Rab32-based vesicles coordinate mitochondria and actin for spindle migration and organelle rearrangement in oocyte meiosis

Graphical abstract

Highlights

• •Rab32 depletion or disruption by Rab32^Q83L/T37N led to oocyte maturation defects.
• •Rab32 regulates ROCK-based actin for spindle migration during oocyte asymmetry.
• •Rab32 determines DRP1 phosphorylation for mitochondria in oocytes.
• •Rab32 affects Golgi apparatus for CG migration, and ER/lysosome distribution in oocytes.

Abstract

Introduction

Rab32 is a part of the Rab GTPase family, which is known as the regulator of vesicle transport for an array of cellular functions including endosomal transport, autophagy, generation of melanosomes, phagocytosis and inflammatory processes.

Objective

However, the role of Rab32 in oocyte meiosis is still not well-defined.

Methods

We depleted Rab32 expression by knock down approach, and we also disrupted Rab32 function by exogenous Rab32^Q83L/T37N mRNA injection for mutation.

Results

In our current investigation, we delved into its impacts on the cytoskeleton dynamics and the functionality of organelles during the meiotic maturation process in mouse oocytes. Rab32 expressed during oocyte meiosis and deletion of Rab32 or the expression of exogenous Rab32^Q83L/T37N led to oocyte polar body extrusion defects or symmetric division. We showed that Rab32 was essential for ROCK1-based actin assembly which further led to spindle migration for the asymmetry. Besides, perturbation of Rab32 affected DRP1 phosphorylation for the spatial arrangement and functionality of mitochondria in mouse oocytes. And we found that Rab32 disruption caused the miscarriage of membrane organelles such as Golgi apparatus, ER, lysosome and CGs during oocyte meiosis, leading to ER stress and autophagy.

Conclusions

In summary, our study unravels the critical functions of Rab32 for the interplay between actin and mitochondria, which further facilitates movement of the spindle apparatus and organelles arrangement in mouse oocyte meiotic development.

Introduction

In mammals, the formation of a potentially developing oocyte depends on two vital and fully synchronized maturation processes, which are nuclear maturation and cytoplasmic maturation [1]. During this process, GV (germinal vesicle) oocytes undergo nuclear membrane dissolution and subsequently after the germinal vesicle breakdown, oocytes progress to the metaphase I (MI), metaphase II (MII) to extrude the first polar body. During the MI and MII stages, the spindle migrates with the assistance of kinetic action of actin and eventually takes on cortical localization [2]. Meanwhile, oocyte maturation also involves a variety of cellular processes in the cytoplasm, including changes of mitochondrial distribution and activity [3], endoplasmic reticulum organization [4], calcium ion signaling, and redistribution of other organelles such as Golgi apparatus and lysosomes. These maturation pathways are essential for laying the foundation of oocytes for fertilization and embryonic development.

At the end of MI stage, the chromosome-carrying spindle migrates into the cortex of oocyte to achieve asymmetric division. The cytoplasmic actin network around the spindle is capable of pushing the spindle toward the cortex [5]. Besides the actin nucleators Arp2/3 and Formin family, the members of the GTPase Rho family have become key regulatory factors of actin cytoskeleton [7]. During polar body formation, Cdc42, a member of Rho family, was necessary for the recruitment of N-WASP and actin assembly [8]. RhoA was found to regulate LIMK1/2-cofilin to mediate actin dynamics during mouse oocyte meiosis [9]. Additionally, the Rab family protein Rab35 was related to RhoA and regulates actin assembly through the ROCK-cofilin pathway, further promoting spindle assembly [10]. In addition to participating in the spindle migration, actin network also plays a necessary role in the transport of organelles include mitochondria. Functional mitochondria provides ATP during oocyte development to support cytoplasmic and nuclear maturation [11]. Some GTPases have been reported to be involved in mitochondrial movement by mediating actin filaments. Miro1 and Miro2, the Rho-GTPases, could fine-tune actin-dependent mitochondrial motility and distribution [12]. Moreover, the Roco GTPase family member, Leucine-rich-repeat kinase 2 (LRRK2), was shown to be involved in actin assembly for mitochondrial function in mouse oocyte maturation [13]. Therefore, actin filaments assume essential responsibilities during oocyte meiosis for spindle migration and organelle transport.

Rab GTPases are membrane proteins that specifically label organelles and vest the membrane organelle identity by recruiting transport and trafficking proteins [14]. As with other small GTPase, Rab GTPase function is regulated by guanine nucleotide exchange factor (GEF) and GTPase activating protein (GAP) to switch between inactive GDP-binding and active GTP-binding states [15]. Previous studies have shown that Ras-related GTP-binding proteins (RABs) regulate organelles through actin filaments. Rab7 exhibits a vital role on actin dynamics, and depletion of Rab7 could disturb mitochondrial membrane potential and its distribution [16]. Our previous results demonstrated that the Rab23-Kif17-cargo complex regulated tubulin acetylation for spindle organization and drove actin-mediated spindle migration during meiosis [17]. It was also reported that Rab8a organized Golgi distribution via ROCK-mediated actin assembly during mouse oocyte maturation [18]. Rab32 belongs to Rab superfamily and several cellular processes have been shown to be related with it. Rab32 was involved in autophagy via autophagosome components recycling (ACR) [19]. And Rab32 was also found to regulate ER contact with other organelles such as mitochondria, as a regulator of the mitochondria-associated membrane (MAM) properties [20]. Additionally, it has been suggested that Rab32 is a dual-function protein, which is involved in both mitochondrial anchoring of PKA and mitochondrial dynamics [22]. Although Rab32 is reported to participate in various biological processes in somatic cells, its role in mouse oocyte meiosis is still unclear.

In present study, via the knockdown and overexpression approaches, we confirmed that Rab32 regulates spindle migration and the functions of other organelles through actin-mediated mitochondrial dynamics during mouse oocyte meiotic maturation.

Material and methods

Antibodies and chemicals

Mouse monoclonal anti-Rab32 was purchased from Santa Cruz biotechnology (Cat# sc390178), beta-actin and α-tubulin were purchased from Cell Signal Technology. Fluo-4 AM was purchased from Beyotime Biotechnology (Nantong, China). Mito-Tracker Red CMRos (1:500, Cat# M7521, Invitrogen, Eugene, OR, USA) was used to detect the distribution of mitochondria with an ultimate density of 2  μmol/L. TMRE (1:200, Cat# C2001S) was used to detect the mitochondrial membrane potential. Golgi-tracker Green (1:100, Cat# C1045S-1), ER-tracker Green (1:100, Cat# C1042M-1), and lysosome-tracker Red (1:12000, Cat# C1046) were purchased from Beyotime Biotechnology (Nantong, China). TRITC Phallodin was purchased from SAITONG (Beijing Pusitang Biotechnology Co., LTD, Cat# T10446-300 T). Mouse monoclonal anti-α-tubulin-FITC antibodies were from Sigma-Aldrich Corp (St. Louis, MO, USA, Cat# F-2168-2ML, 1:400). Goat anti-rabbit IgG/Alexa Fluor 488 (Cat# ZF-0511, 1:200) and TRITC-conjugated goat anti-rabbit IgG (Cat# ZF-0316, 1:200) were from Zhongshan Golden Bridge Biotechnology, Co., Ltd. (Beijing, China). Horseradish peroxidase-conjugated goat anti-rabbit/mouse IgG (H + L) antibodies (Cat# CW0102S, 1:2000) were obtained from CWBIO (Beijing, China). All other chemicals were purchased from Sigma (St. Louis, MO, USA), unless otherwise stated.

Oocyte collection and culture

All mouse experiments were conducted in strict compliance with the guidelines set forth by the Animal Research Institution of Nanjing Agricultural University in China, and the experimental protocols were granted approval by the Experimental Animal Research Committee. Oocytes were extracted from 4 to 6-week-old ICR research mice and maintained in M16 medium (Sigma Chemical Co, St. Louis, MO) at a temperature of 37 °C within a 5 % CO2 incubator, with liquid paraffin oil enveloping the culture. Throughout the cultivation period, germinal vesicle (GV)-stage oocytes were harvested at designated time intervals for subsequent analyses.

Plasmid construct and in vitro transcription

Myc-Rab32 vector was generated by Wuhan GeneCreate Biological Engineering Co, Ltd. Total RNA was obtained from 10 ovaries using an RNA Extraction Kit (Takara MiniBEST Universal RNA Extraction Kit), and the cDNA was generated with the PrimeScript RT reagent kit (Takara, Dalian, China). We use following primers to amplify the full-length coding sequence of Rab32 by PCR: F, 5′- GCA GGA TCC CAT CGA TT-3′; R, 5′- GAT GAA CTT CGG CGA GAG-3′. The PCR products were purified and cloned into the Myc-pcDNA3 vector using an In-Fusion HD Cloning Kit, which was purchased from Takara. We used a Fast Mutagenesis System kit (TransGen Biotech, Beijing) and Myc-Rab32 plasmid to synthesize Myc-Rab32 T37N which is a Rab32 GDP-bound mutant plasmid, and Myc-Rab32 Q83L which is a Rab32 GTP-bound mutant plasmid. The mutant primers: Rab32 T37N-F, 5′- CGT GGG TAA GAA CAG CAT CAT CAA GCG CTA CGT GCA-3′; R, 5′- TAG CGC TTG ATG ATG CTG TTC TTA CCC ACG CCG AGC T-3′; Rab32 Q83L-F, 5′- GAT ATC GCG GGA CTG GAA CGG TTT GGC AAC ATG ACT C-3′; R, 5′- TTG CCA AAC CGT TCC AGT CCC GCG ATA TCC CAC AGC T-3′. And we used AvrII to get the linearized constructed plasmids and purified to synthesize Rab32 mRNA. Rab mRNAs were created by HiScribe T7 High Yield RNA Synthesis Kit (NEB), capped with m7G (5′) ppp (5′) G (NEB) and tailed using a Poly(A) Polymerase Tailing Kit (Epicenter), unified using a RNA Clean & Concentrator-25 Kit (Zymo Research) in the end. To ensure the stability of the mRNA, we store the product at − 80 °C.

Microinjection of Rab32 siRNA and mRNA

5–10 picoliters (pl) of Rab32 siRNA were injected into oocytes to knock down (KD) endogenous Rab32 in mouse oocytes. They were then arrested at the GV stage for 20–24 h in M16 medium to facilitate the depletion of Rab32. After that, the oocytes were washed twenty times and transferred to fresh M16 medium for culture at different time points. The Rab32 siRNA sequences are 5′-CCU CUG CCA AGG AUA AUA UTT-3′ and 5′-AUA UUA UCC UUG GCA GAG GTT-3′, diluted with diethyl pyrocarbonate (DEPC)-treated water to a stock concentration of 50 μM (Genepharma, Shanghai, China). As a control group, DNase/RNase-free water was microinjected. For protein expression and mutant experiments, GV oocytes were injected with 5–10 pl of mRNA and cultured in M16 medium containing 5 μM milrinone for 2 h. As a control, DNase/RNase-free water was microinjected.

Real-time quantitative PCR analysis

Real-time quantitative PCR (qPCR) was employed to assess the efficiency of Rab32 mRNA knockdown. Total RNA was isolated from a pool of 50 oocytes using the Dynabeads mRNA DIRECT kit (Invitrogen Dynal AS, Norway). Subsequently, first-strand cDNA synthesis was performed using the PrimeScript RT Master Mix (Takara, Japan) under the following thermal profile: 37 °C for 15 min, followed by 85 °C for 5 s, with a final hold at 4 °C. cDNA fragments of Rab32 were amplified using the following primers: Forward, F, 5′- CCC AAG CTT GAT GGC GGG CGA GGG ACTA-3′; R, 5′- CCG GAA TTC TCA GCA GCA CTG GGA CCT-3′ (GENEWIZ, Suzhou, China). All primers were conducted validated amplification via 'melting curve analysis' (single peak, no primer-dimers). The polymerase chain reaction (PCR) system was executed with the following parameters: an initial denaturation step at 95 °C for 30 s, followed by 40 cycles of quantitative PCR analysis (95 °C for 5 s and 60 °C for 30 s), and a melt-curve analysis consisting of 95 °C for 5 s, 60 °C for 60 s, and a final step at 95 °C for 1 s, with the reaction held at 4 °C. The relative expression levels of Rab32 were calculated by normalizing to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) using the comparative 2^-ΔΔCt method [23].

MtDNA copy number analysis

50 oocytes of every group were collected at MI stage. Subsequently, we employed a mitochondrial DNA extraction kit (Cat# B518749-0050) from Bioengineering to perform the extraction and used qRT-PCR to quantify the mitochondrial copy number.

The β-actin primers were as follows: F, 5′-TGT GAC GTT GAC ATC CGT AA-3′; R, 5′-GCT AGG AGC CAG AGC AGT AA-3′. The MT primers were as follows: F, 5′-CCA ATA CGC CCT TTA ACA AC-3′; R, 5′-GCT AGT GTG AGT GAT AGG GTA G-3′ (Genewiz) [13].

ATP level analysis

To assess the ATP content, the Bioluminescent Cell Detection Kit (Cat# 1003493051) from Sigma was utilized to measure ATP levels in 30 oocytes. The ATP release reagent, which was diluted with Sigma water prior to use, enhances membrane permeability and facilitates the immediate release of ATP from the oocytes. Subsequently, the treated samples were mixed with a luciferase-containing ATP assay mixture, and the bioluminescence intensity was measured using a multimode microplate reader (Tecan Spark) to determine the relative ATP levels.

Immunofluorescent staining and confocal microscopy

A minimum of 30 oocytes were fixed in a 4 % paraformaldehyde solution in PBS (pH 7.4) for 30 min and then permeabilized with 0.5 % Triton-X-100 for 20 min at room temperature. Subsequently, the oocytes were blocked with 1 % bovine serum albumin (BSA) and incubated with mouse monoclonal anti-Rab32 (1:50), acetylation tubulin (1:100), and anti-α-tubulin-FITC (1:100) at 4 °C overnight. Following incubation, the oocytes were transferred to a washing buffer containing 0.1 % Tween 20 and 0.01 % Triton X-100 in PBS and washed three times. Finally, the oocytes were labeled with Alexa Fluor 488 or Alexa Fluor 594 conjugated secondary antibodies (1:100). For F-actin visualization using phalloidin staining, oocytes were incubated with phalloidin-TRITC at room temperature for 1 h. Following this, chromosomes were stained with Hoechst 33342 for 10–20 min. The oocytes were then mounted onto glass slides and examined using a confocal laser scanning microscope equipped with a 40× water immersion lens (Zeiss LSM 700 META, Germany). Additionally, live cell fluorescence staining was utilized to assess the distribution of mitochondria with Mito-Tracker Red CMXRos (Invitrogen, USA), mitochondrial membrane potential (TMRE, Beyotime Biotechnology, Nantong, China), Golgi (Beyotime Biotechnology, Nantong, China), endoplasmic reticulum (Beyotime Biotechnology, Nantong, China), lysosome (Invitrogen, Eugene, OR, USA) and calcium level (Fluor 4.AM, Beyotime Biotechnology, Nantong, China). Oocytes were cultivated to the desired stage of meiosis before being transferred to M16 medium containing fluorescent probes for a 30-minute incubation at 37 °C in a 5 % CO2 atmosphere. They were then promptly imaged using confocal microscopy. For chromosome staining, Hoechst 33342 was applied at a dilution of 1:800. The fluorescence intensity was analyzed using Image J software (US National Institutes of Health, Bethesda, MD, USA). The confocal microscopy parameter settings were standardized for consistent measurements of mitochondrial membrane potential, calcium levels, acetylation of tubulin, and actin fluorescence intensity. In order to count the proportion of spindle migration to the cortex, we fixed the oocytes after 9 h release from Milrinone, and analyzed the ratio in the control group and Rab32 siRNA injection group.

Western blotting

Approximately 200 mouse oocytes were lysed using NuPAGE LDS Sample Buffer and heated for 10 min. The lysates were then stored at −20 °C prior to undergoing 10 % Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE). Electrophoresis was performed at 160 V for 80 min, followed by protein transfer to polyvinylidene fluoride (PVDF) membranes (Millipore, Billerica, MA) at 20 V for 20 min. Membranes were blocked with TBST containing 5 % non-fat milk for 1.5 h, then incubated with mouse monoclonal anti-Rab32 antibody (1:100), rabbit monoclonal anti-α-tubulin antibody (1:1000), mouse monoclonal anti-β-actin antibody (1:1000), anti-ROCK1, anti-p-DRP1, anti-GM130, anti-GRP78, anti-INF2, anti-LC3, and anti-Myc antibodies (1:1000) at 4 °C overnight. Afterward, the membranes were washed five times with TBST for 10 min each. Subsequently, they were incubated with horseradish peroxidase-conjugated goat anti-rabbit/mouse IgG (H + L) antibodies (1:5000) in blocking buffer at room temperature for 1.5 h. Finally, the membranes were developed using a mixture of high-sig and super-sig (1:10) ECL Western blotting substrate (Tanon, China). Band intensity values were quantified using Image J software.

Statistical analysis

For each analysis, a minimum of three biological replicates was conducted. The means and standard error of the mean (SEM) were determined, and the t-test was utilized to assess the statistical significance between the treatment and control groups. The statistical analysis was performed using paired t-tests within GraphPad Prism 5 software. Cohen’s d analysis was executed with R 4.3.2 in appropriate places. Results were considered statistically significant at a P-value of less than 0.05 (denoted by *) and highly significant at a P-value of less than 0.01 (denoted by **).

Results

Expression and subcellular distribution of Rab32 during oocyte meiosis

To explore the role of Rab32 during oocyte maturation, we examined the expression and localization in mouse oocytes at different stages. The western blotting was performed by using germinal vesicle (GV) stage, germinal vesicle breakdown (GVBD), MI and MII-stage oocytes, cultured for 0, 4, 8, 12 h, respectively. The results indicated that Rab32 expressed at all stages of oocyte maturation (Fig. 1A). Meanwhile, immunofluorescence staining results showed that Rab32 broadly distributed in the cytoplasm of oocytes (Fig. 1B). To further explore the role of active and inactive forms of Rab32 in oocytes meiosis, we constructed two Rab32 mutants of GTP-bound Rab32 and GDP-bound Rab32, named Myc-Rab32 Q83L and Myc-Rab32 T37N, respectively. We verified their localization in oocytes and found that it was widely distributed in the cytoplasm (Fig. 1C). The specificity of antibody and exogenous mRNA were also identified (Fig. S1).

Fig. 1.

Expression and subcellular distribution of Rab32. (A) The expression levels of Rab32 at GV, GVBD, MI, MII stages were detected by western blotting in mouse oocytes. (B) Subcellular localization of Rab32 was marked in mouse oocytes. Rab32 was widely distributed in the cytoplasm. Red, Rab32; Green, tubulin; blue, DNA; scale bar, 20 μm. (C) The exogenous Rab32 exhibited a broad distribution within the cytoplasm. Red, Myc-Rab32; Green, tubulin; blue, DNA; scale bar, 20 μm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Disruption of Rab32 function impairs mouse oocyte maturation

To explore the function of Rab32 in the development of oocytes, we utilized the microinjection of siRNA to knock down Rab32 in GV stage oocytes. We first validated the knock down efficiency at mRNA level (Fig. 2A). And the significant decrease of Rab32 protein level was shown in Rab32-KD oocytes compared to control group in Fig. 2B. Following the knockdown of Rab32, the protein expression level was observed to be diminished relative to that of the control group. Subsequently, we found that oocytes lacking Rab32 showed a failure of polar body extrusion and an increased rate of large body production (Fig. 2C). The statistical analysis results confirmed this finding (Fig. 2D, E). We then examined the effects of disruption of Rab32 function. Overexpression of Rab32 Q83L and Rab32 T37N both led to a decrease in oocyte maturation, manifested by the failure of the extrusion polar body in Rab32-mutantion mRNA injection oocytes, with a prominent asymmetric division in the Rab32 T37N-overexpression group (Fig. 2F). And the statistical results confirmed this data (Fig. 2G).

Fig. 2.

Disruption of Rab32 function impairs oocyte maturation. (A) mRNA level of Rab32 was counted after Rab32-siRNA injection. Control (n = 30), Rab32-KD (n = 30), 1 vs. 0.18 ± 0.05. (B) Rab32 protein level was examined by western blot after Rab32 siRNA injection. The band intensities were analyzed by Image J software. 1 vs. 0.43 ± 0.04. (C) Rab32 defects caused the failure of polar body extrusion and proportion of large polar body. scale bar, 20 μm. (D) The statistical analysis of the percentage of polar body extrusion. Control (n = 129), Rab32-KD (n = 141), polar body extrusion: 67.35 ± 2.51 % vs. 46.10 ± 1.01 %, Cohen’s d = 5.54, 95 % CI [1.74, 9.35]. (E) The statistical analysis of large polar body rates. Control (n = 129), Rab32-KD (n = 141), large polar body: 16.03 ± 3.52 % vs. 44.80 ± 2.89 %, Cohen’s d = 5.15, 95 % CI [0.44, 9.86]. (F) Injection of Rab32 Q83L and Rab32 T37N caused the failure of polar body extrusion, accompanied by a slight asymmetric division. scale bar, 20 μm. (G) A decline of polar body extrusion rates after overexpression of GTP- or GDP-bound Rab32. Control (n = 154), Rab32 Q83L (n = 136), Rab32 T37N (n = 142), polar body extrusion: control group: 83.27 ± 4.22 %, vs. Rab32 Q83L overexpression group: 61.13 ± 5.567 %, Cohen’s d = 2.59, 95 % CI [0.49, 5.66] vs. Rab32 T37N overexpression group, 54.93 ± 2.521 %, Cohen’s d = 4.71, 95 % CI [0.31, 9.11]; large polar body: control group: 0.63 ± 0.63 %, n = 154 vs. Rab32 Q83L overexpression group: 4.40 ± 0.67 %, Cohen's d = 3.35, 95 % CI [0.17, 6.86] vs. Rab32 T37N overexpression group: 5.90 ± 0.12 %, Cohen's d = 6.70, 95 % CI [0.87, 12.49]. * P < 0.05, ** P < 0.01.

Rab32 regulates actin-based spindle migration in mouse oocyte meiosis

To investigate the causes of large polar body during oocyte maturation, we tested spindle migration and actin assembly after knocking down Rab32. In Fig. 3A, we found that a large proportion of oocytes after Rab32 knockdown showed central localization of meiotic spindle at MI stage, while most spindles migrated to the cortex in control group. We analyzed the ratio of spindle position relative to cortex between the control oocytes and the Rab32-KD group. The statistical data showed that the rate of spindle migration in Rab32 siRNA injection oocytes was significantly lower than that of control group (Fig. 3A). Since actin network plays a vital part in spindle migration, we examined actin distribution in oocytes. Our results showed that the fluorescence intensity at the cortical region had no significant discrepancy between these groups (Fig. S2A, B). However, cytoplasmic F-actin aggregation around the circumference of spindle was decreased in the Rab32-KD oocytes, and the statistical analysis data confirmed the fluorescence intensity differences (Fig. 3B).

Fig. 3.

Rab32 regulates meiotic spindle migration during oocyte maturation. (A) The spindle stayed in the center of the oocyte after Rab32 knockdown instead of migrating to the cortex. Green, tubulin; blue, DNA; scale bar, 20 μm. The statistical analysis of spindle migration. Control (n = 135), Rab32-KD (n = 137), 67.80 ± 3.31 % vs. 37.57 ± 3.45 %. (B) Deletion of Rab32 decreased the cytoplasmic actin around the spindle. Red, actin; blue, DNA; scale bar, 20 μm. The statistical analysis of cytoplasmic actin fluorescence intensity showed significantly reduced after Rab32 knockdown. Control (n = 135), Rab32-KD (n = 137), 1 vs. 0.71 ± 0.06. (C) Quantitative analysis of the extent of spindle migration indicated that the L/D ratio was significantly higher after injection of Rab32 Q83L and Rab32 T37N. Control (n = 93), Rab32 Q83L (n = 97), Rab32 T37N (n = 93), control group: 1 vs. Rab32 Q83L: 1.06 ± 0.17 vs. Rab32 T37N, 1.17 ± 0.01. Pink, actin; Green, spindle. (D) The cytoplasmic actin enhanced in the Rab32 Q83L mRNA injection group, and Rab32 T37N mRNA injection decreased cytoplasmic actin. Red, actin; green, spindle; blue, DNA; scale bar, 20 μm. The statistical analysis of the cytoplasmic actin fluorescence intensity after overexpressing Rab32. Control (n = 93), Rab32 Q83L (n = 97), Rab32 T37N (n = 93), control group: 1 vs. Rab32 Q83L: 1.15 ± 0.03 vs. Rab32 T37N, 0.87 ± 0.02. (E) Co-IP results showed that Rab32 was correlated with ROCK1. (F) ROCK1 expression levels after overexpression of Rab32 Q83L or Rab32 T37N, and the quantification of western bolt bands. Control group: 1 vs. Rab32 Q83L overexpression group: 1.36 ± 0.04 vs. Rab32T37N overexpression group, 0.53 ± 0.02. (G) The reduction of Rab32 led to decreased localization of ROCK1 around the spindle. Control (n = 15), Rab32-KD (n = 14), 26630 ± 523.90 vs. 24693 ± 546.70. (H) Overexpression of Rab32 Q83L mRNA elevated the intensity of ROCK1 at the periphery of spindle, while Rab32 T37N overexpression diminished ROCK1. Control (n = 15), Rab32 Q83L (n = 14), Rab32 T37N (n = 14), control group: 34582 ± 560.30, vs. Rab32 Q83L: 36042 ± 297.00 vs. Rab32 T37N, 30445 ± 594.80. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001, and ns, no significant difference. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

We also disrupted Rab32 function, and the data indicated that the spindle position was further away from the cortex in the Rab32-mutantation injection oocytes than the control oocytes (Fig. 3C) (The oocyte diameter as D and the distance from one of the spindle poles to the oocyte cortex as L, and the ratio of L/D was used to represent the degree of spindle migration to the cortex). Similarly, the fluorescence intensity of cortical actin filaments showed no change (Fig. S2C, D). However, the cytoplasmic actin fluorescence signals at the periphery of spindle decreased in the Myc-Rab32 T37N injection oocytes compared with the control oocytes, and Myc-Rab32 Q83L mRNA injection contributed to a significant increase on cytoplasmic actin surround the spindle in Fig. 3D. And the fluorescence intensity analysis demonstrated the staining results (Fig. 3D). Utilizing Co-IP assays, we unveiled an interaction between Rab32 and ROCK1 (Fig. 3E). Besides, as shown in Fig. 3F, overexpression of Rab32-GTP (Rab32 Q83L) significantly resulted in an increase in the expression level of ROCK1, while overexpression of Rab32-GDP (Rab32 T37N) led to a decline in the expression level of ROCK1. Accordingly, we assessed the alterations in ROCK1 fluorescence intensity within oocytes derived from the Rab32 knockdown and overexpression, respectively. As displayed in Fig. 3G, the attenuation of Rab32 expression correlated with a notable reduction in the peri spindle localization of ROCK1. Within the overexpression groups, the introduction of mRNA encoding the active Rab32 isoform was associated with an elevation in ROCK1 protein intensity. In contrast, the incorporation of mRNA encoding the inactive Rab32 isoform corresponded to reduction in ROCK1 protein levels (Fig. 3H). These findings were consistent with the results obtained from our Western blot analyses.

Rab32 maintains mitochondrial function in mouse oocyte meiosis

Since actin dynamics take an important part in the transport of mitochondria in mammalian oocytes, we next explored the effects of Rab32 on mitochondria, and our results revealed that mitochondrial signals around the spindle was weakened in the Rab32-KD oocytes (Fig. 4A), as indicated by the statistical data for fluorescence intensity in Fig. 4B. Mitochondrial membrane potential (MMP) is an important index to characterize mitochondrial function. Decreased TMRE accumulation means reduced depolarization or inactive mitochondrial membrane potential. TMRE fluorescent probe was used to detect MMP, which was markedly declined in Rab32-KD oocyte (Fig. 4C), which was confirmed by the fluorescence intensity analysis (Fig. 4D). Subsequently, we evaluated the relative mtDNA copy number and found a marked decrease following Rab32 depletion (Fig. 4E). Next, we examined the ATP synthesis level and the ATP relative content in the Rab32 siRNA-injected oocytes was decreased compared with the control oocytes (Fig. 4F). We also monitored the calcium level in mitochondria by Fluo-4 AM probe, and the calcium level around the spindle decreased after Rab32 siRNA injection (Fig. 4G). The fluorescence intensity analysis corroborated the findings (Fig. 4H). Injecting Myc-Rab32 Q83L mRNA resulted in an abnormal aggregation of mitochondria, while Myc-Rab32 T37N mRNA injection caused a decrease of mitochondria clustered around the spindle (Fig. 4I). Statistical analysis showed that the aberrant rate of mitochondrial distribution in Myc-Rab32 Q83L and Myc-Rab32 T37N mRNA injection groups was remarkably higher than that in control group (Fig. 4J). Then we detected the expression level of p-DRP1, and it showed that overexpression of Rab32 mutants declined p-DRP1 expression level (Fig. 4K).

Fig. 4.

Rab32 maintains mitochondrial function in mouse oocyte meiosis. (A) Injection Rab32 siRNA caused aberrant mitochondrial distribution. Red, mitochondria; blue, DNA; scale bar, 20 μm. (B) The fluorescence intensity analysis of mitochondria showed a significant descend after Rab32 knocking down. Control (n = 73), Rab32-KD (n = 69), 1, vs. 0.55 ± 0.08. (C) TMRE was used to mark the mitochondrial membrane potential (MMP) showing an obvious decline in Rab32 depletion. Red, TMRE; blue, DNA; scale bar, 20 μm. (D) The statistical analysis of MMP fluorescence intensity showing with a significant decrease on the Rab32-KD group. Control (n = 75), Rab32-KD (n = 76), 1 vs. 0.60 ± 0.05. (E) The relative content of mtDNA copy number in the Rab32 siRNA-injected group was apparent reduced. Control (n = 200), Rab32-KD (n = 200), 1 vs. 0.80 ± 0.03. (F) The relative content of ATP level in the Rab32 siRNA-injected group was significantly decreased. Control (n = 90), Rab32-KD (n = 90), 1 vs. 0.58 ± 0.05. (G) Fluo-4 AM was used to label the calcium level in mitochondria showing with a decline of calcium level in Rab32-KD oocytes. Green, calcium; blue, DNA; scale bar, 20 μm. (H) The statistical analysis of the calcium fluorescence intensity, and Rab32 knockdown caused significantly weakening. Control (n = 75), Rab32-KD (n = 78), 1 vs. 0.55 ± 0.10. (I) Compared with the control group, injection of GTP-bound Rab32 induced mitochondria remarkably clustered around the spindle, and injection of GDP-bound Rab32 induced mitochondria dispersed to cytoplasm. Red, mitochondria; blue, DNA; scale bar, 20 μm. (J) The statistical analysis of the aberrant mitochondrial distribution, and it elevated sharply in the overexpression groups. Control (n = 72), Rab32 Q83L (n = 68), Rab32 T37N (n = 69), control group: 25.10 ± 3.69 % vs. Rab32 Q83L: 51.10 ± 3.19 % vs. Rab32 T37N, 56.00 ± 1.10 %. (K) Western blot results showed p-DRP1 expression levels after overexpression of Rab32 Q83L or Rab32T37N, and the quantification of western bolt bands. Control group: 1 vs. Rab32 Q83L overexpression group: 0.79 ± 0.02 vs. Rab32 T37N overexpression group, 0.69 ± 0.08. * P < 0.05, ** P < 0.01, *** P < 0.001. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Rab32 regulates Golgi-based cortical granule transport during oocyte maturation

As the actin filament network undertakes the function of short distance transport during oocyte maturation, we then observed the distribution of other organelles in cytoplasm. Upon Rab32 siRNA injection, we observed a notable reduction in Golgi aggregation at the spindle periphery in Rab32-KD oocytes (Fig. 5A). The aberrant distribution rate of Golgi increased after Rab32 siRNA injection, and the fluorescence intensity of the Golgi apparatus declined in Rab32-KD group (Fig. 5B). Compared to control oocytes, GTP-bound Rab32 (Myc-Rab32 Q83L) injection gave rise to the Golgi apparatus to spread throughout the cytoplasm, however, GDP-bound Rab32 (Myc-Rab32 T37N) injection has no significant effect on the distribution of Golgi apparatus (Fig. 5C), which was shown by the statistical data (Fig. 5D). Besides, we found that overexpression of Rab32 Q83L or Rab32 T37N both led to the down-regulation of GM130, showing with decreased density, respectively (Fig. 5E).

Fig. 5.

Rab32 regulates Golgi-based cortical granule transport during oocyte maturation. (A) The accumulation of Golgi around the spindle was reduced with an abnormal distribution. Green, Golgi; blue, DNA; scale bar, 20 μm. (B) The statistical analysis of Golgi abnormal distribution, and Rab32-KD significantly disturbed Golgi localization and the relative fluorescence intensity analysis of Golgi. Control (n = 87), Rab32-KD (n = 90), abnormal Golgi: 28.33 ± 1.67 % vs. 58.57 ± 8.00 %; fluorescence intensity: 1 vs. 0.83 ± 0.02. (C) Injection of Rab32 Q83L led to the evenly distribution of Golgi into cytoplasm, while injection of Rab32 T37N resulted in slight agglutination of Golgi in the cytoplasm. Green, Golgi; blue, DNA; scale bar, 20 μm. (D) The statistical analysis of Golgi abnormal distribution. It demonstrated injection of Rab32 Q83L disturbed Golgi localization, and injection of Rab32 T37N had no significant effects on Golgi distribution. Control (n = 42), Rab32 Q83L (n = 39), Rab32 T37N (n = 44), control group: 27.08 ± 2.08 % vs. Rab32 Q83L: 63.57 ± 2.85 % vs. Rab32 T37N, 34.20 ± 2.27 %, no significance. (E) Western blot results showed GM130 expression levels after overexpression of Rab32 Q83L or Rab32T37N and the quantification of western bolt results. Control group: 1 vs. Rab32 Q83L: 0.73 ± 0.07 vs. Rab32 T37N, 0.74 ± 0.02. (F) Injection of Rab32 siRNA led to errors in the transport of cortical granules (CGs). Green, CGs; blue, DNA; scale bar, 20 μm. (G) The graph showed a failure of CGs migration to the cortex and the related statistical analysis. Control (n = 73), Rab32-KD (n = 79), 72.90 ± 0.20 % vs. 49.00 ± 4.06 %. (H) The cortical granules failed to migrated properly to the cortex and form an empty area after injection of GTP- and/or GDP-bound Rab32 mRNA. Green, CGs; blue, DNA; scale bar, 20 μm. (I) The statistical analysis of CGs aberrant migration rate. Control (n = 54), Rab32 Q83L (n = 56), Rab32 T37N (n = 45), control group: 27.23 ± 2.11 % vs. Rab32 Q83L: 45.73 ± 2.40 % vs. Rab32 T37N, 68.38 ± 3.25 %. * P < 0.05, ** P < 0.01, **** P < 0.0001, and ns, no significant difference. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

The migration of cortical granules (CGs) is related to the Golgi-based vesicle transport, and CG free domain is related to actin-based spindle migration. In most of the Rab32-KD oocytes, there was no CG free domain at the cortex (Fig. 5F). The statistical data also showed that CGs migration failed in large proportion of the Rab32 siRNA injection oocytes (Fig. 5G). Analogously, fluorescence-based observations highlighted abnormal migration of CGs in response to Myc-Rab32 Q83L or Myc-Rab32 T37N mRNA injection, diverging from the control group (Fig. 5H), which was shown by the statistical analysis (Fig. 5I).

Rab32 regulates ER and lysosome distribution during oocyte meiosis

Thereafter, we scrutinized the endoplasmic reticulum (ER) staining signal in oocytes at the MI stage. Intriguingly, the ER in oocytes subjected to Rab32 siRNA injection exhibited a dispersed cytoplasmic distribution, rather than the typical clustering around the spindle observed in control oocytes (Fig. 6A). In the Rab32-KD group, the incidence of aberrant distribution patterns was significantly accentuated when juxtaposed with the control group, concurrently, the fluorescence intensity emanating from the ER was conspicuously attenuated in contrast to the control group (Fig. 6B). In the Myc-Rab32 Q83L mRNA injection group, the ER had large clumps in the cytoplasm, while Myc-Rab32 T37N mRNA injection reduced the distribution area of the ER (Fig. 6C). Overexpression of Rab32 increased the abnormal ratio of ER (Fig. 6D). Then we examined the expression levels of GRP78 and LC3 respectively (Fig. 6E). The findings revealed that overexpression of Rab32 Q83L led to the down-regulation of GRP78, and overexpression of Rab32 T37N could increase the expression levels of GRP78 (Fig. 6F). For LC3, there were no significant differences observed between groups (Fig. 6F).

Fig. 6.

Rab32 regulates ER and lysosome distribution during oocyte meiosis. (A) The ER around spindle declined after Rab32 knocking down, and floated to the cytoplasm. Green, ER; blue, DNA; scale bar, 20 μm. (B) The statistical analysis of ER abnormal distribution, and Rab32-KD significantly disturbed ER localization. And the relative fluorescence intensity analysis of ER showed decreased after Rab32 siRNA injection. Control (n = 87), Rab32-KD (n = 90), abnormal ER: 18.90 ± 1.10 % vs. 60.40 ± 3.53 %; fluorescence intensity:1 vs. 0.46 ± 0.07. (C) Overexpression of GTP-bound Rab32 induced aberrant aggregation of ER in cytoplasmic, and overexpression of GDP-bound Rab32 caused contracts area of ER around the spindle. Green, ER; blue, DNA; scale bar, 20 μm. (D) The statistical analysis of ER abnormal distribution, showing with higher abnormal rates. Control (n = 60), Rab32 Q83L (n = 55), Rab32 T37N (n = 57), control group: 19.40 ± 3.96 % vs. Rab32 Q83L: 47.00 ± 3.57 % vs. Rab32 T37N: 50.00 ± 2.59 %. (E) Western blot results showed LC3 and GRP78 expression levels after overexpression of Rab32 Q83L or Rab32 T37N. (F) Quantification analysis of western bolt results. GRP78: control group: 1 vs. Rab32 Q83L: 0.57 ± 0.02 vs. Rab32 T37N: 1.67 ± 0.10; LC3: control group: 1 vs. Rab32 Q83L: 1.28 ± 0.13 vs. Rab32T37N, 1.52 ± 0.40, no significance. (G) Rab32-KD induced an aggregated lysosome mass in cytoplasm. Red, lysosome; blue, DNA; scale bar, 20 μm. (H) The statistical analysis of lysosome abnormal distribution. (I) Overexpression of Rab32 resulted in excessive agglutination and/or decreased lysosomal vesicles in oocyte. Control (n = 57), Rab32-KD (n = 52), 19.83 ± 1.63 % vs. 73.43 ± 5.50 %. Red, lysosome; blue, DNA; scale bar, 20 μm. (J) The statistical analysis of lysosome aberrant distribution. Control (n = 59), Rab32 Q83L (n = 56), Rab32 T37N (n = 57), control group: 27.13 ± 3.13 % vs. Rab32 Q83L: 59.87 ± 3.14 % vs. Rab32 T37N, 53.63 ± 2.05 %. * P < 0.05, ** P < 0.01, *** P < 0.001, and ns, no significant difference. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

We also detected lysosome by lysosome-tracker, and there was obviously abnormal distribution of lysosome after Rab32 siRNA injection, and instead of small vesicles being uniformly distributed in the cytoplasm, they were excessively clustered (Fig. 6G). Quantitative assessment via statistical methods highlighted a pronounced augmentation in the frequency of atypical distribution patterns, diverging significantly from the normative values observed within the control group (Fig. 6H). Besides, both Myc-Rab32 Q83L and Myc-Rab32 T37N mRNA expression caused lysosome vesicles almost disappear (Myc-Rab32 Q83L overexpression group) or erupt (Myc-Rab32 T37N overexpression group) (Fig. 6I). The overexpression groups exhibited a conspicuously heightened frequency of lysosome mislocalization, a deviation that starkly contrasted with the baseline established by the control group (Fig. 6J).

Discussion

In present study, we revealed the essential roles of Rab32 for regulating mitochondrial distribution and actin assembly, and further participated in mediating the organelles by energy production and cytoskeleton organization during oocyte meiotic maturation (Fig. 7).

Fig. 7.

Roles of Rab32 during mouse oocyte maturation. Schematic representation of Rab32 regulating the crosstalk between mitochondria and actin for spindle migration and organelles rearrangement during oocyte meiotic maturation.

We found that Rab32 widely distributed in the cytoplasm of oocytes, and it also accumulates in the cortex area. Evidence has shown that Rab32 is localized to the cytoplasmic matrix, ER, Golgi, and lysosomal associated organelles (LRO), such as melanosomes and autophagosomes, as well as in mitochondria [24]. The localization pattern of Rab32 in oocyte is similar to that reported for several GTPase, such as Rab23 [17], Rab35 [25], Rab11 [26], and Rac1 [27] in oocytes. The different localization patterns of Rab32 staining with anti-Rab32 antibody and Myc-Rab32 T37N/Q83L labeling with anti-MYC antibody maybe due to 1) overexpression systems may lack proper post-translational modifications, 2) high overexpressed exogenous protein caused nonspecific aggregation, and 3) tag antibodies presented stronger and more specific signals. To explore the roles of Rab32 in meiosis, we used microinjection to knock down Rab32 and overexpression GTP-bound and GDP-bound Rab32, respectively. We showed that Rab32 depletion caused two types phenomenon: failure of polar body extrusion and large polar body generation, while in the Rab32 Q83L- and Rab32 T37N-expressing groups polar body extrusion rates decreased. Actin plays an essential role in assisting spindle migration, ensuring the asymmetric division of cytoplasm and polar body extrusion [28,29]. Then, we first examined the effect of Rab32 on actin filaments, and the data showed that Rab32 knockdown or excessive GDP-binding Rab32 could lead to a decrease in cytoplasmic actin, and active GTP-bound state of Rab32 accelerated cytoplasmic actin assembly. The excessive aggregation or reduction of actin both induced spindle migration disorder. Existing evidence indicates that the function of Rab32 is associated with actin. In the human melanoma cells MNT1, Rab32 was demonstrated to bind to Myosin Vc, the actin-based motor, for the biogenesis and secretion of melanosome [30]. As vital factors in vesicle transport, Rab GTPase members have been reported to involved in regulating the actin network to achieve correct spindle migration in oocyte meiosis. Rab11a-mediated vesicles drove the dynamics of the actin network by recruiting Myosin Vb to reassemble actin filaments, and regulated the network density by isolating and aggregating actin nucleation during mouse oocyte maturation [31]. In mouse oocytes, absence of Rab14 caused actin-mediated spindle migration defects via ROCK-cofilin signal pathway [32]. Next, we explored the potential mechanism by which Rab32 affects actin in oocytes. We found that ROCK expression was elevated with the active state of GTP-bound and decreased after injection with the inactive GDP-bound state. ROCK was an effector of Rho-GTPase which localized around spindle, and it has been demonstrated to regulate cytoplasmic actin assembly and meiotic spindle migration [33]. Meanwhile, there was also evidence that small GTPase played a role in actin accumulation via ROCK [34,35]. Based on our results and previous studies, we concluded that Rab32 regulated actin-mediated spindle migration in mouse oocyte meiosis, during which ROCK may play an important moderating role.

Mitochondria are highly dynamic organelles and the maturation of oocyte meiosis occurs in parallel with the trafficking of mitochondria surrounding various subcellular compartments [36]. We found that defects of Rab32 functions decreased mitochondria around spindle and brought about mitochondrial dysfunction. Moreover, the expression level of p-DRP1 was reduced in the overexpression group, indicating that exogenous supplementation of Rab32 would impair mitochondrial dynamics. As the rare Rab GTPase that resides in the mitochondria [37], effects of Rab32 on mitochondria has been reported in several models. Evidence suggested that inactive Rab32 disrupted mitochondrial dynamics by sequestering its effector DRP1 to block mitochondrial fission in HEK293T cells [38]. In human SH-SY5Y neuroblastoma cells, active Rab32 augmented mitochondrial fragment, and inactive Rab32 impeded mitochondrial motility [39]. In the pyroptosis model, Rab32 was capable of triggering pyroptosis in Schwann cells through the modulation of mitochondrial oxidative stress [40]. Besides, other members of Rab family are also involved in mitochondrial function during oocyte meiosis. In a CCCP-induced mitophagy, declined Rab7 activity triggers activation of the Parkin-mediated mitochondrial autophagy pathway in mouse oocytes [41]. In aging mouse oocyte, Rab9 accumulation activated the PRKN-mediated mitochondrial autophagy and led to mitochondrial dysfunction [42]. There is a close relationship between actin and mitochondria [43]. INF2, a protein that catalyzed the formation of actin filaments, was crucial for correct mitochondrial division [44]. Blocking the assembly of actin on mitochondria could interfere with mitochondrial fusion and fission [45]. An interacting role of Rab11a was reported on the trafficking of DRP1 for mitochondrial division and actin remodeling [46]. Rab32 transported DRP1 to the mitochondria and activated the ser616 site of DRP1 for mitochondrial fission in glial cells [47]. We speculate the reason for Rab32 interplaying in mitochondria and actin: the aberrant function of mitochondria caused the defective actin assembly, while actin was also involved in mitochondrial fission and transport, and thus the migration of mitochondria is hindered during oocyte meiosis.

In addition, we found that Rab32 was related with the quantity and spatial distribution of organelles within oocytes, encompassing Golgi apparatus, ER, and lysosome. And the expression of exogenous Rab32 Q83L/T37N induced the altered expression of proteins associated with organelle functions, including GM130 [48], GRP78 [49], and LC3 [50]. Notably, in the oocytes of expressing Rab32 T37N, the spatial organization of the Golgi apparatus was preserved, indicating that the introduction of inactive Rab32 selectively perturbed Golgi functionality without exerting influence on its positional integrity. Rab32 facilitated the degradation of mitochondria-proximal ER membranes through the process of autophagy, with the assistance of the long isoform of reticulon-3 (RTN3L) [51]. In C. elegans, GLO-1 was the counterpart to Rab32 in mammals, and activation of GLO-1 accelerated the LRO maturation [52]. Following the injection of Rab32 Q83L/T37N, our results indicated that overexpression of the active form of Rab32 results in a decrease in lysosome maturation, whereas expression of inactive form and knockdown of Rab32 led to lysosome aggregation. We speculated that it may serve as a compensatory response to sustain autophagic activity. In the mouse melanocytes, Rab32 regulated the key step of melanogenesis enzyme transport, particularly tyrosinase from the trans-Golgi network to melanosomes [53]. Rab proteins serve as pivotal regulators of vesical transport between various subcellular compartments during oocyte meiosis. In Drosophila oocytes, Rab5 and Rab11 were found to participate in the biogenesis of yolk granules, a process regulated by endolysosomal trafficking [54]. In addition, Rab3a and Rab27a were proved to respectively be involved in the regulation of CG migration and exocytosis in mouse oocytes [55]. These data suggested that Rab32 plays a role in regulating the distribution and relevant functions of membrane-associated organelles.

In conclusion, our study reveals that Rab32 mediates spindle migration via regulating the crosstalk of mitochondria and actin, and completes the trafficking of membrane-associated organelles during oocyte meiotic maturation.

Compliance with ethics requirement

All mouse experiments were conducted in strict compliance with the guidelines set forth by the Animal Research Institution of Nanjing Agricultural University in China, and the experimental protocols were granted approval by the Experimental Animal Research Committee (SYXK2023-007).

Availability of data and materials

All data in this study are included in this published article.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Supplementary Fig. S1.

Antibody specificity validation. (A) The full-length band of endogenous Rab32. (B) The localization of Rab32 staining with Rab32 antibody after overexpression Rab32 mutants. (C) The immunofluorescence images of Rab38, which indicated the localization of Rab38 associated with the spindle. (D) The separate staining experiment of the secondary antibody, serving as a negative control to demonstrate the specificity of the Rab32 antibody. (E) The target recognition site of mouse monoclonal anti-Rab32 antibody, which was distinct from Rab38 and Rab7.

Supplementary Fig. S2.

(A) Rab32 defects had no significant influence on cortical actin. Red, actin; blue, DNA; scale bar, 20 μm. (B) The statistical analysis of the fluorescence intensity of cortical actin, it showed no significance between control and Rab32-KD groups. Control (n = 135), Rab32-KD (n = 137), 1 vs. 0.95 ± 0.04, no significance. (C) Overexpression of Rab32 had no significant effects on cortical actin. Gray, actin; blue, DNA; scale bar, 20 μm. (D) The statistical analysis of the cortical actin after overexpressing of Rab32 showing with no difference. Control (n = 93), Rab32 Q83L (n = 97), Rab32 T37N (n = 93), control group: 1 vs. Rab32 Q83L: 0.92 ± 0.04, no significance vs. Rab32 T37N, 1.02 ± 0.04, no significance.