Mass Spectrometrybased Methods

Affinity measurement techniques based on mass spectrometry (MS) are of increasing interest due to the exquisite sensitivity and unique selectivity possible with MS [8]. In addition to low protein consumption, these techniques enjoy the benefit of having all reaction components in solution. Affinity selection-MS (ASMS) screening methods have been implemented using a number of hardware configurations [9-21] and all include the following steps: (1) an affinity selection step, where the protein is equilibrated with one or more potential ligands, leading to the formation of a complex of the protein with any compound capable of binding; (2) the resulting receptor-ligand complexes are separated from non-binding mixture components; and (3) ligands are identified by MS or MS/MS [9-21]. Since its first description in 1991 [22], a number of researchers have reported methods that use the direct analysis of non-covalent protein-ligand complexes by electrospray ionization-MS (ESI-MS), especially ultra-sensitive nanospray techniques, to study binding interactions [23, 24]. However, these methods require the non-covalent complexes to survive the transition to the gas phase, and there is conflicting data on the correlation of gas-phase affinity measurements with solution-phase interactions [25]. Also, these and other related MS affinity measurement techniques do not tolerate non-volatile salts or buffers, or high co-factor, metal ion, or detergent concentrations that may be necessary for proper protein folding and stability. Though not rigorously affinity selection methods, techniques that are based on hydrogen-deuterium exchange-MS [26], including the PLIMSTEX [28] method described in Chapter 11, and the SUPREX [29, 30] method, enable thermodynamic and equilibrium binding affinity estimates using high-sensitivity MALDI-MS analysis. Diffusion-based MS methods using laminar flow features in capillaries also enable the measurement of protein-ligand binding constants [32].

To take advantage of the high sensitivity and selectivity inherent to MS, while permitting greater flexibility in binding reaction conditions, hyphenated methods based on multidimensional chromatography-MS have been developed to study small molecule-protein interactions [33]. Several variants, both step-wise and integrated, have been reported and are described in detail in this book, including size-exclusion chromatography (SEC) coupled with reverse-phase chromatogra-phy-MS (RPC-MS; this Chapter), gel filtration ''spin-column''-MS (Chapter 2 [13, 34, 35]) ultrafiltration-MS (Chapter 4), frontal affinity chromatography-MS (Chapter 6 [36]), and affinity capillary electrophoresis-MS. While most reports demonstrate these methods for screening small molecule combinatorial libraries for affinity selection-based drug discovery [37, 38], Blom and co-workers described a way to quantify the binding affinities of individual compounds for their protein target by SEC-RPC-MS, and researchers at NeoGenesis demonstrated a mixture-based, competitive binding method using SEC-RPC-MS to rank binding affinities and classify ligand-ligand competition as direct or allosteric (see below) [12, 39].

An important advantage of MS-based techniques lies in their ability to simultaneously distinguish multiple components from complex reactions, enabling mixture-based analysis. As mentioned above, this feature has been exploited primarily for the discovery of ligands from pools of compounds synthesized by combinatorial chemistry techniques. However, this advantage is also useful for evaluating the binding properties of multiple protein-ligand interactions in compound mixtures, including simultaneous affinity measurements, binding site classification by ligand-ligand competition analysis, and mixture-based dissociation rate determination. A multiplexed approach to evaluating these binding characteristics enables combinatorial synthesis methods to be applied to the affinity optimization process. Medicinal chemists can thereby optimize the structure and affinity of lead compounds while minimizing the need for synthesis and purification of individual ligands, which is the most time-consuming aspect of the affinity optimization process. This approach can dramatically decrease the time, expense, and effort required to optimize a lead molecule, as the synthesis and purification of discrete compounds is reserved for only the most interesting ligands that require more detailed functional studies. Also, the ability to multiplex compounds for follow-up evaluation enables the rapid triage of multiple hits from a high-throughput screening (HTS) campaign. HTS often generates multiple compound series for which no objective assessment can be made a priori as to the likelihood of any one series progressing through medicinal chemistry optimization. A well designed, mixture-based optimization can enable the collection of critical data that can be used to identify the most promising series from an HTS screen for further follow-up.

This chapter describes a hardware platform for affinity selection-MS using continuous SEC, and the application of this platform to characterizing the binding interactions that most directly impact the medicinal chemistry component of the drug development process. The first application is a technique for quantitatively measuring absolute protein-ligand binding affinities, commonly expressed as the equilibrium binding constant Kd. The second method relies on ligand-ligand competition to yield two valuable results from mixtures of compounds: (1) to distinguish same-site versus different, or allosteric, binding by multiple ligands to the same target, thus providing insight into the location of ligand binding; and (2) to simultaneously measure the affinities of multiple ligands to the target. Last, a method is presented for measuring the dissociation rates of small molecule ligands from their protein targets, either as individual compounds or as pools. Examples are shown for several drug targets of contemporary interest in the pharmaceutical industry.

Was this article helpful?

0 0

Post a comment