Distribution of Bacteriophage +3T Homologous Deoxyribonucleic Acid Sequences in BaciZZus subtihs 168, Related Bacteriophages, and Other BaciZZus Species IiVONA T. STROYNOWSKIt Department of Genetics, Stanford Unkersity School of Medicine, Stanford, California 94305 Received 10 October 1980/Accepted 12 June 1981 The Bacillus subtilis 166 chromosome was found to share extensive homology with the genome of bacteriophage +3T. At least three different regions of the bacterial genome hvdridized to ribonucleic acid complementary to +3T deoxyri- bonucleic acid (DN.1). The thymidylate synthetase gene, thyA, of B. subtilis and the sequences adjacent to it were shown to be homologous to the region in the +3T DSA containing the phage-encoded thymidylate synthetase gene, thyP3. SP,f?, a temperate bacteriophage known to be integrated into the B. subtilis 168 chromosome, was demonstrated to be closelv related to 03T. Other regions of the bacterial genome were also found to hybridize to the +3T probe. The nature and location of these sequences in the bacterial and phage chromosomes were not identified. It was shown, however, that they were not homologous to either the th~P3 gene or the DXA surrounding the thyP3 gene. The chromosomes of other Bncillus species were also screened for the presence of +3T homologous se- quences, and the th~P3 gene was localized in the linear genomes of phages 43T and pll by heteroduplex mapping. It is suggested that the presence of sequences of phage origin in the B. subtilis 168 chromosome might contribute to the restructuring and evolution of the viral and bacterial DNAs. In general, Bacillus species contain a substan- rial number of temperate phages, and many of the straim are known to be polylysogens. They also harbor defective phages and genes which e.re esprejsed when the ceh are subjected to mitomycin C or UV light (1, 11, 20, 25, 33). In most cases, the biological function of the inducible genes is not know-n, nor is it known v;hJ the>- haFe b:en retaired in the bacterial chromosomes during evolution. The defective phages and cryptic genes have recently been the subject of intensive study. Some of the functions expressed in the induced cells include modifica- tion methylases and DXA repair enzymes (13, 36). In certain instances, the bacterial lysogens are known to benefit from functional comple- mentation by phage genes. In particular, an in- teresting case is the host-temperate phage rela- tionship obsemed between Bacillus subtilis and tivo of its temperate phages. &T and pll (5,35). These phages c~_structursl genes for thymi- dylare synthetases, which can complement a thymine deficiency in their host. These genes are designated thyP3 and th>Pll, respectively (12). The phages integrate into the I?. subtilis 91 chromosome at sites different from the two loci, thyA and thyB, encoding bacterial thymidylate synthetases (39). The thyP genes are expressed during lysogeny (35). They can also be intro- duced into B. subtilis by DNA-mediated trans- formation, which does not require the successful integration of the entire prophage (39). Thus, lysogenic infection and DNA-mediated transfor- mation appear to utilize different pathways for uptake of genes into the cells. As a preliminary step in the exploration of the mechanism by which the thyP3 gene transforms bacterial aux- otrophs, the B. subtilis 168 chromosome was analyzed for the presence of DNA sequences homologous to the +3T genome. The relation- ship between the thyP3 gene of (p3T and the thyPll gene of pll was also examined. To gen- eralize my observations, the genomes of several other Bacillus species were screened for the presence of sequences homologous to the thyP3 gene. MATERIALS AND METHODS Bacterial strains. Plasmids pFT23, -33, -401, and -502 (7) and pFT thyP3 were propagated in Bsche- richia coli strain W5543 (hsdR hsdM leu thi thy rpsL trp tonB2) or 1V5545 (hsdR hsdM' thr Ieu thi sup&H rpsL Zac tonA pro). B. subtilis strains SB168 (trpC2), 92 STROYNOWSKI J. BACTERIO~. SB1200 (iltd8 thyB1 citB1 gapA2), and SB591 (IhyA rhyB) and BaciUus strains SB510, SB511, SB512, SB513, SB514, SB515, SB519, SB522, SB727, SBi30, SB734, SD1096, SB1098, SB1099, SBllOO, and SBlllO were from the Stanford University collection. Strain SB1207 (leu met thr SPp-) was obtained from S. A. Zahler, and strain SB1213 (trpC2) is a SB168 deriva- tive which was cured spontaneously of SPB DNA sequences but was still resistant to infection by SP/3. Phage stocks. +3T DNA was prepared after mi- tomycin C (Sigma Chemical Co.) induction of B. sub- G/is as outlined below. One colony of an SB168 deriv- ative strain, cured of SP/J' and lysogenized with phage &T, was inoculated into 50 ml of L broth supple- mented with lo-' M M&la. The overnight-grown culture was diluted 10 times into the same medium and incubated with shaking at 37oC until early log phase (40 Klett units). At this stage, 0.5 pg of mito- mycin C per ml was added, and the cells were incu- bated at 37oC until lysis occurred (2 to 3 h). The culture was briefly centrifuged (20 min at 8,000 rpm) to remove bacterial debris. The lysate was filtered, incubated with DNase (5 pg/ml) for 20 min at 37"C, and heated for 5 min at 60oC to inactivate the enzyme. The phage were pelleted by overnight centrifugation at 8,000 rpm in a Sorvall centrifuge. The pellet was suspended in 10 mM Tris-1 mM EDTA, pH 8.5. The phage-containing solution was adjusted to a density of 1.50 (refractive index = 1.3810) with CsCl and centri- fuged for 40 h at 48,ooO rpm. Further purification of the phage was obtained by centrifugation in a step CsCl gradient for 12 h at 37,OOO rpm. The pooled fractions from several gradients were dialyzed against 3 liters of 10 mM Tris-I mM EDTA, pH 8.5, DNA extracted with 1% sodium lauryl sulfate for 30 min at room temperature, and then extracted three times with phenol. The phage DNA was extensively dialyzed against IO mM Tris-1 mM EDTA, pH 8.5, and stored in 4oC. The phage preparation was also examined by electron microscopy to verify that it did not contain phage particles with a morphology different from that of +3T. T4 phage was prepared as described by Kim and Davidson (15). Phage pll DNA was a generous gift from J. C. Orrego. Bacterial DNA and plasmid DNA preparation. Bacterial DNAs were prepared by a modification of the method of Klotz and Zimmer (16) as described by R. M. Harris-Warrick (Ph.D. thesis, Stanford Univer- sity, Stanford, Calif., 1976). The pronase treatment was omitted. Closed circular plasmid DNAs were iso- lated from cleared lysates, prepared according to Cle- well and Hetinski (3). DNA was centrifuged in ethid- ium bromide-c&urn chloride gradients (23). Ethidium bromide was removed by chromatography on Dowex A6 5OW-XB. The DNA was separated from oligori- bonucleotides by gel filtration on Bio-Gel A-l!jM. The plasmid DNA was extracted three times with phenol, ethanol precipitated, and dialyzed against IO mM Tris-1 mM EDTA, pH 8.5. Transformation bioassay in B. subtilis. B. sub- Mis cells were brought to competence as described by Harris-Warrick (Ph.D. thesis, 1976). The competent cells were concentrated IO-fold by centrifugation, sus- pended in 5% glycerol and frozen in liquid nitmet, Thawed competent cells were diluted IO-fold in c&r. ma! medium supplemented with 20 mM MgCl,. ar,+ DNA was added in various final concentrations (0.1 ti, 1 pg/ml). The mixture was incubated for 60 min i1 37oC with gentle shaking and then plated on se!cc:i~e media (Harris-Warrick, Ph.D. thesis, 1976). Restriction enzyme cleavage of DNA. Res:fic. tion enzyme digestions were according to the speci6. cations recommended by the vendors of the enzhaies. BamHI, SmnI, PatI, B&D, Sir/I, HpaI, and Hind111 enzymes were purchased from New England Biot&. Inc.; EcoRI was purchased from Miles Laboratories, Inc. A two- to threefold enzyme excess was used mu. tinely for most of the experiments. Reactions were terminated by heating the -sampIes at 65oC for 7 mm. Agarose gel electrophoresis. Electrophoresis of DNA was done in 0.7% agarose horizontal slab gels I 1-1 by 20 by 0.3 cm) at 60 V overnight or at 175 V for 3 to 4 h. Paper wicks connected the gel to tanks containing 0.089 M Tris-O.002 M EDT&O.089 M borate, pH 8.5. The gels contained 1 pg of ethidium bromide per ml. DSA was visualized by U\' illumination of the gels. and the fluorescent bands were photographed by using a yellow filter (Kodak no. 9 Wratten filter) and Pola- roid film (type 55P/S). 1Ieasurements of the molec- ular weights of the DNA molecules were made relative to o3T DNA cleaved by EcoRI (7) and SPPl DS.4 cleaved by EcoRI (10). Nucleic acid hybridization. Hybridization exper- iments were carried out ssentially as described b\ Southern (29). The h>-bridizations were done at 3iiC in a solution containing 50% formamide, 0.75 Al Sac!. and 0.075 M sodium citrare. "P-labeled complemen- tary RSA (cRNA) was us.& as the probe. It ws synthesized on plasmid and chromosomal DNA tem- plates in a reaction mixture containing 50 to 100 ,ul of [u-~`P]GTP (1 mCi/mll, 0.5 mM each ATP, CTP, and UTP, 2 pg of DNA, 40 mM Tris (pH 8),10 mM MgCb. 10 mM B-mercaptoethanol, and E. coli core RN.4 polmerase in a total volume of LOO pl. RNA polm- erase was a gift from D. Brurlag. The reaction u-as carried out at 37oC for 3 h. DBase was then added to 20 pg/ml together with carrier RNA. The reaction was terminated by phenol extraction. [3'P]cRNA was sep- arated from unincorporated nucleoside triphosphates by Sephadex G-56 chromatography. A total of 5 x IO" to 5 x 10" cpm of J'P-labeled cRNA probe was used per 2 ml of hybridization solution containing a nitro- cellulose filter (8 by 2 cm). After hybridization, the washed Nters were blotted dry and autoradiographed at -90oC. using Du Pont Crone-x 4 or Kodak XR-5 X- ray film and Du Pont Crones Lightning Plus intensi- fying screens. Electron microscopy. Preparation and spreading of heteroduplex molecules WCS done by the method of Davis et aI. (4). except that the molecules were spread on water. No DNA extraction on +3T or pll phage was performed before using the phage in heteroduples experiments. The photographs were taken at a mag- nification of ~4,500. The lengthi of molecules were determined by measuring projections of 35-mm nega- tives on the surface of a Hewlett-Packard 9864 A digitizer interfaced with a Hewlett-Packard 9810A cal- 3 -8' ) -_ 146. 1961 PHAGE +3T Dh'A SEQUENCES IN B. SC'BTILIS 93 ,J~:J~. The projections were meaxued at a tota! ,,,;7lficarion of X135,ooO. All me%urements were ,-& relative to a standard II?;.\ molecule, pSC101, r,Ju,lted on the same grid. The molecular weight of :...+ I >lgCl:, 10 rnY T-A> 1pH 6.5). and 10 I.i of EcoRI enzyme (Lliles j'~ratories) at 37'C in a tozzl volume of 50 4. +nples of 10 4 each were removed after 1, 3, 7.5, and ;; h. and the reactions were temkared by heating the L&es to 6joC for 7 min. A O.>;y amount of EcoRI'- .$:.ed pFT4OI from each sample was rnised with 0.2 c EcoRI-cleaved pMB9 in a w-4 volume of i0 ~1. 1 01 kza:ion ms carried out at 10 pg of DS.4 per d for I 5 i! ~I)`C, using Ta ligase and th.e buffer described by .;zuamella et al. (26). E. coli iv.%-%3 competent cella Tire prepared as described elsewhere (X1;). ne rransformants (Thy' Tcrl Kere --leeted on aa - .. &?!PS ~qxzIi!m minimal medium [SO] supplemented %j:h 2.5 pg each of the common amino acids per ml, +-o-q, 30 yg of tetracycline per ml, and a&. RESULTS Construction of chimeric plasmids con- taining 03T DNA sequences. Most of the recombinant plasmids (pl?I?3, pFT-24, pFT33, pF'T401, and pFT.502) used in this work were constructed previously in our laboratory. Figure i3 summarizes their molecular structures. The Thimeras contain different but overlapping +3T DS.4 sequences joined to E. coli plasmid pSC101 or pLIB9. A DNA segment of +3T origin s common to all of the recombinant plasmids and complements the thyA thy-mine deficiency mnuration in E. coli. It it ak~ able to transform Thr.- ausotrophs of B. subt2i.s to prototrophy. Therefore, this segment of c;3T DSX is assumed 10 contain the thyP3 gene. To achieve higher purification of this specific gene, it was necessary i0 diminate nonessential DS;X sequences flank- ing the thyP3 gene. This was accomplished by subcloning ofEcoR1' fragments (22) of pFT401 in the EcoRI site of plasmid pBR322. Figure lb shows the structure of the resulting chimer% p!=mid, plYI' thyP3. The molecular size of the EroRI DSA insert carried by pFT thyP3 is 0.5i 51dal. Sequences homologous to 03T DNA in the B. subti~is chromosome. StructuraI anal- !.*s of DSA sequence relationships were per- formed by using the technique of cRKX-DNA hybridization coupled with restriction enzyme and!.sis (29). At least three different regions of ihe E. subtih chromosome capable of hybrid- !% o3T cRNX were identiried. Dx.\ homology of the bacterial thy-4 and ?hage th~P3 genes was demonstrated by hybrid- IzXion of cRNX prepared on pFT Ih3,P3 to the b) L- ** P-3 - FIG. 1, Molecular structure of chimeric plasmids. All DNAs are aligned from the BarnHI site located in the vectors. The vector components of the hybrids are represented by heavy lines: E, pSClO1; @ RSF2122; Kl, pBR322. The inserts are shoun as thin lines. There is, in all cases, an EcoRI site at the boundary between the vectors and inserts. Additional EcoRI sites are denoted as arrows. The direction of the arrows indicates the orientation of the inserts. (Q) Diagrams of plasmids described earlier (7) but used in this work. The sizes (in megadaitons) of the seg- ments are (A) 4.5, (B) 0.67, (C) 0.24, (0) 0.9, (E) 0.9, (A') 2.1. A' is homologous to left part of segment A. (b) Diagram of the chimeric plasmid constructed by cloning the EcoRI* fragment of pFT401 in the pBR322 vector. The size of pFT thyP3 insert is 0.57 Mdal. EcoRI-cleaved fragments of B. subtilis DNA which are associated with th.yA biological activ- ity. The data showing partial purification of B. subtilis segments containing thyA and thyB genes are presented in the accompanying paper, where it is shown that wild-type B. subtilis contains two EcoRI fragments associated with Thy' activity (32). The thyA' gene resides on an EcoRI fragment larger than 10 Mdal, whereas the thyB' gene is carried on a 4.2-Mda.l EcoRI fragment. RNA complemenkry to pFT thyP3 hybridized to only one band of the B. subtilis EcoRI restriction digest (Fig. 2). The molecular weight of the DNA in this band corresponded to the segment containing thyA biological activity. The size of the DNA insert in pFT thyP3 was only 0.57 Mdal (-950 base pairs), and therefore it contained very little, if any, of the DNA se- quences other than the UzyP3 gene. Thus, it is assumed that the observed hybridization occurs between the phage and bacterial thy genes. Con- trol experiments involving similar hybridizations between cRNA transcribed from the plasmid vector pBR322 gave negative results. To construct a physical map of the thyA re- gion, B. subtilis DNA was cleaved with different restriction enzymes and then hybridized to Pas- labeled cRNA prepared from several of the re- combinant plasmids. Only the Hind111 enzyme 94 STROYXOWSKI cleaved this region internally (Fig. 3). The dif- ference in the hybridization patterns observed in Fig. 3a and 3b is due to the different lengths of the probes used in these two experiments. It follows that the B. subfilis chromosome contains one or more other genes which have homology to +3T sequences. This gene (or genes) was linked to thyp3 in the phage and to thyA in B. subtilis, as evidenced by the unsuccessful at- - >10 FIG. 2. Hybridization of cRNA pFT thyP3 to EcoRI-digested B. subtilis 168 DNA. Two micro- grams of SB168 DNA u,as digested uith EcoRI en- zyme, electrophoresed through a 0.7% agarose gel, transferred to a nitrocellulose filter (29), and hybrid- ized to `"P-labeled RL\`A complementary to plasmid pFT thyP3. The conditions of hybridization are de- scribed in Materials and Methods. Sizes of the stained DNA bands are expressed in megadaltons. J. Bacx~,,; tempts to separate them by digestion wirh ttt following enzymes: EcoRI, BumKI, SmaT, Prti, &$I, S&I, and HpaI (dara nor &OFIT~). n.., size of the restriction fragment thar con:ainbj the thyA region was 7.8 >ldal for PstL 5.4 116~ for BgZII, 6.9 M&J for HpaI, and -more rhzn 1:: Mdal for EcoRI, BamHI, San, and SmaI. DO;. ble digestion of B. subtilis 165 D?iX Cth Eco& and Hind111 enzymes, follou-ed by hybridizatic,; with cRNA pFT.502, was compatible uith the physical map of the thyP3 homologous regics shown in Fig. 4. In addition to the segment ass0cia:e-d with the a) b) A 6 A B FIG. 3. Hybridization of la) cRS-4 pFT tk~P.3 and (bj cRN.4 pFT502 to HindHI-cleaved D.Y--lO i'rOducts of a cr"yptic prophage, SP/3, which is knolvn to be integrated into the B. subtilis chro- - nosome between &A and kcrrtil (40). Figure 5 - 5.6 presents an example of the experiment support- :ng this conclusion. The EcoRI band of 1.25 &WV hi&l, which hybridized to cRNA pFT502 in sBl6.8 (Fig. 5A), was missing in the EcoRI digest "f srrain SB1207 cured of SPJ3 (Fig. 5B) and reappeared after this strain was lysogenized with z; 5Pp (Fig. 32. .Qproximately 20 bands appeared in the au- torndiogram shown in Fig. 6 when the RNA Ienrplementary to the whole 03T genome was used as a probe in hybridization with the restric- tion digest of the SPP lysogen. The bands mere .lO ++ + 5.6 ++ +++ + + + +++ + +++ + + + + 1.3 + 1.25 i- 1.15 + 0.9 + 0.63 + 0.53 + 52.06 1.45 + 1.3 ++ " If strong (++i) bands are counted as doublets, the total size of the hybridizing fragments is 70.4 Mdal. In addition to the bands listed above, EcoRI fragments of 1.45 and >lO Mdal appeared on the autoradiogram upon longer exposure. +, weak. *The total size of the bands other than the SPP- specific or thyA region DNA is 8.35 Mdal. ++, Strong; +, weak (>I0 Mdal) corresponded to the thyA region. The nature and location of the other three seg- ments of 5.6, 1.45, and 1.3 Mdal were not iden- tified. When the P3*-labeled cRNA was transcribed from a B. subtilis chromosome containing SPj3 prophage (SB581) and hybridized to an EcoRI digest of +3T DNA, not all of the bands were labeled (Table 2). The pattern described in Ta- ble 2 indicates the portions of the +3T genome that were detected by this technique. They cor- responded to the portions of the #3T genome present in the 3. subtilis chromosome. Sequences homologous to +3T in other Bacillus species. Table 3 summarizes the re- sults of screening the chromosomes of several Bacillus species for the presence of thyP3 ho- mologous sequences. In most cases, the presence of the thyP3 homologous gene was accompanied by the presence of the linked 43T homologous sequence (B. subtilis Marburg, B. globigii, B. subtilis niger, and B. cougufans). B. mycoides and B . pumilus DNAs did not hybridize cRS+.\ pFT thyP3 but showed hybridization !virh cRNA pFT502, indicating the presence of & gene or genes linked to thyP3. Only B. subtiIi,q Marburg and 8. coagulans were found to CC,>- tain a SPP-specific EcoRI band which h-h& ized to cRNA pFTjD2 (Fig. 5). Thus, these t;r-0 species are assumed to be SPP lysogens. - >lO - 5.6 :1 FIG. 7. Hybridization of cRNA +3T to EcoRl- cleaved SB1207 (SP/3-). An identical h@idizatiol; pattern was observed by using strain SBI213 D-V.4 Sizes of stained bands are expressed in megadalton.+. TABLE 2. Hybridization of RNA compiementaq to SB591 (SPb+) to EcoRI-cleaved +.3T D.%`A" Band no. Hybridization 1 2 3 4 + 5ab + 6 + 7 + 8 ol- 9 10 + n EcoRI bands of p3T were numbered according to Ehrlich et al. (7). Detection of hybridization to bands smaller than band 10 was below the resolution of the technique used. +, Hybridization present; -, no hy- bridization. I-OL. 148, 1981 PHAGE: +3T DNA SEQUENCES IN B. SUBTXLIS 97 TABLE 3. Screening of bacterial chromosomes for the presence of the thyP3 gene and linked @T homologous sequence Hybridiza- Hybridization of tion of cRNA pFT502 to Transformation efficiency (no. of tran$formants/ cRNA pFT EcoRI band" no. of viable cells per pg of DSA)" thyP3 to EcoRI band." >I0 hldal >lO Mdal 1.2 hldal Thy His Phe ,;Blx3 (SP/- Q-2) sg3~2 (B. subtilis Marburg) 5B315 (B. megatherium) SB.515 (B. polymyxa) SBjIl (B. subtilis subsp. Cer. i-llidis) SB513 (B. cereus) SBj12 (B. globigii) SBjll (B. mycoides) SBjIO (B. subtilis subsp. ni- ger) SF31110 (B. amy!oIiquefu- ciens) SBllW (B. snbtilis subsp. - 6 x 1O-4 3 x lo-* 3 x lo-' + 5 x lo-' 3 x lo-' 3 x 1o-4 3 x 10-T 5 x 10-1 1.5 x lo-' I.5 x lo-' natto) SBIEEI (B. coagulans) + + + 8 x lO-4 6 x lo-* 6 x 1o-4 SBlO98 (B. brevis) SB1096 (B. pumilus) + 1 x 1o-3 5 x lo-< 5 x lo-' SBZ.4 (B. subtilis N) - 10-' SBX4 (B. subtilis H) SBiS (B. subtilis K) b'.%j (E. coli K-12) NT NT `I'1 phage NT XT " Xn EroRI band 2.5 Mdal from SBi27 (B. subtilis K) hybridized both cRNA pFT thyP3 and cR?!A pFT502 probes. +. Hybridization present; -, no hybridization. `The recipient strain in the transformation as-says was SB748 (thyA thyB his-2 w-o-2). Transformation &ciencv is listed only when the number of transformants in the assays was significantly higher than the number bf revertants for the marker tested. The numbers shown are averages of six sets of experiments. NT, Sot tested. The DSA extracted from the Bacillus species studied was also tested for its ability to trans- form ihe B. subtilis thbmine auxotroph SB591 to prototrophy. In many cases, the high trans- formation efficiency to the Thy+ phenotype was correlated with the presence of the thyP3 ho- mologous sequences in the donor species (B. subtilis niger, B. coagulans, B. subtilis Mar- burg, and SBl213). In the case of DNA from B. globigii, which was found to hybridize the thyP3 probe, the transformation efficiency was below the rel-ersion frequency of the recipient strain. This indicates that the DNA sequences sur- rounding the thy genes in the donor and recipi- ent strains are nonhomologous. An alternative esplanation which proposes the existence of a B. s1:btili.s restriction system that degrades B. glo- b&i DS.1 was ruled out by the results of Harris- l\`arrick (Ph.D. thesis, 1976). A few strains (B. megatherium, B. subtilis subsp. terminalis, B. .~ubtilis subsp. natto, and B. pumilus) did not hybridize to the thyP3 piobe, but their DNA still transformed the B. subtilis 168 thymine auxotroph to prototrophy. This might have been due to transformation with the thyB locus which was contained in the donor strains. Another possibility is that the thyA genes in the donor strains diverged considerably from the B. sub- tilis thyA sequence and therefore were not de- tected under the hybridization conditions used. Neither E. coli nor T4 phage DNA was able to transform B. subtilis strain SB591 to Thy+ or hybridize to thyP3 sequences. These results demonstrate specificity of the Southern hybrid- ization technique and transformation assay used in these studies. Mapping of the thyPl1 gene in pll. The thyP3 gene has been mapped in the +3T genome by heteroduplex analysis of hybrids formed be- tween the chimeric plasmid pFT25 and phage DNA (7). In view of the known sirniIarit,ies be- tween +3T and pll(5), I decided to use the same technique to IocaIize thyPl1 in ~11. The pFT21 plasmid containing thyP3 of the 63T phage was 98 STROYNOWSKI linearized with the BarnHI restriction enzyme. After denaturation, this plasmid DNA was an- nealed to the complementary single strands of the pll DNA (Fig. 8). A double-stranded region of homology with a size of 3.0 +_ 0.2 Mdal was found at a distance of 50 to 54% of the total length from one of the termini of the pll DNA molecule. The standard error of this measure- ment was 3% of the fractional pll length. The double-stranded region found in the heterodu- plex of pll with segment A of plasmid pFT24 was not perfectly homologous. Two small loops could be detected. The right part of segment A (1.5 Mdal) did not hybridize to the pll DNA. This indicates that the DNA sequences sur- rounding the thyP gene in the +3T and pll phages are different. Figure 8 shows also the localization of the thyP3 gene in the +3T ge- nome. The thyP3 gene was situated at a distance J. BACTERIOL. 44 to 48% from one of the +3T ends. Statistical error of this measurement was 2% of the frac. tional03T length. DISCUSSION The results described here provide evidence for the existence of extensive DNA homologies between the chromosomes of B. subtilis 168 and its temperate bacteriophage +3T. Three distinct regions of the bacterial chromosome were iden- tified that were capable of hybridizing RNA complementary to the +3T genome. One of these regions contains the structural gene for thymidylate synthetase A and wss shown to be homologous to the phage-encoded thyP3 gene. In addition, another homologous DNA sequence (equal to or less than 2.15 Mdal) was located next to thyA in B. subtilis and next to thyP3 in +3T. The nature of this other gene i thyPl1 d) I I I I 0 1) 20 40 60 80 MOLECULAR WEIGHT IMDAL) FIG. 8. Localization of the thyP gene in pll and +3T. (a) Histogram of the position ofpoint X (see belou ti shown in relation to the ends of plI DNA. X corresponds to the far left end of the +jlT insert in pFT24 and pFT25. (b) Diagram of pll Dh'A heteroduplexed with BamHI-linearizedpFT24 D.U. Segments B, C, and D represent the regions of homology between the two DNA strands. Loops E and F correspond to regions of nonhomology. The sizes of heteroduplex regions are (in megadultons): (A) 0.2; (B) 1.i3; (C] 0.11; (D) 1.1; (E) 0.13; (F) 0.26; (G) 6.39; (H) 41.1; (I) 38.8. Secenteen molecules with the sfructure represented by this diagram were analyzed. (c) Histogram of the position of point X is shoun in relation to the ends of +3T D.VA. (d) Diagram of +3T DNA heterodupiexed with BarnHI-tinearizedpFT25 DNA. Segment B represents the region of homology between +`T andpFT25. The sizes of heteroduplex regions are (in megadaltons): (A) 0.2; (B) 6.3; (C) 4.7; (D) 34.8; (E) 41.1. Fifteen molecules with the structure represented by this diagram were measured. \`OL. 146, 1951 i". iAGE +3T DNA SEQUENCES IN B. SUBTZLIS 99 (or genes) was not identified. No markers closely Linked to thyA or thyP3 have been reported in H. subtilis of +3T. In other systems, such as T4 phage, DNA sequences close to the structural gene for thymidylate synthetase code for pro- teins involved in DNA metabolism (37). In con- trast, the E. coli or Salmonella typhimurium th?A Iocus maps between the lys and arg loci, awav from other genes involved in DNA synthe- .cis (1;). The second region of extensiv*e homology be- tween the bacterial genome and the +3T genome spas identified as an SPP prophage. SPP is a temperate, cryptic bacteriophage present in B. &tiZis strains derived from Spizizen's trans- formable strain 168 (31). The discovery by War- ner et al. (36) of a cured strain has permitted the characterization of SPP. It is a fairly large phage of complex structure similar morphologically to @T and pll (36) but not to PBSX, a defective bacteriophage also known to lysogenize all B. subtilis 168 strains (25). The prophage attach- ment site for SPB lies between ilcA and kauA (10). SPj? does not convert B. subtilis Thy- aurotrophs to prototrophy upon lysogenization. It can, however, carry out a specialized trans- duction of the citK and hau.4 genes (40). Its mechanism of specialized transduction was pro- posed by Zahler et al. (40) to resemble closely the E. coli phage X dgal system (2). The results described in this communication demonstrate that the phages SB/3 and +3T are closely related. When RNA complementary to the +3T genome was hybridized to the EcoRI restriction digest of a SP/3 lysogen, more than 20 bands (total of 50 to SO Mdal) were homologous to the radio- active RN.4 probe. The restriction fragments which did not cross-hybridize between the two genomes presumably code for the traits that are different in the two phages, such as immunity. In addition, it was shown that SPP lacks the th~P3 sequence (0.57 Mdal), although the DNA sequence located next to this gene in +3T is still present in the SPB genome. Finally, it was shown that the B. subtilis duo- mosome contains yet another region homolo- gous to 03T which is different from ThyA or s1'/3 DSA. Two strains independently cured of SPb still hybridized +3T probe. Three EcoRI fragments of B. subtilis 168 DNA (molecular sizes of 5.6, 1.45, and 1.3 Mdal) showed homology to sequences other than the thyP3 region of +3T. The nature and chromosomal locations of these sequences were not identified. The presence of $3T homologous sequences scattered at different locations in the B. subtilis chromosome might promote site-specific recom- bination and, in consequence, restructuring and evolution of the bacterial and phage chromo- somes. The ability of SPP to recombine with +3T (described in the accompanying paper [32]) supports this hypothesis and suggests that. thy-transducing phages such as 43T and pll could have been created during recombination events between SPP phage and the thyA region of B. subtilis. B. subtilis is not the only organism known to carry multiple sequences of viral origin in its genome. This has been shown to be a common property of many eucaryotic systems. Simian virus 40 (14, 17) and adenovirus (6, 9, 24) se- quences are present in many copies in the ge- nomes of their transformed mammalian hosts. The thymidine kinase gene coded by herpes simplex virus can integrate stably into mouse DNA at many different sites (21). Numerous species of vertebrates also contain endogenous latent RNA tumor viruses present in the cells as proviral DNA copies (8,18,23). The distribution and functions of these proviruses in the host genomes are not known at present., nor is it known whether they can promote site-specific recombination of the eucaryotic chromosomes. In view of the complexity and inherent difficul- ties in studying the structure of the mammalian genome, the arrangement and role of cryptic genes and phages in B. subtilis might be of general interest as a model system for under- standing the virus-host relationship. ACKNOmEDGMEhTS I am grateful to J. Lederberg, in whose laboratory this work was done, for his continuous support and many helpful sug- gestions. I also thank him, M. Winkler, and A. T. Ganesan for critical reading of this manuscript. I am indebted to H. Bursz- tyn-Pettegrew for sharing her unpublished data with me and to P. Evans for excellent technical assistance. This work wa; supported t,y Public Health Service prec'oc- toral fellowship GM 00295 and grant CA 16896 from the National Institutes of Health. LITERATURE CITED 1. Arwert, F., and L. Rutberg. 1974. Restriction and mod- ification in B. subtilis induction of a modifying activity in Bacillus subtilis 168. Mol. Gen. Genet. 133:175-177. 2. Campbell, A. 1962. Episomes. Adv. Genet. ll:lOl-145. 3. Clewell, D., and D. R. Helinski. 1969. Supercoiled cir- cular DNA-protein complex in E. coli: purification and induced conversion to an open circular DNA form. Proc. Natl. Acad. Sci. U.S.A. 62:1159-1162 4. Davis, R.. M. Simon, and N. Davidson. 1971. Electron microscope heteroduplex methods for mapping regions of base sequence homology in nucleic acids. Methods Enzymol. X:413-428 5. Dean, D. H., J. C. Orrego, K. W. Hutchison, and H. 0. Halvorson. 1976. New temperate phage for Bacillus subtih, pll. J. Viral. 20:509-519. 6. Doerfler W., H. Burger, J. Ortin, E. Fanning, D. T. Brown, M. Westphall, U. Winterhoff, B. Weiser, and J. Schick. 1955. Inrergration of adenovirus DNA into the cellular genome. Cold Spring Harbor Slmp. Quant. Biol. 39:129-136. 100 STROYNOWS&I J. BACTF.RICII. 7. Ehrlich, S. D., H. Bursztyn-Pettegrew, I. Stroy- nowski, and J. kderberg. 1976. Expression of the thlmidylate synthetase gene of the Bacillus subtilis bacteriophage +3T in Eschen'chia coli. Proc. Natl. Acad. Sci. U.S.A. 73:4145-4149. 8. Evans, R. M., hi. Shoyab, and M. A. Baluda. 19%. Studies on characterization of the integration sites of a&n RNA tumor virus specific DNA. Cold Spring Harbor Symp. Quant. Biol. 39:1005-1007. 9. Fujinaga, K., I. Sekikawa, H. Yamazaki, and M. Green. 1975. Analysis of multiple viral genome frag- ments in adenovirus 7-transformed hamster cells. Cold Spring Harbor Symp. Quant. Biol. 39:633-636 10. Gancsan, A. T., J. J. Andersen, J. Luh, and M. Effron. 19i6. Deoxyribonucleic acid metabolism in Bacillus subtilis and its phage SPPl, p. 319-325. In D. Schles- singer fed.), hiicrobiology-19i6. American Society for Microbiology, Washington, D.C. 11. Garro, A. J., P. Hammer, and B. Recht. 1976. Biochem- ical and genetic analysis of the defective BaciIIus sub- tilis bacteriophage PBSX, p. 340-349. In D. Schlessin- ger fed.), Microbiology-1976. American Society for Microbiology, Washington, D.C. 12. Graham, E. S., F. E. Young, and G. A. Wilson. 19i7. Effect of site specifc endonuclease digestion on the flzyP3 gene of bacteriophage @T and the thyPl1 gene of bacteriophage pll. Gene 1:169-180. 13. Gunther&V., B. Pavlek, J. Stutz, and T. A. Trautner. 1976. Restriction and modification in Bacillus subtilis inducibility of a DNA methylation activity in non- modifying cells. J. Viral. 20:188-195. 14. Kelly, T. J.. Jr., and D. Nathans. 1977. The genome of simian virus 40. Adv. Virus Res. 21:85-173. 15. Kim, J. S., and N. Davidson. 1974. Electron microscope heteroduplcx study of sequence relations of T2. T4 and T6 bacteriophage DNAs. Virology 57:93-111. 16. Klotz, L. C., and B. H. Zimmer. 1972. Size of DNA determined by viscoelastic measurements: results on bacteriophages, B. subtilis and E. roli. J. Mol. Biol. 72: 779-800. 17. hfartin, hf. A., G. Khoury, and G. C. Fareed. 1975. Specific reiteration of viral DNA sequences in mam- malian cells. Cold Spring Harbor Symp. Quant. Biol. 39:129-136. 18. Neiman, P. E., S. E. Wright, and H. G. Purchase. 1975. Studies of the interrelationship of chicken leukosis virus and host cell genomes by RNA-DNA hybridization. Cold Spring Harbor Symp. Quant. Biol. 39:875-884. 19. O'Donovan, G. A., and J. Neuhard. 1970. Pyrimidine metabolism in microorganisms. Bacterial. Rev. 34:278- 343. 20. Okamoto K., J. A. Mudd. J. Mangan. W. M. Huang, T. V. Subbaiah. and J. Marmur. 1968. Properties of the defective phage of Bacillus subtilis. J. Mol. Biol. 34:413-428. 21. Pellicer, A., hf. WigIer, R. Axe], and S. Silverstein. 1978. The transfer and stable integration of the HSV thymidine kinase gene into mouse celIs. Cell 14:133- 141. 22. Polisky, B., P. Greene, D. Garfm. B. J. McCarthy, H. M. Goodman, and H. Boyer. 1975. Specificity of sub- strate recognition by the EcoRI restriction endonucle- me. Proc. Natl. Acad. Sci. U.S.A. 72:33lC-3314. 23. Radloff, R., W. Bauer. and J. Vinograd. 1967. A dye- buoyant-density method for the detection and isolation of dosed circul~ duplex DNA: the dosed circular DS.i in Hela cells. Proc. Natl. Acad. Sci. C.S.A. 5i:1.5::- 1521. 24. Sambrooli, J., 31. Botchan, P. Gallimore. B. Ozatme. U. Petersson, J. Williams, and P. ;L Sharp. 1":~ Viral D9A -seququences in cells transformed by s&t. virus 40, adenorirus type 2 and adenovinx type 5. Cc.t Spring Harbor Smp. Quant. Biol. 39:615-632. 25. Seaman, E., E. Tarmy, and J. Marmur. 1964. Inducti,:+ phages of E. subtilk. Biochemisrry 3:60;-613. 26. Sgaramella, V, H. Bursztyu-Pettegrew, and S. D. Ehrlich. 1977. U-se of the T4 ligate to join flush-endti DNA wgments, p. 57-68. In R. F. Beers, Jr., and E. G Basset ted.), Recombinant molecules: impact on scienc-e and sociery. Raven Press, New York. 27. Sgaramella, V., S. D. Ehrlich, H. Burszt-yn-Pettr. grew, and J. Lederberg. 1976. Enhancement of cram- fecting activity of bacteriophage P22 DSA upon eron- ucleolytic erosion J. hlol. Biol. 105:58i-&92. 28. Sinsbeimer, R. L. 1977. Recombinant DSA Annu. Re:-. B&hem 46:415-%38. 29. Southern, E. M. 19%. Detection of speciSc .sequence- amorg D% fragments separated by gel electrophorc- sic+. J. Mol. Biol. 98x503-517. 30. SpIzizen, J. 1958. Transformation of biochemicaUy defi- cient strsins of Bacillus subtilis by deox~x-ibonucleate Proc. Natl. Acad. Sci. U.S.A. 44:1052-1058. 31. Stewart, C. 1969. Physical heterogeneity among Eac;Nu.~ subtilis dronyribonucleic acid molecules czrrying par- ticular genetic markers. J. Bacterial 98:11&1247. 32. Stroynowski, I. T. 1981. Integration of rhe bacteriophage +T3-coded thymidylare synthetase gene into the Bacil- Zus subtilis chromosome. J. Bacterial. 144:101-108. 33. Subbaiah, T. V., C. D. Goldthwaite, and J. hfarmur. 1965. Nature of bacteriophages induced in BaciNus subtilis, p. 435-446. In W. Byron and J. d. Vogel cd.). Evolving genesand proteins. Academic Prw, Inc., PieN- York. 34. Teich, N. hf.. and R A. Weiss. 1977. Beware the lurking virogene. p. 471-484 In R. F. Beers, Jr.,and E. G. Basset ted.), Recombinant molecules: impact on .tience and society. Raven Press, Sew York. 35. Tucker, R. G. 1969. Acquisition of th>lr,iiylate synthe- tase activi:? by a th:mine-requiring mutant ofBacillus subtitis folloalng infection by the temperate phage +3, J. Gen. Viral. 4:489jo4. 36. Warner, F. D., G. A. Kites, M. P. Ebmano, and H. E. Hemphill. 1976. Characterization of SP/3: a xemperare bacteriophage from Bacillus subtilis 168 M. Can. J. Microbial. 23:45-51. 37. Wood, W. B.. and H. R ReveI. 1976. The genome of bacteriophage T4. Baaeriol. Rev. 40:847-868. 38. Yasbin, R E. 19X. DSA repair in Bacillus subtilis. I. The presence of an inducible system. hiol. Gen. Genet. 153:211-218. 39. Young, F. E., X T. Wiiams, and G. A. Wilson. 19i6. Development of biochemical genetics in Bocilhs sub- tilis, p 5-13. In D. SchIessinger ted.), Microbioloa- 1976. American Society for Microbiology, Wa&ington, D.C. 40. Zahler, S. A., R Z. Korman, R. Rosenthal, and H. E. Hemphill. 1977. Bocilhs subtifis bacteriophage SP,f?. Localization of the prophage attachment sire and spa- cialized trawduction. J. Bscteriol. 129:556558. i