Kisspeptin prevention of amyloid-β peptide neurotoxicity in vitro

Abstract

Alzheimer's disease (AD) onset is associated with changes in hypothalamic-pituitary-gonadal (HPG) function. The 54 amino acid kisspeptin (KP) peptide regulates the HPG axis and alters antioxidant enzyme expression. The Alzheimer's amyloid-β (Aβ) is neurotoxic, and this action can be prevented by the antioxidant enzyme catalase. Here, we examined the effects of KP peptides on the neurotoxicity of Aβ, prion protein (PrP), and amylin (IAPP) peptides. The Aβ, PrP, and IAPP peptides stimulated the release of KP and KP 45-54. The KP peptides inhibited the neurotoxicity of Aβ, PrP, and IAPP peptides, via an action that could not be blocked by kisspeptin-receptor (GPR-54) or neuropeptide FF (NPFF) receptor antagonists. Knockdown of KiSS-1 gene, which encodes the KP peptides, in human neuronal SH-SY5Y cells with siRNA enhanced the toxicity of amyloid peptides, while KiSS-1 overexpression was neuroprotective. A comparison of the catalase and KP sequences identified a similarity between KP residues 42-51 and the region of catalase that binds Aβ. The KP peptides containing residues 45-50 bound Aβ, PrP, and IAPP, inhibited Congo red binding, and were neuroprotective. These results suggest that KP peptides are neuroprotective against Aβ, IAPP, and PrP peptides via a receptor independent action involving direct binding to the amyloid peptides.

Figure 1

Effects of amyloid peptides on KP 1–54 and KP 45–54 release from SH-SY5Y neurons. Neuronal SHSY-5Y cell cultures were exposed to Aβ 1–42, Aβ 25–35, PrP 106–126, PrP 118–135, IAPP 1–37, IAPP 20–29, A-Bri 1–34, and A-Dan 1–34 peptides (50 nM each) for 2 h. The release of ir-KP 1–54 (A) and ir-KP 45–54 (B) into the cell culture media was determined by EIA. All results are expressed as the mean ± SEM (n = 8). (* = P < 0.05 vs control (media alone); one-way ANOVA.)

Figure 2

Effects of KP peptides on Aβ, PrP, A-Bri, A-Dan, and IAPP neurotoxicity in SH-SY5Y neurons. The effects of 10 μM KP 1–54 on the toxicity of Aβ 1–42, Aβ 1–40, Aβ 25–35, Aβ 29–40, Aβ 31–35, PrP 106–126, PrP 118–135, IAPP 1–37, IAPP 8–37, IAPP 20–29, A-Bri 1–34, and A-Dan 1–34 peptides (5 μM each) were tested in human SH-SY5Y neuroblastoma cell cultures (A), with cell viability determined by the MTT assay. The effects of KP 1–54, KP 27–54, KP 42–54, KP 45–54, KP 45–50, KP 45–47, KP 47–50, and NPFF peptides (10 μM each) on the toxicity of 5 μM Aβ 1–42 (B), 5 μM PrP 106–126 (C), and 5 μM IAPP 1–37 (D) were tested in human SH-SY5Y neuroblastoma cell cultures, with cell viability determined by the MTT assay. All results are expressed as a % control (SH-SY-5Y cells in media alone) and are expressed as the mean ± SEM (n = 8). (* = P < 0.05 vs control (media alone); † = P < 0.05 vs amyloid fibrils alone; one-way ANOVA.)

Figure 3

Effects of KP peptides on Aβ and PrP neurotoxicity in rat cortical neurons. The effects of 10 μM KP 42–54 (closed blue circles) or 10 μM KP 45–50 (closed red squares) on the toxicity of Aβ 1–42 (A) and PrP 106–126 (B) were tested in rat cortical neuron cell cultures, with cell viability determined by the MTT assay. The effects of 10 μM KP receptor antagonist P234 or 10 μM NPFF receptor antagonist RF9 on 2.5 μM KP 42–54 or 2.5 μM KP 45–50 protection against the toxicity of 5 μM Aβ 1–42 (C) and 5 μM PrP 106–126 (D) were tested in rat cortical neuron cell cultures, with cell viability determined by the MTT assay. All results are expressed as % control (rat cortical neurons in media alone) and are expressed as the mean ± SEM (n = 8). (* = P < 0.05 vs Aβ or PrP alone; † = P < 0.05 vs Aβ or PrP plus KP 42–54 or KP 45–50; one way ANOVA.)

Figure 4

Effects of KP receptor knockdown on KP protection against Aβ, PrP, and IAPP neurotoxicity in SH-SY5Y neurons. Neuronal SHSY-5Y cell cultures were treated with control siRNA or GPR-54 siRNA and allowed to recover for 48 h prior to exposure to Aβ 1–42, PrP 106–126, or IAPP 1–37 (5 μM each) with or without 10 μM KP 45–54. Cell viability was determined by the MTT assay. All results are expressed as % control (siRNA treated SH-SY-5Y cells in media alone) and are expressed as the mean ± SEM (n = 8).

Figure 5

Effects of KP peptides on IAPP toxicity in pancreatic islet cultures. The effects of 10 μM KP 45–54 (closed red circles) on the toxicity of 5 μM IAPP 1–37 was tested in human pancreatic islet cell cultures (A), with cell viability determined by the MTT assay. The effects of 10 μM KP receptor antagonist P234 or 10 μM NPFF receptor antagonist RF9 on 2.5 μM KP 45–54 protection against the toxicity of 5 μM IAPP 1–37 (B) was tested in human pancreatic islet cell cultures, with cell viability determined by the MTT assay. All results are expressed as % control (pancreatic islet cells in media alone) and are expressed as the mean ± SEM (n = 8). (* = P < 0.05 vs IAPP alone; † = P < 0.05 vs IAPP and KP 45–54; one-way ANOVA.)

Figure 6

Effects of modified KiSS-1 expression and anti-KP on Aβ, PrP, and IAPP neurotoxicity in SH-SY5Y neurons. Neuronal SHSY-5Y cell cultures (A) were treated with control siRNA or KiSS-1 siRNA and allowed to recover for 48 h prior to exposure to Aβ 1–42, Aβ 25–35, PrP 106–126, or IAPP 1–37 (5 μM each). Cell viability was determined by the MTT assay. The effects of 10 μg/mL anti-KP 45–54 antibody, 10 μM KP receptor antagonist P234, or 10 μM NPFF receptor antagonist RF9 on the toxicity of 5 μM Aβ 1–42 (B) were tested in human SH-SY5Y neuroblastoma cell cultures, with cell viability determined by the MTT assay. Neuronal SHSY-5Y cells transfected and stably overexpressing (C) either control expression vector (PCont) or expression vector containing the KiSS-1 gene (PKiSS) were treated with Aβ 1–42, Aβ 25–35, PrP 106–126, or IAPP 1–37 (5 μM each) and cell viability determined by the MTT assay after 24 h. The effects of 10 μg/mL anti-KP 45–54 antibody, 10 μM KP receptor antagonist P234, or 10 μM NPFF receptor antagonist RF9 on the toxicity of 5 μM Aβ 1–42 (D) were tested in KiSS-1 overexpressing SH-SY5Y neuroblastoma cell cultures, with cell viability determined by the MTT assay. All results are expressed as % control (SHSY-5Y neurons, siRNA treated SHSY-5Y neurons, or transfected SHSY-5Y neurons in media alone) and are expressed as the mean ± SEM (n = 8). (* = P < 0.05 vs control (media alone); † = P < 0.05 vs control siRNA (A), Aβ alone (B and D), and PCont vector (C); one-way ANOVA.)

Figure 7

Binding of KP to Aβ, PrP and IAPP peptides. Alignment of the human metastasis-suppressor KiSS-1 preproprotein sequence (NP_002247.3) with the human catalase sequence (NP_001743.1) is shown in A. The red box highlights the region of KP that prevents Aβ, PrP, and IAPP toxicity; blue and green boxes highlights the Gly-Ala-Ile-Ile region that binds catalase in schemes A and B respectively; the black box highlights Aβ 31–35, which inhibits Aβ 1–42 binding to catalase. Immunoplates were coated with Aβ 1–40, Aβ 1–28, Aβ 29–40, Aβ 25–35, PrP 106–126, PrP 118–135, IAPP 1–37, IAPP 20–29, A-Bri 1–34, or A-Dan 1–34 fibrils. Coated plates were incubated with either biotinylated KP 45–54 (B) alone (open columns) or in the presence of unlabeled KP 45–54 (closed blue columns) and bound material determined by EIA. Plates coated with KP 1–54, KP 27–54, KP 42–54, KP 45–54, KP 45–50, KP 45–47, KP 47–50, or NPFF (C) were incubated with biotinylated Aβ 1–42 (open columns), biotinylated PrP 106–126 (red columns), or biotinylated IAPP 1–37 (black columns) and bound material determined by EIA. All results are expressed as the mean ± SEM (n = 8). (* = P < 0.05 vs control (buffer alone); † = P < 0.05 vs biotinylated KP 45–54; one-way ANOVA.)

Figure 8

Effects of KP on Congo red binding to Aβ, PrP, and IAPP. The Aβ 1–40, Aβ 1–28, Aβ 25–35, Aβ 29–40, PrP 106–126, PrP 118–135, IAPP 1–37, IAPP 20–29, A-Bri 1–34, and A-Dan 1–34 peptides were dissolved in PBS and incubated at 37 °C for 24 h, with constant oscillation, with or without KP 45–50. Congo red binding assays (A) were performed with amyloid peptides alone (open columns) or in the presence of KP 45–50 (closed blue columns). The effect of Congo red (0–100 mM) on biotinylated Aβ 1–42 binding to KP 45–54 (B) was determined by EIA. All results are expressed as the mean ± SEM (n = 8). (* = P < 0.05 vs amyloid fibrils alone; one-way ANOVA.)