Effects of COX‑2 depletion on the learning and memory ability.
Firstly, it was observed that after genotyping COX-2KO (COX-2-/-) mice by PCR (Fig. 1A), no COX-2 protein was detected by Western blot (Fig. 1B).Then a comparison between WT and COX-2KO mice revealed no differences in the body weights, body temperatures, or basal inflammation (TNF-a, iNOS, and IL-1β) in the hippocampus (Fig.1C-E). Furthermore, to determine whether COX-2 contributes to learning and memory ability, we used Morris water maze (MWM) memory test and novel object recognition task (NORT). In the MWM test, the results showed that the escape latency of mice in each group decreased with the increase in training times, indicating that the training of mice was effective. Compared with WT group, the escape latency at training day 4 and 5 were significantly shorter for COX-2 knocked out mice (Fig.1C). Furthermore, to measure spatial learning and memory retention, we next performed probe trials on the 6th day. In the comparison with WT mice, COX-2KO ones failed to find the target platform with less time spending (Fig.1D, F) and fewer platform crossings (Fig.1E) in the target quadrant. In addition, no differences were observed between WT and COX-2KO mice in terms of swimming speed (Fig.1I) and total distance traveled (Fig.1J).Next, we used NORT to assess the recognition memory of different groups. As illustrated in Figure 1G, the recognition index of COX-2KO mice were reduced compared to the WT group at the test of both 1h and 24h. Moreover, the total distance traveled by COX-2KO and WT mice in the NORT (Fig.1M) was also not different. These results suggested that COX-2KO mice were impaired in learning and memory ability.
Effects of COX‑2 depletion on the SYP and PSD95.
Then we detected synaptophysin (SYP), PSD95 by double immunofluorescence and Western blot. We stained PSD95 with red and SYP with green in the hippocampus CA1 (Fig. 2A). We found there was almost no change in the fluorescence intensity of SYP between WT and COX-2KO those two groups, however, compared with WT group, the expression of PSD95 was almost absent in COX-2KO samples with a faint green staining (Fig. 2B). Consistent with the immunofluorescence result, we found COX-2 depletion could significantly downregulate the expression of PSD95 protein but not SYP in the hippocampus (Fig. 2C). These results indicated that COX-2KO could influence the expression of PSD95 instead of SYP in the brain.