Applied Biophysics for Drug Discovery E-Book

• Herausgeber: John Wiley & Sons
• Kategorie: Wissenschaft und neue Technologien
• Sprache: Englisch

Seitenzahl: 608

Veröffentlichungsjahr: 2017

Beschreibung

Applied Biophysics for Drug Discovery is a guide to new techniques and approaches to identifying and characterizing small molecules in early drug discovery. Biophysical methods are reasserting their utility in drug discovery and through a combination of the rise of fragment-based drug discovery and an increased focus on more nuanced characterisation of small molecule binding, these methods are playing an increasing role in discovery campaigns. 

This text emphasizes practical considerations for selecting and deploying core biophysical method, including but not limited to ITC, SPR, and both ligand-detected and protein-detected NMR.

Topics covered include:

•          Design considerations in biophysical-based lead screening

•          Thermodynamic characterization of protein-compound interactions

•          Characterizing targets and screening reagents with HDX-MS

•          Microscale thermophoresis methods (MST)

•          Screening with Weak Affinity Chromatography

•          Methods to assess compound residence time

•          1D-NMR methods for hit identification

•          Protein-based NMR methods for SAR development

•          Industry case studies integrating multiple biophysical methods

This text is ideal for academic investigators and industry scientists planning hit characterization campaigns or designing and optimizing screening strategies.

Leseprobe

Table of Contents

Cover

Title Page

List of Contributors

1 Introduction

References

2 Thermodynamics in Drug Discovery

2.1 Introduction

2.2 Methods for Measuring Thermodynamics of Biomolecular Interactions

2.3 Thermodynamic‐Driven Lead Optimization

2.4 Enthalpy as a Probe for Binding

2.5 Enthalpy as a Tool for Studying Complex Interactions

2.6 Current and Future Prospects for Thermodynamics in Decision‐Making Processes

References

3 Tailoring Hit Identification and Qualification Methods for Targeting Protein–Protein Interactions

3.1 Introduction

3.2 Structural Characteristics of PPI Interfaces

3.3 Screening Library Properties

3.4 Hit‐Finding Strategies

3.5 Druggability Assessment

3.6 Allosteric Inhibition of PPIs

3.7 Stabilization of PPIs

3.8 Case Studies

3.9 Summary

References

4 Hydrogen–Deuterium Exchange Mass Spectrometry in Drug Discovery ‐ Theory, Practice and Future

4.1 General Principles

4.2 Parameters Affecting Deuterium Incorporation

4.3 Utilization of HDX MS

4.4 Practical Aspects of HDX MS

4.5 Advantages of HDX MS

4.6 Perspectives and Future Application of HDX MS

References

5 Microscale Thermophoresis in Drug Discovery

5.1 Microscale Thermophoresis

5.2 MST‐Based Lead Discovery

References

6 SPR Screening: Applying the New Generation of SPR Hardware

6.1 Platforms for Screening

6.2 SensiQ Pioneer as a “OneStep” Solution for Hit Identification

6.3 Deprioritization of False Positives Arising from Compound Aggregation

6.4 Concluding Remarks

References

7 Weak Affinity Chromatography (WAC)

7.1 Introduction

7.2 Theory of WAC

7.3 Virtual WAC

7.4 Equipment and Procedure

7.5 Validation of WAC

7.6 Applications

7.7 Conclusions and Future Perspectives

Acknowledgments

References

8 1D NMR Methods for Hit Identification

8.1 Introduction

8.2 NMR Methods for Quality Control

8.3 NMR Binding Assays

8.4 Multiplexing

8.5 Specificity

8.6 Automation

8.7 Practical Considerations for NMR Binding Assays

8.8 Conclusions

References

9 Protein‐Based NMR Methods Applied to Drug Discovery

9.1 Introduction

9.2 Chemical Shift Perturbation

9.3 Methods for Obtaining Structural Information on Protein–Ligand Complex

9.4 Recent and Innovative Examples of Protein‐Observed NMR Techniques Applied Drug Discovery

9.5 Conclusions and Future Perspectives

References

10 Applications of Ligand and Protein‐Observed NMR in Ligand Discovery

10.1 Introduction

10.2 Ligand‐Observed NMR Experiments Based on the Overhauser Effect

10.3 Protein‐Observed NMR Experiments: Chemical Shift Perturbations

10.4 Conclusion

Acknowledgments

References

11 Using Biophysical Methods to Optimize Compound Residence Time

11.1 Introduction

11.2 Biophysical Methods for Measuring Ligand Binding Kinetics

11.3 Measuring Structure–Kinetic Relationships: Some Example Case Studies

11.4 Effects of Conformational Dynamics on Binding Kinetics

11.5 Kinetic Selectivity

11.6 Mechanism of Binding and Kinetics

11.7 Optimizing Residence Time

11.8 Role of BK in Improving Efficacy

11.9 Effect of Pharmacokinetics and Pharmacodynamics

11.10 Summary

References

12 Applying Biophysical and Biochemical Methods to the Discovery of Allosteric Modulators of the AAA ATPase p97

12.1 p97 and Proteostasis Regulation

12.2 Structure and Dynamics of p97

12.3 Drug Discovery Efforts against p97

12.4 Uncompetitive Inhibitors of p97 Discovered by High‐Throughput Screening

12.5 Fragment‐Based Ligand Screening

12.6 Conclusions

References

13 Driving Drug Discovery with Biophysical Information: Application to

Staphylococcus aureus

Dihydrofolate Reductase (DHFR)

13.1 Introduction

13.2 Results and Discussion

13.3 Conclusion

References

14 Assembly of Fragment Screening Libraries: Property and Diversity Analysis

14.1 Introduction

14.2 Physicochemical Properties of Fragments

14.3 Molecular Diversity and Its Assessment

14.4 Experimental Evaluation of Fragments

14.5 Assembling Libraries for Screening

14.6 Concluding Remarks

References

Index

End User License Agreement

List of Tables

Chapter 03

Table 3.1 Structural classification of protein–protein interfaces. Example structures illustrated in the columns to the right are marked in bold. PDB coordinates for example structures: bromodomain–histone complex: 3JVK, Bcl‐2–Bax complex: 2XA0, CD80–CD28 complex: created by overlay of CD28 structure 1YJD with CD80–CTLA4 complex 1I8L. Abbreviations: Bak, Bcl‐2 homologous antagonist/killer; Bax, Bcl‐2‐associated X; Bcl, B‐cell lymphoma; HIF1α, hypoxia‐inducible factor 1α; CD28, cluster of differentiation 28; CD80, cluster of differentiation 80; HIV, human immunodeficiency virus; ICAM1, intercellular adhesion molecule 1; IL‐2, interleukin‐2; IL‐2Rα, interleukin‐2 receptor; IL‐17A, interleukin‐17A; IL‐17RA, interleukin‐17 receptor; LEDGF, lens epithelium‐derived growth factor; LFA1, leukocyte function‐associated antigen‐1; MDM2, murine double minute 2 homologue; p53, tumor protein p53; SMAC, second mitochondria‐derived activator caspase; VHL, Von Hippel–Lindau disease tumor suppressor; IAP, inhibitor of apoptosis protein.

Chapter 10

Table 10.1 NOE‐based experiments for analysing protein‐ligand complexes

List of Illustrations

Chapter 02

Figure 2.1 The Δ

H

and

K

D

for the interaction of the parent inhibitor KNI‐10026 (left) and two derivatives, KNI‐10007 (top right) and KNI‐10006 (bottom right) with plasmepsin II.

Figure 2.2 Examples of ITC data for complex interactions (not 1 : 1 binding). (a) Example of high‐resolution ITC data collected with MicroCal™ PEAQ‐ITC for a titration of a model racemic mixture (160 μM total ligand concentration) and 10 μM protein. For experimental details, see Malvern White paper [45]. (b) Example of a ligand titration into a protein dimer with two identical sites that display negative cooperativity (previously unpublished data).

Figure 2.3 Example of ITC data for the titrations of a protein kinase with a compound conducted at 1 and 5% v/v DMSO concentrations. (a) An overlay of the raw data. The color green is the thermogram for the titration in the presence of 1% DMSO. (b) The binding isotherms for the two titrations overlaid with the best‐fit curves obtained by global fit of the two datasets with

N

value shared and association constant (

K

) and Δ

H

set as local/specific to each dataset.

Figure 2.4 Global analysis of Mg

2+

and Ca

2+

binding to (a) pig b–parvalbumin, (b) rat a‐PV 49–60/85, and (c) Phl p 7 polcalcin. ITC binding isotherms overlaid with the best‐fit curve for titrations of 50–60 μM protein with (A); 1.10 mM Ca

2+

/1.0 mM Mg

2+

(B); 1.10 mM Ca

2+

/5.0 mM Mg

2+

(C); 1.09 mM Ca

2+

/10.0 mM Mg

2+

(D); 1.08 mM Ca

2+

/20.0 mM Mg

2+

(E); 2.06 mM Mg

2+

(F).

Chapter 03

Figure 3.1 Mechanisms of PPI modulation: A, orthosteric inhibition; B, allosteric regulation—inhibition; C, orthosteric stabilization; D, allosteric regulation—stabilization.

Figure 3.2 Property plots showing how PPI fragments and inhibitors are heavier and more lipophilic, contain more acid/basic groups, and have a higher hydrophobic proportion than their non‐PPI counterparts. However, there is little difference in the three‐dimensionality of the PPI sets compared with their non‐PPI counterparts from the normalized PMI plot. Data is based on four sets of 100 known inhibitors or fragments.

Figure 3.3 Case studies for each epitope type. (a) Crystal structure of BRD4 bromodomain (white surface) bound to JQ1 (light blue sticks; PDB: 3MXF); (b) Crystal structure of Bcl‐2 (white surface) bound to ABT‐199 (light blue sticks; PDB: 4MAN); (c) Crystal structure of MDM2 (white surface) bound to a Nutlin compound (light blue sticks; PDB: 1RV1); (d) Crystal structure of IL‐17A homodimer (white cartoon) bound to the HAP peptide (light blue cartoon; PDB: 5HHX).

Chapter 04

Figure 4.1 Analytical workflow of HDX.

Figure 4.2 Differential HDX MS of mapping conformational changes and binding pocket of the anticancer drug vemurafenib on the protein kinase B‐Raf. Regions presenting lower deuteration (blue and purple) and those unchanged (green and orange) are superimposed on the corresponding crystal structure.

Chapter 05

Figure 5.1 (a) Schematic representation of MST instrumentation. MST is measured in capillaries with a total volume of 10 µl. The fluorescence within the capillary is excited and detected through the same objective and coupled with an IR laser to locally heat a defined sample volume. Thermophoresis of fluorescent molecules through the temperature gradient is detected over time. (b) Schematic of MST traces. Prior to IR laser activation, fluorescent molecules are homogeneously distributed and a constant initial fluorescence is detected. After activation of the IR laser, a rapid fluorescence change is observed, followed by a slower thermophoretic redistribution of the fluorescent‐labeled molecules. The thermophoretic movement is detected for a defined MST‐on time. Deactivation of the IR laser leads to back‐diffusion of molecules, which is solely driven by mass diffusion. MST, MicroScale Thermophoresis; IR, infrared. (c) Typical MST binding experiment. The thermophoretic movement of a fluorescent molecule (black trace; “unbound”) changes upon binding to a non‐fluorescent ligand (red trace; “bound”). (d) For analysis, the change in thermophoresis is expressed as the change in the normalized fluorescence (Δ

F

norm

), which is defined as

F

1

/

F

0

(

F

values correspond to average fluorescence values between defined areas marked by the red and blue cusors, respectively). Titration of the non‐fluorescent ligand results in a gradual change in thermophoresis, which is plotted as Δ

F

norm

versus ligand concentration to yield a binding curve that can be fitted to derive binding constants. (e) MST instruments NT.115/LabelFree and NT.Automated family and the respective capillary trays are shown. The NT.Automated instrument uses capillary chips that can be handled by commercial robotic platforms. The table summarizes basic instrument specifications.

Figure 5.2 Size‐change independency of MST. (a) Schematic representation of a streptavidin subunit in complex with a biotin‐AHX‐glycine peptide (modified from pdb 3RY2) and chemical structures of biotinylated peptides used for interaction studies. ΔMW is the difference in molecular weight compared to the control peptide with a single glycine residue. (b) MST traces and

F

norm

values of NT647‐labeled streptavidin in complex with the biotinylated peptides shown in (a) in 1× PBS + 0.05% Tween20 at medium MST power, measured on a Monolith NT.115.

F

norm

values were measured as triplicates.

Figure 5.3 Application examples for MST. (a) Crystal structure of the transmembrane protein NRT1.1 in complex with nitrate ions. (b) Quantification of the binding affinity of nitrate ions to GFP‐tagged BRT1.1 wild‐type protein and point mutants. The results show that His356 is essential for interaction, while the positively charged Arg45 and Lys164 are dispensable (from Ref. [23]). (c) Quantification of the interaction of NT647‐labeled protein kinase A (PKA) and its small‐molecule inhibitor quercetin in buffer and at different concentrations of human serum. The affinity decreases with increasing serum concentrations. (d) Quantification of the binding of quercetin to NT647‐labeld human serum albumin (HSA). (e) PKA–quercetin affinity is reduced upon addition of HSA concentrations equivalent to the concentrations of HSA in 5 and 30% serum, respectively.

Figure 5.4 Detection of protein aggregation by MST. (a) MST traces and fluorescence microscopy image (inset) of homogeneous, non‐aggregating samples. The red line corresponds to a double‐exponential fit of the MST trace; corresponding residuals are shown below. (b) MST traces and fluorescence microscopy image (inset) of samples containing protein aggregates. Arrowheads highlight aggregates in the microscopy image. The red line corresponds to a double‐exponential fit of the MST trace; corresponding residuals are shown below. (c) Time‐lapse micrographs detected by fluorescence microscopy of MST of an aggregated and non‐aggregated solution of NT647‐labeled rituximab antibody in PBS. While the non‐aggregated sample shows a homogeneous decrease of fluorescence intensity upon activation of the IR laser, several bright particles start traversing the heated spot during MST in the aggregated sample, thereby causing irregular MST traces shown in (b). Scale bars = 30 µm.

Figure 5.5 MST‐based single‐point and affinity screening. (a) Schematic overview of automatic sample preparation, capillary chip filling, and execution of MST experiments for single‐point screenings. (b) Schematic overview of automatic preparation of serial dilutions, capillary chip filling, and execution of MST experiments for affinity screenings. (c) Representative data from an MST noise test and adsorption tests using buffer with (buffer B) and without 0.05% Tween 20 (buffer A). Note that in the absence of detergent, the MST signal noise is significantly increased and protein adsorption occurs to the capillary walls indicated by aberrant capillary shapes in the capillary scan. (d) Establishment of optimal MST setting for screening campaigns using a positive control interaction. MST settings are chosen in such a way that measurement time is minimal, while the signal‐to‐noise ratio of the binding amplitude is maximal.

Figure 5.6 Cascade of a small‐scale single‐point and affinity screening on p38

alpha

inhibitors. (a) MST data of duplicate measurements of the interaction of small molecules with NT647‐labeled p38

alpha

kinase. Dotted lines represent hit thresholds. Five potential binders have been identified using a double‐blind approach. (b) Affinity determination of small molecules identified in (a). (c) Comparison of

K

d

values of the characterized binders with literature values after unblinding.

Chapter 06

Figure 6.1 Stages in early lead generation and potential use of various platforms.

Figure 6.2 Parts of the SensiQ OneStep binding sensorgram.

Figure 6.3 Various kinetic approaches on the SensiQ Pioneer yield similar kinetic parameters as the traditional approach on the Biacore T200. (a) and (b) Multiple‐cycle kinetics on the T200 and Pioneer, respectively. (c) Equivalent single‐cycle kinetics (termed FastStep on the SensiQ). (d) OneStep approach to obtain kinetic information.

Figure 6.4 Differentiating specific from nonspecific binding using SensiQ OneStep. (a) Two phases in the association and dissociation. (b) A 10‐fold lower screening concentration results in the disappearance of the nonspecific aspect.

Figure 6.5 Compounds that bind with low surface activity often do not correlate well with

in vitro

data. (a) Traditional dose–response sensorgrams demonstrating non‐saturation at higher concentrations. (b) OneStep analysis with fixed

R

max

providing similar results to (a). However, as can be seen in (c), allowing the

R

max

to float allows for more accurate data fitting.

Figure 6.6 On–off rate plot for compounds tested as part of a secondary HTS screen. The on rate is on the

y

‐axis, while the off rate is on the

x

‐axis. The diagonal lines represent various

K

D

s. A graph of this type shows that at a given

K

D

, two compounds could have vastly different kinetic profiles.

Figure 6.7 Determination of CAC and comparison with various IC

50

s. The blue curve is cell‐based assay, while the yellow and green curves are from two different biochemical assays. The aggregation threshold is derived from the mean + 3 SD from control wells. The concentration at which the Epic response crosses the threshold is the aggregation point.

Chapter 07

Figure 7.1 The artist’s view of biological interactions on a cell surface, many of which are of a weak or transient nature.

Figure 7.2 WAC of the anomeric forms of α‐ and β‐

p

‐nitrophenyl glycosides of maltose on a monoclonal antibody HPLC column. The maltose components were injected in the presence of bovine serum. The initial void peak is serum components devoid of any affinity, whereas the maltose components are binding in the range of

K

d

 = 0.1–1 mM.

Figure 7.3 Experimental and virtual weak affinity separations of carbohydrates on a wheat germ agglutinin (WGA) affinity column. 3′SL, 3‐sialyllactose.

Figure 7.4 WAC analysis of three small carbohydrate derivatives. CapGal,

N

‐(ε‐aminocaproyl)‐β‐galactosylamine; MNPG, 3‐nitrophenyl α‐

D

‐galactopyranoside; and ONPG, 2‐nitrophenyl β‐

D

‐galactopyranoside.

Figure 7.5 Mass spectra of six drugs on a human serum albumin column (50 × 2.0 mm I.D.) by isocratic separation in a mobile phase of 96% 50 mM ammonium acetate and 4% isopropanol.

Figure 7.6 Clinical application of WAC: isocratic analyses of the corticosteroids (cortisol and cortisone) by direct injection of human serum (2 µl diluted 1 : 10 in the mobile phase of 5 mM ammonium acetate pH 7.4; 0.2 ml/min) on immobilized human serum albumin (HSA) (50 × 2.1 mm; 118 mg HSA/g silica) (a and b). (c) represents the analysis of five clinical plasma samples. One analytical cycle amounts in this case to ~2.5 min (marked by the line in (c)) if only cortisol is monitored.

Figure 7.7 Chiral separation of

D

,

L

‐phenylalanine by WAC on anti‐

D

‐phenylalanine (a) and anti‐

L

‐phenylalanine (b) antibodies.

Figure 7.8 (a) Results from fragment screening with WAC of 566 compounds and two reference binders on thrombin (3‐ABA, 3‐aminobenzamidine; 4‐ABA, 4‐aminobenzamidine). The horizontal line indicates the cutoff retention time and the star indicates a good hit. (b) Example of extracted ion chromatograms from 14 fragments and dimethyl sulfoxide (DMSO) in one mixture.

Chapter 08

Figure 8.1 Applications of 1D NMR along the lead discovery and optimization time line.

Figure 8.2 Effect of salt concentration on aqueous solubility. The amount of compound that remains in solution can be greatly affected by the composition of the buffer. This figure shows an example for which even a small addition of salt (0.125 M NaCl) leads to the complete precipitation of compound, resulting in potential false negatives in binding assays.

Figure 8.3 Popular 1D NMR assays for detecting intermolecular interactions between small molecules and their macromolecular target.

Figure 8.4 Example data from the STD assay, showing the aromatic region of the

1

H NMR spectrum for two compounds, one that binds target (left) and one that does not bind target (right). In the top spectrum for each panel, the saturation pulse was set to a region of the spectrum containing no resonances (−10 ppm); the resultant spectrum reflects the non‐attenuated 1D

1

H NMR spectrum for each compound. In the middle spectra, a saturation pulse train with an excitation bandwidth of ~1 ppm was set to 0.5 ppm, thereby directly saturating the methyl region and indirectly saturating the entire spectrum for the macromolecular target via “spin diffusion.” Compound binding leads to a reduction in compound resonance intensity, which is best seen in each difference spectrum generated by subtracting the saturated spectrum from the unsaturated spectrum (lower spectra).

Figure 8.5 The T2R assay exploits the difference in relaxation rates of a small molecule and its macromolecular target. For a weak interaction between compound and target, the NMR signal is the weighted average of the free and bound ligand.

Figure 8.6 Multiplexing. The addition of several compounds into one sample can significantly reduce reagents required for the experiment and data acquisition time. Multiplexing is best achieved when chemical shift encoding is used to select the compound mixtures, such that each compound has at least one uniquely resolved/isolated resonance in the composite spectrum of the mixture.

Figure 8.7 Confirming specificity of binding. Fragments 1 and 2 both show evidence of binding, as evidenced by the large peaks in the NMR STD difference spectra (left panel). Using a mutated protein that partially blocks the binding site (middle panel) or blocking the binding site with a tightly binding inhibitor (right panel) eliminates binding of the specific compound (disappearance of peaks in the difference spectra of middle and right panels for fragment 1), but not for the nonspecific compound (no signal in difference spectra of bottom middle and right panels for fragment 2).

Chapter 09

Figure 9.1 Binding site mapping using CSP data. (a) Overlay of [

15

N,

1

H]‐HSQC spectra recorded on a

15

N‐labeled sample of the target protein with increasing concentrations of unlabeled peptide (binding partner). Spectra were recorded at the following protein : peptide molar ratios: 1 : 0 (blue spectra), 1 : 1 (cyan), 1 : 2 (orange), and 1 : 5 (red). For one peak the direction of the shift is indicated with a red arrow and the chemical shift differences in proton (Δ

δ

H

) and nitrogen (Δ

δ

N

) are indicated by black arrows. (b) The proton and nitrogen chemical shift differences were measured comparing the spectra recorded on samples of the target protein and of the target protein plus fivefold molar excess of peptide. The weighted chemical shift difference (Δ

δ

weighted

) was calculated for each assigned N─H backbone group of the target protein using the following equation:

. Where 0.14 is a scaling factor required to account for the difference in the range of amide proton and amide nitrogen chemical shifts [1]. The Δ

δ

weighted

values found were plotted, and an empirical threshold (red dashed line) was set at the average of all the Δ

δ

weighted

values plus two times their standard deviation to distinguish between perturbed and unperturbed residues. (c) Surface view of the target protein structure with residues perturbed by peptide binding highlighted in red.

Figure 9.2 Schematic representation of the AIDA method [17]. [

15

N,

1

H]‐HSQC spectra are recorded before and after complex formation between a small

15

N‐labeled protein (below ~20 kDa)—reporter protein—and its larger (above ~30 kDa) protein binding‐partner that is left unlabeled. Formation of the complex induces broadening of the signals of the reporter protein that will become weaker and eventually disappear because of a faster relaxation rate due to the increase in molecular weight. The addition of a molecule that disrupts the complex will result in the recovery of the spectrum that will display peak shifts (highlighted by red asterisks) if the antagonist binds to the reporter protein. Otherwise, when the antagonist binds to the larger member of the complex the spectrum of the reporter protein will be recovered unchanged.

Figure 9.3 Schematic representation of the SOS‐NMR method [23]. STD‐NMR spectra recorded on 2‐(3′‐pyridyl)‐benzimidazole with different FKBP samples (i–vi). The residual peaks observed in spectra (ii–v) correspond to specific protons of the ligand that are in close contact with non‐deuterated residues in the protein.

Figure 9.4 Schematic representation of Abl conformational change induced by myristate binding and monitored by [

15

N,

1

H]‐HSQC. (a) [

15

N,

1

H]‐HSQC spectrum recorded on free Abl (left) and Abl crystal structure (PDB: 2G2H) (right). The C‐terminus of helix_I (shown in red), which includes Val525, is unstructured and not visible in the crystal structure. (b) [

15

N,

1

H]‐HSQC spectrum recorded on Abl with bound myristate (left) and Abl–myristate crystal structure (PDB: 1OPK) (right). Myristate is shown in sticks (grey carbons), helix_I in red, and position of Val525 is indicated by an arrow.

Figure 9.5 Schematic representation of how time‐resolved NMR can be used to monitor a PTM reaction. The enzyme that catalyzes the PTM reaction is incubated with its protein or peptide substrate, and NMR spectra are recorded at regular time intervals to monitor the signal of the residue that is modified in the reaction. (a) As the reaction progresses, the intensity of the signal of the unmodified residue decreases and in parallel the signal of the modified residue increases. (b) Signals detected in (a) are quantified and plotted to obtain graph as the one showed into the picture to then extrapolate information such as the reaction rate.

Chapter 10

Figure 10.1 INPHARMA‐based optimization of protein inhibitors using known protein–peptide interactions. The protein–peptide interactions are compared with the protein–inhibitor interactions using INPHARMA experiments. The peptide region that is missing in the inhibitor is identified by INPHARMA and modelization, and an optimized inhibitor is synthesized.

Figure 10.2 STD experiment. (a) Principle of the STD‐based epitope mapping assessment. The relative intensities of the peaks are compared in the STD and STD

off

(1D) spectra. Protons can be highlighted that are buried (protons with STD effect of 100%), or solvent exposed (STD intensities weaker than the STD

off

intensities). (b) STD spectra recorded on the 4′‐fluoro‐[1,1′‐biphenyl]‐4‐carboxylic acid compound bound to Bcl‐xL. Protons located at proximity of the fluorine are shown to be buried by STD, in agreement with the 1YSG structure shown in (c).

Figure 10.3 Comparison of CSP signs observed for a series of analogous ligands bound to peroxiredoxin 5. (a) Comparison of

1

H CSPs (both magnitude and signs are shown) between compounds

1

and

2

(top); compounds

1

,

3

, and

4

(middle); and compounds

1

and

5

(bottom). (b) Differences of CSP magnitudes are mapped on the protein 3D structure. (c) Opposite CSP signs are highlighted using red spheres on the protein 3D structure. When compounds

1

and

5

are compared, opposite signs are mainly observed for CSPs in the protein region 30–60. By contrast, compounds

1

,

2

,

3

, and

4

display similar CSP signs in this region. For residues in loop 113–125 (in blue in c), opposite CSP signs are observed when compound

1

is compared with compounds

3

or

4

. While comparing CSP magnitudes does not lead to any conclusion, comparing CSP signs highlight key differences: (i) the catechol moiety in all compounds but five displays a similar orientation in the protein–ligand complexes; (ii) the ter‐butyl and phenyl groups of compounds

3

and

4

are located near the loop 113–125 (colored in blue in c).

Figure 10.4 Principle of the post‐docking CSP‐based filter. Experimental

1

H CSPs (CSP

exp

) are compared with

1

H CSPs back‐calculated (CSP

calc

) using docking positions of the ligand in the protein structure. The positions are ranked according to the agreement between CSP

exp

and CSP

calc

values.

Chapter 11

Scheme 11.1 Kinetic scheme for a one‐step binding mechanism.

Figure 11.1 (a) Schematic representation of ligand recognition models. Conformational selection and induced fit are often linked, and complex formation can proceed via both pathways simultaneously. (b) Typical dependence of pseudo‐first‐order rate constant,

k

obs

, on ligand concentration, observed for different binding mechanisms. The curves depict the situation where the conformational transition step is significantly slower than ligand binding.

Scheme 11.2 Kinetic scheme for a two‐step binding mechanism, with receptor isomerization occurring after initial binding (induced fit).

Scheme 11.3 Kinetic scheme for a two‐step binding mechanism, where receptor isomerization occurs prior to ligand binding (conformational selection).

Figure 11.2 Effect of binding kinetics and affinity on target engagement and selectivity. Plasma drug concentration is shown along with the fractional occupancy of the primary target and two secondary targets. The kinetic constants, according to Scheme 11.2, for the primary and secondary targets 1 and 2, respectively, are

k

1

: all 1 × 10

9

 M

−1

 s

−1

;

k

2

: 10, 10, 50 s

−1

;

k

3

: 1, 1, 0;

k

4

: 0.01, 0.5, 0.

Figure 11.3 Free energy diagram for a typical one‐step binding mechanism, as shown in Scheme 11.1. Kinetic rates are proportional to the activation energy barrier between the two states, such that increasing residence times (decreasing

k

diss

) can be achieved by increasing the energy barrier between the complex and the transition state,

R.L

‡

. This may be achieved by stabilizing the complex or by destabilizing the transition state. Destabilizing the transition state will tend to decrease

k

ass

(leading to slower association) unless the free energy of the ligand can also be increased.

Figure 11.4 Simplest model for describing the interplay between binding kinetics and pharmacokinetics. The administered dose is distributed to the target location with rate constant

k

dist

and cleared from the target location with rate constant,

k

cl

. The rate constants for compound–target association and dissociation, according to Scheme 11.1, are given by

k

ass

and

k

diss

, respectively. The target, compound, and compound–target complex concentrations are given by C, T, and C.T, respectively. All four rate constants are important in determining the compound–target complex concentration.

Chapter 12

Figure 12.1 Structure of p97. (a) Domain architecture of a p97 protomer, each consisting of an N‐domain (PPI site), tandem ATPase domains D1 and D2, and a C‐terminal domain (PPI site). (b) Cartoon (PDB 3CFC) showing top and side views of p97 bound to ADP. The major domains of every other protomer are colored as depicted in (a). (c) Space‐filled cryo‐EM models of p97 in three distinct nucleotide‐dependent conformations. Arrows highlight conformational rearrangements that occur in the D2 domain between conformations I and II and in the N‐domain between conformations II and III, respectively (PDB 5FTL, 5FTM, 5FTN).

Figure 12.2 Conformational preference of p97 PPIs is altered in MSP1 mutations. (a) SPR sensorgrams showing tight binding of p47 to WT p97‐ATP and weaker binding to WT p97‐ADP. (b) Normalized SPR response fit to a two‐site equilibrium model for p47 binding to WT p97 and a MSP1 mutant in the absence or presence of ADP and ATP. ADP weakens the binding of p47 to WT‐p97 by eightfold (from 26 to 184 nM), but has no effect on p47 binding to MSP1‐p97. (c) Model indicating the preference p47 to bind to p97 in the ATP‐bound N‐domain “up” conformation over the ADP‐bound N‐domain “down” conformation. In MSP1 mutations, p47 loses this conformational selectivity.

Figure 12.3 Summary of HTS and fragment screens used to target specific domains of p97.

Figure 12.4 Biochemical mechanism of action (MOA) studies on two p97 inhibitors identified by HTS. (a) Chemical exploration around two hits from HTS yielded compounds with improved potency for both the indole amide and phenyl indole series. (b) The MOAs for the two HTS hits were determined to be uncompetitive based on a shift in the IC

50

value at various ATP concentrations. (c) Lineweaver–Burk plots on SAR compounds within each series confirmed an uncompetitive mechanism of action for both chemical series.

Figure 12.5 SPR experiments determine mode of binding of the uncompetitive p97 inhibitor. (a) SPR sensorgrams (black lines) for the binding of UPCDC30245 (0–3.33 μM) to p97 in the absence of nucleotide and presence of ADP or ATP. Kinetic analysis (orange lines) was used to determine

K

D

values. (b) Normalized binding curves for an equilibrium fit, illustrating the binding preference for the ADP conformation of p97. (c) UPCDC30245 binding was measured to a WT, K521A‐D1, and K524A‐D2 in the presence of ADP. The K521A‐D1 and K524A‐D2 mutations limit ADP binding to the D1 and D2 domains, respectively. (d) Normalized binding curves for an equilibrium fit show a reduced affinity to the K524A‐D2 mutation and indicate that ADP binding to D2 is important for tight compound binding.

Figure 12.6 NMR experiments shed light into the binding mode of p97 inhibitors. (a) 1‐D

1

H NMR group epitope mapping (GEM) experiment shows line broadening in the presence of the D2 domain of p97, indicating binding. The strongest intensity changes occurred at methyl protons at positions

a

and

b,

suggesting that these groups are in close contact with p97. (b) 2D

13

C methyl‐TROSY NMR of ligands bound to U‐[

2

H], [

1

H,

13

C]‐ILVMA isotope‐labeled p97 D2. Excerpts from the spectra show changes in isoleucine δ1 methyl resonances observed upon the addition of different ligands. Data courtesy of Michael Chimenti and Mark Kelly.

Figure 12.7 Cryo‐EM structure of UPCDC30245 bound to p97. (a) Cartoon model of p97 with UPCDC30245 (red) bound at the interface of the D1 and D2 domains (PDB 5ftj). (b) Close‐up view of UPCDC30245 bound to the D2‐ADP conformation of p97 (PDB 5FTJ) and (c) superimposition of UPCDC30245 onto the D2‐ATPγS conformation of p97 (PDB 5FTM). Movement of the D2 domain when ATPγS binds causes rearrangement of the UPCDC30245‐binding site. Residues exhibit clashes (<3 Å; black lines) between UPCDC30245 and the ATPγS‐bound form of p97.

Figure 12.8 2D methyl‐TROSY NMR spectra of U‐[

2

H], [

13

CH

3

]‐ILVMA p97 ND1 domain in the absence and presence of nucleotides and fragments. Full spectra (left) of the p97 ND1 domains (black) overlaid with spectra of ND1 + ADP (red) show the chemical shift perturbations upon ADP binding. Residues in the isoleucine, alanine, and leucine regions all show changes. Panels depicting parts of the spectra (left) of ND1 (black) and ND1 in the presence of fragment (red) show that a competitive fragment exhibits similar shifts as ADP and AMP, whereas a noncompetitive fragment shows different chemical shifts. ADP competition was determined by SPR [88].

Chapter 13

Scheme 13.1 NADPH‐dependent catalysis of dihydrofolate to tetrahydrofolate by DHFR.

Figure 13.1 Overlay of crystal structures of NADPH‐TMP‐bound structures of SA WT and the S1 mutant DHFR indicating a similar binding mode of TMP as determined by X‐ray crystallography.

Figure 13.2 Thermodynamic parameters of TMP binding to SA WT and S1 mutant DHFR in presence and absence of cofactor, NADPH [15], are shown in this bar graph representing change in free energy (dG), enthalpy (dH), and entropy (TdS). The binding affinity and stoichiometry of binding (

n

b

) are listed in the table.

Figure 13.3 Chemical structure of Trimethoprim (TMP) used as prototype ligand.

Figure 13.4 (a)

15

N–

1

H HSQC NMR correlation spectrum for 250 μM uniformly

13

C,

15

N labeled apo WT DHFR, (b) broadened resonances, which are not observed for apo SA WT DHFR in 3D heteronuclear NMR experiments are highlighted in red and mapped onto the SA WT‐NAPDH‐TMP X‐ray structure. Binding sites of NADPH and TMP are shown for reference but were not included in the sample. Observed resonances for apo WT protein are colored in green, (c)

15

N–

1

H HSQC NMR correlation spectrum for ternary WT‐NADPH‐TMP complex containing 250 μM uniformly

13

C‐/

15

N‐labeled apo WT DHFR with 500 μM NADPH and 250 μM Trimethoprim.

Figure 13.5 Residues undergoing extreme line broadening (purple) and relaxation dispersion (pink) are shown for (a) SA WT and (b) S1 mutant–NADPH binary complexes. These residues are mapped onto the ternary (NAPDH‐TMP) crystal structures. TMP is shown in the figure to illustrate its binding pocket but was not in the sample. For S1, mutations in α‐helix B are highlighted in yellow.

Figure 13.6 NMR Relaxation dispersion results for SA WT‐NADPH binary complex (blue diamonds), SA WT‐NADPH‐TMP complex (blue circles), S1 mutant–NADPH binary complex (red diamonds), and S1 mutant–NADPH‐TMP complex (red circles) for (a) H30, (b) R57, and (c) G119at 25°C.

15

N

R

2

relaxation rates were measured on Bruker 600 and 700 MHz NMR spectrometers at 25°C using a relaxation‐compensated Carr–Purcell–Meiboom–Gill (CPMG) pulse sequence [23a]. Data were recorded on a sample containing 0.25 mM uniformly

15

N‐enriched DHFR (WT or S1), 1 mM NADPH, and 0.5 mM Trimethoprim (only in ternary complexes). Relaxation decay curves were measured for five values of

τ

cp

ranging from 0.5 to 5.33 ms.

Figure 13.7 Binding affinities and kinetic profiles for TMP binding to S1 mutant and SA WT DHFR in the presence of NADPH. Binding affinity of TMP to S1 mutant is determined by equilibrium binding methods using six ligand concentrations with the highest concentration =1 μM and in threefold dilutions. Binding affinity of TMP to SA WT is determined using three ligand concentration (5.5, 1.8, and 0.62 nM) and performing a global fit (

K

D

determined by the standard equation

K

D

 = 

k

off

/

k

on

).

Figure 13.8 SPR binding sensorgrams shown for S1 mutant DHFR and four inhibitors with differences in antibacterial activity as determined by MICs and IC

50

s. Off rates in SPR were determined at one concentration for each compound. Ligand concentration of 50 nM, 5.5 nM, 5.5 nM, and 1 μM were run for BAL0030545 [26], PF‐1, PF‐2, and TMP, respectively.

Figure 13.9

15

N–

1

H HSQC correlation spectrum for 250 μM uniformly

15

N‐labeled TMP and SA WT (green) and S1 mutant DHFR (red). Chemical shift of N‐4‐Ha is shifted downfield in the presence of WT DHFR relative to that observed for N‐4‐Ha in complex with S1 mutant. The samples contained 250 μM of uniformly

13

C‐/

15

N‐labeled WT or S1 DHFR along with 500 μM NADPH in 10 mM HEPES, pH 8.0 buffer with 50 mM NaCl.

Figure 13.10 Naphthalene series rationale. Cinnamate compound (

1

) exhibited great intrinsic potency but poor MIC values. A naphthalene scaffold was designed to bring additional lipophilicity into the core. Early naphthalene–acid‐containing compounds, exemplified by compound (

2

), exhibited excellent binding and improved MICs.

Figure 13.11 Library design strategy. Orthogonal library design strategies carried out to select monomers to couple to the naphthyl moiety. Strategies included exploring flexibility of DHFR by interacting with the lid (based on NMR dynamics data), exploiting protein desolvation effects (based on WaterMap and X‐ray data), isosteric replacement of acids, identifying slow

k

off

compounds (based on SPR data), and identifying monomers in favorable physicochemical property space to enhance antibacterial properties.

Figure 13.12 Library design results. Shown here is a histogram of results from the prospective biophysics library efforts. Hypotheses/strategies that were prosecuted are shown in the

x

‐axis. Number of compounds is indicated along the

y

‐axis. Bars are color coded based on numbers of compounds evaluated,

Staphylococcus aureus

wild‐type active compounds (cut‐off 5 nM), S1 mutant active compounds (cut‐off 500 nM), and whole‐cell active compounds.

Chapter 14

Figure 14.1 Physicochemical property profiles of the Monash 2011 fragment library with typical properties for Ro3‐compliant fragments, lead‐like molecules, and Ro5 drug‐like compounds shown with dotted lines. Fsp3, fraction of sp

3

‐hybridized carbons from total carbons; HAC, heavy atom count; HBA, hydrogen bond acceptors; HBD, hydrogen bond donors; MW, molecular weight (Da); NRotB, number of rotatable bonds;

S

log

P

, calculated logarithm of the octanol/water partitioning coefficient; and TPSA, polar surface area (Å

2

). Properties were calculated in KNIME v2.11.3 with RDKit.

Figure 14.2 Diversity properties of the Monash 2011 fragment library (1137 compounds). (a) ECFP4 and MACCS fingerprint profiles of the library showing the distributions of maximum (green) and average (red) Tanimoto coefficient for each compound in the library compared with the rest of the library. Principal component analysis (PCA) of molecular quantum number (MQN) space for (b) all fragment space (red, a one million fragment‐like subset of GDB17) with density contours shown in black lines and the Monash 2011 fragment library (blue) and (c) commercially available fragment space (red, 717,262 compounds with MW < 300 Da, TPSA < 100 Å

2

, and

S

log

P

 < 3 from the eMolecules database of commercially available compounds) with density contours shown in black lines and the Monash 2011 fragment library (blue). (d) The normalized principal moment of inertia (nPMI) plot for the Monash 2011 fragment library with conformers within 5 kcal/mol of the lowest energy conformer colored on a gradient from blue (lowest energy) to red. Conformers, MQN, and fingerprint profiles were calculated in KNIME v2.11.3 with RDKit with an MMFF94s force field using up to 10 conformers and merging conformers with RMSD of ≤1.0 Å. PCA was calculated in R v3.2.2.

Figure 14.3 QC workflow for assessing fragments. Only compounds that pass the QC process are deemed suitable for inclusion in the screening library.

Figure 14.4 Fragment QC by

1

H NMR spectroscopy. Exemplar spectra are shown for compounds that (a) pass QC and fail QC based on (b) solubility in aqueous buffer or missing, (c) identity, and (d) purity.

Figure 14.5 Property plots of compounds that pass, fail, and were missing in the initial QC by NMR. (a) Scatter plot highlighting the enrichment of missing compounds at low molecular weight and low TPSA. (b) Box plot comparison of MW,

S

log

P

, and TPSA for pass (

N

 = 704), fail (

N

 = 159), and missing (

N

 = 270) compounds with

t

‐test

p

‐values indicated comparing the pass and missing datasets. Box plots show minimum, 25th, 50th, 75th, maximum, and mean as the lower whiskers, lower box, midline, upper box, upper whisker, and cross, respectively. MW, molecular weight (Da); TPSA, polar surface area (Å

2

);

S

log

P

, calculated logarithm of the octanol/water partitioning coefficient. Properties were calculated in KNIME v2.11.3 with RDKit and statistics calculated in R v3.2.2.

Figure 14.6 Analysis of properties of frequent hitters. Comparison of property profiles of frequent hitters (

N

 = 241, fit an identified frequent hitter scaffold or have hit in ≥50% of screens) and non‐frequent hitters (

N

 = 895) for (a) molecular weight (MW, Da) and calculated partitioning coefficient in octanol/water (

S

log

P

). (b) Fraction of sp

3

‐hybridized carbons from total carbons (Fsp3) and polar surface area (TPSA, Å

2

). The average number of

1

H NMR STD screens per compound = 18.6. Properties were calculated in KNIME v2.11.3 with RDKit and statistics calculated in R v3.2.2.

Guide

Cover

Table of Contents

Begin Reading

Pages

iii

iv

xiii

xiv

xv

xvi

xvii

1

2

3

4

5

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

61

62

63

64

65

66

67

68

69

70

71

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

285

286

287

288

289

290

291

Applied Biophysics for Drug Discovery

Edited by

Donald Huddler

Widener University Delaware Law SchoolWilmington, USA

Edward R. Zartler

Quantum Tessera ConsultingCollegeville, USA

This edition first published 2017© 2017 John Wiley & Sons Ltd

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions.

The right of Donald Huddler and Edward R. Zartler to be identified as the authors of the editorial material in this work has been asserted in accordance with law.

Registered OfficesJohn Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USAJohn Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK

Editorial OfficeThe Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK

For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com.

Wiley also publishes its books in a variety of electronic formats and by print‐on‐demand. Some content that appears in standard print versions of this book may not be available in other formats.

Limit of Liability/Disclaimer of WarrantyIn view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of information relating to the use of experimental reagents, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert or instructions for each chemical, piece of equipment, reagent, or device for, among other things, any changes in the instructions or indication of usage and for added warnings and precautions. While the publisher and authors have used their best efforts in preparing this work, they make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives, written sales materials or promotional statements for this work. The fact that an organization, website, or product is referred to in this work as a citation and/or potential source of further information does not mean that the publisher and authors endorse the information or services the organization, website, or product may provide or recommendations it may make. This work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for your situation. You should consult with a specialist where appropriate. Further, readers should be aware that websites listed in this work may have changed or disappeared between when this work was written and when it is read. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.

Library of Congress Cataloging‐in‐Publication Data

Names: Huddler, Donald Preston, 1971– editor. | Zartler, Edward, editor.Title: Applied biophysics for drug discovery / edited by Donald Huddler, Edward R. Zartler.Description: Hoboken, NJ : Wiley, 2017. | Includes bibliographical references and index. | Identifiers: LCCN 2017013165 (print) | LCCN 2017014921 (ebook) | ISBN 9781119099499 (pdf) | ISBN 9781119099505 (epub) | ISBN 9781119099482 (hardback)Subjects: | MESH: Drug Discovery–methods | Biophysical PhenomenaClassification: LCC RS420 (ebook) | LCC RS420 (print) | NLM QV 745 | DDC 615.1/9–dc23LC record available at https://lccn.loc.gov/2017013165

Cover image: Provided by Edward Zartler; (Background) © hakkiarslan/Getty ImagesCover design: Wiley

List of Contributors

Timon AndréNanoTemper Technologies GmbHMunichGermany

Current address: Heidelberg UniversityGermany

Michelle R. Arkin, Ph.D.Department of Pharmaceutical ChemistrySmall Molecule Discovery CenterUniversity of California, San FranciscoUSA

Juan Astorga‐Wells, Ph.D.Biomotif AB & HDXperts AB and Department of Medical Biochemistry and BiophysicsKarolinska InstitutetStockholmSweden

Philipp Baaske, Ph.D.NanoTemper Technologies GmbHMunichGermany

Tanja BartoschikNanoTemper Technologies GmbHMunichGermany

Michal Bista, Ph.D.Structure and BiophysicsDiscovery SciencesAstraZenecaCambridgeUK

Alessio Bortoluzzi, Ph.D.Division of Biological Chemistry and Drug Discovery, James Black CentreSchool of Life SciencesUniversity of DundeeUK

Current address: Immunocore LtdMilton ParkAbingdonOxfordshireUK

Kris A. Borzilleri, B.S.Pfizer Worldwide Research and DevelopmentGroton, CTUSA

Susan M. Boyd, D.Phil.IOTA Pharmaceuticals Ltd.St. John’s Innovation CentreCambridgeUK

Dennis Breitsprecher, Ph.D.NanoTemper Technologies GmbHMunichGermany

Stacie L. Bulfer, Ph.D.Department of Pharmaceutical ChemistrySmall Molecule Discovery CenterUniversity of California, San FranciscoUSA

Current address: Deciphera PharmaceuticalsLawrence, KSUSA

Steven S. Carroll, Ph.D.Merck & Co.West Point, PAUSA

Jaehyun Cho, Ph.D.Department of Biochemistry and Molecular BiophysicsColumbia University, NYUSA

Boris A. Chrunyk, Ph.D.Pfizer Worldwide Research and DevelopmentGroton, CTUSA

Alessio Ciulli, Ph.D.Division of Biological Chemistry and Drug Discovery, James Black CentreSchool of Life SciencesUniversity of DundeeUK

Bradley C. Doak, Ph.D.Medicinal ChemistryMonash Institute of Pharmaceutical SciencesMonash UniversityVictoriaAustralia

Stefan Duhr, Ph.D.NanoTemper Technologies GmbHMunichGermany

Minh‐Dao Duong‐ThiSchool of Biological SciencesNanyang Technological UniversitySingapore

Kelly Fahnoe, B.S.Pfizer Worldwide Research and DevelopmentGroton, CTUSA

Caroline A. Fanslau, M.S.Bristol‐Myers SquibbPrinceton, NJUSA

Alessandra Fenoli, Ph.D.NanoTemper Technologies GmbHMunichGermany

Current address: University of SalernoItaly

Mark Flanagan, Ph.D.Pfizer Worldwide Research and DevelopmentGroton, CTUSA

Mary J. Harner, Ph.D.Bristol‐Myers SquibbPrinceton, NJUSA

Melissa Harris, B.S.Pfizer Worldwide Research and DevelopmentGroton, CTUSA

Holly Heaslet, Ph.D.Pfizer Worldwide Research and DevelopmentGroton, CTUSA

Geoffrey A. HoldgateStructure and BiophysicsDiscovery SciencesAstraZenecaCambridgeUK

Donald Huddler, Ph.D.Computational and Structural ChemistryGlaxoSmithKline plcCollegeville, PAUSA

Current address: Widener University Delaware Law School, DEUSA

George Karam, Ph.D.Pfizer Worldwide Research and DevelopmentGroton, CTUSA

Isabelle Krimm, Ph.D.Institut des Sciences Analytiques, UMR5280 CNRSUniversité Lyon 1, Ecole Nationale Supérieure de LyonFrance

Brian Lacey, B.S.Pfizer Worldwide Research and DevelopmentGroton, CTUSA

Thorleif LavoldBiomotif AB & HDXperts ABDanderydSweden

Thomas V. Magee, Ph.D.Pfizer Worldwide Research and DevelopmentGroton, CTUSA

Natalia Markova, Ph.D.Scientific Marketing BiosciencesMalvern InstrumentsStockholmSweden

Melanie MaschbergerNanoTemper Technologies GmbHMunichGermany

Alex McColl, B.S.Pfizer Worldwide Research and DevelopmentGroton, CTUSA

Guille Metzler, B.S.Eng.PharmaCadence Analytical Services, LLCHatfield, PAUSA

William J. Metzler, Ph.D.Bristol‐Myers SquibbPrinceton, NJUSA

Current address: PharmaCadence Analytical Services, LLCHatfield, PAUSA

J. Richard Miller, Ph.D.Pfizer Worldwide Research and DevelopmentGroton, CTUSA

Craig J. Morton, Ph.D.Australian Cancer Research Foundation Rational Drug Discovery CentreSt. Vincent’s Institute of Medical ResearchVictoriaAustralia

Luciano Mueller, Ph.D.Bristol‐Myers SquibbPrinceton, NJUSA

Kartik Narayan, Ph.D.Sanofi PasteurSwiftwater, PAUSA

Ronan O’Brien, Ph.D.Business Development‐MicroCalMalvern InstrumentsNorthampton, MAUSA

Sten OhlsonSchool of Biological SciencesNanyang Technological UniversitySingapore

Arthur Palmer III, Ph.D.Department of Biochemistry and Molecular BiophysicsColumbia University, NYUSA

Henry Putz, B.S.Pfizer Worldwide Research and DevelopmentGroton, CTUSA

Anil Rane, Ph.D.Pfizer Worldwide Research and DevelopmentGroton, CTUSA

Philip RawlinsStructure and BiophysicsDiscovery SciencesAstraZenecaCambridgeUK

Parag Sahasrabudhe, Ph.D.Pfizer Worldwide Research and DevelopmentGroton, CTUSA

Ron Sarver, B.S.Pfizer Worldwide Research and DevelopmentGroton, CTUSA

Martin J. Scanlon, Ph.D.Medicinal ChemistryMonash Institute of Pharmaceutical SciencesMonash UniversityVictoriaAustralia

Veerabahu Shanmugasundaram, Ph.D.Pfizer Worldwide Research and DevelopmentGroton, CTUSA

Jamie S. Simpson, Ph.D.Medicinal ChemistryMonash Institute of Pharmaceutical SciencesMonash UniversityVictoriaAustralia

Christopher J. Stubbs, Ph.D.Structure and BiophysicsDiscovery SciencesAstraZenecaCambridge, UK

Tim Subashi, B.S.Pfizer Worldwide Research and DevelopmentGroton, CTUSA

Chakrapani Subramanyam, Ph.D.Pfizer Worldwide Research and DevelopmentGroton, CTUSA

Andrew P. Turnbull, Ph.D.Cancer Research Technology Ltd.London Bioscience Innovation CentreLondonUK

Björn Walse, Ph.D.SARomics Biostructures ABLundSweden

Jane M. Withka, Ph.D.Pfizer Worldwide Research and DevelopmentGroton, CTUSA

Roman Zubarev, Ph.D.Department of Medical Biochemistry and BiophysicsKarolinska InstitutetStockholmSweden

1Introduction

Donald Huddler*

Computational and Structural Chemistry, GlaxoSmithKline plc, Collegeville, PA, USA,

Over the last two decades, biophysics has reemerged as a core discipline in drug discovery. Many may argue that biophysical methods never truly left discovery, but all will note the renewed present importance and central role of such methods. This reemergence is driven by three primary forces: the birth of fragment‐based drug discovery schemes, the recognition of and desire to mitigate artifacts in traditional biochemical screening, and a desire to accelerate the transition from first‐in‐class to best‐in‐class molecules by focusing on hit and lead kinetics. Each of these strategies or goals requires various information‐rich biophysical methods to experimentally execute. This text aims to summarize some of the key methods emerging from these three broad enterprises. First, though, it will map the contours of these three drivers of biophysics’ reemergence and link them to the chapters that follow.

Fragment‐based drug discovery and fragment‐based lead discovery are slightly different names for the same discovery approach: using a library of relatively small compounds to probe the surface of a target protein for binding sites. Fragment‐based discovery approaches are animated by the information theory‐based idea that relatively simple, small compounds sample chemical space more effectively than larger, more complex molecules [1, 2]. In practice, this approach drives one to develop low complexity screening libraries [3, 4]; consequently, the binding interactions with target proteins are generally very weak. Weak interactions require sensitive methods to unambiguously detect the binding event [5]. In simple bimolecular binding, the concentration of the complex is driven by the concentration of the ligand; this drives many scientists to screen their fragment libraries at relatively high concentrations. Effective screening methods must both be able to detect relatively weak interactions in the context of relatively high compound concentrations; several biophysical methods are well suited for this demanding screening campaign [6]. Various NMR approaches have been successfully applied to identify and characterize weak small molecule–protein interactions [7]. This text explores both traditional protein‐detected NMR [8] approaches in Chapters 9 and 10and nontraditional NMR [9, 10] approaches in Chapter 8. Both approaches have merit and are usefully applicable in partially overlapping circumstances. Surface plasmon resonance (SPR) [11, 12] and microscale thermophoresis (MST) [13] have also been successfully deployed in fragment screening campaigns to detect weak interactions. Chapters 5 and 6 explore applications of MST and SPR beyond fragment‐based discovery, respectively.

A second force driving the reemergence of biophysical methods in drug discovery has been the desire to identify and eliminate high‐throughput screening hits that operate through uninteresting nuisance mechanisms. Brian Schoichet recognized and characterized some commonly observed nuisance phenomena; many of these nuisance mechanism enzymatic assay hits had weak micromolar activities and showed either a flat or highly irregular SAR [14]. Schoichet’s team determined that the aberrant behavior in biochemical screening assays was driven by poor solubility resulting in compound aggregate formation. These compound aggregates, present in extremely low concentration, serve as protein sinks, adsorbing most of the target protein, yielding what appeared to be detectable but weak inhibition [15]. His team demonstrated that many of these aggregation‐based inhibitors could be culled from screening hits by comparing activity in an assay with no or very low detergent to a high detergent assay condition. Compounds that lose activity in the high detergent assay were likely to be uninteresting nuisance hits.

Several biophysical methods complement the differential detergent biochemical assay [16]. In the biochemical assay approach, the presence of aggregates is inferred, whereas in the biophysical approaches, the aggregates are directly detected. SPR is uniquely suited such direct detection of nuisance behavior in a buffer matched to the original biochemical screening buffer [17]. Aggregated compounds generate complex binding responses that are not simple 1 : 1 interactions but rather reflect the partitioning of the aggregated compound between the free buffer and the protein captured on the sensor chip. Aggregated compounds also show complex binding to the sensor surface with no target protein captured, providing a simple, parallel means to detect nonideal interactions in real time during library screening. Hit validation workflows now commonly employ SPR, mass spectrometry, and other biophysical methods to remove nuisance mechanism hits [18].

A third trend driving the reemergence of biophysics in drug discovery is the desire to optimize kinetic or thermodynamic properties with an aim to rapidly progress from a first‐in‐class compound to a best‐in‐class compound. When comparing a first‐in‐class compound to a best‐in‐class compound, the best‐in‐class molecule generally has high selectivity for the pharmacologic target and consequently a lengthy residence time with that target [19]. Detailed understanding of compound binding kinetics [20] and inhibitory mechanism leads to better candidates with properties more like an ideal best‐in‐class compound [21]. SPR allows real‐time analysis of binding kinetics [22]; streamlined experimental approaches allow rapid compound sorting based on kinetic parameters [23]. Combining thermodynamic data with affinity and kinetic data further characterizes the intermolecular interactions, enabling detailed SAR and further compound optimization [24]. This idea is explored and different methods applied inform interaction quality in Chapters 2, 4, 7, and 11.

The text concludes with a case study in Chapter 14 that joins many of the methods and concepts discussed in earlier chapters. The Pfizer research team used a combination of traditional biochemical analysis, focused structural information derived from NMR, SPR kinetics, and NMR dynamics to optimize a Staphylococcus aureus DHFR inhibitor. Data from no one method assured success; it was the conjunction of data from the several biophysical techniques that enabled their focused, hypothesis‐driven prospective library design that ultimately yielded novel, nonacid cell‐active inhibitors. Importantly, the dynamics and kinetic data incorporated common resistance mutations, informing the library design and ultimately the candidate compounds. This discovery case study exemplifies the fully integrated discovery approach where data‐rich biophysical techniques continually inform discovery. This approach enables research teams to target transient protein conformations, protein–protein interaction surfaces, or complex enzyme targets—all examples of targets that have met will have little success with traditional high‐throughput enzymatic screening [25].

This text is a survey of contemporary biophysical methods in drug discovery. Biophysical methods report on intermolecular interactions directly with rich detail; these methods naturally complement traditional high‐throughput screening [26, 27], particularly when attacking irregular, nonenzymatic [28, 29], or membrane protein [30, 31] targets.

References

1. Leach, A. R. and Hann, M. M. Molecular complexity and fragment‐based drug discovery: ten years on.

Curr. Opin. Chem. Biol.

15:489–496 (2011).

2. Hann, M. M., Leach, A. R., and Harper, G. Molecular complexity and its impact on the probability of finding leads for drug discovery.

J. Chem. Inf. Comput. Sci.

41:856–864 (2001).

3. Boyd, S. M., Turnbull, A. P., and Walse, B. Fragment library design considerations.

WIREs Comput. Mol. Sci.

2:868–885 (2012).

4. Lau, W. F., Withka, J. M., Hepworth, D., Magee, T. V., Du, Y. J., Bakken, G. A.,

et al.

Design of a multi‐purpose fragment screening library using molecular complexity and orthogonal diversity metrics.

J. Comput. Aided Mol. Des.

25:621 (2011).

5. Mashalidis, E. H., Sledz, P., Lang, S., and Abell, C. A three‐stage biophysical screening cascade for fragment‐based drug discovery.

Nat. Protoc.

8:2309–2324 (2013).

6. Joseph‐McCarthy, D., Campbell, A. J., Kern, G., and Moustakas, D. Fragment‐based lead discovery and design.

J. Chem. Inf. Model.

54:693–704 (2014).

7. Kim, H. Y. and Wyss, D. F. NMR screening in fragment‐based drug design: a practical guide.

Methods Mol. Biol.

1263:197–208 (2015).

8. Dias, D. M. and Ciulli, A. NMR approaches in structure‐based lead discovery: recent developments and new frontiers for targeting multi‐protein complexes.

Prog. Biophys. Mol. Biol.

116:101–112 (2014).

9. Pilger, J., Mazur, A., Monecke, P., Schreuder, H., Elshorst, B., Bartoschek, S.,

et al.

A combination of spin diffusion methods for the determination of protein‐ligand complex structural ensembles.

Angew. Chem.

54:6511–6515 (2015).

10. Cala, O. and Krimm, I. Ligand‐orientation based fragment selection in STD NMR screening.

J. Med. Chem.

58:8739–8742 (2015).

11. Perspicace, S., Banner, D., Benz, J., Müller, F., Schlatter, D., and Huber, W. Fragment‐based screening using surface plasmon resonance technology.

J. Biomol. Screen.

14:337–349 (2009).

12. Kreatsoulas, C. and Narayan, K. Algorithms for the automated selection of fragment‐like molecules using single‐point surface plasmon resonance measurements.

Anal. Biochem.

402:179–184 (2010).

13. Jerabek‐Willemsen, M., Wienken, C. J., Braun, D., Baaske, P., and Duhr, S. Molecular interaction studies using microscale thermophoresis.

Assay Drug Dev. Technol.

9:342–353 (2011).

14. McGovern, S. L., Caselli, E., Grigorieff, N., and Shoichet, B. K. A common mechanism underlying promiscuous inhibitors from virtual and high‐throughput screening.

J. Med. Chem.

45(8):1712–1722 (2002).

15. McGovern, S. L., Helfand, B. T., Feng, B., and Shoichet, B. K. A specific mechanism of nonspecific inhibition.

J. Med. Chem.

46(20):4265–4272 (2003).

16. Feng, B. Y., Simeonov, A., Jadhav, A., Babaoglu, K., Inglese, J., Shoichet, B. K., and Austin, C. P. A high‐throughput screen for aggregation‐based inhibition in a large compound library.

J. Med. Chem.

50(10):2385–2390 (2007).

17. Giannetti, A. M., Koch, B. D., and Browner, M. F. Surface plasmon resonance based assay for the detection and characterization of promiscuous inhibitors.

J. Med. Chem.

51:574–580 (2008).

18. Lee, H., Zhu, T., Patel, K., Zhang, Y.‐Y., Truong, L., Hevener, K. E.,