Nicotinamide mononucleotide biosynthesis and the F-actin cytoskeleton regulate spindle assembly and oocyte maturation quality in post-ovulatory aged porcine oocytes

Abstract

Background

Post-ovulatory aging (POA) is associated with reduced fertilization rates and poor embryo quality both in vivo and in vitro. However, the relationship between nicotinamide adenine dinucleotide (NAD⁺) and the filamentous actin (F-actin) cytoskeleton in POA-induced oocytes remains unknown. Here, we investigated the mechanisms by which the NAD⁺ salvage pathways function in poor oocyte maturation upon POA through the F-actin cytoskeleton.

Methods

Porcine oocytes were aged by extending in vitro maturation (IVM) for an additional 24 h to create a POA model. F-actin and adducin 1 (ADD1)-related spindle assembly were analyzed using immunofluorescence, western blotting, and RNA sequencing to identify key gene categories in the POA and IVM groups. To assess NAD⁺ function in restoring oocyte maturation, nicotinamide mononucleotide (NMN) was added and the maturation efficiency was evaluated. Expression of spindle assembly factors, F-actin cytoskeleton factors, aging markers, and NAD⁺-related genes was analyzed via quantitative polymerase chain reaction, immunofluorescence, and western blotting.

Results

We revealed unique interactions between the F-actin/ADD1-related cytoskeleton and aging factors (clusterin (CLU) and FAM111 trypsin-like peptidase A (FAM111A)) in poor-quality oocytes. POA oocytes were established with an extension of 24 h based on 44 h of IVM. They exhibited actin collapses and abnormal cortical F-actin, ADD1, and acetyl(Ac)-α-tubulin protein levels, which resulted in defective spindle assembly. RNA sequencing analysis was performed to identify differentially expressed genes involved in the oocyte viability response to aging, the cytoskeleton, and NAD metabolic processes using IVM and/or POA oocytes. This showed that NAD-binding genes were differentially expressed after POA induction, eight of which were downregulated compared with IVM oocytes. Importantly, activation of NAD⁺ pathways upon addition of NMN to the medium at 24 h after IVM rescued the maturation capability of POA oocytes with perturbations of spindle assembly and cortical F-actin.

Conclusion

F-actin polymerization through NAD⁺ generated from NMN is an essential factor in determining oocyte quality. This effect is mediated by microtubules related to spindle assembly in POA oocytes.

Introduction

The maturation of mammalian oocytes, which is crucial for fertilization, involves two consecutive meiotic cell divisions, transitioning from the germinal vesicle (GV) stage to the metaphase II (MII) stage [37]. During oocyte maturation, the correlation between spindle formation and microtubules is pivotal for successful meiotic maturation during first polar body (1st PB) extrusion [26]. Acetylation (Ac) of α-tubulin, a component of various microtubule structures, maintains microtubule stability and spindle integrity [10]. The dysregulation of Ac-tubulin levels in oocytes is closely associated with spindle assembly defects and chromosome misalignment during in vitro maturation (IVM) [16]. Therefore, the organization and dynamics of microtubules and cytoskeletal structure are fundamental for efficient spindle assembly and functionality during the early stages of embryonic development. The maturation capability of oocytes following IVM is assessed based on the spindle assembly and arrangement of microtubules by cytoskeletal factors [17]. This serves as a key criterion for evaluating oocyte quality. In particular, with the decline in the capacity of female germ cells due to aging, disruption of the spindle apparatus is one piece of evidence indicating a decrease in oocyte quality with age [31].

Post-ovulatory aging (POA) refers to the progressive deterioration of oocytes both in vivo and in vitro when optimal fertilization at the MII stage is delayed after ovulation [8]. This deterioration adversely affects oocyte viability and the potential for successful fertilization because the interval between ovulation and fertilization increases [29]. Oocyte maturation is governed by complex interplay of intra- and extra-ovarian factors; however, the precise mechanisms underlying POA remain unclear. POA serves as a model for in vitro aging, affects assisted reproduction technology procedures, and leads to reduced fertilization rates and impaired embryonic development [8, 40]. Structural and functional changes induced during aging, including chromosomal and spindle anomalies, cortical granule exocytosis, and zona pellucida hardening contribute to a decline in oocyte quality [1]. Low-quality oocytes share common characteristics with POA oocytes, including diminished spindle assembly and reduced nuclear maturation efficiency [8]. During in vitro aging of oocytes, abnormalities in the cytoskeleton emerge, which are characterized by imbalances in spindle morphology and alterations in cortical filamentous actin (F-actin) expression [38]. In particular, changes in F-actin levels are correlated with nicotinamide adenine dinucleotide (NAD⁺) levels in aging oocytes, and NAD⁺ depletion leads to cytoskeletal disorganization [12]. F-actin disruption affects spindle assembly and contributes to a decline in oocyte quality with aging.

NAD⁺ is an important cofactor in cellular energy metabolism, DNA repair, and other biosynthetic pathways [7]. A recent study showed that depletion of nicotinamide phosphoribosyltransferase (NAMPT), the main enzyme in the NAD⁺ biosynthetic pathway, in mouse oocytes leads to cytoskeletal abnormalities and impaired asymmetry [46]. NAMPT deficiency in aging mouse models results in meiotic defects in oocytes. Nicotinamide mononucleotide (NMN) is a key intermediate in the NAD⁺ biosynthesis pathway and is formed through the action of NAMPT [33]. Many studies have indicated that NMN supplementation shows promise in reversing deficiencies related to mitochondrial homeostasis, reactive oxygen species (ROS) production, DNA repair, and cell survival caused by inadequate NAD⁺ levels [2, 20, 41, 45]. Recently, treatment methods aimed at increasing NAD⁺ levels in aged oocytes were shown to enhance oocyte quality in various mammals, including mice [50], cattle [13], and pigs [34]. Consequently, numerous studies have suggested that generation and synthesis of NAD⁺ play an indispensable role in oocyte maturation [25, 33, 34]. Despite significant progress in understanding the role of NMN and NAMPT in the aging model, the protective effect of NMN on oocyte maturation in the porcine POA-induced model [27], particularly in regulating cytoskeletal dynamics and quality control, has not been fully determined.

In the current study, we performed mRNA transcriptome analysis to confirm that POA reduces porcine oocyte quality by affecting F-actin-mediated microtubule stabilization, cytoskeletal dynamics, and NAD⁺ biosynthesis. Furthermore, NMN supplementation enhanced the quality of aged oocytes by restoring F-actin-mediated spindle assembly and rescuing compromised NAD⁺ levels or biosynthesis associated with meiotic maturation during aging. In vitro administration of NMN restored NAD⁺ levels in porcine POA oocytes, resulting in an increased rate of meiotic maturation and improved oocyte quality through the regulation of microtubule-related spindle assembly and cytoskeletal dynamics.

Materials and methods

Experimental designs

Experiment I: Oocytes were classified into Grade 1 (G1) and Grade 2 (G2) based on cell number, expansion degree, and cytoplasmic lipid content after 44 h of in vitro maturation. We analyzed aging-inducing factors in poor-quality oocytes. Experiment II: We investigated actin collapse, abnormal cortical F-actin, and protein levels of ADD1 and Ac-α-tubulin in POA-induced oocytes, as well as changes in NAD synthesis through NAD-binding genes. Experiment III: We focused on the recovery of the NAD⁺ pathway in POA oocytes through NMN supplementation to rescue the spindle assembly and cortical F-actin.

Chemicals and animals

All chemicals and reagents used in this study were purchased from Sigma-Aldrich (St. Louis, MO, USA). Porcine ovaries were obtained from prepubertal sows (6-month-old female pigs; Yorkshire/Landrace (♀) × Duroc (♂), 100 kg) at a local slaughterhouse (Gyeongsan and Daegu, Korea). No experiments were performed on live animals.

Porcine immature oocyte collection and IVM

Prepubertal porcine ovaries were obtained from a local slaughterhouse and transported in saline containing antibiotics. Immature cumulus-oocyte complexes (COCs) were aspirated from follicles (3–6 mm in diameter) and washed in Tyrode’s lactate-N-2-hydroxyethylpiperazine-Nʹ-2-ethanesulfonic acid (TL-HEPES). Immature COCs were selected under an optical microscope, as previously described [32]. Approximately 50 immature COCs were matured in 500 µL of IVM medium at 38.5°C in 5% CO₂. North Carolina State University-23 medium supplemented with 10% follicular fluid (v/v), 0.57 mM cysteine, 10 ng/mL β-mercaptoethanol, 10 ng/mL epidermal growth factor, 10 IU/mL pregnant mare’s serum gonadotropin (PMSG), and 10 IU/mL human chorionic gonadotropin (hCG) was used for oocyte maturation. After culture for 22 h, COCs were matured in IVM medium without PMSG and hCG for an additional 22 h at 38.5°C in 5% CO₂. COCs were categorized based on the morphology of the cumulus cell layers: grade 1 (G1, oocytes with a minimum of three layers of compact cumulus cells) and grade 2 (G2, oocytes with a thin layer or partially denuded cumulus cells).

In vitro aging and NMN treatment

Good quality porcine oocytes that had been matured in vitro for 44 h were further cultured for 24 h at 38.5°C in 5% CO₂. NMN (25, 50, and 100 μM; N3501) was added to IVM medium. NMN treatment did not affect meiotic maturation (Supplementary Fig. 1). Thus, 25 μM NMN was chosen as the appropriate concentration for subsequent recovery experiments.

Assessment of cumulus cell expansion and orcein staining

Cumulus expansion in porcine COCs was assessed using a microscope (Leica, Solms, Germany) after 44 and 68 h of IVM in the G1/G2 and aged groups, respectively. Following previously described methods [30, 32], the process consisted of four steps (step 0 is no expansion, step 1 is separation of only the outermost layer of cumulus cells, step 2 is further expansion involving the outer half of the cumulus oophorus, and step 3 is further expansion up to complete expansion including corona radiate cells). After 44 h, meiotic maturation was determined based on the nuclear stage. Oocytes were denuded by pipetting in TL-HEPES medium supplemented with 0.1% hyaluronidase, rinsed with 0.1% polyvinyl alcohol prepared in phosphate-buffered saline, and mounted on microscope slides. Samples were fixed overnight in acetic acid/ethanol (1:3, v/v) and stained with 1% acetic orcein (v/v) for 5 min. The meiotic stages of the samples were examined under a microscope (Leica).

Measurement of perivitelline space (PVS) area of porcine oocytes

The images of IVM and POA groups in each experimental group were acquired using a microscope (Leica). To measure the PVS area, we measured cytoplasmic diameter, zona pellucida thickness, and total oocyte diameter respectively. The size of each part of the matured oocyte was measured using ImageJ software (NIH). The general measurement formula is: "PVS = (Total Oocyte Diameter − Cytoplasmic Diameter − Zona Pellucida Thickness)*1/2”.

F-actin and immunofluorescence staining

FITC-Phalloidin (P5282) and an anti-α-tubulin-FITC antibody (F2168) were prepared according to the manufacturer’s instructions. Porcine oocytes were fixed overnight with 3.7% formaldehyde at 4°C, permeabilized with 0.5% Triton X-100, and incubated with anti-adducin 1 (ADD1; Santa Cruz, Dallas, TX, USA; sc-33633) and anti-γ-H2AX (Abcam, Cambridge, England; ab11174) primary antibodies (1:100 dilution) for 1 h. Thereafter, oocytes were incubated with Alexa Fluor 555-conjugated goat anti-mouse IgG (Cell Signaling, Danvers, MA, USA; #4409) and Alexa Fluor 555-conjugated goat anti-rabbit IgG (Cell Signaling, #4413) secondary antibodies (1:200 dilution). The stained oocytes were mounted with 4ʹ,6-diamidino-2-phenylindole (DAPI; Vector Laboratories, Newark, CA, USA; H-1000–10) and imaged using an LSM800 confocal microscope (Zeiss, Jena, Germany). Chromosome misalignment was assessed by checking whether DAPI-stained chromosomes were symmetrically aligned relative to white arrows crossing their centers. Spindle abnormalities were evaluated based on α-tubulin staining, focusing on the accuracy of tubulin alignment with centrosomes as previously described [49, 52]. Histograms for densitometry analysis were obtained using ImageJ (NIH). All data were plotted using GraphPad Prism 5.0 (San Diego, CA, USA). Differences were considered significant at *p < 0.05, **p < 0.01, and ***p < 0.001.

RNA isolation and quantitative polymerase chain reaction (qPCR)

A total RNA extraction kit (TRI solution™; BSK Bio, Daegu, Korea) was used to extract total RNA from 100 denuded oocytes (DOs) of each group (G1, G2, M, A, M + NMN, A + NMN, and M-A + NMN groups). qPCR was performed using Power SYBR® Green PCR Master Mix (SYBR Green with low ROX; Applied Biosystems, MA, USA) containing specific primers (Table 1) in a Quant Studio 3 quantitative PCR machine (Applied Biosystems). The PCR conditions were as follows: 95℃ for 10 min, followed by 45 cycles of 95℃ for 15 s, 61–63℃ for 30 s, and 72℃ for 30 s. PCR specificity was identified by analyzing the melting curve data, and mRNA levels of specific genes were normalized to GAPDH. Gene expression was calculated using the 2^‒ΔΔCT method.

Table 1 Information of primers in qPCR

Western blotting

Mature COCs (50 per group) and DOs (150 per group) were collected and placed in PRO-PREP protein lysis buffer (iNtRON, Daejeon, Korea). Total proteins were separated on an 8–10% gradient gel and transferred to nitrocellulose membranes (Pall Life Sciences, NY, USA). The membranes were then incubated with anti-ADD1 (Santa Cruz), anti-Ac-α-tubulin (Cell Signaling), and anti-β-actin (Santa Cruz) primary antibodies. Thereafter, membranes were probed with horseradish peroxidase-conjugated secondary antibodies (Cell Signaling) overnight at 4°C. The blots were developed using an ECL kit (Cytiva, Amersham, UK) following the manufacturer’s instructions. The amount of protein was calculated based on band densities using the Fusion Solo software (Vilber Lourmat, Collégien, France), and the signal bands were quantified by scanning the bands using ImageJ software (NIH).

RNA sequencing (RNA-seq) analysis

Total RNA was isolated from porcine oocytes (500 DOs per group) in the IVM and POA groups using TRIzol reagent (Invitrogen, Waltham, MA, USA). RNA quality was assessed using an TapeStation 4000 system (Agilent Technologies, Amstelveen, The Netherlands) and the RNA Integrity Number equivalent (RIN^e) of the sequenced samples ranged from 6.2—7.0. RNA was quantified using an ND-2000 Spectrophotometer (Thermo Inc., DE, USA). Libraries were constructed for each RNA sample using a QuantSeq 3ʹ mRNA Seq Library Prep Kit (Lexogen Inc., GmbH, Austria) according to the manufacturer’s instructions. High-throughput single-end 75 bp sequencing was performed using NextSeq 500 (Illumina Inc., CA, USA). Differentially expressed genes (DEGs) that displayed a greater than 1.5-fold change in expression after POA in the entire transcriptome were analyzed using ExDEGA software (Excel-based Differentially Expressed Gene Analysis version 3.0; Ebiogen Inc., Seoul, Korea). The False Discovery Rate (FDR) was calculated with a significance threshold set at 0.05. Only genes with an FDR < 0.05 were considered significantly differentially expressed. Gene classification was based on searches of the Database for Annotation, Visualization, and Integrated Discovery (DAVID, http://david.abcc.ncifcrf.gov/, accessed on March 15, 2025) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway (http://www.genome. jp/kegg/mapper. com, accessed on March 3, 2024). GO and KEGG parameter settings both used a p-value < 0.05. Heatmap analysis of related genes in embryos was performed using Multi Experiment Viewer software. RNA-seq analysis was performed in three independent replicates to ensure reproducibility and reliability of the results.

NADH activity assay

The porcine oocytes were collected from Con, POA and NMN treated groups. Then, the NAD⁺/NADH ratio was determined using the NAD/NADH Assay Kit following the manufacturer’s instructions (ab65348; Abcam, ambridge, UK) as described by Kulthawatsiri et al. [21]. The samples were mixed with reagents following the manufacturer’s instructions (Abcam). The reactions were determined at 450 nm using InfiniteM200pro (Tecan, Männedorf, Switzerland).

Statistical analysis

Each experiment was repeated at least three times, except for RNA-seq analysis, which was performed based on two biological replicates. All percentage and image data are presented as mean ± standard deviation. Western blot and qPCR data are presented as mean ± standard error of the mean. The results were analyzed using a t-test and one-way ANOVA followed by Tukey’s multiple comparison test. All data were analyzed using GraphPad Prism software (version 5.0; San Diego, CA, USA). Histogram values in densitometric analysis were measured using ImageJ software (National Institutes of Health, Bethesda, MD, USA). Differences were considered significant at *p < 0.05, **p < 0.01, and ***p < 0.001.

Results

F-actin-mediated spindle assembly and DNA damage in low-quality oocytes during porcine oocyte maturation

Our study evaluated porcine oocyte quality based on the cytoplasmic lipid content and cumulus cell expansion at the beginning of IVM. Oocytes were categorized into two groups (G1 and G2) according to morphological criteria outlined by Park et al. [32] and Yang et al. [48]. G1 oocytes exhibited a high cytoplasmic content and 5.4 × 10³ cumulus cells, whereas G2 oocytes displayed a low cytoplasmic content and 1.2 × 10³ cumulus cells (Fig. 1A and B, Supplementary Tables 1 and 2; 44 h matured porcine oocytes). As depicted in Fig. 1C and D, the 1st PB extrusion rate (G1: 80.30% vs. G2: 43.13%, p < 0.001) and meiotic maturation rate (G1: 82.73% vs. G2: 64.58%, p < 0.05; Supplementary Table 3) were higher in the G1 group than in the G2 group. Histone H2AX recruits repair factors to DNA damage sites, with each γH2AX focus representing a single double-strand break [44]. γH2AX fluorescence was increased (p < 0.01) in G2 oocytes (Fig. 1E). ADD1 regulates mitotic spindle assembly as a centrosome protein [53]. Figure 1F demonstrates the alignment of ADD1 staining with DAPI staining, indicating that ADD1 expression was reduced in nuclei of low-quality oocytes in the G2 group. As shown in Fig. 1F, the intensity of ADD1 in nuclei was decreased in the G2 group, whereas the rates of misaligned chromosomes and abnormal spindle formation were increased in the G2 group (Fig. 1G). Additionally, the rates of misaligned chromosomes (G1: 36.47% vs. G2: 62.33%, p < 0.05) and abnormal spindles (G1: 28.89% vs. G2: 54.00%, p < 0.01) were significantly higher in the G2 group than in the G1 group (Fig. 1H and I). Abnormal spindle morphology with misaligned chromosomes is a key indicator of low oocyte quality [51]. Fluorescent expression of F-actin in the cortex of mature oocytes exhibited the same pattern as ADD1 expression (Fig. 1J). These findings suggest that ADD1 plays a role in regulating spindle assembly as a component of the oocyte microtubule-organizing center. Regulation of α-tubulin Ac is crucial for oocyte meiotic maturation and influences spindle morphology and chromosome alignment [35]. The ADD1 protein level was significantly higher (p < 0.01) in the G1 group than in the G2 group (COCs: Fig. 1K and DOs: Fig. 1L). Ac-α-tubulin expression was significantly increased (p < 0.001) in the COCs and DOs of G2 group. Furthermore, the gene expression levels of aging markers (FAM111 trypsin-like peptidase A (FAM111A) and clusterin (CLU)) were significantly higher (p < 0.05) in the G2 group than in the G1 group according to qPCR (Fig. 1M). The gene expression levels of factors related to the cytoskeleton, including microtubule-associated protein 7 (MAP7) and spindle- and kinetochore-associated protein 1 (SKA1), were significantly (p < 0.05) decreased in the G2 group (Fig. 1N). Collectively, these findings imply that insufficient microtubule stabilization stemming from spindle assembly leads to lower-quality pig oocytes. This deficiency may contribute to the increased induction of aging markers in porcine oocytes during IVM.

Fig. 1

Assessment of the maturation capacity of low-quality porcine oocytes. G1 represents high-quality oocytes based on cumulus cell expansion, while G2 represents low-quality oocytes. A Changes in cumulus cell expansion percentages in mature porcine COCs in the G1 and G2 groups. Scale bar, 500 μm. B Number of cumulus cells surrounding oocytes in the G1 and G2 groups. C Most oocytes in the G1 group extruded the 1st PB, in contrast to the significant reduction in extrusion observed in the G2 group (p < 0.001). D Meiotic stages classified as GV, GVBD, MI, and MII, with representative images of porcine oocyte meiotic maturation by orcein staining in the G1 and G2 groups after 44 h of IVM. Scale bar, 20 μm. Values with different superscript letters are significantly different (p < 0.05). E γ-H2AX expression levels in the nucleus measured by immunofluorescence staining in the G1 and G2 groups. Scale bars, 20 μm. F Fluorescent images of ADD1 in nuclei of oocytes in the G1 and G2 groups with pixel intensities measured along the lines drawn across the oocytes. Scale bar, 10 μm. G–I Comparative analysis of spindle assembly, chromosome alignment, and microtubule stability in porcine oocytes between the G1 and G2 groups, with representative images of the spindle/chromosome structure in both groups. Scale bar, 10 μm. J Fluorescence intensity of cortical F-actin in G1 and G2 oocytes with pixel intensities measured along the lines drawn across the oocytes. K and L ADD1 and Ac-α-tubulin in G1 and G2 porcine COCs and DOs examined by Western blot analysis, respectively. M and N Transcriptional expression level of aging markers (FAM111A and CLU) and spindle assembly-related cytoskeletal factors (MAP7 and SKA1) measured by qPCR analysis in G1 and G2 porcine oocytes. * p < 0.05, ** p < 0.01, and *** p < 0.001

Reduction of the maturation rate of POA oocytes

Figure 2A illustrates a schematic timeline for meiotic maturation and POA (68 h; additional 24 h after 44 h of IVM) in in vitro studies of porcine oocytes. Here, we explored meiotic maturation and cumulus cell expansion in porcine POA oocytes (Fig. 2B–D, Supplementary Table 4). POA oocytes displayed a non-significant trend towards a lower meiotic maturation rate (control (Con): 80.28% vs. Aged: 68.95%), decreased cumulus cell expansion (as assessed through three steps; Con: 69.30% vs. Aged: 3.56%, p < 0.001), and abnormalities in 1st PB extrusion (Con: 84.33% vs. Aged: 61.43%, p < 0.01). Increased γH2AX expression (Con: 28.94% vs. Aged: 61.29%, p < 0.01) in nuclei of POA oocytes indicated DNA damage similar to oocytes with degraded quality, suggesting POA oocytes had a reduced maturation capacity (Fig. 2E). As described in Fig. 2F, nuclear expression of ADD1 was significantly reduced in POA oocytes (p < 0.001), similar to oocytes in the G2 group. The spindle apparatus was labeled using an α-tubulin-FITC antibody, while chromosomes were counterstained with Hoechst [4]. As shown in Fig. 2G-I, the frequencies of α-tubulin-mediated abnormal spindle assembly (p < 0.01) and chromosome misalignment (p < 0.001) were significantly higher in the aged group than in the IVM group. Low-quality oocytes in the G2 group exhibited a phenotype strikingly similar to that of in vitro aged oocytes in the POA group. A scatter plot generated through RNA-seq analysis revealed that the expression levels of aging-related transcripts (CLU and FAM111A) and transcripts encoding spindle assembly-related cytoskeletal factors (MAP7 and SKA1) were changed more than 1.5-fold change in the aged group (Fig. 2J). Normalized log2 data show gene expression patterns, with lines highlighting changes in the POA group vs. Con (Fig. 2K). MAP7 promotes microtubule dynamics for intracellular transport in oocytes [43], while SKA1 is involved in spindle microtubule formation during mouse oocyte meiosis [22]. To confirm these findings, we conducted qPCR analysis of these genes (Fig. 2L and M). Expression of aging markers (CLU and FAM111A) was increased (p < 0.05), while expression of MAP7 (p < 0.001) and SKA1 (p < 0.05) was decreased in the aged group. These results suggest that maturation progression and the meiotic apparatus are disrupted in the POA model, resulting in low-quality porcine oocytes during IVM.

Fig. 2

Evaluation of meiotic maturation of aged porcine oocytes. A Schematic diagram of the experimental design. Porcine oocytes underwent 44 h of IVM in the M (Con) group. In the A group (aged: POA), oocytes underwent 68 h of IVM, i.e., an additional 24 h of culture after the initial 44 h of IVM. B Changes in cumulus cell expansion percentages in porcine COCs from the M (Con) and A (POA) groups. Scale bar, 500 μm. C Analysis of the porcine oocyte meiotic stage by orcein staining in the M (Con) and A (POA) groups. D Investigation of 1.^st PB extrusion in the A (POA) model (p < 0.01, statistical significance). E Fluorescence expression of γ-H2AX in the nucleus of oocytes measured by immunofluorescence staining in the M (Con) and A (POA) groups. Scale bar, 20 μm. F Fluorescent images showing ADD1 in nuclei of oocytes from the M (Con) and A (POA) groups, with pixel intensities analyzed along the lines drawn across the oocytes. Scale bar, 10 μm. G–I Representative images of α-tubulin and chromosomes in M (Con) and A (POA) oocytes. Scale bar, 10 μm. Investigation of spindle assembly, chromosome alignment, and microtubule stability in the M (Con) and A (POA) groups. J Scatter plot of global gene expression in the aging model (POA) compared to control oocytes, with a fold-change threshold (≥ 1.5 fold, p < 0.05). Genes are represented by dots (Red: up-regulated, Blue: down-regulated). Gene names associated with aging and microtubule-cytoskeleton functions are shown. K Changes in the expression of four genes (CLU, FAM111A, MAP7, and SKA1) are depicted, with CLU and FAM111A upregulated and MAP7 and SKA1 downregulated in the POA group. (L and M) qPCR analysis of selected genes (aging category: CLU and FAM111A; microtubule-cytoskeleton category: MAP7 and SKA1) from the list of DEGs determined by RNA-seq analysis. The mRNA levels of CLU, FAM111A, MAP7 and SKA1 in matured porcine denuded oocytes in POA group were measured using q-PCR, where GAPDH was used as the internal control. Expression in M (Con) oocytes (white bars) was compared with that in A (POA) oocytes (orange bars). All expression levels are relative to those in M (Con) oocytes, which were arbitrarily set to onefold. * p < 0.05, ** p < 0.01, and *** p < 0.001

Deficient microtubule-driven meiotic maturation of porcine POA oocytes

Thickened cortical F-actin in oocytes contributes to force balance and cortical softening, thereby enhancing oocyte quality [9]. Expression of cortical F-actin in mature oocytes was lower in the POA group than in the Con group (Fig. 3A). Moreover, western blotting indicated that the ADD1 protein level was significantly decreased (p < 0.001), whereas the Ac-α-tubulin protein level was significantly increased (p < 0.01), in POA oocytes (Fig. 3B). We measured the number of all genes that were changed in the POA group in three subcategories: Molecular Function, Cellular Component, and Biological Process using DAVID analysis. We also indicated the names of the five genes with the greatest significant changes. This analysis was conducted in two categories: Cytoskeleton (Fig. 3C) and Aging (Fig. 3D). Our results identified DEGs in the comparison between M (Control) and A (POA), showing that the transcriptome profiles of MII-stage oocytes differed from those of POA-aged oocytes (Fig. 3E-H). Cytoskeleton-related genes were significantly enriched in the microtubule-cytoskeleton organization category, which included 42 DEGs, including MAP7 and SKA1 (Fig. 3E). Aging-related genes were significantly enriched in the cell aging and age-related behavioral decline category (Fig. 3F), which included 32 genes, including FAM111A, B2M, and CLU. As shown in Fig. 3H, the scatter plot illustrates DEGs that met the criteria of adjusted fold change > 1.5 in NAD binding and NAD metabolic processes. In total, 25 DEGs were identified, including downregulated genes, such as sirtuin 1 (SIRT1), SIRT3, nicotinamide nucleotide adenylyltransferase 1 (NMNAT1), NMNAT2, NMNAT3, and NAMPT (Fig. 3H). To elucidate the increasing of niacinamide level in POA-induced porcine oocytes, we measured the NAD⁺/NADH ratio. Our results exhibited that POA oocytes significantly suppressed the NAD⁺/NADH ratio (Fig. 3I) compared with control group. These data suggest that genes linked to the microtubule-derived cytoskeleton and NAD-binding metabolism are crucial factors influencing porcine POA oocytes.

Fig. 3

Identification of ADD1/F-actin-mediated cytoskeleton stabilization and analyses of DEGs in POA oocytes. A Representative images of cortical F-actin signals in porcine oocytes from the IVM and aged groups. Porcine oocytes were stained with phalloidin-TRITC to label actin filaments. Scale bar, 50 μm. B Western blot analysis of ADD1 and Ac-α-tubulin in porcine oocytes following POA induction. The fold changes of protein levels were obtained by normalizing the signals to that of β-actin. Histograms show the densitometric data obtained using ImageJ software. C and D The graphs indicate Changes in gene count through DAVID functional annotation by A (POA) group compared to the M (Con) group. Each of the “cytoskeleton” and “aging” categories is divided into three subcategories: molecular function, cellular component, and biological process. The total number of the top five genes in each selected subcategory is shown. E–H The Scatter Plot shows differentially expressed genes in M (Con) and A (POA) meeting the conditions of adjusted Fold > 1.5. Comparison of the expression patterns of DEGs between M (Con) and A (POA) oocytes. The scatter plot shows DEGs that met the criteria of adjusted p < 0.05 and fold-change ≥ 1.5. (I) NAD.⁺ content in M (Con) and A (POA) oocytes. ** p < 0.01 and *** p < 0.001

RNA-seq transcriptome profiling and DEG analysis of POA oocytes after NMN exposure

NMN supplementation enhances in vitro oocyte quality and elicits anti-aging effects [11]. This led us to question whether artificially replenishing NMN to enhance NAD biosynthesis in vitro elicits protective effects in POA oocytes. However, NMN supplementation did not change the oocytes nuclear maturation (Supplementary Fig. 1). As shown in Fig. 4A, we divided oocytes into three groups according to the NMN treatment time, namely, IVM for 44 h in the presence of NMN (M + NMN), IVM for 44 h in the absence of NMN followed by aging for 24 h in the presence of NMN (Aged + NMN: A + NMN), and IVM and aging for 68 h in the presence of NMN (IVM–Aged + NMN: M–A + NMN), to verify the protective effects of NMN on NAD synthesis in the POA model. To analyze the effects of NMN on POA-induced aging oocytes, we compared the gene expression profiles of M (Con) with all NMN-treated groups based on the NMN supplementation treatment period (Fig. 4B). GO and KEGG enrichment analyses were conducted to delve deeper into the functional roles of the DEGs associated with the cytoskeleton and aging in oocytes of groups with differential NMN support based on the duration of IVM (Supplementary Fig. 2). The downregulated genes were mainly enriched in microtubule cytoskeleton organization, actin filament organization, and mitotic cell cycle (Supplementary Fig. 2). In the A + NMN group, DEGs were predominantly enriched in categories such as actin filament organization, mitotic spindle assembly, multicellular organism aging, and microtubule bundle formation (Supplementary Fig. 2). The heatmap cluster revealed the number of genes whose expression increased or decreased upon NMN treatment (fold change > 1.5, p < 0.05, Fig. 4C–F, Supplementary Table 5). The “Cytoskeleton” category included 42 DEGs and the “Aging” category included 71 DEGs (Fig. 4C and D) when enriched genes in the NMN-treated groups were compared with those in the control group. The RNA expression levels of FAM111A and SIRT1 were elevated in the A + NMN group (Fig. 4C). As shown in Fig. 4E and F, 39 DEGs were identified in the “NAD binding” category in the POA model, including up-regulated genes such as SIRT1 in POA group. In addition, 21 key genes, including NMNAT1–3 (Fig. 4F), showed decreased mRNA levels in POA oocytes (Fig. 5L). A protein–protein interaction (PPI) network of genes was constructed using the STRING database. The network includes cytoskeleton, aging, NAD binding, and the NAD metabolic process categories, with edges representing predicted functional associations. The PPI network was built at a mean confidence level of 0.4, resulting in a total of 42 nodes and 66 edges (Fig. 4G). These data suggest that genes linked to the microtubule cytoskeleton and NAD-binding metabolism play a significant role in influencing the anti-aging and maturation abilities of porcine POA oocytes upon NMN treatment, which aims to enhance NAD levels.

Fig. 4

Analyses of DEGs according to NMN treatment in IVM and POA oocytes. A Schematic diagram illustrating the NMN treatment method and the POA induction process of porcine oocytes. Oocytes were classified into five groups, namely, IVM (M: Con), aged (A: POA), M + NMN, A + NMN, and M-A + NMN, based on the periods of aging and NMN treatment. B RNA-seq analysis was performed to assess transcriptomes in the A, M + NMN, A + NMN, and M-A + NMN groups compared with the M (Con) group, followed by scatter plot analysis to confirm key factor expression after IVM or upon POA. The Scatter Plot shows differentially expressed genes in M(Con), A(POA) and those of NMN treated groups meeting the conditions of adjusted at least fold-change ≥ 1.5 and p < 0.05. Red and blue dots show up-regulation and down-regulation genes, respectively. The graph includes the name of key factor genes. C-F Heatmap cluster analysis of significantly up- or downregulated genes in POA oocytes. Heatmaps show the changes in the levels of representative transcripts in the indicated functional categories (Cytoskeleton, Aging, NAD binding, and NAD metabolic process). G Protein–Protein Interaction (PPI) network analysis of DEGs from RNA-seq data of IVM, POA and MNM treated groups. The network illustrates the interactions between key proteins, with edge thickness reflecting the strength of the interaction. The analysis was performed using Ingenuity Pathway Analysis (IPA) to identify significant molecular interactions related to the DEGs

Fig. 5

Enhancement of NAD biosynthesis by NMN supplementation improves cytoskeletal integrity and protects against aging-related abnormalities in POA oocytes. A Representative images of PVS morphology. PVS sizes in the M (Con), A (POA), A + NMN, and A-M + NMN groups. *** p < 0.001. Scale bar, 10 μm. B–D Comparative analysis of spindle assembly, chromosome alignment, and microtubule stability in porcine oocytes, with representative images of spindle/chromosome structure in the M (Con), A (POA), A + NMN, and M-A + NMN groups. Scale bar, 10 μm. E Fluorescent images of ADD1 in nuclei of oocytes in each group, with pixel intensities measured along the lines drawn across the oocytes. Scale bar, 10 μm. F Representative images of the F-actin distribution in all groups. Scale bar, 50 µm. The graphs on the right show fluorescence intensity profiling of phalloidin in oocytes. Lines were drawn through the oocytes, and pixel intensities were quantified along these lines. G and H Western blot analysis of ADD1 and Ac-α-tubulin in porcine COCs and DOs of each group, respectively. The fold changes of protein levels were calculated by normalizing the signals to that of β-actin. Histograms show the densitometric data obtained using ImageJ software. I–L The qPCR analysis of selected genes (aging category: CLU and FAM111A, microtubule-cytoskeleton category: MAP7 and SKA1, and NAD associated category: SRIT1 and NMNAT1–3) from the list of DEGs determined by RNA-seq analysis. Values from qPCR were normalized to GAPDH expression. M Quantitative analysis of NAD.⁺ content in porcine oocytes of M (Con), A (POA), and A (POA) + NMN (200 oocytes for each group). The p-values determined by the two-tailed Student’s t-test are indicated. Data are presented in arbitrary units as the mean ± S.E.M. *p < 0.05, **p < 0.01, and ***p < 0.001

Elevation of NAD biosynthesis increase microtubule and F-actin dynamics in porcine POA oocytes

POA oocytes were treated with 25 μM NMN in IVM medium for IVM or POA (in vitro aging; A, 44–68 h) and both IVM and POA (M + A; 0–68 h). To assess oocyte cytoplasmic maturation based on morphology and phenotype, we examined changes in the perivitelline space (PVS) in POA oocytes after NMN supplementation (Fig. 5A). PVS is the acellular region between the oocyte plasma membrane and zona pellucida, which is crucial for cytoplasmic maturation [31]. As described by Miao et al. [28], the PVS area was larger in the POA group than in the IVM group. However, the PVS area significantly shrank (p < 0.05) upon NMN supplementation during in vitro aging. The maturation arrest or low quality of oocytes relates to spindle assembly disruption; therefore, we confirmed the protective effect of NMN treatment in POA oocytes by staining them with an anti-α-tubulin-FITC antibody. POA oocytes exposed to NMN during in vitro aging displayed a typical barrel-shaped spindle with symmetrically aligned chromosomes, similar to oocytes in the IVM group (Fig. 5B–D). NMN restored the spindle/chromosome structure of oocytes compared with that in the POA group. Nuclear ADD1 is associated with mitotic spindles during entry into mitosis [5]. Furthermore, we examined ADD1 expression in nuclei of oocytes and verified the specific localization of ADD1 to meiotic spindles through colocalization analysis of α-tubulin and ADD1 in POA oocytes (Fig. 5E). Fluorescence imaging and quantification data confirmed the impairment of cortical F-actin in POA oocytes (Fig. 5F). Conversely, NMN-exposed POA oocytes exhibited significant recovery of cortical F-actin on the plasma membrane compared with POA oocytes. ADD1 and Ac-α-tubulin protein levels were recovered in porcine COCs (Fig. 5G) and DOs (Fig. 5H) of the A + NMN group, in which in vitro aging was conducted with NMN supplementation, compared with the POA group. Based on the RNA-seq analysis shown in Fig. 4, transcriptional expression of aging-, cytoskeleton-, and NAD biosynthesis-related genes was determined by qPCR in POA oocytes treated with NMN (Fig. 5I-L). The qRT-PCR analysis of the oocyte aging markers CLU and FAM111A showed that their elevated expression in POA oocytes was significantly downregulated in the NMN-treated groups similar to the control (Fig. 5I). NAD associated genes (SIRT1, NMNAT1, NMNAT2, and NMNAT3) were upregulated in these oocytes, and differences in expression of MAP7 and SKA1, which are related to spindle microtubules, were confirmed by qPCR (Fig. 5J‒L). Additionally, the NMN supplementation in POA-induced porcine oocytes was found to restore the decreased NAD⁺ levels in the POA aging model (Fig. 5M). NMN supplementation mitigates the meiotic defects induced by POA by upregulating NAD-synthesizing enzymes and preventing the accumulation of critical aging markers, thereby inhibiting the deterioration of porcine oocyte quality. Collectively, these observations suggest that NMN supplementation improves microtubule-mediated spindle assembly and cytoskeletal dynamics in porcine POA oocytes.

Discussion

Our previous study indicated that microtubule-mediated cytoskeleton stabilization positively affects embryonic developmental competence in pigs during in vitro culture (IVC) [18]. Several studies have demonstrated that damage of cortical F-actin and microtubule-related spindle assembly caused by oocyte aging can affect oocyte quality [36]. Based on these reports, in the present study, we hypothesized that enhancement of microtubule-derived spindle assembly and F-actin stabilization will improve the quality of porcine oocytes.

In our investigation, analysis of transcriptional expression levels in low-quality oocytes revealed the upregulation of FAM111A and CLU, both of which are markers of oocyte aging (Fig. 1M). The involvement of FAM111A in chromosome segregation and cell cycle regulation highlights its potential as a novel marker of oocyte aging [47]. Recently, the interaction of CLU with molecules involved in intracellular signaling and DNA repair was demonstrated, and its overexpression during cellular aging has been reported [42]. Moreover, we observed the increased occurrence of abnormal spindle assembly related to nuclear maturation, which was associated with both F-actin and microtubules. This was particularly evident in the G2 group (low-quality oocytes), mirroring the patterns observed under POA conditions (Figs. 1 and 2). These findings suggest that low-quality oocytes have elevated levels of prominent aging factors associated with POA.

The actin cytoskeleton plays a crucial role in maintaining cellular homeostasis. In particular, the cell structure, which is composed of microtubules and microfilaments, ensures precise chromosome alignment and segregation during meiotic cell division. Ac of α-tubulin, which is vital for microtubule function, is observed in mammalian oocytes during meiotic maturation. Ac-α-tubulin levels increased in aging oocytes, as previously described by He et al. [14], and this increase may reflect an adaptive response to aging, potentially contributing to the formation of astral microtubules [6, 24]. In the present study, Ac-α-tubulin was significantly upregulated in POA oocytes, consistent with previous results. POA occurs when an ovulated oocyte awaits fertilization by a sperm, during which high levels of tubulin Ac lead to abnormal spindle elongation during the MII-anaphase transition [3].

In mammals, SIRT1 and NAMPT form a novel regulatory network known as the NAD metabolic pathway. SIRT1, a member of the sirtuin (SIRT) family, acts as an NAD-dependent deAcase and is a key player in oocyte aging, suggesting it is a promising target to modulate age-related metabolism [15]. Our RNA-seq data showed altered expression of NAD synthesis factors (NMNAT1–3) and SIRT1 in POA oocytes (Fig. 3), with qPCR confirming increased SIRT1 and decreased NMNAT1–3 mRNA levels during aging. However, the specific roles of actin dynamics in NAD⁺ biosynthesis, particularly during in vitro aging of oocytes, remain unclear. Given that the characteristics of low-quality oocytes resemble the damage observed during both POA and in vitro aging, we hypothesized the potential for modulation of this regulatory network to improve oocyte quality and overcome POA. Therefore, we demonstrated that NAD⁺ content, which was reduced in POA-induced aging porcine oocytes, was restored after NMN treatment.

NMN, a precursor of NAD, was recently identified as an anti-aging factor or protective agent against POA, as well as an agent to improve porcine oocyte quality [39]. Previously, it was reported that NMN supplementation delays the aging process in porcine oocytes, leading to improved oocyte quality [23]. Furthermore, we suggest a close association between NAD⁺ biosynthesis and oocyte maturation capacity. However, a strategy to address the role of POA-related microtubule stabilization issues in spindle assembly perturbation has not been reported, other than leveraging the relationship between NAD⁺ biosynthesis and the oocyte maturation capacity during IVM. The age-associated depletion of NAD⁺ induces cytoskeletal changes, which particularly affect F-actin organization [19]. These alterations potentially compromise cellular functions and contribute to the overall decline in cellular health observed with age. Therefore, our findings on these mechanisms are important to understand spindle assembly integrity and maturation function in aging oocytes during in vitro processes.

Supplementation of aged mouse oocytes with the endogenous compound NMN improves fertility, as evidenced by the restoration of free NAD(P)H levels. Recent evidence showed that in vivo NMN supplementation restores NAD⁺ levels and enhances the quality of oocytes from aged mice by maintaining chromosomal euploidy and fertilization ability [27]. Supplementation of IVM culture media does not affect IVM of oocytes or the efficiency of subsequent early embryonic development (Pallard et al., 2021). Our results are consistent with this finding and show that NMN supplementation during IVM does not affect the meiotic maturation rate. However, RNA-seq analysis revealed significant differences in the transcriptional expression levels of enzymes such as NMNAT1 and NMNPT upon exposure to NMN during induction of POA following IVM (Fig. 4).

Importantly, the key finding of this study is that expression of the oocyte aging markers CLU and FAM111A was increased in POA oocytes with decreased quality. Validating its protective effect, we showed that NMN mitigated POA-induced meiotic defects and poor cytoplasmic maturation by upregulating cortical F-actin and ADD1, while reducing Ac-α-tubulin compared to POA group. Consequently, this intervention resolved the low oocyte quality resulting from the accumulation of damage due to in vitro aging of porcine oocytes (Fig. 5). These results indicate that the decline in oocyte quality due to POA is overcome by supporting the crucial NAD regulation involved in cytoskeletal and spindle-based chromosomal abnormalities.

Our study showed that POA oocytes exhibit characteristics similar to low-quality oocytes due to deficiencies in cytoskeleton-based cortical F-actin and spindle assembly, along with stabilization of Ac-α-tubulin. Additionally, increased levels of the aging markers CLU and FAM111A, which are elevated in POA oocytes, were observed in the G2 group with few cumulus cells and a low cytoplasmic lipid content. Based on these results, we concluded that low-quality oocytes were similar to oocytes with POA-induced damage (Fig. 6). Transcriptome sequencing analysis of POA and IVM oocytes revealed that NMN treatment restored expression of SIRT1 and NMNAT1–3, which encode enzymes related to NAD biosynthesis, confirming age-related changes in abnormal oocytes and cumulus cells. Biosynthesis of NAD from NMN mitigated abnormal spindle assembly and expression of the key aging factors CLU and FAM111A in POA oocytes.

Fig. 6

Graphical summary. I. Experiment (top panel): Schematic diagram depicting the roles of cytoskeletal stabilization in spindle assembly and aging in G1 (good-quality) and G2 (poor-quality) porcine oocytes. G2 oocytes demonstrate similarities to POA oocytes, characterized by upregulated expressions of the major aging markers CLU and FAM111A, thereby confirming their aging-associated characteristics. II. Experiment (middle panel): Schematic diagram depicting the major types of aging-related damage in porcine POA oocytes determined by transcriptome analysis. The diagram illustrates characteristics of aging related to the oocyte maturation ability, F-actin/ADD1-related cytoskeletal dynamics, aging-related gene expression, and salvage pathways (NAD⁺). Specifically, reduced NAD synthesis is associated with upregulation of aging factors and defects in F-actin-mediated microtubule assembly. III. Experiment (bottom panel): Schematic diagram illustrating the protective effect of NMN treatment on aging damage in porcine POA oocytes. NMN treatment effectively protects against aging-related damage in POA-induced oocytes. Specifically, NMN treatment during in vitro aging most effectively restores oocyte function, reducing the PVS area and abnormal spindle assembly. These improvements occur concomitant with normalized levels of nuclear ADD1 and Ac-α-tubulin. Furthermore, transcript levels of the NMN synthesis enzymes NMNAT1–3 and the longevity-related protein SIRT1, which are typically reduced during aging, are significantly restored by NMN treatment. Similarly, expression of the aging markers CLU and FAM111 is significantly mitigated, highlighting the role of NMN in a novel mechanism to overcome POA

This study emphasizes that depletion of NAD⁺ is a key characteristic of aged oocytes and suggests that this characteristic is similar to the increased expression of aging factors in low-quality oocytes. It is exciting to investigate how activation of NAD⁺ biosynthesis or enhancement of NAD⁺ availability affects the developmental efficiency of oocytes whose quality has declined due to aging up to the blastocyst stage using in vitro fertilized embryos. NMN reduces ROS levels in aged porcine oocytes, thereby improving mitochondrial function, which is closely related to enhancement of embryonic developmental competence [23]. These findings suggest that NMN supplementation during IVM can positively impact further embryonic development by enhancing NAD⁺ biosynthesis during IVC. Additionally, while this study used porcine POA oocytes, it provides essential resources for in-depth investigation of POA in clinical studies and for identifying the causes of infertility in older women.