Electronic Letters to:

Other Original Articles: Hidefumi Inaba, William Martin, Anne S. De Groot, Shuwen Qin, and Leslie J. De Groot Thyrotropin Receptor Epitopes and Their Relation to Histocompatibility Leukocyte Antigen-DR Molecules in Graves’ Disease J Clin Endocrinol Metab 2006; 91: 2286-2294 [Abstract] [Full text] [PDF]

eLetters: Submit a response to this article

Electronic letters published:

Response to letter from Prof RG Phelps -

Leslie J De Groot, H Inaba, W Martin, A S De Groot, S Qin   (18 January 2007)

IC50 comparisons of binding to different class II using different reference peptides are flawed

Richard G Phelps   (9 November 2006)

Response to letter from Prof RG Phelps - 18 January 2007
Leslie J De Groot, Research Brown University, H Inaba, W Martin, A S De Groot, S Qin Send letter to journal: Re: Response to letter from Prof RG Phelps - ldegroot{at}earthlink.net Leslie J De Groot, et al.	We thank Professor Phelps for his comments, and agree that comparisons of epitope affinities for DRs on the basis of IC50s rather than Kis could introduce an error when comparing affinities across different DRs. However, relative affinities of epitopes for one particular DR are not affected at all. We note that other research groups present their comparisons in the same IC50 format that we used (1), as does Professor Phelps in both the abstract and figure I of his own manuscript (2). We observed that the IC50s of the indicator peptides we used were similar for all DRs, ranging from 0.3 to 0.5 µM. This suggests, but does not prove, that the Kis would also be similar. We measured the Ki for HSP3-13, the indicator peptide for DR3, and found it to be 0.08 µM. Thus, using the relation noted in Phelps’ E-letter, we can derive the Kis for our data on DR3 in the conditions described in our report, as follows: Ki = IC50 for DR3/3.5. Phelps et al also used the same indicator peptide (HA307-319) that we used in our study, and reported the following relations for DR7 and DR1: Ki = IC50 for DR7/6.87 and IC50 for DR1/6.33. This suggests that in making cross DR comparisons using our IC50s, true affinities of peptides for some DRs would increase by a factor of about 2 from those reported. We believe this would not significantly alter the overall interpretation of our study. References 1. Geluk A, van Meijgaarden KE, Schloot J, Drijfhout JW, Ottenhoff THM, Roep BO. 1998. HLA-DR Binding Analysis of Peptides From Islet Antigens in IDDM. Diabetes 47:1594-1601 2. Phelps RG, Jones V, Turner AN, Rees AJ. 2000. Properties of HLA class II molecules divergently associated with Goodpasture’s disease. Int Immunol 12:1135-1143.
IC50 comparisons of binding to different class II using different reference peptides are flawed 9 November 2006
Richard G Phelps, Doctor University of Edinburgh Send letter to journal: Re: IC50 comparisons of binding to different class II using different reference peptides are flawed richard.phelps{at}ed.ac.uk Richard G Phelps	In their paper, Inaba et al. (1) sought to compare the affinity of TSH receptor peptides for class II molecules with divergent associations with Graves' disease using IC50 values. We believe that this approach is flawed. Even using a simplified model for the complex interactions in these assays, affinity is related to IC50 through a function that depends upon the concentration and affinity of the reference peptide. The authors do not indicate the affinity of their various reference peptides for the various class II molecules, so the IC50 values given cannot be accurately compared. The possible degree of error is considerable. Under conditions where the concentration of reference peptide is high relative to class II, such that its concentration may be considered constant and the concentration of any peptides displaced from the class II/ peptide complexes used in the assay may be ignored, Kd = IC50/(1+c/k), where c and k are, respectively, the reference peptide concentration and Kd for the class II under assessment. The authors give c as 0.2 μM, but not k. Commonly used reference peptides have k values between 2 and 200 nM for different class II molecules, so the resulting Kd could be between 0.5 and 0.005 times the IC50. Therefore, IC50 data permit comparison of the peptides’ relative affinities for a particular class II, but not comparison of the affinities of particular peptides to different class II molecules, nor the overall affinity of the peptides for class II molecules with different disease associations. We find these shortcomings unfortunate, particularly because we agree with the authors’ conclusion that disease-protective class II have higher affinity for autoantigen-derived peptides; this echoes our own observation with respect to Goodpasture's disease (2) and could be very pertinent to the way class II molecules influence susceptibility to autoimmune diseases. References 1. Inaba H, Martin W, De Groot AS, Qin S, De Groot LJ. 2006. Thyrotropin Receptor Epitopes and Their Relation to Histocompatibility Leukocyte Antigen-DR Molecules in Graves’ Disease. J Clin Endocrinol Metab 91:2286-2294 2. Phelps RG, Jones V, Turner AN, Rees AJ. 2002. Properties of HLA class II molecules divergently associated with Goodpasture's disease. Int Immunol 12:1135-1143