Novel 6-aryl-1-(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine derivatives: design, synthesis, and biological evaluation as potent tubulin polymerisation inhibitors with anticancer activity Authors: Jiake Gao, Yujing Zhang, Rui Qu, Qianqian Xu, Zhenjuan Sun, Liya Cui, Chao Wang Published: 2026 ABSTRACT AbstractMicrotubules assembled from α/β-tubulin heterodimers are critical for cell division and well-established anticancer drug targets, making tubulin polymerization inhibitors a viable route for new chemotherapeutics. Guided by structural analysis of colchicine-site binders and tubulin-ligand computational simulations, we rationally designed and synthesized a series of 6-aryl-1-(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidines as novel colchicine-binding site tubulin inhibitors. Derivative 9t displayed the strongest antiproliferative potency, with IC₅。 values of 0.065-0.096 μM across tested cancer lines. It exerted minimal toxicity to normal L929 fibroblasts, yielding a selectivity index over 300. Mechanistic assays confirmed 9t suppresses cell-free tubulin polymerization, destroys cellular microtubule architecture, induces persistent G₂/M cell cycle arrest, and activates cancer cell apoptosis. Overall, 9t serves as a promising dual-function tubulin inhibitor with both cytostatic and cytotoxic anticancer effects, meriting further preclinical investigation. Introduction Microtubules, filamentous polymers assembled from α/β-tubulin heterodimers, constitute a fundamental component of the eukaryotic cytoskeleton. Beyond providing structural support, these dynamic filaments orchestrate a range of essential cellular processes, including intracellular vesicular transport, maintenance of cell polarity, and, most critically, the faithful segregation of chromosomes during mitosis1,2. The hallmark of microtubule function—termed dynamic instability—refers to the stochastic switching between phases of polymerisation and depolymerisation, an equilibrium that is indispensable for proper cell division3,4. Pharmacological perturbation of this delicate balance triggers sustained mitotic arrest, ultimately committing tumour cells to programmed cell death5. Although microtubule-targeting agents (MTAs) such as taxanes and vinca alkaloids have achieved notable clinical success, their long-term utility is frequently compromised by the emergence of acquired resistance mechanisms and dose-limiting off-target toxicities, including peripheral neuropathy and myelosuppression6. These limitations have spurred sustained efforts to discover agents that bind to alternative sites on tubulin, particularly the colchicine-binding pocket on β-tubulin, as a strategy to circumvent the drawbacks associated with conventional MTAs7. Over the past two decades, this line of inquiry has yielded a diverse repertoire of colchicine-binding site inhibitors (CBSIs) featuring a wide array of chemical scaffolds, several of which have progressed into clinical evaluation8. Representative examples include colchicine (1) itself, combretastatin A-4 (CA-4, 2), the 4-substituted methoxybenzoyl-aryl-thiazole (SMART, 3), and ABI (4, Figure 1)9–12. Despite their potent preclinical antitumor activity, the clinical translation of many CBSIs has been hindered by suboptimal pharmacokinetic properties, inadequate metabolic stability, or dose-limiting toxicities, underscoring the ongoing need for new chemotypes with improved therapeutic indices13,14. The chemical structure of the well-known tubulin inhibitors. The SMART pharmacophore consists of three conjugated aromatic units: a 3,4,5-trimethoxyphenyl ring (A-ring), a substituted phenyl moiety (B-ring), and a five-membered heterocyclic core (C-ring), connected via a ketone linker between the B- and C-rings. Systematic structure-activity relationship (SAR) studies (including SMART analogs 5–8, Figure 1) have established that the “bent” molecular conformation is a critical determinant of potent antitubulin activity, whereas the alternative “straight” conformation correlates with markedly reduced bioactivity15–21. Density functional theory (DFT) calculations have further corroborated this conformational requirement, demonstrating that the energy preference for the bent conformation directly parallels antiproliferative potency across a series of SMART analogs17. Beyond conformational considerations, the ketone linker in SMART derivatives is metabolically labile, being susceptible to in vivo reduction to the corresponding secondary alcohol, which results in substantial loss of biological activity18. Thus, both conformational instability and metabolic vulnerability represent significant challenges that must be addressed in the design of next-generation CBSIs. The ring-tethering strategy has emerged as a powerful and widely adopted approach to simultaneously address these two limitations. By replacing a labile or rotatable linker with a conformationally rigid heterocyclic system, this strategy locks the molecule in its bioactive conformation while simultaneously enhancing metabolic stability. This design principle has been successfully exemplified in the development of rigidified SMART analogs, wherein the ketone linker has been embedded within five-membered heteroaromatic rings, yielding CBSIs with substantially improved pharmacological profiles19–21. In this context, the pyrazolo[3,4-d]pyrimidine nucleus represents a privileged yet previously unexplored scaffold for tubulin-targeted drug discovery. Although this fused heterocyclic system has been extensively investigated in other areas of medicinal chemistry—most notably as kinase inhibitors—owing to its rigid planar topology, favourable physicochemical properties, and versatile hydrogen-bonding capacity, its potential as a tubulin polymerisation inhibitor has not, to the best of our knowledge, been reported22–24. We hypothesised that incorporation of this scaffold into the SMART pharmacophore would not only stabilise the bioactive bent conformation and protect the metabolically vulnerable carbonyl group but also introduce unique protein–ligand interactions not accessible with previously described rigidified chemotypes. Preliminary molecular docking studies suggested that the pyrazolo[3,4-d]pyrimidine core, in combination with appropriate B-ring substitution, could engage tubulin through a hydrogen bond with β-Asn349—an interaction not observed with SMART or its vinylogous analogs. Based on these considerations, we herein report the rational design, synthesis, and comprehensive biological evaluation of a focused library of 6-aryl-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine derivatives as novel tubulin polymerisation inhibitors (Figure 2). Through systematic SAR investigations, compound 9t, featuring an indole-based B-ring, was identified as a highly potent lead with nanomolar antiproliferative activity across multiple cancer cell lines. Mechanistic studies revealed that 9t potently inhibits tubulin polymerisation, disrupts the intracellular microtubule network, induces G2/M phase cell cycle arrest, and triggers apoptotic cell death. Notably, 9t exhibited reduced cytotoxicity towards normal human cells compared with the reference agent SMART. Collectively, these findings establish the pyrazolo[3,4-d]pyrimidine scaffold as a promising new platform for the development of next-generation tubulin-targeting anticancer agents. The rational design of target compounds. Result and discussion Chemistry The synthetic route to the target 6-aryl-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine derivatives (9a-t) is outlined in Scheme 1. Commercially available pyrimidine-2,4(1H,3H)-dione (10) was subjected to bromination using molecular bromine in DMF at 121 °C to afford 5-bromopyrimidine-2,4(1H,3H)-dione (11) in 63% yield25. Chlorination of 11 with phosphorus oxychloride in the presence of N,N-diisopropylethylamine at 102 °C gave 5-bromo-2,4-dichloropyrimidine (12) in 80% yield25. Subsequent halogen-lithium exchange using i-PrMgCl·LiCl followed by trapping with dry DMF provided 2,4-dichloropyrimidine-5-carbaldehyde (13) in 35% yield26. Condensation of 13 with hydrazine hydrate in THF at 0 °C afforded 6-chloro-1H-pyrazolo[3,4-d]pyrimidine (14) in 40% yield27. Reagents and conditions (a) Br2, DMF, 121 °C, 4 h, 63%; (b) POCl3, iPr2EtN, 102 °C, 18 h, 80%; (c) i-PrMgCl.LiCl, dry DMF, THF, −78∼-42 °C, 26 h, 35%; (d) N2H4.H2O, THF, 0 °C, 1.5 h, 40%; (e) 3,4,5-trimethoxyphenylboric acid, pyridine, cupric acetate monohydrate, DCM, 80 °C, microwave (M.W), 15 min, 60%; (f) Substituted phenylboronic acid, Pd(PPh3)4, K2CO3, 1,4-dioxane/H2O, N2 atmosphere,126 °C, M.W., 25 min, 48–95%. With the core heterocycle in hand, N-arylation was accomplished under Chan-Lam type cross-coupling conditions. Thus, treatment of 14 with 3,4,5-trimethoxyphenylboronic acid in the presence of copper(II) acetate and pyridine in dichloromethane under microwave irradiation at 80 °C for 15 min delivered the key intermediate 6-chloro-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (15) in 60% yield28. The final diversification was achieved via palladium-catalyzed Suzuki-Miyaura cross-coupling. Intermediate 15 was coupled with a panel of commercially available or synthetically accessible arylboronic acids using Pd(PPh3)4 as the catalyst and K2CO3 as the base in a 1,4-dioxane/water mixture under microwave irradiation at 126 °C for 25 min29. This versatile transformation smoothly produced the target compounds 9a-t in moderate to excellent yields (48–95%). The structures of all final compounds were unambiguously confirmed by 1H NMR1,³C NMR, and HRMS analyses. Biological evaluation In vitro anti-proliferative activity The anti-proliferative activity of the newly designed and synthesised 6-aryl-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine derivatives (9a-t) was systematically evaluated against three human cancer cell lines: HeLa (cervical carcinoma), MCF-7 (breast adenocarcinoma), and A549 (non-small cell lung carcinoma). The MTT assay was employed following 72 h of continuous compound exposure, with the well-established CBSI SMART serving as the positive control. The half-maximal inhibitory concentration (IC50) values for all target compounds and the reference agent are summarised in Table 1. Table 1.In vitro anticancer activity (IC50 in μMa) of the compounds (9a-t).Compounds(IC50 ± SD, μM)HeLaMCF-7A549 9a 0.090 ± 0.0720.65 ± 0.590.23 ± 0.16 9b 3.14 ± 1.451.70 ± 1.722.62 ± 2.13 9c 11.8 ± 2.5326.5 ± 7.0716.3 ± 8.26 9d 2.27 ± 1.350.68 ± 0.411.77 ± 0.93 9e 14.1 ± 6.2919.5 ± 9.2820.3 ± 10.8 9f 23.9 ± 5.9828.6 ± 8.1325.3 ± 7.29 9g 0.28 ± 0.221.45 ± 0.676.16 ± 0.94 9h 26.7 ± 9.88>30>30 9i 22.8 ± 3.38>30>30 9j 7.18 ± 2.934.23 ± 2.318.72 ± 2.95 9k 3.54 ± 2.694.82 ± 4.626.48 ± 2.12 9l 9.55 ± 3.371.45 ± 0.677.90 ± 1.30 9m 2.36 ± 0.653.53 ± 0.885.36 ± 1.31 9n 0.41 ± 0.283.64 ± 1.831.21 ± 0.78 9o 0.13 ± 0.0701.52 ± 1.100.94 ± 0.63 9p >306.05 ± 3.04>30 9q 0.10 ± 0.0230.35 ± 0.330.67 ± 0.19 9r >30>30>30 9s 13.09 ± 1.88>30>30 9t 0.065 ± 0.012 0.083 ± 0.019 0.096 ± 0.022 SMART b 0.046 ± 0.0330.063 ± 0.0250.076 ± 0.037 a IC50: the half maximal inhibitory concentration. b Used as positive controls. In vitro anticancer activity (IC50 in μMa) of the compounds (9a-t). IC50: the half maximal inhibitory concentration. Used as positive controls. As shown in Table 1, the anti-proliferative potency varied considerably depending on the nature and substitution pattern of the B-ring introduced via Suzuki–Miyaura cross-coupling. Among the twenty analogs, compound 9t, which incorporates an indole moiety as the B-ring, exhibited the most potent activity across all three cancer cell lines, with IC50 values of 0.065 ± 0.012 μM (HeLa), 0.083 ± 0.019 μM (MCF-7), and 0.096 ± 0.022 μM (A549). These values are remarkably comparable to those of SMART (IC50 = 0.046–0.076 μM), underscoring the potential of the pyrazolo[3,4-d]pyrimidine scaffold as a bioisosteric replacement for the core heterocycle in classical SMART analogs. A detailed structure-activity relationship (SAR) analysis was conducted to delineate the key structural determinants governing antiproliferative activity. For monosubstituted phenyl B-rings, the position and electronic nature of the substituent played a critical role. Compounds bearing para-methyl (9d) or para-methoxy (9 g) groups retained moderate to good activity, with IC50 values ranging from 0.68 to 2.27 μM across the three cell lines. In contrast, the ortho-methyl analog (9b) and meta-methyl analog (9c) exhibited significantly reduced potency, particularly in HeLa and A549 cells (IC50 up to 11.8 μM), suggesting that para-substitution is preferred for optimal interaction with the tubulin colchicine-binding pocket. The introduction of electron-withdrawing groups, such as fluoro (9j), chloro (9k), nitro (9 l), or cyano (9 m), generally led to decreased antiproliferative activity compared to their electron-donating counterparts, although the fluoro-substituted derivative 9j still showed moderate activity (IC50 = 7.18 μM in HeLa). Notably, the 3,4-dimethoxyphenyl derivative 9o displayed promising activity in HeLa cells (IC50 = 0.13 ± 0.070 μM), second only to 9t within the phenyl-substituted series, indicating that a 3,4-dioxy substitution pattern may also be favourable. Further comparison of isomeric pairs revealed consistent trends. The para-substituted derivatives (9d, 9 g, 9j, 9k, 9 l, 9 m) consistently outperformed their ortho- and meta-substituted counterparts, confirming that the para position provides optimal spatial orientation for engaging key residues within the colchicine-binding pocket. Among the electron-donating substituents, the methoxy group (9 g) conferred superior activity compared to the methyl group (9d), suggesting that enhanced hydrogen-bonding capacity at the para position improves target engagement. Among the electron-withdrawing substituents, the fluoro derivative (9j) exhibited the best activity, likely due to its favourable balance of lipophilicity and electronegativity, while the strongly withdrawing nitro group (9 l) and the hydrogen-bond-accepting cyano group (9 m) significantly reduced potency. SAR analysis of disubstituted phenyl B-rings further supported these conclusions. The 3,4-dimethylphenyl derivative (9n) showed moderate activity (IC50 = 0.41 μM in HeLa), while the 3,4-dimethoxyphenyl derivative (9o) displayed markedly enhanced potency (IC50 = 0.13 μM in HeLa), indicating that electron-donating dioxy substitution at the 3,4-positions is more favourable than dimethyl substitution. This observation suggests that both electronic effects and hydrogen-bonding capacity at the B-ring contribute to antiproliferative activity. The 4-hydroxyphenyl derivative (9i) and 4-ethoxyphenyl derivative (9h) exhibited poor activity (IC50 > 22 μM in most cases), indicating that free hydroxyl or longer alkoxy groups at the para position are detrimental to activity, likely due to unfavourable steric or electronic effects within the confined binding pocket. Expanding the B-ring to bicyclic aromatic systems yielded mixed outcomes. The naphthyl derivative 9p exhibited poor activity (IC50 > 30 μM in HeLa and A549), likely due to steric hindrance within the confined colchicine-binding site. Among heteroaromatic B-rings, the thiophene analog 9q showed modest activity (IC50 = 0.10–0.67 μM), whereas pyridyl derivatives 9r and 9s were essentially inactive (IC50 > 30 μM in most cases), highlighting the critical role of hydrogen-bonding capacity and ring electronics. The exceptional activity of the indole-containing 9t can be rationalised by its ability to form an additional hydrogen bond with β-Asn349, as suggested by molecular docking studies), a feature that distinguishes it from other B-ring substitutions. To evaluate the selectivity profile of the most promising compound 9t, its cytotoxicity was further assessed against normal fibroblasts L929 cell line under identical experimental conditions. As compiled in Table 2, 9t exhibited an IC50 value greater than 30 μM against L929 cells, whereas SMART showed an IC50 of 16.3 μM. Notably, the IC50 values of 9t against the three tested cancer cell lines (HeLa, MCF-7, and A549) ranged from 0.065 to 0.096 μM, resulting in selectivity indices (IC50 in L929/IC50 in cancer cells) exceeding 300-fold. This markedly higher selectivity contrasts favourably with SMART, which displayed narrower differential cytotoxicity under the same experimental conditions. These preliminary data suggest that 9t possesses a potentially broader therapeutic window than the reference agent. Nevertheless, it should be noted that these selectivity estimates are derived from a single normal cell line and a single time point; comprehensive evaluation of the therapeutic index will require testing against a broader panel of normal cell types, additional exposure durations, and ultimately, in vivo toxicological studies. Table 2.In vitro anticancer activity and cytotoxicity test of 9t and SMART.Compounds(IC50, μM)aL929 9t >30 SMART 16.3 In vitro anticancer activity and cytotoxicity test of 9t and SMART. In summary, the systematic SAR investigation of this novel pyrazolo[3,4-d]pyrimidine-based library has led to the identification of compound 9t as a highly potent and selective antiproliferative agent. The marked enhancement in activity conferred by the indole B-ring, combined with the favourable selectivity profile, validates the pyrazolo[3,4-d]pyrimidine scaffold as a promising new chemotype for the development of tubulin polymerisation inhibitors. These findings provided a strong rationale for further mechanistic characterisation of 9t, as described in the following sections. Effect on tubulin polymerisation A tubulin polymerisation assay was conducted in vitro following a previously established protocol28 to determine whether the candidate compound 9t affects microtubule dynamics. SMART, a well-recognised microtubule-destabilizing reference, and paclitaxel (PTX), a known microtubule stabiliser, served as positive and negative controls, respectively. The results, summarised in Figure 3, showed that PTX dramatically promoted microtubule assembly, leading to a substantial increase in the polymerisation rate relative to the vehicle control. In contrast, both 9t and SMART, tested at the same concentration of 1 μM, effectively suppressed tubulin polymerisation kinetics. The observed anti-polymerisation activity was concentration-dependent, with a stronger inhibitory effect detected at the higher dose (3 μM) compared to the lower one (1 μM). Effects of compound 9t and control compounds on tubulin polymerisation. Taken together, these findings indicate that 9t acts as a direct and potent microtubule-destabilizing agent. Furthermore, the pyrazolo[3,4-d]pyrimidine scaffold emerges as a promising structural template for the development of novel small-molecule tubulin polymerisation inhibitors, warranting further structural optimisation and in-depth mechanistic studies. Analysis of immunofluorescence staining The effect of compound 9t on the cellular microtubule network was evaluated in HeLa cells by immunofluorescence staining following a previously described protocol29. As shown in Figure 4, untreated control cells retained an intact and well-organised filamentous microtubule array, characteristic of normal cytoskeletal architecture. In contrast, exposure to 9t at 65 nM resulted in pronounced disruption of the microtubule system. Under these conditions, the typically elongated cytoplasmic microtubules underwent substantial shortening and collapsed into dense perinuclear clusters encircling the nucleus—a morphological hallmark indicative of active microtubule depolymerisation. Immunofluorescence analysis showing the effects of compound 9t (65 nM) and SMART (65 nM) on the cellular microtubule network and microtubule reorganisation. Notably, the cytoskeletal alterations induced by 9t were highly comparable to those elicited by the reference agent SMART (65 nM), a well-established ligand targeting the colchicine-binding pocket of tubulin. This striking phenotypic consistency strongly suggests that 9t exerts its effects through a microtubule-destabilizing mechanism. Thus, the immunofluorescence data provide direct cellular-level evidence that 9t acts as a potent microtubule-destabilizing agent. Furthermore, consistent with the biochemical findings from the microtubule polymerisation assay, these observations support the conclusion that 9t effectively interferes with microtubule dynamics. Collectively, these results further validate the pyrazolo[3,4-d]pyrimidine scaffold as a promising chemotype for the development of novel microtubule-targeting compounds. Cell cycle analysis Microtubule-disrupting agents typically exert their anti-mitotic effects by arresting cells at the G2/M checkpoint, a well-established strategy for suppressing tumour growth30. To characterise the anti-mitotic properties of the newly identified compound 9t, its influence on the cell cycle profile of HeLa cervical carcinoma cells was examined following 24 h of exposure. Flow cytometric quantification revealed that 9t produced a concentration-dependent accumulation of cells in the G2/M phase over a concentration range of 33–130 nM (Figure 5). At the maximum dose tested (130 nM), the proportion of cells in G2/M rose from a baseline value of 4.9% to 50.0%. This marked cell cycle redistribution indicates that 9t effectively blocks mitotic progression, a pattern consistent with that of established microtubule-targeting drugs. The ability of 9t to induce substantial G2/M arrest aligns with its previously observed inhibition of tubulin polymerisation, supporting a unified mechanism in which direct interference with microtubule assembly leads to cell cycle blockade. Collectively, these results not only underscore the anti-proliferative potential of 9t but also highlight the pyrazolo[3,4-d]pyrimidine core as a valuable scaffold for the development of novel anti-mitotic agents. Effects of compound 9t on the cell cycle distribution in HeLa cells. Induction of cell apoptosis Tubulin-targeting agents have emerged as a promising class of anticancer drugs, largely owing to their capacity to provoke programmed cell death in tumour cells30. To assess the pro-apoptotic potential of the novel tubulin inhibitor candidate 9t, HeLa cervical carcinoma cells were exposed to increasing concentrations of the compound for 48 h. Apoptotic cell death was subsequently quantified using Annexin V-FITC/propidium iodide dual staining in conjunction with flow cytometric analysis. The results, presented in Figure 6, demonstrate that 9t induced a marked and concentration-dependent rise in apoptosis. At a concentration of 33 nM, 9t produced moderate apoptotic effects, with 13.1% of cells undergoing programmed cell death. When the dose was increased to 65 nM and 130 nM, the apoptotic rates rose substantially to 23.9% and 38.0%, respectively. This dose-responsive elevation in apoptosis highlights the potent pro-apoptotic activity of 9t in HeLa cells. Notably, the degree of apoptosis observed at the highest concentration is comparable to that reported for several clinically relevant tubulin inhibitors, further emphasising the therapeutic potential of this compound. The ability of 9t to efficiently trigger programmed cell death is consistent with the well-established mechanism of action of microtubule-destabilizing agents and provides additional evidence supporting its continued development as a promising lead compound for anticancer therapy. Analysis of apoptosis induction by compound 9t in HeLa human cervical carcinoma cells. Inhibit cell migration To evaluate the effect of compound 9t on cancer cell migration, a wound-healing assay was performed in HeLa cells. As shown in Figure 7, confluent cell monolayers were scratched to create a uniform wound area, and the cells were then treated with vehicle control, SMART (65 nM), or compound 9t (65 nM) for 24 h. At 0 h, the wound gaps in all groups were comparable, confirming the consistency of the initial wounding. After 24 h of incubation, cells in the control group migrated extensively into the denuded area, resulting in a marked reduction of the wound gap, indicating active cell migration under normal conditions. In contrast, treatment with SMART at 65 nM significantly inhibited cell migration, as evidenced by a wider remaining wound gap compared to the control group. Notably, compound 9t at the same concentration (65 nM) exhibited a more pronounced inhibitory effect on cell migration, with the wound gap remaining substantially wider than that observed in both the control and SMART-treated groups. These results demonstrate that compound 9t effectively suppresses the migratory capacity of HeLa cells, with an inhibitory effect superior to that of the reference agent SMART at the same concentration. The potent anti-migratory activity of 9t further supports its potential as a promising anticancer agent, as inhibition of tumour cell migration is a critical attribute for preventing cancer metastasis. Effects of compound 9t and SMART on HeLa cell migration in a wound-healing assay. Representative images of wound gaps at 0 h and 24 h after treatment with vehicle control, SMART, or compound 9t. Molecular modelling study To gain structural insight into the remarkable biological activity of the most potent congener 9t, molecular docking studies were performed using the tubulin in complex with a colchicine-site ligand (PDB ID: 5LYJ). The binding modes of 9a (a representative moderately active analog) and 9t were compared with that of the reference compound SMART to identify key intermolecular interactions governing potency. As depicted in Figure 8, superimposition of the docked poses revealed that SMART (pink), 9a (green), and 9t (yellow) share an overlapping binding region within the colchicine-binding pocket, suggesting a conserved orientation for this scaffold family. The central pyrazolo[3,4-d]pyrimidine core of both 9a and 9t participates in a cation-π interaction with the ε-ammonium group of β-Lys352, a residue known to play a critical role in ligand recognition at this site. More importantly, 9t establishes an additional hydrogen bond between the indole N-H moiety and the carbonyl oxygen of β-Asn349—a polar contact absent in the docking models of SMART and 9a. This supplementary interaction is likely responsible for the enhanced binding affinity of 9t, as evidenced by its more favourable computed docking score (−10.573) compared to SMART (−8.782) and 9a (−9.088). At the colchicine site, a comparison diagram of the fitting binding patterns of 9a and 9t with SMART. Carbon atoms are shown in pink for SMART, in green for 9a, and in yellow for 9t. The residues from the α, β-tubulin chain are shown in platinum-blond. Collectively, these docking results offer a plausible structural rationale for the superior antiproliferative activity of 9t, indicating that its improved potency stems from additional polar contacts within the colchicine-binding site. Furthermore, these findings reinforce the notion that 9t acts as a tubulin polymerisation inhibitor and highlight the utility of structure-guided design in advancing this promising chemotype for further optimisation. Molecular dynamics simulation To further validate the binding stability of compound 9t within the colchicine-binding pocket and to complement the docking analysis, a 100 ns molecular dynamics (MD) simulation was performed using the GROMACS 2025 software package. As shown in Figure 9(A), the protein backbone RMSD stabilised at 0.15–0.25 nm after the initial 10 ns equilibration period, while the ligand RMSD remained remarkably low at 0.05–0.10 nm throughout the simulation, confirming that compound 9t is tightly anchored within the binding pocket without significant positional deviation or dissociation. RMSF analysis (Figure 9B) revealed that the majority of residues exhibited low fluctuation values (0.1–0.2 nm), while residues lining the colchicine-binding pocket displayed consistently low RMSF values, indicating a rigid binding site microenvironment that provides stable interactions for the ligand. The radius of gyration (Rg) remained stable at approximately 3.0 nm throughout the simulation (Figure 9C), with no increasing or decreasing trend, confirming that the protein maintained a compact folded conformation without unfolding or abnormal expansion. The solvent-accessible surface area (SASA) fluctuated within 330–355 nm2 (Figure 9D), with a gradual decrease during the initial phase followed by stabilisation, further supporting the conformational stability of the protein. Molecular dynamics simulation results for the 9t-tubulin complex over 100 ns. (A) RMSD plots. (B) RMSF per residue. (C) Radius of gyration (Rg). (D) Solvent-accessible surface area (SASA). Collectively, the MD simulation results provide robust evidence that compound 9t forms a dynamically stable complex with tubulin at the colchicine-binding site, corroborating the docking predictions and further substantiating the structure-activity rationale for its potent antiproliferative activity. Physicochemical properties The drug-likeness of the synthesised 6-aryl-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine derivatives was evaluated through in silico prediction of key physicochemical descriptors in accordance with Lipinski’s rule of five. For the reference compound SMART and the representative derivatives 9a and 9t, the molecular weight (MW), calculated logarithm of the octanol–water partition coefficient (CLog P), number of hydrogen bond donors (HBD) and acceptors (HBA), and topological polar surface area (TPSA) were computed using the publicly accessible SwissADME platform (http://www.swissadme.ch) and ChemBioDraw Ultra 14.0 software. As summarised in Table 3, all examined compounds fully complied with Lipinski’s criteria, exhibiting appropriate molecular size, balanced lipophilicity, favourable hydrogen-bonding capacity, and acceptable polar surface characteristics. These computational predictions suggest that derivatives 9a and 9t possess physicochemical profiles compatible with satisfactory oral bioavailability and drug-like properties, thereby supporting their continued evaluation as potential orally administrable anticancer candidates. Table 3.Predicted physicochemical properties of SMART, 9a, and 9ta.CompoundsCLogPTPSANatomsMWHBAHBDStandard<5<140 <500<10<5 SMART 3.7557.1242355.4150 9a 2.9568.0145362.3970 9t 2.9480.0449401.4371 a CLogP: Calculated LogP; TPSA: topological polar surface area; Natoms: No. of atoms; MW: molecular weight; HBA: hydrogen-bond acceptor atoms. HBD: hydrogen-bond donor atoms. Predicted physicochemical properties of SMART, 9a, and 9ta. CLogP: Calculated LogP; TPSA: topological polar surface area; Natoms: No. of atoms; MW: molecular weight; HBA: hydrogen-bond acceptor atoms. HBD: hydrogen-bond donor atoms. Conclusion In this work, the pyrazolo[3,4-d]pyrimidine scaffold was explored for the first time as a novel chemotype for tubulin polymerisation inhibition. Through systematic SAR investigations, compound 9t—bearing an indole moiety as the B-ring—emerged as the most potent derivative, displaying nanomolar antiproliferative activity across a panel of human cancer cell lines (HeLa, MCF-7, and A549), with IC50 values ranging from 0.065 to 0.096 μM. Mechanistically, 9t potently suppressed tubulin assembly in cell-free polymerisation assays, disrupted the intracellular microtubule network, induced sustained G2/M phase arrest, and subsequently triggered apoptotic cell death. Molecular docking studies provided a structural rationale for these observations: the pyrazolo[3,4-d]pyrimidine core, in combination with the indole B-ring of 9t, facilitates a key hydrogen-bonding interaction with β-Asn349—a contact absent in the binding models of the reference agent SMART and its vinylogous analog 9. This additional polar interaction likely underpins the enhanced potency of 9t. Importantly, 9t exhibited markedly lower cytotoxicity against normal L929 fibroblasts (IC50 > 30 μM) compared to SMART (IC50 = 16.3 μM), yielding selectivity indices exceeding 300-fold. This favourable safety profile, together with its potent antiproliferative activity, positions 9t as a promising lead candidate for further development. While the current study provides robust in vitro evidence for the potent anticancer activity, favourable selectivity, promising resistance profile, and anti-migratory properties of compound 9t, we acknowledge that several important studies are warranted to fully evaluate its drug development potential. These include: (i) comprehensive in vitro ADME profiling, including metabolic stability and CYP450 interaction assessment; (ii) investigation of reactive oxygen species (ROS) generation to clarify whether oxidative stress contributes to its anticancer activity; (iii) preliminary pharmacokinetic studies to evaluate bioavailability and systemic exposure; and (iv) in vivo efficacy and toxicity studies in suitable tumour models to validate its therapeutic potential. These investigations are currently being planned and will be reported in due course as part of our ongoing efforts to advance this promising compound towards preclinical development. Although the rigidification strategy itself is not unprecedented, the identification of the pyrazolo[3,4-d]pyrimidine scaffold as a new platform for tubulin-targeted drug discovery expands the available structural diversity in this field. Compared to previously reported rigidified SMART bioisosteres—such as imidazoles, thiophenes, and pyridines—the pyrazolo[3,4-d]pyrimidine core offers distinct hydrogen-bonding capabilities and a unique three-dimensional topology. The exceptional activity of 9t, particularly its ability to engage β-Asn349 through a hydrogen bond, suggests that this scaffold may enable protein–ligand interactions not readily accessible with other chemotypes. While direct comparative studies are needed to definitively establish superiority, the potent in vitro profile of 9t validates the pyrazolo[3,4-d]pyrimidine scaffold as a valuable addition to the tubulin inhibitor arsenal and provides a solid foundation for future optimisation efforts. Collectively, these findings support the continued investigation of this chemotype towards the development of novel anticancer therapeutics. Experimental Chemistry Materials and methods Unless otherwise specified, all commercial reagents and solvents were used as received. Nuclear magnetic resonance (NMR) spectra (1H at 500 MHz and1³C at 126 MHz) were recorded on a Bruker AVANCE spectrometer in deuterated chloroform (CDCl3, Aladdin, Shanghai, China), with tetramethylsilane (TMS) as the internal standard. High-resolution mass spectrometry (HRMS) data were obtained using an Agilent Accurate-Mass Q-TOF 6530 mass spectrometer equipped with an electrospray ionisation (ESI) source operating in positive-ion mode. Reaction progress was monitored by thin-layer chromatography (TLC) on silica gel 60 HSGF254 plates (Energy Chemical, Anhui, China), and the spots were visualised under UV light at 254 nm. Flash column chromatography purification was performed on silica gel (200–300 mesh, Aladdin, Shanghai, China). Microwave-assisted reactions were carried out in sealed vessels using a CEM Discover single-mode microwave reactor. The purity of all final compounds 9a‑t was determined by high-performance liquid chromatography (HPLC) analysis and found to be greater than 95%. The human cancer cell lines employed in this study were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China) and were maintained following the supplier’s recommended protocols. General synthetic procedure for 5-bromopyrimidine-2,4(1H,3H)-dione (11) To a stirred solution of pyrimidine-2,4(1H,3H)-dione (10, 1.0 g, 8.9 mmol) in N,N-dimethylformamide (7 ml) at 121 °C, bromine (0.50 ml, 9.8 mmol) was added dropwise. The mixture was stirred for an additional 4 h, then cooled to room temperature and triturated with dichloromethane. The resulting precipitate was collected by filtration and recrystallized from water to afford the title compound (11, 1.1 g, 63%) as a white solid. General synthetic procedure for 5-bromo-2,4-dichloropyrimidine (12) To a solution of compound 11 (0.50 g, 2.6 mmol) in phosphorus oxychloride (0.75 ml, 7.9 mmol) at 0 °C was added N,N-diisopropylethylamine (0.90 ml, 5.3 mmol) slowly. The resulting mixture was heated at 102 °C for 18 h. After cooling to room temperature, ice-water (80 ml) was added, and the aqueous layer was extracted with diethyl ether (30 ml × 3). The combined organic phases were dried over anhydrous MgSO4 and concentrated under reduced pressure. The residue was purified by flash column chromatography (ethyl acetate/hexane, 1:9) to afford the title compound 12 as a colourless oil (0.48 g, 80%). General synthetic procedure for 2,4-dichloropyrimidine-5-carbaldehyde (13) To a solution of compound 12 (0.64 g, 2.8 mmol) in dry THF (8 ml) under N2 at −78 °C was added isopropylmagnesium chloride-lithium chloride complex (1.3 M in THF, 2.2 ml, 2.9 mmol). The mixture was stirred for 2 h, then dry DMF (0.45 ml, 5.6 mmol) was added dropwise, and the temperature was allowed to rise to −42 °C over 24 h. Upon completion, 1 N HCl (8 ml) was added, and the mixture was extracted with diethyl ether (three times). The combined organic layers were dried (MgSO4), concentrated under reduced pressure, and the residue was purified by silica gel column chromatography to afford the title compound 13 as a white solid (0.17 g, 35%). General synthetic procedure for 6-chloro-1H-pyrazolo[3,4-d]pyrimidine (14) To a solution of compound 13 (0.17 g, 1.0 mmol) in THF (5 ml) at 0 °C was added 80% hydrazine hydrate (2.1 mmol). The mixture was stirred for 1.5 h. Upon completion, the mixture was filtered, and the filtrate was extracted with DCM (three times). The combined organic layers were dried (MgSO4), concentrated under reduced pressure, and the residue was purified by silica gel column chromatography to give the title compound 14 as a yellow solid (0.062 g, 40%). General synthetic procedure for 6-chloro-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (15) Compound 14 (0.046 g, 0.30 mmol), 3,4,5-trimethoxyphenylboronic acid (0.13 g, 0.60 mmol), pyridine (0.047 g, 0.60 mmol), and copper(II) acetate (0.12 g, 0.60 mmol) were combined in anhydrous 1,4-dioxane (10 ml). The mixture was irradiated under microwave conditions at 80 °C for 15 min. After cooling, the reaction was quenched with ethyl acetate (10 ml) and washed sequentially with water and brine. The organic layer was dried over anhydrous MgSO4, filtered, and concentrated under reduced pressure. Purification by silica gel column chromatography (petroleum ether/ethyl acetate = 4:1) gave the title compound 15 as a white solid (0.028 g, 60%). General synthetic procedure for 6-chloro-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (9a-t) A suspension of compound 15 (0.032 g, 0.10 mmol), the corresponding arylboronic acid (0.15 mmol), Pd(PPh3)4 (0.012 g, 0.010 mmol), and K2CO3 (0.041 g, 0.30 mmol) in degassed 1,4-dioxane/water was subjected to microwave irradiation at 126 °C for 25 min. After cooling, the mixture was diluted with water and extracted with ethyl acetate (3 × 8 ml). The combined organic layers were washed with brine, dried over anhydrous MgSO4, and concentrated under reduced pressure. The residue was purified by silica gel column chromatography to afford the title compounds 9a-t. 6-phenyl-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (9a) White solid; yield: 85%; 1H NMR (500 MHz, CDCl3) δ 9.33 (s, 1H), 8.64 − 8.56 (m, 2H), 8.29 (s, 1H), 7.79 (s, 2H), 7.56 − 7.51 (m, 3H), 4.04 (s, 6H), 3.94 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 162.07, 153.50, 153.26, 152.72 (2 C), 137.56, 136.42, 134.83, 133.72, 131.09, 128.68 (2 C), 128.63 (2 C), 114.35, 98.55 (2 C), 61.07, 56.28 (2 C). HRMS calcd for C20H19N4O3 [M + H]+ 363.1457, found 363.1459. 6-(o-tolyl)-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (9b) White solid; yield: 82%; 1H NMR (500 MHz, CDCl3) δ 9.36 (s, 1H), 8.32 (s, 1H), 8.02 (d, J = 7.3 Hz, 1H), 7.67 (s, 2H), 7.42 − 7.37 (m, 1H), 7.35 (t, J = 6.7 Hz, 2H), 3.97 (s, 6H), 3.91 (s, 3H), 2.71 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 164.72, 153.51, 152.80, 152.33 (2 C), 137.98, 137.54, 136.61, 134.72, 133.62, 131.49, 131.24, 129.70, 125.98, 113.58, 98.90 (2 C), 61.03, 56.33 (2 C), 21.68. HRMS calcd for C21H21N4O3 [M + H]+ 377.1614, found 377.1615. 6-(m-tolyl)-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (9c) White solid; yield: 55%; 1H NMR (500 MHz, CDCl3) δ 9.30 (s, 1H), 8.44 (s, 1H), 8.40 (d, J = 7.8 Hz, 1H), 8.28 (s, 1H), 7.82 (s, 2H), 7.42 (t, J = 7.6 Hz, 1H), 7.35 (d, J = 7.5 Hz, 1H), 4.05 (s, 6H), 3.94 (s, 3H), 2.47 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 162.12, 153.47, 153.23, 152.65 (2 C), 138.26, 137.42, 136.29, 134.92, 133.70, 131.88, 129.26, 128.62, 125.81, 114.31, 98.32 (2 C), 61.08, 56.21 (2 C), 21.51. HRMS calcd for C21H21N4O3 [M + H]+ 377.1614, found 377.1617. 6-(p-tolyl)-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (9d) White solid; yield: 90%; 1H NMR (500 MHz, CDCl3) δ 9.30 (s, 1H), 8.50 (d, J = 8.2 Hz, 2H), 8.28 (s, 1H), 7.79 (s, 2H), 7.34 (d, J = 8.0 Hz, 2H), 4.04 (s, 6H), 3.94 (s, 3H), 2.46 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 162.20, 153.48, 153.31, 152.67 (2 C), 141.51, 136.34, 134.89, 134.87, 133.72, 129.46 (2 C), 128.60 (2 C), 114.19, 98.49 (2 C), 61.07, 56.26 (2 C), 21.51. HRMS calcd for C21H21N4O3 [M + H]+ 377.1614, found 377.1614. 6-(2-methoxyphenyl)-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (9e) White solid; yield: 83%; 1H NMR (500 MHz, CDCl3) δ 9.36 (s, 1H), 8.31 (s, 1H), 7.89 (dd, J = 7.6, 1.8 Hz, 1H), 7.77 (s, 2H), 7.49 − 7.46 (m, 1H), 7.10 (ddd, J = 8.3, 5.3, 1.6 Hz, 2H), 3.97 (s, 6H), 3.90 (s, 3H), 3.90 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 162.89, 158.10, 153.48, 152.97, 152.27 (2 C), 136.45, 134.85, 133.56, 132.25, 131.33, 128.41, 120.70, 113.66, 112.21, 98.67 (2 C), 61.02, 56.27 (2 C), 56.12. HRMS calcd for C21H21N4O4 [M + H]+ 393.1563, found 393.1562. 6-(3-methoxyphenyl)-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (9f) White solid; yield: 82%; 1H NMR (500 MHz, CDCl3) δ 9.31 (s, 1H), 8.29 (s, 1H), 8.21 (d, J = 7.8 Hz, 1H), 8.19 − 8.13 (m, 1H), 7.77 (s, 2H), 7.44 (t, J = 7.9 Hz, 1H), 7.09 (dd, J = 7.8, 2.3 Hz, 1H), 4.04 (s, 6H), 3.94 (s, 3H), 3.93 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 161.87, 160.03, 153.50, 153.21, 152.63 (2 C), 139.00, 136.42, 134.82, 133.71, 129.65, 121.13, 117.23, 114.41, 113.67, 98.56 (2 C), 61.07, 56.28 (2 C), 55.42. HRMS calcd for C21H21N4O4 [M + H]+ 393.1563, found 393.1562. 6-(4-methoxyphenyl)-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (9 g) White solid; yield: 53%; 1H NMR (500 MHz, CDCl3) δ 9.27 (s, 1H), 8.56 (d, J = 8.9 Hz, 2H), 8.26 (s, 1H), 7.77 (s, 2H), 7.04 (d, J = 9.0 Hz, 2H), 4.04 (s, 6H), 3.94 (s, 3H), 3.92 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 162.21, 161.96, 153.48, 153.36, 152.64 (2 C), 136.34, 134.90, 133.75, 130.31 (2 C), 114.03 (2 C), 109.98, 98.55 (2 C), 96.31, 61.07, 56.28 (2 C), 55.44. HRMS calcd for C21H21N4O4 [M + H]+ 393.1563, found 393.1589. 6-(4-ethoxyphenyl)-1-(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (9h) White solid; yield: 50%; 1H NMR (500 MHz, CDCl3) δ 9.27 (s, 1H), 8.55 (d, J = 9.0 Hz, 2H), 8.26 (s, 1H), 7.77 (s, 2H), 7.02 (d, J = 9.0 Hz, 2H), 4.15 (q, J = 7.0 Hz, 2H), 4.04 (s, 6H), 3.94 (s, 3H), 1.48 (t, J = 7.0 Hz, 3H); 13C NMR (126 MHz, CDCl3) δ 162.00, 161.63, 153.86, 153.47, 152.61 (2 C), 134.90, 133.76, 130.30 (2 C), 129.97, 114.52 (2 C), 113.86, 98.55 (2 C), 96.31, 63.65, 61.07, 56.27 (2 C), 14.78. HRMS calcd for C22H23N4O4 [M + H]+ 407.1719, found 407.1730. 4-(1-(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidin-6-yl)phenol (9i) White solid; yield: 88%; 1H NMR (500 MHz, CDCl3) δ 9.27 (s, 1H), 8.52 (d, J = 8.8 Hz, 2H), 8.26 (s, 1H), 7.76 (s, 2H), 6.97 (d, J = 8.8 Hz, 2H), 6.72 (s, 1H), 4.03 (s, 6H), 3.94 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 161.84, 158.46, 153.48, 152.61 (2 C), 136.54, 134.85, 133.78, 130.59 (2 C), 129.90, 116.15, 115.59 (2 C), 113.91, 98.61 (2 C), 61.07, 56.27 (2 C). HRMS calcd for C20H19N4O4 [M + H]+ 379.1406, found 379.1409. 6-(4-fluorophenyl)-1-(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (9j) White solid; yield: 91%; 1H NMR (500 MHz, CDCl3) δ 9.31 (s, 1H), 8.66 − 8.57 (m, 2H), 8.30 (s, 1H), 7.73 (s, 2H), 7.21 (t, J = 8.7 Hz, 2H), 4.03 (s, 6H), 3.94 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 164.87, 161.17, 153.51, 153.19, 152.75 (2 C), 136.54, 134.71, 133.76, 130.72 (2 C), 129.72, 115.68 (2 C), 114.24, 98.71 (2 C), 61.08, 56.28 (2 C). HRMS calcd for C20H18FN4O3 [M + H]+ 381.1363, found 381.1358. 6-(4-chlorophenyl)-1-(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (9k) White solid; yield: 81%; 1H NMR (500 MHz, CDCl3) δ 9.31 (s, 1H), 8.55 (d, J = 8.6 Hz, 2H), 8.30 (s, 1H), 7.73 (s, 2H), 7.51 (d, J = 8.6 Hz, 2H), 4.03 (s, 6H), 3.94 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 161.10, 153.52, 153.13, 152.77 (2 C), 137.37, 136.56, 136.08, 134.67, 133.77, 129.91 (2 C), 128.93 (2 C), 114.41, 98.70 (2 C), 61.08 (2 C), 56.28. HRMS calcd for C20H18ClN4O3 [M + H]+ 397.1067, found 397.1061. 6-(4-nitrophenyl)-1-(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (9 l) Yellow solid; yield: 62%; 1H NMR (500 MHz, CDCl3) δ 9.39 (s, 1H), 8.79 (d, J = 9.0 Hz, 2H), 8.39 (d, J = 9.0 Hz, 2H), 8.37 (s, 1H), 7.70 (s, 2H), 4.04 (s, 6H), 3.96 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 159.77, 153.84, 153.57, 152.96 (2 C), 152.92, 149.42, 143.34, 141.66, 133.86, 129.44 (2 C), 123.85 (2 C), 114.84, 98.93 (2 C), 61.12, 56.33 (2 C). HRMS calcd for C20H18N5O5 [M + H]+ 408.1308, found 408.1312. 4-(1-(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidin-6-yl)benzonitrile (9 m) White solid; yield: 64%; 1H NMR (500 MHz, CDCl3) δ 9.37 (s, 1H), 8.72 (d, J = 8.6 Hz, 2H), 8.36 (s, 1H), 7.83 (d, J = 8.7 Hz, 2H), 7.68 (s, 2H), 4.03 (s, 6H), 3.95 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 156.59, 153.55, 152.94, 152.92 (2 C), 150.24, 143.27, 141.67, 136.74, 133.85, 132.47, 129.06 (2 C), 119.14 (2 C), 114.80, 98.95 (2 C), 61.11, 56.32 (2 C). HRMS calcd for C21H18N5O3 [M + H]+ 388.1410, found 388.1408. 6–(3,4-dimethylphenyl)-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (9n) White solid; yield: 81%; 1H NMR (500 MHz, CDCl3) δ 9.29 (s, 1H), 8.39 (s, 1H), 8.33 (d, J = 9.5 Hz, 1H), 8.27 (s, 1H), 7.83 (s, 2H), 7.30 (d, J = 7.9 Hz, 1H), 4.05 (s, 6H), 3.94 (s, 3H), 2.38 (s, 3H), 2.37 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 162.27, 153.47, 153.31, 152.63 (2 C), 140.24, 136.87, 136.25, 135.14, 134.97, 133.71, 130.09, 129.74, 126.20, 114.16, 98.30 (2 C), 61.07, 56.22 (2 C), 19.93, 19.87. HRMS calcd for C22H23N4O3 [M + H]+ 391.1770, found 391.1774. 6-(3,4-dimethoxyphenyl)-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (9o) Light yellow solid; yield: 56%; 1H NMR (500 MHz, CDCl3) δ 9.27 (s, 1H), 8.27 (s, 1H), 8.25 (dd, J = 8.5, 2.0 Hz, 1H), 8.16 (d, J = 2.0 Hz, 1H), 7.73 (s, 2H), 7.01 (d, J = 8.5 Hz, 1H), 4.03 (s, 6H), 4.02 (s, 3H), 3.99 (s, 3H), 3.93 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 161.91, 153.47, 153.34, 152.55 (2 C), 151.80, 150.15, 136.42, 134.84, 133.81, 122.20, 113.95, 112.34, 111.39, 110.92, 98.74 (2 C), 61.08, 56.54, 56.29 (2 C), 55.81. HRMS calcd for C22H23N4O5 [M + H]+ 423.1668, found 423.1665. 6-(Naphthalen-2-yl)-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (9p) White solid; yield: 79%; 1H NMR (500 MHz, CDCl3) δ 9.35 (s, 1H), 9.15 (s, 1H), 8.67 (dd, J = 8.6, 1.6 Hz, 1H), 8.30 (s, 1H), 7.97 (d, J = 8.2 Hz, 2H), 7.91 (d, J = 7.2 Hz, 1H), 7.84 (s, 2H), 7.58 − 7.53 (m, 2H), 4.08 (s, 6H), 3.96 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 162.03, 153.52, 153.29, 152.74 (2 C), 136.39, 134.89, 134.84, 134.82, 133.77, 133.28, 129.35, 129.15, 128.32, 127.76, 127.45, 126.51, 125.13, 114.34, 98.48 (2 C), 61.10, 56.28 (2 C). HRMS calcd for C24H21N4O3 [M + H]+ 413.1614, found 413.1615. 6-(Thiophen-3-yl)-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (9q) Light yellow solid; yield: 83%; 1H NMR (500 MHz, CDCl3) δ 9.26 (s, 1H), 8.43 (dd, J = 3.1, 1.1 Hz, 1H), 8.27 (s, 1H), 8.00 (dd, J = 5.0, 1.1 Hz, 1H), 7.74 (s, 2H), 7.43 (dd, J = 5.1, 3.1 Hz, 1H), 4.03 (s, 6H), 3.93 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 159.06, 153.48, 153.07, 152.81 (2 C), 141.61, 136.43, 134.80, 133.79, 128.82, 127.39, 126.31, 113.96, 98.58 (2 C), 61.07, 56.30 (2 C). HRMS calcd for C18H17N4O3S [M + H]+ 369.1021, found 369.1021. 6-(Pyridin-3-yl)-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (9r) Light yellow solid; yield: 86%;1H NMR (500 MHz, CDCl3) δ 9.81 (s, 1H), 9.34 (s, 1H), 8.85 (d, J = 8.0 Hz, 1H), 8.76 (s, 1H), 8.33 (s, 1H), 7.92 − 7.76 (m, 1H), 7.73 (s, 2H), 4.03 (s, 6H), 3.94 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 160.14, 153.56, 152.95 (2 C), 152.91, 151.49, 150.13, 136.66, 135.98, 134.57, 133.81, 132.90, 123.59, 114.76, 98.68 (2 C), 61.07, 56.35 (2 C). HRMS calcd for C19H18N5O3 [M + H]+ 364.1410, found 364.1410. 6-(Pyridin-4-yl)-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (9s) Light solid; yield: 89%; 1H NMR (500 MHz, CDCl3) δ 9.38 (s, 1H), 8.83 (s, 2H), 8.43 (d, J = 4.9 Hz, 2H), 8.36 (s, 1H), 7.71 (s, 2H), 4.04 (s, 6H), 3.95 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 159.87, 153.58, 152.96 (2 C), 152.90, 150.49, 144.94, 138.23, 136.80, 134.45, 133.78, 132.90, 122.35, 115.19, 98.84 (2 C), 61.09, 56.33 (2 C). HRMS calcd for C19H18N5O3 [M + H]+ 364.1410, found 364.1405. 6-(1H-indol-4-yl)-1–(3,4,5-trimethoxyphenyl)-1H-pyrazolo[3,4-d]pyrimidine (9t) Yellow solid; yield: 80%; 1H NMR (500 MHz, CDCl3) δ 9.39 (s, 1H), 8.45 (dd, J = 7.6, 0.9 Hz, 2H), 8.30 (s, 1H), 7.77 (ddd, J = 3.0, 2.2, 0.8 Hz, 1H), 7.74 (s, 2H), 7.60 (d, J = 8.0 Hz, 1H), 7.39 − 7.35 (m, 2H), 4.00 (s, 6H), 3.95 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 163.72, 153.54, 153.33, 152.38 (2 C), 137.04, 136.44, 134.96, 133.77, 129.68, 126.81, 125.46, 122.75, 121.84, 114.11, 113.56, 105.02, 98.98 (2 C), 61.07, 56.38 (2 C). HRMS calcd for C22H20N5O3 [M + H]+ 402.1566, found 402.1566. Biological evaluation In vitro anti-proliferative activity The anti-proliferative effects of the synthesised derivatives and the reference compound SMART were assessed using the MTT colorimetric assay according to a previously reported procedure. Briefly, target cells were seeded into 96-well plates (Corning, USA) at a density of 5 × 10³ cells per well and incubated for 24 h to allow adhesion. Subsequently, the cells were exposed to serial dilutions of the test compounds or vehicle control (0.1% DMSO) for 72 h. After the treatment, the culture medium was removed and replaced with a fresh MTT solution (Roche, 11465007001). Following an additional 4 h incubation at 37 °C, the resulting formazan crystals were dissolved in 150 μL of dimethyl sulfoxide (DMSO, 150 μL/well, Rhawn Chemicals, R012315, 99.7%) under continuous shaking for 10 min on a microplate shaker (BioTek, Synergy H1). The absorbance was measured at 492 nm using a microplate spectrophotometer. The half-maximal inhibitory concentration (IC50) values were calculated by nonlinear regression curve fitting using IBM SPSS software. All cell lines, including HeLa (NCACC, SCSP-504), A549 (NCACC, SCSP-503), and MCF-7 (NCACC, SCSP-531) were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China) and maintained under recommended culture conditions. Effect on tubulin polymerisation The effect of compound 9t on tubulin polymerisation was assessed using a commercial kit (BK001, Cytoskeleton Inc., Denver, CO, USA) according to the manufacturer’s protocol. Briefly, purified tubulin was reconstituted in ice-cold G-PEM buffer (2 mM MgCl2, 80 mM PIPES, 1 mM GTP, 0.5 mM EGTA, and 15% glycerol). Aliquots of the tubulin solution were treated with compound 9t at the indicated concentrations, paclitaxel (Sigma-Aldrich, MFCD00869953), or SMART (InvivoChem, Cat No.:V3031) prior to polymerisation. The reaction mixtures were then transferred to a pre-warmed 96-well plate at 37 °C, and polymerisation was monitored kinetically over 62 min using an Infinite® F500 microplate reader. Fluorescence intensity was recorded at 60 s intervals (excitation 360 nm, emission 420 nm). The resulting kinetic curves enabled comparative analysis of the effects of each compound on tubulin assembly and polymerisation rate. Analysis of immunofluorescence staining Microtubule organisation was examined by immunofluorescence staining following a previously described protocol. HeLa cells seeded in 24-well plates were treated with vehicle or compound 9t for 24 h. After treatment, the cells were fixed with 4% formaldehyde, permeabilized with 0.2% Triton X-100 (Sigma-Aldrich, MFCD00128254), and blocked with 5% bovine serum albumin (BSA, Beyotime, ST023) for 30 min. Subsequently, the cells were incubated overnight at 4 °C with a mouse anti-α-tubulin primary antibody (Beyotime, AF2827; 1:100 dilution in PBS containing 2% BSA) at a 1:100 dilution in PBS (Beyotime, C0221B) containing 2% BSA. Following three washes with PBS, the samples were incubated with an FITC-conjugated anti-mouse secondary antibody (Abclonal, GTX300120; 1:100 in PBS with 2% BSA) for 3 h at 37 °C. Nuclei were counterstained with DAPI (4′,6-diamidino-2-phenylindole, Beyotime, C1006). After final washes, images were acquired using a fluorescence microscope to evaluate microtubule architecture and nuclear morphology. Cell cycle analysis HeLa cells were seeded in 6-well plates at a density of 2 × 105 cells per well and treated with compound 9t (33, 65, or 130 nM) or vehicle control (0.05% DMSO) for 24 h. Both adherent and floating cells were collected, washed with PBS, and fixed in ice-cold 70% ethanol at −20 °C for 2 h. After fixation, the cells were washed twice with PBS and stained with propidium iodide (PI) at 50 μg/mL in the presence of 100 μg/mL RNase A (Beyotime, C1734) for 30 min at room temperature in the dark. Flow cytometric analysis was carried out using a FACSVerse flow cytometer (BD Biosciences, USA) with excitation at 488 nm and emission detected through a 585/42 nm bandpass filter. Debris and doublets were excluded by gating on FSC-A versus SSC-A and FSC-A versus FSC-H, and 10,000 single cells were acquired per sample. Data were analysed using FlowJo software (version 10.8, TreeStar, USA) employing the Dean-Jett-Fox model for cell cycle phase quantification. Induction of cell apoptosis Apoptosis was evaluated using Annexin V-FITC/PI dual staining according to a previously described method. HeLa cells were seeded in 6-well plates at a density of 2 × 105 cells per well and treated with compound 9t (33, 65, or 130 nM) or vehicle control (0.05% DMSO) for 48 h. Both floating and adherent cells were collected, washed with PBS, and resuspended in 500 μL of 1× Annexin V binding buffer (10 mM HEPES, pH 7.4, 140 mM NaCl, 2.5 mM CaCl2). The cells were then stained with 5 μL of Annexin V-FITC (Beyotime, C1062S) and 5 μL of PI (Beyotime, ST1569) for 15 min at room temperature in the dark. Flow cytometric analysis was carried out using a FACScan flow cytometer (BD Biosciences, USA) with excitation at 488 nm. Annexin V-FITC fluorescence was detected through a 525/25 nm bandpass filter, and PI fluorescence through a 610/20 nm filter. Debris was excluded by gating on FSC-A versus SSC-A, and 10,000 events were acquired per sample. Quadrant gates were set using untreated control cells to distinguish viable (Annexin V-/PI-), early apoptotic (Annexin V+/PI-), late apoptotic (Annexin V+/PI+), and necrotic (Annexin V-/PI+) populations. Wound-healing migration assay The anti-migratory effect of compound 9t was evaluated using a wound-healing assay. HeLa cells were seeded in 6-well plates and cultured to 90–95% confluence. A straight scratch wound was created across the centre of each well using a sterile 200 μL pipette tip. The detached cells were removed by washing twice with PBS, and the remaining cells were incubated with serum-free medium containing vehicle control (0.1% DMSO), SMART (65 nM), or compound 9t (65 nM) for 24 h. Images of the wound gaps were captured at 0 h and 24 h using an inverted microscope (Olympus, IX73) equipped with a digital camera. The wound closure was assessed by measuring the remaining wound area using ImageJ software. All experiments were performed in triplicate. Molecular docking analysis To explore the possible binding conformations and intermolecular contacts between the synthesised compounds and tubulin, a docking study was performed using the Schrödinger software package. Three-dimensional structures of the ligands were constructed with ChemBioDraw Ultra, converted to SDF format, and subsequently optimised through energy minimisation employing the LigPrep module. The crystallographic coordinates of tubulin co-crystallized with a known ligand (PDB entry: 5LYJ) were downloaded from the RCSB Protein Data Bank. The protein structure was prepared using the Protein Preparation Wizard, which involved adding hydrogen atoms, setting correct bond orders, and refining the geometry with the OPLS3 force field to alleviate steric clashes and ensure proper structural orientation. A docking grid box of cubic shape with side lengths of 15 Å was defined and centred on the experimentally characterised colchicine-binding pocket of tubulin, thereby delimiting the region for ligand placement. Docking simulations were run with the Glide algorithm implemented in Schrödinger, operating under standard precision (SP) mode. The resultant binding poses and key ligand–protein interactions were subsequently examined and visualised using PyMOL. Molecular dynamics simulation To further validate the binding stability of compound 9t within the colchicine-binding pocket, molecular dynamics (MD) simulations were performed using the GROMACS 2025 software package. The amber99sb-ildn force field was applied for the protein, and the General Amber Force Field (GAFF) was used for the ligand. The TIP3P water model was employed to solvate the system within a periodic boundary box with a 1 nm buffer distance. Sodium and chloride ions were added to neutralise the system and adjust the ionic strength to 0.15 M to mimic physiological conditions. Prior to production simulation, the complex system was subjected to energy minimisation using the steepest descent method for 10,000 steps. Subsequently, the system was equilibrated under NVT ensemble at 300 K for 2 ns, followed by NPT ensemble for an additional 2 ns. Finally, a 100 ns production MD simulation was performed at 300 K and 1 atm. Trajectories were saved every 10 ps for subsequent analysis. Post-simulation analyses, including root-mean-square deviation (RMSD), root-mean-square fluctuation (RMSF), radius of gyration (Rg), and solvent-accessible surface area (SASA), were performed using GROMACS built-in utilities.