c-KIT inhibitors reduce pathology and improve behavior in the Tg(SwDI) model of Alzheimers disease

BK40143 and BK40197 are selective for specific tyrosine kinases

We performed screening analysis via KinomeScan, a competitive binding assay, to determine the selectivity profiles of BK40143 (Fig 1A) and BK40197 (Fig 1B) against over 450 distinct kinases (Fig 1C and D). Primary screening results and corresponding compound/target affinities indicated POC (percentage of control) values with tighter binding (higher affinity) interactions associated with lower POC values and weaker binding (lower affinity) associated with higher POC values. We found that the lowest POC range of 0 ≤ x < 0.1 corresponded with BK40143 and BK40197 (Fig 1C and D) completely abolishing ligand binding of human c-KIT at 0 and 0.3 POC, respectively. The gain-of-function c-KIT (V559D) mutation (Hirota et al, 1998) that is associated with c-KIT activation resulting in aberrant mast cell growth and accumulation of mast cells in tissues (El-Agamy, 2012), showed similar binding affinity of BK40143 (0.25 POC) and BK40197 (0.35 POC). BK40143 also prevented ligand binding against PDGFRβ (0.1 POC), whereas BK40197 showed lower affinity with a POC 1.3. Interestingly, BK40197 displayed 0.4 POC against the autophagy-related protein complex, FYVE finger-containing phosphoinositide kinase (PIKFYVE) (Rivero-Rios & Weisman, 2022). BK40143 also displayed a POC range of 1 ≤ x < 10 against several other targets, including PDGFRα, phosphorylated Abl, p-Abl (2.5), the serine/threonine Citron Rho kinase, CIT (2.8), the FMS-like tyrosine kinase-3, FLT3 (4.7), Src (5.3), and SRC family tyrosine kinases that activate T-cell signaling, including spleen tyrosine kinase, SYK (5.5), mitogen-activated protein (MAP)4K4 (6.8), lymphocyte-specific protein tyrosine kinase, LCK (7.1), and Unc-51-like autophagy kinase, ULK3 (9.8), whereas BK40197 displayed affinity to a plethora of other autophagy and inflammation related kinases, including the anti-inflammatory serine/threonine-protein kinase RIO3 (RIOK3) (4.8), Phosphatidylinositol-4-phosphate 5-kinase type-1 alpha (PIP5K1A) (5.1), interleukin-1 receptor-associated kinase 1(IRAK1) (5.3), ULK3 (6.1), the threonine/tyrosine mitogen-activated protein kinase kinase (MEK5) (6.9), and the mammalian Target of Rapamycin, mTOR (9.3).

BK40143 and BK40197 induce autophagic clearance of α-synuclein in vitro

Binding specificity of BK40143 and BK40197 revealed binding to major protein kinases involved in autophagy, including ULK3 (Gonzalez-Rodriguez et al, 2022), mTOR (Wang & Zhang, 2019), PDGFRs (Zhang et al, 2021b), PIKFYVE (O’Connell & Vassilev, 2021), as well as c-KIT (Lee et al, 2013). To establish that treatment with BK40143 and BK40197 induces autophagy, human SH-SY5Y neuroblastoma cells were transfected with an α-synuclein plasmid before being treated with rapamycin/chloroquine as a positive control, DMSO as a negative control, or an ascending dose (10, 1, or 100 μM) of either BK40143 or BK40197 for 8 h. Cells were then harvested and treated using the Abcam Autophagy Assay kit, which uses a cationic amphiphilic tracer to stain autophagic vacuoles, before analysis via flow cytometry. We found that treatment with 10 μM of both BK40143 (Fig 2A and C) and BK40197 (Fig 2B and D) resulted in significant increases in labelled cells, indicating induction of autophagy in cells stressed with α-synuclein.

BK40143 and BK40197 inhibit peripheral mast cell maturation in a mouse model of AD

c-KIT is responsible for mast cell development and homeostasis (Lake et al, 1978; Linnekin, 1999; Kitamura et al, 2006). TKs are the central driving force for processes leading to mast cell activation (Parravicini et al, 2002; Gomez et al, 2005). Multiple receptors are capable of mediating or modifying mast cell activation (Kraft & Kinet, 2007; Kuehn & Gilfillan, 2007), the major receptor responsible being FcεRI, the high affinity receptor for immunoglobulin E (IgE) (Rivera & Gilfillan, 2006). Mast cells are tissue-resident cells of hematopoietic lineage which are derived from CD13⁺CD34⁺KIT(CD117)⁺ bone marrow pluripotent progenitors (Kirshenbaum et al, 1999). Although mast cells play an important role in the innate immune system (Marshall, 2004; Galli et al, 2005), these cells are also responsible for the detrimental exaggerated reactions to antigen (Galli et al, 2008) that follow the release of potent inflammatory mediators as a consequence of receptor-mediated mast cell activation (Metcalfe et al, 1997). The progression of these cells to fully mature mast cells is dependent on c-KIT activation which occurs because of stem cell factor-induced c-KIT dimerization and auto-phosphorylation (Kirshenbaum et al, 1999). To verify the functional effects of BK40143 and BK40197 on c-KIT inhibition, we investigated whether inhibition of c-KIT with BK40143 and BK40197 affects peripheral mast cell populations in vivo. We collected whole blood samples from Tg(SwDI) (referred to as TgAPP) mice treated with 5 or 45 mg/kg of each drug, respectively, or DMSO for 6 wk before isolating mast cells for analysis via flow cytometry (Fig 3A–C). Whole blood was sorted for mast cells using a triple gating strategy: 647 anti-CD45 to isolate leukocytes, PE anti-c-KIT to isolate mast cells, and APC/fire anti-FcεR1α to identify mature mast cells. The ratio of immature-to-mature mast cells (c-KIT+ versus c-KIT+/ FcεR1α+) was compared between c-KIT inhibited and DMSO-treated controls to determine the functional consequence of c-KIT inhibition on peripheral mast cell maturation. We found that mice treated with 5 mg/kg BK40143 (Fig 3B and D) exhibit a significantly greater ratio of immature-to-mature mast cells when compared with DMSO-treated mice (Fig 3A and D), indicating that c-KIT inhibition with BK40143 prevents maturation of peripheral mast cell progenitors. In mice treated with BK40197, we saw a similar, albeit not statistically significant (P = 0.07) increase in immature-to-mature mast cell ratio (Fig 3C and E), further signifying prevention of mast cell progenitor maturation. We further demonstrated c-KIT inhibition in the brains of our mice by performing Western blot for phospho-c-KIT on lysates from drug treated versus control mice. We observed a significant decrease in phospho-c-KIT levels in BK40143-treated mice (Fig 3F and G), whereas BK40197-treated mice displayed a similar, albeit not statistically significant, trend (Fig 3F and H). Collectively, the data confirm the screening experiment and indicates that treatment with BK40143 and BK40197 is sufficient to prevent mast cell proliferation in vivo, via c-KIT inhibition.

Pharmacokinetics

We previously demonstrated that BK40143 was brain penetrant (Fowler et al, 2020). Pharmacokinetic (PK) analysis of BK40197 showed it to also be brain penetrant (Fig 4A and B), with a brain to serum ratio of 25.1% at 20 mg/kg and 20.1% at 45 mg/kg, indicating that both our compounds cross the blood-brain barrier (BBB). We then treated WT mice with ascending doses of BK40143 and BK40197 to determine the maximum tolerable dose for each drug in vivo. Multiple ascending dose treatments revealed the maximum tolerable dose of BK40143 to be 7.5 mg/kg injected IP (Fig 4C), whereas a similar paradigm revealed the maximum tolerable IP dose of BK40197 to be 45 mg/kg (Fig 4D). This identified our compounds as brain penetrant and tolerable in mice, allowing us to test them as potential treatments for neurodegenerative diseases.

BK40143 and BK40197 improve behavior in mouse models of neurodegeneration

To examine the behavioral outcomes following c-KIT inhibition with BK40143 and BK40197, we treated TgAPP mice with either 5 or 45 mg/kg IP, respectively, or DMSO for 6 wk, before performing behavioral analysis. We used the novel object recognition test, a memory assessment that indicates mild cognitive impairment, the Morris water maze, a measure of spatial memory that demarcates more severe memory impairments, the elevated plus maze, a measure of anxiety-like behavior, and nesting, a metric of normative mouse behavior. We discovered that mice treated with both BK40143 (Fig 5A) and BK40197 (Fig 5B) spent significantly greater amounts of time exploring the novel object than did DMSO-treated controls, indicating improved recall on the novel object recognition test. We also found that mice treated with both drugs were able to locate the platform significantly quicker on later training trials than the control mice (Fig 5C), indicating improved spatial memory after drug treatment. Interestingly, only mice treated with BK40143 showed improvement on measures of memory on the probe day of the water maze (time in the correct quadrant [Fig 5D], number of quadrant entries [Fig 5E], and latency to the platform [Fig 5F]), demonstrating that treatment with BK40143 correlates with greater improvement in spatial memory. Furthermore, we found that only TgAPP mice treated with BK40197 spent significantly greater time in the open arms of the elevated plus maze than the DMSO-treated controls (Fig 5G), indicating that treatment with BK40197 may be sufficient to alleviate cognitive symptoms associated with AD pathology, such as anxiety. Lastly, we examined memory and nesting behavior in A53T mice treated with our drugs and observed significant improvements on novel object recognition in mice treated with BK40143 (Fig 5H), as well as significant decreases in nest weight, indicative of decreased nestlet shredding behavior, in mice treated with both BK40143 (Fig 5I) and BK40197 (Fig 5J). Together, these data emphasize the potential of these two compounds in alleviating cognitive-behavioral deficits associated with AD and identifies c-KIT inhibition as a putative method for therapeutic intervention when treating neurodegenerative disorders.

BK40143 and BK40197 reduce concentrations of amyloid-beta and pTau via autophagy

To determine the effects of c-KIT inhibition with our compounds on brain amyloid-beta and pTau concentrations in vivo, we first performed immunohistochemistry to visualize amyloid deposition in our treated and untreated animals. We used immunofluorescence and 3, 3′ diaminobenzidine staining using a 6E10 antibody (BioLegend) to observe whether treatment with our compounds had any outcome on amyloid levels. We discovered that treatment with BK40143 (Fig 6C, D, and G) and BK40197 (Fig 6E–G) robustly decreased amyloid concentrations in the dentate gyrus of the hippocampus in treated animals compared with DMSO controls (Fig 6A, B, and G). We further confirmed that TK inhibition reduces concentrations of CNS neurotoxic proteins via ELISA for amyloid-beta and tau using brain lysates taken from both TgAPP and Tg4510 mice. We found that both BK40143 (Fig 6H and I) and BK40197 (Fig 6J and K) treatment significantly reduced Aβ and pTau (396) concentrations compared with DMSO-treated controls. To confirm our screening data that BK40143 and BK40197 bind to various autophagy proteins (mTOR, ULK3, PIKEFYVE, etc.) and induces autophagy in cell-stressed with alpha-synuclein in vitro (Fig 2), we performed protein analysis on brain samples harvested from TgAPP mice. Western blot revealed significant increases in levels of beclin-1, a multi-domain protein integral for the induction of autophagy, in mice treated with BK40143 (Fig 6L and N) and BK40197 (Fig 6M and O) compared with DMSO-treated controls, indicating that drug treatment is sufficient to induce autophagy in vivo, in association with reduced amyloid and pTau. This further identifies c-KIT inhibition as an effective mechanism for clearing neurotoxic protein aggregates in AD brains.

BK40143 and BK40197 reduce microglial inflammation

To examine the effects of c-KIT inhibitors on microglial activation and subsequent neuroinflammation in vivo, we performed immunohistochemistry and morphological analysis of microglia from the brains of treated and untreated animals. We stained for IBA1, a calcium-binding protein used to identify microglia, before analyzing microglial morphology, a representative feature of activated microglia, using the “surfaces” plug-in in Imaris. We observed a significant reduction in microglial activation in both BK40143 (Fig 7E and F) and BK40197 (Fig 7I and J) treated microglia compared with DMSO controls (Fig 7A and B). We also found significantly reduced surface area of microglia from BK40143 (Fig 7G, H, and M) and BK40197 (Fig 7K, L, and N) treated versus untreated (Fig 7C and D) animals. This identifies c-KIT inhibition as a mechanism for reducing microglial activation and neuroinflammatory phenotypes in AD brains and suggests a multiple mechanism process by which TKIs may be promoting therapeutic benefit.

Sex as a biological variable

Sex was not considered as a biological variable, as both male and female mice were used and showed similar results.

Synthesis of BK40143 and BK40197

We prepared a small group of N-heterocyclic aromatic scaffolds with the goal to improve activity with physicochemical characteristics favorable for brain uptake and crossing of the BBB as we previously reported. Neurotherapeutics with a molecular weight below 500 Da and structures that form less than eight hydrogen bonds with solvating water molecules are generally considered more likely to cross the BBB via lipid-mediated free diffusion. To achieve favorable PK properties for efficient brain uptake, we attempted to offset lipophilic features with a moderate molecular dipole moment and hydrogen bond capacities. All products were prepared in high purity according to ¹H & ¹³C nuclear magnetic resonance spectroscopy and chromatographic methods.

Specificity binding kinase screening assay

The KINOMEscan screening platform (Eurofins DiscoverX Corporation) was performed using active site-directed competition binding assay to quantitatively measure interactions between BK40143 and BK40197 and more than 450 human kinases and disease relevant mutant variants. KINOMEscan assays do not require adenosine triphosphate and thereby report true thermodynamic interaction affinities, as opposed to IC₅₀ values, which can depend on adenosine triphosphate concentration (https://www.eurofinsdiscovery.com/solution/scanmax). BK40143 and BK40197 that bind the kinase active site and directly (sterically) or indirectly (allosterically) prevent kinase binding to the (proprietary Eurofins DiscoverX Corporations) immobilized ligand reduce the amount of kinase captured on a solid support, whereas BK40143 and BK40197 that do not bind the kinase have no effect on the amount of kinase captured. Screening “hits” are identified by measuring the amount of kinase captured in test (BK40143 and BK40197) versus control samples by using a quantitative and ultra-sensitive qRT-PCR method that detects the associated DNA label. BK40143 and BK40197 were screened at 10 µM concentration, and results for primary screen binding interactions were reported as “% Ctrl” (POC), where lower numbers indicate stronger hits and calculated as follows:[test compound signal−positive control signalnegative control signal−positive control signal]×100

test compound = BK40143 or BK40197

negative control = DMSO (100% Ctrl)

positive control = control compound (0% Ctrl)

Based on screening data from of profiled compounds, a proportional relationship between primary screening results and corresponding compound/target affinities can be described as binding constants (Kd values) for the indicated ranges of POC values with tighter binding (higher affinity) interactions associated with lower POC values and weaker binding (lower affinity) associated with higher POC values. Briefly, kinase-tagged T7 phage strains were grown in parallel in 24-well blocks in an E. coli host derived from the BL21 strain. E. coli were grown to log-phase and infected with T7 phage from a frozen stock (multiplicity of infection = 0.4) and incubated with shaking at 32°C until lysis (90–150 min). The lysates were centrifuged (6,000g) and filtered (0.2 μm) to remove cell debris. The remaining kinases were produced in HEK-293 cells and subsequently tagged with DNA for qRT-PCR detection. Streptavidin-coated magnetic beads were treated with biotinylated small molecule ligands for 30 min at RT to generate affinity resins for kinase assays. The liganded beads were blocked with excess biotin and washed with blocking buffer (1% BSA, 0.05% Tween 20, 1 mM DTT; SeaBlock [Pierce]) to remove unbound ligand and to reduce non-specific phage binding. Binding reactions were assembled by combining kinases, liganded affinity beads, and test compounds in 1x binding buffer (20% 0.17x PBS, 0.05% Tween 20, 6 mM DTT; SeaBlock). Test compounds were prepared as 100x stocks in 100% DMSO and directly diluted into the assay. All reactions were performed in polypropylene 384-well plates in a final volume of 0.02 ml. The assay plates were incubated at RT with shaking for 1 h and the affinity beads were washed with wash buffer (1x PBS, 0.05% Tween 20). The beads were then resuspended in elution buffer (1x PBS, 0.05% Tween 20, 0.5 μM non-biotinylated affinity ligand) and incubated at RT with shaking for 30 min. The kinase concentration in the eluates was measured by qRT-PCR.

In vitro autophagy assay

Human SH-SY5Y neuroblastoma cells were grown and cultured in F12 media with 10% fetal bovine serum, 1% L-glutamine, and 1% pen/strep at normal atmospheric conditions (37°C, 21% O₂, 5% CO₂). Upon reaching 50–70% confluence, cells were transduced with a lentivirus containing an alpha-synuclein plasmid and allowed to grow for 8 h. Cells were then treated with an ascending dose (100 nM–10 μM) of BK40143 or BK40197, or DMSO or rapamycin/chloroquine overnight before being harvested and treated with the Abcam Autophagy Assay kit (cat. #ab139484) according to manufacturer’s protocols. Cells were then resuspended in cell staining buffer before being analyzed on a Becton Dickinson LSRFortessas flow cytometer.

In vivo mast cell analysis

Whole blood samples were harvested from the right ventricle of treated and untreated mice and added to 1,000 U/ml heparin-coated tubes to prevent coagulation. 100 μl of blood was then treated with three flow-conjugated antibodies: 647 anti-mouse CD45 (cat. #103123; BioLegend), PE anti-mouse CD117 (c-KIT) (cat. # 105807; BioLegend), and APC/fire 750 anti-mouse FcεR1α (cat. #134339; BioLegend) for 30 min, then treated with 1x RBC lysis buffer (cat. #420301; BioLegend) for 20 min. Cells were washed and resuspended in cell staining buffer before being analyzed on a Becton Dickinson LSRFortessas flow cytometer. Immature-to-mature mast cell ratio was calculated by dividing the percentage of single-labelled (CD117+) mast cells by the total number of c-KIT expressing cells (CD117+ plus CD117+ FcεR1+). All animal work was approved by Georgetown University Animal Care and Use Committee (GUACUC).

Pharmacokinetic analysis

The PK profile of BK40197 was investigated after single IP doses of 20 and 45 mg/kg BK40197 to male and female 5–6 mo old C57BL/6 mice compared with DMSO. A solution at 5 ng/ml isotope labelled reference compounds d8-BK40197 was prepared. The solvent used for internal standard (ISTD) dilution is (200 ml/50 ml) acetonitrile/ethyl acetate (ACN/EtOAc). Brain tissue was homogenized in ice water bath and was prepared using weight accurately on analytical balance with 4 decimals. A volume water (MilliQ) was added by adjustable 100–1,000 μl to a concentration at 100 mg brain tissue/ml water. Brain homogenate (25 μl) and plasma sample (25 μl) that was thawed to RT and gently mixed were pipetted to eppendorf tube with lid and 75 μl ISTD added, mixed on vortex and centrifuged for 5 min at 12, 3 RCF. A volume of 25 μl milli-Q water and 75 μl brain or plasma samples were added into a 200 μl PE vial and mixed on vortex and 10 μl and injected on LC-MS/MS. All animal work was approved by GUACUC.

Drug tolerability assessment

3–4-mo-old WT mice (male and female, C57BL/6) were IP injected with starting doses of 10 mg/kg of BK40143 or BK40197 daily for 5 d before a 2-d washout period. BK40143 dosage was reduced to from 10 to 7.5 mg/kg doses due to rodent mortality, and this dose was continued in 5 d increments up to 70 d. BK40197 dosage was increased in 10 mg/kg increments for 5 d each until the maximum tolerable dose of 50 mg/kg was reached. All animal work was approved by GUACUC.

Mice

Mice used for experiments and analysis were primarily 12–14-mo-old TgAPP mice expressing the neuronally derived human APP gene, 770 isoform, containing the Swedish, Dutch, and Iowa familial Tg(SwDI) mutations driven by a Thy1 promoter. A mixture of males and females were used for each experiment, as sex was not considered as a biological variable for these experiments. Cohorts consisted of 10 mice per treatment group, and 2–3 cohorts for each drug were used. To investigate the effects of BK40143 and BK40197 on levels of amyloid-beta, we used an additional cohort of 4-mo-old TgAPP mice. To investigate drug effects on levels of phospho-Tau, we used cohorts of 3-mo-old Tg4510 tauopathy mice that express the human P301L tau mutation driven by a CAMKIIα promoter. For behavioral analysis, we also treated human α-synuclein mutant (A53T) mice possessing a mutation to the SNCA gene driven by a prion (PrP) promoter to overexpress α-synuclein, of which equal numbers of male and female mice were used. All animal work was approved by GUACUC.

Drug treatment

BK40143 and BK40197 were kept in powder form at -20°C until ready for use. Drugs were dissolved in DMSO at 1.25, 2.5, or 5 mg/ml for BK40143 and 2.5, 5, 10, or 45 mg/ml for BK40197. Mice were then weighed, and the corresponding volume was IP injected daily within 2 h of the previous day’s injection for 6 wk in the treated mice.

Novel object recognition

Novel object recognition was carried out using a 50 × 50 cm arena with clear walls covered by brown construction paper. Mice were habituated to the room for 30 min each day, before being allowed to freely explore the arena for 5 min two times, for 3 d. On the 4th d, two identical objects were placed in the arena in opposite adjacent corners, 10 cm from the corner of the arena. Mice were placed along the wall opposite from the objects and allowed to freely explore for 10 min. Approach behavior (extending nose toward the object) toward both objects was recorded manually using handheld stop watches and compared with ensure there was no preference for either neutral object. For the probe trial, one object was removed and replaced with a novel object, and mice were again allowed to freely explore for 10 min, with approach time towards the novel and familiar objects recorded. Trials in which the mice explored for less than 1 s were excluded. Discrimination index was calculated by dividing time spent exploring the novel object by total time exploring both objects and was compared for treated versus control animals.

Morris water maze

The Morris water maze was performed using a water bath 6 ft in diameter and filled with roughly 1 ft of water. The maze was divided into four quadrants, each with a visual stimulus positioned on the wall of the maze in the center of the quadrant arc. A clear platform was placed in one quadrant just beneath the surface of the water. Mice were then placed in a random non-platform quadrant and allowed to swim freely for 60 s (time was stopped and recorded if the mouse found the platform) before being placed on the platform for 15 s and returned to their cage. This was repeated for each of the three quadrants each day for 4 d. On the 5th d, the platform was removed and mice were placed in the quadrant opposite of where the platform had been and allowed to swim freely for 1 min before being returned to their cage. Time spent in the correct quadrant, latency to the platform, number of platform entries, and number of quadrant entries were recorded.

Elevated plus maze

The elevated plus maze was performed using a standard set up, with two enclosed arms and two open arms. Mice were placed at the intersection of the open and closed arms and allowed to freely explore for 5 min, two times. Time spent in the open and closed arms and number of entries into the open and closed arms were recorded.

Nesting

Mice were singly housed in standard cages overnight (6 pm–6 am) with a 3-gram cloth nestlet and allowed to freely behave. The next day, nestlets were weighed and photographed, and amount of nestlet shredding was recorded by three observers based on the following scale: 0—nestlet untouched, 1—nestlet hardly touched (more than 90% intact), 2—nestlet partially torn (50–90% intact), 3—mostly shredded but no identifiable nest, less than 50% intact but less than 90% in a single quarter of the cage, 4—An identifiable but flat nest, more than 90% torn and material is gathered in a single quarter of the cage, 5—A near perfect nest, more than 90% of the nestlet is torn and the nest is a crater, with walls higher than the mouse body height.

Immunohistochemistry

Staining for amyloid-beta and microglia was performed on left brain hemispheres collected from treated and untreated mice. Brains were cryoprotected with 30% sucrose and sectioned at 30 μm on a cryostat. Sections were then treated with an anti-amyloid beta (6E10, cat. #803001; BioLegend) or anti-IBA1 (cat. # 018-28523; Wako Chemicals) primary followed by either an AlexaFluor secondary (Thermo Fisher Scientific) and DAPI mounting medium (Vector Labs) or a biotinylated secondary and 3, 3′ diaminobenzidine chromogen substrate. Slides were imaged on either a Leica DMi8 light microscope or a Zeiss LSM 800 confocal microscope.

Western blot

Brain hemispheres from treated and untreated animals were lysed in 1x STEN buffer and protein concentrations were measured via BCA assay. Samples were normalized to 1 μg/μl and added to 2x Laemmli loading buffer + β-mercaptoethanol and boiled for 10 min to denature the protein. 15 μg of sample were then loaded in 4–12% Bis–Tris gels with MES buffer and electrophoresed, before transfer to a nitrocellulose membrane, where they were blocked with 5% dry milk in 1x TBST and incubated overnight with an anti-Beclin-1 primary antibody (cat. #3739S; Cell Signaling) or a phospho-c-KIT antibody (cat. #44-496G; Invitrogen). The next day, the membrane was washed and treated for 1 h with a goat anti-rabbit HRP-linked secondary followed by SuperSignal West Dura Extended Duration Substrate (cat. #37071; Thermo Fisher Scientific) and imaged on an Amersham Imager 600. This procedure was repeated the next day with an anti-mouse actin primary as a loading control.

ELISA

All samples were analyzed in parallel using the same reagents. A total of 25 μl soluble protein was incubated overnight at 4°C with 25 μl of a mixed bead solution. After washing, samples were incubated with 25 μl detection antibody solution for 1.5 h at RT. Streptavidin–phycoerythrin (25 μl) was added to each well containing the 25 μl of detection antibody solution. Samples were then washed and suspended in 100 μl of sheath fluid. Samples were then run on MAGPIX with Xponent software. The median fluorescent intensity data were analyzed using a five-parameter logistic or spline curve-fitting method for calculating analyte concentrations in samples. Specific pTau ser396 (KHB7031; Invitrogen), aggregated Aβ (KHB3491; Thermo Fisher Scientific), and Aβ42 (KHB3442; Invitrogen) were performed according to manufacturer’s protocol on tissue soluble extracts from brain lysates in 1XSTEN buffer.

Microglial morphological analysis

63x z-stack images (0.19 μm intervals) of IBA1 stained microglia from treated and untreated mice were collected on a Zeiss LSM confocal microscope. 3D images were then analyzed using the “surfaces” plug-in in Imaris 10.0 to ascertain cell-surface area. Individual microglial surface areas were calculated in the advanced statistics tab of Imaris.

Proximity ligation assays were performed using the Millipore Duolink PLA protocol

Slides containing two consecutive brain slices were blocked in Duolink blocking solution for 1 h in a slide chamber containing water at 37°C. Blocking solution was then removed and PAR2 (1:250) and tryptase (1:250) primary antibodies were added at the concentrations specified above and incubated at 4°C overnight. The next day, primaries were removed and slides were washed twice for 5 min in wash buffer before addition of plus and minus probes for incubation at 37°C in a humid chamber for 1 h. Slides were then washed again twice for 5 min in wash buffer before addition of ligation buffer for incubation at 37°C in a humid chamber for 30 min. Next, slides were washed twice for 5 min in wash buffer, and polymerase was added for incubation at 37°C in a humid chamber for 100 min. Finally, slides were washed twice for 10 min in wash buffer and mounted using DAPI mounting medium and a glass coverslip. Slides were imaged on a Leica Dmi8 light microscope.

Statistical analysis

All statistics were calculated in GraphPad Prism 9. Two-tailed unpaired t tests were used to compare treatment groups. Ordinary one-way ANOVAs were used to examine the effects of our drugs at different doses in vitro. Two-way ANOVA was used on the Morris water maze to compare group differences over the course of the trials.

Study approval

All animal studies were approved by the Georgetown Animal Care and Use Committee GUACUC. This is not a clinical trial.