and then John is going to tell you about future directions that are very exciting that the field is going in so let’s get started and ask what is synthetic biology if you get any you know self-proclaimed group of synthetic biologists in a room they will never agree on a definition so instead of trying to nail one down I’m actually going to ask where did synthetic biology come from because i think if we think about what led to what we now call synthetic biology will understand what it means and so the answer to where did synthetic biology come from is peas peas you ask the synthetic biology is in the P Dan well not quite so there was a gentleman named Gregor Mendel who was an Austrian monk and he loved gardening but unlike you I who might have a vegetable garden you know grow some herbs and spices he was kind of obsessive compulsive about his garden and he groupies and he loved them so much that he wrote down everything that he could about these peas how tall the plants are what color their seeds are what shape the seeds are so this obsessive-compulsive behavior actually led him to notice certain things about the peas and that is that these traits that they have when they go from generation to generation these traits are passed on in a predictable fashion and this observation when contrary to the thinking at the time which was that if you make two plants together their traits like color will just blend so if you have a yellow and a green plant and you cross them you’ll get a yellowish green plant and mr. Mendel’s observation was that that’s not actually the case so if you have a yellow pea and a green pea in fact that yellow color is what he called dominant so if you cross the yellow and a green you’ll actually get yellow and so in this example if we had a yellow pea that in its kind of family history also had a green pea that yellow will dominate but if we again cross it with a green pea you can predict the inheritance of these color traits and we’ll get a predictable ratio of yellow to green peas this laid the foundation for the study of inheritance and really the birth of the modern the field we today called genetics so that’s great we now understand that traits are heritable in predictable ways they’re discrete units of inheritance but how how are these traits inherited and for that one of the major breakthroughs came from a gentleman named Thomas Morgan who worked at Columbia University and Thomas was not so much of a plant guy he was instead a fly guy he loved flies so much that he generated a room full of flies at Columbia University that later became world-famous and me personally I don’t like flies you can pay me enough money to be in a room full of thousands of flies let alone spend my whole PhD in that room but you know different strokes for different folks Thomas Morgan really liked those flies and he really got that question of how our traits inherited and what he was able to observe is that flies their traits are stored on particular locations of their chromosomes so people we for example have two copies of all of our chromosomes one from our mother one from our father and so we have two copies of every gene and he was able to determine that one of those copies for example the dominant yellow example I was talking about earlier is passed on again in a predictable way and it happens via these chromosomes so he was able to predict how flight rates were passed on this was around nineteen fifty additionally he went further and looked at the structure of these chromosomes so here I’m showing you this X shaped thing which is what a chromosome looks like when cells are dividing when the chromosomes are really densely packed and he went beyond just saying that these traits are on the chromis but actually figuring out where on the chromosome they’re stored and today this concept r is known as genes and by measuring these locations he was able to more tightly define where these traits are and today we still measure these distances on chromosomes in units called Morgan’s named after Thomas Morgan so he was able to map out where the purple I trait is stored on a chromosome where certain wing shapes body shapes and so

on are stored but this left one more big mystery chromosomes are made of proteins and DNA and the thinking at the time was that proteins probably are what what are responsible for transmitting these traits people were more familiar with protein they had studied proteins extensively they knew they did things they had functions so it made logical sense that they were responsible for transmitting traits DNA was not a household word like it is today then not too long thereafter Alfred Hershey and Martha chase working at Cold Spring Harbor did a very elegant experiment using viruses and these viruses really consist of only two things a protein structure and then they have their genome their traits encoded as DNA which they carry with them and they’re very elegant experiment they were able to track the protein or the DNA and by labeling either the protein or the DNA and then taking these labelled viruses and infecting cells they were able to track what the protein and DNA are doing over time and so they let these cells grow for a long enough time that the original viruses were no longer around and found that the only thing that remained was the labeled DNA and not the labeled protein thereby they demonstrated that it was the DNA that was responsible for passing on the traits in this case the viral infection so now we know that traits are heritable in the street units that specific regions of a chromosome carry these traits and further that the DNA on the chromosome is what’s responsible for transmitting that information shortly thereafter a big breakthrough instead in the study of DNA happened when James Watson and Francis Crick who are credited with this discovery published a model for the structure of the DNA molecule and they proposed that DNA forms this double helix structure understanding this concept was very key to understanding how DNA interacts with everything else in biological cells an additional crucial piece of information came from erwin chargaff he was a biochemist and he studied what DNA is made of which are base pairs essentially biochemical building blocks and by studying the DNA from very basic organisms he found that there are these four building blocks and that they occur and again very specific ratios so adnan occurred in the same ratio as thymine and guanine occurred this is the same ratio as cytosine and this insight was actually also key for Crick and Watson to understand how the base is paired in their proposed double helical structure so we know that the DNA is responsible for these traits and from this DNA eventually we get protein there’s also another intermediate called RNA which is another type of molecule like DNA but it wasn’t clear back in the day what the relationship between these three classes were can you make protein from DNA can you make RNA from protein all possibilities were initially considered Francis Crick of Crick and Watson in 1956 published a paper laying out what we now call the central dogma of molecular biology any proposed that from DNA you make RNA and from RNA you make protein and this basic relationship was key to understanding so much of what happens inside of cells at the molecular level at the time when he published it was still unclear whether you could go back from RNA to DNA and it was even questioned whether you can make protein straight from DNA today we don’t believe that to be the case we know the order as DNA to RNA to protein so are there any questions up to this point yes that’s a great question so oh so the repeat the question it was how did Morgan know we’re on the chromosomes these traits were located and so there is a process in the fly breeding where you do very specific crosses of flies with the trade and without the trait that you’re looking for and then by

analyzing the chromosomes that result from kind of their offspring you can actually tell which regions were responsible for those traits that’s the short version i’ll tell you the long version in the break if you’d like any more questions great so now we understood that DNA is key for inheriting traits and we understood something about its structure so we wanted to move from observing this to actually manipulating this and this is where the beginnings of synthetic biology could be traced back to so a class of proteins known as endonucleases enzymes that cut DNA in very specific ways were discovered and using these enzymes scientists were able to make predictable cuts to DNA which formed the foundation for then manipulating that DNA further a process called polymerase chain reaction that I won’t elaborate on right now was developed to copy DNA so now we had cut copy and also paste so all your basic word editing functions were there so with cut copy paste you can already do a lot and so what did people do with this knowledge well they asked could we do something useful with it for example making insulin so in the early 20th century insulin which is used to trace treat diabetes was terribly expensive you had to isolate it from animal pancreases there was the desire to make insulin much more cheaply so we think again about this central dogma that Creek proposed well we want to make insulin protein and we know proteins are encoded in the DNA so we found out what gene was responsible for coding for insulin maybe we could cut and paste that and use that somehow to make insulin and that’s exactly what people did they cut out the insulin gene they pasted it into another piece of DNA and then and this is key they use that cut and pasted insulin gene to put it into bacteria and the nice thing about bacteria is they grow fast and they can make a lot of protein so unlike growing up a whole animal and then harvesting a little bit of insulin you could just grow bacteria in huge vats isolate the insulin from these large amounts of bacteria which are just churning out insulin day and night and now make insulin much more cheaply and make it available to diabetics around the world this was the first huge success for what we now call static biology insulin went from an expensive medical treatment not available to everyone to a commodity we don’t even call it a medication anymore it’s sold as a commodity like paper or timber is so it has been incredibly successful and that was just the very first time people try to apply the synthetic biology mindset to a problem in this case making insulin so I hope I’ve showed you that synthetic biology if we put it all together is really just really really advanced botany we went from studying peas and understanding that their traits were inherited in predictable ways and then Thomas Morgan showed us that these traits are inherited in specific regions of chromosomes and then we developed techniques to cut and paste DNA and and for that we use that to solve a very pressing problem of making insulin so that’s great but what other problems can we apply can we solve by applying these tools can we make other medications more cheaply beyond insulin can we fight the spread of malaria and other similar infectious diseases and most importantly can we make better peace and my answer is yes we can and so on the rest of this talk you’re going to hear about some really exciting work that’s addressing all these problems all of these things are happening right now today thank you in any questions god this is what with it in that year in this unity what was the nature of scientific collaboration so abroad it’s interesting of course science was a smaller world there were fewer people involved actually in reading up for this talk again so chargaff who you recall established those ratios between the base pairs he didn’t like quicken Watson he met them and he actually told him about their idea and then they got the Nobel Prize for the double helix structure and he was left out so he actually went around

complaining that everybody that you know they left him out took credit so it was a collaborative and even people who didn’t really like each other did communicate however it’s always a question of who gets the credit and that you know problem still exists today there’s one more question now this for example is a DNA love you a putting it into a bacteria in the bacteria national manufacturing plant to produce the of the Jews video compound characters today when you make that modification the only sense of how that affects the light and reproductive capabilities of bacteria is it is that something usually possess a huge problem or is that easily certainly so first let me say that justin is going to go into way more detail on kind of this process but it absolutely represents a burden to the bacterium and this is a challenge that needs to be overcome by science and and engineers okay so oh sorry what so Dee asked me size of 40 vr missions question so I think it really it’s a more of a case by case basis so this is things people are struggling with today there are chemicals that are available that are made through chemical synthesis and that people are now thinking about making biologically and oftentimes it comes down to how well can you make the biological organism make that and at what cost you might be able to make it but it might be more expensive than making it chemically so on a case-by-case basis people evaluate does this make sense okay I’ll turn it over Justin hi guys I’m Justin I’m currently a PhD student finishing my third year in the lab George church in today I’m going to be talking about how we’re engineering life to make renewable chemicals to improve our health and the environment so this part of the talk is actually split into two parts so first is about a bad bug and an ambitious idea that led to the explosion of the field of metabolic engineering and this work is actually what inspired me to pursue a PhD in synthetic biology so hopefully you guys are also a little bit inspired or the very least learn something new so we’re going to be covering the main ideas behind metabolic engineering as well as the challenges we’re currently facing the field in the second part of the talk we’re going to be going over some new technologies that are being worked on now that will really push the boundaries of metabolic engineering and hopefully make a direct impact in our lives in the near future and this is going to include some research on developing biosensors i’m personally working on for my thesis and so to begin with plasmodium commonly known as malaria is a genus of single-celled parasites it’s carried by mosquitoes and affects humans and other animals and it’s an extremely infectious and dangerous disease it causes all kinds of terrible side effects and in many cases death in 2010 there were over 200 million cases worldwide and an estimated 700,000 deaths and it’s particularly widespread in the developing world so there are currently many strategies employed to deter the spread of malaria and treat infected patients and these include the use of bed nets and sprays and repellents such as DDT there are some drugs such as chloric win and there’s also some kind of further ideas such as gene drives that jon is going to go over later in the talk but the state-of-the-art right now is artemisinin-based combination therapy so artemisinin artemisinin is a highly efficacious drug derived from the plant artemis iya annua or sweet wormwood and it’s kind of interesting because this plan was originally discovered through the use in traditional Chinese medicine but it’s sense proved to be a potent modern medicine as well so there are a few major problems with artemisinin supply that lead to instability in providing

enough treatments worldwide it’s difficult to grow can only be grown in specific places and actually takes two years to cultivate so these have led to shortages in the past which is a really big problem there’s also a limited number of suppliers which has turned the medicine into a commodity with large price fluctuations and suppliers hoarding material and waiting for prices to increase so as I said before it’s a very big problem for the developing world so because these problems people were looking for another way to supply artemisinin for malaria treatments so to quickly recap some of what Dan just went over let’s recall that life depends on DNA to code for proteins molecular biology techniques have a lot of snip you lady na in useful ways such as expressing useful proteins like insulin one thing we didn’t go much into you was what proteins do within the cell so one of the important functions is to seyn the metabolism of cells so metabolism is basically the chemistry of life so proteins help make the chemicals that cells need to live and grow such as caffeine ADP and ethanol and proteins that carry out chemical reactions are also called enzymes so metabolisms and cells are incredibly diverse this is just a tiny snippet of some of the proteins involved in one specific pathway in a cell and we’re constantly researching more about how organisms break down and build up different chemicals as part of their metabolism and we have much more to learn if we’re easy to understand we’d probably have a much better idea on how to get super skinny or super buff really quickly but we’re not quite there yet one day so with the advancements of molecular biology techniques people began to wonder what if we could engineer the metabolisms of organisms to make useful chemicals so the basic premise is not really a new one and some of you have probably benefited or suffered from this actually the classic example is yeast converts sugar into ethanol which what makes it bubbles in beer they’re also important for making bread and it’s what makes bread rise so the idea behind metabolic engineering is that we can use the same process to develop other useful chemicals such as fuels flavors fragrances or in this case pharmaceuticals such as hardness in in so to do this genes to make a chemical Pinterest like those found in the art of Mesilla plant are moved into a simpler easier to work with organism like baker’s yeast but doing this is not always easy and there’s actually a few problems that we need to overcome so the first problem is identifying what genes carry out the chemical reactions of interest we know the Artemisium plant produces artemisinin but how it turns out to be a really difficult problem because there are so many genes in a given organism and there are many different techniques to find out which ones are relevant but they’re also often time consuming and take many years to discover the second problem is that organisms and proteins do not always have the exact traits we want them to have so we look for we have to look for variants of them that do or create new variants that have the desirable traits one example this is that people hate well most people hate seeds and watermelons most sane people hate seeds and watermelons so farmers found ways to grow seedless watermelons and the reason that this is is that enzymes and organisms are not naturally optimized for producing chemicals sort of enzymes are optimized but they’re optimized for growing and sustaining life as you might recall from such scientists as Darwin archaean Malcolm’s in Jurassic Park so the problem here is enzymes are not optimized for having maximum activity in a Cell nature’s idea of optimal is not the same as a metabolic engineers idea of optimal this makes sense because oftentimes an enzyme might be producing an intermediate that the cell doesn’t need large quantities of or it might even be toxic to the cell so to produce more of our desired chemical using metabolic injuring often need to do optimizations on these enzymes so we can take this idea and use modern techniques to find useful mutations there are two common ways of generating these useful mutations first as a subject cells to something that mutates DNA such as UV light radiation or chemicals such as benzene or sodium azide you guys might have heard of these referred to as carcinogens because cancer often arises for mutations in the DNA this is often represented by this little cartoon lightning bolt a second more targeted technique is the use of error-prone PCR so Dan previously mentioned PCR as a method for amplifying DNA by replicating it many times within a test tube the process goes something like this we have DNA you prime it for replication with another small piece of DNA and extend the primed DNA with an enzyme called DNA polymerase so this is done cyclically where the DNA polymerase extends and replicates the DNA so you get many

copies so error-prone PCR has the same process as if we use a polymerase that has had its proof reading ability removed so this leads to errors accumulating as the replication occurs and this creates genetic diversity so using these two techniques we can generate many variants of enzymes and test them for our desired activity such as enhanced production of a chemical the third problem is that even after we combine the genes from one organism into another to Bruce a new chemical so just hard to miss it in we still have a lot of work to do that’s because as you’ll recall in organisms metabolism is extremely complicated we need to balance the fluxes so we don’t accumulate high levels of toxic intermediates we also often up to make sure that they’re not enzymes that interact with the new chemicals you’re producing in the cell um so you might have to remove off-target interactions and this is all very difficult and time consuming and requires up-regulating or down regulating the expression of different genes including those that you added to the organism or that are inherent in the genome of the organism that you’re using and sometimes you have to remove remove genes completely so it was facing these challenges that the synthetic artemisinin project or the art project was born in 2001 with the goal of using yeast to produce artemisinin at the time the idea of using a simple microorganism to produce something as complicated as artemisinin was an extraordinarily ambitious project so who was crazy enough to try this this person turn out BJ kiesling who was a professor of chemical engineering at UC Berkeley he also wasn’t alone with us ambition because he had to have another equally crazy person to support him the project so luckily for him there was an extremely generous and wealthy person by the name of Bill Gates who funded the project in 2004 through his fluent philanthropic excuse me foundation so Jane is live were able to successfully identify the key genes in Artemis in a pathway port them to yeast which I said before is used commonly to make things such as bread and beer optimize the enzymes and then balance the the new pathways the new metabolism to increase the amount of chemical made so the C strain developed is able to take glucose a common sugar they can grow you can get glucose from different plants and use it to make artemisinin now so the drug producing strain has actually been licensed to sanofi-aventis and the synthetically derived artemisinin is now being used it for treatments around the world today as we speak so this work was a tremendous amount of effort but it’s pretty amazing that you can the synthetic artemisinin is being produced through fermentation this totally novel process to make a real world impact with hundreds of billions treatments currently happening so quick recap metabolic engineering is used to make useful chemicals which is awesome but it’s difficult you need to identify the useful genes enzymes need to be optimized and then metabolisms need to be balanced secondly metabolic engineering was successful used to change the way the world produces an important anti-malarial drug and this is just a huge success story for synthetic biology so I can take a quick break for any questions that you guys have about this for the artemisinin pathway I don’t have the exact number but i think there was something like 15 max has a yeah thumbs up about 15 inch between 10 and 15 which doesn’t sound like a lot but I mean that’s the final product they end up having 10 or 15 new ones but there’s a lot of testing that goes on so you have one enzyme and doesn’t work quite right so you find something that’s similar and then you test it for activity it’s just a very iterative process yes extremely cheap and it’s also because it was founded through the bill and funded through the Bill and Melinda Gates Foundation the deal they have with sanofi aventis is to keep the prices down said it together using like 20 million doses that they need beautiful yeah what has to happen to get to the right so I don’t know the exact details oh sorry the question was why are we only producing a hundred million doses instead of two full 300 million so I don’t know the exact details of the licensing but I know there was some concern about displacing the farmers that are already growing the Artemis and implants are the Artemisia plants so that might be part of why they’re limiting it part of it could be the amount of fermenters that sanofi has allocated to artemisinin and other

things like that any other questions Oh you showed how many isolated so there’s so how do we isolate the artemisinin or the chemical produced by the yeast and we knew metabolic engineering so that’s actually a big problem when you’re starting a new metabolic engineering process because if the chemical is trapped in the yeast then you have to spend a lot of time energy to break open the east and then isolate the chemical of interest so sometimes there’s a lot of engineering that goes into making a new yeast that produces a chemical to actually have pumps that export the drug that you’re interested in making so that’s one technique others are a lot of times these chemicals when you produce such a high amount they just diffuse out because there’s so much being produced there’s also things you can do like add what’s called an overlay to the media that the east are growing in and this will kind of soak up the chemical interest because it’s more soluble in the overlay than in the East media yeah pumps proteins that the yeast express for the purpose of moving chemicals out of the East cell cool so where are we going now I very briefly touched on it and the art project will extraordinarily worthwhile was actually a tremendous effort that took the better part of a decade and many millions of dollars to complete so obviously we can’t use the exact same approach for everything but some very smart and innovative people with a lot of capital are still working on this both in academic labs as well as in commercial corporations so these are just a few of the organizations in the field of metabolic engineering now so there’s obviously a lot of optimism that metabolic engineering is a viable approach that can be improved drastically so going back to the three main problems genes enzymes and pathways there are a few key emerging technologies that are going to help face these challenges so these include DNA synthesis so we can more easily engineer larger pieces of DNA as well as create more variants more rapidly will allow us to test genes and optimize enzymes DNA sequencing is also being used to gather information about organisms which will yield information about genes and pathways so this picture shows a line this line is the rate at which computing power has been increasing so you can imagine how much faster computers and phones have gotten in the past ten years but this line is the rate at which DNA sequencing is increasing so you can definitely imagine that there’s gonna be a lot of new possibilities in the new future that are being opened up by the rate of increase of DNA sequencing bioinformatics and other modeling developments will also allow us to better predict what genes have useful activities and will also allow us to better optimize and balance pathways new technologies to quantify metabolites will also address all three problems and lastly there’s a lot of excitement about the biosensors which is what I personally work on my for my thesis and I think this will be a great tool for optimizing enzymes of pathways so in the church lab we decided to make a progesterone biosensor progesterone is an intermediate to hydrocortisone also known as cortisol and his in itself an important clinical steroid and both are actually on the World Health Organization’s list of in central medicines both have been produced from in yeast at very low levels with progesterone being a limiting intermediate so we decided to try to use biosensors to optimize conversion of pregnenolone to progesterone by the enzyme three beta hsd so going back to artificial selection and directed evolution we wanted to have a way to look for a desired trait more rapidly this is pretty difficult to do because there’s not always an obvious way to link our desire tree so there’s chemical production to something that’s very easily measured like cellular growth however if we could link a chemical signal to gene expression we could potentially force a cell to become dependent on a chemical of interest one idea that we’re working on to try this in the church lab is to use conditionally destabilized proteins the idea behind this is you can take a protein that naturally binds the chemical and find a mutant of it that requires the chemical to be stable so the mutant when expressed in the cell is Miss folded cells have protein quality control machinery which detect this ms folding and degrade that protein however when the chemical interest is present the protein associates with it and this binding provides enough energy to stabilize the fold so the proteins no longer degraded by the cell so this

conditional stability and degradation behavior is able to be conferred to any fusions so we can actually link the biosensor to any other gene that we want this means we can use a protein like GFP which flores’s and link it to the biosensor and have those fluorescence levels report on the amount of chemical present in the cell so theoretically this technique can be applied to many different kinds of chemicals so to develop a conditionally destabilized protein biosensor we start with a protein that binds our chemical interest we then utilize it using error-prone pcrs previously described and then use it to gfp we then use a technique called flow cytometry activated cell sorting or so fax works by rapidly flowing cells one at a time path delays are in a very narrow stream of liquid the laser is used to measure each cells fluorescence and the stream can be instantaneously adjusted to sort or capture individual cells based on their fluorescence so this is an extremely powerful technique that can measure millions of cells in a few hours and we can use it to identify proteins that have our desired biosensor behavior so once we have our biosensor protein instead of just using it to gfp to measure fluorescence we can actually link it to other proteins that activate gene expression so here the idea is the same without the target chemical the entire protein is degraded due to the previously discovered mutations addition of the target chemical is able to stabilize the protein allowing our entire protein biosensor construct to activate expression of a gene such as his three which is an enzyme that the east needs to grow so to use our biosensor to optimize activity of the three beta hsd enzyme we make the cells dependent on increased activity of free beta agency so increase progesterone production so we do this by minimizing the 3 beta hsd enzyme in a yeast that is unable to make an essential metabolite called histidine which is a type of amino acid so our biosensor is then linked to the activation of the history protein which allows the cells to make histidine recipes growth so here we feed the cells with pregnenolone had mutants of three beta HSC that are not active enough will not activate cell growth however cells with an enhanced version of three beta HSD are able to make enough progesterone to activate the biosensor and then make enough histidine to grow so this allowed us to isolate improve enzyme variants from a very large pool very rapidly so we use this method to find mutants of three beta HSC and measure their using standard methods so this graph shows the amount of progesterone produced when cells were fed pregnenolone and this is the original wild-type protein and enhanced versions were found through directed evolution and they show it significantly enhanced activity so this was all done actually in just a matter of a couple of weeks and the biosensors we think they’re extremely powerful tool for metabolic engineering and enzyme our pathway optimization so just a quick recap new technologies are very rapidly advancing our ability to produce new molecules a novel renewable way and I think actually in the next five or ten years the way we think about making materials will begin to change quite a bit as we can produce more and more things biologically so instead of just harvesting them from hard to grow plants or animals or petrochemicals we can use more simple organisms to produce them more rapidly there are also a lot of different biosensor ideas and other technologies that are coming to fruition and all these are making synthetic biology a really exciting space right now so if you have any questions before the break help you create the progesterone ages help you select things that are our yeah the question was how do we actually is the biosensor so before I showed we made a big library are a large number of variants of the three beta HSC enzyme through air pro pcr so some of those are to be much worse and some of them are very much better and using traditional methods we have to look at each variant one at a time with a biosensor we put them all together put them on the media and everything that is worse dies because our biosensor is not activated and everything that is better produces enough progesterone to activate our biosensor to activate his sitting growth our expression and growth easily screen them and pick out the correct one yeah exactly what are you asking what is secreted what are you asking here is are you asking his progesterone secreted so in this case our biosensor part of the question is is

the producer and secreted are we how is this all happening in the cellar out of the cell the progesterone is being produced in the cell by the three beta HS the enzyme and the biosensor is also being producing the cell because it’s integrated into the yeast genome and expressed so when those two come together in the East cell then it tells the yeast to activate expression of an enzyme such as history which allows the cells to grow how do we purify the progesterone is closin so in this case when I measured the final progesterone activity I just took the whole media that the cell was in and then highlights the cells so this is what it’s talking about before where sometimes you have to do energy to get all the chemical out so in this case we lice the cells and then measured everything that came out we disrupted we broke the cells and then everything’s filled out and then that’s how we isolated it like physically broke the cells does that answer cause me look oh how do we measure specifically as a question yeah the final measurements for the comparison of the wild-type versus the enhanced variance we’re done using a method called chromatography and mass spectrometry so these are methods for taking a very dirty sample and spreading it all out so you can isolate one chemical at a time and then identify what chemical that is so I can explain that later in the break if you want but that’s the method we used process go back and find out which units actually reduce a chemical that you wanted yeah the question is do we go back and look at what specific mutations help produce the camo chemical of interest and yet absolutely so when we do the selection when we do the the live versus dead growth um we get individual colonies that are we call it ISIL clonal so they came from one individual cell and they replicate many times so then you can see it very easily by I and you can take a bit of that and then send it for DNA sequencing and find out what allowed that specific version to grow better can you predict what changes to the protein you might want and force the desired change directly yeah so the question is instead of using air prone PCR which is what we call a non-rational way of generating diversity because it’s pretty much random can we use more rational or more predictive methods for doing this so yeah absolutely in some cases it’s very hard to know what changes are going to enhance your desired activity so that’s why we actually use error-prone pcr but the cool thing about this technique is you can use either one um to generate the diversity and select for the enhanced version using the growth selection and then you can map it back a pue sequence the variations and try to guess try to look in hindsight to see what made that better and then do that in an interview in melton i’m gonna ask how dangerous this process in the handling etcetera creating broke out random mutations is that something that needs to be taken into account so it’s something something you don’t want that yeah so for these specific cases that I mentioned it’s extremely not dangerous and I’m not just saying that so through aircrew pcr most mutations are actually going to break enzymes and that’s why you need something like our biosensor or other techniques to look for the enhanced activity there are some changes that could be potentially dangerous um John’s actually going to get into some things that you would want to potentially release in a while to affect the the ecosystem and there you need some more thought into what changes you’re making and he’ll get into that maxim so the question is are any of the compounds we make too big for bio sensor to detect and or activate gene expression in the nucleus no animalism

so the biosensor is actually originally um when it’s being expressed it’s in the cytoplasm and then can be targeted to the nucleus if it is stabilized if that makes sense so there is a nuclear targeting sequence but it’s being made and then probably miss folded when the chemical is not present in the cytoplasm your license how do you come up with a version of a molecule of that only become stable and bound with this particular other audio it’s just random how to drive so the question is how do we sort of getting how do we find our biosensor trade where it’s dependent on the small molecule to be stable so I had a slide on that where I very briefly went into it we do airpro pcr because we don’t we can’t predict what mutations will create that trait so we have to look for it by fusing it to gfp so when we make our variance that we’re testing we were making like 10 to the 6 so a million sometimes a bit more so 10 million variants that we test very rapidly using the flow cytometry so if we didn’t have this fax method we would not be able to make new biosensors in any sort of timely fashion oh one more question what is the most difficult thing I do in the lab every day well I mean it’s it’s very hard to think okay well I’ll say the joke anyways the food here is really bad like don’t come here for the food I would say the most difficult thing is trying to spend the time to think very deeply about problems because in the day-to-day lab there’s a lot of busy work you have to do there’s a lot of like oh you got to meet with your professor or other people in the lab you got to go to class and other things but I think the point of grad school is to learn how to think very critically about a problem and address problems that will make it impact in the lives of scientists or other people so it’s hard to really take the time and step back and focus on that alright moving on how can synthetic biology help solve this problem more specifically how do we bring synthetic biology from these vats and laboratory into the beautiful pristine nature and try to preserve it well as with any synthetic biology project we’re going to have to take closer look at DNA and for the remainder of this talk I’m going to use the simplified format of DNA show on the Left which kind of looks like a ladder so we tweaked some DNA in the lab by inserting a mutant gene or a laboratory design gene that is showing green here which let’s say effectively makes a boring old gray mosquito green and does some special stuff that we wanted to and then as dan went over earlier in his pee mating example from Greg Mendell if our green special engineered mosquito meets a gray wild-type one they’re going to mate and their children is going to inherit one chromosome or one copy of each gene from its parents so it’s going to have a gray Jing and one engineer gene and honestly we don’t know about gene engineering that well to say that the green engineer copy will be dominant so let’s say it will be either recessive or maybe there’ll be some kind of mixing going on either way we can’t say for sure all the children will express the physical traits that we wanted to for whatever purpose we’re engineering these mosquitoes for let’s go for one more moving on for one more generation you may notice that if all the children of the previous generation were to mate with each other we get something like a quarter of them should inherit two copies of the engineer gene and therefore express the traits we want but this is the

most hypothetical ideal case and you can imagine that if any of these mosquitoes were to mate with more wild types our engineered gene will proceed to get diluted even further and then some of them may die so we will essentially lose the gene in the wild and that’s not good because we spent all that money and effort engineering these genes and it’s not going to give us an effect in the environment so we need to tweak inheritance be cut and at the end the day this is necessary since engineering genes are super expensive and just difficult so what if in the same punnett square instead of this kind of boring old Mendelian inheritance we can engineer our special gene to not only express the traits we want but also go back in on its own genetic background and edit any other copy that does not contain itself to contain the new engineer gene in this way we can ensure that any mating event that involves one parent with carrying our engineer gene will produce only children that carry both copies of our engineer gene and I’m going to tell you that synthetic biology can do this scientists have known about some naturally-occurring g elements that can bias inherence its own favor for probably 50 and 60 years by now and a very clever guy named austin burt was a one of the first people to propose to make use of these selfish seeing elements to petty piggyback more engineered DNA and introduce them into other organisms using some newer technology discovered only about three years ago we now have a better system to do what dr. Burt proposed and this technology can effectively allow us to engineer organisms in nature and this technology is known as string drives that is the buzzword here feel free to google it and again to reiterate the way this whole process works starts with point one where a mating event occurs with something in the wild and something we made in a laboratory and released into the wild it will produce children that carry one copy each of the genes and then the engineered genes will activate and copy itself onto the other chromosome and create essentially more engineered organisms that can go on to further propagate these engineered genes until an entire population carries the genes we wanted to and do the things we wanted to do are there any questions silence all right well moving on though why bother doing this also just you know again isn’t editing danger that nature a little bit dangerous well let’s take a look at malaria again as a Justin one fully Illustrated this is a disease that affects millions to hundreds millions people around the world has huge financial burden and causes a huge amount human suffering around the world every year and the key to the malaria parasite passing from people to people is actually its host organism the mosquito since people cannot transmit malaria to other people directly so if we can stop mosquitoes from being this intermediate host organism for the parasite malaria by say editing out some jeans are just tweaking them a little bit to prevent malaria from being able to reproduce inside the mosquito then we can effectively eradicate malaria completely and this is in fact a real project that’s currently under active investigation here at the Harvard Medical School and the harvard school of public house here’s another case while my favorite personal favorites involving rabbits so to start with the question of why rabbits were introduced to Australia some of you may know Australia doesn’t really have native mammals they have marsupials and then rabbits were introduced through English colonists as game animals for hunting and they brought about two dozen of them over now why are rabbits a problem well they look cute and they’re extremely fun to play with in Europe Asian Americas but they’re very aggressive eaters they tend to eat everything which causes the topsoil to become exposed and to erosion and other problems and they multiply really really fast to just give you a basic idea of how fast one single pair

of rabbits can produce over a thousand progeny in just one year and they can live up to seven years so extrapolate that yourself and each pair of rabbits can theoretically give rise to 40 million or so rap baby rabbits this was a real study done by the government of New Zealand to try to understand just the scale of their rapid problem and therefore rabbits are actually and considered the number one cause of environmental degradation and driver of species extinction in Australia and New Zealand well what have we done about this we built a giant fence that would make Donald Trump proud this thing spans the length of over 2,000 miles across the constant of ocean the western portion of the continent of Oceania and it didn’t work we also tried some other extreme methods in 1950s dr. Frank’s fenner a veterinarian introduced my socks my viruses that are make rabbits kind of really sick and I really limit their ability to kind of hop around reproduce and such and this caused a brief crash in Australian rapid population dropped from its peak of 600 million or so to about 100 mm in in the late 90s early 2000s but it’s gone back up again because the remaining rabbits are apparently immune to this virus and yeah as EML come says life finds a way so and this is why i am here talking about this rather radical synthetic biology solution to try to solve the problem of the rapid expansion in oceania as well as malaria now of course many of you may have some concerns about this well what about the long-term effects and experiments are in progress to find out about the long-term effects of these things and of course is what have to be accelerated and this is actually going to be my thesis project to use model organisms within proper laboratory containment to figure out what are the long-term effects of using these engineer gene elements 50 100 generations after initial introduction so just so we know before we actually do anything and some of you may be concerned about experiments getting out you can say well fine you know these tested well designed jeans are okay but what if something a lab accidentally escapes well we try our best to prevent that by both engineering our genes such that the ones we’re experimenting on cannot actually pass these genetic elements onto wild animals if they do escape we use a lot of barrier containment and that’s exactly what it sounds like doors and cages and we use environmental containment which means most of the organisms we work on we try to work on them in a laboratory situated in locations where these organisms are not native to and the environment is hostile for their survival and of course nothing is fun without an undo button at most of you guys know from working on computers so we’re actively researching to figure out designs for reversal gene drives and immunization programs to ensure that if even after all the precautions we take something does go wrong we can hit on do button and start again so the vision is to eradicate many infectious diseases around the world including a malaria yellow fever dengue and lyme disease locally and to restore the environment by removing a lot of the pests invasive species that we’ve brought around the world and I hope I’ve at least convinced some of you that synthetic biology can not only create a lot of useful products but potentially save lives and preserve the beautiful Earth’s that we live on and in summary I guest and Justin and I have talked about using some synthetic biology to make useful products fight disease restore the ecosystem and potentially make dinosaurs which is going to be another science in Newstalk three weeks from now and Ian Malcolm is going to want to shush me from talking say anything more about t-rex’s so I’d like to thank you all very much for coming here braving the storm and I’m happy to take any questions sorry go ahead cycle where the breaks

that cycle watch for them eventually falling to be a more direct again nature oh it’s fine away yes and we hope that in the kind of down period where there’s very little malaria that is immune to our design happening we can hopefully identify these new organisms that are immune to art design and design eugene drives to re-engineer other organisms or things and how we can hopefully eliminate the disease and again sorry good so I was thinking again with the studio’s if you look at the malaria malaria can take a variability is there are you looking for any evidence that that capability to perform some other function get into mosquitoes the co-sleeping performs some other biological function either the US over its environment other creatures etc so to reiterate the question the question is basically are in the proposed malaria eradication project using gene drives are we trying to use gene drives to confer new functions on to mosquitoes no I’m asking is your bilingual malaria-carrying nobility the sku of the earth does that ability is that wait is that now is that for fun performing another any of the functions are you looking to whether it’s performing other functions that if yes stop this because it’s got something else you break into that Genie destroyed by by malaria to in fact mosquito will bring that gene break something else in this kid was that gene doing something else useful so uh i’m not the biggest expert on anopheles mosquitoes but as far as literature i’ve read and the scientists are actively pursuing this tells me to our knowledge the Jing is not doing anything significant to the mosquito life cycle so they can hopefully continue their row as a I guess food species for many frogs birds and other things so the question is I propose that we do something about rabbits and oceania presumably desired gene to kill the rabbits and actually no I’d like to just figure out a way to slow down their reproductive rate it is currently insane one pair a thousand rabbits per year if we can reduce that even by a half or reduce it down to a quarter of that rate than the rapid population will be much more manageable by methods we currently have welcome did you have you can explain that so yeah excellent questions so the question was I guess kind of glossed over and the Skip tagine editing process that introduces our engineered gene into wild-type jeans and this uses the new crisper based genome editing process I really don’t want to get into the details feel free to talk to me afterwards and I’m there as a Stefan vinny introduced earlier there’s going to be a podcast talk specifically about crisp air technology so I’ll let those guys explain that there’s that questions oh yeah to be optimized why make it a weekend to that easily kill it you want to optimize it total sweat within also the specific you know till it in it over lunch about because the genius but it cost a fair so the question was the things I’m proposing seem to involve limiting

reproductive ray or trying to kill an organism and isn’t that kind of counterintuitive to natural selection and how genes spread naturally and how are we proposing to actually spread these jeans around and I like to say that most of the systems we are currently designing do not actually involve things that would be really detrimental to the organism survival and this is one of the things I’m actively testing in model organisms was in the laboratory so ideally these organisms would still be competitive enough now this means it doesn’t have to be as competitive to the natural wild type maybe seventy eighty percent competitiveness and now be enough to allow it to have a few mating events past the gene on and as I explained once the gene has been passed on it’s very good at spreading itself to other copies but I is there any promising promising that’s possible so the question is are there any promising leads on the research for the undo button and I don’t want to get too much into details of this but the short answer is ill involve designing new organisms in a live laboratory that are based on editing the gene that we introduced to change it back to what it was before and theoretically by doing that these organisms should be more evolutionarily competitive than the engineered process these we introduced and that combined was the gene dry factor should allow us to change things back to what it was in a somewhat time effective manner any other questions all right thank you