THE JOURNAL OF BIOLOGICAL CHEMISTRY _ © 1992 by The American Society for Biochemistry and Molecular Biology, Inc. Vol. 267, No. 13, Issue of May 5, pp. 8778-8784, 1999 Printed in U.S.A. Features of Replication Fork Blockage by the Escherichia coli Terminus-binding Protein* Eui Hum Lee and Arthur Kornberg (Received for publication, July 15, 1991) From the Department of Biochemistry, Stanford University School of Medicine, Stanford, California 94305-5306 Blockage of the progress of a DNA replication fork in Escherichia coli can be ascribed to an inhibition of helicase action at the orientation-specific binding of a termination sequence (ter) by the ter-binding protein (Lee, E. H., Kornberg, A., Hidaka, M., Kobayashi, T., and Horiuchi, T. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, 9104-9108). These observations have been ex- tended to include the PriA helicase, thus confirming that blockage is general for helicases. The site of arrest of synthesis by a replication fork is at the very first nucleotide of the 22-base pair FE. coli-terB sequence. Strand displacement by DNA polymerases is also in- hibited, but is less profound and is orientation-specific. The ter sequences of plasmids Rl-terR and -terL and of plasmids R6K and R100 have been compared with those of E. coli-terA and -terB. In the replication of the circular Escherichia coli chromo- some, two forks move bidirectionaily from the unique origin (oriC) and terminate in a region that contains five terminus (ter) sites that are inverted repeats of a 22-bp' sequence (1, 2). Binding of ter by the E. coli terminus-binding protein (TBP), encoded by the tau (or tus) gene, is essential (3-5). One orientation of the bound sequences (terB and terC) blocks the clockwise (CW)-moving fork, whereas the other sequences (terA, terD, and terE) block the counterclockwise (CCW) fork (2). Homologous functional and structural ter sites have also been demonstrated in E. coli plasmids R6K (6, 7), R1 (7, 8), and R100 (8) and in Bacillus subtilis (9, 10). Orientation- specific blockage of the replication forks observed in vivo (3, 4,7, 8) has been observed with a reconstituted purified enzyme system for replication of the minichromosomal oriC plasmid (11). The component in the system most immediately affected is the DnaB helicase (12, 18). However, the obstruction to helicase action appears to be general inasmuch as two other helicases are also blocked in an orientation-specific manner (11). This study was undertaken to explore other features of the TBP-ter complex (its capacity to block the action of the PriA helicase and strand displacement by DNA polymerases), to determine precisely the location of the blockage in or near the bound ter sequence, and to compare the relative blocking strengths of the various ter sequences. * This work was supported by Grants GM07581 and AG02908 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accord- ance with 18 U.S.C. Section 1734 solely to indicate this fact. *The abbreviations used are: bp, base pair(s); TBP, terminus- binding protein; CW, clockwise; CCW, counterclockwise; SDS, so- dium dodecyl sulfate; Hepes, 4-(2-hydroxyethyl)-1-piperazineeth- anesulfonic acid; kb, kilobase(s). EXPERIMENTAL PROCEDURES Strains, Enzymes, and Other Reagents—E. coli strains used were DH5e (F°, ¢80lacZA115, endAl, recAl, hrs”, hsm*, supE44, thi-1, gyrA, d-, A(lacZ YA-ArgF), U169) and JM83 (ara, A(lac-proAB), rpsL, @80lacZA115). Restriction endonucleases were Accl, Clal, Dral, EcoRV, Nael, PstI, Smal, Sfil-, Sacl, SacII, and SnaBI; and other enzymes used were T4 DNA ligase, T4 DNA polymerase, T4 poly- nucleotide kinase, and DNA polymerase I (large fragment), all from New England BioLabs, Inc. Calf intestine alkaline phosphatase was from Boehringer Mannheim. T7 DNA polymerase was from U. S. Biochemical Corp. T5 DNA polymerase was a gift from Dr. R. K. Fujimura (Oak Ridge National Laboratory). The enzymes used in the oriC reconstitution assay were as described (11, 12, 14). Other re- agents were unlabeled deoxynucleoside and ribonucleoside triphos- phates (Pharmacia LKB Biotechnology Inc.(, [y-"’P]ATP (6000 Ci/ mmol, 1 Ci = 38 GBq) and [a-*=P]dTTP (800 Ci/mmol) (Amersham Corp.), and bovine serum albumin (Pentex). TBP buffer contained 550 mM NaCl, 50 mM Tris-HCl (pH 7.5), 20% (v/v) glycerol, 1 mm EDTA, and 1 mM dithiothreitol. Plasmids, Phage DNA, and Synthetic Oligonucleotides—Plasmids pTZ18R and pTZ19R were from Pharmacia. pTB101 contains a 678- bp Hincll-PstI fragment spanning oriC (positions —189 to +489) of pCM959 (15) cloned in the pBluescript vector (Stratagene). Plasmid pKHG300 (obtained from Dr. Horiuchi, Kyushu University, Fukuoka, Japan) is a pUC9-based vector carrying the tau gene and overproduces TBP >2000-fold (3). M13mp18 and M13mp19 replicative form DNAs were from New England BioLabs, Inc. M1l3mp18-.ter CCW and M13mp19.ter CW DNAs were as described (11). Synthetic oligomers were synthesized by Alan Smith in the Protein and Nucleic Acid Facility at Stanford University. In Vitro oriC Reconstitution Assay: Coupled and Staged—The cou- pled reconstituted oriC DNA replication reaction was essentially as described (11). In brief, the reaction mixture (40 ul) contained 30 mm Hepes/KOH (pH 7.6), 8 mM magnesium acetate, 2 mM ATP, template DNA (600 pmol as nucleotide), 0.8 mM each CTP, GTP, and UTP, 0.1 mm each dATP, dCTP, dGTP, and dTTP, 16 nm [a-“P]TTP (100-300 cpm/pmol), 150 mM potassium glutamate, 80 ng of DnaA, 900 ng of E. coli single-stranded DNA-binding protein, 8 ng of DnaB, 27 ng of DnaC, 50 ng of gyrase A, 30 ng of gyrase B, 8 ng of HU protein, 17 ng of primase, 100 ng of DNA polymerase IIT holoenzyme, and 26 ng of the 8-subunit of DNA polymerase III holoenzyme. After incubation at 30 °C for 30 min, reactions were stopped by adding 100 ul of 100 mM pyrophosphate containing 20 ug/ml carrier DNA (calf thymus). The extent of DNA synthesis was measured by liquid scintillation counting after precipitation with 10% (w/v) trichloroa- cetic acid and filtration on Whatman GF/C glass filters. Progress of bidirectional DNA replication was assayed in a staged reaction. The prepriming complex was formed in 20 yu] containing 40 mM Hepes/KOH (pH 7.6), 15% (w/v) glycerol, 200 ng/ml bovine serum albumin, 0.007% Brij 58, 0.25 mm EDTA, 200 mM potassium glutamate, 3.3 mm ATP, template DNA (600 pmol as nucleotide), 8 ng of HU protein, 8 ng of DnaB, 27 ng of DnaC, and 80 ng of DnaA. After 20 min at 37 °C followed by 2 min at 18 °C, 20 ng of TBP (or TBP buffer as a control) was added and incubated for 1 min. The omitted replication components (primase, gyrases A, and B, polym- erase III holoenzyme, §-subunits, E. coli single-stranded DNA-bind- ing protein, GTP, CTP, UTP, dATP, dCTP, dGTP, dTTP, and [a- *P]dTTP), in a volume of 10 yl that also contained magnesium acetate at 10 mM and potassium glutamate at 200 mM, were added. Incubation was continued for 2, 4, and 6 min; and the reactions were stopped by adding EDTA to 20 mm. A small portion of each sample 8778 Replication Fork Blockage in E. coli was used to determine DNA synthesis as described above. The re- mainders of the samples were extracted twice with phenol and twice with chloroform and isoamyl alcohol, (24:1) and precipitated with ethanol. The DNA pellets were redissolved in buffer containing 10 mM Tris-HCl (pH 8.0), 0.2 mm EDTA and digested with the desired restriction endonucleases following the manufacturer’s instructions. One unit of DNA replication activity promotes the incorporation of 1 pmol of nucleotide/min at 30 °C. Purification of TBP—All operations were at 4°C. E. coli strain JM83 (pKHG800) (3) was grown in 1.5 liters of L-broth to optical density (Acoonm) of 0.5, harvested by centrifugation in a Beckman JA- 10 rotor for 10 min at 6000 rpm, and frozen in liquid nitrogen; frozen cells were resuspended in 20 ml of lysis buffer containing 50 mM Tris-HCl (pH 7.5), 10% sucrose, 5 mM dithiothreitol, 10 mm EDTA, 100 mM NaCl, 20 mM spermidine HCl (pH 7.5), (NH4)2SO, to 5% saturation, and 4 mg of lysozyme and were incubated on ice for 45 min. Cells were incubated at 37°C for 5 min (inverting tubes once every minute), incubated on ice for 10 min, and centrifuged in a Sorvall SS-34 rotor for 1 h at 14,000 rpm (Fraction IJ, 20 ml). Ammonium sulfate (0.313 g/ml) was added during a 30-min period to Fraction I; stirring was continued for another 30 min. The precipitate was collected by centrifugation in a Sorvall SS-34 rotor for 20 min at 18,000 rpm and resuspended in 4 ml of buffer A (50 mm Tris-HCl (pH 7.5), 20 mm NaCl, 20% glycerol, 1 mm dithiothreitol, and 1 mM EDTA) (Fraction II, 4.2 ml). Fraction II was dialyzed against buffer A overnight and applied at 7.2 ml/h to a 13-ml column (2.6 cm high, 2.5-cm diameter) of DEAE-Sephacel equilibrated with buffer A. Flow- through fractions that contained TBP were collected (Fraction III, 21.8 ml). Fraction III] was applied at 7.2 ml/h to a 15-ml heparin- agarose column (3.0 cm high, 2.5-cm diameter) equilibrated with buffer A. The column was washed with 30 ml of buffer A plus 280 mM NaCl prior to elution using a linear gradient of 280-980 mM NaCl in buffer A (45 ml). The peak of TBP activity collected in 12- ml fractions eluted between 500 and 700 mM NaCl. Protein concen- trations were measured by the Bradford assay (28). One unit of TBP causes a 50% inhibition of replication of the oriC-ter CW-CCW plasmid (Fig. 1c) in vive in 1 min at 30 °C. The purification data are shown in Table I. Purified TBP is near homogeneity (>95%) as judged from Coo- massie Blue staining of an SDS-polyacrylamide electrophoresis gel (16) and binds to the ter site in a 1:1 molar stoichiometry ratio when assayed by oriC reconstitution in vitro. Purified TBP does not inhibit DNA replication of oriC plasmids lacking ter sites even at a 30-fold molar excess. Preparation of DNA Polymerase Substrates—A 64-mer oligonucle- otide (40 ng) containing the complementary ter sequence was 5’-end- labeled with **P using T4 polynucleotide kinase (17) and mixed with M13mp18-ter CCW or M13mp19-ter CW single-stranded DNA (10 #g) in 100 wl of annealing buffer containing 10 mm Tris-HCl! (pH 8.0), 100 mm NaCl, and 1 mM EDTA. The mixtures were heated at 100 °C for 2 min, cooled slowly to 50 °C, and incubated for 1 h and then cooled slowly to room temperature. The template, with the oligomer annealed to it, was purified by a Bio-Gel A-5 spun column (17), extracted with phenol and chloroform, and precipitated with ethanol. Annealing of the substrates was confirmed by electrophoresis on a 0.8% agarose gel followed by autoradiography. Similarly, a 30- mer (or 28-mer) oligonucleotide containing the ter sequence was annealed to M13mp18-ter CCW (or M13mp19-ter CW) DNA as described above. Assay for Strand Displacement of DNA Polymerases—The reaction mixture (30 41), containing 10 mM Tris-HCl (pH 7.5), 50 mm NaCl, 0.1 »g of DNA substrate (with annealed 5’-°’P-end-labeled ter oligo- mer, 6000 cpm), and 0.2 ug of M13 sequencing primer (17-mer), was incubated at 37°C for 30 min and cooled to room temperature. The Teaction volume was brought to 100 ul by adding 50 mm Tris-HCl (pH 7.5), 10 mM MgSO, 1 mm dithiothreitol, 50 ug/ml bovine serum albumin, and 250 uM each dATP, dCTP, dTTP, and dGTP. Samples i. divided into two portions. To one was added 30 ng of TBP; to ane other was added TBP buffer as a control. After incubation at . C for 1 min, DNA polymerase was added; and at indicated time of portions of each sample were transferred to an equal volume 0 oon solution containing 60 mm EDTA, 1% SDS, 20% glycerol, el @ bromphenol blue, and 0.02% xylene cyanol. Samples were and ephoresed on a nondenaturing 7% polyacrylamide gel, dried, ol autoradiographed. Slices of the dried gel that contained undis- ae (substrate) and displaced (oligomer) bands were cut out; their be l0activity was measured in a liquid scintillation counter; and the Teentage of displaced oligonucleotide was calculated. The DNA 8779 ) (b) ro) (Nael) “a, Snabi SacT AP pBSoriC PCM959 oriC-terCW E.coli-ter A E.coli-terB R6K-terR R6K-terL Ri-terA Ri-terL R100-terA R100-terL B.Subtilis-teril oriC-terCW:-CCW R6K-terF tert. Ri-terA terL oriC-terCW:CCW E.coli-terBterB Fic. 1. Construction of oriC-ter plasmids. Pairs of comple- mentary synthetic oligomers were annealed to generate ter duplexes with cohesive ends for ClaI and EcoRV restriction endonucleases. (i) E. coli-terA, 5'-ATCAATTAGTATGTTGTAACTAAAGTAT-3’ (28 mer) and 5’-CGATACTTTAGTTACAACATACTAATTGAT-3’ (30-mer); (ii) R100-terR, 5’-ATCATTATGAATGTTGTAACTAC- TTCAT-3’ (28-mer) and 5’-CGATGAAGTAGTTACAACATTCA- TAATGAT-3’ (30-mer); (iii) R100-terL, 5’-ATCTGTCTGAGTGT- TGTAACTAAAGCAT-3’ (28-mer) and 5’-CGATGCTTTAGTT- ACAACACTCAGACAGAT-3’ (30-mer); and (iv) B. subtilis-terlIT, 5’- ATCATTGAATATTTAGTACATAGTGTAT-3’ (28-mer) and 5’-C- GATACACTATGTACTAAATATTCAATGAT-3’ (30-mer). To gen- erate a unique Clal restriction site in pTB101 DNA (15), the addi- tional Clal site (at nucleotide 764) was inactivated by T4 DNA polymerase treatment after partial digestion of the plasmid DNA with Clal. a, the ter sequences listed were inserted into a plasmid vector cleaved with Clal and EcoRV restriction enzymes to generate the oriC-ter CW plasmids. Other oriC-ter plasmids were constructed by inserting the ter DNA fragments into pTB101 DNA; E. coli-terB, 204-bp Accl-PstI DNA fragment of the oriC-ter CW plasmid (11); R6K-terR, 74-bp Alul-HaelJI DNA fragment of pUC-R6K-ter (26); R6K-terL, 142-bp Alul-HaeIII DNA fragment of pUC-R6K-ter, R1- terR, 195-bp Dral-Sfil DNA fragment of pHM6050 (27); and R1-terL, 105-bp Sfil DNA fragment of pHM6050 (27). The above DNA frag- ments containing ter sites were blunt-ended with T4 DNA polymerase and inserted into the EcoRV site of pTB101 DNA to generate oriC- ter CW plasmids. b, the R6K-ter sites (terR and terL) containing the 216-bp AluI DNA fragment of the pUC-R6K-ter plasmid (26) and the Rl-ter sites (terR and terL) containing the 554-bp Dral-EcoRV DNA fragment of pHM6050 (27) were isolated, blunt-ended, and inserted into the Nael site of pTB101 to generate oriC-ter CW - CCW plasmids. c, the oriC-ter CW-CCW (E. coli-terB-terB) plasmid was constructed by inserting the synthetic ter oligomers used for construction of oriC- ter CCW plasmids into the oriC-ter CW plasmid (11) that was cleaved with SacI and SacII restriction enzymes. The T-like symbol represents a ter sequence oriented such that the head of the T should block the progress of the replication fork. TABLE I Purification of TBP Steps oe Total activity Specific activity Yield mg units x 10° units x 10°°/mg % I. Lysate* 210 Il. Ammonium sulfate 113 14 1.23 (100) III. DEAE-Sephacel 11.8 11 9.3 78 IV. Heparin-agarose 0.96 8 83 57 * The activity of this fraction could not be determined accurately. polymerases used were DNA polymerase I (large fragment), (5 units), T5 DNA polymerase (4.4 units), and T7 DNA polymerase (2 units). Construction of PriA Helicase Substrates (See Fig. 8)—The 56-mer phage #X174 primosome assembly site (pas) recognized by priA protein (n’ protein) was synthesized with an extension of 4 residues at the 5’-end to generate the 60-mer: 5’-CCTTGAGGTTATTA- ACGCCGAAGCGGTAAAAATTTTAATTTTTGCCGCTGAGGG- GTTGACC-3’. This 60-mer was annealed (by a 4-bp match) and 8780 ligated to the pair of 36-mer complementary synthetic oligomers containing the 22-bp E. coli-terB sequence: oligomer 1, 5’- AAGGGGCACATAATAAGTATGTTGTAACTAAAGTGT-3’; and oligomer 2, 5/-TCACACACTTTAGTTACAACATACTTAT- TATGTGCC-3’. To verify the ligation, the 96-mer was isolated by gel electrophoresis in 7 M urea, 10% polyacrylamide and then an- nealed with a 5’-°*P-end-labeled complementary ter 36-mer (oligomer 1) to generate the CCW ter duplex, the sequence of which is oriented to block the PriA helicase movement from pas. In a similar way, the oppositely oriented E. coli-terB duplex substrate was prepared from the pair of synthetic ter oligomers: oligomer 3, 5’-AAGGGGCACAT- CATTTAGTTACAACATACTTATTGT-3’; and oligomer 4, 5’- TCACACAATAAGTATGTTGTAACTAAAGTATGTGCC-3’. The T-like symbol (see Fig. 8) enclosing the ter sequence is oriented such that the head of the T blocks the progress of the replication fork. Two-dimensional Gel Electrophoresis—In vitro replicated DNA samples (3000-20,000 cpm) were cut with EcoRV and SnaBI restric- tion endonucleases and electrophoresed in two dimensions as de- scribed (18). The first dimension was in 0.4% agarose-for 18 h at 1 V/cm at 4 °C in buffer containing 37 mM Tris borate (pH 8.3), 1 mM EDTA; DNA size markers (HindIII-digested ) DNA) were included. The second dimension was in 1% agarose for 4 h at 8 V/cm at 4°C with 0.3 ng/ml ethidium bromide. Gels were dried, autoradiographed, and scanned by densitometry. SDS, Agarose, and Polyacrylamide Gel Electrophoresis—-SDS-poly- acrylamide gel electrophoresis was as described (16). HaellI and EcoRV restriction endonuclease digests of replicated DNA and reac- tion samples from strand displacement of DNA polymerase assays were brought to 0.5% SDS, 10% glycerol, 0.01% bromphenol blue, and 0.1% xylene cyanol and were loaded on 7% polyacrylamide gel (acrylamide/N,N’-methylenebisacrylamide ratio of 30:1). Gels were electrophoresed in buffer containing 37 mm Tris borate (pH 8.3), 1 mM EDTA for 4 h at 10 V/cm, dried, autoradiographed, and scanned by densitometry. Also, slices of dried bands were measured for their radioactivity in a liquid scintillation counter. Agarose and polyacry]- amide gel electrophoresis were essentially as described (17). DNA Sequencing—DNA sequencing was by the dideoxynucleotide sequencing method (19) using the Sequenase system (U. S. Biochem- ical Corp.) according to the instructions of the manufacturer. In brief, - the complementary ter oligomers (100-fold molar excess) and se- quencing primer were annealed to template substrates by boiling for 2 min and cooling to room temperature slowly; a 20-fold molar excess of TBP was added; and the reaction mixture was incubated at room temperature for 1 min. The rest of the procedure was as described by the manufacturer. RESULTS Effectiveness of Several ter Sequences in Inhibition of DNA Replication—Among the 17 identified ter sequences (7-9), the capacity to block DNA activity in vitro has been reported for only the E. coli-terB (5, 11) and R6K-terR (20) sequences. oriC plasmids containing these or seven other ter sequences (Fig. 1a) were compared at an equimolar ratio of TBP to the ter sequence. Those of E. coli and R1 were more effective than those of R6K and R100 (Table II). The dissociation constant (Kp) for E. coli-terB is 10°” mM (5 x 10° M for R6K-terR) (1). Possibly the 2-3-fold stronger inhibition of DNA replication by £. coli-terB compared with R6K-terR (Table II) reflects differences in their affinity for TBP. However, Kp values for the other ter sequences have yet to be determined. The incomplete blockage of replication can be attributed to the bidirectional mode of oriC replication (11). Thus, a ter se- quence oriented to block the CW fork fails to prevent move- ment of the CCW fork, and vice versa. We tested B. subtilis terII sequence, mindful of the fact that the E. coli TBP is twice the size of that from B. subtilis (14.5 kDa) (21). An inhibition of only 10% was observed at a concentration 30-fold higher than that of the EF. coli TBP (data not shown). Limits of Progress of Replication Forks in oriC-ter Plas- mids—The extent of the movement of a replication fork with an oriC-ter plasmid, with or without TBP, was analyzed by Haelll and EcoRV restriction enzyme cleavage as previously Replication Fork Blockage in E. coli TABLE II Comparative effectiveness of several ter sequences in the inhibition of DNA replication Plasmid DNA synthesis, +TBP? % pBS-oriC 97 E. coli-terA 35 E. coli-terB 23 R6K-terR 60 R6K-terL 64 Rl-terR 33 Rl-terL 34 R100-terR 58 R100-terL 57 “For each plasmid, the value for DNA synthesis with TBP js compared to that without it. Values, determined after incubations of 15 and 30 min, were similar; and the average is given. All the ter plasmids contained the ter sequence in pBS-oriC (Fig. 1a). The coupled reconstitution assay (see “Experimental Procedures”) con. tained 3 ng of TBP and 75 fmol of DNA, as plasmid. The DNA synthesis of oriC-ter plasmids averaged 200 and 400 pmol for 15- and 30-min incubations, respectively, in the absence of TBP. terCW-CCW TBP None + None ~- DNA SYNTHESIS (pmol) 2 a Fs +4+ E.coll-8-B TIME (min) Fic. 2. Inhibition of DNA synthesis of oriC-ter CW.CCW plasmids by TBP-ter complex. The DNA substrate (75 fmol as circle) was used with or without TBP (30 ng). All plasmids were oriC with ter CW.CWW sequences, except where indicated (None). The ter sequences were a pair of E. coli-terB, a pair of Rl-terR and -terL, and a pair of R6K-terR and -terL, as indicated. described (11). In the absence of TBP, the fragments gener- ated by digestion of the oriC-ter plasmids indicated a bidirec- tional mode starting at or near oriC (Fig. 3), a pattern similar to the symmetrical bidirectional progress for the control oriC plasmid, whether or not TBP was present (11). With TBP, replication of an oriC-ter CW plasmid was unidirectional and blocked in the CW direction; with the oriC-ter CCW plasmid, movement was blocked in the CCW direction (Fig. 3). In these replications, full-length DNA appears within 2 min (11, 12). Thus, there was ample time for a replication fork to proceed through a CW ter site from the CCW direction, and vice versa. This problem was circumvented by constructing oriC-ter CW: CCW plasmids (Fig. 1, b and c), which blocked replication fork movements from oriC in both directions. Replication of a plasmid with ter sequences oriented in opposite directions on both sides of oriC was almost com- pletely inhibited (Fig. 2). The three constructions included 4 plasmid with two E. coli-terB sequences, one with Rl-terR and -terL, and one with R6K-terR and -terL, each pair S° oriented as to prevent both CW and CCW movements. Rep- lication was sharply limited by the bound ter sequences to the oriC region (Fig. 3). Replication Fork Blockage in E. coli 8781 = wo Li19, eau a Ee Mm oR AS tu sagan ye pel mM om ORS U0 ae Near mM oR oriC oric oric ortterCWw 20 15 10 5 0 LABEL IN FRAGMENT (% of total) oriC-terCWsTBP 1 2 3 LENGTH (kb) oriC-terCCW oriC-terCwCCW 20 15 10 0 oriC-terCCW+TBP oriC-terCWCCW+TBP LENGTH (kb) LENGTH (kb) Fic. 3. Progress of replication forks of oriC and oriC-ter plasmids. The oriC in vitro reconstitution assay was staged, and the replication products were separated on 10% glycerol, 7% polyacrylamide gels and identified by autoradiography. DNA fragments were generated by HaeJII and EcoRV cleavage of a 4-min DNA replication product; replication fragments RO-R3 are in the CW order from oriC; fragments LO-L10 are in the CCW order. TBP (30 ng) was in 10-fold molar excess over plasmid DNA (75 fmol as circle). The autoradiograph was scanned by densitometry, and the peaks were cut out and weighed. These intensity measurements were corrected for fragment size, and the percent of the total intensity of the lane in each fragment was calculated. The positions of minimal oriC and ter (T-shaped area) are shown. The fragments generated by HaellI and EcoRV digestion were as follows: RO, 785 bp; R1, 250 bp; R2, 926 bp; R3, 216 bp; LO, 785 bp; L1, 108 bp; L2, 9 bp; L3, 448 bp; L4, 484 bp; L5, 76 bp; L6, 85 bp; L7, 374 bp; L8, 22 bp; L9, 45 bp; and L10, 204 bp. The DNA size of L3 is 262 bp for oriC-ter CCW and oriC-ter CW.CCW plasmids. (a) FIRST DIMENSION ——— ere SPOT 1 LINEAR Fic. 4. Analysis of replicated molecules by two-dimensional gel electrophoresis. The EcoRV- and SnaBl-digested replication products were analyzed as described under “Ex- perimental Procedures.” a, the diagram SECOND DIMENSION 1kb 4 2.87 kb == BUBBLE SPOT3 LINEAR SPOT 4 Y AND DOUBLE Y ——___ indicates the sizes and migration pat- terns expected of the products (18). Spot J, oriC-containing linear DNAs: 1.14 kb from oriC, 1.16 kb from oriC-ter CW, 0.92 kg from oriC-ter CCW, and 0.95 kb from oriC-ter CW-CCW; spot 2, repli- cation intermediates of spot 1 DNAs (bubble-shaped molecules); spot 3, 2.87- kb linear DNAs; spot 4, replication in- termediates of spot 3 DNAs (double-Y- and Y-shaped). b, two-dimensional gel profile of the replication products after 4 min of synthesis in a staged oriC in vitro assay. TPB, where indicated, was Present in a 10-fold molar excess relative to ter, (b) sig Pes of Replicated Molecules Analyzed by Two-dimen- mn i Gel Electrophoresis—The first dimension separates Olecules based on mass and the second, according to shape n autoradiographs of the two-dimensional gels of kinds - and EcoRV-digested replication products (Fig. 4), two oriC linear DNA molecules were observed from both the oriC-en oriC-ter plasmids replicated in the absence of TBP: inear ouiming small linear molecules (spot 1) and large bubbles ees (spot 3) (Fig. 4). In addition, spot 2 contains small lie aped molecules, the replication intermediates of the near molecules; and spot 4 contains the Y- and double SnaBl oriC-terCCW +7 orlC-terCW +TBP —-TBP oriC-terCW-CCW -TBP +TBP Y-shaped molecules, the replication intermediates of the large linear molecules. In the presence of TBP, a similar pattern was obtained with the control (oriC) plasmid. However, with the ter CW and ter CCW plasmids, spot 2 molecules were enriched 5-6-fold, and the spot I molecules decreased drasti- cally; with the ter CW-CCW (Fig. 1c) plasmid, only spot 2 molecules were detected by autoradiography (Fig. 4). These results further demonstrate that the DNA replication forks were confined between the two ter sites and terminated their DNA synthesis at the ter sites. Exact Location of Replication Block—Since replication forks 8782 do not proceed beyond the ter site, the precise position of the blockage can be determined in a sequencing gel after cleavage with an appropriate restriction endonuclease. Smal digestion of replication products of oriC-ter CW-CCW (Fig. 1c) gener- ated a 120-bp DNA fragment (Fig. 5) that corresponds exactly to the distance from the Smal site to the first nucleotide position of the E. coli-terB (CCW) site. A similar result was obtained from a digestion with AccI (Fig. 5), which yielded a 137-bp fragment that also corresponds to the distance from the first nucleotide position of the E. coli-terB (CW) site to the Acc! site (Fig. 5). In addition to these 120- and 137-bp DNA fragments, others of 106, 463, and 476 bp were seen, presumably generated by synthesis from initiation sites in oriC. To confirm that the 120- and 137-bp DNA fragments are true replication-arrest bands, other restriction enzymes such as HindIll, BamHI, Aatil, and BglII were used. For example, HindIII-Smal double cleavage generated 120-bp (Smal-terB (CCW)), 180-bp (HindIII-terB (CW)), 106-bp, and 290-bp fragments, all of which are readily interpreted as true DNA replication-arrest bands (data not shown). Also deduced from several restriction-enzyme analyses and com- parisons of fragment sizes, two potential initiation sites were located at positions —70 and —100 downstream from oriC for the CW direction and 10 or more potential initiation sites within oriC for CCW synthesis (data not shown). Effect of TBP-ter Complex on Strand Displacement by DNA Polymerase—Three DNA polymerases (T5, T7, and E. coli polymerase I (large fragment)), each lacking a 5’ — 3’ exo- nuclease but able to displace the strand annealed to the template (22), were examined for their ability to dissociate a M M_~ AccI Smal Fic. 5. Site of arrest of DNA replication in oriC-ter CW. CCW plasmid. The oriC in vitro reconstitution assay was staged after 6 min of DNA synthesis as described under “Experimental Procedures.” The replication products of an oriC-ter CW-CCW plas- mid (Fig. 1c) were digested with AccI or Smal and electrophoresed on a7 M urea, 8% polyacrylamide sequencing gel (0.4 mm x 60 cm). The predominant bands observed with Smal were at 120, 106, and 470 bp and with Acc] at 137, 453, and 465 bp. The 120-bp DNA band from Smal and the 137-bp DNA band from Accl were resistant to treatment with 0.05 M NaOH, but other bands were decreased slightly, suggest- ing that the 120- and 137-bp DNA bands are not RNA-primed DNA strands. The two lanes of size markers are indicated (lanes M). Essentially the same experiment was repeated six times with virtually identical results. The figure was from overexposed autoradiograph film to show all possible bands. Replication Fork Blockage in E. coli ter sequence bound by TBP. The template-primers (Fig. 6a) possessed ter sequences oriented to block either the CCW or CW replication forks. Displacement of the annealed oligonu- cleotide was measured electrophoretically. In the absence of TBP, all three DNA polymerases displaced 60-85% of the annealed oligomers (Fig. 6b). In the presence or TBP, strand displacement by DNA polymerases was largely inhibited on both templates (CW and CCW). However, with Mi3mpi8. ter CCW, in which ter is oriented to block fork movement, strand displacement was inhibited twice as much as with M13mp18 ter CW (Fig. 6b). The data also show the inhibition of strand displacement activity of DNA polymerases when the ter sequence is correctly oriented and TBP is present. The location at which the DNA polymerase was blocked by the TBP-ter complex was readily determined with T7 DNA polymerase and chain-terminating dideoxynucleoside triphos- phates. The blockage sites were at the fourth nucleotide from the 5’-end of ter CCW and at the fifth nucleotide from the 5’-end of ter CW (Fig. 7). Thus, this DNA polymerase pene- trates several nucleotides into the complex, unlike the com- plete blockage of the forks observed in replication of the plasmids (Fig. 5). The greater effectiveness of a ter sequence against a replication fork is attributable to blockage of the DnaB helicase and the relative lack of strand-displacing ac- tivity possessed by the DNA polymerase II holoenzyme (23). Penetration of T7 DNA polymerase 3 bp from one end of the 22-bp ter sequence and 4 bp from the other leaves a 15-bp (a) M13mp13 (b) 100 DNA Polymerase = (TEP) 13 (TEP) T7 DNA Polymerase (TBF) (Large Fragment) ONA Polymerase 4 CW) 0 omy cow) ccw(-) cow) cw} r) L Cw} a - L, cCow(+) cw) CWle} cows) CCW(s) 20 oft tt if Jit tbo} @ 5 10 15 2% ao s 0 15 2 o 5 @ 18 ~T T DISPLACED OLIGOMER (percent) TIME (min) Fic. 6. Effects of TBP-ter complexes on strand displace- ment by DNA polymerase. a, the partial duplexes were constructed from M13mp18-ter CCW and M13mp19-ter CW circles as described (11) with annealed 64-mer oligomers containing the complementary ter sequence; a 7-nucleotide tail was present at the 5’-end of the partial duplexes and a 27-nucleotide tail was at the 3’-end. A 17-mer sequencing primer was annealed to each plasmid as described under “Experimental Procedures.” b, The assay conditions were as described under “Experimental Procedures.” The products of the DNA polym- erase reactions with DNA polymerase I (large fragment, 5 units), T5 (4.4 units), and T7 (2 units) were separated by electrophoresis on 4 10% glycerol, 7% polyacrylamide gel and detected by autoradiography. These amounts of the DNA polymerases were based on previous titrations of their strand displacement activity. TBP, where indicated, was present in a 20-fold molar excess over ter. Slices of the gel containing the displaced oligomers were measured for their radioac- tivity in a liquid scintillation counter. The percentage of displaced oligomer was based on the total radioactivity (displaced plus undis- placed oligomers). Replication Fork Blockage in E. coli Substrate: M13mp19-terCW M13mp18-terCCW TBP: - + 20 - + cos. TT) ATC GATCGATC oa E.coli-terB Fic. 7. Site of arrest of replication by TBP-ter complex. Using the partial duplexes with a sequencing primer (Fig. 6a) and T7 DNA polymerase, dideoxynucleotide sequencing (19) was performed according to the instructions of U. S. Biochemical Corp. Where indicated, TBP was present in a 20-fold excess over the ter sequence. The predominant termination sites are indicated. The T-like symbol representing the ter (E. coli-terB) sequence is oriented such that the head of the T should block the progress of a replication fork. length between the two DNA polymerase termination sites protected by TBP. Orientation-specific Blockage of PriA Helicase by TBP-ter Complex—The orientation-specific inhibition of helicases, in- cluding DnaB, Rep, and UvrD (helicase H), by a TBP-ter complex (11) was tested with PriA protein (24). The strand displacement activity of PriA protein was almost completely prevented by the TBP-terB complex and only in the orienta- tion expected to block the 3’ —» 5’ movement characteristic of this helicase (Fig. 8). This finding makes an earlier sugges- tion that the blockage of DnaB helicase is due to a specific TBP-helicase interaction (20) even more unlikely. DISCUSSION The termination of a replication fork by a protein-bound sequence has been observed for some circular genomes in vivo (3, 4, 7, 8) and with a reconstituted system of purified proteins (5, 11). The inverted repeats of a certain sequence (ter), when bound by TBP, block helicase actions in an orientation- specific manner (11, 20). In this study, we have further ex- amined features of the capacity of a TBP-ter complex to prevent the separation of strands essential at a replication ork. In a previous study (11), we used the oriC plasmid, which exploits the unique origin of the E. coli chromosome, and Placed the terB sequence of E. coli on either side of it to block one or the other of the two replication forks. Inasmuch as the unblocked fork might come full circle, the inhibition of rep- lication was incomplete. When oppositely oriented ter se- quences were placed one on each side of oriC, both forks were stopped, and replication was confined to the minimal oriC Tegion (Figs, 2 and 3). As discussed previously (11), the TBP- ter complex is especially effective in blocking the progress of the fork in the rolling-circle mode of replication that super- Venes in the reaction. The precise mechanisms involved in this Stage are unclear, as are the means by which the TBP- ter complex inhibits fork movement, presumably by prevent- ing helicase action on this substrate. The arrest of the replication fork by a TBP-ter complex was reflected in the migrations of replication intermediates analyzed by two-dimensional gel electrophoresis. A single ubble-shaped structure within oriC appeared when synthesis Was blocked by complexes oriented in opposite directions on oth sides of oriC (Fig. 4). : lockage of the replication forks in the reconstituted oriC ystem could be attributed to inhibition of the DnaB helicase 8783 (a) Le 3g gg 0S NUCLEOTIDES (b) 40 = = 5 Substrate TBP w 30 7 ccw - = woo ° cw + Ww a 3 S 20 oO 9 = °o @ wt $ a a ccw + oO J L L 1 0 25 50 75 100 125 PriA PROTEIN (ng) Fic. 8. Blockage of PriA helicase by TBP-ter complex. a, the helicase substrate, oriented to block the CCW movement of a replication fork, should prevent the action of the PriA helicase that moves in that direction; the CW substrate should offer no such impediment. b, assay of the PriA helicase. The reaction mixtures (20 ul) contained 20 mm Tris-HCl (pH 7.5), 9 mm MgCh, 4% (w/v) sucrose, 100 ug/ml bovine serum albumin, 1 mm ATP, 0.4 ug of E. coli single-stranded DNA-binding protein, 1.4 fmol of DNA substrate (linear oligomer), and 30 fmol of TBP. The amounts of PriA protein are indicated. The reaction was incubated at 30°C for 10 min and terminated by adding 20 mm EDTA, 0.25% SDS. Samples were loaded on a nondenaturing 10% glycerol, 10% polyacrylamide gel; dried; and subjected to autoradiography. Slices of the dried gel that contained the undisplaced (substrate) and displaced (oligonucleotide) bands were cut out, and their radioactivity was measured in a liquid scintil- lation counter. The percentage of oligonucleotide displaced is re- corded. component of the system (11). However, the inhibition of two other helicases, Rep and UvrD (helicase II), by the same TBP-ter complex, also in an orientation-dependent way, in- dicated that the bound ter sequence (rather than specific TBP-helicase interactions) was responsible. The opposite conclusion was reached in another study (20), which reported that a TBP complex with a ter sequence from plasmid R6K blocked the actions of DnaB, but not those of Rep or helicase II. The suggestion that this discrepancy might be due to a tighter association of TBP with the E. coli-terB sequence compared with that of R6K (1) led us to examine the relative strengths of a variety of ter sequences. All were found to be rather similar in their capacity to inhibit a replication fork (Table I). The claim that the TBP-ter (R6K) complex fails to block helicase II has since been withdrawn (25). The generality of the TBP-ter complex inhibition has been ex- tended to the helicase action of the T antigen of SV40 (25) and to that of the PriA helicase (Fig. 8). The capacity of the TBP-ter complex to impede strand separation was demonstrable with DNA polymerases that have strand-displacing activity. The DNA polymerases of 8784 phages T5 and T7 and the large fragment of E. coli polymerase I were all affected (Fig. 6), but the blockages were less severe and orientation-specific than those for the helicases. The exact location at which a replication fork, advanced by DnaB helicase, is blocked by a TBP-ter complex was placed at the very first nucleotide on either side of the 22-bp E. coli- terB sequence (Fig. 5). In the replication of a ColE1-ter plasmid (5), arrest was observed, as in our studies (Fig. 5), at the very first nucleotide of the 22-bp terB sequence, but also at the second nucleotide and prematurely at sites 6 and 47 bp upstream. Separation of strands by a DNA polymerase was not only less orientation-dependent, but was also not as severely restricted; displacements appeared to be limited by a 15-bp region of the complex (Fig. 7). These effects need to be reconciled with a footprint of DNase I inhibition that showed that the entire 22-bp sequence was affected by TBP binding.” Clearly, more extensive and refined analyses of TBP binding are needed to account for the extent and particularly the orientation specificity of the binding of a ter sequence. Acknowledgment—We are grateful to LeRoy L. Bertsch for his help in preparing the manuscript. REFERENCES 1. Kuempel, P. L., Pelletier, A. J., and Hill, T. M. (1989) Cell 59, 581-583 2. Hidaka, M., Kobayashi, T., and Horiuchi, T. (1991) J. Bacteriol. 173, 391-393 3. Hidaka, M., Kobayashi, T., Takenaka, S., Takeya, H., and Hor- iuchi, T. (1989) J. Biol. Chem. 264, 21031-21037 4. Sista, P. R., Mukherjee, S., Patel, P., Khatri, G. S., and Bastia, D. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, 3026-3030 5. Hill, T. M. and Marians, K. J. (1990) Proc. Natl. Acad. Sci. U.S. A, 87, 2481-2485 6. Bastia, D., Germine, J., Crosa, J. H., and Ram, J. (1981) Proce. ? FE. H. Lee and A. Kornberg, unpublished data. 10. 11. 12. 13. 14. 15. 16. 17. 7 Sanger, F., Nicklen, S., and Coulson, A. R. (1977) Proc. Natl. 20. 21. 22. 23. 24. 25. 26. 27. . Bradford, M. M. (1976) Anal. Biochem. 72, 248-256 Replication Fork Blockage in E. coli Natl. Acad. Sci. U. S. A. 78, 2095-2099 . Horiuchi, T., and Hidaka, M. (1988) Cell 54, 515-523 . Hill, T. M., Pelletier, A. J., Tecklenburg, M. L., and Kuempel, p, L. (1988) Cell 55, 459-466 . Williams, N. K., and Wake, G. R. (1989) Nucleic Acids Res. 17, 9947-9956 Lewis, P. J.. and Wake, R. G. (1989) J. Bacteriol. 171, 1409_ 1408 Lee, E. H., Kornberg, A., Hidaka, M., Kobayashi, T., and Horj- uchi, T. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, 9104-9108 Baker, T. A., Funnell, B. E., and Kornberg, A. (1987) J. Biol. Chem. 262, 6877-6885 Matson, S. W., and Kaiser-Rogers, K. A. (1990) Annu. Rep, Biochem. 59, 289-329 Kaguni, J. M., and Kornberg, A. (1984) Cell 38, 889-900 Meijer, M., Beck, E., Hansen, F. G., Bergmann, H. F., Messer, W., von Meyenberg, K., and Schaller, H. (1979) Proc. Nati. Acad. Sci. U. S. A. 76, 580-584 Laemmli, U. K. (1970) Nature 227, 680-685 Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY Brewer, B. J., and Fangman, 8. L. (1987) Cell 51, 463-471 Acad. Sci. U. S. A. 74, 5463-5467 Khatri, G. S., MacAllister, T., Sista, P. R., and Bastia, D. (1989) Cell 59, 667-674 Lewis, P. J., Smith, M. T., and Wake, R. G. (1989) J. Bacteriol. 171, 3564-3567 Kornberg, A. (1982) DNA Replication, W. H. Freeman & Co., New York O’Donnell, M. E., and Kornberg, A. (1985) J. Biol. Chem. 260, 12884-12889 Lasken, R. S., and Kornberg, A. (1988) J. Biol. Chem. 263, 5512- 5518 Bedrosian, C. L., and Bastia, D. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 2618-2622 Horiuchi, T., Hidaka, M., Akiyama, M., Nishitani, H. and Seki- guchi, M. (1987) Mol. Gen. Genet. 210, 394-398 Masai, H., and Arai, K. (1989) J. Biol. Chem. 264, 8082-8090