Gene Expression Profiling

---

By Mehrdad Hariri

The new millennium started in the final stages of the Human Genome sequencing project, which symbolized the beginning of a new chapter in biomedical research. A new “post-genomic” era has already started, during which understanding gene function is becoming the most important challenge. Up to now, the genomes of 800 organisms have been sequenced — indeed, sequencing is the first step toward understanding the molecular mechanisms of living organisms. The next immediate and more important question concerns the biological function of these genes in different situations and conditions.
The study of gene function is called functional genomics. Common approaches to studying functional genomics include measuring gene expression using high throughput techniques such as oligonucleotide and cDNA microarrays. Until recently, functional genomics was limited to elucidating only one gene at a time. But during the mid-90s, scientists developed high throughput techniques that provide a global picture of expression of thousands of genes in a single experiment. Microarrays provide scientists with the tools needed to observe changes in expression in parallel throughout cells in culture or in tissues. 1, 2, 3

Technology
Development of the first genomic microarrays goes back to the early ‘90s and was first reported by Foder and colleagues,4 followed by the use of cDNA clones by Brown.5 Since then, the technology, its usage and application of the techniques have been hugely modified.

There are several different types of genomic microarrays currently in use: short oligos, cDNA microarrays, and long oligos made with Agilent Technologies Inc. (Palo Alto, CA) inkjet technology. Affymetrix Inc. (Santa Clara, CA) was the first company to commercialize short oligo microarray technology, trademarked as GeneChip®. Affymetrix arrays are manufactured by a light-directed chemical synthesis process, which combines a series of photolithographic masks to define chip exposure sites with specific chemical synthesis steps, to produce high-density arrays of 25 base-pair oligonucleotides. An alternative method, cDNA microarray, is manufactured by a pin-based robotic system dispensing 1.5- to 2.5-kb-long cDNA clones onto a glass slide. cDNA arrays are a considerably cheaper form of microarray production, compared to Affymetrix’s proprietary process. A more recent technique for manufacturing arrays, also developed by Agilent, uses inkjet printing technology to manufacture long oligos (60 mer), which claims to increase sensitivity and specificity compared to shorter probes. Finally, Amersham Biosciences Corp. (now part of the GE Healthcare, Piscataway, NJ) has produced the most recent gene microarray technology (CodeLink™), which features 30-mer long oligonucleotides in a novel 3-D gel matrix. The matrix is comprised of a long-chain hydrophilic polymer. The company claims this 3-D gel matrix environment provides lower background and higher sensitivity for the arrays. 6, 7, 8

In all of these technologies, each spot identifies all, or part, of one gene. The basis of analysis is a relative comparison between two conditions in order to measure the differential expression in one condition versus another — for example, disease versus normal. In general, what is measured in microarray tests is not the absolute activity, but rather a relative expression of genes.

There are several steps in the analysis of microarray experiments:

1) RNA isolation and labelling the generated cDNA

2) hybridization of labelled sample to array, in Affymetrix gene chips, a cRNA copy of cDNA that is biotinlated and hybridize to the target chip

3) data analysis, a process in which data are filtered, normalized, and statistically significant genes with differential expression are selected

4) functional analysis or data mining

The material and expertise needed for any of these steps has generated a huge market for reagents and computational software, as well as expert individuals to perform, analyse and interpret the generated data. Many companies have started commercializing this software. There are also many public online resources available for data analysis.

Application
Within the last decade, our view of gene function has changed considerably. Profiling gene expression has helped to illustrate a genetic signature of particular cell lines, tissue or biopsy samples from patients in response to a treatment or specific condition, or to classify groups of patients according to their response or disease progress. These elements, either at the laboratory bench or in clinics, have facilitated a deep understanding of molecular mechanisms and have proposed new links, interactions and functions for genes. Gene-expression profiling using microarrays is the subject of numerous publications. A search for “gene expression profiling” in Medline indicates a dramatic increase in the number of publications within the last six years (Figure 1).

Some of the of gene expression profiling applications include:

Cancer
Gene expression profiling is used for almost all types of cancers, including breast, lung, colon, renal, prostate and lymphoma.

In general, studies on cancer can be categorized into different classes:
1)     Gene expression profiling as a prognosis tool: the comparison between different stages of particular tumour, or between patients with bad or better outcome, and those who have developed metastasis and those who have not. These studies help identify different molecular signatures of good and bad prognosis and may lead to the identification of new prognostic markers.9
2)     Identify the heterogeneity among tumours. One interesting area is to examine if histologically similar tumours share similarities in molecular signature. It is crucial to understand why tumours that share similar histopathologic criteria behave differently either in response to treatment or disease progression. This provides a different angle to our knowledge of tumours and can lead to the identification of molecular signature as one of the classification indicators.
3)     Gene expression as a diagnostic tool. Investigating the gene that plays an important role in tumour development as well as metastasis will help to identify diagnostic indicators of tumours.
4)     Defining the treatment strategy; many studies have been focused on understanding the response to drug treatment, and if it can be predicted according to expression profile.10-13

Pharmacology and drug discovery
Gene expression profiling is being used to improve lead optimizations, to characterize clinical development candidates, and to determine the differences in gene expression of tissues exposed to various doses of different drugs. This provides a better understanding of pharmacodynamics and molecular mechanism of drug action. In pharmacodynamics, gene expression profiling can help assess whether the drug is targeting the right molecular and biochemical pathway.

Gene expression profiling is also a useful tool in measuring drug sensitivity and toxicity. This area, called toxicogenomics, is used to find gene-expression patterns in a model tissue or organism exposed to a compound and their use as early predictors of adverse events in humans. Gene expression profiling has also been used in the discovery of new drugs. Screening biochemical compounds and measuring the gene expression response to them is one approach in drug discovery investigations. 1-7, 9-13

SNPs
Single nucleotide polymorphisms (SNPs) are single DNA base changes within the genome, at every 100 to 300 bases along the three-billion base human genome. SNPs are considered a source of genetic difference among individuals, making them important in pharmacogenomics and gauging individual responses to a given drug. Microarrays offer a very effective way of studying SNPs. Affymetrix has recently launched a GeneChip Mapping 100K of human SNPs. The company claims this chip can be used for other applications, including population genetics, and whole-genome association studies. Affymetrix researchers are developing SNP chips with 500,000 to one million SNPs.

Other areas of biomedical sciences
Microarrays have provided new insight into biomedical sciences and gene expression profiling and has become an essential part of many research laboratories. Measuring gene expression can potentially provide clues about common regulatory mechanisms, biochemical pathways and broader cellular functions. Investigations using microarrays have already provided a more comprehensive view on the complex regulation of the cell cycle, cell proliferation, differentiation and cell death. This will lead to the identification of many new functions for known genes as well as the discovery of new genes. In vitro and in vivo studies, as well as clinical sciences, have been filled with research elucidating differential gene expression patterns and conveying new insight into gene function. In infectious disease, microarrays characterize the pathogen or the host response to the pathogen. In other diseases, scientists are now able to visualize the global expression of target tissue in patients and elucidate a signature profiling of each disease. 13-16



Future Direction
Use of microarrays has led to expanding databanks, which has facilitated the exchange of information. It has also prompted global network mapping of functional genome analysis. Expression profiling has enriched biomedical science’s understanding of disease pathology and is helping identify points for potential therapeutic intervention. Despite this, microarrays have to overcome some technical difficulties. Reproducibility, robustness and standardizations are the most immediate deficiencies of this technique. There are collective efforts to address these issues from different sections of government, industry and academia. In the near future, the use of the microarray as a functional genomics tool will go beyond the benches and inflowing clinics to be used as a prognostic and diagnostic tool. This projects a promising future that will move toward personalized medicine by providing enormous data in diagnostic, prognostic, drug discovery and other areas of biomedical research in general. Personalized medicine will enable us to identify the genetic map of every patient, allowing us to predict responses to treatment options, identify possible risks and determine the most effective means of treatment.

Mehrdad Hariri is a PhD candidate at the Ontario Cancer Institute, University of Toronto (Toronto, ON). He can be reached at mhariri@uhnres.utoronto.ca.


References:
1.     Butte A. "The use and analysis of microarray data." Nat Rev Drug Discov 2002, 1:951-960.
2.     King HC, AA Sinha. "Gene expression profile analysis by DNA microarrays: promise and pitfalls. JAMA 2001, 286:2280-2288.
3.     Macgregor PF, JA Squire. "Application of microarrays to the analysis of gene expression in cancer." Clin Chem 2002, 48:1170-1177.
4.     Pease AC, D Solas, EJ Sullivan, MT Cronin, CP Holmes, SP Fodor. "Light-generated oligonucleotide arrays for rapid DNA sequence analysis." Proc Natl Acad Sci U S A 1994, 91:5022-5026.
5.     Shalon D, SJ Smith, PO Brown. "A DNA microarray system for analyzing complex DNA samples using two-color fluorescent probe hybridization." Genome Res 1996, 6:639-645.
6.     Barrett JC, ES Kawasaki. "Microarrays: the use of oligonucleotides and cDNA for the analysis of gene expression." Drug Discov Today 2003, 8:134-141.
7.     Cao W, C Epstein, H Liu, C DeLoughery, N Ge, J Lin, R Diao, et al. "Comparing gene discovery from Affymetrix GeneChip microarrays and Clontech PCR-select cDNA subtraction: a case study." BMC Genomics 2004, 5:26.
8.      van de Goor TA. “Using Microarrays to Detect Disease and Tailor Therapy.” Pharmaceutical Discovery Sep 1, 2005.
9.     Simon R, DG Altman. "Statistical aspects of prognostic factor studies in oncology." Br J Cancer 1994, 69:979-985.
10.     Winegarden N. "Microarrays in cancer: moving from hype to clinical reality." Lancet 2003, 362:1428.
11.     Ntzani EE, JP Ioannidis. "Predictive ability of DNA microarrays for cancer outcomes and correlates: an empirical assessment." Lancet 2003, 362:1439-1444.
12.     Simon R, MD Radmacher, K Dobbin, LM McShane. "Pitfalls in the use of DNA microarray data for diagnostic and prognostic classification." J Natl Cancer Inst 2003, 95:14-18.
13.     Clarke PA, R te Poele, P Workman. "Gene expression microarray technologies in the development of new therapeutic agents." Eur J Cancer 2004, 40:2560-2591.
14.     Stoughton RB. "Applications of DNA microarrays in biology." Annu Rev Biochem 2005, 74:53-82.
15.     Afshari CA. "Perspective: microarray technology, seeing more than spots." Endocrinology 2002, 143:1983-1989.
16.     Petricoin EF, 3rd, JL Hackett, LJ Lesko, RK Puri, SI Gutman, K Chumakov, et al. "Medical applications of microarray technologies: a regulatory science perspective." Nat Genet 2002, 32 Suppl:474-479.