Tip:
Highlight text to annotate it
X
Hi, Ira Mellman again from Genentech. I'd like to continue our discussion
of the cell biology of the immune response, this time turning to specifically
the problem of antigen presentation and the role of dendritic cells
in linking the two aspects of the immune response
that we've already discussed, innate immunity and adaptive immunity.
Innate immunity having been discovered by Metchnikoff
and adaptive immunity by Ehrlich all about 100 years ago.
Now in order for T-cells to do their job helping B-cells respond by
making antibodies and also by helping to kill virus infected cells and other types of pathogens,
they themselves require a significant amount of help, which is provided
by the so-called antigen presenting cells or APCs. Now recognition of T-cell
receptors of their targets does not necessarily lead to cell killing.
In fact this is an interaction that has to take place in order to generate a sufficient number
of T-cells to mediate the immune response in the first place. So the signal
that is sent by the T-cell receptor to the T-cell is a consequence
of seeing its ligand serves to activate the cell to not only to kill but also to differentiate
and divide, again depending on the type of T-cell you happen to be.
These types of signals together with other types of signals provided by other types of
membrane proteins referred to as co-stimulatory molecules,
are provided by a class of cells referred to as APCs or antigen-presenting cells.
Now as we reviewed in the last lecture, T-cell receptors see two different types
of peptides, peptides bound either to MHC class II molecules or peptides bound
to MHC class I molecules. CD8 T-cells recognizing the class I molecules,
CD4 T-cells recognizing the class II molecules. This is a structural
representation of how the recognition of the peptide MHC complexes by T-cell receptors actually occurs.
So here in the upper level you see the T-cell receptor itself,
the two major antigen recognition elements of it, the alpha chain and the beta chain,
which form a complex that is specifically adapted to recognize that a particular peptide
bound to a particular MHC molecule so here you can see the peptide,
which usually ranges in 9 to 10 or more of amino acids in length,
nestled within a very specific peptide binding cleft found in this case in a MHC class I molecule
that is expressed by the antigen presenting cell expressing in turn the antigen from
which this particular peptide is derived. In order for this system to work,
it has to be amplified, it has to be selected, the T-cells have to be amplified
and have to be selected and have to be affinity matured in such a way
that they can detect their peptides with higher and higher affinity
continuously again sculpting these T-cell receptors. Now producing
the peptide MHC complex, which is obviously key to this entire process
is a job of the antigen presenting cell. Now how does the antigen presenting cell do this?
We've already talked about the fact that there is both a class I and a class II
dependent pathway or restricted pathway of antigen presentation.
The class II pathway is predominantly adapted to presenting those molecules
derived from extracellular pathogens such as extracellular bacteria or proteins,
toxins, whatever are released from bacteria. So in those cases what happens
as we already discussed in the case of B-cells, the antigens bind to the surface
of the cell, the antigen presenting cells are taken up and deposited
within endocytic vesicles within the cytoplasm of the APC.
Here the antigens are unfolded, unwound, and eventually degraded into peptides
that are loaded also in endocytic vesicles onto the class II molecules themselves.
Now how the class II molecules get there is a rather remarkable story
and a remarkable piece of cell biology and membrane traffic in its own right.
Class II molecules are membrane proteins and like almost all other membrane proteins
begin their lives at the level of the rough endoplasmic reticulum,
where they are synthesized and inserted across the ER membrane
and assembled together with a chaperone called the invariant chain in the ER
that then after assembly takes place gets transported from
the endoplasmic reticulum into the cis-golgi, through the golgi stack, emerging at the
trans-side of the golgi complex but unlike proteins that are destined for
secretion or destined to be inserted into the plasma membrane,
MHC molecules are diverted from this constitutive secretory pathway and instead
are taken into the endocytic compartment where the invariant chain is removed
and we'll come back to that in a moment, and the class II molecule rendered accessible
to the peptides that can be derived from the incoming antigen.
Following binding of the peptide to the class II molecule, this complex is taken to the cell surface
where it can now be recognized by CD4+ T-cells.
Now the class I pathway as we've discussed services predominately those antigens
that are self-synthesized by a given cell. This of course can include
any membrane protein or cytosolic protein that is synthesized as a
consequence of the viral infection. So in the case of cytosolic proteins, these proteins are made,
ubiquitinated, degraded in the cytosol by the proteosome and peptides
that are produced by the proteosome were transported into the ER, where they are then bound
to class I molecules that are again, transported from the ER to the golgi
but now rather than going to the endocytic compartment,
instead the class I molecules go directly to the cell surface. Now all antigen presenting cells
are not created equal. There are amateurs and there are professionals.
Amateurs can make the class I response in most cases, because MHC class I molecules
are expressed by virtually all nucleated cells in the body so therefore virtually
all nucleated cells are protected by the immune system against infection
by viruses, which is a good thing. The MHC class II system on the other hand
is synthesized and expressed only by a relatively restricted number of cells
in the body. In most cases our cells that are specialized cells in the immune system
B-cells, macrophages, and most importantly these dendritic cells
that we mentioned earlier. Now dendritic cells are really special
and in fact are the professionals professional. They are the Tiger Woods
of the antigen presenting cell universe. Why? Because they are by far the most efficient.
They can capture almost meaningless and non-existent amounts of antigens
and turn those antigens into small peptides that can stimulate T-cell responses. They have
the rather unique capacity to be able to capture antigen from
wherever in the body antigen is first introduced,
be it in the skin, in the lung, in the intestine wherever the antigen is captured
and then its not simply left to passive transport through the lymphatics to go back to the lymph nodes,
but rather these cells are adapted to hone in on the lymphatics and make a bee line directly
into the lymph nodes where the dendritic cells can find a huge accumulation
of both T-lymphocytes and B-lymphocytes and help their stimulatory events take place.
Perhaps most importantly, dendritic cells are the only cells in the immune system,
the only antigen presenting cell that can actually initiate the antigen specific
immune response. In other words, prior to the advent of the first of an antigens kind
coming in before you had your very first infection with influenza
you may have T-cells that are capable of responding to influenza virus,
but they are naïve, they don't really know what to do. Only the dendritic cell can wake
them up. If you delete dendritic cells from a mouse using a variety
of genetic knockouts, you find those mice are almost completely incapable
of mounting antigen specific immune responses. Why? Because only
the dendritic cell can present an antigen at a sufficient level of efficiency and with
a sufficient amount of stimulation provided to the T-cell in order to wake the T-cell up
and get the T-cell response going. Now the other side of the coin here though
is the issue of tolerance in the sense that remember one of the key abilities
of the immune system is to be able to mount virulent cytotoxic responses
protective responses to invading pathogens, but somehow minimize
if not avoid entirely destructive components that are directly against our self antigens
in other words our own tissues and selves. Dendritic cells play also a key role
in ensuring that our immune system maintains tolerance to self antigens
and we'll come back at the very end to discuss some of the more recent ideas
on how we think that takes place. Now probably though the most important
key conceptual element of how the dendritic cell system works
and why dendritic cells play such a role that is so important in linking the innate and
adaptive response is that like cells of the innate response, innate immune response,
dendritic cells can respond to exactly the same types of microbial signals
as do the macrophages. Again, by virtue of the fact that they express
the same classes of Toll-like receptors that macrophages do, but instead of under
most circumstances emitting cytotoxic compounds as a consequence of that,
they turn the information and the advent of microbial pathogens
coming into the system into the peptide language that can be understood by T-cells,
thereby linking the activation of the adaptive response
to the activation of the innate response. So then as I'm mentioning in words,
as you can see here, dendritic cells do play this really important missing link
that intimately connects innate immunity with adaptive immunity responding
to the innate signals and turning those signals into the language of the adaptive immune response.
Now this is a concept that is relatively new and certainly not nearly as
old meaning 100 years, as the first discovery of the adaptive response
and the innate response. It took almost 80 to 100 years later to really figure this out,
and that was done by this gentleman with no beard, Ralph Steinman,
who works at the Rockefeller University who was really among the very very first
to appreciate that dendritic cells have this remarkable role in being able to have a critical element
in linking the innate and adaptive responses by being just so powerful and so special
and so adept at generating T-cell responses in response to antigens
and in response to adjuvants or microbial products. Now the basic logic of
the dendritic cell system is shown here, and this is terrifically rich and complex
but indeed quite understandable. The idea is as follows, dendritic cells exist
as immature sentinels in a variety of peripheral tissues, in fact all of our peripheral tissues
contain dendritic cells. Here we are looking at the skin, at the epidermis,
where dendritic cells are intercalated in various levels in the skin,
most interestingly in the epidermis itself where these long stellate cells
are intercalated among the far more numerous keratinocytes. Indeed, the fact that these dendritic cells
existed in the skin is actually a fairly old observation made by Paul Langerhans
also who was responsible for having first identified the islets of Langerhans in the pancreas,
but Langerhans didn't know what these cells did, but indeed he identified that they were there.
We now know that they are present in the skin for the purposes of immune surveillance.
They are there to capture incoming antigens, to capture incoming pathogens,
and after that capture takes place they migrate out from the skin, enter the lymphatics,
and eventually as I already mentioned find their way into lymphoid organs
where they also now start to intercalate together with lymphocytes T-cells and B-cells.
Now by the time they get to lymphoid organs under most circumstances,
these dendritic cells change in their characteristics, they become mature.
The difference between an immature cell and a mature cell turns out to be key
in understanding exactly what happens and we'll come to that in just a moment.
Immature dendritic cells and mature dendritic cells differ from one another
in some very very important ways. Immature dendritic cells are shown here
in an immunofluorescence picture. What you can see is that all the MHC molecules,
particularly the MHC class II molecules that are expressed by
immature dendritic cells are sequestered inside the cell in lysosomes and late endosomes.
They are not at the surface, so as a result these immature DCs are relatively speaking
incapable of stimulating T-cells. In addition they don't express co-stimulatory
molecules, they're very poor at secreting cytokines and they are
relatively non-motile, so as a result they are very poor at T-cell stimulation.
What they are good at though is antigen accumulation. So we view these
cells as the sentinels that are the first ones that encounter antigen in the periphery
and then as a consequence of detecting antigen and also having the ability
to detect the innate signals encompassed or encoded in those antigens
via Toll-like receptors and other inflammatory product receptors these
dendritic cells change their morphology and also change their function
dramatically and really can do so in a remarkably short period of time.
We find that only a relatively few hours is required to transform a cell
that looks like this to a cell that looks like this, one that extends out enormous
dendrites that give them their name, that now relocate all of the
class MHC molecules that were present in lysosomes to the surface of the cells
and also induce the expression of a variety of other important molecules
such as these co-stimulatory molecules that are necessary for optimal T-cell stimulation.
So the mature dendritic cell is the cell that is most adept at antigen presentation
and antigen stimulation. This flip from the immature to the mature state
is intimately linked to the fact that the immature dendritic cell expresses
these Toll-like receptors. Were it not for that fact, not for the fact that
these cells that are intimately associated with the adaptive response
also have the capacity to respond to the most fundamental and
elemental element of the innate response we would not have this link. So it's actually
maturation that does this. Now as cell biologists interested in membrane traffic
we've been very interested over the years as have been many other groups
in trying to understand what's responsible for this dramatic morphogenetic
change that dendritic cells exhibit and how does it relate to the function
of these cells and how to these functions relate to the overall control
of the immune response. So here on the left you are looking at a diagram
of the membrane traffic phenotype if you will of an immature dendritic cell.
These are cells that are highly endocytic, they take up lots of antigen
by a variety of different mechanisms of endocytosis, those antigens come in from
the outside, find their way to endosomes and finally to lysosomes
where quite remarkably they sit unlike most other cells that degrade
very rapidly proteins and nucleic acids and lipids and carbohydrates
that make it into lysosomes in immature dendritic cells the antigen that enters lysosomes
is protected and protected from degradation. At the same time,
these cells are making a large number of MHC molecules, particularly MHC class II molecules
which instead of being taken to the cell surface where they sit,
which is what happens in most other cells, now these molecules as well are targeted to the lysosomes
but basically nothing happens, the MHC molecules sit there, the antigen sits there,
a little bit of degradation, a little bit of loading of peptide onto the MHC molecules
but really not too much takes place until these cells are exposed to a TLR, a Toll-like receptor ligand.
In other words, one of the preserved molecular patterns that are conserved
from microbe to microbe, then everything starts to change and starts to change really fast.
One of the first things that occurs is that endocytosis, at least many forms
of endocytosis are shut off to a very large extent, this reflects
the fact that these Rho-family of GTPases particularly of the molecules
Cdc42 rather than being present in a constitutively active form
is now de-activated and present not as the GTP active form but rather as the GDP inactive form.
So antigen uptake is not cut-off completely but is diminished.
Next thing that happens is that newly synthesized class II molecules,
rather than being directed entirely to lysosomes are now taken from the golgi to endosomes
and like in other cells for other types of molecules are targeted to the cell surface.
There is a lot of reasons for this, we'll come back to one of the most important
in a minute but much of it has to do as well with changes in the metabolism
of this class II associated chaperone referred to as the invariant chain. As long
as the class II molecules associated with the invariant chain the class II molecule
is taken to lysosomes. If the invariant chain is removed, the class II molecule can
now proceed out to the cell surface. This happens for a variety of reasons,
not the least of which is due to down regulation of an antiprotease
called cystatin c, which turns off the proteolytic enzyme that is normally responsible
for degrading this invariant chain. In fact the lysosomes are activated
completely in this case for a variety of reasons, probably one of the most
interesting is the activation of lysosomal acidification. Normally
lysosomes are acidic vesicles, as Metchnikoff first told us, but in immature dendritic cells
they're less acidic than they need to be. The reason for that has been described
over the last two or three years as reflecting two key events.
One, the vacuolar proton pump or the ATPase that is required to move protons
from the cytosol into the lumen of the lysosome thereby dropping its pH,
is inactive in the immature dendritic cell. As a consequence of maturation,
as a consequence of TLR stimulation, this proton pump
is activated by turning on an assembly process, now allowing protons
to be translocated from the cytosol into the lysosomal lumen
in exchange for ATP hydrolysis thereby acidifying the interior of the lysosome.
Sebastian Amigorena in Paris has further found that in some ways like
macrophages, dendritic cells are indeed capable of generating active
oxygen species but one of the most important features here
is not so much to kill the incoming pathogen but rather to further regulate
the ability of lysosomes and incoming phagocytic vesicles to acidify their lumens,
again emphasizing how important it is to the dendritic cell
to ensure that the pH of the compartments involved in antigen presentation
and antigen processing are indeed carefully regulated.
So this is basically how it works, very simply the lysosomal pH of
immature dendritic cells is slightly acidic, it has a pH of 5.5.
The lysosomal pH found in organelles of mature dendritic cells
is more acidic by one whole pH unit, 4.5.
Doesn't sound like a lot but it turns out that most of the lysosomal
proteases and nucleic acid degrading enzymes and lipid degrading enzymes
that are found in lysosomes have a very sharp pH optimum
and they really don't work very well unless the pH that they are working in
is below pH 5, so the maturation process then drives the pH down
from a pH that's too high for optimal activity of the lysosomal proteases
to a pH that now is just right, the goldilocks effect allowing these
lysosomal proteases to do their work at optimal levels increasing
the efficiency at which peptides can be generated for association
with MHC class II molecules. Now this is a diagram just quickly as to how this works.
So here you can see a class II molecule that begins its life associated
with the invariant chain as you've seen in previous diagrams,
it consists of two membrane proteins, an alpha chain and a beta chain,
here is the invariant chain in green together with its lysosomal targeting signal.
The invariant chain is degraded by a series of proteolytic cleavages
most important of which is mediated by an enzyme called cathepsin s,
which is found most prevalently in professional antigen presenting cells such as dendritic cells.
This removes the lysosomal targeting signal from the invariant chain,
leaving a small segment of the invariant chain that is rapidly removed
by virtue of the activity of another class II associated chaperone,
not invariant chain, but something called HLA-DM,
that destabilizes the affinity of this remaining invariant chain fragment
to the peptide binding cleft of the class II molecule,
allowing the peptide to bind and displace the invariant chain derived peptide
and this then peptide MHC complex can go up to the cell surface.
I mentioned cystatin c before, it works at this stage by inhibiting
the activity of cathepsin s, it slows the cleavage, renders the cleavage less efficient
of the invariant chain making these molecules less accessible to peptide loading.
In a flow diagram in terms of what this means for membrane traffic,
here emanating from the golgi complex is the invariant chain class II complex,
enters endosomes and in immature dendritic cells nothing happens
because the level of cathepsin s is low, the activity of cathepsin s is low because
the activity of cystatin c is high and also the pH of these structures is not optimal,
and instead these class II molecules go to lysosomes. Maturation reduces
cystatin c activity, increases cathepsin s activity playing a role in helping
these class II molecules proceed on to the cell surface. Now the molecules
that had made it to lysosomes are there and also as long as the dendritic cell
remains immature not much happens. Here is a video taken by Amy Chow
in our laboratory a few years ago from a MHC molecule that has been linked
to the green fluorescent protein in immature dendritic cells, and what you
are looking at are lysosomes that are just sort of bouncing around, aimlessly
in these immature cells. So as I've mentioned earlier,
the class II molecules are there, the antigen is there and nothing is happening.
But very soon after adding LPS to this system, a ligand for a Toll-like receptor,
specifically TLR-4, you get a very different effect. Now what you can see
is that these lysosomes begin shooting out tubules and the lysosomes
start accumulating in a small dot depleting the amount of class II that is
associated with them and you can begin to see class II appearing on the surface
of the cell, all these events taking place really just over a time course of
a couple of hours. This is a video in which we are just able to visualize
the movement of some of these lysosome derived tubules from their site of
formation in lysosomes out to the periphery of cells and while I'm going
to show this to you, there are various biophysical techniques that you can use
to actually show that these tubules and vesicles derived from them will
physically fuse with the plasma membrane of the cell, delivering not only
the MHC class II molecule, but obviously also the antigen that has bound
to it as a consequence of the peptide loading event that took place
in the lysosome just at the moment of dendritic cell maturation.
All of these events take place as I said in just a couple of hours
if you wait overnight you can now see live dendritic cells that look ever much like
the static images I showed you earlier. Here a cell with all of its MHC class II molecules
on the surface, lysosomes are now stained red because none of the GFP
or green fluorescent protein coupled class II molecules are present within them anymore.
So what that means then is here one more unexpected feature of the system,
which is that MHC class II molecules can escape from lysosomes
to dendritic cell surface, by a pathway which we refer to as your retrograde transport pathway.
Anterograde transport refers to what happens when molecules
such as antigens come from the outside by endocytosis into the lysosomal compartment,
retrograde refers to what happens when they go out. Again, this is probably,
or almost certainly I should say, a microtubule driven process,
but what's most remarkable is that it happens at all. Our conventional view of
lysosomes is really kind of summarized here in this early medieval representation,
even prior to Metchnikoff, showing that there is no escape.
This is at least the conventional view. Proteins, antigens, whatever microbes are collected,
delivered into lysosomes and simply degraded. So we had not really anticipated
that there was going to be a pathway anywhere in cell biology
that was going to allow us to recover or allow a cell such as a dendritic cell
to recover molecules in a very selective and very efficient fashion
so they can move from this degradative compartment out to the cell surface.
Now when they get out to the surface, why do they stay there?
Why don't they just come back into lysosomes by endocytosis.
Well you might say endocytosis is shut off and indeed, as I already mentioned,
it is as a consequence of down regulating active forms of Rho family GTPases,
such as Cdc42 and Rac, all of which are involved in actin assembly. So the normal
ability of immature dendritic cells to capture antigen by such processes
as macropinocytosis or phagocytosis, which are both strongly actin driven
processes as we've seen in the first video, indeed are turned off. But uptake
via endocytosis by clathrin coated pits, which are much smaller vesicular carriers
that can be formed about only 0.2 microns in diameter, as opposed to the
phagosome, which can be one or two or three or even five microns in diameter,
this pathway continues. And indeed we know that MHC class II molecules
can enter cells, and indeed can enter dendritic cells by clathrin mediated
endocytosis. So why is it that class II molecules do not enter in the mature cell?
That brings us to another emerging fundamental feature of membrane traffic,
and that is the role of protein based ubiquitination in controlling the movement
of membrane proteins from one compartment to the next. Over the last several
years, a large number of investigators have demonstrated that ubiquitin plays
a critical role in signaling, either the endocytosis of endocytic receptors into
clathrin coated pits and/or the ability of those receptors at the level of endosomes
to be sequestered into these rather oddly shaped structures
called multivesicular bodies. We know from the work of Scott Ember and others
for example that when ubiquitin is modified or a membrane protein is
modified by ubiquitin, after delivery or upon delivery to the endosomal compartment,
these ubiquitinated membrane proteins are further captured
by the endosome sequestered in these little internal vesicles that then
are delivered to late endosomes and eventually to lysosomes where they can finally
be degraded. Now this happens in dendritic cells because if you look actually
by electron microscopy in a picture taken by the late Marc Pypaert is shown here
that the class II molecules are not found on the limiting membrane of lysosomes,
but rather are found associated to a first instance with these internal vesicles.
Now this was a real surprise, a double surprise because we further thought
that of course not only everything that would go into a lysosome would be
degraded, but certainly everything that was associated with the multivesicular body
would be degraded somehow dendritic cells and class II molecules have figured this out.
Now this is how the ubiquitin system works, and I'll show you the solution
to this problem at least the first solution to the problem. Ubiquitin is a small
protein that is now well known as a consequence of a series or cascade of events
to be covalently linked to a variety of different acceptors both cytosolic proteins
as well as membrane proteins by virtue of the activity of at least two large
families of so called E3 ligases that fall either into the HECT family or
the RING family, the details of this at the moment are not important,
but what is important is that both of these E3 ligases can affix ubiquitin molecules
usually the lysine acceptors on a variety of different proteins.
Now it turns out that of all the MHC class II molecules that have been
sequenced so far there's enormous diversity in terms of the sequences
that one finds in the cytoplasmic domains of these molecules
except for a single conserved lysine residue shown here at position 4 in the cytoplasmic domain
of the beta chain of the MHC class II molecule. When Jeoung-Sook Shin
who is now at UCSF as a faculty member was in my lab, she made this
realization and further surmised that gee if there is such an important
and well conserved lysine residue present in class II molecules,
perhaps class II molecules are subject to ubiquitination. And, indeed, not only
did she find that they are subject to ubiquitination, but that ubiquitination
is exquisitely well regulated. So here you are looking
at an SDS gel which was subjected to western blot antibody labeling procedure
to detect molecules of MHC class II that are either ubiquitinated or not ubiquitinated.
So here as you can see in immature dendritic cells class II molecules
on the beta chain are nicely ubiquitinated, but again, shortly after maturation
of these cells by stimulation of the Toll-like receptor system,
you now find that those ubiquitin molecules are no longer present on class II
we believe because they are no longer added but the most important
conclusion is that they are not there, and because they are not there,
they now lack the ability to enter the cell and become sequestered
in multivesicular bodies and in lysosomes. So the bottom line is that in
short dendritic cell because ubiquitination does not occur,
those MHC class II molecule peptide complexes that are recovered by retrograde transport
from the lysosomal compartment to the plasma membrane stay put
because they lack the information, namely the ubiquitin molecule
linked to the class II that happens in immature dendritic cells.
They lack this ubiquitin molecule and as a consequence these class II peptide complexes
stay where they can best serve the interest of the immune system
and be available for recognition by Cd4 positive T-cells.
And this is what that looks like. Dendritic cells shown here in red
can complex with an enormous number of T-cells because they express
at such high levels these MHC peptide complexes in the case of class II because
those complexes cannot be internalized after maturation by endocytosis.
Class I is a different story, and I'd like to turn to that now because it illustrates
another aspect of the system as why dendritic cells are indeed so special,
not only in terms of the role they play in the immune response,
but how that role is determined by alterations, in fact some rather
unexpected alterations in membrane traffic.
Now class I as I've already mentioned is a pathway that is mostly adept
to dealing with endogenous antigens. So the best example I can give you
is if you have an influenza virus infection and epithelial cells in your airway
are infected by influenza virus, those cells are going to be making
lots of proteins that are encoded by the virus. Those proteins are going to be degraded
in the cytosol, again following a ubiquitination event they'll be degraded
by the proteosome in the cytosol, small peptides generated from
those ubiquitinated proteins and then those small peptides translocated
into the lumen of the endoplasmic reticulum by virtue of the activity of a rather
remarkable ATP-driven membrane peptide transporter called TAP, actually TAP1 and TAP2.
So the peptide that enter into the ER are loaded onto class I molecules
and then they make their way out to the cell surface by the constitutive secretory pathway.
Now there's a flaw in this logic. Remember I told you that only dendritic cells
can start an immune response. And I also just told you that the predominate cell,
in fact possibly the only cell that is infected after you become infected by
influenza are epithelial cells. How do we guard ourselves against
the possibility that the dendritic cell doesn't become infected?
The epithelial cells are incapable of generating a robust T-cell response,
only the dendritic cells are capable of doing that but the dendritic cells are not infected,
they're not making the influenza virus specific proteins. So how do
dendritic cells deal with this? They deal with it by having developed
a rather remarkable system of membrane transport,
which is classically referred to as cross presentation. It was really first described
by an immunologist, Mike Bevan. Here it was thought to be the case
that antigens coming in from the outside rather than being restricted
to the MHC class II pathway can cross over and in fact have access to the class I pathway
and do so by somehow breaking out of the endosome lysosome system,
entering into the cytosol and becoming accessible to the ubiquitination proteosome
degradation system that then is also responsible for servicing
those antigens that are endogenously synthesized by the cell. So the peptides
that form from these cross presented antigens can enter into the ER lumen
via the TAP1 TAP2 translocator, be loaded onto class I molecules
and then as I've been describing, make it out to the cell surface.
So the way this probably works in practice in the case of virus infections
is diagrammed here, immature dendritic cells will capture and take up
just as another phagocytic load a virus infected cell, which has been killed
by virtue of its virus infection. So dead cells, apoptotic cells, necrotic cells
can be nicely recognized by dendritic cells, enter into phagosomes,
these cells are then degraded and then antigens derived from the infected cell,
most importantly the virus encoded antigens now come out into the cytosol
and can be degraded by the proteosome system and be presented on the surface
of the dendritic cell on class I molecules. Now remember while all this is going on,
proteins that are also intrinsic to this self of the infected cell,
in other words our own proteins, are not going to be immune to this,
they will also be degraded in the phagosome and some portion of them
we have to imagine will also come into the cytosol and be degraded
and be loaded onto class I molecules and presented to now
CD8 positive T-cells on the cell surface. So how is it then that
the dendritic cell can distinguish between the viral antigen and the self antigen?
How does the immune system do this? How does the immune system do that?
Now it may be apparent to some of you, but I actually had to go off
and think about this for a little bit, but it turns out that to understand
how the immune system balances tolerance and immunity responses
to self versus non-self actually has a lot again to do with dendritic cells,
specifically the property of dendritic cell maturation which we now should
come back to and look at in a little bit more detail to understand just how
these cells via the process of maturation controls the progress of the immune response.
Now remember that maturation links the two major arms of the immune system,
the innate immune response to the adaptive immune response
by detecting microbial pathogens and then turning those pathogens into
the peptides that are required to be presented to T-cells to generate adaptive immunity.
Now the maturation process itself in its first instance is controlled by this family
of Toll-like receptors that I've mentioned, that essentially act as barcode readers
and I'll come back to that in a moment, that identify and deconstruct exactly
what type of pathogen happens to be present at the time the maturation event is induced.
Now there are a number of different Toll-like receptors. Some of them are
diagrammed here. About 12 or more, I've lost count since it changes periodically,
and what these Toll-like receptors are for is that they are specific for a variety of
different components that one finds in a wide variety of different pathogens.
You find some of the Toll-like receptors expressed on the plasma membranes
of cells, other Toll-like receptors are expressed in endosomes and lysosomes.
They are specific for bacterial proteins, bacterial lipids, one of the proteins
is one called flagellin, which is the major component of the bacterial flagella.
LPS is a major lipid species found in cell walls of many bacteria,
but many of the intracellular Toll-like receptors in particular actually react with specifically
and identify specifically nucleic acids, both single stranded and double stranded RNA and DNA.
In all cases, the general property that is elicited is this property of maturation,
but not all forms of maturation are created equal because not all
Toll-like receptors are created equal, and in fact they transmit different types
of signals to the dendritic cell. So as I said, these receptors work together
in the sense of a bar code. Some firing, others not firing, depending upon
what pathogen happens to be present, and the dendritic cell reacts to this information
and undergoes a path of its own maturation that adapts itself specifically
to the type of immune threat that is coming in from the environment
again and identifies the nature of the pathogen, the nature of the threat based
on the combinatorial array of Toll-like receptor ligands that happen to be present
and associated with that particular bacterium. Now as a consequence
of detecting these different arrays of Toll-like receptor ligands,
the dendritic cell matures and what it does as a consequence of that is
of course is to secrete a wide variety of different cytokines, which are again
essentially immunological hormones. Now T-cells are very smart, they are
almost infinitely capable of recognizing a wide array of different types
of antigens. But they have to essentially be told what to do by the dendritic cell
and this is not only by virtue of identifying the particular peptide MHC complex
that is presented by the dendritic cell that actually gets the T-cell responses going,
but this mixture of cytokines that is released by dendritic cell specifically
and in a customized fashion depending upon the type of microbe that was detected
is that which actually determines what the overall differentiation of the T-cell is going to be.
So its not enough simply to stimulate T-cells as a consequence
of having them detect via the T-cell receptor, the peptide MHC complexes
that are formed by dendritic cells, but the dendritic cells add to that process
by transmitting their experience of what type of pathogen had come in,
and they do this in this case by secreting the characteristic cytokines.
So one example of this is certain types of Toll-like receptors, or stimulation by
Toll-like receptors will cause dendritic cells to release the very
potent immunogenetic cytokine, IL-12, or interleukin 12.
T-cells that detect antigen exposed on the surface of
a dendritic cell that is secreting interleukin 12 undergo a type of differentiation
that allows them to become a particular sub-class of T-cell called Th1 T-cells,
which are highly inflammatory and highly immunogenetic cells.
Now the way this looks actually in situ is shown nicely here. This is a
scanning electron micrograph showing a dendritic cell in the background
with a number of T-cells that are attached very closely, as I've shown you
earlier in one of the video images, T-cells move around quite a bit across
the surface of dendritic cells, but when they finally find a good match,
a peptide MHC to a given T-cell receptor, they have a tendency to stay there
and stay there for a fairly long period of time, hours if not more,
and over this period of time they are literally bathed in the cytokine mix
that is being released by the dendritic cell, instructing them to become
a wide array of T-cells that all recognize antigen but all have a very very different
functional outcome with respect to how the immune response works.
I've lifted some of the more popular T-cell types that one can find.
We won't go through them in any detail at all, but just simply to say that
there are many many different possible T-cell outcomes that have different effects
on the so CD8 killer cells, or cytotoxic T-lymphocytes,
CTLs are ones that actually kill their target, so a cell that is infected by a virus
as we've discussed before will be killed by a cytotoxic T-cell of this particular type.
Not any T-cell will do this, but in fact only these will. CD4 helper cells help
generate antibody responses by working in collaboration with B-cells.
Inflammatory cells, central memory, effector memory cells are all T-cells that
circulate retaining the information of the immune account that had occurred,
retaining the lessons learned from the dendritic cell, and finally one
that we'll come back to in just a moment is the regulatory T-cell,
which rather than promoting immunity, actually dampens immunity and
probably assists or plays a central role in assisting the process of tolerance.
Again, often these cells are generated by dendritic cells, but not always,
and I think here I'd like to turn to just this very issue of tolerance and recognition
of self or non-self. How does it work? Basically it works in two settings,
before birth and early after birth, in the organ called the thymus where all
T-cells have their origin, prior to the exposure of the developing fetus or organism
to any exogenous antigens at least under normal circumstances,
a critical process called negative selection occurs. Now during negative selection,
either dendritic cells in the thymus or a related but nevertheless different
type of cell which has the same function or is presumed to have the same function
in the thymus called thymic epithelial cells have the remarkable function
of being able to turn on at the level of transcription, the expression of a wide array
of almost all of the proteins that we know that will be expressed in differentiated cells
later in life in the pancreas, in the liver, in the kidney, all of these are
expressed by these cells early on in the thymus, creating peptide MHC complexes
that are recognized by these thymocytes or forming T-cells that are being born
and developing in the thymus at this very very early stage of life. Now what
happens at this stage though is really quite interesting
and also quite important, and indeed quite profound.
Rather than the recognition event of the cognate peptide MHC by the T-cell receptors
on these developing T-cells, rather than causing an immune response or
an unrestricted proliferation of the T-cells, instead the T-cells
are induced to undergo apoptosis and die. In immunological parlance,
these cells are referred to as deleted. So any T-cell that recognizes
its antigen in the environment of the thymus early in development
is negatively selected and removed from the repertoire of
all of the antigens that could possibly be seen by the T-cell receptor later in life.
So this is a terrifically important first pass, whereby the immune system,
thymic and epithelial cells and dendritic cells due to the special properties of the thymus,
which are still not quite understood, are capable of removing a wide array
of T-cell specificities that would otherwise recognize host or self proteins
causing auto-reactivity and auto-immunity. As powerful and as important
as this process is, its not 100% efficient. Some self antigens are missed,
but another critically important class of antigen that's missed
is environmental antigens, since after birth we are all bombarded
and in fact bathed in a wide number of environmental allergens such as pollen in the air,
food allergens of various types, things that may penetrate through the skin,
and if we were to amount an immune response to each one of these foreign
antigens we would be hyper-allergic and not be in a very good state.
Now there is no way that the thymus can educate our T-cell responses or
the dendritic cells can educate our T-cell responses in the thymus
to delete T-cells that might be specific for a pollen or environmental antigens.
That has to occur after birth because obviously as a fetus we are not generally
speaking exposed to too many environmental antigens that are allergic in this sense.
So this is a process that's been left to the formation of this final and critically
important and yet incredibly poorly understood form of T-lymphocyte
called the T reg, or regulatory T-cell. Now a lot of these are in fact formed
in the thymus, so in addition to deletion of T-cell reactivities, one also finds
that the thymus will produce an array of regulatory T-cells that recognize
a variety of self antigens that will then have a tendency to help turn off
T-cell responses that occur inappropriately later. Okay, but many regulatory T-cells,
or T regs, are not produced in the thymus but rather are produced in the periphery
as a consequence of not so much a negative selection process but a tolerogenic
process which in this case is mediated almost exclusively by the dendritic cell
not by the thymus. Now this happens under steady-state condition.
So what I mean by that is if an antigen is encountered by dendritic cells that have not
received a stimulus via a microbial type Toll-like receptor ligand to mature,
what one finds then is that all of the same processes of antigen processing
and presentation and transport of the peptide of MHC complexes
to the surface take place, dendritic cell develops to a form that is capable of efficiently
doing this and efficiently generating T-cell recognition, but nevertheless
under these conditions in the absence of a Toll-like receptor stimulus
or another inflammatory stimulus, the type of T-cell that emerges is a
regulatory T-cell, or an induced regulatory T-cell, totally as a consequence
of peripheral recognition events. So these T regs instructed again and formed
almost uniquely by peripheral dendritic cells found in our lymph nodes,
lymphoid organs and indeed all of our peripheral tissues are our last line
and in many cases our most important line of defense against mistakes
that can be made by the immune system between self and non-self,
between foreign and endogenous antigens. Again, serving to control
this balance between tolerance and immunity. Again, it is the T-cell that does it,
although to be fair we don't really understand very much about how T regs work,
this is an emerging field, an emerging problem at the moment.
But what it does seem to be quite clear based on genetic deletion results
and antibody blocking results that have been done quite recently, it is these
dendritic cells that exist under steady state non-inflammatory conditions
that really are responsible for generating these T regs.
Now this is critically important for a variety of reasons because every time a
dendritic cell matures as a consequence of being stimulated by Toll-like receptor
ligand, those dendritic cells present not only the foreign antigen
but also present every self antigen in the body that they can come in contact
with over that period of time. So every time we respond to a foreign stimulus,
we run the risk of responding to one of our own self antigens
and run the risk of developing auto-immunity. So in order to maintain
this very very careful relationship, this very very careful balance
that must occur lest auto-immune pathologies set in between immunity and tolerance,
there is this continuing production of T regs that occurs that provide with us
this important break, this important line of defense to maintain equilibrium,
maintain homeostasis, and maintain health. So the way I like to frame this
is as a hypothesis, which I stress by saying is really no more than a hypothesis
at this point, but nevertheless I think summarizes quite well the process as many
of us really believe it occurs at this point. So as I was saying,
under conditions of no infection, under the steady-state, one finds
immature dendritic cells in peripheral tissues, as shown here in the skin,
again using the same diagram we've been looking at during the course of these two lectures.
No infection, immature in the periphery, and at some point these cells
either just due to stochastic processes or due to some inductive process
migrate from the skin via the lymphatics into the lymphoid tissues.
Along the way, they undergo a type of maturation, because now in lymphoid
organs they are capable of presenting antigens, all self-antigens in this case,
or all environmental antigens, but nevertheless the type of maturation
that they undergo is one that leads to tolerance. So they are not antigen
presenting, they are not very effective antigen presenting cells as immature cells
in the periphery, they are much better at it when they are in lymphoid organs
but they are nevertheless still tolerogenic. Again, these are conditions under
steady state, meaning no infection, no inflammation. Everything changes though
when we go to the condition of infection, or inflammation. Here now
what you find is that again the dendritic cells are still immature in the periphery,
but now they encounter a Toll-like receptor stimulus as a consequence
of the advent of one or more microbes as we've been discussing. The migration
process is the same, the delivery to lymphoid organs is more or less the same,
but now the maturation that takes place is one which is not tolerogenic but rather immunogenetic.
Okay, so the T-cells that are produced by this same progenitor population of
dendritic cells, possibly, possibly there are subsets, but possibly the same
progenitor population of dendritic cells under conditions of infection,
under conditions of inflammation, yields T-cells that undergo development
not to produce T regs, but rather to produce one of the many types of
inflammatory or immunogenetic T-cells that I listed for you just a moment ago.
Now this to me comprises probably one of the most profound of all problems that
remain in the immune system. I've stated it in what may appear to some of you
at least to be relatively reasonable terms, but the fact of the matter
is we have almost no idea how these events are interconnected.
We know small details, all of which are indeed enticing, beginning with why
is it that transcriptionally one can find so many different differentiated gene
products expressed early on in the thymus? We know something about what
the transcription factors are that do this, but how all of this is regulated,
how it actually works, very very few of the details are really known,
and it really represents a terrific area for research of basic biology and also to
come up with still solutions to one of the great problems left in immunology.
There are many, but this is certainly one that tops my list. Now another reason
why this is so important is not just because of the basic biological aspect,
but also because of the disease aspect. I think increasingly as size progresses
in our understanding and our ability to do more and more complex experiments,
particularly at the systems level, an equally valid path to take in studying basic
biology is to understand disease processes. This of course has been done
by many in the past, but I think increasingly so at earlier and earlier stages
in ones scientific career and ones scientific interests, its possible to begin
to do real basic solid experiments where your question is what happens during
a particular disease process. So how does tolerance and immunity
fit into this? I've already hinted at it several times, but if you have a situation
where there is too little tolerance in other words the dendritic cells
or the thymus were not optimally efficient at deleting auto-reactive T-cells
or turning auto-reactive T-cells into T regs, or regulatory T-cells,
one can find a wide variety of diseases that fall into the broad class of auto-immune disorders,
such as auto-immune diabetes, lupus, or Myasthenia gravis. These are mediated
either by the production of pathogenic antibodies to self proteins,
or T-cells that exert direct cytotoxic effects on normal host tissues.
Another possibility is chronic inflammation, so diseases such as arthritis,
or asthma, Crohn's disease, ulcerative colitis, Multiple Sclerosis, possibly
have to do with the fact that an inflammation starts and then can't be turned off.
These may not be strictly auto-immune in many cases because for a lot of these
diseases there may not be a single antigen against which T-cells continuously
are producing new antibody via B-cell production or new cytotoxic secretions
as a consequence of the T-cells own activities, but nevertheless these
are processes that keep going because of disregulation of the balance
between tolerance and immunity. And again, in many ways one can attribute
the brute cause of all of this to dendritic cells misbehaving,
presenting antigens in the wrong context, producing the wrong type
of T-cells under a condition that doesn't call for that type of T-cell response,
and then the do-loops that emerge simply don't get turned off.
So how do we intervene in all of these things and how can we do this not only to
understand the biology, which is of course paramount, but also to understand
how therapeutically we can begin to intervene in these diseases processes
with ever higher degrees of specificity and exactness so that we can turn off
just the disease process and not interfere with normal ongoing processes
or actually do more harm than good. So now that I've moved myself from
academia to a biotech company, these are problems that are coming to the fore
on a daily basis, and its of no small matter to try and understand and grapple
with these problems, not only as a basic scientist but also as someone
who is now committed to understand how it is that you can turn that
basic science knowledge into dealing with major major health problems such as these.
It's also possible that you have too much tolerance and two sets
of very bad things can occur under these circumstances. This is different from diseases
that lead to immunodeficiency. Here you have diseases where the immune system
is often intact, but has been educated by the pathogenic organism
or as shown here by cancer cells, to evade the immune response. Cancer is I think
a particularly challenging example. Immunotherapy in cancer is something
that is now just starting to gain steam now with the first immunotherapeutic
to prostate cancer just having been approved this year, but what we
understand about cancer and the immune system tells us, at least to a first approximation,
that many cancers in fact are capable of generating immune responses,
either due to mutation or to ectopic expression of proteins that are not normally produced by a given cell.
Cancer cells in fact can elicit T-cell responses, but they've figured out,
or if they haven't figured out at least they've been selected for cells that are
capable of subverting those T-cell responses, either by turning them off,
such that when the T-cell penetrates into a tumor bed and tries to kill its target,
the target protects itself by secreting or placing on its surface molecules that in
fact will abrogate T-cell responses rendering them as immunologically, I'll just say, anergic.
Another possibility though is that cancer cells in fact by much the same way that
dendritic cells seem to do it, will seem to generate a T-regulatory or T reg responses,
again having the same effect subverting T-cell responses to cancerous cells
that would otherwise be amenable to be controlled by those T-cells, at least in theory.
Another, and I think in many ways more clear example, occurs in the case of many
chronic viral infections, such as CMV or ***. CMV is a particularly good example,
but many other chronic viruses are as well. What happens in these cases is that
viruses have figured out how to down regulate proteins on the surface of
the infected cell, or in some cases even on the surface of dendritic cells
in such a way as to prevent, again, T-cell recognition or in some cases even T-cell responses.
The immune system in a sense gets educated to understand that the viral proteins
are not actually foreign, but indeed are part of the own host protein repertoire,
and as a consequence by tricking the immune system in this way,
the virus can replicate and can maintain its infection with impunity without
risk of detection by the immune system. So in the case of diseases where
there is too much tolerance, therapeutic intervention that one can imagine,
is how is it that you reactivate T-cell responses by convincing the dendritic cells
to break tolerance as we say and re-introduce antigens either derived from cancer cells
or from viruses under conditions that now can generate positive
immunogenic immune responses rather than just tolerogenic immune responses.
So both of these types of disease states, again, embody I think some of the most exciting
biology, both immunology and cell biology that one can think of
in the immune system, plus also offer the opportunity to the lucky and interested
scientists, which hopefully includes me, to understand how it is one can
actually make either biological agents or even small molecule drugs
to either induce tolerance under conditions where you would like to turn off
chronic inflammation or turn off auto-immunity or overcome tolerance
under conditions where you would like to re-activate the immune system,
re-educating it to go about its job and combating what are effectively
foreign agents such as cancer or chronic viruses both for therapeutic benefit. Thank you.