Reprint from Genetics, Vol. 47, No. 10, October, 1962. Printed in U.S.A. PATTERNS OF SEXUAL RECOMBINATION IN ENTERIC BACTERIA? P. H. MAKELA,? J. LEDERBERG anp E. M. LEDERBERG Department of Genetics, Stanford University School of Medicine. Palo Alto, California Received May 29, 1962 TRAIN K-12 of Escherichia coli has played a preeminent role in the study of bacterial sexuality. New knowledge of the mechanism of sexual differentiation and the development of more sensitive techniques and test strains have subse- quently brought many more bacteria within the orbit of this breeding system. The immunogenetics of Salmonella poses many interesting problems (LEDERBERG and Irno 1956; Inno 1958, 1961a,b; LepERBERc 1961) that could be only partly analyzed by methods of phage-mediated transduction. This paper presents a sur- vey of crossing behavior in Salmonella and some other enteric bacteria which was conducted as a basis for the further study of flagellar phase variation in Salmonella. Sexual recombination in E. coli is dependent on a fertility factor F which confers the property of maleness on cells carrying an F particle either in the cytoplasm or fixed to the chromosome (LEDERBERG, CavaLLI and LEDERBERG 1952; Jacop and Wotitman 1961). The impact of F is expressed in at least two ways: the modification of the cell surface allowing for the conjugation reaction of an F+ with F- acceptor cells, and the impulse to the chromosome to migrate from the male partner via the conjugal bridge to synapse and crossover with the corresponding chromosome of the female partner. Even in the F+ cell where the F particle is characteristically extrachromosomal, it probably forms at least a temporary association with the chromosome in those cells actually involved in conjugation. In general, the point on the chromosome at which the F particle is located tends to be the last segment to be transferred during an orderly progressive process of conjugal exchange, perhaps on account of a fixation of the F factor that binds the chromosome to a position on the cell surface whose modification is involved in the formation of the conjugal bridge. The F particle sometimes acquires a translocated fragment of chromosome, a few recognizable markers now sharing the contagious transmission of the F ele- ment (Jacop and AprLBerG 1959; Hirota 1959). These compound F ele- ments, designated F” (F prime) have the advantage that their transmission can be more readily followed through the diagnosis of the translocated markers. They 1 This work has been supported in part by grants from the National Science Foundation from a Graduate Research Training Grant in Genetics (G2-295) and research grants C-4496 from the National Institutes of Health, U.S. Public Health Service. ? Present address, Department of Serology and Bacteriology, University of Helsinki, Helsinki, Finland. Genetics 47: 1427-1439 October 1962. 1428 P. H. MAKELA, J. LEDERBERG AND E. M, LEDERBERG also confer a very efficient transfer of the translocated markers. Following the preliminary reports of fertility of Z. coli with Salmonella (Baron, SprLMan and Carry 1959; Baron, Carey and SpILMAN 1959a,b; Miyake and Demerec 1959; ZINDER 1960a,b) we thought to explore the range of sexual competence in Sal- monella by following the transmission of an F’ that efficiently transfers the Lac markers. MATERIALS AND METHODS The cultures used in this investigation are listed in Table 1. TABLE 1 Strains used Our strain . number Description* References Escherichia coli, K-12 derivatives W 6 F*Lac*M- LEDERBERG, ef al. (1952) W 1895 Hir, LactM- Cava, LEDERBERG and LEDERBERG (1953) W 3287+ F,,*LactM-S" W 3637 LactM-S* Orsxov, et al. (1961) W 3747 F,.’LactM- Hirota (1959) W 3876 Q ,Lac-S* Ricuter (1961) W 4145 Lac~,, Cook and LEDERBERG (1962) W 4678 Lac” ,,P-S* Cook and LEDERBERG (1962) W 4680 Lacy 49S" Cook and Leperserc (1962) Salmonella T 2 typhimurium (Lilleengen No. 85) Srocxer, Zinper and Leperserc (1953) its derivatives: S™(SW 1342), F,,° (SW 1346) SW 685 paratyphi B (derivative of SW 543 SrockeEr, ef al. (1953) of Srocker et al., originally Kauffmann No. 223) its deriva- tives: S7(SW 1390), F,..*(SW 1343) SW 753 bovis-morbificans 3640 Epwarps and Bruner (1942) SW 764 enteritidis 1891 Epwarps and BRUNER (1942) SW 776 london 1446 Epwarps and BruNER (1942) SW 777 give 316 Epwarps and BRUNER (1942) SW 779 muenster 4546 Epwarps and BruNER (1942) SW 787 senftenberg 3007 Epwarps and BruNER (1942) SW 790 aberdeen Eowarps and Bruner (1942) SW 791 poona Epwarps and Bruner (1942) SW 795 hvittingfoss Epwarps and Bruner (1942) SW 803 abony 74 Epwarps and Bruner (1942) its derivatives: S"(SW 1353), Ft (SW 1351, 1364, 1463), F,,*(SW 1485).F 55 stable’ (SW 1365, 1486), Hir(SW 1462), M-S"(SW 1361), P-S*(SW 1355), Gal-H-i:1,2 S?*(SW 1464), Mal-Ara-S” (SW 1417), etc. INFECTIBILITY BY FACTOR F 1429 TABLE 1-—Continued SW 1214 typhimurium TM-9S"-2 Baron, et al. (1959b) its derivatives: T-Tyr-(SW 1259), F,,*T-Tyr (SW 1372) SW 1338 adelaide Nossat and Leperperc (1958) SW 1394 java No. 5, obtained from Orskov, et al. (1961) F. Orskov as fertile with Hfr colt SW 1395 miami No. 187 obtained from Orskov, et al. (1961) F. Orsxov as fertile with Hfr coli Shigella W 1779 sonnei S3 (P9) is Niacin- Fréprricg (1948) its derivatives S*™(W 4973), F,..* H 1 flexneri its derivatives: S’, F,,* Klebsiella K 1 Serratia marcescens SM 6 its derivative: F,.,*, Fa.xow, et al. (1961) very unstable SM 6-S7-11 Sr Fa.xow, et al. (1961) W 2745 fecal isolate, also listed as Waisman and Stone (1958) CDC 184/55 * All stocks prototroph, streptomycin sensitive, Zac-, and without a demonstrable F factor if not otherwise indicated. Abbreviations: S' =resistant to streptomycin 200ug/ml Lac=lactose, Gal= galactose, Ara= arabinose, Ma/— maltose. += fermenting, ~==-nonfermenting. Growth factor requirements: M-=methionine, P-=proline, H-=histidine, T- = threonine, Ty7-= tyrosine requiring. Mating types: F, F+, Hfr, F,;*, Frastapie’ ave described in the text. The references should be consulted for the presence in some strains of additional markers immaterial to the present work. . 7 Test strain used by Orsxov, et al. (1961) W 3287 and 3747 both acquired their F,, from a strain W 3213 isolated in 1955 by Leprrserc (unpublished) as an unstable, hyperfertile male derivative of W G. Cultural procedures are detailed elsewhere (LEDERBERG 1950). “EM” agar (EMS agar without succinate) was frequently used as a combined selective and indicator medium. It is a synthetic medium with a given sugar as sole carbon source, and also contains eosin and methylene blue to delineate prototrophic sugar-positive colonies, Recombinants were selected on minimal agar plates at the intersection of drops or streaks of the parent cultures, or from cell suspensions mixed in broth for 30 minutes at a density of about 5 X 108 male and 2 x 10? ‘female cells per ml. The plates were scored after two days incubation at 37°C. To test for F” infection, spot tests equivalent to crossing tests were made on lactose selective media. For greater encouragement of F’ transfer, mixed cultures were incubated in broth either for 30 minutes or overnight, centrifuged and spread on EM Lac plates. In the latter case, 10° cells of the minority parent were plated. Since in this experiment, the F’ can migrate not only from the donor to recipient, but also from one infected recipient to others, it is not possible to give precise frequencies of infection and the results are expressed as plus or minus. Several mutually confirmatory tests were routinely conducted for the success- ful transfer of F to a new Salmonella culture (A) by crossing with known female indicator strains, (B) by infective transfer to known female strains, (C) by irans- fer to a special indicator strain, 2, (RicuTER 1961) which is especially advan- tageous as the acquisition of F results in an unusually fertile é, that can be 1430 P. H. MAKELA, J. LEDERBERG AND E. M, LEDERBERG efficiently detected in situ (SNEaTH and LepErBerc 1961), (D) by a staining reaction on EMB agar plates (compare ZrnpER 1960b): F* strains of many Salmonella types can be distinguished on EMB agar without fermentable sugar giving purplish as compared to white or bluish colonies of F-. The color difference was best seen after 18 hours at 37° followed by 24-48 hours at room temperature, the plates being observed by oblique lighting. For the disinfection of F by acridine orange (Hirota 1960) these conditions were used: overnight incubation in acridine orange-nutrient broth, pH 7.6 start- ing from a small inoculum of 100-10,000 cells/ml. For f-agglutination test, the Hfr and F- sera described by @nsKov and OxsKov 1960, and kindly furnished by them were used according to their instructions in both slide and tube agglutination tests. EXPERIMENTS AND CONCLUSIONS Many previous attempts to demonstrate sexuality in Salmonellas were un- successful (ZINDER and LepERBERG 1952), Later successes depended on a fortu- nate choice of Salmonella strain as the initial female parent and on the use of appropriate highly fertile male strains of E. coli in conjunction with suitable diagnostic markers. We are particularly indebted to Dr. L. S. Baron and Dr. N. Zrnver for early information on their findings. Further matings in Salmon- ella depend on the successful transmission of the F particle, whose provenience was usually strain K-12 of E. coli, to competent Salmonellas which would then act as males. The first report of E. coli x Salmonella, (Baron e¢ al. 1959a) involved the unique strain of Salmonella typhimurium, TM 9-S'-2 which was highly fertile with FE. coli W1895, an Hfr,; male. However, the progeny of this cross were generally interfertile with many other Salmonellas. The immunogenetic factors in which we are especially interested, H, and H,, had not been definitively mapped, nor could we succeed in demonstrating the segregation of H, or H, in these crosses which did involve a substantial segment including the Lac marker. This provocative cross was therefore futile for these purposes and further work was focused on achieving (a) general fertility of Salmonella x Salmonella mat- ings and (b) segregation of a wide range of markers, especially H, and H,. To establish appropriate strains it appeared necessary first to introduce a typical infectious F particle. In due course it was found possible to do this with a number of Salmonellas. However, the initial survey stressed the behavior of the technically favorable F,, particle which is readily recognized by the associated transmission of the lactose positive phenotype. This character is especially apt for work with Salmon- ella as most naturally occurring serotypes of Salmonella are inherently lactose negative. They therefore require a minimum of prior laboratory manipulation to make them ready for experimental tests. The experimental regime was to culti- vate an auxotrophic F,;Lact donor strain with a prototrophic Lac” acceptor strain and then selectively search for prototrophic Lact progeny by plating on EM lactose agar. INFECTIBILITY BY FACTOR F 1431 In K-12, F,, infection leads to the establishment of moderately stable hetero- genotes, i.e., partially diploid cells carrying the F,, with its attached segment in addition to the original haploid chromosome. The heterogenotic state is revealed by subsequent segregation of new phenotypes: Lacy F- (F’ lost); Lac’ F+ (F’ particle broken with disappearance of Lac+ but retention of F+); Lac+ stable (by integration of Lact into the chromosome), in addition to the parental Lact F’ type. All Lact progeny from F,,-Lact+-infected Salmnonellas, and more than 1,000 have been purified and examined on EMB lactose agar, have been heterogenotic in respect to Lac. Furthermore, they have been much less stable than correspond- ing K-12 F,,Zact, the degree of stability varying with different recipient species. The segregants have all been Lac-, either F- or F+; in no case have stable Lact been observed which would correspond to the integration of the Lac+ fragments in the chromosome. In Table 2 are given the proportions of Lac” segregants when F,,-infected clones are transferred in broth. In K-12, an occasional clone is TABLE 2 Segregation of Lac- from F’ . Lac*-infected clones Infected with F,, stable clones Unstable clones Tefected with Strain cette one Jaws Nemes Tact MenesEaoe Salmonella abony SW 803 1 day t 0.5 tt 4-50 12 99 99 99 <1 Purified F’ containing clones were grown in broth with daily transfers to fresh medium, and periodically plated on lactose-indicator media for counting the proportion of Lac- to total colonies. One day’s growth corresponds to 20 generations, observed from which F,, has disappeared, while most continue to segregate Lac™ at a frequency of less than one percent, suggesting a stable equilibrium between the F’ particle and its host cell. In most Salmonellas on the other hand, F,, gradu- ally disappears. Because it would have had ample opportunity of infecting new cells and spreading through the culture this would suggest that cells that have lost F,, remain immune to it, or that F;, multiplies more slowly than the host and is gradually diluted out. The first possibility is contraindicated since isolated Lac” segregants from such cultures are readily reinfectible with F,;. Some clones of Salmonellas, as shown in Table 2, show more stable associations of F,,. In fact, Salmonella miami, which is also rather easily infected with F from 1432 P. H. MAKELA, J. LEDERBERG AND E. M. LEDERBERG K-12, gives a majority of stable clones. One such was also picked in S. abony after 13 Lact reisolations and subjected to further study. In this case the F,,; seems to have been modified permanently: infecting almost any strain it would give a heterogenote in stable equilibrium continuously segregating Lac” at a low rate of about 0.5 percent. This F,; mutant was called Fy; stanie and for infectibility study was transferred into suitable Salmonella and E. coli stocks. F-infectibility of various enteric bacteria: A number of Salmonella, Shigella, Serratia and Klebsiella strains were tested with F,, and Fy; s:apie from K-12 and from Salmonella abony. As usual Lact transfer was used as an indication of the TABLE 3 F’ . Lac’ infectibility of various enteric bacteria Donors Lact Af- 1 2 3 4 K-12 K-12 S.abony S. abony Recipient Lace Fy Fis atable Wa 13 stable E. coli K-12 W 4145 10-- 10-1 10-2 10-1 Salmonella Group B abony SW 803 10-6 10-% 10-8 10-2 typhimurium TM 2 10-7 10-6 10-4 10-* TM 9-S'-2 SW 1214 10-" 10-2 10-4 10-# paraty phi B SW 685 + . 10-6 >10-5 java SW 1394 10-7 . : 10-5 Group C bovis-morbificans SW 753 + . . 10-6 Group D enteritidis SW 764 — + 10-7 10-* miami SW 1395 10-5 .. 10-2 Group E london SW 776 10-8 . . >10-5 give SW 777 — + 10-6 10-4 muenster SW 779 10-8 10-6 10-# 10-8 senftenberg SW 787 10-7 10-6 .. 10-° Groups F,G,Letc. aberdeen SW 790 10-7 . . >10-5 poona SW 791 — + 10-7 10-6 Avittingfoss SW 795 10-7 . >10-5 adelaide SW 1338 — + 10-7 10-7 Shigella flexneri Ht 10-3 10-2 10-3 10-2 sonnet SW 1779 . a 10-3 10-1 Klebsiella K 1 + . a + Serratia marcescens SM 6 + -+- + + SM 6-S'-11 10-7 10-7 107 10-7 SW 2745 10-6 10-6 . 10-6 Donor (100 parts) :recipient (1 part) mixtures were plated after 30 min contact in broth on minimal lactose agar. Number of infected cells (=colonies growing) is expressed as fraction of recipient cells plated. If this test was negative. a_ prolonged time of incubation was used, and approximately 108 recipient cells were plated; the results in this case are given as + or —. INFECTIBILITY BY FACTOR F 1433 F’ infection, The results are shown in Table 3. In column 1 are given the frequen- cies of infection with K-12 F,,+ as donor. Large differences in the infectibility of the various species are evident, ranging from 10~ to less then 10-* under the experimental conditions. Eighteen of 23 strains tested could, however, be infected. With Fy; stavie (column 2) frequencies are augmented 10 to 100-fold except for the Serratia species. Thus F,, .:an1. originally adapted to S. abony has an advantage in other Salmonellas and also in Shigella and £. coli K-12. The same advantage is seen as the difference between columns 3 and 4 where F’-infected S. abony was the donor. E. coli K-12 can be compared directly with S. abony as an F donor: columns 1 and 3 versus 2 and 4 of Table 3. When S. abony is the donor, every one of the 23 strains tested can be infected with F,,. K-12 is equally well infected from either donor, but in all Salmonella * Salmonella combinations there is a difference of 10-10* in favor of the Salmonella donor. This may be attributed to a specific surface compatibility of Salmonellas in conjugation with other Salmonella. A comparison of reciprocal crosses suggests a complex pattern of breeding compatibility: Therefore, the reciprocal infections were expanded to include addi- tional donors (Table 4). The first two columns come directly from Table 3, the third column refers to S. typhimurium TM2 as the donor, This strain, which is widely used in transduction studies in combination with phage P22, is a very poor donor of F’ either in homologous or heterologous combinations. However, F,, is quite stable in TM2 in these conditions and one can only guess that the effectiveness of the F particle in altering the surface for male conjugal function varies from one background genotype to another. In the fourth column is repre- sented S. typhimurium TM9-S’-2, the recipient strain of Baron et al. 1959b, which is a good donor both to F- forms of the same strain and to K-12. The same pattern is shown by S. paratyphi B. The Shigella strains were very effectively infected from F’ K-12 and Salmonella, as well as in the homologous combination (frequencies 10-°-10~), while Serratia is very poorly infected in all combinations. TABLE 4 F ,, infection in homologous and heterologous combinations Donors: F,,-infected clones of E. coli Salmonella Shigella Serratia . ™. para- Recipients K-12 abony TM2 = 9-St-2 typhi B flexneri sonnei marcescens E. coli K-12 10-2. 106 102 10-4 —(<10-7) .. 10-§ Salmonella abony 10-6 10 10-7 106 107 396104 10-8 ™ 2 10-7 10-4 10-7 +107 TM 9-St-2 10-3 10-*+ 10-6 10-7 paratyphiB +(<10-8) 10-6 10-6 . Shigella flexneri 10-3 10-3, .. .. 10-3 sonnei . 103, .. .. 10-4 . Serratia marcescens 10-7 10-7 .. .. . . .. —(<10)] Infection done by growing together (30 min) 100 parts of donor and one part of recipient, plated on minimal-lactose- streptomycin medium. Number of growing colonies expressed as fraction of recipient cells. Homologous combinations are in squares along a diagonal. 1434. P. H. MAKELA, J. LEDERBERG AND E. M. LEDERBERG There thus seem to be three relevant genotypic statements for a given mating test: (1) female strain, (2) the strain which has acquired an F particle to become male, and (3) the quality (origin and history) of this F particle. To recapitulate, K-12 seems to be a universally good recipient for F infection; one is bound to recall that the F particle used in all these experiments originates in this strain. No universally competent donors have been found. Most strains are effective donors in a homologous combination and to strain K-12. Their ability to accept F from K-12 or from other species varies subject to alteration by complementary mutants like the aforementioned TM9-S"-2. These can be selected for by crossing E. coli with the species in question. Baron has described these variants which are more fertile than the original population as “F- mutants” from an F° status (Baron et al. 1959b). However, this designation applies peculiarly to the reaction with K-12 as the source of F. While more compatible mutants must be assumed to occur, they are not the principal factor in the frequency of F transfer. Disinfected Lac segregants ob- tained from a number of F,,-Lact heterogenotes of E. coli X Salmonella exhibit the same fertility as the original Salmonella strain. This was also true of Lac” segregants from S. abony Lact derived from a cross with E. coli Hfr W1895 (Table 5). Infection with wild-type F: There is no prior basis to expect different compati- bilities of F from F,,, but more effective methods are needed to detect the F- infected cells. The presence of standard F need not always be manifested by observable fertility of the F-carrying strain in a new species, and, as in Shigella (Luria and Burrous 1957) may have to be demonstrated by transfer back to TABLE 5 Fertility of Salmonella clones after a previous mating with E. coli K-12 Frequency of Lact progeny from crasses with donors: K-12 Hfr K-12 F+ S. abony Fy seapie* Recipients W 1895 W 3747 SW 1365 S. abony clone 1 ow. 10-6 10-2 clone 2 coe. 2x 10-6 3x 10-2 Lac” segregant from clone 1 F,,** Loa. 5 «x 10-7 3x 10-2 clone 2 F,,** Loe 10-6 10-2 clone 1 Lact+ . 2 > 10-6 10-2 S. typhimurium TM-9-S*_2 clone 1 10-4 2x 10-3 2x 10-2 clone 2 5 x 10-5 4x 10-3 10-2 Lac~ segregant from clone 1 F,,** 2x 10-4 3 x 10-3 3x 102 clone 2 F,,** 2x 10-5 3 10-3 10-2 * From a cross with W 3747. 7 From a cross with W 1895. fac+ progeny was selected on minimal lactose streptomycin plates afler 30 min incubation of mating mixtures with a 100-fold donor excess in broth. Results are expressed as fraction of Lact of recipient cells plated. As recipients we used two single colony isolates of S$. abony and S. typhimurium each, and a fac” segregant from Lact derivatives of these obtained after crossing with either Fy, or Hfr K-12, INFECTIBILITY BY FACTOR F 1435 E, coli, Unlike the F-associated characters which mark the F’ particles, standard F confers no advantage we might readily use to select a small number of infected cells. However, we have to rely on the contagiousness of F to gradually enrich for F+ cells in mixed populations even though its rate of spread might be even- tually limited (Table 2). For the detection of F in new strains back transfer to E. coli was customarily used as an ultimate criterion and was greatly facilitated by the use of the ?; detector (Ricuter 1961). The color differential on EMB agar, which was more reliable in S. abony than in S. typhimurium, also was particularly helpful in detecting the segregation of F- in F+ clones. F- could always be found to the extent of at least one percent in F-infected S. abony. The color test corresponded very well with other tests for F as repeatedly confirmed (Table 6). However, it can be easily confused with other sources of color variation (e.g. S>R (smooth to rough) shows a color difference in this system) and it cannot be relied upon for the diagnosis of F independently of other evidence. Experiments on the transfer of F are summarized in Table 7. F is quickly transmitted from S. abony to the homologous recipient as well as to S. muenster TABLE 6 Correlation of different tests for the presence of F Agglutination in No. of Infective sen serum Staining on Stock colonies transfer to? , 1:80 1:10 EMB ++ ++ + + ? _— S. abony F- 95 0 0 0 0 9 86 Ft 36 36 36 0 31 5 0 F,, 5 5 5 0 5 0 0 Hir 1 1 1 0 i 0 0 S. typhimurium TM-9-S7-2 F- 7 0 0 0 0 7 0 a 8 8 0 0 0 8 0 Single colonies from EMB plates, where the staining reaction (purple +, white —) was scored, were picked up in broth and grown overnight. Drops of these were tested by infective transfer to @,, and in tube agglutination tests. For details see Materials and Methods. TABLE 7 Infection of wild-type F into various species . . Donor K-12 F+ W 6 Donor S. abony F+ SW 1364 Time of snixed culture (number of transfers, each 9 generations) 1 2 3 7 1 2 3 7 Recipients E, coli W 4145 +t t+ th te th $+ t+ 44 S. abony SW 803 — + +h ++ $+ +4 44 44 S. typhimurium T™ 2 — — — — — +e 4" 4% S. muenster SW 779 — — —_— — t+ 44 44 44 Serratia marcescens SM 6-S’-11 — — — — * Weak reaction. Donor (in 100-fold excess) and recipient grown together in broth with serial transfers of 0.1 ml to 30 ml of fresh broth. F character of recipient cells reisolated at various times is tested by infective transfer to 2, (see Materials and Methods), recorded as ++ if majority of cells are F+, as +, if less than ten percent are H+, and —, if none of 500 cells tested are male. 1436 P, H. MAKELA, J. LEDERBERG AND E. M. LEDERBERG and E. coli but only slowly to S. typhimurium TM2. However, the assay of TM2 F+ might be hindered as already noted with F,,;. S. abony is also infected from E. coli K-12 although more slowly than from the homologous donor. We did not demonstrate infection of Serratia marcescens. All in all these results agree with the data from F,; infection studies taking into account that wild-type F is less readily detected than F,;. Properties of F-infected strains: Most of these studies have been carried out in S. abony; the male and female cultures have behaved rather as in E. coli K-12. The F+ strain is infective at high efficiency to the homologous strain, less to others (Tables 4 and 7). F+ can be disinfected with acridine orange but higher concentrations are required than with K-12 and full disinfection of the culture is not achieved (Table 8). In E. coli, female cells have been observed to move less rapidly than males (Skaar, RicuTer and Lepersere 1957). Although the differ- ence in motility is not impressive, it permits the practical selection of F- by pass- ing the stock through one or several motility agar columns of 5 cm. In S. abony the inhibition of motility by F+ is much more marked, in fact, male cultures are usually very poorly agglutinated by antiflagellar antiserum (which does not bode well for the use of F-mediated crossing for studies of the immunogenetics of the H antigen). The agglutination can be restored by selection through 5 cm of motil- ity agar, but the F character is usually lost at the same time. OrsKov and Orsxov (1960) have demonstrated a new antigen on the surface of male £. coli by means of specific agglutinating antisera (F*+ or Hfr sera). This Hfr serum also agglutinates S. abony males to a titer of less than 1-100 (36 Ft, five F,, and one Hfr were tested). The reaction of these with F- serum as well as the reaction of eight female strains with both these sera were negative in a serum dilution of 1-10 (Table 7). The f+ antigen could not be detected in F+ S. typhimurium TM9-S"-2. Colonies of male strains also tend to be rougher than female ones. This has been described by Maccacaro (1955) for K-12 and has been our experience in S. abony as well, The effect might be partly explained by the selection of rougher TABLE 8 Effect of acridine orange on F-infected coli and Salmonella cultures Fraction of F+ cells after overnight growth with the concentrations of acridine indicated: Experiment Stock 0 20 xg /ml 40u¢/ml 80 ze/ml 160 ze /ml growth I E.coli Ft 140/140 20/456 9/306 2/201 inhibition Il E.coli F* 97/97 44/71 I S. abony F* 158/158 80/89 29/81 It S. abony F* 199/201 161/162 258/260 53/133 I E.coli F,,* 98/98 24/36 3/60 Wl S.abony Fy, ccante” 782/795 165/169 88/106 60/161 37/116 All the cultures were recently infected with the F agent in question, They were grown from a small inoculum of approximately 10* cells overnight in 1 ml of nutrient broth pH 7.6 with varying concentrations of acridine orange, streaked out on EMSLac plates and replica plated on F indicator 9, LacS"+F-M-LactS" spread on EMLacSm, (for details see Materials and Methods). Number of colonies giving a -+ reaction on these plates is given as fraction of total number of colonies tested. Blank entries signify not done. INFECTIBILITY BY FACTOR F 1437 cells, which probably are more effective recipients (O@rsKov and OrsKov 1961) but was also observed as an immediate consequence of F transfer. F and F,, both make S, abony able to donate chromosomal markers to acceptor cells. The recombinants appear at a low frequency, about 5 X10-7, observed as 10-20 recombinants in a simple spot test (Table 9). Recombinants have thus been obtained for various auxotrophic markers, sugar fermentation markers and H antigen markers (linked to the histidine or methionine markers). Thus larger segments of the chromosome can be transferred than in phage-mediated trans- duction. Hfr variants showing a more stable attachment of the F particle to the S. abony chromosome can also be obtained. A more comprehensive account of such strains and their application to immunogenetic studies will be forthcoming. DISCUSSION With the use of a wide range of fertile males and more effective methods of detection, the scope of conjugal interaction among enteric bacteria has continu- ously increased. In the present report it has been possible to establish interspecies hybrids by means of a mutant fertility factor F’ in every one of 23 strains of Salmonella, Shigella, Serratia and Klebsiella tested. This experiment has been paralleled or anticipated by several other authors and will doubtless continue to constitute the basis of exciting work on the molecular basis of evolutionary dif- ferentiation. An outstanding example of the materialization of such a hope is the study of the DNA of intergeneric hybrids of S. typhi X Serratia marcescens (Marmur et al, 1961; FaLkow et al. 1961). In these hybrids, the DNA pycno- gram shows a new band whose density suggests that it can be attributed to the F fragment ultimately derived from E. coli, whose DNA has a different base com- position and characteristic density than that of S. marcescens. In this particular hybrid combination the exogenotic material associated with the F particle appears not to have successfully integrated with the acceptor chromosome. A failure of integration is also indicated in ZrnpER’s (1960a) studies on phage-mediated transduction from E. coli x Salmonella hybrids, the genes of E. coli origin being poorly transduced to Salmonella by phage grown on the hybrid. It is often per- TABLE 9 Fertility of Salmonella abony F+ Number of colonies growing in drops with 7 Recipient Marker F-SsSW 803 strain scored Medimmn (=control) F+SsSW 1351 F+SsSw 1463 SW 1361 M DOSm 0 10 15 1355 P DOSm 4 15 25 1464 H DOSm 5 20 20 1464 Gal EMGalSm + Histidine 0 10 8 1447 Mal EMMalSm 0 15 15 Ara EMAraSm 0 12 15 All crosses were done by dropping approximately 3% 107 of both recipient and donor bacteria from an overnight broth culture onto the selective plates. Number of colonies growing within each drop after 48 hours’ incubation is given in the table, as mean values of 3-100 experiments. 1438 P. H. MAKELA, J. LEDERBERG AND E, M. LEDERBERG plexing to determine whether, from a genetical standpoint, a stable association with integration of the genetic material into the chromosome has taken place or whether the cell remains heterogenotic. A good indication of integration would be the transfer of genes of Salmonella origin with equal efficiency as those of E. coli origin. OnsKov, Orskov and KaurMaAnn (1961), concur in reporting the fertility of a wide range of Salmonella serotypes with a culture designated as W3287, K-12, Hfr. This strain is closely related to strain W3747 used in the present investiga- tion and like it, carries the F,,;Zact fragment (Table 1). SUMMARY Twenty-three strains of Salmonella, Shigella, Serratia and Klebsiella have been tested for infectibility by the sex-fertility factor, F, from Escherichia coli K-12, Large differences were observed in the ability of the various strains to be infected with F, due partly to differences in their ability to support the growth of F, partly and perhaps mainly to differences in their mating ability. Apart from the requirement for F-determined maleness of one partner specific com- patibilities were observed in several cases, homologous strains showing the high- est degree of F transfer and fertility. In addition, the F factors varied in their capacity to infect Salmonella strains and all 23 strains could be infected with a mutant F factor designated F,; stanie- The F factor introduced from EZ. coli confers on the infected cells very much the same properties of sexual compatibility as it does in #. coli K-12. In this way it is possible to obtain a complete sexual recombination system in Salmonella abony and other serotypes. ACKNOWLEDGMENTS These studies were conducted in 1959-1960 in the course of an exchange visit by Dr. P. H. Maxexd under a U.S. government grant program, public law 265 /584, LITERATURE CITED Baron, L. S., W. F. Carry, and W. M. Spirman, 1959a Genetic recombination between Escherichia coli and Salmonella typhimurium. Proc. Natl. Acad. Sci. U.S. 45: 976-984. 1959b Characteristics of a high frequency of recombination (Hfr) strain of Salmonella typhosa compatible with Salmonella, Shigella and Escherichia species. Proc. Natl. Acad. Sci. 45: 1752-1757. Baron, L. S., W. M. Spruman, and W. F. Carey, 1959 Hybridization of Salmonella species by mating with Escherichia coli. Science 130: 566-567. Cavaniy, L. L., J. Leperserc, and E. M. Leperserc, 1953 An infective factor controlling sex compatibility in Bacterium coli. J. Gen. Microbiol. 8: 89-103. Coox, A. K., and J. Leperserc, 1962 Recombination studies of lactose nonfermenting mutants of Escherichia coli K-12. Genetics 47: 1335-1353. Epwarps, P. R., and D. W. Bruner, 1942 Serological identification of Salmonella cultures. Cire. Kentucky Agr. Expt. Sta. 54. Fatxow, S., J. Marmur, W. F. Carney, W. M. Sprtman, and L. S. Baron, 1961 Episomic transfer between Salmonella typhosa and Serratia marcescens. Genetics 46: 703-706. INFECTIBILITY BY FACTOR F 1439 Frévenicg, P., 1948 Actions antibiotiques réciproques chez les Enterobacteriaceae. Rev. Belge Pathol. Med. Exptl. (Suppl. 4) 19. Hrrora, Y., 1959 Mutants of the F factor in Escherichia coli K-12. Records Genet. Soc. Am. 28: 75. (Abstr.) Genetics 44: 515. 1960 The effect of acridine dyes on mating type factors in Escherichia coli. Proc. Natl. Acad. Sci. U.S. 46: 57-64, Irno, T., 1958 Immunogenetics of Salmonella. Thesis. University of Wisconsin. 1961a A stabilizer of antigenic phases in Salmonella abortus-equi. Genetics 46: 1465-1469. 1961b Anomalous homology of flagellar phases in Salmonella. Genetics 46: 1470-1474. Jacos, F., and E. A. Avennerc, 1959 Transfert de caractéres génétiques par incorporation au facteur sexuel d’Escherichia coli. Compt. Rend. 249; 189-191. Jacos, F., and E. L. Wortman, 1961 Sexuality and the Genetics of Bacteria. Academic Press. New York and London. LEDERBERG, J., 1950 Isolation and characterization of biochemical mutants of bacteria. Methods Med. Res. 3: 5-22. 1961 A duplication of the H, (flagellar antigen) locus in Salmonella. Genetics 46: 1475-1481. Leprnserc, J., L, L. Cavauur, and E. M. Leperperre, 1952 Sex compatibility in Escherichia coli. Genetics 37: 720-730. Leperserc, J., and T. uno, 1956 Phase variation in Salmonella. Genetics 41: 743-757. Lunta, S. E., and J. W. Burrous, 1957 Hybridization between E. coli and Shigella. J. Bacteriol. 74: 461-476. Maccacaro, G. A., 1955 Cell surface and fertility in Escherichia coli. Nature 176: 125-126. Marmur, J., R. Rownn, S. Fatxow, L. S. Baron, C. ScurrpKraut, and P. Dory, 1961 The nature of intergeneric episomal infection. Proc. Natl. Acad. Sci. U.S. 47: 972-979. Mivyaxe, T., and M. Demerec, 1959 Salmonella-Escherichia hybrids. Nature 183: 1586. Nossat, G. J. V., and J. Leperserc, 1958 Antibody production by single cells. Nature 181: 1419-1420. Orsxov, I, and F. OrsKov, 1960 An antigen termed f+ occurring in F+ Escherichia coli strains. Acta Pathol. Microbiol. Scand. 48: 37-46. 1961 Fertility of Escherichia coli antigen test strains in crosses with K-12. Acta Pathol. Microbiol. Scand. 51: 280-290. Orsxov, F., 1. OxsKov, and F. Kaurmann, 1961 The fertility of Salmonella strains determined in mating experiments with Escherichia strains. Acta Pathol. Microbiol. Scand. 51: 291-296. Ricurer, A., 1961 Attachment of wild-type F factor to a specific chromosomal region in a variant strain of Escherichia coli K-12: the phenomenon of episomic alternation. Genet. Res, 2: 333-345, Sxaar, P. D., A. Ricurer, and J, Leperserc, 1957 Correlated selection for motility and sex- incompatibility in Escherichia coli K-12. Proc. Natl. Acad. Sci. U.S. 43: 329-333. Sneatu, P. H. A., and J. Leperserc, 1961 Inhibition by periodate of mating in Escherichia coli K-12. Proc. Natl. Acad. Sci. U.S. 47: 86-90. Stocker, B. A. D., N. Zrnver, and J, Leperserc, 1953 Transduction of flagellar characters in Salmonella. J. Gen. Microbiol. 9: 410-433. Waisman, Harry A., and W. H. Sronz, 1958 The presence of Serratia marcescens as the pre- dominating organism in the intestinal tract of the newborn. The occurrence of the “red diaper syndrome.” Pediatrics 21: 8-12. Zinver, N. D., and J. Leperperc, 1952 Genetic exchange in Salmonella. J. Bacteriol. 64: 679-699. Zinper, N., 1960a Hybrids of Escherichia and Salmonella. Science 131: 813-815. 1960b Sexuality and mating in Salmonella. Science 131: 924-996.