The answer is a historical one. The best approach for decades to determine the quality of mRNA was no simply run out a sample on an agarose gel and examine the rRNA bands (which were visible due to high concentration). Almost invariably, when the rRNAs bands looked tight, the subsequent probed (usually by radioactive tracing) mRNA bands were tight on a Northern gel. Likewise, any degradation of rRNA observed meant that the mRNA bands would be degraded. The assumption is that the same RNases target both rRNA and mRNA at the same rate. Of course, once you take additional steps to further purify your mRNA, all bets are off.

You can't make any assumptions on the quality/integrity of enriched mRNA or small RNA - in order to be certain you would have to run your sample again on a Bioanalyzer and check for signs of degradation (or run a gel in case you have huge amounts of RNA, which is usually unlikely).

However, the RIN score is not based on the quality/integrity of the rRNA alone (at least as calculated in the Bioanalyzer software) but rather integrates the information from the whole electropherogram (http://www.agilent.com/about/newsroom/lsca/background/rna_integrity.html), therefore at least some component of the RIN score is derived from the quality of the mRNA.

Basically you have to take a leap of faith, based on the assumption that you are able to process RNA without contaminating the samples. The fact that you have produced good quality total RNA in the first isolation is an indicator that you are doing a good job of working in a relatively RNase-free environment using good reagents.

The answer is a historical one. The best approach for decades to determine the quality of mRNA was no simply run out a sample on an agarose gel and examine the rRNA bands (which were visible due to high concentration). Almost invariably, when the rRNAs bands looked tight, the subsequent probed (usually by radioactive tracing) mRNA bands were tight on a Northern gel. Likewise, any degradation of rRNA observed meant that the mRNA bands would be degraded. The assumption is that the same RNases target both rRNA and mRNA at the same rate. Of course, once you take additional steps to further purify your mRNA, all bets are off.

I don't always trust the RIN score, sometimes I get a bad RIN but it happenes that some of the mRNA are not degraded and vis versa. You can run a qPCR of some of the genes that you know they are expressed in this population of RNA and see how good it's

RNA degradation affects mRNA expression results (not in major cases), therefore, the verification of RNA integrity prior to downstream applications like RT-qPCR and mircroarrays is indispensable.

to minimize the interference of RNA integrity on relative quantification models, you can normalize gene expression by an internal reference gene!!

If you consider that the proportion of the rRNA, accounts for the 90% of the total RNA, and the mRNA accounts barely 1%, then is clear that if the quality of your rRNA is poor due to degradation by RNAses, the quality and probability to have a confident representation of any gene that you has been looking for in your mRNA became uncertain. For that reason the correlation among the quality of rRNA and mRNA is important for the determination of a trusty expressión from your mRNA. Simple; if the 90% (rRNA) of your Total RNA it's starting to lost, the 1% (mRNA) of the Total RNA maybe already doesn't exist.

I will also add to all these nice comments that it is difficult to assess directly the quality of the purified mRNA fraction. The mRNA migrate roughly as a smear and some times few specific bands can be detected for the highly abundant mRNA, but how would you estimate the quality of mRNA from this smear?

Well, I think that you have an a priori problem. The RNA degradation. There's several issues that you should have to consider to obtain a good quality RNA. The sample origin, like the organism and tissue itself, the method that do you use to extract your RNA and the way of preservation of your extracted RNA. If there are, as you say, "abundant RNA" in your samples, you should do dilutions, to better observe rRNA bands on a gel and get a better readings on the method that you performed for quantification (RIN scores, RiboGreen, spectrophotometric readings). If since the beginning you have a smear of your RNA, is an indication that your RNA is being degraded. I add you a gel of good quality RNAs from shrimp (Personal image) and from mouse (Ambion; Life Technologies.)

http://es-mx.invitrogen.com/site/mx/es/home/brands/Product-Brand/rnalater.html

The strong bands are rRNA 18S (stronger) and 28S (less stronger) and tRNAs and probably the 5S rRNA are the smaller bands. The gel was a native gel only to see the preservation of the RNA extraction and for that reason the migration of the rRNA bands it's not the expected.

Hi Hever,

Of course I agree with your comments. But the point I wanted to make is the following: if you purify mRNA for instance by depleting rRNA with ribozero kit, how would you assess quality of the resulting mRNA? To my point of view this is not trivial (QPCR is possible but most people will not do that at this stage) and actually most people assess mRNA quality a priori from total RNA quality assuming that no degradation occurs during the rRNA depletion step. I do not say that this is the right way to do it. Simply assessing mRNA quality is indeed more problematic. When possible I check mRNA quality and want to see on the electrophoregram a big nice smear with a max near 1-4kb. If I see a shift towards smaller size (< 500bp-1kb) I assume degradation occurs. Examples http://www.chem-agilent.com/pdf/5988-3001EN.pdf

But coming back to the subject of this post, this is not quantitative and very often one have not even enough mRNA to do such QC, as pointed several time in this thread. So yes many times one do not have choice and simply check total RNA quality and not mRNA quality.

I think you already have the answer.. rRNA integration just gives an idea of mRNA integration. rRNA is high in quantity and hence visible on the gel but mRNA being low in quantity is not visible..

One of the best methods is to normalize using geNorm. A normalization validated with few genes (V2 / 3

All of the above suggestions are good and true considerations. They point out that there are many assumptions that are made when isolating nucleic acid from samples. There are various efficiencies of lysis, the purification of RNA, degradation (specific mRNAs degrade more easily than others), reverse transcription method, and PCR efficiency. It then becomes important to treat all samples the same, assuming they have the same efficiencies of all the above steps.

One thing to track the yields and efficiency is to have an endogenous control gene or RNA you detect for consistent results (as noted above). Another possibility is to spike a known amount of an RNA or even DNA into your samples, either at the tissue lysis step and/or later at a step you choose. This is the best way I know to track what happens to your samples, though it takes effort to set up.

Total RNA quality as indicated by the RIN value >7 is just an estimate that RNA isolation went well / RNA is intact and not degraded. THis also assumes the fact that if the rRNA is not degraded, the mRNA might not be . But the fact is that mRNA is more labile when compared to rRNA. Still, we have only total RNA estimate prior to mRNA enrichment for assessing the quality.

Total RNA (which actually is evaluation of relative abundance of rRNA) is used for quality assessment because it's quick and easy. Historically established, and with some degree of correlation between this rRNA profile and mRNA degradation.

But it is by no means a good indicator of mRNA quality. However, given the very low abundance of total mRNA, and given the diversity of mRNA species, it is quite difficult to establish a common technique to specifically measure mRNA integrity.

Given that degradation is not only an indicator of "improper" bench-work, but also a tightly regulated and important gene expression modulation mechanism, each mRNA species might have a specific degradation dynamics.

Alternatives to rRNA abundance profile have been sought, but have not reached a level of general use and acceptance. So, for the time being, rRNA abundance profiles remains the golden standard.

correlating the quality of Total RNA with the quality of the mRNA?

Since rRNA species are significantly more abundant than mRNA, the probability of degradation of the rRNAs is significantly higher than degradation of the mRNAs. Additionally, when one thinks of targets for degradation, the probability of degrading the higher molecular weight RNAs is higher than the probability of degrading smaller molecular weight RNAs. This is clearly seen when RNA integrity is analyzed by gel or capillary electrophoresis. The 28S rRNA band is reduced before the 18S rRNA band begins to show any visible sign of degradation. Once the 28S rRNA band starts to degrade, the integrity of the HMW mRNA species would have to be considered as possibly degraded as well (Santiago et al 1986).

As Sebastian stated "The fact that you have produced good quality total RNA in the first isolation is an indicator that you are doing a good job of working in a relatively RNase-free environment using good reagents."

Therefore unless you introduce RNases in your downstream reactions, your mRNA should be intact.

If you are comfortable with the results to the end of total isolation, then you could spike your sample at this point. If you are uncomfortable from the very start, then you could spike your samples from the very start of isolation as already suggested.

Since your mRNA is very precious and you may have very little, I would recommend that you spike with a known quantity of cRNA of about 5 Kb (if interested in all size ranges). Many of the ion channel and 7TMR G-protein coupled receptor cDNAs are available under the control of T7 or T3 phage promoter. This makes it a simple in vitro transcription with storage of the cRNA for inclusion in a number of isolation procedures.

You could then examine your sample as Alice and Elizabeth have stated by qRT-PCR.

Check out this article by Fleige and Pfaffl (2006). It cites some of the primary refererences on RNA size and degradation.

http://www.gene-quantification.de/fleige-pfaffl-rin-2006.pdf

It's like a relative qualification of your sample. As said by many, if your rRNA bands remain, your mRNA should be intact as well. In gene expression experiment, rRNA has been used a a reference gene because of its abundant. Therefore, by running an agarose gel with formaldehyde, the rRNA bands should be clearly seen and could indicate the status of your mRNA in the sample.

As Gary says its historical . Buts its also fair to say its effective and so persists. I was once advised by a guy who ran a microarray facility (so had handled all different kinds of samples with varying quality) that a good RIN invariably gave good results. A low RIN sample may also give nice results but not every time.