DNA Microarray with Colorimetric Detection for Gene Expression Analysis

Konan Peck

《Journal of Genetics and Molecular Biology》 11卷3期 (2000/09) Pp. 114-116

https://doi.org/10.30047/JGMB.200009.0005

並列摘要

The cDNA microarray provides an efficient and powerful tool to solve the difficulties in quantifying expression of a large number of genes. Enzyme colorimetric detection method is an alternative to fluorescence detection for measuring multiple parameters simultaneously. Microarray with colorimetric detection (microarray/CD) makes the already powerful method more accessible to academic laboratories. The system utilizes an enzyme-linked colorimetric detection method to identify differentially expressed genes on Nylon filter membranes.
The technology of the cDNA microarray with colorimetric detection system is consisted of four major components:
1. Array Fabrication:
A UniClone system that archives non-redundant gene clones was constructed by using 2 million cDNA clones obtained from the IMAGE consortium human eDNA libraries. The non-redundant human gene clones were identified by applying relational database method to analyze databases such as GenBank,
dbEST, Unigene, IMAGE, and others. The non-redundant cDNA clones were PCR amplified and the results were verified by slab gel electrophoresis. The PCR products are concentrated and arrayed on Nylon filter membranes by an arraying machine fitted with steel pins. The arraying machine is capable of holding up to 100, 000 samples at a time and able to perform walk-away operations. By using 24 spotting pins, the arraying machine is able to produce microarrays with density exceeding 6,000 spots per cm2 Flowcharts of the constructions of the UniClone system and the fabrication of arrays containing non-redundant gene sets are attached as illustrations.
2. cDNA probe preparation:
To prepare hybridization probes, messenger RNAs are extracted from cell cultures. For clorimetric detection, labeled cDNA probes are generated by random primed labeling during the reverse transcription reaction using both biotin-dUTP and digoxigenin-dUTP labels. These labeled cDNA probes once hybridized and processed by the enzyme colorimetric method yield differential colors on target gene sports on Nylon membranes. Different combinations of enzyme/substrate pairs yield different chromogens. Streptavidin-ß-galactosidase binds to biotin and reacts with X-gal (5-bromo-4-chloro-3-indolyl-ß-D-galactopyranoside) to yield blue chromogen and anti-DIG-alkaline phosphatase binds to digoxigenin (DIG) and reacts with Fast Red TR/AS-MX to yield red chromogen.
3. Expression level measurement:
Colorimetric detection is different from laser induced fluorescence detection that flatbed scanners or drum scanners commonly used in image digitization or pre-press industry are employed. These devices are highly accessible and much less costly than the laser induced fluorescence scanner. Both single and dual color enzyme colorimetric detection methods can be applied to microarray measurements. Single color detection is feasible if the variation from one array to the other is much lower than the expected difference of gene expression. For single color enzyme colorimetric detection, the coefficient of variation (CV) among arrays determines the discrimination limit of differential expression. For dual color detection, the discrimination limit is bound by the resolution of colors and is about 70% in expression levels.
The characteristics of colorimetric measurement in terms of sensitivity, throughput, and dynamic range will be discussed and compared with the laser induced fluorescence detection and autoradiography in the presentation. To profile gene expression from small amount of samples such as tissue specimens, signal amplification methods such as the catalyzed reporter deposition method and the RNA amplification method will be demonstrated.
4. Expression data analysis:
The algorithm for analyzing a dual color image obtained by enzyme colorimetric detection is different from the algorithm for a pseudo-color encoded image obtained by laser induced fluorescence detection. To identify differentially expressed genes, the color of each spot in an array is separated into three primary colors. The coordinate of 3-D space represents the composition of the three primary colors for each spot. The differential expression ration is interpolated from a calibration of the distances of a series of control spots of various expression ratios in a 3-D coordinate of primary colors.
Different applications of microarray require different strategies for data analysis. Statistical tools are needed to analyze vast amount of expression data. Data preprocessing is needed to assess the quality of the data sets. Array fabrication using solid pins or inkjet method to deposit samples results in around 7% coefficient of variation (CV) while array fabrication based on quail pins usually results in more than 10% CV among arrays of the same batch. The greater variation of expression data comes from the culture conditions and factors associated with variations of gene expression will be discudssed.