I'm modeling kinetic SPR data for a set of peptides. Experimentally, I immobilized a large protein ligand on 3 cells and had one reference cell (Biacore 2000). Then I injected a series of peptides (triplicate sets of six varying concentrations) with Kinject function (4min association w/10min dissociation) and of course regenerating the surface after each run. I used Scrubber 2 to clean up my data to acquire double-referenced sets for kinetic modeling. I'm using a heterogeneous ligand model in ClampXP; I know my ligand has 2 binding sites per molecule. My question is when fitting the parameters, should I keep the max [L] and [L*] concentration the same for each peptide? All peptides were run across the same surface, so theoretically a "global" [L] and [L*] makes sense. But I get best fits for each peptide when I also fit [L] and [L*]...
I think it is correct to have [L] and [L*] as global parameters. Both are the amount of binding places available for the peptides and with a proper regeneration this will not change during the experiments.
Making parameters local will improve your fit because it will relax the constraints.
Some remarks about ClampXP. Apparently the program is not available any more. It seems there is a problem with (some of) the fitting procedure(s) and I heard there were some legal problems with the used algorithms. When it is possible, check your result with the Biacore software.
Thanks for the comments. Couple of follow up questions...
1. I use global fitting for [L] and [L*] within each peptide simulation. But should I fix the [L] and [L* ] for a different set of peptides. For example, fitting for peptide A yields [L] = 10 and [L* ] = 20. When fitting for peptide B, should I fix the [L] and [L*] to the values determined for peptide A or again include those parameters in a new fit?
2. I'm unaware of any issues w/Clamp XP and acquired it recently from the original authors. I run my experiments off-site then use the freeware to analyze my data. Should I be concerned about the validity of my data or publishing?
1: The [L] and [L*] values are in Response units. Although they represent the amount of ligand binding places the actual value is a conversion of the actual SPR angle shift. In principle, the response is linear with the amount ligand for protein-protein interactions. For small peptides this is a little more difficult as pointed out by some authors. I will include the references below.
I think I would have [L] and [L*] global per peptide. So A and B have their own [L] and [L*] value. Some quality control should be in place. Like [L]pepA should be close to [L]pepB.
2. If you acquired the software from the original authors I trust they have fixed the problems.
1. Davis, T. M. and Wilson, W. D.; Determination of the Refractive Index Increments of Small Molecules for Correction of Surface Plasmon Resonance Data. Anal.Biochem. (284): 348-353; 2000.
2. Di Primo, C. and Lebars, I.; Determination of refractive index increment ratios for protein-nucleic acid complexes by surface plasmon resonance. Anal.Biochem. (368): 148-155; 2007.
3. Schuck, P. and Minton, A. P.; Analysis of mass transport-limited binding kinetics in evanescent wave biosensors. Anal.Biochem. (240): 262-272; 1996.
4. Stenberg, E. et al; Quantitative determination of surface concentration of protein with surface plasmon resonance by using radiolabelled proteins. J.Colloid Interface Sci. (143): 513-526; 1991.