Don't understand why we need RT-PCR? - Slight Confusion guys, please clear me up (Mar/03/2010 )

Hi, I am not asking for any answers to any question like a kid. I just want to understand a bit.

On our assessment for Identification of Viruses by Genome Sequences, we were given the following info:

1.) People are dying of high fever and specimens were cuultured for bacteria,....etc but none of the infectious agents were isolated. Suggested that its the virus that caused the outbreak.

2.) Samples of rodents that could serve as viral reservoirs or vectors + patient samples were sent for analysis by RT-PCR.


NOW, THIS IS WHERE I AM CONFUSED!!!

WHY DO WE NEED TO DO RT-PCR ?? I MEAN WHY DO WE NEED TO FIND cDNA from mRNA then amplify it rather than-->CAN WE NOT simply USE THE DNA IN THE RODENTS AND PATIENTS SAMPLES AND AMPLIFY THEM, THEN LOOK UNDER SEQUENCING GEL TO GET THE SEQUENCE, and find it on the BLAST Search????

When I look on the internet to clarify myself, all i get is the definition and process or RT-PCR or RFLPs!

The simplest explanation I can think of is that many viruses are RNA viruses, so if you want to detect their presence you need to look for their RNA, not their DNA, and for that you need RT-qPCR.

ivanbio on Mar 3 2010, 03:33 PM said:

MAN, YOU ARE GENIUS! SO SIMPLE, WHY DIDNT I THINK THAT!! I'M NEVER GOING TO SUCEED! BUT THANKS.

There will be more copies of mRNA from a single virus than of the DNA (or RNA) used to encode it. So the cDNA is an amplification step that is useful in picking up lesser amounts of virus.

The presence of a virus does not indicate that it's causing a disease. Basic PCR will only show presence or absence of an amplicon, but not so much the viral load. Real time can give you viral load, giving a better idea of whether or not the virus is the actual cause of the disease or just happens to be there.

Actually sequencing the PCR product is not really needed. You're not using random primers for this stuff, it should be primers that have been shown to be diagnostic for a particular organism or type of organism. That way, presence or absence is pretty specific. There should be no need to compare the sequence to the genetic database.

Sorry but that's confused me again.

How do you extract mRNA from rodents and human samples directly without extracting DNA? We extract DNA first, then RNA, dont we? I've just realised I'm really poor at Genetics and Molecular Biology

This is the whole thing:

People are dying of fever(haemorrhagic) because of an outbreak of possibly a virus. A guy gets samples from rodents and patient specimens and he performs RT-PCR on them(HOW? how to extract RNA?), with positive controls (dont get this bit!) consisting of genomes from viruses known to cause this fever. Nucleotide sequencing reaction (which is this? how we do it?) were then performed on small fragments of these RT-PCR products. Now, there is an accident and origin of sample was destroyed.

I now need to determine identities of viral genome amplicons. I have been provided a diagram of sequencing gel, labelled A-G.

This is what I did:


From the diagram of sequencing gel, I found my sequence and did a BLAST Search on NCBI Website and got a Virus with 79.8 Score.

how do i know this virus is reponsible for the outbreak? what do i look for?

COULD ANYONE HELP ME PLEASE? I REALLY WANT TO LEARN THIS AND UNDERSTAND. MY FRIENDS ARE JUST WORRIED ABOUT THE ACTUAL MARKS, BUT I REALLY WANT TO UNDERSTAND THIS BIT AS ITS CRUCIAL AND MUST BE SIMPLE TO UNDERSTAND. THANKS A LOT FOR ALL THE PEOPLE WHO REPLIED ABOVE.

jamescorden on Mar 3 2010, 01:46 PM said:

First, by RT-PCR, do you mean real-time PCR or reverse-transcriptase PCR?

I will assume reverse transcriptase for this response (I assumed real time above).

You can purify RNA by a number of methods, and it's not dependent on DNA purification. Positive controls are known samples used to compare the products of the unknown samples. The sequencing reaction, which I'm not as familiar with, is likely done to verify sequence and match to the database to differentiate between species that have similar but not exact sequences.

The original sample may not be necessary if the cDNA product was archived. That can be used as PCR template if a second reaction is needed, which can then be used as template for a sequencing reaction.

When you do your blast search, what are the % identity and similarity? If it's a 100% (or close to it) match, there's a good chance that is your virus.

Anyway, that's a pretty general answer. Without more specific questions, I can't give any more specific answers. I don't think the details of the method of PCR are as important as figuring out the sequence and blasting it for the identity. You could use regular PCR, but might suffer some loss in sensitivity. That is, you may not get a product to sequence without amplifying the signal by first using reverse-transcriptase.

We are not trying to identify the human or the rodent -- we know what they are (if we did want to identify them, we would use their DNA genome to do so). However, we are trying to identify a virus that is infecting these mammals, and is most likely an RNA virus -- we can't sequence the viral DNA, because there is none -- the viral genome is RNA.

So, in order to get the nucleotide sequence of the viral genome (or at least a significant enough portion of it to be confident in the identification of the virus), we need to turn its RNA into DNA, using an enzyme called reverse transcriptase, an RNA-dependent DNA polymerase (BTW, this is exactly what RNA viruses do in order to replicate in a mammalian host -- mammals don't know how to make copies of RNA, so the virus turns its RNA into DNA with reverse transcriptase so that the host can make copies of the now-DNA genome).

To do this, we first recover all the nucleic acid (DNA and RNA) from the samples (human or rodent). The DNA in this case is a contaminant (we know what a human and a mouse is, therefore their DNA is not needed), so we get rid of the mammalian DNA using a DNA-specific endonuclease (DNAse) that destroys the DNA but leaves the RNA intact. Now we have our DNA-free RNA samples.

These samples will contain host-derived RNA in huge excess, but also will contain RNA from the infecting virus as well. We turn all this RNA into cDNA using reverse transcriptase (mammalian systems turn DNA into RNA by a process called transcription; here we're turning RNA into DNA, thus the enzyme that does this is called "reverse transcriptase"), leaving us with a pool of cDNA which matches the original RNA in our host sample. The cDNA derived from the viral RNA is in this pool, but likely is present in very small amounts -- how can we get more?

We use PCR, which can turn several copies of a segment of DNA (or cDNA) into trillions of copies of that DNA. Once we perform PCR on our cDNA sample using primers specific to the viral genome, the situation is neatly reversed -- now we have many, many more copies of the viral DNA than of the mammalian cDNA, just what we need.

We purify the amplified copies of the viral DNA away from other DNA, and use that purified sample in a sequencing reaction. Sequencing allows us to know the alphabet of the viral genome, or whatever portion thereof we PCR amplified. By comparing this sequence to other known sequences in a database, we can identify the virus that is causing the fevers...

It's been years since I've actually done this, so if I've got something wrong, others with more recent or extensive experience are invited to jump in and correct any errors I've made...

Hope this helps!