Improving Quantification Accuracy For Western Blots

by Kristi L. H. Ambroz

Figure 1. Nitrocellulose and PVDF membrane were scanned on the Odyssey Infrared Imaging System (LI-COR Biosciences, Lincoln, NE) at an intensity = 5 for both 700 and 800 nm wavelengths. The same membranes were scanned at 532 nm and 635 nm wavelengths with a PMT=500 on a GenePix 4100A (Molecular Devices, Sunnyvale, CA). Click here to enlarge.

Imaging systems used for Western blot analysis are often described as 'quantitative'. While it is true that the instrumentation used to image the blots does acquire signal in a quantitative way, data produced by these systems can only be quantitative if the signal generated by the chemistry is proportional to the amount of sample on the blot in a linear fashion.

Historically, the most popular protein detection method in Western blotting applications has been the use of chemiluminescence with either a CCD camera or exposure to film. Chemiluminescent detection has been the chemistry of choice for Western blotting due to its sensitivity. Chemiluminescent detection relies on an enzymatic reaction that produces light, which is detected by a CCD camera that records photons and displays an image based on the amount of light generated. Alternatively, the membrane can be placed on film and later developed into a viewable picture. The enzymatic reaction used to produce the light is dynamic, constantly changing over time. Some samples produce bright light for a short time, and others produced comparatively dim light, but for a long period of time. Therefore, images must be collected at an optimized time. This time-dependence of signal compromises quantification and accuracy.

Fluorescent detection, by comparison, is static. Light produced from the excitation of a fluorescent dye can be compared to a light bulb. When a fluorescent dye is excited, or 'on', the amount of light produced is constant. This makes fluorescent detection a more precise and accurate measure of the differences in signal produced by labeled antibodies bound to proteins on a Western blot. Proteins can be accurately quantified because the signal generated by the different amounts of proteins on the membranes is measured in a static state, as compared to chemiluminescence, in which light is measured in a dynamic state.

Secondary antibodies labeled with visible fluorophores have been available for Western blotting, however, their performance has been poor. More recently, near-infrared (NIR) fluorophores have been reported to provide excellent sensitivity, with all the advantages of fluorescent detection.

Advantages of NIR fluorescence detection

In addition to providing accurate quantification, NIR detection has other advantages.

High sensitivity. NIR detection is more sensitive than visible fluorescence and equal to or more sensitive than chemiluminescent detection (sensitivity with chemiluminescence depends on the substrate and detection methods that are used). At longer wavelengths, membrane surfaces and biomolecules exhibit greatly reduced autofluorescence, resulting in lower background and enhanced sensitivity when NIR fluorophores are used for detection. Untreated nitrocellulose and PVDF membranes have much lower autofluorescence when scanned in the NIR than in the visible range of the spectrum (Figure 1). As a result, there is a dramatic decrease in membrane associated background in the NIR. The reduction of background using NIR detection directly addresses the primary caveat of membrane-based protein detection with either chemiluminescence or visible fluorophores — low signal-to-noise ratio. NIR detection dramatically increases the signal-to-noise ratio for membrane-based applications.

Figure 2. Near-infrared versus chemiluminescent detection of transferrin at concentrations of 1000 pg, 32 pg and 1 pg on nitrocellulose membrane. (A) Odyssey Infrared Imaging System, 800 nm scan, with an intensity setting of 4. Chemiluminescent blot detected after (B) 30 seconds, (C) 1 minute, (D) 5 minutes, and (E) 10 minutes of exposure to film. Click here to enlarge.

Wide linear detection range. NIR-labeled secondary antibodies yield a much wider linear detection range than chemiluminescence on Western blots.⁽¹⁾ The practical result is illustrated in Figure 2. With chemiluminescence, optimizing exposure times for low protein concentrations may make bands with high protein concentrations unquantifiable (Figure 2E). Similarly, optimizing for high protein concentrations may make bands with low protein concentrations undetectable (Figure 2B). When visualizing uncharacterized proteins, a given protein may not be detected at all using chemiluminescence, simply because the exposure was not optimal. With NIR detection, the wider linear detection range and static nature of fluorescence detection means that all protein concentrations within the detection limits of the instrument are visible in the scanned image.

Multiplex detection. Cell signaling pathways are often regulated by protein phosphorylation cascades. A protein such as a receptor tyrosine kinase becomes phosphorylated following ligand binding setting off multiple signaling pathways. Highly sensitive two-color NIR fluorescence detection enables the simultaneous detection of the total protein present as well as the amount of protein phosphorylation that has occurred without the need for stripping and reprobing — something that cannot be done with chemiluminescence. This approach to multiplexed detection has been implemented in fluorescence imaging instruments such as the Odyssey Infrared Imaging System from LI-COR Biosciences. The Odyssey system uses two separate NIR lasers and detectors to image labeled antibodies at 710 and 805 nm, simultaneously (Figure 3).

This simultaneous ratiometric approach to multiplexed detection greatly increases the accuracy of quantitative immunoblotting. One detection channel can be used to detect a protein of interest and a second channel to normalize for sample loading; for example, detecting the amount of phosphorylated protein in one channel and a housekeeping protein in the second. The ratio of the signal from an unknown protein to a housekeeping protein can be used to accurately normalize the signal intensities and correct for loading and sampling errors.

Two-channel NIR fluorescent detection provides the unique capability to accurately quantify proteins over a large dynamic range. Very low amounts of proteins have been accurately quantified using this ratiometric approach.⁽²⁾ This accuracy is difficult to achieve with other detection methods. While chemiluminescent detection has good sensitivity, its limited dynamic range, and lack of a second detection channel for ratiometric analysis make it very difficult to accurately quantify low amounts of proteins.

Simplified procedures. A final advantage of NIR detection is the simplified way Western blot imaging is conducted. Substrates, film, and a darkroom are no longer needed. NIR fluorescent blots are stable for months and can be re-scanned without any loss of signal.

Employing NIR technology in other applications

NIR fluorescence detection has been an enabling technology for a variety of applications beyond Western blotting.

Figure 3. Anti-EGFR and anti-phospho-EGFR antibody specificity in A431 cells. Two-fold serial dilutions of unstimulated and EGF-stimulated A431 cell lysates are shown in two-color images collected with an Odyssey Infrared Imaging System. Single color images (B and C) can be overlaid (A) to show both total protein and phosphorylated protein (yellow indicates overlapping red and green signals). The mobility shift caused by phosphorylation is visible (A) as indicated by the red bands above the yellow bands. Click here to enlarge.

In-cell Western assays. An immunocytochemical assay, termed the In-cell Western (LI-COR), has been developed that uses NIR fluorescence to detect and quantify proteins in fixed cells. Detecting proteins in their cellular context further increases quantification precision. Proteins in fixed, cultured cells are detected directly in microplates, which yields higher throughput compared to Western blotting and eliminates typical Western blotting steps such as cell lysate preparation, electrophoresis, and membrane transfer. In-cell Western assays are widely applicable and have been used in a variety of studies including cell signaling, analysis of G-protein coupled receptor function, neurology, and oncology.^{(3, 4, 5)} A novel variation of the In-cell Western procedure has been used to study the internalization and recycling of the cannabinoid receptor 1 (CB1), a GPCR class receptor.⁽⁶⁾

In Vivo imaging. NIR fluorescent detection has proven useful for in vivo imaging of small animals. Low tissue autofluorescence at 800 nm makes it possible to use probes with NIR labels to image tumors and organs.⁽⁷⁾ In vivo imaging is an important tool for any research that uses animal models to study diseases, such as Alzheimer's disease.⁽⁸⁾

NIR detection has also been used for novel applications in tissue imaging,⁽⁹⁾ protein arrays,⁽¹⁰⁾ and EMSA analysis.⁽¹¹⁾

About the authors

Kristi L. H. Ambroz, Ph.D. is a Senior Scientist in Molecular Biology Product Development at LI-COR Biosciences. Her email address is Kristi.ambroz@licor.com.

References

1. Schutz-Geschwender, A. et al. Quantitative, two-color Western blot detection with infrared fluorescence. http://www.licor.com/bio/PDF/IRquant.pdf (25 Jul. 2006).

2. Picariello, L. et al.A comparison of methods for the analysis of low abundance proteins in desmoid tumor cells. Anal. Biochem. 354:205-212 (2006).

3. Chen, H. et al. A cell based immunocytochemical assay for monitoring kinase signaling pathways and drug efficacy. Anal. Biochem. 338:136-142 (2005).

4. Wong, S. K.-F. A 384-well cell-based phospho-ERK assay for dopamine D2 and D3 receptors, Anal. Biochem. 333:265-267 (2004).

5. Dickey, C. A. et al. Development of a high throughput drug screening assay for the detection of changes in tau levels—proof of concept with HSP90 inhibitors. Curr. Alzheimer Res. 2:231-238 (2005).

6. Miller, J. Tracking G protein-coupled receptor trafficking using Odyssey Imaging. http://www.licor.com/bio/PDF/Miller_GPCR.pdf (25 Jul. 2006).

7. Houston, J.P. et al. Quality analysis of near-infrared fluorescence and conventional gamma images acquired using a dual labeled tumor targeting probe. J. Biomed. Optics. 10:054010-1-11 (2005).

8. Skoch, J. and Bacskai, B. The LI-COR Odyssey as a near-infrared imaging platform for animal models of Alzheimer's disease. http://www.licor.com/bio/PDF/MassGen.pdf (25 Jul. 2006).(2004)

9. Hawes, J. et al. GalR1, but not GalR2 or GalR3, levels are regulated by galanin signaling in the locus coeruleus through a cyclic AMP-dependent mechanism. J. Neurochem. 93:1168-1176 (2005).

10. Yeretssian, G. et al. Competition on nitrocellulose-immobilized antibody arrays: From bacterial protein binding assay to protein profiling in breast cancer cells. Mol. Cell. Proteomics 4:605-617 (2005).

11. Geddie, M. et al. Rational design of p53, an intrinsically unstructured protein, for the fabrication of novel molecular sensors. J. Biol. Chem. 280:3564-3566 (2005).