Best Practices for Quantitative Western Blotting
Western blotting, now a ubiquitous and routine technique used in life science research and medical diagnostics laboratories, was first described as a qualitative protein-analysis method almost 40 years ago.1,2 Earlier Western blot protocols were cumbersome and required protein radiolabeling for detection, adding an element of danger to an otherwise-routine assay. Understandably, this resulted in a push to develop non-radioactive alternatives with improved safety, flexibility, and detection sensitivity. Although a reliable and wide-ranging protein detection method, qualitative Western blotting is limited by its inability to accurately quantitate changes in protein expression.
Efforts to address the shortcomings of qualitative Western blots have produced not only a streamlined qualitative method, but also advanced Western blotting as a quantitative tool.
In Western blotting, proteins are classically represented as laddered bands, where the observed density of each band proportionally correlates to protein abundance. However, a signal-emitting tag (preferably visualized and analyzed by digital imaging) is required to determine ratiometric changes of a particular protein’s abundance. This prerequisite for higher imaging sensitivity is characteristic of quantitative Western blotting, and it is the added procedural-step that necessary to overcoming the inherent limitations of the traditional qualitative method.
Qualitative information may be sufficient for proof-of-concept studies where the experimental goal is to obtain a yes or no answer for protein-presence. However, quantitative Western blotting has become essential to credibly implicating the involvement of a particular protein in a physiological response or disease pathology. Quantitative Western blotting methods define the amount of protein present in either relative or absolute terms, where protein abundance is determined based on band intensity and is resolved by two principle categories of detection: chemiluminescence or fluorescence.
Relative quantitation is dependent on the proper implementation of baseline references, commonly a housekeeping protein or total-transferred protein. By obtaining signal values for known amounts of a purified protein and generating a standard curve based on serial dilutions of a known quantity of, for example, bovine serum albumin, the absolute abundance of a particular protein of interest can be determined.3 Both absolute and relative quantitation methods strive to achieve the same goal: to precisely measure changes in protein abundance.
Housekeeping proteins are proteins that are commonly and constitutively expressed in the experimental cell or tissue model (e.g., actin, tubulin, GAPDH).4 These proteins are generally unaffected by experimental manipulations and are therefore frequently used as Western blot loading controls for relative protein quantitation. Their emitted signal is interpreted to be proportional to the amount of protein loaded in the corresponding gel lane and is used to determine the fold change of the target protein(s). However, this approach requires two fundamental assumptions: 1) that the housekeeping protein quantity is equal across all experimental conditions, and 2) that it has a linear spectrum of detection to the target protein(s).3 Said differently, this approach is often argued to, at best, reliably confirm protein loading success and volume equivalency across all sample-containing lanes.
Alternatively, total-protein staining has emerged as a valuable method of protein normalization that both verifies the amount of protein loaded and the transfer efficiency from gel to membrane. This relative quantitation approach labels all proteins on the gel or membrane using various commercially available stain-based or stain-free methods, from which protein abundance can be quantitated by either choosing several bands or an entire lane containing a ladder of bands for densitometry analysis.3,5 Total-protein staining has the advantage of not requiring antibodies for quantitation and is therefore not subject to experimental manipulation. In addition, because it also verifies protein transfer efficiency, total-protein staining serves as an internal trouble-shooting control.
Some research applications require the quantitation of a protein of interest in absolute terms rather than relative. To obtain absolute protein-mass measurements, the appropriate purified protein standards, which have been compared to and calibrated against a purified protein of known mass (e.g., bovine serum albumin), are required.3 A standard curve can be generated using densitometry data from multiple protein concentrations, allowing for bands to be absolutely compared against those bands representing a known quantity of the purified protein of interest. Although a highly precise method that is likely to produce new insights into protein function and cell-signaling processes, absolute quantitation is the more time-consuming option and may reduce throughput in conventional academic laboratories.
Regardless of your quantitative approach (relative or absolute), the primary path to densitometry analysis is identical; it is essentially made up of the traditional qualitative Western blotting protocol with a signal-emitting tag to the target protein as a final step. A look at the countless number of Western blots published since the technique’s inception illustrates the evolution of detection methods to favor chemiluminescent- and fluorescent-dependent tagging over radioactive protein-labeling. Chemiluminescence is an indirect protein-detection method that depends on the conjugation of a secondary antibody to an enzyme, usually horseradish peroxidase (HRP). With the addition of an enhancing reagent containing hydrogen peroxide, luminol is oxidized to produce transient photons for detection by film or digital imaging.2 Fluorescence, as its name implies, utilizes fluorescent dyes as secondary-antibody tags for target protein detection by an imaging workstation equipped with a combination of excitation and emission filters.
Fluorescent labeling eliminates the need for a hydrogen peroxide-containing substrate to generate a detectable light signal. Moreover, unlike chemiluminescent signals, fluorescently produced emissions are not transient and will not fade over time if stored properly. These procedural advantages not only simplify the data collection process, but also provide for a more direct, stable, and consistent choice of detection and analysis. Furthermore, multiplexing, the ability to detect multiple proteins at once, requires only that the primary antibodies originate from different species and that the corresponding (secondary antibody) fluorescent dyes are easily differentiated on a wavelength-based detection spectrum. Multiplexed analysis can visualize an ever-expanding number of fluorochromes at once, representing an added time-saving advantage of fluorescent labeling over chemiluminescence. In contrast, the only way to detect and quantitate multiple proteins using HRP-conjugated chemiluminescent secondary antibodies is through a series of laborious membrane-stripping and re-probing steps using primary antibodies targeting each protein of interest one by one. Hence, the time and reproducibility constraints of this approach are readily recognizable. In short, multiplexing via fluorescent labeling accelerates time to results and streamlines the leap from qualitative to quantitative Western blotting, a process that may otherwise be viewed as tedious and multi-faceted.
To select the right imaging method for your protein quantitation purposes, consider the following: the detection system sensitivity with regard to the signal type emitted, the detection range, signal longevity and associated background noise, and multiplexing potential. Note that detection and quantitation using fluorescence is dependent on access to a committed digital workstation equipped with the right light sources and excitation and emission filters. Furthermore, whereas transfer membrane material may not be of primary concern to the researcher interested in qualitative or quantitative Western blotting by way of chemiluminescence, note that both nitrocellulose and polyvinylidene fluoride (PVDF) are not without flaws if using fluorescence-based protein quantitation. To minimize the autofluorescent background noise produced by traditional membrane materials, custom low-fluorescence membranes are commercially available for fluorescence-based protein detection studies. Your choice of normalization, staining, and Western blot protocol will depend on the level of detection and quantitation precision desired, combined with the available laboratory resources for detection and analysis.
Quantitative Western blotting is instrumental to accurately delineating the biochemical attributes of a protein, and, therefore, upstream cell-signaling processes important in both normal and disease physiology. For this reason, quantitation of protein expression is quickly becoming a standard requirement for publication in many peer-reviewed scientific journals aiming to improve data reproducibility by other researchers. Measuring protein abundance is critical to, for example, precisely modeling 3-dimensional structures of multiprotein complexes and for protein phosphorylation stoichiometry, each relevant to a spectrum of biological pathways. In addition, antibody-binding affinities are more accurately compared if quantitated, as are various preliminary analyses of human serum-antibody titers, among other applications. Having an understanding of appropriate standards, reagents, and methods, in addition to what signal-detection imaging techniques are available is key to proper protein quantification and reproducible data.
References
1. Renart J et al., “Transfer of proteins from gels to diazobenzyloxymethyl-paper and detection with antisera: A method for studying antibody specificity and antigen structure,” PNAS 76:3116-3120, 1979.
2. Towbin H et al., “Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications,” PNAS 76:4350-4354, 1979.
3. James K, “An analysis of critical factors for quantitative immunoblotting,” Sci Signal 8:rS2-rS11, 2015.
4. McDonough AA et al., “Considerations when quantitating protein abundance by immunoblot,” Am J Cell Physiol 308:C426-C433, 2015.
5. Posch A, Taylor CS, “The design of a quantitative western blot experiment,” Biomed Res Intl 2014:361590, 2014.