21 1 57 _N~_~___ v_o_L._m __ -~--~_L_ ~------NEWSANDVIEWS------'- species exhibiting resistance, as the use of the agent continued. It became clear that two processes are in­ Antibiotic resistance and the volved in the of resistant popula­ tions: the 'invention' of the resistance genes themselves, and their multiplication evolution of and spread. Relatively little is known even from Mark Richmond now about the first of the two stages, though there are some clues. As for these­ A PAPER by Victoria Hughes and Naomi within the population and capable of being cond, the mechanisms of genetic exchange Datta in this week's issue of Nature (p.725) transferred within the members of the discussed above, coupled with selection, fills a gap in our knowledge of the population. Furthermore, that integration can provide a complete explanation. emergence of antibiotic resistance in and excision ofplasmids into and from the One part of this story has, however, bacteria. The study makes use of a sometimes transferred blocks always been taken for granted, and it is remarkable collection of bacteria made of information from the chromosomal to here that the paper appearing in this week's between 1917 and 1954-before the use of the extrachromosomal state underlined the issue of Nature finally provides definitive antibiotics became widespread and at a fact that a bacterial population as a whole evidence. For the scenario outlined above time when little was known of bacterial had considerably genetic fluidity even if, to be correct, bacterial strains isolated genetics - and kept in sealed vessels since under normal circumstances, this fluidity before the advent of large-scale use of anti­ that period. was constrained. When it was realized that biotics should exhibit all the genetic flexi­ The study of bacterial plasmids and this potential fluidity extended to popula­ bility of the antibiotic-resistant bacteria we episomes, vital to an understanding of the tions containing different species and see nowadays, but the incidence of spread of antibiotic resistance, did not real­ genera it was clear that one of the main pro­ resistance genes themselves should be low. ly begin until the 1960s. Early on, most ex­ cesses determining the means of bacterial Using a large collection of Gram-negative perimental work was focused on a few evolution had been discovered and that a bacterial isolates made by R.G.E. Murray bacterial species - notably Escherichia much deeper understanding of the from before the antibiotic , Naomi Dat­ coli and Staphylococcus aureus - and the emergence of antibiotic resistance could be ta and Victoria Hughes have now shown plasmids/episomes involved were limited gained. that this is indeed the case. to a few types of phage and antibiotic­ The emergence of resistance had already In a sense, therefore, Murray's collec­ resistance factors and to some colicin fac­ been broadly defined: the introduction of a tion is representative of the kinds of tors. What emerged was a picture of the novel antibacterial agent was followed by bacterial population that existed when bacterial genome in which most of the the rapid appearance of resistant bacterial antibiotic therapy started in the early genetic information in the organism was populations which then became more and 1940s. Plasmid transfer and gene exchange seen to be concentrated as a single large cir­ more widespread, both geographically and supported by appropriate pressure have cular DNA molecule - the chromosome also with respect to the bacterial strains and done the rest. D - of molecular weight about 2 x 109 , but some was carried as small circular DNA genetics molecules which did not seem absolutely necessary for bacterial growth and survival in most environmental circumstances. Is there a human locus? When these extrachromosomal DNA TIt elements carried the relevant genetic infor­ from P.N. Goodfellow and P. W. Andrews mation they behaved as bacterial viruses, capable of being triggered to multiply CHROMOSOME 17 of wild mice frequently lethality. It might also be expected that an rapidly to burst the host cell. contains a region of mutant genes, known equivalent of the TIt complex would be A further characteristic of many as the TIt complex, maintained in associa­ linked to the human equivalent of H-2, plasmids and phage was their ability to tion by local suppression of recombina­ HLA. This possibility has provided the transfer from one bacterial strain or species tion. Within the region, which stretches starting point for several searches for the to another and, in some cases, to catalyse from the centromere to beyond the H-2 human Tit complex. Genes known to be the transfer of further independent circular locus, different t-haplotypes are charac­ linked to the HLA region include loci con­ DNA molecules. At the extreme, certain terized by different H-2 types and dif­ trolling class 1 and class 2 cell-surface an­ bacterial plasmids, following integration ferent recessive lethal with strik­ tigens, loci controlling complement com­ 1 into the bacterial chromosome, could even ing effects on embryogenesis • The ponents and a locus controlling the enzyme cause the transfer of a copy of a whole t-haplotypes also carry a group of inter­ glyoxalase (GLO). At least two alleles are chromosome from one bacterial cell to esting mutations which, when homo­ known for each of the four major histo­ another to give rise, at least transiently, to a zygous, cause male sterility and when compatibility complex (MHC)-linked diploid organism. heterozygous cause carrier males to complement loci, but these are so closely Our awareness of the range of possible transmit the chromosome carrying the linked that recombination between them interactions of bacterial plasmids and the !-haplotype to nearly 100 per cent of their has not been observed and only 14 common effects these interactions had on the offspring. A similar constellation of muta­ cis combinations of alleles (complotypes) bacteria radically altered our view of the tions known as Segregation Distorter has occur. individuality of single bacterial cells. Popu­ been observed in wild populations of In a survey of families drawn from the lations of bacteria could now be seen as a Drosophila melanogaster. Evidence for a Boston area, Awdeh eh af.3 observed that single gene pool, with most of the infor­ locus homologous to the TIt locus in mam­ several complotypes showed non-random 2 mation segregated for much of the time in mals other than mice, and perhaps rats , association with alleles at the HLA-D and the individual members of the population, has been harder to find. Several recent HLA-B loci and less often with alleles of but with extra DNA molecules distributed papers suggest that a human homologue of the TIt complex may indeed exist Hi. P. N. Goodfellow is in the Laboratory of A human equivalent of the murine TIt Human Molecular Genetics, Imperial Cancer Research Fund, Lincoln's Inn Fields, London Mark Richmond is Vice-Chancellor and Pro­ complex would be expected to be WC2A JPX and P. W. Andrews in the Wistar fessor ofMolecular Microbiology in the Univer­ associated with segregation distortion, Institute, 36th Street and Spruce, Philadelphia, sity of Manchester, Manchester M13 9PL. recombination suppression and recessive Pennsylvania 19104.

0028-0836/83/160657-02SOI.OO © 1983 Macmillan Journals ltd