Transduction by Bacteriophage Ti * Henry Drexler

Proceedings of the National Academy of Sciences Vol. 66, No. 4, pp. 1083-1088, August 1970

DEPARTMENT OF MICROBIOLOGY, BOWMAN GRAY SCHOOL OF MEDICINE, WAKE FOREST UNIVERSITY, WINSTON-SALEM, NORTH CAROLINA Communicated by Edward L. Tatum, May 25, 1970 Abstract. Amber mutants of the virulent coliphage T1 are able to transduce a wide variety of genetic characteristics from permissive to nonpermissive K strains of Escherichia coli.

The virulent coliphage T1 is not known to be related to any temperate phage. Certain of the characteristics of T1 seem to be incompatible with its potential existence as a temperate phage; for example, the average latent period for T1 is only 13 min' and about 70% of its DNA is derived from the host.2 In order to demonstrate transduction by T1 it is necessray to provide condi- tions in which potential transductants are able to survive; this is accomplished by using amber mutants of T1 to transduce nonpermissive recipients. Experi- ments which show that T1 is a transducing phage are presented and have been designed chiefly to illustrate the following points: (1) Infection by T1 is able to cause heritable changes in recipients; (2) the genotype of the donor host is im- portant in determining what characteristics can be transferred to recipients; (3) the changes in the recipients are not caused by transformation; and (4) there is a similarity between the transducing activity and the plaque-forming ability of T1 with respect to serology, host range, and density. Materials and Methods. Bacterial strains: The abbreviations and symbols of Demerec et al.3 and Taylor and Trotter4 are used to describe all pertinent genotypes. The phenotype of each strain is given below, together with the symbols used to identify them, and are repeated at each point where their inclusion is useful for clarity. All strains used are members of the species Escherichia coli. KB-5 is a streptomycin-resistant (StrR) strain which is a permissive host (Su+) for all the amber mutants of T1 used here. KB-3 is able to ferment lactose (Lac+) and galactose (Gal+) and is independent of any requirement for arginine (Arg+), tryptophan (Trp+), serine (Ser+), threonine (Thr+), and biotin (Bio+). A tryptophan-dependent (Trp-) variant of KB-3 which requires tryptophan to satisfy its growth requirement was used in one experiment. CS1006 is a streptomycin-sensitive (StrS) strain which is Su+. A spontaneous StrR variant of CS100 was isolated in this laboratory. The strain W33507 is a nonpermissive host (Su-) for the amber mutants used here; it is StrS and is unable to ferment either galactose (Gal-) or lactose (Lac-). Mutants of W3350 which are dependent on either arginine (Arg-), tryptophan (Trp-), or serine and threonine (Ser-Thr-) were obtained by exposing W3350 to nitrosoguanidine. W3350/1 is a spontaneous, T1- resistant mutant isolated in this laboratory. Strain R9018 was derived from W3350 (X); it has a deletion from galE through XQ and also from XB through chlA including one or more genes controlling the biosynthesis of biotin; phenotypically R901 is Su-Gal-Bio-. Strains KB-3, CS100, W3350, and R901 were kindly provided by Drs. W. Michalke, J. R. Christensen, E. Six, and A. Campbell respectively. 1083 Downloaded by guest on September 29, 2021 10841084GENETICS: H. DREXLER PROC. N. A. S.

Phage strains: T1 amber mutants (am) are from Michalke.' Efficiencies of plating strains Tlamll and Tlam5amll on W3350 compared to KB-3 were 10-6 and less than 10-8 respectively. T1 plaques were less than 4 mm in diameter on the various selective media. The infected recipient cells were diluted to give less than 40 plaques per plate. Materials: L broth,9 L agar,9 saltless nutrient broth,10 synthetic T medium with glucose and agar (TMGA),"1 and complete eosin methylene blue agar (EMB)12 have been described elsewhere. T1 resuspension medium is a sterile solution of 1.21% tris(hydroxymethyl)aminomethane (Trizma Base, Sigma Chemical Corp.), 0.05% CaCl2, 0.06% MgSO4, and 0.15% gelatin. Streptomycin is an aqueous solution of streptomycin sulfate (Eli Lilly and Co.). Specific amino acids were used to supplement the T medium in 0.002% amounts. Biotin, when needed, was used at concentrations of 5 ,ug/ml. DNase was a solution of a crystalline preparation of Nutritional Biochem. Corp. CsCl solution contained 1.12 g/ml of CsCl dissolved in 1.21% Trizma base. Methods: In experiments, cells were grown to log phase in L broth and harvested at a titer of about 5 X 108/ml. Cells were centrifuged three times and resuspended each time in saltless nutrient broth and finally concentrated to about 1-2 X 109/ml. Cells and phage (multiplicity of infection of about 1) were incubated at 37°C for 7 min, chilled, and a representative sample was centrifuged in order to test the supernatant for un- adsorbed phage. The adsorption mixture was maintained in an ice bath during plating. In experiments where the transfer of the streptomycin-resistant phenotype was being selected, a sample of the adsorption mixture was plated by overlay on a L agar plates; these plates were incubated at 37°C for 3 hr and streptomycin solution was added via an additional overlay and incubation continued.'0 In experiments other than the streptomycin experiments, the adsorption mixture was spread on the surface of either EMB plates or TMGA plates. A volume of uninfected recipient cells equal in number to the number of cells exposed to T1 were plated as controls in all experiments. Anti-Ti serum wasl)repared by injecting wild type T1 which had been grown on Shi- gella dysenteriae into a rabbit; before use the complement was inactivated by incubation at 56°C for 30 min. Before use, all stocks of Ti were washed several times in Ti resuspension medium, re- frigerated overnight, and filtered. All stocks were routinely tested for sterility. With CsCl density gradient centrifugation, 2 ml of stock phage (about 1-3 X 1011 plaque-forming units/ml) were mixed with 3 ml of CsCl solution. The mixture was centrifuged for 18 hr in a model L, Beckman utracentrifuge in a SW50L rotor at 100,000 X g (36,000 rpm). Results: Transduction of various markers: Table 1 presents data from experiments in which Tlam phage grown on the Su+, Gal+, Arg+, Trp+, and StrR strain KB-3 were used to infect variants of W3350; all W3350 variants are Gal- and StrS; Arg- and Trp- isolates of W3350 were also used. In experiment 3 it can be observed that strain CS100 Su + StrR is also able to serve as a donor for Ti-mediated transduction. The results in Table 1 demonstrate the generalized nature of Tl transduction. Since the amber mutants of T1 which were used here (i.e., Tlamll and Tlam- 5aml 1) kill W3350 even though no phage are produced, transductants must arise from cells which obtained a transducing particle (containing donor DNA but little or no phage DNA") but no phage particle. The efficiency of transduction is given as the number of tranductants per milliliter less control values per singly infected cell as calculated using the multiplicity of infection, the number of cells, and the Poisson distribution. When KB-3Su+ is used as the donor, the number of transductants per singly infected cell varies from about 3 X 10- to 1 X 10-. Importance of the previous host: Tlam5amll was grown on CS100SutStrR Downloaded by guest on September 29, 2021 VOL. 66, 1970 GENETICS: H. DREXLER 1085

TABLE 1. Generalized tranduction by Ti; importance of previous host; host range of transducing particles. * Colonies Transductants/ Phenotype Su + Su - with selected singly infected Expt. selected donor recipient phenotype cells 1 Gal + KB-3Gal + W335OGal- 157 5 X 10-7 Control 0 2 Arg+ KB-3Arg + W3350Arg- 211 3 X 10-7 Control 0 3 StrR CS10OStrR W335OStrS 2155 9 X 10-6 CSLOOStrS W335OStrS 35 7 X 10-8 Control 13 4 Trp + KB-3Trp+ W3350Trp- 129 1 X 10-6 KB-3Trp- W3350Trp- 21 7 X 10-8 Control 1 5 StrR KB-3StrR W335OStrS 248 1 X 10-6 Control 0 W335OStrS/1 0 Control 0 * Tlamll was used in Expts. 1, 2, and 5. Tlam5amll was used in Expts. 3 and 4. and CS100Su+StrS respectively. The third experiment in Table 1 shows that phage grown on CS100Su+StrR can transfer streptomycin resistance to W3350- Su-StrS but that phage grown on CS100Su+StrS cannot. Experiment 4 of Table 1 shows that if a tryptophan-dependent auxotroph (KB-3Su-Trp-), which requires tryptophan to satisfy its requirement, is used as the host for Tlam5amll, the phage is either not able to transfer tryptophan independence to a recipient with a similar requirement (W3350Su-Trp-) or it does so at a very low rate compared to phage grown on Trp+ donors. Since KB-3Trp- donors and W3350Trp- recipients do not necessarily represent identical point mutations, a small amount of transduction may occur in this instance. Host range of transducing activity: Experiment 5 in Table 1 gives the results of a representative experiment in which attempts were made to transduce the Tl-resistant, Su-, StrS strain W3350/1 using Tlaml which had been grown on KB-3Su+StrR. It can be seen that while W335OSu-StrS could be transduced for StrR by the phage, no W3350/1 transductants were observed. Effect of specific anti-Ti serum: Antiserum was used in concentrations which, on the one hand, would inactivate 70-90% of the phage and which, on the other hand, would inactivate less than 10% of the phage after a 1:10 dilution of the antiserum. The first experiment in Table 2 gives the results of a representative TABLE 2. The effect of specific antiserum and DNase on the ability of Tlamll to transduce.* Special Colonies Transductants/ Phenotype treatment with selected singly infected Expt. selected of phage phenotype cells 1 StrR None 269 8 X 10-7 Exposed to antiserum 1 <10-8 Control 1 2 Trp + None 34 3 X 10-7 Exposed to DNase 48 3 X 10-7 Control 4 * KB-3Su +Trp+StrR was used as donor in both experiments. As recipients W3350Su -StrS was used in Expt. 1 and W335OSu-Trp- in Expt. 2. Downloaded by guest on September 29, 2021 1086 GENETICS: H. DREXLER PROC. N. A. S.

experiment. The antiserum was able to reduce the plaque-forming ability of T1 from 4 X 1010 to 2 X 109 units/ml with a concomitant reduction in efficiency from 8 X 10-7 to 10-8. Effect of DNase on transducing activity: It is generally accepted that the transforming principle is sensitive to DNase14 while transducing particles are not. 15 Treatment with DNase affected neither the plaque-forming ability of Tlamll nor the efficiency of transduction. The second experiment in Table 2 gives the results of such an experiment. Effect of ultraviolet (UV) irradiation on transduction: Transforming principle is sensitive to UV.16 The plaque-forming ability of bacteriophage is sensitive to UV but the transducing activity is typically more resistant. 17 It can be seen in Figure 1 that the ability to transduce is far more resistant to UV than is the ability to form plaques. CsCl density gradient centrifugation: Tlam phage were centrifuged at high speed in CsCl solution. Representative data are plotted in Figure 2 and show

120

k FIG.I 1.-A comparison of the effects of ultraviolet irradiation (UV) on the transduc- 80- \ing and plaque-forming ability of T1. T1- am5amll which had been grown on KB- 3Su+StrR was exposed to UV by a 15 W, Germicidal lamp at a distance of 100 cm c\Forming for 0, 2, 4, and 6 mill respectively. After zt40_\T1 UV the various samples were tested for their 40 ability to produce plaques on KB-3Su+ and loo transduce W335OSu-StrS for StrR. Each 20 point on the StrR curve represents over 200 StrR recipients counted.

0 2 4 6 Exposure60\Plaque-GEto UV Light (minutes) 70_ Ploque- I\ .Ga + Fbrming / 60 -T I I FIG. 2.-Sedimentation of Tl and 50INl1 CsCI. The distribution of plaque- Str forming T1 and various Ti transducing ° 40 activities after centrifugation for 18 hr / l| atl100,000\ \ X g in a Beckman model L 30 ultracentrifuge. The distribution of 3C-/ \|il transducing activities for Gal+, Lac+, and StrR were based on 157, 736, and 20 1422 transductants respectively. --\t, Plaque-forming T1; ---, Gal+ 10 / Lac t transductants; -0-, StrR transductants; -A-, Lac + transductants.

.3 .4 Distance from the Miniscus Total Distance Downloaded by guest on September 29, 2021 VOL. 66, 1970 GENETICS: H. DREXLER 1087

the position of the transducing activities of several markers in the gradient compared to the position of the plaque-forming ability. It may be deduced from the data that particles possessing transducing activity have densities which are similar to but not necessarily identical to the plaque- forming particles of T1. Cotransduction: The first experiment in Table 3 presents data in which

TABLE 3. Cotransduction by Tlam5amll.* Colonies Colonies Transductants/ with donor Phenotype with selected simply infected phenotype Expt. selected phenotype cells (%) 1 Gal+ 81 .5X10-7 100 Gal+ Control 0 Bio+ 57 4 X 10-7 100 Bio+ Control 0 2 Thr+ 45 5 X 10-8 49 Thr+ Control 5 Thr+Ser+ 9 1 X 10-8 100 Thr+Ser+ Control 0 * KB-3Su +Gal +Bio +Thr +Ser + was used as donor in both experiments. The recipient in Expt. 1 was R901Su -Gal -Bio - and in Expt. 2, W3350Su -Thr -Ser- was used.

R9OiSu-Gal-Bio- was infected with Tlam5amll which had been grown on KB-3Su+Gal+Bio+. Each transductant, whether selected for either Gal+ or Bio +, was found to be transduced for both phenotypes. The second experiment in Table 3 shows the results of infecting a W3350Su-- Ser-Thr- strain with Tlam5amll which had been grown on KB-3Su+Ser+Thr+. In the data presented it can be seen that about half of the cells transduced for Thr+ (i.e., selected on TMGA + serine plates) are also found to be Ser+. In other experiments, for which data are not given, the percentage of transductants selected for Thr+, which are also found to be Ser+, varied between 10 and 60%. A small but significant number of colonies which are both Ser+ and Thr+ are invariably observed on experimental TMGA plates, but have never been observed on control TMGA plates. Significant numbers of Ser+Thr- colonies have not been observed. This selection-dependent variation in transduction is not understood but is not a unique observation with transduction. 18 Discussion. The following characteristics of the reaction leading to transfer of genes from one host to another show that transduction is involved: absence of viable bacteria in the donor stocks, dependence of the genes transferred on the genotype of the donor (Table 1), and resistance of the transfer to both DNase (Table 2) and UV irradiation (Fig. 1). The wide variety of genes which can be transferred (Tables 1, 2, and 3 and Fig. 2) demonstrate the generalized nature of the transduction. The sensitivity of the genetic transfer to the action of specific anti-T1 serum, its inability to affect cells which do not adsorb T1, and the similarity of its density to the density of T1 (Fig. 2) all serve to illustrate the close relationship of the transducing ability to T1. Cotransduction of closely linked genes (Table 3) shows that the transducing particles contain small fragments of the bacterial chromosome and is consistent Downloaded by guest on September 29, 2021 1088 GENETICS: H. DREXLER PROC. N. A. S.

with the view that transducing particles produced by generalized transducing phage usually contain substantial amounts of bacterial DNA. 13 Phage such as T1 which are not circulary permuted5 are believed to package their DNA by a self-determined mechanism. ' Generalized transducing particles are believed to be formed by the externally determined mechanism.'3"9 The observation that a nonpermuted phage mediates generalized transduction is unique and raises a question as to how T1 packages its DNA. At present there is no direct evidence which can be used to determine the mechanism by which T1 loads its heads with DNA. If T1 transducing particles contain all or nearly all bacterial DNA as is true of the generalized transducing phage P1,13 then each transducing particle carries about i% of the E. coli chromosome.20'2' The observation that T1, a virulent phage which degrades the bacterial chromo- some2 and rapidly produces a burst of phage,' is able to transduce broadens the horizon of transduction because it serves to illustrate that even this type of phage may have redeeming qualities which give the virus-host cell system selective advantages. Dr. J. R. Christensen graciously made his laboratory in the University of Rochester Medical Center available for some of the work used in preparing this report. I would like to thank Dr. Rolf Benzinger for several valuable suggestions and for his encouragement. Without the competent technical assistance of Miss Laura Winstead and Mrs. Lindsay R. Lambe this work could not have been done. * This work was supported by grant Al 07107 from the National Institute of Allergy and Infectious Diseases, National Institutes of Health. I Delbrijck, M., J. Bacteriol., 50, 131 (1945). 2 Labaw, L. W., J. Bacteriol., 66, 429 (1952). 3 Demerec, D., E. A. Adelberg, A. J. Clark, and P. E. Hartman, Genetics, 54, 61 (1966). 4 Taylor, A. L., and C. D. Trotter, Bacteriol. Rev., 31, 332 (1967). 6 Michalke, W., Mol. Gen. Genet., 99, 12 (1967). 6 Skaar, P. D., and A. Garen, these PROCEEDINGS, 42, 619 (1956). 7 Campbell, A., Virology, 14, 22 (1961). 8 Campbell, A., S. Adhya, and K. Killen, "The Concept of Prophage," in Bacterial Episomes and Plasmids, eds. G. E. W. Wolstenholme and M. O'Connor, CIBA Foundation Symposium (Boston: Little, Brown and Co., 1969), p. 12. 9 Bertani, G., J. Bacteriol., 62, 293 (1951). 10 Drexler, H., Virology, 33, 674 (1967). 1" Melechen, N. E., and P. D. Skaar, Virology, 16, 21 (1962). 12 Lederberg, J., Methods in Medical Research, ed. J. H. Comrie, Jr. (Chicago: Year Book Publishers, 1950), vol. 3, p. 5. 13 Ikeda, H., and J. Tomizawa, J. Mol. Biol., 14, 85 (1965). 14 Avery, 0. T., C. M. Macleod, and M. McCarty, J. Exp. Med., 79, 137 (1944). 15 Zinder, N. D., and J. Lederberg, J. Bacteriol., 64, 679 (1952). 16 Zamenhof, S., H. E. Alexander, and G. Leidy, J. Exp. Med., 98, 373 (1953). 17 Garen, A., and N. E. Zinder, Virology, 1, 347 (1955). 18 Hartman, P. E., "Genetic Studies with Bacteria," Carnegie Inst. Wash. Publ., 612, 35 (1956). 19 Ozeki, H., and H. Ikeda, Annu. Rev. Genet., 2, 245 (1968). 20 Lang, D., H. Bujard, B. Wolfe, and D. Russell, J. Mol. Biol., 23, 163 (1967). 21 Cairns, J., J. Mol. Biol., 6, 208 (1963). Downloaded by guest on September 29, 2021