Alcoholic Liver Disease Symptoms

>> okay. so good afternoon. i trust that the good weather will bring more folks but i'm sure you're all dedicated to the toip itopic today so we'redelighted that you're here. i would remind you later when we have questions or discussions,

just wait for a minute until you get the microphone. each week there are several hundred people at the nih who are online and they can't hear the questions, so that's why we do it that way. okay. so i won't go through the

brooklyn bridge, we're a little late and i think you've all heard the reason why this is the logo to cross science and medicine, and the topic of today is the next one, if . so although single cells have functions which are delineated in specific parts or directions

of those cells, and probably every cell in the body i guess with the exception of the red blood cell has that kind of regionalized activity which would be defined at least to biologists as polarity, physicists have a different definition of polarity, having

to do with pluses and minuses, but it's probably in the mesoso mesosoaic time when they got together to form groups of cells which became organs and then became organisms and species, and that process of cells coming together to form an organ require junctions and it

required separation of one domain of the plasma membrane from the other. and that is tied up with specific functions of a cell. and so today we're going to really have a discussion of two different systems, the nervous system and the hepatocyte system

which illustrate different aspects of this overall concept of the vital importance of polarity to maintain functional integrity and life. so our first speaker today, juan bonifacino, received his ph.d. in biochemistry at the university of buenos aires,

where he still is an honorary professor, and following a postdoc fellowship with rick klausner, he became a permanent person here at the nih, and has been here since 1981. he's the associate scientific director of the cell biology and neurobiology branch in nichd,

and juan's work is known worldwide. he is one of clearly the leading thinkers and scientists in cell biology in general, but specifically with respect to intracellular trafficking and its regulation, which is a key element in the polarization

process, as you will see if you don't know already. and in recent years, has focused on studies in the central nervous system, which he's going to speak about today. now juan goes way beyond the basic biology, which he hopes to delineate, and others follow

along, because he reaches out and collaborates with people who come forth with diseases, not the patients but the physicians, diseases in the beginning that are inheritable and then become focused into the pathway that he has defined, and then it gets to be broader when you begin to

think of acquired diseases, drugs, so forth, which are not so specific with respect to an individual gene being affected or its transcription or anything like that. that's the same thing that's going to involve later with the liver.

so juan has trained a huge number, 70 or 80 investigators here at nih who have populated the world and we're really very grateful to you for taking the time to present. >> well, thank you very much, win, for your very, very kind introduction and for inviting me

again to speak at this course. i think i spoke about two or three years ago about a different topic. and thank you all for being here this afternoon. so in the first part of this talk, i'm going to talk about mechanisms of polarized sorting

in neurons and then win will talk about the liver, i guess, or epithelial cells. and so we very nicely introduced the topic of cell polarity and he actually had an analogy of cell polarity. i have my own, which is -- it's the earth. the earth is a polarized object.

it has a north poll and it has a south pole, and both poles are very different. the north pole has the sea, the arctic sea, antithe sout and thesouth pole has a continent, antiarctic, in the north pole and the north pole and the south pole, polar bears jumping from a chunk of

ice on to the next one. and whereas in the south pole, or close to the south pole anyway, there are penguins. so both poles are very different and they have different properties. so likewise as win mentioned in our body and in nature in

general, all cells are organized, they have in many cases a north pole and a south pole, and two of the most commonly studied cell types that are polarized are epithelial cells and neurons. and so epithelial cells have an apical pole, an apical domain

and they have a basal domain. and they also have lateral domains, and for many purposes, the basal and the lateral domains are considered as a single basal lateral domain. so this is called ap cal bay says polarity. they have axons and they have

dendrites. in epithelial cells and in other cell types, particularlparticularly during development, there is an additional axis, called planar cell polarity, in which the lateral domains are different, there is an anterior and posterior domain that also have

different properties. so in neuron, although this axis is not so clearly delineated, also there is a role for components of the anterior/posterior polarity machine. now, as i said before, these dough manies ardomains aredifferent, they have

different protein and li lipid compositions, different organelles and different properties, in both epithelial cells and in neurons. now how is this polarity established and then maintained? this is a very complex process and there are many determinants

of cell polarity. first of all, there are extracellular signals that tell the cells to become polarized and how they shall become polarized. signals coming from the extracellular matrix, from other cells, or from grad gradients --many

others. so there are extracellular signals that tell the cells how to become polarized. those signals are received by surface receptors on the polarized cells for the extracellular matrix, for other cells and for the cellular

mediators. in addition, there are polarity modules or molecules that are intracellular and that mark the site where one of the poles of the cell will be established. that includes -- in epithelial cells, many others for planar cell polarity.

then there are signaling pathways that coordinate all of these processes. there is also a trafficking machinery, trafficking pathways that deliver the components to the different domain, so in the case of epithelial cells, to the apical domain or to the basal

lateral domain. and finally, the cytoskeleton also contributes to the maintenance of cell polarity by supporting the transport of those components to the different domains or the organization of the different domains, and in particular, the

micro tubule cytoskeleton and the acting cytoskeleton. now, mutations in some of these four determinants, how have they been found to underlie various diseases. including some very common diseases such as cancer, in fact, one of the hallmarks

of oncogenic transformation is the loss of the polarized organization of cells, for instance, intestinal epithelial cells, and in many cases, that linked to mutations or changes in the expression levels of various polarity determindeterminants

such at pten, p13k, scribble and many others. there are polarity determinants in other hereditary diseases including polycystic kidney disease, tube russ sclerosis in which there are benign tumors but there are also problems in axonal polarity in neurons,

neural tube defects, myoclonic epilepsy, the crash syndrome, which is a form of spastic paraplegia, and several others. so all of these are caused by mutations or defects in these polarity determinants, but for the remainder of my talk, which win told me should be about 40

minutes, otherwise he says he's going to pull a hook because he has to leave, anyway, so for the remainder of the talk, i'm going to discuss the trafficking machinery and the cytoskeleton that underlies polarized sorting in neurons, and the defects in these determinants that are

responsible for some diseases. on to neurons. as i mentioned before, neurons are polarized cells, they have dendrites and they have axons. and then driets receive the nerve impulse from other cells and axons convey the action potential that signals the other

neurons or to muscle cells in the neuromuscular junction junction. now these specialized functions of dendrites and axons are under underlie -- neurotransmitter receptors, for instance, glutamate receptors, whereas axons have neurotransmitter

transporters that package those neurotransmitters into synaptic vesicles, and they also have different types of adhesion molecules, for example -- that interact with each other to form synapsisnapsynapsis -- so onefundamental question then in cell polarity is, how are thes is thispolarized

distribution achieved, because all of these proteins are included in the dna, they are transcribed into messenger rna, they are translated in association with the -- they're all together at some point, and yet some of them accumulate in the dendrites and others

accumulate in the axon. now, the mechanisms of polarized sorting are complex and i will not be able to address the multiple factors that contribute to this polarized distribution of proteins, but one important factor is the packaging of these two types of proteins into

different transport carriers in the soma of the dendrite. although -- the soma of the neurons. although they are all made together in the endoplasmic reticulum and they all travel together, at some point these two groups of proteins are

sorted into different transport carriers. dendritic carriers and axonal carriers, so how does that sorting occur? but before i address that question, i would like to show you a little movie made by one of my postdoctoral fellows which

shows the model system that we use for many of our studies and this is the soma of the neuron and these are the dendrites. and in red is a -- receptor which as you can see is a somatodendritic -- in contrast, the adhesion molecule ngcam is an axe only protei axonalprotein labeled with

green and is for the most part not found in the soma and the dendrite, so you can see how you can visualize the polarized distribution of two types of protein in neurons, one is axonal and the other is somatodendritic. now, if we now do lie sal

imaging this, is a live neuron, and we are going to -- -- then you're going to see the vesicles entering the axon and entering the neighboring dendrite, and lu be able to appreciate now we will photobleach that, that green depose into the green, into the axon, and red goes into

red into the dendrite. for the most part. i hope you can pir receive that with this lighting. so you can see sorting now in play so emanating from the soma, some carriers go into the axon while others go into the dendrite, so sorting has

occurred in the s soma of the neuron. so how does that sorting occur? how are these two types of proteins packaged into different carriers? here i have to talk about our work because this is a question that we began addressing about

four or five years ago now. we had prophetized that this sorting likely involved sorting signals in the transmembrane proteins, so axonal proteins and dendritic proteins have intrinsic signals that tell them where to go, and those signals would be recognized by adaptor

proteins that are components of protein coats that this recognition would lead to the packaging of those proteins into either a dendritically targeted vesicle or an axonally targeted vesicle. so that was a hypothesis that we set out to test, and we did it

for many different plasma membrane proteins and i'm just going to show you one example of the work that we have done using as a model protein the copper transporting atp7b. so this is a multi-spanning membrane protein, it crosses the membrane many times and its role

is to transport copper across membranes. now this protein is expressed mainly in the liver, but it's also in the kidneys and in the brain, but not in other tissues. so it has a rather restricted pattern of expression. mutations in this copper

transporting atpas are the cause -- okay, so this is what it does. it loads copper into copper-dependent enzymes and mutations in this protein cause -- disease, which is a disease in which copper accumulates in the liver,

causing liver disease, but there are also neurological and psychiatric symptoms that could either be secondary to the liver dysfunction or they could have a primary origin because of the presence of this protein in the brain as well. so this is the protein, atp7b,

so one thing that we do in our work is we take these neurons, we transect them with a green fluorescent protein. this is what you see. these are such beautiful cells. this is a neuron, here is the soma, these are the dendrites, and here labeled with red

arrowheads, that's the axon. as you can see, this protein is mainly found in the soma and the dendrites but there is very little in the axon. it's polarized. it doesn't mean that there is nothing in the axon. there is a little bit.

but the ratio of dendrite to axon intensity is approximately tenfold. now, this protein has a -- tail, and in that tail, there are two -- that we like to refer to as the die lucent signal. and when we meu take those two -- we express a mutant

protein, this is what happens. now the protein is not only in the soma and the dendrites but now it appears in the axon, which is very long and highly branched. so i hope you can appreciate the striking contrast between these two images.

this is just, just by mutating two amino acids in a person that has more than a thousand amino acids. so this dileucine is -- when you mutate it, the protein loses its polarity and it's both in the dendrite and the axon, for these proteins to be restricted to a

a -- that's part of the signal. now, we hypothesize that that signal would be recognized by an adapter protein. and over the years, in my lab we have stud dee studied manydifferent adapter -- in particular -- we already knew that. we focused on one of these

adapter protein complexes named as ap-1 as a candidate for the recognition of the dileucine signal because it was already known to bind signals in other context. also because we have done work in epithelial cells, in sorting to the basal lateral domain, so

we suspected that these could also be involved in polarized sorting in neurons. this complex has four subunits. named gamma, beta 1 -- it's a clathrin associated protein that forms around vesicles and it localizes to the network which in neurons is found in the soma

of the cell, and that's where the sorting generally occurs. so in order to test whether ap-1 was responsible for the recognition of that dileucine signal, we used a -- negative approach, so we again used -- this is again atp7b, in the soma and the dendrites, very little

in the axon, and when we overexpress the signal 1 subunit of ap1, the normal -- the one that recognizes dileucine signals, nothing happened, atp7b was still somatodendritic, but if we overexpress the signal 1 mutant that was unable to recognize the dileucine signals,

it had a dominant negative effect and atp7b now appeared in the axon. so this then proves the interaction -- was responsible for the packaging of this protein into somatodendritic transport carriers for restriction of these proteins to

the somatodendritic domain, so ap-1, therefore, functions in the soma of the neuron to recognize the di dileucinesignals for transport to the dendrites. if you either mutate the signal or interfere with the ap-1 complex now, atp7b is no longer restricted to the

dendrites and it appears in the axon. so in addition our colleague steve taylor also found a ubiquitous form of the corporate transport atp7a also behaves in a similar matter, by virtue of the interaction between the dileucine signal and the ap --

adapter complex. and we also demonstrated the same for various other cargoes, other plasma membrane proteins that go through the dendrite, so ap1 is really a global regulator of polarized sorting in neurons, as well as in epithelial cells. its function is to make sure

that a subset of plasma membrane proteins go to one membrane, and not to the other domain, the axonal or the apical domain. so what is the relevance to disease, which i think is why many you have are attending this course, and the connections are now againin beginning to emerge,some

diseases are identified in which there is a mutation in some components of the sorting machinery. in particular, mutations in each of the three isoforms of the sigma 1 have now been shown to be the cause of a different heritable disorder.

so these diseases are known as mednik syndrome, caused by mutations of ap-1, fried-pettigrew syndrome, and pustular sor psoriasis, so each causes a different genetic disorder. now, interestingly, two of them, mednik and fried/pettigrew

present with -- and -- it's mainly a disease of the central nervous system. but in addition, two of them, mednik and postular sore eyepsoriasis, present in cells, tissues or organs that have polarized cells within them. for instance, the the inner ear

where there are hair cells, there are polarized as well, because there is deafness, and problems in the skin, and the same in pustular psoriasis. so i think the core lace between the symptoms that are seen in these ap1 deficiency disorders and the role of ap-1 in

polarized cells are really striking, and suggest that these diseases result from failure of polarized sorting of some cargo, some plasma membrane proteins into both neurons and polarized epithelial cells. i would like to tell you a little more about mednik

syndrome where that connection may be a little clearer. this is a mednik patient seen here at the clinical center in building 10 by steve kaylor, and he kindly provided me with these pictures. what you will be able to appreciate in this patient is

mainly the skin symptoms, there is rash and there is keratodermia, so the baby has a hearing aid because of the deafness. you cannot see the other symptoms. what are those other symptoms? interestingly, mednik is not the

name of the physician who described it, it's an acronym for the main symptoms, which are mental retardation, enteropathy, deafness, neuropathy, ichthyosis, and keratodermia. interestingly -- previous lab had found that in addition to these problems, these patients

also had hepatomegaly, cholestasis, low serum and csf copper, high liver copper, and other defects of liver and copper metabolism that are reminiscent mainly of wilson's disease, which as i mentioned before is due to mutations in atp7b, so here you can see the

really strong correlation between the mutation in atp7b and the adapter protein that recognizes the die lo dileucine signal -- so if you have -- or the acan'te adapter that sortsit, you have some of the symptoms similar. now these patients,

unfortunately, have a lot of other problems. those are probably due to the sorting of other proteins that are also dependent on ap-1 and we are now interested in identifying those other -- so you may be able to help these patients with some therapies.

in the case of the copper metabolism defects in these patients, they are treated with zinc acetate, which is standard therapy for wilson's disease. it decreases copper uptake in the intestine and the same therapy can be used for these patients, and actually their

liver function and copper metabolism improves quite significantly, and when it, a lot of secondary problems also improve. so if we could identify those other ap-1 dependent cargoes then there might be other things we can do for these patients.

so i'm thinking of the hook. so now, this is about an ap-1 dependent sorting. however, there is another adapter protein complex related to ap-1 known as ap-4, which has also been shown to play a role in polarized sorting in neurons, but it sorts another subset

of -- to the somatodendritic domain. it sorts a different type of glued matbleut mate receptorsdependent on ap4, and it functions similarly to ap-1 by recognizing signals in the tail of these receptors, probably at the -- for sorting to the somatodep dri

tick domainsomatodendriticdomains, result in missorting in a similar manner to the defective expression of ap-1. now, interestingly, other diseases have been associated with ap-4 defects. ap-4 has four subunits, and mutations in each of the four

subeuptsubunits of ap-4 have nowbeen shown to be the cause of hereditary spastic paraplegia, these types: 47, 50, 51 and 52. so this is a movement disorder, there is initially hypotonia, that later vel develops into hypertonia, there is also mental retardation, and mutations in

each of the four subeupts have now been found in patients and they all have the same manifestations. also -- who's here in the audience recently identified variations in the -- ap-4 in heterozygous patients that have -- so the homozygous

mutations result in this very severe spastic paraplegia and mental retardation, these heterozygous variations are associated with familial stuttering. so here again, you can see that another adapter protein complex involving polarized sorting in

neurons is defective in diseases that mainly compromise the function of the central nervous system. so we think that these diseases are also going to be defects in polarized sorting. so ap-1 and ap-4 function in sorting to the soma -- dough

maip. so so far i've been telling you about sorting into this type of carrier, the somatodendritic carrier, how about the axonal carriers? unfortunately i don't have much to offer you. this is not noan known verywell. believe it or not, some of you

may be surprised that we do not understand very well how proteins are packaged into axonal transport carriers. what are the signals? what are the adapters that recognize those signals? still not very well understood and i will leave it at that.

so i think it's a very exciting goal of future research. i would like to now, in the next five or 10 minutes, like to address a question, so now i've told you how proteins are packaged into two different types of carriers. here we understand the

mechanism, here we don't understand it very well but they are packaged into two types of vesicular transport carriers. but now there is an additional problem in polarity. how are the carriers themselves targeted to the axonal or dendritic -- why don't these go

to the axon, right? and you saw in the movie, the green went into green and red went into red. they are being sorted. how do they know where to go? that's a question that i would like to very briefly address now now.

before i address that question, i would like to pass this question, what is the boundary -- win already alluded to this in his introduction, it must be the boundary that separates the two domains. in epithelial cells, are the cell -- junction, the tight

junctions separated the apical from the -- domain, where is that boundary in the neurons? well, for plasma membrane protein, that boundary is in the so called axon initial seg iment or ais. so it's an area in the first part of the axon, and these axon

initial segment is a very specialized structure in the neuron, which has a very high concentration of voltage -- sodium potassium channels which are responsible for the generation of the action potential. so this is where the neuron

receives all the information coming from the dendrites and from the soma, and here is where the decision is made, okay, i have all this information. -- down the axon for transmission to other cells. so this is really a function -- they have an additional role.

it functions as a barrier for the lateral diffusion of the axonal plasma membrane proteins an the somatodendritic protein, they cannot diffuse to this densely packed plaster of voltage gated channels and adhesion molecules and other things, so that is a boundary

for the axonal and somatodendritic plasma membrane proteins. now, interestingly, the axon initial segment has also been proposed to function as a filter, an actin-based filter, that allo allows -- somehowallows the axonal transport carriers to

go through it and -- into the axon but blocks the transport of dendritic carriers. so we were intrigued by that mechanism and for a while we wanted to study it, what kind of a filter is it, how does it work work? but we did some experiments that

actually persuaded us that the axon initial segment was not -- for cyto -- we couldn't find the dense network of actin filaments in the -- initial segment. there was some actin but it was not what you would expect for a filament, people actually did it much better than we did it and

they demonstrated by very sophisticated techniques, there is no acting filter in the cytoplasm of the axon initial segment, there is no plug, there is no filter there. second, we observed that indeed axonal transport carriers were able to go through this region

of the neuron without any problem, dendritic carriers -- not of the axon initial segment but at an earlier region in what's called the axon -- we observe many instances and i can appreciate the importance form of cortical dysplasia with brain mall for mi malformations,there are

also -- of a kinesin-wide chain chain -- that causes another form of spastic paraplegia, and finally there's another form of spastic paraplegia and finally that there are the tao-opathies. also there is common pathology of tau in -- so i'd like to summarize what i presented, that

polarized sorting then depends on two selective processes. the first one is packaging the proteins into at least two types of vesicles that will go to the somatodendritic or axonal domains. the second process is the delivery of those carriers to

the domain, you first package and then you deliver them. so the first process is mediated by signal adapter interactions, that's how they get packaged, and the second process is mediated by interaction of those carriers with micro tubular motors and the underlying

cytoskeleton, and die fects in either the packaging or the targeting of the carriers is a relatively common cause of neurological disease, certainly imperative forms of neurological disease, but most likely also in other spontaneous diseases, that that isn't as well understood.

i also wanted to mention the fact that the most recent study suggests that the boundary for the somatodendritic and axonal domains are the plasma membrane of the axon initial segment, but in the cytoplasm, it's in this preaxonal exclusion zone and they operate by different

mechanisms w this, i thank you very much for your attention. [applause] i hope we have time for some questions if you have any. >> that was fantastic. we have time for some questions if you wish. or you want to hold on until the

discussion afterwards? >> thank you s m. my question is more on the signal transduction between the two different carriers. could you share some -- as to even for a case where it is related to a disease, is it similar to normal state or is

there a difference between signal -- >> signal transaction, you mean in terms of the sorting signals or kinases and -- >> sorting. >> before i was interested in neuron, i was a biologist, and going into neurobiology, i

figure, well, a lot of these things must be known, as we started going to the literature, we realized that a lot of processes were not well understood. the current state of under understanding is we know a lot about dendritic sorting, we know

a lot about that, but we know very little about axonal sorting signals. and that's something that we are interested in and i'm sure a lot of other people are interested in, so at present i cannot really tell you how the two mechanisms compare because we

know very little about axonal sorting. >> we do not even know whether axe onl sorting will depend on the recognition of the signal by an adapter protein that could be an entirely different mechanism which could be lipid mediated, which could be very different.

but anyway. other questions? yes. >> how does the neuron know which part will come develop -- to the axon, which part to the dendrite? >> so all those determinants that i talked about at the

beginning are probably important, so when you work with -- neurons in primary cull teu you have to isolate the hippocampus in the brain, and then you have to treat it with enzymes to separate all the neurons, then you put them in culture and they are all round.

they don't have axons, they don't have dendrites, just a round cell. and then they attach and they begin to flatten and then they emit newerrites, then one of them will become the axon. so there the decision was made, some of them will become the

axon, and very early on, you have all those polarity determinants that i was talking about, the par proteins and -- and scribble and those play a role in determining which of the neurites will become an axon. then you start polarizing the cytoscel tan.

before you have an axon initial segment or anything, the axe onl -- have redirected towards one of the neurites. then they begin directing a lot of cargoes into that newerrite that becomes an axon. so at some point in that round cell, a decision will be made in

one projection will be labeled as the axon and then that will go. >> do you think that model that you outlined for the segregation where it knows to not send certain things to the axon, do you think that's similar or related to others to cell

differentiation or like -- do you think systems might be similar? >> a lot of these railroad conserved -- i mentioned epithelial cells, other polarized cells so they are quite conserved, and this are also quite conserved in other

organism, including yeast. now -- is a little bit different from other yeasts. it relies more on the acting cytoskeleton -- rather than the micro tubules. so with certain adaptations, the mechanisms are quite similar. -- is a little bit different.

other yeasts, fungi, they use more the micro tubules -- -- a lot of these merck nisms aremechanisms are conserved. yeast has ap-1, for instance -- in the same way that our other questions? okay. thank you very much. >> the -- participate in theliver participates in

polarity -- none of them as elegantly demonstrated as you just saw. bup let's follow through here. so hepatocyte polarization has three features to it which are frankly not well recognized by people who worked in the liver field as distinguished perhaps

from the cell biology field, and that is that the mechanisms are complex, they're essential. the absence of polarity is a signal for death in the hepatocyte and they are virtually neglected in the clinical factors of liver disease and i'd like to point

out to you in the presentation why the basis of these three things. then i'd like to end with something which i will describe but you'll fail to see due to my er ror, which is the introduction of high resolution study of hepatocyte polarization

in living mice, and this is work done by -- who's sitting right over here and we'll get to it. now, for those who don't reside intellectually within the liver, i have to show this slide to point out just the ballpark. arterial and venous blood enters on the left-hand side over there

there, and then passes through the red area, which is the hepatic sinusoid, and that is bathing, for the main part, two rows of hepatocytes which are represented as tuboidal cells. this diagram, they look all alike and histologically, they look alike, but they're carrying

out very different functions and different positions. and then between, the blood is running from here to here and back up to the heart, and between these rows in the opposite direction is the biliary secretion by these hepatocytes.

so they are secrete lipids, particularly phospholipid, detergent, bile acid, which we're going to dwell on a bit, it's the major mechanism by which we eliminate cholesterol, and traditionally has been described as emulsifying fat in the intestinal tract

facilitating its absorption. but in reason years, as i'll show you, it turns out that bioacids are also signaling molecules which play a part in the polarization machinery, then that secretion into the bile is a major route for the elimination of a wide variety of

anionic and cationic molecules, drugs, metabolites, some hormones and so forth. now, you have to keep in mind that hepatocytes are very different from neurons and they're very different from is keepithelial cells such as inthe intestine because they have

unique polarization. one hepatocyte can have up to three apical domains. so here's an hepatocyte, here's one biocanaliculus, and there could be a third one. so a single cell can have three polarized domains. this is in complete comparison

to columnar epithelial, like in the intestinal tract, where these cells are lined up, there are tight junctions that separate them, but they have a brush border on a single surface that faces into the intestinal lumen. here, this biocanaliculous is

formed by the union of two cells forming a tight junction, and then delineating a small capillary-like structure, which is the bile kanaliculous. this is what it looks like in scanning electron microscopy, here would be two hepatocytes and here's this capillary ln

like structure with its micro villi and it's across that plasma membrane that all the secretion in the bile occurs, and this membrane is separated from the basal lateral by tight junctional and other complexes. these pictures are misleading. here you get one picture.

if one were to do staining of the abc transport that's found in the can lick lus, it looks sort of like chicken wire in a section, but in reality, what the canaliculous is, that is, this biliary system; a network as revealed here in a rather elegant cast that was made years

ago. and it's this network that's never been reproduced and has handicapped our knowledge about polarization mechanisms. now the loss of this network results in the inability to secrete particularly the detergent, the bile acid, and it

results in liver failure. colecholestasis, the bile is standing still predominantly in the hepatocytes as it acts as a detergent and has many detox ik effects. why has this whole thing been neglected? this is what the normal liver

looks like when it goes to a pathologist, or for that matter, even an abnormal liver. they stain it, but it doesn't visualize the bile canaliculous, you would never know it was there. frankly, i think this is the main reason why virtually no

attention has been placed to this critical structure which is essential for the secretion of many metabolites and particularly the bile acids. now the system that we use to study this is one that was developed years ago by a group in north carolina and then

discarded. these are primary cultures of rodent or human hepatocytes put in a collagen sandwich, and then they are stained, in this case, with a tight junctional marker and an intrinsic protein or transporter, the canicular membrane.

there's no cell division here, there's virtually no cell death, and what happens sequentially in a remarkably predictable manner is that this network formation occurs other a period of about six days, and this system has proved invaluable in studies of the mechanism by which this

polarization occurs. now another major discovery that put things in a different direction was the finding that there were six atp binding cassette transporters which are localized, restricted to this apical membrane. they each couple the hydrolysis

of atp to transport the different substrate. this is the one we'll talk about mainly, because b11 transports a whole variety of bioacids, the detergents, from the cell into the canaliculous. each -- this one is flipping choline, this is a sterile transporter, this is a

bunch of organic -- sulfates, so forth, and each one of these has been associated with an inheritable disease characterized by a phenotype that's a direct result of the failure of these pro teeps toproteins to be made properly -- this is what we would call functional polarity

as distinguished from one we'll talk about a little later, of structural polarity. structural meaning that basic structure of this apical domain is altered, functional meaning that there are individual components that are affected. so let me tell you a little

about this one. with the discovery of abcd11, many people theorize what would a disease look like if it were unable to secrete bile acids into the intestine, and it was all agreed that children who had this would be very sick, and indeed they are.

there's a varied phenotype that they're all characterized by retention of bile acids. and this transporter, there are now over 100 mutations of a variety of molecular types, basically either isn't made or isn't processed properly, misfolded or isn't trafficked

properly into the apical membrane, and there are mutations affecting all of these these. the net effect is severe liver damage in a child which progresses to scarring and cirrhosis, and the only cure is transplantation.

this gene is only expressed in the hepatocyte. so this is a kind of proof of principle that if the cell is unable to see crete the secretethe detergent, bad things happen. we don't know what. but there are many other situations that are not as

clearly inheritable where the same phenotype occurs and that's very common in ep willing to. epitolo im. y. sepitology. so how does it get from the transgotransgogy to the apical membrane? so we discovered that the atp

binding cassette transporters do not go by that indirect route, they actually are trafficked directly. furthermore, they exist in a large intracellular pool, a recycling endosome pool where there are 10 times as much of these receptors in this

recycling pool as there actually are in the membrane. physiologically, this movement from the pool to the membrane is stimulated by things like eating, because then bile acids return to the liver and bile acids are a stimulus for this movement from the pool, also

various peptide hormones that activate -- do the same thing. so this large pool accommodates the wave of bile acids coming back after a meal, moves to the membrane, facilitates its secretion and then it recycles back. i can't show you the movie but

what this would show you, this is abcb-11 and it is in the micro tubular organizing center and it actually moves as long tubular vesicles which proceed to specific sites in the apical membrane, and those specific sites turn out to be from studies done by several

laboratories to be largely a docking protein, so these vesicles move, they oscillate, and there are smaller vesicles that come in from the side, and all of this is micro tubular-dependent, it's completely obliterated if a micro tubule i inhibitor isadded

went. we then spent the next 10 years trying to identify players that are involved in this movement. and we used a wide variety of techniques. most of them biochemical yeast to hybrid scrien screens -- so our current understanding of the

process is ?a these transporters are made, they're glycosylated, in the golgi, they traffic along micro tubules and they require the dynamic portion of the micro tubule, the plus end, which is marked by a clip is 70 and eb10, which turned out to be porp, and important, and they enter this

large pool which i scribed to you and they move to the apical this movement is regulated by this complex of a rab protein, rab11a, and its adapt e and a myosin motor molecule which we'll talk about in a moment. now, the recycling -- is influenced by these proteins and

is a clat rin-dependent process, and there also is degradation where some of this transporter winds up where it gets degraded. now, some of the things indicated in yellow here, i'll comment about because they turned out to be very important. when we effect these various

parameters indicated in yellow, which i'll show you, the effect is not just one on trafficking, the effect is one on polarization. but this is not just functional polarization, it isn't just that the transporter doesn't get from here to here.

it's that this is converted into like a planar epithelium. there is structural change. so here's a cartoon of what is believed to be post golgi vesicles here carrying car thecart goa, they utilize this rab 11a complex which tbiendz a myosin vb motor which moves along in

this case the actin filaments. there's another adapter protein that was 2keus covered which interactsz with rab 11 and mutations cause a human disease. in studying this, we generated a rab 11a locked-gdp molecule and then transected it into cells. the question was, if you block

rab 11a, or you put a construct of the myosin motor, which is only the tail, we thought that if we expressed these things, we would just be inhibiting trafficking from the recycling thing. but what turned out was, as i anticipated a moment ago for

you, we blocked polarization, completely. so for example this is a n. a cell line that polarizes -- it's the same thing. these cells in culture grow and the increasing number of cells that polarizes get up here, and if we express these constructs,

the ip hib trthe inhibitory --remarkably enough, the cells don't die, growth continues, but they don't polarize. so here is the polarization. this is in these wifb cells, expressing either the gdp-locked rab or the myosin -- it's affecting the whole machinery of

so what sort of a queue exists between the recycling pathway and the polarization machinery? so juan commented about all of these i just mentioned, in the liver, these are the same. these are the hepatocyte components of the whole complex process.

cell adhesion molecules, cell junctions, various cytoskeletal factors, the intracellular trafficking mechanism, and it's all dependent on cellular energy. so what we have done through the years is to inhibit components of these trafficking systems,

the junction, the recycling, energy, and observed using the imaging techniques which i quickly illustrated for you as well as other techniques to show that not only is abcb11 trafficking impaired by these maneuvers, but the cells don't polarize or if they're

introduced when they have been polarized in the cultured system, they depolarize, so these are required for creation and maintenance. so for example, inhibitory inhibition of micro tubules by a variety of means including an adapter protein, ridixon, which

was done in a knockout system, we've done studies with a par 1 protein involved in junctional formation, her knockout and silencing rna and i told you about the -- and we're going to talk for a minute about the role of energy in this entire now all of these things, these

manipulation, led to the failure to polarize, the inability to traffic, and when done in a polarized cell, rapidly to depolarize. now, here's a case history which was presented to me actually when i was visiting in holland a few years ago.

this is an 8-year-old school girl, she had a normal birth, was 6 pounds, but she probably began to lose weight, was dehydrated, had diarrhea on ingestion of any kind of foods, including milk, she was a medical emergency. and she was treated by putting

an indwelling line and maintained with parenteral nutrition which was sustained for a period of four years. at about one month of age, her jaundice, which was present at birth, became much more severe, and she had increasing liver damage over the next three

years. an intestinal biopsy, which i'll show you an example of next, was performed, which provided the diagnosis. at age 4, she had a successful transplantation of the entire small intestine and liver. and has on immunosuppressive

drugs ever since. by age 5 or 6, she had regained more or less normal growth and nutrition. i couldn't believe it when they presented this as a case history, and in walked this 8-year-old girl. she had a disease called micro

vi lus inclusion body disease, in which this is the small intestine from a biopsy. now normally the small intestine has -- the surface of it sort of looks like a picket fence, but all of the surface micro villi are gone. they don't appear.

biologists were imprissed by this inclusion which looks a little bit like the micro villi of the surface, and that's why they call it inclusion disease but it still isn't clear whether this represents a brush border that is inappropriately made,

trafficked, or whether it's more readily re-absorbed but it hasn't been studied. but the net effect is that this young lady and about 200 others lack the brush border on the surface. the polarity is lost. this is converted into a planar

cell, still having tight junctions, but it can't function in a polarized way. now this is recessively inherited and a similar process occurs in the liver, but we don't have as much detail about it as we did. now, this is due to a series of

mutations, there are 16 described so far, in the myosin 5b gene first described in this article in "nature genetics." similarly, this has been observed in mice where this has been knocked out. so here we see evidence of a profound human disorder

affecting liver and kidney due to a disruption of the polarity targeting machinery. now let's turn to the fact that this process and all of the processes that juan showed you basically require cellular now, the two key players that we have been working with are these

two kinases, lkb1 was initially described as a tumor suppressor, it's mutated in the human cancer disease called the puts-jagar syndrome. -- requires the presence of a scaffolding and an adapter protein. compactly what the agonists are

that lead to this phosphorylation, growth factors or anything else, is yet to be really defined. one of the major downstream targets of lkb1 is another unique kinase, ampk. amp-activated pro teeb kinase,protein kinase, there are 14 members of the

family but there's only one which is primarily related to the polarization machinery. it's activated by lkb1 but also by a rise of anp in the cell. so it's thought about as the metabolic sensor. when atb has been consumed and the ratio goes up, that adds the

phosphorylation as well. so the relationship of amp kinase to cellular metabolism has been studied extensively. but it wasn't until about 2004, i guess was the first one, in which it was shown in different cell systems that the ampk and lkb1 is required for polarity.

if i use this to summarize a little background of what amk and lkb1, the activation is regulated in part by bile -- acting through a cyclic amp mechanism, but that's not the only way in which lkb1 is activated and as i pointed out, the real agonist for it have not

been defined. phosphorylation of ampk by amp or activation of lkb1 leads to a profound suppression of all systems that utilize atp within the cell. that is lipid synthesis is inhibited, protein synthesis -- this is the base

>> speaker2: by the way, of the relationship of ampk to the autophagy pathways and to m tour and its relationship to cancer. now the one system that is activated is cellular ka tab lism and we find what's activated is cellular so ampk activation leads to

enhancement of trafficking, tight junction formation and cytoskeletal organization. while doing these studies in the cell cultures, in jennifer's lab where many people were working on mitochondria, we were asked what do the mitochondria look like?

well these cells are isolated on day one, mitochondria, stained here, are all very small. there's little evidence of their degradation my to have gee, but they are all very, very small, and here it's quantified. but then very rapidly over a period of a couple of days, the

mitochondria undergo fusion and they become like giant mitochondria. extending throughout the cell. now vision and fusion are regulated by complex of proteins, some of which are indicated here. fusion is accentuated by

expression of mfn1 and 2, which are outer mitochondrial membrane proteins, and also by opa1, which is intermitochondrial. the vision machinery is largely driven by a complex which is getting more complicated every day involving the protein brp1, which accentuates this part of

the process and tends to lead to the breakdown of mitochondria in that my to have gee of turnover and that activation of biogenesis and also of stimulation to form these long elongated mitochondria. so it was discovered that the process of change from

mitochondrial vision breaking down to fusion was associated with activation of mfn1 and opa, no change in structural proteins indicating this was not a generalized effect, and little demonstrabl alteration in the btn vision process. so something is happening to

activate as a result of the stress of cellular isolation to activate these cells. now what happens when the mitochondria get confused? where their potential is driven upward, their atp synthesis, which is very low, is restored and they're much more efficient

in making atp when they're fused and when they're not, and interestingly enough, the initial atp that's made for the first three days is derived from oxidative phosphorylation as shown here by the fact that that atp is inhibited by -- where it is only after day three where

glycolysis kicks in and we see -- what does this mean? what it suggests that after stre the mitochondria are responding preferentially by fusing using oxidative generating atp by ox daystive phosphorylation and that this is an initial stress response.

well, how does ampk affect mitochondria? the traditional and classic studies by hardy showed that it acts indirectly by increasing glucose uptake affecting glucose transporter and enhancing glycolysis. as a product of metabolism, nad

activates -- leading to mitochondrial biogenesis. but more recently, direct effects on mitochondria were shown by bruce spiegelman who showed that -- which leads to new mitochondria, and more recently, ruben shaw showed that ampk can directly interact with

the mitochondrial -- the conclusion is that ampk is a major regulator of mitochondrial dynamics and metabolism. now the bile acid physiologically plays a part in facilitating the polarized phenotype. but whenever bile acid

accumulates within the cell, it leads to this mitochondrial stress reaction that i mentioned and it leads to fragmentation, reduced beta oxidation, reactive oxygen and ultimately other organelle toxicity. the important point is that bile acid accumulation is the common

pathway for intracellular injury and that leads to loss of polarity and loss of polarity is the beginning of cell death. now this schema has arrows but the arrows should have question marks because we don't know all of the processes that are involved.

but it's clear that bile acids have some regulatory role in polarization that, the two kinases are critical for atp generation and maintenance, and that they are acting directly and indirectly through mitochondrial vision, fusion and other events.

but whether ampk is only acting in this way or whether it cannot be acting on downstream targets in the polarization machinery itself is yet to be worked out. many of the components of the polarization machinery had ampk consensus sites. but the only one where

functional polarization has shown to be important was -- must be phosphorylated in order to maintain its dijunctional formation. now there are a variety of inherited hepatocyte polarity the ones on your left are the ones i referred to before as

functional. so these are the transporter in which the defect resides either in transdescription in misfolding or in some instances in altered traffic. on the right are a series of disorders which mutations exist in the structural machinery, and

i showed you one, myosin 5b. i'll show you another now in the bpar vps33b complex which intd acts with this. there are disorders in which tight junctional protein 2 and claudin are mutated with having clinical consequences in the production of liver disease

particularly in children, and there is also a recently described mutation in -- 3, the docking protein, which has a phenotype virtually identical to that seen in the microvillar inclusion body disease. so mutations in vpar cause this there are contractures of the

joint, there's kidney dysfunction, and there's impaired biliary secretion and colecholestasis, add one initial finding was that the bile acid transporter, instead of normally being seen in this chicken wire-like localization is actually in the bay sew lateralbasal lateral

it's misfolded, misdirected. here's a cartoon that indicates the summation of all of the studies that were done. so under normal conditions, the abc transporter on the apical membrane goes through the apical cycling pool, and a portion of it goes through the endosomol

pool and is efficientleventually degraded. this has been done experimentally in mice and there are eight different mutations that have been described in children so far. what happens is, transporters do not remaip on the surface of the apical membrane for a sufficient

period of time. they are trafficked normally but they are misdirected to the basal lateral domain and a portion go into an intracellular domain that we have yet to identify. now vpar interacts with the b complex.

when it's effective, we get mislocalization of the transporters, and an accelerated loss from the apical membrane. we propose that vpar regulates the rate of recycling. and degradation of memberring proteins and studies are going on regarding its role in

intracellular trafficking. now these are rare inheritable diseases, but in the same way as one demonstrated the studies of the rare inheritable neurologic diseases provide key information as to the function of normal pathways, the same thing is true in the liver but we have a much

longer way to go. on the other hand, the real interest down the line is what happens in acquired disease, which is much more complex? now one example of this are drugs. there are a variety of drugs that are toxic to the liver.

some more severe than others. and a great deal of interest in trying to delineate toxicity before it is given through patients and sold commercially. attention has become focused on the mitochondria in most of these processes for some of the reasons that i showed you a

moment ago. and recently, a colleague in australia has proposed and done some very interesting studies with several hepatotoxic drugs which lead to empairmt of fusion rapidly to mitochondrial figures to autophagy, mitochondrial damage and so forth, and if one

restores the ampk to the system, it enhances biogenesis fusion and prevents and to some extent reverses the toxicity with some well, this is the beginning of a series of studies to relate energy metabolism, polarity, mitochondrial biology, if you will, and drug toxicity.

so the same thing applies with acquired diseases like viral hepatitis. hepatitis c requires attachment to claud1 for entry into the cell. what is the role of either mutations or of toxicity in that process in the disease.

and there are many other examples. now i'll conclude with these studies of what happens now if you have a selective liver specific lkd1 deficiency. so this was generated in our lab a couple of years ago. this is an albumen cre knockout,

and the cre is expressed at about day 6 or 7, at which time then the lkv1 knockout is expressed. and so the animals from that point on, they lose weight, they exhibit a pretty severe phenotype, they're jaundiced, and they die within 28 to 30

days. now the interesting thing is the liver does not show signs of inflammation or necrosis, so this is a functional disorder. michael jarnir has been doing electron microscopic studies of these animals for a couple of years now, and came up with a

fascinating observation. here is the control at birth, the biocanaliculous is discernible, and that's within the normal range here at 4 weeks, these are micro villi, these are tight junctions. this is just exaggerated from here.

in contrast, in the first week, looks perfectly normal because the transgenic has not been but by four weeks, a remarkable finding with some consistency, there are markedly reduced number of to begin with and when they're present, they tend to show the fact they are not

restricted just to this dough they actually extend around the lateral domain in areas where you don't see much if any tight and this is a magnification disk. that led us to wonder whether the role of the tight junction in the phenotype, which then

leads us to -- and regret pli i havregrettably ihave to describethe movies, so what natalie works with is live imaging in 10-gram or even sometimes smaller mice, and these animals then, their liver is placed on a -- is open, the abdomen is open, they're anet anesthetized, and then images

are combined and processed. and so here is an example of when the wild type animal is injected, where they're both injected, you're not going to see it, but -- which normally is traveled through the blood, taken up into the liver, to form carboxy fluorescein, and that's

a substrate for an abc transporter, abc-c2 which pumps it out into the biocanaliculous. what this movie would show, i'll replace it for other purposes, is that in the area between the cell, would you see this fine almost herring bone distribution of the canaliculous.

this is interesting despite the absence of the movie because in the lkb-one knockout, you see evidence is the -- and this is fluorescein, it still remains within hepatocytes and a very small amount ever gains access to the canaliculous. furthermore was observed that

much of the color in the vascular space was not just red or green but they're actually large areas of yellow, wondering how does this fluorescein get back out into the cell, out of the cell into the blood? the two transporters on the basal lateral membrane are not

overexpressed in the cell so we thought it may be related to what happens at the tight so here are components of the tight junction which collectively function to link the actin cytoskeleton with fully impermeable barrier which permits the formation of the

apical domain. here shows the fragmentation, this is the bile acid transport are stain, a lot of it's intracellular and it's very fragmented and that's sort of matched by -- shows a similar pattern, in other words, the wie owe canaliculous is grossly

distorted and the tight junctions are likewise highly fragmented. singulin -- is totally disrupted. it's interesting that it has an -- consensus site and that might be the mechanism by which this is happening.

regrettably once again -- this one works? how do i get it to work? oh, there we go. so here's the wild type, this is the knockout. what you see here is dextrand, this is the vascular space, and the black area is where the

hepatocytes would be, and between that is where the biocanaliculous is but you don't see anything. so here you see the vascular space and you see this. so what is this? this has gone from the blood directly into the biliary

this is the first demonstration to my knowledge ever at the cellular level in an impact organ of the existence of the paracellular pathway of the way in which material gets from the broocanaliculous back into the bloodstream. this has been proposed to be an

extremely important mechanism in various acquired diseases where the biocanaliculous can be so here are the key points which i've tried to present. from the same point of structural polarity, each he hepatocyte contains apical membrane domain which is the one

or more sometimes up to three. now that imposes a whole new dmengs of understanding intracellular events when you have three apical domains within a single sell. they're functionally sealed by tight junctions and they form a very complex interconnected

network they're normally only about 10% of the surface area of the hepatocyte but it's greatly increased by the presence of micro villi op the surface and we don't understand the dynamic nature of those micro villi. now functional polarity is maintained by the polarized flow

of molecules across the can lick l --things are coming into the liver, going out of the liver from the blood side and selectively going out of the liver on the biliary side. when the bile components, particularly the bile acids are retained either due to some

genetic disease or potentially an acquired disease, that leads to cellular damage, to my toe mitochondrial damage and to structural depolarization. these are elements of the polarization system which i won't take the time to describe as we mentioned them before a

couple of times. i would dwell on the fact that mitochondrial -- is essential for polarization and any disorder, particularly an acquired disorder which pro foundly alters energy metabolism has its downstream consequences of depolarization, and that

process is highly regulated by ampk. so there are many more questions than we have answered, and the state of the art and the liver is far afield from where it is in the neuron, but we hope that the twain will meet, and that's the purpose of sessions like

this and this course. so thank you very much. we have time for a few questions if anyone wishes. >> [inaudible] >> we don't know as much about the liver from a morphological standpoint because these are taken from children at a time

where there's no inflammatory pathology so you can see what's going on and they usually don't do a biopsy of the liver. in the knockout mouse that has been generated, there are both liver and gut alteration. the syntaxin3 is interesting because that's strictly a liver

problem. okay? >> that's not quite true, is it? i mean, hepatitis b enters hepatosites. you're the expert. >> juan, do you want to comment a little bit about how viruses get into cells?

i mean, most of the work with viral entry has been with things like vsv and others, and so it's indirect. but i mean, it's quite clear that occludin and clathrin 1 are docking sites for hepatitis c now, how that happens and how it -- and what the consequences

r i don't know. but we have some data which lead to interesting hypothesis, a working model, is that when cells are chronically infected like with hepatitis b, and they're loaded with surface antigen, many of those cells are depolarized.

and so far, depolarization is not a healthy state of affairs. they don't seem to re-polarize, they presumably are eventually going to be replaced. so thank you all very, very much.