Confusion about proper endogenous control for microRNAs - why miR-16? so confused... (Aug/27/2008 )
I saw this new paper in BMC Molecular Biology: "Identification of suitable endogenous control genes for microRNA gene expression analysis in human breast cancer". They explored the use of different endogenous controls in quantifying expression of microRNAs and identified concluded that miR-16 and let-7a are best used for this purpose (better than U6, and others). But so many groups including the one reporting in this paper in cancer research: "MicroRNA Gene Expression Deregulation in Human Breast Cancer" have shown that miR-16 is highly variable between normal and cancerous breast tissue. Similarly let-7a is a known onco-miR! So what the hell is going on here?? Doesn't it make more sense to use snoRNAs and rRNAs such - analogous to using cytoskeletal genes for normal gene controls?? That almost sounds like using BRCA as a control when studying breast cancer. Or how about p53 - nice stable gene.
I'm so confused 
I agree, it doesn't make sense to pick a potential tumor suppressor (miR-16) as your housekeeping if you're looking at cancer samples. Other samples, well maybe it could be ok. But to be honest I'm not entirely convinced by their data. having variations from 12-98% miRNA of total sample is a "rather large" variation and it is not really confined to a particular sample type. Also looking at fig 4A, the samples look really different on the "gel." Anyway, they didn't even look at Rnu6b, which I use and am happy with the results it gives me. I guess I'll stick with Rnu6b until more data comes along!
Yup - I use RNU48 and it's been perfect so far! Thanks for the reply!
Yup - I use RNU48 and it's been perfect so far! Thanks for the reply!
I suppose that there's no such thing as a UNIVERSAL normalizer. Mir-16 is very good in Huvecs in hypoxic conditions, but i agree with you when speaking about breast cancers. In my opinion every condition has its good normalizer. Mir-16 is good in many conditions, I've no experience with RNUs being good normalizers, but again, it depends on what you're studying. Consistent statistical validation is the FIRST thing to do when starting a new screening with miRNAs.
My PERSONAL opinion is that finding a "large spectrum" normalizer is very good (and difficoult!), sticking to one (without validation, obviously) could lead to wrong results.
Fizban