Mode of insertion of the broad-host-range plasmid RP4 and its derivatives into the chromosome of Myxococcus xanthus

PLASMID

18,

111-119 (1987)

Mode of Insertion of the Broad-Host-Range Plasmid RP4 and Its Derivatives into the Chromosome of Myxococcus xanthus SAMIR JAOUA,*

JANINE

F. GUESPIN-MICHEL,*

AND ANNICK

M. BRETON~

*Laboratoire de microbiologic, Faculte des sciences et techniques de Rouen, BP67, 76130 Mont-saint-Aignan, TLaboratoire de Gtknhique Microbienne, Vniversith de Technologie de Compit?gne, BP233. 60206 CompiSgne, France

and

Received March 20, 1987; revised July 1, 1987 The mode of insertion of the broad-host-range plasmid RP4 into the chromosome of Myxococcus xanthus strain DZl has been analyzed. The plasmid integrated in numerous sites of the chromosome and generated insertional mutations. There is a hot spot of integration located between 3 1.5 and 34.5 kb clockwise from the EcoRI site of the plasmid. In the absence of this segment the insertion can, however, take place, but much less efficiently. The presence of transposable elements on the plasmid decreases severely the insertion frequency. Once integrated, RP4 could be transferred back to Escherichia coli, either by precise excision or with a segment of the Myxococcus chromosome. The role of site-specific recombination in RP4 integration is discussed. 0 1987 Academic Press, Inc.

Myxococcus xanthus is a well-studied myxobacterium for which several systems of gene transfer have been established. The coliphage P 1 has been extensively used, first for introducing transposon Tn5 into M. xanthus chromosome (Kaiser, 1984; Kuner and Kaiser, 198 I), and subsequently for transferring M. xanthus cloned genes from Escherichia coli back to their original host (O’Connor and Zusman, 1983; Shimkets et al,, 1983). We have previously shown that the broadhost-range plasmid RP4 can be conjugated into M. xanthus and that it is maintained in a chromosomally integrated state (Breton et al., 1985). These properties have been used to integrate foreign genes into the chromosome of M. xanthus (Breton et al., 1986; Breton and Guespin-Michel, 1987). We have also shown that the inserted plasmid can be transferred back to E. coli without any apparent change, or with an added fragment arising from the M. xanthus chromosome (R prime). Besides Bacteroides spp., M. xanthus is one of the rare known examples into which

RP4 can be conjugated and maintained but cannot replicate autonomously (Shoemaker et al., 1986). In several instances this plasmid can be integrated into the host’s chromosome. Among gram-negative bacteria, in all cases studied, integration is promoted either by transposon Tnl (Harayama et al., 1980) or by an insertion sequence, termed IS21 (or 1%) present on the RP4 molecule. The frequency of these insertions is generally quite low and is significantly increased when a second IS21 is adjacent to the first one, as in plasmid R68.45 (Haas and Holloway, 1978), or by means of a sequence homology to the host chromosome (Julliot and Boistard, 1979). We have studied the mode of integration of RP4 into the chromosome of M. xanthus DZ 1, a strain in which the frequency of integration of RP4 is about lOOO-fold higher than observed with other strains of the same species (Breton et al., 1985). MATERIALS

Bacterial conditions. 111

AND

METHODS

strains, plasmids, and growth Bacterial strains and plasmids 0147-619X/87

$3.00

Copyright 0 1987 by Academic Press, Inc. All rights of reproduction in any form reserved.

JAOUA. GUESPIN-MICHEL.

112

AND BRETON

TABLE I BAC-TERIAL STRAINS AND Relevant characteristics

PLASMIDS Reference

Strains Myxococcu.~

xanthus

DZI CM2003 1 CM2005

Parental strain Sma

1

CM2021 1 CM2022

I

CM2023 CM2024 CM2025 CM2026 CM2027 CM2030 CM2019

I 1 I I I 1

Different clones SmR, OCR, Km’, CbR obtained by transfer of RP4 into DZI

DZ I containing pUZ8 Sm’, KmR, Oca, HgR

Zusman ef al. (1978) Breton c( al. (1985) This work This work This work This work This work This work This work This work This work This work

E. coli

W3101 Nal Plaamids incP- 1 RP4 R68.45 pME305 pUZ8 pCM2003

pCM2005 pCM2019 pCM2023 pCM2030

pULBl13

Ret A13 trp E Nal

Kopecko et a/. ( 1976)

Km, Tc, Ap/Cb Km, Tc, Ap/Cb Tc, Ap/Cb, repts Km, Tc, Hg Km, Tc, Ap/Cb Km, Tc, Ap/Cb Km, Tc, Hg Km, Tc, Ap/Cb Km, Tc, Ap/Cb Km, Tc. Ap/Cb

Dattaef a/. (1971) Haas and Holloway (1978) Rella (1984) Hedges and Matthew ( 1979) Breton ef al. (1985) From CM2005 (this work) From CM2019 (this work) From CM2023 (this work) From CM2030 (this work) van Gijsegem and Toussaint (1982)

No(e. Km, Tc/Oc, and Ap/Cb are markers as described by Novick et al. (1976). Tc/Oc confers resistance to tetracycline in E. co/i and to oxytetracycline in M. xanthus. Ap/Cb confers resistance to ampicillin in E. co/i and to carhenicillin in M. xanrhus. Km confers resistance to kanamycin.

used in this study are listed in Table I. M. xmthus was grown at 30°C in CYE medium: Casitone (Difco Laboratories), 1%; yeast extract (Biomerieux), 0.5%; MgS04, 0.1%; pH 7.6. TT is a 10 mM Tris-HCI, pH 7.6, buffer, containing 1 mM KH2P04/Na2HP04 and 0.2% MgSO.,. E. coli was grown at 37°C in Luria broth (Miller, 1972). The media was solidified with 1.2% agar (Biomerieux). Milk plates were prepared from CYE plates overlaid with 3 ml of a mixture (2:l) of TT soft agar (0.6%) and sterilized skimmed milk. Proteolytic activity was estimated by the size of a lysis halo surrounding colonies toothpicked

onto CYE milk plate. This size is compared to reference strain DZl(?). For M. xanthus, the antibiotics used and final concentrations (in milligrams per liter) were kanamycin sulfate, 75 (Bristol Laboratories); oxytetracycline chloride, 12.5 (Roussel Uclaf); carbenicillin, 500 (Beecham Sevigne); nalidixic acid, 100 (Sigma), and streptomycin, 500 (Diamant). For E. coli, the antibiotics used (in milligrams per liter) were ampicillin, 25 (Bristol Labomtories); kanamycin, 50; nalidixic acid, 100; and tetracycline, 10 (Roussel-Uclaf). Matings. Matings were carried out as described in Breton et al. (1985).

INSERTION

113

OF RP4 INTO THE Myxococcus xunfhus CHROMOSOME

Plasmid detection. Free supercoiled plasmid DNA was visualized by the method of Eckbardt ( 1978). Plasmid DNA isolation. Plasmids were extracted from E. coli by the cleared lysate technique (Cohen and Miller, 1969) and the supercoiled plasmid DNA was separated from linear chromosomal DNA by ultracentrifugation in a cesium chloride-ethidium bromide gradient (Clewell and Helinski, 1969). After removal of the ethidium bromide, the DNA was dialyzed against TE buffer (10 mM Tris-HCl, pH 8.0, 1 mM EDTA, Na3). Isolation of total DNA from M. xanthus. The total DNA was extracted as described by Davis et al. (1980) adapted for M. xanthus by Avery and Kaiser (1983) from strains grown on solid selective medium (CYE, Km, Oc, Cb).’ Restriction enzyme digestion and agarose gel electrophoresis. Restriction buffers were used according to Davis et al. ( 1980). Restriction enzymes ClaI and PstI were purchased from Appligene France and Sac11 was from New England Biolabs. XDNA digested by both EcoRI and Hind111 (BoehringerMannheim) was used as a standard. Southern blots, jilter hybridization, and autoradiography. DNA digested by the ap propriate enzymes was run in a horizontal Tris-acetate agarose (0.7%) gel and transferred from the gel to the nitrocellulose as described by Davis et al. ( 1980). The plasmid DNA to be used as a probe was labeled by nick translation using 2 mCi of [cy-32P]dCTP (Amersham Corp.) per milliliter and 200 ng of DNA in 25 ml final volume incubated at 15°C according to Rigby et al. (1977). The hybridization was done using a slightly modified Denhardt method (1976). The hybridization buffer was 3X SSC ( 1 X SSC is 0.15 M NaCl and 0.015 M sodium citrate, pH 7), ’ Abbreviations used: Ap, ampicillin, cb, carbenicillin, Km, kanamycin, Oc, oxytetracycline, Tc, tetracycline, Sm, streptomycin, dCTP, deoxycytidine triphosphate.

TABLE 2 ~IENOTYPIC CLASSIFICATION TRAN~C~NJUGANTS

OF DZ I

Transconjugants

Color

Proteolytic activity

Percentage of clones

DZI CM202 I CM2022 CM2023 CM2024 CM2025 CM2026 CM2027

T T T/R T/Y T T/R T/Y T/R

f f + ++ ++ + + f

9 I I 29 43 2 15

Nofe. T, tan; R, red: Y, yellow. The total number of clones is 500. The proteolytic activity is scored as described under Materials and Methods.

10X Denhardt (1 X Denhardt is 0.02% bovine serum albumin, 0.02% Ficoll 400, and 0.02% polyvinylpyrrolidone), and 0.1% sodium dodecyl sulfate. The prehybridization lasted 12 h at 70°C and the hybridization in the same buffer with the labeled denaturated probe lasted 24 h at 70°C. After hybridization, the filters were washed four times with 3~ SSC, 0.1% sodium dodecyl sulfate at 70°C and two times with 0.5~ SSC at 70°C. Autoradiography was performed using FUJI X-ray film; films were exposed at -80°C for several days. RESULTS

RP4 Integration Mutations

Create Insertional

The existence of potential insertional mutations due to RP4 integration was checked with a collection of 500 transconjugants displaying the three resistances conferred by RP4 to M. xanthus, Km, Oc, and Cb. Table 2 shows that the transconjugants could be classified into different phenotypic classes according to their color and the size of halo of proteolysis. Strain DZl is a mutant strain of M. xanthus, which forms smooth tan colonies instead of motile yellow ones. It is impaired in the production of several extracel-

114

JAOUA,

GUESPIN-MICHEL,

AND

BRETON

-B 13 RP4

I

, ’

,y

0

3

6

! 9

‘,

12

15

I

I

I

I

16

21

24

27

1” 30 t't 33

‘VI

1’ 36

39

42

I

I

I

45

48

51

1’ 54

’ 56.4 KbP

IHi Ap

Km

Tc

pME305 44.4

Kbp

pUZ8 54.3 Kbp

Hg FIG. 1. Restriction maps of RP4 and its derivatives. This figure is a synthesis of several maps given in Haas and Riess ( 1983) Villarroel et al. ( 1983), and Rella ( 1984). The sizes are in kilobases.

lular enzymatic activities. Often in Myxococcus, a change in colony color may be due to a shift in color phase (Burchard and Dworkin, 1966), but mutations impairing the colony color have already been described (Hagen et al., 1978). In the absence of auxotrophic mutants such mutants are important in M. xanthus genetic studies. We have shown that Tn5 insertions can lead to a decrease or an increase in extracellular activities (Nicaud et al., 1984) and the proteolytic activity (the size of a lysis halo on CYE-milk plates) is an easily scored trait. Although we are aware that these phenotypes are somewhat arbitrary, they suggested the existence of insertional mutations. One strain from each class was kept for further analysis.

D@erent RP4 Insertion Sites into the A4yxococcusxanthus Chromosome Chromosomal DNAs from each of the seven transconjugants described above were digested with PstI, electrophoresed, and hybridized with an RP4 probe (Southern blot (Fig. 2)). In all cases a 6-kb fragment, present in self-hybridization pattern of RP4, is absent from the autoradiogram of the digested chromosomes and replaced by two bands, with molecular weights different for each

strain (Table 3). The sum of the two extra bands differs in each case, which indicates that the insertions occurred at different sites on the chromosome. Two other randomly picked transconjugants (CM2003 and

a

b

c

d

e

f

g

2521 -

6-

2.5-

FIG. 2. Southern blot hybridizations of KmR, OCR,and CbR transconjugants. DNA prepared from several strains was digested with PstI and analyzed by Southern blot hybridization as described under Materials and Methods. Lanes (a) CM202i, (b) CM2022, (c) CM2023, (d) CM2024, (e) CM2025, (f) CM2026, and (g) CM2027. The sizes in kilobases of RP4 PsrI fragments are indicated at the side.

INSERTION

OF RP4 INTO THE Myxococcus xunthus CHROMOSOME

115

TABLE 3 MOLECULAR

WEIGHTS OF THE Two MYXOCD~CVS XANTHVS

Transconjugant

CM202 1

FRAGMENTS HYBRIDABLE CHROMOSOME REPLACING

WITH RP4 WHICH CONTAIN THE 6-KB PstI FRAGMENT

CM2022

CM2024

CM2023

CM2025

A SEGMENT OF RP4

CM2026

OF THE

CM2027

Molecular weight of the two fragments

10.5 9.3

8.2 1.0

8.5 6.2

10.5 2.0

5.9 2.2

14.0 2.3

5.7 2.1

Sum

19.8

9.2

14.7

12.5

8.1

16.3

7.7

Note.

The sizes of fragments were calculated by comparing their mobilities with those of X standards.

CM2005) were also checked by the same technique; the insertions took place in the same 6-kb fragment of the plasmid RP4 and in two other chromosomal loci (result not shown).

prior to Southern blot hybridization (Fig. 4). In all cases, a 2-kb new band appears, while in strains CM2022, CM2023, and CM2027, a 3-kbp ClaI-Sac11 segment (marked with arrows in Fig. 1) is clearly missing and one (or two) band of different size is also present.

The RP4 Primes are D@erent Breton et al. (1985) have studied two RP4 primes arising from the same transconjugant upon conjugation with E. cd. The two plasmids analyzed were different in size, but shared a common segment of the M. xanthus chromosome as judged from their restriction pattern. Figure 3 shows the P.stI digestion pattern of one of those plasmids (pCM20036) and of another RP4 prime (pCM2030) obtained from a different transconjugant. It is clear that the two patterns are completely different, indicating that the two R-prime plasmids carry different segments of the M. xanthus chromosome.

a

b

c

d

The RP4 Site of Integration From Fig. 2, it is clear that in all the cases RP4 is inserted via the 6-kb PstI fragment. The fact that a RP4 fragment was lost on digestion with PstI, which cuts inside IS21 (Fig. I), excludes normal insertion of the plasmid between two copies of IS21; in addition, part of IS21 is included in the 6-kb fragment (Fig. 1). To characterize the mode of insertion, ClaI-Sac11 double digestion of chromosomal DNA from the transconjugants were done

FIG. 3. Restriction analysis of RP4 and its derivative RP4 primes, pCM2003-6 and pCM2030, digested with Pst 1 restriction endonuclease. Lanes (a) XDNA digested by EcoRI showing fragments of 21.8, 7.52, 5.93, 5.54, 4.8, and 3.41 kb and by HindIII showing fragments of 28,23.51,9.59,6.63,4.29,2.28 and 1.94 kb,(b)RP4,(c) pCM2030, and (d) pCM2003-6(h) CM202 1.

116

JAOUA, GUESPIN-MICHEL, abcdefo

17129.3 -

6.5 4.7 3-

1.8FIG. 4. Southern blot hybridization of KmR, OcR, CbR transconjugants. DNA prepared from the strains was digested by both Sac11 and ClaI restriction endonucleases and analyzed by Southern blot hybridization as described under Materials and Methods. Lanes (a) CM2021, (b) CM2022, (c) CM2023, (d) CM2024, (e) CM2025, (f) CM2026, and (g) CM2027. The sizes in kilobases of RP4 .%x11-ClaI fragments are indicated at the side.

The simplest explanation of these results is that the insertion took place preferentially in this 3-kb segment, 2 kb apart from either CluI or Sac11 site, and generated another Sac11 or ClaI site at one of the new junctions. Integration

AND BRETON

different RP4 derivatives, is given in Table 4. Plasmid pUZ8 is very efficiently inserted. The absence of the 3-kb segment decreased the insertion frequency of pME305 but did not prevent the insertion, which showed that this segment was not essential for integration of the plasmid. This was also shown in a Southern blot hybridization analysis of a transconjugant bearing pULB 113 with an RP4 probe in which the insertion took place between the 13.7 and 29.7 kb (result not shown). On the other hand, the presence of one extra copy of IS21 was associated with a decrease in the frequency of integration (R68.45 < RP4 < pUZ8). ModiJication of Plasmids Transferred from M. xanthus back to E. coli A common mechanism of integration of a plasmid into another replicon involves a transposable element. The plasmid is then inserted in between two copies of the element, one copy of which is retained on the plasmid upon excision. When transferred back to E. coli, RP4 plasmids are not conspicuously larger than the original RP4. Among the four retransferred plasmids that we studied, two plasmids (pCM2004 and pCM2005) behave like RP4, having the same frequency of transfer to M. xunthus and restriction patterns identical with those of sev-

of D@erent RP4 Derivatives

In order to confirm these results, we have studied the insertion of several RP4 derivatives described in Table 1. R68.45 possesses two adjacent IS21 sequences and is integrated much more readily than RP4 into the chromosome of many gram-negative bacteria (Haas and Holloway, 1978); pUZ8 and pME305 lack IS21 but pME305 also lacks the hot-spot 3-kb Cl&&.zcII segment and most of the 6-kb PstI segment. In pULB 113, a transposon Mu3A is integrated into the SacII-C/u1 hot spot (Van Gijsegem and Toussaint, 1982). A comparison of the frequencies of transfer, and thus of the integration of these

TABLE 4 TRANSFER

Plasmid RP4 R6845 pME305

pUZ8 pULB113 pCM2023 pCM2019

FREQUENCY PLASMIDS

OF DIFFERENT INTO DZ 1

IncP- 1

Selective agent

FrequencyY

Km, Sm Km, Sm Oc, Sm Km, Sm Km, Sm Km, Sm Km, Sm

5.10-* 5.10-4 6.10-’ 6.10-l 5.10-* 6.10-l 6.10-l

a Number of cells obtained on selective media per number of recipient cells. The donor strain of plasmid is W3101 NaI.

INSERTION

OF RP4 INTO THE Myxococcus

eral restriction enzymes (results not shown). One retransferred plasmid, pCM2023, acquired a higher frequency of insertion into the DZl chromosome (Table 4). Figure 5 compares the restriction profile of pCM2023 with that of the parental RP4. pCM2023 has lost the 3-kb SucII-CluI segment of RP4, which is replaced by a 4-kb band. Finally, pCM2019 was retransferred from CM20 19, which contains pUZ8. Its frequency of transfer to DZl did not increase, but pUZ8 was very efficiently transferred into this strain (Table 4). However, when other strains were used as recipients, such as strain DK 10 1 that was shown to be a much less efficient receptor (Breton et al., 1985), pCM20 19 was transferred 1OO-fold more efficiently than pUZ8. No difference, however, could be detected between the restriction profiles of both pUZ8 and pCM20 19. DISCUSSION

Insertion of RP4 into the M. xanthus chromosome occurs via a specific mechaa

b

17 12 -

6.5-

FIG. 5. Restriction pattern of a modified plasmid transferred from M. xunthus transconjugant CM2023 to E. cob. Lanes (a) pCM2023 restricted by both SzcII and CIaI endonucleases and (b) RP4 restricted by Sac11 and CIaII endonucleases. The sizes in kilobases of the RP4 SucII-ClaI fragments are indicated at the side.

xanthus CHROMOSOME

117

nism. We present evidences indicating the lack of involvement of transposon Tnl, and the resident IS21, although the integration takes place preferentially in a region of the plasmid close to that sequence. Indeed, the deletion of these sequences (in plasmid pUZ8) considerably increases the frequency of integration of the plasmid. There is a hot spot of integration in a 3-kb segment of RP4; deletion of that segment, even when IS21 is also deleted, such as in plasmid pME305, decreases the efficiency of integration. A simple hypothesis would be that the integration of RP4 can be mediated by an insertion sequence from the M. xunthus chromosome, having a preferential insertion site in the 33-kb region of the plasmid. This hypothesis is supported by three facts: a hot spot of insertion on the plasmid, numerous integration loci on the Myxococcuschromosome, and modification of the retransferred pCM2023 plasmid. The occurrence of insertional mutations may be, to some extent, inconsistent with this hypothesis. However, they could arise in one of two ways: several copies of the postulated IS may exist on the M. xunthus chromosome, and mutations could arise if RP4 insertion generates transcriptional modifications of chromosomal rearrangements. This possibility is supported by studies on the structural instability of inserted incP-1 plasmids (Jaoua et al., 1986). Alternately, transfer of the plasmid to a new host might activate the transposition mechanism of the putative IS that is then transposed to other parts of the chromosome as well as to the plasmid. The very high frequency of transfer observed with plasmid pUZ8 could be explained by such type of activation. Other mechanisms of integration are equally possible. The lack of a conspicuous new DNA sequence in both pCM2005 and pCM20 19 might indicate that in M. xunthus strain DZl, recombination occurs between very small regions of homology, undetectable by the classical hybridization techniques. Since there is no change in the

118

JAOUA, CXJESPIN-MICHEL,

transfer frequency of pCM2005, the excision process is likely precise, whereas in pCM2019 a modification, undetected by restriction analysis, might be responsible for increasing recombinations upon retransfer to h4yx~1coccu.~.Detailed molecular analysis of the insertion regions of retransferred plasmids is in progress in our laboratory in order to determine precisely the nature of the mechanism responsible for plasmid integration in the M. xunfhus chromosome. ACKNOWLEDGMENTS

REFERENCES AVERY, L., AND KAISER, D. (1983). In sifu transposon replacement and isolation of a spontaneous tandem genetic duplication. Mol. Gen. Gena. 191, 99-109. BRETON, A. M., JAOUA, S., AND CUESPIN-MICHEL, J. (1985). Transfer of plasmid RP4 to Myxococcus xanrhus and evidence for its integration into the chromosome. J. Bacterial. 161, 523-528. BRETON, A. M., YOUNES, G., VAN GLISEGEM, F., AND GUESPIN-MICHEL, J. (1986). Expression in Myxococcus xanthus of foreign genes coding for secreted pectate lyases of Erwinia chrysanthemi. J. Biofechnol. 4,303-3 1 I. BRETON, A., AND GUESPIN-MICHEL, J. F. (I 987). Escherichia coli pH 2.5 acid phosphatax and &lactamase TEM2 are secreted into the medium by Myxococcus FEMS

Microbial.

Lett. 40, I83-

188.

BURCHARD, R. P., AND DWORKIN, M. (1966). Light induced lysis and carotenogenesis in Myxococcus xanthus.

J. Bacferiol.

91, 535-545.

CLEWELL, D. B., AND HELINSKI, D. R. (1969). Supercoiled circular DNA-protein complex in Escherichia co/i: Purification and induced conversion to an open circular DNA. Proc. Natl. Acad. Sci. USA 62, 1159-I 166. COHEN, S. N., AND MILLER, C. A. (I 969). Multiple molecular species of circular R factor DNA isolated from Escherichia

coli. Naure

(London)

DENHARDT, D. T. (1976). A membrane filter technique for the detection of complementary DNA. Biochcnr Btophys. Rcs. Commun. 23, 64 l-646. &KARDT, T. (1978). A rapid method for the identification of plasmid deoxyribonucleic acid in bacteria. Plasmid 1. 584-588. HAAS, D.. AND RIESS,G. (1983). Spontaneous deletions of the chromosome-mobilizing plasmid R68.45 in Pseudomonas aeruginosa PAO. Plasmid 9,42-52. HAAS. D., AND HOLLOWAY, B. W. (1978). Chromosome mobilization by R plasmid R68.45: A tool in Pseudrr monas genetics. Mol. Gen. Gena. 158,229-237. HAGEN, D. C., BRETSCHER, A. P., AND KAISER, D. ( 1978). Synergism between morphogenetic mutants of Myxocaccu.~

This work has been supported by a grant from the MRT (bourse de docteur ingenieur to S.J.) and by a grant from the “Conseil Regional de Picardie” for purchasing the equipment.

xanthus.

AND BREI‘ON

224, I276-

1277.

DA~-TA, N., HEDGE& R. W., SHAW, E. J., SYKES, R. B., AND RICHMOND, M. H. (1971). Properties of an R factor from Pseudomonas aeruRinosa. 1. Baderiol. 108, 1244- 1249. DAVIS, R. W., BOTSTEIN, D., AND ROTH, J. R. (1980). “Advanced Bacterial Genetics. A Manual for Genetic Engineering.” Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.

xunthus

Dev. Biol. 64, 284-296.

HAR~YAMA, S., TSUDO, M., AND LINO, T. (1980). High frequency mobilization of the chromosome of Escherichia co/i by a mutant of plasmid RP4 temperaturesensitive for maintenance. Mol. Gen. Genef. 180, 47-56. HEDGES, R. W.. AND MATTHEW, M. (1979). Acquisition by Escherichia co/i of plasmid-borne fl-lactamases normally confined to P.wudomonas spp. Playmid 2,269-278.

JAOUA, S., BRETON, A. M., YOUNES, G., AND GUESPIN-MICHEL, J. F. (1986). Structural instability and stabilization of IncP-I plasmids integrated into the chromosome of Myxococcu~ xanthus. J. Biotechnol. 4.3 13-323. JULLIOT, J. S.. AND BOISTARD, P. (1979). Use of RP4prime plasmids constructed in vitro to promote a polarized transfer of the chromosome in E. co/i and Rhizobium

mdiloti.

Mol.

Gen. Genet.

173, 189-298.

KAISER, D. (1984). Genetics of myxobacteria. In “Myxobacteria: Development and Cell Interactions” (E. Rosenberg, Ed.), pp. 163- 184. Springer-Verlag, Berlin/New York. KOPECKO, D. J., BREVET, J., AND COHEN, S. N. (1976). Involvement of multiple translocating DNA segments and recombinational hot-spots in the structural evolution of bacterial plasmids. J. Mol. Biol. 108, 333-360. KUNER, J., AND KAISER, D. (1981). Introduction of transposon Tn5 into Myxococcus for analysis ofdevelopmental and other non selectable mutants. Proc. Nat/

Acad. Sci. USA 78,425-429.

MILLER, J. H. (1972). “Experiments in Molecular Genetics.” Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. NICAUD, J. M., BRETON, A. M., YOUNES, G., AND GUESPIN-MICHEL, J. (1984). Mutants of Myxococcus xanthus impaired in protein secretion: An approach to study of a secretory mechanism. Appl. Microbial. Biotechnol.

20, 344-350.

NOVICK, R. P., CLOWES, R. C., COHEN, S. N., CURTISS, R. III, DATTA, N., AND FALKOW, S. (1976). Uniform

INSERTION

OF RP4 INTO THE Myxococcu.r xunrhus CHROMOSOME

nomenclature for bacterial plasmids: A proposal. Bucteriol.

Rev. 40, 168-I

89.

O’CONNOR, K. A., AND ZUSMAN, D. R. (1983). Coliphage Pl-mediated transduction of cloned DNA from Escherichia coli to Myxococcus xanrhus: Use for complementation and recombinational analysis. J. Bacterial.

155, 3 17-329.

RELLA, M. (1984). “Transposon Insertion Mutagenesis of the Pseudomonas aeruginosa Genome.” Ph.D. thesis. RIGBY, D. W. J., DIECKMANN, M., RHODES, C., AND BERG, P. (1977). Labelling deoxyribonucleic acids to high specific activity in vitro by nick-translation with DNA polymerase I. J. Mol. Biol. 113, 237-25 I. SHIMKETS, L. J., GILL, R. E., AND KAISER, D. (1983). Developmental cell interactions in h4yxococcu.r xanthus and the spoC locus. Proc. Narl. Acad. Sci. USA 80, 1406-1410. SHOEMAKER, N. B., Gtz’t-rv, C., GARDNER, J. F., AND SALYERS, A. A. (1986). Tn4351 transposes in Bacfer-

119

aides spp. and mediates the integration of plasmid R75 I into the Bacteroides chromosome. J. Bacterial. 165,929-936.

VAN GUSEGEM, F., AND TOUSSAINT, A. (I 982). Chromosome transfer and R-Prime formation by an RP4::miniMu derivative in Escherichiu coli, Salmonella typhimurium. Klebsiella pneumoniae and Proteus mirabilis.

Plasmid

7, 30-44.

VILLAROEL, R., HEDGES, R. W., MAENHAUT, R., LEEMANS, J., ENGLER, G., VAN MONTAGU, M., AND SCHELL, J. (1983). Heteroduplex analysis of P-plasmid evolution: The role of insertion and deletion of transposable elements. Mol. Gen. Genef. 189, 390-399.

ZUSMAN, D. R., KROTOSKI, D. M., AND CUMSKY, M. (1978). Chromosome replication in Myxococcus xanthus. J Bacterial. 133. 122- 129. Communicated

by Julian

Davies