NATURE VOL. 227 AUGUST 8 1970 581 Central Dogma of Molecular Biology by The central dogma of molecular biology deals with the detailed FRANCIS CRICK residue-by-residue transfer of sequential information. It states MRC Laboratory of Molecular Biology, that such information cannot be transferred from protein to either Hills Road, protein or nucleic acid. Cambridge CB2 2H “The central dogma, enunciated by Crick in 1958 and the keystone of molecular biology ever since i likely to prove a considerable over simplification. ‘Tis quotation is taken from tho beginning of an unsigned artielot headed “Contral dogma reversed”, recounting the very important work of Dr Howard ‘Temin? and others® showing that an RNA tumour virus can use viral RNA fas a template for DNA synthesis. This is not tho first time that the idea of tho contral-dogma has been mis- understood, in one way or another. In this article I explain why the torm was originally introduced, its true meaning, and stato why I think that, properly under- stood, it is still an idea of fundamental importance. ‘The contral dogma was put forward® at a poriod whon much of what wo now know in molecular genetics was not established. All we had to work on were certain frag- mentary experimental results, themselves often rather uncertain and confused, and a boundless optimism that the basic concepts involved were rather simple and probably much the same in all living things. In such a situation well constructed theories can play’a really useful part in stating problems clearly and thus guiding experi- ment. ‘The two contral coneepts which had ben produced, originally without any explicit statement of the simplifica tion being introduced, were those of sequential information and of defined alphabets. Neither of these steps was trivial. Beeauso it was abundantly elear by that timo that a protein had a woll defined threo dimensional strac- ture, and that its activity depended crucially on this structure, it was necessary to put the folding-up process on one side, and postulate that, by and large, the poly- peptide chain folded itself up. ‘This temporarily reduced tho central problem from three dimensional one to ‘ono dimensional ono. It was also necessary to argue that in spito of the miscellaneous list of amino-acids found in proteins (as then given in all biochemical text- books) some of them, such as phosphoserine, were socond- ary modifications; and that there was probably a universal feb of twonty used throughout nature. Tn the same way minor modifications to the nucleic acid bases wore ignored; uracil in RNA wns considered to be informationally CO DNA Z “ peers ig. The arrows show all the posible simple transfers between the fice fants of polymers. “They represent the directional ow of “etaled sequence taformation, analogous to thymine in DNA, thus giving four standard symbols for the components of nucleic acid. ‘The principal problem could then be stated as the formulation of the goneral rules for information transfor from ono polymer with a defined alphabet, to another. ‘This could be compactly represented by the diagram of Fig. 1 (which was actually drawn at that timo, though I ‘am not suro that it was ever published) in which all possible simplo transfors wore represented by arrows. ‘Tho arrows do not, of course, reprosent the flow of matter but the directional flow of detailed, residue-by-residue, sequence information from ono polymer molecule to another. ‘Now if all possiblo transfers commonly occurred it would have been almost impossible to construct: useful theories. Nevertheless, such theories wore part of our overyday discussions. ‘This was becanse it was boing, tacitly assumed that certain transfers could not occur. Tt ocurred to me that it would be wise to state these preconceptions explicitly. PROTEIN gd any meat eee ae ee ee it nf A little analysis showed that the transfer could be divided roughly into three groups. ‘The first group was ‘those for which some evidence, direct or indirect, seemed to exist. ‘These are shown by the solid arrows in Fig. 2 ‘They were: I (a) DNA-DNA 10) DNA+RNA I (c) RNA-+Protein I(@) RNA+RNA ‘The last of those transfers was prosumed to oceur because of the existence of RNA viruses. Next thero were two transfors (shown in arrows) for which there was noithor a1 evidence nor any strong theoretical requirement. II (a) RNA-+DNA (see the reference to ‘Temin’s work") II (0) DNA-Protein 2as dotted oxperimental "They562 ‘The Intter was the transfer postulated by Gamow, from (doublo stranded) DNA to protein, though by that time his particular theory had been disproved. ‘Tho third class consisted of tho three transfe arrows of which have beon omitted from Fig. 2 sore the tranafers: III (a) Protein-Protein IIL (6) Protoin+RNA. TI (e) ProtoinsDNA ‘Tho general opinion at the time wes that claes I almost certainly exuted, clan 11 mas probably rare oF abvont, find thas lass IL waa very unlikely to eesur.. Ths Aocison had to bo made, therefore, whothor to asssme ‘int only cass I traatoreOeourred. ‘There were, however, ho overwhelming structaral reason why tho tanafer in clus IT should ‘ot be innpossble. In thet, for all wo Imnow, the mpliention of al RNA viruses could have gone Sry of ADA intone, Orth ther ands ere ‘wero good gonoral reasons against all tho threo pose transfers in clase TEI. In bof, i was most unlly, for ereochermicaltoasoos, that protein-protein traafer ‘ould be dono in tho simplo way that DNA™>DNA tranafor ‘vas eavinged. ‘Tao transfer protein-oRNA (and the {nalogous protein-DNA) would havo required. (back) {raslation, that fs, tho teatafer from one alphabet to rutorally quitofferent ono. esas roalied: that forward translation involved very complex. machinery. Moreover it seomed unlikely on gonral grounds that this ‘machinery oul ealy work bacevard. ‘Tho only reason. able alternative wou thas the eell had evolved an entrely teparnte st of complicated machinery for back transltion, ‘sul of this there was mo truce, and'no reewon to believe that it might bo nocded. T decided, therefore, to play aafo, and to stato as the basis ammption of the now molooar biology the non. xistonco of transfers of class TI Becaute tito wero all tho possible transfers from, protein, the centsat dogs ould be stated in tho form “neo (soquentil) information Tin paated into protein it ennnot got out agein”®, About claan If T decided to romain ecroty elent. "At (his stago I moet make fur pointe about the formula- tion of the eontral dogma which have occasionally pro- Alvced mimmderstandings. (Seo, for exarnpl, Commoner: iis ror has boon possted out by Hleachinan and. on sore gonoral grounde by Hershey®) (1) Te saya nothing about what. the machinery of transfor in'mado of, and in pattcalar nothing bout tors.” (Tt wna asrumed tha, in goneral, the astray of {anafor ena high) (2) Te says nothing about control mechaniems—that i, stout the mate at whieh the processes work. {3} Te waa intended to” apply only, to present-day ergnniems, and not to vente im tho remote pasty wach a8 ‘the origin of lfo or the origin of tho code. (4) Tels not tho samo, at is commonly assumed, as tho sequence hypothesis, which was clearly distinguished fom i in the amo ssticlot In particule the eoquenco Ihypotheris ‘war a pontive statement, saying’ thatthe (Overall) transfer nuviie acid-sprotein did exe, whereas the contral dogun was a nogative statement, saying that {rensfers from protein did not exe. Tr looking ace T am etree not only by the brashnoss whic allowed us to venture powerful statements of & ery general nature, ut alko by the rather delieate iscrisination used i selecting what statements to make, ‘Timo has shown that not everybody. appreciated. ou restraint. 'So much for the history of tho subject, What of tho present ? I think itis clear that the old clasieation, Trough useful nt the timo, could ‘be Improved, and. smiggnt that the ine possible transfor bo regrouped tentatively into threo class, I propose that these be NATURE VOL, 227 AUGUST 8 1970 DNA 4. called goneral transfers, special transfers and unknown ‘ranafors, General and Special Transfers ‘A goneral transfer is one which ean oeour in all cells. ‘The obvious cases aro DNA~DNA DNA-RNA RNA-Protein Minor exceptions, such as the mammalian reticulocyte, hich probably Ideks the first two of thes, should not "A special transfer is ono which does not oseur in most cells, but may occur in special circumstances. Possible candidates aro RNA-RNA RNA+DNA DNA—Protoin At the present time the first two of thoso have only been shown in certain virus-infected colle. As far ax I know thero is no evidence for the third oxeopt in a special eell- free system containing neomycin’, though by a trick it could probably be made to happen, using neomyein, in an intact bacterial cell Unknown Transfers ‘Those aro the throo transfors which tho contral dogma postulates never occur: Protein-Protein Protein-+DNA Protein>RNA Stated in thin way it jn lear that the spocial transfers ace chon about which there isthe moet uncertainty. Tt ght indeed have "profound implications for molecular Biology" if any’of thes special trantere could be shown to bo goneral on—-if notin all cella atleast to bo widely Alstributed. So for, hoscover, then i no evidence forthe first two of thoes oxoopt in a ocllinfoctod wth en RNA virus," sucha oll cho contral dogrnn demands that at Toast ono of the frst two specie! transfers should occur {his statement, incidentally, shows ‘the power ‘of the central dogma in making thabretieal predictions. Nor, es Thhave indented, thers any good theoretical reason why tho transfor RNA_-DNA shotld not sometimes bo used, Thavonnever uggeted that it eannet occur, nor s far as T know, have any ot my’ ellegros, ‘Aithvgh te deta of the sieation hero sre plausible, our kmowiedgn of molecular biology, oven fra callie atone for the organs in fae itil far too completo to allow us to assert dogmatically tht ier comet. (Therein for oxannplo, tho problem of the chemical nature of the agent of the diseese scrape:NATURE VOL. 227 AUGUST 8 1970 500 the articles by Gibbons and Hunter* and by Griffith Nevertheless, we know enough to say that a non-trivial) ‘example showing that the classification was wrong could be an important discovery. It would eertainly be of great interest to find a coll (as opposed to a virus) which had RNA as its gonetic material and no DNA, or a coll which used single-stranded DNA as messenger rather than RNA. Perhaps tho so-called repotitive DNA is produced by an RNA—DNA transfer. Any of these would bo of the greatest interest, but they could be accommodated into our thinking without undue strain. On the other hand, tho discovery of just one type of present day cell which could carry out any of the three unknown transfers would shake the whole intellectual basis of molecular biology, 563 and it is for this reason that the central dogma is as important today as when it was first. proposed. Received July 8, 1070. 4 Nature, 286, 1198 (970). ‘emia; He A and Minuten, S Natere 886, 1211 (1970), ‘This artltecon- "aint I rtornces tobe edn’ eer work dating buck 10100, ‘Baltimore, Dy Natur, 26, 1200 (1070). See also the brit account of Sofi ent vo nage 8, OS “Grey BI Gin Symp, Soe, Bp le Bieloseal Repaton Ocrometican S18 95} v + commoner hy Nature, 280,994 (2068). ‘Mls, Pe Nature 28,90 0070) *Hrahey, A, Di, Nature, 298,603 1070), f afecar ad Helland, J, J Proc, US at Asad. Set 84, 80 (1005. ions, Re A ad Hunter, G. Dy Notur, 18, 1041 106). Geitth 3.8. Setar, 905,20 Characterization of the Products of RNA-directed DNA Polymerases in Oncogenic RNA Viruses by S. SPIEGELMAN ‘A. BURNY MR. DAS 1. KEYDAR 1. SCHLOM M. TRAVNICEK K. WATSON Institute of Cancer Research, Columbia University, and College of Physicians and Surgeons, 99 Fort Washington Avenue, New York, NY 10032 ‘Town's DNA provirus hypothesis', according to which the replication of the RNA or RNA tumour viruses takes placo through a DNA intermediate, explained the following “unique features of infections with RNA oncogenic viruses: (a) the heritably stable transformation of normal cells induced with these viruses; (6) the apparent. vertical transmission of high leukaemia frequency in reciprocal crosses between high and low frequency strains of mice*; and (c) the requirement for DNA synthesis® in the carly stages of infection. The hypothesis makes two specific predictions amenable to_exporimental test. DNA complementary to viral INA should appear after infection and therefore should be detectable by molecular hybridization. Suggestive but not decisive experiments supporting this prediction have been reported, Further, Temin invokes the existence of an enzyme that can carry: out a reversal of transcription. by catalysing the synthesis of DNA on an RNA template. Evidence for such an enzyme has been presented recently by Baltimoret and Temin and Mizutani’, who found a DNA-polymerizing activity in both avian and murine tumour viruses. ‘The enzyme was deteeted by the incor- poration of tritium-abelled thymidine triphosphate GEL-TTP) into an acid-insolublo product, that ean be destroyed by deoxyribonuclease. Maximal activity required tho presence of all four deoxyriboside triphos- phates and magnesium. ‘The fact that the activity is mhibited by ribonuclease implies that the RNA of the virion is necessary for tho reaction. ‘These findings are clearly pregnant with implications for the molecular details of viral oncogenesis. Their potential importance demands quick confirmation and extension, @ task the present work undertook to fulfil, Several RNA tumour viruses contain an enzyme that synthesizes a DNA-RNA hybrid using the single stranded Hybridization experiments confirm that the DNA strand is comple- mentary to the I RNA as template. al RNA. Wo report here the finding of DNA polymerase activity in all of the seven tumour viruses we have examined and establish by physieal and chemical charaeterization that the product is in fact a DNA hoteropolymer, Further, we show that the DNA synthosized is complementary to viral RNA by demonstrating its ability to hybridi specifically with homologous viral RNA. Finally, we find the expected nascent. RNA-DNA complexes in the reaction. ‘These have been detected and characterized in glycorol and Cs,S0, gradients and shown to be sensitive to denaturation procedures whieh disrupt’ RNA-DNA hybrids. Preparation of Viruses for Enzyme Test Ramschor murine leukaemia virus (RLV) was obtained as a ten-fold mouse plasma concentrate. Vicus lob RPV-HL-67-5 (infectivity titre of 340 log spleen weight enlarging units per ml.) prepared from CifW-S mice was used. All procedures following the original thawing of the plasma ‘wore conducted at 0°-4° C. Plasma was first clarified at 16,000g for 10 min. The resulting supernatant was layered ‘on & 100 per eent glycerol cushion and centrifuged at 95,000g for 70 min, ‘Phe material obtained on and just above the glycerol cushion was then layered over a preformed 25-50. per cant sitcrose gradient and contrifuged at 95,000 for 3 h. ‘The resulting virus band (1-16. gjern® was diluted in 0-01 M ‘Tris-HCl (pH 8:3), O-1 MNaCl, 0-002 M EDTA buffer (INE) and recentei: fuged for 2 h at 95,0009. ‘The resulting pellet was resus: pended in TNE and assayed for protein content. A similar procedure served to purify RLV harvested from SLS.VS tissue culture supematants grown in our labora- tory.