Microglial DBP Signaling Mediates Behavioral Abnormality Induced by Chronic Periodontitis in Mice

Abstract

Several lines of evidence implicate that chronic periodontitis (CP) increases the risk of mental illnesses, such as anxiety and depression, yet, the associated molecular mechanism for this remains poorly defined. Here, it is reported that mice subjected to CP exhibited depression-like behaviors and hippocampal memory deficits, accompanied by synapse loss and neurogenesis impairment in the hippocampus. RNA microarray analysis disclosed that albumin D-site-binding protein (DBP) is identified as the most prominently upregulated target gene following CP, and in vivo and in vitro immunofluorescence methods showed that DBP is preferentially expressed in microglia but not neurons or astrocytes in the hippocampus. Interestingly, it is found that the expression of DBP is significantly increased in microglia after CP, and knockdown of microglial DBP ameliorated the behavioral abnormality, as well as reversed the synapse loss and hippocampal neurogenesis damage induced by CP. Furthermore, DBP knockdown improved the CP-induced hippocampal inflammation and microglial polarization. Collectively, these results indicate a critical role of DBP in orchestrating chronic periodontitis-related behavioral abnormality, hippocampal synapse loss and neurogenesis deficits, in which the microglial activation may be indispensably involved.

Keywords: albumin D‐site‐binding protein; chronic periodontitis; depression‐like behaviors; hippocampus; microglia.

Figure 1

Chronic periodontitis induces anxiety‐ and depression‐like behaviors in mice. A) Experimental schedule of chronic periodontitis (CP) modeling and tests of emotional behaviors. B) Changes of the average body weight during the 28‐days CP modeling. C‐D) The sucrose preference of Ctrl mice and Sus or Res mice subjected to CP in SPT. E‐F) The immobility time of Ctrl, Sus, and Res mice in TST (E) and FST (F). G) The movement traces, open‐arm entries, open‐arm duration, and open‐arm distance of Ctrl, Sus, and Res mice in EPM. H) The movement traces, center crossing times, center duration, and center distance of Ctrl, Sus, and Res mice in OFT. I) The buccal and palatal structure of teeth and periodontal tissue from Ctrl, Sus, and Res mice. J‐K) The statistical results of CEJ‐ABC on buccal J) and palatal K) sides of Ctrl, Sus, and Res mice. L) The representative H&E staining of periodontal tissues from Ctrl, Sus, and Res mice. All data are presented as mean ± SEM. ns, no significant difference. Student's t‐test for (B), two‐way ANOVA for (D‐F, G, H, J, K). n = 25, 12, 14 per group for (C‐H); n = 6 per group for (J, K). *p < 0.05, **p < 0.01 and ***p < 0.001 versus Ctrl group; ^# p < 0.05, ^## p < 0.01 and ^### p < 0.001 versus Sus group.

Figure 2

Chronic periodontitis induces hippocampal memory deficits in mice. A) Experimental schedule of CP modeling and behavioral tests related to learning and memory. B) Data analysis of Morris water maze test of Ctrl and CP mice, including the movement traces (B1, B3), the latency to the platform area (B2, B4) during training and testing, the platform crossing times (B5), the time spent in each quadrant (B6) and average speed (B7) on the probe trail at Day 8. C) The freezing time duration of Ctrl and CP mice in cue (C1) and contextual (C2) fear condition test at Day 3. D) Data analysis of Barnes maze test of Ctrl and CP mice, including the movement traces (D1, D3), the latency to target hole for the first time (D2, D4) during training and testing, the time spent nosing the target hole (D5), target hole entries (D6), accuracy of target hole (D7), distance traveled (D8) and average speed (D9) at Day 5. The accuracy of target hole is calculated as the ratio of time duration in target hole area/ the total time spent nosing the target and non‐target holes. E) Time‐course of LTP induced by HFS in hippocampal slices from the Ctrl and CP groups (E1), and the statistical results relative to baseline in Ctrl and CP groups (E2). All data are presented as mean ± SEM. Student's t‐test for all. n = 10 per group for (B); n = 6 per group for (C); n = 12, 13 per group for (D); and n = 10 per group for (E). *p < 0.05, **p < 0.01 and ***p < 0.001 versus Ctrl group.

Figure 3

Chronic periodontitis induces synapse loss and neurogenesis impairment in the hippocampus. A) Gene ontology (GO) enrichment bubble map of differentially expressed genes (DEGs) between Ctrl and CP mice showing the top 20 enriched biological process terms. B) GO enrichment histogram of DEGs between Ctrl and CP mice showing the top 20 enriched biological process terms. C) GO functional enrichment network of DEGs between Ctrl and CP mice. D) GO enrichment upset map of DEGs between Ctrl and CP mice. E) Representative images of Nissl staining in the hippocampus of Ctrl and CP mice. Scale bar, 300 µm in left and 100 µm in enlarged images. F) Statistical results of Nissl's staining in the hippocampal sub‐regions of CP mice relative to Ctrl mice. G) Double‐staining of PSD‐95 (blue) and synaptophysin (green) in the hippocampus of Ctrl and CP mice. The right are magnified images showing the colocalization of PSD‐95 and synaptophysin. Scale bar, 300 µm in left and 100 µm in enlarged images. H‐J) Statistical results of PSD‐95 intensity (H), synaptophysin intensity (I) and colocalized ratio between PSD‐95 and synaptophysin (J) in the hippocampal sub‐regions of CP mice relative to Ctrl mice. K‐M) Representative Western blots and quantification of PSD‐95 and synaptophysin expression in the CA1 (K), DG (L), and CA3 (M) of CP mice relative to Ctrl mice. N) Representative BrdU‐ and DAPI‐staining of hippocampus from Ctrl and CP mice. Scale bar, 100 µm. O) Representative Ki67‐ and NeuN‐stained images of hippocampus from Ctrl and CP mice. Scale bar, 100 µm. P) Representative DCX‐stained images of hippocampus from Ctrl and CP mice. Scale bar, 100 µm. Q) Quantification of the relative number of BrdU‐, Ki67‐ or DCX‐ positive neurons in hippocampus from Ctrl and CP mice. All data are presented as mean ± SEM. Student's t‐test for all. n = 12 per group for (F); n = 9 per group for (H‐J); n = 6 per group for (K‐M); and n = 8 per group for (Q). *p < 0.05, **p < 0.01 and ***p < 0.001 versus Ctrl group.

Figure 4

Microglial DBP signaling is required for depression‐like behaviors and memory deficits induced by chronic periodontitis. A) The principal component analysis (PCA) table showing the sample repeatability of Ctrl and CP mice. B) Clustering heat map of DEGs in the hippocampus of CP mice relative to Ctrl mice. C) Correlation matrix showing the correlation between samples of hippocampus from Ctrl and CP mice. D) Volcano map showing the number of DEGs in hippocampus between Ctrl and CP mice. E) Immunofluorescence of DBP colocalized with cell specific markers for neuron (NeuN), astrocyte (GFAP) and microglia (Iba1) in hippocampus of C57 mouse, respectively. Scale bar, 100 µm. F) Pie chart showing the distribution of DBP among astrocyte, neuron and microglia in the hippocampus. G) Immunofluorescence assay results showing the colocalization of DBP within BV2 cell cultured in vitro. Scale bar, 100 µm. H) Representative Western blots and quantification of DBP expression in the neuron, microglia, and astrocyte of hippocampus samples from Ctrl, Sus, and Res mice. I) Experimental schedule of viral microinjection in the hippocampus of Cx3cr1‐iCre mice, CP modeling, and tests of emotional behaviors and memory. J‐L) Effects of specific microglial DBP knockdown on depression‐like behaviors, such as sucrose preference in SPT (J), immobility time in TST (K) and FST (L). M) Representative movement traces of Morris water maze test on learning stage. N) Effects of specific microglial DBP knockdown on average latency to platform in Morris water maze test on learning stage from Day 1 to Day 6. O) Representative movement traces of Morris water maze test on the probe trail at Day 8. P‐S) Effects of specific microglial DBP knockdown on the latency of the first reach to the platform area (P), platform crossing times (Q), total time spent in different quadrants (R) and average speed (S) in Morris water maze test on Day 8. All data are presented as mean ± SEM. Two‐way ANOVA for all. n = 6 per group for (H), and n = 13 per group for (J‐L, N, P‐S). *p < 0.05, **p < 0.01 and ***p < 0.001 versus Ctrl group or AAV‐eGFP + Ctrl group. ^# p < 0.05, ^### p < 0.001 versus Sus group or AAV‐eGFP + CP group.

Figure 5

Knockdown of microglial DBP reverses synapse loss and neurogenesis impairment following chronic periodontitis. A‐D) Representative immunofluorescence images and statistical results of Nissl staining in the three sub‐regions of hippocampus from the Ctrl and CP mice treated with AAV‐eGFP or AAV‐DBP. The right is magnified staining images of hippocampal sub‐regions of CA1, DG, and CA3. Scale bar, 300 µm in left and 100 µm in enlarged images. E) Double‐staining of PSD‐95 (blue) and synaptophysin (green) in the hippocampus of Ctrl and CP mice treated with AAV‐eGFP or AAV‐DBP. The right is magnified staining images of hippocampal sub‐regions of CA1, DG, and CA3. Scale bar, 300 µm in left and 100 µm in enlarged images. F‐H) Statistical results of PSD‐95 intensity (F), synaptophysin intensity (G), and colocalized ratio between PSD‐95 and synaptophysin (H) in the hippocampal sub‐regions of CP mice relative to AAV‐eGFP + Ctrl mice. I‐K) Representative Western blots and quantification of PSD‐95 and synaptophysin expression in the CA1 (I), DG (J), and CA3 (K) sub‐regions of CP mice relative to AAV‐eGFP + Ctrl mice. L) Representative BrdU‐ and DAPI‐stained images, Ki67‐ and NeuN‐stained images, and DCX‐stained images of hippocampus from Ctrl and CP mice treated with AAV‐eGFP or AAV‐DBP. Scale bar, 100 µm. M‐O) Quantification of the relative number of BrdU positive (M), Ki67 positive (N), and DCX positive (O) neurons in hippocampus from Ctrl and CP mice treated with viral injection. All data are presented as mean ± SEM. Two‐way ANOVA for all. n = 12 per group for (B‐D, F‐H, M‐O), and n = 6 per group for (I‐K). *p < 0.05, **p < 0.01 and ***p < 0.001 versus AAV‐eGFP + Ctrl group. ^# p < 0.05, ^## p < 0.01 and ^### p < 0.001 versus AAV‐eGFP + CP group.

Figure 6

Chronic periodontitis regulates microglial phenotype and neuroinflammation through DBP. A‐L) Statistical analysis of mRNA changes of IL‐1β (A), IL‐6 (B), TNF‐α (C), IFR‐γ (D), CCL2 (E), CXCL10 (F), iNOS (G), IL‐4 (H), IL‐10 (I), IL‐13 (J), TNF‐β (K) and Arg‐1 (L) in the hippocampus tissues from the Ctrl and CP mice treated with AAV‐eGFP or AAV‐DBP. M, N) Representative immunofluorescence images and statistical results of CD68 in the hippocampus from the Ctrl and CP mice treated with AAV‐eGFP or AAV‐DBP. The right is magnified staining images of hippocampal sub‐regions of CA1, DG, and CA3. Scale bar, 300 µm in left and 100 µm in enlarged images. O, P) Representative immunofluorescence images and statistical results of CD206 in the hippocampus from the Ctrl and CP mice treated with AAV‐eGFP or AAV‐DBP. The right is magnified staining images of hippocampal sub‐regions of CA1, DG and CA3. Scale bar, 300 µm in left and 100 µm in enlarged images. Q, R) Representative immunofluorescence images and statistical results of Iba1 in the hippocampus from the Ctrl and CP mice treated with AAV‐eGFP or AAV‐DBP. The right is magnified staining images of hippocampal sub‐regions of CA1, DG and CA3. Scale bar, 300 µm in left and 100 µm in enlarged images. S‐V) Representative Western blots and quantification of CD68, CD206 and Iba1 in hippocampus from the Ctrl and CP mice treated with AAV‐eGFP or AAV‐DBP. All data are presented as mean ± SEM. Two‐way ANOVA for all. n = 12 per group for (A‐L, N, P, R), and n = 6 per group for (T‐V). *p < 0.05, **p < 0.01 and ***p < 0.001 versus AAV‐eGFP + Ctrl group. ^# p < 0.05, ^## p < 0.01 and ^### p < 0.001 versus AAV‐eGFP + CP group.

Figure 7

Schematic illustration of the mechanism of CP‐induced abnormality in emotional behaviors and memory deficits. Chronic periodontitis specifically increases the expression of DBP in the hippocampal microglia, subsequently induces the activation of resting microglia, resulting in the secretion of multiple inflammatory factors, and then mediates synapse loss and neurogenesis impairment in the hippocampus, and finally, facilitates the anxiety‐ and depression‐like behaviors and memory deficit in mice.