
And that's--
that's the answer here as well.
And so while I was unable to answer the question is Sweden,
Martin Schmeing was back at work in the lab and came up with the answer
by the time I got back.
And so
what he found was that
if you don't have the A-site substrate bound,
this peptide-linked
group, the peptidyl group,
is buried.
So this is where a water molecule would have to be, either here or here, and it can't get there;
it's sterically blocked by--
sort ofÉ
So what he did is he looked at several substrates,
and I'll show you a movie-- this will probably be a little easier to understand--
but the major point is that if he has
CA in the A-site there's not much of a conformational change
but if he has CCA--
and there are a lot of biochemical data for why one did this--
what happens is this
protecting base moves out of the way
and there's a reorientation--
and it's actually--
of the peptidyl group--
and reorients so that it's
now ready
for chemistry.
So what happens--
ok, here's the uninduced state.
Wrong orientation, really.
But then, when you get induced state, this moves out of the way,
and you get this
in position to attack,
which it can then do, and you get
this tetrahedral intermediate, and the oxyanion
that's formed,
and that's what it would look like
with a substrate side-chain on.
So that's--
that's what it looks like.
The ground state where it starts.
And so there are two
groups that could be
involved in the chemistry of the catalysis--
now all catalysis by enzymes
is activated by,
or enhanced
by, the orientation of two substrates. It's entropy.
describe this
nearly forty years ago.
So that's the biggest effect. All enzymes do that. But what else do they do?
Well, there's a hydroxyl group here from the
P-site sugar,
and there is a base here
with a nitrogen.
Well, initially we thought, "Well,
we wonder if that could be the base."
Not our best hypothesis, I have to say, 'cause it turns out to be wrong, as I'll mention.
And the other possibility is that
this hydroxyl has an important role, and that
turns out to be right.
So, Rachel Green
mutated that
particular A to all the other--
all other-- three other bases,
and found
that with a full-length tRNA, it made no difference, so
clearly that wasn't important for catalysis.
However, Scott Strobel,
also working with Rachel Green,
showed that if they removed that
hydroxyl group
from the P-site substrate,
the rate went down
a very large
amount. So clearly it was important for something.
A similar experiment, but using fragments,
had been done
by Barta
in Austria
and she found
just a couple hundredfold difference, but she found, also,
that when this hydroxyl was removed,
the rate went down, and so she hypothesized--
using our structural information and her
result-- she hypothesized
that this hydroxyl is acting as sort of a shuttle.
So it's
acting as a general base, namely picking up the proton from this attacking alpha
amino group
and then acting as a general acid
to provide it
to the leaving
three prime hydroxyl group.
And there have been some experiments that have been done subsequently, and that's
pretty well, I think, agreed upon
as the mechanism
and what the enzyme is doing.
Then another thing that enzymes do is stabilize transition states, lower the activation barrier,
and is it doing that?
Well, we used a substrate--
again, synthesized
in Scott Strobel's lab
by Kevin Huang,
and this is an analog of the intermediate,
and the carbon is of mimicked by a phosphorus
of the oxygen by sulfur so we could see it--
and then there's a mimic of a side chain.
And this is quite high resolution,
considering it's a ribosome,
I might say.
So we could see where all the atoms are, and if we look,
we see that there's a water molecule interacting
with this
oxyanion mimic
being held in place
by the RNA,
or the ribosome.
And so that
could indeed be stabilizing, because
there's quite a dipole moment
on the RNA--
around the water, rather. Sorry.
So, what's the source of catalytic power?
Substrate orientation by the 23S
RNA. Very important.
A proton shuttle
by the two prime hydroxyl.
And probably transition state
Well, I think everybody's ready for a movie.
And so this movie was made by Martin Schmeing.
Music was
composed by others.
And what he did is he took
each of the many structures that he
and then morphed in between these structures, so every time there's a little R-factor down
this corner,
that's a real structure.
and things in between might be
a little Disney-esque.
Here's the
large ribosomal subunit
and there's the peptidyl transferase center,
and so
we're going to land
at the peptidyl transferase center...
And we'll concentrate
on the peptidyl transferase center.
We have now
the P loop
and the A loop.
We're just showing that.
Now we have the
peptidyl CCA
P site substrates coming in, making its--
two base pairs.
And you see the carbonyl group
is protected
by these two bases.
Now we have the attacking A-site substrate
making one hydrogen bond--
and this is the uninduced structure-- and then
now there's a conformational change-- you get the induced structure, so this gets oriented properly.
There's a stabilizing water molecule.
Now we're ready
for the attack
of the
alpha amino group.
The alpha amino group--
ok, it seems to have worked.
So here's a tetrahedral carbon with the oxyanion
with a water molecule,
and then it breaks down to get the product.
And then the diacylated
or actually, tRNA in the real case--
goes to the E-site, the exit site.
So now it's--
this is made up, of course, we don't know how it gets there exactly.
going over two the E-site and the CCA will be
interacting with the E-site.
The A76 is going to be stacked in here
Now there's no room for an amino acid, which is why only the diacylated tRNA
gets in here.
And now-- now there's one more trip-- at the rotation of 180 degrees,
the P-site
CCA goes--
or A-site, rather, goes to the P-site.
So that's...
that's what we know about it.
SoÉ I gave a lecture once in Cambridge at the hundredth-- fiftieth anniversary of
the structure of DNA, and
I got to this point I asked Richard Henderson, who's the head of the laboratory,
"Should I continue?" And he said, "I don't think you can pick top that."
I'm going to continue because I want to move on now
and talk about the
practical applications
of the ribosome
And that is:
it's a major target
of antibiotics
and we can understand how it works
from our structural studies, and
the information can be used, and is being used,
to design new antibiotics--
not by us, but
by Rib-X Pharmaceuticals, as I'll get to.
So as you all know,
resistant strains of bacteria are a major problem. This was in the New York Times a couple years
about MRSA-- 0:32:56.289,0:32:58.970 methicillin-resistant staphylococcus aureus--
the number here is 19,000 deaths in US hospitals alone.
I've seen numbers up to 99,000
for all resistant strains
Anyhow, that's a large number of people who die from
lack of ability for the antibiotics to work.
most of the antibiotics--
sorry, about half the antibiotics target the
large ribosomal subunit, the one we're studying
actually about half of them do.
So it's a major, major target.
Now this work was done by
Jeff Hansen, who also did some of the substrate work.
This is one of our summer picnics.
One of the nice things about living in New Haven, as many of you know,
is the shoreline is gorgeous, and so it's
really very nice.
So Jeff
started by binding macrolides to
the ribosome large subunit.
And I'll--
erythromycin is probably one that you know about--
We've taken you know what it is-- maybe all of you--
a Z-pack,
now it turns out that
eubacteria-- and
Haloarcula marismortui is not a eubacteria,
but an archaebacteria.
are really eukaryotes in disguise.
They hung out with the
eukaryotes for about a billion years after they split off from the eubacteria,
and so they're very similar and--
and so how can we bind antibiotics
to this--
eukaryote lookalike? Well it turns out you just go to very high concentrations.
And with some of them we've been able to do it.
And so here
that's where the macrolide ring binds.
Peptide bond formation occurs up here.
This is down in the tunnel
there's three of them here-- the rings bind more or less the same way and they differ
by what the substitutions are.
That's how the
the different antibiotics differ from
one from the other.
So how do they function?
Well, again, this is a split-open ribosome.
That's the position of the macrolide.
In green are residues of mutation--
make the
ribosome resistant
to the antibiotic--
and now if we look up the tunnel,
we can see-- there's where peptide bond formation occurs and peptide comes out this tunnel, but--
there's what happens
there's where the antibiotic binds,
blocking the exit of the polyeptide...
I call this molecular constipation.
Actually, I said this in Wweden and nobody laughed. I think they
all understand--
their English
is really good-- but I don't think they understood constipation.
The word, that is.
So we did some studies that try and understand how these mutations actually
effect the
binding of the antibiotics.
Now it turns out that
in E. Coli,
mutation-- this A
reduces to the binding constant of
erythromycin by 10,000 fold.
And it turns out that in
Haloarcula marismortui,
that A is a G.
Just as it is in eukaryotes.
And so it doesn't bind to a large
of antibiotics
called MLSK,
including erythromycin.
Why is that?
Well, when we look
at where that
nitrogen that
on the G but not on the A
it's right under this hydrophobic ring,
it has to get these solvated,
and what's more,
it doesn't fit as tightly.
So that, we hypothesized, was the reason why
this bound
much less tightly.
So what was done by Daqi Tu, who's a joint graduate student between Peter's lab
and mine,
and Gregor Blaha, a post doc in my lab,
was to
mutate the
Haloarcula marismortui--
that G
into an A.
Sort of the reverse of a resistance mutation.
A non-resistance mutation.
And when they did that,
they found that all
of this
family of antibiotics
that didn't bind previously
bound. Erithromycin,
just a whole bunch of them bound very well,
and if you look--
with erythromycin,
there's a G here
even at three Millimolar, which is saturated
it didn't bind.
with the mutation
.003, which is about the lowest we could use in the crystal,
So we estimated
about a ten thousand fold in the binding counts. And then if you look
at one compound, Azithromycin, that we could bind
at very high concentrations, 10 Millimolar--
both Haloarcula marismortui
and to the mutated
you see that it moves down.
By about an angstrom. But not much.
It's roughly in the same place
And there is where that pesky little nitrogen is that gets in the way of its binding.
So that's why, I
should say,
use of these structures--
well, it is useful to use these structures
to design new antibiotics, in spite of Haloarcula marismortui not being a eubacteria.
I might say it's more important to have an accurate structure,
perhaps, than a
eubacteria structure.
I just want to point out that
Ada Yonath's work
early on--
she had the antibiotic
erythromycin oriented this way, and
it's actually
oriented this way.
And we've now looked at
another eubacteria and see exactly the same thing, so this is the correct one.
Now, different families antibiotics bind in nearby places-- I'm not going to go through them all--
here's a peptidyl transferase center.
Peptide bond formation.
Here's one that binds to what's called an A-site
because that's where the A-site subject binds.
There are many that bind here.
Here is Sparsomycin,
Chloramphenicol, which also binds up here,
So what do you do with this information?
And here is where I'm going to transition
into what
is being done at
What you can do is you can take a piece of one of these molecules
and tie it
chemically to piece of another one to make a hybrid
You use that using the structure,
developed initially in Bill Jorgensen's lab in the Chemistry department--
and these methods work very well.
And so Rib-X,
which is on George Street-- 300 George Street--
has been working since 2001, when
I and several of my colleagues-- Peter
and Bill and others--
founded it,
and there they're getting good
So here's an example.
Never mind the chemical details here, you can't see them anyhow and it doesn't really matter-
so what they're doing is they're taking
a one-compound linezolid
which is
sold by Pfizer, it's a major new antibiotic-- new--
some years old, but--
and combined it sparsomycin, which is basically a rat poison-- it's not discriminating.
So if you look at the selectivity,
linezolid does select eubacteria but
not eukaryotes,
whereas sparsomycin isn'.t
And then what is measured by the microbiologist
is something called the MIC.
It's not--
M-I-C-K-E-Y is what many of us know,
M-O-U-S-E, but--
this is--
M-I-C is the minimum inhibitory concentration.
And you have to be below 4 to be effective.
And linezolid, as you can see, is quite effective in many of these cases.
But not all of them.
And so these are the various combinations they made, just tying them together, never mind the details--
and they found,
for example, one here that looks pretty good. 0.5, 0.5,
16 to high 128, still not good.
So then what you do is you do cycles
of this. And again--
a whole bunch of cycles,
let's say 260ish,
and then you look at the MICs
against two tough strains.
So, you see, this this was one that
they didn't target;
neither did linezolid.
But there it is; it's down to two, and this is very good.
So they made compounds, now--
they call it--
they've got one that they call Radezolid
and it will work against all kinds of resistance bacterial strains.
Here is Zithromax Z-pack,
doesn't work against any of these,
and here's a Vancomycin
and linezolid resistant--
Vancomycin resistant
works very well against--
these compounds from Rib-X work very well.
And it's not just MRSA that's a problem; there are
other other kinds of--
of resistant strains.
Again, enterococci--
here's what Radezolid does,
all fine;
not good enough...
So Radezolid's now made it through phase two clinical trials,
and so it's looking pretty good.
Many of us are very enthusiastic about that.
The company has several compounds
that are in pre-clinical
trials that were looking good,
getting lined up
to go into the clinic, including
one that was a de novo design, just starting from scratch,
that works against gram negative bacteria, which is also a major problem.
So I think it's going to be very good.
And then I'm going to end
the science part with just one last thing:
where do we go from here? I always get asked, "Well, what are you going to do next?"
And I always say, "The future is hard to predict."
let me talk about tuberculosis.
192 people will die from tuberculosis during the next hour. It's a major,
problem, not in this part of the world, but certainly in the third world.
Although I think we
could get into trouble as well.
So what we have done
is we've bound two
compounds that are active against TB
to 70S ribosome structure that we have done
with three tRNA's bound.
And let me just say--
and this is in press, it's
not out yet--
so it binds between the two subunits,
and there's the tRNA-- that's in the A-site-- now we're zooming in.
So here's the tRNA
that's being--
it's interacting with the messenger RNA--
this is what was found initially by Venki Ramakrishnan--
and there are two bases that are flipped out--
1492 and 1493--
that are stabilizing this interaction.
And what this compound Vyomicin is doing is stabilizing
this whole ensemble, just gluing it right there
and stopping it.
Now why is this important?
Well, it turns out that
here's Viomycin--
Capreomycin binds in the same place--
and there are two compounds that
found were
bound to the small subunit:
Hygromycin B
and Paromomycin.
And, once again, one can only think:
well, maybe
the Rib-X trick will work here, and maybe one can make new antibiotics.
And there's a strain called XDR
that is arising in south--
southern Africa.
A major resistance strain,
because resistance can--
all antibiotics, no matter what the target molecule is,
and the question is:
can that be
The challenge, of course, is going to be
to get funding for that because
this cures poor people, but
hopefully we can
do something about that. So I'm going to stop there. I'm going to make some thank you.
If I--
if I may.
A few people I didn't mention
in Peter Moore's group: Betty Freeborn
was the technician who made the ribosome early on, and Larysa
made them subsequently.
Ah, let's see, who else didn't I mention...
Dan Klein worked on many of the structural aspects
in the later stages
and I have to mention that Scott Strobel
had his
lab synthesize some of the compounds.
And I should also mention my fellow
Chemistry winners,
Ada Yonath
and Venki Ramakrishnan. And this is before he discovered he needed to shave in order to get through
He said he always was, you know, patted down and had all kinds of trouble,
but as soon as he shaved,
boom! Right through.
So in case you think there's no profiling, you're wrong.
So this was at
another one of those meetings in Sweden
a few years ago, very nice
And then I should
mention that I was able to take
fourteen guest plus
Joan, of course,
with me to Sweden, and
I had a number of family members.
But I was also able to take a Yale--
associated Yale people who our President wants.
Peter, of course,
a pivotally important person-- Peter Moore.
And then there's Nenad Ban
and Poul Nissen.
Nenad's now the head of a
very successful laboratory in Zurich,
and Paul is doing very well
as the head of a structural biology
in Arhus.
And then this is Jeff
and Martin Schmeing.
And I also
Peggy Eatherton to come along because she--
she handles
so much of my workload, particularly the paper
workload, and
she stabilizes the whole lab crew, making them work together, so that's very helpful.
So we had a good time.
And I also have to mention
the environment in which all this happened.
In 1995,
here were the people who were in
structural biology
at Yale.
There was the the WERMS group,
called WERMS because it's Wyckoff,
Moore, and Steitz.
And then
in the mid- to late eighties,
HHMI started supporting
and by 1995, there was Jennifer Doudna,
Paul Sigler,
and Axel Brunger.
This was a great group
seven out of the eight
are or were in the National Academy of Science.
Very, very distinguished group, and
I have to say,
science doesn't happen in a vacuum.
It's very important to have a
good environment within
which to work.
And this was a great
And this is the hooha
that happened
the presentation--
a wonderful banquet.
And the servers, I don't know they did it-- there were
fifteen hundred people there,
gorgeous tables and
music and
and lots of singing and dancing
some of us--
there's Venki,
there's Ada Yonath--
little hard to see.
The Crown Princess.
I'm behind the candlestick here--
sitting at the main table.
And here's what was on the
front page of the Swedish newspaper--
the Stockholm newspaper
the day after the Awards ceremonies.
I naturally thought--
"Gosh, isn't the Crown Princess lucky that Venki and I got her on the front
And then--
then somebody pointed out that maybe it was the other way around.
So, in the future--
who knows when the future is--
the Nobel Prize will be given out
by the now-Crown Princess, who will become
the Queen of Sweden. So I'll stop there. Thank you.