Hi Stefanie,

I'm not sure what is your aim, are you checking enrichment in the fractions (also not sure what fractions you are talking about, cytoplasmic and nuclear, or organelles and membrane fractions, such as Golgi, ER, cell membrane...) compared to the WCL within each transfected line or are you interested in differences in the WCL and in each of the fractions between the GOI and the EV transfected lines ?

The reason for this question is that if you are checking enrichment between the WCL and the fractions within each transfected line, you are supposed to load same amount of protein from each sample (WCL and each fraction) but you don't have any gene for calibration (no gene can be assumed to distribute in the same amount in different fractions/cell compartments), so you will have to rely on Coomassie staining of your lanes on the membrane after you finish with your WBs to calibrate your samples. But if you are checking the same fractions between the GOI and EV transfected cells, you should use specific fraction markers to calibrate the samples from the different cell lines. In other words, no fractionation will be equal, in terms of purity of fractions and final yield of each fraction, so you should check all your fractionations with specific markers for the different fractions, and use these markers to calibrate among the different fractionations (GOI and EV transfected cells).

If I knew what fractions you are producing I could possibly suggest some markers, but with some reading on the topic you'll also find them yourself!

Good luck!

Hi,

I am interested in differences in the WCL and in each of the fractions between the GOI and the EV transfected lines. I want to see the difference in protein expression in WCL first and then, in the cytoplasmic/nuclear fractions I want to see where the protein is: cytoplasmic or in the nucleus? So normally, when I see an upregulation in WCL between  GOI and EV lines, I have to see it also in the fractions. For example: One of my proteins shows an upregulation in WCL and I also see that most of the protein is located in the nucleus. Problem: When I compare the nuclear fractions, the bands have the same thickness. So where is my upregulation??

As purity markers, I use GAPDH for WCL and cytoplasmic fraction and Histone H3 for WCL and nuclear fraction. These bands have the same thickness, so besides purity, I see that I have loaded the same amount of protein in each fraction.

Hi Stefanie,

ok, I see.So a few remarks.

1. I guess if you are sure the genes you overexpress are not affecting cell metabolism or cell growth, GAPDH would be fine as a calibration gene... but be careful how much you load, GAPDH is highly abundant and you have to make sure you don't saturate the membrane with it when you transfer. We usually load max. 10 ug of protein per sample if we use it and use the shortest exposure times (making sure you don't see any white or light gray area in the center of the GAPDH band... that's a sign of having too much protein there that burns out very quickly the ECL reagent... but you see it best at short exposure time).

2. You talk about band thickness, not intensity. So are you using a film to acquire your WB signal or a digital system? WB quantification is very delicate, and already using ECL has its issues, due to the amplification reaction. In any cae with a digital system you can have shoert exposures (for example 5 sec) that you can progressively cumulate to see how the intensity of the band increases and make sure you are in the linear range for quantification. Films have a very short linear range, so you have to be very careful with the exposure times...

3. I believe I'm still missing one point... You say that "One of my proteins shows an upregulation in WCL and I also see that most of the protein is located in the nucleus. Problem: When I compare the nuclear fractions, the bands have the same thickness."... Ok, so how do you "see" that the protein is located in the nucleus if your WB does not show it? Do you have a fluorescently labelled protein or, anyway, you stain the protein by IF or ICC, and you see it under a microscope? If you have this evidence, you might want to check the purity of your nuclear fractions in the opposite way... Use GAPDH on your nuclear fractions to see how really "nuclear" they are... if you have cytoplasmic contamination in one of them (keeping in mind that a cytoplasmic fraction has 10-20 times more protein than a nuclear one) you might be diluting out all your nuclear proteins and, therefore, also the possible enrichment of your GOI.

I don't know if any of this will help, possibly some internal help from your lab with a better control of the experimental conditions might be more fruitful! Anyway, good luck!

Hi,

maybe you understood me wrong...

Our lab uses the ECL and digital imaging method. On one Gel, I load WCL, cytoplasmic and nuclear fractions, so I can see where my GAPDH/H3  markers are: GAPDH not in the nucleus (and without white/grey area in the band) and H3 not in the cytoplasm.

What I mean is: When I see an upregulation of a protein in the WCL, I have to see this upregulation also in the fractions where the protein is located. I saw an upregulation of my protein in the WCL (GOI to EV), I saw not so much protein in the cytoplasm (GOI and EV) and much protein in the nucleus (also GOI and EV). So, it is located mostly in the nucleus. But when I have an upregulation that I can see in the WCL, I should see this upregulation also in the nucleus: more protein in the GOI nuclear fraction than in the EV nuclear fraction, as it is in the WCL. And not two similar bands... For help, I attached a picture of my blot: See left to right: Marker, EV WCL; GOI WCL; EV cytoplasm; GOI cytoplasm; EV nuclear fraction; GOI nuclear fraction. I hope it is better understandabe now what I mean.

Thank you for your help! :)

PS: the GOI is not the protein in the picture! I only have cell lines with EV/GOI and I analyze them for other proteins. This one in the picture is one of them.

I have seen varibility in subcelluar fractionation with diffrent styles membrane disruption. The ER can become complicated to fractionate.  I found nitrogrn cavitation to be a handy tool.

Hi Stefanie,

good job, at least your problem is now perfectly clear! But I'm sorry to say that I don't think I have a direct solution to it... I can only try a few wild hypotheses!

For example, in this blot the first sample (EV WCL) looks suspicious... background is lighter compared to the sample on its right (the GOI one) and there seems to be a lighter arch invading more than half the width of the band... maybe some blotting problem? Have you tried running the WB a second time? I've no clue at what height your band is, but would you have another protein close by (a bit higher or a bit lower) to test and see if it behaves similarly (assuming that GAPDH is similar in the first two samples, as you assure...)? Do you have any evidence that you are looking at the specific band? (often Abs, especially commercial ones, are not so specific as it is claimed, with the most intense band not necessarily being the specific one, and non-specific bands can be affected by many uncontrollable factors in the different samples)...

Try putting together all your WBs, cut from the original files the relevant strips for the genes you have tested (like you show above), plus GAPDH and H3 and place them one below each other in a ppt slide or an ai file or whatever you use, and try to see how much makes sense in which blots and what is not convincing... The only way to spot experimental problems is to doubt oneself, and be hypercritical...!

Good luck!

 Hi,

thank you for your ideas.

GAPDH and H3 are always similar and I assume that I am looking at the specific band, because the height is ok...

But I have another idea and I'm not sure if it makes sense:

For preparing the nuclear fraction, I put the nuclei lysis buffer on my pellet (after getting my cellular fraction), lyse it for 30 min on ice, sonicate the lysate and after another centrifugation step, I take the supernatant as nuclear fraction. Question: In this nuclear fraction, there are soluble nuclear proteins and chromatin (mainly euchromatin?), right? And could it be that in the pellet after the last centrifugation step, there is mainly heterochromatin? Could it be that I lose some proteins associated with the heterochromatin because I discard the last pellet? Should I use the whole lysate after sonification without centrifugation for WB?

Well Stefanie, if you were losing material I could have supported your idea, in the sense that protocols indeed count on what you recover at the end, although I would not focus too much on eu- and heterochromatin, since in your nuclear lysis conditions (high salt or urea or SDS) and after sonication (that essentially is one way to fragment gDNA to reduce the viscosity of the extract and facilitate gel loading and run) there will be little chance that any gDNA structure is left.

The difficulty of troubleshooting your problem is that while in the WCL you seem to have an expression different of the tested gene depending on the expression of the GOI, in the nuclear lysate this difference is lost, but not because you lose the signal in both conditions (transfection with EV and GOI), but because you appear to gain signal in the EV condition. This is why my doubts fall on your EV WCL in the specific example you uploaded... The only possible explanation seems to be that something went wrong during the loading or the run or the transfer for that sample, if you are absolutely sure that you didn't inadvertently load twice the GOI nuclear lysate (this you might be able to confirm probing the same membrane for your (over)expressed GOI if also this has nuclear localization)...

Anyway, of course there are different possibilities to obtain cytoplasmic and nuclear fractions. For the cytoplasmic fraction there is an increase in lysis strength going from Tween to TritonX to DOC. And as far as the nuclear fraction is concerned, there are several possibilities, for example: 1. sequential, but partial, extraction with medium ionic strength (250-300 mM NaCl or KCl) and then high ionic strength (450-500 mM salt), which normally leaves the gDNA in the pellet; 2. lysis of the pellet with 8M urea and sonication; 3. lysis in 2x Laemmli buffer and disruption of the gDNA by vortexing for 10-15 sec every 2 min while boiling the proteins for the classical 10 min. No method will solubilize all your proteins, there will always be a pellet, with the last two methods indeed very small! Nevertheless, you don't seem to have a problem there, you have the signal in the nuclear lysates even where you would not want it (EV!)!

Think a little bit about this over the holidays, then, if you have not done it, just repeat the WB to make sure you can rule out technical problems after the lysis. If things remain unchanged, I would make a new culture and new lysates...