Quinoline carboxylic acid derivatives as potent ectonucleotidase inhibitors.

Introduction
Tumor progression is intricately linked to immune cell infiltration and the resulting inflammatory milieu, which collectively promote angiogenesis, immune escape, and metastasis1. Tumors also evolve mechanisms to subvert immune detection by manipulating molecular cues within the tumor microenvironment (TME). A central component of this immune evasion is the accumulation of adenosine, a small molecule with potent immunosuppressive activity that dampens the function of tumor-infiltrating leukocytes through the activation of A2A and A2B adenosine receptors2,3. Adenosine levels in the TME are tightly regulated by ectonucleotidases, cell surface enzymes that sequentially hydrolyze extracellular ATP into AMP and subsequently into adenosine microenvironment4. These enzymes include nucleoside triphosphate diphosphohydrolases (NTPDases), alkaline phosphatases (ALPs), nucleotide pyrophosphatases/phosphodiesterases (NPPs), and ecto-5′-nucleotidase (e5′NT)5,6. NTPDases hydrolyze ATP to ADP and AMP, while e5′NT converts AMP to adenosine, as illustrated in Fig. 1. NPPs, such as NPP1, also participate by hydrolyzing nucleotides to monophosphates and pyrophosphate7. Overexpression of these ectonucleotidases has been reported in multiple types of cancer, such as melanoma8, breast9,10, and prostate cancers11. Their overactivity is associated with tumor growth, suppression of immune responses, and poor clinical outcomes12–14.
Increased enzymatic activity leads to adenosine accumulation in the TME, which promotes tumor tolerance by inhibiting T and B cell function, natural killer (NK) cell cytotoxicity, and dendritic cell (DC) activation15,16. Beyond cancer, aberrant ectonucleotidase expression is implicated in conditions such as insulin resistance17, cardiovascular disease18, and pathological mineralization, reinforcing their therapeutic relevance19. Inhibiting these enzymes offers a promising strategy to reduce adenosine levels, reverse immunosuppression, and enhance anti-tumor immunity20–22.
A diverse array of heterocyclic scaffolds has been explored for the targeted treatment of cancer, with particular focus on ectonucleotidase inhibition as a strategy to modulate purinergic signaling and counteract tumor immune evasion. Both nucleotide- and non-nucleotide-based inhibitors have been investigated in this context23. Among the nucleotide analogs, quemliclustat (AB680) has demonstrated promising preclinical results and is currently undergoing Phase 1 clinical trials for cancer therapy. However, such compounds often suffer from limitations, including poor chemical stability, low membrane permeability, and susceptibility to enzymatic degradation24. To address these shortcomings, various classes of non-nucleotide inhibitors have emerged. Sulfonate derivatives25–29, triazole-based molecules, and pyrazole analogs have shown nanomolar potency against ectonucleotidases30, though issues with metabolic instability and selectivity persist. Despite these efforts, many reported inhibitors exhibit only moderate efficacy and suboptimal pharmacokinetic profiles, underscoring the need for alternative scaffolds with improved biological and drug-like properties31. Representative examples of standard ectonucleotidase inhibitors are illustrated in Fig. 2.
The design of quinoline-6-carboxylic acid amides as potential ectonucleotidase inhibitors is guided by both structural considerations and prior literature. Quinoline is a well-established heterocyclic scaffold with recognized anticancer potential32–35. Its planar, aromatic structure and embedded nitrogen atom make it structurally analogous to purine nucleotides such as ATP, allowing it to interact with nucleotide-binding pockets of ectonucleotidases like CD39 and CD7332–35. Additionally, quinoline derivatives have shown inhibitory activity against cancer-relevant enzymes, including topoisomerases36, and kinases37. Incorporating a carboxylic acid at the 6-position enables the scaffold to mimic phosphate moieties, fostering polar interactions within the enzyme active site, while conversion to a carboxamide improves solubility, stability, and membrane permeability, key features for targeting extracellular enzymes38–43. Despite these favorable characteristics, previously reported quinoline-based ectonucleotidase inhibitors have shown weak potency (IC₅₀ > 100 µM) and poor isoform selectivity. These limitations emphasize the need for improved analogs with enhanced biological activity and drug-like properties44.
In this context, recent studies have reported quinoline-based carboxamides to be effective inhibitors of ATP-hydrolyzing enzymes, suggesting their potential interaction with the catalytic regions of the enzymes involved in purine metabolism38,39. Given their known anticancer properties and favorable drug-like features, we were inspired to design, synthesize, and explore quinoline-6-carboxylic acid amides as inhibitors of ectonucleotidases—another promising target involved in cancer immune evasion. Molecular docking was employed as a post-processing analysis to examine binding orientations and molecular interactions with the target proteins. The biological activity of the synthesized compounds was, then, evaluated through fluorescence assay to assess cellular interaction, MTT assay for measuring cytotoxicity, and apoptosis assay to determine the cell-death inducing potential of these compounds.

Experimental section

General chemistry methods
All the solvents used in this study were of analytical grade and used without further purification. Chemicals were purchased from local suppliers and used directly in the reactions. Melting points of the synthesized compounds were measured using a Gallenkamp melting point apparatus (Model: AAR3235). Reaction progress was monitored by thin-layer chromatography (TLC) using Merck precoated silica gel plates (60 F254), with spot detection under UV light at 254 nm to confirm complete consumption of the starting materials. Structural analysis of the compounds was performed using a Bruker Avance III HD NMR spectrometer operating at 300 and 400 MHz. Both proton (1H) and carbon (13C) NMR spectra were recorded, with chemical shifts reported in parts per million (ppm) and coupling constants (J) in Hertz (Hz). Signal multiplicities were reported as follows: s for singlet, d for doublet, t for triplet, and q for quartet. Elemental analysis (C, H, N, S) was performed using a LECO 630-200-200 TruSpec CHNS microanalyzer to confirm the purity and composition of the synthesized compounds. HPLC analysis was performed on a high-performance liquid chromatograph, LC-QGL3100, Jinan City, Shandong Province, China. The HPLC chromatographs are presented in the supplementary information Figures S25-S36.

General synthetic procedure (A) for quinoline-based amide derivatives (4a-4l)
Quinoline-6-carboxylic acid (1.15 mmol, 200 mg, 1 equiv) was reacted with thionyl chloride (3.46 mmol, 412 mg, 3 equiv) in the presence of DCM (20 mL) as a solvent under reflux at 50 °C for 8 h. The conversion of the acid into the corresponding acid chloride was monitored by TLC. Once the conversion was complete, the acid chloride was treated with various amines and stirred overnight, with the reaction progress monitored through TLC. Upon completion, the reaction mixture was concentrated using a rotary evaporator to remove the solvent. The resulting residue was added to a beaker containing distilled water and stirred until a precipitate formed. The precipitate was filtered and dried in an oven at 37 °C, yielding quinoline-based amide derivatives (4a–4l) in good to excellent yields.

N-(4-chlorobenzyl)quinoline-6-carboxamide (4a)
Using the general synthetic protocol, A, (4-Chlorophenyl)methanamine (1.15 mmol, 164 mg, 140 μL, 1 equiv) was employed as the starting amine, yielding the target compound 4a as a sand brown solid (67%); m.p: 185–187 °C; Rf = 0.3 (hexane/ EtOAc = 3:2); 1H NMR (400 MHz, DMSO) ppm δ 9.32 (t, J = 6.0 Hz, 1H, N12-H), 8.99 (dd, J = 4.2, 1.7 Hz, 1H, C2-H), 8.55 (d, J = 2.0 Hz, 1H, C5-H), 8.48 (dd, J = 8.4, 1.7 Hz, 1H, C4-H), 8.21 (dd, J = 8.7, 2.0 Hz, 1H, C7-H), 8.09 (d, J = 8.8 Hz, 1H, C8-H), 7.61 (dd, J = 8.3, 4.2 Hz, 1H, C3-H), 7.40 (s, 4H, C15, 16, 18, 19-H), 4.53 (d, J = 5.9 Hz, 2H, C13-H); 13C NMR (101 MHz, DMSO) δ 166.0 (s, C-11), 152.1 (s, C-2), 148.7 (s, C-9), 138.6 (s, C-14), 137.1 (s, C-4), 132.0 (s, C-17), 131.4 (s, C-10), 129.2 (s, C-15, 19), 129.1 (s, C-10), 128.3 (s, C-5), 128.1(s, C-8), 127.7 (s, C-16, C18), 127.1 (s, C-7), 122.2 (s, C-3), 42.2 (s, C-13) ; FT-IR (neat) ῡ (cm–1) = 3453 cm-1 (N–H stretching), 1654 cm-1 (C = O stretching); Calculated for C17H13ClN2O, %: C, 68.81; H, 4.42; N, 9.44; Found, %: C, 68.90; H, 4.51; N, 9.47. HPLC: RT (min) = 3.030, Purity (%) = 96.78.

N-(4-fluorobenzyl)quinoline-6-carboxamide (4b)
According to general procedure, A, (4-fluorophenyl)methanamine (1.15 mmol, 145 mg, 141μL, 1 equiv) was employed as the starting amine, yielding the target compound 4b as a chocolate brown solid (64%); m.p: 150–152 °C; Rf = 0.3 (hexane/ EtOAc = 4.6); 1H NMR (300 MHz, DMSO) ppm δ 9.31 (t, J = 6.0 Hz, 1H, N12-H), 8.99 (dd, J = 4.2, 1.8 Hz, 1H, C2-H), 8.55 (d, J = 2.0 Hz, 1H, C5-H), 8.48 (dd, J = 8.2, 2.3 Hz, 1H, C4-H), 8.21 (dd, J = 8.8, 2.0 Hz, 1H, C7-H), 8.09 (d, J = 8.8 Hz, 1H, C8-H), 7.61 (dd, J = 8.3, 4.2 Hz, 1H, C3-H), 7.46–7.33 (m, 2H, C16,18-H), 7.17 (t, J = 8.9 Hz, 2H C15,19-H), 4.53 (d, J = 5.9 Hz, 2H, C13-H); 13C NMR (75 MHz, DMSO) ppm δ 166.3 (s, C-11), 161.7 (d, 1JC-F = 240.8 Hz, s, C-17), 152.6 (s, C-2), 149.1 (s, C-9), 137.9 (s, C-4), 136.2 (d, 4JC-F = 2.0 Hz, C-14), 132.6 (s, C-10), 129.8 (d, 3JC-F = 8.1 Hz, C-15), 129.5 (s, C-5), 128.5 (s, C-8), 128.2 (s, C-7), 127.6 (s, C-6) 122.7 (s, C-3), 115.5 (d, 2JC-F = 21.1 Hz, C-16), 42.6 (s, C-13); FT-IR (neat) ῡ (cm–1) = 3279 cm-1 (N–H stretching), 1635 cm-1 (C = O stretching); Calculated for C17H13FN2O, %: C, 72.85; H, 4.67; N, 9.99; Found, %: C, 72.96; H, 4.74; N, 10.02. HPLC: RT (min) = 2.887, Purity (%) = 93.97.

N-(4-methylbenzyl)quinoline-6-carboxamide (4c)
Using the general synthetic protocol, A, p-tolymethanamine (1.15 mmol, 139 mg, 147 μL, 1 equiv) was employed as the starting amine, yielding the target compound 4c as a light brown solid (70%); m.p: 190–192 °C; Rf = 0.4 (hexane/ EtOAc = 4:1); 1H NMR (300 MHz, DMSO) ppm δ 9.33 (t, J = 6.0 Hz, 1H, N12-H), 8.98 (dd, J = 4.2, 1.7 Hz, 1H, C2-H), 8.57 (d, J = 2.0 Hz, 1H, C5-H), 8.47 (d, J = 8.3 Hz, 1H, C4-H), 8.22 (dd, J = 8.8, 2.0 Hz, 1H, C7-H), 8.08 (d, J = 8.8 Hz, 1H, C8-H), 7.61 (dd, J = 8.3, 4.2 Hz, 1H, C3-H), 7.25 (d, J = 7.8 Hz, 2H, C19-H), 7.14 (d, J = 7.8 Hz, 2H, C16-H), 4.49 (d, J = 5.9 Hz, 2H, C13-H), 2.27 (s, 3H, C18-H); 13C NMR (75 MHz, DMSO) ppm δ 166.1 (s, C-11), 152.2 (s, C-2), 149.1 (s, C-9), 137.6 (s, C-4), 137.0 (s, C-17), 136.3 (s, C-6), 132.7 (s, C-14), 129.4 (s, C-10), 129.3 (s, C-5), 128.5 (s, C-8), 128.2 (s, C-16), 127.8 (s, C-15), 127.6 (s, C-7), 122.6 (s, C-3), 43.0 (s, C-13), 21.1 (s, C18); FT-IR (neat) ῡ (cm–1) = 3296 cm-1 (N–H stretching), 1635 cm-1 (C = O stretching); Calculated for C18H16N2O, %: C, 78.24; H, 5.84; N, 10.14; Found, %: C, 78.38; H, 5.89; N, 10.17. HPLC: RT (min) = 2.873, Purity (%) = 99.15.

N-(2,6 difluorophenyl)quinoline-6-carboxamide (4d)
Using the general synthetic protocol, A, 2,6 difluoroaniline (1.15 mmol, 149 mg, 124 μL, 1 equiv) was employed as the starting amine, yielding the target compound 4d as a light brown solid (70%); m.p: 243–245 °C; Rf = 0.2 (hexane/ EtOAc = 7:3); 1H NMR (300 MHz, DMSO) ppm δ 10.45 (s, 1H, N12-H), 9.04 (dd, J = 4.2, 1.8 Hz, 1H, C2-H), 8.70 (d, J = 2.0 Hz, 1H, C5-H), 8.56 (dd, J = 8.3, 1.8 Hz, 1H, C4-H), 8.29 (dd, J = 8.8, 2.0 Hz, 1H, C7-H), 8.16 (d, J = 8.8 Hz, 1H, C8-H), 7.66 (dd, J = 8.3, 4.2 Hz, 1H, C3-H), 7.44 (tt, J = 8.8, 6.4 Hz, 1H, C16-H), 7.32–7.18 (m, 2H, C17-H); 13C NMR (75 MHz, DMSO) ppm δ 165.5 (s, C-11), 158.21 (d, 1JC-F = 247.4 Hz, C-14), 158.18 (d, 1JC-F = 247.3 Hz, C-18), 152.6 (s, C-2), 149.0 (s, C-9), 137.3 (s, C-4), 131.0 (s, C-6), 129.4 (s, C-10), 129.1 (s, C-5), 129.05 (s, C-8), 128.6 (t, 3JC-F = 9.8 Hz, C-16), 127.6 (s, C-7), 122.5 (s, C-3), 114.7 (t, 3JC-F = 9.3 Hz, C-15,17), 112.0 (d, 2JC-F = 22.3 Hz, C-13); FT-IR (neat) ῡ (cm–1) = 3165 cm-1 (N–H stretching), 1665 cm-1 (C = O stretching); Calculated for C16H10F2N2O, %: C, 67.60; H, 3.55; N, 9.85; Found, %: C, 68.68; H, 3.64; N, 9.89. HPLC: RT (min) = 2.813, Purity (%) = 94.27.

N-(2,4 difluorophenyl)quinoline-6-carboxamide (4e)
Using the general synthetic protocol, A, 2,4-difluoroaniline (1.15 mmol, 149 mg, 117 μL, 1 equiv) was employed as the starting amine, yielding the target compound 4e as a brown (60%); m.p: 186–188 °C; Rf = 0.5 (hexane/ EtOAc = 3:2); 1H NMR (400 MHz, DMSO) ppm δ 10.40 (s, 1H, N12-H), 9.02 (dd, J = 4.2, 1.8 Hz, 1H, C2-H), 8.66 (d, J = 2.1 Hz, 1H, C5-H, ), 8.54 (dd, J = 8.4, 1.7 Hz, 1H, C4-H), 8.26 (dd, J = 8.8, 2.0 Hz, 1H, C7-H), 8.14 (d, J = 8.8 Hz, 1H, C8-H), 7.70–7.59 (m, 2H, C3, C18-H), 7.39 (ddd, J = 10.6, 9.1, 2.9 Hz, 1H, C17-H), 7.15 (tdd, J = 8.5, 2.8, 1.4 Hz, 1H, C15-H); 13C NMR (100 MHz, DMSO)(a) ppm δ 165.1 (s, C-11), 152.4 (s, C-2), 148.9 (s, C-9), 137.3 (s, C-4), 131.6 (s, C-6), 129.2 (s, C-10), 128.8 (s, C-5), 128.4 (dd, J = 10.3, 2.8 Hz, C-18), 128.0(s, C-8), 127.1 (s, C-7), 122.4 (s, C-3), 111.3 (d, 2JC-F = 25.5 Hz, C-13), 104.7–104.0 (m, 2C, C-15,17); FT-IR (neat) ῡ (cm–1) = 3279 cm-1 (N–H stretching), 1635 cm-1 (C = O stretching). Calculated for C16H10F2N2O, %: C, 67.60; H, 3.55; N, 9.85; Found, %: C, 68.70; H, 3.66; N, 9.90. HPLC: RT (min) = 2.907, Purity (%) = 94.55.
(a)Due to the dilute nature of the sample, the signals for the two carbons directly attached to fluorine, expected in the range of 159–160 ppm, are missing.

N-[5-fluoro-2-(trifluoromethyl) phenyl]quinoline-6-carboxamide (4f)
Using the general synthetic protocol, A, 4-fluoro-2-(trifluoromethyl) aniline (1.15 mmol, 206 mg, 150 μL, 1 equiv) was employed as the starting amine, yielding the target compound 4f. as a desert brown (65%); m.p: 200–202 °C; Rf = 0.3 (hexane/ EtOAc = 3:2); 1H NMR (300 MHz, DMSO) ppm δ 10.46 (s, 1H, N12-H), 9.03 (dd, J = 4.2, 1.7 Hz, 1H, C2-H), 8.64 (d, J = 2.0 Hz, 1H, C5-H), 8.60–8.50 (m, 1H, C4-H), 8.26 (dd, J = 8.8, 2.0 Hz, 1H, C7-H), 8.16 (d, J = 8.8 Hz, 1H, C8-H), 7.75 (dt, J = 8.9, 1.8 Hz, 1H, C17-H), 7.70–7.60 (m, 3H, C3, 8, 15-H); 13C NMR (75 MHz, DMSO) ppm δ 166.2 (s, C-11), 160.3 (d, J = 244.6 Hz, C-17), 152.8 (s, C-2), 148.9 (s, C-9), 137.3 (s, C-4), 134.0 (s, C-6), 133.9 (s, C-5), 132.1 (s, C-13), 131.6 (s, C-10), 129.3 (s, C-5), 128.7 (s, C-8), 127.7 (s, C-7), 127.2 (s, C-19), 122.4 (s, C-3), 120.4 (s, C-16), 120.1 (s, C-18), 114.0 (s, C-14); FT-IR (neat) ῡ (cm–1) = 3345 cm-1 (N–H stretching), 1650 cm-1 (C = O stretching); Calculated for C17H10F4N2O, %: C, 61.08; H, 3.02; N, 8.38; Found, %: C, 61.19; H, 3.12; N, 8.41. HPLC: RT (min) = 2.873, Purity (%) = 99.152.

N-(pyridin-2-yl)quinoline-6-carboxamide (4g)
Using the general synthetic protocol, A, pyridine-2-amine (1.15 mmol, 109 mg, 1 equiv) was employed as a the starting amine, yielding the target compound 4 g as a cream yellow (68%); m.p: 97–99 °C; Rf = 0.4 (hexane/ EtOAc = 3:7); 1H NMR (300 MHz, DMSO) ppm δ 11.03 (s, 1H, N12-H), 9.02 (dd, J = 4.2, 1.8 Hz, 1H, C2-H), 8.76 (d, J = 2.0 Hz, 1H, C5-H), 8.51 (ddd, J = 8.4, 1.8, 0.8 Hz, 1H, C4-H), 8.42 (ddd, J = 4.9, 2.0, 0.9 Hz, 1H, C7-H), 8.30 (dd, J = 8.8, 2.1 Hz, 1H, C8-H), 8.26–8.22 (m, 1H, C14-H), 8.11 (d, J = 8.8 Hz, 1H, C3-H), 7.88 (ddd, J = 8.4, 7.3, 2.0 Hz, 1H, C16-H), 7.64 (dd, J = 8.3, 4.2 Hz, 1H, C17-H), 7.19 (ddd, J = 7.3, 4.9, 1.0 Hz, 1H, C15-H); 13C NMR (75 MHz, DMSO) ppm δ 165.6 (s, C-11), 152.4 (s, C-2), 152.2 (s, C-13), 148.9 (s, C-15), 148.1 (s, C-9), 138.3 (s, C-17), 137.4 (s, C-4), 131.8 (s, C-6), 129.2 (s, C-10), 129.0 (s, C-5), 128.2 (s, C-8), 127.0 (s, C-7), 122.3 (s, C-3), 120.0 (s, C-16), 114.7 (s, C-18); FT-IR (neat) ῡ (cm–1) = 3294 cm-1 (N–H stretching), 1683 cm-1 (C = O stretching); Calculated for C15H11N3O, %: C, 72.28; H, 4.45; N, 16.86; Found, %: C, 72.41; H, 4.58; N, 16.90. HPLC: RT (min) = 2.997, Purity (%) = 98.05.

N-(2-iodophenyl)quinoline-6-carboxamide (4h)
Using the general synthetic protocol, A, 2-Iodoaniline (1.15 mmol, 253 mg, 1 equiv) was employed as the starting amine, yielding the target compound 4 h as a Coffee brown solid (65%); m.p: 182–184 °C; Rf = 0.3 (hexane/ EtOAc = 4:6); 1H NMR (300 MHz, DMSO) ppm δ: 10.36 (s, 1H, N12-H), 9.03 (dd, J = 4.2, 1.7 Hz, 1H, C2-H), 8.70 (d, J = 2.0 Hz, 1H, C5-H), 8.58–8.52 (m, 1H, C4-H), 8.32 (dd, J = 8.8, 2.0 Hz, 1H, C7-H), 8.16 (d, J = 8.8 Hz, 1H, C8-H), 7.97 (dd, J = 7.7, 1.2 Hz, 1H, C15-H), 7.65 (dd, J = 8.3, 4.2 Hz, 1H, C3-H), 7.49 (dd, J = 6.7, 1.6 Hz, 2H, C16, 18-H), 7.10 (ddd, J = 7.9, 6.5, 2.4 Hz, 1H, C17-H); 13C NMR (75 MHz, DMSO)(a) ppm δ: 164.9 (s, C-11), 152.4 (s, C-2), 148.9 (s, C-13), 139.8 (s, C-9), 139.0 (s, C-15), 137.3 (s, C-4), 132.0 (s, C-6), 129.3 (s, C-10), 128.9 (s, C-5), 128.7 (2C, C8,17), 128.5 (s, C-7), 127.9 (s, C-16), 127.2 (s, C-18), 122.4 (s, C-3), 99.1 (s, C-14); FT-IR (neat) ῡ (cm–1) = 3392 cm-1 (N–H stretching), 1652 cm-1 (C = O stretching); Calculated for C16H11IN2O, %: C, 51.36; H, 2.96; N, 7.49; Found, %: C, 51.42; H, 2.99; N, 7.52. HPLC: RT (min) = 2.987, Purity (%) = 99.16.

N-(2, 4-dimethoxybenzyl)quinoline-6-carboxamide (4i)
Using the general synthetic protocol, A, (2,4 dimethoxyphenyl)methanamine (1.15 mmol, 193 mg, 174 μL, 1 equiv) was employed as the starting amine, yielding the target compound 4i as a Dark brown solid (65%); m.p: 120–123 °C; Rf = 0.6 (hexane/ EtOAc = 2:8); 1H NMR (300 MHz, DMSO-d6) ppm δ 9.05–8.96 (m, 2H, C2-H, N12-H), 8.56 (d, J = 2.0 Hz, 1H, C5-H), 8.48 (dd, J = 8.4, 1.8 Hz, 1H, C4-H), 8.22 (dd, J = 8.8, 2.0 Hz, 1H, C7-H), 8.08 (d, J = 8.8 Hz, 1H, C8-H), 7.62 (dd, J = 8.3, 4.2 Hz, 1H, C3-H), 7.16 (d, J = 8.3 Hz, 1H, C18-H), 6.58 (d, J = 2.4 Hz, 1H, C15-H), 6.49 (dd, J = 8.3, 2.4 Hz, 1H, C17-H), 4.44 (d, J = 5.6 Hz, 2H, C11-H), 3.82 (s, 3H, C19-H), 3.74 (s, 3H, C20-H); 13C NMR (75 MHz, DMSO) ppm δ 165.8 (s, C-11),, 159.7 (s, C-17), 157.6 (s, C-15), 151.9 (s, C-2), 148.5 (s, C-9), 137.3 (s, C-4), 132.4 (s, C-6), 128.8 (s, C-10), 128.6 (s, C-19), 128.0 (s, C-5), 127.9 (s, C-8), 127.1 (s, C-7), 122.2 (s, C-3), 118.9 (s, C-14), 104.4 (s, C-18), 98.2 (s, C-16), 55.5 (s, C-20), 55.2 (s, C-21), 37.6 (s, C-13); FT-IR (neat) ῡ (cm–1) = 3289 cm-1 (N–H stretching), 1638 cm-1 (C = O stretching); Calculated for C19H18N2O3, %: C, 70.79; H, 5.63; N, 8.69; Found, %: C, 70.88; H, 5.70; N, 8.73. HPLC: RT (min) = 2.900, Purity (%) = 97.54.

N-(2,4 -dichlorobenzyl)quinoline-6-carboxamide (4j)
Using the general synthetic protocol, A, (2,4 dichlorophenyl)methanamine (1.15 mmol, 203 mg, 155 μL,1 equiv) was employed as the starting amine, yielding the target compound 4j as a Light brown solid (67%); m.p: 165–167 °C; Rf = 0.4 (hexane/ EtOAc = 6:4); 1H NMR (300 MHz, DMSO-d6) δ: 9.33 (t, J = 5.7 Hz, 1H, N12-H), 9.03–8.94 (m, 1H, C2-H), 8.58 (d, J = 2.0 Hz, 1H, C5-H), 8.52–8.43 (m, 1H, C4-H), 8.22 (dd, J = 8.8, 2.0 Hz, 1H, C7-H), 8.10 (d, J = 8.8 Hz, 1H, C8-H), 7.64 (d, J = 0.8 Hz, 1H, C15-H), 7.63–7.57 (m, 1H, C3-H), 7.43 (d, J = 1.2 Hz, 2H, C17,C18-H), 4.57 (d, J = 5.7 Hz, 2H, C11-H); 13C NMR (75 MHz, DMSO-d6) δ: 166.1 (s, C-11), 152.2 (s, C-2), 148.9 (s, C-9), 137.2 (s, C-4), 135.5 (s, C-6), 133.0 (s, C-14), 132.3 (s, C-15), 131.8 (s, C-17), 130.3 (s, C-16), 129.2 (s, C-10), 128.7 (s, C-19), 128.2 (s, C-8), 127.8 (s, C-5), 127.4 (s, C-7), 127.2 (s, C-18), 122.3 (s, C-3), 40.5 (s, C-13); FT-IR (neat) ῡ (cm–1) = 3289 cm-1 (N–H stretching), 1638 cm-1 (C = O stretching); Calculated for C17H12Cl2N2O, %: C, 61.65; H, 3.65; N, 8.46; Found, %: C, 61.73; H, 3.72; N, 8.53. HPLC: RT (min) = 3.197, Purity (%) = 97.57.

N-(4-Iodophenyl)quinoline-6-carboxamide (4k)
Using the general synthetic protocol, A, 4-iodoaniline (1.15 mmol, 253 mg, 1 equiv) was employed as the starting amine, yielding the target compound 4 k as a Dark brown (63%); m.p: 182–184 °C; Rf = 0.4 (hexane/ EtOAc = 4:1); 1H NMR (300 MHz, DMSO-d6) δ: 10.72 (s, 1H, N12-H), 9.09 (dd, J = 4.5, 1.7 Hz, 1H, C2-H), 8.75–8.68 (m, 1H, C5-H), 8.69 (d, J = 8.0 Hz, 1H, C4-H), 8.32 (dd, J = 8.8, 2.0 Hz, 1H, C7-H), 8.20 (d, J = 8.9 Hz, 1H, C8-H), 7.90–7.58 (m, 5H, C3, 14, 15, 17, 18-H); 13C NMR (75 MHz, DMSO-d6) δ: 165.0 (s, C-11), 151.2 (s, C-2), 146.8 (s, C-9), 139.2 (s, C-13), 139.0 (s, 2C, C-15,17), 137.4 (s, C-4), 133.0 (s, C-6), 129.0 (s, C-10), 128.8 (s, C-5), 127.6 (s, C-8), 127.2 (s, C-7), 122.5 (s, C-3), 122.5 (s, 2C, C-13,18), 87.7 (s, C-16); FT-IR (neat) ῡ (cm–1) = 3288 cm-1 (N–H stretching), 1678 cm-1 (C = O stretching); Calculated for C16H11IN2O, %: C, 51.36; H, 2.96; N, 7.49; Found, %: C, 51.45; H, 3.05; N, 7.56. HPLC: RT (min) = 3.256, Purity (%) = 97.56.

N-(5-fluoro 2-methylphenyl)quinoline-6-carboxamide (4l)
Using the general synthetic protocol, A, 5-fluoro 2-methylaniline (1.15 mmol, 144 mg, 145 μL,1 equiv) was employed as the starting amine, yielding the target compound 4 l as a brown solid (122 mg, 66%); m.p: 164–166 °C; Rf = 0.4 (hexane/ EtOAc = 3:7); 1H NMR (300 MHz, DMSO) ppm δ: 10.19 (s, 1H, N12-H), 9.02 (dd, J = 4.2, 1.7 Hz, 1H, C2-H), 8.66 (d, J = 2.0 Hz, 1H, C5-H), 8.55 (dd, J = 8.5, 1.8 Hz, 1H, C4-H), 8.28 (dd, J = 8.8, 2.0 Hz, 1H, C7-H), 8.15 (d, J = 8.8 Hz, 1H, C8-H), 7.65 (dd, J = 8.3, 4.2 Hz, 1H, C3-H), 7.34 (ddd, J = 10.3, 8.4, 4.6 Hz, 2H, C18, C16-H), 7.04 (td, J = 8.5, 2.8 Hz, 1H, C15-H), 2.28 (s, 3H, C19-H); 13C NMR (75 MHz, DMSO) ppm δ: 165.1 (s, C-11), 160.2 (d, 1JC-F = 239.1 Hz, C-17), 152.4 (s, C-2), 148.9 (s, C-9), 137.6 (d, 3JC-F = 10.2 Hz, C-15), 137.3 (s, C-4), 132.2 (s, C-6), 131.51 (d, 3JC-F = 8.8 Hz, C-13), 129.2 (s, C-10), 129.2 (s, C-8), 128.7 (s, C-5), 128.1 (s, C-14), 127.2 (s, C-7), 122.4 (s, C-3), 112.69 (d, 2JC-F = 45.0 Hz, C-16), 112.68 (s, C-18), 17.76 (s, C-19) ; FT-IR (neat) ῡ (cm–1) = 3352 cm-1 (N–H stretching), 1649 cm-1 (C = O stretching); Calculated for C17H13FN2O, %: C, 72.85; H, 4.67; N, 9.99; Found, %: C, 72.96; H, 4.79; N, 10.04. HPLC: RT (min) = 2.937, Purity (%) = 98.82.

Density functional theory
The geometry of the selected compounds was optimized using Density Functional Theory (DFT) at the B3LYP/3-21G level to obtain energy-minimized structures for subsequent analysis45–47. The chemical reactivity of the synthesized compounds was determined by analyzing the frontier molecular orbitals. The compounds’ softness and chemical hardness were measured. Gaussian 09 software was used to conduct quantum mechanical research on the multi-core machine, and Gauss View 6 was used to view the optimized structures.

Enzymes preparation
To produce the enzymes, COS-7 cells were transfected with plasmids containing the h-NTPDase1 gene (GenBank accession number U87967)48. The h-NTPDase2 construct (GenBank accession number NM_203468) was generously provided by Aileen F. Knowles49. We received the h-NTPDase3 plasmid (GenBank accession no AF034840) as a generous gift from Terence L. Kirley 49. h-NTPDase8 (GenBank accession no AY430414)50. h-NPP1 (GenBank accession no NM006208) a kind gift of James W. Goding51 and h-e5’NT (GenBank accession no DQ186653)52 as previously described53. Transfection of COS-7 cells was carried out in 10 cm culture plates by incubating them at 37 °C for 5 h using serum-free DMEM/F-12 medium mixed with 6 µg of plasmid DNA and 24 µL Lipofectamine. Following this, the plate received an equal volume of DMEM/F-12 medium containing 20% fetal bovine serum after 6–8 h. After 44–48 h of incubation, the cells were washed 3 times with a harvesting buffer containing 95 mM NaCl, 0.1 mM PMSF, and 45 mM Tris at pH 7.5. After scraping, the cells were centrifuged twice (300 g, 5 min at 4 °C), washed, and resuspended in harvesting buffer with aprotinin (10 µg/mL) before undergoing sonication. Centrifugation at 850 g for 5 min at 4 °C was performed to remove the nuclei and cellular debris. The supernatant obtained was supplemented with 7.5% glycerol and preserved at − 80 °C. Protein concentration was determined using the Bradford assay in a microplate format, with bovine serum albumin (BSA) serving as the standard54.

Enzyme inhibition assays

h-NTPDase activity assay
A significantly modified version of a previously published spectrophotometric technique was used to study the effect of all the synthesized compounds on h-NTPDase1, -2, -3, and -855. The compounds were screened at a concentration of 100 µM. The assay was performed in an incubation medium containing 50 mM of Tris–HCl and 5 mM of CaCl2 maintained at pH 7.4. Reaction medium containing buffer (55 µL), test compound (10 µL, 1 mM working solution; final well concentration 0.1 mM), and 10 µL of enzyme solution of h-NTPDase1 (12 ng/well) or h-NTPDase2 (37 ng/well) or h-NTPDase3 (43 ng/well) or h-NTPDase8 (63 ng/well) was pre-incubated at 37℃. After 10 min, the reaction absorbance was measured (630 nm) with a microplate reader (FLUOStar Omega, BMG Labtech, Germany). 10 µL of substrate i.e. ATP (100 µM) was added to begin the reaction followed by 15 min incubation. The enzymatic reaction was eventually stopped by adding 15 µL of malachite green reagent. By measuring the absorbance at 630 nm, released inorganic phosphate was identified and the percent inhibition was computed. Compounds with > 50% inhibition of any isoform of h-NTPDase, were further diluted and their IC50 values were calculated by generating and fitting the dose–response curves with GraphPad Prism 5.0 (San Diego, CA, USA).

h-NPP1 activity assay
With slight modifications to the previously described procedure, an inhibition assay on h-NPP1 was conducted to verify the inhibitory effect of the synthesized derivatives56. The assay for h-ENPP1 was conducted in a 96-well plate format, with each well containing a total volume of 100 µL. The assay buffer, maintained at pH 9.5, consisted of 50 mM Tris–HCl, 5 mM MgCl₂, and 0.1 mM ZnCl₂. Test compounds were prepared in 10% DMSO and applied at a final concentration of 0.1 mM. Each well received the buffer, 100 µM of the compound, h-ENPP1 enzyme (27 ng/well), and the substrate p-nitrophenyl 5′-thymidine monophosphate. After 25 min of incubation at 37 °C, absorbance was recorded at 405 nm using a BioTek FL*800TM microplate reader (Instruments, Inc., USA). Compounds showing more than 50% inhibition were further evaluated by serial dilution to determine their IC₅₀ values. Data analysis was performed using GraphPad Prism 5.0 (San Diego, CA, USA).

h-e5’NT activity assay
Inhibition assay for human ecto-5'-nucleotidase was carried out according to the previously reported method57. The enzyme inhibition assay was conducted in a final volume of 100 µL per well. The assay buffer consisted of 50 mM Tris–HCl, along with MgCl₂ and CaCl₂, and was adjusted to pH 7.4 before use. Each well was loaded with 55 µL of buffer, 10 µL of e5′NT enzyme solution, and 10 µL of the test compound prepared at a 1 mM working concentration, resulting in a final concentration of 0.1 mM per well. After a 10-min pre-incubation at 37 °C, the enzymatic reaction was initiated by adding 10 µL of 2 mM AMP as the substrate. The mixture was then incubated for an additional 20 min at 37 °C. To detect phosphate release, 15 µL of malachite green reagent was added to each well, and absorbance was recorded at 630 nm. All experiments were conducted in triplicate. Compounds that showed inhibition above 50% were further analyzed to determine their IC₅₀ values using non-linear regression in GraphPad Prism 5.0.

Fluorescence spectroscopy and microscopic analysis
Fluorescence spectroscopy of compound 4d was conducted using a Horiba Fluoromax-4 spectrofluorometer, with excitation wavelength between 400–600 nm, and emission spectra were recorded to characterize its optical properties. Fluorescence imaging experiments were performed using A549 human lung carcinoma cells from American Type Cultures Collection (ATCC, Manassas, VA), while untreated A549 cells were used as controls in fluorescence imaging to account for background autofluorescence and confirm the specificity of compound 4d uptake and signal. To evaluate the effect of compound 4d on enzyme function, a real-time fluorescent-based assay was used to monitor changes in fluorescent intensity, wavelength shifts, and emission patterns to infer binding events. Fluorescence signals were normalized via min–max normalization and analyzed using the microscope’s built-in software.

Molecular docking studies
Quinoline-6-carboxylic acid derivatives 4a, 4b, 4d, 4g, and 4k were docked with respective enzymes. Molecular docking experiments were conducted on the synthesized derivatives (4a-4l) using Autodock Vina® 1.5.4 software58. The crystal structures of human ectonucleotide pyrophosphatase/phosphodiesterase 1 (h-ENPP1) and h-e5′NT were retrieved from the Protein Data Bank https://www.rcsb.org/ using PDB IDs 6WEW and 4H2G, respectively. For h-ENTPDase isoenzymes, homology models were utilized due to the unavailability of their experimental structures. Ligand structures were drawn using ChemDraw software and were then subjected to energy minimization and saved in PDBQT format using Autodock Vina 1.5.4 software. The best pose with minimum binding energy was selected, and re-docking was carried out for validation of the docking results. Visualization of the 2D and 3D protein ligand interactions was done using Discovery Studio 2021.

In silico ADME analysis
The SwissADME web server was employed to assess the pharmacokinetic properties of all the compounds (4a-4l). ChemDraw was used to generate the SMILES representation of the synthesized derivatives, which were then uploaded to an online platform for pharmacokinetic profile prediction.

MTT assay
MTT assay was conducted to check the safety profile of the synthesized derivatives. The assay was conducted in MCF-7 breast cancer cells. Briefly, the cells were seeded at a density of 10,000 cells per well in a sterile 96-well plate and incubated for 24 h in a 5% CO2 incubator at 37 °C. The media was then discarded, and the cells were washed with PBS. Tested compounds were loaded at a concentration of 100 μM/well in triplicate. 1% DMSO and 100 μM cisplatin were used as negative control and positive control, respectively. A blank well that contained only serum-free media served as a reference. The 96-well plate was further incubated for 24 h in 5% CO2 incubator at 37 °C. Following incubation, 100μL of MTT reagent was loaded/well to achieve a final concentration of 0.2 mg/mL. The cells were then placed in an incubator for a further 2 to 4 h and monitored for the formation of formazan crystals. Next, the media was aspirated, and 100 μL/well solubilizing agent (SDS solution) was added. Subsequently, the plates were shaken for 30 min to ensure complete dissolution of the formazan crystals. The absorbance was measured using a microplate reader, with readings taken at 570 nm for the formazan and at 630 nm for the background signal. The conversion of MTT reagent to formazan crystals by mitochondrial dehydrogenases was indicative of cell viability.

Sulforhodamine B (SRB) proliferation assay
The antiproliferative activity of the synthesized compounds was evaluated using the Sulforhodamine B (SRB) colorimetric assay against the MCF-7 human breast cancer cell line. The cell line was originally from ATTC and was validated through STR profiling by Microsynth AG. The cell line used in this study was obtained from ATCC® HTB-22™. Cells were seeded in 96-well plates at an optimized density of 1.5 × 103 cells per well and allowed to adhere overnight under standard culture conditions (37 °C, 5% CO₂). Serial two-fold dilutions of each compound were prepared in complete culture medium to achieve final concentrations ranging from 50 µM to 0.195 µM (9 concentrations).
After 24 h of incubation to allow cell attachment, 25 µL of each compound dilution was added to wells containing 100 µL of cell suspension, and the plates were further incubated for 72 h. Following treatment, cell growth was fixed by adding ice-cold 10% trichloroacetic acid (TCA) to obtain a final concentration of 3% in each well and incubating the plates at 4 °C for at least 2 h. The plates were then washed four times with distilled water and air-dried.
The fixed cells were stained with 0.057% (w/v) Sulforhodamine B (SRB) solution for 30 min at room temperature. Unbound dye was removed by washing four times with 1% acetic acid, and plates were air-dried again. The bound dye was solubilized in 100 µL of 10 mM Tris base (pH 10.5) with shaking for 5 min, and the absorbance was measured at 490 nm using a microplate reader. The percentage of cell viability relative to untreated controls was calculated, and GI₅₀ values (concentration causing 50% growth inhibition) were determined using nonlinear regression analysis in GraphPad Prism software.

Results and discussion

Synthesis
The synthesis of quinoline-6-carboxylic acid derivatives was carried out in accordance with the reported procedure59. First step involved the conversion of quinoline-6-carboxylic acid (1) into quinoline-6-acid chloride (2) using thionyl chloride. The reaction was refluxed for 8 h and monitored by TLC to confirm the consumption of the starting acid and the formation of quinoline-6-acid chloride (2). Subsequent addition of the respective amine led to the formation of the corresponding amides (4a–4l) in good yields, ranging from 60 to 70%. The amidation step was carried out at 50 °C, and the reaction progress was monitored by TLC, which indicated complete conversion of the acid chloride into its respective amide. The resulting amides were fully characterized using FTIR and 1H and 13C NMR spectroscopy.
To assess the versatility of the reaction and introduce structural diversity, a diverse set of amine substrates was explored, as depicted in Fig. 3. This set comprised primary amines, variously substituted benzylamines, functionalized anilines, and heterocyclic aniline derivatives. Among the benzylamines, substituted benzylamines with 4-chloro, 4-fluoro, 4-methyl, and 2,4-dichloro substituents (3a–3c, 3j) produced good to excellent yields of the corresponding amides (4a–4c, 4j), with yields of 67%, 64%, 70%, and 67%, respectively. In a similar fashion, a range of aniline derivatives were tested, including monosubstituted anilines with electron-donating groups, such as chloro (Cl) at the ortho (3h) and para (3k) positions. Additionally, disubstituted anilines with a mix of electron-donating groups (F, OCH3; 3d, 3e, 3i) and both electron-donating and electron-withdrawing groups (F, CF3, CH3; 3f, 3l) were also compatible, yielding the desired products in good to excellent yields. Additionally, heterocyclic pyridyl amine (3g) was also successfully utilized to synthesize the desired compounds (4g), furnishing the products in good yields. For all the synthesized compounds, characteristic peaks were observed in the FTIR spectra. A stretching band appeared in the range of 3165–3462 cm⁻1, indicating the presence of the NH group, while the absorption band at 1628–1682 cm⁻1 confirmed the existence of the C = O group. The structures were further confirmed using 1H, and 13C spectroscopy. The 1H-NMR data of compounds in the δ 9.00–11.06 ppm range confirmed the presence of NH which further confirmed the presence of amide functionality. In 13C-NMR data, the appearance of a signal at δ 164.9–166.3 ppm for C = O further confirmed the formation of quinoline-6-carboxylic acid amide derivatives (4a-4l). All NMR spectra are provided in the Supporting Information (Figure S1-S24).

Enzymatic assays
The ability of the synthetic compounds to inhibit the activity of the h-NTPDase1, -2, -3, and -8 isoenzymes, h-NPP1, and h-e5′NT was evaluated. The newly synthesized compounds showed pronounced inhibition of these enzymes, as listed in Table 1. The enzymatic tests of the compounds (4a–4l) revealed their strong inhibitory effects, with IC50 values ranging from 0.092 µM to 1.00 µM, as shown in Table 2. IC50 values were determined as shown in Table S1 (Supporting Information).
The structure–activity relationship (SAR) analysis demonstrated a substantial influence of substituents on the in vitro activity of the synthesized compounds (4a–4l) against h-NTPDases, h-NPP1, and h-e5′NT. Notably, compound 4a, featuring a 4-chlorobenzyl substitution on the quinoline carboxamide scaffold, exhibited the highest potency against h-e5′NT (IC50 = 0.092 ± 0.02 µM), underscoring the crucial role of the chlorine substituent in enhancing enzyme inhibition. Based on its unique interaction profile and the absence of structurally similar inhibitors in the existing literature, compound 4a may be considered a potential first-in-class h-e5′NT inhibitor. It also demonstrated the strongest overall activity among all tested derivatives. Compound 4j, bearing an additional 2-chloro substituent compared to 4a, exhibited slightly reduced but notable potency, likely due to steric hindrance affecting optimal ligand orientation and diminishing binding efficiency within the h-e5′NT active site. Electron-donating groups (e.g., methyl) at the 4-position (4c) enhanced activity, whereas introduction of electron withdrawing group i.e., fluorine substitution reduced it, emphasizing chlorine’s optimal electronic properties. Pyridyl rings and 2-iodo substitutions maintained good activity, but bulky 4-iodo groups significantly decreased 4 k inhibition potential. We also observed that activity varied with substitution positions. Compound 4e, bearing fluoro substituents at the 2- and 4-positions, retained potency; however, compound 4d, with fluoro groups at the 2- and 6-positions, exhibited reduced activity, emphasizing the importance of precise substituent placement for optimal activity.
This difference in activity between 4e and 4d may be attributed to both steric and electronic factors. In compound 4d, the fluorine atoms at the 2- and 6-positions create steric hindrance, potentially disrupting optimal binding within the enzyme’s active site. Moreover, strong electron-withdrawing effects from fluorine may reduce electron density on the aromatic ring and amide linkage, thereby weakening hydrogen bonding or π-π stacking interactions with key residues. Thus, both the position and electronic nature of substituents critically influence activity.
For h-NPPI, the tested series of compounds exhibited an opposite trend in activity compared to the h-e5′NT enzyme, with the exception of compound 4a, which showed good potency against both enzymes. Notably, the 4-iodo aromatic moiety displayed exceptional efficacy, with an IC50 value of 0.11 ± 0.02 µM. A structural comparison of the inhibitors against h-NPPI and h-e5′NT revealed that electronegative atoms (F, Cl, I, and CF3) at the ortho or para positions of the phenyl ring, either as single or double substitutions, significantly enhanced inhibitory potential against h-NPPI. The IC50 values for these inhibitors ranged from 0.1 to 14 µM, highlighting the importance of both substitution patterns and electronic effects in optimizing enzyme inhibition.
All the compounds (4a–4l) displayed good to moderate inhibitory activities against h-NTPDases, with IC50 values ranging from 0.28 ± 0.03 µM to 6.39 ± 1.07 µM. However, compounds 4a, 4e, 4i, 4j, 4k, and 4l, which include halogen-substituted derivatives and anilines with electron-donating groups, showed poor activity against all four enzymes. Compound 4c, featuring a 4-methylbenzyl substitution, was active across all the enzyme targets, while compound 4d, with 2,6-difluoro substitutions, demonstrated the highest potency specifically against h-NTPDase1 and -2.
In contrast, compound 4e, containing fluoro groups at the 2- and 4-positions, was inactive against all four isoenzymes. This inactivity could be attributed to steric hindrance or an unfavorable electronic interaction at the active site.
Compound 4b, with a 4-fluorobenzyl moiety, exhibited good potency against h-NTPDase1 and -3 and was the most potent inhibitor of h-NTPDase8. Against h-NTPDase3, compounds 4b, 4c, 4d, 4f, 4g, and 4h showed activity, with compound 4g, a pyridyl-substituted derivative, emerging as the most potent inhibitor. Compound 4i, which contains a methoxy substitution on the phenyl ring, showed less than 50% inhibition for all enzymes except h-e5′NT, where it demonstrated good inhibitory activity. Notably, compound 4d emerged as a broad-spectrum inhibitor, demonstrating significant activity against both h-NTPDase1 (IC50 = 0.28 ± 0.03 µM) and h-NTPDase2 (IC50 = 0.92 ± 0.17 µM), thereby highlighting its potential as a multi-target ectonucleotidase modulator. Most potent compounds among quinoline-6-carboxylic acid derivatives based on IC50 measurements are illustrated in Fig. 4.

Fluorescence characterization of the druggable compound 4d and its cellular interaction
Fluorescence spectroscopy of compound 4d revealed distinct excitation-dependent emission profiles between 410–500 nm. The emission spectra displayed strong, well-defined fluorescence peaks across the visible region (400–500 nm), demonstrating excellent photostability and consistent emissive behavior under near-UV to blue light irradiation (Fig. 5A). After cellular interaction, the red shift in the fluorescence spectral profiles likely indicated changes in its local microenvironment. This shift possibly resulted from binding with membrane-associated proteins or polar molecules present in the lipid domains, both of which alter the compound’s electronic environment60. This robust optical performance, characterized by minimal photobleaching, positions 4d as a potential fluorophore for bioimaging applications. Fluorescence microscopy of A549 lung cancer cells treated with 4d provided detailed spatial information about its cellular distribution. Bright-field images showed preserved cell morphology with intact membranes and no signs of cytotoxicity at the tested concentration (10 μM, 24 h incubation). In the red fluorescence channel, intense fluorescence was observed predominantly at the plasma membrane, with some vesicular staining suggesting possible endocytic uptake. Spectral shifts in emission suggest binding interactions between the compound d and the enzyme. The high signal-to-noise ratio (SNR > 15:1) and clear membrane integrity indicated viable colocalization. The merged image in Fig. 5B highlights this exclusive membrane targeting, with the red fluorescence of 4d forming a continuous rim around cells while remaining distinctly separated from cytoplasmic and nuclear compartments. Microscopic analysis further confirmed compound uptake, intracellular localization, and biological relevance, integrating in vitro findings and cellular behavior.

Molecular docking studies for validation

h-ENTPDase1
Compound 4d showed excellent inhibition of h-NTPDase1 (IC50 = 0.28 µM), which is supported by its favorable docking score (− 7.8 kcal/mol). Amino acid components involved in the binding interactions of h-NTPDase1 are: Asp213, Tyr450, Gly178, Ser361, Phe360, and Cys502. Molecular docking study of the compound 4d with h-NTPDase1 revealed conventional hydrogen bonding between oxygen of the carbonyl group and the amino acid Trp450. Other interaction shown by the compound are π-sigma interaction of Phe360 with phenyl ring. Cys502 refers to the amino acid Cysteine (CYS), whereby, cysteine contains a thiol (-SH) group that interacts with benzene ring and show π-sigma interaction with the aromatic (benzene) ring. Another amino acid, Gly178, shows that carbon-hydrogen bond interaction with NH group can act as a hydrogen donor in hydrogen bonding. Asp213 with its carboxyl side chain can act as a hydrogen bond acceptor. Asp213 shows hydrogen bond interaction with NH group. Gly178 also shows halogen interaction with the fluorine group.These molecular docking findings are consistent with the SAR analysis, indicating that the ortho-fluoro substituents in compound 4d facilitate optimal binding orientation and enhance interactions within the active site. The steric fit and electron-withdrawing nature of the 2,6-difluorophenyl moiety contribute significantly to the compound’s strong binding affinity (− 7.8 kcal/mol), supporting its high inhibitory potency (Fig. 6A).

h-ENTPDase2
Compound 4d exhibited strong inhibitory activity against h-NTPDase2, with an IC50 value of 0.92 µM. This potency is further supported by its favorable binding affinity (− 8.5 kcal/mol), indicating stable and effective interaction within the enzyme’s active site. Amino acid components engaged in the binding interactions of h-NTPDase2 are: Asp201, Leu202, Tyr350, His50, Ala347, Arg392, Gly435, and Cys502. Compound 4d, the most active compound against h-NTPDase2 was docked using a homology model to analyze the interactions. Molecular docking of the compound 4d with h-NTPDase2 revealed conventional hydrogen bonding of the oxygen group with Ala347, and Arg392 forms conventional hydrogen bonding with the parent quinoline moiety when the guanidinium groups NH or NH2 act as a hydrogen donor, and the quinoline nitrogen act as the acceptor. His(50) and Tyr350 show π-π stacked and π-π T-shaped with benzene ring. Another amino acid Cys502 shows π-alkyl interaction with the aromatic ring. Another amino acid Gly435 shows carbon-hydrogen bond interaction with carbon atom attached to the fluorine group. Leu202 and Asp201 show halogen bond interaction with the fluorine group (Fig. 6B).

h-ENTPDase3
Compound 4g demonstrated strong inhibitory activity against h-NTPDase3, with an IC50 value of 0.32 µM. This high potency is further supported by its favorable binding energy (− 7.3 kcal/mol), indicating a stable and efficient interaction within the active site of the enzyme. Amino acid components engaged in the binding interactions of h-NTPDase3 are: Leu144, Leu145, Gln148, Gly141, Ser100, and Glu266. Molecular docking of the compound 4g with h-NTPDase3 revealed conventional hydrogen bonding of hydrogen group with Gln148, whereas, parent moiety quinoline ring was observed to have π-alkyl interaction with Leu144 and Leu145. Furthermore, phenyl ring formed π- anion interaction with Glu266 and the nitrogen atom of quinoline ring show carbon-hydrogen interaction with Gly141 also. Ser100 amino acid show carbon hydrogen bond interaction with π-electronic cloud of the quinoline ring. In line with SAR observations, the pyridin-2-yl group in compound 4g enhances binding orientation and electronic complementarity within the h-NTPDase3 active site. The nitrogen atom in the pyridine ring likely facilitates key polar interactions, contributing to its strong binding affinity (− 7.3 kcal/mol) and high inhibitory potency (IC50 = 0.32 µM) (Fig. 6C).

h-ENTPDase8
Compound 4b demonstrated potent inhibition against h-NTPDase8, achieving an IC50 of 0.44 µM. The compound also exhibited a strong binding affinity (− 9.0 kcal/mol), suggesting a stable and efficient interaction with the enzyme’s active site. Amino acid components involved in the binding interactions of h-NTPDase8 are: Tyr402, Tyr357, Trp398, Pro393, and His360. Molecular docking of the compound 4b with h-NTPDase8 revealed conventional hydrogen bonding of oxygen group with Tyr402, and phenyl ring formed π-π T-shaped interaction with Tyr357 and His360. Similarly, Phenyl ring formed π-alkyl interaction with Pro393, and the quinoline moiety formed a π-π T-shaped interaction with Trp398 amino acid. The SAR analysis of compound 4b suggests that the 4-fluorobenzyl substituent enhances its binding interactions and electronic compatibility within the h-NTPDase8 active site. The electron-withdrawing fluorine atom likely contributes to improved stabilization, consistent with its strong binding affinity (− 9.0 kcal/mol) and notable inhibitory potency (IC50 = 0.44 µM) (Fig. 6D).

h-ENPP1
Compound 4k exhibited remarkable inhibitory activity against NPP1, with an IC50 value of 0.11 µM. Its strong binding energy (− 9.9 kcal/mol) indicates a highly stable ligand–enzyme complex, highlighting its potential as a promising NPP1 inhibitor. Amino acid sites contributing to the binding interactions of h-NPP1 are: Tyr371, Tyr340, Lys295, Phe257, His380, and Pro323. Molecular docking of the compound 4k with h-NPP1 revealed conventional hydrogen bonding between the nitrogen of the quinoline ring and the amino acid lys295. Tyr340 and phe257 are both aromatic amino acids. They show π-π stacked interactions with ligands containing aromatic moiety such as the quinoline group. Similarly, Tyr371 can engage in π-π T-shaped interactions with quinoline ring due to perpendicular orientation of the aromatic ring. The SAR analysis is well supported by the docking results for compound 4k, where the 4-iodophenyl moiety contributes to enhanced binding interactions within the h-NPP1 active site. The large size and polarizability of the iodine atom likely promote strong hydrophobic and halogen bonding interactions, resulting in a highly stable complex (− 9.9 kcal/mol) and potent inhibition (IC50 = 0.11 µM) (Fig. 6E).

h-e5′NT
With an IC50 value of 0.092 µM, compound 4a emerged as a highly active inhibitor of ecto-5′-nucleotidase. Its molecular docking score (− 8.7 kcal/mol) reflects a strong and favorable interaction with the enzyme’s binding pocket. Amino acid residues implicated in the bonding interaction of h-e5′NT are Ile91, Trp408, Gln523, Thr90, Lys50, Ala412, and Ser49. Molecular docking of the compound 4a with h-e5′NT revealed conventional hydrogen bonding between Gln523 and the oxygen atom of the carbonyl group. Similarly, Thr90 shows π-sigma interaction with the quinoline moiety. The quinoline ring interacts with the sigma bond of atom like oxygen threonine’s hydroxyl group. While Ile90 and Ala412 amino acids interact with the benzene ring through π-alkyl interaction, where the hydrophobic alkyl side chains of Isoleucine (isopropyl group) and Alanine (methyl group) align with and interact with the π-electrons of the aromatic benzene ring and Lys50 show π-alkyl interaction with quinoline moiety. Also, π-π T-shaped interaction occurs between Trp408 and benzene ring. Additionally, amide-π stacked interaction occurs between Ser49 and the quinoline moiety where the amide group from Ser49 can stack with quinoline aromatic ring via the NH and carbonyl group, stabilizing the interaction. The SAR findings for compound 4a suggest that the 4-chlorobenzyl group plays a crucial role in enhancing its binding orientation and hydrophobic alignment within the ecto-5′-nucleotidase active site. The electron-withdrawing chlorine atom, along with the flexible benzyl linker, facilitates effective positioning and stabilizing interactions, which align with its strong docking score (− 8.7 kcal/mol) and excellent inhibitory activity (IC50 = 0.092 µM) (Fig. 6F).

In-silico pharmacokinetics studies

Prediction of pharmacokinetics properties and drug-likeness of the quinoline-6-carboxylic acid derivatives (4a-4l)
Pharmacokinetics properties and drug-likeness of the quinoline-6-carboxylic acid derivatives were calculated using SwissADME. The pharmacokinetics properties were determined based on gastrointestinal (GI) absorption, permeation to the blood–brain barrier (BBB), and ability to act as a P-glycoprotein (P-gp) substrate. The drug-likeness of these compounds was determined based on Lipinski’s rule of five numbers of rotatable bonds and topological polar surface area (TPSA). The calculated parameters related to pharmacokinetics properties and drug-likeness are presented in Tables 3 and 4, respectively.
All compounds (4a–4l) exhibited high gastrointestinal (GI) absorption and BBB permeability, with no predicted interaction as P-gp substrates. TPSA values were low (41.99–60.45 Å2), supporting high absorption and brain access. All compounds followed Lipinski’s rule with no violations, showed low HBD (1) and moderate HBA (2–6), and MlogP values (1.89–3.84) within the optimal range for drug-likeness. The number of rotatable bonds (3–6) indicates suitable molecular flexibility. These values collectively support the compounds’ potential as orally active, CNS-accessible drug candidates.
The results indicated that the synthesized compounds met all the criteria. SwissADME employs a visual method known as the boiled egg plot, which graphically represents GI absorption and BBB penetration. The analysis revealed that all the synthesized compounds are in the yolk region, except for 4f (Fig. 7). All compounds except 4f exhibit efficient blood–brain barrier permeability and may be considered for further investigation as therapeutic agents against brain tumors.

Density functional theory
In this study, density functional theory (DFT) calculations were employed to analyze the HOMO–LUMO characteristics of quinoline derivatives 4a–4l. The energy gap between the frontier orbitals provided valuable insights into the compounds’ electronic properties, including their chemical reactivity, molecular stability, and optical behavior. The optimized geometries of these derivatives are depicted in Fig. 8.
Table 5 summarizes the calculated physicochemical parameters, including optimized energies, dipole moments, polarizability, frontier molecular orbital (FMO) energies (EHOMO and ELUMO), and the corresponding HOMO–LUMO gaps (ΔE). A higher ΔE value reflects a greater tendency to donate electrons, while a lower ΔE indicates an enhanced ability to accept electrons. The HOMO–LUMO gap plays a critical role in determining the chemical reactivity and kinetic stability of molecules. Compounds with larger ΔE values are generally more kinetically stable, less polarizable, and chemically inert, categorizing them as hard species. Conversely, a smaller ΔE implies greater chemical reactivity, increased polarizability, and lower kinetic stability, indicating a soft molecular nature.

Global chemical reactivity descriptors
The global reactivity parameters for the selected compounds (4a–4l) were calculated from their HOMO and LUMO energy values using standard formulas (Fig. 9). These parameters, which provide insights into the chemical stability and reactivity of the molecules, are shown in Table 6.
All the studied compounds exhibited moderate HOMO–LUMO energy gaps. Notably, derivatives with relatively moderate ΔE values, such as 4d (4.51 eV) and 4g (4.47 eV), also showed strong inhibitory activity against h-NTPDase1 and h-NTPDase3, respectively. For instance, compound 4d demonstrated dual potency against h-NTPDase1 and -2, suggesting that enhanced electronic polarizability, as reflected by a lower HOMO–LUMO gap, may promote favorable interactions with polar or charged residues within the enzyme active site, thereby improving binding affinity. Similarly, compounds 4a, 4b, and 4k, which exhibited moderate HOMO–LUMO gaps, also showed promising inhibitory activity. Compound 4a displayed potent inhibition of e5′NT (IC₅₀ = 0.092 ± 0.02 μM), compound 4b inhibited h-NTPDase8 (IC₅₀ = 0.44 ± 0.08 μM), and compound 4k significantly inhibited h-NPP1 (IC₅₀ = 0.11 ± 0.02 μM). HOMO–LUMO structures of the quinoline derivatives (4a-4l) are illustrated in Fig. 9.

Cytotoxicity assessment using MTT assay
The MTT assay was performed to observe the cytotoxic effects of various compounds (4a–4l) on the MCF-7 cells, compared to cisplatin (positive control) and serum-free media (negative control). Among the test compounds, 4k and 4j exhibited the highest cytotoxicity (56.9% and 55.0%, respectively), with corresponding viability values around 43–45%, suggesting strong anti-proliferative effects. In contrast, compound 4e showed the lowest cytotoxicity (14.6%) and highest cell viability (85.4%), indicating minimal toxicity. Notably, cisplatin displayed very high cytotoxicity (86.29%), validating the assay, while the negative control showed 0% cytotoxicity, confirming the specificity of the assay. Some compounds like 4g, 4h, and 4d also showed substantial cytotoxicity (around 47–49%), suggesting potential for further investigation (Table 7). Overall, compounds 4j and 4k appear the most promising for anti-cancer activity, though additional validation (e.g., apoptosis assays). Graphical Representation of the % Cytotoxicity and % Viability of the Studied Compounds were shown in Fig. 10.

Sulforhodamine B (SRB) proliferation assay
Sulforhodamine B (SRB) Proliferation Assay was conducted at 24, 48, and 72 h to verify the cytotoxic activity of the quinoline-6-carboxylic acid derivatives observed in the MTT assay. A time-dependent decline in cell viability was noted for all compounds. Consistent with the MTT cell viability dropping below approximately 50% after 72 h. In contrast, compound 4a exhibited weak cytotoxicity in the MTT assay (37% inhibition), it showed a stronger antiproliferative effect in the SRB assay, particularly after 72 h of exposure. This difference arises because the SRB assay measures long-term inhibition of cell growth, while MTT primarily reflects short-term metabolic activity. Thus, the enhanced SRB response suggests that 4a may act through a delayed cytostatic mechanism rather than immediate cytotoxicity Fig. 11.

Conclusion
In this study, a series of quinoline-6-carboxylic acid derivatives was successfully synthesized and evaluated for their inhibitory potential against key ectonucleotidases, including h-NTPDase isozymes, h-NPP1, and h-e5′NT. The compounds were obtained in good yields (60–70%) and accommodated a diverse range of substituents (F, Cl, I, CH₃, CF₃, OCH₃). Structural characterization was confirmed by NMR spectroscopy. Biological screening revealed that compound 4d was the most potent dual inhibitor of h-NTPDase1 and h-NTPDase2 (IC50 = 0.28 µM and 0.92 µM), with docking scores (− 7.8 and − 8.5 kcal/mol). The enhanced activity was attributed to a favorable balance of structural, steric, and electronic features. The 2,6-difluorophenyl moiety facilitated optimal binding orientation and engaged in stabilizing interactions (hydrogen bonding, π–π stacking, halogen contacts). Fluorescence imaging showed selective plasma membrane localization, indicating possible target engagement and suggesting its potential as a bioimaging agent. In addition, compounds 4g, 4b, 4a, and 4k showed selective inhibition of h-NTPDase3, h-NTPDase8, h-e5′NT, and h-NPP1, respectively. Molecular docking revealed that binding was stabilized by hydrogen bonds, π-anion, π-alkyl, and π–π stacking interactions. Complementary DFT calculations suggested a correlation between LUMO stabilization and inhibitory potency, consistent with observed SAR trends. The cytotoxicity profiles of selected compounds (4j, 4k, and 4d) in MCF-7 breast cancer cells demonstrated moderate anti-proliferative effects (48–52% cell viability at 100 µM), with 4j and 4k showing the highest cytotoxicity among the tested derivatives. While these values are lower than the positive control (cisplatin, 86% cytotoxicity), the results suggest a potential for further optimization. Importantly the MTT and SRB results collectively indicate that compounds 4j, 4k, 4g, and 4h possess moderate time-dependent cytotoxic effects against MCF-7 cells. Notably, compound 4a, which showed limited activity in the MTT assay, exhibited enhanced antiproliferative potential in the SRB assay, suggesting a delayed cytostatic mechanism rather than immediate cytotoxicity. Taken together, although the cytotoxic effects observed were moderate, the combined enzymatic inhibition, molecular docking, and SRB data support the role of quinoline-6-carboxylic acid derivatives for the development of ectonucleotidase inhibitors with potential applications in cancer immunotherapy and membrane-targeted diagnostics. Further studies are needed to improve their anticancer potency and selectivity.

Supplementary Information
Below is the link to the electronic supplementary material.