Making cDNA, or complementary DNA, means using an enzyme called reverse transcriptase to copy an RNA molecule into a stable DNA version that you can then amplify, clone, or sequence. The process boils down to a handful of core steps: prepare clean RNA, pick your primers, run a reverse transcription reaction, and optionally synthesize a second strand. Each step has pitfalls that can quietly wreck your results, and the choices you make at every stage depend on what you plan to do with the cDNA afterward.
Why Bother Converting RNA to DNA
RNA is the molecule cells use to carry genetic instructions from genes to the protein-making machinery. If you want to study which genes a cell is actively using, RNA is the direct readout. The problem is that RNA is fragile. It breaks down easily on the bench, it cannot be amplified by standard PCR, and it is a poor substrate for most cloning workflows. Converting RNA into a DNA copy gives you a durable molecule you can amplify millions of times over, insert into a plasmid, or feed into a sequencing instrument. The discovery of the enzyme that makes this possible, reverse transcriptase, was reported simultaneously by Howard Temin and David Baltimore in 1970 and fundamentally changed molecular biology by showing that genetic information could flow from RNA back to DNA.
1PubMed Central. 50th anniversary of the discovery of reverse transcriptaseStep One: Preparing High-Quality RNA
Everything that follows depends on starting with clean, intact RNA. After you extract total RNA from your cells or tissue, one critical issue is genomic DNA contamination. Even a trace of leftover genomic DNA will produce false signals in PCR, making it look like a gene is being expressed when really you are just amplifying the DNA that was already there. The standard fix is to treat your RNA sample with DNase I, an enzyme that chews up DNA while leaving RNA intact.
Protocols for DNase treatment follow a similar logic. You incubate the RNA with DNase I, then stop the enzyme’s activity and remove or inactivate it before moving to reverse transcription. One well-tested approach treats one microgram of RNA with one unit of DNase I for 30 minutes at 37°C, then heat-inactivates the enzyme at 75°C for five minutes. This was shown to destroy all contaminating DNA while completely preserving the target mRNAs.
2PubMed. Optimization of Dnase I removal of contaminating DNA from RNA for use in quantitative RNA-PCRAn alternative approach adds a cleanup step after DNase treatment. In yeast RNA preparations, for example, researchers have compared a protocol that purifies the DNase-treated RNA with phenol-chloroform extraction and ethanol precipitation against a simpler protocol that skips the cleanup and goes directly into reverse transcription. Both approaches efficiently eliminate DNA contamination.
3PubMed Central. Comparing protocols for preparation of DNA-free total yeast RNA suitable for RT-PCRThe phenol-chloroform route adds time and loses some RNA to extra handling, but it removes the DNase enzyme entirely and gives you a cleaner template. The direct route is faster and simpler, and works well when you are processing many samples. Your choice depends on how sensitive your downstream assay is and whether residual enzyme components might interfere.
Step Two: Choosing Your Primers
Reverse transcriptase needs a short stretch of nucleotides, a primer, to grab onto the RNA template and begin copying. There are three common options, and each one suits different situations.
- Oligo(dT) primers: These are short strings of thymine nucleotides that bind to the poly(A) tail found at the end of most messenger RNAs. They selectively target mRNA and tend to produce cDNA that is biased toward the 3′ end of transcripts, since the enzyme starts copying from the tail and may not reach the far end of long messages. Oligo(dT) priming is a natural fit when you only care about protein-coding genes and want to avoid copying ribosomal RNA or other non-polyadenylated species.
- Random hexamers: These are mixtures of all possible six-nucleotide sequences, so they bind throughout any RNA molecule. They give you broader coverage, including non-coding RNAs and ribosomal RNA, and they can prime from internal regions of transcripts. This makes them a better choice when your target lacks a poly(A) tail or when you need more even representation across the full length of a transcript.
- Gene-specific primers: If you already know exactly which transcript you want to measure, you can design a primer that matches a specific sequence within that gene. This produces the most focused cDNA and minimizes background, but it limits you to one target per reaction.
A direct comparison of oligo(dT) and random hexamer priming for environmental cDNA libraries found that, on average, both methods recovered comparable richness and composition of target gene sequences, though the random hexamer approach introduced more variability between replicates.
4PubMed. Comparative assessment of fungal cellobiohydrolase I richness and composition in cDNA generated using oligo(dT) primers or random hexamersMany researchers hedge their bets by combining oligo(dT) and random hexamers in the same reaction. This gives reasonably comprehensive coverage of polyadenylated transcripts while still catching some non-poly(A) species. When studying non-coding RNAs specifically, a method that incorporates random hexamer priming preceded by poly(A)-tailing and adaptor-anchoring steps has been shown to produce lower quantitative PCR cycle-threshold values for the majority of targets tested, indicating more efficient detection.
5Europe PMC. Quantification of long non-coding RNAs using qRT-PCR: comparison of different cDNA synthesis methods and RNA stabilityStep Three: Picking the Right Reverse Transcriptase
The enzyme doing the actual work is reverse transcriptase, and not all versions behave the same way. The two workhorses are derived from avian myeloblastosis virus (AMV) and Moloney murine leukemia virus (M-MLV). Both copy RNA into DNA, but they handle heat and RNA structure differently.
AMV reverse transcriptase binds more tightly to its template and tends to stay attached during synthesis, which gives it better tolerance of higher temperatures. M-MLV reverse transcriptase has a looser grip and is more sensitive to heat. However, M-MLV has become the more popular backbone for engineered enzymes because it is easier to modify.
6PubMed Central. The role of template-primer in protection of reverse transcriptase from thermal inactivationOne key modification involves removing the RNase H activity that wild-type reverse transcriptases carry. In nature, the enzyme copies RNA into DNA and simultaneously degrades the RNA template behind it. That is fine for the virus but problematic in the lab, because premature RNA degradation can cause the enzyme to fall off before it finishes copying long transcripts. Researchers engineered M-MLV variants that lack RNase H activity, and these enzymes can synthesize cDNA at much higher temperatures without losing function.
6PubMed Central. The role of template-primer in protection of reverse transcriptase from thermal inactivationRunning the reaction at a higher temperature helps melt through RNA secondary structures that would otherwise stall the enzyme. This is especially important when your target RNA has complex folding patterns or is very long. Engineered thermostable and highly processive M-MLV variants have been developed specifically for faithful synthesis of long cDNA on structured RNA templates.
7Protein Engineering, Design and Selection. Generation and characterization of new highly thermostable and processive M-MuLV reverse transcriptase variantsMost commercial kits today use an engineered M-MLV derivative. The exact identity varies by manufacturer, but they generally share these traits: reduced or absent RNase H activity, tolerance of reaction temperatures up to about 50–55°C, and improved processivity compared to the wild-type enzyme.
Step Four: Running the First-Strand Synthesis Reaction
With your RNA cleaned, primers chosen, and enzyme in hand, the actual reaction setup is straightforward. A typical first-strand synthesis involves combining your RNA template, primers, reverse transcriptase, a buffer containing magnesium, and a supply of the four deoxynucleotide building blocks (dNTPs). Many protocols also include an RNase inhibitor protein to protect the RNA template during the reaction.
The reaction usually begins with a brief primer-annealing step: you heat the RNA and primers together to around 65°C to open up secondary structures, then cool on ice to let the primers bind. After that, you add the enzyme and buffer components and incubate at the enzyme’s optimal temperature, typically 37–50°C depending on the enzyme variant, for 30 to 60 minutes. A final heat-inactivation step at around 70–85°C for a few minutes stops the reaction and denatures the enzyme. What you are left with is a hybrid molecule: the original RNA strand paired with a newly synthesized cDNA strand.
A few practical tips can make the difference between a great prep and a failed one. First, keep everything RNase-free. Wear gloves, use dedicated pipettes, and work with nuclease-free water. Second, do not skimp on the denaturation step before adding enzyme. RNA folds on itself readily, and skipping the heat step means the primers may land in the wrong place or not bind at all. Third, if your downstream application is quantitative PCR, run a no-reverse-transcriptase control alongside your real reaction. This control tells you whether any signal in your PCR comes from leftover genomic DNA rather than genuine cDNA.
When You Need Double-Stranded cDNA
For many applications, first-strand cDNA is all you need. Quantitative PCR, for instance, uses first-strand cDNA directly as a template. But if you are building a cDNA library for cloning or for certain sequencing protocols, you need a second DNA strand to make the molecule fully double-stranded.
The classical method for second-strand synthesis uses a combination of RNase H, E. coli DNA polymerase I, and E. coli DNA ligase. RNase H nicks and partially degrades the RNA strand of the RNA-DNA hybrid left over from first-strand synthesis. The resulting short RNA fragments serve as primers for DNA polymerase I, which fills in the gaps with DNA. DNA ligase then seals the remaining nicks to produce a continuous double-stranded cDNA molecule.
8Nature Methods. Synthesis of complementary DNAAdding E. coli DNA ligase during this process has been shown to increase the length of the resulting cDNA clones, particularly for long RNA templates, because it repairs nicks that would otherwise cause the molecule to break during subsequent handling.
9PubMed Central. Second-strand cDNA synthesis with E. coli DNA polymerase I and RNase H: the fate of information at the mRNA 5′ terminus and the effect of E. coli DNA ligaseAfter second-strand synthesis, the ends of the double-stranded cDNA are usually blunted with a DNA polymerase and phosphorylated with T4 polynucleotide kinase to prepare them for ligation into a cloning vector or for adapter ligation in a sequencing library prep.
Template-Switching Methods for Full-Length Capture
One persistent challenge in cDNA synthesis is capturing the very beginning (5′ end) of the original mRNA. Reverse transcriptase starts at the 3′ end and works its way toward the 5′ end, but it sometimes falls off before reaching the far end of a long transcript. Even when it does reach the end, the information at the extreme 5′ cap of the mRNA can be lost during second-strand synthesis and cloning.
Template switching, often marketed under the name SMART (Switching Mechanism at the 5′ end of RNA Template), elegantly solves this. When M-MLV reverse transcriptase reaches the 5′ end of the RNA, it adds a few extra non-templated cytosines to the cDNA. A specially designed oligonucleotide with guanines at its 3′ end base-pairs with those extra cytosines, and the enzyme switches templates and continues copying along the oligonucleotide. This anchors a known sequence at the 5′ end of every full-length cDNA in a single step.
10PubMed. Reverse transcriptase template switching: a SMART approach for full-length cDNA library constructionOne limitation of early template-switching protocols was that the reverse transcriptase could keep extending beyond the oligonucleotide, forming unwanted concatenated copies of the switching oligo that increased background noise. Incorporating non-natural nucleotide bases at the 5′ end of the template-switching oligonucleotide has been shown to block the enzyme from reading through, which reduces background and improves cDNA yield.
11PubMed Central. Incorporation of non-natural nucleotides into template-switching oligonucleotides reduces background and improves cDNA synthesis from very small RNA samplesTemplate switching is now the basis of most single-cell RNA sequencing workflows, where you need to capture full-length transcripts from tiny amounts of starting material.
Working with Degraded or Low-Input RNA
Not all RNA comes out of cells in pristine condition. Formalin-fixed, paraffin-embedded (FFPE) tissue samples, the kind archived in hospitals for pathology, are notorious for producing severely fragmented RNA. Clinical biopsies and old stored samples present similar challenges. If you only have a few nanograms of total RNA, or if your RNA is partially degraded, the standard protocol needs adjustment.
Random hexamer priming becomes essential with degraded RNA, because oligo(dT) primers require an intact poly(A) tail that fragmented transcripts often lack. For FFPE-derived RNA, a common approach involves depleting ribosomal RNA through hybridization to specific probes and RNase H digestion, then converting the remaining transcripts to cDNA using random primers.
12PubMed Central. Optimization for Sequencing and Analysis of Degraded FFPE-RNA SamplesAnother strategy for severely degraded material is 3′ mRNA-Seq, which focuses sequencing reads on the 3′ end of transcripts rather than trying to capture the full length. Because degradation tends to chew RNA from the 5′ end, the 3′ end is often the last fragment standing. This approach, combined with unique molecular identifiers that tag each original cDNA molecule, has been shown to generate high-quality sequencing data even from severely degraded FFPE-derived RNA.
13PubMed Central. Application of the 3′ mRNA-Seq using unique molecular identifiers in highly degraded RNA derived from formalin-fixed, paraffin-embedded tissueFor very low-input samples, such as single cells or laser-capture microdissected tissue, template-switching methods with PCR amplification of the cDNA are the standard. The key is to minimize the number of handling steps and tube transfers, since every manipulation loses material. Many commercial kits designed for low-input work combine reverse transcription, template switching, and cDNA amplification in a streamlined protocol that keeps the sample in one tube as long as possible.
Common Mistakes and How to Avoid Them
Certain errors show up again and again in labs that are new to cDNA synthesis. Knowing what to watch for can save weeks of troubleshooting.
- Skipping the no-RT control: If you do not include a reaction that omits the reverse transcriptase enzyme, you have no way to distinguish genuine cDNA signal from amplification of contaminating genomic DNA. This is the single most important negative control in any RT-PCR experiment.
- Using too much RNA: Overloading the reaction with RNA can inhibit the reverse transcriptase. Most enzymes work best with about 10 nanograms to 5 micrograms of total RNA per reaction, depending on the kit. More is not better.
- Ignoring RNA integrity: Running a quick gel or using a bioanalyzer to assess your RNA before starting saves you from synthesizing cDNA off a degraded template and then wondering why your quantitative PCR results look strange. If your RNA is degraded, switch to random hexamers and adjust expectations accordingly.
- Freeze-thaw cycling your RNA: Every time you thaw and refreeze an RNA sample, you risk degradation. Aliquot your RNA into single-use tubes immediately after extraction.
- Forgetting RNase-free technique: RNases are extraordinarily stable enzymes that are present on skin, in dust, and on unclean labware. A single fingerprint on a tube rim can destroy your RNA sample. Treat everything as potentially contaminated until proven otherwise.
What Happens After cDNA Synthesis
The cDNA you have made is a means to an end, and the end shapes how you handle the product. For quantitative PCR, the first-strand cDNA is typically diluted and used directly as template. A study comparing qPCR and RNA-seq for quantifying gene expression used a standard commercial RNA-to-cDNA kit followed by amplification with SYBR green master mix, a workflow that represents the most common use of cDNA in routine molecular biology labs.
14PubMed Central. Comparison between qPCR and RNA-seq reveals challenges of quantifying HLA expressionFor RNA sequencing, the cDNA goes through adapter ligation, size selection, and sometimes additional PCR amplification before loading onto a sequencer. The details vary by platform. For cDNA library construction and cloning, the double-stranded cDNA is ligated into a vector and transformed into bacteria. And for newer long-read sequencing technologies, the emphasis is on generating full-length cDNA with minimal fragmentation, which makes enzyme choice and template-switching efficiency especially important.
Regardless of the application, storing your cDNA at -20°C is standard. Unlike RNA, cDNA is reasonably stable at this temperature and tolerates a few freeze-thaw cycles without significant degradation. Still, keeping aliquots and minimizing unnecessary handling is good practice, particularly if your starting material was precious or irreplaceable.
How cDNA Synthesis Fits into Single-Cell Genomics
Perhaps the most demanding application for cDNA synthesis today is single-cell RNA sequencing. Here you start with the RNA content of one individual cell, which can be as little as about 10 picograms of total RNA. Standard protocols simply do not work at that input level. The field depends heavily on template-switching reverse transcription followed by PCR amplification to generate enough material for sequencing. Every inefficiency in the reverse transcription step gets amplified along with the cDNA, so enzyme performance and priming strategy matter more than in any bulk application.
Droplet-based platforms isolate thousands of individual cells in oil-water emulsions, each containing a barcoded oligo(dT) primer bead. The reverse transcription happens inside the droplet, tagging every cDNA molecule from that cell with a unique barcode. Plate-based methods process cells individually in wells, often using template-switching with full-length transcript coverage. The choice between these approaches involves trade-offs in the number of cells profiled versus the depth of transcript information per cell, but both rely on the same fundamental cDNA synthesis chemistry described above, just pushed to its performance limits.