Reverse transcription
A primer of 18–25 thymidines anneals to the poly-A tail, so only polyadenylated RNA is copied and every cDNA starts at the 3′ end. It is the cleanest way to focus the reaction on mRNA, ribosomal RNA is largely excluded, but it cannot see a transcript without a tail, and coverage falls away towards the 5′ end in a way that gets much worse as the RNA fragments.
Our ruleSimulations prepare you for the bench but don't replace hands-on practice.
Oligo(dT)
A primer of 18–25 thymidines anneals to the poly-A tail, so only polyadenylated RNA is copied and every cDNA starts at the 3′ end. It is the cleanest way to focus the reaction on mRNA, ribosomal RNA is largely excluded, but it cannot see a transcript without a tail, and coverage falls away towards the 5′ end in a way that gets much worse as the RNA fragments.
How it differs from the standard method
- Needs a poly-A tail: histone mRNAs, most lncRNAs, rRNA, mature miRNA and bacterial mRNA are invisible.
- Anchored or locked variants (oligo-dT with a V or VN at the 3′ end) force priming at the tail–body junction rather than somewhere inside the tail, which tightens the start point.
- Priming has its own temperature ceiling: oligo(dT)₂₀ melts off around 56 °C, so a very hot reaction loses the primer even if the enzyme is happy.
- On degraded RNA the 3′ bias becomes severe, so comparisons must either use 3′-proximal amplicons or RNA of matched integrity.
1One enzyme, one pass, one decision that matters most
Reverse transcriptase is an RNA-dependent DNA polymerase: give it a primer annealed to an RNA template and a supply of dNTPs and it will copy that RNA into a complementary DNA strand. That is the whole reaction.
Oligo(dT)
Selects mRNA. Needs a poly-A tail, so histone mRNAs, most lncRNAs, rRNA and bacterial transcripts are invisible. Because synthesis starts at the tail and the enzyme is not infinitely processive, coverage falls away towards the 5′ end, and on fragmented RNA it collapses.
Random hexamers
Copies everything, tail or no tail, and gives even coverage along a transcript. The price is that most of the cDNA is ribosomal, each strand is only a few hundred bases, and the yield number means almost nothing about your gene.
Gene-specific primer
The most sensitive option for one target: all the enzyme points at it and nothing competes. The tube then contains only that cDNA, so a reference gene needs its own reaction, and you cannot come back to it for a second gene.
Minus-RT
The same RNA, the same mix, no enzyme. Anything that amplifies afterwards was DNA to begin with. It is the one control that separates your measurement from genomic DNA, and the one most often left off the plate.
2Why the 3′ end reads higher than the 5′ end
Two independent things have to be true for an assay to find its template in oligo(dT)-primed cDNA: the stretch between the poly-A tail and the far end of the amplicon has to be intact in the RNA, and the enzyme has to have copied all of it in one go. Both get less likely with distance.
3Structure, temperature and the enzyme you have
RNA is single-stranded and therefore folded. Hairpins and more complex structure hide priming sites and physically stall the polymerase, which is both why protocols start with a 65 °C denaturation and a snap-cool on ice, so the primer binds before the RNA refolds, and why running the reaction hot helps.
- First-strand cDNA
- The DNA copy of the RNA made in this reaction. A DNA:RNA hybrid until the RNA is removed or displaced; usually used directly as a PCR template without any second-strand step.
- Processivity
- How far the enzyme travels before letting go. It sets how far from the primer you can still measure, and it is what temperature, time and enzyme choice buy you.
- 3′ bias
- The fall in coverage with distance from the poly-A tail in tail-primed cDNA. Reported as a 3′/5′ assay ratio; a measure of RNA integrity and processivity, not of expression.
- Minus-RT (−RT) control
- An otherwise complete reaction with the enzyme left out. Signal in it comes from DNA, so it is what makes a cDNA measurement interpretable.
- RNase H activity
- Degradation of the RNA strand of an RNA:DNA hybrid. Native M-MLV has it; RNase H-minus mutants keep the template intact and so make longer cDNA.
- RIN
- RNA integrity number, 1–10, from the electrophoretic trace. It stands in for mean fragment length, which is what actually decides how far an oligo(dT) cDNA reaches.
- Template switching
- The enzyme jumping to a second template, used deliberately in single-cell protocols to add an adapter at the 5′ end, and a source of artefact when it happens by accident.
- One-step / two-step
- Whether the reverse transcription shares a tube with the PCR that follows. Two-step makes cDNA first and keeps it; one-step does both in sequence in a sealed tube.
4How it fails
- Genomic DNA. A column leaves a few per cent behind, and that DNA is a perfectly good template for any assay whose amplicon lies inside one exon. DNase the RNA, design across an exon junction where you can, and run a minus-RT control for every sample: the per-sample control, because contamination varies between extractions.
- The gene is not polyadenylated. Replication-dependent histone mRNAs end in a stem-loop; most long non-coding RNAs, all rRNA, mature miRNA and bacterial mRNA have no tail. With oligo(dT) they read as absent. Check before concluding anything.
- Degraded RNA plus a 5′ assay. Reported as a change in expression, caused by fragment length. Measure RIN, keep it comparable across the samples you are comparing, and prefer 3′-proximal amplicons when it is not.
- Unequal input. Reverse transcription is roughly proportional to the RNA that went in, so a tube with 30% more RNA gives 30% more of everything. That is arithmetically identical to a 1.3-fold induction, and no downstream normalisation separates them unless the reference gene behaves exactly like the target.
- Carry-over inhibition. RT buffer, leftover primer and the enzyme all slow a downstream polymerase. Keep the cDNA to a tenth or less of the PCR volume, or dilute it; it affects every tube equally, so a ratio can look fine on a reaction whose absolute numbers are badly wrong.
- Priming at too high a temperature. With oligo(dT) above about 55 °C, and with random hexamers at any temperature much above room temperature, the primer is simply not bound when the enzyme needs it.
Key terms
Practise these as flashcards →A primer of 18–25 thymidines that anneals to the poly-A tail. Selects polyadenylated RNA and starts every cDNA at the 3′ end.
A mixture of all 4096 six-base sequences, annealed cold. Primes any RNA at many points, giving flat coverage and short cDNA.
Usually the reverse primer of the assay itself. Gives the lowest detection limit for one target and no cDNA for anything else.
The reaction without the enzyme. Anything amplifying from it was DNA, so it is the control that validates a cDNA measurement.
An engineered M-MLV variant that works at 50–60 °C, where RNA secondary structure melts; usually RNase H-minus as well.
Chilling the denatured RNA–primer mix on ice so the primer anneals before the RNA refolds into the structure you just melted.
Signal from a 3′-proximal assay divided by signal from a 5′ assay on the same cDNA: a direct readout of degradation and processivity.
Putting the same mass of RNA into every reaction that will be compared, measured fluorometrically rather than by A₂₆₀.
When to use it
- Any time an RNA measurement has to be made with a DNA method: qPCR, digital PCR, amplicon sequencing or cloning a transcript all need the cDNA step first.
- Cloning a coding sequence as it is actually spliced in a tissue, rather than from genomic DNA: the cDNA carries the mRNA's exon structure.
- Making a reusable archive: one oligo(dT) or hexamer reaction gives a cDNA stock that dozens of different assays can be run on later, which is why most labs reverse transcribe in bulk and store it.
- Reaching the lowest possible detection limit for a single transcript in a precious sample, where a gene-specific primer is worth the loss of flexibility.
- Measuring non-polyadenylated RNA (histone mRNAs, many lncRNAs, bacterial transcripts, viral genomes) where random or specific priming is the only option.
Compared with alternatives
The usual destination for the cDNA. Prefer thinking of them as one workflow: the Cq you eventually report carries whatever the reverse transcription did, and its two enzymatic steps compound.
Also starts with a reverse transcription, but after rRNA depletion or poly-A selection and with fragmentation built in. Choose it when you do not know in advance which transcripts matter.
Measures RNA directly, without copying it, so it reports transcript size and isoform state. Prefer it when the question is what the RNA looks like rather than how much of it there is.
Run before this step, not after. It gives the RIN and the 28S/18S ratio that tell you whether a 5′ assay on tail-primed cDNA can mean anything.
The reverse direction, DNA to RNA. Its product is often useful here as a spike-in control or as the template for an absolute standard that does include the RT step.
Sub-techniques
Choosing one here switches the whole page, including the protocol and the lab, to it.
A primer of 18–25 thymidines anneals to the poly-A tail, so only polyadenylated RNA is copied and every cDNA starts at the 3′ end. It is the cleanest way to focus the reaction on mRNA, ribosomal RNA is largely excluded, but it cannot see a transcript without a tail, and coverage falls away towards the 5′ end in a way that gets much worse as the RNA fragments.
- Needs a poly-A tail: histone mRNAs, most lncRNAs, rRNA, mature miRNA and bacterial mRNA are invisible.
- Anchored or locked variants (oligo-dT with a V or VN at the 3′ end) force priming at the tail–body junction rather than somewhere inside the tail, which tightens the start point.
- Priming has its own temperature ceiling: oligo(dT)₂₀ melts off around 56 °C, so a very hot reaction loses the primer even if the enzyme is happy.
- On degraded RNA the 3′ bias becomes severe, so comparisons must either use 3′-proximal amplicons or RNA of matched integrity.