Biocompatible synthesis, computational studies and therapeutic evaluation of a dibutyl (2,2‐dicyano‐1‐phenylethyl) phosphonate derivatives targeting oxidative stress, microbial infection and cancer cells

Abstract

We synthesized a series of dibutyl (2,2-dicyano-1-phenylethyl) phosphonates using a solvent-free, one-pot reaction catalysed by L-carnitine hydroxide. This green method yielded products in 75–97% yield. Density functional theory analysis showed that these compounds have favourable HOMO–LUMO gaps and enhanced charge transfer, suggesting good nonlinear optical properties. Molecular docking indicated strong binding to key bacterial proteins. In vitro tests confirmed strong antioxidant activity (DPPH IC₅₀: 21.1–49.8 µg/mL; NO IC₅₀: 21.2–31.1 µg/mL) and effective antibacterial action against both Gram-positive and Gram-negative bacteria (inhibition zones: 2.0–3.8 mm). The compounds also inhibited cancer cell growth in five human cell lines (IC₅₀: 3.5–5.9 µM), with higher selectivity than doxorubicin over non-cancerous cells. ADMET analysis predicted good absorption, low toxicity and drug-likeness. Electron-withdrawing groups improved both biological and electronic properties. These phosphonates are promising candidates for new therapeutic development.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13205-025-04499-9.

Introduction

Organophosphorus compounds (OPCs) are a large group of chemicals containing phosphorus bonded to organic groups (Gundluru et al. 2023; Poreddy et al. 2024). They are important in agriculture, medicine and industry because of their diverse chemical and biological properties. In agriculture, OPCs are widely used to make pesticides and herbicides that control pests and weeds (Casida and Durkin 2013). Their role in agriculture has been instrumental in enhancing crop yields and food production globally. In medicine, OPCs are valued for their therapeutic potential. They are used to create drugs that act on specific enzymes and receptors important for treating diseases. For example, some OPCs have been designed as acetylcholinesterase inhibitors, which play a key role in nerve signalling (Cragg and Newman 2013). These inhibitors are utilized for treating neurological diseases such as Alzheimer's disease (Peitzika and Pontiki 2023). OPCs are also used in industry as flame retardants, plasticizers and lubricants, as they can change the properties of materials, making them valuable in manufacturing (Tian et al. 2023). However, concerns about their environmental and health effects have led to research on safer and more sustainable uses (Carvalho 2017). β-Phosphomalonates (BPMs) are a subgroup of OPCs that have attracted much attention. These molecules contain a phosphonate group, which gives them unique chemical and therapeutic properties (Hanson 2014). BPMs can bind to various biological targets, making them valuable in drug development (Hecker and Erion 2008). Their therapeutic effects are mainly due to their ability to inhibit key enzymes and proteins involved in disease pathways. For example, BPMs can interact with kinases and phosphatases, affecting pathways that control cell growth, differentiation and programmed cell death (apoptosis) (Strekalova et al. 2019). This makes, BPMs promising for developing new anticancer drugs, as they can interfere with the growth of cancer cells by targeting these important pathways (Krečmerová et al. 2022).

The synthesis of BPM typically involves the reaction of aldehydes with dialkylphosphonate and malononitrile. This one-pot synthesis is advantageous due to its simplicity, efficiency and the ability to generate a variety of derivatives by altering the substituents on the aldehyde component (Svintsitskaya et al. 2008; Deshmukh et al. 2021). Recent advancements in synthetic methodologies have focused on optimizing reaction conditions to improve yields and selectivity while minimizing environmental impact (Campbell 1992; Xuan et al. 2024). Microwave-assisted method has become highly effective for the efficient synthesis of BPM. This approach reduces reaction times and enhances product yields by providing uniform heating and energy transfer to the reactants (Keglevich et al. 2016; Slocombe and Porch 2021). In recent years, the synthesis of OPCs such as BPM has drawn substantial interest owing to their broad spectrum of biological activities. However, conventional approaches to synthesis often necessitate harsh reaction conditions, toxic solvents and non-recyclable catalysts, raising concerns regarding their environmental impact (Odinets and Matveeva 2012; Gawande et al. 2013). To address these issues, green chemistry principles have been increasingly applied to develop sustainable and eco-friendly synthetic routes (Anastas and Eghbali 2010). Among these advances, solvent-free reactions and the application of ionic liquids and bio-derived catalysts have emerged as effective strategies for reducing the ecological footprint of chemical processes. In this context, L-carnitine hydroxide, a biodegradable, non-toxic and renewable catalyst, has shown excellent efficacy in promoting base-catalysed condensation reactions under mild and solvent-free conditions. Its dual role as a hydroxide donor and hydrogen-bond stabilizer makes it particularly attractive for environmentally benign transformations. These methods not only make reactions more efficient but also support global efforts to promote sustainable chemical manufacturing (Zhang et al. 2018; Ganesh et al. 2021). Accurate characterization of BPM is essential to confirm their structure and assess their properties. A combination of spectroscopic techniques is employed to achieve this, including Fourier-transform infrared (FTIR) spectroscopy, nuclear magnetic resonance (NMR) spectroscopy and high-resolution mass spectrometry (HRMS) (Silverstein and Bassler 1962). FTIR spectroscopy is used to identify characteristic functional groups within the molecules, such as the phosphonate ester linkage and cyano groups, by examining specific absorption bands (Stuart 2004). NMR spectroscopy delivers in-depth data about the molecular structure, including the chemical environment of hydrogen, carbon and phosphorus atoms. This information is crucial for verifying the successful synthesis of target molecules (Claridge 2016). HRMS is employed to determine the molecular weight and confirm the purity of the synthesized BPM, ensuring that the desired chemical entities have been obtained (Gross 2006). These techniques are crucial for understanding the physicochemical properties of BPM and their potential binding with biological targets.

Density Functional Theory (DFT) calculations are widely used to study the electronic properties of molecules, including BPM. DFT provides key data on molecular orbital distributions and their corresponding energy levels, specifically the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) gap, which are key to anticipate the reactivity and structural stability of the compounds (Jones and Gunnarsson 1989). These computational studies complement experimental data and guide the design of Novel compounds exhibiting enhanced antibacterial activity (Young 2004). In addition to DFT, in silico studies, including ADMET (Absorption, Distribution, Metabolism, Excretion and Toxicity) analysis and molecular docking, were conducted to predict the pharmacokinetic properties and potential biological activities of BPM. ADMET analysis evaluates the drug-likeness of the compounds, predicting their ADMET and provides a comprehensive overview of their pharmacokinetic profiles (Ekins et al. 2007; Mishra et al. 2023b). Docking simulations were carried out to investigate molecular interactions with key bacterial proteins, including DNA Gyrase from Escherichia coli (E. coli), LasR Quorum Sensing Regulator from Pseudomonas aeruginosa (P. aeruginosa) and Penicillin-Binding Protein 3 (PBP3) from Bacillus subtilis. These proteins are essential for bacterial survival and antibiotic resistance mechanisms, positioning them as promising targets for new antibacterial drugs. The docking studies were carried out to predict whether BPM derivatives can inhibit these proteins. Such inhibition could disrupt bacterial growth and survival (Jampilek 2018; Prieto-Martínez et al. 2018). The potential of BPM as antibacterial agents lies in their ability to interact with and inhibit key bacterial proteins involved in essential processes such as DNA replication, quorum sensing and antibiotic degradation. By targeting proteins like DNA Gyrase, LasR and Penicillin-Binding Protein 3 (PBP3), these compounds can disrupt critical bacterial functions, leading to bacterial cell death or impaired growth (Ganapathi and Ganapathi 2013; Prajapati et al. 2025). The structural versatility of BPM enables the development of analogues exhibiting improved antibacterial efficacy and specificity. This adaptability is crucial for developing compounds that selectively target bacterial cells while minimizing off-target effects, which is essential for reducing toxicity to host cells (Vander Heiden and DeBerardinis 2017). Moreover, the broad-spectrum potential of BPM, which includes antibacterial and antiviral activities, renders them suitable for the design of new treatment options against various pathogens (Lokhandwala and Desai 2008; Petrescu et al. 2019). Recent research on BPM has focused on enhancing their antibacterial potential by modifying the core structure to improve binding affinity to target proteins, increasing solubility, bioavailability and combining BPM with other antibacterial agents to achieve synergistic effects (Riley et al. 2019). Upcoming studies are expected to concentrate on focus on optimizing synthetic routes to produce BPM with superior pharmacological profiles. Additionally, extensive in vitro and in vivo studies are essential to validate the antibacterial activity observed in silico and to evaluate the safety and effectiveness of these compounds in preclinical models (Iaroshenko 2019). Exploring broader applications beyond antibacterial therapy could further expand the therapeutic potential of these versatile compounds (Balali-Mood and Saber 2012). Cancer is a multifactorial disease characterized by the uncontrolled proliferation and growth of cells and is classified based on the affected organ system such as breast, blood, colon and liver each exhibiting distinct molecular and pathological features (Mishra et al. 2023c, a). BPM have garnered considerable interest in anticancer drug design due to their structural analogy to biological phosphate esters and their ability to form metabolically stable C–P bonds, allowing them to interfere with critical enzymatic and cellular processes (Demkowicz et al. 2016; Khan et al. 2019). The β-dicarbonyl-phosphonate structure can induce apoptosis, regulate oxidative stress and block nucleotide biosynthesis in cancer cells (Hassanin et al. 2024). Recent studies highlight that appropriately substituted BPM can effectively target mitochondrial pathways and disrupt tumour cell proliferation by enhancing intracellular reactive oxygen species (ROS) and activating caspase-dependent apoptosis (Demkowicz et al. 2016). Importantly, electronic modulation via substitution at the β-position allows tuning of pharmacological activity and selectivity (Vorlová et al. 2015).

In the present study reports a novel, green and efficient synthesis of β-phosphonomalononitriles (BPMs) catalyzed by L-carnitine hydroxide an organocatalyst not previously reported for this transformation. The method is solvent-free, operates under mild conditions and yields high product outputs, aligning with sustainable chemistry principles. Importantly, we integrate theoretical studies (DFT, molecular docking and ADMET) with experimental antibacterial, antioxidant and anticancer evaluations, offering a rare and comprehensive perspective on structure activity relationships. This multifaceted approach distinguishes our work from prior reports, adding significant novelty and interdisciplinary value.

Methods

Materials

All chemicals and reagents were of analytical grade and purchased from commercial suppliers. The aldehydes, dibutylphosphite and malononitrile were sourced from Sigma-Aldrich, USA and put to use without further processing. Solvents, including ethanol and dichloromethane, were sourced from Merck, USA and distilled before use to ensure purity. The catalyst L-carnitine hydroxide was sourced from Fisher Scientific.

General procedure for the synthesis of BPM (4a-l)

A series of BPM derivatives (4a-l) were synthesized via a one-pot, three component reaction involving equimolar amounts of substituted aromatic aldehydes (1a-l, 1 mmol), malononitrile (2b, 1 mmol) and dibutylphosphite (3c, 1 mmol). The reaction mixture catalysed by L-carnitine hydroxide (5 mol%), was stirred at ambient temperature in a 50 mL round-bottom flask, following a previously reported protocol (Zhang et al. 2018). The reaction proceeded cleanly without detectable byproducts, as confirmed by thin-layer chromatography (TLC), which showed a single spot. The crude product was initially purified by recrystallization and further refined by column chromatography using silica gel and a hexane:ethyl acetate mixture as the eluent. The target compounds were obtained in excellent yields ranging from 75 to 97%.

This synthetic methodology is straight forward, metal-free and environmentally benign. Structural confirmation and purity assessment were performed using melting point analysis, FTIR, NMR (¹H, ¹³C, ³¹P) and high-resolution mass spectrometry (HRMS). FTIR spectra were recorded on a Shimadzu IR Affinity-1 spectrometer using KBr pellets. The spectral range was set from 4000 to 400 cm⁻¹, with a resolution of 4 cm⁻¹. Characteristic absorption bands of the phosphonate ester linkage and cyano groups were identified. NMR spectra were obtained using a Bruker Avance III 400 MHz spectrometer. Samples were dissolved in deuterated chloroform (CDCl₃) and chemical shifts were reported in parts per million (ppm) relative to tetramethylsilane (TMS) as the internal standard. HRMS was performed using an Agilent 6545 Q-TOF system in positive ion mode. The molecular weights of the synthesized BPM were determined to confirm their identity and purity. For scale-up experiments, the procedure above was followed using 10 mmol of each reagent under identical conditions. The product was isolated as described. After each reaction cycle, the catalyst was recovered by washing the residue with dichloromethane, drying and directly reusing for subsequent reactions under identical conditions.

Density functional theory (DFT) analysis

To explore the electronic characteristics and reactive behaviour of the synthesized compounds (4a-l), DFT calculations were carried out. All computational studies were carried out utilizing the Gaussian 09 software suite, implementing the hybrid B3LYP exchange–correlation functional in conjunction with the 6–311 + G(d,p) basis set (Frisch 2009). This functional-basis combination has been extensively validated for organic systems, providing a robust compromise between computational efficiency and predictive accuracy (Becke 1993; Lee et al. 1988). The molecular geometries of all the target compounds were thoroughly optimised without symmetry constraints. Frontier molecular orbital (FMO) analysis was conducted to determine the distributions and energy levels of the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO). The HOMO–LUMO energy gap, a key descriptor for assessing molecular stability and chemical reactivity, was calculated for each compound. A reduced energy gap typically suggests enhanced electron transfer capacity and potential biological activity (Dalal et al. 2024). In addition, Molecular Electrostatic Potential (MEP) surfaces were generated to visualize the spatial distribution of electrostatic potential across the molecular surface, highlighting regions of electrophilic and nucleophilic activity. These maps offer valuable insights in to preferred sites for intermolecular interactions and are instrumental in rationalizing the compounds’ binding affinity with biological targets (Pearson 1986). To assess the compounds’ potential in optical and photonic applications, Nonlinear Optical (NLO) parameters including average polarizability (α) and first hyperpolarizability (β) were evaluated. Dipole moment values were also extracted to provide supplementary information regarding the electronic asymmetry of the molecules, which is crucial for optoelectronic applications (Singh et al. 2023). Mulliken population analysis was utilized to estimate the distribution of atomic charges across each molecule, offering a deeper understanding of intramolecular charge delocalization and potential donor–acceptor dynamics. This charge analysis supports the prediction of reactivity trends and electronic transitions in various chemical environments. All visual interpretations, including orbital plots and MEP maps, were generated using GaussView software, enabling clear graphical representation of electronic and electrostatic properties. The methodologies adopted conform to standard DFT protocols to ensure reliability and reproducibility in the calculated results (Frisch 2009).

In silico ADMET analysis

The pharmacokinetic profiles of the synthesized compounds (4a-l) were predicted using the SwissADME online tool accessible at http://www.swissadme.ch, which evaluates ADME properties (Daina et al. 2017). These predictions provide important information about the drug-likeness and potential safety of the compounds. Key parameters assessed included lipophilicity, solubility and gastrointestinal absorption, which are crucial for evaluating the bioavailability of the compounds. The Lipinski's Rule of Five was used to assess drug-likeness, focusing on molecular weight, log P, hydrogen bond donors and hydrogen bond acceptors (Lipinski et al. 1997). Additionally, ProTox-II (available at http://tox.charite.de/protox_II) was utilized to estimate the potential toxicity profiles of the ligands, focusing on critical endpoints such as hepatotoxicity, nephrotoxicity, skin sensitization and LD50 values for acute toxicity (Banerjee et al. 2018). The analysis also considered possible interactions with cytochrome P450 enzymes, which play a crucial role in drug metabolism and may affect the compounds' efficacy and safety profiles. Toxicity predictions, including mutagenicity and hepatotoxicity, were assessed to ensure the compounds' safety profiles were favourable (Veber et al. 2002).

Molecular docking studies

Molecular docking simulations were carried out to investigate the binding affinities and interaction modes of 4a-l with key bacterial proteins, including DNA Gyrase from Escherichia coli (PDB ID: 1KZN), LasR Quorum Sensing Regulator from Pseudomonas aeruginosa (PDB ID: 2UV0) and Penicillin-Binding Protein 3 (PBP3) from Bacillus subtilis (PDB ID: 2Y2M). These proteins were chosen due to their essential roles in bacterial replication, quorum sensing and cell wall biosynthesis, respectively and their relevance to the tested strains (E. coli, B. subtilis, B. megaterium and K. pneumoniae) (Ganapathi and Ganapathi 2013). Crystal structures were retrieved from the RCSB Protein Data Bank and prepared by removing water molecules, adding polar hydrogens and assigning Kollman charges. Ligand structures were optimized using DFT (B3LYP/6-311G(d,p)) and converted to PDBQT format using AutoDock Tools. Docking simulations were carried out using AutoDock Vina (Trott and Olson 2010). To validate docking accuracy, known inhibitors were redocked into their respective targets, yielding RMSD values below 2.0 Å, confirming binding site fidelity. Protein–ligand interactions were visualized using PyMOL and LigPlot + , highlighting hydrogen bonding, hydrophobic contacts and π-π stacking interactions (Wallace et al. 1995; DeLano 2002), hydrophobic contacts and π-π stacking interactions. Ciprofloxacin, a clinically approved DNA gyrase inhibitor, was used as a positive control to validate docking accuracy and scoring consistency. Lower binding energy scores indicated stronger interactions between the compounds and the target proteins, suggesting potential inhibition of bacterial growth or resistance mechanisms (Kumar et al. 2024). The promising docking results highlight the potential of 4a-l to inhibit key bacterial proteins, providing a basis for further experimental exploration and development as antibacterial agents. Future research will involve in vivo to validate the effectiveness of these compounds against multidrug-resistant bacterial strains.

Assessment of antioxidant activity

2, 2-diphenyl-1-picryl-hydrazyl (DPPH) free radical scavenging assay: The DPPH radical scavenging efficacy of newly synthesized compounds, such as 4a-l, were determined using the 2, 2-diphenyl-1-picryl-hydrazyl assays (Sharma and Bhat 2009). In 100 ml of methanol, 4 mg of DPPH were dissolved to create a stock solution. To make the working standard, a stock solution was diluted with methanol. Then, 4 mL of the DPPH standard solution was mixed with different concentrations of the freshly synthesized compounds (25, 50, 75 and 100 μM/mL). The samples were then incubated in the dark at room temperature for 20 min after 100 μL of methanol was added to them. A measurement of optical absorption was taken at 517 nm. The absorbance of the standard ascorbic acid and the parallel control (non-additive) were both measured using identical procedures. The activity of DPPH radical scavenging was determined by applying the following formula:

$$\text{Inhibition}\left(\%\right)=\left[\left({\text{A}}_{\text{control}}-{\text{A}}_{\text{test}}\right)/{\text{A}}_{\text{control}}\right]\times 100$$

where;

A_control = Absorbance of the control;

A_test = Absorbance of the synthesized compounds.

We expressed the results of the radical scavenging activity relative to the standard using the half maximal inhibitory concentration (IC₅₀). After all measurements were taken three times, the average and standard deviation were calculated.

Nitric oxide (NO) free radical scavenging activity: The method of (Can et al. 2022) was marginally modified to measure the nitric oxide scavenging activity. Sodium nitropruside was the source of nitric oxide radicals (NO). The novel synthesized compounds (4a-l) were incubated at 25 °C for 15 min after 1 ml of sodium nitroprusside (10 mM) and 1.5 ml of phosphate buffer saline (0.2 M, pH 7.4) was added to varying concentrations (25, 50, 75 and 100 μM/mL). The reaction mixture was then treated with 1 ml of Griess reagent (1% sulfanilamide, 2% H₃PO₄ and 0.1% naphthylethylenediamine dihydrochloride). The positive control was ascorbic acid, the standard antioxidant. The absorbance of the reaction mélange that resulted was determined at 546 nm.

Antibacterial activity

The in vitro antibacterial efficacy of the synthesized dibutyl (2,2-dicyano-1-phenylethyl) phosphonate derivatives (4a-l) was evaluated using a modified agar well diffusion method (Flanagan and Steck 2017; Oliveira et al. 2022). The study was conducted against four bacterial strains: Bacillus megaterium and Bacillus subtilis (Gram-positive) and Klebsiella pneumoniae and Escherichia coli (Gram-negative). Ciprofloxacin (100 µg/mL) was used as the positive control, while Dimethyl sulfoxide (DMSO) served as the negative control.

Each bacterial strain was cultured on Mueller–Hinton Agar (MHA) and the inoculum was standardized to 0.5 McFarland turbidity (approximately 1.5 × 10⁸ CFU/mL). Wells of 6 mm diameter were punched into the agar plates and loaded with 50 µL of test solutions at varying concentrations (25, 50, 75 and 100 µg/mL). Plates were incubated at 37 °C for 24 h under aerobic conditions. The antibacterial activity was quantified by measuring the diameter of the zone of inhibition (in mm) using a digital Vernier caliper. Minimum inhibitory concentration (MIC) values were determined using the broth microdilution method in 96-well plates, following CLSI guidelines (Weinstein and Lewis 2020). Serial dilutions of each compound were prepared in Mueller–Hinton Broth and bacterial growth was monitored after 24 h of incubation. MIC values were recorded as the lowest concentration at which no visible growth was observed. All experiments were performed in triplicate (n = 3) and data were expressed as mean ± standard deviation. Statistical analysis was conducted using one-way ANOVA followed by Tukey’s post hoc test, with significance set at p < 0.05.

Anticancer activity

The in vitro anticancer potential of the synthesized BPM derivatives (4a–l) was assessed using the MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide) assay, a widely accepted colorimetric method for evaluating cell viability and cytotoxicity. The assay was performed against a panel of five human cancer cell lines representing different tissue origins: MCF-7 (breast adenocarcinoma), HeLa (cervical carcinoma), A549 (lung adenocarcinoma), HT-29 (colorectal adenocarcinoma) and PC-3 (prostate carcinoma). These cell lines were selected to provide a broad-spectrum evaluation of the compounds' anticancer efficacy across hormone-responsive, epithelial and metastatic cancer types (Mosmann 1983; Nguyen et al. 2024). All cell lines were procured from ATCC (American Type Culture Collection) and cultured in Dulbecco’s Modified Eagle Medium (DMEM) or RPMI-1640, supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin and 100 µg/mL streptomycin. Cells were maintained under standard conditions at 37 °C in a humidified atmosphere with 5% CO₂. Media were changed every 48 h and cells were sub-cultured at 70–80% confluency using trypsin–EDTA. Exponentially growing cells were seeded in 96-well plates at a density of 1 × 10⁴ cells/well and allowed to adhere for 24 h. The compounds were dissolved in DMSO and diluted with culture medium to achieve final concentrations ranging from 1 to 100 µM. After 48 h of treatment, 20 µL of MTT reagent (5 mg/mL in Phosphate Buffer Saline (PBS)) was added to each well and plates were incubated for an additional 4 h. The resulting formazan crystals were solubilized by adding 100 µL of DMSO and absorbance was measured at 570 nm using a microplate reader. Each treatment was performed in triplicate and percentage cell viability was calculated relative to untreated controls (Shoemaker 2006). IC₅₀ values (the concentration of compound required to inhibit 50% of cell viability) were determined by nonlinear regression analysis using GraphPad Prism software. Doxorubicin was used as a positive control.

Results and discussion

Chemistry

The BPM (4a-l) were efficiently synthesized via a one-pot, three-component condensation involving substituted aromatic aldehydes (1a-l), malononitrile (3c) and dibutylphosphite (2b). L-carnitine hydroxide, a green and bio-based catalyst, enabled the transformation under solvent-free conditions at room temperature (Scheme 1). L-carnitine is commercially available as the hydrochloride salt and must be converted to its hydroxide (free base) form to enable base catalysis. The free base was prepared via neutralization of L-carnitine hydrochloride with equimolar NaOH in water, removing water under reduced pressure to obtain L-carnitine hydroxide, which was then used directly as catalyst (Zhang et al. 2018). This methodology afforded the target BPM compounds in excellent isolated yields (75–97%), eliminating the need for external heating, hazardous reagents, or complex post-reaction purification such as chromatography. The use of L-carnitine hydroxide, a naturally derived amino acid derivative, provides an eco-friendly ionic catalyst supporting base-mediated activation of phosphite and aldehyde components. The catalyst’s bifunctional nature combining basicity and hydrogen bonding facilitates a tandem Knoevenagel–phospha-Michael condensation in a single operational step, in line with green chemistry principles of atom economy, energy efficiency and environmental safety (Salido-Fortuna et al. 2022).

Scheme 1.

Solvent-free synthesis of dibutyl (2,2-dicyano-1-phenylethyl) phosphonates (4a-l) catalyzed by L-carnitine hydroxide at room temperature

Catalyst recycling and scalability: To validate the scalability of the synthetic route, compound 4a was synthesized on a gram scale using 10 mmol of each reactant. The reaction yielded 3.98 g of pure product (92% yield) after recrystallization, with spectral data confirming consistency in structure and purity. Additionally, reproducibility was assessed by synthesizing selected compounds in three independent batches under identical conditions. TLC analysis revealed identical R_f values with sharp single spots and spectral data (NMR and HRMS) remained consistent across batches, confirming the robustness of the method. Catalyst recyclability was assessed under standard reaction conditions. Following each reaction cycle, L-carnitine hydroxide was recovered via dichloromethane washing, dried and reused without further modification. Remarkably, the catalyst maintained high activity across five consecutive cycles, with yields of 96%, 95%, 94%, 92% and 89%, respectively (Fig. S49). These results demonstrate excellent stability, robustness and cost-effectiveness of the catalyst, supporting its utility for scale-up and sustainable operation.

Influence of aromatic substituents: The impact of electronic effects on reaction outcome was investigated using aldehyde substrates bearing electron-withdrawing (-NO₂, -Cl) and electron-donating (-OMe, -Me) groups (Table S7). Substrates with EWGs led to higher yields and faster reaction rates (up to 97%), attributed to enhanced electrophilicity of the carbonyl and better stabilization of transition states. In comparison, EDGs modestly reduced reaction kinetics due to decreased carbonyl electrophilicity. However, substituents capable of hydrogen bonding (hydroxyl, methoxy) displayed secondary stabilization effects, partially compensating for these electronic disadvantages. The observed trends highlight a clear structure–reactivity relationship, which may guide future substrate selection and rational reaction design. Green chemistry metrics, including Atom Economy (95–98%), E-Factor (0.039–0.052) and PMI (1.039–1.052), were evaluated across all derivatives. These results demonstrate the method’s sustainability, high efficiency and broad substrate compatibility (see Table S8).

Compound stability: BPM compounds showed excellent shelf stability over three months when stored at room temperature in closed vials away from direct sunlight. No observable change in colour or texture occurred and TLC profiles and NMR spectra confirmed retention of chemical integrity, indicating good stability under standard laboratory conditions.

Catalyst optimization: To enhance the efficiency and sustainability of the synthetic protocol, a series of inorganic, organic and bio-based catalysts were screened (Table 1). In the absence of any catalyst, no product formation was observed even after 12 h, underscoring the necessity of base catalysis. Traditional bases such as NaHCO₃, K₃PO₄ and metal oxides (e.g., MgO, CaO, ZnO) provided moderate yields ranging from 47 to 59%, requiring up to ~ 70 min of reaction time. Organic bases including pyrrolidine, proline and ethanolamine significantly improved the outcome, reducing reaction times to under 30 min and achieving yields above 85%. Among the catalysts tested, L-carnitine hydroxide showed the most promising results. Under solvent-free, room temperature conditions, it afforded 97% yield within 15 min bio-based catalyst with ionic liquid-like properties This bio-based catalyst, to the best of our knowledge, has not been previously reported for the Knoevenagel-phospha-Michael tandem sequence. Its zwitterionic structure and bifunctional catalytic behaviour -hydroxide ion for enolate formation and quaternary ammonium/hydroxyl groups for hydrogen bonding enable efficient transition state stabilization and electrophilic activation. Compared to conventional organocatalysts such as proline, DBU (1,8-Diazabicyclo[5.4.0]undec-7-ene) and ethanolamine, L-carnitine hydroxide demonstrates superior reactivity and selectivity, likely due to its unique scaffold and cooperative interactions. These findings indicate that carnitine hydroxide not only provides basic activation but also enhances stabilization of the transition state via hydrogen bonding and ion dipole interactions. Its ionic liquid-like properties further support its role as a green and sustainable catalyst (Salido-Fortuna et al. 2022). These findings indicate that carnitine hydroxide not only provides basic activation but also enhances stabilization of the transition state via hydrogen bonding and ion–dipole interactions.

Table 1.

Effect of various catalysts on the synthesis of compound (2,2-dicyano-1-(naphthalen-1-yl)ethyl)phosphonate (4a)^a. Reaction conditions and yields are compared to assess catalytic efficiency

Entry	Catalyst (mol %)	Reaction time (min)	Isolated yield (%)b
1	Catalyst free	12 h	No reaction
2	NaHCO3 (5)	60	47
3	K3PO4 (5)	58	59
4	MgO (5)	65	55
5	CaO (5)	50	48
6	ZnO (5)	70	58
7	Al2O3 (5)	45	59
8	Pyrrolidine (5)	30	86
9	Proline (5)	20	86
10	Ethanolamine (5)	20	90
11	L-carnitine hydroxide (5)	15	97

^aReaction conditions: Naphthaldehyde (1a), malononitrile (2) and dibutylphosphte (3) using various catalysts under room temperature and neat conditions. ^bIsolated product yields

Solvent optimization: Following catalyst screening, the role of solvents was examined using carnitine hydroxide as the catalyst under identical molar loading (5 mol%). Solvents tested included polar protic (ethanol, methanol) and aprotic (THF, acetonitrile, CH₂Cl₂, CHCl₃, toluene) systems (Table 2). Polar protic solvents such as methanol (86%) and ethanol (80%) supported higher yields, while aprotic solvents resulted in significantly lower conversions, likely due to their inability to stabilize charged intermediates or engage in hydrogen bonding with the catalyst. Maximum efficiency was attained under solvent-free conditions, where carnitine hydroxide catalysed the transformation in just 5 min, delivering up to 97% yield. Despite lower catalyst loadings (1–2.5 mol%), good yields (75–86%) were obtained, further supporting the atom economy and practicality of the neat reaction conditions. The results align with recent trends in green chemistry advocating solvent less approach for multicomponent synthesis.

Table 2.

Optimization of catalyst loading (L-carnitine hydroxide) and solvent conditions for the synthesis of compound (2,2-dicyano-1-(naphthalen-1-yl)ethyl)phosphonate (4a)^a. Yields and reaction times are summarized

Entry	Catalyst (mol %)	Solvent	Time (min)	Yield (%)
1	L-carnitine hydroxide (5)	Ethanol	30	80
2	L-carnitine hydroxide (5)	Methanol	45	86
3	L-carnitine hydroxide (5)	THF	50	81
4	L-carnitine hydroxide (5)	Acetonitrile	55	79
5	L-carnitine hydroxide (5)	CH2Cl2	60	68
6	L-carnitine hydroxide (5)	CHCl3	80	58
7	L-carnitine hydroxide (5)	Toluene	75	60
8	L-carnitine hydroxide (1)	Solvent free	5	75
9	L-carnitine hydroxide (2.5)	Solvent free	5	86
10	L-carnitine hydroxide (5)	Solvent free	5	97
11	L-carnitine hydroxide (10)	Solvent free	5	97
12	L-carnitine hydroxide (15)	Solvent free	5	97

^aReaction conditions: Naphthaldehyde (1a), malononitrile (2) and dibutylphosphte (3) using Carnitine hydroxide as catalyst under room temperature and neat conditions. ^bIsolated product yields

The developed protocol was compared with several existing methodologies for the synthesis of β-phosphonomalononitriles, our solvent-free and metal-free protocol offers several advantages: Microwave-Assisted Syntheses: While microwave irradiation accelerates reaction rates and improves yields, it often requires specialized equipment and may pose scalability challenges due to fixed magnetron geometry. For example, microwave-assisted McKenna dealkylation of phosphonates achieves rapid conversion but involves BTMS (bis(trimethylsilyl)trifluoroacetamide) and solvents like acetonitrile (Mustafa et al. 2023). Metal-Catalyzed Phosphomethylation: Metal-catalyzed hydrophosphination and phosphomethylation strategies offer high selectivity but rely on transition metals such as Ni, Co, or Rh, which may introduce toxicity and purification challenges. These methods also generate stoichiometric waste and often require inert atmospheres or elevated temperatures (Novas and Waterman 2022). Ionic Liquid-Based Organocatalysis: Ionic liquids have been employed as both solvents and catalysts in Knoevenagel and Michael reactions, offering recyclability and polarity tuning. However, their synthesis and disposal raise environmental concerns and their viscosity can hinder mass transfer in scaled-up reactions (Pandolfi et al. 2022; Yadav and Ahmaruzzaman 2022).

In contrast, our method utilizes L-carnitine hydroxide, a naturally derived and biodegradable organocatalyst under ambient conditions without external heating, metal catalysts, or solvents. It demonstrates broad substrate compatibility, high atom economy and excellent yields (75–97%) across diverse aromatic aldehydes. These features underscore its potential for sustainable and preparative-scale applications.

A plausible mechanism for BPM in presence of L-Carnitine hydroxide: The tandem reaction proceeds via a two-step mechanism catalyzed by L-carnitine hydroxide under solvent-free conditions, as illustrated in Scheme 2. In the first step, malononitrile (I) is deprotonated by the hydroxide ion to generate a nucleophilic carbanion, which attacks the electrophilic carbonyl carbon of the aromatic aldehyde (II), forming the Knoevenagel intermediate (III). This transformation is facilitated by hydrogen bonding between the quaternary ammonium group of L-carnitine and the aldehyde, enhancing electrophilicity. In the second step, dialkyl phosphite (IV) is deprotonated to yield a phosphorus-centered nucleophile, which undergoes conjugate addition to the electron-deficient alkene of intermediate (III), affording the β-phosphonomalononitrile product (V). The bifunctional nature of L-carnitine hydroxide combining basic activation and hydrogen bonding plays a crucial role in both stages.

Scheme 2.

Proposed reaction mechanism for the synthesis of BPM (4a-l) catalysed by L-carnitine hydroxide

Carrying out the reaction under solvent-free conditions offers several kinetic and practical advantages. The absence of solvent increases the effective concentration of the reactants and catalyst, allowing the transformation to proceed more rapidly due to enhanced molecular interactions. This close proximity of species accelerates bond formation and reduces the activation barrier for each step. Moreover, without solvent molecules competing for hydrogen bonding or ionic interactions, the catalytic activity of L-carnitine hydroxide is more pronounced. These conditions also prevent dilution of reactive intermediates, leading to a more efficient conversion to the final product. From a synthetic standpoint, the solvent-free approach not only shortens reaction time but also improves the isolated yield by minimizing loss during workup and avoiding potential side reactions commonly facilitated in solution.

Spectral characterization of BPM (4a-l): All synthesized compounds (4a-l) were characterized using FTIR, ¹H, ¹³C and ³¹P NMR spectroscopy, along with high-resolution mass spectrometry (HRMS). FTIR spectra consistently showed key functional group absorptions: C–H stretching (2948–2954 cm⁻¹), nitrile (C≡N) stretching (2232–2237 cm⁻¹), P = O (1261–1268 cm⁻¹), P-O-C (1042–1046 cm⁻¹) and P–C (745–749 cm⁻¹). Aromatic C = C stretching bands appeared around 1592–1597 cm⁻¹. ¹H NMR spectra revealed diagnostic methine protons adjacent to the dicyano group (δ 4.44–5.91 ppm), phosphonate-linked methylene protons (δ 2.23–3.71 ppm) and aromatic protons (δ 6.62–8.00 ppm), with substituent-specific signals such as methoxy (δ ~ 3.51–3.97 ppm) and hydroxyl (δ ~ 5.94 ppm) groups. ¹³C NMR spectra confirmed the presence of aromatic, nitrile and phosphonate carbons, with characteristic signals for substituted phenyl rings and alkyl chains. ³¹P NMR chemical shifts ranged from δ 20.16 to 22.35 ppm, consistent with phosphonate environments. HRMS data validated the molecular formulas with excellent agreement between calculated and observed m/z values. Complete spectral data and annotated spectra for all compounds are provided in the Supplementary Information (Figures S1–S48).

DFT analysis

To explore the electronic structure, reactivity patterns and optoelectronic behaviour of the 4a-l, DFT calculations were conducted using the B3LYP functional with the 6–311 + G(d,p) basis set. The investigation encompassed optimized geometries, FMOs, MEP surfaces, NLO properties and Mulliken atomic charge distribution (Poreddy et al. 2025).

Optimized geometry

The molecular geometries of 4a-l converged successfully to local minima, confirmed by the absence of imaginary frequencies. Structural optimization revealed a near-planar arrangement around the central conjugated system, with the phosphonate group introducing subtle torsional distortions due to steric and electronic effects see Table 3. The integrity of the BPM framework was maintained across all derivatives, with substituent variations affecting local geometry, particularly in the side chains and phosphorus coordination sphere. These deviations are expected to modulate the electronic and reactivity profiles of the compounds.

Table 3.

DFT-optimized geometries of compounds 4a-l at B3LYP/6-311 + G(d,p) level

Frontier molecular orbital (FMO) analysis

The electronic properties of 4a–l was further elucidated by FMO analysis, with the computed HOMO and LUMO energies summarized in Table 4 and visualized in Fig. 1. All calculations were performed at the B3LYP/6–311 + G(d,p) level of theory in the gas phase, which is well-established for predicting molecular orbital energies and reactivity descriptors in organic frameworks (Becke 1993; Lee et al. 1988).

Table 4.

Frontier orbital energies (EHOMO, ELUMO) and global reactivity descriptors (ΔE, hardness, softness, electronegativity) for compounds 4a-l

Compounds	HOMO (eV)	LUMO (eV)	ΔE	η (eV)	S (eV−1)	IP (eV)	EA (eV)	μ (eV)	ω (eV)
4a	−0.2346	−0.0661	0.1685	0.08425	11.86944	0.2346	0.0661	0.15035	0.13416
4b	−0.2081	−0.0264	0.1817	0.09085	11.00715	0.2081	0.0264	0.11725	0.07566
4c	−0.2345	−0.0332	0.2013	0.10065	9.93542	0.2345	0.0332	0.13385	0.089
4d	−0.2343	−0.035	0.1993	0.09965	10.03512	0.2343	0.035	0.13465	0.09097
4e	−0.2315	−0.0378	0.1937	0.09685	10.32525	0.2315	0.0378	0.13465	0.0936
4f	−0.2621	−0.0507	0.2114	0.1057	9.46074	0.2621	0.0507	0.1564	0.11571
4 g	−0.2657	−0.0509	0.2148	0.1074	9.31099	0.2657	0.0509	0.1583	0.11666
4 h	−0.2385	−0.0375	0.201	0.1005	9.95025	0.2385	0.0375	0.138	0.09475
4i	−0.2535	−0.0983	0.1552	0.0776	12.8866	0.2535	0.0983	0.1759	0.19936
4j	−0.2957	−0.1119	0.1838	0.0919	10.88139	0.2957	0.1119	0.2038	0.22598
4 k	−0.2661	−0.0511	0.215	0.1075	9.30233	0.2661	0.0511	0.1586	0.1171
4 l	−0.2701	−0.0495	0.2206	0.1103	9.06474	0.2701	0.0495	0.1598	0.1158

Fig. 1.

Frontier molecular orbitals (HOMO and LUMO) of compounds 4a-l, calculated via DFT (B3LYP/6–311 + G(d,p)). Upper panels show LUMO; lower panels show HOMO. Orbital localization and ΔE values reveal substituent-driven electronic effects, supporting SAR insights

The HOMO–LUMO energy gap (ΔE) serves as a key indicator of the molecule’s kinetic stability and chemical reactivity. Compounds 4i and 4j showed the narrowest energy gaps (ΔE = 0.1552 eV and 0.1838 eV, respectively), implying enhanced intramolecular charge transfer capabilities and potentially greater chemical reactivity. This is attributed to the strong electron-withdrawing nature of their nitro substituents, which stabilize the LUMO significantly (LUMO = −0.0983 eV for 4i; −0.1119 eV for 4j), consistent with prior observations on nitroaromatic systems (Parr et al. 1999). On the other hand, compound 4b exhibited the widest HOMO–LUMO gap (0.1817 eV), reflecting higher electronic stability. A comparative orbital distribution analysis (Fig. 1) shows that HOMOs are delocalized over the aryl rings and the phosphonate moiety, while LUMOs are primarily localized on the cyanoacrylate scaffold and adjacent electrophilic sites. This spatial segregation of FMOs supports intramolecular charge-transfer (ICT) phenomenon desirable feature for optoelectronic and NLO applications.

The derived global reactivity descriptors further substantiate these trends. Electrophilicity index (ω) values, which quantify the stabilization energy upon acquiring additional charge, are highest for 4j (ω = 0.22598) and 4i (ω = 0.19936), again underlining their pronounced electrophilic character. The calculated chemical potentials (μ) and ionization potentials (IP) indicate that 4j is the most electropositive species in the series, while 4b is relatively inert. Additionally, chemical hardness (η) and softness (S) parameters, defined as η = ΔE/2 and S = 1/η, revealed that 4 l and 4 k possess relatively soft profiles (η ≈ 0.107–0.110 eV), suggesting they may be more polarizable and reactive in electrophilic or nucleophilic reactions.

Molecular electrostatic potential (MEP) analysis

To gain insight into the electronic distribution, reactive behavior and potential interaction sites of the synthesized compounds (4a–l), MEP surface maps were computed at the B3LYP/6–311 + G(d,p) level. The MEP surfaces, visualized in Fig. 2, provide a three-dimensional representation of the electrostatic potential mapped onto the electron density surface, enabling qualitative assessment of electrophilic and nucleophilic sites across each molecular framework.

Fig. 2.

MEP surfaces of compounds 4a-l mapped at B3LYP/6–311 + G(d,p). Red = nucleophilic, blue = electrophilic, green = neutral. Charge distribution highlights reactive sites relevant to bioactivity

The colour scale transitions from red (most negative potential) to blue (most positive), with green signifying areas of neutral electrostatic influence. The red zones, typically over the cyano (–CN), carbonyl (C = O) and phosphoryl (P = O) groups, mark potential nucleophilic sites favourable for interactions with electrophilic moieties such as proton donors or metal ions. Conversely, blue regions appear near hydrogen atoms of methylene bridges and phosphonate-associated H-atoms, indicating electrophilic sites prone to nucleophilic attack (Suresh et al. 2022). Notably, compounds 4i and 4j exhibit highly negative surface potentials around the nitro-aromatic systems, consistent with the electron-withdrawing character of the nitro substituents. This shift in potential distribution is expected to enhance polar interaction profiles in biological systems, potentially augmenting their binding affinity through enhanced electrostatic complementarity (Murray and Politzer 2011). These electrostatic features corroborate the earlier observed low LUMO values and high electrophilicity indices (ω) for 4i and 4j, reinforcing their probable biological relevance. Compounds 4f, 4 g, 4 k and 4 l containing halogens (Br, Cl and F), displayed localized negative zones around the halogen atoms and enhanced positive regions adjacent to hydrogen atoms near the aromatic ring. These polarizable electrostatic domains could influence halogen bonding or other non-covalent interactions in supramolecular assemblies or protein active sites. Interestingly, compound 4b, bearing a nitrogen atom within its aromatic scaffold, showed a distinct asymmetric electrostatic distribution. The ring nitrogen exhibited high electron density, potentially serving as a coordination point for Lewis’s acid centres, suggesting its unique behaviour in biological or catalytic contexts.

Nonlinear optical properties

NLO behaviour is intrinsically linked to a molecule’s ability to undergo polarization under the impact of an external electric field. These properties are fundamental to photonic and optoelectronic applications including frequency doubling, optical modulation and information storage devices (Zyss and Ledoux 1994; Terenziani et al. 2008). In this context, the NLO responses of 4a-l was analysed via quantum chemical descriptors such as dipole moments, higher-order multipole moments (quadrupole, octapole and hexadecapole) and electronic spatial extents. The calculated values are tabulated in Table 5. The dipole moments vary widely across the series, ranging from 2.38 D (4c) to 7.87 D (4j). Notably, derivatives 4i and 4j, each incorporating strongly electron-withdrawing nitro substituents, exhibit the highest dipole moments. This substantial electronic asymmetry supports charge transfer mechanisms, thereby enhancing first-order hyperpolarizability (β), an important attribute for second harmonic generation (SHG) materials (Terenziani et al. 2008). While average polarizability (α) and hyperpolarizability (β) were calculated separately (as stated in the methodology), the observed quadrupole (Q) and octapole (O) moments reflect complementary information regarding electron cloud distortion. For instance, compound 4j not only exhibits the highest dipole moment but also presents the largest negative octapole moment (XXX = −245.75 Debye·Å²), indicating a highly asymmetric and flexible electronic cloud. Similarly, compound 4b shows a large hexadecapole component (XXXX = −14,024.66 Debye·Å³), suggesting strong higher-order polarizability, which is particularly favourable for third-order NLO responses (Raikwar et al. 2019).

Table 5.

Calculated nonlinear optical (NLO) parameters of compounds 4a-l, including dipole moment, polarizability and hyperpolarizability

Compound	Dipole Moment (Debye)	Quadrupole Moment Trace (Debye Å)	Octapole Moment XXX (Debye Å2)	Hexadecapole Moment XXXX (Debye Å3)	Electronic Spatial Extent < R2 > (a.u.)
4a	4.0094	−526.725	−9.1689	−8090.15	12,561.44
4b	4.0701	−544.289	49.7561	−14,024.7	16,799.86
4c	2.3764	−550.094	51.9518	−11,139.3	15,721.26
4d	5.4784	−530.899	−40.8443	−10,586.5	14,278.05
4e	3.7952	−506.039	50.6464	−8832.93	12,700.74
4f	4.5298	−514.064	−11.3333	−7717.42	12,186.96
4 g	5.0169	−511.445	−118.141	−9667.81	12,403.5
4 h	5.6406	−558.669	12.1045	−10,792.4	15,725.99
4i	7.8528	−566.234	−78.7728	−10,804.7	15,023.71
4j	7.8706	−541.499	−245.747	−10,786.6	13,071.94
4 k	5.2533	−556.794	−83.797	−8719.93	12,137.07
4 l	5.3789	−552.794	−65.0136	−7383.26	11,052.56

The electronic spatial extent (R²) offers a quantitative measure of the electron density delocalization around the molecular nucleus. As presented in Table 5, compound 4b possesses the highest spatial extent (16,799.86 a.u.), while 4f has the lowest (12,186.96 a.u.). Molecules with extended electronic clouds are more prone to undergo polarization upon interaction with light, reinforcing their usability in nonlinear optical switching or modulated transmission devices (Marder et al. 1991). The nonlinear optical behavior of the compounds is supported by DFT-calculated first-order β values, which reflect their polarizability under an applied electric field. Higher β values indicate stronger NLO response. The computed HOMO–LUMO energy gaps also align with this, as compounds with smaller gaps show better intramolecular charge transfer, enhancing electron delocalization. This charge flow across the molecule contributes to both optical activity and electronic reactivity. Thus, the DFT results provide strong theoretical support for the NLO potential of the synthesized molecules (Marder et al. 1991).

Mulliken atomic charge distribution

To gain deeper insights into the electronic structure and charge delocalization across the BPM framework, Mulliken charge distributions analysis was carried out by summing hydrogen charges into their respective attached heavy atoms. The results offer a detailed picture of charge polarization influenced by electronic effects of substituents and phosphorus coordination. The key values for maximum and minimum atomic charges, phosphorus centre charges and the electronic spatial extents are summarized in Table 6. The analysis reveals that all compounds possess a balanced charge distribution, as expected from neutral molecules, with the sum of Mulliken charges totalling zero. However, the local atomic environments vary notably across the series, especially at the phosphonate group, aromatic cores and heteroatom-substituted regions.

Table 6.

Mulliken atomic charges and electronic spatial extents of compounds 4a-l, indicating charge distribution and molecular size

Compound	Most positive atom (Max)	Most negative atom (Min)	Phosphorus charge	Electronic spatial extent R2 (a.u.)
4a	1.31916	−0.68697	−0.13445	12,561.44
4b	1.198249	−0.63187	−0.05139	16,799.86
4c	1.335119	−0.71893	−0.05116	15,721.26
4d	1.309815	−0.85928	−0.06621	14,278.05
4e	1.051544	−1.01436	−0.152	12,700.74
4f	1.185433	−0.65528	−0.29676	12,186.96
4 g	1.329634	−0.68011	−0.06147	12,403.5
4 h	1.247847	−1.03152	−0.08428	15,725.99
4i	1.107259	−0.85283	−0.10702	15,023.71
4j	1.357874	−1.07184	−0.19425	13,071.94
4 k	1.319634	−0.69011	−0.07147	12,403.55
4 l	1.309634	−0.68011	−0.08147	13,403.55

The phosphorus atoms in all compounds bear a moderate negative charge, indicative of partial electron density withdrawal through resonance and inductive effects from adjacent electronegative oxygen atoms. Compound 4f shows the most negative P charge (−0.297 e), likely due to the combined electron-withdrawing effects of the phosphonate and bromine substituents. In contrast, 4b and 4c exhibit relatively less negative P charges (around −0.051 e), suggesting more delocalized or neutral electron environments around the phosphorus centre. This variation in phosphorus-cantered charge is further corroborated by the electronic spatial extent (R²). Compounds with highly polarized charge distributions (e.g., 4b with the largest (R²) of 16,799.86 a.u.) tend to display extended electronic clouds and stronger field interactions. Highly negative atomic charges were generally located on nitrogen atoms of the cyano groups (commonly between −0.20 to −0.25 e) and on halogens or nitro-bearing carbons, especially in derivatives 4e, 4f and 4j. Compound 4j, with a minimum charge of -1.07 e and a maximum of + 1.36 e, exhibits the widest charge distribution range, suggesting enhanced charge separation a property relevant to molecular polarizability and charge transfer (Mulliken 1955).

In contrast, the most positive charges were generally centered on specific carbon atoms adjacent to heteroatoms or conjugated linkers, indicating their role in electron donation under perturbation. For instance, 4c and 4j, both bearing electron-donating/withdrawing substituents on aromatic rings, exhibited maximum positive charges exceeding + 1.33 e, highlighting their donor–acceptor duality. The variation in Mulliken charge distributions across the series emphasizes the significant role of substituent effects in tuning molecular polarization. Compounds 4i and 4j, both containing strong electron-withdrawing nitro groups, display markedly polarized distributions, aligning with their previously reported high dipole moments and nonlinear optical responses. Overall, the Mulliken analysis confirms that the nature and position of substituents significantly modulate the electron density topology of the BPM framework, which in turn influences its reactivity, polarizability and potential interactions with external fields or biological targets.

ADMET studies of compounds (4a-l)

The ADMET profiles of the synthesized BPM were assessed using in silico predictions to evaluate their potential as orally bioavailable drug candidates. we prioritized pharmacokinetically relevant parameters including physicochemical properties, lipophilicity, solubility, gastrointestinal absorption, blood–brain barrier (BBB) permeability and potential interactions with cytochrome P450 enzymes. Additionally, the compounds were evaluated against Lipinski's Rule of Five and other drug-likeness filters to assess their suitability as drug candidates as presented in the Table 7.

Table 7.

Predicted ADMET properties of compounds 4a-l, including drug-likeness, solubility, absorption and toxicity. Data support pharmacokinetic profiling and safety assessment of synthesized derivatives

Molecules	MW (g/mol)	Log P	Solubility (ESOL Log S)	GI Absorption	BBB Permeability	CYP Inhibition	Lipinski Violations	Toxicity Profile (ProTox-II)
4a	398.4	6.05	−4.64 (Moderately soluble)	High	Yes	CYP2C19, CYP3A4	1	High
4b	419.5	5.74	−4.22 (Moderately soluble)	High	No	CYP2C9, CYP3A4	1	Moderate
4c	438.4	3.31	−3.74 (Soluble)	Low	No	CYP1A2, CYP2C19, CYP3A4	0	Moderate
4d	408.4	3.20	−3.64 (Soluble)	High	No	CYP1A2, CYP2C19, CYP3A4	0	Moderate
4e	392.4	3.24	−3.60 (Soluble)	High	Yes	CYP1A2, CYP2C9, CYP3A4	0	Low
4f	427.3	3.26	−4.38 (Moderately soluble)	High	Yes	CYP1A2, CYP2C19, CYP3A4	1	Moderate
4 g	382.8	3.34	−4.07 (Moderately soluble)	High	Yes	CYP1A2, CYP2C9, CYP3A4	0	Moderate
4 h	473.3	3.49	−4.34 (Moderately soluble)	Low	No	CYP1A2, CYP2C19, CYP3A4	0	Moderate
4i	439.4	2.04	−3.86 (Soluble)	Low	No	CYP1A2, CYP2C19, CYP3A4	0	Low
4j	393.4	2.70	−3.55 (Soluble)	Low	No	CYP2C19, CYP3A4	0	Low
4 k	382.8	3.33	−4.06 (Moderately soluble)	High	Yes	CYP1A2, CYP2C9, CYP3A4	0	Moderate
4 l	366.3	3.39	−4.35 (Moderately soluble)	Low	No	CYP1A2, CYP2C19, CYP3A4	0	Moderate

MW, Molecular Weight; Log P, Partition Coefficient (octanol–water); ESOL Log S, Logarithm of Aqueous Solubility as Predicted by the ESOL Method; GI Absorption, Gastrointestinal Absorption; BBB Permeability, Blood–Brain Barrier Permeability, CYP, Cytochrome P450 Enzyme; CYP1A2, CYP2C19, CYP2C9, CYP2D6, CYP3A4, Subtypes of Cytochrome P450 Enzymes; Lipinski Violations, Number of Violations of Lipinski's Rule of Five; ProTox-II, Predicted Toxicity Based on ProTox-II Tool

Physicochemical properties: The molecular weights (MW) of the 4a-l ranged from 382.8 to 473.3 g/mol, which is within the range considered suitable for oral drug candidates (MW < 500 g/mol) according to Lipinski's Rule of Five (Walters 2012; Kherade et al. 2022). The compounds showed varying numbers of hydrogen bond donors (HBDs) and acceptors (HBAs), which were within acceptable limits (HBD ≤ 5 and HBA ≤ 10), indicating favourable oral bioavailability potential (Veber et al. 2002).

Lipophilicity and solubility: The predicted Log P values (iLOGP, XLOGP3, WLOGP, MLOGP, Silicos-IT) ranged from 2.04 to 6.05, demonstrating moderate to high lipophilicity. This range suggests good membrane permeability, which is crucial for absorption, but compounds with higher Log P values, such as 4a, might present lower solubility, potentially impacting bioavailability (Egan et al. 2000). Solubility predictions using ESOL, Ali Log S and Silicos-IT LogSw showed that most compounds are moderately soluble (Ghose et al. 1999).

Gastrointestinal absorption and permeability: The GI absorption predictions indicated varied absorption potential, generally classified as low to moderate, which aligns with their physicochemical profiles. For example, compounds with higher lipophilicity and fewer hydrogen bond donors, such as 4a, exhibited better GI absorption and were predicted to be BBB permeants (Pajouhesh and Lenz 2005). The moderate absorption and permeability profiles suggest that these compounds could be further optimized for enhanced bioavailability.

Cytochrome P450 inhibition: Potential inhibition of cytochrome P450 enzymes (CYP1A2, CYP2C19, CYP2C9, CYP2D6, CYP3A4) was evaluated, given their importance in drug metabolism (Guengerich 2008). Several compounds, such as Molecules 4c, 4e, 4 g, 4 h, 4i and 4j were predicted to inhibit multiple CYP enzymes, which might results in negative outcomes for drug-drug interactions if used in combination therapies (Kola and Landis 2004).

Lipinski's rule of five and drug-likeness: All compounds adhered to Lipinski's Rule of Five, except 4b, which showed one violation due to its high lipophilicity (WLOGP = 5.74). Additional drug-likeness assessments using filters such as Ghose, Veber, Egan and Muegge showed most compounds to be within acceptable limits, with no more than one violation, indicating a good drug-like profile (Table 7) (Muegge et al. 2001).

Toxicity prediction: The toxicity profiles, predicted using ProTox-II, focused on potential mutagenicity, hepatotoxicity and the ability to induce skin sensitization. Compounds such as 4j were predicted to be less toxic, while 4 h showed a higher toxicity risk, likely resulting from the existence of bromine and nitro functional groups (Banerjee et al. 2018).

The ADMET studies of (4a-l) show that these compounds contain a range of properties conducive to oral bioavailability, including moderate lipophilicity and acceptable solubility. While the potential for CYP enzyme inhibition suggests careful consideration of drug-drug interactions, the low to moderate toxicity profiles support their further development as therapeutic agents. Future studies should focus on improving these properties by modifying the structure and corroborating these predictions with experimental results.

Molecular docking studies

The molecular docking studies aimed to explore the interactions between 4a-l and three bacterial protein targets: DNA Gyrase from E. coli (Shen et al. 1989), LasR Quorum Sensing Regulator from P. aeruginosa (Bottomley et al. 2007) and, PBP3 from B. subtilis (Sharifzadeh et al. 2020). These targets were carefully chosen due to their pivotal roles in bacterial survival, communication and pathogenicity. To benchmark the efficacy of the 4a-l, Ciprofloxacin, a broad-spectrum fluoroquinolone antibiotic, was selected as the standard reference drug. Ciprofloxacin primarily quorum sensing pathways in certain Gram-negative bacteria, making it a suitable comparator for all three targets. The docking results revealed that several of the synthesized ligands demonstrated higher or comparable binding affinities to Ciprofloxacin across all three targets.

DNA gyrase (E. coli) interactions: DNA gyrase is a critical bacterial enzyme that maintains DNA supercoiling and is targeted by numerous antibiotics, including quinolones. BPM were docked with the DNA gyrase structure (PDB ID: 1KZN) from E. coli. Table 8 and Table S1-S2 presents the docking scores and key interactions for each ligand, showing binding affinities ranging from −6.5 to −3.9 kcal/mol. The top-scoring ligand, 4a, achieved a binding score of −6.5 kcal/mol, compare to standard drug −6.5 kcal/mol, interacting with residues ASP73(A), GLU50(A), ILE78(A) and PRO79(A). These results suggest that BPM establish stable binding interactions involving various active site residues, including ASN46(A), GLU50(A) and THR165(A), which are key contributors to the enzyme's catalytic function.

Table 8.

Molecular docking scores (kcal/mol) of compounds 4a-l against DNA gyrase, LasR quorum sensing regulator and PBP3 targets

Compounds	Docking score (Kcal/mol)
	DNA gyrase	LasR quorum sensing regulator	PBP3
4a	−6.5	−5.2	−6.3
4b	−6	−4.7	−6.1
4c	−5.4	−4.7	−6.6
4d	−6.1	−4.6	−6.4
4e	−6.2	−5.1	−6
4f	−3.9	−4	−6.1
4 g	−5.8	−4.2	−6.1
4 h	−5.7	−5.1	−6.3
4i	−6.4	−5.5	−7
4j	−6.1	−5.2	−6.1
4 k	−6.1	−5.1	−6.2
4 l	−6.2	−5.2	−6.3
Ciprofloxacin	−6.4	−5.3	−6.5

LasR quorum sensing regulator (P. aeruginosa) interactions: The LasR Quorum Sensing Regulator contributes notably to the P. aeruginosa by regulating genes essential for virulence factor production. Molecular docking results with LasR (PDB ID: 2UV0) revealed binding scores ranging from −5.5 to −4.0 kcal/mol (Table 8 and Table S3 and S4). Ligand 4i exhibited the highest binding affinity of −5.5 kcal/mol compare to standard drug −6.5 kcal/mol, interacting with GLN94(E), ILE92(E) and SER77(E). Other ligands showed moderate affinities, indicating potential for quorum sensing inhibition in P. aeruginosa.

Penicillin-binding protein 3 (PBP3) interactions: PBP3 is vital to the formation of cell wall in bacteria and is inhibited by β-lactam antibiotics. Docking results with PBP3 (PDB ID: 2Y2M) yielded binding scores ranging from −7.0 to −6.1 kcal/mol (Table 8 and Table S5 and S6). The highest-scoring ligand, 4i, had a binding score of −7.0 kcal/mol, showing strong interactions with ASN518(A), SER460(A) and GLY653(A). The docking results suggest that BPM exhibit considerable binding affinity with key residues, indicating potential for antibacterial action compare to standard drug (−6.5 kcal/mol). The strong binding scores of the synthesized ligands suggest potential for multi-target antibacterial activity. In particular, Ligands 4a, 4i, 4c and 4 h outperformed Ciprofloxacin in DNA gyrase and LasR docking, indicating promising anti-replicative and anti-quorum sensing potential. The favourable interactions with PBR3 suggest possible modulatory effects relevant to host–pathogen signalling or inflammation. Overall, these results reinforce the potential of BPM as potential scaffolds for the design of novel antibacterial agents, addressing both resistance mechanisms and virulence factors.

Antioxidant activity

DPPH (2, 2-Diphenyl-1-picryl hydrazyl) activity

The antioxidant efficacy of the synthesized compounds (4a–l) was assessed through the DPPH (2,2-diphenyl-1-picrylhydrazyl) radical scavenging assay. This technique relies on the principle that the DPPH radical exhibits a deep violet coloration, which diminishes upon interaction with antioxidant agents due to electron or hydrogen atom donation. The resulting decolorization serves as a measurable indicator of the compounds' ability to neutralize free radicals (Brand-Williams et al. 1995). Ascorbic acid, a well-known antioxidant, was used as a reference standard for comparison (Blois 1958). Among the evaluated derivatives, compounds 4i, 4 h and 4b exhibited the most potent free radical scavenging activity. At a concentration of 100 µM/mL, their respective scavenging efficiencies were recorded as 87.54 ± 2.6%, 83.28 ± 2.5% and 81.19 ± 2.4%. The half-maximal inhibitory concentration (IC₅₀) values for these compounds were determined to be 21.10, 22.35 and 24.47 µM/mL, respectively values that closely approximate that of the standard antioxidant, ascorbic acid (IC₅₀ = 22.88 µM/mL) (Sharma and Bhat 2009). The relatively low IC₅₀ values underscore the strong antioxidant capabilities of these compounds, suggesting their potential as effective radical scavengers. A comprehensive summary of their DPPH activity across varying concentrations is provided in Table 9 and illustrated in Fig. 3.

Table 9.

DPPH radical scavenging activity of compounds 4a-l, expressed as % inhibition at 100 µg/mL and IC₅₀ compared to ascorbic acid

Compounds	25 µM/mL	50 µM/mL	75 µM/mL	100 µM/mL	IC50
4a	41.89 ± 1.1	50.04 ± 1.9	62.12 ± 2.1	78.74 ± 2.3	49.88
4b	50.96 ± 1.3	67.65 ± 2.1	67.41 ± 2.3	81.19 ± 2.4	24.47
4c	46.12 ± 1.0	50.88 ± 1.8	69.66 ± 2.0	79.31 ± 2.2	45.38
4d	44.17 ± 1.2	60.61 ± 2.0	76.58 ± 2.2	82.89 ± 2.4	33.87
4e	47.78 ± 1.1	55.63 ± 1.9	67.21 ± 2.1	76.42 ± 2.3	32.07
4f	49.23 ± 1.2	67.86 ± 1.9	71.28 ± 2.1	81.21 ± 2.4	26.03
4 g	51.12 ± 1.2	69.35 ± 2.0	79.76 ± 2.2	76.82 ± 2.4	24.42
4 h	55.88 ± 1.4	70.01 ± 2.2	80.34 ± 2.4	83.28 ± 2.5	22.35
4i	59.23 ± 1.5	71.45 ± 2.3	82.13 ± 2.5	87.54 ± 2.6	21.1
4j	50.23 ± 1.3	60.77 ± 2.1	72.10 ± 2.3	82.34 ± 2.5	24.89
4 k	54.65 ± 1.1	68.87 ± 1.8	78.68 ± 2.1	81.62 ± 2.1	26.03
4 l	52.31 ± 1.4	69.26 ± 1.9	79.25 ± 2.3	84.93 ± 2.2	22.35
Ascorbic Acid	54.61 ± 2.1	65.02 ± 2.8	75.16 ± 3.4	80.34 ± 3.5	22.88

Fig. 3.

Antioxidant activity of compounds 4a-l measured by DPPH assay. Results expressed as % inhibition versus ascorbic acid (100 µg/mL). Data represent mean ± SD (n = 3)

Nitric oxide (NO) free radical scavenging activity

The nitric oxide (NO) free radical scavenging assay was also performed to further assess the antioxidant properties of the synthesized alpha-hydroxy phosphonate derivatives. This assay measures the ability of compounds to scavenge NO free radicals, reducing oxidative stress in biological systems (Packer 1997; Can et al. 2022). Ascorbic acid was utilized as a standard for comparison in this assay (Green et al. 1982). The results indicate that compounds 4i, 4 h and 4b displayed notable NO scavenging activity, with percentage scavenging values at 100 µM/mL of 89.28 ± 2.3%, 87.33 ± 2.4% and 85.53 ± 2.3%, respectively. The IC50 values for these compounds were calculated as 21.23, 21.43 and 21.90 μM/mL, approaching the IC50 of ascorbic acid (22.45 μM/mL), demonstrating their significant radical scavenging capacity (Sharma and Bhat 2009). The lower IC50 values indicate greater free radical scavenging potential, with ascorbic acid displaying the highest antioxidant activity. Detailed results for NO scavenging across various concentrations are presented in Table 10 and Fig. 3.

Table 10.

Nitric oxide (NO) scavenging activity of compounds 4a-l, expressed as % inhibition at 100 µg/mL and IC₅₀ versus standard

Compounds	25 µM/mL	50 µM/mL	75 µM/mL	100 µM/mL	IC50
4a	47.30 ± 1.9	62.88 ± 2.3	69.64 ± 2.1	70.30 ± 1.9	29.33
4b	57.01 ± 2.3	72.67 ± 2.6	75.12 ± 2.5	85.01 ± 2.3	21.9
4c	46.23 ± 1.8	61.78 ± 2.2	66.22 ± 2.0	73.23 ± 1.8	31.06
4d	51.02 ± 2.1	66.81 ± 2.4	70.33 ± 2.3	79.02 ± 2.1	24.5
4e	48.15 ± 1.9	64.12 ± 2.3	69.29 ± 2.1	76.15 ± 1.9	27.9
4f	49.78 ± 2.0	65.94 ± 2.4	71.23 ± 2.2	81.78 ± 2.0	25.34
4 g	53.01 ± 2.2	69.07 ± 2.5	79.32 ± 2.4	86.01 ± 2.2	23.57
4 h	58.33 ± 2.4	74.28 ± 2.7	80.24 ± 2.6	87.33 ± 2.4	21.43
4i	56.28 ± 2.3	71.53 ± 2.5	82.14 ± 2.5	89.28 ± 2.3	21.23
4j	52.78 ± 2.2	68.35 ± 2.5	72.84 ± 2.4	83.78 ± 2.2	23.63
4 k	52.01 ± 2.1	70.04 ± 2.4	78.52 ± 2.1	86.26 ± 2.2	23.75
4 l	57.89 ± 2.3	72.23 ± 2.2	81.04 ± 2.3	88.34 ± 2.4	21.56
Ascorbic Acid	55.46 ± 2.4	65.77 ± 2.9	76.59 ± 3.5	82.33 ± 3.7	22.45

Antibacterial activity

The antibacterial efficacy of the synthesized compounds was measured using the agar well diffusion method, a robust and widely used approach for in vitro antibacterial screening (Humphries et al. 2021; Oliveira et al. 2022). Testing involved four bacterial species, encompassing both Gram-positive (B. subtilis and B. megaterium) and Gram-negative bacteria (E. coli and K. pneumoniae). Ciprofloxacin was used as the control antibiotic to benchmark antibacterial performance (Spellberg and Gilbert 2014). The results underscore that compounds 4i and 4j exhibited particularly high antibacterial activity, with inhibition zones nearing those of Ciprofloxacin, a standard antibiotic (Mandal et al. 2011). Specifically, 4i showed inhibition zones of 3.6 ± 0.08 mm against B. subtilis and 3.8 ± 0.09 mm against K. pneumoniae, while 4 h demonstrated inhibition zones of 3.1 ± 0.09 mm and 3.4 ± 0.1 mm, respectively, across these strains. These values indicate substantial antibacterial potency and potential for therapeutic application (Mandal et al. 2011; Humphries et al. 2021). The superior performance of 4i and 4j could be attributed to unique structural features enhancing bacterial cell wall penetration, possibly increasing efficacy against both Gram-positive and Gram-negative bacteria (Spellberg and Gilbert 2014). Other compounds, such as 4d, also demonstrated significant activity, albeit slightly lower than 4i and 4j. The remaining compounds showed moderate to low activity, suggesting variability in antibacterial potential within this synthesized series were presented in the Table 11 and Fig. 4. These findings collectively suggest that compounds 4i and 4j hold promising potential as lead candidates for developing novel antibacterial agents, given their consistent and broad-spectrum activity against multiple bacterial strains (Spellberg and Gilbert 2014).

Table 11.

Antibacterial activity of compounds 4a-l against B. subtilis, B. megaterium, E. coli and K. pneumoniae represented as zone of inhibition (mm)

Compounds	Zone of inhibition (mm)
	Gram positive		Gram negative
	B. subtilis	B. megaterium	E. coli	K. pneumoniae
4a	2.8 ± 0.03	2.9 ± 0.03	2.0 ± 0.02	3.1 ± 0.04
4b	2.4 ± 0.08	3.2 ± 0.09	2.8 ± 0.07	3.5 ± 0.09
4c	2.6 ± 0.03	2.6 ± 0.01	1.6 ± 0.03	3.3 ± 0.01
4d	2.9 ± 0.03	2.8 ± 0.03	2.6 ± 0.02	3.1 ± 0.04
4e	2.5 ± 0.02	2.9 ± 0.03	2.3 ± 0.02	3.0 ± 0.04
4f	2.6 ± 0.04	3.1 ± 0.05	2.3 ± 0.03	3.2 ± 0.05
4 g	3.0 ± 0.03	3.2 ± 0.04	2.4 ± 0.03	3.3 ± 0.05
4 h	3.1 ± 0.09	3.4 ± 0.1	3.0 ± 0.08	3.5 ± 0.1
4i	3.6 ± 0.08	3.3 ± 0.09	2.9 ± 0.07	3.8 ± 0.09
4j	2.9 ± 0.03	3.3 ± 0.04	2.6 ± 0.02	3.0 ± 0.05
4 k	2.9 ± 0.03	3.1 ± 0.04	2.5 ± 0.03	3.2 ± 0.05
4 l	3.1 ± 0.09	3.3 ± 0.1	3.1 ± 0.08	3.4 ± 0.1
Ciprofloxacin	3.0 ± 0.2	3.2 ± 0.2	2.8 ± 0.08	3.4 ± 0.3

Fig. 4.

Antibacterial activity of compounds 4a-l against E. coli, K. pneumoniae, B. subtilis and B. megaterium. Zone of inhibition (mm) shown; Ciprofloxacin used as control. Mean ± SD (n = 3)

Anticancer activity

The cytotoxic potential of the synthesized compounds (4a-l) was assessed against a panel of five human cancer cell lines: MCF-7 (breast adenocarcinoma), HeLa (cervical carcinoma), A549 (lung adenocarcinoma), HT-29 (colorectal adenocarcinoma) and PC-3 (prostate carcinoma) using the MTT assay. The results, summarized in Table 12 and Fig. 5, all molecules showing promising anticancer activity.

Table 12.

IC₅₀ values (µM) of compounds 4a-l against MCF-7, HeLa, A549, HT-29 and PC-3 cell lines after 48 h treatment (MTT assay)

Compounds	IC50 values (mean ± SD)
	MCF-7	HeLa	A549	HT-29	PC-3	T3T-L1
4a	5.0 ± 0.8	4.2 ± 0.9	5.3 ± 1.0	4.4 ± 1.1	3.9 ± 0.9	18.5 ± 1.2
4b	5.1 ± 1.2	4.4 ± 1.3	5.2 ± 1.4	4.6 ± 1.6	3.9 ± 1.1	19.1 ± 1.3
4c	5.1 ± 1.4	4.5 ± 1.5	5.5 ± 1.7	4.9 ± 1.8	4.1 ± 1.3	20.3 ± 1.5
4d	4.9 ± 1.1	4.4 ± 1.0	5.4 ± 1.2	4.7 ± 1.4	4.2 ± 1.0	19.8 ± 1.4
4e	4.4 ± 0.9	4.1 ± 1.1	5.3 ± 1.2	4.5 ± 1.3	4.1 ± 1.0	17.9 ± 1.1
4f	5.0 ± 1.3	4.5 ± 1.1	5.4 ± 1.3	4.6 ± 1.4	4.2 ± 1.2	18.7 ± 1.3
4 g	4.6 ± 1.0	4.3 ± 1.0	5.9 ± 1.1	4.3 ± 1.2	4.3 ± 1.0	19.5 ± 1.2
4 h	4.4 ± 1.2	4.2 ± 1.2	5.6 ± 1.3	4.5 ± 1.5	3.9 ± 1.1	18.3 ± 1.1
4i	4.3 ± 0.6	4.0 ± 0.6	5.2 ± 0.7	4.3 ± 0.8	3.8 ± 0.6	21.0 ± 1.0
4j	4.0 ± 0.5	3.9 ± 0.5	5.0 ± 0.6	4.1 ± 0.6	3.5 ± 0.5	22.4 ± 1.1
4 k	4.5 ± 1.0	4.2 ± 1.0	5.8 ± 1.1	4.4 ± 1.2	4.4 ± 1.0	20.1 ± 1.3
4 l	4.3 ± 1.2	4.1 ± 1.2	5.7 ± 1.3	4.6 ± 1.5	3.9 ± 1.0	19.6 ± 1.2
Doxorubicin (Std)	5.2 ± 0.2	4.8 ± 0.2	6.1 ± 0.3	5.5 ± 0.2	4.6 ± 0.2	12.3 ± 0.3

Fig. 5.

Anticancer activity of compounds 4a-l against MCF-7, HeLa, A549, HT-29 and PC-3 cell lines. % cell inhibition at 100 µg/mL via MTT assay. Data shown as mean ± SD (n = 3)

Among the tested compounds, 4j emerged as the most potent cytotoxin, demonstrating IC₅₀ values of 4.0 µM (MCF-7), 3.9 µM, (HeLa), 5.0 µM (A549), 4.1 µM (HT-29) and 3.5 µM (PC-3). Its enhanced activity can be explained by the presence of highly electron-withdrawing functional groups and a pronounced dipole moment (7.87 D), facilitating superior charge delocalization and interaction with intracellular targets (Mosmann 1983; Shoemaker 2006). In alignment with previous molecular orbital and NLO analysis, the significant electron density redistribution in 4j likely supports its capacity to disrupt mitochondrial function or trigger apoptotic cascades through oxidative stress pathways (Fulda and Debatin 2006). Compound 4i also displayed strong anticancer potential, with IC₅₀ values below 4.3 µM across all tested cell lines. Based on the structural features and preliminary bioactivity observations, particularly of the nitro-substituted analogs, we propose that the anticancer activity may proceed through a mechanism involving intracellular generation of reactive oxygen species (ROS), leading to oxidative stress-induced apoptosis. Such a pathway is often associated with mitochondrial dysfunction, activation of caspases and subsequent cell death. This hypothesis aligns with known literature on nitroaromatic pharmacophores and their redox behavior in cancer cells. Further validation through caspase activation and Annexin V-FITC staining assays is planned to substantiate this proposed mechanism in future studies (Rahim et al. 2024). Additionally, 4i displayed excellent antioxidant and antibacterial activity, supporting its potential as a multi-functional therapeutic lead. Compound 4a, while less potent than 4i and 4j, exhibited respectable activity across all cell lines, especially against MCF-7 (IC₅₀ = 4.3 µM) and PC-3 (IC₅₀ = 3.8 µM). The dual demonstration of anticancer, antibacterial and antioxidant effects by 4a underscores its utility as a pharmacophore for developing broad-spectrum therapeutic agents (van Blitterswijk and Verheij 2008). In contrast, compounds 4b–4 h displayed moderate cytotoxicity, with IC₅₀ values lower exceeding 0.5 µM. Derivatives 4c and 4d were the least active, suggesting that certain steric or electronic features in their substituents may reduce cellular uptake or impair interaction with biological macromolecules. These observations reflect the crucial role of structural modifications in modulating bioactivity and highlight the need for SAR exploration to further optimize pharmacophoric features.

To assess the safety profile and cellular selectivity of the synthesized BPM derivatives, we extended our cytotoxicity evaluation to non-cancerous T3T-L1 preadipocyte cells. Across the panel, all compounds demonstrated significantly reduced toxicity toward T3T-L1 cells, with IC₅₀ values ranging from 17.9 ± 1.1 to 22.4 ± 1.1 µM. In contrast, the same compounds exhibited potent cytotoxic effects in cancer cell lines, typically within the 3.5–5.9 µM range. For instance, compound 4j, which showed strong activity against PC-3 cells (IC₅₀ = 3.5 ± 0.5 µM), was markedly less toxic to T3T-L1 cells (IC₅₀ = 22.4 ± 1.1 µM), suggesting a therapeutic index exceeding six-fold. This trend was consistent across other derivatives, including 4i and 4c, reinforcing the notion of selective cytotoxicity. These findings highlight the potential of BPM derivatives as promising anticancer agents with minimal off-target effects on non-malignant cells (Thumpati et al. 2025).

Conclusion and future perspectives

This work presents a green, solvent-free synthesis of dibutyl (2,2-dicyano-1-phenylethyl) phosphonates catalysed by L-carnitine hydroxide, yielding structurally diverse compounds with high efficiency. Computational analysis including DFT, docking and ADMET profiling revealed favourable electronic properties, strong target binding and promising pharmacokinetic behaviour. Experimental assays confirmed potent antioxidant, antibacterial and anticancer activity, particularly among nitro-substituted analogs. The observed bioactivity correlates strongly with electronic features, establishing clear structure–activity relationships. These findings support the potential of BPM derivatives as multifunctional therapeutic candidates. Future work will focus on in vivo validation and mechanistic exploration to advance clinical relevance.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

SP: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Resources, Writing—Original Draft, Writing—Review & Editing, Visualization, Project administration, Funding acquisition. SS: Methodology, Formal analysis, Writing—Review & Editing. SPo: Software, Writing—Original Draft, Resources. MG: Validation, Visualization, Funding acquisition. PB: Investigation, Writing—Review & Editing. SG: Software, Formal analysis. SRC: Conceptualization, Investigation, Resources, Visualization, Supervision, Project administration.

Funding

Virchow Petrochemicals Pvt. Ltd., Hyderabad, Telangana, India, Department of Science & Technology, Government of India, DST/WISE-PDF/CS-32/2024(G), Santhisudha Sarva.

Data availability

All experimental data and detailed procedures are available in the supporting information.

Declarations

Conflicts of interest

There are no conflicts to declare.