Gene expression analysis reveals chemicalspecific profiles

1
Gene expression analysis reveals
chemical-specific profiles
Hisham K. Hamadeh, Pierre R. Bushel, Supriya
Jayadev, Karla Martin, Olimpia DiSorbo, Stella
Sieber, Lee Bannett, Raymond Tennant, Raymond
Stoll, J. Carl Barrett, Kerry Blanchard, Richard
S. Paules, and Cynthia A. Afshari
Jonathan Thomulka and Michelle Tsai
2
What is this used for?
Gene expression profiling technology is used to
examine multiple genes and signaling pathways
Toxicogenomics field of science dealing with
information about gene and protein activity in
response to toxic substances
3
Toxicity Techniques
One way is to study the change of gene expression
in response to chemical exposure using classical
methods of association Problem This cant be
used because chemical responses tend to be
frequently modulated by MANY compounds You would
have to look at many genes for these changes by
many compounds and at their collective state of
expression
Microarray
Concerted expression pattern across genes
constitutes the expression profile of a compound
at a certain dosage and time
4

• Hypothesis
• Whether structurally unrelated compounds from the
  same chemical class produced similar, yet
  distinguishable, gene expression profiles
• Compounds studied
• (all liver toxicants)
• 3 peroxisome proliferator class
• Clofibrate
• Wyeth 14,643
• Gemfibrozil
• 1 Enzyme inducer
• Phenobarbital

5

• Methods
• Sample
• Rats each had a different dosage of a single
  compound and how long they were administered
  either 24 hours or 2 weeks
• Histopathological analysis
• Physically looked at liver tissue
• RNA isolation
• RNA isolated using
• QIAGEN kits for real-time PCR

(normal rat liver tissue)
6

• Methods cont.
• cDNA microarray hybridization and analysis
• cDNA Rat Chip, M13 primers
• (Control Cy3, sample Cy5)
• RNA labeled by reverse transcription
• reaction with Cy3 and Cy5 conjugates
• Fluorescently labeled cDNAs were mixed
• and hybridized simultaneously to the cDNA
  microarray chip.
• Chips scanned and analyzed by a computer
• Locates targets on the array
• Gets rid of background
• Identifies differentially expressed genes using a
  probability-based method.
• Calculated a 95 confidence interval for the
  ratio intensity values
• (Genes having a value outside this interval were
  considered to be significantly differentially
  expressed)

7

• Methods cont.
• Real-time quantitative PCR
• RNA samples representing animals were used to
  validate the expression profile of 10 genes
  obtained from using cDNA microarray data
• Performed reverse transcription and real-time PCR
  of RNA at the same time, using custom primers
  (reverse and forward)
• Used a kit to detect double-stranded DNA
  generated during PCR amplification which also
  looked at fluorescent signals to monitor the
  reaction

8

• Results
• Just from looking at tissue
• 24 hour dosed animals No physical changes
• 2 week dosed animals these livers showed
  hypertrophic hepatocytes (increased liver cell
  size)

Animals treated with the 3 peroxisomes abundant
microvesiculated cytoplasm (lots of vesicles in
cytoplasm)
Animals treated with phenobarbital basophilic
stippling (dots)
9
(No Transcript)
10
(No Transcript)
11
Table 2 cont.
12
Validation of Experiment 1. Replicate analysis
3 animals were used for each compound and their
gene expression was measured on 3 chips. This
approach gave 9 measurements for each gene -gt
reduced the probability of getting false
positives 2. Routinely conducted re-sequencing
of the clones that were found to have significant
change. The clone set showed an accuracy of
90 3. Validated the expression profile of 10
genes across samples through high correlation
between cDNA microarray and RT-PCR (Table 3)
13
Figure 1
RT-PCR products of the genes indicated to show
the quality of the reaction (validate)
14
Compared cDNA microarray data with RT-PCR which
showed a high correlation -gt validated their
technique Induction or repression of each gene
was confirmed across multiple samples
15
Figure 2
Application of hierarchical cluster analysis
confirmed that individual animals could be
distinguished by the class of toxicants to which
they were exposed
Revealed 2 distinct nodes containing animals
treated with either of the 2 classes of chemicals
Gene
Different Animals
Clustering based on expression pattern
Red gene induction Green gene repression
16
Figure 3
Principal components analysis demonstrated close
proximity in the gene expression pattern between
the peroxisome chemicals, but distinct compared
to phenobarbital-exposed animals
17
Figure 4 Correlation of gene expression profiles
of individual animals exposed to different
chemicals
18
Figure 5
Animals
Genes
Red gene induction Green gene repression
19
Genes whose transcripts were increased in
phenobarbital-exposed animals over controls (time
independent)
20
Genes that are regulated in the same way for the
24 hours and 2 week test can be used in further
experiments as biomarkers when looking at gene
expression profiles
Below are different genes that were modulated by
peroxisome proliferators collectively in a
time-independent fashion
21
Figure 6
Red Gene induction Green Gene
repression (relative to control)
22
Transient and delayed alterations in gene
expression in response to the daily exposure to
the chemicals used in this study
A transiently altered transcripts (most
represented signaling related genes)
23
B genes that required a delayed period of time
for regulation. Associated with adaptation events
24
Figure 7
Yellow Past findings Blue New mechanistic
insights
25
There is high concordance of the expression
changes found in the microarray analysis in
phenobarbital-exposed rats with the results
obtained by scientists using other methodologies
over many years of study
26
Regulatory and molecular pathways affected by
phenobarbital
27
Gene expression profile data has supported past
findings and revealed new relations
28

• Potential Problems
• Whether animals should be grouped together as a
  pool or examined individually in toxicogenomics
  studies
• Pooling and then using individual animals for the
  verification steps can make microarray less
  costly, but can cause misinterpretation of data
  if one animal shows a distinct response or lack
  of response
• (This study analyzed individual chemically
  exposed animals against a pool of control animals)

29

• Conclusion
• Expression profiling can give a more
  comprehensive overview of molecular responses to
  toxicant exposure by revealing the coordinate
  expression of multiple genes in homeostasis and
  metabolic pathways
• The agreement of the expression profiles for
  these two chemical classes with past studies lend
  confidence in the use of these gene expression
  profiles in further pattern recognition
  applications
• There is an influence of the time of exposure on
  gene expression
• Data revealed gene expression that has not been
  previously associated with the compounds they
  used -gt could provide targets of mechanism

30
Further Reading Eisen, M.B., Spellman, P.T.,
Brown, P.O., Botstein, D. (1998). Cluster
analysis and display of genome-wide expression
patterns. Proc. Natl. Acad. Sci. U.S.A.
9514863-14868. Fielden, M.R., Zacharewski, T.R.
(2001). Challenges and limitations of gene
expression profiling in mechanistic and
predictive toxicology. Toxicol. Sci.
606-10. Nuwaysir, E.F., Bittner, M., Trent, J.,
Barrett, J.C., Afshari, C.A. (1999). Microarrays
and toxicology The advent of toxicogenomis. Mol.
Carcinog. 24153-159. Torres, T.T., Muralidhar,
M., Ottenwalder, B., Schlotterer, C. 2008. Gene
expression profiling by massively parallel
sequencing. Genome Res. 18172-177.