Nitration of protein tyrosine residues (nY) is a marker of oxidative stress and may alter the biological activity of the modified proteins. The aim of this study was to develop antibodies toward site-specific nY-modified proteins and to use histochemistry and immunoblotting to demonstrate protein nitration in tissues. Affinity-purified polyclonal antibodies toward peptides with known nY sites in MnSOD nY-34 and of two adjacent nY in the sarcoplasmic endoplasmic reticulum calcium ATPase (SERCA2 di-nY-294,295) were developed. Kidneys from rats infused with ANG II with known MnSOD nY and aorta from atherosclerotic rabbits and aging rat skeletal and cardiac sarcoplasmic reticulum with known SERCA di-nY were used for positive controls. Staining for MnSOD nY-34 was most intense in distal renal tubules and collecting ducts. Staining of atherosclerotic aorta for SERCA2 di-nY was most intense in atherosclerotic plaques. Aging rat skeletal muscle and atherosclerotic aorta and cardiac atrium from human diabetic patients also stained positively. Staining was decreased by sodium dithionite, which chemically reduces nitrotyrosine to aminotyrosine, and the antigenic nY-peptide blocked staining for each respective nY site but not for the other. As previously demonstrated, immunoblotting failed to detect these modified proteins in whole tissue lysates but did when the proteins were concentrated. Immunohistochemical staining for specific nY-modified tyrosine residues offers the ability to assess the effects of oxidant stress associated with pathological conditions on individual proteins whose function may be affected in specific tissue sites.
- manganese superoxide dismutase
- sarcoplasmic endoplasmic reticulum calcium adenosinetriphosphatase
oxidative stress affects the function of cells in tissues affected by a variety of diseases. 3-Nitrotyrosine is a posttranslational oxidant modification of proteins, which has increased abundance in diseased tissues, offering a marker of oxidant stress (3). Tyrosine nitration does not appear to occur indiscriminately because screens for proteins nitrated in animal models of sepsis or diabetes detect on the order of 100 affected proteins (6, 10). Nitration can influence the function of proteins, and some proteins such as manganese superoxide dismutase (MnSOD; SOD2) (8) and prostacyclin synthase (9) that are nitrated in their active catalytic sites are inactivated by nitration of single tyrosine. Other proteins such as synuclein (4) and the sarcoplasmic reticulum calcium ATPase (SERCA) (1, 11) appear to be nitrated in diseased tissues on multiple tyrosines. Nitrotyrosine usually occurs on a fraction of the total tyrosine residues on any one protein, making it difficult to detect protein-specific nitration. Western blotting is often negative or shows nonspecific staining, but immunoprecipitation together with immunoblotting has permitted detection of nitration on specific proteins (1, 5). Studies of rats infused with ANG II for 7 days showed that MnSOD accounted for 20% of the total tyrosine-nitrated proteins in renal tissue and was associated with a 50% reduction of enzymatic activity. In addition, atherosclerotic rabbit and human aorta showed tyrosine nitration of SERCA that was associated with impaired vasodilator function and oxidative stress.
The purpose of this study was to develop and test polyclonal antibodies to specific peptides containing nitrotyrosine at MnSOD-Tyr34, shown to be the residue most susceptible to nitration by peroxynitrite (8), and a di-nY in SERCA Tyr-294,Tyr-295 identified in aging skeletal (11) and cardiac muscle (7). Tyrosine nitration of both of these proteins appears to be associated with protein dysfunction.
Affinity-purified rabbit anti-MnSOD nY-34, and anti-SERCA2 di-nY-294,295 were provided by Bethyl Laboratories (Montgomery, TX). Antibodies were raised against the tyrosine-nitrated peptides: MnSOD, 25LHHSKHHAA(nY)VNNLNV40; SERCA, 281DPVHGGSWIRGAI(nY)(nY)FKIAV300. The peptides were chemically synthesized with the constituent amino acids including nitrotyrosine. Each antigen was injected into rabbits, and serum was obtained approximately every 2 wk over a 5-mo period. To remove antibodies to the nonnitrated sequences, antisera were processed over immunosorbents consisting of nonnitrated peptides immobilized on agarose. Subsequently, specific antibodies against nitrated peptides were column purified from the processed antisera by using immunosorbents consisting of the nitrated peptides immobilized on agarose. Yields of antibody ranged from 5 to 20 μg/ml. Polyclonal anti-nitrotyrosine antibody was obtained from Upstate Biotechnology (Lake Placid, NY). A polyclonal anti-SERCA antibody conjugated to Sepharose was obtained from Bethyl Laboratories. A polyclonal anti-MnSOD antibody was obtained from Upstate Biotechnology, and a monoclonal anti-SERCA2 antibody (clone IID8) was obtained from Affinity Bioreagents (Golden, CO).
Kidney from rats infused with ANG II (0.72 mg·kg−1·day−1) for 7 days and thoracic aorta from New Zealand rabbits fed a diet with 0.5% cholesterol and 4% peanut oil for 13 wk were obtained as described previously (1, 5). Samples of aging rat skeletal muscle and cardiac sarcoplasmic reticulum membrane fraction were obtained from 34-mo-old male Fischer 344 × Brown Norway F1 hybrid rats purchased from the National Institute on Aging, from colonies maintained at Harlan Sprague-Dawley (Indianapolis, IN). These have previously been shown by mass spectrometry studies to contain increased amounts of SERCA di-nY-294,295 (7, 11). The rats were allowed to adapt for 2 wk after arrival in a 12:12-h light-dark cycle and were provided with water and food ad libitum. The animals were anesthetized by CO2 and killed by decapitation, the heart was obtained, and hindlimb skeletal muscle was rapidly sliced into 5-mm-thick samples and fixed in 10 vol of 10% buffered formalin acetate (Fisher). Human aortic punch biopsies, discarded femoral artery segments, and cardiac atrial biopsies were obtained at the time of surgery from patients with atherosclerosis and diabetes mellitus. The tissues were fixed in 10% buffered formalin acetate and embedded in paraffin. Study of these specimens was approved by the Boston University Medical Center Institutional Review Board.
After removal of paraffin and rehydration, tissue sections (5 μm thick) were treated with 10 mmol/l citric acid (pH 6.0) and were microwave heated (2 min, 3 times at 700 W) to recover antigenicity. Nonspecific binding was blocked with 10% normal goat or horse serum in PBS (pH 7.4) for 30 min before incubation with individual primary antibodies, i.e., polyclonal anti-nitrotyrosine antibody (1 μg/ml) termed “pan-nY” here, polyclonal anti-MnSOD antibody (3 μg/ml), monoclonal anti-SERCA2 antibody (2.8 μg/ml), polyclonal anti-MnSOD nY-34 antibody (4 μg/ml), or polyclonal anti-SERCA2 di-nY-294,295 antibody (2 μg/ml), each in PBS with 1% BSA overnight at 4°C. Tissue sections were then incubated for 30 min at room temperature with a biotinylated anti-rabbit IgG secondary antibody (1:800 for nitrotyrosine or 1:200 for others) by using the Vectastain ABC kit (Vector Laboratories, Burlingame, CA). Vector Red alkaline phosphatase substrate (Vector) was used to visualize positive immunoreactivity. Specificity of anti-pan-nY, anti-MnSOD nY-34, or anti-SERCA2 di-nY-294,295 antibodies was confirmed by treating sections for 40 min with sodium dithionite (100 mmol/l in 100 mmol/l sodium borate, pH 9.0) to reduce nitrotyrosine to aminotyrosine or by preincubating the antibodies with free MnSOD nY-34 peptide or SERCA2 di-nY-294,295 peptide (antibody:peptide, 1:5 by weight, respectively). Antibodies against nonnitrated MnSOD (polyclonal) or SERCA2 (monoclonal, clone IID8) were used for testing specificity of staining. Each tissue section was stained in parallel with an adjacent section of the same tissue stained by using a nonimmune IgG (Vector Laboratories, Burlingame, CA) as the primary antibody as a negative IgG isotype control. The development of staining with the immune IgG was stopped before any nonspecific staining occurred with the nonimmune IgG to ensure specificity.
A homogenate was prepared from an atherosclerotic human femoral artery obtained at the time of leg amputation necessitated by severe peripheral arterial atherosclerotic disease. Homogenization and immunoprecipitation of SERCA was performed with a polyclonal anti-SERCA antibody (Bethyl) conjugated to Sepharose beads as previously described (2) and after SDS-PAGE was detected with the anti-SERCA2 di-nY-294,295 antibody. To eliminate staining of both the immunoprecipitation antibody and human IgG by the secondary antibody, the SERCA2 di-nY antibody was conjugated to horseradish peroxidase, used at a concentration of 0.2 μg/ml and directly visualized with ECL solution (Amersham). Protein concentrations were estimated by the bicinchoninic acid (BCA) assay (Pierce, Rockford, IL) by using bovine serum albumin as the standard. To convert nitrotyrosine to aminotyrosine, blots were incubated for 5 min with sodium dithionite (100 mmol/l in 100 mmol/l sodium borate, pH 9.0) and rinsed 2 × 10 min with 0.1 mmol/l PBS with 0.1% Tween-20 before immunoblotting.
Immunohistochemical staining for 3-nitrotyrosine.
Positive staining with polyclonal anti-nitrotyrosine (pan-nY) antibody was demonstrated in tubules within the renal inner medulla of an ANG II-infused rat (Fig. 1A), as well as in atherosclerotic rabbit aorta (Fig. 1C) as shown previously (1, 5). The staining was abolished if the antibody was preincubated with 3-nitrotyrosine (10 mmol/l, data not shown) (1, 5). In addition, staining throughout the kidney (Fig. 1B) and in the aorta (Fig. 1D) was largely prevented by pretreatment of the tissue sections with sodium dithionite (100 mmol/l, 40 min) to reduce nitrotyrosine to aminotyrosine.
ANG II infusion for 7 days dramatically increased staining of the renal medullary tubular cells with the MnSOD nY-34 antibody (Fig. 2, A and B). Lesser increases were noted in the cortical distal tubules and glomeruli (Fig. 2, C and D). Treatment of the renal tissue sections with dithionite nearly eliminated staining by the MnSOD nY-34 antibody (Fig. 3, A and B). Dithionite treatment had no apparent effect on staining the sections with a polyclonal antibody against MnSOD (Fig. 3, C and D), indicating the specificity of dithionite treatment to eliminate nitrotyrosine. Staining of ANG II-infused rat kidneys with anti-MnSOD nY-34 antibody was also blocked by prior treatment of the antibody with the MnSOD nY-34 peptide but not with the SERCA2 di-nY-294,295 peptide, further demonstrating specificity of the antibody (Fig. 4A).
Immunoblotting of homogenates of normal or ANG II-infused rat kidney failed to detect any MnSOD nY-34 (Fig. 4B). The MnSOD nY-34 antibody did detect a 26-kDa protein, as well as several higher molecular weight species in recombinant MnSOD treated with peroxynitrite (50 μmol/l) (Fig. 4B, rightmost lane). The nitrated recombinant protein could also be detected when spiked into normal kidney homogenate (Fig. 4B, second lane from left), and the staining was prevented by dithionite.
Staining of the media and intimal lesion of atherosclerotic rabbit aorta with the SERCA2 di-nY-294,295 antibody (Fig. 5A) was similar to those in Fig. 1C and previously observed with the polyclonal anti-pan-nY antibody (1). Dithionite strongly decreased staining by the anti-SERCA2 di-nY-294,295 antibody throughout the aortic cross section (Fig. 5B). Dithionite did not affect staining of the aortic media by the monoclonal antibody for SERCA2 (Fig. 5, C and D). Staining of atherosclerotic rabbit aorta with anti-SERCA2 di-nY antibody was blocked by pretreating the SERCA2 di-nY antibody with the di-nY SERCA2 peptide but not with the MnSOD nY-34 peptide (Fig. 6A).
An immunoblot performed on sarcoplasmic reticulum obtained from aging rat cardiac muscle with the SERCA2 di-nY antibody showed a prominent 110-kDa band. The staining was decreased by pretreating the antibody with the SERCA2 di-nY-294,295 antigenic peptide but not the nonnitrated SERCA2 peptide (Fig. 6B).
Strong immunohistochemical staining for SERCA2 di-nY was observed in aging rat skeletal muscle, and the staining was decreased by prior incubation of the antibody with the di-nY-294,295, but not the nonnitrated peptide, as well as by treating the tissue section with dithionite (Fig. 7).
Human diabetic atherosclerotic aorta biopsies stained positively with the anti-SERCA2 di-nY antibody in a pattern similar to that seen in the atherosclerotic rabbit aorta (Fig. 8A). Both plaque and smooth muscle beneath it were stained. Dithionite treatment eliminated staining (Fig. 8B). In addition, SERCA that was immunoprecipitated with Sepharose-conjugated anti-SERCA antibody from a homogenate of a segment of human atherosclerotic superficial femoral artery stained positively with anti-SERCA2 di-nY antibody (Fig. 8C), and detection by the antibody was prevented by treatment of the blot with dithionite.
Biopsies of human diabetic atherosclerotic cardiac atrium stained positively both for MnSOD nY-34 and SERCA2 di-nY (Fig. 9). The staining was located in myocardial cells and was prevented by pretreatment of the tissue sections with dithionite.
This study demonstrates the feasibility of developing antibodies toward oxidant modifications of specific proteins, identifying not only the type of modification, in this case nitrotyrosine, but also the protein affected. Staining for such oxidant-modified proteins is widespread in tissues of pathological and aging animal models and humans with cardiovascular disease. The two antibodies introduced here were shown to specifically detect nitrotyrosine in MnSOD and SERCA in tissues in which the modification had previously been demonstrated by other methods. Tyrosine-nitrated MnSOD was demonstrated to increase in ANG II-infused rat kidney where it represented ∼20% of the total nitrated proteins in the kidney and was implicated in decreased MnSOD activity (5). However, as in the present study, Western blotting of whole tissue homogenate failed to detect the nitrated protein unless the protein was first immunoprecipitated and then immunoblotted. Immunoprecipitation of SERCA from homogenates of atherosclerotic rabbit aorta was also required for detection of nitrotyrosine (1). The requirement for immunoprecipitation to reach the required sensitivity for detection of protein-specific nitrotyrosine may be explained by the fact that despite the high fraction of tyrosine nitration of some specific proteins like MnSOD or SERCA, the total amount of nitrotyrosine on any one protein is very low. Very often when nitrotyrosine is reported on the basis of whole tissue immunoblots, the staining has not been verified with the multiple controls used here. Another problem with SERCA in particular is that because it is prone to aggregation on heating, immunoblots are usually performed on unboiled protein, and immunoglobulin heavy chain dimers under these conditions migrate at the same apparent molecular weight. Special attention has to be paid to this possibility. Immunohistochemistry is far more sensitive than Western blotting for the detection of nitrotyrosine. The sites of tyrosine nitration to which the antibodies utilized in this study were directed were previously shown to be nitrated in the models studied. MnSOD nY-34 was shown to be the most sensitive to nitration by peroxynitrite, and the protein modification was demonstrated by mass spectrometry (8). Likely due to the fact that the tyrosine residue is within the active site of the enzyme and involved in its catalytic activity, nitration of this tyrosine residue is associated with inactivation of the enzyme. In the ANG II-infused rat kidney, MnSOD activity was decreased by 50% without a change in protein expression (5), consistent with Tyr-34 nitration demonstrated in this study. In addition, by using the sequence-specific MnSOD-nY antibody described here, an increase in staining was demonstrated in the kidneys of diabetic apolipoprotein E-deficient mice (12). The increase in staining correlated with a 30% decrease in MnSOD activity, further linking tyrosine nitration of MnSOD to its inactivation in vivo.
SERCA2 di-nY-294,295 was demonstrated by chemical methods and mass spectrometry to be present in aging skeletal and cardiac muscle sarcoplasmic reticulum (7, 11), and in unpublished studies we have identified by mass spectrometry the di-nY-294,295 in SERCA from atherosclerotic rabbit aorta (T. Adachi and J. Ying, unpublished data). Earlier studies showed that tyrosine nitration of SERCA is associated with decreased activity of the enzyme (1), although the specific functional effect of di-nY-294,295 is not yet known. This pair of tyrosine residues is located on the inner aspect of the sarcoplasmic reticulum membrane on one of the transmembrane loops that constitute the calcium pore of the enzyme, suggesting the potential functional importance of this nitrated pair of tyrosines. When tyrosine nitration was decreased in atherosclerotic rabbit aorta after treatment with an antioxidant, SERCA activity was restored, suggesting a relationship between tyrosine nitration and activity (1).
The specificity of the two antibodies used in this study was demonstrated by showing that staining was decreased by dithionite, which reduces nitrotyrosine to aminotyrosine. In addition, we considered whether an antibody raised to the tyrosine-nitrated peptides used here might be immunoreactive toward nitrotyrosine residues in general. However, this possibility was discounted by showing that immunogenicity of the antibody was blocked by the specific nitrated peptide but not the nitrated peptide of the other protein. Giasson et al. (4) took another approach in raising monoclonal antibodies directed toward synuclein that was tyrosine nitrated in vitro. Unlike the anti-peptide antibodies developed here, those antibodies showed reactivity toward several sites of tyrosine nitration on synuclein.
Use of antibodies directed toward site-specific protein oxidation uniquely compliments methods already developed. These include immunochemical, biochemical, and mass spectral techniques. For instance, gel electrophoresis with immunoblotting has the potential to identify proteins with nitrotyrosine modifications but not the specific sites of nitration. Mass spectral and chemical techniques are able to definitively identify sites of tyrosine nitration, although the sensitivities of the methods are, so far, less than immunochemical detection. In addition, there is an added advantage to immunohistochemical detection because the tissue required is minimal, as demonstrated in this study by the positive staining detected in diseased patient tissue biopsies.
The proteins to which the antibodies were developed here were chosen for their suspected relevance for function in tissues affected by oxidant stress. The ability to detect tyrosine nitration of specific proteins like MnSOD and SERCA in tissues will allow investigators to detect if these or other proteins are affected by oxidants in vivo. The advantage afforded by antibodies toward specific oxidant-modified sites provides the ability to infer a change in protein function that has been identified to be associated with the modification. This should add significantly to our knowledge of the effects of oxidant stress on tissues and cells in vivo.
The studies were supported by National Institutes of Health (NIH) Grants R01-HL-31607–21, R01-HL-55620–08, and R01-AG-27080–01, the NIH Boston University Cardiovascular Proteomics Center (N01-HV-28178), American Heart Association Grant-in-Aid 0455799T, and NIH Grant 2P01-AG-12993 (V. Sharov, C. Schöneich, and D. Bigelow).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 by the American Physiological Society