Cis-P tau extraction

Pathogenic P-tau formation was induced by the traumatic brain injury (TBI) model [11]. In this model, TBI occurred by dropping a 450 g weight from 2 m height on skull of anesthetized adult male Wistar rats [47, 48]. It was purified from the brain extract after cis pT231- tau accumulation confirmation. The cortex tissues were separated and lysed in an extraction buffer (2 ml/g tissue). The contents of extraction solution include 20 mM PIPES pH 6.9, 1 mM MgSO4, 1 mM EGTA in the presence of 2 mM DTT, 1 mM PMSF, 1 mM EDTA, and 1 M NaCl. Then we pellet the homogenate by centrifugation at 6000 × g for 20 min at 4 ℃, and we sonicated the supernatant on ice four times (15 s on, the 30 s off) and boiled in 5 M NaCl for 10 min. We chilled the extract on ice and then ultra-centrifuged (Beckman coulter optima L-100XP) at 100,000 × g for 60 min at 4 ℃. The supernatant was dialyzed against PEM buffer (3 × 1). Finally, we purified tau protein by ion-exchange chromatography, as explained [49], and stored it at—80 ℃ until use.

Animals

Thirty adult male C57BL/6 mice (4 months old, weighing 25–27 g) were obtained from Tarbiat Modares University (Tehran, Iran) and housed in 21 ± 2 ℃, 12 h light–dark cycle, with free access to food and water. Animals were divided into control and cis-P tau groups. Cis-P tau or its solvent (saline) were injected intra-hippocampally in cis-P tau and control groups respectively. All experiments were run 7 months after chemical injection.

Stereotaxic surgery

Mice were anesthetized with an intraperitoneal injection of ketamine/xylazine (100 mg/kg to 10 mg/kg) and immobilized in a stereotaxic frame (Stoelting, USA). AD model was induced by cis-P tau injection (1 μg/1 μl) into the dorsal hippocampus (stereotaxic coordination: 2 mm posterior and 1.7 mm bilaterally to the right and left of bregma and—1.6 mm below dura [50]) in a volume of 2 μl over 6 min via a 10 μl Hamilton syringe using a microsyringe pump (WPI, UK). All microinjections were done at a speed of 0.5 μl/min, and the injection needle was left in place for an additional 10 min to allow the solution to diffuse from the tip entirely.

Y- maze test

We selected a Y-shaped gray Plexiglas maze with 30 cm length, 10 cm width, and 15 cm height for the working memory task. The animal was put on the end of one arm to explore for 8 min session freely. An entry happens when all four mouse limbs are inside an arm. An alternation is determined as successive entries into all three arms. Next, after recording the number of arm entries and alternations, the percentage of the alternation behavior was calculated by the below formula:

$$Alternation \,precentage=\frac\times 100$$

A spontaneous alternation happens when a mouse enters a different arm of the maze in each of 3 consecutive arm entries (i.e., visit from A to B or C, which are designated to the other arms, respectively). In addition, incorrect trials are considered to travel back to a previously experienced arm, such as CBC moving. All movements were recorded using a computer-linked video camera mounted above the platform [51], and data were analyzed by Ethovision software 11 (Noldus Information Technology, Wageningen, The Netherlands).

Barnes maze test

We assessed spatial hippocampal-dependent learning and memory using by Barnes maze test. The maze includes a circular platform (92 cm in diameter) with 20 holes (hole diameter: 5 cm) along with the surroundings. During the experiment, the mouse learned the spatial position of the goal box (17.5 cm in length, 7.5 cm in width, and 8 cm in height). Three-maze cues were placed all around the room to show the location of the goal box hole. In the pre-training trial, the mouse was put in the maze's center in a white-colored cubed start box (12.5 cm × 8 cm). After 10 s, the start box was raised, and the mouse learned to enter the goal box by guiding it to the goal box and staying there for 2 min. After the pre-training trial, the first trial began. At the onset of each trial, the mouse was put in the start box, and 10 s later, a light was turned on, the box was raised, and the mouse explored the maze freely. The trial finished when the mouse entered the goal box or after 5 min had elapsed. After entering the mouse into the goal box, the light was turned off, and the mouse remained in the goal box for 1 min.

Mice trained for four trials (at 15 min intervals) per day for 4 days. After each trial, the maze was cleaned with 70% ethylic alcohol solution. A probe trial was done on the 5 day when the goal box was closed to assess maze learning and memory retention. The probe experiment allows determining whether trained animals use environmental cues to create a spatial map of their environment and find a hole that was previously a goal box. The delay and time spent to find the last correct hole was measured. Total trials were recorded by using a ceiling-mounted video camera.

The measured behavioral parameters included: 1—Primary and total latency evaluated as the time spent by the mouse to find the goal box for the first time (primary latency) and entering (total latency) during a learning trial; 2—Errors measured as the number of incorrect holes explore before finding (primary errors) and entering (total errors) the goal box. Errors are explained as exploring any hole that does not contain the goal box; 3—Total distance and velocity for each trial are also calculated by using EthoVision XT; 4—For each trial, the search strategy (exploration patterns) is classified as direct (moving directly to the target hole), serial (systematic search of sequential holes in a clockwise or counterclockwise direction), and random (unordered and random exploration of the maze); 5—The frequency of target hole exploration was assessed by the goal sector (GS) parameter, and it is the sum target and a neighbor right or left holes explorations divided by 3; 6—The frequency of non-target hole exploration was assessed by the non-goal sector (NGS): the sum of explorations of the 17 non-goal holes divided by 17; 7—Goal sector preference: the ratio of GS to NGS explorations; 8—Target-seeking activity: the total explorations for whole, divided by 20 [24].

Field potential recording in the hippocampal slices

The mice were anesthetized with carbon dioxide (CO2) and decapitated at 7 months after bilateral hippocampal cis-P tau injection. We removed the mouse brain rapidly, and then the fresh brain was transferred into a chilled artificial cerebrospinal fluid (aCSF) chamber. This chamber bubbled with carbogen (95% O2 and 5% CO2). The aCSF contained (in mµ): NaCl 124, NaHCO3 26, KH2PO4 1.25, KCl 5, CaCl2 2, MgCl2 2.06, and d-glucose 10 and its pH was 7.3–7.4. Next, we prepared coronal 400 µm thick slices containing the hippocampus using a vibratome (model VT 1200, Leica, Germany). After slice preparation, we put the slices in a recovery chamber for at least 60 min at room temperature. Then we transferred slices (one by one) to an interface-type recording chamber containing 32 ℃ aCSF solution in a warm, humid oxygenated environment.

We recorded field potentials from the stratum radiatum in the dorsal and ventral hippocampus. We used a stimulating electrode (stainless steel, Teflon coated, A–M Systems, USA) that was placed on the Schaffer collateral path and a recording glass electrode (borosilicate, O.D.: 1.5 mm, I.D.: 0.86 mm, Sutter instrument, USA). The recording electrode (2–5 MΩ) was filled with aCSF and was placed on the stratum radiatum of hippocampal CA1. A reference electrode was also put in the recording chamber. The recording electrode transferred signals to an amplifier (ME208300, Nihon-kohden, Japan), and the signals were visualized by custom-made software (Potentialize; ScienceBeamCo., Iran). We plotted the Input/output curve to calculate test pulse intensity. The evoked field potential was recorded from the CA1 area at the test pulse intensity (50% of an intensity producing maximum response) for 20 min. Then, primed-burst stimulation (PBS; a single pulse followed 170 µs later by a burst of 10 pulses at 200 Hz, and the entire train was repeated ten times) was applied, and post-PBS responses were recorded for 60 min.

Tissue processing and sectioning

The mice were deeply anesthetized with ketamine/xylazine (100 mg/kg to 10 mg/kg) 7 months after cis P-tau injection. Then transcardial perfusion was performed with phosphate-buffered saline (20 mL) followed by 4% phosphate-buffered formalin (15–20 mL). Brain tissues were fixed overnight in the solution (4% paraformaldehyde in phosphate-buffered saline), next embedded in OCT compound (Sakura; Finetek; Torrance, CA), and cut into 8 μm thick serial sections.

Nissl staining

The number of survived cells was determined using the Nissl staining method. In brief, the sections were rehydrated with graded series alcohols (96%, 80%, and 70%) and stained with 0.1% Cresyl Fast Violet (Merck, Germany) at room temperature for 2 min. After washing, the sections were dehydrated by a graded series of alcohols (70%, 80%, 96%, and 100%). Then they were cleaned in xylene, cover slipped with Entellan (Merck, Chemical, Germany), and photographed. An Olympus BX-51 microscope and DP72 camera captured consecutive images at 400 × magnification. Using a grid (200 μm × 200 μm), the images were randomly assigned, and six squares were counted to measure survived cells, which were calculated as the number of cells/mm2.

Immunofluorescence

After washing with PBS-Tween, brain sections were permeabilized for 10 min with 0.2% (v/v) Triton X-100 and blocked for 1 h with NGS 10%. Afterward, the samples were incubated with the following primary antibodies: cis pT231-tau mAb (1:500, gift from KP. Lu), and Aβ oligomers (1:500, Abcam) at 4 ℃ in a moist and humid chamber overnight. A secondary antibody anti-rabbit or anti-mouse was added after washing the samples at 37 ℃ for one hour (Alexa Fluor 488, Thermo Fisher Scientific, Rockford, USA). DAPI was used for staining the nuclei. The samples were visualized by a fluorescent microscope (Olympus, BX51 with Olympus DP72 digital camera), and the images were analyzed by using ImageJ software v1.43 (NIH, Bethesda, MD, USA).

Statistical analysis

Spontaneous alternation, percent time in center point, total distance, test pulse, and percentage of potentiation were analyzed by unpaired t-test. Primary and total errors and latency, as well as field potentials (before and after LTP induction), were analyzed using two-way ANOVA followed by Sidak's multiple comparisons test. The correlation between the direct strategies in trial days was analyzed using the correlation test and Pearson correlation coefficient. The values were expressed as means ± standard error of the mean (SEM). All statistical analyses were conducted using Graphpad Prism (version 6.0). The probability level is interpreted as statistically significant when P < 0.05.