Nicotinamide mononucleotide protects ovarian function and oocyte developmental competence during chemotherapy

Abstract

Background

Cyclophosphamide (CTX)-induced ovarian dysfunction and infertility represent significant concerns for reproductive-age or younger female cancer patients. Although various fertility preservation techniques are currently accessible, there remains a pressing demand for an efficient, non-invasive strategy to protect ovarian function that can be employed concurrently with chemotherapy. Considering the significance of nicotinamide adenine dinucleotide (NAD+) in regulating DNA damage and apoptosis, we aimed to examine the protective effects of nicotinamide mononucleotide (NMN, an NAD⁺ precursor) on ovarian function against CTX-induced damage.

Results

Eight-week-old female C57 mice were underwent to a 14-day treatment protocol, receiving either saline, CTX, or CTX combined with NMN supplementation. The protective effects of NMN supplementation during CTX treatment on ovarian reserve, oocyte quality, and developmental competence were evaluated. NMN supplementation during CTX treatment increased NAD⁺ content in the ovary, improved ovarian reserve, enhanced endocrine function, reduced reactive oxygen species (ROS) levels, alleviated DNA damage, and reduced apoptosis. Furthermore, this supplementation improved the rates of two-cell embryo and blastocyst formation, increased total cell counts, while decreasing ROS levels, DNA damage, and apoptosis in blastocysts. Moreover, the protective mechanisms of NMN may involve key genes such as Banp and Rbm47 in the ovarian tissue, along with serum/glucocorticoid-regulated kinase 1 (Sgk1) in oocytes.

Conclusions

Collectively, our results highlight the protective effects of NMN against CTX-induced damage to the reproductive function, thus addressing a critical gap in fertility preservation. We present a potential non-invasive strategy that does not interfere with cancer therapy timelines.

Background

Despite revolutions in diagnostic and therapeutic methods that have substantially enhanced clinical remission rates for cancer, fertility perseveration remains a significant challenge for patients receiving anticancer therapy, with fertility reduced by approximately 30–50% compared to the general population [1–3]. Chemotherapeutic agents vary in their effects on ovarian function, with alkylating agents being particularly detrimental. Cyclophosphamide (CTX), widely employed in cancer treatment and for managing autoimmune diseases, is known for its strong reproductive toxicity and is classified as a high-risk agent [4, 5]. CTX induces granulosa cell apoptosis and follicular atresia in both human and mouse ovarian tissues, leading to a reduction in the number of primordial follicles (PmFs) [6]. This damage can result in amenorrhea, diminished ovarian function, and even premature ovarian failure [7–9].

The primary approach to fertility preservation before chemotherapy involves the cryopreservation of oocytes, embryos, or ovarian tissue. However, these methods frequently exhibit limited efficacy and may cause delay in the initiation of cancer treatment. Moreover, their applicability is particularly constrained in paediatric patients [10]. As a result, there is a critical need for an efficient, non-invasive strategy to protect ovarian function that can be administered concurrently with chemotherapy. Recent research has focused on identifying ovarian function protectors against CTX. Potential candidates include phosphatidylserine I, AS101, mTORC inhibitors, tamoxifen, tyrosine kinase inhibitors, and melatonin [11, 12]. Nevertheless, the safety and efficacy of these agents have yet to be fully validated, highlighting the urgent need to develop effective and reliable ovarian protectors to preserve reproductive health in female cancer patients receiving chemotherapy.

Nicotinamide mononucleotide (NMN) is enzymatically transformed into nicotinamide adenine dinucleotide (NAD⁺) by NMN adenosine transferase within the cell [13–16]. This enzymatic pathway constitutes the primary mechanism of NAD⁺ production within cells, with NMN acting as a rate-limiting intermediate and thus being a vital precursor in NAD⁺ synthesis [16, 17]. The coenzyme NAD⁺ serves as a crucial element in cellular metabolic processes, enabling various metabolic and redox reactions while assisting hydrogen transport throughout crucial metabolic pathways [18].

NAD⁺ has an essential regulatory role in several tissues and organs by reducing DNA damage and apoptosis. NMN supplementation, a precursor of NAD⁺, is reported to effectively increase the NAD⁺ content in organisms, leading to a significant decrease in γH2AX levels (a biomarker for DNA double-strand breaks) in fibroblasts and alleviating DNA damage in these cells [19]. Moreover, NMN utilisation via the NAD⁺/SIRT1 pathway has been shown to decrease platinum compound-induced apoptosis and attenuate disorder-induced oxidative damage in renal cells [20, 21]. Additionally, the replenishment of NAD⁺ levels can effectively restore reproductive function in aged mice [15].

Considering these effects of NAD⁺, NMN supplementation during chemotherapy may protect ovarian function. Therefore, this study examined the efficacy of NMN supplementation during CTX chemotherapy in preserving ovarian function and improving the quality and developmental competence of mature oocytes following chemotherapy.

Materials and methods

Ethical approval

Mouse experiments were conducted according to the guidelines and regulations approved by the Institutional Review Board of Tongji Hospital (approval number TJH-202305017). All animal experiments were performed based on the Animal Research: Reporting In Vivo Experiments (ARRIVE) guidelines.

Animals and treatments

Seven-week-old female C57 mice and 11-week-old male C57 mice were purchased from Jiangsu Jicui Animal Company. The mice were housed under a regulated environment (temperature 22 °C, humidity 60%, and a 12-h light-dark cycle) in the Experimental Animal Center of Tongji Hospital, and had free access to food and water. Before the commencement of the experiment, all mice were acclimatized for a week, and then randomly assigned to different experimental groups.

The 8-week-old female mice were randomly divided into three groups: blank control (control group), CTX-treated (CTX group), and NMN-protected (NMN + CTX group) (Fig. 1A). Mice in the CTX group and NMN + CTX group were intraperitoneally injected 150 mg/kg aqueous CTX on the morning of Day 1 (D1) and Day 6 (D6) [22]. Meanwhile, mice in the control group were injected with an equal volume of saline at the corresponding time points (Fig. 1B). In the NMN + CTX group, mice received intraperitoneal injections of NMN (200 mg/kg) once per night from Day 0 (D0) to Day 13 (D13) [23]. Meanwhile, mice in both the control and CTX groups were injected with an equal volume of saline during the same period (Fig. 1B). On the morning of D14, mice from each group were to be euthanized by cervical dislocation (a method used to rapidly euthanize the animals). Thereon, the left ovary, right ovary, and serum samples were collected separately for further analysis; alternatively, some mice underwent superovulation to obtain MII oocytes for in vitro fertilization (IVF) and subsequent embryo culture (Fig. 1A). Detailed information regarding the number of mice, ovaries, oocytes and/or blastocysts utilized in each individual experiment is included in the figure legends.

Fig. 1.

Outline of the study design (Fig. 1A) and Animal treatment (Fig. 1B). (Control: control group; CTX: CTX group; NMN + CTX: NMN + CTX group). Mice in the CTX and NMN + CTX groups received intraperitoneal injections of 150 mg/kg aqueous cyclophosphamide on the morning of Day 1 (D1) and Day 6 (D6), while mice in the control group received an equal volume of saline at the same time points. In the NMN + CTX group, mice were also injected intraperitoneally with NMN (200 mg/kg) once nightly from Day 0 (D0) to Day 13 (D13), while an equal volume of saline was injected simultaneously into mice in both the control and CTX groups (Fig. 1B). On the morning of D14, mice from each group were to be euthanized by cervical dislocation (a method used to rapidly euthanize the animals). Thereon, the left ovary, right ovary, and serum samples were collected separately for further analysis; alternatively, some mice underwent superovulation to obtain MII oocytes for in vitro fertilization (IVF) and subsequent embryo culture (Fig. 1A)

Measurement of NAD(H) content in the ovaries

On Day 14, an NAD/NADH quantification kit (Beyotime, Shanghai, China) was applied to determine the concentration of NAD⁺ of the right ovaries of each euthanized mouse from each experimental group, following the manufacturer’s instructions (Fig. 1A). Briefly, the mouse ovaries were weighed, and NAD(H) was extracted using a NAD(H) extraction solution. Subsequently, 100 µL of the supernatant was subjected to a 30-min incubation in a water bath at 60 °C to obtain the NAD(H) sample. Then, each well was supplemented with 20 µL of the sample and 90 µL of alcohol dehydrogenase, and incubated at 37 °C for 10 min. Subsequently, 2 µL of colour development solution was added to each well and incubated at 37 °C for 1 h. The absorbance at a wavelength of 450 nm was measured using an ELx808IU microplate reader (BioTek, Winooski, VT, USA). Finally, the concentrations were established by extrapolating the standard curves.

Follicle number assessment

On day 14, the left ovaries collected from each mouse were fixed, embedded, and serially sectioned into a series of 5-µm-thick sections for haematoxylin and eosin staining (Servicebio, Wuhan, China) (Fig. 1A). Follicles were counted in every fifth section to avoid duplication and ensure representative sampling. Follicles were categorized into developmental stages based on morphological features, including PmFs, early growing follicles (EgFs), antral follicles (AnFs), and atretic follicles (AtFs). Primary and secondary follicles were classified as EgFs [24]. The total amount of follicles was the sum of PmFs, EgFs and AnFs, excluding AtFs. Subsequently, the proportion of follicles at various developmental stages in each ovary was analysed. Additionally, the ovary index (OI) was used to accurately assess changes in ovarian mass relative to body weight ratio [25].

Enzyme-linked immunosorbent assay (ELISA)

Serum was extracted from the venous blood of mice in each experimental group on the morning of D14 (Fig. 1A). Serum concentrations of anti-Müllerian hormone (AMH) and follicle-stimulating hormone (FSH) were assessed using enzyme-linked immunosorbent assay (Elisa) kits for AMH (CEA228Mu, Cloud-CLONE, Houston, TX, USA) and FSH (CEA830Mu, Cloud-CLONE) following the manufacturer’s instructions.

After collecting the right ovaries of each mouse on the morning of D14 (Fig. 1A), the protein expression levels of RNA-Binding Protein 47 (RBM47) and BTG3 Associated Nuclear Protein (BANP) were evaluated using mouse RBM47 ELISA Kit (abx541619, Abbexa Ltd., Cambridge, United Kingdom) and mouse BANP ELISA Kit (abx392280, Abbexa Ltd.) following the manufacturer’s guidelines.

Ovarian transcriptome sequencing

The total RNA was extracted from the right ovaries of the mice on day 14, after euthanasia (four ovaries per group) (Fig. 1A). Subsequently, RNA libraries were then constructed and quantified using Qubit 3.0 and quantitative polymerase chain reaction (qPCR). Following successful quality control of the libraries, sequencing was carried out using the Illumina NovaSeq 6000 sequencing platform (Illumina, San Diego, CA, USA) in PE150 mode. Initial quality control of the raw sequencing data was performed, and high-quality reads were aligned to the reference genome (GRCh38) was performed for gene expression profiling and structural analysis.

Sequencing data underwent further quality control for downstream bioinformatic analyses. All downstream transcriptome analyses were conducted in R (version 4.3.1; Vienna, Austria) using the official DESeq2 pipeline (version 1.42.1). Differentially expressed genes (DEGs) were identified based on an absolute fold change of > 1.5 and a p-value < 0.05. The results were visualised using heatmap (version 1.0.12) and ggplot2 (version 3.5.0). Gene ontology (GO) enrichment analysis was performed utilizing the R package clusterProfiler (version 4.10.1). Enriched pathways with p-values < 0.05 were identified, and the intersection of different datasets was visualised by the R packages UpsetR (version 1.4.0) and VennDiagram (version 1.7.3).

Ovarian stimulation, in vitro fertilisation, and embryo culture

On D14, female mice from each group received an intraperitoneal injection of 10 IU of pregnant mare serum gonadotropin (Solarbio, Beijing, China), followed by an injection of 10 IU of human chorionic gonadotropin (HCG) (LIVZON, Zhuhai, China) after 48 h. The mice were euthanised 14–16 h post-HCG injection, and the cumulus-oocyte complexes (COCs) were collected from the oviducts in M2 medium (Aibei, Nanjing, China) (Fig. 1A).

Sperm was initially released from the cauda epididymis of both testes into a pre-equilibrated 900 µL of G-IVF solution (Vitrolife, Stockholm, Sweden) and capacitated for 1 h (37 °C, 5% CO₂, 5% O₂, and N₂ balance). Then 10 µL of sperm was added to the pre-equilibrated 90 µL drops of G-IVF medium containing up to 20 COCs per drop. They were incubated together for 4 h in a humidified environment with 37 °C, 5% CO₂, 5% O₂, and N₂ balance, after which the putative zygotes were denuded. Subsequently, ten zygotes were placed into a 20 µL drop of G1 solution (Vitrolife), which was covered with paraffin oil (Sigma-Aldrich, St Louis, MO, USA), within a humidified atmosphere (37°C, 5% CO₂, 5% O₂, and N₂ balance). Two-cell embryo rate was assessed on IVF-Day 2 (IVF-D2) (IVF day is considered to be IVF-D1), whereas the blastocyst formation rate was assessed on IVF-D6 [26].

ROS level assessment

After collection, live MII oocytes or IVF-D6 blastocysts were washed three times in M2 medium (Aibei), followed by incubation with 20 µL of M2 medium containing 0.1 µL of 2’,7’- dichlorodihydrofluorescein diacetate (Beyotime) for 30 min at 37 °C in the dark. Subsequently, they were washed three additional times in M2 medium and mounted on glass slides. Fluorescence was observed and recorded by a fluorescence microscope under consistent conditions. The fluorescence intensity of ROS within the oocytes/blastocysts was measured on the ImageJ software (Version 1.53, NIH, Bethesda, MD, USA) [27].

γH2AX level assessment

After collection, The MII oocytes or IVF-D6 blastocysts were fixed in 4% paraformaldehyde (Servicebio, Wuhan, China) for 30 min, and permeabilizated in 0.1% Triton-X-100 (Sigma-Aldrich, St Louis, MO, USA) for 20 min at room temperature. After blocking in 3% BSA (Sigma-Aldrich, St Louis, MO, USA) for 1 h, they were washed with 1% BSA and incubated with an anti-γH2AX antibody (1:200; CST, Toronto, Canada) at 4 °C overnight. Subsequently, they were washed with a 1% bovine serum albumin (BSA) solution for three times. Oocytes/blastocysts were incubated with Cy3-labeled Goat Anti-Rabbit IgG (H + L) (1:200; Servicebio, Wuhan, China) for 1 h at 37 °C, and then were stained with Hoechst 33,258 (Servicebio, Wuhan, China) for 10 min at 37 °C. Finally, they were mounted on glass slides and examined under identical conditions using a fluorescence microscope (Zeiss Axio Observer. A1, Oberkohen, Germany). Fluorescence intensity was analysed on ImageJ (version 1.53, NIH, Bethesda, MD, USA). The relative fluorescence intensity of γH2AX was determined as the ratio of fluorescence intensity at 570 nm to that at 460 nm.

Oocytes apoptosis assessment

The MII oocytes were stained using an Annexin-V Staining Kit (Beyotime). After washing with M2 medium, the oocytes were incubated in 20 µL of M2 medium containing 2 µL of Annexin-V-FITC for 10 min in darkness. Subsequently, they were washed three times in M2 medium and mounted on glass slides for observation under identical conditions using a fluorescence microscope. Fluorescence intensity was analysed using ImageJ software (Version 1.53, NIH).

Number of cells in blastocysts assessment

IVF-D6 blastocysts were fixed in 4% paraformaldehyde (Servicebio, Wuhan, China) for 30 min, and permeabilizated in 0.1% Triton-X-100 (Sigma-Aldrich, St Louis, MO, USA) for 20 min at room temperature. Subsequently, the blastocysts were stained with Hoechst 33,258 (Servicebio, Wuhan, China) for 10 min at 37 °C. Then they were mounted on glass slides and examined under identical conditions using a fluorescence microscope at 570 nm (Zeiss Axio Observer. A1, Oberkohen, Germany). Finally, the number of cells in blastocysts was counted using ImageJ software (version 1.53, NIH, Bethesda, MD, USA).

Blastocysts apoptosis assessment

The terminal deoxynucleotidyl transferase dUTP nick-end labelling (TUNEL) apoptosis detection kit (Beyotime) was used to assess apoptosis in IVF-D6 blastocysts, following the manufacturer’s instructions. Briefly, blastocysts were incubated with the TUNEL assay solution and subsequently stained with Hoechst 33,258. Subsequently, the labelled blastocysts were examined under a fluorescence microscope using the same parameters. Images were analysed using ImageJ software (Version 1.53, NIH).

Oocytes transcriptome sequencing

The oocyte transcriptome was sequenced using the Smart-seq 2 methodology. Three sets of samples, each containing eight MII oocytes, were collected from each group. Nucleic acid sequences were reverse transcribed using Oligo dT to generate first-strand cDNA. Subsequently, PCR amplification was performed to enrich the nucleic acids. The library was constructed through various steps, such as DNA fragmentation, end repair, the addition of ‘A’ junctions, and PCR amplification. Quality control measures were implemented during library preparation. An Illumina platform with a PE 150 sequencing strategy was employed to sequence the libraries, and raw reads obtained from Hi-Seq sequencing were subjected to preprocessing steps to remove low-quality sequences and splice contamination, resulting in high-quality clean reads that served as the basis for all subsequent analyses. Bioinformatic analysis was conducted following the previously described methods outlined in the section ‘Ovarian transcriptome sequencing’.

RNA isolation and quantitative real-time PCR (RT-qPCR)

Oocyte RNA was exacted and amplified with a Single Cell Sequence Specific Amplification Kit (Vazyme, Beijing, China) based on the manufacturer’s instructions. Ovary RNA was exacted with the RNA-easy Isolation Reagent and reversed to cDNA by HiScript ll Q RT SuperMix (Vazyme). PCR amplification was performed by Taq Pro Universal SYBR qPCR Master Mix (Vazyme). Table 1 presents the primer sequences.

Table 1.

Sequences of PCR primers used for RT-PCR

Gene	Gene ID	Forward primer (5′→ 3′)	Reverse primer (5′ → 3′)
Gapdh	14,433	AGGTCGGTGTGAACGGATTTG	TGTAGACCATGTAGTTGAGGTCA
Rbm47	245,945	AGCCATGAACAGCGATCCAAC	CCGGTGCGCTCTATCAGTG
Banp	53,325	AAGCGCCAGCGACTAGAGA	GCTATCCAACCGCAAACATATCG
F13a1	74,145	GAGCAGTCCCGCCCAATAAC	CCCTCTGCGGACAATCAACTTA
Sgk1	20,393	CTGCTCGAAGCACCCTTACC	TCCTGAGGATGGGACATTTTCA

RT-PCR: reverse transcription polymerase chain reaction

Statistical analysis

Data were analysed using GraphPad Prism 9 (GraphPad Software, San Diego, CA, USA). Normally distributed data were analysed using Student’s t test, whereas non-parametric data were assessed with a two-tailed Mann–Whitney U test. Dichotomous data were assessed using the two-sided Fisher’s exact test. The data are presented as means ± standard errors of the mean (SEMs). Statistical significance was set at p < 0.05.

Results

Replenishment of NMN mitigates the decline in ovarian NAD⁺ content induced by CTX

The NAD⁺, NADH, and NAD(H) levels in the ovaries were assessed across the three groups. Both NAD⁺ and NAD(H) levels were significantly decreased in the CTX group compared with those observed in the control group (Fig. 2A; p < 0.05). Conversely, NMN supplementation with CTX led to a significant increase in the NAD⁺ and NAD(H) content in the ovaries (Fig. 2A; p < 0.05). Moreover, neither CTX treatment nor NMN supplementation affected the NADH content in the ovaries (Fig. 2A; ns).

Fig. 2.

Protective effects of NMN on ovarian reserve during CTX treatment

(A) NMN supplementation during CTX treatment significantly increased NAD⁺ and NAD(H) content in the ovarian tissue. Control group (n = 6), CTX group (n = 6), and NMN + CTX group (n = 6). (B) NMN supplementation during CTX treatment significantly increased the count of PmF, AnF, TF, and decreased AtF. Control group (n = 6), CTX group (n = 6), and NMN + CTX group (n = 6). (C) NMN supplementation during CTX treatment significantly increased the percentage of PmF and decreased the percentage of EgF. However, it had no effect on the percentage of AF. Control group (n = 6), CTX group (n = 6), and NMN + CTX group (n = 7). (D) NMN supplementation during CTX treatment significantly increased AMH level and OI, and decreased FSH level. Control group (n = 6), CTX group (n = 6), and NMN + CTX group (n = 6). The data are presented as means ± SEMs. (****p < 0.0001, **p < 0.01, *p < 0.05, ns: no significance)

AnF: antral follicle, AtF: atretic follicle, EgF: early growing follicle, PmF: primordial follicle; TF: total follicle

Repletion of NMN demonstrated a protective effect against the detrimental impact of CTX on ovarian function

Compared with control treatment, CTX treatment significantly reduced the number of PmFs, AnFs, and total follicles (TFs), while increased the number of AtFs. (Fig. 2B; PmFs, TFs and AtFs: p < 0.01; AnFs: p < 0.05). Conversely, NMN supplementation during CTX treatment significantly increased the number of PmFs, AnFs, and TFs, while reducing AtFs compared to CTX treatment alone. (Fig. 2B; PmFs: p < 0.01; AnFs, TFs and AtFs: p < 0.05)

No significant difference was observed in the quantity of EgFs between the groups (Fig. 2B). However, compared with the control group mice, the CTX group displayed a significant reduction in the PmFs/TFs ratio and an increase in the EgFs/TFs ratio (Fig. 2C; p < 0.0001). Furthermore, NMN supplementation during CTX treatment significantly increased the PmFs/TFs ratio while suppressing the EgFs/TFs ratio (Fig. 2C; p < 0.0001). Notably, no variations were observed in the AnFs/TFs ratios among the experimental groups (Fig. 2C).

The serum AMH and FSH levels, which are crucial indicators of ovarian function, were measured in all experimental groups. In comparison to the control and NMN + CTX groups, the CTX group exhibited a significant decrease in the AMH (p < 0.01) levels and an increase in the FSH (p < 0.05) levels (Fig. 2D). Additionally, the OI was markedly lower in the CTX group compared to in the control (p < 0.0001) and NMN + CTX groups (p < 0.01) (Fig. 2D).

Characterisation of NMN target effectors in the ovaries during CTX treatment

Ovarian transcriptomic sequencing of the control group and CTX group revealed distinct patterns of gene expression differences between the two groups (Fig. 3A and B). Similarly, the expression patterns of DEGs between the CTX group and NMN + CTX group are shown in Fig. 3D and E. To further investigate the functions of DEGs between groups, we performed enrichment analysis of these genes using the GO framework. Compared with the control group, genes upregulated in the CTX group were enriched in redox and cell lysis-related pathways; whereas compared with the CTX group, genes upregulated in the NMN + CTX group were enriched in cell activation and cytokine-mediated signaling pathways (Figs. 3C and F). Given previous reports that CTX treatment enhances oxidative stress and DNA damage in injured tissues, we obtained DNA damage response genes (DDRGs) and oxidative stress genes (OSGs) from GeneCard, and identified their intersection with intergroup DEGs to determine the effects of NMN supplementation during CTX treatment on ovarian genes involved in oxidative stress and DNA damage response (Fig. 3G). Rbm47 and Banp emerged as key candidate genes, as their gene expression was significantly downregulated in CTX-treated mice, while NMN supplementation significantly restored their expression levels (Fig. 3H). The expression levels of Banp and Rbm47 were measured via reverse transcription polymerase chain reaction (RT-PCR), and ELISA results confirmed the corresponding protein levels; both sets of results were consistent with sequencing data (Fig. 3I–L). In summary, these findings highlight Rbm47 and Banp as potential effectors mediating CTX-induced ovarian damage and suggest that NMN supplementation can effectively mitigate its toxicity.

Fig. 3.

Characterisation of NMN target effectors in the ovaries during CTX treatment

A) Heat map illustrating the differential expression of ovarian genes between the control (n = 4) and CTX (n = 4) groups. Blue indicates down-regulated differentially expressed genes (DEGs), whereas red represents upregulated DEGs. B) Volcano plot of differential gene expression in ovarian tissue between the control and CTX groups. C) Gene Ontology analysis was performed to investigate the differential expression patterns in ovarian tissues between the control and CTX groups. D) Heat map illustrating the differential expression of ovarian genes between the CTX (n = 4) and NMN + CTX (n = 4) groups. E) Volcano plot of differential gene expression in ovarian tissue between the CTX and NMN + CTX groups. F) Gene ontology analysis was performed to investigate the differential expression patterns in ovarian tissues between the CTX and NMN + CTX groups. G) Upset plot showing the DNA damage response gene set and oxidative stress gene set. DDRGs: DNA damage response genes; OSGs: oxidative stress genes. H) Relative expression of the enriched genes Banp and Rbm47 in the control (n = 6), CTX (n = 6), and NMN + CTX (n = 6) groups. I) Ovarian Banp mRNA expression in the control (n = 6), CTX (n = 6), and NMN + CTX (n = 6) groups. J) Ovarian Banp expression in the control (n = 6), CTX (n = 6), and NMN + CTX (n = 6) groups. K) Ovarian Rbm47 mRNA expression in the control (n = 6), CTX (n = 6), and NMN + CTX (n = 6) groups. L) Ovarian Rbm47 protein expression in the control (n = 6), CTX (n = 6), and NMN + CTX (n = 6) groups. (****p < 0.0001, ** p < 0.01, * p < 0.05, |FC| >1.5)

Replenishment of NMN during CTX treatment reduces ROS levels in mature oocytes

Following ovarian stimulation, MII oocytes were collected from each experimental group. Significantly fewer MII oocytes were collected from the CTX group than those from the control and NMN + CTX groups (Fig. 4A; p < 0.0001 for CTX vs. control; p < 0.001 for CTX vs. NMN + CTX). The intensity of 2′,7′-dichlorodihydrofluorescein diacetate fluorescence was used to determine ROS levels in mature oocytes (Fig. 4B). CTX treatment significantly elevated ROS levels in oocytes compared with control treatment (Fig. 4C; p < 0.0001), whereas NMN supplementation during CTX treatment significantly attenuated ROS levels in oocytes (Fig. 4C; p < 0.01).

Fig. 4.

Replenishment of NMN during CTX treatment enhances the yield and quality of mature oocytes

A) CTX significantly reduced the number of retrieved mature oocytes; replenishment of NMN during CTX treatment effectively prevented the decrease in the number of retrieved mature oocytes induced by CTX (control: n = 6; CTX: n = 6; NMN + CTX: n = 6). B) Images of ROS fluorescence staining in mature oocytes. Scale bar = 50 μm. C) The fluorescence intensity of ROS signals was measured in the control, CTX, and NMN + CTX oocytes. CTX treatment significantly elevated ROS levels in mature oocytes compared with that in the control group, while NMN supplementation during CTX treatment significantly attenuated ROS levels in mature oocytes. (Control: n = 32 oocytes; CTX: n = 34 oocytes; NMN + CTX: n = 30 oocytes). D) Images of γH2AX fluorescence staining in mature oocytes. Red fluorescence represents γH2AX, blue fluorescence represents DNA, scale bar = 10 μm. E) The relative fluorescence intensity of γH2AX (red-fluorescence intensity/blue-fluorescence intensity) in the control (n = 36 oocytes), CTX (n = 34 oocytes), and NMN + CTX (n = 31 oocytes) oocytes. CTX treatment significantly increased the relative intensity of γH2AX in mature oocytes compared with that in the control group, while NMN supplementation during CTX treatment significantly attenuated the relative intensity of γH2AX in mature oocytes. F) Representative images of Annexin-V fluorescence staining in the control, CTX, and NMN + CTX group oocytes. Scale bar = 50 μm. G) The fluorescence intensity of Annexin-V signals in the control (n = 36 oocytes), CTX (n = 33 oocytes), and NMN + CTX (n = 29 oocytes) groups. Annexin-V level was significantly higher in the oocytes from the CTX group than in those from both the control and NMN + CTX groups. The data are presented as means ± SEM. (****p < 0.0001, *** p < 0.001, ** p < 0.01, * p < 0.05)

Replenishment of NMN-protected oocytes from DNA damage and CTX-induced apoptosis

The levels of DNA damage were assessed by quantifying the relative intensity of γ-H2AX in oocytes (Fig. 4D). The results revealed significant elevation in γ‐H2AX level in the oocytes from the CTX group (Fig. 4E p < 0.0001), whereas a considerable decrease in γ‐H2AX level was observed in oocytes from the NMN + CTX group (Fig. 4E; p < 0.01).

Furthermore, the fluorescence intensities of Annexin-V were used to determine the apoptotic status of mature oocytes (Fig. 4F). Annexin-V levels were significantly higher in oocytes from the CTX group compared to those from both control and NMN + CTX groups (Fig. 4G, p < 0.05).

Characterisation of NMN target effectors in mature oocytes by single-cell transcriptome analysis

Heat maps and volcano plots revealed significant differences in gene expression among the groups, indicating distinct transcriptome profiles among oocytes in the three groups (Fig. 5A, D). The distribution of differential genes was visualised using volcano plots (Fig. 5B, E). GO analysis of these DEGs revealed their involvement in biological processes such as oxidative stress, DNA damage response, apoptosis, and NAD⁺ metabolism (Fig. 5C, F). This suggests a close relationship between the effect of CTX/NMN + CTX on oocytes and oxidative stress/DNA damage.

Fig. 5.

Characterisation of NMN target effectors in mature oocytes during CTX treatment

For Oocytes transcriptome sequencing, three sets of samples (each containing eight MII oocytes) were collected from each group. (A) Differential gene heat map of the control and CTX groups, with differentially expressed down-regulated genes, are presented in blue, while differentially expressed elevated genes are indicated in red. (B) Volcano plot of DEGs in oocytes from the control and CTX groups. (C) GO enrichment analysis of DEGs in oocytes from the control and CTX groups. (D) Heat map of DEGs in oocytes from the CTX and NMN + CTX groups. (E) Volcano plot of DEGs in oocyte from the CTX and NMN + CTX groups. (F) GO enrichment analysis plot of DEGs in oocytes from the CTX and NMN + CTX groups. (G) Upset analysis of DNA damage response gene set and oxidative stress gene set. (H) Relative expression of the enriched genes Sgk1 and F13a1 in the control, CTX, and NMN + CTX groups. (I) Oocyte Sgk1 gene mRNA expression in the control (n = 6), CTX (n = 6), and NMN + CTX (n = 6) groups. (J) Oocyte F13a1 gene mRNA expression in the control (n = 5), CTX (n = 5), and NMN + CTX (n = 5) groups

DDRGs: DNA damage response genes, GO: gene ontology, OSGs: oxidative stress genes

We obtained DDRGs and OSGs from GeneCard and identified their intersections with the DEGs between the groups. The genes within these intersections were involved in oxidative stress and DNA damage responses in oocytes. Additionally, these intersected genes were affected by both CTX treatment and NMN supplementation (Fig. 5G). The results indicated that serum/glucocorticoid-regulated kinase 1 (Sgk1) and protein-glutamine gamma-glutamyltransferase A (F13a1) were potential target genes. These genes exhibited suppressed expression due to CTX treatment but increased expression upon NMN supplementation (Fig. 5H). To validate the sequencing findings, RT-PCR was used to assess Sgk1 and F13a1 expression levels in oocytes. The results demonstrated the stable expression of Sgk1 in oocytes, which was downregulated by CTX treatment and upregulated following NMN supplementation (Fig. 5I). Conversely, F13a1 exhibited unstable expression patterns in oocytes, making it undetectable in several experiments owing to its low expression levels, with significant heterogeneity observed among individual samples within the group (Fig. 5J).

Repletion of NMN enhances blastocyst production

After in vitro fertilisation, CTX treatment significantly impaired the IVF-D2 two-cell embryo rate and IVF-D6 blastocyst rate compared with control treatment (Fig. 6A, B; p < 0.05). Conversely, NMN supplementation during CTX treatment significantly improved both the IVF-D2 two-cell embryo rate and IVF-D6 blastocyst rate compared with CTX treatment alone (Fig. 6A, B; p < 0.01). Notably, NMN supplementation also had a positive effect on embryo quality.

Fig. 6.

Replenishment of NMN during CTX treatment improved MII oocyte developmental competence and embryo quality

(A) Images of 2-cell embryo and blastocyst formation in the control, CTX, and NMN + CTX groups. Scale bar = 50 μm. (B) The IVF-D2 two-cell embryo rate (A total of 113–135 embryos (control: 113; CTX: 128; NMN + CTX: 135) were assessed per group individually in 5 replicate experiments) and IVF-D6 blastocyst rate (A total of 96–108 embryos (control: 101; CTX: 96; NMN + CTX: 108) were assessed per group individually in five replicate experiments) in the groups. CTX treatment significantly reduced the IVF-D2 two-cell embryo rate and IVF-D6 blastocyst rate compared with that in both the control and NMN + CTX groups. (C) The total cell number of IVF-D6 blastocysts in the control (n = 33 embryos), CTX (n = 31 embryos), and NMN + CTX (n = 33 embryos) groups. CTX treatment significantly reduced the total cell number compared with that in both the control and NMN + CTX groups. (D) Images of ROS fluorescence staining in blastocysts from the control, CTX, and NMN + CTX groups. Scale bar = 25 μm. (E) The fluorescence intensity of ROS signals in the control (n = 33 embryos), CTX (n = 36 embryos) and NMN + CTX (n = 39 embryos) group. ROS fluorescence intensity in the control and NMN + CTX group were significantly lower than that in the CTX group. (F) Images of γH2AX fluorescence staining in blastocysts in the control, CTX, and NMN + CTX groups. Red fluorescence represents γH2AX, blue fluorescence represents DNA. Scale bar = 20 μm. (G) The relative fluorescence intensity of γH2AX (red-fluorescence intensity/blue-fluorescence intensity) in control (n = 33 embryos), CTX (n = 31 embryos), and NMN + CTX (n = 33 embryos) blastocysts. The relative fluorescence intensity of γH2AX in the control and NMN + CTX groups was significantly lower than that in the CTX group. (H) Image of TUNEL fluorescence staining of blastocysts from the control, CTX, and NMN + CTX groups. Green fluorescence represents TUNEL-positive cells, and blue fluorescence represents DNA. Scale bar = 20 μm. (I) The TUNEL-positive cell percentage in blastocysts in the control (n = 29 embryos), CTX (n = 24 embryos), and NMN + CTX (n = 20 embryos) groups. TUNEL-positive cell percentage in the control and NMN + CTX groups was significantly lower than that in the CTX group. The data are presented as means ± SEMs. (****p < 0.0001, *** p < 0.001, ** p < 0.01, * p < 0.05)

NMN supplementation improves embryo health

In comparison to the control group, the CTX group exhibited a significant reduction in the overall cell count of blastocysts (p < 0.05), as depicted in Fig. 6C. Conversely, the NMN + CTX group had a significantly higher overall cell count in blastocysts compared to CTX group (Fig. 6C; p < 0.01).

Additionally, CTX treatment increased ROS level, γH2AX fluorescence intensity, and proportion of TUNEL-positive cells in blastocysts compared with control treatment (ROS: p < 0.0001; γH2AX: p < 0.0001; TUNEL-positive cell percentage: p < 0.0001), whereas NMN supplementation reduced ROS generation, γH2AX fluorescence intensity, and TUNEL-positive cell percentage compared to CTX group (ROS: p < 0.001; γH2AX: p < 0.0001; TUNEL-positive cell percentage: p < 0.001) (Fig. 6D–I).

Discussion

NAD⁺ plays a crucial role in ovarian reproductive function [28, 29]. Several pathophysiological processes are associated with reduced NAD⁺ levels in the ovaries [30, 31]. Miao et al. reported that the highest number of mature oocytes was obtained from aged mice at doses of 200 and 500 rather than 1,000 mg/kg body weight/day of NMN [23]. Moreover, local NMN accumulation because of administration of high doses of NMN might have negative effects on tissues and cells [32, 33]. So, we initially used a dose of 200 mg/kg body weight/day. At the beginning of this research, we further examined the content of NAD⁺ in the ovarian tissue of the three groups (Fig. 2A) at this dose. The results showed that CTX treatment resulted in a significant decrease in NAD⁺ levels within ovarian tissues. Supplementation with NMN, an NAD⁺ precursor, effectively alleviated the CTX-induced decline in NAD⁺ levels in the ovaries. Interestingly, in Fig. 2A, in relation to NAD + content, the NMN + CTX group appeared to perform better than the control group. So, we compared the NAD⁺ content between NMN + CTX group and control group. However, there is no significant difference between these two groups in statistical analysis (NAD⁺, P = 0.39). This means that supplementation with NMN at this dose restored NAD⁺ in the ovaries effectively. These results are consistent with those reported by Ma et al. recently [34]. So, we used a dose of 200 mg/kg body weight/day for our subsequent study.

Administration of CTX in mice resulted in a notable decrease in the number of PmFs, AFs, TFs, serum AMH levels, and the OI, and an increase in the number of AtFs and serum FSH levels. These results indicated that CTX can significantly damage ovarian function in mice and exhibit strong reproductive toxicity in the ovaries, consisting with previous researches [35, 36]. To clarify the protective effects of NMN on ovarian function during CTX chemotherapy, we supplemented NMN along with CTX treatment. NMN supplementation significantly preserved the quantity of PmFs, AFs, TFs, serum AMH levels, and OI, and reduced the number of AtFs and serum FSH levels. These results suggest that NMN has the potential to mitigate CTX-induced injuries to the ovarian function.

Notably, the ratio of growing follicles to TFs changed significantly among the three groups, whereas the ratio of AFs to TFs did not. This may be attributed to CTX-induced injury and apoptosis of growing follicles, resulting in the recruitment of numerous PmFs to the growth stage. However, NMN reduced CTX-induced injury to growing follicles, reducing the recruitment of PmFs. In adult female mice, it takes at least 7 days from PmF activation to transition to EgF, and at least 37 days to transition to AF [37, 38]. The time point of observation in this study was set at 14 days after intraperitoneal injection of CTX, when the activated PmF entered the early growing follicle phase but had not yet progressed to the AF development stage. This explains why no differences were observed in the proportions of AFs.

To clarify the protective mechanisms and targets of NMN in ovarian function, we performed transcriptome sequencing of mouse ovaries. Our finding revealed that a significant proportion of DEGs were associated with responses to DNA damage and oxidative stress. Significantly reduced expression of Banp and Rbm47 was observed in the CTX group, whereas a significant increase in their expression was observed in the NMN + CTX group. These findings indicate that NMN may protect ovarian function from CTX-induced injury by targeting genes, such as Banp and Rbm47. Previous studies have shown that Bamp and Rbm47 were expressed in human and mouse ovaries [39–41]. Banp plays essential role in DNA damage repair of zebrafish cell and Rbm47 inhibited ubiquitination and degradation of Foxo3a, which inhibits PmF activation [42–44]. Nevertheless, the existing evidence concerning Banp and Rbm47 mainly stems from studies in other tissues and species, and remains rather limited. This study highlights the importance of Banp and Rbm47 in mouse ovaries, suggesting they may play a role in NMN-mediated protection against CTX damage. This offers us direction for further in-depth exploration of the mechanism by which NMN safeguards the ovaries from CTX-induced damage.

Notably, both clinical and basic studies have consistently demonstrated that, after exposure to CTX, a subset of ovarian follicles can still progress through ovulation, fertilisation, and embryonic development. However, the functionality of the surviving oocytes remains compromised, with a significant reduction in their quality. This impairment subsequently affects the overall quality of the resultant embryos. To investigate the impact of NMN supplementation during CTX treatment on the quality of oocytes, we assessed ROS levels, DNA damage, and apoptosis in MII oocytes.

The accumulation of ROS in oocytes has the potential to hinder the oocyte maturation process [45], induce DNA fragmentation and apoptotic cell death [46]. We evaluated ROS levels in oocytes and observed that CTX increased ROS levels and DNA damage, leading to early apoptosis. However, supplementation with NMN exhibited protective effects in oocytes by mitigating the detrimental effects of CTX. This was primarily evidenced by a substantial increase in the number of mature oocytes obtained through ovulation promotion, as well as a substantial reduction in ROS levels, DNA damage, and apoptosis levels within oocytes.

The sequencing results showed that DEGs among the three groups were associated with oxidative stress and DNA damage response. Notably, the expression of Sgk1 was significantly reduced in CTX group and markedly increased in NMN + CTX group compared with control group (Fig. 5). Furthermore, a study of the oocyte translatome revealed that Sgk1 exhibits a higher translation rate in young female oocytes compared with that in older oocytes [47]. In vitro studies showed that the use of inhibitors targeting Sgk1 kinase activity in young oocytes results in a diminished maturation rate and an increased proportion of spindle abnormalities in mature oocytes [47]. In addition, Sgk1 has the ability to attenuate oxidative stress and increase cell survival [48–50]. In zebrafish embryos, CTX inhibits the expression of Sgk1 [51]. Additionally, Sgk1-knockout mice exhibit impaired expression of genes associated with defects in oxidative stress defence, leading to developmental impairment and abortion of embryos [52]. Taken together with our findings, these results highlight the role of Sgk1 in alleviating CTX-induced damage in oocytes by reducing oxidative stress and DNA damage. More studies are needed to investigate the potential role of Sgk1 in protecting oocytes from chemotherapy-induced damage and its correlation with NMN.

Successful fertilisation and embryonic development necessitate the high-quality oocytes, which are essential for the initiation of embryogenesis [53]. In this study, CTX reduced both the cleavage rate and blastocyst formation of fertilized eggs. Although few oocytes managed to form blastocysts, they exhibited significantly elevated ROS levels, DNA damage, and apoptosis. Additionally, a significant reduction in cell count was observed within the blastocysts. These findings indicate that the detrimental effects of CTX on oocytes persisted throughout the embryonic period, consistent with the findings from previous toxicological studies of CTX [54, 55]. Notably, NMN supplementation during CTX treatment alleviated the adverse effects of CTX on embryos, resulting in a significant improvement in fertilisation and blastocyst formation rates, enhanced blastocyst cell number, reduced ROS levels, DNA damage, and apoptosis. NMN supplementation enhanced the developmental competence of oocytes and improved blastocyst quality.

Recent studies have demonstrated that NMN serves as an emerging protector of ovarian fertility against various forms of damage, including obesity, aging and chemotherapy especially [15, 56]. NMN has been shown to provide protective effects on ovarian function against chemotherapy (doxorubicin or cisplatin)-induced damage in mice [57]. However, the effect of NMN in protecting ovarian fertility against CTX-induced damage remains controversial [34, 58]. Stringer et al. reported that NMN failed to enhance the ovarian reserve impaired by CTX, while Ma et al. found that NMN protected the ovarian reserve against CTX-induced injury. The disparities may be attributed to variations in experimental designs, particularly in the timing and administration of NMN supplementation relative to CTX treatment. Specifically, Ma et al. continued NMN supplementation following the CTX administration, while Stringer et al. did not. Moreover, neither study investigated the protective effect of NMN on oocyte and embryo against CTX-induced damage. In our study, we used CTX as the chemotherapeutic agent and found that NMN supplementation provided protective benefits to ovarian reserve, oocyte quality, and embryo competence. Additionally, we conducted further investigations to elucidate the underlying mechanisms.

This research has some limitations. Firstly, the primary goal of procreation is to produce healthy offspring. Therefore, both the birth rate and long-term health outcomes of the offspring should be examined to assess the long-term effects of NMN. However, such evaluations were constrained by time and resource limitations. Secondly, although we have demonstrated the protective effects of NMN against CTX treatment in mice, it remains unclear whether supplementing NMN would interfere with the efficacy of CTX. Recently, Ho WJ et al. demonstrated that NMN does not interfere with the efficacy of chemotherapy agents such as doxorubicin, cisplatin, bleomycin, cisplatin, doxorubicin, etoposide, gemcitabine, methotrexate, paclitaxel, and vincristine [57]. Nevertheless, CTX was not included in their study. Although CTX, cisplatin and doxorubicin all lead to rapid depletion of the ovarian follicular reserve [11], the underlying mechanisms differ significantly among these agents. Before NMN can be clinically applied as an ovarian protective agent, further investigation, including clinical trials, are required to evaluate whether NMN administration during CTX chemotherapy affects drug efficacy or potentially promotes tumor growth.

Finally, this study primarily investigated the protective effects of NMN against CTX-induced ovarian damage and therefore did not evaluate the independent stimulatory effects of NMN. In the study by Ho WJ et al., the oocyte yield in 8-week-old mice treated with NMN alone showed no significant differences compared to that in healthy control mice [57]. Recently, we applied NMN in aged mice and demonstrated that reversing an age-dependent decline in NAD(P)H restored oocyte quality, embryo development, and functional fertility [15]. Moreover, Miao et al. reported that in vivo supplementation of NAD⁺ precursor NMN effectively improves the quality of maternally aged oocytes by restoring their mitochondrial function and enhancing meiotic competency, fertilization ability, and subsequent embryotic development potential [23]. As this was a preliminary investigation, further studies are necessary to clarify the standalone biological effects of NMN and to assess the safety of its supplementation in healthy mice, as current evidence remains insufficient to draw definitive conclusions. This issue is crucial for understanding the broader implications and safety profile of NMN supplementation.

Nonetheless, our findings suggest that NMN supplementation may serve as a non-invasive approach for preserving ovarian function during chemotherapy, providing a strong foundation for its clinical application.

Conclusion

In conclusion, the biochemical and genetic mechanisms highlight the protective effects of NMN against cyclophosphamide-induced injury to the reproductive function, addressing a critical gap in fertility preservation. We propose a non-invasive strategy for protecting ovarian function that does not interfere with cancer treatment timelines.

Abbreviations

AnF

Antral follicle

AtF

Atretic follicle

AMH

Anti-Müllerian hormone

COCs

Cumulus-oocyte complexes

CTX

Cyclophosphamide

DDRGs

DNA damage response genes

EgF

Early growing follicle

F13a1

Protein-glutamine gamma-glutamyltransferase A

FSH

Follicle-stimulating hormone

GO

Gene Ontology

HCG

Human chorionic gonadotropin

NAD+

Nicotinamide adenine dinucleotide

NMN

Nicotinamide mononucleotide

OI

Ovary Index

OSGs

Oxidative stress genes

PCR

Polymerase chain reaction

PmF

Primordial follicle

ROS

Reactive oxygen species

Sgk1

Serum/glucocorticoid-regulated kinase 1

TF

Total follicle

Author contributions

Lin Shen: Conception, design of the work, analysis, and drafted the work. Hemei Li: Acquisition and drafted the work. Xueqi Gong: Acquisition, analysis. Hanwang Zhang: Interpretation of data. Yiqing Zhao: Conception, design of the work, interpretation of data and substantively revised the manuscript.

Funding

This research was supported by the Basic Research Project for Young and Middle-Aged Physicians of the Beijing Health Promotion Association (grant number: BJHPA-2022-SHZHYX ZHQNYJ-JICH-002), Scientific Research Starting Foundation for Returned Overseas Scholars (grant number: 2020HGRY004) of Tongji Hospital, and Wuhan Knowledge Innovation Special Basic Research Project (grant number: 2022020801010535). The funders had no role in the conceptualisation, design, data collection, analysis, decision to publish, or preparation of the manuscript.

Data availability

Data is provided within the manuscript, and the data used and/or analysed in this study are available from the corresponding author on request.

Declarations

Ethics approval and consent to participate

This is an animal study. Mouse experiments were conducted following the guidelines and regulations approved by the Institutional Review Board of Tongji Hospital (approval number TJH-202305017). All animal experiments were performed based on the Animal Research: Reporting In Vivo Experiments (ARRIVE) guidelines. Consent to participate is not applicable.

Competing interests

The authors declare no competing interests.

Clinical trial number

Not applicable.