EUROPEAN SERVICE

First Broadcast 22,1.62, THE FRONTIERS OF KNOWLEDGE NO,256

 

English Service (40000000 he
at 1800 GMT . ‘Cracking the Genetic Code!

 

by Dr. Francis H.C. Crick, F.R.S.,
ef the Mepartment of Physics,
Cavendish Laboratory, University
ef Cambridge.

Pre-rec: 16,1.62.
72 TBU 135549
Dur ;

Why do people resemble their parents? Ina broad way we already
know the answer, We all start from the union of one sperm cell with
one egg cell, each of which carries about half the genetic information.
Now there must be a lot of this information since we're very
complicated objects. Not only do we have many different organs, such
as eyes and ears and a liver, but each of these is made from special
cells, in a very intricate way, and each cell is also in itself g&

very compu
=~ So clearly there must be a very great deal of
genetic information stored in both the egg and the sperm,

And yet it's been knewn for a long time that a sperm is a very
small object. Its head, which carries the genetic information, is
no mere than one hundredth of a millimetre across, The actual
genetic material weighs only a few times a millionth of a millionth
ef agram. However the information is recorded, it must be written
on a very small scale if so much is to be got into such a very tiny
space.

The obvious conclusion is that the genetic information is
likely to be written at the atomic level, in terms of atoms and small
molecules, and indeed all the experimental work of the last ten years
suggests that this is true,

However, a general argument ef this kind can only be suggestive.
We shall need many more detaiged facts if we're to know just exactly
hew the genetic message is stored, and how it's read.

Fer some time we've had a good idea of the sort of way in which
the message is stored, In most organisas it's written on @ leng thin
thtinaar melecules known cellectively as the family deexyhAibonucleic

acids, or & DNA. DNA is a polymer. That is to say, it has a regular
. 25 Lines : Cont'd...
~ 2

repeating backbone, with side groups called "bases" projecting at
regular intervals. Hewever, the bases are not all the same. There
are four kinds of them, and the genetic information is conveyed by
the precise order of the different sorts of bases along the DNA.
In other words the genetic message is written in a language of four
letters, Incidentally, the tetal length of the message, for man, is
net short, It is probably more than a thousand million letters leng.

But what does the message mean? How does it act? Here again

MESSEG 2,
, each

we know the answer in broad terms, Each section of the
paragraph we might say, contrels the production of a particular
protein, Proteins are large molecules and their great importante in
the living cell is that they are the "catalysts", which decide which
chemical reactions will take place, For every chemical reaction the
cell carrie& out, there's a special protein which accelerates that
reaction and that reaction alone,

New all proteins are built on the same ground plan, Each is
a polymer, a long molecule, again with a regular backbone (although
quite different from the backbone ef DNA), a backbone to which side
groups are attached at regular intervals, However, in protein there are
20 kinds of side-groups, so that proteins, the machine tools of the
cell, are written in a 20-letter language.

What we did not know, until recently, was how the cell managed
the translation from the 4-letter language of the genetic material
to the 20-letter language of the gene preducts, the proteins.
Within the last six or eight months tiere's been a spectacular
breakthrough in this preblem and it now looks as if we shall know
the answer within a year or so.

Although it's obviously important to know the details of the
reading machanism ~ in other words to discover the various
biochemical steps - and although we know quite a bit about some of
these steps, especially thoxe leading up to the actual act of protein
synthesis, the part of the problem I want to spotlight is what one
might call the "dictionary" used in the process, Each of the 20 units —

31 Lines: Conttd..
ao 3 -
theryp're called amine acids, by the way - which go to make up the BREAK
protein must be represented by a greup of letters in the four-
letter genetic language. Let's call these letters, A,B,C, and Dy
to make it easier, Then ene amine acid, one of the twenty, might
be coded by ABB, and another by CDA and se ferth, Se: the first
question we can ask is hew many letters stand for ene amino acid,
It could hardly be enly twe, because there are enly 16 possible
different pairs ef letters te be made frem A,B,C, am D, whereas we
need 20 combinations. Se three is the minimum number, Actually,
as we shall see, it leoks as if three is cerrect, and that the
genetic cede has te be read eff in triplets.
. You may wonder why one can'$ get at the cede in a different
x manner, One would need te knew the amino acid sequence of a
particular protein and also the base~sequence of the piece of DNA
which carried the message for that protein, and then Simply cumpare
the two. But unfortunately this is not technically possible.
In favourable cases, and with a lot of hard work, one may be able
to find the sequence of amino acids in a certain protein, but it's
almost impossible at the present time to find the base sequence
of a piece of DNA, or even to obtain pure the particular piece
one is after.

One indirect method is to make a chemical change in the INA,
and see how much of the protein is changed. This has been done
recently, particularly by Dr. Wittmann at Tubingen, by using the
plant virus, Tobacco Mosaic Virus. This has the related
ni lronucleic acid, RNA, as its genetic material. Using nitrous
acid it's possible to alter just one base of the RNA, at random,
and then to find how the protein produced by the virus, which
makes up its shell, is altered. The typical change is to one
amino acid only.

This result is important because it shows that one letter of

the RNA only affects a single amino acid. It's quite easy to
30 lines (Cont'd.)
- 4 —

imagine a case where, say, the first three letters would code the
first amino acid, and the second amino acid would be coded by
letters 2, 3 and 4 in the sequence, not by no's 4, 5 and 6.

Such overlapping codes now look highly unlikely.

But how do we know that a set of three letters codes an amino
acid, and not a set of four, or five or any other numberg? Some
very recent work by my Cambridge colleagues, Mrs. Leslie Barnett,
Dre Sydney Brenner, Dr. Richard Watts-Tobin, and myself have

very probable

made it that the correct answer is three. Our

   

work was not biochemical but genetical, and was done for convenience
on a virus, or bacteriephage as it is called, which attacks bacteria;
the T4 virus. We had been studying the effects produced by a
certain chemical which we suspected added or removed a base from
the DNA of the virus. We found we had several different viruses
which seemed to be altered in the same sort of way, but in
different places in the DNA, all rather close together. For
simplicity let's suppose that in each case a single base had been
added to the DNA, but at different points.

Now by genetic methods we can put together viruses which
have any two of these defects in their DNA, or even all three.
If the virus had any of the single defects, or any of the double
defects, it wouldn't grow on a perticular strain of bacteria.
The gene in which the defects occurred would not function. But
if we put all three defects into one gene the function came
back again.

We explained this as follows. We assumed that the code was
read in groups of three letters, starting from a fixed point
and working along three at q time. There is nothing in the string
of letters to show where one triplet ends and the next begins.
Now if a letter is added anywhere, the whole reading goes wrong
from there on, since after the added letter, the letters fall

into the wrong sets of three. Thus mest of the message becomes
31 lines (Cont'd.)
-5 =
nonsense. The effect is just the same if two letters are added.
However, if three letters are added, although the message is
wrong in that neighbourhood, the whole of the rest of the message
now falls correctly into sets of three and makes sense again, so
that in favourable cases the gane will work, after a fashion.

Our results also suggest that mest of the 64 possible triplets
stand for an amino acid, so that usually an amino acid can be
coded by more than one triplet, but exactly how this is done we
can't say.

However, the real break-through in the problem is due to
two young biochemists at the National Institutes of Health (at
Bethesda in America) called Nirenberg and Matthaei. This August,
at the Biochemical Cengress in Moscew, they reported a spectacular
result. They were working on a cell-free system, made from broken
bacterial cells, which will synthesise protein. This system has
been developed in a number of laboratories. To this system they
added a special RNA. ENA is a close relatiom of DNA and is
believed to carry the genetic message from the DNA, in the nucleus,
to the cytoplasm where the protein is actually made. The DNA
is the master-cepy and the RNA is the working copy - "DNA makes RNA
and HINA makes protein" as the eco goes.

The RNA they added, which was made by a special enzyme, was
peculiar in that all the bases were the same. That is, the
messenge# read BBBRBB... and so on, let us p¢way. To everyone's
surprise this simple repetitive message was translated by the
fragments of the cell, and a protein was produced which had all
its amino acids the same, so that it read ¥s¥ 3... let us say.

In other words the triplet BBB stands for’. In strict chemical
terms, three uracil bases code for one phenylalanine Aodsautes,

This result has now been repeated in labs all over the w_orld.
Moreover, other artificial RNAts can be produced and tested

on the system. Already very interesting results have been reperted,
31 lines (Cont'd.)
-~6-

especially by Ochea and his colleagues in New York, and the general
feeling is that with luck the whole code may be obtainable in
this way.

We still don't know whether the code is universal. ‘The same
20 amino acids are used in proteins throughout nature, from
virus to man, but it is not yet certain that the same triplets
code them in all erganismg although preliminary evidence suggests
this is probable. If so, we shall have the key to the mokecular
organisation of all living things on Earth.

But on Mars, I wonder? ‘Will there be life, or the remains
of life, on Mars? And will it be DNA and RNA amd protein all
over again? ‘The same languages perhaps, with the same code
connecting them? Who knows?

EN D
12 lines
TOTAL 160 lines