Upregulation of SNAP25 by HDAC inhibition ameliorates Niemann‐Pick Type C disease phenotypes via autophagy induction

Niemann-Pick Type C disease (NPC) is a rare fatal neu-rodegenerative disorder caused by mutations in the NPC1 or NPC2 gene, 1 leading to abnormal accumulation of non-esterified cholesterol in lysosomes. Defective autophagy caused by the failure of com-posed of SNARE 2 has been reported in NPC


Upregulation of SNAP25 by HDAC inhibition ameliorates Niemann-Pick Type C disease phenotypes via autophagy induction
Dear Editor Niemann-Pick Type C disease (NPC) is a rare fatal neurodegenerative disorder caused by mutations in the NPC1 or NPC2 gene, 1 leading to abnormal accumulation of nonesterified cholesterol in lysosomes. Defective autophagy caused by the failure of autolysosome formation composed of SNARE machinery 2 has been reported in NPC disease and synaptosomal-associated protein 25 (SNAP25) has been highlighted as one of the important components of SNARE machinery. 3,4 Currently, there are no Food and Drug Administration (FDA)-approved treatments and a recent study shows that histone deacetylase (HDAC) inhibitors may be promising therapeutics for NPC disease. 5 To investigate the effects of HDAC inhibition in NPC-iNSCs, suberoylanilide hydroxamic acid (SAHA), N-hydroxy-7-(2-naphthylthio) heptanomide (HNHA), FK228, and valproic acid (VPA) were treated at a nontoxic concentration (Figure S1A-E) and non-esterified cholesterol was stained with filipin III ( Figure 1A,B). HDAC inhibitor-treated NPC-iNSCs had lower levels of non-esterified cholesterol and among them, HNHA showed the most effective cholesterol reduction. Likewise, the free cholesterol level was notably reduced by the HDAC inhibitor-treatment including SAHA which has previously exhibited an ameliorated activity in NPC disease 6 together with a new synthetic HDAC inhibitor, HNHA 7 ( Figure 1C).
We then compared the effects of the long-term treatment of compounds in NPC1 KO mice. SAHA and HNHA (HDACi)-treated mice showed improvement in body weight relative to vehicle-treated NPC1 KO mice ( Figure 1D) Figure 1E). A pathological phenotype of NPC is the progressive death of cerebellar Purkinje cells, 8 so we used Nissl-and calbindin-positive staining to measure neurodegeneration of Purkinje cells. The total number of Nisslor calbindin-positive cells in the HDACi-treated NPC1 KO mice increased 2.3-fold (SAHA) and 1.8-fold (HNHA) or 3fold (SAHA) and 2.3-fold (HNHA) in the cerebella, respectively ( Figure 1F-I).
Next, the mode of action of HDACi in NPC-iNSCs was analysed through total gene expression analysis using RNA-seq (Figure 2A-C). Heat maps showed differential expression of genes in wild type (WT), NPC and HDACitreated NPC-iNSCs ( Figure 2A). The 13 genes expressing the same tendency were selected in both HDACi-treated groups compared to the dimethyl sulfoxide (DMSO)treated group ( Figure 2B). Furthermore, we analysed gene profiles by using the evolutionary genealogy of genes: Non-supervised Orthologous Groups (eggNOG) browser ( Table 1). The top eight categories showing significant differences were selected and an in-depth analysis of the factors for category U, intracellular trafficking, secretion, and vesicular transport, was conducted ( Figure 2C). In the list of genes showing the greatest difference, SNAP25 increased the most ( Table 2).
We confirmed that the protein levels of SNAP25 were increased by HDACi treatment ( Figure 2D,E). In addition, overexpression of SNAP25 in NPC-iNSCs decreased free cholesterol, demonstrating that upregulation of SNAP25 in NPC-iNSCs can result in a reduction of lipids ( Figure 2F,G). Each compound was then treated on both control and SNAP25 knocked down cells to compare the effect of compounds on SNAP25. Notably, the effect of compounds was significantly decreased when SNAP25 was silenced compared to when it was not ( Figure 2H,I), implying that SNAP25 plays as a key target gene of HDACi treatment.   Referring to previous studies, 3,4 we hypothesized that the potential mechanism by which SNAP25 regulates cholesterol levels in NPC-iNSCs might be through autophagy. We found that SNAP25 was deficient in NPC-iNSCs ( Figure S3A,B) and when SNAP25 was overexpressed in NPC-iNSCs, p62, LC3-I/II, and LAMP1a were markedly reduced ( Figure 3A,B). Next, we examined whether SNAP25 directly interacts with the components of the SNARE complex by co-immunoprecipitation analysis for endogenous SNAP25, and it was revealed that HDACitreatment increased the interaction between SNAP25, STX17, and VAMP8 in the SNARE complex ( Figure 3C). A strong interaction between Vamp8-SNAP25-STX17 in HDACi-treated NPC-iNSCs was also verified by proximity ligation assay (PLA, Figure 3D). Although SNAP25 was increased by HDACi-treatment compared to DMSO, Rapamycin (Rapa) and Bafilomycin (Baf) treatment did not decrease autophagy markers ( Figure 3E,F). After Baf treatment, p62 and LC3-II were more accumulated and both markers were decreased by HDACi treatment which were administrated after Baf treatment while increasing SNAP25 ( Figure S3E). Next, we used a tandem fluorescent-tagged mRFP-GFP-LC3 reporter to assess autophagic flux. 9 Autophagosomes (mRFP + -GFP + -LC3) were abnormally accumulated in NPC-iNSCs, whereas HDACi or Rapa-treated cells exhibited increased autolysosomes (mRFP + -mRFP − -LC3), suggesting that HDACi can facilitate autophagy-inducing activity and autophagic flux ( Figure 3H,I). Elevated levels of autophagy markers in cerebellar lysates of NPC1 KO mice were reduced in HDACi-treated mice and SNAP25 was upregulated in the cerebellum of HDACi-treated mice ( Figure 3J,K). These results verified the effect of HDACi related to autophagic flux with SNAP25 upregulation in NPC1 KO mice, as well as in vitro using NPC-iNSCs.
Previous studies have reported that NPC-iNSCs are defective in neuronal differentiation. 10 NPC-iNSCs exhib-ited reduced levels of TUJ1 (early neuron marker), neurofilament (NF), and MAP2 (mature neuron markers) compared to WT-iNSCs. Notably, the number of TUJ1-and NF-positive cells was significantly upregulated by HDACitreatment in NPC-iNSCs ( Figure 4A-F). In addition, SNAP25 overexpression could rescue the neuronal differentiation defects of NPC-iNSCs by upregulating the level of NF in NPC-iNSCs ( Figure 4G,H). These results demonstrated that enriched SNAP25 via either pharmacological inhibition of HDAC or overexpression of SNAP25 alleviates defective neuronal differentiation in NPC-iNSCs.
In conclusion, this study demonstrated that SNAP25 is a key player in alleviating the pathological phenotypes of NPC disease. Upregulation of SNAP25 by HDACitreatment rescued impaired autophagic flux and reduced abnormally accumulated cholesterol in NPC-iNSCs cells via compensating for deficient STX17-SNAP29-Vamp8 complexes with STX17-SNAP25-Vamp8 complexes. Furthermore, increased SNAP25 recovered deficient neuronal differentiation capacity in NPC-iNSCs and also improved survival of Purkinje cells in NPC1 KO mice, demonstrating that SNAP25 could enhance autophagy pathways and neuronal differentiation in vitro and in vivo ( Figure 4I). Collectively, SNAP25 could be a novel therapeutic target for NPC disease since the upregulation of SNAP25 through HDACi treatment has shown both the increase of neuronal differentiation and the reduction of cholesterol accumulation.