miR-135a-5p mediates memory and synaptic impairments via the Rock2/Adducin1 signaling pathway in a mouse model of Alzheimer’s disease

Zheng, Kai; Hu, Fan; Zhou, Yang; Zhang, Juan; Zheng, Jie; Lai, Chuan; Xiong, Wan; Cui, Ke; Hu, Ya-Zhuo; Han, Zhi-Tao; Zhang, Hong-Hong; Chen, Jian-Guo; Man, Heng-Ye; Liu, Dan; Lu, Youming; Zhu, Ling-Qiang

doi:10.1038/s41467-021-22196-y

Download PDF

Article
Open access
Published: 26 March 2021

miR-135a-5p mediates memory and synaptic impairments via the Rock2/Adducin1 signaling pathway in a mouse model of Alzheimer’s disease

Kai Zheng¹^na1,
Fan Hu²^na1,
Yang Zhou²^na1,
Juan Zhang²,
Jie Zheng¹,
Chuan Lai²,
Wan Xiong²,
Ke Cui²,
Ya-Zhuo Hu³,
Zhi-Tao Han³,
Hong-Hong Zhang³,
Jian-Guo Chen⁴,
Heng-Ye Man ORCID: orcid.org/0000-0002-3530-3066⁵,
Dan Liu⁴,
Youming Lu ORCID: orcid.org/0000-0002-6754-4411⁴ &
…
Ling-Qiang Zhu ORCID: orcid.org/0000-0001-9964-9229^2,4

Nature Communications volume 12, Article number: 1903 (2021) Cite this article

10k Accesses
43 Citations
14 Altmetric
Metrics details

Subjects

Abstract

Aberrant regulation of microRNAs (miRNAs) has been implicated in the pathogenesis of Alzheimer’s disease (AD), but most abnormally expressed miRNAs found in AD are not regulated by synaptic activity. Here we report that dysfunction of miR-135a-5p/Rock2/Add1 results in memory/synaptic disorder in a mouse model of AD. miR-135a-5p levels are significantly reduced in excitatory hippocampal neurons of AD model mice. This decrease is tau dependent and mediated by Foxd3. Inhibition of miR-135a-5p leads to synaptic disorder and memory impairments. Furthermore, excess Rock2 levels caused by loss of miR-135a-5p plays an important role in the synaptic disorder of AD via phosphorylation of Ser726 on adducin 1 (Add1). Blocking the phosphorylation of Ser726 on Add1 with a membrane-permeable peptide effectively rescues the memory impairments in AD mice. Taken together, these findings demonstrate that synaptic-related miR-135a-5p mediates synaptic/memory deficits in AD via the Rock2/Add1 signaling pathway, illuminating a potential therapeutic strategy for AD.

Receptor tyrosine kinase ROR1 ameliorates Aβ1–42 induced cytoskeletal instability and is regulated by the miR146a-NEAT1 nexus in Alzheimer’s disease

Article Open access 28 September 2021

SIRT1 deficiency increases O-GlcNAcylation of tau, mediating synaptic tauopathy

Article 25 July 2022

Synaptosome microRNAs regulate synapse functions in Alzheimer’s disease

Article Open access 08 August 2022

Introduction

Alzheimer’s disease (AD) is the most common neurodegenerative disorder in aged people with an increasing incidence with age in the population worldwide¹. The best characterized clinical manifestation of AD is progressive cognitive decline, which appears at the very early stages of AD, in the absence of the two pathological hallmarks: senile plaques and neurofibrillary tangles^2,3. Although the underlying mechanisms involved in the two pathological changes have been well studied, therapeutic strategies targeting these pathologies have failed. To date, there have been no approved drugs to block the decline in cognitive function in AD.

Synapses are considered the fundamental units in the brain, and synaptic activity can stimulate the maturation of mushroom-shaped spines and form new synapses⁴ such that synaptic strength can be adapted to environmental changes and in turn play an important role in learning and memory^5,6. In the brain of an individual with AD, both the disruption of synaptic activity and loss of synapses were found, especially at the early disease stage⁷. Compared with senile plaques and neurofibrillary tangles, synaptic disorders, characterized by loss of synapses and decreased synaptic activity, have higher correlations with cognitive impairments in AD⁸. Thus, understanding the underlying mechanisms of synaptic disorders will be beneficial to the development of therapeutic strategies for AD at an early stage.

MicroRNAs (miRNAs) are a group of small noncoding RNAs ~22 bp in length that silence targeted genes at the posttranscriptional level⁹. miRNAs participate in many physiological processes and pathological pathways, including embryonic development, tumorigenesis, and cardiac diseases^9,10. The fine-tuning ability of miRNAs permits them to regulate the local translation of synaptic-associated proteins at synapses^11,12. In brain tissue or sera from individuals with AD, numerous miRNAs have been identified to be deregulated and proposed to be involved in the formation of senile plaques and neurofibrillary tangles¹³. Recently, the abnormal regulation of miRNAs in the synaptic disorder of AD was also investigated. Some brain-enriched miRNAs, such as miR-124, miR-125b, and miR-132^12,14,15, were aberrantly expressed in AD brains and mediated impairments to synaptic plasticity. However, most of these miRNAs expression levels are not themselves regulated by synaptic activity. Thus, how these miRNAs contribute to synaptic disorders in AD remains elusive.

Here, we first screened for alterations in the levels of miRNAs associated with synaptic plasticity in the hippocampus of AD model mice and found that miR-135a-5p is abnormally downregulated in excitatory pyramidal neurons. The loss of miR-135a-5p is tau dependent and mediated by the reduction of the transcriptional factor Forkhead box D3 (Foxd3). Artificial inhibition of miR-135a-5p results in synaptic disorder and memory impairments via activation of the Rho-associated coiled-coil containing protein kinase 2 (Rock2)/Adducin 1 (Add1) signaling pathway. Overexpressing miR-135a-5p or silencing Rock2 not only rescues dendritic spine maturation but also attenuates synaptic disorders and memory deficits in AD model mice. Finally, a peptide that can block the direct phosphorylation of Add1 exerts similar neuroprotective effects, which provides the basis for developing a novel therapeutic approach for AD.

Results

Alterations in synaptic activity-dependent miRNAs in the hippocampus of AD model mice

To understand the potential role of synaptic activity-associated miRNAs in the pathogenesis of AD, we first performed a systematic analysis of the expression profile of synaptic activity-related miRNAs in the hippocampus of AD model mice at different stages. We focused on 12 miRNAs related to synaptic plasticity as reported in previous literature^16,17 and compared the expression of those miRNAs in the hippocampus of APP/PS1 AD model mice to that of wild-type mice at 9 months and 12 months. We found that at 12 months of age, the AD model mice showed apparent downregulation of miR-135a-5p and upregulation of miR-136-3p, miR-19a-3p, miR-125b-5p, and miR-26a-5p (Fig. 1a). However, at 9 months of age, only a decrease in miR-135a-5p and an increase in miR-125b-5p were detected (Fig. 1b). These data suggested that miR-135a-5p downregulation or miR-125b-5p upregulation may play an important role in the relatively early stages of AD. As the dysfunction of miR-125b-5p in AD has been well documented¹⁴, we focused on the possible role of miR-135a-5p in the synaptic disorder of AD in this work.

**Fig. 1: Alteration of synaptic activity-associated miRNAs in the hippocampus of AD mice.**

The miR-135a-5p expression is abnormally reduced in the hippocampus of AD model mice and the human frontal cortex of patients with AD

We further examined changes in miR-135a-5p levels in the hippocampus of APP/PS1 mice of different ages. We found that the level of miR-135a-5p was not changed at 3 months in APP/PS1 mice but showed a decrease at 6 months (Fig. 1c). We also performed fluorescence in situ hybridization and detected that the decrease in miR-135a-5p expression is prominent in CaMKII-positive neurons (Fig. 1d, e) but not in GAD67-positive neurons (Supplementary Fig. 1a), suggesting that the loss of miR-135a-5p in AD mainly occurs in excitatory neurons. In the cultured primary neurons, the miR-135a-5p level was reduced upon overexpression of human Tau, APP, or PS1 (Fig. 1f, Supplementary Fig. 1b, c). However, in primary neurons from Tau knockout mice, only Tau overexpression could induce the reduction in miR-135a-5p (Supplementary Fig. 1d, e). In line with this, downregulation of miR-135a-5p was also detected in the hippocampus of P301S tau mice at 6 months (Supplementary Fig. 1f). Moreover, the extent of Tau expression was negatively correlated with miR-135a-5p levels in the hippocampus of AD model mice (Fig. 1g, h) but not the wild-type mice (Supplementary Fig. 1g, h). These data suggested that the loss of miR-135a-5p in AD is tau dependent, which leads to suppressed synaptic strength (Supplementary Fig. 1i–k). Finally, in the frontal cortex of patients with AD and AD mice, the level of miR-135a-5p is decreased when compared to age-matched control subjects (Fig. 1i, Supplementary Fig. 1l).

Loss of miR-135a-5p expression is mediated by the reduction of Foxd3

We then investigated why miR-135a-5p is downregulated in AD. Previous studies have shown that RNA degradation is increased in the brains of individuals with AD¹⁸, and we first predicted that the loss of miR-135a-5p is caused by RNA degradation. To test this hypothesis, we examined the stability of miR-135a-5p by blocking new RNA synthesis with actinomycin D¹⁹ and measuring the loss of miR-135a-5p and U6 levels over a 12-h period in primary hippocampal neurons infected with adeno-associated virus (AAV) containing human Tau or empty vector (Fig. 2a, b). We found that the half-life of miR-135a-5p in Tau-overexpressing neurons was identical to that in control virus-infected neurons (Fig. 2b). Our data are reliable because the half-life of miR-132-3p in our system was approximately 4 h (Fig. 2a), which is consistent with a previous study²⁰. Thus, the suppression of miR-135a-5p by Tau overexpression was not caused by dysfunction in RNA degradation. Next, we considered whether the transcription of miR-135a-5p is suppressed in AD. To address this, we measured the levels of primary miR-135a-5p transcript (pri-miR-135a-1/2) in the nuclear fractions of primary hippocampal cultures with Tau overexpression. We found that pri-miR-135a-1 but not pri-miR-135a-2 was reduced in Tau-overexpressing neurons (Fig. 2c). Consistent with that result, pri-miR-135a-1 was reduced in the hippocampus of APP/PS1 AD model mice from 6 months (Fig. 2d, Supplementary Fig. 2a, b). These data suggested that the transcription of pri-miR-135a-1 is inhibited, which possibly leads to the loss of miR-135a-5p in AD. To determine how this transcription was inhibited, we cloned different fragments (1 kb, 2 kb, 3 kb) of the promoter region of pri-miR-135a-1 and inserted them into the pGL3-reporter luciferase construct (Fig. 2e). We found that the construct containing the 1 kb promoter region exhibited the highest luciferase intensity (Fig. 2f). We further analyzed the potential transcriptional factor binding sites within the promoter region of pri-miR-135a-1 (chr9:106153125–106154124) by implementing three independent website tools (RegRNA2.0²¹, PROMO²², and LASAGNA²³) and identified the transcriptional factors Max, Gata-1, Foxd3, Ahr, Arnt, Olf-1, Foxo1, and HNF-3β in the predicted results of all three databases (Supplementary Data 1–3). By examining their expression in the brain (www.uniprot.org and www.biogps.org), we observed that the expression levels of Max, Ahr, Arnt, Olf-1, and HNF-3β are very low in the brain, which rules out their potential roles in regulating the transcription of pri-miR-135a-1 in the brain. We then focused on the remaining three transcriptional factors: Gata-1, Foxd3, and Foxo1. By performing cotransfection of these three factors with the reporter construct, we determined that overexpression of Foxd3 but not of Foxo1 or Gata-1 resulted in dramatically increased luciferase activity in cells transduced with the reported construct containing the pri-miR-135a-1 promoter (Fig. 2g). Consistently, existing chromatin immunoprecipitation (ChIP)-seq data (GSE58407)²⁴ indicated the physical binding of Foxd3 with the pri-miR-135a-1 promoter (Fig. 2h). The results of our ChIP assay demonstrated that Foxd3 could bind with the pri-miR-135a-1 promoter (P-135a), and the binding affinity was attenuated by Tau in vitro (Fig. 2i, j). Moreover, the protein level of Foxd3 was decreased to 53% and 57% in the hippocampus of APP/PS1 mice and P301S tau mice compared to that in wild-type littermates (Fig. 2k, Supplementary Fig. 2c). Upon overexpressing Foxd3 in primary hippocampal neurons transfected with Tau, we discovered that the level of miR-135a-5p was restored (Fig. 2l). Thus, we concluded that the loss of miR-135a-5p induced by Tau is due to the reduction of Foxd3.

**Fig. 2: Loss of miR-135a-5p in AD is caused by the reduction of Foxd3.**

Inhibition of miR-135a-5p expression induces memory impairments in vivo

We then asked whether the loss of miR-135a-5p expression is involved in the pathogenesis of AD. Because the decrease of miR-135a-5p was first detected in the 6-month AD mice, we then performed bilateral intrahippocampal injection of an AAV-miR-135a-5p (1 × 10¹²) sponge (S-135a) to simulate the loss of miR-135a-5p in the hippocampus of 6-month wild-type C57BL/6 mice (Fig. 3a, b). We evaluated spatial memory and synaptic plasticity of these mice one month later and observed that the S-135a-infected mice displayed longer latency from the fourth day during the learning stage in the Morris water maze task (Fig. 3c, d). During the probe trial test, compared with the control mice, S-135a-infected mice showed longer latency for the first crossing, fewer crossings to the platform region, and shorter swimming time in the target quadrant with no changes in swimming speed (Fig. 3e–i, the original results were shown in Supplementary Fig. 3). Furthermore, in the Barnes maze task, S-135a-infected mice displayed fewer nose poke numbers within the target hole, longer latency to the target hole, and more errors made during the search stages for the target hole than control mice (Fig. 3j–n). Finally, in the fear conditioning test, S-135a-infected mice showed a significant reduction in freezing time in the contextual conditioning paradigms, while only a slight difference was detected in tone conditioning (Fig. 3o, p), which relies on the ventral part of the hippocampus^25,26. Together, these results indicated that loss of miR-135a-5p expression induced Alzheimer-like learning and memory deficits in mice.

**Fig. 3: Inhibition of miR-135a-5p induces memory impairments in vivo.**

Inhibition of miR-135a-5p induces synaptic disorder in vivo

Because hippocampal synaptic plasticity is believed to be the basis of learning memory²⁷, we then examined alterations in synaptic plasticity in S-135a-treated mice. We first evaluated hippocampal LTP in the CA3–CA1 circuit and found that the I–O curve in S-135a-infected mice was lower than that in control virus-infected mice (Fig. 4a), indicating that infection of S-135a altered basal synaptic transmission. In addition, S-135a-infected mice displayed impairment in LTP because the normalized fEPSP slope was dramatically suppressed (Fig. 4b, c). Moreover, paired-pulse facilitation (PPF) experiments suggested that S-135a did not impair presynaptic vesicle release (Fig. 4d). Golgi staining revealed that S-135a not only decreased the density of dendritic spines but also reduced the percentage of mushroom-type spines (Fig. 4e–g); furthermore, dendritic complexity, which was evaluated by Sholl analysis and DCI scores, was stunted in response to S-135a (Fig. 4h–j). These data suggested that the loss of miR-135a-5p expression results in synaptic disorder in vivo, which was observed in AD model mice²⁸.

**Fig. 4: Inhibition of miR-135a-5p induces the synaptic disorder in vivo.**

Loss of miR-135a-5p expression results in the activation of the Rock2 signaling pathway

It is known that miRNAs control gene expression by either inhibiting the translation or promoting the degradation of their target mRNAs²⁹. Based on this premise, we sought to identify the direct target of miR-135a-5p that mediates synaptic disorders in AD. We used TargetScan³⁰ and miRDB³¹ databases to predict the potential targets of miR-135a-5p (Supplementary Fig. 4a, Supplementary Data 4 and 5) and chose candidates with a Pct over 0.6 or miRDB scores over 70. Then, we selected 194 common candidate genes and conducted a KEGG analysis (Supplementary Fig. 4b). We found that 15 signal pathways were highly enriched, and the adjusted p value was less than 0.05 (Supplementary Table 1). Among the candidates in the 15 signal pathways, four of them (Rock1, Rock2, Cacna1d, and Pik3r2) were the most attractive because they exist in at least four pathways and are associated with actin cytoskeleton dynamics or calcium signaling. Given that Rock1 protein is mainly distributed in nonneuronal cells³² and showed no change in the hippocampus of 9-month-old APP/PS1 mice (Fig. 5a), we excluded the possible involvement of Rock1 in miR-135a-5p loss-induced memory/synaptic disorder though it plays an important role in actin remodeling and spine density^33,34. We also detected alterations in Rock2, voltage-dependent L-type calcium channel subunit alpha-1D isoform a (encoded by Cacna1d) and p85β (encoded by Pik3r2) in the hippocampus of APP/PS1 mice. We found that only Rock2 expression was dramatically upregulated (Fig. 5a), which indicates that Rock2 is the potential downstream target that mediates the synaptic disorder induced by the loss of miR-135a-5p expression.

**Fig. 5: Loss of miR-135a-5p results in the activation of the Rock2 signal pathway.**

To verify the direct regulation of Rock2 by miR-135a-5p, we constructed the wild-type (WT) 3′UTR region of Rock2 (NM_009072.2), which contains the miR-135a-5p binding site, and a 3′UTR variant with a mutation at the miR-135a-5p binding site; cloned them both in a luciferase reporter gene and cotransfected the reporter vectors into HEK293T cells with either miR-135a-5p mimic or scrambled control (Fig. 5b). We found that the miR-135a-5p mimic suppressed the luciferase intensity in cells transfected with the WT construct but not the mutated construct (Fig. 5c). Meanwhile, by transfecting the mimic and inhibitor of miR-135a-5p into Neuro2a cells, we found that the protein levels of Rock2 were downregulated and upregulated, respectively, while mRNA levels were unaffected (Supplementary Fig. 4c, e). These data suggested the direct posttranscriptional regulation of Rock2 by miR-135a-5p. More importantly, we detected a negative correlation of miR-135a-5p and Rock2 in the hippocampus of APP/PS1 and C57 mice (Supplementary Fig. 4f, g).

Next, we explored the potential downstream effectors of Rock2 activation in mediating synaptic disorders in AD. Considering the above data, Rock2 activation may result in morphological changes in dendrites and dendritic spines, which depend on actin cytoskeleton remodeling. By referring to a previous report³⁵, we performed bioinformatics analysis for the potential substrates of Rock2 and found that the phosphorylation of adducin 1 (Add1) proteins is potential of interest (Supplementary Table 2). By combining this information with a phospho-proteomics study of a brain from an individual with AD³⁶, we selected Add1, Add2, and Crmp2 as candidates for further biological experiments. We found that in the hippocampus of APP/PS1 mice at 9 months, phosphorylation at Thr445 and Ser726 of Add1 and at Thr555 of Crmp2 were dramatically increased (Fig. 5d). However, in cultured hippocampal neurons, application of the miR-135a-5p inhibitor only markedly increased the phosphorylation of Add1 at Thr445 and Ser726, but not the Crmp2 at Thr555 (Fig. 5e), nor the phosphorylation of Limk1/2 (Supplementary Fig. 4h, i), two most important proteins downstream Rock regulating actin stabilization and remodeling^37,38. Furthermore, in the frontal cortex of patients with AD and AD mice, the levels of phosphorylated Ser726-Add1 are increased, which are accompanied by an increase in Rock2 expression (Fig. 5f, Supplementary Fig. 4j, k). Importantly, we detected a relatively strong negative correlation of miR-135a-5p and both Rock2 and phospho-S726-Add1 but a positive correlation of Rock2 and phospho-S726-Add1 (Fig. 5g-i). Thus, the loss of miR-135a-5p expression results in aberrant Rock2 activation and the subsequent deregulation of the cellular cytoskeleton via phosphorylation of Add1 at Ser726.

Overexpression of miR-135a-5p or silencing of Rock2 rescues memory impairments and synaptic disorders in AD model mice

We then asked whether upregulation of miR-135a-5p or silencing of Rock2 could rescue memory impairments and synaptic disorders in AD model mice. We used 8-month-old APP/PS1 mice for the injection, because the memory impairment was apparent in 9 months but not 6 months mice^39,40, and the agomir of miR-135a-5p only lasts up to 6 weeks^41,42. Upon infusion of the agomir of miR-135a-5p or the lentivirus containing Rock2-specific shRNA into the hippocampus of 8-month-old APP/PS1 mice (Supplementary Fig. 5a), we found that both agents were able to not only suppress Rock2 expression and the phosphorylation of Add1 at Ser726 (Fig. 6a, Supplementary Fig. 5a) but also restore the density and maturation of dendritic spines (Fig. 6b–d, Supplementary Fig. 5b–d) and the complexity of neuronal dendrites (Fig. 6e–g, Supplementary Fig. 5e–g). Moreover, the LTP/LTD deficits (Fig. 6h, i, Supplementary Fig. 5h, i, Supplementary Fig. 6) and memory impairments in AD model mice could also be partially rescued (Fig. 6j–m, the original data was shown in Supplementary Fig. 7, Supplementary Fig. 5j–m). These data suggested that restoring the miR-135a-5p/Rock2 signaling pathway rescues memory impairments and synaptic disorders in AD.

**Fig. 6: Overexpression of miR-135a-5p rescues the memory impairments and synaptic disorder in AD mice.**

A peptide blocking the phosphorylation of Add1 rescues memory impairments and synaptic disorders in AD model mice

Finally, we tested whether blocking the phosphorylation of Add1 can rescue memory impairments and synaptic disorders in AD model mice. According to our previous experience⁴³, we generated a series of peptides of different lengths that overlap the phosphorylation sites (Supplementary Fig. 8a). We found that only peptide 3 could attenuate the phosphorylation of Add1 in vitro (Fig. 7a, Supplementary Fig. 8b); thus, we named this peptide siP-Add1 and administered it to 9-month-old APP/PS1 or wild type mice at a dose of 15 mg/kg for 14 consecutive days via i.p. injection (Fig. 7b). We observed that siP-Add1 had no effect on WT mice, including the bodyweight, open field test, dendritic spines and trees, and LTD (Supplementary Fig. 8c-m). Our results showed reduced levels of phosphorylated S726-Add1 in vivo (Fig. 7c). Moreover, siP-Add1 administration not only significantly restored the density of dendritic spines (Fig. 7d–f) and the complexity of neuronal dendrites (Fig. 7g-i) but also ameliorated LTP and LTD deficits (Fig. 7j, k, Supplementary Fig. 8k–m) and memory impairments (Fig. 7l–o) in AD model mice. Thus, blocking Rock2-mediated phosphorylation of Add1 effectively rescues AD-like synaptic and memory impairments in mice.

**Fig. 7: A peptide blocking the phosphorylation of Add1 rescues the memory impairment and synaptic disorder in AD mice.**

Discussion

In this study, we identified that miR-135a-5p, a synaptic-associated miRNA, is abnormally downregulated in AD. Loss of miR-135a-5p expression results in an increase in Rock2 activity and hyperphosphorylation of Add1 at Ser726, which in turn leads to dendritic abnormalities and memory impairments. Blocking the miR-135/Rock2/p-Add1 axis effectively rescued the AD-like synaptic and memory deficits in the APP/PS1 mouse model (Fig. 8).

**Fig. 8: miR-135a-5p/Rock2/Add1 signal mediates the synaptic/memory impairments in AD.**

Recently, we have demonstrated the critical roles of multiple miRNAs in mediating the synaptic disorders of AD^12,44,45, but most of them are not associated with synaptic activity. Here, we first screened the alterations in miRNAs associated with the synaptic activity, and of those miRNAs, miR-135a-5p was of particular interest because it showed the most dramatic change in expression in the hippocampus of AD model mice at 9 months. Although the expression profile of miR-135a-5p in the hippocampal of patients with AD is not clear, some of the high throughout data suggested that miR-135a-5p was downregulated in the gray matter and serum of AD paitents^46,47. miR-135a-5p is upregulated by long-term depression (LTD), a form of synaptic plasticity proposed as the primary cellular substrate of learning/memory⁴⁸. Our data demonstrated that the downregulation of miR-135a-5p in AD is tau dependent because the loss of miR-135a-5p expression in MAPT KO neurons can be induced by tau overexpression but not by either APP or PS1. In line with this, tau pathology (hyperphosphorylation of Tau) in the hippocampus and cortex of APP/PS1 mice occurs no earlier than 6 months^49,50, at which point miR-135a-5p levels began to decrease and the mice displayed significant synaptic disturbance. Soluble tau aggregates inhibit synaptic LTD in vivo⁵¹, and Tau is required for the synaptic toxicity of Aβ⁵². Therefore, tau pathology might play a critical role in mediating miR-135a-5p suppression in AD. Moreover, we also found no negative correlation of Tau with miR-135a-5p in normal mice, suggesting that the synergism of Aβ and Tau⁵³ might be important to mediate the downregulation of miR-135a-5p in AD. In addition, we found that the decrease in miR-135a-5p expression is prominent in excitatory neurons and caused by a decrease in Foxd3 levels in AD. Foxd3 is a well-known transcriptional factor that is important for the development of the vertebrate nervous system, including the determination, migration, and differentiation of neural crest lineages⁵⁴. There are many synaptic-related (or brain-enriched) genes, including many miRNAs⁵⁵ that are transcriptionally regulated by Foxd3⁵⁶. Importantly, a study of genome-wide identification of the binding region of Tau in neurons found that an AG-rich motif was present within Tau-binding regions, and Tau acted as a repressor of genes related to Tau-binding⁵⁷. Meanwhile, we found that two of the variant AG-rich motif (GGAGGGAGAG, GGAGAGAGAG) were present in the promoter region of Foxd3 at −3807 to −3793 in the mouse genome. It is thus possible that Tau protein especially the pathological Tau could bind and repress the transcription of Foxd3, resulting in a reduction of Foxd3 protein. Consistently, the direct transcriptional regulation of miR-135a-5p by Foxd3 was identified in our study based on a series of experiments and previous array data. Thus, the loss of miR-135a-5p expression in AD is Tau dependent and mediated by Foxd3.

Numerous reports have revealed the crucial roles of miRNAs in controlling neuronal morphology and maintaining synaptic plasticity⁵⁸. An image-based miRNome-wide functional miRNA screen in an induced neuronal differential cell model suggested that miR-135a-5p is the potent stimulator for developmental axon growth and branching, cortical neuronal migration, and axon regeneration⁵⁹. Intravitreal injection of miR-135a/b mimic promoted the axon regeneration of retinal ganglion cell (RGC) after optic nerve injury in adult mice. While miR-135a/b sponge further destroyed the axon regeneration of RGC. This was consistent with our result that miR-135a-5p promoted dendritic complexity of hippocampal pyramidal neurons. Administration of miR-135a inhibitors effectively blocked NMDA-induced spine retraction¹⁶. These lines of evidence strongly suggest that miR-135a-5p deregulation is correlated with abnormalities in neuronal morphology. In this study, we found that the loss of miR-135a-5p expression results in a decrease in dendritic complexity and spine maturation, while overexpression of miR-135a-5p in vivo rescues those abnormalities. Moreover, loss of miR-135a-5p expression induces the reduction in LTP and basal transmission in the CA3–CA1 projection. This inhibition seems to be a postsynaptic process because we did not observe any differences in the PPF experiment. However, a recent study reported that overexpression of pre-miR-135a leads to a significant reduction in the number of spines in mouse primary hippocampal neurons⁶⁰. We propose that this discrepancy is due to the different circumstances between the in vitro and in vivo experiments, and the CMV or constitutively active mutant Mef2-vp16 driven overexpression of pre-miR-135a would overwhelm the endogenous miRNA processing machinery^61,62 and result in significant increases of both miR-135a-5p and miR-135a-1/2-3p, which is different from our findings here showing upregulation of miR-135a-5p alone. Similar to our research, Liu et al.⁶³ reported decreased miR-135a-5p in the hippocampus of APP/PS1 mice and suggested that this is responsible for the overproduction of Aβ, the prime contributor to the synaptic disorder in AD. In another study, miR-135a-5p knockdown enhances the frequency of sEPSC events in BLA pyramidal neurons⁶⁴, suggesting a regulatory role of miR-135a-5p in presynaptic function in the amygdala. Here, we demonstrated the critical role of miR-135a-5p in regulating postsynaptic maturation in the hippocampus. Interestingly, targeted knockdown of miR-135a-5p in the amygdala results in anxiety-like behavior in mice⁶⁴. As anxiety is one of the most prevalent psychiatric manifestations in the early stage of AD, administration of miR-135a-5p might be a therapeutic approach for both emotional and cognitive disorders relating to AD. Besides, a series of important studies have demonstrated several other microRNAs, including miR-132-3p, were differentially expressed in response to LTP induction and regulated synaptic proteins relevant to persistent LTP^65,66. Given the important roles of miRNA in the fine-tuning of gene regulation⁶⁷, targeting these synaptic activity-regulated miRNAs might be promising therapeutic approaches for synaptic disorder-related neuropsychiatric diseases.

Our study further demonstrated the important role of Rock2-mediated hyperphosphorylation of Add1 in the synaptic/memory disorder caused by reduced miR-135a-5p expression in AD. It is well known that posttranscriptional regulation by miRNAs has pleiotropic effects, and it is reasonable to speculate that miR-135a-5p participates in synaptic/memory disorders by targeting different target mRNAs. A previous KEGG analysis in the rat hippocampus suggested that the predicted target genes of miR-135a-5p might be important for LTP and the synaptic vesicle cycle of glutamatergic synapses in AD⁶⁸. Here, we pursued Rock2 as a direct target of miR-135a-5p based on a series of bioinformatics analyses and biochemical validation. The in vivo experiments further confirmed the remedial effects of silencing Rock2 on dendritic abnormalities and synaptic/memory deficits. Our study demonstrated that the miR-135a-5p/Rock2 signaling axis is important for actin cytoskeleton dynamics in dendrites and is consistent with a previous in vitro study⁶⁹. Coincidentally, overexpression of miR-135a-5p results in the downregulation of Rock1 and Rock2 expression, which in turn suppresses the in vivo invasion abilities of prostate cancer cells, indicating that this pathway is also implicated in actin cytoskeleton organization in tumors⁷⁰. Moreover, a recent screening series including quantitative proteomic screens and in silico analysis revealed Rock2 as one of six potential main regulators of synaptic and axonal degeneration in vivo⁷¹. Furthermore, Rock2 knockout mice displayed significant deficits in spine properties and synapse density as well as impairments in basal synaptic transmission and hippocampal LTP⁷². Our data provide evidence that overexpression of Rock2 can also impair dendritic morphology and LTP and affect basal transmission. Thus, maintaining a certain level of Rock2 is essential for physiological synaptic plasticity. We also identified the downstream targets of Rock2, such as Add1, that promote synaptic disorders, and disruption of the phosphorylation of Add1 by a specific peptide also rescued dendritic abnormalities and learning memory deficits in AD. It is known that there are many other substrates of Rock2, such as Limk, Mlc, Crmp2⁷³, which mediate a variety of biological processes including cytoskeletal regulation, cellular responses and neuronal degeneration. The application of TAT peptide could specifically block the interaction between Rock2 and Add1 and phosphorylation of S726-Add1 without affecting other substrates^74,75. This result provided evidence of a novel therapeutic approach for this devastating disease. Finally, we speculate that Rock2 might not be the sole downstream target of miR-135a-5p in AD. Our predicted miR-135a-5p targets include the following genes that are related to the disruption of neuronal morphology remodeling and learning memory in AD: glycogen synthase kinase-3 beta (Gsk3b), beta-site APP cleaving enzyme 1 (Bace1), glutamate receptor, ionotropic, NMDA2B (epsilon 2) (Grin2b), very low-density lipoprotein receptor (Vldlr), Golgi-associated, gamma adaptin ear containing, ARF binding protein 1 (Gga1), fermitin family member 2 (Fermt2), NADH:ubiquinone oxidoreductase subunit B9 (Ndufb9), and Ndufa4, mitochondrial complex associated (Ndufa4) (see Supplementary Table 3 for additional putative targets of miR-135a-5p that are associated with AD).

In summary, the results here provide evidence for an important role of the miR-135a-5p/Rock2 signaling pathway in mediating synaptic disorders in AD. As decreased miR-135a-5p levels have been observed in the gray matter of patients with AD⁴⁶ and Rock2 levels are increased in the earliest stages of AD and remain elevated throughout disease progression, our study proposed a therapeutic target for overcoming synaptic and memory deficits in AD.

Methods

Animals and cells

Male APP/PS1 mice were purchased from the Jackson Laboratory (Bar Harbor, ME, NO. 034832). Adult male C57BL/6 mice were purchased from the National Resource Center of Model Mice (Nanjing, China). The Tau-Knockout (Tau−/−) mouse was generated using the CRISPR–Cas9 system by specifically knocking out the sequence of exon 5 (Beijing Biocytogen Co., Ltd., Beijing, China). Genotyping for Tau−/− mice was performed by multiplex polymerase chain reaction using a pair of primers (Forward primer: 5′-GCTACAGTGTGAGTGAGGTTCTAGC-3′; Reverse primer: 5′-GAACCACAGCTGCTTAGGGAAAAC-3′). The genetic background of Tau−/− mice is C57BL/6. All these mice and their nontransgenic littermates were bred in the Experimental Animal Central of Tongji Medical College, Huazhong University of Science and Technology. All these mice were housed under a 12-h light/dark cycle in a temperature (22–24 °C) and humidity (40–60%) controlled room with enough food and water. This study was approved by the Institutional Animal Care and Use Committee of the Huazhong University of Science and Technology. Mouse N2a cells and Human 293T cells were bought from the American Type Culture Collection (ATCC) bank (Manassas, VA, USA).

Plasmids and viruses

miR-135a mimic/inhibitor/agomir and the scrambled control were purchased from RiboBio (Guangzhou, China). Mimic and inhibitor was used for transfection on cells, and agomir was used for stereotaxic injection in mice. 3′UTR of Rock2 was amplified and cloned into psiCHECK-2 (Promega, Madison, WI). The promoter sequence of miR-135a-1 was downloaded from UCSC, and the 3 kb, 2 kb, 1 kb sequence immediately upstream to the transcription start site were cloned into pGL3-Basic vector. The coding sequences for the mRNAs of Foxd3, Gata-1, and Foxo1 were cloned into pcDNA3.1(−) or pcDNA3.1(−)3xflag. Rock2 was cloned into a pcDNA4 vector. AAVs for miR-135a sponge and lentivirus for Rock2 short hairpin RNA was purchased from OBiO Technology (Shanghai, China). The target sequence of Rock2 shRNA was 5′-AACAATAGAGATCTACAAGAT-3′.

Morris water maze

The Morris water maze was performed as previously described⁴⁵. Briefly, Marks of different shapes were pasted around the swimming tank, the temperature of the water was kept between 20 and 22 °C and was made opaque with moderate titanium dioxide powder. The mice were trained for six consecutive days in the afternoon to find a platform hidden that submerged 1 cm below the water surface. Mice were trained from three different quadrants to find the platform for up to 60 s every day. Each mouse remained on the platform for 15 s if they found it or they were guided to the platform and stayed for 15 s. The movements of mice were recorded by a video camera connected to a computer. There was no task on the seventh day and mice got a rest. On the eighth day, the hidden platform was removed, mice were tested to find the platform from the opposite quadrant for 60 s (probe trial). The latency to reach the place of the platform, the percent time spent in the target quadrant, the crossing times to the platform regions, and the swimming velocity was recorded.

Luciferase reporters assay

The pGL3 containing different promoter sequences of miR-135a was cotransfected into HEK293T cells with pRL-TK. And the vector containing 1 kb promoter of miR-135a was cotransfected with Foxd3, Gata-1 or Foxo1 in addition to pRL-TK. The psiCHECK2 containing WT or mutated (Mut) 3′UTR of Rock2 was cotransfected into HEK293T with mimic or scramble of miR-135a. Cells were harvested after 48 h and cell lysates were collected with 1× PLB buffer for firefly and renilla luciferase activities using the dual-luciferase reporter assay system (Promega, USA) according to the manufacturer’s protocol.

Primary hippocampal neuron culture

The isolated embryonic hippocampal neurons were cultured as previously described⁴⁵. Briefly, pregnant mice between 15 and 17 days were anesthetized and hippocampi of fetal mice were isolated under a dissection microscope at 4 °C. The collected hippocampi were cut into small pieces with scissors, digested with 4 ml of 0.125% trypsin for 10–15 min, and stopped by adding 5 ml of the neuronal planting medium (10% fetal bovine serum in DMEM/F12). After centrifugation, resuspending, and trituration with the planting medium, the neurons were filtered with a 200-mesh sieve and plated onto a 60 mm plastic culture dish or 6-well cell culture plate coated with poly d-lysine. The neurons were incubated in the humidified incubator at 37 °C for 4 h with 5% CO₂, and the medium was replaced with a Neurobasal medium supplemented with 2% B27 (maintenance medium). The medium was changed every 3 days for half volume.

Electrophysiological recording

The electrophysiological recording of hippocampal sections was performed as previously described⁴⁴. Briefly, mice were decapitated and the brains were immediately immersed in ice-cold ACSF (125 mM NaCl, 2.0 mM KCl, 2.5 mM CaCl₂, 1.2 mM MgSO₄, 1.2 mM KH₂PO₄, 26 mM NaHCO₃, and 11 mM glucose), which was continuously bubbled with 95% O₂ and 5% CO₂. The mouse brain was cut into 300-μm-thick transverse slices in a horizontal plane with a vibrating microtome (Leica). Slices were preincubated in ACSF at 32 °C for 30 min and then at room temperature for 30 min. A multi-electrophysiological recording setup (MED64; Alpha Med Sciences, Tokyo, Japan) was used to record the field excitatory postsynaptic potential (fEPSP). Slices were placed on 64 microelectrodes in the center of a recording chamber and perfused constantly with ACSF at 32 °C. I/O curves were obtained with an incremental stimulation intensity from 10 to 50 μA. The fEPSPs evoked at CA3-CA1 synapses were recorded from the dendritic layer of CA1 neurons. LTP was induced using a standard high-frequency stimulation paradigm consisting of 2 trains of stimuli at 100 Hz (each for 1 s, pulse duration 200 μs) with 30 s intervals. PPF was accomplished with two consecutive stimuli pulses separated by 50 ms. LTD was induced using low-frequency stimulation at 1 Hz (pulse duration 200 μs) with 900 pulses for 15 min. For mEPSCs recordings, hippocampal neurons were held at −70 mV in voltage-clamp mode and were recorded in 10 s epochs for a total duration of at least 200 s per recording⁷⁶. The data were collected and analyzed with pClamp 10 and ClampFit 10.2, respectively. Statistical analysis was performed by using the data from the last 5 min of recording.

Statistical analysis

All data are presented descriptively as the mean ± SEM and analyzed using SPSS 18.0 and Prism 6.0. The difference between the two groups was assessed using an unpaired Student’s t test. The variance among multiple groups was assessed by a one- or two-way analysis of variance with/without repeated measures followed by a post hoc test. All experiments were repeated three times except those specified, and p < 0.05 was considered statistically significant. All the statistical analysis data was supplied as Supplementary Table 4.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Data availability

The source data for figures and supplementary figures of this paper are available in the source data file. Source data are provided with this paper.

References

Fratiglioni, L. & Qiu, C. Prevention of common neurodegenerative disorders in the elderly. Exp. Gerontol. 44, 46–50 (2009).
Article PubMed Google Scholar
West, M. J., Kawas, C. H., Stewart, W. F., Rudow, G. L. & Troncoso, J. C. Hippocampal neurons in pre-clinical Alzheimer’s disease. Neurobiol. Aging 25, 1205–1212 (2004).
Article CAS PubMed Google Scholar
Jessen, F. et al. A conceptual framework for research on subjective cognitive decline in preclinical Alzheimer’s disease. Alzheimers Dement. 10, 844–852 (2014).
Article PubMed PubMed Central Google Scholar
Saneyoshi, T. et al. Activity-dependent synaptogenesis: regulation by a CaM-kinase kinase/CaM-kinase I/betaPIX signaling complex. Neuron 57, 94–107 (2008).
Article CAS PubMed PubMed Central Google Scholar
Ryan T. J., Roy D. S., Pignatelli M., Arons A., Tonegawa S. Memory. Engram cells retain memory under retrograde amnesia. Science 348, 1007–1013 (2015).
Humeau, Y. & Choquet, D. The next generation of approaches to investigate the link between synaptic plasticity and learning. Nat. Neurosci. 22, 1536–1543 (2019).
Article CAS PubMed Google Scholar
Selkoe, D. J. Alzheimer’s disease is a synaptic failure. Science 298, 789–791 (2002).
Article ADS CAS PubMed Google Scholar
Terry, R. D. et al. Physical basis of cognitive alterations in Alzheimer’s disease: synapse loss is the major correlate of cognitive impairment. Ann. Neurol. 30, 572–580 (1991).
Article CAS PubMed Google Scholar
Winter, J., Jung, S., Keller, S., Gregory, R. I. & Diederichs, S. Many roads to maturity: microRNA biogenesis pathways and their regulation. Nat. Cell Biol. 11, 228–234 (2009).
Article CAS PubMed Google Scholar
Boon, R. A. et al. MicroRNA-34a regulates cardiac ageing and function. Nature 495, 107–110 (2013).
Article ADS CAS PubMed Google Scholar
Banerjee, S., Neveu, P. & Kosik, K. S. A coordinated local translational control point at the synapse involving relief from silencing and MOV10 degradation. Neuron 64, 871–884 (2009).
Article CAS PubMed Google Scholar
Wang, X. et al. Melatonin attenuates scopolamine-induced memory/synaptic disorder by rescuing EPACs/miR-124/Egr1 pathway. Mol. Neurobiol. 47, 373–381 (2013).
Article CAS PubMed Google Scholar
Chopra N. et al. MicroRNA-298 reduces levels of human amyloid-β precursor protein (APP), β-site APP-converting enzyme 1 (BACE1) and specific tau protein moieties. Mol. Psychiatry. https://doi.org/10.1038/s41380-019-0610-2 (2020).
Banzhaf-Strathmann, J. et al. MicroRNA-125b induces tau hyperphosphorylation and cognitive deficits in Alzheimer’s disease. EMBO J. 33, 1667–1680 (2014).
Article CAS PubMed PubMed Central Google Scholar
Edbauer, D. et al. Regulation of synaptic structure and function by FMRP-associated microRNAs miR-125b and miR-132. Neuron 65, 373–384 (2010).
Article CAS PubMed PubMed Central Google Scholar
Hu, Z. et al. miR-191 and miR-135 are required for long-lasting spine remodelling associated with synaptic long-term depression. Nat. Commun. 5, 3263 (2014).
Article ADS PubMed CAS Google Scholar
Gu, Q. H. et al. miR-26a and miR-384-5p are required for LTP maintenance and spine enlargement. Nat. Commun. 6, 6789 (2015).
Article ADS CAS PubMed Google Scholar
Alkallas, R., Fish, L., Goodarzi, H. & Najafabadi, H. S. Inference of RNA decay rate from transcriptional profiling highlights the regulatory programs of Alzheimer’s disease. Nat. Commun. 8, 909 (2017).
Article ADS PubMed PubMed Central CAS Google Scholar
Guo, Y. et al. Characterization of the mammalian miRNA turnover landscape. Nucleic Acids Res. 43, 2326–2341 (2015).
Article ADS CAS PubMed PubMed Central Google Scholar
Sethi, P. & Lukiw, W. J. Micro-RNA abundance and stability in human brain: specific alterations in Alzheimer’s disease temporal lobe neocortex. Neurosci. Lett. 459, 100–104 (2009).
Article CAS PubMed Google Scholar
Chang, T. H. et al. An enhanced computational platform for investigating the roles of regulatory RNA and for identifying functional RNA motifs. BMC Bioinform. 14, S4 (2013).
Article CAS Google Scholar
Messeguer, X. et al. PROMO: detection of known transcription regulatory elements using species-tailored searches. Bioinformatics 18, 333–334 (2002).
Article CAS PubMed Google Scholar
Lee, C. & Huang, C. H. LASAGNA-Search: an integrated web tool for transcription factor binding site search and visualization. Biotechniques 54, 141–153 (2013).
Article CAS PubMed Google Scholar
Krishnakumar, R. et al. FOXD3 regulates pluripotent stem cell potential by simultaneously initiating and repressing enhancer activity. Cell Stem Cell 18, 104–117 (2016).
Article CAS PubMed PubMed Central Google Scholar
Wilmot, J. H., Puhger, K. & Wiltgen, B. J. Acute disruption of the dorsal hippocampus impairs the encoding and retrieval of trace fear memories. Front. Behav. Neurosci. 13, 116 (2019).
Article CAS PubMed PubMed Central Google Scholar
Hunsaker, M. R. & Kesner, R. P. Dissociations across the dorsal-ventral axis of CA3 and CA1 for encoding and retrieval of contextual and auditory-cued fear. Neurobiol. Learn Mem. 89, 61–69 (2008).
Article PubMed Google Scholar
Parra-Damas, A. et al. CRTC1 function during memory encoding is disrupted in neurodegeneration. Biol. Psychiatry 81, 111–123 (2017).
Article CAS PubMed Google Scholar
Viana da SilvaS., et al. Early synaptic deficits in the APP/PS1 mouse model of Alzheimer’s disease involve neuronal adenosine A2A receptors. Nat. Commun. 7, 11915 (2016).
Pasquinelli, A. E. MicroRNAs and their targets: recognition, regulation and an emerging reciprocal relationship. Nat. Rev. Genet. 13, 271–282 (2012).
Article CAS PubMed Google Scholar
Agarwal V., Bell G. W., Nam J. W. & Bartel D. P. Predicting effective microRNA target sites in mammalian mRNAs. Elife 4, e05005 (2015).
Article PubMed Central Google Scholar
Chen, Y. & Wang, X. miRDB: an online database for prediction of functional microRNA targets. Nucleic Acids Res. 48, D127–D131 (2020).
Article CAS PubMed Google Scholar
Hashimoto, R. et al. Distribution of Rho-kinase in the bovine brain. Biochem. Biophys. Res. Commun. 263, 575–579 (1999).
Article CAS PubMed Google Scholar
Penzes, P., Buonanno, A., Passafaro, M., Sala, C. & Sweet, R. A. Developmental vulnerability of synapses and circuits associated with neuropsychiatric disorders. J. Neurochem. 126, 165–182 (2013).
Article CAS PubMed PubMed Central Google Scholar
Penzes, P. & Cahill, M. E. Deconstructing signal transduction pathways that regulate the actin cytoskeleton in dendritic spines. Cytoskeleton (Hoboken) 69, 426–441 (2012).
Article CAS Google Scholar
Amano, M. et al. Kinase-interacting substrate screening is a novel method to identify kinase substrates. J. Cell Biol. 209, 895–912 (2015).
Article CAS PubMed PubMed Central Google Scholar
Xia, Q. et al. Phosphoproteomic analysis of human brain by calcium phosphate precipitation and mass spectrometry. J. Proteome Res. 7, 2845–2851 (2008).
Article CAS PubMed PubMed Central Google Scholar
Lin, T. et al. Rho-ROCK-LIMK-cofilin pathway regulates shear stress activation of sterol regulatory element binding proteins. Circ. Res. 92, 1296–1304 (2003).
Article ADS CAS PubMed Google Scholar
Henderson, B. W. et al. Pharmacologic inhibition of LIMK1 provides dendritic spine resilience against beta-amyloid. Sci. Signal 12, eaaw9318 (2019).
Article PubMed PubMed Central CAS Google Scholar
Chen, S. Q. et al. Age-related changes in brain metabolites and cognitive function in APP/PS1 transgenic mice. Behav. Brain Res. 235, 1–6 (2012).
Article CAS PubMed Google Scholar
Minkeviciene, R. et al. Age-related decrease in stimulated glutamate release and vesicular glutamate transporters in APP/PS1 transgenic and wild-type mice. J. Neurochem. 105, 584–594 (2008).
Article CAS PubMed Google Scholar
Landgraf, D. et al. Genetic disruption of circadian rhythms in the suprachiasmatic nucleus causes helplessness, behavioral despair, and anxiety-like behavior in mice. Biol. Psychiatry 80, 827–835 (2016).
Article PubMed PubMed Central Google Scholar
Li, H. et al. A novel microRNA targeting HDAC5 regulates osteoblast differentiation in mice and contributes to primary osteoporosis in humans. J. Clin. Investig. 119, 3666–3677 (2009).
Article CAS PubMed PubMed Central Google Scholar
Su, Y. et al. MicroRNA-26a/death-associated protein kinase 1 signaling induces synucleinopathy and dopaminergic neuron degeneration in Parkinson’s disease. Biol. Psychiatry 85, 769–781 (2019).
Article CAS PubMed Google Scholar
Wang, X. et al. A novel MicroRNA-124/PTPN1 signal pathway mediates synaptic and memory deficits in Alzheimer’s disease. Biol. Psychiatry 83, 395–405 (2018).
Article CAS PubMed Google Scholar
Liu, D. et al. Targeting the HDAC2/HNF-4A/miR-101b/AMPK pathway rescues tauopathy and dendritic abnormalities in Alzheimer’s disease. Mol. Ther. 25, 752–764 (2017).
Article CAS PubMed PubMed Central Google Scholar
Wang, W. X., Huang, Q., Hu, Y., Stromberg, A. J. & Nelson, P. T. Patterns of microRNA expression in normal and early Alzheimer’s disease human temporal cortex: white matter versus gray matter. Acta Neuropathol. 121, 193–205 (2011).
Article PubMed Google Scholar
Liu, C. G. et al. MicroRNA-135a and -200b, potential Biomarkers for Alzheimers disease, regulate beta secretase and amyloid precursor protein. Brain Res. 1583, 55–64 (2014).
Article ADS CAS PubMed Google Scholar
Ge, Y. et al. Hippocampal long-term depression is required for the consolidation of spatial memory. Proc. Natl Acad. Sci. USA 107, 16697–16702 (2010).
Article ADS CAS PubMed PubMed Central Google Scholar
Kurt, M. A., Davies, D. C., Kidd, M., Duff, K. & Howlett, D. R. Hyperphosphorylated tau and paired helical filament-like structures in the brains of mice carrying mutant amyloid precursor protein and mutant presenilin-1 transgenes. Neurobiol. Dis. 14, 89–97 (2003).
Article CAS PubMed Google Scholar
Inestrosa, N. C. et al. Tetrahydrohyperforin prevents cognitive deficit, Aβ deposition, tau phosphorylation and synaptotoxicity in the APPswe/PSEN1ΔE9 model of Alzheimer’s disease: a possible effect on APP processing. Transl. Psychiatry 1, e20 (2011).
Article CAS PubMed PubMed Central Google Scholar
Ondrejcak, T. et al. Soluble tau aggregates inhibit synaptic long-term depression and amyloid β-facilitated LTD in vivo. Neurobiol. Dis. 127, 582–590 (2019).
Article CAS PubMed Google Scholar
Spires-Jones, T. L. & Hyman, B. T. The intersection of amyloid beta and tau at synapses in Alzheimer’s disease. Neuron 82, 756–771 (2014).
Article CAS PubMed PubMed Central Google Scholar
Busche, M. A. et al. Tau impairs neural circuits, dominating amyloid-beta effects, in Alzheimer models in vivo. Nat. Neurosci. 22, 57–64 (2019).
Article CAS PubMed Google Scholar
Sommer, L. Generation of melanocytes from neural crest cells. Pigment Cell Melanoma Res. 24, 411–421 (2011).
Article CAS PubMed Google Scholar
He, G. Y. et al. The FOXD3/miR-214/MED19 axis suppresses tumour growth and metastasis in human colorectal cancer. Br. J. Cancer 115, 1367–1378 (2016).
Article CAS PubMed PubMed Central Google Scholar
Xiang, J., Wen, F., Zhang, L. & Zhou, Y. FOXD3 inhibits SCN2A gene transcription in intractable epilepsy cell models. Exp. Neurol. 302, 14–21 (2018).
Article CAS PubMed Google Scholar
Benhelli-Mokrani, H. et al. Genome-wide identification of genic and intergenic neuronal DNA regions bound by Tau protein under physiological and stress conditions. Nucleic Acids Res. 46, 11405–11422 (2018).
CAS PubMed PubMed Central Google Scholar
Siegel, G., Saba, R. & Schratt, G. microRNAs in neurons: manifold regulatory roles at the synapse. Curr. Opin. Genet Dev. 21, 491–497 (2011).
Article CAS PubMed Google Scholar
van Battum, E. Y. et al. An image-based miRNA screen identifies miRNA-135s as regulators of CNS axon growth and regeneration by targeting Krüppel-like factor 4. J. Neurosci. 38, 613–630 (2018).
Article PubMed PubMed Central Google Scholar
Vangoor, V. R. et al. Antagonizing increased miR-135a levels at the chronic stage of experimental TLE reduces spontaneous recurrent seizures. J. Neurosci. 39, 5064–5079 (2019).
Article CAS PubMed PubMed Central Google Scholar
Chen, Y., Gao, D. Y. & Huang, L. In vivo delivery of miRNAs for cancer therapy: challenges and strategies. Adv. Drug Deliv. Rev. 81, 128–141 (2015).
Article CAS PubMed Google Scholar
Borel, F., Kay, M. A. & Mueller, C. Recombinant AAV as a platform for translating the therapeutic potential of RNA interference. Mol. Ther. 22, 692–701 (2014).
Article CAS PubMed PubMed Central Google Scholar
Liu, C. G. et al. MicroRNA-135a and -200b, potential Biomarkers for Alzheimer׳s disease, regulate β secretase and amyloid precursor protein. Brain Res. 1583, 55–64 (2014).
Article ADS CAS PubMed Google Scholar
Mannironi, C. et al. miR-135a regulates synaptic transmission and anxiety-like behavior in amygdala. Mol. Neurobiol. 55, 3301–3315 (2018).
Article CAS PubMed Google Scholar
Ryan, B., Logan, B. J., Abraham, W. C. & Williams, J. M. MicroRNAs, miR-23a-3p and miR-151-3p, are regulated in dentate gyrus neuropil following induction of long-term potentiation in vivo. PLoS ONE 12, e0170407 (2017).
Article PubMed PubMed Central CAS Google Scholar
Ryan, B. & Williams, J. M. Novel microRNA revealed by systematic analysis of the microRNA transcriptome in dentate gyrus granule cells. Neurosci. Lett. 707, 132280 (2019).
Article PubMed CAS Google Scholar
Ryan, B., Joilin, G. & Williams, J. M. Plasticity-related microRNA and their potential contribution to the maintenance of long-term potentiation. Front. Mol. Neurosci. 8, 4 (2015).
Article PubMed PubMed Central CAS Google Scholar
Chen, C. L., Liu, H. & Guan, X. Changes in microRNA expression profile in hippocampus during the acquisition and extinction of cocaine-induced conditioned place preference in rats. J. Biomed. Sci. 20, 96 (2013).
Article PubMed PubMed Central CAS Google Scholar
Swanger, S. A., Mattheyses, A. L., Gentry, E. G. & Herskowitz, J. H. ROCK1 and ROCK2 inhibition alters dendritic spine morphology in hippocampal neurons. Cell. Logist. 5, e1133266 (2015).
Article PubMed Google Scholar
Kroiss, A. et al. Androgen-regulated microRNA-135a decreases prostate cancer cell migration and invasion through downregulating ROCK1 and ROCK2. Oncogene 34, 2846–2855 (2015).
Article CAS PubMed Google Scholar
Wishart, T. M. et al. Combining comparative proteomics and molecular genetics uncovers regulators of synaptic and axonal stability and degeneration in vivo. PLoS Genet. 8, e1002936 (2012).
Article CAS PubMed PubMed Central Google Scholar
Zhou, Z., Meng, Y., Asrar, S., Todorovski, Z. & Jia, Z. A critical role of Rho-kinase ROCK2 in the regulation of spine and synaptic function. Neuropharmacology 56, 81–89 (2009).
Article CAS PubMed Google Scholar
Feng, Y., LoGrasso, P. V., Defert, O. & Li, R. Rho Kinase (ROCK) inhibitors and their therapeutic potential. J. Med. Chem. 59, 2269–2300 (2016).
Article CAS PubMed Google Scholar
Shaw, A. E. & Bamburg, J. R. Peptide regulation of cofilin activity in the CNS: a novel therapeutic approach for treatment of multiple neurological disorders. Pharm. Ther. 175, 17–27 (2017).
Article CAS Google Scholar
Ji, S. P. et al. Disruption of PTEN coupling with 5-HT2C receptors suppresses behavioral responses induced by drugs of abuse. Nat. Med. 12, 324–329 (2006).
Article CAS PubMed Google Scholar
Ma, M. et al. A novel pathway regulates social hierarchy via lncRNA AtLAS and postsynaptic synapsin IIb. Cell Res. 30, 105–118 (2020).
Article CAS PubMed PubMed Central Google Scholar

Download references

Acknowledgements

This study is supported partially by the National Key Research and Development Program of China (Grant No. 2019YFE0121200), the National Natural Science Foundation of China (82030032, 31721002, 81771492, 81961128005, 81829002, and 81871108), Top-Notch Young Talents Program of China of 2014, and Academic Frontier Youth Team of Huazhong University of Science and Technology to Dr. Ling-Qiang Zhu.

Author information

These authors contributed equally: Kai Zheng, Fan Hu, Yang Zhou.

Authors and Affiliations

Department of Geriatrics, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
Kai Zheng & Jie Zheng
Department of Pathophysiology, Key Lab of Neurological Disorder of Education Ministry, School of Basic Medicine, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, P. R. China
Fan Hu, Yang Zhou, Juan Zhang, Chuan Lai, Wan Xiong, Ke Cui & Ling-Qiang Zhu
Beijing Key Laboratory of Aging and Geriatrics, National Clinical Research Center for Geriatric Disease, Institute of Geriatrics, Chinese PLA General Hospital and Chinese PLA Medical Academy, Beijing, P. R. China
Ya-Zhuo Hu, Zhi-Tao Han & Hong-Hong Zhang
The Institute of Brain Research, Collaborative Innovation Center for Brain Science, Huazhong University of Science and Technology, Wuhan, P. R. China
Jian-Guo Chen, Dan Liu, Youming Lu & Ling-Qiang Zhu
Department of Biology, Boston University, Boston, MA, USA
Heng-Ye Man

Authors

Kai Zheng
View author publications
You can also search for this author in PubMed Google Scholar
Fan Hu
View author publications
You can also search for this author in PubMed Google Scholar
Yang Zhou
View author publications
You can also search for this author in PubMed Google Scholar
Juan Zhang
View author publications
You can also search for this author in PubMed Google Scholar
Jie Zheng
View author publications
You can also search for this author in PubMed Google Scholar
Chuan Lai
View author publications
You can also search for this author in PubMed Google Scholar
Wan Xiong
View author publications
You can also search for this author in PubMed Google Scholar
Ke Cui
View author publications
You can also search for this author in PubMed Google Scholar
Ya-Zhuo Hu
View author publications
You can also search for this author in PubMed Google Scholar
Zhi-Tao Han
View author publications
You can also search for this author in PubMed Google Scholar
Hong-Hong Zhang
View author publications
You can also search for this author in PubMed Google Scholar
Jian-Guo Chen
View author publications
You can also search for this author in PubMed Google Scholar
Heng-Ye Man
View author publications
You can also search for this author in PubMed Google Scholar
Dan Liu
View author publications
You can also search for this author in PubMed Google Scholar
Youming Lu
View author publications
You can also search for this author in PubMed Google Scholar
Ling-Qiang Zhu
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

L.Q.Z. initiated and designed the study, L.Q.Z., K.Z., D.L., Y.L, and H.M. supervised the study; K.Z., F.H., Y.Z., J.Z., and Jie Z., performed the molecular biological experiments and animal experiments, Y.Z.H., Z.T.H., and H.H.Z. provided the human samples. C.L., W.X., and K.C. performed the electrophysiological recording. F.H., Y.Z., J.Z., J.G.C., Y.L., D.L., and L.Q.Z. analyzed the data. L.Q.Z., K.Z., and F.H. wrote the paper.

Corresponding authors

Correspondence to Youming Lu or Ling-Qiang Zhu.

Ethics declarations

Competing interests

The peptide used to block the direct phosphorylation of Add1 submitted to the Patent Office of the People’s Republic of China by Dr. Ling-Qiang Zhu, Dr. Fan Hu, Dr. Dan Liu, Dr. Jie Zheng, Dr. Yang Zhou (Application No. 202011632141.7). The remaining authors declare no competing interests.

Additional information

Peer review information Nature Communications thanks Cliff Abraham, Chu Chen, Alfredo Miñano-Molina and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Description of Additional Supplementary Files

Supplementary Data

Reporting Summary

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Zheng, K., Hu, F., Zhou, Y. et al. miR-135a-5p mediates memory and synaptic impairments via the Rock2/Adducin1 signaling pathway in a mouse model of Alzheimer’s disease. Nat Commun 12, 1903 (2021). https://doi.org/10.1038/s41467-021-22196-y

Download citation

Received: 25 June 2020
Accepted: 25 February 2021
Published: 26 March 2021
DOI: https://doi.org/10.1038/s41467-021-22196-y

This article is cited by

Short-Term Memory Deficit Associates with miR-153-3p Upregulation in the Hippocampus of Middle-Aged Mice
- Francesca Stabile
- G. Torromino
- A. Rinaldi
Molecular Neurobiology (2024)
MiR-135a-5p regulates window of implantation by suppressing pinopodes development and decidualization of endometrial stromal cells
- Yunan He
- Ying Ju
- Xiaohong Wang
Journal of Assisted Reproduction and Genetics (2024)
CircAKT3 alleviates postoperative cognitive dysfunction by stabilizing the feedback cycle of miR-106a-5p/HDAC4/MEF2C axis in hippocampi of aged mice
- Xuan Wang
- Xiaole Tang
- Ailin Luo
Cellular and Molecular Life Sciences (2024)
N6-methyladenosine-modified circRIMS2 mediates synaptic and memory impairments by activating GluN2B ubiquitination in Alzheimer's disease
- Xiong Wang
- Jiazhao Xie
- Liming Cheng
Translational Neurodegeneration (2023)
Cathepsin B S-nitrosylation promotes ADAR1-mediated editing of its own mRNA transcript via an ADD1/MATR3 regulatory axis
- Zhe Lin
- Shuang Zhao
- Yong Ji
Cell Research (2023)

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.