Baylor College of Medicine (BCM) Institutional Animal Care and Use Committee (IACUC) approved all mouse care and manipulation. Mice were housed in an American Association for Laboratory Animal Science-certified level three facility. All mice were maintained in a 14 h light/10 h dark cycle at 68–72 °F and 30–70% humidity, with standard mouse chow and water ad libitum. Mice were monitored daily by veterinary staff. All procedures to maintain and use these mice were reviewed and approved animal protocol (AN-1013) by the BCM IACUC in accordance with the guidelines of the US National Institutes of Health (NIH).

Materials

The materials used in the present study are given in Supplementary Table 2.

Mouse models

CFW mice were used for in vivo validation of Tyk2. Timed pregnant Swiss Webster (CFW) mice or FVB mice were used for embryonic primary cortical neuronal culture experiments. Tau P301S transgenic mice (PS19 line, (C57BL/6 x C3H) F1) were bred in our laboratory. Tyk2 knockdown mice in PS19 background were generated by ICV injection of the recombinant AAV8 harboring Tyk2 shRNA. Transgenic offspring for these experiments were generated by mating PS19 males with FVB females. Pathogenic tau mice (P301S tau 1N4R) were generated by ICV injection of AAV harboring tau441-P301S or tau441-Y29F/P301S with/without human TYK2KD in FVB background.

Antibody dilution

The following primary antibodies were used: anti-TYK2 (Abcam, ab303500; 1:1,000); antiphospho-TYK2 (p292; Abcam, ab138394; 1:1,000); PHF1 (Peter Davies, 1:2,000); antitau (Abcam, ab80579; 1:4,000); antitau (Agilent, A0024; 1:10,000); anti-HA (BioLegend, 901514; 1:4,000); anti-Flag (M2; Sigma-Aldrich, F3156, 1:4,000); anti-ptau (pT205; Thermo Fisher Scientific, 44-738G; 1:2,000); antitau, oliogomeric (Merk Millipore, ABN454-I; 1:1,000); antiphospho-Tyr (Merk Millipore, 05-321; 1:1,000); anti-FYN (Cell Signaling Technology, 4023; 1:1,000); antivinculin (Sigma-Aldrich, V9131; 1:2,0000); anti-GAPDH (ImmunoChemical, 2-RGM2; 1:20,000); anti-GFAP (Norus Biologicals, 53809; 1:1,000); anti-IBA1 (Wako, 019-19741; 1:1,000); antiphospho-tau (pTyr29; not available, 1:1,000); antiphospho-tau (pTyr18; MediaMabs, MM-0194-P; 1:1,000); antiphopsho-tau (pS396); PHF13 (Cell Signaling Technology, 9632S; 1:1,000); anti-DUSP (ABclonal, A2919; 1:1,000) and anti-β-catenin (Cell Signaling Technology, 8480; 1:1,000).

The following secondary antibodies were used: donkey antirabbit IgG, Alexa Fluor 555 (Thermo Fisher Scientific, A-31572; 1:500); donkey antigoat IgG, Alexa Fluor 594 (Jackson ImmunoResearch, 705-585-003; 1:500); donkey antirabbit IgG, IRDye 800CW (Li-COR Bioscience, 926-32213; 1:10,000); goat antirabbit IgG, IRDye 680RD (LI-COR Bioscience, 926-68071; 1:10,000); goat antimouse IgG, IRDye 800CW (LI-COR Bioscience, 926-32210; 1:10,000) and goat antimouse IgG, IRDye 680RD (Li-COR Bioscience, 926-68072; 1:10,000).

Method detailsCloning

We designed four to five individual shRNA sequences per target gene, using the SplashRNA algorithm64. The shRNA sequence was amplified by PCR, using one pair of oligomers set (miRE-Gib-forward, 5′-ttcttaacccaacagaaggctcgagaaggtatattgctgttgacagtgagcg; miRE-Gib-reverse, 5′-gtaaacaagataattgctcgaattctagccccttgaagtccgaggcagtaggca), followed by integration into pAAV-YFP-miRE vector through Gibson cloning as previously described65.

Precision LentiORFs are human cDNA open reading frames (ORFs) cloned into a lentiviral expression vector to overexpress the given human proteins in mammalian cells.

Human TYK2 and TYK2KD, containing flag tag at the C-terminal, were amplified by PCR and cloned into the linearized pLenti-EF1a or pAAV-chicken β actin promotor cut by BamH1/EcoR1 or EcoR1/Not1 using a Gibson Assembly reaction. For tau441 WT containing mutation (Y18F, Y29F, Y197F, Y310F or/and Y397F), tau441-P301S or tau441-Y29F/P301S, two split DNA fragments were amplified, introducing mutation by mutagenic primer, and were joined by assembly reaction to generate full-length tau cDNA. The primers used are given in Supplementary Table 3.

Viral production

Recombinant virus containing shRNA or ORF was produced in low-passage HEK293T cells via triple transfection at 80–90% confluency, using TransIT-293 transfection reagent according to the manufacturer’s instructions. For lentiviral packing, cells were transfected with viral shuttle vector (pGIPz, pLenti and pLOC), psPAX2 and pMD2.G at a 4:3:1 ratio. The medium was collected at 48 and 72 h post-transfection. Lentivirus was concentrated 50-fold using a Lenti-X concentrator (Clontech, 631231) in case a higher titer and more purified viruses were required. Recombinant AAV8 was produced using a triple transfection with pAAV, capsid and helper plasmid in a 150 mm dish format, and the resultant virus was purified and concentrated on an iodixanol step gradient, as previously described66,67. To increase the yield of virus particles, cell-associated and medium-containing secreted AAVs were collected separately at 72 h post-transfection and combined before purification.

Cell culture, transfection, infection and drug treatment

All cell lines were cultured according to the manufacturer’s protocol. Cells were transfected with plasmids encoding cDNA or shRNA using Lipofectamine 2000 or TransIT-293 transfection reagent, according to the manufacturer’s protocol, and incubated for 48–72 h followed by the designed experiments. Cells were infected with lentivirus expressing ORF at 3–10 multiplicity of infection (MOI) in the various cell culture size formats. After 24–48 h of infection, cells were incubated with the appropriate antibiotics at least for 72 h. Cells were further maintained until additional treatment or collection.

Cortical neurons isolated from E16.5 FVB embryos were plated at a density of ~300,000 per well on poly-d-lysine (0.1 mg ml−1)-coated 24-well plates. Neurons were maintained in NB-B27 medium (neurobasal plus medium + 1X B27 supplement, 0.5 mM glutamax and 0.2× penicillin–streptomycin; Invitrogen). At day in vitro 3 (DIV3), neurons were infected with recombinant lentivirus at 3–5 MOI. At DIV10, neurons were treated with DOX at a concentration of 50 nM to initiate transgene expression for 24 h.

ICV injection of AAV at postnatal day 0 (P0-ICV injection of AAV)

The knockdown efficiency of each shRNA in pAAV was tested in Neuro2A cells as previously described68. The two most potent shRNAs targeting Tyk2 or Luciferase gene as NT control were packed into AAV8 (ref. 68). AAV8 harboring a shRNA expression cassette was bilaterally infused into the mouse brain at 4–6 × 1010 viral genomes per hemisphere using ICV injection at postnatal day 0 as previously described69. For protein overexpression in the brain, AAV8 containing cDNA expression cassette injection at 1–2 × 1010 viral genomes per hemisphere. For tau pathology, transgenic offspring were generated by mating PS19 males with FVB females.

Mouse brain sample preparation

Mice were killed by isoflurane inhalation at 3 weeks or 2 months postinjection. The forebrain region was collected and immediately frozen on dry ice. For pathology, PS19 mice and age-matched WT littermate mice were killed by sodium pentobarbital overdose and transcardially perfused with ice-cold PBS at 9 months postinjection. The whole hemisphere was isolated and fixed by 4% paraformaldehyde (PFA) for 48 h followed by cryopreserving in 1× PBS containing 30% sucrose at 4 °C for histological study. For the biochemical study, cortex and hippocampus from the leftover hemisphere were dissected together and frozen immediately. Frozen samples were homogenized using an electric pestle (handheld polytron; WPR, 47747-370) in 10× volumes/weight of cold resuspension buffer (1× PBS containing 5 mM EDTA, protease inhibitor cocktail and phosphatase inhibitor cocktail). Part of the homogenate was then diluted 1:1 with RIPA buffer (1× PPS containing 5 mM EDTA, protease inhibitor cocktail, phosphatase inhibitor cocktail, 1% deoxycholate, 1% Triton X-100 and 0.5% SDS). After mixing well, the supernatant was collected by centrifugation at 15,500g for 20 min for IB analysis. For the preparation of detergent-insoluble tau, brain homogenate removed tissue debris by centrifugation (2,000g for 5 min at 4 °C) was mixed with the same volume of 2× concentrated RIPA buffer without SDS and was centrifuged at 100,000g for 30 min at 4 °C. The supernatant was collected (S1), and the pellet was dissolved by incubation with 1× RIPA containing 1% sarkosyl (without SDS) for 1 h at room temperature on an orbital shaker (P1). The soluble fraction (S2) was collected by centrifugation at 200,000g for 30 min at 4 °C. Pellet was dissolved in 1× PBS by sonication (probe sonicator, 30% amplitude, pulsed ten times with 2′/1′ on/off; P2). Each sample was mixed with protein-loading dye and boiled at 95 °C for 5 min. RNA was further purified from homogenate using the RNeasy mini kit (Qiagen) as needed.

Human tissue sample preparation and IB analysis

Postmortem brain tissues from participants with AD and control participants were provided in the form of frozen blocks by the Massachusetts Alzheimer’s Disease Research Center, which has Institutional Review Board (IRB) approval for the tissue bank and for sharing deidentified autopsy samples (IRB 1999P009556). Specifically, frozen tissues from the middle temporal cortex of patients with AD (n = 17) and control individuals (n = 12) were obtained (Supplementary Table 1). Tissue donor anonymity was assured by the Massachusetts Alzheimer’s Disease Research Center. AD cases consisted of pathologically severe AD, stages V–VI. Each brain was homogenized in RIPA buffer with a protease inhibitor cocktail (Roche) and diluted in RIPA to 1:10 (wt/vol). Samples were then centrifuged at 16,363g for 15 min at 4 °C. The supernatants were portioned into aliquots, snap-frozen and stored at 80 °C until analyzed. Samples were mixed with running buffer, run on a gel and analyzed by IB. Primary antibodies used were anti-pTyk2 (pY292; 1:1,000), PHF1 (1:2,000, gift from P. Davies), antivinculin (1:20,000) and GAPDH (1:50,000).

Cell lysate preparation

Frozen cells in plates were thawed on ice briefly and were lysed by gentle mixing with ice-cold cell-lysis buffer (50 mM Tris, 150 mM NaCl, 2 mM CaCl2, 1% Triton X-100, 5 mM EDTA and 5% glycerol (pH7.5)) containing 1× protease and phosphatase inhibitor cocktails on shaker for 30–60 min at 4 °C. Cell supernatant was collected by centrifugation at 15,000g for 30 min and then subjected to further experiments. The total protein concentration of brain tissue and cell lysate was determined with a BCA Protein Assay Kit. For detergent-insoluble tau faction, cell lysate in ice-cold lysis buffer was subjected to spinning down at 1,250g for 10 min at 4 °C, and then the supernatant was further centrifuged at 100,000g for 30 min at 4 °C. Supernatant was collected as detergent-soluble tau fraction. The pellet was dissolved in 1× PBS by sonication (probe sonicator, 30% amplitude, pulsed ten times with 2′/1′ on/off; P2). Each sample was mixed with protein-loading dye and boiled at 72 °C for 10 min.

Immunoprecipitation

Supernatant from one well of a six-well plate format was incubated with 0.75–1 µg of antitau antibody or 1 µg of anti-Flag antibody conjugated with protein G-conjugated Dynabead (15 µl slurry) for 4 h at 4 °C. After washing with lysis buffer four times, immunoprecipitated proteins were eluted in 30 µl of 1.5× Laemmli sample buffer by boiling 75 °C for 10 min.

IB analysis

Lysate (1–1.5 μg μl−1) in 1× sample buffer was denatured at 75 °C for 10 min. To detect tau from the cultured cells, 10–15 μg of the sample was incubated with antitau antibody (Dako; 1:8,000), whereas 5–10 μg or 2–5 μg of sample from WT or PS19 mouse brain was detected by antitau antibody (1:4,000). Protein samples were resolved on precast 4–12% Bis–Tris gels (3-(N-morpholino)propanesulfonic acid running; Invitrogen) and transferred onto nitrocellulose membranes (Bio-Rad, 1620145) in Tris–glycine buffer supplemented with 10% methanol at 120 mV for 100 min at 4 °C. After blocking 30 min with 5% bovine serum albumin, the membrane was incubated with the indicated primary antibodies overnight at 4 °C. The protein bands were visualized using corresponding fluorescent secondary antibodies. Infrared fluorescence was measured with the Odyssey CLx imager (LI-COR) and quantified using Image Studio software version 5.2 (LI-COR Biosciences).

FRET measurement of seeding activity

Tau-seeded transduction of the tau RD P301S FRET biosensor cell line and flow cytometry analysis were conducted as previously described12. Lipofectamine diluent consisting of 7.5 μl Opti-MEM (Gibco, 31985070) and 0.1−0.4 μl Lipofectamine 2000 (Invitrogen, 11668-500) mixed with proteopathic seed diluent consisting of 7.5 μl Opti-MEM and 0.5–2.0 μg protein extract from brain homogenate of tauopathy mouse model was incubated for 30 min at room temperature and added to cells in a 96-well plate format with 60–70% confluency. After 24–48 h of seed transduction, cells were dissociated into single cells in 1× PBS containing 2 mM EDTA and 3% fetal bovine serum. Upon excitation by a 405 nm laser, we measured the FRET signal with the BD LSRFortessa cell analyzer. For data collection, BD FACSDiva Software (v9.0) was used. For each experiment, we recorded cyan fluorescent protein (CFP) and FRET-YFP signals of 50,000–80,000 singlet events at 485/22 nm filter and 525/50 nm filter, respectively. In the bivariate plot, to assess the number of FRET-positive cells, we created a gate by using the FRET-negative signal exhibited by biosensor cells treated with Lipofectamine alone and the FRET-positive signal exhibited by HEK293T cells expressing CFP fused with YFP.

In vitro tau aggregation assay in a cellular model

To assess in vitro tau aggregation upon extracellular tau seeding, we used sarkosyl-insoluble tau fraction purified from 8- to 10-month-old PS19 mice as seeding material, as previously described (‘Mouse brain sample preparation’). The concentration of tau was determined by the tau ELISA kit.

The stable cell line (293T: tau441-P301S-YFP or 293T: tau441-Y29F/P301S-YFP) as seed-recipient cells were plated at 50,000 cells per well in a 24-well format plate and were reverse-transfected with plasmids containing cDNA of RFP, TYK2, TYK2KD or FYN with or without CAPON, using TransIT-293 transfection reagent according to the manufacturer’s protocol. Twenty-four hours later, seed transduction complexes (mixture of 15 μl Opti-MEM + 0.5 μl Lipofectamine 2000 and 15 μl Opti-MEM + proteopathic seeds) were incubated for 30 min at room temperature and added to cells. At 24–48 h after seed transduction, the cell medium was replaced with 2% Triton X-100 in 1× PBS, and cells were incubated for 1 min at room temperature. Then cells were added with the same volume of 2% hexadecyltrimethylammonium bromide and further incubated for 10 min at room temperature to remove soluble proteins. Cells were then fixed with 4% PFA for 20 min at room temperature. The whole plates were then scanned to capture tau-YFP aggregates using a cell imaging microplate reader (BioTek, Cytation5 Cell Imaging Reader) at ×25 magnification (3,402 × 3,001 pixels, 13.2 × 11.6 μm per field) and analyzed using Fiji software. For image-data acquisition, BioTek Gen6 Data Analysis Software was used. For total tau control, the total YFP signal from each well was captured before the extraction of soluble protein.

Immunofluorescence and microscopy assay

Fixed brains were sagittally sectioned (40 µm thickness) using a sliding microtome (Leica, SM2010 R). Floating sections located 1–1.5 mm from the midline were selected for staining and analysis based on landmarks in the hippocampus, lateral ventricle and striatum. A total of three to four sections were imaged per animal. Brain sections were immunostained with GFAP (1:2,000) and IBA1 (1:2,000) antibodies at 4 °C overnight. Sections were then incubated in Alexa Fluor-conjugated goat secondary antibodies corresponding to primary antibodies and Hoechst (1:2,000) at 4 °C overnight. Fluorescence images were collected from nonoverlapping fields within the somatosensory cortex (above CA1), hippocampus CA1 and the dentate gyrus. A single optical plane of 0.977 µm in depth was collected in blue (Hoechst) and red (GFAP or IBA1) channels using fluorescence microscopy (Carl Zeiss) at ×100 magnification (895.26 µm × 670.8 µm per field). For representative images, z-stacked fluorescence images (14 optical images) were collected from the same field of the cortex and hippocampus CA1 region using confocal microscopy (Carl Zeiss) at ×200 magnification (416.80 µm × 416.80 µm). For image-data acquisition, ZEN microscopy software was used. Then the collected image data were analyzed using Fiji software.

Data collection and statistical analyses

All statistical analyses were done using GraphPad Prism 9.0. No statistical methods were used to predetermine sample sizes, but our sample sizes are similar to those reported in previous publications12. All data collection and analysis were performed blind to the condition of the experiments. For the purpose of data analysis, all data were collected, excluding data from experimental groups where the positive and negative controls did not function correctly. For animal experiments, because our study involved the entire population of mice rather than a sample subset, traditional randomization techniques were not applicable. Our analysis was conducted on the full cohort, ensuring that all data points contributed to the final results. For the scatter dot plot, data distribution was assumed to be normal, but this was not formally tested. For other types of plots, normality and equal variances were tested by the Shapiro–Wilk test. All comparisons between groups were two-sided. Comparisons between the two groups were analyzed using unpaired t test. All comparisons involving more than two groups were analyzed using one-way analysis of variance (ANOVA) followed by Bonferroni post hoc tests, Dunnett’s multiple comparison test, Tukey’s multiple comparison test and Sidak’s multiple comparison test as indicated in figure legends. All reported P values are for post hoc comparisons. Adjustments were made for all multiple comparisons. All graphs display group mean ± s.e.m. (*P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001).

Inclusion and ethics

We worked to ensure sex balance in the selection of nonhuman participants. The author list is gender balanced, and we worked to achieve gender balance in our reference list.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.