UNIT 17.15 High-Throughput, Multiplexed Analysis of 3-Nitrotyrosine in Individual Proteins

 

Introduction

Protein modifications are a key factor in cell signaling and toxicity. Classic methods of protein detection, such as western blotting, are time-consuming and only semi-quantitative. Enzyme-linked immunosorbent assays (ELISAs) are higher-throughput and potentially quantitative. Unfortunately, traditional microplate-based ELISAs are optimized for single-antigen detection, and, hence, even with 384-well plates, high-throughput screening of multipy modified proteins is impractical.

Detection of modified proteins (i.e., phosphorylated or halogenated species) is problematic because of a shortage of good antibodies, requiring antibody validation for specificity for the targeted protein modification (e.g., 3-nitrotyrosine) versus a structurally similar modification (e.g., 3-chlorotyrosine). In addition, the relative scarcity of these modified proteins increases the difficulty of their detection. Sandwich ELISAs overcome many of these roadblocks by concentrating the target protein (modified or unmodified) with a single adsorbed antibody and using a second antibody to recognize only the target epitope. Recent advances in detection methods have also greatly enhanced sensitivity (Woodbury et al., 2002). The reduction in assay size also reduces the amount of sample required for analysis and increases the number of identical microarray chips that can be printed onto a single 25 × 75–mm slide, allowing for increased throughput. The combination of increased throughput and decreased sample size creates the potential for the multiplexed analysis of large archives of human samples, but creates the need for bioinformatics tools to efficiently process of the large quantities of data that are generated. We therefore developed a suite of software for ELISA microarray data processing, as well as data management and quality control. We recently utilized this ELISA microarray platform to analyze nitrotyrosine levels in 24 proteins in 458 human samples. All samples were analyzed in triplicate, such that this single study undertook over 30,000 ELISA analyses (Jin et al., 2011). This study provided an unprecedented examination of nitrotyrosine levels in smokers and subjects with inflammatory lung disease.

The basic sandwich ELISA analysis is a widely used assay system for both clinical and research purposes. In this protocol, we describe our adaptation of the basic sandwich ELISA protocol into a multiplexed system that allows for rapid and quantitative screening of multiple protein targets. The combination of a sandwich ELISA approach with a microarray platform makes this a suitable approach for measuring both modified and unmodified proteins in small sample volumes. Similar to DNA microarrays, a fluorescent probe is used to generate the final signal, providing an additional benefit of wide dynamic range. We describe the use of this ELISA microarray platform to detect proteins that are modified with 3-nitrotyrosine (nTyr) (Fig. 17.15.1).

Figure 17.15.1 Comparison of 3-nitrotyrosine ELISA microarray procedure with typical multiplex sandwich ELISA microarray analysis. In the standard, multiplexed ELISA microarray analysis, the various detection antibodies are combined into a single solution (left column). In the 3-nitrotyrosine (PTM) ELISA microarray, a single detection antibody for 3-nitrotyrosine is used (right column). Based on the fluorescence intensity in the array spots, the relative amount of 3-nitrotyrosine modifications is estimated.

Basic Protocol 1 outlines the process of slide manufacture, sample incubation, and data acquisition. Basic Protocol 2 describes the production of nTyr standards that can be used as positive controls, and for generation of standard curves. Basic Protocol 3 details a high-throughput method for screening antibodies targeted against specific protein modifications—nTyr in this protocol—since we have observed that commercial monoclonal antibodies do not always demonstrate their advertised degree of specificity towards the target modification versus other possible modifications or even the unmodified protein. Finally, Basic Protocol 4 illustrates a strategy for coping with situations where the target modified protein concentrations across different samples differ by several orders of magnitude. Such situations are not uncommon with human samples, and a logical means to efficiently and systematically dilute samples to bring signals within the assay's dynamic range is required.

NOTE: Due to the complexity of this protocol, the authors assume that end users have a basic understanding of microarray printer setup, experiment design, dynamic range validation, sample preparation, and data analysis.

 

Basic Protocol 1: Production of ELISA Microarray Slides and Their Use in Assay for 3-Nitrotyrosine

In this section, we detail the production of slides divided in 16 identical wells, with each well containing one enzyme-linked immunosorbent assay (ELISA) microarray chip (Fig. 17.15.2). These slides are subsequently used for the detection of 3-nitrotyrosine modifications of specific proteins. The entire process of slide manufacture, sample preparation, detection, and data analysis requires 3 days of work, although the first day (slide manufacture) can be completed far in advance.

Figure 17.15.2 ELISA microarray slide and chip format. The arrows and circles identify four replicate assays that, like all assays, are printed once in each quadrant in each well of the 16-well glass slide.

 Materials

• Capture antibodies (Table 17.15.1)
• Phosphate-buffered saline (PBS; Fisher Scientific)
• Alexa Fluor 546–conjugated goat anti—rabbit IgG (Invitrogen, cat. no. A-11035),
• Nonimmune (negative control) rabbit IgG
• 2% and 0.1% (w/v) bovine serum albumin (BSA, IgG-free, Jackson ImmunoResearch Laboratory, catalog number 001-000-162) in PBS
• PBS-T: 0.05% (v/v) Tween-20 (Sigma, cat. no. P7949) in PBS
• Desiccant
• Plasma samples for analysis
• Green fluorescent protein (GFP; Prospecbio, cat. no. PRO-687; http://www.prospecbio.com/)
• Biotinylated anti-3-nitrotyrosine monoclonal antibody (Hycult Biotechnology, cat. no. HM5002; http://www.hycultbiotech.com)
• Biotinylated anti–green fluorescent protein (GFP) antibody (Rockland Immunochemicals, Inc, cat. no. 600-401-215)
• Streptavidin-conjugated horseradish peroxidase (HRP) (Jackson ImmunoResearch Laboratory, catalog number 016-030-084)
• 1 µg/ml biotinylated tyramide (PerkinElmer)
• 100 mM sodium borate buffer, pH 8.5 (see recipe)
• Hydrogen peroxide (H₂O₂; Sigma, cat. no. H0904)
• Streptavidin-Cy3 (Jackson ImmunoResearch Laboratory, cat. no. 016-160-084)

• Microarray printer (GeSiM Nanoplotter, http://www.gesim.de/)
• 384-well microtiter plates
• Polylysine-coated glass slides (Fisher Scientific, cat. no. 22-037-216), separated into 16 separate but identical wells
• Super Pap Pen (Beckman Coulter; IM3580)
• Airtight moisture chamber associated with Microarray Printer (GeSiM Nanoplotter, http://www.gesim.de/)
• Dissecting microscope (optional)
• Laser scanner: ScanArray Express HT laser scanner and data acquisition software (Perkin-Elmer)
• Slide rack (Pacific Southwest Lab Equipment Inc., cat. no. 4465; http://www.psl-equip.com/)
• Centrifuge (IEC Cetra-MP4R with CAT-244 slide adapter)
• Vacuum desiccator and vacuum source
• Vacuum sealer and bags
• 96-well assay plates (BD Falcon, cat. no. 353075)
• Tightly sealable moisture container and lid (Pacific Southwest Lab Equipment Inc., http://www.psl-equip.com/)
• Oscillating mixer (Stovall Life Science, Inc., http://www.slscience.com/)

   

  Table 17.15.1 Capture Antibodies for nTyr PTM ELISA Microarray^a

  ---

  Capture antibody	Abbreviation	Catalog no.
  Alpha-lactalbumin	ALB	sc-53151^b
  Angiotensinogen	AGT	MAB3156^c
  Aminopetidase N	APN	AF3815^c
  Amphiregulin	AmR	MAB262^c
  CD14	CD14	MAB3833^c
  Clusterin	Clu	AF2747^c
  Ceruloplasmin	Cp	sc-69767^b
  C-reactive protein	CRP	MAB17071^c
  Epidermal growth factor (EGF)	EGF	DY236 kit^c
  EGF receptor (extracellular domain)	EGFR	AF-231^c
  E-selectin	Esel	AF-724^c
  Basic fibroblast growth factor	FGFb	MAB233^c
  Fibrinogen	Fibr	ID6-250310^d
  Heparin-binding epidermal growth factor	HBEGF	AF-292^c
  Hepatocyte growth factor	HGF	MAB694^c
  Intercellular adhesion molecular 1	ICAM	MAB720^c
  Insulin-like growth factor 1	IGF-1	MAB291^c
  Lactoferrin	LFn	sc-14431^b
  Leptin	Leptin	MAB398^c
  Galectin-3BP/MAC-2BP	MAP-2BP	AF2226^c
  Matrix metalloprotease 1	MMP1	AF901^c
  Matrix metalloprotease 2	MMP2	AF902^c
  Matrix metalloprotease 9	MMP9	AF911^c
  Osteopontin	OPN	MAB14332^c
  Platelet-derived growth factor AA	PDGF	MAB221^c
  Polymeric immunoglobulin receptor	pIgR	AF2717^c
  RANTES	RANTES	MAB678^c
  Lung surfactant protein A	SP-A	LS-C17957^e
  Transforming growth factor alpha	TGFa	AF-239^c
  Tumor necrosis factor alpha	TNFa	MAB610^c
  u-Plasminogen activator/urokinase	uPR	MAB1310^c
  Vascular endothelial growth factor	VEGF	AF-293^c

  ---

   ^aParts of this table have been previously published. These parts are reproduced with permission from Journal of Proteome Research (Gonzalez et al., 2008a) and Environmental Health Perspectives (Jin et al., 2011).

   ^bSanta Cruz Biotechnology, Inc.

   ^cR&D Systems.

   ^dABBiotec (http://www.abbiotec.com/).

   ^eLifespan Biosciences (http://www.lsbio.com/).

 First day: Manufacture chips

The selection of the antibodies, and thus the targeted antigens, will depend on the focus of the study. Care must be taken to ensure that the antibodies will work as capture reagents in ELISA microarrays. Generally, antibodies that work well for immunoprecipitation will also work well as capture antibodies in sandwich ELISAs. However, antibodies that only work well under conditions where the proteins are denatured, such as immunoblot or immunohistochemistry, are commonly not suitable capture reagents.

1.	Dissolve (or dialyze) commercial individual capture antibodies in phosphate-buffered saline (PBS). p type = annotation The final concentration of the capture antibodies (in PBS) should be approximately 0.5 to 1.0 µg/µl. p type = annotation IMPORTANT NOTE: Antibody selection will depend on the particular experiment (for examples, see Gonzalez et al., 2008a; Jin et al., 2011). Regarding the volume for printing, it varies considerably depending on the printer, but for the GeSiM Nanoplotter printer that we use, 12 µl of capture antibody is enough to produce two or three print jobs of up to 36 slides per printing (with 16 chips per slide, or 576 chips per print job).
2.	Set up a microarray printer. p type = annotation We use a piezoelectric, noncontact, GeSiM Nanoplotter to print our slides, as previously described (Zangar et al., 2006; Gonzalez et al., 2008a; Jin and Zangar, 2010).
3.	Start with 12 µl of each capture antibody (0.5 to 1 µg/µl) placed in a well of a 384-well microtiter plate. p type = annotation Capture antibodies, along with orientation spot A546 (0.1 mg/ml, Alexa Fluor 546 labeled goat anti-rabbit IgG), PBS (buffer control or blank), non-immune rabbit IgG (negative control for non-specific protein binding), and GFP antibody (used for a calibrant assay) (Daly et al., 2010) are printed on polylysine-coated glass slides that are preprinted with hydrophobic barriers to create 16 separate, but identical, wells on each slide (Gonzalez et al., 2008b). p type = annotation For the preprinted polylysine slides, preprinted barriers should be reinforced using a Super Pap Pen by drawing directly on top of the printed lines. One coat is usually sufficient.
4.	Print each capture reagent in quadruplicate on each chip, once in each of the four identical quadrants on every chip (Fig. 17.15.1). p type = annotation In the example shown in Figure 17.15.1, there are total of 36 × 4 = 144 spots per chip, and with the inclusion of 4 × 4 = 16 control spots, there are 32 × 4 = 128 spots for the antibodies used in the assays. This number of assays can be increased substantially, as there is sufficient room for many more spots.
5.	After printing, leave the slides in the printer at 60% humidity and room temperature for at least 3 hr to allow the proteins to adsorb onto the slide surface. p type = annotation Although it is possible to covalently cross-link the capture reagents to the slide surface, we have systematically tested this approach and found that it does not improve assay characteristics (Seurynck-Servoss et al., 2007).
6.	Prior to blocking, visualize the printed antibodies with PBS as buffer under a dissecting microscope, or more systematically verify by reflection scanning (excitation 633 nm and emission 633 nm, sometimes referred to as “RedReflect” scanning) using the ScanArray Express HT laser scanner.
7.	To block the slides, first place them in a rack that only touches the edges of the slides, and therefore does not damage the printed spots. Without any prior washing steps, block the slides by dipping the rack in 2% BSA in PBS. Complete the blocking by incubating the slides for 2 hr at room temperature or overnight at 4°C. p type = annotation The dipping of the slides should be done at a steady, moderate speed. If done too fast, the printed spots will streak, creating “comets” (Gonzalez et al., 2008a).
8.	Wash the slides by submerging the rack in PBS-T three times for 2 min each. Dry the slides by first centrifuging 30 sec at 500 × g (1700 rpm in IEC Cetra-MP4R with CAT-244 slide adapter) to remove free liquid from the slide, and then placing under an ∼20 mm Hg vacuum for 10 min to remove any residual water.
9.	Immediately use the dry slides or vacuum pack with desiccant and store at −20°C. p type = annotation Our experience is that antibody microarray chips are stable for 6 months or more when stored under these conditions.

 Second day: Process samples and initiate microarray analysis

10.	Allow frozen slides to warm to room temperature before opening vacuum-packed containers.
11.	Centrifuge all plasma samples 30 min at 12000 × g, 4°C, to remove any precipitants.
12.	Dilute the plasma samples 5-fold with 0.1% BSA/PBS. Spike a final concentration of 100 pg/ml GFP into each sample for evaluation of data quality and, if needed, normalization (Daly et al., 2010). p type = annotation To facilitate sample organization, the diluted samples are typically prepared in 96-well plates in the same order in which they will be analyzed. The use of 96-well plates also allows for the sample to be efficiently transferred to the microarray slides, thereby minimizing differences in incubation start times across samples. p type = annotation A 5-fold dilution has worked for human plasma and a similar dilution has worked for mouse plasma and presumably would work for other species. However, this would need to be determined empirically for each additional sample type or species (see Basic Protocol 4).
13.	Prior to loading, lay the microarray slides flat in sealable moisture container(s) with damp paper towels on the bottom and a porous plastic barrier between the towels and the slides. Using a multichannel pipettor, transfer 15 µl of each diluted sample from a 96-well microtiter plate (see step 13), which contains the samples in the desired order, to the ELISA microarray slides (see step 10). p type = annotation Ideally, the samples are organized in a randomized block design to minimize the possibility of any systematic bias entering the analysis due to variations in printing or process. Samples are commonly analyzed on three or more identical chips, but the replicate chips are typically located on different positions on different slides to minimize the possibility of any slide or positional bias.
14.	Once the samples are placed on the chips, be sure the whole chip is covered with sample. Seal the moisture containers immediately after the samples are placed on the slides, thereby ensuring a chamber with saturated humidity throughout the sample incubation period. p type = annotation Once the samples are placed on the slides, do not let the slides dry until processing is complete, as premature drying will increase background signal. We recommend using room humidifiers to increase humidity throughout the slide processing steps, especially during periods of low ambient humidity.
15.	After loading the samples, put the slides on a rotating or oscillating mixer with very slow movement (we use a setting of 0.5 or 1) and, typically, incubate for 16 hr (overnight) at room temperature. p type = annotation Higher rates of mixing do not improve results, but the increased agitation may lead to sample mixing between wells, and is generally associated with a loss of specific signal (unpub. observ.), possibly due to fluid shear effects that disrupt protein bonds.

 Third day: Complete microarray processing

 16.

Prior to washing the slides, gently remove the samples from the chips by aspiration with a house vacuum line fitted with a single pipet tip placed at the end of the line to reduce flow.

p type = annotation

The aspiration rate can be increased by removing the tip end with a clean razor blade. The aspiration tip is carefully slid along the edge of the slide to remove the bulk of the sample solution from each well, but care is taken not to touch the chip surface. That is, aspiration should be done from the edges of the wells to prevent physical damage to the printed reagents. To prevent sample mixing at the first wash step while still ensuring that the chips do not dry out, samples are quickly aspirated from one slide at a time, and each slide is immediately placed in a slide rack that is submerged in PBS-T.

 17.

Leave the rack in the PBS-T for 2 min after the last slide is added. Repeat 2-min wash steps twice more in fresh PBS-T. Only soak, do not mix during any wash steps. After washing, to minimize excess liquid on the slides so that the detection antibody is not randomly diluted by residual wash buffer (but preventing drying), quickly aspirate off excess wash buffer as described in step 11.

 18.

Load a mixture of biotinylated anti-3-nitrotyrosine monoclonal antibody diluted to 0.1 mg/ml in 0.1% BSA/PBS and 25 ng/ml biotinylated anti-GFP, for calibration assay (Daly et al., 2010), onto the array slides at 15 µl/chip.

 19.

Incubate the slides with the detection antibodies at room temperature for 2 hr with very gentle shaking on an oscillating mixer.

 20.

Soak the slides three times in PBS-T, each time for 2 min.

 21.

Incubate each array with 30 µl streptavidin-conjugated HRP diluted to a final concentration of 1 µg/ml in PBS-T for 30 min.

 22.

Soak in PBS-T three times, each time for 2 min.

p type = annotation

It is not necessary to aspirate off the excess reagent at this or subsequent steps.

 23.

Enhance the signal with the biotinylated tyramide as described in Kodama et al. (2004) and Daly et al. (2010). To do this, dilute biotinyltyramide to a final concentration of 1 µg/ml in 100 mM borate buffer (pH 8.5) containing 0.001% H₂O₂. Put the slides on an oscillating mixer with moderate shaking (we use a setting of 2), and (typically) incubate for 10 min.

p type = annotation

The signal amplification steps (steps 21 to 24) may increase the signal-to-noise ratio by as much as 100- to 1000-fold (Woodbury et al., 2002), but may not be required when analyzing relatively high levels of antigen.

 24.

Soak in PBS-T three times, each time for 2 min.

 25.

Incubate each array with 20 µl of streptavidin-Cy3 (diluted to 1 µg/ml in PBS-T buffer) for 45 min in dark and incubated with gentle mixing on an oscillating mixer.

 26.

Soak in PBS-T three times, each time for 2 min.

 27.

Rinse the slides by quickly dipping in pure water twice. Centrifuge slides to remove excess water, and vacuum dry for 10 min under ∼20 mm Hg.

p type = annotation

At this point, slides can be stored in the dark at −20°C, sealed under vacuum and with desiccant for long periods of time, if desired. A very modest loss of fluorescent signal can be observed after months of storage, but relative signal intensities are unchanged and quantitative results (when using an independent standard curve) are not significantly changed.

 28.

Scan slides with a laser scanner.

p type = annotation

The slides are imaged using appropriate excitation and emission wavelengths for the fluorophore used, with the scanner settings (photomultiplier tube gain and laser power) set to produce signal in the usable range of the scanner. We use ScanArray Express software to quantify the spot fluorescence intensity from the scanned images. Typical ELISA microarray images are shown in Figures 17.15.2 and 17.15.3.

Figure 17.15.3 Evaluation of 3-nitrotyrosine antibodies for assay sensitivity and specificity. (A) Diagram of antigens printed on the slides. (B, C, and D) Three different antibodies reacted with printed antigens. The antibody in panel D is selected for the best specificity and sensitivity to 3-nitrotyrosine modification. Key: KLH: keyhole limpet hemocyanin; OVA: Ovalbumin; NTO: 3-nitro-4-hydroxybenzoic acid–labeled OVA. This antigen mimics 3-nitrotyrosine–modified OVA; BTK: 3-bromo-4-hydroxybenzoic acid–labeled KLH. This antigen mimics 3-bromotyrosine modified KLH; BTO: 3-bromo-4-hydroxybenzoic acid–labeled OVA. This antigen mimics 3-bromotyrosine modified OVA; CTO: 3-choloro-4-hydroxybenzoic acid–labeled OVA. This antigen mimics 3-chlorotyrosine modified OVA. BSA-nTyr: peroxynitrite-treated BSA. This antigen is a positive control for 3-nitrotyrosine; BSA-BrO: sodium hypobromite–treated BSA.

 29.

For the general microarray data analysis, refer to previous publications on this topic (Zangar et al., 2006; Daly et al., 2008, 2010; Jin and Zangar, 2009; White et al., 2011).

p type = annotation

For 3-nitrotyrosine data analysis, the fluorescence intensity of each spot in the laser-scanned image is quantified by ScanArray Express software or a similar, commercial program. The data are evaluated and, if needed, calibrated using the data from the internal control ELISA for GFP and the program ProMAT Calibrator (Daly et al., 2010; Zangar et al., 2009). Diagnostic images from this program are used to determine if data normalization is likely to be beneficial. The data are then processed with ProMAT (White et al., 2006), which quickly organizes and summarizes the data based on replicates at the assay and sample level. If independent standards (i.e., purified antigens of known concentrations) are analyzed, this program will also automatically generate standard curves and calculate the sample concentration values based on the fluorescent signal for each assay. The ProMAT bioinformatics programs can be obtained free of charge by following the directions at http://www.pnl.gov/statistics/ProMAT/.

 

Basic Protocol 2: Generation of 3-Nitrotyrosine Standards by Modification of BSA by Peroxynitrite

This protocol describes the production of nitrated BSA for testing antibody reactivity and specificity. Similar protocols can be used to generate 3-bromotyrosine and 3-chlorotyrosine, except for differences in the chemicals used to modify the protein (Kato et al., 2005; van Dalen et al., 2009).

 Materials

• Fatty acid-free BSA (Sigma, cat. no. A8860)
• Phosphate-buffered saline (PBS; Fisher Scientific)
• Peroxynitrite stock solution (Millipore, cat. no. 20-107)
• 10 M NaOH
• 12 M HCl

• NanoDrop UV-Vis ND-2000 Spectrophotometer (Thermo Scientific)
• 12- to 14-kDa MWCO dialysis tubing (Spectrum Laboratories, Inc., cat. no. 132678)

• Additional reagents and equipment for protein nitration (Beckman et al., 1994)

 Perform protein nitration

Protein nitration is accomplished by mixing fatty-acid-free BSA and peroxynitrite, as described in Beckman et al. (1994), with minor modifications.

1.	Completely dissolve 5 mg of fatty-acid-free BSA 1 ml PBS containing 50 mM NaHCO3.
2.	Dilute peroxynitrite stock (use within a week of production by the vendor) to 20 mM. p type = annotation Before use, the stock concentration of peroxynitrite is determined by absorbance at 302 nm in a solution of 0.3 M NaOH and using an extinction coefficient 1670 M^−1 cm^−1 (Beckman et al., 1994).
3.	Add 100 µl of 20 mM peroxynitrite onto the side of the tube of the BSA solution right above the liquid, and vortex immediately. After vortexing, simply place the BSA on the bench and let sit for 5 min, at which time the reaction is complete.

 Quantify nitration and remove unreacted chemicals

4.	To quantify the amount of nitration, make an aliquot of the reaction protein alkaline with the addition of ≤20 µl of 10 M NaOH and a second aliquot acidic by addition of ≤20 µl of 12 M HCl and quickly shaking, which prevents the formation of a precipitate. Scan the protein solutions from 300 nm to 500 nm. p type = annotation The absorbance at 430 nm at acidic pH is subtracted from the absorbance at 430 nm at alkaline pH, and the concentration of nitrotyrosine is calculated by using the extinction coefficient of 4400 M^−1 cm^−1. PBS is used a blank solution for the spectrophotometer before scanning. In our experience, a 10-fold molar excess of peroxynitrite to BSA produced the highest 3-nitrotyrosine signal in our ELISA assay.
5.	Transfer the modified BSA into MWCO 12- to 14-kDa dialysis tubing to remove unreacted chemicals. Dialyze against three changes of PBS, each of 12-hr duration at 4°C.
6.	After dialysis, aliquot the nTyr-BSA into multiple tubes, and store at −80°C. p type = annotation The nitrated BSA is stable for years, and can be used to serve as a positive control for evaluation of the sensitivity and specificity of antibody and chips as a test protocol. The nitrated BSA is used in Basic Protocol 3, which tests antibody reactivity and specificity.

 

Basic Protocol 3: 3-Nitrotyrosine Antibody Evaluation

We have found that commercial antibodies against 3-nitrotyrosine do not always perform as advertised, and that they need to be tested prior to use for both the desired binding activity and for absence of cross-reactivity against proteins with related modifications (i.e., 3-chlorotyrosine, 3-bromotyrosine) (Jin et al., 2011). Some commercial antibodies react with the unmodified antigen but show little, if any, reactivity against the same protein post-modification (nitration of tyrosine, in this case).

Overall, this protocol combines a classic sandwich ELISA method (capture of the target antigen(s) onto the substrate surface followed by addition of a detection antibody), but leverages the throughput capabilities and minimal sample and reagent usage afforded by the microarray platform.

 Materials

• Antigens needed to print and test (see Kato et al., 2005, and van Dalen et al., 2009, for modification procedures; the only difference is the use of 4-hydroxy-3-nitrobenzaldehyde to modify BSA):
  • KLH: keyhole limpet hemocyanin (Pierce, cat. no 77600): this antigen is a control for unmodified protein
  • OVA: ovalbumin (Pierce, cat. no. 77120): this antigen is a control for unmodified protein
  • NTO: 4-hydroxy-3-nitrobenzaldehyde (Sigma-Aldrich, cat. no. 55971)–labeled OVA: this antigen mimics 3-nitrotyrosine modified OVA
  • BTK: 3-bromo-4-hydroxybenzoic acid (Indofine Chemical Company, Inc., cat. no. 19-155)–labeled KLH: this antigen mimics 3-bromotyrosine-modified KLH
  • BTO: 3-bromo-4-hydroxybenzoic acid (Indofine Chemical Company, Inc., cat. no. 19-155)–labeled OVA: this antigen mimics 3-bromotyrosine modified OVA
  • CTO: 3-chloro-4-hydroxybenzoic acid (Sigma-Aldrich, cat. no. C44605)–labeled OVA: this antigen mimics 3-chlorotyrosine–modified OVA
  • BSA-nTyr: peroxynitritetreated BSA: this antigen is a positive control for 3-nitrotyrosine
  • BSA-BrO: sodium hypobromite (Fisher Scientific, ca. no. NC9754116)–treated BSA: this antigen is a control for bromated protein
• Phosphate-buffered saline (PBS; Fisher Scientific)
• PBS-T: 0.05% (v/v) Tween-20 (Sigma, cat. no. P7949) in PBS
• 1% (w/v) bovine serum albumin (BSA) in PBS-T
• 3-nitrotyrosine antibodies used for detection antibodies:
  • Antibody 1: goat anti-nTyr, polyclonal (Abcam, cat. no. Ab27648)
  • Antibody 2: goat anti-nTyr, polyclonal (Meridian Life Science, cat. no. K97520G; http://meridianlifescience.com/)
  • Antibody 3: rabbit anti-nTyr, polyclonal (Molecular Probes, cat. no. A-21285)
  • Antibody 4: mouse monoclonal antibody to nTyr (monoclonal strain HM11; Thermo Scientific, cat. no. MA1-35729 or Hycult Biotechnology, cat. no. HM5001; http://www.hycultbiotech.com/)
• Secondary antibodies (depending on source of primary antibody):
  • Mouse anti-rabbit IgG (Jackson ImmunoResearch Laboratory, cat. no. 211-035-109)
  • HRP-conjugated rabbit anti-goat IgG (Jackson ImmunoResearch Laboratory, cat. no. 305-035-003)
  • HRP-conjugated goat anti-mouse IgG (Jackson ImmunoResearch Laboratory, cat. no. 115-035-003)

• Additional reagents and equipment for printing array slides (Basic Protocol 1)

 Print and block slides

The nitrated, chlorinated, and brominated antigens, along with a calibrant (GFP), orientation spot A546, and vehicle control (PBS) are printed to the slides. Details on generation of BSA-nTyr are described in Basic Protocol 2, where references are provided for generation of chlorinated and brominated BSA antigens as well. Additional information on these antigens is included below.

1.	Dilute the protein antigens to 0.5 mg/ml in PBS and print onto polylysine-coated slides. For details of printing protocol, see Basic Protocol 1. p type = annotation This printing step is basically the same as one would use for the capture antibodies.
2.	Block the printed slides in 1% BSA in PBS-T overnight at 4°C.
3.	Wash the slides three times, each time for 2 min with PBS-T.

 Treat with primary antibody

4.

Add 20 µl of a single, diluted 3-nitrotyrosine antibody. Incubate the slides with the antibody at room temperature for 2 to 16 hr. p type = annotation In general, detection antibody concentrations from 0.1 to 1 µg/ml are optimal for detection antibodies. We have found it is best to empirically test a range of detection antibody concentrations in order to determine which dilution produces the best signal-to-noise ratio (Gonzalez et al., 2008a).

5.

Wash the slides three times, each time for 2 min with PBS-T.

 Treat with secondary antibody

6.

Depending on the source of the primary antibody used in step 4, choose the appropriate secondary antibody. Incubate slides with secondary antibody diluted to 1 µg/ml in PBS-T for 2 hr at room temperature.

7.

Wash the slides three times in PBS-T as described in step 5. Perform steps 20 to 29 of Basic Protocol 1 on the third day. p type = annotation Representative results of different 3-nitrotyrosine antibodies obtained with our ELISA microarray chips are shown in Figure 17.15.3. The monoclonal antibody (Hycult Biotechnology) was selected for our 3-nitrotyrosine ELISA microarray because it preferentially reacts with nitrated BSA and the nitrotyrosine mimic, NTO, while other antibodies show cross reactivity with other modified proteins, including bromotyrosine or chlorotyrosine.

 

Basic Protocol 4: Identifying a Useful Dilution of Human Plasma for Measuring 3-Nitrotyrosine

The measurement of nitrotyrosine levels in specific proteins can potentially be influenced by varying concentrations of the targeted antigen. This is a common problem in human studies, as antigen concentrations commonly vary 100-fold or more between individuals. To prevent this potentially confounding effect, we try to saturate our antibodies with antigen whenever possible. Under these conditions, the amount of antigen captured is dependent on the amount of antibody printed on the spot, which should be a constant; the nitrotyrosine signal will therefore be proportional to a constant amount of captured antigen. However, the examination of dilutions above the saturation point is necessary to demonstrate that the signal is maximal and therefore saturated. In this section, we describe a simple protocol for determining a sample dilution that will result in most of the antigen concentrations being essentially saturating. Once a usable dilution is identified, only that dilution is used for all samples in the subsequent study.

 Materials

• Frozen plasma samples

• Additional reagents and equipment for 3-nitrotyrosine microarray ELISA (Basic Protocol 1)

 1.

Thaw frozen plasma samples on ice. Centrifuge 20 min at 12000 × g, 4°C.

p type = annotation

The supernatant is used for analysis.

 2.

Dilute plasma samples 2-, 5-, 10-, 20-, 100-, 1000-, 5000-, and 25000-fold in 0.1% BSA/PBS.

 3.

Analyze the diluted samples by microarray ELISA as described in Basic Protocol 1.

p type = annotation

The fluorescence intensity as a function of dilution for different samples and different capture antibodies is shown in Figure 17.15.4. The results suggest that when the plasma samples are diluted 2×, 5×, 10×, or 20×, the resulting fluorescence intensity signal is at approximately the same level regardless of the dilution for most of the assays, suggesting that most of the capture antibodies are saturated at a 20-fold dilution. We typically use a 5-fold dilution of plasma samples for our studies to ensure saturation of the capture antibodies.

Figure 17.15.4 Defining a useful dilution of human plasma for measuring 3-nitrotyrosine. ELISA microarray results for the dilution of human plasma. Four samples were diluted with 0.1% BSA/PBS at 2-, 5-, 10-, 20-, 100-, 1000-, 5000-, and 25000-fold. The fluorescence intensity of 3-nitrotyrosine from each diluted assay was plotted with folds of dilution (X-dimension label). The different capture antibodies are color coded in the inset key. The results suggest that the 3-nitrotyrosine level is saturated across all measured samples when plasma samples are 20-fold or less diluted.

sect1 type = reagents

 

Reagents and Solutions

p type = annotation

Use Milli-Q-purified water or equivalent in all recipes and protocol steps. For common stock solutions, see appendix 2A; for suppliers, see suppliers appendix.

 Borate buffer, 100 mM, pH 8.5

Dissolve 38.1 g sodium tetraborate decahydrate (Na₄B₂O₇ċ10H₂O; Sigma, cat. no. S-9640) in 900 ml Milli-Q-purified water. And 37% hydrochloric acid dropwise until pH is 8.5, and make up to 1000 ml with water. Filter the solution through 0.22-µm membrane, and store up to 6 months at room temperature.

sect1 type = commentary

 

Commentary

 

Background Information

Nitric oxide (NO) is an important cellular messenger that participates in paracrine signaling. NO rapidly reacts with superoxide radical to yield peroxynitrite (ONOO⁻), a strong oxidant that nonezymatically reacts with protein tyrosine residues to form nitrotyrosine (Beckman et al., 1994; Gow et al., 1998). The aromatic nitro (-NO₂) group is formed at position 3 of the tyrosine ring. There are also alternative ways that 3-nitrotyrosine can be formed, such as by eosinophil peroxidase or myeloperoxidase activity (Kodama et al., 2004; Mohiuddin et al., 2006).

Protein nitration is associated with many neurodegenerative and inflammatory diseases, including Parkinson's disease, Alzheimer's disease, transplant rejection, lung infection, central nervous system and ocular inflammation, shock, and cancer (Beckman et al., 1994; Ischiropoulos, 1998a,b; Peluffo and Radi, 2007). Nitration can increase or decrease protein activity, depending on the protein, and may contribute to disease pathogenesis (Pehar et al., 2004). Hence, there is a need to understand which proteins are nitrated during pathogenesis of these diseases.

The traditional sandwich ELISA uses a matched pair of capture and detection antibodies to measure each antigen (Fig. 17.15.2). For the multiplex microarray platform, the various capture antibodies are physically separated by printing in isolated spots. The ELISA microarray is an efficient tool for measuring the levels of up to ∼25 proteins with excellent sensitivity (single-digit pg/ml), dynamic range (∼ 3 orders of magnitude), and low coefficients of variation between replicate measurements (5% to 20%) in small volumes (20 µl or less of diluted sample) (Varnum et al., 2004; Zangar et al., 2004).

In this manuscript, we describe the application of our custom ELISA microarray platform for assessing trace 3-nitrotyosine levels of proteins in complex biological samples. We have found that this new application integrates well with the current sandwich ELISA microarray and is very powerful for analyzing protein modifications for trace amounts of 3-nitrotyrosine studies (Jin et al., 2011; Zangar et al., 2011). A previous study demonstrated its ability to measure nitration levels in 24 proteins in 458 plasma samples, with each analysis done in triplicate (Jin et al., 2011). This single, 24-hr experiment represents a total of over 30000 ELISAs (Jin et al., 2011). Thus, it is clear that the throughput capabilities of this platform are substantial.

 

Anticipated Results

The raw data from this nitrotyrosine analysis are raw fluorescent signal intensity (arbitrary units) for each spot. There are four spots per chip for each capture antibody, from which the ProMAT program automatically derives the median value for each assay from that chip. As each sample is typically analyzed on three replicate chips, the average of the signal intensity of these three replicates is derived from a standard spreadsheet and used as the single value for that sample. Thus, in contrast to the traditional ELISAs, which rely on a standard curve generated from analysis of a serial dilution of purified antigen(s) to calculate the antigen concentration in the samples, the nitrotyrosine assay only provides a semi-quantitative data (i.e., raw fluorescence) for each protein.

 

Troubleshooting

See Table 17.15.2 for problems, likely causes, and possible solutions for problems with the 3-nitrotyrosine ELISA.

 

Table 17.15.2 Common Problems with and Likely Causes of These Problems

---

Problems	Possible reasons	Solutions
No fluorescence signal with orientation spots	Problem with printing or scanner	Scan slides known to have a positive signal from previous analysis. If scanner appears to be working fine, test the printer using only fluorescently labeled proteins.
No fluorescence except the orientation spot	Missed steps or bad reagents	Repeat the complete procedure with freshly prepared reagents
No fluorescence of internal control but other spots okay	Missing either GFP antigen or detection antibody	The data can be analyzed without the internal calibrant assay, but data quality may suffer. It may be possible to normalize the data based on the signal from the orientation spot or the PBS-blank spot, depending on the source of systematic bias.
High background	a. Blocking reagent	a. Test different blocking reagent
	b. Samples dilution	b. Titrate different dilution of samples
No signal after a full complete ELISA procedure	Degraded streptavidin-HRP; degraded Biotinylated tyramide; degraded Cy3- streptavidin	All reagents should be stored in the dark. Reconstitute fresh solution and store in the dark.

---

 

Time Considerations

Setting up the printer may take up to 3 hr, depending on the number of capture antibodies and slides to be printed. Printing of the slides may take up to 12 hr, depending on the number of slides printed, but this task is can be done overnight and does not require an active researcher's time. A single slide from each print job should be tested prior to an experiment that consumes valuable samples, which means completing Basic Protocol 1 with a single, positive-control sample. Although incubations times extend over an approximately 24-hr period, actual active labor time is probably 3 hr.

 

Acknowledgments

This research was supported by the National Institute of Environmental Health Sciences (U54/ES016015) and a U.S. Department of Defense postdoctoral fellowship (W81XWH-10-1-0031 for H. Jin).