Construction and evaluation of new drug-resistance cassettes for gene disruption mutagenesis in Streptococcus pneumoniae, using an ami test platform

Gene, 164 (1995) 123-128 © 1995 Elsevier Science B.V. All rights reserved. 0378-1119/95/$09.50

123

GENE 09202

Construction and evaluation of new drug-resistance cassettes for gene disruption mutagenesis in Streptococcus pneumoniae, using an ami test platform (cat; erm; lacZ reporter; pEVP3; synthetic promoter)

Jean-Pierre Claverys a, Agnes Dintilhac a, Ekaterina V. Pestova b, Bernard Martin a and Donald A. Morrison b aMolecular Genetics and Microbiology CNRS-UPR 9007, Universitk Paul Sabatier, 118 route de Narbonne, 31062 Toulouse, France. Tel. (33-61) 335-983; and bLaboratory for Molecular Biology, Department of Biological Sciences, University of Illinois, Chicago, IL 60612, USA

Received by R.E. Yasbin: 24 March 1995; Revised/Accepted: 8 May/10 May 1995; Received at publishers: 13 July 1995

SUMMARY

Although drug-resistance markers have been used frequently for gene-disruption mutagenesis in Streptococcus pneumoniae, none has yet been shown to be free of dependence on local transcription for its expression. Indeed, the erythromycin-resistance marker (erm), originating in pAMI31, has been used as an indicator of local transcription on several occasions. A procedure is demonstrated for evaluation of the autonomous expression of such a marker by placing it in a consistent background, at the pneumococcal ami (aminopterin resistance) locus, in combination with active or inactive alleles of the ami promoter (pA). Using this test platform, a chloramphenicol-resistance marker (cat) and a spectinomycinresistance marker used in streptococcal gene disruption studies and derived from pJS3 and pDL269, respectively, were shown to depend on local transcriptional signals for expression when placed in the pneumococcal chromosome as single-copy genes. To overcome this limitation, new drug-resistance cassettes were designed and constructed, using pA as a model for synthetic promoters for the erm and cat genes. Both new cassettes were shown, by the same procedure, to be expressed after insertion in the pneumococcal chromosome, independent of local transcription. A new insertionduplication vector, pEVP3, incorporating the new cat cassette and a lacZ reporter derived from pTV32, was also constructed. The ami test platform was used to demonstrate both the autonomous expression of cat and the reporter function of lacZ in chromosomal copies of pEVP3.

Correspondence to: Dr. D.A. Morrison, 900 South Ashland, Chicago, IL 60612 USA. Tel. (1-312) 996-6839; Fax (1-312)413-2691; e-mail: [email protected] The sequences of all new plasmids detailed herein are available by anonymous file transfer (ftp) from the directory /pub/depts/bios/ morrison, at ftp.uic.edu

Abbreviations: A, absorbance (1 cm); aad9, spectinomycin-resistance gene; amiACDEF, aminopterin- and Mtx-resistance genes; 13Gal, 13-galactosidase; bp, base pair(s); cat, gene encoding Cm acetyltransSSDI 0378-1119(95)00485-8

ferase; Cm, chloramphenicol; E., Escherichia; Er, erythromycin; erm, Er R gene; kb, kilobase(s) or 1000 bp; MCS, multiple cloning site(s); Mtx, methotrexate; nt, nucleotide(s); Nv, novobiocin; ORF, open reading frame(s); ori, plasmid origin of replication; pA, promoter of amiACDEF locus; pC, promoter of cat in pC194; PCR, polymerase chain reaction; pE, promoter of erm in pAMI31; Pollk, Klenow (large) fragment of E. coli DNA polymerase I; pT, promoter of tet; R resistance/resistant; RBS, ribosome-binding site(s); Sin, streptomycin; Sp, Streptococcus pneumoniae; syn, synthetic; Tc, tetracycline; tet, TcR gene; wt, wild type; ' (prime), denotes a gene truncated at the indicated side.

124 INTRODUCTION Gene-disruption mutagenesis using transposons or integrative vectors has been a valuable and popular tool in bacterial genetics. In Streptococcus pneumoniae (Sp), the mechanisms of natural genetic transformation provide an efficient path to gene disruptions in which an insertion vector carrying a small fragment of chromosomal D N A is integrated by insertion-duplication at a specific homologous target site (Vasseghi et al., 1981). If designed to include a lacZ reporter, such insertion vectors would also permit monitoring the target gene's activity by creating transcriptional or translational fusions, at either r a n d o m or specific chromosomal insertion sites. A disruption vector is not fully satisfactory if its drugresistance marker is not autonomously expressed, since the resulting bias could prevent recovery of mutations in genes that are not permanently active. Although drugresistance markers derived from multicopy plasmids are commonly used for pneumococcal gene disruption, from the earliest applications it has appeared that such markers might depend on external promoters for expression when present in a single copy. For example, expression of the Er R marker, erm, from pAM[31 (Martin et al., 1987) depends on cotranscription with a local operon or on amplification after insertion (Vasseghi and Claverys, 1983; Alloing et al., 1990). The aim of this work was to show how the well-studied Sp ami locus can be used as a convenient test-platform for evaluation of single-copy gene expression in this host, to develop new, autonomously expressed, drug-resistance cassettes, and to assemble a new gene disruption plasmid for pneumococcus carrying a lacZ reporter. These improved basic genetic tools will be useful in molecular studies of pneumococcal pathogenesis such as those reviewed by Paton et al. (1993).

EXPERIMENTAL AND DISCUSSION

(a) Evaluation of cat and aad9 markers as gene-disruption tools in S p Only a few drug-resistance markers have been used for gene disruption in pneumococcus, and none has been directly evaluated for autonomous expression after insertion in a silent locus. A priori, the cat cassette of the plasmid pJS3 appeared a good candidate for gene disruption in Sp, as it is expressed in E. coli, Bacillus subtilis and Sp on a broad-host-range plasmid (Ballester et al., 1986). This DpnI cassette, including cat of pC194 (Horinouchi and Weisblum, 1982) linked to a tet promoter derived from the streptococcal plasmid pLS1 (Ballester et al., 1986), was incorporated by Lacks and

Greenberg (1991) into a chromosome walking vector, pWG5. However, several incidental observations suggested to us that a single copy of the pJS3 cat depended on local transcription for expression. Thus, insertions of cat were preferentially recovered in the O N orientation (cat in the same polarity as the target gene) and expression of the cat in a chimeric (pC194/pLS1) pneumococcal plasmid was low but could be increased by rearrangements bringing the tet promoter near cat, as in pJS3 (Ballester et al., 1986). The ami locus offers a convenient and valuable context for evaluation of the expression of introduced elements in this species. Its sequence is known; phenotypes of many mutants are known; selections exist for both ami- and ami + alleles; it is not essential; a down-promoter mutation is available in strain R142; and plasmids carrying parts of the locus are available for directing insertion of genes as single copies at the chromosomal ami locus by recombination. To ask directly whether the cat of pJS3 requires cotranscription with a target of integration, we used the ami locus as a test platform (Fig. 1), with cat inserted in the arniC gene in two orientations (Fig. I). Transformation

pR269 pR268

R142

,o-

arnioperon

p+ p-

C'

I

catcassette

I

'C

£~

E

F

Fig. 1. Use of an ami test platform to evaluate expression of single-copy genes. In the illustrated application of the ami test-platform strategy, targeted disruption of the amiACDEF locus was achieved by homologous exchange during genetic transformation of competent cells with a donor plasmid carrying a selectableinsert flanked by ami sequences. Elements indicated on the thin line designated R800/R142 represent the ami locus of pneumococcal strains RS00 (wt pA) and R142 (mutant pA). R142 lacks 11 bp including the -10 region of the ami promoter (p) (Alloing et al., 1990). The donor scaffoldingplasmids illustrated above the line contained the pJS3 cat cassette (in orientations indicated by arrows) within the amiC gene of a cloned segment of the ami locus. The product of homologous exchange between donor plasmid and recipient chromosome is shown below. MtxR of an amiC disruption permitted scoring of insertion-bearing RS00 recombinants independent of the expression of the inserted gene, while scoring of CmR recombinants reflected cat expression. Methods: The MboI fragment of pJS3 was ligated to BamHI-digested plasmid pR243 (pR243 differs from pR244 (Alloing et al., 1990) by the absence of the PstI fragment ending in the ApR and amiA genes). Among ErR transformants of E. coli DH1, two carrying cat in opposite orientations were retained and named pR268 and pR269, as shown. The structures of these plasmids were verified by analysis of HindllI and HaelII sites.

125 of Sp strain R800 with plasmids pR268 (ON orientation of cat) and pR269 (OFF orientation of cat) illustrated the approach, and showed the cat cassette to be inactive unless transcribed from an external promoter (Table I). While the reasons for the unsatisfactory level of autonomous expression of a single copy of the pJS3 cat are unknown, we show below that addition of a native pneumococcal promoter relieved the requirement for an active flanking promoter. Similar experiments were undertaken to evaluate the spectinomycin-resistance marker (aad9) from pDL269 (LeBlanc et al., 1991), which was reported to be suitable as a single-copy marker in streptococci (Tao et al., 1992). A B s t U I aad9 fragment of pDL269 which included both promoters described by LeBlanc et al. (1991) was inserted at the B a m H I site in the amiC gene of pR244 to make pR324 (ON) and pR325 (OFF) plasmids. With the former, pneumococcal transformants were obtained at a normal rate, but with the latter, there were none (data not shown). Thus, the pDL269 aad9 gene also depended on an external promoter for efficient expression when integrated into the pneumococcal chromosome in a single copy.

(b) Design, construction and activity of new cat and erm cassettes allowing expression of single-copy cat and erm genes in S p As previous reports of orientation-dependent expression of the pAMI31 erm gene and the circumstantial evidence described above for the pC194 cat gene in Sp suggested that transcriptional signals, rather than translational ones, were limiting in their use in Sp as single-

copy inserts, we designed new cassettes that conserved translational signals but incorporated upstream sequences from the native pneumococcal promoter, pA. The synthetic (syn) cat cassette, constructed by PCR from a portion of the pJS3 cat cassette, was cloned in the MCS of BlueScribe SK (+) (Stratagene, La Jolla, CA, USA) to form pR326 (Fig. 2). A new erm cassette was made in a similar way, combining the 5' part of pA with a 3' part of the erm promoter (pE) in M13mpl9, to produce M13eryAD (Fig. 3). The ami-targeting plasmids pR327 and pR328, carrying the syn cat cassette within amiC in either orientation, were constructed as shown in Fig. 2. These plasmids transformed R142 with equal efficiencies, and at expected frequencies (Table I). Since R142 carries a defective pA, comparison with the results for plasmids pR268 and pR269 shows that the transcriptional bias of the cat from pJS3 was relieved in the syn cassette. The new cassette required neither a specific orientation nor an active upstream promoter for effective expression. To test the utility of the syn erm cassette, an 892-bp B a m H I fragment of M13eryAD was ligated with PvulI ami fragments from pR244, after filling its ends; the mixture of products readily transformed R142 to Er R (data not shown). Both orientations of erm were recovered, in equal frequencies, as the Er R transformants were half Mtx R, half Mtx s. As non-polar amiA mutants are Mtx s (Alloing et al., 1994), but a m i C D E F mutants are Mtx R, it was expected that one orientation (OFF) of erm insertion would not affect the Mtx R of R142 (Alloing et al., 1990), while the other (ON) could reactivate the a m i C D E F genes to give the Mtx s phenotype.

TABLE I Expression of cat cassettes after insertion in amiC Transforming D N A Plasmid a

Transformants (normalized to SmR)a Source of cat cassette

cat

promoter

orientation b

Recipient (pA allele) c

Cm a

Mtx R

pR268 pR269

wt wt

pJS3 pJS3

ON OFF

R800 (+) R800 (+)

1.7 < 10 '~

2.1 1.0

pR327 pR328

syn syn

This work This work

ON OFF

R142 ( - ) R142 ( - )

cat

2.2 2.1

ND f ND

a Construction of plasmids is given in Figs. 1 and 2. b Indicates concordant (ON) or opposed (OFF) orientations of cat and disrupted target gene, as constructed in donor plasmid and expected in transformant chromosome. c Cells were transformed at approx. (2-4) × 107 cells/ml, as described in Alloing et al. (1994). Promoter alleles were wt (+) or ami142 ( - ) , d For normalization of competence levels, each recipient was transformed in parallel with plasmid D N A and with R 119 (str41) chromosomal DNA. C m R transformants were determined by plating in agar followed by challenge with an overlay of 9 m g Cm/1. The yield of ami insertions (ami = Mtx R) was determined after expression for 120 min at 37°C in solid medium, followed by challenge with an overlay of 5.4 m g Mtx/1. e No Cm R transformant were obtained at a dilution at which 10 -4 would have been detected. r Not done, since strain R142 is already fully Mtx R due to the ami142 promoter down mutation.

126 -35

-10

~A

5'gaaaatttgtTTGATTtttaatggataatgtgaTATAATggttcaacaaaegaaaattggataaagt3 ~

-10

-35 ~T

BM19

I'*''''''~''''*''''~'~''',~,~,,*,~,j~,,~`+,,,,~

-35 ~C2

780aagaatTTGAAA~ttagtttatatgt~gTAAAATgtttta.~bp~aaa~acaaaagacca~attttttaatgtg~tcttttatt~ttcaactaaa9cacccatta~tt~aa~aaa~9aaaattggataaa~tgg~atatttTTAAAAtatatattta pJS3

-I0

-35 ~CI

RSSl

-I0

RBS2

982 t9ttaca~TAATATt9acttttaaaaaag~attgattctaatgaagaaagcagacaa9taagcct~ctaaattca~tttagataaaaatttagg~ggcatat~aa[~cat Leader: MetLysLysALaAspLys***===========> <===========

BlueScribe SK (+)

PCR cassette

3'aactgcaactgccagccttgggS'BM18 l'"'"'ll'"ll'"'"',,,,,,, ,,, ,,,,,,, coding region]-3Obp-ttgacgttgagcctcggaacccttaac

pJS3

pR244

ApaI . )~hoI ~!Acc I pR327 pR328

"~pR326'la~~

Hindlll

(ON)(OFF)

Bsp M1 EcoRI Pst I

Sma I BamHl Fig. 2. Design and construction of a cat cassette for gene disruption in Sp. (Upper part) Design of syn cat cassette constructed by PCR, showing alignment of primers with the sequence of pJS3 (Ballester et al., 1990). BM 19 included 40 nt of pA sequence ( - 3 5 and - 1 0 elements shown in upper case). Three promoters preceding cat are indicated by ' - 1 0 ' and ' - 3 5 ' and are designated as: pT, the tet promoter of pLS1 (upper case), and pCl (underlined) and pC2 (upper case), promoters originating in pC194. The new cassette contained pA in place of pT and a terminator element (arrows) found between pT and pC2. Primer BM18 was complementary to a sequence downstream from the cat terminator. Primer nt matching pJS3 are indicated by a vertical line (I). RBS of the leader peptide and cat genes are overlined, pJS3 nt numbering shows distance from the unique BglI site. (Lower part) Incorporation of syn cat cassette in new vectors. Open arrows indicate a mutant amiA gene promoter and other ami genes in pR244 and the portions of lacZ in pR326. Solid arrows denote drug-resistance markers; the position of pA is shown by p and the extent of the syn cat cassette, by double lines. Methods: The PCR product amplified from pJS3 with BM18 and BM19 was cloned, after blunting with T4 polymerase, in BlueScribe SK (+) (Stratagene, La Jolla, CA, USA) at its EcoRV site, to form pR326 (3992 bp). cat orientation was verified by digestion with AseI. To make ami-scaffold vectors, the same PCR product was cloned in the amiC gene of pR244 (Alloing et al., 1990), at a similarly blunted BamHI site; two 11589-bp plasmids obtained, with cat in the ON and OFF orientations were named pR327 and pR328, respectively. Their structures were verified by digestion with NcoI.

(c) Incorporation of the syn cat cassette in a lacZ reporter plasmid suitable for insertion-duplication mutagenesis in Sp As a transcriptional reporter would be useful in Sp, we designed and constructed an insertion vector to make lacZ transcriptional fusions, pEVP3 combined the ori of pBR322, the new syn cat cassette, a derivative of the E, coli lacZ gene preceded by translational stops in all reading frames and by Gram + translation start signals, and a MCS to allow cloning of targeting DNA fragments upstream from lacZ (Fig. 4). pEVP3 was stable in E. coli DH1, conferring Cm R. Expression of a single copy of pEVP3 in pneumococcus was evaluated by inserting it into amiA at loci with mutant or wt promoters via insertion-duplication using plasmids pEVP4 and pEVP5 (Fig. 4). Cm Rtransformants were equally frequent for both loci and both orientations (data not shown), confirming that in this context, too, transcriptional target-site bias was eliminated for the syn cat cassette. Function of the pEVP3 lacZ as a reporter

was also demonstrated: pEVP5 inserts (lacZ in the ON orientation) were Cm R, despite the OFF orientation of cat (data not shown), and [3Gal activity of lysates remained at background levels after insertion in the promoterless R142 ami locus, but increased (threefold) after insertion in the wt locus of R800.

(d) Further discussion The results described above imply that the markers examined will have several useful applications in the examination of the pneumococcal chromosome. As the present results show that the pJS3 cat does not confer Cm R in silent chromosomal sites, it is probable that it is an excellent reporter gene in the Sp chromosome. The low endogenous expression of the cat, erm and aad9 genes implies that obtaining transformants following random insertion of one of these markers is strongly indicative that the target site is transcribed throughout colony growth, and indicates the orientation of uncharacterized

127 Hin

dIIl

Sphl

-35

~

...........

5,ccgggcccAAAATTTGTTTGATTTgtatcttaaaattttgtataatagg3'

281

OAMlO4 ( p $ )

[ l l 'I I"I I I" i l ' " . . . . I I " , , , , ,. ,. . . acttaagagtgtgTTGATAgtgcagtatcttaaaattttgTATAATaggaatt'48 -35 ~E

-I0

>

<...........

31ctttattaagatactcagcgacggctgacatgaaaaatgtcagccgtcaaaaacctagggg5

iSse 8387 PstI Sail YbaI 8am HI ~maI Apa I

Y%

~ DAWrOS

M 13eryAD

I"'"'"'"'"II"II,,,,,,,,,,,,,, ,,

blO-atg-726 er~

bp*aggaa~ataattctatga9tcgcttttgtaaatttggaaagttacacgtt

erm

pR243 '

"'

BamH Sma 1 - Kpn 1 ~: S a c 1 E c o R1

Pvu 1 F~pl B.dl

Fig. 3. Design and construction of a new erm cassette for Sp. (Left part) Design of a syn promoter for erm in Sp. The nt sequences of the erm region of pR243 (numbered from the unique K p n I site) and the primers used for synthesis of the new cassette are shown. DAMI04 combined the - 3 5 region of pA, including 9 upstream nt and 1 downstream nt (upper case), with the - 1 0 region and 16 upstream nt of the pAMI31 erm promoter to form a hybrid syn promoter (pS). The distal primer DAM105 was complementary to 21 nt near the end of erm and to the pJS3 cat terminator (arrows). The first and last codons of the erm gene are underlined. The native erm promoter is indicated by ' - 10', ' - 3 5 ' and pE. The nt of primers matching pR243 are indicated by a vertical line (I). (Right part) Incorporation of the erm cassette in a cloning vector. Filled arrow indicates the erm gene; the interrupted lacZ~ coding sequence is shown by open arrows, pS is shown by p. Pairs of B a m H I and SmaI sites and several unique restriction sites flanking the cassette are shown. Methods: The PCR product amplified from pR243 with primers DAM104 and DAM105 was cloned, after blunting with T4 DNA polymerase in the Sinai site of M13mpl9. The structure of the 8142-bp product in one clone, named M13eryAD, was verified by sequencing of the new junctions; however, it revealed that a 1-bp deletion (removing a G residue located 21 bp from the 3' end of the PCR cassette) had occurred in the cat terminator sequence, possibly destabilizing the potential stem-loop terminator structure. Bglll Nsil iSph I ~AspTl8

ixpm

~Sma 1 =Xbal Xhol~ BomHI

Fig. 4. Construction of insertion-duplication l a c Z reporter plasmids for Sp. Plasmid pEVP3 was constructed by replacement of the kanamycinresistance gene of pLGW300 (Van Sinderen et al., 1990) by the cat cassette of pR326, including p A (p) and the cat terminator. Open arrow denotes the promoterless lacZ reporter of pLGW300 (derived from pTV32), which is preceded by the RBS of the Bacillus subtilis spoVG gene and by stop codons in all reading frames (Perkins and Youngman, 1986). Filled arrow denotes cat gene, double lines, the extent of the syn cat cassette. Unique restriction sites between cat and lacZ are shown. Open box, ori from pBR322. Methods: The cat cassette obtained on a 1073-bp S m a I - X h o I fragment from pR326 was ligated to the 5227-bp X h o I - B s p H I fragment of pLGW300; a 6302-bp product was named pEVP3, after verification of its structure by analysis of 26 restriction sites. Derivative amiA-targeting plasmids contained internal fragments o f a m i A . To produce pEVP4, a 1013-bp SpeI-FspI fragment from pR327 was cloned between the Sinai and X b a I sites of pEVP3, pEVP5 was constructed by cloning a 742-bp S p h I - S p e l fragment from pR327 between the SphI and X b a I sites of pEVP3. The structure of pEVP4 was verified by restriction with D r a I + P s t I , and that of pEVP5 by restriction with PstI + H i n d l l I .

ORFs. Indeed, erm has already been successfully used to deduce the orientation of several genes. A significant background level of 13Gal activity in R800

suggests the presence of an endogenous enzyme in this strain: for detection of moderate or low expression, selection of a strain with lower endogenous enzyme levels would improve the sensitivity of this l a c Z reporter. (e) Conclusions (I) The ami locus provides a convenient test platform for evaluating expression of single-copy genes in Sp. (2) The activity of the tested cat. erm, and aad9 genes depends on the site of insertion in the chromosome; in silent sites, they do not afford drug resistance. (3) New erm and cat gene cassettes incorporating parts of the pneumococcal ami promoter are active even at silent sites of insertion. (4) A derivative of lacZ from pTV32 acts as a transcriptional reporter in Sp. (5) New tools for use in analysis of pneumococcus resulting from this work include two cassettes with improved single-copy expression (erm and cat), convenient sources of these cassettes in plasmids of known sequence (pR326 and M13eryAD), a lacZ reporter insertion vector (pEVP3) and lacZ reporter derivatives (pEVP4, pEVP5) which can be used to assess the activity of promoters in a constant ami context.

ACKNOWLEDGEMENTS

This work was supported in part by the US National Science Foundation (MCB-9106847), by an award (to D.A.M.) from the Philippe Foundation for Franco-

128

American Research Exchange, and by a short-term Poste Rouge (to D.A.M.) from the CNRS. We thank D. Van Sinderen for a sample of pLGW300.

REFERENCES Alloing, G., Trombe, M.C. and Claverys, J.P.: The ami locus of the gram-positive bacterium Streptococcus pneumoniae is similar to binding protein-dependent transport operons of Gram-negative bacteria. Mol. Microbiol. 4 (1990) 633-644. Alloing, G., de Philip, P. and Claverys, J.P.: Three highly homologous membrane-bound lipoproteins participate in oligopeptide transport by the Ami system of the gram-positive Streptococcus pneumoniae. J. Mol. Biol. 241 (1994)44-58. Ballester, S., Lopez, P., Alonso, J.C., Espinosa, M. and Lacks, S.A.: Selective advantage of deletions enhancing chloramphenicol acetyltransferase gene expression in Streptococcus pneumoniae plasmids. Gene 41 (1986) 153-163. Ballester, S., Alonso, J.C., Lopez, P. and Espinosa, M.: Comparative expression of the pC194 cat gene in Streptococcus pneumoniae, Bacillus subtilis, and Escherichia coll. Gene 86 (1990) 71-79. Horinouchi, S. and Weisblum. B.: Nucleotide sequence and functional map of pC194, a plasmid that specifies inducible chloramphenicol resistance. J. Bacteriol. 150 (1982) 815-825. Lacks, S.A. and Greenberg, B.: Sequential cloning by a vector walking along the chromosome. Gene 104 (1991) 11-17.

LeBlanc, D.J., Lee, L.N. and Inamine, J.M.: Cloning and nucleotide base sequence analysis of a spectinomycin adenyltransferase AAD(9) determinant from Enterococcus faecalis. Antimicrob. Agents Chemother. 35 (1991) 1804-1810. Martin, B., Alloing, G., Mejean, V. and Claverys, J.P.: Constitutive expression of erythromycin resistance mediated by the ermAM determinant of plasmid pAMI31 results from deletion of 5' leader peptide sequences. Plasmid 18 (1987) 250-253. Paton, J.C., Andrew, P.W., Boulnois, G.J. and Mitchell, T.J.: Molecular analysis of the pathogenicity of Streptococcus pneumoniae: the role of pneumococcal proteins. Annu. Rev. Microbiol. 47 (1993) 89-115. Perkins, J.B. and Youngman, P.J.: Construction and properties of Tn917-1ac, a transposon derivative that mediates transcriptional gene fusions in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 83 (1986) 140-144. Tao, L., LeBlanc, D.J. and Ferretti, J.J.: Novel streptococcal-integration shuttle vectors for gene cloning and inactivation. Gene 120 (1992) 105-110. Van Sinderen, D., Withoff, S., Boels, H. and Venema, G.: Isolation and characterization of coraL, a transcription unit involved in competence development of Bacillus subtilis. Mol. Gen. Genet. 224 (1990) 396-404. Vasseghi, H. and Claverys, J.P.: Amplification of a chimeric plasmid carrying an erythromycin-resistance determinant introduced into the chromosome of Streptococcus pneumoniae. Gene 21 (1983) 285-292. Vasseghi, H., Claverys, J.P. and Sicard, A.M.: Mechanism of integrating foreign DNA during transformation of Streptococcus pneumoniae. In: Polsinelli, M. (Ed.), Transformation - 1980. Cotswold Press, Oxford, 1981, pp. 137-154.