Nucleic Acids

NMR SpectroscopyMarkus W. Germann

Departments of Chemistry and BiologyGeorgia State University

Based in part on Pascale Legault lectures

Structure Determination:

I) Assignment

II) Local Analysis•glycosidic torsion angle, sugar puckering,backbone conformationbase pairing

III) Global Analysis•sequential, inter strand/cross strand, dipolar coupling

Nucleic Acids have few protons…..•NOE accuracy

> account for spin diffusion•Backbone may be difficult to fully characterize

> especially α and ζ. •Dipolar couplings

123

45

6

Numbering

9

78

321

654

Numbering

N

NN

N

OH

N

N

O

NHH

H3C

N

N

N

N

NHH

NH

H

N

NN

N

OH

O

H

N

NN

N

O

NH

H

CH3

N

NN

N HH

N

NN

N

InosineBase: Hypoxanthine

Xanthosine

NebularineO6 Me Guanosine

2Amino Adenosine

7deaza Adenosine

N

N

N

N

NCH3CH3

N

NN

N NH

H6 Dimethyl aminopurinee 2 Aminopurine 5Me Cytosine

α Adenosine

N

N

N

N

NHH

OH

OHO

Alternate bases and modifications (small selection)

HETERO BASE PAIRS

NN

O

OH3C N

NN

N

NH

H

HN

N

NN

O N

H

H

H

NN

O

N H

H

H+

Hoogsteen

NN

O

OH3C

N

N

N

N

NHH

H

N

N

N

N

O

NH

H

HN

N

O

N H

H

Watson-Crick

N

N

O

OH3C

N

N

N

N

NHH

HN

N

N

N

O

NH

H

H

N

N

O

N HH

Reverse Watson-Crick

NN H

O

OH3C

NNH

O

O CH3N

N H

O

OCH3

NNH

O

O CH3

HOMO BASE PAIRS

NN

O

NHH

NN

O

N HH

H+

CC+

TT(I) TT(II)

N

N

N

N

N HH

N

N

N

N

NH H

AA(I)

N

N

N

N

O

N

H

H

H

N

N

N

N

O

N

H

H

H

GG(I)

N

N

N

N

N HH

N

N

N

N

NHH

AA(II) GG(II)

N N

N

NO

H

NN

N

NO

H

NH2

NH2

α

β

δ

γ

ε

ζ

χ

O5’

O3’

ν3ν0

ν1ν2

ν4O4’

nucle

otid

e un

it

α (n-1)O3'-P-O5'-C5'β P-O5'-C5'-C4'γ O5'-C5'-C4'-C3'δ C5'-C4'-C3'-O3'ε C4'-C3'-O3'-Pζ C3'-O3'-P-O5'(n+1)

Glycosidic torsion angle χO4'-C1'-N1-C2 (pyrimidines)O4'-C1'-N9-C4 (purines)

Torsion angles in nucleic acids

H1' 5-6H2' 2.3-2.9(A,G) 1.7-2.3(T,C)H2'' 2.4-3.1(A,G) 2.1-2.7(T,C)H3' 4.4-5.2H4' 3.8-4.3H5' 3.8-4.3H5'' 3.8-4.3

H1' 5-6H2' 4.4-5.0H3' 4.4-5.2H4' 3.8-4.3H5' 3.8-4.3H5'' 3.8-4.3

C1' 83-89C2' 35-38C3' 70-78C4' 82-86C5' 63-68

C1' 87-94C2' 70-78C3' 70-78C4' 82-86C5' 63-68

RNADNA

9R-Borano DNA•RNA

5’-d(A T G G T G C T C)(u a c c a c g a g)r-5’

Adenine GuanineH2 7.5-8 C2 152-156 - - C2 156H8 7.7-8.5 C8 137-142 H8 7.5-8.3 C8 131-138N6H 5-6/7-8 N6 82-84 N1H 12-13.6 N1 146-149- - - N2H 5-6/8-9 N2 72-76

C4 149-151 C4 152-154C5 119-121 C5 117-119C6 157-158 C6 161N1 214-216 N1 146-149N3 220-226 N3 167

N7 224-232 N7 228-238N9 166-172 N9 166-172

Thymidine Uridine CydidineH6 6.9-7.9 C6 137-142 H6 6.9-7.9 C6 137-142 H6 6.9-7.9 C6 136-144Me5 1.0-1.9 Me5 15-20 H5 5.0-6.0 C5 102-107 H5 5.0-6.0 C5 94-99N3H 13-14 N3 156 N3H 13-14 N3 156-162 - - N3 210- - - - - - - N4H 6.7-7/81-8.8 N4 94-98

C2 154 C2 154 C2 159C4 169 C4 169 C4 166-168C5 95-112 C5 102-107 C5 94-99N1 144 N1 142-146 N1 150-156

No Structure Required!

Often, depending on the question asked, a full structure determination is not required

DNA Hairpin

Does it form a duplex? Which base pairs are thermo labile? Which base pair is which… assignment? Is the loop structured? Structure

DNA Hairpin AT GC T

Thermal lability

DNA Hairpin

1D NOE

9090

O

N

HN

O

O

CH3

O

O

O

N

NH

O

O

CH3

O

O

C1

C1

α

β

C G C T T A A C G-5’

C G C T A A G C G-5’

C G T T A A G C G-5’

C G C T T A G C G-5’

control 5’-G C G A A T T C G C

5’-G C G A A T C G CC G C T A A G C G-5’

αTαTalphaT

5’-G C G A A T T G CαC

αCalphaC

5’-G C G A T T C G CαAalphaA

5’-G C A A T T C G CαGalphaG

5’-G C G A A T C G CαTalphaT2 αT

αG

αA

C G C T T A A G C G-5’

Do the duplexes form, is there base pairing? Does the unusual base pair form?

New DNA constructs

BASE PAIRING AND BACKBONE CONFORMATIONBASE PAIRING AND BACKBONE CONFORMATION

imino 1H NMR 31P NMR

T7T6G9/G1

G3

12.513.013.514.0 ppm

G3T6T7

G9/G1

T7T6 G3G9

G1

T7 T6G3G9

G1

G9 G3T6

T7

G1

control

3´-3´5´-5´

�-1.5�-1.0-0.51.0 0.5 0.0 ppm

alpha T

alpha C

alpha A

alpha G

C G C T T A A G C G-5’5’-G C G A A T T C G C

Local Parallel Stranded Environment is Necessary forLocal Parallel Stranded Environment is Necessary forStable Duplex FormationStable Duplex Formation

15% Native PAGE, 15 mM MgCl2

O

BPB

d-GCGAATTCGCCGCTTAAGCG-d

d-GCG CGC T

A

T

Ad-GCGAATTCGCCGCTTAAGCG-d

d-GCG CGC T

A

T

A

E: alphaT (115 µM)

A: dT11B: d(CCGG)2C: alphaT (0.5 µM)D: control (0.5 µM)

F: control (115 µM)

A

B

Solvent Suppression

The presence of an intense solvent resonance necessitates an impractical highdynamic range. 110 M vs <1mM (down to 5-10 uM)To overcome this problem several methods are currently applied:

1) Presaturation.2) Observing the FID when the water passes a null condition after a 180

degree pulse.3) Suppression of broad lined based on their T2 behavior.4) Selectively excitation, with and without gradients5a) Use of GRASP to select specific coherences thereby excluding the intense

solvent signal. In this case the solvent signal never reaches the ADC. Thisallows the observation of resonances that are buried under the solvent peak.

5b) Use of GRASP to selectively dephase the solvent resonance(WATERGATE)

P18

90

PRESAT

Presaturation field strength:20-40 Hz corresponds to a6-12ms 90deg pulse.

Pros: Easy to set upExcellent water suppression

Cons: Resonances under water signal!(T variation)Labile protons not visible(some GC pairs may be)

90180

d1 td

x

y

z

x

y

zSolvent

WEFT

Method relies on different T1 values forwater and solute.

It fails if the relaxation times are similar.Intensity of the solute resonances may vary.For a selective 180 degree pulse on thesolvent these problems are largely avoided.

d1 td

90x 180y

td

N times

Spin-lock along y.Resonances with short T2 (broadsignals) are suppressed. Notparticularly good for suppressingsolvent signals.

CPMG

d1 td

90x 180y

td

PRESAT with Spin Echo

Presaturation field strength:20-40 Hz corresponds to a6-12ms 90deg pulse.

Pros: Easy to set upExcellent water suppressionFlat baseline

Cons: Resonances under water signal!(T variation)Labile protons not visible(some GC pairs may be)

Selective Excitation

90Selective rf pulse on solvent resonance followedby a z gradient pulse to dephase the water signal.This could be followed by a mild presaturationfield. The selective rf pulse (1-2ms, dependingon width to be zeroed) is usually of the gausstype.Also, the selective rf p ulse z-gradient constructscould be repeated (WET).

Bo

My

Bz

M =0

SINGLE LINE ON RESONANCE

Bz

y

z

x

y

z

x

y

solvent

Jump and return9090

d1

x -x

tdPros: Easy to set up

Excellent water suppression(with proper setup as good as presat)Good for broad signals!

Cons: Non uniform excitationBaseline not flat

Other sequences: 1331 etc

0

21

-2-1

+1+1Gz

Δ Δ Δ Δ-x90 90

x x90 180

-x

Watergate

Pros: Excellent water suppressionUniform excitationBaseline flat

Cons: May loose broad resonances

p1G1 + p2G2 ...... = 0

Water

Structure Determination:

I) Assignment NOESY, COSY, HSQCTOCSY……

II) Local Analysis•glycosidic torsion angle (NOE, COSY)•sugar puckering (COSY, NOE, +)•backbone conformation (COSY, +)•base pairing (NOE, COSY)

III) Global Analysis•sequential (NOE, COSY)•inter strand/cross strand (NOE, COSY)•dipolar coupling (HSQC, HSQC)

Black unlabeled, Blue labeled DNA or RNA

2’

2’’

O

H1'

B

H4'

O H2''

H3' H2'

O H5'

H5''

Stereospecific Assignment

2’

2’’

How do we determine them?

a) Rules of Thumb (5’, Shugar and Remin)

b) Short mixing times NOESYCrosspeak H1’-H2’’ > H1’H2’

Structure Determination:

I) Assignment

II) Local Analysis•glycosidic torsion angle, sugar puckering,backbone conformationbase pairing

III) Global Analysis•sequential, inter strand/cross strand, dipolar coupling

Nucleic Acids have few protons…..•NOE accuracy

> account for spin diffusion•Backbone may be difficult to fully characterize

> especially α and ζ. •Dipolar couplings

α

β

δ

γ

ε

ζ

χ

O5’

O3’

ν3ν0

ν1ν2

ν4O4’

nucl

eotid

e un

it

Sugar puckeringThe five membered furanose ring is not planar. It can be puckeredin an envelope form (E) with 4 atoms in a plane or it can be in atwist form. The geometry is defined by two parameters: thepseudorotation phase angle (P) and the pucker amplitude (Φ).In general: RNA (A type double helix) C3' endo.DNA (B type double helix) C2' endo.

νi = Φm cos (P + 144 (j-2))

δ = ν3 + 125°

3’endo

2’endo

N (Northern)

S

(Southern)

5’

5’

2’endo

3’endo

2’endo sugar H1’, H2’, H2”, H3’ region

H2” H2’H1’ H3’

3’endo sugar H1’, H2’, H2”, H3’ region

H2” H2’H1’ H3’

2’endo sugar H1’, H2’, H2”, H3’ region

H2” H2’H1’ H3’

2’endo sugar H1’, H2’, H2” region

Sugar puckeringUsually (DNA) one observes equilibrium of the S and N forms sugarre-puckering. Unless one form greatly dominates the local analysisrequires quite a few parameters: PN , PS , ΦN , ΦS , fSSeveral methods for analysis exist, graphical and the more rigoroussimulation. In practice the desired outcome determines the effortto be made. Sums of the coupling constants are often easier toobtain.

fS = (Σ 1’ –9.8)/5.9

Σ 1’ = J 1’2’ + J 1’2’’

Σ 2’ = J 1’2’ + J 2’3’ + J 2’2’’

Σ 2’’ = J 1’2’’ + J 2’’3’ + J 2’2’’

Σ 3’ = J 2’3’ + J 2’’3’ + J 3’4’

If fs < 50% J1’ 2’ < J1’ 2’’If fs ca 0% J1’ 2’ very smallIf fs > 70% J1’ 2’ > J1’ 2’’

Sugar puckering

Sugar puckering can also be studied from: CSA-DD and DD-DD cross-correlation data. CT-NOESY experiments ( 01JMR262-153)

control alphaT alphaC alphaA alphaGNt Σ1´ fs Σ1´ fs Σ1´ fs Σ1´ fs Σ1´ fs

G1 15.2 0.92 15.3 0.93 14.6 0.81 14.3 0.76 14.9 0.86

C2 15.1 0.90 14.7 0.83 15.0 0.88 15.0 0.88 (15.3) (0.93)

G3 16.2 1.00 15.9 1.00 14.9 0.86 16.0 1.00 9.4 -

A4 16.2 1.00 15.3 0.93 15.3 0.93 10.7 - 14.5 0.80

A5 15.7 1.00 15.3 0.93 15.1 0.90 12.1 0.39 14.9 0.86

T6 15.1 0.90 15.3 0.93 14.7 0.83 14.6 0.81 14.8 0.84

T7 16.0 1.00 12.3 - (15.3) (0.93) 15.0 0.88 15.6 0.98

C8 15.1 0.90 12.9 0.53 9.5 - 15.0 0.88 14.4 0.82

G9 15.7 1.00 14.7 0.83 13.5 0.63 14.3 0.76 14.9 0.86

C10 (14) (0.7) (14) (0.7) (14) (0.7) (14) (0.7) (14) (0.7)

ΦS,N = 37°fS = 125-165PS = 0.44-0.55

ΦS,N = 37°fS = 130-155PS = 0.78-0.86

Pseurot calculations

Sugar puckeringHOW to obtain?

-COSY, E.COSY, low flip angle COSY

Some referencesSzyperski, T., et al . (1998) Measurement of Deoxyribose 3 JHH ScalarCouplings Reveals Protein-Binding Induced Changes in the Sugar Puckersof the DNA. JACS. 120, 821- 822.

Iwahara J, et al. (2001), An efficient NMR experiment for analyzing sugar-puckering in unlabeled DNA: J.Mag Res. 2001, 153, 262. H3’-H2” and H3’-H2’ couplings via constant time NOESY.

J. Boisbouvier, B. Brutscher, A. Pardi, D. Marion, and J.-P. Simorre, NMRdetermination of sugar-puckers in nucleic acids form CSA—dipolarcrosscorrelated relaxation, J. Am. Chem. Soc. 122, 6779–6780 (2000).

BioNMR in Drug Research 2003Editor(s): Oliver ZerbeMethods for the Measurement of Angle Restraints from Scalar, DipolarCouplings and from Cross-Correlated Relaxation: Application toBiomacromoleculesChapter Author: Christian Griesinger

Distance information determines the glycosidic torsion angle

How do we get distance information?o Nuclear Overhauser effect (< 6Å)

2.5Å

3.8Å

α

β

δ

γ

ε

ζ

χ

O5’

O3’

ν3ν0

ν1ν2

ν4O4’

nucle

otid

e un

it

α and ζ pose problemsDeterminants of 31P chem shift.

ε and ζ correlate. ζ = -317-1.23 ε

Resonance Assignmnent DNA/RNA (Homonuclear)

Resonance Assignmnent DNA/RNA (Homonuclear)

7.4

αC8

A5

ppm

5.45.65.86.06.2 ppm

7.6

8.0

T6

C10T7C2

G1G9G3

A48.2

7.8

7.2

NOESY Connectivity (e.g. α C Decamer)NOESY Connectivity (e.g. α C Decamer)

G1-H1’

G1-H8

7.4

αC8

A5

ppm

5.45.65.86.06.2 ppm

7.6

8.0

T6

C10T7C2

G1G9G3

A48.2

7.8

7.2

7.4

αC8

A5

ppm

5.45.65.86.06.2 ppm

7.6

8.0

T6

C10T7C2

G1G9G3

A48.2

7.8

7.2

7.4

αC8

A5

ppm

5.45.65.86.06.2 ppm

7.6

8.0

T6

C10T7C2

G1G9G3

A48.2

7.8

7.2

2'2''

2'2''

2'2''

G

αC

T3'-3'

5'-5'

H

H

H

5’- CATGCATG GTACGTAC – 5’

DNA Miniduplex

31P NMR

Two- and Three-dimensional 31P-driven NMR Procedures for completeassignment of backbone resonances in oligodeoxyribonucleotides. G.W. Kellogand B.I. Schweitzer J. Biomol. NMR 3, 577-595 (1993).

31P NMR

ppm

4.04.24.44.64.85.05.2 ppm

�-2.0

�-1.5

�-1.0

�-0.5

1.0

0.5

0.0

P7

P8

P3

P2

P6P4

P1

P5

P9

ppm

4.04.24.44.64.85.05.2 ppm

�-2.0

-�1.5

�-1.0

-�0.5

1.0

0.5

0.0

P6

P4

P8P2P9P1P7P5

P3

ppm

4.04.24.44.64.85.05.2 ppm

�-1.5

�-1.0

�-0.5

1.0

0.5

0.0

P2

P3

P5P6 P7

P1P4

P9P8

AlphaC

3’ 4’ 5’,5’’

;****************************************;mwgcorrpt, AMX version;X-H correlation. H-detected;Sklenar et al., 1986, FEBS, 208, 94-98;****************************************d12=20up2=p1*2

1 ze d11 dhi2 d113 d12 p2 ph0 d2 lo to 3 times l1 d3 (p3 ph2):d d0 (p1 ph1) (p3 ph1):d go=2 ph31 d11 wr #0 if #0 id0 ip2 zd lo to 3 times td1 do exit

ph0=0ph1=0ph2=0 0 2 2ph31=2 2 0 0

;>>>>>>>>>>>>>>>DELAYS;d0 = 3us;d2 = 50ms;d3 = 3us;d11= 30 msec

;>>>>>>>>>>>>>>>PULSES;p1 = 90 deg proton pulse hl1 = 1;p2 = 180 deg proton pulse hl1 = 1;p3 = 90 deg X pulse;>>>>>>>>>>>>>>>LOOP-COUNTER;l1 = loop counter for presaturaton;l1*d2 = relaxation delay (l1=40, d2=50ms >>2s);>>>>>>>>>>>>>>>COMMENTS;rd=pw = 0, nd0 = 2, in0 = 1/(2*SW);ns = 4*n, ds = 4, MC2= TPPI;-----------------------END of PROGRAM---------------

Backbone Experiments:

Harald Schwalbe, Wendelin Samstag, Joachim W. Engels, Wolfgang Bermel, and Christian Griesinger,"Determination of 3J(C,P) and 3J(H,P) Coupling Constants in Nucleotide Oligomers", J. Biomol. NMR 3, 479-486

Z. Wu, N. Tjandra, and A. Bax, Measurement of H3’-31P dipolar couplings in a DNA oligonucleotide byconstant-time NOESY difference spectroscopy, J. Biomol. NMR 19, 367–370 (2001).

G. M. Clore, E. C. Murphy, A. M. Gronenborn, and A. Bax, Determination of three-bond H3’-31P couplings innucleic acids and protein-nucleic acid complexes by quantitative J correlation spectroscopy, J. Mag. Reson.134, 164–167 (1998).

BioNMR in Drug Research 2003Editor(s): Oliver ZerbeMethods for the Measurement of Angle Restraints from Scalar, Dipolar Couplings and from Cross-CorrelatedRelaxation: Application to BiomacromoleculesChapter Author: Christian Griesinger:J-Resolved Constant Time Experiment for the Determination of the Phosphodiester Backbone Angles α and ζ.