Reverse transcription is the synthesis of DNA using RNA as a template. It is catalyzed by an enzyme called reverse transcriptase, which assembles a DNA strand complementary to the sequence of an RNA molecule. The immediate product is commonly called complementary DNA, or cDNA. Depending on the biological system, this DNA may subsequently become double-stranded, integrate into a chromosome, remain outside chromosomes, or extend an existing DNA end.
Reverse transcription is essential to the replication of retroviruses and several other groups of viruses, the movement of many genetic elements, and the maintenance of chromosome ends by telomerase. Related enzymes also function in bacterial defense systems. In biotechnology, reverse transcription makes RNA accessible to methods designed for DNA, including amplification, cloning, and sequencing. Its discovery in 1970 established that genetic information can pass from RNA to DNA, substantially expanding the understanding of information transfer in living systems.
Discovery and the direction of genetic information
The conventional pathway of genetic information transfer connects DNA, RNA, and protein. In transcription, an RNA polymerase copies information from DNA into RNA; in translation, ribosomes use an RNA sequence to determine the sequence of a protein. Reverse transcription adds an RNA-to-DNA route. Its name describes the reversal of the template–product relationship in transcription, not a reversal of the same chemical reaction or the operation of an ordinary RNA polymerase backward.
The biological problem that led to its discovery arose from studies of tumor-producing RNA viruses. Howard Temin, investigating Rous sarcoma virus, proposed during the 1960s that its RNA genome passed through a DNA intermediate. His “provirus hypothesis” helped explain why infected cells could stably retain viral information and transmit it to daughter cells. Evidence that DNA synthesis was required early in infection supported this interpretation, but direct demonstration of an RNA-directed DNA-synthesizing enzyme was needed to establish the mechanism.
In June 1970, two independent reports in Nature supplied that demonstration. David Baltimore described RNA-dependent DNA polymerase activity in particles of murine leukemia virus, while Temin and Satoshi Mizutani reported corresponding activity in Rous sarcoma virus. The enzyme was associated with purified viral particles, and the experiments distinguished DNA synthesis directed by viral RNA from ordinary DNA-dependent synthesis. These discoveries linked biochemical activity to the previously proposed DNA stage of retroviral replication.
The results are often described as overturning the central dogma of molecular biology, but that formulation is historically imprecise. Francis Crick’s central dogma concerned restrictions on the transfer of sequence information out of protein; it did not require that RNA-to-DNA transfer be impossible. Reverse transcription contradicted a simplified, strictly one-way picture of DNA-to-RNA-to-protein information flow, rather than demonstrating transfer of protein sequence information back into nucleic acids.
In 1975, Baltimore, Temin, and Renato Dulbecco received the Nobel Prize in Physiology or Medicine for discoveries concerning interactions between tumor viruses and the genetic material of cells. Subsequent work established that reverse transcription was not confined to infectious RNA viruses. The discovery of reverse-transcribing mobile elements, chromosome-end synthesis by telomerase, and bacterial reverse transcriptases transformed it from an unusual viral process into a broadly distributed biochemical strategy.
Catalytic mechanism and enzyme organization
Reverse transcriptases belong to the wider family of DNA polymerases. They select deoxyribonucleoside triphosphates according to base pairing with a template and add their deoxyribonucleotide residues to a growing DNA chain. Synthesis proceeds in the direction, while the template is read in the opposite direction. For most reverse transcriptases, initiation requires a primer with an available -hydroxyl group, although the chemical nature of the primer differs among systems.
Each addition forms a phosphodiester bond and releases pyrophosphate:
The primer’s terminal hydroxyl attacks the incoming nucleotide’s alpha phosphate. Conserved acidic amino-acid residues in the enzyme coordinate divalent metal ions, usually magnesium under physiological conditions, that help organize the substrates and facilitate catalysis. The familiar “fingers,” “palm,” and “thumb” description of polymerase structure refers to regions that position incoming nucleotides, contain catalytic residues, and contact the nucleic-acid duplex. Structural studies using X-ray crystallography and other methods have revealed how these regions rearrange during nucleotide incorporation.
Many reverse transcriptases perform both RNA-dependent and DNA-dependent DNA synthesis. This combination allows a single enzyme to copy an RNA template into DNA and then participate in making the complementary DNA strand. Retroviral enzymes also contain an RNase H activity that cleaves RNA within RNA–DNA hybrids. RNase H removes much of the original RNA template and helps generate or process primers needed for later stages of replication. Polymerase and RNase H activities are coordinated but occur at distinct catalytic sites.
Enzyme organization varies substantially. Human immunodeficiency virus type 1 has a reverse transcriptase formed from two related subunits, conventionally called p66 and p51. The larger subunit contains the functional polymerase and RNase H sites, while the smaller subunit has a predominantly structural role. Other retroviral reverse transcriptases have different subunit arrangements. Telomerase and many bacterial or mobile-element enzymes lack the complete functional organization characteristic of retroviral enzymes.
Most familiar retroviral reverse transcriptases lack the proofreading exonuclease activity found in many replicative DNA polymerases. Misincorporation, template switching, and incomplete extension can therefore influence their products. Fidelity nevertheless depends on enzyme identity, sequence context, nucleotide concentrations, and reaction conditions. Reverse transcription is not uniformly inaccurate: enzymes from different biological families can differ considerably in fidelity, processivity, and ability to traverse structured RNA.
Reverse transcription in retroviral replication
A retrovirus packages its RNA genome together with enzymes needed for the next infection cycle. In human immunodeficiency virus type 1, or HIV, the particle contains two associated copies of positive-sense genomic RNA. After entry into a susceptible cell, reverse transcription takes place within a changing viral nucleoprotein assembly. Its progression is coordinated with capsid remodeling and intracellular transport; the timing and location of individual stages vary among retroviruses.
The first DNA strand begins from a cellular transfer RNA packaged into the virus. Its end pairs with a primer-binding site near the end of the viral RNA. Reverse transcriptase extends this primer toward the RNA’s terminus, producing minus-strand strong-stop DNA. RNase H degrades the RNA portion of the resulting hybrid, exposing DNA complementary to a repeated sequence designated R. This DNA can pair with the corresponding R sequence at the RNA’s end, enabling the first strand-transfer event.
Minus-strand DNA synthesis then continues along the viral RNA. RNase H removes most of the copied RNA, but a purine-rich segment called the polypurine tract persists long enough to prime plus-strand DNA synthesis. Subsequent extension and processing remove the original RNA primers and expose complementary primer-binding-site sequences. Pairing between these sequences enables the second strand-transfer event. Further DNA synthesis produces a double-stranded viral DNA molecule with long terminal repeats at both ends.
These long terminal repeats contain U3, R, and U5 regions, assembled through the copying and strand-transfer steps from sequences originally located near opposite ends of the RNA genome. Their formation illustrates why retroviral reverse transcription is more elaborate than simply making one uninterrupted DNA copy of an RNA molecule. The resulting DNA is a substrate for viral integrase, which inserts it into host chromosomal DNA. Integration and reverse transcription are distinct reactions carried out by different enzymes.
The integrated DNA, called a provirus, can be transcribed by the host’s RNA polymerase II. Some resulting RNAs function as viral messenger RNAs, while others become genomes for new particles. Because reverse transcriptase can switch between the two packaged RNA templates, reverse transcription can also generate recombinant viral genomes. Template switching may help complete synthesis when one template is damaged, but it also contributes to genetic variation when the paired RNAs differ in sequence. Integration establishes a durable DNA form of the viral genome, whereas reverse transcription supplies that form.
Reverse transcription in DNA viruses
Reverse transcription is not restricted to viruses whose mature particles carry RNA genomes. Hepatitis B virus and related hepadnaviruses package DNA but reproduce that DNA through an RNA intermediate. Their replication therefore differs both from ordinary DNA-virus replication and from the retroviral cycle. The infectious particle contains a relaxed circular DNA genome in which one strand is complete and the other is usually incomplete.
After infection, viral DNA reaches the nucleus and is repaired into covalently closed circular DNA, commonly abbreviated cccDNA. This molecule associates with proteins and serves as a transcriptional template rather than as an obligatory chromosomal insertion. Host RNA polymerase II produces several viral RNAs, including a pregenomic RNA longer than one genome unit. Pregenomic RNA serves both as a messenger for viral proteins and as the template that will be reverse-transcribed.
A viral polymerase binds a stem–loop structure called epsilon on pregenomic RNA, and this complex is incorporated into an assembling nucleocapsid. Unlike retroviruses, hepadnaviruses initiate DNA synthesis by protein priming. A tyrosine residue in the polymerase’s terminal-protein domain supplies the hydroxyl group to which the first DNA nucleotide becomes covalently attached. A short DNA sequence is synthesized and transferred to a complementary region near the other end of the RNA template, after which minus-strand DNA synthesis continues.
The polymerase’s RNase H activity degrades the RNA as DNA synthesis advances. A residual RNA fragment supplies the primer for plus-strand DNA synthesis, and additional template-transfer events help produce the characteristic circular genome. Plus-strand synthesis is often unfinished when mature particles are released. The unusual partially double-stranded genome is therefore a consequence of the replication pathway, not simply a damaged version of a conventional circular DNA molecule.
Plant pararetroviruses provide another important example. Cauliflower mosaic virus has a double-stranded DNA genome that is transcribed in the nucleus, while reverse transcription of a genome-length RNA occurs during particle-associated replication. These viruses, like hepadnaviruses, do not require integration into a host chromosome as a normal step in producing progeny.
The contrast among these systems is biologically important. Retroviruses generally reverse-transcribe an incoming RNA genome before establishing an integrated DNA form. Hepadnaviruses and plant pararetroviruses instead use reverse transcription to regenerate DNA genomes from RNA produced within an infected cell. Consequently, the presence of reverse transcriptase does not by itself establish a virus’s genome type, its need for integration, or its evolutionary placement.
Retrotransposons and the evolution of genomes
Reverse transcription allows certain transposable elements to copy themselves through RNA intermediates. Such elements, collectively called retrotransposons, are major components of many eukaryotic genomes. Their expansion differs from the movement of elements that primarily excise and reinsert DNA: transcription produces an RNA copy, reverse transcription produces new DNA, and insertion can increase the number of genomic copies without removing the original element.
Long-terminal-repeat retrotransposons share substantial mechanistic and evolutionary relationships with retroviruses. Their RNA can be packaged into intracellular virus-like particles containing proteins required for reverse transcription. A double-stranded DNA copy is generated and inserted into the genome by integrase. Many such elements lack the machinery needed for transmission as an infectious virus between organisms, although the evolutionary boundaries among retroviruses, endogenous retroviruses, and LTR retrotransposons are complex.
Non-LTR retrotransposons use a different strategy. Human LINE-1 elements encode an RNA-binding protein, ORF1p, and a second protein, ORF2p, with endonuclease and reverse-transcriptase activities. In target-primed reverse transcription, ORF2p cleaves a chromosomal DNA strand, exposing a -hydroxyl group that primes DNA synthesis directly from LINE-1 RNA. The new copy is thus synthesized at its eventual insertion site rather than being fully manufactured elsewhere before integration. Many resulting insertions are truncated at their ends.
LINE-1 machinery can also mobilize RNAs that do not encode their own reverse transcriptase. Human Alu elements are a prominent example. Occasionally, cellular messenger RNAs are copied and inserted, producing processed gene copies that usually lack the introns present in the original gene. Such copies may become nonfunctional processed pseudogenes, but some acquire expression and biological functions. Reverse transcription therefore contributes not only repeated DNA but also new arrangements of gene-derived sequences.
Sequence analyses of the human genome indicate that roughly 45 percent of its sequence is recognizably derived from transposable elements, with most of that fraction attributable to retrotransposons. Approximately 8 percent consists of endogenous retroviral sequences and related LTR-derived material. These figures describe accumulated evolutionary remnants, not the proportion of elements currently capable of movement. Most copies are inactive because of mutation, deletion, or host repression.
New insertions can disrupt genes, alter regulatory regions, or provide substrates for homologous recombination. Conversely, sequences of retroelement origin have sometimes been incorporated into host functions. Cells restrict their activity through mechanisms involving DNA methylation, chromatin regulation, and RNA-based silencing. The distribution of retroelements across genomes consequently records a continuing interaction between molecular copying mechanisms, host regulation, and evolutionary change.
Telomerase and chromosome-end synthesis
Telomerase is a specialized reverse-transcribing enzyme that extends telomeres, the repetitive DNA–protein structures at the ends of linear chromosomes. Conventional DNA replication cannot by itself maintain every terminal sequence indefinitely. Removal of terminal RNA primers and the processing needed to produce protected chromosome ends create an end-maintenance problem. Telomerase counteracts sequence loss by adding repeated DNA using a short RNA template carried within the enzyme complex.
In humans, the catalytic protein is telomerase reverse transcriptase, or TERT, and its associated RNA is called TERC or telomerase RNA. The RNA has both a templating role and structural roles in enzyme assembly and function. Other proteins help stabilize the RNA, assemble the complex, and deliver or regulate it at telomeres. Telomerase therefore operates as a ribonucleoprotein enzyme rather than as a polymerase that encounters an unrelated free RNA template.
The chromosome’s existing DNA end serves as the primer. A portion of this end base-pairs with the template region in telomerase RNA, and TERT adds DNA nucleotides to extend it. Human telomerase produces repeats of the sequence -TTAGGG-. After reaching the end of its short template, the enzyme can reposition the DNA relative to the RNA and begin another repeat. This repeat-addition processivity distinguishes telomerase from reverse transcriptases that mainly copy a long RNA template once.
Extension of the guanine-rich strand does not complete all aspects of telomere replication. Conventional DNA-synthesis machinery, aided by specialized telomere-associated factors, synthesizes complementary DNA and coordinates subsequent processing. Telomere-binding proteins regulate enzyme access and help distinguish natural chromosome ends from broken DNA. Telomerase activity must therefore be understood within a larger system of chromosome-end protection, replication, and DNA repair.
Elizabeth Blackburn and Joseph Gall characterized the unusual repetitive DNA at chromosome ends in the ciliate Tetrahymena in 1978. Carol Greider and Blackburn identified telomerase activity in 1984 and reported it in 1985. Genetic studies by Jack Szostak and collaborators helped establish how telomeric sequences protect linear DNA. Blackburn, Greider, and Szostak received the 2009 Nobel Prize in Physiology or Medicine for discoveries concerning chromosome protection by telomeres and telomerase.
Telomerase activity is regulated differently among organisms and cell types. In humans it is prominent in germline-associated contexts and certain proliferative cell populations, while many somatic cells have little activity. Most human cancers maintain telomeres through telomerase reactivation, although a subset uses recombination-associated alternative lengthening mechanisms. These differences concern the maintenance of chromosome ends; they do not make telomerase a general-purpose copier of cellular RNA.
Bacterial reverse transcriptases and specialized functions
Reverse transcriptases occur widely in bacteria, where they participate in mobile elements and several kinds of antiviral defense. Their diversity demonstrates that RNA-directed DNA synthesis is not exclusively a feature of eukaryotic cells and their viruses. Some bacterial enzymes are evolutionarily associated with mobile introns, while others form components of systems whose products and biological roles differ sharply from retroviral DNA.
Group II introns are RNA elements capable of self-splicing, often assisted by an intron-encoded protein with reverse-transcriptase activity. During mobility, the excised intron RNA and its protein form a ribonucleoprotein complex that recognizes a DNA target. The intron RNA can reverse-splice into one DNA strand, and a DNA end supplies a primer for reverse transcription. Subsequent host-mediated processing produces a DNA copy of the intron at the new location. This mechanism couples RNA catalysis to target-primed DNA synthesis.
Some group II intron reverse transcriptases are unusually processive and can operate at relatively high temperatures. These properties reflect their biological task of copying long, structured intron RNAs and have made them useful starting materials for laboratory enzymes. Group II introns are also important in models for the evolutionary origins of spliceosomal introns and related RNA-processing machinery, although the detailed historical pathway remains an evolutionary reconstruction rather than a directly observed event.
Retrons produce a distinctive nucleic-acid product called multicopy single-stranded DNA, or msDNA. A retron RNA folds so that an internal hydroxyl group can initiate DNA synthesis, producing a branched RNA–DNA structure. In many characterized retrons, the reverse transcriptase and its nucleic-acid product operate with an associated effector as a defense system against bacteriophages. Infection-triggered changes can activate responses that prevent productive phage reproduction, sometimes at the cost of the infected cell.
Diversity-generating retroelements use reverse transcription for another purpose: creating localized sequence variation. An RNA copy of a template region is reverse-transcribed with characteristic mutagenesis associated with adenines in the template. The resulting information is transferred to a related variable region, often within a gene encoding a protein involved in molecular interactions. The extensively studied system in Bordetella phage BPP-1 diversifies a receptor-binding protein and can change host recognition.
Certain bacterial CRISPR systems contain reverse transcriptase fused to, or associated with, the adaptation protein Cas1. In characterized examples, these systems can acquire information from RNA and place corresponding DNA sequences into CRISPR arrays. Bacterial reverse transcriptases thus illustrate several distinct uses of the same fundamental chemistry: mobile-element propagation, production of defense-associated nucleic acids, targeted protein diversification, and acquisition of RNA-derived immune memory.
Laboratory synthesis, amplification, and sequencing
In experimental molecular biology, reverse transcription converts RNA into a more stable and readily manipulated DNA representation. A typical reaction contains RNA, reverse transcriptase, deoxyribonucleoside triphosphates, a primer, suitable salts, and a buffer. Because RNA is vulnerable to degradation by ribonucleases, sample preservation and extraction strongly influence the result. The reaction product represents the RNA molecules successfully copied under those conditions, not automatically a complete or unbiased record of the original sample.
Primer choice determines which RNAs enter the reaction. Oligo(dT) primers bind polyadenylated RNA and are widely used for eukaryotic messenger RNA. Random primers initiate at many sites and can copy non-polyadenylated transcripts, including abundant ribosomal RNA unless it has been removed. Gene-specific primers restrict synthesis to selected targets. Oligo(dT)-primed reactions can favor regions near transcript ends, whereas random priming and gene-specific priming introduce different sequence and accessibility biases.
Reverse transcription followed by the polymerase chain reaction is called RT-PCR. In a two-step workflow, cDNA is synthesized first and aliquots are used in separate amplification reactions. In a one-step workflow, reverse transcription and amplification occur sequentially within the same reaction vessel. Quantitative reverse-transcription PCR, usually abbreviated RT-qPCR, follows amplified product accumulation by fluorescence. “RT” denotes reverse transcription, not real time; conflating the two obscures whether a protocol measures RNA, DNA, or amplification kinetics.
Quantitative interpretation depends on RNA input, reverse-transcription efficiency, amplification efficiency, and normalization. A difference in fluorescence threshold does not by itself establish the same fold difference in starting RNA under every assay condition. Genomic DNA contamination is another important complication: amplification can occur even when no RNA-derived cDNA is present. Controls omitting reverse transcriptase and assays designed around exon junctions help distinguish sources of signal, although their effectiveness depends on the target’s genomic structure.
Most RNA sequencing methods also begin by making cDNA. Subsequent processing introduces adapters, sample identifiers, and sometimes unique molecular identifiers before DNA sequencing. Single-cell workflows may use reverse transcription to attach cell-specific barcodes to transcripts, allowing sequence reads to be assigned to their cell of origin. In contrast, direct RNA sequencing reads RNA molecules without first converting their sequence into cDNA for the actual sequencing measurement.
Enzyme engineering has improved thermostability, tolerance of inhibitors, and the copying of structured or long RNAs. Elevated reaction temperatures can reduce some RNA secondary structures, but temperature must remain compatible with enzyme performance and sample integrity. Template switching is deliberately exploited in some library-construction protocols to attach known sequences to cDNA ends. The same property can also create artificial chimeric molecules, making experimental design and data interpretation inseparable from the enzyme’s mechanism.
Antiviral inhibition and engineered genetic systems
Reverse transcriptase is an important antiviral target because some viruses depend on an activity that differs substantially from the machinery responsible for ordinary host chromosomal replication. Two major inhibitor classes used against HIV act by different mechanisms. Nucleoside and nucleotide analogue reverse-transcriptase inhibitors become active after intracellular processing to phosphorylated forms that compete with natural substrates. Many interfere with continued DNA extension after incorporation because their chemical structure does not support normal chain elongation.
Zidovudine, approved for HIV treatment in 1987, is a historically important nucleoside analogue. Other agents differ in their base structures, activation pathways, persistence within cells, and interactions with viral and host polymerases. Tenofovir is a nucleotide analogue used in treatment regimens for HIV and hepatitis B. The shared involvement of reverse transcription explains this overlap in antiviral activity, but it does not imply that all reverse-transcriptase inhibitors work against all reverse-transcribing viruses.
Non-nucleoside reverse-transcriptase inhibitors used against HIV-1 bind an allosteric pocket in the enzyme rather than acting as nucleotide substrates. Their binding alters enzyme conformation and impairs polymerase function. This pocket and its susceptibility to particular compounds are not universal properties of reverse transcriptases. Consequently, an inhibitor effective against HIV-1 may have little or no activity against a distantly related viral enzyme or against telomerase.
Resistance can arise through changes in the viral polymerase. Some mutations alter substrate discrimination, reducing analogue incorporation; others enhance removal of an incorporated inhibitor or weaken drug binding. The effects depend on the specific drug and combination of mutations. Reverse-transcriptase inhibitors act at the DNA-synthesis stage and do not themselves remove an already integrated HIV provirus or eradicate the nuclear cccDNA reservoir of hepatitis B virus. These distinctions help explain why suppressing replication and eliminating persistent viral genetic material are separate biological problems.
Reverse transcription has also been adapted for genome editing. Prime editing, introduced in 2019, combines a modified Cas9 protein that nicks DNA with an engineered reverse transcriptase and a prime-editing guide RNA. The guide RNA both directs target recognition and carries a primer-binding sequence and a template encoding the intended edit. A nicked genomic DNA end pairs with the primer-binding region, and reverse transcriptase extends that end using the RNA template. Cellular processing of the resulting DNA intermediates can install the programmed sequence.
This engineered use differs from ordinary cDNA synthesis because the primer is part of the chromosome and the RNA template encodes a designed local change. Editing outcomes depend on target sequence, guide design, delivery, cell state, and repair pathways. Bacterial retrons and retroelement-derived proteins have likewise been adapted experimentally to generate DNA donors or record molecular events, extending reverse transcription from an object of biological study into a component of programmable genetic systems.