Expressed gene number can be measured en masse

KEY CONCEPTS:

• "Chip" technology allows a snapshot to be taken of the expression of the entire genome in a yeast cell.
• ~75% (~4500 genes) of the yeast genome is expressed under normal growth conditions.
• Chip technology allows detailed comparisons of related animal cells to determine (for example) the differences in expression between a normal cell and a cancer cell.

Recent technology allows more systematic and accurate estimates of the number of expressed genes. One approach (SAGE, serial analysis of gene expression) allows a unique sequence tag to be used to identify each mRNA. The technology then allows the abundance of each tag to be measured. This approach identifies 4,665 expressed genes in S. cerevisiae growing under normal conditions, with abundances varying from 0.3 to >200 transcripts/cell. This means that ~75% of the total gene number (~6000) is expressed under these conditions (Velculescu et al., 1997). Figure 3.34 summarizes the number of different mRNAs that is found at each different abundance levels.

The most powerful new technology uses chips that contain high-density oligonucleotide arrays (HDAs). Their construction is made possibly by knowledge of the sequence of the entire genome. In the case of S. cerevisiae, each of 6181 ORFs is represented on the HDA by 20 25-mer oligonucleotides that perfectly match the sequence of the message and 20 mismatch oligonucleotides that differ at one base position. The expression level of any gene is calculated by subtracting the average signal of a mismatch from its perfect match partner. The entire yeast genome can be represented on 4 chips. This technology is sensitive enough to detect transcripts of 5460 genes (~90% of the genome), and shows that many genes are expressed at low levels, with abundances of 0.1-2 transcripts/cell. An abundance of <1 transcript/cell means that not all cells have a copy of the transcript at any given moment.

The technology allows not only measurement of levels of gene expression, but also detection of differences in expression in mutant cells compared with wild-type, cells growing under different growth conditions, and so on (Hughes et al., 2000; for review see Young, 2000). The results of comparing two states are expressed in the form of a grid, in which each square represents a particular gene, and the relative change in expression is indicated by color. The upper part of Figure 3.35 shows the effect of a mutation in RNA polymerase II, the enzyme that produces mRNA, which as might be expected causes the expression of most genes to be heavily reduced. By contrast, the lower part shows that a mutation in an ancillary component of the transcription apparatus (SRB10) has much more restricted effects, causing increases in expression of some genes (Holstege et al., 1998).

The extension of this technology to animal cells will allow the general descriptions based on RNA hybridization analysis to be replaced by exact descriptions of the genes that are expressed, and the abundances of their products, in any given cell type (Mikos and Rubin, 1996).