Role of the nitro functionality in the DNA binding of 3-nitro-10-methylbenzothiazolo[3,2-a]quinolinium chloride

Abstract

Interest in DNA binding drugs has increased in recent years due to their importance in the treatment of genome-related diseases, like cancer. A new family of water-soluble DNA binding compounds, the benzothiazolo[3,2- a]quinolinium chlorides (BQCls), is studied here as potential candidates for chemical treatment of solid tumor cells that may encounter low-oxygen environments, a condition known as hypoxia. These compounds are good DNA intercalators; however, no studies have been made of these compounds under hypoxic conditions. This work demonstrates the importance of the nitro-functionality in the DNA binding of 3-nitro-10-methylbenzothiazolo[3,2- a]quinolinium chloride (NBQ-91), which possesses nitro-functionality, and 10-methylbenzothiazolo[3,2- a]quinolinium chloride (BQ-106), which does not. Both NBQ-91 and BQ-106 have similar noncovalent binding affinity toward DNA. Dialysis experiments show that NBQ-91 binds DNA under N2-saturated conditions with increasing concentrations of reducing agent, presumably due to reduction of the nitro-functionality. Conversely, because of the lack of nitro-functionality, the presence of a reducing agent had no effect on BQ-106 binding to DNA under both aerobic and N2-saturated conditions. Clonogenic assays were performed to determine the quinolinium chloride cytotoxicities under both aerobic (95% air and 5% CO2) and hypoxic (80% N2 and 20% CO2) conditions. The calculated ratios of cellular toxicity under aerobic to hypoxic conditions caused by the same concentration of test agent (CTR values) show greater levels of cell death under hypoxia than under aerobic conditions for mitomycin C (MC) (CTR = 0.7 at 1 microM) and NBQ-91 (CTR = 0.4 at 200 microM) than for BQ-106 (CTR = 1.0 at 200 microM), which agreed with the previously reported data for MC and confirmed the importance of nitro-functionality for reactivity under hypoxic conditions. There was no correlation between noncovalent binding affinity constants and their cytotoxicity under hypoxic conditions. Adduct formation between the NBQ-91 and 2'-dG was also assessed by reacting 2'-dG or DNA with NBQ-91 and BQ-106 under N2-saturated conditions in the presence of hypoxanthine and xanthine oxidase (HX/XO). DNA covalent adduct formation was analyzed by two techniques: LC-ESI-MS and Sephadex size exclusion chromatography. LC-ESI-MS results clearly indicate the formation of a prominent molecular ion at masses of 266.0 and 530.58 Da, corresponding to the [M + H](+2) and [M](+) molecular ions of the monitored 2'-dG-NBQ-91 adduct. Results from the Sephadex size exclusion chromatography support these findings because the NBQ-91 elution percentage increases in the presence of HX/XO due to the reduction of the nitro-functionality, which results in covalent binding to DNA. This study reports evidence of the DNA binding capacity of this bioreductive drug. The preferential N2-saturated over aerobic conditions for DNA binding makes NBQ-91 a potential bioreductive compound for hypoxic cell killing.

Figure 1

Quinolinium salts used in this study.

Figure 2

Absorption spectra of solutions containing 30 mM phosphate buffer (pH 7.4) and (A) NBQ-91 without HX/XO (continuous line), NBQ-91 and 2 mol of HX per mol of NBQ-91, and 20 mU XO/mL (broken line) and (B) BQ-106 without HX/XO (continuous line), BQ-106 and 2 mol of HX per mol of BQ-106, and 20 mU XO/mL (broken line).

Figure 3

Uric acid formation after quenching with 100 µM allopurinol a N₂-saturated reaction containing 500 µ HX, 50 mU XO/mL, and 100 µM NBQ-91 (squares) or BQ-106 (circles) in the presence (black symbols) or absence (open symbols) of 2.0 mM DNA in 30 mM phosphate buffer, pH 7.4.

Figure 4

Absorption spectra before (continuous line) and after (broken line) dialysis of N₂-saturated solutions containing 30 mM phosphate buffer (pH 7.4) and (A) 100 µM NBQ-91, 200 µM HX, 20 mU XO/mL, and 2.0 mM DNA and (B) 100 µM BQ-106, 200 µM HX, 20 mU XO/mL, and 2.0 mM DNA. DNA was added first in both samples A and B.

Figure 5

Change in absorbance (ΔA = A_{with HX} − A_{wo HX}) of N₂-saturated (black symbols) and air-saturated (open symbols) solutions containing 30 mM phosphate buffer (pH 7.4), 0–200 µM HX, 2.0 mM DNA, 0–20 mU XO/mL, and 100 µM NBQ-91 (□) or 100 µM BQ-106 (Δ).

Figure 6

Chromatographic analysis for the reaction of NBQ-91 with 2′-dG with the highest HX:NBQ-91 molar ratio (6:1). The presences of XO (2.10 min), HX (2.48 min), NBQ-91 (7.47 min), and 2′-dG (12.20 min) are clearly observed. Small unknown peaks were also present. HPLC analysis was monitored at a wavelength range of 254–370 nm. The stationary phase used was a Supelcosil LC-18-S 15 cm × 3 mm, 5 µM (Supelco). From each reaction, 20 µL injections were eluted using a 30 min nonlinear gradient from 98:2 (H₂O/MeOH) to 100% B in 25 min at a flow rate of 350 µL/min. The adduct detection region refers to the area where the [M + H]²⁺ molecular ion reported in Figure 8 was observed upon LC-ESI-MS analysis. A region rather than a specific RT is mentioned given the slight variations in the exact retention time observed within reactions with different HX:NBQ-91 molar ratios for the detection of the monitored molecular ion.

Figure 7

Full scan positive ESI-MS analysis of the parent compound NBQ-91 through direct infusion through capillary tubing at a flow rate of 5 µL/min. The MW of the parent NBQ-91 containing the Cl-anion (330.7 Da) was not observed. Instead, a molecular ion at m/z 295.4 consistent with the [M]⁺ ion of the NBQ-91 (exact mass, 295.3 Da) after the loss on the chloride anion was observed. Fragments of the parent were also observed (not presented) as the capillary and cone voltages changed. The ESI-MS conditions used were as follows: drying gas, N₂; flow rate, 125 L/h; capillary voltage, 6.08 kV; cone voltage, 70 V; extractor, 2 V; source temperature, 100 °C; and desolvation temperature, 200 °C.

Figure 8

Full scan positive ESI-MS analysis of the reaction of NBQ-91 with 2′-dG after HPLC separation. The molecular ion at m/z 266.0 represents the [M + 1]²⁺ ion of the proposed 2′-dG adduct MW 530.6 obtained upon chromatographic analysis as in Figure 6. The monitored adduct was detected in the reaction where the HX:NBQ-91 molar ratio was the highest (6:1). ESI-MS conditions used were as follows: drying gas, N₂; flow rate, 200 L/h; capillary voltage, 3.08 kV; cone voltage, 70 V; extractor, 5 V; source temperature, 100 °C; and desolvation temperature, 200 °C.

Figure 9

Chromatographic analysis for the reactions of NBQ-91 with DNA with the highest HX:NBQ-91 molar ratio (6:1). Samples were enzymatically digested prior to the HPLC analysis. The chromatographic analysis presents the separation of the digestion enzymes, NBQ-91 and nucleosides. Small unknown peaks (potential adducts) were also observed. HPLC analysis was monitored at a wavelength range of 254–370 nm. The adduct-detected region refers to the area where the molecular ions reported in Figure 10 were obtained upon LC-ESI-MS analysis. A region rather than a specific RT is mentioned given the slight variations in the exact retention time observed within reactions with different HX:NBQ-91 molar ratios for the detection of the monitored molecular ion. The stationary phase used was a Supelcosil LC-18-S 15 cm × 3 mm, 5 µM (Supelco). From each reaction, 20 µL injections were eluted using a 30 min nonlinear gradient from 98:2% (H₂O/MeOH) to 100% B in 25 min at a flow rate of 350 µL/min.

Figure 10

Full scan positive ESI-MS analysis of the reaction mixture of NBQ-91 with DNA after enzymatic digestion and HPLC separation. As with the 2′-dG reactions, ESI-MS analysis was applied to detect the formation of a 2′-dG-NBQ-91 adduct at similar mass. A prominent ion at m/z 266.0 corresponding to the [M + 1]²⁺ molecular ion of 2′-dG-NBQ-91 adduct was clearly detected. The molecular ion at m/z 266.0 represents the ion of the proposed 2′-dG adduct after chromatographic analysis presented in Figure 9. The molecular ion [M]⁺ at 530.7 m/z but could represent the unprotonated adduct carrying its natural positive charge. In this reaction, the HX:NBQ-91 molar ratio was also 6:1 as in the 2′-dG-NBQ-91 reaction presented above. ESI-MS conditions used were as follows: drying gas, N₂; flow rate, 200 L/h; capillary voltage, 3.08 kV; cone voltage, 70 V; extractor, 5 V; source temperature, 100 °C; and desolvation temperature, 200 °C.

Figure 11

Some possible structures of 2′-dG-NBQ-91 adducts with molecular mass of 530.58 Da, which could be produced after reduction of NBQ-91 to the corresponding hydroxylamine followed by reaction with 2′-dG.

Scheme 1

Redox Cycling of NBQ-91 in the Presence of HX/XO