Somatostatin-evoked Aβ catabolism in the brain: Mechanistic involvement of α-endosulfine-KATP channel pathway

Abstract

Alzheimer's disease (AD) is characterized by the deposition of amyloid β peptide (Aβ) in the brain. The neuropeptide somatostatin (SST) regulates Aβ catabolism by enhancing neprilysin (NEP)-catalyzed proteolytic degradation. However, the mechanism by which SST regulates NEP activity remains unclear. Here, we identified α-endosulfine (ENSA), an endogenous ligand of the ATP-sensitive potassium (K_ATP) channel, as a negative regulator of NEP downstream of SST signaling. The expression of ENSA is significantly increased in AD mouse models and in patients with AD. In addition, NEP directly contributes to the degradation of ENSA, suggesting a substrate-dependent feedback loop regulating NEP activity. We also discovered the specific K_ATP channel subtype that modulates NEP activity, resulting in the Aβ levels altered in the brain. Pharmacological intervention targeting the particular K_ATP channel attenuated Aβ deposition, with impaired memory function rescued via the NEP activation in our AD mouse model. Our findings provide a mechanism explaining the molecular link between K_ATP channel and NEP activation, and give new insights into alternative strategies to prevent AD.

Fig. 1. Identification of ENSA as a NEP regulator in vitro.

A-C. NEP activity after treatment of co-cultured cells with 1 µM somatostatin or TT232 for 24 h. A Cortical/hippocampal (Ctx&Hip) neurons (n = 12 wells per treatment), (B) co-cultured neurons (n = 10 wells per treatment), and (C) basal ganglia neurons (n = 8 or 9 wells per treatment) were used. D–F NEP activity in co-cultured neurons after the replacement of the culture medium with conditioned media from (E) Ctx&Hip and (F) basal ganglia neurons treated with 1 µM somatostatin for 0–6 h. n = 6–10 wells per treatment in co-cultured neurons. G–K NEP activity of co-cultured neurons after replacement of the culture medium with separated conditioned media from Ctx&Hip neurons treated with SST or TT232. H, I 10 and (J and K) 30kDa centrifugal filters were used for the separation (n = 7–10 for each group). NEP activity in co-cultured neurons after incubation with (L) ENSA, (M) NSG-1 and (N) NUCKS-1 recombinant proteins for 24 h. n = 8–10 wells per treatment in co-cultured neurons. Data represent the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 (one-way ANOVA with Dunnett’s post-hoc test).

Fig. 2. Elevation of NEP activity in Ensa KO mice.

A Immunostaining of NEP (Red) and VGAT (Green) from hippocampi of 3-month-old WT and Ensa KO mice. Scale bar is 100 µm in low magnification image and 50 µm in high-magnification image. B Statistical analysis of NEP immunoreactivity (n = 5 for each group). LM: lacunosum-molecular layer, Omo: Outer molecular layer and MMo: middle molecular layer. C Statistical analysis of colocalized NEP and VGAT signals (n = 5 for each group). D NEP activity in membrane fractions from hippocampi of 3-month-old WT and Ensa KO mice (WT: n = 7, Ensa KO: n = 6). E. Aβ₄₀ ELISA of hippocampi of 3-month-old WT and Ensa KO mice (WT: n = 5, Ensa KO: n = 6). F. Aβ₄₂ ELISA of hippocampi of 3-month-old WT and Ensa KO mice (WT: n = 5, Ensa KO: n = 6). G Aβ₄₂ ELISA of hippocampi of 3-month-old Mme KO mice and Ensa/Mme dKO (Mme KO: n = 4, Ensa/Mme dKO: n = 5). H and I. Immunostaining of Aβ (Green), NEP (Red) and DAPI (blue) from 18-month-old App^NL-F and App^NL-F/Ensa KO mice. Statistical analysis of amyloid plaque area in 18-month-old App^NL-F and App^NL-F/Ensa KO mice (n = 6 for each group). Scale bar is 100 µm. J Aβ₄₂ ELISA of Tris-HCl-buffered saline-soluble (Ts) hippocampal fractions from 18-month-old App^NL-F and App^NL-F/Ensa KO mice (App^NL-F: n = 7, App^NL-F/Ensa KO: n = 6). K Aβ₄₂ ELISA of guanidine-HCl-soluble (GuHCl) hippocampal fractions from App^NL-F and App^NL-F/Ensa KO mice (App^NL-F: n = 7, App^NL-F/Ensa KO: n = 6). L Immunostaining of NEP (Red) and VGAT (Green) in hippocampi of 18-month-old App^NL-F and App^NL-F/Ensa KO mice. Scale bar is 100 µm in low-magnification image and 50 µm in high-magnification image. M Statistical analysis of NEP immunoreactivity (n = 4 for each group). N Statistical analysis of colocalized NEP and VGAT signals (n = 4 for each group). Data represent the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 (Student’s or Welch’s t test).

Fig. 3. Identification of ENSA as a substrate for NEP.

A Immunoblotting of ENSA incubated with or without NEP and mentioned inhibitors for 24 h at 37 ˚C. Thio: Thiorphan, Phos: Phosphoramidon. B Specific peak of full-length of ENSA after incubation with or without NEP and thiorphan. C Specific peak of cleaved ENSA after incubation with or without NEP and thiorphan. D Sequence of full-length of ENSA. Arrowheads indicate cleavage site by NEP. E, F Immunoblotting of ENSA from cortices and hippocampi of 6-month-old WT and Mme KO mice. Values indicated in the graph show ENSA band intensities normalized to that of β-actin (n = 5 for each group). G–I Immunoblotting of (H) NEP and (I) ENSA from hippocampi of 3-month-old WT mice after overexpression of active or inactive mutant NEP by SFV gene expression system. Values indicated in the graph show NEP and ENSA band intensities normalized to that of β-actin (n = 4 for each group). J Aβ₄₂ ELISA of Tris-HCl-buffered saline-soluble fractions containing 1% Triton-X from hippocampi of WT mice after overexpression of active or inactive mutant NEP by the SFV gene expression system (n = 4 for each group). K Immunostaining of ENSA (Green), NEP (Red) and DAPI (Blue) in CA3 from 3-month-old WT, Ensa KO and Mme KO mice. Scale bar is 50 µm in low-magnification image and 10 µm in high-magnification image. White arrows indicate colocalized signals. Data represent the mean ± SEM. *P < 0.05, ****P < 0.0001 (Student’s or Welch’s t test).

Fig. 4. SUR1/Kir6.2 regulates NEP expression and activity.

A, B Immunostaining of NEP in hippocampi of WT and ABCC8 KO mice (n = 6 for each group). Scale bar represents 500 µm. C NEP ELISA with hippocampi of WT, heterozygous and homozygous ABCC8 KO mice (n = 5 for each group). D NEP activity in hippocampi of 3-month-old WT and Abcc8 KO mice (n = 5 for each group). E Aβ₄₀ ELISA of hippocampi from 3-month-old WT and Abcc8 KO mice (n = 4 for each group). F Aβ₄₂ ELISA of hippocampi from 3-month-old WT and Abcc8 KO mice (n = 4 for each group). G Aβ₄₀ ELISA of hippocampi from 3-month-old Mme KO and Abcc8/Mme dKO mice (n = 4 for each group). H Aβ₄₂ ELISA of hippocampi from 3-month-old Mme KO and Abcc8/Mme dKO mice (n = 4 for each group). I Immunoblotting of APP, CTFs, IDE and ECE-1 in hippocampi of 3-month-old WT and Abcc8 KO mice. J Values indicated in graphs show band intensities for APP, CTFs, IDE, and ECE-1 normalized to that of β-actin (n = 5 for each group). K, L. Immunostaining of NEP in hippocampi of WT and heterozygous Kcnj8 KO and homozygous Kcnj11 KO mice (n = 4 for each group). Scale bar represents 500 µm. M NEP ELISA with hippocampi from WT and heterozygous Kcnj8 KO and homozygous Kcnj11 KO mice (n = 5 for each group). N NEP activity from hippocampi of 3-month-old WT and heterozygous Kcnj8 KO and homozygous Kcnj11 KO mice (WT: n = 8, Kcnj8 KO: n = 6, Kcnj11 KO: n = 5). In (B), (D), (E), (F), the data represent the mean ± SEM. *P < 0.05, **P < 0.01, ****P < 0.0001 (Student’s t test). In (C), (L), (M), (N), the data represent the mean ± SEM.**P < 0.01, ***P < 0.001, ****P < 0.0001 (one-way ANOVA with Turkey’s multiple comparison test).

Fig. 5. Improvement of Aβ pathology and memory function in App^NL-F mice via enhancement of NEP activity by Dz treatment.

A NEP activity after treatment of co-cultured neurons for 24 h with different doses of diazoxide (Dz) (n = 9–10 for each group). B NEP activity in membrane fractions from anterior cortex (Ctx), posterior Ctx and hippocampus (Hip) of 4-month-old WT mice treated with or without Dz (n = 6 for each group). C, D. Immunoblotting of NEP in anterior Ctx of 4-month-old WT mice treated with or without Dz. Values indicated in the graph Values indicated in the graph show NEP band intensities normalized to that of β-actin (n = 4 for each group). E Aβ₄₂ ELISA of GuHCl fractions from anterior Ctx and Hip of 4-month-old WT mice with or without Dz (Dz (–): n = 6, Dz (+): n = 7). F Aβ₄₂ ELISA of GuHCl fractions from anterior Ctx and Hip of 6-month-old Mme KO mice with or without Dz (n = 8 for each group). G Freezing ratio of 18-month-old WT and App^NL-F mice treated with or without Dz (WT Dz (–): n = 12, WT Dz (+): n = 13, App^NL-F Dz (–): n = 10, App^NL-F Dz (+): n = 11). H, I Immunostaining of Aβ (Green) and NEP (Red) in Ctx, Subiculum and Molecular layer from 18-month-old App^NL-F with or without Dz (n = 7 for each group). Scale bar in cortical image = 500 µm and hippocampal image = 200 µm. J Aβ₄₂ ELISA of GuHCl fractions from cortices and hippocampi of 18-month old App^NL-F with or without Dz (n = 8 for each group). K Immunostaining of NEP (Red) and VGAT (Green) in hippocampi from 18-month old App^NL-F with or without Dz. Scale bar is 100 µm in low-magnification image and 50 µm in high-magnification image. L. Statistical analysis of immunoreactivity of NEP (n = 5 for each group). LM: lacunosum-molecular layer, Omo: Outer molecular layer and MMo: middle molecular layer. M Statistical analysis of colocalized signals of NEP and VGAT (n = 5 for each group). In (A), the data represent the mean ± SEM. *P < 0.05, ***P < 0.001 (one-way ANOVA with Dunnett’s post-hoc test). In (B), (D), (E), (I), (J), (L), (M), the data represent the mean ± SEM. *P < 0.05, **P < 0.01, (Student’s t test). In (G) the data represent the mean ± SEM. On day 3, WT Dz (+) vs App^NL-F Dz (–) *P < 0.05. On day 4, WT Dz (–) vs App^NL-F Dz (–) *P < 0.05, WT Dz (+) vs App^NL-F Dz (–) **P < 0.01, App^NL-F Dz (–) vs App^NL-F Dz (+) *P < 0.05 (two-way ANOVA with Turkey’s multiple comparison test).