Zfra Inhibits the TRAPPC6AΔ-Initiated Pathway of Neurodegeneration

Abstract

When WWOX is downregulated in middle age, aggregation of a protein cascade, including TRAPPC6AΔ (TPC6AΔ), TIAF1, and SH3GLB2, may start to occur, and the event lasts more than 30 years, which results in amyloid precursor protein (APP) degradation, amyloid beta (Aβ) generation, and neurodegeneration, as shown in Alzheimer's disease (AD). Here, by treating neuroblastoma SK-N-SH cells with neurotoxin MPP+, upregulation and aggregation of TPC6AΔ, along with aggregation of TIAF1, SH3GLB2, Aβ, and tau, occurred. MPP+ is an inducer of Parkinson's disease (PD), suggesting that TPC6AΔ is a common initiator for AD and PD pathogenesis. Zfra, a 31-amino-acid zinc finger-like WWOX-binding protein, is known to restore memory deficits in 9-month-old triple-transgenic (3xTg) mice by blocking the aggregation of TPC6AΔ, SH3GLB2, tau, and amyloid β, as well as inflammatory NF-κB activation. The Zfra4-10 peptide exerted a strong potency in preventing memory loss during the aging of 3-month-old 3xTg mice up to 9 months, as determined by a novel object recognition task (ORT) and Morris water maize analysis. Compared to age-matched wild type mice, 11-month-old Wwox heterozygous mice exhibited memory loss, and this correlates with pT12-WWOX aggregation in the cortex. Together, aggregation of pT12-WWOX may link to TPC6AΔ aggregation for AD progression, with TPC6AΔ aggregation being a common initiator for AD and PD progression.

Keywords: Alzheimer’s disease; SH3GLB2; TIAF1; TRAPPC6A; TRAPPC6AΔ; WWOX; p53; tumor suppressor.

Figure 1

Protein aggregation in Wwox knockout MEF cells. (A) The expression of endogenous TPC6A, TIAF1, and JNK1 was determined in the wild-type and knockout Wwox MEF cells (with gel run under reducing SDS-PAGE and then Western blotting). These cells were treated with TGF-β1 (5 ng/mL) for 1 h. (B) The presence of NFT was shown in the heterozygous and knockout Wwox MEF cells. TNF-α (100 ng/mL) did not increase NFT expression in both cells during treatment for 1 h. (C) TGF-β1 (5 ng/mL) increased polymerization of wild-type TPC6A in the knockout Wwox MEF cells during treatment for 1 h at 37 °C.

Figure 2

Polymerization of TPC6A∆, TIAF1, and SH3GLB2 under MPP+ stimulation. SK-N-SH cells were seeded to 70% confluence and cultured overnight with 10% fetal bovine serum. Prior to stimulation, cells were synchronized under serum-free conditions for 20 h. Cells were stimulated with MPP+ at the indicated doses for 24 h. By non-reducing SDS-PAGE (6%) and Western blotting, protein polymerization was observed. (A) SH3GLB2 dimerized in a dose-dependent manner. (B) Polymerized TIAF1 was shown upon increasing MPP+ from 125 to 250 µM. (C) TPC6A∆ showed polymerization at 62.5 µM MPP+ treatment, followed by gradual increment of polymerization at 125 and 250 µM MPP+ treatment. (D) TPC6A∆ phosphorylated at Ser35 at 125 and 250 µM MPP+ treatment. Two individual repeats were performed for each experiment. Representative figures are shown. (E) As an internal standard, the level of α-tubulin is shown.

Figure 3

Zfra4-10 peptide restored the learning and memory of 10-month-old 3xTg AD mice. (A,B) 3xTg mice, aged 10 months old, received Zfra4-10 tail vein injections, followed by examination of their learning and memory capabilities by a novel object recognition task (A) and a Morris water maze (B) at 1 month after the injections. The percentages of novel object exploring time (time spent on novel object/time spent on both objects) of acquisition, short-term (2 h) delay, and long-term (24 h) delay are shown (control group: sham, n = 10; Zfra injection group: Zfra4-10, n = 10). In the Morris water maze, the learning curve represents the latency of finding the hidden platform during the training session. The probe test was quantified as time spent in the target quadrant (Bonferroni’s post hoc test and ANOVA: * 0.01< p < 0.05, ** 0.001< p < 0.01, *** p < 0.001, experimental group vs. respective control group. Sham: n = 10; control: n = 10).

Figure 4

Zfra4-10 blocked age-dependent memory loss in 3xTg mice from 3 to 9 months of age. (A,B) Three-month-old 3xTg mice received Zfra4-10 peptide via tail vein injections once per week for 3 consecutive weeks, followed by one-week rest, and then they were subjected to novel ORT (A) and Morris water maze analysis (B). The animal behavior experiments were repeated when mice reached 6 and 9 months old, respectively. The percentages of novel object exploring time of acquisition, short-term (2 h) delay, and long-term (24 h) delay are shown (n = 5 for each group). The learning curve represents the escape latency to find the hidden platform during a training session. The probe test was quantified as time spent in the target quadrant (Bonferroni’s post hoc test and ANOVA: * p < 0.05, ** p < 0.01, *** p < 0.001 vs. the respective control group; n = 5).

Figure 5

WWOX expression was critically associated with memory maintenance. (A,B) Wild-type and heterozygous Wwox mice at age 3 and 10 months were subjected to learning and memory tests by the novel object recognition task (A) and Morris water maze (B). The percentages of novel object exploring time of acquisition, short-term (2 h) delay, and long-term (24 h) delay are shown (n = 5). The learning curve represents the latency to find the hidden platform during a training session. The probe test was quantified as time spent in the target quadrant (Bonferroni’s post hoc test and ANOVA: * p < 0.05, ** p < 0.01, *** p < 0.001 vs. respective control group; n = 20). (C) Micrographs of 11-month-old wild-type and heterozygous hippocampal areas. TPC6AΔ plaques were significantly upregulated in the heterozygous mice (** p < 0.01 vs. respective control group; n = 20). The number of plaques per microscopic field at 200× is shown.

Figure 6

Presence of pS35-TPC6AΔ and SH3GLB2 aggregates in the cortex of 11-month-old heterozygous Wwox mice. (A–F) By immunohistochemistry, protein expression of pS35-TPC6AΔ (A), TPC6A (wild type) (B), pS37-TIAF1 (C), TIAF1 (D), SH3GLB2 (E), and pT181-Tau (F) was examined in the brain cortex of 11-month-old heterozygous Wwox mice and age-matched wild-type mice. The extent of protein aggregation was examined (n = 5 to 10; Student’s t-test). Negative controls without staining with primary antibodies are shown. The digitally enlarged pictures are shown in Figure S3.

Figure 7

Identification of pT12-WWOX as aggregates in the brain cortex of 11-month-old heterozygous Wwox mice. (A–E) Compared to the wild-type mice, increased immunointensity of staining was shown in the cortex of heterozygous Wwox mice using an antibody against pY287-WWOX (A) and pY33-WWOX (B). The presence of pT12-WWOX aggregates (C), but not pS14-WWOX (D), was shown in the cortex of heterozygous Wwox mice. No pT12-WWOX aggregates were found in the wild-type mice. The bar graphs show mean ± standard deviation (n = 5; ** p < 0.001, experimental group vs. respective control group, Student’s t-tests). The number of plaques per microscopic field (100x magnification) is shown for pT12-WWOX (n = 3) (C). No plaques are shown with pS14-, pY33-, and pY287-WWOX. (E) Negative controls without staining with primary antibodies are shown. The digitally enlarged pictures are found in Figure S4.