Transcript Toxicogenomics

Toxicogenomics
Heather Handley
JP Student
Toxicogenomics
“… field of study that combines clinical, genomic, and
proteomic data into a unified framework for understanding
the biochemical and genetic basis for various diseases.”
(Ballatori et al., 2003)
“… a new scientific field that elucidates how the entire
genome is involved in biological responses of organisms
exposed to environmental toxicants/stressors. It combines
information from studies of genomic-scale mRNA profiling,
cell-wide or tissue-wide protein profiling (proteomics),
genetic susceptibility, and computational models to
understand the roles of gene-environment interactions in
disease.” (Selkirk and Tennant, 2002)
Toxicogenomics
Functional
Genomics
Comparative
Genomics
TOXICOGENOMICS
&
PHARMACOGENOMICS
Population
Genomics
Transcriptomics
(Gene Expression)
Agenda
MICROARRAY OVERVIEW
TOXICOGENOMICS EXAMPLES
Sequence Analysis
1. Pharmacogenomics and individualized medicine
2. Comparative and functional toxicogenomics
Gene Expression Profiling
3.
Biomarkers of exposure and effect
4.
5.
Genomic approaches to study of toxic mechanisms
Toxicant “signature” profiling and predictive toxicology
Cytochromes P450
• Monooxygenase enzymes responsible for metabolism of
>80% of all clinical drugs and many organic pollutants
• Known roles in pollutant toxicity/carcinogenicity, drug-drug
interactions, adverse drug effects, drug reactivity
• Large, complex gene superfamily … ~2,500 individual
genes in bacteria, fungi, plants and animals
• Most animals have ~100 genes thought to be derived from
a single common ancestor via extensive gene/genome
duplication events
• Many inducible CYPs regulated by transcription factors in
nuclear receptor or bHLH-PAS gene (super)families
Microarrays 101
Definitions:
•
Platforms for massively parallel hybridization assays
•
High-density arrays of 100s to 1000s of probe-containing
features immobilized on a solid substrate
Terminology:
Traditional hybridizations (e.g. Northern, Southern, Western):
target = immobilized sample (e.g.all RNAs/DNAs)
probe = specific molecule of intrest in liquid phase
Microarrays:
target = sample in liquid phase
probes = molecules of interest immobilized on substrate
Microarrays 101
• Probe type
– cDNA: gene expression profiling
– Genomic DNA: CGH, ChIP-on-CHIP
– Oligonucleotides
25-80mers
spotted or synthesized in situ (photolithography or inkjet)
– Proteins: enzyme activity, protein-protein interactions
– Antibodies: protein expression
– Cells: biochemical functions, gene expression
– Tissue sections (TMA): high-throughput ISH or IHC
Microarrays 101
• Substrate
– Membranes (nylon, cellulose, etc.)
– Coated glass slides
o

o
Poly L-lysine
g-amino poly-silane (GAPS)
sugars
– Membrane-on-slide
• Probe density
– Low-density macroarrays (10-100s of features)
– Moderate-density microarrays (1,000s)
– High-density microarrays (10,000s)
Microarrays 101
• Radioactive detection
– Single sample per array
– Good sensitivity
A
B
• Fluorescent detection
– CAN apply multiple
samples per array
– Less sensitive but more
quantitative for changes
A+B
Pharmacogenomics
• Effect of polymorphisms on drug metabolism and toxic
side effects of pharmaceutical agents
• Oligonucleotide arrays can be used to identify presence of
specific alleles in individuals, or to quantify allele ratios in
populations
e.g. Affymetrix CYP chip (18 known mutations defining 10
alleles of CYP2D6 and 2 alleles of CYP2C19)
CYP2D6 poor metabolizer genotypes protect against
hepatitis C & cyrrhosis progression
• Microelectronic arrays can improve sensitivity/accuracy in
detecting single nucleotide differences
The Thousand
Dollar Genome
Genome Resequencing Technologies
• Sequencing by hybridization
• Solid-phase multiplex PCR
• Solexa TotalGenotyping with Single Molecule Arrays
• Single molecule sequencing
– Protein nanopore
– U.S. Genomics
Comparative
Toxicogenomics
• Species differences may exist at the level of gene
complement, enzyme function (coding sequence), or
transcriptional regulation (flanking genomic or intronic
sequences)
• Improved homology searching tools (combined with gene
prediction) can be used to detect all members of a gene
(super)family in a given genome
• Evolutionary analyses full gene complements facilitate
distinction of orthologous and paralogous relationships in
large gene superfamily
Functional
Toxicogenomics
• Species comparisons can be used to identify regions of
functional constraint or positive selection
• Algorithms for motif detection can be used to predict
regulatory elements (and possibly transcription factors)
• cDNA microarrays can be used to assess gene expression
• Antibody arrays can be used to assess protein levels
• Protein arrays can be used to assess enzyme function
Biomarker Arrays
•
Biomarker = biological response (ideally quantifiable) to
an environmental chemical, which provides a metric of
exposure and sometimes toxic effect(s); may be at the
molecular, cellular or whole organism level.
•
Multiple-gene biomarkers are likely to be more sensitive
and discriminating than single genes
e.g. Larkin et al. (2003)  macroarrays for assessing
exposure of fish to estrogenic compounds
– Cost-effective
– Well implemented
– Non-model organism
Genomic Approaches
to Toxic Mechanisms
• Toxicity of many compounds due to altered transcriptional
regulation of gene expression
• Gene expression profiling provides opportunities to
identify affected molecular pathways and cellular functions
• Platform(s)  cDNAs for oligonucleotides on glass slides
• Competitive hybridization with two-channel fluorescent
detection used to compare gene-specific relative
expression levels between two conditions
(except Affymetrix)
AHR signaling and Dioxin Toxicity
*
AHR-ARNT
Estrogen-responsive genes
?
AHR Gene Battery
• Cytochrome P450 1A (CYP1A)
• Glutathione S Transferase
Oncogenes
Cytokines
TOXICITY
“Adult Heart” Library
Atrium
Bulbus arteriosus
Sinus
venosus
Ventricle
Connective tissue
Smooth Muscle
Endothelium
Blood (>2 cell types)
TISSUE
• 4+ cell-types
• 10,000 genes?
cDNA LIBRARY
• 76,800 clones
– Unsequenced
– Redundant
Adult Heart Microarrays
4,896
+ ~100
AH clones
controls/genes
of interest
5,186 features
(~2,500 genes?)
Gene (mRNA)
Expression Profiling
Control Sample
Experimental Sample
Prepare total RNA
Prepare total RNA
Generate amino-allyl
modified cDNA
Generate amino-allyl
modified cDNA
Label with fluorescent
green dye (Cy3)
Label with fluorescent
red dye (Cy5)
Microarray
Analysis
Control
Equal Expression
More Expression
In Experimental
Experimental
More Expression
In Control
Data Analysis
2-fold
CYP1A
Dioxin-responsive genes
INDUCTION
159
41
94
0.5nM TCDD 5nM TCDD
1
54
58
SUPPRESSION
GENERAL TRENDS
• 361 clones differentially
expressed (p-value  1*10-4)
– 6 previously identified clones
– 99 assembled into 17 contigs
(15 known genes, 2 ESTs)
– ~75 low quality sequence
• Predominantly induction
• Low-dose responses
prevalent
Chemical Profiling
and Predictive Toxicology
• Diagnostic features determined from training set of
compounds with known mechanisms of action
• Methods for determining diagnostic subset include:
– Self Organizing Maps (SOMs) and Neural networks
– Bayesian statistics
• Caveats
– Must be certain about mechanisms of training compounds
– Can only distinguish states represented in training set
Thomas et al., 2001
Thomas et al., 2001