Influence of the length of target DNA overhang proximal to the array surface on discrimination of single-base mismatches on a 25-mer oligonucleotide array

Abstract

Background: The performance of probes on an oligonucleotide microarray can be characterised in terms of hybridisation signal strength and the ability to discriminate sequence mismatches between the probe and the hybridising target strand, such as those resulting from SNPs. Various properties of the probe affect mismatch discrimination, such as probe length and the position of mismatched bases, and the effects of these factors have been well characterised in a variety of array formats.

Results: A low-density microarray was developed to systematically investigate the effect of a probe's position within hybridised target PCR products on the tolerance and discrimination of single-nucleotide mismatches between the probe and target. In line with previous reports, hybridisation signals were attenuated by different degrees depending on the identity of the mismatch, the position of the mismatch within the probe, and the length of the PCR product. However, the same mismatch caused different degrees of attenuation depending on the position of the probe within the hybridising product, such that improved mismatch discrimination was observed for PCR products where a greater proportion of the total length was proximal to the array surface.

Conclusions: These results suggest that the degree of mismatch discrimination can be influenced by the choice of PCR primers, providing a means by which array performance could be fine-tuned in addition to manipulation of the properties of the probes themselves.

Figure 1

Illustration of the position of probes within PCR-amplified product. The section of the product overhanging the probe proximal to the array surface is referred to as the 5′ segment of the product, and the section overhanging the 5′ end of the probe is referred to as the 3′ segment of the product.

Figure 2

Array analysis of PCR products with similar overall length but different 5′ segment lengths: probe set 1. Panels show the signal intensity for perfect match and mismatch probes hybridised to PCR products with total lengths between 293 and 323 bases and 5′ segment lengths ranging from 43 to 272 bases; each panel shows the effect of a different mismatch at three positions within the probe. Approximately 10¹² copies of PCR amplicon were applied to each array. Results shown are mean values for duplicate spots on the same array (error bars show standard deviation). 5AA = AA mismatch between probe and PCR product located in the 5′ third of the probe, etc.

Figure 3

Array analysis of PCR products with similar overall length but different 5′ segment lengths: probe set 2. Panels show the signal intensity for perfect match and mismatch probes hybridised to PCR products with total lengths between 289 and 317 bases and 5′ segment lengths ranging from 23 to 246 bases; each panel shows the effect of a different mismatch at three positions within the probe. Approximately 10¹² copies of PCR amplicon were applied to each array. Results shown are mean values for duplicate spots on the same array (error bars show standard deviation). 5AA = AA mismatch between probe and PCR product located in the 5′ third of the probe, etc.

Figure 4

Array analysis of PCR products with the same 5′ segment length but different total lengths. Arrays were hybridised with PCR products with constant 5′ segment lengths (A: 43 bases; B: 191 bases; C: 272 bases) but different overall lengths ranging from 94 to 978 bases, 242 to 1126 bases, and 323 to 1207 bases, respectively. Signal intensities were normalised relative to the signal for the perfect match probe on each array and are plotted against the 5′ segment length shown as a proportion of the total product length. Approximately 10¹² copies of PCR amplicon were applied to each array. Results shown are mean values for duplicate spots on the same array (error bars show standard deviation). PM = perfect match probe. AA = AA mismatch between probe and PCR product, etc.

Figure 5

Array analysis of symmetrical and asymmetrical PCR products. A. Signal intensity for hybridisation of symmetrical and asymmetrical PCR products of 312 and 317 bases, respectively, both with a 5′ segment length of 43 bases. Asymmetrical PCR was carried out using a forward primer modified for linear-after-the-exponential (LATE) PCR, with a primer ratio of 1:10. Approximately 500 ng total DNA (single-stranded plus double-stranded) was hybridised to each array. Results shown are mean values for duplicate spots on the same array. AA = AA mismatch between probe and PCR product, etc. B. Agarose gel electrophoresis of symmetrical and asymmetrical PCR products. The products of symmetrical and asymmetrical PCR were visualised by agarose gel electrophoresis with GelRed nucleic acid stain before and after treatment with S1 nuclease to confirm that the product of asymmetrical PCR contained single-stranded DNA (removed by S1 nuclease) as well as double-stranded DNA.