Are the proteins you're testing yeast proteins or proteins from another organism?

I've not worked with yeast before.

So, let me know if I have a grasp of what you're attempting to do:

• There are two protein, a Bait and a Prey Protein. It sounds as if they are surface protein that are a receptor and ligand pair.
• The bait protein is expressed on the surface of one yeast and the bait protein is expressed on the surface of the other yeast
• If the yeast come into close proximity, the bait and prey protein can bind.
• If the bait and prey protein bind, a signal is generated with results in the expression of a reporter gene or some sort...

Do I have that correct? If so, it sounds like very cool research.

As I said, I've not worked with Yeast before but I have investigated receptor ligand interactions of cell surface receptors in mammalian cells. I believe the concept is the same...

Watching the interaction of protein on the surface of two different cells is tricky. The cells aren't exactly cooperative, they don't necessarily line up as we want them to, and there's so many other interactions going on that things get convoluted.

Could you replace one of the yeast with a glass-supported lipid-bilayer onto which you put either the bait or prey protein? I'll try to add an image here...

If you've not done this before, it might sound complicated, but, like most techniques, once you learn how it's not terribly difficult but it can be extremely useful if you're interested in protein protein interaction between cells.

There is a huge benefit to monitoring the interaction on a glass suported lipid bilayer. You can use a confocal microscope (or any microscope really) to monitor the interaction real time. If you have a fluorescent reporter gene, then you could use a confocal to record 3D images in real time.

I know you'd prefer to get your current system working, I just thought I'd put this out there in case you needed a backup system. Good luck with it!

Dear Dominik,

You may have many problems. Let us list them:

1. DNA codons compatibility. Yeast will tolerate many DNAs but some are simply not very good. You may overcome it by using a system which is closer to the original DNA than yeast. Another option would be codon optimization which is long and costly.

2. Role of the protein. You may have metabolic toxin/interference by your protein which is not compatible with the Y2H. The simplest way would be to look for a fingerprint of the protein and see if they would be problematic with your assay.

3. You may have transmembrane regions somewhere which are a big problem for a simple Y2H. The easiest way to fix it is to use bioinformatics to learn more about the protein's structure and topology. If they have TM domains, use the soluble part in your bait. Alternatively, use a TM-dedicated Y2H which should work if you set the domains for the assay.

4. Try to express the protein and use a classic binding assay. It should dispell any doubts about interactions but will take more time.

Hope it helps

Wes

Thank you all for your answers!

Waleed Albihlal all proteins we used were from Homo sapiens. So we were not testing yeast proteins.

Tim Tolentino With the yeast two-hybrid system you are not looking at the interaction between two proteins on the surface of two cells, you are looking at the interaction of two proteins inside one yeast cell. If the two proteins interact there will be some reporter genes expressed helping the yeast cell to grow on selective media.

This sounds like a really interesting technic I will have a look at it.

Wieslaw Swietnicki We checked for transmembrane regions and before but none of our proteins have TMs. Codon optimization might be a possibility. As I said, we already validated some of our interactions via biochemical protein interaction assays but we wanted to validate those possible interactions in yeast first because we found them via a previously done Y2H screen.

Hello Dominik,

Y1H, Y2H and Y3H are not as straight forward as they might seem. There are a whole bunch of things that can go wrong in those experiment not because of your experimental design, your handling or anything like that. It is mainly related to nature of the experiment itself and the points @Wieslaw Swietnicki mentioned in his answer. You also have to bear in mind that yeast two-hybrid and its derivatives are notorious for showing false negatives and positives. Having said that, here are a few things you might want to check before you run your next experiment:

1. Always check with western blot that the proteins you're testing are expressed.

2. Do the proteins you're working with have homologs or orthologs in yeast? I've had this issue before with Y1H and Y2H when I was working with plant proteins. They worked very well with plant-specific proteins but not with conserved proteins where there are similar homologs and orthologs in yeast and I believe but not really sure about it that if you're testing highly conserved proteins between human and yeast than those protein will be subject to many other interactions with endogenous yeast proteins that may prevent the test proteins from interacting with each other.

3. You mentioned that those proteins have already been tested with Y2H and you're trying to reproduce the result. Maybe you want to try to follow exactly the same protocol and use the same (yeast strains) they used in that previous study. For instance, I normally don't do Y2H with co-transformation, I use two different mating-type strains such as AH109 and Y187. Transform one with the bait plasmid and they other with the prey. Grow yeast strain in separately in liquid YPD (make sure the plasmid is not episomal) then drop 3 ul of the bait strain on YPD plate, wait for it to dry then drop 3 ul of the prey strain on top of the bait strain. Incubate overnight at 30C then streak on a plates with double selection to select diploid cells. After that you can run the assay you want depending on what reporter your using e.g. HIS3, LacZ ..etc

Best of luck

Waleed Albihlal thanks for your help!

We tested both, co-transformation and mating. Also the protocol and yeast strains (PJ694A/alpha) were the same like in the previously done screen.

Did you ever tried to validate the interaction with full length proteins in Y2H or did you always used domains of the bait and prey proteins?

I used full length proteins in both Y1H and Y2H had good success except with some of the highly conserved proteins such as metabolic enzymes and chaperons, some worked and some didn't work at all.

Maybe you can tell use a little bit more about your experiment:

1. What reporter are you using?

2. Have you tried swapping the GAL domains because as this also can be problematic in Y2H. For example, in your case you have GAL4-BD-Protein1 and Protein2-GAL4-AD, you could try swapping them and making something like GAL4-BD-Protein2 and Protein1-GAL4-AD

best of luck

We are using the HIS, ADE, MEL1 and LacZ as reporters. All reporters worked in our controls (from HybriZAP-2.1).

We are using the Gateway vectors pAD_GAL4_2.1 and pBD-GAL4_Cam as Destination vectors for our proteins. Both vectors code for an N-terminal tag. We actually do not have any other vectors for C-term fusion of Gal4 AD/BD.

All our constructs were sequenced and checked if they are in frame. The expression of our fusion proteins were checked on WB.

Dominik,

I'm sorry. I misunderstood exactly what you were doing...

It sounds interesting...

Good luck with it!

Hi i am Dr. Asif Ali, if you want to know about positive and negative control please go though this link.

It will guide you a detailed protocol

https://youtu.be/c4pIacQ9uM0

https://youtu.be/bJg0eQ7Zc-Y

Regards