Systems Biology of Ageing

1
Systems Biology of Ageing

• Jennifer Hallinan
• Centre for Integrated Systems Biology of
• Ageing and Nutrition
• (CISBAN)
• Newcastle University, UK

2
CISBAN
3
Newcastle University Campus for Ageing and
Vitality

• NIHR Biomedical Research Centre
• Clinical Ageing Research Unit
• Wolfson Research Centre
• Wellcome Biogerontology Building
• Magnetic Resonance/PET Imaging Centre
• Centre for Systems Biology of Ageing and
  Nutrition
• MRC Centre for Brain Ageing and Vitality
• NHS Centre for the Health of the Elderly
• Assistive Technology

4
Why do we age?
5
Continuing increase in life expectancy
Life expectancy is increasing by 5 hours a day
6
Scientific understanding of ageing

• Ageing is caused not by active gene programming
  but by evolved limitations in somatic maintenance

• Animals in nature die young
• No need or opportunity to evolve a program
• Programmed ageing, if it existed, would be
  unstable
• Immortal mutants do not arise

Protected
Survival
Wild
Age
7
Scientific understanding of ageing

• Ageing is a results of a build-up of cellular
  damage
• Complex network of mechanisms contributing to
  cellular ageing
• Telomere erosion
• Oxidative stress
• Mitochondrial dysfunction
• Protein homeostasis
• Multiple, complex and inherently stochastic

8
Scientific understanding of ageing

• To understand the cell and molecular basis of
  ageing is to unravel the multiplicity of
  mechanisms causing damage to accumulate and the
  complex array of systems working to keep damage
  at bay
• Ageing is intrinsically malleable, with important
  effects being mediated by nutrition
• Kirkwood, Cell 2005

9
Model systems for ageing research

• Homo sapiens Mechanisms contributing to cellular
  ageing in vitro (fibroblasts)
• Mus musculus How cell defects contribute to
  ageing in vivo
• Saccharomyces cerevisiae A model for high
  throughput screening of genes involved in damage
  responses, and their susceptibility to nutrition

10
Telomere length homeostasis relevant to ageing
and conserved across species
Telomerase
Telomere binding
H. sapiens
S. cerevisiae
11
In-vitro methodologies
expression arrays ITRAQ proteomics
live cell imaging
Passos et al., Science, submitted Nelson et al.
in prep
Passos et al. PLoS Biol 2007 Ahmed et al. J Cell
Sci 2008
12
Intrinsic ageing of mammalian cells

• e.g. Human fibroblasts in vitro
• Pathways and networks
• Heterogeneity
• Genomic and proteomic analyses
• Functional assays and targeted interventions
• Contribution to in vivo ageing

13
Senescent cell (human fibroblast)

• DNA damage foci
• Telomeres
• Overlap of damage foci with telomeres
• Mitochondria with high membrane potential (good)
• Mitochondria with low membrane potential (bad)

14
In vivo program

• Ageing mice colony
• Enriched environments
• In-vivo program work
• Caloric restriction
• Proteomics
• Transcriptomics

15
Yeast studies on telomere length homeostasis

• Genome-wide transcriptomic response to telomere
  uncapping in cdc13-1 mutants
• Genome-wide screen for proteins that affect
  growth of telomere capping mutants

• High-Throughput Robots
• Inoculate colony to liquid
• Grow to saturation
• Serially dilute
• Spot onto solid media
• Let colonies grow
• Photograph
• Identify and analyse interesting genetic
  interactions

16
Yeast studies on telomere length homeostasis
Image Captured
Lighting Corrected
Spots Located
Growth Quantified
ROD database
Epistasis Quantified
Growth Curves Generated
17
CDC13 epistatic interactions
18
Data integration and modelling is essential in
ageing research

• Multiple mechanisms a clear need for systems
  integration
• Multiple experimental models and human studies
• The big questions cannot be answered by a lot of
  disconnected separate studies
• The added-value of data co-ordination justifies
  the effort required for data integration and
  sharing

19
Portal for Systems Biology of Ageinghttp//cisban
-silico.cs.ncl.ac.uk/index.html
20
Towards a model of a virtual ageing cell

• Telomere loss and oxidative stress Proctor
  Kirkwood Mech Ageing Dev 2001
• Mitochondrial mutation Kowald Kirkwood J Theor
  Biol 2000
• Somatic mutation Kirkwood Proctor Mech Ageing
  Dev 2003
• Telomere capping Proctor Kirkwood Aging Cell
• Extrachromosomal DNA circles Gillespie et al J
  Theor Biol, in press
• Genetic pathways eg Sir2 gene action (in
  progress)
• Protein turnover Chaperones, heat shock proteins
  (in progress)
• Network models
• Mitochondrial mutation, oxidative stress, protein
  turnover (Kowald Kirkwood Mutation Res 1996)
• Somatic mutation, telomere loss, mitochondrial
  mutation, oxidative stress (Sozou Kirkwood
  JTheor Biol 2001)

21
Biology of Ageing e-Science Integration
Simulation (BASIS)

• Systems Biology Mark-up Language (SBML) for
  network representation
• Extend, share, merge models
• Internet-based (web services), database
• Stochastic simulation (compute cluster)
• Further development of SBML, MIRIAM, BioModels

www.basis.ncl.ac.uk, Kirkwood et al Nat Rev Mol
Cell Biol 2003
22
BASIS architecture
Basis architecture (hardware)
23
CaliBayes

• New statistical technology for Bayesian model
  calibration
• CaliBayes Java API
• CaliBayes Web Services
• R packages
• SBML models

Calibration Results
www.calibayes.ncl.ac.uk/
24
Data handlingSystems/Molecular Biology Data
Archive
25
SyMBA Overview

• Based on the Functional Genomics Experiment
  Object Model / Markup Language (FuGE-OM,
  FuGE-ML)?
• http//fuge.sourceforge.net
• Implements FuGE to create a systems biology data
  portal, archive, and data integration tool
• Archive of raw, High-Throughput (HT) data and
  associated metadata
• Useful as input in research into integrative
  bioinformatics within the CISBAN in silico group
• Open source available via SourceForge
• http//sourceforge.net/projects/symba/

26
Semantic data integration to support modelling
SAINT

• Model annotation though data integration
• Ontology based

27
SAINT Example CDC13
Lister et al., Bioinformatics, submitted
28
Lightweight data integration using Probabilistic
Functional Interaction Networks
?
Gene fusion
?
29
What can you do with them?

• Integrate large omics data sets
• Identify interactions which havent made it to
  the literature
• Automatically update
• Inform low-level modelling
• Identify other players in pathways
• Assign putative function to unknown genes
• Identify functional modules (clustering)
• Identify candidate genes
• Investigate lists of genes of interest
• Microarray
• Genetic screens

30
Key Downregulated Upregulated Knockout Both
31
Key Downregulated Upregulated Knockout Both
32
Protein RIF2 (RAP1-interacting factor 2).
DNA-binding protein RAP1 (SBF-E)
(Repressor/activator site-binding protein) (TUF).
Casein kinase II subunit beta' (CK II beta').
Casein kinase II subunit alpha' (EC 2.7.11.1) (CK
II).
33
Protein RIF2 (RAP1-interacting factor 2).
DNA-binding protein RAP1 (SBF-E)
(Repressor/activator site-binding protein) (TUF).
Cocitation, DIP
Casein kinase II subunit beta' (CK II beta').
Casein kinase II subunit alpha' (EC 2.7.11.1) (CK
II).
34
Ageing relevant networks
35
Clustering a relevant network
36
ONDEX for the analysis of epistatic interactions
37
A cross species, ageing relevant interactome
database CID
38
A cytoscape plugin for CID
39
A single direct pathway linking p21 to p38 MAPK
and TGFb
Passos et al., Science (submitted)
40
Conclusions

• Ageing is a complex, multi-factorial process
• Systems approach combines
• Detailed in vitro and in vivo studies
• Annotation and archiving of large amounts of data
• Integration of data generated internally and
  externally
• Statistical and computational analysis
• Dynamic modelling
• Iterative process

41
(No Transcript)
42
Acknowledgements

• Tom Kirkwood
• Daryl Shanley
• Carole Proctor
• Conor Lawless
• Anil Wipat
• Allyson Lister
• Katherine James
• David Lydall
• Steven Addinall
• Amanda Greenall
• Thomas von Zglinicki
• Joao Passos
• Doug Turnbull
• et al!

43
Both
Knockout
Downregulated
Upregulated
44
Saccharomyces cerevisiae
45
Calculate probabilities
Data integration
Eventually