eLife Episode 16: Synthetic cells, antivirals and sex pheromones
n this episode of the eLife podcast we hear about reproducibility, drug resistance, cells without walls, gene transfer, interspecies signalling, and stem cells.
In this episode
00:39 - Cancer Reproducibility Project
Cancer Reproducibility Project
with Tim Errington, The Centre for Open Science
In 2012, the research world was rocked when scientists at Amgen announced they've been unable to repeat 47 out of 53 landmark publications that they've picked out as promising avenues for the development of new therapeutics. Now eLife is supporting an initiative called, the "Reproducibility Project: Cancer Biology", which is a collaboration between the Centre for Open Science and the Science Exchange to independently replicate selected results from 50 leading cancer biology papers. The aim isn't to conduct a witch hunt, but instead to identify the factors that might influence reproducibility more generally as Tim Errington explains to Chris Smith...
Tim - This initiative is to look and do an open investigation of how reproducible certain pre-clinical cancer research papers are. What that means is if I look at the results of a published paper, how likely is it that I can reproduce those exact findings if I do everything the exact way that the original authors did?
Chris - Looking at it another way, Tim. How unlikely is it that were I to take a cancer paper off the shelf today and attempt to replicate what's in it, that I wouldn't be able to do so?
Tim - That gets into some of the barriers that this project is trying to investigate, which is how much information are we actually sharing with our colleagues and with the public? How much methodology, how much data, how much of the actual results are being presented in any given paper? It's kind of analogous to cooking. If I want you to make a cake but I don't tell you all the ingredients and the entire recipe step-by-step, but I only show you a pretty picture of a cake at the end. It's really difficult to know how to recreate that cake.
Chris - The only certainty is you must be Michelin star chef.
Tim - Yes!
Chris - Do you think that there's an endemic problem in science? I mean, if one looks at the volume of material that's published, is there evidence that this is unreliable or untrustworthy, or that people are publishing things for the wrong reasons?
Tim - So, there's a lot of evidence that's suggesting there's a lot of excessive positive results in the literature. And I think what we're seeing is that all the incentive structure is about these results that show that something is occurring, suggesting that researchers are holding back some of the data that they're actually producing.
Chris - But you go further than that in announcing this study. One commentator has gone quite a bit further than that and I will quote. "Some laboratories publish one irreproducible study after another in high-impact journals, collecting data to support their intuition and paying little attention to whether or not the data truly support the conclusion" So it sounds like rather than just people cherry picking, actually, they're actively deceiving.
Tim - There's some evidence in the literature that does suggest that. And this project's intent is if somebody is holding back and only doing the experiments and showing the results that support their hypothesis and if we do a direct replication, how likely is it that we'll see the exact same distribution, the exact same result if we look at every piece of data?
Chris - How are you going to go about doing this?
Tim - We're going to take a subset of experiments in these 50 papers that we've identified and we're going to work with the original authors, and obtain all of that original materials, methodology to get a full description of what was exactly done and to hopefully have the original authors look it over and provide that input that is missing from the publication.
Chris - So, the broad aim of this study is to say, "We have identified 50 big papers in the field of cancer research. We are going to attempt to recreate the results using the methods and the approach taken by the authors of that paper to see whether or not our outcomes agree with theirs."
Tim - Yeah. And so we take it one step further. And I think this is the real kind of crux of the entire project which is if we do that across many experiments and many studies, we get a really rich data set that we can then analyse and ask the simple question, "Is this associated with, or not, the ability to reproduce those results?"
Chris - So say you do reproduce the result. How do you know a true result really is that positive and that you haven't just fallen for the same mistake that they made?
Tim - Hang on. That's a very good question. We designed these experiments in a way that at least tries to minimize that as much as possible for error that occurs in the replication. But, as you pointed out, if we reproduce a given result but we obtain the same materials, so we obtain that same imperfection. And thus, we obtain that same result that's flawed with that imperfection. This project can't necessarily uncover that.
Chris - And equally, the flipside of the coin. Does you failing to reproduce a result mean that the original result was wrong, or does it mean actually that you've just failed to do what someone else did?
Tim - I think what you just said the second time is the absolute right answer. It's just another experiment. It's another experiment that's trying to understand if you can reproduce the original result. It has no bearing on those results. If anything, I think it's exciting if you can't reproduce it because if you can't reproduce it then that means that all the variables that you thought were not important, might actually be important. That this effect actually is not maybe robust as you think it is but there's some underlying biology that is worth following up and understanding. Why did we not recreate what somebody else did?
Chris - And what is this project's ultimate goal? What are you hoping when you've been through these 50 papers and you have or haven't reproduced things? What are you hoping to achieve?
Tim - The main thing that we're trying to do is to create an initial data set that allows us to kind of hold the mirror to ourselves and say, "What are our current research practice and publishing practices that are having a potential influence on the ability to replicate something, and how might we change those?" And I think another very related aim for this project is to demonstrate that you can do science in an open manner and that we can do and make replications, something that we should probably do more often in the greater scientific community, that there's a lot of value in doing direct replications, especially when we're doing it for experiments that have very high therapeutic potential.
07:04 - The one and only
The one and only
with Elizabeth Tanner, Gladstone Institute
Most people tend to agree that multi-drug approaches, hitting the virus from several directions at once, is the way forward but as Elizabeth Tanner explains to Chris Smith, if you choose your target wisely, one drug may well be enough.
Elizabeth - One of the major challenges in using drugs to treat viral infections is that drug resistance tends to emerge very rapidly, sometimes within days, and one of the best ways of circumventing this problem is to use multiple drugs, usually three at a time to prevent resistance. However, there are many viral infections for which we don't even have a single drug, let alone multiple. Finding multiple takes a lot of money and many years, and so we wanted to figure out if there was a way to use just a single drug to simultaneously stop the growth of the virus, cure the infection, and prevent the emergence of drug resistance.
Chris - Why do some viruses - and particularly things like HIV - why do they have such a high tendency to become resistant to these drugs?
Elizabeth - So, the replication of the virus genome, especially when it's an RNA virus - not a DNA virus - is very error prone, there is a lot of mistakes.
Chris - Is it possible to exploit that fact and come up with drugs that, rather than be hampered and hindered by the fact there are so many variants - one of which could be resistant to the agent - is there a way of exploiting that to make a drug that works very well in that setting?
Elizabeth - Yes. So, there are some virus proteins, that as part of their function, must form a complex. So, there's something called the capsid on the virus and that's sort of the outer shell of the virus and it's made up of 60 different copies of a protein. So, you can imagine those 60 different copies can come from 60 different variants. If one of those proteins or even maybe a handful of them come from a drug resistant variant, only a small fraction of proteins in the actual structure will be drug resistant. The majority will be drug susceptible and will still bind the drug and they won't function. So, overall, that entire complex is likely to be non-functional, even though there are drug resistant proteins in there.
Chris - In other words, it should be possible to use your strategy and make a single drug which despite the fact you will get some resistance, it should nonetheless - if chosen carefully in terms of its target on the virus - should be able to bring down the entire shooting match.
Elizabeth - Exactly!
Chris - How have you modelled this, and how have you tested it to see whether or not the theory does play out and it does deliver?
Elizabeth - We set up an infection model like would be in a patient - only we used mice - in this case, they're transgenic for the human polio virus receptor. And we infected them with polio virus and then we treated them with a drug, actually, the drug that's currently in human clinical trials for polio virus. And it inhibits the capsid which has a number of proteins in this structure, and when we treated them with the drug, we were looking to see if over a number of days we could see drug resistant virus emerging from the infection. With the capsid inhibitor, even up to 7 days, drug resistant virus never emerged, compared to a control drug that we didn't think would work in this way. In that case, by 4 days, we could see a significant increase in the amount of virus that's resistant to that other drug.
Chris - Do you nonetheless think that there's a risk of getting some kind of compensatory mutations where, given enough time, the virus will evolve some kind of novel effector mechanism which would, in some way, thwart the blockade of your drug?
Elizabeth - Yes. It's theoretically very possible. For example, if there was a way for the drug resistant variant to only form a complex with other proteins from its own types. So, drug resistant proteins only associate with other drug resistant proteins then yes, it can escape. But that would have to happen independent of the mutation to also make it resistant. So, it's sort of like having multiple drug therapy because the virus will have to simultaneously gain two new abilities. One, it will have to become resistant to the drug, and two, it will have to be able to identify only other proteins like it.
Chris - Are these results therefore, generalizable more broadly, not just polio but many other agents. Could we extrapolate this and take the modus operandi that you have disclosed, showing that this is achievable and say, "Right. If we know what the sort of target is in the virus that we're looking for, we can go and find something similar across the board in different agents and that's the one to go for and we can have mono-drug therapy."
Elizabeth - Yes. Other people in the Kirkegaard lab are actually working on other viruses and actually totally different targets. Someone's working on Hepatitis C, and someone else is working on DNA virus, and they have very, very promising results that you can extrapolate to those viruses.
12:15 - Off the wall
Off the wall
with Jeff Errington, University of Newcastle
Many bacteria are clad in a tough cell wall that provides protection and also stops the cell from bursting through osmosis. The cell wall's also critical for bacteria to divide correctly, or at least that's what we thought. Because now, Newcastle scientist, Jeff Errington has found that by growing bugs in antibiotics for an extended period and then deleting a small clutch of genes, it's possible to produce strains called L-forms that lack a cell wall but can, critically, still reproduce, albeit very slowly. This could be the way their ancestors did it early in evolution. But here in the present, it could now hold the key to producing self-replicating synthetic cells.
Jeff - The cell wall is incredibly important for bacteria and one of the things that it does is to protect the cell from bursting due to the osmotic pressure inside the cell which pushes out on the cell membrane. So, normally bacterial cells rely on the wall for survival. And one of the reasons why antibiotics work so well is that they interfere with the synthesis of the cell wall, and in conditions where a cell wall isn't formed, the cells basically explode. So, we have to establish conditions where we could protect the cells from this sort of explosive event. And what surprised us a little bit to begin with was that the cells that had no cell wall would simply sit there without growing.
Chris - Were these cells actually viable? Could you tell whether they were alive or were they just sitting there not doing anything because they were dead?
Jeff - They're actually alive. You can infect them with viruses, for example, and they'll support the growth of these viruses. They simply don't grow and actually, we don't yet understand why they don't grow. But what we discovered was that if we select for certain mutational events then the cells would grow.
Chris - What are the mutations that are necessary to make these bugs propagate then, and how do they make a difference?
Jeff - Well. The key mutation that we discovered led to a single kind of event that then forced the cells to produce excess amounts of cell membrane. So, they result in increasing the surface area of the cell relative to its volume and surprisingly to us, it turned out that that is sufficient for these cell wall division bacteria to grow.
Chris - Do you have any idea why just having an excess of cell membrane would have that dramatic effect?
Jeff - This project has taken us to very interesting places around various aspects of biology but also to biophysics. And it turns out that there are biophysicists who are working on trying to mimic how they imagine very primitive cells back in the early days of life on the planet would have grown and divided. And the biophysicists had already been thinking that by increasing the surface area of a simple membrane bag, that that might drive the proliferation of those cells.
Chris - That's very surprising, isn't it? That these bacteria should still be able to divide in this way because is there not a very highly evolved mechanism which involves various skeletal elements inside the cells that enable the bacteria to make sure that each of the daughters have got copies of the chromosomes and so on. So, they must be ditching that mechanism in order to still be able to divide in this L-form that you've created.
Jeff - That's right. Probably the biggest surprise that we got very early on - the whole of the cell division machinery that bacterial cells normally use, becomes completely dispensable in the L-form cells. So, we can delete many of the genes that are normally required for division and it makes absolutely no difference to the proliferation of the L-forms.
Chris - Putting this all together then, you have identified the genes which appear to be necessary for you to deactivate. This in turn, leads to a loss of the cell wall, an overproduction of cell membrane, and this restores the ability to these cells to be able to divide by a mechanism as yet we don't really understand. Why is this important, though? And can we do this with any bacterial cell if instead of bacteria subtilis, Staph. aureus, for example. Could we do the same thing with them?
Jeff - The basic principles of L-form biology that we've uncovered seem to apply also to a very wide range of bacteria. In the new paper, we show that Staphylococcus aureus and E. coli seem to be able to undergo very similar L-form switch. And all of this is important for several reasons. There are many clinical case studies implicating L-forms in all kinds of diseases, many of a chronic or persistent nature. One of the reasons why L-forms might be implicated in those diseases is that antibiotics like penicillins are absolutely ineffective against L-forms.
Chris - What about another potential implication and, perhaps, application which is that have heard from the same crew, Craig Ventner, who sequenced the genome in recent years are now saying they made semi-synthetic life. Could we also use this sort of approach in making synthetic microorganisms or just synthetic cells in general?
Jeff - We're very excited by the possibility of using what we've learned about L-forms in that sort of context. It seems to me to be quite daunting to imagine how you could build a machine that would synthesize the wall in the correct configuration and then a division machine that would bring about the accurate division of walled cells. But of course, we now know with the L-forms that we can have pretty normal, in a sense, cells growing in the complete absence of the whole cell wall synthetic machinery and the whole of the cell division machinery. So, that makes the prospects of generating an artificial cell, I think, rather more straight-forward.
18:02 - Spreading far and wide
Spreading far and wide
with Seth Bordenstein, Vanderbilt
Genes are a precious commodity and if an organism evolves the genetic recipe that enables it to do something a bit special, other organisms often, through various mechanisms, acquire the same genetic know-how so that they too can benefit. We had thought that this gene swapping occurred mainly between just a few species or close relatives at a time. But now, Vanderbilt scientist, Seth Bordernstein, has discovered a gene that began life in bacteria infecting viruses, and has since been traded right across the tree of life - from archea, to fungi, plants, and even people.
Seth - So, this gene is part of a lysozyme family. Technically, it's called the GH25 muramidase and it's designed to cut open the cell wall of bacteria. And particularly, the main component of the cell wall of bacteria which is called peptidoglycan. These molecules are actually used by bacteria when they divide and they make two cells, they cut open their cell wall. But other organisms have hijacked these kinds of molecules to kill bacteria. Viruses will use these types of molecules to break open their bacterial cells that they inhabit. In this case, we ended up finding that the other domains of life, Archaea, in particular, were using this particular molecule to also kill bacteria and this is something that hadn't been discovered.
Chris - Now you've got something of a chicken and egg situation though, haven't you? Because the big question must be, "Well. Which of these came first?", "Where did it start?", and then, "Where did it get into all these different parts of the tree of life along the evolutionary pathway?"
Seth - What's so clear in our case is that the only archaea that has this particular gene and makes this lysozyme lives at the bottom of the ocean near boiling vents and no other archaea in the sequence data bases have this gene. So, it's absolutely clear that this gene was horizontally transferred from bacteria, where the gene is more common, into archaea. The same thing can be said for the plant version of this gene. There are no other plant genomes on the data bases that have this gene. More exciting to us was the fact that we could guess what types of interactions led to the exchange of these genes. So, it turns out that bacteria that live with archaea at the bottom of the ocean were the likely source of these genes, and we can show that through sequence comparisons. And soil bacteria were the likely donor to give these genes to plants and fungi.
Chris - Did the viruses then steal it because they were naturally infecting the bacteria and ended up with those genetic elements finding their way into the virus and they were a useful effector. So, the virus clung onto them.
Seth - That's probably correct. In the bacterial world, viruses are constantly exchanging DNA with bacteria and vice versa. There's almost a lawless exchange of genetic information among bacteria and viruses. So, this is a very common phenomenon - horizontal gene transfer. It's not as common in archaea and eukaryotes. And moreover, what we found was the same gene had been transferring independently across the diversity of life. That's really what's new here, is sort of opening up the limits of horizontal gene transfer.
Chris - And what are the implications of this? Do you think that this is a lot broader and a lot more common than we had anticipated? And how might this inform our understanding of horizontal gene transfer more broadly?
Seth - The biomedical application is really clear here. So, we have characterized the first antibacterial gene in archaea and we think this work will open up and energize the pursuit of antibiotic discoveries in archaea which have arguably been a vastly under-tapped source for new antibiotic molecules. And as we race against the antibiotic resistance problem, well, we need all the help that nature's willing to give us. And so we want to pursue this thing a little further with the gene that we discovered. The broader question, the basic science question of how extensive is horizontal gene transfer? We're still learning and we were surprised to find the results that we did because we thought everything had been done on horizontal gene transfer that could possibly be done. But when we looked at this unusual case of the same gene moving between all types of life, we realize horizontal gene transfer was pushing its limits, at least in terms of our knowledge. So, we suspect it will be more to come. We suspect that antibiotic genes will be commonly transferred between different groups of life because of the universal pressure to deal with bacteria, either holding them at bay or turning them into mutualists, or just killing them outright to defend themselves against bacterial pathogens.
23:00 - Fatal attraction
with Jessica K Cinkornpumin, California State University
A case of fatal attraction now, and a gene that enables a nematode to track down a species of oriental beetle using the beetle's sex pheromone. Intriguingly, the gene that the nematode uses also protects it from what is otherwise a toxic effect of the pheromone. Jessica Cinkornpumin is at California State University, Northridge.
Jessicca - The nematode we study is Pristionchus pacificus, and it's specifically attracted to a sex pheromone released by the beetle called, ZDTO. We wanted to study how this interaction actually occurs in a genetic level. We know that the chemical is released by the beetle and the nematode senses it with these neurons that are in their head, and when they sense this pheromone they actually stand up and they are able to actually hop on to the beetle and live on the beetle until the beetle dies. And this is when they actually emerge again and are able to reproduce.
Chris - The big question then is - Well. How are they actually doing that sensing?
Jessica - We don't know. So, we try to identify this gene and how we did it was, since we know that the nematode is very attracted to the pheromone ZDTO, there's a screen to identify Pristionchus pacificus nematode that doesn't sense ZDTO specifically.
Chris - How clever. I'm anticipating then what you're going to say to me is that by looking for a nematode that doesn't do the sensing, you're assuming there must be something missing genetically from that nematode strain. And if you can find out what's missing, that tells you what the gene is that does the sensing normally.
Jessica - Yes. And I think I want to emphasize that the reason why we did the screen this way is because we want to find a gene that's really responsible for seeking hosts and not a gene responsible for just sensing in general, so that we can study this relationship.
Chris - And how did you know that that's what you were going to get when you did this study and you weren't just exploring genes that might abolish the behaviour in some other way?
Jessica - So, we did take the nematode that we study that is missing the attraction to ZDTO and then tested it with all these other pheromones or signals that it's also attracted to, and we found that it's really, really specific to ZDTO.
Chris - Which tells you that gene must be responsible for picking up that signal?
Jessica - Yes.
Chris - Did you then do the obvious experiment which is to take the missing gene from the nematodes that have it or the gene product and put it into the one that doesn't, and prove that you can restore that behaviour?
Jessica - That's exactly what we did. We made a functional copy of the gene and this working copy was then injected into the nematode that lacked it. After it grows and produces its eggs, these eggs actually had the gene rescued.
Chris - Thus, proving that it really does work?
Jessica - Yes.
Chris - Why is this important? Because, yes, it's nice you've got this gene; you've identified how these nematodes track down this beetle target using one of its own pheromones. But what are the implications of this finding, apart from its academic interest?
Jessica - Well, I think, especially since we identified this one gene to do two things. One, sense; and one, protect it from the same attractant, this pheromone ZDTO. This indicates to us that there are many genes and interactions we still don't even know, and when we understand this interaction much better, we can actually use this information that we know to study how parasites also find their host and use that in a, maybe, agricultural sense and use it to protect agriculture from the harmful effects of these pests.
Chris - When you say that the gene can protect the nematode from the beetle, explain how that works a bit. Would the pheromone normally be poisonous, or in some way destructive to the nematode then?
Jessica - Actually, this is very new to us. We actually didn't expect to find this so we are really interested into looking into this more, but as we know so far, nematode that doesn't have ZDTO and is exposed to the pheromone, gets paralyzed by the pheromone.
Chris - Do you have any idea how the gene is doing the protection?
Jessica - We think it's a lipid binding protein that actually responds to the pheromone. So, we either think it actually carries a pheromone to a receptor that does some downstream effects - either protecting or sensing since it does two things, or that it actually works in a function of removing the pheromone from the receptor so that it can continually have another pheromone bind the receptor. And we think somehow, this binding and the downstream effects is how it will protect or sense the ZDTO.
27:57 - Niche work
with Todd Nystul, University of California at San Francisco
Stem cells are what repair and replace tissues as mature cells die or are lost. They do it by dividing in two. But how does one daughter cell know to migrate away and become part of the surrounding tissue while the other remains as a stem cell? The answer might be to do just with the shape of the cells. Todd Nystul has found that when epithelial cells divide, the daughter is pushed out of the stem cell layer and then develops a bottom, sides, and a top, unlike its parent which lacks a top. This top might allow the cell to respond to other signals in the local environment, triggering it to move away and specialize.
Todd - One of the ideas in the stem cell field about how these differences could be specified is that stem cells reside in a specialised micro-environment in the tissue that we call a niche. The niche can be made up of other cell types that are nearby. It can be simply be a consequence of the shape of tissue, and the idea is that this external environment provides signals that act on the stem cell and cause it to remain undifferentiated. Whereas, daughter cells move away from that niche and escape those signals, allowing them to go on to differentiate.
Chris - The daughter cell migrates away, escapes the influence of whatever this local factor is, and that releases it to then follow any other local signals or anything intrinsically programmed into it that it inherits from its stem cell parent?
Todd - That's correct. So, in this case, we were considering the influence of a signal produced by the niche, called EGFR, ligand produced by niche cells and can act on proteins that are present on the surface of the stem cell and then initiate a cascade of events inside the cell. But what is the consequence of activating these proteins on the surface of the cell, and how does that translate to a difference in what makes the stem cell a stem cell versus a daughter, a daughter?
Chris - How did you study that, and in what system?
Todd - We chose to use the fruit fly ovary tissue called the follicle epithelium. Each little unit of the ovary has exactly two epithelial stem cells in them, and we know where the daughter cells that they produced go as they are produced. And so when we look at a tissue, we can really with single cell resolution say, "There's the stem cell! There's the daughter that was produced yesterday. There's one that was produced two days before that," and ask very precisely what the differences are between them.
Chris - And what did they reveal to you?
Todd - We had started with an observation that the stem cell has a different shape than the daughters that it produces. Epithelial cells in general are known to have a very characteristic shape when they take on their differentiated or functional form, they look very rectangular. They have a bottom and a top, and sides, and each of those surfaces has a different function for the tissue. We noticed that the stem cell had a bottom and it had sides, but no top, and so that made it have more of a triangular shape rather than a rectangle like we would expect from an epithelial cell. That was curious to us because we know that that top is a very important part of the cells functions - the place where the cell can receive signals from the external environment. And so, we started to look into the literature about what was known about signals that regulate cell shape and there was one paper, a little over 20 years ago, that suggested to us that this signalling pathway, the EGFR signalling pathway, could be important. So, we could take out the receptor, specifically from stem cells. When we took it away, the stem cells completely lost their shape and they didn't have any recognizable shape at all, and nor did their daughters. But if we took it away from the daughter cells, there was no effect at all.
Chris - That's fascinating on two levels. Because on the one hand, you have something which is controlling the appearance, morphology, of the stem cell itself, but then you've got something which appears to direct the behaviour of the daughter cell based on what the parent cell was doing, too.
Todd - Yeah. That was a big surprise for us.
Chris - That daughter cell is out of range because it's moved away from where this signal must be, to start with. So, there's something going on downstream of the initial signalling effect, isn't there?
Todd - Exactly, yeah, and that was a very big surprise for us. Many of the signals that we study which act on a stem cell tend to act just on the stem cell to promote the stem cell identity. But here we had a signal that was not only promoting the stem-like shape of the stem cell, but also by doing so, was setting the stem cell up to produce a daughter cell that would now take on a new shape. And so, one way that we're thinking about this is that the development of a top to the cell could allow those daughter cells to interact with other cells in the tissue and receive signals from the tissue that tell them to go on to become a functional cell in a tissue. Whereas, stem cells, which lack that top, will not receive those signals and will, therefore, remain undifferentiated in that way.