A density functional theory study of cyclophosphamide and purinethol adsorption on a covalent triazine framework (CTF-2) for drug delivery applications

Abstract

The conventional delivery systems are ineffective in therapeutics because of poor adsorption at target site. Conversely, 2D sheet based materials are the advanced drug delivery systems that provide decent adsorption and good attraction with drug molecules and can be used as a carrier of effective drug delivery. In this case, we show two-dimensional (2D) porous covalent triazine framework (CTF-2) as a Drug Delivery System to cyclophosphamide and purinethol (Anticancer Drugs). The CTF-2 cavity interacts with both the cyclophosphamide (CP) and purinethol (PU) drugs. The interactions of the drug with the surface are stabilized by the calculated binding energies favoring the adsorption of the drug molecules on the CTF-2 surface. The geometric and electronic properties of the both drugs inside the cavity of CTF-2 are used to confirm the interactions among the drugs and their potential delivery. The adsorption energies of CP@CTF-2 and PU@CTF-2 complexes are − 1.04 eV and − 0.82 eV. However, the BSSE corrected energy values of CP@CTF-2 and PU@CTF-2 complexes are − 1.02 eV and − 0.80 eV. The outcome of the NCI and the QTAIM analysis has given the explanation that the van der Waals interactions are an important factor in the stabilization of the CP@CTF-2 and PU@CTF-2 complexes. The transfer of charge between the drug and the carriers is explained through the NBO and EDD analysis. FMO analysis showed that HOMO-LUMO gap of the CP@CTF-2 complex (6.92 eV) and the PU@CTF-2 complex (6.62 eV) were both smaller than that of the bare CTF-2 surface (7.10 eV). Dipole moment (µ) and pH can help one to understand the interaction of the drug molecules with the carrier and the surrounding environment especially in drug delivery. These studies shows how these drugs can be loaded and off-loaded using the CTF-2 carrier. This study demonstrates the potential of CTF-2, which is an innovative and promising drug delivery carrier in the treatment of various diseases.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-36405-5.

Introduction

Cancer, one of the most detrimental diseases, is the leading global cause of death^1–5. Cancer is affecting millions of peoples every year. However, recent advancements in diagnostic and treatment approaches have resulted in a decline in this number^6–9. For this purpose, several therapies, such as hormone-therapy, radio-therapy, and chemo-therapy, have been introduced to treat cancer¹⁰. One of the frequently used methods for the treatment of cancer is chemotherapy¹¹. However, the adverse effects of chemotherapeutic medicine in different tissues¹² include low platelet count, swelling of the throat, blurred vision, anemia, and breast swelling. In tumor cells, anticancer medications have cytotoxicity to healthy or normal cells, less water solubility, and low drug concentration¹³. Moreover, various antineoplastic drugs have been studied to determine their effectiveness in destroying cancer cells^14,15. The most extensively employed classes of anti-cancer drugs are cyclophosphamide (CP) and purinethol (PU) and their derivatives. The 2D structures of cyclophosphamide and purinethol are shown in Fig. 1.

Fig. 1.

Structure of cyclophosphamide and purinethol.

Cyclophosphamide (CP) is a DNA-alkylating agent belonging to the oxazaphosphorines group. CP was synthesized by Arnold and Bourseaux, who performed its early stage clinical trial for cancer treatment in 1958^16–18. In 1959, CP became the eighth FDA approved cytotoxic anti-cancer drug. CP is used to treat various solid tumors, pediatric malignancies, and hematologic cancers. Its therapeutic applications include chronic myeloid leukemia, Hodgkin’s lymphoma, breast cancer, uterine cancer, and small cell lung cancer, Burkitt’s lymphoma, endometrial cancer, and neuroblastoma. Additionally, it effectively manages conditions such as renal disorders like corticosteroids-resistant nephrotic syndrome and multiple sclerosis and focal segmental glomerulonephritis^19–21. It is a chemotherapeutic drug and also considered as a nitrogen mustard oxazaphosphorine compound. CP synthesis was based on the idea of employing enzymatic processes to convert non-toxic pro-drug into its active form (phosphoramide mustard (PM)), predominantly within the cancer cells. This approaches aimed to overcome the severe toxicity of nitrogen mustard²². Cyclophosphamide (CP) as a prodrug is mostly in the neutral form of the molecule when circulating in the system because of its stability at physiological pH (~ 7.4). The main role of this neutral form is to cause interactions with the CTF-2 nanocarrier during this transport process before arriving to the target site and hepatic metabolic activation. Therefore, the neutral state of the system is an important concept that can be modeled to determine the drug@carrier interactions required to analyze adsorption, stability and release properties²³.

Purinethol (PU), or mercaptopurine is a protein-like molecular structure containing nitrogen and sulfur atoms, which enable interaction with various substances²⁴. PU is a widely recognized drug used for the therapy of various cancers and disorders, including Acute Promyelocytic Leukemia (APL), Acute Lymphocytic Leukemia (ALL), ulcerative colitis, and Crohn’s disease^2,24–26. Although PU has illustrated effectiveness in patient treatment, serious side effects such as liver toxicity, vomiting, bone marrow suppression, increase the risks of future cancer, and loss of appetite may limit its widespread prescription for individual²⁷. Optimizing PU could significantly improve its therapeutic-approaches, resulting in more favorable and the consistent treatment outcomes²⁸. Mercaptopurine exhibits pH-dependent protonation-behavior arising from its purine framework containing multiple heteroatoms. Based on prior experimental and theoretical investigations, protonation preferentially occurs at the N1 site under acidic-conditions, yielding the N1-H⁺ species as the most stable form, while minor protonation at N7 or N9 may exist with low population. Therefore, in this study, the N1 protonated form of PU was selected to model the interaction with the CTF-2 nanocarrier, representing the dominant species in biologically-relevant conditions^29,30.

These anti-cancer drugs have therapeutic properties. However, the high dose of these drugs can harm normal cells. Traditional chemotherapy approaches often lack specificity, leading to the damage of healthy cells alongside targeting cancerous melanotic cells^31,32. Therefore, during the targeted drug delivery system (DDS), to be able to achieve the effective treatment, it is important to regulate the optimal plasma concentration of the anti-neoplastic drugs. This approach allows the administration of the right dose depending on the needs of the patient. There by minimizing adverse effects and maximizing therapeutic outcomes^33,34. In nanotechnology, recent advancements have led to exciting improvements in disease diagnosis and treatment^35,36. The nanoparticle—based drug delivery (NDDS) system offer precise control and accuracy in targeting. These systems are designed to improve drug delivery directly to cancer cells while minimizing the risk of toxicity to normal or healthy cells³⁷. In recent times, different type of drug delivery systems are reported for the drug molecules as nanocarrier such as liposomal^38–40, silver and gold⁴¹, dendrimers⁴², and carbon nanotubes⁴³. In general, for drug delivery systems, 2D materials demonstrate great potential due to their larger surface area for drug loading^44–46. In literature, several reports that explain successful 2D material applications in DDS. More recently, phosphorene and hexagonal boron nanosheets has been theoretically use for fluorouracil and mercaptopurine in drug delivery systems⁴⁷. Whereas another DFT (Density Functional Theory) study examines fluorinated graphene oxide as a carrier for camptothecin and doxorubicin drugs as a delivery system^48–50.

CTF (covalent triazine framework) consists of triazine rings, usually composed through the polymerization of organic linkers, which may include nitrile and amine groups. These triazine units are covalently bonded, making rigid and stable structure with appreciable porosity^51–53. This study aims to examine the interactions between CTF-2 surface and anti-cancer drugs (CP and PU). CTF-2 is a novel type of nitrogen-rich porous organic polymers, recognized for their large surface area and stability⁵⁴. It consists of a highly porous network of triazine rings associated by cyanuric chloride linkage. The larger pore size and higher surface area compared to other CTFs make CTF-2 suitable for applications that require higher adsorption capacities, such as drug delivery systems (DDS). Interestingly, CTF-2 has π-conjugated structure and nitrogen sites that ensure good points of interaction with drugs by adsorption which guarantees stability and also effectiveness of controlled release⁵⁵. These properties arise from their effective drug-adsorption and controlled-release capabilities, which are required for enhancing the therapeutic effectiveness of the anti-cancer drugs⁵⁶. The nitrogen-rich framework in CTF-2 surface assets strong binding with drug molecules. Additionally, the capacity to functionalize CTF-2 carrier improve its efficiency and specificity in delivering drugs (CP and PU) to target site^57,58. To the best of our knowledge, this work represents the first theoretical investigation of CP and PU drug adsorption, stability, and release behavior on the CTF-2 surface using combined DFT and MD approaches, highlighting the novelty and contribution of this study^59–61.

DFT calculation have been conducted to study selectivity and adsorption of drugs (cyclophosphamide and purinethol) on CTF-2 surface, including Geometrical Optimization, Non-Covalent Interactions (NCI), Symmetry Adapted Perturbation Theory (SAPT0), Electron Density Differences (EDD), Natural Bond Order (NBO), Quantum Theory of Atoms in Molecule (QTAIM), and Frontier Molecular Orbital (FMO) analyses.

Computational methodology

All DFT calculations are carried out by using the software Gaussian 09⁶². At the ωB97XD functional and 6-31G (d, p) basis set, geometries through CTF-2 surface and Drug@CTF-2 complexes are optimized. The functional ωB97XD is range separated and is considered optimal for the Non-Covalent interactions⁶³. The ω parameter in ωB97XD controls range separation, which effectively minimizes self-interaction error⁶⁴. The initial structure of CTF-2 used in this study was adopted from experimentally reported covalent triazine frameworks available in the literature^58,60,65. Geometry optimizations were performed using DFT, and the stability of each optimized structure was verified through frequency calculations to ensure that the obtained geometries represent minima on the potential energy surface. As this work focuses on computational analysis, no additional experimental characterization of CTF-2 was conducted. Future work will incorporate experimental verification to further support the computational predictions.

For all the complexes, electronic properties, including electron density difference (EDD), natural bond order (NBO), and frontier orbital theory (FMO), are investigated at the same level of theory, ωB97XD/6-31G (d, p)^64,66–69. On the CTF-2 surface, the drug molecules (cyclophosphamide and purinethol) are adsorbed through various orientations to attain the most stable geometric configuration of the complexes. The geometrical and structural parameters of the complexes are visualized via Chemcraft 1.6 and GaussView 5.0 software^70,71. The interaction energies of Drug@CTF-2 are calculated as,

1

In the above equation, energy of the Drug@CTF-2 complex is E_complex, the energy of CTF-2 is E_CTF−2, and the drug energy is E_Drug. Whereas the counterpoise method is utilized to investigate the BSSE (Basis set superposition error) corrected interaction energies to ensure the accuracy of interaction energy.

2

In the above equation, the E_int.CP is counterpoise interaction energy, E_int is interaction energy, and BSSE. The NCI analysis is frequently applied to investigation the nature of interactions among the drug and CTF-2 carrier. The 2D RDG (reduced density gradient) scatter plot and 3D iso-surfaces are visualized via GIMP 2.10.38, VMD 1.9.3 and GNU plots software’s^72,73. On two variables, ρ (electronic density) and 2D RDG, the Non-Covalent Interactions (NCI) index is dependent⁷⁴. These factors are interrelated via mathematical equation.

3

The nature of Non-Covalent Interactions is characterized by the sign and magnitudes of electron density. The presence of either accumulation or diminishment of electron density (ρ) at a particular point is determined via the sign of Laplacian. The negative and large values of (λ₂)ρ shows the electrostatic interactions in the region of RDG, the large and positive values are represents the steric repulsive forces in the red spikes region, and in the green color projection small and positive values shows the van der Waals interactions^76–78.

The Analysis (SAPT0) is computed via PSI4 software⁷⁹. This analysis comprises four contributions, electrostatic (E_elstat), exchange-repulsion (E_exch), induction (E_ind), and dispersion (E_disp), expressed as,

4

In this equation, exchange represents repulsive interaction due to Pauli exclusion, while electrostatic, induction, and dispersion contributions are attractive and stabilize the complex⁶⁰.

The QTAIM (Quantum Theory of Atoms in Molecules) analysis provides us the nature of bonding interactions. This analysis is performed via VMD and Multiwfn 3.7 software^72,80. The QTAIM analysis is performed via different types of parameters, including BCPs (Bond Critical Points), CCPs (Cage Critical Points), RCPs (Ring Critical Points), and NCPs (Nuclear Critical Points).

5

6

7

The nature of Non-Covalent Interactions (NCI) is analyzed via the BCPʹs. Various topology parameters, including Laplacian of electronic density (∇²ρ), electronic density (ρ), total electron energy H(r), kinetic G(r), and potential energy density V(r) are employed at bond critical points to determine the type of interactions^81–83.

Electron Density Differences (EDD) analysis is conducted to visualize the charge transfer, and the Natural Bond Order (NBO) analysis is applied to investigate the charge shifting between the iso-surfaces of drug and carrier by using the software Chemcraft 1.6^71,84. The Frontier Molecular Orbital (FMO) analysis is performed via GaussView 5.0 software. FMO analysis is conducted to explore the alterations in electronic parameters.

8

Here, Egap is the gap between HOMO-LUMO energy, E_HOMO is the highest occupied molecular orbital, and E_LUMO is the lowest occupied molecular orbital. The HOMO-LUMO energy gap is used as a qualitative measure of electronic activity and charge transfer capability of the molecular systems. The present DFT calculations are performed under non-periodic boundary conditions using the Gaussian package, periodic band structure or transport analyses were not included^85,86.

Results and discussion

Geometry optimization and interaction energies

CTF-2 is a highly porous network of triazine rings (C₃N₃) which are connected by aromatic linkers. This study investigates the adsorption of two anti-cancer drugs (CP and PU), on CTF-2 surface to examine their potential on drug delivery system. The CTF-2 surface and structure are shown in Fig. 2.

Fig. 2.

Optimized geometries of CTF-2 carrier (A), and drugs cyclophosphamide (B), purinethol (C).

The most stable geometries of complexes are shown in Fig. 3, and the least stable optimized geometries of the both complexes are represented in Fig. S1. The most stable optimized geometries for CP@CTF-2 and PU@CTF-2 show that drug are slightly away from central cavity due to the electron rich side of the CTF-2 surface. In CP@CTF-2, the carbon atom C1 interacts with H11 at the distance of 2.77 Å, and H10 shows interactions with H2, C3, and N4 atoms of surface at the distances of 2.39 Å, 2.86 Å, and 2.00 Å. The H5 interact with O9 at the distance of 2.43Å, and H8 shows interactions with two carbon atoms (C6 and C7) of the cavity at the distances of 2.85Å and 2.81Å. In PU@CTF-2, the hydrogen atom H1 of the cavity shows interactions with S6 and H5 at the distance of 2.84 Å and 2.33 Å and N2 interacts with H5 at the distance of 2.23Å. The H3 interacts with H6 at the distances of 2.63, and C3 interacts with H5 at the distance of 2.94 Å, respectively.

Fig. 3.

Structures of optimized geometries of complexes CP@CTF-2 and PU@CTF-2.

To realistically evaluate the outcomes of interactions, bond length and interaction energy are two significant aspects⁸⁷. While the drug molecule approaches CTF-2 carrier, the bond length reduces, facilitating the formation of a stronger complex. The interaction energies for the CP@CTF-2 and PU@CTF-2 complexes are − 1.04 eV and − 0.82 eV, respectively (see Table 1). The interaction energy of the most stable orientation cyclophosphamide@CTF-2 is significantly stronger than that of the purinethol@CTF-2 complex. The most stable CP@CTF-2 complex is achieved due to the shorter bond length of 2.00 Å (N4···H10) among the nitrogen atom of CTF-2 surface and the hydrogen atom of cyclophosphamide drug.

Table 1.

Interactions, bond length and BSSE energies of optimized geometries of CP@CTF-2 and PU@CTF-2 complexes with ωB97XD/6-31G (d, p) level of theory.

Interactions	Bond length (Å)	Adsorption energy (eV)	Eint (solvent) (eV)	Eint (Drug@Graphene) (eV)
Cyclophosphamide@CTF-2				CP@Graphene
C1−−H11 H2−−H10 C3−−H10 N4−−H10 H5−−O9 C6−−H8 C7−−H8	2.77 2.39 2.86 2.00 2.43 2.85 2.81	− 1.04	− 0.77	− 1.08
Purinethol@CTF-2				PU@Graphene
H1−−S6 H1−−H5 N2−−H5 H3−−H6 C3−−H5	2.84 2.33 2.23 2.63 2.94	− 0.82	− 0.61	− 0.99

In this work, solvent effects are also considered to assess the adsorption behavior of drug molecules on the optimized CTF-2 structure. The interaction energies of cyclophosphamide (CP) and purinethol (PU) in the solvent phase are calculated as − 0.77 eV and − 0.61 eV, respectively, indicating favorable interactions in an aqueous environment. The adsorption energies of CTF-2 were compared to the energies of interaction of graphene to put the adsorption behavior of the CTF-2 in perspective. Graphene is found to have a higher absolute adsorption energy with both CP (− 1.08 eV) and PU (− 1.99 eV) although the difference is mainly due to the fact that graphene has a much larger surface area that is plain and therefore a much higher number of carbon atoms are involved in the dispersion-driven stabilization.

These interaction and BSSE corrected energy values for CP@CTF-2 and PU@CTF-2 are shown in Table 2. Geometry optimizations are performed using the 6-31G (d, p) basis set for computational efficiency. The BSSE corrected energy values of CP@CTF-2 and PU@CTF-2 complexes are − 1.02 eV and − 0.80 eV. To validate the energies and minimize basis set superposition error (BSSE), single-point calculations are carried out with the larger def2-TZVP basis set. The resulting interaction energies at BSSE are − 1.01 eV in CP and − 0.79 eV in PU at CTF-2 which is nearly similar to the 6-31G (d, p) energy and this proves the accuracy and reliability of the computational approach taken. Further, thermodynamic parameters are also considered with the aim of offering an understanding into the stability and viability of the adsorption of drugs. Gibbs free energies (ΔG_ads) and contributions to entropy are calculated using thermochemical corrections giving − 0.36 eV and − 0.21 eV values of CP@CTF-2 and PU@CTF-2, respectively. These findings affirm that the adsorption is thermodynamically favorable and this validates the possibility of CTF-2 as an effective carrier of targeted drug delivery.

Table 2.

BSSE-corrected interaction energies, thermochemical corrections, Gibbs free energies, and entropy of CP@CTF-2 and PU@CTF-2 that show the stability of drug@CTF-2 interaction.

Complexes	BSSE corrected interaction (eV)	BSSE corrected interaction (def2-TZVP) (eV)	Thermo-chemical correction (eV)	Gibbs free energy (ΔG_ads) (eV)	Entropy (eV)
CP@CTF-2	− 1.02	− 1.01	− 0.96	− 0.36	− 0.98
PU@CTF-2	− 0.80	− 0.79	− 0.68	− 0.21	− 0.71

Comparison of adsorption energies of anti-cancer drug delivery frameworks

Interaction energies of the CP@CTF-2 (− 23.94 kcal/mol) and PU@CTF-2 (− 18.89 kcal/mol) complexes were compared with previously reported adsorption energies of anticancer drugs on the different nanoparticles using alternative DFT methods. Notably, cyclophosphamide (CP) on C₃Al and 6-mercaptopurine (6-MP, also known as Purinethol) on Al₁₂N₁₂ and BN(6,6) exhibited adsorption energies of − 31.4, − 82.3, and − 3.2 kcal/mol, respectively. Similarly, gemcitabine (GB) and flutamide (FLT) on CTF-1 showed adsorption energies of − 28.1 and − 26.5 kcal/mol, while mercaptopurine (MP) and thiotepa (TEP) on CTF − 0 displayed − 25.3 and − 22.7 kcal/mol, respectively. Strong interactions were also reported for fluorouracil (FU) and nilotinib (NU) on C₂N with adsorption energies of − 28.1 and − 27.5 kcal/mol, whereas comparatively weaker adsorption was noted for NU@BC₂ (− 20.3 kcal/mol) and NU@B₄₀ (− 25.1 kcal/mol) (see Table 3). These comparisons indicate that CP and PU exhibit adsorption strengths consistent with the reported range for similar drug-nanocarrier systems, supporting their potential for stable loading and controlled release in targeted drug delivery applications.

Table 3.

Comparison of interaction energies of complexes with reported literature data on anti-cancer drug delivery systems.

Drug delivery system (Drug@CTF-2)	DFT functional	Interaction energies (kcal/mol)
CP@C3Al	–	− 31.488
6-MP@Al12N12	–	− 82.389
6-MP@BN(6,6)	B3LYP	− 3.290
MP@CTF-0	B3LYP	− 25.360
TEP@CTF-0	B3LYP	− 22.760
GB@CTF-1	PBE0-D3	− 28.191
FLT@CTF-1	PBE0-D3	− 26.591
NU@C2N	M06-2x	− 27.51
FU@C2N	M06-2x	− 28.11
NU@BC3	PW91	− 20.392
NU@B40	PBE0-D3	− 25.193

Non-covalent interaction (NCI) analysis

The NCI analysis is frequently applied to explore the nature of non-bonding interactions, including repulsive forces, hydrogen bonding, and van der Waals interactions among drugs (cyclophosphamide, purinethol) and the CTF-2 surface⁹⁴. The NCI analysis comprises 2D-RDG (Reduced Density Gradient) plots and 3D iso-surfaces of the complexes⁹⁵. The 2D-RDG graphs are derived by plotting the sign(λ₂)ρ on x-axis against the Reduced Density Gradient (RDG) on the y-axis. In sign(λ₂)ρ, the sign(λ₂) provides the information about interaction type, and the ρ indicates the strength of bonding. In 3D iso-surfaces, the color patches are also dependent on the sign(λ₂)ρ value. The 2D RDG graph and 3D iso-surfaces are presented in Fig. 4.

Fig. 4.

2D RDG scatter plot and 3D iso-surfaces of CP@CTF-2 (A), and PU@CTF-2 (B) complexes.

2D RDG scatter plots comprise of red, green, and blue color spikes in 0.05 a.u. to − 0.05 a.u. range. The blue color spikes in the range of − 0.02 a.u. to − 0.05 a.u. indicate the existence of strong hydrogen bonding or electrostatic interactions. The green color spikes represents the weak interactions, including dipole − dipole or London dispersion forces in the region of 0 a.u. of (λ₂)ρ between drug molecules and CTF-2 surface. In the region of (λ₂)ρ > 0.01 a.u., appearance of red color spikes shows steric repulsive forces⁹⁶.

In complex PU@CTF-2, 3D iso-surfaces show a predominance of green regions, indicating weak van der Waals interactions. These attractive forces were observed via Sulphur and nitrogen-containing groups of PU. In the 2D-RDG scatter plots of this system, the green color spikes appears in the range of (λ₂)ρ ≈ −0.01 a.u., indicating the existence of weak London dispersion force, and positive values of red color projections show the repulsive interactions. The blue color spikes represent strong hydrogen bonding at the x-axis of the 2D RDG scatter graph. While the complex CP@CTF-2 shows more stability as compared to the PU@CTF-2 complex. Overall, the RDG and 3D iso-surface analyses clearly illustrate the nature and strength of non-covalent interactions in both complexes.

Quantum theory of atom in molecule (QTAIM) analysis

In the Quantum Theory of Atoms in Molecules (QTAIM), all non-covalent interactions are analyzed via Bader’s theory, which has been applied to various molecular systems⁸². This analysis is widely acknowledged for its capability to describe a range of inter and intra-molecular interactions, including ionic bonding, hydrogen bonding, and van der Waals interactions. Various parameters of topology, such as the Laplacian of electronic density (∇²ρ), electronic density (ρ), total electron energy H(r), kinetic G(r), and potential energy density V(r) are employed at BCPs (bond critical points) to assess the nature of interactions^97–101.

For shared shell interactions (covalent bond interactions), electronic density (ρ) values must be positive and greater than the 0.1 a.u., meanwhile, values of ∇²ρ are always negative and large. For closed shell interactions, ρ values must be less than 0.1 a.u, and the values of ∇²ρ (Laplacian) are positive and small^102,103. For Non-Covalent Interactions, the total electronic density H(r) is always less than 0, and the ratio of V/G must be less than 1.

To analyze the QTAIM results, all complex geometries are optimized using the same level of theory ωB97XD/6-31G (d, p). In Table S2, results of topological properties attained via QTAIM analysis for both complexes are reported. 3D topology parameters of QTAIM analysis for Cyclophosphamide@CTF-2 and Purinethol@CTF-2 complexes are shown in Fig. 5 at the same level of theory.

Fig. 5.

3D topology parameters of QTAIM analysis for Cyclophosphamide@CTF-2 and Purinithol@CTF-2 complexes.

In complex CP@CTF-2, a total of 11 BCPs are observed, including five C−−H, three H−−O, two H−−N, and one H−−H bond interaction. The ∇²ρ values range from 0.008 to 0.068 a.u. and the values of ρ vary in the range of 0.002 to 0.027 a.u., respectively. Similarly, in PU@CTF-2 complex, total 7 BCPs are examined, including two H−−S, two C−−C, one N−−H, one C−−N, and one H−−C bond interactions. The ∇²ρ values vary in the range of 0.013 to 0.047 a.u. and values of ρ are in the range of 0.004 to 0.017 a.u., respectively. The QTAIM parameters confirm the presence of weak Non-Covalent interaction between the drug molecules and the CTF-2 carrier. These results collectively indicate relative interaction strengths and confirm that both drugs effectively absorbed within the carrier, with PU show stronger binding compared to CP.

Symmetry adapted perturbation theory (SAPT0) analysis

According to Jeziorski and colleagues, the SAPT0 method is a powerful theoretical model applied to measure intermolecular interactions with high precision¹⁰⁴. This theory allow in-depth understanding of the nature of interactions which controls molecular recognition and adsorption. The interaction energies are divided by SAPT0 into four, the electrostatic (Eelstat) and exchange-repulsion (Eexch) energies, the induction (Eind) and the dispersion (Edisp) energies.

In the case of the CP@CTF-2 complex, the attractive contributions are mainly due to the electrostatic (– 0.98 eV) and dispersion (– 0.96 eV) terms, and this implies that charge-driven interactions are significant factors that stabilize the complex and van der Waals forces (see Table 4). The stabilization is moderately contributed by the induction energy (– 0.28 eV). Unsurprisingly, the repulsive interaction between CP and the CTF-2 framework is explained by the exchange component (1.35 eV).

Table 4.

SAPT0 energy decomposition analysis for CP@CTF-2 and PU@CTF-2 complexes.

Complexes	Eelstat (eV)	Eexch (eV)	Eind (eV)	Edisp (eV)	Total SAPT0 (eV)
CP@CTF-2	– 0.98 (43.92%)	1.35	– 0.28 (12.71%)	– 0.96 (43.37%)	– 0.87
PU@CTF-2	– 0.82 (42.99%)	1.12	– 0.21 (10.72%)	– 0.89 (46.27%)	– 0.79

In the case of the PU@CTF-2 complex, dispersion (– 0.89 eV) is the most significant attractive force, with electrostatic interactions (– 0.82 eV) coming in a close second and induction (– 0.21 eV) providing a significant but weaker stabilizing effect (see Fig. 6). The exchange repulsion (1.12 eV) is lower compared to the CP complex, reflecting reduced steric crowding; however, the overall attractive contributions are also weaker for PU.

Fig. 6.

Graphical representation of the energy components calculated from the SAPT0 analysis of CP@CTF-2 and PU@CTF-2 complexes.

The total SAPT0 energies, calculated as − 0.87 eV for CP@CTF-2 and − 0.79 eV for PU@CTF-2, confirm that cyclophosphamide exhibits stronger binding and forms a more stable adsorption complex with the CTF-2 carrier than purinethol.

Electronic properties

Electron density difference (EDD) and natural bond orbital (NBO) analyses

Electron Density Differences analysis is performed to visualize the charge transfer between the iso-surfaces of drug and the CTF-2 surface. The iso-surfaces of drug@CTF-2 obtained by EDD analysis are shown in Fig. 7. The NBO analysis is used to investigate the charge release between the components of drug@CTF-2. The positive values of the NBO charges indicate that, the charge transfer from drug to surface^105,106. In Table 5, values of NBO charges are reported. In EDD analysis, the red and blue color iso-surfaces indicates the accumulation and depletion of the electronic densities.

Fig. 7.

EDD plots for CP@CTF-2 and PU@CTF-2 complexes with iso-value = 0.0003 a.u.

Table 5.

The results of FMO, NBO charges and dipole moment of complexes CP@CTF-2 and PU@CTF-2.

Complex’s	HOMO energy	LUMO energy	Energy gap (eV)	NBO charge (e−)	Dipole moments (Debye)
CP@CTF-2	− 7.97	− 1.05	6.92	− 0.016	5.35
PU@CTF-2	− 7.64	− 1.03	6.62	0.006	5.49
CTF-2	− 8.82	− 0.98	7.10	−	0.00
CP	−	−	−	−	6.01
PU	−	−	−	−	5.94

In the, CP@CTF-2 and PU@CTF-2 complexes, NBO charge values of the CP is − 0.016 e⁻, while for PU is 0.006 e⁻, respectively. In chlorophosphamide@CTF-2, the negative sign in NBO value shows that the charge shifted from surface CTF-2 to the drug molecule, while in purinethol@CTF-2, positive sign of NBO value reveals that the charge shifted from drug molecule to the surface CTF-2.

In the case of PU@CTF-2, higher charge values due to the shorter interaction distance indicate the stronger attractions among PU and CTF-2 as compared to the CP@CTF-2 complex. In the CP@CTF-2 complex, the blue color iso-surfaces are arrived on Nitrogen and Oxygen atoms of the drug, which represents the accumulation of electronic densities. While the red color iso-surfaces mainly appeared on Carbon and nitrogen of the carrier surface, which shows the depletion of the charge densities. This represents the charges are transfers from surface CTF-2 to CP drug molecules.

Alternatively, in PU@CTF-2 complex, the blue iso-surfaces are predominantly concentrated on the nitrogen atoms on the surface which represent charge accumulated, whereas the red iso-surfaces on the drug represent charge depleted. This is a clear evidence of charge transfer on the PU drug to the CTF-2 surface. The analysis of the EDD highlights a strong electronic communication between the carrier and the drug in the complex of CP@CTF-2. It is notable that the localization of the color is limited to the adsorption region, which indicates that charge redistribution occurs mainly at the drug@carrier interface and therefore it is important to note that noncovalent interactions are essential in stabilizing the system. This is to confirm that, the significant redistribution of electronic density is carried out especially at the binding site which confirms the involvement of charge transfer to stabilize the adsorbed drugs. Such localized features are consistent with the nature of noncovalent interactions governing drug@carrier stabilization.

Frontier molecular orbital (FMO) analysis

The Frontier Molecular Orbital analysis is performed to understand the change in electronic parameters. The drug@CTF-2 and bare surface energy values of HOMO (high occupied molecular orbital), LUMO (low unoccupied molecular orbital), and E_gap (energy gap) are shown in Table 5. Orbital densities of CP@CTF-2 and PU@CTF-2 complexes are shown in Fig. 8. The conductivity of material is analyzed via variation in band gap, the reduction in energy gap leads to increased reactivity and conductivity, and the large energy gap decreases the stability and conductivity¹⁰⁷.

Fig. 8.

Orbital densities of CP@CTF-2 (A) and PU@CTF-2 (B) complexes via FMO analysis (Iso-value = 0.02 a.u).

For the CP@CTF-2 complex, the HOMO energy value is − 7.97 eV, and the LUMO energy value is − 1.05 eV, respectively. For PU@CTF-2 complex, the HOMO and LUMO energies are − 7.64 eV and − 1.03 eV, respectively. The energy gaps after complexation of the CP and PU with CTF-2 are 6.92 eV and 6.62 eV, respectively. For the bare surface CTF-2, the energies of HOMO and LUMO are − 8.82 eV and − 0.98 eV, and the E_gap is 7.1 eV, respectively. The studied complexes observed a decrease in energy gap compared to the bare surface CTF-2.

Upon complexation of CTF-2 surface with the drug CP, the energy of HOMO increases from − 8.82 eV to − 7.97 eV and the energy of LUMO decreases from − 0.98 eV to − 1.05 eV. Furthermore, the E_gap is decrease from 7.1 eV to 6.92 eV. Similarly, in the PU@CTF-2 complex, the energy of HOMO increases from − 8.82 eV to − 7.64 eV and the energy of LUMO decreases from − 0.98 eV to − 1.03 eV. The E_gap (Energy gap) decreases from 7.1 eV to 6.62 eV.

In complex CP@CTF-2, the HOMO electron density is shifted away from the surface, where it is interacting with the drug, while the LUMO electron density is shifted towards the surface. In complex PU@CTF-2, the orbital density of HOMO is located via the drug, whereas in LUMO, the orbital density is located over the CTF-2 surface. This represents the charge distribution between the drug and carrier. Overall, in FMO analysis, the results indicate that the greater reduction in the energy gap for PU@CTF-2 compared to CP@CTF-2 signifies stronger electronic interactions and enhanced charge transfer, suggesting a higher affinity of the CTF-2 surface for PU over CP.

Dipole moment (µ) analysis

In the dipole moment analysis, drug solubility and target release efficiency are evaluated by change in its magnitude before and after complexation. The pristine CTF-2 framework exhibits a dipole moment of 0.00 D due to its symmetric architecture which results in complete cancellation of internal dipoles. Upon adsorption, this symmetry is disrupted by drug@surface interactions, leading to notable polarization in both complexes. For cyclophosphamide, the dipole moment decreases from 6.01 D to 5.35 D, while in purinethol, shift of 5.94 to 5.49 D is observed after adsorption. These changes in dipole moment compared to bare surface and drugs, reflects the enhanced polarization induced by its interaction with the framework¹⁰⁸. These variations arise from shifts in electron density at the binding interface, which generate new localized dipole within the complexes. The resulting polarity is significantly higher than that of the bare CTF-2 surface, supporting improved compatibility with aqueous biological media, the slightly large dipole moment of PU@CTF-2 suggests a more pronounced polarization effect, consistent with its comparatively stronger interaction with the CTF-2 carrier.

Drug release (pH effect)

The drug delivery process is the most essential step to transfer the drug from CTF-2 surface to targeted cells. The normal blood cells pH is in the range of 7.35–7.45. Although the pH level in the surroundings of malignant cells is below 6^82,109–112. Thus, we investigate the pH effect on studied complexes CP@CTF-2 and PU@CTF-2. In an acidic environment, DFT simulations were conducted on the CP and PU drugs loaded on surface CTF-2 (covalent triazine framework). The atoms at the ends of the surface CTF-2 were protonated, followed by structure optimization at the ωB97XD/6-31G (d, p) level of theory. The decrease in adsorption energies and increase in interaction distances observed which are crucial for the efficient release of drugs from carrier surface.

Conclusion

This research systematically examined the prospects of the CTF-2 framework as a delivery system of anti-cancer therapy drugs cyclophosphamide and purinethol using Density Functional Theory. The adsorption behavior, structural configuration and charge transfer between the drug molecules and CTF-2 surface are examined critically. The corrected BSSE interaction energies show that the CP@CTF-2 (− 1.04 eV) is better bonded than PU@CTF-2 (− 0.82 eV), which implies that the CP complex is more stable. Solvent effect is considered using SMD model, and the adsorption energies in water remain favorable (− 0.77 eV for CP@CTF-2 and − 0.61 eV for PU@CTF-2). Thermochemical correction including Gibbs free energies and entropy contribution are calculated, yielding ΔG_ads values of − 0.96 eV and − 0.68 eV, with entropy contribution of − 0.98 eV for CP@CTF-2 and − 0.71 eV for PU@CTF-2, confirming that adsorption is thermodynamically favorable under solvated conditions. Non-covalent interaction (NCI) and QTAIM analyses confirm the presence of weak but significant interactions stabilizing both complexes. Natural bond orbital (NBO) analysis shows small charge transfer from CTF-2 to the drugs, with − 0.016 e⁻ for CP@CTF-2 and 0.006 e⁻ for PU@CTF-2. FMO analysis indicate a slight reduction in the HOMO-LUMO gap for the complexes (6.92 eV for CP@CTF-2 and 6.62 eV for PU@CTF-2) compared to the bare surface (7.1 eV), reflecting increased electronic stability upon drug adsorption. Dipole moment of the complexes increased to the zero in the pristine surface of 5.35 D and 5.49 D of CP and PU of CTF-2 and PU, respectively, which demonstrated stronger polarity and compatibility at aqueous solutions. All these studies point to the loading and offloading of drug molecules using the CTF-2 carrier as being more effective.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

T.T.: Visualization, methodology, data curation, formal analysis, writing-original draft. M.Y.: Conceptualization, data curation, methodology, supervision, writing-original draft. I.B.: Software, resources, writing-review and editing. N.A.: Data curation, formal analysis, software, project administration, funding, writing-review and editing.

Funding

This work was supported by the Deanship of Scientific Research, Vice Presidency for Graduate Studies and Scientific Research, King Faisal University, Saudi Arabia [Grant No. KFU251633].

Data availability

All data generated or analysed during this study are included in this published article [and its supplementary information files].

Declarations

Competing interests

The authors declare no competing interests.