Long Non-coding RNABACE1-AS Functions As A Competing Endogenous RNA Of SERF1a And BACE1 To Promote β-amyloid Aggregation In Alzheimer’s Disease

An estimated 24 million people suffer from dementia worldwide, of which the majority is thought to suffer from Alzheimer’s disease(AD). Therefore, Alzheimer’s disease is a major public health problem, AD and its assotiated research field attracting more and more attention from all over the world. Despite symptomatic treatment can alleviate the symptoms of Alzheimer’s disease, but its pathogenesis urgent further study. One major pathological hallmark of Alzheimer’s disease is the amyloid plaques in brain tissue. The main component of amyloid plaques is the over generated amyloid β(Aβ) and its aggregation, which has negative impact on memory and is the key to the pathogenesis of AD.Many previous investigations focus on the mechanisms regulating Aβ production in AD. It is found that Aβ is produced via sequential proteolytic cleavage of amyloid precursor protein(APP) by BACE1 and γ-secretase. Compared with that of normal controls, the BACE1 concentration in subjects with Alzheimer’s disease is elevated. BACE1 is therefore recognized as a critical regulator in the pathogenesis of AD by controlling Aβ generation. However, BACE1 regulatory factors and the mechanisms mediated its upregulation in AD patients’ brain tissue are still not entirely clear.In another term, an increasing number of reports reveal that Aβ fibrils are highly polymorphic. Most recently, experiments on tissue from two AD patients with distinct clinical histories showed different Aβ fibril structure in each patient. All these findings indicated that Aβ aggregation process and its regulatory mechanism is critical for the understanding of AD development. An evolutionarily conserved protein, called MOAG-4/SERF, was reported to directly affect the aggregation kinetics of Aβ proteins in vitro, although the biological and pathological functions of human SERF1 a in AD patients remain to be defined. Hence, understanding the molecular mechanism of the deregulation of these key regulators, such as BACE1 and SERF1 a, may be important for elucidating the pathology, clinical diagnosis and therapy of Alzheimer’s disease.Changes in the miRNA levels of AD patients indicated that the regulatory function of miRNAs via targeting many important transcripts, such as those of BACE1 and APP, might also play a key role in the pathogenesis of AD. However, despite the important role of these mi RNAs, their regulators in AD brains remains undefined. At the same time, multiple lines of evidence indicate that another type of epigenetic modifier, lncRNA, which are more than 200 nucleotides in length, may also contribute to the pathogenesis of AD. In a previous report, Faghihi and colleagues reported that the level of one lncRNA, named BACE1-AS, which is an antisense transcript of BACE1, was elevated in subjects with AD. BACE1-AS was also reported to form RNA-duplex with BACE1 and increase the expression of BACE1 and BACE1-AS deficiency reduced Aβ production. However, it is unclear whether BACE1-AS has other biological functions and regulatory targets when it is aberrantly elevated in AD patients and mouse AD models.Several recent reports have provided a model suggesting that lncRNA may function as a competing endogenous RNA(ceRNA) to “sponge” mi RNAs and modulate the concentration and biological functions of mRNAs. In our previous work, we also proved that linc-RoR functions as a ceRNA that links the mi RNA and transcriptional networks in human embryonic stem cells. However, the role of ceRNAs in the process of AD development and amyloid aggregation has not yet been elucidated.In our current study, we mainly focus on BACE1-AS and its function. And we found that BACE1-AS greatly promoted Aβ intracellular aggregation in vitro and in vivo in a SERF1a-dependent manner. We also found that BACE1-AS functions as a ceRNA to post-transcriptionally regulate SERF1 a expression. The levels of BACE1-AS and SERF1 a expression were well correlated in the brain tissue of individuals with AD. These data suggested that BACE1-AS has a role in regulating SERF1 a expression and driving AD pathogenesis. PartⅠLong non-coding RNA BACE1-AS promote Aβ aggregation Methods:(1) In the Aβ overproduction cellular models,overexpression and knockdown BACE1-AS,using Western-blot to test Aβ and its aggregation.(2) Excluding the impact of BACE1 and detect Aβ aggregation by Western-blot and immunofluorescence staining.(3)Using in vitro Aβ Kinetic aggregation assay with cell PDS.(4) Test the impact of BACE1-AS on AD-related cellular phenotype.(5)Screen out BACE1-AS target gene.(6)Identify whether mi RNA participate the process of BACE1-AS regulating Aβ aggregation through Dicer knock-down. Results:(1) BACE1-AS promotes Aβ intracellular aggregation by Western-blot and immunofluorescence staining.(2)Kinetics of Aβ aggregation assayed using thioflavin T(ThT) fluorescence proved BACE1-AS promotes Aβ aggregation.(3) BACE1-AS has negative impact on cell viability and promote opoptosis in Aβoverproduction cellular models.(4) BACE1-AS increase Tau phosphorylation.(5) SERF1 a was identified as BACE1-AS regulated gene.(6)Proved miRNA participate the process of BACE1-AS regulating Aβ aggregation. Conclusions: BACE1-AS promotes Aβ intracellular aggregation. BACE1-AS-mediated Aβ aggregate formation was dependent on SERF1 a. MiRNA participate the process of BACE1-AS regulating Aβ aggregation. Part Ⅱ BACE1-AS upregulate SERF1 a through competing endogenous RNA mechanismMethods:(1) Identify mi RNAs shared by BACE1-AS and SERF1 a through bioinformatics tool.(2)Using luciferase reporters assay, qrt-PCR,Western-blot to comfirm these miRNA can target both BACE1-AS and SERF1 a.(3) Construct transcripts with mutation(Mut) in the miRNA response elements(MREs) and test its impaction on SERF1 a.(4) Using real-time qRT-PCR to quantify the exact copy numbers of BACE1-AS, SERF1 a and the three miRNAs per cell.(5) Test the impact of these mi RNAs on AD-related cellular phenotype.(6)Test the inpact of BACE1-AS on these miRNAs.(7) Northern-blot and RNA stability assays test whether BACE1-AS promote miRNA degradation.(8)RIP analysis test whether BACE1-AS directly bind to these miRNAs.Results:(1) Identified miR-1285, miR-1273, and miR-3064 are shared by BACE1-AS and SERF1 a through bioinformatics tool.(2)Luciferase reporters assay, qrt-PCR,Western-blot comfirmed that these miRNA can target both BACE1-AS and SERF1 a.(3) BACE1-AS-mut without MREs formi R-1285, miR-1273, and miR-3064 abolished the effect of BACE1-AS on SERF1 a.(4)The exact copy numbers of BACE1-AS, SERF1 a and the three miRNAs per cell were satisfy the conditions of the ceRNA mechanism.(5)CCK-8, qrt-PCR, Western-blot proved these miRNAs negativelyaffectedAD-related cellular phenotype.(6) BACE1-AS can decrease the mature form of these miRNAs but has no significant effect on their pre-mature form.(7) Northern-blot and RNA stability assays proved BACE1-AS promoted miRNA degradation.(8) RIP analysissuggested BACE1-AS directly bind to these miRNAs.Conclusions: MiR-1285, miR-1273, and miR-3064 target both BACE1-AS and SERF1 a, BACE1-AS play as a competing endogenous RNA(ceRNA) of SERF1 a and bind to these miRNAs to potect SERF1 a, resulting the increase of SERF1 a. Part Ⅲ BACE1-ASupregulate BACE1 through competing endogenous RNA mechanismMethods:(1) Constructed a BACE1-AS-Δ overexpression construct, lacking the complementary region, and decect whether the other part of BACE1-AS has function on regulating BACE1.(2) Constructed a BACE1-AS-complementary overexpression construct, containing only the complementary region in BACE1-AS, and decect whether RNA duplex is nessassary for BACE1 upregulation by BACE1-AS.(3) Identify miRNAs shared by BACE1-AS and BACE1 through bioinformatics tool.(4) Test whether BACE1-AS upregulate BACE1 through ceRNA mechanism following the methods we used in PartⅡ.Results:(1) Overexpression of BACE1-AS-Δ, which cannot form RNA duplexes with BACE1 mRNA, can also increase BACE1 expression.(2)Ectopic expression of BACE1-AS-complementary couldn’t increase BACE1.(3)Identified mir-29 b,mir-107,mir-124,mir-485,mir-761 are shared by BACE1-AS and BACE1.(4) BACE1-AS can decrease the mature form of these miRNAs but has no significant effect on their pre-mature form.(5) Northern-blot and RNA stability assays proved BACE1-AS promoted miRNA degradation.(6) RIP analysissuggested BACE1-AS directly bind to these miRNAs.Conclusions: RNA duplex is nessassary for BACE1 upregulation by BACE1-AS, other part of BACE1-AS besides its segment complementary to BACE1 also has function on upregulating BACE1. BACE1-AS play as a ceRNA of BACE1 and bind to mir-29 b,mir-107,mir-124,mir-485,mir-761 to potect BACE1, resulting the increase of BACE1. Part Ⅳ In vivo AD-mice experiments and associated gene expression in AD patientsMethods:(1) Intracranially injecte BACE1-AS over-expressionlentivirus into the brain of APP/PS1 transgenic mice(ADmodel),and detect the Aβ and amyloid plaques.(2) Test the effect of BACE1-AS on the expression of BACE1, SERF1 a and related miRNA.(3)Find clinicalevidence in the existing array data from AD patients’ samples in GEO database.Results:(1) Treatment with the LV-BACE1-AS increased Aβ aggregation in vivo compared with treatment with the LV-con control.(2) Aβ immunoreactivity and the number and size of amyloid plaques were increased after LV-BACE1-ASinjection compared with treatment with the LV-con control.(3)BACE1-AS, SERF1 a and BACE1 mRNA were all significantly elevated in the AD-mice brain region injected with LV-BACE1-AS compare with that injected with LV-con.(4)The expression levels of BACE1-AS-regulated miRNAs shared by SERF1 a and BACE1 were also validated in vivo.(5)The abundance of miR-485, miR-107, miR-1273 d, and mi R-3064 were significantly decreased while SERF1 a was increased in AD patients than the controls.(6) We found BACE1, PSEN1, PSEN2, and TAU displayed a strong positive correlation with SERF1 a in all AD patients.Conclusions:BACE1-AS promote Aβ generation and aggregation in vivo. The experiment results in AD-mice and the date analysis in AD patients consistent to our results in vitro. SERF1 a is an important factor that may play a role in AD development. SummaryAmyloid-beta(Aβ) aggregates, the major pathological hallmark of Alzheimer’s disease(AD), and modifiers of Aβ production and aggregation have been intensely investigated. However, how these modifiers are post-transcriptionally regulated during AD development is poorly understood. Here, we report evidence that a known AD-related long non-coding RNA, BACE1-AS, greatly promoted Aβ aggregation by inducing elevated SERF1 a, a direct modifier of amyloid fiber assembly, and increased the cytotoxicity of Aβ. Furthermore, BACE1-AS post-transcriptionally increased the expression of SERF1 a and BACE1 by sequestering and depredating the microRNAs targeting to SERF1 a mRNA or BACE1 mRNA. Finally, in in vivo assays, we found that BACE1-AS over-expression induced SERF1 a and BACE1 elevation, micro RNAs decrease and Aβ aggregation in APP/PS1 transgenic mice. Our results suggest that BACE1-AS functions as a competing endogenous noncoding RNA to control SERF1 a and BACE1 expression and that deregulation of this balance may increase Aβ generation and intracellular aggregation, promoting Alzheimer’s disease–associated pathophysiology.