UPMC Physician Resources
Sorry CME is not currently available for this activity.
CFTR Genetics in Pancreatitis and Cystic Fibrosis
In this presentation, Dr. David Whitcomb provides a brief overview of the clinical manifestations and biology of Cystic Fibrosis, as well as Cystic Fibrosis of the pancreas.
Upon completion of this activity, participants should be able to:
- Recognize the role of CFTR in human biology
- Recognize the effects of CFTR gene variance on human health
- Apply new advances in genetics into clinical practice
- Cohn JA, Friedman KJ, Noone PG, Knowles MR, Silverman LM, Jowell PS. Relation between mutations of the cystic fibrosis gene and idiopathic pancreatitis. N Engl J Med. 1998;339(10):653-8. PMID: 9725922
- Sharer N, Schwarz M, Malone G, Howarth A, Painter J, Super M, et al. Mutations of the cystic fibrosis gene in patients with chronic pancreatitis. New England Journal of Medicine. 1998;339(10):645-52
Dr. Whitcomb has financial interests with the following proprietary entity or entities producing health care goods or services as indicated below:
- Grant/Research Support: NIH, DOD, AbbVie, National Pancreas Foundation
- Consultant: AbbVie, Novartis
- Stockholder: SMART-MD, Ambry Genetics
The University of Pittsburgh School of Medicine is accredited by the Accreditation Council for Continuing Medical Education (ACCME) to provide continuing medical education for physicians.
The University of Pittsburgh School of Medicine designates this enduring material for a maximum of .75 AMA PRA Category 1 Credits™. Each physician should only claim credit commensurate with the extent of their participation in the activity. Other health care professionals are awarded (0.075) continuing education units (CEU) which are equivalent to .75 contact hour.
For your credit transcript, please access our website 4 weeks post-completion at http://ccehs.upmc.edu and follow the link to the Credit Transcript page. If you do not provide the last 5 digits of your SSN on the next page you will not be able to access a CME credit transcript. Providing your SSN is voluntary.
Release Date: 9/10/2015 | Last Modified On: 9/10/2015 | Expires: 9/12/2016
I want to begin this talk about 100 years ago with William Osler and he stated that if it were not for the great variability among individuals medicine may as well be a science and not an art. And what I'd like to say is that because we now are beginning to understand the reason for the variability medicine can become a science and no longer has to be an art. So I'm going to talk today a little bit about cystic fibrosis and use that as a background to talk about cystic fibrosis of the pancreas which in fact is the origin of the term cystic fibrosis, it refers to the pathology of the pancreas.
So I fished out a picture of a child recognizing that all adults originate as children and this points out our major view of cystic fibrosis as a lung disease where the airways become infected and eventually scarred leading to cystic fibrosis lung disease. It also affects the digestive tract and other organ systems, a very high rate of sinusitis and the major organ systems are shown here because the gene affects epithelial cells that line the organs of the body and also the sweat glands. By the time individuals approach adulthood there are a large number of complications that are seen in individuals, everything from constitutional growth and vitamin deficiencies to all the major organ systems that use cystic fibrosis for secretion of chloride or bicarbonate. What's interesting is that this is not a uniform process, it turns out there is high variability even among patients with the same underlying gene mutation profile.
I'm going to focus a little bit later on the talk on three organ systems, primarily the pancreas but also the sinuses and the reproductive, male reproductive organs. I'm going to give a brief review now of the CFTR biology and see what the protein actually does.
So the disease is recognized by chronic pancreatitis beginning in infancy, there is an elevated sweat chloride and progressive lung disease and sinusitis. Other features we mentioned are important: meconium ileus for us gastroenterologists, male infertility and liver disease. We recognized the mutations in CFTR discovered by positional cloning in 1989 and what you see is that it forms a chloride anion channel shown here with 12 transmembrane domains and a central channel and then regulatory elements that determine if this is open or closed and some other features as well, so it's a regulated anion channel.
The importance is that there are now recognized as some new bicarbonate - excuse me, new mutations that affect bicarbonate conductance but not chloride conductance. And the original one that was described was the R75Q that was done in a collaboration between my lab and Rafe Razell. The syndrome affects sinuses and male fertility and there is complex genotypes that seem to be associated with it, and we'll talk about that a little bit later.
So this is the functional unit of the pancreas, this is a acinus and what we have is we have an area where the acinar cells they makeup the bulk of the pancreatic parenchyma. These are polarized cells and these little gray dots represent zymogen granules that contain all the digestive enzymes needed for digesting complex proteins, carbohydrates and lipids. When they are secreted they are secreted into the lumen of the duct and then the duct cells secrete a bicarbonate rich fluid that flushes the digestive enzymes out of the pancreas and into the intestine where they become activated and begin the digestive process. The duct cells are quite interesting and what we have shown is that there are just a limited number of molecules that are necessary to secrete bicarbonate and so this can be modeled because it's relatively simple. I draw your attention here to the CFTR molecule that secretes bicarbonate and allows the bicarbonate rich fluid to flush the digestive enzymes out while maintaining a high pH which inhibits activation of the enzymes and protects the pancreas from destruction.
So this is what the CFTR molecule looks like. This was put together for me by Dr. General and Ivet Bahar in our Computational Biology Group and what we see is that it serves primarily as an anion channel, it's located on the apical membrane of the cells and you can see the location of the plasma membrane in relationship to the CFTR molecule with the regulatory elements being the big blob at the very bottom of the molecule and it regulates ion transport. What's interesting is that the molecule is actually different than many of the other molecules in the class in that it actually forms a channel so that there is a high conductance rate of bicarbonate or chloride right through the middle as shown in this illustration. I have a disclaimer in that, the actual molecule is not black and blue.
So let's talk about CFTR and the pancreas. In 1998 there were two papers that were published back to back in the New England Journal of Medicine and what they described is that CFTR variance in patients with chronic pancreatitis but without lung disease was too high to be explained by random chance. So the question we have is what is chronic pancreatitis and why are these mutations important? Now our understanding of chronic pancreatitis comes back from the last 100 years and it's based on the germ theory of disease where a single factor causes a complex disorder. And in fact all of western medicine is built around this paradigm which is critically important to understand how to fight infectious diseases, and now we have a disease of the pancreas where you have inflammation without infection. The features are that there is progressive scarring, there is maldigestion with a loss of acinar cells, there is diabetes with a loss of the islet cells, there is variable and severe pain syndromes and a high risk of pancreatic cancer. The diagnosis and treatment is based on demonstrating irreversible damage to the pancreas, so you test the patient over and over again with blood tests and with CT scans and different types of imaging techniques until you show that there is irreversible damage, you make the diagnosis and then treat symptomatically. So what we have is a hopeless irreversible condition that's expensive to diagnose and to treat.
In 1995 there was a very interesting paper that was published in the New England Journal of Medicine on medical progress in chronic pancreatitis by several of the experts from Harvard University and after carefully studying the previous 100 years of research and effort and all the different models that were made in order to understand this disease they concluded that chronic pancreatitis remains an enigmatic process of uncertain pathogenesis, unpredictable clinical course and unclear treatment.
What we see as illustrated here by hereditary pancreatitis with early age of onset is that it's a progressive disease when you begin with the first symptom of acute pancreatitis and then over a number of years you have a complete destruction of the parenchyma which we describe as chronic pancreatitis and develop malabsorption of food because of maldigestion, a diabetes begins to occur with the destruction of the islet cells and then pancreatic cancer begins to be seen. What's interesting is that not everybody gets the symptoms at the same time at the same severity at the same rate and some individuals will go 100 years with the worst known mutation for pancreatitis and never have a symptom in their entire life, at least 93 years old we published on someone who gave us their pancreas when they died of sepsis. And others start having severe symptoms at age 2 or 3 years old and their pancreas is destroyed by age 5, identical same mutation.
So what is the summary of progress? Well the disease is defined by end stage pathology, traditional methods to find the underlying causing of failed, it is not inflammation from infection, we know that, but when we do the null hypothesis significant testing of the hypothesized etiologies we find that there is no one thing that causes this. The inflammation and fibrosis and organ dysfunction is seen in other organ systems that undergo chronic inflammation and to understand this we really need to have a new paradigm. So the paradigm is gene environment interactions with progressive modeling that replaces the germ theory of disease.
So the 20th century we had the germ theory of disease that was put into place by law, so the Flexner Report published in 1910 stated that using voodoo and bloodletting, leaches, magic potions failed and that we had to use science and education and ethics to reform the medical profession. And the science at the time was the germ theory of disease that a germ causes human disease and their report stated that every state in the United States and every province in Canada should take over the - and regulate medicine and the only way a person can practice medicine is if they had proper education and the only education was in the germ theory of disease and the scientific method was based on Koch's Postulates where you have a single factor that's always there when you have the disease, the disease goes away when you get rid of the agent. You put the agent in a test model, you get the disease again, you remove it from the test model and it gets better. So there can only be one single factor and the method is called the Scientific Method, who can be against that? And by law that has to be taught in every medical school. And so a single agent causes complex disorder.
Let's fast forward to 2015 and every medical school in the United States sends their curriculum to Harrisburg, or to their state capital, ours is Harrisburg, and then elected representatives from Punxsutawney and Altoona go and they see whether or not the University of Pittsburgh's curriculum matches the Flexner Report of 1910, and if it doesn't they have to adjust it. So I think that sort of slows down progress and focuses our thinking because now we have group thinking where everybody is doing all the medicine the same way which is great for infectious diseases but it's a major problem for other diseases. So the scientific method of Koch can only allow one factor, the clinical pathologic disease definitions are textbooks or ICD9 codes or ICD10 codes, everything is based on this single theory.
The results, there's been huge progress in infectious diseases but we really don't understand complex diseases where there is two or more factors that are required to get a syndrome. It can be gene interaction or gene-gene or multiple genes but neither one of the factors is either necessary or sufficient to cause the disease, it's just there in a mixture and somehow messes up a complex process so that you get dysfunction. The problem is that the germ theory gives little guidance for managing complex disorders unless you just do it by random observation, Brownian motion and brute force.
So if we see inflammation, pain and organ dysfunction the assumption is it's a germ and the germ theory allows us to identify bacteria by doing null hypothesis significance testing under the Koch Model. And if we do not find a bacteria then we say oh gee, it must be the gene that caused the disorder characterized by inflammation, pain and organ dysfunction. When we look at the human genome finally what we see is there is a whole bunch of genetic factors and multiple etiologies that are loosely associated with the same symptom and it's actually common in the general population and it turns out that we can't interpret the data and the germ theory failed.
We also find out that we are having problems with evidence based medicine, our electronic medical records and other projects so the electronic medical record is supposed to be meaningful but it's based on an old paradigm and it tells you something about coding and billing rather than etiologies and mechanisms. The evidence based medicine is really - collapsed into the mean of the population, it requires large populations and the data is based on old studies and even older theories and it doesn't give insights or direction into the future.
Genome-wide association studies is the statistical approach to managing massive amount of data with international studies but it has given us limited progress in the clinic and now we have next generation sequencing where data from an individual person is so big and so complex that in most cases it's not helpful and it's almost un-interpretable unless you identify a new Mendelian disorder or a extremely strong factor which we can do in some cases.
So what we need is individualized medicine for complex syndromes where you have multiple etiologies with the same pathology. The same pathology has multiple outcomes. The treatment effects are unpredictable and in order to rough over this we need to focus on the mechanisms rather than mere association, it relies on modeling and simulation not disease classification and it provides guidance to individuals rather than populations.
So it turns out that the pancreas is a perfect model to begin studying the mechanism of complex disease because it's really simple, there is only two cell types and they each do one thing and we don't have to worry about the environment. And it turns out that cystic fibrosis is one of the key genetic factors that helps us unravel this entire complex model.
So what we did is put together a disease model we called the Sentinel Acute Pancreatitis Event Model and it really has three features: asymptomatic people, people with first symptom of inflammation and people who had inflammation and progressed to end stage disease. So in period A you can go for years with really bad genes and some bad habits and nothing happens, bad habits of smoking and drinking for example, and then something happens a stochastic event that initiates acute pancreatitis or injury of the pancreas. And that’s very important because what it does is it activates the stellate cells and the immune cells and so they all come into the pancreas and they set up shop and they sit there and they start doing bad things. Now acute pancreatitis lasts for about a week and afterwards there’s a big clean up process but there are some variables that are extremely important. So you see between B and C these variables are ones that determine whether or not there’s complete healing like we normally see in animal models, or if we have certain cell types malfunctioning. If the stellate cells remain active you get fibrosis. Normally we expect for them to undergo apoptosis and diminish in stop producing fibrosis, but if they don’t you get fibrosis. The acinar cell loss if you can’t save or regenerate the acinar cells then you get maldigestion. If the islet cells have dysfunctional repair or know the process going on, they’re lost. Pain syndrome if the nerves don’t respond normally to chronic inflammation you can get a chronic pain syndrome. And if you can’t repair the DNA, after all the DNA damage, you end up with pancreatic cancer. Now what’s interesting is that these complications are not surrogates of each other. We have a paper in process I think it’s published on line that looked at CAT scans and we measured the amount of fibrosis and we measured the amount of pain and tried to see if there was a correlation between pain and fibrosis could you predict how much pain there was from the CAT scan and the answer is no. There’s no correlation at all because their different cell types and there’s different susceptibility factors.
But now we can begin putting our hands around it because we can sort out each one of these in a systematic way and begin doing a reverse engineering process so that we can do forward predictive modeling.
Now does this actually work? Well it turns out that in a couple different models it’s clear that if a person has acute pancreatitis about a third of them have some change in their pancreas to make them hypersusceptible to additional episodes of acute pancreatitis, which we call recurrent acute. And people with recurrent acute appear to develop chronic pancreatitis and so this has been replicated in several studies. What’s interesting is that alcohol is a major driver of this process and to the right shows a Japanese study and it shows that if you take alcohol away the percentage of people that go from acute to recurrent acute or recurrent acute to chronic, is markedly diminished. So now we know that there are separate processes that drive the progression and are able to identify them. One of the interesting things we have a paper that’s going to be I think there will be a news release Friday we previously showed that there’s a gene on the X chromosome that is very important for progression from recurrent to acute to chronic pancreatitis because it messes up the immune response. We just found the gene that’s linked to smoking. We know now that people that smoke cigarettes are at risk of progressing if they have another gene in the background and they can’t handle the cigarette smoke, it drives the progression to chronic pancreatitis.
Now if you look at the population we see that the likelihood of having recurrent acute pancreatitis depends on what the etiology is. If you have alcohol it’s very high, if you have biliary and you take gallbladder out it’s very low, and if you have idiopathic it’s in the middle. The question is that is there a link between recurrent acute pancreatitis and chronic. Because when I went to medical school that was considered apostasy and in Germany there were people fired for even suggesting that. But what we see is that if you have acute pancreatitis your risk of getting chronic pancreatitis is increased 5 times if you have recurrence. If you have recurrence there’s a 5 times higher risk of developing chronic pancreatitis. So in fact we’re beginning to see that the epidemiology data matches our hypothesis.
Now the key gene turns out to be trypsin and Mark Lowe talked about the trypsin gene which is in the middle and if there are mutations that prematurely activated inside the pancreas prevented from being broken down or affect calcium concentrations which regulate trypsin activation and in
inhibitions or if you can’t flush the trypsin out of pancreas as shown in the bottom right the duct cell for CFTR, or if you can’t have an acute inflammatory response to upper regulate the SPINK which is the trypsin inhibitor, if you have any of those they predispose to recurrent acute pancreatitis and then to chronic pancreatitis. These are either in the acinar cell or the duct cell, either one of them puts you at risk but there’s different combinations of factors in one system versus the other system that will tip you over the edge.
So the risk of trypsin activation is going to occur in the acinus. What we normally see is excitation secretion coupling and the factors in the acinar cell that cause premature trypsin activation, a are related calcium regulation which turns trypsin on and off, and if you have high calcium it turns it on and prevent inhibition, or the number of mutations that effectively change the calcium regulatory regions of the molecules to always on. Or in the duct cell mostly CFTR and there are some other factors as well and then blocking of the duct so that you can’t get the enzymes out. These are the major things that are associated with premature trypsin activation and acute pancreatitis.
Now what we’re able to do using the SAPE hypothesis is to figure out a variety of pathways including alcohol, what are the genes that predispose to alcoholic pancreatitis and what do the ones that drive it to fibrosis. The trypsin pathways, necrosis fibrosis pathway, auto immune, ER stress, all of these are different pathways and different mechanisms and different preventative treatments that can be used to prevent the development of fibrosis.
And I’m just focusing today on the duct cell pathway because that’s where CFTR is. So CFTR is important for bicarbonate transport and so here we see the acinus with the duct cells and the acinar cell and the model here shows that there is secretion of bicarbonate through CFTR. Now CFTR is a chloride channel and so why isn’t it secreting chloride and the answer is on the base of lateral surface there are no chloride transporters, the chloride can’t get in on this side, but either chloride or bicarbonate can get out on the far side and the bicarbonate gets in because there’s a sodium bicarbonate co-transporter. So the only anion that can get in on this side is bicarbonate and the only one that can get out on this side is chloride or excuse me the ones that are available chloride or bicarbonate.
Now what this is saying is that there was a couple mutations that were described that suggested that they disrupted bicarbonate secretion but not chloride. We hypothesized that if there is a mutation that prevented bicarbonate from coming out, then you had a situation where you can’t get chloride in on one side, you can’t get bicarbonate out on the other side, and then that effect is you have no secretions. So we analyze almost 1,000 cases and 1,200 controls from our North American Pancreatitis study for 81 variants and these are the ones that were ever reported in any case study anywhere more than once. So the pre test probability that they were actually functional variants was increased since they had been previously reported and we found 43 of these as well. What we found was actually pretty interesting. We knew that the major cystic fibrosis variant would cause pancreatitis alone or in combination with one of the other factors and that’s SPINK1, N34 acid variant which prevents the activation of the trypsin inhibitor during inflammation. But what we also found is that the bicarbonate defective variants also were associated with chronic pancreatitis especially in the presence of SPINK where there was greater than 4 fold increase risk that was highly significant.
So the question is how in the world does that work? And it turns out that there’s a guy in Seoul, South Korea by the name of Ming Goo Lee and what he was doing was trying to figure out how the ion transport was regulated because there is a internal sensor to the cell for ion concentration called WNK1 and what he showed is that it appears to change the way that CFTR permeability works. And we tried to do a collaborative study with him because I said I think I know what the mutations are that are causing this. And he had spent multiple years doing gene basking experiments trying to figure out where this was and all these different hot spots. And I said I think I know what they are, I’m going to send you the 9 variants, you put them in your model and see whether or not it works. He said okay yeah we’ll be happy to collaborate, blah, blah, blah. And I kept pushing you know how are things coming. Finally, my brother who was doing some consulting work for Boeing in Incheon, South Korea, I had a chance to visit him on my way to Japan. So I called up Min Goo and I said hey I’ll be there in 3 months I’m going to come by I’d like to go over our experiment. So he dropped everything and did the experiments and got them done the day I got there and everyone was walking around stunned. Because the predictions actually worked. It turns out that normally the permeability ratio in wild type CFTR between bicarbonate and chloride is that the bicarbonate is about 20% of the permeability of bicarbonate but if you stimulate it with WNK1 it actually becomes a bicarbonate channel. However, these mutations all blocked the transition point so that it couldn’t secrete bicarbonate with the exception of the M470V that we couldn’t find any statistical evidence for but everybody was claiming was a key molecule. So the effective WNK1 activation both on permeability and conductions was completely messed up by our variants..
So then I went to Ivet Bahar and I said you know we looked at the molecule and there is no clustering. These things are like all over the place. Can you take the protein structure, figure out where the mutations are and wrap this thing up so that we have a 3-dimensional orientation and see what happens. They all came together in a very interesting way and what we found is there were 5 variants in the hinge region that effected the internal diameter of the molecule and had some other effects. There was also some that were in the middle of the channel itself so here’s an example down in panel C and D, where here’s the normal distance between these 2 amino acids and when you put a substitution in here it blocks it. So you can’t get bicarbonate through but you can still get chloride through. And so these variants told us the mechanism for disruptive bicarbonate secretion and there was actually 3 of them including an important area on the nucleotide binding domain 2 that seemed to regulate this as well.
Now another story was that once Ming Goo looked at this and worked with Ivet Bahar they looked at this vestibule that allows ions to come in and figured out for the first time how ion transporters can be selective. How can you have an ion transporter to transports chloride but not bicarbonate or bicarbonate and not chloride, or sodium but not potassium, or potassium but not sodium. And they solved the problem based on this model. They finally figured out what it is and the manuscript is in review as it solves a major problem of how cell activity works and it also has immense pharmacologic implications because now we know what the mechanism is and what to target. Gene bashing experiments failed just doing careful clinical observations and raising a hypothesis succeeded.
So the question is how about other organs that use CFTR to secrete bicarbonates? And it turns out he sinuses do because the bicarbonate is critical for hydrating mucous. When mucous is secreted from goblet cells it’s very thick and tenacious and it has to be in a bicarbonate environment to allow it to unfold and fluff up into a big fluffy slimy thing that will get the bacteria out. Otherwise it’s like glue and sticks there. So you would predict that if the bicarbonate function wasn’t working you’d be predisposed to sinusitis. And it turns out that in the male reproductive organs bicarbonate is important for the formation of the vas deferens but also for the survival of sperm. Sperm has to be in a bicarbonate environment in order to be viable on your way to fertilization. So either of those mechanisms require bicarbonates. So we looked at these two organs and what we found was that in rhinosinusitis if you have these bicarbonate defective mutations you end up with a huge risk of sinusitis, and if you are a male and you have these bicarbonate defective mutations alone or in combination with another one you P value is 1x10 to minus 7 and the odds ratio is not 3 it’s 300. That means we nailed the etiology of these patients with pancreatitis and male infertility, it is these mutations and not other mutations.
We’ve also learned that CFTR can work in combination with anatomic problems if you have pancreatic divisum and have distal resistance and CFTR mutations and can’t generate hydrostatic pressure you end up with stagnation of fluids in the pancreatic duct and you’re susceptible to pancreatitis. It’s also important observation is that smoking is also important in disrupting bicarbonate conductance but the effect is not large enough if you have 2 normal copies of CFTR, but if you have CFTR variants then it increases the risk of pancreatitis and can make you susceptible to pancreatitis. The other CTRC gene that we’re reporting later this week makes you susceptible to fibrosis. So there’s 5 major mechanisms if you have 2 severe CFTR mutations you get cystic fibrosis. If you have a severe and (inaudible) variable you get atypical cystic fibrosis with the same organs affected but with a milder effect. If you have a bicarbonate defective variant you end up with a syndrome of idiopathic chronic pancreatitis which is no longer idiopathic. CBAVD which is male infertility and sinusitis. If you have a CFTR severe or bicarbonate defective mutation and a mutation on the injury response gene SPINK1 you get isolated idiopathic chronic pancreatitis and the question is if you have CFTR and smoking or heavy drinking does that cause chronic pancreatitis.
So in conclusion chronic pancreatitis to the end stage a series of events that leads to irreversible organ damage and susceptibility to chronic pancreatitis really reflects lower thresholds of pancreatic injury and it can happen in multiple sites in multiple ways. But we can figure out what they are now we’ve got a handle on the problem. The CFTR variants are the most important risk factor of all the ones we’ve seen for the pancreatitis, affecting the pancreatic duct and the ones that affect bicarbonate secretion not only leads to recurrent acute and chronic pancreatitis but they also increase susceptibility to sinusitis and male infertility.
So this is the adult pancreatitis group that has been working to collect patients and phenotype them for 20 years and working together at every aspect of the disease in order to improve the management of the disease. And what we’ve been able to do is take a hopeless disease, become a champion and change the way that medicine in practiced. This is my collaborative group. We’ve had over 35 other medical centers that have contributed in different ways but it’s centered here at the University of Pittsburgh. Thank you.