Ensemble Optimization Method on SAXS EOM 2.0 tutorial Giancarlo - - PowerPoint PPT Presentation

Ensemble Optimization Method on SAXS

EOM 2.0 – tutorial Giancarlo Tria giancarlo.tria@embl-hamburg.de BioSAXS group @ EMBL Hamburg

EMBO Practical Course on Solution Scattering from Biological Macromolecular October 21st, 2012

Ensemble Optimization Method on SAXS – EOM 2.0

Kratky Plots to Detect Disorder

2

Unfolded Folded Multi-domains with flexible linkers Kratky plot establishes an approximate relationship between I(s) vs s for folded and unfolded proteins

Ensemble Optimization Method on SAXS – EOM 2.0

► Smooth Scattering profiles and featureless Kratky Plots ► Large Rg and Dmax ► Absence of correlation peaks in the p(r) function ► Low correlation densities in ab initio reconstructions ► Isolated domains in rigid body modelling ► Prediction of disorder using bioinformatics tools

http://www.idpbynmr.eu/home/science/research-tools.html

Indications (not Proofs!!!) of Flexibility

Ensemble Optimization Method on SAXS – EOM 2.0 SAXS curves

Analysis of the overall size descriptors (Rg, p(r), Kratky)

Modelling: ab initio (DAMMIN/DAMMIF) and Rigid body (BUNCH/CORAL) Analysis of the differences

Rigid Scenario Flexible Scenario

Detection of Flexibility: A Crucial Issue

Go for flexibility!

Ensemble Optimization Method on SAXS – EOM 2.0

Flexibility as mixture of different conformations

5

∑

=

k k k

s I v s I ) ( ) (

dr sr sr r p s I

D

∫

= sin ) ( 4 ) ( π

vk = volume fraction Ik(s) = scattering intensity from the k-th component For monodisperse systems the scattering is proportional to that of a single particle averaged

• ver all orientations

Ensemble Optimization Method on SAXS – EOM 2.0

Ensemble methods in SAXS

6

... Rg

1

Rg

2 Rg 3 Rg 4

Rg

5

...

ρ (Rg)

...

∑

=

I N s I

) ( 1 ) (

Genetic Algorithm Pool generation Crysol

The Ensemble Optimization Method (EOM)

Bernadó, Mylonas, Petoukhov, Blackledge, and Svergun. Structural characterization of flexible proteins using small-angle X-ray

• scattering. J Am Chem Soc 2007, 129:5656-64.

Ensemble Optimization Method on SAXS – EOM 2.0

Genetic Algorithm (optimized ensemble size)

7

Mutation Crossing Crossing Mutation Elitism Generation 1 Elitism Generation 2 Chromosome Chromosome

∑

=

=

N n n s

I N s I

1

) ( 1 ) (

Ensemble Optimization Method on SAXS – EOM 2.0

Modelling: Native vs. Random

8

ψ φ

Cα Cα

Kohn et al. PNAS, 2004, 101, 12491 Rg

Rg = R0·Nν R0 Persistence Length ν Solvent ‘quality’

Several experimental and theoretical studies establish ν ≈ 0.588 as an indication of the ‘random coil’ in chemically denatured (Urea or GuHCl) proteins. N

Theoretical distribution of the bond and dihedral angles for random chains Quasi Cα -Cα Ramachandran plot

• G. Kleywegt , Validation of protein models from Cα

coordinates alone, JMB, 1997, 273, 371-376

Bond angles vs. Dihedral angles

Ensemble Optimization Method on SAXS – EOM 2.0

• TAU protein isoform (124AA)

Unfolded protein …

9

> CYLSRKLMLDARENLKLLDRMNRLSPHSCL QDRKDFGLPQEMVEGDQLQKDQAFPVLYE MLQQSFNLFYTEHSSAAWDTTLLEQLCTGL QQQLDHLDTCRGQVMGEEDSELGNMDPIV TVKKYF sequence.seq curve.dat

Inputs for using EOM:

Ensemble Optimization Method on SAXS – EOM 2.0

10

…

<1min per repetition

Ensemble Optimization Method on SAXS – EOM 2.0

… unfolded protein: results

11

0.01 0.02 0.03 0.04 0.05 0.06 0.07 20 40 60 80 Series1 Series2 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 50 100 150 200 250 Series1 Series2 pool ensembles pool ensembles Rg [Å] Dmax [Å] Rg = 45.05 Rg = 32.96 Dmax = 140.38 Dmax = 101.02

Ensemble Optimization Method on SAXS – EOM 2.0

Missing loops (i.e. flat electron density map) …

12

MRIGMV……..GGVQSHVLQ…..VLRDAGHEVS…….PHVKLPDYVS

missing loop 30 AA

Kratky Plot

apoferritin

vs.

pool

Nter.pdb Cter.pdb seq.seq curve.dat

Ensemble Optimization Method on SAXS – EOM 2.0

... Missing loops: results

13

0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 21 23 25 27 Series1 Series2 Rg [Å] pool ensembles Rg = 24.46 Rg = 24.28

Ensemble Optimization Method on SAXS – EOM 2.0

??missing??

31 AA N-terminal tail

??missing??

14

pool

high resolution (MX) N-terminal pentamer domain

Flexible pentamer in solution …

(full length protein measured in two buffers, with low and high ionic strength respectively)

high resolution (MX) C-terminal monomer domain

??missing??

122 AA inter-domains linker

??missing??

Ensemble Optimization Method on SAXS – EOM 2.0

15

0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 20 40 60 80 100 Pool High ionic strength Low ionic strength 0.01 0.02 0.03 0.04 0.05 0.06 0.07 80 130 180 230 280 330 380 Pool High ionic strength Low ionic strength

Rg, Å Dmax, Å

Multi-curves fitting

pool

… Flexible pentamer in solution: results

(full length protein measured in two buffers with low and high ionic strength respectively)

Ensemble Optimization Method on SAXS – EOM 2.0

16

Case extra: dodecamer (P62, 2 domains) + tRNA

158 AA N-terminal tail high resolution (MX) N-terminal monomer domain (141 AA) 9 AA inter-domains linker high resolution (MX) C-terminal monomer domain (270 AA) 30 N single strand tRNA

subUnit contact residues range max distance in

Ensemble Optimization Method on SAXS – EOM 2.0

Number of chains: 1 000

Number of chains: 5 000

Number of chains: 10

Number of chains: 10 000

Number of chains: 64 790

Number of chains: 100

EOM Tests: Size of Pool

17

Ensemble Optimization Method on SAXS – EOM 2.0

Resolution of Subpopulations by EOM …

Generate a pool, select two subpopulations from it and calculate scattering curve for their union

Wide subpopulations

Narrow subpopulations

18

Ensemble Optimization Method on SAXS – EOM 2.0

Scattering curves

20 30 40 50 0.00 0.02 0.04 0.06 0.08 0.10 Rg, Å Density 0% < Rg < 5%, 95% < Rg < 100% 10% < Rg < 15%, 85% < Rg < 90% 5% < Rg < 10%, 90% < Rg < 95% 15% < Rg < 20%, 80% < Rg < 85% 20% < Rg < 25%, 75% < Rg < 80% 25% < Rg < 30%, 70% < Rg < 75% 30% < Rg < 35%, 65% < Rg < 70% 35% < Rg < 40%, 60% < Rg < 65% 40% < Rg < 45%, 55% < Rg < 60% 45% < Rg < 50%, 50% < Rg < 55% Pool

10-15% Well distinguishable ΔRg = 13.02 15-20% Undistinguishable ΔRg = 11.33 100 AA Genetic Algorithm

…resolution of Subpopulations by EOM

19

Ensemble Optimization Method on SAXS – EOM 2.0

Take home messages (EOM 2.0)

• EOM

allows

• ne

to quantitatively characterize the flexibility of a particle (what the conformations that the protein prefers in solution)

• Intrinsically unfolded protein (IDP) can be easily modelled

with EOM (no Size limitations)

• SAXS

in solution can be used as complementary technique to model flexible systems, disordered regions, etc..

20