Sciencemadness Discussion Board

Yeast plasmid max size

Jackson - 12-12-2018 at 10:17

Hi, does anyone know the maximum capacity for a plasmid that yeast could recive. I have a plasmid i want to make and inster in yeast but there are genes to encode for multiple protiens. My plasmid is around 10,000 bp long so indont know if it is too much to insert.

Tsjerk - 12-12-2018 at 10:27

Shouldn't be a problem

The secret is; you only need one. Just make sure your selection is stringent.

[Edited on 12-12-2018 by Tsjerk]

Metacelsus - 13-12-2018 at 00:14

10 kbp plasmids are fine for yeast (just make sure you have all the proper elements, like a yeast origin of replication and a selectable marker). In fact, yeast can have plasmids much larger than this.


Jackson - 13-12-2018 at 12:41

It is three seperate proteins that are each needed for different parts in the synthesis. I am new to plasmid design and I wanted to know if i needed to put in start and stops for each protein if i want them to come out separately as different protiens. Also, in plasmids I have seen premade, they seem to have random dna inbetween each sequence. Is this needed, and if it is, is it just random code or is it an important piece of data. Would it be better to have three individual plasmids or just one Plasmid?

Tsjerk - 13-12-2018 at 13:09

You want to distinguish between plasmids that are maintained in the cells and integration plasmids. They are very different.

A plasmid that is maintained needs to have a yeast (and probably a bacterial) origin of replication. Bacterial and yeast origin of replication are different.

The bacterial origin of replication you need because you are very most likely cloning the bacteria in E. coli.

Integration plasmids don't need a end host origin of replication, as they have either one or two regions that facilitate cross-over. It depends on the organisms whether it can be done in two cross-overs at once (easier) or in two separate crossovers (two step process). A single crossover strain can also be used but is not stable, as the single cross-over can be reversed.

Your start and stop codon question is a bit of a newbie question. Yes! you need a start and a stop codon for all three coding sequences. otherwise you will get one protein, or most likely something between one and one and less than two something...

so, in plasmids I have seen premade, they seem to have random dna inbetween each sequence

that DNA is not random, it comprises ribosome binding sites (RBSs). I don't know much about Eukaryota RBSs, but in bacteria it is quite simple, is has a -35 and a -10 region. Those literally correspond to 35 basepair upstream and 10 basepair upstream of the start codon. Together they are called the RBS region, you can't do without them.

Do you really have a 10kb plasmid that you designed yourself? And you don't know about RBSs and start and stop codons? If so you wasted a ton of time and money. I hope you are in the designing phase.

If you don't get everything I just said and you are in the designing phase, better dig in a bit deeper. You are going to fuck up.

[Edited on 13-12-2018 by Tsjerk]

DoctorOfPhilosophy - 13-12-2018 at 13:16

I think this really depends on how you want your proteins expressed (when and in what quantity). Often you try to control the protein expression so at the beginning of the cell culture there is no expression and the cells use all their energy to multiply and grow a healthy culture with many cells. Once the culture is mature, and has reached a certain density, you induce protein expression by adding a chemical like IPGT.

I think we can rule out having three plasmids. This is difficult because they are going to compete for DNA replication resources during cell division (especially if they have a high copy number) and I don't think they are going be stable in the culture. Even if perhaps you have three orthogonal selectable markers, and that's going to put a lot of stress on your culture.

With three proteins, you have to wonder what ratio do you want the proteins to be expressed at because the cell only has so many resources to go around. If you want to express in similar amounts, you can use one promoter for at least two (and presumably three) by having them share a promoter. What you end up with is a single piece of RNA with all proteins on it, separated by IRES (internal ribosome entry site) or a similar scheme (

I'm going to skip over some details but overall the a gene with a single protein (For expression) might look like this:

Promoter, [RBS, 6-8nt spacer, start Codon, spacer, His-Tag, spacer, Factor Xa recognition site, spacer, start codon, protein sequence, stop codon], terminator

So for multiple proteins you might double or triple up on the RNA part (in the square brackets), interleaved with IRES and spacers.

To answer your question about the spacers, they are often some pseudorandom mix of glycine and serine because these are simple and not likely to cause a side reaction or some unwanted secondary structures. The reason not to just use glycine is because of tRNA depletion (tRNAs are "expensive" so if you have GGGGGGG you're gonna run out of Gs and stall the ribosome)

You might also be seeing restriction sites, which are obviously there to help assemble (or disassemble) the plasmid.

Tsjerk - 13-12-2018 at 13:28

Please tell us what the genes are supposed to do, then I can probably tell you a lot more. A sequence would also help.

Jackson - 13-12-2018 at 21:57

My project is to make 5-methoxy seratonin, which is decarboxilated melatonin. I want to make some to see if it is orally active, and if it has similar sleep inducing effects like melatonin. I am basing the biosynthesis off the biosynthesis of psilocybin in psilocybe cubensis. In psilocybe cubensis the process is: decarboxilate tryptophan, add a hydroxyl group to the 4 position, transfer a phosphate group onto the hydroxyl group, and dimethlyate the amine that is not at the 1 position. My modified biosynthesis for 5-methoxy seratonin is: using human tryptophan hydroxlyase 1 or 2 to make 5-htp, use the psilocybin protein that methylates to methylate the hydroxyl group, use the psilocybin tryptophan decarboxylase to decarboxylate it to 5-MeO seratonin.

I assume that in the psilocybin biosynthesis the phosphotransfer step is done first because if the methylation was done first the hydroxyl group on the 4th position would also get methylated. I use this assumtion for the methylation step in my biosynthesis, but i am probably wrong about it.

Tsjerk - 13-12-2018 at 23:48

Are you doing this in a professional lab or in your own?

Jackson - 14-12-2018 at 00:46

Trying to fuse the plasmid with the yeast in a home lab, but in a sterile environment. The plasmid will be ordered from a synthesis company.

phlogiston - 14-12-2018 at 02:16

Why did you choose yeast? Bacteria are often easier and faster to work with. For this kind of project, I would probably try bacteria first, and only move to yeast if the protein is not expressed or not functional.

Also, yeast being eukaryotic, have you verified that these three enzymes sort to the same subcellular compartment?

I think you are underestimating the difficulty of this project. IMO you should break it up into smaller, more manageable pieces. Divide and conquer.

It would be much easier to design three plasmids, one for each of your proteins, and express each one in a separate strain of yeast/bacteria.
Once you are confident that they are indeed expressed and functional, you can test whether they work with your substrates.
(optionally) you can purify and concentrate the enzymes. Including a tag into the sequence of your proteins will greatly facilitate this. Look that up if you haven't heard of that before.
Then, you can test mixtures containing all three enzymes, either purified, or lysates, and see if the train works.
Then, move to expressing multiple of your enzymes in the same strain of yeast/bacteria.

[Edited on 14-12-2018 by phlogiston]

Jackson - 14-12-2018 at 03:00

In the paper that I read for the psilocybin synthesis that is what they did. They used an organism to synthesize the different proteins and then used the proteins for the reaction. I wanted to use yeast because the genes I plan to be working with are eukaryotic and so I inferred that they would probably be more comparable with a eukaryotic organism. That and also I can more readily get a pure strain of yeast as opposed to a pure strain of E. coli or similar.

phlogiston - 14-12-2018 at 03:56

It is indeed sometimes the case that yeast can assemble a eukaryotic protein that E coli won't. However, many eukaryotic proteins are well expressed in bacteria, and the yield is usually much, much higher. Especially in terms of the volume of culture you need to grow for a given amount of protein. One thing that I sometimes found to help with this is to swap codons in your sequence to codons that are frequent in the species of bacteria that you'll be using. Even only doing this with the first 10 codons or so can have a very profound effect on expression levels. (This is something you can do with yeast as well).
I don't know if it is really any more difficult to prepare pure strains of yeast or E coli. It is not very difficult to do anyway, certainly not compared to the other difficulties you will be facing for this project.

Also, have you considered asking the authors of those papers for their plasmids? It is not uncommon for researchers to share plasmids if you ask kindly.

Tsjerk - 14-12-2018 at 04:08

Whether or not you are going to use a prokaryot or a eukaryot, you definitely want a optimized lab strain, not a bakers yeast or a wild E. coli. Both are easy to get a clean strain out of nature of, but I don't think wild-type strains will give a very high production.

I think if your article uses yeast, they probably also tried in with E. coli, but that didn't work.

[Edited on 14-12-2018 by Tsjerk]

DoctorOfPhilosophy - 14-12-2018 at 06:13

Can you post the paper here? I orginally thought you were going to do the experiement in vitro but if all these proteins are supposed to work in vivo then that would be shooting in the dark. I share a lab with a group that does metabolic pathway engineering and they spend most of their time sitting at a mass spec trying to figure out what's in their lysates. Unless you have a plan to analyze the reaction intermediates and side products this might not be worth your time and money. Depending how close your experiement is to a known working one. I understand in vitro will be difficult without a giant floor centrifuge and expensive nickel beads so I can't even suggest that unless you have a lab.

If you just want to drink some 5-methoxyseratonin to see what happens, you might be better off making it by chemical synthesis.

Tsjerk - 14-12-2018 at 06:17

You need E. coli anyway to prepare the plasmid with. You only get a couple of nanograms when you have it synthesized and you need much more to transform with. Especially when you never done it.

DoctorOfPhilosophy - 14-12-2018 at 06:44

That's the easy part, you can buy DH5 alpha cells from the Odin for 10 bucks

Tsjerk - 14-12-2018 at 10:11

If you need a good protocol to make them nicely competent let me know, I will post it here. They were my pets for a long time.

Jackson - 14-12-2018 at 10:24

The supporting information document listed below the abstract has the full paper that isnt behind the paywall. I will link both the abrstact and the pdf

Paper behind paywall:


Tsjerk - 14-12-2018 at 10:51

I checked the article, and my next question is; do you have access to a mass spec? Do you want to show it can be done or do you want multi-milligram amounts in hand? What is your goal? Because if it is to have a more than 1 mg of pure material I have to disappoint you by telling you that you won't succeed. If you don't have a couple hundred thousand Dollars. But I guess if that were the case you wouldn't be here asking these questions.

In the article they describe the biological pathways by which psilocybin is made in mushrooms. For this you need a shit ton of work, but you don't have to isolate the product, you can show it with mass spec. Therefore the quantities of product produced only have to be minimal, MS is very sensitive. But if it is there with the right genes and experimental background there, you can show your hypothesis true or not.

This has nothing to do with production of viable amounts in real life, please have a look at the first five years of penicillin production... the organism in question was already producing the wanted compound. What I think you are talking about here is an order of a magnitude more work. There are very few organisms used in industry which were genetically modified and are producing economical amounts of secondary metabolites, because it is hard to do.

Please don't spent hundreds of Dollars on a synthetic plasmid if you want to make this compound on a gram scale, go for the chemical route.

Edit; you can easily graduate for a PhD on a project like this.

[Edited on 14-12-2018 by Tsjerk]

Jackson - 14-12-2018 at 12:14

My idea was to have all the protiens be produced inside the yeast and react in the yeast, and then to extract the compound from the yeast. If i end up attempting this i plan to use an ampicillin resistance gene in the plasmid to select for the yeast that took in the plasmid. This way i can use a small amount of plasmid material and just grow the yeast with the plasmid inside them to get a larger ammount of modified yeast without using a large ammount of plasmid material.

Also i dont need a large ammount because i want to test if its bioactivity is similar to that of melatonin.

[Edited on 12/14/2018 by Jackson]

Tsjerk - 14-12-2018 at 14:53

You are talking bullocks. As you didn't get what I was saying I conclude you don't have the knowledge to pull of this project. Please don't spend your money on a self designed plasmid as you will screw it up... You won't even be able to design the plasmid. You obviously didn't get what they were saying in the article you posted.

Jackson - 14-12-2018 at 15:05

I know that in the article they used the proteins to synthesize it externally but it trying to use it so that it is produced in the yeast or other microorganism. I do realize that I don’t have the knowledge to pull this off but I would like to know where I can learn more.


Tsjerk - 14-12-2018 at 15:15

I understand what you try to do, but that is not what they did in the article.

[Edited on 14-12-2018 by Tsjerk]

Metacelsus - 14-12-2018 at 15:23

Quote: Originally posted by Jackson  
If i end up attempting this i plan to use an ampicillin resistance gene in the plasmid to select for the yeast that took in the plasmid. This way i can use a small amount of plasmid material and just grow the yeast with the plasmid inside them to get a larger ammount of modified yeast without using a large ammount of plasmid material.

If you'd read the link I posted earlier, you'd have learned that ampicillin isn't a selection agent for yeast. It only works on bacteria.

Tsjerk - 14-12-2018 at 15:26

Quote: Originally posted by Jackson  

Also i dont need a large ammount because i want to test if its bioactivity is similar to that of melatonin.

[Edited on 12/14/2018 by Jackson]

For that you need large amounts, large amounts compared to what you will be able to extract from the yeast. This point I tried to make when talking about the difference between production and mass spec conditions. What you think are small amounts... are orders of magnitude bigger than what you will be able to make and what was shown in the article.

[Edited on 14-12-2018 by Tsjerk]

phlogiston - 14-12-2018 at 15:32

Jackson, your basic idea is not bad.
It is just that from the questions you are asking it is obvious that you haven't got the slightest idea of the magnitude of the challenge.
Things like ampicillin resistance for selection of transormants is extremely basic stuff, you are missing the actual problems you are facing.

Also, your notion of 'small amounts' is very different from what you can realistically expect to make by this method.
If you are planning to 'assay bioactivity' in the way that I think you are, you'll need many milligrams. But yields would probably be in the ng - ug range. (And it would be in a very dilute solution so you'd have to purify and concentrate the product. But that is a minor challenge after you made the yeast).

Tsjerks assessment that it would cost several 100's of thousands of dollars to do this is very true. Even with that kind of money and a professional lab it would still be a risky project that has a high chance of failure.

Tsjerk - 14-12-2018 at 15:33

Quote: Originally posted by Jackson  
If i end up attempting this i plan to use an ampicillin resistance gene in the plasmid to select for the yeast that took in the plasmid. This way i can use a small amount of plasmid material and just grow the yeast with the plasmid inside them to get a larger ammount of modified yeast without using a large ammount of plasmid material.

If you would believe a graduated molecular biologist... you don't need just a small amount when transforming yeast. Definitely more than you think; multiple micro-grams.

Edit: As I already said; this is PhD stuff. It took me years to master this stuff and I still didn't.

[Edited on 14-12-2018 by Tsjerk]

DoctorOfPhilosophy - 14-12-2018 at 17:25

Quote: Originally posted by Jackson  
This way i can use a small amount of plasmid material and just grow the yeast with the plasmid inside them to get a larger ammount of modified yeast without using a large ammount of plasmid material.

Also i dont need a large ammount because i want to test if its bioactivity is similar to that of melatonin.

One of the many reasons you might want to try in bacteria first is because indeed, you don't need a large amount of plasmid to try a single transformation. Maybe 100 ng (at most). Of course if it doens't go well you have to try again, and considering that you might purchase 250 ng from the supplier (IDT in this example), that only gives you 2.5 tries. That's not including duplicates or triplicates which are usually a good idea to compensate for mistakes or bad luck.

Now the real reason why people put their plasmids into bacteria even if they're working with yeast or higher organisms, is because it's a nice stable way to store plasmids that can readily create more plasmids. It's a lot cheaper to transform some 10 dollar e coli cells than to order plasmids 10 times. And if you try to store the plasmid in the yeast cell then you will have all kinds of drawbacks if you want to make more or extract ("prep") it. Basically it won't multiply fast and it won't be stable, and it may or may not hold well in the freezer, especially a home freezer which you open all the time and put large warm things into.

Also, it's worth considering that 5-Methoxytryptamine is only 30 bucks on sigma, so buying it might be a faster way to research its pharmacodynamics.

[Edited on 15-12-2018 by DoctorOfPhilosophy]

Jackson - 14-12-2018 at 19:18

Oh, if it’s that cheap on sigma and it’s food grade then I guess I don’t need to try and attempt this whole thing.