Circular RNA circTrpc6-induced tau phosphorylation contributes to neuronal injury and cognitive decline via miR-485-5p/eIF4E axis in senescence-accelerated mouse prone 8 mice Authors: Jin-Lan Huang, Qiu-Qing Yan, Yan-Bin Guo, Yu-Xuan Du, Ling-Ling Liu, Xiao-Jing Sun, Chao Yu, Deng-Pan Wu Published: 2026 ABSTRACT Tau hyperphosphorylation, a key neuropathological feature in neurodegenerative disorders, is strongly linked to cognitive decline. CircRNAs are recognized to play a crucial regulatory role in numerous human diseases, yet their impacts on neurodegenerative diseases remain unclear. This study was aimed to characterize and analyze the contribution of circTrpc6 to neurodegeneration and its molecular mechanisms. We found that circTrpc6 was mainly expressed in the hippocampal and cortical regions and was a cytoplastic circRNA with high abundance in neuronal cells, and its expression was upregulated in the SAMP8 hippocampus. CircTrpc6 knockdown attenuated tau phosphorylation and aggregation in N2a cells and the SAMP8 hippocampus, alleviated hippocampal neuronal and synaptic injury and improved cognitive function of SAMP8 mice. Overexpression of circTrpc6 increased tau phosphorylation and aggregation in N2a cells. Mechanistically, circTrpc6 could modulate eIF4E and p-eIF4E expression and the interaction of eIF4E and tau mRNA via interacting with miR-485-5p and influencing miR-485-5p expression, thus regulating tau phosphorylation and aggregation and affecting cognitive function. Collectively, our work indicates a crucial role of circTrpc6 in regulating tau phosphorylation and cognitive function via miR-485-5p/eIF4E axis, and implies that circTrpc6 may serve as an important regulator in tau phosphorylation and represents a potential target for treatment of cognitive decline in neurodegenerative diseases.Supplementary InformationThe online version contains supplementary material available at 10.1007/s10565-026-10214-y. Introduction Aging constitutes a key contributor to cognitive decline in neurodegenerative conditions, such as various dementias (Liu et al. 2024). Despite extensive studies on age-related cognitive decline have been conducted over recent decades, the underlying molecular mechanisms remain poorly understood, and effective treatment options continue to be limited (Liu et al. 2024). Tau hyperphosphorylation is the neuropathological hallmarks of neurodegenerative disorders, including Alzheimer’s disease (AD) (2024; Liu et al. 2024). This pathogenic form of tau has been shown to disrupt microtubule stability and axonal transport, ultimately causing neuronal death, and is strongly correlated with cognitive impairment (Alonso et al. 2016; Huber et al. 2018). Therefore, delineating the regulatory mechanisms controlling tau hyperphosphorylation represents a promising strategic avenue for developing new treatments for age-related cognitive decline in neurodegenerative diseases. Circular RNAs (circRNAs) belong to a distinct category of endogenous RNAs featured by a covalently closed loop configuration, devoid of a 5’ cap as well as a 3’ poly(A) tail. (Kristensen et al. 2019). This unique circular conformation makes them more stable than linear RNAs (Kristensen et al. 2019). CircRNAs were originally regarded as nonfunctional byproducts of aberrant splicing until their capacity to act as microRNA (miRNA) sponges was uncovered (Pisignano et al. 2023). In recent years, circRNAs have attracted growing attention because of their newly recognized functions in various human disorders, such as cancer, cardiovascular disorders, diabetes, and neurological diseases (Mei and Chen 2022; Wang et al. 2025; Yuan et al. 2023; Zhao et al. 2025). Notably, circRNAs are enriched in the mammalian brain and accumulate with advancing age (Gruner et al. 2016; Weigelt et al. 2020), suggesting their potential involvement in age-related neurodegenerative disorders (Wu et al. 2023). Indeed, several reports have indicated that circRNAs are differentially expressed in brain tissues derived from the senescence-accelerated mouse prone 8 (SAMP8) strain, and involved with the pathways associated with neurodegenerative disorders, including AD (Huang et al. 2018a; Zhang et al. 2017). Nevertheless, the role of circRNAs in neurodegenerative diseases remains largely unknown. SAMP8, a naturally occurring mouse strain exhibits a phenotype of accelerated senescence and recapitulates key aspects of age-related neurodegenerative alterations associated with age-related dementia, including age-related cognitive deficits and neuropathological characteristics, such as Aβ deposition, tau hyperphosphorylation, synaptic deficit and neuronal loss, making them an appropriate model for investigating age-related neurodegenerative diseases (Grinan-Ferre et al. 2018; Jiao et al. 2022; Pacesova et al. 2022). In the study, we found that circTrpc6, a circular RNA that derived from exons 4 ~ 10 of the Trpc6 gene, was upregulated in the SAMP8 hippocampus. Functional and mechanistic investigations showed that circTrpc6 interacted with miR-485-5p to regulate eIF4E expression, affecting the binding of eIF4E with tau mRNA and tau expression, thereby influencing tau phosphorylation and cognitive function. Collectively, this study unveils circTrpc6 as a novel regulator of tau phosphorylation and positions circRNA-mediated signaling as a therapeutic avenue for age-related cognitive decline in neurodegenerative diseases. Materials and methods Mice and ethics statement SAMR1 and SAMP8 male mice from Health Science Center of Peking University (China) and Jiangsu Huachuang-Sino Pharma Tech Co., Ltd. (China) were housed under regulated temperature and humidity, exposed to a twelve-hour light/dark cycle, and had free access to both food and water. Ethical approval for the experimental procedures and animal care protocols was granted by the Animal Welfare Committee of Xuzhou Medical University. Cell culture, transfection, real-time polymerase chain reaction (RT-PCR), RNase R treatment and Western blot assays Cell culture, transfection, RT-PCR, RNase R treatment and Western blot assays were performed according to our previous study (Wu et al. 2025). Detailed methods were provided in Supplemental Materials and Methods. Microinjection of adeno-associated viruses and behavioral tests Adeno-associated viruses (AAVs) or LVs were microinjected into 3-month-old SAMP8 hippocampus, following a previously reported protocol with modifications (Huang et al. 2021). Behavioral tests, including Morris water maze (MWM) test, passive avoidance test and novel object recognition task were performed following previously described methods (Huang et al. 2021; Wu et al. 2024). Detailed methods of microinjection of AAVs and behavioral tests were provided in Supplemental Materials and Methods. shRNA sequences are listed in Supplemental Table 1. Label-free quantitative proteomic analysis Label‑free quantitative proteomic analysis was carried out by Aksomics Bioscience Institute (Shanghai, China). Briefly, Protein extraction and sample lysis were performed with an SDT buffer containing 1 mM DTT, 100 mM Tris–HCl, and 4% SDS, pH 7.6. Following digestion by trypsin, the protein digests were measured using a Q Exactive mass spectrometer (Thermo Scientific), which allowed for the identification and quantitation of peptides and proteins. RNA pull-down assay The assay was conducted following a published method with slight alterations (Zhang et al. 2021). After harvesting and lysing N2a cells, the lysates were incubated at 4 °C with streptavidin-coated magnetic beads (BEAVER Biomedical, Suzhou, China) that had been precoated with biotinylated probes specifically targeting miR-485-5p (General Biology Co., Ltd., China) overnight. The probe sequence is listed in Supplemental Table 1. After washing the beads, RNAs were extracted. Subsequently, the expression levels of circTrpc6 and the eIF4E mRNA 3′UTR were measured by RT-PCR. RNA-fluorescence in situ hybridization (FISH) assay Probes labeled with FAM that specifically recognize the back‑splice junction of circTrpc6 were generated by General Biology Co., Ltd. (Chuzhou, China). The probe sequence is listed in Supplemental Table 1. Signal detection was carried out with a Fluorescent In Situ Hybridization Kit following the manufacturer’s recommended protocol (Guangzhou RiboBio Co., Ltd., China). Images were acquired using a fluorescence microscope (Olympus Corporation, Japan). Staining assays Golgi, Nissl and TUNEL staining assays were carried out following the protocol of manufacturers. Detailed methods were provided in Supplemental Materials and Methods. Detection of aggregates in vitro In vitro aggregation was assessed by Thioflavin S (ThS) staining assay following a published method with minor modifications (Khlistunova et al. 2006). Briefly, N2a cells cultured on coverslips were subjected to fixation with paraformaldehyde for 30 min, permeabilizated with Triton X-100 for 6 min at room temperature, incubated with 1% ThS (Sigma) at 37 °C for 20 min, and finally rinsed three times with PBS. Detection of aggregates in brain sections Brain section aggregates were detected using an adapted protocol from a previous study (Liu et al. 2020). Briefly, the sections were first permeabilized and blocked using a solution consisting of goat serum and Triton X-100 for 15 min at room temperature. They were then incubated at 4 °C with an anti-tau primary antibody (HUABIO, China, diluted 1:200) overnight. Following PBS washes, the sections were exposed to a secondary antibody. ThS staining was then performed using 0.3% ThS (Sigma) in 50% ethanol for 10 min. Finally, the brain section was rinsed with 80% ethanol and PBS, then mounted for imaging. RNA Immunoprecipitation (RIP) The interaction between eIF4E and tau mRNA was assessed by a RIP assay based on a previously described method with modifications (Gagliardi and Matarazzo 2016). In brief, cells were collected and lysed, and the resultant lysates were incubated with protein A/G agarose beads (Santa Cruz Biotechnology, USA) along with an eIF4E antibody (ABclonal, China) overnight at 4 °C. Following incubation, the co‑precipitated RNA was extracted, and RT‑PCR was performed to quantify tau mRNA expression. Statistical analysis Data from all experimental groups are expressed as means ± SEM. Statistical calculations were performed using SPSS 20.0 for Windows. Student’s t test was used to evaluate differences between two groups. When three or more groups were compared, one‑way analysis of variance (ANOVA) was used. Statistical significance was defined as a P value < 0.05. Results CircTrpc6 is a stable cytoplastic circRNA that is abundant in neuronal cells and upregulated in the SAMP8 hippocampus According to our previous data of circRNA microarray, we found that circTrpc6 (mmu_circRNA_013636) was reduced in the hippocampus of SAMP8 mice administrated with Panax Notoginseng Saponins (PNS), a traditional Chinese medicine that has been reported to alleviate neurodegenerative changes and improve cognitive impairment in SAMP8 mice (Huang et al. 2014, 2018b; Yang et al. 2024), suggesting that circTrpc6 may play a role in neurodegenerative disease. CircTrpc6 is a circular RNA that is originated from exons 4 to 10 of Trpc6 gene (Fig. 1B). Owing to the more resistance of circRNAs to RNase R exonuclease than linear mRNAs (Kristensen et al. 2019), RNase R treatment assay was performed to measure the stability of circTrpc6. The results revealed that compared with linear ACTB mRNA, circTrpc6 displayed greater resistance to RNase R treatment (Fig. 1A). For further confirming whether circTrpc6 is circular in form, the RT-PCR products were subjected to Sanger sequencing for sequence verification. As demonstrated in Fig. 1B, the RT-PCR products exhibited a sequence that was completely identical to that of circTrpc6 and revealed a back-splice junction, confirming circTrpc6 as a stable circular RNA.Fig. 1CircTrpc6 expression in brain regions, organs and different cells, as well as subcellular location of circTrpc6 and effect of circTrpc6 on cognitive function and hippocampal neuronal and synaptic injury in SAMP8 mice. (A) RNase R resistance of circTrpc6 was detected by RNase R treatment assay, N = 3. (B) Sanger sequencing of the PCR products of circTrpc6. (C) Relative expression of circTrpc6 in the hippocampus, cortex, heart, liver, spleen, lung, and kidney was measured by RT-PCR assay, N = 3 ~ 4 mice. (D) Relative expression of circTrpc6 in 3-month-old SAMP8 and age-matched SAMR1 mice was detected by RT-PCR assay, N = 3 mice. (E) Relative expression of circTrpc6 in N2a, BV-2 and C8-D1A cells was measured by RT-PCR assay, N = 3 ~ 4. (F) Subcellular location of circTrpc6 was detected by RNA-FISH assay. (G) Relative subcellular expression of circTrpc6 was measured by RT-PCR assay, N = 3. (H) The infection efficiency of AAV2/9-EGFP in the hippocampal region of 3-month-old SAMP8 mice was evaluated after injection of AAVs for 6 weeks. AAVs (green), DAPI (blue). (I) Relative hippocampal expression of circTrpc6 was measured using RT-PCR assay after 3-month-old SAMP8 mice were infected with AAVs containing circTrpc6 shRNA for 6 weeks, N = 3 mice. (J ~ M) The effect of circTrpc6 knockdown on the cognitive function of SAMP8 mice was evaluated via the MWM test (J), passive avoidance test (K), and novel object recognition task (L, M), N = 7 ~ 8 mice. (N ~ P) The effect of circTrpc6 knockdown on the percentage of Nissl body (N), neuronal apoptotic rate (O) and spine density (P) in the hippocampus of SAMP8 mice was assessed via Nissl, TUNEL and Golgi staining assays, respectively, N = 3 mice. All data in the figure are presented as mean ± SEM. Student’s t test was used to assess statistically significant differences. *P < 0.05, **P < 0.01 CircTrpc6 expression in brain regions, organs and different cells, as well as subcellular location of circTrpc6 and effect of circTrpc6 on cognitive function and hippocampal neuronal and synaptic injury in SAMP8 mice. (A) RNase R resistance of circTrpc6 was detected by RNase R treatment assay, N = 3. (B) Sanger sequencing of the PCR products of circTrpc6. (C) Relative expression of circTrpc6 in the hippocampus, cortex, heart, liver, spleen, lung, and kidney was measured by RT-PCR assay, N = 3 ~ 4 mice. (D) Relative expression of circTrpc6 in 3-month-old SAMP8 and age-matched SAMR1 mice was detected by RT-PCR assay, N = 3 mice. (E) Relative expression of circTrpc6 in N2a, BV-2 and C8-D1A cells was measured by RT-PCR assay, N = 3 ~ 4. (F) Subcellular location of circTrpc6 was detected by RNA-FISH assay. (G) Relative subcellular expression of circTrpc6 was measured by RT-PCR assay, N = 3. (H) The infection efficiency of AAV2/9-EGFP in the hippocampal region of 3-month-old SAMP8 mice was evaluated after injection of AAVs for 6 weeks. AAVs (green), DAPI (blue). (I) Relative hippocampal expression of circTrpc6 was measured using RT-PCR assay after 3-month-old SAMP8 mice were infected with AAVs containing circTrpc6 shRNA for 6 weeks, N = 3 mice. (J ~ M) The effect of circTrpc6 knockdown on the cognitive function of SAMP8 mice was evaluated via the MWM test (J), passive avoidance test (K), and novel object recognition task (L, M), N = 7 ~ 8 mice. (N ~ P) The effect of circTrpc6 knockdown on the percentage of Nissl body (N), neuronal apoptotic rate (O) and spine density (P) in the hippocampus of SAMP8 mice was assessed via Nissl, TUNEL and Golgi staining assays, respectively, N = 3 mice. All data in the figure are presented as mean ± SEM. Student’s t test was used to assess statistically significant differences. *P < 0.05, **P < 0.01 For investigating the role of circTrpc6, we detected circTrpc6 expression in brain regions, including hippocampus and cortex, and different organs, including kidney, heart, liver, lung, and spleen by RT-PCR assay. The findings revealed that circTrpc6 expression was enriched in hippocampal and cortical regions (Fig. 1C). Due to the relative hippocampal enrichment of circTrpc6, we measured the hippocampal expression of circTrpc6 in 3-month-age SAMP8 and age-matched control strain SAMR1. The findings demonstrated that hippocampal expression of circTrpc6 was elevated in SAMP8 mice relative to SAMR1 controls (Fig. 1D). We further assessed its expression in neuronal, microglial, and astrocytic cells, and found that circTrpc6 was abundant in neuronal cells (Fig. 1E). Then, subcellular localization of circTrpc6 in neuronal cells was detected using RNA-FISH and RT-PCR assays. The findings indicated that circTrpc6 was mainly located in cytoplasm (Fig. 1F and G). Hippocampal circTrpc6 knockdown improves cognitive function and alleviates neuronal and synaptic injury in SAMP8 mice We next determined whether circTrpc6 has a regulatory role in cognitive function. To this end, AAV2/9 vectors carrying either control or circTrpc6‑targeting shRNA (shcircTrpc6) were administered into the hippocampus of SAMP8 mice via microinjection, and the animals were maintained for six weeks. The efficiency of AAV infection was evaluated by morphology and RT-PCR assay, showing that AAVs were mainly in the hippocampal region and a remarkable downregulation of circTrpc6 expression was confirmed in the shcircTrpc6 group (Fig. 1H and I), indicating efficient AAV-mediated knockdown. The cognitive performance of mice was assessed using the MWM and passive avoidance tests as well as novel object recognition task. Specifically, in the MWM test, circTrpc6 knockdown significantly increased platform entries and prolonged target quadrant time (Fig. 1J), suggesting enhanced spatial memory by circTrpc6 knockdown. The passive avoidance test revealed a marked extension in step-through latency (Fig. 1K), implying improved retention memory following circTrpc6 knockdown. Furthermore, in the novel object recognition task, circTrpc6 knockdown led to a remarkable elevation in the percentage of exploration entries and time toward the novel object (Fig. 1L and M), indicating circTrpc6 knockdown ameliorates recognition memory in SAMP8 mice. Subsequently, we assessed whether circTrpc6 knockdown has a role in neuronal and synaptic damage through Nissl, TUNEL and Golgi staining assays. The results revealed that circTrpc6 knockdown elevated the percentage of Nissl body, reduced neuronal apoptotic rate and enhanced synaptic density in the SAMP8 hippocampus (Fig. 1N ~ P), suggesting a significant alleviation of neuronal and synaptic damage following circTrpc6 knockdown. Collectively, our findings indicate that hippocampal knockdown of circTrpc6 improves cognitive performance and mitigates neuronal and synaptic injury in SAMP8 mice. eIF4E is the downstream target of circTrpc6 Next, differential expression of proteins in the SAMP8 hippocampus with hippocampal overexpression of circTrpc6 were identified. Efficiency of infection of AAVs overexpressing circTrpc6 was provided in Supplemental table S1. The results showed that there were 127 dysregulated proteins between the control and circTrpc6 overexpression groups (Fig. 2A), among which 38 upregulated proteins with a fold change > 1.2 and 53 downregulated proteins with a fold change > 1.2 were identified. Then, differential expression of proteins with a fold change > 1.2 were used to build a protein–protein interaction (PPI) network using the STRING database, and Cytoscape software with Degree, MCC, and MNC algorithms was employed to identify the top ten hub proteins based on their ranking scores. The overlapping hub proteins identified through these analyses included Ndufs8, Hnrnpa1, Atp6v1f, Suclg2, Sdhc, Syncrip and eIF4E (Fig. 2B).Fig. 2eIF4E is the downstream target of circTrpc6. (A) A volcano plot showing dysregulated proteins in three hippocampal samples from control and circTrpc6-overexpressing mice via proteomics, N = 3 samples/group. (B) Cytoscape software with Degree, MCC, and MNC algorithms was employed to identify the top ten hub proteins based on their ranking scores. (C) Relative expression of circTrpc6 was evaluated via RT-PCR assay after N2a cells was infected with LVs containing circTrpc6 shRNA, N = 3. (D) The effect of circTrpc6 knockdown on eIF4E and p-eIF4E expression in N2a cells was assessed using Western blot assay. N = 4. (E) The effect of circTrpc6 knockdown on eIF4E and p-eIF4E expression in the hippocampus of SAMP8 mice was assessed using Western blot assay. N = 4. (F) Relative expression of circTrpc6 was evaluated via RT-PCR assay after N2a cells was infected with circTrpc6 overexpression LVs, N = 3. (G) The effect of circTrpc6 overexpression on eIF4E and p-eIF4E expression in N2a cells was assessed using Western blot assay. N = 4. All data in the figure are presented as mean ± SEM. Student’s t test was used to assess statistically significant differences. **P < 0.01 eIF4E is the downstream target of circTrpc6. (A) A volcano plot showing dysregulated proteins in three hippocampal samples from control and circTrpc6-overexpressing mice via proteomics, N = 3 samples/group. (B) Cytoscape software with Degree, MCC, and MNC algorithms was employed to identify the top ten hub proteins based on their ranking scores. (C) Relative expression of circTrpc6 was evaluated via RT-PCR assay after N2a cells was infected with LVs containing circTrpc6 shRNA, N = 3. (D) The effect of circTrpc6 knockdown on eIF4E and p-eIF4E expression in N2a cells was assessed using Western blot assay. N = 4. (E) The effect of circTrpc6 knockdown on eIF4E and p-eIF4E expression in the hippocampus of SAMP8 mice was assessed using Western blot assay. N = 4. (F) Relative expression of circTrpc6 was evaluated via RT-PCR assay after N2a cells was infected with circTrpc6 overexpression LVs, N = 3. (G) The effect of circTrpc6 overexpression on eIF4E and p-eIF4E expression in N2a cells was assessed using Western blot assay. N = 4. All data in the figure are presented as mean ± SEM. Student’s t test was used to assess statistically significant differences. **P < 0.01 It has been demonstrated that cytoplasmic-localized circRNAs primarily function through miRNA sponge mechanisms (Kristensen et al. 2019; Ma et al. 2023; Wang et al. 2025). Given our observation that circTrpc6 is predominantly localized in the cytoplasm as detected by RNA-FISH and RT-PCR assays (Fig. 1F and G), circTrpc6 may sequester miRNAs to alleviate their inhibitory effects on the expression of target genes, thereby upregulating downstream target expression. We ranked the upregulated proteins including Ndufs8, Hnrnpa1, and eIF4E based on their fold change and found that eIF4E exhibited the highest upregulation, suggesting that eIF4E may serve as a downstream target of circTrpc6. Next, stable N2a cell lines with circTrpc6 overexpression and knockdown were constructed (Fig. 2C and F) and whether circTrpc6 impacts eIF4E and phosphorylated eIF4E (p-eIF4E) expression was investigated using Western blot assay in vitro and in vivo. The results indicated that knockdown of circTrpc6 significantly downregulated eIF4E and p-eIF4E expression in N2a cells and the SAMP8 hippocampus (Fig. 2D and E). Overexpression of circTrpc6 remarkedly upregulated eIF4E and p-eIF4E expression in N2a cells (Fig. 2G). Our findings suggest that eIF4E serves as a downstream target of circTrpc6. Hippocampal eIF4E knockdown leads to improved cognitive function and mitigation of neuronal and synaptic injury in SAMP8 mice We next examined the impact of eIF4E on cognitive function and neuronal and synaptic injury in SAMP8 mice. RT-PCR analysis showed that hippocampal eIF4E expression was remarkably upregulated in 3-month-old SAMP8 mice relative to age-matched SAMR1 controls (Fig. 3A). To assess its functional role, LV particles encoding sheIF4E were injected into the SAMP8 hippocampus to knock down eIF4E. Figure 3B confirmed successful eIF4E knockdown, as evidenced by markedly reduced hippocampal eIF4E level in the sheIF4E group. Importantly, eIF4E knockdown improved cognitive performance in SAMP8 mice, indicated by increased platform entries, extended target quadrant time, longer step-through latency, and higher exploration of novel objects (Fig. 3C ~ F). Moreover, eIF4E knockdown increased the percentage of Nissl body, reduced neuronal apoptotic rate and enhanced synaptic density in the hippocampus (Fig. 3G ~ I). Together, these findings suggest that hippocampal eIF4E knockdown enhances cognitive function and alleviates neuronal injury and synaptic deficits in SAMP8 mice.Fig. 3Effect of eIF4E knockdown on cognitive function, hippocampal neuronal and synaptic injury in SAMP8 mice. (A) Relative expression of eIF4E in 3-month-old SAMP8 and age-matched SAMR1 mice was detected via RT-PCR assay. N = 3 mice. (B) RT-PCR assay was used to detect eIF4E expression in the hippocampus after 3-month-old SAMP8 was infected with eIF4E shRNA AAVs for 6 weeks, N = 3 mice. (C ~ F) The effect of eIF4E knockdown on the cognitive function of SAMP8 mice was evaluated via the MWM test (C), passive avoidance test (D), and novel object recognition task (E, F), N = 8 ~ 10 mice. (G ~ I) The effect of eIF4E knockdown on the percentage of Nissl body (G), neuronal apoptotic rate (H) and spine density (I) in the hippocampus of SAMP8 mice were evaluated via Nissl, TUNEL and Golgi staining assays, respectively, N = 3 mice. All data in the figure are presented as mean ± SEM. Student’s t test was used to assess statistically significant differences. *P < 0.05, **P < 0.01 Effect of eIF4E knockdown on cognitive function, hippocampal neuronal and synaptic injury in SAMP8 mice. (A) Relative expression of eIF4E in 3-month-old SAMP8 and age-matched SAMR1 mice was detected via RT-PCR assay. N = 3 mice. (B) RT-PCR assay was used to detect eIF4E expression in the hippocampus after 3-month-old SAMP8 was infected with eIF4E shRNA AAVs for 6 weeks, N = 3 mice. (C ~ F) The effect of eIF4E knockdown on the cognitive function of SAMP8 mice was evaluated via the MWM test (C), passive avoidance test (D), and novel object recognition task (E, F), N = 8 ~ 10 mice. (G ~ I) The effect of eIF4E knockdown on the percentage of Nissl body (G), neuronal apoptotic rate (H) and spine density (I) in the hippocampus of SAMP8 mice were evaluated via Nissl, TUNEL and Golgi staining assays, respectively, N = 3 mice. All data in the figure are presented as mean ± SEM. Student’s t test was used to assess statistically significant differences. *P < 0.05, **P < 0.01 CircTrpc6 regulates tau phosphorylation via eIF4E eIF4E, a eukaryotic translation initiation factor that recognizes the 5’ cap of mRNAs, is critically involved in mRNA translation and degradation (Mars et al. 2024). Reports demonstrated that eIF4E phosphorylation was markedly improved in the brains of individuals with AD, showing a positive correlation with the expression of tau and phosphorylated tau (p-tau), suggesting its involvement in tau phosphorylation (Li et al. 2004). We therefore hypothesized that eIF4E may act as a key mediator of tau phosphorylation and that circTrpc6 might regulate this process through eIF4E. As expected, knockdown of either eIF4E or circTrpc6 significantly reduced tau and p-tau expression in N2a cells and the SAMP8 hippocampus, without altering the p-tau/tau ratio (Fig. 4A ~ D). Conversely, overexpression of either eIF4E or circTrpc6 remarkably elevated tau and p-tau expression in N2a cells, but did not affect the p-tau/tau ratio (Fig. 4E and F). Additionally, circTrpc6 overexpression increased tau and p-tau expression in N2a cells, which could be reversed by eIF4E knockdown (Fig. 4G). Given that tau phosphorylation promotes its aggregation and toxicity (Avila et al. 2004; Liang et al. 2022), we further evaluated the effects of circTrpc6 and eIF4E on aggregate formation using ThS staining assay. The results demonstrated that knockdown of circTrpc6 or eIF4E markedly reduced ThS/tau-positive aggregates in the hippocampus of SAMP8 mice (Fig. 4H and I). In contrast, overexpression of circTrpc6 or eIF4E remarkably increased the level of aggregates in N2a cells (Fig. 4J and K). Furthermore, the facilitatory role of circTrpc6 overexpression in the aggregate level of N2a cells was suppressed by eIF4E knockdown (Fig. 4L).Fig. 4CircTrpc6 regulates tau phosphorylation via eIF4E. (A) The effect of eIF4E knockdown on the expression of eIF4E, p-eIF4E, tau and p-tau (S396) in N2a cells was detected via Western blot assay. N = 4. (B) The effect of eIF4E knockdown on tau and p-tau expression in the hippocampus of SAMP8 mice was assessed using Western blot assay. N = 4. (C) The effect of circTrpc6 knockdown on tau and p-tau (S396) expression in N2a cells was assessed using Western blot assay. N = 4. (D)The effect of circTrpc6 knockdown on tau and p-tau (S396) expression in the hippocampus of SAMP8 mice was assessed using Western blot assay. N = 4. (E) The effect of eIF4E overexpression on the expression of eIF4E, p-eIF4E, tau and p-tau (S396) in N2a cells was detected via Western blot assay. N = 4. (F) The effect of circTrpc6 overexpression on tau and p-tau (S396) expression in N2a cells was detected via Western blot assay. N = 4. (G) The effect of eIF4E knockdown on tau and p-tau (S396) expression regulated by circTrpc6 overexpression in N2a cells was detected via Western blot assay. N = 4. (H, I) The effect of circTrpc6 and eIF4E knockdown on tau aggregates in the hippocampus of SAMP8 mice was evaluated via ThS staining assay. N = 3 mice. (J, K) The effect of circTrpc6 and eIF4E overexpression on aggregates in N2a cells was evaluated via ThS staining assay. N = 3. (L) The effect of eIF4E knockdown on aggregates regulated by circTrpc6 overexpression in N2a cells was detected via ThS staining assay. N = 3. All data in the figure are presented as mean ± SEM. Student’s t test (A ~ F, H ~ K) and one-way ANOVA (G, L) were used to assess statistically significant differences. *P < 0.05, **P < 0.01, ns, not significant CircTrpc6 regulates tau phosphorylation via eIF4E. (A) The effect of eIF4E knockdown on the expression of eIF4E, p-eIF4E, tau and p-tau (S396) in N2a cells was detected via Western blot assay. N = 4. (B) The effect of eIF4E knockdown on tau and p-tau expression in the hippocampus of SAMP8 mice was assessed using Western blot assay. N = 4. (C) The effect of circTrpc6 knockdown on tau and p-tau (S396) expression in N2a cells was assessed using Western blot assay. N = 4. (D)The effect of circTrpc6 knockdown on tau and p-tau (S396) expression in the hippocampus of SAMP8 mice was assessed using Western blot assay. N = 4. (E) The effect of eIF4E overexpression on the expression of eIF4E, p-eIF4E, tau and p-tau (S396) in N2a cells was detected via Western blot assay. N = 4. (F) The effect of circTrpc6 overexpression on tau and p-tau (S396) expression in N2a cells was detected via Western blot assay. N = 4. (G) The effect of eIF4E knockdown on tau and p-tau (S396) expression regulated by circTrpc6 overexpression in N2a cells was detected via Western blot assay. N = 4. (H, I) The effect of circTrpc6 and eIF4E knockdown on tau aggregates in the hippocampus of SAMP8 mice was evaluated via ThS staining assay. N = 3 mice. (J, K) The effect of circTrpc6 and eIF4E overexpression on aggregates in N2a cells was evaluated via ThS staining assay. N = 3. (L) The effect of eIF4E knockdown on aggregates regulated by circTrpc6 overexpression in N2a cells was detected via ThS staining assay. N = 3. All data in the figure are presented as mean ± SEM. Student’s t test (A ~ F, H ~ K) and one-way ANOVA (G, L) were used to assess statistically significant differences. *P < 0.05, **P < 0.01, ns, not significant In summary, these findings imply that circTrpc6 regulates tau phosphorylation through eIF4E. CircTrpc6 regulates eIF4E expression via miR-485-5p Given that cytoplasmic circRNAs have been shown to interact with miRNAs and modulate their expression, thereby influencing the interaction of miRNA with the 3'UTR of downstream target mRNAs and regulating gene expression (Ma et al. 2023). We next aimed to identify miRNAs that might mediate the regulatory effect of circTrpc6 on eIF4E. We predicted miRNAs that could potentially interact with circTrpc6 using miRanda and RNAhybrid and observed that there were 38 miRNAs with binding energy below −20 kcal/Mol. Next, we identified 135 miRNAs potentially binding the 3'UTR of eIF4E mRNA using TargetScan and miRDB databases. Intersection of these two sets revealed a single overlapping miRNA, miR-485-5p, suggesting its potential to interact with both circTrpc6 and the 3'UTR of eIF4E mRNA (Fig. 5A). To test this, we performed an RNA pull-down assay using a biotin-labeled miR-485-5p probe. The assay revealed significant enrichments of both circTrpc6 and eIF4E mRNA 3'UTR, but not the negative control ACTB mRNA (Fig. 5B and C), confirming specific interaction of miR-485-5p with circTrpc6 and the 3’UTR of eIF4E mRNA. Moreover, knockdown of circTrpc6 increased miR-485-5p expression in N2a cells and the SAMP8 hippocampus, while circTrpc6 overexpression reduced miR-485-5p expression in N2a cells, implying that circTrpc6 modulates the expression of miR-485-5p (Fig. 5D ~ F). These findings suggest that circTrpc6 interacts with miR-485-5p and modulates its expression.Fig. 5CircTrpc6 regulates eIF4E expression via miR-485-5p. (A) Prediction of miRNAs interacting with circTrpc6 using miRanda and RNAhybrid and with eIF4E mRNA 3’UTR using TargetScan and miRDB. (B, C) The interactions of miR-485-5p between circTrpc6 (B) and eIF4E mRNA 3’UTR (C) were assessed via RNA pull-down assay, N = 3. (D, E) The effect of circTrpc6 knockdown on miR-485-5p expression in N2a cells (D) and the hippocampus of SAMP8 mice (E) was evaluated via RT-PCR assay. N = 3. (F) The effect of circTrpc6 overexpression on miR-485-5p expression in N2a cells was assessed via RT-PCR assay. N = 3. (G, H) The effect of miR-485-5p overexpression on eIF4E and p-eIF4E expression in N2a cells (G) and the hippocampus of SAMP8 mice (H) was detected via Western blot assay. N = 4. (I) The effect of miR-485-5p knockdown on eIF4E and p-eIF4E expression in N2a cells was detected via Western blot assay. N = 4. (J) The effect of miR-485-5p overexpression on eIF4E and p-eIF4E expression regulated by circTrpc6 overexpression in N2a cells was detected via Western blot assay. N = 4. All data in the figure are presented as mean ± SEM. Student’s t test (B ~ I) and one-way ANOVA (J) were used to assess statistically significant differences. *P < 0.05, **P < 0.01 CircTrpc6 regulates eIF4E expression via miR-485-5p. (A) Prediction of miRNAs interacting with circTrpc6 using miRanda and RNAhybrid and with eIF4E mRNA 3’UTR using TargetScan and miRDB. (B, C) The interactions of miR-485-5p between circTrpc6 (B) and eIF4E mRNA 3’UTR (C) were assessed via RNA pull-down assay, N = 3. (D, E) The effect of circTrpc6 knockdown on miR-485-5p expression in N2a cells (D) and the hippocampus of SAMP8 mice (E) was evaluated via RT-PCR assay. N = 3. (F) The effect of circTrpc6 overexpression on miR-485-5p expression in N2a cells was assessed via RT-PCR assay. N = 3. (G, H) The effect of miR-485-5p overexpression on eIF4E and p-eIF4E expression in N2a cells (G) and the hippocampus of SAMP8 mice (H) was detected via Western blot assay. N = 4. (I) The effect of miR-485-5p knockdown on eIF4E and p-eIF4E expression in N2a cells was detected via Western blot assay. N = 4. (J) The effect of miR-485-5p overexpression on eIF4E and p-eIF4E expression regulated by circTrpc6 overexpression in N2a cells was detected via Western blot assay. N = 4. All data in the figure are presented as mean ± SEM. Student’s t test (B ~ I) and one-way ANOVA (J) were used to assess statistically significant differences. *P < 0.05, **P < 0.01 Next, we examined whether eIF4E expression is modulated by miR‑485‑5p. The results demonstrated that miR-485-5p overexpression reduced eIF4E and p-eIF4E levels in N2a cells and the hippocampus of SAMP8 mice, while knockdown of miR-485-5p increased their levels in N2a cells (Fig. 5G ~ I). Furthermore, we examined whether circTrpc6 modulates eIF4E and p-eIF4E levels through miR-485-5p. The findings indicated that upregulation of eIF4E and p-eIF4E induced by circTrpc6 overexpression was reversed by overexpression of miR-485-5p in N2a cells (Fig. 5J). Taken together, these data demonstrate that circTrpc6 regulates eIF4E and p-eIF4E expression via miR-485-5p. Hippocampal overexpression of miR-485-5p improves cognitive function, suppresses tau phosphorylation and alleviates neuronal and synaptic injury in SAMP8 mice Subsequently, we assessed whether miR‑485‑5p regulates cognitive performance and tau phosphorylation. Firstly, we detected miR-485-5p level in the SAMP8 hippocampus. The findings showed that the hippocampal level of miR‑485‑5p was markedly decreased in 3‑month‑old SAMP8 mice relative to age‑matched SAMR1 mice (Fig. 6A). We then elevated hippocampal miR-485-5p levels in SAMP8 mice by injecting AAV2/9 vectors into the hippocampus. After six weeks, behavioral assays were conducted to assess the effect of miR-485-5p on cognitive performance. The data revealed that miR-485-5p overexpression elevated platform entries, prolonged target quadrant time and step-through latency, and increased the percentage of exploring time and entries on novel objects (Fig. 6B ~ E), implying that miR-485-5p overexpression improves cognitive function of SAMP8 mice. Furthermore, hippocampal miR-485-5p overexpression attenuated the levels of tau and p-tau in N2a cells and the SAMP8 hippocampus, while miR-485-5p knockdown facilitated tau and p-tau levels in N2a cells (Fig. 6F ~ H). However, miR-485-5p did not influence the ratio of p-tau to tau (Fig. 6F ~ H). ThS staining assay demonstrated that miR-485-5p knockdown increased aggregate level in N2a cells and hippocampal miR-485-5p overexpression reduced ThS/tau-positive aggregates in the SAMP8 hippocampus (Fig. 6I and J). Additionally, hippocampal miR-485-5p overexpression increased the percentage of Nissl body, decreased neuronal apoptotic rate and enhanced synaptic density in the SAMP8 hippocampus (Fig. 6K ~ M).Fig. 6Effect of miR-485-5p overexpression on cognitive function, tau phosphorylation, hippocampal tau aggregates, neuronal and synaptic injury in SAMP8 mice, and on tau phosphorylation and aggregation in N2a cells. (A) Relative expression of miR-485-5p in 3-month-old SAMP8 and age-matched SAMR1 mice was detected via RT-PCR assay. N = 3 mice. (B ~ E) The effect of miR-485-5p overexpression on the cognitive function of SAMP8 mice was evaluated via the MWM test (B), passive avoidance test (C), and novel object recognition task (D, E), N = 8 ~ 10 mice. (F, G) The effect of miR-485-5p overexpression on tau and p-tau (S396) expression in N2a cells (F) and the hippocampus of SAMP8 mice (G) was detected via Western blot assay. N = 4. (H) The effect of miR-485-5p knockdown on tau and p-tau (S396) expression in N2a cells was detected via Western blot assay. N = 4. (I) The effect of miR-485-5p knockdown on aggregates in N2a cells was evaluated via ThS staining assay. N = 3. (J) The effect of miR-485-5p overexpression on tau aggregates in the hippocampus of SAMP8 mice was assessed via ThS staining assay. N = 3. (K ~ M) The effect of eIF4E knockdown on the percentage of Nissl body (K), neuronal apoptotic rate (L) and spine density (M) in the hippocampus of SAMP8 mice were evaluated via Nissl, TUNEL and Golgi staining assays, respectively, N = 3 mice. All data in the figure are presented as mean ± SEM. Student’s t test was used to assess statistically significant differences. *P < 0.05, **P < 0.01, ns, not significant Effect of miR-485-5p overexpression on cognitive function, tau phosphorylation, hippocampal tau aggregates, neuronal and synaptic injury in SAMP8 mice, and on tau phosphorylation and aggregation in N2a cells. (A) Relative expression of miR-485-5p in 3-month-old SAMP8 and age-matched SAMR1 mice was detected via RT-PCR assay. N = 3 mice. (B ~ E) The effect of miR-485-5p overexpression on the cognitive function of SAMP8 mice was evaluated via the MWM test (B), passive avoidance test (C), and novel object recognition task (D, E), N = 8 ~ 10 mice. (F, G) The effect of miR-485-5p overexpression on tau and p-tau (S396) expression in N2a cells (F) and the hippocampus of SAMP8 mice (G) was detected via Western blot assay. N = 4. (H) The effect of miR-485-5p knockdown on tau and p-tau (S396) expression in N2a cells was detected via Western blot assay. N = 4. (I) The effect of miR-485-5p knockdown on aggregates in N2a cells was evaluated via ThS staining assay. N = 3. (J) The effect of miR-485-5p overexpression on tau aggregates in the hippocampus of SAMP8 mice was assessed via ThS staining assay. N = 3. (K ~ M) The effect of eIF4E knockdown on the percentage of Nissl body (K), neuronal apoptotic rate (L) and spine density (M) in the hippocampus of SAMP8 mice were evaluated via Nissl, TUNEL and Golgi staining assays, respectively, N = 3 mice. All data in the figure are presented as mean ± SEM. Student’s t test was used to assess statistically significant differences. *P < 0.05, **P < 0.01, ns, not significant Collectively, these findings imply that miR-485-5p overexpression improves cognitive function, suppresses tau phosphorylation and alleviates neuronal and synaptic injury in SAMP8 mice. CircTrpc6 regulates tau phosphorylation via miR-485-5p/eIF4E axis in vitro Given that circTrpc6 and its downstream targets miR-485-5p and eIF4E affects tau phosphorylation, and circTrpc6 regulates tau phosphorylation via eIF4E, we assumed that circTrpc6 regulates tau phosphorylation via miR-485-5p/eIF4E axis. The data demonstrated that circTrpc6 overexpression enhanced tau and p-tau levels and elevated aggregate level in N2a cells, which could be reversed by overexpression of miR-485-5p (Fig. 7A and B). Moreover, the facilitatory function of miR-485-5p knockdown in regulating tau, p-tau and aggregate levels in N2a cells could be reversed by eIF4E knockdown (Fig. 7C and D). These findings indicate a regulatory role of circTrpc6 in tau phosphorylation via miR-485-5p/eIF4E axis.Fig. 7CircTrpc6 regulates tau phosphorylation and the interaction of eIF4E with tau mRNA via miR-485-5p/eIF4E axis. (A) The effect of miR-485-5p overexpression on tau and p-tau (S396) expression regulated by circTrpc6 overexpression in N2a cells was detected via Western blot assay. N = 4. (B) The effect of miR-485-5p overexpression on aggregates regulated by circTrpc6 overexpression in N2a cells was assessed via ThS staining assay. N = 3. (C) The effect of eIF4E knockdown on tau and p-tau (S396) expression regulated by miR-485-5p knockdown in N2a cells was detected via Western blot assay. N = 4. (D) The effect of eIF4E knockdown on aggregates regulated by miR-485-5p knockdown in N2a cells was assessed via ThS staining assay. N = 3. (E ~ G) The effects of eIF4E (E) and circTrpc6 (F) overexpression and miR-485-5p knockdown (G) on the interaction of eIF4E with tau mRNA were evaluated via RIP assay. N = 3. (H) The effect of miR-485-5p overexpression on the interaction of eIF4E with tau mRNA regulated by circTrpc6 overexpression in N2a cells was detected via RIP assay. N = 3. (I) The effect of eIF4E knockdown on the interaction of eIF4E with tau mRNA regulated by circTrpc6 overexpression in N2a cells was detected via RIP assay. N = 3. (J) The effect of eIF4E knockdown on the interaction of eIF4E with tau mRNA regulated by miR-485-5p knockdown in N2a cells was detected via RIP assay. N = 3. All data in the figure are presented as mean ± SEM. Student’s t test (E ~ G) and one-way ANOVA (A ~ D, H ~ J). **P < 0.01, ns, not significant CircTrpc6 regulates tau phosphorylation and the interaction of eIF4E with tau mRNA via miR-485-5p/eIF4E axis. (A) The effect of miR-485-5p overexpression on tau and p-tau (S396) expression regulated by circTrpc6 overexpression in N2a cells was detected via Western blot assay. N = 4. (B) The effect of miR-485-5p overexpression on aggregates regulated by circTrpc6 overexpression in N2a cells was assessed via ThS staining assay. N = 3. (C) The effect of eIF4E knockdown on tau and p-tau (S396) expression regulated by miR-485-5p knockdown in N2a cells was detected via Western blot assay. N = 4. (D) The effect of eIF4E knockdown on aggregates regulated by miR-485-5p knockdown in N2a cells was assessed via ThS staining assay. N = 3. (E ~ G) The effects of eIF4E (E) and circTrpc6 (F) overexpression and miR-485-5p knockdown (G) on the interaction of eIF4E with tau mRNA were evaluated via RIP assay. N = 3. (H) The effect of miR-485-5p overexpression on the interaction of eIF4E with tau mRNA regulated by circTrpc6 overexpression in N2a cells was detected via RIP assay. N = 3. (I) The effect of eIF4E knockdown on the interaction of eIF4E with tau mRNA regulated by circTrpc6 overexpression in N2a cells was detected via RIP assay. N = 3. (J) The effect of eIF4E knockdown on the interaction of eIF4E with tau mRNA regulated by miR-485-5p knockdown in N2a cells was detected via RIP assay. N = 3. All data in the figure are presented as mean ± SEM. Student’s t test (E ~ G) and one-way ANOVA (A ~ D, H ~ J). **P < 0.01, ns, not significant CircTrpc6 regulates the interaction of eIF4E with tau mRNA via miR-485-5p/eIF4E axis eIF4E modulates mRNA processing through its direct interaction with the methyl‑7‑guanosine (m7G) cap located at the 5′ terminus of mRNAs (von der Haar et al. 2004). The findings that either circTrpc6, miR-485-5p or eIF4E regulated tau and p-tau expression but did not affect the ratio of p-tau to tau prompted us to explore whether the modulatory roles of circTrpc6, miR-485-5p or eIF4E in tau expression and phosphorylation is associated with the interaction of eIF4E with tau mRNA. The results of RIP assay revealed that eIF4E could interact with tau mRNA and eIF4E overexpression increased the interaction of eIF4E and tau mRNA (Fig. 7E). Next, we explored whether circTrpc6 and miR-485-5p affects the interaction of eIF4E with tau mRNA. Our findings indicated that circTrpc6 overexpression or miR-485-5p knockdown improved eIF4E and tau mRNA interaction (Fig. 7F and G), indicating that circTrpc6 and miR-485-5p modulate the interaction of eIF4E with tau mRNA. Further studies revealed that the facilitatory role of circTrpc6 overexpression in the interaction of eIF4E with tau mRNA was reversed by either overexpression of miR-485-5p or knockdown of eIF4E (Fig. 7H and I). Additionally, eIF4E knockdown suppressed the interaction of eIF4E with tau mRNA modulated by miR‑485‑5p knockdown (Fig. 7J). Collectively, the above results suggest that circTrpc6 regulates the interaction of eIF4E with tau mRNA via miR-485-5p/eIF4E axis. Discussion Elevated tau phosphorylation correlates with cognitive impairment and neurodegeneration in age-related neurodegenerative diseases (Alonso et al. 2016; Liang et al. 2022). This study elucidates the regulatory mechanism involving circTrpc6, miR-485-5p, and eIF4E in tau phosphorylation and cognitive function. Knockdown of circTrpc6 or eIF4E, or miR-485-5p overexpression improves cognitive function of SAMP8 mice and attenuates the expression and phosphorylation of tau in vitro and in vivo while overexpression of circTrpc6 or eIF4E, or miR-485-5p knockdown facilitates the expression and phosphorylation of tau in vitro. Further results revealed that circTrpc6 could interact with miR-485-5p and suppresses its expression, leading to increased eIF4E and p-eIF4E expression, thereby facilitating tau expression and phosphorylation. These findings establish the circTrpc6/miR-485-5p/eIF4E axis as a critical pathway modulating tau phosphorylation and cognitive function, revealing its novel role and therapeutic potential for cognitive decline in neurodegenerative disorders. Growing evidence has shown that circRNAs cannot be considered as simple by‑products generated by either transcriptional noise or erroneous splicing events; instead, they carry out essential biological functions and exert critical regulatory roles in diseases (Kristensen et al. 2019; Wang et al. 2025; Wu et al. 2023). Therefore, we focused on the circular transcriptome associated with neurodegenerative diseases and found that circTrpc6 level was reduced in the SAMP8 hippocampus after PNS treatment relative to control mice in our previous study (Huang et al. 2018b). In the present work, we observed increased expression of circTrpc6 in the SAMP8 hippocampus. Further study revealed that circTrpc6 was mainly expressed in neuronal cells but not in microglial and astrocytic cells. Nevertheless, circTrpc6 expression in neurons and glial cells in vivo requires further investigation. Furthermore, hippocampal circTrpc6 knockdown remarkably improved cognitive function and ameliorated hippocampal neuronal injury and synaptic deficit in SAMP8 mice, implying an alleviatory role of circTrpc6 in neuronal and cognitive function. Subsequently, proteomics and bioinformatics were used to identify the downstream targets of circTrpc6 and found that eIF4E is the potential downstream target. Further studies revealed that circTrpc6 could regulate eIF4E and p-eIF4E expression in vitro and in vivo. Evidence indicates that eIF4E phosphorylation was elevated and significantly correlated with the expression of both tau and p-tau in the brains of patients with AD, implying the potential involvement of eIF4E in tau phosphorylation (Li et al. 2004). This study demonstrated that eIF4E level was elevated in the SAMP8 hippocampus and eIF4E knockdown decreased tau and p-tau levels in N2a cells and the hippocampal region of SAMP8 mice, indicating a regulatory role of eIF4E in tau expression and phosphorylation. Inconsistent with our findings, Veo, et al. found that eIF4E knockdown has no effect on human tau expression in SK-N-SH cells (Veo and Krushel 2009). This difference might be attributable to different cell line involved in the study. Due to the regulatory role of eIF4E in tau expression and phosphorylation, we assumed that circTrpc6 might regulate tau expression and phosphorylation via eIF4E. As speculated, circTrpc6 knockdown inhibited tau and p-tau levels in N2a cells and the hippocampus of SAMP8 mice, and circTrpc6 overexpression facilitated tau and p-tau expression, which could be attenuated by eIF4E knockdown in N2a cells. Further studies revealed that circTrpc6 or eIF4E knockdown could attenuate tau aggregation in the hippocampus of SAMP8 mice, while circTrpc6 or eIF4E overexpression led to increased aggregation in N2a cells, as assessed by ThS staining assay. Together, these data imply that circTrpc6 regulates tau phosphorylation and aggregation via eIF4E. In the cytoplasm, circRNAs can function as decoy molecules to modulate target gene expression through a “miRNA sponge” mechanism. Specifically, circRNAs harbor miRNA binding sites that allow them to sequester cognate miRNAs, thereby blocking the interaction between miRNAs and their target mRNAs (Chen 2020; Kristensen et al. 2019). In our study, we found that circTrpc6 was predominantly localized to the cytoplasm using RNA‑FISH and RT‑PCR assays, implying that it may interact with miRNAs to modulate eIF4E expression. Online databases were employed to predict miRNAs capable of binding both circTrpc6 and eIF4E, identifying miR‑485‑5p as a candidate. The interaction of miR‑485‑5p with circTrpc6 and eIF4E was confirmed using an RNA pull‑down assay. Moreover, miR-485-5p could retard the expression of eIF4E, and miR‑485‑5p could reverse the regulatory effect of circTrpc6 on eIF4E expression, suggesting the interaction of circTrpc6 with miR-485-5p and regulation of circTrpc6 in eIF4E expression. Reports have shown that circRNAs interacting with miRNAs can modulate miRNA expression levels through the regulation of miRNA degradation (Chen et al. 2019). We investigated whether circTrpc6 influences miR-485-5p expression. The results revealed that circTrpc6 overexpression could reduce miR-485-5p expression. Nevertheless, it remains to be determined whether circTrpc6 modulates the degradation of miR‑485‑5p. Next, we assessed whether miR-485-5p modulates tau phosphorylation and cognitive function. The results demonstrated that miR-485-5p level was reduced in the SAMP8 hippocampus. miR-485-5p overexpression attenuated tau and p-tau levels in N2a cells and the hippocampal region of SAMP8 mice, and alleviated hippocampal tau aggregation of SAMP8 mice. miR-485-5p knockdown promoted tau, p-tau and aggregate levels in N2a cells. Furthermore, miR-485-5p overexpression improved hippocampal neuronal injury and the deficit of synaptic morphology and cognitive function of SAMP8 mice, indicating that miR-485-5p regulates tau phosphorylation and cognitive performance. The impact of the circTrpc6/miR-485-5p/eIF4E axis on tau phosphorylation was further confirmed by our in vitro experiments. It is worth noting that despite overexpression of miR‑485‑5p or eIF4E knockdown could largely phenocopy the effects of circTrpc6 knockdown, the degree of improvement in tau phosphorylation and cognitive function was not entirely equivalent, indicating that there might be other potential targets governed by circTrpc6, which need to be further investigated. The present work demonstrated that, in our in vitro and in vivo studies, circTrpc6, miR-485-5p and eIF4E could regulate tau and p-tau expression, but did not impact the ratio of p-tau to tau, indicating the regulatory roles of circTrpc6, miR-485-5p and eIF4E in tau phosphorylation may be associated with their modulation of tau expression. eIF4E is a multifunctional factor regulating mRNA processing, export, degradation and translation in many organs by directly binding to m7G cap at the 5’end of mRNAs (Mars et al. 2024; von der Haar et al. 2004). Therefore, we hypothesized that circTrpc6, miR-485-5p and eIF4E regulating tau expression and phosphorylation might be related with the interaction of eIF4E and tau mRNA. As expected, either overexpression of circTrpc6 and eIF4E or miR-485-5p knockdown could elevate the interaction of eIF4E and tau mRNA. The facilitatory role of circTrpc6 overexpression in the interaction of eIF4E and tau mRNA could be retarded by either miR-485-5p overexpression or eIF4E knockdown. Furthermore, eIF4E knockdown could reversed the interaction of eIF4E and tau mRNA regulated by knockdown of miR-485-5p, indicating that circTrpc6/miR-485-5p/eIF4E axis modulates eIF4E and tau mRNA interaction. However, the interaction of eIF4E and m7G cap at the 5’end of tau mRNA, and whether circTrpc6/miR-485-5p/eIF4E axis regulates tau mRNA processing, export, degradation or translation need to be investigated in further studies. It is important to note that circRNAs and miRNAs have numerous downstream targets, which can form complex regulatory networks and thus play important roles in diseases, including neurodegenerative disorders (Dong et al. 2023; Zhang et al. 2017; Zhong et al. 2018). Additionally, tau phosphorylation can be regulated by multiple phosphokinases or pathways (Tao et al. 2022; Wang and Mandelkow 2016). Therefore, this study cannot rule out the possibility that circTrpc6 regulates tau phosphorylation through other downstream targets that influence phosphokinases or pathways. It would be interesting to clarify whether additional pathways are involved in the regulation of tau phosphorylation by circTrpc6 via using inhibitors or activators of these phosphokinases or pathways in future studies. Several limitations have to be interpreted. First, in the study, a hippocampal injection technique was performed, viral vectors may impact other brain regions because of cerebrospinal fluid circulation. Therefore, the impacts of circTrpc6, miR-485-5p and eIF4E in other brain regions should be considered. Second, in current study, AAV2/9 or LV preparations were used to explore whether circTrpc6, miR-485-5p and eIF4E impacts tau phosphorylation and cognitive function. In further studies, AAV2/9 preparations driven by neuron-specific protomer should be performed. Third, despite the study provide the evidence demonstrating the key functions of circTrpc6, miR-485-5p and eIF4E in inhibiting tau phosphorylation in SAMP8 mice, the direct relevance to AD and other tauopathies is still lacking, which requires investigation in future research. Fourth, it has been demonstrated that miR-485-5p level was reduced in the brains of AD subjects and miR-485-5p could bind to the open reading frame of the beta-secretase-1 (BACE1) mRNA, thus affecting BACE1 expression (Faghihi et al. 2010). BACE1 is a critical enzyme responsible for Aβ production (Hampel et al. 2021). Therefore, miR-485-5p might modulate Aβ production via regulating BACE1 expression. Further studies in this regard are needed. Taken together, our results indicate a novel circRNA-driven pathway regulating tau phosphorylation and cognitive function, in which circTrpc6 acted as “decoy” sponging miR-485-5p to regulate eIF4E expression, thereby regulating tau phosphorylation and cognitive function. This work broadens our knowledge of how circRNAs function in the brain and offers novel perspectives for developing potential preventive strategies and therapies against cognitive decline associated with neurodegenerative diseases. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (PDF 4448 KB)Supplementary file2 (DOCX 205 KB) Supplementary file1 (PDF 4448 KB) Supplementary file2 (DOCX 205 KB)