Reprinted from the Proceedings of the NaTionaL ACADEMY OF SCIENCES Vol. 47, No. 1, pp. 52-55. January, 1961. LINKAGE OF GENETIC UNITS OF BACILLUS SUBTILIS IN DNA TRANSFORM ATION* By Eucene W. Nestert anp Josuua LEDERBERG DEPARTMENT OF GENETICS, STANFORD UNIVERSITY MEDICAL CENTER, PALO ALTO, CALIFORNIA Communicated November 25, 1960 Bacterial transformation, the transfer of genetic markers by DNA, provides a mechanism by which closely linked genetic loci may be ordered in a segmental map. Since a unit of DNA transferred from a donor bacterium to the recipient cell represents only a small fragment of the entire genome,' it is not possible to arrange many randomly distributed loci in any single transfer experiment. By the same token, when a pair of genes shows an appreciable degree of joint trans- formation, it points directly to their genetic contiguity. During a genetic and chemical study of the system of DNA transfer in Bacillus subtil?s described by Spizizen,? it was observed that two markers could be simultaneously incorporated into a doubly auxotrophic recipient. DNA isolated from a wild type organism will transfer two independent characters—one concerned with an enzyme of trypto- phan (more precisely indole glycerol phosphate) formation and the other with an enzyme of histidine biosynthesis. The evidence for their linked transfer is pre- sented below. Materials and Methods.—The origin and genotype of the principal strains employed in this study are given in Table 1. The liquid media employed include: Pen, Penassay Broth currently TABLE 1 List or Srrarns or Bacillus subtilis Strain Genotype Origin 168 ind- Burkholder and Giles? 23 thr— Burkholder and Giles SB19 Reference prototroph 23 — X 168* SB1 ind— hisy— UV treatment‘ of strain 168 SB25 ind~ his.~ UV treatment?! of strain 168 SB48 hisg~ SB19 — X 168 (nitrous acid treatment)* SB60 ind hiss 168 — x SB485 SB32 hiso~ SB19 — x SB25 SB33 ind~ SB19 — X SB25 SB31 Prototroph S8B19 — < SB25 * The symbol 23 — x 168 indicates that DNA from strain 23 was used to transform the recipient strain 168. 35B31, SB32, and 8B33 are transformants selected to insure an isogenic background for the ind and Ais markers. sold as Difco Antibiotic Medium 3; CHT-2, minimal medium? containing 0.02 per cent acid hydrolyzed casein and 20 sg per ml of pL-tryptophan; CHT-10, minimal medium containing 0.1 per cent casein hydrolysate and 200 ug per ml of pi-tryptophan. For platings, we used minimal agar (histidine agar or tryptophan agar) containing 40 yg per ml L-histidine or 20 ug per ml pr- tryptophan, respectively. The recipient cells are grown aerobically’ to the completion of exponential growth in Pen medium. The culture is washed and resuspended, 1:10 in a 15 K 125 mm test tube containing 5 ml of CHT-2 medium. After four hours of further incubation at 37°, the cells are washed and diluted 1:10 into fresh CHT-10 medium precooled to 30°, and 0.9 ml aliquots are dispensed into 15 X 125 mm test tubes. After incubation at 30° for 90 minutes, 0.1 ml of a solution of DNA (containing in most instances 1 yg per ml of DNA) is added to each tube, and incubation is con- tinued for an additional 30 minutes. Deoxyribonuclease (Worthington Biochemical Corp.) 52 VoL. 47, 1961 GENETICS: NESTER AND LEDERBERG 53 20 ug, and magnesium sulfate to give a concentration of 0.01 Mf are added. After an additional 10 minutes incubation, 0.1 ml aliquots of the cells are plated on the indicated minimal agar. These plates are incubated at 37°, and read at 40 to 48 hours. This procedure, a modification of the technique devised by Spizizen,® has given consistent results in our hands. For the preparation of DNA, the donor cells are grown in Pen and washed twice in 0.14 4f NaCl, and the packed cells are resuspended at a concentration of 1 gm (wet weight) per 4 ml of a solu- tion containing 0.14 M sodium chloride, 0.01 47 sodium citrate, and 0.01 M phosphate buffer solu- tion, pH 6.6. Lysozyme (Armour) is added to give a concentration of 1 mg per ml and the sus- pension stirred at 37° for 10 minutes. The now viscous lysate is suspended in 10 volumes of a solution of 0.14 M sodium chloride and 0.01 M sodium citrate at 0°. The nucleic acid-protein complex is then sedimented at 0° at 30,000 x g for 20 minutes. The pellet is suspended in 4 volumes of 2 M NaCl relative to the original volume of the lysate and extracted by agitation with a magnetic stirrer for 80 minutes at 4°. The suspension is then centrifuged at 20,000 x g for 20 minutes at 0° and the pellet reextracted with 2.@ NaCl. The pooled extracts are added to 2 volumes of 95 per cent ethanol previously cooled to —20°. The fibrous precipitate of DNA is collected on a hooked rod and resuspended in 2 M NaC}. After storage overnight at 4°, the DNA is deproteinized by gentle manual shaking with an equal volume of a chloroform-octanol (5:1) mixture for 10 minutes, followed by centrifugation at 2,500 x g for 5 minutes. The upper, aqueous layer is transferred to fresh chloroform-octanol and the process repeated (usually seven times) until no gelatinous interface between the two layers appears. The DNA is reprecipitated in 2 volumes of 95 per cent ethanol and dissolved in 2 M NaC) + 0.01 M phosphate buffer, pH 7.4. In this medium, the DNA has been stored for at least 6 months at 4° and has retained full transforming activity. However, repeated pipetting of the DNA solution often results in a loss in activity, perhaps due to shear forces.® The procedure as described here is a modification of the procedures employed by Spizizen? and Ephrati-Elizur.“ When prepared by this method, 0.02 to 0.05 ug of DNA will saturate 7 X 107 recipient cells in one ml. Experimental Results—Nonidentity of his mutants: The relationship of the single histidine markers to each other was studied by determining whether the DNA extracted from one Ais~ mutant could transform another his— mutant to prototrophy (Table 2). If prototrophs do result, the mutant loci are recombina- TABLE 2 DNA TransFrers INVOLVING Ais Loci Recipient Bacteria SB25 Donor DNA from SB1 SB48 SBI (hisi7) 0 60 86 SB25 (hise7) 21 0 15 SB48 (hiss) 33 3l 0 SB19 (hist) 226 98 190 The numbers are Ais* transformants per ul aliquot of recipient cell suspension (5 & 107 per ml). They were scored on tryptophan agar. The quantitative significance of the numbers is uncertain on account of possible variations in the competence of the recipient cultures and the absolute activity of the donor DNA. tionally distinct, i.e., they do not mark the same point of the chromosome. The inference from Table 2 is that these three his~ strains carry nonallelic mutations. Linkage: Although a comparison of the absolute frequencies of transformation to prototrophy could serve to estimate the relative distances of the mutant loci, as in Table 2, there are two important reservations: (1) the need to compare trials involving different, perhaps not comparable, batches of cells and of DNA and (2) the untested assumption that many transformants are produced with the donor markers which cannot be counted in the scoring system. A more reliable method depends on the actual recovery and estimation of single and multiple marker trans- fers. The linkage relationship can be expressed as the cotransfer index, r, a measure 54 GENETICS: NESTER AND LEDERBERG Proc. N. A. 8. of the frequency of joint transfer of two markers compared to the total number of recombinant genotypes measured by the transformation experiment. To use a general notation,"' in a system a'b\—X a°b°, giving transformant types ab}, a'b°, and a°b!, a'b} a'b} "* abt + alb? + ab! al + bE — ad! (since at = a1b° + a'b!). For example, in Table 3, ind+ hist ind+ + hist — ind+ hist’ In other experiments, it may be possible to estimate a/b! and a'b° but not @°b!. In this case, we may assume a°b'!2¢a'b°? and write an estimated index a'p} "o* abt + 2 alb™ For unlinked markers, r can be estimated as a1b°/2N, ie., one-half the efficiency of transformation for either marker, N being the total number of (transformable) cells. The ability of DNA extracted from wild type cells to accomplish the joint trans- formation of the three strains of ind—his— cells to prototrophy is shown in Table 3. TABLE 3 TRANSFORMATION OF ind~ his AUXOTROPHS TO PRoTOTROPHY Transformant Classes, per 104 recipient cells ind + hist i i in ind+ his Caleulated asrandom Cotransfer Donor DNA Recipient Cells Found coincidence Index r SB31 ind+ hist — X SB25ind-his.— 70 72 48 0.5 0.51 SB31 ind+ hist -—— xX SB60 ind his;s- 3.9 5.0 0.03 0.02 0.003 SB31 indt hist — xX SBI ind-his.- 50 100 0.8 0.5 0.005 8B32 indt his —_— SB25 ind- his. 46 20 0.2 0.09 0.003 SB33 ind~ hist The transformation procedure was as given in the protocols except that DNA was used at a concentration of 20 wg per ml for his: and Ais: and at a concentration of 0.1 ug per ml for hiss. The transformant classes were counted on minimal agar (for ind* his*), histidine agar (for ind*) and tryptophan agar (for hist), respectively. The calculated random coincidence is the product of ind+ & hist. The cotransfer index for the appearance of double transformants indthist+ varies from 0.004 to 0.51 with different his mutants. This result is taken as evidence of linkage between the ind and his. loci. Introducing the ind+ and his+ markers on different DNA molecules by mixing DNA preparations from indthis.- gave no evidence of linkage. This argues against the possibility that the linkage can be accounted for by the random coincidence of unlinked markers. Additional auxotrophic markers have been introduced into strain 168 after UV irradiation or nitrous acid treatment of DNA in vitro or by selection of spontaneously occurring mutants as previously described. These markers included growth re- quirements for methionine, methionine plus lysine, glutamic acid or proline, valine plus isoleucine, cystine, and a streptomycin resistance marker. When limiting concentrations of SB19 DNA were used, the cotransfer index never exceeded 0.005. Thus, while ind is closely linked to his:, it is not linked to any of six other markers. Vou. 47, 1961 GENETICS: NESTER AND LEDERBERG 55 Discussion.—The foregoing evidence indicates that the loci his; and ind stand in @ special relationship to each other in contrast with several other pairs of loci studied in the same way. The cotransfer index for these markers might be accounted for if the transform- able cells were a very small fraction of the total. The estimation of competence by the addition of increasing amounts of DNA is not a satisfactory measure owing to the possible interference of one absorption with another. The main argument against this interpretation of the high cotransfer index for ind/his, is its unique value for these markers compared to other sets and to the introduction of ind* and hist from separate DNA preparations. The evidence is closely analogous to that for the linkage of the Fla and H, loci in Salmonella transduction! and for mannitol dehydrogenase and streptomycin resistance in Pneumococcus transforma- tion.! The present findings, in conjunction with those of Ephrati-Elizur et al.,'° are the first evidence of linkage in B. subtélis. This type of analysis, hopefully, may open the way to more intensive studies such as have been done on the genetic chemistry of DNA in Pneumococcus on the one hand and on the sequential ar- rangement of genes and their physiological function in the enteric bacteria on the other. In fact, the histidine and tryptophan pathways have been studied especially extensively in Salmonella, where a striking correlation between linkage and bio- synthetic sequence has been observed.“ At first sight, such a correlation could not be inferred from the present data, but this question requires more complete biochemical and genetic study. * This work was supported by training grant 2G295 and research grant C4496 from the National Institutes of Health and by a grant from the National Science Foundation. + The work was done while the author was on a fellowship awarded by the American Cancer Society. 1 Hotchkiss, R. D., J. Cellular Comp. Physiol., 45 (Supplement 2), 1 (1955). 2 Spizizen, J., these PRocrEpiNGs, 44, 1072 (1958). 3 Burkholder, P. R., and N. H. Giles, Jr., Am. J. of Botany, 34, 345 (1947). 4 Lederberg, J., in Methods in Medical Research, ed. R. W. Gerard (Chicago: The Year Book Publishers, Inc., 1950). Vol. 3, p. 5. Lederberg, J., and E. Lederberg, J. Bact., 63, 399 (1952). Strain 168 cells were irradiated with ultraviolet light on a Difco nutrient agar plate to a survivor- ship of 10-* and the colonies screened by replica plating onto minimal medium supplemented with tryptophan. 5 Schuster, H., and G. Schramm, 7. Naturforsch., 13 B, 697 (1958). The wild type DNA was treated in vitro with nitrous acid, 1 M for 30 minutes at 22°, and used to transform strain 168 to indole independence.