I am wondering what is the best method for normalization of microRNA in RT-qPCR experiments. My samples are plasma samples so can I use a reference gene for normalization? And what is the most stable miRNA in the plasma in your opinion to be used?
The answer is that depends.... If you are using profiling methods in which you analyze a lot of miRNAs at the same time (more than 50), the best method to normalize is to use the global normalization. You can use several free software for doing that. I personally like DataAssist from Life:
If you are looking to a few miRNAs you can use preferentially the spike-in method adding a synthetic unrelated small RNA to your RNA sample before the extraction. I personally do not like to use internal miRNAs for normalization because they tend to vary a lot in circulating fluids. Some people like to use miR-16 to normalize, but in some situations it changes a lot among samples.
Like Paco says, if you are studying few microRNAs spike in controls work fine. Although, you still find some variability. We experienced that miR-16 abundance can negatively correlated with inflammatory markers. But to determine it, still helps us to search for biomarkers.