|Home | About | Journals | Submit | Contact Us | Français|
Biomolecule-centered radicals are intermediate species produced during both reversible (redox modulation) and irreversible (oxidative stress) oxidative modification of biomolecules. These oxidative processes must be studied in situ and in real time in order to understand the molecular mechanism of cell adaptation or death in response to changes in the extracellular environment. In this regard, we have developed and validated immuno-spin trapping to tag the redox process, tracing the oxidatively-generated modification of biomolecules, in situ and in real time, by detecting protein- and DNA-centered radicals. The purpose of this method article is to introduce and update the basic methods and applications of immuno-spin trapping for the study of redox biochemistry in oxidative stress and redox regulation. We describe in detail the production, detection and location of protein and DNA radicals in biochemical systems, cells, and tissues, and in the whole animal as well, by using immuno-spin trapping with the nitrone spin trap 5,5-dimethyl-1-pyrroline N-oxide (DMPO).
The oxidative modification of biomolecules has been systematically observed under normal and pathological conditions [1–3]. However, to understand the role that redox biochemistry has in health and disease requires the identification of the biomolecule(s) that is/are targets of such modification(s), the specific residue(s) where the radical was first generated, the nature of such modifications, the cellular and subcellular locations of biomolecule-centered free radicals in metabolically active systems, and their localization in the whole animal.
Oxidatively-generated damage to biomolecules, e.g., proteins and DNA, produce protein- and DNA-centered radicals that generally occur through initial abstraction of an electron or hydrogen atom [4, 5] (Scheme 1). These biomolecule-centered radicals often decay quickly; they can react either with oxygen, resulting in fragmentation or post-translational modifications [6, 7], or with neighboring biomolecules or antioxidants, resulting in aggregation or scavenging, respectively . A free radical is an atom, ion or molecule that is usually, very reactive and unstable because it has one and only one unpaired electron in an outer orbital, which explain its paramagnetic properties. Exceptions include paramagnetic transition metals like copper. The unpaired electron gives these species paramagnetic properties that make them suitable for detection by electron spin resonance (ESR) spectroscopy, the “gold standard” technique used to detect free radicals . However, because of the high reactivity of protein- and DNA-centered radicals, they are generally stable for only microseconds to seconds before they decay to produce diamagnetic (ESR-silent) species; although stable protein radicals such as the tyrosyl radical of ribonucleotide reductase do exist .
In the spin-trapping technique, a reactive radical (R•) adds across the double bond of a diamagnetic compound, known as a spin trap, to form a much more stable free radical, a nitroxide radical adduct or radical adduct, which can then be examined by ESR [9, 11] (Scheme 1). This technique is called ESR-spin trapping. Spin trapping was a critical technical advance in the detection of free radicals in biology since the radical adducts, for example lipid-radical adducts, have lifetimes of minutes and, in a few cases, even hours, which means that biological free radicals can be detected in many biological systems in vitro and, in some cases, even in biological fluids (bile, blood, and urine) from living animals [9, 12, 13]. The analyses of protein and DNA radicals by ESR or ESR-spin trapping are usually performed in chemical systems by exposing the isolated cellular biomolecules [14, 15] or their components (amino acids, fatty acids, bases, nucleosides, nucleotides, and sugars) [16, 17] to oxidizing conditions (i.e., peroxidases/peroxides, hypohalous acids, Fenton systems, ozone, and irradiation) in the absence or presence of a spin trap followed by analysis by ESR  (for an example, see Figure 1A). However, as a practical matter, the ESR or ESR-spin trapping analysis of protein and DNA radicals and their radical adducts produced in functioning cells is complex because the time required to prepare homogenates or to isolate the DNA from the biological matrix is typically much longer than the decay of the parent radical(s) or radical adduct(s) .
Previously, we have published step-by-step protocols for the immuno-spin trapping analysis of protein-  and DNA-centered  radicals. Those protocols have been used as a basis for expanding the field of biomolecule-centered free radical detection in cell, tissue, and whole animal models (Table 1). In this update, the production and detection of protein and DNA radicals in biochemical, cell, tissue, and whole animal systems using immuno-spin trapping with the nitrone spin trap 5,5-dimethyl-1-pyrroline N-oxide (DMPO) is described. Immunoanalyses, mass spectrometry (MS), and molecular magnetic resonance imaging (mMRI) are used to investigate the resulting nitrone adducts.
Among the spin trap compounds, the nitrone spin trap DMPO is the least toxic, and has convenient pharmacokinetics (uptake, distribution, metabolism, and excretion) in biological systems, i.e., cells, parasites, and animals . DMPO, which is soluble in water and organic solvents, can access any cellular compartment and thus can trap, in situ and in real time, protein- and DNA-centered radicals whenever and wherever they are produced. The adducts thus formed (DMPO-biomolecule adducts) remain stably bound in most cases, thereby facilitating their extraction and immunoanalysis as nitrone adducts, which are usually as stable as DMPO itself [19, 20, 22] (Figure 1C). Accordingly, we have developed a new technology to detect protein [22–24] and DNA radicals [19, 25] which we have named immuno-spin trapping (Scheme 1). See Table 1 for a complete list of references on immuno-spin trapping.
Immuno-spin trapping combines the specificity of spin trapping with the specificity and sensitivity of antigen-antibody interactions by detecting the nitrone moiety in DMPO-protein or DNA- radical-derived nitrone adducts (hereafter referred to as nitrone adducts) with a rabbit antiserum against DMPO (Scheme 1). Protein and DNA radicals are trapped in situ and in real time by DMPO and form radical adducts. The radical adduct decays by oxidation, to form a nitrone adduct that is recognized by the anti-DMPO antiserum. The anti-DMPO antiserum recognizes DMPO, but not the molecules to which DMPO is bound. The nitrone moiety in DMPO makes it highly antigenic.
The anti-DMPO antibody was produced as described in Detweiler et al. . Briefly, DMPO was conjugated to an octanoic acid (OA) chain in the C2 position of the pyrroline N-oxide ring. The resulting hapten (DMPO-OA) simulates the trapping of a radical by DMPO. The DMPO-OA complex was then conjugated to ovalbumin (OVA) to produce the immunogen (DMPO-OA-OVA) that was used to immunize rabbits following an immunization plan. This antiserum was used to develop and validate immuno-spin trapping  and recognizes the hapten (DMPO-OA), free DMPO, and protein- or DNA-DMPO nitrone adducts. Basic step-by-step protocols for immuno-spin trapping of protein - and DNA -centered radicals have been published elsewhere and updated in the present method review paper.
Because the anti-DMPO antiserum recognizes DMPO, the immunoanalyses of DMPO-biomolecule nitrone adducts are heterogeneous, i.e., they require separation of biomolecule-centered radicals from free DMPO and DMPO-small molecule-centered radicals (for example, DMPO-dithiotreitol nitrone adducts which binds to solid supports) . It is a major advantage, however, that the nitrone group does not exist in nature, thus eliminating the possibility of a false positive from endogenous cellular components. Thus immuno-spin trapping involves three main steps [19, 20]: i) trapping macromolecule-centered radicals to form stable nitrone adducts in situ and in real time; ii) separation and/or extraction of the nitrone adducts, and iii) immuno-detection/localization of the nitrone adducts.
NOTE: Not all the reagents and equipment that follow are needed for each variant of immuno-spin trapping assays. Before starting, read the entire protocol to determine which reagents and equipment will be needed for a particular option of the immuno-spin trapping assay.
NOTE: Among redox biochemical systems, the reaction of hemoproteins such as hemoglobin or myoglobin with H2O2 is the best characterized. Hemoglobin or myoglobin nitrone adducts are prepared by reacting the hemoproteins (usually 1–10 μM) with a 1- or 10-fold excess of H2O2 in phosphate buffer and in the presence of 1–100 mM DMPO. Amine-containing buffers (for example, tris(hydroxymethyl)amino-methane, TRIS) should be avoided because of their ability to react with ROS. Preparation of nitrone adducts in 10–50 mM ammonium bicarbonate, pH 8.0, is used in some superoxide dismutase (SOD) investigations, but not, for example, in hypochlorous (HOCl) oxidations due to the scavenging of HOCl by the tertiary amine, forming chloramines that are also oxidants . We recommend using a phosphate buffer; however, note that phosphate at high concentrations can chelate metals such as iron and copper, thus modulating Fenton chemistry. The reactions are stopped by removing excess H2O2 with catalase or by inhibiting the heme with cyanide or azide in peroxidase/peroxide systems or chelating agents in Fenton systems. The removal of excess reagents and DMPO by dialysis with an appropriate cut-off membrane is the safest choice. Other materials and reagents are included in the protocol.
Three μl of a hemoprotein solution hundredfold more concentrated than the final heme concentration (100X stock) is added to 291 μl of 100 mM chelexed phosphate buffer, pH 7.4, containing 0.1–0.5 mM DTPA followed by addition of 3 μl of a 1M DMPO solution (in water or the same buffer; the neat compound is ~10 M), and the reaction is started by adding 3 μl of freshly prepared 100X H2O2 solution to the reaction mixture. The reaction mixture is incubated at 37 °C for 1 h with agitation in a thermomixer and stopped by removal of excess H2O2 with 3 μl of a catalase solution in phosphate buffer, pH 7.4 (Roche Applied Bioscience, 100 or 500 IU in 0.1 M phosphate buffer, pH 7.4). Placing the tubes containing the reaction mixture into an ice bucket will slow the radical chemistry; however, if the reaction components in excess are not removed or the reaction is not stopped, further formation of nitrone adducts continues during the analysis. Do not add stop solution to the samples that are prepared for mass spectrometric (MS) analyses. Reactions for MS analysis should be quenched by dialysis at 4 °C against 10–50 mM ammonium bicarbonate buffer, pH 8.0, and then stored at −20 °C or −80 °C until analysis (see analysis of nitrone adducts). Dialysis is preferred over the addition of quenching agents as these compounds can interfere with subsequent MS analyses. DMPO is covalently bound to the atomic site of highest electron spin density. If frozen at −20 or −80 °C, nitrone adducts are generally stable for years without significant alterations of the nitrone epitope(s). Representative results of ELISA and Western blot analyses of Mb-DMPO nitrone adducts and controls are shown in Figures 2A–2D.
Although there are several ways to produce DNA nitrone adducts, the reaction of calf-thymus DNA or DNA isolated from animal tissues with copper and H2O2 produces high amounts of nitrone adducts (see  for a detailed step-by-step protocol). DNA nitrone adducts are prepared as follows: 10 μM calf-thymus DNA as nucleotide (1 unit of absorbance at 260 nm is ~50 μg DNA/ml which, in turn, is ~150 μM of DNA as nucleotides) in 10 mM chelexed sodium phosphate buffer is incubated with 10 μM Cu2+ (chloride or sulfate), and the reaction is initiated by addition of H2O2 to a final concentration of 100 μM. The total volume of the reaction is usually 300 μl. After 30 or 60 minutes of incubation at 37 °C, the reaction is stopped either by removing the excess H2O2 with 5 μl of catalase (10 IU), inhibiting Cu2+ redox cycling with 1 mM cyanide, or by chelating Cu2+ with DTPA (10 mM). DNA nitrone adducts can be extracted from tissues and cells exposed to oxidizing or inflammatory conditions in the presence of DMPO with the most up-to-date and detailed protocols for production, extraction, analysis, and representative results published elsewhere [19, 20].
This is a heterogeneous immunochemical technique (i.e., a procedure that requires separation between bound and unbound material in some step(s)) that allows simultaneous screening of many samples . In this technique, 10 μl of the protein nitrone adduct solution or homogenate (1 μg/10 μl) is added to 190 μl of a 0.1 M bicarbonate buffer, pH 9.6, or PBS, pH 7.4 in 96-well microtiter plates. The protein binds to the bottom of the well through hydrophobic interactions . Depending on whether the development method is chromogenic or chemiluminescent, we recommend white or transparent, respectively, high-protein-binding, flat-bottom, 96-well microplates. After coating, the non-bound material is removed by a washing step (300 μl of washing buffer per well) with a buffer containing a detergent and an inert protein to remove or avoid, respectively, any nonspecific interactions. We recommend a washing buffer composed of PBS, 0.05 % Tween-20 and 0.1 % fat-free milk and the use of a microplate washer. This step allows the removal of any non-protein nitrone adducts that can interfere with the detection of protein nitrone adducts because the anti-DMPO antibody recognizes both protein-bound and free DMPO . These characteristics represent one of the most important advantages of heterogeneous assays like ELISA compared to homogeneous assays.
Next, block the nonspecific binding sites with 100 μl of blocking buffer (1% fat-free milk in PBS, pH 7.4) and incubate the microplate for 40 to 60 minutes at 37 °C. After blocking, wash the plate one time with washing buffer as before. Remove the last washing solution and add 100 μl of the rabbit anti-DMPO serum diluted in washing buffer (1:10,000). After 60 minutes of incubation, nonspecifically-bound and unbound antiserum is removed by two or three washing steps followed by the addition of 100 μl of a goat anti-rabbit IgG conjugated with HRP. Typical dilutions of the secondary antibody for ELISA are 1:10,000, and incubations are performed at 37 °C for 60 minutes. After incubation, non-bound and weak nonspecific interactions are removed by two or three washing steps with 300 μl of washing buffer per well. Finally, immuno-complexes (nitrone adduct/anti-DMPO antibody/anti-rabbit IgG-HRP) are detected by adding 50 μl of the substrate solution that produces luminescence (VisiGlo Chemilu HRP substrate kit; Amresco; cat. no. 218-kit) or soluble colored products (1-Step™ Ultra TMB-ELISA, Pierce; cat. no. 34028) that can be detected and measured using a microplate reader (Infinite M200, Tecan, RTP, NC). Representative results of the analysis of Mb-DMPO nitrone adducts by ELISA are presented in Figures 2A and 2C.
To perform the Western blot analysis of protein nitrone adducts in cell or tissue homogenates, mix 30 μl of the reaction mixture or cell homogenate (0.1–1 mg/ml) with 10 μL of NuPage™ LDS Sample Buffer (4X) (Invitrogen; cat. no. NP0007) and 4 μL 10X NuPAGE® Sample Reducing Agent (Invitrogen; cat. no. NP0004). Heat the sample in a water bath at 90 °C for 7–10 min, and then let cool to room temperature. Load 10 μl of sample (1–40 μg proteins) per lane in 1.0 mm 10- well 4–12 % NuPAGE® Novex Bis-Tris Gels (Invitrogen; cat. no. NP0321BOX). For better standard molecular weight band visualization, we add 1 to 3 μl SeeBlue® Plus2 Pre-Stained Standard (Invitrogen; cat. no. LC5925) in the first and in the middle lane of the gel. Alternatively, the use of MagicMark™ XP Western standard (Invitrogen; cat. no. LC5602) helps to locate the approximate molecular weight in the developed X-ray film or imaged membrane. Perform the protein separation using an XCell SecureLock™ Mini-Cell system under continuous voltage conditions for 42 min at 200 V. Blot the separated proteins onto a nitrocellulose membrane following the manufacturer’s instructions. We use a nitrocellulose membrane filter paper sandwich, 0.45 μM pore size (Invitrogen; cat. no. LC2001), and the blotting is performed in an XCell II™ Blot Module from Invitrogen (cat. no. EI9051).
After assembling the blot and inserting the module, the proteins are blotted to a nitrocellulose membrane at 40 V for 20–30 min. A semi-dry transfer apparatus (BioRad Laboratories, Hercules, CA) is advantageous with respect to time (~15 V for 20 min) and reproducibility when blotting the proteins to the membrane. After transfer is completed, the blotting efficiency can be examined by exposing the membrane to a solution of 1% ponceau S red in 5% acetic acid for 30 seconds and then washing the membrane with distilled water until red bands appear in a clear background. The red bands will disappear during the blocking of the membrane.
To identify protein DMPO nitrone adducts, incubate the membrane in an appropriate, clean glass or plastic container with 10–25 ml of blocking buffer for 40–60 minutes at room temperature with low agitation in an orbital shaker. Blocking and washing buffer are prepared as described for the ELISA procedure. Remove the blocking buffer, add 10–25 ml of washing buffer, incubate the membrane for 10 minutes, and remove the washing buffer. Then add 10 ml of the anti-DMPO antiserum (1:5,000 in washing buffer), and incubate for 1 h at room temperature or overnight in the refrigerator with agitation. Wash the membrane three or four times with washing buffer with 10 minutes of agitation each time. Remove the last washing buffer and add 10 ml of a goat anti-rabbit IgGFc conjugated with alkaline phosphatase (AP) or horse radish peroxidase (HRP). In either case, the dilution of the secondary antibody is 1:5,000 in washing buffer. HRP produces more sensitive readings; however, some cell peroxidases and pseudo-peroxidases might react with H2O2 and peroxidase substrates and produce luminescence. AP is not suggested for analysis of nitrone adducts in lung tissue and macrophages, which are known to have high endogenous AP content.
Incubate the membrane for 60 minutes with the secondary antibody, and wash three to four times as before. Remove the last washing solution and incubate in a clean tray with 3–5 ml of the development reagent per membrane. If an AP conjugate is used as a the secondary antibody, we recommend the use of CDP-star solution (Roche, 1:1,000 in Tris-HCl, pH 9.6) ; however, if an HRP conjugate is used, we recommend the use of a VisiGlo Plus™ HRP chemiluminescent kit from Amresco. In both cases, incubate the membrane with the reagent for 5 minutes, and then record the luminescence by using X-ray film or an imager. For a complete protocol see , and for representative results see Figures 2B and 2D.
Mass spectrometry approaches to the identification of proteins and the specific site in the primary structure of a protein with a covalently-attached DMPO (amino acids where the radical was trapped) are continually evolving [30, 31]. Two ionization techniques are routinely used for the MS analysis of biomolecules: electrospray ionization (ESI) and matrix assisted laser desorption ionization (MALDI). MALDI analysis is very fast and is routinely used for protein identifications. ESI analysis is compatible with on-line HPLC, and, therefore, is used in conjunction with HPLC for the analysis of complex mixtures. Both ionization techniques are compatible with tandem mass spectrometry (MS/MS) – the gas-phase fragmentation of biomolecules that allows for the sequence determination of peptides/proteins and the identification of specific sites of radical adduct formation.
For the identification of proteins corresponding to positive bands in Western blots of tissues or homogenates treated with an oxidant and DMPO, Western blots are compared and/or overlayed with a Coomassie-stained gel of the same sample performed in parallel [20, 23, 30]. Following excision of the gel band and in-gel digestion, the resulting peptide mixture is analyzed by mass spectrometry. MALDI/MS analyses were used to confirm the identity of Mb from rat heart supernatant . Subsequent in vitro reactions of horse heart metMb treated with H2O2 in the presence of DMPO followed by ESI/MS analyses confirmed the formation of a Mb-DMPO radical adduct (Figure 1B) . In the deconvoluted mass spectrum of the intact protein, an ion was observed that corresponded in mass to the addition of a DMPO molecule to the protein. Following enzymatic digestion of the reaction mixture, LC/ESI/MS/MS was used to unequivocally assign the DMPO spin trap in horse heart myoglobin to tyrosine 103 (Tyr 103) [32, 33].
We have also shown that our immuno-spin trapping approach and MS analyses are complementary. Using immuno-spin trapping, we provided the first evidence of formation of a protein radical in cells . In this study we identified hemoglobin nitrone adducts in human red blood cells exposed to H2O2 and DMPO, and data from this study suggested that tyrosine and cysteine residues were the radical sites trapped by DMPO . Later, LC/ESI/MS/MS of human oxyHb treated with H2O2 and DMPO confirmed these data. Using LC/MS/MS, DMPO adducts of human oxyHb were identified on cysteine 93 in the beta chain and on tyrosine 24 and tyrosine 42 in the alpha chain. Additionally, a novel histidine residue (His 20) in the alpha chain was identified as a site of DMPO spin trapping (Figure 3) . In two recent studies, following Western blot analyses with the anti-DMPO antibody, myeloperoxidase was identified from the corresponding Coomassie-stained gel from HL60 cells [34, 35].
Moreover, we have used mass spectrometry to study the mechanism of these free radical- mediated processes. From LC/MS/MS analyses, the identification of the primary and secondary trapping sites of radical formation in human heart Mb was determined; thereby providing insight into a potentially important mechanism of free radical damage operating in ischemia-reperfusion in the heart .
The MS of nitrone adducts can be analyzed from either a solution (especially in the case of in vitro experiments) or from SDS-PAGE gels (i.e., first located by Western blot with the anti-DMPO antiserum) stained with an MS compatible stain. For staining of the gel, we suggest Coomassie blue, Simply Blue, or Sypro Ruby protein gel stain (Invitrogen; cat. no. S12001). Of note: silver staining is less compatible with MS due to a variety of reasons which, in some instances, results in the protein being covalently bound to the acrylamide gel. If possible, the samples should be free of additives such as nonvolatile salt buffers (e.g., sodium and phosphate buffers), detergents, and glycerol as these chemicals can interfere with the analyses. We suggest that the reactions be performed in volatile buffers (e.g., 50 mM ammonium bicarbonate buffer, pH 8.0) or, alternatively, in 10 mM phosphate buffer followed by dialysis at 4° C against 10 or 50 mM ammonium bicarbonate buffer, pH 8.0. Extensive prewashing of the dialysis cassette or tube membrane with ultrapure water to eliminate preservatives is highly recommended to avoid potential interference. In biochemical systems, the reaction does not require the addition of stopper (for example, catalase), but it does require dialysis overnight at 4° C against 2 L of 10–50 mM ammonium bicarbonate buffer, pH 8.0, to eliminate excess reagents.
Hemoglobin control samples (Hb, Hb+H2O2, Hb+DMPO) and the hemoglobin+DMPO+H2O2 reaction sample are either dialyzed against or prepared in 50 mM ammonium bicarbonate buffer, pH 8.0 . Samples are subjected to digestion using porcine trypsin (Promega Corporation, Madison, WI) at a protein:enzyme ratio of 20:1. The reactions are allowed to proceed overnight at 37 °C.
Electrospray ionization (ESI) mass spectra and tandem mass spectra are acquired, in our case using a Micromass Q-Tof Ultima Global (Waters/Micromass, Milford, MA) hybrid tandem mass spectrometer. This instrument is equipped with a nanoflow electrospray source and consists of a quadrupole mass filter and an orthogonal acceleration time-of-flight mass spectrometer. The needle voltage is ~3000 V and the collision energy is 10 eV for the MS analyses. For the LC/MS/MS analyses, a Waters CapLC HPLC system (Waters, Milford, MA) consisting of binary pumps and a micro autosampler is used to deliver the gradients. Injections of 1–8 nl are made on the column and a linear gradient of 5–75% acetonitrile (0.1% formic acid) over 35 min at a flow rate of ~300 nl/min is used for the chromatographic separations. The column is a 15 cm × 75 μm id Hypersil C18 (“pepmap”) column (LC Packings, San Francisco, CA).
During the LC/MS/MS experiments, we employed automated data-dependent acquisition software. For these acquisitions, the instrument can switch from the MS mode to the MS/MS mode and then return back to the MS mode based on predetermined or operator-entered parameters such as abundance and time. The advantage of this software is that both MS and MS/MS data can be acquired from a single chromatographic separation of the mixture. The collision energy for these experiments is set according to the charge state and the m/z of the precursor as determined from a charge state recognition algorithm. Data analysis is accomplished with a MassLynx data system, MaxEnt deconvolution software, and ProteinLynx software supplied by the manufacturer.
The first protein radical adducts produced and trapped by DMPO inside a cell with an intact membrane were identified as hemoglobin nitrone adducts in intact red blood cells treated with H2O2 and DMPO . The tissue  and subcellular [38, 39] localization of DMPO-protein nitrone adducts is an extension of this study and the result of over seven years of careful validation of the immuno-spin trapping technique in biochemical systems [19, 20, 23–25, 40–42]. Immuno-spin trapping represents a major breakthrough for the study of biological free radical reactions because biomolecule-centered radicals are trapped by DMPO to form protein-nitrone adducts, which can be detected in situ or isolated and characterized in detail [34, 35]. For instance, with the anti-DMPO antiserum, Bonini et al. used confocal microscopy to observe that catalase-DMPO nitrone adducts were localized inside the peroxisomes of hepatocytes exposed to hypochlorite . Confocal microscopy was also used to detect MPO-protein nitrone adducts in the liver in an acetone model of ketosis . In this study, Stadler et al. showed that protein-DMPO nitrone adducts were located inside hepatocytes lining the centrolobular vein .
For fluorescence microscopy studies, we culture RAW 264.7 macrophages as previously described . We use DMEM without phenol red and with 10 % heat-inactivated fetal calf serum, referred to as complete medium. The cells are collected by centrifugation (200×g for 10 minutes at room temperature) and resuspended in complete medium at 103 cells/ml. A two-hundred μl aliquot of the cell suspension is plated onto a glass-bottomed 8-well glass-chambered slide (Lab-Tek Chamber Slide System; Cat. no. 177402, Nalge Nunc International, Rochester, NY). After 24 h in a cell incubator (37 °C, 5 % CO2 and 99 % humidity), the monolayers are washed three times with pre-warmed (37 °C) PBS, pH 7.4. Then, DMPO (10–80 mM) in HBSS, pH 7.4, with Ca2+ and Mg2+ (HBSS+) is typically added 15–20 minutes prior to the treatment of choice. DMPO promptly diffuses throughout cell compartments where it can trap, in situ and in real time, the biomolecule-centered free radicals producing nitrone adducts.
After the treatment is completed, the monolayers are washed three times with HBSS+ and immediately fixed with methanol-free paraformaldehyde solution (4 %) dissolved in PBS. The fixative chemistry of paraformaldehyde can be stopped by incubating the slides for 5 minutes with 100 mM Tris-HCl, pH 8.0. After fixation, the cells are permeabilized with cold methanol (− 20 °C) for 2 minutes. Methanol is then disposed of and the cells are washed with HBSS+ four times. Paraformaldehyde is a cell fixative that produces cross-links between the proteins and the plate, and methanol dissolves the membrane lipids. After methanol permeabilization, the fixated cell structures are treated with a blocking buffer of choice for 2 h at room temperature or overnight at 4 °C. We recommend the use of a source of proteins of the same origin as the secondary antibody. For example, if a horse or goat anti-rabbit IgG conjugated to HRP is used, a 1% solution of normal horse or goat serum, respectively, would be the one of choice.
After blocking and washing three times with washing buffer, the anti-DMPO antiserum, diluted 1:500 to 1:2,000 in wash buffer, is added. Finally, after four washes the Alexa Fluor® 488 goat anti-rabbit IgG SFX kit (Invitrogen; cat. no. A31628 ) is added to indicate anti-DMPO binding to specific sites within the cells. Figures 4A and B contain examples of DNA radical detection using immunochemistry. RAW 264.7 cells were loaded with copper or iron (100 μM) for 4 h at 37 °C. The medium was removed and the monolayer was washed 3 times with HBSS+ with 1 mM DTPA to remove the metals from the cell surface. Then 1 ml of HBSS+ containing different concentrations of tert-butyl hydroperoxide (tert-BOOH) was added. Figure 4C shows the detection of nitrone adducts in macrophages treated as in Figures 4A and 4B using immunofluorescence.
Figure 4C shows a immunohistochemistry assay of nitrone adducts in sections of spinal cords of symptomatic SOD1G93A transgenic rats . Interestingly, a serial injection of low amounts of DMPO (doses of 50 mg/Kg) in saline 48 h and 24 h before the sacrifice of symptomatic transgenic SOD1G93A rats has proven to restore the function of mitochondria isolated from the spinal cords of these animals . Using Western blot analysis with the anti-DMPO antibody, the authors also detected increased nitrone adducts in mitochondria from the spinal cords of sick animals compared to those isolated from control animals. There were also increases in nitrone adducts in sections of the spinal cords of those sick animals injected with DMPO . More recently, Stadler et al.  observed DMPO nitrone adducts by immunohistochemistry with the anti-DMPO antiserum in sections of livers of mice receiving acetone in their drinking water and intraperitoneal injections of DMPO (20 μl of the pure compound/every 6 h for the period that the acetone treatment lasted) .
Current methods for the in vivo detection of free radicals include electron paramagnetic resonance imaging (EPRI), fast EPRI, Overhauser magnetic resonance imaging (OMRI) and conventional T1-weighted MRI. EPRI often requires the use of narrow-line spin probes, such as trityl radicals . These free radical nitroxide probes are often non-toxic and bio-compatible . Continuous wave (CW) EPRI has a high sensitivity and general applicability for a wide range of free radicals and paramagnetic species, but it also requires long image acquisition times and, therefore, is limited in its use for most in vivo applications where rapid changes in the magnitude and distribution of spins occur . Recently, a fast 3D EPRI system at L-band frequency using spiral magnetic field gradients has been developed, and shows promise for in vivo imaging of free radicals .
A comparison of EPRI, OMRI and MRI in phantom solutions was recently performed by Hyodo et al. . The authors found that the sensitivity and resolution of images obtained by EPRI and OMRI greatly depended on the linewidth of the reporting nitroxides (e.g., 3-carbamoyl-2,2,5,5-tetramethylpyrrolidine-1-oxyl (3CP)), whereas images obtained by MRI were essentially independent of the EPR line width . Temporal images were obtained using the three imaging modalities that monitored the reduction of 3CP by ascorbic acid (ASA) .
OMRI is a double resonance technique that combines the advantage of MRI for spatial resolution with the sensitivity of EPR by taking advantage of the Overhauser enhancement dependent on the linewidth of a paramagnetic agent, which is dependent on oxygen concentration [45, 46]. The advantage of using standard MRI was that it provided detailed anatomical information, as well as being able to detect in vivo tissue redox status, without a dependence on the linewidth of a reporting agent . The image enhancement that is observed in MR images is dependent on the effect of free radicals on the T1 relaxation time of surrounding water protons.
Another group used the OMRI approach via dynamic nuclear polarization (DNP), which enhances the sensitivity of nuclear magnetic resonance (NMR) spectroscopy and the contrast of MRI by transferring the significantly higher electron spin polarization of stable free radicals to nuclear spins . Stable radicals such as 2,2,6,6-tetramethyl-pypiridine-1-oxyl (TEMPO) radicals isotopically labeled with 15N and deuterium are employed as the source for the DNP polarization transfer at a magnetic field relevant for MRI (0.35-1 T). These polarized molecules, in the form of an agarose-based porous material that is covalently spin-labeled with the stable radicals, are introduced to a biological system sensitive to the presence of radicals, ensuring an Overhauser effect induced by DNP without release of radicals into the biological sample . Phantom solution studies indicate that to provide an optimal DNP polarization medium, it is necessary that there be high radical loading, sufficient water access to the radicals, high water mobility, and a close contact between the radicals and the water protons .
An extension of MRI involves the introduction of a molecular probe that is made up of a MRI contrast agent as the signaling component and an affinity component that could be a peptide or antibody for a molecular marker. We (Towner laboratory) have developed a molecular MRI approach for the in vivo detection of DMPO-protein adducts in disease models that involve oxidative stress processes such as amyotrophic lateral sclerosis (ALS) (SOD1G93A transgenic mouse model) and streptozotocin (STZ)-induced diabetes . In this approach we synthesized a Gd-DTPA (gadolinium-diethylenetriaminepentaacetic acid)-albumin-biotin MRI contrast agent that is easily distributed to a tissue/organ of interest via the vascular system (i.v. injection via a tail-vein catheter), and anti-DMPO antibodies covalently bound to the albumin moiety, which allows in vivo immuno-spin trapping . Specific binding of the anti-DMPO molecular MRI probe is characterized by a sustained increase in MR signal intensity in T1-weighted images and a corresponding decrease in T1 relaxation within affected regions . A biotin component is added for ex vivo fluorescence labeling (streptavidin-Cy3) of an anti-DMPO mMRI probe bound to a tissue/organ that generated a high level of protein radicals (trapped by administration of DMPO) .
There are also numerous developments in the use of iron oxide or magnetite nanoparticles as MRI contrast agents for mMRI probes, which may increase the sensitivity of detection. The presence of the molecular-targeted magnetite nanoparticles results in a sustained decrease in MR signal intensity and a corresponding decrease in T2 relaxation. The mMRI approach can be applied to many experimental models for diseases that involve oxidative stress processes, taking advantage of the high anatomical resolution of MRI and the specific detection of protein radicals in vivo.
Among numerous factors that need to be considered are the omission of components, type, concentrations and pH of buffers, chelators of metals and solvents, order of addition of reagents, time and temperature of incubation, target/oxidant ratio, and concentration of DMPO.
For some applications, DMPO is purified by vacuum sublimation at room temperature and stored under an argon atmosphere at −70 °C. The vacuum-sublimation purification of DMPO and the use of DTPA in the phosphate buffer are not required for cell or animal experiments. DMPO is sensitive to light, temperature, metals, and oxygen, forming H2O2 via auto-oxidation as for any organic compound. Thus, precautions should be taken during its handling. The spin trap is available in caramel-colored flasks as a crystalline solid and requires thawing at 37 °C. Once liquified, the pure DMPO is aliquoted into 50 μl per tube, and a stream of an inert gas such as argon or nitrogen over the liquid is used to evacuate oxygen. The aliquoted tubes are kept frozen at −20 °C or −80 °C and protected from light until use.
In spin-trapping experiments, because free radicals are very reactive and hence decay very fast, DMPO is added prior to or during their formation. DMPO diffuses easily through all cell compartments and must be present both in high enough concentration to outcompete the mechanisms of radical decay (for example, radical-radical, radical-oxygen and electron transfer reactions) and when and where protein and DNA free radicals are formed . After the preparation of the nitrone adducts, it is convenient, but not necessary, to eliminate excess reagent (i.e., DMPO, H2O2, and small molecular weight inhibitors) by dialysis, using a membrane with a cut-off appropriate for the molecular weight of the protein or DNA under analysis.
To analyze protein-centered radicals in biochemical systems, it is essential to control the protein purity in the preparation, which can vary from provider to provider and from batch to batch. For example, proteins and DNA can be contaminated with bound metals that, when reacted with H2O2, can produce extensive fragmentation by Fenton mechanisms [25, 42]. In this case, the biomolecule must be exhaustively dialyzed against chelexed buffers to eliminate the metals. Before use, metHb or metMb are passed through PD-10 desalting columns (Amersham Biosciences) to remove trace contaminants or dialyzed against 100 mM phosphate buffer, pH 7.4, using a 3.5 kDa cut-off dialysis cassette (Slide-A-Lyzer, Dialysis Cassette; cat. no. 66330, Pierce). Observation of a pure protein in silver- or Coomassie blue-stained gel is important to avoid misinterpretations.
Typically, in chemical systems, for example peroxidases or pseudoperoxidase/peroxide [24, 50] or Fenton systems (see Table 1), a concentration of DMPO between 5 and 100 mM is used, whereas in cell culture, depending on the type of cells, 10 to 100 mM for 24 h can be used without significant cytotoxicity. High concentrations of DMPO are needed in chemical systems to outcompete the radical decay mechanisms. In immunohisto/cytochemistry, the interference by free DMPO and small DMPO-molecule nitrone adducts is not a problem due to the fixation, permeabilization, and extensive washing carried out during the analysis before the exposure to the anti-DMPO antiserum.
When DMPO was added to cell culture with phenol red at a concentration between 50 and 100 mM, the color of the medium changed to the alkaline range (bright red). However, we have not found significant cytotoxicity in RAW 264.7 macrophages incubated for 24 h with up to 50 mM DMPO in culture medium with phenol red and 10 % fetal calf serum (FCS) . In addition, DMPO can trap superoxide radical anion , but very slowly, and therefore should not interfere with superoxide signaling or with cell development, proliferation, or physiology .
In cell experiments, it is imperative to include cells with and without treatment and with and without the spin trap. Antibiotics, normally used in cell culture, can affect redox processes, and, thus, must be omitted whenever possible . If antibiotics must be used, we recommend the inclusion of proper controls. We suggest a careful validation, on a case-by-case basis, of the spin trap concentration to be used in cell experiments with and without the treatment compound or condition. A series of controls has been previously published [19, 20].
In addition, for in vivo experiments, the inclusion of untreated and animals treated with injection of saline (vehicle) or DMPO alone is necessary to avoid misinterpretations. In any case, DMPO-specific immunoreactivity may be suppressed by competing DMPO and the nitrone adducts in the sample for the anti-DMPO antibody [22, 23]. The control experiments must be included to enhance the significance of the immuno-spin trapping experiments.
Although less sensitive than ELISA, Western blot is the only heterogeneous immunochemical technique that allows us to investigate and separate more than one protein-centered, radical-derived nitrone adducts in the same system . We can use ELISA to quantify and screen protein radical-derived nitrone adduct generation, but to study and identify any structural effect on the protein (e.g., aggregation or fragmentation), we must use Western blot analysis to detect and mass spectrometry analyses to identify the protein and the location of the DMPO in its primary structure. Although often omitted in the published literature, a Western blot of the whole homogenate should always be included. Once protein nitrone adducts are detected as positive bands in a Western blot, the corresponding bands in gels stained by Coomassie blue are cut, digested and analyzed by mass spectrometry to identify the protein(s) that form nitrone adducts. It should be noted that although MS is generally less sensitive than Westerns/ELISAs, it is a universal detector rather than a selective detector. Many gel bands that appear “pure” by Western blot analysis result in the detection of multiple proteins by mass spectrometry. Therefore, it is imperative to perform additional biochemical experiments to verify which protein is involved in the free radical processes. Additionally, it is quite difficult to determine the specific residue where DMPO is bound, i.e., where the radical has been trapped, especially from an in vivo system, due to the low abundance of the trapped protein [9, 52]. In order to avoid interference by DNA, which may aggregate proteins by forming a sticky net that modifies migration of proteins in the gel, we recommend treating the homogenate with benzonase (Novagen; cat. no. 70746-3) or similar DNAse enzymes to digest genomic DNA. Moreover, we add a cocktail of protease inhibitors (Protease Inhibitor Cocktail, Amresco; cat. no. M250-1ML) to the homogenization buffer to avoid the proteolysis of proteins by lyzosomal enzymes released during cell or tissue homogenization.
In animals, for example rats and mice, a single dose of 1 to 2 g of pure DMPO/kg body weight 2 h before the sacrifice and tissue removal for analysis of nitrone adducts is well tolerated. A vehicle such as saline or PBS is used as a control. Like antibiotics, some analgesics can affect the redox process. Thus, in order to avoid misinterpretations, the use of analgesics in animal experiments must include appropriate controls, for example DMPO, analgesic, analgesic/DMPO, treatment/DMPO and analgesic/treatment/DMPO.
Immuno-spin trapping is the only technique available to study protein and DNA radicals generated inside functioning cells or animals and to determine where they are located. This technology has quickly evolved as shown by the extent of recently published data (Table 1). See Scheme 2 for a rationale for using immuno-spin trapping tools (immunochemistry, mass spectrometry, and molecular MRI) in order to identify targets of oxidation that may be useful biomarkers of oxidative and inflammatory diseases. Immuno-spin trapping is a straightforward technique that requires proper controls and careful considerations, some of which we have previously described [19, 20] and highlighted in this method article. Some of the advantages of using immuno-spin trapping rather than ESR and ESR-spin trapping for the detection of protein- and DNA-centered radicals are: i) the feasibility of applying this technology in any research, clinic, or academic laboratory without requiring complicated equipment or specialists in physical chemistry and quantum mechanics as is needed for ESR; ii) the small quantity of sample required (μg), which is usually critical for exploring new systems; and iii) the possibility of detecting, characterizing, identifying, and localizing the distribution of biomolecule-centered radicals in multiple systems (biochemical, cell, tissue, and whole animal). See Table 1 for a complete list of references.
Due to their role in a number of biological processes and conditions, ROS and drug-induced macromolecule-centered radicals can now be detected and identified using immuno-spin trapping. Upon detection of the target(s) of oxidation by immuno-spin trapping, the identity of the protein radical and site of radical location are identified using the power of mass spectrometry techniques. This knowledge can be important in elucidating the molecular mechanisms of damage and functional consequences. We can also study the subcellular localization of macromolecule-centered radicals and infer biological consequences. In vivo immuno-spin trapping is still in development and may be a powerful tool to analyze the anatomical distribution of protein radicals in the living whole animal . Finally, immuno-spin trapping is a powerful technique for studying oxidative modification of biomolecules by reactive chemical species and their role in post-translational modifications in physiology and diseases.
The project described was supported by Award Number R00ES015415 from the National Institute of Environmental Health Sciences and, in part, by the Intramural Research Program of the NIEHS/NIH. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Environmental Health Sciences or the National Institutes of Health. DCR also acknowledges the start-up grant from the Presbyterian Health Foundation (PHF) to OMRF. The authors acknowledge Dr. Kalina Ranguelova, Dr. Hugo Cerecetto, Dr. Luke Szweda and Dr. Michael Kinter for critical review and to M.S. Quentin Pye, Dr. Ann Motten, and Ms. Mary J. Mason for helping in the editing, of this manuscript.
I. CONFLICT OF INTEREST
The authors state no conflict of interest.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.