Single-molecule Detection in situ

University dissertation from Uppsala : Acta Universitatis Upsaliensis

Abstract: The human body contains a variety of different cell types that share a common genome, but differ in how they use the information encoded therein. Variation in molecular content exists even at the level of individual cells, and to provide deeper insight into complex cellular processes methods that permit analysis of each cell on its own are needed. This thesis presents molecular methods for localized detection of individual nucleic acid molecules. The developed methods are based on padlock probes and target-primed rolling circle amplification. Single-molecule detection sensitivity in combination with single-nucleotide genotyping selectivity enables detection of allelic DNA variants and closely related target sequences directly in cells. Padlock probes further enable multiplex detection of targets, and in combination with image analysis quantitative molecular data for individual cells can be acquired for large cell populations at a resolution that no other in situ detection method can provide at present. In this thesis, the in situ target-primed rolling circle amplification technique was first used for genotyping of a point mutation in the mitochondrial genome with padlock probes. This displayed mitochondrial DNA heterogeneity in cell populations. Application of the method on comet assay preparations showed that mitochondrial genomes are lost from these samples prior to analysis. Nuclear DNA targets, however, can be efficiently detected in corresponding samples. Padlock probes and rolling circle amplification are thus an attractive alternative to FISH analysis for localized DNA detection in comet assay samples. A method was also developed for localized detection of individual mRNA molecules with padlock probes and rolling circle amplification. This method provides unique possibilities to genotype allelic variants of transcripts in situ. mRNA expression is associated with substantial cell-to-cell variation and our presented method permits simultaneous visualization of multiple transcripts directly in complex tissue samples. Application of the methods presented in this thesis will enable new types of studies of biological samples from both normal and disease states.