Next Article in Journal
CD44 Targeting of Cisplatin-Loaded Hyaluronic Acid-Modified Mesoporous Silica Nanoparticles for Lung Adenocarcinoma: Synthesis, Characterization, In Vitro and In Vivo Evaluation
Previous Article in Journal
Polynuclear Silver(I)–Quinoxaline Complex: Comprehensive Structural Characterization, Antimycobacterial Properties and DNA/BSA Binding Study
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Pharmaceutics
Volume 18
Issue 2
10.3390/pharmaceutics18020170
pharmaceutics-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Pharmaceutical Binary and Ternary Complexes of Gemcitabine with Aluminum Metal–Organic Framework: Mechano-Chemical Encapsulation, Delayed Drug Release, and Toxicity to Pancreatic Cells

by
Kamala Panthi
1,
Sheriff Umar
1,
James Wachira
2 and
Alexander Samokhvalov
1,*
1
Department of Chemistry, Morgan State University, 1700 East Cold Spring Lane, Baltimore, MD 21251, USA
2
Department of Biology, Morgan State University, 1700 East Cold Spring Lane, Baltimore, MD 21251, USA
*
Author to whom correspondence should be addressed.
Pharmaceutics 2026, 18(2), 170; https://doi.org/10.3390/pharmaceutics18020170
Submission received: 4 December 2025 / Revised: 13 January 2026 / Accepted: 22 January 2026 / Published: 28 January 2026
(This article belongs to the Section Drug Delivery and Controlled Release)
Browse Figures
Versions Notes

Abstract

Background: gemcitabine is a cytidine analog and major anticancer drug functioning as an antimetabolite. However, its administration by systemic route is accompanied by “burst” and side effects. To limit this, drugs are encapsulated in matrices; metal–organic frameworks (MOFs) are coordination polymers with strong potential for drug encapsulation and delayed release. Methods: mechano-chemical synthesis of solid-state binary complex lag(CYCU-3)(Gem) is described from aluminum MOF (Al-MOF) CYCU-3 and gemcitabine free base (Gem). Synthesis is conducted by liquid-assisted grinding (LAG) with dimethyl sulfoxide (DMSO) followed by its outgassing. The alternative “dry” synthesis results in dry(CYCU-3)(Gem). Materials were characterized by FTIR spectroscopy and XRD, and delayed Gem release was tested to phosphate buffered saline (PBS) at 37 °C. The in vitro toxicity to pancreatic cancer PANC−1 and healthy cells hTERT−HPNE E6/E7/K−RasG12D was assessed by fluorometric assay. Results: in lag(CYCU-3)(Gem) interactions MOF-drug are via non-covalent bonds at O-H and COO− groups of CYCU-3 as found by FTIR marker peak shifts and crystal structure is retained, while dry(CYCU-3)(Gem) shows significant amorphization and loss of functional groups. The lag(CYCU-3)(Gem) but not dry(CYCU-3)(Gem) shows delayed Gem release for 6000 min. The suppression of PANC−1 cells by lag(CYCU-3)(Gem) is time-dependent and it correlates with delayed Gem release. For the first time, a concept of ternary stoichiometric complex lag(CYCU-3)1(Gem)1(CIT)2 is tested that also contains natural organic compound citronellol (CIT), and its structure, bonding and release of Gem are compared to those of binary complex. Bonding is at the O-H groups of CYCU-3 and this complex shows delayed Gem release. Conclusions: binary and ternary complexes of Gem with CYCU-3 yield delayed release and cytotoxicity. LAG is promising for synthesis of solid-state complexes of gemcitabine for delayed release and time-dependent suppression of cancer cells.

1. Introduction

Considerable progress has been made in reducing cancer incidence and mortality through modification of behavior, screening and improved therapies, but cancer treatment remains a major unmet medical need, with over 600,000 deaths annually and an incidence exceeding 2 million in the U.S. alone [1]. Globally, the occurrence of cancer is increasing rapidly, and it is predicted [2] that the rate will rise to 35 million cases in 2050 from 20 million in 2022. While cancer of the pancreas is the 10th most frequent cancer in the U.S., it is the third leading cause of mortality [3] with a survival rate of only 13%. The data show increases in the incidence of pancreatic cancer, amid stabilized declines in the number of cases of all cancers [1]. In advanced disease, its main treatments include chemotherapy with gemcitabine [4] alone and in combination with other drugs, e.g., taxanes and FOLFIRINOX [5].
Gemcitabine [4] is a major drug approved by U.S. Food and Drug Administration (FDA) as hydrochloride salt GemHCl (Gemzar) for systemic chemotherapy, including that of pancreatic cancer. Gemcitabine free base (Gem), Figure S1 is approved by FDA for injections and studied as active ingredient in chemotherapeutic agents, e.g., [6]. Despite the presence of tautomeric forms in crystalline Gem [7], see Figure S1, this compound and GemHCl are converted to the same molecular form in aqueous solution at physiological pH. Gemcitabine is phosphorylated intracellularly to gemcitabine diphosphate which inhibits nucleoside diphosphate reductase, and its triphosphate causes cytotoxicity by interfering with DNA replication [8]. Despite the clinical efficacy of gemcitabine, its adverse effects include hematological suppression, cardiotoxicity, and potentially mitochondrial toxicity [9,10].
The systemic administration of gemcitabine and other drugs has a disadvantage of the “burst” [11], a fast increase in drug concentration in tissues after administration, followed by metabolization and loss of efficiency. Also, the systemic administration requires large doses of drugs [12] that potentially increase toxicity, promote resistance to gemcitabine [13] and other drugs [14]. Studies for improving therapeutic value have included the encapsulation of gemcitabine in lipid nanocapsules [15] and liposomes [16] for drug delivery. Local delayed release also known as “local drug delivery”, e.g., see in ref. [17] minimizes the “burst” and provides sustained pharmaceutical action, e.g., via drug-eluting implants. Materials for pharmaceutical implants are usually of generic formula (matrix) (drug) where the “conventional” matrix can be biopolymer e.g., chitosan [18] or synthetic polymer [17]. In such matrices, drug molecules are weakly attached by physisorption which results in limited stability of composite materials and quick drug release.
Among advanced matrix materials, metal–organic frameworks (MOFs) stand out as hybrid organic–inorganic coordination polymers which gained wide recognition for encapsulation and release of various drugs, from anticancer agents [19] to antibiotics [20]. MOFs contain the repeating secondary building units (SBUs) that consist of metal cation and organic linker. MOFs offer an outstanding structural variability and very wide selection of linkers, which can be tuned for affinity to encapsulate drug molecules and capability to release them. For biomedical applications, MOFs are needed that have large nanopores for “guest” molecules and a synergistic combination of functional units. CYCU-3 is aluminum MOF (Al-MOF) with large ca. 3 nm pores, and its linker features the combination of polar COO− and H-O groups as well as non-polar aromatic phenyl and olefinic C=C groups, see Figure 1 (its full 3D structure was published elsewhere [21]).
For drug encapsulation, a common method is sorption from diluted solution in water [22] or organic solvent [23] on a micrometer size powder. However, most small-molecule organic anticancer drugs have very low solubility in any solvent, so encapsulation by sorption requires large volumes of solvents. As greener alternative, mechano-chemical encapsulation has been considered. The mechano-chemical processing results in activation [24] of powders [25] and their reactions; this is widely utilized in synthesis of drugs [26] and processing of pharmaceuticals [27]. The mechano-chemical reaction can be conducted between two dry powders or in the presence of grinding fluid [28] also known as liquid-assisted grinding (LAG).
The common pharmaceutical composites contain two ingredients (matrix and drug), however their chemical formulae, e.g., (matrix)x(drug)y are usually not reported. Beyond this, multi-component materials were reported which also contain additives; e.g., the folate conjugated dual drug formulations in [29] contain poly(lactic-co-glycolic acid) matrix, gemcitabine drug, curcumin and folate. Similarly, for the multi-component materials, chemical formulae such as (matrix)x(drug)y(additive)z are seldom reported. Terpenes are natural small-molecule organic compounds with a variety of pharmaceutical and biological applications. Citronellol (CIT) is a terpene alcohol with pharmacological properties [30] and it was studied as an anticancer [31] agent.
For the multi-component materials, knowledge of bonding between their components is essential. Molecular spectroscopy is a common approach to determine mechanism of bonding. In particular, attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectroscopy is a powerful method of characterization of pharmaceuticals [32] including intact and ground pharmaceutical drugs [33]. Delayed drug release is commonly studied in model physiological buffer solution [34] such as PBS at 37 °C with or without mechanical stirring. PANC−1 is pancreatic ductal epithelial cell line commonly used for the in vitro studies of cytotoxicity, e.g., [35]. We report the following:
  • the mechano-chemical synthesis of the binary complex lag(CYCU-3)(Gem) from CYCU-3 (Figure 1) and Gem by LAG, followed by outgassing a grinding fluid;
  • spectroscopic and structural analysis of lag(CYCU-3)(Gem);
  • mechano-chemical “dry” synthesis of related material dry(CYCU-3)(Gem), its analysis and comparison of composition and structure with lag(CYCU-3)(Gem);
  • delayed release of gemcitabine from lag(CYCU-3)(Gem) and dry(CYCU-3)(Gem) to PBS at 37 °C and its timescale;
  • toxicity of lag(CYCU-3)(Gem) to pancreatic cancer cells PANC−1;
  • a concept and synthesis of the ternary complex lag(CYCU-3)1(Gem)1(CIT)2 using LAG with CIT as grinding fluid, its instrumental analysis and delayed release of gemcitabine.

2. Materials and Methods

2.1. Chemicals

For synthesis of CYCU-3, precursors were 4,4′-stilbenedicarboxylic acid H2SDC of a 95% purity (from Ambeed Inc., Buffalo Grove, IL, USA, product A705140) and anhydrous AlCl3 of a 99% purity (from Beantown Chemical Corporation, Hudson, NY, USA, product 218275-250G). Glacial acetic acid CH3COOH was from J.T. Baker (now part of Avantor), Phillipsburg, USA (product 9508-33). Solvents were N,N-dimethylformamide (DMF) of a ≥99.5% purity (from TCI America, Portland, OR, USA, product D0722), acetone of a 99.5% purity (from Thermo Fisher Scientific, Waltham, MA, USA, product 030698.K2), and anhydrous ethanol (from Thermo Fisher Scientific, product A405F-1GAL). Gemcitabine free base (Gem) of a 98.0% purity (from TCI America, Portland, OR, USA, product G0544) was stored in a freezer at −80 °C after receiving. In the LAG, one grinding fluid was dimethyl sulfoxide (DMSO) of a 99.9% purity (from Thermo Fisher Scientific, Alfa Aesar brand, product 42780) and the other was citronellol (CIT) of a >95.0% purity by GC (from TCI America, product C0370). Unless otherwise noted, all chemicals were purchased from commercial sources and used as received.

2.2. Synthesis and Activation of CYCU-3

The as-synthesized MOF (asCYCU-3) was prepared according to ref. [21]. Namely, samples of H2SDC at 107.3 mg (0.4 mmol), AlCl3 at 53.33 mg (0.4 mmol) and glacial acetic acid CH3COOH at 120 mg (2 mmol) were mixed in a 10 mL DMF and heated at 180 °C in an autoclave for 3 days. The autoclave was allowed to cool naturally, the obtained yellow powder was filtered on a fiber glass filter with ca. 1.2 μm pores (from Omicron Scientific, Inc., Marietta, GA, USA, product 133047) and washed with 20 mL DMF. The obtained asCYCU-3 was activated, to remove volatile impurities, at 90 °C for 12 h in a vacuum oven (from Across International, Livingston, MT, USA, product AT09e.110) that was connected to a two-stage vacuum oil pump (from Xtractor Depot, Montebello, CA, USA, pumping speed 12 cfm, base pressure ca. 15 mTorr). This resulted in the activated form denoted below CYCU-3.

2.3. Instrumental Analysis

ATR-FTIR spectra were collected by infrared spectrometer (from Thermo Fisher Scientific, Waltham, MA, USA, model Nicolet iS20) equipped with an ATR attachment (from Thermo Fisher Scientific, model Smart iTX). Data acquisition software OMNIC Specta Nicolet iS50 Edition was used, the optical aperture was set “Open” and variable gain was used. Each spectrum was averaged 512 times, optical slit was 4 cm−1 and an increment of the wavenumber was 0.5 cm−1. To minimize effects of humidity in ambient air, the interior of the FTIR spectrometer was continuously purged with dried air at a flow rate of 30 scfh (standard cubic feet per hour) using a flowmeter (from Dwyer Instruments LLC, Michigan City, IN, USA, product RMA-7). Dried air was obtained using an FT-IR Purge Gas Generator (from Parker Hannifin Corporation, Cleveland, OH, USA, product 74-5041) which yields the water vapor content (per specifications) equivalent to a dewpoint of −73 °C (relative humidity RH < 1%). To monitor the quality of spectra continuously and automatically remove artifacts due to trace amounts of water vapor, the parameter “Atmospheric Correction” was enabled, and parameter “Spectral Quality Results” was set at “H2O level” ≥ 95%. ATR-FTIR spectra were presented in absorbance mode.
Patterns of powder X-ray diffraction (XRD) were obtained by diffractometer MiniFlex (from Rigaku Corporation, Tokyo, Japan) that was equipped with nickel foil to remove a K-beta artifact. In all experiments, a Cu K-alpha line at 0.15418 nm was used and increments of the 2θ angle were 0.02 deg. Numeric fitting of ATR-FTIR and XRD peaks and removal of baseline from XRD patterns were conducted using Microcal Origin 2016 program. SEM images were collected using JCM-7000 NeoScope benchtop instrument from JEOL USA Inc., Peabody, MA, USA.

2.4. Mechano-Chemical Synthesis of Binary Complexes of CYCU-3 with Gem

The first complex denoted lag(CYCU-3)(Gem) was prepared by LAG as below, using mechanical ball mill (from Retsch GmbH, Haan, Germany, model Retsch Qiagen TissueLyser) equipped with two stainless steel grinding vials of 5 mL in volume.
Prior to LAG, a mechanical pre-mixing of dry powders was conducted (for better uniformity). CYCU-3 has Hill formula C16H11O5Al1 based on its SBU, and formal molar mass 310 mg/mmol, while gemcitabine free base Gem has molar mass 263 mg/mmol. In a glove bag filled with dried air, a 310 mg (1 mmol) CYCU-3 and 263 mg (1 mmol) Gem were added to a pre-weighted stainless steel grinding vial. This specimen had formula (CYCU-3)1(Gem)1 and, based on the equimolar amounts of compounds, corresponded to formal molar mass 310 + 263 = 573 mg/mmol. The grinding vessel was closed with a stainless-steel screw cap, attached to a mechanical ball mill, and automatic shaking was performed at 30 Hz frequency for 30 min.
After pre-mixing, the grinding vessel was opened in a glove bag in dried air, and a 0.125 mL DMSO (LAG fluid) was added. Then, one stainless steel grinding ball (7 mm diameter) was added, the grinding vessel was closed, and an assembly was ground at 30 Hz. The grinding was conducted by repeating the sequence “grind-idle” (5 min. “grind” followed by 5 min. “idle”). This was used to minimize potential over-heating of specimens, and this sequence was repeated until the total “grind” time of 1 h. The grinding vessel with obtained wet mixture was opened, placed into the same vacuum oven, in which activation of MOF was conducted, and outgassed at room temperature to completely remove DMSO. The first criterion of removal of LAG fluid was achieving a constant base pressure of the vacuum oven at ca. 40 mTorr. The second criterion was achieving a constant mass (grinding vial + grinding ball + MOF + Gem) as measured immediately after venting the vacuum oven with dried air. The obtained binary complex lag(CYCU-3)(Gem) was obtained as white powder and stored in a closed specimen jar. Reactions of its synthesis are as follows (the number before each substance is its amount in mmol):
0.5 CYCU-3 (s) + 0.5 Gem (s) + 1.8 DMSO (l) → GR→ [0.5 CYCU-3 (s) + 0.5 Gem (s) + 1.8 DMSO (l)]
where GR is grinding, followed by full outgassing (OUTG) of DMSO based on observed loss of mass:
[0.5 CYCU-3 (s) + 0.5 Gem (s) + 1.8 DMSO (l)] → OUTG → 0.5 lag(CYCU-3)(Gem) (s) + 1.8 DMSO (v)
The other binary complex of CYCU-3 with Gem denoted dry(CYCU-3)(Gem) was a white powder and it was prepared by similar mechano-chemical procedure, except that no DMSO was used.

2.5. Mechano-Chemical Synthesis of Ternary Complex of CYCU-3 with Gem and CIT

The ternary complex was prepared similarly to the binary complex lag(CYCU-3)(Gem) using LAG except the following important details. First, the amounts of added chemicals were CYCU at 155 mg (0.5 mmol), Gem at 132 mg (0.5 mmol) and CIT at 239 mg (1.53 mmol). Second, the CIT was outgassed in a vacuum oven until a constant mass, but the cumulative mass (grinding vial + grinding ball + MOF + Gem) was not achieved. Instead, the obtained constant mass was larger, indicating the loss of 83 mg CIT and presence of CIT in the complex, which corresponded to formula lag(CYCU-3)1(Gem)1(CIT)2. Its synthesis is as follows:
0.5 CYCU-3 (s) + 0.5 Gem (s) + 1.53 CIT (l) → GR → Mixture
where Mixture is the obtained wet mixture of compounds. In the outgassing of excessive CIT, the following reaction occurs:
Mixture → OUTG → 0.5 lag(CYCU-3)1(Gem)1(CIT)2 (s) + 0.53 CIT (v)
The obtained ternary complex (white powder) was stored in a closed glass specimen jar. Collection of FTIR spectra of freshly prepared samples and those after storage for a few months did not reveal differences, and compounds were stable at room temperature and in dry environment.

2.6. Delayed Release of Gemcitabine to PBS

The automated drug dissolution system (ADDS) automatically collects, at pre-programmed time intervals [36], small aliquots from the medium (pharmaceutical material + PBS) and stores them in an autosampler for an offline quantitative analysis. Tests were performed in an ADDS model VK 7000 that is equipped with heater/circulator VK 750D, peristaltic pump VK-810, and automatic dissolution sampling station VK 8000 (all from Vankel Technology Group Inc., Cary, NC, USA). In drug release experiments, the typical specimen contained 33 mg Gem. The dissolution medium consisted of 750 mL of 1X PBS without calcium and magnesium, prepared by diluting a PBS powder (from Albert Bioscience Inc., Laguna Hills, CA, USA) in DI water and adjusting pH to 7.4. A paddle method was employed [37] at a stirring speed of 60 revolutions per minute (rpm). The medium was maintained at a constant temperature of 37 ± 0.5 °C using a one-liter glass dissolution vessel in the VK 7000 instrument and the integrated water thermostat bath with heater/circulator VK 750D. The sampling cannulas of the VK 7000 apparatus were furnished with 10 μm porous ultra-high molecular weight polyethylene filters (from Quality Lab Accessories LLC, Telford, PA, USA, product FIL010-01-a). Namely, stirring of PBS in the vessel was started, and the powder of specimen was gently added to it. Then, 2 mL aliquots were withdrawn automatically at pre-programmed intervals and collected in the VK 8000 sampling station. The collected specimens were frozen at −80 °C until an offline analysis of gemcitabine by an HPLC-UV method. Experiments were conducted under sink conditions, and molar concentration of gemcitabine in PBS was at least three times lower than its solubility.

2.7. Chromatographic Analysis of Gemcitabine in Release Media

First, frozen specimens collected by the ADDS were thawed, and filtered by a PTFE syringe filters that have Luer lock connectors 4 mm in diameter and 0.22 μm pore size (from Tisch Scientific, Cleves, OH, USA, product SF17504), using disposable 1 mL Luer lock tip syringes (from BH Supplies, Berwick, PA, USA).
An HPLC-UV analysis was conducted by instrument of series 1100 (from Agilent Technologies Inc., Santa Clara, CA, USA) and the software was ChemStation for LC 3D systems of version B.04.02. This instrument was equipped with guard cartridge ZORBAX Eclipse XDB-C18 (4.6 × 12.5 mm, 5 µm) and column ZORBAX Eclipse XDB-C18 (4.6 × 150 mm, 5 µm), both from Agilent. An isocratic HPLC mode was used, and mobile phase was prepared per [38] as a mixture (40 mM aqueous ammonium acetate/acetonitrile) at vol./vol. ratio 97.5/2.5. Injection volume was 1 µL, flow rate was 0.5 mL/min, and the detection wavelength was 254 nm. The HPLC-UV calibration curve was obtained from standard solutions of Gem in PBS upon numeric integration of its HPLC peaks.

2.8. Cell Viability Assay of PANC−1

PANC−1 human pancreatic cancer cells were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA, product CRL-1469) and cryopreserved in liquid nitrogen vapor phase at the Morgan State University Core Laboratory. The cells were cultured following recommendations of the supplier. Briefly, the cells were thawed rapidly and diluted in culture medium consisting of Dulbecco’s Modified Eagle Medium DMEM (product 30-2002) supplemented with 10% fetal bovine serum FBS (product 30-2021) and 1% penicillin-streptomycin antibiotic solution (product 30-2300), all from ATCC. The cells were harvested by centrifugation at 400× g for 10 min and the supernatant was discarded. The pellet was resuspended in fresh culture medium and transferred to a humified incubator held at 37 °C and containing 5% CO2. Passaging was conducted on pre-confluent cells by washing them with serum-free DMEM and then adding trypsin-EDTA. Detachment was monitored microscopically, and the cells diluted with fresh culture medium were counted and replated at a dilution of 1:4 based on the original volume. A single batch of cells was used for all experiments within a three-month period, to minimize variability.
Cell proliferation was assessed by the Alamar Blue (AB) assay in fluorescence mode. PANC−1 cells were seeded in 96-well plates at a density of 1 × 104 cells/well and incubated for 24 h. CYCU-3 and its complex are not soluble in PBS, so the cytotoxicity test was designed as modification of procedure for water-insoluble Yb3+ MOF [39]. The modification involved suspending the chemical agent in a calculated volume of DMSO (from MP Biomedicals LLC, Irvine, CA, USA, product 194819) per [40] to achieve desired concentration (in μM). Then the sample was sonicated (sonicator model VGT−1620QTD from GT SONIC, Shenzhen City, China) for 30 s to 1 min. to achieve fine suspension, as assessed by visual clarity and the lack of immediate precipitation, and the obtained suspension was immediately added to the PANC−1 cell cultures in the 96-well plates. The final concentration of DMSO in the cell culture medium for all treatment and vehicle control (DMSO) wells was maintained at 0.1% (v/v). The cells were incubated at 37 °C in a CO2 incubator for 48 and 72 h. After each specified period, the culture media were replaced with 10% (v/v) AB solution (from G-Biosciences, St. Louis, MO, USA, product 786-922) prepared in complete growth media, and subsequently incubated for another 2 h. Finally, fluorescence readings were taken using a BioTek Synergy H1 microplate reader (from BioTek Instruments, now part of Agilent Technologies, Bozeman, MT, USA) at 560/590 nm excitation/emission wavelengths. All cytotoxicity tests were conducted in sextuplicate (n = 6). Cell viability was then expressed as a percentage of fluorescence signal measured in the vehicle-treated control wells. GraphPad Prism version 8.4.3 software was used in statistical analysis. To assess differences in cytotoxicity, cell viability data from the different treatment groups were analyzed with one-way ANOVA followed by post hoc Tukey’s test with the statistical significance cutoff set at p < 0.05.

2.9. Cell Viability Assay of hTERT−HPNE E6/E7/K−RasG12D

The hTERT−HPNE E6/E7/K−RasG12D, also known as hTERT human pancreatic cells, were obtained from ATCC (CRL-4038) and cryopreserved in liquid nitrogen vapor at the Translational Core Facility of the University of Maryland Marlene and Stewart Greenebaum Cancer Center. Cell culture media was made up first of a base medium, which was a mixture of 75% DMEM without glucose (from Sigma-Aldrich, now Merck, Saint Louis, MO, USA, product D-5030) supplemented with 2 mM L-glutamine and 1.5 g/L sodium bicarbonate and 25% Medium M3 Base (from INCELL Corporation LLC, San Antonio, TX, USA, product M300F-500). To make the complete growth medium, fetal bovine serum at 5% (from ATCC, product 30-2021), 10 ng/mL human EGF recombinant protein (from Thermo Fisher Scientific, Waltham, MA, USA, of Gibco brand, product PHG0314), 5.5 mM D-glucose (from Thermo Fisher Scientific, product MP21946721) and 750 ng/mL puromycin dihydrochloride (from Thermo Fisher Scientific, product AAJ67236XF) were added to the base medium. The hTERT cells were seeded in 96-well plates at a density of 2500 cells/well and incubated for 24 h. before treatment. The addition of chemicals and colorimetric assay were performed as for PANC−1 cells.

3. Results and Discussion

3.1. The IR Spectra and XRD Patterns of CYCU-3 and Gemcitabine Free Base (Gem)

Figure 2 shows IR spectra of thermally activated CYCU-3 as drug encapsulation matrix in this work; they are consistent with those in [21], however the IR peaks of CYCU-3 were not assigned to specific functional groups in [21]. Also, peaks in Figure 2 are consistent with published IR spectra of related aluminum MOFs with carboxylate linkers. In Figure 2a, the peak at 3710 cm−1 is due to the O-H stretch of the bridging H-O(-Al) group aka the μ-OH group. The benchmark Al-MOF MIL-53(Al) has an IR peak at ca. 3710 cm−1 due to the same group [41]. The corresponding deformation peak is in Figure 2c at ca. 983 cm−1 consistently with IR peaks reported for many Al-MOFs [41,42,43]. In Figure 2, major IR peaks (sharp ones of high intensity) are marked with frames, and they can serve as spectroscopic “markers” to learn about interactions of CYCU-3 with Gem in the complex.
The IR peaks in Figure 2 can be assigned based on spectra of the structurally similar Al-MOFs [41,44,45] and related small molecule compounds [46], Table S1. Figure S2 shows powder XRD pattern of CYCU-3 and it is consistent with references [21,47]. In Figure S2, the strongest and narrow characteristic peak at 2θ = 6.1 deg. was fitted with Gaussian function, and the Scherrer analysis was conducted: D = k λ/β cos(θ) where D is nanocrystal size, k is the shape factor of 0.9, λ is an X-ray wavelength, β is the full width at the half-maximum (FWHM) of diffraction peak (in radian), and θ is the Bragg angle. This results in the average nanocrystal size of CYCU-3 at ca. 25 nm. Figure S3 shows SEM of CYCU-3 with characteristic needle-shaped crystals at ca. 3 μm in length and ca. 250 nm in cross section. A high form-factor of CYCU-3 crystals is consistent with its anisotropic orthorhombic lattice [21] with parameters a = 33.776, b = 59.003 and c = 6.742 Å that differ significantly from each other.
Now, it is of interest to explain IR spectra of Gem drug and detect its spectral markers that do not overlap with marker peaks of CYCU-3 matrix. Figure S4 shows ATR-FTIR spectrum of Gem and it is consistent with its spectrum reported previously [7]. Judged by the strong peak at ca. 3400 cm−1 the imino-keto tautomer of Gem [7] is detected; see assignments of Gem vibrations in [48,49,50] and in Table S2. The marker peaks of CYCU-3 and Gem are in solid frames in Figure 2 and Figure S4, respectively, and they can be used to study interactions of CYCU-3 and Gem in complex lag(CYCU-3)(Gem).
Figure S5 shows the XRD pattern of Gem; it is consistent with the literature data [51] of Gem crystals of the orthorhombic lattice and the Pmna space group. The notable reflections are of Miller index (002) at ca. 9.6 deg., the (301) at 15.9 deg., the (210) at 16.3 deg., the (303) at 20.9 deg., and the (122) at 27.9 deg. The numeric peak fitting of strong sharp peak at 2θ = 27.9 deg. and Scherrer’s analysis yields an average nanocrystal of ca. 27 nm. Strong XRD peaks of Gem will be used in analysis of its reaction with CYCU-3, see in the next section.

3.2. The Binary Complexes Prepared by Liquid-Assisted Grinding lag(CYCU-3)(Gem) and by Dry Grinding dry(CYCU-3)(Gem)

In complex lag(CYCU-3)(Gem), the interactions between CYCU-3 matrix and Gem drug are analyzed by comparison of FTIR spectra of each component to the spectra of another component, and using their “marker” peaks. The shift in an IR peak of the given functional group of that component in the complex vs. pure compound indicates binding at that group.
Figure 3 shows ATR-FTIR spectra of lag(CYCU-3)(Gem) vs. CYCU-3 that are scaled for comparison as in Figure 2 (CYCU-3) and Figure S4 (Gem). First, in spectra of lag(CYCU-3)(Gem) there are no additional peaks due to DMSO, consistently with its removal based on gravimetric analysis. Importantly, the presence of all major IR peaks of CYCU-3 in the complex indicates preservation of this MOF after LAG followed by DMSO outgassing. This finding is consistent with reported [21] good thermal stability of CYCU-3 at 150 °C in a vacuum. Further, in Figure 3a the peak due to the stretch vibration of the H-O(-Al) group is broadened and red shifted, from 3710 cm−1 for pure CYCU-3 to 3700 cm−1 for CYCU-3 in the complex. Strong broadening of this stretching peak is consistent with loss of the deformation peak of the same H-O(-Al) group in Figure 3c at 984 cm−1.
Hence the H-O(-Al) group in CYCU-3 interacts with Gem molecules. The related ca. <20 cm−1 red shift of an IR peak of the H-O(-Al) group in the structurally similar Al-MOF MIL-53(Al) upon sorption of carbon monoxide CO is due to its interaction with the H-O site as weak Brønsted acid [52]. Carbon monoxide has properties of weak Brønsted base [53]. Gem is a weak base due to its NH2 group, so in the complex lag(CYCU-3)(Gem) it interacts by this group (weak Brønsted base) with weak Brønsted acid site of the H-O(-Al) group in CYCU-3. Also, in Figure 3b the peak at 1560 cm−1 due to the asymmetric vibration νasym(COO−) of CYCU-3 is red shifted to 1555 cm−1 (shown by an arrow) indicating that the COO− group interact with Gem. This is consistent with red shift of peak due to the symmetric vibration νsym(COO−) of the same group from 1435 cm−1 to 1426 cm−1. In Figure 3c one can see the change in the Al-O peaks of CYCU-3 at ca. 550 cm−1 for the complex vs. pure MOF, suggesting interaction at the Al-O bond. However, low wavenumbers range is not among the “fingerprint” part of the IR spectra (the latter is usually in mid-IR) and pure Gem has group vibrations in this range [7] for both tautomers.
In Figure S6 one can see the zoomed marker peaks of CYCU-3 for pure CYCU-3 and CYCU-3 in the complex. Figure S7 shows ATR-FTIR spectra of lag(CYCU-3)(Gem) vs. Gem with its marker peaks in the frames. In spectrum of Gem in the complex, all Gem peaks are present, so it is not decomposed during LAG or the subsequent outgassing of DMSO. This is consistent with TGA and DSC studies [54] showing that Gem has good thermal stability and decomposes only above ca. 235 °C.
In Figure S7, changes in the marker peaks of Gem show a moderate degree of its tautomerization in the complex, from the imino-keto to the amino-keto form, consistently with tautomerization of pure Gem under LAG [7]. This observation alone is not indicative of bonding Gem to CYCU-3 in the complex; however, when combined with spectral shifts in IR peaks of CYCU-3 in the complex, this indicates bonding of the drug to the matrix.
Powder XRD pattern of complex lag(CYCU-3)(Gem) is in Figure S8. For its CYCU-3 component one can see the significant modifications of XRD peaks vs. intact CYCU-3 (in Figure S2). The characteristic XRD peak at 6.1 deg (in solid frame) is retained and it has about the same width as for intact CYCU-3 (Figure S2). This indicates the lack of amorphization of CYCU-3 under LAG with Gem. This also indicates that the complex lag(CYCU-3)(Gem) contains some free CYCU-3 not bonded to Gem. On the other hand, other peaks of CYCU-3 undergo changes; a new peak at 5.3 deg appears (in solid frame) that can be assigned to the bonded form of CYCU-3. Additionally, a peak at ca. 3 deg is lost (in dashed frame) confirming the interaction of CYCU-3 with Gem. For Gem component of lag(CYCU-3)(Gem), the characteristic XRD peak at 27.9 deg is lost (in dashed frame) indicating some amorphization of Gem or its interaction with CYCU-3. Importantly, a new peak at 10.5 deg is present that is absent for pure CYCU-3 or Gem, indicating formation of their binary complex. SEM of lag(CYCU-3)(Gem) in Figure S9 shows ca. 250 nm nanocrystals fused into microcrystalline agglomerates, and the absence of typical needles of CYCU-3, consistently with its grinding and reaction during LAG.
Bonding in the complex lag(CYCU-3)(Gem) in Figure 4 is consistent with structure of CYCU-3 and molecular size of Gem; red dashed lines indicate noncovalent bonds between CYCU-3 matrix and Gem drug.
Namely, hexagonal channels in CYCU-3 have cross-sectional sizes [21] of ca. 28.3 × 31.1 Å, while the largest dimension in the Gem molecule is 8.9 Å (per estimate by Chem3D Pro version 12.0 program) so the Gem molecule can fit the nanopore of CYCU-3 in the complex.
As control, CYCU-3 was ground with Gem without any liquid, but otherwise at the same conditions as for lag(CYCU-3)(Gem). The XRD pattern of obtained material dry(CYCU-3)(Gem) is in Figure 5.
For CYCU-3 component of dry(CYCU-3)(Gem) its characteristic XRD peak at ca. 5.5 deg. (in black frame) is strongly broadened vs. pure CYCU-3, and the Scherrer analysis gives just a 6.5 nm particle size. This indicates a strong amorphization of CYCU-3 upon dry grinding with Gem. For Gem in dry(CYCU-3)(Gem), its XRD peak at 27.9 deg (in violet frame) remains, indicating an incomplete bonding to CYCU-3. Its Scherrer analysis gives 20 nm particle size, indicating moderate changes after dry grinding with CYCU-3. Hence, amorphization of CYCU-3, rather than its bonding to Gem, is observed upon dry grinding. The observed data of dry(CYCU-3)(Gem) are in stark contrast to lag(CYCU-3)(Gem), which has a bond between Gem to CYCU-3. The relative degree of interaction of CYCU-3 with Gem in each material is assessed below by the in vitro release of Gem to PBS.

3.3. The In Vitro Delayed Release of Gemcitabine from Binary Complexes

In the quantitative determination of gemcitabine released to PBS, the HPLC-UV analysis was performed based on Gem calibration plot, Figure S10. Figure 6 shows the temporal trace of delayed release of gemcitabine from lag(CYCU-3)(Gem) to PBS at 37 °C under stirring in the ADDS. In Figure 6a, after a quick initial increase in concentration [Gem] within ca. 60 min., delayed release occurs within ca. 60–6000 min., and the concentration of Gem steadily increases (a “slow component”). Figure 6b shows a detailed view of the initial stage, with quick dissolution of non-bonded Gem as follows:
Gem (s) → 1 Gem (aq)
where (s) is solid phase and (aq) is solution in PBS. Figure 6b also shows numeric fitting (solid red line) of this dissolution curve using kinetic formula of the pseudo-first rate law [55]. Here, the concentration of product is modeled as y(t) = A + B × (1 − exp(−k × t)) where the k is rate constant, B is a proportionality coefficient, and A is a numeric offset. The dissolution profile is fitted at a very good parameter R2adj = 0.99 and the k = 1.6 ± 0.1 min−1. Next, the “slow” component indicates that lag(CYCU-3)(Gem) contains bonded Gem, consistently with FTIR data, and this Gem undergoes delayed release:
lag(CYCU-3)(Gem) (s) → 1 Gem (aq) + CYCU-3 (s)
From the applications perspective, the range of molar concentration of released gemcitabine in Figure 6a is close to highest concentration at ca. 100 μM reported [35] in tests of toxicity to pancreatic cancer PANC−1 cells. Further, dynamics of drug release in Figure 6 is observed under the in vitro conditions with mechanical stirring, i.e., under the significantly accelerated mass transfer. In contrast, the in vivo experiments do not have forced mass transfer, so a much longer release time scale is anticipated. The presented data are of potential interest to prepare drug-eluting implants.
After delayed release of Gem, the PBS medium contains Gem and the anion of 4,4′-stilbenedicarboxylic acid (by HPLC analysis), i.e., anion of linker of CYCU-3; this indicates partial hydrolysis of CYCU-3. The concentration of linker released from the complex is in Figure S11a and it can be compared with the concentration of released Gem in Figure S11b. The time correlation in Figure S11a,b indicates that Gem is bonded to linker in the complex, consistently with FTIR spectral data.
Figure S12 shows temporal release of gemcitabine from complex dry(CYCU-3)(Gem). No delayed release is observed and there is a moderate decrease in drug concentration at the longer time. This is consistent with partial amorphization of CYCU-3 matrix in this complex per XRD data, and an apparent lack of interaction between Gem and CYCU-3. In summary, for the preparation of solid-state materials for delayed drug release, LAG is preferred to dry grinding.

3.4. The In Vitro Toxicity of lag(CYCU-3)(Gem) to Pancreatic Cells

Pancreatic cancer cells PANC−1 were cultured in the presence of increasing concentrations of lag(CYCU-3)(Gem) to assess its cytotoxicity (Figure 7 and Figure S13). In Figure 7, the mean values ± standard errors of the mean (SEM) of four or six replicates were plotted; a moderate suppression of cell viability is found at increasing concentration and a plateau within 1.2–30 μM. This plateau is likely due to delayed release of Gem from lag(CYCU-3)(Gem) and/or the effect of CYCU-3 matrix.
The ANOVA treatment of these data is in Figure S13, and it shows significant cytotoxicity; differences with p-values of ≤ 0.05 are shown with ****. The cells were also treated with CYCU-3 in its molar amounts equivalent to those in the complex lag(CYCU-3)(Gem), see Figure 7 and Figure S13, and CYCU-3 is not independently cytotoxic to PANC−1 cells (Figure S14). For comparison, the viability data of healthy pancreatic cells hTERT−HPNE E6/E7/K−RasG12D (also known as hTERT) with CYCU-3 are in Figure S15 and CYCU-3 has no toxicity. This MOF can be regarded as a promising drug encapsulation matrix which is not toxic to pancreatic cells.

3.5. The Ternary Complex lag(CYCU-3)1(Gem)1(CIT)2

The ternary complex lag(CYCU-3)1(Gem)1(CIT)2 was prepared from CYCU-3 and Gem by LAG with citronellol (CIT), Figure S16, as grinding fluid. To our knowledge, this is the first report of encapsulation of gemcitabine into the ternary complex using mechano-chemistry, and the first stoichiometric ternary complex of gemcitabine.
Figure S17 shows FTIR spectrum of CIT which is scaled for direct comparison with IR spectra of CYCU-3 (Figure 2) and Gem (Figure S4). The spectrum in Figure S17 is consistent with previously published data [56,57], and Table S3 has major IR peaks of CIT with their assignments. In Figure S17, the two strong sharp peaks at 1376 cm−1 and 1057 cm−1 due to the deformation C-O-H vibration and the ν(C-O) stretch, respectively, are characteristic to CIT (its markers) since they are not present in the IR spectra of CYCU-3 or Gem. The FTIR marker peaks of CIT are used below in the identification of mode of its bonding in the complex. Specifically, analysis of bonding in lag(CYCU-3)1(Gem)1(CIT)2 is conducted by a pair-wise comparison of its FTIR spectra with spectra of each component: CYCU-3, Gem and CIT.
The first step is to monitor changes in spectra of CYCU-3 component of the ternary complex. Figure 8 shows FTIR spectrum of lag(CYCU-3)1(Gem)1(CIT)2 versus CYCU-3 with its markers.
In the spectrum of lag(CYCU-3)1(Gem)1(CIT)2 all marker peaks of CYCU-3 are retained, so the matrix is not decomposed. In Figure 8a the peak of the stretch vibration of the H-O(-Al) group is red shifted (shown by an arrow) from 3710 cm−1 for pure CYCU-3 to 3695 cm−1 for CYCU-3 in the complex (by 15 cm−1) and it retains its shape. The FTIR spectra of lag(CYCU-3)1(Gem)1(CIT)2 versus CYCU-3 are consistent with reported [58] FTIR peaks of MIL-53(Al) before and after sorption of ethanediol which, similarly to CIT, has the O-H groups. This indicates that in lag(CYCU-3)1(Gem)1(CIT)2 the molecules of CIT are bonded by their O-H groups at the H-O(-Al) groups of CYCU-3. This is consistent with partial loss of the deformation H-O(-Al) peak of CYCU-3 in the complex in Figure 8c. On the other hand, in Figure 8b there are no shifts in either asymmetric or symmetric stretch vibrations of the COO− group of CYCU-3. The observed lack of bonding CIT at the COO− group is consistent with absence of peak shifts in carboxylate group in MIL-53(Al) after sorption of isopropanol, methanol and ethanol as found by FT-Raman spectra [59]. This indicates bonding of CIT and Gem molecules in lag(CYCU-3)1(Gem)1(CIT)2 only at the O-H group of CYCU-3. This is in contrast to the binary complex lag(CYCU-3)(Gem) in which Gem molecules are bonded to CYCU-3 via its O-H and COO− groups. Hence, Gem molecules are bonded more weakly to CYCU-3 in the ternary complex than in the binary complex. The second step is to analyze how the IR markers of Gem in the ternary complex change vs. those for pure Gem, Figure S18. For lag(CYCU-3)1(Gem)1(CIT)2 all marker peaks of Gem are retained so it is not decomposed. Figure 9 shows FTIR spectrum of lag(CYCU-3)1(Gem)1(CIT)2 versus CIT with its marker peaks in frames. In Figure 9a, the complex has a strong multiplet within 3000–2800 cm−1 (dashed frame) due to several C-H vibrations of CIT molecules (see Table S3). This is consistent with data by gravimetric analysis that shows two CIT molecules per SBU of CYCU-3 in this complex. In Figure 9a the sharp peak at ca. 3400 cm−1 is due to the individual O-H groups of CIT in the complex, and it is very different from a wide band at ca. 3450–3150 cm−1 for the interacting O-H groups in pure liquid CIT. Hence, in the ternary complex CIT molecules interact by their (C)-O-H groups with the μ-OH groups of CYCU-3 and possibly with Gem. Further, in Figure 9b one can see the marker peak of CIT (in solid frame); for pure CIT the (C)-O-H peak is at 1377 cm−1 while for the complex it is blue shifted to ca. 1381 cm−1. Again, CIT interacts by its (C)-O-H groups with the O-H groups of CYCU-3. To our knowledge, this is the first report of the stoichiometric complex in which the modality of Gem bonding is affected by an extra component.
The XRD pattern of ternary complex lag(CYCU-3)1(Gem)1(CIT)2 is in Figure S19. Similarly to XRD of lag(CYCU-3)(Gem), peaks are narrow so there is no amorphization (destruction) of CYCU-3 in the ternary complex. Minor modifications of XRD pattern of CYCU-3 after LAG vs. pure CYCU-3 (Figure S2) include relative intensity of peaks at ca. 2.9 and 6.0 deg that indicates interaction of CYCU-3 with Gem or CIT. The largest dimension in CIT molecule is 8.8 Å per estimate by ChemDraw 3D program, and this is very close to the largest molecular dimension in Gem molecule at ca. 8.9 Å. Hence both Gem and CIT molecules can fit the nanopores of CYCU-3. Finally, there is some not-bonded Gem (marked “Gem”) based on its XRD peaks. SEM images of lag(CYCU-3)1(Gem)1(CIT)2 are similar to those of lag(CYCU-3)(Gem).
In Figure 10, lag(CYCU-3)1(Gem)1(CIT)2 shows delayed release of gemcitabine up to ca. 2500 min.
The full timescale of release is shorter than that of the binary complex lag(CYCU-3)(Gem). At the timescale of hundreds to thousands of min. (Figure 10a), an overall faster release of Gem from the ternary complex versus the binary complex is consistent with weakened bonding of Gem to CYCU-3 in the former, only at the O-H group of matrix as observed by FTIR data. To our knowledge, this is the first report of the stoichiometric complex of gemcitabine in which an extra component modulates drug release. In Figure 10a, there is the initial fast dissolution of Gem (red trendline), followed by delayed (an increasing concentration, pink trendline) and then sustained release. The early step in Figure 10b (at ca. 0–45 min.) is close to that of dissolution of free Gem in PBS in similar experiment [60] and starting from ca. 45 min., delayed release of Gem from the complex takes place. For the ternary complex, delayed release of gemcitabine occurs despite the large amount (a double molar excess) of citronellol.

4. Conclusions

Solid-state binary complex lag(CYCU-3)(Gem) prepared by LAG retains functional groups and crystal structure of CYCU-3, and has non-covalent bonds of its O-H and COO− groups with Gem. This is detected by analysis of marker FTIR peaks of each component of the complex. The lag(CYCU-3)(Gem) shows in vitro delayed release of gemcitabine to PBS for up to 6000 min. Further, delayed release of Gem and linker of CYCU-3 from the complex to PBS are correlated in time, which indicates that Gem is bonded to CYCU-3, and its release involves destruction of these bonds. The lag(CYCU-3)(Gem) at 1.2–30 μM concentrations shows the substantial, up to 30% suppression of pancreatic cancer cells PANC−1 in a colorimetric assay. This cells suppression is time-dependent, and its timescale is correlated with that of delayed release of Gem from the same complex. In contrast, dry grinding of CYCU-3 and Gem results in solid-state material dry(CYCU-3)(Gem), in which CYCU-3 matrix is strongly amorphized, and which shows no delayed release of gemcitabine. For the first time, a new concept of the multicomponent stoichiometric pharmaceutical material is demonstrated which is synthesized by LAG: ternary complex lag(CYCU-3)1(Gem)1(CIT)2 with natural terpene citronellol. This ternary complex is formed via interactions at the O-H group of CYCU-3 based on a strong ca. 15 cm−1 shift in its IR peak. The complex lag(CYCU-3)1(Gem)1(CIT)2 shows delayed release of gemcitabine to PBS within 2500 min. despite the presence of a 2-fold molar excess of co-encapsulated CIT. Encapsulation of drugs by LAG, followed by a complete or partial removal of grinding fluid, is a promising method to prepare solid-state binary and ternary pharmaceutical materials. They can show in vitro delayed drug release and time-dependent suppression of cancer cells, and they are promising for emerging drug-eluting implants to suppress tumor growth.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/pharmaceutics18020170/s1, Figure S1: Molecular structure of gemcitabine free base Gem. (a) The imino-keto tautomer; (b) the amino-keto tautomer. Table S1: Assignments of IR peaks of CYCU-3 Al-MOF. Figure S2: Powder XRD of CYCU-3. Figure S3: SEM of CYCU-3. Figure S4: ATR-FTIR spectra of Gem. (a) High wavenumbers; (b) mid-IR; (c) low wavenumbers. Table S2: Major IR peaks of Gem. Figure S5: XRD pattern of Gem. Figure S6: Zoomed marker IR peaks of CYCU-3 in lag(CYCU-3)(Gem) vs. pure CYCU-3. (a) The H-O(-Al) stretch; (b) νasym(COO−); (c) νsym(COO−); (d) the H-O(-Al) deformation. Figure S7: Spectra and marker IR peaks of Gem in lag(CYCU-3)(Gem) vs. pure Gem. (a) High wavenumbers; (b) mid-IR; (c) low wavenumbers. Figure S8: XRD of lag(CYCU-3)(Gem). Figure S9: SEM of lag(CYCU-3)(Gem). Figure S10: The HPLC-UV calibration plot of gemcitabine in PBS. Figure S11: Dynamics of release from lag(CYCU-3)(Gem). (a) Linker; (b) gemcitabine. Figure S12: Release of gemcitabine from dry(CYCU-3)(Gem). Figure S13: The ANOVA analysis of viability of PANC−1 cells treated with lag(CYCU-3)(Gem). (a) 48 h.; (b) 72 h. Figure S14: The viability of PANC−1 cells treated with CYCU-3 at 48 h. Figure S15: The viability of healthy cells hTERT−HPNE E6/E7/K−RasG12D treated with CYCU-3 at 48 h. Figure S16: Molecular structure of citronellol (CIT). Figure S17: ATR-FTIR spectra of CIT. (a) High wavenumbers; (b) mid-IR; (c) low wavenumbers. Table S3: Assignments of IR peaks of CIT. Figure S18: ATR-FTIR spectra of lag(CYCU-3)1(Gem)1(CIT)2 vs. Gem. (a) High wavenumbers; (b) mid-IR; (c) low wavenumbers. Figure S19: XRD of lag(CYCU-3)1(Gem)1(CIT)2.

Author Contributions

Conceptualization, A.S.; Methodology, A.S.; Investigation, A.S.; Formal analysis, A.S., S.U. and J.W.; Data curation, K.P. and S.U.; Writing—original draft preparation, A.S.; Writing—review and editing, A.S. and J.W.; Supervision, A.S.; Project administration, A.S. All authors have read and agreed to the published version of the manuscript.

Funding

Research reported in this publication was supported by the National Institute Of General Medical Sciences of the National Institutes of Health under Award Number SC3GM136647. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The authors acknowledge the use of core facilities supported by the National Institute on Minority Health and Health Disparities through grant number 5U54MD013376 and National Institute of General Medical Sciences through grant number 5UL1GM118973.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The original contributions presented in this study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author.

Acknowledgments

A. S. thanks Brandon Carter-Cooper from Translational Laboratory Shared Services (TLSS) at the University of Maryland School of Medicine’s and Greenebaum Comprehensive Cancer Center for measuring suppression of hTERT−HPNE E6/E7/K−RasG12D cells by colorimetric Alamar Blue assay.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Siegel, R.L.; Giaquinto, A.N.; Jemal, A. Cancer statistics, 2024. CA A Cancer J. Clin. 2024, 74, 12–49. [Google Scholar] [CrossRef]
  2. Bray, F.; Laversanne, M.; Sung, H.; Ferlay, J.; Siegel, R.L.; Soerjomataram, I.; Jemal, A. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA A Cancer J. Clin. 2024, 74, 229–263. [Google Scholar] [CrossRef] [PubMed]
  3. Siegel, R.L.; Kratzer, T.B.; Giaquinto, A.N.; Sung, H.; Jemal, A. Cancer statistics, 2025. CA A Cancer J. Clin. 2025, 75, 10–45. [Google Scholar] [CrossRef] [PubMed]
  4. Noble, S.; Goa, K.L. Gemcitabine. A review of its pharmacology and clinical potential in non-small cell lung cancer and pancreatic cancer. Drugs 1997, 54, 447–472. [Google Scholar] [CrossRef] [PubMed]
  5. Haggstrom, L.; Chan, W.Y.; Nagrial, A.; Chantrill, L.A.; Sim, H.W.; Yip, D.; Chin, V. Chemotherapy and radiotherapy for advanced pancreatic cancer. Cochrane Database Syst. Rev. 2024, CD011044. [Google Scholar] [CrossRef]
  6. Wang, Y.; Fan, W.; Dai, X.; Katragadda, U.; McKinley, D.; Teng, Q.; Tan, C. Enhanced tumor delivery of gemcitabine via PEG-DSPE/TPGS mixed micelles. Mol. Pharm. 2014, 11, 1140–1150. [Google Scholar] [CrossRef]
  7. Henry, B.; Samokhvalov, A. Characterization of tautomeric forms of anti-cancer drug gemcitabine and their interconversion upon mechano-chemical treatment, using ATR-FTIR spectroscopy and complementary methods. J. Pharm. Biomed. Anal. 2023, 226, 115243. [Google Scholar] [CrossRef]
  8. Mini, E.; Nobili, S.; Caciagli, B.; Landini, I.; Mazzei, T. Cellular pharmacology of gemcitabine. Ann. Oncol. 2006, 17, v7–v12. [Google Scholar] [CrossRef]
  9. Beutel, A.K.; Halbrook, C.J. Barriers and opportunities for gemcitabine in pancreatic cancer therapy. Am. J. Physiol. Cell Physiol. 2023, 324, C540–C552. [Google Scholar] [CrossRef]
  10. Fowler, J.D.; Brown, J.A.; Johnson, K.A.; Suo, Z. Kinetic investigation of the inhibitory effect of gemcitabine on DNA polymerization catalyzed by human mitochondrial DNA polymerase. J. Biol. Chem. 2008, 283, 15339–15348. [Google Scholar] [CrossRef]
  11. Bhattacharjee, S. Understanding the burst release phenomenon: Toward designing effective nanoparticulate drug-delivery systems. Ther. Deliv. 2020, 12, 21–36. [Google Scholar] [CrossRef]
  12. Molehin, O.R.; Idowu, K.A.; Olaoye, A.B.; Fakayode, A.E.; Adesua, O.O. Influence of Clerodendrum volubile leaf extract on doxorubicin-induced toxicity and inhibition of carbonyl reductase mediated metabolism. J. Complement. Integr. Med. 2022, 19, 937–946. [Google Scholar] [CrossRef]
  13. Saiki, Y.; Hirota, S.; Horii, A. Attempts to remodel the pathways of gemcitabine metabolism: Recent approaches to overcoming tumours with acquired chemoresistance. Cancer Drug Resist. 2020, 3, 819–831. [Google Scholar] [CrossRef] [PubMed]
  14. Bai, L.; Li, X.; Yang, Y.; Zhao, R.; White, E.Z.; Danaher, A.; Bowen, N.J.; Hinton, C.V.; Cook, N.; Li, D.; et al. Bromocriptine monotherapy overcomes prostate cancer chemoresistance in preclinical models. Transl. Oncol. 2023, 34, 101707. [Google Scholar] [CrossRef] [PubMed]
  15. Pitorre, M.; Gazaille, C.; Pham, L.T.T.; Frankova, K.; Béjaud, J.; Lautram, N.; Riou, J.; Perrot, R.; Geneviève, F.; Moal, V.; et al. Polymer-free hydrogel made of lipid nanocapsules, as a local drug delivery platform. Mater. Sci. Eng. C 2021, 126, 112188. [Google Scholar] [CrossRef] [PubMed]
  16. Tang, H.; Zhang, Z.; Zhu, M.; Xie, Y.; Lv, Z.; Liu, R.; Shen, Y.; Pei, J. Efficient delivery of gemcitabine by estrogen receptor-targeted pegylated liposome and its anti-lung cancer activity in vivo and in vitro. Pharmaceutics 2023, 15, 988. [Google Scholar] [CrossRef]
  17. Wolinsky, J.B.; Colson, Y.L.; Grinstaff, M.W. Local drug delivery strategies for cancer treatment: Gels, nanoparticles, polymeric films, rods, and wafers. J. Control. Release 2012, 159, 14–26. [Google Scholar] [CrossRef]
  18. Obeidat, W.M.; Lahlouh, I.K. Chitosan Nanoparticles: Approaches to preparation, key properties, drug delivery systems, and developments in therapeutic efficacy. AAPS PharmSciTech 2025, 26, 108. [Google Scholar] [CrossRef]
  19. Horcajada, P.; Chalati, T.; Serre, C.; Gillet, B.; Sebrie, C.; Baati, T.; Eubank, J.F.; Heurtaux, D.; Clayette, P.; Kreuz, C.; et al. Porous metal-organic-framework nanoscale carriers as a potential platform for drug delivery and imaging. Nat. Mater. 2010, 9, 172–178. [Google Scholar] [CrossRef]
  20. Qin, J.; Jiang, K.; Yuan, Z.; Zhang, L.; Chen, B. A Multifunctional MOF heterojunction for combating antibiotic resistance and promoting regenerative immunity. ACS Nano 2025, 19, 39768–39782. [Google Scholar] [CrossRef]
  21. Lo, S.-H.; Chien, C.-H.; Lai, Y.-L.; Yang, C.-C.; Lee, J.J.; Raja, D.S.; Lin, C.-H. A mesoporous aluminium metal–organic framework with 3 nm open pores. J. Mater. Chem. A 2013, 1, 324–329. [Google Scholar] [CrossRef]
  22. Lucena, F.R.S.; de Araújo, L.C.C.; Rodrigues, M.D.D.; da Silva, T.G.; Pereira, V.R.A.; Militão, G.C.G.; Fontes, D.A.F.; Rolim-Neto, P.J.; da Silva, F.F.; Nascimento, S.C. Induction of cancer cell death by apoptosis and slow release of 5-fluoracil from metal-organic frameworks Cu-BTC. Biomed. Pharmacother. 2013, 67, 707–713. [Google Scholar] [CrossRef] [PubMed]
  23. Samokhvalov, A. Adsorption on Mesoporous Metal-Organic Frameworks in Solution for Clean Energy, Environment and Healthcare; CRC Press: Boca Raton, FL, USA, 2017. [Google Scholar]
  24. Uzunova-Bujnova, M.; Dimitrov, D.; Radev, D.; Bojinova, A.; Todorovsky, D. Effect of the mechanoactivation on the structure, sorption and photocatalytic properties of titanium dioxide. Mater. Chem. Phys. 2008, 110, 291–298. [Google Scholar] [CrossRef]
  25. Kaneva, N.; Bojinova, A.; Papazova, K.; Dimitrov, D.; Zaharieva, K.; Cherkezova-Zheleva, Z.; Eliyas, A. Effect of thermal and mechano-chemical activation on the photocatalytic efficiency of ZnO for drugs degradation. Arch. Pharm. Res. 2016, 39, 1418–1425. [Google Scholar] [CrossRef] [PubMed]
  26. Sović, I.; Lukin, S.; Meštrović, E.; Halasz, I.; Porcheddu, A.; Delogu, F.; Ricci, P.C.; Caron, F.; Perilli, T.; Dogan, A.; et al. Mechanochemical preparation of active pharmaceutical ingredients monitored by in situ Raman spectroscopy. ACS Omega 2020, 5, 28663–28672. [Google Scholar] [CrossRef]
  27. Shah, U.V.; Wang, Z.; Olusanmi, D.; Narang, A.S.; Hussain, M.A.; Tobyn, M.J.; Heng, J.Y.Y. Effect of milling temperatures on surface area, surface energy and cohesion of pharmaceutical powders. Int. J. Pharm. 2015, 495, 234–240. [Google Scholar] [CrossRef]
  28. Denlinger, K.L.; Ortiz-Trankina, L.; Carr, P.; Benson, K.; Waddell, D.C.; Mack, J. Liquid-assisted grinding and ion pairing regulates percentage conversion and diastereoselectivity of the Wittig reaction under mechanochemical conditions. Beilstein J. Org. Chem. 2018, 14, 688–696. [Google Scholar] [CrossRef]
  29. Mukhopadhyay, R.; Sen, R.; Paul, B.; Kazi, J.; Ganguly, S.; Debnath, M.C. Gemcitabine co-encapsulated with curcumin in folate decorated PLGA nanoparticles; a novel approach to treat breast adenocarcinoma. Pharm Res 2020, 37, 56. [Google Scholar] [CrossRef]
  30. Santos, P.L.; Matos, J.P.S.C.F.; Picot, L.; Almeida, J.R.G.S.; Quintans, J.S.S.; Quintans-Júnior, L.J. Citronellol, a monoterpene alcohol with promising pharmacological activities—A systematic review. Food Chem. Toxicol. 2019, 123, 459–469. [Google Scholar] [CrossRef]
  31. Rajendran, J.; Pachaiappan, P.; Thangarasu, R. Citronellol, an acyclic monoterpene induces mitochondrial-mediated apoptosis through activation of proapoptotic factors in MCF-7 and MDA-MB-231 human mammary tumor cells. Nutr. Cancer 2021, 73, 1448–1458. [Google Scholar] [CrossRef]
  32. Tiernan, H.; Byrne, B.; Kazarian, S.G. ATR-FTIR spectroscopy and spectroscopic imaging for the analysis of biopharmaceuticals. Spectrochim. Acta A 2020, 241, 118636. [Google Scholar] [CrossRef] [PubMed]
  33. Grisedale, L.C.; Jamieson, M.J.; Belton, P.; Barker, S.A.; Craig, D.Q.M. Characterization and quantification of amorphous material in milled and spray-dried salbutamol sulfate: A comparison of thermal, spectroscopic, and water vapor sorption approaches. J. Pharm. Sci. 2011, 100, 3114–3129. [Google Scholar] [CrossRef] [PubMed]
  34. Du, P.-Y.; Gu, W.; Liu, X. A three-dimensional Nd(III)-based metal-organic framework as a smart drug carrier. New J. Chem. 2016, 40, 9017–9020. [Google Scholar] [CrossRef]
  35. Réjiba, S.; Bigand, C.; Parmentier, C.; Hajri, A. Gemcitabine-based chemogene therapy for pancreatic cancer using Ad-dCK::UMK GDEPT and TS/RR siRNA strategies. Neoplasia 2009, 11, 637–650. [Google Scholar] [CrossRef]
  36. Li, X.; He, G.; Su, F.; Chu, Z.; Xu, L.; Zhang, Y.; Zhou, J.; Ding, Y. Regorafenib-loaded poly (lactide-co-glycolide) microspheres designed to improve transarterial chemoembolization therapy for hepatocellular carcinoma. Asian J. Pharm. Sci. 2020, 15, 739–751. [Google Scholar] [CrossRef]
  37. Rongthong, T.; Pongjanyakul, T. Quaternary polymethacrylate−magnesium aluminum silicate film formers: Stability studies for tablet coatings. J. Drug Deliv. Sci. Technol. 2021, 62, 102389. [Google Scholar] [CrossRef]
  38. Betsiou, M.; Bantsis, G.; Zoi, I.; Sikalidis, C. Adsorption and release of gemcitabine hydrochloride and oxaliplatin by hydroxyapatite. Ceram. Int. 2012, 38, 2719–2724. [Google Scholar] [CrossRef]
  39. Foucault-Collet, A.; Gogick, K.A.; White, K.A.; Villette, S.; Pallier, A.; Collet, G.; Kieda, C.; Li, T.; Geib, S.J.; Rosi, N.L.; et al. Lanthanide near infrared imaging in living cells with Yb3+ nano metal organic frameworks. Proc. Natl. Acad. Sci. USA 2013, 110, 17199–17204. [Google Scholar] [CrossRef]
  40. Lapidus, R.G.; Carter-Cooper, B.A.; Sadowska, M.; Choi, E.Y.; Wonodi, O.; Muvarak, N.; Natarajan, K.; Pidugu, L.S.; Jaiswal, A.; Toth, E.A.; et al. Hydroxylated dimeric naphthoquinones increase the generation of reactive oxygen species, induce apoptosis of acute myeloid leukemia cells and are not substrates of the multidrug resistance proteins ABCB1 and ABCG2. Pharmaceuticals 2016, 9, 4. [Google Scholar] [CrossRef]
  41. Salazar, J.M.; Weber, G.; Simon, J.M.; Bezverkhyy, I.; Bellat, J.P. Characterization of adsorbed water in MIL-53(Al) by FTIR spectroscopy and ab-initio calculations. J. Chem. Phys. 2015, 142, 124702. [Google Scholar] [CrossRef]
  42. Banga-Bothy, G.-A.; Samokhvalov, A. Porphyrin aluminum MOF with ultra-high water sorption capacity: In-situ time-dependent ATR-FTIR spectroscopy and gravimetry to study mechanism of water bonding and desorption. Vib. Spectrosc. 2022, 119, 103356. [Google Scholar] [CrossRef]
  43. Henry, B.; Samokhvalov, A. Hygroscopic metal-organic framework MIL-160(Al): In-situ time-dependent ATR-FTIR and gravimetric study of mechanism and kinetics of water vapor sorption. Spectrochim. Acta A 2022, 267, 120550. [Google Scholar] [CrossRef] [PubMed]
  44. Niekiel, F.; Ackermann, M.; Guerrier, P.; Rothkirch, A.; Stock, N. Aluminum-1,4-cyclohexanedicarboxylates: High-throughput and temperature-dependent in situ EDXRD studies. Inorg. Chem. 2013, 52, 8699–8705. [Google Scholar] [CrossRef]
  45. Hoffman, A.E.J.; Vanduyfhuys, L.; Nevjestić, I.; Wieme, J.; Rogge, S.M.J.; Depauw, H.; Van Der Voort, P.; Vrielinck, H.; Van Speybroeck, V. Elucidating the vibrational fingerprint of the flexible metal–organic framework MIL-53(Al) using a combined experimental/computational approach. J. Phys. Chem. C 2018, 122, 2734–2746. [Google Scholar] [CrossRef] [PubMed]
  46. Silverstein, R.M.; Webster, F.X.; Kiemle, D.J.; Bryce, D.L. Spectrometric Identification of Organic Compounds, 8th ed.; Wiley: Hoboken, NJ, USA, 2014. [Google Scholar]
  47. Zhao, X.; Gao, Z.; Li, Z.; Huang, H. A highly fluorescent Al3+-based metal–organic framework (CYCU-3) for selective and sensitive sensing of 2,4,6-trinitrophenol. J. Porous Mater. 2018, 25, 1597–1602. [Google Scholar] [CrossRef]
  48. Florián, J.; Baumruk, V.; Leszczyński, J. IR and Raman spectra, tautomeric stabilities, and scaled quantum mechanical force fields of protonated cytosine. J. Phys. Chem. 1996, 100, 5578–5589. [Google Scholar] [CrossRef]
  49. Ten, G.N.; Baranov, V.I. Calculation and analysis of the IR spectra of cytosine in various phase states. J. Appl. Spectrosc. 2005, 72, 155–163. [Google Scholar] [CrossRef]
  50. Rezkallah, E.; Ibrahim, A.; Dahy, A.; Hakiem Ahmed, A.; Mahfouz, R. DFT and thermal decomposition studies on gemcitabine. Z. Phys. Chem. 2019, 233, 1503–1507. [Google Scholar] [CrossRef]
  51. Wu, D.; Tang, P.X.; Li, S.S.; Luo, H.Z.; Li, H. X-ray powder diffraction data for gemcitabine, C9H11F2N3O4. Powder Diffr. 2015, 30, 76–78. [Google Scholar] [CrossRef]
  52. Ravon, U.; Chaplais, G.; Chizallet, C.; Seyyedi, B.; Bonino, F.; Bordiga, S.; Bats, N.; Farrusseng, D. Investigation of acid centers in MIL-53(Al, Ga) for Brønsted-type catalysis: In situ FTIR and ab initio molecular modeling. ChemCatChem 2010, 2, 1235–1238. [Google Scholar] [CrossRef]
  53. Stoyanov, E.S.; Malykhin, S.E. Carbon monoxide protonation in condensed phases and bonding to surface superacidic Brønsted centers. Phys. Chem. Chem. Phys. 2016, 18, 4871–4880. [Google Scholar] [CrossRef]
  54. Wu, C.; You, J.; Wang, X. Thermal decomposition mechanism and kinetics of gemcitabine. J. Anal. Appl. Pyrolysis 2018, 130, 118–126. [Google Scholar] [CrossRef]
  55. Simonin, J.-P. On the comparison of pseudo-first order and pseudo-second order rate laws in the modeling of adsorption kinetics. Chem. Eng. J. 2016, 300, 254–263. [Google Scholar] [CrossRef]
  56. Songkro, S.; Hayook, N.; Jaisawang, J.; Maneenuan, D.; Chuchome, T.; Kaewnopparat, N. Investigation of inclusion complexes of citronella oil, citronellal and citronellol with β-cyclodextrin for mosquito repellent. J. Incl. Phenom. Macrocycl. Chem. 2012, 72, 339–355. [Google Scholar] [CrossRef]
  57. Baranska, M.; Schulz, H.; Walter, A.; Rösch, P.; Quilitzsch, R.; Lösing, G.; Popp, J. Investigation of eucalyptus essential oil by using vibrational spectroscopy methods. Vib. Spectrosc. 2006, 42, 341–345. [Google Scholar] [CrossRef]
  58. Cao, N.; Liu, J.; Wang, Y.; Zhou, Y.; Zhao, M.; Ban, Y.; Yang, W. MIL-53 and its OH-bonded variants for bio-polyol adsorption from aqueous solution. RSC Adv. 2022, 12, 21681–21689. [Google Scholar] [CrossRef]
  59. Rehman, T.U.; Agnello, S.; Gelardi, F.M.; Calvino, M.M.; Lazzara, G.; Buscarino, G.; Cannas, M. Unveiling the MIL-53(Al) MOF: Tuning photoluminescence and structural properties via volatile organic compounds interactions. Nanomaterials 2024, 14, 388. [Google Scholar] [CrossRef]
  60. Umar, S.; Samokhvalov, A. Encapsulation of gemcitabine on porphyrin aluminum metal-organic framework by mechano-chemistry, delayed drug release and cytotoxicity to pancreatic cancer PANC−1 cells. Molecules 2024, 29, 3189. [Google Scholar] [CrossRef]
Pharmaceutics 18 00170 g001
Figure 1. The simplified 2D structure of SBU in Al-MOF CYCU-3.
Figure 1. The simplified 2D structure of SBU in Al-MOF CYCU-3.
Pharmaceutics 18 00170 g001
Pharmaceutics 18 00170 g002
Figure 2. ATR-FTIR spectra of CYCU-3. (a) High wavenumbers; (b) mid-IR; (c) low wavenumbers.
Figure 2. ATR-FTIR spectra of CYCU-3. (a) High wavenumbers; (b) mid-IR; (c) low wavenumbers.
Pharmaceutics 18 00170 g002
Pharmaceutics 18 00170 g003
Figure 3. ATR-FTIR spectra of complex lag(CYCU-3)(Gem) vs. CYCU-3. (a) High wavenumbers; (b) mid-IR; (c) low wavenumbers.
Figure 3. ATR-FTIR spectra of complex lag(CYCU-3)(Gem) vs. CYCU-3. (a) High wavenumbers; (b) mid-IR; (c) low wavenumbers.
Pharmaceutics 18 00170 g003
Pharmaceutics 18 00170 g004
Figure 4. Proposed bonding in lag(CYCU-3)(Gem).
Figure 4. Proposed bonding in lag(CYCU-3)(Gem).
Pharmaceutics 18 00170 g004
Pharmaceutics 18 00170 g005
Figure 5. XRD pattern of dry(CYCU-3)(Gem).
Figure 5. XRD pattern of dry(CYCU-3)(Gem).
Pharmaceutics 18 00170 g005
Pharmaceutics 18 00170 g006
Figure 6. Release of gemcitabine from lag(CYCU-3)(Gem). (a) The complete (long) timescale. (b) The initial (short) timescale.
Figure 6. Release of gemcitabine from lag(CYCU-3)(Gem). (a) The complete (long) timescale. (b) The initial (short) timescale.
Pharmaceutics 18 00170 g006
Pharmaceutics 18 00170 g007
Figure 7. The viability of PANC−1 cells treated with lag(CYCU-3)(Gem).
Figure 7. The viability of PANC−1 cells treated with lag(CYCU-3)(Gem).
Pharmaceutics 18 00170 g007
Pharmaceutics 18 00170 g008
Figure 8. ATR-FTIR spectra of lag(CYCU-3)1(Gem)1(CIT)2 versus CYCU-3. (a) High wavenumbers; (b) mid-IR; (c) low wavenumbers.
Figure 8. ATR-FTIR spectra of lag(CYCU-3)1(Gem)1(CIT)2 versus CYCU-3. (a) High wavenumbers; (b) mid-IR; (c) low wavenumbers.
Pharmaceutics 18 00170 g008
Pharmaceutics 18 00170 g009
Figure 9. ATR-FTIR spectra of lag(CYCU-3)1(Gem)1(CIT)2 versus CIT. (a) High wavenumbers; (b) mid-IR; (c) low wavenumbers.
Figure 9. ATR-FTIR spectra of lag(CYCU-3)1(Gem)1(CIT)2 versus CIT. (a) High wavenumbers; (b) mid-IR; (c) low wavenumbers.
Pharmaceutics 18 00170 g009
Pharmaceutics 18 00170 g010
Figure 10. Release of gemcitabine from lag(CYCU-3)1(Gem)1(CIT)2. (a) Full timescale; (b) the initial period.
Figure 10. Release of gemcitabine from lag(CYCU-3)1(Gem)1(CIT)2. (a) Full timescale; (b) the initial period.
Pharmaceutics 18 00170 g010
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Panthi, K.; Umar, S.; Wachira, J.; Samokhvalov, A. Pharmaceutical Binary and Ternary Complexes of Gemcitabine with Aluminum Metal–Organic Framework: Mechano-Chemical Encapsulation, Delayed Drug Release, and Toxicity to Pancreatic Cells. Pharmaceutics 2026, 18, 170. https://doi.org/10.3390/pharmaceutics18020170

AMA Style

Panthi K, Umar S, Wachira J, Samokhvalov A. Pharmaceutical Binary and Ternary Complexes of Gemcitabine with Aluminum Metal–Organic Framework: Mechano-Chemical Encapsulation, Delayed Drug Release, and Toxicity to Pancreatic Cells. Pharmaceutics. 2026; 18(2):170. https://doi.org/10.3390/pharmaceutics18020170

Chicago/Turabian Style

Panthi, Kamala, Sheriff Umar, James Wachira, and Alexander Samokhvalov. 2026. "Pharmaceutical Binary and Ternary Complexes of Gemcitabine with Aluminum Metal–Organic Framework: Mechano-Chemical Encapsulation, Delayed Drug Release, and Toxicity to Pancreatic Cells" Pharmaceutics 18, no. 2: 170. https://doi.org/10.3390/pharmaceutics18020170

APA Style

Panthi, K., Umar, S., Wachira, J., & Samokhvalov, A. (2026). Pharmaceutical Binary and Ternary Complexes of Gemcitabine with Aluminum Metal–Organic Framework: Mechano-Chemical Encapsulation, Delayed Drug Release, and Toxicity to Pancreatic Cells. Pharmaceutics, 18(2), 170. https://doi.org/10.3390/pharmaceutics18020170

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop