Dopamine stabilized in ultra-nanoreservoirs for controlled delivery in parkinson’s disease

ABSTRACT

Aims

Parkinson’s disease (PD) is a neurodegenerative disorder caused by the progressive degeneration of dopaminergic neurons in the substantia nigra pars compacta, leading to impaired dopamine (DA) signaling and motor control. Intermittent dosing of current DA precursors results in side effects, prompting research into controlled drug release mechanisms for sustained and targeted delivery of DA.

Materials & methods

In this work, we stabilized DA within a nanostructured silicate matrix (nanoreservoir) using the sol-gel method. We examined the physicochemical properties, kinetics of drug release, and biocompatibility in dopaminergic neurons and fibroblasts.

Results

The optimized synthesis method allowed for the stabilization of DA by preventing its oxidation. The physicochemical and controlled release analysis showed a direct relationship between the mesoporous structure, interaction of the DA with the matrix, and the release kinetics followed, proving the possibility to modify the rate of release by adjusting the synthesis parameters. Furthermore, the nanoreservoirs were biocompatible with dopaminergic neurons and fibroblasts in vitro.

Conclusions

The research sets the stage for potential in vivo evaluations and new strategies for managing PD, offering hope for improved treatments based on DA and not derivatives.

Graphical Abstract

1. Introduction

Beyond the perception of Parkinson’s disease (PD) as a movement disorder (tremor, bradykinesia, rigidity, etc.), scientific advances have made it evident that a multitude of non-motor features, such as cognitive impairment, autonomic dysfunction, sleep disturbances, depression, and insomnia, are part of the condition [1]. Despite being among the more effectively treatable neurodegenerative diseases, PD remains inexorably progressive, lacking a definitive cure and culminating in disability and mortality. The pathological underpinning of PD involves the degeneration of dopaminergic neurons in the substantia nigra pars compacta (SNpc), responsible for dopamine (DA) synthesis [2]. The consequential reduction in DA concentrations within the nigrostriatal pathway, connecting SNpc to the striatum, caudate nucleus, and putamen, results in the characteristic motor imbalances seen in PD [3,4]. Similarly, although DA is usually degraded into inactive metabolites by oxidoreductase enzymes, the neurotransmitter is susceptible to oxidation, generating quinones and free radicals as products [5], which have been linked to the neuronal loss observed in PD [6]. Furthermore, damage to the nigrostriatal pathway results in functional changes of the remaining dopaminergic neurons to preserve DA availability in the striatum [7], wherein surviving dopaminergic neurons increase DA synthesis to counterbalance the depletion resulting from the loss of their counterparts [8–10]. Based on the above, a restoration of physiological DA levels by exogenous administration could reduce the stress associated with DA overproduction by surviving neurons, provided that DA oxidation is avoided.

Addressing the challenge of restoring physiological DA levels, recent breakthroughs in nanomedicine have propelled controlled drug delivery (CDD) to the forefront of therapeutic strategies. Nanocarriers, nanostructures designed for drug transport, have emerged as key players due to their substantial surface area, sub-micron diameters, and customizable surface chemistry, enhancing drug stability and bioavailability [11,12]. For DA, polymeric nanostructures, including chitosan, poly(D, L)-lactic-co-glycolic acid, polyvinylpyrrolidone-acrylic acid, and alginates, have been explored for controlled delivery, exploiting their biodegradability and bioactivity [13]. However, concerns about the degradation processes and potential harmful residual components derived from the delivery mechanisms of polymeric nanostructures have prompted investigations into inorganic nanocarriers [14,15].

Among them, mesoporous silicate nanoparticles have garnered attention for their low toxicity, wide availability, rich functionality, good biocompatibility, targeted delivery capability, and controlled drug release properties [16]. The synthesis method used, the “sol-gel” technology, allows the obtention of nanomaterials as drug reservoirs for CDD systems, given the possibility of modifying the structure depending on synthesis parameters, allowing the immobilization of active biological compounds [17,18]. López et al. [19] achieved the first known stabilization of DA in an oxide nanocarrier: in that work, the authors optimized the synthesis parameters of the formulation to obtain a silicate capable of preventing DA oxidation.

Building on this foundation, our study delves into the physicochemical characteristics of these silicate nanocarriers, examining their relationship with DA release kinetics and cellular uptake processes. Furthermore, we investigate the biocompatibility of these nanoparticles in terms of cell survival. The correlation between nanoreservoir physicochemical qualities and DA release kinetics identified in our research lays the groundwork for defining the pharmacodynamics of these nanoreservoirs, a crucial step before their in vivo evaluation.

2. Materials and methods

2.1. Chemicals used

Dopamine hydrochloride (DA·HCl, Sigma-Aldrich), silicate precursor, tridistilled water (H₂O, Zeyco), absolute ethanol (EtOH, Sigma-Aldrich), and nitrogen gas (N₂, 99.99%, Infra) were used as received.

2.2. Synthesis of NP-DA nanoreservoirs

The nanoreservoirs were synthesized using the sol-gel method following the synthesis parameters previously standardized [20]. Three samples were prepared with DA at different concentrations (5, 10, and 15 mol%) and one sample without DA as a reference: DA concentrations were determined based on previous studies [20]. For each sample, a solution of DA·HCl in deionized H₂O was prepared in a 500 mL three-hole flask, so that this reagent had no impurities that could interfere with the proper stabilization of DA in the nanoreservoir. In an inert atmosphere at 25°C, a solution of silicate precursor was added by slow drip into an addition funnel for 4 h for sol formation. To prevent evaporation of the solvents, a coil condenser sealed with a rubber stopper was incorporated, through which a constant flow of water was passed. The solution was kept under constant stirring (400 rpm) until gelation. The same process was followed for the preparation of a fifth sample with 10 mol% of DA stabilized in silicate, in whose synthesis process the inert atmosphere was not preserved, to have a reference with the oxidized neurotransmitter (this sample was only used for comparison). Once the gel was obtained, the different samples were subjected to a vacuum drying process for 4 h in a Rotavapor® R-100 (Büchi Labortechnik) incorporated with a heating bath. The flasks of each sample were placed in a water bath with a constant rotation of 150 rpm. A constant flow of water was passed through the recirculating cooler to facilitate condensation of the solvents. Once the drying was completed and the solvents evaporated, the nanoreservoirs went through a fine maceration process in Agate mortar until fine powders were obtained. From each sample, 15 g of NPs were synthesized. The powdered samples were subjected to a pressure packing process (7 tons in a hydraulic press, Carver Inc.) to obtain pellets of 1 cm in diameter and 1 mm thick.

2.3. Nanoreservoirs characterization

A full operating description of each of the physicochemical techniques carried out is offered in the Supplementary Information.

2.4. Release kinetics determination

The DA release rate by our nanoreservoirs was measured by colorimetry at 280 nm (DA signal) using a NanoDrop 2000 UV-Vis spectrophotometer (Thermo Scientific). Twenty standard solutions of DA were prepared in triplicate (0.1 to 1 × 10⁻³ mol/L) in a range of 10–100% of the theoretical amount stated as 100% (1 × 10⁻³ mol/L) to construct a calibration curve and subsequently read the samples. The release was carried out by weighing 200 mg of each of the samples (5, 10, and 15 mol%) in powder form, which were packed to form 1-mm-thick tablets by applying 7 tons of pressure. The tablets were placed in deionized water and 2 μL aliquots were taken for each reading over a period between 0 and 100 h, in triplicate. For the first hour, release was assessed every 5 minutes. Then, for the second hour, every 10 minutes. Between the 3rd and 48th h, measurements were taken every 30 minutes and then every hour thereafter. A kinetic analysis was performed by fitting several semi-empirical models into the release curve: zero-order, first-order, Higuchi, Hixson-Crowell, and Korsmeyer-Peppas. R² closest to 1 was determined as the best fit. Once a model was selected from the other, an adjustment was made according to the mass quantity of DA released, so that the necessary parameters could be obtained to substitute in the model equation and thus be able to predict the long-term release behavior. For this purpose, the values of the variables were optimized using the “Solver” function of Excel (version 2204), so that the sum of the standard deviations between the real data and the adjustment would be the minimum. Iterations were performed until the fit yielded an R² >0.98.

2.5. Qualitative cytotoxicity evaluation

For qualitative analysis (impact in morphology), the human neuroblastoma SH-SY5Y cell line (ATCC, Manassas USA) donated by José Víctor Segovia Vila (CINVESTAV) [21], was maintained in Dulbecco’s modified Eagle medium (DMEM) high glucose (Gibco, USA) supplemented with 10% heat-inactivated fetal bovine serum (FBS), l-glutamine (2 mm), penicillin (100 U/ml), and streptomycin (100 μg/ml) under an atmosphere of 5% CO₂, in a humidified incubator at 37°C. Cells were maintained in the proliferation phase for 5 days before the differentiation protocol was applied. For dopaminergic differentiation, a standardized protocol was followed [22,23]. In brief, the medium was changed, and DMEM supplemented with retinoic acid (10 μM) was added to the cells. These cells were then maintained in the new medium for 3 days. Differentiation was corroborated by the morphological appearance of the cells as indicated in the protocol. The studies were conducted at a cell density of 100,000 cells per well. The dissolutions of nanoreservoirs were prepared with DMSO (90 mL) and kept under refrigeration. The culture medium in the 24-well cell culture multi-well plates was removed with a Pasteur pipette connected to a Büchner flask with a vacuum. The bottom of the plates was not touched as cells had adhered to it. The wells were rinsed with 200 μL of PBS and extracted with the Pasteur pipette. The cells were exposed to the nanoreservoirs [NP-DA-0%, NP-DA-10%, and NP-DA-Ox] at concentrations of 5, 10, 25, 50, 100, and 250, and 500 μg/mL with DMSO for 24 h at 37°C in a 5% CO₂ atmosphere. The study was carried out by triplicates. The control condition only contained 200 μL of the supplemented DMSO used to dilute the nanoreservoirs.

2.6. Quantitative cytotoxicity evaluation

On the other hand, for quantitative analysis, we opted for a different cell line to extend the in vitro viability study. Mouse embryonic fibroblasts of stage 14.5 days post coitus were used as reported in previous works [24], given its easy handling and high representation in various endothelial tissues, which makes it an excellent ex vivo model that is widely used [25]. Briefly, the fibroblasts were seeded in 24-well boxes previously treated with 0.1% porcine skin gelatin at a density of 1 × 10⁷ cells per well. M10 medium (DMEM 10% SFB, GPS 1×, sodium pyruvate 1×, non-essential amino acids 1× and 180 µL/100 ml of β-Mercaptoethanol) was used, 12 h later the medium was aspirated, and fresh medium previously tempered to 37°C was added. Prior to the beginning of the treatment, the medium was recovered from each well. Fresh medium was placed, and the nanoreservoirs were applied at 50, 100, and 200 μg/mL and incubated for 24 h. Once the incubation with the treatment was finished, cells adhered to the plate were recovered by washing each twice with 1X PBS and then 100 μL of 0.25% trypsin previously tempered was added and incubated for 7 min at 37°C. Trypsin was inactivated with 500 μL of M10 medium and the cells were centrifuged at 1000 rpm for 5 min, resuspended in 500 μL of fresh medium and the number of live and dead cells was calculated with trypan blue.

2.7. Data analysis

To determine whether significant statistical differences were present both in the triplicate assays for each sample and between the samples, the data were analyzed by ANOVA with a 95% confidence interval (p < 0.05). Tukey’s range test was utilized as a post hoc when significant differences were identified by the ANOVA analysis.

3. Results

DA nanoreservoirs with different basal DA concentrations (NP-DA-0%, NP-DA-5%, NP-DA-10%, and NP-DA-15%) and one intentionally oxidized sample (NP-DA-Ox) were obtained following the sol-gel method, as shown in the schematic in Supplementary Figure S1 and the detailed description incorporated in the Experimental Section. Drip incorporation of the alkoxide allows individual molecules to hydrolyze into orthosilicic acid, a transient state with a coordination number of 5 that rapidly undergoes an alkoxylation process to form the Si-O-Si and Si-OH bridges. Drugs and other compounds added in the initial solution act as nucleation centers for the formation of the silicate network around them through stabilization by hydrogen bonds. Supplementary Figures S1c and S2 show the importance of the synthesis process followed for preserving DA in its basal state. As described above, DA is oxidized in the presence of oxygen, generating highly reactive species (dopaminechrome, dopamine quinone, H₂O₂, OH•, O), which have been linked to apoptosis in dopaminergic neurons [26]. All DA oxidation derivatives show a significant color change (gray or brown) from the characteristic bright white of DA; thus, the color of the sample was the first indicator of the state of the DA (basal and/or oxidized state) in silicate (also with a characteristic white color), both during the synthesis process (Supplementary Figure S2a) and as a powder once the solvents were evaporated (Supplementary Figure S2b). Samples were kept in normal atmospheric conditions of temperature and humidity, and sample color has been preserved for 36 months since the synthesis, suggesting the conservation of DA in its basal state at least for this time. To corroborate the state of stabilized DA, a full physicochemical characterization was carried out. The analyses were divided into five groups according to the physicochemical features to analyze: (i) particle size and structure, (ii) atomic composition, (iii) superficial properties, (iv) DA stabilization, and (v) thermal stability.

3.1. Particle size and crystalline structure

In Figure 1(a), a panoramic view in scanning electron microscopy (SEM) reveals the formation of conglomerates (<100 μm) comprising nanoreservoirs with 10 mol% DA. These conglomerates exhibit elongated structures resembling femurs, as depicted in Figures 1(b,c). A more detailed examination reveals the existence of smaller spherical conglomerates ranging from 190 to 380 nm in size, as depicted in Figure 1(d). The granular topology of these smaller conglomerates strongly implies the presence of individual particles at a smaller scale, as further supported by transmission electron microscopy (TEM) analysis.

Figure 1.

SEM (a–d) images of NP-DA-10% nanoreservoirs at magnifications (a) x100, (b) x5,000, (c) x25,000, and (d) x40,000. (e) HR-tem image of NP-DA-10% sample and (f) amplification with particle sizes delimited with red arrows. (g) SAED pattern of nanoreservoir depicting the amorphous formation of crystalline phases. (h) X-ray diffraction patterns of samples reveal an amorphous structure for the sample. SED: secondary electron detection.

In Figures 1(e,f), brightfield TEM micrographs of the NP-DA-10% sample (with additional samples displayed in Supplementary Figure S3) are presented. These micrographs reveal that the nanoreservoirs consist of individual amorphous nanoparticles spanning sizes within the range of 7 to 50 nm. These nanoparticles cluster together to form larger amorphous aggregates, resembling the structures observed in SEM micrographs. Interestingly, there is no notable disparity among the samples, irrespective of the concentration of stabilized DA, as illustrated in Supplementary Figure S3. This observation suggests that the presence of DA does not exert a significant influence on particle size or morphology during the synthesis process. Neither the concentration of DA nor its oxidation state significantly alters these parameters. However, it is worth noting that the coloration in the TEM micrograph appears to intensify with increasing DA concentration. This visual effect, associated with sample density, implies that samples with higher DA concentration exhibit greater densities compared to the reference sample NP-DA-0%, as anticipated.

Selected area electron diffraction (SAED) analysis carried out through TEM indicated the amorphous distribution of the atoms in the nanoreservoir’s network (Figure 1(g)). The above was corroborated by X-ray diffraction (XRD). Figure 1(h) shows the presence of a weak peak around 2θ = 23°: this signal indicates that all nanoreservoirs present an amorphous structure without any degree of crystallinity. That is expected for nanostructured silica below 823 K [27]. Notably, stabilized DA did not appreciably alter the lattice ordering. Amorphous silicate is associated with the presence of point defects (oxygen vacancies, free electrons, etc. [28]) which, in the case of CDD nanocarriers, translates into an optimization of their stabilization and release properties [29].

3.2. Atomic composition

For the detailed analysis and comparison of the surface chemical state, composition, and electron interaction between the three main samples (NP-DA-0%, NP-DA-10%, and NP-DA-Ox), X-ray photoelectron spectroscopy (XPS) measurements were carried out, in which signal assignation and atomic concentration (Table 1, Supplementary Figure S4) were calculated. Supplementary Figure S4 clearly indicates the presence of Si, O, C, N, and Cl (the latter two only observed in NP-DA-10% and NP-DA-15%). The matrix of the NP-DA-0% sample is composed exclusively of Si, O, and C, as expected (Supplementary Figures S4e, S4j, S4o). The signal deconvolution for Si 2p (Supplementary Figure S4e) indicates the presence of different oxidation states: Si⁺ (suboxides SiO_x), Si³⁺ (Si-O-Si), and Si⁴⁺ (silicon dioxide, Silicate) [30–32], the latter is the preponderant one in the nanoreservoir matrix. No signals were observed for Si⁰ or Si²⁺, indicating that all Si in the nanoreservoir matrix is oxidized and that no organic remnants are directly bound to Si, respectively [30,33]. For the NP-DA-10% sample (Supplementary Figure S4g), signals for Si⁺ and Si³⁺ showed a chemical shift with respect to their counterparts in the sample without DA, suggesting an interference of the molecule on the matrix, as well as an effect on the reorganization of the electronic structure of Si. Finally, in contrast to the above, the oxidized sample (Supplementary Figure S4i) exhibited an intense signal associated with the Si²⁺ oxidation state (silicon monoxide, SiO) [34]. The absence of this signal for the rest of the samples suggests that oxidized DA exerts a significant effect during the hydrolysis and alkoxylation processes, resulting in the appearance of this oxidized species.

Table 1.

Atomic percentage of the samples by deconvolution of atomic signals and their components in different oxidation states according to XPS analysis.

		Atomic percentage (%)
Component		NP-DA-0%	NP-DA-5%	NP-DA-10%	NP-DA-15%	NP-DA-Ox
Si 2p		31.51	28.75	24.63	18.06	27.81
	Si0	0	0	0	0	0
	Si1+	3.01	0	13.58	0.04	4.70
	Si2+	0	0	0	1.49	15.35
	Si3+	10.67	23.59	6.57	5.69	7.12
	Si4+	17.83	5.16	4.48	10.84	0.64
O 1s		65.50	61.00	52.09	43.37	57.70
	O−	9.52	0	25.36	5.73	0
	Si-O	55.98	36.95	13.34	30.72	41.10
	Si-O-H	0	20.58	11.29	6.63	15.01
	O-H	0	3.47	2.10	0	1.54
C 1s		2.99	9.49	20.64	32.91	13.38
	C-Csp2	0	5.65	7.29	3.66	3.66
	C-Csp3	2.77	2.59	6.72	5.05	5.05
	C-O	0	1.25	4.93	3.08	3.08
	C=O	0.22	0	1.71	1.59	1.59
N 1s		0	0.76	1.76	4.05	1.11
	N-Si3	0	0	0.55	1.62	0
	C-N+	0	0.43	0.86	1.48	0.43
	C-NH2	0	0.33	0.35	0.94	0.68
Cl 2p		0	0	0.87	1.61	0
	Cl−	0	0	0.87	1.61	0

^‡Sample not characterized.

*Note: For sample NP-DA-0%, the total concentration of O bonded to Si according to the deconvolution for Si 2p, yields a total contribution of 53.16%, while the percentage calculated from the deconvolution for O 1s shows a total of 55.98%. The percentage error between the estimated and actual value is 5.03%, indicating a good calculation in the Si 2p deconvolutions. Similarly, the atomic percentages for the bonds associated with DA increased proportionally to the stabilized DA concentration, which corroborates the correct assignment and, above all, the correct stabilization of DA at the theoretical concentrations.

In the O 1s region, the NP-DA-0% sample (Supplementary Figure S4j) exhibits two signals: one attributed to the Si-O bonding [35], and the other is attributed to oxygen vacancies [36]. Sample with 10 mol% DA (Supplementary Figure S4l) also exhibits, apart from the Si-O signal, two other signals. The first one is attributed to surface silanol groups [34,37], which constitute 20.58% of the atomic total of the sample. The presence of this signal for the nanoreservoir with respect to the reference suggests a relationship between the increase of Si-O-H groups and the presence of DA. On the other hand, the second signal corresponds to chemisorbed OH groups resulting from small concentrations of water contamination [38]. Notably, unlike the oxidized sample (Supplementary Figure S4n), the NP-DA-10% sample exhibits the oxygen vacancy contribution.

Regarding the C 1s region, Supplementary Figure S4o shows the signals for the bonds C-C and C=O in NP-DA-0% [30]. The first signal can be attributed to organic remnants generated during the hydrolysis process of the precursor alkoxide in the synthesis of the nanoreservoir, while the second signal corresponds to ambient CO₂ deposited on the surface of the nanoreservoir. Notably, unlike the reference, in samples NP-DA-10% and NP-DA-Ox (Supplementary Figures S4q, S4s) it is observed that the adventitious carbon signal exhibits additional components. Firstly, the C-C signal is divided into two:one for the sp² hybridization state (attributable to the aromatic ring of the DA), and the other for the sp³ (attributable to the bonds of the aliphatic chain of the DA) [35]. The other additional signal corresponds to the C-O bond which is associated with the OH groups bonded to carbons 1 and 2 of the benzenediol ring of DA [39].

In both samples doped with DA, the presence of the neurotransmitter is confirmed due to the presence of nitrogen (N) (Supplementary Figures S4v, S4×), together with the triad Si, O and C, which corresponds to the amine group (NH₂) of DA. In the sample with 10 mol% DA, two signals are appreciated corresponding to the C-N⁺ and the -NH₂ bonds [35]. Both signals indicate the presence of an amino functional group, such as the one found in the DA molecule. Furthermore, a third signal is observed only in the oxidized sample at 398.43 eV, which is attributed to an N-Si₃ compound [40]. Although the proportion of this signal is quite small (0.55%), its presence indicates the existence of two possible situations: (i) a covalent bond between the silicate matrix and a certain percentage of DA molecules or (ii) a covalent bond between the silicate matrix and the nitrogen used as an inert gas to inhibit DA oxidation. Given that such signal was not observed in the reference (inert atmosphere) nor the oxidized sample (non-inert atmosphere), the first hypothesis could be the most plausible. Finally, although only observed in the non-oxidized sample (Supplementary Figures S4aa), a signal for Cl⁻ was observed, indicating ionic remnants of the DA precursor used for the synthesis.

3.3. Superficial properties

Having identified the elemental composition and stoichiometry of the nanoreservoirs, the functional groups were determined as a function of their vibrational interactions through the FTIR spectroscopy (Figure 2(a-c)). The identification of functional groups by signal is shown in Table 2.

Figure 2.

(a–c) IR spectra of the samples showing the main signals associated with functional groups: (a) region 2750–2350 cm⁻¹, (b) region 1600–1250 cm⁻¹ and (c) region 1050–500 cm⁻¹. (d-i) Superficial properties of the samples. (d) Zeta potential distribution with surface charge (mV) and conductivity (mS/m). (e) Pyridine adsorption FTIR spectra of all the samples with identification of Lewis and acid sites, (f) pyridine adsorption FTIR spectra of NP-DA-0% sample at different temperatures, and (g) quantification of acid sites concentration per mass for the NP-DA-0% sample at different temperatures. (h) Nitrogen adsorption-desorption isotherms and (i) BET specific surface area (SSA) and pore size per sample.

Table 2.

IR signals and vibration assignment.

Wavenumber (cm−1)						Vibration assignment
DA	NP-DA-0%	NP-DA-5%	NP-DA-10%	NP-DA-15%	NP-DA-Ox
–	3732(w)	3732(w)	3732(w)	3732(w)	3732(w)	νs(O-H), superficial
2779(w)	–	–	2733(w)	2733(w)	–	νas(C-H), CH2
2661(w)	–	2661(w)	2661(w)	2661(w)	2661(w)	νas(C-H), CH2
2580(sh)	–	–	2580(sh)	2580(sh)	–	νas(C-H), CH2
2571(w)	–	–	2567(w)	2567(w)	2567(w)	νs(N-H), NH2
2509(w)	–	–	2509(w)	2509(w)	2509(w)	νs(N-H), NH2
2484(w)	–	–	2484(w)	2484(w)	–	νs(N-H), NH2
2457(sh)	–	–	2457(sh)	2457(sh)	2457(sh)	νs(N-H), NH2
2409(sh)	–	–	2409(sh)	2409(sh)	2409(sh)	νs(C-H), alkane
2382(w)	–	–	2382(w)	2382(w)	2382(w)	νs(C-H), alkane
–	1849(w)	1849(w)	1849(w)	1849(w)	1849(w)	δs(C-H)
1581(w)	–	1581(w)	1581(w)	1581(w)	1581(w)	νas(C-C), aromatic
1562(sh)	–	1562(sh)	1562(sh)	1562(sh)	1562(sh)	νas(C-C), aromatic
1560(s)	–	–	1560(s)	1560(s)	1560(s)	νas(C-C), aromatic
1472(w)	–	–	1472(w)	1472(w)	1472(w)	νs(C-C), aromatic
1421(sh)	–	1421(w)	1421(w)	1421(w)	1421(w)	νs(C-C), aromatic
1408(s)	–	–	–	–	–	νs(C-C), aromatic
1377(s)	–	–	–	–	–	δas(C-O-H)
1332(w)	–	–	1332(w)	1332(w)	–	δas(C-O-H)
1307(s)	–	–	1307(w)	1307(w)	–	δs(O-H), phenol
–	–	1276(w)	1276(w)	1276(w)	1276(w)	δs(C-O-H)
989(w)	989(s)	986(s)	983(s)	983(s)	981(s)	νas(Si-O), O-Si-O
898(s)	–	898(sh)	898(sh)	898(sh)	898(sh)	δs(C-H)
–	860(s)	854(s)	850(s)	850(s)	854(s)	νs(Si-O)
806(s)	–	806(w)	806(w)	806(w)	806(w)	δas(C-C-C)
763(s)	–	–	761(w)	761(w)	761(w)	δas(C-C-C)
661(w)	661(w)	661(w)	661(w)	661(w)	661(w)	δs(C-C-C) and/or δs(Si-O),-O-Si-O−
594(w)	–	–	594(w)	594(w)	594(w)	δs(C-C-C)
–	514(w)	514(w)	514(w)	514(w)	514(w)	δs(Si-O), Si-O−

s = strong, w = weak, sh = shoulder, ν_s = symmetrical elongation, ν_as = asymmetrical elongation, δ_s = symmetrical bending, δ_as = asymmetrical bending.

The first signal identified at 3732 cm⁻¹ (Table 2) corresponds to a stretching (ν) of the O-H bond present in the hydroxyl groups of the surface of the silicate matrices [41], necessary for hydrogen bond formation with biological structures. In the region between 2750 and 2350 cm⁻¹ (Figure 2(a)), a series of signals corresponding to DA side chain C-H₂ and the amino group are present [42], these are observed mainly for the sample doped with 10 and 15 mol% DA, mainly, and to a lesser extent in that doped with 5 mol% DA and in the oxidized sample. In the 1600–1250 cm⁻¹ region (Figure 2(b)), several signals are present, which correspond to different vibrational modes of the aromatic C-C and phenol C-O-H bonds in the benzene ring of DA [42,43]. To conclude this region of the spectrum, the peak at 1276 cm⁻¹ must be highlighted: although characteristic of the symmetric δ C-O-H bond in DA [42], the signal only appears in the doped nanoreservoirs, which suggests a possible relationship between DA and the silicate matrix that leads to the generation of the signal. Notably, as expected, all of these signals are absent in the reference sample due to the lack of the neurotransmitter. Finally, the last region of the spectrum (Figure 2(c)) corresponds to the characteristic signals of silicates, such as the O-Si-O bridge (989 cm⁻¹) [44], the Si-O bond (860 cm⁻¹) [44], the oxygen vacancy ⁻O-Si-O⁻ bond (661 cm⁻¹) [41], and the Si-O⁻ bond (514 cm⁻¹) [45].

It is significant to highlight that the doped nanoreservoirs present redshifts proportional to the DA concentration, suggesting an impact of their stabilization on this bond. The rest of the signals correspond to the C-C-C bond in the DA side chain [42].

The zeta potential was calculated for the surface electrical properties. Previous work has shown that an anionic profile (negative charges) is necessary for the optimization of drug stabilization properties [45]. Figure 2(d) shows the zeta potential distributions for each of the samples analyzed, as well as the values of zeta potential and conductivity obtained. All zeta potential values are above the aggregation threshold (−10 mV), indicating low aggregation of individual NPs [46]. Similarly, negative charges between −15 and −21 mV correspond to intermediate anionic profiles [47]. An increase of at least one order of magnitude in the conductivity of the nanoreservoirs was observed as the DA concentration increased with respect to the reference sample. This phenomenon has been observed in other nanostructures functionalized with polydopamine [48]. In contrast, the oxidized sample showed a milder increase in conductivity, which could be due to some interaction of oxidative derivatives of the neurotransmitter that could interfere with the conductivity.

Pyridine adsorption analysis by FTIR was used to distinguish the presence of Lewis and/or Brønsted acid sites in the samples [49,50]. Figure 2(f) shows the FTIR spectra with pyridine adsorption for all samples at 50°C. For NP-DA-0%, pyridine adsorption resulted in the appearance of the band characteristic of pyridine coordinated with Lewis’s acid sites at 1450 cm⁻¹ and the band corresponding to pyridine adsorbed on Lewis and Brønsted acid sites at 1490 cm⁻¹ [51]: No Brønsted acid sites (1541 cm⁻¹) were observed. Notably, no acid sites were observed for the DA-doped samples, suggesting that the molecule is stabilized at these sites. The above could be due to the nature of the aromatic ring of DA, which is conserved in pyridine. Such chemical structure could endow the neurotransmitter to be stabilized on the acidic sites of the nanoreservoirs, thus preventing its characteristic signals from being seen. Notably, the acid sites in the NP-DA-0% sample disappear from 100°C (Figures 2(g,h)).

Finally, to evaluate the porosity of the samples, a Brunaeur-Emmett-Teller (BET) analysis was applied. Figure 2(i) shows the nitrogen adsorption-desorption isotherms for the samples NP-DA-0%, NP-DA-10% and NP-DA-Ox; likewise, Figure 5(f) shows the textural parameters of specific surface area (SSA) and pore size. All the isotherms present a type IV hysteresis loop, according to the IUPAC classification [52]. This type of isotherm presents an initial behavior of monolayer formation, followed by multilayer formation until reaching a maximum multilayer thickness at maximum pressure P₀ [53]. The NP-DA-10% sample exhibits the slowest slope, indicating a large heterogeneity in pore size. Interestingly, this sample exhibited higher specific surface areas and smaller average pore diameters than the reference and the oxidized samples. Regarding hysteresis, the NP-DA-0% and NP-DA-Ox samples present hysteresis of the H1 type, which is associated with almost uniformly spherical particle agglomerates with very narrow pore distributions [54]. In contrast, the isotherm for NP-DA-10% exhibits H2-type hysteresis, which corresponds to layered sheets or ink-bottle-shaped pores [52].

Figure 5.

Cytotoxicity evaluation of nanoreservoirs NP-DA-0% and NP-DA-10% in the SH-SY5Y cell line showing morphological change progression at (a) 1 h, (b) 6 h, and (c) 24 h incubation. (d) Quantification of cell viability in fibroblasts at 24 h determined by trypan blue assay. No significant difference in cell viability was observed between cultures treated with NP-DA-0% (F [3,8] = 1.08144, p = 0.4104) and NP-DA-10% (F [3,8] = 1.65803, p = 0.2521).

3.4. Dopamine stabilization

Knowing the presence of stabilized DA in the silicate network, it is necessary to corroborate the state in which the neurotransmitter is found. As previously mentioned, the first control point to evaluate whether DA was oxidized during the synthesis process is the coloration of the nanoreservoirs obtained after the drying process. However, for a better corroboration of the above, UV-Vis spectroscopy was performed. The spectrum obtained for the nanoreservoirs is shown in Figure 3(a). All the spectra of the DA-doped samples exhibited two clear signals in the mid-ultraviolet, one at 230 nm and the other at 280 nm. The signal at 280 nm is characteristic of the π(π* transition observed in the dihydroxyphenyl group of DA [55]. As expected, no signals were observed in the visible region of the spectrum [56]. After this signal, for the case of the oxidized sample, a region is found in the 300–400 nm interval with not well-defined bands corresponding to the different products of DA autooxidation: dihydroxyindole (299 nm), aminochrome (300 nm) and dopamine o-quinone (370 nm) [57]. The absence of these signals for the rest of the samples, as well as the difference in intensities for the characteristic DA signal (280 nm), corroborate the stabilization of oxidized DA for the NP-DA-Ox sample, as well as the stabilization of DA in its basal state for the rest of the samples. It is interesting to highlight that the signal observed at 230 nm does not correspond to any known signal for DA, so that it could indicate a charge transfer between the p-orbitals of the oxygens of the silicate matrix with some of the s-orbitals in the hydroxyl groups of the DA; notably, there is a tendency for the signal intensity to decrease as the DA concentration increases. In contrast, the NP-DA-0% showed two signals quite similar to those observed for the samples with DA, one at 220 nm and the other at 280 nm. The first signal corresponds to paramagnetic defects, such as (Si•Si) (positively charged paramagnetic oxygen vacancies), Si (neutral dangling silicon bond) centers, oxygen vacancies, and free electrons [58–60]. On the other hand, the signal located at 284 nm corresponds to the absorption peak of bulk silicate materials [61].

Figure 3.

(a) uv-vis spectra of samples doped with DA compared to the reference. (b-d) NMR spectra of the samples NP-DA-0%, NP-DA-10%, and NP-DA-Ox. (b) NMR MAS for ¹³C core, (c) NMR MAS for ²⁹Si core and (d) NMR CP/MAS for ²⁹Si core. (e) Descriptive diagram of the species in the spectra for ²⁹Si core.

The type of interaction between DA and the nanoreservoirs matrix was measured by Nuclear Magnetic Resonance (NMR) to interpret the type of release the nanostructure will follow. The spectra obtained are shown in Figure 3(b–d). The ¹³C MAS-NMR spectrum of DA (Figure 3(b)) shows seven signals associated with the different types of carbon present in the molecule. Signals for ¹³C MAS NMR of DA were observed, as expected, in the samples doped with the neurotransmitter (Figure 3c), thus confirming its stabilization at room temperature and pressure. Similarly, in the oxidized sample a signal was identified at 203.90 ppm, which can be associated with carbonyl carbons in quinones, corroborating the presence of the oxidized DA [62]. The rest of the signals observed, both in the samples with DA and in the reference silicate matrix correspond to aliphatic carbons remaining on the surface of the nanoreservoirs, mainly ethyl products of the hydroxylation of the silicate precursor during the hydrolysis of the sol-gel method [63].

Regarding the silicate matrix, the ²⁹Si MAS NMR spectra were evaluated to identify the degree of condensation and connectivity in the nanoreservoirs (Figure 3(d)). The silicon (Si) sites were labeled according to the usual NMR spectroscopy notation (Figure 3(f)): Qⁿ represents the connectivity of Si atoms bonded to quaternary oxygen, i.e., bonded to (4-n) OH groups [64], such that the higher the value of n, the greater the connectivity in the matrix, i.e., the greater the condensation [65]. Chemical shifts (δ) around −83 ppm, −93 ppm, and −102 ppm were assigned to the hydrated silicates Q¹ (Si(OSi)(OH)₃), germinated silanols Q² (Si(OSi)₂(OH)₂) and isolated silanols Q³ (Si(OSi)₃)(OH)), respectively [66]. This assignment is supported by the ²⁹Si CP/MAS NMR spectra (Figure 3(e)), where the resonances of hydroxyl-bearing species (in particular, Q³) appear stronger than in the MAS spectrum. This is because, in this technique, the dilute ²⁹Si spins are polarized during contact time due to the abundant presence of ¹H spins, improving the NMR sensitivity [67,68]. The ²⁹Si MAS NMR spectrum (Figure 3(d)) shows that the Q² structure is the dominant Q species in all samples, with the presence of Q¹ and Q³ signals, with no detectable Q⁴ signals (siloxane bridges, (Si(OSi)₄)), indicating a moderately condensed structure [65,69]. In the sample with basal DA, a significant difference in Q²/Q³ ratio is observed with respect to the ratios of the reference and the sample with oxidized DA. This phenomenon could be due to a higher formation of hydroxyl groups during the condensation step of the sol-gel method due to the presence of dipoles induced by the DA molecule.

In addition to the above, it is important to highlight the absence of signals in the samples with DA in the region δ = [−70.0, −50.0] ppm of the ²⁹Si CP/MAS NMR spectrum. According to the literature [70], in this region the Si sites T⁰, T¹, T² and T³ take place, in which n corresponds to the number of -Si-O-R bonds linked to the Si site at Tⁿ. This lack of signals indicates that the DA molecule was not stabilized by covalent bonds, thus suggesting that the interaction of DA with silicate is of a weak type.

3.5. Thermal stability

The integrated thermogravimetric analysis/differential scanning calorimetry (TGA/DSC) technique was used to verify the thermal stability of DA in nanostructured sol-gel silicate in atmospheric conditions (Supplementary Figure S5a, S5c, S5e, S5g, S5i). All samples exhibited a mass loss caused by endothermic evaporation of the water and ethanol used during the sol-gel synthesis, and evaporation of the water absorbed on the silicate surface [71]. DA oxidation was observed from 150°C to 350°C, as well as its subsequent decomposition (~240°C). Interestingly, this exothermic process was similar in the 5 mol% and oxidized samples, while the 10 mol% and 15 mol% samples exhibited higher mass losses. The similarity between the mass losses of the oxidized sample and the sample with DA at 5 mol% suggests that the concentration of oxidized DA in the NP-DA-Ox sample corresponds to ~5 mol%. Sample with 15 mol%, on the other hand, presented two mass losses (5.67% and 12.10%), indicating a rearrangement in the matrix structure, possibly due to the disintegration of DA. Losses after ~400°C are due to the removal of organic residues (carbon dioxide, amorphous carbon, etc.) [72]. The total mass loss for the four samples was 32.44% (NP-DA-0%), 31.13% (NP-DA-5%), 37.54% (NP-DA-10%), 38.06% (NP-DA-15%) and 32.39% (NP-DA-Ox). Finally, Supplementary Figures S5b, S5d, S5f, S5h, S5j shows the thermograms with inert atmosphere (nitrogen). The behavior was quite similar to that observed for the samples with air.

3.6. In vitro drug delivery kinetics determination

The neurotransmitter release kinetics were evaluated by introducing the nanoreservoirs into deionized water [45]. Before the assay, a calibration curve was established to convert UV-Vis absorbance at 280 nm into concentration values (Supplementary Figure S6). Various semi-empirical release kinetic models were used to identify the one that most accurately characterized and predicted the release of DA. Figure 4 showcases the release curves of DA over a 100-h evaluation period, with Figures 4(a–e) representing linear fits for the various models. Supplementary Table S1 summarizes the parameters obtained for each adjustment. The assay indicated a clear trend in the release rate as a function of DA concentration, releasing more abruptly for higher concentrations, and retaining slower releases for lower concentrations. Remarkably, the Korsmeyer-Peppas model emerged as the most suitable model for describing DA release in all nanoreservoirs: this model aptly represents drug release in polymeric structures, such as nanostructured mesoporous silicate. Next, an adjustment was made according to the mass quantity of DA released, so that the necessary parameters could be obtained to substitute in the Korsmeyer-Peppas equation and thus be able to predict the long-term release behavior. The fitted parameters and the substitution in the Korsmeyer-Peppas equation are presented in Table 3. Similarly, Figure 4(f) shows the DA release curve as a function of mass released with the corresponding Korsmeyer-Peppas fit in dotted line. Notably, all nanoreservoirs yielded a value of n < 0.5 for their Korsmeyer-Peppas equation (F = K_mtⁿ) (Table 3), indicating that the solvent diffusion is faster than the relaxation process of the polymer chain, so the kinetics of this phenomenon is characterized by diffusivity [73,74], particularly, a Fick diffusion [75,76]. Notably, Fick diffusion depends on the pore size of the matrix [76]. In this regard, manipulation of the pore size distribution would allow the manipulation of the release rate that the nanoreservoir will have. With the above, the nanoreservoir can be custom synthesized by modifying the water/alkoxide ratio and the pH of the reaction to obtain desired release kinetics (Supplementary Figure S7) [77]. In future work, further analysis will be presented regarding release kinetics in vivo with Wistar rats and PD models.

Figure 4.

Controlled release in distilled water of DA stabilized in nanoreservoirs over a period of 100 h and fit by different controlled drug release models: (a) fit to zero-order model, (b) Fit to first-order model, (c) fit to higuchi model, (d) fit to hixson-crowell model and (e) Fit to korsmeyer-peppas model. In (f) The korsmeyer-peppas model is shown for DA mass release as a function of time. R² closer to 1 indicates best fit.

Table 3.

Adjustment to release as a function of DA mass quantity with Korsmeyer-Peppas model and substitution of parameters in the equation.

Sample	Km	n	Equation with parameters (F = Km tn)	R2
NP-DA-5%	10.8715	0.0318	F = 10.8715 t0.0318	0.9801
NP-DA-10%	24.6782	0.0233	F = 24.6782 t0.0233	0.9981
NP-DA-15%	39.5288	0.0082	F = 39.5288 t0.0082	0.9994
NP-DA-Ox	17.5328	0.0151	F = 17.5328 t0.0151	0.9984

F = Release mass of DA [mg]; Km = Release constant [mg/h]; t = time [h].

3.7. Cytotoxicity evaluation in cell culture

The potential cytotoxicity of the nanoreservoirs was evaluated in cultures of SH-SY5Y differentiated to dopaminergic neurons and fibroblasts, with the nanoreservoirs dissolved in DMSO. Figures 5(a-c)(a-c) illustrate the temporal morphological changes of SH-SY5Y dopaminergic cells following the application of NP-DA-0% and NP-DA-10% nanoreservoirs, as well as a DMSO control, at concentrations ranging from 50 to 250 μg/mL (additional concentrations, as well as morphological evaluation of sample NP-DA-Ox, are shown in Supplementary Figure S8). The initial assessment was conducted within 1 h of treatment application. At all tested concentrations, the cultured cells exhibited morphologies comparable to their respective controls: neuronal-like shapes with non-polarized bodies and a few processes [78]. The cells remained adhered to the well base with high confluence. Expectedly, cell death was observed in H₂O₂ controls (not shown), where cells appeared spherical with apoptotic bodies.

Further evaluation at 6- and 24-h post-treatment revealed cell morphologies were maintained at all concentrations up to 250 μg/mL. A trypan blue quantification assay conducted in fibroblasts after 24 h of exposure (Figure 5(d)) confirmed no significant difference in cell viability between cultures treated with NP-DA-0% (F [3,8] = 1.08144, p = 0.4104) and NP-DA-10% (F [3,8] = 1.65803, p = 0.2521) across all measured concentrations, as determined by a one-way ANOVA study. However, although not significant, a decreasing trend in cell viability associated with increasing nanoreservoir concentration is noticeable, especially in the sample with 10 mol% DA. These results indicate that the nanoreservoirs are biocompatible with fibroblasts and dopaminergic cells at concentrations up to 250 μg/mL. However, it is noteworthy that at 500 μg/mL, a morphological change in dopaminergic neurons began to appear after 6 h (Supplementary Figure S8b), with cells becoming more spherical, suggesting a cytotoxic effect at this concentration. This effect was more pronounced at 24 h, where only non-viable cells were observed at 500 μg/mL, similar to the H₂O₂ control. Notably, at 24 h, treatment with NP-DA-Ox induced spherical morphology in culture at concentrations as low as 5 μg/mL (Supplementary Figure S8c), demonstrating the intrinsic toxicity of oxidized DA derivatives.

The evaluation of dopaminergic cells and fibroblasts cell viability in the presence of various concentrations of the nanoreservoirs confirmed their biocompatibility, as no significant cell death occurred at concentrations up to 250 μg/mL (a high concentration in comparison to other studies involving silicate nanostructures in the brain). However, cell death observed at 500 μg/mL and above suggests this may be the maximum administrable concentration in cell culture, warranting further investigation in in vivo studies. Moreover, the oxidized sample exhibited clear cytotoxicity from the minimum concentration at 24 h, unlike its counterparts. Given that the state of DA is the only variable between samples, it is evident that the toxic effect is linked to the oxidation of the neurotransmitter. This underscores the significance of stabilizing basal DA within the silicate matrix to protect it from oxidation and eliminate its intrinsic toxicity.

4. Discussion

PD remains an incurable, progressive syndrome primarily characterized by the progressive loss of dopaminergic neurons in the SNpc. This degeneration disrupts DA signaling within pathways critical for motor control, as well as other functions. While methods exist for the controlled administration of DA precursors, such as L-DOPA, there were previously no customizable mechanisms for the direct delivery of DA itself that could effectively prevent its oxidation into toxic by-products, hence reducing their adverse effects. This study demonstrates the successful stabilization of DA within mesoporous silicate nanoreservoirs, offering a promising approach for controlled DA release in PD treatment. The findings highlight the potential for these nanocarriers to maintain DA in its basal state over extended periods, preventing oxidation – a critical advancement given DA’s inherent instability. By testing various DA concentrations within the nanoreservoirs, this study establishes a framework for understanding the physicochemical properties, release kinetics, and biocompatibility of these formulations.

This work adds to a growing body of research on nanomedicine-based approaches for PD, addressing key limitations in traditional polymeric carriers for controlled delivery of both precursors and DA itself [13]. While previous studies have explored polymeric nanostructures like chitosan and poly(lactic-co-glycolic acid) (PLGA) for DA encapsulation [79–83], challenges such as slower degradation rates, limited biocompatibility, and potential toxicity have persisted. These polymeric carriers often degrade too slowly, disrupting the controlled release of DA and creating therapeutic lags [14,15]. In contrast, the use of inorganic mesoporous silicates – demonstrated in this study – overcomes these issues by providing a stable, biocompatible platform that supports sustained DA release with customizable surface chemistry. Silicate nanocarriers exhibit a highly stable structure, which improves the protection of sensitive drugs such as DA against degradation, especially due to temperature (up to 150°C) and oxidation, while their porous nature allows for high drug loading capacity and a tunable release profile (up to 600 mg of drug per gram of nanoreservoir) [84]. This makes them especially useful for applications requiring prolonged release and targeted delivery. In contrast to polymeric nanoparticles, which may suffer from instability in harsh physiological environments, silicates provide enhanced durability and chemical stability, minimizing premature release, as shown here for the first 10 hours of drug release. Additionally, these nanoreservoirs can be functionalized with various surface modifications, enabling precise control over release kinetics and improving bioavailability. Unlike liposomes, which can face challenges related to leakage and rapid clearance by the reticuloendothelial system [85], silicate nanocarriers are less prone to such issues due to their robustness and modifiable surface characteristics. Studies have demonstrated that these attributes make these carriers particularly promising for the stabilization and controlled release of drugs in various biomedical applications [86].

The silicate matrix used in this research builds upon prior work by López et al. [19,45] on oxide nanocarriers but introduces notable improvements, particularly in the synthesis and stabilization process. By fine-tuning sol-gel synthesis parameters, this study optimizes nanoreservoir properties, enabling more effective DA stabilization. Moreover, the incorporation of higher DA concentrations into the matrix, without significant oxidation, advances previous research that achieved DA stabilization at lower concentrations and with less predictable release kinetics. This study also contributes to understanding how the nanoreservoir’s pore size and structural characteristics influence DA release, providing a foundation for further refinement in future designs. In this regard, these nanocarriers could be tailored to individual patient needs, given the flexibility of the synthesis process [87,88]. For instance, adjusting the water-to-alkoxide ratio or modifying the pH during synthesis could fine-tune DA release rates. This adaptability could support personalized approaches to PD management, accommodating variations in patients’ responses to DA therapies.

Building on this study’s foundation, several avenues for future research can further explore DA nanoreservoirs’ therapeutic potential. First and foremost, in vivo studies are essential for validating their efficacy in animal models of PD. Such studies should assess how nanoreservoirs distribute across the blood–brain barrier [89], their pharmacokinetics [90], and their long-term effects on both motor and non-motor symptoms [91]. Additionally, examining how nanoreservoirs affect biomarkers of oxidative stress and inflammation in PD models could provide insights into their neuroprotective potential [92]. Further research is also needed to optimize the nanoreservoir design for enhanced therapeutic efficacy. For instance, coating the nanoreservoirs with polymers that facilitate blood–brain barrier penetration or adding targeting ligands specific to dopaminergic neurons could improve delivery efficiency [93]. Future studies could also explore incorporating other neuroprotective agents into the nanoreservoirs, creating a multifunctional platform capable of addressing various aspects of PD pathology. Additionally, adjusting DA concentrations and exploring co-delivery strategies with antioxidants could help mitigate potential cytotoxic effects observed at higher doses. Finally, as nanomedicine progresses toward clinical application, regulatory and safety considerations will become increasingly important. Long-term studies should examine DA nanoreservoirs’ stability under physiological conditions, their biodegradability, and any potential immunogenic responses. By addressing these factors, future research can advance the clinical feasibility of DA nanoreservoirs as a novel approach to PD therapy. Furthermore, beyond motor symptom management, stabilizing DA with nanoreservoirs may impact non-motor symptoms in PD, such as cognitive decline and mood disturbances, which are also linked to dopaminergic dysfunction [94]. The steady DA release from these nanocarriers could positively affect DA-dependent neurocircuits involved in these functions, as long as the nanocarriers are optimized to release DA following the physiological requirements of the patient. Although further research is needed to confirm this hypothesis, it represents a compelling direction for future studies.

Notably, our study has several limitations. The study primarily evaluates the nanoreservoirs in vitro. Although in vitro results are promising, they may not accurately reflect in vivo conditions due to the complex environment within a living organism. In this regard, although the release kinetics can be modified, the study’s findings are based on the Korsmeyer-Peppas model and in a simplified aqueous environment. Real biological environments could affect release rates differently due to factors like enzymatic activity, pH variations, and interaction with other cellular components. Further in vivo studies are needed to confirm the safety, efficacy, and potential side effects of these nanocarriers. Furthermore, while we achieved successful stabilization of dopamine, the long-term stability and resistance to oxidation in different biological environments remain uncertain. Finally, the nanoreservoirs showed cytotoxic effects at concentrations above 250 μg/mL. This indicates a threshold for safe use, limiting the potential dosage that can be safely administered, which may affect therapeutic effectiveness for certain applications.

5. Conclusions

In conclusion, this study provides a first proof-of-concept for DA-loaded mesoporous silicate nanoreservoirs as a controlled delivery system for PD treatment. The sol-gel synthesis method enables DA stabilization and sustained release while mitigating oxidation, something not previously reported, thus addressing critical challenges in current DA replacement therapies. Although there are areas that require further investigation, these findings lay a strong foundation for future work aimed at advancing DA nanoreservoir technology for therapeutic applications in PD and beyond.

Funding Statement

This project was funded by DGAPA-PAPIIT, grant number IN219623 and by CONAHCyT, grant number A1-S-10064.

Article highlights

• Parkinson’s disease (PD) is an incurable motor disorder that rely in the administration of dopamine (DA) precursors to reduce symptomatology.

• DA is not directly administered given its high oxidation rates, making it neurotoxic.

• We achieved the stabilization of DA in optimized sol-gel silicate nanoservoirs whilst preventing its oxidation.

• Nanoreservoirs exhibit nanoparticle sizes (below 30 nm), mesoporores (4 nm), superficial negative charges, and Lewis acid sites.

• DA is stabilized through weak bonds thus allowing for its controlled release.

• Nanoreservoirs’ controlled release of DA follows the Korsmeyer-Peppas model.

• The DA release kinetics can be adjusted by modifying synthesis parameters.

• DA-loaded nanoreservoirs are biocompatible and nontoxic to dopaminergic neurons.

• This work lay a strong foundation for future research on DA nanoreservoir technology in vivo and clinical trials for therapeutic applications in PD and beyond.

Author contributions

F.J.P.-G. did the material synthesis, full characterization, release kinetics determination, formal analysis, data curation, visualization, and manuscript writing. T.L.-G. designed the methodology, assisted with the synthesis, and is associated with revision, resources, project administration, and overall supervision of the work. E.Y.C.-M. did the TGA/DSC measurements. M.A.A.-L. did the BET and ELS measurements. J.N.-B. did the pyridine adsorption measurements. M.V. is associated with the release kinetics evaluation. O.C.-N., O.R.L.-M., and M.C.C.-A. are associated with cytotoxicity evaluation assays. M.G.-C. designed the methodology, is associated with writing, revision, funding acquisition, resources, project administration, and overall work supervision. The manuscript was written through the contributions of all authors. All authors have approved the final version of the manuscript.

Disclosure statement

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.

Declaration of interest

The authors declare no competing financial interest.

Data availability statement

All data for the replication of this work are given in the supplemental information or can be obtained by the lead contact upon reasonable request.

Resource availability

Lead contact. Further information and requests for resources should be directed to and will be fulfilled by the lead contact Magdalena Guerra-Crespo (mguerra@facmed.unam.mx).

Materials availability

Full experimental procedures can be found hereby and in the supplemental information.

Data and code availability

All data for the replication of this work are given in the supplemental information or can be obtained by the lead contact upon reasonable request.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/17435889.2025.2460228