Proteomics, Genomics & Microarrays
Data Normalisation - An analytical approach for reliable western blot analysis
Deanna Woo, Katherine Schaefer, Ken Oh, Bio-Rad Laboratories Inc
Western blotting is a fundamental technique used widely across the life sciences sector by supporting the detection, identifi cation, and quantitation of target protein(s) within a complex sample. Its affordability, specifi city and widespread accessibility makes this technique prevalent in protein biology research by offering critical insights into protein expression, localisation and post-translational modifi cations. However, to obtain reliable western blot results and data integrity requires that researchers must follow best practices for data normalisation. This analytical process facilitates meaningful and accurate comparisons of different samples, minimising variability from experimental errors and sample types [1].
Data Normalisation enables accurate
comparison of changes in protein expression By facilitating precise comparisons of changes in protein expression levels within a given sample set, western blot data normalisation promotes reliable reporting of experimental outcomes, fostering scientifi c progress and guiding future research [2]. In other words, researchers should be confi dent that variation observed between target protein abundances and different samples is the result of the actual biology and not due to the data being skewed by common errors, such as inconsistent sample preparation, uneven protein transfer, or pipetting errors.
Data normalisation using loading controls is essential for unbiased western blot experiments, eliminating process errors that could lead to variations in sample amounts across lanes. Although experiments should aim to minimise such inconsistencies through proper design and execution, data normalisation corrects for any non-biological differences between test samples and target proteins prior to comparison.
Western Blot normalisation methods
Despite the different approaches available for researchers to normalise their data, there has been a growing crisis of confi dence within the scientifi c community that has led many scientifi c journals to increase the stringency of their publication guidelines [3]. Researchers should thoroughly evaluate available normalisation methods and associated guidelines surrounding their use. Regardless of the chosen technique, achieving reliable results hinges on establishing a linear detection range in data normalisation, to demonstrate that any protein signal changes align proportionally with the amount of protein contained within test samples run on the same blot [1].
There are two main types of normalisation that can be applied when performing western blots: single protein detection normalisation and total protein normalisation (TPN) - each with its own set of advantages and potential drawbacks for researchers to consider.
Single protein detection normalisation
Single protein detection uses proteins, such as ‘housekeeping’ proteins (HKPs) (i.e., internal normalisation standards) or exogenous proteins as loading controls [4]. Immunoprobing these proteins determines their abundance, serving as a proxy for the entire protein population within a sample.
To qualify as a loading control, an internal protein must meet specifi c criteria: it cannot be your target protein and must exhibit abundant, ubiquitous, and consistent expression within the sample. HKPs, such as β-actin, GAPDH, and β-tubulin, are commonly chosen as internal normalisation standards due to their involvement in basic cellular functions and widespread expression. Consequently, antibodies for them are widely accessible, and their expression is easily detectable.
To ensure the reliability of housekeeping protein (HKP) expression, it is crucial to validate positive and negative controls experimentally, confi rming consistent expression levels across sample populations, and establishing the linear range of detection [5]. This latter step is essential because while HKPs are typically abundantly expressed, target proteins often exhibit low expression levels. Overloading samples to detect low-abundance target proteins can saturate HKP reference bands beyond the linear range, undermining their function as loading controls and potentially leading to inaccurate detection.
Despite their ubiquitous expression, recent studies have also shown that the expression levels of HKPs themselves can vary across different cell types and are infl uenced by different biological factors, including the cell cycle, cell density, developmental cycle, tissue type, subject age, and response to treatment [6]. In addition, HKP expression has been shown to be upregulated in certain cancers (Figure 1) [7]. As a result, it is important for researchers to validate their HKP for consistent expression across different sample types and experimental conditions. This can be achieved by conducting a linear range determination test, which may involve several rounds of primary and secondary antibody dilution ratio optimisation.
Figure 1: Differences between HKP expression and total protein staining between tumour (T) and non-tumour (N) cancerous tissues. Taken from Hu et al [7].
Exogenous protein normalisation is a less common technique for single protein detection, whereby a known quantity of purifi ed or recombinant protein is added into the samples. Given that the primary antibody is capable of solely detecting this exogenous protein, this type of control offers the ability to precisely control the amount of protein added and ensures consistency across all experimental conditions. Whilst control tests are also necessary for exogenous proteins, they bypass the need for an expression level test.
When using an exogenous control, however, it is important to take the point at which the protein is introduced into consideration. For example, when the protein is introduced after sample extraction from cells or tissues, it can only be used to normalise experimental error introduced through sample loading and gel-to-membrane transfer and not for variation in sample preparation.
Total Protein Normalisation overcomes
linearity challenges in immunodetection TPN provides an alternative approach whereby all sample proteins are visualised, and total protein levels are quantifi ed. This eliminates the reliance on a single control protein to represent the entire protein population. Instead, TPN utilises the total protein abundance measured from the blot as the basis for normalisation.
Colorimetric dyes such as Ponceau S, Coomassie Blue and Amido Black can be measured via densitometry or imaged with a white light source of a digital imager, whilst stains such as Sypro RUBY can be measured via a fl uorescent light source from a digital imager. As Ponceau S and Sypro RUBY are reversible dyes, staining can be performed prior to immunodetection, but staining and destaining steps are time sensitive so researchers must perform these steps with care. On the other hand, Coomassie Blue and Amido Black are irreversible stains and can therefore be used after immunodetection steps [2].
Total protein stains are typically less sensitive than antibody-based immunodetection used for detection of HKPs, overcoming issues related to oversaturation and therefore demonstrate good linearity within a typical loading range of 10–50 μg of cell lysate enabling a more accurate determination of low-abundance target proteins. In addition, normalising target protein levels against total protein expression prevents results being affected by variable HKP expression (Figure 1). TPN is also less likely to be impacted by variability in the reference signal, which is more likely to occur using HKP normalisation due to the need for stripping and reprobing steps. Consequently, despite the historical precedence of HKP normalisation, there is increasing advocacy for TPN as the method of choice for normalisation in western blotting.
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48
