Introduction

Depression is a common mental health disorder characterized by persistent feelings of sadness, loss of interest or pleasure in activities, and a range of physical and cognitive symptoms1,2. As a major public health concern, depression affects individuals, families, and society as a whole, increasing the burden on healthcare systems and society in general3,4,5. However, currently available treatments for depression face several challenges and issues, including limited treatment options, severe side effects of medications, high drug tolerance, and a high rate of disease recurrence, in part due to an incomplete understanding of its pathogenesis6,7. Therefore, elucidating the mechanisms underlying depression is imperative to enhance our comprehension of this complex mental disorder, advance diagnosis and therapeutic interventions, and ultimately mitigate the profound impact of depression on individuals and society.

MicroRNAs (miRNAs) are a class of short noncoding RNA molecules that play crucial roles in various physiological processes and contribute to the pathogenesis of some diseases8. Targeting and modulating disease-associated miRNAs holds great potential for the development of innovative therapeutic interventions8,9. Among these miRNAs, miR-143-3p has attracted attention due to its enrichment in the brain and its pivotal roles in the progression of Alzheimer’s disease (AD), as well as acute ischemic stroke. It is upregulated in the serum of patients with mild cognitive impairment associated with AD, acute ischemic stroke patients, and mice models of stroke10,11. The inhibition of miR-143-3p expression has demonstrated potential in alleviating neurological deficits and providing protection against blood-brain barrier damage in mice12. Additionally, silencing miR-143-3p can prevent neuroblastoma cell apoptosis induced by nitric oxide and safeguard neurons from ischemic brain injury in rats13. In addition to the pathogenic role of miR-143-3p in the aforementioned brain disorders, our lab recently observed the differential expression of miR-143-3p during the erasure of fear conditioning memory when compared with the extinction of fear memory14. Abnormalities in fear learning and memory processes have been implicated in the development and maintenance of mental disorders such as depression and anxiety15,16. Individuals with depression may experience heightened fear responses in certain contexts and struggle to overcome learned fear responses. Hence, it is plausible to posit that certain molecules are shared between fear memory processes and the onset of depression. However, whether miR-143-3p represents one such molecule that may contribute to the development of depression remains unknown.

The purpose of this study was to investigate in depth the role of miR-143-3p in the onset of depression and explore the potential mechanisms underlying the miR-143-3p/Lasp1 axis, with the aim to providing a promising direction through which to enhance our understanding of the pathogenesis of depression and developing novel therapeutic approaches.

Results

miR-143-3p was upregulated in the VH of mice treated with CRS or LPS

To investigate the potential involvement of miR-143-3p in stress-induced depressive-like behaviors, we first established a chronic restraint stress (CRS) induced depressive mice model following the experimental procedure (Fig. 1A). Compared to the controls, there was no significant change in the distance traveled by the mice treated with CRS in the open field test (OFT) (Supplementary Fig. 1A). The OFT results demonstrated that CRS significantly reduced the time spent in the center areas compared with that of the CON mice (t = 3.448, P = 0.0039, Supplementary Fig. 1B). However, the sucrose preference index was significantly reduced in the sucrose preference test (SPT) (t = 4.830, P = 0.0003, Supplementary Fig. 1C). In addition, the immobility time was markedly increased in both the tail suspension test (TST) (t = 4.512, P = 0.0005, Supplementary Fig. 1D) and the forced swimming test (FST) (t = 4.817, P = 0.0003, Supplementary Fig. 1E), indicating the successful establishment of the depressive mice model. Next, the levels of miR-143-3p in various depression-related brain regions, including VH, DH and PFC, were assessed using RT-qPCR assay. A repeated measures ANOVA (RM ANOVA) showed a significant main effect of CRS treatment on miR-143-3p expression (CRS: F(1,15) = 39.945, P < 0.0001). The results showed a significant increase in the expression of miR-143-3p in mice treated with CRS in the VH (P = 0.0164, Fig. 1B), DH (P = 0.0048, Fig. 1C), and PFC (P = 0.0361, Fig. 1D) compared with CON mice. To further validate the role of miR-143-3p in depression, another acute depression mice model was employed where mice were intraperitoneally injected with Lipopolysaccharide (LPS), whereafter the expression of miR-143-3p was detected following the experimental procedure (Fig. 1E). Compared to the PBS-treated mice, there was no significant change in the distance traveled by the mice treated with LPS in the OFT (Supplementary Fig. 1F). The OFT results demonstrated that LPS treatment significantly reduced the time spent in the center areas compared with that of the PBS-treated mice (t = 4.251, P = 0.0008, Supplementary Fig. 1G). Similar to previous reports, mice treated with LPS exhibited increased immobility in the TST (t = 3.149, P = 0.0071, Supplementary Fig. 1I) and FST (t = 4.394, P = 0.0006, Supplementary Fig. 1J), while the sucrose preference index was decreased comparison with PBS-treated mice (SPT, t = 7.070, P < 0.0001, Supplementary Fig. 1H), indicating the establishment of an acute depressive mice model. Although an RM ANOVA revealed that LPS treatment did not induce changes in miR-143-3p compared to PBS across different brain regions (LPS: F(1,14) = 2.069, P = 0.174), Bonferroni’s multiple comparisons test showed a significant increase in miR-143 levels in the VH brain region of the LPS treatment (P = 0.0498, Fig. 1F). However, there were no significant changes were observed in the DH (P > 0.9999, Fig. 1G) or PFC (P > 0.9999, Fig. 1H). These results suggest a strong association exists between upregulated miR-143-3p in the VH and depressive-like behaviors, warranting further investigation into this specific relationship.

Fig. 1: miR-143-3p was upregulated in the depressive-like mice models.
figure 1

A Timeline of the experimental procedure for CRS treatment in mice. B miR-143-3p mRNA level in the ventral hippocampus (VH) (n = 7–8). C miR-143-3p mRNA level in the dorsal hippocampus (DH) (n = 7–8). D miR-143-3p mRNA level in the prefrontal cortex (PFC) (n = 7–8). E Timeline of the experimental procedure for LPS treatment in mice. F miR-143-3p mRNA level in the VH (n = 7–8). G miR-143-3p mRNA level in the DH (n = 7–8). H miR-143-3p mRNA level in the PFC (n = 7–8). Data were presented as mean ± SEM. **P < 0.01 versus CON group or versus PBS group.

Full size image

Knockdown of miR-143-3p in the VH alleviated depressive-like behaviors in mice treated with CRS

To explore the specific roles of miR-143-3p in the VH on depressive-like behaviors in mice subjected to CRS, an AAV-miR-143-3p-sponges virus was used to interfere with the expression of miR-143-3p in VH 14 days before CRS treatment, after which depressive-like behaviors were assessed (Fig. 2A). The AAV-miR-143-3p-sponges virus was specifically and highly expressed in VH 14 days after stereotaxic injection (Fig. 2B). The knockdown efficacy of miR-143-3p was evaluated by RT-qPCR and demonstrated a significantly decrease compared with AAV-Ctrl treated mice (t = 3.949, P = 0.0168, Fig. 2C). The OFT results demonstrated that CRS/AAV-Ctrl treatment significantly reduced the time spent in the center areas compared with that of the CON/AAV-Ctrl mice. Moreover, this deficit was significantly reversed by the knockdown of miR-143-3p in the VH (CRS: F(1,31) = 4.811, P = 0.037; AAV microinjection: F(1,31) = 7.209, P = 0.012; interaction: F(1,31) = 94.806, P = 0.011, Fig. 2E). No difference was observed in the total distance traveled among all groups (CRS: F(1,31) = 0.041, P = 0.842; AAV microinjection: F(1,31) = 0.063, P = 0.804; interaction: F(1,31) = 0.000, P = 0.987, Fig. 2D), indicating that the respective treatments alter the state of anxiety without affecting spontaneous activity in the mice.

Fig. 2: Knockdown of miR-143-3p in the VH rescued depressive-like behaviors in CRS mice.
figure 2

A Information on recombinant AAV-143-3p sponges and experimental paradigm for CRS in the present study. B Representative site of virus injection into the VH (Scale bar: 2 mm for the representative image and 100 μm for the enlarged image). C RT-qPCR assay to validate the efficiency of miR-143-3p knockdown (n = 3). D The locomotion of total distance in open field test (OFT) (n = 8). E The time spent in center in OFT (n = 8). F The percentage of sucrose preference in the Sucrose preference test (SPT) (n = 8). G The immobility time in tail suspension test (TST) (n = 8). H The immobility time in forced swimming test (FST) (n = 8). Data were presented as mean ± SEM. *P < 0.05, **P < 0.01 versus CON/AAV-Ctrl group, #P < 0.05, ##P < 0.01 versus CRS/AAV-Ctrl group.

Full size image

Three other behavioral tests (SPT, TST, and FST) were employed to evaluate the effect of miR-143-3p on the depressive-like phenotype. The results from the SPT showed a significant decrease in the sucrose preference index in CRS/AAV-Ctrl mice compared to CON/AAV-Ctrl mice, which was rescued by the knockdown of miR-143-3p (CRS: F(1,31) = 18.874, P < 0.0001; AAV microinjection: F(1,31) = 8.781, P = 0.006; interaction: F(1,31) = 16.150, P < 0.0001, Fig. 2F). Similarly, increased immobility time was observed in mice exposed to CRS in both TST and FST, however, the effects were blocked by the knockdown of miR-143-3p in the VH (TST, CRS: F(1,31) = 12.561, P = 0.001; AAV microinjection: F(1,31) = 9.574, P = 0.004; interaction: F(1,31) = 10.527, P = 0.003; FST, CRS: F(1,31) = 17.301, P < 0.001; AAV microinjection: F(1,31) = 8.434, P = 0.007; interaction: F(1,31) = 13.786, P = 0.001, Fig. 2G, H). These results suggest that miR-143-3p in the VH is involved in the occurrence of depressive-like behaviors induced by chronic stress.

Knockdown of miR-143-3p in the VH alleviated depressive-like behaviors in mice treated with LPS

Next, we validate the involvement of miR-143-3p in the pathogenesis of depressive-like behaviors using another mouse model, specifically the LPS-induced acute depression model. An illustration of the corresponding experimental procedure is provided in Fig. 3A. The OFT results demonstrated that the LPS/AAV-Ctrl treatment significantly reduced the time spent in the center areas compared with PBS/AAV-Ctrl treated mice. Furthermore, this effect was ameliorated by miR-143-3p knockdown in the VH (LPS: F(1,43) = 8.295, P = 0.006; AAV microinjection: F(1,43) = 6.697, P = 0.013; interaction: F(1,43) = 6.046, P = 0.018, Fig. 3C). No difference was observed in the total distance among all groups (LPS: F(1,43) = 0.131, P = 0.719; AAV microinjection: F(1,43) = 0.750, P = 0.392; interaction: F(1,43) = 1.616, P = 0.211, Fig. 3B). The SPT results revealed a significant decrease in the sucrose preference index in mice treated with LPS/AAV-Ctrl compared with those treated with PBS/AAV-Ctrl. More importantly, the knockdown of miR-143-3p in the VH significantly increased the sucrose preference index compared with the LPS/AAV-Ctrl treated mice (LPS: F(1,43) = 19.839, P < 0.0001; AAV microinjection: F(1,43) = 8.463, P = 0.006; interaction: F(1,43) = 19.051, P < 0.0001, Fig. 3D). Additionally, both TST and FST showed that the immobility time was significantly higher in mice receiving LPS/AAV-Ctrl treatment compared than in those receiving PBS/AAV-Ctrl treatment, whereas the effects were significantly ameliorated by miR-143-3p knockdown in the VH (TST, LPS: F(1,43) = 12.293, P = 0.001; AAV microinjection: F(1,43) = 11.807, P = 0.001; interaction: F(1,43) = 13.093, P = 0.001; FST, LPS: F (1,43) = 4.368, P = 0.043; AAV microinjection: F(1,43) = 4.104, P = 0.049; interaction: F(1,43) = 4.311, P = 0.044, Fig. 3E, F). These results suggest that miR-143-3p is involved in depressive-like behaviors induced by both chronic and acute stress.

Fig. 3: Knockdown of miR-143-3p in the VH rescued depressive-like behaviors in LPS mice.
figure 3

A Information on recombinant AAV-143-3p sponges and experimental paradigm for LPS in the present study. B The locomotion of total distance in OFT (n = 11). C The time spent in center in OFT (n = 11). D The percentage of sucrose preference in SPT (n = 11). E The immobility time in TST (n = 11). F The immobility time in FST (n = 11). Data were presented as mean ± SEM. *P < 0.05, **P < 0.01 versus PBS/AAV-Ctrl group, #P < 0.05, ##P < 0.01 versus LPS/AAV-Ctrl group.

Full size image

miR-143-3p participates in the modulation of depressive-like behaviors by regulating synapses

In both patients with depression and animal models of depression, a reduction in dendritic spine density and alterations in their morphology is associated with depression17,18. In this context, the effects of miR-143-3p on synapse changes were investigated using TEM and Golgi staining analysis. We found that the density of the postsynaptic density (PSD) was significantly lower in mice treated with CRS/AAV-Ctrl compared to CON/AAV-Ctrl mice in the VH. However, the knockdown of miR-143-3p significantly attenuated the reduction induced by CRS (CRS: F(1,23) = 42.157, P < 0.001; AAV microinjection: F(1,23) = 0.376, P = 0.547; interaction: F(1,23) = 13.951, P = 0.001, Fig. 4B, C). No difference was observed in the length of active zone among groups (CRS: F(1,23) = 0.428, P = 0.520; AAV microinjection: F(1,23) = 0.157, P = 0.696; interaction: F(1,23) = 0.159, P = 0.695, Fig. 4D). Furthermore, Golgi staining analysis revealed that treatment with CRS/AAV-Ctrl decreased the spine density on apical dendrites compared with the CON/AAV-Ctrl mice, and this change was significantly ameliorated by miR-143-3p knockdown (CRS: F(1,19) = 48.56, P < 0.0001; AAV microinjection: F(1,19) = 59.93, P = 0.000; interaction: F(1,19) = 63.69, P < 0.0001, Fig. 4E, F). Next, a correlation analysis was performed to further explore a possible association between the observed altered behavior and histological changes in the VH. To this end, we evaluated the correlation of the sucrose preference index, which reflects depressive-like behavior, with spine density (Golgi staining). As shown in Fig. 4G, a linear correlation (R = 0.9457) was observed between spine density and sucrose preference, indicating that spine density was positively correlated with sucrose preference in individual mice. These results indicate that chronic stress-induced synaptic loss in the VH is regulated by miR-143-3p, which may be related to the appearance of depressive-like behaviors in mice.

Fig. 4: Knockdown of miR-143-3p in the VH rescued CRS induced synapse alterations.
figure 4

A The experimental paradigm in the present study. B Representative transmission electron microscopy (TEM) images of ultrastructure synapses of the VH region (Scale bar, 200 nm). C The density of synapses of VH (n = 3). D The length of the synaptic active zone of VH (n = 3). E Representative Golgi staining images of VH dendritic spine morphology (Scale bar, 10 μm). F Spine density of dendrites in the VH (n = 5). G Correlation between spine density in the VH and sucrose preference in individual mice (n = 20). Data were presented as mean ± SEM. **P < 0.01 versus CON/AAV-Ctrl group, #P < 0.05, ##P < 0.01 versus CRS/AAV-Ctrl group.

Full size image

As synaptic morphology and density are regulated by synaptic-associated proteins, we further examined the changes of postsynaptic density protein 95 (PSD-95) and synaptophysin (Syn) in the VH. Immunofluorescence analysis revealed a significant decrease in the expression of PSD-95 and Syn following CRS/AAV-Ctrl treatment compared with CON/AAV-Ctrl mice. However, miR-143-3p knockdown in the VH rescued the decrease in PSD-95 and Syn caused by CRS (PSD-95, CRS: F(1,15) = 9.777, P = 0.052; AAV microinjection: F(1,15) = 355.0, P < 0.0001; interaction: F(1,15) = 14.05, P = 0.033; Syn, CRS: F(1,15) = 30.27, P = 0.012; AAV microinjection: F(1,15) = 6.166, P = 0.089; interaction: F(1,15) = 3.434, P = 0.161, Fig. 5B–D). We also investigated the changes in inhibitory synapses by gephyrin immunofluorescence, and found no differences in the mean intensity of gephyrin (CRS: F(1,15) = 2.488, P = 0.141; AAV microinjection: F(1,15) = 0.204, P = 0.660; interaction: F(1,15) = 0.001, P = 0.972; Supplementary Fig. 2A, B). Western blot analysis showed a similar result (PSD-95, CRS: F(1,23) = 6.980, P = 0.016; AAV microinjection: F(1,23) = 7.835, P = 0.011; interaction: F(1,23) = 3.426, P = 0.079; Syn, CRS: F(1,23) = 10.715, P = 0.004; AAV microinjection: F(1,23) = 0.959, P = 0.339; interaction: F(1,23) = 12.279, P = 0.002, Fig. 5E, F). These findings support our previous observations on synaptic density and suggest that miR-143-3p is involved in the regulation of depressive-like behaviors induced by chronic stress via its impact on synaptic-associated components.

Fig. 5: Knockdown of miR-143-3p in the VH rescued CRS induced synaptic-associated proteins decrease.
figure 5

A The experimental paradigm in the present study. B Representative immunofluorescence images of postsynaptic density protein 95 (PSD-95) and synaptophysin (Syn) in the VH (Scale bar, 20 μm). C Quantitative analysis of the PSD-95 mean intensity of immunofluorescence in the VH (n = 4). D Quantitative analysis of the Syn mean intensity of immunofluorescence in the VH (n = 4). E Representative western blotting of PSD-95 and Syn in the VH. F Quantitative analysis of the expression levels of PSD-95 and Syn in the VH (n = 6). Data were presented as mean ± SEM. *P < 0.05, **P < 0.01 versus CON/AAV-Ctrl group, #P < 0.05, ##P < 0.01 versus CRS/AAV-Ctrl group.

Full size image

Lasp1 is a crucial downstream target gene of miR-143-3p involved in the regulation of depressive-like behaviors induced by CRS

MicroRNAs participate in a wide range of physiological processes by exerting their regulatory functions through complementary binding to target sites, thereby suppressing gene expression. The downstream target genes of miR-143-3p were predicted using a bioinformatics website ENCORI/starBase (https://rnasysu.com/encori/). Subsequently, only those target genes predicted by more than three miRNA-mRNA programs and supported by AGO CLIP-seq experiments were selected for further analysis. Among the candidates, LIM and SH3 domain protein 1 (Lasp1) attracted our attention due to its ability to regulate cytoskeletal protein, potentially affecting synaptic structures or functions. To confirm the relationship between Lasp1 and miR-143-3p, we first constructed and transfected the fluorescent recombinant plasmids containing wild-type (WT) or mutant Lasp1 3′ UTR region into 293K cells. After the administration of a miR-143-3p mimic, a significant decrease in the fluorescence intensity was observed in the WT Lasp1 group, while no changes were observed in the mutant Lasp1 group (F(1,19) = 11.90, P < 0.0001 Fig. 6A, B). These results suggest that Lasp1 serves as a potential target of miR-143-3p. In addition, in the VH of mice, CRS treatment significantly decreased the mRNA levels (CRS: F(1,23) = 4.841, P = 0.040; AAV microinjection: F(1,23) = 7.045, P = 0.015; interaction: F(1,23) = 4.143, P = 0.055, Fig. 6C) and protein expression (CRS: F(1,23) = 7.228, P = 0.014; AAV microinjection: F(1,23) = 6.105, P = 0.023; interaction: F(1,23) = 3.143, P = 0.091, Fig. 6D, E) of Lasp1 compared with the CON/AAV-Ctrl group, while the knockdown of miR-143-3p rescued the CRS-induced decrease in Lasp1 expression. We further investigated whether the suppression of Lasp1 could reverse the behavioral consequences caused by the knockdown of miR-143-3p. To this end, AAV-143-3p-sponge (143-sponge) and AAV-shLasp1 (shLasp1) were used to knock down the expression of miR-143-3p and Lasp1, respectively. The AAV-shLasp1 virus was specifically and highly expressed in VH 14 days after stereotaxic injection (Supplementary Fig. 3A). The results demonstrated a significant reduction in the protein expression levels of LASP1 after treatment with shLasp1 (t = 4.717, P = 0.003, Supplementary Fig. 3B, C). The results from the OFT demonstrated that inhibiting miR-143-3p expression in the VH effectively reversed the decreased time spent in the center areas induced by CRS treatment. However, the knockdown of Lasp1 significantly prevented this phenomenon from occurring (F(3,31) = 8.004, P = 0.001, Fig. 7C, D). No difference in the locomotor activity was observed among these groups (F(3,31) = 0.554, P = 0.650, Fig. 7B). Next, depressive-like behaviors were examined using SPT, TST and FST. When the expression of miR-143-3p was knocked down in CRS-treated mice, the decreased sucrose preference index in SPT, as well as increased immobility in TST and FST were all reversed. Furthermore, the knockdown of Lasp1 was found to prevent these changes from happening (SPT, F(3,31) = 13.235, P < 0.0001, TST, F(3,31) = 5.403, P = 0.005, FST, F(3,31) = 7.382, P = 0.001, Fig. 7E–G). These results suggest that miR-143-3p is implicated in the manifestation of depressive-like behaviors in mice via the regulation of Lasp1 expression.

Fig. 6: Lasp1 is a target gene of miR-143-3p.
figure 6

A miR-143-3p and Lasp1 binding site prediction. B Luciferase activity in luciferase reporter gene assay (n = 5). **P < 0.01 versus miR-143-3p mimic control in WT group, ##P < 0.01 versus miR-143-3p mimic in WT group. C Lasp1 mRNA level in the VH (n = 6). D Representative western blotting of LASP1 in the VH. E Quantitative analysis of the expression levels of LASP1 in the VH (n = 6). Data were presented as mean ± SEM. *P < 0.05, **P < 0.01 versus CON/AAV-Ctrl group, #P < 0.05, ##P < 0.01 versus CRS/AAV-Ctrl group.

Full size image
Fig. 7: Knockdown of LASP1 could reverse the antidepressant effects of knockdown miR-143-3p in CRS and LPS mice.
figure 7

A The experimental paradigm for CRS treatment in the present study. B The locomotion of total distance in OFT in CRS model (n = 8). C The time spent in center in OFT in CRS model (n = 8). D The track maps during 10 min in OFT. E The percentage of sucrose preference in SPT in CRS model (n = 8). F The immobility time in TST in CRS model (n = 8). G The immobility time in FST in CRS model (n = 8). H The experimental paradigm for LPS in the present study. I The locomotion of total distance in OFT in LPS model (n = 8). J The time spent in center in OFT in LPS model (n = 8). K The track maps during 10 min in OFT. L The percentage of sucrose preference in SPT in LPS model (n = 8). M The immobility time in TST in LPS model (n = 8). N The immobility time in FST in LPS model (n = 8). Data were presented as mean ± SEM. *P < 0.05, **P < 0.01 versus CRS/AAV-Ctrl group or versus LPS/AAV-Ctrl group, #P < 0.05, ##P < 0.01 versus CRS/143-sponge group, or versus LPS/143-sponge group.

Full size image

The miR-143-3p/Lasp1 axis is involved in the regulation of depressive-like behaviors induced by acute LPS treatment

To further investigate the role of the miR-143-3p/Lasp1 axis in the pathogenesis of depression, an LPS acute treatment-induced depression model was used. Consistent with the results observed in mice treated with CRS, the inhibition of miR-143-3p in the VH effectively reversed the LPS-induced reduction in time spent in the center areas in OFT (F(3,31) = 11.165, P < 0.0001; locomotor: F(3,31) = 0.940, P = 0.435, Fig. 7I–K), as well as decreased sucrose consumption rate in SPT (F(3,31) = 23.238, P < 0.0001, Fig. 7L) and increased immobility in TST (F(3,31) = 4.905, P = 0.007, Fig. 7M) and FST (F(3,31) = 6.412, P = 0.002, Fig. 7N). More importantly, the synchronous knockdown of Lasp1 significantly prevented these aforementioned phenomena from occurring. In summary, the antidepressant effects induced by miR-143-3p knockdown require the assistance of its target gene Lasp1.

The miR-143-3p/Lasp1 axis is involved in the occurrence of depressive-like behaviors by regulating synapse alterations in VH

Lasp1 plays an important role in the synaptic development of rat hippocampal neurons19. On this basis, we used immunofluorescence staining, WB and Golgi staining to analyze the synapse number, morphology, and related proteins in VH. Immunofluorescence analysis showed that CRS treatment led to a decrease in the expression of some synapse-associated proteins, including PSD-95 and Syn. However, the knockdown of miR-143-3p effectively reversed this effect induced by CRS. Furthermore, the simultaneous knockdown of Lasp1 abolished the beneficial effects of miR-143-3p in CRS-treated mice (PSD-95, F(1,15) = 22.69, P < 0.0001, Syn, F(1,15) = 43.42, P < 0.0001, Fig. 8B–D). Consistent results were obtained using WB analysis (PSD-95, F(1,23) = 6.306, P = 0.004; Syn, F(1,23) = 6.845, P = 0.002, Fig. 8E, F). Additionally, Golgi staining demonstrated that CRS treatment resulted in a decrease in spine density of apical dendrites in VH. However, the knockdown of miR-143-3p effectively reversed this change. Additionally, microinjection with shLasp1 was found to prevent the rescue of spine density by miR-143-3p (F(1,19) = 70.67, P < 0.0001, Fig. 8G, H). Next, we evaluated the correlation between the sucrose preference index and spine density and found a significantly positive correlation between sucrose consumption and spine density (R = 0.8653, Fig. 8I). Considering the role played by LASP1 as an actin-binding protein responsible for the formation and maintenance of dendrites and dendritic spines19, the changes in F-actin after the different treatments were examined to fully elucidate the pathway affected by the miR-143-3p/Lasp1 axis. Immunofluorescence staining revealed decreased fluorescence signals of F-actin induced by CRS treatment, while the knockdown of miR-143-3p was found to effectively reverse these changes. Furthermore, this reversible effect was abolished when simultaneously knocking down Lasp1 (F(1,15) = 46.08, P < 0.0001, Fig. 8J, K). In conclusion, these results demonstrate that the miR-143-3p/Lasp1 axis can regulate depressive-like behaviors via the modulation of cytoskeletal proteins, subsequently affecting the synapse number and function in VH.

Fig. 8: Knockdown of LASP1 could reverse the synapse alterations of knockdown miR-143-3p in CRS mice.
figure 8

A The experimental paradigm for CRS treatment in the present study. B Representative immunofluorescence images of PSD-95 and Syn in the VH (Scale bar, 20 μm). C Quantitative analysis of the PSD-95 mean intensity of immunofluorescence in the VH (n = 4). D Quantitative analysis of the Syn mean intensity of immunofluorescence in the VH (n = 4). E Representative western blotting of PSD-95 and Syn in the VH. F Quantitative analysis of the expression levels of PSD-95 and Syn in the VH (n = 6). G Representative Golgi staining images of the VH dendritic spine morphology (Scale bar, 10 μm). H Spine density of dendrites in the VH (n = 5). I Correlation between spine density in the VH and sucrose preference in individual mice (n = 20). J Representative immunofluorescence images of F-actin in the VH (Scale bar, 20 μm). K Quantitative analysis of the F-actin mean intensity of immunofluorescence in the VH (n = 4). Data were presented as mean ± SEM. *P < 0.05, **P < 0.01 versus CRS/AAV-Ctrl group, #P < 0.05, ##P < 0.01 versus CRS/143-sponge group.

Full size image

Discussion

miRNAs are found in a variety of body fluids, including serum and cerebrospinal fluid, as well as specific tissues, making them promising biomarkers for different physiological and pathophysiological conditions. In turn, they show promising applications in achieving the goal of disease diagnosis and prognosis prediction20,21. Importantly, miRNAs that are secreted, particularly from extracellular vesicles (EVs), including exosomes, can facilitate paracrine and endocrine communication between different tissues or subregions, potentially influencing gene expression and the function of distant cells22. Given the ease of obtaining EV donors and the availability of techniques with which to manipulate miRNAs in EV, EV-mediated miRNA therapy holds great promise as a breakthrough therapeutic strategy in pharmacology23. Therefore, identifying specific miRNAs involved in diseases can provide valuable insights for subsequent research and therefore the development of new drugs and diagnosis methods.

In this study, two mouse models were employed to elucidate the role of miR-143-3p in the pathogenesis of depression. Our findings revealed that the expression of miR-143-3p was upregulated in the VH of mice with CRS- or LPS-induced depression. Notably, inhibiting miR-143-3p expression in the VH effectively ameliorated depressive-like behaviors in both CRS- and LPS-treated mice. Furthermore, we validated Lasp1 as one of the downstream target genes of miR-143-3p and demonstrated the involvement of the miR-143-3p/Lasp1 action axis in mediating depressive-like behaviors in mice. Taken together, our results indicate that alterations in synapse numbers in VH were influenced by the miR-143-3p/Lasp1/F-actin axis. Therefore, this axis may represent a critical factor contributing to the pathogenesis of depression.

Previous studies have reported on the dysregulation of several miRNAs, including decreased miR-21 and miRNA-184 and increased miRNA-15a, miR-214-3p, miR-204-5p, miR-483-5p, and miR-124, in depression-related regions of the brain, including the PFC, amygdala, and hippocampus24,25,26,27,28,29. As a widely and highly expressed noncoding RNA in the brain, miR-143-3p plays an important role in the pathogenesis and prognosis of cerebral ischemia10,30. However, its involvement in depression remains unknown. In this study, miR-143-3p was found to be enriched and upregulated in the subregion of the hippocampus, the VH, following psychological stress. MiR-143-3p knockdown in the VH of mice was sufficient to exert an antidepressant effect under conditions of chronic and acute psychological stress. These findings demonstrate previously unknown disease events associated with miR-143-3p, namely involvement in the regulation of synapse numbers and its contribution to the development of depression.

Synaptic changes in response to experience are often input-specific, which may not be explained by widely regulation of translation in the brain but rather depend on local regulatory mechanisms2,31. These mechanisms are necessary for translating external stimulus into plasticity-related changes in neural synapses, including structural remodeling and alterations in synaptic numbers, which are largely driven by the regulation of the dynamics of the actin cytoskeleton through end capping, severing, bundling, and crosslinking actin filaments32,33,34. The LIM and SH3 domain-containing protein LASP1, a member of the nebulin family, is a small multidomain actin-binding protein involved in the assembly and disassembly of actin filaments35,36,37. Previous reports have demonstrated a close association between LASP1 and many neurological disorders, including autism, schizophrenia, and medulloblastoma38,39,40,41. Additionally, although downstream signaling pathways involving LASP1 have been identified, the direct upstream regulatory mechanism of LASP1 remains unclear. In this study, we found that miR-143-3p is constitutively present and upregulated preferentially in the VH under stress conditions. Moreover, increased miR-143-3p was found to directly target and inhibit the LASP1-mediated dynamics of F-acting, leading to altered synaptic plasticity in this region, contributing to depressive-like phenotypes. It is worth that there continues to be a lack of direct evidence proving whether miR-143-3p can regulate functional synaptic transmission, as well as the specific characteristics of the synapses it may influence. As such, electrophysiological studies are needed in the future.

To summarize, this study provides unprecedented evidence on the role of miR-143-3p in the VH in the context of depressive-like phenotypes induced by different stress stimuli. MiR-143-3p was found to trigger the LASP1/F-actin signaling axis and play a crucial role in mediating alterations in synapse density in the VH, resulting in the subsequent development of depressive-like behaviors. These results suggest that targeting the key molecules in the miR-143-3p/LASP1/F-actin pathway may represent a novel therapeutic approach for the treatment of depression, as well as offering valuable insights into the effects of cytoskeleton-related proteins on the pathogenesis of depression.

Methods

Animals

Male C57BL/6 mice (25.0–30.0 g, 10 weeks old) were purchased from Pengyue Laboratory Animal Breeding (Jinan, China). All animals were housed under a 12 h light/dark cycle at 18–22 °C with food and water available ad libitum unless noted otherwise. All animal procedures were in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the institutional animal care and use committee of Shandong University.

Chronic restraint stress (CRS) procedure

Mice were divided randomly into chronic control group (CON) and CRS group. The restraint stress procedure was performed as previously described42,43. Briefly, CRS mice were restrained for 2 h per day for 14 consecutive days in ventilated restrainers, CON mice were handled daily.

Drugs and treatments

Lipopolysaccharide (Cat#L3129, Sigma-Aldrich, USA) was dissolved in phosphate-buffered saline (PBS). The mice were intraperitoneally (i.p.) injected with LPS (0.83 mg/kg) or PBS 24 h before behavioral testing44.

Stereotaxic hippocampal injection

The surgery was performed as described previously45. The mice were anesthetized with isoflurane and placed in the stereotaxic apparatus (RWD Life Science, China). Viruses were injected bilaterally using a 33-gauge injection needle with a 5 μl Hamilton syringe coupled to an automated microinjection pump (RWD Life Science, China); 0.5 μl virus was delivered at a rate of 0.1 μl/min. The VH according to the Mice Brain Atlas, the coordinates were as follows from bregma: ML ±2.5 mm bilateral; AP −2.8 mm; DV −3.0 mm. All AAVs were diluted to 5 × 1012 TU/ml before infusion using the enhanced infection solution (Genechem, China). After injection the needle was kept at the site for 5 min, followed by slow withdrawal. Infection efficiency assays and injection sites were observed a minimum of 14 days after virus injection. Mice with inaccurate injection sites had their corresponding experimental results excluded. Production of AAV-Control-shRNA-mScarlet, AAV-Lasp1-shRNA-mScarlet were performed by OBiO Technology (China). The targeting sequence of Lasp1-shRNA was 5′-GAGGTCATTTTTCGTTGGC-3′. Production of AAV-Scramble-mScarlet, AAV-miR-143-3p-sponges-mScarlet were performed by Genechem (China).

Behavioral tests

The behavioral tests were performed after the 24 h of 14 days of restraint stress or 24 h after LPS injection. Animals were excluded based on the following criteria: in the open field test, animals with a movement distance of less than 1500 cm or greater than 5000 cm were excluded. For animals subjected to viral injection, those with inaccurate injection sites were also excluded.

Open field test (OFT)

OFT was performed as in our previous report to evaluate the spontaneous locomotive activity and anxiety-like behavior42. Each mouse was placed in the center of the open field (40 × 40 × 40 cm) and their free-moving behaviors for 10 min were recorded using a video tracking system (Smart V3.0). The total distance and the crossing lines were analyzed. Animals with a movement distance of less than 1500 cm or greater than 5000 cm were excluded.

Sucrose preference test (SPT)

SPT was assessed for depressive-like behavior based on our previous report with minor modifications42. Mice were accustomed to sucrose solution (1%) for 3 consecutive days prior to the test. On the test day, mice were water-deprived for 22 h, and given two bottles: one containing 1% sucrose solution and the other water for 2 h. Then, the weights of the bottles were measured and fluid consumption was determined. Sucrose preference index was calculated as: sucrose preference (%) = sucrose intake/total fluid intake (sucrose + water) × 100%.

Tail suspension test (TST)

TST was assessed for depressive-like behavior based on a previous report44. Each mouse was hung upside down by its tail, with its head 5 cm from the bottom. The test lasted for 6 min and the activity of mouse was video-recorded for analysis (Smart V3.0). Immobility was defined as the absence of any limb or body movements besides respiration.

Forced swimming test (FST)

FST was also for assessed depressive-like behavior based on our previous report42. Each mouse was placed individually into glass cylinders. The test lasted for 6 min and the activity of mouse was video-recorded for analysis (Smart V3.0). Immobility was defined as the absence of any movement except for the head above water.

Sample collection

Immediately after behavioral tests, animals were sacrificed. Brain sample collection was performed as in our previous report42. In summary, brains were removed after decapitation, and then coronal sections (1 mm thick) were obtained using a mouse brain slicer (RWD Life Science, China). The prefrontal cortex (PFC), dorsal hippocampus (DH) and ventral hippocampus (VH) were punched according to the Mice Brain Atlas using a blunt-end 17 gauge syringe needle (1 mm inner diameter) and frozen in liquid nitrogen, then stored at −80 °C for further PCR, western blot analysis.

Real-time quantitative polymerase chain reaction (RT-qPCR) analysis

Total RNA in PFC, DH and VH was isolated using an Animal Total RNA Isolation kit (ForeGene, China). Complementary DNA (cDNA) was synthesized from total purified RNA using miRNA First Strand cDNA Synthesis (Sangon Biotech, China) or HiScript II QRT SuperMix (Vazyme, China) for qPCR. RT-qPCR reaction was performed using UltraSYBR Mixture (ComWin Biotech, China). Primer sequences used were as follows: mouse LASP1: 5′-TCCCCACTGTGTTTATTAGGGG-3′ and 5′-TTGCAGGTCTCGCAGTGAAA -3′; mouse GAPDH: 5′-AGGTCGGTGTGAACGGATTTG-3′ and 5′-TGTAGCCATGTAGTTGAGGTCA-3′. Specific primers for mature miR-143-3p and U6 were obtained from RiboBio (China). miR-143-3p took U6 as the internal reference, and LASP1 took GAPDH as the internal reference. 2−ΔΔCt showed the relative expression level of each target gene.

Transmission electron microscopy (TEM)

TEM assay was conducted as previously described46. In brief, small VH tissues (1 × 1 × 1 mm) were immediately fixed in 2.5% glutaraldehyde for 2 h at RT and 22 h at 4 °C. The following protocol was carried out at Weiya Electron Microscopy (China). The tissues were embedded and cut into ultrathin sections (~70 nm) along the coronal plane. All TEM images were obtained by JEM-1200EX TEM (JEOL Ltd., Japan) and quantified blindly and independently by investigators without knowledge of the samples. Three brains were collected from each group (n = 3), with two images taken from randomly selected fields of view within the VH for each brain.

Golgi staining

Golgi staining was conducted using the FD Rapid GolgiStain kit (FD NeuroTechnologies, USA) according to the manufacturer’s instruction42. Briefly, brains were immersed in impregnation solution (mixture of A and B solutions from kit) for 2 weeks in the dark. Then, brains were transferred to Solution C and kept in the dark for 3 days. Brains were cut at 100-μm thickness in coronal and collected using VT1200S Vibratome (Leica, Germany) filled with Solution C. After drying, slides were stained with a mixture of Solution D, Solution E and distilled water. Sections were rinsed twice in double distilled water, dehydrated with ethanol, cleared with xylene and coverslipped in resinous mounting medium. Spines were counted under 40× magnification by Microscope (Olympus, Japan).

Immunofluorescence

For histological analysis, mice were anesthetized with isoflurane. Mice were transcardially perfused with 4% paraformaldehyde in PBS. Brains were post-fixed in 4% paraformaldehyde at 4 °C for 24 h. Coronal sections were continuously cut at 30 μm using a vibrating microtome (Leica VT1200, Germany). Immunofluorescence staining was performed as described previously46. The coronal sections were permeabilized with 0.3% Triton X-100 in PBS for 1 h at RT, blocked with 10% NGS for 1 h at RT, incubated with primary antibodies consisting of anti-Synaptophysin (1:100, Cat#ab32127, Abcam, United Kingdom), anti-PSD-95 (1:100, Cat#ab13352, Abcam, United Kingdom), anti-gephyrin (1:500, Cat#147018, SYSY, Germany) for 24 h at 4 °C, incubated with secondary antibodies, stained with DAPI, and then mounted with antifading mounting media. The images were taken on a fluorescence microscope (confocal microscope LSM900, ZEISS, Germany). The mean fluorescence intensity (MFI) was quantified using ImageJ software. The average of at least three slices in each mouse was calculated in the VH and used for statistical analysis. The experimenters were blind to mice groups during the analysis of images.

FITC-phalloidin staining

Sections were washed with PBS two times and then permeabilized in 0.3% Triton X-100 for 30 min at room temperature. After washed three times with PBS, sections were incubated with FITC-phalloidin (1:500, Cat#P5282, Sigma-Aldrich, USA) and 2.5% DAPI at room temperature in the dark for 2–3 h. The images were taken on a fluorescence microscope (confocal microscope LSM900, ZEISS, Germany). The MFI was analyzed using ImageJ software. The experimenters were blind to mice groups during the analysis of images.

Western blotting

The brain tissues were homogenized with QIAGEN TissueLyser II in RIPA buffer containing 1 mM PMSF (Millipore, USA) and protease inhibitor cocktail (Bimake, China). Protein samples were subjected to 10% SDS-PAGE gel and transferred to PVDF membranes. The membrane was blocked with 5% non-fat milk in TBST and incubated with primary antibodies (anti-Tubulin, mouse, 1:5000, Cat#T5168, Sigma-Aldrich; anti-GAPDH, mouse, 1:3000, Cat#60004, Proteintech; anti-PSD-95, rabbit, 1:2000, Cat#2507, CST; anti-Synaptophysin, rabbit, 1:2000, Cat#ab32127, Abcam; anti-LASP1, rabbit, 1:2000, Cat#10515, Proteintech) overnight at 4 °C. After rinsing with TBST, the membrane was incubated with secondary antibodies diluted in TBST for 1 h at RT. Then, the blots were developed with Western Chemiluminescent HRP Substrate (Millipore, USA). Images were taken using the Tanon 5200 Chemiluminescent Imaging System. Protein band densities were quantified by Quantity One software. All unprocessed western bolt images in this study are shown in Supplementary Fig. 4.

Luciferase reporter assay

The luciferase reporter plasmid containing the wild-type (WT) or mutant (MUT) 3′UTR fragment of LASP1 were synthesized by RiboBio (China). The correctly sequenced luciferase reporter plasmids, WT, and Mut were cotransfected, respectively, with miR-143-3p mimic into the HEK-293T cells. After 48 h, the cells were collected and lysed. The luciferase activity was measured using the duo-lite luciferase assay system (Vazyme, China). This experiment was repeated three times.

Statistical analysis

Statistical analyses were performed by GraphPad 8.0 software. Student’s t-test was used to compare two groups; one-way or two-way ANOVA was used for multiple comparisons followed by the Bonferroni test. The levels of miR-143-3p in VH, DH and PFC of treatment between CON and CRS or LPS mice were analyzed by a two-way RM ANOVA followed by the Bonferroni test. Correlations were assessed using the nonparametric Spearman’s rank correlation test. All data are presented as the means ± SEM and P ≤ 0.05 was considered as significant.

Reporting summary

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