Transcript Microarray Applications

Course LMP 1506S
Clinical Applications of Genomics and
Proteomics
Thursday, April 15, 2004, 9-11 am
MSB Room 6205
DNA
mRNA Protein
Eleftherios P. Diamandis MD,Ph.D
([email protected])
Website:www.acdclab.org
Where could you find my lecture slides?

Go to my website:
www.acdclab.org

Click on “Teaching”

Find lecture title and follow instructions on how
to download it
My Objective (unique for this course)

To demonstrate how new technological
advances in genomics and proteomics can be
used to help patients

A link between discovery and clinical
applicability is important and constitutes what is
currently known as “Translational Research”
Microarrays
What is a microarray?

A microarray is a compact device that contains a large number of
well-defined immobilized capture molecules (e.g. synthetic oligos,
PCR products, proteins, antibodies) assembled in an addressable
format.

You can expose an unknown (test) substance on it and then
examine where the molecule was captured.

You can then derive information on identity and amount of
captured molecule.
Microscope slide
DNA
microarray
16
17
18
7
Actin
DNA
CyclinD
DNA
DHFR
DNA
8
RB
DNA
E2F1
DNA
tubulin
DNA
9
control
DNA
Myc
DNA
Src1
DNA
Microarray Technology
Manufacture or Purchase Microarray
Hybridize
Detect
Data Analysis
Advantages of Microarrays

Small volume deposition (nL)

Minimal wasted reagents

Access many genes / proteins simultaneously

Can be automated

Potentially quantitative
Limitations of Microarrays

Relatively new technology (10 years old)

Still has technical problems (background)

Poor reproducibility between investigators

Still mostly manual procedure

Relatively expensive
Applications of Microarrays

Gene expression patterns

Single nucleotide polymorphism (SNP) detection

Sequence by hybridization / genotyping / mutation detection

Study protein expression (multianalyte assay)


Protein-protein interactions
Provides: Massive parallel information
If Microarrays Are So Good Why
Didn’t We Use Them Before??




Not all genes were available
No SNPs known
No suitable bioinformatics
New proteins now becoming available
Microarrays and associated technologies should be
regarded as by-products of the Human Genome
Initiative and bioinformatics
Reviews on Microarrays

A whole issue on Microarray Technology has been
published by Nature Genetics, Dec. 2002 (Vol. 32)

Books:
 Bowtell D. Sambrook J. DNA Microarrays. Cold Spring
Harbor Laboratory Press, 2003

Schena M. Microarray Analysis. Wiley Liss, 2003
History
1991 - Photolithographic printing (Affymetrix)
1994 - First cDNA collections are developed at Stanford.
1995 - Quantitative monitoring of gene expression patterns
with a complementary DNA microarray
1996 - Commercialization of arrays (Affymetrix)
1997- Genome-wide expression monitoring in S. cerevisiae (yeast)
2000 – Portraits/Signatures of cancer
2003 - Introduction to clinical practice
Microarray Fabrication
Two Major Methods:
[a]
Affymetrix  Photolithography
(400,000 spots in 1.25 x 1.25 cm area!)
[b] Everybody else  Mechanical
deposition (printing) [0.5 - 2nL] on
glass slides, membranes,etc
Principles of DNA Microarrays
(printing oligos by photolithography)
(Fodor et al. Science 1991;251:767-773)
Microarrays, such as Affymetrix’s
GeneChip, now include all 50,000
known human genes.
Science, 302:211, 10 October, 2003
Affymetrix Expression Arrays


They immobilize oligonucleotides (de novo synthesis; 25
mers)
For specificity and sensitivity, they array ~10 or more oligos
per gene

Latest version covers 50,000 genes (whole human genome)
in one array (Agilent Technologies has the same density
array; G4112A)

They label-test RNA with biotin and detect with streptavidinfluor conjugates
Preparation of Labeled mRNA
for Hybridization

Use oligo-dT with a T7 RNA polymerase promoter for
reverse transcription of extracted mRNA
(procedure makes cDNA)

Use T7 RNA polymerase and biotin-labeled
ribonucleotides for in vitro transcription (produces
biotinylated, single-stranded cRNA)

Alternatively: You can directly label cRNA with Cy-3
and Cy-5 fluors using T7 RNA polymerase
Microarray Applications
Differential Gene Expression
sample 1
(tumor
tissue)
RNA extraction and labeling
to determine expression level
RNA
cDNA
cRNA
Cy5-UTP
red fluorescence
reverse transcriptase,
T7 RNA polymerase
cDNA
RNA
sample 2
(reference)
cRNA
Cy3-UTP
green fluorescence
sample of interest
compared to
standard reference
Reference tissue
cRNA (green)
Tumor tissue
cRNA (red)
1
2
3
4
5
6
7
8
9
10
1
2
3
4
5
6
7
8
9
10
1
2
3
4
5
6
7
8
9
10
1
2
3
4
5
6
7
8
9
10
1
2
3
4
5
6
7
8
9
10
Human genes
on a microarray slide
Differential Gene Expression
(Budding vs Non-Budding Yeast)
Normal vs. Normal
Normal vs. Tumor
Lung Tumor: Up-Regulated
Lung Tumor: Down-Regulated
Lung Tumor: Up-Regulated
Signal transduction
Cytoskeleton
Proteases/Inhibitors
Kinases
Lung Tumor: Up-Regulated
Signal transduction
Cyclin
B1
Cytoskeleton
Cyclin-dependent
kinase
Tumor expressionrelated protein
Proteases/Inhibitors
Kinases
Lung Tumor: Down-Regulated
Signal transduction
Proteases/Inhibitors
Cytoskeleton
Kinases
Lung Tumor: Down-Regulated
Signal transduction
Cytoskeleton
Tumor necrosis
factor-related protein
Proteases/Inhibitors
Kinases
Genes Common to Many Tumors
(e.g.Kidney; Liver; Lung)
Up-regulated
Down-regulated
Microarray Applications
Whole Organism Biology
Whole Genome Biology With Microarrays
Cell cycle in yeast
Study of all yeast genes
simultaneously!
Red: High expression
Blue: Low expression
Lockhart and Winzeler Nature 2000;405:827-836
Microarray Applications
Single Nucleotide Polymorphism (SNP) Analysis
Single Nucleotide Polymorphisms (SNP)

DNA variation at one base pair level; found at a frequency of
1 SNP per 1,000 - 2,000 bases

A map of 1.42 x 106 SNPs has been described in humans
(Nature 2001; 409:928-933) by the International SNP map
working group

Identification: Mainly a by-product of human genome
sequencing at a depth of x10 and overlapping clones

60,000 SNPs fall within exons; the rest are in introns
Why Are SNPs Useful?

Human genetic diversity depends on SNPs between
individuals (these are our major genetic differences, plus
micro/minisatellites)

Specific combinations of alleles (called “Haplotypes”) seem
to play a major role in our genetic diversity

How does this genotype affect the phenotype
Disease predisposition?
Why Are SNPs Useful?

Diagnostic Application
Determine somebody’s haplotype (sets of SNPs) and assess
disease risk.

Be careful: These disease-related haplotypes are not as yet
known!
Nature 2003 426: 789-796
Genotyping: SNP Microarray



Immobilized allele-specific oligo probes
Hybridize with labeled PCR product
Assay multiple SNPs on a single array
TTAGCTAGTCTGGACATTAGCCATGCGGAT
GACCTGTAATCG
TTAGCTAGTCTGGACATTAGCCATGCGGAT
GACCTATAATCG
Many other methods
For SNP analysis have
been developed
SNP Analysis by Microarray
GeneChip® HuSNPTM Mapping Assay (Affymetrix)
More than 10,000 single nucleotide polymorphisms
(SNPs) covering all 22 autosomes and the X
chromosome in a single experiment.
Coverage:1 SNP per 210 kb of DNA
Needs:250 ng of genomic DNA-1 PCR reaction
Commercial Microarray for Clinical Use
(Pharmacogenomics)
Roche Product
CYP 450 Genotyping
(drug metabolizing system)
FDA Confusion
Class 1 medical device? (no PMA)
Class 2 or 3 medical device?
(requires pre-market approval)
From: Nature Biotechnology 2003 21:959-60
“The US government has blocked the sale of a
new kind of DNA diagnostic test, putting up an
unexpected barrier to the marketing of technology
to distinguish genetic differences in how patients
metabolize certain drugs.”
Science 2003 302: 1134
SNP Detection by Mass Spectrometry

High throughput detection of SNPs can be
achieved by mass spectrometry

SNP Center in Toronto (PMH) runs a
Sequenom Mass Spectrometry system
Microarray Applications
Sequencing by Hybridization
Sequencing By Hybridization


Address the need for high-speed, low-cost sequencing of
large sequences in parallel.
Example:
Consider examining 50Kb of sequence for 1,000 individuals.
Conventional Method
50Kb x 1,000 = 50 Mb of
sequence. At a rate of 500
bases per lane and 30
sequencing lanes, you can
produce 15 Kb of sequence per
day. You need 10 years for the
project.
Microarray
With one microarray of 1.25 x 1.25
cm dimension, you can scan 50 Kb
of sequence at once. You need
1,000 microarrays to complete task.
This may be completed in a few
days.
Sequencing by Microarray Technology
GeneChip p53 Assay Reagents

p53 Primer Set:
PCR primer pairs of exons 2-11 optimized for a
single-tube multiplex reaction

Fragment Reagent:
DNase 1 for DNA fragmentation

Control Oligonucleotide F1:
Positive hybridization control

p53 Reference DNA:
Human placental DNA
GeneChip p53 Assay
Performance Characteristics

Bases of genomic DNA analyzed
1262 bp
Base calling accuracy for missense
mutations
> 99.9%
Time from purified DNA to data
4.5 hrs
Maximum steady state throughout
equivalent to 6310 bp/hr
As validated on a set of 60 human p53 genomic DNA samples. “Maximum
steady state through-put based on one GeneChip analysis system.
Microarray Applications-Non Human - Chips
Avaliable Now (2004)

Pathogens (detection of Bird-Flu Virus strains)

Smallpox (bioterrorism)

Malaria (Plasmodium anopheles)

Zebrafish/Xenopus laevis (model organisms)

SARS Virus sequencing
Microarray Applications

Food Expert-ID (available by Bio-Merieux;2004)

DNA chip can verify quickly the animal species
composition and the authenticity of raw or processed
food and animal feed

By providing multi-species identification, FoodExpert-ID
will help to improve safety of food for human and animal
consumption, thereby contributing to consumer health
protection
Microarray Applications
Protein Microarrays
Protein Microarrays

Protein microarrays are lagging behind DNA microarrays

Same idea but immobilized elements are proteins instead of
nucleic acids

Number of elements (proteins) on current protein microarrays
are limited (approx. 500)

Antibodies for high density microarrays have limitations (crossreactivities)

Aptamers or engineered antibodies/proteins may be viable
alternatives
(Aptamers:RNAs that bind proteins with high specificity and affinity)
Applications
Screening for:
 Small molecule
targets
 Post-translational
modifications
 Protein-protein
interactions
 Protein-DNA
interactions
 Enzyme assays
 Epitope mapping
High-throughput proteomic analysis
Label all Proteins in Mixture
Protein array now commercially
available by BD Biosciences(2002)
Haab et al. Genome Biology 2000;1:1-22
Cytokine Specific Microarray
(Microarray version of ELISA)
IL-1 
IL-6
IL-10
marker protein
cytokine
Detection system
BIOTINYLATED MAb
ANTIGEN
CAPTURE MAb
VEGF
MIX
Competing High Throughput Protein Technologies
Bead-Based Technologies
 Luminex-flow cytometry
 Illumina-bead chips
Microfluidics
 Zyomyx
Mass spectrometry
 Ciphergen-protein chips
Microarray Clinical Applications
Cancer Diagnostics
Molecular Portraits of Cancer
Rationale:
The phenotypic diversity of breast and other tumors
might be accompanied by a corresponding diversity in
gene expression patterns that can be captured by using
cDNA microarrays
Then
Systematic investigation of gene expression
patterns in human tumors might provide the basis
of an improved taxonomy of breast cancers
Perou et al. Nature 2000;406:747-752
Molecular Portraits of Cancer
Breast Cancer
Perou et al. Nature 2000;406:747-752
Green: Underexpression
Black: Equal expression
Red: Overexpression
Left Panel: Cell Lines
Right Panel: Breast Tumors
Figure Represents 1753 Genes
Differential Diagnosis of
Childhood Malignancies
Ewing Sarcoma: Yellow
Rhabdomyosarcoma: Red
Burkitt Lymphoma: Blue
Neuroblastoma: Green
Khan et al. Nature Medicine 2001;7:673-679
Differential Diagnosis of Childhood Malignancies
(small round blue-cell tumors, SRBCT)
EWS = Ewing Sarcoma
NB = Neuroblastoma
RMS = Rhabdomyosarcoma
BL = Burkitt’s Lymphoma
Note the relatively small number of
genes necessary for complete
discrimination
Khan et al. Nature Medicine 2001;7:673-679
Microarray Milestone:
June 2003 Start Date
Question:
Can microarray profiling be used in clinical practice?
Prognosis/Prediction of therapy/Selection of patients who
should be treated aggressively?
Nature 2002; 415: 530-536
NEJM 2002; 347: 1999-2009
Van’t Veer and colleagues are using microarray profiling as a routine
tool for breast cancer management (administration of adjuvant
chemotherapy after surgery).
Their profile is based on expression of 70 genes
Treatment Tailoring by Profiling
premenopausal, lymph node negative
Gene Expression profiling
60%
40%
Poor signature
~ 56 % metastases at 10 yrs
~ 50 % death at 10 yrs
Good signature
~ 13 % metastases at 10 yrs
~ 4 % death at 10 yrs
Adjuvant chemo- and
hormonal therapy
No adjuvant therapy
or hormonal therapy only
295 patients
survival
metastases-free
Kaplan-Meier Survival Curves
time (years)
time (years)
Profiling in Clinical Practice

Metastatic potential is an early and inherent ability rather
than late and acquired

Predictive power of prognostic signature confirmed in
validation series

Prognostic profile outperforms clinical parameters

~30-40% reduction of unnecessary treatment and
avoidance of undertreatment (LN0 and LN+)
Therapeutic Implications

Who to treat:

Prognostic profile as diagnostic tool


Prognostic profile implemented in clinical trials


improvement of accurate selection for adjuvant therapy
(less under- and over-treatment)
reduction in number of patients & costs (select only
patients that are at metastatic risk)
How to treat:

Predictive profile for drug response

selection of patients who benefit
Commercial Clashes





Oncotype DX by “Genomic Health Inc”, Redwood
City, CA
A prognostic test for breast cancer metastasis based
on profiling 250 genes; 16 genes as a group have
predictive value; $3,400 per test
215,000 breast cancer cases per year (potential
market value > $500 million!)
No validation of test; No FDA approval
Test has no value for predicting response to treatment
Science 2004;303:1754-5
Commercial Clashes





Mammaprint marketed by Agendia, Amsterdam,
The Netherlands
Based on L.Van’t Veer publications
Test costs Euro 1650; based on 70 gene
signature
Prospective trials underway
Celera and Arcturus developing similar tests
(prognosis/prediction of therapy)
Science 2004;303:1754-5
Tissue Microarrays

Printing on a slide tiny amounts of tissue

Array many patients in one slide (e.g. 500)

Process all at once (e.g. immunohistochemistry)

Works with archival tissue (paraffin blocks)
Gene Expression Analysis of Tumors
cDNA Microarray
Lakhani and Ashworth Nature Reviews Cancer 2001;1:151-157
Tissue Microarray
Alizadeh et al. J Pathol 2001;195:41-52
Histochemical staining of microarray tissue
cores of ovarian serous adenocarcinoma

H&E

hK6
Histochemical staining of a microarray tissue
core of ovarian clear cell adenocarcinoma

H&E

hK6
Histochemical staining of a microarray tissue
core of ovarian serous adenocarcinoma

H&E

hK6
Microarray Future: Conclusions








Must go beyond describing differentially expressed genes
Inexpensive, high-throughput, genome-wide scan is the end
game for research applications
Protein microarrays will be deployed within the next few years
Publications are now being focused on biology rather than
technology
SNP analysis-population surveys, SNP map
Pharmacogenomics
Diagnostics
Industrialized biology: Rapid replacement of single-gene
experiments
CIPHERGEN.com
The ProteinChip® Company
ProteinChip® Arrays:
SELDI affinity chip surfaces (Ciphergen)
Reverse Phase Anionic
Cationic
NR
SO4
SO4
Receptor Ligand
SO4
NR
Enzyme
IMAC
Normal Phase
NR
Me(II)
Me(II)
Me(II)
Antibody Protein A/G
DNA
The SELDI Process and ProteinChip® Arrays
• Sample
goes directly onto the ProteinChip Array
• Proteins are captured, retained and purified directly on the chip (affinity capture
• Surface is “read” by Surface-Enhanced Laser Desorption/Ionization (SELDI)
®
)
Laser
Molecular Weight
100 m2
to
1 mm2
Sample
ProteinChip® Array
TOF-MS Detection PBS II System
7.5
5
2.5
TOF-MS
2000
Detector
Ionized proteins are detected and their mass
determined by Time-of-Flight Mass Spectrometry
8000
Serum Fingerprint by Mass Spectrometry
Results
Classification by Proteomic Pattern
Cancer
Unaffected New Cluster
Unaffected Women
No evidence of ovarian cysts
Benign ovarian cysts <2.5cm
Benign ovarian cysts >2.5cm
Benign gynecological
inflammatory disorder
2/24
1/19
0/6
0/7
22/24
18/19
6/6
0/7
0/24
0/19
0/6
7/7
Women with Ovarian Cancer
Stage I
Stage II, III, IV
18/18
32/32
0/18
0/32
0/18
0/32
Petricoin III EF, et al. Lancet 2002;359:572-577
Serum Proteomic Patterns for
Detection of Prostate Cancer
(Petricoin et al. JNCI 2002;94:1576-1578)
Comparison of the actual histopathologic diagnosis
following a single sextant biopsy set with the predicted
diagnosis from proteomic pattern analysis of patients’
serum samples obtained prior to biopsy
Predicted diagnosis by
proteomic pattern analysis
Actual histopathologic diagnosis
N
Prostate cancer
Stage I
Stage II
38
7
31
Benign disease PSA level, ng/mL
<4
4 - 10
> 10
75
137
16
Cancer
N (%)
36 (95)
7
29
5 (7)
40 (29)
6 (37)
Benign
N (%)
2 (5)
0
2
70 (93)
97 (71)
10 (63)
Petricoin et al. JNCI 2002; 94: 1576-1578
Current Reviews/Opinions/Commentaries
Diamandis, EP
Clin Chem 2003; 49: 1272-1275
Diamandis EP
J Natl Cancer Inst 2004; 96: 353-356
Diamandis EP
Mol Cell Proteomics 2004;3:367-78
Other
Technological
Advances
Comparative Genomic Hybridization

A method of comparing differences in DNA copy number
between tests (e.g. tumor) and reference samples

Can use paraffin-embedded tissues

Good method for identifying gene amplifications or
deletions by scanning the whole genome
Comparative Genomic Hybridization
Cot1DNA blocks repeats)
Label with Cy-3
Label with Cy-5
Nature Reviews Cancer 2001;1:151-157
Comparative Genomic Hybridization
Arrayed CGH

Same as previous slide but use arrays of BAC
clones instead of chromosomes
Laser Capture Microdissection
An inverted microscope with a low intensity laser that
allows the precise capture of single or defined cell
groups from frozen or paraffin-embedded histological
sections
Allows working with well-defined clinical material
Tumor Heterogeneity (Prostate Cancer)
Tumor Cells: Red
Benign Glands: Blue
Rubin MA J Pathol 2001;195;80-86
Laser Capture Microdissection
LCM uses a laser beam and a special thermoplastic polymer transfer cap (A).The cap is set on
the surface of the tissue and a laser pulse is sent through the transparent cap,expanding the
thermoplastic polymer. The selected cells are now adherent to the transfer cap and can be lifted
off the tissue and placed directly onto an eppendorf tube for extraction (B).
Rubin MA, J Pathol 2001;195:80-86
Molecular Profiling of Prostate Cancer
Rubin MA,
J Pathol 2001;195:80-86
The Future??
Cancer Patient
 Surgery/Biopsy
Cancerous Tissue
 Array Analysis
Tumor Fingerprint

Individualized Treatment
The Future??
General Population



Imaging
Multiparametric/miniature testing of
serum on a protein array
Mass spectrometric serum/urine
proteomic pattern generation
Screen-positive patients
Prevention; Effective Therapy
The Future??
Asymptomatic individuals

Whole genome SNP analysis
Predisposition to certain disease
Prevention (drugs; lifestyle)
Surveillance
Systems Biology
A “new” buzzword
 Aims to explain biological phenomena by combining:
* Biology/Medicine
* Mathematics
* Physics
* Engineering
* Chemistry
* Computer Science
See: http://www.systemsbiology.org/
http://systems-biology.org/

Systems Biology
Venter Re-Enters Sequencing
Craig Venter …….
experimental laboratory dedicated to evaluating
breakthrough sequencing technology with the goal of
sequencing a person’s genome in a single day for
only $1,000 - a fraction of current costs.
Nature Biotechnology 2002;20:965
If I want my DNA sequenced…how much would it
cost today?
Fundraising campaigns often repay donors with
mugs, buttons or books as a token of thanks, but
DNA sequencer J. Craig Venter is offering something
more personal. People who donate $500,000 to his
recently formed J. Craig Venter Science Foundation
can have their genome analyzed and get the results
on a disk.
Science 2002;298:947
New slides-updates
Mass spetrometric tissue imaging
See Nature Med 2001;7:493-496
Prostate Gene Expression Profiles
The top 210 genes with a statistically significant difference in expression between prostate cancer and BPH. Data are organized in a matrix format following hierarchical clustering analysis of
the 210 genes. Each row represents a single gene; each column represents a prostate sample. Normalized ratios correlating to the abundance of mRNA relative to a common reference are
represented by colors; red, down-regulated relative to reference; green, up-regulated relative to reference; black, approximately same as reference. Color saturation represents the magnitude of
deviation from the reference. Selected clusters of genes are listed with corresponding gene symbol and IMAGE clone ID.
Luo et al. Cancer Res 2001; 61: 4683-4688
Prostate Gene Expression Profiles
Overview of experimental procedures for gene expression profiling of prostate tissues. Prostate samples were trimmed and sectioned to enrich epithelial content in each
specimen and to facilitate sample homogenization. Total RNA was extracted from the samples and labeled with Cy3-dUTP in a RT reaction. RNA from a pool of two
BPH specimens was labeled in parallel with Cy5-dUTP and used as reference sample for all of the 16 prostate cancer and nine BPH samples. Labeled products from the
test samples were mixed with the labeled reference and cohybridized to microarrays containing cDNAs for 6500 human genes. Images were scanned, and data were
analyzed to study the gene expression patterns.
Luo et al. Cancer Res 2001; 61: 4683-4688