to BioTechniques free email alert service to receive content updates.
Kristie Nybo, Ph. D.
BioTechniques, Vol. 54, No. 4, April 2013, pp. 189–190
Full Text (PDF)

This month's question from the Molecular Biology Forums (online at comes from the “RNA Methods” section. Entries have been edited for concision and clarity. Mentions of specific products and manufacturers have been retained from the original posts, but do not represent endorsements by, or the opinions of, BioTechniques.

Molecular Biology Techniques Q&A

How can I troubleshoot a failed 5’ RACE? (Thread 32143)

Q I am trying to do 5’ RACE for plant genes using a Clontech kit. For one gene, the procedure worked well, but I'm having some problems with the other. I synthesized a new cDNA in case the gene was expressed at very low levels, but I still don't see any bands on the gel. What can I do to improve my results?

A Since the kit worked for one gene, you should not have any technical problems. Is it possible that the one that failed was already a full-length transcript? mRNA degradation might also cause 5’ RACE to fail. Try repeating the procedure with freshly prepared RNA.

A Can you PCR amplify this gene from your cDNA template? It is possible that there is a secondary structure or a high GC stretch that is difficult to amplify by PCR. Try raising or lowering the annealing temperature or use a cosolvent like 3-10% DMSO or betaine.

Q I tried using three different sources of RNA and altering the annealing temperatures. I was able to get multiple bands on the gel, so I will try different annealing temperatures now to optimize the conditions.

Does anyone have experience with Primescript Reverse Transcriptase from Takara Bio? It's an MMLV alternative to the RT provided by Clontech. I used it for my cDNA synthesis and wonder if it could be contributing to the problem.

I can't amplify the gene since I have only a partial EST. Genomics resources for trees are scarce, so I don't have much information. I'm trying to complete the EST by 5’ RACE. The GC content of the EST is fine (40-50%), so I don't expect that to cause any problems.

A The Clontech 5’ RACE kit incudes the SMARTScribe Reverse Transcriptase. Why didn't you use that? A variant may behave differently and alter your results.

Q At the time I purchased the kit, the SMARTScribe RT was not provided.

A Can you get a PCR product based on the original EST with one of the RACE primers? You might be able to use that as a control to find out if you have enough template. If a RACE primer can create an amplicon, that will show that at least one primer works under the given PCR conditions.

Multiple bands are not necessarily a bad result. Some genes have alternate start sites or alternate splices that would create more than one 5’ length.

Even if the percentage of GC in the EST is favorable, that doesn't give any indication of the GC content of other regions. 5’ ends sometimes have a higher GC content than the rest of the sequence. It is sometimes possible to pull data from multiple bands even if some are not specific. The bands can be cut out of a gel, purified, cloned, and sequenced. In some cases it works to cut out a band and run the PCR again on that product, although you won't want to use too many cycles because you may end up with multiple bands again. Then you can sequence the clean PCR product at the end.

Q I am strictly following the protocol provided in the manual. I now repeatedly see multiple bands, but the desired band is too faint to be extracted properly. When I repeat the PCR, the band pattern changes, with smaller sized bands becoming more intense. In the end, I just find very little of the expected product.

A How did you determine the size of band to expect?

It is likely that your procedure includes a nested PCR step on the side with known sequence. But to improve extremely faint bands, you might try an additional nested step with a primer inside the last 3’ primer used. If the multiple bands are specific products resulting from alternate transcripts, your results will not change. But if not, the additional nest step should increase specificity and eliminate nonspecific bands.

Alternatively, you might try to cut out the correct band and do another PCR amplification. You might even try PCR without gel purification if there is not enough product to purify. If you need to use that approach, it would probably work best if you use a low melting agarose gel or biotechnology grade agarose designed not to inhibit enzymes. PCR will also work with a small percentage of regular agarose in the mix as well. Taq is relatively tolerant of impurities, but I'm not sure how sensitive other proofreading enzymes are to agarose contamination. Be sure not to use too many cycles because you will amplify traces of other bands as well, potentially resulting in a smear of products in the gel. Even if the result is somewhat messy, at least you will have increased the amount of product. Then if you clone, you can check for the correct product.

It's also possible that the faint band you are looking at is actually nonspecific also.

Even when following the protocol perfectly, the results will not be good if the starting template was weak or degraded. Try conventional PCR on the EST sequence alone as a control to see if the template can be readily amplified by PCR and that it contains the target message

Q I compared the size of my EST with sequences from other organisms to approximate the size of band I expected to see.

I plan to extract the band and look at it, but it was very faint and was loaded on a regular 0.8% gel, so I'm not sure that I will have a useful outcome. I can't do nested PCR at this point since I loaded all of my product on the gel, which was the only way I was able to see the band.

Since that time, I repeated my reactions and only see non-specific bands or no bands at all.

I think my cDNA is fine since I was able to amplify two genes completely from the same cDNA.

I called the manufacturer, but it didn't help much since I'm using an older version of the kit, which has now been replaced.

A I would guess that the ends of the gene are more likely to vary from other species than the center of the transcript because alternate splices are most prevalent near the termini and the untranslated region often varies in sequence more than the center. Why don't you try blotting the gel? Then you can use a cold probe to see which bands are specific products.

You might also check the bands with a quick PCR if you have internal primers in the known area.

Another possible cause of the problem would be that the RT stops at regions of high GC content. If that is the case, a higher temperature in the RT step might help.