SACRAMENTO, Calif. — The Association of Biomolecular Resource Facilities' Nucleic Acid Research Group this week presented the results of its 2009-2010 study, which assessed the impact of various priming strategies on generating cDNA for use with quantitative PCR.
In the study, three different RNA templates of varying RNA quality were sent to 20 different labs, which performed the reverse transcriptase using the primers and enzymes of their choice and then sent the cDNA plates to a single lab for qPCR processing on a Life Technologies AB 7900 HT.
Although the labs used a total of six different RT enzymes at a range of temperatures, there were very few differences across the groups in terms of the effect on Cq value, according to Katia Sol-Church of the A.I. DuPont Hospital for Children, who presented the results of the NARG study at ABRF's annual conference here.
"We were impressed" that all 20 labs produced essentially the same results, with very few differences, Sol-Church said, noting the Cq value ranged from 24 to 27 for the 20 labs. "We didn't expect that."
One reason NARG members did not expect such reproducible results was because the effort built upon a preliminary, "in-house" study among seven labs conducted last year that resulted in Cq values that were "all over the place," Sol-Church said. In comparison to the 20-lab study, the Cq range for the smaller study was between 21 and 29, and spanned five levels of "statistically significant difference," she said.
The key difference between the two studies, she noted, was that each lab in the preliminary effort did its own qPCR, while that step was centralized in the larger study. That fact led the group to conclude that any differences that were seen in the first study were due to the qPCR step, not the RT priming step.
Beyond reproducibility, the 20-lab study reached essentially the same conclusion regarding specific primer strategies as the preliminary study did — if the cDNA is only going to be used for one gene, then a gene-specific primer is the best option. If a gene-specific primer isn't available, however, a randomer-oligo(dT) combination is the best bet. In that case, the closer the assay is to the 3' end, the better the results will be, NARG members said.
In terms of specific enzymes, Clontech's SmartScribe appeared to perform the best, with a Cq of 23.9, while Bio-Rad's iScript was the worst, with a Cq of 26.77. However, Sol-Church stressed that only one participating lab used each of those enzymes so it would be difficult to draw any broad conclusions from the results. By comparison, 10 labs used Applied Biosystems' High Capacity RT kit; four used Invitrogen's SuperScript III; while two labs each used AB's standard RT kit and Roche's Transcriptor.
The choice of temperature made no measurable difference in the results, Sol-Church said.
The study also set out to determine the effect of RNA quality on Cq. RNA was degraded to three different levels — an RNA integrity number of 8, a RIN of 4, and a RIN of 2. However, due to various delays the labs were only able to analyze the samples with RIN of 8 and 4.
Preliminary results with those two samples indicate that the mildly degraded RNA may not have as much of a negative impact as some may have though. The Cq for the RIN 8 sample was 26, while that of the RIN 4 sample was 27, which is "not so bad" and indicates that while mildly degraded RNA does impact results, it could still be useful for certain applications, according to Sol-Church.
NARG plans to present the full results with the RIN 2 sample at next year's ABRF meeting.