3 8 7
THE CHEMICAL BASIS OF HEREDITY-
THE GENETIC CODE*
Department of Biochemistry
New York University School of Medicine, New York
'ORK of recent years has thrown considerable light on the
way in which the genetic information stored in DNA**
Ri is transmitted to the protein-making machinery of the
cell, so that a specific nucleotide sequence in DNA gives
,rSSE5.5ele52 rise to a unique amino acid sequence in the polypeptide
chain of a protein. It may be said that DNA contains a coded message,
the genetic code, with instructions for the manufacture of specific pro-
teins. Wie know now that RNA participates in this process as a mes-
senger1' 2 between the DNA and the protein. DNA directs the syn-
thesis of a specific messenger RNA and this in turn directs the syn-
thesis of a specific protein.
The relation between DNA, messenger RNA, and protein is illus-
trated schematically in Figure I in which portions of the nucleotide and
amino acid chains of these compounds are shown as strips of recording
tape. Moreover, under the assumption that the genetic code is a triplet
code, the continuous sequence of nucleotide bases in DNA and RNA
is represented as a sequence of separate triplets for better visualization
of the correspondence between the individual base triplets of messenger
RNA and the various amino acids. The amino acids on the protein strip
composition, but not necessarily the sequence, shown on the RNA strip.
An account of the experimental work on which this correlation is based
I are known to correspond to the triplets having the base
*This article is based upon a lecture given before a meeting of The Rudolf Virchow Medical
Society in the City of New York, held at The New York Academy of Medicine, November 4,
1963, and upon the Paul Karrer Lecture, given at the University of Zurich, Switzerland, June 26,
1963. The lecture has been previously published in Experientia,
here by permission of the editor and publisher of Experientia.
**The following abbreviations are used: RNA, ribonucleic acid; DNA, deoxyribonucleic acid; A, C,
G, U, T, the bases adenine, cytosine, guanine, uracil, thymine, also the corresponding nucleosides
or nucleotides; in the case of polyntucleotides, poly A means polyadenylic acid, poly UG, poly-
uridylic-guanylic aci(l, etc.;
5'-diphosphates of adenosine,
5'-triphosphate; GTP, guanos'ne 5'-triphospbate; T.MV, tobacco mosaic virus. The amino acids are
abbreviated by using the first three letters of their names
asp, aspartic acid) with the exception of asparagine and glutamine which are abbreviated as asn
and gin, respectively. In the shorthand writing of polynucleotides, as in pApUpUpU
the letter p to, the left of the nucleoside initial indicates a 5'-phosphate; the letter p to the right,
1964 and is printed
i, hyypoxanthine; poly I, tsolyinosinc acid; ADP, CDP. GDP, UDP,
cytidine, guanosine,uridine; ATP, adenosine
(e.g., thr, threonine; try, tryptophan;
Vol. 40, No. 5, May 1964
catalyzes the DNA-dependent synthesis of RNA, is the key enzyme in the transcription
of the genetic message from DNA to RNA (step 1). The code reald-out step (step 2)
can be considered as a translation from
the fouir character language of RNA into the
twenty character language of proteins. nm-RNA stands for messenger RNA.
1. Two-step transcription of genetic code. RNA nucleotidyl transferase, which
is the main purpose of this lecture. 'When the base composition and
sequence of the coding units for all of the amino acids in proteins is
known, the genetic code will have been deciphered. \Ve may not be
far from this goal.
CODE READ-OUT IECHANISAI
RNA nucleotidyl transferase, the enzyme catalyzing the synthesis
of messenger RNA with DNA as template, operates through a A\Vatson-
Crick base-pairing mechanism and yields, as shown in Figure I, com-
plementary copies-or negatives-of one of the DNA strands.3 In this
first step the DNA code is transcribed into an RNA code represented
by a unique sequence of bases. In what follows the words genetic code
or amino acid code will be used to denote the messenger RNA code.
In the read-out step the messenger RNA serves as a template for the
alignment of the amino acids in a sequence prescribed by its own base
sequence. For this purpose each amino acid is linked through its car-
boxyl group to one end of a specific amino acid acceptor or transfer
RNA (s-RNA) of small molecular wveight (about 25,000). This reac-
tion is catalyzed by a specific enzyme which recognizes both the amino
acid and its s-RNA. Energy is provided each time by the splitting of
one molecule of ATP to AMP and inorganic pyrophosphate. There are
as many amino acid activating enzymes (aminoacyl-s-RNA synthetases)
and at least as many species of s-RNA as there are different amino
acids, i.e., at least 20. The coding triplets of messenger RNA are recog-
nized by complementary adaptor triplets (e.g. AAA for UUU) in
s-RNA.4' " The adaptors are believed to be responsible for the specific
attachment, through a base-pairing mechanism, of the amino acid-
bearing s-RNA to positions prescribed by the base sequence of the
Bull. N. Y. Acad. Med.
3 8 8
THE GENETIC CODE
3 8 9
UUG AUU GAG UUU
--------UUG AUU GAG UUU+
UUA ...................... UUG AUU GAG UUU
AUU GAG UUU
Fig. 2. Scheme illustrating the assembly of polypeptide chains andmessenger read-out.
The free amino group of the amino acids attached to s-RNA
visualization of the polarity of chain growth. The s-RNA is represented as a T with
the adaptor triplet on top of the horizontal limb and the amino acid linked (through
its carboxyl group) at the free end of the vertical limb. The rectangles represent the
ribosomal binding site. In the diagram the polypeptides are assumed to be assembled
on the ribosomes as they move along the messenger RNA chain but it is equally possible
that the messenger itself moves over the surface of static ribosomes.
is shown for easier
Fig. 3. Schematic model of polysome function (Courtesy of Dr. A. Rich).
Vol. 40, No. 5, May 1964
messenger RNA which, on interaction with the ribosomes, becomes a
template for protein synthesis.
As shown for hemoglobin synthesis68 the polypeptide chain
assembled sequentially from the N-terminal end. The s-RNA is released
stepwise from the amino acid (a) last linked to the growving polypeptide
chain, as a newv aminoacyl-s-RNA (b) is placed on the next position of
the template and a peptide bond is formed involving the acyl group of
a (with cleavage of the aminoacyl-s-RNA linkage) and the amino
group of the amino acid carried by b (Figure 2).
Recent work has shown9'1 that during protein synthesis the ribo-
somes form aggregates (polysomes) which are apparently held together
by messenger RNA for they are dissociated by ribonuclease. These
observations have been interpreted as meaning that several ribosomes
are simultaneously engaged in making protein on the same messenger
as they move along its chain or as the messenger moves over the ribo-
somal surface. This is illustrated diagramatically in Figure 3.
As discussed elsewhere"2 it has until recently been a matter for
speculation whether the ribosomes deliver ready-made protein or only
the unfolded polypeptide chains, i.e., the protein's primary structure.
The demonstration13 that ribonuclease and other enzymes can assume
their native globular conformation in aerated buffer solutions, following
unfolding through reduction of their disulfide bridges in urea solutions,
proves that the secondary and tertiary structure of proteins may be
formed spontaneously and are determined by their primary structure,
i.e., the amino acid composition and sequence. Folding of the poly-
peptide chains might occur partially on the ribosomes and be com-
pleted after release.
Another question of interest is the maximum size of individual poly-
peptide chains manufactured by the ribosomes. There is increasing
evidence that native proteins are aggregates of subunits of relatively
small size. The molecular weight of the subunits of various proteins
(hemoglobin, apoferritin, proteins of plant and bacterial viruses), from
I7,000 to about 25,000, would correspond in the upper range to chains
of 200 amino acid residues. For a triplet code this would require a
messenger 6oo nucleotides long (molecular weight about i8o,ooo).
No small subunits have been detected in enzymes. How\vever, several
enzymnes and various other proteins are known to be polymers of two
to four units each ranging in molecular weight from 40?000 to 2s,010
Bull. N. Y. Acad. Med.
THE GENETIC CODE
It appears entirely possible that these units may in turn consist of a
number of primary subunits. If so, most proteins would be formed by
the-spontaneous or catalyzed-aggregation of small subunits produced
under genetic control by the protein-making machinery of the cell. In
some cases the primary subunits might aggregate first into larger, sec-
ond-order units which would finally polymerize to the native protein.
Messenger RNA, like DNA, has only four different bases, whereas
proteins may have up to twenty different amino acids. Therefore, at
least three bases (triplet code) would be needed to specify one amino
acid. Combination of four elements three at a time yields 43 or 64
triplets, more than enough to determine 2o amino acids. A doublet code
would not supply enough coding units, since it would yield only 42 or
i6 doublets. It appears, therefore, that at least a code of triplets is
required for transcription of the genetic message. Genetic experiments
of Crick and co-workers14 have provided elegant evidence in favor of a
SYNTHETIC POLYNuCLEoTIDES As MESSENGERS
The question arose whether synthetically prepared polyribonucleo-
tides would substitute for natural messenger RNA in cell-free systems
of protein synthesis and would promote the incorporation of amino
acids into proteins. If so, any correlation between the base composition
of such polynucleotides and the nature of the amino acids incorporated
might open the way to the deciphering of the genetic code. This proved
to be the case.
The key to the experiments with artificial messengers was the
enzyme polynucleotide phosphorylase discovered in our laboratory in
1955.15 This enzyme catalyzes the synthesis of polynucleotides from
ribonucleoside5'-diphosphateswith liberation of orthophosphate. The
over-all, reversible reaction catalyzed by polynucleotide phosphorylase
may be represented by the equation below, in whichP.Pstands for the
pyrophosphate moiety of nucleoside diphosphates.
n Base-ribose-PP (Mg++) (base-ribose-P)
In the direction to the left the enzyme catalyzes the phosphorolysis of
polyribonucleotides with formation of the corresponding ribonucleoside
diphosphates. Homopolynucleotides, i.e. polymers of only one kind of
nucleotide (poly A, poly U, etc.) are produced from individual nucleo-
side diphosphates (ADP, UDP, etc.). Copolynucleotides containing
. + Pi
Vol. 40, No. 5, May 1964
Fig. 8. Amino acids released by acid hydrolysis of tryptic peptides from products of
incorporation of vairious aiioiho acids
tracings of paper chroinatogranis. A and B, peaks
isoleucine-C14 and lysine-C14. C and D, peaks 1 and 3 froni experiment with asparagine-
C14 and lysine-C14. The position of amiino acid markers is slhwn on the middle strip.
thie presence of poly AU
1 and 3 from experiment with
and trilysine, respectively.
The radioactive peptide peaks were eluted, hydrolyzed with hydro-
chloric acid, and the free amino acids separated by paper chroma-
tography. The actigraph tracings of Figure 8 show the distribution of
radioactivity in the HCI hydrolysates of peptide peaks
experiments with both isoleucine and lysine (sections A and B) and
asparagine and Iysine-labeled (sections C
course converted to aspartic acid by hydrolysis. Both peaks yielded
radioactive lysine and isoleucine in one case and lysine and aspartic
acid in the other. It may be seen, however, that the proportion of
lysine relative to that of the other amino acid was higher in peak 3
(the peak closer to the cathode) than in peak
amino acids in a ratio of
Further work aimed at closer identification of the polypeptides
formed with synthetic copolynucleotide messengers is in progress. In
the meantime the above, preliminary experiments leave little doubt that
copolynucleotides do indeed direct the cell-free synthesis of copoly-
I and 3 from the
nald D)). Asparagine
i. This suggests that
I peptides may contain labeled lysin-e and isoleucine (or
: I or higher.
3 peptides probably contain these
Bull. N. Y. Acad. Med.
4 I 0
THE GENETIC COD)E
4 I I
peptides and lend firm support to the results and deductions of the work
on the genetic code reviewed in this lecture.
The fact that synthetic polyribonucleotides can take the place of
natural messenger RNA and direct the cell-free synthesis of polypep-
tides, the amino acid composition of which depends on the base com-
position of the polynucleotides, has brought us surprisingly close to the
solution of the genetic code puzzle. The enzyme polynucleotide phos-
phorylase has, as we have seen, been the key to this development.
Progress in our understanding of the code, together with the spectacular
advances in our knowledge of the basic, chemical and enzymatic proc-
esses underlying the replication of DNA, the synthesis of messenger
RNA, and the biosynthesis of protein, have given us deep insight into
the molecular mechanisms of heredity.
There is evidence that the genetic code is of the nonoverlapping
type.3" For correct transcription the message must be read through
from one end of the messenger to the other. For a triplet code the
coding ratio, i.e. the ratio of nucleotide residues of the polynucleotide
template to amino acid residues of the polypeptide synthesized, should
have a value of 3. Direct determination of this ratio is barred at present
due to degradation of polynucleotides and polypeptides by nucleases
and peptidases present in the crude cell fractions available for protein
synthesis studies. As already pointed out, these enzymes also interfere
with the experimental determination of base sequence of the coding
units. It thus appears that the preparation of purified systems of protein
synthesis wvill be a prerequisite for successful experimental attack of
some of the most important problems still awaiting solution in the study
of the genetic code.
Various phases of the work at New York University reported in this paper
were carried out by Drs. P. Lengyel, J. F. Speyer, C. Basilio, A. J. Wahba, W.
Szer, Y. Kaziro, and A. Grossman, and Messrs. R. S. Gardner, J. J. Protass, and
R. S. Miller, with the technical help of M. C. Schneider and H. Lozina. We are
grateful to Drs. R. C. Warner and C. Gilvarg for many helpful discussions and
suggestions. The work was aided by Grant AM-01845 and 1-SOI-FR-05099-02 from
the National Institutes of Health, U. S. Public Health Service, and by fellowships of
the U. S. Public Health Service (C.B.) and the Jane Coffin Childs Fund for Medical
Research (A.J.W., W.S.).
Vol. 40, No. 5, May 1964
4 1 2
R E FE R E N CE S
1. Jacob, F. and Monod, .J. Genietic regi--
proteins, .J. Molec. Biol. 3:318-56, 1961.
Brenner, S., Jacob, F. and Meselsoni, M.
ma ltion from genes to ril)osomes for pro-
tein sy3nthesis, Nature (Lontdon)
576-81, May 13, 1961.
3. Marmur, J. and others. Specificity of
complementary RNA formed by B. sub-
tilis infected with bacteriophage SP 8,
(oldSpringHarb)or Symp. Quanit. Riol.
Sgpmp.Soc.Exp.Biol. 12:138, 1958.
5. Chapeville, F. and others. On the role
of soluble ril)onucleic acid in coding for
aimino acids, 17roc. X(it. A cad. SCi. USA
Formation of the peptidechain of hemo-
globin, Jiroc. Nfat. Acad. Sdi. USA
Dintzis, 11. M. Assembly of the peptide
chains of hemoglobin, T'roc. Nat. A cad.
Sci. USA 47/:.247-61, 1961.
8. Goldstein, A. and Brown, B. J. Addition
growing protein chains in E. coli, Rio-
(him.Biophgs.Acta 53':438-39, 1961.
9. Warner, J. R., Knopf, P. M. and Rich,
A. Multiple ribosomial strncture in pro-
USA 49:122-29, 196:3.
10. Gierer, A. Function of aggregated reti-
enlocCyte rihosomies in protein synlthesis,
J. 31olec. Biol. 6:148-57, 196:3.
12. Ochmo, S. Opening A(ldress. In Infoi-ma-
tional Macromolecules. H. J. V\ogel, V.
Bryson and J. 0. Lminpen, eds. New
York, Lcadlemmic Press, 1963, p. 3.
p)rotein structure and biosynthesis.
Vogel, V. B3ryson, and J. 0. L(aimpen,
eds. New York, Academic Press, 1963.
14. Crick, F. II. C. and others. General na-
in the sinthesis of
11. C. Onlproteinsynthesis,
J., Leahy, J. and Schwveet, R.
aci(l to C-terminal ends
J. Molec. Biot. 6:374-88,
tire of the genetice code for proteins,
Nature (Lowdoni) 19.2:1227-32, Dec. 30,
Ochoa, S. I)ie enzyniatische Synthese von
11ibonueleinsaluitire (lRNS), Ag(/ ew. (hem.
Ileppel, L. A., Ortiz, P. J. and Ochoa,
sized by p)olnlltcleotide phosphorylase.
II. Structure of polymers containing a
mixture of bases. J.
polynuicleotides synthesized by polynui-
1)oly-nlcleoti(les and the amino acid code.
Bretscber, M. S. and Grumnberg-Manago,
Pol rilhoutcleoti(le-directed p)rotein
synthlesis Iusino an E. coli cell-free sys-
tem, Natu re (Loiidoni)
20. Nirenberg, M. W. and Matthaei,
Iependence of cell-free protein synthe-
sis in E. coli upon naturally occurring
or synthetic polyribonuicleotides,
Not. Acaod. Sci. USA ;47:1588-1602, 1961.
21. Lengyel, P., Spever, J. F. and Ochoa, S.
Synthetic polyniieleotides and the amtino
acid code. Proc. i'ot. A c(((d. Sci. USA
22. Ochoa, S. Synthetic lyolmnicleotid1es andl
the genietic code, Fed. Proc. 2.2:i2-74,
polynuicleotides and the antino acid co(le.
VHIII. Jroc. Aot. A cod. Sci. UTSA 49:
2k. Spever, J. F. and others. Synthetic poly-
nucleotides and the amino acid code. IV.
Ochoa, S. Synthetic polunticleotides and
the genetic code. In Informo tional Ma-
0. Lanipen, eds. New York, Aca-
demic Press, 1963, pp. 437-49.
,J. and Ochmoa, S. Studies on
S. and others. Synthetic
Vaet. Aalcd. Sci. UJSA ;8:2087-91,
.19. :283-81, July
II. J. 'Vogel, V. Bryson,
Bull. N. Y. Acad. Med.
THE GENETIC CODE
4 1 3
code. IX, Proc. Nat. A cad. Sci. USA
27. Jukes, T. H. The genetic code, Amer.
Sci. 51:227-45, 1963.
28. Eck, R. V. Genetic code: emergence of
a symmetrical pattern, Science 140:477-
80, May 3, 1963.
29. Doctor, D. P., Apgar, J. and Holley,
acid-acceptor ribonucleic acids by coun-
tercurrent distribution, J. Biol. Chem.
30. Sueoka, N. and Yamane, T. Fractiona-
tion of aminoacyl-acceptor RNA and
the coding problem. In IThformational
Macromolecules. H. J. Vogel, V. Bry-
son, J. 0. Lampen,
Academic Press, 1963, pp. 205-27.
:31. Weisblum, B., Benzer,
R. W. A physical basis for degeneracy
Acad. Sci. USA 48:1449-54, 1962.
Basilio, C. and others. Synthetic poly-
nucleotides and the amino acid code.
V. Proc. Nat. Acad. Sci. USA 48:613-16,
3.3. Jones, 0. W., Jr. and Nirenberg, M. W.
Qualitative survey of RNA code words.
Proc. Nat. Acad. Sci. USA 48:2115-23,
34. Nirenberg, M. W. and Jones, 0. W., Jr.
Current status of the RNA code. In
Vogel, V. Bryson, and J. 0. Lamnpen,
eds. New York, Acadeniic Press, 1963,
35. Wittmann, H. G. Proteinuntersuclhtungen
an Mutanten des Tabakniosaikvirus als
Beitrag zum Problem des Genetischen
Codes, Z. Vererbungsl. 93:491-530, 1962.
rstigita, A. and Fraenkel-Conrat, H.
Composition of proteins of chemically
Molec. Biol. 4:73-82, 1962.
37. von Ehrenstein,
G. and Lipmann,
Experiments on hemoglobin biosynthesis,
Proc. Nat. Acad. Sci. USA 47:941-50,
38. Smith, E.
Nucleotide base coding
and amino acid replacements in proteins.
II, Proc. Nat. Acad. Sci. USA 48:677-
39. Arnstein, H. R.
of yeast amino
eds., New York,
S. and Holley,
S., Cox, R. A. and
Hunt, J. A. Function of polyuridylic
acid and ribonuicleic acid in protein bio-
synthesis by ribosomes from mammalian
1042-44, June 16, 1962.
40. Maxwell, E.
S. Stimulation of amino
acid incorporation into protein by nat-
ural and synthetic polyribonucleotides
in a mammalian cell-free system, Proc.
Nat. Acad. Sci. USA 48:1639-43, 1962.
41. Weinstein, I. B. and Schechter, A. N.
Polyuridylic acid stimulation of phenyl-
alanine incorporation in animal cell ex-
tracts, Proc. Nat. Acad. Sci. USA 48:
Griffin, A. C. and O'Neal, M. A. Effect
of polyuridylic acid upon incorporation
in vitro [14 C] phenylalanine by ascites
tumor components, Biochim.
Acta 61:496-71, 1962.
43. Weinstein, I. B. Discussion of Part IV.
In Informational Macromolecules. H. J.
V\ogel, V. Bryson and J. 0. Lampen,
eds., New York, Academic Press, 1963,
41. Protass, J. J., Speyer, J. F. and Leng-
yel, P. Amino Acid code in alcaligenes
faecalis, Science 143:1174-76, 1964.
45. Singer, M. F., Heppel, L. A. and Hil-
moe, R. J. Oligonucleotides as primers
Biol. Chem. 235:738-50, 1960.
polynucleotides and the amino acid code.
VI. Proc. Nat. Acad.
acid by a potassium-activated phospho-
diesterase from Escherichia coli, J. Biol.
Chem. 238:2251-53, 1963.
48. Kaziro, Y., Grossman, A. and Ochoa,
S. Identification of peptides synthesized
by the cell-free E. coli system with poly-
nucleotide messengers, Proc. Nat. Acad.
Sci. USA 50:54-61, 1963.
49. Waley, S. G. and Watson, J. The action
of trypsin on polylysine, Biochem. J.
50. Stewart, J. W. and Stahmann, M. A.
Chromatography of polylysine, J. Chro-
matogr. 9:233-35, 1962.
51. Smith, M. A. and Stahmann, M. A.,
Biochem. Biophys. Res. Comm. In press.
Vol. 40, No. 5, May 1964