Insight into the Phase Inversion of Myristic Acid In Situ Gels for Drug Delivery Applications

Abstract

In situ gels (ISGs) are well-known as smart drug delivery systems due to their inherent gel-forming ability and controlled drug release properties. This study presents comprehensive molecular-level insights into the physical properties and phase transformation processes of the formulation of 10 to 60% w/w myristic acid (MYR)-based ISGs using ethanol (EtOH), N-methyl-2-pyrrolidone (NMP), and dimethyl sulfoxide (DMSO) as solvents, investigated using conventional experimental techniques, molecular dynamics (MD) simulations, and density functional theory (DFT). The results showed that 40% w/w myristic acid in NMP (NM40) exhibited swift gel formation with low water tolerance to induce phase inversion into the matrix. Scanning electron microscopy revealed that NM40 had multilayer, sheet-like structures, in which NM40 displayed a denser and less porous topography than 40% w/w myristic acid in DMSO (DM40). NM40 also had efficient antimicrobial efficacy against various microbes; thus, NM40 was the appropriate ISG as an antimicrobial drug delivery system. Moreover, MD simulation demonstrated that during the initial stage of gel formation, MYR began to agglomerate and arrange in an orderly manner, resembling crystallization. A higher concentration of MYR promoted a higher compactness of the MYR structure and the lower solvent exchange rate. Additionally, DFT calculations demonstrate that hydrogen bonding is the key interaction, contributing to the orderly arrangement of molecules. These significant findings pave the way for the tailoring and optimization of ISG formulations in drug delivery systems.

1. Introduction

In situ gels (ISGs) have gained attention as drug delivery systems due to their ability to enable minimally invasive drug administration for localized treatment at targeted therapeutic sites. − Solvent exchange-induced ISGs serve as an injectable drug delivery system, where active ingredients are dissolved in biocompatible solvents. Upon administration into an aqueous environment, the solution undergoes a phase transition into a gel or matrix-like state due to the simultaneous influx of antisolvent (aqueous fluid) and efflux of solvent. Consequently, the entrapped drugs are gradually released from the formed matrix over time. −

Among the various materials used in ISG formulations, saturated fatty acids, including phospholipids, cholesterol, and other fatty acids, have gained prominence due to their biocompatibility, gel-forming properties, and ability to modulate the sustained release of hydrophilic drugs. − The hydrophobic properties of these fatty acids play a crucial role in facilitating drug encapsulation and modulating the drug release rate, , thereby protecting the drug from premature degradation and preserving the gel structure at the targeted site. Myristic acid (MYR) was selected as the fatty acid in this study due to its high hydrophobicity as well as its solid state at both room and body temperature. This characteristic is essential for maintaining the matrix mass after the phase transformation from a solution state in solvent exchange-induced ISGs.

Molecular dynamics (MD) simulation is a powerful technique for predicting the structural and functional properties of molecules, providing valuable insights into dynamic systems. In drug delivery research, MD simulations have been widely utilized not only to investigate physicochemical properties at the molecular level, such as molecular structures and intermolecular interactions, − but also to optimize conditions for designed systems, − ultimately reducing experimental costs. For instance, both conventional and MD simulation experiments were used to investigate the mechanism of borneol matrix formation in an antisolvent-induced borneol-based in situ matrix system.

Density functional theory (DFT) is widely used in drug delivery systems to understand molecular interactions at the atomic level, such as the behavior of drug molecules in different environments. The DFT approach represents a transformative paradigm, furnishing valuable insights into molecular interactions. , By predicting electronic properties, DFT helps optimize drug formulations for improved stability and bioavailability. − It is intriguing to apply DFT for describing the interactions in in situ gels using this quantum mechanical technique as such an application has not been previously reported.

This study aims to develop a novel, nonpolymeric MYR-based ISG drug delivery system. The physical properties and phase transformation behavior of 10–60% w/w MYR-based ISGs in various solvent environments were explored in order to evaluate their potential for clinical applications, particularly in treating localized infections such as periodontitis within the periodontal pocket. Ethanol (EtOH), dimethyl sulfoxide (DMSO), and N-methyl pyrrolidone (NMP) were used as solvents for MYR in this investigation. Additionally, MD simulations were performed to gain further insights into the early stages of ISG formation, the molecular flow of components, and the interactions between the system’s components. Furthermore, the formation energy for all components in the ISG systems was calculated using DFT. The results of this study provide a foundation for future advancements in ISG drug delivery systems and pave the way for the development of next-generation therapeutics tailored to meet specific clinical needs.

2. Experimental Section

2.1. Materials

Myristic (MYR) (lot no. FPLK437X4S, P.C. drug center, CO.LTD, Bangkok, Thailand) was used as the matrix-forming agent. NMP (≥99.5%, Lot No. 144560-118, QreC, Auckland, New Zealand), DMSO (≥99.9%, Lot No. 1862992, Fisher Chemical, Horsham and Loughborough, UK), and ethanol (EtOH) (Absolute, Lot No. 12020052, RCI Labscan Limited, Bangkok, Thailand) were used as the solvents. Potassium dihydrogen orthophosphate (Lot No. E23W60) and sodium hydroxide (Lot No. AF310204) from Ajax Finechem, New South Wales, Australia, were used as the components of phosphate-buffered saline (PBS pH 6.8). Agarose (lot no. H7014714, Vivantis, Selangor Darul Ehsan, Malaysia) was used to determine the gel formation behavior. Staphylococcus aureus ATCC 6538, Escherichia coli ATCC 8739, Candida albicans ATCC 10231, Candida krusei TISTR 5259, Candida lusitaniae TISTR 5156, and Candida tropicalis TISTR 5306 were obtained from the Department of Medical Sciences, Ministry of Public Health, Nonthaburi, Thailand. Porphyromonas gingivalis ATCC 33277 (Microbiologics Inc., St. Cloud, MN, USA) was purchased from Thai Can Biotech Co., Ltd., Bangkok, Thailand. Tryptic soy agar (TSA) and Sabouraud dextrose agar (SDA) (Difco, Detroit, MI, USA) were used as the culture media for antimicrobial and antifungal testing, respectively. Sheep blood agar (M&P IMPEX, Bangkok, Thailand) was used for culturing P. gingivalis. All solvents and chemicals were of analytical grade and used as received.

2.2. Preparation of ISG

MYR-based in situ gel systems were formulated by dissolving MYR at varying concentrations (ranging from 10 to 60% w/w) in three different solvents: DMSO, NMP, and EtOH. The preparation involved thoroughly mixing the components using a magnetic stirrer at room temperature until the MYR was completely dissolved. Details regarding the specific concentrations and solvent ratios used in each formulation are listed in Table .

1. Composition of ISM Formulations (%w/w).

formulation	myristic acid	NMP	DMSO	EtOH
NM10	10	90
NM20	20	80
NM30	30	70
NM40	40	60
NM50	50	50
NM60	60	40
DM10	10		90
DM20	20		80
DM30	30		70
DM40	40		60
DM50	50		50
DM60	60		40
EM10	10			90
EM20	20			80
EM30	30			70
EM40	40			60

2.3. Density and Viscosities

The density of the formulations was determined using a pycnometer (Densito 30PX, Mettler Toledo Ltd., PortableLab TM, East Bunker Ct Vernon Hills, IL, USA), with each measurement replicated three times (n = 3). Viscosity and shear stress measurements were performed by using a viscometer (Brookfield Engineering Laboratories Inc., Middleborough, MA, USA) equipped with a CP-40 spindle. These measurements were conducted at a constant temperature of 25 °C and a shear rate of 250 s^–1, also with three replicates for each sample (n = 3).

2.4. Surface Tension and Contact Angle

Surface tension for each formulation was assessed by using a goniometer (FTA 1000, First Ten Angstroms, Newark, CA, USA). This measurement involved the observation of changes in the shape of a pendant drop, which was suspended in air and injected at a controlled rate of 1.9 μL/s using the goniometer’s integrated pump. The experiment was replicated three times (n = 3) to ensure the accuracy and reliability of the surface tension data. Additionally, the contact angles of the formulations on different surfaces of glass, paraffin, and agarose gel were measured using the same goniometer. For each measurement, a droplet was pumped out at a rate of 1.9 μL/s, and the contact angle was recorded at a time point of 5 s, based on the first automatically captured image of the droplet. These measurements were also conducted in triplicate (n = 3) to validate the consistency of the results.

2.5. Investigation of ISG Transformation

The transformation process of the in situ gel (ISG) systems was meticulously studied by injecting the gel using a 1 mL syringe equipped with an 18 gauge needle into a phosphate-buffered saline solution (PBS) with a pH of 6.8. To monitor the gelation process, images were captured at predefined intervals of 1, 5, 10, 15, and 30 min after introducing the ISG into the surrounding agarose, which facilitated phase separation and the subsequent solution-to-gel transformation. This transformation was visibly marked by the development of a turbid layer. Furthermore, the cross-sectional view of the gel formation was examined by placing 150 μL of ISG into a 300 μL hole (7 mm in diameter) within an agarose well. The progression of gel or matrix formation was documented under a stereo microscope (SZX10, Olympus Corp., Japan) at time points of 0, 1, 5, 10, 20, and 30 min. Additionally, to explore the microscopic interactions at the interface between the ISG and the aqueous phase, plain (0.6% w/w) agarose gel and 0.4 μg/mL sodium fluorescein-loaded agarose gel were prepared by dissolving in PBS (pH 6.8) at 60 °C. These plain and colored agarose gels were cut to the edge, and 50 μL of ISG formulations was dropped closely to the agarose rim. Phase transformation at the interface was observed under an inverted microscope (TE-2000U, Nikon, Kaw, Japan) for visible light and under an inverted fluorescent microscope (TE-2000U, Nikon, Kaw, Japan) using blue (B2A) filter with excitation at 450–490 nm for probing green color of sodium fluorescein by capturing the image at different time intervals (1, 5, and 15 min). These images helped further elucidate the dynamics of the gel formation process at the molecular interface.

2.6. Water Tolerance Measurement

The water tolerance test was performed to determine the tolerability of the system to transform from the liquid phase to the gel phase against of water. , To assess the water tolerance of the in situ gel (ISG) formulations, the capacity of the formulations to maintain clarity upon mixing with an aqueous phase was evaluated. Initially, a sample amount of 2.5 g of each ISG formulation was placed into a test tube. Using a micropipet, 20 μL of deionized water was then carefully added to the sample, followed by vortex mixing at controlled temperatures of 25 and 37 °C. Deionized water was added until turbidity appeared, which was the end point of the test, and the water amount was noted. The water tolerance was calculated using the equation shown below. To ensure reliability and reproducibility of our findings, each test was performed in triplicate (n = 3).

$$\%\mathrm{water}\mathrm{tolerance}=\frac{\mathrm{water}\mathrm{amount}(\mathrm{g})}{\mathrm{sample}\mathrm{amount}(\mathrm{g})+\mathrm{water}\mathrm{amount}(\mathrm{g})}\times 100\%$$

2.7. Wide Angle X-ray Scattering (WAXS)

The crystallization of ISG systems was systematically investigated using three distinct levels of deionized water, which corresponded to 10, 80, and 100% of their respective water tolerance values. These levels were chosen to replicate different stages of the crystallization process. The formulations prepared were labeled as formula-W1, formula-W8, and formula-matrix, representing the initial, late, and crystal stages of ISG crystallization, respectively. A comparative analysis was then conducted on the crystal properties of these water-incorporated formulas against a control formula with no added water and no raw matrix-forming material. To examine the crystallinity properties of these samples, wide angle X-ray scattering (WAXS) data were collected at Beamline 1.3 W of the Synchrotron Light Research Institute (SLRI) in Nakhon Ratchasima, Thailand. The WAXS setup involved a multipole wiggler as the radiation source. The Q range for the study was set from 0.00 to 32.93 nm^–1, which corresponds to a 2θ range of 0.00 to 47.59 degrees. The incident X-rays were generated at an energy that corresponded to a wavelength of 0.13776 nm. For the experiments, a sample–detector distance of 182.097 mm was used, allowing the measurement of characteristic d-spacing ranging from 1 to 33 nm.

2.8. FTIR Spectroscopy Analysis of ISG Components

A sample of the ISG solution was smeared onto the surface of a KBr tablet. Subsequently, FTIR was utilized to investigate the compatibility and interactions between components of the gels. The percentage of FTIR transmittance for the sample was recorded by using an FTIR spectrometer (Nicolet 4700, Madison, USA).

2.9. Scanning Electron Microscope (SEM)

The morphology of dried ISG remnants was thoroughly analyzed using scanning electron microscopy (SEM) (TESCAN MIRA3, Brno-Kohoutovice, Czech Republic). Samples were initially subjected to a gel formation experiment in phosphate-buffered saline (PBS) at pH 6.8 and left undisturbed for 7 days. Following this period, the gel remnants were rigorously washed with 200 mL of distilled water and subsequently freeze-dried. The dried samples were stored in a desiccator for 72 h to ensure complete dehydration. Prior to SEM analysis, each sample was sputter-coated with gold to enhance electron conductivity and image clarity. The examination was conducted at an accelerating voltage of 15 kV, optimizing the resolution and depth of the field to capture detailed surface structures.

2.10. Antimicrobial Activities Test

The standard strains used in this experiment included S. aureus ATCC 6538, E. coli ATCC 8739, C. albicans ATCC 10231, C. krusei TISTR 5259 C. lusitaniae TISTR 5156, and C. tropicalis TISTR 5306. The antimicrobial activities of the formulations were assessed employing the cylinder plate method, , using the cylinder plate method, with bacterial and fungal inocula prepared to match the 0.5 McFarland standard and spread on TSA (for bacteria) or SDA (for fungi). For P. gingivalis, the bacterial inoculum was spread on sheep blood agar. A 100 μL aliquot of each formulation was placed in a 6 mm diameter cylinder on the inoculated plates, and cultures were incubated: P. gingivalis under anaerobic conditions using an anaerobic incubator (Forma Anaerobic System, Thermo Scientific, Ohio, USA) and other microbes under aerobic conditions. Conversely, other microbial cultures were incubated under aerobic conditions. After 18 h of incubation at 37 °C, inhibition zones were measured, with each experiment conducted in triplicate.

2.11. Molecular Dynamics Simulation for Phase Transformation Study

Computational dynamics modeling was employed to explore the phase inversion and molecular movement within the ISG formulations upon exposure to an aqueous environment. Atom coordinates for each substance were sourced from databases such as PubChem and the Cambridge Crystallographic Data Centre (CCDC). For substances not available in databases, molecular optimization was performed using Gaussian09 (Gaussian, Inc., Wallingford, Connecticut, United States) to derive the starting geometry for MD simulations. The chemical 3-D structure of all components is displayed in Figure S1.

Based on the molar ratios of the components in each formulation shown in Table , the simulation models of two states of formulations: without water (DM10, DM40, NM10, and NM40) and with water (DM10W, DM40W, NM10W, and NM40W) were constructed, and their molecular dynamic box details are displayed in Table . The minimum water content around the interface that induces phase transformation in these simulation models was determined based on the water tolerance value.

2. MD Simulation Box Details of ISM Formulations.

molecular dynamic box details	DM10	DM10W	DM40	DM40W	NM10	NM10W	NM40	NM40W
amount of MYR molecules	300	300	1200	1200	300	300	1200	1200
amount of DMSO molecules	7800	7800	6000	6000
amount of NMP molecules					6300	6300	4800	4800
amount of WAT molecules		12,200		5810		12,000		6900
mole ratio of MYR:(DMSO/NMP):WAT	1:26	1:26:40	1:5	1:5:4.8	1:21	1:21:40	1:4	1:4:5.8
total amount of molecule in system	8100	20300	7200	13,010	6600	18,600	6000	12,900
total amount of atom in system	91,238	127,838	112,800	130,230	94,164	149,964	129,667	150,367

MD simulations of each system were conducted using Amber 20 software (University of California, San Francisco, California, USA) for 200 ns. The force field parameters for the molecules were generated using the Antechamber module with the general AMBER force field. The modeled system, comprising MYR and organic solvents, was placed within a periodic boundary box and solvated with TIP3P water. Energy minimization and subsequent heating to 310 K were performed using the sander module, followed by MD simulation at the same temperature using the pmemd module. Analysis of the simulations was conducted using Visual Molecular Dynamics (VMD) software (Theoretical and Computational Biophysics group, The Beckman Institute, University of Illinois at Urbana–Champaign, Illinois, United States). H-bond occupancy and the density of molecules were calculated, with H-bond formation determined when the distance between acceptor and donor atoms was less than 3.5 Å, and the angle formed by the acceptor, donor, and hydrogen atoms was less than 60°.

For simulation analysis, the root mean square deviation (RMSD) of MYR, solvents, and water was obtained using the cpptraj module. Diffusion constants were calculated using the ptraj module with the time-evolved trajectory, providing insights into the migration distance from the initial position and the diffusive properties of MYR, organic solvents, and water. For visual clarity and to better understand the configuration of MYR, certain molecules were repositioned in the x-, y-, or z-direction based on the length of the periodic boundary box in the visualization of simulation structures. Irrelevant molecules were omitted to ensure clarity in the displayed simulation structures, which were visualized using VMD with a focus on intermolecular interactions.

2.12. Formation Energy Evaluation

The optimization process utilized density functional theory (DFT), specifically integrating Becker’s three-parameter exchange and Lee–Yang–Parr correlation potentials (B3LYP). This approach was further refined with Grimme’s D3BJ dispersion correction (B3LYP-D3BJ) and employed the 6-31G­(d,p) Pople basis set. This combination is crucial for accurately modeling van der Waals interactions and achieving precise molecular geometries. All simulations were conducted in vacuum to eliminate the influence of water, as comparative studies with the conductor-like polarizable continuum model (C-PCM) indicated only minor qualitative differences. Computational analyses were performed using Gaussian09 software. The formation energy (E _f), which evaluates the stability of the MYR complex with a solvent, was calculated. This was done by determining the energy difference between the complex (E _MYR‑SOL) and the sum of the energies of the isolated adsorbate (E _adsorbate) and the MYR molecule (E _MYR) in their fully relaxed gas-phase geometries. The following equation was used:

$$E_{\mathrm{f}}=E_{\mathrm{MYR}-\mathrm{SOL}}-E_{\mathrm{SOL}}-E_{\mathrm{MYR}}$$

2.13. Statistical Analysis

All data were examined using the one-way analysis of variance ANOVA, followed by the LSD posthoc test. The analysis was conducted using SPSS for Windows (version 11.5). The significant level was set at p < 0.05.

3. Results and Discussion

3.1. Physical Appearance, Density, and Viscosity

The prepared MYR-based ISGs appeared as clear and transparent solutions, regardless of MYR concentration or solvent type. Density measurements at room temperature revealed a decreasing trend with increasing MYR content in both DMSO (1.0678–0.9587 g/cm³) and NMP (1.0265–0.9412 g/cm³) series, likely due to MYR’s lower density compared to these solvents. Conversely, the density of the EtOH ISG series (0.7950–0.8273 g/cm³) increased with higher MYR concentrations, reflecting MYR’s higher density than ethanol. Comparing solvent types, density followed the order DMSO > NMP > EtOH, mirroring the trends of their pure forms (DMSO: 1.101 g/cm³; NMP: 1.027 g/cm³; EtOH: 0.789 g/cm³ at 25 °C). Notably, the overall ISG densities remained close to that of water, suggesting potential for gradual settling within the periodontal pocket’s crevicular fluid.

The ISG formulations displayed an interesting relationship among MYR concentration, solvent type, and resulting viscosity. Across all formulations, the viscosity increased markedly with higher MYR content. This trend was especially pronounced in the NM and DM series, where viscosity rose from 3.55 cP for NM10 to 11.46 cP for NM60, and from 3.68 cP for DM10 to 12.12 cP for DM60. Even in the EM series, which showed the least dramatic increase, viscosity rose from 1.66 cP for EM10 to 3.18 cP for EM40. Compared to previous studies on solvent exchange-induced ISGs, polymer-based ISGs such as PLA, PLGA, Eudragid RS, and nitrocellulose exhibited much higher viscositiesapproximately 2040, 530, 3600, and 4500 cP, respectively. − In contrast, nonpolymer-based ISG materials like borneol and ibuprofen demonstrated viscosities of around 5 and 14 cP, respectively. , These comparisons highlight a key advantage of small-molecule ISGs, namely, their significantly lower viscosity profiles compared to those of polymer-based systems. Moreover, the MYR-based ISGs exhibited lower viscosities than ibuprofen-based ISGs, possibly due to MYR’s lower steric hindrance and weaker molecular interactions with solvent molecules. ,

This observed trend in viscosity can be attributed to the unique properties of MYR. As a long-chain fatty acid, MYR’s increasing concentration reduces the proportion of solvent molecules within the formulation, leading to increased steric hindrance and molecular entanglement. This restricts the free movement of solvent molecules, thereby increasing the viscosity of the system. The extent of this steric effect varies, depending on the solvent type. EtOH containing an −OH group is a polar protic solvent and participates in hydrogen bonding, while NMP and DMSO belong to a polar aprotic solvent and cannot form a hydrogen bond between themselves. Therefore, EtOH, with its lower molecular size and hydrogen bonding capability, is less susceptible to MYR’s steric hindrance, resulting in the significantly lower viscosity observed in these formulations compared to NMP and DMSO (2.04 cP for NMP, 1.98 cP for DMSO, and 1.28 cP for EtOH). Additionally, at higher MYR concentrations, the lower ratio of solvent molecules to MYR amplifies the impact of steric interactions and intermolecular forces, further contributing to the viscosity increase.

It is important to note that this analysis focuses primarily on the viscosity data. Incorporating insights from other characterization techniques, such as gelation behavior, alongside relevant citations would provide a more comprehensive understanding of the results and enhance the depth of the discussion.

3.2. Surface Tension and Contact Angle

Unlike the consistent trend observed in viscosity, the surface tension profiles of the ISG formulations revealed a nuanced relationship with the MYR concentration and solvent type. While increasing MYR content resulted in a predictable increase in viscosity, surface tension exhibited a more intricate and solvent-dependent response, as shown in Table .

3. Physical Properties of MYR-Based ISG Systems (Mean + S.D.) (n = 3) .

				contact angle (degree)
formula	density (g/cm3)	viscosity (cP)	surface tension (mN/m)	glass slide	agarose gel	paraffin
NM10	1.0116 ± 0.0001	3.55 ± 0.05	34.59 ± 0.40	14.45 ± 0.77	NA	44.15 ± 1.82
NM20	0.9945 ± 0.0001	4.37 ± 0.34	33.80 ± 0.47	13.37 ± 1.14	NA	43.94 ± 0.72
NM30	0.9807 ± 0.0001	5.72 ± 0.22	32.00 ± 0.61	17.11 ± 0.95	16.75 ± 0.57	37.65 ± 0.88
NM40	0.9675 ± 0.0001	6.68 ± 0.23	32.23 ± 0.03	20.04 ± 0.43	30.05 ± 0.17	45.82 ± 0.80
NM50	0.9541 ± 0.0000	9.33 ± 0.45	31.00 ± 0.00	20.93 ± 0.67	34.52 ± 1.09	45.04 ± 0.71
NM60	0.9412 ± 0.0000	11.46 ± 0.29	31.16 ± 0.18	26.37 ± 0.41	38.86 ± 0.53	42.42 ± 1.70
DM10	1.0678 ± 0.0004	3.68 ± 0.15	29.08 ± 0.28	15.35 ± 0.88	12.09 ± 0.14	42.26 ± 0.54
DM20	1.0427 ± 0.0004	4.41 ± 0.17	29.50 ± 0.50	13.80 ± 0.41	14.50 ± 0.36	42.03 ± 1.18
DM30	1.0201 ± 0.0001	5.37 ± 0.19	29.29 ± 0.29	14.68 ± 0.51	18.65 ± 0.24	42.27 ± 0.12
DM40	0.9980 ± 0.0002	7.04 ± 0.12	27.99 ± 0.52	24.47 ± 0.25	41.17 ± 1.33	41.39 ± 0.33
DM50	0.9844 ± 0.0094	9.14 ± 0.14	28.10 ± 0.54	21.02 ± 0.59	43.70 ± 0.54	42.27 ± 0.94
DM60	0.9587 ± 0.0001	12.12 ± 0.17	27.38 ± 0.45	21.23 ± 0.61	41.61 ± 0.68	41.59 ± 0.49
EM10	0.7950 ± 0.0000	1.66 ± 0.07	23.35 ± 0.19	10.73 ± 0.49	NA	20.19 0.12
EM20	0.8041 ± 0.0000	2.14 ± 0.10	23.62 ± 0.69	6.53 ± 0.00	NA	11.32 0.68
EM30	0.8138 ± 0.0001	2.54 ± 0.07	23.86 ± 0.11	NA	NA	NA
EM40	0.8273 ± 0.0001	3.18 ± 0.25	24.12 ± 0.27	NA	NA	NA
NMP	1.0265 ± 0.0008	2.04 ± 0.13	39.31 ± 0.28	31.08 ± 0.40	7.12 ± 1.51	54.94 ± 1.31
DMSO	1.0935 ± 0.0007	1.98 ± 0.09	43.95 ± 0.13	33.98 ± 1.25	4.30 ± 1.19	64.70 ± 0.81
EtOH	0.7858 ± 0.0000	1.28 ± 0.10	28.36 ± 0.27	15.16 ± 0.10	NA	23.19 ± 0.63

^aNA is present with nonavailable data; it indicates that the sample did not form a drop shape on the surface being tested.

In both the NMP and DMSO series, a compelling decrease in surface tension was observed with increasing MYR concentration. The NM10 formulation, starting with a surface tension of 34.59 mN/m, gradually decreased to 31.16 mN/m for NM60. A similar trend was observed in the DM series with DM10 starting at 29.08 mN/m and ending at 27.38 mN/m for DM60. However, the EtOH series defied this trend, exhibiting a subtle, yet distinct, increase in surface tension as the MYR concentration rose. EM10 began at 23.35 mN/m and gradually climbed to 24.12 mN/m for EM40, highlighting a unique dynamic at play. Further investigation of the influence of different solvents revealed that NMP consistently held the highest surface tension across all MYR concentrations, ranging from 33.80 mN/m for NM20 to 24.12 mN/m for EM40, mirroring its inherently high surface tension of 39.31 mN/m. DMSO followed closely, exhibiting values between 27.99 mN/m for DM40 and 23.62 mN/m for EM20, reflecting its intrinsic surface tension of 43.95 mN/m. Finally, ethanol, with its inherently low surface tension of 28.36 mN/m, stood out with the lowest overall value, ranging from 23.35 mN/m for EM10 to 24.12 mN/m for EM40.

This intricate relationship among these factors can potentially explain the observed trends. The decreasing surface tension in NM and DM series might be attributed to the incorporation of MYR molecules, possessing hydrophobic hydrocarbon chains, at the air–liquid interface. This could disrupt the hydrogen bonding networks present in NMP and DMSO, leading to a lower overall surface tension. , Conversely, the slight increase in the EM series might be due to the limited ability of ethanol to disrupt hydrogen bonding and its relatively low surface tension value. As the MYR concentration increases, its presence at the interface becomes more prominent, leading to a small rise in surface tension compared to pure EtOH.

The contact angles of the ISG formulations revealed intriguing interactions with various surfaces, offering valuable insights into the interplay among MYR content, solvent selection, and surface type, as presented in Table . On glass slides, the ethanol series displayed a notable decrease in contact angle with increasing MYR concentration, culminating in the complete loss of a stable droplet shape at higher concentrations, indicating exceptional spreadability. This trend can be attributed to MYR’s increasing hydrophobicity and its ability to disrupt hydrogen bonding at the glass–liquid interface, thereby promoting spreading. In contrast, the NMP and DMSO series exhibited a slight increase in contact angle, likely due to their higher viscosity, suggesting reduced spreadability.

For agarose gels, which were designed to mimic tissue at the administration site, both the DM and NM series exhibited significantly higher contact angles compared to those on the glass surface. This difference is likely due to matrix formation when the formulations came into contact with the water in the agarose gel, physically hindering the spreading of the droplets. Higher MYR concentrations in these series further increased the contact angle. The swift gel formation of DM60, as shown in Figure B, produced a skin-like surface around the droplet, which could hinder solvent exchange and reduce the contact angle. In contrast, droplets in the EM series expanded and disappeared within 5 s of being dropped, making it impossible to measure their contact angles. This behavior may be due to the properties of the EM formulation, which had very low viscosity and an insufficient gel formation rate, causing the droplets to spread thin across the surface. Additionally, EtOH’s higher hydrophilicity compared to DMSO and NMP made it more miscible with the water phase, allowing it to penetrate the aqueous agarose surface before the EM formulation could form a droplet.

1.

In vitro matrix formation behavior of MYR-based using three different solvents (EtOH (A), DMSO (B), and NMP (C)) in phosphate-buffered saline (PBS) at 25 and 37 °C.

The DM and NM series displayed contact angles exceeding 40° on paraffin, indicating their overall hydrophilic character. This suggests that despite the presence of MYR, the formulations retain a net attraction to water, as evidenced by the relatively high contact angle observed on the hydrophobic paraffin surface. This comprehensive analysis of contact angles highlights the diverse interactions between MYR-based ISG formulations and various surfaces, providing valuable insights for optimizing the formulation design for specific applications.

Interestingly, the surface tension and contact angle profiles showed trends similar to those observed in the viscosity profiles when compared to borneol-based and ibuprofen-based ISGs. The surface tension of MYR-based ISGs was lower than that of borneol-based ISGs but higher than ibuprofen-based ISGs. This reflects the molecular interactions within the ISG system, ordered by ibuprofen-NMP > myristic acid-NMP > borneol-NMP. , Moreover, the contact angle values of ibuprofen-based ISGs against glass, agarose, and paraffin were higher than those of MYR-based ISGs, indicating that MYR-based ISGs have better spreadability than ibuprofen-based ISGs. This suggests that MYR-based ISGs could diffuse more effectively into small pockets, offering an advantage for localized applications.

3.3. In Situ Gel Transformation

The in situ gel transformation of the myristic-acid–based formulations revealed significant insights into the dynamics of gel formation. Upon administration into phosphate-buffered saline (PBS) at pH 6.8, which mimics the fluid conditions of the periodontal pocket, all formulations initially remained as clear liquids, as shown in Figure . As the interaction between the ISG and the aqueous phase progressed, visible phase separation occurred, indicating gel transformation. The kinetics of gel transformation varied significantly across different formulations, with the MYR concentration playing a critical role in this behavior. In formulations with higher MYR concentrations, a rapid transition to a gel state was observed, attributed to increased hydrophobic interactions within the matrix. The EM series did not immediately transform into a gel after being injected into the bottom of the test tube. Instead, it began transforming into a gel after floating to the surface of the PBS. The appearance of the EM gel, resembling white sediment suspended at the surface of the PBS, may be due to the formula’s low density, as mentioned earlier, causing it to float and disperse before transforming into a gel. In contrast, the DM and NM series began transforming immediately after injection into the bottom of the test tube, forming a single unit without the dispersion seen in the EM series. Both the DM and NM formulations at 40% w/w exhibited complete gel transformation within 1 min.

Moreover, the DM formulations appeared to undergo faster transformations than the NM formulations, likely due to DMSO’s higher miscibility with water, which facilitates rapid solvent exchange and accelerates gelation. This observation is crucial, as the rate of solvent exchange is closely linked to the viscosity of the formulation. Our viscosity results showed that formulations with DMSO, despite their rapid gelation, maintained a manageable viscosity, which enhances the ease of administration compared to the more viscous polymer-based systems.

Previous studies on polymer-based ISGs, including PLGA, PLA, Eudragit RS, and nitrocellulose, demonstrated slower gel formation due to the high viscosity of these formulations. The high viscosity reduced the mobility of water and solvent molecules, slowing down the solvent exchange process and leading to incomplete phase transformation. ,, This finding aligns with prior research suggesting that solvent systems that balance rapid gel transformation with manageable viscosities are preferable for ensuring ease of administration and effective drug entrapment.

Our observations of in situ gel transformation revealed that increasing the temperature from 25 to 37 °C notably affected the gelation behavior of the EM, NM, and DM series. Contrary to expectations that higher temperatures might enhance the entropy of the ISG system during the crystallization process of MYR, our results indicate that gelation was actually impeded at the elevated temperature across all of the solvent systems studied. At 25 °C, the formulations generally displayed a gradual increase in opaque matrix formation, with those containing higher concentrations of MYR transitioning more rapidly to a gel or matrix-like state. This suggests that at room temperature, phase inversion for gel formation was more effectively driven by the solvent–water miscibility properties and MYR concentration. Specifically, the EM series demonstrated progression to a turbid dispersion layer, particularly at higher MYR concentrations, indicating ineffective gel formation. In contrast, the NM and DM series showed a steady increase in gelation with increasing MYR concentrations, with the gel transformation visibly advancing over the 30 min observation period.

However, at 37 °C, the transition to gel was significantly slower. Despite initially being in a clear state, the formulations exhibited delayed turbidity and gelation across all series. For instance, the EM10 to EM30 formulations showed significantly less opacification at higher temperatures, suggesting that the elevated temperature may have affected MYR solubility, thereby delaying the phase inversion process. This pattern was also observed in the NM and DM series, where the higher temperature appeared to disrupt typical gel transformation kinetics, potentially due to changes in solvent–antisolvent dynamics, which are critical for gel formation. These findings are particularly important because they suggest that the physical properties of the solvents, combined with the solubility of the gel-forming agents, are altered at higher temperatures in a way that inhibits gel transformation. This demonstrates the sensitivity of MYR’s phase transformation behavior to temperature changes, which could be explored in future work for developing MYR-based ISGs for temperature-sensitive drug delivery systems.

Figure shows the cross-sectional view at 12× magnification of MYR gel formation under a stereomicroscope, illustrating the completion of the gel formation process in the MYR-based ISG. Phase inversion began rapidly around the interfacial region of the agarose well. Even at a 10% MYR concentration, the EM series demonstrated rapid and complete gel formation surrounding the agarose rim. As MYR content increased, the transformation rate at the interface increased accordingly. However, the transformation at the center region was delayed and incomplete, as the MYR matrix around the interface blocked solvent exchange between the ISG and agarose gel. For the 20 and 30% w/w EM formulations, gel formation was observed on the surface of the ISG solution, potentially due to solvent evaporation, particularly EtOH. Over time, MYR gel growth extended into the agarose gel.

2.

Cross-sectional view of matrix formation of MYR-based ISG systems using NMP, DMSO, and EtOH as the solvents in the agarose hole under a stereo microscope at a magnification of 12× (B).

In comparison, the DM and NM series displayed slower transformation rates than did the EM series, but the overall completion of gel formation was better in the DM and NM formulations. Increasing the MYR content accelerated the gel transformation, similar to the EM series. However, at 50 and 60% w/w MYR in the DM formulations, a hollow center was observed within the ISG, indicating that excessively rapid phase transformation may not be beneficial. For comparison, commercial ISG products like Atridox contain 36.7% poly­(DL-lactide) (PLA) dissolved in 63.3% NMP. Previous research on polymer-based and small-molecule-based ISG systems suggests that an optimal concentration of gel-forming agents around 30–40% is appropriate for drug delivery system formulations.

A microscopic view of the interface region between MYR-based ISG systems and agarose gel, mimicking the periodontal pocket environment, is shown in Figure A,B. After MYR-based ISG came into contact with agarose, MYR gel mass formed and became denser and more compact as the MYR concentration increased. Interestingly, at 20% w/w, gel transformation was completed within 5 min. However, at concentrations above 30% w/w, gel transformation slowed, although it accelerated at even higher concentrations. This slower transformation at higher MYR concentrations is due to the dense packing of MYR crystals at the interface, which tightly blocked water diffusion into the ISG phase, thus retarding the solvent exchange process.

3.

Matrix formation at the interface of agarose gel and MYR-based ISG, observed at 100× magnification in bright-field (A and B) and fluorescent modes (C) under an inverted microscope.

To further investigate the antisolvent process, sodium fluorescein was used to track the phase inversion process in the 40% w/w DM and NM formulations, as shown in Figure C. MYR at 40% (w/w) was the minimum concentration capable of self-gel formation in both macroscopic and microscopic studies. Both DM40 and NM40 exhibited green fluorescence moving from the agarose gel side toward the ISG, indicating water moving into the ISG system. MYR crystal growth followed the water movement into the ISG. The DM40 formulation exhibited lower fluorescence intensity in the gel transformation region compared to that of NM40, indicating that less water was required for the phase inversion process in DM40. This finding explains why the phase inversion rate of DM40 was faster than NM40. , Overall, our findings suggest that the use of MYR and specific organic solvents in ISG formulations significantly enhances the gel transformation process. This improvement not only provides a more robust mechanism for drug delivery but also offers advantages over traditional polymer-based systems by enabling a more predictable and sustained drug release profile. This research paves the way for further innovations in the design of ISG systems, particularly for applications requiring precise control over the drug release kinetics.

3.4. Water Tolerance

This study revealed distinct patterns in water tolerance values primarily influenced by the solvent type and temperature, as shown in Figure . EtOH-based formulations exhibited the highest water tolerance, requiring 24.30–36.10% water to induce phase separation. This can be attributed to ethanol’s lower dielectric constant of 24.3, which enhances MYR–solvent interactions by reducing polarization between EtOH and MYR, thus maintaining MYR solubility in the formulation. In contrast, DMSO- and NMP-based formulations demonstrated significantly lower water tolerance values due to the higher dielectric constants of DMSO and NMP, approximately 47 and 32, respectively. , These higher dielectric constants promoted phase separation of MYR when water was added, as lower MYR–solvent interactions were observed. This behavior corresponds to the decreased surface tension when MYR was added to ISG formulations, with MYR–DMSO interactions being weaker than MYR–NMP interactions. , These water tolerance results support the findings from the gel formation experiments, as the DMSO series required less water than the NMP series to induce MYR phase transformation, correlating with faster gel formation and water-induced regions observed in water-tracking experiments.

4.

%Water tolerance of MYR-based ISG (n = 3) (A) and water tolerance MYR-based ISG systems after titration with deionized water at temperatures of 25 and 37 °C (n = 3) (B).

Focusing on the NMP series, temperature had a noticeable effect on water tolerance when comparing room temperature (25 °C) to body temperature (37 °C). At 25 °C, the water tolerance was lower, indicating a quicker onset of turbidity upon the addition of water. However, at 37 °C, water tolerance increased significantly. This improvement at higher temperatures can be attributed to the increased solubility of MYR in NMP, aligning with thermodynamic principles that predict solubility increases with the temperature. This increased solubility facilitates nucleation, crystal growth, gel formation, and MYR crystallization within the system. Lower water tolerance is crucial for ensuring phase inversion through a solvent exchange with the aqueous phase. This factor is particularly important in drug delivery applications, such as periodontal treatments, where matrix formation is necessary for effective drug encapsulation and controlled release. In environments such as the periodontal pocket, where the formulation comes into contact with bodily fluids, low water tolerance promotes phase inversion and enables sustained drug release.

By systematically comparing water tolerance across different solvent types and examining temperature variations within a specific solvent series, this study provides comprehensive insights into the formulation dynamics that influence the performance and efficacy of ISG drug delivery systems based on water sensitivity.

3.5. Wide Angle X-ray Scattering (WAXS) and Fourier-Transform Infrared Spectroscopy (FTIR)

The phase inversion characteristics of the MYR-based ISG system, evolving through the gradual introduction of water, were thoroughly analyzed using WAXS. This study aimed to explore the intricate transformations that occur when water is incrementally introduced into the ISG system to reach the water tolerance threshold. In the WAXS experiment, the NM series were focused due to interference caused by DMSO in the DM series, which affected the software’s ability to calculate the scattering pattern. The initial WAXS profiles of intact MYR, NMP, and NM40 solution are shown in Figure S2. In these profiles, the discernible peaks of NM40 align with the corresponding NMP peaks, indicating the amorphous nature of the solvent within the NM40 ISG.

As the water content increased progressively, the WAXS peak patterns of water-induced ISG at 10–80% (NM40W1–NM40W8) levels continued to match the NMP peaks. Remarkably, at the 100% water tolerance end point (NM40 matrix), the WAXS pattern exclusively mirrored that of the MYR pattern. This result indicates that MYR molecules within the NM40 gels begin nucleation and crystallization, forming a gel and progressing to a crystalline structure, as monitored through WAXS signals. , This finding confirms that MYR undergoes nucleation, gel formation, and crystallization when the water tolerance point is reached, transforming into a crystalline matrix after water diffuses into the ISG system, consistent with the fluorescence tracking of gel formation shown in Figure S2.

The FTIR spectra revealed several characteristic peaks of the solvents, intact MYR, and MYR-ISG solutions, as shown in Figure S3. These results were essential for confirming the presence of MYR and organic solvents in the formulations as well as for detecting any potential chemical interactions between MYR and the solvents. Each formulation consistently displayed the characteristic peaks of its respective solvent without any shift in peak positions, indicating stability and the absence of reactive interference.

For example, all formulations exhibited a strong, broad peak around 3300 cm^–1, indicative of O–H stretching vibrations, suggesting the presence of hydroxyl groups typically found in components like ethanol. Additionally, the FTIR spectra showed peaks around 2920 and 2850 cm^–1, corresponding to the asymmetric and symmetric stretching vibrations of CH ₂ groups, respectively. These peaks, representative of alkyl chains, were sharper and more intense in the formulations (DM40, NM40, and EM40) compared to intact MYR, likely due to the higher concentration of solvents relative to MYR in these systems.

In the NMP-based system (NM40), a peak near 1700 cm^–1, attributed to C=O stretching vibrations, was prominent, confirming the ester functionalities of both MYR and NMP. In the DMSO-based formulations, both DMSO and DM40 exhibited a strong peak at 1044 cm^–1, characteristic of the S=O stretching vibration unique to the DMSO structure. These FTIR findings confirmed that no molecular alterations occurred in MYR when it was incorporated into the ISG formulations. Furthermore, the results demonstrate that the solvents did not engage in chemical interactions with MYR, providing crucial evidence for the structural and chemical stability of the MYR-based ISGs.

3.6. Surface Topography

The surface topography of dried gel formulations, DM40 and NM40, after 7 days of phase transformation in PBS pH 6.8, revealed distinct structural characteristics, as shown in Figure , highlighting their potential as drug delivery systems. Using SEM at a magnification of 2000x, both formulations displayed a multilayer sheet-like structure, with DM40 exhibiting a noticeably denser packing compared to NM40. This suggests that the matrix structure in DM40 was more tightly arranged, possibly due to differences in the physical properties of the solvent systems used.

5.

SEM photographs of MYR-based ISM remnants using DMSO and NMP as the solvents after phase inversion in PBS at pH 6.8 for 7 days.

Further examination of the gel surfaces at a higher magnification of 10,000× provided deeper insights into the porosity of each formulation. The SEM images showed that NM40 exhibited lower porosity compared to DM40, which can be attributed to the solvent exchange rate during the gelation process. The phase transformation results indicate that the slower solvent exchange rate in NM40 led to more nucleation origins for MYR crystallization compared to DM40. This resulted in a more compact and less porous structure in NM40, which could influence the diffusion pathways for any entrapped drug, potentially affecting the drug release kinetics. , Moreover, the morphology of the ISG also presents a crystalline structure that corroborated the crystalline pattern observed in the WAXS analysis of the MYS-based system. This integrated evidence confirmed that MYS is an effective gel-forming material for solvent-induced ISG drug delivery systems. The layered, sheet-like morphology, with its well-structured internal architecture, suggests that the formulations can effectively entrap drugs within the ISG matrix, even after an extended period of drug release.

It is particularly important to conduct drug release studies after loading the drugs into the system to understand how the unique structural properties of DM40 and NM40 influence the release behavior under physiological conditions. These studies will provide crucial insights into the practical applications of these formulations in clinical settings and help optimize the system for specific therapeutic needs.

3.7. Antimicrobial Activities

The antimicrobial effectiveness of various organic solvents and MYR-based ISG formulations is shown in Table and Figure S4. The antimicrobial activities were assessed across a range of bacterial and fungal pathogens, providing a comprehensive understanding of each formulation’s therapeutic potential.

4. Clear Zone Diameter of MYR-Based ISG Systems against S. aureus ATCC 6538, E. coli ATCC 8739, C. albicans ATCC 10231, C. krusei TISTR 5259, C. lusitaniae TISTR 5156, and C. tropicalis TISTR 5306 (n = 3) .

	clear zone diameter (mm) (mean ± S.D.)
formula	S. aureus ATCC 6538	E. coli ATCC 8739	C. albicans ATCC 10231	C. krusei TISTR 5259	C. lusitaniae TISTR 5156	C. tropicalis TISTR 5306
NMP	17.0 ± 1.0a	17.7 ± 1.5b	28.0 ± 1.0	23.7 ± 1.2	28.3 ± 1.5d	25.8 ± 0.8
DMSO	12.2 ± 1.0	12.0 ± 1.0c	16.3 ± 0.6	15.8 ± 1.3	20.8 ± 0.8	19.8 ± 1.3
NM40	15.3 ± 1.5a	16.3 ± 2.1b	25.0 ± 1.0	21.0 ± 1.0	27.0 ± 1.0d	23.3 ± 0.6
DM40	0.0 ± 0.0	10.0 ± 1.0c	11.7 ± 0.6	11.3 ± 1.3	12.2 ± 2.2	15.3 ± 1.5

^aThe superscripts a–d indicate a no significant difference (p ≥ 0.05) by using one-way ANOVA followed by an LSD posthoc test.

NMP demonstrated robust activity against both Gram-positive bacteria (Staphylococcus aureus) and Gram-negative bacteria (Escherichia coli), with inhibition zone diameters of 17.0 and 17.7 mm, respectively. In this study, NMP also exhibited significant antifungal activity against Candida albicans, producing a clear zone of 28.0 mm. NMP is well-recognized not only as an effective solvent in pharmaceutical applications but also for its inherent antimicrobial properties, particularly against fungi. Previous studies have noted that NMP exhibits fungicidal activity, likely due to its ability to disrupt fungal cell membrane integrity and interfere hyphal development with fungal metabolism. , When comparing NMP alone to the MYR-based NM40 formulation, no significant changes in activity were observed against S. aureus, E. coli, C. krusei, and C. lusitaniae, with inhibition zones of 15.3, 16.3, 21.7, and 27.0 mm, respectively. This indicates that the addition of MYR did not interfere with the antimicrobial properties of NMP, and MYR-based ISG formulations maintained the antimicrobial activity of NMP, despite the retardation of NMP diffusion after the gel transformation of the MYR-based ISG.

DMSO, in comparison, showed generally lower antimicrobial activity than NMP. It produced smaller inhibition zones against S. aureus and E. coli, measuring 12.2 and 12.0 mm, respectively. The antifungal activity against C. albicans was also lower with an inhibition zone of 16.3 mm. According to Hassan’s work, DMSO can inhibit microbial growth, though typically at higher concentrations than those used in this study. This could explain the lower inhibition zones observed with the DMSO-based formulations, which managed a 16.3 mm zone against only C. albicans. This suggests that while DMSO is an effective solvent for drug delivery, its antimicrobial properties are less potent under the conditions tested. Consequently, the DM40 formulation did not produce any inhibition zone against S. aureus (0.00 mm), indicating a lack of antibacterial activity in this particular setup. Its antifungal activity was also significantly reduced across all Candida strains (p < 0.05), with the smallest clear zone of 11.3 mm observed against C. krusei. Compared to DMSO alone, the DM40 formulation showed a significant reduction in antimicrobial activity (p < 0.05). This reduction could be linked to the retardation of DMSO diffusion due to MYR gel formation after contact with the aqueous phase of the media.

The addition of MYR, particularly at concentrations as high as 40%, has been previously documented to enhance antimicrobial effects due to its ability to alter the permeability of microbial cell walls and membranes. In this study, the NM40 formulation showed a slight decrease in antifungal activity compared to that of NMP alone against the three Candida species. However, it still demonstrated broad-spectrum antifungal potential, with clear zones of 27.0 mm against C. lusitaniae and 23.3 mm against C. tropicalis. The reduction in activity with MYR addition in NM was less pronounced than that in DM, likely due to the slower gel transformation of NM40 compared to DM40, which affects the retardation of solvent diffusion and confirms the antifungal effects of NMP. On the other hand, it is noteworthy that the DM40 formulation did not exhibit significant antibacterial activity and showed reduced antifungal efficacy. This could be attributed to specific interactions between DMSO and MYR, which may influence the release dynamics and bioavailability of the active compounds.

In conclusion, these results illustrate the antimicrobial potential of NMP and MYR, particularly against fungal pathogens, and emphasize the importance of carefully selecting and optimizing solvent systems in ISG formulations. The unique properties of NMP as both a solvent and an antimicrobial agent offer significant advantages for developing effective antifungal therapies, as demonstrated by the superior antifungal activity observed in this study. These findings not only expand on existing knowledge but also pave the way for further research into tailoring ISG formulations for targeted antimicrobial applications.

3.8. Computational Results of MD Simulation

To explain the gel transformation and observe the molecular packing during solvent exchange at the initial stage of MYR transformation, the models of formulations were designed, and MD simulations were performed by using AMBER20 for 200 ns. The RMSD values of the MD systems (Figure ) were then investigated and compared in order to determine the stability and conformational changes of each formulations. Moreover, to observe the molecular behavior of each component in the system upon the diffusion of water, the RMSD of myristic acid, solvent, and water molecules relative to the starting point in each MD simulations of all formulations was investigated and is shown in Figure S5. The snapshots of the systems after contact with water (DM10W, DM40W, NM10W, and NM40W) at 0, 5, 10, 50, 100, 150, and 200 ns were also taken and are displayed in Figure .

6.

Root mean square deviation (RMSD) of the MD simulations of all formulations (DM10, DM40, NM10, and NM40) (A) before and (B) after contact with water (DM10W, DM40W, NM10W, and NM40W).

7.

Snapshots of molecular dynamic simulations of the solvent exchange mechanism of (A) DM10W, (B) DM40W, (C) NM10W, and (D) NM40W, where green, orange, yellow, and cyan molecules represent MYR, DMSO, NMP, and water molecules, respectively.

As can be seen in Figure , DM systems took a shorter time to reach equilibrium (20 ns), compared with NM systems (75 ns). The RMSD of DMSO molecules in DM systems also became stable by 20 and 25 ns for DM40 systems and DM10 systems, respectively, while the RMSD of NMP molecules in NM systems attained the steady value in 40 and 50 ns for NM40 systems and NM10 systems (Figure S5C,D). This confirms that the MYR concentration is a key factor influencing gel transformation even in the initial 200 ns of the solvent exchange process. The RMSD of the organic solvents (both DMSO and NMP) also decreased upon adding water, likely due to MYR molecule packing obstructing solvent mobility in the ISG system. The RMSD of water also decreased, influenced by MYR concentration, as shown in Figure S5. This indicates that the gel-forming agent concentration is crucial in the gel transformation process as it affects solvent exchange and the completion of gel formation. Thus, the concentration should be a key consideration in the development of ISG drug delivery systems.

The 3-D snapshots from the MD simulation of each molecular component in DM10W, DM40W, NM10W, and NM40W (Figure A–D) showed significant differences in water molecule behavior between the 10 and 40% formulas. Water molecules were obstructed by MYR, leading to slower diffusion in the 40% MYR formula compared to the 10% MYR formula. The DMSO series showed a slightly slower diffusion than the NMP series. The MYR molecule compaction in each formula indicated orderly arrangement at the molecular level, suggesting MYR nucleation and crystallization within 200 ns. This arrangement involves H-bonding interactions among the carboxyl groups in the MYR molecule’s hydrophilic head and hydrophobic interactions between the tails. When comparing MYR to other small-molecule ISG materials like ibuprofen and borneol, which do not form rigid structures as MYR does during the initial phase of solvent exchange, , MYR’s ability to entrap active drugs at the molecular level is evident. Increasing MYR content from 10 to 40% (w/w) allowed larger MYR group formation in simulations, with the 40% MYR formula showing no splitting into multiple groups as observed in the 10% formula. Comparing DMSO and NMP series, MYR in DMSO formed bulk agglomerations more efficiently, likely due to DMSO’s superior water miscibility. This aligns with the matrix formation study results. The radius of gyration (R _g) confirmed MYR compaction, showing decreased R _g values with increased MYR content (Figure ), indicating tighter packing after solvent exchange. H-bonding interactions among MYR molecules were more significant in the DMSO series than in the NMP series, as shown in Figure S6, reinforcing the role of H-bonding in MYR crystallization. The mobility of molecules, especially solvents and water, was reduced by MYR agglomeration, as shown by decreased diffusion constants with a higher MYR content (Table ).

8.

Radius of gyration of MYR of the MD simulations of (A) DM10, DM40, DM10W, and DM40W and (B) MN10, MN40, MN10W, and MN40W.

5. Diffusion Constant of Each Substance in Formulations Box before and after Contact with Water.

	composition
formulation	MYR	DMSO	NMP	WAT
DM10W	0.2868	11.7213		16.4977
DM40W	0.0325	3.607		5.6541
NM10W	0.4078		3.0772	6.5883
NM40W	0.0157		1.2814	3.2496

In summary, MD simulation provided insights into MYR molecular arrangement as an orderly crystal packing at the initial stage. This supports the microscopic study showing MYR crystal growth around the interface and the crystal topology observed in the SEM studies. These findings indicate that MYR concentration and solvent type are critical factors in ISG system behavior, influencing gel formation and drug release controllability, which are essential for developing ISG formulations.

3.9. Computational Results of Formation Energy via DFT Calculations

To gain insights into the interaction between MYR and various solvents, we computationally investigated the adsorption mechanism using density functional theory (DFT) calculations. Initially, the geometry optimization was performed at the B3LYP-D3BJ/6-31G­(d,p) level, and the optimized structures of MYR, NMP, DMSO, and EtOH are presented in Figure S7. To further explore the interaction sites of these molecules, we explored their electrostatic potential maps (ESPs), which measure the strength of the charges from nearby nuclei and electrons at specific positions. ESPs facilitate visual assessments of a molecule’s charge distribution, enabling the identification of both Lewis acid and Lewis base sites. The calculation results revealed that the hydroxyl, carboxylic, and carbonyl groups of both MYR and solvents had negative electrostatic potential (red region in Figure S7); thus, they acted as electron-donating parts to other molecules. The structures of molecular complexes involving a 1:1 ratio of MYR with both solvents and MYR itself at various interaction sites were then designed, as shown in Figure S8, and their formation energies were computed and compared, as depicted in Figure .

9.

Formation energy of all configurations calculated at the B3LYP/6-31g­(d,p) level.

For the MYR-MYR interaction, a complex presenting a hydrophilic head-to-head interaction emerged as the most stable, with a formation energy of −0.98 eV, as shown in Figure . This configuration included two hydrogen bonds between the carboxylic groups with bond distances of 1.6 Å, as depicted in Figure . This result agreed with the MD simulation, which presented the orderly arrangement by the majority interaction as H-bonding among the carboxyl groups at the hydrophilic head of MYR.

10.

Most stable configuration with the ESP maps of (1:1) MYR-MYR, MYR-NMP, MYR-DMSO, and MYR-NMP. The molecular structures are calculated at the B3LYP/6-31g­(d,p) level.

For the NMP-MYR complex, four stable configurations were identified: (i) the OH functional group of MYR paired with the carbonyl group of NMP, (ii) the same groups paired but in a different orientation, and for types (iii) and (iv), the hydrophobic tail of MYR paired with different orientations of NMP. Of these, configuration (i) proved to be the most stable, displaying a formation energy of −0.61 eV. The optimized structure featured two hydrogen bonds with distances of 2.34 and 2.74 Å, as shown in Figure .

Similarly, four configurations were considered for the MYR-DMSO complex: (i) the hydroxyl group of MYR paired with the sulfoxide group of DMSO, (ii) the carbonyl group of MYR paired with the methyl group of DMSO, (iii) the hydrophobic tail of MYR paired with the sulfoxide group of DMSO, and (iv) the hydroxyl group paired with the sulfoxide group in a different orientation from (i). It was found that configurations (i) and (iv) had the lowest formation energy (−0.82 eV), indicating that the hydroxyl group of MYR formed strong hydrogen bonding interaction with the sulfoxide group of DMSO.

For the MYR-EtOH complex, the most stable formation involved pairing the carbonyl and hydroxyl groups of MYR with the hydroxyl group of EtOH, resulting in a formation energy of −0.73 eV and a bond distance of 1.72 Å, as shown in Figures and , respectively. ESP analysis indicated strong interactions between the negatively charged regions at the hydroxyl group of MYR and the carbonyl group of NMP, the hydroxyl group of EtOH, and the sulfoxide group of DMSO. These results confirm that hydrogen bonding plays a crucial role in the solubilization and gel transformation of MYR in these systems and explain the orderly arrangement of MYR molecules after transformation. The stability and integrity of these hydrogen bonds not only influence the initial solubilization process but also significantly affect the subsequent gelation behavior of MYR, as evidenced by the uniform and structured gel matrices observed in SEM analyses. This molecular arrangement is critical for ensuring the controlled release of active ingredients encapsulated within the gel, contributing to the efficacy of drug delivery.

Furthermore, MYR’s ability to form strong hydrogen bonds and orderly crystalline structures becomes apparent. This property of MYR enables it to entrap active drugs at the molecular level during the initial phase of the transformation. The increased MYR content from 10 to 40% w/w in our simulations further demonstrated MYR’s capacity to form larger agglomerations, as seen in the simulation snapshots. These findings, combined with MD simulations and DFT calculations, provide a comprehensive understanding of the molecular interactions and transformation mechanisms in MYR-based ISG systems. This synergistic effect of computational techniques offers detailed molecular-level insight, confirming the pivotal role of hydrogen bonding in the effective gelation and drug delivery applications of MYR-based systems.

4. Conclusions

This study provides a thorough investigation of solvent exchange MYR-based ISG systems using physicochemical analyses, matrix formation studies, MD simulations, and DFT calculations. The results showed that the higher MYR concentrations led to the formation of larger MYR agglomerates and a more compact matrix structure. FTIR spectroscopy confirmed the pivotal role of hydrogen bonding between MYR and solvents such as NMP and DMSO. SEM analyses also demonstrated uniform and structured MYR matrices essential for controlled drug release. Particularly, NMP-based formulations with higher MYR content showed swift gel formation with low water tolerance to induce phase inversion into the matrix and showed efficient antimicrobial efficacy against various microbes, making them promising candidates as an antimicrobial drug delivery system. Moreover, MD simulations provided the structural dynamics underpinning gel formation and demonstrated how solvent and MYR interactions at the molecular level influence the macroscopic properties of the ISG systems, while DFT calculations highlighted the critical role of hydrogen bonding in solubilizing MYR and stabilizing the gel matrix, thus supporting the experimental observations. Overall, this research underscores the importance of careful solvent selection, optimal MYR concentration, and the role of hydrogen bonding in developing effective ISG formulations for further research on drug delivery systems in specific therapeutic applications.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.5c00187.

• 3-D structure of MYR, DMSO, NMP, and ethanol; WAXS of MYR-based ISG formulations during MYR phase separation by water induction; FTIR spectra of organic solvents, intact MYR and MYR-ISG solutions; clear zone of MYR-based ISG systems against S. aureus ATCC 6538, E. coli ATCC 8739, C. albicans ATCC 10231, C. krusei TISTR 5259, C. lusitaniae TISTR 5156 and C. tropicalis TISTR 5306 (n = 3); root mean square deviation (RMSD) of MYR, DMSO, NMP, and water molecules relative to the starting point in each MD simulations of all formulations before (DM10, DM40, NM10, and NM40) and after contact with water (DM10W, DM40W, NM10W, and NM40W); change in number of hydrogen bonds of MYR-MYR molecule of formulations box contact with water; optimized structure of MYR, EtOH, DMSO and NMP with their electrostatic potential maps, calculated at the B3LYP-D3BJ/6-31g­(d,p) level; and configurations considered for molecular complexes involving a 1:1 ratio of MYR with both solvents and MYR itself at various interaction sites (PDF)

The authors declare no competing financial interest.