The design of primers for microRNA RT-qPCR is challenging as the average microRNA is only 22 nucleotides long, which is the same length as a traditional PCR primer. However there are methods to overcome these limitations and make primer designing. This excellent paper below may be helpful in this context.

As alternative approach, you may use the Life technology website and order a premade and validated miRNA Taqman probe.

http://www.biomedcentral.com/1471-2105/15/29

Thanks for your answer. I am perfectly aware of the techniques and this paper in particular. However, I tried to get others (your) experience with qPCR experiments that failed because of inappropriate primer design that was due to an incorrect microRNA annotation (too short).

Thanks

When we design qPCR primers for miRNA experiments we used a synthetic miRNA to crossvalidate the prmer design. You can add the synthetic miRNA at three log concentration (10, 1 and 0.1 nM) and see if your probe will detect. If the design fails we simply design another set of probes. For the design of synthetic miRNA you may check in the reference below. If you are interested to a mismatch of the sequence you may incorporate this in the synthetic miRNA and check if your primer will detect correctly. I do not know what do you mean to an incorrect annotation. We check the sequence of the miRNA using miRNA-seq before to perform primer designing.

http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0114750

Thank you very much. 

As for the incorrect annotatioan I am talking about cases were smallRNA seq is not available directly and annotations (and microRNA lengths) are simply taken from miRBase. Those might be incorrect in some cases...

There is no way to design primers if you do not know the sequence of the target. You have to sequence the target first and design the primers next. If you are working on a species where no public data are available, you have to sequence the small RNA first and identify the specific sequence of your target. However, if you did not follow such golden rule and you have the results and now the sequence of your reference changed, you may use two synthetic miRNAs (the old and the new sequence) to see which is the impact on the results. Most likely the impact will be relevant. This is the best advice I can give you. If you are working in humans there are tens of public projects where you can check the actual sequences and see which is the right reference.  Good luck!

Thanks for your answer!

I address the fact that many companies offer qPCR plate systems that have prepipetted primers for "known" microRNAs that are based on miRBase annotations. I know for some cases that miRBase annotations are erroneous and have microRNA sequences that are too short in comparison to what I see in our data. The question therefore is: what happens if the primerdesign is based on such a too short target sequence.  

Oh and I am working on human.

There are many examples of commercial products carrying old references. Our work is difficult and if for the purpose of a fast screening a pre-designed plate can be fine. Then I would recommend to validate the results in a very robust way. Using library of miRNA-seq is now possible to double check every reference in human and check the actual sequence even in miR-Base. There are hundreds of miRNAs poor investigated and you do not have to take the Reference as "perfect". An emerging field  is to look at the SNP in miRNAs which looks very promising in some diseases and whose quantification via qPCR requires specific primer design. As every SNP, in some populations given miRNAs sequence may be frequently different from those present in miRBASE. In the case of shorter references used in a plate, it may be possible that they are not specific and crosshybridize with other miRNAs . This is also the case of some probes in the microarray chips. When you get the results from a screening you need always a crossvalidation step, whose primer design has been carefully validated.

Dear Bastian,

Depending on the actual shortening of the sequence (how many nucleotides ,shortening  5' or 3')  and depending on the PCR system you are using you may get an amplification with low efficiency or no amplification at all.However you will still get amplification of the target the PCR is design for. If you know that the standard primers are not detecting the main dominant fraction of your miRNA of interest I'd suggest to test the standard primers against some custom primers designed for the sequence you expect to be dominant in your sample. Ideally test both primersets with their corresponding synthetic miRNA and the other target sequence to get an idea in how far the primers detect the RNA they are not intended to detect.  

"The design of primers for microRNA RT-qPCR is challenging as the average microRNA is only 22 nucleotides long"

You must be referring to mature miRNA sequence, with pre-miRNA sequence you can get >70 bps which is a reasonable size to detect (and I have detected several low copy miRs this way such as miR-34a without the use of Taqman or LNA probes)

Yes indeed. I am talking about mature microRNAs here and the method I refer to is RT-qPCR.

Hi Dr. Bastian Fromm

i suggest you these two good papers:-

1- A tool for design of primers for microRNA-specific quantitative RT-qPCR

Article A tool for design of primers for microRNA-specific quantitat...

2- MicroRNAs Specific Primer Design using miRNA Design Tool

http://www.isca.in/IJBS/Archive/v6/i8/6.ISCA-IRJBS-2017-061.pdf

Thank you for your answer but that was not my question.

The point is that many miRBase microRNAs are incorrectly annotated (too short) and thus this affects qPCR negatively.

miRBase: annotating high confidence microRNAs using deep sequencing data

Article RBase: annotating high confidence microRNAs using deep seque...