Introduction
Until recently, all microplate spectrofluorometers were filter-based instruments equipped with a limited number of excitation/emission filter pairs. For a given application, the wavelengths of the filter pair did not necessarily correspond to the optimal excitation/emission wavelengths of the fluorophore. The best available filter was often 20, 30 or even 50 nm away from the optimal wavelength of the fluorophore. If the existing filters could not be used, it was necessary to purchase and install different ones.
With the introduction of the SPECTRAmax GEMINI, it is no longer necessary to use bandpass filters or to settle for suboptimal wavelength compromises. The GEMINI contains both excitation and emission monochromators such that any combination of wavelengths between 250 and 850 nm is easily obtained. Given the flexibility, the user naturally wishes to choose the best possible combination. For a particular fluorophore, the choice will depend on the location of its excitation and emission maxima and the separation between them (Stokes’ shift). If the two wavelengths are far apart the choices are easy: the excitation and emission maxima are selected, along with an appropriate cutoff filter to remove residual excitation illumination and optimize sensitivity. If the two wavelengths are close together, the selection/optimization process is less straightforward, because of the need to minimize excitation light carryover. This application note gives a basic procedure for optimization of excitation and emission wavelengths of the SPECTRAmax GEMINI microplate spectrophotometer. An example is given for application of the procedure to a fluorophore with a relatively large Stokes’ shift (quinine).
Overview
The first step in developing fluorescence analysis methodology is to select the excitation wavelength. The next step is to select the optimum combination of emission wavelength and cutoff filter that gives the highest possible signal/background ratio. Generally, samples containing 10-8 M to 10-6 M fluorophore will give sufficient signal. For optimal signal/background results, data should be acquired with the highest PMT (voltage) setting. Thus the fluorophore concentration should be low enough such the emission scans can be done with high PMT without saturating the detector.
Procedure
A: Microplate Preparation
- Put 200 µL sample into one or more wells of a microplate and 200 µL buffer or solvent into one or more other wells.
B: Selection of Excitation Wavelength
- In SOFTmax PRO, set up a plate section for an excitation scan with no cutoff filter and use the medium PMT setting.
- Set the emission wavelength based on the tentative value from the literature (or from a customary filter set used to measure your fluorophore). If there is no literature value available for the emission wavelength, see Step 5 of Section F, below.)
- Set the excitation scan to start/stop approximately 30 nm below/above the tentative excitation value obtained from the literature (or customary excitation filter) and use a step increment of 1-2 nm.
- Perform the scan and note the excitation wavelength at the peak (lambda max), as well as the maximum RFU value.
- Select the optimal excitation wavelength.
- If the excitation lambda max and emission wavelength are separated by more than 80 nm, select lambda max as the excitation wavelength.
- If the excitation and emission wavelengths are less than 80 nm apart, Molecular Devices recommends that you increase the separation by choosing the lowest excitation wavelength that gave 90% maximal RFU.
C: Emission Scan without cutoff filter
You almost certainly will need a cutoff filter, but this scan is useful to help you understand and interpret the subsequent emission scan(s) and thus help you to select a cutoff filter.
- In SOFTmax PRO, set up a plate section for an emission scan with no cutoff filter and use the high PMT setting. Set the excitation wavelength to the value determined in Step 5 of the excitation scan above
- Set the emission scan to start/stop approximately 50 nm below/above the tentative emission value obtained from the literature (or existing filter pair) and set the step increment to 1-2 nm. Perform the scan.
D: Selection of Emission Wavelength and Cutoff Filter
Cutoff filters block light below their cutoff wavelength and transmit light above their cutoff wavelength. The nominal cutoff value is the wavelength at which 50% transmission occurs. You should select an emission cutoff filter to block as much residual excitation light as possible without unduly reducing the fluorescent signal from your sample. The cutoff value should be near the maximum emission wavelength and at least 35 nm greater than the excitation wavelength. (Preferably, the cutoff value should be between the excitation and emission wavelengths, but in cases of very short Stokes’ shifts, the best cutoff filter may actually be higher than the emission maximum!) SPECTRAmax GEMINI’s wavelength choices for cutoff filters are 420, 435, 455, 475, 495, 515, 530, 550, 570, 590, 610, 630, 665 and 695 nm.
- Set up a plate section for a second emission scan as in Step C above, except for the added Emission Cutoff Filter. Perform the scan and note the wavelength that gives maximum emission. Check the plot from the blank to be assured that there is no unexpected fluorescence in that region. The wavelength giving the maximum emission for the sample will then be your Optimal Emission Wavelength for the chosen cutoff filter.
E: Emission scan with different cutoff filter (optional)
If you are unsure of your choice of cutoff filter, you may wish to repeat the emission scan with one or more different cutoff filters (or excitation wavelength/cutoff filter combinations). If desired, create signal/background plots to aid in the decision process (See. Maxline App. Note #31).
F: Comments
- The wavelength with maximum emission is not necessarily the wavelength with highest signal-to-background ratio. Depending on possible fluorescence contained in the blank, the Optimal Emission Wavelength may be shifted away from the emission maximum.
- Once an emission wavelength is designated, the “Autofilter” feature will generally select the same cutoff filter wavelength as that used in the above optimization method.
- For emission wavelengths less than 420 nm, optimal emission and excitation wavelengths will best be determined by experimental iteration. Try emission and excitation wavelength combinations with the 420 nm cutoff or with no cutoff filter. Similarly for excitation wavelengths greater than 660 nanometers, try emission and excitation wavelength combinations with the 695 nm cutoff filter or with no cutoff filter.
- If the fluorophore has an extremely narrow Stokes’ shift, you may wish to repeat the emission scans, using an excitation wavelength even lower than the 90% max value.
- If the emission wavelength is not known, select a tentative emission wavelength to be about 50 nm greater than an absorbance maximum of the fluorophore. The absorbance maximum may be determined by performing a spectral scan (on a more concentrated fluorophore solution) in a UV/Vis spectrophotometer. If all else fails, you may need to make a guess and go through several iterations of Sections B-E.
Examole of optimization procedure using 1.24 µM quinine sulfate dissolved in 0.01 N HCl
For the excitation scan, the initial emission wavelength was arbitrarily set to 500 nm in order to demonstrate that the method can work even if the initial emission wavelength is not close to the emission maximum. The excitation scan between 300 and 400 nm revealed an excitation peak at approximately 350 nm (Figure 1).
Figure 1: Excitation scan of 1.24 µM quinine with emission wavelength set to 500 nm.
The separation (~150 nm) between the excitation wavelength and emission peak was greater than 80 nm, so the chosen excitation wavelength was 350 nm. Emission scans were then performed, first without a cutoff filter and then with three different cutoff filters (Figure 2). Without a cutoff filter, the emission maximum was approximately 500 RFU at 443 nm. The cutoff filters progressively shifted the peak to the right and decreased the maximum signal. The 455 nm cutoff filter especially blocked most of the peak.
Figure 2: Emission scans of 1.24 µM quinine (excitation set to 350 nm). The plots from left to right are: No cutoff filter, 420 cutoff, 435 cutoff, 455 cutoff.
The plot with the 420 nm filter was considerable higher than that with the 435 nm filter, but before selecting the 420 nm filter, it was important to check the scans of the blank solution (0.01 N HCl) (Figure 3). As expected, the scan with no cutoff filter had the highest signal because of scattered excitation light. The 420 nm plot was higher than the 435 nm plot until approximately 460 nm, above which they were essentially identical.
Figure 3: Emission scans of 0.01 N HCl (excitation wavelength set to 350 nm)
The final decision process was facilitated by preparing plots of signal/ background ratio as a function of emission wavelength. (The formulas necessary for these plots can be found in the SOFTmax PRO Formula Reference Guide and also in MAXline App. Note No. 31.) The signal/background plot for the 420 nm cutoff filter is shown in Figure 4. (The plot from the 435 nm filter was slightly, but consistently lower.) The maximum ratio was located at approximately 470 nm - higher than the location of the peak fluorescence, but coinciding with a dip in background signal seen in Figure 3.
Figure 4: Plots of signal/background for emission scans with 420 nm cutoff filter.
The optimal wavelength settings for analyzing quinine sulfate on the GEMINI were chosen to be 350/470 (Ex/Em) with a 420 nm cutoff filter.