Synthesis, Characterization, and Analysis of Probenecid and Pyridine Compound Salts

Abstract

This study aimed to address the issue of the low solubility in the model drug probenecid (PRO) and its impact on bioavailability. Two salts of probenecid (PRO), 4-aminopyridine (4AMP), and 4-dimethylaminopyridine (4DAP) were synthesized and characterized by PXRD, DSC, TGA, FTIR, and SEM. The crystal structures of the two salts were determined by SCXRD, demonstrating that the two salts exhibited different hydrogen bond networks, stacking modes, and molecular conformations of PRO. The solubility of PRO and its salts in a phosphate-buffered solution (pH = 6.8) at 37 °C was determined, the results showed that the solubility of PRO salts increased to 142.83 and 7.75 times of the raw drug, respectively. Accelerated stability experiments (40 °C, 75% RH) showed that the salts had good phase stability over 8 weeks. Subsequently, Hirshfeld surface (HS), atom in molecules (AIM), and independent gradient model (IGM) were employed for the assessment of intermolecular interactions. The analyses of salt-forming sites and principles were conducted using molecular electrostatic potential surfaces (MEPs) and pK_a rules. The lattice energy (E_L) and hydration-free energy (E_HF) of PRO and its salts were calculated, and the relationships between these parameters and melting points and the solubility changes were analyzed.

1. Introduction

The synthesis of the multi-component crystals (such as cocrystals, salts, and solvates) [1] of active pharmaceutical ingredients (APIs) in order to enhance their physicochemical properties, including solubility, dissolution rate, bioavailability, stability, and hygroscopicity, has received increasing attention and found widespread application in the pharmaceutical industry [2,3]. Notably, 40% of currently marketed drugs exhibit poor water solubility [4], and the formation of salts is the preferred approach for addressing the low solubility of pharmaceutical compounds [5]. According to the IUPAC definition, a salt is “the substance formed by the combination of cations and anions”. The formation of a drug molecule as a salt requires the presence of an ionized API along with another counterion. Therefore, for drugs with ionization ability, synthesizing salts represents a highly established method for enhancing their physicochemical properties [6]. The formation of salt derivatives has been shown to significantly enhance the solubility, dissolution rate [7], stability [8,9,10], and bioavailability [11] of pharmaceutical compounds while also extending the patent exclusivity period for marketed drugs. It is estimated that more than 50% of drug molecules are administered in the form of salts [12].

Probenecid (PRO; Figure 1) (Chemical name: p-[(dipropyl amino) sulfonyl] benzoic acid, molecular formula: C₁₃H₁₉NO₄S, CAS: 57-66-9), a synthetic sulfonamide with the dual effects of promoting uric acid excretion and inhibiting penicillin excretion, has long been used in the treatment of chronic gout [13,14]. The latest research has demonstrated that bumetanide effectively modulates human renal physiological functions by inhibiting ATP transporter proteins in the renal collecting tubules and proximal tubules. Additionally, its potential to mitigate transient global cerebral ischemia/reperfusion injury in mice underscores its significant pharmaceutical value and wide-ranging applications [15,16]. However, according to the biopharmaceutical classification system (BCS) [17,18], PRO is a BCS II drug with good permeability but poor water solubility. According to the literature reports, its solubility in water at 37 °C is only 72.2 μg/mL [19], which influences its bioavailability. Although the water solubility of the PRO is significantly limited, there has been limited progress in terms of developing methods for synthesizing salts to enhance the water solubility of PRO. Recently, there have been studies on the elastic crystals formed by the coprecipitation of PRO with 4,4-azopyridine [20,21], 4,4-bipyridine [22], azacytidine, and piperazine [23]. Additionally, research has shown an enhancement in solubilitiy through the coprecipitation of PRO with benzamide [19,24] and 1,2-bis(4-pyridyl) ethene (2.43 times increase) [25]. Furthermore, there have been investigations into the N, N-dimethylformamide solvate and the pyridine solvate of PRO [22,23]. Due to the presence of a carboxyl group that can dissociate in PRO molecules, we aim to enhance its solubility through the synthesis of PRO salts. This study computed and analyzed the 3D full interaction map [26] (FIM) of PRO and selected pyridine-based basic compounds that are prone to proton transfer and form salts for experimental screening. Subsequently, we carried out experimental screening of the coformers by means of a liquid-assisted grinding approach, which is more efficient than conventional solvent-free mechanical grinding [27,28], and characterized them using powder X-ray diffraction (PXRD) and differential scanning calorimetry (DSC). We finally found that PRO can form a new phase with 4-aminopyridine (4AMP) and 4-dimethylaminopyridine (4DAP). As commonly basic coformers [29,30], 4AMP and 4DAP have been employed in the formation of salts with Furantoin [31], Piroxicam and Meloxicam [32], Ibuprofen [33], and other drugs to enhance the solubility and other physicochemical properties of these pharmaceutical compounds.

In this investigation, PRO was chosen as the model drug, and its salts were synthesized with coformers 4-aminopyridine (4AMP, CAS: 504-24-5) (PRO)⁻(4AMP)⁺ (1:1) and 4-dimethylaminopyridine (4DAP, CAS:1122-58-3) (PRO)⁻(4DAP)⁺ (1:1) (Figure 1). Moreover, the salts were comprehensively characterized by powder X-ray diffraction (PXRD), differential scanning calorimetry (DSC), thermogravimetric analysis (TGA), Fourier transform infrared spectroscopy (FTIR), and scanning electron microscopy (SEM). The results revealed distinct PXRD patterns, FTIR spectra, melting points, decomposition temperatures, and crystal morphologies for the salts compared to the raw materials. The crystal structures of the salts were determined by single crystal X-ray diffraction (SCXRD), revealing distinct hydrogen bond networks, packing modes, and PRO molecular conformations. Given the primary absorption site of PRO in the small intestine, equilibrium solubility data for PRO and its salts were obtained in a phosphate-buffered solution (pH = 6.8) at 37 °C to simulate intestinal conditions. Results indicated that the solubility of (PRO)⁻(4AMP)⁺ and (PRO)⁻(4DAP)⁺ increased to 142.83 and 7.75 times, respectively, compared to the raw drug, significantly improving the poor solubility of PRO. Accelerated stability experiments were conducted on the salts at 40 °C and 75% relative humidity (RH), and no observable crystalline changes were observed in either salt within 8 weeks, indicating excellent phase stability. The molecular interactions within the salts were analyzed using Hirshfeld surface (HS), atom in molecules (AIM), and independent gradient model (IGM). The salt-forming sites and principles were analyzed using molecular electrostatic potential surfaces (MEPs) and pK_a rules. The lattice energy (E_L) and hydration-free energy (E_HF) of PRO and its salts were calculated, and the relationships between these parameters and melting points and solubility changes were analyzed. It is particularly necessary to explain that 4AMP and 4DAP have a certain toxicity and are not suitable for human medication. However, the focus of this study is concept verification in regard to improving the solubility of soluble drugs through salt formation, which can provide new ideas for improving the solubility of drugs by designing salt formation.

2. Materials and Methods

2.1. Materials

The PRO (purity: 98.0%) and coformers (purity: ≥98.0%) used in this study were purchased from Shanghai Dibai Biotechnology Co., Ltd., Shanghai, China and Tianjin Xiensi Biotechnology Co., Ltd., Tianjin, China. All solvents were of analytical grade and provided by Tianjin Jiangtian Chemical Technology Co., Ltd., Tianjin, China. All chemicals can be directly used without further purification. Deionized water is prepared in the laboratory, as shown in Table S1.

2.2. Screening of Salts

The 3D full interaction map (FIM) of PRO were computed and generated by importing the .CIF format files of the PRO’s parent crystal structure (CCDC No: 987956) into Mercury 4.2.0 software (Cambridge Crystallographic Data Centre, Cambridge, UK) (Figure 2a) [34], which shows the potential hydrogen bond donor and acceptor sites in the PRO molecule. The dark outlined (red/blue) regions indicate a higher tendency for synthon formation, while the transparent regions have a lower tendency. The PRO molecule contains hydrogen bond donors and acceptors, such as carboxylic acid and sulfonamide groups (Figure 2b), that can form salts with basic compounds containing pyridine rings or other nitrogen-containing heterocycles [35] (Figure 2c). The coformers selected in this experiment include 4-aminopyridine (4AMP), 4-dimethoxypyridine (4DAP), 4-hydroxypyridine (4HDP), 2-pyridinecarboxamide (2PCA), 3-aminopyridine (3AMP), and 2,6-diaminopyridine (26DAP).

Experimental screening of the coformers was conducted using liquid-assisted grinding (LAG). An amount of 1 mmol of PRO and 1 mmol of the coformer were weighed using an analytical balance, then added to an agate mortar. Subsequently, 40 μL of ethanol was added in four equal portions, and the sample was dried at 50 °C for 24 h after grinding for 1 h. Finally, PXRD and DSC analyses were employed for characterization, as depicted in Figures S1 and S2. From these analyses, it can be observed that new peaks emerged in the PXRD patterns of PRO after grinding with 4AMP and 4DAP, and the DSC curves indicated the existence of new melting points. Consequently, we are convinced that PRO formed two new phases with 4AMP and 4DAP.

2.3. Synthesis of Salts

2.3.1. Synthesis of (PRO)⁻(4AMP)⁺

The PRO (285.4 mg, 1 mmol) and 4AMP (94.2 mg, 1 mmol) were weighed and added to 10 mL of acetone. The slurry was suspended at room temperature for 12 h, filtered, and then dried at 50 °C for 24 h to yield a white solid powder.

Due to the formation of oil during the slow solvent evaporation method for single crystal preparation, single crystal was obtained using the slow cooling crystallization method. PRO (57.1 mg, 0.2 mmol) and 4AMP (18.8 mg, 0.2 mmol) were weighed and added to 10 mL of acetone, the mixture was heated to dissolve at 50 °C before being cooled at a rate of 5 °C/h until colorless rod-shaped single crystals formed at a temperature of 20 °C.

2.3.2. Synthesis of (PRO)⁻(4DAP)⁺

The PRO (285.4 mg, 1 mmol) and 4DAP (122.2 mg, 1 mmol) were weighed and added to 10 mL of ethyl acetate. The slurry was then suspended at room temperature for 12 h, filtered, and then dried at 50 °C for 24 h to yield a white solid powder.

Single crystal was obtained using the slow solvent evaporation method. Additionally, PRO (28.6 mg, 0.1 mmol) and 4DAP (12.3 mg, 0.1 mmol) were dissolved in ultrasonicated ethyl acetate and slowly evaporated to obtain colorless rod-shaped single crystals after several days.

2.4. Characterization

2.4.1. Single Crystal X-ray Diffraction (SCXRD)

SCXRD measurement was conducted on a diffractometer (Saturn 70CCD, Rigaku, Japan) using Mo Kα radiation (λ = 0.71073 Å). The integration and scaling of the intensity data were performed by the CrysAlisPRO 1.171.39.46 program (Rigaku Oxford Diffraction, 2018). The structure was solved using direct methods in SHELXT [36] and refined using the full matrix least-squares method in SHELXL [37], both of which were performed under OLEX2-1.2 [38]. The non-H atoms were refined using anisotropic parameters. The riding model with U_iso = 1.2 − 1.5U_eq(C) was employed for the geometric localization and refinement of H atoms on C atoms, while the H atoms bonded to N and O atoms were located based on the residual peaks of electron density in the Fourier maps and refined with isotropic parameters. Absorption effect data were corrected using SADABS [39] and hydrogen bond data were detected using PLATON [40]. Mercury 4.2.0 software was used for visualization of the crystal structure [41].

2.4.2. Powder X-ray Diffraction (PXRD)

PXRD was measured using a powder X-ray diffractometer (MiniFlex600, Rigaku, Japan) with Cu Kα radiation (λ = 1.54178 Å). The working voltage and current were 40 kV and 100 mA, respectively. The samples were measured within the 2θ range of 2–35° with a step size of 0.01° and a scanning rate of 10°/min. The data were collected at room temperature. The simulated PXRD patterns were obtained using Mercury 4.2.0 software.

2.4.3. Differential Scanning Calorimetry (DSC)

DSC was performed on a Mettler DSC 1 STARe system (Mettler–Toledo, Greifensee, Switzerland) that was calibrated with indium standards prior to analysis. A sample of 5–10 mg was heated in a standard aluminum crucible with a needle hole in a nitrogen atmosphere at a heating rate of 10 °C/min and a nitrogen flow of 70 mL/min. An empty aluminum crucible with a needle hole was used as the reference.

2.4.4. Thermogravimetric Analysis (TGA)

TGA was performed using the Mettler TGA/DSC 1 STARe system (Mettler–Toledo, Greifensee, Switzerland) under a nitrogen atmosphere with a flow rate of 40 mL/min. Each analysis utilized approximately 5–10 mg of a sample placed in an alumina crucible. The standard uncertainty was ±0.00001 g and the sample heating rate was set at 10 °C/min.

2.4.5. Fourier Transform Infrared Spectroscopy (FT-IR)

The ALPHA II infrared instrument (Bruker, Ettlingen, Germany) equipped with an ATR attachment was used to collect infrared spectra in the scanning range of 4000 to 400 cm⁻¹ under ambient conditions. At least 16 spectra were collected for each sample, and the average value was taken. The instrument resolution was 4 cm⁻¹.

2.4.6. Scanning Electron Microscopy (SEM)

To analyze the crystal morphology of the samples, images of the samples were recorded using a scanning electron microscope (SEM, TM3000, Hitachi Corporation, Tokyo, Japan).

2.4.7. Equilibrium Solubility Measurement

The equilibrium solubility of PRO and its salts in a phosphate-buffered solution (pH = 6.8, simulating intestinal fluid) was determined at 37.0 ± 0.1 °C using the shake flask method. The phosphate-buffered solution (pH = 6.8) was prepared according to the Chinese Pharmacopoeia (2015 edition) [42]. An excess sample was suspended in the buffer solution at 37.0 ± 0.1 °C for 48 h to achieve equilibrium, followed by filtration of the supernatant through a PTFE filter membrane (0.45 μm) and appropriate dilution. The liquid chromatography peak areas of a series of PRO solutions with known concentrations ranging from 10 to 400 μg/mL were determined. Subsequently, with the concentrations and chromatographic peak areas serving as the abscissa and ordinate, respectively, the least square method was employed to linearly fit the scattered data, thereby obtaining the working curve for solubility testing. The experiment were performed using high-performance liquid chromatography (Agilent 1220, Shanghai, China) under specific test conditions, including an Agilent Extend C18 Column (250 × 4.6 mm, 5 μm), a column temperature of 30 °C, a UV detection wavelength of 245 nm, a methanol–water mobile phase (75:25), and a flow rate of 1.0 mL/min [43]. The PXRD analysis was employed for monitoring phase transitions. The equilibrium solubility experiment was repeated three times for accuracy.

2.4.8. Accelerated Stability Experiment

In order to test the stability of PRO and its salts, accelerated stability experiments were conducted in a medicine stability test chamber (HWS-70B, Taisite Instrument Co., Ltd., Tianjin, China) at 40 ± 1 °C and 75 ± 1.5% RH. After 8 weeks, the samples were removed from the chamber and analyzed using PXRD to evaluate their phase stability.

2.5. Computational Details

2.5.1. Hirshfeld Surface (HS)

The Hirshfeld surface (HS) and two-dimensional (2D) fingerprint diagram were computed and generated by importing the .CIF format files of the PRO and its salts into CrystalExplorer 17.5 software for a clear and intuitive study of molecular interactions within the crystal structure as well as for quantitative analysis [44,45].

2.5.2. Atom in Molecules (AIM) and Independent Gradient Model (IGM)

The crystals structures of PRO [34] and its salts were optimized using the CASTEP module of Materials Studio 6.0 (Accelrys, San Diego, CA, USA) [46]. Structural relaxation of the positions of all atoms was performed by plane wave density functional calculations, the cell parameters were fixed during the optimization process. The Perdew–Burke–Ernzerhof (PBE) generalized gradient approximation (GGA) was used as an exchange–correlation density functional [47] along with ultra-soft pseudopotential [48] plus Grimme’s D2 [49] dispersion correction. A cut-off energy of 780 eV was set, and a Brillouin zone integration was performed using discrete 2 × 2 × 2 k-point sampling of the original cell. For structural relaxation, thresholds for energy, force, and atomic shift were set to 10⁻⁵ eV, 5 × 10⁻² eV/A, and 10⁻³ A, respectively. After geometry optimization, molecular clusters required for calculations were extracted from crystal supercells. Single point energy levels of the molecular cluster were calculated at the RI [50]-wB97M-V [51]/def2-TZVP [52] level using ORCA 4.2.0 [53] to generate wave functions for AIM [54] and IGM [55] analysis.

2.5.3. Molecular Electrostatic Potential Surfaces (MEPs)

The geometric optimization of raw compounds was conducted using ORCA 4.2.0 software at the RI-B3LYP-V/def2-TZVP level while obtaining wavefunction files for both the molecule and salts to facilitate analysis of molecular electrostatic potential surfaces (MEPs). Subsequently, visualization of the MEPs, AIM, and IGM was achieved using Multiwfn 3.8 [56,57,58] and VMD 1.9.3 [59].

2.5.4. pK_a

The pK_a values of PRO and coformers were computed utilizing MarvinSketch (ChemAxon, Budapest, Hungary) [60,61].

2.5.5. Lattice Energy (E_L) and Hydration-Free Energy (E_HF)

In this study, the Forcite module under Materials Studio 6.0 (Accelrys, San Diego, CA, USA) was utilized to compute the lattice energy (E_L) for assessing the lattice strength of PRO and its salts. The geometric optimization of the lattice was accomplished with the Compass force field method [62,63], and the calculation of E_L can be shown in Equation (1) as follows:

E_L = E_(bulk)/Z − E_A − E_B

(1)

where Z represents the number of asymmetric units, E_(bulk) represents the energy of a crystal unit cell, and E_A and E_B represent the relaxed energy of A and B molecules (kcal/mol), respectively.

The uESE method is a simple, efficient and accurate method for calculating the solvation free energy of ions and neutral molecules. For anionic systems in particular, the uESE method is more convenient and accurate than the traditional SMD method [64]. The specific procedural steps involve geometric optimization of cations, anions, and PRO molecules at the RI-B3LYP-D3(BJ)/def2-TZVP level using ORCA 4.2.0, followed by inputting the optimized structures into Multiwfn 3.8 software to generate calculation files for the uESE input file. Finally, the uESE program is utilized for computing hydration-free energy.

3. Results and Discussion

3.1. Crystal Structure Analysis

3.1.1. Single Crystal X-ray Diffraction Analysis

The single crystal structures of (PRO)⁻(4AMP)⁺ and (PRO)⁻(4DAP)⁺ were determined through SCXRD analysis, with the corresponding crystallographic data and refined details being presented in

Table 1. Crystallographic data and structural refinement details of (PRO)⁻(4AMP)⁺ and (PRO)⁻(4DAP)⁺.

Compound	(PRO)−(4AMP)+	(PRO)−(4DAP)+
Empirical formula	C18H25N3O4S	C20H29N3O4S
Formula weight	379.47	407.52
Temperature/K	293 (2)	293 (2)
Crystal system	orthorhombic	monoclinic
Space group	P212121	P21/c
a/Å	7.1146 (1)	6.8830 (1)
b/Å	11.2314 (2)	7.1512 (2)
c/Å	24.3877 (4)	43.1607 (8)
α/°	90	90
β/°	90	93.680 (2)
γ/°	90	90
Volume/Å3	1948.75 (5)	2120.06 (8)
Z	4	4
ρcalc/(g/cm3)	1.293	1.277
F (000)	808.0	872.0
2θ range for data collection/°	3.340 to 72.258	6.002 to 57.394
Goodness-of-fit on F2	1.025	1.100
Rint	0.0613	0.0502
R1 indexes [I ≥ 2σ (I)]	0.0590	0.0534
ωR2 indexes [all data]	0.1280	0.1496
CCDC No.	2283803	2360637

. These specific datasets have been deposited in the Cambridge Crystallographic Data Centre under CCDC numbers 2283803 and 2360637. Additionally, the C-O bond length data for the carboxyl group of PRO salts were summarized in

Table 2. C-O bond length distribution of the salts of PRO.

Compound	DC-O (Å)	ΔDC-O (Å)	Proton Transfer
(PRO)−(4AMP)+	C1-O1:1.251, C1-O2:1.263	0.012	Yes
(PRO)−(4DAP)+	C1-O1:1.265, C1-O2:1.241	0.024	Yes

.

Table 3. Hydrogen bonding parameters of (PRO)⁻(4AMP)⁺ and (PRO)⁻(4DAP)⁺.

D-H⋯A	D-H(Å)	H⋯A(Å)	D⋯A(Å)	D-H⋯A(°)	Symmetry
(PRO)−(4AMP)+
N2+-H2⋯O2−	0.861	1.885	2.732 (3)	167.36
N3-H3A⋯O1	0.860	1.965	2.812 (3)	167.77	x, −1/2 + y, 3/2 − z
N3-H3B⋯O2−	0.861	2.042	2.888 (3)	167.51	1 − x, −1/2 + y, 3/2 − z
C12-H12B⋯O3	0.970	2.585	3.268 (3)	127.55	1 + x, y, z
C14-H14⋯O4	0.930	2.529	3.203 (3)	129.62	−1/2 + x, 1/2 − y, 1 − z
C18-H18⋯O4	0.930	2.556	3.357 (3)	144.50	1/2 + x, 1/2 − y, 1 − z
(PRO)−(4DAP)+
N2+-H2⋯O1−	0.859	1.858	2.7130 (19)	172.85
C10-H10C⋯O4	0.961	2.573	3.465 (19)	154.49	x, 1 + y, z
C14-H14⋯O2	0.930	2.350	3.213 (2)	154.29	−1 + x, y, z
C18-H18⋯O2	0.930	2.356	3.036 (2)	129.74

, Tables S2 and S3 contain information on hydrogen bonds and other weak interactions information within the salts’ crystal structure.

The bond length of carboxyl groups in single crystal data is frequently utilized for distinguishing between cocrystals and salts. The disparity in bond length between the two C-O bonds within the carboxyl group, denoted as ΔD_C-O, serves as a crucial metric for discerning proton transfer. A smaller ΔD_C-O value signifies the formation of a salt via proton transfer, whereas a larger value (>0.08 Å) indicates cocrystal formation without proton transfer [65]. Both single crystal structures in this study exhibit relatively small ΔD_C-O values (Table 2), corroborating proton transfer and confirming the formation of salt rather than cocrystals.

(PRO)⁻(4AMP)⁺ salt (1:1) The crystal structure of (PRO)⁻(4AMP)⁺ salt crystallizes in the orthorhombic crystal system with a space group of P2₁2₁2₁ (Z = 4). The asymmetric unit consists of one (PRO)⁻ anion and one (4AMP)⁺ cation. In the asymmetric unit, the H atom on the carboxyl group of the PRO molecule transfers to the N atom on the pyridine ring of the 4AMP molecule, forming the salt. Since the benzene ring of the (PRO)⁻ anion and the pyridine ring of the (4AMP)⁺ cation were reversed by 36.43° (Figure S3), no cyclic hydrogen bond motif was formed. The (PRO)⁻ anion and (4AMP)⁺ cation are connected through a charge-assisted hydrogen bond N₂⁺-H₂⋯O₂⁻ (1.885 Å; 167.36°) (Figure 3a). The asymmetric units are connected into a 1D chain through hydrogen bond C₁₂-H_12B⋯O₃ (2.585 Å; 127.55°; 1 + x, y, z) interactions (Figure 3b). The 1D chains are further connected into a double 1D chain through hydrogen bond (C₁₄-H₁₄⋯O₄ (2.529 Å; 129.62°; -1/2 + x, 1/2 − y, 1 − z) and C₁₈-H₁₈⋯O₄ (2.556 Å; 144.50°; 1/2 + x, 1/2 − y, 1 − z)) (Figure 3c). In addition, weak interactions, such as C₇-H₇⋯π (2.721 Å, 148.95°; 1/2 + x, 1/2 − y, 1 − z), and C₁₃-H_13A⋯π (2.862 Å, 154.75°; 1/2 + x, 1/2 − y, 1 − z) (Figure 3c), play a crucial role in the crystal structure. Observed from the a-axis direction, the double chains are connected into a 3D structure through hydrogen bond (N₃-H_3B⋯O₂⁻ (2.042 Å; 167.51°; 1 − x, −1/2 + y, 3/2 − z) and N₃-H_3A⋯O₁ (1.965 Å; 167.77°; −x, −1/2 + y, 3/2 − z)) (Figure 3d).

(PRO)⁻(4DAP)⁺ salt (1:1) The crystal structure of (PRO)⁻(4DAP)⁺ salt crystallizes in the monoclinic P2₁/c (Z = 4) space group, with one (PRO)⁻ anion and one (4DAP)⁺ cation in the asymmetric unit (Figure 4a). There is overall disorder at the terminal propyl group of PRO, with PART1 accounting for 0.585 (6), and the crystal structure is resolved using the larger PART1 fraction. Proton transfer is detected from the carboxyl group of PRO to the pyridine N atom of 4DAP, leading to the formation of salt. The phenyl ring of the (PRO)⁻ anion and the pyridine ring of the (4DAP)⁺ cation exhibit a slight tilt, with only a 5.26° rotation occurring between their planes (Figure S3), so (PRO)⁻ anions and (4DAP)⁺ cations form a cyclic motif $R_2^2$ (7) through hydrogen bond C₁₈-H₁₈⋯O₂ (2.356 Å; 129.74°) and charge assisted hydrogen bond N₂⁺-H₂⋯O₁⁻ (1.858 Å; 172.85°). Observed from the b-axis, the asymmetric units are connected by hydrogen bonds C₁₄-H₁₄⋯O₂ (2.350 Å; 154.29°; −1 + x, y, z) to form 1D chains (Figure 4b). Then, observed from the a-axis, the 1D chains are connected by C₁₀-H_10C⋯O₄ (2.573 Å; 154.49°; x, 1 + y, z) to form 2D planes (Figure 4c). The 2D planes are connected by C₂₀-H_20A⋯π (2.653 Å; 148.86°; 1 − x, 1 − y, 1 − z) and π⋯π (3.800 Å; 0.03(9)°; −x, 2 − y,1 − z) weak interactions to form 2D double-layer planes. The centroid–centroid distance within the π⋯π stacking amounts to 3.800 Å; the plane–plane distance is 3.448 Å; the angle between the normal of the plane and the centroid connection is 24.90°; the slippage is 1.597 Å. According to Christoph Janiak et al., the aforementioned π⋯π stacking represents a typical parallel-displaced π⋯π stacking of rings and encompasses the contribution of π⋯σ attraction [66]. Subsequently, the 2D double-layer planes are interconnected along the c-axis via weak interlayer interactions, thereby forming a complete 3D crystal structure (Figure 4d).

3.1.2. Conformation and Packing Similarity Analysis

The flexibility of PRO molecular conformation can significantly impact the assembly of multi-component crystals. In this investigation, we examined the diverse conformations of PRO in both its parent crystal [34] and salt crystals, followed by overlaying and analyzing the conformations using Mercury 4.2.0 software (Figure 5a). It is evident from the data that the terminal dipropyl amino group is particularly prone to distortion within the PRO molecule. Furthermore, proton transfer during salt formation induces specific alterations in PRO conformation. Notably, compared to the PRO parent crystal, the conformation of PRO in the (PRO)⁻(4AMP)⁺ salt crystal exhibits a certain degree of resemblance; conversely, significant differences are observed between the conformation of PRO in the (PRO)⁻(4DAP)⁺ salt crystal and that in the parent crystal. Specifically, within the (PRO)⁻(4DAP)⁺ salt structure, rotation of the carboxyl group occurs relative to both (PRO)⁻(4AMP)⁺ salt and pure PRO crystalline forms; consequently, these two salts are formed through charge-assisted hydrogen bonds N₂⁺-H₂⋯O₁⁻ and N₂⁺-H₂⋯O₂⁻ (Figure S3).

The molecular packing similarity of two salts was further evaluated using Mercury 4.2.0 software, taking into consideration various structural and ensemble factors. Comparison of the analysis of seven molecules revealed that only one molecule exhibited similar packing positions (Figure 5b). This suggests a poor similarity in crystal packing among them, possibly due to variances in the hydrogen bond networks of the two salts (Figure 5c–d).

3.2. Powder Characterization Analysis

3.2.1. Powder X-ray Diffraction Analysis

The experiment involved the preparation of powder samples of (PRO)⁻(4AMP)⁺ and (PRO)⁻(4DAP)⁺ salts using the slurry suspension method, followed by their characterization using PXRD. The single crystal structures obtained from SCXRD for the salt crystals were utilized to simulate the PXRD pattern of the salts, as shown in Figure 6. It is evident from the figure that the characteristic peaks of the obtained salts are distinctly discernible from those of the initial raw materials, thereby confirming the formation of new high-purity phases rather than physical mixtures. Furthermore, there is a fundamental consistency between the experimental PXRD patterns and simulated PXRD patterns based on a single crystal structure, thus substantiating both representativeness and high crystallinity.

3.2.2. Thermal Analysis

DSC is a commonly used characterization method for analyzing the thermal behavior of drugs. The formation of salts can change the original chemical structure and cause changes in thermal properties, so the generation of new phases can be determined by the differences in thermal characteristics during drug heating. The DSC curves of the raw materials and salts are shown in Figure 7, and the results show that the melting point (T_onset) of PRO, 4AMP, and 4DAP are 198.73 °C, 157.42 °C, and 111.75 °C, respectively. The experimental melting point of (PRO)⁻(4AMP)⁺ is 132.04 °C, which is lower than the melting point of raw materials. However, the melting point of (PRO)⁻(4DAP)⁺ is 141.54 °C, which is between the melting points of the two starting materials. The melting enthalpies of (PRO)⁻(4AMP)⁺ and (PRO)⁻(4DAP)⁺ are −9.57 kcal/mol and −10.51 kcal/mol, respectively. By analyzing the results together with the PXRD results, it was shown that the formation of salts, rather than physical mixtures, was responsible for the new phase.

TGA is a widely utilized method for thermal analysis, providing valuable insights into the thermal stability and solvent composition of pharmaceuticals. The TGA-DSC curves in Figure 8 reveal that (PRO)⁻(4AMP)⁺ salt exhibits no weight loss prior to decomposition, possesses a melting point lower than the decomposition temperature, and demonstrates robust thermal stability. Conversely, (PRO)⁻(4DAP)⁺ salt undergoes simultaneous melting and decomposition, indicating inferior thermal stability. These observations may be attributed to the lower boiling point of 4DAP (162 °C) compared to that of 4AMP (273 °C). Additionally, the absence of residual solvents (including water) in the medicine salts is evident from the figure.

3.2.3. Spectral Analysis

In order to gain deeper insights into the salt formation process, FTIR spectroscopy was employed to analyze the salts of PRO and their respective raw materials. The comparative results are presented in Figure S4, with detailed characteristic peak information being provided in Table S4. For pure PRO, the characteristic peaks were observed at 1686, 2964, 1284, and 1341 cm⁻¹, corresponding to C=O stretching vibration, O-H stretching vibration, C-N stretching vibration, and S=O stretching vibration, respectively. As for coformers, the C=N stretching vibrations of 4DAP and 4AMP were identified at 2900 cm⁻¹ and 3003 cm⁻¹, respectively. No N-H vibration peak was detected in pure 4DAP; however, a peak appeared at 3072 cm⁻¹ in the formed salt, indicating N-H stretching vibration and confirming proton transfer occurrence. A strong characteristic stretching vibration corresponding to NH₂ was detected at 3432 cm⁻¹ in pure 4AMP. Nevertheless, upon salt formation, the NH₂ stretching vibration vanished, and, subsequently, a broad peak emerged within the range of 3362–3318 cm⁻¹. These manifestations suggest that the amino groups have associated. In combination with the SCXRD analysis in Section 3.1.1, since the acidic proton forms a hydrogen bridge between the N atom on the 4AMP pyridine ring and the O atom of the PRO carboxyl group and the NH₂ of 4AMP and the carboxyl group of PRO form strong N-H⋯O hydrogen bonds, these factors lead to the reduction of the stretching vibration intensity and frequency of NH₂. Furthermore, the shift of the C=O peak to a lower frequency subsequent to salt formation further implies that the original hydrogen bond network is disrupted and substituted by a newly formed hydrogen bond network, thereby further validating the formation of the salt.

3.2.4. SEM Analysis

The prepared powders were subjected to SEM analysis to assess alterations in crystal morphology compared to the starting material (Figure 9). As depicted in the picture, PRO demonstrates elongated plate-like crystals, 4AMP displays block-shaped crystals, and 4DAP exhibits thick plate-like crystals (Figure 9a–c). Interestingly, the (PRO)⁻(4AMP)⁺ salt manifests a sleek rod-like crystal structure on its surface (Figure 9d), whereas the (PRO)⁻(4DAP)⁺ salt presents a rugged and fragmented plate-like crystal structure on its surface (Figure 9e). These phenomena imply that the salts synthesized experimentally exhibit distinct crystal morphologies compared to the raw materials.

3.2.5. Equilibrium Solubility Analysis

By determining the liquid chromatographic peak areas of a series of PRO solutions with known concentrations (10–400 μg/mL) and then taking the solution concentration C (μg/mL) as the abscissa and the chromatographic peak area A (mAU*s) as the ordinate, as well as performing linear fitting using the least squares method on the data, the working curve A = 40.4293 × C + 23.0174 for the solubility test was obtained, and the value of R² is 0.9999. The working curve is shown in Figure S5. Moreover, since it is known that the main absorption site of PRO is in the small intestine, the equilibrium solubility data of PRO, (PRO)⁻(4AMP)⁺ and (PRO)⁻(4DAP)⁺ were measured in phosphate-buffered solution (pH = 6.8) at 37 °C, and the residual solids after dissolution experiment were filtered and analyzed using PXRD to determine the type of solid phase at equilibrium. The results showed that both salts retained their original crystal patterns after the dissolution experiment (Figure S6). PRO, (PRO)⁻(4AMP)⁺ and (PRO)⁻(4DAP)⁺ were 4.61 mg/mL, 658.45 mg/mL, and 35.71 mg/mL, respectively, in phosphate-buffered solution at 37 °C (pH = 6.8). Among them, the solubility of (PRO)⁻(4AMP)⁺ and (PRO)⁻(4DAP)⁺ increased to 142.83 and 7.75 times that of PRO, respectively, as shown in

Table 4. (PRO)⁻(4AMP)⁺ and (PRO)⁻(4DAP)⁺ equilibrium solubility data (PBS 6.8).

Compound	Equilibrium Solubility (mg/mL)	The Increasing Factor
PRO	4.61
(PRO)−(4AMP)+	658.45	142.83
(PRO)−(4DAP)+	35.71	7.75

. The increase in solubility may be due to ionization and the reduction in lattice energy after salt formation, while water is a polar solvent and easily dissolves various ionic compounds.

3.2.6. Accelerated Stability Analysis

The stability of PRO and the salts (PRO)⁻(4AMP)⁺ and (PRO)⁻(4DAP)⁺ was assessed under accelerated conditions of 40 ± 1 °C and 75 ± 1.5% RH for 8 weeks (Figure S7). The PXRD patterns remained unchanged before and after the accelerated stability experiment, indicating high stability in the drug salts under elevated humidity without undergoing phase transformation or decomposition, which is crucial for sample storage.

3.3. Computational Analysis

3.3.1. HS Analysis

In order to visualize the molecular interactions within the multi-component crystals, the 3D Hirshfeld surfaces of (PRO)⁻(4AMP)⁺ and (PRO)⁻(4DAP)⁺ were computed using CrystalExplorer 17.5 software. The resulting 3D Hirshfeld surface is depicted in Figure 10a,b, with distinct colors denoting various interaction types: red signifies strong interactions, white denotes moderate interactions, and blue indicates negligible interactions. Analysis of the 3D Hirshfeld surface diagram reveals that the red regions are predominantly localized around the carboxyl group of the PRO molecule, signifying its a primary hydrogen bonding site involved in intermolecular interactions. In order to further investigate the weak interactions within the crystal, additional analysis of the molecular packing was performed, resulting in the generation of color mapping maps for shape index and curvedness on the Hirshfeld surface (Figure 10a,b). The red regions and blue regions in the shape index map correspond to depressions and protrusions, respectively. Alternating red and blue regions on the PRO surface indicate various weak interactions connecting PRO molecules with ligand molecules. Furthermore, analysis of the curvedness map reveals that the PRO molecule exhibits a broad, flat surface, suggesting planar stacking interactions on its sides (e.g., π⋯π interaction).

To further quantify the degree of molecular interactions, 2D fingerprint plots were computed for (PRO)⁻(4AMP)⁺ and (PRO)⁻(4DAP)⁺, with d_i and d_e denoting the distances from points on the Hirshfeld surface to the nearest nucleus inside and outside the surface, as depicted in Figure 10c,d. The predominant portion of the fingerprint spectrum corresponds to H-H contacts, while the protruding segment represents O-H contacts that contribute to O⋯H hydrogen bonds within the crystal structure. Due to nitrogen atoms in the pyrimidine ring being involved in salt formation, N-H contacts do not exhibit a protruding segment; instead, C-H contacts primarily contribute to another region.

Furthermore, the 2D fingerprint plots in Figure 10c,d illustrate varying proportions of interactions, with detailed data being provided in Table S5. In the case of salts (PRO)⁻(4AMP)⁺ and (PRO)⁻(4DAP)⁺, the predominant H-H contacts denote van der Waals interactions, constituting the highest percentage among all interactions. The 2D plane of (PRO)⁻(4DAP)⁺ primarily forms a 3D structure through van der Waals forces, resulting in significantly larger H-H contacts (55.3%) compared to those of (PRO)⁻(4AMP)⁺ (49.6%). Additionally, O-H contacts exhibit higher percentages (24.0% and 32.5%), signifying the crucial role of O⋯H hydrogen bonds in crystal packing. The C-H contacts (22.8%) observed in (PRO)⁻(4AMP)⁺ are largely attributed to C-H⋯π weak interactions that contribute to structural stability. Both compounds have minimal N-H contacts (1.2% and 0.5%, respectively) as a result of nitrogen atoms participating in salt formation within the pyrimidine ring structure; however, there is a higher proportion of C-C contacts for (PRO)⁻(4DAP)⁺ at 0.5%, indicating that the π⋯π weak interactions contributes to structural stabilization.

3.3.2. MEPs Analysis

Molecular electrostatic potential surfaces (MEPs) has been widely utilized for predicting molecular interaction sites and characterizing molecular recognition patterns within crystal structures. The MEP value reflects the strength of these interactions, which is crucial for understanding the formation of multi-component crystals. Figure 11a–c illustrates the MEPs of PRO, 4AMP, and 4DAP before salt formation, with orange denoting positive electrostatic potential and blue representing negative electrostatic potential. The peaks and troughs in the electrostatic potential are highlighted by yellow and blue spheres, respectively.

The global maximum electrostatic potential (+63.06 kcal/mol) for the PRO molecule is located at the O-H bond of the carboxyl group, indicating its propensity to donate a proton or act as a hydrogen bond donor. The primary hydrogen bond acceptor is the O atom of the carboxyl group, with an electrostatic potential of −33.10 kcal/mol. Conversely, the O atom of the sulfonyl group exhibits the global minimum electrostatic potential (−41.99 kcal/mol). According to the principle of electrostatic potential complementarity, the area with the highest electrostatic potential in PRO tends to be bind with the regions with the lowest electrostatic potential in 4AMP and 4DAP. Both 4AMP (−45.14 kcal/mol) and 4DAP (−47.06 kcal/mol) exhibit the lowest electrostatic potential at the N atom of the pyridine ring, thereby attracting the carboxyl group in PRO to facilitating proton transfer, leading to the formation of (N₂⁺-H₂⋯O₂⁻) and (N₂⁺-H₂⋯O₁⁻) salts. The H atom in the amino group of 4AMP exhibits the highest electrostatic potential (+43.35 kcal/mol), whereas in 4DAP, the H atom of the pyridine ring demonstrates a relatively higher electrostatic potential (+19.15 kcal/mol) and tends to form hydrogen bonds with the carboxyl group and sulfonic group of PRO.

Additionally, following the formation of salts, as shown in Figure 11d,e, the positive potential becomes concentrated on the 4AMP and 4DAP cations while the negative potential is primarily focused on the PRO anion due to proton transfer neutralizing the electrostatic potential in PRO, 4AMP, and 4DAP. After salt formation, PRO engages with 4AMP and 4DAP through a highly compatible electrostatic complementary interaction, thereby promoting overall system stability.

3.3.3. pK_a Analysis

According to the ΔpK_a empirical theory (pK_a (base) − pK_a (acid)), a cocrystal is formed when ΔpK_a < 0 and a salt is formed when ΔpK_a > 3. When 0 < ΔpK_a < 3, either a salt or a cocrystal may be formed, which is in the intermediate state of proton transfer. The ΔpK_a values of PRO, 4AMP, and 4DAP are 3.3 and 3.9, respectively, indicating that salts are formed, which is also confirmed by the SCXRD results (Table S6). However, in a recent study, Cruz-Cabeza re-examined the ΔpK_a rule by statistically analyzing the dataset of 6465 crystal complexes and discovered that the region of the intermediate state of proton transfer is broader, ranging from −1 < ΔpK_a < 4 [67]. Based on this, we can infer that the two new solid forms prepared in this experiment are in the intermediate state of proton transfer, being salt–cocrystal continua [68] rather than complete salts.

3.3.4. AIM Analysis

The purpose of AIM analysis is to investigate the characteristics of molecular or intermolecular interactions, with a particular focus being placed on the essential method for understanding hydrogen bond properties. Quantification of the topological parameters at the bond critical point (BCP) yields crucial insights into describing potential intermolecular interactions. In this study, we assessed the strength of hydrogen bonds in PRO and its salts using electron density (ρ), Laplacian operator (∇²ρ), electron kinetic density (G), electron potential density (V), and total electron energy density (H) [69]. The AIM topological pathway is illustrated in Figure 12, while the topological parameters of the hydrogen bond BCP are presented in

Table 5. Hydrogen bond topological parameters for PRO, (PRO)⁻(4AMP)⁺ and (PRO)⁻(4DAP)⁺ in AIM analysis.

D-H⋯A	H⋯A (Å)	ρ (a.u.)	∇2ρ (a.u.)	G (a.u.)	V (a.u.)	H (a.u.)	EH (kcal/mol)
PRO
O1-H1⋯O2	1.796	0.0338	0.1434	0.0343	−0.0328	0.0016	−6.79
(PRO)−(4AMP)+
N2+-H2⋯O2−	1.885	0.0297	0.1267	0.0288	−0.0260	0.0028	−10.95
N3-H3A⋯O1	1.965	0.0224	0.1078	0.0228	−0.0186	0.0042	−8.50
N3-H3B⋯O2−	2.042	0.0195	0.0850	0.0180	−0.0147	0.0033	−7.54
C12-H12B⋯O3	2.585	0.0076	0.0273	0.0059	−0.0049	0.0010	−3.59
C14-H14⋯O4	2.529	0.0080	0.0308	0.0064	−0.0052	0.0013	−3.71
C18-H18⋯O4	2.556	0.0080	0.0271	0.0059	−0.0049	0.0009	−3.74
(PRO)−(4DAP)+
N2+-H2⋯O1−	1.858	0.0311	0.1332	0.0305	−0.0277	0.0028	−11.41
C10-H10C⋯O4	2.573	0.0008	0.0039	0.0007	−0.0004	0.0003	−1.33
C14-H14⋯O2	2.350	0.0104	0.0397	0.0081	−0.0062	0.0019	−4.52
C18-H18⋯O2	2.356	0.0128	0.0473	0.0100	−0.0082	0.0018	−5.30

.

As shown in Figure 12, the BCPs and bond path are marked with orange dots and lines to indicate the complex interactions between the PRO parent crystal and its salt crystals. Additionally, Table 5 indicates that the ranges of ρ and ∇²ρ are 0.0008–0.0311 a.u. and 0.0039–0.1332 a.u., respectively. The values of BCPs are within the acceptable range for hydrogen bonds, signifying the presence of hydrogen bonds in PRO, (PRO)⁻(4AMP)⁺, and (PRO)⁻(4DAP)⁺. Furthermore, the nature of hydrogen bonds can be assessed through ∇²ρ and H values. Rozas et al. [70] have established the following standard guidelines for hydrogen bond interactions: when ∇²ρ < 0, it represents a strong covalent interaction; when ∇²ρ > 0 and H < 0, it represents a partially covalent interaction; when ∇²ρ > 0 and H > 0, it represents a weak electrostatic interaction. Theoretical foundations are provided by these standards for distinguishing the strength levels of hydrogen bonds. Emamian and Lu’s method for calculating the hydrogen bonding energy (E_H) in a neutral system is shown in Equation (2) [54].

E_H = −223.08 × ρ_(BCP)/(a.u.) + 0.7423

(2)

The calculation method for the hydrogen bond energy (E_H) of an electrostatic system such as salt is shown in Equation (3).

E_H = −332.34 × ρ_(BCP)/(a.u.) − 1.0661

(3)

The calculated hydrogen bond energies (E_H) of the original PRO crystal form, and its salts are summarized in Table 5.

In the crystal structure of the PRO parent crystal, the primary hydrogen bonds occur between the carboxyl groups of PRO dimers with a strength of −6.79 kcal/mol. Upon salt formation, disruption of the carboxyl dimer in PRO occurs. In (PRO)⁻(4AMP)⁺, the main hydrogen bonds are formed between the carboxyl group of PRO and the N atom on the pyridine ring of 4AMP through charge-assisted hydrogen bonding (N₂⁺-H₂⋯O₂⁻), as well as between the carboxyl group of PRO and the amino group of 4AMP through hydrogen bonding (N₃-H_3A⋯O₁ and N₃-H_3B⋯O₂⁻), with strengths of −10.95 kcal/mol, −8.50 kcal/mol, and −7.54 kcal/mol, respectively. In the crystal structure of (PRO)⁻(4DAP)⁺, the primary hydrogen bond is the charge-assisted hydrogen bond (N₂⁺-H₂⋯O₁⁻) between the carboxyl group of PRO and the N atom on the pyridine ring of 4DAP, its strength is −11.41 kcal/mol. Computational results indicate that that the main hydrogen bond strength of (PRO)⁻(4DAP)⁺ is greater than that of (PRO)⁻(4AMP)⁺, so the melting point is higher; however, the latter exhibits a more robust N-H⋯O hydrogen bond network, resulting in enhanced thermal stability.

3.3.5. IGM Analysis

Molecular cluster fragments containing all types of hydrogen bonds were extracted from PRO and its salts. The intermolecular interactions in PRO and the two salt molecular clusters were analyzed using an IGM. A δ_ginter contour map filled with sign (λ2) ρ colors was generated, as depicted in Figure 13. Different interaction types were distinguished by varying ellipsoid colors: green denoting van der Waals forces and blue representing hydrogen bonds. In the PRO single crystal dimer, green-wrapped blue oblate ellipsoids indicated the presence of van der Waals interactions and hydrogen bonds, with the latter being predominantly present.

After the formation of salt, the carboxyl dipolar groups of PRO are disrupted, leading to the development of a more intricate hydrogen bond network. Additionally, there are predominant blue ellipsoids and smaller green isosurfaces present between molecules in both salts, indicating that van der Waals forces contribute to the stacking of PRO salts alongside hydrogen bonds. The structural analysis of the δ_ginter contour set at 0.01 a.u. reveals that reducing this value results in the emergence of extensive green isosurfaces (Figure S8). This suggests an abundance of weak interactions, such as C-H⋯π and π⋯π, within the PRO salt structure, potentially leading to reduced crystal strength and a lower melting point compared to PRO alone. Furthermore, (PRO)⁻(4AMP)⁺ exclusively exhibits N₂⁺-H₂⋯O₁⁻ hydrogen bonds corresponding to blue-dominated ellipsoidal bodies, while, in (PRO)⁻(4AMP)⁺, there are a greater number of blue-dominated ellipsoidal bodies, indicative of a more robust network of N-H⋯O hydrogen bonds, resulting in enhanced thermal stability.

3.3.6. Lattice Energy (E_L) and Hydration-Free Energy (E_HF) Analysis

In general, enhanced solubility is typically associated with the increase in the Gibbs free energy of solvation. From a thermodynamic standpoint, the process of hydration can be delineated into sublimation, where individual components of the crystal are transported from an infinite distance into the gas phase and, subsequently, hydration occurs when molecules or ions are enveloped by interacting with water molecules. Consequently, the Gibbs free energy of solvation should equate to both the sublimation free energy and the hydration-free energy (E_HF). Lattice energy (E_L) denotes the absorbed energy when a crystal transitions to a gaseous state under standard conditions; thus, sublimation free energy can be approximated as negative E_L. By computing E_L and E_HF, we gain deeper insights into PRO’s solubility behavior after salt formation. As shown in

Table 6. Calculated values of E_L and E_HF.

Compound	EL (kcal/mol)	EHF (kcal/mol)
PRO	−64.87	−14.33
(PRO)−(4AMP)+	−41.84	−96.95
(PRO)−(4DAP)+	−50.62	−95.84

, salt formation disrupts PRO’s original crystal form dimer structure, leading to reduced E_L after the salt formation, which is a significant factor contributing to the decrease of melting points. Furthermore, there is consistency between E_L order and melting point: (PRO)⁻(4AMP)⁺ < (PRO)⁻(4DAP)⁺ < PRO.

Furthermore, the data indicate that the E_HF after salt formation of PRO is higher than that of its original crystalline form. However, due to the use of different methods for calculating E_L and E_HF in this study, the Gibbs free energy of solvation cannot be simply obtained by their summation. Nevertheless, salt formation results in a decrease in E_L and an increase in E_HF. This leads to a significantly lower Gibbs free energy for the salt solution compared to PRO. Additionally, (PRO)⁻(4AMP)⁺ exhibits lower E_L and higher E_HF than (PRO)⁻(4DAP)⁺, which contributes significantly to its enhanced solubility.

4. Conclusions

This study calculated the FIM of PRO and selected appropriate coformers based on supramolecular synthon rules, subsequently identifying (PRO)⁻(4AMP)⁺ and (PRO)⁻(4DAP)⁺ salts through liquid-assisted grinding (LAG). The salts underwent comprehensive characterization using PXRD, DSC, TGA, FTIR, SEM, etc., revealing that PRO displayed various PXRD patterns, melting points, thermal decomposition temperatures, IR absorption peaks, and crystal morphology after salt formation. SCXRD analysis demonstrated distinct hydrogen bond networks, molecular conformations, and packing modes in the two salts. In the phosphate-buffered solution (pH = 6.8), the solubility of the two salts increased to 142.83 and 7.75 times, respectively, compared to that of the raw drug; no crystalline transformation occurred during the 8 weeks of accelerated stability experiments at 40 °C with 75% RH. Molecular intermolecular interactions in the salts’ crystal structure were analyzed using HS; AIM and IGM were employed to analyze the salts’ hydrogen bond networks; MEPs and pK_a rules were utilized to investigate salt formation mechanisms and binding sites. Additionally, after salt formation, PRO exhibited decreased lattice energy but increased hydration-free energy, resulting in a lower melting point than that of PRO while demonstrating improved solubility in water. In summary, in this paper concerning the salt formation of soluble drugs, a comprehensive characterization and crystal structure analysis of the prepared PRO salt were conducted. Simultaneously, the principle underlying the changes in the melting point and solubility was deeply explored through quantitative analysis, providing novel insights for enhancing the solubility of drugs by salt formation through the selection of appropriate co-formers. Nevertheless, 4AMP and 4DAP possess certain toxicity, thus more consideration should be given to compounds that comply with GRAS safety certification when choosing cofomers.