Therapeutic Intervention of Neuroinflammatory Alzheimer Disease Model by Inhibition of Classical Complement Pathway with the Use of Anti-C1r Loaded Exosomes

Abstract Alzheimer’s disease (AD) is a complex neurodegenerative disease associated with memory decline, cognitive impairment, amyloid plaque formation and tau tangles. Neuroinflammation has been shown to be a precursor to apparent amyloid plaque accumulation and subsequent synaptic loss and cognitive decline. In this study, the ability of a novel, small molecule, T-ALZ01, to inhibit neuroinflammatory processes was analyzed. T-ALZ01, an inhibitor of complement component C1r, demonstrated a significant reduction in the levels of the inflammatory cytokines, IL-6 and TNF-α in vitro . An LPS-induced animal model, whereby animals were injected intraperitoneally with 0.5 mg/kg LPS, was used to analyze the effect of T-ALZ01 on neuroinflammation in vivo . Moreover, exosomes (nanosized, endogenous extracellular vehicles) were used as drug delivery vehicles to facilitate intranasal administration of T-ALZ01 across the blood-brain barrier. T-ALZ01 demonstrated significant reduction in degenerating neurons and the activation of resident microglia and astrocytes, as well as inflammatory markers in vivo . This study demonstrates a significant use of small molecule complement inhibitors via exosome drug delivery as a possible therapeutic in disorders characterized by neuroinflammation, such AD.


Introduction
Alzheimer's disease (AD) is a global health and socioeconomic concern, contributing to one of the most common causes of dementia [1], [2].Currently, over 50 million people are globally affected with AD, and this is expected to rise to 152 million people by 2050 [1], [2].AD is a multifactorial and polygenic neurodegenerative disease that begins in the hippocampus and cerebral cortex and is characterized by cognitive decline and memory loss.Major hallmarks of AD include β-amyloid (Aβ) plaque formation, neuro brillary tangles (NFTs) and neuroin ammation [1], [2].Over the last decade, few FDA-approved therapeutics have been established for the treatment of AD including acetylcholinesterase inhibitors (donepezil, rivastigmine) and non-competitive N-methyl-D-aspartate antagonists (memantine).Most recently, monoclonal anti-amyloid therapeutics -aducanumab and lecanemab -have been developed for the treatment of this neurodegenerative disease [3], [4].However, the associated possibility of brain swelling increases health and nancial burden on AD patients [3], [5], [6].The success rate of AD drugs is low, stemming from the complex pathophysiology and a poor understanding of the disease progression.
The role of neuroin ammation in various neurodegenerative diseases, including AD, has been alternatively considered.Epidemiological evidence of neuroin ammation is demonstrated in the protective use of non-steroidal anti-in ammatory drugs (NSAIDs) and moderate reduction in the risk of AD [7]- [9].Furthermore, genetic and bio uid markers indicative of neuroin ammation were shown to increase in AD [10], [11].Neuroin ammation is also present in AD patients regardless of Aβ load, indicating a possible initiator role for neuroin ammation in the development of AD [12]- [14].
Neuroin ammation occurs in the central nervous system (CNS) in response to several factors such as CNS injury, infections, toxins, or autoimmunity [13], [14].The duration, type and actions of in ammatory factors can further determine the severity of the neuroin ammatory response.Neuroin ammation plays both a protective and pathological role in the CNS.It is important for brain development, tissue repair after injury, brain plasticity and restoration.However, neuroin ammation can lead to cognitive impairment, neuronal damage, reduced plasticity, anxiety, and depression.The in ammatory response is mediated by the cells of the CNS through the production of cytokines, chemokines, reactive oxygen species and other messengers that contribute to several pathways [15], [16].However, targeting these in ammatory pathways have not yielded effective AD therapeutics and thus, investigators have begun to consider the role of the in ammatory Pathways of Complement Fixation in the development of AD [17]- [21].
The complement system is comprised of over 40 proteins that play a dominant role in initial activation of an in ammatory response and, antibody-mediated pathogen opsonization and clearance [18], [19].The complement system is activated by three different recognition pathways (classical, alternative and lectin) which lead to sequential enzyme activation, protein cleavage and induced function-enabling protein conformational changes.Regardless of the activation mechanism, the three major pathways result in: (1) opsonization, via the deposition of the activation dependent cleavage fragments, which tag pathogens for more e cient phagocytosis, (2) leukocyte/microglia recruitment to the site of injury via production of chemotactic peptides and (3) targeted death of pathogens due to the creation of a membranolytic pore in pathogen cell membranes [17], [22], [23].The generation of the diffusible chemotactic peptides can also lead to the alteration of immune (and other) cellular activities such as production of reactive oxidation species and secretion of pro-in ammatory cytokines that have evolved to aid in the e cient clearance of pathogens.The complement system is normally controlled by several circulating (C1 inhibitor, factor H, C4b-binding protein) and membrane-associated (CD46, CD55, CR1, CD59) complement regulators [17], [19], [23].However, in AD, the excessive activation and poor regulation of the complement system can lead to neurodegeneration, cognitive decline, and memory loss [17], [19].
Numerous small-molecule anti-complement drugs are in development or currently in use for several neurodegenerative and in ammatory diseases such as AD [17]- [19], [22], [23].Of note, three drugs are currently licensed for use in in ammatory diseases -C1INH (Cinryze, Berinert), C5 monoclonal antibody (Eculizumab) and FUT-175.C1INH, similar to the intrinsic C1 inhibitor, reduces duration and severity of acute immune attacks and Eculizumab prevents MAC formation [18], [19].The small molecule FUT-175 is a potent serine protease inhibitor that affects enzymatic activities of C1r, C1s, factor D and C3/C5 [18].However, the off-target effects of these molecules result in toxicity, making them poor candidates for clinical use as a therapeutic in AD.
Analysis of the complement pathway has regarded the initiating complex of the classical complement pathway as a possible therapeutic target [24], [25].Research by Hong et al. analyzed the e cacy of an anti-C1q inhibitor for the treatment of neuroin ammation [25].Results demonstrated a reduction in neuroin ammation but the inhibitor of C1q also reduced opsonization and immune clearance [25].In 2020, Rushing et al. discovered two small molecules (CMP-1611 and CMP-1696) that selectively inhibit C1r through fragment-based discovery analysis [26].C1r consists of two N-terminal CUB domains, an epidermal growth factor (EGF)-like domain, two complement control modules (CCP) and a C-terminal serine protease (SP) domain.Though the mechanism of action of CMP-1611 is unknown, CMP-1696, referred to as T-ALZ01, was found to competitively inhibit C1r through binding of the CCP2-SP region, a functional region conserved in different species such as humans, mice, and rats [26].Though T-ALZ01 is a possible therapeutic, its inability to cross the blood-brain barrier warranted the use of an alternative method of drug delivery.Exosomes are nanosized, extracellular vesicles characterized by a lipid bilayer encompassing biological components [27]- [29].Exosomes demonstrate speci c targeting capabilities to the cell type from which they were derived, thus, microglia derived exosomes were utilized to ensure targeted delivery of T-ALZ01 to the microglia [27]- [29].Exosomes as a drug delivery tool would prove bene cial in reducing economic and social burden, making the administration of therapeutic easier on patient and doctor.
The primary objective of this study focuses on the hypothesis that T-ALZ01, loaded into microglia-derived exosomes, would inhibit the in ammatory processes that cause neurodegeneration and interrupt cellular communication in AD.Given the mechanism of action and speci city of T-ALZ01, several in ammatory markers were assessed in an in vitro and in vivo neuroin ammatory model of AD where the candidate drug was delivered by microglia-derived exosomes.T-ALZ01 demonstrated promising results as a potential AD therapeutic through the reduction of in ammatory markers.

T-ALZ01 demonstrates low IC 50 in SD cells exposed to LPS
To rst determine how T-ALZ01 affects primary Sprague Dawley (SD) microglia cells and to establish the range of subtoxic concentrations for future studies, the MTT cell viability assay was carried out.Primary SD microglia cells were exposed to increasing concentrations (5-500 µM) of T-ALZ01 for 72 hours.Donepezil (DPZ), an FDA approved therapeutic in neuroin ammatory conditions, was used to compare the e cacy of T-ALZ01.Following the 72-hour exposure period, the MTT cell viability substrate was added to the cultures.Only metabolically active cells can reduce the substrate and, thereby, generate a product that allows for the measurement of absorbance.Any decrease in absorbance was considered an indication of cell death or toxicity.In contrast, increases in absorbance readings are typically due to proliferation of cells in the treated cultures.
A non-linear curve analysis was carried out to calculate the IC 50 value allowing determination of the concentration of T-ALZ01 at which the cell viability is reduced by 50%.The IC 50 value for T-ALZ01 was 388.8 µM (Fig. 1A).These results indicate that concentrations of T-ALZ01 from 5 µM to 100 µM are nontoxic under healthy conditions.To further investigate the effect of T-ALZ01 in an LPS-induced neuroin ammatory environment, the SD microglial cells were exposed to 100 ng/mL LPS.The LPS was added 30 minutes prior to the addition of T-ALZ01.In a neuroin ammatory environment, T-ALZ01 demonstrated a IC 50 value of 99.83 µM (Fig. 1B).The lower IC 50 observed after exposure to LPS indicates a more sensitive response to T-ALZ01.Taken together, these data demonstrate that T-ALZ01 is safe to use at concentrations at 10-50 µM.These experiments were also conducted in an immortalized mouse cell line (SIM-A9) which yielded lowered IC 50 after LPS exposure (IC 50 = 326.7,healthy; IC 50 = 33.26,LPS) (Supplementary Figure S1A-B).

T-ALZ01 affects the complement pathway in a dosedependent manner in SD microglia cells
To investigate the effect of T-ALZ01 on the expression of C1r in LPS-induced neuroin ammation, the levels of C1r in primary SD microglia cells treated with T-ALZ01 (10, 30 and 50 µM) was assessed by Western blot analysis (Fig. 2A-B).Whole cell lysates of treated SD microglia cells were probed for C1r using an anti-C1r antibody.C1r levels were normalized against GAPDH.The LPS-only microglia cells served as positive control and unstimulated cells helped to establish the baseline levels of C1r.T-ALZ01 exhibited a dose-dependent inhibitory response.The higher the concentration of the drug, the lower the expression of C1r.Though there was no statistical signi cance between the LPS-only group and 10 µM T-ALZ01, signi cant reduction of C1r was observed in 30 µM (***p < 0.005) and 50 µM (****p < 0.0001) compared to LPS-only (Fig. 2A-B).Western blot analysis of the effect of T-ALZ01 on the expression of C1r in SIM-A9 microglia cells were also conducted and demonstrated a similar dose-dependent inhibitory effect compared to the LPS-only group (10, 30 and 50µM, ****p < 0.001) (Supplementary Figure S2A-B).
To establish T-ALZ01 effect on the complement pathway, ELISA of complement protein C4 and the complement hemolysis 50% (CH50) in SD microglia cells (Fig. 2C-D).Primary SD microglia cells exposed to 100 ng/mL LPS were treated with T-ALZ01 (10, 30 and 50 µM) for 72 hours.In contrast to the expression of C1r, the 10 µM T-ALZ01 demonstrated a signi cant decrease of the protein levels of C4 (**p < 0.01) and CH50 (*p < 0.05) compared to the untreated and LPS groups respectively (Fig. 2C-D).T-ALZ01 also demonstrated a signi cant decline in CH50 protein levels compared to LPS-only group (**p < 0.01) (Fig. 2C-D).

Microglia-derived exosomes can be utilized as a drug delivery tool
Considering the inability of the novel anti-C1r inhibitor to cross the blood-brain barrier (BBB), microgliaderived exosomes were isolated using the polyethylene-glycol method for use in drug delivery.Exosomes are nanosized, extracellular vesicles that demonstrate a heterogenous population (Supplementary Figure S3).Primary SD microglia cells were cultured in normal media (Control) or exosome-depleted media (EXO-Dp) for 5-7 days and the supernatant was collected for exosome isolation.Quantitative analysis of the exosome samples revealed a signi cant decrease in protein concentration (*p < 0.05), exosome concentration (***p < 0.005) and abundance (****p < 0.001) of the EXO-Dp group compared to the Control group (Fig. 3A-C, respectively).Characterization of the exosome samples indicated successful isolation of microglia-derived exosomes.The ExoCheck Antibody Array con rmed the sample contained exosomes with a high expression of ALIX, an exosomal marker (Fig. 3D).Particle size analysis showed the size of the isolated exosomes in the Control (172.4 nm) and EXO-Dp (182.4 nm) groups (Fig. 3E-F, respectively).
Cryo-Transmission electron microscopy demonstrated a larger population of exosomes in the Control group compared to the EXO-Dp group (Fig. 3G-H, respectively).These results indicate successful isolation of exosomes and removal of the exosomes found intrinsically in FBS.EXO-Dp exosome samples were used in the drug loading process of T-ALZ01.The structure of T-ALZ01 was identi ed and the success of the drug loading was evaluated using HPLC and LC-MS (Supplementary Figure S4).

T-ALZ01 reduces neurodegeneration in the hippocampus and cortex of male and female SD rats
To examine the ability of T-ALZ01 to reduce neurodegeneration in vivo, an LPS-induced neuroin ammatory model mimicking AD was established in male and female SD rats.The animals were exposed to LPS (1 mg/kg) for 1 hour followed by treatment with DPZ (1 mg/kg) or T-ALZ01 (0.2 mg/kg) every day for 7 days.Phenotypic analysis of the e cacy of T-ALZ01 in reducing neuroin ammation was conducted by counting fecal pellets (Supplementary Figure S5).LPS affects the gut and thus causes frequent bowel movements.The analysis of fecal pellets following treatment can be used to determine the effect of the drug to alleviate LPS-related symptoms.Complement analysis of the CNS tissue homogenates indicated a signi cant reduction in the protein levels of CH50 was observed in the cortex (****p < 0.001) and hippocampus (***p < 0.005) of female SD rats but no statistical signi cance in the protein levels of C4 (Fig. 4A-D, respectively).Western blot analysis of the tissue homogenates indicated that the concentration of T-ALZ01 did not signi cantly reduce the relative protein expression of C1r (Supplementary Figure S6).ELISA analysis of tissue homogenates demonstrated signi cant decline of the protein levels of TNF-α and IL-6 in the cortex (****p < 0.001) and hippocampus (****p < 0.001) of male SD rats (Fig. 4E-H respectively).Similarly, female SD rats revealed a signi cant decrease in the levels of TNF-α and IL-6 in the cortex (****p < 0.001) and hippocampus (****p < 0.001) (Fig. 4E-H, respectively).The levels of the anti-in ammatory IL-10 were signi cantly increased in male and female in the cortex (****p < 0.001) and hippocampus (****p < 0.001) (Fig. 4I-J, respectively).

Discussion
The aim of this research was to highlight the therapeutic ability of T-ALZ01, delivered by microglia-derived extracellular vesicles called exosomes, in the alleviation of neuroin ammation in Alzheimer's disease (AD).It was hypothesized that T-ALZ01, based on its mechanism of action, would inhibit the in ammatory processes that lead to neurodegeneration.First, the therapeutic effect of T-ALZ01 was examined in vitro through the analysis of in ammatory cytokines and complement pathway activity.This was followed by the analysis of T-ALZ01 loaded exosomes in vivo in LPS-induced neuroin ammation rat model to determine its effectiveness in the brain microenvironment.The complement system pathway was considered an important target as it is a major contributor to the in ammatory processes that cause neurodegeneration [19], [20], [23].The complement protein inhibitors that have been tested in in ammatory diseases such as FUT-175, C3b inhibitor, C5a inhibitor and ANX-M1 (C1q inhibitor) have shown e caciousness in the treatment of such diseases but also show detrimental side effects [18], [25].
The results of this study showed that the administration of T-ALZ01 in SD microglia cells demonstrated effective inhibition of C1r in a dose-dependent manner, as similarly demonstrated by Rushing et al. [26].However, the concentrations of T-ALZ01 that did not completely inhibit C1r showed reduced neuroin ammation which is corroborated by Hong et al. whereby low doses of a C1q inhibitor were utilized to reduce neuroin ammation [25].The analysis of the in ammatory cytokine levels in vitro and in vivo showed that T-ALZ01, compared to donepezil (DPZ), resulted in a signi cant reduction of proin ammatory cytokines (TNFα, IL-6).These cytokines are common markers of neuroin ammation as their mechanism of action exacerbates the macrophagic activity of the microglia further leading to astrogliosis and neuronal damage.In addition, the anti-in ammatory cytokine (IL-10) demonstrated increased protein levels in treated groups in the brain tissue, which symbolizes a more homeostatic brain environment [30], [31].In contrast to SD microglial cells, cytokine analysis of an immortalized mouse microglial cell line (SIM-A9) showed increased sensitivity to T-ALZ01 and thus, did not demonstrate effective reduction of cytokine levels.These results are possibly due to the structural differences in the C1r protein and the low intrinsic activity of the complement pathway in mouse microglia cells compared to rat microglia cells [32]- [34].Unlike rat, and human, the C1r protein in mice are composed of two isoforms (C1r-a and C1r-b) and thus, the use of T-ALZ01 could potentially affect the complement pathway in mice greater than rat, and human.
It has been reported that females are more prone to the development of AD compared to males [1], [2].Females are exposed to more AD-associated risk factors such as stress, menopause-related hormone deprivation and diabetes, and thus have a more active in ammatory complement system compared to males [35]- [37].As demonstrated by the results in the study, in ammatory dysregulation was stronger in females than males.Few animal studies have utilized females in the analysis of the e cacy of possible AD therapeutics.The ability of T-ALZ01 to alleviate neuroin ammation in the female groups would prove bene cial in the treatment of AD in a susceptible population.The e cacy of T-ALZ01 at a wider dosing range and different dosing regimen was not evaluated in this study, hence future research could analyze these conditions in male and female groups.Analysis of the long-term effect of T-ALZ01 would also prove bene cial in understanding its effect in both genders.Females are more sensitive to T-ALZ01 and thus, long-term inhibition of the C1r protein may pose substantial effects.Though many complement protein inhibitors have been developed in the treatment of various neuroin ammatory diseases, the inability of these compounds to cross the blood-brain barrier (BBB) has limited usage.Unlike research conducted by Hong et al.where intravenous administration of a C1q inhibitor was conducted, this study utilized nanosized extracellular vesicles called exosomes for intranasal drug administration [25].Similar to a study conducted by Haney et al., the biocompatible, pharmacokinetic, and targeted ability of exosomes proved bene cial in ensuring successful delivery of the candidate drug to the brain [38].Exosomes, derived from healthy SD microglia, can travel to the microglia in the brain without being rejected by the immune system.However, off-targets to other resident brain cells are possible.The use of exosomes in the delivery of complement protein inhibitors would prove bene cial as a non-invasive method of treatment.The potential ease of administration would bene t the patient and doctor.Though the study utilized microglia-derived exosomes, it provides possible avenues that could be considered in drug delivery.Astrocyte-derived exosomes could be a potential candidate for the delivery of therapeutics to astrocytes, another contributor of neuroin ammation.
Exosome-liposome hybrids would provide use of the bene ts of both vesicles -improved targeted ability, biocompatibility, and drug loading [29], [39].
In summary, the study underscores the signi cant contribution of T-ALZ01 in the treatment of neuroin ammation in AD.Mechanistically, low concentrations of T-ALZ01 alleviate neuroin ammatory conditions without complete inhibition of the CCP2-SP domain of the C1r protein in male and female rats.Exosomes provide a non-invasive, biocompatible, and targeted alternative for the delivery of AD therapeutics to the brain.These ndings provide insights into the potential therapeutic use of T-ALZ01 in mitigating the neuroin ammation associated with neurodegenerative diseases such as AD.

Animals
All animal experimental protocols and guidelines were approved by the St. John's University Institute Animal Care and Use committee (IACUC Protocol #2035).Wild-type Sprague Dawley (SD) rats (9-10 weeks old) were treated in strict accordance with recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institute of Health.The study was reported in accordance with ARRIVE guidelines.The animals were housed in a pathogen-free facility at 22 ± 2°C, 50 ± 5% humidity and 12-hour light/dark cycle with free access to chow diet and water.Males and females were housed separately.

Cell viability analysis
The Sprague Dawley (SD) microglial cells were seeded in 96 well plates at 1 x 10 5 cells/mL, with a nal volume of 100 µL per well, and incubated at 37°C in 5% CO 2 for 1 hour.The microglial cells were divided into two groups: (1) Control -cells were incubated with DPZ and T-ALZ01 (5-500 µM) and ( 2) Treatedcells were exposed to 100 ng/mL LPS for 30 minutes followed by the addition of DPZ and T-ALZ01 at a range of 10-100 µM.The cells were incubated at 37°C in 5% CO 2 for 72 hours.After 72 hours, supernatant was discarded and 100 µL of 0.5% MTT solution was added to the 96-well plates and incubated for 2 hours.After dissolving the formazan product crystals in 100 µL of DMSO for 10 minutes at 37°C, the absorbance at 570 nm was measured using the Molecular Devices FilterMax F5 microplate reader.For Group 1, the viability of the untreated cells was set as 100%.For Group 2, the viability of the LPS treated cells was set as 100%.IC 50 was calculated using the log inhibition curve generated through GraphPad Prism 9 software.The cell viability assay was performed in triplicate and each assay was run three times.

Cytokine and complement activity analysis
The primary SD microglial cells were seeded at 1x10 5 cells/mL in 24-well plates, with a nal volume of 500 µL per well.Next, 100 ng/mL LPS was added, and the cells were further incubated for 30 minutes.The cells were then incubated with DPZ (50 µM) or TALZ-01 (10-50 µM) for 72 hours.The supernatant was collected and used in the following ELISA according to manufacturer instructions: (1) cytokine analysis -TNF-α, IL-6 and IL-10 from Boster Biological Technology (Pleasanton, CA, USA) and, (2) complement pathway activity -CH50 and C4 from AFG Scienti c (Northbrook, IL, USA).Three independent experiments were conducted in duplicate.The generated standard curve was used to determine the concentration of each cytokine and complement protein with GraphPad Prism 9 software.
The detection and quanti cation of in ammatory cytokines and complement pathway activity in the hippocampal and cortical homogenates were conducted using ELISA according to manufacturer instructions from AFG Scienti c (Northbrook, IL, USA).Three independent experiments were conducted in duplicate.The generated standard curve was used to determine the concentration of each cytokine and complement protein with GraphPad Prism 9 software.

Preparation of cell lysates and tissue homogenates
The primary SD microglial cells were rst seeded at 1x10 5 cells/mL in a T25 ask.Next, 100 ng/mL LPS was added, and cells were incubated for 30 minutes.The cells were then exposed to 10, 30 and 50 µM of T-ALZ01 and incubated at 37°C in 5% CO 2 for 24 hours.After 24 hours, the culture medium was aspirated and 3 mL of 1X PBS was added to the asks.The cells were scraped, collected in a 15 mL centrifuge tube, and pelleted at 500xg for 10 minutes at 4°C.The supernatant was discarded and 50 µL 1X RIPA lysis buffer, supplemented with 1 mM PMSF, was added to the pellet.The resuspended cells were transferred to labelled microcentrifuge tubes and kept on ice for 30 minutes.The cells were then centrifuged at 12000 rpm for 20 minutes at 4°C.After 20 minutes, cell debris was removed, and the cell lysates were stored at -80°C for future western blot analysis.Protein concentrations were determined using bicinchoninic based (BCA) protein assay from ThermoFisher Scienti c (Waltham, MA, USA).
The isolated cortex and hippocampus from one of the mid-sagittal hemisections (stored at -80°C) were thawed and weighed prior to the homogenization process.For every 0.5 g of tissue, 4 mL of the Homogenization Buffer (20 mM HEPES/0.32 M sucrose) in double-distilled water was added.At the time of homogenization, 1 mM PMSF was added.The homogenate was transferred to 15 mL tubes and spun for 5 minutes at 1000xg.The supernatant was transferred to a new 15 mL tube and spun for 20 minutes at 13300xg.The supernatant (S2 fraction) was transferred to a new 15 mL tube.The pellet (P2 fraction) was resuspended in P2 buffer at a volume equal to the initial weight of the tissue.Protein assays were carried out on each fraction.Each fraction was stored at -80°C.

Western blot analysis
Whole cell lysates or tissue homogenates (50 µg) were diluted with 4X Laemmli sample buffer to nal 1X dilution and western blot procedure was conducted according to Naja et al. [40].The C1r primary antibody (1:1000) and the HRP-conjugated secondary antibody (1:1000) were used.The HRP signals on the blots were developed using the SuperSignal West Pico Chemiluminescent Substrate and analyzed by capturing the chemiluminescence signal using the Omega Lum ™ G Imaging System (Rockford, IL, USA).Quanti cation of the western blots were carried out using ImageJ (Image Processing and Analysis in Java 1.8.0_112) developed at NIH.

Immunolabelling
In a 24-well cell culture plate, glass coverslips (12mm/Fisher Scienti c, Pittsburgh, PA) were coated for 1 hour at room temperature with 500 µL poly-D-lysine (PDL at 0.1 mg/mL/0.1Mboric acid).Next, the coverslips were washed 4 times with cell culture-grade distilled water (Millipore, MA, USA) and sterilized under UV light for 30 minutes.Primary SD microglia cells were seeded onto the PDL-coated coverslip at a density of 1x10 5 cells/mL.The cells were then exposed to 100 ng/mL LPS.The cells were then treated with DPZ (50 µM) or T-ALZ01 (10, 30 and 50 µM) for 72 hours.The primary cells were then xed and prepared according to Habeeb et al. [41].The coverslips were incubated in the primary antibody (1:200) followed by the Cy3-conjugated secondary antibody (1:500) and then mounted on Fisherbrand glass microscope slides with a drop of VectaShield Mounting Medium with DAPI.Slides were stored at 4°C and sealed with nail polish the next day to prevent drying of the samples.
One of the mid-sagittal hemisections of the brains isolated from SD rats were xed in ice cold 4% PFA/PBS in a 50 mL tube, with gentle agitation, for 2 hours at 4°C.Next, the brain tissue was transferred to 30% sucrose in PBS and incubated overnight at 4°C, with occasional gentle inversion of the tube.Once tissue was fully equilibrated, the brain tissue was embedded in OCT and then dredged in a 1:1 mixture of OCT and 30% sucrose until it was fully covered.The mold was then placed in a at bed of pulverized dry ice and once frozen, the tissue block was wrapped in para lm and stored at -80°C.One hour before sectioning, the tissue blocks were allowed to equilibrate in the Microm HM 525 cryostat microtome (Waltham, MA, USA).The xed brain tissues were sliced at 14 µm (12-15 slices per mid-sagittal cut).The sections were then dipped into PBS to hydrate the slides and excess was dabbed off.The sections were blocked with 10% heat inactivated goat serum (HIGS) in PBT for 1 hour at room temperature then dipped once in PBT and excess was dabbed off.Following the blocking, the sections were stained with the primary antibodies (GFAP 1:1000, CD11b 1:500) in antibody dilution (100 µL HIGS/PBT) at 4°C overnight.
After washing 3x with PBT for 5 minutes each, the tissues were incubated with uorochrome-conjugated secondary antibodies (Cy2 anti-mouse 1:200, Cy3 anti-rabbit 1:500) in antibody dilution for 1 hour at room temperature.The immunostained tissue was washed 3x with PBT for 5 minutes then the VectaShield mounting solution with DAPI (Burlingame, CA, USA) was added and coverslip was placed.
The slides were stored at 4°C until imaging.

Exosome isolation and quantitation
FBS was spun at 100,000xg for 2 hours to remove exosomes before being added to the DMEM.SD microglial cells were then plated (1:25) in T225 asks in media containing regular FBS (Control) or media containing exosome-depleted FBS (EXO-Dp) for 5-7 days.The vesicle-containing medium from cell culture was collected in 50 mL centrifuge tubes and exosomes were isolated according to Rider et al. [42].
Samples were centrifuged in a Beckman Coulter Allegra X-30R centrifuge (Jersey City, NJ, USA) at 4255xg for 1 hour at 4°C.The resulting pellet was suspended in 500 µL of PBS.Protein concentrations were determined using bicinchoninic based (BCA) protein assay from ThermoFisher Scienti c (Waltham, MA, USA).
The control and EXO-Dp exosome samples (stored at -80°C) were thawed and gently vortexed before use in quantitation experiments analyzing uorescence intensity and acetylcholinesterase activity.Coumarin 6 was used to analyze uorescence intensity in exosome samples.Coumarin 6 powder was added to 200 µL of sample in a light-sensitive microcentrifuge tube and rocked overnight on a shaker at 37°C.The samples were then centrifuged at 6000 rpm for 2 minutes.Next, 50 µL of the supernatant was added to 450 µL of acetonitrile and mixed.Then, 50-100 µL was added to each well of a light-sensitive 96-well plate and read at 450 nm using the Molecular Devices FilterMax F5 microplate reader.
Acetylcholinesterase activity was analyzed using the EXOCET Exosome Quantitation Kit from System Biosciences (Palo Alto, CA, USA).Instructions were followed according to the manufacturer.Experiments were carried out in duplicate from three independent experiments.

Exosome characterization
The control and EXO-Dp exosome samples (stored at -80°C) were thawed and gently vortexed before quanti cation using the Exo-Check Antibody Array from System Biosciences (Palo Alto, CA, USA).Each array is comprised of 12 pre-printed spots and features 8 antibodies for known exosome markers (CD63, CD81, ALIX, FLOT1, ICAM1, EpCAM, ANXA5 and TSG101), a GM130 cis-Golgi marker for cellular contamination, two positive controls and a blank control.Instructions were followed according to the manufacturer.The arrays were developed using the SuperSignal West Pico Chemiluminescent Substrate and analyzed using the Omega LumTM G Imaging System (Rockford, IL, USA).Particle size analysis was conducted through dynamic light scattering (DLS) using the Zetasizer instrument (Malvern Panalytical).Samples were diluted in sterile, particle-free PBS at ratios of between 1:250 and 1:1,000 for optimum analysis.PBS was tracked before each experiment to ensure that it was particle-free.Cryo-TEM electron microscopy was conducted on exosomes (CUNY Advanced Research Center, NY, USA).Samples were rst cryo-xed (plunge-freeze) to eliminate water crystals that can affect structure, thus creating vitreous ice.
Next, samples were transferred to the TEM for imaging.Approximately 10 images were taken of each sample.

Exosome drug loading
T-ALZ01 was loaded into exosomes through sonication.T-ALZ01 (0.2 mg) was added to 200 µL of exosomes (~ 1x10 8 exosomes) and the mixture was sonicated using a Cole-Parmer Ultrasonic Processor (Vernon Hills, IL, USA).The sonication settings used were 500 v, 2 kHz, 20% power, 6 cycles by 4 seconds pulse and 2 seconds pause according to Haney et al. [38].The mixture was allowed to cool on ice for 2 minutes, and then the sonication cycle was repeated.The success of the drug loading was analyzed using high performance liquid chromatography (HPLC).

Lipopolysaccharide (LPS)-induced neuroin ammation rat model
Eighteen male and eighteen female (9-10 weeks old) wild-type SD rats were purchased from Taconic Biosciences (Germantown, NY, USA).They were randomly and equally allocated to three treatment groups (n = 12/group) -PBS, DPZ and T-ALZ01.All animals received intraperitoneal injection of 0.5 mg/kg LPS daily.Four hours after each daily LPS injection, animals received treatment based on their allocated group.DPZ was administered orally at 1 mg/kg daily.T-ALZ01 was administered intranasally at 0.2 mg/kg.The animal experiment was carried out for 7 days.The rats were weighed daily and checked for any signs of toxicity or distress such as lethargy, loss of appetite and motor/behavioral changes (ataxia or lack of self-grooming).Observations were also made on the number of fecal pellets produced and reported.Once the experiment was completed, animals were sacri ced in 24-48 hours.Animals were exposed to 2% iso urane for 2 minutes followed by sacri cial decapitation.The brains were isolated, and the brains were hemisected down the mid-sagittal plane.One hemisection was subjected to cryosectioning for use in immunocytochemistry and the second hemisection was used for biochemical and protein analysis.The brains of four untreated male SD rats (4 months old) were isolated and used as background control.

Imaging
Fluorescent images were obtained using a Zeiss Axioplan 200M upright uorescent microscope, AxioCam digital camera and AxioVision 4.8.2.0 software.Phase contrast images were collected using a Zeiss Axiovert 25CFL inverted microscope equipped with a Luminera In nity 3 − 1 CCD camera and In nity capture software (version 6.5.4).

Statistical analysis
Signi cance is de ned as p ≤ 0.05.Levels of signi cance are identi ed as follows: * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001 and **** p ≤ 0.0005.All statistical analyses were conducted using GraphPad Prism 9 (San Diego, CA, USA).Unpaired two-tailed t-tests were used to compare two groups; to compare more than two groups, one-way analysis of variance (ANOVA) Tukey test was performed.T-ALZ01 reduces neuroin ammation in tissue homogenates of male and female SD rats.One of the midsagittal hemisections of the isolated brains from the treated (n=18) and female (n=18) wild-type SD rats were homogenized and used for the analysis of cytokine protein levels and complement activity.Untreated male SD rats (4-month-old) were used as negative control.(A-D) ELISA analysis of complement activity in the tissue homogenates indicated signi cant reduction in the protein levels of CH50 in the Figure 5 T-ALZ01 reduces neurodegeneration in the cortex and hippocampus male and female SD rats.One of the mid-sagittal hemisections the isolated brains from the treated male (n=18) and female (n=18) wildtype SD rats were sectioned (14 µM; n=12-15 slices/group) for analysis of degenerating neurons.Untreated male SD rats (4-month-old) were used as negative control (n=4).The tissue sections were labelled with a histo uorescent anionic dye (green) speci c for degenerating neurons and DAPI to label nuclei (blue).(A) T-ALZ01 exhibited reduced uorescence intensity of the anionic dye in male and female

Figure 1 T
Figure 1

Figure 2 T
Figure 2