LICOR white logo

The Photochemistry of X-Ray Film

Before you head to the developer, it is important that you understand the variables that X-ray film detection introduces to your experiment.

Film’s response to light

  • Response of film to light exposure is not linear. The linear relationship holds true across a narrow range1 of values (called dynamic range), but becomes inaccurate outside that range.
  • Figure 1. Characteristic curve of X-ray film (adapted from Kodak, 2007). The "toe" region of the curve indicates very low exposures. The center or "straight line" region is the approximate linear response range, where OD is proportional to the log of the exposure. The "shoulder" region represents higher exposures and is relatively flat. The slope of the "straight line" region indicates the film's scale of contrast. A steeper curve indicates a shorter scale of contrast and narrower dynamic range.

  • Signals above and below the dynamic range cannot be accurately detected by film. For example, this occurs when strong and faint signals are detected simultaneously.
  • The tendency of strong bands to blur and spread on film obscures adjacent bands and makes quantitation difficult. This is especially problematic when stronger bands are located near faint bands, because the longer exposures required for detection of faint bands increase the spreading of stronger bands.
  • Figure 2. A strong signal quickly activates the majority of silver grains in that region. Film response slows and eventually saturates as new photons of light fail to activate new silver grains.

    Figure 3. Faint bands are not effectively detected. When the rate of light emission is too low to activate silver grains, the signals cannot be detected - even with long exposures.

    Figure 4. On film, strong signals blur and spread to obscure adjacent bands. ARNO protein was detected in cell lysates, using ECL Plus substrate and 90-sec film exposure. Spreading of bands is especially problematic when longer exposures are required to detect faint bands (right side of blot).

  • Unevenly distributed substrate, pooled substrate, and bubbles affect signal intensities. Strong bands with high concentrations of HRP enzyme may cause rapid local depletion of the substrate, generating central "white-outs" in strong bands where light can no longer be generated.
  • Figure 5. Substrate availability can affect signal intensity on chemiluminescent Westerns. The same pair of samples was loaded three times on a single blot (10 XX or 5 XX of C32 cell lysate; circled areas). SuperSignal West Pico substrate was used for detection, with 5 min film exposure. Non-uniform distribution of substrate across the blot made some signals noticeably stronger than others (yellow circle).

  • When using densitometry, it is important to remember that optical density is not a direct function of light generation by the chemiluminescent substrate2. Optical density is an indirect function that depends on the response of film to light, which is non-linear. The accuracy of densitometry is therefore dependent on the sensitivity, linear response range, and exposure time of the film.
  • Figure 6. Densitometry of film is limited by signal saturation. A) Purified AKT was detected on a Western blot with chemiluminescence and exposure to film. B) Signals were quantified by densitometry of the exposed film. The sigmoid curve indicates signal saturation and limited dynamic range using densitometry and film. Reprinted from Wang et al, 2011.9

The photochemistry of film has several inherent limitations that affect data output and can introduce error.

References

  1. Laskey, R.A. Efficient detection of biomolecules by autoradiography, fluorography or chemiluminescence. Methods of detecting biomolecules by autoradiography, fluorography and chemiluminescence. Amersham Life Sci. Review 23:Part II (1993).
  2. Baskin, DG and WL Stahl. Fundamentals of quantitative autoradiography by computer densitometry for in situ hybridization, with emphasis on 33P. Journal of Histochemistry and Cytochemistry 41(12):1767-76 (1993).