Affinity Ranking in Compound Mixtures

Advances in chemical synthesis have enabled considerable sophistication in the construction of diverse compound libraries to probe protein function [61, 62]. However, few general techniques exist that can directly assess binding mechanisms and evaluate ligand affinities in a multiplexed format. To realize the full potential of combinatorial chemistry in the drug discovery process, generic and efficient tools must be applied that combine mixture-based techniques to characterize protein-ligand interactions with the strengths of diversity-oriented chemical synthesis.

ALIS-based techniques enable researchers to rank the affinity of multiple li-gands for a protein receptor while simultaneously showing whether the ligands bind the same site as a competitor ligand or bind an allosteric site. As a simple example to describe the basis of the method, the warfarin versus warfarin-D6 competition data shown previously in Fig. 3.9 yields sigmoidal curves when normalized and plotted on a logarithmic axis (Fig. 3.12). The total competitor concentration at which each protein-ligand complex concentration (here, warfarin-D6) is reduced to half its value in the absence of the competitive ligand is defined as the affinity competition experiment 50% inhibitory concentration (ACE50 value) and is dependent upon the Kd of the ligand and other experimental parameters (Fig. 3.12A). The ACE50 value, which describes the concentration of the competitor required to compete out 50% of the ligand of interest, is the converse of the ordinary definition of a biochemical or biophysical IC50, which describes the concentration of the ligand of interest required to compete out 50% of a

Fig. 3.12 The ALIS affinity competition experiment 50% inhibitory concentration (ACE50) method. (A) The warfarin versus warfarin-D6 ALIS competition data from Fig. 3.9, normalized and plotted on a logarithmic axis, yields the ACE50 value, which is the titrant concentration at which the ligand binding is reduced by 50%. (B) Simulated

ACE50 experiment for a three-component mixture. Dashed lines indicate variation of that component's concentration by +3-fold (an overall 9-fold difference) highlighting that the method is insensitive to ligand concentration. See text for details. Reprinted from [39] with permission from the American Chemical Society.

Fig. 3.12 The ALIS affinity competition experiment 50% inhibitory concentration (ACE50) method. (A) The warfarin versus warfarin-D6 ALIS competition data from Fig. 3.9, normalized and plotted on a logarithmic axis, yields the ACE50 value, which is the titrant concentration at which the ligand binding is reduced by 50%. (B) Simulated

ACE50 experiment for a three-component mixture. Dashed lines indicate variation of that component's concentration by +3-fold (an overall 9-fold difference) highlighting that the method is insensitive to ligand concentration. See text for details. Reprinted from [39] with permission from the American Chemical Society.

known compound, for example, a radioligand. In contrast to a conventional IC50 value, a higher ACE50 value indicates a higher-affinity ligand: greater competitor concentration is required to displace the compound of interest from the binding site.

Though the ALIS ACE50 method resembles a radioligand displacement assay, the MS-based readout enables multiple components to be measured simultaneously, an advantage which is not possible using radiochemical or fluorescence methods. Fig. 3.12B demonstrates a simulated titration of a three component mixture where the total concentration of all pool components (1 mM each, 3 mM total) is less than the total receptor concentration (simulated at 5 mM). Under these conditions, individual library components bind independently to the excess receptor and compete only with the titrant, and not with one another, and the ACE50 value of each component depends upon its Kd. The dashed lines simulate variation in the concentration of each component by a factor of +3 (an overall nine-fold difference in concentration). As the simulation shows, the ACE50 values are insensitive to ligand concentration: Over a nine-fold variation in the concentration of any ligand, the ACE50 values are virtually unchanged. This feature is valuable because it allows the ACE50 method to be applied where the concentrations of the mixture components are not accurately known; for example, to the direct products of a mixture-based combinatorial chemical synthesis. As such, the method enables unpurified combinatorial mixtures to be used for the affinity optimization of a lead compound's chemical structure, a problem of great importance in the pharmaceutical industry.

The ACE50 method for ranking compounds by their protein-ligand binding affinity is demonstrated in Fig. 3.13 for a mixture of ligands to the M2 receptor. This mixture contains representatives of compounds of different structural classes, including analogs of NGD-3346, discovered by ALIS screening of combinatorial libraries. The known M2 active site inhibitor atropine was used as the titrant against 2.0 mM M2 in the presence of 0.5 mM per component compound pool. The ACE50 curves indicate clear differences in affinity, with NGD-3350 exhibiting a higher affinity than its structural congeners NGD-3348 and NGD-3346. This result correlates well with those from independent biochemical activity measurements and ALIS-based Kd titration experiments. ALIS saturation binding experiments with the individual M2 ligands yield the same rank-order of affinities as revealed by the ACE50 experiment: Kds of 0.7, 2.1, 2.9, and 6.2 mM were measured for NGD-3350, NGD-3348, NGD-3346, and NGD-3344, respectively. The compound with the highest ACE50 value, NGD-3350, has the best Kd at 0.7 mM, and this compound also shows the best biochemical activity in a cell-based cAMP assay [63]. In a tissue-based assay for M2 antagonism, NGD-3350 yields an IC50 of 9.6 mM [64, 65]. Only this compound shows significant activity in tissue, consistent with the remaining compounds all having lower affinity as determined by ACE50 ranking, ALIS Kd titration, and M2 antagonist activity in the cAMP assay

It is also noteworthy that the ACE50 technique for affinity ranking also allows mixture components to be classified as either allosteric or directly competitive with another ligand of interest. In the M2 example, reanalyzing the sigmoidal ACE50 curves in Fig. 3.13 as the ratio plots instead shows that the response ratios

Fig. 3.13 The ACE5o method demonstrated for a mixture of ligands at 1 |mM per component to the M2 receptor at 5 |mM concentration. (A) NGD-3350 requires the greatest competitor concentration to be competed from the receptor, indicating that it is the highest affinity ligand. (B) Ratio plots indicate direct binding competition with atropine. (C) Select compound structures. Reprinted from [39] with permission from the American Chemical Society.

Fig. 3.13 The ACE5o method demonstrated for a mixture of ligands at 1 |mM per component to the M2 receptor at 5 |mM concentration. (A) NGD-3350 requires the greatest competitor concentration to be competed from the receptor, indicating that it is the highest affinity ligand. (B) Ratio plots indicate direct binding competition with atropine. (C) Select compound structures. Reprinted from [39] with permission from the American Chemical Society.

are linear, indicating that all the ligands examined are directly competitive with atropine. Consistent with this result, the biochemical assays mentioned above all show that the ligands tested, like atropine, are antagonists of M2.

These results highlight the ability of the ACE50 method to simultaneously rankorder compounds by affinity, especially mixtures of structural analogs synthesized by combinatorial chemistry techniques. The method is particularly valuable for identifying those compounds with improved affinity relative to a progenitor, for example, the improved affinity of NGD-3350 relative to its parent NGD-3346. Through multiple iterations of combinatorial analog synthesis and ACE50 analysis, structure-activity relationships can be developed for the compound series and the potency of the initial hit can be optimized, even in the absence of a biochemical assay

Was this article helpful?

0 0

Post a comment