alf) A. TSUGITA AND H. FRAENKEL-CONRAT 



A-coiitaininji imclcotidc, according to Ijoth groups) must be attiil)utcd 

 to mutational events other than C— > U and A — > G wliich, as stated, are 

 definitely possil)le. Altei-nately, the degeneracy of the code might supply 

 an explanation. It is encouraging, however, that at least 55% of the 

 amino acid exchanges observed by us can be attributed to cytosine 

 deaminations on the basis of present coding data. About 15% of the 

 changes suggest an A -^ G exchange, but an ecjual fraction is indicative 

 of C-^G and A^U exchanges. On the other hand, the fact that the 

 allotment of code symbols to amino acids has been influenced by ex- 

 change data obtained with nitrous acid mutants naturally renders the 

 "agreement" less surprising. It also suggests that the limits in the useful 

 exploitation of random nucleotide copolymers for coding information 

 have now been reached, if not overstepped. 



To conclude this short survey of the present concepts concerning the 

 code that relates RNA to protein structure, one is now in the fortunate 

 position that much experimental data has recently become available 

 which brings this problem from the realm of speculation to that of 

 factual information. It appears that three or possibly a slightly greater 

 number of nucleotides code for one amino acid; it appears probable that 

 several oligonucleotide units may represent the same amino acid. It is 

 certain that a few uridylic acids represent one phenylalanine, and the 

 approximate nucleotide composition of the words spelling 16 other amino 

 acids are known with varying certainty. The code seems to be quite 

 similar for very different organisms. It is probably deciphered starting 

 from an end of the nucleotide chain and thus requires no markers 

 (commas) between "words." 



The main projects for the immediate future are (1) the preparation, 

 by chemical or biosynthetic means, of polymers of known repeating 

 nucleotide sequences and their systematic testing for messenger activity, 

 (2) the establishment of the nature of many decidedly single-hit, one- 

 step mutations in terms of resulting amino acid exchanges, and (3) the 

 development of methods to determine the nucleotide exchange produced 

 by the mutation, be it in the DNA or in the messenger RNA. 



It would appear that within a few years it may become possil)le to 

 produce a known modification in one of the simpler DNA or RNA 

 genomes and find it to result in a predicted protein alteration. The 

 natural mode of transfer of information from the genetic DNA to the 

 messenger RNA is now reasonably well understood. There remains only 

 one major obstacle to overcome before we can fruitfully exploit our new 

 knowledge toward the manipulation of even the simplest genetic ma- 

 terial, which is that of the small viruses. The remaining hurdle is the 



