Protein concentration is measured by exploiting a physical or chemical property of proteins and comparing the resulting signal to a set of known standards. The most common approaches fall into two broad camps: ultraviolet absorbance, which reads how much light a protein sample absorbs at 280 nm, and colorimetric assays like the Bradford, BCA, and Lowry methods, which generate a colored product proportional to the amount of protein present. Choosing the right method depends on your sample type, the buffer it sits in, and how much protein you expect to find.
Measuring Protein by UV Absorbance at 280 nm
The simplest way to estimate protein concentration is to shine ultraviolet light through your sample and measure how much gets absorbed at a wavelength of 280 nm. Proteins absorb at this wavelength because of specific amino acids in their chains, primarily tryptophan, tyrosine, and to a lesser degree the disulfide bonds formed by cysteine residues.1PubMed Central. How to measure and predict the molar absorption coefficient of a protein You place your sample in a spectrophotometer, record the absorbance reading, and then convert that reading to a concentration using either a standard curve or a published absorptivity value for the specific protein you are working with.2PubMed. Spectrophotometric determination of protein concentration
The appeal of this method is speed. There is no reagent to mix, no incubation step, and no color development to wait for. You pipette a drop of sample, take a reading, and you have a number in seconds. It is also nondestructive: you can recover your sample afterward if you need it for downstream experiments. The downside is that different proteins absorb 280 nm light to very different degrees, depending on how many tryptophan and tyrosine residues they contain. A protein with very few of those amino acids will give a weak signal that underestimates the true concentration, while a protein rich in them will read higher than expected. If you are measuring a pure protein whose absorptivity is published, this is not a problem. If you are measuring a crude mixture of many proteins, the reading becomes an approximation.
Nucleic acid contamination is the classic pitfall here. DNA and RNA absorb strongly at 260 nm and still absorb at 280 nm, so even a small amount of nucleic acid in your sample will inflate the protein reading. One useful quality check is the 260/280 absorbance ratio. For a reasonably pure protein sample, a ratio around 0.6 is expected; higher values suggest nucleic acid contamination is skewing the measurement.3DeNovix. Nucleic Acid Quantification, Protein Quantification, UV-Vis – Section: Protein 260/280 Purity Ratio
The Bradford Assay
The Bradford assay is probably the most widely used colorimetric protein method in research labs, and for good reason: it is fast, inexpensive, and tolerates many common buffer components. The assay uses a dye called Coomassie Brilliant Blue G-250. In its free, unbound state the dye absorbs light at 465 nm and appears reddish-brown. When the dye binds to protein, its absorption maximum shifts to 595 nm, turning the solution blue.4PubMed. Bradford Assay for Determining Protein Concentration The more protein in the sample, the more blue color develops, and you read the intensity at 595 nm to quantify it.5Analytical Biochemistry. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding
The dye preferentially interacts with arginine residues, though histidine, lysine, and the aromatic amino acids contribute a small amount to the signal. The binding involves a mix of electrostatic interactions, van der Waals forces, and hydrophobic contacts.6Analytical Biochemistry. Mechanism of dye response and interference in the Bradford protein assay This means the Bradford assay responds differently to different proteins: a protein with abundant arginine will give a stronger color per microgram than one with very little arginine. The standard curve is typically made with bovine serum albumin (BSA), so the concentration you calculate is really “BSA equivalents” unless you build a curve with your specific protein of interest.
One practical advantage of the Bradford assay is that the color develops within about two minutes and remains stable for a reasonable window, making it easy to read plates or cuvettes without rushing. It also tolerates moderate concentrations of many reducing agents and denaturants. Its Achilles heel is detergents: even the nonionic detergent Tween 80 can inflate Bradford readings by roughly two to three times the true value, and the problem is not just the detergent producing color on its own but also altering how the dye interacts with protein.7Analytical Biochemistry. Interference of the detergent Tween 80 in protein assays If your lysis buffer contains SDS or other ionic detergents, the Bradford assay will usually give unreliable results.
The BCA Assay
The bicinchoninic acid (BCA) assay works by a two-step reaction. First, protein reduces copper ions from Cu²⁺ to Cu⁺ in an alkaline environment, much like the older Lowry and Biuret reactions. Then, bicinchoninic acid chelates the resulting Cu⁺ ions to form a purple-colored complex that absorbs at 562 nm. The intensity of the purple color is proportional to protein concentration.8PubMed Central. Competitive Binding to Cuprous Ions of Protein and BCA in the Bicinchoninic Acid Protein Assay
Compared to the Bradford, the BCA assay responds more uniformly across different proteins, partly because the copper reduction step is driven primarily by the peptide backbone itself rather than specific side chains. It also handles most detergents well, which makes it a better choice when you are working with membrane proteins solubilized in detergent-containing buffers. The tradeoff is that the BCA assay is sensitive to reducing agents. Thiol-containing reagents like dithiothreitol (DTT) and beta-mercaptoethanol reduce Cu²⁺ to Cu⁺ on their own, producing color that has nothing to do with protein and artificially inflating your reading.9Analytical Biochemistry. Protein determination using bicinchoninic acid in the presence of sulfhydryl reagents High concentrations of urea, another common component of lysis buffers for two-dimensional gel electrophoresis, also interfere with the BCA assay.10PubMed. Protein quantification and its tolerance for different interfering reagents using the BCA-method with regard to 2D SDS PAGE
The BCA assay requires a longer incubation than the Bradford, typically 30 minutes at 37 °C for the standard protocol, and the color continues to develop slowly over time. Reading all samples at the same time after a consistent incubation period matters more here than it does for the Bradford, where the color is relatively stable.
The Lowry and Biuret Methods
The Lowry method predates both the Bradford and BCA assays and was for decades the default way to measure protein in a research lab. It combines the Biuret copper-binding reaction with the Folin-Ciocalteu reagent, which is reduced by tyrosine and tryptophan residues as well as by the copper-peptide complex, producing a blue color. The chemistry involves transfer of electrons to components of the Folin reagent, and the contribution from aromatic side chains makes the Lowry assay somewhat protein-dependent in its response.11Analytical Biochemistry. On the chemical basis of the Lowry protein determination It is also notorious for being sensitive to a long list of interfering substances, including detergents, sugars, lipids, and buffer salts. For these reasons it has largely been displaced by the Bradford and BCA assays in everyday use, though it still appears in older protocols and specific applications.
The Biuret method is the simplest of the copper-based assays. It measures the purple complex formed when copper ions bind to peptide bonds in alkaline solution, without the Folin-reagent amplification step. That simplicity makes it the least sensitive of the group: it works best when you have relatively concentrated protein, roughly 0.075 mg or more per assay. On the other hand, it is highly tolerant of non-protein components and handles samples containing DNA at concentrations up to about 0.7 mg/mL without significant interference.12Analytical Biochemistry. A micro-biuret method for estimating proteins A systematic comparison of spectrophotometric methods found that the Biuret method offered the widest working range for purified protein and complex mixtures, spanning roughly 0.4 to 100 mg/mL.13PubMed. Systematic comparisons of various spectrophotometric and colorimetric methods to measure concentrations of protein, peptide and amino acid If you have plenty of protein and need a quick, interference-resistant measurement, the Biuret method is underrated.
How to Choose Between These Assays
The decision tree usually starts with your buffer. If your sample contains detergent, the Bradford assay is risky; go with BCA. If your sample contains reducing agents or high urea, BCA is risky; go with Bradford. If your sample contains both detergents and reducing agents at once, you are in a bind and may need to remove those components before measuring. One practical workaround for stubborn buffers is to precipitate the protein out of solution first, wash the pellet, and then redissolve it in a compatible buffer before running the assay. Trichloroacetic acid precipitation followed by a Lowry measurement has been shown to give concentrations nearly identical to amino acid analysis, considered the gold standard, even for dilute membrane proteins in detergent-heavy solutions.14Analytical Biochemistry. Efficient precipitation and accurate quantitation of detergent-solubilized membrane proteins
The second consideration is sensitivity. If you have limited material, the Bradford and BCA assays work well in the low-microgram range. The Biuret method requires at least an order of magnitude more protein. UV absorbance at 280 nm sits somewhere in between and depends heavily on the instrument and pathlength. Microvolume spectrophotometers that use surface tension to hold tiny droplets between two measurement surfaces can quantify protein in as little as half a microliter, greatly expanding the practical dynamic range of absorbance-based measurements.15PubMed Central. Microvolume protein concentration determination using the NanoDrop 2000c spectrophotometer
The third consideration is throughput. If you are processing hundreds of samples, the Bradford and BCA assays adapt easily to 96-well microplate formats. UV absorbance at 280 nm is fast per sample but reads one cuvette at a time on most instruments, unless you have a plate reader with UV capability.
Fluorometric Approaches
Fluorescence-based protein assays are less common in routine work but can be useful when you need higher sensitivity or when colorimetric assays fail because of buffer interference. These methods rely on a fluorescent dye that changes its emission properties upon binding protein. One example is the BisANS dye, which showed a detection limit of about 0.65 micrograms per milliliter in a simple aqueous medium and actually performed slightly better in a lysis buffer containing chelators, detergent, and protease inhibitors, where the detection limit dropped to around 0.28 micrograms per milliliter despite an overall decrease in fluorescence intensity.16PLoS ONE. Application of BisANS fluorescent dye for developing a novel protein assay That combination of high sensitivity and compatibility with common lysis buffers addresses a gap that colorimetric assays struggle with.
The practical barrier is instrumentation: you need a fluorescence plate reader or spectrofluorometer, which not every lab has. For labs that do, fluorometric assays can be a good alternative when sample volume is limited or when classic assays give inconsistent results due to interfering components.
Why Your Standard Protein Matters
Every colorimetric or dye-binding assay produces a number that is only as meaningful as the standard curve it was calibrated against. Because each assay responds differently to different proteins, measuring your unknown against BSA gives you “BSA-equivalent” concentration. For many purposes this is perfectly adequate, especially if you are loading equal amounts of protein across gel lanes or normalizing enzyme activity. But if you need the actual mass of a specific protein in solution, BSA may not be the right standard.
This problem becomes especially visible with unusual protein types. Mucins, for example, are heavily glycosylated proteins that respond poorly to standard BSA-calibrated Bradford assays. Using a mucin-based standard instead, such as bovine submaxillary mucin, gave nearly twice the measured concentration compared to the same samples read off a BSA standard curve.17PubMed Central. Simple and accessible methods for quantifying isolated mucins for further evaluation If your protein of interest has a structure substantially different from BSA, whether because of heavy glycosylation, unusual amino acid composition, or membrane-spanning domains, consider building your standard curve with a protein that more closely resembles what you are measuring.
Removing Interfering Substances Before Measurement
When your buffer is incompatible with every available assay, precipitation is the standard rescue strategy. You crash the protein out of solution using an organic solvent or acid, spin it down, wash away the offending reagents, and resuspend the pellet in something compatible. The most common precipitation agents are acetone, trichloroacetic acid (TCA), and methanol-chloroform mixtures.
Recovery is not guaranteed, though, and some protocols lose a meaningful fraction of your protein. TCA-acetone precipitation, for instance, can be hard to redissolve and has been reported to recover only about 78% of the starting material from cell homogenates, compared to roughly full recovery with optimized acetone precipitation protocols.18Electronic Journal of Biotechnology. Comparison of protein precipitation methods for sample preparation prior to proteomic analysis of Chinese hamster ovary cell homogenates Optimizing the salt concentration and incubation temperature during acetone precipitation can push recovery to about 98% in as little as two minutes, even for dilute protein samples starting as low as 1 microgram per milliliter.19PubMed. Rapid and Quantitative Protein Precipitation for Proteome Analysis by Mass Spectrometry Ammonium sulfate fractionation is another option and has the added benefit of efficiently depleting albumin, which can dominate a plasma or serum sample and mask the proteins you actually want to measure.20PubMed. Comparison of protein precipitation methods for sample preparation prior to proteomic analysis
Low-molecular-weight proteins and peptides (below about 15 kDa) deserve caution: several precipitation methods have been shown to recover them less efficiently than larger proteins, which can bias your results if small proteins make up a significant portion of your sample.18Electronic Journal of Biotechnology. Comparison of protein precipitation methods for sample preparation prior to proteomic analysis of Chinese hamster ovary cell homogenates
Reference and High-Accuracy Methods
When the question is not just “roughly how much protein is here” but “what is the exact protein content of this sample,” the standard spectrophotometric assays are not accurate enough. Their protein-to-protein variability and susceptibility to interference make them estimation tools, not precision instruments.
Total amino acid analysis (AAA) is considered the gold standard for protein quantification. The protein is completely hydrolyzed into individual amino acids, which are then separated and quantified. This gives you the actual mass of protein regardless of its amino acid composition or structure. A simplified version of this method, called aromatic amino acid analysis, skips the complete hydrolysis and instead measures just phenylalanine and tyrosine after a shorter separation step, achieving a recovery of about 97% and a detection limit in the single-digit milligrams-per-liter range.21PubMed Central. Determination of the protein content of complex samples by aromatic amino acid analysis, liquid chromatography-UV absorbance, and colorimetry These reference methods are used to validate other assays and to assign certified concentrations to protein standards.
Mass spectrometry offers another route to precise protein quantification, particularly when you need to measure one specific protein in a complex mixture. The general approach is to digest the protein of interest into peptide fragments with an enzyme like trypsin, then use synthetic isotope-labeled versions of those same peptides as internal standards. The mass spectrometer can distinguish the labeled standard from the natural peptide by their slight mass difference, and the ratio between them gives a precise concentration.22PubMed Central. Mass spectrometry based targeted protein quantification: methods and applications One critical detail: the enzymatic digestion step must be driven to completion, or the amount of released peptide will underestimate the true protein concentration. Work with recombinant growth hormone showed that an optimized digestion protocol running just 30 minutes, compared to the 18-plus hours some older protocols use, produced results that matched amino acid analysis within measurement uncertainty.23PubMed. Protein quantification by isotope dilution mass spectrometry of proteolytic fragments: cleavage rate and accuracy These mass-spectrometry-based methods are overkill for routine lab work but indispensable in clinical biomarker validation, pharmaceutical quality control, and metrology.
Common Mistakes That Quietly Ruin Results
Plenty of labs run protein assays daily and still make errors that inflate or deflate their numbers without anyone noticing. A few of the most frequent problems are worth calling out because they are easy to avoid once you know about them.
Reading the standard curve outside its linear range is the first. Every colorimetric assay has a range over which color intensity increases proportionally with protein concentration. Above that range, the curve flattens out and differences in protein amount produce barely any difference in absorbance. If your unknown sample gives an absorbance reading near or above the highest standard, the interpolated concentration will be an underestimate. Dilute and re-read rather than trusting the math.
Using water as the blank instead of your actual buffer is the second. Many buffer components absorb light or react with assay reagents. If you zero the instrument against water but your sample contains buffer salts, detergent, or carrier molecules, the blank is wrong and every reading shifts. Always blank with the same buffer your protein is dissolved in, minus the protein.
Ignoring the timing of color development is the third, particularly for the BCA assay. Because the copper-reduction reaction continues slowly over time, two samples read five minutes apart will give different values even if they contain the same amount of protein. Either use a plate reader to measure everything simultaneously, or be meticulous about staggering your additions so that each tube gets the same incubation time.
Finally, assuming that a detergent or reducing agent “probably won’t matter” at low concentrations is a gamble that frequently goes wrong. Even Tween 80 at concentrations routinely used for stimulating protein secretion can inflate Bradford readings by more than three-fold and alter the chromatographic behavior of the protein.7Analytical Biochemistry. Interference of the detergent Tween 80 in protein assays Running a control with buffer plus detergent, no protein, is the simplest way to catch these problems before they corrupt an entire experiment’s worth of data.