Proc. Natl. Acad. Sci. USA Vol. 87, pp. 1278-1282, February 1990 Biochemistry Pseudomonas chromosomal replication origins: A bacterial class distinct from -type origins (initiation/DnaA-binding sites/GATC sites/13-base-pair repeats/autonomously replicating sequence/dnaA-dnaN-rpmH-rnpA genes) THOMAS W. YEE* AND DOUGLAS W. SMITHt Department of Biology, B-022, and Center for Molecular Genetics, University of California, San Diego, La Jolla, CA 92093 Communicated by Donald R. Helinski, October 13, 1989 (received for review June 12, 1989)

ABSTRACT The bacterial origins ofDNA replication have Nevertheless, viable dam- insertion and deletion mutants been isolated from Pseudomonas aeruginosa and Pseudomonas of E. coli have been isolated (20, 21); hence, GATC meth- putida. These origins comprise a second class of bacterial ylation is a dispensable cellular function. Dam origins distinct from enteric-type origins: both origins function appeared quite recently in the E. coli lineage (22). How in both Pseudomonas species, and neither functions in Esch- similar or different are initiation and the origin in that erichia coli; enteric origins do not function in either pseudo- have no Dam methylation system? This question is examined monad. Both cloned sequences hybridize to chromosomal here for two such bacteria (23), Pseudomonas aeruginosa fragments that show properties expected ofreplication origins. and Pseudomonas putida. We find that isolated pseudo- These origin plasmids are highly unstable, are present at low monad origins function in both P. aeruginosa and in P. copy number, and show mutual incompatibility properties. putida, but not in E. coli, and the enteric-type origin does not DNA sequence analysis shows that both origins contain several function as an origin in P. aeruginosa. This pseudomonad 9-base-pair (bp) E. coli DnaA protein binding sites; four of origin class is then the second class of bacterial origins these are conserved in position and orientation, two of which analyzed in detail. The P. aeruginosa origin contains no resemble the R1 and R4 sites of the E. coli origin. Conserved GATC sites and the P. putida origin contains only four. Thus, 13-bp direct repeats adjacent to the analogous R1 site are also the abundance ofGATC sites in the origins ofenteric bacteria and only is indeed related to a fundamental (although dispensable) found. No GATC sites are in the P. aeruginosa origin Dam in the normal initiation process. four are in the P. putida origin; no other 4-bp sequence is function of methylation present in high abundance. Both origins are found between MATERIALS AND METHODS sequences similar to the E. coli and Bacillus subtilis dnaA, Bacterial Strains and General Procedures. P. aeruginosa dnaN, rpmHf, and rnpA genes, a gene organization identical to PA02003 argH- recA- (24) and P. putida KT2440 r- m+ (25) that for B. subtilis and unlike that of E. coli. A second were used as sources of chromosomal DNA and for cloning. autonomously replicating sequence was obtained from P. aeru- E. coli K-12 strains were: LE392 hsdR514 supE44 supF58 ginosa that has some properties of bacterial origins. lacYJ galK2 gaI722 metBI trpR55 F- (26), JM101 A(lac-pro) supE thi/F' traD36 proAB+ 1aCIq A(lacZ)M15 (27), JM83 ara Central to the process regulating prokaryotic DNA replica- A(lac-pro) rpsL thi [480 A(lacZ)M15] (28), and JZ294 polAI tion are the events that occur at the chromosomal replication argH hsdR rpsL thyA36 (29). Pseudomonas growth, plasmid origin, oriC. The oriC regions of Escherichia coli and other isolation, and transformation were as described (30). Anti- enteric species have been isolated as autonomous replicons biotics used were tetracycline (100 Ag/ml) for P. aeruginosa (1, 2), functional as replication origins in E. coli. The oriC and P. putida and carbinicillin (500 tug/ml) for P. putida. E. region of the gram-positive bacterium Bacillus subtilis has coli growth and transformation, stability measurements, and been identified and analyzed (3-7). Enteric bacterial origins recombinant DNA methodologies were essentially as de- differ from each other primarily in base substitutions that scribed (31). Nested deletion series fragments for dideoxy DNA sequence determination were generated as described preserve orientation and spacing of the conserved elements and (for a review, see ref. 8). Within the minimal origin [-245 (32), and the two-dimensional method of Brewer Fang- base pairs (bp)], five 9-bp direct and inverted repeats, the "R man (33) was used to probe Pseudomonas chromosomal sites," show sequence conservation and identical spacing DNA with cloned putative origins. the ofDnaA Copy Number Determination. The Yee and Inouye two- and orientation (2, 8) and are required for binding dimensional DNA fingerprint protocol (34) was modified to a protein to DNA (9, 10). DnaA protein also mediates strand horizontal gel format with thin combs and by use of simul- opening at three conserved 13-bp direct repeats, prior to taneous in situ digestion with the second of the assembly of a "prepriming complex" (11). Each of the unseparated adjacent first-dimension lanes sealed in a Ziploc analyzed enteric minimal origins contains 9-14 GATC sites, bag (Dow). Copy number per chromosome was estimated by and eight are positionally conserved (2). comparing plasmid mass (easily identifiable) with differing This conservation of GATC sequence, the recognition site molecular weight chromosomal fragments by using an LKB for the dam gene product, a DNA methylase (12), Ultrascan densitometer. implies an important functional role for Dam methylation in initiation. Evidence indicates a role for Dam methylation in RESULTS timing between initiation events (13-17) and in segregation of Origin Cloning and Subclone Isolation. Complete Pst I, daughter chromosomes (18); however, GATC function in EcoRI, and HindIII digests of P. aeruginosa or P. putida mismatch repair (for a review, see ref. 19) is different from its chromosomal digests were ligated with pBR322 linearized function in initiation (8, 16, 17). Abbreviations: ORF, open reading frame; ARS, autonomously rep- The publication costs of this article were defrayed in part by page charge licating sequence. payment. This article must therefore be hereby marked "advertisement" *Present address: Department of Biology, FO-31, University of in accordance with 18 U.S.C. §1734 solely to indicate this fact. Texas at Dallas, Richardson, TX 75083. 1278 Downloaded by guest on September 27, 2021 Biochemistry: Yee and Smith Proc. Natl. Acad. Sci. USA 87 (1990) 1279

P E C CHEP pae301 3.0kb Table 1. Incompatibility of origin plasmids and plasmid (pTWY312) copy numbers Resident plasmid and origin source P SpPv B Sp.S KD B SpE Sp E P pae31 0 V'rtl~.' ---- X ..1 A 6.1 kb (pTWY470) New ppu332 pae310 (pTWY322) plasmid pTWY451 pTWY461 pTWY470 pTWY310 E P B SP PvKPv SH BP E pTWY451 0.66 <0.04 ppu332 I II Id 4.35kb pTWY461 0.12 0.20 pTWY470 0.67 0.24 pTWY310 0.17 0.08 - FIG. 1. Pseudomonas-origin fragment restriction maps. Plasmids pTWY301 0.12 0.41 pTWY312 tet and pTWY322 tet were generated by Hae III partial digestion plus addition of Pst I linkers and were inserted into the Pst Copy number: 1.0 0.7 1-2 3-5 I site of pBR322. Plasmid pTWY470 bla carries the 3.35-kilobase (kb) Comparative transformation efficiencies of P. aeruginosa pae310 EcoRI-Sal I fragment inserted into EcoRI/Sal I-digested PA02003 containing the indicated resident plasmids relative to cells pBR322. Plasmids pTWY451 bla and pTWY461 tet carry the ppu332 containing no resident plasmid are shown. Correction is made for EcoRI fragment inserted with identical orientation into the EcoRI site variations in competence as measured by transformation with of pBR322 derivatives harboring inactivated tet (pTWY451) or amp pMW79. (pTWY461) genes. Shading: minimal origin regions. Restriction sites: C, Cla I; D, Dra I; E, EcoRI; H, HindIlI; K, Kpn I; P, Pst I; (36) transformed both E. coli hosts equally well, as did the Pv, Pvu lI; S. Sal I; Sp, Sph I. enteric oriC plasmids pJZ19 (37) and pJZ70 (29). Thus, the cloned Pseudomonas sequences, although not detrimental to with the corresponding restriction enzyme and transformed E. coli viability, do not function as origins in E. coli. into E. coli LE392. Plasmid DNA was isolated from pooled Conversely, the two enteric oriC plasmids failed to transform colonies, and competent P. aeruginosa and P. putida cells either P. aeruginosa or P. putida, whereas the three Pseu- were transformed. Inability ofthe narrow-host-range pBR322 domonas-origin plasmids transformed both species at rates origin to function in pseudomonads (25) was used to select comparable to that observed for pMW79 (36). Thus, Pseu- Pseudomonas autonomously replicating sequences (ARS). domonas candidate oriC plasmids are nonfunctional in E. Of 10 P. aeruginosa transformants, all isolated ARS frag- coli, and E. coli-type oriC plasmids are nonfunctional in these ments overlapped with one of two Pst I inserts called pae301 two pseudomonads, as observed (38). and pae310 (Fig. 1; in plasmids pTWY301 tet and pTWY310 To determine the minimal DNA sequences required for tet). Although a P. putida recA- strain was unavailable, an origin function, deletion derivative plasmids (Fig. 1) were EcoRI ARS fragment called ppu332 (Fig. 1; in plasmid constructed by partial restriction with Hae III, addition of pTWY332 tet) was isolated. This ppu332 insert showed linkers, and insertion into pBR322. The minimum-size deriv- sequence similarity (demonstrated by Southern hybridiza- ative so obtained for pTWY332 spanned Sal I, HindIII, and tion) with the pae310 insert (data not shown). BamHI sites. Functional testing of subclones of the Sal I- No tetracycline-resistant colonies were obtained by trans- BamHI and HindIII-BamHI fragments showed that the Sal I- formation of E. coli JZ294 (29), a polA- strain in which the BamHI fragment, but not the HindIII-BamHI fragment, con- pBR322 origin is nonfunctional (35), with pTWY301, tained a functional origin. Transformation experiments pTWY310, pTWY332, or pBR322, whereas all of these showed that all tested plasmids carrying functional subclones plasmids transformed the isogenic polA+ parent of JZ294, of pae310 or ppu332 were able to replicate within both P. LE392, at similar rates. The broad-host-range vector pMW79 aeruginosa and P. putida but not within E. coli JZ294. Sizes

Symmetric Asymmetric Symmetric Asymmetric Asymmetric Cross-Over Barrier Simple Y Double Y Bidirect. Bidi rect Unidirect Unidirect 1 Unidirect 2 - C>- x < -c z- - <> > C>- Ocm = A. B. C. D. E. 'F. IG. H.

'I/

2Z~~~~-----~ ~ kb2kb 2 2k

kbK~ 1kb1k I L. M.

FIG. 2. Two-dimensional gel analysis ofPseudomonas-origin candidates. (A-I) Predicted streak patterns for different types ofbranched DNA molecules (33, 39). (J) Pst I-digested P. aeruginosa DNA hybridized with the pae310 Pst I fragment. (K) EcoRI-digested P. putida DNA hybridized with the ppu332 EcoRI fragment. (L) EcoRI-digested P. aeruginosa DNA hybridized with the 1.2-kb pae31O EcoRI fragment. (M) Pst I-digested P. aeruginosa DNA hybridized with the pae301 Pst I fragment. Downloaded by guest on September 27, 2021 1280 Biochemistry: Yee and Smith Proc. Nati. Acad. Sci. USA 87 (1990)

of the minimal origins so obtained were 800 bp for pae301, 502 hibit mutual incompatibility, independent of the source spe- bp for pae310, and 657 bp for ppu332. cies. Stability, Incompatibility, and Copy Number of Pseudomo- Two-dimensional DNA fingerprinting was used to deter- nas-Origin Plasmids. In the absence of selection, all tested mine plasmid copy numbers; these were all low (Table 1). Pseudomonas-origin plasmids were highly unstable. Even Some correlation is seen between size-of-origin-region pre- when plated directly from selective medium, about 40% ofP. sent, copy number, level of incompatibility, and plasmid aeruginosa cells initially containing pTWY310 and 12% of P. stability. In general, the lowest copy-number plasmids are putida cells initially containing pTWY332 were tetracycline also the least stable. sensitive. Decrease of plasmid-bearing cells in nonselective Identity ofthe Chromosome Origin with the Cloned Origins. medium occurred at a rate of 10-12% per generation for both In the two-dimensional gel method of Brewer and Fangman cases and at a rate of 15-20% for either pTWY470 in P. (33), branched molecules show retarded mobility in the aeruginosa or pTWY451 in P. putida. These rates are com- second, vertical dimension (shape separation) compared with parable to those observed for enteric origin plasmids in E. coli the first, horizontal dimension (mass separation). When (31). probed, these generate streak patterns characteristic of the Incompatibility (Table 1) was examined via transformation type of branched molecule (Fig. 2 A-I). Pseudomonas chro- interference (lowered transformation by an incoming plasmid mosomal DNA from balanced-growth cultures was probed due to a resident plasmid). Plasmids showed moderately with the cloned ARS DNA. Because of unit copy of chro- strong interference with each other and with themselves. mosomal sequences, attainable signal-to-noise ratios were Transformation efficiencies were depressed from 4-fold to relatively poor. The 1.2-kb EcoRI fragment to the right of the more than 25-fold in all cases except for pTWY461 as resident minimal pae310 origin (Fig. 1) provided a nonorigin sequence plasmid. Thus, all three Pseudomonas-origin plasmids ex- control. A simple arc is observed (Fig. 2L), arguing that this A. P. aeruginosa Autonomously Replicating Sequence: -60 -50 -40 -30 -20 -10 0 10 20 30 -70 CACCGCCGATCACGATTACGTCAAAACGGGTAGGGAAATCCACCACGCACCTCGGCCTGTTCAGAAGCAGGAATTCAGGAAAGCCGGACAGTATAAGGAT 30 Sad3A I I I I a I EooRI I I 31 TTAGCGAACCTGAATGAAGACTCCTGGACAAAAAATAGCCAATCCAGATCGAGGGAGCAAAACGGCTCGGCGGGCTTACAAAAAATAAAAGGAAAGGAAT 130 I sinfI I S 5u3A a 13merl,?? I 13mor-m , 131 GTTTATATCTTTTTAGAACTTGTGTTTATTACTTAGCAGCGGCTTTTCFGTTGATAitTACGTCGAGCCCAGTAAAGACATAGCCTGTAGCGAATGATAA 230 13mer-r RI 231 GGCTGTCCCAAACCCTATCTGGGCCClrG-TTG-A-TA~CGTTTTTTAGCTGTGGATGGAATGGCTCfT-A-TC-AACErGTGGAtTATCAC~AGCTTCGGC 330 I I ApuI I I I I AMIm' ' 331 GGGTTTATCTACCCTCCTGTCCCTGGG( ACGTTTCAGTCGAGCGTATCCGGGCCGTGGCAGGACCGCAGCCAGGTAGGAAGAAGGAGATGGA 430 I I I R4 I I AvaII 431 AAGGCGTGGGAAAGGCTCAGTCGTTGCCGCTTTCCAGCGTATCGATGGATGACGACAGGCGCTCGAAGCGGCGCTTGGCGGAGACCTGGAAATCCCGCTC 530 531 GGAGCCCTCGTTAGCGGGTTTGCGCAGCAGTAGCCGAGCACTGAGCCGCTAGTGCGGTCGTAACTGCCGAACCAGTCGATCATCTGTCCGACCTGGCATT 630 a 9 I I FspI I I a I I Sau3N 631 CGCCGCCGAGCCACACCGCCAGGTTGGCTACGATCTCCATGCTGTAGTGTCCGAGTGCTCGCCGTGGGACGCCTTGGGAGCCGAAGGCGAGGGCCCCGCC 730 I Sau3A I I Apal 10--T ~EWL -50 GGCGCGACGACGCGACAGAACCTGACGGCCGTTCTTGGTG CATACGGGC 0 B. P. aeruginosa Origin of DNA Replication: rpmH-0 AlaArgArgArgSerLeuValGlnArgGlyAsnLysThrA aMetArgAla

1 GCGGAAACCGTGGACGCGAGCGCGCTTGAGGGTGCTGGGTTGGAAAGTACGTTTCATGATTCGGTACCTGGGTTGACGACTTGAGGTCGCAGTGACCCCG 100 ArgPheGlyHisValArgAlaArgLysLeuThrSerProGlnPheThrArgLysMet KI BU 101 TTTAAAGAGACCGGCGATTCTAGTGAAATCGAACGGGCAGGTCAATTTCCAACCAGCGATGACGTAATAGATAGATACAAGGAAGTCATTTTTCTTTTAA 200 I , DrI I I I I I I 13mo-I I 13mer-m Dral 201 AGGATAGAAACGGTTAATGCTCTTGGGACGGCGCTTTTC CAT TCGATGAAGCCCAGCAATTGCGTGTTTCTCCGGCAGGCAAAAGGTTGTCGA 300 I I 1 , 13mer-r I I a RI I I I 301 GAACCGGTGTCGAGGCTGTTTCCTTCCTGAGCGAAGCCTGGGGATGAACGAGATGCTACCAA|CGGTlTTTT AA|GGCTGTGCGCAGGGATGTA 400 401 CCCCCTTCAAAGCAAGGGfrTTCAC|^GTCCAGGACGACCGTCCGTCGGCCTGCCTGCTTTTATTAAGGTCTTGATTTGCTTGGM CTACGCATC 50O F @ ~~R4 BstNdQ' Ddd MetSerValGluLeuTrP-r 600 :)U1;niI nraF-riST-r-r.AT-A;lf.-TrrnnCcTCCCGCTACAATAGGCGCTTATTTCGTTGTGCCGCCTTTCCAATCTTTGGGGGATATCCGTGTCCGTGGAACTTTGGI fl%.Imn 4r,.& L - & . I I I I 8 I I EcoRV

-50 GCTTGATGGTGCTTGGTTGGAAAGTACGTTTCATGGCGTGTTACCTGGTTT 0 C. P. putida Origin of DNA Replication: rpmH ~- LysIleThrSerProGlnPheThrArgLysMet BaMN

1 GTCGACGACGGGCCGGGATGGCCCCCTTTTTAAGAGACCGGCGATTCTAGAGAAAGCAAGCCCATAGGTCAATTTCCAACCAGTCTTTCCATATAGAGCA 100 101 TGTGATGGACGGTGCCTGTTGATCAGTGCCCAAGGGGTGCTTGATCGGACACACGGATCGGGGACAACATAAAAGGGCTTAAGTT 200 I Su3A Sau3A ''u 113I l'-m3 201 TTTTGAAGAACTTATAACTCTT GTGGATA^CCTT tGTGATA^CCTGCGCTGGCCCATGAATTACGGGGTGTACAGAGTTTTACAACTTTGTTCTGA 300 13mer-re- | ' RI ' ' ' SuA3k 301 TCCCGTGCTGCGCTTGTTCCAATCGTGAGCGAAAGCEGTGGATG F CACCTTTATCCACg-CGGAGiTTAA GCTAAG.GGGTGGGGTTGTGCA 400 ' ' ' ' ' ' ' Ddel_ 500 PPP^oP401 -- 50oR0OP srrelr fr rra5rzr-an50AAAAnrr50TTTTGTCGATAAATGGCTGTTTTGTCGTGGTlCTAACGTGTCCACAI ------T,. - TAGCCCTCATGGTCGGTCGTI*RA4I --~~ R4 I SAu--Wufvimmu%.;bX0 I]_ MetSerValGlu:iLeuTrpGln--dPaA a ra- vavorACenAAC 600 501 rGTGGATAPZTGAACGCTCGACCGGTACAATGGCGGTTTGTiTTGLCCTCATCGCTTvlllI-IAIU'.T- Abbaso vT.-nt I U%3v v .CTTTGGCAGC Cfrl3I I I I DdeI EwRV m

FIG. 3. Pseudomonas-origin DNA sequences. (A) The pae301 origin. Numbering is relative to the EcoRl site. (B) The pae31O origin. The minimal Hae Ill origin is enclosed. (C) The ppu332 origin. The enclosed origin is based on the Sal l-BamHI minimal ppu332 origin and the Hae III minimal pae310 origin. N-terminal ends ofthe dnaA and rpmH genes are shown. Numbering is relative to the Sal I site. Putative DnaA binding sites are enclosed, probable R1 and R4 sites are so labeled, and 13-bp direct repeats are underlined. GenBank (40) accession numbers: M30124 (A), M30125 (B), and M30126 (C). Downloaded by guest on September 27, 2021 Biochemistry: Yee and Smith Proc. Natl. Acad. Sci. USA 87 (1990) 1281 fragment is an "ordinary" fragment (Fig. 2A). In contrast, the subtilis (5). (iii) Each origin contains three 13-mer direct pae310 Pst I fragment as probe yielded a pattern of retarded repeats and several 9-bp DnaA binding sites. Two of these streaks characteristic ofgrowing replication bubbles convert- DnaA sites have the same spacing and orientation of the R1 ing to forks (Fig. 2J). The presumptive pae310 origin region and R4 sites in the enteric origin, and the position of the is closely centered within the Pst I fragment; thus, the streak 13-mer direct repeats relative to R1 is the same as for the of forks should be very short (see Fig. 2C). This appears to enteric origins (Figs. 3 and 4). Transformation using isogenic be the case (Fig. 2J). However, there is an abundance of strains and positive and negative control plasmids demon- molecules characteristic of intermediates containing short strated that these origins function as origins in both Pseu- bubble regions, and the number of molecules with bubbles domonas species and not in E. coli, and that the enteric nearly the size of the complete Pst I-fragment is very small. bacterial origin is nonfunctional in P. aeruginosa. The Pseu- Similarly, the P. putida ppu332 EcoRI fragment as probe domonas origins thus comprise a class of bacterial origins yielded a discontinuous streak pattern characteristic of a functionally different from the enteric-origin class. functioning origin (Fig. 2K). Probing with the P. aeruginosa The bacterial origin has not yet been isolated from B. pae301 Pst I fragment yielded puzzling results (Fig. 2M), subtilis as an autonomously replicating sequence. However, somewhat suggestive ofa replication terminus (Fig. 2B). This three regions are likely candidates. The first two, rich in the pattern is neither that of a typical origin-carrying fragment 9-bp DnaA binding sequence, are found between the B. nor that for an ordinary chromosomal fragment. subtilis dnaA-like and rpmH-like open reading frames (ORFs) DNA Sequence Determination. Using deletion-derivative and between the dnaA-like and dnaN-like ORFs (5). Hybrid- origin plasmids (see Fig. 1) as DNA sources, nucleotide ization experiments with initially replicating fragments from sequences were determined for both strands and are shown germinating spores favor the latter as the origin (3), although in Fig. 3. recent mutant studies indicate strong incompatibility be- tween the first region and chromosomal initiation (7). This DISCUSSION incompatibility is overcome by mutations in the 9-bp DnaA binding sites (7). In contrast, similar hybridization experi- Two distinct DNA sequences were isolated from P. aerugin- ments with B. subtitis dnaB37 cells suggest that initiation osa PA02003 and one from P. putida KT2440 that are begins 10 kb downstream toward the rrnO operon (4), and it functional origins in each Pseudomonas species but not in E. may be that B. subtilis uses two closely linked origins (3, 6). coli. The following evidence argues that the P. aeruginosa The E. coli-type and pseudomonad-type origins are to date pae310 sequence and the P. putida ppu332 sequence are the the only two classes of bacterial origins precisely located true bacterial replication origins. (i) These sequences both functionally and physically. hybridize to chromosomal DNA sequences that show prop- The pae301 origin contains only two 13-mer repeats, and erties expected oforigins ofDNA replication (Fig. 2J and K); none of its DnaA sites show nine of nine homology to the this is not true of the pae301 sequence (Fig. 2M). The profile DnaA consensus sequence (Fig. 3). Further, pae301 se- observed for an "ordinary" DNA fragment (Fig. 2L) is quences show no significant similarity to any GenBank (40) decidedly different from those obtained using the three DNA sequences. Thus, although this origin shows more putative origin fragments. None of the profiles is like that features ofbacterial origins than do bacteriophage or plasmid expected for unidirectional initiation events (see Fig. 2 E-G). origins, the pae31O sequence is the likely chromosomal (ii) Sequences of both minimal origin regions show definitive origin. homology to the E. coli and B. subtilis rpmH and rnpA genes Common features of the pseudomonad origin are seen in a to one side and to the E. coli and B. subtilis dnaA and dnaN consensus sequence (Fig. 4), obtained with the PIR ALIGN genes on the other side (T.W.Y., unpublished data). The P. program (42). The pae310 Hae III minimal origin includes putida dnaA-like and upstream sequences agree completely part ofthe rpmH-like ORF (Fig. 3B), whereas the ppu332 Sal with those found by Fujita et al. (41). These upstream I-BamHI minimal origin includes part of the dnaA-like ORF sequences contain the P. putida origin region, but origin (Fig. 3C). No known bacterial origins contain coding se- function was not demonstrated (41). Thus, chromosomal quences. A 73-bp insertion in the ppu332 origin relative to the genes bracket these origins, and these genes are those pae31O origin is an unusual feature. This insertion shows no expected for a genetic organization similar to that of B. significant similarity to GenBank sequences and is unlikely to

1 G..CGAC..... GG.CG..... TG.CCCC. TTT ..AAGAGACCGGCGATTCTAG.GAAA.C.A.C .... AGGTCAATTTCCAACCAG ...... 100 a ttga t cag a ....g -a t t g a gggc c------t gacg c gga g c tt a g a g ccat tctttccatatagagca * I I .m.r 13mer-I 101 AT.. .G.A .A.AT ~~~13mm-rnm .. 200 ...... A.A.AA.GAAG....T...... T. . ____-______gac t at g ag t ca g tc tt t cttttaaag tgtgatggacggtgcctgttgatcagtgcccaaggggtgcttgatcggacacacg cgg g ca c ga a aa a ag ca a aaaaagctt 201 13mea-r__ ..T.GAA ....TTA...CTCTT ..G..G...... TTqGTG.ATAAC...... GCCCA ..AATT.CG.G.T ...C.G...... A.AA...... GA 300 ga a acGg atg gg ac gcgctt cI tc atgaa gc g t t tct c gcaggc a gg cga tt t gaac taa aa tg ataacc RI ct cgctg tg a g g gta a agtttt c ct tct 301 .CC.GTG ..G.G. .TGTT.C ..TC.TGAGCGAA ..CTG.GGATGAA T.TC.AC g G TATCCAC CGG.. .G.....G.G. .GGG.TGT.C. 400 a g tc a gc t ct c gc cgagatg tt t c gctgt c ca a a c t c ct c ct c aa g ag t aacacct ag a ag ctaag g tg t g a 401 .....C.A.. G.TTATCCACA .G..A.A. . .G.C. GCC.... TG.T... AT.AA.G.CT.. .TT..C.TGG.. :CONSENSUSSEQUENCE ccctt aa gcaag g R4 aa Tcc ggacg cc t cgtcg tgcc c ttt t g t tga tg t gg : P. aeruginosa tagcc tc tggtc t_ R4 gg ctt ttcac ga g gaaaa gttt g cg- a t g gtt gt g tt : ;P.pulisa FIG. 4. Pseudomonas-origin sequence comparisons. In the consensus sequence, capital letters indicate identical nucleotides in both origins; no letter indicates different nucleotides in the two origins, and dashes indicate a relative deletion. Sequences within the defined minimal pseudomonad origin (ppu332 Sal I site to pae310 Hae III site) are shown; the probable true minimal origin, based on the positions of 13-mers (underlined), DnaA sites (boxed), the 73-bp relative insertion, and the enteric bacteria origin is enclosed. Other features are as in Fig. 3. Numbering is relative to the Sal I site in the P. putida sequence. Downloaded by guest on September 27, 2021 1282 Biochemistry: Yee and Smith Proc. Natl. Acad. Sci. USA 87 (1990) be in the true minimal origin. The probable true minimal 7. Moriya, S., Fukuoka, T., Ogasawara, N. & Yoshikawa, H. origin is boxed in Fig. 4. Four ofthe eight 9-bp DnaA sites of (1988) EMBO J. 7, 2911-2917. the ppu332 origin align perfectly with each ofthe four present 8. Smith, D. W. & Zyskind, J. W. (1989) in Chromosomes: Eu- karyotic, Prokaryotic, and Viral, ed. Adolph, K. W. (CRC, in the pae310 origin (Fig. 4; boxed sites), as well as to four Boca Raton, FL), Vol. 2, in press. found in the pae301 origin. 9. Fuller, R. S., Funnel, B. E. & Kornberg, A. (1984) 38, These origin regions are A+T-rich (48%, 51%, and 46% for 889-900. pae310, ppu332, and pae301, respectively, versus 35% for the 10. Matsui, M., Oka, A., Takanami, M., Yasuda, L. & Hirota, Y. pseudomonad chromosome), and all three origins abound in (1985) J. Mol. Biol. 184, 529-533. potential inverted repeat structures. The pae310 minimal 11. Bramhill, D. & Kornberg, A. (1988) Cell 52, 743-755. 12. Geier, G. E. & Modrich, P. (1979) J. Biol. Chem. 254, 1408- origin contains no GATC sites and there are only four in each 1413. of the ppu332 and pae301 minimal origin regions; no other 13. Smith, D. W., Garland, A. M., Herman, G., Enns, R. E., 4-bp sequence is present in unusually high abundance. The Baker, T. A. & Zyskind, J. W. (1985) EMBO J. 4, 1319-1327. most abundant are the nonpositionally conserved pairs of 14. Messer, W., Belledes, U. & Lother, H. (1985) EMBO J. 4, sequences GAAA/TTTC and AAAG/CTTT, found 11/7 and 1327-1332. 9/9 times in the pae310 and ppu332 origin regions. No 15. Russell, D. W. & Zinder, N. D. (1987) Cell 50, 1071-1079. 16. Skarstad, K., von Meyenburg, K., Hansen, F. G. & Boye, E. methylation process comparable to enteric GATC methyl- (1988) J. Bacteriol. 170, 852-858. ation is known for the pseudomonads (22, 23). 17. Bakker, A. & Smith, D. W. (1989)J. Bacteriol. 171, 5738-5742. Three 13-bp direct repeats in the E. coli origin provide the 18. Ogden, G. B., Pratt, M. J. & Schaechter, M. (1988) Cell 54, site to open oriC duplex DNA for entry of DnaB to 127-135. form a "prepriming complex" (11). Similar direct repeats are 19. Marinus, M. G. (1987) Annu. Rev. Genet. 21, 113-131. found in other enteric origins (2) as well as in the B. subtilis 20. Marinus, M. G., Carraway, M., Frey, A. Z., Brown, L. & Arraj, J. A. (1983) Mol. Gen. Genet. 192, 288-289. origin region (5). One of two sets of positionally conserved 21. Parker, B. & Marinus, M. G. (1988) Gene 73, 531-535. direct repeats in the pseudomonad-origin regions (Fig. 4) may 22. Barbeyron, T., Kean, K. & Forterre, P. (1984) J. Bacteriol. 160, serve a similar initiation function. The first set (positions 586-590. -70, -31, and 37) is highly conserved between the two 23. Brooks, J. E., Blumenthal, R. M. & Gingeras, T. R. (1983) origins but is outside the minimal Sal I-BamHI ppu332 origin Nucleic Acids Res. 11, 837-851. and includes the 73-bp relative insertion in the P. putida 24. Chandler, P. M. & Krishnapillai, V. (1974) Mutat. Res. 23, 15-23. origin sequence. The second set (positions 156, 180, and 205; 25. Bagdasarian, M. & Timmis, K. N. (1982) Curr. Top. Microbiol. underlined in Figs. 3 and 4) is a highly A+T-rich region, and Immunol. 96, 47-67. the spacing between these repeats and to the DnaA sites 26. Murray, N. E., Brammer, W. J. & Murray, K. (1977) Mol. mimics those of the enteric origins (8, 11). Gen. Genet. 150, 53-61. Fundamental features of any bacterial origin likely then 27. Messing, J. (1983) Methods Enzymol. 101, 20-78. include: (i) presence of at least four 9-bp DnaA sites; (ii) 28. Vieira, J. & Messing, J. (1982) Gene 19, 259-268. 29. Harding, N. E., Cleary, J. M., Smith, D. W., Michon, J. J., spacing of -180 bp between two of these (R1 and R4), which Brusilow, W. S. A. & Zyskind, J. W. (1982) J. Bacteriol. 152, are inverted repeats; and (iii) presence of three or more 13- 983-993. to 16-bp A+T-rich direct repeats adjacent to the R1 site. 30. Potter, A. A. (1985) in Recombinant DNA Methodology, eds. Spacing between the 13-bp repeats and between the DnaA Dillon, J. R., Nasim, A. & Nestmann, E. R. (Wiley, New sites, plus sequence differences and DNA modification York), pp. 147-156. mechanisms, may then the host for each 31. Zyskind, J. W., Deen, L. T. & Smith, D. W. (1979) Proc. Nail. provide specificity Acad. Sci. USA 76, 3097-3101. class. 32. Dale, R. M. K., McClure, B. A. & Houchins, J. P. (1985) Plasmid 13, 31-40. We thank B. J. Brewer for detailed protocols and preprints, and 33. Brewer, B. J. & Fangman, W. L. (1987) Cell 51, 463-471. D. 0. Wood for plasmid pMW79. This work was supported by 34. Yee, T. & Inouye, M. (1982) J. Mol. Biol. 154, 181-196. National Institutes of Health Grant GM31839 to D.W.S. 35. Kingsbury, D. T. & Helinski, D. H. (1973) J. Bacteriol. 114, 1116-1124. 1. Yasuda, S. & Hirota, Y. (1977) Proc. Natl. Acad. Sci. USA 74, 36. Wood, D. O., Hollinger, M. F. & Tindol, M. B. (1981) J. 5458-5462. Bacteriol. 145, 1448-1451. 2. Zyskind, J. W., Cleary, J. M., Brusilow, W. S. A., Harding, 37. Zyskind, J. W. & Smith, D. W. (1980) Proc. Natl. Acad. Sci. N. E. & Smith, D. W. (1983) Proc. Natl. Acad. Sci. USA 80, USA 77, 2460-2464. 1164-1168. 38. O'Neill, E. A. & Bender, R. A. (1988) J. Bacteriol. 170, 3. Ogasawara, N., Mizumoto, S. & Yoshikawa, H. (1984) Gene 3774-3777. 30, 173-182. 39. Brewer, B. J. & Fangman, W. L. (1988) Cell 55, 637-643. 4. Sdror-Laurent, S. J. & Henckes, G. (1985) Proc. Natl. Acad. 40. Bilofsky, H. S. & Burks, C. (1988) Nucleic Acids Res. 16, Sci. USA 82, 3586-3590. 1861-1864. 5. Moriya, S., Ogasawara, N. & Yoshikawa, H. (1985) Nucleic 41. Fujita, M. Q., Yoshikawa, H. & Ogasawara, N. (1989) Mol. Acids Res. 13, 2251-2265. Gen. Genet. 215, 381-387. 6. Levine, A., Henckes, G., Vannier, G. & Sdror, S. J. (1987) 42. George, D. G., Barker, W. C. & Hunt, L. T. (1986) Nucleic Mol. Gen. Genet. 208, 37-44. Acids Res. 14, 11-15. Downloaded by guest on September 27, 2021