Hi Christoph,

There are a couple of things that come to mind when seeing this. Firstly, how do you prepare your protein lysate? Has the protein been purified using this FLAG-tag on a column using FPLC (or similar), or is the protein being expressed in a mammalian cell?

There are a number of things that can go wrong in either of these cases. For purified protein, if too concentrated, it may precipitate out of solution, once aggregated it will sit in the top of the gel and not migrate. For mammalian cells, firstly does the protein have a propensity to aggregate? and secondly how have you lysed/sheared DNA and then cleared the lysate. If you are not careful you can have incomplete lysis of cells and be adding cells or genomic DNA to the gel which will definitely cause that top band that you are seeing. Trying to get higher concentrations of protein by reducing the amount of lysis buffer just doesn't work, and will give you more grief later on.

In western blot, I usually find the old saying of 'less in more' very true. As an undergraduate I would try to load more protein thinking that obviously if there is more protein that I should get better bands, however this isn't the case at all. Lanes run wonky, bands are uneven and sometimes there is so much protein there that the HRP uses up the substrate while imaging and you get those fat bands with the white spot in the middle. Try and load no more than 30ug, I currently load around 15ug of whole cell lysate.

Looking at how the bands ran into the gel, I would say that the 5% stacking gel is not the problem either. The bands are straight and clean, if it was a problem of the gel being too dense, then you would see a smear not two distinct bands. In saying this, I do agree with the others that 4% or lower would generally be a better choice for the stacking gel. Also, unless there is a reason for not using the reducing agents (either DTT or b-mercap) such as looking at protein interactions then it is good idea to always use them with heating.

I apologise for the long winded post, but I hope it helps. Western blots are the bane of many molecular biology students, but once you have cracked it, they aren't so bad.

Good luck

Do you add some reduction agent as DTT or b-mercaptoethanol when you prepare your sample ? It might be some inter-chain disulfide bond.

You could decrease your percentage of acrylamid in both stacking and running gel to decrease the position of your protein in the gel (the better position is in the middle of the gel). For the stacking do you try at 3% ?

I didn't add b-mercaptoethanol. There shouldn't be disulfide bonds. This SDS-PAGE was run with a 5% stacking and a 12% running gel.

I'll give it a shot with your suggestions.

Do you think it could be a problem of simply to much protein loaded?

Your answer means that you have no cystein in your protein. If you have some, I will try with beta-Me in case of...

The stacking at 5% seems to be a little high, I prefer 3 or 4%. 3% is good the penetration of proteins in gel but is sometime not easy to use (too flabby....).

I think I will decrease the running speed. If you do 100V for 15min and 200V for 1H, you could try to do 70V for 20min and 150V until the end of gel.

For the amount of proteins, you could decrease the quantity but I think it is not the major difficult.

Thanks! Sounds good.

I'll give it a try..

I agree with above suggestions and also you may need to heat your sample in boiling water bath for 3-5 min.

Good luck

I agree with reducing your stacking to a lower percentage (not lower than 3% though), also I suggest increasing your running time and voltage, I have been successful with running to the end of the gel but it takes a while. In addition, the amount of proteins should be lower, have you tried reading the amount of protein you are loading?

Hi Christoph,

There are a couple of things that come to mind when seeing this. Firstly, how do you prepare your protein lysate? Has the protein been purified using this FLAG-tag on a column using FPLC (or similar), or is the protein being expressed in a mammalian cell?

There are a number of things that can go wrong in either of these cases. For purified protein, if too concentrated, it may precipitate out of solution, once aggregated it will sit in the top of the gel and not migrate. For mammalian cells, firstly does the protein have a propensity to aggregate? and secondly how have you lysed/sheared DNA and then cleared the lysate. If you are not careful you can have incomplete lysis of cells and be adding cells or genomic DNA to the gel which will definitely cause that top band that you are seeing. Trying to get higher concentrations of protein by reducing the amount of lysis buffer just doesn't work, and will give you more grief later on.

In western blot, I usually find the old saying of 'less in more' very true. As an undergraduate I would try to load more protein thinking that obviously if there is more protein that I should get better bands, however this isn't the case at all. Lanes run wonky, bands are uneven and sometimes there is so much protein there that the HRP uses up the substrate while imaging and you get those fat bands with the white spot in the middle. Try and load no more than 30ug, I currently load around 15ug of whole cell lysate.

Looking at how the bands ran into the gel, I would say that the 5% stacking gel is not the problem either. The bands are straight and clean, if it was a problem of the gel being too dense, then you would see a smear not two distinct bands. In saying this, I do agree with the others that 4% or lower would generally be a better choice for the stacking gel. Also, unless there is a reason for not using the reducing agents (either DTT or b-mercap) such as looking at protein interactions then it is good idea to always use them with heating.

I apologise for the long winded post, but I hope it helps. Western blots are the bane of many molecular biology students, but once you have cracked it, they aren't so bad.

Good luck

Thanks for the help guys!!

I ran a new gel yesterday. I used half the amount of protein and I boild the samples with b-mercaptoeth. for 5 min. Unfortunately the I didn't have the opportunity to change the percentage of the SDS. I'll do this the next time. I'll upload the pic of the WB later this afternoon.

Regarding your questions:

I'm expression a membrane protein in mammalian cells and I'm purifiying it with the "Mem-PER Eukaryotic Membrane Protein Extraction Kit".

We are currently using the bradford assay in our lab, which is not compatible with the high amount of detergent in the hydrophobic phase. So I don't know exactly how much protein I'm loading.. Anyway the ponceau staining is showing me that it was really a lot..

I also read that the high amounth of detergent in the samples after using the kit could cause the bad running of the samples.

Here the result, after I loaded half the protein and added beta-mercaptoeth.

In the next run I'll decrease the stacking gel concentration and also I'll dilute the sample more for having a lower concentration of detergent.

Hi Christoph,

Getting one clean band is half the battle won so good work.

Another suggestion though, when you load the gel do you have any empty wells?

Looking at the new blot I think you loaded empty well/sample/ladder/sample/empty well. Often (depending on gel percentage) as the proteins run down the gel they start to get narrower just as a force of being push through the gel, similar to your ladder. However when a sample has a empty well next to it, the sample will start to spread out like in your sample lanes (compare the width of your sample to the width of your ladder).

When you run the gel next time, make up some 1x loading dye and, for example, if you put 15uL of sample in your wells, add 15uL of 1x loading dye to all empty wells. This helps the proteins migrate evenly across the whole gel and will give you even sized bands.

Note that this band size effect can also be seen if loading a small volume next to a large volume. Try to keep all samples/wells roughly the same volume, just adjust with the 1x loading dye.

Good Luck

Nice suggestion Jacob. Thanks for that. I'll try it out asap.

There are two more points that may help you: first try to sonicate the sample (after boiling), cool it to 4 degree using ice and centrifugate it and then load the supernatant into the wells.

digest your protein more & dilute it too before run on SDS page

Another way is to change the protocol

detergent in your solution sometimes interfere with SDS making crosslinks and blocking proteins from passing through stacking gel

Hi Guys,

after I ran the nice gel (see above). I ran yesterday a gel for the detection of a different membrane protein with 4% stacking and 10% resolving gel. Also I diluted the lysate from the membrane prep 1:2 to reduce the concentration of detergent and added beta-mercaptoeth.

I ran the gel at 100V for 1,5 h.

I guess I have the same problem again..

Next time I'll sonicate the sample and spin it down as Mohammad suggested.

Any other suggestion??

Thanks,

Chris

Hi Christoph,

If you think that the detergents in the membrane extraction kit are causing problems, you could try to get rid of the detergent after the extraction. There are a number of different methods, you could use a spin column with a low molecular weight cut off and dilute your samples in a more appropriate buffer. A second option could be to dialyse the samples or a third would be to precipitate the protein extracts with TCA, and then resuspend in a different buffer.

After that, measure protein concentration and the proceed how you have been.

Good luck

Hi Christoph,

I work in the field of membrane protein crystallography and we never boil the samples when running a SDS-PAGE gel. We just the mix the protein with the loading buffer (with SDS and beta-mercaptoethanol) and it works perfectly fine. The only thing is that may be you won't disassemble membrane proteins that form oligomers (I work with a membrane protein that forms trimers and if I don't boil the sample I have a trimer and if I boil it I have a monomer).

Some membrane proteins when being boiled they aggregate and don't want to enter the gel or migrate properly.

In fact, I know someone who had a similar problem. He wanted to detect an eukaryotic membrane protein from a cell lysate by Western blot but the migration was not good. The protein seemed to have a higher molecular weight. He always boiled the sample. When he didn't boil the sample, the membrane protein migrated properly.

May be your problem is due to this. Give it a try.

Good luck

Thanks a bunch for this suggestion Patricia. Yesterday I run a WB and I sonicated the sample after boiling. Unfortunately I got the same result as posted before. So I won't boil it this time and do the WB again..

add DNase or sonicate, spin down insoluble debris at 1400 rpm for 20 min at 4 Deg C. It seems like a lot of unsheared DNA or cellular debris. Your protein may also be partially insoluble, in which case you need to play with buffers.

I did the Western Blot without boiling the sample. Still the same result.

Any other suggestions?

Hi Christoph

You may need more SDS, and an incubation at 60 degrees or so rather than boiling. The attached paper might help.

http://www.malariaresearch.eu

Whoops here it is

Hi Christoph,

Sorry for all the problems you seem to be having.

Have all of the western blots been from the same extraction, or have you been doing multiple extractions?

If you have re-extracted you could check that something hasn't gone wrong there. If you happen to have kept the latest membrane membrane (and you used PVDF) you could always re-probe the membrane for the flag tagged protein to see if it migrated properly. To see if it is a sample prep issue or if it has something to do with this particular protein.

If you know of someone in your institute that uses BCA to measure protein concentration, it might be worth while asking to use some. Just as a point of reference as to how much protein you are adding. If your protein extracts are too concentrated, you may not be able to reduce the proteins efficiently.

Do you have access to an ultra centrifuge? If so, you may want to try extracting the membrane fraction using this ultracentrifugation. If your problem is from the solvents/ detergent etc from your extraction kit, you might avoid issues that way. Also, I read through the manual for the kit, it suggests that you can dilute down as much as 1:5 so if the detergents are really the problem that might help.

A last thought, what are you trying to show using the membrane fraction? Are you using this fraction just to confirm that your proteins are in the membrane or are you doing something else with the proteins? For example, if it is localisation that you want, you could try immunocytochemistry as an alternative,

Obviously not getting the kit to work is frustrating, but if you tell us more about what you want to see/show, we could potentially come up with an alternative solution if this problem can't be resolved quickly.

Good luck

Hi there,

to answer Jacob's questions:

-I did different extraction. They all look the same. Unfortunately i don't have excess to an ultra centrifuge.

The whole point of the extraction and WBs is just to show that there is an expression on the protein I'm working with. This protein is a ion channel and is located therefore in the membrane. This is the only reason why I'm doing a membrane preparation.

Also I'm already seeing an physiological impact, but now I need to show the expression of the protein to make the data more conclusive.

What are your thoughts about that?

Hi Christoph,

While I have have heard that some membrane proteins can be quite difficult to detect with western blot due to their low amounts, you have a protein that you are over expressing you shouldn't have too much trouble doing a whole cell lysate. This will show you that you have overexpression.

Overepression sometimes isn't enough though, so since you are looking at an ion channel and it should be located in the plasma membrane then you should show that it is there; which is why you would be doing the membrane fractionation kit. If you cannot get the fractionation kit to work, then I can only suggest immunocytochemistry for localisation.

Sorry that I can't be of more help.

Good luck

Hi Christoph,

I have one question for you: where did you find procedure for preparation gels for electrophoresis?

In the Roche Lab FAQ

you need to check the % of acrylamide that you use and it is possible that your protein is as aggregate and the migration in the gel is different. have you another method that you can check if it is as aggregate or not? have you check the protocol for sample preparation before SDS-PAGE?

I have experience that one wrong chemicals can stop migrating of protein trough SDS-PAGE gel. Therefore you should check what reagents you use for gel preparation.

All above suggestions are valid and based on their  extensive research experience. 

If you have time: You may also try

(i) Ultrapure ammonium persufate and SDS, Polyacrylamide, Bisacrylamide from Bio-Rad as they specialize in SDS-PAGE biotechnology. 

(ii) Use Whatman filter to purify your buffer before use. Usually proteins migrate better in an alkaline (7.8-8.5 pH) better. 

(ii) The migration will also depend opn the net surface charge on the protein.

(iii) The migration will also depend on the molecular weight on the protein.

(iv) If you you do not want to sonicate briefly at 4C, you  may pass the lysate through 23 gauage needle by a microsyringe. 

(iv) I fully aggree with one of the expert that loading of proteins in a lysate should not be molre than 5-15 microgram/lane in a minigel. 

(v) Use a full antioxidant cocktail containing DTT, PMSF, beta mervaptoethanol. You may make you own or purchase ready made. 

(vi) Run  say 60v first for 15 min without loading the  protein sample to make the tracks free of any hindrance during electrophoretic run. 

Keep the entire electrophoretic unit  on ice pack or in the cold room at 4C. 

Usualy 150V for 1.5 hrs at 4C or maximum 200V for 1 hr at 4C should work. 

Please do not waste your antibody. First make sure that your transfer on to the PVDV or nitrocellulose membrane is complete by doing a Ponceao Red stain. 

The success of blot also depends on your basic knolwedge of preparing the proper buffers (pH, molarities, temp etc). Keep  all the solutions at 4C except SDS solution. 

Please estimate your protein concdntration by using sensitive Bradfor's reagent (1-20) by preparing the BSA standards.  These steps may help you in the long run. 

Most Imp: Make sure that the electrical connections are correct and your electrophoretic appratrus ( voltage supply unit is providing the desired and constant voltage during the entire run). 

Good Luck. 

Thanks a lot for the detailed suggestions. 

I'll update you as soon as I figured something out. 

Cheers, 

Hey Christoph,

I'm currently fighting with the same problem than you. All my blots were fine but as soon as I started with FLAG-Tag, I got the same problem than you. 

I tried with more SDS, longer boiling, some DTT but it's still stack between stacking and resolving gel.

Have you found a solution between your last post and now?

Thank you

Hi Celine, 

Now I'm using a 4% stacking gel and a 7.5% resolving gel. I'm adding beta-mercapto to the Lämmli buffer and I'm incubating at 37°for 15 min.  Also I changed the extraction from mem-per buffer to m-per buffer. I think that there are just too many membrane proteins in the prep and they agglomerate. With this procedure I'm getting the band at the right height, but still there is a significant part stuck at the top of the gel. 

I hope this helps a bit. 

Cheers, 

chris

I have some question. how you make your lyasate.  I mean after cell lysis, do you spin your samples so that all the cell debris are gone. Otherwise they create a lot of problem while loading in the gel. You can heat the sample to 90 degree for 7-10 minutes after mixing with Laemmli sample buffer. I hope this helps

Good luck

 Hey Christoph,

I have the same problem with you. The protein aggregate on the top of the running gel. In your lammli buffer, you add both DTT and beta-mercapto? 

Thank you.

hi everybody

I have a problem as you. i have pepsin-solubilized collagen from sea cucumber body wall.

then i hydrolyzed it by Glu-c V8 protease according to protocol that used by (Feng-xia Cui et al., 2004) and run the hydrolysate on SDS-PAGE 5% stacking and 15% separating.voltage was 120 v and electricity current intensity was 0.6 mA .but after passin 1hr not samples, markers, and blank loading buffer moved.

what is the problem in your idea?

please help me.

thanks