Reprinted from the Proceedings of the NATIONAL ACADEMY OF SCIENCES, Vol. 42, No. 5, pp. 245-249. May, 1956. A BINARY MUTABILITY SYSTEM IN ESCHERICHIA COLI By P. D. Sxaar BIOLOGICAL LABORATORY, COLD SPRING HARBOR, NEW YORK,* AND DEPARTMENT OF GENETICS, UNIVERSITY OF WISCONSIN, MADISON, WISCONSIN F Communicated by M. Demerec, March 8, 1956 Although genetic instability is often largely a function of the gene which itself mutates, there is abundant evidence that the residual genome may exert an effect. This effect may be strong and highly localized, as if at a single locus.!_ Such binary mutability systems in maize have been investigated extensively and have led to important concepts of gene mutation and cellular differentiation.2 Among micro- organisms, differences in spontaneous mutation rates have been analyzed genetically in only a few instances,? and in no case has a binary system been revealed. Treffers et al.‘ discovered that one substrain of Hscherichia coli, strain K-12, exhibits a mutation rate from streptomycin sensitivity to streptomycin resistance that is about 100-fold greater than that observed in other K-12 substrains. The present investigation reveals that the basis of mutability in this strain is binary, involving genetic determinants which are not closely linked. 246 GENETICS: P. D. SKAAR Proc. N. A. 8. MATERIALS AND METHODS The mutable strain, C519, employed in these experiments was derived from 58-278,* ® but no longer requires biotin. Strain CS19 carries the following perti- nent markers: F+ Pa-~ T+ L+ Tht Lacyt S* Vii Muit. Additional K-12 sub- strains employed are W677 (F- Pat T- L- Th- Laey~ S* Vy" Mut-), W1177 (an S* derivative of W677), and CS11 (an F'+ derivative of W1177). The designations F+ and F7- refer to mating type;* Pa-, T-, E-, and Th-, to phenylalanine, threonine, leucine, and thiamine growth requirements, respectively; Zac—, to an inability to ferment lactose; S’ and V1’ to resistance to streptomycin and coliphage Tl (and T5). The determinant of mutability is symbolized as Mut+. Like other nonmutable strains, the rate with which W677 undergoes mutation from S° to S’ is about 4 * 10-'° per bacterial division, whereas the CS19 mutation rate is about 4X 107. The B/r strains employed have been described elsewhere’ and may be character- ized here by the location of their auxotrophic mutations with relation to two important and more or less distinct segments of the H. coli genome. IMN27 requires arginine and methionine because of two mutations in the 14—Th region; IMNG64 requires tryptophan, tyrosine, phenylalanine, and p-aminobenzoic acid, owing to a single mutation in the S—Xyl region. Both are F- S* Mut~. The experimental procedures employed follow descriptions given elsewhere.® For crosses, 16-20-hour Difco Penassay broth cultures were washed twice in saline, and Q.i-ml. samples were mixed on minimal (or supplemented minimal) agar medium. Recombinants were examined after single-colony isolation on complete medium. Streptomycin, threonine, leucine, and thiamine were made up to con- centrations of 200, 20, 10, and 0.01 mg/1, respectively, where indicated. To score recombinants for mutability, about 100 cells were inoculated into 10 ml. of Penassay broth and incubated for 24 hours at 37° C. without aeration. Then 0.1-ml. samples were plated onto nutrient agar containing streptomycin. If 10 or more resistant colonies appeared, the recombinant was scored as mutable. If less than 10 colonies appeared, two new broth cultures were initiated from small inocula and similarly assayed; if each of these two samples contained less than 10 resistant cells, the recombinant was scored as nonmutable. In no case was further testing necessary. This is sufficient to distinguish parental types with a low probability of error, since it was found empirically that among samples of 50 W677 cultures similarly grown, 45 contained no resistant cells and none contained more than 4, whereas, among samples of 50 CS19 cultures similarly grown, none contained less than 25 resistant cells. RESULTS The Site of Instability Determination.—CS19 was crossed with W677 on un- supplemented medium and medium singly supplemented with threonine, leucine, and thiamine. Recombinants were scored for Lac, Vi, Mut, and unselected nutritional markers. Within the limitations of the method of testing, the segrega- tion of mutability was unambiguous. In the unsupplemented cross (Table I), the high frequency of mutable recombinants indicates a close linkage of Mut to T, L, or Th. In the thiamine-supplemented cross, the equally high incidence of mutability among both the Tht and Th- recombinants indicates that the close linkage is not VoL, 42, 1956 GENETICS: P. D. SKAAR 247 to Th. A comparison of the E- and T~ recombinants from the third and fourth crosses reveals that, whereas 5 out of 8 7'— recombinants are mutable, only 2 out of 24 L~ recombinants are mutable. Thus Mut appears to be more closely linked to Z than to T. The sequence Lac:—V;—L—T is extensively documented” and is consistent with the present data. A decision as to whether Mut lies to the left or to the right of Z is not easily made on the basis of the limited data available. How- ever, the fact that, of the 9 nonmutable recombinants issuing from the threonine- supplemented cross, all involve a crossover between V, and L, whereas only 3 involve a crossover between L and 7’, suggests that Mut lies to the left of Z. If so, the frequencies of single crossovers in the three regions, Lac:x——V1—"_-Mut__L, provide a crude estimate of relative map distances. Considering only prototrophic recombinants and excluding those obtained in the presence of leucine, these fre- quencies are 73, 22, and 6, respectively. Prototrophs obtained on leucine are excluded because of the suggestion that some of them may be inhibited by this amino acid.}° TABLE 1 SEGREGATION OF Mut anp AssociaTED Factors IN CRossES BETWEEN CS19 (P+ Pa~ T+ L+ Tht Lacyt Vis 8* Mut*) ann W677 (F- Pat T- L- Tho Laeg~ Vit & Mut~) SuppLEMENT UnseLecTeD MARKERS None THIAMINE LEUCINE THREONINE Lac Vi Mut Tht The + L- Tt T- + s + 15 5 9 25 2 49 1 - s + 18 4 16 21 0 51 3 - r + 6 2 6 6 0 14 1 + r + 0 0 0 2 0 1 0 + 8 ~ 0 0 0 0 1 0 0 - 8 _ 0 0 0 0 1 0 0 - r - 1 0 1 1 20 5 2 + r _ 0 0 0 0 0 1 1 Total 40 43 7 129 The Site of Mutation.—Two independent S’ derivatives of CS19 were crossed with W677. Owing to the much higher incidence (at least 100) of S’ mutants in Mutt cultures than in Mui- cultures, it is likely that the S’ mutations were associated with Mutt. The frequency of mutable types among the S* recombinants (Table 2) was comparable to that observed in the cross of CS19 S* with W677, indicating that mutation from S* to S’ does not inactivate Mui*+. Further, very few of the re- combinants were S’, indicating not only the binary nature of the mutability system but also that the components are not closely linked. The absence of S’ recombinants in a cross of one CS19 S’ with W1177 (S’) suggested that the two mutations were isolocal or closely linked. The question whether most of the S’ mutations in Muit populations also occur at this site was approached by a somewhat more sensitive test. The aromatic deficiency of the B/r strain, IMN64, is closely linked to the S locus (of W1177), whereas the de- ficiencies of IMN27 are not. In K-12 (F*+) < B/r (F7-) crosses, only those K-12 unselected markers which are closely linked to the deficiency (selective marker) of the B/r parent occur frequently among recombinants. Thus the frequency of S’ recombinants from the cross of CS11 (W1177 F+) and IMN64 is higher than from the cross of CS11 and IMN27 (Table 2). Six independent S’ mutants of CS19 248 GENETICS: P. D. SKAAR Proc N, A. 5S. were crossed with IMN64. The high frequency of 8’ recombinants in all these crosses is consistent with the conclusion that most Mué+-associated mutations to TABLE 2 SEGREGATION OF Muti anD S IN SeLEcTED CrossEs PrototTroPHic RECOMBINANTS Per Cent Ss F+ Parent F- Parent EXAMINED Per Cent Sr Mut+ wt CS19 S* #1 W677 40 13 100 0 Sr #2 Wo677 64 22 96 4 CS11 IMN64 72 56 Lee wee eA 48 17 Lee wee N64 48 vee 6 94 CS19 IMN27 48 - 4 96 CS19 Sr #2 IMN64 50 60 a a Sr #3 IMN64 50 70 Sr #4 IMN64 50 60 Sr #5 IMN64 50 62 S” $6 IMN64 50 62 Sr #7 - IMN64 50 58 CS19 Sr #2 IMN27 50 18 streptomycin resistance occur at or near the same site. Since the deficiencies of IMN64 and IMN27 are not closely linked to L, the low frequency of Mutt re- combinants in crosses of these strains with CS19 is expected. DISCUSSION Examination of mutable recombinants from crosses involving CS19 immediately revealed that mutants other than S’ were abnormally frequent. This is also true of CS19 populations.* The range of action of Mut+ is not the immediate concern of this report, but it is appropriate to remark that the abnormally frequent mutants include Tht in Th- Mut* recombinant populations and 7+ in T-> Mué+ recom- binant populations. If these represent reversions, three widely separated loci are Mutt-influenced. The remoteness of the two components of the S mutability system lends credence to the view that instability is manifested throughout the bacterial genome in this mutable strain and possibly in others.!!_ This approximates the situation in Drosophila populations containing mu-F" and is not inconsistent with the situation expected of maize populations containing Ac and Ds.? That Mut* acts by inducing mutations has been assumed in this presentation. Evidence has been presented‘ that differences in relative growth rates cannot explain the high frequency with which S” mutants appear in Mutt strains. The objection that mutations might be equally frequent in Mut- strains, but more often lethal, is applicable, of course, to many situations where differences in mutation rates are inferred. SUMMARY The high rate of spontaneous mutation to streptomycin resistance (S") exhibited by the K-12 substrain CS19 has been shown to involve a binary mutability system. Mutations occur at or near the same S locus, but their high frequency depends upon a remotely linked factor, Mut+. The segregation of Mut+/Mut~ is well defined and is consistent with a close linkage to the Z (leucine) locus. * Present address. The work at Cold Spring Harbor was supported under contract by the United States Army Chemical Corps Biological Warfare Laboratories, Fort Detrick, Frederick, Maryland. VoL. 42, 1956 GENETICS: P. D. SKAAR 249 t Paper No. 610 from the Department of Genetics. The work at Madison (1951-1953) was supported by grants from the National Cancer Institute (C-2157), Public Health Service, and from the Research Committee, Graduate School, University of Wisconsin, with funds allotted by the Wisconsin Alumni Research Foundation. The counsel of Dr. J. Lederberg is gratefully acknowledged. 1M. Demerec, these Procrepines, 15, 834-838, 1929; M. M. Rhoades, J. Genetics, 33, 347- 354, 1936; H. H. Plough, Cold Spring Harbor Symposia Quant. Biol., 9, 127-137, 1941; J. V. Neel, Genetics, 27, 519-536, 1942; P. T. Ives, Evolution, 4, 236-252, 1950. 2 B. McClintock, Cold Spring Harbor Symposia Quant. Biol., 16, 13-46, 1951. 3N. H. Giles, Jr., Cold Spring Harbor Symposia Quant. Biol., 16, 283-313, 1951; E. M. Leder- berg, Genetics, 37, 469-483, 1952. ‘H. P. Treffers, V. Spinelli, and N. O. Belser, these ProcrEepines, 40, 1064-1071, 1954. 5K. L. Tatum, these Procrrepinas, 31, 215-219, 1945. 6 J. Lederberg, L. L. Cavalli, and E. M. Lederberg, Genetics, 37, 720-730, 1952; W. Hayes, Cold Spring Harbor Symposia Quant. Biol., 18, 75-93, 1953. 7V. Bryson, P. D. Skaar, H. Davidson, J. Hadden, and G. Bompiani, Ann. Rept. Biol. Lab. Cold Spring Harbor, N.Y., 1954-55, pp. 16-22. 8 J. Lederberg, E. M. Lederberg, N. D. Zinder, and E. R. Lively, Cold Spring Harbor Symposia Quant. Biol., 16, 413-443, 1951; J. D. Watson and W. Hayes, these ProceEpines, 39, 416-426, 1953. 9 J. Lederberg, Genetics, 32, 505-525, 1947; Methods in Medical Research, 3, 5-22, 1950. 10K, H. Rothfels, Genetics, 37, 297-311, 1952. 1A. Goldstein and J. 8. Smoot, J. Bacteriol., 70, 588-595, 1955. 12M. Demerec, Genetics, 22, 469-478, 1937.