I am having some trouble with RNU6B as endogenous control for several ovarian cancer cell lines. Could you recommend me other stable reference miRNAs that have worked in your experimental approaches?
Thank you so much for all your ideas and suggestions. Now I have a couple of candidates to think about or going back to my data files to check the most stable one!!!!!
The reference gene(miRNA in your case) is usually chosen based solely on the fact that it is not affected by or doesn't show variability in its expression pattern under the defined experimental conditions. If you are using mammalian cell lines thee expression of several RNAs/miRNAs remain stable for a long time but it is always advised that you take an miRNA or another small RNA of comparable length ( due to various effects on PCR efficiency of the probes used). If possible it would always be good to add an external "spike in" control (an artificial RNA externally added and later quantified as the reference RNA/miRNA) as it is most likely not to get affected by your experimental conditions. There are now standard spike in controls commercially available being used widely.
In case you would still like to go for an endogenous miRNA make sure you consult literature for what others have used and check technical specifications from companies like ABI (now Life technologies) as you would find useful information there. But make sure you verify that there is no effect on your reference gene by the experimental condition you provide..
Thank you again for all your comments. Next week I will have some results from the new control RNU48 in comparisson with RNU6B. I will let you know....Enjoy your experiments in this friday!!!
Dear Francesca, RNU48 is dysregulated in cancer and should not be used for normalization of qPCR data. You can check this paper: Gee et al (2011) The small-nucleolar RNAs commonly used for microRNA
normalisation correlate with tumour pathology and prognosis, British Journal of Cancer (2011) 104, 1168 – 1177.
I am happy to let you know that I have the results comparing RNU6B and RNU48 as endogenous controls with the same experimental groups from several human cancer cell lines under different conditions. The best results with less dispersion are those associated with RNU48 as endogenous control. Actually RNU48 appears in much earlier CTs (around 15) thanRNU6b.
In the cell lines I have tested, the expression pattern of RNU48 is not affected and doesn't show variability under the defined experimental conditions.
I am a bit lost trying to normalize microRNA qPCR results and I just found your conversation... Do you expect your reference microRNAs have a stable Ct across all the samples? I was intended to use miR-191 as a normalizer in plasma, but I have great differences in Cts in my samples (from 32 to 40). My study is too small to use the global mean normalization approach. What do you suggest with your experience??
Unfortunately these things happens specially with plasma miRNAs. Try to use an internal spike-in control by adding a totally unrelated miRNA, and use it as internal control. There are several of those spike-in RNAs commercialized by Exiqon. This should circumvent your variability problems.
Thanks for your comments guys. I was using RNU48 for glioblastoma cell lines but it failed to amplify in some samples. But i tested the same samples against RNU44, RNU48 and RNU6b and the same was observed for 44 and 48. Now i am using RNU6b. seems to work for me? Anyone else used in brain tumors?