Use of Chromosomal Integration in the Establishment andExpression of blaZ, a Staphylococcus aureus rB-Lactamase Gene, in
Bacillus subtilisCHARLES W. SAUNDERS, BRIAN J. SCHMIDT, MAX S. MIROT,t LEO D. THOMPSON, AND MARK S. GUYER*
Genex Corp., Gaithersburg, Maryland 20877
Received 24 June 1983/Accepted 23 November 1983
With several different vectors, attempts were made to establish blaZ, a Staphylococcus aureus -
lactamase gene, in Bacillus subtilis. Stable establishment of blaZ in B. subtilis was achieved by use of a
vector that allowed the integration of a single copy of the gene into the chromosome of that host. blaZ was
expressed in the heterologous host since B. subtilis strains carrying integrated blaZ produced ,-lactamaseand were more resistant to ampicillin than was wild-type B. subtilis. blaZ was stably inherited in suchstrains, as no loss of the ability to produce 1-lactamase was observed after growth in nonselective liquidmedium or on solid medium. In contrast, a blaZ-containing restriction fragment could not be established inB. subtilis with either pUB110- or pC194-based vectors. Similarly, a pC194-based shuttle vector (pGX318)containing the 5' end of blaZ (including the promoter and the coding region for the signal sequence and thefirst few amino acids of the mature protein) was unable to transform B. subtilis. Two derivatives of pGX318that could be stably established in B. subtilis were isolated. The structures of these derivatives suggestedthat inactivation of the blaZ promoter was associated with the acquisition of the ability to be established.
The analysis of the expression of genes in a heterologoushost has become an important area of study for reasons ofboth fundamental and commercial interest. The in vitroexpression of blaZ, a Staphylococcus aureus 1-lactamasegene, in a coupled transcription-translation system derivedfrom Bacillus subtilis was demonstrated by McLaughlin etal. (24). Our attempts to extend this result with in vivostudies were initially unsuccessful because we were not ableto introduce blaZ into B. subtilis with plasmid vectors.Analysis of blaZ mutations on plasmids that were obtainedfrom these attempts suggested that the problem in establish-ment was associated with the expression of blaZ. Subse-quently, the technique of chromosomal integration (14, 47)was found to allow blaZ to be established in B. subtilis. TheB. subtilis strains isolated in this way carry blaZ stably andexpress the gene efficiently. Thus, chromosomal integrationappears to be a useful technique for stabilizing the inheri-tance of genes that are not easily established on plasmidvectors.
MATERIALS AND METHODSStrains and plasmids. Escherichia coli HB101 (hsd-20
recA13 ara-14 proA2 lac YJ galK2 rpsL20 xyl-5 mtl-i supE44)(4) was used throughout. B. subtilis BR151 (trpC2 metB10lys-3) (20) and 1A169 (trpC2 hisA sacB182) (Bacillus GeneticStock Culture) were used as noted. E. coli plasmids pSC122(42), pMB9 (37), and pBR322 (5) and B. subtilis vectorspE194, pC194, and pUB110 (6), have been described previ-ously. pGX145 was constructed (D. M. Anderson and E. A.Kalk, unpublished results) by inserting the multisite linkerregion of pUC9 (44) as an EcoRI-HindIII fragment intopKG1816 (a derivative of pKG1800 [23] containing the A totranscriptional terminator; K. McKenney, personal commu-nication). pGX345 was constructed by inserting a 1-kilobase(kb) MspI-MboI fragment from pC194 (18) between the ClaI
* Corresponding author.t Present address: Stritch School of Medicine, Loyola Universi-
ty, Chicago, IL 60626.
and BamHI sites of pGX145; the inserted fragment containscat, but lacks a replicon functional in B. subtilis (18).pGX310 and pGX311 were constructed as follows: pSC122contains a 7-kb EcoRI fragment from the S. aureus plasmidp'258 (34), which carries blaZ. We recloned (24) this EcoRIfragment into the higher-copy-number replicon pMB9, gen-erating pGX300. pGX300 was digested into four fragmentswith HindIII and EcoRI, and the digest was self-ligated.Among the ampicillin-resistant (Apr) transformants recov-ered were two strains containing smaller plasmids, pGX310and pGX311 (which each contain 3.4 kb of pI258-derivedDNA).Media and reagents. The compositions of L broth and L
agar (13) have been described previously. S7 is the syntheticmedium of Vasantha and Freese (43) in which the morpho-linepropanesulfonic acid buffer was replaced by 100 mMpotassium phosphate (pH 7.0).For plasmid preparations, M9 buffer (36) was supplement-
ed with 0.6% Casamino Acids, 1 pg of thiamine per ml, 45puM FeSO4, 2 mM MgSO4, and 0.2% glucose. Minicelllabeling medium consisted of morpholinepropanesulfonicacid medium supplemented with amino acids, purines, py-rimidines, and vitamins as described previously (30), withthe following modifications: arginine, 6 mM; cysteine, 0.5mM; riboflavin, 1 ,ug/ml; folate and niacin, 2 ,ug/ml; pyridineand pyridoxal, 4 ,ug/ml; biotin and pyridoxamine, 8 ,ug/ml;methionine, p-hydroxybenzoic acid, and 2,3-dihydroxyben-zoic acid were omitted.
Restriction endonucleases were purchased from New En-gland Biolabs and Boehringer Mannheim and used as de-scribed by Maniatis et al. (21). Nuclease BAL-31 waspurchased from Bethesda Research Laboratories and wasused according to the supplier's recommendations. Lyso-zyme, RNase A, proteinase K, and agarose (type II) werepurchased from Sigma Chemical Co. [a-32P]dCTP and[35S]methionine were obtained from Amersham Corp. Acryl-amide-bisacrylamide (29:1) was purchased from Bio-RadLaboratories.
718
Dow
nloa
ded
from
http
s://j
ourn
als.
asm
.org
/jour
nal/j
b on
24
Nov
embe
r 20
21 b
y 18
3.96
.70.
151.
INTEGRATION OF blaZ 719
Analytical gel electrophoresis. Samples were electrophor-esed as described by Gellert et al. (8).
Transformation. Competent B. subtilis cells were preparedas described previously (39), centrifuged, suspended in 0.1volume of spent growth medium, frozen in liquid nitrogen,and stored at -80°C. For transformation, frozen competentcells were thawed quickly at 37°C and diluted with 1 to 3volumes of medium SpC (39). A 0.3-ml sample of cells wasmixed with DNA and incubated at 37°C on a reciprocalshaker for 30 min. Samples (0.1 ml) were plated in a 2.5-ml0.7% agar overlay on an L plate, which was incubated at37°C for 1 h before being overlaid with an additional 2.5 ml of0.7% agar containing chloramphenicol or kanamycin to givea final concentration in the plate of 10 pLg of chloramphenicolor kanamycin per ml.Competent E. coli cells were prepared as described previ-
ously (21), except that cells were treated and suspended in100 mM CaCl2-50 mM MgCI2. Glycerol was then added to20%, and 0.5-ml samples were frozen and stored at -80°C.Transformation was performed as described previously (21)with cells that had been thawed on ice. Transformants wereselected by spreading samples on L agar containing 30 ,ug ofampicillin per ml alone or with any one of the followingadditions: 12.5 ,ug of tetracycline per ml, 10 ,ug of chloram-phenicol per ml, or 10 ,ug of kanamycin per ml.
Preparation of plasmid DNA. Small amounts of plasmidDNA were prepared by the method of Holmes and Quigley(17) from E. coli cells grown overnight at 37°C in L brothcontaining the appropriate antibiotic(s). For preparation ofsmall amounts of plasmid DNA from B. subtilis, the methodof Birnboim and Doly (3) or a modification of the method ofHolmes and Quigley was used. For the latter, an overnightculture, grown at 37°C in Penassay broth (Difco Labora-tories) containing antibiotic(s), was diluted 20-fold into thesame medium and incubated for 4 h at 37°C. A 5-ml sampleof the culture was centrifuged, and the cells were suspendedin 350 Ru of STET buffer (8% sucrose, 5% Triton X-100, 50mM EDTA, 50 mM Tris-hydrochloride [pH 8.0]). A 50-,ulsample of lysozyme (10 mg/ml in STET) was added, and thesample was incubated for 30 min at 37°C. After boiling for 1min and centrifuging in a microfuge for 10 min, the superna-tant was removed and treated with S ,ul of proteinase K (5mg/ml), first at 37°C for 15 min and then at 75°C for 20 min.The DNA was precipitated with an equal volume of isopro-panol and suspended in 50 pul of water.
Large amounts of plasmid DNA were prepared from B.subtilis as described previously (11). From E. coli, largeamounts of plasmid DNA were prepared by the method ofHolmes and Quigley (17). DNA prepared by either methodwas purified by dye-buoyant density equilibrium centrifuga-tion as described previously (12). After removal of theethidium bromide, the DNA was dialyzed into 10 mM Tris-hydrochloride (pH 8.0)-i mM disodium EDTA.
Preparation of B. subtilis chromosomal DNA. A culture wasgrown at 37°C in in 100 ml of L broth to an absorbancy at 600nm of about 0.4. The cells were spun down and suspended in2 ml of 150 mM NaCI-100 mM EDTA, pH 8.0. A 0.25-mlsample of lysozyme (4 mg/ml in 150 mM NaCI-100 mlEDTA, pH 8.0) was added, and the sample was incubatedfor 25 min at 37°C. After the addition of sodium dodecylsulfate (SDS) to 1.3% and vortexing, the sample cleared andbecame viscous. The sample was extracted with an equalvolume of phenol-chloroform (1:1). The DNA was precip-itated with 2.5 volumes of 95% ethanol, suspended in 1 ml ofwater, and treated with 5 p.1 of boiled RNase A (10 mg/ml).
Gel transfer hybridization. A 5-pI sample of a chromo-
somal DNA preparation (above) was digested with restric-tion enzyme(s) in a total of 50 [LI. The digested DNA waselectrophoresed through 0.8% agarose in E buffer (50 mMTris base, 30 mM sodium acetate, 3 mM disodium EDTA,adjusted to pH 7.8 with glacial acetic acid), transferred tonitrocellulose paper (Schleicher & Schuell Co., BA85), andhybridized to 32P-labeled DNA as described by Wahl et al.(45). DNA was labeled with [oa-32P]dCTP with a nick transla-tion kit from Bethesda Research Laboratories.DNA sequence analysis. The DNA sequence was deter-
mined by the procedure of Maxam and Gilbert (22).P-Lactamase determinations. The ability of strains to pro-
duce ,-lactamase was assessed either by pipetting 100 [L1 ofan overnight culture or by streaking a colony smear on aCefinase disk, according to the instructions of the manufac-turer (BBL Microbiology Systems).
Immunoprecipitation of ,-lactamase. E. coli strains weregrown to an absorbancy at 600 nm of 0.4 to 0.6 in L brothcontaining 30 p.g of ampicillin per ml. Cells from 1 ml ofculture were washed twice with 0.5 ml of 200 mM NaCl andsuspended in 400 p.1 of minicell labeling medium. A 100-pAsample of the cell suspension was incubated at 37°C for 10min. A 50-,uCi sample of [35S]methionine was added, andincubation was continued for 3 min. A 0.5-ml sample of cold10% trichloroacetic acid was added, and the sample was kepton ice for at least 1 h.The sample was then centrifuged at 4°C for 10 min in a
microfuge. The pellet was washed with 0.5 ml of cold 5%trichloroacetic acid and suspended in 50 ,u1 of SP buffer (50mM Tris-hydrochloride [pH 6.8], 1% SDS, 2 mM disodiumEDTA, 1% P-mercaptoethanol, 10% glycerol) plus 2 pA of 1M Tris base. The sample was boiled for 5 min, vortexed, andboiled again for 5 min. A 150-pu sample of IPD buffer (50 mMTris-acetate [pH 7.4], 0.5% Triton X-100, 2 mM disodiumEDTA, 200 mM NaCl) was added.The sample was pretreated with activated staphylococcal
cells. For activation, the cells (Pansorb; Calbiochem) werecentrifuged and suspended in the original volume of 50 mMTris-acetate (pH 7.4)-0.5% Triton X-100-2 mM disodiumEDTA-150 mM NaCl. After incubation for 15 min at roomtemperature, the cells were again centrifuged and suspendedin the original volume of IPD-SP (3:1) buffer and stored onice. For pretreatment of the sample, 100 plA of activatedstaphylococcal cells was added, and the mixture was incu-bated for 30 min on ice. The cells were then removed bycentrifugation, and the supernatant was used for immuno-precipitation.A 5-pA sample of antibody to staphylococcal ,-lactamase
was added (50 pl of antibody precipitates 10 ,ug of ,1-lactamase; J. Nielsen, personal communication), and themixture was incubated overnight at 4°C. A 100-pA sample offreshly activated staphylococcal cells was added, and themixture was incubated for 30 min on ice. The cells werepelleted for 90 s in a microfuge and washed three times withIPW buffer (10 mM Tris-hydrochloride [pH 7.4], 150 mMNaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.5%SDS), each time completely suspending the pellet. After thefinal wash, the cells were suspended in 50 ,ul of SP buffer andboiled for 5 min. The cells were removed by centrifugation,and the supernatant was saved for analysis.For immunoprecipitation of B. subtilis extracts, cultures
were grown to an absorbancy at 600 nm of 0.3 to 0.6 in S7medium supplemented with tryptophan and lysine (50 ,ug/mleach) and methionine (5 ,ug/ml). The cells were centrifugedat room temperature and suspended in the original volume ofS7 supplemented with tryptophan and lysine only (50 p.g/ml).
VOL. 157, 1984
Dow
nloa
ded
from
http
s://j
ourn
als.
asm
.org
/jour
nal/j
b on
24
Nov
embe
r 20
21 b
y 18
3.96
.70.
151.
720 SAUNDERS ET AL.
)pCl 94 blaZ\ , tet p(2.9kb) kan pUB110
H B (4-5 kb)
R R
P~~~~~~~~
pGX314 pGX312 ..Bb/Z (11kb) (13kb)blaZ X blaZ @ R
BH
P cat H ka
FIG. 1. Construction of bIaZ-containing shuttle vectors. pGX300(not shown) is a chimera of a 7-kb blaZ-containing EcoRI fragmentof pSC122 with pMB9; pGX300 contains two HindIII sites and twoEcoRI sites. As described in the text, pGX311 was generated frompGX300 after digestion with both HindlIl and EcoRI. pGX311contains a 3.4-kb fragment of staphylococcal origin ( ) thatcontains blaZ (c=.) and the large EcoRI-HindIII fragment of pMB9(- - ). pGX314 is a chimera of pGX311 and pC194 (ED), joinedat the unique HindIlI site of each plasmid. pGX312 is a chimera ofpGX311 and pUB110 ( "'-t- ),joined at the unique BamHI site of eachplasmid. Restriction endonuclease recognition sites are indicated asfollows: E, EcoRI; H, HindlIl; B, BamHI; P, PvuII. In a typicalexperiment, no transformants (<10 per ml of transformed cells)were recovered after exposure of BR151 to approximately 1 ,ug ofeither pGX312 or pGX314 DNA. Approximately 4,000 Kmr coloniesper ml of transformed cells were obtained after exposure to about0.3 p.g of pUB110 DNA in the same experiment. Mixing eitherpGX312 or pGX314 DNA with pUB110 DNA did not prevent thelatter plasmid from transforming BR151. A total of four attemptswere made using strain 1A169 as the recipient, and one experimentused strain BR151.
After shaking for 15 min at 37°C, 100 [L1 of the culture weretransferred to a microfuge tube, and 50 RCi of [35S]methio-nine was added. Samples were incubated for 5 min at 37°Cand then spun for 2 min in a microfuge at 4°C. The superna-tant was removed, and 1 ml of cold 10% trichloroacetic acidwas added; this was considered the extracellular fraction.The cell pellet was suspended in the original volume of lysineplus tryptophan-supplemented S7 medium, and 5 RI oflysozyme (20 mg/ml in water) was added. The mixture wasincubated for 2.5 min at 37°C, and 1 ml of cold 10%trichloroacetic acid was added. The samples were thentreated as described above for E. coli samples.
Immunoprecipitated samples (5 ml) were counted in 3 mlof 3a70B Complete Counting Cocktail (Research ProductsInternational Corp.) in a Beckman LS6800 liquid scintillationcounter. Samples of 2 to 20 p.1 were subjected to electropho-resis on a 15% SDS-polyacrylamide gel as described previ-ously (41). After staining and destaining, the gel was pre-pared for fluorography with En3Hance (New EnglandNuclear Corp.). The dried gel was exposed to Kodak XAR-2film at -80°C.
RESULTSInability to establish blaZ in B. subtilis on a multicopy
plasmid. pGX312 and pGX314 (Fig. 1) are blaZ-containingplasmids that were designed to be shuttle vectors as they
contain replicons capable of extrachromosomal maintenancein both E. coli (the pMB9 replicon) and B. subtilis (the pC194or pUB110 replicons). However, although each plasmidcould be easily established and maintained in E. coli (datanot shown), no colonies containing either plasmid wererecovered from attempts to transform competent cells ofeither of two B. subtilis strains, BR151 and 1A169 (Fig. 1).
(a) _ B
(4.7)I"pGX2428 an(84b4k)
*f R
blaZ
R XB
kapBlpUB11 0
(b)
kan
B
R
(2.4) k
pGX2430 kan
(64kb)
B
R X B
kan
pUB1 1 0
x
FIG. 2. Analysis of the ability of portions of pGX311 to beestablished on plasmids in B. subtilis. Restriction endonucleaserecognition sites are indicated as in Fig. 1; also X, XbaI. (a) A 1-p.gsample of pGX311 that had been digested with EcoRl and BamHIand 0.1 ,ug of pUB110 that had been digested with EcoRI, BamHI,and XbaI were incubated with DNA ligase in a volume of 10 ,ul for4 h at 12°C (XbaI was used to digest pUB110 to reduce thebackground of transformants carrying pUB110). Thirty Kmr trans-formants were obtained upon mixing with competent BR151 cells.Of 11 transformants examined, 10 had plasmids identical topGX2430; the other carried pUB110. In a control experiment 1,200transformants were obtained with 0.1 ,ug of undigested pUB110. (b)A 1-p.g sample of pGX311 that had been digested with EcoRl andXbaI and 0.06 p.g of pUB110 that also had been digested with EcoRIand XbaI were incubated with DNA ligase in a volume of 14 ,ul for 2h at 12°C. Fifty Kmr transformants were obtained upon mixing withcompetent BR151 cells. No transformants were observed to containthe 6.0-kb fragment of pGX311. All 12 transformants examined hadplasmids identical to pGX2428. In a control experiment 0.1 p.g ofundigested pUB110 yielded 2,700 transformants.
kan
J. BACTERIOL.
Dow
nloa
ded
from
http
s://j
ourn
als.
asm
.org
/jour
nal/j
b on
24
Nov
embe
r 20
21 b
y 18
3.96
.70.
151.
INTEGRATION OF blaZ 721
(a)
f cat
.-*----9blaZA31 , . fbiaZ::IS1
pGX328 pGX2427
., cat * a cat
(b)
Hindill p.. S/D TTG V Sa/lI I I
50 100 150 200 240FIG. 3. Derivatives of pGX318 that can be established in B.
subtilis. (a) pGX318 is a pC194-based shuttle vector that carriesabout 240 base pairs of the 5' end of blaZ (see b). It was constructedby ligating PvuIl-digested pC194 with an NruI digest of a mixture ofseveral E. coli plasmids that contain the promoter of blaZ anddiffering amounts of the blaZ-coding sequence (unpublished data).After transformations of B. subtilis with pGX318; only plasmidscontaining rearrangements in the blaZ component of pGX318 (repre-sented by pGX328 and pGX2427) were recovered. (b) Diagram ofthe region of blaZ present in pGX318. The locations of the promot-er, Shine-Dalgarno sequence (S/D), the TTG initiation codon forblaZ and the signal sequence processing site (V) are as described byMcLaughlin et al. (25).
Although a few kanamycin-resistant (Kmr) colonies wereobtained in the experiments with pGX312, these all harboredplasmids that were considerably smaller and differed signifi-cantly in restriction pattern from the parental plasmids. Inparticular, blaZ did not appear to be intact in any of thesetransformants.Although these results suggested that some shuttle vectors
based upon pGX311 could not be established in B. subtilis,fragments comprising most of pGX311 could be establishedin that host. Figure 2 outlines two experiments; in each,pGX311 was digested with two restriction enzymes to gener-ate a pair of fragments. Attempts were then made to clone
each fragment. However, only one fragment of the pair wasrecovered in transformants in each instance (Fig. 2). Theonly region of pGX311 that was not established on vectorpUB110 was the 1.2-kb segment extending from the BamHIsite to the XbaI site (Fig. 2b), a region which includes thepromoter and 5' end of blaZ.
Inability to establish promoter-bearing fragments of blaZ ona multicopy plasmid in B. subtilis. The previous resultssuggested that the 5' end of blaZ may have been involved inpreventing the inheritance in B. subtilis of plasmids carryingblaZ. Further evidence in support of this hypothesis camefrom experiments with pGX318 (Fig. 3), a pC194-basedshuttle vector that contains only a small region of the 5' endof blaZ. In two different transformation experiments with B.subtilis as a recipient, no colonies containing pGX318 wererecovered. In one experiment, the plasmids from the sixtransformants appeared to be similar to one another, butdifferent from pGX318 in that they contained an 800-base-pair insertion in the blaZ promoter region (Fig. 3). By its sizeand the presence of single PstI and PvuII recognition sites(data not shown), the insert appeared to be ISJ (35; the ISJinsertion presumably occurred during passage of pGX318 inE. coli; see below).
In another experiment, in which a different preparation ofpGX318 was used for transformation, five chloramphenicol-resistant (Cmr) transformants were analyzed. Plasmids infour of these were found to have large deletions in the blaZregion; the fifth plasmid (designated pGX328) contained asmall deletion. DNA sequence analysis of pGX328 revealedthat there was a'31-base-pair deletion (designated AblaZ31)in the blaZ promoter region (Fig. 4). This deletion removedthe -10 region of the blaZ promoter and the site of initiationof transcription (25).As implied by its stability and in contrast to pGX318,
pGX328 was able to transform B. subtilis. With competentBR151 cells and with 1 pg of DNA per ml in the transforma-tion tube, 510 Cmr transformants per 0.1 ml of culture wereobtained with pGX328; under the same conditions, pGX318generated only 10 Cmr transformants per 0.1 ml (the plas-mids in these few Cmr transformants had suffered rearrange-ments, as noted above). Thus, although we have been unableto establish pGX318 in B. subtilis, we have isolated twoderivatives of it which readily transform this host. As bothderivatives contained alterations in the promoter region ofblaZ, we conclude that this region of the gene is responsiblefor the inability of pGX318, and presumably other blaZ-containing plasmids, to transform B. subtilis.
Integration of blaZ into the B. subtilis chromosome. Incontrast to our inability to isolate a B. subtilis strain carryingblaZ on a plasmid, we were readily able to isolate B. subtilisstrains that stably carried blaZ integrated into the chromo-
FIG. 4. DNA sequence of the 5' nontranslated regions of blaZ in pGX318 and pGX328. The DNA sequence of this region in pGX318 wasdetermined and found to be identical to the sequence presented by McLaughlin et al. (25). Several features identified by these authors areindicated: the -35 and -10 regions are indicated in large bold type, the site of transcription initiation is indicated by an arrow, the Shine-Dalgarno sequence is denoted with italicized type, and the TTG initiation codon is underlined. Note that the sequence deleted from pGX318 isflanked by an 8-base-pair direct repeat, which is part of a larger (15 of 16-base-pair) direct repeat. Deletion of the region between short directrepeats has been observed frequently (1).
VOL. 157, 1984
Dow
nloa
ded
from
http
s://j
ourn
als.
asm
.org
/jour
nal/j
b on
24
Nov
embe
r 20
21 b
y 18
3.96
.70.
151.
722 SAUNDERS ET AL.
R H
/ tet
pGX310 '(8.9kb) J
bla-'H R
, R
Icat
-Rcat
R
pGX2434
amp ..
(N/P)
FIG. 5. Construction of blaZ-containing chromosomal intetion vectors. Chromosomal DNA fro,m B. subtilis was digestedPvuII and cloned into pGX345 that had been digested with NpGX281 and pGX282 (data for pGX282 not shown; it car
different B. subtilis chromosomal DNA than pGX281) were rec
ered among Apr Cmr transformants of E. coli. The B. subchromosomal DNA is indicated by a heavy line. A blaZ-contailEcoRI fragment from pGX310 was then cloned into the EcoRI sitpGX281 (pGX310 was derived from pGX300 as describedpGX311 [see legend to Fig. 1]). pGX310 differs from pGX31containing the small EcoRIbHindIII fragment of pMB9 (ca. 400 tpairs) at both junctions between the S. aureus and pMl39 sequenNote that EcoRI digestion of pGX310 generates a 3.8-kb fragncomprised of the 0.4-kb EcoRI-HindIII fragment of pMB9 and3.4-kb HindIII-EcoRl blaZ fragment. pGX2413 and pGX2which differ in the orientation with which the blaZ fragmentinserted into pGX281, were identified by screening the Aprtransformants of E. coli. Most of the amp gene of pGX2413deleted by ligation at the unique XhoI and BgIl sites of the plasafter digestion with those enzymes and removal of the sinstranded ends with nuclease BAL31. pGX2423 was identifiedscreening Apr Cmr transformants of E. coli. pGX2434 was isolafter digestion of pGX2413 with EcoRI and self-ligation. Symused are described in the legend to Fig. 1. Restriction enz
recognition sites are designated as in Fig. 1 and 2; also N, NruI;XhoI; Bg, BglI; (N/P), a fused half NruI and PvuII site that i'longer recognized by either enzyme.
some. pGX281 (Fig. 5) was chosen to test the ability of bto be established in B. subtilis upon integration. This plaslacks a replicon active in B. subtilis so it cannot be mtained extrachromosomally. It does, however, contain Dhomologous to a region of the B. subtilis chromosome at
genetic marker that is active in that host; these featsallow pGX281 to transform B. subtilis by integration.
A bIaZ-containing fragment from pGX310 (Fig. 5) wasinserted into pGX281 in each of two orientations, generatingpGX2413 and pGX2414, respectively. In addition to blaZ,both of these plasmids also carry the amp gene of pBR322(Fig. 5). Consequently pGX2423, a plasmid carrying blaZbut not amp, was also constructed (Fig. 5) as a control.Cmr transformants were obtained from transformation of
BR151 with pGX281 and its blaZ-carrying derivativespGX2413, pGX2414, and pGX2423 (Table 1). Transformantsthat arose from those plasmids containing blaZ produced -lactamase (Table 1). This Bla+ phenotype must be due to thepresence and expression of blaZ rather than ofamp, because(i) transformants generated from pGX2423, which carriesblaZ but lacks most of amp, were Bla+; and (ii) pGX281,which contains amp but not blaZ, did not produce Bla+transformants. Furthermore, removal of the blaZ-containingEcoRI fragment of pGX2413, which generated the plasmidpGX2434 (Fig. 5), resulted in the loss of the ability to
cat transform B. subtilis to Bla+ (Table 1). Bla+ transformantswere also obtained with blaZ-containing derivatives ofpGX282 (pGX271, pGX272; Table 1), presumably due tointegration of blaZ at a different location on the B. subtilischromosome.To test the stability of strains carrying an integrated
vector, we subcultured each of several transformants by
at patching sequentially six times onto drug-free L agar. Allcolonies that had been derived from a Bla+ transformantretained that phenotype during growth in the absence of
-(NIP) selection for the transforming plasmid (Table 2). As anothertest, we grew cultures in drug-free liquid medium for 30generations. As was the case with passage on plates, nocured derivatives were observed (Table 2). These results
gra-with TABLE 1. Transformation of BR151 with chromosomalFrul. integration vectorsrriesCov-
None <1pC194 - 60 0/6pGX345 - <1pGX281 - 5 0/25pGX2413 + 2 27/30epGX2414 + 3 11/11pGX2423 + 6 7/1oepGX2434 - 9 0/20pGX282 - 50 0/2pGX271 + 180 8/8pGX272 + 500 8/8a pC194 and pGX345 are not integration vectors since they do not
contain any DNA homologous to the B. subtilis chromosome; pC194is maintained extrachromosomally in B. subtilis. All other plasmidslisted contain B. subtilis chromosomal DNA and are integrationvectors (Fig. 5; pGX271 and pGX272 are blaZ-containing deriva-tives of pGX282).
b +, Plasmid carries blaZ; -, plasmid does not carry blaZ.' The number of Cmr transformants per 0.1 ml of transformed
cells plated is given. Donor DNA (1 p.g/ml in the transformationtube) was used to transform competent BR151 cells. Cmr transform-ants were selected as described in the text.
d The number of Bla+ transformants per number of Cmr trans-formants tested in given. A representative sample of the Cmr trans-formants was scored for p-lactamase production with Cefinasedisks.
e The few Cmr Bla- colonies were not reproducibly observed intransformations with pGX2413 and pGX2423. We suspect that theyrepresented the rare appearance of colonies on chloramphenicolplates that are unrelated to the presence of transforming DNA.
R.... I.. cat
pGX281
(5.7kb)
amp
(N/P)
pGX242358.5kb)
(Xh/Bg)
J. BACTERIOL.
Dow
nloa
ded
from
http
s://j
ourn
als.
asm
.org
/jour
nal/j
b on
24
Nov
embe
r 20
21 b
y 18
3.96
.70.
151.
INTEGRATION OF hlaZ 723
TABLE 2. Stability of Bla+ in B. subtilis strains carrying integrated plasmidsPlate passageb Liquid passage"
a All plasmids listed except pC194 are integration vectors. pC194 is maintained extrachromosomally in B. subtilis.b A colony of each transformant was transferred by patching daily onto drug-free L agar for 6 to 10 days, after which samples were streaked
onto drug-free L agar. Single colonies were then picked onto L agar with and without chloramphenicol (10 p.g/ml) and tested forchloramphenicol resistance and P-lactamase production.
' A colony of each transformant was inoculated into L broth and subcultured for a total of 30 generations. Samples were then streaked on Lagar, and single colonies were scored for chloramphenicol resistance and f3-lactamase production.
demonstrate that blaZ was stably inherited by strains carry-
ing an integrated vector.Location of vector DNA in B. subtilis transformants. To
confirm that blaZ was integrated into the chromosome of thestable transformants and to analyze the structure of theintegrated vector, cellular DNA from several transformantswas examined by gel transfer hybridization (45). With 32p-labeled, linearized pGX2413 as a probe, DNA from strainsGX2499, GX2498, and GX2497 (generated by transformationwith pGX281, pGX2413, and pGX2414, respectively) was
studied. The hybridization patterns of DNA digested with a
number of restriction enzymes (Table 3) were consistentwith the hypothesis that, in each case, a single copy of theentire plasmid had integrated at the site of homology be-tween the chromosome and plasmid. No evidence for thepresence of free plasmid or multiple tandem integratedcopies in a significant fraction of the population was found.The results of similar experiments with undigested chromo-somal DNA (data not shown) were consistent with thisinterpretation.
Expression of blaZ in B. subtilis. The synthesis of S. aureus
1-lactamase was demonstrated in B. subtilis strains labeled
with [35S]methionine for 5 min. After lysis of the cells,extracts were treated with antibody to the S. aureus 13-
lactamase (kindly provided by J. 0. Lampen). Analysis ofthe labeled immnunoprecipitated protein from the Bla+strains GX2492 and GX2498 on SDS-polyacrylamide gelsshowed the presence of a band that migrated at about 38kilodaltons (Fig. 6). This band was Absent in an extract of theBla- strain GX2499 (Fig. 6). On the basis of a comparison ofthe number of immunoprecipitable counts obtained fromGX2492 and GX2499, we estimated that about 1% of thetrichloroacetic acid-insoluble counts (total incorporationinto protein) was specifically immunoprecipitable with theanti-1-lactamase antibodies. Most of the immunoprecipita-ble material from B. subtilis was cell associated (Fig. 6).
In an E. coli strain carrying blaZ on pGX271, the immuno-precipitable counts were also found to represent about 1% ofthe total synthesis (data not shown). The immunoprecipitat-ed material from E. coli also migrated as a 28-kilodaltonprotein.Growth of B. subtilis in the presence of ampicillin. Imanaka
et al. (19) reported that B. subtilis strains carrying thepenicillinase gene (penP) of Bacillus licheniformis were
TABLE 3. Summary of results of Southern blot experimentsSize (kb) of fragments predicted to hybridize if the transforming
B. subtilis strain used to Enzyme used to Size (kb) of plasmid were established as:prepare chromosomal digest chromo- fragments Multiple tan-
DNA" somal DNA pGX2413 Single chromo- dem chromo- Free plasmid'somal copiesb
GX2498(pGX2413) and EcoRI 3.8 3.8 3.8 3.8GX2497(pGX2414) 8 Two junction 5.7 5.7
a The integration vector used to transform BR151 to generate each strain is indicated in parentheses.b The predicted fragment sizes are calculated for head to tail repeats.c In addition to the predicted fragments listed, which correspond to plasmid-derived fragments, a fragment derived from the B. subtilis
chromosome would be expected. This would be 14 kb for the EcoRI digest, 4 kb for the BglII digest, and 1.1 kb for the PvuII digest.d The mobility of the band indicated would be that of undigested plasmid DNA.The mobility of the band indicated would be that of monomer length linear DNA.
f The BgIl + PvuII experinment was not done with GX2498 or BR151.
VOL. 157 1984
Dow
nloa
ded
from
http
s://j
ourn
als.
asm
.org
/jour
nal/j
b on
24
Nov
embe
r 20
21 b
y 18
3.96
.70.
151.
724 SAUNDERS ET AL.
A B C D E F
I43
0)- 20)
FIG. 6. SDS-gel electrophoresis of [355]methionine-labeled B.subtilis extracts that were immunoprecipitated with antiserum tostaphylococcal P-lactamase. Cultures of GX2499 (lanes A and B),GX2498 (lanes C and D), and GX2492 (lanes E and F) werefractionated into supernatants (lanes B, C, and E) and cell-associat-ed material (lanes A, D, and F) as described in the text. The mobilityof protein standards (egg white ovalbumin, bovine erythrocytecarbonic anhydrase, soybean trypsin inhibitor) is indicated. GX2499(Bla-) was generated by transformation of BR151 with pGX281,GX2498 (Bla+) was generated by transformation of BR151 withpGX2413, and GX2492 (Bla+) was generated by transformation ofBR151 with pGX2423.
unable to form a colony on L agar containing 20 ,ug ofampicillin per ml. Such strains, however, could grow in thepresence of ampicillin after replica plating, as also noted byGCray and Chang (10). B. subtilis strains containing integratedblaZ were able to grow on L agar containing ampicillin atconcentrations up to 500 jig/ml, but only in the region of astreak where the inoculum was heavy. These strains wereunable to form single colonies on medium containing ampi-cillin at a concentration of 5 jig/ml or more. B. subtilis strainsthat did not carry blaZ were unable to grow at all in thepresence of 10 ,ug of ampicillin per ml.
DISCUSSIONThe results described in this paper demonstrate that an S.
aureus 1-lactamase gene (blaZ) can be established, stablymaintained, and expressed in B. subtilis. The expression ofblaZ in transformants generated from both pGX2413 andpGX2414, which differ in the orientation of the blaZ frag-ment, suggests that the native blaZ expression signals,rather than signals supplied by an adjacent DNA sequence,were being used in B. subtilis. Furthermore, we have foundthat a 130-base-pair region of blaZ, extending from theupstream HindIII site (Fig. 3b) through the promoter, butcontaining no structural gene information, was sufficient toallow expression of a cat gene lacking a promoter (46) in B.subtilis (data not shown). These results strongly suggest thatthe S. aureus blaZ promoter is active in vivo in B. subtilis.With respect to the protein that is produced in B. subtilis,
the mobility of the 35S-labeled material precipitated by theanti-S. aureus 1-lactamase serum suggested a molecularmass (28 kilodaltons) for the blaZ-specific product that issimilar to that expected for an S. aureus 1-lactamase (2).However, since the S. aureus 1-lactamase is a processedlipoprotein in its native host (32), the blaZ product made in
either B. subtilis or E. coli may exist in any one of severalforms, and further characterization of the protein will benecessary to determine the form(s) of staphylococcal -lactamase that is made in these heterologous hosts.
In S. aureus, the blaZ product is largely secreted to theextracellular medium, although about one-third is membranebound (32). We have observed only cell-bound 1-lactamasein B. subtilis. However, the extracellular staphylococcal 3-lactamase is very sensitive to proteases (J. 0. Lampen,personal communication). As we have not yet attempted tominimize proteolysis, we cannot draw any conclusions aboutthe ability of B. subtilis to secrete this heterologous geneproduct.The introduction of blaZ into B. subtilis was accomplished
by the use of a vector that must integrate into the B, subtilischromosome to be inherited (successful establishment ofblaZ was obtained with each of four different integrationvectors; Table 1 and data not shown). In contrast, blaZcould not be established on several shuttle vectors basedupon plasmids maintained at high copy number in B. subtilis.This inability appeared to be associated with the presence ofthe blaZ promoter. Stassi and Lacks (40) among others (9),have described a similar correlation between the presence ofa strong promoter on a plasmid and the failure of thatplasmid to be established.The behavior of blaZ in B. subtilis appears to parallel the
behavior of the B. lichenformis ,B-lactamase gene (penP) inthe same host. Imanaka et al. (19) were unable to clone anEcoRI fragment containing a highly expressed allele ofpenPinto the EcoRI site of pUB110 (i.e., no Bla+ colonies weredetected among Kmr transformants). However, the samefragment could be cloned into a low-copy-number plasmidvector. In contrast to the behavior of this highly expressedallele, an allele of penP that was not so highly expressedcould be cloned into either vector. Several analogous exam-ples of restriction fragments that cannot be cloned on highcopy number vectors, but that can be cloned on lower-copy-number vectors, have been reported in E. coli (15, 16, 27).These results suggest that both the degree of expression of
the gene being cloned and the nature of the vector being usedcan play a role in whether a cloned gene will be successfullyestablished in a particular host on a particular vector. In thecase of blaZ, we cannot determine at present whether thefailure to be established on plasmids in B. subtilis is due tosome toxic effect(s) of a product of the cloned DNA, todisruption of some essential plasmid function as a result ofthe presence of the cloned insert, or to some other cause.However, any interpretation will need to account for the factthat neither the intact blaZ gene (on pGX312 and pGX314)nor a small fragment from the 5' end (on pGX318) could beestablished.Chromosomal integration vectors have been previously
used in B. subtilis to map genes for which there is noselectable marker (14, 47) or for complementation analysis(7). In this paper, we have demonstrated the additional valueof such vectors in the stabilization of a highly expressedgene. Although existing low-copy-number plasmid vectorsmay also be useful for this purpose (19), the chromosomalintegration vectors are attractive alternatives for severalreasons. They have been extensively characterized bothgenetically and physically. They are able to replicate in E.coli, where plasmid constructions and purification can oftenbe more easily done than in B. subtilis. Chromosomalintegration vectors such as pGX281 also contain manyunique restriction sites to facilitate plasmid manipulations.Finally, low-copy-number plasmids may be sensitive to the
J. BACTERIOL.
Dow
nloa
ded
from
http
s://j
ourn
als.
asm
.org
/jour
nal/j
b on
24
Nov
embe
r 20
21 b
y 18
3.96
.70.
151.
INTEGRATION OF blaZ 725
effects of expression of some cloned genes. Although DNAanalysis of the integrated sequences is not straight-forward,reestablishing an integrated vector as a plasmid in E. coli hasbeen reported (26, 31, 38; C. Rudolph, personal communica-tion); this would allow the vectors to be analyzed rapidly.
Finally, with regard to the origin of pGX2427 (Fig. 3), it islikely that the IS] insertion into pGX318 occurred in E. coli.First, there appear to be no IS] sequences in B. subtilis (33).Second, when the DNA preparation (from E. coli) used inthe transformation from which pGX2427 had been recoveredwas examined carefully, it was found to be a mixture ofpGX318 and a small amount of pGX318::ISI. The latter wasapparently identical to pGX2427. We have observed severalindependent IS] insertions into blaZ segments in E. coli(data not shown) in addition to that described in pGX2427.Although we have not sequenced the IS] insertion site in anyof these cases, the target sites for IS] have been reported tobe AT rich (28). The blaZ promoter, like many otherpromoters from gram-positive bacteria (29), is very AT rich(25). Thus, IS] insertion into such promoters may not be aninfrequent event during work with genes of gram-positivebacterial origin in E. coli.
ACKNOWLEDGMENTSWe thank Carl Banner, Steve Fahnestock, Kathryn Fisher, Ethel
Jackson, Vasantha Nagarajan, Linc Sonenshein, and Bob Straus-berg for very helpful discussions. DNA sequence analysis ofAblaZ3l was done by Paul Reed. We are also grateful to TeresaMancusi, Shirley Trattner, and Laurie Hinds for their cooperationand skill in the preparation of the manuscript and to Joy Boudreauxand Dale Windsor for their help in the preparation of the figures.
LITERATURE CITED1. Albertini, A. M., M. Hofer, M. P. Calos, and J. H. Miller. 1982.On the formation of spontaneous deletions: the importance ofshort sequence homologies in the generation of large deletions.Cell 29:319-328.
2. Ambler, R. P. 1975. The amino acid sequence of Staphylococ-cus aureus penicillinase. Biochem. J. 151:197-218.
3. Birnboim, H. C., and J. Doly. 1979. A rapid alkaline extractionprocedure for screening recombinant DNA. Nucleic Acids Res.7:1513-1523.
4. Bolivar, F., and K. Backman. 1979. Plasmids of Escherichia colias cloning vectors. Methods Enzymol. 68:245-280.
5. Bolivar, F., P. J. Greene, M. C. Betlach, H. L. Heyneker, andH. W. Boyer. 1977. Construction and characterization of newcloning vehicles. Gene 2:95-113.
6. Ehrlich, S. D., B. Niaudet, and B. Michel. 1982. Use of plasmidsfrom Staphylococcus aureus for cloning of DNA in Bacillussubtilis. Curr. Top. Microbiol. Immunol. 96:19-29.
7. Ferrari, E., and J. A. Hoch. 1983. A single copy, transduciblesystem for complementation and dominance analyses in Bacil-lus subtilis. Mol. Gen. Genet. 189:321-325.
8. Gellert, M., K. Mizuuchi, M. H. O'Dea, and H. A. Nash. 1976.DNA gyrase: An enzyme that introduces superhelical turns intoDNA. Proc. Natl. Acad. Sci. U.S.A. 73:3872-3876.
9. Gentz, R., A. Langner, A. C. Y. Chang, S. N. Cohen, and H.Bujard. 1981. Cloning and analysis of strong promoters is madepossible by the downstream placement of a RNA terminationsignal. Proc. Natl. Acad. Sci. U.S.A. 78:4936-4940.
10. Gray, O., and S. Chang. 1981. Molecular cloning and expressionof Bacillus licheniformis P-lactamase gene in Escherichia coliand Bacillus subtilis. J. Bacteriol. 145:422-428.
11. Gryczan, T. J., S. Contente, and D. Dubnau. 1978. Characteriza-tion of Staphylococcus aureus plasmids introduced by transfor-mation into Bacillus subtilis. J. Bacteriol. 134:218-329.
12. Guyer, M. S. 1978. The -yi sequence of F is an insertionsequence. J. Mol. Biol. 126:347-365.
13. Guyer, M. S. and A. J. Clark. 1975. cis-Dominant, transfer-deficient mutants of the Escherichia coli K-12 F sex factor. J.Bacteriol. 125:233-247.
14. Haldenwang, W. G., C. D. B. Banner, J. F. Ollington, R. Losick,J. A. Hoch, M. B. O'Connor, ai;d A. L. Sonenshein. 1980.Mapping a cloned gene under sporulation control by insertion ofa drug resistance marker into the Bacillus subtilis chromosome.J. Bacteriol. 142:90-98.
15. Harding, N. E., J. M. Cleary, D. W. Smith, J. J. Michon,W. S. A. Brusilow, and J. W. Zyskind. 1982. Chromosomalreplication origins (oriC) of Enterobacter aerogenes and Klebsi-ella pneumoniae are functional in Escherichia coli. J. Bacteriol.152:983-993.
16. Henning, U., H. Royer, R. M. Teather, I. Hindennach, and C. P.Hollenberg. 1979. Cloning of the structural gene (ompA) for anintegral outer membrane protein of Escherichia coli K-12. Proc.Natl. Acad. Sci. U.S.A. 76:4360-4364.
17. Holmes, D. S., and M. Quigley. 1981. A rapid boiling method forthe preparation of bacterial plasmids. Anal. Biochem. 114:193-197.
18. Horinouchi, S., and B. Weisblum. 1982. Nucleotide sequenceand functional map of pC194, a plasmid that specifies induciblechloramphenicol resistance. J. Bacteriol. 150:815-825.
19. Imanaka, T., T. Tanaka, H. Tsunekawa, and S. Aiba. 1981.Cloning of the genes for penicillinase, penP and penI, ofBacillus licheniformis in some vector plasmids and their expres-sion in Escherichia coli, Bacillus subtilis, and Bacillus licheni-formis. J. Bacteriol. 147:776-786.
20. Lovett, P. S., E. J. Duvall, and K. M. Keggins. 1976. Bacilluspumilus plasmid pPL10: properties and insertion into Bacillussubtilis 168 by transformation. J. Bacteriol. 127:817-828.
21. Maniatis, T., E. F. Fritsch, and J. Sambrook (ed.) 1982. Molecu-lar cloning. Cold Spring Harbor Laboratory, Cold Spring Har-bor, N.Y.
22. Maxam, A. M., and W. Gilbert. 1976. A new method forsequencing DNA. Proc. Natl. Acad. Sci. U.S.A. 74:560-564.
23. McKenney, K., H. Shimatake, D. Court, U. Schmeissner, C.Brady, and M. Rosenberg. 1981. A system to study promoterand terminator signals recognized by Escherichia coli RNApolymerase, p. 383-415. In J. G. Chirikjian and T. S. Papas(ed.), Gene amplification and analysis, vol. 2. Analysis ofnucleic acids by enzymatic methods. Elsevier-North Holland,Amsterdam.
24. McLaughlin, J. R., C. L. Murray, and J. C. Rabinowitz. 1981.Plasmid-directed expression of Staphylococcus aureus P-lacta-mase by Bacillus subtilis in vitro. J. Biol. Chem. 256:11273-11282.
25. McLaughlin, J. R., C. L. Murray, and J. C. Rabinowitz. 1981.Unique features in the ribosome binding site sequence of theGram-positive Staphylococcus aur s ,-lactamase gene. J.Biol. Chem. 256:11283-11291.
26. Mejean, V., J. P. Claverys, H. Vasseghi, nd A. M. Sicard. 1981.Rapid cloning of specific DNA fragm nts of Streptococcuspneumoniae by vector integration into the chromosome fol-lowed by endonucleolytic excision. Gene\ 15:289-293.
27. Michaelis, S., L. Guarente, and J. Beckwith. 1983. In vitroconstruction and characterization of phoA-lacZ gene fusions inEscherichia coli. J. Bacteriol. 154:356-365.
28. Miller, J. H., M. P. Calos, D. Galas, and M. Hoffer. 1980.Genetic analysis of transpositions in the lac region of Escherich-ia coli. J. Mol. Biol. 144:1-18.
29. Moran, C. P., N. Lang, S. F. J. LeGrice, G. Lee, M. Stephens,A. L. Sonenshein, J. Pero, and R. Losick. 1982. Nucleotidesequences that signal the initiation of transcription and transla-tion in Bacillus subtilis. Mol. Gen. Genet. 186:339-346.
30. Neidhardt, F. C., P. L. Bloch, and D. F. Smith. 1974. Culturemedium for enterobacteria. J. Bacteriol. 119:736-747.
31. Niaudet, B., A. Goze, and S. D. Ehrlich. 1982. Insertionalmutagenesis in Bacillus subtilis: mechanism and use in genecloning. Gene 19:277-284.
32. Nielsen, J. B. K., and J. 0. Lampen. 1982. Membrane-boundpenicillinases in Gram-positive bacteria. J. Biol. Chem.257:4490-4495.
33. Nisen, P., M. Purucker, and L. Shapiro. 1979. Deoxyribonucleicacid sequences homologies among bacterial insertion sequenceelements and genomes of various organisms. J. Bacteriol.140:588-596.
VOL. 157, 1984
Dow
nloa
ded
from
http
s://j
ourn
als.
asm
.org
/jour
nal/j
b on
24
Nov
embe
r 20
21 b
y 18
3.96
.70.
151.
726 SAUNDERS ET AL.
34. Novick, R. P., E. Murphy, T. J. Gryczan, E. Baron, and I.
Edelman. 1979. Penicillinase plasmids of Staphylococcus au-
reus: restriction-deletion maps. Plasmid 2:109-129.35. Ohtsubo, H., and E. Ohtsubo. 1978. Nucleotide sequence of an
insertion element IS]. Proc. Natl. Acad. Sci. U.S.A. 75:615-619.
36. Roberts, R. B., R. H. Abelson, D. B. Crowie, E. T. Bolton, andR. S. Buttin. 1963. Studies of biosynthesis in Escherichia coli.Carnegie Institution publication no. 607. Carnegie Institution,Washington, D.C.
37. Rodriguez, R. L., F. Bolivar, H. M. Goodman, H. W. Boyer, andM. Betlach. Construction and characterization of cloning vehi-cles, p. 471-477. In D. P. Nierlich, W. J. Rutter, and C. F. Fox(ed.), Molecular mechanisms in the control of gene expression.Academic Press, Inc., New York.
38. Roeder, G. S., and G. R. Fink. 1980. DNA rearrangementsassociated with a transposable element in yeast. Cell 21:239-249.
39. Sonenshein, A. L., B. Cami, J. Brevet, and R. Cote. 1974.Isolation and characterization of rifampin-resistant and strepto-lydigin-resistant mutants of Bacillus subtilis with altered sporu-lation properties. J. Bacteriol. 120:253-265.
40. Stassi, D. L., and S. A. Lacks. 1982. Effect of strong promoters
on the cloning in Escherichia coli of DNA fragments fromStreptococcus pneumoniae. Gene 18:319-328.
41. Studier, F. W. 1973. Analysis of bacteriophage T7 early RNAsand proteins on slab gels. J. Mol. Biol. 79:237-248.
42. Timmis, K., F. Cabello, and S. N. Cohen. 1975. Cloning, isola-tion, and characterization of replication regions of complexplasmid genomes. Proc. Natl. Acad. Sci. U.S.A. 72:2242-2246.
43. Vasantha, N., and E. Freese. 1980. Enzyme changes duringBacillus subtilis sporulation caused by deprivation of guaninenucleotides. J. Bacteriol. 144:1119-1125.
44. Vieira, J., and J. Messing. 1982. The pUC plasmids, anM13mp7-derived system for insertion mutagenesis and sequenc-ing with synthetic universal primers. Gene 19:259-268.
45. Wahl, G. M., M. Stern, and G. R. Stark. 1979. Efficient transferof large DNA fragments from agarose gels to diazobenzyloxy-methyl-paper and rapid hybridization by using dextran sulfate.Proc. Natl. Acad. Sci. U.S.A. 76:3683-3687.
46. Williams, D. M., E. J. Duvall, and P. S. Lovett. 1981. Cloningrestriction fragments that promote expression of a gene inBacillus subtilis. J. Bacteriol. 146:1162-1165.
47. Wilson, F. E., J. A. Hoch, and K. Bott. 1981. Genetic mapping ofa linked cluster of ribosomal ribonucleic acid genes in Bacillussubtilis. J. Bacteriol. 148:624-628.
