484
Biochimica et Biophysica Acta, 521 (1978) 484--492 © Elsevier/North-Holland Biomedical Press
BBA 99356
TRANSFORMATION IN BA CILL US SUB TILLS WITH NITROGEN MUSTARD CROSSLINKED DNA EFFECT ON COTRANSFORMATION AND MUTATION FREQUENCIES
SAUL SCHEINBACH and RIVKA RUDNER
Department of Biological Sciences, Hunter College of the City University of New York, New York, N.Y. 10021 (U.S.A.) (Received March 20th, 1978)
Summary
Bacillus subtilis DNA Was treated in vitro with nitrogen mustard and the crosslinked molecules were purified, after alkali denaturation, by hydroxyapatite chromatography. When tested for the ability to transform the trpC2hisB2 segment, these molecules exhibited a decrease in the cotransformation index (r) as compared to native or renatured DNA. The decrease in r was not accompanied by an increase in mutagenicity.
Introduction Covalent crosslinks between complementary DNA strands induced in vitro or in vivo have been detected following treatment with a variety of agents including the bifunctional mustard compounds [1], psoralen plus light [2], mitomycin C [3], nitrous acid [4], ultraviolet irradiation [5] and low pH [6]. In addition, bacterial DNA preparations contain a small fraction of 'naturally' crosslinked molecules, apparently resulting from shear forces generated during purification [7]. Upon exposure to denaturants, both naturally and chemically induced crosslinked molecules undergo strand unwinding, but separation is prevented by the crosslinks, thereby allowing the strands to rewind once the denaturant is removed [7,4]. Irrespective of their origin, crosslinked molecules have been shown to transform well in both Haernophilus influenzae [8,9], and Bacillus subtilis [10,11]. In B. subtilis, studies recently reviewed by Dubnau [12] have shown that donor DNA is converted to single strand intermediates by exonucleolytic degradation of one strand to 5'-mononucleotides. Fragments of the single strand then replace an equivalent segment of recipient DNA by a non-covalent association to form a donor recipient complex which is subsequently covalently
485 joined to the recipient chromosome. Conversion of crosslinked molecules to single strands could n o t occur w i t h o u t removal of the covalent bridge. Removal (or repair} of crosslinks involving b o t h excision and recombination was observed in Escherichia coli [13--15] and could generate mutations in ways similar to that which occurs during 'SOS' repair of overlapping daughter strand gaps [16,17]. We report studies on the effect of nitrogen mustard crosslinks on genetic linkage and mutagenicity of B. subtilis transforming DNA designed to establish whether c o m p e t e n t cells remove crosslinks via an error-free or errorprone mechanism. Materials and Methods
Bacterial strains. The recipient strain used in this study was SB-25, a derivative of B. subtilis 168 which carries the closely linked hisB2 and trpC2 markers and was kindly provided by D. Dubnau. The prototrophs W23 (SrEWI r) and SS-1 were used as DNA sources, the latter was obtained by transformation of SB-25 with W23 DNA. Strain H3 (hisB2) was used as the recipient for mutagenesis assays and was similarly prepared by transfromation of SB-25. DNA preparation and hydroxyapatite chromatography. DNA isolation was carried out according to a modification [18] of the Marmur [19] method. DNA diluted to 25 pg/ml in triethanolamine-HC1 buffer (0.025 M}, pH 7.3, was denatured by gentle addition of 1/4 volume NaOH (0.16 M), EDTA (0.002 M) to yield pH 12.3. After 3 min at room temperature without stirring, an equal volume of citric acid (0.2 M)-Tris (0.06 M) was added to neutralize the solution. Renaturation was performed in 2 × SSC (SSC, 0.15 M NaC1-0.015 M sodium citrate, pH 7.0) at 68°C for 4--5 h at a concentration of a b o u t 18 gg/ml DNA [18]. Hydroxyapatite was prepared according to the method of Miyazawa and Thomas [20]. Columns were packed at room temperature to a height of 1.0 cm per A26onm units of the DNA to be applied. A flow rate of 23--25 ml/h was maintained by a peristaltic pump and 4-ml fractions were collected. Absor'bance of each fraction was read at 260 nm. Fractions used for transformation were sterilized over chloroform and used immediately or stored at 4°C. For additional information see Fig. 2 legend. Nitrogen mustard treatment. N-Methyl-bis-(2-chloroethyl)-amine-HC1, a bifunctional nitrogen mustard, donated b y Merck, Sharpe and Dohme Research Laboratories, was hydrated to 400 ~M and used fresh. Crosslinking was done by modification of an earlier m e t h o d [ 11 ]. Nitrogen mustard was diluted in 0.025 M triethanolamine-HC1, pH 7.3, and allowed to stand 30 min at 25°C to yield the ethyleneimmonium ion. DNA was then added to yield 25 pg DNA/ ml and 10 pM nitrogen mustard. The solution was denatured as described above and the reaction terminated by addition of 1.25 volumes Na2S203 (1.0 M). Undenatured samples were treated in an identical manner except that alkali and acid were mixed before being added. Samples to be used for transformation were added directly to recipient cells after being sterilized with a drop of chloroform. Transformation and mutagenesis assays. C o m p e t e n t cells were prepared by a modification [18] of the two-step growth cycle [21]. Medium I and Medium II
486 were previously described [18]. The cells were collected following growth in Medium II by centrifugation, concentrated 20-fold in Medium II containing 10% glycerol, quickly frozen at --20°C and stored at --60°C. Transformations were performed with thawed cells diluted 10-fold in Medium II by adding 0.1 ml transforming DNA to 0.9 ml of cells and incubating at 37°C for 30 min. The reaction was terminated by adding 10 ~g pancreatic DNAase/ml (Worthington Biochemical Corp.) in 0.2 M MgSO4. Samples were stored on ice until plated. Selective plates contained minimal medium [21], 1.5% Bacto Agar (Difco) and 50 ~g/ml of L-tryptophan or L-histidine or were unsupplemented. D o n o r DNA from strain SS-1 was used to transform SB-25 and the cotransformation index, r, as defined by Nester and Lederberg [22] was measured by either of the following methods: In the first method, the transformed cells were plated directly on three selective media to distinguish each of the three possible transformant types. In the second, transformants were selected for either single marker, and 50--150 well spread colonies were replica plated to a second selective medium so as to measure the fraction of double transformants among the singles. In some cases colonies were streaked onto nutrient agar plates prior to replica plating. For mutagenesis assays, frozen c o m p e t e n t H3 cells were transformed with crosslinked or renatured SS-1 DNA and plated on minimal medium supplemented with 2.5 pg L-tryptophan/ml and 250 pg histidine assay/ml (Difco). Mutations in trpC, trpD or trpF can be detected on this medium because the resulting colonies excrete fluorescent intermediates [23--25]. Fluorescent colonies among the his + transformants were scored after 40 h incubation at 37°C by viewing them under a ultraviolet at 254 nm. Results and Discussion
Production and purification of crosslinked transforming DNA The cotransformation index and mutagenicity of nitrogen mustard-treated transforming DNA were examined for crosslinked molecules obtained b y hydroxyapatite chromatography. Treatment of DNA with 10 #M nitrogen mustard (HN2) for 100 min yielded maximum transforming activity following transient exposure to alkali and was denoted HN2-D-DNA. Fig. I(A--D) illustrates hydroxyapatite chromatograms of HN2-D-DNA and three control preparations, i.e. native, alkali denatured and thermally renatured DNA. As shown in Fig. 1B, a b o u t 40% of the recovered HN2-D-DNA eluted in the second peak like the bulk of native DNA (Fig. 1A) and contained the transforming activity. A comparison with a chromatogram of alkaii-denatured DNA (Fig. 1C), indicates that prior treatment with nitrogen mustard resulted in a 4-fold increase in the native-like fraction. In the case of denatured DNA this second peak represents the naturally crosslinked molecules and the relative proportion was comparable to that observed for denatured H. influenzae DNA [8]. Renatured DNA served as a control to distinguish between effects caused by crosslinks and those due to strand unwinding and rewinding. Fig. 1D shows that approx. 75% of the recovered material became double stranded as a result of thermal renaturation.
487
OA
80
0.3
• 60
0,4
E
30
Q3 0.40
0.2
- 40
0.1
- 20
~
== 20 ~,
Y
0.2
z
Biochimica et Biophysica Acta, 521 (1978) 484--492 © Elsevier/North-Holland Biomedical Press
BBA 99356
TRANSFORMATION IN BA CILL US SUB TILLS WITH NITROGEN MUSTARD CROSSLINKED DNA EFFECT ON COTRANSFORMATION AND MUTATION FREQUENCIES
SAUL SCHEINBACH and RIVKA RUDNER
Department of Biological Sciences, Hunter College of the City University of New York, New York, N.Y. 10021 (U.S.A.) (Received March 20th, 1978)
Summary
Bacillus subtilis DNA Was treated in vitro with nitrogen mustard and the crosslinked molecules were purified, after alkali denaturation, by hydroxyapatite chromatography. When tested for the ability to transform the trpC2hisB2 segment, these molecules exhibited a decrease in the cotransformation index (r) as compared to native or renatured DNA. The decrease in r was not accompanied by an increase in mutagenicity.
Introduction Covalent crosslinks between complementary DNA strands induced in vitro or in vivo have been detected following treatment with a variety of agents including the bifunctional mustard compounds [1], psoralen plus light [2], mitomycin C [3], nitrous acid [4], ultraviolet irradiation [5] and low pH [6]. In addition, bacterial DNA preparations contain a small fraction of 'naturally' crosslinked molecules, apparently resulting from shear forces generated during purification [7]. Upon exposure to denaturants, both naturally and chemically induced crosslinked molecules undergo strand unwinding, but separation is prevented by the crosslinks, thereby allowing the strands to rewind once the denaturant is removed [7,4]. Irrespective of their origin, crosslinked molecules have been shown to transform well in both Haernophilus influenzae [8,9], and Bacillus subtilis [10,11]. In B. subtilis, studies recently reviewed by Dubnau [12] have shown that donor DNA is converted to single strand intermediates by exonucleolytic degradation of one strand to 5'-mononucleotides. Fragments of the single strand then replace an equivalent segment of recipient DNA by a non-covalent association to form a donor recipient complex which is subsequently covalently
485 joined to the recipient chromosome. Conversion of crosslinked molecules to single strands could n o t occur w i t h o u t removal of the covalent bridge. Removal (or repair} of crosslinks involving b o t h excision and recombination was observed in Escherichia coli [13--15] and could generate mutations in ways similar to that which occurs during 'SOS' repair of overlapping daughter strand gaps [16,17]. We report studies on the effect of nitrogen mustard crosslinks on genetic linkage and mutagenicity of B. subtilis transforming DNA designed to establish whether c o m p e t e n t cells remove crosslinks via an error-free or errorprone mechanism. Materials and Methods
Bacterial strains. The recipient strain used in this study was SB-25, a derivative of B. subtilis 168 which carries the closely linked hisB2 and trpC2 markers and was kindly provided by D. Dubnau. The prototrophs W23 (SrEWI r) and SS-1 were used as DNA sources, the latter was obtained by transformation of SB-25 with W23 DNA. Strain H3 (hisB2) was used as the recipient for mutagenesis assays and was similarly prepared by transfromation of SB-25. DNA preparation and hydroxyapatite chromatography. DNA isolation was carried out according to a modification [18] of the Marmur [19] method. DNA diluted to 25 pg/ml in triethanolamine-HC1 buffer (0.025 M}, pH 7.3, was denatured by gentle addition of 1/4 volume NaOH (0.16 M), EDTA (0.002 M) to yield pH 12.3. After 3 min at room temperature without stirring, an equal volume of citric acid (0.2 M)-Tris (0.06 M) was added to neutralize the solution. Renaturation was performed in 2 × SSC (SSC, 0.15 M NaC1-0.015 M sodium citrate, pH 7.0) at 68°C for 4--5 h at a concentration of a b o u t 18 gg/ml DNA [18]. Hydroxyapatite was prepared according to the method of Miyazawa and Thomas [20]. Columns were packed at room temperature to a height of 1.0 cm per A26onm units of the DNA to be applied. A flow rate of 23--25 ml/h was maintained by a peristaltic pump and 4-ml fractions were collected. Absor'bance of each fraction was read at 260 nm. Fractions used for transformation were sterilized over chloroform and used immediately or stored at 4°C. For additional information see Fig. 2 legend. Nitrogen mustard treatment. N-Methyl-bis-(2-chloroethyl)-amine-HC1, a bifunctional nitrogen mustard, donated b y Merck, Sharpe and Dohme Research Laboratories, was hydrated to 400 ~M and used fresh. Crosslinking was done by modification of an earlier m e t h o d [ 11 ]. Nitrogen mustard was diluted in 0.025 M triethanolamine-HC1, pH 7.3, and allowed to stand 30 min at 25°C to yield the ethyleneimmonium ion. DNA was then added to yield 25 pg DNA/ ml and 10 pM nitrogen mustard. The solution was denatured as described above and the reaction terminated by addition of 1.25 volumes Na2S203 (1.0 M). Undenatured samples were treated in an identical manner except that alkali and acid were mixed before being added. Samples to be used for transformation were added directly to recipient cells after being sterilized with a drop of chloroform. Transformation and mutagenesis assays. C o m p e t e n t cells were prepared by a modification [18] of the two-step growth cycle [21]. Medium I and Medium II
486 were previously described [18]. The cells were collected following growth in Medium II by centrifugation, concentrated 20-fold in Medium II containing 10% glycerol, quickly frozen at --20°C and stored at --60°C. Transformations were performed with thawed cells diluted 10-fold in Medium II by adding 0.1 ml transforming DNA to 0.9 ml of cells and incubating at 37°C for 30 min. The reaction was terminated by adding 10 ~g pancreatic DNAase/ml (Worthington Biochemical Corp.) in 0.2 M MgSO4. Samples were stored on ice until plated. Selective plates contained minimal medium [21], 1.5% Bacto Agar (Difco) and 50 ~g/ml of L-tryptophan or L-histidine or were unsupplemented. D o n o r DNA from strain SS-1 was used to transform SB-25 and the cotransformation index, r, as defined by Nester and Lederberg [22] was measured by either of the following methods: In the first method, the transformed cells were plated directly on three selective media to distinguish each of the three possible transformant types. In the second, transformants were selected for either single marker, and 50--150 well spread colonies were replica plated to a second selective medium so as to measure the fraction of double transformants among the singles. In some cases colonies were streaked onto nutrient agar plates prior to replica plating. For mutagenesis assays, frozen c o m p e t e n t H3 cells were transformed with crosslinked or renatured SS-1 DNA and plated on minimal medium supplemented with 2.5 pg L-tryptophan/ml and 250 pg histidine assay/ml (Difco). Mutations in trpC, trpD or trpF can be detected on this medium because the resulting colonies excrete fluorescent intermediates [23--25]. Fluorescent colonies among the his + transformants were scored after 40 h incubation at 37°C by viewing them under a ultraviolet at 254 nm. Results and Discussion
Production and purification of crosslinked transforming DNA The cotransformation index and mutagenicity of nitrogen mustard-treated transforming DNA were examined for crosslinked molecules obtained b y hydroxyapatite chromatography. Treatment of DNA with 10 #M nitrogen mustard (HN2) for 100 min yielded maximum transforming activity following transient exposure to alkali and was denoted HN2-D-DNA. Fig. I(A--D) illustrates hydroxyapatite chromatograms of HN2-D-DNA and three control preparations, i.e. native, alkali denatured and thermally renatured DNA. As shown in Fig. 1B, a b o u t 40% of the recovered HN2-D-DNA eluted in the second peak like the bulk of native DNA (Fig. 1A) and contained the transforming activity. A comparison with a chromatogram of alkaii-denatured DNA (Fig. 1C), indicates that prior treatment with nitrogen mustard resulted in a 4-fold increase in the native-like fraction. In the case of denatured DNA this second peak represents the naturally crosslinked molecules and the relative proportion was comparable to that observed for denatured H. influenzae DNA [8]. Renatured DNA served as a control to distinguish between effects caused by crosslinks and those due to strand unwinding and rewinding. Fig. 1D shows that approx. 75% of the recovered material became double stranded as a result of thermal renaturation.
487
OA
80
0.3
• 60
0,4
E
30
Q3 0.40
0.2
- 40
0.1
- 20
~
== 20 ~,
Y
0.2
z
