Intranasal delivery of rapamycin via brain-targeting polymeric micelles for Alzheimer’s disease treatment

Abstract

Alzheimer’s disease (AD) is a long-term neurological disorder associated with neuroinflammation and amyloid-beta (Aβ) aggregation, which leads to a decline in cognitive and behavioral changes. Rapamycin (Rapa) is an immunosuppressive drug effective in preventing organ rejection after a kidney transplant. In the last few years, orally delivered Rapa has emerged as a potential candidate for improving cognitive function in patients with AD. However, it is evident that long-term oral treatment of Rapa causes systemic toxicity, and although controversial, it may even trigger the aggregation of Aβ deposition. This study investigated the therapeutic potential of intranasally delivered brain-targeting polymeric micelles carrying Rapa. We successfully prepared Fibronectin CS1 peptide-conjugated poly(ethylene glycol)-block-poly(D, L-lactic acid) (FibCS1-PEG-b-PLA) micelles carrying Rapa, which were 98.08 ± 1.15 nm in particle size with a polydispersity index of 0.21± 0.01. FibCS1-PEG-b-PLA micelles showed a significant improvement for nasal permeation of Rapa across RPMI-2650 epithelial cells. Behavioral studies such as corner, novel object recognition and Morris Water Maze tests showed promising results towards the improvement of cognitive function in a 3xTg-AD mice model when treated with intranasal FibCS1-PEG-b-PLA micelles carrying RAPA at a dose of 0.2 mg/kg (q4d × 5). The western blot and ELISA results of the brain tissues of 3xTg-AD mice treated with intranasal FibCS1-PEG-b-PLA micelles carrying Rapa showed significant reductions in Aβ and two pro-inflammation markers (e.g. interleukin (IL)-1β, tumor necrosis factor (TNF)-α). Here, we conclude that brain-targeting FibCS1-PEG-b-PLA micelles carrying Rapa were effective in reaching the brain via intranasal route, reduced pro-inflammatory markers and Aβ, and improved cognitive function in AD-induced mice.

Graphical Abstract

Introduction

Alzheimer’s disease (AD) is a neurological disorder that impairs neurocognitive function. The main pathological features found in AD are the presence of amyloid β (Aβ) plaques and tau neurofibrillary tangles in the brain, resulting in neuronal loss and brain atrophy [1]. The presence of Aβ plaque is considered the primary pathological condition of AD. It has been reported that the mechanistic target of rapamycin (mTOR) activity is closely related to Aβ deposition and clearance [2]. Under normal circumstances, the brain rapidly clears insoluble Aβ. However, for patients suffering from AD, the imbalance between the production and clearance of Aβ in the brain causes an increased accumulation of Aβ [3]. For these patients, mTOR is abnormally activated, resulting in the inhibition of Aβ clearance. Therefore, inhibiting mTOR promotes the Aβ clearance. Inversely, the accumulation of Aβ is known to alter the function of mTOR, which is directly linked to learning and memory. Although not clearly elucidatedas of yet, it has been believed that inhibiting mTOR increases the expression of sirtunin 1 (SIRT1), a key regulator of α-secretase, which, as a result, inhibits the production of Aβ. Rapamycin (Rapa) is an mTOR inhibitor produced by Streptomyces hygroscopicus. Caccamo et al reported that AD-induced mice (3xTg) fed with Rapa-containing food (2.24 mg/kg) for 10 weeks showed a significant reduction in Aβ immunoreactivity of neurons in the CA1 region of the hippocampus [4]. Chronic inflammation is another major factor in the progression of AD. Nuclear factor-κB (NF-κB) is a key mediator known for the anti-inflammatory effect of Rapa. Rapa exerts anti-inflammatory effects by down-regulating p65, interleukin (IL)-1β, tumor necrosis factor-α (TNF- α) and other factors associated with NF-κB. IL-6 takes a complex role by stimulating the synthesis of Aβ precursor protein while exerting a neuroprotective effect by activating the phagocytic activity of microglia to degrade Aβ. Although controversial, it has been reported that Rapa upregulates IL-6 expression in astrocytes in a Parkinson’s disease mouse model, showing a neuroprotective effect [5]. It is evident that Rapa improves AD-induced cognitive dysfunction [6]. The restoration of mTORC1 activity after Rapa treatment demonstrated a significant effect on cognitive performance in vivo [6]. Rapa is reported to enhance synaptic plasticity, which plays a key role in the development of the nervous system, learning and memory, and cognitive function [7–9]. Rapa also increases synaptic protein expression by increasing mitochondrial autophagy and preventing cytochrome C-mediated apoptosis [7]. Rapa is known to exert a major effect on aging in rodents [10,11]. The NIA’s intervention testing program (ITP) identified Rapa as the compound that extends lifespan in rodents when given orally via the mouse chow (14 ppm) and the effect was noticeable across both males and females. The primary mechanism of extending lifespan appears to be the inhibition of lethal neoplastic disease [12].

Although Rapa has been proven effective in AD, several questions remain to be addressed [4]. First, Rapa lacks specific targeted effects and as a result potential side effects may occur. There are a few instances regarding Rapa’s adverse effects associated with oral Rapa monotherapy. The most common side effects at the highest dose tested (20 mg/week for 6–16 weeks) were mouth ulcers, headache, fatigue, and neutropenia [13]. As per Shi et al, it was found that the long-term oral use of rapamycin for 2–3 months inhibited mTOR activity and reduced TREM2 expression in microglia of AD mice, reducing the uptake and clearance of Aβ by microglia and aggravating AD-like pathological changes in the brains of 5xFAD AD mice [14]. The animals received Rapa via daily diet at 143 ppm encapsulated Rapa (14 ppm active Rapa; 2.24 mg/kg, 5 g/day). Additional side effects such as glucose intolerance, diabetes, and immunosuppression resulting from long-term use of Rapa are of concern to AD patients [4]. In this study, to address the non-targeted effect of Rapa and its side effects owing to the long-term oral use, a low dose (0.2 mg/kg) of Rapa was delivered to AD-induced mice intranasally via brain-targeting polymeric micelles. We hypothesized that brain-targeting polymeric micelles carry Rapa effectively to the brain via nose-to-brain route at the lower dose of Rapa, and brain-targeting polymeric micelles carrying Rapa offer the greater therapeutic efficacy on short-term treatment regimen, thereby reducing systemic side effects. The current AD therapeutics are mainly delivered via oral and parenteral routes. The major drawback of these approaches is poor drug concentration, reduced therapeutic efficacy, and greater side effects of systemic toxicity [15].

Intranasal “nose-to-brain” drug delivery route has been discovered as an alternative approach that addresses these issues and treats the neurological disorders such as migraine (e.g. Onzetra Xsail^®) and Parkinson’s disease at greater efficacy and improved patient compliance [16]. Intranasally delivered therapeutics can bypass the blood brain barrier (BBB) and reach the central nervous system (CNS) via the olfactory and trigeminal neural pathways [17]. Several pathways for nose-to-brain delivery have been proposed based on pre-clinical studies including the olfactory pathway. To date, olfactory pathway, transporting drugs directly to the brain from the nasal cavity along the olfactory and trigeminal nerves, is considered to be the most direct “nose-to-brain” pathway. Once inhaled via nasal pathway, the formulation enters the nasal vestibule where vibrissae, turbulence, and mucosal contact filter particles > 12 μm in size [18]. It then arrives in a respiratory region which is lined with ciliated pseudostratified columnar epithelium and contains the highly vascularized nasal turbinates. The formulation then reaches the olfactory region located on the roof of the nasal cavity. The olfactory nerve (the maxillary branch of the trigeminal nerve) entering the CNS through the pons provides the formulation with direct CNS access by evading the BBB. Such direct CNS access can be achieved via intracellular and extracellular pathways [19]. The intracellular pathway begins with endocytosis by olfactory sensory cells, followed by axonal transport to the synaptic clefts in the olfactory bulb. The drug undergoes exocytosis onto the olfactory bulb, where neurons projecting to brain regions repeat this process. In the extracellular pathway, drugs are transported directly into the cerebral spinal fluid across the paracellular space in the nasal epithelium, then pass through the perineural space to reach the subarachnoid space of the brain.

To further improve the efficacy of drug delivery, nanoparticles have been utilized as the emerging vehicles. As the axonal transport of rodents is < 100 nm, nanoparticles in average particle size < 100 nm were detected inside olfactory epithelial cells [20]. As the nasal mucous membrane is negative in charge, zeta potential of the nanoparticle at > 30 mV appears to stabilize the nanoparticles, avoiding agglomeration due to the electrostatic repulsion within the nasal cavity and facilitating the interaction between the nanoparticles and the mucosal cells. It is noteworthy that higher positive zeta potential may cause greater toxicity in nasal mucous membrane.

Polymeric micelles are spherically shaped nanoparticles composed of amphiphilic block copolymers, featuring a hydrophobic core and a hydrophilic shell [21]. The most widely adopted hydrophilic block is PEG and the dense brush of PEG on the surface of the micelles ensure that the micellar network embedded with hydrophobic compounds are soluble in water. The modified PEG such as N-Hydroxysuccinimide-PEG (NHS-PEG), permits the versatile conjugation of targeting moieties on the surface of the PEG-based polymeric micelles. The hydrophobic blocks impart unique features and predictive properties of polymeric micelles for drug delivery. For example, poly(ε-caprolactone) (PCL) is a semicrystalline polymer with a melting temperature (T_m) of 55°C, whereas PLA is amorphous with a glass transition temperature (T_g) of 50°C. PEG-b-PCL is expected to offer a greater stability for the loaded drugs in a liquid form due to the semicrystalline core of the micelles, however, lyophilization of PEG-b-PCL micelles is rather challenging due to the fragile semicrystalline core. PEG-b-PLA is expected to change the average particle size depending on the drug loading; however, it is easier to prepare lyophilized PEG-b-PLA micelles with greater stability.

In this study, we prepared fibronectin CS1 peptide-conjugated poly(ethylene glycol)-block-poly(D,L-lactic acid) (FibCS1-PEG-b-PLA) micelles carrying Rapa. The primary focus of the study was to demonstrate the effectiveness of the tumor-targeting micellar formulation in reaching the brain via intranasal route, reducing pro-inflammatory markers and Aβ and improving cognitive function in AD-induced mice. Most of the studies evaluating the efficacy of Rapa on improving cognitive functions and extending lifespan have been conducted with the long-term oral administration of Rapa. In this study, we highlighted the feasibility of short-term intranasal administration of brain-targeting micelles carrying Rapa for improving cognitive behavior including motor coordination, recognition memory, and special learning and memory in an AD-bearing mouse model.

Materials and Methods

Preparation of FibCS1-PEG-b-PLA micelles carrying Rapa

PEG-b-PLA micelles carrying Rapa were prepared using a solvent evaporation method modified from the previous report [22]. Briefly, NHS-PEG_5k-b-PLA_16k (10 mg) and Rapa (1 mg) were dissolved in 3 mL acetone, followed by a drop-wise addition of 1.0 mL of pre-warmed phosphate buffer (10 mM, pH 8.4) at 60 °C with vigorous mixing. Acetone was then evaporated under reduced pressure using a rotatory evaporator at 60 °C. Non-encapsulated Rapa molecules were removed by centrifugation for 5 min at 10,000 rpm, followed by filtration through a 0.22 μm nylon syringe filter. The content of Rapa encapsulated in micelles was quantified by HPLC analysis. The chromatographic separation of Rapa was accomplished using an Agilent system 1220 Infinity LC equipped with a UV detector and Extend C₁₈ ODS column (100 mm × 3.0 mm, 3.5 μm) (Fig 3–4). A mixture of acetonitrile: 0.1 % ortho phosphoric acid in water (75:25) was used as a mobile phase at a flow rate of 0.7 mL/min, and the injection volume was 10 μL. The Openlab 3.0 software was used to monitor and process the output signal. The temperature of the column was kept at 40°C, and detection was made at 277 nm. Rapa was eluted within a 5 min run time, giving a sharp peak for the drug at 1.9 ± 0.2 min. The method was linear over the working concentration range of 1 to 1000 μg/mL. Fibronectin CS1 peptide (100 μg) was dissolved in phosphate buffer (10 mM, pH 8.4), and the peptide solution was added to the PEG-b-PLA micelles carrying Rapa. Then, the mixture was kept at room temperature for 24 h while stirring. The mixture was then centrifuged and filtered (Amicon filter MWCO 10 kD) at 10,000 rpm for 30 min to remove unconjugated, free peptides. The content of the unconjugated peptide was quantified by HPLC analysis with the above chromatographic system with Mobile Phase A: 0.1 % Ortho phosphoric acid in water and Mobile Phase B: acetonitrile. Gradient starting at 90% Mobile phase-A for 10 min, followed by a linear extension to 90% Mobile phase B over 15 min at a flow rate of 0.7 mL/min at 40 °C at 209 nm. The percentage entrapment efficiency (%EE) and Rapa-loading (%DL) for conjugated FibCS1-PEG-b-PLA micelles were calculated based on the equations below.

Figure 3.

The cumulative Rapa release profile for FbCS1-PEG-b-PLA micelles carrying Rapa in nasal simulated fluid at pH 6.5.

Figure 4.

The cumulative Rapa permeated across the RPMI-2650 epithelial cells after 3 h treated with Rapa solution or FibCS1-PEG-b-PLA micelles carrying Rapa at 11 μg/mL (A) and (B) 23 μg/mL and the calculated apparent permeability for Rapa solution or FibCS1-PEG-b-PLA micelles carrying Rapa at 11 μg/mL (C) and (D) 23 μg/mL.

$$\text{EE(\%w/w)}=\frac{\text{Drug encapsulated in FibCS1}-\text{PEG}-b-\text{PLA}}{\text{Total drug added}}\times 100$$ 1

$$\text{DL(\%)}\frac{\text{Drug encapsulated in FibCS1}-\text{PEG}-b-\text{PLA}}{\text{Total}(\text{Drug}+\text{polymer})\text{added}}\times 100$$ 2

Physicochemical characterizations

The z-average diameters and polydispersity indices (PDI) of FibCS1-PEG-PLA micelles carrying Rapa and PEG-PLA micelles carrying Rapa at 25 °C were determined by dynamic light scattering (DLS) measurements using a Zetasizer Nano-ZS (Malvern Instruments, UK) at a 173° detection angle and using a He−Ne ion laser incident beam (4 mW, λmax = 633 nm). This system employed a photodiode detector that captures the scattering intensity and was subsequently analyzed using a digital autocorrelator. The analysis was performed based on the Stokes=Einstein equation to produce a correlation function. The Z-average was derived from a Cumulants analysis of the measured correlation curve, wherein a single particle size was assumed, and a single exponential fit was applied to the autocorrelation function. The samples used for this process were previously diluted (1:99 v/v in buffer) by considering the count rate falls within the instrument’s optimal range (e.g., 100–300 kcps). The samples were diluted with Milli-Q water to adjust to a suitable scattering intensity and were well dispersed before the measurements. All the experiments were performed in triplicates (n=3).

For cryo-Electron Microscopy (cryo-EM) experiment, all grids were treated for 35 sec in Fischione Nanoclean 1070 (70% power) with a mixture of Argon (75%) and Oxygen (25%). Cryo-EM grids were prepared in Vitrobot Mark IV (Thermo Fisher Scientific) at 21 °C with the following settings: the relative humidity 100%, wait time 5 sec, and blot time 4 sec, blot force 4. Three microliters of sample solution were pipetted onto a freshly treated lacey carbon grid. The sample solution was incubated on EM grid, blotted with filter paper before being plunged into liquid ethane that was pre-cooled by liquid nitrogen. The cryo-EM grids were then transferred to and stored in liquid nitrogen. The cryo-EM grid was transferred in liquid nitrogen into a Gatan 626 cryo-specimen holder that was then inserted into the microscope stage. The specimen temperature was maintained at about −170 °C during data collection. Cryo-EM images were taken with CETA camera (Thermo Fisher Scientific, Waltham, MA) and in the low-dose mode of Titan Halo TEM operating at 300 kV (Thermo Fisher Scientific, Waltham, MA).

The iodine probe methodology was implemented to determine the critical micelle concentration (CMC) of FibCS1-PEG-PLA micelles. A KI/I2 solution was prepared and placed in the dark. Following the preparation of a range of polymeric dilutions with varying concentrations, 25 μL of KI/I2 solution was transferred to each polymeric dilution, and the mixture was incubated for 12 h in the dark. Further, UV-vis spectroscopy was used to determine the absorbance of each concentration at 366 nm (Double beam UV-Vis spectrophotometer, Perkin Elmer). A significant rise in the value was regarded as the CMC of PEG_5k-b-PLA_16k. The graph was created on the logarithm of PEG_5k-b-PLA_16k concentration (x-axis) versus absorbance (y-axis) coordination.

The drug release property was assessed using a dialysis method. Briefly, FibCS1-PEG-PLA micelles carrying Rapa were filtered (0.22 μm) and put into the dialysis cassette (2000 MWCO, Thermo Scientific, Waltham, MA, USA). Then, the dialysis cassettes were submerged in a beaker filled with 100 mL of nasal simulated fluid (pH 6.5). At predetermined intervals of 0.5, 1, 2, 4, 8, 12, and 24 h, 1 mL of the sample was removed from the cassette and replaced with fresh nasal simulated fluid. The validated HPLC method was then used to quantify Rapa from the collected samples diluted with the mobile phase.

Nasal permeability study

RPMI-2650 epithelial cells were seeded on polyester Transwell^™ inserts (12-well plates, area 0.33 cm², 0.4 μm pore size) [23]. In short, the air-liquid interface model was created by plating 100 μL of cell suspension at 8.5 × 10⁴ cells/Transwell^™ polyester inserts, and monolayer development was ensured by adding 600 μL of medium to the basolateral chambers. Plates were then incubated for 24 h at 37°C, 95% air humidity, and 5% CO₂. The integrity of the cell layer was evaluated using transepithelial electrical resistance (TEER) testing. This method assesses the ionic resistance over the cell layer during various phases of differentiation and maturation. To verify the formation of the cellular monolayer, the TEER of RPMI-2650 cell layers was measured every 2–3 days during growth using an Endohm^® chamber and an EVOM^® resistance meter (World Precision Instruments, Sarasota, FL, USA). As the pore size and concentration of the cells at membranes affect the TEER, the technique of measurement involves evaluating the blank resistance (RBlank) of the membrane alone (without cells) as well as the resistance across the cell layer on the insert membrane $\left({\text{R}}_{\text{Total}}\right)$. The following formulas were used to get the final layer of cell resistance.

$${\text{R}}_{\mathit{\text{Cells}}\Omega }={\text{R}}_{\text{Total}}\text{-}{\text{R}}_{\text{Blank}}$$ 3

$${\text{TEER}}_{\text{Cells}}=R_{\text{Cells}\Omega }\times M_{\text{area}}\left({\text{cm}}^2\right)$$ 4

The permeation examination was conducted using RPMI-2650 cells monolayer after 16 days with suitable TEER plateau values (80–85 Ω.cm²). Cells were treated with the conjugated formulation that contained Rapa. Prior to treatment, the EMEM medium in the Transwell^™ basal chamber (acceptor) was removed from each well and changed out with new media. Rapa solution (in 0.1% DMSO) and FibCS1-PEG-b-PLA micelles carrying Rapa were added to the apical chamber (donor) of Transwell^™ inserts. The mixture was then incubated at 37 °C, 5% CO2, and 95% air humidity. Following treatment, a 50 μL sample was removed from the acceptor chamber once every 3 h and replaced with the equivalent amount of new media.

$$\mathit{\text{Papp}}=\frac{dQ}{dt}\times \frac{1}{A*C}$$ 5

$$Jss=\mathit{\text{Papp}}\times C$$ 6

where dQ/dt is the amount of Rapa present in the apical compartment a function of time, A is the membrane surface area (cm²), C is the initial Rapa concentration in the apical chamber, and J_ss is steady-state flux.

Immunofluorescence analysis

For immunofluorescence analysis, glass coverslips (12 mm, Fisher Scientific, Pittsburgh, PA) were coated with 500 μL poly-D-lysine (at 0.1 mg/mL/0.1M boric acid) on a 24-well cell culture plate and allowed to sit at ambient temperature for 1 h. The PDL-coated coverslip was seeded with 1×10⁵ cells/mL of Sim-9 CRL-3265 microglia cells. The cells were then treated with 100 ng/mL of LPS and received a 48h treatment with free Rapa at 25 μM or with FibCS1-PEG-b-PLA micelles carrying Rapa at 25 μM. Then, the cells were fixed for 10 min at room temperature using ice-cold 4% paraformaldehyde. The coverslips were then moved to a humidified chamber after being cleaned once with ice-cold 1x PBS. Then coverslips were soaked in a permeabilization solution (0.1% Triton x-100/PBS) for 15 min at ambient temperature. After decanting the solution, the coverslips were left to sit at room temperature for 30 min in a blocking buffer (5% heat-inactivated goat serum/0.3% Triton x-100/PBS). The coverslips were blocked and then incubated overnight at 4°C in the primary antibody diluted in the antibody dilution buffer (1% bovine serum albumin/0.3% Tween 20/PBS). The Cy3-labeled secondary antibody was then diluted in the antibody dilution buffer and applied to the coverslips for 1 h at room temperature after the coverslips had been cleaned 4 times with 1x PBS. Lastly, a drop of the mounting medium containing DAPI was applied to the coverslips before placing them on the glass microscope slides. The slides were kept at 4°C and covered with nail polish the following day to keep the samples from drying out. A Zeiss Axioplan 200M upright fluorescent microscope, an AxioCam electronic camera, and an AxioVision 4.8.2.0 program were used to capture fluorescent images. A Zeiss Axiovert 25CFL inverted microscope with a Luminera Infinity 3–1 CCD camera and Infinity collection software (version 6.5.4) were used to take phase contrast images.

Preparation of cell lysates

Free Rapa at 25 μM and FibCS1-PEG-b-PLA micelles carrying Rapa at 25 μM were added to microglia cells (1×10⁵ cells/mL) in the presence of LPS (100 ng/mL) in order to assess their impact on CD11b expression in LPS-induced neuroinflammation. In a T25 flask, the cells were initially grown at a density of 1×10⁵ cells/mL. Following confirmation of adhesion, cells were treated for 30 min with 100 ng/mL of LPS. The treated cells were incubated for 48 h at 37°C with 5% CO₂. The scraped cells were centrifuged at 5,000 rpm for 10 min at 4°C. After discarding the supernatant, 50 μL of 1x RIPA lysis solution supplemented with 1 mM phenylmethylsulfonyl fluoride was added to the pellet. After being moved to labeled microcentrifuge tubes, the resuspended cells were placed on ice for 30 min. After that, the cells were centrifuged for 20 min at 4°C at 12,000 rpm. Cell debris was eliminated, and the cell lysates were kept for further western blot analysis at −80°C.

Sample preparation for Western Blots

Following the treatment, the cells were rinsed twice with ice-cold Dulbecco’s phosphate buffered saline (DPBS) and lysed using radio-immunoprecipitation assay (RIPA) buffer (Thermo Scientific, Waltham, MA) containing 1% Triton X 100, phosphatase inhibitor cocktail, and protease inhibitor cocktail (Thermo Scientific, Waltham, MA). Cells were lysed and centrifuged for 5 min at 14,000 g. The supernatant was collected, and a BCA assay was performed for protein measurement (Protein Assay Kit, Thermo Fischer, Waltham, MA).

Similarly, brain tissue was homogenized using T-PER protein extraction buffer (Thermo Scientific, Waltham, MA) containing phosphatase and protease inhibitors. Collected brain homogenates were centrifuged for 10 mins at 14000g, and the supernatant protein concentration was measured.

Western Blot Analysis

Cell lysate or tissue homogenate was denatured using 4X Laemmli buffer for 5–7 mins at 95°C. Samples were either used for conducting western blots the same day or kept at −80°C for later use. Primary antibodies used were Anti-amyloid precursor protein (Abcam, Cambridge, UK), IL-6 monoclonal antibody (Invitrogen, Waltham, MA), TNF alpha polyclonal antibody (Invitrogen, Waltham, MA), Beta-actin monoclonal antibody (Proteintech, Rosemont, IL). Western blotting was performed using either cell lysate or brain tissue homogenate. Blots were transferred to a polyvinylidene difluoride (PVDF) membrane and blocked using the appropriate blocking buffer, either 5% w/v milk or 5% w/v BSA in TBST buffer, before being incubated with primary antibodies overnight on a shaker at 4°C. The proteins were then incubated with species-specific secondary antibodies anti-mouse (Cell Signaling, Danvers, MA) and anti-rabbit (Cell Signaling, Danvers, MA) for 2 h at room temperature. Clarity^™ Western ECL Substrate (Bio-Rad, Hercules, CA) was used to detect protein bands, and a ChemiDoc imaging system (Bio-Rad, Hercules, CA) was used to visualize the bands. The quantification of the bands was done using the ImageJ software.

Enzyme-linked immunosorbent assay (ELISA)

ELISA was performed using either cell lysate or brain tissue homogenate. The levels of Aβ42 (Invitrogen, Waltham, MA), TNF-α (Invitrogen, Waltham, MA), and IL-6 (Biomatik, Ontario, Canada) were determined according to the manufacturer’s instructions [24]. In brief, each sample was measured against a standard curve. Samples were diluted using the provided standard diluent buffer in the kit (1:4 dilutions) and incubated for varying times at room temperature, allowing the samples to bind the capture antibody pre-coated to each well. The sample plate underwent extensive washing and varying incubation times at either room temperature or 37°C with a biotin-conjugated detection antibody (same as the capture antibody). After the removal of the excess antibody, horseradish peroxidase-labeled streptavidin (SAV-HRP) was added and incubated again. At last, the plate was washed, and a tetramethylbenzidine (TMB) substrate was added to produce a color change. Results were interpreted based on the intensity of color that was directly proportional to the antigen of interest in the sample. The standards provided a linear curve, and the best-fit line determined by linear regression was used to calculate the concentration of Aβ42, TNF-α, and IL-6 in samples.

Fluorescence optical imaging

The Fluobeam^® 800 system (Fluoptics, Cambridge MA) was used to detect the fluorescence signals of DiR in 3xTg AD mice in a real-time manner. It consists of a filtered camera with both a white light and near-infrared (NIR) light source and a computerized screen that displays NIR fluorescent images in real-time at 700–800 nm wavelength. The fluorescence signal of DiR molecules appeared to be white whereas non-fluorescent parts of the animals remained dark (in black). After imaging live animals, they were sacrificed and brain and nasal tissues were removed and imaged using IVIS (Lumina LT series III, Perkin Elmer, Waltham, MA). The biodistribution of intranasally delivered FibCS1-PEG-b-PLA micelles carrying DiR (3.5 μg/mouse, n=4 female) and non-targeted PEG-b-PLA micelles carrying DiR (3.5 μg/mouse, n=4 female) were monitored in 3xTg-AD mice (12–15-week-old). Brain and nasal tissues, removed at 24, 48, 96, and 120 h post intranasal administration, were scanned. All images were acquired by back-thinned, back-illuminated grade 1 CCD camera with the following parameters: exposure time = 1 sec; binning = medium; f/stop 2. Filter sets were fixed with the following parameters for DiR: excitation at 745 nm and emission at 800 nm. Acquired images were measured and analyzed with Living Imaging^® software. The distribution of DiR in the brain and nasal cavity was quantified by average radiant efficiency, total photons per second per square centimeter per steradian in the irradiance range (microwatts per square centimeter): [p/s/cm²/sr]/[μW/cm²]. The brain-specific accumulation of DiR was evaluated by signal-to-noise ratio: average radiant efficiency at the region of interest (ROI) was drawn around the brain to calculate the average radiant efficiency.

Animal behavior studies

FibCS1-PEG-b-PLA micelles carrying Rapa were delivered intravenously (n=4 female) at 20 mg/kg or intranasally (n=5 female) at 0.2 mg/kg in 3xTg-AD mice (12–15-week-old) [25]. For control, 3xTg-AD mice received empty FibCS1-PEG-b-PLA micelles (n=5 female) (12–15-week-old). The 3xTg-AD mice received a total of 5 doses of treatments every 4 days (q4dx5). Animals were anesthetized using isoflurane delivered at a flow rate of 2.5 mL/min with oxygen for 2 min to ensure minimal movement and stress during the procedure. Immediately after anesthesia induction, the formulation was administered intranasally using a calibrated 10 μL pipette fitted with sterile, low-retention pipette tips. The mice were held in a supine position, and the formulation was gently instilled into each nostril in small aliquots to allow for natural inhalation and optimal nasal absorption.

The temporal sequence of behavioral testing was performed based on the degree of stress in each test, with the most stressful one at the end: the CT and NOR at the initial level and the MWM on days 17–21 (Fig. 10). Tests were recorded by a digital USB camera and processed with the EthoVision XT software for evaluation of data. All the material involved in the behavioral testing was thoroughly cleaned with 70% ethanol between trials to ensure the absence of olfactory cues.

Figure 10.

Schematic illustration of the in vivo behavior studies: AD mouse model (3xTg) was given a total of 5 doses (20 mg/kg IV and 0.2 mg/kg IN) of treatments every 4 days (q4dx5). The temporal sequence of behavioral testing was practiced based on the degree of stress in each test, with the most stressful one at the end: the corner test (CT), the Novel object recognition (NOR) test, and the Morris water maze (MWM) test.

Corner test (CT):

The CT was performed by placing each mouse in the middle of the arena (square 40 cm × 40 cm × 40 cm) for 30 sec. Horizontal activity distance traveled and the number of visits to the corner were recorded for evaluation.

Noval Object Recognition (NOR):

In this study, each mouse was placed in a square box (square 40 cm × 40 cm × 40 cm). To habituate the mice, each mouse was placed in the center of the arena for 5 min. After 5 min, the mouse was taken out and put it in the holding cage. The box was thoroughly cleaned with 70% ethanol each time when new animal was placed. For training 1(T1), two similar objects were placed in the inverse direction of the arena NE corner and SW corner. For testing (T2), one of the two objects was replaced with a new object. Each mouse was given 10 min to explore the objects. For both T1 and T2, the score was recorded for the first 5 min. If the mouse did not meet the minimum exploration time of 20 s for both objects, scoring continued past 5 min until total exploration exceeded 20s.

Morris water maze (MWM) test:

In the Morris water maze (MWM) paradigm, the capacity to locate a hidden platform in a pool of opaque water stained with non-toxic white tempera was used to evaluate hippocampus-dependent learning and memory. The MWM took place in a 1.5-meter-diameter (44 cm depth) circular pool. The MWM test consisted of three paradigms:

Visible platform: The visible platform stage took place on the first day, with a colored flag highlighting the platform. To make sure the mice understood the platform’s existence, the animals conducted 4 trials with a 10-min break in between.

Acquisition: The platform was hidden about 5 mm beneath the water’s surface throughout the acquisition phase, and cues were incorporated to enable spatial orientation. There were 4 trials every day for 5 days. The animals were to arrive at the platform after becoming oriented with the cues. When the animals did not locate the platform in the 60s, they were manually led to the platform. To guarantee distinct sequences throughout stages and days, starting points/locations were randomly arranged.

Probe: The platform was removed for the final probe. In order to accurately assess long-term memory, this phase involved 60 sec of navigation without the platform. The animals began moving from the furthest point to the typical platform site. To avoid hypothermia, animals were placed under a heating lamp (red-light, 100 W) after being taken out of the water. The distance moved in the arena was measured across trials during the acquisition stages. The time traveled within a zone arena/center point on the former platform location was measured in the final probe.

Hematoxylin-eosin (H&E) staining

Brain tissues were preserved in a 10% (v/v) phosphate-buffered formalin solution and incubated overnight to perform H&E staining. Sections of brain tissue fixed in paraffin (5 μm thick) were utilized. In short, 5-μm sections were deparaffinized and dehydrated using xylene and various alcohol concentrations before being rehydrated with water. Hematoxylin was applied for 3 min, followed by water rinsing, 70% acid-alcohol differentiation, eosin staining for 30 sec, 95% ethanol rinsing, absolute ethanol dehydration, and xylene clearing for 15 min before cover-slipping with Permount^® (Fisher Scientific, Waltham, MA). Photographs of H&E-stained slides were captured in 6–10 random locations at 40x magnification under a light microscope (Life technologies, EVOS FL Auto Microscope, Waltham, MA).

Statistical analysis

Significance is defined as p < 0.05. Levels of significance are identified as follows: * p < 0.05, ** p < 0.01, *** p < 0.001 and **** p < 0.0005. All statistical analyses were conducted using GraphPad Prism V. 9.00 X (San Diego, USA). Unpaired two-tailed t-tests were used to compare two groups: to compare more than two groups, one-way analysis of variance (ANOVA).

Results and Discussion

Physicochemical characterizations

PEG-b-PLA micelles carrying Rapa (pre-conjugation of peptide) were 87.6 ± 1.24 nm in average particle size (PDI 0.31 ± 0.01), whereas FibCS1-PEG-b-PLA micelles carrying Rapa (post-conjugation of fibronectin CS1 peptide) were 98.08 ± 1.15 nm (monomodal) in average particle size (PDI 0.21 ± 0.01) (Fig. 1 and 2). Fibronectin CS-1 peptide, a water-soluble peptide that helps in the reduction of amyloid plaque accumulation in the brain’s hippocampal region, was used as a targeting moiety. The conjugation of fibronectin CS1 peptide contributed to the slight increase in particle size. Zeta potential of PEG-b-PLA micelles carrying Rapa (pre-conjugation) was −3.58 ± 0.66 mV whereas that of FibCS1-PEG-b-PLA micelles carrying Rapa was −12.59 ± 0.87 mV. The change in zeta potential is attributed to the negative charges of fibronectin CS1 peptide, H-Glu-Ile-Leu-Asp-Val-Pro-Ser-Thr-OH, at pH 8.5,At this pH aspartic acid (Asp; pKa of 3.9) and glutamic acid (Glu; pKa of 4.6) have negative charges as their side chain carboxylic group is deprotonated. The percent drug encapsulation efficiency of FibCS1-PEG-b-PLA micelles carrying Rapa was 94.23 ± 3.4%, and the percent drug loading of those was 9.06 ± 2.12%. The amount of peptide conjugated to the surface of FibCS1-PEG-b-PLA micelles carrying Rapa was approximately 35 μg per 10 mg of PEG-b-PLA.

Figure 1.

Size Distribution by Intensity for FibCS1-PEG-b-PLA micelles carrying rapamycin (A) and cryo-TEM analyses for FibCS1-PEG-b-PLA carrying rapamycin (left) and FibCS1-PEG-b-PLA micelles carrying DiR (right) (B)

Figure 2.

The plot of absorbance of iodine vs. log concentration of PEG_5k-b-PLA_16k determined by iodine probe method.

As shown in Fig. 2, the CMC performed by using iodine probe method was 1 mcg/mL. The low CMC value indicates high thermodynamic stability of micelles and strong hydrophobic strength of the micellar core, which may lead to greater entrapment of hydrophobic compounds such as Rapa. The low CMC is attributed to the greater length of hydrophobic PLA blocks from PEG_5k-b-PLA_16k, which promotes self-assembly of the amphiphilic copolymers into micelles.

As shown in Fig. 3, the drug release pattern assessed in nasal simulated fluid demonstrated that 73.21 ± 6.5% of Rapa was released from FbCS1-PEG-b-PLA micelles within 8 h whereas free Rapa (log P 4.3) dissolved in 0.1% DMSO solution showed the rapid precipitation of Rapa within 1 h (data not shown).

Nasal permeability study

The nasal permeation across the monolayer RPMI-2650 cells was carried out in two groups: Rapa solution (0.1% DMSO) and FibCS1-PEG-b-PLA micelles carrying Rapa. Transepithelial resistance (TEER) tests were conducted every 2–3 days until day 16 when TEER readings plateaued, noting that day 1 TEER value was 35.2 Ω*cm², while day 16 TEER value was 85.4 Ω*cm². FibCS1-PEG-b-PLA micelles carrying Rapa transported across RPMI-2650 cell monolayers with improved permeability compared to Rapa solution (Fig. 4). Following a 3 h incubation time, 11.27 ± 0.12 μg/mL and 19.78 ± 0.22 μg/mL of Rapa crossed the membrane when incubated with FibCS1-PEG-b-PLA micelles carrying Rapa at 11 (half of IC₅₀) and 23 (IC₅₀) mcg/mL, respectively. It is noted that the IC₅₀ value of FibCS1-PEG-b-PLA micelles carrying Rapa on C8D1A astrocytes was 22 ± 0.5 μM and that on CRL-3265 microglial cells was 25 ± 0.3 μM (Supporting information, Fig. S1, Table S1). When incubated with free Rapa at 11 and 23 μg/mL, 5.61 ± 0.01 μg/mL and 8.62 ± 0.15 μg/mL of Rapa crossed the membrane. The apparent permeability (P_app) calculated for Rapa from FibCS1-PEG-b-PLA micelles (2.43 cm/h*10⁴) was notably higher than that for Rapa from 0.1% DMSO solution (1.38 cm/h*10⁴). Presumably, polymeric micelles carrying Rapa crossed the lipid bilayer membrane at a greater efficacy due to the improved lipophilicity-to-hydrophilicity balance. It is assumed that the lipophilic nature of Rapa prevented it from permeating the nasal membrane whereas Rapa carried via FibCS1-PEG-b-PLA micelles was able to permeate the membrane and detected in the reservoir.

Neuroinflammation analyses

Cluster of differentiation 11b (CD11b) is a transmembrane glycoprotein, a reliable marker of activated microglia in AD. CD11b is primarily present on the surface of microglia cells, immune cells, and macrophages. In the study, microglia activation was done by incubating microglial cells with lipopolysaccharide (LPS) (100 ng/mL) for 30 min, which generated inflammation and released cytokines. The activation of CD11b was followed by the treatment of free Rapa at 25 μM and FibCS1-PEG-b-PLA micelles carrying Rapa at 25 μM. The activated microglia cells released the pro-inflammatory marker, and Rapa appeared to assist in the reduction of the CD11b expression. LPS-treated microglia cells showed the elevated CD11b expression from 15 ± 2.1% (control) to 78 ± 1.2% (Fig. 5). LPS-treated microglia cells treated with free Rapa showed a notable reduction of CD11b expression at 38 ± 4.6%. LPS-treated microglia cells treated with FibCS1-PEG-b-PLA micelles carrying Rapa showed the most significant reduction of CD11b expression at 20 ± 0.86%.

Figure 5.

The immunofluorescence analysis and the quantification of the CD11b expression on LPS-treated microglia cells after being treated with free Rapa or FibCS1-PEG-b-PLA micelles carrying Rapa at 25 μM.

Tumor Necrosis Factor-alpha (TNF-α) and Interleukin-6 (IL-6) are key pro-inflammatory cytokines that play a significant role in neuroinflammation, contributing to the activation of astrocytes and microglia in the brain. The elevated levels of these cytokines are associated with neurodegenerative process, which leads to neuronal damage and dysfunction in conditions of AD. The most significant reductions in both pro-inflammatory markers, TNF-α and IL-6, were observed on LPS-treated microglia cells when treated with FibCS1-PEG-b-PLA micelles carrying Rapa at 25 μM. (Fig. 6) (western blot) that free Rapa at 25 μM was shown to decrease IL-6 (from 0.273 ± 0.23 pg/mL to 0.230±0.22 pg/mL) and TNF-α (from 0.952 ± 0.23 pg/mL to 0.490 ± 0.02 pg/mL) cytokines compared to the LPS-treated microglia cells. FibCS1-PEG-b-PLA micelles carrying Rapa at 25 μM significantly downregulated IL-6 (from 0.273 ± 0.23 pg/mL to 0.195 ± 0.12 pg/mL) and TNF-α (from 0.952 ± 0.23 pg/mL to 0.230± 0.12 pg/mL) cytokines compared to the LPS-treated microglia cells.

Figure 6.

Western blot analysis on LPS-treated microglia cells for protein expression and relative band density of pro-inflammatory markers TNF-α and IL-6.

ELISA (Enzyme-linked immunosorbent assay) in the (Fig. 7) also shows that free Rapa at 25 μM was shown to decrease IL-6 (from 600.35 ± 13.3 pg/mL to 524.24 ± 10.4 pg/mL) and TNF-α (from 59.76 ± 4.1 pg/mL to 46.96 ± 7.1 pg/mL) cytokines compared to the LPS-treated microglia cells. FibCS1-PEG-b-PLA micelles carrying Rapa at 25 μM significantly downregulated IL-6 (from 600.35 ± 13.3 pg/mL to 338.24±7.23 pg/mL) and TNF-α (from 59.76 ± 4.1 pg/mL to 19.73±1.9 pg/mL) cytokines compared to the LPS-treated microglia cells.

Figure 7.

ELISA analysis on LPS-treated microglia cells for protein expression of pro-inflammatory markers TNF-α (left) and IL-6 (right).

Nose-to-brain micellar drug delivery in vivo

FibCS1-PEG-b-PLA micelles carrying DiR (3.5 μg /mouse) were delivered intranasally to 3xTg AD-bearing mice. (Fig. 8) demonstrates the intensity of fluorescence (Ex 745 nm Em 800 nm) at 1, 4, 24, and 48 h post intranasal delivery of FibCS1-PEG-b-PLA micelles carrying DiR. The most prominent fluorescence was detected around the nostrils within 1 h. As time progressed, presumably, the formulation traveled towards the brain region, showing the reduced fluorescence signals in the nostrils. Later, at 24 and 48 h, there were much weaker fluorescence signals detected in the nasal cavities. It is noted that Fluobeam^® was not able to detect and capture the DiR fluorescence signals through the live mouse skull in vivo. To visualize the DiR molecules traveled to the brain, 3xTg AD-bearing mice were sacrificed at 24, 48, 96, and 120 h post intranasal administration of FibCS1-PEG-b-PLA micelles carrying DiR (3.5 μg /mouse) or that of non-targeted PEG-b-PLA micelles carrying DiR (3.5 μg /mouse) (Fig. 9). The strongest fluorescence signal of DiR (white in color) in the removed brain ex vivo was at the average radiant efficiency of 5.75×10⁷ [p/s/cm²/sr]/[μW/cm²] on 48 h post intranasal delivery of FibCS1-PEG-b-PLA micelles carrying DiR. The fluorescence signal of DiR in the removed AD brain was at the average radiant efficiency of 7.61×10⁶ [p/s/cm²/sr]/[μW/cm²] on 96 h post intranasal delivery of FibCS1-PEG-b-PLA micelles carrying DiR. No notable fluorescence was detected in the removed AD brain when delivered intranasally by non-targeted micelles (below the average radiant efficiency of 2×10⁶ [p/s/cm²/sr]/[μW/cm²] on 48 h). Additionally, two 3xTg mice were injected with FibCS1-PEG-b-PLA micelles carrying DiR (3.5 μg/mouse) intravenously, and there was no fluorescence of DiR detected in the dissected brain throughout the experiment in 120 h post-treatment This implies that DiR itself was not able to cross the BBB when delivered intravenously whereas the targeted FibCS1-PEG-b-PLA micelles carrying DiR reached the AD brain, showing the strongest DiR signal in 48 h. This result proves that the Fibronectin CS1 peptide is effective in targeting the brain, and the intranasal drug delivery route is effective in delivering the compounds to the brain at an improved efficacy.

Figure 8.

Real-time, live near-infrared (NIR) optical imaging of AD mouse model (3xTg) treated with intranasal FibCS1-PEG-PLA micelles carrying DiR (3.5 μg/mouse) captured by Fluobeam^®.

Figure 9.

Relative radiant efficiency of DiR distributed in the removed brain and nasal tissues of AD mouse model (3xTg) treated with intranasal FibCS1-PEG-PLA micelles carrying DiR (3.5 μg /mouse) or non-targeted PEG-b-PLA micelles carrying DiR (3.5 μg /mouse) on 48, 96, and 120 h captured by IVIS^® (3.5 μg /mouse).

Animal behavior studies

The exploratory behavior, anxiety levels, and cognitive behaviors (e.g. motor coordination, recognition memory, special learning and memory) of the AD mouse model (3xTg) treated with FibCS1-PEG-PLA micelles carrying Rapa (20 mg/kg intravenous (IV) and 0.2 mg/kg intranasal (IN); q4dx5) was predicted from the corner test (CT), Novel object recognition (NOR), and Morris Water Maze (MWM) test (Fig. 10)

The exploratory behavior, anxiety levels, and motor coordination were tested by the CT (Fig. 11). Horizontal activity such as distance travelled and number of visits to the corner was recorded and analyzed by the EthoVision XT software. In the control groups, empty micelles were intranasally given to 3xTg (without Rapa). Mice in the control group showed a lack of locomotor function in the arena after treatments, staying at 1–2 corners of the arena (distance moved: 747.4 ± 207.6 cm, frequency of visiting corners: 1.75 ± 0.50 times). This is presumably due to the high anxiety levels in the AD mouse model. Compared to control, 3xTg mice treated with IV FibCS1-PEG-PLA micelles carrying Rapa at 20 mg/kg (distance moved: 1963.75 ± 222.45 cm, frequency of visiting corners: 2.5 ± 0.58 times) and 3xTg mice treated with IN FibCS1-PEG-PLA micelles carrying Rapa at 0.2 mg/kg (distance moved: 2134.62 ± 333.50 cm, frequency of visiting corners: 4.25 ± 0.96 times) showed more active movement, visiting all 4 corners, which can be attributed to reduced anxiety. The treatments, FibCS1-PEG-PLA micelles carrying Rapa, appeared to help increase the exploratory behavior, enhancing the drive to explore the arena. As Rapa reduces neuroinflammation, treated mice demonstrated the improvement of neuronal circuits and the desire to travel across the arena. Additionally, the improvement in movement was seen more in intranasal (IN) treatment group, which can be attributed to increased autophagy, which clears the Aβ protein and improves cognitive function, proving that intranasally delivered Rapa is effective in reducing anxiety-like behavior and increasing cognitive and motor coordination.

Figure 11.

The CT results (A) Heat map and (B) Horizontal activity (distance moved, number of visits) for AD mouse model (3xTg) treated with empty micelles, IV FibCS1-PEG-PLA micelles carrying Rapa at 20 mg/kg and IN FibCS1-PEG-PLA micelles carrying Rapa at 0.2 mg/kg (q4dx5).

The recognition memory was tested by the NOR (Fig. 12). In this study, the AD mice were placed in a square box (called an arena). To habituate the mice, the mice were placed in the center of the arena for 5 min. After 5 min, each animal was taken out and put in the holding cage. The arena was thoroughly cleaned with 70% ethanol each time before a new animal was placed. For training 1, two similar objects were placed in the inverse direction of the arena. For testing (T2), one of the two objects was replaced with a new object. Each mouse was given 10 min to explore the objects. For both T1 and T2, the score was recorded for the first 5 min if the mouse did not meet the minimum exploration time of 20s for both objects. It was anticipated that the treated AD mice would show a high inclination towards a novel object over a familiar one; however, it was observed that the treated AD mice spent more time for exploring a familiar object (IV: 6.08 ± 1.34 sec for identical object versus 1.53 ± 1.07 sec for the novel object; IN: 5.95 ± 0.82 sec for identical object versus 2.30 ± 0.89 sec for novel object) whereas non-treated AD mice (in control) spent the equal amount of time between novel and identical objects (6.68 ± 1.34 sec for identical object versus 5.95 ± 0.82 sec for novel object). It is suspected that the treated AD mice had better cognitive function, recognizing the novel object and avoiding it due to fear and as a result, they spent less time on the new object compared to the familiar object. The AD mice in control presumably did not recognize new versus old objects therefore spent an equal amount of time exploring the objects.

Figure 12.

Total exploration time (time spent exploring identical (left) or novel objects (right) during T2) measured in the NOR study for AD mouse model (3xTg) treated with empty micelles, intravenous (IV) FibCS1-PEG-PLA micelles carrying Rapa at 20 mg/kg and IN FibCS1-PEG-PLA micelles carrying Rapa at 0.2 mg/kg (q4dx5).

The MWM test is a highly verified and widely used behavioral experiment that tests hippocampus-dependent memory in mice (Fig. 13, Supplementary videos 1–3). First, the AD mice were trained to assess a memory of the pool of water and the hidden stage in the center of the pool. In this study, it was observed that the time spent finding the hidden stage in water for AD mice in control was significantly higher (time spent to find the stage: 42.51 ± 13.01 sec, distance moved: 1298.75 ± 174.86 cm) when compared to IN (time spent to find the stage: 14.97 ± 6.21 sec, distance moved: 532 ± 105.48 cm) and IV (22.94 ± 4.89 sec, moved 788.0 ± 170.65 cm) treated 3xTg mice. This can be attributed to enhancing the learning memory of the treated animals by restraining the activity of mTOR pathway, increasing synaptic modeling in neurons, and enhancing learning power in the AD mouse model.

Figure 13.

The MWM results (A) Heat map and (B) distance moved in the water maze and time spent finding a stage for the AD mouse model (3xTg) treated with empty micelles, IV FibCS1-PEG-PLA micelles carrying Rapa at 20 mg/kg and IN FibCS1-PEG-PLA micelles carrying Rapa at 0.2 mg/kg (q4dx5).

Additionally, IN FibCS1-PEG-PLA micelles carrying Rapa at 0.2 mg/kg (q4dx5) was most effective in reducing neuroinflammation and Aβ protein. (Fig. 14) (western blot) shows that FibCS1-PEG-b-PLA micelles carrying Rapa administered intravenously (20 mg/kg) decreased Aβ protein expression level (from 0.963 ± 0.02 pg/mL to 0.478 ± 0.24 pg/mL) and two pro-inflammatory markers, IL-6 (from 0.360 ± 0.01 pg/mL to 0.14 ± 0.006 pg/mL) and TNF-α (from 0.278± 0.11 pg/mL to 0.106± 0.01 pg/mL) cytokines, compared to control that received empty FibCS1-PEG-b-PLA micelles. FibCS1-PEG-b-PLA micelles carrying Rapa administered intranasally (0.2 mg/kg) significantly decreased Aβ protein expression level (from 0.963 ± 0.02 pg/mL to 0.053 ± 0.01 pg/mL), IL-6 (from 0.360 ± 0.01 pg/mL to 0.15 ± 0.01 pg/mL) and TNF-α (from 0.278± 0.11 pg/mL to 0.035± 0.01 pg/mL) cytokines compared to control that received empty FibCS1-PEG-b-PLA micelles.

Figure 14.

Western blot analysis on the removed AD mouse brain for protein expression and relative band density of pro-inflammatory markers, TNF-α and IL-6. The AD mice (3xTg) were treated with empty micelles, IV FibCS1-PEG-PLA micelles carrying Rapa at 20 mg/kg and IN FibCS1-PEG-PLA micelles carrying Rapa at 0.2 mg/kg (q4dx5).

ELISA also shows (Fig. 15) that FibCS1-PEG-b-PLA micelles carrying Rapa administered intravenously (20 mg/kg) decreased Aβ protein expression level (from 426.66 ± 20.3 pg/mL to 358.87 ± 19.6 pg/mL), IL-6 (from 4013.6 ± 108.2 pg/mL to 2545.9 ± 232.2 pg/mL) and TNF-α (from 691 ± 23.2 pg/mL to 364.92 ± 42.2 pg/mL) cytokines compared to control that received empty FibCS1-PEG-b-PLA micelles. FibCS1-PEG-b-PLA micelles carrying Rapa administered intranasally (0.2 mg/kg) significantly decreased Aβ protein expression level (from 426.66 ± 20.3 pg/mL to272.36 ± 23.2 pg/mL), IL-6 (from 4013.6 ± 108.2 pg/mL to 2165.6 ± 450.2 pg/mL) and TNF-α (from 691 ± 23.2 pg/mL to 267.92 ± 26.23 pg/mL) cytokines compared to control that received empty FibCS1-PEG-b-PLA micelles.

Figure 15.

ELISA analysis on the removed AD mouse brain for protein expression and relative band density of pro-inflammatory markers, TNF-α and IL-6. The AD mice (3xTg) were treated with empty micelles, IV FibCS1-PEG-PLA micelles carrying Rapa at 20 mg/kg and IN FibCS1-PEG-PLA micelles carrying Rapa at 0.2 mg/kg (q4dx5).

Histopathology

The notable Aβ accumulation (called amyloidosis; blue arrow) was observed in the hippocampus region of 3xTg-AD mice (Fig. 16A). There was a noticeable decrease in extracellular Aβ deposition in 3xTg-AD mice who received IN FibCS1-PEG-PLA micelles carrying Rapa at 0.2 mg/kg (Fig. 16B). Fig. 16A shows nuclear pyknosis (red arrow) which is a hallmark feature of apoptotic cell death, illustrating the nucleus shrinking and neuronal loss. This can be associated with neuroinflammation, which initiates oxidative stress and mitochondrial dysfunction. There was a noticeable reduction in nuclear pyknosis (red arrow) in 3xTg-AD mice who received IN FibCS1-PEG-PLA micelles carrying Rapa at 0.2 mg/kg (Fig. 16B). The H&E staining of the cortex of 3xTg-AD mice showed damaged neurons and vacuolar bodies (Fig. 17). Disordered morphological changes were seen with neuronal loss; this can be attributed to Aβ accumulation and oxidative stress on neurons. Granulovacuolar degeneration (GVD), where dense, dark-staining granules were enclosed by transparent vacuoles (about 3–5 μm in diameter), as depicted by the red arrows in Fig. 17A. Upon treatment, the GVD and the neuroinflammation appeared to be reduced (Fig. 17B).

Figure 16.

The H&E staining of hippocampal gyrus in 3xTg-AD mice who received IN empty FibCS1-PEG-PLA micelles (A) and who received IN FibCS1-PEG-PLA micelles carrying Rapa at 0.2 mg/kg (B). The red arrows indicate characteristic nuclear-pyknosis cells. The blue arrow indicates Aβ plaque accumulation.

Figure 17.

The H&E staining of cortex in 3xTg-AD mice who received IN empty FibCS1-PEG-PLA micelles (A) and who received IN FibCS1-PEG-PLA micelles carrying Rapa at 0.2 mg/kg (B). The red arrows indicate granulovacuolar degeneration.

Conclusion

The nose-to-brain route has been reported as a promising alternative to conventional routes for effective drug delivery to the brain. The main advantages lie in the nose-to-brain drug delivery pathway, which can be further improved by employing adequately formulated polymeric micelles that enhance effective permeation of drugs across the nasal mucosa and ultimately improve therapeutic efficacy/treatment outcomes.

The aim of this research was to highlight the therapeutic ability of brain-targeting polymeric micelles carrying Rapa and evaluate therapeutic effects including the reduction of pro-inflammatory response and Aβ and improvement of cognitive function in AD. Fibronectin CS1 peptide-conjugated PEG-b-PLA (FibCS1-PEG-b-PLA) micelles carrying Rapa. FibCS1-PEG-b-PLA carrying Rapa enhanced the nasal permeability of Rapa through the RPMI 2650 epithelial cells compared to free Rapa. FibCS1-PEG-b-PLA carrying Rapa reduced Aβ protein levels in the AD brain in vivo and pro-inflammatory markers (TNF-α and IL-6) in microglia cells in vitro and in vivo. FibCS1-PEG-b-PLA carrying an NIR imaging agent, DiR, confirmed that DiR successfully travelled to the brain of the AD mouse model with the strongest DiR signal in the brain at 48 h post intranasal delivery of FibCS1-PEG-b-PLA carrying DiR. Finally, intranasally delivered FibCS1-PEG-b-PLA carrying Rapa showed significant improvement in cognitive behavior such as motor coordination (CT), recognition memory (NOR), and special learning and memory (MWM) in the AD mouse model in vivo. A histopathological study showed notable decreases in extracellular Aβ deposition in the hippocampus region and granulovacuolar degeneration and in the cortex.

The present study focused on improving drug delivery to the brain via intranasal administration of brain-targeting polymeric micelles. The future work will focus on 1) the assessment of how rapamycin is distributed after the formulation enters the brain, 2) the in vivo toxicity study/pathophysiological analysis for long-term use of the intranasal formulation and 3) the pharmacokinetics profiles of targeted and non-targeted micelles carrying rapamycin.

Highlights.

• Brain-targeting FibCS1-PEG-b-PLA micelles carrying rapamycin enhanced the nasal permeability of rapamycin through the nasal epithelial cells.

• FibCS1-PEG-b-PLA micelles carrying rapamycin reduced Aβ protein levels and pro-inflammatory markers (TNF-α and IL-6) in the brain of the Alzheimer’s disease-bearing mouse model in vivo.

• Intranasally delivered FibCS1-PEG-b-PLA micelles carrying rapamycin showed significant improvement in cognitive behavior such as motor coordination, recognition memory, and special learning and memory in vivo.