Complementary DNA (cDNA) is DNA synthesized from an RNA template by the enzyme reverse transcriptase. In laboratory practice, the term commonly refers to DNA copied from messenger RNA (mRNA), although other types of RNA can also serve as templates. The initial product is a DNA strand complementary to the RNA; a second DNA strand may subsequently be synthesized to produce double-stranded cDNA. Because cDNA preserves sequence information from RNA in a form compatible with many DNA-based methods, it is central to the analysis of gene expression, molecular cloning, and transcript sequencing.
Molecular basis
Ordinary transcription produces RNA using DNA as a template. Reverse transcription transfers sequence information in the opposite direction, from RNA to DNA. Reverse transcriptase extends a primer by adding deoxyribonucleotides to its end, synthesizing DNA in the direction while reading the RNA template in the opposite direction.
The first cDNA strand is antiparallel and complementary to the RNA template. Adenine in the RNA pairs with thymine in DNA, uracil with adenine, and cytosine with guanine. For example:
If a second DNA strand is synthesized, its sequence corresponds to that of the RNA template, with thymine replacing uracil. Consequently, descriptions of a “cDNA sequence” often present this RNA-equivalent strand rather than the complementary first strand.
The word complementary describes the relationship between the initial DNA copy and its RNA template. Although any DNA strand can have a complementary sequence, cDNA conventionally denotes DNA derived from RNA, rather than any complementary DNA strand.
Relationship to genomic DNA
Genomic DNA represents an organism’s genome, including expressed and unexpressed regions. A cDNA preparation instead represents the RNA molecules present in a particular sample and captured by a particular preparation method.
In eukaryotes, most protein-coding genes contain introns that are removed from precursor RNA by RNA splicing. A cDNA copied from a fully processed mRNA therefore normally contains joined exon sequences rather than the intron-containing organization of the corresponding genomic region. This distinction makes cDNA useful for identifying splice junctions and studying alternative splicing.
However, cDNA is not necessarily an intron-free copy of a protein-coding sequence:
- RNA samples may contain unspliced or partially spliced transcripts.
- Mature transcripts can retain introns.
- cDNA can be synthesized from noncoding RNA.
- An mRNA-derived cDNA can include untranslated regions as well as the protein-coding region.
- Many cDNA molecules are incomplete copies of their templates.
An mRNA-derived cDNA does not ordinarily include the upstream promoter or other genomic regulatory regions that are not part of the transcript. It may also reflect RNA processing or editing, so differences between cDNA and genomic sequence do not necessarily indicate a genomic mutation.
Laboratory synthesis
RNA selection and primer choice
cDNA synthesis begins with an RNA preparation. Depending on the purpose, this may be total RNA, enriched mRNA, or a selected RNA species. Reverse transcriptase requires a primer with an available hydroxyl group; it does not normally begin a new DNA strand without one.
Three widely used priming strategies capture different parts of the RNA population:
| Primer type | Binding target | Main characteristics |
|---|---|---|
| Oligo(dT) | Poly(A) tails of many eukaryotic mRNAs | Enriches for polyadenylated transcripts but can favor their regions and miss non-polyadenylated RNA |
| Random primers | Multiple sites across RNA molecules | Can copy many RNA classes, including abundant ribosomal RNA, and generate overlapping partial copies |
| Gene-specific primers | A selected RNA sequence | Restricts synthesis to a chosen target or group of targets |
Primer choice therefore influences which transcripts are represented and how evenly their lengths are covered. Oligo(dT) priming can also occur at internal adenine-rich sequences rather than exclusively at the poly(A) tail.
First- and second-strand synthesis
During first-strand synthesis, reverse transcriptase produces a DNA copy associated with the RNA template. Depending on the method, the RNA is then removed or degraded, and DNA polymerase synthesizes a second DNA strand. Other workflows amplify the first-strand product directly or add sequence adapters before further processing.
A cDNA preparation may therefore consist of single-stranded DNA, double-stranded DNA, or DNA incorporated into a more elaborate sequencing or cloning construct. “Full-length cDNA” specifically indicates a copy intended to span an entire RNA transcript; ordinary cDNA synthesis does not guarantee full-length products.
Major applications
Analysis of gene expression
In reverse-transcription PCR (RT-PCR), RNA is first converted into cDNA, which is then amplified by the polymerase chain reaction. This allows RNA sequences to be studied using conventional DNA amplification.
When combined with quantitative PCR, the method is commonly called RT-qPCR. It estimates target RNA abundance through measurements made on the resulting cDNA. Interpretation depends on RNA quality, reverse-transcription efficiency, amplification performance, and normalization. The abbreviation RT-PCR denotes reverse-transcription PCR and should not be treated as synonymous with real-time PCR.
Cloning and recombinant protein production
A cDNA can be inserted into a plasmid or another vector as part of recombinant DNA work. For many eukaryotic genes, using a mature mRNA-derived coding sequence avoids the problem that bacterial expression hosts generally cannot remove eukaryotic spliceosomal introns.
A suitable cDNA coding sequence is nevertheless only one requirement for expression. The construct also needs regulatory elements compatible with the host, and the host may not reproduce all of the folding, processing, or modifications required for a functional protein.
cDNA libraries and sequencing
A cDNA library is a collection of DNA copies derived from an RNA population. Classical libraries contain cDNA inserts cloned into vectors. Their contents reflect the tissue, cell type, developmental stage, or experimental condition from which the RNA originated, as well as biases introduced during preparation.
In RNA sequencing, many methods convert RNA into cDNA and prepare those molecules for DNA sequencing. The resulting data can reveal transcript abundance, splice junctions, transcript boundaries, and previously uncharacterized transcripts. Not all RNA-sequencing methods require cDNA: direct RNA sequencing reads RNA molecules themselves.
Single-cell RNA-sequencing methods also commonly use reverse transcription. Molecular barcodes can be introduced during preparation to associate sequences with individual cells or original captured molecules.
Historical development
The discovery of reverse transcriptase in 1970, independently reported by Howard Temin and Satoshi Mizutani, and by David Baltimore, established an enzymatic mechanism for RNA-directed DNA synthesis. The work arose from studies of RNA tumor viruses and helped explain how retroviral genetic information could pass through a DNA intermediate.
During the 1970s, reverse transcription became an important laboratory route to obtaining DNA copies of expressed genes. cDNA cloning made it possible to isolate and characterize protein-coding sequences without first reconstructing their exon structure from genomic DNA.
Later developments in PCR, automated sequencing, and high-throughput transcript analysis expanded cDNA use from the investigation of individual genes to large-scale studies of RNA populations.
Limitations and interpretation
cDNA is an experimentally generated representation of RNA, not a complete or unbiased inventory of it. Several limitations affect its interpretation:
- RNA integrity: Degraded RNA produces fragmented templates and can cause uneven transcript coverage.
- Incomplete copying: RNA secondary structure, transcript length, and enzyme properties can prevent synthesis of full-length copies.
- Representation bias: Primer selection, RNA enrichment, and sequence-dependent reaction efficiencies influence which molecules are captured.
- Sequence artifacts: Reverse transcriptase can introduce errors or switch templates. Subsequent amplification can introduce further errors or distort relative abundance.
- Genomic DNA contamination: Residual DNA can be amplified alongside cDNA. No-reverse-transcriptase controls help distinguish such contamination from RNA-derived signals.
- Limited biological inference: RNA abundance does not directly establish protein abundance or activity, because translation and protein turnover are separately regulated.
A cDNA library’s failure to contain a sequence therefore does not, by itself, establish that the corresponding gene is absent from the genome or never expressed. Likewise, an apparent sequence variant or splice form may require independent validation to distinguish biological variation from preparation artifacts.