Formulation design and mechanism study of hydrogel based on computational pharmaceutics theories

Highlights

• A computational pharmaceutics strategy was built to guide the rational formulation design of polymer-based drug delivery systems (DDSs) such as hydrogel.

• Drug releasing mechanism of hydrogel was elucidated from the molecular lever by coarse-grained molecular dynamic simulations.

• The computational pharmaceutics methods for the study of formulation factors were proved to be accurate and feasible by various experiments.

Abstract

Formulation design and mechanism study of the drug delivery system (DDS) is an important but difficult subject in pharmaceutical research. The study of formulation factors is the most time- and labor-consuming work of formulation design. In this paper, a multiscale computational pharmaceutics strategy was developed to guide the systematic study of formulation factors of a typical polymer-based DDS, hydrogel, and further to guide the formulation design. According to the strategy, the combination of solubility parameter (δ) and diffusion coefficient (D) calculated by the AA-MD simulation was suggested as the general evaluation method for the matrix screening of the hydrogels at the pre-formulation stage. At the formulation design stage, the CG-MD simulation method was suggested to predict the morphology and drug-releasing behavior of the hydrogels under different formulation factors. The influence mechanism can be explained by the combination of multiple parameters, such as the microstructure diagram, the radius of gyration (Rg), the radial distribution function (RDF), and the free diffusion volume (V_diffusion). The simulation results are in good agreement with the in vitro release experiment, indicating that the strategy has good applicability.

Introduction

Drug delivery system (DDS) refers to a technical system that comprehensively regulates the distribution of drugs in organisms in space, time and dosage. Polymer is an important type matrix of DDS, which has played an integral role in the advancement of the DDS by acting as a drug carrier and providing the controlled release of drugs. Essentially, the drug release performance of a DDS is decided by the morphology of the polymer matrix and the drug-matrix interactions, which are deeply affected by the formulation factors, such as the degree of polymerization (DP) of the matrix materials, solvent, temperature, pH, the concentration of drug, matrix, and additives, and so on. A systematic understanding of the drug release mechanism of the polymer matrix under different formulation conditions will help guide the development of polymer-based DDS more efficiently and intelligently. However, systematic research on formulation factors often requires an enormous amount of labor and time, which limits the efficient development of DDS. However, the developing process is still always laborious, time-consuming, costly, and lacks mechanistic understanding, because the trial-and-error method, which is based on a huge number of experiments, is still the most conventional research method [1]. Therefore, the development of new research methods for the formulation design and mechanism study of DDS has become an important subject in pharmaceutics.

Computational pharmaceutics, which integrates multi-scale modeling, artificial intelligence, and big data techniques into drug formulation development, has been considered to be able to accelerate and simplify the development process of formulation, and lead the development from personal experience to knowledge base [2]. Formulation design and mechanism study of the DDS is one of the important applications of computational pharmaceutics. With the help of computational methods, pharmaceutics scientists can predict the properties and functions of DDSs with different formulation factors, as well as the interaction mechanisms between drugs, excipients and the human body can also be visually and dynamically explained at the molecular level. Successful applications of computational pharmaceutics have been reported on cyclodextrin-based formulations to improve the solubilization of poorly soluble drugs [[3], [4], [5]], polymer micellar drug delivery systems [[6], [7], [8], [9], [10], [11]], liposomes for drug delivery [[12], [13], [14]]and so on. The previous studies conducted by the investigators also employed coarse-grained molecular dynamics (CG-MD) simulation, a popular computational modeling method, to study the solubilization effect of the natural bio-surfactant of saponin [[15], [16], [17], [18]], and the effect of edge activators on the skin permeation behavior of transfersomes [19,20]. However, most of these researches are focused on a certain aspect of formulation development, and rarely provide regular and integrated guidance strategies of formulation design. The present study hopes to take a step in this area.

Hydrogel is a typical DDS with a three-dimensional network, and the interstitial space was filled up with water and other biological fluids. Due to the wide spectrum of both chemical and physical characteristics, hydrogels offer various advantageous properties, such as being usually non-toxic, biocompatibility, controllable drug release rate, various dosage form, and flexible route of administration, such as transdermal administration, transmucosal administration, oral administration, injection administration, etc.

Similar to most DDS, the screening of matrix materials at the pre-formulation stage and the research on the influence of formulation factors are very crucial works in hydrogel formulation design, because the matrix not only acts as a drug carrier but also affects the release of the drug. There are many polymer materials that can be used to prepare a hydrogel, such as carbopol, methylcellulose, chitosan, polyvinyl alcohol, polyvinylpyrrolidone, etc. However, the present screening methods for the selection of matrix materials mostly depend on experience, or a large number of experiments, and the R&D efficiency is usually low. The compatibility of the drug with the matrix is the most important index for the screen of the matrix. Conventional experimental methods to study drug-polymer compatibility include differential scanning calorimeter (DSC), optical microscope observation, and so on. In computational methods, solubility parameters (δ) and Flory-Huggins parameters (χ) are often used to evaluate the compatibility of substances, which can be easily calculated by much mature software [19,20]. In general, drugs and polymers with similar δ values or lower χ are more likely to be miscible. In the present study, a comparative study was conducted on the applicability of δ and χ to determine the optimal method for the study of drug-polymer matrix compatibility.

At the formulation design stage, the study for formulation factors of hydrogel is more complicated, because there are so many factors, as mentioned before, that can affect the performance of the hydrogel. Although these factors often work together, in the actual preparation process, it usually needs to adjust a certain formulation factor according to the specific needs, such as rapid or sustained drug release. Therefore, it is necessary to correctly evaluate the influence and mechanism of every formulation factor. This will inevitably require a large amount of experimental data for support, which consumes a lot of labor, material resources, and financial resources. Many researchers have attempted to use computational methods to study the performance of hydrogels and established many great numerical models. Although these work well in predicting the drug release from hydrogels, they are relatively abstract in mechanism interpretation and have many restrictions. In addition, no systematic method has been developed to guide the rational formulation design of hydrogels [[21], [22], [23], [24], [25]].

To systematically study the influence of formulation factors on the performance of the hydrogel, explain the influence mechanism, and further guide the design of the hydrogel, here, a multiscale dynamic simulation strategy was developed. According to the strategy, at the pre-formulation stage, the compatibility of the drug with different matrices was predicted by all-atoms molecular dynamics (AA-MD) simulation methods to screen the optimal matrix material of hydrogel; at the stage of formulation design, the effect of formulation factors, such as the DP of the matrix material solvent, pH, matrix concentration, drug concentration, and additives, on drug diffusion, as well as the interaction mechanisms between the drug and matrix in the hydrogels were explained by the CG-MD simulations, which simplify the details in the description on the atomistic models into coarse-grained representations, and is more effective and suitable for the study of formulation factors’ effects of polymer drug delivery systems. Capsaicin (CAP), a natural active pharmaceutical ingredient that is mostly developed as a hydrogel formulation for transdermal administration to avoid the strong liver first-pass effect and short half-life and achieve rapid onset [[26], [27], [28]], was taken as a model drug to investigate the drug-releasing mechanism of the hydrogel in this study. The classic in vitro experiment was conducted to evaluate the reliability of simulation results at the same time.