Transcript Document

Conformation of Ligands on Macromolecules
•
Introduction: Diversity of Complexes Studied by NMR
•
Influence of Chemical Exchange on NMR Spectra of ReceptorLigand Complexes
– Fast vs. Slow Exchange
•
Characterizing Exchange Rates by NMR
•
NMR Techniques for Studying Receptor-Ligand Interactions and
Ligand Conformations on Receptors
– Deuteration
– Isotope Filtering
– Transfer NOE
References
General
Craik, D. J. and J. A. Wilce (1997). “Studies of protein-ligand
interactions by NMR.” Methods in Molecular Biology 60: 195232.
Methods of Enzymology 239 (1994)
Section IV, pp. 657-767, Protein-Ligand Interactions
Feeney, J. and B. Birdsall (1993). NMR Studies of Protein-Ligand
Interactions. NMR of Macromolecules: A Practical Approach. G.
C. Roberts, Oxford: 183-215.
Wand, A. J. and S. W. Englander (1996). “Protein complexes studied
by NMR spectroscopy.” Current Opinion in Biotechnology 7(4):
403-8.
Ligand Binding and Chemical Exchange
“Biomolecular NMR Spectroscopy” by Evans (1995)
Chapter 1.3.1 – 1.3.3, pp. 43-46, Kinetics
Lian, L. Y. and G. C. Roberts (1993). Effects of Chemical Exchange on NMR
Spectra. NMR of Macromolecules: A Practical Approach. G. C. Roberts,
Oxford: 153-182.
Isotope Filters
Breeze, A. L. (2000). “Isotope-filtered NMR methods for the study of
biomolecular structure and interactions.” Progress in Nuclear Magnetic
Resonance Spectroscopy 36(4): 323-372.
Gronenborn, A. M. and G. M. Clore (1995). “Structures of protein complexes by
multidimensional heteronuclear magnetic resonance spectroscopy.” Critical
Reviews in Biochemistry & Molecular Biology 30(5): 351-85.
Otting, G. and K. Wuthrich (1990). “Heteronuclear filters in two-dimensional
NMR” Quarterly Reviews of Biophysics 23(1): 39-96.
Transferred NOE
“Biomolecular NMR Spectroscopy” by Evans (1995)
Chapter 6.5, pp. 246-247, Transferred NOE
Ni, F. (1994). “Recent Developments in transferred NOE methods.” Progress in
Nuclear Magnetic Resonance Spectroscopy 26: 517-606.
Campbell, A. P. and B. D. Sykes (1993). “The two-dimensional transferred
nuclear Overhauser effect: theory and practice.” Annual Review of
Biophysics & Biomolecular Structure 22: 99-122.
Deuteration
LeMaster, D. M. (1990). “Deuterium labelling in NMR structural analysis of
larger proteins.” Quarterly Reviews of Biophysics 23(2): 133-74.
Diversity of Complexes Studied Using NMR
•
protein-peptide
CD4 36-59 peptide/HIV gp120 (transfer noe)
mellitin peptide/calmodulin (deuteration)
pY peptide from PDGF receptor/SH2 of PLCC (isotope filter)
•
receptor-ligand
cyclophilin/cyclosporin A (isotope filter)
FKBP/FK506 (isotope filter)
•
enzyme-substrate/inhibitor
NAD+/lactate dehydrogenase (transfer noe)
glutathione/glutathione transferase (transfer noe)
•
protein-carbohydrate
Slex tetrasaccharide/E-, P-, L-selectin (transfer noe)
•
antibody-antigen
Fab/Fv fragment
•
protein-nucleic acid
homeodomain/DNA fragment (isotope filter)
U1A protein/ RNA hairpin (isotope filter)
•
protein-protein
p53 tetrameric oligomerization domain (isotope filter)
Binding Equilibria
kon
E + L
EL
koff
where: kon is a bimolecular rate constant that is a measure of probability of
productive encounter between free receptor and ligand – kon is limited by
diffusion controlled rate 107 – 108 M-1 s-1
koff is a unimolecular rate constant that is inversely proportional to t – the
lifetime of the complex
The binding affinity is given by equilibrium dissociation constant:
KD = [E][L]/[EL] = koff/kon
As a rough approximation:
• tight binding (slow exchange) is characterized by KD << mM
• weak binding (fast exchange) is characterized by KD >> mM
Effect of Exchange on NMR Spectra
Suppose a nucleus is able to experience two different environments,
what would its nmr signal look like?
1.5ppm
1.0ppm
H
H
If exchange between the two environments was turned off, we
would see two signals – the time domain signal or FID would be a
superposition of the FID of each signal
low freq. component = 1.5ppm signal
FT
1.5ppm
high freq. component = 1.0ppm signal
1.0ppm
The frequency domain spectrum is a superposition of the individual
spectra.
Slow Exchange
What happens if we turn exchange between the two sites on?
1.5ppm
1.0ppm
H
H
If exchange between the two environments is very slow (lifetime of
complex is long):
kex << Dn
the frequency (or chemical shift) of each environment is sampled
low freq. component = 1.5ppm signal
before exchange
exchange event
FT
1.5ppm
high freq. component = 1.0ppm signal
1.0ppm
The frequency domain spectrum contains two signals representing
a superposition of the individual signals (the intensities of the signals
would represent the relative amounts of each species).
Fast Exchange
1.5ppm
1.0ppm
H
H
If exchange between the two environments is very fast (lifetime of
complex is short):
kex >> Dn
each sampled point is a weighted average of the points from the free
and bound states
FT
weighted average of free and bound points
1.25ppm
The frequency domain spectrum contains a single signal whose
chemical shift is a weighted average of the chemical shifts of the free
and bound states:
dobs = pfree*dfree + pcomplex*dcomplex where p represents mole fraction
Intermediate Exchange
As the lifetime of complex approaches the frequency difference of
the signals corresponding to the free and bound states:
kex ~ Dn
the signals become “exchange broadened”:
slow exchange limit
intermediate exchange
fast exchange limit
Characterizing Fast/Slow Exchange by NMR
Before we attempt to study the structure of a ligand in a complex, it
is useful to determine if the complex is in fast or slow exchange on
the NMR timescale. This is often done by titrating the ligand (L) into
a sample containing the receptor (R) and monitoring the NMR
spectrum:
Titration under conditions of:
slow exchange
fast exchange
[L]:[R]
0:1
0.5:1
1:1
1.5:1
2:1
Rf
Rb
Lf Lb
Rf
Rb
Lf Lb
NMR Techniques for Determining the Structure
of a Ligand in a Complex
To determine the structure of a bound ligand using NMR methods,
we must be able to assign the signals belonging to the ligand within
the complex and determine structural constraints belonging to those
signals (ie. NOEs). To determine the structure of a ligand bound to a
receptor, we must be able to distinguish the ligand signals from
receptor signals:
+
spectrum of ligand
=
spectrum of receptor
spectrum of complex
Note: not necessarily a
superposition of ligand
and receptor spectra
How to separate signals of receptor from those of ligand within the
complex?? To distinguish ligand signals from receptor signals, the
following techniques are available:
o DEUTERATION
o ISOTOPE LABEL/FILTER
o TRANSFER NOE
Deuteration of Receptor
replacement of the protons by deuterons in the receptor will
eliminate the signals of the receptor within the complex:
1H
1H
2H
1H
deuteration of receptor
spectrum of complex
spectrum of ligand
bound to deuterated
receptor
Deuteration of the receptor would normally be used to study the
structure of a ligand in a complex under conditions of slow exchange.
This method has largely been superseded by isotope filtering
methods because:
 expense of deuteration
 ability to edit spectra of complexes for the ligand signals or
receptor signals in isotope filtering experiments
 exchangeable protons on receptor (ie. amide protons) are not
removed from spectra collected in water
Hsu, V. L. and I. M. Armitage
“Solution structure of cyclosporin A and a nonimmunosuppressive
analog bound to fully deuterated cyclophilin.”
Biochemistry 31, 12778 (1992).
objective: determine the structure of the immunosuppressive drug
cyclosporin bound to the immunophilin cyclophilin.
complex: Cyclosporin A (CsA, cyclic 11-mer peptide) + cyclophilin (CyP,
17.7 kDa)
background:
• cyclosprin A exhibits immunosuppressive activity (prevents organ/bone
marrow transplant rejection).
• CyP, a cytosolic protein immunophilin, binds to CsA, and has proline cistrans isomerase activity that is inhibited by CsA.
• Structures reported for CsA in crystal and in CHCl3 solution indicate
presence of a cis peptide bond. Is cis bond an important feature of
complexed CsA?
complex: Cyclosporin A (CsA, cyclic 11-mer peptide) + cyclophilin (CyP,
17.7 kDa)
binding: KD ~ 10-8 M
sample and methods:
• perdeuterated CyP by recombinant expression in bacteria grown in
deuterated algal hydrolysate in D2O
• NMR sample consisted of 0.4mM complex with excess CyP
• 2D homonuclear experiments run (COSY,TOCSY,NOESY)
• 66 intraresidue and 55 interresidue NOEs obtained and used in XPLOR
(simulated annealing protocol)
result:
• structures generated had 0.54A rmsd for backbone
• bound structure contains only trans peptide bonds and shows no regular
secondary structure (free CsA in organic solvents and in the crystal has a b
sheet structure)
• intermolecular nOes were identified by comparing NOESY spectra of the
CsA/[2H]-CyP complex with a fully protonated complex – supports
prediction that the CyP Trp indole ring is located in the binding site.
Isotope Filtering
labeling of the receptor (or ligand) with 13C/15N permits the use of
an “isotope filtering” nmr experiment to select for signals in the
spectrum from protons that are either bonded to 13C/15N or 12C/14N:
1H
1H
13C/15N
1H
12C/14N
12C/14N
isotope filter to select for 12C/14N
spectrum of complex
1H
13C/15N
1H
12C/14N
spectrum of ligand
1H
13C/15N
isotope filter to select for 13C/15N
spectrum of complex
spectrum of receptor
The Isotope Filter
In an isotope filtered experiment, a 90o pulse is replaced by:
1H
1/2JXH
1/2JXH
receiver
X
phase x
x x x
-x
x
-x
or
x
x
selects X-1H
LB
RB
selects for 1H
that are not
X-1H
If the receptor (R) or ligand (L) is isotopically labeled, an isotope
filter at each end of a NOESY can be use to select for:
 NOEs between protons on the ligand ( )
 NOEs between protons on the receptor ( )
 NOEs between ligand protons and receptor protons ( )
depending on the phase cycling scheme used. Isotope filtering can be
used to study the structure of a ligand in a complex under conditions
of fast or slow exchange.
RA
LA
2D NOESY spectrum
RA
LA
LB
RB
R. R. Rustandi, D.M. Baldisseri, and D. J. Weber
“Structure of the negative regulatory domain of p53 bound to
S100B(bb)”
Nature Structural Biology 7, 570 (2000).
objective: determine the structure of p53 peptide – S100B(bb) complex to
gain insight on how to develop inhibitors that block the Ca2+ dependent
interaction between the two proteins.
complex: peptide derived from p53 (22 residues, S367-E388) bound to
S100B(bb) (~13 kDa) forms a quaternary complex consisting of two p53
peptides per S100B(bb) dimer.
background:
• the tumor suppressor protein p53 is a transcription activator that signals for
cell cycle arrest and apoptosis
• the c-terminal negative regulatory domain (residues 367-392) contains a site
that activates p53 when phosphorylated
• members of the S100 protein family are overexpressed in tumor cells and
their function is tightly regulated by Ca2+ concentrations
• upon binding Ca2+, S100B(bb) undergoes a large conformational change
which exposes a hydrophobic patch that is required for binding to p53
• the Ca2+ dependent interaction of p53 with dimeric S100B(bb) causes
inhibition of p53 dependent transcription
binding: KD ~ 20 mM
sample and methods:
• 3-6 mM [13C/15N]-S100B(bb) + 5-10mM unlabeled peptide with 6-13mM
Ca2+
• asymmetrically labeled S100B(bb): 50% unlabeled S100B(bb) + 50%
[13C/15N]-S100B(bb)
• peptide + [2H]-S100B(bb)
• assign peptide in complex using 15N-filtered & 13C-filtered NOESY/TOCSY
and 1H-1H NOESY/TOCSY on peptide + [2H]-S100B(bb)
result:
• determine and compare S100B(bb) structures in the apo, Ca2+ and p53
peptide bound states
• p53 peptide is random coil in absence of S100B(bb) but adopts a helical
conformation when bound to Ca2+ loaded S100B(bb)
The Transfer NOE
The transfer NOE is typically used to study the structure of a
ligand in a complex under conditions of fast exchange. No isotopic
labeling of the ligand or receptor is necessary. Even under conditions
where there is only a small percentage of bound ligand in solution,
the “memory” of NOEs present within the bound state conformation
of the ligand are carried over to the ligand free in solution under
conditions of fast exchange:
Ha
Ha
Hb
Ha
Hb
2D NOESY spectrum
before addition of receptor
Hb
Ha
Ha
Hb
Hb
Ha
Hb
2D NOESY spectrum
after addition of small amount
of receptor (a few percent)
Gizachew et al.
“NMR studies on the conformation of the CD4 36-59 peptide
bound to HIV-1 gp120”
Biochemistry 37, 10616-25 (1998)
objective: Determine the gp120-bound conformation of CD4 to better
understand the gp120-CD4 receptor interaction at the molecular level and
help elucidate the mechanism of HIV entry into the cell.
complex: CD4 36-59 peptide with HIV-1 gp120. The peptide is known to
block the interaction with gp120.
background:
• The HIV-1 gp120 is a viral surface glycoprotein that binds to CD4
receptor and allows the virus to gain entry into T-lymphocyte CD4+ cells.
• Peptide of human CD4 receptor (435 residues) - structures of two
different crystal forms are different.
• Peptide is random coil free in solution.
binding: KD ~ 10-9 of extracellular domain of receptor
KD ~ 100-500 mM for peptide
methods:
• NMR sample consisted of 2mM peptide and 66mM gp120
• NOESY at 6 different mixing times
• simulated annealing with 107 constraints (4 long-range and 42 medium
range)
result:
• structures different than those found in crystal
• see decreases in NOE as CD4 titrated in
NMR-Based Screening in Drug Discovery
References
Reviews
Peng, J. W., J. Moore and N. Abdul-Manan (2004). "NMR experiments for
lead generation in drug discovery." Progress in Nuclear Magnetic Resonance
Spectroscopy 44(3-4): 225-256.
Lepre, C. A., J. M. Moore and J. W. Peng (2004). "Theory and applications
of NMR-based screening in pharmaceutical research." Chemical Reviews
104(8): 3641-3675.
Meyer, B. and T. Peters (2003). "NMR Spectroscopy techniques for
screening and identifying ligand binding to protein receptors." Angewandte
Chemie-International Edition 42(8): 864-890.
Stockman, B. J. and C. Dalvit (2002). "NMR screening techniques in drug
discovery and drug design." Progress in Nuclear Magnetic Resonance
Spectroscopy 41(3-4): 187-231.
Hajduk, P. J., R. P. Meadows and S. W. Fesik (1999). “NMR-based screening
in drug discovery.” Quarterly Reviews of Biophysics 32(3): 211-240.
Techniques
• SAR by NMR
Shuker, S. B., P. J. Hajduk, R. P. Meadows and S. W. Fesik (1996).
"Discovering high-affinity ligands for proteins: SAR by NMR." Science
274(5292): 1531-4.
Hajduk, P. J., G. Sheppard, D. G. Nettesheim, E. T. Olejniczak, S. B. Shuker,
R. P. Meadows, D. H. Steinman, G. M. Carrera, P. A. Marcotte,
J. Severin, K. Walter, H. Smith, E. Gubbins, R. Simmer, T. F. Holzman,
D. W. Morgan, S. K. Davidsen, J. B. Summers and S. W. Fesik (1997).
“Discovery of Potent Nonpeptide Inhibitors of Stromelsin Using SAR by
NMR.” Journal of the American Chemical Society 119: 5818-5827.
• STD NMR
Mayer, M. and B. Meyer (1999). “Characterization of ligand binding by
saturation transfer difference NMR spectroscopy.” Angewandte Chemie
International Edition 38(12): 1784-1788.
• Relaxation Edited NMR
Hajduk, P. J., E. T. Olejniczak and S. W. Fesik (1997). “One-dimensional
relaxation- and diffusion-edited NMR methods for screening compounds
that bind to macromolecules.” Journal of the American Chemical Society
119(50): 12257-12261.
• Diffusion Edited NMR
Bleicher, K., M. Lin, M. J. Shapiro and J. R. Wareing (1998). “Diffusion
Edited NMR: Screening Compound Mixtures by Affinity NMR to Detect
Binding Ligands to Vancomycin.” Journal of Organic Chemistry 63:
8486-8490.
Techniques
• H2O-editing
Dalvit, C., P. Pevarello, M. Tato, M. Veronesi, A. Vulpetti and M.
Sundstrom (2000). “Identification of compounds with binding affinity to
proteins via magnetization transfer from bulk water.” Journal of
Biomolecular Nmr 18(1): 65-68.
NOE-based Methods
• Transfer NOE
Meyer, B., T. Weimar and T. Peters (1997). “Screening mixtures for
biological activity by NMR.” European Journal of Biochemistry 246(3): 7059.
• NOE pumping:
Chen, A. and M. J. Shapiro (1998). “NOE pumping: A novel NMR
technique for identification of compounds with binding affinity to
macromolecules.” Journal of the American Chemical Society 120(39):
10258-10259.
NMR screening in Drug Discovery
Many therapies rely on the use of a drug that interacts with a cellular
target and alters its activity. Drug discovery is the process of
identifying and optimizing drug candidates and is usually a time
consuming task that consists of several critical stages:
• target selection
• lead generation
• lead optimization
• preclinical development
The lead generation step identifies starting compounds or “leads”
with binding activity against the target and usually requires the
screening of a very large number of compounds (>106).
It is important to note that most compounds in a library will not bind
to the target and those that interact, bind only weakly (KD ~ uM-mM)
Compounds in the library that interact with the target are likely to bind
only weakly (uM – mM). NMR is ideally suited for screening a library of
compounds for weak to tight binding in a high throughput fashion and is
utilized in one of the following modes:
• target or receptor-based screening - does ligand interact with target by
following changes in the target chemical shifts. Observe and compare
chemical shifts of target in the absence and presence of ligand (SAR by
NMR)
• ligand-based screening – does ligand interact with target by following
changes in ligand NMR parameters upon adding target:
• saturation transfer difference (STD)
• relaxation editing
• diffusion editing
• hydration
• NOE based methods
SAR by NMR
SAR by NMR is carried out in five main steps:
1.
Identification of ligands with high binding affinity from a library of
compounds by using 2D 1H-15N HSQC
2.
Optimization of ligands by chemical modification
3.
Identification of ligands binding in the presence of saturating amounts of
optimized ligands from step 2 by using 2D 1H-15N HSQC
4.
Optimization of ligands for the second site
5.
Tethering the two ligands from step 2 and step 4 in various positions and
checking again for binding
Shuker et al. (Science 274, 1531 (1996)) used SAR by NMR to
screen 1000 substances for FK506-like binding activity. FK506 is an
important drug that binds FKBP and suppresses rejection reactions
after organ transplants but it has high toxicity. SAR by NMR yielded
a tethered ligand with high affinity for FKBP (KD=19 nM). The
complex was then subjected to a detailed structural analysis using
NMR.
STD NMR – Saturation Transfer Difference
STD involves selectively saturating a resonance that belongs to the
receptor (must find a region of the spectrum that contains only resonances
from receptor such as 0 ppm to -1 ppm). If ligand binds, saturation will
propagate from the selected receptor protons to other protons of receptor
via spin diffusion and then the saturation is transferred to binding
compounds by cross relaxation at the ligand-receptor interface. A control
experiment is done by saturating a region of the spectrum that does not
contain signal. The resulting difference spectrum yields only those
resonances that have experienced saturation, namely those of the receptor
and those of the compound that binds to the receptor. The receptor is
typically present at very small concentrations so its resonances will not be
visible and, if needed, can be eliminated by a relaxation filter.
Mayer and Meyer (Angewandte Chemie 38, 1784 (1999) used STD
NMR to screen a library of carbohydrate molecules for binding
activity towards a carbohydrate binding protein, wheat-germ
agglutinin (WGA). STD can also be used for determining the
binding epitope of the ligand, information that is of prime
importance for the directed development of drugs.
Epitope Mapping Using STD NMR
STD NMR can be used to identify the ligand fragments or epitopes
that are in direct contact with the receptor because these components
receive the highest degree of to saturation.
Diffusion Edited Screening
A variety of NMR pulse sequences has been developed that allow the
investigation of diffusion constants in solution. On the basis of such
experiments it is possible to discriminate compounds in mixtures
according to their diffusion properties. This so-called diffusion editing has
been successfully applied to deconvolute compound mixtures, and to
detect molecular association processes. For small- and intermediate-size
molecules the approach works well, and in principle, diffusion editing is
also applicable to the detection of binding of low-molecular-weight
compounds to large protein receptors.
First, a PFG-STE or PFG-SE spectrum of the chemical mixture in the
absence of the protein is recorded at a low gradient strength. Next, the
same spectra for the chemical mixture in the presence of the protein are
recorded at low and high gradient strengths and subtracted to produce a
spectrum that contains only the signals of the compounds not interacting
with the protein. The resulting subtracted spectrum is then subtracted from
the spectrum of the chemical mixture recorded in the absence of the
protein to obtain a spectrum that contains only the signals of the molecule
that binds to the receptor.
Hajduk et al. (JACS 119, 12257 (1997)) used diffusion editing to
demonstrate that 4-cyano-40-hydroxybiphenyl, which binds to
stromelysin with a dissociation constant of 20 mM, was easily
identified from a mixture containing eight other non-binding
compounds.
A comparison of methods for ligand-based and target-based NMR
screening in drug discovery: