Secondary Structure Circular Dichroism Analysis - Guidance on using CDNN CD Spectra Deconvolution Software
Introduction
CDNN is a program which can be used to analyze circular dichroism (CD) spectra. It was written by Dr. Gerald Böhm, Institut für Biotechnologie, Martin-Luther Universität Halle-Wittenberg, who has kindly agreed to let Applied Photophysics distribute the software. Please note that any publication of results calculated using CDNN should refer to the author (see CDNN manual on disk for details).
CDNN is just one of a large number of circular dichroism analysis tools which are available. The University of Medicine and Dentistry of New Jersey (UMDNJ) Circular Dichroism Facility website has a list of free CD analysis software. Applied Photophysics also recommends the excellent DichroWeb program at the Centre for Protein and Membrane Structure and Dynamics at the School of Crystallography Birkbeck College. A small list of alternative CD secondary structure analysis programs is provided at the end of this document. Other tools can be found by searching through Google.
CDNN is provided on the CD-ROM which was provided with your Chirascan circular dichroism spectrometer. If you do not have access to the disc, you can contact Applied Photophysics Technical Support at techsup@photophysics.com to request a copy of the software.
Installation
Installation of the CDNN software is straightforward. Run the SETUP.EXE program and follow the instructions. Please read the Software Licence Agreement. Remember, although the software is provided by Applied Photophysics it is Copyright © Dr. Gerald Böhm.
Getting Started
Start the CDNN software from the shortcut on the Start Menu. You will be presented with a dialogue (see figure 1).
Figure 1. CDNN CD Spectra Deconvolution Software
Click on File>Open… On first use, the program will open at the Sample Files folder (C:\Program Files\CD deconvolution\Sample files\). Select one of the sample files and click on Open. In this example, I have used the Lysozyme example file lysozym.txt. (A copy of the file lysozym.txt is available here.) CDNN will present the user with a dialogue asking the user to select the CD signal type (see figure 2).
Figure 2. CD Signal Type dialogue
All the examples in the Sample Files folder are converted to Delta Epsilon, the differential extinction coefficient (in this case further normalised for the number of amino acid residues, of which more later). Click on OK. The panel will disappear and you will be left with the start-up panel. Click on Deconvolute to calculate the contributions of the various components to the protein secondary structure.
Figure 3. CDNN CD Spectra Deconvolution Software
The units, delta epsilon, are per amino acid residue. Once you are familiar with the use of the program and what the results mean, move on to your own samples. Note, it is highly recommended that you read the notes that come with the program itself!
Using CDNN with your own results
Measure the circular dichroism (CD) spectrum of your protein and carry out any manipulations, such as averaging, baseline correction and smoothing. Note that baseline correction is essential to get reasonable results. Figure 4 shows some repeat spectra of lysozyme done on a Chirascan instrument, and displayed in the Pro-Data Viewer program. The experimental conditions are listed in Table 1 below.
|
|
|
Figure 5 (below) shows the resultant curve after averaging, baseline correction and smoothing.
Figure 5. Lysozyme spectrum, after averaging, baseline correction and smoothing.
CDNN requires a text-file input format, with wavelengths and associated CD values at 1nm intervals. It does not matter if you have measured the spectrum at a different step-size (the above spectra were measured at 0.5nm steps) because there is a conversion program that generates the correct input format. In Pro-Data Viewer Click on File>Save current plot… to call up the following window and choose CDNN files (*.txt) from the drop-down list under Save as type. Specify a filename and location and click on Save.
Launch CDNN as previously described and open the file that you have just saved. This example uses the lysozyme data described above. Note that, unlike the examples in Sample Files folder, these CD data are in millidegrees. Care is needed to ensure that the results are meaningful.
Figure 6. CDNN Signal Type with options
With the Millidegrees radio button selected, the Essential Parameter dialogue box appears. It is imperative that the information requested be supplied. For lysozyme measured under the conditions given above, the fields are correctly filled in. Click on OK and then Deconvolute.
Figure 7. CDNN Deconvolution results
Comparison of the example and real data will show that the results are broadly similar. There are small differences in the spectra and it is not therefore surprising that there are small differences in the calculated contributions of the various secondary structure components.
Dealing with uncharacterised proteins
Not all proteins are well characterised and for example, the molecular weight may not be known. Fortunately that does not prevent you calculating the secondary structure components. In brief it is sufficient to replace the molecular weight of the protein with the average molecular weight of the amino acid residues (generally accepted to be about 113) and to replace the actual number of amino acids with the number 1.
Figure 8. CDNN CD Signal Type options for an uncharacterized protein.
Click on OK and Deconvolute to generate an identical result to that using the correct molecular weight and the correct number of amino acid residues.
Normalising for concentration and number of amino acid residues in an unknown protein
Concentration, generally expressed in mg/ml for proteins, is numerically equivalent to concentration in g/l. The molarity of the solution is therefore concentration in mg/ml divided by the MW of the protein.
Molarity = conc(mg/ml) / MW
A commonly accepted value for the MW of an amino acid residue is 113. The number of residues in a protein is therefore its molecular weight divided by 113.
Number of amino acid residues = MW/113 = N
If your protein is not fully characterised, then the MW is unknown. However, that does not stop you from normalising for both concentration and the number of amino acid residues.
Measured CD normalised for concentration = (Measured CD)/Molarity
Measured CD normalised for number of residues = (Measured CD)/N
Measured CD normalised for concentration and number of residues =
(Measured CD)*(1/Molarity)*(1/N)
= (Measured CD)*(MW/mg)*(mean residue MW/MW)
= Measured CD*(113/mg)
You can see that the calculation does not require knowledge of the MW of the protein nor its molar concentration. All that remains to be done is to normalise for pathlength – NB in centimetres.
If you know your protein, then converting to molar delta A (Δε) or molar ellipticity ([Θ]) in the Pro-Data (Chirascan) program would be the normal route, prior to going in to CDNN. If you wish to normalise for the number of residues as well, which is what is usually done, then the number of amino acid residues can be calculated by dividing the true MW by 113 and this number should be input to the CDNN program in the appropriate field.
Transferring CDNN results to Microsoft® Excel
Because of its quirky nature, CDNN print functions do not work correctly and once the secondary structure calculation is done, you may wish to copy data to another program. We recommend that Excel be used as the recipient program. Use the Edit>Copy Table and Edit>Copy Graph commands in the Deconvolution window to copy the results and the graph respectively and then paste them into Microsoft® Excel.
Final note
Be aware that all secondary structure calculations are notoriously ‘soft’ (because the calculations are under-determined) and results will depend on the basis sets chosen as well as the quality and extent of the data.
Another useful resource is the Protein Circular Dichroism Data Bank, hosted by The School of Crystallography, Birkbeck, and the School of Biological and Chemical Sciences, Queen Mary, University of London.
Chirascan - the new standard in circular dichroism
 |
 |
 |