Nuclear Magnetic Resonance (NMR) Data Protein - PowerPoint PPT Presentation


PPT – Nuclear Magnetic Resonance (NMR) Data Protein PowerPoint presentation | free to download - id: 7fed7-ZDc1Z


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation

Nuclear Magnetic Resonance (NMR) Data Protein


Bio-molecular structure determination. Structure/imaging from ... relaxation, TROSY (molecular weight 100 kDa) ... binding and molecular interactions in ... – PowerPoint PPT presentation

Number of Views:337
Avg rating:3.0/5.0
Slides: 80
Provided by: niv2
Learn more at:


Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Nuclear Magnetic Resonance (NMR) Data Protein

Nuclear Magnetic Resonance (NMR) Data
ProteinProtein Docking
  • Presented by
  • Nivya Papakannu
  • ECE Department, UMASS Amherst

  • Introduction to protein structure
  • Peptide (protein) sequences
  • Primary, Secondary, Tertiary and Quaternary
  • Introduction to NMR
  • Protein-Protein Interaction.
  • Mapping Protein-Protein Interactions in Solution
    by NMR Spectroscopy
  • Nuclear Overhauser Effect (NOE)
  • Chemical shift
  • Titrations with NMR

  • Fast Mapping of Protein-Protein Interfaces by NMR
    Spectroscopy(chemical shifts and unassigned NMR
  • Residual Dipolar Coupling
  • Structures of Protein-Protein Complexes Are
    Docked Using Only NMRRestraints from Residual
    Dipolar Coupling and Chemical Shift Perturbations
  • Other NMR Methods
  • High-throughput inference of proteinprotein
    interfaces from unassigned NMR data
  • Summary

  • Peptides (and proteins) are composed of amino
    acids interlinked by amide bonds.
  • Amides are made by condensing together a
    carboxylic acid and an amine
  • a -COOH, which is a carboxyl group (acidic).
  • a -NH2, which is an amino group (basic).
  • an -H hydrogen.
  • a residue R which varies depending on the amino
  • Any number of amino acids can be joined together
    to form peptides
    of any length.
  • The peptide has a "backbone" and side-chains. The
    backbone atoms consist of the peptide amide units
    and the carbons the side-chains consist of the
    remaining atoms in the molecule (i.e. the "R"
    groups of each amino acid)

Types of Amino Acids
  • Proteins are not linear molecules as suggested
    when we write out a "string" of amino acid
    sequence, -Lys-Ala-Pro-Met-Gly- etc., for
  • This "string" folds into an intricate 3-D
    structure, unique to each protein .It is this 3-D
    Structure that allows proteins to function.
  • To understand the details of protein functions,
    we must understand the protein structure.
  • The Protein structure is broken down into four
  • Primary Structure refers to linear" sequence of
    amino acids.
  • Secondary structure is the local spatial
    arrangement of a polypeptides backbone atoms
    without regard to the conformations of its side
  • Tertiary structure refers to the
    three-dimensional structure (global folding) of
    an entire (single) polypeptide.
  • Quaternary structure involves association of two
    or more polypeptide chains, loosely referred to
    as subunits (refers to the spatial arrangement of
    the subunits).

Primary Structure of Protein
  • Proteins, like peptides, are composed of amino
    acids joined together via amide linkages.
  • The only difference between peptides,
    polypeptides and proteins is the number of amino
    acid residues in the chain.
  • Generally, peptides are small 10 or 20 residues
  • polypeptides might range up to 50 or 60 residues,
    with anything larger considered a protein.
  • Primary structure is sometimes called the
    "covalent structure" of proteins because, with
    the exception of disulfide bonds all of the
    covalent bonding within proteins defines the
    primary structure.
  • In contrast, the higher orders of protein
    structures (i.e. secondary, tertiary and
    quaternary) involve mainly non-covalent

Primary Structure
Secondary Structure of protein
  • Local spatial arrangement of a polypeptides
    backbone atoms without regard to the
    conformations of its side chains.
  • The most common secondary structure elements in
    proteins are the alpha-helix and the Beta-sheet.
  • Alpha-helix
  • Right-handed it turns in the direction that the
    fingers of a right hand curl when its thumb
    points in the direction that the helix rises.
  • The alpha helix has 3.6 residues per turn and a
    pitch (the distance the helix rises along its
    axis per turn) of 5.4 Å.
  • The alpha helices of proteins have an average
    length of ,12 residues, which corresponds to over
    three helical turns, and a length of ,18 Å.
  • Stabilized by hydrogen bonds between the carbonyl
    oxygen of one amino acid and the backbone
    nitrogen of a second amino acid located four
    positions away.
  • Amino acid side chains project outward and
    downward from the helix to avoid interference
    with the polypeptide backbone and with each

  • Stabilized by hydrogen bonds between the carbonyl
    oxygen of an amino acid in one strand and the
    backbone nitrogen of a second amino acid in
    another strand. (hydrogen bonding occurs between
    neighboring polypeptide chains rather than within
    one as in alpha helix)
  • Beta sheets can be either parallel or
  • Anti-parallel beta sheet - neighboring
    hydrogen-bonded polypeptide chains run in
    opposite directions.
  • Parallel beta sheet - hydrogen-bonded chains
    extend in the same direction.
  • Beta Sheets in proteins contain 2 to greater
    than12 polypeptide strands, with an average of 6
  • Each strand may contain up to 15 residues, the
    average being 6 residues.

Tertiary Structure of protein
  • The tertiary structure of a protein describes the
    folding of its secondary structural elements and
    specifies the positions of each atom in the
    protein, including those of its side chains.
  • The atomic coordinates of these structures are
    deposited in a database known as the Protein Data
    Bank (PDB) which allows the tertiary structures
    of a variety of proteins to be analyzed and
  • The common features of protein tertiary structure
    reveal much about the biological functions of the
    proteins and their evolutionary origins.
  • Structures are determined through X-ray
    crystallographic or nuclear magnetic resonance
    (NMR) studies.
  • Amino acid side chains in globular proteins are
    spatially distributed according to their
  • Non-polar amino acids are "hidden" within the
    structure, out of contact with the aqueous
    solvent (hydrophobic).
  • Charged polar residues are exposed on the outer
    surface, in contact with the aqueous solvent
  • Uncharged polar groups are usually on the protein
    surface but also occur in the interior of the
    molecule. When buried in the protein, the
    formation of a hydrogen bond neutralizes their

  • Certain groupings of secondary structural
    elements,called supersecondary structures or
    motifs, occur in many unrelated globular
  • Most common form of supersecondary structure is
    the bab-motif, in which an a-helix connects two
    parallel strands of a b-sheet.
  • Another common supersecondary structure, the
    b-hairpin motif, consists of anti-parallel
    strands connected by relatively tight reverse
  • An aa-motif, two successive anti-parallel
    a-helices pack against each other with their axes
  • Extended b-sheets often roll up to form bb
  • Motifs may have functional as well as structural
    significance. For example, a babab unit, in which
    the b strands form a parallel sheet with helical
    connections, often acts as a nucleotide binding
    site. In most proteins that bind dinucleotides
    two such babab units combine to form a motif
    known as a dinucleotide-binding fold.

Quaternary Structure of Protein
  • Consists of more than one polypeptide chain with
    multi subunits.
  • Usually, each polypeptide within a multi-subunit
    protein folds more-or-less independently into a
    stable tertiary structure and the folded subunits
    then associate with each other to form the final
  • Stabilized mainly by non-covalent interactions
    all types of non-covalent interactions like
    hydrogen bonding, Vander walls interactions and
    ionic bonding, are involved in the interactions
    between subunits.
  • Hemoglobin is one example of a multi-subunit
    protein. It has an a2b2 structure consisting of
    four polypeptides, two alpha subunits and two
    beta subunits.

Nuclear Magnetic Resonance (NMR)
NMR is a physical phenomenon based upon the
magnetic property of an atom's nucleus. NMR
studies a magnetic nucleus, like that of a
hydrogen atom, by aligning it with an external
magnetic field and perturbing this alignment
using an electromagnetic field. The response to
the field (the perturbing), is what is exploited
in NMR spectroscopy.
NMR Data
Bio-molecular structure determination
  • Structure/imaging from molecules to animals

History of nuclear magnetic resonance
  • 1946 Bloch, Purcell First nuclear magnetic
  • 1955 Solomon NOE (nuclear Overhauser effect)
  • 1966 Ernst, Anderson Fourier transform NMR
  • 1975 Jeener, Ernst Two-dimensional NMR
  • 1985 Wüthrich First solution structure of a small
  • from NOE-derived distance restraints
  • NMR is about 25 years younger than X-ray
  • 1987/8 3D NMR 13C, 15N isotope labeling
  • 1996/7 New long-range structural parameters
  • - residual dipolar couplings (also anisotropic
  • - cross-correlated relaxation, TROSY (molecular
    weight gt 100 kDa)
  • 2003 First solid-state NMR structure of a small
  • Nobel prizes
  • 1952 Physics Bloch (Stanford), Purcell (Harvard)
  • 1991 Chemistry Ernst (ETH)

Structure determination by NMR
Why bio-molecular NMR?
  • Structure determination of bio-macromolecules
  • no crystal needed, native-like conditions
  • nucleic acids difficult to crystallize,
    affected by crystal packing
  • Ligand binding and molecular interactions in
  • NMR fingerprint - with residue/amino acid
    resolution !!!
  • Characterization of dynamics and mobility
  • conformational dynamics ? enzyme turnover,
    kinetics, folding
  • Molecular weight X-ray gt200 kDa, NMR lt 50-100
  • NMR and X-ray crystallography are complementary

Nuclear spin
NMR nuclear spins, magnetic moments and
nuclear magnetic dipole
A nuclear spin of I gt 0 is associated with a
magnetic dipole moment µ ?L
Why do we need a magnet? - Spin up, spin down
Nuclear Magnetic Resonance
  • Radio frequency pulses
  • Induce resonance by applying an external
    magnetic field that oscillates with the
    precession frequency of the spins (radio
    frequencies MHz)

A radio frequency pulse
Energy levels - why do we need a BIG magnet?
  • Spins in the a (up) and ß (down) states populate
    the energy levels according to a Boltzmann
  • This leads to a small macroscopically observable
    magnetization along the z-axis Mz (parallel to
  • No x- or y-magnetization is observed since the
    spin vectors are not phase coherent,
  • i.e. they precess independently from each other
    around B0, the x, y components average to zero
  • The energy difference between the two states
    scales with the magnetic field strength B0
  • A larger population difference resulting from
    this yields more nuclear magnetization!

The NMR experiment
A one-dimensional NMR experiment
Chemical shift - one-dimensional (1D) NMR
A 1D NMR spectrum of a protein
2D NMR COSY through bond correlation
Protein-Protein Interactions
  • The study of protein interactions has been vital
    to the understanding of how proteins function
    within the cell.

Protein A
Protein B
There are two classes of protein docking a)
Protein-Protein b) Protein-Ligand
Mapping Protein-Protein Interactions in Solution
by NMR Spectroscopy
NMR for protein-protein interactions
  • Is it desirable to use NMR for studying
    protein-protein interactions?
  • To understand protein-protein interactions is
    very important for biochemical processes in
    living organisms.
  • We need to know the fine details of the interface
    in protein complexes to understand how life is
    encoded and how diseases can be cured.
  • Individual structures do not disclose where the
    interactions in the complexes take place because
    many proteins adapt their conformations
    dramatically to improve the fit.
  • Resolving the structures of the complexes poses a
    great challenge for researchers though high
    resolution individual structures are available.
  • NMR is very well suited for the study of
    especially weak protein-protein interactions
    (i.e. non-covalent interactions), as no
    crystallization is required.
  • Scientists have used different NMR techniques to
    resolve these complex structures.

Nuclear Overhauser Effect (NOE)
  • This is a special kind of NMR experiment that can
    be done in 1D or in 2D.
  • It also gives information about how far apart the
    different nuclei are from the irradiated nuclei
    through the 1/r6 relation of the distance between
    the nuclei to the intensity of the NOE signal.

Nuclear Overhauser Effect
  • NOE is an unambiguous way of mapping
    bio-macromolecular interactions.
  • A full three-dimensional structure of the complex
    is determined, using many precise NOE-derived
    distance constraints between the two interacting
  • This method is only applicable when the
    interaction between the molecules is relatively

Nuclear Overhauser Effect
  • A powerful method to obtain the intermolecular
    NOE is the isotope-edited NOE.
  • Requires that the two interacting macromolecules
    should have different isotopic labeling patterns.
  • For instance, one partner has no labeling (i.e.,
    1H, 14N, 12C) while the other is labeled with
    stable isotopes, e.g., 1H, 15N, 12C or 1H, 15N,
    13C or sometimes even 1H, 2H,15N,13C.
  • Principle
  • Distinguishes between protons residing on labeled
    and unlabeled macro molecules which is an
    essential ingredient to map the structures.
  • The isotope editing method is very well suited
    for the study of labeled proteins and unlabeled
    nucleic acids.

Cross Saturation
  • Obtains low-resolution, but gives highly
    relevant, interface information quickly.
  • Governed by the same physical processes as the
    NOE experiment.
  • Donating partner protein is not labeled, while
    the observed protein is per-deuterated and
    15N-labeled, but its amide deuterons are
    exchanged back to protons.
  • Cross-relaxation carries the saturation from the
    donor to the acceptor protein amide protons,
    where it is detected using a 15N-1H HSQC or, for
    larger proteins, 15N-1H TROSY.
  • Those acceptor 15N-1H cross-peaks that change in
    intensity upon the donor saturation are very
    likely to be close to the intermolecular

Cross Saturation
  • This experiment is robust because only protons
    close to the interface will light up, even if
    long-range conformational changes occur.
  • Advantage of cross saturation over NOE
  • Because of its experimental simplicity, the
    method can be widely applied.
  • Disadvantage
  • NOE-based methods work only when complexes are
    relatively tight . Reason Weaker complexes are
    best described by an ensemble of interconverting
    structures, for which the concentrations of the
    individual structures are too small to give rise
    to detectable NOEs.
  • The cross-saturation experiment promises to be
    more sensitive for weaker interactions than the
    two- or three-dimensional isotope-edited NOE,
    because longer magnetization transfer times can
    be used.

Chemical Shift Perturbation Mapping
  • Chemical shift perturbation is the most widely
    used NMR method to map protein interfaces.
  • In this technique the 15N-1H HSQC spectrum of one
    protein is monitored when the unlabeled
    interaction partner is titrated in, and the
    perturbations of the chemical shifts are
  • The interaction causes environmental changes on
    the protein interfaces and, hence, affect the
    chemical shifts of the nuclei in this area.

Chemical Shift Perturbation Mapping
  • Chemical shift perturbations are very sensitive
    to subtle effects
  • Generally, Shift perturbation measurements just
    yield the locations of the interfaces on the
    individual binding partners. It is then still
    unknown how the partners interact on an
    atom-to-atom basis.
  • Disadvantage
  • In some cases, the entire protein may change
    conformation, and all shifts may be affected and
    perturbation fails as a mapping.

Titrations with NMR
  • Titrations with NMR allows affinity estimation,
    stoichiometry and kinetics of binding apart from
    mapping of the interface.
  • How the chemical shifts of the labeled protein
    change during the titration is determined by the
    kinetics of the interaction.
  • If the complex dissociation is very fast, there
    is, only a single set of resonances whose
    chemical shifts are the fractionally weighted
    average of the free and bound chemical shifts.
  • This regime is referred to as fast chemical
    exchange and is often observed for weaker
  • If all two-dimensional trajectories are linear
    and occur at the same rate, a single binding
    event is indicated.
  • If the trajectories for different, resonances
    occur at a different rate, and/or if they are
    curved, more than one binding site is implicated.

Titrations with NMR
  • If the complex dissociation is very slow, one
    observes one set of resonances for the free
    protein and one set for the bound protein. During
    the titration, the free set will disappear and
    will be replaced by the bound set.
  • This regime is referred to as slow chemical
  • In the intermediate chemical exchange case, the
    frequencies of the changing resonances become
    poorly defined, and extensive kinetic broadening
    sets in.

Fast Mapping of Protein-Protein Interfaces by NMR
  • NMR allows the study of protein interaction in
  • Major problem in NMR

  • - assignment of data before using the
    results from the NMR spectra.
  • Example NOE NMR data provides interatomic
    distance restraints in order to use these
    distance restraints in structure determination,
    we must first assign each restraint to a pair of
    nuclei in the protein
  • The assignment is typically done manually and is
    time consuming.
  • Example E1NHPr complex required 2 years
    of data analysis.

Ability of Mapping through NMR
  • Advantage of NMR
  • - Ability to map interfaces efficiently.
  • A prerequisite for using the chemical shift NMR
    mapping method, is that the assignment of the
    nuclei showing chemical shift changes be known.
  • Here a combinatorial approach was used to map the
    interface without chemical shift assignments.

Ability of Mapping through NMR
  • The method is based on preparing several protein
    samples, each one selectively 15N labeled with
    one particular amino acid.
  • The peaks of shifted amino acids of a certain
    type can be identified.
  • Combined results, from several differently
    labeled samples allows to define the minimum
    number of a certain type of amino acid located
    in the interface.
  • Comparison of this list with the known structure
    of the protein is then used to identify the
    interface (Figure 1).

Ability of Mapping through NMR
  • This approach is divided into two phases
  • - 1) Identify amino acid types with low
    abundance in the sequence. The combination of
    data from the rare residues allows us to identify
    the location of the binding site.
  • -2) Use more common residues to characterize the
    extent and shape of the interface.

  • Here in nNOS PDZ domain was studied.
  • Five lysines, four phenylalanines, two
    histidines, and one tyrosine as initial probes to
    identify the site of the interface.
  • 15N-lysine labeled nNOS PDZ was titrated with
    unlabeled PSD-95 PDZ2.
  • Two changes in the spectrum occurred.
  • One peak disappeared while another peak appeared,
    characteristic of a complex in slow exchange.
  • Two other peaks exhibited spectral broadening,
    characteristic of a complex in intermediate
  • This shows 2 distinct interfaces exist.
  • To distinguish these two interfaces,
  • - Binding site exhibiting slow exchange
    termed primary
    - Site exhibiting intermediate
    exchange - secondary.

Ability of Mapping by NMR
  • From the combinatorial approach makes lysine
    most likely candidate for the primary binding
  • To further investigate the interface, we labeled
    the nNOS with 15N-tyrosine, histidine, and
  • The results predict that the primary binding site
    contains at least one lysine, one histidine, and
    two phenylalanines, but does not contain
  • The secondary binding site contains two lysines,
    one tyrosine, no histidines, and one
  • Thus mapping for this complex was carried out.

(A) shows the 15N,1H-HSQC spectrum of the free
lysine-labeled PDZ domain from nNOS with five
peaks, corresponding to the five lysines. (B)
shows the spectrum of a sample in which 50 of
the nNOS PDZ domain molecules are in complex with
the PDZ2 domain from PSD-95. (C) shows a
spectrum of a 11 complex of both PDZ domains.
(D) His are magenta, Lys are red, Phe are green,
and Tyr are gold. (E). Arg are red, His are
magenta, Met are cyan, Phe are green, and Tyr are
gold. In summary, 2 His, 2 Tyr, 1 Phe, 1 Met, and
no Arg were seen to be involved in the primary
interface, and 1 His, 1 Tyr, 1 Arg, and neither
Phe nor Met were involved in the secondary
Residual dipolar couplings in NMR structure
Long-range information in NMR
  • a traditional weakness of NMR is that all the
    structural restraints are short-range in nature
    (in terms of distance, not in terms of the
    sequence), i.e. NOE restraints are only between
    atoms lt5 Å apart, dihedral angle restraints only
    restrict groups of atoms separated by three bonds
    or fewer
  • over large distances, uncertainties in
    short-range restraints will add up--this means
    that NMR structures of large, elongated systems
    (such as B-form DNA, for instance) will be poor
    overall even though individual regions of the
    structure will be well-defined.

long-range structure bad
to illustrate this point, in the picture at left,
simulated nOe restraints were generated from the
red DNA structure and then used to calculate the
ensemble of black structures
best fit superposition done for this end
short-range structure OK
Zhou et al. Biopolymers (1999-2000) 52, 168.
Residual dipolar couplings
  • The spin dipolar coupling depends on the distance
    between 2 spins, and also on their orientation
    with respect to the static magnetic field B0.
  • In solution, the coupling averages to zero as the
    molecule tumbles, so that splittings in resonance
    lines are not observed--i.e. we cant measure
    dipolar couplings.
  • In solids, on the other hand, the couplings dont
    average to zero, but they are huge, on the order
    of the width of a whole protein spectrum. This
    is too big to be of practical use in
    high-resolution protein work
  • compromise it turns out that you can use various
    kinds of media, for eg. liquid crystals, to
    partially orient samples, so that the dipolar
    coupling no longer averages to zero but has some
    small residual value

Residual dipolar couplings
Proteins tumbling isotropically in solution No
orientational bias Dipolar interaction averages
to zero with tumbling (act as free dipoles) No
observable dipolar coupling. Too small!
Proteins in a single crystal Complete
orientational bias Enormous dipolar coupling. Too
big! (Dipolar couplings as big as entire proton
spectral range)
filamentous phage, lipid bilayer
fragment, cellulose crystallite
Proteins dissolved in liquid but oriented
medium Some liquid crystals acquire macroscopic
order in a magnetic field e.g. bicelles,
filamentous phage, cellulose crystallites Collisio
ns with protein impart a slight orientational
bias A small residual dipolar coupling
results giving interpretable information
Measurement of Residual Dipolar Couplings
--regular HSQC --decoupled in both
dimensions --15N-1H splittings not observed
--HSQC without decoupling in 15N dimension --
isotropic solution --15N-1H splittings observed,
(92-95 Hz)
--HSQC without decoupling in 15N
dimension --partly oriented --15N-1H splittings
observed, with RDC!
  • This picture illustrates measurement of 15N-1H
    residual dipolar couplings for a protein in a 7
    bicelle (fragments of lipid bilayer) solution.
  • The bicelle preparation is isotropic (not
    ordered) at 25 C (left), allowing measurement of
    the scalar couplings. Upon heating to 35 C, the
    bicelle preparation becomes anisotropic
    (ordered) such that the measured coupling now
    includes an RDC component.
  • RDCs can therefore be measured by comparing
    spectra taken at the different temperatures.
  • RDCs can often be tuned by adjusting the
    composition of the liquid
  • crystal mixture.

SAG Strain induced alignment in a gel
pores in gel contain protein
axially compressed, radially stretched oblate
ellipsoid pores
radially compressed, axially stretched prolate
ellipsoid pores
regular polyacrylamide gel
  • Proteins can be incorporated into cylindrical
    polyacrylamide gels within NMR tubes.
  • If the gel is stretched or compressed, the pores
    become anisotropic and can impart partial order
    to a protein.

Interpretation of RDCs--what do they mean?
  • Spin dipolar interaction between two nuclei
    depends upon their relative position with respect
    to an external magnetic field.
  • The residual dipolar coupling will therefore be
    related to the angle between the internuclear
    axis and the direction of the partial ordering of
    the protein.
  • Because the internuclear axis will have a
    different orientation for different bonds in the
    protein, the RDCs will exhibit a broad range of

internuclear axis (bond vector)
axis of partial ordering
15N-1H residual dipolar coupling will differ for
these two residues.
RDCs give information about long-range order in
Note that the relative values of 15N-1H RDCs for
a set of amide nitrogen hydrogen pairs do not
depend upon the distance between those pairs,
only on their relative orientation with respect
to a common axis system!
two NH bond vectors far apart, but with same
two NH bond vectors close together
In other words, RDCs can in principle tell us the
relative orientation of two bond vectors even if
they are on opposite ends of the molecule.
Contrast this with NOE distance restraints and
dihedral angle restraints which define
short range order.
Illustration of effect of using residual dipolar
couplings on the quality of nucleic acid
structure determination by NMR
a) without rdc b) with rdc
Zhou et al. Biopolymers (1999-2000) 52, 168.
A problem with dipolar couplings is that one
cannot distinguish the direction of an
internuclear vector from its inverse. Thus the
two opposite orientations below give the same RDC
  • This ambiguity makes calculating a structure de
    novo (i.e. from a random starting model) using
    only residual dipolar couplings very
    computationally difficult.
  • If there is a reasonable starting model,
    however, this is not a problem. Thus residual
    dipolar couplings are especially good for
    refining models/low resolution structures.

Structures of Protein-Protein Complexes Are
Docked Using Only NMR Restraints from Residual
Dipolar Coupling and Chemical Shift Perturbations
Orientation restraints from RDC
  • A significant barrier to NMR structure
    determination of complexes using NOE is the
    reliance on the intermolecular distance
    restraints between atoms lt5 Å apart.
  • Clearly, for elongated systems it is desirable
    to use additional non-NOE-based restraints to
    determine structures of protein complexes.
  • The measurement of RDC from partially oriented
    protein complexes allows for the determination of
    long range order over the entire complex, but
    places no restraint on intermolecular
  • The orientation information contained in the RDC
    restraints can replace the orientation
    information in intermolecular NOEs

Restraints from DCS
  • Chemical shift perturbations offer another
    alternative to intermolecular NOEs.
  • Structures of protein-ligand complexes can be
    determined by minimizing the difference between
    experimental and simulated chemical shift
  • (DCS ?CSexp - ?CSsim)
  • The DCS restraints alone are effective toward
    restraining intermolecular translations, but less
    effective at restricting intermolecular
  • In this study the relative values of DCS and RDC
    combined together can be used to interpret
    protein-protein complexes (avoiding the need for
    NOE restraints).

  • EIN-HPr complex (3EZA) is used as a model system
    for the structure calculations.
  • There are a number of important assumptions made
    in performing calculations
  • 1) Assume that there is no substantial change in
    the backbone structure of EIN or HPr upon complex
  • 2) A simplification that no conformational
    changes in aromatic side chains, that may occur
    upon binding.
  • 3) Only ??CSexpt data for EIN residues were used
    to calculate the structures.

(No Transcript)
  • Alignment using RDC data for the complex
    establishes the relative orientations of EIN and
    HPr, but does not restrain intermolecular
  • The fit of the RDC data allows for two EIN-HPr
    orientations, only one of which places the
    experimental chemical shift perturbations for EIN
    and HPr in close proximity.
  • To restrain the EIN-HPr complex using chemical
    shifts, we minimize the difference between
    experimental and simulated chemical shift
  • Simulations of chemical shift differences are
    made by taking the difference between a reference
    (EIN and HPr separated by 300 Å) and bound
    complex (EIN and HPr in close proximity).

  • The experimental ??CSexpt data ranges from -0.38
    to 0.37 ppm. The calculated ?CSsim data has a
    similar range for the best structures.
  • DCS values for 945 EIN-HPr structures ranged from
    1.71 (best fit) to 6.17 (worst fit).
  • The mean DCS value for all 945 structures is
  • The DCS values for the 12 best structures range
    from 1.71 to 1.94
  • The mean structure of the 12 best structures has
    a DCS of 1.76.
  • The best Fit is the structure for which DCS is

Other NMR Methods
Mapping with Dynamics
  • 15N relaxation has been extensively used to
    describe the dynamics (motion) of the protein
    backbone. Information is obtained on rapid
    fluctuations at the pico- to nanosecond time
  • Here, we are interested in whether changes in
    dynamics upon intermolecular interactions can be
    used to map intermolecular interfaces.
  • In this vision, Upon interaction, one sub
    conformation is stabilized, corresponding to a
    change in population levels of the different
    conformations available to the protein.
  • Nevertheless, it appears that protein dynamics is
    too much an intrinsic part of the binding process
    to make it a reliable tool for interface mapping.

Mapping with Amide-Proton Exchange
  • Prediction of the protein interfaces done using
    the amide-proton exchange.
  • Amide-proton exchange in proteins occurs upon
    transient unfolding of structure, ranging from
    local to global.
  • Solvent accessibility to the transiently unfolded
    areas allows the exchange of the amide protons
    for other protons, or deuterons.

Mapping with Paramagnetics
  • Map the interfaces by comparing results in the
    presence and absence of paramagnetic ions.
  • Paramagnetic ions in solution broaden the NMR
    resonances of the solute. The effect is caused by
    transient dipolar interaction between the
    unpaired electron spins and the nuclei in the
  • When the electronic spins behave as free magnetic
    dipoles, no shifts of NMR resonances occur and
    the NMR broadening follows a simple r-6 law.
  • If a protein is placed in a solution containing
    paramagnetic species, only its surface residues
    will be affected by the paramagnetic broadening.

Mapping with Site-Specific Spin-Labeling
  • Mapping of interfaces by comparing results in the
    presence and absence of free stable radicals.
  • Free stable radical nitroxide (spin-label) TEMPO
    can interact specifically with cysteine groups
    engineered on the protein surface.
  • This method was used to determine the antibody
    interface with a spin-labeled peptide antigen .

High-throughput inference of proteinproteininter
faces from unassigned NMR data
  • In this paper the authors have put forth a
    computational approach where an efficient
    algorithm is developed for identifying the
    interface (not structure) between proteins using
    unassigned NMR data.
  • To identify the interface region in this
    approach we use data from
  • - unassigned chemical shifts
  • - unassigned RDCs
  • - Structural model of the protein

Unassigned chemical shifts
  • The algorithm uses
  • 1H-15N 2D HSQC data for a set of pairs (one pair
    per residue that contains the chemical shifts of
    the amide proton and nitrogen) of the protein.
  • NMR data from either amide exchange or water
    HSQC experiments identify the chemical shifts
    from the given HSQC spectrum associated with
    surface, or solvent accessible residues in the

Unassigned RDCs
  • The algorithm uses
  • RDC data that gives global orientation
    restraints on inter-nuclear vectors.
  • Here NH RDCs, give orientation information about
    backbone amide bond vectors.
  • Each RDC D is a real number, where
  • D DmaxvT Sv.
  • Dmax is the dipolar interaction constant
  • v is the internuclear vector
  • S is the 3 3 Saupe order matrix, or alignment
    tensor, which specifies the orientation of the
    protein in the laboratory frame (i.e. magnetic
    field in the NMR spectrometer).

Geometric Clustering Problem
  • The HSQC experiment data and RDCs are recorded.
  • One RDC for every backbone amide bond.
  • For each amide bond, a pair of (HN, N) chemical
    shifts is also measured.
  • We let R be the set of RDC values for the
    backbone amide bond vectors of our protein.
  • We take the set R to be the RDCs associated
    with amide perturbed chemical shifts the apo and
    holo form of protein.
  • Residues of set R are likely candidates for the
    interface region.
  • R subset clustered on the protein surface .

Geometric Clustering problem
  • Geometric optimization problem corresponds to
    identifying a set of
  • candidate NH bond vectors and their residues
    that map to, within
  • experimental error, a set of RDCs R that is
    a subset of R that are
  • clustered on the protein surface.

  • As proteins achieve their functions by
    interacting with other molecules, thus knowledge
    about the biological function of protein will be
    enhanced with structures of protein complexes.
  • NMR was found to be a unique technique in
    understanding these complex motifs. Nuclear
    Overhauser Effect (NOE), chemical shift
    perturbations and residual dipolar couplings
    (RDCs) are some of the major NMR methods to
    understand and resolve the protein-protein
  • When unassigned NMR data is used, computational
    and combinatorial approaches can be combined to
    further interpret the structures.
  • Thus NMR combined with computational algorithm
    holds a unique and useful approach in future for
    understanding the protein-protein complex