221: Weapon of mucus destruction, WMD

This week in microbiology is brought to you by the American Society for Microbiology at ASM.org/twim. This is "Twim" this week in microbiology, episode 221, recorded on July 9, 2020. I'm Vincent Rackeniello and you're listening to the podcast that explores unteen life on earth, joining me today from small things considered, alio-shector. Hello there, how are you? I'm very well, thank you. I am COVID-free. How are you? I'm doing fine. I tested negative. Very good. Actually, I haven't been tested, but I feel fine, but who knows, I could be asymptomatically infected. Yeah, that's great. Also joining us from Ann Arbor, Michigan, Michelle Swanson. Hello, it's a pleasure to be here. Is it snowing in Ann Arbor? It would be 90 degrees Fahrenheit here, currently. So, these are the 90s over the summer there. Is it really hot all the time? It's been hot the last week, but it's also been cooled off at night, so it's been really pleasant. All right. And from Charles to South Carolina, where I'm guessing it's 95, Michael Schmitt. Hello, everyone. No, it's only 30 degrees centigrade or 86 degrees Fahrenheit. So, it's not so bad, but it's principally overcast, guys, and we're sending you up a tropical depression, Vincent, so you're going to get a bunch of rain this weekend. It rained every day this week and last week, actually. We had six inches of rain on Tuesday. Wow. That's Helios, you know, last two years of rain in California. We had it on Tuesday. Well, here we span the country. Yes, we got the West Coast and we have the middle. Isn't that so? That's really cool. And we do microbiology all the way in between. So, today we have a snippet and a paper, and I'm happy to say both are not anything related to SARS-CoV-2, except one of them is about to respire. Listen, can I interject something? For listeners who may not be aware that we are a sister podcast to this week in virology, or run by Vincent, that is a fantastic website, a fantastic resource for anything involving the coronavirus. It is absolutely incredible. And it has a tremendous recognition. So, I'm really very happy to be close to that. I agree. And it's up to date. They stay right on top. And Vincent, I was delighted to hear that reporters are now recognizing you as a resource and contact you after listening to TWIV. Thank you. It never happened before. And I can only offer my condolences. Your inbox is just probably filling up in your voicemail as well. Yeah, but we have an obligation, right, to share our knowledge with our stakeholders. We do indeed. The cycle's the latest is, well, there have been a lot of activity about school openings, but the one that's, their reporters are all in a flurry about is this supposed mutation that increases transmissibility, which of course is nonsense. No, it is indeed nonsense. And I have learned that first fact, listening to your TWIV podcast, and I have been screaming from the rooftops that an amino acid substitution does not convert a fundamental change in how the virus is infecting the human race. The data are just not there. I mean, you got to go and look at the primary data. You don't take the poll quote from the reporter saying, "A mutation has occurred." And they're effectively drawing a logical conclusion without data simply because you're seeing a spike. And again, we have to be better at teaching people to interpret experimental data. Yeah, for sure. And you know, these reporters see what we're saying and they come and they want the other side of the story. So they talk to us, which is good. Right. But yes, it's always been our goal to try and get the right information out there. Well, nextstrain.org has sequenced over 3200 SARS-CoV-2 isolates. And they have a timeline of how those isolates are changing with time and they have a clock. And their clock is offering that the virus, at the last time I think I looked, it was 21 nucleotides are changing per year. Wow. That's nothing. Out of a virus of 30,000 that is not very many nucleotide alterations. And again, remember, the genetic code is redundant. And so consequently, a change in a nucleotide is not necessarily going to change the protein. And it may be in a silent area of the protein itself. So what is the selective pressure that is forcing the virus to change? And right now the virus seems to be going through people quite easily. It's infecting us quite easily. And here on Twim, we are not redundant. Yes. We are not. In fact, we have two stories for you. We're going to start with a snippet from Michael. So the paper I pulled is entitled "Sutomonas Originosa" Biofilm eradication via nitric oxide releasing cycle dextrins. It's by Rulard, Markovitz, Bacado Hill, and Shoeenfish. And they are located at the University of North Carolina Chapel Hill. And they are in the Department of Chemistry and Physics and Astronomy and Pharmacy. And it's a really fascinating paper. The bottom line up front is found in the image associated with the abstract. And what we are offered in that image is a pure culture of a "Sutomonas Originosa" Biofilm growing on a surface. And as the first line of the author's abstract offers us, this microbe is the principal contributor to the morbidity and mortality of cystic fibrosis patients. Which as many of you know is a genetic disorder that impacts the lungs, the pancreas, the liver kidneys, and testin of individuals afflicted by this autosomal recessive genetic lesion, which biochemically is pretty simple. It's manifested by the presence of mutations in both copies of the gene for the cystic fibrosis transmembrane conductance regulator, commonly abbreviated CFTR protein. And it's taught in probably high school biology today because it was one of the first diseases for which the molecular biology was worked out. And this protein is principally involved in the production of sweat, digestive fluids, mucus, which is really the focus of today's presentation. And when presence, this lesion substantially impacts on the life expectancy of the person carrying this lesion, resulting in an average life expectancy of between 42 and 50 years of age. And that has gone up with the advent of effective antimicrobial therapy, but as we well know, 42 to 50 years of age, we'd all be dead on this podcast. So it is really quite scary. The other thing about CF is it has a penetrance of approximately one in 3,000 in the newborns of parents of northern European descent. And it's less common in Africans and Asians. And the main signs in symptoms are salty tasting skin, poor growth, poor weight gain, and why we are here is due to the accumulation of this thick, sticky mucus, which leads to frequent chest infections. And lung problems are really responsible for over 80% of the deaths. Now we have discussed CF and pseudomonas in previous twins. And what I'm going to remind everyone, it's a core-immediated event with two communication signals that are controlling two unique virgins, regulants, that enable the pseudomonas originosa really to cause the profound pathology that is seen in these patients. And I'll place into the show notes a nice review that appeared in protein and cell in 2015 that is in the post-culture. of domain that will provide you an overview of the biology of this forminal pathogen that a snippet won't allow me to expand upon that is simply entitled the hierarchy, quorum-sensing network in pseudomonas originosa. So now back to the story. The challenge before microbiologists and infectious disease physicians treating children, adolescents, and young adults is that this chronic respiratory infection caused by pseudomonas originosa and the lungs of these individuals are rarely eradicated due to the protection that the cystic, fibromatic mucus actually affords the biofilm matrix. And the composition of the biofilm matrix determines its viscoelastic properties and then controls or effects antibiotic efficacy. In other words, the antibiotics simply don't have enough oomph in order to eliminate the infection. So the biofilm, once it's effectively whacked down, it will actually come back sort of like mowing the grass at our homes. The bottom line is that the virulence programming in a CF individual allows this microbe to elaborate a unique extracellular matrix which is made of water and then something that's termed generically extracellular polymeric substances or EPS. What we need to appreciate is that this matrix contains large concentrations of DNA, negatively charged molecule, and polysaccharides and proteins that elevate and here's a term they introduce, the viscoelastic moduli which we have all empirically experienced as young children when we explored our own EPS matrix by placing the mucus coming out of our noses between our thumb and index finger and marvel at the elastic mixture of. You know, it's over lunchtime in California. I'm sorry, I know this is gross, but this is audio and not video, so I can't show you the elastic nature of this mucus, of snot. So the DNA and the EPS actually increases the sniff stiffness of the biofilm and as DNA is negatively charged it can tie up cationic or positively charged antibiotics again reducing the effectiveness of antibiotic therapy being offered in the CF patient. Again, the consequence here is that the debris and the viscoelastic properties confer a reduced susceptibility of this biofilm community to antibiotics. Now this is really bad news if you're trying to limit the spread and effects of infection. And again, the review on pseudomonas and the virulence regulant and all the virulence factors that this microbe elaborates really highlight how this microbe is able to really cause systemic pathology in not only the lung but the rest of the hose. Now the other problem we have is gone unchallenged the biofilm when it gets too big for a particular location, the film effectively decamps and that's part of the quorum release mechanism. So we have planktonic cells that we even then find new ground which effectively expands the biofilm. And we have spoken previously about strategies where therapies penetrate or disrupt the biofilm and we know that those generally work well but clinically they have been hampered by the tenacity of the mucus produced by the cystic fibrosyket individual. So you guess correctly the microbe takes advantage of this niche and the biofilm infections are often inherently heterogeneous which in turn again limits the effect of the traditional antibiotic therapies. Now enter the story. So now you should have a strong frame of reference so you can follow along as you're listening. Enter nitric oxide. So nitric oxide which they abbreviate NO can both disrupt the physical structure of a biofilm and eradicate the biofilm principally the microbes namely pseudomonas originosa because in most CF individuals the biofilm is often a pure culture. Now the nitric oxide has two principal activities. First it's mucolytic. It will literally take the snot out of snot and secondly it's bacterial side. So step one, bust up the mucus and step two as the nitric oxide is a strong oxidant this will result in the killing of the biofilm slash microbial cells. So it's a bacteria. Beyond that by removing the biofilm it would allow the antibiotics which are being used to be effective. Absolutely. Absolutely. The mucolytic activity is really one way of doing this. So you see it's the equivalent of a molecular grenade which you can visualize how it might go off. Just like in battle you need to deploy the grenade to the proper location where it can affect the most destruction and we want to destroy the biofilm. We want to destroy the mucous. Now previously this group found a way to deliver the nitric oxide and they were able to do this with a kytosan which was shown to have significant anti-biofilm action. But its limitation was on the chemical side because it's a it's a linearly it's a linear positively charged molecule where the delivery vehicle caused the contraction of the biofilm and thus limited its overall mucolytic. Even though the nitric oxide was coming out but because the biofilm effectively dried itself down like concrete it was effectively very challenging for the nitric oxide to bust up the mucous and then it would limit its effectiveness. So enter the molecule the authors investigated. The cycle dextrin scaffold to which the nitric oxide was added. Now this is a good model of the molecule and that the parent the beta cycle dextrin is less than two nanometers in size which means it can easily penetrate the film which in the presence of a proton and recall that protons are routinely ejected with each electron that goes into a bacterial membrane and their presence as ubiquitous and the presence of the proton is effectively the equivalent of pulling the pin under grenade. And you guess right a eradication of the biofilm regardless of the mechanical strength leading the authors to propose that nitric oxide therapies especially one where this delivery mechanism is small like psychodextrin maybe able to eradicate mechanically robust cystic fibrodotomy and these mucous layers that routinely develop in advance CF lung disease. Michael. Yes. Beta cyclodextrin are these used in people for anything else? I don't know. I didn't dig into the pharmacology but they say they have a favorable toxicity profile but I just wonder if it's used for anything else. I don't know. I didn't have time to get into the pharmacology of the beta cyclodextrin as to see where they're also used. We also need to appreciate then since the nitric oxide that's coupled to the cyclodextrin answers yes. Okay. There used to be pharmaceutical drug delivery agriculture environmental engineering so yes. It's a good choice. That's a good choice. So in addition we can't forget that nitric oxide since it's in the presence of proton is going to be released. This will then serve to deaminate the DNA that is lending this negative charge to the mucous so that it has a beneficial effect of limiting the effectiveness of the biofilm at being able to resist the antimicrobial therapies. It's going to deaminate all the purines and pyrimidines as necessary. So the experiments that this group conducted are the ones that you likely have already thought of yourself. First what influence do growth conditions? have on biophysical properties. I'm not going to burden you with all those experiments other than to share with you that the first thing that they established is that the beta-cyclo dextran, what it does in the presence of a proton or an acidic environment approximating that of a biofilm is it quickly releases the nitric oxide effectively releasing their grenade from the sores and the concentration goes quickly up to a clinically relevant concentration and that concentration is sustained over a seven hour time frame where it can do some clinical good and that's measured in the form of micromole per milligram. So the next group of experiments that they did is they evaluated the pseudomonas originosa's behavior when it was grown on different media. And I'm only going to explain one of their specific experiments, but you need to understand that the media that they use to evaluate the behavior of pseudomonas originosa in the context of their beta-cyclo dextran releasing NO. They also looked at tobermysin and coliston, which are two antibiotics that are routinely used to try to bust up, at least tobermysin is routinely used to try to bust up CF-based biofilms. So the four media that they tried was one of the favorites of microbiologist, Triptocase Soybroth, which is a generic medium that is used, it's buffered, it's got glucose, it's a nice happy medium. The second medium that they used is they supplemented it with poor sign gastric musins at a final concentration of four grams per liter, which makes the medium a little more viscous or mucousy. The third medium was an artificial sputum medium and as you can imagine it's it's simulating what sputum is going to behave like. And the fourth one, the one that I want you to focus on when you look at the experiments is something called sputum filtrate medium and the sputum was isolated from a CF patient. They had two CF patients who offered to donate their sputum and then that sputum was effectively collected. It was diluted twofold in PBS. It was homogenized to effectively bust it up by pipetting and work-taxing. And then that was spun down, getting rid of the solids and the supernaten was then filter sterilized and stored until it was needed where it was then diluted one in ten with TSP. And that medium was simply termed S F M or for sputum free medium or sputum filtrate medium. So they second and third and fourth experiments are they're looking at the snot properties of how the pseudomonas is behaving. And I'm not going to go into detail of those experiments, but if you're interested you can see it. I'm just going to cut to the chase and go to the experiment that really wowed me. And that's in their fifth figure and they simply asked the question of what the antibiotics did and they're calling the cyclodextran nitric oxide releasing substance, they're antibiotic. And they again had three very good controlled experiments. They had cyclodextran, tobermysin and calliston. And here they monitored the viability of the biofilm as a function of growth. And again they were using the 4Media I just described. And what they learned is that the antibiotic minimum biofilm eradication concentrations that they abbreviate MBEC varied between the biofilms formed in different growth media. That's not surprising because their previous experiments were able to show that the snot characteristics were changing. And when you look at the biofilms grown in the artificial sputum medium and the sputum free or the sputum filtrate medium, they required a greater concentration of tobermysin for complete eradication relative to TSP or TSP plus the gastric musins. And the experiments with the anti biofilm action of the cyclodextran nitric oxide was not influenced by the inherent biofilm matrix composition or the snot-like characteristics. And the the nitric oxide eradicated all the biofilms at the same concentration of two megs per mill irrespective of growth conditions. And in fact in the ASM plus the biofilms which had the greatest measured snot property, they were more susceptible to the cycle dextran nitric oxide releasing compound at the lower concentration of one meg per mill compared to the other biofilms. So what do we learn? Nitric oxide is an attractive molecular grenade or as I would argue a smart weapon where the material delivered can go to a complex biofilm, busing up its primary protective layer that mucus and as a consequence the bacteria side on the community because we know strong oxidants really work well against biofilms. The presence of the environmental DNA such that we see a lot in cystic fibrosis mucus translates to large DNA concentrations in this EPS matrix. It elevates the viscoelastic moduli and it results in decreased antibiotic efficacy which then leads to pseudomonas becoming more tolerant because it's a selective pressure to the organism as you need to keep increasing the concentrations. And the exogenous nitric oxide represents a powerful anti-biofilm therapy because it can simultaneously disrupt the biofilm regardless of the matrix composition such as DNA or its stickiness or mucus like property that is technically referred to as the rheology and it can most importantly eradicate all surviving colonies. But finally the small size of the nitric oxide in this matrix being at two nanometers and its mechanism of bacteria side all action allow for effective biofilm penetration first and foremost and then eradication of the biofilm regardless of the mechanical strain. So I looked at this and I said there may be hope for the poor kids who have CF because it can be easily delivered and I'm hoping to see a clinical trial soon as to whether or not this material can be used. As far as you know that there's not one published. No, I have I I looked for that and I didn't find one. So those say you need to do some animal studies. Yes, they need to do animal studies and there are good animal models for CF because we understand the CF and we understand the process so I'm certain that they'll go to animal trials first. So Michael one of the reasons antibiotics are such great drugs is we can target them to proteins or pathways that are unique to the microbe whereas NO is really pretty reactive. So how would they target it? How would you deliver this to a kid who's got CF biofilms in their lungs without doing damage to the animal? As a nebulized vapor, they would take it as a breathing treatment. But wouldn't it do damage to everything it touches? I mean well the biofilm you target it to the biofilm but you know there's a lot between that is true. And so you would use it in concert with an antibiotic is what I would assume would have. Or direct or direct the agent to the biofilm matrix so that it's not too hard to touch. You want to attach a monocolonal antibody that would be specific but the problem there is again it's got to get through the mucus in order or maybe make a monocolonal against mucus to see whether or not you could effectively attach it down. - Right. - The cyclic dextrin and then it generates the NO, which could then penetrate. - Well, the NO isn't released unless the pH is sufficiently low. - Yes. - And maybe the answer to the question, because you need a lower pH in order to release the nitric oxide from the cyclic dextrin. - Yeah. - Maybe I didn't explain that well enough in my attempt to snippetize this to-- - Right, but the nitric oxide, once released, is could get a specific, that is correct. - Could get into the circulation. So I don't know how they're gonna deal with that. - Yeah. - That's a good point, Michelle. And oh, even we'll have to, we'll get into the bloodstream, for example. - Oh, yes. - Oh, yes. - I'm wondering if it could be used in an environmental remediation type strategy where you have to clean out hospital supply of biofilms, for example. - Or better yet, the brewing industry, where there's biofilms in the plumbing. - Yeah, or Legionella in pipes. - All right. - Interesting. Michael, it occurs to me, this is a WMD, a weapon of mucus destruction. - It isn't WMD, a weapon of mucus destruction. That is great, Vincent. (laughing) - Oh, well, my mind is just-- - You're mind-blowing to be on COVID. - Yeah, well, I didn't influence by Alan Dove. Michael, that was terrific. I think it's a really clever approach. I hope it moves forward. - Yes. All right, next on our menu for today, we have a paper from, you know. - And the paper is titled, "Colestorometabolism by Uncultured Human Gut Bacteria Influences Host Collestoral Level." And it is published as an open access paper in Cell Host and Microbe, in August of 2020, so coming up. And it is from a collaborative group at the Broad Institute, which is a partnership between MIT and Harvard in Cambridge. And the authors are, first co-authors are Douglas Kenny and Damian Plicta. And other authors include Shungan, Copal, Hall, Fu, Vazen, Shaw, Vlamakis, Balkas and Xavier. And it is about microbial dark matter. And although we're going to be talking about feces that dark here doesn't refer to fecal material, but instead emphasizes that there are a vast number of microbes, gazillion microbes that we have yet to culture or characterize, but they're out there. So they are going to probe this dark matter and identify an amazing class of enzymes that are encoded by microbes that colonize our gut and look as though they can modulate cholesterol levels in our blood serum. So to do this, they leveraged knowledge of cholesterol metabolism from the literature. They also had some amazing databases of bacterial genomes and also some really valuable paired sets of human fecal metagenomes and their matched human fecal metabolome data as well. They also applied some classic molecular genetics and biochemistry and importantly, a great deal of really insulicoingenuity to identify this new family of intestinal sterile metabolism genes that likely decrease total cholesterol in the serum of the human host. So this is important because it's well established that high levels of serum cholesterol caused cardiovascular disease, which in industrial countries is responsible for 25% of deaths. So this report is really a major step in fulfilling a promise made more than a hundred years ago by Dr. Austin Flint in his presentation to the American Medical Association and published in 1897 JAMA, which you can access. It's available on the World Wide Web. So in that talk in 1897, Dr. Flint described the conversion of cholesterol in the gut by the action of fecal bacteria. And he concluded that our knowledge-- - It's a very different reason. - Yeah, it's a great reason. And it's a lovely plenary talk. You can read his whole presentation, but he closes by saying our knowledge of the physiological chemistry of the feces is only just begun. We may look to future investigations for much that will be most important as well as interesting. So let's just see how important and interesting this is. But let's start with cholesterol. So of course it's gotten a bad rep because it's known for clogging arteries. It's a reason we try not to overindulge in fast food. But what's known? Certainly our diet is a major source of cholesterol. If you eat animal products that can get into your small intestine. But also our liver synthesizes cholesterol, which is transported to the gallbladder, secreted into the small intestine. And then we have epithelial cells in our intestine that can package and secrete cholesterol into blood plasma and there it travels around the body to the sites that we need. And of course we need it to build membranes of our cells. And also cholesterol is needed to synthesize steroid hormones, vitamin D, and bile acids, which are important for our digestion. So the hunt is on now. Let's find microbes and microbial products that can digest metabolize a cholesterol inner gut. And they were motivated to do this by a old literature and some recent papers that were very exciting, including a paper published by Park at All, a group from Korea from the Handog Global University. They published in PLOS-1 that a particular lactic acid bacteria, when it colonized mice that were fed a high cholesterol diet, this particular strain of lactic acid bacteria reduced serum cholesterol of the mice, increased short-chain fatty acids, to the degree that was similar to their positive control, which is the drug statins. And we've probably heard of people taking statins to reduce their serum cholesterol. So that statin reduces our inhibits a particular enzyme in the cholesterol biosynthetic pathway. So there's definitely reason to believe that there are microbes out there that can metabolize cholesterol. And this group is determined to locate them in the dark matter. So I thought this was really a tour de force. Their whole strategy is laid out and figured two of the paper, but they also provide in figure one a nice schematic of the metabolism of cholesterol by these enzymes. So they're in particular interested in enzymes that can metabolize cholesterol to coprostonal, because once that is in our gut, our epithelial transporters cannot ferry it into our bloodstream. So if the microbes can essentially take the cholesterol out of that pool and reduce serum cholesterol then indirectly. So as a first step in this hunt, they leveraged more than 3,000 human gut microbiome data sets. And they assembled this into a haystack, let's call it, of 6 million different genes from the human gut microbiome. So all the collective gut microbes collect their genes into one big data set. And then as a way to begin to whittle this down to a reasonable number of candidates, they then regrouped that this 6 million list of 6 million genes into groups by homology. And in particular, different functional classes based on about 50% amino acid identity. So that grouped it now into a million genes. So it's getting a little easier to work with. But that essentially is their haystack. So they've got a haystack now of a million different groups of functional genes from guts that live in human bugs that live in the human gut. So now what does the needle look like? To begin to get clues to that, they tapped a study that this group had contributed to and published in a pair of 2019 papers in nature and nature of the microbiology, where they had 625 samples from humans, where they had determined the metagenome sequence. And also from the same sample had profiled the metabolism. So they had these two big data sets where they can go back and forth between 625 different human samples. And what they started to do first was ask of the 625, how many of these samples have, co-prostinol, the metabolite of cholesterol contained in them. And when they pulled that subset out, that then whittled their haystack of genes down to a more manageable group of about a million genes. So we're only a million. And a little smaller. So then they used their knowledge of this particular class of genes, and they came up with query criteria. And they could adjust both the sensitivity and specificity to try to logically come up with a more narrow set. So when they did that, initially, no candidate genes popped out. So instead they began to look at the literature and think about the types of enzymes that could be working. And they also adjusted their sensitivity and specificity criteria in ways that they describe in the paper to whittle this one million pieces of straw down to 30,000 candidates. So then to narrow it more further they then went to the literature and there are reports of four different genuses of bacteria that have been reported to metabolize cholesterol. So in the laboratory they got their hands on one of each of this type and they tried to confirm that the particular strain could in fact metabolize cholesterol and generate this metabolite. So they were able to demonstrate that for one of the bacteria. It's a U-bacterium strain that had been originally isolated from hog sewage lagoon. So once they had that and it verified that it could break down cholesterol they then determined its whole genome sequence. And when they then queried that whole genome sequence against their remaining haystack they did find about 300 proteins that met particular sensitivity and specificity criteria. So by using their knowledge of the biochemistry and their bioinformatic skills they're now whittling this haystack down. But they still have now 328 candidates. So at this point they went into the wet lab and decided to start with one of the one U-bacterium strain that had previously been reported to break down cholesterol and that they had verified in their lab. And they studied its ability to break down cholesterol in an in-vitual reaction. So they prepared lysates from this U-bacterium, fed the lysates cholesterol and low and behold did generate one of the metabolites of cholesterol. And they also learned in those in-vitual reactions that the enzyme activity in this crude lysate required NADP but it didn't require oxygen. So these are key features of the enzyme that they then use to narrow the candidates even further. So again based on their knowledge of cholesterol metabolism and encymology and the literature they reasoned that the enzyme could be for example a cholesterol oxidase could do this. But when they queried their 300 genes there were no cholesterol oxidases. Likewise they thought maybe it could be a cholesterol dehydrogenase but again they had no cholesterol dehydrogenases left in their pool. So they then thought about hydroxy steroid dehydrogenase enzymes. And these seem like good candidates because they are known to be produced by bacteria from human gut and they're known to require NADP and they're known not to require oxygen. So in fact they found in their pool of 300 when they looked specifically for hydroxy steroid dehydrogenase enzymes they did find for protein clusters that met these criteria. So now this is a much more reasonable number. They could go back to the wet lab and use classical methods to study these candidate enzymes. So in particular they expressed any coli, the four candidate hydroxy steroid dehydrogenase enzymes and made lysates and did find in fact that they could oxidize cholesterol cholesterol. And they also found that when they cultured the bacteria and added cholesterol the bacteria induced expression of this particular enzyme. Consistent with that enzyme being responsible for the enzymatic activity that they had quantified in their crude lysates. So they decided to really focus on this particular enzyme. They cloned it. They epitope tagged it and that allowed them then to purify the enzyme. And again doing biochemical assays in the lab they demonstrated that it could convert cholesterol to cholesterol and found that it required NADP. So they are feeling really confident that they've identified an enzyme that can break down cholesterol and they've named it now intestinal sterile metabolism A or ismiae is how I'll refer to it. So now that they have one gene they went back and again use their bioinformatic wizardry to ask if there are similar genes elsewhere in their big database. And in fact they did find more in the microbial dark matter. All of them were orphan genes in the NCBI database so they had not been previously characterized and they were encoded by microbes that had not yet been cultured. So clearly new material to study. And they also verified by expressing each of these new candidates any coli and making lysates that they each enzyme could oxidize cholesterol. So they're again feeling really confident that they've gotten their hands on an enzyme that can perform this first step in the cholesterol metabolism in the gut. They also by aligning the sequences of these six ismiae homologs identified a really tempting catalytic triad that would be predicted to be responsible for the enzymatic activity. And indeed when they did site-directed mutagenesis they demonstrated that that motif was critical. So then they use this ismiae sequence to search for related enzymes in their database. And as I mentioned they found more of these orphan genes they verified the sequence. And then to use phylogenetic analysis and found additional microbes that encode ismiae homologs. And by mapping them they found that they all fell into a clade in the clostridium genus which again makes good sense because the clostridium are known to be in the gut and known to produce short-chain fatty acids. And in that by doing that analysis they were able to identify additional bacterial strains that encoded more ismiae homologs. And then went into the published databases and did searches and in fact found yet 14 more ismiae homologs. So there seemed to be multiple ismiae genes spread in this dark matter that have been previously characterized from metagenomes. So now they get into the really exciting biological tests. They've shown in vitro that these particular enzyme can oxidize cholesterol but is there any evidence that it can impact our serum cholesterol levels. So as their first biological test they took stool from eight healthy donors, cultured at anaerobically, and then quantified cholesterol and its metabolic byproducts. And indeed found that those healthy donors who encoded ismiae in their gut metagenome did have more of the cholesterol metabolite, postinol. And then when they looked at human fecal microbiomes that are available in the literature, they found anywhere from 37 to 92 percent of the samples of human fecal microbiomes. In fact encode ismiae homologs. So this doesn't seem to be a rare class. It's just has not been studied. It hasn't been cultured. I haven't had a way to really get at it. So now the icing on the cake, they went back to this data set that they had published in 2019 where they had hundreds of paired human genome gut metagenomes samples and human gut fecal metabolone data sets. And again they found that the presence of oxidized cholesterol in their metabolism individuals also encoded an ismiae homolog. So again the statistics were really quite strong. So again consistent with their hypothesis. Not only that, but the samples from people who had a microbe that encoded the ismiae gene also had lower stool cholesterol and increased levels of the cholesterol metabolite cholesterol. So now is the grand finale. They again went to the literature and there are three large data sets of human stool samples where we have the metagenome profile and also quantified the amount of serum in serum cholesterol in those same human donors. And these are from more than 1400 people from the US, the Netherlands and China, three distinct data sets. And what they found is that in stool metagenomes that encoded an ismiae homolog, there was a decrease in total cholesterol compared to those metagenomes samples that did not encode the ismiae homolog, ICMA homolog. And the decrease was on the order of 0.15 millimole per liter. Now, to put that in context, it's known that there's a genetic predisposition, human mutation, in the HMGCR gene, that alters cholesterol by about.05 to.06 millimole per liter. So there's in that same order of magnitude, but the microbe is correlated with a.15 millimole reduction. And to put that in greater context, if you take a statin to reduce your serum cholesterol, you can get a 1.2 millimolar per liter serum decrease. So all of this together, I think, shows that there is not uncommon microbes in our gut, in particular, there's this U-bacterium species, that produce an enzyme that metabolize cholesterol to a form that our epithelial cells cannot transport into our serum and therefore reduces our total serum cholesterol. And they've done so by what I've found to be just a really insightful, rigorous kind of systems biology approach, where they've taken advantage of these large data sets of bacterial genomes, metagenomes from human feces, and also metabolome profiles from the same human species. And logically came up with steps that they could narrow down candidates and then take their reasonable list of candidates and actually do the functional tests in the laboratory. So it's really a beautiful demonstration of the marriage of bioinformatics, computational biology, and classic biology, and microbiology and molecular genetics and really showcases what places like the Broad Institute and other groups that really combine these genomic approaches with classic microbiology and molecular genetics where they can take us. And of course, the next big step, of course, would be how can we now culture these organisms and or promote their colonization, stable colonization in humans as an alternative to therapeutic approach to control dangerous levels of cholesterol in our serum that can cause. Maybe we could put the oxidase in some other cut-backed terium and just these probiotics that you take that are okay to eat, maybe we just put it in there. Yeah, so I did simplify this a bit. They focused on the first enzyme, the first step of that metabolism. So it takes at least three different steps, but they now have their hands on the pathway and can begin to think about those steps. What is the benefit to the bacteria to oxidizing cholesterol? Is there some? You know, they didn't discuss that, but I presume they're using it to either for energy or for biosynthesis. Yeah, to get the fatty acid, the short-chain downstream, right? Yeah, yeah, for built-in. I don't know what's good for us, but they have no what's good for us. You know, it's the bacteria that do indeed know what's good for us. It's a great study. Yeah, really exciting. I wondered why in the beginning they didn't just look for those enzymes, but I guess they didn't know the enzymes involved in the oxidation, right? It's the chicken and the egg story. Yeah. The pathway or the process had been described, but people had not been able to connect it to genes or enzymes. So that's really the tour de force here. What percent of the population do you think have these cholesterol oxidizing bacteria? So they had two different large data sets, and I believe it was somewhere between, yeah, in one data set, it was 37% of the human fecal microbiome samples. And then the other data set was 92% had homologues of ISMEA, ICMA, the first enzyme in this pathway. Pretty cool. Yeah. So let me say something about the two senior authors. Please do. This is unusual. We normally just choose one author, the very first one. But in this case, it makes sense to use the first two because they were totally symbiotic. They were interchangeable, apparently. And they did a lot of work. In fact, let me tell you what the second author, Damien Plick, I had to say about the first author who was Douglas Kenny. This is what Damien said. The project was really Doug's baby. He nourished it, thought hard about the chemistry, thought hard with many experiments and never stopped to be excited. His excitement spilled over the list of the team. Not bad. So now, what about Douglas Kenny? Well, he was an undergraduate at the University of Delaware. He went to the Broad Institute. And he started science by enjoying science classes as he was growing up. So by the time he reached college, he knew that's what he wanted to do. He fell in love with research and he's been lucky, he feels, to have had support of help from many others. Memory that he was asked, "What is his strong memory, especially excited you?" In the lab, he says, "Most memorable day in the lab was when he first detected the activity of the ISM aging from a ubiquitarian, copper, or starin or ligands." This was the first solid evidence that it were on the right track. So finding the enzyme is really what they did. So when asked about what encouragement he has for more junior colleagues, he says, "The most important thing he helped me in your graduate school is to find hobbies to do with friends outside the lab." He said, "You chose to go to grad school because you love research, but it's just as important to have other hobbies." So he says, "And he has a picture that shows us he is stealing a boat for the dragon boat. It's a higher-ward boat, probably in the Charles River. And it looks like a dragon. It really is quite a sight." And he says, "I don't remember dragon boats from our time in Boston, do you, Elio?" "No, I don't. I don't. I don't remiss them. We blew it." But let's say what the Daemian has to say. Daemian has an interesting background in that. He grew up in Poland and then went to Denmark to do some work. He was a postdoc in a lab by Henry Bjorn Nielsen, Systems Biology, Micro Biome Systems Biology Group. He then worked for a. Before that, he had worked for a biotech company which was. Well, the Center for Biological System Analysis, Sequence Analysis. He then started to do a postdoc at MIT, where he is now. And he says, "If I can find it in one second, he finds that in a way doing research is for fulfilling the American dream." I'm not going to make it to the point where I'm going to make it to the point where I'm going to do research. I thought it was a very clever, very interesting way of putting it to fulfill the American dream by doing research. What do you think? Not bad. Anyhow, who and Kenny and Daemian are a team? And there's another guy involved, so that they're really doing very, very well. The other person is a postdoc in the Ramnex lab, the Mietri Shungen. So it's really quite a group. They seem to interact extremely well and there's a seem to be very nice people. Yeah, it was very interdisciplinary, wasn't it? This study. That's right. For those of you at Harvard who routinely read the Harvard Gazette, this story was picked up by the Harvard Gazette. We'll put the link in the show notes. It's elegantly entitled "Microbes Might." I think you can strike the word "might" and say, "Microbes manage your cholesterol." I mean, before the advent of statins, they were in charge. We just ate the wrong food and depleted the guys who were managing our cholesterol and then things got out of control. That's the only quibble I have with the Harvard Gazette story is they should strike the word "might." I also can add that dragon boat racing is a traditional Chinese watercraft activity. It's been going on for more than 2,000 years. And people, there's some 20 rowers, 20 paddlers to abreast. And of course, there's a drummer in the front to keep the cadence and it's a race. Cool. I like this idea about being very enthusiastic. I think that's so important because so many people get kind of glum when their experiments don't work and they lose their enthusiasm. It's hard to keep it up, I understand. But it's really important. Oh no, absolutely. You can influence other people by staying in Tuziastic. People draw off you. I know that because I did that. I had it in the lab when I was a student. It was a guy who was so enthusiastic. It was infectious. It was just great. Yeah, because you've got to be able to imagine possibilities. You can talk yourself out of any experiment, but if you can suspend this belief and think big. Well, the only thing more contagious than COVID-19 is hope. And so if we can have hope and the science that we're pursuing, I think our science will ultimately go much further than if you're so negative all the time. And you're just gwom and nothing's working. You just have to work through it and keep your spirits up. Well, I am so excited about this particular finding and the realization that there are human microbe gut bacteria that can break down cholesterol. I'm super excited. I can only imagine how this team must have felt as this story unfolded. It's really exciting. I noticed that Doug is defending in August. Yeah. Well, it couldn't have been working for a while. That's tough. Because that's the fact that-- Maybe he's been writing at home, right? And when you've got the paper published, that's big news in itself. Yeah. And say what he's going to do next, but I wonder if he's not going to stay put and finish up a little bit of what he's telling. He, um, I love it. And he's going to be virtual for sure. And then because-- No, because Harvard's not having classes in the fall. Or football. Zoom away. Zoom away. Zoom away. Maybe you could have virtual, you know, Zoom football games. No. No. No. It won't work, Michelle. It's like Zoom golf. It just doesn't have the same panache. Michelle, no football ad Michigan, right? They have not yet announced what the plan is. I'm sure they'll cancel. The clock is ticking. Yeah. All right, thank you, Michelle. It was lovely. I enjoyed it. Thank you. All right, let's read an email or two. One is from Gordon, just two episodes of "Twin" under my belt and have been delighted. It's amazing how well and non-chemist. And visualized chemical cellular processes from the "Twin" podcast. Thanks. Episode 216 starts with a cough had me captivated. This question is about one small part of 216. The practical implications for ordinary patients that may follow from the email discussed from Harry. In particular, Harry's comments on present surrology testing limitations and his questions about where to search for molecules, which in Antiviral Drug Good Target, the excerpt begins at minute 59.27. So that was the one where the component from TB makes animals cough, right? Right. Because I think staff at my doctor's office might enjoy the excerpt, I want to be sure I grasp the main points. Besides offering insight possibly new to them, maybe this excerpt will also alert them to the practical benefits of "Twin" here, here. Back to Harry's email, I may have misunderstood, never studied microbiology, but it sounds like present surrology tests are barely useful. They can't be relied upon by medical staff to accurately evaluate a patient stage of infection or post-infection. From that, I infer that patients should be skeptical of them too, especially of a negative result. Was that your takeaway from his comments? Well, I don't remember what he said, but they're not completely useless. It really depends on the particular microbe, right? Yeah, many companies make tests and their accuracy varies from company to company. So it really depends, and the FDA has a wonderful page of the reliability of COVID surrology tests. So I think they can be useful, but of course they're not 100% accurate. You have to remember that. One of the biggest ones that's problematic is the surrology test for a Lyme disease, because it takes a while for you to develop IGG against Lyme. And so you just have to test at the proper time. And the IGM, which is the first antibody produced, goes away fairly quickly, and then it takes a while for the IGG levels to manifest themselves. So, but that's generally known by the clinician ordering the test. And when they ordered the test and you sit with them and they interpret the result, they should offer you the grain of salt that you will interpret. I recently had a surrology test for one of the tick-borne diseases that frequent South Carolina. And I was negative for that, but there are lots of tick-borne diseases in South Carolina. I had tick-bite and I had the rash and all the other associated symptoms. So they just broke out the antibiotic and said, "Well, the surrology didn't tell us what we wanted because it was specific for that one rachetsia that they were specifically testing." All right, continuing with Harry. IGM assays are especially unhelpful due to low specificity. Even IGG test results which have marked better specificity must be interpreted with many caveats, is the right antigen being tested. The full negative rate for IGG assays is high. So, a natural question is what's the full negative rate for a given serologic test? As an of the time the lab for the assay to detect latent presence of M, but I guess he means that the assay will be allowed to culture for weeks. No, the assay doesn't take weeks. It takes hours at best. But to get IGM, you have to go weeks after the infection because IGM comes first and then weeks later IGG. That's probably what he was referring to. Yeah. From all that, it appears that Harry and late April did not see a serologic test that could tell one very much. I wonder if that still holds for the commonly given test in the US. No, there are some good ones now and they're not just in the US, but in other countries that are giving good data. Also, Harry linked to Australia Public Health Lab Network, Point of Care, testing guidelines from March 20th. Quote, PHL, and recommends that mass surveys of immunity to retrospectively determine the true prevalence of infection should be done with ELISA, and that will not lateral flow devices. Is the upshot that until we learn otherwise, patients should be seeking an ELISA-based serology test. It sounds like having an answer to this question would help patients get more quickly to a trustworthy test result. Yes, I think the ELISA tests, which are done in laboratories, which are done with more blood or serum are in general more accurate than the lateral flow assays, which are done with the pinprick of blood and you can do in your home theoretically. So, I would seek out an ELISA test. Again, I will put in the show notes a link to the FDA site that we've mentioned over on TWIV about how which ones are good and which ones are less good. Okay, I think we'll stop there because the next ones are rather long, we'll save those for next time and that is TWIM-221. You can find the show notes, which includes links and the letters at microbe.tv/twim. You can subscribe on any podcast player. We'd love you to do that. Tell us how many people are listening and if you have a question, comments, TWIM-Microbe.tv. If you really like what we do, consider supporting us. microbe.tv/contribute. Michelle Swanson's at the University of Michigan. Thank you, Michelle. Thank you. I enjoyed it. Alio Shactor is at Small Things Considered. Thank you, Alio. Michael Schmidt is at the Medical University of South Carolina. Thank you, Michael. Thanks, everyone. And I am Vincent Racken-Ello. You can find me at virology.blog. I'd like to thank the American Society for Microbiology for their support of TWIM and Ronald Jankies for the music. This episode of TWIM was edited by Ray or Taygun. Thanks for listening, everyone. We'll see you next time on this week in Microbiology.