miRNA spike-in control - (Aug/11/2010 )

After reading many papers, I arrived to a conclusion that spike-in control is really necessary on extraction of miRNA in some cases (for ex., from plasma). People say that they use synthetic non-related miRNAs (say, C.elegans in case of human samples) but usually don't give a reference of supplier. So, can anybody give a reference to pre-designed synthetic miRNA for using as a spike-in control?
Another way is to order the synthesis of such miRNA from arbitrary supplier, but unfortunately I am not really sure how to design it to make adequate control. I.e., should I just take the full length of mature sequence? or how many nucleotides? without adding anything on 3'/5'? without complement chain?
I would be grateful for any advice. Thanks in advance.

comp3v on Wed Aug 11 09:58:39 2010 said:

The spike in issue is somewhat controversial to me. What i mean is that doing spik-in is sometimes indicated as a control and even used as a normalizer. I'm not so comfortable with this kind of thinking. in a practical way it must be said that it just gives you a control of extraction but only of the synthetic miRNA extraction. in addition just take into account that it 's not usable for normalization because you add a fixed amount of exo miRNA to your sample and different samples have variable RNA content. in conclusion my opinion is that spike in is just useless. In theory one should also note that plasmatic miRNAs could have different sources (exosome/microparticles/protein bound) while the spiked in is just free RNA. it could mean some differences in yield.
by the way if you want to use a spike in just take the complete mature sequence and and just ask your supplier to make it (RNA i would say but DNA works just the same, personal observation).
BE CAREFUL: once you have a synthetic miRNA in your lab you'll be prone to SEVERE contamination of your samples. you'll want to handle it VERY carefully
bye
Fiz

Hello Fizban,
I agree spike in only gives control of extraction, but in this case what do you recommand for qPCR normalization?
I think some used miR-16 or others, but because it is present in red cells, I'm not sure it is a good one. Its expression could vary notably because of hemolysis, which is difficult to control. Or could we add spike in just before retrotranscription starting from same totalRNA quantity ?
Regards,

Thanks for the answer!

Fizban on Thu Aug 19 10:22:24 2010 said:

Indeed, I was mainly speaking about control of extraction, especially considering that efficiency of extraction from plasma can vary significantly (due to different level of lipids/proteins etc). At the same time, it is known that in many pathologies total microRNA content in plasma can be changed drastically - i.e. it is worth to estimate the amount of microRNA and not only relative expression of some miRNA species. And of course, the question of choosing the right normalization control is still open: I could not find any final conclusion for reference microRNA in case of plasma microRNA.

yes, that was my question - if I can just take mature sequence. At the same time, could you explain more about using DNA? I thought that if I perform normal RNA extraction, I should selectively extract only RNA and not DNA, and thus DNA oligo cannot be correct control of RNA extraction - or I am wrong here?



With best regards,
V.

justme on Fri Aug 20 15:20:31 2010 said:

there's no such thing as a "global normalizer" so no, you can't use miR 16 just because others did use it (some without even explaining why, but this is a whole different story).
My opinion (personal) is that there are only 3 ways: 1) you are able to quantitate you RNA (difficoult when working with plasma, probably feasible with a bioanalyzer 2100) 2) deep sequencing, you have numbers of reads, it is quantitative for definition, but expensive as far as i know (anyone with cheap deep sequencing available out there? ) 3) screening. you measure a lot or all miRNAs in each sample (or a good n) and see which are the most stable. not so subtle and neat but sometimes you have to make it work.

i'm open to new suggestions/ alternative ways/ criticisms
Fiz

comp3v on Mon Aug 23 12:49:29 2010 said:

Yep, sorry. i was just thinking about DNA because when i bought synthetic miRNAs 1st time they sent me DNA miRNAs instead of RNA. they work just the same with RT and qPCR when used as a free substrate but of course you have to make RNA if you are extracting that nucleic acid, of course!!!!

bye Fiz

For human samples, we used U47 small RNA as endogenous control.