Hi Joana, it is not unusual to have different levels of GAPDH or actin in different cell types, for this reason it is very difficult to quantify among different cell types or tissues using single protein references, although I agree that you have quite dramatic differences and the lack of actin in some of your cells is a bit suspicious. Yet, keep in mind these two points:

GAPDH is involved in the energy production cycle, glycolysis, so it is typically more expressed in cells with high energy requirement (muscle, brain, tumors) then cells with lower energy requirement.

Actin exists in two forms, filamentous (F-actin) and monomeric (G-actin) in constant and dynamic equilibrium, but with the difference that monomeric actin is soluble in most buffer conditions while the filamentous is triton/NP-40/tween insoluble and only solubilizes in strong detergents such as SDS and partially at high pH. So, depending on the cell type, the growth conditions, the extracellular stimuli and the lysis you will extract different amounts of actin that don't reflect the total protein content unless you are doing a total cell lysis in SDS (for example boiling cells in Laemmli buffer). The strange thing with your HL60 cells is that you normally always have a good amount of G-actin, so although it might vary depending on many factors, there should still be enough to detect it on a WB.

So, two suggestions:

I would grow a new culture of HL60 (and the other cell lines you are unsure of) being careful tto the culture conditions and test them with a total cell lysate.

If you can do total cell lysates (dierectly in Laemmli) for your quantification experiments, then I would use g-tubulin for calibration: this is found only in the centrosome, and in non dividing cells there is only one, so it is mostly proportional to the number of your cells. But you need a total lysate for this because contrary to alpha-tubulin that collapses completely at 4 C, gamma-tubulin is only partially soluble in mild detergents and 4 C... Anyway, even better if your Ponceau works well (on PVDF it is not that reliable, but maybe you use nitrocellulose) that is also a very good quantification method, since you can measure the total protein transferred to the membrane in each lane.

Good luck.

Hi Joana, it is not unusual to have different levels of GAPDH or actin in different cell types, for this reason it is very difficult to quantify among different cell types or tissues using single protein references, although I agree that you have quite dramatic differences and the lack of actin in some of your cells is a bit suspicious. Yet, keep in mind these two points:

GAPDH is involved in the energy production cycle, glycolysis, so it is typically more expressed in cells with high energy requirement (muscle, brain, tumors) then cells with lower energy requirement.

Actin exists in two forms, filamentous (F-actin) and monomeric (G-actin) in constant and dynamic equilibrium, but with the difference that monomeric actin is soluble in most buffer conditions while the filamentous is triton/NP-40/tween insoluble and only solubilizes in strong detergents such as SDS and partially at high pH. So, depending on the cell type, the growth conditions, the extracellular stimuli and the lysis you will extract different amounts of actin that don't reflect the total protein content unless you are doing a total cell lysis in SDS (for example boiling cells in Laemmli buffer). The strange thing with your HL60 cells is that you normally always have a good amount of G-actin, so although it might vary depending on many factors, there should still be enough to detect it on a WB.

So, two suggestions:

I would grow a new culture of HL60 (and the other cell lines you are unsure of) being careful tto the culture conditions and test them with a total cell lysate.

If you can do total cell lysates (dierectly in Laemmli) for your quantification experiments, then I would use g-tubulin for calibration: this is found only in the centrosome, and in non dividing cells there is only one, so it is mostly proportional to the number of your cells. But you need a total lysate for this because contrary to alpha-tubulin that collapses completely at 4 C, gamma-tubulin is only partially soluble in mild detergents and 4 C... Anyway, even better if your Ponceau works well (on PVDF it is not that reliable, but maybe you use nitrocellulose) that is also a very good quantification method, since you can measure the total protein transferred to the membrane in each lane.

Good luck.

Hi, thank you for your advice!

Well yes I use PVDF...but in fact I repeated the Ponceau for other membrane yesterday and this time I really took a closer look and saw that those lanes had slightly fainter bands, but the most noticeable difference is the absence of a band above de 40KDa which I believe is the b-actin. I already repeated the extraction from HL60 and gave me the same thing! What seemed to help (because now I can see a band but not as thick anyway) was to increase the concentration of protein in the sample by lysing in less volume. Can this have some influence? I had a sample of around 1 ug/uL and other of 2.5 ug/uL, but loaded the same amount (10 ug), one shows a band the other don't. 

Or can something be interfering with the quantification method? I wash the cells twice with PBS, so it can't be from serum. Neverthless, the concentrations that I measure are very similar to the other cell lines with the same concentration of cells and the cells are roughly the same size, so it shouldn't be that.

But I will definitely follow your suggestion of the total cell lysate. Thank you. 

Hi again, just to answer your questions, although this does not yet solve your problem. Bradford, BCA, Lawry quantification are of course all approximate, normally overestimating your real concentration due to the effect of detergents such as Triton or SDS, unless you are careful to dilute your extract 10x in water and quantify 10 ul (it's a good trick to increase accuracy). Anyway, as you say, this happens for all your extracts, and if your Ponceau is comparable as I understand, that's not an issue. It could in some cases be an issue if you load too little protein because Western blotting, for several reasons depending on the proteins, antibodies and the conditions used for the incubation, work up to a threshold level of the protein on the membrane, which means below this threshold quantity that is specific for every antigen-antibody interaction you lose the signal, non-linearly. But I doubt this happens with actin. Anyway it is interesting that also with Ponceau you don't see the actin band in the HL60 cells, it seems therefore it is related to the cells. It is possible that they are very rich in F-actin and have reduced G-actin availability... I would have not guessed it from the cell type (growing in suspension normally requires less F-actin because there is no adhesion and no cell crawling), but who knows... Total extract will tell you...