Increased Kallistatin was presented in AD patients and could impair cognitive memory in mice.

(A-B) Serum Kallistatin(A), fasting blood glucose (FBG), triglyceride (TG), and total cholesterol (TC). (B) of AD patients and their corresponding normal control subjects. (C-D) Serum Kallistatin(C), TG, and TC(D) of AD patients with DM and their corresponding normal control subjects. The behavioral performance of KAL-TG mice was assessed through the Morris water maze test. (E-H), Y-maze test(I), and electrophysiology (J). (E)The escape latency time of different months of KAL-TG mice (3M, 6M, 9M, 12M) and corresponding WT mice were presented during 1-6 day. (F-H) Cognitive functions were evaluated by spatial probe test at day 7, the representative each group mice traces were shown (F), then analyzing each group mice crossing platform times (G) and time percent in the targeted area (H), n=4 to 9 per group. (I) Spontaneous alternation of Y-maze test. (J) LTP was measured by whole-cell voltage-clamp recordings of CA1 neurons in acute hippocampal slices of KAL-TG (3M, 6M, 12M) and WT mice, n = 6-12 cells from 3 mice per group. Error bars represent the standard deviation (SD); one asterisk, p < 0.05, two asterisks, p < 0.01; three asterisks, p < 0.001; four asterisks, p < 0.0001.

Kallistatin promoted Aβ generation.

(A-B) Immunohistochemistry staining of Aβ(A) was carried out in KAL-TG and WT mice hippocampal tissue. Scale bar, 100μm. The statistical analysis of Aβ plaques (B) in hippocampal tissue of KAL-TG and WT mice, n=4-5 per group. (C) Protein levels of Aβ were tested by western blot analysis in hippocampal tissue, n=3 per group, then statistically analyzed the above results. (D) Hippocampal tissue Aβ42 contents were performed by ELISA in KAL-TG and WT groups, n=3 per group. (E) Primary mouse neurons were isolated, then infected with adenovirus to overexpress Kallistatin for 72h. Aβ42 concentration of primary hippocampal neurons supernate and cell lysate were quantified by ELISA, n=3 per group. (F-G) Western blot analysis of Aβ protein level in primary hippocampal neurons infected with overexpressing Kallistatin adenovirus and control groups, then statistical analysis of Aβ protein levels, n=3 per group. Error bars represent the standard deviation (SD); one asterisk, p < 0.05; two asterisks, p < 0.01; three asterisks, p < 0.001.

Kallistatin promoted tau phosphorylation.

(A-B) Immunohistochemistry staining of phosphorylated tau (p-tau S396, p-tau T231, p-tau S202) and tau(A) was carried out in KAL-TG and WT mice hippocampal tissue. Scale bar, 100μm. The statistical analysis of phosphorylated tau(B) in hippocampal tissue of KAL-TG and WT mice, n=3 per group. (C-D) Protein levels of phosphorylated tau (p-tau S396, p-tau T231, p-tau S202) and tau were tested by western blot analysis in hippocampal tissue, then statistically analyzed the above results. Error bars represent the standard deviation (SD); one asterisk, p < 0.05; two asterisks, p < 0.01; three asterisks, p < 0.001.

Kallistatin transgenic mice exhibited increased BACE1 expression and activity in the hippocampus.

(A) Western blot analysis of relevant proteins, such as APP, PS1, and BACE1 during Aβ generation in hippocampal tissue of each time point (6M, 9M, 12M) KAL-TG mice and corresponding WT control groups, n=3 per group, then statistical analysis of APP, PS1 and BACE1 protein levels. (B) Immunohistochemistry staining of BACE1 was carried out in KAL-TG and WT mice hippocampal tissue at each time point (6M, 9M, 12M). n=3 to 5 per group. Scale bar, 100 μm. (C) Statistical analysis of BACE1 immunohistochemistry staining, n=3 to 4 per group. (D) ELISA measured the β-secretase activity of each group’s hippocampal tissue, n=3 per group. Error bars represent the standard deviation (SD); one asterisk, p < 0.05, two asterisks, p < 0.01.

In vitro, Kallistatin promoted BACE1 expression to augment Aβ by suppressing PPARγ activation.

(A) The relevant protein levels in primary mouse neurons infected with overexpressing Kallistatin adenovirus during Aβ generation were determined by western blot analysis. (B) Statistical analysis of BACE1 expression in primary neurons. (C-D) β-secretase (C) and γ-secretase (D) activity of primary hippocampal neurons infected with overexpressing Kallistatin adenovirus and control adenovirus was measured by ELISA. (E) Primary hippocampal neurons were treated with BACE1 inhibitor verubecestat (50nM), then infected with adenovirus to overexpress Kallistatin. Western blot analysis of Aβ, BACE1, and Kallistatin protein levels, β-actin served as a loading control. (F) HT22 cells were infected with BACE1 siRNA, then infected with adenovirus to overexpress Kallistatin. Western blot analysis of Aβ and BACE1 protein levels, β-actin served as a loading control. (G) The relevant proteins involved in BACE1 transcriptional expressions, such as PPARγ, YY1, and SP1 were measured by western blot analysis in hippocampal tissue. β-actin served as a loading control. (H) Statistical analysis of PPARγ in hippocampal tissue of each group. (I) The representative diagrams of PPARγ expression in hippocampal tissue were presented in the above graphs. Scale bar, 100μm. (J) Statistical analysis of PPARγ immunohistochemistry staining in hippocampal tissue of each group, n=3 to 4 per group. (K)Primary hippocampal neurons were treated with PPARγ agonist rosiglitazone (10nM) for 12h, then infected with adenovirus to overexpress Kallistatin for 72h. Western blot analysis of Aβ and BACE1 protein levels. β-actin served as a loading control. (L) Statistical analysis of PPARγ protein levels in each group. Error bars represent the standard deviation (SD), one asterisk, p < 0.05, two asterisks, p < 0.01; three asterisks, p < 0.001; ns means no significant difference.

Kallistatin directly bonded to the Notch1 receptor, which activated the Notch1 pathway to promote Aβ production.

(A) Notch1 expression was measured by western blot analysis in hippocampal tissue. β-actin served as a loading control. (B) Statistical analysis of Notch1 in hippocampal tissue of each group. (C) The representative diagrams of Nocth1 expression in hippocampal tissue were presented in the above graphs. Scale bar, 100μm. (D) Statistical analysis of Notch1 immunohistochemistry staining in hippocampal tissue of each group. (E-F) Primary hippocampal neurons were infected with overexpressing Kallistatin adenovirus for 72h, then Co-IP analysis (E) and membrane extraction experiment (F) was performed to verify whether Kallistatin can bind to the Notch1 receptor. β-actin served as a loading control. (G-H) HT22 cells were treated with siRNA (Notch1) and shRNA (HES1) to knock down Notch1 and HES1 for 12h, then infected with adenovirus to overexpress Kallistatin for 24h. Western blot analysis was used to detect the Notch1 signaling pathway. Error bars represent the standard deviation (SD), one asterisk, p < 0.05.

Kallistatin promoted phosphorylation of tau by suppressing Wnt signaling pathway.

(A-B) GSK-3β and p-GSK-3β expression was measured by western blot analysis in hippocampal tissue, then statistically analyzed the above results. (C-D) Primary hippocampal neurons were treated with overexpressing Kallistatin adenovirus for 72h, then western blot analysis was used to detect the content of GSK-3β, p-GSK-3β, tau, p-tau (Ser9, T231, S396), and statistically analyzed the above results. (E-F) Primary hippocampal neurons were treated with overexpressing Kallistatin adenovirus for 48h, then treated with LiCl (10mM) for 24h, western blot analysis was used to detect the content of GSK-3β, p-GSK-3β, p-tau (Ser9, T231, S396), and statistical analysis of the above results.

Fenofibrate could alleviate memory and cognitive impairment of KAL-TG mice.

(A) Illustration of experimental protocols. Fenofibrate (0.3 g/kg/d × 5week, i.g.) or rosiglitazone (5mg/kg/d × 5week, i.g.) were given to KAL-TG mice. The serum for Kallistatin measuring was collected at week 0 and week 4. And Morris water maze and Y-maze test were performed at week 5. (B-E) Behavioral performance was assessed through the Morris water maze test (B-E) and the Y-maze test (F). (B) The escape latency time was presented during 1-5 day. (C-E) Cognitive functions were evaluated by spatial probe test at day 6, then analyzing each group of mice crossing platform times(C), time percent in the targeted area (D), and the path traces heatmap (E), n=5 to 6 per group. (F) Spontaneous alternation of Y-maze test. (G) Kallistatin decreased ratio was calculated by dividing the serum Kallistatin concentration of KAL-TG mice before Fenofibrate/ rosiglitazone treatment by the serum Kallistatin concentration of KAL-TG mice after a-month treatment, and serum Kallistatin concentration was measured by ELISA. (H) Protein levels of Aβ, BACE1, p-tau, tau, p-GSK-3β and GSK-3β were tested by western blot analysis in hippocampal tissue. Error bars represent the standard deviation (SD); one asterisk, p < 0.05.

(A-B) GAD disease enrichment analysis (A) and PFAM analysis (B) result. Differentially expressed genes in neurons were obtained from GSE161355, and GAD disease enrichment was analyzed on David database. (C-D) Western blot analysis of Kallistatin expression in aging model SAMP8 and corresponding control SAMR1 mice hippocampal tissue samples, then statistically analyzing the above results. β-Actin served as a loading control. Error bars represent the standard deviation (SD); three asterisks, p < 0.001.

(A) HT22 cells were infected with adenovirus to overexpress Kallistatin for 48h. Aβ42 concentration of supernate and cell lysate was quantified by ELISA. (B-C) Western blot analysis of Aβ protein level in HT22 cells infected with overexpressing Kallistatin adenovirus and control groups for 48h, then statistical analysis of Kallstatin protein levels. (D) BACE1 mRNA expression in hippocampal tissue. (E) α-secretases (ADAM9, ADAM10, ADAM17) mRNA expression in hippocampal tissue. (F) γ-secretase activity of each group’s hippocampal tissue was measured by ELISA. (G) Primary hippocampal neurons were identified with MAP2. Error bars represent the standard deviation (SD), one asterisk, p < 0.05, two asterisks, p < 0.01.

(A) The relevant protein levels in HT22 cells infected with overexpressing Kallistatin adenovirus during Aβ generation were determined by western blot analysis. (B) Statistical analysis of BACE1 expression in HT22 cells. (C) β-secretase activity of HT22 cells infected with overexpressing Kallistatin adenovirus and control adenovirus was measured by ELISA. (D) Primary hippocampal neurons were infected with BACE1 siRNA for 72h. Western blot analysis of BACE1 protein levels. (E) HT22 cells were treated with PPARγ agonist rosiglitazone (10nM) for 12h, then infected with adenovirus to overexpress Kallistatin for 48h. Western blot analysis of Aβ and BACE1 protein levels. β-actin served as a loading control. Error bars represent the standard deviation (SD), two asterisks, p < 0.01.

(A-B) Western blot analysis of Notch1 protein level in primary hippocampal neurons and HT22 cells infected with overexpressing Kallistatin adenovirus and control groups. (C) HT22 cells were infected with overexpressing Kallistatin adenovirus for 48h, and Co-IP analysis was conducted to verify whether Kallistatin can bind to the Notch1 receptor. β-actin served as a loading control. (D) Primary hippocampal neurons were treated with Kallistatin protein for 72h, then IP analysis. (E) Primary hippocampal neurons were infected with Notch1 siRNA for 72h. Western blot analysis of BACE1 protein levels. (F) HT22 cells were infected with HES1 shRNA for 48h. Western blot analysis of HES1 protein levels.

(A) Western blot analysis of GSK-3β, p-GSK-3β, tau, p-tau(Ser9, T231, S396) in HT22 cells infected with overexpressing Kallistatin adenovirus and control groups. (B) Western blot analysis of GSK-3β, p-GSK-3β, p-tau(Ser9, T231, S396) in HT22 cells infected with overexpressing Kallistatin adenovirus and control groups for 24h, then treated with LiCl(10 mM) for 24h.

Clinical characteristic of AD patients

Clinical characteristic of AD patients with DM