Upload
boris-fackovec
View
111
Download
2
Tags:
Embed Size (px)
Citation preview
Intramolecular Interactions in Globular Proteins
Boris Fa£kovec
Advisor: RNDr. Ji°í Vondrá²ek, CSc.
Katedra fyzikální a makromolekulové chemie
P°írodov¥decká fakulta UK
21st December 2011
Boris Fa£kovec ([email protected]) Protein modeling 21st December 2011 1 / 17
Introduction - physical studies of proteins
Sequence-structure-stability-function relationships in globular proteins -central to biochemistry
1931 �rst theory of protein denaturation (Wu)
1958 statistical thermodynamics of polymers (Zimm)
1958 �rst resolved protein X-ray structure (Kendrew)
1961 An�nsen's experiments
1977 �rst protein simulation (Karplus)
1983 Go model
80's-90's intensive development of force �elds (Kollman, Jorgensen)
1985 knowledge-based force �elds (Jernigan, Miyazawa)
90's - extensive calorimetric (DSC) studies (Privalov, Makhatadze)
Denatured state investigations (Shortle)
Boris Fa£kovec ([email protected]) Protein modeling 21st December 2011 2 / 17
Introduction - recent development
1993 sidechain atlas (Thornton)
90's lattice simulations (Shakhnovich)
1995 energy landscape perspective (Bryngelson, Onuchic)
1995 First force �eld decomposition of interactions (Lazaridis)
2000 Variational theory, spin glasses, principle of minimal frustration(Wolynes)
2005 stabilization of rubredoxin by strong dispersion interactions in itscore (Vondrasek)
2008 identifying stabilizing residues by IEM calculations(Biedermannova)
2010 IEM development - fragmentation and QM calculations ofprotein molecules (Berka)
2010 FF calculations surprisingly good agreement with benchmark QM(Kolar)
Boris Fa£kovec ([email protected]) Protein modeling 21st December 2011 3 / 17
Introduction - Interaction energy matrix (IEM)
Fragmentation of a protein native structureQM calculationsClassi�cation of fragments - backbone BB, sidechain charged CH,polar PO, non-polar NPPair additivity!
Figure: Interaction energy matrix of backbone-backbone interactions for a shortpeptide
`interaction' = inter-residual non-covalent interaction in single structureBoris Fa£kovec ([email protected]) Protein modeling 21st December 2011 4 / 17
Introduction - Types of intramolecular interactions
Charged ion-ionI high values of IEs in IEMs not in correspondence with real stabilization
e�ectI high compensation of attractive and repulsive interactionsI high compensation of interactions in native and unfolded stateI exceptionally high enthalpic-entropic compensation
Charged multipoles
Backbone
van der Waals, stackingI short ranged - small compensation inI always attractive - small compensationI probably undervalued in IEMsI hydrophobic residues burial - folding driving force
Boris Fa£kovec ([email protected]) Protein modeling 21st December 2011 5 / 17
Introduction - Objectives
Characterization of magnitudes and distributions of inter-residualnon-covalent interaction energies
Development of uni�ed in silico treatment - solution to problems withcharged residues
Decomposition of stabilizing energy
Boris Fa£kovec ([email protected]) Protein modeling 21st December 2011 6 / 17
Methods
Structure set selection - X-ray resolution <2 Å, single-stranded, noligands → 1358 structures
Optimization of hydrogen atoms - GROMACS, OPLS FF
Fragmentation and IEM calculation → 10 IEMs
Terminal backbones not considered,
HIS double protonated → charged residue
Only interactions with IE<-0.05 kcal/mol or IE>0.05 kcal/mol weresampled
Subsets - size and secondary structure content
Boris Fa£kovec ([email protected]) Protein modeling 21st December 2011 7 / 17
Results - Residue interaction energy distributions
0
0.2
0.4
0.6
0.8
1
-25 -20 -15 -10 -5 0 5 10
BB
0
0.2
0.4
0.6
0.8
1
-80 -60 -40 -20 0 20 40 60
BBCH
0
0.2
0.4
0.6
0.8
1
-25 -20 -15 -10 -5 0 5 10 15 20
BBNP
0
0.2
0.4
0.6
0.8
1
-25 -20 -15 -10 -5 0 5 10
BBPO
0
0.2
0.4
0.6
0.8
1
-250 -200 -150 -100 -50 0 50 100 150
CHCH
0
0.2
0.4
0.6
0.8
1
-100 -80 -60 -40 -20 0 20 40 60
CHNP
0
0.2
0.4
0.6
0.8
1
-50 -40 -30 -20 -10 0 10 20 30
CHPO
0
0.2
0.4
0.6
0.8
1
-20 -15 -10 -5 0 5
NPNP
0
0.2
0.4
0.6
0.8
1
-10 -8 -6 -4 -2 0 2 4
PONP
0
0.2
0.4
0.6
0.8
1
-6 -5 -4 -3 -2 -1 0 1 2 3
POPO
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0.55
-15 -14 -13 -12 -11 -10
BB
Figure: Distribution of RIE for all types of interactions. Various curves representsecondary structure particular classes.
Boris Fa£kovec ([email protected]) Protein modeling 21st December 2011 8 / 17
Results - Domain size in globular proteins
E =
{ED(1− kN
13 ) N ≤ ND
E = ED N ≥ ND
(1)
-4
-3.5
-3
-2.5
-2
-1.5
-1
50 100 150 200 250 300
avera
ge R
IE / [kcal/m
ol]
protein chain length
NPNP calculated
model
Figure: Average RIE - size dependence of NPNP interactions}HCIE of BB-BBinteractions.
Boris Fa£kovec ([email protected]) Protein modeling 21st December 2011 9 / 17
Results - Interaction energy distributions
number of contacts increases immensely with decreasing IE de�nition,diverges to ∞ at IE=0Cumulative distribution of contributions to sum of IEs (DCIE) and itsderivative (HCIE)
0.2
0.4
0.6
0.8
1
1.2
1.4
-4 -3 -2 -1 0 1 2
DC
IE
IE / [kcal/mol]
Figure: Cumulative distribution curve for IE of SER-TYR pair. BIE is the value ofthe interaction energy where the curve intersects 1 for the �rst time (-0.32kcal/mol), BIE0.5 is the value where it intersects 0.5 (-1.58 kcal/mol).
Boris Fa£kovec ([email protected]) Protein modeling 21st December 2011 10 / 17
Results - Interaction energy distributions
Random energy model:
HCIE = IE
(a0e
−(IE
σ0
)2+
t∑i=1
aie−(IE−IEi
σi
)2)(2)
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
-7 -6 -5 -4 -3 -2 -1 0
HC
IE
IE / [kcal/mol]
Figure: HCIE of BB-BB interactions. Red line represents calculated data, greenline represents �t using 8 Gaussians.
Boris Fa£kovec ([email protected]) Protein modeling 21st December 2011 11 / 17
Results - Optimum de�nition of interresidual contact basedon interaction energy matrix calculations
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
-7 -6 -5 -4 -3 -2 -1 0
BB-BB
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
-30 -25 -20 -15 -10 -5 0
BB-CH
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
-8 -7 -6 -5 -4 -3 -2 -1 0
BB-PO
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
-5 -4 -3 -2 -1 0
BB-NP
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
-140 -120 -100 -80 -60 -40 -20 0
CH-CH
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
-25 -20 -15 -10 -5 0
CH-PO
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
-10 -8 -6 -4 -2 0
CH-NP
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
-4 -3.5 -3 -2.5 -2 -1.5 -1 -0.5 0
PO-PO
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
-3 -2.5 -2 -1.5 -1 -0.5 0
PONP
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
-2.5 -2 -1.5 -1 -0.5 0
NPNP
Figure: Contact de�nitions from HCIE curves for each type of interaction
Boris Fa£kovec ([email protected]) Protein modeling 21st December 2011 12 / 17
Results - Optimum contact de�nitions and their protperties
Table: compens = ratio of the sum of all negative IEs to the sum of all IEs.Columns 3�6 show the order contributions of a particular type of interaction to aparticular fragment type. x(BIEx) is the ratio of the energy content of theproductive and all the interactions.
IE type CD BBCO CHCO POCO NPCO BIE compens x(BIEx)
BBBB -1.6 1.08 -0.28 1.14 0.72BBCH -10 0.09 0.32 -3.5 4.04 0.39BBPO -3 0.1 0.37 -0.4 1.26 0.28BBNP -1.8 0.07 0.15 -0.1 1.05 0.08CHCH -82 0.16 -69 11.7 0.81CHPO -12 0.1 0.11 -4 3.03 0.47CHNP -3 0.16 0.1 -1.37 2.5 0.44POPO -0.8 0.33 -0.5 1.29 0.9PONP -0.4 1.44 0.82 -0.1 1.06 0.82NPNP -0.3 1.48 -0.19 1.09 0.96
Boris Fa£kovec ([email protected]) Protein modeling 21st December 2011 13 / 17
Conclusion
RIE characterized by magnitudes and distributions
No correlation between sidechain IEs and secondary structure content
Typical one-domain protein length - 110 residues
Random energy model can be very successfully applied for IE statistics
Compensation of positive and negative interactions characterized
Contact de�nitions for each type of IEs → contact orders 1.34 for BB,0.74, 2.25 and 2.56 for the CH, PO and NP sidechains
Boris Fa£kovec ([email protected]) Protein modeling 21st December 2011 14 / 17
Future directions
Web application on IOCB's site - IEM calculation, contact matrix
Hydrophobic core de�nition - clusters in contact graphs
Scaling factors → energy functions
Boris Fa£kovec ([email protected]) Protein modeling 21st December 2011 15 / 17
References
�Decomposition of Intramolecular Interactions Between Amino-Acidsin Globular Proteins - A Consequence for Structural Classes ofProteins and Methods of Their Classi�cation�, Fackovec B andVondrasek J, 2011, chapter 20, �Systems and Computational Biology -Molecular and Cellular Experimental Systems�
�Optimal de�nition of inter-residual contact in globular proteins basedon pairwise interaction energy calculations�, Fackovec B andVondrasek J, Bioinformatics, submitted
Boris Fa£kovec ([email protected]) Protein modeling 21st December 2011 16 / 17
Acknowledgement
Ji°í Vondrá²ek Ji°í Vym¥tal Karel BerkaJi°í Kysilka
Boris Fa£kovec ([email protected]) Protein modeling 21st December 2011 17 / 17