1

An Introduction to DNA microarrays

Rebecca Fry, Ph.D.

http://www.buffalo.edu/UBT/UBT

What is a DNA Microarray?

genes or gene fragments attached to a substrate (glass)

Hybridized slide

Two dyes

Image analyzed

Tens of thousands of spots

2

The Beginnings of Microarray Technology

Lockhart et al., 1996Nature Biotechnology“Expression monitoring by hybridisation to high-density oligonucleotidearrays”

Schena et al., 1995 Science“Quantitative monitoring of gene expression patterns with a complementary DNA microarray”

8 years later4162 references

3

Uses and Applications

Disease characterizationGene screening

Pathway mapping Target identification

Developmental BiologyPersonalized medicine

Mechanism of Action Studies Molecular Diagnosis of Disease

Toxicology Prediction of Drug Efficacy/Toxicity

A model experiment: Two samples of interest

Q: Which genes are differentially expressed in EGFP cells versus EGFP KD cells?

EGFP KDEGFP expressing cells

http://www.oardc.ohio-state.edu/plantranslab/gfp-a.gif

Differential

Gene

Expression?!

4

Cell

nucleus

cytoplasm

DNA

Inside every cell: DNA Serves as a genetic blueprint

RNA

Protein

Chem.ox.ac.uk

Dna molecule: berkeley.edu

Process of transcribing deoxyribonucleic acid toribonucleic acid is“TRANSCRIPTION”

Nucleic acidsmust be translated intoamino acids that make up proteins

Process: “TRANSLATION”

RNA

Protein

DNA

Relating Gene Expression

Is a gene expressed? Is protein produced

Ideally measure protein levels

High throughput protein assays complicated

We measure transcript level

5

Two Popular Microarraying Platforms

www.molgen.mpg.de

Spotted microarrays

cDNA: PCR products (500-2,000bp)

synthesized oligos>10,000 probes

Affymetrix“Gene Chip”500,000 probes25 mer (represents a fragment of a gene)

Introduction

Commercially available microarray

www.the-scientist.com

EGFP ORF

1 846

ATTCTGCAGTCGACGGTACCGCGGGCCCGGGATCCACCGGTCGCCACCATGGTGAGCAAGGGCGAGGAGCTGTTCACCGGGGTGGTGCCCATCCTGGTCGAGCTGGACGGCGACGTAAACGGCCACAAGTTCAGCGTGTCCGGCGAGGGCGAGGGCGATGCCACCTACGGCAAGCTGACCCTGAAGTTCATCTGCACCACCGGCAAGCTGCCCGTGCCCTGGCCCACCCTCGTGACCACCCTGACCTACGGCGTGCAGTGCTTCAGCCGCTACCCCGACCACATGAAGCAGCACGACTTCTTCAAGTCCGCCATGCCCGAAGGCTACGTCCAGGAGCGCACCATCTTCTTCAAGGACGACGGCAACTACAAGACCCGCGCCGAGGTGAAGTTCGAGGGCGACACCCTGGTGAACCGCATCGAGCTGAAGGGCATCGACTTCAAGGAGGACGGCAACATCCTGGGGCACAAGCTGGAGTACAACTACAACAGCCACAACGTCTATATCATGGCCGACAAGCAGAAGAACGGCATCAAGGTGAACTTCAAGATCCGCCACAACATCGAGGACGGCAGCGTGCAGCTCGCCGACCACTACCAGCAGAACACCCCCATCGGCGACGGCCCCGTGCTGCTGCCCGACAACCACTACCTGAGCACCCAGTCCGCCCTGAGCAAAGACCCCAACGAGAAGCGCGATCACATGGTCCTGCTGGAGTTCGTGACCGCCGCCGGGATCACTCTCGGCATGGACGAGCTGTACAAGAAGCTTAGCCATGGCTTCCCGCCGGCGGTGGCGGCGCAGGATGATGGCACGCTGCCCATGTCTTGTGCCCAGGAGAGCGGGATGGACCGTCACCCTGCAGCCTGTGCTTCTGCTAGGATCAATGTGTAGGCGGCCGCGACTCTAGATCATAATCAGCCATACCACATTTGTAGAGGTTTTACTTGCTTTAAAAAACCTCCCACACCTCCCCCTGA

Designing Oligos

70 mer oligo

specific to gene

of interest

6

Spotted microarrays

www.biorobotics.com

Robotic Printing

Robotics: Liquid HandlingResuspension of oligos

www.qiageninstruments.com

Overview of fabrication of spotted microarraysIntroduction

BRCA2

BRCA1

BLM

BID

ATR

ATM

APEXL2

APEX

ADPRTL3

ADPRTL2

ADPRT

ABH3

ABH2

ABH

DDB1

DCLRE1A

CETN2

CDK8

CDK7

CDK6

CDK4

CDK2

CCNH

CASP9

CASP8

CASP3

CASP2

BTG1

EXO1

FANCA

FANCC

FANCE

FANCF

ERCC5

ERCC4

ERCC3

ERCC2

ERCC1

ENDOG

EGFP

DUT

DMC1

MAD2L2

LIG4

LIG3

LIG1

JUN

HUS1

HSU24186

HCNP

HAP1

GTF2H4

GADD45G

GADD45B

GADD45A

G22P1 POLQ

PRKDC

PRSS25

RAD1

RAD17

RAD18

RAD23A

RAD23B

RAD50

RAD51

RAD51C

RAD51L1

RAD51L3

RAD52

RAD54B

alkB homolog

Breast cancer 1

Excision repair

Growth arrest and DNA-damage-inducible

7

A closer look at Spotted microarraysSome nomenclature

www.molgen.mpg.de

each spotrepresents a gene or gene fragment

gene

RNA

“probe”

“target”

Introduction

562 nm

cy3

cy5

www.amersham.com

cy3 cy5

Differential dye incorporationcy5 less well than cy3Light sensitivity: cy5 more easily degraded

664 nm

510 nm

emission

emission

Introductioncy3 and cy5: Commonly used dyes

8

www.genetics.ucla.edu

Labeled cDNA preparation

RNA

Sample 2

Spotted MicroarrayTarget preparation

Reverse transcriptionFlourescent dyes

Introduction

EGFP KD

Combined in equal amounts

Spotted microarray target preparationDirect labeling

cy5cy3

cDNAcDNA

EGFP

Co-hybridized to array

yellow cy3=cy5red cy5>cy3

green cy3>cy5

DNA

Direct Labeling (Spotted Arrays)

RNA

RT

Reverse transcriptionRNA

Enzyme (Superscript RT)Dye

Oligo d(T)nucleotides

AAAAAAAAA AAA

AAAAAA

EGFP KDEGFP

9

28S

Liver

18S

28S

6 kb

4 kb

2 kb

1 kb0.5 kb0.2 kb

Electropherogram (28S/18S Ratio~2)

18S

28S

Fluo

resc

ence

Time (seconds)

0

10

20

30

40

50

60

70

19 24 29 34 39 44 49 54 59 64 69

Sample 1 Ladder Sample 2

RNA quality control

Lad, L, P

Gel Image (in silico) Sharp, Clear Bands

Pre-labeling quality control:

Determine RNA Quality Agilent Bioanalyzer: 50-500 ngNo more formaldehyde gels!!

Microarray Measurements

Scanner Image Analysis .txt or .xls file

Image Analysis: Spotted arrays

10

Spotted microarrays

www.molgen.mpg.de

Signal is average of pixel intensities of spot

2 numbers per spot

Microarray Measurements

cy5

cy3

Signal: Spotted arrays

Information:

Gene nameOligo idCy3 intensityCy5 intensity

Image Analysis: Spotted arraysWhat information do we see?

117.8919272.5676H200008289ADPRT

108.9855259.5362H200008289ADPRT

108.5797265.7391H200008289ADPRT

131.2286324.886H200008289ADPRT

115.6119296.5224H200014101PRKDC

117.8676306.8235H200014101PRKDC

129.5333340.933H200014101PRKDC

132348.522H200014101PRKDC

2351.6976242.5control1control1

2370.5756434.93control1control1

2068.515786.72control1control1

2371.1636642.5control1control1

Spot Mean Intensity (w685)

Spot Mean Intensity (w595)Clone IDName

Steps:Normalize globally

Calculate average of cy3 and cy5Bring cy5 numbers to cy3 by multiplying by common factor

Ratios (take ratio of wt/KD…look for 2 fold cutoff)Log2 (calculate log 2 of ratio to differentiate increase or decrease)Reproducibility (how did the four replicates perform?)

11

S c a tte r P lo t

dChip e xpre s s iontis s diff_ID0 2000 4000 6000 8000 10000 12000

-9

-8

-7

-6

-5

-4

-3

-2

-1

0

1

2

3

4

5

6

7

8

9

2 fold increase

2 fold decrease

Gene population

Plots

EGFP

ver

sus

EGFP

KD

Rat

io L

og2

Data AnalysisRequires software: Spotfire

Requires ability to search for Patterns and Trends

12

mammalian DNA damage response pathway

The DNA damage response: puttingcheckpoints in perspectiveBin-Bing S. Zhou* & Stephen J. Elledge²NATURE |VOL 408 | 23 NOVEMBER 2000 |

Good luck!