Vol. 175, No. 17

JOURNAL OF BACTERIOLOGY, Sept. 1993, p. 5529-5538

0021-9193/93/175529-10$02.00/0 Copyright © 1993, American Society for Microbiology

Transfer Functions of the Conjugative Integrating Element pSAM2 from Streptomyces ambofaciens: Characterization of a kil-kor System Associated with Transfer JULIET1FE HAGEGE,t JEAN-LUC PERNODET,* GUENNADI SEZONOV,4 CLAUDE GERBAUD, ANNICK FRIEDMANN, AND MICHEL GUERINEAU Laboratoire de Biologie et Gene'tique Moleculaire, Institut de Genetique et Microbiologie, Unite de Recherche Associe'e 1354, Centre National de la Recherche Scientifique, Bdtiment 400, Universite Paris-Sud" F-91405 Orsay Cedex, France Received 11 March 1993/Accepted 22 June 1993

pSAM2 is an 11-kb integrating element from Streptomyces ambofaciens. During matings, pSAM2 can be transferred at high frequency, forming pocks, which are zones of growth inhibition of the recipient strain. The nucleotide sequences of the regions involved in pSAM2 transfer, pock formation, and maintenance have been determined. Seven putative open reading frames with the codon usage typical of Streptomyces genes have been identified: traSA (306 amino acids [aal), or,f4 (84 aa), spd4 (224 aa), spdB (58 aa), spdC (51 aa), spdD (104 aa), and korSA (259 aa). traSA is essential for pSAM2 intermycelial transfer and pock formation. It could encode a protein with similarities to the major transfer protein, Tra, of pUlOl. TraSA protein contains a possible nucleotide-binding sequence and a transmembrane segment. spd4, spdB, spdC, and spdD influence pock size and transfer efficiency and may be required for intramycelial transfer. Akil-kor system similar to that of pUlOl is associated with pSAM2 transfer: the korSA (kil-override) gene product could control the expression of the traS4 gene, which has lethal effects when unregulated (Kil phenotype). The KorSA protein resembles KorA of pUlOl and repressor proteins belonging to the GntR family. Thus, the integrating element pSAM2 possesses for transfer general features of nonintegrating Streptomyces plasmids: different genes are involved in the different steps of the intermycelial and intramycelial transfer, and a kil-kor system is associated with transfer. However, some differences in the functional properties, organization, and sizes of the transfer genes compared with those of other Streptomyces plasmids have been found.

Streptomyces spp. are gram-positive mycelial soil bacteria that differ from other bacterial groups by the high G+C content of their DNA, by their complex life cycle, and by producing a large variety of antibiotics and other useful secondary metabolites. Research on Streptomyces spp. has identified a large number of plasmids (reviewed by Hopwood et al. [19]), including integrating plasmids capable of site-specific chromosomal integration (37). Most of the Streptomyces plasmids are self-transmissible and can mobilize chromosomal markers. Transfer regions have been localized on several of them (pIJ101 [27-29], pSN22 [24, 25], SCP2 [4], SLP1 [37], and pSAM2 [46]) but sequenced only in the case of some nonintegrating Streptomyces plasmids (pIJ101 [28] and pSN22 [23-25]). Very few genes seem to be involved in plasmid transfer in Streptomyces spp. Plasmid transfer in Streptomyces spp. is almost always associated with pock formation (lethal zygosis) (19). Pocks are zones of growth inhibition of the recipient strain. It has been suggested that pock formation is a manifestation of intramycelial transfer of plasmid (spreading) within the mycelium of the recipient culture after primary transfer (intermycelial) from the donor. A kil-kor system has been implicated in the transfer of some plasmids in Streptomyces spp. (27): certain genes (kil genes) specify functions lethal to either the host or the plasmid, *

Corresponding author.

t Present address: Department of Genetics, Stanford University

School of Medicine, Stanford, CA 94305. I: Permanent address: Institute of Genetics and Selection of Industrial Microorganisms, 113545 Moscow, Russia.

while others (kor genes, for kil override) encode products that control expression of the Kil phenotype. In pIJ101, the most studied Streptomyces plasmid, only one gene, called tra, is essential for conjugative transfer (intermycelial) and mobilization of chromosomal markers (28, 29). Two genes, spdA and spdB, are responsible for the spreading within the recipient mycelium (28). A kil-kor system (kiUA-korA and kilB-korB) associated with the transfer of pIJ101 has been identified (27). Most integrating plasmids of Streptomyces spp., including SLP1 (37), pIJ408 (47), pIJ110 (18), pSE211 (7, 26), pMEA100 (33), and pSAM2 (40), are also self-transmissible. However, little is known about the genes involved in their transfer. We are interested in studying the transfer functions of integrating plasmids in Streptomyces spp. and the mechanism leading to their transfer. pSAM2 is an 11-kb element originally isolated in Streptomyces ambofaciens, the producer of the macrolide antibiotic spiramycin (40). It can be maintained in two different forms. In the wild-type strain S. ambofaciens ATCC 23877 (strain B1) and in a survivor of exposure to UV light, S. ambofaciens ATCC 15154 (strain B2), pSAM2 is found integrated in the chromosome. In another survivor of UV treatment, S. ambofaciens JI3212 (strain B3), pSAM2 is found simultaneously in the integrated state and as an autonomous plasmid at about 10 copies per chromosome (40). The pSAM2 element can replicate, is self-transmissible, elicits the lethal zygosis reaction (pock formation) (40), and mobilizes chromosomal markers (46). Both the free form of pSAM2 and the integrated sequences of strains B1 and B2 can be transferred very efficiently. pSAM2 also has a site-specific recombina5529

HAG2GE ET AL.

5530

J. BAcT1ERIOL. TABLE 1. Bacterial strains and plasmids

Strain or plasmid

Description

Source or reference

S. ambofaciens J13212 (B3)

UV derivative of S. ambofaciens ATCC 23877; contains pSAM2 in a free and in an integrated form

John Innes collection 40

pro-2 str-6 pro-2 str-6; deficient in intraplasmid recombination

20 54

pBR322 with a 1.8-kb BamHI fragment containing the thiostrepton resistance gene (tsr) pIJ39 inserted in EcoRI(21) of pSAM2 fQ interposon in BstEII(4) of pTS39 Deletion SacI(12)-BglII(15) and Q interposon in BglII(25) of pTS39 Deletion KpnI(9)-KpnI(17) of pTS39 Klenow filing in in BstEII(4) site 0.5-kb deletion from BstEII(4) of pTS39 1.0-kb deletion from BstEII(4) of pTS39 2.5-kb deletion from BstEII(4) of pTS39 5.0-kb deletion from BstEII(4) of pTS39

53

S. lividans TK64 JT46 Plasmids pIJ39

pTS39 pTS68 pTS105 pTS89 pTS59A pD6A pDlO pD19 pD8

tion system very similar to that of temperate phages (6). It consists of an attachment site, attP; an int gene, whose product promotes site-specific integration of pSAM2 by recombination with the chromosomal attachment site (atfB); and axis gene, whose product could be involved in excision of pSAM2. pSAM2 regions involved in transfer, pock formation, and plasmid maintenance have been localized on a functional map (46). We report here the nucleotide sequences of these regions of pSAM2. We have identified the genes involved in intermycelial and intramycelial transfer. We have also shown that a ki-kor system, similar to that of pIJlOl, is associated with pSAM2 transfer, and we have identified the genes involved in this system.

MATERIALS AND METHODS Bacterial strains, plasmids, culture conditions, and transformation. The Streptomyces strains and plasmids used are listed in Table 1. Streptomyces lividans 66 derivatives (20) are the commonly used host strains for cloning experiments. Escherichia coli JM101 and JM109 (58) were used as hosts for the M13 vectors, and E. coli DH5a (obtained from BRL) was used as host for the pUC19 vector (58). General culture and genetic techniques for Streptomyces spp. were as described by Hopwood et al. (17). Streptomyces transformants carrying the thiostrepton resistance (tsr) gene (53), which also confers resistance to nosiheptide, were selected by using 50 ,g of nosiheptide per ml. The omega (Q) interposon containing the aadA gene conferring streptomycin and spectinomycin resistance was obtained from plasmid pHP45Q1 (41). DNA isolation and manipulation. Plasmid DNA was isolated from E. coli by the clear-lysate method (42) and from Streptomyces spp. by alkaline lysis (17). M13mpl8 and M13mpl9 phage DNAs carrying inserts as well as singlestranded DNA sequencing templates were prepared as described by Messing (35). DNA for double-strand sequencing was isolated as described by Sambrook et al. (42), using a QIAGEN kit. Genetic crosses. The abilities of different pSAM2 derivatives to yield pocks and the plasmid transfer efficiencies were determined as described previously (46). DNA sequencing. For DNA sequencing, we used the dideoxy method of Sanger et al. (43), as modified by Biggin

46 46 46 46 This paper 46 46 46 46

et al. (5). Universal M13 primers, -40 primer, or synthetic

oligonucleotides synthesized on an Applied Biosystems model 381A DNA synthesizer were used. We used sequencing kits obtained from Pharmacia LKB (either the T7 sequencing kit or, to overcome the problems of band compressions in sequencing gels due to the high stability of G+C-rich structures, Deaza T7 sequencing mixes and the Gene-ATAQ sequencing kit). cx-35S-dATP was obtained from Amersham. Sequences were determined for both strands. Computer-assisted sequence analysis. DNA and protein sequences were analyzed by using the following programs: PC/GENE package (IntelliGenetics, Inc., Mountain View, Calif.), PCFOLD (60), and DNA Strider (34). Identification of open reading frames (ORFs) was done by the methods (included in the PC/GENE package) of Fickett (12) and Kolaskar and Reddy (31). A program similar to the FRAME program developed by Bibb et al. (3) was also used for identification of ORFs. The FRAME program is based on the biased codon usage that occurs in organisms having a high G+C content. Predicted proteins were studied and analyzed for the presence of hydrophobic domains, predicted to be transmembrane segments (30), and for transmembrane alphahelices (2, 9, 10) in order to determine if these proteins could be associated with membranes. The hydropathy of protein sequences was calculated by the algorithm of Kyte and Doolittle (32) with a window of nine amino acids (aa). The predicted proteins were also scanned for similarities to the Swiss-Prot data bank version 22 by using the FASTA program (39). The TFASTA program was used to compare the sequences of the different predicted proteins with the DNA sequences present in the EMBL data base (version 31) by translating the DNA sequences to protein in all six frames. The predicted proteins were subjected to searches for the presence of DNA-binding domains (helix-turn-helix, Znbinding domains, and leucine-zipper coiled coil) (15) and for the presence of nucleotide-binding fold domains (55). Nucleotide sequence accession numbers. The EMBL accession number of the sequence of pSAM2 regions involved in transfer and pock formation is Z19593, that of the sequence of pSAM2 regions involved in maintenance is Z19589.

pSAM2 TRANSFER AND kil-kor FUNCTIONS

VOL. 175, 1993

5531

Transfer and pock formation

1.85

(7) 2.37

NruI (8) 3.07 Bgl II (25) 7.77 BamHI

(24)

7.35

~

ttr

KpnI (9) 3.30

,, _

A / /

(10) 3.62 ~~~~~~~~~~~~~~NruI (0

BamHI (11) 3.67

BamHI (23) 7.27

SacI(12)4.31

SmaI (22) 6.93

SacI (14) 4.68 BglII (15) 5.11

EcoRI (21) 6.66 BamHI (20) 6.26

Plasmidmaintenance and free) (integrated

BamHI(16)5.26

EcoRV (19) 5.58

5.29

BamHI (18) 5.41 FIG. 1. Map of pSAM2 showing the regions involved in transfer, pock formation, and plasmid maintenance and major ORFs identified in these regions. Only the restriction sites mentioned in the text are shown. The locations of the attachment site attP and the site-specific recombination genes int and xis are also shown. The numbers in parentheses refer to the numbering of the restriction sites. Other numbers indicate the positions (in kilobases) of the restriction sites in the sequence of the whole plasmid.

RESULTS Nucleotide sequences of the regions involved in pSAM2 transfer, pock formation, and plasmid maintenance. Previously, it was shown that the SmaI(2)-NnrI(8) region of pSAM2 (Fig. 1) is involved in conjugal transfer and pock formation (46). It was also shown that the BamHI(11)SacI(14) region of pSAM2 is involved in its maintenance (either integrated or free) (Fig. 1); deletion of this region is lethal for the plasmid or for the host, but if the BstEII(4)NruI(8) region is deleted simultaneously, Streptomyces transformants can be obtained. Between these two regions, a region involved in pSAM2 maintenance as a free molecule is present (46). To refine the genetic analysis, the DNA sequences of the regions involved in transfer, pock formation, and plasmid maintenance were determined. DNA of the shuttle plasmid pTS39 (composed of pSAM2, pBR322, and the thiostrepton resistance gene [Table 1]) was digested with various restriction enzymes, and the different fragments obtained were cloned by insertion into the appro-

priate sites of M13mpl8, M13mpl9, or pUC19 and sequenced. The sequence of the SmaI(2)-NruI(8) region involved in pSAM2 transfer and pock formation is given in Fig. 2a, and that of the BamHI(11)-SacI(14) region involved in plasmid maintenance is given in Fig. 2b. Analysis of the nucleotide sequences of the transfer, pock formation, and plasmid maintenance regions of pSAM2. The G+C content of the sequenced regions involved in pSAM2 transfer and plasmid maintenance is 71.6 and 66.5%, respectively, which is a little lower than the average Streptomyces G+C content of 73% (11). The nucleotide sequences were analyzed for the presence of ORFs by three different methods, as described in Materials and Methods. Because of the high G+C content of Streptomyces DNA, codon usage tends to be severely biased (3). In the region involved in pSAM2 transfer and pock formation (Fig. 1, 2a, and 3), we identified six ORFs with codon biases characteristic of Streptomyces spp. They were

5532

HAGEGE ET AL. 10

J. BACTERIOL.

20

30

40

50

60

a I.R. D.R.1 so. 1 TGACGTGGTTCATGGTCCr G~ GGTCGTCGCTGCTGCGGGTCTdCTGCGGTGGC _

R A Q L T G R T A H R V N D E T S A N M 1021 CCGTGCCCAGCTCACCGGCCGTACCGCCCACCGCGTCAACGACGAAACGTCCGCGAACAT

orf84

NruI (8) 61 GGCGCCCCGCCTGGTACTGGCTCACCTTCGGGGCCCTGGTCGCGACGGTGCGGGTCCTGG

CCGCGGGGCGGACCATGACCGAGTGGAAGCCCCGGGACCAGCGCTGCCACGCCCAGGACC

S

A

V

T

fM S P D A V L A A I Q I P T D T C H P M P S W L P S R S P P T H

1081 GGCTTCGGTGACGTGTCACCCGATGCCGTCCTGGCTGCCATCCAGATCCCCACCGACACA

121 TCCGTACGCCTCGGTCATGGAAGCGTGCGGGCTGACGGTCCGCCCTCACGCTGGCGGCTG

CCGAAGCCACTGCACAGTGGGCTACGGCAGGACCGACGGTAGGTCTAGGGGTGGCTGTGT

AGGCATGCGGAGCCAGTACCTTCGCACGCCCGACTGCCAGGCGGGAGTGCGACCGCCGAC G

P

181 CTCTGGCCCGGATGGCGAATGCCGCGCCTGAGTCCCGGCCGCCGCGCATCTTGCGGTTAC

V

A

P

A

S

V

G

T

S

L

P

D

T

S

T

V

R

G

Q

G A

A

W A R I R A P H G P V S A P R T

GAGACCGGGCCTACCGCTTACGGCGCGGACTCAGGGCCGGCGGCGCGTAGAACGCCAATG

1141 CCCGGCGTCGCTGTCACCGGTGACTCGACAGGCGGCTGGGCCCGTATCCGCGCCCCGCAC

S.D.? 241 GTCCCACTCGTACCGGCCTGGTCTGCGGCTCAAGCTCCGGCCGGGACAGGATGCCTTCGA

GGGCCGCAGCGACAGTGGCCACTGAGCTGTCCGCCGACCCGGGCATAGGCGCGGGGCGTG

CAGGGTGAGCATGGCCGGACCAGACGCCGAGTTCGAGGCCGGCCCTGTCCTACGGAAGCT traSA fM A

D.R.2

A

T

T

D

301 CGTGGCCGCCA

R L d

R

H

S

F

G

V

Y

G

V

T

S

T

T L R E A V N I C N K H A D R T P D V P R C A R P-D.R.3 1201 ACCACGCTGCGCGAGGCCGTGAACATCTGC AP RCCGCACACCTGATGTG

R

TGGTGCGACGCGCTCCGGCACTTGTAGACGTTGTTCGTGCGGCTGGCGTGTGGACTACAC

CCACTCCTTCGGCGTCTACGGCGTCACCTCCCG P

A

L

A

A

F

R

P

A

L

A

P

L

Q A

P

SmaI (6) R_V P L

1261 CCCGCGCTGGCCGCCTTCCGGCCTGCCCTCGCCCCGCTGCCCCAGGCCCGGGTGCCGCTG E L R S G V V E V R M T G Y D V L Q R V 361 TGAACTGCGCTCAGGTGTGGTCGAGGTGCGGATGACCGGCTACGACGTCCTCCAGCGGGT

ACTTGACGCGAGTCCACACCAGCTCCACGCCTACTGGCCGATGCTGCAGGAGGTCGCCCA

GGGCGCGACCGGCGGAAGGCCGGACGGGAGCGGGGCGACGGGGTCCGGGCCCACGGCGAC S K T A P A T A --I.R. _ N 1321 TCCAAGACAGCCCCCGCCACCGCCTGACCTACCCGCTCCTCGGTCGGCGCGACCACCCTT

Q M P A T A E T R P M R I P V A L R E D 421 GCAGATGCCCGCCACAGCCGAGACGCGGCCGATGCGCATCCCTGTGGCGCTGCGGGAGGA

AGGTTCTGTCGGGGGCGGTGGCGGACTGGATGGGCGAGGAGCCAGCCGCGCTGGTGGGAA

spdA

CGTCTACGGGCGGTGTCGGCTCTGCGCCGGCTACGCGTAGGGACACCGCGACGCCCTCCT

S.D.? 1381

G

A

V

H

Y

R

D

Y

R

A

V

P

H

G

L

T

L

G

A

-

-

fM A --

--

R

P

---

T

481 CGGTGCGGTGCACTACCGCGACTACCGTGCCGTCCCGCACGGTCTGACCCTGGGGGCCAC

GCCACGCCACGTGATGGCGCTGATGGCACGGCAGGGCGTGCCAGACTGGGACCCCCGGTG Domain A E

541

S

G K S.

V

Y

Q

R

N

L

V

A

G

L

A

P

H

H

V

GGAGTCCGGGAAGTCCGTCTACCAGCGCAATCTGGTCGCCGGGCTCGCTCCGCACCATGT CCTCAGGCCCTTCAGGCAGATGGTCGCGTTAGACCAGCGGCCCGAGCGAGGCGTGGTACA Domain B A

L

G I D C K Q G V E L F P L A R R F 601 TGCCCTGGTCGGCATCGACTGCAAGCAAGGCGTGGAACTGTTCCCGCTGGCCCGCCGGTT

V

S A L A D N P D T A L D L L E A L V G H 661 CTCCGCGCTCGCCGACAACCCCGACACCGCCCTCGATCTCCTCGAAGCGCTCGTCGGACA

GAGGCGCGAGCGGCTGTTGGGGCTGTGGCGGGAGCTAGAGGAGCTTCGCGAGCAGCCTGT M

E

D

V

Y

Q

L

I

R

A

E

Q

R

I

S

V

A

V

P

D

721 CATGGAGGACGTCTACCAGCTCATCCGGGCCGAGCAGCGCATCAGCGTCGCCGTGCCGGA

GTACCTCCTGCAGATGGTCGAGTAGGCCCGGCTCGTCGCGTAGTCGCAGCGGCACGGCCT EcoRV (7) Transmembrane segment A

E

I

A

A

D

I

W

D

L

R

E

D

L

R

P

V

P

V

V

781 TGCGGAGATCGCCGCCGATATCTGGGACCTGCGCGAGGACCTGCGGCCAGTGCCGGTCGT

ACGCCTCTAGCGGCGGCTATAGACCCTGGACGCGCTCCTGGACGCCGGTCACGGCCAGCA V 841

R 901

V

D

E

V

A

E L

A

L

F

A

T

K

D

E

E

K

R

D

R

I

I

T

A

L

V

R

L

A

Q

L

G

R

A

A

G

I

CCGCGACCGCATCATCACTGCCTTGGTCCGCCTCGCCCAGCTCGGCCGCGCCGCTGGCAT GGCGCTGGCGTAGTAGTGACGGAACCAGGCGGAGCGGGTCGAGCCGGCGCGGCGACCGTA start of pDl9 Deletion I C G Q F G S

E L G K G I T M L RV CTACCTCGAAATCTGCGGGCAGCGCTTCGGCTCCGAACTCGGCAAGGGAATCACCATGCT Y

961

L

GGTCCTGGTCGACGAGGTCGCCGAACTCGCTCTCTTCGCCACCAAGGACGAAGAGAAGCG

L

E

A L R I D A V L V O A V I A G A L S F A 1441 CGCTCTCCGCATCGACGCCGTGCTCGTCCAGGCCGTCATCGCCGGGGCACTGTCCTTCGC

GCGAGAGGCGTAGCTGCGGCACGAGCAGGTCCGGCAGTAGCGGCCCCGTGACAGGAAGCG membrane associated helices H L H D L A A A A G Q D G W K A W A Y P 1501 CCACCTTCACGACCTGGCCGCCGCTGCCGGACAGGACGGCTGGAAAGCCTGGGCCTACCC

GGTGGAAGTGCTGGACCGGCGGCGACGGCCTGTCCTGCCGACCTTTCGGACCCGGATGGG V S V D L L L V A A W R R L R T D G P S 1561 CGTCTCGGTCGACCTGCTCCTGGTCGCAGCCTGGCGCCGACTGCGCACCGACGGACCGTC

GCAGAGCCAGCTGGACGAGGACCAGCGTCGGACCGCGGCTGACGCGTGGCTGCCTGGCAG membrane associated helices R L A W S W F V I A L V A S L G A N V A 1621 CCGGCTGGCCTGGTCCTGGTTCGTCATCGCCCTGGTCGCCTCGCTCGGCGCCAACGTCGC

GGCCGACCGGACCAGGACCAAGCAGTAGCGGGACCAGCGGAGCGAGCCGCGGTTGCAGCG T A G L L D L N D V P A W L R I L V A A 1681 CACCGCCGGACTCCTCGACCTGAACGACGTCCCGGCCTGGCTTCGCATCCTCGTCGCCGC

GTGGCGGCCTGAGGAGCTGGACTTGCTGCAGGGCCGGACCGAAGCGTAGGAGCAGCGGCG membrane associated helices W P A L A F M G G T L L A H T A T H H E 1741 CTGGCCCGCCCTGGCCTTCATGGGCGGCACCCTCCTCGCCCACACCGCCACCCACCACGA

GACCGGGCGGGACCGGAAGTACCCGCCGTGGGAGGAGCGGGTGTGGCGGTGGGTGGTGCT P E A P A P T Q P A P E P P A F T D E H 1801 GCCGGAAGCGCCGGCACCCACCCAACCGGCGCCCGAACCCCCGGCCTTCACGGACGAACA

CGGCCTTCGCGGCCGTGGGTGGGTTGGCCGCGGGCTTGGGGGCCGGAAGTGCCTGCTTGT D L V R V D D T E E P P E L P A P G L O 1861 CGACCTCGTGCGCGTGGACGACACCGAAGAACCCCCGGAACTCCCCGCACCCGGACTCCA GCTGGAGCACGCGCACCTGCATGTGGCTTCTTGGGGGccTT GAGaC,r,rrr-T.arTnAnaT

GATGGAGCTTTAGACGCCCGTCGCGAAGCCGAGGCTTGAGCCGTTCCCTTAGTGGTACGA

FIG. 2. Nucleotide sequences of pSAM2 regions involved in transfer and pock formation (a) and plasmid maintenance (b). The numbers on the left indicate nucleotide positions. Major restriction sites are shown; the accompanying numbers in parentheses refer to the numbers in the restriction map of pSAM2 (Fig. 1). The predicted amino acid sequences encoded by ORFs are shown below or above the nucleotide sequence, depending on the coding strand (see Results). fM, first translated amino acid for both AUG and GUG codons; S.D., potential Shine-Dalgarno region preceding the initiation codon with base complementary to the S. lividans and S. ambofaciens 16S rRNAs. The most significant direct repeats (D.R.) are boxed, and the inverted repeats (I.R.) are indicated by facing arrows. D.R.1 and D.R.2 are direct repeats present upstream of the korSA and traSA genes. Amino acids of possible transmembrane segments, membrane helices, or helices from possible helix-turn-helix motifs are underlined. The start of the pD19 (Table 1) deletion is shown.

named, for reasons given below, traSA, orf84, spdA, spdB, spdC, and spdD. In the region involved in pSAM2 maintenance, one ORF, named korSA for reasons given below, is present (Fig. 1, 2b, and 4). All the predicted ORFs are transcribed in the same direction. Strohl (51) has analyzed the putative Shine-Dalgarno sequences of 44 Streptomyces genes and has identified the conserved sequence to be (a/g)-G-G-A-G-G. In most of the ORFs identified, one such sequence was present near the predicted initiation codon

(Fig. 2). The characteristics of each ORF are summarized in Table 2. Functional and sequence analysis of the traSA gene and traS4-encoded protein: involvement in pSAM2 intermycelial transfer. Previous studies (46) had shown that pSAM2 derivatives carrying deletions in the SmaI(6)-NruI(8) region (Fig. 1) could not form pocks and were highly affected in their transfer efficiencies. In this region, we identified one large ORF extending over 918 nucleotides, from nucleotide

pSAM2 TRANSFER AND kil-kor FUNCTIONS

VOL. 175, 1993 membrane associated helices

10

P A P P A V P A P P V P A A L I_ D H 1921 GCAGGCCCCGGCTCCGCCCGCGGTTCCCGCCCCGCCGGTCCCGGCTGCACTGA.,TCGACCA CGTCCGGGGCCGAGGCGGGCGCCAAGGGCGGGGCGGCCAGGGCCGACGTGACT. 'AGCTGGT

20

40

30

5533

50

60

A

0

A R K V A A D H E H R T G S P I D T D T 1981 CGCCCGCAAGGTCGCCGCCGACCACGAACACCGCACCGGCAGCCCTATCGACA *CCGACAC GCGGGCGTTCCAGCGGCGGCTGGTGCTTGTGGCGTGGCCGTCGGGATAGCTGT 'GGCTGTG L R T R L G V P P Q L A D A I A A Q L A 2041 ACTCCGCACCCGCCTCGGCGTCCCGCCCCAGCTCGCCGACGCCATCGCCGCCC. :AGCTCGC TGAGGCGTGGGCGGAGCCGCAGGGCGGGGTCGAGCGGCTGCGGTAGCGGCGGG'ITCGAGCG

epdB --fM P A R D H F H S V M R S.D.? 2101 CTGACCCGAAAGGAGAACGACGCCTCATGCCTGCCCGCGACCACTTCCACTCC :GTGATGC GACTGGGCTTTCCTCTTGCTGCGGAGTACGGACGGGCGCTGGTGAAGGTGAGG CACTACG Sau3A (5) D.R.3 I G T H R D R H G R T K H A A V C T N D 2161 GGATCGGCACCCACCGCGACCGCCACGGCCGCAC g iCGTGIGC LACC;AACG CCTAGCCGTGGGTGGCGCTGGCGGTGCCGGCGTGGTTCGTGCGGCGGCACACGI ITGGTTGC

R C G W S A D Y T A Q S A A Q L A A R T 2221 ACCGCTGCGGCTGGTCCGCCGACTACACCGCCCAGTCCGCCGCACAGCTCGCC :GCCCGCA TGGCGACGCCGACCAGGCGGCTGATGTGGCGGGTCAGGCGGCGTGTCGAGCGG CGGGCGT

spdC H

R

C

K

V

S

--- S.D.?

fM D

V

P

L

W

F

SacI (14) 1 TCTTGTAGACGCCGGTGAGCTCGTCGACCTCGACGCGGATGCCTGTCTCTTCCCAGACTT AGAACATCTGCGGCCACTCGAGCAGCTGGAGCTGCGCCTACGGACAGAGAAGGGTCTGAA

Sacl (13) 61 CGCGGGCTACGCCGGTCTCCGGGGTCTCGTCCAGTTCGAGTACACCGCCGGGGAGCTCCC GCGCCCGATGCGGCCAGAGGCCCCAGAGCAGGTCAAGCTCATGTGGCGGCCCCTCGAGGG D.R. 2 121 AGGTGCCGTTGTCGGCTCGGCGGATCGCCAGGAGACGCCCGTCTTCGC ACCACC. .w

D.R. 1

181

V

G

V

L

G

V

K

L

I

R

P

P

W

W

L

241

CCTCGTATCCTCCGTCTCTTCCTGATACCCTTGGTGCCATCTTCCCCCGTCCAGCCCCGG R Y V Q I A D E I V Q Q I R A G V L K P 301 TCGATACGTGCAGATCGCGGACGAGATCGTGCAGCAGATCCGGGCTGGTGTCCTCAAGCC

AGCTATGCACGTCTAGCGCCTGCTCTAGCACGTCGTCTAGGCCCGACCACAGGAGTTCGG G

I .

2341 TCTGCGTCGGAGTCCTCGGCGTCAAGCTCATCCGCCCGCCCTGGTGGCTCATC 'GCCGTTC AGACGCAGCCTCAGGAGCCGCAGTTCGAGTAGGCGGGCGGGACCACCGAGTAG CGGCAAG membrane associated helices Y L L A_D S L L A P V I D T A V L L G 2401 TCCTCCTCGGCGGCTACCTCCTCGCCGACAGCCTCCTGGCCCCGGTCATCGAC. :ACCGCCG AGGAGGAGCCGCCGATGGAGGAGCGGCTGTCGGAGGACCGGGGCCAGTAGCTG ITGGCGGC

1korSA

S.D.?fM G T T V E G G R S G P GGAGCATAGICAGAGAAGaCTATGGGAACCACGGTAGAAGGGGGCAGGTCGGGGCC S.D.?

2281 CCCACCGCTGCAAGGTCAGCTIGAGG:CCCCCGTGGACGTCCCGCTCTGGTTC :GCCCTGC GGGTGGCGACGTTCCAGTCGATCCTCCGGGGGCACCTGCAGGGCGAGACCAAG ;CGGGACG C

&GCAACGGATACGGAGTGCAGCGGCGGCGACGTGGCTTdt(i£:.TTshCTCATGTACA GTCGTTGCCTATGCCTCACGTCGCCGCCGCTGCACCGAAGGACACCAACTGAGTACATGT

D

M

V

P

S

E

Helix S E

SacI (12) L V D

Helix

R

Y

fz V i G G I

361

CGGCGACATGGTGCCAAGTGAATCGGAGCTCGTCGATCGCTACGGCGTGTCCGGCGGCAC GCCGCTGTACCACGGTTCACTTAGCCTCGAGCAGCTAGCGATGCCGCACAGGCCGCCGTG

421

AATCCGCAAGGCCATGGTCGAGGTGCGAGCGAGCGGACTCGTCGAGACCCGGCACGGCAA TTAGGCGTTCCGGTACCAGCTCCACGCTCGCTCGCCTGAGCAGCTCTGGGCCGTGCCGTT

I R

K

A

M

V

E

V

R

A

S

G

L

V

E

T

R

H

G

K

spdD --- S.D..?

fM F R P K H P T M P Q P T G 2461 TCAAGTAAGGATATCCGCCCATGTTCCGGCCCAAGCACCCGACCATGCCCCAG CCCACCG AGTTCATTCCTCTAGGCGGGTACAAGGCCGGGTTCGTGGGCTGGTACGGGGTC :GGGTGGC BstEII (4) K

G S I V K D R P P V R H R S S D R F R R 481 AGGCTCGATCGTGAAGGACCGGCCGCCGGTACGGCACCGCTCCTCCGACCGCTTCCGGCG

TCCGAGCTAGCACTTCCTGGCCGGCGGCCATGCCGTGGCGAGGAGGCTGGCGAAGGCCGC

T1 P A P 2521 GCACGGTCACCCCGCCCGCCGTCGTCGAGCCGACGACCATCACGCCCGGCACC CCGGCCC

S L R Q G G K A A Y L A E S A Q S G A T 541 CTCGCTCCGCCAGGGCGGCAAGGCCGCCTACCTCGCCGAGTCCGCGCAGTCCGGAGCCAC

CGTGCCAGTGGGGCGGGCGGCAGCAGCTCGGCTGCTGGTAGTGCGGGCCGTGG rGGCCGGG Si,maI (3)

GAGCGAGGCGGTCCCGCCGTTCCGGCGGATGGAGCGGCTCAGGCGCGTCAGGCCTCGGTG

T 2581 CGTCACCGGCCCCCACCGCCCCGGCTCCGTCCCGCCCGGTCTTCCACCTCACC :CCGGGCA GCAGTGGCCGGGGGTGGCGGGGCCGAGGCAGGGCGGGCCAGAAGGTGGAGTGGC GGCCCGT

A K V S V L Y I G P M E A P A D A A Q R 601 GGCCAAGGTGAGCGTCCTCTACATCGGCCCCATGGAGGCCCCCGCGGACGCCGCCCAGCG

T

S

A

V

P

L

T

P

A

P

L

A

P

T

V

A

A

G

V

P

G

V

A

G

E

P

T

P

S

A

T

R

V

T

P

V

I

V

L

T

F

V

P

H

V

G

L

G

T

A

G

P

V

CCGGTTCCACTCGCAGGAGATGTAGCCGGGGTACCTCCGGGGGCGCCTGCGGCGGGTCGC

L V

2641 CCGCGCTCGCCCTCGTCGGCGGCGGCACCGCCGTCGTCCTGGTCGTCGGCGCC GTCCTGG GGCGCGAGCGGGAGCAGCCGCCGCCGTGGCGGCAGCAGGACCAGCAGCCGCGG CAGGACC

661

membrane associated helices

L G V P A G T Q V L A R R R L Y F R N G ACTGGGCGTCCCCGCCGGCACTCAGGTGCTCGCCCGGCGGCGCCTCTACTTCCGCAACGG TGACCCGCAGGGGCGGCCGTGAGTCCACGAGCGGGCCGCCGCGGAGATGAAGGCGTTGCC

L A A A V T A A S L A I C A L V I 2701 TCTCCATGCTCCTCGCGGCCGCCGTCACCGCCGCCTCGCTGGCCATCTGCGCC CTGGTCA

721 CACCCCGGTCGAGACCGCCTCCTCCTACCTCCCGTGGGACGTCGTCAAGGACATCCCCGA

AGAGGTACGAGGAGCGCCGGCGGCAGTGGCGGCGGAGCGACCGGTAGACGCGG GACCAGT

GTGGGGCCAGCTCTGGCGGAGGAGGATGGAGGGCACCCTGCAGCAGTTCCTGTAGGGGCT

S

R

I

T

L

T.L V

N

A

Q

H

R

R

---

2761 TCCGCTCACTCGTCAACGCCCAGCACCGCCGCTGAACGAATCACCGGGGCGAC GCACAAG AGGCGAGTGAGCAGTTGCGGGTCGTGGCGGCGACTTGCTTAGTGGCCCCGCTG CGTGTTC D.R. 3

SzmaI (2) 2821 CC CCGCCCCCGGG

GGCGGTTCGTGCGGCGGCGGGGGCCC

P

V

E

T

A

S

S

Y

L

P

W

D

V

V

K

D

I

P

E

L F A E N P G G G I Y A R L E D H G H 781 GCTGTTCGCCGAGAACCCCGGCGGCGGTGGCATCTACGCCCGACTCGAAGACCACGGGCA

CGACAAGCGGCTCTTGGGGCCGCCGCCACCGTAGATGCGGGCTGAGCTTCTGGTGCCCGT E F A E F V E T L Q A R P A S K A E A T 841 CGAGTTCGCCGAGTTCGTCGAGACGCTGCAAGCACGGCCGGCCTCTAAGGCGGAAGCCAC

GCTCAAGCGGCTCAAGCAGCTCTGCGACGTTCGTGCCGGCCGGAGATTCCGCCTTCGGTG E L A L S P G A P V V H L I R E A R T T. 901 CGAACTGGCTCTCAGCCCAGGCGCCCCGGTCGTTCACCTGATCCGGGAAGCCCGCACCAC

GCTTGACCGAGAGTCGGGTCCGCGGGGCCAGCAAGTGGACTAGGCCCTTCGGGCGTGGTG A G L V V E V C D T L M A A D Q F V F E 961 GGCCGGGCTCGTCGTCGAGGTCTGCGACACCCTCATGGCCGCTGACCAGTTCGTTTTCGA

CCGGCCCGAGCAGCAGCTCCAGACGCTGTGGGAGTACCGGCGACTGGTCAAGCAAAAGCT

1021

BamHI (11) Y R I P A A D --I.R. GTACCGGATCCCAGCAGCCGACTGACGACCGCTCAACTCCTCACAGCCCGTCGCAGTTC CATGGCCTAGGGTCGTCGGCTGACTGCTGGCGAGTTGAGGAGTGTCGGGCAGCGCTCAAG

1081

TCTGTCGCGGCGGGTTGACTCATGTATAGGAGTGGTGTACTCTTCTTCATGTCACTCATA AGACAGCGCCGCCCAACTGAGTACATATCCTCACCACATGAGAAGAAGTACAGTGAGTAT

1141 TACATGAGTGACGGAGTCCAGCCTCTATAGAGGAGTGATCCGCTGTGCGTCAGATCCCCG

ATGTACTCACTGCCTCAGGTCGGAGATATCTCCTCACTAGGCGACACGCAGTCTAGGGGC

Sa1I 1201 TCGAC AGCTG FIG. 2-Continued.

5534

J. BACTERIOL.

HAGEGE ET AL. TABLE 2. Characteristics of ORFs involved in pSAM2 transfer, pock formation, and plasmid maintenance

ORF

in Coordinates sequence (bp)a

Overall G+C (%)

traSA orf84 spdA spdB spdC spdD korSA4

302-1219 1093-1344 1430-2102 2127-2300 2313-2465 2481-2792 266-1042

69.0 81.5 74.0 72.2 68.5 76.8 69.4

position: 3 2 1 G+C (%) at codon

mol wt Predicted of protein

306 84 224 58 51 104 259

33,441 8,613 23,575 6,531 5,512 10,399 27,868

84.8 93.0 87.4 96.5 94.1 97.0 86.4

49.2 68.8 53.9 57.6 38.4 61.5 49.7

72.7 82.7 80.7 62.7 73.0 72.1 72.1

No. acids of amino

a Data for korSA refer to nucleotide positions in Fig. 2b; those for all other ORFs refer to nucleotide positions in Fig. 2a.

302 to 1219 (Fig. 2a). Derivatives carrying a deletion in the C-terminal part of this ORF (pD19) or in the entire ORF (pD8) did not form pocks, and their transfer abilities were very low (Fig. 3). No other deletions had such dramatic effects on pSAM2 pock formation, suggesting that this ORF encodes a protein required for pSAM2 intermycelial transfer from the donor to the recipient. Therefore, we called the gene traSA (for transfer gene of S. ambofaciens) and called its product TraSA. traSA cannot be introduced into Streptomyces spp. in the absence of the BamHI(11)-SacI(14) region (46) (Fig. 4), indicating that it has a lethal effect (kil locus). The functional properties of traSA are similar to those of transfer genes from other Streptomyces plasmids, such as tra of pIJ101 (28) or traB of pSN22 (24), which are essential for transfer and pock formation and which also have lethal effects. The study of the TraSA amino acid composition revealed the presence of one hydrophobic stretch of amino acid residues (17 aa in length; aa 176 to 193) predicted to be a transmembrane segment by the method of Klein et al. (30). Furthermore, it contains two short sequence motifs, A (aa 78 to 85) and B (aa 102 to 106) (Fig. 2a and Sa), similar to nucleotide-binding domains (55). The deduced traSA product has significant similarity (23.1% identity and 61% overall similarity in a 225-aa segment) to Tra of pIJlOl, the major protein involved in pIJ101 transfer (Fig. 5a). Interestingly, pIJ101 Tra protein may also be a nucleotide-binding protein Sau3A (5)

NruI (8)

(28). However, it should be noted that traSA of pSAM2 is predicted to encode a protein half the size (306 aa) of that encoded by tra of pIJ101 (621 aa). TraSA has similarities only with the C-terminal half of Tra of pIJ101 (Fig. Sb). Downstream from traSA lies another ORF. There are two possible start codons for this ORF. If the first initiation codon is used, the ORF would encode an 84-aa protein that overlaps the last 43 aa of traSA. The second possible start codon overlaps the TGA stop codon of traSA. If this second initiation codon is used, this small ORF, encoding a 41-aa protein, and traSA might be translationally coupled. Such a situation is quite common in Streptomyces operons (28). No proteins with significant sequence similarity to the putative products of these ORFs were identified in the data bases examined. Functional and sequence analysis of spd4, spdB, spdC, and spdD: involvement in pSAM2 spreading within the recipient mycelium. Previously it was shown that derivatives carrying deletions in the BstEII(4)-SmaI(6) region were affected in their pock sizes and transfer abilities (46) (Fig. 1). Analysis of this region revealed the presence of four putative ORFs: spd4, spdB, spdC, and spdD (Fig. 2a and 3). Derivatives carrying a deletion in spdD (pD6A) or in which the BstEII(4) site located in this ORF was destroyed by Klenow fragment (pTS59A) yielded almost normal pocks like those of pTS39. However, pD6A transfer efficiency was rather low (Fig. 3). Uninterrupted transcription through spdD is necessary for BstEIl (4) SmiaI (2) I

ftrSA

oiT8

spdA

~Sparsic 7RMb Omega interposon Klenow Deletion

plasmid pTS68

pock

small

transfer

efficiency -4

4x10

pTS59A almost normal

ND

pD6A pD5

almost norrmal

8x10

small

pD1O

very small

1.4x102 2x102

pD19

no

2x10-6

pD8

no