On the Translation of the Genetic Code F 2 MARSHALL I NIRENBERG National Heart Institute - National InstituteS of Health /+ ae poe vod dige to take tare oppurtun hy vo. relate some thig. Cale - 2? No wed Or Know Lease of the genetic language. | For some two to four’ v a billion years some such Language has probabd.y provided the ba sis for a continuous dialogue between ceils and their descend- , cv naamscogern UE at fr fa ants. Fossil records aff bacteria. about 3 billion years etd~- y ‘ave been reported (ey Baghoorn’Schopé) the first vertebrate appeared approximately 500 million years ago; ana amphibians and mammals about 350 ana 180 million years ago, respectively. seat The presence of bacteria 3 billion years ago nay-indicate the presence of an operational code at that tinep Almost surely “ne code nas Zunctioned for more than 500 million years. The oy remarkable similarity in codetwords used in bacterial, amphi- bian and mammalian replicative processes suggests that most, if not all, forms of life on this planet use almost the same genetic language, and that this language has been used, pos- sibly wich few major changes, for at least 500 miliion years. 7 a s it is by virtue of this language that pach generation is e to pass to the next generation a library of information t- ab which specifies in detail how to make the many kinds of protein catalyst) that the cells will need for theizx development. And on Le . ai. thous t now seems clear that all, or aimost all.forms of 4 r-' ife on this planet use virtualiy the same language, ecently a number of dialects" have been found. I shall describe this Geer. Ipea 1 {tae : 4 + *4 = The elucidation of the genetic code has been “he subject / 7 ins of much intensive work, particularly in the past four or five years, and I wouid sixe to stress -atuthe-oursety, that this work, ane-particularly-the work with which I have been associ- o . tet . a e ated ‘nas been, in a very real sense a collaborative project. ? . Zé : 2 ‘Here ig. show de wy lg Yee nM of - pfoby mA beep ves nied Pol” an a ail a, f Na ef TE ; ‘schematically the aobte- -soranded Sir vist -egastes wit be i “* oe foot v enzyme,- RNA~polymerase ,+ cataiyses the synthesis of messen er yn A. . ; ae 4 ; Bf RNa, osing the DNA as a template. Only one strand of twe DNA wo is copiea by Sse RNA-polymerase; amd the copying process is se- quential) and erere-are signals, whose exact nature is unkown, ok amb Witeeh specify she begir naing and the end of ‘the messeag#Z-RN ae e eRe i ‘ gytithesic. | She next diagram (Fig. 2), shows. schematicallyy the process of protein synthesis. In the DNA shown here, the dir- ferent cross-hatchings represent various segments of DNA, each “eg corresponding tc a specific protein,’ or group of eroteins. Ri- = bosomes are shown,schematicaliy, attached to the messenger=ANA — 1 o bist of collaborators at Bethesda. ee rs Fae basic features of the JOwick- Po. Sem ie ett we TL wacle reading, or translation, begins} amr as soon as one ribo- some moves down the MeSGeRSOLARNA, another becomes attached un- cil the messes alae i is EN covered with [a gomes Bren, ee n LF MND in pgly rater Ce | Tae aettar rea ding Yaleowplice” @ ean e% RN . (soluble-- L. maA}} which ee speciiic amino¥acids and recognizeg parti- t cad BA Dy cular m¢RNA code words\en_the-vibesemes. Thus the codenwere, ¢ é “OF codo..f is recognized not by the amino- acidg, per-se, but by Aire ete Am a ME OS IN I coer coleaiese-the $HRNAD- _y+-—--Fig..- illustrates, again diagramatically but in more cetai 4 ae ty the codon recognition process as exemplified by that most inten- & Ce be . sively sfudied organism, oti... fi Tt f f. : th oe pe, AR “gt, Cron i Pee MA tg al. 4 Tae _gibosome’ of Beeehi cbiptisexyewo su sublunits: the larger. 305 aad tae smailex 30$- The messempewRNA iies on the smaller the Cart & {tens ‘perc of the ribosome, @ and presumably [three bases)in 1 the ACh ot ot ek ARE te ge MomeerpeteeRNA molecule te—-eeden!) are > recognized by{thred, Basel 2 can then bind . ahecrae Pomel | fas) : fat one of two possible binding sites on the Larger ribosome’ sub- unit, a@-particular amino-acid, (aa). One of these binding sites . ae 70s . is for the peptidai S+RNA, so the STRNA which is at tached to the growing (protein) Pee reneae. molecule; and the other) for the incoming amino-acid s}RNA,. hus the ree _ enzymes (Sae—two-s-RNA ls Ghost a, i ~ ore Bae oh plustict Ph which supplies tne acti- 2 vacion energy, are cequired for the transfer of the growing poiy- e. pepcide chain to the next (incoming) amino~a eid S+RNA complex, Wnen this is accomplished the SrRNA required for the previous &; YALE howe’ amino“acid is discarded, and a shift in-seme-way occurs so that the next codon (triplet of bases) on the m#RNA can be recognized 7 fT by 4 new stRNA. In this way the protein synthesis starts at a tebe are ran Py we . ae a RE cmt Ph ARS ye ee 4 aoc wore - . given place (on the m+RNA) , reads groupings of three bases se~ quentiaity and with a given polarity. Fo = f “ : \ oe" ols Co i ‘ “ in an accual living cell, even “thre~smettest bacterial’ ‘cell, trey, me Cty ae copumowetsoe~byochemical ai processes axe simultaneous ly, in-process. all part of the cell metabolism. The Synthesis of even a single Bes a € protein is quice an elaborate process involving, Ninter-alia, the cru.csfer of a long DNA message to an m+RNA molecule which Yas cypneaily suffic tent nucleotides (about 1,500) to code seme ': he cee Pe fe Pope ea aminc~acids for the-protein. polypeptide chains, Moreover, in an actual celi these 1,500 nucleotides will not be arranged at " a zn any cae sequence, refie cing ene ta ‘that there is a hog ean oad tact, jaan on eS qe ut “ ’ aeons Ly, Cog pit Et So away great number o£ different sequences o£;amino-acids ‘(of -whieh-20. i, hes va tovfvaclid, oo . . different—varieties.-are ) whith constitute different proteins.. Nonetheless,oy a gveat' variety of biochemical and genetic LK investigations, especia ity with bacteria and viruses, a=g@eeat N “many features of She protein synthesis, f includi ing { an particular|y el i ry sea@n. information sbout the code, nas been obtainedY ine work f snall Me cescribag is, however, characterized by the use of much Simpier, in vitro systems, where the essentially chemical feacures QD of some of the basic stens in the whole process are studied. Th , the success of these methods, « and the concurrenc. of, vesults £23 fiem-with those from in vivo experiments, where—beth-are- avai. 5 SRR cemenmeemicinetiine ape wiil I hope demonstrate how a physio-chemical or mole- cular basis can be found for the base processes governing such fundamentally biological phenomena as cell metabolism and repli- cation. The basis for our earlier work on the DNA-RNA code was the use of synthetic messages,’ (in place, thec is, of actual m+RNA) which were randoml % oriented sequences of the four code letters, UJ CA Lae . aga cdl } ; cyfoes e), _ACengne) Gtpanine) , the four bases AGS Z ow of m#RNA. In this racteristic of the code could be “ha - ee t ~ Cerexminedey inp router. the. base compositions Of dae code- *. to.y 1 Pk x. * Le fit 4 Tek i fw. a words, Dut not t Sequente-pi-the-bases ia-the. in the 4 words. Thus the orobien/ up to two or three years ago was like Fi that of an anagram: we Knew the letters comprising the codewords but ndé¥ the order of the Peteers within each word. hs i Et~hes beennweke established’ i in several laboratories that tole ct-one~edded’ 4 syntheti c messenrger=RNA, in particular polyuri- Ce - Af dah dylic acid™ @ sy nthetic RNA with entirely U bases) ¥’to a suit-~ adie mixture of ribosomes, StRNA's >» enzymes, ATH, G GTP and amino- a acids Ses the poly-~U vweebd selectively bind® phenylalanine s+RNAY/ (l.e., the particular st RNA associated with the incorporation of + amino-acid ohenylalanine in protein),. to sae ribosomes. My coi- league Philip weaer and f then speculated sow small a message (c~ che RNA type) would direct the binding of s+RNA to the ribo- Some, Experiment showed thac only three bases were needed, chat is, very small molecules comprising only the triplet itsel¢ would direct the binding of of the appropriate amino-acid StRNA to tne ribosomes. This provided a,rather Simple route towards the determination of the sequence of letters in the RNA code-° words. Our main problem was to devise Suitable techniques for Syn- thesizing triplets. At the time we started our work with such triplets, methods had been reported for making some 20 or 25 of the 64 (=4) triplets which can be constructed from the four d nucleotides U, Cc, Ay G. These had been prepared by enzymatic breakdown of RNA, ox by chemical Synthesis, in the latter case using some of the very elegant techniques devised by Khorana and his associates. Iwo general techniques were developed in our laboratory, the first by Leder, Singer and Brimacombe, and the second by “erton Bernfieid. The first employed polynucleotide phosphory- L s e ey _ OA Tee ca = A " 4 \ te A get Pa ve . 1ase £522 enzyme Ach Ahr jul prcetailiis, tt padi, >, ? of . / ° | a a ead coe ; _s en de 4 : [ry Log. un pomGewl3- Single nucleotides to ditnucleotides to make Fig. ? 7 trimers , tetramers, pentamers, ete. The second method em- t eh, a] ployed the enzyme pancreatic RNA-ase, w nough normally pa tt ry pale a breakdown or degradative enzyme, will also catalyze an ex- change reaction between polynucleotides and car be used to make . . “ . ee Ls . tripiezs with well-defined sequences. Using the methods of ‘ ee Khorana and these two enzymatic tecaniques, it was possible to Synthnesize amost all of the 64 reap tees In connection with the use of thie small polynucicocice or oligonucleotide" molecules such as the trinucleotides, it is important to. point out BReme that any given sequence of nucleo- tides can exist, when incorporated in actual m¢RNA in three chemicaiiy distinct forms, depending on the location of the se- quence in the wacle messenger molecule. The chemical forms re- sate to the three positions (a) as an internal codon (trinucieo- tide) ox as one of the other of the terminal groups - so called t w 3'-terminal codon and 5'-terminal codon. This is illustrated hy n Fig. . e Fig. fie ft. - All of the evidence te-date suggests that the biological char- a&cceriscics oF codon recognition may in some, perhaps in many, a wh AE cases be influenced by the particular position of the codon in che m#RNA (or equivalently in the DNA). Thus each of the 64 triplets referred to above thay exist in three effect ively dif- ferent structural forms. tae significance of these ‘'secondary" chemical features is of en) indicated by exzeriments / in vitro, witn the o.igonucleotides, The heiical RXA (or DNA) has a definite sense or direction - with a cefinite "beginning" and a definice "snaing', 3! and 5! refer to features of the chemical structure at these espective cexyminalis. = eS Ma ay _ t a ? * and specifically by studying the influence of various (phos- phoYiacing) Suosctitutions on either the 3' or 5' terminal hy- droxyi groups of the sugar in the trinucleotides. Thus Fig. Fig. Shows the binding of phenylalanine s+RNA to ribosomes as a function of the concentration of the trinucleotide. A simple ¥ we triplet, UUU, has an activity shown by (a). I one adds a ob. pnosphate to the 5! hydroxyl group te the Sugar the activity is greatly inex “eased, i. e., the binding or template effectiveness r of the trinucleotide is greatly enhanced;”"(b) . A phosphate at- tached to the 3' terminal lowers the tempiate effectiveness, (c}). Recently, Fritz Rotman prepared some analogues of UUU Pettey ClpELG. hase with a methyl group attached to the 5! phosphate, and also, wich & methyl group attached at both terminals, i.e. both 5' and 3! phosphate. The methyl group at the 3' phosphate terminal great- iy reduced the template effectiveness. A triplet with 2'; 3! cyclic phosphate shows very little template activity. it seems possible that'significant terminal variations of this sore may occur in different biological circumstances, and that ota? rane Poa chesefmay possibiy regulate the template activicy of the codons. For example, the terminal hydroxyls of the sugars (ribose) may The bincing of che StRNA to the ribosome is determined by tech- L i Gioactive tracer is incorsorated in the SPRNA, ivity associated finaily with the ribosome of this binding. It is in thet the term |. enoces the effectiveness o2 che binding. pbk oo 5 fe he be modixied in such a manner. Certainly a substitution at the 5'-terminus may be important because this could furnish a Ssig- nal which specifies the attachment and/or the detachment of the ribosome from the message, (m+RNA or substitute}. Recently Mitra and Hurwitz, and also Stent, have shown that, in vitro at least, méessenges-RNA contains a triphosphate attached to the terminal nycroxyi; and aithough it is not clear what physiological func- tion this triphosphate serves, it is highly plausible that it may in some way specify the initiation of reading the message. It couid aiso determine the first (three letter) word to be % reac, phase the reading, and, perhaps affect the susceptibility “GO enzymes thac could attack the termini of the messengex-RNA. ancernal codons may also be modified by these secondary chemical changes; the 2' hydroxyl or the base could be modi- fied and such cnanges may be relevant to the punctuation of the faa he bt mie, £ a message. it-a@iso- cannot be excluded that the codon recognition process is in some instances affected by the particular neigh- bors of that codon on the message. Cus. at Lt-should also. be pointed.out that there could ‘poseibty be a difference between internal initiation and termination (i.e. initiation or csermination of polypeptide sequence (5 rotein) by & codon interne..y locazec in the message) and serminc1 initia- ction and termination (the same Process effected by terminal co- Gons). Consicer the situatio where. ‘the meseencves<"0/. appears Tes) \ LO to contain the information for the assembly of more than one protein, (or more than one polypeptide chain of « protein). If one starts to read (from the left in Fig. ) the codon for Fig. the terminal initiation, one then reads in the message unti.z one veaches the word that says stop") and chen trere-wiii-be an unknown mecaanism See starting the second message at an in- 2 terval position. It seems quite piausible, although not known, that chese terminal and internal initiation and termination me Q 3 % ry a ” rH Nn Q ould be different --possibly different codons. Anotner feature of codon recognition concerns tne degener- acy of the code, or the existence of synonyms, i.e. different codons which code the same amino~Acid in the polypeptide se- quence. With che appropriate oligonucleotides, one can examine, t.. in vitro, the effectiveness of different synonym messages in &£ binding the particular amino- acidg : s+RNA! s to the ribosomes. The results of such are illustrated in Fig. . For example, a Fig. phenylalanine SP RNA responded to both the oLigonucleotides UUU and UUC, but UUC was slightly more active than UUU. Sinlarly fy Lysine+ £s¢RNA xesconced to both AAA and AAG but hese-cthere-ts Tu 2a = - - a * A ek at ane rence in the templace activity betweem the tua we gauke vo kede two synonymg, Tne first of these degeneracies, that between the wna a bh o t -- : te (smaller) pyrimicine bases C and U when they occur* as, third let- ter of the cocon,is universal throughout the code. The second Potes. af wept hoe sf? A Cex tp itae deg cneracy, @& the (large) purine bases A and G in f coir "+88 occurs in all but two or three words (c.f. Fig. ) we turn now from these refinements end cetailed feacures oi the triplec-binding method to the actual results obtained rocecure. Shmee the triplets have a weil-defined se- ee. val quence or nucleotide es 7 there are 64 possible ~qmh t tripl ets} asa) . ‘fe we have synthesized 63 of these and determined the amino-acids which they code The results are summarised ) in Fig. “ “* 2 APs : Pe a Se aR FR Fig. 2 ERS “4 The astezisks indicate base compositions oz codons which < . A o - o 7 * a # 77 . were cetermineda by directing protein synthesis an Eecoli- extracts . . hbo as with syntnetic randomly-ordered polynucleotides. jt is clear. caat—there is~a-very..close/ sexsespendenee with the results of “HE earlier work. it is interesting to notice the types of syno- nyms which occur (some of which have aiready been mentioned). jail - “3 qed ; vf poe cua uutene yet Sies_glutamic acid eetincopcndtentn | ‘the codons” CAA and GAGF- an ex- Se sre meent ey ermine Te mn ie “ CC is tay vet se ee &, af wd ampie a ‘A=G degeneracy in the third Place. bikewise Lspar ahd coclenso, woth. acid and; ea C }-eorrespondina-te U=C degeneracy in the third piace, Another. ctype of degeneracy is illustrated by Threonine sirconine | 4 hird bas) ct wnich is coded sy AC and any os the four U, C, A,.Gi place, —Besakos ine, on the other hand, is one of the rare cases (irypt Ops> may be another) in which GReec—ee 0 cord piace ce- 12 Oe LD ee Le Lol racy AUG gexicsenfee but AUA codes for Asoleucine. yen Mead 03 (oO “his degeneracy of the code can have many consequences, One of the more obvious is the possibility of a great deal of wat k a Sage ohh tae oy OF polio. tah me "silent" mutation, that is fon one of the code-words, or grouss wets ‘ of synonymous coce-words, there may be CONVCY Saga .0f~a~—bage--j.n- oy the—third-positioa to anotner base without resulting in an amino-~ . . . - . a, acic rep.acement. Another obvious conclusion is that amino i) cics which are very similar chemically, such as the dicarboxylic acids ( aspartic acid and glutamic acig, have closely related co- 2 Gons. This may reflect the evolution of the code, but whether or not this is so, one consequence would certainly be that when an error in repiication does occur, usually the first two bases are read correctly and the third one incorrectly. And very of- cen the result of an error in reading will be the substitution pe nitenn : q, in a protein of a chemically reiated aminoacid, thus the general picture of the code igs that it is quite conservative-- in the sense that it usually minimizes error or the consequences of er- ror. The various patterns df Synonym codons are summarized in Fig. - <(N-formyimethionine S#RNA shown here is the initiator), ta] ig +S. g in addition to the 0 odons for the specific amino“acids, ‘dere-as~as.has-been-mentioned earliex, some code-words Sneek. b's “ppear to serve special functions ("punctuation" etc.). For ex- ampie, the recenz work oi Brenner, Garen and Zinder, and of others, ae indicaces chat UAA and UAG may indicate the end of a message - although the precise mechanism for punctuation is unknown. 1 ULG, CUG, AUG and in some cases GUG may specify the initiation bh © 4 fo ry * of a message. Our recent scudies, and also those o and Marxer in England, have indicated that these codons - at 1east when in terminal positions ~ are recognized by formylmethi- onine and this may serve as an initiator of protein synthesis. Some possidie szecial function codons are listed in Fig. . 4 nh At beg ‘) fos! Sanger ,first * Observed imiengli that one of toe two. saN ra et a og yee w a * yr 8 “Pocmyt Stoup: x WS that is tne -amino “group~—of bee - methionine, after the methionine- a wo ' eS oa + LU ie COROT gy was Tinked-to-the-s+RNA-ecoutd-be fomaylated. The work o£ Capec- chi and colleagues, and of Zinder, kawe suggested that this may mentioned HI specify initiation of Message translation. And as already, UUG, AUG, CUG and to some extent GUG are recognized by Kom formylmechiensae WPRNA; also that UAA and UAG may serve as ter- minators. It also appears likely that the words AG F wistin ending» he ~~ o JU, ©, A or G may also serve aS speciai function words; but s€= 9.5 Aa these functions have not sé-- -farc been found. The present situa- at pote tion in this field is e most interesting one, in that.the neces- sary tools for deciphering che special functiozx WOLGS _G52~£6 sccm, wed, and ic should soon be possible to understand more about the mechanism of these special words and the sole they play in pro- tein synthesis, be oi E=-would aike-to turn now-to- a variation of the triplet- Ear cg el Be binding methoc,- whi ch throws. further \etgisteont the coding mechan- ism. D. Hatfield has recently prepared some radioactive trip- Lets / Cin the earlier experiments it was the StRNA which con- tained the racicactive tracer), “and has studied the binding of these tripiets to the ribosomes in the presence of the amino- acid S4RNA. Fig. shows both the binding of the triplet and of the S+RNA (here phenylalanine S#RNA) to the ribosome. Fig. Asean 5eseen, in the presence of the appropriate triplet polynucleotide phenyiaianine s#RNA binds to the ribosome; in the absence of the s}RNA very little triplet binds to the ribo- some. Because of this, in the presence of the s4RNA both the triplet polynucleotide and the phenylalanine s+RNA bind to the ribosome 4c approximately the same rate. Thus the compiex on cne ribosome may well be a one-to-one association of triplet and XNA. this technique provides a very simple and quite sensitive method for detecting codon recognition by s7RNA which is not t hea a } acylated with amino-acids. Thus some special function words may not be recongized by activating enzymes, s+RNA's, which are e wo x . n > : ™ : : Owes . 4 Bue . ; po tee : | ek wd 5 if : 2 we OTE OS NO et ee vg is “f ok ey Joe fhe alt ee : e not acylaced, and this method would provide a relatively simsle route towards detecting such recognition. We have also made investigations (in collaboration with B. P. Docter and Waicer Reed) with purifies s+RNA fvactions, i.e., media containing essentially only 4 singie type of s7RNA, de- ah Tees ra , ; rived from Esecoli fractions. We find that ycosine-sfRNA re- cognizes both UAC ana UAU, which again exemplifies the C=U de- generacy in oJ the third place. (There are two types of Tyrosine= ° t da S#RNA, difi rh er in ; both types recognize UAC and UAU.) Si milarly Vaiine/s?RNA recognizes both GUA and GUG (G=A degeneracy) but the GUG to a much lesser extent than GUA. The Ercoli fraction leucine-1-s-RNA and Leucine-2-s?RNA both xecognize the leucine codons (UUA, UUG, CUU, CUC, CUA, CUG). Recently, however, J. A. Carbon has reported that in mammal Liver one species of Leucine=s~-RNA preferentially recognizes AAG, and the other pre- ferentially recognizes AAA. There are also types of Leucine-s#RNA which recognize CUG, and others which recognize UUG. The major variant of methionine~s-RNA which, as mentioned previously, wili accept a formyl group recognizes UUG and CUG, Joe Tay out a less prominent methionine+s+ -RNA recognizes AUG preferern- tially. Uikewise there is a Mryptovhan s#RNA which recognizes UGG, CGG and co a smaller extent AGG. The pattern here is clear: a close relationship between U, C and A in the first place of 16 tne coding criplet. R. nolley, working with purified fractions cecal Z C2OD of yeast s?RNA, found “alan ine=s#RNA recognized “@é,GCC and CCA -- again the group U, C, or A but now in the third place of sane coding tviplec. It should also be pointed out srominent binds to ribosomes very weakly in response to the ucleotice triplets: it is possible that this tyve of weak re- cognition involves only two of the three nucleotide basis in the * This work with pure fcactions = such és +e élanine~s+RNA prepared from yeast, can afford some further insigh iacevaction witn the messenger would be a week interaction: ~ ? buc it is also quite possible that a U or C in a terminal po- sition wouid noc greatly inhibit the interac A metal group on a 2'-nydroxyl ceoxyribose (sugar) mighc also result in 4 weaker interaction, and furthermore, by permitting a greater freedom of motion on the ribosome, such a modification might resuit in greater ambiguity, i.e.) lower specificity of These resuits with infrequently occurring (or “trace'y bases, and particularly those with Inosine, cohen St rongly sug- gest that s#RNA may be modified enzymatically, _ after it is xre- Leased from the DNA template (where it is assembied in the cell). Since the level of "trace" bases is quite high in an actual cell, it seems Likely chat there exists a whole spectrum of interme- diates, stRNA's in various stages of successive modification. the consequences of this are f£ “r. easy to visualize. For ex- “A =a ° ~ . s oN ° a a a > ample, if an acenine(A) in stRNA is de-aminalid and so converted > + » which would normally recognize the Gri- oy dylic acid base in the message, would now be revlaced by some- into an inosine(i), the thing (the I) which can recognize U, C or A. Simila: intercon- versions would result from the desamination of a C or the conver- Sion of G to i. It is possible, althoucn perhaps rather pre- Mature LO speculate, that this type of interconversion plays an umportant tiologiecal role. There nascextaialy-beem a great deal 02, Work [recen em ve Cuarhce b K Lure vida of ¢ whey Ate a, _Wietekt sugzests vnat, £t ,is possible in actual cells, fesome 7 yea ON ee “ wey—te modify che specificity of codon. recognition and—this is—eertai. ~taveprofound biological consequences. An example of this is the effect of the anti- biotic screptomycin, tt—has—been—shown,—by Davis, Gilbert growed ana Gorini +ychas streptomycin will bind on to the 30+ 8 part of the ribosome (the small subfunit) , and all the available evi- dence suggests this binding of @ie- streptomycin to the ribosome may in some way cistort the topography of the codon recognition site so thet greater ambiguity in coGon recognition results. This (Gna y oe one mechanism . cage reater degree of error in pro- S \iia y tein Sell ee ee to ae ~eason-—-te-aeeount™for-the action of streptomycin on bacterial celis. tps wee ee eee 7 | There are other exampies{ dn addition to streptomycin, of i = ae the modification of the specificity of ‘codon ~ recognition. Aree f -aeent, comparative study /he R. Marshall and T. Kaskemadt i*= of the specificity of codon recognition with StRNA from amphi- Ge iy bian, Xenopus ~aevus / siver, from guinea pig liver and from Ee ee A SSI coli. Eecolli exgiaineds?RNA does not recognize AGG and recognizes ete eel, eR et Ay > CGG only very slichtly, whereas *-r both amncithian and mammalian = : an meat 7 > fb GQ & The contrast setween alanineés{RNA's from yeast, mentioned cax- lier, anc Eecoii is also shown in this diagram. Ata both amphibian Liver aan a uinea- ~pig Liver’ GCG is a very 2 See seam ae a eaatts ct nee ee ee tt ef . “rt active codonl whereas, 2 the the amphibian liver SCG has no activity an . Spe for alanine¢sf- WAY. This contrasts with tne activity for seco? oem? hs alanine-sfRNA. + ~~: fe show cipal As Gee ks. J seust-—t-have-shewa;—by the examples I-heve-priefly- ~sketensd; how some features of the complex machinery for protein synthesis in celis can be studied by means of relatively much simpler ae Thus it has been established that the same sequences of three nmucieotide bases ddge the same amino-"/ acids throughout the whole range of organisms, from bacteria to mammalian Livers. And this universal code has been explored by molecular biochemistry in vitro. Zowever, we have seen that there aze seconaary features, such as tne se_ctive responses to cifferent synonym codons, and the subcie modifications of the s+RNA's waich can be of great . ww f. cae he. importance in actual, complex living organisms. Features suchas 23 say pséy important biological roles; by selectively controlling nes a bia UO: (b wae vate pxocein syntnesis they may be an important factor wi the general process of ceil differentiation. These are cer-~ tainly problems for the future. Finally, I would drew-attenttorr-to-the-ace that even, in vitro, at its simplest, the whole detailed process of coding in protein synthesis - involving DNA-m}RNA-s#RNA-ribosomes, activa- ction enzymes, ATP, etc. is far from fully understood. Even the basic underlyiny questions - why, for example, does a triplet o Q eo) Pu @ Oo Kh (+ ry? rn n 1 ¢p) Oo He ct @m xist, why should not phenylalanine instead of a2.anine corsespond to GCU and GCC? Is there a basic chemical teason for this, or is it to some degree a mattex of Caistorical) chancel My personal belief is that there is an underlying mean- ing for this and chat it will be found.