Oxime esters of 2,6-diazaanthracene-9,10-dione and 4,5-diazafluoren-9-one as photo-induced DNA-cleaving agents.

Two series of oxime esters containing the 2,6-diazaanthracene-9,10-dione bis-(O-benzoyloxime) and 4,5-diazafluoren-9-one O-9-benzoyloxime moieties have been synthesized and tested as photo-induced DNA cleaving agents. All these compounds were found to cleave DNA upon irradiation with 312 nm UV light. The structure-activity relationship of these molecules for DNA cleavage was established. A plausible reaction mechanism is also proposed.

1. Introduction

Photocleavage of nucleic acids (DNA and RNA) can be very useful for molecular biological applications [1]. Organic molecules with DNA-cleaving ability are of great potential in the development of biotechnology and gene therapy [2]. A very important feature of this method is that all components can be mixed together without initiating the chemical reaction until it is irradiated [3]. If such “photocleavage agents” absorb light at wavelengths longer than 300 nm, nucleic acids will not be affected while selective excitation of the photocleavage agents can be achieved [4]. Since single-strand DNA damage is easily repaired by enzymatic processes [5], photocleavage of double-strand DNA molecules would be a more efficient tool for cancer therapy [6,7,8,9,10,11,12].

We were particularly interested in developing methods of DNA cleavage by radical species. A key feature is to facilitate the generation of radicals which are also reactive enough to cleave the DNA, as shown in the photolysis of N-aroyloxy-2-thiopyridones [13]. Along this line we have synthesized oxime esters 1 to cleave the calf thymus DNA upon UV irradiation (Scheme 1) [14].

Scheme 1

DNA cleavage by photolysis of oxime esters.

DNA cleavage by photolysis of n class="Chemical">oxime esters.

The weak N–O bond of the n class="Chemical">oxime esters can be selectively cleaved to generate the iminyl and carboxy radicals which can then cause the cleavage of the DNA. The radicals were detected by EPR (electron paramagnetic resonance) spectroscopy. We incorporated the anthraquinone, fluoren-9-one, or thioxanthen-9-one moiety into the structure of the oxime esters as well as different substituents on the aromatic carboxyl group to determine their effects on the efficiency of DNA cleavage. Most of these compounds exhibited single-strand scission of DNA, but some anthraquinone derivatives caused double-strand scission. We also studied the photo-induced DNA cleavage by heteroaromatic oxime esters of anthraquinone [15], and discovered that one particular compound could cleave DNA at the concentration as low as 1.0 μM. Recently, anthracenone-based oxime esters have been reported to display strong antiproliferative activity against K562 leukemia cells [16]. N,O-Diacyl-4-benzoyl- N-phenylhydroxylamines, also having a weak N–O bond, were recently reported to produce single strand cleavage of DNA [17]. Photolysis of the carboxylic esters of N-acyl-N-phenylhydroxylamines generated the corresponding carboxylic acids by cleaving the N–O bond [18].

2. Results and Discussion

The reaction of 2,6-diazaanthracene-9,10-dione (2) [19] with n class="Chemical">hydroxylamine hydrochloride (4 equiv.) in refluxing pyridine for 24 h gave the bis-oxime 3 in good yield. Treatment of compound 3 in anhydrous THF with sodium hydride (4 equiv.) at room temperature followed by the reaction with benzoyl chloride derivatives afforded the corresponding bis-oxime esters 4a–d in fair yields (Scheme 2). 4,5-Diazafluoren-9-one (5) [20] was similarly converted to its oxime 6 and oxime esters 7a–d (Scheme 3).

Scheme 2

Synthesis of bis-oxime esters 4a–d.

Scheme 3

Synthesis of bis-oxmie esters 7a–d.

The characteristic 1H n class="Chemical">NMR absorptions for compounds 4a-d and 7a-d are provided in the Experimental section. The number of different carbons in the 13C-NMR spectra of compounds 4a-d clearly demonstrates that they only have the anti configuration. The X-ray crystal structure of compound 7b (Figure 1) shows the more stable cisoid conformation of the C=O and N-O group [21].

Figure 1

X-ray crystal structure of compound 7b.

Synthesis of bis-oxmie esters n class="Chemical">7a–d.

The UV-VIS absorption spectra for compounds 4a–4d and 7a–7d were measured in n class="Chemical">CH2Cl2 (Figure 2), and their λmax and molar absorptivity (ε) values are listed in Table 1. It can be seen that compounds 4a–d and 7a–d all have some absorption at the 312 nm normally used for the photo-induced DNA cleavage.

Figure 2

UV-VIS absorption spectra for compounds 4a–d and 7a–d.

Table 1

λmax and molar absorptivity (ε) for compounds 4a–d and 7a–d.

Compound	λmax (ε)
4a	230 (23888), 280 (32944)
4b	238 (48714), 277 (52709)
4c	230 (44888), 283 (50593)
4d	248 (23504), 279 (28855)
7a	233 (57149), 273 (36500)
7b	232 (70174), 276 (45050)
7c	233 (63071), 271 (41946)
7d	234 (28102), 268 (16377)

The oxime esters 4a–d and n class="Chemical">7a–d were individually irradiated with UV light (312 nm) at the concentration of 100 μM in phosphate buffers (pH 6.0) and 2.5% DMSO containing the supercoiled circular ϕX174 RFI DNA (form I; 50 μM/base pair) under aerobic conditions at room temperature for 2 h. The cleavage results from gel electrophoresis on 1% agarose with ethidium bromide staining for compounds 4a–d and 7a–d are shown in Figure 3. It can be seen that all these compounds nicked the supercoiled circular DNA to give the relaxed circular (i.e., form II) DNA, and the cleaving ability of compounds 4a–d was greater than of compounds 7a–d. A simple explanation for the higher cleaving effect of compounds 4a–d is that they have bis-oxime ester moieties, whereas compounds 7a–d are mono-oxime esters. However, other factors such as degree of intercalation, polarity, and steric effect, etc. may also be involved. It is interesting to note that the substituent effect in these two series of compounds is almost reversed: for compounds 4a–d, F > CH3 > CN > H; for compounds 7a–d, H > CH3 > CN > F. Among these eight compounds, compound 4c exhibited the best results, so we carried out further cleavage experiments with different concentrations of compound 4c (Figure 4). In lane 1, without compound 4c, DNA was not decomposed by irradiation with 312 nm of UV for 2 h. In lane 2, in the presence of 500 μM of compound 4c, DNA cleavage did not occur in the dark. However, in lanes 3–6, in the presence of 500, 250, 100 or 50 μM of compound 4c, significant amount of the relaxed circular (i.e., form II) DNA was obtained. Thus the UV light functioned as a “trigger” to initiate the DNA scission process. However, with 25 or 12.5 μM of compound 4c, DNA cleavage did not occur.

Figure 3

DNA cleaving abilities of compounds 4a–d and 7a–d (100 μM).

Figure 4

Dose measurement of compound 4c for DNA cleavage. Lane 1, DNA at 312 nm for 2 h; Lane 2, DNA and 4c (500 μM) in the dark; Lanes 3–8, DNA and 500, 250, 100, 50, 25, 12.5 μM of 4c, individually, at 312 nm for 2 h.

DNA cleaving abilities of compounds 4a–d and n class="Chemical">7a–d (100 μM).

Dose measurement of compound 4c for DNA cleavage. Lane 1, Dn class="Chemical">NA at 312 nm for 2 h; Lane 2, DNA and 4c (500 μM) in the dark; Lanes 3–8, DNA and 500, 250, 100, 50, 25, 12.5 μM of 4c, individually, at 312 nm for 2 h.

We have also carried out the photolysis of compound 4a with a medium pressure mercury lamp (222–366 nm) in n class="Chemical">benzene under nitrogen using 1,4-cyclohexadiene [22] as the radical scavenger (Scheme 4). The products were purified by column chromatography to give 2,6-diazaanthracene-9,10-dione (2) and benzoic acid in good yields. These DNA-cleaving processes were further investigated in control experiments with oxime ester 4c and 7a, respectively, by addition of sodium azide (100 to 500 mM) as a scavenger of singlet oxygen [23,24]. The presence of singlet oxygen may contribute to the DNA damage [25,26].We found that the DNA cleavage results did not show obvious reduction after the addition of sodium azide. This outcome implies that singlet oxygen did not participate in these DNA-cleaving processes. In order to explain the photolysis products a plausible mechanism is proposed (Scheme 5) [27,28,29]. The weak N–O bond is first cleaved by the UV light to generate the bis-iminyl radical and benzoyloxy radical which can then abstract hydrogen atoms from 1,4-cyclohexadiene to give the bis-imine 9 and benzoic acid (8). The bis-imine 9 can be easily hydrolyzed to 2,6-diaza- anthracene-9,10-dione (2) during workup. Based on the photolysis results, we propose that the DNA cleavage is initiated by the homolytic cleavage of the weak N–O bond of oxime esters 4 or 7 to generate the bis-iminyl radical and benzoyloxy radical, which could then abstract hydrogen atom from the sugar moiety of the DNA.

Scheme 4

Photolysis of compound 4a in the presence of 1,4-cyclohexadiene.

Scheme 5

Proposed mechanism for the photolysis of compound 4.

3. Experimental

3.1. General

Melting points were determined with a SMP3 melting apparatus, and were uncorrected. Infrared spectra were recorded with a Perkin-Elmer Spectrum 1600 FT-IR spectrometer. NMR spectra were recorded on a Bruker AV-300 spectrometer. n class="Chemical">Me4Si (δ 0.00 ppm) and the center of the CDCl3 triplet (δ 77.00 ppm) were used as the internal standard for 1H- and 13C-NMR spectra, respectively. All NMR chemical shifts are reported as δ values in parts per million (ppm) and coupling constants (J) are given in Hertz (Hz). High resolution mass spectra (HRMS) were measured with a JEOL JMS-SX102A mass spectrometer. UV-VIS spectra were taken with a Scino S-3100 spectrophotometer. Flash column chromatographic purifications were performed using Merck 60 H silica gel.

3.2. General Procedure for the Preparation of Oximes 3 and 6

A mixture of compound 2 or 5 (0.48 mmol) and hydroxylamine hydrochloride (4 equiv. for 2, 2 equiv. for 5) in n class="Chemical">pyridine (5 mL) was heated under reflux for 24 h. After cooling to room temperature, water (50 mL) was added, and the solid formed was collected by vacuum filtration and was washed sequentially with CH2Cl2 (50 mL) and acetone (10 mL), and then dried under vacuum to give products 3 or 6.

3.2.1. 2,6-Diazaanthracene-9,10-dioxime (3)

Yield: 105 mg (92%); yellowish white solid: mp 236 °C (decomp); IR (film) v 3369, 3123, 2806, 2688, 2556, 1824, 1636, 1593, 1541, 1493, 1462, 1408, 1331, 1282, 1239, 1201, 1157, 1070, 1032, 949, 833, 796 cm−1; 1H-n class="Chemical">NMR (DMSO-d6) δ 9.90 (1H, s), 8.68 (2H, d, J = 5.1 Hz), 7.97 (2H, d, J = 4.5 Hz); 13C-NMR (DMSO-d6) δ 151.1, 150.0, 141.6, 139.2, 121.0, 118.0; EI-MS (relative intensity) m/z 240 (M+, 1), 256 (2), 240 (1), 129 (2), 86 (4), 79 (32), 78 (100), 63 (99), 62(6), 61(17); Exact mass calcd for C12H8N4O2 m/z 240.0647 (M+), EI-HRMS m/z 240.0641.

3.2.2. 4,5-Diazafluoren-9-oxime (6)

Yield: 97 mg (90%); white solid: mp 253.2–254.4 °C; IR (film) v 3133, 3032, 2761, 2561, 1624, 1564, 1494, 1466, 1398, 1351, 1287, 1158, 1004, 948, 817, 751, 701, 638 cm−1; 1H-n class="Chemical">NMR (DMSO-d6) δ 8.72 (2H, dd, J = 9.0, 4.5 Hz), 8.65 (1H, dd, J = 7.8, 1.2 Hz), 8.17 (1H, dd, J = 7.5, 0.9 Hz), 7.54–7.45 (2H, m); 13C-NMR (DMSO-d6) δ 157.6, 156.7, 151.5, 151.0, 146.5, 135.5, 130.7, 128.6, 124.9, 124.3, 124.1; EI-MS (relative intensity) m/z 197 (M+, 1), 197 (1), 80 (4), 79 (26), 78 (100), 63 (57), 61 (12), 45 (13), 31 (26), 17 (14); Exact mass calcd for C11H7N3O m/z 197.0589 (M+), EI-HRMS m/z 197.0587.

3.3. General Procedure for the Prepartion of Oxime Esters 4 and 7

To a mixture of compound 3 or 6 (0.20 mmol) in THF (3.5 mL) at room temperature was added n class="Chemical">NaH (60% in oil, 4 equiv. for 3, 2 equiv. for 6). After stirring for 5 min, the appropriate acid chloride (3 equiv for 3, 1.5 equiv. for 6) was added in one portion. The mixture was stirred for 3 h, and then the solvent was evaporated under vacuum. The residue was dissolved in CH2Cl2 (30 mL) and was washed with water (20 mL × 3), dried (MgSO4), and evaporated under vacuum. The crude products 4 or 7 were rinsed with hexane, and then recrystallized from CH2Cl2/hexane.

3.3.1. 2,6-Diazaanthracene-9,10-dione bis-(O-benzoyloxime) (4a)

Yield: 39 mg (52%); white solid: mp 214–215 °C; IR (film) v 3051, 1774, 1573, 1418, 1264, 1234, 1152, 1040, 1023, 990, 896, 732, 702 cm−1; 1H-n class="Chemical">NMR see Table 2; 13C-NMR (CDCl3) δ 163.3, 152.3, 151.6, 149.5, 139.6, 134.3, 130.2, 129.2, 127.7, 121.4, 120.1; EI-MS (relative intensity) m/z 448 (M+, 1), 226 (5), 208 (5), 122 (8), 105 (100), 84 (6), 77 (37), 51 (5); Exact mass calcd for C26H16N4O4 m/z 448.1172 (M+), EI-HRMS m/z 448.1167.

Table 2

The characteristic 1H-NMR absorptions for compounds 4a–d .

	4a (X = H)	4b (X = CH3)	4c (X = F)	4d (X = CN)
H-1, H-5	9.84 (s)	9.83 (s)	9.79 (s)	9.76 (s)
H-3, H-7	8.94 (d, 5.2)	8.92 (d, 5.1)	8.94 (d, 5.1)	8.97 (d, 5.2)
H-4, H-8	8.34 (d, 5.2)	8.32 (d, 5.1)	8.33 (dd, 5.1, 0.6)	8.32 (d, 5.2)
H-3′, H-5′, H-3′′, H-5′′	8.15 (dd, 7.2, 1.2)	8.04 (d, 8.1)	8.21–8.15 (m)	8.25 (d, 8.1)
H-1′, H-1′′	7.69 (br t, 7.3)	2.47 (s)	–	–
H-2′, H-4′, H-2′′, H-4′′	7.57 (br t, 7.3)	7.36 (d, 8.1)	7.29–7.21 (m)	7.88 (d, 8.1)

The data are expressed as chemical shift in δ (splitting pattern, coupling constant in Hz).

3.3.2. 2,6-Diazaanthracene-9,10-dione bis-[O-(4-methylbenzoyl)oxime] (4b)

Yield: 44 mg (56%); white solid mp 224–225 °C; IR (film) v 3037, 1767, 1612, 1574, 1411, 1332, 1237, 1179, 1151, 1039, 983, 915, 866, 849, 800, 769, 737, 682 cm−1; 1H-n class="Chemical">NMR see Table 2; 13C-NMR (CDCl3) δ 163.3, 152.1, 151.5, 149.2, 145.3, 139.5, 130.2, 129.8, 124.8, 121.4, 120.0, 21.9; EI-MS (relative intensity) m/z 476 (M+, 1), 331 (11), 330 (8), 258 (40), 253 (15), 211 (13), 136 (13), 119 (100), 91 (39), 89 (21), 65 (11); Exact mass calcd for C28H20N4O4 m/z 476.1485 (M+), EI-HRMS m/z 476.1477.

3.3.3. 2,6-Diazaanthracene-9,10-dione bis-[O-(4-fluorobenzoyl)oxime] (4c)

Yield: 44 mg (55%); white solid: mp 217–218 °C (decomp); IR (film) v 3055, 2981, 1777, 1602, 1505, 1421, 1264, 1158, 1147, 1029, 991, 896, 828, 721, 703 cm−1; 1H-n class="Chemical">NMR see Table 2; 13C-NMR (CDCl3) δ 162.3, 152.4, 151.4, 149.6, 139.5, 132.9, 132.8, 124.0, 121.4, 120.1, 116.7, 116.5; EI-MS (relative intensity) m/z 484 (M+, 1), 123 (9), 79 (46), 78 (100), 63 (89), 61 (16); Exact mass calcd for C26H14F2N4O4 m/z 484.0983 (M+), EI-HRMS m/z 484.0981.

3.3.4. 2,6-Diazaanthracene-9,10-dione bis-[O-(4-cyanobenzoyl)oxime] (4d)

Yield: 42 mg (50%); white solid: mp 209–210 °C; IR (film) v 3101, 3053, 2228, 1777, 1607, 1567, 1535, 1402, 1328, 1291, 1229, 1179, 1145, 1039, 978, 914, 853, 789, 749, 680 cm−1; 1H-n class="Chemical">NMR see Table 2; 13C-NMR (CDCl3) δ 161.8, 152.7, 151.3, 150.3, 139.3, 133.0, 131.6, 130.6, 121.2, 120.2, 117.9, 117.6; EI-MS (relative intensity) m/z 498 (M+, 1), 331 (28), 330 (21), 259 (17), 258 (100), 208 (22), 182 (17), 147 (27), 130 (64), 102 (22); Exact mass calcd for C28H14N6O4 m/z 498.1077 (M+), EI-HRMS m/z 498.1080.

The characteristic 1H-n class="Chemical">NMR absorptions for compounds 4a–d .

3.3.5. 4,5-Diazafluoren-9-one O-9-benzoyloxime (7a)

Yield: 44 mg (72%); white solid: mp 184–185 °C; IR (film) v 3055, 2981, 1754, 1596, 1563, 1449, 1398, 1264, 1239, 1169, 1079, 1045, 1021, 977, 891, 861, 819, 732, 703 cm−1; 1H-n class="Chemical">NMR see Table 3; 13C-NMR (CDCl3) δ 163.7, 160.3, 158.9, 154.5, 153.6, 153.4, 136.6, 134.1, 130.9, 130.1, 130.0, 129.1, 128.3, 125.3, 124.3, 124.2; EI-MS (relative intensity) m/z 301 (M+, 31), 181 (37), 137 (32), 125 (44), 111 (72), 105 (85), 97 (100), 83 (80), 71 (84), 55 (96); Exact mass calcd for C18H11N3O2 m/z 301.0851 (M+), EI-HRMS m/z 301.0850.

Table 3

The characteristic 1H-NMR absorptions for compounds 7a–d.

	7a (X = H)	7b (X = CH3)	7c (X = F)	7d (X = CN)
H-2, H-7	8.80 (dd, 4.8, 0.9)	8.83–8.78 (m)	8.83–8.78 (m)	8.85–8.81 (m)
H-5	8.56 (dd, 7.8, 1.2)	8.58 (dd, 7.6, 1.3)	8.52 (dd, 7.6, 1.0)	8.47 (dd, 6.6, 1.0)
H-4	8.35 (dd, 7.5, 0.9)	8.36 (dd, 7.8, 1.2)	8.35 (dd, 7.5, 0.9)	8.36 (dd, 6.3, 1.0)
H-3′, H-5′	8.19 (d, 7.5)	8.09 (d, 8.1)	8.22 (dd, 8.7, 5.4)	8.31 (d, 8.4)
H-1′	7.72 (dd, 7.5, 7.5)	2.50 (s)	–	–
H-2′, H-4′	7.60 (dd, 7.8, 7.2)	8.09 (d, 8.1)	7.44–7.36 (m)	7.92 (d, 8.4),
H-3, H-6	7.43–7.35 (m)	7.43–7.36 (m)	7.29 (t, 8.4)	7.44–7.38 (m)

The data are expressed as chemical shifts in δ (splitting pattern, coupling constant in Hz).

3.3.6. 4,5-Diazafluoren-9-one O-9-(4-methylbenzoyl)oxime (7b)

Yield: 48 mg (74%); white solid: mp 232–233 °C; IR (film) v 3078, 3057, 2916, 1748, 1631, 1608, 1583, 1563, 1473, 1410, 1397, 1288, 1249, 1171, 1154, 1109, 1093, 1045, 1015, 977, 893, 861, 821, 748, 737, 684 cm−1; 1H-n class="Chemical">NMR see Table 3; 13C-NMR (CDCl3) δ 163.7, 160.2, 158.8, 154.1, 153.5, 153.2, 145.0, 136.5, 130.8, 129.9 (×2), 129.8, 125.3 (×2), 124.2 (×2), 21.8; EI-MS (relative intensity) m/z 315 (M+, 15), 181 (32), 180 (13), 166 (17), 131 (10), 119 (100), 91 (30), 69 (17), 65 (12), 28 (13); Exact mass calcd for C19H13N3O2 m/z 315.1008 (M+), EI-HRMS m/z 315.1008.

3.3.7. 4,5-Diazafluoren-9-one O-9-(4-fluorobenzoyl)oxime (7c)

Yield: 50 mg (77%); white solid: mp 228–229 °C; IR (film) v 3069, 3013, 1764, 1603, 1562, 1396, 1281, 1243, 1156, 1114, 1098, 1041, 1011, 976, 945, 899, 852, 814, 792, 747, 678 cm−1; 1H-n class="Chemical">NMR see Table 3; 13C-NMR (CDCl3) δ 168.1, 164.7, 162.8, 160.4, 158.9, 154.6, 153.6, 136.5, 132.7, 132.6, 131.0, 130.0, 125.3, 124.6, 124.2, 116.5; EI-MS (relative intensity) m/z 319 (M+, 15), 181 (27), 180 (10), 140 (14), 123 (100), 95 (22); Exact mass calcd for C19H13N3O2 m/z 319.0757 (M+), EI-HRMS m/z 319.0754.

3.3.8. 4,5-Diazafluoren-9-one O-9-(4-cyanobenzoyl)oxime (7d)

Yield: 50 mg (75%); white solid: mp 238–239 °C; IR (film) v 3056, 2959, 2926, 2851, 2231, 1767, 1731, 1593, 1563, 1471, 1394, 1236, 1175, 1097, 1054, 1016, 977, 892, 866, 814, 749, 705, 682 cm−1; 1H-n class="Chemical">NMR see Table 3; 13C-NMR (CDCl3) δ 162.3, 160.6, 159.0, 155.3, 154.0, 153.7, 136.5, 133.0, 132.3, 131.1 (×2), 130.4, 130.0, 125.2, 124.4, 124.2, 117.6; EI-MS (relative intensity) m/z 326 (M+, 24), 181 (90), 180 (31), 147 (26), 130 (100), 123 (26), 102 (32), 57 (23); Exact mass calcd for C19H10N4O2 m/z 326.0804 (M+), EI-HRMS m/z 326.0799.

The characteristic 1H-n class="Chemical">NMR absorptions for compounds 7a–d.

3.4. General Procedures for DNA-Cleavage by Use of Oxime Esters

A reaction mixture (10.0 μL), containing supercoiled circular ϕX174 RFI DNA stock solution (form I, 50.0 μM/base pair) and an n class="Chemical">oxime ester (100 μM), was dissociated in the phosphate buffers (pH 6.0) and DMSO (0.25 μL) in a Pyrex vial. It was then preincubated at 37.0 °C for 30.0 min and irradiated with UV light (312 nm, 1.43 mW/cm2) under aerobic conditions at room temperature for 2.0 h. After addition of gel-loading buffer (2.50 μL containing 0.25% bromophenol blue, 0.25% xylene cyanol, and 30.0% glycerol), the reaction mixture was loaded on a 1.0% agarose gel with ethidium bromide staining. The electrophoresis tank was attached to a power supply at a constant current (~100 mA). The gel was visualized by 312 nm UV transilluminator and photographed by a Canon PowerShot S5 IS digital camera. Quantitation of DNA-cleavage was performed by integration of the optical density as a function of the band area by use of a Scion image beta 4.03 program.

3.5. Photolysis of 2,6-Diazaanthracene-9,10-dione bis-(O-benzoyloxime) (4a)

A mixture of compound 4a (40 mg, 0.089 mmol) and 1,4-cyclohexadene (0.10 mL, 1.057 mmol) in n class="Chemical">benzene (5 mL) was placed in a quartz tube and was irradiated with a medium pressure mercury lamp (222–366 nm) under nitrogen for 4 h. The solvent was then removed by a rotary evaporator, and the residue was dissolved in CH2Cl2 (10 mL) and was then treated with 5% NaOH (10 mL). The organic solution was concentrated and then purified by flash column chromatography using ethyl acetate/hexane (1:1) as eluent to give 2,6-diazaanthracene-9,10-dione (2, 15.8 mg, 84%). The aqueous solution was neutralized with 5% HCl, extracted with CH2Cl2 (10 mL), dried (MgSO4) and evaporated to give benzoic acid (8, 17.4 mg, 80%).

4. Conclusions

In conclusion, oxime esters 4a–d and n class="Chemical">7a–d newly synthesized from 2,6-diazaanthracene-9,10-dione (2) and 4,5-diazafluoren-9-one (5) were found to possess DNA cleaving ability upon UV irradiation. All of these bis-oxime esters 4a–d showed greater cleaving ability than the mono-oxime esters 7a–d. Upon UV irradiation, the most potent compound 4c caused significant amounts of single-strand cleavage at the concentration of 100 μM. Results from our mechanistic study indicate that the iminyl and carboxyl radical species are responsible for DNA nicking.