Finding Better Compounds: Labelfree Assays in Compound Identification

1
Finding Better Compounds Label-free Assays in
Compound Identification Profiling
Charles A. Lunn Schering-Plough Research
Institute Kenilworth, New Jersey 07033
Woodbridge Hilton Sept 15-16, 2009
2
Emergence of high throughput biochemistry
Calcium mobilization using FLIPR Tetra
Technical advances allow rapid analysis of most
biochemical activities
3
Development of automation strategies improve
quality speed of screening campaigns
Cytomat 6000 Incubtor, Under Table
Custom Air Knife Plate Cleaning Device
Echo555-1
Envision Left access
Envision Right access
Echo555-1
O
O
O
O
FLIPR Tetra
O
O
Thermo CytoMat 44 compound library
Ambient Automated Hotel
O
VSpin
AquaMax
O
DM Delidder 8 Position
Cytomat 6000 Incubtor, Under Table
4
Evolution of Primary Screening Automation
High Throughput Biochemistry
2010s
Strategies to find BETTER compounds compounds
for non-traditional targets

• Label-free technologies can overcome limits /
  expand reach of classic biochemistry

5
Label-free technology can overcome limits /
expand reach of classic biochemistry

• Definition
• Ligand used for screening assay is presented to
  target free of external, unnatural label
  (fluorophore, etc)
• Native ligand can be used
• Detection does not impact interaction between
  ligand and target
• Strategies
• Use biophysical properties of substrate / product
  / system to measure biochemistry
• Use biology of cell to measure ligand / target
  interaction

6
Strategies for no-label screening

• Cell-based
• Cell-substrate sensing
• resonance waveguide grating (Corning EPIC)
• peak wavelength value modulation (SRU BIND)
• cellular dielectric spectroscopy (MDS Sciex)
• electric impedance sensing (ACEA)
• Pathway screens using biosensor
• target activation utilizes native ligand,
  detection separate

• Biochemistry
• altered surface optical property with hit binding
• surface plasmon resonance (BiaCore)
• resonance waveguide grating (Corning EPIC)
• biolayer interferometry (BioForte)
• altered protein confirmation with hit binding
• Surface Enhanced Raman Spectroscopy
• ligand effect on protein thermal stability
• ThermoFluor dye binding
• high throughput microcalorimetry (Vivactis)

4
MASS REDISTRIBUTION
3

• compound identification
• bound hit identification by LC/MS
• Automated Ligand Identification System
  (Schering Cambridge)
• high throughput substrate / product
  quantitation (BioTrove)

1
2
7
Mass spectrometry as no-label detection platform

• Biologic molecules can be ionized, providing a
  form that can be detected by appropriate
  detector.
• The mass spectrometer sorts ions according to
  their mass/charge ratios with high accuracy
  each mass/charge signal specific for specific
  biomolecule.
• The detector converts the ion signal to a
  proportional electrical current. Quantitating the
  magnitude of these electrical signals as a
  function of m/z identifies the specific
  compound/compounds in question.

8
MS to identify bound library compound(s)
Example 1

Current Opinion in Chemical Biology 2007,
11518526 Affinity selection-mass spectrometry
screening techniques for small molecule drug
discoveryD Allen Annis, Elliot Nickbarg, Xianshu
Yang, Michael R Ziebell and Charles E
WhitehurstSchering-Plough Research Institute,
Cambridge, MA 02139
Cmpd. library in mass encoded pools
9
Example of Affinity selection- mass spectrometry

• J Biomol Screen. 2006 Mar11(2)194-207.
• Discovery and characterization of orthosteric and
  allosteric muscarinic M2 acetylcholine receptor
  ligands by affinity selection-mass spectrometry.
• Whitehurst CE, Nazef N, Annis DA, Hou Y, Murphy
  DM, Spacciapoli P, Yao Z, Ziebell MR, Cheng CC,
  Shipps GW Jr, Felsch JS, Lau D, Nash HM.
• NeoGenesis Pharmaceuticals, Inc., Cambridge, MA
  02139

M2 AChR ligand NGD-3366 shows allosteric binding
versus N-methyl-scopolamine.
Allosteric ligand NGD-3366 decreases dissociation
rate of N-methyl-scopolamine from M2 AChR.
10
Characteristics of affinity selection-mass
spectrometry screening

• Identify compounds which bind target
• Platform independent of target bioactivity
• Binding to novel sites on target protein
• e.g., allosteric sites, etc.
• Biologic efficacy of hits determined by 2o
  analysis
• Binders with no bioactivity common
• Requires specialized, mass encoded compound
  library
• Requires purified, soluble protein target
• Compounds that destabilize target conformation
  may represent false positives

11
Example 2
Technique for rapid sampling / purification /
analysis of MS samples
Parallel columns
Staggered injection
Mass Spectrometric Techniques for
Label-free High-Throughput Screening in Drug
Discovery Roddy, et al (2007) Anal Chem. 79(21)
8207-13. (Novartis Institute for Biomedical
Research)
11
12
Technique for rapid sampling / purification /
analysis of MS samples
US20050123970 A1 High throughput
autosampler Inventor(s) Ozbal, Can Green,
Donald

• US20050194318 A1
• High throughput autosampler
• Inventor(s) Ozbal, Can
• Green, Donald

• Linear sample processing, each separated by air
  bubble
• Integrated sample purification using surface
  adsorption chemistry
• Proprietary valves allow switch from sample to
  buffer without pressure drop

13
BioTrove RapidFireTM technology accurately
measures enzymatic activity / kinetics.

• RESULTS FROM ANTIBACTERIAL LIPID DEACYLASE
  PROGRAM
• Classic assay (charcoal adsorption of
  14C-acetate) not appropriate for HTS
• Mass spectrometry can easily detect ?42 amu

• Reaction started by adding 20 ml enzyme to
  compounds (10 ml) plus substrate (20 ml) using
  PLATEMATE PLUS.
• After 60 minutes at room temperature, HCl added
  (20 ml 0.25 N) to stop reaction.
• Plates sealed, then placed at - 80oC (minimum
  2 hours)
• Plates shipped to BioTrove, Inc. Woburn, MA for
  analysis.

14
Hit identification from orthogonal mixtures
Quantitate substrate, product conversion _at_
5-8 seconds per sample.
median z' for mixture screening was 0.74
Hit validation/dose-response in duplicate
15
No-label thermodenaturation method to confirm
ligand binding to lipid deacylase
Example 3

• Bound ligand stabilizes protein, increasing
  denaturation temperature exposing hydrophobic dye
  binding sites

Ligand
Increasing temperature
Dye
J Biomol Screen. 2006 11(7)854-63. Universal
screening methods and applications of
ThermoFluor MD Cummings, MA Farnum, MI Nelen
16
Temperature-dependent Fluorescent (TdF) Assay A
Thermal Shift Assay

• A typical TdF assay
• 0.5-2 µM protein 5-50 µM compound, 5-20 µl
  (96- or 384-well)
• 10 pmol (0.3 to 1 µg) protein per Kd
  determination
• Dose-dependent melting temperature (Tm) shifts
  were fitted to calculate TdF Kd values at 25oC.
• Assuming -7 kcal/mol enthalpy of binding
• 50 error margin for -4 to -11 kcal/mol enthalpy
  of binding
• 3-fold error margin for 0 to -14 kcal/mol
  enthalpy of binding

17
Selected hits from Rapid FireTM screening bind to
enzyme
Good correlation between enzymatic IC50 and
biomolecular Kd values
J Biomol Screen. 2006 11(7)854-63. Universal
screening methods and applications of
ThermoFluor MD Cummings, MA Farnum, MI Nelen
17
18
Example 4
Confirmation testing following traditional
screening campaign
Fluorescence detection
Mass detection
BioTrove technology eliminates false positives
due to fluorescence artifacts
19
The evasive nature of drug efficacy implications
for drug discovery
Trends in Pharmacological Sciences Volume 28,
Issue 8, August 2007, Pages 423-430 Special
Issue on Allosterism and Collateral Efficacy
Ségolène Galandrin, Geneviève Oligny-Longpré and
Michel Bouvier Department of Biochemistry and
Groupe de Recherche Universitaire sur le
Médicament, Institute for Research in Immunology
and Cancer, Université de Montréal, Montréal
(Québec), H3C 3J7, Canada

• Biased agonism / functional selectivity

• ligands that behave as agonists toward a given
  receptor can act, through the same receptor, as
  antagonists or even inverse agonists on a
  different pathway in the same cell. These
  observations have important implications for the
  molecular definition of efficacy and the process
  of drug discovery

Target-based strategies may not be sufficient in
the search for better compounds
20
Cannabinoid CB2 receptor-specific agonists
inverse agonists modulate chemotaxis - HOW?
Example 5 Relevant Pharmacology
CB2 Receptor AGONISTS
CB2-selective INVERSE AGONISTS
WIN
CP55,940
Sch.319
Anandamide
Triaryl bis sulfone CB2-specific inv. ag.
JTE-907

• Agonists induce
• chemotaxis
• Gonsiorek et al, 2007
• Tanikawa et al, 2007
• Jiang et al, 2007
• Kishimoto et al, 2006
• Berghuis et al, 2005
• Oka et al, 2004
• Jorda et al, 2002

• Agonists inhibit
• chemotaxis
• Montecucco et al, 2008
• McHugh et al, 2007
• Mukhopadhyay et al, 2007
• Coopman et al, 2007
• Nilsson et al, 2006
• Ghosh et al, 2006
• Sacerdote et al, 2000

P
b-arrestin
MAPK
Recycling
21
Chemotaxis exhibits bell-shaped dose-response
curve
Fig. 4.   Chemotaxis induction of CCR7
transfectants by SLC and ELC. Parental L1.2 cells
(closed squares) or transfected L1.2 cells stably
expressing CCR7 (circles) were placed in upper
wells and allowed to migrate toward SLC (closed
circles and closed squares) or ELC (open circles)
in the lower wells. Migrated cells were collected
and counted by flow cytometry. The assay was done
in duplicate. Representative results from two
separate experiments are shown.
Dias-Baruffi M, Roque-Barreira MC, Cunha FQ,
Ferreira SH. Biological characterization of
purified macrophage-derived neutrophil
chemotactic factor. Mediators Inflamm.
19954(4)263-269.
Yoshida R, Nagira M, Kitaura M, Imagawa N, Imai
T, Yoshie O. Secondary lymphoid-tissue chemokine
is a functional ligand for the CC chemokine
receptor CCR7. J Biol Chem. 1998 Mar
20273(12)7118-22.
No clear correlation between chemokine conc. and
biology
21
22
Cannabinoid CB2 receptor-specific agonists
inverse agonists modulate chemotaxis - HOW?
Example 5 Relevant Pharmacology
CB2 Receptor AGONISTS
CB2-selective INVERSE AGONISTS
WIN
CP55,940
Sch.319
Anandamide
Triaryl bis sulfone CB2-specific inv. ag.
JTE-907

• Agonists induce
• chemotaxis
• Gonsiorek et al, 2007
• Tanikawa et al, 2007
• Jiang et al, 2007
• Kishimoto et al, 2006
• Berghuis et al, 2005
• Oka et al, 2004
• Jorda et al, 2002

• Agonists inhibit
• chemotaxis
• Montecucco et al, 2008
• McHugh et al, 2007
• Mukhopadhyay et al, 2007
• Coopman et al, 2007
• Nilsson et al, 2006
• Ghosh et al, 2006
• Sacerdote et al, 2000

P
b-arrestin
MAPK
Recycling
Unique Biology
23
Potential therapeutic significance of ß-arrestin
pathways

• beta-Arrestin 1 mediates nicotinic acid-induced
  flushing but not its antilipolytic effect, in
  mice.
• - Walters et al, J Clin Invest. 2009 May 1
  119(5) 13121321.
• A ß-arrestin 2 Signaling Complex Mediates Lithium
  Action on Behavior
• - Beaulieu et al. 2008 Cell January 11 132(1)
  125-136.
• CXCR4 dimerization and beta-arrestin-mediated
  signaling account for the enhanced chemotaxis to
  CXCL12 in WHIM syndrome.
• - Lagane et al 2008 Blood Jul 1112(1) 34-44.
• beta-Arrestin-1 mediates glucagon-like peptide-1
  signaling to insulin secretion in cultured
  pancreatic beta cells.
• - Sonoda et al. (2008) PNAS May 6105(18)6614-9
• A unique mechanism of ß-blocker action
  Carvedilol stimulates ß-arrestin signaling
• - Wisler et al. 2007 PNAS October 16 104(42)
  1665716662.

24
Comparative potency of cannabinoid compounds in
second messenger and receptor ß-arrestin
interaction assays
Selected cannabinoid compounds showed good
correlation in efficacy the three assays tested.
No significant bias toward any activity.
24
McGuinness, et al , J Biomol Screen. 2009, 14,
49-58
25
Technologies sensitive to changes in signal
transduction paths may resolve complexity
CB2 Receptor AGONISTS
CB2-selective INVERSE AGONISTS
WIN
CP55,940
Anandamide
Sch.319
JTE-907
McGuinness, et al , J Biomol Screen. 2009, 14
49-58


ß-arrestin
ß-arrestin
Raborn, et al J. Neuroimmune Pharm. 2008, 3,
117-129
Src
Recycling
Cotton Claing Cellular Signalling2009, 21
1045-1053
RhoA
Raf
RhoA
DeFea, Brit. J.Pharma, 2008, 153 S298S309
ROCK
MEK1
Rac
ROCK
MAPK
ERK 1/2
Actin Stress Fibers Adhesion, motility etc
Membrane Ruffling
Actin Reorganization
Actin Stress Fibers Adhesion, motility etc
25
26
Mass changes at interface detected using CORNING
Epic resonance waveguide grating technology
Penetration limit
Stimulation

• Operating Principal Cell-based Assays
• Measures changes in local index of refraction
  resulting from ligand-induced mass redistribution
  within bottom region (150 nm) of the cell
  monolayer
• Change in index measured as shift in resonance
  wavelength

FEBS Letters 579 (2005) 41754180 Probing
cytoskeleton modulation by optical biosensors Ye
Fang, Ann M. Ferrie, Guangshan Li Biochemical
Technologies, Corning Incorporated, Corning, NY
14870
27
Effect G-protein coupling on EPIC signal
J. Pharmacological and Toxicological
Methods(2007) 55 (3) 314-322 Non-invasive
optical biosensor for assaying endogenous G
protein-coupled receptors in adherent cells Ye
Fang, Guangshan Li and Ann M. Ferrie, Corning
Incorporated, Corning, NY
Gq receptor
Gs receptor
Gi receptor
Agonist added
Agonist added
Agonist added
Fig. 1. The GPCR signaling and its DMR signal.
(a) thrombin (40 unit/ml) in CHO cells. (b)
epinephrine (25 nM) in A431 cells. (c) LPA
(200 nM) in A431.
28
Profiling biology of b2 adrenoceptor ligands
using optical biosensor
FEBS Lett. (2008), 582(5)558-64. Label-free
optical biosensor for ligand-directed functional
selectivity acting on b2 adrenoceptor in living
cells Ye Fang, Ann M. Ferrie Biochemical
Technologies, Corning Incorporated, Corning, NY
14831
Fig. 1. The structures of b2AR ligands and their
DMR in quiescent A431 cells. The ligands included
(-) epinephrine (8 nM), (-) isoproterenol (10
nM), norepinephrine (100 nM), dopamine (32 µM),
halostachine (500 µM), catechol (500 µM),
tyramine (125 µM), phenylethylamine (500 µM),
salmeterol (8000 nM), salbutamol (164 nM),
labetalol (2 µM), xamoterol (1 µM), pindolol (8
µM), S (-) pindolol (8 µM), CGP12177 (100 nM),
and alprenolol (4 µM). The grey arrows indicated
the time when the agonist was introduced.
29
The grating of the BIND biosensor reflects only a
single wavelength (Peak Wavelength Value or
PWV). When a biomolecule or cell binds to the
biosensor surface, this reflected wavelength
increases. Real time binding can be observed by
measurement of the shift in PWV over time.
30
CONCLUSION Roles for no-label technologies in
drug discovery effort

• Identification of chemotypes that might be missed
  using non-biological ligands
• Intermediate throughout screening tool for
  refractory targets
• Targets with complex ligands
• Targets where fluorophore alters ligand binding
• Compound profiling
• Identification of unknown properties of chemotype
• Characterizing efficacy within primary cell
  preparations

31
Acknowledgements
SPRI / Infectious Disease Erik Langsdorf Lynn
Miesel Dayna Daubaras Cynthia Kravec Todd
Black SPRI / New Lead Discovery Asra
Malikzay Steve Cifelli Jane Wen Debra
McGuinness Ellen Barrabee Richard Hart Frederick
Monsma Marvin Bayne SPRI / Cambridge D. Allen
Annis Charles E Whitehurst SPRI / Structural
Chemistry Rumin Zhang BioTrove, Inc. Maxine
Jonas William A. LaMarr Can Ozbal
The Drug Discovery Facility Schering-Plough
Research Institute, Kenilworth, NJ 07033