ROBINSONROBINSON Three Biological Systems: DNA, RNA, Membrane-binding Proteins Using EPR as a probe of the Structure-function relation Dynamics-function.

ROBINSON

Three Biological Systems:DNA, RNA, Membrane-binding

Proteins

Using EPR as a probe of the Structure-function relation Dynamics-function

relation

Graduate Students:

Tamara Okonogi

Robert Nielsen

Thomas E. Edwards

Faculty:

Snorri Sigurdsson

Michael Gelb

Kate Pratt

Post Docs:

Andy Ball

Ying Lin

Stephane Canaan

Supported by NSF and NIH

ROBINSON

Biological Applications of the Spin Label Method

Bending (Dynamics) of native DNApolymorphic nature of DNA’s motions

Response of the TAR (to binding proteins)Structural (and dynamic) response of

RNA

Membrane-Binding ProteinsRelation of active site to membrane

surface

Comments on EPR’s futureTime Domain, Low Field, High Field

ROBINSON

A Spin Labeled Base Pair

Replace a natural base pair with a spin labeled one.

Using phosphoramadite chemistry, construct DNAs of any length and sequence.

Make the duplex from xs complement.

ROBINSON

EPR 101

The slower moving the label the wider the spectral width.

Sorry, we have to look at squiggly lines.

ROBINSON

CWEPR Spectra for sl-DNAs

Two different isotopes of spin labels. For duplex DNAs of different lengths, with the spin label uniquely in the middle of each DNA.

ROBINSON

Flexible AT Sequences Inserted in 50mer Duplex DNA

Label at position 6

Distance of AT sequences from probe

ROBINSON

Methylphosphonates replace Phosphates

•Place a line of 10 MPs in a row (UNB)

•Place a Patch of 6 MPs together (AP)

Removes the negative charge locally (due to the phosphates).

MPs cause DNA to bend toward the patch.

Is DNA more flexible (bendable)?

MPs are a “phantom model” for protein binding

ROBINSON

Move the Neutral Patch Away From the Label

ROBINSON

Close Up of High Field Lines

ROBINSON

MPs Are More Flexible

ROBINSON

Does the DNA sequence determine flexibility?

•We examined many (40) different sequences.

•Measured the dynamics for each sequence

•All duplex DNAs were 50 base pairs long

•All duplex DNAs had the first 12 base pairs constant

•The probe was always at postion 6.

As a sequence is moved further from the duplex DNA its effect falls off.

ROBINSON

Sequences Of Duplex DNA

1. NT 0 --- --- 0.087 ± 0.002

2. AT4 8 12 19 0.090 ± 0.001

3. AT10 20 12 31 0.093 ± 0.001

4. AT15 30 12 41 0.097 ± 0.003

5. AT7A 15 12 26 0.093 ± 0.001

6. AT7A_s5 15 17 31 0.091 ± 0.001

7. AT7A_s12 15 24 38 0.088 ± 0.001

8. AT7A_s24 15 36 50 0.087 ± 0.002

9. AA7A or AAA5 15 12 26 0.083 ± 0.001

10.AA10 20 12 31 0.089 ± 0.001(0.084)

11.AA7A_s5 15 17 31 0.086 ± 0.001(0.084)

12.CG7C 15 12 26 0.085 ± 0.002

13.CG10 20 12 31 0.086 ± 0.001

14.CG7C_s5 15 17 31 0.086 ± 0.001

15.CC7C or CCC5 15 12 26 0.086 ± 0.002

16CC10 20 12 31 0.084 ± 0.001

17.CC7C_s5 15 17 31 0.087 ± 0.001

18.AC7A 15 12 26 0.088 ± 0.002

19.AG7A 15 12 26 0.089 ± 0.002

20.AAT5 15 12 26 0.089 ± 0.004

21.AAC5 15 12 26 0.088 ± 0.001

22.AAC5_s5 15 17 31 0.087 ± 0.001

23.AAG5 15 12 26 0.087 ± 0.002

24.AGG5 15 12 26 0.089 ± 0.001

25.AGG5_s5 15 17 31 0.087 ± 0.001

26.ACG5 15 12 26 0.089 ± 0.001

Number, Name L N1 N2

2

6 exp 26

exp

2

6 mod

rad2

ROBINSON

Sequences Of Duplex DNA cont’d

27.ACG5_s5 15 17 31 0.085 ± 0.001

28.ACT5 15 12 26 0.088 ± 0.001

29.ATC5 15 12 26 0.087 ± 0.002

30.CAG5 15 12 26 0.089 ± 0.002

31.CCG5 15 12 26 0.087 ± 0.001

32.CCG5_s5 15 17 31 0.082 ± 0.002

33.1/2CAP: TGTGACAT 8 12 19 0.089 ± 0.002

34.TATA: TATATAAA 8 12 19 0.093 ± 0.002(0.088)

35.G3C3-motif 6 12 17 0.084 ± 0.001(0.083)

36.G3C3-motif_s1 6 13 18 0.081 ± 0.001(0.077)

37.G3C3-motif_s8 6 20 25 0.076 ± 0.001(0.070)

38.G3C3-motif_s0_s10 G3C3-motif_s10

6 6

12 17 22 27

0.088 ± 0.001(0.083)

39. G3C3-motif/A5-tract G3C3 A5-tract G3C3 A5-tract

6 5 6 5

2 17 18 22 23 28 29 33

0.089 ± 0.001(0.086)

40. A5-tract/G3C3-motif A5-tract G3C3 A5-tract G3C3

5 6 5 6

12 16 17 22 23 27 28 33

0.087 ± 0.001(0.072)

Number, Name L N1 N2

2

6 exp 26

exp

2

6 mod

rad2

ROBINSON

Goodness of Fit

ROBINSON

Models for the DNAs flexing

Considered 3 different types of flexibility in A Nearest Neighbor picture (a di-nucleotide model)

• 3 parameters: pur-pur (same as pyr-pyr), pur-pyr, and pyr-pur are the three distinct steps

• 6 parameters: AT is different from GC and order doesn’t matter. (Hogan-Austin Model)

• 10 Parameter: All dinucleotide steps are unique (the two stiffest were so stiff we had to fix them)

Pur = A or GPyr = T or C

ROBINSON

The Goodness of Fit Using Different Models

ROBINSON

Flexibility: Force Constant Ratios for different numbers of 50-mer DNAs

ROBINSON

Conclusions about DNA dynamics

DNA (measured by EPR, fast time-scale) is three times stiffer than that measured by traditional methods:

Demonstrate polymorphic nature of duplex DNA and suggests the existence of slowly relaxing structures.

Certain sequences are inherently more flexible.

Eg: AT runs and charge neutral (MP) sequences.

Sequence dependent DNA flexibility does not discriminate between AT vs GC (regardless of order).

The Hogan-Austin hypothesis is wrong.

Sequence does discriminate between purines and pyrimidines.

The step from (5’) CG to a GC (3’) is most flexible (CpG step)

The step from (5’) CG to a GC (3’) is most flexible

The step from (5’) TA to a AT (3’) is next-most flexible

ROBINSON

TAR RNA

PNAS 1998, 95, 12379

TAR RNA and Replication of the HIV

ROBINSON

ODMTO N

NH

O

O

O

PN

OCN

HN CF3

O

RNA synthesis

RNA deprotection

OO N

NH

O

O

O

PO

OO

NH2

-RNA

RNA

OO N

NH

O

O

O

PO

OO -RNA

RNA

O

HNHN

NO

N

O

NH2

N

O

NCO

Cl OCCl3

O

Edwards, T. E., et. al. J. Am. Chem. Soc. 2001, 123, 1527-28

Preparation of Spin-Labeled RNA

ROBINSON

GCGAG

AGACCGG

GCUC

UCUGGCCC5'

3'

5'3'

C

40U23

U25

38

EPR Spectra of Spin-Labeled TAR RNAs

ROBINSON

• Interactions of metal ions with the TAR RNA

• Binding of Tat-derivatives to the TAR RNA

• Inhibition of the TAR RNA by small molecules

EPR Studies of TAR RNA

ROBINSON

High-Resolution Structures of TAR RNA

ROBINSON nativ

eCa2+

Na+

GCGAG

AGACCGG

GCUC

UCUGGCCC5'

3'

5'3'

C

40U23

U25

38

Edwards, T. E., et. al. Chem. Biol. 2002, 9(6), in press

EPR of TAR RNAs in the Presence of Cations

ROBINSON

EPR Spectra: “Dynamic Signature”

ROBINSON

• Interactions of metal ions with the TAR RNA

• Binding of Tat-derivatives to the TAR RNA

• Inhibition of the TAR RNA by small molecules

EPR Studies of TAR RNA

ROBINSON

Argininamide:H2N

NH2O

HN NH2

NH

Tat Derived Peptide (mutant): YKKKKRKKKKA

Tat Derived Peptide (wild type): YGRKKRRQRRR

Structural Requirements for Tat Binding

ROBINSON

High-Resolution Structures of TAR RNA

ROBINSON

Edwards, T. E., et. al. Chem. Biol. 2002, 9(6), in press

Dynamic Signatures for TAR RNA Binding

ROBINSON

• Interactions of metal ions with the TAR RNA

• Binding of Tat-derivatives to the TAR RNA

• Inhibition of the TAR RNA by small molecules

EPR Studies of TAR RNA

ROBINSON

Small Molecule Inhibitors of TAR

ROBINSON

Dynamic Signatures for TAR RNA Binding

ROBINSON

No calcium-specific change, as suggested by crystallography, was observed in solution by EPR

The wild-type Tat peptide causes a dramatic decrease in the motion of U23 and U38, implying that in addition to R52 other amino acids are important for specific binding

EPR can predict specific site binding

Taken together, our results provide evidence for a strong correlation between RNA-protein interactions and RNA “dynamic signature”

Conclusions

ROBINSON

Amino acid effect: green = strong pink = weak black = none

spin-labeled RNT 1p RNA-protein complex

AG U*

A

C

A

G

U*

G

C

U*U

no effect

no effect

effect

5'

RNT 1p RNA

RNT 1p protein

NMR: HSQC

ROBINSON

Membrane Binding Proteins

Bee venom phospholipase

Oriented on a membrane surface by

Site Directed Mutagenesis

EPR spin relaxant method

ROBINSON

Human Secretory Phospholipase sPLA2

A highly charged (+20 residues) lipase

ROBINSON

Spin Lattice Relaxation and Rotational Motion of the

Molecule How CW spectra change with viscosity

How Relaxation Rate R1 changes with viscosity

ROBINSON

Labeling sPLA2 with a Spin Probe

Use site directed mutagenesis techniques to prepare proteins with a single

properly placed cytsteine. General Reaction for adding relaxants

H3C S S CH2

N OO

OS CH2

N O

PLA2 C SHPLA2 C S

+

..

The protein should contain only one cysteine for labeling.

Protein labeled at only one site at a time per experiment.

ROBINSON

Relaxant Method: Nitroxide Spectra depend on concentration of

relaxants

Spin-Spin (T1 or R1 processes)

Spin-Lattice (T2 or R2 processes)

1 1

2 2

o

o

R R Rlxnt

R R Rlxnt

Rates are increased by the same amount due to additional relaxing agents (relaxants).

2 1 2 1 2

2 1 2

02 2 2 1 2

o o

o o o

o o

P R R R Rlxnt R Rlxnt

R R R Rlxnt

P P P R R Rlxnt

ROBINSON

CW-EPR Saturation Method

• Measure the Height

• Plot as a function of field or Incident Power

• Extract the P2 parameter..

ROBINSON

Obtaining Relaxation Information

• Time Domain (Saturation Recovery or Pulsed ELDOR) depends on R1, directly.

• CW method (progressive saturation or rollover”) depends on P2.

• Signal Height is a function of incident microwave power:

12

2

1

o

o

cPY

P

P

312 2

ROBINSON

Relaxant effects for sl-sPLA2and Salt Effects

Spectra for spin labeled sPLA2 as a function of ionic strength of NaCl

ROBINSON

sPLA2 CW Curves with Membrane

ROBINSON

Direct measurement of Spin-Spin Relaxation Rates

Bound to membrane (DTPM) vesicles

Bound to Mixed Micelles

ROBINSON

Effect of Membrane on Crox Concentration

Exposure factor as a function of distance from the membrane surface. Crox is z=-3 and the membrane is negatively charged.

1

1

membranemembrane

no membraneno membrane

CroxR

R Crox

ROBINSON

sPLA2 on Membrane

View from membrane

Yellow: Hydrophobic Residues

Blue: Charged (pos) residues

Orientation perpendicular to that predicted by M. Jain.

Anchored by hydrophobic residues. Charges not essential

ROBINSON

Salt EffectCrox salted off protein by addition of NaCl

ROBINSON

sPLA2 Conclusions

sPLA2 causes the vesicles to aggregate.

Explains much other data and misconceptions about the kinetics and processive nature of sPLA2 action.

sPLA2 was oriented on micelles (instead) using spin-spin relaxation rates alone.

Orientation different from that of other model.

Hydrophobic residues are the main points of contact.

Charges provide a general, non-specific attraction.

ROBINSON

Extra Thoughts: Model Spin Label All Four First Harmonic Signals

ROBINSON

Model Spin Label: All four second harmonic signals

ROBINSON

Model Spin Label:Hyperfine Interaction With Protons

and FID