Change in Superscript lot influences transcript level of one gene but not others - (Jan/18/2013 )

We've recently switched lots of Superscript III reagents and have observed a ~2.5 fold decrease in the Ct value of one specific transcript. We have a big stock of two different dilutions of RNA from a cell line which expresses an oncogene diluted into RNA from a cell line which does not. This stock is aliquoted into single use tubes and stored at -70 °C. After switching lots from Superscript III, we observe a 2.5 Ct drop in the new cDNA preps compared to the old, but only for this specific oncogene, not for the control gene. And this happens in both the high and low dilutions. For other genes in other RNA samples, we do not observe any difference. An accurate quantification is extremely important as we are working with patient samples and now our internal control samples are suddenly giving a ~5 fold increase in expression, so it's dangerous to compare values from old samples from our patients with their new samples. Does anybody have experience with something like this, a transcript-specific change in reverse transcription efficiency? I can understand it's possible due to secondary structures of the RNA, but I don't remember reading something like this (or more important, how we can solve this).

Do you have chance to compare old and new batch of Superscript side by side (i.e one reaction with old, one with new in the same run)?
If not, there is possibility the transcript selectively degraded (that can happen) before the switch.

You can always try a third RT system (side by side) to check if there is variability among different RTs.

Hi Trof, thanks for the suggestion. Unfortunately, the old lot is finished (also at the company) so we can't do side by side comparison. The fact that we aliqout the RNA in single use tubes and that we see an increase in copy number for this oncogene in the two different RNA dilutions, makes it hard for me to believe that it's due to selective degradation. If it would only have happened in one of the two RNA dilutions, degradation would also been my guess, although I would then expect that the copy number would decrease instead of increase. Theoretically, partial degradation may remove some secondary structures that are inhibiting cDNA synthesis, so one could speculate on a mechanism like this resulting in a lower Ct and higher copy numbers.

I think the only option will be to order a new lot of this kit and hope this is working more similar to the old lot than the one we have now. I'm working in a diagnostics lab and it's not so easy to change protocols, it requires extensive validation. One normally hopes/sees that different lot numbers behave similarly, but apparently that's not always the case.