deuteration: structural studies of larger proteinsstudies of larger proteins • problems with...

Deuteration: Structural Studies of Larger Proteins

• Problems with larger proteins• Impact of deuteration on

relaxation rates• Approaches to structure

determination• Practical aspects of producing

deuterated proteins

(See Gardner and Kay (1998) Ann. Rev. Biophys. and Biomol. Str. 27, 357-406 for a general review)

6.57.07.58.08.59.09.510.0

0

1

2

3

4

6.57.07.58.08.59.09.510.0

0

1

2

3

4

Tendamistat 8 kDa

Cdc42 21 kDa

6.57.07.58.08.59.09.510.0

108

110

112

114

116

118

120

122

124

126

128

6.57.07.58.08.59.09.510.0

108

110

112

114

116

118

120

122

124

126

128

HR1b9 kDa

ττττc = 6nsec

Cdc42/ACK26 kDa

ττττc=18nsec

Effects of IncreasingMolecular Size

•number of resonances- crowded spectra

•faster relaxation- broad lines- low intensity- overlap- long experiment time

Transverse relaxation times as a function of protein size

0

50

100

150

200

250

300

4 8 12 16 20 24 28

tran

sver

se re

laxa

tion

time

T2 [m

s]

correlation time [ns]

Cαααα(D)N(H)

Cαααα(H)

Hn

Hαααα

Hββββ

Cx CyNz ∆∆∆∆= 20-24 msecCx Cy ∆∆∆∆=20-27 msec

but T2 often less than 30 msec

Sensitivity α Πα Πα Πα Πnsin(ππππJ∆∆∆∆) ΠΠΠΠmcos(ππππJ∆∆∆∆) exp(-R2 Σ∆Σ∆Σ∆Σ∆) }}}}

Linewidth ∆ν∆ν∆ν∆ν1/2 = 1/ππππT2

789 ppm 789 ppm 789 ppm 789 ppm

ττττc 5nsMW 10kDa

10ns20kDa

15ns30 kDa

25ns50kDa

Linewidth vs. Correlation Time

coherence transfer relaxation

active passive

Relaxation Mechanisms

•Dipole-Dipole interactions (DD)•Chemical shift anisotropy

HX=13C,15N

X

HH

H

internal DD contributions

external DDcontributions

R1,2 (X) ~ D.I(I+1).(γγγγXγγγγY)2 . ΣΣΣΣ J(ωωωωi)dip

i

γγγγD/γγγγH = 1/6.5

dCD/dCH = 1/16

Impact of deuteration on relaxation rates

Cαααα N Hαααα HN

rela

xatio

n ra

te c

onst

ant/H

z

0

20

40

60

80

100

removed by sidechain deuteration

fixed

ττττc =12 nsec

Impact of deuteration on relaxation rates

T2

T1

ττττc = 18 nsec

2

4

6

8

10

12

14

16

0 5 10 15 20

overall correlation time [ns]

Rat

io

T2(CD) / T

2(CH)T

2(CD) / T

2(CH)T

2(CD) / T

2(CH)

T1(CD) / T

1(CH)

Correlation time (nsec)

Rat

io

T2(CD)/T2(CH)

T1(CD)/T1(CH)

Effects of deuteration on Cαααα relaxation times

Isolated C-H pair

J(0) dominate

J(ωωωωC-ωωωωD) effects

Approaches to the Structure Determination of Larger

Proteins

• For ττττc up to ~12 ns (20 kDa) -13C/15N-labelling

• For ττττc up to ~18 ns (35 kDa) -fractional deuteration and 13C/15N-labelling

• For ττττc above ~18 ns - selective protonation and 13C/15N-labelling

Triple Resonance NMR and Random Fractional Deuteration

• Backbone assignments• Side-chain assignments• Measurement of NOE contacts

Backbone Assignments

• Deuteration reduces relaxation• Maximum sensitivity with 100%

deuteration- Grzesiek et al., (1993) J. Am. Chem. Soc.

115, 4369-4370.- Yamazaki et al., (1994) J. Am. Chem. Soc.

116, 6464-6465.- Yamazaki et al., (1994) J. Am. Chem. Soc.

116, 11655-11666.

HNCA HN(CO)CA

N

H

Cα

C

O

H

N

DCβ

HβCγ

Hγ

N

H

Cα

C

O

H

N

DCβ

HβCγ

Hγ

12

3

4

123

4 5

6

Backbone Experiments with Deuterated Proteins

•Out and back

•CT for 13C evolution (1/Jcc)

•1H T1s:longer recycle delaypreserve water

•Deuterium decoupling

•Removal of residual CααααH

•Sensitivity enhancement (20 nsec limit?)Shan et al (1996) JACS 118 6570-79

y -yWALTZ16x

1H15N

13Cαααα

2H

t1/2

ττττd

Tc Tc-t1/2

2Tc = 26.6 ms = 1/Jccττττd = 1.7ms = 1/4JCH

•2H restored to z-axis (lock stability)•CααααH evolves for 1/2JCH - removed by 1H decoupling

Backbone Assignments

Nietlispach et al., (1996) J. Am. Chem. Soc. 118, 407-415.

Hββββ

Hββββ

CααααC

O

H

N

Cββββ

H/D

Cαααα

Cββββ H/D

H/D

H/D

CααααC

O

H

N

Cββββ

H/D

Cαααα

Cββββ H/D

H/D

H/D

HBCB/HACANNH HBCB/HACA(CO)NNH

Isotopomer Fraction Contributionfor Cββββ peaks

CββββH2-CααααH 12.5% 4%

CββββHD-CααααH 25% 16%

CββββH2-CααααD 12.5% 17%

CββββHD-CααααD 25% 63%

CββββD2-CααααH 12.5% 0%

CββββD2-CααααD 12.5% 0%

Isotopomers in 50% 2H Proteins

Effect of Deuteration on HBCB/HACA(CO)NHfor a protein with ττττc~18ns (30kDa)

0% 2H 50% 2H 75% 2H

7.58.08.59.0

20

30

40

50

60

70

ppm7.58.08.59.0

0% 2H 50% 2H

Effect of Deuteration on HBCB/HACANNH for a protein with

18 nsec correlation time

• Backbone assignments• Side-chain assignments• Measurement of NOE contacts

Triple Resonance NMR and Random Fractional Deuteration

Side-chain Assignments

• Deuteration reduces relaxation• Maximum sensitivity with 50%

deuteration– Nietlispach et al., (1996) J. Am. Chem.

Soc. 118, 407-415.

HCC(CO)NNH

N

H

Cαααα

C

O

H

N

HααααCββββ

HββββCγγγγ

Hγγγγ

1

2

3 4

5

1

2

ppm7.58.08.59.0 7.58.08.59.0

Effect of Deuteration on HCC(CO)NNH for a Protein with a

12 nsec Correlation Time

0% 50%

Side-chain Assignments

• Deuteration reduces relaxation• Deuteration removes both

protons

C

HC

HC D

C

HC

HC

D

HCCH-COSY HCCH-TOCSY

D

D

D

D

H

Effect of Deuteration on HCCH-TOCSY for a Protein with 18 nsec

Correlation Time

0.51.01.52.02.53.03.54.04.5

4.2

4.3

4.4

4.5

4.6

0.51.01.52.02.53.03.54.04.5

4.2

4.3

4.4

4.5

4.6

0.60.81.01.21.41.61.82.0

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

0.60.81.01.21.41.61.82.0

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

0%-2H

50%-2H

0%-2H

50%-2H

Valα α

α β α γ1

α γ2

Ile

δ δδ γ2δ γ1

δ β

Isotope Shifts

Isotope effects:

1∆∆∆∆ 13C ~ -0.25 ppm2∆∆∆∆ 13C ~ -0.1 ppm3∆∆∆∆ 13C ~ -0.07 ppm1∆∆∆∆ 15N ~ -0.3 ppm2∆∆∆∆ 15N ~ -0.05 to -0.1 ppm

e.g 13Cαααα in 100% D protein -0.5 ppm isotope shift

weak secondary structure dependency.

Isotopomers not resolved in most experiments

• Backbone assignments• Side-chain assignments• Measurement of NOE contacts

Triple Resonance NMR and Random Fractional Deuteration

Measurement of NOE Contacts

• Deuteration affects the measurement of HN - HN, HN - HC and HC - HC cross peaks in different ways

– Nietlispach et al., (1996) J. Am. Chem. Soc. 118, 407-415.

Distance Restraints and Fractional Deuteration

0% 2H 50% 2H 75% 2H

NH/NH

NH/sidechain

ττττc = 12 nsec

Measurement of NOE Contacts

• Similar contributions from different isotopomers lead to comparable chemical shifts

CHD

CH2

δ (ppm)

J-correlated NOESY

CHD

CH2

δ (ppm)

Structure Calculations

• Use of ARIA for structure determination

– Nilges (1995) J. Mol. Biol. 245, 645-660

• Global fold from selectively protonated data

• Use Talos to estimate backbone torsion angles from chemical shifts

Fractional Deuteration at 50%

•Improved sensitivity and linewidth

•No complications from different isotopomers

•Backbone and sidechain assignments possible

•1H/1H NOEs can be observed

•Applicable to proteins up to 30 kDa

Approaches to the Structure Determination of larger

Proteins

• For proteins of up to ττττc ~ 12 ns, use 13C/15N-labelling

• For proteins of up to ττττc ~ 18 ns, use fractional deuteration and 13C/15N-labelling

• For proteins ττττc ~ 18 ns and above, use selective protonation and 13C/15N-labelling

Approaches for Proteins Larger than 30 kDa

Complete deuterationmaximum sensitivity for

backbone experiments but limited NOE information (HN <-> HN)

Selective protonation of residues e.g AILV in deuterated background

Selective protonation of Methyl-groups(Gardner J. Am. Chem. Soc. 119, 7599)

Triple Resonance NMR and Selective Protonation

• Label proteins with ILV+FY• Methyl and aromatic 13C and 1H

nuclei relax more slowly– Allows measurement of NH-NH,

NH-methyl and methyl-methyl NOE contacts

• These residues are typically found in the protein core or interfaces

• Biosynthetic pathways allow straightforward labelling

• Can adjust the number of residue types labelled

Triple Resonance NMR and Selective Protonation

• Backbone assignments• Side-chain assignments• Measurement of NOE contacts

Backbone Assignments

• Deuteration reduces relaxation• Maximum sensitivity with 100%

deuteration• ILV - Cαααα mainly deuterated• Me-selective - Cαααα 100% deuterated

HNCA HN(CO)CA

N

H

Cαααα

C

O

H

N

DCββββ

HββββCγγγγ

Hγγγγ

N

H

Cαααα

C

O

H

N

DCββββ

HββββCγγγγ

Hγγγγ

12

3

4

123

4 5

6

Triple Resonance NMR and Selective Protonation

• Backbone assignments• Side-chain assignments• Measurement of NOE contacts

Side-chain Assignments

• Deuteration reduces relaxation• Maximum sensitivity with 100%

deuteration– Yamazaki et al., (1994) J. Am. Chem. Soc.

116, 11655-11666.– Farmer et al., (1995) J. Am. Chem. Soc.

117, 4187-4188.

N

H

Cαααα

C

O

H

N

DCββββ

DCγγγγ

D

1

2 3

4

1

CC(CO)NNH

N

H

Cαααα

C

O

H

N

DCββββ

DCγγγγ

D

12

34

5

6

HNCACB

Side-chain Assignments

(H)CC(CO)NNH

N

H

Cαααα

C

O

H

N

DCββββ

DCγγγγ

Hγγγγ

2

3 4

5

1

2

Gardner et al (1996) J. Biomol. NMR 8, 351-356.

Triple Resonance NMR and Selective Protonation

• Backbone assignments• Side-chain assignments• Measurement of NOE contacts

Measurement of NOE Contacts

• 3D 15N- and 13C-NOESY

• 3D Val/Ile (HM)CMCB(CMHM) NOESY

– Zwahlen et al., (1998) J. Am. Chem. Soc. 120, 4825-4831.

• 3D 13C/13C-NOESY– Zwahlen et al., (1998) J. Am. Chem.

Soc. 120, 7617-7625.

• 3D HQQF-NOESY

Gz

13C

1H

CT

∆∆ ∆ ∆φrec

1φ

ξ 2

φ3

BB

13C

1H

∆ ∆ ∆φrec

1φ

ξ 4

φ3

BB

φ2

φ4

φ5

QQF

3∆ − t 1/2

t 1/4t 1/4 ∆/2

∆t 1/4 + ∆t 1/4 +

CT − t 1/2t 1/2

G3 G3 G4 G4G2 G2G1

2 CT

DEPT-HQQC

Methyl-Selective Correlation Experiments

HQQF

(Kessler, 1989)

all 1H transverse (16 msec)

a Hyb 2H1xCyc 8H1xCyH2zH3z (5∆∆∆∆ evolution of JCH)

6∆∆∆∆ ≈ 1/Jcc = 24 msec

only 1 proton transverse between b and c

a

b

c

-0.50.00.51.01.5H (ppm)1

(a) HQQF

(b) HQQC

(c) DEPT-HQQC

(c) QQF-HSQC

Sensitivity Improvement in HQQF

0.00.20.40.60.81.01.21.41.6

10

2

4

6

8

0

2

4

6

8

C (

ppm

)13

H (ppm)1

I126γ2

I117γ2

I46γ2

A142 β

V85γ2

I4γ2V77

γ1

L165δ2

L112δ2

L165δ1

L53δ2

L112δ1

Resolution Improvement in HQQFCdc42 methyl region

Structure Calculations

• Use ARIA for structure determination

– Nilges (1995) J. Mol. Biol. 245, 645-660

• Restrain φφφφ and ϕϕϕϕ angles in early stages and slacken off later

Limited Restraints in Structure Determination

Ras:•Mixed αααα and ββββ structure•Reasonable size (21kDa)•Single domain

Numbers of Methyls and Aromatics:

11 Ile11 Leu15 Val5 Phe9 Tyr0 Trp

NHs only - 13Å RMSD

HN/NH2

Ile, Val, Leu

Aromatic

13C/15N ILVFY2.0Å RMSD101 HN-HN (ass-ass)390 HN-S/C (ass-amb)431 S/C-S/C (ass-ass)

13C/15N ILV 15N FY5.0Å RMSD101 HN-HN (ass-ass)390 HN-S/C (ass-amb)231 ILV-ILV (ass-ass)174 ILV-FY (ass-amb)26 FY-FY (amb-amb)

13C/15N ILV 15N FY1.7ÅNH NOEs to 5ÅAs above with 330 HN-HN NOEs

Structures from Limited Restraints

Ras NMR Structure0.6Å RMSD 101 HN-HN 1171 HN-S/C1862 SC/SC

Structures from Limited Restraints:Effect of Dipolar Couplings

Clore et al (1999) JACS 121 6513-14

BAF - 89 residues all ααααCVN - 101 residues 2 domains - all ββββ

BAF CVNHN-HN 106 101HN-Me 40 84HN-arom 18 18Me-Me 25 70Me-arom 51 53arom-arom 5 5Total 245 331

Hbonds 37 40

RDC 259 334

H-bonds + HN + Me ~4Å accuracyAdding aromatics - 1.37Å (BAF) 1.53Å (CVN)Adding RDC - 0.91Å (BAF) 1.10Å (CVN)

RDC: HN,N-C�,HN-C�,Cαααα-H,Cαααα-C�

Conclusions

• Selective Protonation– optimises the sensitivity of experiments

used to correlate side-chain and backbone resonances

– allows one to obtain limited NOE data and structural models

– use of orientational restraints should improve structures

Practical Aspects of Producing Deuterated Proteins

• Random fractional deuteration

LB/H2O LB/50% D2O

M9 in 50% D2O witheither [50% 2H, 100% 13C/15N]algal hydrolysate or[100% 2H/13C] glucose

OD600 = 0.4

Practical Aspects of Producing Deuterated Proteins

• Selective protonation• Use 2H/13C glucose, amino acids and:

CH3-CD2-CO-COO- [3,3-2H]-13C-2-ketobutyrate

CH3-CD2-CD(CD3)-CD(ND3+)-COO- isoleucine

Gardner and Kay (1997) J. Am. Chem. Soc. 119, 7599-7600.

LB/H2O LB/95-99% D2O

M9 in 95-99% D2O

[100% 2H/13C] glucose

[100% 3,3−3,3−3,3−3,3−2H2, 100% 13C] 2-ketobutyrate

[100% αααα/ββββ-2H, 100% 13C/15N] Val

OD600 = 0.4