Benzodioxin-Annulated Naphthalimides as Potent DNA Replication Stress Inducers with Dual p53-Dependent and Independent Antitumor Activity

Abstract

Background/Objectives: The development of small-molecule agents that selectively target DNA replication remains a central strategy in anticancer drug discovery. In this study, we report the biological characterization of a novel 6-nitro-benzodioxin-naphthalimide (NI) derivative (compound 5a), evaluated as a potential DNA-targeted anticancer lead. Methods/Results: The antiproliferative activity of 5a was assessed in a small panel of human lung carcinoma cell models (A549, H1299) and a non-malignant control (MRC-5), revealing pronounced cytotoxic effects in tumor cells, accompanied by favorable selectivity indices. Mechanistic investigations demonstrated that treatment with 5a results in strong inhibition of DNA synthesis, as evidenced by a marked reduction in EdU incorporation and a robust induction of the DNA damage marker γH2AX. These effects were associated with cell-cycle perturbations characterized by accumulation in G₁ and G₂/M phases, followed by activation of apoptotic pathways. Importantly, clonogenic survival assays confirmed that even transient exposure to 5a leads to a sustained loss of proliferative capacity, indicating irreversible long-term cellular damage. These results support a replication stress-driven mechanism of action for compound 5a, consistent with interference in DNA-associated processes during S phase. Conclusions: While the precise molecular initiating event remains to be elucidated, the observed biological profile positions 5a as a promising DNA-targeted lead structure with potential for further pharmaceutical optimization. These findings provide a solid foundation for the continued development of naphthalimide-based compounds as anticancer agents within a pharmaceutically relevant framework.

1. Introduction

Over the past five decades, 1,8-naphthalimide derivatives have attracted sustained scientific interest owing to their unique DNA-interactive properties, structural tunability, and broad biological potential. The search for affordable, potent, and selective anticancer agents remains a central objective in medicinal chemistry, particularly in light of persistent challenges related to drug resistance, off-target toxicity, and unfavorable pharmacokinetic profiles of many clinically used chemotherapeutics [1,2]. DNA-interacting small molecules constitute one of the most extensively investigated classes of anticancer agents. Typically, these compounds contain planar tri- or tetracyclic chromophores that enable π–π stacking interactions with DNA base pairs. Such non-covalent electrostatic and hydrophobic interactions can induce DNA unwinding, elongation, or stiffening, thereby perturbing essential enzymatic processes including replication and transcription [3,4,5]. Consequently, many DNA-interacting agents inhibit the activity of topoisomerase I and II, enzymes that play a critical role in maintaining DNA topology during cell-cycle progression [6,7]. Within this context, the 1,8-naphthalimide pharmacophore represents one of the most versatile scaffolds among DNA-interacting compounds and has been intensively explored for anticancer drug design [8,9]. The flat aromatic core of naphthalimides favors DNA association, while substitution at the imide nitrogen enables the introduction of diverse side chains to modulate physicochemical and pharmacological properties [10,11]. Classical representatives such as mitonafide and amonafide advanced to clinical evaluation; however, their development was ultimately discontinued due to dose-limiting hematological and neurotoxic side effects [12,13,14,15].

Consequently, current research efforts focus on rational structural modification of the naphthalimide scaffold, including heterocyclic fusion, incorporation of electron-donating or -withdrawing substituents, and hybridization with additional bioactive motifs, with the aim of enhancing anticancer efficacy while reducing systemic toxicity [11,16,17]. Recent advances in synthetic methodologies have enabled gram-scale preparation of heterocyclic-extended naphthalimide derivatives, further expanding their accessibility for biological evaluation [11,18]. Despite these advances, the biological properties of dioxin-extended naphthalimides remain insufficiently explored, particularly with respect to structure–activity relationships, mechanisms of cytotoxicity, and selectivity toward tumor cells with different p53 status.

The tumor suppressor protein p53 plays a pivotal role in coordinating cellular responses to DNA damage by regulating cell-cycle checkpoints, DNA repair pathways, and apoptosis [19]. Given that p53 mutations or deletions occur in more than half of human cancers, including lung carcinoma [20,21], the identification of agents that retain anticancer activity independently of p53 status is of substantial therapeutic interest. In this regard, naphthalimide derivatives that activate both p53-dependent and p53-independent cytotoxic pathways may overcome a major limitation of conventional DNA-targeting chemotherapeutics.

Beyond classical DNA interactions, 1,8-naphthalimide derivatives exhibit multifaceted biological activities, including modulation of p53 signaling [22], activation of checkpoint pathways, such as ATM/ATR–Chk2, leading to S- or G₂/M-phase arrest [23], and induction of mitochondrial apoptosis, with subsequent caspase-3/7 activation [16,24]. Alternative cytotoxic mechanisms, such as lysosomal membrane permeabilization and ferroptosis, have also been reported for selected analogs [25]. In parallel, the intrinsic fluorescence of the naphthalimide scaffold has facilitated the development of theranostic platforms combining imaging and cytotoxicity, including enzyme-activated probes, peptide-conjugated systems, and multifunctional fluorescent agents [26,27,28,29]. In this study, we report the design, synthesis, and comprehensive biological evaluation of a new class of heterocyclic-extended 1,8-naphthalimide derivatives, exemplified by the lead compound 5a (benzodioxino-6-nitro-NI). Compound 5a exhibits sub-100 nM cytotoxicity against A549 and H1299 lung carcinoma cells while maintaining favorable selectivity toward non-malignant MRC-5 fibroblasts. Using an integrated set of cellular assays—including proliferation analysis, clonogenic survival, cell-cycle profiling, EdU incorporation, γH2AX foci formation, caspase activation, and p53 accumulation—we delineate a replication stress-associated mechanistic cascade culminating in cell-cycle arrest and caspase-dependent apoptosis. Notably, 5a elicits distinct cellular responses in p53-proficient (A549) and p53-deficient (H1299) cells, indicating the engagement of both p53-dependent and p53-independent pathways. Collectively, these findings position compound 5a as a potent and selective anticancer candidate within the naphthalimide family and place the present work within the broader framework of next-generation naphthalimide-based therapeutics [16].

2. Materials and Methods

2.1. Synthesis

All starting materials and solvents were commercially available and used without additional purification Fluorochem (Glossop, UK), Across (Antwerpen, Belgium), and Fisher Scientific (Hampton, NH, USA).

NMR spectra were recorded on a Bruker Avance 500 MHz instrument (Bruker, Karlsruhe, Germany) operating at 500 and 126 MHz for ¹H and ¹³C, respectively. Chloroform-d and trifluoroacetic acid-d were used as solvents. Chemical shifts are reported in δ units (ppm) and referenced to the residual solvent signals (¹H at 7.26 ppm and ¹³C at 77.160 ppm for Chloroform-d and ¹H at 11.50 ppm and ¹³C at 164.20 ppm for trifluoroacetic acid-d). HRMS were recorded on a ThermoFisher Scientific Orbitrap Exploris 120 (Source—HESI APCI, Comby Nozzle, Bremen, Germany). Elemental analyses were carried out on a Leco CHNS-932 (Leco Europe, Geleen, The Netherlands). Thin layer chromatographic (TLC) analysis was performed on silica gel plates (Macherey-Nagel F60 254 40 × 80; 0.2 mm, Macherey-Nagel, Duren, Germany). HPLC analysis was performed on a Purospher^® STAR RP-18 column (150 × 4.6 mm, 5 µm). The mobile phase consisted of 0.1% TFA in water and acetonitrile (50:50, v/v), delivered at a flow rate of 1.0 mL/min. The column temperature was maintained at 25 °C, and UV detection was carried out at 254 nm.

General procedure for the preparation of esters 3a–e: A mixture of dibutyl 3,4-dibromo-6-nitronaphthalene-1,8-dicarboxylate (15.0 mmol, 7.97 g) or dibutyl 3,4-dibromo-naphthalene-1,8-dicarboxylate (15.0 mmol, 7.29 g), catechol or 2,3-dihydroxinaphthalene or 4-nitrocatechol (18.0 mmol) and potassium carbonate (36 mmol, 4.97 g) in 40 mL NMP was stirred and heated at 150 °C for 1–2 h. The mixture was cooled down to room temperature and poured into 50 mL of cold water containing 5 mL of concentrated hydrochloric acid. The precipitation was filtered, washed with water, and dried. The crude product was purified by column chromatography on silica gel using cyclohexane/dichloromethane as the eluent; the chromatographic system was started with pure cyclohexane, and the polarity was gradually increased by addition of dichloromethane up to 40%.

2.1.1. Synthesis of Dibutyl 2-Nitrobenzo[b]naphtho [1,2-e][1,4]dioxine-4,5-dicarboxylate 3a

Yield 6.69 g (93%) as a yellow solid. ¹H NMR (Chloroform-d, 500 MHz): 9.15 (d, J = 2.3 Hz, 1H), 8.58 (d, J = 2.3 Hz, 1H), 7.76 (s, 1H), 7.08–7.06 (m, 1H), 7.03–6.99 (m, 2H), 6.94-6.90 (m, 1H), 4.32 (dt, J = 13.9, 6.8 Hz, 4H), 1.81–1.73 (m, 4H), 1.52–1.43 (m, 4H), 1.00 (t, J = 7.4 Hz, 3H), 0.99 (t, J = 7.4 Hz, 3H). ¹³C{1H} NMR (Chloroform-d, 126 MHz): 167.11, 166.93, 145.03, 141.57, 141.34, 139.75, 138.56, 132.86, 127.34, 126.20, 125.59, 125.22, 124.64, 124.57, 122.11, 120.52, 117.09, 116.93, 66.19, 65.89, 30.70, 30.65, 19.35, 13.90. Anal. calcd. C₂₆H₂₅NO₈ C, 65.13; H, 5.26; N, 2.92%; found C, 64.98; H, 5.47; N, 3.13%.

2.1.2. Synthesis of Dibutyl 2-Nitrodinaphtho[1,2-b:2′,3′-e][1,4]dioxine-4,5-dicarboxylate 3b

Yield 7.15 g (90%) as a yellow solid. ¹H NMR (Chloroform-d, 500 MHz): 9.09 (d, J = 2.4 Hz, 1H), 8.51 (d, J = 2.4 Hz, 1H), 7.80 (s, 1H), 7.65–7.63 (m, 1H), 7.59–7.58 (m, 1H), 7.44 (s, 1H), 7.36–7.32 (m, 2H), 7.20 (s, 1H), 4.34 (dt, J = 8.9, 6.8 Hz, 4H), 1.83–1.76 (m, 4H), 1.54–1.46 (m, 4H), 1.01 (t, J = 7.4 Hz, 6H). ¹³C{1H} NMR (Chloroform-d, 126 MHz): 167.09, 166.93, 144.87, 140.41, 140.36, 138.86, 137.76, 132.73, 131.27, 131.09, 127.21, 127.18, 127.15, 126.28, 126.23, 126.06, 124.58, 124.54, 122.05, 120.24, 113.32, 113.00, 66.18, 65.92, 30.72, 30.68, 19.37, 19.35, 13.92. Anal. calcd. C₃₀H₂₇NO₈ C, 68.05; H, 5.14; N, 2.65%; found: 68.22; H, 5.28; N, 2.83%.

2.1.3. Synthesis of Dibutyl 2,10-Dinitrobenzo[b]naphtho[1,2-e][1,4]dioxine-4,5-dicarboxylate 3c

Yield 5.74 g (73%) as a yellow solid. ¹H NMR (Chloroform-d, 500 MHz): 9.14 (d, J = 2.4 Hz, 1H), 8.63 (d, J = 2.4 Hz, 1H), 8.02 (d, J = 2.5 Hz, 1H), 7.80 (s, 1H), 7.07 (d, J = 8.8 Hz, 1H), 4.36–4.30 (m, 4H), 1.81–1.73 (m, 4H), 1.52–1.43 (m, 4H), 0.99 (t, J = 7.4 Hz, 3H), 0.98 (t, J = 7.4 Hz, 3H). ¹³C{1H} NMR (Chloroform-d, 126 MHz): 166.83, 166.57, 146.79, 145.42, 144.64, 141.43, 138.53, 137.45, 133.13, 127.91, 127.47, 124.42, 124.00, 122.74, 121.68, 120.19, 117.20, 113.21, 66.36, 66.12, 30.66, 30.64, 19.33, 13.88. Anal. calcd. C₂₆H₂₄N₂O₁₀ C, 59.54; H, 4.61; N, 5.34%; found C, 59.43; H, 4.49; N, 5.23%.

2.1.4. Synthesis of Dibutyl 2-Aminobenzo[b]naphtho[1,2-e][1,4]dioxine-4,5-dicarboxylate 3d

Yield 6.00 g (89%) as a yellow solid. ¹H NMR (Chloroform-d, 500 MHz): 7.36 (s, 1H), 7.33 (d, J = 2.4 Hz, 1H), 7.32 (d, J = 2.4 Hz, 1H), 6.97–6.92 (m, 3H), 6.89-6.87 (m, 1H), 4.26 (td, J = 6.8, 2.9 Hz, 4H), 4.06 (bs, 2H), 1.76–1.69 (m, 4H), 1.49–1.39 (m, 4H), 0.97 (t, J = 7.4 Hz, 3H), 0.96 (t, J = 7.4 Hz, 3H). ¹³C{1H} NMR (Chloroform-d, 126 MHz): 168.78, 168.14, 144.09, 142.24, 142.14, 137.20, 136.81, 131.95, 127.26, 125.61, 124.68, 124.36, 121.27, 119.72, 117.68, 116.70, 104.22, 65.40, 65.28, 30.78, 30.71, 19.39, 19.37, 13.93. Anal. calcd. C₂₆H₂₇NO₆ C, 69.47; H, 6.05; N, 3.12%; found C, 69.25; H, 5.92; N, 3.33%.

2.1.5. Synthesis of Dibutyl Benzo[b]naphtho[1,2-e][1,4]dioxine-4,5-dicarboxylate 3e

Yield 5.80 g (89%) as a yellow solid. ¹H NMR (Chloroform-d, 500 MHz): 8.27 (dd, J = 8.5, 1.3 Hz, 1H), 7.89 (dd, J = 7.1, 1.3 Hz, 1H), 7.63 (s, 1H), 7.54 (dd, J = 8.5, 7.1 Hz, 1H), 7.01–6.95 (m, 3H), 6.90–6.89 (m, 1H), 4.30 (t, J = 6.8 Hz, 2H), 4.29 (t, J = 6.8 Hz, 2H), 1.78–1.71 (m, 4H), 1.50–1.42 (m, 4H), 0.98 (t, J = 7.4 Hz, 3H); 0.97 (t, J = 7.4 Hz, 3H). ¹³C{1H} NMR (Chloroform-d, 126 MHz): 168.91, 167.85, 142.14, 141.94, 138.39, 136.99, 130.58, 129.31, 125.84, 125.68, 125.31, 124.95, 124.57, 124.27, 121.33, 116.86, 116.76, 65.44, 65.41, 30.76, 30.73, 19.38, 13.93. Anal. calcd. C₂₆H₂₆O₆ C, 71.87; H, 6.03; found C, 72.00; H, 5.88%.

General procedure for the preparation of anhydrides 4a–e: To a solution of diesters 3a–e (6 mmol) in 40 mL of acetic acid, 10 mL of concentrated sulfuric acid were added dropwise. The reaction mixture was stirred vigorously and heated at 110 °C for 1 h. After cooling, about 30 g of crushed ice was added to the reaction mixture and the resulting crystals were filtered, washed abundantly with water and dried. The resulting anhydrides were of high purity (HPLC, 96–99.8%) and were used without further purification.

2.1.6. Synthesis of 2-Nitro-4H,6H-benzo[de]benzo[5,6][1,4]dioxino[2,3-g]isochromene-4,6-dione 4a

Yield 2.07 g (99%) as a yellow solid. Anal. calcd. C₁₈H₇NO₇ C, 61.90; H, 2.02; N, 4.01%; found C, 61.99; H, 2.20; N, 4.15%.

2.1.7. Synthesis of 2-Nitro-4H,6H-benzo[de]naphtho[2′,3′:5,6][1,4]dioxino[2,3-g]isochromene-4,6-dione 4b

Yield 2.30 g (96%) as a yellow solid. Anal. calcd. C₂₂H₉NO₇ C, 66.17; H, 2.27; N, 3.51%; found C, 66.40; H, 2.29; N, 3.35%.

2.1.8. Synthesis of 2,11-Dinitro-4H,6H-benzo[de]benzo[5,6][1,4]dioxino[2,3-g]isochromene-4,6-dione 4c

Yield 2.36 g (100%) as an orange solid. Anal. calcd. C₁₈H₆N₂O₉ C, 54.84; H, 1.53; N, 7.11%; found C, 55.03; H, 1.67; N, 6.89%.

2.1.9. Synthesis of Dibutyl 2-Aminobenzo[b]naphtho[1,2-e][1,4]dioxine-4,5-dicarboxylate 4d

Yield 1.74 g (91%) as a yellow solid. Anal. calcd. C₁₈H₉NO₅ C, 67.72; H, 2.84; N, 4.39%; found C, 67.83; H, 2.97; N, 4.61%.

2.1.10. Synthesis of Dibutyl Benzo[b]naphtho[1,2-e][1,4]dioxine-4,5-dicarboxylate 4e

Yield 1.73 g (95%) as a yellow solid. ¹H NMR (Chloroform-d, 500 MHz): 8.52 (dd, J = 7.2, 1.0 Hz, 1H), 8.47 (dd, J = 8.5, 1.0 Hz, 1H), 8.14 (s, 1H), 7.78 (dd, J = 8.4, 7.3 Hz, 1H), 7.05–7.00 (m, 3H), 6.95–6.93 (m, 1H). ¹³C{1H} NMR (Chloroform-d, 126 MHz): 160.61, 159.56, 142.81, 141.54, 141.01, 139.65, 132.78, 128.36, 128.33, 127.74, 125.95, 125.15, 123.76, 122.52, 118.96, 117.09, 116.95, 113.79. Anal. calcd. C₁₈H₈O₅ C, 71.06; H, 2.65; found C, 71.19; H, 2.49%.

General procedure for the preparation of imidies 5a–e: To a suspension of the corresponding anhydride 4a–e (5 mmol) in 20 mL of 2-methyl-2-butanol was added N,N-dimethyl ethylenediamine (7.5 mmol, 0.66 g). The reaction mixture was refluxed for 30 min. After cooling, about 20 g of crushed ice was added to the reaction mixture, and the resulting crystals were filtered, washed abundantly with water, and dried. The resulting imides were of high purity (HPLC, 95–99.9%) and were used without further purification.

2.1.11. Synthesis of 5-(2-(Dimethylamino)ethyl)-2-nitro-4H-benzo[de]benzo[5,6][1,4]dioxino[2,3-g]isoquinoline-4,6(5H)-dione 5a

Yield 1.97 g (94%) as a yellow solid. ¹H NMR (Trifluoroacetic acid-d, 500 MHz): 9.50 (s, 1H), 9.38 (d, J = 1.5 Hz, 1H), 8.49 (s, 1H), 7.21–7.25 (m, 3H), 7.09 (dd, J = 6.3, 2.7 Hz, 1H), 4.93 (d, J = 4.4 Hz, 2H), 3.95 (s, 2H), 3.40 (d, J = 4.3 Hz, 6H). ¹³C{1H} NMR (Trifluoroacetic acid-d, 126 MHz): 167.69, 167.46, 148.69, 148.03, 144.18, 143.31, 142.79, 130.20, 129.08, 128.95, 128.01, 127.16, 127.07, 125.47, 124.60, 119.30, 119.13, 118.51, 60.30, 46.56, 38.84. Anal. calcd. C₂₂H₁₇N₃O₆ C, 63.01; H, 4.09; N, 10.02%; found C, 63.12; H, 4.21; N, 10.23%. HRMS (ESI) m/z 420.1180 (calcd for C₂₂H₁₈N₃O₆ [M+H]⁺ 420.1190).

2.1.12. Synthesis of 5-(2-(Dimethylamino)ethyl)-2-nitro-4H-benzo[de]naphtho[2′,3′:5,6][1,4]-dioxino[2,3-g]isoquinoline-4,6(5H)-dione 5b

Yield 1.97 g (94%) as a yellow solid.

HRMS (ESI) m/z 470.1334 (calcd for C₂₆H₂₀N₃O₆ [M+H]⁺ 470.1347).

2.1.13. Synthesis of 5-(2-(Dimethylamino)ethyl)-2,11-dinitro-4H-benzo[de]benzo[5,6][1,4]-dioxino[2,3-g]isoquinoline-4,6(5H)-dione 5c

Yield 2.14 g (92%) as yellow solid¹H NMR (Chloroform-d, 500 MHz): ¹H NMR (500 MHz): 9.51–9.51 (m, 1H), 9.36 (d, J = 2.1 Hz, 1H), 8.49 (s, 1H), 8.10–8.09 (m, 2H), 7.23 (d, J = 9.6 Hz, 1H), 4.81–4.78 (m, 2H), 3.81–3.79 (m, 2H), 3.28 (s, 6H). ¹³C{1H} NMR (Chloroform-d, 126 MHz): 167.32, 166.97, 149.66, 149.29, 146.66, 146.30, 143.32, 142.83, 130.66, 128.76, 127.85, 126.89, 125.58, 125.09, 124.55, 120.00, 119.74, 115.43, 60.57, 46.49, 38.83. Anal. calcd. C₂₂H₁₆N₄O₈ C, 56.90; H, 3.47; N, 12.06%; found C, 57.03; H, 3.29; N, 11.93%. HRMS (ESI) m/z 465.1027 (calcd for C₂₂H₁₇N₄O₈ [M+H]⁺ 465.1041).

2.1.14. Synthesis of 2-Amino-5-(2-(dimethylamino)ethyl)-4H-benzo[de]benzo[5,6][1,4]dioxino[2,3-g]isoquinoline-4,6(5H)-dione 5d

Yield 1.75 g (90%) as a yellow solid. ¹H NMR (Trifluoroacetic acid-d, 500 MHz): 8.74 (d, J = 2.1 Hz, 1H), 8.67 (d, J = 2.2 Hz, 1H), 8.27 (s, 1H), 7.04–6.99 (m, 3H), 6.94–6.92 (m, 1H), 4.70–4.68 (m, 2H), 3.70 (t, J = 5.2 Hz, 2H), 3.16 (s, 6H). ¹³C{1H} NMR (Trifluoroacetic acid-d, 126 MHz): 167.45, 167.40, 146.66, 143.91, 143.12, 142.61, 130.37, 128.43, 127.91, 127.80, 127.63, 127.60, 125.82, 125.57, 125.09, 118.92, 118.74, 60.43, 46.16, 38.46. Anal. calcd. C₂₂H₁₉N₃O₄ C, 67.86; H, 4.92; N, 10.79%; found C, 67.99; H, 4.81; N, 10.93%. HRMS (ESI) m/z 390.1436 (calcd for C₂₀H₂₂N₃O₄ [M+H]⁺ 390.1448).

2.1.15. Synthesis of 5-(2-(Dimethylamino)ethyl)-4H-benzo[de]benzo[5,6][1,4]dioxino[2,3-g]isoquinoline-4,6(5H)-dione 5e

Yield 1.68 g (90%) as a yellow solid. ¹H NMR (Trifluoroacetic acid-d, 500 MHz): 8.66 (d, J = 7.2 Hz, 1H), 8.64–8.61 (m, 1H), 8.20 (d, J = 9.5 Hz, 1H), 7.92–7.89 (m, 1H), 7.13–7.03 (m, 3H), 6.97 (d, J = 7.2 Hz, 1H), 4.83–4.81 (m, 2H), 3.87–3.84 (m, 2H), 3.34 (s, 6H). ¹³C{1H} NMR (Trifluoroacetic acid-d, 126 MHz): 169.75, 168.48, 147.08, 143.57, 143.00, 142.56, 142.37, 135.45, 132.39, 132.22, 129.97, 128.51, 128.34, 127.40, 125.93, 124.91, 119.04, 118.96, 60.34, 46.53, 38.73. Anal. calcd. C₂₂H₁₈N₂O₄ C, 70.58; H, 4.85; N, 7.48%; found C, 70.44; H, 4.89; N, 7.37%. HRMS (ESI) m/z 375.1328 (calcd for C₂₂H₁₉N₂O₄ [M+H]⁺ 375.1339).

2.1.16. Synthesis of Dibutyl 6-Amino-3,4-dibromonaphthalene-1,8-dicarboxylate 6

To a solution of dibutyl 3,4-dibromo-6-nitronaphthalene-1,8-dicarboxylate (20 mmol, 10.62 g) in 100 mL of acetic acid, iron powder (75 mmol, 4.19 g) was added in portions at 90 °C. The reaction mixture was stirred for 2 h at the same temperature (monitoring by TLC). After cooling, the reaction mixture was poured into ice. The resulting precipitation was filtered, washed with water, and dried. The crude product was purified by column chromatography on silica gel using cyclohexane/dichloromethane as the eluent; the chromatographic system started with pure cyclohexane, and the polarity gradually increased by addition of dichloromethane up to 80%. Yield 9.62 g (96%). ¹H NMR (Chloroform-d, 500 MHz): 7.86 (s, 1H), 7.60 (d, J = 2.4 Hz, 1H), 7.44 (d, J = 2.4 Hz, 1H), 4.27 (t, J = 6.8 Hz, 2H), 4.26 (t, J = 6.8 Hz, 2H), 4.20 (s, 2H), 1.72 (h, J = 7.2 Hz, 4H), 1.44 (dh, J = 14.6, 7.4 Hz, 4H), 0.96 (t, J = 7.4 Hz, 3H), 0.96 (t, J = 7.4 Hz, 3H). ¹³C{1H} NMR (Chloroform-d, 126 MHz): 168.19, 167.71, 145.67, 136.60, 132.49, 130.79, 129.78, 126.82, 123.92, 122.22, 121.12, 112.21, 65.71, 65.62, 30.67, 19.33, 13.90. Anal. calcd. C₂₀H₂₃Br₂NO₄; C, 47.93; H, 4.63; N, 2.79%; found. C, 47.77; H, 4.78; N, 3.00%.

2.1.17. Synthesis of Dibutyl 3-Nitronaphthalene-1,8-dicarboxylate 8

A mixture of 3-nitro-1,8naphthal anhydride (50.0 mmol, 12.16 g) and potassium hydroxide (120.0 mmol, 7.92 g) in 200 mL of water was stirred at 90 °C for 15 min. Aliquat 336 (1.0 mL) and 1-bromobutane (200 mmol, 21.58 mL) was added, and the reaction mixture was refluxed for 2 h. After cooling, the reaction mixture was extracted with dichloromethane. The organic layer was dried and evaporated to dryness. The crude product was purified by column chromatography on silica gel using cyclohexane/dichloromethane as the eluent; the chromatographic system started with pure cyclohexane, and the polarity gradually increased by addition of dichloromethane up to 50%. Yield 18.11 g (97%) as a brownish oil. ¹H NMR (Chloroform-d, 500 MHz): 8.91 (d, J = 2.0 Hz, 1H), 8.71 (d, J = 1.8 Hz, 1H), 8.18 (d, J = 7.7 Hz, 2H), 7.72 (t, J = 7.7 Hz, 1H), 4.36 (t, J = 7.2 Hz, 2H), 4.33 (t, J = 7.2 Hz, 2H), 1.82–1.74 (m, 4H), 1.53–1.43 (m, 4H), 0.99 (t, J = 7.3 Hz, 3H), 0.98 (t, J = 7.3 Hz, 3H). ¹³C{1H} NMR (Chloroform-d, 126 MHz): 168.10, 167.32, 144.61, 134.05, 133.63, 133.35, 132.68, 131.10, 130.19, 127.84, 127.56, 123.02, 66.16, 65.83, 30.70, 30.68, 19.34, 13.90. Anal. calcd. C₂₀H₂₃NO₆ C, 64.33; H, 6.21; N, 3.75%; found C, 64.56; H, 6.41; N, 3.70%.

2.1.18. Synthesis of Dibutyl 3-Aminonaphthalene-1,8-dicarboxylate 9

To a suspension of iron powder (135 mmol, 7.54 g) in a mixture of 100 mL acetic acid, 60 mL of ethanol and 40 mL of water a solution of dibutyl 3-nitronaphthalene-1,8-dicarboxylate (45 mmol, 16.80 g) in 50 mL of acetic acid was added in portions at 90 °C. The reaction mixture was stirred for 30 min at 90 °C. After cooling, the mixture was poured into ice, and the mixture was extracted with dichloromethane. The crude product was purified by column chromatography on silica gel using cyclohexane/dichloromethane as the eluent; the chromatographic system was started with pure cyclohexane, and the polarity gradually increased by addition of dichloromethane up to 80%. Yield 13.45 g (87%) as brownish oil. ¹H NMR (Chloroform-d, 500 MHz): 7.73 (dd, J = 7.1, 1.3 Hz, 1H), 7.70 (dd, J = 8.4, 1.1 Hz, 1H), 7.45 (d, J = 2.5 Hz, 1H), 7.39 (dd, J = 8.3, 7.1 Hz, 1H), 7.09 (d, J = 2.5 Hz, 1H), 4.28 (t, J = 6.9 Hz, 4H), 1.76–1.70 (m, 4H), 1.48–1.40 (m, 4H), 0.96 (t, J = 7.4 Hz, 3H), 0.95 (t, J = 7.4 Hz, 3H). ¹³C{1H} NMR (Chloroform-d, 126 MHz): 169.39, 168.96, 143.16, 136.29, 131.56, 130.50, 130.17, 126.86, 125.67, 122.23, 121.94, 112.57, 65.34, 65.18, 30.77, 30.73, 19.38, 19.36, 13.92. Anal. calcd. C₂₀H₂₅NO₄ C, 69.95; H, 7.34; N, 4.08%; found C, 69.88; H, 7.49; N, 3.99%.

2.1.19. Synthesis of Dibutyl 3-Amino-4-bromonaphthalene-1,8-dicarboxylate 10

To a solution of dibutyl 3-aminonaphthalene-1,8-dicarboxylate (35 mmol, 12.02 g) in 100 mL acetic acid, a solution of bromine (40 mmol, 6.40 g) in 20 mL of acetic acid was added dropwise in a period of 5 min at room temperature. The mixture was stirred for an additional 15 min and then poured into ice (300 g). The mixture was extracted with dichloromethane, washed with water, and evaporated to dryness. The crude product was purified by column chromatography on silica gel using cyclohexane/dichloromethane as the eluent; the chromatographic system started with pure cyclohexane, and the polarity gradually increased by addition of dichloromethane up to 50%. Yield 14.63 g (99%) as a brownish oil. ¹H NMR (Chloroform-d, 500 MHz): 8.29 (dd, J = 8.7, 1.3 Hz, 1H), 7.81 (dd, J = 7.1, 1.2 Hz, 1H), 7.59 (s, 1H), 7.54 (dd, J = 8.6, 7.1 Hz, 1H), 4.28 (t, J = 6.9 Hz, 4H), 1.80–1.68 (m, 4H), 1.51–1.37 (m, 4H), 0.96 (t, J = 7.4 Hz, 3H), 0.95 (t, J = 7.4 Hz, 3H). ¹³C{1H} NMR (Chloroform-d, 126 MHz): 168.92, 168.28, 140.54, 134.16, 130.92, 130.69, 129.63, 127.38, 127.11, 124.53, 123.14, 121.69, 65.57, 65.39, 30.73, 30.69, 19.37, 19.34, 13.91. Anal. calcd. C₂₀H₂₄BrNO₄; C, 56.88; H, 6.73; N, 3.32%; found. C, 56.70; H, 6.49; N, 3.19%.

2.1.20. Synthesis Dibutyl 3,4-Dibromonaphthalene-1,8-dicarboxylate 11

To a cold (0–5 °C) solution of dibutyl 3-amino-4-bromonaphthalene-1,8-dicarboxylate (30.0 mmol, 12.67 g) in 100 mL acetic acid, 50 mL of acetonitrile and 30 mL of 70% sulfuric acid, a solution of sodium nitrite (40.0 mmol, 2.76 g) in 20 mL of water was added dropwise in a period of 1 h. The mixture was stirred an additional 15 min and then poured into a solution of copper (I) bromide (30.0 mmol, 4.30 g) in 50 mL of 48% hydrobromic acid. The mixture was diluted with water up to 600 mL and stirred for 2 h. The mixture was extracted with dichloromethane, dried and evaporated to dryness. The crude product was purified by column chromatography on silica gel using cyclohexane/dichloromethane as the eluent; the chromatographic system started with pure cyclohexane, and the polarity gradually increased by addition of dichloromethane up to 30%. Yield 11.96 g (82%) as a colorless oil. ¹H NMR (Chloroform-d, 500 MHz): 8.55 (dd, J = 8.6, 1.1 Hz, 1H), 8.14 (s, 1H), 8.04 (dt, J = 6.8, 3.4 Hz, 1H), 7.64 (dd, J = 8.6, 7.2 Hz, 1H), 4.31 (t, J = 6.8 Hz, 3H), 4.30 (t, J = 6.8 Hz, 3H), 1.79–1.71 (m, 4H), 1.49–1.41 (m, 4H), 0.97 (t, J = 7.4 Hz, 3H); 0.96 (t, J = 7.4 Hz, 3H). ¹³C{1H} NMR (Chloroform-d, 126 MHz): 168.29, 167.33, 134.63, 133.64, 132.45, 131.09, 131.05, 130.63, 129.38, 127.76, 127.39, 123.71, 65.89, 65.66, 30.70, 30.67, 19.35, 19.32, 13.90. Anal. calcd. C₂₀H₂₂Br₂O₄ C, 49.41; H, 4.56%; found C, 49.35; H, 4.67%.

2.2. Cell Cultures

The human non-small cell lung cancer cell lines A549 (ATCC^®CCL-185™) and NCI-H1299 (ATCC^®CRL-5803™), along with the normal lung fibroblast line MRC-5 (ATCC^®CCL-171™), were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). A549 cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM), H1299 cells in RPMI-1640, and MRC-5 cells in Minimum Essential Medium (all acquired from Thermo Fisher Scientific, Waltham, MA, USA). The growth media for all cell lines were supplemented with 10% fetal bovine serum (Thermo Fisher Scientific, Waltham, MA, USA) and a combination of 100 units of penicillin, 100 µg streptomycin, and 0.25 µg amphotericin B per mL (Merck KGaA, Darmstadt, Germany) to prepare complete culture media. All cells were maintained at 37 °C in a humidified atmosphere with 5% CO₂.

2.3. Compounds in Solutions

Stock solutions of doxorubicin (hydrochloride) (Cayman Chemical Company, Ann Arbor, MI, USA), amonafide (Merck KGaA, Darmstadt, Germany), mitonafide (MedChemExpress LLC, Monmouth Junction, NJ, USA), and all in-house naphthalimides were prepared at 2 mM in DMSO and diluted in cell culture media immediately before use. The final DMSO concentration did not exceed 0.5% (v/v) in any experiment. Corresponding vehicle (DMSO-only) controls were included for all assays and cell lines.

2.4. Cytotoxicity Test

The cytotoxic activity of the compounds was tested using the MTT-dye reduction assay. [30]. Cells were seeded at optimal densities for each cell line (H1299: 1.5 × 10³; A549: 2.5 × 10³; MRC-5: 4.5 × 10³ cells per well) in 96-well plates, allowed to attach for 24 h at 37 °C with 5% CO₂, and then treated for 72 h. The maximum DMSO concentration in the growth medium at the highest drug dose was 0.5%. The control group for each experiment was treated with medium containing 0.5% DMSO. After treatment, the media was replaced with phenol-red-free media containing MTT (Thermo Fisher Scientific, Waltham, MA, USA) at a final concentration of 0.5 mg/mL. The plates were then incubated for 4 h at 37 °C, and the resulting formazan crystals were dissolved by adding DMSO to each well. Absorbance was measured using an ELISA plate reader Varioscan (Thermo Fisher Scientific, Waltham, MA, USA) at a test wavelength of 570 nm. IC₅₀ values were calculated using the “log of concentration vs. normalized response (variable slope)” model in GraphPad Prism software version 8.0.1 for Windows (GraphPad Software, Boston, MA, USA). The selectivity index (SI) was calculated as the IC₅₀ value for normal cells divided by that for cancer cells (SI = IC₅₀ normal/IC₅₀ cancer), with higher SI values indicating greater anticancer selectivity. Light microscopy images were acquired using a Zeiss Axiovert 200M microscope (Carl Zeiss, Oberkochen, Germany) equipped with a 10× PlanApochromat objective (NA = 0.45) and a CCD camera (AxioCam MRm, Carl Zeiss).

2.5. Clonogenic Assay

Cells were seeded at optimal densities for each cell line (H1299: 600; A549: 800 cells per well) in 6-well plates and allowed to adhere overnight prior to treatment. Subsequently, the medium was replaced with fresh medium containing concentrations corresponding to the compounds’ calculated IC₁₀, IC₁₅, and IC₂₀ for a treatment period of 24 h. Afterward, the plates were returned to the incubator (37 °C, 5% CO₂) for an additional 10–14 days. Cells were fixed with 3.7% paraformaldehyde and stained with 0.2% crystal violet. Colonies were counted using Fiji (ImageJ Software 2.15.1 [31], and plating efficiency (PE) and survival fraction (SF) were calculated as described [32]. Graphs were made using GraphPad Prism software version 8.0.1 for Windows (GraphPad Software, Boston, MA, USA).

2.6. Flow Cytometry

2.6.1. Apoptosis Detection

An Annexin V Apoptosis Detection Kit (FITC/PI; Thermo Fisher Scientific, Waltham, MA, USA) was used to assess apoptosis by flow cytometry, following the manufacturer’s guidelines. Cells were seeded at optimal densities for each cell line (H1299: 6 × 10⁴; A549: 10 × 10⁴ cells per well) in 12-well plates, allowed to adhere for 24 h, and exposed to the calculated IC₅₀ concentrations of naphthalimides for 24 and 48 h. For the analysis, the cells were resuspended in Annexin V Binding Buffer (AVBB), followed by the addition of Annexin V-FITC. After a 20 min incubation, the samples were washed with AVBB. Next, they were resuspended in AVBB with PI and incubated on ice in the dark for one hour. The labeled cells were subsequently analyzed using a Becton Dickinson FACScalibur flow cytometer (BD Biosciences, San Jose, CA, USA). The percentages of viable, early apoptotic, late apoptotic, and necrotic cells were quantified using FlowJo v.10.8.1 software (BD Biosciences, Ashland, OR, USA).

2.6.2. Cell Cycle Analysis

For cell cycle analysis, cells were seeded at a density of 7 × 10⁴ cells per well in 6-well plates, allowed to attach for 24 h, synchronized by incubation with low- or serum-free medium for an additional 24 h, and subsequently treated with the calculated IC₅₀ and IC₇₅ concentrations of naphthalimides for 24 and 48 h. Cells from each sample were collected and then fixed in 70% ethanol. The cells were washed with PBS and resuspended in 0.5 mL of 0.1% Triton X-100 in PBS containing 200 µg/mL RNase (Merck KGaA, Darmstadt, Germany) and 25 μg/mL propidium iodide (Merck KGaA, Darmstadt, Germany). They were incubated in the dark at room temperature for 1 h. Flow cytometric analysis was performed using a Becton Dickinson FACScalibur instrument (BD Biosciences, San Jose, CA, USA). Prior to DNA content analysis, cell populations were gated to exclude debris and cell aggregates, and singlet discrimination was performed using FSC-A versus FSC-H parameters. Where applicable, a sub-G₁ region was monitored as a quality-control indicator of DNA fragmentation, but cell-cycle phase distributions (G₀/G₁, S, and G₂/M) were quantified from the gated singlet population. Data acquisition and analysis were performed using FlowJo v10.8.1 software (BD Biosciences, Ashland, OR, USA). The percentages of cells in each cell-cycle phase were calculated from PI fluorescence histograms using the Dean-Jett-Fox model.

2.7. Immunofluorescence Microscopy

Cells were seeded at a density of 4 × 10⁴ on 12 mm coverslips, allowed to adhere for 48 h, and then treated with different concentrations for 24 and 48 h. Samples were then fixed in 3.7% paraformaldehyde in PBS for 5 min at room temperature, followed by fixation with methanol for 7 min at −20 °C. The fixed cells were washed with 100 mM Tris-HCl (pH 7.2) for 5 min prior to permeabilization with 0.1% Triton X-100 in PBS for 5 min. Coverslips were then blocked for one hour in blocking buffer containing 3% BSA and 0.1% Tween 20 in PBS.

2.7.1. Nuclear p53 Accumulation

Coverslips were subsequently incubated overnight at 4 °C with the primary antibody—monoclonal mouse anti-p53 (Abcam, Cambridge, UK) at a dilution of 1:100. After washing, cells were incubated with the secondary antibody—donkey anti-mouse Alexa 555 antibody (Thermo Fisher Scientific, Waltham, MA, USA) at a 1:2000 dilution for 30 min at room temperature.

2.7.2. EdU Incorporation and γH2AX Labeling

For γH2AX labeling coverslips were incubated overnight at 4 °C with the primary antibody—monoclonal rabbit anti-phospho-histone H2AX (Cell Signaling Technology, Danvers, MA, USA) at a dilution of 1:200. After washing, cells were incubated with the secondary antibody—goat anti-rabbit Alexa 555 antibody (Thermo Scientific, Waltham, MA, USA) at 1:2000 dilution, for 30 min at room temperature.

For EdU incorporation, coverslips were incubated with 25 µM EdU for 30 min prior to fixation. After blocking, a “click” reaction was carried out in a reaction mixture containing 100 mM Tris-HCl, pH 7.6, 4 mM CuSO₄, 10 µM Alexa Fluor 488 azide, and 100 mM sodium ascorbate for 30 min at room temperature.

2.7.3. Caspase-3/7 Detection

Caspase activation was detected using CellEvent Caspase-3/7 Green Detection Reagent (Thermo Fisher Scientific, Waltham, MA, USA). Cells were seeded at a density of 4 × 10⁴ on 12 mm coverslips, allowed to adhere for 48 h, and then treated with concentrations corresponding to IC₅₀ and IC₇₅ for 24 h. Coverslips were then incubated with 5 µM of the caspase-3/7 detector for 30 min. Samples were then fixed in 3.7% paraformaldehyde in PBS for 15 min at room temperature.

All coverslips were mounted using ProLong Diamond Antifade mounting media (Thermo Fisher Scientific, Waltham, MA, USA) containing 10 µg/mL DAPI (Sigma-Aldrich, St. Louis, MO, USA). Fluorescent images were acquired using a Zeiss Axio Imager.A2 microscope equipped with an Axiocam 705 mono CCD camera, with the EC Plan-Neofluar 20×/0.5 M27 and Plan-Apochromat 63×/1.4 Oil DIC M27 objectives (Carl Zeiss, Oberkochen, Germany). All images from control and treated cells used for the analyses were acquired with the same exposure time. Image analysis was performed with CellProfiler (version 4.2.6) as well as Fiji (ImageJ Software 2.15.1 [31], and the resulting data were used to generate graphs in RStudio (version 2025.9.2.418) (Posit team (2025). RStudio: Integrated Development Environment for R). Quantitative immunofluorescence data (γH2AX, EdU, Caspase-3/7, and p53) were obtained from at least three independent experiments, with a minimum of 10 randomly selected microscopic fields analyzed per condition in each experiment. Statistical analysis was performed using one-way ANOVA, followed by post hoc multiple comparison tests.

2.8. Statistical Analysis

All statistical analyses were performed using RStudio (version 2025.9.2.418) (Posit team, 2025). RStudio: Integrated Development Environment for R or GraphPad Prism software (version 8.0.1 for Windows). All experiments were performed in at least three independent biological replicates, unless otherwise stated. Quantitative data are presented as mean ± standard deviation (SD).

Data distribution was assessed for normality where applicable, and parametric statistical tests were applied accordingly. For all analyses involving more than two groups, appropriate multiple-comparisons corrections were applied as specified below. A p-value < 0.05 was considered statistically significant.

For cytotoxicity assessment (MTT assay), IC₅₀ values were calculated by nonlinear regression using a four-parameter logistic model. Comparisons between compounds and cell lines were performed using one-way analysis of variance (ANOVA), as appropriate, followed by Tukey’s multiple comparison post hoc tests.

Clonogenic survival data were analyzed by one-way ANOVA followed by Dunnett’s post hoc test, comparing each treatment condition to the corresponding untreated control within the same cell line.

Cell cycle distribution data obtained by flow cytometry were analyzed using one-way ANOVA, followed by Tukey’s multiple comparison post hoc tests. For cell cycle analysis, at least 100,000 events per sample were collected.

Quantitative immunofluorescence data from γH2AX, EdU, caspase-3/7, and p53 assays were analyzed using one-way ANOVA, followed by Tukey’s multiple comparison post hoc tests. At least three independent experiments were analyzed, with a minimum of 10 randomly selected microscopic fields per condition per experiment.

For apoptosis analysis by flow cytometry (Annexin V/PI staining), 20,000 events per sample were acquired, and the percentages of viable, early apoptotic, late apoptotic, and necrotic cells were quantified and statistically analyzed using one-way ANOVA followed by Tukey’s post hoc test.

3. Results

3.1. Chemistry

Our primary synthetic objective was to obtain heterocyclic-extended mitonafide and amonafide derivatives featuring fused dioxin rings at positions 3 and 4. A crucial consideration in our retrosynthetic analysis was the initial preparation of the corresponding anhydrides. This approach offers several advantages: (i) introducing the dimethylaminoethyl fragment in the final synthetic step simplifies the purification of intermediate compounds; and (ii) it readily allows for modifications with other pharmacophore fragments, thus widening the scope for future applications. The annulation of additional rings onto the naphthalimide core can be readily achieved through aryl nucleophilic substitution (SnAr), facilitated by electron-withdrawing groups. Consequently, we selected 3,4-dibromo-6-nitronaphthalic anhydride as a suitable starting material (Scheme 1), given its recently developed facile gram-scale synthesis by our group [33]. To enhance the solubility and simplify the purification of intermediates, we opted to utilize their diester analogs, a strategy previously employed by us in the synthesis of dibenzofuran-annulated naphthalimides [34].

We converted anhydride 1 to the corresponding dibutyl ester via an aqueous alkylation reaction (Scheme 2), isolating the desired product, 2, in a high yield of 92% on a gram scale, as described previously [33]. Dibutyl ester 2 appeared as a pale brownish oil, exhibiting excellent solubility in most organic solvents. The nucleophilic substitution reaction of diester 2 with catechol or 2,3-dihydroxynaphthalene was conducted in NMP at 150 °C using potassium carbonate as a base in an inert medium. Under these conditions, the reaction proceeded quantitatively (monitored by TLC) within just a few hours. The resulting benzo- and naphthodioxine derivatives, 3a,b, were isolated and purified on a gram scale in 93% and 90% yield, respectively.

Esters 3a,b were readily converted to anhydrides upon heating in glacial acetic acid in the presence of concentrated sulfuric acid. The reactions proceeded via acid-catalyzed dealkylation, with the resulting anhydride precipitating directly from the reaction mixture. This method yielded anhydrides 4a,b with very high purity and in near-quantitative yields. The thus obtained pure anhydrides were then subjected to imidization with unsymmetrical N,N-ethylene-1,2-diamine. This reaction was carried out in tertiary amyl alcohol under reflux for 30 min. The target imides, 5a,b, were isolated in quantitative yields.

To examine the influence of a second nitro group, we applied the same procedures to obtain a nitro-substituted analog of 5a. 4-nitrocatechol reacted successfully with 2, affording the corresponding ester 3c in high yield as a mixture of regioisomers (Scheme 3). The consequent dealkylation and imidization proceeded smoothly in 100 and 92% yields, respectively, to afford the target nitrobenzodioxin derivative 5c.

For the synthesis of the amino analog of 5a, we initially attempted direct reduction. However, this approach proved unsuccessful due to the lower solubility and strong retention of imide 5a on silica, which significantly hindered purification. Consequently, we reverted to an earlier synthetic stage and reduced diester 2. This was achieved using iron powder in acetic acid under heating (Scheme 4). The reaction proceeded rapidly, affording the desired product in a very high yield of 96%.

The subsequent steps, including the aryl nucleophilic substitution with catechol, dealkylation of ester 3d, and the subsequent imidization of anhydride 4d to afford the target benzodioxin-annulated amonafide 5d, were all carried out in an analogous manner to the procedures described above.

To better understand the influence of the heterocyclic-extended system alone, we aimed to synthesize, for comparison, a 3,4-benzodioxine-annulated naphthalimide lacking a substituent at position 6. This, however, necessitated starting from a 3,4-dibromo derivative, the synthesis of which presents a challenge in terms of achieving the precise substitution pattern. Fortunately, we recently published a method for obtaining 3,4-dihalonaphthalimides [35], which we modified for this specific case by again utilizing diester intermediates (Scheme 5). As a starting material, we employed 3-nitronaphthalic anhydride 7, which is both readily available and inexpensive, and can also be easily prepared by the nitration of naphthalic anhydride.

Alkylation of the potassium carboxylate formed in situ from 3-nitro-1,8-naphthalene anhydride was carried out using 1-bromobutane in the presence of Aliquat 336 as a phase-transfer catalyst in an aqueous medium under reflux for 2 h. This reaction proceeded with an almost quantitative yield (97%), affording product 8 as a crystallizing oil. The nitro group was subsequently reduced to an amino group with iron powder in acetic acid. The resulting amino ester 9 was then brominated with an excess (1.5 equiv.) of elemental bromine in glacial acetic acid. The bromination was completely selective, yielding dibutyl 3-amino-4-bromo-1,8-naphthalenecarboxylate 10 in quantitative yield. Finally, the conversion of amino ester 10 into the corresponding dibromo derivative was carried out under Sandmeyer reaction conditions. 3,4-dibromoester 11 was isolated by column chromatography in good yield (82%) as a colorless oil.

From this point, we proceeded with synthetic steps analogous to those described above: aryl nucleophilic substitution with catechol to ester 3e, followed by dealkylation to yield anhydride 4e, and finally, imidization to afford the target N-(2-(dimethylamino)ethyl)-3,4-benzodioxine annulated naphthalimide 5e (Scheme 6).

3.2. Biology

3.2.1. Cytotoxicity and Antiproliferative Effect

MTT Assay

The cytotoxic potential of the investigated compounds was evaluated in human lung adenocarcinoma A549 and H1299 cell lines, as well as in non-malignant MRC-5 fibroblasts, using the MTT assay. Half-maximal inhibitory concentration (IC₅₀) values, calculated from at least three independent experiments, are summarized in

Table 1. Calculated IC₅₀ [nM] values for doxorubicin, naphthalimides, and dioxin–naphthalimides. Data show mean ± SD from at least n = 3 independent experiments (see Figure S32 for detailed dose–response curves of 5a).

Compounds	A549		H1299		MRC-5
	IC50 ± SD	SI	IC50 ± SD	SI	IC50 ± SD
doxorubicin	242 ± 3.8	1.8	120 ± 9.5	3.6	441 ± 3.4
mitonafide	644 ± 4.1	4.8	490 ± 2.3	6.2	3064 ± 4.8
amonafide	3228 ± 6.3	1.8	1853 ± 8.1	3.2	5908 ± 8.3
5a	80 ± 7.5	5.6	68 ± 6.6	6.6	449 ± 3.6
5b	50,911 ± 2.3	1.3	32,004 ± 4.8	2	65,748 ± 6.2
5c	163 ± 5.6	6.9	153 ± 2.3	7.4	1128 ± 4.2
5d	303 ± 4.8	3	132 ± 4.3	7	931 ± 9.3
5e	882 ± 4.3	1.5	1080 ± 5.6	1.2	1332 ± 7.6

. Tumor selectivity was assessed using the selectivity index (SI), defined as the ratio of IC₅₀ values in normal MRC-5 fibroblasts to those in cancer cells (Table 1). Compounds with SI values greater than 2 are generally considered to exhibit preferential cytotoxicity toward tumor cells. Among the reference compounds, doxorubicin displayed high cytotoxic potency, with IC₅₀ values of 242 ± 3.8 nM and 120 ± 9.5 nM against A549 and H1299 cells, respectively. In contrast, the clinically evaluated naphthalimides mitonafide and amonafide showed lower activity, with IC₅₀ values in the low micromolar range.

Introduction of the benzodioxin moiety into the naphthalimide scaffold markedly modulated biological activity. The unsubstituted benzodioxin-NI analog (5e) exhibited moderate cytotoxicity toward both tumor cell lines, with IC₅₀ values of 882 ± 4.3 nM in A549 cells and 1080 ± 5.6 nM in H1299 cells. Notably, compound 5a (6-nitro-benzodioxin-NI) emerged as the most potent derivative in the series, displaying IC₅₀ values of 80 ± 7.5 nM in A549 cells and 68 ± 6.6 nM in H1299 cells, thereby exceeding the antiproliferative activity of doxorubicin in both models.

A pronounced decrease in cytotoxic activity was observed upon introduction of bulky substituents, as exemplified by compound 5b, which showed IC₅₀ values above 30 µM in both tumor cell lines. This finding suggests that steric effects negatively influence productive interactions with cellular targets. In contrast, the dinitro- and amino-substituted analogs 5c and 5d retained submicromolar activity, with IC₅₀ values ranging from 132 to 303 nM in cancer cells.

Among the tested compounds, 5a and 5c demonstrated the highest tumor selectivity, with SI values of 5.6 and 6.9 in A549 cells, and 6.6 and 7.4 in H1299 cells, respectively. This pronounced selectivity was particularly evident in the p53-deficient H1299 cell line. The benzodioxin-annulated amonafide analog 5d also exhibited favorable selectivity (SI = 3–7), whereas compounds 5b and 5e showed low selectivity (SI ≤ 2), indicative of non-discriminatory cytotoxic effects. The substitution at the C-6 position of the benzodioxin–naphthalimide core exerted a strong influence on both cytotoxic potency and tumor selectivity. Nitro substitution enhanced both parameters, whereas amino or bulky substituents reduced antiproliferative activity while partially preserving selectivity. These results identify nitro-substituted benzodioxin–naphthalimides, exemplified by compound 5a, as promising lead structures for the further development of selective anticancer agents targeting lung carcinoma cells.

Morphological Changes in H1299 Cells upon Treatment with Dioxin-Annulated NI Derivatives

To further characterize the cellular response to treatment, morphological alterations of H1299 lung carcinoma cells were examined by phase-contrast light microscopy following exposure to 5a and 5d for 24, 48, and 72 h (Figure 1).

Treatment with compound 5a induced pronounced, time-dependent morphological changes characteristic of cytotoxic stress. After 24 h, cells exhibited early alterations, including mild cell shrinkage and partial loss of adherence. At 48 h, a marked reduction in cell density was observed, accompanied by extensive cell rounding and detachment. By 72 h, the monolayer was largely disrupted, with the remaining cells displaying condensed and fragmented morphology, consistent with advanced loss of cellular viability.

Comparative evaluation of the effects of the nitro-substituted derivative 5a and the amino-substituted analog 5d revealed clear differences in cytotoxic potency. While treatment with 5a resulted in extensive morphological deterioration and near-complete loss of adherent cells by 72 h, exposure to 5d produced less pronounced effects under identical conditions. Specifically, 5d treatment led to partial cell rounding and reduced confluency at 48–72 h, while a substantial fraction of adherent cells remained viable, indicating lower cytotoxic activity. These qualitative morphological observations are in good agreement with the MTT-based cytotoxicity data, corroborating the higher antiproliferative and cytotoxic efficacy of the nitro-substituted benzodioxin–naphthalimide derivative. The progressive nature of the observed morphological changes further supports a time- and structure-dependent cellular response, likely involving regulated cell death pathways triggered by compound 5a.

Doxorubicin- and Naphthalimide-Induced Inhibition of Clonogenic Survival

The long-term proliferative capacity of lung carcinoma cells following treatment with doxorubicin, mitonafide, and compound 5a was evaluated using a colony formation assay in A549 and H1299 cell lines (Figure 2). Representative images (Figure 2a) demonstrate a clear concentration-dependent reduction in both colony number and colony size upon treatment with all tested compounds compared to untreated controls. Under control conditions, both cell lines formed numerous dense colonies, whereas exposure to increasing concentrations corresponding to IC₁₀, IC₁₅, and IC₂₀ values resulted in progressive suppression of clonogenic growth. The most pronounced effect was observed for compound 5a, which nearly abolished colony formation in both A549 and H1299 cells at IC₂₀. In contrast, treatment with doxorubicin and mitonafide also reduced clonogenicity, but residual surviving colonies remained visible, particularly at lower concentrations.

Quantitative analysis of clonogenic survival (Figure 2b, the full uncut graphs are present in Figure S33) corroborated these qualitative observations. In both cell lines, the surviving fraction decreased significantly in a concentration-dependent manner following treatment with all compounds. Notably, compound 5a induced the steepest decline in clonogenic survival, reducing the surviving fraction to below 20% at IC₂₀, whereas mitonafide and doxorubicin produced more moderate inhibition at comparable doses. Comparison between the two lung cancer models revealed that A549 cells (p53-wild-type) exhibited slightly higher sensitivity to all tested agents than H1299 cells (p53-null). These results demonstrate that compound 5a exerts a strong and concentration-dependent suppression of clonogenic survival, surpassing the activity of the reference compounds and confirming its potent long-term antiproliferative effects in both p53-dependent and p53-independent cellular contexts.

3.2.2. Cell Cycle Perturbation

Effect of 5a on Cell Cycle Progression

The impact of 5a on the cell cycle distribution of A549 and H1299 lung carcinoma cells was analyzed by flow cytometry following propidium iodide (PI) staining (Figure 3 and Figure 4). Cells were treated for 24 and 48 h with concentrations corresponding to IC₅₀ and IC₇₅ values and compared with untreated controls.

In A549 cells (Figure 3), treatment with 5a resulted in a pronounced accumulation of cells in the G₁ phase at both 24 h and 48 h, accompanied by a significant decrease in the S-phase population (p < 0.05), consistent with inhibition of cell cycle progression. At the higher concentration (IC₇₅), this effect was particularly evident, with more than 80% of cells remaining in G₁ after 48 h of exposure, indicating a dominant G₁-phase arrest. A modest increase in the G₂/M fraction was observed at the lower concentration (IC₅₀), suggesting partial engagement of G₂/M checkpoint control under submaximal stress conditions. Time-dependent analysis further revealed a significant increase in the G₁ population in A549 cells treated with IC₇₅ between 24 and 48 h, confirming a sustained arrest response (combined graph is present in Figure S34a).

In contrast, H1299 cells displayed a distinct cell-cycle response pattern (Figure 4). Treatment with 5a induced a marked accumulation of cells in the G₂/M phase, particularly after 48 h at IC₇₅, where up to 75% of cells were arrested in G₂/M. This redistribution was accompanied by a concomitant decrease in both G₁ and S-phase populations, indicative of a robust G₂/M checkpoint activation.

Consistent with these observations, treatment of H1299 cells with both IC₅₀ and IC₇₅ concentrations resulted in a time-dependent increase in the G₂/M fraction (combined graph is present in Figure S34b), underscoring a sustained G₂/M arrest in the p53-deficient cellular context.

Together, these results demonstrate that compound 5a induces cell line-specific cell cycle arrest, characterized by G₁-phase arrest in p53-proficient A549 cells and G₂/M-phase arrest in p53-deficient H1299 cells. This differential response indicates that 5a interferes with DNA replication and checkpoint regulation in a manner that is influenced by p53 status. The accumulation of cells in either the G₁ or G₂/M phase is indicative of activation of DNA damage response pathways, leading to checkpoint engagement and cell cycle blockade. Such a cell cycle arrest pattern is consistent with replication-stress-associated mechanisms previously described for DNA-interacting naphthalimide derivatives. While the present data do not directly identify the primary molecular target, the enhanced activity of the nitro-substituted benzodioxin scaffold suggests an increased capacity to induce replication-associated DNA damage or replication fork perturbation, thereby promoting cytostatic effects and facilitating the activation of downstream cell death pathways.

Replication Stress and DNA Damage Induced by Compound 5a

To further elucidate the mechanism underlying the antiproliferative activity of 5a, we assessed DNA damage formation and replication activity using a dual γH2AX/EdU immunofluorescence assay in A549 and H1299 cells (Figure 5).

In A549 cells, untreated controls exhibited strong nuclear EdU fluorescence with low basal γH2AX signal, consistent with active DNA synthesis and minimal endogenous DNA damage (Figure 5a,b). Treatment with 5a at IC₅₀ and IC₇₅ concentrations resulted in a marked, time- and dose-dependent increase in nuclear γH2AX signal, accompanied by a pronounced reduction in EdU incorporation. After 48 h of exposure at IC₇₅, the majority of nuclei displayed intense γH2AX foci together with near-complete suppression of EdU labeling, indicative of profound replication inhibition and accumulation of DNA damage. Quantitative analysis confirmed a significant increase in mean γH2AX fluorescence intensity alongside a concomitant decrease in EdU signal (p < 0.05). Importantly, EdU staining and image acquisition were performed using identical exposure times and imaging settings across all experimental conditions, and EdU suppression reflects a marked reduction in DNA synthesis rather than an absolute absence of EdU-positive cells.

In H1299 cells, treatment with 5a similarly induced a substantial increase in γH2AX levels, detectable as early as 24 h and persisting after 48 h of exposure (Figure 5c,d). EdU incorporation was strongly suppressed at both time points, indicating effective inhibition of DNA synthesis and accumulation of DNA damage in the p53-deficient cellular context. Notably, the relative increase in γH2AX signal was moderately higher in H1299 cells compared to A549 cells, suggesting that p53 deficiency may contribute to enhanced persistence of DNA damage and/or reduced repair capacity (Figure S35).

These data demonstrate that compound 5a induces a replication stress-associated DNA damage phenotype, characterized by elevated γH2AX signaling and markedly reduced DNA synthesis in both lung carcinoma models. The concurrent increase in γH2AX and loss of EdU incorporation is consistent with replication interference followed by DNA break formation, potentially arising from impaired replication fork progression and/or DNA–enzyme conflicts. While direct evidence for topoisomerase inhibition was not obtained in the present study, the observed pattern closely aligns with mechanisms previously reported for structurally related naphthalimide derivatives that disrupt DNA processing during S phase.

3.2.3. Study of Apoptosis and Cell Death

Annexin V-FITC/Propidium Iodide (PI) Dual-Staining Flow-Cytometric Analysis

To determine whether the cytotoxic activity of compound 5a is associated with apoptosis, Annexin V-FITC/propidium iodide (PI) dual-staining flow-cytometric analysis was performed in A549 and H1299 cells after 24 and 48 h treatment with IC₅₀ concentrations (Figure 6).

In A549 cells, untreated controls contained predominantly viable cells (>90%) with minimal apoptotic or necrotic fractions (Figure 6a,b). Exposure to 5a resulted in a modest increase in early apoptotic cells after 24 h, followed by a more pronounced rise in apoptotic and PI-positive populations after 48 h, indicating a time-dependent activation of cell death pathways.

In H1299 cells, the compound produced a similar but slightly stronger effect (Figure 6c,d). A clear increase in the early apoptotic fraction was observed after 24 h, progressing to extensive apoptosis by 48 h, accompanied by a substantial reduction in the viable cell population. The overall apoptotic response in H1299 cells exceeded that observed in A549 cells, suggesting that p53 deficiency does not impair, and may even enhance, susceptibility to 5a-induced cell death.

Comparative treatment with the reference compound mitonafide resulted in a smaller increase in apoptotic fractions in both cell lines (Figure S36), highlighting the superior proapoptotic efficacy of compound 5a.

Taken together, these findings demonstrate that compound 5a efficiently induces apoptosis in lung carcinoma cells in a time-dependent manner. This apoptotic response is consistent with a downstream consequence of the replication stress and DNA damage phenotypes observed in preceding experiments.

Activation of Executioner Caspases by 5a

To further substantiate the involvement of apoptotic pathways in the cytotoxic activity of compound 5a, the activation of executioner caspases-3 and -7 was evaluated in A549 and H1299 cells following 24 h treatment with IC₅₀ and IC₇₅ concentrations (Figure 7).

In untreated A549 cells, basal caspase-3/7 activity was low and restricted to a small fraction of the cell population. Exposure to 5a resulted in a pronounced, dose-dependent increase in green fluorescence intensity corresponding to activated caspase-3/7 (Figure 7a). Quantitative analysis confirmed a statistically significant increase in caspase activity at both concentrations compared with untreated controls (p < 0.05), with the strongest activation observed after 24 h at the IC₅₀ (Figure 7b).

A comparable pattern of caspase-3/7 activation was detected in H1299 cells, indicating that executioner caspase activation occurs irrespective of p53 status. Both the proportion of caspase-positive cells and the overall fluorescence intensity were markedly increased following treatment with 5a, confirming robust activation of downstream apoptotic machinery (Figure 7c,d).

Taken together, these results demonstrate that compound 5a induces caspase-dependent apoptosis in lung carcinoma cells. The activation of executioner caspases corroborates the Annexin V/PI flow-cytometric findings and functionally links the replication stress and DNA damage phenotypes observed in earlier experiments to the execution phase of programmed cell death.

Induction and Nuclear Accumulation of p53 by 5a

To assess whether the DNA damage response induced by compound 5a involves p53 signaling, p53 expression, and subcellular localization were analyzed by immunofluorescence microscopy in A549 cells following 24 h treatment with IC₅₀ and IC₇₅ concentrations (Figure 8).

In p53-proficient A549 cells, untreated controls exhibited weak and diffuse nuclear p53 staining. Exposure to 5a resulted in a pronounced, concentration-dependent increase in nuclear p53 signal intensity. After 24 h at IC₅₀, moderate nuclear accumulation of p53 was observed, which further intensified at the higher concentration (IC₇₅). Quantitative analysis confirmed a statistically significant increase in mean nuclear p53 fluorescence intensity compared with untreated controls (p < 0.05), consistent with p53 stabilization in response to DNA damage.

These findings indicate that treatment with compound 5a is associated with activation of p53-dependent DNA damage response signaling, leading to nuclear accumulation of p53. This response is consistent with the observed G₁-phase cell cycle arrest in A549 cells and likely contributes to the downstream induction of apoptosis, as supported by Annexin V/PI staining and caspase-3/7 activation.

3.3. Proposed Mechanistic Model of Action of 5a

Together, the results obtained from cytotoxicity assays, cell cycle profiling, EdU incorporation, γH2AX immunostaining, Annexin V/PI flow cytometry, caspase activation, and p53 immunofluorescence converge to define a coherent mechanistic framework for the action of compound 5a in lung carcinoma cells (Figure 9, schematic summary). Compound 5a exhibits potent antiproliferative activity against both A549 (p53 wild-type) and H1299 (p53-null) cells, surpassing the efficacy of the classical naphthalimides mitonafide and amonafide.

The initial cellular response to 5a involves a pronounced inhibition of DNA replication, as evidenced by strong suppression of EdU incorporation. This replication blockade is accompanied by the accumulation of DNA damage, reflected by robust γH2AX signaling. In p53-proficient A549 cells, these genotoxic events are associated with nuclear accumulation of p53, leading to G₁-phase cell cycle arrest, consistent with checkpoint engagement and p53-mediated regulation of cell fate. In contrast, in p53-deficient H1299 cells, the damage response bypasses G₁ control and manifests predominantly as G₂/M-phase arrest, indicating that checkpoint activation occurs through p53-independent mechanisms as well.

Sustained replication stress and DNA damage ultimately trigger apoptotic cell death, as demonstrated by Annexin V/PI staining and robust activation of executioner caspases-3 and -7, marking the terminal phase of apoptosis. Collectively, these findings indicate that compound 5a exerts its cytotoxic effects through a replication stress-associated DNA damage response, leading to checkpoint activation and caspase-dependent apoptosis in both p53-dependent and p53-independent cellular contexts.

This integrative analysis supports a mechanistic model in which 6-nitro substitution within the benzodioxin–naphthalimide scaffold contributes to enhanced genotoxic stress and antiproliferative activity, positioning compound 5a as a potent and selective anticancer candidate against lung carcinoma cells (Figure 9).

4. Discussion

In this study, we demonstrate that compound 5a exerts potent antiproliferative activity in lung carcinoma cells through a multi-step mechanism initiated by replication interference and replication stress, followed by DNA damage accumulation, checkpoint engagement, and caspase-dependent apoptosis. The convergence of suppressed EdU incorporation, γH2AX accumulation, and cell-cycle perturbation supports a replication stress–associated mode of action, consistent with mechanisms reported for DNA-interacting naphthalimide derivatives. Compared with the reference naphthalimides mitonafide and amonafide, compound 5a displays markedly lower IC₅₀ values while maintaining improved selectivity indices within the tested cell panel, particularly toward p53-deficient H1299 cells. These findings underscore the importance of benzodioxin annulation and C-6 nitro substitution as key structural determinants of enhanced biological activity within the naphthalimide scaffold.

The classical 1,8-naphthalimide nucleus has long been recognized for its DNA-intercalating properties and topoisomerase inhibition [2,9,36], yet its clinical development has been limited by non-specific toxicity and poor selectivity. Structural modification at the C-6 position is known to modulate electronic properties and π–π stacking interactions with nucleic acids [37,38]. In this context, the nitro substituent in 5a likely enhances electron-withdrawing character, contributing to altered DNA interactions and pronounced cytotoxicity. While direct biophysical evidence of DNA binding or topoisomerase inhibition was not obtained in the present study, the observed biological effects are consistent with enhanced interference with DNA replication processes.

Mitonafide and amonafide were selected as reference compounds because they represent clinically investigated naphthalimide derivatives with established DNA-interacting properties, allowing direct benchmarking within the same chemical class. Classical replication stress inducers such as hydroxyurea or aphidicolin, which directly inhibit nucleotide metabolism or DNA polymerase activity, were not included because the primary objective of this study was to assess whether benzodioxin annulation enhances replication-associated stress relative to structurally related naphthalimides rather than to mechanistically distinct replication inhibitors. Accordingly, EdU suppression and γH2AX induction by 5a are interpreted within this comparative framework.

Functional assays provide coherent support for a replication stress-associated mechanism of action. Suppression of EdU incorporation and accumulation of γH2AX were detected prior to extensive apoptotic commitment, indicating that impaired DNA synthesis represents an early cellular response rather than a secondary consequence of cell death. Importantly, the mechanistic conclusions presented herein are based on cellular markers of replication stress and DNA damage response activation, rather than on direct identification of a specific molecular target or replication fork component. Similar γH2AX-associated phenotypes have been reported for naphthalimide derivatives that interfere with DNA processing during S phase [39,40,41]. Accordingly, the replication stress-associated mechanism proposed here is inferred from the convergence of reduced DNA synthesis, DNA damage accumulation, and checkpoint engagement at defined IC₅₀/IC₇₅ concentrations and early time points (24–48 h), rather than from direct interrogation of replication fork dynamics or replication stress signaling intermediates.

Cell-cycle analysis revealed distinct responses in A549 and H1299 cells, highlighting the influence of cellular context on checkpoint engagement following replication stress. In p53-proficient A549 cells, DNA damage was associated with nuclear p53 accumulation and G₁-phase arrest, consistent with p53-dependent checkpoint control. In contrast, p53-deficient H1299 cells predominantly accumulated in the G₂/M phase, indicating engagement of p53-independent mechanisms. Given the intrinsic genetic and molecular differences between these cell lines, these phenotypes cannot be attributed exclusively to p53 status; however, they support a contributory role of p53 in shaping the checkpoint response to compound 5a. While direct biochemical interrogation of checkpoint regulators was not performed, the observed perturbations are consistent with differential engagement of DNA damage checkpoints rather than assignment to a single canonical pathway.

Sustained replication stress and DNA damage ultimately culminated in apoptotic cell death in both cellular models. Annexin V/PI staining and robust activation of executioner caspases-3 and -7 confirmed caspase-dependent apoptosis. The induction of apoptosis in both p53-proficient and p53-deficient cells indicates that persistent genotoxic stress is sufficient to trigger the execution phase of programmed cell death even when canonical p53-mediated transcriptional pathways are compromised. These apoptotic outcomes likely represent downstream consequences of sustained replication interference rather than primary drivers of early DNA synthesis inhibition.

It should be noted that selectivity was evaluated using a single non-malignant fibroblast line (MRC-5) and therefore reflects selectivity within the tested experimental panel rather than definitive tumor specificity. Differences in growth characteristics and culture conditions between normal and cancer cells may also influence absolute IC₅₀ values, representing a limitation of the present study.

Taken together, the present data support a mechanistic model in which compound 5a initiates a cascade beginning with replication interference and replication stress-associated DNA damage, followed by checkpoint activation and culminating in caspase-dependent apoptosis. Notably, this sequence operates in both p53-dependent and p53-independent contexts, suggesting that 5a may circumvent a common resistance mechanism limiting the efficacy of DNA-targeting chemotherapeutics. Further mechanistic resolution—such as DNA fiber analysis, interrogation of RPA signaling, or synchronized cell-cycle progression assays—will be required to distinguish primary replication fork perturbation from downstream cytotoxic effects and represents an important direction for future studies.

5. Conclusions

In conclusion, the benzodioxin-annulated 1,8-naphthalimide derivative 5a exhibits potent antiproliferative activity and improved selectivity in the tested cell panel against lung carcinoma cells, through a replication stress-associated mechanism involving DNA damage, checkpoint arrest, and caspase-dependent apoptosis. Notably, the compound remains effective in both p53-proficient and p53-deficient cellular contexts, highlighting its potential to overcome a key resistance mechanism in DNA-targeting anticancer therapy. Together, these findings position compound 5a as a promising lead for further preclinical development of naphthalimide-based anticancer agents.