Amplicon secondary structure prevents target hybridization to oligonucleotide microarrays

DNA microarrays that are used as end-point detectors for PCR assays are typically composed of short (15-25 mer) oligonucleotide probes bound to glass. When designing these detectors, we have frequently encountered situations where a probe would not hybridize to its complementary, terminally labeled PCR amplicon. To determine if failures could be explained by general phenomenon such as secondary structure, we designed a microarray to detect eight regions of the Escherichia coli 16S rDNA gene. We then amplified eight amplicons of different lengths using a biotin conjugated, antisense primer. Amplicons were then hybridized to the microarray and detected using a combination of signal amplification and fluorescence. In most cases, probe sequences complementary to the 5' region of the amplified products (sense orientation) did not hybridize to their respective amplicon. We tested for positional bias and showed that a biotin conjugated sense primer mirrored the same probe failures. Nick translated products, however, hybridized to all probes. Because nick translation generates many labeled fragments of random length, we concluded that this method disrupted secondary structure that otherwise prevented the amplicons from hybridizing to their respective probes. We also show that nick translation does not compromise detector sensitivity even when used with long PCR amplicons (ca. 1.5 kbp). Despite the increased cost of the nick translation, we concluded that this labeling strategy will reduce the time needed to design new assays as well as avoid possible false negatives during field applications. Alternative labeling strategies are also discussed. (C) 2004 Elsevier B.V. All rights reserved.