Science Buddies: "Ask an Expert"

Hello,
My idea is trying to produce venom by inserting the gene for the venom into a K12 strain of E. Coli bacteria. I would be following a method similar to that of the PGLO lab (the only difference here would be the gene that was added would be of the venom instead of a glowing protein). I am wondering, however, why no one has done this before. When researching this I could not find a single resource where someone even attempted to try to create venom by inserting a plasmid with a gene for any venom. Is there some problem that I can't see that makes this impossible, because I feel like someone should have tried this if there was no problem that would render this experiment useless? Also, assuming, that there is no problem, the venom I would be trying to produce would be that of the western black widow. I have the genetic sequence for the venom however I'm unsure if I need something like a protomer region that I have to make, or if the plasmid will just fuse into other plasmids and the protein will be transcribed/translated automatically. My other idea was to actually use the pBAD complete sequence with a custom insert of the Black Widow venom to make sure it would then we transcribed and translated. This is what I have done in the PDF attached below. I am then however unsure if that would just combine the glowing protein and the venom into one giant mess that would then have no function at all. The other problem with this is that I'm working on a limited budget and I either need a short sequence that doesn't cost thousands of U.S. dollars, or a company that would be willing to give me a student discount. So if you know of any of these places were I could get this, that would also be really helpful assuming I really have to have a giant sequence. Attached is the sequence that I am currently working with, however, I either need to shorten it, because it costs to much, or if the pBAD sequence is not going to be helpful, then everything capital can be ignored. Everything that is lowercase is the actual sequence for the Black Widow venom. I was also thinking of possibly switching the venom to something more toxic to the E. Coli bacteria, perhaps like a cytotoxin or a hemotoxin, and use that toxin to prove that the venom is really what was produced by the death of new E. Coli. Your thoughts on this would also be really wonderful. To whoever reads this, Thank you for taking the time to read and respond and I hope you are having a good day,
-Michael

Hi Michael,

A rather interesting project idea! Great way to learn about genetics. Let's unpack it a bit:

1) Your first question, regarding why no one's done something like this before, is probably most easily answered by ethical considerations. Why do you want to create black widow venom specifically? It seems unnecessarily dangerous, and with few beneficial uses. Why not a less toxic compound?

2) I should also note here that "venom" is rarely ever accomplished by a single protein encoded by a single gene. Venoms are usually concoctions of several proteins, although yes, there is often a dominant "active ingredient" of sorts. That brings me to ask the question: where did you get the gene sequence you've attached, and what exactly is it encoding?

3) More broadly, what you are referring to here -- inserting DNA into bacteria so that the bacteria synthesizes things for you -- is now an accepted, routinely used procedure in biology laboratories! This is what's known as "transformation," and it's so common & useful that there are companies that sell dedicated kits to do it. You may be able to find more resources using this keyword search term. Understanding how plasmids work will also be vital. Here, for instance, is a tutorial on how to perform a transformation (which assumes you have access to a fully stocked lab): https://www.addgene.org/protocols/bacte ... formation/
("Transfection" is also a commonly used term, although this is more specific, because it only refers to transformation that is performed using a bacteriophage (virus) to insert the DNA.)

4) Finally, regarding your point about making the target protein something toxic to the E. coli: this is a creative solution, but unfortunately it has problems of its own. Think about it this way: if this variant experiment succeeded, how would you be able to tell the difference from the experiment failing because all the cells died due to an incubation error (without going in to chemically test for the presence of toxin)? The reason why glowing proteins are often used in these experiments is because it's super easy to figure out if it worked or not. If it glows green, you did it! If it doesn't, you failed. That's a simplification, but nonetheless very handy in the lab.
(And one tiny nitpick -- a hemotoxin wouldn't necessarily kill E. coli! "Hemo" refers to blood, which E. coli and single-celled organisms don't have. Hemotoxins might disrupt cells in a way that would also affect E. coli, but it's not explicitly part of its definition.)

I don't mean to intimidate you -- you clearly have ambition and I don't want to take away from that. Having this much of a plan already shows that you've done your research. But I'd advise you to readjust your focus a little bit, and ask yourself why you want to do this experiment -- what question you'd be answering. Then, you'll want to consider what materials & equipment you'll realistically have access to, as that tends to be the critical deciding factor in many high school science fair projects. If you can let us know roughly what you have access to, we can help you carve out a project from there.

In case it's helpful, I'm also linking our Project Guide, which I recommend to anybody on this forum starting out a new project: https://www.sciencebuddies.org/science- ... ience-fair Give it a read-through to familiarize yourself with the full process. The Advanced Project Guide tab may also be of interest to you at your level.

All in all, this is definitely an interesting idea, and it's a great way to learn about plasmid genetics (a topic I have specifically avoided getting into the weeds of in this post, lest it get way too long). Key takeaways are:
-> Trying to synthesize anything remotely harmful, like a fully functioning toxin, will be difficult simply because scientific ethics will bar you from even attempting it. My advice is to find something else you'd want to synthesize.
-> I'd recommend that your first priority is figuring out how you're going to get your hands on the materials -- the plasmids in particular (regardless of what they'd be encoding). Do you have access to a university lab or something comparable?

Hope that helps -- that was mighty long and I fear I may have rambled for a bit. Keep us posted on what you're thinking. And keep up the good work!

Hello again,
First off thank you so much for responding so quickly and with such a detailed response and my apologies for taking so long to respond back. My reasoning for using a toxic compound is for the end goal of this experiment. I trying to be able to produce antivenom more safely and efficiently. My goal was to produce the Black Widow venom with E. Coli so that the people that make antivenom could get the venom without having to milk a spider, or a snake, or whatever organism is being used. I was thinking that this would also be more efficient, based off the research I've done, since milking spiders and snakes for venom is very dangerous and time consuming. Therefore, if it could just be produced continually by E. Coli we could increase safe working conditions well also increasing productivity. This then would hopefully mean that more antivenom could be produced and distributed to more hospitals around the world and hopefully save a few more people from death then otherwise is happening right now. I know this is kind of a big goal so I am doing this experiment solely to see if this would be a feasible way to increase safety and productivity of antivenom and then use the implications to take it further. The place where I got this sequence was the pBAD sequence (in capitals) from the the Bio-Rad website https://www.bio-rad.com/en-us/applicati ... D=NISQOC15 and then the alpha latrotoxin sequence (in lower case, also the venom produced by the western black widow) from the NCBI website, specially here https://www.ncbi.nlm.nih.gov/protein/AGD80171.1 . With the amino acid sequence given I then used this website to get to the full base sequence https://www.bioinformatics.org/sms2/rev_trans.html. The whole gene sequence is encoding the pBAD glowing protein where the alpha latrotoxin venom is attached to end of the second protein coding region. I am a high school senior and we have access to a chemistry lab, a small biotech program (we have all the tools needed in the video that Bio-Rad has made on how to perform a bacterial transformation, https://www.youtube.com/watchv=DHVlSDXu ... adExplorer) and otherwise what is usually available to high school students. However, we do have couple interesting things like a PCR machine, and an inverse microscope so if something exotic is needed we might actually have it. Also since I am in IB Bio, we have permission to use whatever resources are at the school so trying to get approval to use one of these things is not going to be a problem. Once again, thank you so much for taking the time to read this and I hope you are having a good day and staying healthy,
- Michael

Hey there,

Sorry about the delayed response -- I got distracted while typing up a reply, then never got back to it...Apologies.

Your reasoning with the antivenom is interesting, and I suppose it's defensible; nonetheless, I'd still say it feels a bit ethically risky for a science fair level project. If you are confident that the paperwork is doable and this won't raise eyebrows when you deliver the project (or proposal) to your teachers and/or fair judges, I suppose I have no further reason to stop you. The only point I would raise is to get a good sense of the danger levels of black widow bites...they're certainly not as lethal as most people probably think, but it doesn't mean they're harmless. If you want to scale down the danger, you could easily substitute the toxin for a different target protein. At the end of the day, this is still what I would recommend.

I am quite impressed that your high school provides access to all the equipment and facilities you would need -- in that regard, you are very fortunate! You know better than I do about what your material constraints are, so I am leaving that to your judgment then. It seems you're in good hands though. Just make sure you know you have everything you need before you begin performing any experiment, as it is quite annoying to have to hunt something down mid-experiment.

Let me go back to your original question about feasibility of the project as described. From an overview perspective, what you are doing is completely plausible and it is done quite often in labs. However there's a bit more to it than pasting in a reverse-translated DNA sequence into a plasmid. (Side note: can I ask how you intend to obtain this custom plasmid?) Here's a link that highlights more of what you'd expect from a lab: https://blog.addgene.org/plasmids-101-r ... on-cloning

The gist of it is that something like your pBAD sequence is used as an empty template, or "backbone," that does not contain the gene of interest but has restriction sites that can be used to paste in a gene of interest. Restriction sites are the ticks you see in the sequence diagram that have names like BamHI and EcoRI; these correspond to specific restriction enzymes that make cuts at these specific points. This is where your gene of interest (also cut open with the same restriction enzyme) can be ligated in. Simply put, at a molecular level, here's what needs to happen before transformation can begin:
1) Empty backbone gets cut open
2) Gene of interest gets cut open (in a way that is compatible with step 1)
3) Gene of interest gets annealed into the open empty backbone
4) Desired backbone+gene plasmid is purified from all the other crap that's floating around after all that molecular cutting and pasting

If I understand correctly, what you are proposing is skipping all of that and somehow (I'm not sure how) ending up with a plasmid that has your gene insert already in it, possibly by manufacturing a custom sequence from scratch. Which would indeed be quite expensive.

I have no idea whether that will work; my expertise is not in genetics, but I am leaning towards "no." Intuition tells me that if it were simple (or if it were cost-effective), all these labs that depend on bacterial transformation would be doing it that way. The science tells me that, while you probably won't need a separate promoter, the reverse translation you've done is probably not going to work the way you intend. Remember, mRNA to protein is not 1:1; considerable processing happens both before and after transcription and translation. You lose information when going from RNA to protein. The reverse translation resource you have used seems to be designed for building PCR primers, which only need to bind to a short segment. I doubt it functions the same for building an entire protein coding sequence. The plasmid may also have difficulty parsing where GFP ends and your gene begins, but that's a lesser problem.

Let me know what you are thinking. I suspect you'll have to use a more tried-and-true method for plasmid bacterial transformation; I'd hate to have you risk so much money, time, and resources.

Hello again,
My apologies for the late response, I decided to take a break to destress over thanksgiving, however, I'm now getting back to work. The way I intended to get the plasmid was to order it from possibly genewiz or some other manufacture that might be willing to either sell it at less than a thousand U.S. dollars or give me some student discount or something like that. If you have any suggestion on where I might be able to get this that would super helpful, if not, do not worry about it. Thank you for the explanation of how the gene would be inserted because I had found several times in my research that you have to place the gene of interest into a vector, yet, I had no idea what a vector was and deemed it unimportant. Now that you have explained how that works, it makes much more sense how labs can easily do this. Yet even then I suppose I really am stuck in a loop hole over what I do with the sequence now that you pointed out that it is not a 1:1 ratio of translation. I do understand what you mean I just have no idea what to do at this point. I know that genes usually have an n number of A's (adenine) on each side of any sequence that is going to be translated yet they are all destroyed in the cytoplasm. Yet I believe each gene needs a different number of A's (adenines) at the beginning and end of the sequence and for each different cell this will also change the number of Adenine needed. Unless, would it be possible to find the number of adenine based off of Chargaff's rule or does that only apply to the actual gene after it has been damaged by the cytoplasm and therefore is fully functional? This is the only way that I think that I might be able to repair the sequence. Thank you for your consideration and thoughtfulness in your responses, if you have an idea on putting me back on the right track that would be awesome, yet I fear I may have to trash this idea over the complexity.
-Michael

Hi Michael,

Apologies for the late response. I've tried to think about how we could take this project in a direction that would make it more feasible for you to execute, but I'm not confident there is one (without considerable expense). It is clear to me that you have done a lot of fantastic independent research into this topic, from genetics & the central dogma to plasmid microbiology specifically, and I truly applaud you for putting in that effort! At the moment, however, I suspect we've hit a wall in the theory behind the project. I encourage you to take what you've learned and re-apply it with a fresh project concept -- the time you've spent on this one will not be "wasted" because you have learned from the experience. Perhaps you can develop a project that uses a more mundane transformation protocol. For instance, an engineering-oriented project where you can identify how certain various factors affect the speed or success rate of a transformation. (Just spitballing here.)

Please keep me posted, and don't hesitate to ask further questions, even with a barebones idea. I look forward to hearing about your innovations in the near future.