Principles of Piqor Technology

DNA microarrays are miniaturized devices for the parallel analysis of ribonucleic acids by hybridization. The technology essentially reverses the setup of the classical hybridization methodology originnally invented by Southern [4]. DNA microarrays are widely used for expression analysis because they require only small amounts of sample material. Since - generally speaking - microarrays consist of immobilized probes spotted on a solid substrate which allows for automation, they meet the demands for high-throughput screening: low need of sample material and the possibility to automate the process. Microarray formats differ regarding the substrates, the probe selection strategies, and the way the probes are immobilized. Several aspects of the technology are discussed in detail below.

Memorec produces low-density microarrays suitable for two-colour experiments, that is, direct comparison of treated and control samples ending in the determination of expression ratios (Figure 2.2). The PIQOR (parallel identification and quantification of mRNA) system so far comprises a cDNA collection for the human, mouse, and rat species, a set of predefined arrays for specific subjects, and the possi-

Fig. 2.2 Outline of expression profiling using cDNA microarrays.

Fig. 2.2 Outline of expression profiling using cDNA microarrays.

i ^ Discrimination of mismatches i ^ Discrimination of mismatches

Oligonucleotides

bases

Oligonucleotides

bases

Fig. 2.3 Hybridization efficiency as a function of > cDNA chain length.

cDNAs bility to customize each array configuration with respect to the kind, number, and replicates of the probes. PIQOR arrays carry highly specific cDNA fragments 200-400 base pairs long, which are handpicked following complex bioinformatics methods and the amplification of suitable gene regions by PCR.

The length of the employed cDNA fragments is of particular interest regarding the robustness of the hybridization process. With increasing length of the used probes, the kinetics become more stable. The asymptotic curve (Figure 2.3) illustrates that a length of 200 bases is adequate to guarantee a stable hybridization independent of single nucleotide polymorphisms and varying GT contents of the single probes. The latter permits the hybridization conditions to be optimised for all of the probes that are hybridized in parallel. Furthermore, the sensitivity of DNA arrays increases the longer the probes are, since more labelled samples may hybridize with the matching immobilized probes. The length of the probes should be limited to a maximum of about 400 base pairs to avoid the probability of cross hybridization caused by repetitive elements and unspecific interactions. Furthermore, limitation to ~400 base pairs provides the ability to distinguish even genes from highly homologous gene families like those for the cytochrome P450 enzymes by choosing fragments from appropriate (i.e., poorly conserved) gene regions (see Section 2.3.2).

Compared to cDNA arrays, the spotting or in situ synthesis of oligonucleotides is a completely different manufacturing strategy having a deep impact on the way the results are analyzed. The arrayed oligonucleotides range from some 25 to 70 residues per probe. Depending on the length of the probes, a set of different oligonucleotides is necessary to unambiguously detect particular genes. Thus, several signals are gained for one gene, which sometimes may be contradictory due to differing hybridization kinetics or splice variants and therefore hard to interpret.

In general, the probe selection strategy for microarrays is subject to several prerequisites such as the objective of the experiment, the available sequence information regarding the organism to be investigated, and the efforts one is willing to supply in advance of the actual hybridization. If one were interested in overviewing the expression levels of as many genes as possible in a small set of experiments, it would be sufficient to spot cDNA libraries, uncharacterized ESTs, or oligonucleotides derived therefrom. This postpones in-depth sequence analysis and annotation to the point when hybridization has been performed and the resulting clones/genes of interest have been identified. This strategy reduces the expenses for prechip sequencing and bioinformatics to nothing. It may be most appropriate for high-throughput arraying projects. In contrast, if the experimental setup calls for accurate identification and quantification of particular mRNA species by hybridization, one can either buy ready-to-spot sets or go the time and effort of cloning and sequence-verifying the cDNAs oneself. However, these processes should be accompanied by extensive quality management (especially when targeting diagnostics).

0 0

Post a comment