Guidelines for the Detection of NADPH Oxidases by Immunoblot and RT-qPCR

Abstract

The identification of NADPH oxidase (NOX) isoforms in tissues is essential for interpreting experiments and for next step decisions regarding cell lines, animal models, and targeted drug design. Two basic methods, immunoblotting and reverse transcriptase-quantitative polymerase chain reaction (RT-qPCR), are important to monitor NOX protein and messenger RNA (mRNA) levels, respectively, for a range of investigations from understanding cell signaling events to judging NOX inhibitor efficacies. For many other genes that are expressed in high abundance, these methods may seem rather simple. However, detecting the low expression levels of endogenous NOX/DUOX is difficult and can be frustrating, so some guidelines would be helpful to those who are facing difficulties. One reason why detection is so difficult is the limited availability of vetted NOX/DUOX antibodies. Many of the commercial antibodies do not perform well in our hands, and dependable antibodies, often generated by academic laboratories, are in limited supply. Another problem is the growing trend in the NOX literature to omit end-user validation of antibodies by not providing appropriate positive and negative controls. With regard to NOX mRNA levels, knockdown of NOX/DUOX has been reported in cell lines with very low endogenous expression (C_q values ≥30) or in cell lines devoid of the targeted NOX isoform (e.g., NOX4 expression in NCI-60 cancer cell panel cell line 786–0). These publications propagate misinformation and hinder progress in understanding NOX/DUOX function. This chapter provides overdue guidelines on how to validate a NOX antibody and provides general methodologies to prepare samples for optimal detection. It also includes validated methodology to perform RT-qPCR for the measurement of NOX mRNA levels, and we suggest that RT-qPCR should be performed prior to embarking on NOX protein detection.

1. Introduction

NADPH oxidases (NOX) are the major producers of purpose-driven ROS production. Unlike other enzymes that produce ROS as a side reaction during the course of catalysis, the NADPH oxidase enzymes use NADPH to reduce molecular oxygen to form superoxide and hydrogen peroxide as their sole known enzymatic function. NOX enzyme expression is complex and dependent upon factors such as species, tissue, cell type, and physiological or pathological status. Thus, the cataloguing of NOX expression has often been reported by specific tissue type. While immunoblotting and PCR are seemingly simple methods, NOX detection presents a special case requiring knowledge of certain precautions and preparations prior to embarking on these methods. Membrane proteins, in general, are more difficult to detect, and NOX proteins have adorned themselves with an “invisibility cloak” [a la Harry Potter] which may be due to their low abundance, glycosylation, resistance to standard extraction methods, susceptibility to proteolytic degradation, and perhaps a variety of other factors. Many manufacturers market “anti-NOX isoform antibodies,” but the majority of their validation data (and/or literature referenced in their product data sheets) are incomplete (e.g., lack of appropriate positive and negative controls, no information about isoform specificity). In our hands many of these antibodies fail to identify the claimed NOX isoform. Experts in the field are frustrated by the poor results obtained when using such antibodies. Astonishingly, a large number of manuscripts are accepted without proper antibody validation. Thus, guidelines are needed in order to properly evaluate immunoblot data to avoid misleading new investigators in this field who are searching for appropriate reagents.

A number of validated NOX antibodies have emerged from academic laboratories. Some of them are listed in Tables 5, 6, 7, 8, 9, 10, and 11, and additional ones are described in [1]. NOX and DUOX antibodies listed in Tables 5, 6, 7, 8, 9, 10, and 11 are not all-inclusive, and the reader is advised to judge other antibodies in the literature based on the methods of validation described in these guidelines. Exact dilutions for each antibody are not provided since it will vary depending on the characteristics of the sample.

Table 5.

NOX1 and NOXA1 antibodies

NOX1 source/catalog #	Host and clonality	Immunogen (aa)	Is the target detected in WB (other applications)?	Species detected (human, mouse, rat)	Reference
Dr. J. Doroshow	Mouse Monoclonal Clone 22	224–564	Yesa (IF, IHC)	H	[11, 12]
Dr. C. Yabe-Nishimura	Rabbit Polyclonal	139–152 (mouse)	Yesb (IF, IHC)	M	[7]
Dr. C. Yabe-Nishimura	Rat Monoclonal Clone 31C1	139–152 (mouse)	Yesb (IF, IHC)	M	Unpublished
Dr. J. David Lambeth	Mouse Monoclonal Clone E39	3rd extracellular loop	Yesc (IF, IHC)	H	[13]
OriGene Technologies #TA349080	Rabbit Polyclonal	Proprietary	Nod	H
Novus #NBP1–31546	Rabbit Polyclonal	Proprietary	Nod	H, M, R
NOXA1 subunit Santa Cruz sc-398873	Mouse Monoclonal Clone H-6	1–200	Yese	H, M, R

Evaluation of academic and commercial NOX1 antibodies

^aSpecificity in cell lines is shown in Ref. [11]; specific in CHO cells overexpressing NOX1 (Dr. V. Jaquet, pers. communication)

^bAuthors stated that the polyclonal antibody detected Nox1 in mouse colon but not in mouse small intestine or lung. Their mAb detects Nox1 in mouse colon, small intestine, and lung (personal communication Dr. M. Matsumoto) (see Fig. 8)

^cThere was a limited quantity of this antibody available for WB verification (see Fig. 5d). This antibody was made by diaDexus (out of business). The specific epitope was not disclosed, only the location (3rd extracellular loop)

^dUnable to differentiate between NOX1 stably overexpressed in HEK 293 cells and empty vector-transfected controls

^eProduct datasheet indicates that NOXA1 was detected in HEK 293T cells overexpressing NOXA1, but not in vector control

Table 6.

NOX2 antibodies

NOX2 source/atalog #	Host and Clonality	Immunogen (aa)	Is the target detected in WB (other applications)?	Species detected (human, mouse, rat)	Reference
Dr. A. Jesaitis Abcam #ab80897 Santa Cruz #sc-130543	Mouse Monoclonal Clone 54.1	383–390	Yesa (IF)	H, M	[14]
Dr. D. Roos, LSbio #LS-C85347	Mouse Monoclonal Mo 48	Purified cytochrome b558	Yes (IHC)	H	[17, 18]
Dr. A. Jesaitis Santa Cruz #sc-130549	Mouse Monoclonal Clone CL5	135–147	Yesc	H	[19]
BD Transduction Laboratories, #BD61141	Mouse Monoclonal Clone 53	450–556	Yesd (IF rat)	H, M, R	[2, 20]
LifeSpan Biosciences #LS-C179418	Mouse Monoclonal Clone 7D5	143–151	Yesb (Flow)	H [15, 16]

Evaluation of academic and commercial NOX2 antibodies

^aCross-reactivity with NOX1–4 and GRP58

^bCross-reactivity with gelsolin

^cIsoform specificity not tested. Other applications not tested

^dAccording to the manufacturer, a 58 kDa band is detected in mouse macrophages

Table 7.

NOX3 antibodies

NOX3 source/catalog #	Host and clonality	Immunogen (aa)	Is the target detected in WB (other applications)?	Species detected (human, mouse, rat)	Reference
Bioss bs-3683R	Rabbit Polyclonal	70–120	Yesa	H, M,R	Unpublished

At this time there is little information on NOX3 antibodies and their validation. Several NOX3 antibodies are on the market (see www.abcam.com and www.antibodies-online.com), but none of these have, thus far, been validated nor tested for isoform specificity

^aWorked well in detecting NOX3 overexpression in HEK 293 cells for one group but not for another (Dr. Ulla Knaus, Dr. Vincent Jaquet, pers. communication), and specificity over other isoforms has not been tested

Table 8.

NOX4 antibodies

NOX4 source/catalog #	Host and clonality	Immunogen (aa)	Is the target detected in WB (other applications)?	Species detected (human, mouse, rat)	Reference
Dr. A. Shah	Rabbit Polyclonal	556–568	Yesa (IF) (see Fig. 5a)	H, M	[21]
Dr. P. Jansen-Dürr	Rabbit Monoclonal (27–6)	335–578	Yesb (see Fig. 5f)	H, M	[22]
Dr. U. Knaus	Rabbit Polyclonal (7843)	251–266	Yesc (IF, Flow)	H	[23]
Dr. U. Knaus	Rabbit Monoclonal (20.1, 37.7)	251–266	Yesc (IF, Flow)	H	[23]
Dr. J. Doroshow	Rabbit Monoclonal (47–6)	242–262	Yesc (IHC) (see Fig. 5b)	H	[12]
Abcam ab133303	Rabbit Monoclonal	Unknown	Nod, e (see Fig. 5h)	H, R
Abcam ab109225	Rabbit Monoclonal	500–578	Nod,e	H, R
Millipore ABC459	Rabbit Polyclonal	C-terminus	Nod, f (see Fig. 5g)	H, M
Novus Biologicals #NB110–58851	Rabbit Polyclonal	500–578	Nod, g (see Fig. 5c)	H, M, R

Evaluation of academic and commercial NOX4 antibodies

^aExcellent sensitivity for mouse and human NOX4 although many bands are detected, some of which may be NOX4 variants; does not cross-react with NOX1 in HEK 293 cells stably overexpressing NOX1 (see Fig. 5a)

^bDetects human NOX4 in Trex cells overexpressing NOX4; detects Nox4 in mouse kidney cortex, multiple bands

^cNOX4-specific (epitope is absent in other NOX isoforms) but only detects human NOX4

^dDoes not distinguish between HEK293 cells expressing an empty vector vs. HEK293 cells stably overexpressing NOX4

^eIdentical results were obtained by two independent laboratories (Dr. J.D. Lambeth and Dr. J. Doroshow)

^fWebsite states that application is not for WB, but for IHC and ICC

^gUnspecific in mouse; older lots may have been specific for NOX4 in overexpressing cells, but recent lots are not specific

Table 9.

NOX5 antibodies

NOX5 source/catalog #	Host and Clonality	Immunogen (aa)	Is the target detected in WB (other applications)?	Species detected (rodents lack NOX5 gene)	Reference
Dr. W. Nauseef	Rabbit Polyclonal	1–169	Yesa (IF, IP)	H	[6, 24, 25]
Dr. J. D. Lambeth	Rabbit Polyclonal (#8579)	491–506 and 577–588	Yesb (IHC)	H	[4]
Dr. J. Doroshow	Mouse Monoclonal	600–746	Yesb (IF, IHC)	H	[26]

Evaluation of NOX5 antibodies

^aReference [24] shows a MW of 75 kDa in K562 cells stably transfected with NOX5β. Detects a 70 kDa band and a smaller band at 43kDa (degradation product) in HEK cells expressing NOX5. Antibody works for immunoprecipitation

^bReference [4] shows a MW of 75kDa in HEK 293 cells transiently transfected with NOX5

Table 10.

DUOX antibodies

DUOX1, DUOX2, DUOXA1, DUOXA2 source/catalog #	Host and clonality	Immunogen (aa)	Is the target detected in WB (other applications)?	Species detected (human, mouse, rat)	Reference
DUOX1 and 2 Dr. U. Knaus	Rabbit Polyclonal	DUOX1 775–1026	Yesa (Flow, IHC)	H, M	[27]
DUOX2 specific Dr. U. Knaus	Rabbit Polyclonal	634–648	Yes (Flow, IHC)	H	[28]
DUOXA1 specific Dr. U. Knaus	Rabbit Polyclonal	280–299	Yes	H	[28]
DUOX1 and 2 Dr. X. De Deken	Rabbit Polyclonal I2	DUOX1 618–1004	Yesa	H, M	[29]
DUOX1 and 2 Dr. C. Dupuy	Rabbit Polyclonal	DUOX2 410–423	Yesa	H, M	[30]
DUOX1 specific Dr. C. Dupuy	Rabbit Polyclonal	DUOX1 988–1011	Yes	H, M	[31]
DUOX1 and 2 Dr. C. Dupuy	Mouse Monoclonal Clone	DUOX2 21–589	Yesa	H	[32]
DUOX1 and 2 Dr. J. Doroshow	Mouse Monoclonal Clone S-40	DUOX2 131–540	Yesa	H, M	[33]
DUOX1 Bioss bs-11431R	Rabbit Polyclonal	Unspecified	Nob	H, M, R
DUOX2 Santa Cruz (Y-15) sc-49939	Goat Polyclonal	Unspecified	Yesa	H, M, R
DUOX2 Santa Cruz (E-8) sc-398681	Mouse Monoclonal	DUOX2 626–680	Not testedc	H, M, R
DUOX1 Sigma-Aldrich HPA023544	Polyclonal	910–1011	Yesa (IHC)	H
DUOXA1 specific Sigma-Aldrich HPA041578	Polyclonal	270–326	Yesd (IHC)	H

Evaluation of DUOX1, DUOX2, DUOXA1, and DUOXA2 antibodies

^aCross-reactivity between DUOX1 and DUOX2. Dr. Dupuy’s monoclonal antibody has ~tenfold preference for DUOX2, and in porcine tissues this antibody does not cross-react with Duox1 (see reference [32])

^bDetects a non-specific band at a similar molecular weight as DUOX in non-transfected cells

^cRecommended by Santa Cruz as a replacement for Santa Cruz (Y-15) sc-49939. See website for images

^dDetects additional non-specific bands

Table 11.

Antibodies for subunits (p47^phox, p67^phox) and for p22^phox

Subunits and p22phox source/catalog #	Host and Clonality	Immunogen (aa)	Is the target detected in WB (other applications)?	Species detected (human, mouse, rat)	Reference
p47phox Dr. M. Quinn Millipore 07–001	Rabbit Polyclonal	Full-length human p47phox	Yesa, b (IHC, IP)	H, M, R	[34]
p47phox Dr. M. Quinn	Mouse Monoclonal Clone 43.12	Full-length human p47phox	Yes (others not tested)	H, M, R	Unpublished
p67phox Dr. M. Quinn Millipore 07–002	Rabbit Polyclonal	Full-length human p67phox	Yesa, b (IHC, IP)	H, M, R	[34]
p67phox Dr. M. Quinn	Mouse Monoclonal Clone 81.1	Full-length human p67phox	Yes (IHC)	H, M, R	Unpublished
p47phox Santa Cruz sc-17844	Mouse Monoclonal Clone A-7	196–390	Yesa (others not tested)	H, M, R.
p67phox Santa Cruz sc-374510	Mouse Monoclonal Clone D-6	Full-length p67phox	Yesa (others not tested)	H, M, R
p40phox Dr. M. Quinn	Mouse Monoclonal Clone 1.9	Full-length human p40phox	Yes (IHC)	H
p22phox Dr. A. Jesaitis Dr. M. Quinn Santa Cruz sc-130550	Mouse Monoclonal Clone 44.1	29–33 and 182–188	Yesa, b (IP)	H, M, R	[14]

Many academic laboratories have published antibodies generated for these subunits. We have used antibodies generated in the laboratory of Dr. Mark Quinn

Also listed are commercial antibodies that worked in our conditions for immunoblots

^aTested with human neutrophil cytosolic fractions

^bTested in HEK 293 cells overexpressing cytosolic components

Given the lack of dependable commercial antibodies, many researchers have relied upon PCR as a surrogate for the detection of NOX expression. Yet even with the specificity of NOX primers, conflicting results can be obtained regarding the presence of specific NOX isoforms in identical tissues under basal conditions. Thus, this chapter also includes a section describing one method for RT-qPCR. The basal levels of NOX mRNA are often low, indicated by quantification cycle (C_q) values in the range of 30–32. Thus, relying solely on reverse transcriptase PCR (RT-PCR) without performing reverse transcriptase-quantitative PCR (RT-qPCR) does not provide a secure sense of the quantity of NOX mRNA present. In addition, the high C_q values, indicating very low mRNA quantities, should heighten the level of concern about false-positive results arising from trace contamination from genomic DNA (gDNA), NOX-bearing plasmids, and other sources. Thus, the inclusion of appropriate negative controls is crucial, and the handling of samples requires rigorous protocols to avoid contamination. In addition, several NOX isoforms have variants due to alternative splicing. Thus, an added layer of precision is required in designing primers that detect the full-length form. We recommend that RT-qPCR be performed prior to immunoblotting to obtain an indication of whether the level of protein is likely to be detectable by immunoblot. Here, we share what we have learned and hope that these guidelines will improve the validity of published NOX immunoblots and PCR data in the future.

2. Materials

2.1. RNA Isolation and Reverse Transcriptase-Quantitative PCR (RT-qPCR)

1. RNaseZap™ RNase decontamination solution.

2. Decontaminated workspace and equipment free of NOX plasmids.

3. Nuclease-free water.

4. Nuclease-free pipette tips with barrier filters.

5. RNeasy Plus Kit containing gDNA eliminator columns.

6. 70% ethanol prepared in DEPC-treated water.

7. DEPC-treated water.

8. β-mercaptoethanol.

9. 1.5 mL micro-centrifuge tubes.

10. Phosphate-buffered saline (PBS).

11. Cell scrapers.

12. 200-proof ethanol (100%) to prepare 70% ethanol in DEPC-treated water.

13. 1 mL syringes and 21-gauge needles for homogenization or Qiashredder columns.

14. qScript XLT™ cDNA SuperMix.

15. 0.2 mL PCR tubes.

16. A cold 4 °C cube designed to hold small tubes and 96-well plates for PCR setup.

17. Mini-centrifuge capable of briefly spinning down 0.2 mL PCR tubes.

18. RT-qPCR instrument capable of accepting 96-well plates such as Bio-Rad’s CFX96.

19. 96-well PCR plate suggested by the manufacturer of the PCR instrument.

20. Adhesive sealing film for PCR plates.

21. A paddle to apply the adhesive sealing film.

22. Primers for the reference gene and for the specific NOX isoform(s).

23. PerfeCta® SYBR® Green PCR SuperMix or similar.

24. A centrifuge capable of spinning 96-well plates.

2.2. SDS-PAGE and Immunoblotting

1. SDS-PAGE, appropriate percentage or gradient gels.

2. Standard running buffer: 0.025 M Tris-base, do not adjust pH, 0.192 M glycine, 0.1% SDS.

3. Power supply capable of attaining 200 volts.

4. Pre-stained MW protein ladder.

5. Polyvinylidene fluoride (PVDF) membrane, filter papers, sponges, and transfer cassettes and tanks (standard wet transfer).

6. Standard transfer buffer: 0.025 M Tris-base, do not adjust pH, 0.192 M glycine, 20% (v/v) of 100% methanol, chilled to 4 °C.

7. 100% methanol.

8. TBST: 10 mM Tris–HCl, pH 7.4, 0.15 M NaCl. 0.1% Tween-20.

9. Blocking buffer: 5% nonfat dry milk in TBST. Store at 4 °C.

10. Primary antibodies: See Tables 5, 6, 7, 8, 9, 10, and 11 for recommendations.

11. Secondary antibodies: HRP-conjugated goat anti-rabbit, HRP-goat anti-mouse, or fluorescent-labeled antibodies.

12. ECL reagent such as SuperSignal West Pico Chemiluminescent Substrate.

13. Plastic sheets to encase the immunoblot.

14. Autoradiography cassette equipped with intensifying screen son each side.

15. Autoradiography film and processor (or other film-less detection system).

16. Optional: LI-COR Odyssey fluorescent near-infrared (IR) scanner.

17. Optional: LI-COR Odyssey Blocking buffer (TBS).

18. Optional: LI-COR Odyssey IRDye 680LT (red)-labeled secondary antibody.

19. Optional LI-COR Odyssey IRDye 800Cw (green)-labeled secondary antibody.

20. Optional LI-COR Odyssey fluorescent IRDye-labeled protein ladder.

21. Plasmids for positive control antigens: Mammalian expression plasmid containing the NOX cDNA sequence for a specific NOX isoform.

22. Negative control plasmid: Same plasmid as above without the NOX sequence.

23. Human embryonic kidney cell line (HEK 293 cells) or any other cell line with very low or no endogenous expression of NOX/DUOX enzymes.

24. DMEM medium, high glucose (4.5 g/L), containing 10% FBS, or similar.

25. Cell line to be tested, appropriate culture medium.

26. Transfection reagent, such as X-tremeGENE 9, Lipofectamine, etc.

27. OptiMEM medium without any supplements.

28. RIPA buffer: 50 mM Tris–HCl, pH 7.5, 150 mM NaCl, 0.5%sodium deoxycholate, 0.1% SDS, 1.0% Triton X-100.

29. Protease and phosphatase inhibitor cocktail or tablets.

30. Sonicator equipped with a pencil-tip probe or 1 mL syringes and 21-gauge 1-in. needles for homogenization.

31. Micro-centrifuge, cooled to 4 °C.

32. 4× Laemmli’s sample buffer: 200 mM Tris–HCl, pH 6.8, 20% glycerol, 0.4% SDS, 4% β-mercaptoethanol.

33. Protein determination kit, such as the BCA protein determination kit.

3. Methods

3.1. RNA Isolation

1. Use a PCR-clean workspace and tools that have been washed with soap and have been decontaminated from DNA (e.g., NOX plasmids and amplicons) by treatment with UV light. Spray the workspace, tools, and your gloves with ZAP spray. Wipe off any excess with a paper towel (see Note 1, Fig. 1).

2. Use barrier-filtered, PCR-clean pipet tips for all the following steps. Spray the surface of the pipette tip boxes with RNase Zap spray, especially where the thumb is placed.

3. Follow the protocol from Qiagen RNeasy Plus Kit described briefly below for cell lines. Refer to the manual for isolation of RNA from other sources such as tissues. Be sure to wipe the interior and exterior surfaces of the micro-centrifuge with Zap spray, especially knobs, dials, buttons, and the lid where the thumb is placed. Clean any rings of precipitated material from previous spins from the interior walls of the micro-centrifuge.

4. Wash adherent cells twice with PBS (see Note 2).

5. Add 600 μL of resuspension buffer containing fresh 5% β-mercaptoethanol to a 10-cm plate of rinsed cells (about 2.5 × 10⁶ cells).

6. Scrape the cells off the plate, and transfer to the next plate and scrape again. A total of 1 × 10⁷ cells can be processed in 600 μL. Transfer to a 1.5 mL micro-centrifuge tube.

7. Homogenize the cells using a Qiashredder column (Qiagen),or manually homogenize by using a 1-mL syringe and 21-gauge needle, 5 passes. Take 600 μL and add it to the gDNA eliminator column. Spin 10,000 rpm (8000 × g) for 30 s in a micro-centrifuge. Collect the flow through. Dispose of the used gDNA column, and reapply the flow through to a fresh gDNA column, and spin again to ensure that the gDNA is removed (see Note 3).

8. Precipitate the RNA by adding an equal volume (600 μL) of 70% ethanol prepared in DEPC-treated water.

9. Take 700 μL of the 1200 μL sample and add it to the RNA-binding column. Spin as before. Discard the flow through. Add the remainder of the sample to the same column and spin again.

10. Discard the flow through, and wash the column with the wash buffers as directed by the kit.

11. Dry the column by performing the optional spin at 13,200 rpm (15,000 × g), 1 min without buffer added to the column. Elute with 30–50 μL of nuclease-free water to keep the RNA concentrated. Determine the RNA concentration using a Nanodrop spectrophotometer or by other standard methods (see Note 4).

12. Store RNA samples at −70 to −80 °C or proceed to cDNA synthesis on the same day (see Note 5).

Fig. 1.

Preventing contamination in RT-qPCR. (a) Wash the bench and surfaces of tools and equipment with a paper towel soaked in diluted detergent and water. If possible unscrew the lower parts of the pipette barrel and wash in soapy water. (b) The dial and top of the pipette where the thumb is placed is a large source of contamination. (c) Treat the washed pipettes and test tube racks with UV light in a biosafety cabinet or in a UV light box for 1 h. (d) Use a gDNA eliminator column twice to ensure gDNA removal

3.2. cDNA First-Strand Synthesis

1. On the day of cDNA first-strand synthesis, thaw the RNA and keep it on ice. Place 2 μg of RNA into each 0.2 mL PCR tube (or well of a 96-well PCR plate for many samples). Prepare the negative control (Tube 1) first by adding nuclease-free water to the RNA to 20 μL total. To the test sample (Tube 2), add nuclease-free water to the RNA for a final volume of 16 μL. Add 4 μL of 5× qScript XLT cDNA SuperMix only to Tube 2. Each tube should have 20 μL total volume. Of these two tubes, 10 μL will be used for one NOX PCR master mix of 100 μL (4 replicate wells) and 10 μL will be used for one reference gene PCR master mix (4 replicate wells). Prepare enough cDNA first strand (and negative controls) if additional conditions are intended. Spin the PCR tubes briefly in a mini-centrifuge to bring the liquid to the bottom of the tubes before placing them in the thermocycler. If using a 96-well plate for many cDNA first-strand samples, seal the plate and centrifuge the plate at 400 × g for 5 min at 4 °C. Place in a standard thermocycler (it is not necessary to use a real-time PCR instrument at this step), and set the program according to the manufacturer’s instructions. If using the qScript XLT cDNA SuperMix, set the program for 5 min at 25 °C, 50 min at 42 °C, 5 min at 85 °C, and hold at 4 °C (see Note 6).

2. Return the RNA to the −70 °C freezer. Plan the experiment so that the first-strand cDNA can be used immediately, but if necessary freeze the first-strand cDNA tubes at −20 °C overnight (see Note 7).

3.3. Setting Up the RT-qPCR Master Mix and 96-Well Plate

1. Set up the following master mix RT-qPCR reactions (100 μL total volume in 1.5 mL tubes or in a 96-well plate if many samples are being processed). Avoid working on ice since ice may accidently contaminate the samples. Instead use a 4 °C test tube rack (cube) designed to hold PCR tubes or plates. If using a SuperMix containing a hot-start DNA polymerase, then the setup may be performed at 25 °C. Each 100 μL of master mix reaction will be pipetted into 4 replicate wells of the PCR plate (20 μL/well plus 20 μL extra).

2. Always start with the no template control (NTC) tubes first, then the negative control (NC) tubes, and finally the unknown (UNK) tubes (see Note 8).

3. No template controls (NTC): Add 50 μL of 2× PerfeCtA® SYBR® Green SuperMix (see Note 9). Add 30 μL of nuclease-free water. Add 10 μL of 10× (2 μM) NOX forward primer and 10 μL of 10× (2 μM) NOX reverse primer to NTC-1. The NOX primers should be isoform-specific and able to detect the full-length variant. Table 1 lists some of the variants of human NOX isoforms, their main differences, and their accession numbers. Table 2 lists the human NOX primers that we use. All of them detect the full-length isoform as well as all the variants known to date (see Note 10). Table 3 lists mouse NOX primers from a reference [2]. Add 10 μL of 10× (2 μM) reference gene forward primer and 10 μL of 10× (2 μM) reference reverse primer to NTC-2. Do not add any cDNA or RNA to these NTC tubes. Table 4 lists a selection of human primers for reference genes that have been suggested by Integrated DNA Technologies’ RT-qPCR tutorial manual available on their website (see Note 11). Primers for other species can be purchased from Integrated DNA Technologies or other companies and validated as we have done for human NOX RT-qPCR primers.

4. Negative controls (NC): Add 50 μL of 2× PerfeCtA® SYBR® Green SuperMix. Add 20 μL of nuclease-free water. Add 10 μL 10× of forward and 10 μL of 10× reverse NOX primers to NC-1. Add 10 μL (1 ug RNA) of the negative control first-strand cDNA synthesis tube prepared during the cDNA synthesis step. Those tubes only contain RNA and no cDNA. Each tube will now have 100 μL total volume. Continue with NC-2, which will have the reference gene primers instead of NOX primers.

5. Unknowns (UNK’s): Add 50 μL of 2× PerfeCtA® SYBR® Green SuperMix. Prepare one UNK tube for each set of primers to be tested. Add 20 μL of water, 10 μL of 10× forward and 10 μL of 10 × reverse primer, and 10 μL of cDNA (1 ug RNA equivalents) that was synthesized in the presence of first-strand cDNA synthesis SuperMix. The total volume is 100 μL.

6. Use a clean, decontaminated vortex to gently vortex all the PCR master mix tubes for a few seconds, and then spin in a clean, decontaminated micro-centrifuge for 30 s at 10,000 rpm (8000 × g) to bring the liquid to the bottom of the tubes. Starting with the NTC tubes, pipet 20 μL into each of 4 wells of a 96-well PCR plate. There will not be enough volume for a fifth well. Proceed then with the NC tubes; pipet 20 μL into 4 wells. Pipet the UNK wells last, and then change gloves again.

7. Seal the PCR plate with a clear adhesive film made for PCR plates. Use a plastic PCR paddle, and drag it over the surface of the film to ensure that each well is sealed. Use the edge of the paddle and drag it in between column, rows, and all along the edge of the outer rows and columns to ensure that each individual well is sealed tightly (see Note 12 and Fig. 2). Centrifuge the sealed plate at 400 × g (2000 rpm) for 5 min at 4 °C to bring all the liquid to the bottom of the wells.

Table 1.

Human NOX variants, accession numbers, and main characteristics

NOX isoform	Variant number (V) and NCBI accession #	Notes about the variant (see accession # for full details)
NOX1	(V1) NM_007052.4 (V2) NM_013955.2 (V3) NM-001271815	Full length Deletion of 1503–1610 (exon 10) Deletion of 459–543 (exon 4)
NOX2 (CYBB)	NM_000397.3	Full length
NOX3	NM_015718.2	Full length
NOX4	(V1) NM_016931.4 (V2) NM_001143836.2 (V3) NM_001143837 (V4) NM_001291926 (V5) NM_001291927.1 (V6) NM_001291929.1 (V7) NR_120406.1 (V8) NM_001300995.1	Full length Deletion of 1546–1656 (exon 14) Deletion of 1–375 (exon 1) Deletion of 472–475 (part of exon 3) Deletion of 1–375 (exon 1) Deletion of 1–471 (exon 1 and 2) Non-coding Deletion of 1–375 (exon 1) and of nucleotides corresponding to exon 14 of full length
NOX5 [8, 9]	(α) NM_001184779.1 (β) NR_033672.1 (γ) NM_024505.3 (δ) NR_033671.2 (ε) AF317889 (ζ) NM_001184780.1	NOX5α; 18 aa extension at the start of the N-terminus (exon 3); deletion of 27 aa in exon 6; predominant isoform in spleen NOX5β; deletion of 27 aa in exon 6; predominant isoform in testis NOX5γ complete exon 6 without the 18 aa extension of the N-terminus NOX5δ complete exon 6 with the 18 aa extension of the N-terminus NOX5ε (a.k.a. NOX5-S); truncated from exon 3 through most of exon 7 NOX5ζ: contains an extra 25 aa at the start of the N-terminus
DUOX1 NM_017434.4		Full length
DUOX2 NM_014080.4		Full length

Table 2.

Primers used in the detection of human NOX isoforms by RT-qPCR

NOX isoform	Human sequence of primer	Exon location and Tm	Integrated DNA technologies Cat. #
NOX1	GGTCAACACGAGGAGAGC CAAGGATCCACTTCCAAGACTC	E7F, 60.7 E8R, 61.4	Hs.PT.49a.435525.4
NOX2 (CYBB)	CCCAATCCCTCAGTTTGCT CCTTCTGTTGAGATCGCCAA	E7F, 61.1 E8R, 61.2	Hs.PT.51.3983106
NOX3	ACCTTCTGTAGAGACCGCTAT TCACATGCATACAAGACCACA	E7F, 61.2 E8R, 61.1	Hs.PT.51.3065606
NOX4	CTGTGGTGTTACTATCTGTATTTTCTC CTTGCTGCATTCAGTTCAACA	E4F, 60.7 E5R, 60.8	Hs.PT.51.22695360.g
NOX5	GCCAGTGCCTCAACTTCG CCACTACCACGTAGCCCATA	E7F, 62.1 E8R, 62.1	Hs.PT.58.40874887
DUOX1	GCGTCTACATGAGAAATGCCA GCAGCAGTGCATCCACAT	E10F, 61.8 E12R, 61.9	Hs.PT.51.21346453
DUOX2	CGCCACCTACCAGAACATC GGTAGAGAAGAACTGC TCAGAG	E7F, 61.1 E9R, 61.3	Hs.PT.51.20690545

Primers from PrimeTime™ qPCR Pre-designed Assays (without probes) were purchased from Integrated DNA Technology (IDT) and may have been modified slightly to obtain desired Tm. Primers are listed in the 5′ to 3′ direction (R = reverse and F = forward). The exon number may be different for different accession numbers. The T_m values were determined using IDT’s OligoAnalyzer tool (http://www.idtdna.com/calc/analyzer) using the following final concentrations in the qPCR reaction: [primers] = 200 nM, [Na⁺] = 55 mM, [Mg²⁺] = 1.5 mM, and [dNTP] = 0.2 mM. The actual annealing temperatures were determined by performing a gradient RT-qPCR for each primer set using the methods described in the text. These primers detect variant 1 (the full-length variant). To determine whether these primers detect other variants, the user should check the accession number of the variant, some of which are provided in Table 1

Table 3.

Primers used in the detection of mouse Nox isoforms by RT-qPCR

Nox isoform	Mouse primer sequence	Exon location and Tm	Accession no.
Nox1	CCCAGCAGAAGGTCGTGATT GCTAAAGCCTCGCTTCCTCAT	E8F, 63.2 E9R, 63.5	NM_172203.2
Nox2 (Cybb)	CAGGAACCTCACTTTCCATAAGAT AACGTTGAAGAGATGTGCAATTGT	E4F 61.5 E5R 63.0	NM_007807.5
Nox3	CGACGAATTCAAGCAGATTGC AAGAGTCTTTGACATTGCTTTGG	E12F. 61.3 E13R, 61.0	NM_198958.2
Nox4	CCGGACAGTCCTGGCTTATCT TGCTTTTATCCAACAATCTTCTTGTT	E3F, 64.3 E4R, 61.7	NM_015760.5
Duox1	CACCAGGAACGGATTGTTCT CCTGCAAGCCAAAAGAAGAC	E12–13F, 61.2 E13R, 60.7	NM_001099297.1
Duox2	TGCAACAGCTACTGGATTCG ATCCTGTCCTCCAGCTCTGA	E9F, 61.4 E10R, 63.3	NM_001362755.1

These primers were selected from [2]. Primers are listed in the 5′ to 3′ direction (R = reverse and F = forward). The exon number may be different for different accession numbers. The T_m values were determined using IDT’s OligoAnalyzer tool (http://www.idtdna.com/calc/analyzer) using the following final concentrations in the qPCR reaction: [primers] = 200 nM, [Na⁺] = 55 mM, [Mg²⁺] = 1.5 mM, and [dNTP] = 0.2 mM. These primers detect the full-length variant 1 sequence. The listed accession numbers contain details of existing variants

Table 4.

Some suggested human reference primers

Gene	Primer sequence	Exon location ID	Accession #T Cat. #
HMBS	GCAACTGTACCTGACTGGA TCAGGGCCATCTTCATGC	E12–13F, 60.1 E14R, 60.7	NM_000190.3 Hs. PT.58.242481164
HPRT1	CATCAAAGCACTGAATAGAAATAGTGA CCAATTACTTTTATGTCCCCTGTT	E3F, 60.5 E4R, 61.1	NM_000194 Hs. PT.53a.20881146
HPRT1	AGGTATGCAAAATAAATCAAGGTCAT TTCTCCTGAGCAGTCAGC	E1F E2R	NM_000194 Hs. PT.53a.2145446
YWHAZ	TCTGATCCCCAATGCTTCAC TGGTATGCTTGTTGTGACTGA	E8F, 60.9 E9R, 60.9	NM_003406 Hs. PT.53a.20437075

A discussion of reference gene choice can be found in Refs. [3, 10]. Primer sequences are 5′ to 3′. Catalog numbers are for IDT’s Prime Time Pre-Designed qPCR Assays. The exon number may be different for different accession numbers. The T_m values were determined using IDT’s OligoAnalyzer tool using the following final concentrations in the qPCR reaction: [primers] = 200 nM, [Na⁺] = 55 mM, [Mg²⁺] = 1.5 mM, and [dNTP] = 0.2 mM. The actual annealing temperatures were determined by performing a gradient RT-qPCR for each primer set using the methods described in the text. Primers for mouse or rat can be obtained by entering the name of the gene and desired species on the IDT website

Fig. 2.

Correct manual application of the adhesive seal to PCR plates. A plastic paddle is available commercially to assist in the proper application of the adhesive seal to PCR plates. Large standard deviations result from evaporation from individual wells that are not sealed completely

3.4. Running the RT-qPCR Program and Obtaining the Data

1. Insert the 96-well plate into a RT-qPCR instrument, and enter the plate layout using the software of the instrument. Figure 3a shows an example of a 96-well PCR plate layout using Bio-Rad’s CFX 96 instrument and CFX Manager software. Choose a two-step PCR program followed by a melt curve. An example of a two-step program that we use is 3 min at 95 °C for 1 cycle (hot start), then 40 cycles of 95 °C for 15 s and 55 °C for 30 s, plus a melt curve (Fig. 3b). The optimal annealing temperature is determined by performing a gradient RT-qPCR as explained next.

2. To optimize the annealing temperature for each primer set, perform a gradient RT-qPCR by inserting a gradient step in the program such that each row represents a different annealing temperature (Fig. 3b). Then prepare a master mix for 25 identical reactions, and place 8 samples in 1 column of the plate, or if tubes, place the tubes in one column of the instrument. Replicates are recommended and should be placed in additional columns. Run the two-step program with the gradient step inserted. The annealing temperature yielding the lowest C_q value should be chosen. For the human primers in Table 2, the optimal annealing temperatures are within 55–58 °C using the reagents and instrument we used. Examples of the amplification curves obtained using the human NOX4 primer set E4F, E5R in Table 2 and reference gene HPRT1 primer set E3F, E4R in Table 4 for RCC4 and HEK 293 cells are shown in Fig. 3c, d, respectively.

3. The software of the instrument calculates the threshold cycle (formerly known as “C_t”). The correct terminology according to the MIQE rules is quantification cycle (C_q) [3]. The C_q values are dependent upon the value of the threshold level which is either set by the user or determined by the instrument. Consult the instrument’s manual for directions. Preferably, obtain the C_q values by linear regression of the amplification curves which does not depend on the user’s input of the threshold level. Consult the instrument’s manual on how to choose this setting. Observe the software’s calculated mean and standard deviation of the replicate wells for each sample. If necessary, eliminate one outlier out of the four replicates. If more than one well is an outlier, or the standard deviation is too large (>0.5), it is recommended that additional replicates be used and that the experiment repeated until technical errors (pipetting, mixing, sealing) are minimized as determined by the standard deviation. If a C_q value <35 is obtained in the NTC or NC wells, this indicates either a primer-dimer problem, or DNA contamination. In this case, the melting temperature of the amplicons of the NTC and NC wells should be examined to see if the values are due to NOX contamination or primer-dimers. For NOX DNA contamination, the melting temperature will be identical to that obtained from a positive control (plasmid containing the NOX isoform). The PCR products should be run on an agarose gel to check that the anticipated amplicon size is correct. Primer-dimers run less than 100 bp.

4. Sequencing of the PCR products from the UNK wells adds another layer of confidence that the products are correct.

5. Obtain the ΔC_q values for each well by subtracting the C_q value of the reference sample from the C_q value of the NOX sample. The mean ΔC_q value can be used to compare relative NOX mRNA levels between different samples if the reference gene C_q value does not differ significantly between samples. These values can also be used as an indication of whether or not protein detection will be feasible (see Note 13).

6. If two conditions are being compared, e.g., treated vs. untreated, then the software will also calculate the relative fold change, R, using the ΔΔC_q (Livak) method. If the efficiency of the NOX primers is determined to be quite different (10% or more) from the reference gene, then the Pfaffl method should be used as well and the results compared using the two methods. The primer efficiency values can be entered into the dialogue box (see Notes 14 and 15).

7. To determine the efficiency of the NOX and reference gene primer sets, generate a RT-qPCR standard curve using linearized plasmids that contain the NOX cDNA or the reference gene. Be careful not to use the PCR-designated bench and tools to make a series of ten 10-fold dilutions of the linearized plasmid in nuclease-free water starting with a known amount of the plasmid, about 1 μg/μL. Leave the dilutions at the non-PCR bench. Change gloves and go to the PCR-designated bench. Set up ten master mix PCR tubes for one primer set without cDNA. Add 10 μL of one of the plasmid dilutions per tube. Vortex gently and centrifuge the tubes briefly. Pipet four replicate wells per column of the PCR plate (40 wells). Seal the plate well. It is not necessary to mix the wells since centrifugation of the sealed plate (400 × g, 2000 rpm, 5 min, 4 °C) will induce mixing. Freeze the diluted stocks at −20 °C so that they can be used again. Run the qPCR program under the same conditions, but label each well as STD-1, STD-2, STD-3, etc., and enter the dilution-fold values associated with each set of wells. The software will generate a standard curve of C_q values vs. dilution-fold as shown in Fig. 4. The efficiency is calculated from the slope, which ideally would be −3.32 (the binary exponent required for a tenfold change in C_q). The standard curve can also be used to estimate copy number by converting the mass of the plasmid (plus insert) to moles and then to the number of molecules, which is the number of copies since each plasmid molecule contains one copy of the gene (see Note 16).

Fig. 3.

Example of a basic qPCR setup using a Bio-Rad CFX 96 instrument and CFX manager software. (a) Plate layout of the no template control (NTC), negative control (NC) and unknown (UNK) wells, 4 replicate wells each for the NOX primers, and for the reference primers. (b) Example of a two-step PCR program with an annealing temperature gradient inserted. (c) Representative amplification curves (3–4 replicate wells per set of curves) for the reference gene (HPRT1, left set of curves) and NOX4 (right set of curves) for untreated RCC4 clear cell renal cell carcinoma cells. (The cells were a kind gift from Dr. William Kaelin, Harvard University). (d) Amplification curves for untreated HEK 293 cells. Note that for the negative controls (NC) and no template controls (NTC) wells, the curves should resemble the baseline (flat lines) up to cycles 35–40. Primers (200 nM) for NOX4 (E4F and E5R) from Table 2 and HPRT1 (E3F and E4R) from Table 4 were used. Each UNK well contained 200 ng of cDNA (RNA equivalents) or 200 ng RNA in place of cDNA in the NC wells. The optimal annealing temperature was determined from a gradient RT-qPCR experiment

Fig. 4.

Example of a RT-qPCR standard curve used for the determination of primer set efficiency. Ten tenfold serial dilutions were made from pcDNA3.1-NOX4 variant 1 at a starting concentration of 1 μg/μL. Primers (200 nM) for NOX4 (E4F and E5R) from Table 2 were used. The C_q values were plotted vs. log (dilution-fold), and the efficiency of the primer set was determined from the slope

3.5. Immunoblotting: Preparing Lysates of Positive and Negative Control Cells

1. The overexpression of NOX isoforms in a cell line having little or no NOX expression is used as a positive control, and the overexpression of the corresponding empty vector in the same cell line is used as the negative control. NOX1–4 require p22^phox for maturation and expression, so use a cell line that expresses p22^phox for these isoforms. We recommend HEK 293 cells because they contain sufficient endogenous p22^phox protein to support stable NOX isoform expression [4], and they have a transfection efficiency close to 100%. If using a different cell line, the presence of p22^phox protein should be confirmed. If the cell line does not express p22^phox protein (e.g., COS7 cells), then the CYBA gene that codes for p22^phox should be co-transfected along with NOX plasmids. DUOX1 and DUOX2 require the co-transfection of their maturation proteins, DUOXA1 and DUOXA2, respectively. Harvest HEK 293 cells from freshly seeded master plates prepared 1–2 days before the day of seeding. Seed the HEK 293 cells at 2 × 10⁶ cells (high confluency for X-tremeGENE 9, ~80%) in 10 mL of DMEM containing 10% FBS per 10 cm dish without addition of antibiotics the day before transfection.

2. On the day of transfection, add a total of 6 μg of positive control plasmid (e.g., pcDNA3.1-NOX) to 600 μL of OptiMEM in a 1.5 mL micro-centrifuge tube and vortex gently. The DNA concentration should be kept at 0.01 μg/μL. Prepare a second tube in the same way for the negative control plasmid (e.g., pcDNA3.1 plasmid lacking the NOX cDNA as a negative control). If desired, a GFP plasmid (0.5 μg/10 cm plate) can be co-transfected to estimate transfection efficiency (see Note 17).

3. Transfect the cells following the manufacturer’s directions. A number of transfection reagents are suitable such as Lipofectamine, but we use the X-tremeGENE 9 reagent because it does not require any media changes after transfection. If using X-tremeGENE 9, it should be at room temperature and gently vortexed for a few seconds just prior to use. For a 10-cm dish of cells containing 10 mL of media (no antibiotics), add 18 μL of X-tremeGENE 9 to 6 μg of DNA in 600 μL OptiMEM, keeping the ratio of X-tremeGENE 9:DNA at a ratio of 3 μL:1 μg DNA. Be sure not to touch the walls of the tube while depositing the transfection reagent directly into the liquid. Vortex the mixture gently, and spin briefly in a micro-centrifuge to collect all the liquid to the bottom of the tube. Incubate at 25 °C for 15 min to allow lipid micelles to incorporate the DNA.

4. Add the mixture dropwise to the 10 cm plate while swirling the plate. Swirl or rock the plate a few more times prior to placing the plate(s) back in the incubator. Incubate for 24–48 h for protein expression. No media change is required if using XtremeGENE 9 (see Note 18).

5. Remove the medium from the plates containing the transfected HEK 293 cells. Wash the cells twice with PBS. Remove as much of the wash as possible. Place the plates on ice. To the first 10 cm plate, add a minimal volume (300–400 μL) of ice-cold RIPA buffer containing protease and phosphatase inhibitors. Scrape the plate, collect the scraped cells, and keep on ice. If multiple plates are used per condition, then add the collected cells from the first plate to the next plate, and scrape using the same 300–400 μL of RIPA buffer (see Note 19).

6. Sonicate or homogenize both the control and test sample lysates gently by placing the micro-centrifuge tube in a beaker of ice, and administer two 1 s bursts from a sonicator equipped with a pencil-tip probe. Avoid foam by using a low-output setting of 30%. If a sonicator is not available, homogenize 5 times by using a 1-mL syringe and a 21-gauge needle, avoiding foam formation.

7. Clarify the lysate by centrifuging the samples at 13,000 rpm (15,000 × g), 10 min, 4 °C.

8. Collect the clarified lysate, and determine the protein concentration using the BCA protein determination kit or similar methods.

9. Storage of samples: Do not use all of the lysate for SDS-PAGE gel sample preparation because the NOX proteins degrade within 1 week if stored in Laemmli’s sample buffer. Prepare only as much gel sample as needed on the day of use. Aliquot and store the lysates at −20 °C for up to 2 weeks or for longer storage at −70 °C. If the protein precipitates after thawing, re-sonicate or homogenize, re-clarify, and determine the protein concentration again for positive and negative controls. For test samples, fresh lysates should be prepared each time and stored for no more than 1–2 weeks or as determined by each user.

10. To prepare concentrated SDS-PAGE gel samples, use a concentrated recipe for Laemmli’s sample buffer, for example, 4× sample buffer. Take three parts of sample (e.g., 75 μL of the cell lysate), and add one part (25 μL) of 4× Laemmli’s sample buffer. The protein concentration of the SDS-PAGE sample is now 75% of the original lysate. For a 4 mg/mL sample, it is now 3 mg/mL. For a 3 mg/mL sample, load 33 μL which is 100 μg of the test sample on the SDS-PAGE gel. For the positive and negative control lysates, load 0.1, 1, and 10 μg initially, and then determine the best amount to use in your conditions (exposure time, ECL reagent strength, etc.).

11. For NOX, do not boil the samples (see Note 20).

3.6. Preparing Lysates of Cells with Endogenous NOX

1. For the cells to be tested (e.g., RCC4 renal clear cell carcinoma cells), grow enough cells to obtain 2–3 × 10⁷ cells/mL of lysis buffer for each test condition.

2. If desired, treat the cells with a stimulus that can induce or upregulate NOX isoform expression (e.g., TGFβ1, 5 ng/mL final in complete media for 24 and 48 h). Leave some plates untreated or treated with the vehicle.

3. Harvest the cells, and prepare the lysates from the test plates as done in steps 5–11 in Subheading 3.5 above.

3.7. SDS-PAGE and Transfer

1. Use an appropriate percentage SDS-PAGE gel. A 1.5-mm-thick gel and comb will permit large sample volumes to be loaded (e.g., 100 μg protein in 30–50 μL) (see Note 21).

2. Load the pre-stained or fluorescently labeled protein ladder and 100 μg of test samples (30–50 μL). Also load the positive and negative control cell lysates (between 0.1 and 10 μg, determined by user). Run the 1.5-mm-thick mini-gels (about 8 × 10 cm plus the stacking gel) at 25 °C using constant amperage (30 mA/gel) for 15 min to allow stacking, and then increase the amperage to 40 mA/gel for about 1–1.5 h until the dye front reaches the very bottom edge of the gel to obtain optimal separation of the bands. After the run, rinse the gels briefly by dipping them in transfer buffer, and place them on the PVDF membrane. The PVDF membrane should be pre-wetted in 100% methanol and equilibrated in transfer buffer. Remove any bubbles by smoothing with a gloved finger or roller tool.

3. Transfer the proteins to the PVDF membrane. For a standard wet transfer, set the voltage at 90 V for 1 h and transfer at room temperature or in a cold room, use chilled transfer buffer containing 20% methanol, add a small magnetic stir bar to the tank, and place the tank on a magnetic stir plate.

4. Block the PVDF membrane after the transfer for 1 h at RT on a rocking platform. Use 5% nonfat dry milk in TBST, or use LI-COR Odyssey Blocking Buffer without Tween-20. Incubations of up to 3 days (over the weekend) at 4 °C on a rocker should not result in high background if PVDF membranes are used. If using 5% nonfat dry milk in TBST as a blocking buffer, wash the blot with TBST 4 times (5 min each) before adding a primary antibody that is diluted in LI-COR Odyssey Blocking Buffer plus 0.1% Tween-20 (see Note 22).

5. For most antibodies, dilute the antibodies in 3–5% nonfat dry milk in TBST or use LI-COR Odyssey’s Blocking Buffer (TBS) plus 0.1% Tween 20 which works for both fluorescent and non-fluorescent antibodies. A dilution of 1000-fold to 2000-fold (0.5–1 μg/mL final) is a good starting point for immunoblot optimization (see Note 22).

6. Incubate the blot with the primary antibody overnight or longer at 4 °C on a rocking platform. Detection of endogenous NOX proteins may require 72 h for detection (see Note 23).

7. Choice of primary antibody: Some recommended antibodies are listed in Tables 5, 6, 7, 8, 9, 10, and 11. Some commercial antibodies that did not pass validation under the conditions we used (HEK 293 cells overexpressing NOX isoforms) are also listed. Figures 5, 6, 7, and 8 show representative immunoblots using some of the selected antibodies from Tables 5, 6, 7, 8, 9, 10, and 11. Additional information about specific antibodies is provided in the footnotes of Tables 5, 6, 7, 8, 9, 10, and 11.

8. Recover and save the primary antibody and wash the blot 4 times, 5 min each with TBST.

9. Add the secondary antibody to the blot. Dilute HRP-conjugated goat anti-rabbit or mouse antibodies according to the manufacturer’s instructions (see Note 22). Incubate the blot with the secondary antibody solution for 1 h at room temperature. If using LI-COR Odyssey’s fluorescent secondary antibodies, dilute them 20,000-fold in LI-COR blocking buffer plus 0.1% Tween 20 and 0.01% SDS as recommended by LI-COR. Protect the solution from light during incubation. Incubate the blot with the secondary antibody solution for 1 h at room temperature.

10. Discard the used solution of HRP-conjugated secondary antibody. Save fluorescent secondary antibody solutions in a foil-wrapped tube at 4 °C, and reuse until the signal fades (~3 repeats).

11. Wash the blots 4 times with TBST, 5 min per wash, and then add the ECL reagent or proceed to LI-COR Odyssey scanning.

12. Prepare the ECL reagent as directed by the manufacturer. For SuperSignal West Pico Enhanced Chemiluminescent Substrate, we do not dilute the reagent. For an 8 × 10 cm blot, add 2 mL of the SuperSignal West Pico Enhanced Chemiluminescent Substrate, and leave it on for 5 min on a platform rocker (or rock occasionally manually).

13. Remove the blot from the ECL detecting solution, and blot lightly between two pieces of filter paper so that the ECL solution is not dripping. Tape the blot in a sheet protector that has been cut (with the folded edge intact) to the shape of a (used) autoradiography film. Open the sheet protector and tape the blots between the plastic sheets. Close the sheet protector over the blots, and insert it into the autoradiography cassette containing intensifying screens on both sides.

14. Expose the membrane for 1, 10, and 30 s and then 1, 3, 5, and 10 min. If no bands appear by 30 min, then remove the blots and place them in an ECL reagent with higher sensitivity such as SuperSignal West Dura solution. No rinsing of the blot is required before going to the next level of SuperSignal West ECL reagent. SuperSignal West Femto usually results in high background during long exposures.

15. For LI-COR Odyssey scanning, the blot can be wet or lightly dried between two filter papers. Place the wet blot facedown on the glass bed of the scanner. For dried blots, place a glass gel plate or similar weight on top of the membrane to keep it flat. Be sure to choose the membrane mode (169 μm thickness). Preview under an initial intensity of 3.0 for each IRDye wavelength. If the preview looks fine, then proceed to scan the membrane, and adjust the signal to noise ratio as desired using the settings of the instrument. Consult the instrument’s manual for further instructions. If necessary, scan again at a higher intensity for increased sensitivity. The membranes can be re-scanned up to 6 months if stored in the dark.

Fig. 5.

Examples of using positive and negative control lysates for the validation of NOX4 antibodies. Two laboratories tested several NOX4 antibodies from Table 8. Left column: NOX4 antibodies tested in the laboratory of Dr. J. D. Lambeth. Right column: NOX4 antibodies tested in the laboratory of Dr. J. Doroshow. Left column: Identical PVDF immunoblots consisting of 10 μg of the following non-heated cell line lysates. Lane (1) non-transfected HEK 293 cells, lane (2) HEK 293 cells stably transfected with pcDNA3.1-NOX4, untagged (a kind gift from Dr. R. Brandes (Goethe-University Frankfurt), lane (3) HEK 293 cells stably transfected with pcDNA3.1-NOX1, untagged (also a gift from Dr. Brandes). The blots were probed with the following NOX4 antibodies from Table 8 (a) Dr. A. Shah’s NOX4 rabbit pAb, 1:2000 (a kind gift from Dr. A. Shah, University College London); (b) Dr. J. Doroshow’s NOX4 rabbit mAb, clone 47–6, 1 μg/mL (1:1000); (c) NOVUS NOX4 rabbit pAb #NB110–58849 Lot F1 (1:1000, 1 μg/mL); (d) green, Dr. J. D. Lambeth’s NOX1 E39 mAb, 1:1000, as a positive control for NOX1; red, Dr. Shah NOX4 rabbit pAb, 1:2000. (e) β-actin mouse mAb (Sigma-Aldrich), 1:10,000. Blocking buffer, 5% nonfat dry milk in TBST; antibody diluent, LI-COR Odyssey blocking buffer +0.1% Tween-20; primary antibodies incubated overnight; secondary antibodies (HRP-conjugated goat anti-rabbit or anti-mouse), 1 h. ECL reagent: SuperSignal West Pico, 1 s. Right column: identical nitrocellulose membranes consisting of 40 μg of the following cell lysates. Lane (1) non-transfected HEK 293 cells, lane (2) HEK 293 transfected with pCMV6-empty vector, lane (3) HEK 293 stable clone (WT1) overexpressing N-terminal tagged-Myc-DDK-NOX4, lane (4) HEK 293 stable clone (WT2) overexpressing N-terminal tagged-Myc-DDK-NOX4. The following antibodies from Table 8 were tested: (f) Dr. P. Jansen-Dürr’s NOX4 mAb 1:500; (g) Millipore NOX4 pAb (#ABC549), 1:500; (h) Abcam NOX4 antibody (ab133303), 1:1000; (i) anti-Myc antibody; (j) anti-vinculin antibody. Conditions: similar to left panel, except the antibody was diluted in 5% milk in TBST and the ECL reagent was ECL Plus reagent (GE Healthcare)

Fig. 6.

Validation of DUOX antibodies. HEK293 cells were transfected with pcDNA3 HA- DUOX1(hD1) or DUOX2(hD2) and pcDNA3 V5-DUOXA1,2 constructs or empty vectors (EV) for the control condition. COS7 cells were transfected with pcDNA3 HA-mouse Duox2 (mD2) and pcDNA3 V5-mDuoxa2. Primary human thyroid cells were incubated 48 h with or without 14 ng/mL human IL-4. Lysis buffer: 50 mM phosphate buffer, pH 7.5, containing 20 mM EDTA, 1% Triton X-100 plus protease inhibitors (Complete; Roche Applied Science) and phosphatase inhibitors (1 mM vanadate, 100 mM NaF, and 50 nM okadaic acid). DUOX proteins (20–30 μg) in Laemmli buffer were heated at 80 °C for 5 min and were separated on a 6% gel and transferred to a low-fluorescence PVDF membrane (Immobilon-FL, Millipore, Billerica, USA). Blocking: LI-COR Odyssey blocking buffer, 2 h RT. Primary antibody: Dr. De Deken’s anti-DUOX pAb I2 1:5000 or anti-HA-mAb 1:5000 (Covance) overnight at 4 °C. Secondary antibodies: anti-rabbit IRDye800 (green) and anti-mouse IRDye680 (red) from LI-COR, 1 h RT. This DUOX antibody recognizes both human DUOX1 and DUOX2 as well as mouse Duox2 in overexpressing cell lines. DUOX proteins were also detected in human thyrocyte extracts, with higher expression after IL-4 treatment

Fig. 7.

Correlation between ΔC_q values and endogenous NOX4 protein detection in RCC4 cells. (a) C_q, ΔCq, and R (fold-change) values for untreated RCC4 parent cells and for TGFβ1-treated cells (5 ng/mL in complete medium). (b) NOX4 immunoblot probed with Dr. Shah’s NOX4 antibody (1:2000 for 72 h at 4 °C), then developed with SuperSignal West Pico Chemiluminescent reagent, and exposed for 30 min. Lanes (1–4) lysates of parent RCC4 cells (100 μg) untreated or treated with TGFβ1 for 0–72 h as indicated. Lanes (5–8) lysates of a representative RCC4 stable NOX4 knockdown (KD) clone untreated or treated with TGFβ1 for 0–72 h as indicated. Lane (9) positive NOX4 control (1.0 μg) of HEK293 cells stably expressing NOX4. The ΔC_q values corresponding to each condition are shown above each lane, except for lane 4 (72 h) for which the C_q was not determined (n.d.). (c) Immunoblot using Dr. Doroshow’s NOX4 rabbit mAb clone 47–6 (1:1, 000 for 72 h at 4 °C). Lane (1) 48 h, TGFβ1-treated parent RCC4 cells; lane (2) 48 h treated, representative RCC4 stable scramble shRNA clone; lane (3) 48 h treated, representative RCC4 stable NOX4 knockdown (KD) clone. The ΔC_q values corresponding to each condition are shown above each lane. NOX4 protein could not be detected in untreated RCC4 cells in these conditions using Dr. Doroshow’s NOX4 antibody. (e–g) Amplification curves for untreated and 24–72 h TGFβ1-treated parent cells corresponding to the C_q values in panel a; (h) amplification curves for untreated parent and representative clones of scramble shRNA and NOX4 shRNA (KD) corresponding to the C_q values in panel d.The scramble shRNA plasmid had an effect of decreasing NOX4 mRNA expression by eightfold, but the effect on protein expression could not be evaluated without TGFβ1 treatment. When the scramble clone was treated with TGFβ1 for 48 h, the NOX4 protein expression was similar to that of the parent cells treated for 48 h (lane 1 vs. 2, panel c). The NOX4 primer sets were NOX4 E4F/E5R and HPRT1 E3F/E4R as indicated in Table 2. The NOX4 shRNA (KD, retroviral) plasmid (OriGene Technologies, #TG302911) contained the target sequence 5′CACTGAACTATGAGGTCAGCCTCTGCGTG. The scramble (Scr) shRNA (retroviral) plasmid (OriGene Technologies, #TR30013) contained the non-targeting sequence: 5′ GCACTACCAGAGCTAACTCAGATAGTACT 3′

Fig. 8.

Validation of a Nox1 rat monoclonal antibody. A rat monoclonal antibody against mouse Nox1 (clone 31C1) detects Nox1 in wild-type (WT) mouse colon (lane 1) and cecum (lane 3) membrane fractions (70 μg) solubilized in 2% n-dodecyl β-D-maltoside in 10 mM Hepes pH 7.4, 10 mM NaCl, 100 mM KCl, 1 mM EDTA, and 1 mM PMSF. This monoclonal antibody is more specific and sensitive than a previous polyclonal antibody to the same epitope (see Table 5) [7]. The Nox1-specific bands are observed diffusely at 56–80 kDa due to differently glycosylated forms of Nox1 (lane 1). The specificity for Nox1 was confirmed using solubilized membrane fractions from the colon from Nox1-deficient mice (KO) as a negative control (lane 2, 4). This rat monoclonal antibody also detects Nox1 in solubilized membrane fractions of mouse small intestine and mouse lung (unpublished, personal communication from Dr. M. Matsumoto). However, Nox1 protein was not detected in other mouse tissues, possibly due to its low abundance

Fig. 9.

786–0 cells do not express NOX4 and NOX1. 786–0 cells have been used as a model for studying NOX4 and NOX1, but this cell line does not express NOX4 mRNA (a) or NOX1 mRNA (b) using the methods in this chapter. The y-axis scales differ in (a) and (b) due to the use of two different commercial SYBR Green reagents from Quanta in (a) and Bio-RAD in (b)