Transcript N - Academia Sinica

CBMB2008
III. Structure determination: Nucleic Acids
by NMR spectroscopy
Iren Wang
王怡人
Institute of Biomedical Sciences
Academia Sinica
2008. May 8
Outline
I. Nucleic acids hold diverse structures and functions
a. In vitro SELEX
b. Diverse structures of nucleic acids
II. NMR Spectroscopy for Nucleic Acid Assignment
a. The building blocks of nucleic acids
b. Resonance assignment in nucleic acids
III. Some application cases for protein-nucleic acids complexes
IV. Others (Advanced developments in NMR Spectroscopy)
a. Residual Dipolar Coupling (RDC)
b. Transverse Relaxation-Optimized Spectroscopy (TROSY)
c. Paramagnetic spin labeling
I. Nucleic acids hold diverse structures and functions
Nucleic acids hold diverse functions
** the genetic information carriers
** tRNA: transporters of genetic information
mRNA: a copy of the information carried by a gene on the DNA
rRNA: a component of the ribosomes
snRNA (small nuclear RNA): important in a number of processes including RNA splicing and
maintenance of the telomeres, or chromosome ends
** targets for proteins and/or drugs interaction
Postulated stem-loop diagram of the 5’ untranslated region of
HIV-1HXB2 genomic RNA
Nucleic Acids Res. (2008) May, p.1-17
Biochemistry (2008) 10, p.3283-93.
In vitro SELEX (systematic evolution of ligands by exponential enrichment)
an excellent tool for finding nucleotide molecules that have a high affinity for a particular target
from a random pool under specific conditions.
Three processes:
Selection of ligand sequences, Partitioning of aptamers, amplification of bound aptamers
Anal Bioanal Chem (2007) 387, p.171-82.
Nucleic acids hold diverse structures
The possible conformations formed by poly-nucleotides in solution are flexible, “unstructured”
single strands, stacked helical single strands, hairpins, regular duplexes formed by
complementary strands, and a variety of aggregates between partially complementary strands,
which may contain bulges, dangling ends, or stacked single stranded ends.
DNA duplexes, triplex, multi-stranded G-quadruplex structures
RNA structural elements: helices, hairpins, bulges, junctions, pseudoknots
** non-helical conformations and tertiary structure are stabilized by:
metal ions, water-mediated H-bonds, and stacking interactions
** formation of a double-stranded helix is driven by cooperative attractive hydrogen bonding and
stacking interactions
** U-turn / reversed U-turn of RNA: sharp turns in hairpin loops
** the diversity of RNA structures compared to DNA is a result of non-helical secondary structure
** non Watson-Crick base pairs are important for RNA/RNA and RNA/protein recognition
Telomeric DNA quadruplex structures
The arrangement of
hydrogen bonds between
guanines in a G-tetrad
Current Opinion in Structural Biology (2003) 13, p.275-83.
Secondary structure of the T arm and pseudoknotted acceptor arm
of the tRNA-like structure of TYMV genomic RNA
Stem 2
Loop 1
Loop 1
Science (1998) 280, P.434-8.
RNA structure
By Michael Sattler
Proteins recognize unusual RNA structural elements
RNA structure: protein recognition
By stabilizing an adjacent interaction surface, bulges
can participate in complex protein binding sites.
Structure (2000) 8, R47-R54.
RNA bugles as architectural and recognition motifs
Bulges: unpaired stretches of nucleotides located within one strand of a nucleic acid duplex
-sizes: vary from a single unpaired residue up to several nucleotides
stacked
flipped-out
groove-binding
flap residues
Structure (2000) 8, R47-R54.
RNA bugles as architectural and recognition motifs
Bugles stabilization by metal ions
Bugles distortions
Structure (2000) 8, R47-R54.
Non-Watson–Crick base pairs employ non-standard H-bonds
bifurcated H-bonds
cis Watson-Crick G*A
water-mediated
C-H N/O H-bond
open, water-mediated
Watson-Crick G*A
Structure (2000) 8, R55-R65.
II. NMR Spectroscopy for Nucleic Acid Assignment
Nucleic acid bases, nucleosides, nucleotides
Nomenclature, structures, and atom
numbering for the sugars contained in
common Nucleotides.
Nomenclature, structures, and atom numbering for the bases contained
in common Nucleotides.
Labile protons
Torsion Angles in Nucleic Acids
By Michael Sattler
Sugar pucker, pseudorotation
A. Puckering of five-membered ring into envelope (E) and twist (T) forms.
B. Definition of sugar puckering modes
C. Pseudorotation cycle of the furanose ring in nucleosides.
(A)
(B)
(C)
NMR of Proteins and Nucleic acids (1986) by Kurt Wüthrich
Syn/anti conformations – the χ torsion angle
NMR of Proteins and Nucleic acids (1986) by Kurt Wüthrich
1D 1H NMR spectrum in Nucleic Acids (in D2O)
H2 H6 H8
H1’ H5
H2’ H3’ H4’ H5’ H5’’
CH3
1D 1H NMR spectrum in Nucleic Acids (in H2O)
imino
amino
aromatic
Flowcharts for resonance assignment in nucleic acids
A. NOE-based assignment in unlabeled nucleic acids
I (H2O)
Assignment of imino (and amino) resonances to establish base pairing
NOESY imino-imino, amino-imino
II (H2O)
Partial resonance assignment of non-exchangeable protons via NOE
connectivities to amino and/or imino protons
NOESY imino-H2/H6/H8/H5/H1’
III (D2O)
Identification of sugar proton spin systems
(mainly H1’/H2’/H2’’/H3’) (1H, 1H) COSY/TOCSY
Identification of aromatic spin systems
(Cytosine/Thymine H5/H6) (1H, 1H) COSY/TOCSY
Sequential resonance assignment
NOESY H6/H8-H1’, H6/H8-H2’H2’’
IV (D2O)
Assignment of 31P resonances and confirm/extend H3’,H4’,H5’,H5’’
assignments (1H, 31P) HETCOR/HETTOC
Progress in Nuclear Magnetic Resonance Spectroscopy (1998) 32, p.287-387.
Flowcharts for resonance assignment in nucleic acids
B. NOE-based assignment in labeled nucleic acids
I (H2O)
Exchangeable proton/nitrogen correlation
2D 15N-HMQC imino 1H optimized G N1H, U N3H
amino 1H optimized C N4H2, G N2H2, A N6H2
Exchangeable proton/nitrogen sequential assignment
3D 15N-NOESY-HMQC (imino 15N edited NOESY)
imino-imino, amino-imino
15
3D N-NOESY-HMQC (amino 15N edited NOESY)
amino-imino
II (H2O)
Partial resonance assignment of non-exchangeable proton
from NOE connectivities with amino and/or imino protons
3D 15N-NOESY-HMQC (imino 15N edited NOESY)
aromatic-imino
Progress in Nuclear Magnetic Resonance Spectroscopy (1998) 32, p.287-387.
Flowcharts for resonance assignment in nucleic acids
B. NOE-based assignment in labeled nucleic acids
(continued)
III (D2O)
Identification of sugar proton spin systems
3D HCCH-COSY H1’-H2’
3D HCCH-RELAY H1’-H2’/H3’
3D HCCH-TOCSY
Identification of sugar carbon spin systems
2D 13C-CT-HSQC/HMQC
3D HCCH-COSY H1’-C2’
3D HCCH-RELAY H1’-C2’/C3’
3D HCCH-TOCSY H1’-C2’/C3’/C4’/C5’
Identification of proton/carbon aromatic spin systems
2D 13C-CT-HSQC/HMQC H6-C6, H8-C8, H5-C5, H2-C2
2D/3D HCCH-COSY H6-H5, H6-C6/ C5, H5-C6/ C5
Sequential resonance assignment
3D 13C-NOESY-HMQC H6/H8-H1’, H6/H8-H2’H2’’
IV (D2O)
Assignment of 31P resonances e.g. (1H, 31P) HETCOR/HETTOC
Progress in Nuclear Magnetic Resonance Spectroscopy (1998) 32, p.287-387.
Flowcharts for resonance assignment in nucleic acids
C. Assignment via through-bond coherence transfer in labeled nucleic acids
I (H2O)
Exchangeable proton/nitrogen correlation
2D 15N-HMQC imino 1H optimized G N1H, U N3H
amino 1H optimized C N4H2, G N2H2, A N6H2
II (H2O)
Through-bond amino/imino to non-exchangeable base proton correlations
HNCCH/HCCNH
III (D2O)
1. Through-bond H2-H8 correlations {HCCH-TOCSY/(1H,13C) HMBC}
2. Through-bond base-sugar correlations
{HCN (base) with HCN (sugar), HCNCH,
HCNH, {HCN (sugar) with H8N9(H8)C8H8},
{HCN (sugar) with (Hb,Hb) HSQC},
{(H1’, C8/6) HSQC with (H8/6, C8/6) HSQC}
3. Through-bond sugar correlations
{HCCH-COSY/ HCCH-TOCSY}
4. Sequential resonance assignment via through-bond
sugar-phosphate backbone correlations (1H, 13C, 31P)
HCP/ PCH/ PCCH-TOCSY/ HPHCH
Progress in Nuclear Magnetic Resonance Spectroscopy (1998) 32, p.287-387.
NOE-based assignment in unlabeled nucleic acids
I. Assignment of imino (and amino) resonances in H2O
Imino Proton Assignments by 2D NOESY spectrum
1
10
5
15
20
8G
5'–CGACGATGACGTCATCGTCG-3'
3'-GCTGCTACTGCAGTAGCAGC-5'
10
5
1
20
5G
15
11G
H
Me
O
T
H
N
H
H
2G
N
N
N A
N
N
R
H
O
R
H
N
b
imino proton
17G
7T
15T
H
H
O
N
N
R
G N
H
H
N
12T
H
N
H
C
a
18T
N
N
N
H
O
R
imino proton
H
J. Chin. Chem. Soc. (1999) 46, p.699-708.
II. NOESY imino-H2/H6/H8/H5/H1’ in H2O
Imino Proton to Amino to Aromatic Protons
H
Me
d. Aromatic to H2’/H2”
T
H6
N
N
b. Imino to amino/aromatic
H
O
N
N
N A
H
N
1
R
H2
O
R
H8
N
6
c. Amino to aromatic
H8
H
O
N
2
N
R
H
5
3
G N
H
7
H5
N
4
N
H6
C
N
N
a. Imino to imino
N
H
H
O
R
II. NOESY imino-H2/H6/H8/H5/H1’ in H2O
b. Imino to amino/aromatic
Imino Proton to Amino and Aromatic
H
Me
T
6H
H5
H
O
N
H
N
N
N A
N
H8
N
N
H2
O
R
R
H
H8
O
N
N
R
G N
H
H
H5
N
N
H6
C
N
N
N
H
O
R
H
H2
J. Chin. Chem. Soc. (1999) 46, p.699-708.
III. Identification of aromatic spin systems in D2O
Only intra-strand aromatic to aromatic connectivities
J. Chin. Chem. Soc. (1999) 46, p.699-708.
III. Sequential resonance assignment in D2O
NOESY H6/H8-H1’, H6/H8-H2’H2’’
Cytosine: CH5-CH6
JMB (1983) 171, p.319-36.
Duplex-hairpin 5'–CGCGTATACGCG-3'
Nucleic Acids Res. (1985) 13, p.3755-72.
III. Sequential resonance assignment in D2O
NOESY H6/H8-H1’
Only intra-residue cross peaks were marked.
a-f. are the six big CH5-CH6 cross peaks.
J. Chin. Chem. Soc. (1999) 46, p.699-708.
III. Sequential resonance assignment in D2O
NOESY H6/H8-H2’H2’’
Only intra-residue cross peaks were marked.
J. Chin. Chem. Soc. (1999) 46, p.699-708.
1H-31P
Correlation Spectrum
2
3
4
5
6
7
8
11 10
9
8
7
6
5
4
9
10 11
5’ G-p-C-p-G-p-A-p-T-p-A-p-G-p-A-p-G-p-C-p-G
G-p-C-p-G-p-A-p-G-p-A-p-T-p-A-p-G-p-C-p-G 5’
(n-1) H3'- (n) P
b)
5T
8A
1G
9G
11p
10p
5'
7G
3G
2C
2p, 5p, 9p
2
(n) P - (n) H4'
6A
7p
3p
6p
3
H2"
6A
9G 2C
4A
10C
5T
11G
10C
H4'
H2'
O
H3'
(t)
31P
P
O
O

H2"
H5" 
Phosphate buffer
7G
3G
H1'
O
O
H5'
8p
4p
Base
8A
4A
H4'
H3'
O
Base
H2'
H1'
3'
JACS (1992) 114, 3114-5.
NOE-based and via through-bond coherence transfer
assignment in labeled nucleic acids
RNA synthesis by in vitro transcription
Transcription
dT
D
dC
N
A dG
dA
rA
R
N
rC A
RNA samples at natural isotopic abundance and enriched
in 15N and 13C can be prepared with T7 RNA polymerase.
rG
rU
T7 polymerase
DNA template
Transcription starting
3’ ATTATGCTGAGTGATATCCTTATACTATGTAAACTAGTCATATAGG 5’
5’ TAATACGACTCACTATAG 3’
Top strand
RNA synthesized
5’ GGAUUAUGAUACAUUUGAUCAGUAUAUCC 3’
Heteronuclear Chemical Shifts in Nucleotides
Current Protocols in Nucleic Acid Chemistry (2000) 7.7.1-7.7.30
2D 15N–1H HMQC spectra of RNA imino
resonances at different conditions
PNAS (1997) 94, p.2139-44.
The 1H-13C HSQC spectra of labeled nucleic acids
(A) H6/H8-C6/C8, (B) H1’-C1’, (C) H2’/H2’’-C2’, and (D) H3’-C3’
Intraresidue correlation
via through-bond coherence transfer NMR experiments
Adenine
H2-H8 or H5-H6
correlation
H1’,H2’, H3’ H4’H5’H5”
correlations, HCCH-TOCSY
H8-H1’ correlation, HCN
Nucleotide spin system
Correlation in the base-sugar: HCN 3D spectra
2D and 3D TROSY-HCN for obtaining ribose base and intra-base
correlations in the nucleotides of DNA and RNA.
Dotted arrows indicate the intra-base transfers and solid arrows the
ribose-base transfers.
H6/H8
H1’
JACS. 2001, 123, 658-64.
3D HCCH-TOCSY
H1’
Progress in Nuclear Magnetic Resonance Spectroscopy (1998) 32, p.287-387
Interresidue correlation through bond (HCP)
2 spin systems can be linked
Adenine
H3’-C3’-P(n-1)
Residue n
Guanine
H5’,H5’’(n-1)-C5’(n-1)-P(n-1)
OH
Progress in Nuclear Magnetic Resonance Spectroscopy (1998) 32, p. 287-387
Direct observation of H-bonds in nucleic acid base pairs
by inter-nucleotide 2JNN couplings
the connectivity between 1H3(U329)
and the 1H2-13C2 (A32)
JNN HNN-COSY
1H3-15N3(U)
to
JHN HSQC
3D13C-NOESY
15N1(A)
1H2
(A) to
15N1(A)
and
15N3(A)
JACS. (1998) 120, 8293-7.
Structural Determination of Nucleic Acids by NMR
1)
Similar to those used in protein
2)
First build a nucleic acid sequence template
3)
Input H-bonded constraints
4)
Input all exchangeable and non-exchangeable distance constraints and/or dihedral
constraints
5)
Use Distance Geometry calculation to get some initial structures
6)
Use Molecular Dynamics method to refine the structures
3’
5’
Distance constraints (NOEs)
Dihedral angles constraints (J-coupling)
Distance geometry calculator
(XPLOR-NIH)
3’
5’
III. Some application cases for protein-nucleic acids
complexes
NMR spectroscopy as a tool for secondary structure
determination of large RNAs.
Annu. Rev. Biophys. Biomol. Struct. (2006) 35,p.319-42.
The structure of HCV IRES domain II
Dependence of RDC values on the orientation of the
interdipolar vector (C-H) and the alignment tensor
Refinement of the HCV IRES domain II structures
calculated by the use of different sets of RDCs
Rmsd = 7.48Å
Rmsd = 5.79Å
Rmsd = 2.18Å
Annu. Rev. Biophys. Biomol. Struct. (2006) 35,p.319-42.
The common DNA recognition motifs
Helix-turn-helix (HTH) domain
Zinc-finger (ZF) domain
By Dr. Song Tan
The common DNA recognition motifs
Winged-helix (WH) domain
S204
Chemical shift change plot based on NMR titration data
NCa3
N196
a1
Recognition helix
G213
E172
Wing
a2
E172
N196
S204
G213
The common RNA recognition motifs
RNP/RRM domain RNP: ribonucleoprotein
RRM: RNA-recognition motif
Structure (1997) 5, p.559-70.
The common RNA recognition motifs
The KH domains
KH: K-homology motif
Structure (1999) 7, p.191-203.
Structure of the HIV-1 Nucleocapsid protein with SL3
C-RNA recognition element
Science (1998) 279, p. 384-8.
Structure of the HIV-1 Nucleocapsid protein with SL3
C-RNA recognition element
Science (1998) 279, p. 384-8.
The best fit superposition and space-filling representation
of the SL3 RNA in the NC-SL3 complex
Science (1998) 279, p. 384-8.
Recognition of the mRNA AU-rich element by the
zinc finger domain of TIS11d
(a) The ensemble of best 20 structures superposed on backbone heavy atoms in ordered regions of the
protein and RNA. (b) Ribbon representation of a single structure with the addition of green side chains for
the zinc-coordinating ligands. (c) Backbone superposition of the structure ensembles of fingers 1 and 2.
Finger 1 (Arg153–Phe180) is dark blue (backbone), green (zinc coordinating side chains) and red
(intercalating aromatic rings); the bound RNA (U6, A7, U8, U9) is orange. The corresponding colors for
finger 2 are light blue, yellow, pink and yellow.
NSMB (2004) 11, p. 257-64.
Comparison between nucleolin RBD12-sNRE complex
and the other RBD-RNA complexes
RBD: RNA biding domain
EMBO J (2000) 19, p.6870-81.
IV. Advanced developments in NMR Spectroscopy
New Techniques in NMR Spectroscopy
(1). Residual Dipolar Coupling (RDC)
NOE, dihedral angle and H-bond are short-range
restraints and have limitations for some structure
determination, like extended structures or
multiple-domain structure. RDC is a novel
restraint and provides global structure information.
or
(2). Transverse Relaxation-Optimized Spectroscopy (TROSY)
TROSY, which was developed by K. Wüthrich, can select one
fourth of the signals that relax more slowly than the others.
The utilization of TROSY techniques push the size limit of NMR
spectroscopy to 30~50 kDa.
(3). Other Applications
Paramagnetic Spin Labeling.
Residual Dipolar Coupling (RDC)
Residual dipolar couplings arise
from dipole-dipole interactions
between nuclei. In aqueous
solution, the isotropic
orientation of the molecules
average out the dipolar
couplings. However, in oriented
media, the molecular tumbled
anisotropically. The order of 104 to 10-3 of anisotropy tuned the
dipolar coupling constant to be
a residual value of few Hz,
which are well detectable by
NMR spectroscopy.
Values of static dipolar coupling constant
of two-spin systems in protein backbone
Residual Dipolar Coupling (RDC)
The residual dipolar coupling between two spins A and B are given by :
<DAB> = - C(Bo) [ a(3cos2 -1) + 3/2 r(sin2 cos2) ]
where
C(Bo) = S(Bo2/15kT)[AB h/(4p2rAB3).
A and B are gyromagnetic ratios of A and B.
rAB is the distance between A and B.
So, Bo
, DAB
S (order parameter) , DAB
Alignment Media
• Phages (Pf1, fd, TMV) (Zweckstetter, JBNMR)
• Bicelles (Sanders & Schwonek, Biochemistry, 1992;
Ottiger&Bax, JBNMR 1998)
• Polyacrylamide gels (Tycko,JBNMR; Grzesiek
JBNMR; Chou, JBNMR)
• Paramagnetic tagging (Opella , Griesinger,
Byrd)
• CPBr/hexanol (Barrientos, J. Mag. Res, ~2000)
• C12E5/hexanol (Ruckert&Otting, JACS 2000)
• Cellulose crystallites (Matthews, JACS, ~2000)
Alignment of Molecules in Anisotropic Solutions
The most-used media for RDC measurement are :
(a). Phospholipid bicelles and (b). Filamentous phage
HSQC for HN RDC values
Isotropic solution
+ 5.3 mg/ml Pf1
JNH
15N
JNH + DNH
Chemical Shift (ppm)
15N-IPAP
1H
Chemical Shift (ppm)
3D HNCO for C’N RDC values
Isotropic solution
+ 5.3 mg/ml Pf1
JC’N + DC’N
13C
Chemical Shift (ppm)
JC’N
1H
Chemical Shift (ppm)
Structure Refinement with RDC Restraints
<DAB>(,) = Da [ a(3cos2 -1) + 3/2 Dr(sin2 cos2) ]
X
Z
Y
Transverse Relaxation-Optimized Spectroscopy (TROSY)
(a). None-decoupled HSQC
(b). Decoupled HSQC
(c). TROSY-HSQC
15N
1H
1. Main relaxation source for 1H and 15N: dipole-dipole (DD) coupling and,
at high magnetic fields, chemical shift anisotropy (CSA).
2. Different relaxation rates (line width) for each of the four components
of 15N-1H correlation.
3. The narrowest peak (the blue peak) is due to the constructive
canceling of transverse relaxation caused by chemical shift
anisotropy (CSA) and by dipole-dipole coupling at high magnetic field.
4. TROSY selectively detect only the narrowest component (1 out of 4).
Interference between DD and CSA Relaxation
DD + CSA
(large)
(large)
DD – CSA
(large)
DD
(large)
DD
(large)
(A) At High Magnetic Field
(TROSY line-narrowing
effect)
(large)
+ CSA
(small)
– CSA
(B) At Low Magnetic Field
(almost no TROSY
line-narrowing effect)
(small)
•DD relaxation is field-independent. However, CSA relaxation  B02,
therefore at high magnetic fields, CSA relaxation can be comparable to
DD relaxation, and the interference effect on relaxation can be observed.
TROSY Effect is Field Dependent and Motion Dependent
800 (kDa)
Linewidth
150 kDa
50 kDa
Magnetic field strength
•Optimal field strength: 1 GHz for amide NH;
600 MHz for CH in aromatic moieties (500-800 MHz applicable).
Deuteration- NMR structural study of larger proteins
Deuteration is also an important techniques for NMR study of larger proteins (> 20kDa).
It is achieved by raising the E. coli. in D2O medium (NT$ 10,000 / 1L D2O).
Because of the significantly lower gyromagnetic ratio of 2H compared to 1H ([2H] /
[1H] = 0.15), replacement of protons with deuterons removes contributions to proton
linewidths from proton-proton dipolar relaxation and 1H-1H scalar couplings.
The effect of deuteration is similar with that of TROSY and both techniques are
frequently used for NMR study of larger proteins.
The Sensitivity and Resolution Gain by
TROSY and Deuteration
2H,15N-Gyrase-45
(45 kDa), 750 MHz
Current Opinion in Structural Biology (1999) 9, p,594-601.
Paramagnetic Relaxation Enhancement (PRE)
Paramagnetic spin labeling
MTSL