The fast-paced, continually evolving field of quantitative PCR (qPCR) will be showcased at a CHI meeting in San Diego next month. Presenters will describe cutting-edge methodologies and emerging technologies such as digital microfluidics, nanopore and single-molecule sequencing, and improved ways to standardize data.
Traditionally, the first step in gene-expression studies is to purify RNA from samples. However, RNA isolation can be a problem for some applications. Gregory L. Shipley, Ph.D., assistant professor and director, Quantitative Genomics Core Laboratory, University of Texas Health Science Center, says that utilizing cell lysates for real-time qPCR is a helpful alternative. “Isolating RNA can be an expensive and time-consuming process. For small amounts of tissue or low numbers of cells, RNA purification isn’t practical.”
Dr. Shipley suggests that a better way is to utilize cell lysates directly for cDNA synthesis. “Lysates not only act as effective templates for amplification, they provide some added advantages. First, one can easily utilize a very small number of cells for real-time qPCR. If you tease out cell populations from a lung tissue sample, there are lots of cell types present. Typically, you might get a specific population of only about 20,000 to 100,000 cells using flow cytometry. It isn’t practical to isolate RNA from such a small number of cells, especially for analyzing multiple transcripts. Secondly, isolation of RNA can potentially skew the population whereas with cell lysates nothing is lost.”
According to Dr. Shipley, there are some drawbacks that must be taken into consideration. “When using cells directly, the benefit is that everything is in there. However, the disadvantage is that everything is in there. It is important to lyse cells in the presence of agents that will inhibit RNA degradation and to have a way to get rid of contaminating genomic DNA.
“Several manufacturers offer kits for making cell lysates for real-time qPCR and more are jumping on the bandwagon. The challenge for the future will be to figure out how to use small amounts of tissue directly for qPCR. Currently, this works less well in our hands.”
Although gene-expression analysis by reverse transcription quantitative PCR (RT-qPCR) provides an accurate and sensitive means to measure gene expression, inherent biological variability can cause problems. “In the process of establishing a reliable qPCR, I found that biological variation between different samples taken at different days or in different animals was causing problems,” explains Erik Willems, Ph.D., research associate at the Neuroscience, Aging and Stem Cell Research Centre of Burnham Institute for Medical Research.
Dr. Willems then began working with an expert in biostatistics, Professor Jo Vandesompele from the Ghent University in Belgium and co-founder of the qPCR data-analysis company Biogazelle. “We came up with several ways to address these interexperimental variations and developed a set of easy corrections that anyone can perform on their own data. This is a simple but very important contribution to the field as people with difficult biological samples often tend to show their ‘best’ graph, rather than showing the average of a larger sample set.”
The standardization procedure involves three steps according to Dr. Willems. “The first step is to use a logarithm transformation. Basically by putting all the data to a log scale, you reduce the effect of outliers because such a scale gives equal weight to all data points.
“The second step is to perform a mean centering. This basically means that we correct for differences in the control levels of a certain gene, so that the untreated conditions are all leveled for their gene expression. Each data point is divided by the average of all data points, a typical approach used in microarray data correction.
“The third step is to perform autoscaling. Here all data points are divided by the experimental standard deviation. This step corrects for differences in magnitude of differential gene expression between a set of biological replicates.”
After tweaking the data, Dr. Willems says the final analysis can be run in any statistical test of choice such as a simple t-test. “This three-step standardization procedure can be easily performed by anyone somewhat knowledgeable in a spreadsheet program such as Excel. It should allow people to show their qPCR data with much more confidence.”