Reprinted from the Proceedings of the Nationa ACADEMY OF SCIENCES Vol. 46, No. 1, pp. 57-64. January, 1960. THE KFFECT OF ACRIDINE DYES ON MATING TYPE FACTORS IN ESCHERICHIA COLI*} By Yuxrnort Hirota DEPARTMENT OF GENETICS, STANFORD UNIVERSITY MEDICAL CENTER, PALO ALTO, CALIFORNIA Communicated by Joshua Lederberg, November 30, 1959 In “Escherichia colt, female strains are designated as F-; male strains are of two kinds, designated F*+ and Hfr, respectively. The male determinant in Hfr strains behaves as a chromosomal factor allelic to F-.'~* In F+ strains, however, maleness is determined by a factor “F,” with remarkable properties, notably its easy, con- tagious transmission to F~ cells.4~* Cells carrying F can be disinfected by treat- . ments with cobalt ion and with acridine dyes.7.* These properties support the con- clusion that F is a plasmid, an extrachromosomal particle, which is readily trans- ferred during mating contacts. It has been suggested that H{fr strains represent the incorporation of F as an element of the chromosome." ® * * ® Jaeob and Woll- man introduced the term “episome” for a plasmid that has a facultative association with the chromosome.? ‘Che present paper reports further evidence for this con- ception, namely ou the mechanism by which F is eliminated by the acridine dyes. Materials and Methods.---Two acridine dyes, proflavine (PF, 2:8-diamino- acridine), and acridine orange (AO, 2:8-b¢sDimethylaminoacridine) were used. Stock solutions containing LOO wg per ml of PF or 500 yg per ml of AO in water were autoclaved and stored in the dark for periods up to a week. EM-sugar agar!! was used as a selective medium. To grow on this medium, a recombinant must be prototrophic and also be able to ferment the sugar, e.g., lac- tose. Nutrient medium used for acridine treatment consisted of Difco peptone, 10 and Difco meat extract, 10 gm per liter. The pH of the medium was adjusted with sodium hydroxide solution using the Beckman pH meter. Difco penassay broth (Antibiotic assay medium number 3) was used routinely for bacteriological work. Strains of #’. colt used in these experiments are mutants derived from strain K-12, The productiou, origin, and characteristics of these mutants are summarized in Table I. The strain used for acridine treatment was mainly W6. Recombination technique: Overnight cultures of tester strains are streaked on EM-sugar medium and they are cross-brushed against one loopful of the culture being tested. Recombinants arise at the junction of the two cultures only in com- patible combinations.? Acridine method: An overnight Ft culture is diluted to 104 cells ner ml in a 58 GENETICS: Y. HIROTA Proc, N. ALN. TABLE 1 A SUMMARY OF THE Strains Usgep Strain Number Genotype Original Strain Reference W6 F+M- K-12 4 W1895 Hfr.M~ W6 Type Hfré W2979 F-Mal, ~Xyl:~Gale~Aray~ J-11 (supplied by Dr. Cavalli)! W3133 F-Lac™ Ann Cook W4164 FtLac~ W3133 Obtained by F infection W4171 FtM- W1895 Spontaneous reversion (given by Dr. A. Novick) W 4399 FM~ W1895 F-refractory® (selected in soft agar);!® Low fertility X F~ and does not infect F~ to produce F+ nutrient broth (pH 7.6) containing 20 micrograms per ml of acridine orange (ab- breviated AO-20) and incubated overnight at 37°C. The use of acriflavine for the removal of F has been I , | briefly reported in an r _ earlier publication. AO has proved to be more use- ful than acriflavine, pro- flavine, or acridine yellow by virtue of its low toxic- {00 80 % bo ity. s | The F~ clones obtained = in these experiments have “ been repeatedly studied. ry i ‘| The F- strains so obtained | are genetically stable for | this mating type, like the other F- reported. If any Ft cells had reappeared in the F~ culture by spon- 14 taneous mutation, they 0 50 100 would be amplified by CONCENTRATION OF DYES (x/mi) eross-infection during re- Fic. 1.—Effect of pH on the effective concentration of peated passages, acridines. (a) Proflavine, pH 7.65. (6) Proflavine, pH Eexpere ; 7.20. (c) Acridine orange, pH 7.60. nonce orange, pH xperimental Results. 7.20. An overnight culture of W6 was inoculated into brot 1 Environmental effects (107 cells/ml, pH 7.20 or pH 7.6) containing the indicated dye , . . Se and incubated at 37° C for 20 hours before plating. on the action of acridines: The conversion of sex- compatibility after growth in acridine broth was observed by a standard test of sex-compatibility on isolated clones. The results are generally expressed in terms of the conversion fraction, i.e., the proportion of clones, after treatment which have become stably F~. The pll influences the effect of the dyes; for ex- ample, AO-50 gives 100 per cent F- at pH 7.6 but none at pH 7.2. Generally speaking, the minimum effective concentration of acridine is low at a high pH and vice versa (Fig. 1). All the colonies formed in the untreated control remained F+. Proflavine exerts its action at lower concentrations than that of AO at the same pH. However, the rate of conversion is decreased at higher (bacteriostatic) con- centrations of PF. These pH effects indicate that either the drug cation or an VoL. 46, 1960 GENETICS: Y. HIROTA 59 ~~ T T T a T T OO 1000 e e ° e 4 e 80L 4 se. 1 © 60 « a : a & 40h 4 & 20- 4 L | t | | \ 40° 10! to? 10? to* 108 10° 107 BACTERIA INOCULATED Fic, 2.—Effect of inoculum size on conversion of sex-compatibility. Various dilutions of an overnight culture of W6 were inoculated into broth containing AO (pH 7.6, 20 xg. /ml) and incubated overnight at 37°C. anionic bacterial receptor is effective in the elimination of F. The same relationship between bacteriocidal action of acridine ion and pH effect has been extensively studied by Albert.'? The inoculum size of F* cells also influences the rate of con- version, which decreases with larger inocula (Fig. 2). With an inoculum of about {cell per ml, 100 per cent F~- can be obtained in nutrient broth at pH 7.6 with AO 10 to AO-20. The lowest concentration at which any effect was noted was AO-1 to AO-5. The rate of elimination is reduced at lower temperatures and was essentially zero at 56°C (Table 2). The viability of the cells obtained from the treated culture and the untreated culture at low temperature is the same. TABLE 2 Errect or TEMPERATURE ON CoNnvERSION oF Sex-CompaTIBILITY FROM F+ To F- -——Treated with AO ~ -—— Untreated Control——. Temperature (°C): 37 5 37 5 Total colonies tested: 110 110 110 110 F— obtained: 91 4 0 3 Per cent of F-: 82.7 3.6 0.0 2.7 10? cells per ml of an overnight culture of W6 were inoculated into nutrient broth pH 7.6 and treated for about 20 hours with 50 micrograms of acridine orange, or without it as control. Figure 3 shows the conversion fraction in treated cultures plated on EMB lactose agar at various times. F~ clones first appear after a lag of 1 to 2 hours, and their incidence then increases rapidly. The concentration of AO, pH of the medium, and cultural age of the treated cells influence the rate of conversion and the time for removal of F. Acridines are ineffective on cells growing in a synthetic medium." However, peptone 0.02 per cent, restored the effect. (Table 3). A systematic analysis of 60 GENETICS: Y. HIROTA Proc. N. A. S. 30 fT T T T T “J e— 1 20- 4 ‘ 0 A 8 e = w & & sob - Untreated cortro/ 0} ep——e-== 0 - ener ere Ono ee nnn ee e+ ] L | L 0 60 120 180 240 TIME OF TREATMENT (minutes) Fic. 3.—Time analysis of acridine treatment. An exponentiall growing, W6 culture was inoculated into penassay broth (pH 7.2) wit or without AO (36 yug./ml). The F* cultures (10’ cells/ml), were shaken gently at 37°C. Samples were diluted and plated on EMB agar. The following morning single colonies were tested for their com- patability. The figure summarizes five experiments. About 2,000 colonies from untreated controls were all F+. likely constituents of complex media showed the necessity for the simultaneous presence of several co-factors: amino acids plus a nuclein base. AO eliminated F from F+ bacteria in a defined medium containing the following supplements: serine, aspartic acid, isoleucine, valine, and cytosine. The significance of these co-factors for the cure of maleness remains for future studies. The specific supplements re- ° quired for growth of auxotrophic mutants were also essential for F elimination. TABLE 3 Co-Factor FoR REMOVAL OF F sy AO Minimal plus , —— Per Cent of F~ Obtained ———— Indicated Supplement W6 (M-) W4164 (Prototroph) No supplement 0 0 Peptone* plus AO 100 Lee Mixturet. 7 100 “ minus AO 0 0 “ minus serine Ad 10.1 “ “ eytosine 3.7 0 “ *¢ methionine 0 33.4 * 200 micrograms per ml. + Concentration of each supplement: cytosine 100, serine 20, aspartic acid 20, isoleucine 20, valine 20, methio- nine 20, AO 7.5 (micrograms/ml). Overnight cultures of W6 and W4164 were washed with sterilized water, treated in several media, purified on EMB agar, and their compatibility tested by the cross-brushing method. The experimenta) conditions were as follows: inoculum size about 104 cells per ml, time of treatment 20 to 24 hours at 37°C. Methylene blue and thionine which are structurally analogous (Fig. 4) to AO antagonize the elimination of F but not the inhibition of growth by AO. These dyes alone have no dramatic activity on F— disinfection; F~ clones have appeared in- frequently in treatments with methylene blue. Albert!? has summarized analogous examples of ‘therapeutic interference’ in the typanocidal action of these dyes. Vou. 46, 1960 GENETICS: Y. HIROTA Gl Ss Acridine orange Methylene blue N S "CL" “CL YO" N Proflavine Thionine Fic. 4—Comparison of the chemical structure of several dyes. Propamidine isethionate (Baker and May Ltd., 4:4’-diamidino-diphenoxypropane di-(8-hydroxyethane sulfonate)) is another chemotherapeutic agent which sup- posedly reacts with nucleic acids. At 50 micrograms per mi in penassay broth it was found to produce about 15 per cent F— after overnight treatment, and.is there- fore a weak agent affecting F. It is concluded that acridine mainly produces its effect on actively. multiplying cells: in the absence of cell multiplication, i.e., with large inocula, low temperature, nutritional limitation, or bacteriostasis by excess dye, there is little or no loss of F. 2. Mode of action of acridine: The increase in the proportion of F~ after acridine treatment could be attributed in principle either to mutagenic effects of acridine dyes or selective enrichment of F~ cells already present among the F+ population. No differential in growth of F+ vs. F- cells could be found even remotely approach- ing the dramatic differences needed to account for complete conversions (see Table 4). However, the most decisive experiments concerned the fate of small numbers of cells inoculated in AO medium. TABLE 4 RECONSTRUCTION EXPERIMENT _ —-— Input Mixtures - Mixed Cultures after Incubation-———~ Lac*+ F~ Lact F+ Lac~ F- Lac7 Ft Lac* F- Tact F+ Tac” F~ Lacy F* A Treatment 0.329 0 0 0.671 0.471 0 0.529 0 (W2979) (W4164) (W2979) F-elimi- nation Untreated 0.363 0 0 0.637 0.246 0.152 0 0.602 Control (W2979) (W4164) (W2979) F-infec- (W4164) tion Overnight cultures of W4164 (F +) and W2979 (F—) were mixed and about 104 cells/m Jinoculated into broth, pH 7.6, with or without AO, and incubated overnight at 37°C. The proportion of Lac + and Lac™ bacteria in the mixed cultures was diagnosed on EMB lactose agar before and after the treatments; the F status of samples of isolated colonies was tested by crosses against a standard F~ strain (see Materials and Methods). The figures in this table show thé fraction of colonies in each class. The mutagenic effect of AO was tested more directly by cultivating one or two cells in a microdroplet of penassay broth, with or without acridine dye, in an oil- chamber as described by Lederberg.'4 The microcultures were then allowéd to grow for several generations. Each microclone was transferred to EMB lactose agar by a capillary pipette and plated out to give single colonies arising from in- dividual cells. The colonies were then tested for sex-compatibility. The entire clone was plated; thus, differential growth of spontaneous F~ mutants could alter the proportion of F- progeny within a clone but should not affect the number of 62 GENETICS: Y. HIROTA Proc. N. A. 8. TABLE 4 InpuctIon or F~ From Sinezy Isovatep F+ Cent anp F+ CEs Number of Cells Number of Drops Number of Drops in Drop Initially with Indueed F- Tested Acridine-treated cell(s) 1 3* 25 2 bt 20 Untreated Control 1 0 25 2 0 20 An exponential culture of strain W6 was grown at 37°C in penassay broth with gentle shaking. This culture contained 109 cells per m}. Single cells were then isolated in micro- droplets hy a simple modification of De Fonbrune’s oil chamber method. !4 Droplets were dis- pensed by free hand manipulation from a suspension containing 10¢ cells per ml. with or without AQ-40, The droplets were then examined microscopically, and those which were verified to contain precisely one cell or two cells then recorded. They were incubated at 37°C for 4hours. At the end of this time, the cells that had grown in each drop were counted. then collected by a capillary pipette and spread on EMBagar. All the colonies formed on the agar were tested for sex-compatibility. If one or more F~ colonies were observed, this was recorded as ‘‘containing induced F >." By Fisher’s exact test, P = 0.0056. Total numbers of F ~ cells in “induced drops’’ were: * 11,1. tT 20,2,4,1,2. and the total cell nuinbers of those induced droplets were: 33,20,10; 72,10,34,11,71, respectively, clones in which F~ progency are found. As shown in Table 5, AO increases the proportion of F--containing clones from 0/45 in controls to 8/45 in the treated series. This supports the conclusions that AO induces the loss of F from growing FF cells. 3. The action of AO on various male strains: A wide variety of F+ strains mu- tants of E. cola K-12 have been tested. The markers such as nutritional require- ments, drug resistance, lysogenicity, phage resistance, sugar fermentation seem to have no bearing on the effect of AO nor are they altered by the treatment. How- ever, Ifr males differ markedly from F+. For example, Table 6 shows a result of TABLE 6 ACCESS'BILITY OF Two Staves oF F in Various STRAINS TU DISINFECTING ACTION OF ACRIDINE TREATMENT W4171 We W1895 I+ Reversion W4399 Fe Her from W1895 F_Refractory Can infect standard F~ + _ b _ Mode of inheritanee of Nonsegregational Segregational Nonsegregational Segregational muleness Compatibility with + +++ +4 + standard F~ Compatibility status yes no yes no affected by acridine treatment acridine treatment on four male types all derived from W6. Mutants which show chromosomal segregation and non-infectivity of the male character are not sensitive to the disinfecting action of acridine dyes. This inaccessibility is observed both in stable Hfr and the F" mutants (summarized by Lederberg and Lederberg). Hirota and lijima’ reported that AO gave F~ types not only from F* but also from Hfrz* However, this culture has proven to be extremely mutable and sponta- neously produces many F+ reversions. The F~ obtained from cultures labelled Hfr, can be accounted for by their origin from such F* reversions. The F*+ state of the mutable Hfr;* is also accessible to this treatment. Stable Hfr strains, e.g., Hfr, have not been influenced by exposure to AQ. On the other hand, the other Vou. 46, 1960 GENETICS: Y. HIROTA 63 mutants which maintain a contagious F are readily disinfectable by this treatment. F elimination by AO thus differs from the correlated selection of highly motile F- bacteria in soft agar, which selects F- not only from F+ but also from stable Hfr strains. '6 Discussion..-In this phenomenon, many hypotheses may be considered. From the above experiment, however, it seems difficult to escape the conclusion that induc- tion of F~ from F+ is the result of the loss of F particles endowed with genetic continuity. Whether the relative multiplication of F is decreased, or the F particles are agglutinated by combination with the acridinium cation, growing cells exposed to AO eventually produce bacteria without F. Stability of F~ means a permanent loss of F, and conversely the stable main- tenance of the F'* trait and the remarkable infectivity of F in usual media signify that F multiplies faster than the host cell. The transfer of F unlinked to the host chromosome, and the non-segregational uniform inheritance of F in the cross F+ x F-,*~6 would most logically correspond to cytoplasmic transfer of the male deter- minant.'7 '§ That is, the factor in an F+ cell is a plasmid. 7 18 Other plasmids, such as the respiratory factors in yeast, kinetoplasts in trypanosome are also known to be accessible to acridine treatment. !9—22 On the other hand, maleness in Hfr cells is inherited chromosomally as a gene linked to certain markers in genetic recombination tests.'~? Correlated with this is the fact that Hfr males are resistant while infective F+ males are accessible to the acridine treatment (Table 6). These facts suggest that in Hfr strains this male determinant is bound to the particular chromosome site, and thus accounts for the non-infectivity and segregational inheritance of maleness, and its inaccessibility to the disinfecting action of acridine dyes. This situation parallels the lysogenic system of bacteria in which there are two states, vegetative phage, free in the cytoplasm and prophage, bound to the chromo- some. The term episome has been used for this type of particle.® Acridine dyes are reported to cure virus infection, produce defective virus, and induce virus mutants respectively.23-27 However, they do not affect the chromo- somally bound prophage of temperate phages.> The detailed mechanism of action is, however, not known. These dyes do possess the unique property of straining certain nucleic acid-containing particles, nuclei, and the other particles of living cells, in several organisms.*° Nucleotides, nucleic acid, and polyadenylic acid form complex salts with acridine dyes.*° 8! The effect of acridine orange on the function of DNA as a primer for its enzymatic replication has not been studied. The co-factors required for F-disinfection may act in part. by supporting the growth of the host cell in its particulate components in the presence of AO. The F particle itself may have a negative charge, so that acridine exerts its effect only as a cation, Summary.-- Acridine orange converts F+ (male) clones of &. cold into stable F- (female) forms. The main factors which affect the rate of conversion of F+ to F- are the concentration of acridinium ions in the treating medium and the growth of the treated cells in the presence of the dye. Furthermore, the accessibility of F to acridine treatment depends upon the state of F in the host cell. Acridine orange increases the frequency of conversion from F+ to F- directly without appreciable selective growth. The conversion from F+ to F~is irreversible, 64 GENETICS: Y. HIROTA Proc, N. A. 38. as expected for the loss of a genetic particle. Hfr males are resistant to the disinfect- ing action of acridine dyes. . These results are well accounted for by the dual nautre of F, chromosomal F, and plasmid F. I wish to express my heartiest thanks for the guidance given by Professor Joshua Lederberg in this study at the University of Wisconsin. I am also indebted to Dr. Esther M. Lederberg for her helpful criticism of the manuscript. I also wish to thank Professor Hideo Kikkawa of the University of Osaka for his advice and encouragement during the course of this study. * This work has been supported by research grants from the National Cancer Institute of the U.S. Public Health Service and from the National] Science Foundation. + This study was done in partial fulfillment of the requirements for the degree of Doctor of Science at the University of Osaka. 1 Cavalli-Sforza, L. L., and J. L. Jinks, J. Genet., 54, 87 (1956). 2 Richter, A., M.S. thesis, University of Wisconsin, Madison, Wisc. (1957). 2 Wollman, E. L., F. Jaccb, and W. Hayes, Cold Spring Harb. Symp. Quant. Biol., 21, 141 (1956). 4 Lederberg, J., L. L. Cavalli-Sforza, and E. M. Lederberg, Genetics, 37, 720 (1952). 5 Cavalli-Sforza, L. L., J. Lederberg, and &. M. Lederberg, J. Gen. Microbiol., 8, 89 (1953). 5 Hayes, W., J. Gen. Microbiol., 8, 72 (1953). 7 Hirota, Y., Nature, 178, 92 (1956). 8 Hirota, Y., and T. lijima, Nature, 180, 655 (1957). 9 Jacob, F., and E. L. Wollman, Compt. rend. Acad. Sct. Paris, 247, 154 (1958). Cook, A., M.S. thesis, University of Wisconsin, Madison, Wisc. (1959). 4 EM agar is a synthetic medium containing eosin and methylene blue. With the omission * of succinate it corresponds to EMS agar as described by Lederberg, J., in Methods in Medical Re- search, ed. R. W. Gerard (Chicago: The Year Book Publishers, Inc., 1950), vol. 3, p. 5. 12 Albert, A., Selective Toxicity (New York: John Wiley and Sons, Inc., 1951). 44 Davis, B. D., and E. 8. Mingioli, J. Bacteriol., 60, 17 (1950). 14 Lederberg, J., J. Bacteriol., 68, 258 (1954). 15 Lederberg, J., and E. M. Lederberg, in Cellular Mechanisms in Differentiation and Growth. ed. D. Rudnick (Princeton: Princeton University Press, 1956), p. 101. 16 Skaar, P. D., A. Richter, and J. Lederberg, these PRocEEDINGS, 43, 329 (1957). 7 Lederberg, J., Abstr. 7th Int. Cong. of Microbiol., Stockholm, 58 (1958). 18 Lederberg, J., in The Harvey Lectures (New York: Academic Press, 1959), 69. 1 Ephrussi, B., Nrucleo-cytoplasmic Relations in Microorganisms (London: Oxford University - Press, 1953). % Lederberg, J., Physiol. Rev., 32, 402 (1952). 21 Lwoff, A., New Phytologist, 49, 72 (1950). 22 Inoki, 8., Proce. Int. Genetics Symposia, Tokyo, 550 (1956). 23 De Mars, R. 1, 8. E. Luria, H. Fisher, and C. Levinthal, Ann. Inst. Pasteur, 84, 113 (1953). 24 Dulbecco, R., and M. Vogt, Virology, 5, 236 (1956). % Foster, R. A. C., J. Bacteriol., 56, 765 (1948). % Franklin, R. H., Virology, 6, 525 (1958). 27 Briody, B. A., and C. Stannard, J. Immunol., 67, 423 (1951). 8 Bertani, E. L., Virology, 4, 53 (1957). 29 Armstrong, J. A., Exptl. Cell Research, 11, 640 (1956). %* Morthland, F. W., P. P. H. De Bruyn, and N. H. Smith, Epil. Cell Research, 7, 201 (1954). 31 Beers, R. F., D. D. Hendley, and R. F. Steiner, Nature, 182, 242 (1958).