{
  "id": "central-dogma",
  "title": "Central Dogma",
  "lang": "en",
  "source_lang": "en",
  "summary": "The central dogma describes the permitted directions of biological sequence-information transfer and excludes its transfer from proteins back into nucleic acids.",
  "category": "science",
  "subcategory": "biology",
  "type": "concept",
  "aliases": [
    "Central dogma of molecular biology"
  ],
  "generated_at": "2026-11-03",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/central-dogma/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/central-dogma/",
      "json": "https://aiwiki.page/zh/central-dogma.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "The central dogma is a principle of molecular biology concerning the transfer of sequence information among DNA, RNA, and proteins. Proposed by Francis Crick in 1958 and clarified in 1970, it states that once sequence information has entered a protein, it cannot be transferred back into a nucleic acid or used to specify another protein’s amino-acid sequence. Its familiar textbook representation, DNA → RNA → protein, describes the principal route of gene expression, but is not the complete statement of the principle.\n\nThe distinction between the original principle and its simplified diagram is essential. RNA can serve as a template for DNA synthesis, RNA genomes can be copied into RNA, and proteins can regulate almost every stage of gene expression without contradicting the central dogma. The principle concerns the residue-by-residue specification of one biological polymer by another, not every form of cellular influence, inheritance, or communication. It provides a framework for understanding genetic continuity, protein synthesis, viral replication, and the experimental manipulation of genes."
    },
    {
      "heading": "Historical origins and formulation",
      "text": "The central dogma emerged during the effort to explain how hereditary material determines the molecular properties of organisms. Experiments preceding its formulation had established that genetic specificity could reside in DNA. In 1944, Oswald Avery, Colin MacLeod, and Maclyn McCarty identified DNA as the substance responsible for transforming one bacterial type into another. In 1952, Alfred Hershey and Martha Chase used radioactive labeling in experiments with a bacteriophage to distinguish the behavior of viral DNA from that of its protein coat during infection. These results helped shift attention toward nucleic-acid sequences as carriers of hereditary information.\n\nThe double-helical model proposed by James Watson and Crick in 1953 supplied a plausible mechanism for copying DNA. Complementary base pairing suggested that each strand could specify the sequence of a new partner. It did not, by itself, explain how a DNA sequence determined the sequence of a protein. Proteins contain chemically diverse amino acids, whereas DNA contains four principal types of nucleotide. A mechanism was therefore needed to connect two different molecular alphabets.\n\nCrick discussed this problem in a 1957 lecture and in his 1958 paper “On Protein Synthesis.” He distinguished the sequence hypothesis, according to which nucleic-acid sequence specifies protein sequence, from the central dogma, which restricted the possible directions of information transfer. The latter was not simply a statement that DNA produces RNA and RNA produces protein. Its decisive claim was that sequence information, once transferred into protein, could not pass out again in a form that specified another polymer’s sequence.\n\nThe word dogma subsequently caused confusion because it appeared to imply an unquestionable doctrine. Crick later explained that he had intended a strong organizing hypothesis rather than a proposition insulated from evidence. The principle was formulated before many of the relevant mechanisms had been established experimentally, making its predictive content particularly important.\n\nIn 1970, following the discovery of RNA-dependent DNA synthesis in tumor viruses, Crick published a clarification entitled “Central Dogma of Molecular Biology.” He distinguished general transfers, special transfers, and transfers for which no mechanism was known. This account explicitly accommodated RNA-to-DNA information transfer. The historical claim that reverse transcription overturned Crick’s original dogma therefore confuses his restricted principle with the more familiar, one-way textbook diagram."
    },
    {
      "heading": "Sequence information and the scope of the principle",
      "text": "In the central dogma, “information” has a specific molecular meaning: the ordered arrangement of residues in a polymer. A nucleic acid consists of a sequence of nucleotides, and a polypeptide consists of a sequence of amino acids. Information transfer occurs when one sequence specifies another through a reproducible correspondence. DNA replication uses complementary base pairing, whereas translation uses the correspondence between nucleotide triplets and amino acids established by the genetic code.\n\nThis definition separates sequence specification from causal influence. A protein can bind DNA, alter its chemical state, activate a gene, or catalyze nucleotide incorporation without its own amino-acid sequence being copied into the DNA product. For example, a polymerase determines how synthesis proceeds and contributes to its accuracy, but the template normally determines which nucleotide is incorporated at each position. The enzyme’s participation does not make the process protein-to-DNA sequence transfer.\n\nWith three broad classes of information-bearing polymers, there are nine possible directional transfers. Crick’s 1970 account treated DNA → DNA, DNA → RNA, and RNA → protein as general transfers. RNA → RNA, RNA → DNA, and direct DNA → protein were placed in a special category rather than excluded by the principle. The prohibited class consisted of protein → DNA, protein → RNA, and protein → protein, understood specifically as transfer of primary-sequence information.\n\nDirect DNA-to-protein transfer illustrates why this classification is not simply a catalog of normal cellular pathways. Some experimental systems have supported unusual DNA-dependent protein synthesis, but direct translation from DNA is not a general route of cellular gene expression. Its logical compatibility with the dogma does not establish that it is physiologically important. Conversely, the RNA-to-DNA pathway is both compatible with the principle and widespread in nature.\n\nThe dogma also does not equate a gene with a protein. Many genes produce functional RNAs rather than protein-coding messages. Nor does it imply that a sequence uniquely determines an organism’s characteristics. A phenotype depends on gene regulation, molecular interactions, developmental history, and environmental conditions. The central dogma addresses a narrower question: which molecular sequences can serve as templates or coded specifications for others."
    },
    {
      "heading": "DNA replication and the preservation of hereditary information",
      "text": "DNA replication is the DNA → DNA transfer that preserves genetic sequence across cell generations. In the ordinary double helix, adenine pairs with thymine and guanine with cytosine. Separation of the parental strands exposes two templates, each of which can guide the formation of a complementary strand. The resulting molecules each contain one parental strand and one newly synthesized strand, a pattern called semiconservative replication.\n\nMatthew Meselson and Franklin Stahl demonstrated this pattern in 1958 using Escherichia coli grown with different nitrogen isotopes. After cells were shifted from a medium containing heavy nitrogen to one containing light nitrogen, DNA density changed in the pattern predicted by semiconservative copying. The experiment helped connect the structural model of DNA with the physical transmission of hereditary material.\n\nReplication requires coordinated enzymatic activities. Helicases separate the strands, primases supply short primers, and DNA polymerases extend those primers by adding nucleotides to an existing \\(3'\\) end. New DNA is synthesized in the \\(5' \\rightarrow 3'\\) direction. Because the two template strands are antiparallel, one new strand can be synthesized relatively continuously, while the other is formed discontinuously as Okazaki fragments. Primer removal, gap filling, and DNA ligase activity complete the daughter molecules.\n\nCopying is accurate but not perfect. Many replicative polymerases proofread newly incorporated nucleotides, removing mismatches before synthesis continues. Mismatch repair provides an additional layer of correction after replication. Other DNA repair pathways remove chemically damaged bases, repair broken strands, or restore DNA using a homologous template. Their substrates and mechanisms differ, but most preserve sequence by reference to a complementary strand or another DNA molecule.\n\nErrors that escape correction can become stable mutations. This distinguishes information transfer from exact duplication: a copying process may transmit a sequence while occasionally changing it. Such changes supply part of the variation on which natural selection acts. Recombination can also rearrange existing sequences without requiring information to flow from proteins back into nucleic acids.\n\nReplication therefore links molecular templating with biological inheritance, but inheritance need not follow only a parent-to-offspring route. DNA acquired through horizontal gene transfer can be copied and expressed in a new host. The ecological route by which a gene moves and the molecular direction in which its sequence is copied are separate questions."
    },
    {
      "heading": "Transcription and the production of functional RNA",
      "text": "Transcription transfers sequence information from DNA to RNA. An RNA polymerase reads one DNA strand and synthesizes a complementary RNA strand, using uracil in place of thymine. As in DNA synthesis, the new polymer grows in the \\(5' \\rightarrow 3'\\) direction. Unlike the principal replicative DNA polymerases, cellular RNA polymerases can generally initiate synthesis without a pre-existing nucleic-acid primer.\n\nTranscription begins at regulated DNA regions called promoters. In bacteria, promoter recognition commonly involves a sigma factor associated with the polymerase. In eukaryotes, transcription initiation requires combinations of general transcription factors and other regulatory proteins. The same DNA region need not be transcribed continuously: recruitment, initiation, elongation, and termination can all be regulated in response to cellular conditions.\n\nNot all transcription produces messenger RNA. Cells also transcribe ribosomal RNA, transfer RNA, and numerous regulatory or structural RNAs. In eukaryotic nuclei, RNA polymerase I produces the precursor of most major ribosomal RNAs, RNA polymerase II produces protein-coding transcripts and several classes of noncoding RNA, and RNA polymerase III produces transfer RNAs and certain other small RNAs. This division of labor makes RNA production much broader than an intermediate step in protein synthesis.\n\nMany eukaryotic protein-coding transcripts undergo extensive processing. A modified guanosine cap is added at the \\(5'\\) end, most acquire a polyadenylated \\(3'\\) end, and introns are removed through RNA splicing. The identification of split genes by Phillip Sharp, Richard Roberts, and their colleagues in 1977 showed that a mature messenger RNA need not correspond to one uninterrupted stretch of genomic DNA. Alternative splicing permits different combinations of exons to appear in mature transcripts from the same gene.\n\nRNA editing can further alter a transcript after it is synthesized. In a well-characterized example, editing of a cytidine to uridine creates a premature stop codon in apolipoprotein B messenger RNA in the human intestine, contributing to production of a shorter protein than the form produced in the liver. Such processing changes how DNA information is represented in a particular RNA product; it does not involve reading an existing protein sequence back into RNA.\n\nTranscription thus produces a diverse and regulated RNA population rather than a uniform set of passive messages. Some transcripts carry protein-coding instructions, while others function through their own sequences, structures, or catalytic properties."
    },
    {
      "heading": "Translation and the genetic code",
      "text": "Translation converts a nucleotide sequence into an amino-acid sequence. Unlike replication and transcription, it does not produce a chemically similar complementary polymer. It depends on the genetic code, in which three-nucleotide units called codons correspond to amino acids or termination signals. With four possible nucleotides at each of three positions, there are \\(4^3 = 64\\) possible codons.\n\nIn the standard code, 61 codons specify amino acids and three—UAA, UAG, and UGA—function as stop signals. Most amino acids are encoded by more than one codon, a property called degeneracy. This redundancy means that a protein sequence does not uniquely identify the nucleic-acid sequence that encoded it. Different DNA sequences can produce the same polypeptide, even before introns, untranslated regions, and regulatory sequences are considered.\n\nThe decoding machinery depends on transfer RNA molecules. Each has an anticodon that recognizes an appropriate messenger-RNA codon and an attachment site for an amino acid. Aminoacyl-tRNA synthetases establish the link between these two kinds of information by attaching the appropriate amino acid to its corresponding transfer RNAs. Many synthetases also possess editing activities that reduce errors in amino-acid attachment. Codon recognition alone cannot ensure accurate translation if a transfer RNA carries the wrong amino acid.\n\nThe ribosome coordinates messenger RNA, charged transfer RNAs, and accessory factors. During elongation, incoming transfer RNAs are selected, the growing peptide is transferred to the newly arrived amino acid, and the ribosome moves along the message. Formation of the peptide bond is catalyzed at a center formed primarily by ribosomal RNA. The ribosome therefore exemplifies the functional importance of RNA beyond its role as a carrier of coding information.\n\nExperimental decoding of the code began decisively in 1961, when Marshall Nirenberg and Heinrich Matthaei showed that synthetic polyuridylic acid directed production of polyphenylalanine in a cell-free system. Subsequent experiments established the assignments of the remaining codons. The resulting code is nearly universal, but not identical in all organisms and organelles. Some mitochondrial codes differ from the standard code, and specialized mechanisms can recode particular codons to incorporate selenocysteine or pyrrolysine.\n\nTranslation specifies a polypeptide’s primary sequence, not its complete mature state. Folding, cleavage, assembly with other subunits, and chemical modification determine how that polypeptide functions. These downstream processes extend the consequences of sequence information without introducing a pathway for copying protein sequence back into DNA or RNA."
    },
    {
      "heading": "Reverse transcription and RNA-based replication",
      "text": "Reverse transcription transfers sequence information from RNA into DNA. In 1970, Howard Temin and Satoshi Mizutani, and independently David Baltimore, reported RNA-dependent DNA polymerase activity in RNA tumor viruses. The responsible enzyme, reverse transcriptase, provided a mechanism for a process that Temin had proposed in developing the provirus hypothesis. These discoveries contradicted a strictly one-way DNA → RNA model, but not Crick’s original restriction on information leaving proteins.\n\nIn retroviruses, an RNA genome is copied into DNA after infection. Reverse transcriptase synthesizes DNA from the viral RNA template, its RNase H activity degrades RNA in RNA–DNA hybrids, and further synthesis generates double-stranded viral DNA. An integrase then inserts this DNA into the host genome. The integrated form, or provirus, can be transcribed by host machinery to produce viral RNAs and the messages needed for viral protein synthesis.\n\nRNA-to-DNA transfer is not confined to infectious viruses. Many transposable elements move through RNA intermediates. LINE-1 elements, for example, encode proteins that help reverse-transcribe their RNA into genomic DNA. Other repetitive elements can exploit this machinery. Such processes have made substantial contributions to the structure and evolution of eukaryotic genomes.\n\nTelomerase is another important RNA-dependent DNA polymerase. Its RNA component supplies a template for adding short DNA repeats to chromosome ends. Rather than copying an entire free RNA genome, telomerase repeatedly uses a limited internal RNA template. This illustrates how the same permitted direction of information transfer can serve very different biological functions.\n\nRNA → RNA transfer occurs in many RNA viruses, whose genomes are copied by RNA-dependent RNA polymerases. Positive-sense RNA genomes can generally serve directly as messenger RNA, whereas negative-sense RNA genomes must first be transcribed into complementary positive-sense RNA. Double-stranded RNA viruses likewise require polymerases capable of producing transcripts from RNA templates. In some nonviral organisms, RNA-dependent RNA polymerases also amplify regulatory RNA responses.\n\nThese mechanisms show why the central dogma cannot be reduced to a rule that DNA must always precede RNA. A viral life cycle can maintain hereditary information in RNA, generate a DNA intermediate, or alternate among several nucleic-acid forms. What remains excluded is the use of a protein’s amino-acid sequence as the template for reconstructing a nucleic-acid sequence."
    },
    {
      "heading": "Regulation, cellular organization, and information flow",
      "text": "The central dogma specifies possible directions of sequence transfer but does not determine how much RNA or protein a cell produces. Gene expression is controlled at multiple stages: access to DNA, transcription initiation and elongation, RNA processing, RNA stability, translation, and protein degradation. Consequently, two cells with essentially the same genome can maintain very different molecular compositions and functions.\n\nRegulatory proteins provide conspicuous examples of influence flowing toward DNA without sequence information flowing backward. A transcription factor can recognize a DNA sequence and alter recruitment of transcriptional machinery. A signaling pathway can change that factor’s activity by phosphorylation or other modifications. In neither case is the factor’s amino-acid sequence copied into DNA. The interaction changes the use of an existing sequence rather than specifying a new one.\n\nIn eukaryotes, DNA packaging adds another level of control. DNA associates with histones to form nucleosomes and higher-order chromatin structures. Accessibility depends partly on nucleosome arrangement, histone modifications, and DNA methylation. Some chromatin states can persist through cell divisions and contribute to epigenetic inheritance. Persistence of a regulatory state is distinct from transfer of protein primary-sequence information into a nucleic acid.\n\nRNA can also regulate other RNA molecules. In RNA interference, small RNAs guide protein complexes to complementary targets. Depending on the organism and pathway, the result may be messenger-RNA cleavage, translational repression, or changes in chromatin. The guide RNA supplies sequence specificity through base pairing. Proteins provide catalytic and organizational functions, but the sequence comparison remains a nucleic-acid interaction.\n\nCellular organization shapes how these reactions are coupled. In bacteria, transcription and translation can occur together, allowing ribosomes to engage a message before its transcription has finished. In eukaryotes, the nucleus separates most nuclear transcription and RNA processing from cytoplasmic translation. Mitochondria and chloroplasts retain their own genomes and gene-expression systems, while depending heavily on proteins encoded by nuclear genes.\n\nThese distinctions also explain why transcript abundance is not a direct measurement of protein abundance. Messenger RNAs differ in lifetime and translation efficiency, and proteins differ in degradation rate. A rapidly synthesized but short-lived protein can remain scarce, while a stable protein can persist after its messenger RNA has disappeared. The central dogma identifies the molecular route by which a coding sequence is expressed, but quantitative expression requires kinetic and regulatory explanations."
    },
    {
      "heading": "Apparent exceptions and conceptual boundaries",
      "text": "Prions are often presented as exceptions to the central dogma because a protein can promote conversion of other molecules of the same protein into a self-propagating conformation. In mammalian prion disease, an abnormal conformational state of prion protein promotes the conversion of normally folded molecules. Related protein-based inheritance systems occur in fungi. The propagated property is a structural state or assembly pattern, not an amino-acid sequence copied residue by residue.\n\nThis distinction matters because Crick’s excluded protein → protein transfer referred to primary sequence. In a prion process, the amino-acid sequences of newly produced protein molecules are still specified through ordinary gene expression. An existing protein influences how those molecules fold or assemble. Prions therefore broaden the study of biological inheritance without demonstrating reverse transfer of protein sequence information.\n\nProtein-guided changes to nucleic acids require similar care. Enzymes can introduce mutations, select recombination sites, or alter DNA repair outcomes. Such processes can make DNA change in a protein-dependent manner, but dependence is not equivalent to templating. A convincing violation would require a protein’s sequence to specify a corresponding DNA or RNA sequence through an established decoding mechanism, rather than merely influencing where or how a change occurs.\n\nModern genome-editing systems illustrate the distinction. In many CRISPR applications, a guide RNA identifies a DNA target through sequence complementarity, while a protein supplies cleavage activity. Other engineered systems combine reverse transcription with an RNA template to install DNA changes. These pathways can be highly programmable, but their sequence instructions reside in nucleic acids rather than in a protein sequence being reverse-translated.\n\nThe central dogma also does not describe the ultimate origin of translation. The RNA world hypothesis proposes that early biological systems relied extensively on RNA for both hereditary information and catalysis. The existence of catalytic RNAs, or ribozymes, and the RNA-centered catalytic core of the ribosome support the plausibility of important RNA roles in early evolution. They do not establish every step by which the modern coding system arose.\n\nFinally, computational reconstruction must be distinguished from molecular information transfer. A program can infer possible coding sequences from a protein sequence or use evolutionary evidence to estimate an ancestral gene. Such inference is possible because the genetic code is known, but it does not constitute a biological pathway that copies protein sequence into nucleic acid. The central dogma concerns physical mechanisms in biological systems, not every operation that can be performed on sequence data."
    },
    {
      "heading": "Experimental and technological applications",
      "text": "The central dogma supplies the logic for many methods in biotechnology. In recombinant protein production, a coding sequence is inserted into an expression construct, introduced into a suitable host, transcribed, and translated. Production depends not only on the amino-acid sequence but also on promoter choice, RNA stability, codon usage, folding conditions, and protein toxicity. A DNA sequence that expresses efficiently in one organism may perform poorly in another.\n\nReverse transcription makes RNA accessible to many DNA-based analytical methods. In reverse-transcription polymerase chain reaction, RNA is first converted into complementary DNA, which is then amplified. Reverse-transcription quantitative PCR can estimate the abundance of selected RNA targets when accompanied by appropriate controls and calibration. RNA-sequencing workflows commonly also use complementary-DNA intermediates, although direct RNA-sequencing approaches avoid that particular conversion.\n\nThe distinction among DNA, RNA, and protein measurements is fundamental to interpreting molecular data. DNA sequencing identifies genomic variants and potential coding capacity. RNA measurements reveal which sequences are transcribed and how transcripts are processed. Proteomics, often using mass spectrometry, measures proteins and can identify features such as cleavage products or chemical modifications. These approaches interrogate related but nonidentical layers of biological organization.\n\nCell-free systems allow particular information-transfer steps to be studied outside intact cells. Extracts or purified components can support transcription, translation, or both. Such systems were crucial to deciphering the genetic code and remain useful for examining translation initiation, ribosome activity, folding, and engineered synthesis. Removing membranes and selected regulatory pathways makes it possible to isolate mechanisms that are difficult to distinguish inside living cells.\n\nSynthetic biology exploits the modularity of these processes while also exposing their interdependence. Engineered genetic circuits use transcriptional regulators, regulatory RNAs, and protein turnover to control outputs. Genetic-code expansion can introduce noncanonical amino acids by supplying altered transfer-RNA and aminoacyl-tRNA-synthetase systems. These interventions modify how nucleic-acid instructions are interpreted, while retaining nucleic-acid-directed synthesis of protein.\n\nA recurring experimental problem is to separate sequence effects from changes in expression or stability. Two synonymous coding sequences may produce different quantities of the same protein, while a single amino-acid substitution may alter folding without greatly changing messenger-RNA abundance. Measurements at multiple molecular levels are therefore needed to determine whether an observed effect arises during transcription, RNA processing, translation, or the subsequent life of the protein."
    }
  ],
  "links": [
    "molecular-biology",
    "dna",
    "rna",
    "protein",
    "gene-expression",
    "bacteriophage",
    "amino-acid",
    "nucleotide",
    "gene",
    "phenotype",
    "dna-replication",
    "dna-polymerase",
    "dna-ligase",
    "dna-mismatch-repair",
    "dna-repair",
    "mutation",
    "natural-selection",
    "horizontal-gene-transfer",
    "transcription-biology",
    "rna-polymerase",
    "promoter-biology",
    "messenger-rna",
    "rna-splicing",
    "alternative-splicing",
    "translation-biology",
    "genetic-code",
    "codon",
    "transfer-rna",
    "ribosome",
    "peptide-bond",
    "ribosomal-rna",
    "protein-folding",
    "reverse-transcription",
    "reverse-transcriptase",
    "transposable-element",
    "virus",
    "transcription-factor",
    "histone",
    "chromatin",
    "dna-methylation",
    "epigenetics",
    "rna-interference",
    "eukaryote",
    "cell-nucleus",
    "prion",
    "crispr",
    "ribozyme",
    "polymerase-chain-reaction",
    "dna-sequencing",
    "proteomics",
    "mass-spectrometry"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}