(ter function/recognition site/rotational symmetry/coliphage)

Analyses of radioactive oligonucleotides in endonuclease digests of 3'-terminally labeled X DNA revealed the 3' terminal sequence -GTTACG for the I strand and -ACCCGCG for the r strand. These sequences, together with those previously known for the 5' cohesive ends, provide a total of 25 known base-pairs in the vicinity of the termini. When the cohesive ends are paired, the sequence between the nicks can be bisected by a 2-fold rotational axis of symmetry. Five of the first eight base-pairs, on either side of the axis, are rotationally symmetric. This symmetry may be involved in the recognition of the site by an enzyme responsible for formation of the cohesive ends. The DNA of bacteriophage X is a duplex molecule of about 46,500 base-pairs, with single-stranded projections of 12 nucleotides at each 5'-terminus. These projections, or cohesive ends, are of complementary sequence and, by their specific base-pairing, the DNA may form either a ring or a linear aggregate (1, 2). The mechanism by which the cohesive ends are formed is not completely understood. The mature viral DNA molecule may be produced from a replicative intermediate, possibly a concatemer, by endonucleolytic attack at the site of the cohesive ends (for a review, see ref. 3). This endonuclease is the ter function, whose existence was inferred from genetic studies (4) and which has been detected recently in extracts of Escherichia coli infected with phage X (5). Presumably, this endonuclease binds to a unique sequence of nucleotides in the region of the cohesive ends and introduces two nicks, 12 base-pairs apart, on opposite strands. The mature X DNA molecules may then be released by disruption of the base pairs between the nicks. Our objective was to identify the nucleotide sequence recognized by the ter enzyme. The sequence of bases between the two nicks is known (2). To determine sequences beyond these nicks, we labeled A DNA at specific sites at or near its 3'-termini (6) and determined the sequence of oligonucleotides produced by degradation of the DNA with a nonspecific endonuclease (7). The 3'-terminus of the 1 strand is -GTTACG, and that of the r strand is -ACCCGCG. Brezinski and Wang have recently informed us that they have independently obtained sequences in agreement with these results (8). MATERIALS AND METHODS DNA. Phage were produced from the X-lysogen E. coli N1383, (XcI857 Sam7), a gift of Dr. John Little (9, 10), and DNA was isolated (6). Analysis of the DNA on an alkaline 1151 sucrose gradient indicated that more than 90% of the strands were intact. 3'-Terminal Labeling. DNA termini were labeled with ['2P]nucleotides in a reaction catalyzed by the T4 DNA polymerase (6). The DNA [3.7 Amol of nucleotide in 1.95 ml containing 46 mM NaCl-86 mM Tris HCl (pH 8.0)] was heated to melt cohered ends (740, 10 min, followed by quick cooling to 0°). A solution (0.55 ml) containing polymerase (6), a single [a-32Pjdeoxynucleoside triphosphate (6), MgCl2, and 2-mercaptoethanol was added immediately after cooling. Final concentrations were: 1.5 mM DNA, 67 mM Tris HCl (pH 8.0), 36 mM NaCl, 6 mM MgCl2, 6 mM 2-mercaptoethanol, 0.1 mM triphosphate (2 X 1010 cpm/ Atmol), and 600 units/ml of polymerase. Incubation was at 110, and when acid-insoluble radioactivity in an aliquot reached a limiting value (usually 60 min), the reaction was terminated by addition of EDTA (to 20 mM) and by extraction with phenol. Triphosphates were removed by exhaustive dialysis against 1.0 M NaCl-10 mM Tris HCl (pH 8.0)-1.0 mM EDTA. The DNA was then purified by zone sedimentation in an alkaline sucrose gradient (5-20% sucrose in 0.2 M NaOH-0.8 M NaCl-1 mM EDTA; 25,000 RPM, 15 hr, 50, Spinco SW25 rotor). This purification removed small labeled DNA fragments, which contained about 30% of the radioactivity, but which were not detectable by A260 measurements. These fragments were probably not a consequence of endonuclease activity in the incubation mixture, because similar fragments were not observed in identical experiments with T7 DNA (6); they were probably present in the DNA used for labeling. The purified intact [12P]DNA strands were then dialyzed against 50 mM NaCl-10 mM Tris*HCl (pH 8.0)-1.0 mM EDTA. A portion of the DNA was used for strand separation (11), and a sample of each isolated strand was used for nearest-neighbor analysis (12). The remainder of the purified [82P]DNA was used for sequence analysis. Endonuclease Digestion of the Labeled DNA. Labeled DNA was transported from Baltimore to Edinburgh in evacuated, sealed tubes. The solutions were concentrated to 1.0 ml by rotary evaporation, dialyzed against 0.3 mM NaCl, and further concentrated to 0.1 ml. The labeled DNA was then digested with pancreatic DNase (Worthington Biochemical Corp.). Conditions were chosen to achieve an equable distribution of radioactive oligonucleotides from trial digests analyzed byelectrophoresis onWhatman AE81 paper (pH 3.5). Typical conditions were 50 Ahg/ml of DNase in 20 mM Tris HCl (pH 7.5)-10 mM MgC12 at 370 for 3-12 hr. The 1152 Biochemistry: Weigel et al.