Development and In Vivo Assessment of 4-Phenyltellanyl-7-chloroquinoline-loaded Polymeric Nanocapsules in Alzheimer’s Disease Models

Alzheimer’s disease (AD) is the most common form of dementia in older people, and available treatments are palliative and produce undesirable side effects. The 4-phenyltellanyl-7-chloroquinoline (TQ) is an organochalcogen compound studied due to its pharmacological properties, particularly its antioxidant potential. However, TQ possesses some drawbacks such as low aqueous solubility and high toxicity, thus warranting the search for tools that improve the safety and effectiveness of new compounds. Here, we developed and investigated the biological effects of TQ-loaded polymeric nanocapsules (NCTQ) in an AD model in transgenic Caenorhabditis elegans expressing human Aβ1–42 in their body–wall muscles and Swiss mice injected with Aβ25–35. The NCTQ displayed good physicochemical properties, including nanometer size and maximum encapsulation capacity. The treatment showed low toxicity, reduced Aβ peptide-induced paralysis, and activated an endoplasmic reticulum chaperone in the C. elegans model. The Aβ injection in mice caused memory impairment, which NCTQ mitigated by improving working, long-term, and aversive memory. Additionally, no changes in biochemical markers were evidenced in mice, demonstrating that there was no hepatotoxicity in the tested doses. Altogether, these findings provide insights into the neuroprotective effects of TQ and indicate that NCTQ is a promising candidate for AD treatment.


Introduction
Alzheimer's disease (AD) is the most common form of dementia in the elderly population. The main hallmarks of AD are extracellular amyloid-β peptide (Aβ) aggregation and neurofibrillary tangles formation [1]. Episodic memory impairment is one of the main aspects of this disease, and other symptoms include delayed recall and impaired language skills. Therapeutic alternatives attempting to counteract the damage caused by AD have yet to achieve permanent successful results, which include acetylcholinesterase inhibitors (AChE) (tacrine, donepezil, rivastigmine, and galantamine) and N-methyl-D-aspartate yet to achieve permanent successful results, which include acetylcholinesterase inh (AChE) (tacrine, donepezil, rivastigmine, and galantamine) and N-methyl-D-aspart ceptors (NMDARs) (memantine). However, these treatments produce undesirable s fects, thereby hindering patient adherence to treatment. Consequently, the search fo forms of medication is currently an urgent need.
Organochalcogen compounds comprise organic molecules that contain seleniu lurium, sulfur, or oxygen in their structure. These compounds have gained notoriet the years due to their pharmacological properties against neurodegenerative dis mainly in reversing and/or protecting memory impairment in animal models of AD Nevertheless, the utilization of organochalcogens in biological applications presen tain challenges, including concerns about their toxicity and specific physicoche properties (such as low hydrophilicity, which affects its bioavailability). Recent s have provided evidence that incorporating organoalcogen compounds into nanopa significantly enhances their selectivity and pharmacological effectiveness [5][6][7][8].
Our research group has been studying the pharmacological, toxicological, an lecular effects of organochalcogen compounds in Caenorhabditis elegans and rodent 13], and tellurium organic compounds have demonstrated neuroprotective prop against Mn-induced neurotoxicity and in the AD model in mice [12,14,15]. Howeve important to note that pharmacologic studies performed with organotellurium pounds are relatively scarce, since their pharmacological window is narrow [16,17 result, the development of new organotellurium compounds still needs to be explor this sense, we highlight the TQ (Figure 1), which showed promising pharmacol properties, since TQ modulated transcription factors SKN-1 (homologous to NRF2 mans) and DAF-16 (homologous to FOXO3a in humans) and its downstream eff including the antioxidant enzymes SOD-3 and GCS-1 [13], which are both pathwa quired for the beneficial effects exerted by other neuroprotective therapies in tran C. elegans [18,19] and mammals [19,20]. Given the above, we hypothesized that nanoencapsulation of TQ could reduc icity and optimize its effectiveness by modulating physicochemical and biopharma cal properties or minimizing adverse effects and maximizing therapeutic efficacy. meric nanocapsules have been shown to improve drug solubility, increase their bio ability, and target specific tissues or cells, thereby reducing systemic toxicity and im ing therapeutic efficacy [5,21,22].
The use of nanoencapsulated drugs is an opportune strategy for brain deliver ferent nanoparticles are described in the literature, including nanosystems such as tubes, nanospheres, liposomes, and nanocapsules. The use of polymeric nanocapsul been reported as a reproducible method with good stability and biocompatibility [ 25]. Studies have demonstrated that polysorbate 80 (nonionic surfactant) causes a t rary and partial disruption of the blood-brain barrier, decreasing P-glycoprotein efflux activity and apolipoprotein-E (ApoE) levels which facilitate endocytosis in the [24,26], which may lead to substantial concentrations of these drugs in the central ne system. In addition, studies have demonstrated an improved response in vivo using encapsulated drugs targeting the brain, such as curcumin and meloxicam [27][28][29]. given that nanoencapsulated TQ may be a worthwhile strategy in AD drug develop Given the above, we hypothesized that nanoencapsulation of TQ could reduce toxicity and optimize its effectiveness by modulating physicochemical and biopharmaceutical properties or minimizing adverse effects and maximizing therapeutic efficacy. Polymeric nanocapsules have been shown to improve drug solubility, increase their bioavailability, and target specific tissues or cells, thereby reducing systemic toxicity and improving therapeutic efficacy [5,21,22].
The use of nanoencapsulated drugs is an opportune strategy for brain delivery. Different nanoparticles are described in the literature, including nanosystems such as nanotubes, nanospheres, liposomes, and nanocapsules. The use of polymeric nanocapsules has been reported as a reproducible method with good stability and biocompatibility [21,[23][24][25]. Studies have demonstrated that polysorbate 80 (nonionic surfactant) causes a temporary and partial disruption of the blood-brain barrier, decreasing P-glycoprotein (P-gp) efflux activity and apolipoprotein-E (ApoE) levels which facilitate endocytosis in the brain [24,26], which may lead to substantial concentrations of these drugs in the central nervous system. In addition, studies have demonstrated an improved response in vivo using nanoencapsulated drugs targeting the brain, such as curcumin and meloxicam [27][28][29]. Thus, given that nanoencapsulated TQ may be a worthwhile strategy in AD drug development, this study aimed to develop and characterize the physicochemical properties of polymeric nanocapsules containing TQ (NCTQ) and test the potential neuroprotective effect of TQ and NCTQ in different models of AD.

Nanocapsules Preparation
The NCTQ and nanocapsules without TQ (NCBR) were obtained by interfacial deposition of poly(ε-caprolactone) (PCL) [21]. The organic phase was composed of PCL polymer, Span 60 ® surfactant, TQ (1 mg/mL), caprylic/capric triglyceride, acetone, and ethanol and maintained at 35 ± 1 • C under magnetic stirring to solubilize all compounds. Afterward, this phase was injected into the aqueous phase containing distilled water and polysorbate 80 surfactant to produce the nanocapsules. The mean diameter (D [4,3] ) and particle size distribution (span) of nanocapsules suspension were evaluated by laser diffraction analysis (Mastersizer 2000 ® , Malvern Instruments, Worcestershire, UK) (n = 3). The superficial charge was analyzed as a zeta potential by electrophoretic migration (NanoBrook 90Plus, BrookHaven ® ) (n = 3) from a 1:1000 dilution of samples in a pre-filtered NaCl solution (1 mM). The pH of the nanoformulations was evaluated immediately after preparation using a previously calibrated potentiometer (Hanna Instruments, São Paulo, Brazil) (n = 3).

Drug Content Determination and Encapsulation Efficiency
Sample solutions containing NCTQ were prepared at the theoretical concentration of 10 µg/mL in a volumetric flask and maintained in an ultrasonic bath for 30 min to break the nanocapsules. The samples were filtered (0.45 µm, Millipore) before HPLC-PDA quantification (a method previously developed and validated according to Section 2.3.4). Encapsulation efficiency (EE%) was assessed by the ultrafiltration centrifugation method (Ultrafree, Millipore). The EE% was determined by the difference between the total concentration of each drug and its concentration in the aqueous phase using Equation (1) and given as a percentage of recovery (%) relating to the theoretical concentration.
where TQ total is the theoretical concentration of TQ and TQ f ree is the concentration of free TQ in the ultrafiltrate.

HPLC-PDA Apparatus and Chromatographic Conditions
A Shimadzu ® High-Performance Liquid Chromatography (HPLC) system (Kyoto, Japan) equipped with a photodiode array (PDA) detector was used for separation and absorbance analysis, respectively. Data acquisition and system control were performed by analytical software (LC Solution, Release 1.22 SP1). The mobile phase comprised acetonitrile:water:triethylamine (90:10:0.3 v/v/v). The aqueous phase was adjusted to pH 7.0 with phosphoric acid. The TQ was separated using a chromatographic column at 25 ± 1 • C (5 µm, 4.6 × 150 mm; Nano Separation Technologies RP-18). The flow rate was defined as 1.0 mL/min, and TQ was detected at 325 nm after injecting 20 µL.

Method Validation and Sample Preparation
The proposed method was validated based on the determination of specificity, linearity, precision, accuracy, robustness, and system suitability, according to general guidelines [31][32][33].

Linearity and Sensitivity
Linearity was analyzed by preparing three calibration curves of TQ solution in three different assays (5,10,15,20,30,40, and 50 µg/mL) (n = 9). The correlation coefficient (R 2 ) obtained by linear regression was evaluated to verify the suitability of the method. The calibration line was used to assess the limit of detection (LOD) and limit of quantitation (LOQ) [31][32][33].

Precision
Precision was determined by processing six independent TQ samples (10 µg/mL) on the same day (intra-day precision) or three different days (inter-day precision). The results were expressed as relative standard deviation percentage (RSD%).

Accuracy
Accuracy was determined by adding a single TQ concentration to 100 µL of NCBR in acetone, corresponding to 10 µg/mL (n = 3). The results were evaluated according to the recovery percentage of each sample.

Robustness
Robustness was determined by minimally changing some parameters of the analytical method, such as flow change (0.9 and 1.1 mL/min) and pH (6.8 and 7.2) (n = 3). These data were then analyzed according to values of Rt (retention time), T (tailing factor ≤ 2.0), k (retention factor ≥ 2.0), and N (theoretical plate number ≥ 2000) [31,32].

Specificity
Specificity was evaluated from samples containing NCBR to investigate possible interference of the excipients present in the nanoformulation. For this, 100 µL was added to a volumetric flask and diluted in acetonitrile to obtain 10 mL. The samples were maintained in an ultrasonic bath for 30 min and filtered before HPLC injection. This study used strains Bristol N2 (wild-type), SJ4005 (hsp-4::GFP), SJ4100 (hsp-6p::GFP + lin-15(+)), and CL2006 (pCL12(unc-54/human Aβ 1-42 ) + pRF4), worms genetically modified for human Aβ expression. These animals were obtained from the Caenorhabditis Genetics Center (CGC, Minnesota, USA). Worms were maintained in nematode growth medium (NGM) plates seeded with Escherichia coli OP50 at 15 or 20 • C. After reaching the adult stage, pregnant hermaphrodites were washed using a bleaching solution (2.4% NaOH and 1% NaClO in distilled H 2 O) to obtain eggs. At the first larval stage (L1), 1500 worms were acutely exposed to TQ, NCTQ, or NCBR at 1 and 10 µM for 30 min. The control group was exposed to dimethyl sulfoxide (DMSO) at 5% (TQ vehicle). Afterwards, the worms were washed three times with M9 buffer (KH 2 PO4 0.02 M, Na 2 HPO 4 0.04 M, and NaCl 0.085 M) to remove the treatment. The animals were then placed on NGM plates seeded with E. coli OP50 until they reached the L4 larval stage (48 h after L1 stage). In the strain CL2006, the worms were exposed to an up-shift in temperature to a non-permissive 20 • C after the treatment to express the transgenic Aβ peptide in the body wall muscles (Figure 2). the worms were washed three times with M9 buffer (KH2PO4 0.02 M, Na2HPO4 0.04 M, and NaCl 0.085 M) to remove the treatment. The animals were then placed on NGM plates seeded with E. coli OP50 until they reached the L4 larval stage (48 h after L1 stage). In the strain CL2006, the worms were exposed to an up-shift in temperature to a non-permissive 20 °C after the treatment to express the transgenic Aβ peptide in the body wall muscles ( Figure 2).

Figure 2.
Scheme of the experimental protocol to assess the toxicity and efficacy of NCTQ in C. elegans. After 14 h of synchronization, worms in the first larval stage (L1) were acutely (30 min) exposed to TQ, NCTQ, or NCBR at 1 and 10 µM. Afterward, the worms were washed three times to remove the treatment. The animals were then placed on NGM plates seeded with E. coli OP50 as a food source until they reached the L4 larval stage. In this larval stage, we performed the safety evaluation through survival rate, brood size, and body length assays in the strain CL2006. In addition, in the L4 stage, worms of strains SJ4005 and SJ4100 were submitted to a heat shock for 4 h to posteriorly evaluation of chaperones HSP-4 and HSP-6 expression tagged with GFP. At the adult stage, we followed the Aβ aggregation through paralysis phenotype evaluation in the strain CL2006.

Toxicity Endpoints in C. elegans
Nanoformulation safety was evaluated by survival assay, counting the number of live worms in the plates after 48 h of treatment. Animal development was analyzed by obtaining images from five worms of each group and measuring the body length using the software ImageJ. One animal (n = 3) was transferred to new NGM plates with E. coli OP50 to determine brood size during 4 days of reproductive period, which is a parameter of reproductive toxicity and development. The results were normalized and expressed as a percentage of control. All assays were performed in duplicates, except for brood size (triplicates). All assays were repeated in three independent experiments.

Chaperones HSP-4 and HSP-6 Expression
The strains SJ4005 and SJ4100 present a GFP reporter transgene controlled through hsp-4 and hsp-6 promoter, and the expression of these chaperones is localized in the endoplasmic reticulum and mitochondria, respectively. When the worms reached the larval stage L4, they were exposed to a temperature up-shift at 37 °C (heat shock) for 4 h to induce chaperone expression [34]. After this period, the worms were collected and washed with distilled water, then worms were anesthetized with levamisole (1 mM), placed on slides, and the images were obtained using fluorescent microscopy (Floid Cell Imaging Station ® , Thermo Fisher Scientific, Waltham, MA, USA). The intensity fluorescence of 5 worms was quantified using the ImageJ software. This assay was conducted in triplicates and repeated three times. In addition, this method was performed in accordance with previously published studies by our research group [35]. After 14 h of synchronization, worms in the first larval stage (L1) were acutely (30 min) exposed to TQ, NCTQ, or NCBR at 1 and 10 µM. Afterward, the worms were washed three times to remove the treatment. The animals were then placed on NGM plates seeded with E. coli OP50 as a food source until they reached the L4 larval stage. In this larval stage, we performed the safety evaluation through survival rate, brood size, and body length assays in the strain CL2006. In addition, in the L4 stage, worms of strains SJ4005 and SJ4100 were submitted to a heat shock for 4 h to posteriorly evaluation of chaperones HSP-4 and HSP-6 expression tagged with GFP. At the adult stage, we followed the Aβ aggregation through paralysis phenotype evaluation in the strain CL2006.

Toxicity Endpoints in C. elegans
Nanoformulation safety was evaluated by survival assay, counting the number of live worms in the plates after 48 h of treatment. Animal development was analyzed by obtaining images from five worms of each group and measuring the body length using the software ImageJ. One animal (n = 3) was transferred to new NGM plates with E. coli OP50 to determine brood size during 4 days of reproductive period, which is a parameter of reproductive toxicity and development. The results were normalized and expressed as a percentage of control. All assays were performed in duplicates, except for brood size (triplicates). All assays were repeated in three independent experiments.

Chaperones HSP-4 and HSP-6 Expression
The strains SJ4005 and SJ4100 present a GFP reporter transgene controlled through hsp-4 and hsp-6 promoter, and the expression of these chaperones is localized in the endoplasmic reticulum and mitochondria, respectively. When the worms reached the larval stage L4, they were exposed to a temperature up-shift at 37 • C (heat shock) for 4 h to induce chaperone expression [34]. After this period, the worms were collected and washed with distilled water, then worms were anesthetized with levamisole (1 mM), placed on slides, and the images were obtained using fluorescent microscopy (Floid Cell Imaging Station ® , Thermo Fisher Scientific, Waltham, MA, USA). The intensity fluorescence of 5 worms was quantified using the ImageJ software. This assay was conducted in triplicates and repeated three times. In addition, this method was performed in accordance with previously published studies by our research group [35].

Aβ Peptide Aggregation Model in C. elegans
The strain CL2006 expresses the Aβ peptide in muscle cells, causing a locomotor impairment mentioned, (i.e., paralysis). Therefore, to verify the effects of the nanoformulations in an AD model, we performed a paralysis assay that measured worms' response to mechanical stimuli. If the worms moved from the point of origin after stimuli (touch with a brush), they were not considered paralyzed, and if the animals only moved the head or Brain Sci. 2023, 13, 999 6 of 20 pharynx, they were considered paralyzed. All assays were performed with 25 animals in duplicates and three independent experiments [35,36].

Alzheimer's Disease Model in Mice
Three-month-old male Swiss albino mice weighing 25-35 g were acquired from the Federal University of Pelotas (Brazil). We chose Swiss mice based on studies that demonstrated the effect of compounds in animal models of AD in this strain [9,26]. The animals were maintained at a constant temperature (22 ± 1 • C) and in a 12 h dark/light cycle and provided with food and water ad libitum. Animal care and experimental procedures were conducted in compliance with the Guide for the Care and Use of Laboratory Animals (NIH publication no. 8023) [37] and approved by the Committee on Care and Use of Animal Resources of the Federal University of Pelotas, Brazil (CEEA: 7046/2016). The number of animals and intensity of noxious stimuli used were the minimum needed to demonstrate the consistent effects of the treatments.

Experimental Protocol
Mice were randomly divided into six experimental groups (6-8 animals/group): Sham, TQ, NCTQ, Aβ, Aβ + TQ, and Aβ + NCTQ. Thirty minutes before Aβ 25-35 exposure, mice from the sham and Aβ groups received the NCBR dose (10 mL/kg), mice from the TQ and Aβ + TQ groups received the free TQ dose (1 mg/kg), and mice from the NCTQ and Aβ + NCTQ groups received the NCTQ dose at the same dosage. All treatments were intragastrically administered (i.g.) via oral gavage. After treatments, mice belonging to the Aβ, Aβ + TQ, and Aβ + NCTQ groups received Aβ (3 nmol/3 µL/per site by intracerebroventricular injection; i.c.v) [26], while the sham, TQ, and NCTQ groups received saline (3 µL/per site; i.c.v). The i.c.v infusion of Aβ or vehicle (saline) was administered according to a previous study [38]. Mice were treated with TQ, NCTQ, or NCBR every day until the 14th day. On the 15th day, mice were anesthetized by isoflurane inhalation for blood collection by a cardiac puncture, which was removed for ex vivo experiments. The experimental protocol is demonstrated in Figure 3. . Scheme of the experimental protocol used to assess the general toxicity and efficacy of NCTQ in mice. On the first day, thirty minutes after intragastric treatments, mice received amyloid Aβ or vehicle (saline) by the intracerebroventricular route. I.g. treatments with TQ, NCTQ, and NCBR were performed every day until the fourteenth day of the experimental protocol. After four days of Aβ injection, the animals were submitted to the Y-maze, object recognition task, and stepdown inhibitory avoidance tests. On the fifteenth day, blood was collected to evaluate general toxicity markers. Figure 3. Scheme of the experimental protocol used to assess the general toxicity and efficacy of NCTQ in mice. On the first day, thirty minutes after intragastric treatments, mice received amyloid Aβ or vehicle (saline) by the intracerebroventricular route. I.g. treatments with TQ, NCTQ, and NCBR were performed every day until the fourteenth day of the experimental protocol. After four days of Aβ injection, the animals were submitted to the Y-maze, object recognition task, and step-down inhibitory avoidance tests. On the fifteenth day, blood was collected to evaluate general toxicity markers.

Behavioral Tests Y-Maze Task
The Y-maze task was performed as described by Sarter et al. (1988) [39] and was used as a measure of spatial and working memory (8th day; Figure 3). The Y-maze apparatus consisted of a three-arm horizontal maze measuring 40 cm in length, 3 cm in width, and with walls 12 cm high. The three arms were positioned at 120 • angles to each other, radiating out from a central point. The Y-maze test was performed on the eighth day of the experimental protocol ( Figure 3). Mice were initially placed within one arm (A), and the sequence of arm entries (i.e., ABCCAB, where letters indicate arm codes) and the number of arm entries were manually recorded for each mouse over an 8-min period. Alternation was determined by observing successive entries into the three arms on overlapping triplet sets, where three different arms were entered. An actual alternation was defined as consecutive entries into all three arms (i.e., ABC, CAB, or BCA, but not BAB). An entry was counted when all four paws were placed within the boundaries of the arm. The percentage of alternation was calculated as follows: % Alternation = [(Number of alternations × 3)/(Total arm entries − 2)] × 100.

Object Recognition Task
To assess long-term memory (LTM), the object recognition task was used (12th day; Figure 3) [40]. Each mouse was submitted to a habituation session, and the LTM was performed 24 h after training (13th day; Figure 3), where mice explored a familiar object (A1) and a new object (B) for 5 min, and the total time spent exploring each object was determined. Cognitive performance was analyzed by calculating the exploratory preference, and data were expressed as a percentage as training = (A2/(A1 + A2)) × 100; LTM = (C/(A1 + C)) × 100.
Step-Down Inhibitory Avoidance A step-down inhibitory avoidance task evaluated non-spatial and aversive LTM (14th day; Figure 3) [41]. In this one-trial learning task, the animals were put on a platform and received an electric shock (0.5 mA) for 2 s after stepping off the platform onto the grid. When the animals were tested 24 h later, they were exposed to the training apparatus, although no shock was delivered, and the transfer latency time was measured. The maximum transfer latency time was 300 s.

Ex-Vivo Assays
Mice were anesthetized with isoflurane, and blood samples were collected from the heart ventricle to obtain plasma and determine aspartate (AST) and alanine (ALT) aminotransferases (15th day; Figure 3). Plasma was obtained by centrifugation (900× g, 15 min). AST and ALT activities were determined in the plasma using a commercial kit (Bioclin, São Paulo, SP, Brazil).

Statistical Analysis
The C. elegans results were analyzed using one-way analysis of variance (ANOVA) and Tukey's post hoc test for survival, body length, brood size, and chaperone expression. The paralysis rate was assessed by repeated measures ANOVA and Dunnett's post hoc test (GraphPad Prism 7.04, GraphPad, San Diego, CA, USA). Mice data were analyzed by the GraphPad Prism 5 software, and data normality was evaluated by the D'Agostino and Pearson omnibus normality test. Statistical analysis was performed using one-way ANOVA followed by Tukey's post hoc test. The data are expressed as mean ± standard error of the mean (SEM), and p < 0.05 values were considered statistically significant.

Physicochemical Characterization of Nanocapsules
Both NCTQ and NCBR presented a white opalescent aspect commonly observed in nanosuspensions, known as the Tyndall effect. Laser diffraction analysis demonstrated particles on a nanometric scale with a small standard deviation in the diameter of their nanoparticles and span < 2.00. The NCTQ showed a small variation in D [4,3] (240 ± 6 nm) regarding NCBR (231 ± 6 nm; Table 1). Both NCTQ and NCBR presented a negative zeta potential (−25.37 ± 1.52 and −24.36 ± 5.54, respectively; Table 1). Additionally, the pH evaluation of both formulations was approximately 6.00 (Table 1). The values obtained for EE% and drug content were close to the theoretical concentration of NCTQ formulation (1 mg/mL; Table 1).

Validation of the Analytical Methodology
The aqueous phase was composed of water containing 0.03% of triethylamine (v/v) (pH 6.2 adjusted with phosphoric acid) and the organic phase was acetonitrile (85:15 v/v). The flow rate was established as 1.0 mL/min with a total run time of 10 min. The chromatograms obtained showed a sharp and symmetric peak with good resolutions in these conditions. The maximum absorption of TQ was observed at 325 nm, as evidenced by the increase in the peak area at 8.20 min.
Linearity showed a linear relationship with the increase in concentration during drug quantification. The results demonstrated a linear relationship (R 2 = 0.9996) with a representative linear equation established as y = 16,501x − 18,105. The precision was demonstrated as repeatability (intra-day) and intermediate precision (inter-day) ( Table 2). The results showed that our method was found to be repeatable, according to guidelines (RSD < 2.0%) [31,32]. The intra-day precision was carried out by preparing six samples of TQ containing 10 µg/mL. The inter-day precision was assessed on three different days (n = 18). The intra-day precision was assessed on the same day and with the same experimental conditions (n = 6).
The robustness assay demonstrated minimal variations in the method caused expressive changes in the analyte chromatogram. The data show that slight pH and flow rate variations did not cause significant chromatographic changes to the proposed method, thus demonstrating its robustness. These values follow the guidelines and are summarized in Table 3 [31,32].

NCTQ Did Not Elicit Toxicity in C. elegans
The results obtained for survival rate, body length, and the number of larvae are shown below ( Figure 4A-C). The results revealed that none of the acute treatments reduced worm survival rate at 1 or 10 µM in both free and nanoencapsulated forms of TQ ( Figure 4A). In addition, no signs of toxicity in development and reproduction were observed ( Figure 4B,C).

TQ and NCTQ Increased Reticular Chaperone (HSP-4) Expression
Corroborating the paralysis results, TQ (1 and 10 µM) and the NCTQ (10 µM) increased GFP intensity of the reticular chaperone HSP-4 ( Figure 5A-C), with NCBR (1 µM) also demonstrated this effect. However, when we evaluated GFP intensity in the strain SJ4100, which presents GFP-tagged HSP-6 and is localized in the mitochondria, we did not find any significant results in the modulation of this heat shock protein ( Figure 5B-D).

TQ and NCTQ Increased Reticular Chaperone (HSP-4) Expression
Corroborating the paralysis results, TQ (1 and 10 µM) and the NCTQ (10 µM) increased GFP intensity of the reticular chaperone HSP-4 ( Figure 5A-C), with NCBR (1 µM) also demonstrated this effect. However, when we evaluated GFP intensity in the strain SJ4100, which presents GFP-tagged HSP-6 and is localized in the mitochondria, we did not find any significant results in the modulation of this heat shock protein ( Figure 5B-D).

Treatment Reduced Aβ Peptide-Induced Paralysis Rate
The strain CL2006 was bioengineered with a constitutive muscle-specific promoter, accumulating β-immunoreactive deposits and intracellular amyloid (centrally involved in the AD pathogenesis), resulting in a progressive paralysis phenotype [42]. When we evaluated the locomotor dysfunction in the worms, we observed that both TQ and NCTQ reduced the paralysis rates by 10 µM (Figure 4D,E).

Number of Arm Entries and Spontaneous Alternation Behavior
The effects of treatments on the behavioral parameters, specifically the number of arm entries and spontaneous alternation behavior in the Y-maze task of mice, are depicted in Figure 6A,B. The results demonstrated that the treatments did not significantly impact the number of arm entries ( Figure 6A), suggesting that the treatment did not induce changes in motor function among the mice. Furthermore, the findings indicated that Aβ reduced spontaneous alternation behavior by approximately 25% compared to the sham group ( Figure 6B). Another noteworthy finding was that the NCTQ treatment effectively prevented the reduction of Aβ-induced alternations in the Y-maze task, whereas TQ did not exhibit any effects ( Figure 6B). Mice solely pretreated with TQ did not display any discernible differences in spontaneous alternation behavior during the Y-maze task (Figure 6B). However, the treatment with NCTQ alone significantly increased the number of

Treatment Reduced Aβ Peptide-Induced Paralysis Rate
The strain CL2006 was bioengineered with a constitutive muscle-specific promoter, accumulating β-immunoreactive deposits and intracellular amyloid (centrally involved in the AD pathogenesis), resulting in a progressive paralysis phenotype [42]. When we evaluated the locomotor dysfunction in the worms, we observed that both TQ and NCTQ reduced the paralysis rates by 10 µM (Figure 4D,E).

NCTQ Ameliorates Memory Impairment in Mice Number of Arm Entries and Spontaneous Alternation Behavior
The effects of treatments on the behavioral parameters, specifically the number of arm entries and spontaneous alternation behavior in the Y-maze task of mice, are depicted in Figure 6A,B. The results demonstrated that the treatments did not significantly impact the number of arm entries ( Figure 6A), suggesting that the treatment did not induce changes in motor function among the mice. Furthermore, the findings indicated that Aβ reduced spontaneous alternation behavior by approximately 25% compared to the sham group ( Figure 6B). Another noteworthy finding was that the NCTQ treatment effectively prevented the reduction of Aβ-induced alternations in the Y-maze task, whereas TQ did not exhibit any effects ( Figure 6B). Mice solely pretreated with TQ did not display any discernible differences in spontaneous alternation behavior during the Y-maze task ( Figure 6B). However, the treatment with NCTQ alone significantly increased the num-ber of spontaneous alternations ( Figure 6B). These findings strongly suggest that NCTQ actively mitigates spatial and working memory impairments induced by Aβ.
i. 2023, 13, x FOR PEER REVIEW spontaneous alternations ( Figure 6B). These findings strongly sugge mitigates spatial and working memory impairments induced by A The effects of the treatments on the exploratory preference u the object recognition task (ORT) are illustrated in Figure 7A,B. Th in the exploratory preference of objects among groups in the trainin

Object Recognition Task
The effects of the treatments on the exploratory preference using the new object in the object recognition task (ORT) are illustrated in Figure 7A,B. There was no difference in the exploratory preference of objects among groups in the training phase ( Figure 7A).
In the probe test, mice injected with Aβ showed reduced exploratory preference for the new object compared to the sham group ( Figure 7B). Both TQ and NCTQ prevented the reduction of the preference exploratory for LTM ( Figure 7B). These findings imply that TQ and NCTQ act in spatial and LTM impairment caused by Aβ. In the probe test, mice injected with Aβ showed reduced exp the new object compared to the sham group ( Figure 7B). Both TQ the reduction of the preference exploratory for LTM ( Figure 7B). Th TQ and NCTQ act in spatial and LTM impairment caused by Aβ.
Step-Down Inhibitory Avoidance Task Figure 8A,B present the effect of treatments in the step-down task. In the training phase, there was no difference in the transfe groups ( Figure 8A). In the test phase, Aβ decreased (approximatel Step-Down Inhibitory Avoidance Task Figure 8A,B present the effect of treatments in the step-down inhibitory avoidance task. In the training phase, there was no difference in the transfer latency time among groups ( Figure 8A). In the test phase, Aβ decreased (approximately 75%) the transfer latency time compared to the sham group ( Figure 8B). NCTQ treatment significantly prevented this reduction ( Figure 8B). Furthermore, NCTQ and TQ alone did not modify the time in step-down inhibitory avoidance ( Figure 8B). These data demonstrated that only nanoencapsulated TQ protected against the impairment of non-spatial long-term aversive memory.
i. 2023, 13, x FOR PEER REVIEW Figure 8. Effects of TQ, NCTQ, and/or Aβ on step-down inhibitory avoida tency (s) to fall from the platform in the step-down inhibitory avoidance i section. Data are reported as mean ± standard error of the mean (SEM) group (one-way analysis of variance/Tukey's test). (*) p < 0.05 as compared p < 0.05 as compared with the Aβ group. Table 4 shows the effect of treatments on the plasma biochemi was no difference among groups in the ALT and AST activities. In that mice treated with TQ and NCTQ for 14 days at 1 mg/kg did no and ALT activities, suggesting that TQ and NCTQ were not hepatot   Table 4 shows the effect of treatments on the plasma biochemical markers, and there was no difference among groups in the ALT and AST activities. In this study, we verified that mice treated with TQ and NCTQ for 14 days at 1 mg/kg did not show changes in AST and ALT activities, suggesting that TQ and NCTQ were not hepatotoxic at the tested dose. Data are reported as mean ± standard error of the mean (SEM) of six to eight animals per group. Statistical analysis was performed using one-way analysis of variance/Tukey's test. Aspartate (AST) and alanine (ALT) aminotrasferases.

Discussion
This study aimed to develop and characterize physicochemical aspects of polymeric nanocapsules for TQ brain delivery, as well as the biological activity in a mouse model and the invertebrate C. elegans. The analytical method was developed and validated following the guidelines of the International Conference of Harmonization and the Brazilian Health Regulatory Agency [31,33]. NCTQ exhibited adequate physicochemical properties, including nanometer size and high encapsulation capacity. Treatment with NCTQ demonstrated low toxicity, reduced Aβ-induced paralysis in C. elegans, and activated an endoplasmic reticulum chaperone response. In the mouse model, Aβ treatment caused memory impairment, mitigated by NCTQ, improving working, long-term, and aversive memory. Furthermore, no significant changes in biochemical markers of hepatotoxicity in mice were observed.
Previously, our research group demonstrated that clozapine-loaded nanocapsules increased plasma exposure in rats treated with a single intravenous dose. The pharmacodynamic study also showed that nanoencapsulation improved the pharmacological effect in terms of antipsychotic potency as well as the action duration [22]. In this regard, Dimer et al. (2014) [43] demonstrated that nanoencapsulation of olanzapine increased brain exposure to the drug, corroborating the findings of our previous study. Different methods for obtaining nanocapsules are described in the literature, which are generally classified as in situ polymerization of monomers (alkyl cyanoacrylate) or the precipitation of pre-formed polymers (PCL, poly(lactic acid) (PLA), poly(lactic-co-glycolic acid) (PLGA), and methacrylate copolymers) [25,44]. The choice of preparation method, as well as the monomer or polymer to be employed, will depend on the study's objectives, drug characteristics (if applicable), and desired production time [45]. In this study, the choice of nanoprecipitation method using PCL as the pre-polymer and Span 60 ® as the surfactant was based on the previously observed good stability, reproducibility, and absence of toxicity [21,25].
Moreover, the nanoencapsulation of compounds has been associated with a decrease in compound toxicity. For instance, the co-encapsulation of an antimalarial agent (quinine) with curcumin demonstrated a protective effect on worms exposed to free quinine, as evidenced by survival and reproductive parameters [24]. Similar results were reported by Moraes et al., 2016 [46], with the nanoencapsulation of clozapine. Furthermore, clozapineloaded nanocapsules demonstrated a lower incidence of oxidative damage in the brains of rats compared to the same dose of free drug. Additionally, blank nanocapsules were used as a control, showing no oxidative damage [23]. In this regard, the behavioral assessment of rats treated with blank nanocapsules with different coatings demonstrated no behavioral effects, especially those with anionic characteristics [25].
The nanometer size and polymer coating of polymeric nanocapsules positively improved physicochemical aspects and pharmacological efficacy [24,27]. These data corroborate other studies that developed PCL nanocapsules using polysorbate 80 as a stabilizer [22,24,27,47]. The obtained encapsulation efficiency values and drug content of approximately 100% demonstrated maximum drug entrapment capacity. These results are attributed to the oily core of the nanocapsules [22], which improves TQ solubilization. Considering the non-ionic nature of Span 60 ® , the negative zeta potential obtained was due to the ionization of the PCL carboxylic groups and agree with the previously reported results [25]. Furthermore, values close to −30 mV are considered adequate to ensure the balance between stability and cytotoxicity [48,49]. The zeta potential evaluation is important for the stability of the nanoparticle systems since particles with similar charges can repel each other, so with adequate charges, the samples are more resistant to flocculation, sedimentation, and aggregation processes [48]. The increase in drug exposure by nanoencapsulation emphasizes the importance of toxicological studies [22,50]. Alternative models provide preliminary information for initial safety screenings without using a significant number of mammals [24].
Notably, tellurium-containing organic compounds can cause toxic effects on animals, and some species may be more sensitive than others [16]. Given the absence of data on the effects of NCTQ in living organisms, we evaluated its toxicity in one nonmammalian model: C. elegans transgenic model of AD. Free TQ has been reported to lack toxicity at 1 µM [13], and our data demonstrated the safety of using TQ nanoformulations ( Figure 4A-E). C. elegans has been proven to be a powerful complementary biological model during the preclinical drug development stage and studying senile diseases [51]. This is due to several mammalian orthologous genes, including numerous genes associated with human diseases [52]. When these human genes are not present in the worms, it is possible to generate transgenic animals that express these genes, presenting phenotypical characteristics that resemble human diseases. In this context, the strain CL2006 was bioengineered with a constitutive muscle-specific promoter, accumulating Aβ-immunoreactive deposits and intracellular amyloid (centrally involved in the AD pathogenesis), resulting in a progressive paralysis phenotype. Indeed, we observed that these worms presented high paralysis rates and that both TQ and NCTQ at 10 µM attenuated this phenotype ( Figure 4E).
In a recent study, we found that TQ (1 µM) reversed oxidative damage and restored life expectancy reduced by the deleterious effects of paraquat by modulating the DAF-16/FOXO transcription factor [13]. The translocation of DAF-16 to the nucleus induced by TQ may activate antioxidant, pro-longevity, and proteostasis-promoting genes, which may help reduce the deleterious effects of Aβ. For instance, when DAF-16 is activated, it promotes a cellular survival response to the proteotoxicity induced by Aβ aggregation and polyglutamine expansion aggregation, acting in association with heat shock factor-1 (HSF-1) [53]. Therefore, based on the TQ mechanism previously evidenced, we hypothesized that this molecule could improve the phenotypical alterations caused by Aβ aggregation by activating molecular chaperones, also known as heat shock proteins (HSPs), which can degrade protein aggregates. Many chaperone families are involved in this process, such as the HSP70. Studies have demonstrated the role of HSP70 in preventing Aβ formation [54]. In this study, we demonstrated that TQ (1 and 10 µM) and NCTQ (10 µM) ( Figure 5A-D) increased the activation of HSP-4 expression.
In C. elegans, the HSP-4 is localized in the endoplasmic reticulum (ER), homologue to HSP70 family chaperones in humans, and is activated in response to stress conditions. Modulation in the DAF-16/FOXO through TQ treatment may be involved in the response of this chaperone [13]. However, mitochondrial HSP-6 did not increase by any of the treatments ( Figure 6B-D). It is plausible that the treatments induced antioxidant enzymes due to DAF-16/FOXO activation and did not induce appropriate conditions for HSP-6 activation [13,55]. In addition, our data suggest that HSP-4 induction seems a compensatory response to contain Aβ formation and reduce the aggregates ( Figure 5). It is important to emphasize that worms do not have as many cellular barriers as mammals, and their digestive system is quite different, which may explain the lack of differences between the free TQ and NCTQ in this study.
The pathogenesis of AD is closely associated with the presence of Aβ, which aggregates in the brain and forms senile plaques and intracellular neurofibrillary tangles, leading to neural loss and oxidative damage [1]. The most prominent Aβ fragments implicated in AD pathology are Aβ 40 and Aβ 42 , which consist of 40 and 42 amino acids, respectively. For instance, Aβ 25-35 is a widely investigated peptide fragment due to its ability to form aggregates and induce neurotoxic effects [56]. Therefore, an Alzheimer Aβ peptide-based model for mice was established and used to assess the effect of NCTQ. Indeed, this i.c.v. injection model with Aβ 25-35 is commonly used as an AD model because it represents the physical and biological properties of diseases in mice (i.e., investigating exploratory behavior or cognitive function) related to spatial learning and memory [9,26].
The Y-maze is commonly used to evaluate spatial learning and working memory; this method is useful for assessing hippocampal damage [57]. A relevant finding was that the NCTQ treatment prevented the reduction of Aβ-induced alternations in the Y-maze task. At the same time, TQ had no effects ( Figure 6B), proving that NCTQ could exert action in spatial and working memory impairment caused by Aβ. In this same sense, Ianiski et al. (2012) [26] reported the neuroprotective effects of nanoencapsulated meloxicam in mice with Aβ 25-35 -induced memory impairment, but not for free drugs, which was attributed to improved brain exposure to the drug. No locomotor alterations were observed since the treatments did not significantly affect the number of arm entries in the Y-maze test ( Figure 6A).
The ORT is a valuable method for assessing LTM in mice, primarily involving hippocampal neuronal circuits in the consolidation of LTM [57]. The TQ and NCTQ prevented the reduction of the exploratory preference behavior associated with LTM ( Figure 7B). These results further support previous studies indicating the potential of tellurium organic compounds as mnemonic enhancers [5][6][7][8]. Additionally, the step-down inhibitory avoidance task, used to assess aversive memory (also related to the hippocampus), demonstrated that the NCTQ protected the memory damage caused by the Aβ-peptide ( Figure 8B). Notably, NCTQ treatment was more effective in protecting against impairment in aversive memory in animals pretreated with Aβ injection than those treated with TQ, thereby corroborating the data observed in the Y-maze test.
Corroborating these findings, our research group recently published a study demonstrating the enhanced effectiveness of anti-inflammatory agents (meloxicam and curcumin) when nanoencapsulated. This novel approach significantly improved the memory of mice exposed to Aβ 25-35 compared to the effects observed with non-encapsulated drugs in the object recognition test [28].
Today, the drugs available for treating AD have several side effects that limit their use, such as hepatotoxicity [58,59]. Our results did not indicate any change in the plasma levels of these liver markers; despite being preliminary, these data, coupled with the absence of observed changes in locomotion or mortality, suggest the low toxicity of both TQ and NCTQ. Our research group recently conducted a study on a compound analogous to TQ, specifically 7-chloro-4-(phenylselanyl) quinoline, which similarly exhibited low toxicity in rodents [9].
This study also has other limitations that should be taken into consideration. Firstly, although the study suggests that nanoencapsulated TQ exerts neuroprotective effects by activating antioxidant pathways and the endoplasmic reticulum, the precise molecular mechanisms have not been fully elucidated. Furthermore, the study did not compare the effectiveness of TQ nanoencapsulation with existing treatments for AD. Moreover, the shortterm evaluation in animal models does not account for the chronic nature of AD, warranting further investigation into long-term effects and potential disease-modifying properties. These limitations highlight the necessity for additional research to comprehensively assess the therapeutic potential of nanoencapsulated TQ to treat AD. Despite its limitations, this pioneering study on nanoencapsulated TQ provides crucial data, unlocking new possibilities for therapies and future research.

Conclusions
We developed NCTQ with satisfactory physicochemical characteristics. The preliminary analysis of C. elegans demonstrated its safety and efficacy in reducing Aβ peptideinduced paralysis rates in worms. The obtained data in the AD model in mice showed that NCTQ treatment did not present hepatotoxicity and attenuated the memory impairment caused by Aβ peptide. Notably, our data demonstrated that TQ nanoencapsulation improved the compound efficacy in rodents, suggesting that it is a promising candidate for further AD treatment assessments.