Protein Oxidation (Carbonyl Groups)


Because proteins are major components of biological systems and because proteins play an important role in a variety of cellular functions such as signal transduction, mitosis, cellular transport systems, chaperone activity, etc., an age-related increase in oxidative damage to proteins could be important physiologically to an organism.  Stadtman’s laboratory was the first group to show that proteins were oxidatively damaged in vivo (Stadtman, 1990), and the physiological importance of oxidative damage to protein was recognized demonstrated over 30 years ago when Levine (1983) showed the oxidative modification of one of the histidine residues of glutamine synthetase to a carbonyl group inactivated the enzyme.  Currently, protein oxidation is considered to be a potentially important marker for oxidative stress and is considered an important factor in various diseases as well as in aging.  

Almost all amino acids in proteins are potential targets for oxidation by ROS (Huggins et al., 1993).  For example, phenylalanine and tryptophan are converted to hydroxyl derivatives, tyrosine is converted to nitrotyrosine, and histidine is converted to 2-oxohistidine by oxidation. Methionine is unique in that its oxidation to methionine sulfoxide is reversible and can be converted back to methionine by methionine sulfoxide reductase. Cysteine can be oxidized to a series of products, which also include reversibly oxidized forms (disulfide, S-thiolated, S-nitrosylated, and sulfenic acid) and irreversibly oxidized forms (sulfinic and sulfonic acids). However, the most widely studied product of protein oxidation is the formation of carbonyl derivatives to side chains of certain amino acids, e.g., lysine, arginine, praline, and cysteine (Amici et al., 1989).  Lipid peroxidation products and advanced glycation end products also can react with proteins forming carbonyl derivatives to the side chains of lysine and cysteine by the Michael reaction (Doorn and Petersen, 2002).

Stadtman’s laboratory developed a spectrophotometric based assay to detect protein carbonyls in proteins in biological samples (Levine et al., 1990).  The assay uses 2,4,-dinitrophenylhydrazine (DNPH), which reacts with carbonyl groups and has an absorbance maxima at 370nm and a molar extinction coefficient of 22000 M-1 cm-1.   The level of carbonyl groups in proteins is determined by the absorbance at 370nm associated with the acid precipitable material after reacting with DNPH.  Using this assay, Stadtman’s group were the first to show that protein oxidation increased with age (Oliver et al., 1987).   Over the past 30 years, the effects of aging and various pathological processes on the levels of carbonyl groups in protein extracts from tissues of laboratory rodents and human subjects have been studied using this assay (see Bokov et al., 2004 for a review). However, the major obstacle in the spectroscopic measurement of protein carbonyl content by the DNPH assay is the interference from the unreacted, free DNPH associated with the acid precipitates (Goto et al., 1999). Subsequently, assays using Western blots have been developed to identify proteins showing increased levels of carbonyl groups. Carbonyl groups are derivatized with DNPH or other groups and the derivatives are identified using antibodies directed against DNPH (Nakamura and Goto, 1996). However, the sensitivity of this assay is limited by the limited specificity of the available antibodies to DNPH-proteins.  

To address this issue,our laboratory developed an assay to quantify protein carbonyl groups in proteins by derivatization of carbonyl groups with a fluorescent probe, fluorescein-5-thiosemicarbazide (FTC) (Chaudhuri et al., 2006). The hydrazine group on FTC reacts with the carbonyl to form a Schiff base covalently linking the fluorescein moiety to the protein. The fluorescence associated with the acid-precipitable protein pellets is used to measure the level of protein carbonyls fluorometrically with an excitation wavelength of 492 nm and emission wavelength of 516 nm. We showed the reaction of FTC with protein carbonyls was specific for carbonyl groups. However, the inability to remove the free FTC from acid-insoluble material by washing resulted in a great deal of variation, and we found that it was virtually impossible to detect with any consistency an increase in protein carbonyl groups formed in vitro or in vivo after an oxidative stress.  Therefore, we resolved the proteins on SDS-polyacrylamide gel electrophoresis, which separated the free FTC from FTC covalently bound to the oxidized proteins.  Using gel electrophoresis, we are able to determine the carbonyl content (nmoles/mg of protein) of the proteins by measuring the fluorescence associated with the proteins on the gels.  As shown in the Figure below, the protein carbonyl levels increased 2-fold in the liver of old mice compared to young mice.  The major advantage of the FTC labeling assay is that it allows one to directly measure changes in the levels of carbonyl groups in specific proteins using 2-D gel electrophoresis as shown in the Figure. The current methodology used to measure changes in carbonyl groups in specific proteins employs antibodies to the DNP-protein adducts (Nakamura and Goto, 1996). However, an accurate quantification of protein levels by western blots is limited by several factors including the specificity of the antibodies for the DNP-protein adducts and non-quantitative transfer of proteins from SDS-polyacrylamide gel into the PVDF membrane. In contrast, the FTC assay allows one to directly determine the levels of carbonyl groups in specific proteins by measuring the fluorescence associated with each spot. We have shown that FTC-bound proteins are neither lost during 2D gel electrophoresis nor show a differential migration on 2-D gels compared to the proteins free of FTC. An additional advantage of the FTC labeling assay is that one can measure on the same gel both the incorporation of FTC into each protein and the level of protein in the spot using Sypro Ruby. Thus, by measuring the ratio of the fluorescence intensities of FTC and Sypro Ruby, one can calculate the pmoles of carbonyl/µg of protein. Measuring this ratio is important for several reasons. First, it allows one to minimize gel to gel variation because of unequal loading of proteins on the gel. Second, it allows one to more directly compare changes in the carbonyl content of abundant to lesser abundant proteins. Third, the ratio allows one to eliminate the possibility that the changes in carbonyl content are secondary to changes in the expression of protein, i.e., either decreased or increased levels of the specific protein in the extract.  Finally, one can calculate the moles of carbonyl groups/mole of protein from this ratio once the protein is identified.  In other words, it is possible to determine the molar carbonyl content in specific proteins. 


Amici, A., Levine, R.L., Tsai, L., and Stadtman, E.A. (1989). Conversion of amino acid residues in proteins and amino acid homopolymers to carbonyl derivatives by metal-catalyzed oxidation reactions, J. Biol. Chem. 264, 3341-3346.

Bokov, A., Chaudhuri, A., and Richardson, A. (2004). The role of oxidative damage and stress in aging.  Mech. Age. Dev., 125, 811-26.

Chaudhuri, A.R., de Waal, E.M., Pierce, A., Van Remmen, H., Ward, W.F., and Richardson, A. (2006). Detection of Protein Carbonyls in Aging Liver Tissue: A Fluorescence-based Proteomic Approach.  Mech. Age. Dev., 127, 894-861. 

Doorn, J. A. and Petersen, D. R. (2002). Covalent modification of amino acid nucleophiles by the lipid peroxidation products 4-hydroxy-2-nonenal and 4-oxo-2-nonenal. Chem. Res. Toxicol. 15, 1445-1450.

Huggins, T. G., Wellsknecht, M. C., Detorie, N. A., Baynes, J. W., and Thorpe, S. R. (1993). Formation of O-Tyrosine and Dityrosine in Proteins during Radiolytic and Metal-Catalyzed Oxidation. J. Biol. Chem. 268, 12341-12347.

Levine, R. L., (1983). Oxidative modification of glutamine synthetase. I. Inactivation is due to loss of one histidine residue. J. Biol. Chem. 258, 11823-11827.

Levine, R.L., Garland, D., Oliver, C.N., Amici, A., Climent, I., Lenz, A.G., Ahn, B.W., Shaltiel, S., and Stadtman, E.R. (1990). Determination of carbonyl content in oxidatively modified proteins. Methods Enzymol. 186, 464-478.

Nakamura, and Goto, S. (1996). Analysis of protein carbonyls with 2,4-dinitrophenyl hydrazine and its antibodies by immunoblot in two-dimensional gel electrophoresis, J. Biochem. 119, 768-774.

Nakamura, A.and Goto, (1996). Analysis of protein carbonyls with 2,4-dinitrophenyl hydrazine and its antibodies S. by immunoblot in two-dimensional gel electrophoresis, J. Biochem. 119, 768-774.

Oliver, C.N., Ahn., B,, Moerman, E.J., Goldstein, S., and Stadtman, E.R. (1987). Age-related changes in oxidized proteins. J. Biol. Chem. 262, 5488-5491.

Stadtman, E.R. (1990). Metal ion-catalyzed oxidation of proteins: biochemical mechanism and biological consequences, Free Radic Biol Med. 9, 315-25.

Experimental Methods

The tissue sample is homogenized in deaerated buffer [20 mM sodium phosphate buffer pH 6.0 containing 0.5 mM MgCl2, 1 mM EDTA and protease cocktail inhibitors (500 µM AEBSF, HCl, 150 nM aprotinin, 0.5 mM EDTA, disodium salt and 1 µM leupeptin hemisulfate)] and centrifuged at 4oC for 1 hr at 100,000g. The resulting supernatant (cytosolic fraction) is treated with 1% streptomycin sulfate and incubated at 37oC for 10 min. The solution is centrifuged at 11,000g for 10 min at room temperature to remove nucleic acids which contain reactive carbonyl groups. The protein concentration in the supernatant was measured by the Bradford assay, and used to measure protein carbonyl groups either before or after exposing the cytosolic extracts to an oxidative stress.

The cytosolic extracts are diluted to 1 mg/ml, and the extracts are then mixed with FTC (1 mM) and incubated at 37oC for 150 min in the dark. The proteins are precipitated with an equal volume of 20% chilled TCA (v/v) and centrifuged at 16,000g for 5 min at 25oC. The pellets are then re-suspended and washed five times with 100% ethanol/ethyl acetate (1:1) to remove the unbound FTC. The final pellets were then dissolved in phosphate buffer pH 8.0 containing 0.5 mM MgCl2, 1 mM EDTA and 8M urea. The concentration of the protein in each sample is measured by Bradford assay, and approximately 15-25 µg of protein is subjected to 12% SDS-PAGE (sodium dodecyl sulfate-polyacrylamide gel electrophoresis). After electrophoresis, the image of the fluorescent protein on the gel is captured with the Typhoon 9400 using an excitation wavelength of 488 nm and an emission filter at 520 nm with a 40nm bandpass. The intensity of fluorescence for each lane (from top to bottom of the lane) was calculated using ImageQuant 5.0 (Molecular Dynamics, Amersham) software.

Carbonyl groups in specific proteins are determined using 2D gel electrophoresis as described by Chaudhuri et al., (2006). Proteins are separated in the first dimension by isoelectric focusing (IEF) system using pH 3-10 Imobiline dry strips (GE Healthcare). The proteins are then separated in the second dimension using 12% SDS-polyacrylamide gel (w/v). For 2-D gel electrophoresis, the preparation of the FTC-labeled samples is slightly different from the protocol described for 1-D gel electrophoresis. For 2-D gel electrophoresis, the final washed FTC-labeled pellets are dissolved in rehydration buffer containing 8M urea, 2% CHAPS, 0.5% IPG and incubated at 37oC for 5 min followed by sonicating samples three times for 3 min. The samples are centrifuged at 16,000g for 2 min, and the protein in the supernatant applied to the gels. After electrophoresis, the gels are scanned using the Typhoon 9400 variable mode imager with an excitation wavelength of 488 nm and an emission filter at 520 nm with a 40 nm bandpass. The image is captured in a digital format, and the data are imported into the ImageQuant 5.0. After measuring the fluorescence from FTC in the proteins, the gels are fixed with 10% methanol and 7% acetic acid for 10 min followed by staining overnight with the fluorescent dye Sypro Ruby. The stained gels are washed in 10% methanol and 7% acetic acid for 30 min to remove residual dye from the polyacrylamide matrix and placed in water. The Typhoon 9400 variable mode imager with an excitation wavelength of 532 nm and an emission filter of 610BP30 is used to scan the gels to measure the amount of Sypro Ruby associated with each spot. The gels are scanned for FTC and Sypro Ruby fluorescence, and the fluorescence intensity quantitated from 16-bit grayscale images using ImageQuant v5.0. To calculate the carbonyl content per individual protein (pmoles of carbonyl/µg of protein), the pixel intensity of FTC bound to each spot can be converted to pmoles of carbonyl using the standard curve described by Chaudhuri et al., (2006).  


Chaudhuri, A.R., de Waal, E.M., Pierce, A., Van Remmen, H., Ward, W.F., and Richardson, A. (2006). Detection of Protein Carbonyls in Aging Liver Tissue: A Fluorescence-based Proteomic Approach.  Mech. Age. Dev., 127, 894-861. 

Sample Preparation Guildelines

We require snap frozen tissue stored at -80oC for the protein carbonyl assay.  It is best not to store the tissue for more than a few months before conducting the assay to minimize potential oxidation of DNA during storage. For most tissues, 50 mg of tissue is optimal for analysis.