Improved Western Blot Detection using CCD Digital Imaging

Comparison of cooled-CCD digital imaging versus X-ray film for sensitivity, dynamic range and signal linearity with enhanced chemiluminescence (ECL)

by Eric Hommema, Ph.D.; Nikki Jarrett, M.S.; Steve Shiflett, Ph.D.; Suk Hong, Ph.D.; Priya Rangaraj, Ph.D.; Brian Webb, Ph.D. - 06/26/13

Chemiluminescent Western blotting is a popular protein detection method for studying proteins. The typical protocol for measuring chemiluminescent signal exposes the Western blot to laboratory-grade X-ray film, but digital imaging using a cooled-CCD camera is proving to be a more functional choice for researchers. The Thermo Scientific myECL Imager offers digital imaging with a number of improvements over film detection, including greater signal sensitivity, linearity and dynamic range.

Western blot analysis has been used as a qualitative assessment of the presence of a specific protein of interest in a sample with a target-specific antibody. Relative differences in protein amounts can be assessed by the signal intensity as it typically correlates to the protein amount in the blot. Scanning of the film with a basic document scanner and the use of image analysis software has been used to extract some quantitative information; however, densitometry from the scanned X-ray film image can be challenging (Ref.1) and chemiluminescent western blot analysis is generally regarded as not being quantitative.

An alternative method to X-ray film for capturing luminescent signals from western blots is to use a cooled-CCD camera. Charge-coupled device (CCD) cameras use a light-sensitive silicon chip that converts photons into digital signals. Imagers with early generation CCD cameras designed for low-light applications were not capable of matching the speed or sensitivity of X-ray film.

Recent improvements in CCD technology have enabled the development and commercialization of sensitive, cooled-CCD cameras with higher light-capturing performance than film. High-performance CCD cameras cool the silicon chip to sub-zero temperatures to reduce dark current, which produces background noise. To enhance detection sensitivity further, the pixels, or light capturing units of the CCD chip, can be combined or binned. Binning increases the size of each pixel, which effectively increases the amount of light collected in the pixel area.

Figure 1. How binning can improve sensitivity.

Combining high-sensitivity CCD camera and digital image analysis software, researchers who perform chemiluminescent western blot analysis can have the ability to extract more accurate qualitative and quantitative information than was previously available with film. In this article, we demonstrate that the myECL Imager is superior to X-ray film in sensitivity, dynamic range and signal linearity.

To compare CCD camera and X-ray film imaging, we examined signals produced by a NIST-traceable Luminometer Reference Microplate (Harta Instruments, #RM-168) and by our own chemiluminescent Western blots containing serially diluted samples spanning the detectable range for imagers and film. We captured signals digitally using the myECL Imager and by exposure to X-ray film. (Developed films were then scanned at high resolution to produce digitized versions for analysis.) Finally we analyzed the images using the Thermo Scientific myImageAnalysis Software.

Greater sensitivity and dynamic range compared with X-ray film

Images acquired on the myECL Imager are more sensitive than film. Seven of the eight reference plate spots are visible and distinguishable above background on the 300-second exposure acquired by the myECL Imager (Figure 2). In contrast, only five of the eight spots are visible on a 300-second exposure to film. Densitometry of these images shows a 1000-fold dynamic range (maximal signal intensity/minimal signal intensity) in the myECL Imager compared to a 1.5-fold dynamic range for the film, where dynamic range is calculated from the density of spot 1 divided by the density of spot 8 (Figure 2; Table 1). A qualitative assessment of the western blot examples shows that the myECL Imager detects an additional band of the HeLa lysate dilution series when compared to film (Figure 3).

Table 1. Densitometry values for 300-second exposures of reference plate. Results show a 1000-fold (45017/45) dynamic range for the myECL Imager and a 1.5-fold (62734/41286) dynamic range for X-ray film.

Signal linearity

Signal linearity is an important factor for quantitative measurement. When the relationship between signal intensity and sample quantity is linear, unknown sample amounts can be calculated using a simple linear regression model. Visually, western blot exposures on X-ray film may appear to have a broad range of signal linearity; however, densitometry indicates signals on film have narrow linear dynamic range. There are signal linearity differences between the myECL Imager and X-ray film (Figure 4). Images acquired on the myECL Imager show a linear increase in signal for GFP-HA amounts up to 1ng, whereas the linear signal increase on film images is up to 0.125ng. Signal saturation was reached on film at 0.125ng of GFP-HA as opposed to 2ng on the myECL Imager.

High pixel binning in the myECL Imager increases imaging sensitivity

Pixel binning is the ability to combine signals from adjacent pixels to increase the light capturing ability (sensitivity) of a CCD chip. Conversely, pixel combination will result in a larger pixel, which then results in a lower resolution of the captured image.

Each binning increase results in an increase in light capturing ability or sensitivity (Figure 5). Adjustment of the binning setting allows the researcher to find the desired balance between sensitivity, resolution and exposure time. We found the default 3 × 3 setting generally offers the best combination of sensitivity and resolution when using the myECL Imager.

Life science researchers have traditionally used X-ray film for Western blot analysis. This provides qualitative data for protein detection, but film limits protein quantitation because of narrow dynamic range and low sensitivity. Additional drawbacks of film, such as dealing with hazardous developing solutions, maintaining film processors and needing a dark room, make digital imaging easier and more practical. The myECL Imager, with a broad and linear dynamic range, allows researchers to extract quantitative data from Western blots with the additional benefits of convenience and sensitivity.

Luminometer reference plate time course

A Luminometer Reference Microplate (Harta Instruments, #RM-168) was used to demonstrate the imaging speed and sensitivity of the myECL Imager in comparison to standard X-ray film (Thermo Scientific CL-XPosure Film, Part No. 34090). The reference plate contains eight lights (spots) of varying, known signal intensities and is typically used for luminometer calibration. Images were captured on film and on the myECL Imager using all five binning settings (1 × 1, 2 × 2, 3 × 3, 4 × 4, and 8 × 8) at exposure times of 10, 30, 60, 120 and 300 seconds. The film was developed in a Konica Minolta™ SRX-101A Film Processor with developer reagents. To create digital images of the film for analysis, each film was scanned with an Epson™ 4990 Photo Scanner as a 300 PPI (pixels/inch), 16-bit grayscale TIFF image to match the image output from the myECL Imager. Image analysis was performed using the Thermo Scientific myImageAnalysis Software. For analysis, a manual region of equal size was placed over each of the eight reference plate light spots and the density (pixel intensity/region area) was plotted against the observed RLU (Relative Light Units) based on the manufacturer’s Certificate of Analysis. For publication, the white level contrast was adjusted on the images acquired with the myECL Imager to display low-intensity pixels.

Western blots

Fifty micrograms of HeLa lysate was serially diluted (2-fold in each lane) in 4X LDS Sample Buffer (Product # 84788) supplemented with DTT at a final concentration of 50mM. The amount of sample loaded onto the gel was 50, 25, 12.5, 6.25, 3.13, 1.56, 0.781, 0.391, 0.195 and 0.0977µg of lysate. Samples were loaded onto 4-20% Thermo Scientific Precise Tris-Glycine Gels (Part No. 25249) and transferred to Thermo Scientific PVDF Transfer Membrane (Part No.88518) using the Thermo Scientific Pierce G2 Fast Blotter (Part No. 62288) and 1-Step Transfer Buffer (Part No.84731). The blots were probed with mouse anti-PLK1 (Ab No. MA1-848) or rabbit anti-Cyclophilin B (Ab No. PA1-027A) according to instructions supplied with the Thermo Scientific Fast Western, SuperSignal West Dura Kit (Part No. 35070) for 1 hour. The anti-mouse or anti-rabbit fast western optimized horseradish peroxidase (HRP) reagents were added for 10 minutes, followed by four 5-minute washes with Thermo Scientific Fast Western Wash Buffer. SuperSignal West Dura Substrate was added for 5 minutes before imaging on film, followed by image capture using the myECL Imager. Signal loss between imaging events (approx. 5 minutes) was negligible (data not shown). Images were acquired on the myECL Imager in Chemi mode with the default binning setting (3 × 3). X-ray film was developed and scanned as described above.

In vitro protein expression

The Thermo Scientific 1-Step Human In Vitro Protein Expression Kit (Part No. 88882) was used to express green fluorescent protein (GFP) with a C-terminal HA tag. XhoI and NdeI restriction sites were added to the 5’ and 3’ ends, respectively, to the open reading frame of GFP from Pontellina plumata. This fragment was inserted into the multiple cloning site of the pT7CFE-CHA expression vector, which contains the upstream transcription and expression elements necessary for in vitro translation and a C-terminal HA (Human Influenza Hemagglutinin-YPYDVPDYA) tag. Protein expression was performed as described in the kit protocol. After protein expression, the concentration of GFP was determined by measuring fluorescence (482ex/502em) in a Safire™ Microplate Reader and compared to a GFP standard.

Signal linearity

A serial dilution of 2.0 to 0.004ng of in vitro-expressed GFP was loaded onto 4-20% Precise Tris-Glycine Gels and transferred to nitrocellulose membrane using the Pierce G2 Fast Blotter and 1-Step Transfer Buffer. The blot was probed with Thermo Scientific Mouse anti-HA Antibody (Part No. 26183) according to instructions supplied with the Fast Western, SuperSignal West Dura Blot Kit for 1 hour. The anti-mouse fast Western optimized HRP reagent was added for 10 minutes, followed by four 5-minute washes with Fast Western Wash Buffer. SuperSignal West Dura Substrate was added for 5 minutes before imaging on film, followed by the myECL Imager. Signal loss between imaging events (approx. 5 minutes) was negligible (data not shown). For analysis, a manual region of equal size was used over each of the GFP-HA bands in all 10 lanes plus a representative background region. The Global Background Subtracted Density (pixel intensity/region area) was plotted against nanograms of GFP-HA. For publication, the white-level contrast was adjusted on the images acquired with the myECL Imager to display low-intensity pixels.

1. Gassmann, M., Granacher, B., Rohde, B. and Vogel, J. (2009) Quantifying Western blots: Pitfalls of densitometry. Electrophoresis 30:1845-55.

This article was first printed as Application Note 1602635, April 2013.