Helical Peptides, Amyloid Inhibitors and Prediction of Protein Function Andrew Doig - PowerPoint PPT Presentation

1 / 20
About This Presentation
Title:

Helical Peptides, Amyloid Inhibitors and Prediction of Protein Function Andrew Doig

Description:

Helical Peptides, Amyloid Inhibitors and Prediction of Protein Function. Andrew Doig ... At least 20 peptides or proteins aggregate to form toxic amyloid ... – PowerPoint PPT presentation

Number of Views:93
Avg rating:3.0/5.0
Slides: 21
Provided by: andre162
Category:

less

Transcript and Presenter's Notes

Title: Helical Peptides, Amyloid Inhibitors and Prediction of Protein Function Andrew Doig


1
Helical Peptides, Amyloid Inhibitors and
Prediction of Protein FunctionAndrew Doig
2
1. The a-Helix
3
Surveying the PDBExample Side Chains at the
N-Terminus of the Helix
  • Gln, Glu, Asp, Asn, Ser, Thr and His are favoured
    at N1 and N2
  • Hydrogen bond to NH groups at the N-terminus

4
Experiments on Helical PeptidesExample
Hydrophobic-Polar Interactions (Ile-Lys)
  • Peptide Sequence Helix Content
  • Ac-AKAIAAAKAIAAAKAKAGY-CONH2 51
  • Ac-AKIAAAAKIAAAAKAKAGY-CONH2 37
  • Ile-Lys energy is -0.25 kcalmol-1

5
Cooperativity in Side Chain Interactions
Arg c1 trans
  • Arg-Phe -0.08 kcalmol-1
  • Phe-Met -0.35 kcalmol-1
  • Arg-Phe-Met -1.39 kcalmol-1
  • Side chain interactions are much stronger when
    present together
  • Fixing side chain into one conformation costs
    energy
  • But in triplet, only pay this once

Phe c1 trans
Met c1 gauche
6
Phosphorylated Helical Peptides
  • Helix Content () ?G phosphorylation
  • Site Sequence OH OPO3H- OPO32- OPO3H- OPO32-
  • N-cap SAAAAQRAAAARAGY-NH2 31 24 47 0.3 -0.6
  • N1 Ac-SAAAAQRAAAARAGY-NH2 50 51 61 -0.4 -1.6
  • N2 Ac-ASAAAAQRAAAARGY-NH2 42 56 68 -1.1 off scale
  • N3 Ac-AASAAAAQRAAAARGY-NH2 42 49 54 -0.7 -2.3
  • Mid Ac-AAAQRAAAASAAAARGY-NH2 38 25 17 0.6 1.2
  • Phosphorylation is very stabilising at N-terminus
    (more than any other amino acid when 2-)
  • Destabilising in helix interior
  • Phosphorylation may affect protein function by
    inducing or breaking helical structure

7
2. Inhibitors of Amyloid Formation
8
b-Sheets Aggregate
Designed b-sheet peptides usually aggregate At
least 20 peptides or proteins aggregate to form
toxic amyloid Probable cause of Alzheimers
disease, BSE, type II diabetes, Parkinsons etc.
9
Inhibition of Amyloid Formation
Association of free peptides inhibited
Association of N-methylated peptide
b-sheet
Association of N-methylated peptide
Association of free peptides inhibited
10
N-Methylation to Prevent b-Sheet Assembly
Natural peptides
Both edges free to associate with other b-strands
Backbone methyl groups block one edge of peptide
strand
Methylated peptides
This edge blocked
This edge free to hydrogen bond as normal
11
N-Methylated Inhibitors of b-Amyloid (Ab)
  • Aggregation of 40/42 amino acid peptide Ab is
    probable cause of Alzheimers disease
  • Hundreds of N-methylated peptide fragments
    investigated as potential inhibitors
  • Compared to other known derivatised peptide
    inhibitors
  • Work done in collaboration with start-up company
    - Senexis

12
Inhibition of 100mM Ab(1-40) aggregation
Thioflavin T assay
13
Electron Microscopy
Ab(1-40) Ab(1-40) NMe1
Ab(1-40) NMe3 Wild type fibrils
Altered morphology Few fibrils
14
Inhibition of 100mM Ab(1-42) toxicity by selected
Meptides at 11 stoichiometry
15
3. Prediction of Protein Function from Structure
? Non-Enzyme
? Enzyme
16
Orphan Genes and Structures
  • ? 200 genomes (? 25 eukaryotes, 150 bacteria, 20
    archaea)
  • ? 50 unknown function
  • PDB (July 6th 2004)
  • 26000 structures
  • 408 unknown function

17
Function from Structure
  • Aim - assigning function for orphan structures
    (no similar sequence or fold)
  • Two Classes - Enzyme/Non-Enzyme
  • Support Vector Machines machine learning

18
Structural Features
  • Amino acid composition
  • Amino acid composition of surface
  • Cofactors (NAD, ATP, FAD)
  • Metals (Fe, Mg, Cu, Ca)
  • Disulphides
  • Size of largest cleft
  • Secondary structure
  • Surface fractal dimension

19
Accuracy
  • 52 features 77
  • 36 features 80
  • Easier to predict an enzyme (90) than a
    non-enzyme (69)
  • Extended to predicting EC number and to predict
    functional class from sequence for orphan genes

20
(No Transcript)
Write a Comment
User Comments (0)
About PowerShow.com