Evelyn McGown, Ph.D. and Anna Lam, B.S.
Molecular Devices Corporation, 8/99
Modern assays for proteolytic activity typically use synthetic chromogenic or flu-orogenic peptide substrates. Enzyme selectivity is obtained by choosing a peptide sequence uniquely recognized by the catalytic site of a particular enzyme. The chromophore or fluorophore is attached near the peptides cleavage site and enzymatic activity is detected by the increase in color or fluorescence. Chromogenic assays are easy in that one simply monitors the rise in absorbance above a (usually) low background. Fluorometric assays are much more sensitive than chromogenic assays, but are less straightforward. The substrate itself can be fluorescent, causing a high background. If the absorption spectra of the substrate and product overlap, the substrate can quench the signal from the product by absorbing either excitation or emission light. Therefore, both the excitation and emission wavelengths should be optimized to minimize interference from the substrate, while maintaining sufficiently high product fluorescence. This application note describes how to optimize a protease assay in the SPECTRAmax GEMINI micro-plate spectrofluorometer. We chose to use the protease caspase3 and a fluorogenic peptide substrate containing a coumarin derivative.
Coumarin derivatives are among the most popular fluorophores for peptide based protease assays because they are more sensitive and more watersoluble than other fluorescent derivatives.13 A 7amino4methylcoumarin (AMC) derivative was first used as a substrate for chymotrypsin1. Since then, AMC and the 7amino4trifluoromethylcoumarin (AFC) fluorophores have been used for various protease assays.14 In each case, cleavage of the fluorophore results in increased fluorescence, with a shift of excitation and emission maxima to longer wavelengths. AMC a nd AFC appear to give identical sensitivity.3
Caspase3 is one of the family of cysteinecontaining proteases that cleave their substrates at an aspartate residue.4,5 It appears to have several natural substrates6, including poly(ADPribose) polymerase, in which case the cleavage site is at the tetrapeptide AspGluValAsp (DEVD).4,5 Although caspases are well known for their key roles in programmed cell death (apoptosis), an extracellular caspase3 like activity has been recently discovered in suspensions of live cells.7
Several coumarinlabelled DEVD substrates have been used to measure caspase3 activity. Typically AMC4 or AFC7,8 is attached to the Cterminus, and acetyl (Ac) or benzyloxycarbonyl (Z) is attached to the amino terminus. Enzymatic activity results in the release of free AMC (or AFC) product. There are several commercial sources of these substrates, as well as caspase3 assay kits containing them (including Molecular Probes, Enzyme Systems Products, Clontech, R&D Systems, GenoTechnology, CalBiochem, Peptides International, BioRad, and Pharmingen).
This application note describes an assay for caspase3 in the SPECTRAmax GEMINI microplate spectrofluorometer using the substrate ZDEVDAMC.
MATERIALS AND METHODS
ZDEVDAMC substrate was purchased as part of Molecular Probes EnzChek Caspase3 Assay Kit #1 (catalog #E13183). Recombinant caspase3 enzyme was purchased from Calbiochem (catalog #235417), and 7Amino4Methylcoumarin (AMC) from Sigma (catalog #A 9891). Black 96-stripwell plates and white-96 well plates were purchased from Corning Costar.
All dilutions of the substrate and enzyme were made with 1X Reaction Buffer, (prepared by diluting 200 L of the 5X Reaction Buffer and 5 L of DTT, both included in the kit, with 795 L of deionized water.) Stock ZDEVDAMC (10 mM) was prepared by adding 520 L of DMSO into the vial of ZDEVD AMC. Working solutions of ZDEVDAMC stock were typically 1:100 dilutions of the stock, i.e. 5 L of the stock solution into 495 L of 1X Reaction Buffer. The stock enzyme was frozen in 2 or 4 L aliquots (to avoid repeated freeze/thaw cycles) and working enzyme solutions were prepared by diluting the stock 1:10 with 1X Reaction Buffer. The stock AMC solution was 2 mM in DMSO.
Background fluorescence of ZDEVDAMC
ZDEVDAMC substrate is fluorescent, with excitation/emission maxima of ~330 nm/390 nm.9 The excitation/emission maxima of the AMC product are ~342 nm/441 nm. Therefore, both the excitation spectra and emission spectra overlap. Figure 1 shows the substrate and product emission spectra, without (left) and with (right) a 435 nm emission cutoff filter. The substrate has a lower lambda maximum (~390 nm) than the product, and it is less intensely fluorescent, but its signal overlaps significantly with that of the product, even with the cutoff filter.
Because the substrate must be present in a relatively high concentration in the reaction mixture in order to maintain maximal enzyme activity (Molecular Probes recommends 100 M), it contributes a high background to the assay.
Effect of substrate on fluorescence of the product and selection of the excitation wavelength
The absorption spectrum of ZDEVDAMC substrate also affects the caspase3 assay. The substrate absorbs light at 342 nm, the excitation wavelength associated with maximal AMC signal (Figure 2.)
As a result, the substrate absorbs the excitation light and decreases (quenches) the fluorescence intensity of the product. The higher the substrate concentration, the more severe the quenching. The quenching effect of ZDEVDAMC is eliminated by employing a longer excitation wavelength at which substrate absorbance is minimized. This will cause some loss in product fluorescent intensity. Figure 3 shows the fluorescence emission (at 441 nm) of the AMC product in the presence of graded concentrations of substrate and with excitation wavelengths ranging from 345 nm to 370 nm.
The quenching effect was especially severe at excitation wavelengths 345 nm to 355 nm, and was still appreciable at 360 nm. However, it was essentially eliminated at excitation wavelengths greater than 364 nm. As the excitation wavelength increased, the fluorescence of the product predictably decreased. At an excitation wavelength of 368 nm, however, AMC fluorescence was decreased by only 40%.
Recommended instrument settings, choice of microplates, and estimated assay sensitivity
Based on the results shown in Figure 3, 368 nm was selected as the excitation wavelength. The next step was to select the emission wavelength and cutoff filter. To help in the selection process, (signal-background)/background1/2 was plotted versus wavelength. The plots were prepared from scans with no emission cutoff filters, with a 420 nm cutoff filter and a 435 nm cutoff filter (Figure 4.) The optimum emission wavelength appeared to be approximately 467 nm. The cutoff filters gave comparable results in that region and both were somewhat better than no cutoff filter. This was confirmed experimentally by running the enzyme assay and blanks at different emission wavelengths with multiple replicates and estimating limits of quantitation. Based on the results above, we recommend that Z-DEVD-AMC-based caspase-3 assays be run on the SPECTRAmax GEMINI with excitation/emission wavelengths set to 368/467 nm with a 435 nm or a 420 nm emission cutoff filter.
CHOICE OF MICROPLATES
White plates give a higher fluorescent signal (and background) than do black plates because the light is reflected, rather than absorbed. (The background is also higher because white plates are inherently fluorescent.) Figure 5 gives an example of a kinetic caspase3 assay run in both a black and a white microplate. The slope of the reaction in the white plate was ~2.7 times that in the black plate. The initial fluorescence (background) was also ~2.5 times higher in the white plate. Despite the higher background, the white plate gave lower detection limits than did the black plate. Using the criterion of three positive SDs of the minus enzyme blanks, the estimated limits of detection for the assay were ~0.02 ng and 0.10 ng recombinant enzyme per well in the white and black plates respectively.
Fluorogenic peptide substrates can interfere with the measurement of their own hydrolysis products. If the emission spectra overlap, the assay may have a high background. If the absorption spectra overlap, or if the absorbance of the substrate overlaps with the emission of the product, the substrate can quench the signal from the product. Therefore, both the excitation and emission wavelengths must be optimized to minimize interference from the substrate, while maintaining sufficiently high product fluorescence. SPECTRAmax GEMINI microplate spectrofluorometer facilitates the process by having dual monochromators that allow optimal excitation and emission wavelengths to be easily determined.
1. Zimmerman, M., E. Yurevicz and G. Patel. 1976. A new fluorogenic substrate for chymotrypsin. Anal. Biochem. 70: 258262.
2. Gray, C.J. and J.M. Sullivan. 1989. Synthesis of 7amino4methylcoumarin (AMC) derivatives and their hydrolysis by plant cysteine proteinases. J. Chem. Tech. Biotechnol. 46: 1126.
3. Smith, R.E., E.R. Bissell, A.R. Mitchell and K.W. Pearson. 1980. Direct photometric or fluorometric assay of proteinases using substrates contai ning 7 amino4trifluoromethylcoumarin. Thrombosis Research 17: 393402.
4. Nicholson, D.W. et al. 1995. Identification and inhibition of the ICE/CED3 protease necessary for mammalian apoptosis. Nature 376: 3743.
5. GarciaCalvo, M. et al. 1999. Purification and catalytic properties of human caspase family members. Cell Death and Differentiation 6: 362369.
6. Villa, P., S.H. Kaufmann and W.C. Earnshaw. 1997. Caspases and caspase inhibitors. Trends Biochem. Sci. 22: 388393.
7. Davis, D. and S. Wells. 1999. Caspase3like activity appears in buffer containing live intact cells. Bioradiations #103. (BioRad Laboratories; 18004246723).
8. Gurtu, V., S.R. Kain and G. Zhang. 1997. Fluorometric and colorimetric detection of caspase activity associated with apoptosis. Anal. Biochem. 251: 98102.
9. Molecular Probes Product Information sheet for EnzCheck Caspase3 Assay Kit #1.
This model for calculation of signal-to-noise ratio is appropriate when the background signal is high, such that its variance is much larger than other sources of error. For details, see Ingle, J.D and S.R. Crouch. 1988. Signal-to-noise expressions for emission and luminescence measurements. In Spectrochemical Analysis. Prentice Hall, Inglewood Cliffs, NJ. pp. 146-150.