You might not stop to think about it much, especially with all the talk of coronavirus transmission, but it’s pretty amazing that people don’t get more respiratory infections. We’re constantly breathing in microbes including viruses, bacteria, and fungi. Most of these are (at least usually) harmless, but some have the potential to cause disease. Yet, thanks to our body’s initial lines of defense, they don’t get the chance to. No, I’m not talking about all those antibodies you might be hearing a lot about lately. Antibodies (little proteins that specifically “recognize” parts of invaders) are only made later, as part of the adaptive immune response. That takes time and your body needs to act fast. So, before the microbes have a chance to do damage, the lining of your respiratory tract (nose, trachea, lungs) traps the buggers in a coat of mucus, and the cells lining the tract beat their hair-like cilia to sweep that mucus away. 

Mucus is made up of proteins called mucins which are glycoproteins, meaning that they are proteins coated with sugar chains (glycans). Sugar and water like each other, so mucins secreted by epithelial cells (cells lining body surfaces) are able to trap water to form a gel-like substance that serves as a barrier between the air we breathe in and our cells. The mucus traps the wanna-be invaders and then, when we swallow, cough, or nose-blow out that mucus we get rid of it (in case you’re wondering, the harsh environment of the stomach will destroy the stuff in it). So, although you might think of mucus as “gross,” it’s actually kinda a biological superhero. 

But in order to have superpower, it needs to be fairly fluid so that the cilia can sweep it away. And, for patients with the disease cystic fibrosis (CF), their mucus is really thick and sticky, so the cilia get kinda “stuck” in a mat of it. Microbes keep getting trapped, but they can’t get removed, so they’re often able to take root in the lungs, leading to frequent, sometimes life-threatening, infections which often require antibiotics and/or hospitalization. And in a cruel irony, the immune system’s attack on these invaders can end up causing permanent damage to the lung tissue. 

Mucus isn’t only important in your respiratory tract, it’s also important in your digestive system, so patients with CF have digestive problems as well. In particular, their thick mucus can clog up the ducts through which the pancreas excretes enzymes which help break down food. This can cause structural damage to the pancreas including formation of fibrous cysts, which is where the name comes from. So patients with CF often have difficulty gaining weight and pancreatic enzymes are often taken as part of a therapeutic regimen. Just a small part of that regimen, I should add. CF typically requires a LOT of intervention – dozens of pills and nebulizer treatments, physical therapy sessions in which therapists (or friends, parents, or special electric vests) pound strategically on the back to break up mucus so it can be coughed up, frequent clinic and/or hospital visits, and, if lung capacity gets too low, double lung transplant. New, targeted therapies can be game-changers for some CF patients, but they’re only possible thanks to the tireless work of researchers to discover the root cause of CF and search for treatment strategies which aim at that root. 

I want to tell you a bit about that work but first, a little backstory. I went to college with Tiffany Rich, a woman with CF who, several years ago (after having to take an oxygen tank to our graduation ceremony), had a lifesaving double-lung transplant. She has documented her journey with CF and raised money to help support patients and families with the disease. I highly recommend checking out her account @tiffrich22 and @saltycysters. Here’s more about her story: 

It is partly thanks to Tiffany that I have developed an interest in CF research, which led me to ask for the book “Breath from Salt” by Bijal P. Trivedi for Hanukkah – it was the first book I’d read cover-to-cover in a couple years (PhD life has kept me pretty busy!) but its compelling chronicling of the history of CF research was fascinating. One thing it emphasized was that CF research has in large part been made possible by the work of many many many people – from the patients and their families organizing together to raise funds and awareness to the scientists around the world working together to find solutions. I am going to focus on just some of the scientists the book talks about, but know that there are many more and I don’t mean to discount their roles in any way but rather to draw attention to a couple of the lesser-known heroines of the story. Much of my information comes from that book and I will also link to some more articles at the end, but this one is a great overview from Science announcing the discovery, and I include some graphics from it in the figures: 

CF has been around for a long time – you can find references to case stories of likely CF patients going back centuries and the first medical report of CF was written in 1938 by Dorothy Anderson. But its genetic cause, mutations in a gene known as CFTR (cystic fibrosis transmembrane conductance regulator), wasn’t discovered until 1989. That groundbreaking discovery was through a collaborative effort between groups of researchers at the Hospital for Sick Children (aka SickKids – seriously) in Toronto, Canada led by Lap-Chee Tsui and at the University of Michigan, led by Francis Collins (that name might sound familiar because he went on to become director of the U.S. National Institutes of Health (NIH)). Tsui’s team included two talented post-doctoral researchers (scientists who had recently earned their PhDs and were looking for further training) whom I created Wikipedia articles for and will tell you more about in a follow-up post – Johanna Rommens and Batsheva Kerem. Tsui himself was recruited to Toronto by geneticist Manuel Buchwald, who was collaborating with biochemist John (aka Jack) Riordan (biochemistry – yay!). Buchwald and Riordan would be involved in the later work as well. Much of the grunt work from Collins’ lab was carried out by grad student Mitchell Drumm. And they all were funded largely by grants from patient advocate groups, so lots of thanks to go around!

Even before the gene was discovered, there were hints as to what the problem was. One strange clue was that the sweat of untreated patients with CF is salty, as reported by Paul di Sant’Agnese in 1953. Scientifically-speaking, a “salt” is a neutral (chargeless) combo of a positively-charged molecule (a cation) partnered with a negatively-charged molecule (an anion). “Table salt” for example is sodium chloride (NaCl), which consists of sodium ions (Na⁺) counterbalanced by chloride ions (Cl⁻). In addition to on your table, this salt is in your sweat – even if you’re healthy. But CF patients have way too much of it in their sweat and this is a problem because salt is attractive to water, so water follows the salt. When the equivalent situation occurs in your respiratory tract, water gets pulled out of the mucus, making it gunkier and causing it to plug up airways and ducts. (note: this isn’t the whole story of why the mucus is gunky, and recent research has pointed to a bigger role for bicarbonate as I will discuss later).

Since opposites attract, oppositely-charged ions like to follow one another. But was sodium following chloride out of cells and into the secretions? or was chloride following sodium? Was the problem too many ions getting pumped out or ions getting stuck outside, unable to get in? By conducting experiments with cells in dishes and electrodes in nostrils, scientists were able to figure out that the problem was with chloride getting out of cells. Chloride ions were getting stuck inside of cells, leading sodium ions to rush in, dragging water along with them. Thus the salty sweat and dry mucus. This all made a lot more sense once they figured out that the mutation was in an anion channel protein that works to conduct chloride and bicarbonate (HCO₃⁻), so let’s talk about how they found that mutation. Well, mutationS… plural… there have been hundreds of different mutations in that channel discovered as I will get into (and which make treatment strategies trickier for some).

These were the days before high-throughput DNA sequencing, so the scientists searching for the genetic cause of CF had to resort to “old-school” molecular biology techniques. I therefore need to step back for a little molecular biology primer: the instructions for proteins are written in sequences of DNA called genes. These genes are present in really really long pieces of coiled-up DNA called chromosomes. You can think of genes as recipes and chromosomes like cookbooks. You get one copy of each somatic (non-sex) chromosome from each biological parent and these chromosomes are kept locked up in a membranous compartment in the cell called the nucleus

The nucleus is a bit like the reference section of a library – you can make copies of the stuff in there, but you can’t take the originals outside. So, when a cell wants to make a protein, it first makes RNA copies of the genetic recipe and “edits” those copies in a process called “splicing” to remove regulatory regions called introns (which are a bit like extra pages in between the instruction pages). With a little more processing (involving the addition of a molecular “cap” and “tail”) you get mature messenger RNA (mRNA), which is shipped out of the nucleus and into the general interior of the cell (the cytoplasm) where protein-making complexes called ribosomes are standing by to read the instructions and piece together the specified amino acids (protein letters) into a long polypeptide chain that folds up (based on those amino acids’ unique properties) to form a functional protein. 

Each amino acid is “spelled” by a 3-RNA-letter “word” referred to as a codon. So, for example, GCA calls for alanine (Ala, A) whereas GAA calls for lysine (Lys, K). Things get a little complicated terminology-wise because those RNA codons are themselves the complements of DNA codons, but hopefully the pictures can help you understand what I mean by that. 

The key thing to remember is that mutations are changes to the DNA (that permanent copy of the recipe) and, if these mutations occur in a region of the gene that gets copied into mRNA, the mutations will also be present in the mRNA copies. There’s some degeneracy (redundancy) in the genetic code, meaning that some amino acids are spelled by multiple codons. For example, GCA and GCU both spell alanine. Therefore, some letter swaps don’t cause any difference in the recipe (we call these synonymous mutations). But, if the  mutations cause the mRNA to spell a different protein letter or specify extra letters or missing letters, you can get dysfunctional proteins. There’s also something called a “nonsense mutation” whereby an amino-acid-spelling codon gets changed to a “stop codon” which tells the ribosome that its job is done before it gets finished making the protein, leading to truncated proteins being made and often the faulty mRNA being destroyed through a process called nonsense-mediated decay. Mutations outside of the protein-coding part can also cause issues because they can lead to problems with things like splicing. Sorry for all this terminology but it will be relevant I promise and hopefully the figures can help you follow along. 

Speaking of complicated, going back to our CF story, the researchers were looking for a gene they didn’t know what was or where was. And, thanks to that whole “splicing” thing, the gene would likely be spread out over a long segment of DNA containing large introns (those regulatory regions) interspersing the protein-coding regions (exons). 

They started by using genetic linkage analysis – looking for genetic markers that are present in patients with cystic fibrosis but are not present in their non-affected relatives. By “markers” I just mean detectable features, such as the presence or absence of restriction enzyme cut sites. Restriction enzymes (site-specific endonucleases) are bacterial enzymes that recognize and cut specific DNA sequences. There are lots of different ones and they recognize and cut different sequences. Since they’re sequence-specific, changes in the sequence can alter the cut-ability and scientists can use this as a marker by testing whether or not a piece of DNA gets cut (they can tell by separating the DNA pieces by size using electrophoresis and then visualizing the band pattern). Other markers involve whether they contain a sequence that matches specific labeled DNA probes that bind in regions that are known to differ between individuals (i.e. polymorphic regions).

Why do markers matter? Due to the phenomenon of recombination (aka “crossing over”), whereby parts of chromosomes swap segments during germ cell development, each chromosome a child inherits is a mix of the both of that parent’s copies of that chromosome. Going back to our cookbook analogy, it’s as if in each of the biological mother’s (or father’s) cells there are 2 copies of each cookbook (one from their mom and one from their dad) and, during the production of gametes (eggs or sperm), some chapters (stretches of DNA) are swapped between the books. Therefore, even though the gamete will only contain a single copy of the cookbook, some of the recipes that get passed on will be from the grandpa and some from the grandma. Because the swapping often involves large regions of DNA including multiple recipes, those recipes that are close together are more likely to be co-inherited. So, if the cookie and the brownie recipe were close together in the cookbook, and a kid gets their paternal grandma’s cookie recipe, they likely also got that grandma’s brownie recipe. But recipes in the beginning and the end of the book are less likely to always be passed down together.

Bringing it back to genes, genetic markers would only be consistently co-inherited with the gene behind cystic fibrosis if they were close together on the chromosome. So researchers used markers to find the approximate location of the gene. This allowed them to narrow things down to the long arm of chromosome 7 in 1985. Additional markers got them closer and closer until they were ready to start “cloning” the region of DNA surrounding the markers, cutting the DNA into pieces (some overlapping) and sticking them into circular pieces of DNA called plasmids (plasmids can only hold a certain amount of DNA so the piecewise approach was necessary). They could then stick those plasmids into bacteria to make lots of copies of that DNA, thus generating a “plasmid library” containing (in pieces) that entire long stretch. With some cleverness and hard work, this library would allow them to test to see if that stretch of DNA contained genes or was just an in-between-gene (intergenic) region. 

Even if they could see all the DNA sequences, the location of genes wouldn’t be obvious, since DNA is “just” a string of letters without punctuation. So they used radioactively-labeled pieces of DNA from their clones as probes to screen genomic DNA of other animals in order to check for the presence of genes. This approach, referred to as cross-species hybridization, or “zoo blots,” can work because genes are more likely to be evolutionarily conserved than non-genes. After cutting up the genomic DNA of different animals, using electrophoresis to separate those pieces by size, and transferring (blotting) them to a membrane, they added the probes. If there was matching DNA, the probes would stick and, when they checked for radioactivity, they’d see bands and follow-up with further research on those clones.

This strategy might allow them to identify candidate genes, but they’d then have to find *the* gene that was implicated in CF. To do that, they would start by using a cDNA library from sweat gland cells, which came from the lab of John Riordan. cDNA stands for complementary DNA and it’s DNA that is complementary to (can base-pair with) the messenger RNAs in a cell. Thus, an cDNA library from sweat gland cells would show which proteins are expressed in those sweat gland cells (thankfully this was a smaller number than the number of proteins expressed in many other cell types). And, since the disease is known to affect those cells, those genes in the library might be relevant to the disease. 

But there was still a lot of physical ground to cover before they got there – hundreds of thousands of DNA letters (kilobase (kb) pairs). Tsui recruited two talented post-doctoral researchers to help him in the search – Johanna Rommens, who had earned a PhD from the University of New Brunswick, and Batsheva Kerem, who had earned a PhD from Hebrew University in Israel. They were able to find closer markers, but they were still about 280 kb away from the start of the true CF gene when news came of a team claiming to have discovered the CF gene. That turned out to be a “false alarm” – the London-based team led by Bob Williamson had only found a nearby gene, but clearly the race was on. 

So, in 1987, Tsui and Collins decided their labs should team up, each focusing on their specialty. For Tsui’s team, this meant “chromosome walking” and for Collin’s team this meant “chromosome hopping” or “jumping.” Basically, Collin’s team would split up the DNA into linear pieces of ~100,000 bases and glue the ends together (ligate them) with a selectable linker piece. Then they would take all those circles, cut them into smaller pieces, and make another library. Thanks to that selectable linker, they could select for only those segments of DNA containing junction regions, separated by the linker. Then they could clone and test those junction pieces. This strategy allowed them to search in 2 different areas with a single clone and let them jump over large stretches of DNA which might contain tricky-to-clone areas. Tsui’s team would use those jumped clones as starting points to do “chromosome walking” – they’d use the end of one clone as a probe to “fish out” the neighboring, overlapping DNA segment from their library, allowing them to creep their way along and test the clones for genes. Through some mathy “linkage analysis” stuff, they could tell if they were going in the “right” or “wrong” direction (towards or away from the CF gene) and they’d be able to tell if they’d jumped over the CF gene and needed to backtrack. 

They went through a lot of cycles consisting of Drumm (a grad student in Collins’ lab) sending DNA from his jumping libraries to Rommens (in Tsui’s lab), who would use that DNA to find new, closer, markers that she’d send back to Drumm who would use those as bait to find closer DNA in his libraries and then send that DNA back to Rommens and… 

All the while, in parallel, Kerem was identifying overlapping pieces of DNA and aligning them to generate a continuous “map” of the chromosome. She also had the task of using the markers Rommens found to test DNA from CF patients and their non-affected relatives. CF was known to be an autosomal recessive disease. This means that you need 2 faulty copies of the gene (one from biological mom & one from biological dad) in order to get the disease. Barring spontaneous mutations, each parent would thus be a healthy “carrier” with one faulty copy and one good one and would thus have a 50/50 chance of passing down the faulty copy to each child. So, if both parents are carriers, each child has a 1/4 chance of getting 2 good copies, a 1/2 chance of getting 1 good copy & 1 faulty copy (thus being a carrier themselves), and a 1/4 chance of getting 2 faulty copies and thus having CF.

Same goes for markers as long as they’re close enough to the CF gene and the DNA around the gene isn’t too variable. If a marker was linked to CF, “all” the people with CF should have 2 copies of it, parents of people with CF should have 1 copy of it and healthy siblings of CF patients should have 0 or 1 copies of it. Kerem did pioneering work on something called haplotype analysis where she looked at the genetic landscape (e.g. which combinations of markers were present) on each copy of the chromosome and even inferred information about the evolutionary history of the mutations. It’s pretty neat stuff that I won’t get into here, but wanted to mention.

After lots of walking, jumping, fishing, and haplotype analyzing, they got a potential candidate. Well, candidateS, plural. There were 4 potential gene candidates within the clone they’d gotten the hit with. They ruled out 3 of them based on other evidence, which left them with just one. But was it really *the* one? They took 2 pieces of DNA that had matched animal genes in their cross-species hybridization experiments (indicative of conserved genes) and they used them as probes to screen cDNA libraries from different cell and tissue types. Quick refresher: cDNA is the complementary DNA version of the mRNA, so it will stick to the edited version of protein recipes. And, since the libraries are made by reverse-transcribing the mRNAs which are present in the sample (making DNA from RNA templates), and then cloning those cDNAs into plasmids, the cDNA libraries will only contain DNA matching the genes that were actually expressed in those samples. 

They got a bit unlucky in that their probe only contained ~100bp worth of the CF gene, so it was like fishing for a giant trout with very little bait. But they got a bite! The probe matched a cDNA clone from sweat gland cells. And when they used that cDNA clone as a probe to go fishing in tissue and cell samples, they got hits (it was much easier to detect now that they had a bigger bait). They found the gene expressed in many tissues, including strong expression in the types of cells most affected by CF.

However, that clone was only ~900bp long and they knew based on RNA hybridization experiments that the mRNA they were looking for was ~6500 nucleotides long. The clone only contained part of the sequence, the very beginning of it. Now they needed to find the rest. So they kept screening – and screening, and were able to find 18 more matching clones. None contained the full cDNA, but many they had overlapping parts and they were able to fill out the ends using “primer extension experiments” which involve using the ends of the pieces of DNA “bookending” the currently-cloned region as starting points for polymerases to copy the matching DNA from the genomic DNA. Once they had all those pieces, they had to sequence them. DNA sequencing at the time was hard, slow, work. Although the cloned region was over 60000 nucleotides in length, each sequenced piece was only a couple hundred nucleotides long, and they had to sequence each piece multiple times to make sure they got it right so they’d be able to tell “normal” from “mutated.” 

When they compared the sequence of a piece of the candidate gene from a CF patient with the sequence from the patient’s healthy relative they found that 3 DNA letters were missing in patient’s gene – TTC. This deletion mutation would make it so that the protein produced from the gene’s instructions would be missing a single amino acid (protein letter), phenylalanine, in the 508th position. Phenylalanine is abbreviated F, so this mutation is frequently referred to as F508del or ΔF508, with the delta symbol (Δ) signifying a deletion. 

The loss of a single amino acid didn’t seem like that big a deal – I mean, proteins have hundreds, if not thousands, of them! The candidate gene was predicted to have 1480 – could the loss of one really be a big enough deal to cause a deadly disease? Later, biochemists would show that yes, indeed, that single amino acid change could disrupt the protein’s folding, leading it to get disposed of by the cell’s quality control system or, if it did make it to the cell surface, preventing it from opening and closing effectively to let ions through. But first, it was Kerem’s time to shine. 

The teams had collected DNA from multiple members of many Canadian CF families and Kerem went through and checked them all for that mutation. She found that 70% of CF patients had 2 copies of ΔF508, each of those patient’s parents had a single copy of it, and NO healthy people had 2 copies of it. The remaining 30% of patients had 2 other mutations or one ΔF508 and one other mutation, but they weren’t able to identify those mutations yet. Further work (largely by Kerem) would show that there are hundreds of CF-causing mutations and, although ΔF508 is the most common globally, other mutations are more common in certain populations. For example, Kerem discovered that a mutation called W1282X is more common in Ashkenazi Jews and her work led to the implementation of nationwide genetic carrier testing in Israel. More on Kerem’s story in a later post, but let’s get back to the main story.

Now that they had the sequence, other than the presence or absence of mutations, was their more they could learn from it? It was time to call a biochemist. They shared their data with their colleague, Jack Riordan, and asked him if he could tell them anything about what this protein might be. Riordan was adept at predicting a protein’s secondary structure (special folds such as helices, sheets, etc.) based on the sequence. There are 20 (common) amino acids and they have different unique properties, varying in size, charge, and “hydrophilicity” (how much they like to hang out with water, such as the water they’d face inside the cell or on the surface of the cell, versus lipids like those in membranes). Based on those characteristics, scientists can get an idea about how the protein will fold, and which parts might be embedded in membranes, without actually “seeing” it. Pretty wild, right?!

Anyways, when Riordan analyzed the CFTR sequence, he found that, among other things, the protein was predicted to have 2 membrane spanning domains, each with 6 transmembrane helixes, as well as ATP-binding folds. This protein architectural layout was common among many transporters, membrane-spanning proteins that help transport things into and out of cells. Consistent with this finding, when Riordan compared the sequence to a database of other protein sequences called SWISS-PROT, looking for similarities, he found striking similarities to transporters including P-glycoproteins. Since much was known about those other transporter proteins, they were able to make some further inferences about their mystery protein.

Based on the earlier works showing that chloride transport is faulty in CF cells, it seemed highly probable that CFTR was a chloride channel.* But they didn’t have proof. So, they went with a safe naming option and called it cystic fibrosis transmembrane conductance regulator (CFTR). 

* CFTR was also later found to also channel bicarbonate ions (HCO₃⁻). Bicarbonate is used by our body as a pH buffer, helping maintain the optimal pH’s for molecules to function. Among other things, optimal pH’s are needed for pancreatic enzymes to work and for mucins to fold properly. Messing up bicarbonate transport can therefore mess up mucus. So, although chloride was the center of attention for many years, more research is pointing to a key role of bicarbonate. 

Now they had all the evidence they needed to share their results, which they published as 3 back-to-back papers in the September 1989 edition of the journal Science. The first paper, “Identification of the Cystic Fibrosis Gene: Chromosome Walking and Jumping,” which Rommens is the first author on, describes how they located a piece of the putative CF gene. DOI: 10.1126/science.2772657 

The second paper, “Identification of the Cystic Fibrosis Gene: Cloning and Characterization of Complementary DNA,” which Riordan is first author on, describes how they used that piece they’d found to fish out, piece together, and sequence the entire CF gene. And how they discovered that 3 DNA letters were missing in the gene of many CF patients (the ΔF508 mutation). It also describes how the sequence looks like it corresponds to a transporter protein. DOI: 10.1126/science.2475911 

The third paper, “Identification of the Cystic Fibrosis Gene: Genetic Analysis,” which Batsheva Kerem is first author on, contains the strong evidence that the gene they found is *the* gene – the one that, when 2 mutated copies are inherited, causes CF. It’s in this paper that they announce their finding that NO healthy people had 2 copies of the ΔF508 mutation but that ~70% of CF patients did. And that the parents of those patients each had one copy of it. The paper also contains Kerem’s detailed haplotype analysis of the chromosomes of CF patients and family members and speculates on the evolutionary history of the mutations. DOI: 10.1126/science.2570460 

This was all great for science – but what about patients? Thankfully, for many patients, this information, and the flood of results that came once scientists knew what gene to investigate, led to the successful design, manufacture, and approval of targeted medical treatments that dramatically improve their quality of life. If patients have the “right” types of mutations that is… 

Turns out there are a lot of CF mutations that cause different sorts of problems. Some mutations affect how well the protein’s channel opens and closes (but they do so in ways that might respond different to the same potential drug). Other mutations affect how well the protein is able to fold stably enough to get trafficked to the membrane. In the above types of mutations, since full-length protein is being made, there’s a chance to rescue it by helping it fold properly (using drugs called correctors) and/or open better (using drugs called potentiators).  

A company called Vertex pharmaceuticals, with substantial funding and urging from the Cystic Fibrosis Foundation (CFF), got on the hunt. The first CFTR-targeted drug to be developed was a potentiator called ivacaftor (Kalydeco). It came to market in 2012 and could help patients with a mutation that prevented channel opening, but only ~2% of patients had a mutation that *only* prevented the opening. Patients with the most common mutation, ΔF508, had a protein that had problems opening *and* getting to the membrane. So they’d need corrector(s) too. 

A first-generation combo of ivacaftor & a corrector called tezacaftor was approved in 2015 (under the brand name Orkambi) for patients with 2 copies of ΔF508. In 2018 they came out with a duo containing a better corrector, tezacaftor. This second-gen combo was approved under the trade name Symdeko for patients with at least one ΔF508 and the other mutation could be one of 17 other mutations. The latest CF innovation from Vertex is Trikafta, which has a 2nd corrector, elexacaftor (so it has ivacaftor, tezacaftor, & elexacaftor). This trio was FDA-approved in 2019 for patients with at least one ΔF508 mutation – no matter what the second mutation was. It isn’t a cure – it’s a lifelong treatment that costs $$$$$. And it doesn’t fully restore lung function, but it was found to increase lung function and the hope is that if patients are started on the drug early enough, before their lungs have gone through the hell of CF, their lung function might never have a chance to decline too much. 

Since ΔF508 is the most common mutation globally, about 90% of CF patients are potentially treatable with Trikafta. Some other rare mutations which cause similar defects also respond to the drug. But, cost aside, that still leaves lots of others who have different types of mutations, especially mutations that don’t lead to protein production (there’s nothing to fix if it isn’t made).

Some mutations, like the W1282X which Kerem found was common in Ashkenazi Jewish patients, cause the production of partial proteins that can trigger a process called nonsense-mediated decay in which both the protein and the mRNA get destroyed. Protein can also be prevented from being produced due to mRNA splicing problems. Refresher: mRNA splicing is where regulatory regions called introns are removed from the protein-instruction-containing exons in order to make mature mRNAs that ribosomes can use as instructions for making proteins. Mutations at or around splice sites (intron:exon junctions) can alter the splicing pattern, messing up protein production. 

Researchers, including Kerem, are looking at ways to counteract these mutations and get full protein made. For example, pharmaceutical compounds that promote read-through of PTCs  (i.e. get the ribosome to blow through the bad stop sign) might be able to counteract problems caused by nonsense mutations. And antisense oligonucleotides (ASOs) may be able to counteract mRNA splicing mutations by using small segments of DNA complementary to specific regions of the RNA in order to hide improper splice sites and promote proper splicing.

A “one size fits all” treatment strategy would be some sort of gene therapy to provide patients with a copy of the normal CF gene. This strategy is being investigated, and it’s an “ultimate goal” for many, but it’s a lot harder than it sounds. DNA can be delivered to cells using harmless viruses, but at the amounts needed to have an effect, these viruses can trigger potentially severe immune responses. And then there’s the issue of the DNA getting into all the cells that need it and staying there long-term. Here’s a good article on where things stand research- and trial-wise with gene therapy: 

The life expectancy of patients with CF has increased dramatically over the last several decades, but it still takes the lives of many too many people much too soon. And those who are affected face a lifetime of intense therapy regiments – no breaks. Suffering continues, but so does research and I sincerely hope that further treatment strategies to improve both quantity and quality of life are on the horizon.

more information on the discovery: 

Key discovery papers:

1. Rommens, J. M. et al.(1989). Identification of the cystic fibrosis gene: chromosome walking and jumping. Science (80-. ). 245, 1059–1065 

2. Riordan, J. R. et al.(1989). Identification of the cystic fibrosis gene: cloning and characterization of complementary DNA. Science (80-. ). 245, 1066–1073 

3. Kerem, B. et al.(1989). Identification of the cystic fibrosis gene: genetic analysis. Science (80-. ). 245, 1073–1080 

good review article: 

Leave a Reply

Your email address will not be published.