Synthesis and Cytotoxicity Evaluation of Naphthalimide Derived N-Mustards

A series of N-mustards, which was conjugated to mono- or bis-naphthalimides with a flexible amine link, were synthesized and evaluated for cytotoxicity against five cancer cell lines (HCT-116, PC-3, U87 MG, Hep G2 and SK-OV-3). Several compounds displayed better activities than the control compound amonafide. Further evaluations by fluorescence spectroscopy studies and DNA-interstrand cross-linking assays revealed that the derivatives showed both alkylating and intercalating properties. Among the derivatives, the bis-naphthalimide N-mustard derivative 11b was found to exhibit the highest cytotoxic activity and DNA cross-linking ability. Both 11b and 7b induce HCT-116 cell apoptosis by S phase arrest.


Introduction
As the earliest agents used for chemotherapy, DNA alkylating nitrogen mustards have been widely utilized in oncology treatment, and many such agents are still in clinical use ( Figure 1) [1][2][3][4][5]. However, progress in developing new N-mustard agents is limited due to its drawbacks [6]. Its necessarily high chemical reactivity leads to serious adverse effects [3,7,8] by randomly alkylating other cellular nucleophiles. They lack specific affinity to tumor cells and induce bone marrow toxicity [6,9]. To overcome those drawbacks, one of the strategies has been to synthesize bifunctional compounds by linking N-mustards with DNA-affine molecules, such as DNA-intercalators (e.g., acridines [6,9,10], cyclopentanthraquinone [11]) or DNA minor groove binders (e.g., distamycin A and related analogues [12][13][14]). Previous research has demonstrated that linking with an appropriate carrier can modify the specificity of DNA alkylation and thus improve the therapeutic efficacy of N-mustard agents. This strategy has been widely applied in the search for new drugs [9,15,16]. Naphthalimide, a well-defined DNA-intercalator, has been extensively investigated as an antitumor agent [17]. Several naphthalimide derivatives such as mitonafide, amonafide, ethonafide, elinafide, and bisnafide ( Figure 2) are currently undergoing clinical trials, [17]. Elinafide and bisnafide, which are bisintercalators prepared by linking two naphthalimide groups with a polyamine linker, showed much higher cytotoxicity than the mono-intercalators mitonafide and amonafide [18,19]. However, only limited research have been done in the area of linking N-mustard with DNA-binding naphthalimide, and they showed enhanced cytotoxicities [20][21][22][23].
Herein, we designed and synthesized N-mustard derivatives of naphthalimides by linking the N-mustard moiety with naphthalimides. The conjugates acted as both alkylating agents and intercalators. Encouraged by the generally stronger cytotoxicities and high binding capacities of the bis-intercalators, we decided to synthesize both mono and bis-naphthalimide [18] conjugates to compare their cytotoxic activities.

Chemistry
To attach the N-mustard moiety to the naphathalimide core, we synthesized compound 2 from naphthalic anhydride [24,25]. Compound 2 was refluxed with an aminoalcohol in CH 3 CN to yield the key precursor compound 3, which was converted to compound 4 by refluxing in HCHO and HCOOH. The chlorination of alcohol 4 in the presence of SOCl 2 to produce 5 was successful [26]. The N-mustard precursor 6 was synthesized by condensing compound 5 with diethanolamine in the presence of K 2 CO 3 /KI [27]. The final target compound 7 was obtained from 6 by reacting with SOCl 2 (Scheme 1).

Cytotoxic Activity
The N-mustard naphthalimide derivatives 7a-b, 11a-b and the precursors 6a-b, 10a-b in their dihydrochloride forms were screened for their cytotoxic activities against five cancer cell lines, which were HCT-116 (colorectal carcinoma), PC-3 (prostate carcinoma), U87 MG (brain tumor), Hep G2 (liver cancer), and SK-OV-3 (ovarian cancer), with an in vitro cell growth assay using amonafide as positive control. The results were listed in Table 1. The activity of compound 11b was the highest among all the compounds and was increased by several fold (1.4-11) compared to amonafide, while compound 7b showed a similar activity with amonafide. N-mustards 7b and 11b displayed more activity than their precursors 6b, 10b. Furthermore, compared with compounds 7b and 11b, 7a and 11a unexpectedly showed much lower activities (>10 μM). Because quaternary ammonium salts of 7a and 11a were used in this cytotoxicity assay, the six-membered rings in 7a and 11a, which make them less flexible, may be the reason for the lower activities. We also screened all the synthetic intermediates in its mono-hydrochloride form against the same five cancer cell lines (Table 2). It was unexpected to find that compounds 8a and 8b which acted only as DNA-intercalators, exhibited close activities to 11b. These results indicated that N-mustards can increase the cytotoxic activity of the naphathalimide derivatives, but not efficiently.

Fluorescence Studies
A fluorescence study was employed to evaluate the interaction properties of the synthesized derivatives with DNA. Figure 3 illustrates the changes of fluorescence properties of compounds 6b, 7b, 10b and 11b before and after being mixed with calf thymus DNA (ctDNA).The solutions of various concentrations of compounds with ctDNA were incubated at 30 °C for 3 days, then the florescence spectra were recorded. We found that the fluorescence intensity of compound 6b showed little change with or without ctDNA at all five concentrations, indicating a weak intercalating ability. On the contrary, compound 7b exhibited an evident decrease of fluorescence intensity when ctDNA was added, which demonstrated the N-mustard is essential to increase the binding strength to DNA. Both compounds 10b and 11b showed obvious decreases of fluorescence intensity, but the difference between them is not significant. Those results indicate that: (1) an N-mustard residue is able to stabilize the DNA-compound complex after the intercalating process, and the binding potency is enhanced for N-mustard derivatives; (2) bis-naphthalimides 10b, 11b show generally much higher intercalating potential than mono-naphthalimides such as 6b, 7b, which is in accordance with previous research [19]. Conditions: compound and 2 μL supercoiled plasmid DNA (pUC-19, 0.16 μg/μL) were incubated at 37 °C for 1 h in Tris-HCl buffer (0.05 M, pH = 7.4). Con: precursor of each specific nitrogen mustard (6a/b, 10a/b). Concentrations of drugs, from left to right: 1, 5, 10, 50 μM. All the compounds used in the assay are in their dihydrochloride form.

DNA Interstrand Cross-Linking Assay
To evaluate the potency of DNA interstrand cross-linking brought by the N-mustard moiety [15], compounds 7a, b and 11a, b were selected to undergo an agarose gel cross-linking assay, at four concentration levels ( Figure 4). Considering that naphthalimide chromophore can intercalate with DNA and so slow down DNA migration on the gel, 6a,b and 10a,b were employed, respectively, as controls for each mustine derivative. Compounds 7a, 7b and 11b displayed DNA cross-linking potency at 1, 1 and 5 μM, respectively; while 11a displayed almost no DNA cross-linking ability. The results for 7b, 11a and 11b were consistent with their cytotoxic data. However, different from the cytotoxicity data, 7a exhibited high DNA cross-linking activity, even at 1 μM concentration. The different performance of 7a in cells and under agarose gel (with Tris-HCl buffer) conditions may result from its insufficient transmembrane access to the nucleus. The controls 10a and 10b displayed no DNA cross-linking abilities when compared with the N-mustards 11a and 11b, which showed similar activities in the previous tests. This result suggested the possibility that the N-mustard moiety did not show significant differences in in vitro tests, but played an efficient role in interacting with DNA.

Flow Cytometry Assay
Compounds that target DNA by intercalating or cross-linking often interfere with cellular DNA replication, thus inhibiting cell proliferation by inducing cell cycle arrest. To evaluate the effects of the compounds on cell cycle progression, the cell cycle distribution of HCT-116 cells was examined by flow cytometry after treatment with indicated concentrations of compounds 7b, 9b and 11b for 24 h. As shown in Figure 5, compounds 7b and 11b induced G2/M and S phase cell cycle arrest, suggesting that these compounds could interfere with DNA synthesis. Compound 11b showed a more pronounced arrest in G2/M phase at 1 μM than 7b, which is in agreement with the higher cytotoxicity associated with 11b. However, compound 9b, a homo-N-mustard, exhibited marginal effects on the cell cycle distribution, even at 5 μM, implying the N-mustard moiety is essential for inhibition of DNA synthesis.

Figure 5. Effects of compounds on cell cycle distribution of HCT-116 cells.
Conditions: Cells were incubated with indicated concentrations of 7b, 9b, 11b for 24 h, then cells were stained by PI (20 μg/mL) and tested by flow cytometry.

Immunoblotting Assay
Poly(ADP-ribose) polymerase (PARP-1) is specifically cleaved by caspases during the execution phase of apoptosis in response to DNA damage, thus its cleavage has been regarded as a biomarker of apoptosis [28]. Therefore, the cleavage of PARP-1 in HCT-116 cells treated with various compounds was detected by western blotting. As shown in Figure 6, both 7b and 11b induced significant cleavage 7b 9b 11b of the original 116 kDa PARP-1 into 89 kDa and 24 kDa fragments, suggesting that these two compounds induced DNA-damage related apoptosis. In agreement with the results above, the homomustard 9b showed no observable effect on the cleavage of PARP-1. Taken together, the above results suggest that compounds 7b and 11b induced DNA damage-related cell cycle arrest and apoptosis in HCT-116 cells, but 9b might have a different mechanism of action. Conditions: HCT-116 cells were treatment for 24 h in the presence of indicated concentrations of 7b, 9b, 11b, and then cleavage of PARP1 was detected by immunoblotting. GAPDH, a house-keeping gene product, was used as loading control.

General Information
All the solvents were of analytical grade. 1 H-NMR and 13 C-NMR spectra were obtained with JEOL-300 all examples below are 300 MHz but in Supplementary file spectra are labelled as 400 Mhz. Explain this confusing situation spectrometers. The chemical shifts were reported in ppm using TMS as internal standard. High resolution mass spectrometry was measured on a Bruker MicrOTOF-Q. Column chromatography was conducted on silica gel (200~300 mesh). Temperature range for rt is 25 ± 2 °C. [24]. A mixture of 1,8-naphthalic anhydride (4 g, 20.8 mmol) and 3-amino-1-propanol (1.86 mL, 24.6 mmol) in ethanol (40 mL) was heated under reflux for 5 h. The resulting mixture was concentrated by evaporating ethanol under reduced pressure to afford a solid residue, which was purified by column chromatography over silica gel (CH 2 Cl 2 /MeOH = 1:1) to afford compound 1 as an off-white solid (4.27 g, 83%). (2) [25]. To a solution of compound 1 (2 g, 7.84 mmol) and NBS (2.12 g, 11.9 mmol) in dichloromethane (20 mL, dry) was added PPh 3 (

N,N-Bis[2-(1,3-dioxo-2,3-dihydro-1H-benz[de]-isoquinolin-2-yl)propyl]-N',N'-bis(2-hydroxyethyl)-
ethylenediamine (10a). The synthesis of compound 10a was divided into two parts. Firstly, compound 8a (560 mg, 1.05 mmol) was employed to produce 9a through a similar procedure as 9b. Then compound 9a was directly used for the next step. A mixture of 9a (350 mg, 0.63 mmol), K 2 CO 3 (540 mg, 3.91 mmol), KI (100 mg, 0.60 mmol) and diethanolamine (0.8 mL) in CH 3 CN (10 mL) was refluxed under N 2 for 60 h. The resulting mixture was filtrated to remove solid K 2 CO 3 , and then concentrated under reduced pressure. The crude product was purified by column chromatography over silica gel (CHCl 3 /MeOH = 30:1) to yield 10a (light yellow oil, 86.0%). 1 (1 mL) was stirred at room temperature overnight. The remaining SOCl 2 was completely removed by adding Et 2 O and then condensed for several times in vacuo. The residue was dissolved in a mixed solution of methanol and a little methylene chloride, then a drop of concentrated hydrochloric acid was added to afford a white solid precipitate. The suspension was filtered to yield crude product 11a hydrochloride as white solid (30 mg, 51.0%). 1 . To a solution of compound 10b (190 mg, 0.30 mmol) in CHCl 3 (25 mL) was added SOCl 2 (0.15 mL) dropwise and the mixture stirred at 65 °C for 2 h. The solvent was removed under reduced pressure, then water (20 mL) and solid Na 2 CO 3 were added to neutralize the solution, which was extracted with CH 2 Cl 2 (50 mL). The organic layer was washed with saturated NaCl, dried over MgSO 4 , concentrated and then purified by column chromatography over silica gel (CHCl 3 /MeOH = 50:1) to yield 11b (yellow oil, 77.0%). 1

Fluorescence Study Experiment
Each compound was divided into two groups: one was added with a constant concentration of ctDNA (calf thymus DNA, 50 μM) and the other was without the ctDNA as control. Various concentrations of compounds ranging from 1 μM to 25 μM (1, 5, 10, 20 and 25 μM) were used. To every sample was added Tris-HCl buffer (25 mM, pH = 7.4) to get a final volume of 4 mL, which was stirred constantly at 30 °C in the dark for 3 days before recording each spectrum.

Agarosegel Cross-Linking Assay
Each compound was tested at four concentrations (1, 5, 10, 50 μM). To each sample was added 2 μL plasmid DNA (PUC-19, 0.16 μg/μL) and various concentrations of compounds, diluted by Tris-HCl buffer (0.05 M, pH = 7.4) to reach a final volume of 15 μL. Every group has a pure DNA sample as blank control and its own control sample by using its own precursor in concentration of 10 μM. All the above samples were incubated at 37 °C for 1 h and separated on an agarose gel (120 V, 45 min).

In Vitro Cytotoxicity Evaluation
All the compounds were tested in their mono-or dihydrochloride form. The samples were prepared by adding a stoichiometric amount of 1 M HCl to the solution, and then removing the solvent to obtain the target compounds. All cells used in the research were prepared at 3.5 × 10 4 cells/mL concentration and each 100 μL cells suspension was seeded in 96-well cell incubated for 24 h (37 °C, 5% CO 2 ). Then each solution was added and incubated for another 72 h. For the control group, equivalent concentration of DMSO (final concentration 0.5%) was added. MTT (3-[4,5-dimethylthiazol-2yl]diphenyltetrazolium bromide) method was employed to measure the number of surviving cells and the OD value was recorded at 492 nm/620 nm. The IC 50 values were calculated using Prism Graphpad software of the triplicate experiment.

Flow Cytometry Assay
Cells were trypsinized and washed twice with PBS, then fixed in 75% alcohol for at least 30 min at 4 °C. The fixed cells were recovered by centrifugation and washed twice with PBS, and then stained with PI in PBS containing RNAase and 0.1% Triton-X100 for 30 min at 37 °C. The cellular DNA content was measured by a FACScalibur flow cytometer (BD Bioscience, San Jose, CA, USA).

Immunoblotting Assay
Cells were lysed using RIPA lysis buffer containing PMSF and phosphatase/protease inhibitors. The total cellular protein extracts was qualified by BCA assay. 20 μg of total protein was resolved on 10% SDS-PAGE, then electro-transferred onto PVDF membranes, and incubated with appropriate antibodies overnight at 4 °C, then washed in PBS containing 0.1% Tween 20 and incubated with corresponding secondary antibodies for 1 h at room temperature. After that the membranes were developed with the enhanced chemiluminescence Western Blotting detection reagent (Amersham-Pharmacia, Piscataway, NJ, USA).

Conclusions
A series of mono-and bisnaphthalimide N-mustard derivatives were synthesized as potent antitumor agents. Through conjugating N-mustard to naphthalimide, improved cytotoxicity was obtained. Compound 11b exhibited considerably cytotoxicity compared with amonafide, and showed potency as an antitumor agent. Even though more investigation is still needed to evaluate the efficiency of the N-mustard moiety, the bioactivity results suggest that the length of the flexible amine link and naphthalimide core could improve the cytotoxicity. Our work is meaningful in compensating for the lack of research in this field of combining N-mustards with naphthalimides.