YouSub
Untitled

en English

Untitled English subtitles

0 downloads · 466 lines · added Nov 22, 2015

Report a problem
What's wrong?
Advertisement

English transcript

I'm Bob Lefkowitz, an investigator of the Howard Hughes Medical Institute
and a professor of medicine and biochemistry at Duke University Medical Center in Durham, North Carolina.
Today, I'm talking to you about drug and hormone receptors,
and in particular, about one remarkable, huge family of such receptors, called the seven transmembrane receptors,
also known as the G-protein coupled receptors.
In my first lecture, I told you in some detail about the history of how this field evolved
and how we traveled from being quite skeptical about the mere existence of these receptors,
to a fairly detailed understanding of how they function.
Today, I want to bring you right up to the minute, talking about interesting and exciting discoveries
over the last five to ten years, which are completely reshaping our understanding of how the receptors function,
and are regulated, and which can be leveraged to develop entirely new types of drugs.
The focus is going to be on a class of molecules called beta-arrestins.
Now, I introduced this in the first lecture, but to just briefly review,
the classical signaling paradigm for understanding how G-protein coupled receptors work is as shown here,
and was derived over the years largely from studies on two model systems:
the beta-2 adrenergic receptor, for catecholamines, like adrenaline and noradrenaline,
and the so-called visual receptor, rhodopsin, for which the stimulant is really photons of light.
Now, the key feature, as shown here, is that when the receptor is activated, by adrenaline or by a photon of light,
it changes conformation, becomes activated, stimulates G-proteins, leading to activation of second messenger-generating enzymes,
and those second messengers activate kinases leading to cellular responses.
But as we talked about, an opposing system immediately is engaged, which tends to shut down,
or desensitize receptor action.
This involves a two-step mechanism. In the first step, a G-protein coupled receptor kinase,
such as GRK2, shown here, phosphorylates the active conformation of the receptor,
leading to a second step, in which a beta-arrestin molecule binds to the phosphorylated receptor,
physically blocks G-protein coupling, sterically interdicts it, thereby leading to desensitization.
But the exciting thing in this field, which has developed over the last few years,
is that an entirely new signaling paradigm is emerging.
And that is shown here. The notion is that the GRK/beta-arrestin system is really bi-functional.
At the very same time that it desensitizes G-protein signaling, it also serves as a signaling unit,
a signal transducer, in its own right, which is able to lead to signaling to a growing list of biochemical pathways,
some of which are listed here, which in turn lead to a variety of important cellular, physiological consequences.
This system also can lead to endocytosis of the receptors through clathrin.
So, today, we'll be talking about this new appreciation of the signaling and other functions of beta-arrestins,
and how this can be leveraged for drug discovery.
Let's start with the first, or core function, of the beta-arrestins, which I've told you about already:
their ability to sterically block G-protein signaling, thereby slowing the rate of second messenger generation.
But quite recently, we discovered, that this mechanism is actually more complex.
At the very same time that the arrestins are slowing the rate of G-protein activation,
they also serve as adapters which can recruit the very enzymes responsible for the degradation of the second messenger.
Now, in the case of the beta receptor and Gs, which leads to cyclic AMP stimulation,
that would be the cyclic AMP phosphodiesterase, shown here.
In the case of a Gq coupled receptor, like the muscarinic cholinergic receptor, which leads to diacyl glycerol formation,
that would lead to the recruitment of diacyl glycerol kinase by beta arrestin.
Now, this function essentially serves to bring these degradative enzymes into close proximity
with those microdomains of the plasma membrane where the second messenger is being generated.
So, this new information suggests that even the core function of the beta-arrestins,
the desensitization, is actually more complex than we originally perceived.
Specifically, it involves a concerted action whereby the beta arrestins both slow the rate of second messenger generation,
while enhancing the rate of degradation.
Now, the second function of the beta-arrestins
is to serve as adapters which facilitate the clathrin mediated endocytosis of the receptors in response to agonist.
This core function requires the ubiquinitation of beta-arrestins by E3 ubiquitin ligases,
MDM-2 in particular is important here.
And that leads to the interaction of beta-arrestins, as an adapter, not only with clathrin,
but with the clathrin adapter AP-2 and with a growing list of other elements of the endocytic machine.
Now, several years ago, we were carrying out a proteomics screen, which I'll tell you more about in a few minutes,
looking for molecules which bound to beta-arrestins. And in the course of doing that screen,
Kunhong Xiao in our group discovered that a variety of cellular motors bound to the beta-arrestins,
in particular, as one example, the kinesin, Kif3a.
This led to a very interesting line of research in our laboratory by Jeff Kovacs which involves Smoothened.
Now, Smoothened, as some of you may know, is a very atypical, non-canonical, if you will,
seven-transmembrane receptor which may not even couple to G-proteins at all.
Together with its inhibitory co-receptor, Patched, it mediates the important effects of the ligand Sonic hedgehog,
to lead to cellular proliferation and embryonic pattern formation.
Now, several years ago, a number of laboratories showed that in order for Smoothened to function,
it must be translocated to the primary cilium of cells, so that it can be in close proximity with the Gli family of transcription factors.
Now, since it was known that the kinesin Kif3a was necessary for primary cilia formation,
and moreover, that it was somehow, by unknown mechanisms, required for Smoothened function,
and given that we had discovered that there was an interaction between barrestins and Kif3a,
we wondered whether this interaction of barrestin and Kif3a might somehow be involved in the translocation of Smoothened,
which was required, to the cilia, in order for it to function.
And in fact, we were able to find that when the system is stimulated by Sonic hedgehog,
barrestin-2 is recruited to Smoothened, thereby bridging a ternary complex of Smoothened, barrestin and Kif3a.
The Kif3a motor then translocates the entire unit to the primary cilium,
into proximity with Gli transcription factors, which can then be activated, leading to Smoothened signaling.
And in fact, we demonstrated, as has Marc Caron's group, that in this case,
barrestin-2 is required for Smoothened signaling.
Interestingly, GRK-2 appears to be required as well.
These interesting, and somewhat unexpected results,
have raised the possibility that barrestins might function even more widely, as adapters to motors,
to facilitate the cellular trafficking of other G-protein coupled receptors, and perhaps for other cargoes as well.
Now, the most recently appreciated function of the beta-arrestins,
and one which I'll be spending the most time on, is their ability to serve as signal transduction units in their own right.
And this was first discovered about a decade ago, for the non-receptor tyrosine kinase c-Src,
but subsequently has been very widely studied for MAP kinases of various types,
AKT, and a growing list of other signaling systems.
Again, ubiquitination of barrestins appears to be necessary for these functions.
My laboratory, in recent years, has been particularly interested in learning about other pathways
which are activated through this mechanism and also, what pathways or consequences might lie downstream of activation of the pathways we already know about,
specifically the MAP kinases.
Even more remarkably, it turns out that the barrestins can serve as signaling adaptors,
not just for G-protein coupled receptors, but it is turning out for specific examples of receptor tyrosine kinases,
like the IGF receptor, cytokine receptors, ion channel receptors, such as the nicotinic acetylcholinergic receptor,
and even Notch.
Now, in the course of studying the signal transduction properties of beta arrestins,
we were led to an interesting discovery, which has turned out to have markedly facilitated this aspect of the work,
while at the same time, serving to revise in significant ways
some of our most fundamental understanding of how receptor biology works,
and as you'll see toward the end of the lecture, has actually led to a new approach to developing therapeutics.
I'm speaking here of our discovery of what we refer to as biased agonists.
Now, a biased agonist is a ligand which stabilizes a particular active conformation of a receptor,
thus stimulating some responses, but not others. Seven transmembrane receptors, for example,
can be biased toward a particular G-protein or beta arrestin. Mutated receptors can also be biased.
Now, I told you before, in the first lecture, that the classical, core concept, central dogma, if you will,
of pharmacology, is that receptors can exist in inactive 'R' and an active state, R*, and that agonists,
here displayed as 'A', stabilize or induce the activated conformation,
with all signaling consequences then being a downstream result, or a function, of the concentration of active R*.
But the existence of biased agonists, which I've just described to you, immediately says that this is too simple.
That there must be multiple active states of the receptor, at minimum, for the purposes of this discussion, two.
One of which mediates G-protein signaling and the other which would mediate barrestin signaling.
And presumably there are a significantly larger number of active states than two.
But of course we can't know how many at this point.
Let me illustrate this idea with some data about the very first biased agonist and biased receptor which we discovered.
And these both relate to the actions of angiotensin, the most potent vasoconstrictor which is made in the human body.
We had read about two reagents which seemed interesting to us based on our own previous work.
One was a mutated angiotensin octapeptide which is referred to as S-I-I angiotensin,
for the 3 mutated residues in the polypeptide chain.
The second one was a mutated receptor, angiotensin receptor, shown here,
which had mutations in two of the three highly conserved residues, which I mentioned to you in the first lecture,
the so-called 'dry', D-R-Y, motif. Two of these are mutated to alanines and this is called the DRY -AAY receptor.
Now, the common feature of the mutant angiotensin peptide and the mutant angiotensin receptor
is that they're unable to activate G-protein signaling.
But, both can lead to internalization.
These interested us, because, as I've told you,
we were aware of the fact that barrestins are able to serve as adapters, which mediate endocytosis of the receptors.
So, we obtained some of the peptide, had it synthesized,
and we obtained the receptor mutant and we looked at their ability to activate G-protein signaling.
Now, what you can see here is that, in confirmation of results which had been previously published in the literature,
neither the mutant receptor nor the mutant peptide, when applied to a wild-type receptor,
could lead to G-protein signaling, as assessed in a classical second messenger generation.
So, here, we're looking at PI turnover
and when a wild-type angiotensin receptor stimulated with wild-type angiotensin, a classical dose-response curve is observed.
But absolutely no activity, for either the mutant peptide, applied to a wild-type receptor,
or a mutant receptor stimulated with wild-type angiotensin.
We also found no calcium release, no GTP-gammaS binding to G-proteins,
so clearly, as reported, the mutant peptide and mutant receptor are devoid of G-protein stimulating activity.
However, both are very robust in their ability to recruit beta-arrestins.
Now, shown here are confocal micrographs. In this upper panel are shown HEK cells,
which express the angiotensin receptor and which are also expressing GFP beta-arrestin 2.
As you can see, the barrestin is diffusely expressed in the cytoplasm, but is excluded from the nucleus.
When such a cell is stimulated with wild-type angiotensin, twenty minutes later,
the barrestin 2-GFP can be found concentrated in these endocytic vesicles.
That beta-arrestin had first been translocated to the plasma membrane,
which we would have seen at an earlier time point,
and has now been internalized. What you can't see is that the receptors are also colocalized with it.
However, both the mutant peptide, when applied to a wild-type receptor, or the mutant receptor,
stimulated with angiotensin, also robustly recruit barrestin and lead to its colocalization with the receptors in these endocytic vesicles.
This despite their complete inability to activate G-protein signaling.
So, clearly, these mutant peptide and receptor are devoid of G-protein stimulating activity,
but do have robust ability to recruit beta-arrestin. Now, of course, the question becomes
can they use that barrestin as a means of signaling into the cell?
And to initially look at that, we chose a highly conserved signaling system
which I'm sure is familiar to most of you, namely the MAP-kinase cascades.
Now, as you'll recall, the MAP-kinase cascades are highly conserved from yeast to humans,
they consist of three kinases in sequence: a MAP kinase, a MAP kinase kinase and MAP triple kinase.
The MAP kinases in turn are in several families, such as the ERKs, the JNKs and the p38 MAP kinases.
There are at least a dozen or so kinases at every level of this cascade.
So, there's enormous complexity, many different types of G-protein coupled receptors activate these pathways,
as do all sorts of other growth factor receptors, cytokines, et cetera.
And of course, the classical pathways that have been worked out involve phosphorylation by the MAP kinases,
of transcription factors, leading to their activation, translocation to the nucleus,
and regulation of all manner of transcriptional programs, shown here.
Now, a conundrum during the '90s was, which was investigated, was the question of how,
with all this complexity, with all these welter of enzymes, dozens of enzymes,
how did the cell ever organize any specific pathway, such as, for example, RAF, MYC, ERK1,
how did it ever organize such a pathway with any fidelity or reproducibility?
This led to the idea that there might be scaffolding proteins which would bind together several members
of a particular pathway and in yeast, a molecule, sterile-5, was discovered which appeared to function in this way.
We, in fact, were able to demonstrate that beta-arrestins, in particular beta-arrestin 2,
was able to serve as a scaffold for several of these MAP kinase pathways.
For example, it was able to, it was able to scaffold a RAF-MYC-ERK pathway
and an ASK-MKK4-JNK3 pathway.
So, it was able to serve as a scaffold for such pathways.
Let me show you one of the key experiments which led to these types of discoveries.
This shows that in terms of ERK MAP kinases, angiotensin gives robust activation of ERK MAP kinases
in a cell which has been transfected to express the angiotensin receptor.
These are HEK cells.
However, the SII angiotensin, when applied to this cell, also leads to ERK MAP kinase activation,
to the tune of about fifty percent of the wild-type and the same is true for the mutated angiotensin receptor.
Now remember, in this case, there's absolutely no G-protein activation and nonetheless,
we're getting half as much ERK activation, assessed by a simple assay of phospho- or activated ERK,
we're getting half as much ERK activity as we did with the wild-type system.
Now, if we use RNAi to knock down beta-arrestin 2, we lose about half the activity of the angiotensin stimulated system,
but we completely ablate the activity of SII angiotensin or the mutated receptor,
suggesting that they are activating the ERK system exclusively via beta-arrestins.
In the case of the wild-type system, if we block the G-protein arm with a PKC inhibitor, this activity goes to zero.
So, again, even though there's no G-protein activity here, we are getting barrestin mediated activity.
So, to summarize many, many more experiments by my laboratory and other laboratories
working now with dozens of different types of G-protein coupled receptors, have led to this type of understanding.
In the case of a GQ coupled receptor, such as the angiotensin receptor,
it appears to be able to activate this specific effector system, ERK-MAP kinase,
by two distinct, parallel and independent pathways.
One is the classical G-protein pathway, leading to second messenger generation,
ERK activation and translocation to the nucleus to regulate transcriptional programs.
In the case of the barrestin pathway, in this case barrestin-2, this leads to activation of a distinct pool of ERK,
which is sequestered and restrained exclusively in the cytosol and prevented from going to the nucleus.
This undoubtedly leads to the activation of distinct cytosolic substrates with distinct cellular physiological consequences.
We've been able to demonstrate this biology, as have other laboratories now,
very analogous biology with quite a large number of other receptors.
And as I said, the mechanism appears to be a physical scaffolding of the different components of the ERK-MAP kinase.
Now, in our effort to understand the downstream consequences of this barrestin mediated signaling,
in this case through ERK, but other pathways as well, we've recently been using a variety of techniques,
some of them are shown here. We use these barrestin-biased ligands, which I've already mentioned to you,
RNA interference, knock-out cells from barrestin-1 or barrestin-2 knock-out animals,
the knock-out mice themselves and, as I will show you, a variety of proteomics techniques.
What I'd like to do is give you several examples, just examples, of novel pathways, barrestin mediated-pathways,
which we've discovered just in the past several years.
So, for example, we have found a mechanism which operates downstream of the beta-1 adrenergic receptor in cardiomyocytes,
or the angiotensin receptor in vascular smooth muscle cells.
It leads to transactivation of the EGF receptor and consequent activation of ERK,
and it goes through a barrestin mediated activation of Src,
then of metalloproteinases, which release the EGF receptor ligands, like HB-EGF.
This pathway appears to be cardio protective.
We found in vascular smooth muscle cells that through a barrestin mediated activation of ERK-MAP kinases,
angiotensin is able to lead to ERK mediated phosphorylation and activation of a cytosolic kinase called Mnk,
which phosphorylates and activates EIF-4E, a rate-limiting step of protein translation,
leading to an increased rate of protein synthesis.
We found that in vascular smooth muscle cells, again, the angiotensin receptor,
operating through barrestin-mediated signaling, is strongly anti-apoptotic.
And this anti-apoptotic activity is mediated through a concerted mechanism in which the barrestins activate ERK
and PI3-kinase-AKT, two survival pathways, both of which lead to and converge on
phosphorylation and inactivation of pro-apoptotic BAD.
Note that the ERK arm proceeds through phosphorylation of p90-RSK.
Finally, an interesting pathway, which I'll say more about in a clinical context momentarily,
operates downstream of what was previously an orphan receptor, GPR109A,
now known to be a receptor for the very important cholesterol-lowering drug niacin.
This pathway leads through beta-arrestin-1 to activation of PI3-kinase and AKT
and then on to activation of phospho
…

Need subtitles from another video?