Design and development of metal-organic framework-based nanocomposite hydrogels for quantification of deferiprone in exhaled breath condensate

Abstract

In this study, a novel fluorescence nanoprobe based on Materials of Institute Lavoisier (MIL-101) metal-organic frameworks embedding into the agarose hydrogel is fabricated using a hydrothermal technique. It uses for sensitive quantification of deferiprone in exhaled breath condensate (EBC) samples. The morphology and characterization of MIL-101/agarose nanocomposite hydrogel is studied by transmission electron microscopy, dynamic light scattering instrument, powder X-ray diffraction analysis, and Fourier transform infrared spectroscopy. The probe shows a reasonable fluorescence intensity quenching in the presence of deferiprone due to the interactions between iron centers in MIL-101 (Fe) and deferiprone, which likely form non-fluorescent complexes. The proposed nanoprobe demonstrates a linear calibration curve from 0.005 to 1.5 µg mL^− 1 with a detection limit of 0.003 µg mL^− 1. The intra- and inter-day precision of the reported method are 0.3% and 0.4% (n = 5, deferiprone concentration = 1.0 µg mL^− 1), respectively. This method demonstrates high sensitivity and specificity towards deferiprone in the EBC samples and also presents a sensing platform with simplicity, convenience, fast implementation, and cost-effective in medical monitoring.

Introduction

Deferiprone as an orally active iron chelator was developed as a new therapy for iron overload. It is used globally to treat various diseases, including cancer, leukemia, and hemodialysis [1]. Patients who are being treated with deferiprone may experience some adverse effects including neutropenia/agranulocytosis, gastrointestinal problems, raised alanine amino transferase levels, and zinc deficiency. Deferiprone reacts with ferric ions to create 1:3 iron-deferiprone complexes that are neutral and hydrophilic [2]. Monitoring patients with liver or kidney problems is vital when administering medication with deferiprone. Measuring the drug levels in the blood or other body fluids is essential to determine the optimal dosage and frequency of administration, enhancing the effectiveness of the treatment while minimizing the risk of side effects [3]. Several quantitative analytical techniques have been developed for the determination of deferiprone, including capillary electrophoresis-frontal analysis [4], electrochemical [5, 6], reverse phase-high performance liquid chromatography with a UV detector (RP-HPLC) [7, 8], and spectroscopy [9] procedures. Chromatographic techniques are time-consuming processes and require lengthy sample preparation steps and the use of high-cost reagents. Additionally, electrochemical methods often exhibit limited reproducibility, which can hinder the development of new methods. Although these developed methods have their advantages and disadvantages, developing a new and easy-to-use method for drug concentration measuring in complex biological samples is a challenging issue for researchers. In contrast to other methods, optical-based methods are rapid and straightforward to operate, making them suitable for deployment in remote areas, but they lack sufficient selectivity and sensitivity. To enhance the selectivity and sensitivity of optical methods, the implementation of reliable platforms [10] or advanced computational processing techniques [11, 12] is crucial. In this regard, metal-organic frameworks (MOFs) are crystalline frameworks with high porosity, created by assembling inorganic metal-based nodes or clusters with organic ligands through strong coordination bonds [13]. The diverse array of metal ions, organic linkers, and structural motifs enables an almost limitless number of potential combinations. Additionally, the ability to modify structures after synthesis introduces an extra layer of complexity. When combined with the rapidly growing library of experimentally determined structures, it becomes feasible to accurately predict, using computational methods, the affinity of guests for host frameworks. This has significant implications for designing frameworks that can deliver specific properties [14]. The large specific surface area, considerable thermal and chemical stabilities, manageable compositions, and uniform pore size distribution are some of the superior properties of MOFs [15]. These aspects have made MOFs ideal candidates for various applications [16,17,18]. The Materials Institute Lavoisier’s (MILs) MOFs are a well-known class of materials characterized by their outstanding stability, permanent porosity, and extensive specific surface area [19]. Second-generation MOFs, exemplified by MILs, are formed via the combination of phthalic acid or terephthalic acid with trivalent metal ions like (Cr³⁺, Fe³⁺, Al³⁺, etc.). These materials are known for their suitable thermal and chemical stability and many uncoordinated transition metal sites. MIL-101 serves as a prototypical representative of the MIL series, whose versatility allows for the incorporation of various functional groups to enhance its selectivity [20]. For instance, MOFs have the potential to be combined with various materials such as metal nanoparticles, polymers, graphene, and carbon nanotubes to produce various composites that possess unique functionality not present in pure MOF. Recent advances in MOF-polymer composites have led to the development of a new generation of materials with expanded practical applications, featuring desirable properties such as improved processability, electrical conductivity, molecular recognition capabilities, chemical and colloidal stability, and biocompatibility [21]. Recent research has focused on the development of MOF-based hydrogels, which have shown significant improvements over traditional MOFs in terms of mechanical strength, absorption capacity, and total pore volume [22]. Hydrogels are polymeric materials that can absorb and retain a significant amount of water without dissolving [23]. Hydrogels are being extensively researched for their superior softness, wetness, responsiveness, biocompatibility, and bioactivity. Agarose is a type of natural hydrogel derived from a polysaccharide found in the cell walls of marine red algae. Agarose can form a gel by aggregating its hexagonal fibers, which consist of six double helices, and absorbing a large amount of water within its network. Due to its hydrophilic polymeric framework, chemical stability, and tunable pore size, this material has become a popular choice as a supporting medium for optical sensors [24].

The objective of this study was to develop a sensing platform for determining deferiprone levels in exhaled breath condensate (EBC) by leveraging the unique properties of MIL-101/agarose nanocomposite hydrogels (NCH). Cooling-down strategy was utilized to synthesize MIL-101/agarose NCH to achieve the desired structure, stability, and morphology. Herein, agarose as the polymeric matrix, MIL-101 (Fe) as nanofiller, and reinforcement of hydrogel were employed. Deferiprone was selected as a representative analyte due to its wide range of therapeutic applications. In a subsequent step, the probe was used to investigate its ability to detect deferiprone in the EBC of patients undergoing treatment with this medication (Fig. 1). The uniqueness of the method lies in its sustainable and biocompatible materials, ease of synthesis, and rapid detection capabilities, making it an innovative solution for sensing deferiprone in EBC.

Fig. 1

A schematic illustration of MIL-101/agarose NCH synthesis and its application for deferiprone determination

Materials and methods

Reagents and solutions

Chemicals were obtained from commercial sources. The chemical and solvents, including N, N-dimethylformamide (DMF), terephthalic acid (H₂BDC, Mw:166.13 g mol^− 1) and agarose NCH were purchased from Sigma-Aldrich (www.sigmaaldrich.com). Sodium dihydrogen phosphate (NaH₂PO₄) with an M_w of 119.98 g mol^− 1 was utilized for buffer preparation. The pH of the buffer was adjusted to various levels using sodium hydroxide (NaOH) and hydrochloric acid (HCl) also FeCl₃.6H₂O and ethanol (EtOH) for the synthesis of MIL-101 (Fe) were supplied from Merck (www.merck.com). Deferiprone (> 99.7%) was achieved from Arastoo Pharmaceutical Co. (Tehran, Iran) and ultrapure deionized water from (Ghazi Pharmaceutical Co., Tabriz, Iran, www.sgco-infusion.com) was used for the dilution process. All reagents were stored at 4 ^◦C before usage.

Apparatus and characterization

The fluorescence spectra were recorded using an FP-750 spectrofluorometer (Jasco Corp., Japan) with a xenon lamp source. A 1 mL standard quartz cell was used for all fluorescence spectra measurements, with a bandwidth of 10 nm set in the excitation and 20 nm in emission paths. The distribution of particle size in the prepared nanoprobe was estimated and their shape and size were evaluated using a CM120 transmission electron microscopy (TEM) and dynamic light scattering (DLS) (Philips, The Netherlands, www.philips.com). The prepared solutions’ UV-vis absorption spectra were recorded on a spectrophotometer model UV-1800 using a micro quartz cell (Shimadzu, Japan, www.shimadzu.com). Powder X-ray diffraction (XRD) patterns were obtained using a Siemens diffractometer with filtered Cu-Kα radiation at 35 kV within the 2θ range 4^°-70^°. Fourier transforms infrared (FT-IR) spectroscopy was used to verify the chemical bonding of the samples. The spectral width was set at 400–4000 cm^− 1, and the instrument used was Bruker Tensor 27 (KBr wafer technique). MIL-101 (Fe) was successfully synthesized using an oven (Initiator 8 EXP, 2450 MHz frequency, Biotage Corp). The pH was adjusted using a model 744 digital pH meter (Metrohm Ltd., Switzerland).

Synthesis of MIL-101 (fe)

MIL-101 (Fe) was synthesized via hydrothermal synthesis, as described in [25]. First, 1.351 g of FeCl₃·6H₂O (5 mmol) and 0.415 g (2.5 mmol) of H₂BDC were mixed in 30 mL of DMF. The mixture was fully suspended by sonication for 20 min. Next, the suspension was placed in a Teflon-lined stainless-steel autoclave, sealed, and heated at 110 °C for 24 h. Then, the obtained solution was centrifuged, and the precipitated orange mud was washed three times with DMF and hot EtOH to remove the raw materials. The orange mud was dried at 70 °C for 60 min and then activated at 150 °C for 12 h.

Synthesis of MIL-101/agarose NCH

To prepare the MIL-101/agarose NCH according to other MOF and hydrogel composite [26], first, 0.0334 g of agarose was dissolved in 50 mL of water with a stirrer until fully dissolved at 80 °C. In the next part, 2.5 mg of synthesized MIL-101 was dispersed in 1 mL of deionized water and sonicated for 10 min. Then, 668 µL of the pre-prepared MIL-101 solution was added to the hot agarose solution and stirred for another 10 min at 90 °C to ensure homogeneity. The solution was cooled to room temperature to form MIL-101/agarose NCH. The obtained hydrogel was kept at 4 °C for further experiments.

Samples preparation

To prepare working standard solutions, our stock solutions were diluted daily with deionized water. To collect human EBC samples, we employed an exhalation collection device [27]. The EBC device featured a cooling trap with an adjustable temperature ranging from 0 °C to − 25 °C. The cooling trap rapidly cooled the blown air and then removed the vaporized particles from the surface of the trap through condensed water vapor particles. The EBC samples were analyzed directly, without any prior preparation steps. The samples used for optimization and calibration methods were collected from healthy volunteers. Additionally, real samples were collected from four patients receiving deferiprone to check the method’s applicability. The sample donors signed a consent form confirmed by the ethics committee of Tabriz University of Medical Sciences, with the code IR.TBZMED.PHARMACY.REC.1402.050.

General procedure

The fluorescence detection was done in a 2 mL vial using a batch procedure. For optimal conditions, 150 µL of EBC was added to the microtube and 10 µL of 0.1 mol L^− 1 phosphate buffer (pH 4.0). Then, 20 µL of MIL-101/agarose NCH was added, and the final volume was adjusted to 0.5 mL using deionized water. The fluorescence intensity of the nanoprobe signal was measured immediately at around λ_max = 410 nm after excitation at 360 nm. The analytical signal was quantified as ΔF, where ΔF is the difference between the probe intensity in the presence (F) and absence (F₀) of deferiprone. The experiments were conducted at room temperature.

Results and discussions

Structural characterization of MIL-101/agarose NCH

TEM was employed to investigate the morphology and particle size of MIL-101/agarose NCH. As depicted in Fig. 2A, MIL-101 (Fe) was dispersed well throughout the agarose matrix without aggregation and showed an intact crystal morphology with a size of lower than 200 nm. Zeta potential measurements in Fig. 2B indicated a negative zeta potential of -21.5 mV for MIL-101/agarose NCH. The negative zeta potential observed in the MIL-101/agarose NCH is mainly due to deprotonated carboxylate groups in the MIL-101 (Fe) structure. Additionally, the hydroxyl groups in the agarose component may play a role in increasing the negative surface charge. The combination of functional groups from the MOF structure and agarose molecules leads to the overall negative zeta potential observed in the NCH.

Fig. 2

(A) TEM image, and (B) zeta potential of MIL-101/agarose NCH

To further verify the successful synthesis of MIL-101/agarose NCH, FT-IR spectroscopy was conducted (Fig. 3A). For as-prepared MIL-101 (Fe), the absorption band at around 747, 1017, 1393, 1507 and 1600 cm^− 1 belong to C-H bending vibrations in benzene, symmetric vibration of the carboxyl group (-COO-), asymmetric vibration of the carboxyl group (-COO-), C = O bonds in free carboxylic groups, respectively [28]. The C-H bending vibrations in benzene are indicated by the peak at 747 cm^− 1. Symmetric and asymmetric vibrations of carboxyl groups (-COO-) are assigned to peaks at 1393 and 1507 cm^− 1, respectively, while the peak at 1600 cm^− 1 is related to C = O bonds in free carboxylic groups, demonstrating the dicarboxylate linker. These FT-IR peaks confirm the organic carboxylate bridging ligand structure and prove that the orange powder is crystal pure MIL-101 (Fe) [25]. The FT-IR spectra of agarose reveals specific absorption peaks at 3439 cm^− 1 (OH stretching of the hydroxyl group), 1646 cm^− 1 (C = O stretching vibration), 1073 cm^− 1 (C-O glycosidic bonding), and 930 cm^− 1 (vibration of C-O-C bridge of 3, 6-anhydro-L-galactopyranose) [29]. In the case of MIL-101/agarose NCH, similar bands with MIL-101 (Fe) and agarose with slightly different positions and intensities are shown, indicating the successful incorporation of MIL-101 (Fe) into the agarose matrix. Also, XRD analysis was performed on the samples to verify the crystallinity and phase purity of the materials (Fig. 3B). The typical characterization peaks of MIL-101 (Fe) can be seen at 9.1°, and 10.14° related to (311) and (511) crystal planes at reasonably high intensity similar to that observed in previous studies [30]. Pure agarose shows a single peak at 2ϴ = 19.24° [31]. By comparing the XRD patterns of MIL-101/agarose NCH with MIL-101 (Fe) and agarose alone, the appearance of the peaks as mentioned earlier evidenced the successful fabrication of the MIL-101/agarose NCH.

Fig. 3

(A) FT-IR spectra and (B) XRD patterns of Agarose, MIL-101 (Fe), and MIL-101/agarose NCH

Sensing mechanism

The performance of the synthesized nanoprobe has been analyzed after ensuring the preparation process. The interaction between the nanoprobe and deferiprone has been distinguished by studying the spectra and analytical response of the nanoprobe. A quenching process was observed in the fluorescence intensity of MIL-101/agarose NCH system after deferiprone adding, which can be attributed to the interactions between iron centers in MIL-101 (Fe) and deferiprone. These interactions are (i) coordination bonds: MIL-101 contains metal centers (chromium ions) that can form coordination bonds with functional groups on deferiprone, such as oxygen atoms from hydroxyl groups (-OH) or nitrogen atoms from the pyridine ring in deferiprone. (ii) hydrogen bonding: Hydrogen bonding can occur between hydroxyl groups (-OH) or nitrogen atoms in deferiprone and hydrogen bond acceptor sites on the MOF, such as hydroxyl or carbonyl groups on the terephthalate ligands and (iii) electrostatic interactions: Deferiprone contains ionizable groups , and MIL-101 may have charged sites or surface charges that can engage in electrostatic interactions with these groups. The interactions may have resulted in the creation of non-fluorescent complexes that do not exhibit fluorescence when excited, thereby reducing the observed fluorescence intensity [32]. The porous structure of MIL-101 (Fe) supports these interactions by facilitating the diffusion of deferiprone into the framework where it can directly interact with iron centers. The embedding in agarose does not impede this interaction but rather contributes to a controlled environment where deferiprone can efficiently access the MIL-101 (Fe) pores. This structural arrangement ensures that the quenching mechanisms are not merely surface phenomena but are deeply integrated into the material’s framework, enhancing the sensitivity and specificity of the system towards deferiprone. The fluorescence signal of the proposed probe decreases in a concentration-dependent manner as deferiprone concentration increases, rendering it a valuable tool for the determination of deferiprone in various matrices.

Optimization of reaction conditions

Optimizing experimental conditions is crucial for probe design as it directly affects method performance. In the early stages of our study, we discovered that three variables (pH, MIL-101/agarose NCH concentration, and incubation time) had a significant impact on the fluorescent intensity. A deferiprone concentration of 1 µg mL⁻1 was set as the target value for the optimization protocol. The pH of the reaction environment plays a crucial role in modulating the complexation reactions between metal ions and ligands [33]. The effect of pH values was investigated on the system’s response using phosphate buffer (PBS, 0.10 mol.L^− 1) ranging from 2.0 to 8.0. As the pH increased, the nanoprobe’s response exhibited a marked enhancement, peaking at pH 4.0, where the highest fluorescence response (ΔF) was observed, and subsequently decreasing in higher pH values, as shown in Fig. 4A. The maximum fluorescence response was achieved at pH 4.0, likely due to the strong interaction between the deferiprone’s N-containing group and the negative surface of nanoparticles, due to the pK_a value of deferiprone [34]. The optimal amount of MIL-101/agarose NCH was identified by varying the amount used in the design process and evaluating its impact on the system. It was observed that the highest fluorescence response of the nanoprobe was obtained at 20.0 µL of MIL-101/agarose NCH. Thus 20.0 µL was selected as the optimal volume, as illustrated in Fig. 4B. The decreasing in fluorescence after this value can be related to probe self-quenching at high concentration. The time taken for the reaction of deferiprone molecules and MIL-101/agarose NCH to reach equilibrium is called the incubation time. For this purpose, MIL-101/agarose NCH was added to a solution containing deferiprone and PBS (pH 4.0) for various periods. As the incubation time increased from 1 to 20 min, the response of the nanoprobe showed a decreasing pattern (Fig. 4C). So, fluorescence intensity was recorded immediately after solution preparation. The obtained optimal conditions for executing the reactions are as follows: pH:4.0, MIL-101/agarose concentration: 20 µL, and incubation time: immediately.

Fig. 4

Effect of (A) pH, (B) amount of MIL-101/agarose NCH, and (C) incubation time on the response of the system

Interference study with coexisting substances

The established fluorescent method needs to be investigated for its specificity and selectivity to study the effect of any potential interference caused by co-administered drugs with deferiprone. To evaluate selectivity, system responses towards potential available interfering substances including losartan, dexamethasone, amoxicillin, ampicillin, alprazolam, glucose, ibuprofen, clonazepam, phenytoin, carbamazepine, phenobarbital, acetaminophen, nicotinamide, caffeine, ascorbic acid, diclofenac, chlordiazepoxide, metoprolol, dextromethorphan, cetirizine, metronidazole, aspirin, and FeCl₃ were assessed in EBC media. The system response was tested using deferiprone and a range of interfering substances at a concentration of 1.0 µg mL^− 1. As shown in Fig. 5, the interferences had a negligible effect on the probe response. Therefore, it is suggested that this method could be best performed for deferiprone tracing in the EBC of patients receiving these drugs.

Fig. 5

Study of the method selectivity under optimal conditions (pH: 4 and 20 µL of nanoprobe) in the presence of some possible interfering drugs in EBC samples with concentrations of 1 µg mL^− 1

Analytical figures of merit

The optimal conditions were employed for the partial validation of the method in accordance with the guidelines set forth by the Food and Drug Administration (FDA). First, the concentration-dependent behavior of the method was obtained in the optimum conditions. Figure 6 illustrates the fluorescence spectra of MIL-101/agarose NCH, which exhibits a peak at 410 nm upon excitation at 360 nm, and this peak intensity decreases as the deferiprone concentration is increased. The results indicated a linear relationship between the intensity and deferiprone concentration within the range of 0.005–1.5 µg mL^− 1, as shown in the inset of Fig. 6. The equation for regression was ΔF (F–F₀) = 103.96 C_DEF + 8.2813 (R² = 0.9992). In this equation, C_DEF represents the concentration of deferiprone, F₀ is the fluorescence intensity in the absence of deferiprone, and F is the fluorescence intensity at 410 nm in the presence of deferiprone. The calculated limit of detection (LOD) based on 3S_b/m (S_b: blank’s standard deviation; m: calibration slope) and limit of quantification (LOQ) based on 10 S_b/m were 0.003 µg mL^− 1 and 0.01 µg mL^− 1. The precision of the proposed method was evaluated by conducting repeat tests on the same and different days during the investigation. The (RSDs%) for 5 determinations of deferiprone (1.0 µg mL^− 1) were 0.3% and 0.4% for intra-day and inter-day measurements, respectively. The respectable repeatability and reproducibility of the method confirmed the applicability of the nanoprobe for deferiprone monitoring. In order to evaluate the accuracy of the method, known amounts of deferiprone were added to EBC samples and analyzed to determine the recovery rates. As shown in Table 1, the recovery rates for the method ranged from 98 to 104%. A statistical analysis using a t-test confirmed no statistically significant differences between the expected and measured values, thereby validating the method’s accuracy and reliability. The method’s robustness was evaluated by examining its performance under varying conditions, when transferred between laboratories or analysts. As shown in Table 2, the results indicate that the method exhibits negligible variability, thereby demonstrating its robustness and reliability. The analytical comparison of the validated method with other reported techniques in literature for the determination of deferiprone was given in Table 3. As can be seen, the current work exhibited comparable sensitivity compared with other systems.

Fig. 6

Fluorescence intensity of MIL-101/agarose NCH in the absence and presence of deferiprone in the range of 0.005–1.5 µg mL^− 1. Inset: Calibration curve obtained for deferiprone in EBC

Table 1 Determination of deferiprone in spiked EBC samples

Table 2 Robustness study for different levels of deferiprone in different conditions

Table 3 Comparison of analytical characteristics validated method with other reported techniques in the literature

Real samples analysis

To evaluate the analytical applicability and feasibility of the proposed MIL-101/agarose NCH, deferiprone determination was carried out in four EBC samples collected from patients receiving deferiprone. Table 4 presents the findings of the real sample analysis. Recovery experiments were completed by spiking to the concentration of deferiprone (0.05 and 0.5 µg mL^− 1) to confirm the accuracy of the reported values for real sample analyses and study the no interference of other co-administered drugs. The results indicate a satisfactory recovery percentage of 94.2–102.3%, demonstrating that this experiment is highly accurate and independent from matrix effects for determining deferiprone in biological matrices such as human EBC.

Table 4 Determination of deferiprone in real EBC samples by validated method

Conclusion

In this research, a simple and efficient method was used to synthesize MIL-101/agarose nanoparticles, which were then utilized as a probe for the detection of deferiprone in EBC samples. Based on the results, the fluorescence emission of the nanoprobe is gradually quenched by increasing the deferiprone concentration. The quenching in the fluorescence is mainly attributed to the chelation effect of the functional groups in deferiprone and Fe centers in the structure of the nanoprobe. The study found a linear relationship between the intensity and deferiprone concentration within the range of 0.005–1.5 µg mL⁻¹ with LOD of 0.003 µg mL⁻¹. Furthermore, the method’s precision was 0.3% and 0.4% RSDs% for intra-day and inter-day measurements, respectively. Besides, respectable recoveries of 94.2–102.3% in the EBC samples exhibited great potential in practical application. It is worth noting that the MIL-101/agarose NCH probe boasts a number of advantageous properties, including a short response time, simplicity of operation, high sensitivity, and good reliability, which render it a valuable asset for the detection of deferiprone in medical diagnostic settings. Despite the promising results, the presented method has some limitations that could impact its suitability for routine clinical use. Despite its advantages, the method’s long-term storage stability of the nanomaterials is a concern, which may affect its reliability and consistency over time. Furthermore, the EBC sampling process is inherently subject to some variability, which may impact the method’s reliability in clinical settings Additionally, biological samples often contain matrix compounds that can significantly affect the sensitivity of optical methods, making it challenging to achieve extremely low LODs, which can be a significant limitation for the practical application of these probes.