Title: Determination of Protein Concentration 

Techniques to Master:

  1. Using a spectrophotometer to measure absorbance
  2. Prepare dilution series for a standard curve
  3. Perform and interpret a protein concentration assay
  4. Analysis of data that includes a graph, linear regression and calculations related to dilutions. 

Learning Objectives:

  1. Identify a variety of factors that must be considered when choosing an appropriate assay for determination of protein concentration.
  2. Interpret absorbance measurements in order to determine concentration of a protein solution.
  3. Explain how a standard curve is used to determine concentration of a protein solution.

Background: 

A. Introduction to Protein Assays. Many biochemical experiments require accurate determination of protein concentration, for example for determining purification yields or to prepare a sample of appropriate concentration for analysis.  Protein quantitation usually relies on absorbance measurements; these measurements can be grouped into two broad categories: intrinsic absorbance and colorimetric/dye-binding (Table 1).

Intrinsic absorbance measures absorbance of a protein without addition of any other reagents.  Most proteins absorb UV light with maximum absorbance at 280 nm due to absorbance of aromatic amino acid side chains  (mainly tryptophan) and disulfide bonds; hence this assay is sensitive to the amino acid composition of the proteins.1,2 

Colorimetric and dye binding assays are based on the change in absorbance of metal ions or organic dyes upon binding to proteins.  These methods are less sensitive to amino acid composition than intrinsic absorbance. When choosing an appropriate method, one must carefully consider potential interfering compounds.

                      

A number of factors must be considered when choosing an appropriate assay to determine protein concentration. These include: sensitivity and working/linear range; interfering substances; effect on the protein (e.g. A280 does not destroy protein, but other methods do); time and cost constraints; and availability of protein standard (solution of known concentration).  Table 1 compares several methods; additional information can be found in the original references or in the product information for commercially available reagents.  

Table 1: Summary of methods for protein determination

Category

Method

Range

Based on

Interference

Intrinsic Absorbance

280nm [1,2]

ε =  0 – 2 (mg/mL-cm)-1

0.050-2.0 mg/mL

Absorbance of Trp, Tyr, and disulfide bonds

Detergents

Nucleic acids

Colorimetric

     ∙Chemical

Biuret [3]

1.0-6.0 mg/ml

Cu+2 reduced to Cu+1 by peptide bond nitrogens

Ammonium salts

Lowry [4]

0.050-1.5 mg/ml

Biuret Cu+1-catalyzed reduction of Mb and W salts

Amino acid

Nucleic acids

Detergents

Ammonium sulfate

Lipids

Bicinchoninic acid (BCA) [5]

0.05-1.5 mg/ml

BCA binding to Cu+1 in peptide bonds

Reducing sugars

Copper chelating agents

Colorimetric

  ∙Dye

   binding

Bradford [6]

0.010-1.5 mg/ml

Coomassie Blue G-250 binding to proteins (preferentially to aromatic and Arg)

Detergents

NanoOrange® fluorescence based [7]

0.01-10.0 μg/ml

Binding to hydrophobic regions of proteins or detergent-coated proteins

Strong bases

Reducing agents (DTT, 2-ME)

 EDTA

B. Beer-Lambert’s Law: When we use spectrophotometry to determine concentration, we are relying on a mathematical relationship between the amount of light absorbed by the substance and the concentration of the substance (the Beer-Lambert’s law).  

 

A = ecl

 

Where A is absorbance, c is the concentration of the solution, l is the pathlength (usually 1 cm), and e is the extinction coefficient or absorptivity. Absorptivity is a number specific to the wavelength at which absorbance is measured and the substance in solution that absorbs light. Absorptivity also known as the extinction coefficient has unique units.  Because absorbance, A, has no units, the units on the right hand of the Beer-Lambert equation must cancel.  Thus, the units of e must cancel those of c and l, and consequently, become the inverse:  c -1 · l -1  (e.g.  M-1· cm-1  or  (mg/mL)-1· cm-1).  Note two confusing pairs of terms: absorptivity should not to be confused with absorbance, and pathlength (measured in cm) should not be confused with wavelength (measured in nm).

 

If an absolutely pure protein is being used, one could simply determine the concentration using the Beer-Lambert Law and the extinction coefficient for the protein (if it’s known or can be calculated).

For example, if a solution of unknown concentration has intrinsic absorbance at 280 nm of  0.600, we cannot conclude anything about the concentration without additional information.  If we also know that the same protein in solution of 0.50 mg/mL at 280 nm has absorbance of 0.500, we can calculate e = A/cl = 0.500/ (0.50mg/ml*1 cm) = 1 (mg/ml)-1cm-1. The value of e allows us to calculate concentration of the unknown to be 0.60 mg/mL.   

Note that the use of e in calculating concentration relies on the assumption of the linear relationship between absorbance and concentration.  The linear relationship defined by the Beer-Lambert’s law, however, has limits, which can be revealed using standard curve and graphical analysis.

C. Use of a standard curve 

A standard curve is generated by measuring the absorbance of a standard protein at a variety of known concentrations.  The data is plotted with concentration (independent variable) on the x-axis, and absorbance (dependent variable) on the y-axis.  This graph can be used to estimate concentration of the unknown as shown in figure 1.  Note that part of the data appears to follow linear trend. If that data is fitted to a straight line, the equation of the line can be used to calculate the concentration of the unknown sample.  Only the linear portion of the standard curve is fitted.

For example, if the absorbance of the unknown is 0.750, the estimate from figure 1, suggests concentration of the unknown around 0.65 mg/mL.  The equation in figure 2 can be used to calculate concentration:  x = (0.750 + 0.0504) / 1.1535 = 0.694 mg/mL 

If the standard protein behaves similarly to the unknown protein(s), and both are handled in the assay the same way, the inferred concentration should be accurate.  When designing an experiment, you should carefully consider the choice of standard protein and the choice of the assay.  The standard protein of choice is often bovine serum albumin (BSA).  It is small, soluble, well-behaved and relatively inexpensive. Other issues to consider when choosing an assay include: time required, sensitivity, difficulty, cost, and unfavorable interactions with components in the solution (detergents, reducing agents, etc.).  

Some assays are straightforward, such as the intrinsic ultraviolet absorbance assay (UV Assay), which simply measures the absorbance of the aromatic amino acids in the protein at 280 nm.  This method requires standard protein (solution of known concentration) to be the same protein as the unknown.  When such a standard is not available, two assays that are commonly used are the Bradford Assay and the bicinchoninic acid (BCA) Assay.  Unlike the UV assay, these assays rely on reacting the protein with a dye (or other compound) to produce a color change.  As a result, the protein is irreversibly denatured and thus only a small aliquot of the sample should be analyzed with these colorimetric assays. The amount of color change is dependent upon the concentration of the protein in the sample.

D. Bradford Assay. The Bradford Assay is a simple one-step method that can be performed as a standard assay (3.1 mL assay volume, requires 0.1 mL of protein, and a working range of approximately 0.1-1.4 mg/mL), in 96-well plates (same working range, but only 5 uL of protein is required) or as a micro assay (2 mL assay volume, requires 1 mL of dilute protein, 1-10 ug/mL working range).  In this assay, Coomassie Brilliant Blue G dye is mixed with the protein under acidic conditions.  The protein interacts with the dye via ionic and hydrophobic interactions of amino acid side chains. The interaction stabilizes anionic form of the dye, causing a color change from red to blue and a corresponding shift in the λ max from 465 nm (free dye) to 595 nm (dye-protein complex).  The reaction can occur in as little as 5 minutes.  

Purpose: We will determine protein concentration at the end of purification to calculate the protein yield (e.g. mg of protein per ml of culture or per ml of crude lysate).  Tracking the protein yield is helpful for evaluation of success of purification procedures (e.g. when optimizing purification procedure). Knowing the final protein concentration will also help you set up enzymatic activity assays: you will want to know how much protein is added to each assay.

 If the activity of the enzyme can be measured (in addition to measuring protein concentration), specific activity of the enzyme can be calculated in terms of activity units per mg of total protein.  Specific activity is another measure of protein purification success.

 

Safety considerations: Bradford reagent contains phosphoric acid and methanol. Safety Data Sheet includes the following warnings: It may be corrosive to metals. Causes severe skin burns and eye damage. Causes damage to organs.

Wear personal protective equipment.

Experimental Design Considerations: When calibrating the spectrophotometer, students often use water as a “blank.” If you want to measure absorbance contributed by the protein in solution, is water the most appropriate sample for calibration? Could there be any other molecules in your assay mix that absorb some light at the wavelength you are using?  The “blank” should contain everything except protein (i.e. water, buffer, reaction mixtures, etc.).  It is important to perform the assay on the standards, as well as the samples, at the same time to account for any issues such as temperature or instrument variations. The Bradford reagent is stored at  2–8 °C. If a student in a hurry does NOT wait to warm the reagent to room temperature before use, how would this change in the procedure affect results?

Consider other possible sources of error or uncertainty (assume you are careful with pipetting and can ignore pipetting error). Is your standard protein the same as your unknown? If not, would the differences in amino acid composition affect your results? Are there any reagents in your buffer that could interfere with the assay (e.g. detergent; see this complete list of interfering substances [8])? Do your standard concentrations cover the expected linear range of the assay? Did you plan for more than one dilution of the unknown? Remember, you need at least one unknown measurement to be in the linear range of the standard curve.

 

Supplies:

Equipment

Test tubes

Glass pipettes and pipettors

Micropipettors and tips

Plastic cuvette (1 mL or 3 mL)

Spectrophotometer

Computer with Excel for data analysis

 

Reagents:

Bradford reagent (Sigma cat # B6916)   

Standard BSA solution (e.g., 2.00 mg/ml)

Unknown protein

Buffer (for dilutions and a blank; same as buffer used for unknown protein

solution)

 

Procedure:

Based on product information from Sigma-Aldrich [8].

A. Preparing BSA Standards. Using the buffer that your protein is stored in, prepare a series of 5-7 solutions of bovine serum albumin (BSA).  In a standard cuvette-based assay, standard solutions should range from 0.10 to 1.50 mg/mL.                    

B. Bradford Assay

Bradford Assay in 3.1 mL volume

 

Bradford 96-well Plate Assay (requires plate reader)

 

Clean-up:

Dispose of all solutions containing the Bradford reagent in a hazardous waste container.

        

Interpreting Results:   

Generating and using a standard curve in Excel. You will need the following data: calculated concentration of protein and measured absorbance for each sample.  

  1. Calculate concentration (in mg/mL) of protein for each standard sample using the stock concentration of your standard and volumes of reagents used to prepare the sample.  Note, since the volume of Bradford reagent added is the same for all samples, you only need to calculate concentration of diluted protein (in 0.100 mL volume), and not the concentration in the cuvette.
  2. Use Excel to graph absorbance versus mg/mL protein for standard tubes (make sure to use scatter plot!).
  3. Add a linear trendline, the equation of the line and R2 value for the graph (you will need to click “options”, then check a box next to “display equation” and “display R2”).
  4. To determine the concentration of protein (mg/mL) in your unknown, use the equation of the line from the graph, substituting the unknown absorbance for Y and solving for X.  Be sure to use the correct units and number of significant figures.
  5. If you diluted your protein prior to performing the assay, calculate the concentration before the dilution using C1V1 = C2V2 formula.

 

Considering the working range of the assay: Consider an unknown that has absorbance 1.500 and a standard curve in figure 2. Could you use the equation/standard curve in figure 2 to determine concentration of this unknown?  Notice that the absorbance value is higher than any value on the graph in figure 2.  If you use equation to calculate concentration, you are making the assumption that the linear trend can be extrapolated, but that assumption may not be correct, especially when absorbance values exceed 1.

When using a standard curve, one should consider the working range of the assay carefully.  We use the linear portion of the graph to determine concentration, but if you test a wide range of concentrations, you may find that the graph is actually a curve not a straight line.  At higher concentrations of protein, the assay may become saturated. Also, at low concentrations, you absorbance measurements may include large error (e.g. absorbance values below 0.05).  In figure 1, can you estimate the working (linear range) of the Bradford assay?  Linear fit of subset of data, reveals that concentrations up to around 1 mg/ml fit well to a straight line, but at higher concentrations, the slope of the line changes (Figure 3).  

You also should not force the entire data set to linear fit. For the sample with absorbance 1.500, you can estimate concentration from the graph in figure 1, but cannot use the equation in figure 2 (and either equation in figure 3). The best course of action is to make several dilutions of your unknown (hoping that at least one will fall on the linear range of the assay) and repeat the experiment if necessary.

References:

  1. Wetlaufer, D. (1962) Ultraviolet spectra of proteins and amino acids. Adv Protein Chem. 17, 303-90
  2. Pace, C., Vajdos, F., Fee, L., Grimsley, G., and Gray, T. (1995) How to measure and predict the molar absorption coefficient of a protein. Protein Science. 4, 2411–2423
  3. Gornall, A.G., Bardawill, C.J., David, M.M. (1949) Determination of serum proteins by means of the biuret reaction.  J Biol Chem. 177(2), 751-66
  4. Lowry, O., Rosenbrough, N., Farr, A., and Randall, R. (1951) Protein measurement with the Folin phenol reagent. J Biol Chem. 193, 265-275
  5. Smith PK, Krohn RI, Hermanson GT, Mallia AL, Gartner FH, Provenzano MD, Fujimoto EK, Goeke NM, and Klenk DC (1985) Measurement of protein using bicinchoninic acid. Anal Biochem. 150, 76–85
  6. Bradford MM (1976) A rapid and sensitive method for quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem. 72, 248–254
  7. Jones LJ, Haugland RP, and Singer VL (2003) Development and characterization of the NanoOrange protein quantitation assay: a fluorescence-based assay of proteins in solution. BioTechniques. 34(4), 850-4, 856, 858
  8. Bradford Reagent Technical Bulletin (2016) [online] http://www.sigmaaldrich.com/content/dam/sigma-aldrich/docs/Sigma/Bulletin/b6916bul.pdf (Accessed July 15, 2016)

Creative Commons LicenseThis work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.