Gene expression profiling with real-time PCR

Doktorsavhandling, 2005

The sequence of the entire human genome is available today, and this information can be used to understand, characterize, and treat complex diseases such as cancer. Many cellular functions related to survival, growth and differentiation are reflected in the gene expression pattern, and the ability to quantify transcription levels of specific genes is, and always has been, central to research on gene function. By means of gene expression profiling, it will be possible to diagnose an individual’s state of health and also to monitor individual patients’ responses to medication, treatment, and altered living conditions. Real-time PCR, where the amount of DNA product is measured during ongoing amplification, is the most sensitive and accurate method to determine the amount of a specific DNA in a sample. We have contributed to the development of both the technology and its many applications. Our first invention was a fluorescent reporter molecule, the LightUp probe, that binds target DNA in a specific way. The LightUp probe is a peptide nucleic acid (PNA) to which an asymmetric cyanine dye is coupled. LightUp probes are excellent reporters in real-time PCR for nucleic acid quantification. For gene expression profiling to work, mRNA must be reverse transcribed to cDNA before PCR amplification. We evaluated the reverse transcription reaction for different parameters, such as reverse transcriptase, priming strategy, and use of background DNA. We found that the reverse transcription reaction is the most uncertain step in mRNA quantification and that reverse transcription yield varies more than 100-fold between different set-ups. The first application was an assay for B-cell monoclonality, which can be used for lymphoma diagnostics. B-cells produce immunoglobulins with a heavy chain and either a  or  light chain. In healthy individuals, 60% of the B-cells produce  chains. Lymphomas, like all malignant tumors, are clonal and arise from one transformed cell. Thus, in lymphoma tissue, where tumor cells dominate, the  :  ratio is different from 60 : 40. By comparing the expression of the  and  light chain genes, we could assess sample clonality as indicator of lymphoma. Complex samples, like tumor biopsies, usually contain inhibitors that may interfere with the PCR and cause false sample classification. To account for the presence of contaminants, we developed a method to test the quality of each individual sample. Gene expression measurements are usually performed on large cell populations, and based on the expression levels conclusions are made about the individual cells. We collected single pancreatic -cells and measured the transcript levels of some key target genes. Interestingly, our data revealed that the transcript levels of different genes are lognormally distributed among individual cells, and not Gaussian distributed, as expected. We also applied gene expression profiling to identify downstream target genes of N-CAM and FUS-DDIT3 in tumor development, and to monitor short-term as well as long-term effects of glucose stimuli in Saccharomyces cerevisae.