IGEM:Caltech/2008/Ideas

{|cellspacing="5" cellpadding="10" style="background:#ffffff; width: 750px;"
 * -valign="top"
 * style="background:#ffffff"|

Bioremediation (or not)

 * So I guess I had failed to communicate my bigger picture about this idea. So removing materials is one thing that can happen, but the larger picture is getting a lot of independent bacteria (all in different states) together into one location such that information can be easily transmitted amongst the bacteria (through conjugation, etc). So say for instance that a particular bacteria has an important info that needs to be relayed to its neighbors. It can send a flag that will cause all the other bacteria is aggregate around it and pick up the signal. The neat thing about it would be that any bacteria can send a flag, so what we would get is a very dynamic network that has great potential. After the signal is sent, then all the bacteria can disperse and continue what they were doing before. So what we get is a bag of independent modules that can talk to each other (at least to its neighbors). I don't have an application for this, so maybe this idea is for the future.

Cholesterol Degradation

 * Could we make a strain of E. coli that circulates in the blood stream and feeds on cholesterol plaques as its food source? I know part of the 2007 UC Berkeley project, the bactoblood one, was to make E. coli not trigger sepsis when in the bloodstream. This chassis would be very useful for a project like this.
 * Similar idea, but use it for diabetes also and have E. coli feed off sugar. Maybe both?

Bacterial Mouthwash

 * Disrupt biofilms
 * Prophage targeting other bacteria
 * lamB is sufficient for lambda infection.
 * Looking at other receptors
 * SPO2 is a lysogenic subtilis phage
 * How do we get the phage inside the cell? Take a resistant strain (coli w/o lamB for lambda, coli for a subtilis phage), add the necessary receptor on a plasmid. Infect, select for lysogens. Grow under nonselective conditions, then counterselect for loss of the plasmid. Voila. Tetracycline can be counterselected bochner.
 * Would it be possible to design our strain to prevent biofilm development? It might be a lot easier to target a few major strains in biofilm formation on our enamel w/o the biofilm to protect them. Would this be a possibility? Reduction of S. mutans adherence
 * Josh K. Michener 22:23, 11 May 2008 (EDT): See #kruger, kelly, younson, iwaki
 * Robert 20:23, 12 May 2008 (EDT): After reading #5, if I understood correctly, the pac gene producing the PAc cell surface protein antigen binds to our tooth's surface. This in turn blocks most of the binding of S. mutans? So, if one would to have a "special" teeth cleaning removing bacteria from our mouth, and then have toothpaste (or mouthwash, etc.) of our non plaque forming strain (they used S. lactis), this would block the formation of S. mutans? If we were able to develop a steady growth rate of our strain (lets say L. lactis), we could hopefully keep a steady amount of S. lactis on our teeth blocking S. mutans from binding, and preventing plaque formation (I think!).
 * Josh K. Michener 22:23, 12 May 2008 (EDT): S. mutants binds through PAc. So you can get competitive inhibition by expressing exogenous PAc. Either your engineered bacteria could attach using the same receptor, or you could bind something else and pump out PAc to compete off mutans. I think it's an open question whether you'd want to use S. lactis for this or try to get things working in E. coli. S. lactis would be easier initially, but any future work would be tough. I haven't been able to find anything on the functional expression of PAc in E. coli. Best I could find was some stuff that targeted it to the periplasm.
 * Robert 02:05, 13 May 2008 (EDT): McAb. Disregard my previous question about this article. I didn't catch a key word in the article, but I understand it now.
 * H2O2
 * Turn on xanthine oxidase, turn off catalase
 * Minty smell
 * Taste good?
 * 
 * Orange
 * Vanilla
 * Damascenone (fruity)
 * Ethyl-2-methyl butyrate (apple)
 * More important than avoiding having to floss: prevent dental problems in poor/underserved communities where lack of dental care can cause serious problems.

Gut microbiota

 * Food poisoning
 * Curing superbug infections
 * Doug Tischer 20:51, 10 May 2008 (EDT) We could make a gut microbacteria that acts as a guard against the overgrowth of drug resistant bacteria. We could engineer E. coli that turns "hostile" if it detects the presence of say, a tetracycline resistant strain. Bacteria normally love to try to spread their antibiotic resistances, even between species, so I think it would be relatively easy to detect if there is a new bacteria that becomes tetracycline resistant. We could use a riboswitch to detect the presence of a resistance gene within the cell. Then we could engineer in one of several responses. I still like the idea of having an inducible phage. So maybe one response would be to have our E. coli pump out a phage for the other strain. This could be a phage for E. coli (in which case we would engineer our bacteria to lack the viral receptor) or for another strain like Salmonella. Maybe the E. coli could pump out H2O2 if it finds a clump of the drug resistant cells. Or maybe we could take advantage of the fact that during conjugation there is a physical connection between the cells. While the drug resistant strain is passing its plasmid to our engineered strain, we could make our engineered strain pass a lethal gene back (although I know conjugation in a one-way process). I like this idea because it seems to me there are several different strategies (and projects) for detecting the drug-resistant bacteria and killing them.
 * Doug Tischer 21:00, 10 May 2008 (EDT)If the drug resistant bacteria don't readily conjugate to our E. coli, we could still free their DNA into the surrounds by occasionally sending out a set of phages that would lyse the foreign cells and expose the drug-resistant plasmids to the surrounds. Again, we could put the "prophage" into the E. coli, but it would only be occasionally turned on through slipped-strand-mispairing. Once the drug resistant plasmids are in the open, it would be easier for our cells to take them up and detect. Granted E. coli aren't naturally competent, so this could pose a bit of a problem. Perhaps we could use B. subtilis, which I think is naturally competent.
 * Josh K. Michener 02:00, 11 May 2008 (EDT): Take a look at this
 * Doug Tischer 05:07, 11 May 2008 (EDT)Well, that's a little disappointing. However, I still think we could do them one better. I didn't like the fact that they had to modify the other strain to get a selective pressure against having the tetracycline resistance gene. I guess the project I had more in mind would be turning a bacteria into something like a white blood cell. It would have the ability to sense a foreign virulent bacteria based on plasmid genes or specific transcripts. It could then kill the bacteria in a number of ways. One might be to generate an oxidative burst with H2O2, similar to what a nuetrophil does. This would kill our bacteria, and hopefully a neighboring bad bacteria. I'm sure there are many other ways we could make one bacteria kill another if they are close to each other.
 * Doug Tischer 05:40, 11 May 2008 (EDT)Looks like there are a couple of options, if we wanted to do an oxidative burst. There are a couple of enzyme I've found besides xanthine oxidase. NADH oxidase has the advantage/complication that it is integral, 7-subunit protein. The nice thing about this is that it would pump H2O2 outside of the cell directly onto the bad cell nearby. Although, I don't know how successfully we could express a integral protein, let alone in a good stoichiometric ratio. It would definitely take a lot of playing with. Glucose oxidase looks like an easier enzyme. It's a dimeric cytosolic enzyme that consumes D-glucose and produces H2O2. Maybe we could also combine this oxidative burst strategy with protease cascade idea, to get a lot of the H2O2 producing enzyme out before the H2O2 kills the cell.
 * Josh K. Michener 21:01, 11 May 2008 (EDT): Take a look at these tiina, stevens


 * Vitamins
 * Beta carotene?
 * Lactose intolerance

Target Specific Drugs via Bacteria Carrier

 * So something that has frustrated me about drugs is just how non-localized they are. They simply flood the entire bloodstream, hence all the negative side effects. I would be up for investigating if we can get bacteria (which would carry the drug) to locate specific targets (and thus release the drugs at specific locations in the body).... I don't know that much about blood capillaries, so I don't know if it would be possible to determine from inside the capillaries where one is in the body (maybe from solute concentration? ex. low oxygen). But I think this idea is worth investigating (at least for feasibility).
 * Josh K. Michener 16:43, 9 May 2008 (EDT): I'm having trouble thinking of a reason that you'd use bacteria for targeting rather than a surface-coated capsule. The cancer targeting bacteria is one example, but overall I think it would be tough.
 * I'm not quite sure myself. Maybe if it invovles some sort of advanced calculation such as taking into account multiple factors... okay, nevermind then
 * Doug Tischer 20:52, 10 May 2008 (EDT)I had an idea just like this after our meeting last night. What if we could use a virus to target drugs to a particular cell type. I know a lot of viral capsids assembly spontaneously from their monomers, whether or not the viral DNA is present. Could we make the capsid enclose some small molecule (a stand in for an actual drug) or small peptide (like insulin) as it forms, and thus the virus would release it when it enters the target cell. We could take this one step further, and put the genes for a capsid protein from a human viruse in an E. coli under an inducible promoter (kind of like a prophage). When the promoter is activated, the E. coli starts pumping out capsid proteins which then self assembles and hopefully encloses some of our molecule or peptide. The capsid would not enlose the viral DNA, because it would not be present.Once the E. coli lyses, the virus would find its way to its target cell and release its contents. Taken one step further, this E. coli could circulate in the blood stream and produce its drug filled viruses on command. In terms of getting a peptide (like insulin) to stick to the viral capsid, we could use a sort of biotinylation system or other strong protein protein interaction. Tell me if this is just crazy or if you guys like part of it.

Synthetic Biology Platform

 * So this is something that I'm quite interested in, although it might or might not be appropriate in this context. So currently, most of synthetic biology is following the platform of transmitting signals from one module to another via PoPS on DNA chains. My question is if there are any other models that we might want to explore. Or, if PoPS is the correct method, can we somehow expand the library of standard tools (inverters, etc.). Can we look through what's currently available in computer architecture and see what might be implementable in cells (not saying that computers are the right model for syn bio)? If we find a new component and implement it inside cells, although this would not be a big idea, this would certainly be quite novel and have broad uses within the scientific community. This is rather vague - I'll try to find a concrete idea of what I mean.
 * Jason R. Kelly 00:46, 9 May 2008 (EDT): Probably not what you're looking for exactly, but MIT folks had a brainstorming session a bit back about the most-wanted parts that aren't in the Registry. You could cross-ref your list against that one if you want.

Ultrafast Color Changing Cells

 * Changes within a cell are not usually instantaneous. Transcriptional and translational change can take on the order of minutes or hours to produce enough of the desired protein. Could we make this change seem faster by engineering in a cascade of proproteins? We could make the cells (either yeast or E. coli) express a GFP that had a quencher domain linked by a protease site. The GFP wouldn't fluoresce until cleaved by the appropriate protease. This event could be made sudden if the protease itself was an inactive proprotease. With several levels like this, with one proprotease activating the next, it would only take one or two molecules of an initiating factor, like IPTG to make the cells glow green. The cells would both glow green very quickly, and would be very sensitive to any traces of IPTG. Let me know what you guys think.
 * Could we make them change from one color to another really quickly by having different cascades for each color? Then it'd be like a single color bacteria TV or something.

Bacteriofood

 * After reviewing UC Berkeley's BactoBlood in the 2007 iGEM Jamboree, Bacteriofood could be used to produce and carry vital nutrients that can maintain a healthy human. Most useful in third world countries where food is hard to come around. Bacteria being easy to grow and maintain, could be a simple way to feed the underfed. Just trying to throw out any idea that comes across.
 * Some thoughts - could we make bacteria more nutritious? I.e. you'd probably need an E. coli culture plus some list of vitamins/minerals/etc. Could we come up with that list of necessary extras and then engineer E. coli to remove a couple items from the list?

Endosymbiosis

 * Attempt to recreate endosymbiosis by introducing a bacteria (E. coli or B. subtilis or other) into yeast. This would essentially be creating an artificial organelle. I know yeast can grow without mitocondria, being called petite yeast. What new things could they do with an entirely new organelle? I'd imagine this could be done with some sort of double selection. We could take an auxotrophic yeast strain, say for uracil, and then "inject" a bacteria that secretes the missing enzyme in the uricil pathway. The missing enzyme would be secreted directly into the yeast cytoplasm by the bacteria. The bacteria in turn could gain shelter from some antibiotic in the medium. Only by living inside the yeast cell could both the yeast and bacterium survive.
 * Could we hijack an existing organelle rather than introducing a new one? Hacking mitochondria, say?
 * Sounds interesting? What do you mean by "hacking?" What would the new organelle be hacked to do? Would it be like transplanting an organelle from one species to another? (ie - giving chloroplasts to yeast?)
 * Well, there's always two questions - first, what can we do in a summer? And second, how can you spin the summer project as a model system for something more exciting? For the summer - depends on what people have done previously (and I'm not really sure what the current state of the field is). Might be as simple as showing that we can reprogram a specific organelle (worst case, just adding a marker). Longer term, you can make the same argument as UCSF last year - segregate your engineered system off into a separate compartment will less interaction with the host.
 * Okay, looking into this a little more, transforming mitochondria is tough. It seems that you attach your DNA to a 'microprojectile,' physically shoot it into the cell, and hope that it lands in the mitochondria. It's a 'gene gun' - I'd heard of it for transforming plants, but apparently the same is true for mitochondria. Interesting, but a little unrealistic for a summer project.
 * What about yeast parasites?

Taking advantage of noise

 * From an engineering perspective, noise is usually a problem we need to overcome. But in biology, a uniform response simply means that your entire population gets wiped out simultaneously (think agricultural monocultures). Biology often hedges its bets by producing a diversity of phenotypes, some fraction of which are favored in any given environment fluct1, fluct2 . Can we do the same - continually producing a range of phenotypes and allowing the environment to select the ones best suited at that moment?
 * I've always liked the idea of directed evolution. Usually mutation occur at random throughout the genome, but a lot of the time it seems researchers only want a few genes to change. Instead of using the time consuming site directed or random mutagenesis, could we engineer a bacteria mutation a specif portion of its genome at an accelerated rate? I'm thinking something akin to the cre-lox in which the researcher could flank the stretch of DNA to be mutated by two DNA sequences, and then an enzyme or set of enzymes would catalyze the accelerated mutation of only that portion. The ter sites and Tus proteins are normally used to terminate E. coli genome replication by slowing down and kicking off the polymerase. Maybe a toned down ter-Tus system would be enough to screw up the polymerase as it passes through, but not totally derail it.
 * There are mutator strains that increase the genome-wide mutation rate. Ignoring implementation, I think it would be tricky (not impossible, but tricky) to find a niche for targeted in vivo mutation. Mutator strains are generally used when you don't know what your target is. If you know what your target is, the question is what the advantage is over error prone PCR. Best I can come up with is that you might want to hit multiple genomic locations simultaneously, such that error prone PCR would be really time consuming. But those situations seem relatively rare.
 * "In 2006, Collins's team described engineering mutations into the control region of a gene that confers antibiotic resistance to create two strains of the yeast Saccharomyces cerevisiae, one with noisier expression of the gene, one with something more steady. Faced with a lethal antibiotic, the noisier strain survived better5. This result supports the idea that noise is a form of 'bet hedging' for cells: a population is more likely to survive in a changing environment if its members are noisy because some are likely to be making the quantity of a protein best suited to that situation. “A system that is covering more possibilities has a greater chance of survival in unpredictable settings,” says Collins."

CO2 Bacteria

 * It seems many researchers are looking to find or develop a biofuel that can sustain our cars, planes, machines, etc. As time passes by, we are still polluting our atmosphere with greenhouse gases, so why not develop a bacteria that will *chew up* or turn CO2 into a less harmful substance. (Is there any bacteria that consumes CO2?) Maybe we could create filters of bacteria to be put along our car exhausts, etc. I don't have much knowledge about greenhouse gases, but just an idea to throw out.
 * You're talking about photosynthesis here - fixing CO2. It's an important problem, but one that industry is already working on (see this for instance).
 * CO2 mineralization - can we precipitate it (limestone?).

Pimp my E. coli

 * We can trick out our E. coli similar to our car -- spinners, rims, spoilers, etc. One of many modifications is have the cells express reflectins on their surface. Reflectins are highly reflective proteins only found in squid reflective tissue reflect.
 * Random thought - we can probably localize proteins to the poles of the cell. So make the tips a different color from the body
 * If I remember correctly from last summer, UCSF used Pleckstrin Homology (PH) Domains for localizing different fluorescent proteins to different parts of the cell, such as the cell membrane.
 * Could we do polka dots? Make protein clusters in the membrane?

Changing shape

 * Can we make E. coli become helical using crescentin crescentin from Caulobacter?

Motility

 * As an output, we could turn motility on and off gallivan.

Stickiness

 * Make the cells express an adhesive protein mussel or turn on biofilm EPS expression (absent any other biofilm phenotype).

Taste

 * Natural flavoring!

A more analog device

 * Digital and analog responses, a common feature of electrical circuits, are also displayed by biological networks. While recent research has focused on engineering a more digital response using cooperativity or transcriptional cascades, we go the other way and engineer a more analog device.
 * One way to do this is to express mutiple tetR variants that have different affinities for aTc, the primary inducer used in bacteria. I have found one paper that reports tetR variants with different affinities tetR, although I'm sure more can be tracked down.

Others

 * Using network motifs? Uri Alon has done some interesting stuff with it. Should we need to use time-delayed releases, might be handy. http://www.nature.com/nrg/journal/v8/n6/pdf/nrg2102.pdf
 * Using ribozymes? No idea where this is going or if its valid, but who said that we needed to follow how the cell does it.... perhaps RNAi or one of those unique forms could be used as a branching statement in what seems to be a rather linear DNA programming methodology (or at least from first glance).
 * Kind of like this win, an ?:

Random Number Generator

 * FimE inverts a specific stretch of DNA, defined by a pair of sequence elements (IRR and IRL), forming a DNA loop between the two elements fim . If we add multiple copies of one of these elements (one IRR, two IRL), would FimE randomly choose one of the sites (one IRL out of the pair) to invert between? Either choose one of several promoters to attach to a given gene, or one of several genes to attach to a given promoter.
 * These folks holden tried replacing the wildtype 314bp sequence with a 1kb sequence from lambda and still got recombination at a reasonable frequency. So it'll happen over a decent range of lengths.
 * Then, can we tune the probability (from, say, 60:40 to 80:20 to 20:80)? Ideally do this dynamically (based on some small molecule) - use proteins that bend DNA to affect the probability of loop formation.
 * In another study gally, they showed that a natural protein, leucine-responsive regulatory protein (Lrp) helped promote recombination, presumably by bending the DNA.

Population Variability

 * Slipped-strand mispairing (SSM) phase can produce a heritable variation in the expression from a promoter. Roughly one in 1000 divisions produces a change in expression. Couple this expression to a selectable/counterselectable marker. Under any given condition (selection, say), the population thrives, but with a small group of the opposite phenotype (non-expressing). Switch conditions (to counterselecting), and the population can use these revertants to recover. The switching is stochastic by nature and can be directly compared to both natural fluct1 and synthetic fluct2 systems that utilize stochastic switching to adapt to variable and fluctuating environments.
 * Under constantly varying conditions, most circuits would die. These cells, though, can adapt and pass that adaptation on to their descendants.

E(motional) coli

 * In the spirit of Sims, we can engineer our bacteria to be emotional. The emotions can come from environmental effects (turn red upon DNA damage or turn green after nutrient addition) or unknown/seemingly-random forces (use SSM to randomly change the state of bacteria).
 * To introduce bacterial interactions, the emotion of one bacteria can affect all other bacteria in the direct vicinity, This could be accomplished using quorum sensing or conjugation. The sky's the limit on the types of emotions the bacteria display, the degree of emotion, and how these emotions can affect neighboring bacteria.

System Order E. coli

 * I am not sure if this is a problem that needs to be addressed, but here it is! In Syn Bio, there are a lot of designs of systems that could be soon put in humans. For all those systems, there needs to be some sort of order. There are many problems we need to take care of such as; making sure the bacterium is not destroyed by our immune system, will replicate and die at a wanted rate, will have an error rate that will not effect the system (maybe have the e. coli undergo death if there is a mutation in a vital gene), etc. I don't know if these problems will be a factor, but if they are, I thought it would be cool to design a strain of E. coli to be the standard of this system order.