Bacteria: Recombination (IV) 



371 



tide sequence, since the complexity, and hence 

 the specificity, of a cistron's product would be 

 expected to be directly correlated with the 

 number of nucleotides in the cistron. 



It can also be hypothesized that most, if 

 not all, of the heterochromatin in a cell is 

 composed of a monotonous repetition of the 

 same nucleotide pair, usually A-T. This 

 view receives support from the following 

 observations: (1) The sequence of A, T, C, 

 G on a single strand of DNA is far from 

 random; at least 70% of the bases are dis- 

 tributed so that three or more pyrimidines 

 (and hence purines, too) occur in successive 

 nucleotides; sequences of five successive T's 

 have been identified; 4.9% of a tested DNA 

 contains sequences of eight or more pyrimi- 

 dines in succession; (2) a polymer of A-T 

 alone may make up as much as 30% of cer- 

 tain crab sperm DNA; (3) the amount of 

 thymine is increased in Drosophila cells carry- 

 ing an extra, almost entirely helerochromatic, 

 Y chromosome; (4) 5-bromo deoxyuridine, 

 which is known to replace thymine in DNA, 

 causes a high frequency of breaks near or in 

 heterochromatic regions. 



Suppose, to exemplify these suggestions, 

 that all nucleotides in a particular stretch of 

 heterochromatin have the same formula. A, 

 and that three in succession (AAA) comprise 

 a cistron. (We can ignore, at this time, the 

 complementary chain composed of T's.) 

 Since the number of successive A's is expected 

 to be considerable, it is reasonable that a 

 particular A may be used cistronically, some- 

 times with two A's to its right, and other 

 times with two A's to its left. In other words, 

 in heterochromatin, a nucleotide can be 

 shared by adjacent cistrons. On the other 

 hand, a typical, highly specific cistron in 

 euchromatin would be expected to be com- 

 posed of several hundred or a thousand 

 linearly arranged nucleotides. Suppose a 

 particular euchromatic cistron terminates in 

 -TAA. If a break occurs between the -TA 

 and A, and another occurs almost anywhere 



in the heterochromatin, and cross-union of 

 broken ends occurs, the euchromatic cistron 

 will have its proper nucleotide sequence com- 

 pleted. But, since any three successive A's 

 can serve as template for making hetero- 

 chromatic product, there could be nucJeotide- 

 sharing between the now-adjacent euchro- 

 matic and heterochromatic cistrons. In this 

 case, whenever the heterochromatic cistron 

 was used to make product, no complete prod- 

 uct would be formed from the euchromatic 

 cistron. Thus, so long as the adjacent 

 heterochromatic cistron made its product, 

 the euchromatic cistron would be functionally 

 suppressed, and would be scored as an 

 amorph (see p. 210). If conditions changed, 

 so that the euchromatic cistron was able to 

 function as usual, the normal euchromatic 

 phenotype would result. If the heterochro- 

 matic and euchromatic cistrons were func- 

 tional alternately, no detectable phenotypic 

 effect would be expected with regard to the 

 heterochromatic product, but phenotypic 

 mosaicism could result because of the inter- 

 mittently produced euchromatic product. In 

 these ways, suppressed or variegated pheno- 

 typic effects, which are known to be due to 

 the placement of heterochromatin near eu- 

 chromatin, may be explained. Such position 

 effects (cf. Chapters 22 and 25) are frequent 

 following structural changes in chromosomes 

 of Drosophila. Some of these cases of pheno- 

 typic suppression involve genetic elements 

 like Segregation-Distorter in Drosophila and 

 Dissociation in corn, which are associated 

 with heterochromatin, can also cause break- 

 age, and can change their location in the 

 genome. Since such factors resemble epi- 

 somes, it may not be too far-fetched to study 

 these and known episomes for the occurrence 

 and possible consequences (phenotypic sup- 

 pression, organelle movement, and chromo- 

 somal breakage) of nucleotide-sharing. 



Many of the ideas relative to homologous 

 organelles, nucleotide-sharing, and suppres- 

 sive and variegated position effects discussed 



