Another option is to separate the protein sample on a SDS-PAGE and then also add a standard protein (with known concentration at different dilutions). You can then scan the gel and relate the band intensity to the protein concentration (to get a calibration curve of the standard) and finally relate the intensity of the protein of interest to the standard bands. There is free software to analysis the band intensity in the internet.

You could maybe try adding EDTA to your sample after IMAC purification and perform a buffer excahnge by SEC or dialysis and see if you can get rid of your contaminant in this way.

Alternatively I am wondering how much error your contaminant would account for in a Bradford assay. I assume the colour intensity is concentration dependent. So if you check a 1:100 dilution in a Bradford assay I would think it should not interfere with your readings?

Unfortunately the colored pigment is within the size range of the proteins I am isolating, so SEC and dialysis isn't an option. (I'm isolating a family of proteins, with varying MW (30kD - 250kD) and PI (anywhere in the neutral and basic range)

I'm isolating my protein from sea urchins, and each urchin tends to have a different concentration of this colored pigment, unassociated with the concentration of my protein of interest. This effet on the coloration ranges from almost clear to grape-juice purple, and I expect to have a low protein concentration to begin with (with the colored pigment, some concentrations are as low as 0.04mg/ml - I got similar readings of the sample on both A280 and Bradford) so I don't think diluting the sample further is an option (the minimum concentration Bradford will detect something like is 0.002mg/ml I think).

Why do you think it should matter? The protein preparation is usually colored in different ranges from yellowish-orange to brown.

Visible light corresponds to a wavelength range of 400 to 700 nm, why do you care measuring at 280nm??

Well, as Bradford assay detect color changes, adding a really dark sample versus adding a light sample will likely alter the readout. The pigment also interferes with the A280 reading (based on an attempt to isolate just the pigment....)

I once had a sample with somewhat purple concentration and a high protein concentration readout; I made serial dilutions and read them by both Bradford and A280, and neither set produced accurate concentration dilution factors in the readouts.

did you try to measure absorbance only of your sample, what spectrum it has? The sample is much diluted in Brandford, so I highly doubt, that it will have any significant effect. Just try to dilute your sample into pure water just as you're diluting it into Bradford and try to measure the absorbance.

My most recent sample was a very deep purple (could not see through it) and I tried quantifying it by A280. The readout was obscenely high relative to my expected ballpark value (4.3mg/ml vs 0.5-1.5mg/ml, respectively), so I tried diluting it 1:10 and 1:100 in ddH2O, which resulted in the absorbance reading 1.91mg/ml and 0.6mg/ml respectively. I don't have enough of any given sample to do a series of serial dilutions for Bradford and then run my experiments...

We do not know the absorbance coefficient for our protein.

I'm not suggesting to measure in UV with such sample, but for Brandford it should not be a serious problem.

(Also, Bradford measures A595, which is pretty far from UV, so it should not have significant absorbance, even if it has strong reading at A280.)

Another option is to separate the protein sample on a SDS-PAGE and then also add a standard protein (with known concentration at different dilutions). You can then scan the gel and relate the band intensity to the protein concentration (to get a calibration curve of the standard) and finally relate the intensity of the protein of interest to the standard bands. There is free software to analysis the band intensity in the internet.

you can use Qubit or an alternative - it is based on fluorometric quantification. And it is quite precise, similar to Nanodrop

As mentioned just above, we also measure protein concentration by SDS-PAGE along with a BSA or Lysozyme standard (depending on the size of the intended protein). If your protein levels are really low and not visible on commassie (what they should if you are going to do something else with your target except for measuring protein concentration), Western blot with a His-tagged protein of known concetration would also be an alternative.. However, I wonder why the colour is still there when you protein elutes from IMAC and its not associated with the dye...

methanol/chloroform precipitation of the protein is worth trying. Alternatively, find a nearby lab using the Invitrogen EZQ quantification kit. It is blot-based with an extensive wash of the membrane that usually gets rid of detergents and small molecules.

I hope this helps

Hi, I think you should make a dialysis of your sample prior to quantification , logically with a membrane cut off below the weight of your protein. Another procedure is to make a sample blank with buffer or water plus sample. the absorbance of this blank must be subtracted of the value of your blank.

Best regard.

It is possible the micro-biuret method will be helpful. Ruth F. Itzhaki, D.M. Gill A micro-biuret method for estimating proteins // Analytical Biochemistry — Volume 9 — Issue 4 — 1964 — 401—410. Absorbance at 310 - 340 nm is measured. Mayby contaminant does not absorb at this wavelenght range.

Best wishes

The ideal would be to use a refractometer. No need of any wavelenght based measurement. It is particularly useful when you have a very low absorbance due to the lack of tryptophans and tyrosines. You can make a standard curve with BSA and then measure different dilutions of your protein. However, I don't know if you can get one available...

Try a modified Lowry assay. http://www.mendeley.com/research/assay-proteins-presence-interfering-materials/

You can use Pierce Slide A lyser Dialyse cassettes to remove unwanted dyes.

Good Luck

Just use nanodrop!!

Nanodrop is based on absorbance, just like a spectrophotometer, so that is NOT good advice.

Different proteins stain differently with Coomassie, so comparing with a different protein standard will not give you absolute values.

Dialysing to try to remove the contaminant is a good idea.

Also, if the protein is pure apart from the contaminant, then you can get amino acid analysis done on a sample, and use this as your absolute standard against which to calibrate Coomassie stained gels, Bradford assay, Lowry assay or other non absolute-measures.

What does 'within the size range of the proteins I am isolating mean?' Any more specific data on that.

I think that it is a pain and a blessing you know your protein is there just by looking at the sample and the and just run a standard SDS page gel with a BSA standard when you fix your gel by boiling you can remove all dye even running dye before you stain with Denville blue, you can quantify if you scan the gel on LiCOR or regular scanner the quantify with Odyssey software or ImageJ.

We used fluorometric assay to quatify bile proteins...it was helpful with biliary pigments in bile when other methods failed... Plesae refer to our paper:

Fluorometric assay of protein in native human bile.

P R Harvey, G A Upadhya, J L Toth, S M Strasberg

Clinica chimica acta; international journal of clinical chemistry. 09/1989; 183(2):147-54.

It is very hard to suggest, since information on the protein is no sufficient. Only thing I suggest is to purify your protein further by the standard method of protein purification such as anionic ion exchange chromatography, cationic ion exchange chromatography, and molecular exclusion chromatogaraphy. After these procedure, you can test the purity by SDS/PAGE. If the purified protein is still colored, your protein must have a colored cofactor.

My preferred approach would be to ignore the Bradford for now and try to increase the purity of your sample If you are isolating from sea urchins (presumably via the ability of the protein to bind to nickel without a tag?) I doubt that it is very clean after just one step. An ion exchange step can work wonders, either for resolving your protein from a pigment or from another protein, and can be done with a syringe or gravity flow on a small volume if you don't have an FPLC available. Have you actually tried SEC? Even if they run the same by SDS-PAGE, SEC can still work as it separates on overall size and shape. Also, make sure you wash the nickel beads extensively before you elute your protein, maybe even increasing the stringency (amount of imidazole) .

dialysis is the best method to remove the color contaminants from the protein sample. after dialysis run proten through SDS PAGE and used software to quantify the protein. you should do ion exchange chromatography.

I agree with those who say that additional purification could be helpful. Using different buffer conditions with SEC might get you the separation that you need. Also, using a gel-based approach, 2-D gel electrophoresis might provide the separation that you need and then quanitify the band intensity with standards, as Uwe Bornscheuer suggested.

Fai

A simple method may be organic nitrogen and transform in protein x6,22

For me, dialysis or centrifugal ultrafiltration never worked to remove the color from my protein preparation. On the other hand, ionex was pretty useful for that.

UF/DF is much faster technique instead of dialysis.

PolyAmide CC6 resin is good to remove color impurities.

Total Keldal nitrogen x6,22 is ok but too much labour work.

If you have your standard sample you can run HPLC-analysis.

Best way to go is to try desalting your protein sample to remove the coloured impurity, using G10 - G25 Sephadex works wonders sometimes particularly with highly concentrated samples and a long thin column. I have used this routinely to remove unconjugated probes from proteins after labelling and it works fine. Now you can spectrophotometrically determine protein concentration. Similarly, dialysis may work. Otherwise, try HPLC if you can establish that your impurity has a different absorbance maximum somewhere in the spectrum other than your protein (ie 214 nm) although a HPLC equipped with a PDA makes this easier (190-900 nm) or find a fluorimeter and play around with different parts of the spectrum and then jump on the HPLC with wavelength discrimination between your protein and the calorimetric contaminant. Gel densitometry is also handy using BSA as a standard curve analyte. However, what if your protein is bound to a cognate/coloured co-factor? If it so tightly associated it may be a co-factor or metabolite of relevance to your protein of interest.

You might try one of the versions of the Lowry assay in which the protein is first precipitated with trichloroacetic acid (sometimes with sodium deoxycholate) to get rid of interfering substances, which are left behind in the supernatant.

If you go for dialysis or some other chromatographic techniques for discoloration, you may loss some of the quantity of your protein. Better to go directly for TCA Precipitation method. then check the protein quantity of total protein quantification method. It may give you better results.

@all: the problem is that it is associated with proteins.

Doesn't TCA precipitation denature proteins?

Thanks for all the ideas and suggestions. Sorry for my delayed response but its been a busy couple of days.

I think I should have provided some additional clarification.

I am working with a family of closely related proteins (the isoforms vary by a few amino acids in specific regions, and have different post-translational modifications resulting in different MW and pI). My end goal is to load a consistent amount of protein on an IEF strip.

The predicted size of most of these monomers is ~38-50kD, but we get multimers and/or modified proteins as high as >200kD (the proteins are predicted to be disordered,

As to fluorometric assays, does anyone have any suggestions for protocols or reagents or kits and what not to look at? Does one need a specialized reader like the Qubit, or will any fluorescent plate reader work?

Gundumi, I can't seem to find that paper online; would you be willing to send me a copy and/or supply a brief summary of the method?

Have you heard of Direct Detect: http://www.millipore.com/life_sciences/flx4/direct_detect

This is using IR not UV/colour and thus colour is not an issue.

thanks for the tip.

I actually came up with such an obvious approach that I could kick myself: The pigment is very dilute in my initial sample, so much so that it seems to allow me to (ballpark) determine the protein concentration. I can then measure the protein content of the concentrater flow through, and subtract that from the initial protein concentration to get a ballpark concentration of my concentrated sample (recognizing that there is some protein loss, but I only need a ballpark value).

Hi Lauren,

Try this, use a C18-column (SesPak has several size according to your needs) to adsorb your pigment by reverse chromatography. I guess that your proteins are more soluble than this pigment, so add your sample to the column, and then add water or a saline solution. The pigment must be adsorbed by the column and the proteins should elute or could elute by adding methanol 10% or 30%. With methanol 50% the pigment beging to elute too.

Probably your protein sample will be diluted with this methodology, but you can concentrate adding a precipitating agent to resuspended laterly by another solvent volumn.

good luck,

You can remove the colored compound from your sample using gel filtration. I guess the molecular weight of that compound will be different from your protein molecular weight. it is one of the method to remove salt from protein sample after precipitation with salt. Hope I am helpful. thanks. Simon

You can use the standard amino acid analysis, and then calculate the protein content by the sum of amino acids with small correction for destruction of aa.

Lauren,

its depending on the amount you have and you might loose for the measurement, plus the money you might spent for the analyses.

- Protein size is needed (AA or MW).

- Folding if possible might be of interest going in more detail (is however not be needed right now).

- Is only the protein quantification of your interest ?

Please send me some more details by eMail and I will reply as soon as possible.

Regards!

Lauren,

the purple color could be from the IMAC column. They tend to leach metal ions at higher imidazole concentrations, especially if you use reducing agents. Try adding EDTA (to leach it off the tags) and one run on a desalting column to get rid of it.

If your dye is a protein the color should be gone in 6M Guanidinium-HCL or 7M Urea. That is what you observed, yes?

The most precise A280 readings are done in 6M Guanidinium-HCL when the aromatic residues are "free" in solution. If this also removes the dye, just measure all your samples in Guanidinium-HCL. A nanodrop reader reduces sample loss to a minimum.

In your question it is not properly mentioned whether the protein itself is a colored protein or there is actually an associated colored co-isolate or isolates. Without knowing this, the rest cannot be discussed.

Hi,

The first thing you should do is to identify at which wavelength the colored pigment in your protein sample absorb. After that you can easily pick one method for protein quantification. hope my answer is helpful. thanks. Simon

How accurate must your measure be? We will, with difficult samples, simple spot a small amount onto a PVDF or nitrocellulose membrane or cellulose acetate membrane, and spot a few strandard solutions on (BSA). Then, wash the filter paper well, to remove coloured and other contaminants, and finally, stain briefly with Ponceau Red. It is good enough for many purposes.

Hi, I regularly measure proteins in wine so I have interference of phenolics and other coloured substances. A method that I find excellent in this conditions is the EZQ quantitation kit from Invitrogen. It is a fluorescent based method and it does not seem to be bothered by colour interference, at least in my case. It has also the advantage to be a high throughput method as it is done in a 96 wells plate.

Cheers

I'd be a bit wary of an SDS-PAGE determination of the protein concentration, depending on the revelation method you use. Coomassie blue fixes proteins differently, and you'd really need to have a standard of your own protein to be sure of the concentration. I don't now how silver staining works in that aspect. (wikipedia: http://en.wikipedia.org/wiki/Talk%3ACoomassie_Brilliant_Blue)

Very good answers above. I suppose the easiest way indeed is to just run the protein on SDS-PAGE and include different quantities of control protein (like BSA) and quantify from the gel.

In addition to everything mentioned above you might try refractometry or mass spec quantification if you have the instrumentation. It's quite accurate in concentration determination and independent of absorption etc.

You can dialyse your protein in a buffer which contain active charcoal...i think your dye will be adsobed by active charcoal..

In these cases I have simply diluted my protein sample 1:10 or 1:100 and then used a high sensitivity detection bca kit to detect protein specifically. Especially if you are using a highly concentrated protein.

If the colored contamination does not absorb at 280 nm, simply measure the absorbance of your proteins soln at 280 nm, then calculate its concentration taking into account the molar extiction coefficient of your protein (depends on the exact amino acid sequence). For help with calculations check http://expasy.org

Good idea, Nausheen! Just enter your protein sequence at http://web.expasy.org/protparam/ and it'll do the calculation for you.

the answer by Uwe Bornscheuer is good

so is Nausheen but if you are going to use A280 and a calculated E you must denature the protein in 20mM sodium phosphate pH 6.5 with 6M guanidinium HCl

Dear Lauren: The most reliable method to quantify protein is amino acid analysis after hydrolysis acid and hydrolysis basic (for tryptophan). In the lacking the equpment (amino acid analyser) soma suggestion can be done: chromatography your sample in adsorptive material ( activated charcoal, for example) it is necessary that protein are not linke in stationary phase. Other solutions is to dilute the sample to diminish the color since that protein presents a relevant quantitie and to use a very sensitive spectrophotometric method. One question this contaminant colored absorbs in UV (205 - 225 nm) region? If not, this range can be used. Good look for you.

Do TCA precipitation, if you can, and then redissolve and quantify with standard type color assays. SDS is good too as noted in other answers.

You can try to polish your sample with Vivaspin 1kDa ultrafiltration device (http://www.vivaproducts.com/centrifugal-filtration/vivaspin-20.html). I suppose, that contaminant has much lower MW than your desired protein.

Does your protein have any activity? If so use it's own functional assay to characterize it. Does it have a tag or antibodies to it or its tag? Probe for that tag via Western?

The best way to do that is to remove the colored contaminant by gel filtration chromatography before spectrophotometry quantification. Thanks. Simon

When we have protein in solution with color, we try to do a wash using special tubes with filter in the middle. We put the protein in this tubes and centrifuged. And we start adding PBS buffer as the colored liquid goes down with the centrifugal force. We do this until the solution of the protein turned colorless. It is important to choose a tube with a suitable pore filter, because you want to lose the solution, but keep the protein.