Sunday, 19 May 2019

Gene Expression Profiling

Gene Expression Profiling

The efforts of the human genome project have led to the identification of thousands of genes in many different organisms, though functional information for most of the newly identified genes are lacking. Quantitative analysis of gene expression is one step towards an understanding of gene function. DNA microarrays were initially developed for analyzing the expression of large numbers of genes simultaneously (Schena et al., 1995). The microarray technique uses a robot for precise application of
micro-volumes of DNA solutions to a matrix (i.e. glass slide or nylon membrane). The total pool of mRNA from experimentally manipulated cells, tissues or animals can be labeled using an oligo deoxy-thymidine (dT)-primed reverse transcriptase reaction and fluorescent (or radioactive) nucleotides (Freeman et al., 2000; Hegde et al., 2000). The labeled cDNA targets (synthesized from mRNAs) are allowed to bind to the DNA probes immobilized on the glass slide (or nylon membrane). 

The intensity of the hybridization signal is related to the amount of RNA that is present for the corresponding species. Thus, one can readily evaluate the differential expression of thousands of genes in a single microarray experiment. A challenge for functional genomics will be the reverse engineering and modeling of gene regulatory networks that are revealed by global gene expression profiling.

Expressed Sequence Tags (ESTs)
The initial step in functional genomics is to make an extensive catalog of genes expressed in each tissue. Microarray analysis requires careful preparation of tissue-specific cDNA libraries for high-throughput DNA sequencing: normalization of abundant clones (Soares et al., 1994) and eventual subtraction of redundant clones (de Fatima Bonaldo et al., 1996). High-throughput, single-pass DNA sequencing from either the 5′- or 3′-end of randomly picked cDNA clones yields ESTs, representing 500–800 bp reads, that are then used in BLAST searches for gene identity (Altschul et al., 1997). Typically, the approach of sequencing from the 5′-end of EST clones favors gene identity because of the likelihood of finding coding region sequence. The overlapping EST sequences from the same gene (contigs) can be used to assemble full-length in silico cDNA sequences. The number of contigs assembled by this bioinformatic process also gives a good estimate of the number of genes expressed in a given tissue.

(Fig.) Principles of DNA microarray technology. (a) Oligonucleotides (oligos) are directly synthesized on to glass slides. Alternatively, cDNA inserts are PCR amplified, using vector-specific primers, and printed at high density on a solid matrix by a computer-controlled high-precision XYZ robot. (b) Pools of RNA are reverse transcribed in the presence of fluorescently labeled (or radiolabelled) nucleotides (i.e. targets) and allowed to hybridize to the cDNA sequence immobilized on the array (i.e. probes). Hybridizing species are detected with a laser scanner (or a phosphor-imager for nylon membrane arrays). (Originally published in an article by van Hal et al., 2000; reproduced with permission of Elsevier Science, BV.)

Genome-scale cDNA sequencing and DNA microarray technology, initiated by the Human Genome Project, have enabled the exploration and discovery of thousands of genes in model organisms (humans, mice, fruit flies, yeast, etc.). Until now, the great- est obstacle to implementing similar studies of functional genomics in chickens has been the lack of a complete catalog of tissue-specific gene sequences.

(Fig.) CAP3 assembly of ESTs sequenced from different chicken cDNA libraries in the University of Delaware collection. The CAP3 sequence assembly program (Huang and Madan, 1999) was used to build contigs of EST sequences using a 40-base overlap and 95% sequence identity. The overlapping sequence of each EST is shown by the vertical dashed lines and the percentage identity is indicated above each bar. The BLASTX score for each individual EST is given within the bar. The BLASTX score for the high fidelity in silico cDNA was 535. UTR, untranslated region; bp, base pairs.

Before 2001, there was only a modest collection of chicken ESTs (< 22,000) cataloged in public databases maintained at EMBL/GenBank, Roslin Institute, University of Delaware and University of Hamburg. Although the chicken once lagged behind other animals in genomic research, a milestone in avian genomics was reachedin2001with the culmination of several independent chicken EST sequencing efforts. Using high-throughput DNA sequencing, the University of Delaware has identified 37,388 ESTs from several primaries and normalized tissue-specific chicken cDNA libraries: lymphoid tissue, liver, abdominal fat, breast and leg muscle/bone growth plate, pituitary/hypothalamus/ pineal, and reproductive tract (testis/ovary/ oviduct) prepared mainly from broiler chickens. These EST sequences were entered into GenBank and the original chicken EST. 

Another chicken EST database of clones that were sequenced from chicken bursa cells ( presently lists 11,260 ESTs sequenced from the unnormalized dkfz426 library and 12,982 ESTs from the normalized riken1 library (Abdrakhmanov et al., 2000; Buerstedde
et al., 2002). On 17 December 2001, a consortium of British institutions, funded by the British Biotechnology and Biological Sciences Research Council (BBSRC), unveiled a much larger chicken EST database ( of 299,506 clones that were derived from high-throughput sequencing of 21 tissue-specific cDNA libraries prepared from Leghorn (layer type) chickens (Chambers, 2002). The addition of chicken EST sequences to the BBSRC database since the first release now brings the total to 330,388chickenESTs.The combined efforts from these high-throughput chicken EST sequence projects have provided thousands of ESTs from most all chicken tissues, with minimal overlap. The next goal will be to develop a complete gene catalog for the chicken genome from this bonanza of ESTs.

No comments:

Post a comment