Fig 2: PGC-1a ameliorates novel object learning deficits and sensorimotor gating impairment displayed in AD mice. (A) The top panel showed the design of the standard novel object recognition task. During the habituation phase, the animal was allowed to explore an empty arena for 10 min. 24 h later, the animals were exposed to two objects (A,B) (Sample phase), and then after a 5 min interval, they received a 5 min test in which they were allowed to explore a duplicate of the familiar object (A) and a novel object (C) (Test phase). (B) The time spent exploring the novel object (C) for AD mice during the test phase was indicated as Discrimination index (DI). [The test phase: DI = (DC - DA)/(DC + DA)]. (C) The treatment (Vector/Pgc-1alpha) effect on distinguishing objects A and B for AD mice during the sample phase was indicated by Exploring preference (EP). [The sample phase: EPA = (DA)/(DA + DB); EPB = (DB)/(DA + DB)]. (D) The treatment (Vector/Pgc-1alpha) effect on % EP for AD mice distinguishing object A and C during the test phase was calculated using the formula [The test phase: EPA = (DA)/(DA + DC); EPC = (DC)/(DA + DC)]. (E) The treatment (Vector/Pgc-1alpha) effect on % Preference index (PI) for AD mice recognize the object A during the sample phase was calculated using the formula [The sample phase: PIA = (DA)/(DA + DB)], while during the test phase, it was calculated using the formula [The test phase: PIA = (DA)/(DA + DC)]. (F) The treatment (Vector/Pgc-1alpha) effect on % PI for AD mice recognize the object B during the sample phase was calculated using the formula [The sample phase: PIB = (DB)/(DA + DB)], while recognizing object C during the test phase was calculated using the formula [The test phase: PIC = (DC)/(DA + DC)]. (H) The mean startle magnitudes obtained in the first six and last six pulse-alone (120 dBA) trials are shown in the bar graph. (G) % PPI was expressed as a function of prepulse intensity (69, 73, 77, 81, and 85 dBA) for mice of the two treatments. % PPI = [(pulse alone-(prepulse plus pulse))/pulse alone × 100%]. Values were expressed as means ± S. E. M. For each group, n = 7–8. Significance levels were set at * p < 0.05, ** p < 0.01, *** p < 0.001 for noted differences between two treatments/objects/phases.

Fig 3: PGC-1a ameliorates apoptosis-relevant AD pathology. Expression patterns and qualification of (A,C) Bcl-2 and (B,D) BAX from the cortical samples of AAV-Pgc-1alpha-treated AD brains are showed. (E) Anti-apoptotic index, defined as the expression ratio of Bcl-2 to BAX was illustrated by bar graph. (F) Representative images of Nissl staining from the parietal cortex samples of Vector- and Pgc-1alpha- treated AD mice. Scale bar = 100 µm. (G) Nrs of dark neurons of Nissl substance between the two treatments were qualified and showed by bar graph. (H) Apoptosis rate was analyzed with flow cytometry after overexpressing pEnCMV/Pgc-1alpha and plasmid-encoding APPswe for 48 h in N2A cells. (I) Quantification of proportion of apoptotic cells of the two treatments were displayed by bar graphs. Values are expressed as means ± S.E.M. For each group, n = 6–8/group. Significance levels were set at ** p < 0.01, *** p < 0.001 for noted differences between APPswe + pEnCMV and APPswe + Pgc-1alpha groups, or APP/PS1 + Vector and APP/PS1 + Pgc-1alpha groups. Tubulin was used as the loading control.

Fig 4: PGC-1a rescues mutAPP-triggered dynamic imbalance of mitochondrial fission and fusion. N2A cells were transfected with pEnCMV/Pgc-1alpha plasmid and plasmid-encoding APPswe for 48 h. (A) Flag-labeled PGC-1a and EGFP-labeled APPswe were successfully overexpressed in N2A cells. Expression patterns and qualification of the fusion proteins, including (B,D) OPA1, (F,I) MFN1 and (G,J) MFN2, as well as the fission proteins, including (L,N) DRP1 and (P,R) FIS1, were studied with Western blot. Expression patterns and qualification of (C,E) OPA1, (H,K) MFN2, (M,O) DRP1 and (Q,S) FIS1 from the parietal cortex samples of 2×Tg-AD mice treated with Vector/Pgc-1alpha were also examined with immunohistochemistry. Scale bars = 100 µm. (T) Representative electron microscopes showing stage of mitochondria in cortical neurons of AAV-Vector- and AAV-Pgc-1alpha treated APP/PS1 mice. Scale bar = 0.5 µm. Yellow stars: the fissive stage of mitochondria. Significance levels were set at ** p < 0.01, *** p < 0.001 for noted differences between APPswe + pEnCMV and APPswe + Pgc-1alpha groups, or APP/PS1 + Vector and APP/PS1 + Pgc-1alpha groups. Tubulin was used as the loading control.

Fig 5: PGC-1a ameliorates spatial learning disability displayed in AD mice. Performance of path length to reach the hidden platform during the acquisition (Day 1–Day 6) and reversal phases (Day 8–Day 10) for animals was expressed as (B) a function of days and (A) across the 4 trials per day. Distance % in the target zone of the water maze pool during (C) probe test 1 (Day 7) and (D) probe test 2 (Day 11) are shown in the bar graphs. (E,F) present the search patterns made by individual animals during probe tests 1 and 2. Performance of escape latency to reach the hidden platform during the acquisition and reversal phases for animals was expressed as (H) a function of days and (G) across the 4 trials per day. Duration in the target zone of the water maze pool during (I) probe test 1 and (K) probe test 2 are shown in the bar graphs. Time was recorded in each of the four quadrants to index search preference for the initial platform location during (J) probe test 1 and (L) probe test 2 for the two treatments. Values were expressed as means ± S.E.M. For each group, n = 7–8. Significance levels were set at * p < 0.05, ** p < 0.01, *** p < 0.001 for noted differences between APP/PS1 + Vector and APP/PS1 + Pgc-1alpha groups.