Due to the inadequacy of traditional medications and treatment methods, millions of people worldwide are dying. There has always been a tradeoff between accumulating drugs in the target cell and achieving the desired distribution, solubility, and selectivity (Mortazavifar et al., 2019). There is a possibility of accumulation of drug carriers at the target site when passively targeting the pathological state of cancer cells, but effectiveness is not guaranteed. Such problems have significantly been solved by advancing drug delivery capabilities using new generations of drugs (Emanet et al., 2017; Ahmad et al., 2010). The recent development of nanocarrier-based medicines has opened a new window to a more efficient therapy method as they can be widely manipulated, resulting in greater selectivity and lower toxicity. Nanotubes are perfect nanocarrier candidates for enhancing the therapeutic effects of drugs because their wide inner and outer surface area and corresponding internal volume fundamentally open many doors for the functionalization and even encapsulation of effective drug molecules since their endcaps can be removed (Iijima, 1991; Ferreira and Sousa, 2016).

Genistein (Gen), one of the natural and active soy isoflavones, has attracted great scientific interest due to its antitumor properties. Gen exhibits various pharmacological activities through its cancer prevention and anticancer efficacy in various preclinical model systems at low and high concentrations. Previous studies, in both in vivo animal tests and in human clinical trials, have shown that when consumed at low concentrations, similar to those achieved through dietary consumption, Gen inhibits the promotion of cancer cell migration, detachment, and invasion. While at high concentrations, it suppresses several proteins associated with tumorigenesis and apoptosis (Pavese et al., 2010). The broad anticancer properties of Gen include its effectiveness in inhibiting matrix metalloproteinase type 2 (MMP-2) and type 6 (MMP-6) (Huang et al., 2005; Xu et al., 2019). It also prohibits protein tyrosine kinase (PTK) and angiogenesis-promoting variables such as vascular endothelial growth factors (VEGF), which trigger the expression of the breast tumor suppressor protein BRCA. Additionally, it inhibits prostate-specific antigen (PSA) (Tyagi, Song, and De, 2019). In addition to reducing the chemoresistance and radioresistance of cancer cells, Gen is known for its antioxidant properties (Banerjee et al., 2008), which help in the treatment and control of various types of cancer, including breast cancer, prostate cancer, ovarian cancer, and colon cancer (Tyagi, Song, and De, 2019; Huang et al., 2005; Chen et al., 2020; Lee et al., 2012).

The theoretical prediction of boron nitride nanotubes (BNNTs) was made in 1994, which led to their synthesis in 1995 (Rubio et al., 1994; Blase et al., 1994). The tremendous properties of BNNTs include mechanical strength, chemical and thermal stability, electrical insulation, and biocompatibility, making them potential for numerous applications including pharmaceuticals, chemistry, high-temperature technologies, and cosmetics (Han, 2008; Lipp et al., 1989). Although they are structurally similar to carbon nanotubes (CNTs), BNNTs are considered better candidates than CNTs for various applications due to their superior properties. BNNTs are semiconductors with a wide band gap of around 5.5 eV, while CNTs can be either semiconductors or metals based on the chirality of the nanotubes (Ahmad et al., 2015). Moreover, BNNTs demonstrate higher resistance to oxidation and fewer toxic effects than CNTs in previous investigations, which makes them a preferable choice for biomedical applications (Kostoglou et al., 2020; Simon et al., 2019; Kakarla and Kong, 2022). In the context of the biological application of BNNTs, they provide promising advances in medical imaging, molecular biology, and biomedical technologies as well as theoretically supporting the identification and treatment of a range of diseases, including cancer (Ciofani et al., 2009; Ciofani et al., 2013). Although BNNTs have become a controversial figure due to their effectiveness and toxicity, many recent experimental studies suggest that they have little adverse side effects on the response of cells and organisms (Ferreira et al., 2015); for example, Chen et al. (2009) have shown that it delivers DNA into cells efficiently and without apparent toxicity.

While gold and silver nanoparticles have been studied as carriers of Gen (Ahmad et al., 2010), no research has examined BNNTs as carriers for them. When exploring a new carrier-biomaterial system, it is wise to assess the system’s properties computationally before the experimental phase to save time and money. BNNTs have diverse applications in various cancer therapy methods, including hyperthermia, boron neutron capture therapy, gene transfection, and photothermal therapy (Ferreira and Sousa, 2016). As the length of a BNNT is positively correlated with its cytotoxicity (Merlo et al., 2018), we limited the BNNT length to no more than 10Å in this study (Kakarla and Kong, 2022; Kim et al., 2018). Doping BN nanotubes with Fe induces magnetic properties, making Fe-doped BNNTs suitable for hyperthermia and suggesting the potential of this system as an efficient drug in a multitherapeutic approach (Ciofani et al., 2009; Zhi et al., 2005). Fe can be doped in either the B-site or N-site and the magnetic moment is higher when doping a Fe atom at the B site. Doping Fe atoms in BNNTs increase their chemical reactivity compared to pure BNNTs due to their higher adsorption energy, as shown by electronic analyses. While a doped Fe atom improves adsorption performance, this atom has the greatest local reactivity and is the most reactive site. In addition to having a lower formation energy and significantly larger binding energy compared to N-site doping, the formation energy and binding energy of a Fe atom-substituted BNNT at the B site are lower, indicating a more stable structure. Among BNNTs substituted with transition metals at the B site, Fe doping achieves the shortest bond length between the doped atom and neighboring N atoms and the highest magnetic moment (Farmanzadeh and Rezainejad, 2016; Ferreira et al., 2015; Jalalinejad et al., 2021; Li et al., 2011; Thapa et al., 2021).

The study used Density Functional Theory (DFT) calculations with the Dmol3 module in Materials Studio software version 23 (Basiuk and Henao-Holguín, 2013). Dmol3 is commonly employed for various DFT calculations of transition metal-functionalized BNNTs (Zhang et al., 2012). This research focused on single-walled armchair BNNTs with three different chirality indices: (5,5), (6,6), and (7,7), which were doped with a Fe atom at the B site. These nanotubes contained 80, 96, and 112 atoms, with respective diameters of about 6.88 Å, 8.25 Å, and 9.63 Å. Additionally, the BNNTs selected had a length of approximately 7.48 Å and an average B-N bond length of 1.44 Å. Both ends were saturated with hydrogen atoms to prevent dangling bonds. The shorter length of the nanotubes provides a more favorable environment for functionalization in biomedical applications (Juárez et al., 2013).

The exchange-correlation function was estimated using the generalized gradient approximation (GGA) and the Perdew–Burke–Ernzerhof functional (PBE) (Perdew et al., 1996; Lee and Martin, 1997). GGA is preferred over local density approximation (LDA) for its more accurate estimation of interaction energies (Liang et al., 2009). For improved accuracy, unrestricted DFT with dispersion (DFT-D) developed by Grimme was employed (Delley, 2002). The DFT semi-core pseudopotentials (DSPP) with the double numerical polarization (DNP) basis set were utilized (Luo et al., 2014). The DNP basis set, available in Dmol3, is comparable to the 6-31G (d,p) and 6-31G** Gaussian basis sets and is considered one of the best basis sets (Wang et al., 2017; Hosseinzadeh et al., 2020). The maximum energy tolerance, maximum force, and maximum displacement convergence during geometry optimization were set at 10 − 6 Ha, 0.004 Ha/Å, and 0.005Å, respectively (Warrag et al., 2020; Hatab et al., 2020; Lemaoui et al., 2021). To ensure accuracy, the multipolar expansion analysis was employed in the calculations for both gas and solution phases. For considerations in a biological context, the dielectric constant of water was used for the aqueous environment. Consequently, the conductor-like screening model for real solvation (COSMO-RS) in Dmol3 was applied for solubilization analysis (Klamt and Schüürmann, 1993).

Our proposed Gen/FeBNNT complex is formed by conjugating an active Gen molecule with the Fe atom doped in the BNNTs structure through a single covalent bond. The charge distribution, analyzed using the Hirshfeld method, and the density of states (DOSs) of the Gen/FeBNNT are discussed. To analyze the electronic properties, we calculate the energy gap ( E g ) of the different configurations using E H O M O and   E L U M O , as follows in Equation 1: E g = E L U M O − E H O M O (1)

Where E H O M O and   E L U M O represent the highest energy level of the occupied molecular orbital and the lowest energy level of the unoccupied molecular orbital, respectively.

The chemical reactivity and stability properties of the complex were investigated in both solution and gas phases by calculating the electrochemical potential (µ) (Klamt and Schüürmann, 1993), global hardness (η) (Chattaraj and Roy, 2007), and electrophilicity index (ω) (Sheela et al., 2014) as shown in Equations 2–4: µ = 1 2 E L U M O + E H O M O (2) η = 1 2 E L U M O − E H O M O (3) ω = µ 2 2 η (4)

The stability of Gen/FeBNNT complexes in the solvent model is calculated through Equation 5: E s o l v a t i o n = E a q u e o u s − E g a s (5)

Where E s o l v a t i o n , E a q u e o u s , and E g a s are the solvation energy of the complex, the total energy of the complex in the aqueous phase, and the total energy of the complex in the gas phase, respectively.

Equation 6 yields the adsorption energy of the optimized structure of the investigated complex (E_ads) as follows: E a d s = E G e n / F e B N N T − E F e B N N T + E G e n (6)

Where E G e n / F e B N N T is the energy of optimized complex, E F e B N N T and E G e n are the energies of the BNNT doped with a Fe atom and Gen molecule, respectively.

In the present study, we have proposed a new complex involving boron nitride nanotubes and a biomolecule for treating cancer through multiple methods. We used DFT calculations to study boron nitride nanotubes with armchair structure and chirality indices of (5,5), (6,6), and (7,7). These nanotubes act as carriers for the anti-cancer biomolecule, genistein. We also studied the effects of introducing iron (Fe) into the nanotubes, which enhanced their electronic and magnetic properties and facilitated bonding with the genistein molecule. We investigated three different configurations of covalent bonding between the active oxygen in genistein and the doped Fe atom, labeled C1, C2, and C3. We examined the proposed complexes in both solution and gas phase. Our findings indicated that the adsorption of genistein on the nanotube surface led to a reduction in the band gap. This reduction subsequently led to a higher excitation wavelength in the infrared region, pointing to the exceptional potential of the complex for photothermal treatment. Moreover, the inclusion of Fe resulted in the addition of a magnetic moment to the structure, potentially reducing the necessary magnetic field application to the human body in hyperthermia treatment and enabling more precise cell targeting. Notably, the studied complex also presents a significant advantage in providing an adequate dose of boron atoms for boron neutron capture therapy, underscoring its potential as a multi-therapeutic treatment.

The comprehensive analysis of the quantum molecular descriptors of the complexes revealed that the (7, 7) C3 and (6, 6) C3 configurations exhibited the greatest stability and the least chemical reactivity in both phases. All interactions were found to be exothermic and stable, with high adsorption energy values for the mentioned complexes. The solvation energy comparison showed that (7, 7) genistein-nanotube complexes were the most soluble structures. The total density of states (DOS) of the complexes displayed a reduction in the band gap after the interaction, which is in agreement with the energy gap values calculated by the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO). The sigma profile from COSMO-RS files showed that the hydrophilic behavior and dispersion of the complexes in water were enhanced after interaction with genistein molecules. Ultimately, the rigorous analysis led to the unequivocal conclusion that the (7, 7) C3 complex emerges as the optimal genistein carrier within the biological system. Further experimental investigations are encouraged to synthesize and characterize the proposed complex.