The sequences of the synthesized TAMRA-labeled 13A, 13S and 13L were 5-TAMRA-KKWAAAAAAAAAAAAAKK-NH2, 5-TAMRA-KKWSSSSSSSSSSSSSKK-NH2 and 5-TAMRA-KKWLLLLLLLLLLLLLKK-NH2 (GL Biochem, Shanghai, China), respectively. The purity of 13A and 13S were more than 95%. However, that of 13L was 75% because purification of the peptide was difficult. The stock solutions of the peptides were prepared by dissolving the powder in a 1:1 mixture of trifluoroacetic acid and hexafluoroisopropanol. The appropriate conditions to induce aggregation of the peptides were determined in our previous literature22,24. 13L had highest aggregation property among the three peptides. It aggregated immediately after addition into culture media at a concentration of 10 µg/ml. By contrast, 13A and 13S needed long time incubation in aqueous solution for aggregation. Prior to addition to culture media, 13A and 13S were incubated in aqueous solution (1 mg/ml) for 7 days at 37 °C without shaking and for 2 days at 37 °C with shaking at 206 rpm/min., respectively, to induce aggregation.
Preparation of CM
BV2 microglial cell line was kindly provided by Dr. Choi (Korea University). The cell line was cultured in DMEM containing 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin mixed solution at 37 °C with 5% CO2. BV2 cells were plated on micro cover glass in 6-well plates at density of 8.0 × 104 cells per well. 13A, 13S and 13L were applied to the culture medium to be a final concentration of 10 µg/ml. The next day, the medium was removed and cultured cells were washed with PBS once to remove the peptides. Then, the cells were further cultured for 3 days more in the same culture medium without peptide to collect CM. As a different protocol, the culture medium was changed 4 days after the addition of 13L to BV2 cells. Then, BV2 cells were cultured for 2 days more to collect CM (Supplementary Fig. S2a). ELISA assay was performed using ELISA MAX Deluxe Set Mouse TNFα and IL-6 (BioLegend, San Diego, CA) according to the manufacturer’s instruction.
PC12 cell culture in the presence of CM
PC12 cell was purchased from RIKEN BRC. PC12 cells were plated on micro cover glass coated with laminin in 24-well plates at density of 5 × 103 cells per well. Differentiation of the cells was induced for 5 days in the presence of 50 ng/ml NGF in DMEM containing 1% FBS and 0.25% BSA. Then, the cells were further cultured for 4 days in the presence of CM without NGF.
Quantification of cultured cells and mouse neurons
Morphology of cultured cells and mouse neurons were quantified using Image J software as previously reported22. The original images were acquired by the software and the actual scale in the images was reflected to the acquired images. Then, length and area of interest were measured by tracing them using tools of the software. Because the length less than 1 µm was difficult to trace, the processes longer than 1 µm were measured. The area of cell body, total length of branches and thickness of proximal branches of BV2 microglia, the total length of neurites and number of branch points of the neurites of PC12 cells and area of cell body of the PnC neurons in mice were measured. Intensity of CD68 signals in BV2 cells was also measured using Image J software. Viability of cultured cells was determined using Cell Counting Kit-8 (Dojindo Molecular Technologies Inc., Rockville, MD) according to the manufacturer’s instruction.
All experimental protocols were approved by the Animal Resource Committees of Gunma University. All methods were carried out in accordance with relevant guidelines and regulations (NIH) and were reported in accordance with ARRIVE guidelines for the reporting of animal experiments. The number of mice used for experiments was bare minimum to obtain reliable data and we made every effort to minimize the suffering of mice during experiments. Mice were kept in specific pathogen-free conditions in a room where the temperature and light/dark cycle were constant (23 °C and 12 h, respectively). Male and female ICR mice 2–3 months of age were anesthetized with isoflurane and were fixed by stereotaxic instrument. Then, the mice received injection of CM. Two µl of CM or 13L (100 µg/ml) was injected into the PnC (AP − 5.35 mm; ML + 0.5 mm; DV − 5.6 mm)48. Five µl of CM was injected into the right lateral ventricle (AP + 0.2 mm; ML + 0.8 mm; DV − 2.5 mm)22. Behavioral tests were performed on the following day.
Fluorescent staining was done essentially as described previously17,22,49. BV2 cells were fixed with 4% paraformaldehyde (PFA). Mice were also transcardially infused with 4% PFA. After postfixation with same fixative solution overnight and dehydration with 30% sucrose in PBS, coronal brain sections 25 µm in thickness were prepared using cryostat.
The cytoplasm of BV2 cells was stained with Phalloidin-iFluor 488 or 647 conjugate (Cayman Chemical, Ann Arbor, MI) that binds to actin.
For immunofluorescence staining for anti-CD68 antibody (Abcam, Cambridge, UK) and anti-LAMP1 (Merck, Boston, MA) antibody, BV2 cells were incubated with the primary antibodies for 1 h in RT and overnight at 4 °C, respectively. Then, the cells were incubated with Alexa fluor 488-labeled secondary antibody for 1 h in RT. When anti-Iba1 antibody (Gene Tex, Irvine, CA) was used as primary antibody, 2 N HCl was applied to the brain sections for 10 min prior to the addition of the primary antibody overnight at 4 °C. Then, the sections were incubated in HRP-labeled secondary antibody solution for 1 h in RT. Finally, tyramide 488 solution (Thermo Fisher Scientific, Waltham, MA) was applied for 10 min in RT. The fluorescent signals were detected using LSM 880 confocal microscope (Zeiss, Oberkochen, Germany). When the confocal images were acquired, the frame size was 512 × 512 pixels and the scan time was 7.45 s. Line mode and bit depth of 8 bit were applied for averaging. The serial z-stack images of BV2 microglial cells at every 1 µm were also obtained by the confocal laser scanning microscopy.
Nissl staining of brain sections was also carried out essentially as reported previously22. The sections were stained with 0.1% cresyl violet solution for 20 min at 60 °C. The visible images were taken by BZ-9000 microscope (Keyence, Osaka, Japan) or ECLIPS 80i microscope (Nikon, Tokyo, Japan).
Resident-intruder test was performed essentially as described previously52. An intruder mouse that had not been a resident in the cage was put in the cage where a target mouse with the same sex has been a resident for 2 days. The total time the resident mice was showing active behavior (approaching, chasing, sniffing) against the intruder mouse during 5 min was measured.
Size of the box used for three-chambered social approach task (O’HARA & CO., LTD., Tokyo, Japan) is 20 × 40 × 23 cm. There are three chambers (left, center, right) with an equal size in the box. Mice can enter each chamber through the entrances (5 × 3 cm) between the chambers. Each subject mouse was first put in the center chamber and was allowed to move for 5 min (first trial). After 5 min, the subject mouse was taken out. After a different mouse was placed in the cage in one side of the chamber, the subject mouse was again put in the center chamber and the mouse behavior was automatically measured for 5 min (second trial). Social approach was estimated by the time subject mouse spent in the chamber with the mouse. Then, third trial was conducted by adding a novel mouse in the cage that had been empty in the second trial for 5 min, leaving the old mouse in another chamber. Social memory was evaluated by the time subject mouse spent in the chamber with the novel mouse.
The elevated plus maze test was performed as previously reported22. Two open and two closed arms of the maze were 25 cm long and 5.5 cm wide. The closed arms but not open arms had transparent walls with 14.5 cm in height at both sides. The center area of the maze was 5.5 cm × 5.5 cm. The maze was elevated above the ground (55 cm). Mice were first placed on the center area facing an open arm. Entering into the arms was defined when all four paws were on the arms. The total time spent in the open arms and total number of entries into the 4 arms were measured during 10 min.
The open field test was carried out as previously reported24. The apparatus (50 cm × 50 cm × 50 cm, O’HARA & CO., LTD., Tokyo, Japan) automatically measured spontaneous locomotion of mice for 10 min. The floor of the open field was covered with black paper to detect movement of the white mouse. Mice were initially placed in the center area and were allowed to walk freely. The parameters measured were duration of movement during test session, total walking distance, total number of movements, average speed of locomotion during test session, speed during movement, walking distance per movement, duration per movement and the percentage of time the mice stayed in the center area.
The acoustic startle response and PPI of the response were recorded using an acoustic startle reflex measurement system (O’HARA & CO., LTD., Tokyo, Japan). Mice were put into a cylinder which was placed on an acceleration sensor in a chamber. A loudspeaker in the chamber produced sounds as the startle stimulus (40 ms) and the software automatically detected the startle responses. For mice given CM into the lateral ventricle, prepulses (20 ms) with 74, 78 or 82 dB was applied 100 ms before the startling acoustic stimuli. Startle responses at 110 and 120 dB and six combinations of PPI (74–110, 78–110, 82–110, 74–120, 78–120, 82–120 dB) were presented 6 times in a pseudorandom order. The intensities of startling stimuli applied (110 and 120 dB) were chosen because the tone at 120 dB was extensively used for 11 inbred strains of mice50 and the startle response of ICR mice became big using tones at 110 dB and 120 dB compared to those at 80–100 dB51. The prepulse levels (74, 78, 82 dB) were chosen because 74–90 dB were applied for the 11 inbred strains50 and 73–86 dB were used for ICR mice51. The averaged value after each stimulus type was calculated for each mouse. The interval time of each trial ranged from 10 to 20 s.
The behavioral tests using mice given CM in the lateral ventricle were done in the following order with intervals of 10 min; resident-intruder test, three chamber test, elevated plus maze test, open field test, startle response and PPI. Different sets of mice were used for startle responses of mice given 13L-CM in the PnC. All behavioral tests after injection into the lateral ventricle were done using 5 males and 5 females, each group. For startle responses after injection into the PnC, 5–10 males and 5 or 6 females per groups were used.
The values expressed are the mean in the graphs. The error bars represent SE. Statistical significances were examined using one-way ANOVA except three chamber test to which two-way ANOVA was applied. Tukey and Tukey–Kramer posthoc tests were applied when the sample sizes were same and different, respectively. p values less than 0.05 were considered as statistically significant.