As evidence mounts supporting the role that microRNAs play in a variety of biological processes, researchers are looking for new ways to detect and quantify the short, non-coding, single-stranded RNAs.
In line with this trend, a research team at Ben-Gurion University of the Negev in Israel has been looking into the role of miRNAs in viral infection and pathogenesis. However, they found that current methods for detecting miRNAs by real-time quantitative PCR were lacking for their purposes.
For example, one method, stem-loop qPCR, uses stem-loop primers to reverse-transcribe the miRNA, which is followed by hydrolysis probe qPCR, but this approach requires a unique primer and probe for each miRNA assay, which makes it unsuitable for screening large numbers of miRNAs.
Another method extends the 3' end of the RNA through polyadenlyation, followed by reverse-transcription and amplification. However, this approach uses SYBR Green to detect the PCR products, which can lead to false-positive results.
In a paper currently in press with the Journal of Virological Methods, the researchers describe a new assay they developed that combines the polyadenylation approach with a cDNA library that uses a reverse-transcription primer containing poly(T), a universal probe library sequence, and a RACE (rapid ampliﬁcation of complementary DNA ends) adaptor.
The researchers note in the paper that the assay, which they dubbed UPLpolyA, is sensitive, specific, easy to prepare, and lower in cost than the other two methods. "It utilizes RT, reverse primer, and probes that are all generic; therefore it is highly versatile and cost effective, and can be used for high-throughput expression proﬁling of miRNAs," they wrote.
The team used the assay to detect Epstein-Barr virus-encoded microRNAs in EBV-infected cells from various sources and was able to demonstrate differential expression of several miRNAs in patients with infectious mononucleosis versus patients with latent infection. For example, the miRNAs BART-2, -4, and -6 "are clearly expressed in healthy carriers while BART7 is expressed in IM only," they wrote.
PCR Insider recently spoke to lead author Yonat Shemer Avni about the assay and the study's findings. The following is an edited version of the transcript.
Why are you looking at quantifying miRNAs during viral infections? What are the benefits of measuring miRNA as opposed to mRNA or DNA?
For viruses, like the DNA viruses of the herpes family, what people have showed — and this was known before microRNAs were discovered — was that there were certain RNA transcripts that exist, and people were all the time trying to look for open reading frames, putting meaning into [these transcripts] during latency. And for a long time, we tried and we failed [to discover their function]. So eventually it was discovered that these transcripts that are expressed during latency are a precursor of microRNAs.
So now the next step is to try and find out what the microRNA is doing during the latency period of the virus — how does it contribute to the suppression of the viral replication, which is a hallmark of the herpesviridae family.
This is typical for the Epstein-Barr virus, which we described in our paper; and for herpes I and herpes II. For them, they are situated in the latent state in the neurons and the ganglia. There it was known that there was a certain transcript, called latency-associated transcript. This transcript is in fact a precursor for microRNA.
MicroRNAs usually regulate transcription expression of genes, so people are looking into what kinds of genes are suppressed by this virally encoded microRNA, and through this they can tell now which genes are important for reactivation.
So specifically for Epstein-Barr virus, the microRNAs that we tested, the ones where we see differential expression come from a region that was really well known, first, during latency as the RNA, and now it is known that is was actually a precursor for microRNA.
[ pagebreak ]
There are about 20 different microRNAs that are expressed in this region, and there is a lot of alternative splicing of RNA there. So people are looking, for example, [at miRNAs] like the one that we mentioned, BART2, which is in complete antisense to the messenger RNA of the polymerase of the virus, and that's why it's actually suppressing the activity of the translation of the viral DNA polymerase. It is probably suppressing the viral replication during latency.
The [question] is how does the virus get in a latent state? Because … if there is a bad infection of EBV, one out a million lymphocytes will have the genome of the virus. And most of the work that the virus is doing is just to keep itself under the radar of the immune system. So it's like a balance [in] the cells that have the virus: the virus is silencing itself, so proteins of the virus will not be expressed in the cells, and the cells will not be recognized by the immune system.
And if you take out these cells from the body and you don't have the immune system as a player in this game … you observe cell transformation. Because the virus is not under the suppression of the immune system so it can express different genes, and these genes are participating in driving the cell cycle.
So we think — and we are not the only ones of course — that the expression of microRNA during the latency stage is actually a way to suppress the virus.
So it sounds like you're looking at this more as a way to develop better therapeutics or possibly vaccines, as opposed to something on the diagnostics side.
That is also in the back of our minds, but in order to be able to say anything, you first have to analyze what is really the contribution of the microRNA to the latent stage: which microRNAs are expressed during latency, which microRNAs are expressed — if any at all are expressed — during acute infection, when the virus is not latent and you have replication of the virus.
So actually this paper describes the first step. What we are doing now is looking at a lot of people who are healthy carriers of EBV to see what the situation is there.
So the assay was developed as a way to do that quantification at those different stages?
Yes, but we didn't really get so much to the point of quantification because then you have to use synthetic probes and things like that. You can develop it into a really accurate assay if you use the right controls, and if you use standards.
But what we saw with patients versus people who are harboring the disease in the latent stage was that you really have a different population of microRNA. So we didn't really have to do so much, at least at this point, with quantification.
It's one of those situations where it's all or none. So BART7 was expressed only in infectious mononucleosis. We couldn't find it in healthy carriers. So if a pattern like this exists, you don't care so much about [whether] it's two times more or two times less, because it's all or none.
What are benefits of using qPCR for miRNA analysis as compared to, say, microRNA microarrays?
First of all, it allowed us to use only a small amount of RNA that we couldn't have used with a microarray, for which you need much more RNA. If you consider that in healthy carriers, one out of a million cells carries the disease genome, so if you take 10 milliliters from a person, it means you have 107 lymphocytes when you start the experiment, which means that you might have 10 or 20 cells that harbor the virus. So you might not be able to detect that with a microarray.
We also did a lot of microarray analysis because we collaborate with Autogenomics. But what we did there was the analysis of cell lines. We couldn't use it for patients.
We have the advantage that we are a clinical virology lab, so we have access to patients and we can get real samples. Most of the analyses for viral microRNA have been done in tissue culture or mouse models.
In terms of the assay itself, can you walk me through the UPLpolyA method you developed? How does it differ from other qPCR based methods for miRNA analysis?
For stem-loop UPL, you're using a long primer that can fold on itself, and at its 3' [end] the primer matches the microRNA, the specific microRNA that you want to detect. And after you get the cDNA, you [use] the forward primer, which is complementary to part of the microRNA, [and] you [use] the reverse primer, and a [universal probe library], which is sitting on your loop. And then you do the extension, and the UPL is actually a hydrolysis probe. You have a quencher, and a UPL, and the polymerase is exciting the probe, and it is released and you have fluorescence. The more PCR cycles you have, the more fluorescence you have.
[ pagebreak ]
So actually you have a set of three different primers that are specific for the microRNA, because the reverse primer with the loop has to be partially specific and the forward primer is also specific, and the stem loop primer is very long. It costs much more to synthesize it and you usually have to use HPLC purification so it is more costly.
Other people use the stem loop technique without UPL, but they put a small minimal groove primer that is sitting half on the microRNA and half on the synthetic reverse primer, the stem loop primer. Because one of the problems with microRNAs is that they are so tiny that you need a handle to fish them [out] somehow. So people use the stem-loop format in order to … have a long primer on the 3' end that you can use to catch it.
Other people use the polyA technique, where you just come in with the polyadenylation enzyme, which is very efficient, and you're just polyadenylating any RNA that you have in your preparation. So you've extended the 3' part of the microRNA, which is very tiny, and then you have to find ways to amplify it.
So first of all you do cDNA — usually you do it with a RACE primer, some kind of general primer — and you also put 12 Ts in the 5' end of your primer, a repetition of T that will complement the A. For example, one of our RT primers, at the 3' end it has a V [A, G, C] and an N [A, T, G, C]. If you have poly(T), this polyadenylation can be as long as 300 nucleotides, and you don't want to start the reverse transcription 300 nucleotides away from your microRNA, so that's why you use this V and N, and then these 12 Ts in order to put your reverse primer exactly at the end of the microRNA when you do the reverse priming.
So after the polyadenylation you have the microRNA and then the poly(A), and then you [use] a universal adaptor. We added to it this UPL binding region, and then you have the 12 Ts and the two nucleotides that will be sitting at the 3' end of the microRNA, and then you [perform] the reverse transcription and you get your cDNA library.
Then with this library you can look for any microRNA that you want by just changing the forward primer. The forward primer actually represents the microRNA that you want to find.
You can also use this library for messenger RNA detection. It's not specific for microRNAs. It's just the primer that you're using that's for microRNA. And messenger RNAs, usually they're already polyadenylated so they don't need this technique.
So we also used the SYBR Green polyA method, which is done in exactly the same way, only you don't have the UPL binding site. So when you finish amplification you have double-stranded DNA of a certain size, and you have to run it on a gel in order to be sure that it's the right size, because [using] no template control can give rise to all kind of [non-specific PCR] products.
Is there anything that you would need to do to improve this assay?
The problem that we are facing now is that we have predictions for microRNAs that the bioinformatics people at Autogenomics were doing, and then we confirmed some of these predictions with microRNA microarrays that we did together with Autogenomics, but we don't know the exact sequence of these microRNAs.
So with our method — actually for any good real-time PCR — you need to know the exact sequence of your microRNA. So if you have predictions and you have results from a microarray, one of disadvantages of microarrays is that it doesn't give you the exact sequence. You don't know exactly where the dicer was cutting the microRNA. It can be three nucleotides to the right or to the left, something like that.
[ pagebreak ]
And when we don't know the exact cutting point, we can run into problems with the real-time PCR, because then the primer for the microRNA is not accurate and it can give us results that we can't trust.
But when we know the exact sequence, it goes fine. So for microRNAs that have been published and sequenced and so on, it works fine.
So that wouldn't be something that we could address by improving the method. We would have to improve our knowledge.
That the best way is [actually to do] what the people at Autogenomics are doing, but for this method you also need to know the exact sequence. What they do is they also do polyadenylation, and then they do reverse transcription, but they leave part of the microRNA and the RACE primer undetermined by the forward and reverse primer, and they have a specific hydrolysis probe for each microRNA. It's specific then, because it is sitting half on the microRNA and half on the general reverse primer.
This is the best way, you can't argue with it. But then you have to design specific hydrolysis probes, which is rather expensive, for each microRNA that you want to detect. Each probe like this costs at least $300.
So our method allows a lot of versatility, because to order a forward primer, which is $10, is a huge difference. And the UPL you buy once and it's enough for a lot of reactions.
So of course there are better methods. That’s clear. If someone knows that he will always work on the same five microRNAs and that is it, maybe it would be better to order a specific probe and work with that. But if somebody wants to screen lots of microRNA and only afterwards concentrate on certain types, then it's better to use our method and then move on.
So the key to the lower costs is that you only have to design the one primer as opposed to three primers for each microRNA?
Yeah, and with the UPL, instead of buying a hydrolysis probe for each reaction, you buy the UPL for all the reactions. It's completely generic.
How are you using this assay in practice currently?
With this method, we are moving forward with patient samples, and even more interesting people who are just carriers. We are looking into it for the EBV project, and for the rest of the time we are really doing research into the role of microRNA in viral infection.
We're also working on a finding that we discovered together with Autogenomics. It's a newly identified microRNA that was not published before, and we are trying to identify the target, and what is its role in this particular virus's cycle.
This method is not our major interest. It came out of our needs. We are more interested in identifying which microRNAs are expressed at certain stages of infection, and what are their targets.