MCM8 variants in two patients with primary ovarian insufficiency: clinical findings and in vitro defective DNA repair revealed by an MCM8 double-mutant construct Authors: Fei Wang, Shaolian Zang, Pin Li, Xiaoqin Yin Published: 2026 ABSTRACT BackgroundsPediatric-onset primary ovarian insufficiency (POI) presents distinct clinical challenges, including delayed puberty or growth retardation. Previously, we reported a Chinese family with POI harboring compound heterozygous variants (p.C242R and p.S445*) in MCM8 gene, a critical factor for DNA repair and gonadal homeostasis.ObjectiveThis study describes clinical findings of pediatric-onset POI patients with compound heterozygous MCM8 variants and investigates the in vitro functional consequences of an artificial double-mutant MCM8 construct harboring both p.C242R and p.S445*. Using this overexpression model, we assessed DNA damage response, protein interactions, and transcriptomic changes associated with the double-mutant MCM8.MethodsHeLa cell models were used to overexpress double-mutant (p.C242R and p.S445*) or wild-type MCM8. Functional consequences were evaluated using immunofluorescence staining, flow cytometry, and western blotting to assess DNA damage repair, cell cycle and apoptosis. Co-immunoprecipitation coupled with mass spectrometry (CoIP–MS/MS) was used to screen the disrupted protein interactome. Cut&Tag and mRNA-seq analyses were used to examine chromatin-binding-associated changes and transcriptomic alterations.ResultsOvarian tissues from POI patients exhibited reduced MCM8 expression and increased apoptosis. Cells expressing the artificial double-mutant MCM8 construct showed delayed resolution of DNA damage, as indicated by persistent γ-H2AX accumulation, and altered recruitment of DMC1 and RAD51. These changes were accompanied by S-phase accumulation and increased apoptosis. CoIP–MS/MS analysis identified altered interaction between double-mutant MCM8 and MCM6, suggesting possible changes in replication-associated protein interactions. Integrated CUT&Tag and RNA-seq analyses identified MCM8-associated chromatin-binding changes and transcriptomic alterations involving candidate genes enriched in ovarian survival and metabolic pathways, including the PI3K/AKT signaling pathway.ConclusionsThese specific MCM8 variants are clinically associated with pediatric-onset POI. In vitro, expression of double-mutant MCM8 construct was associated with impaired DNA damage resolution, altered MCM8-associated protein interactions, transcriptomic alterations, cell-cycle disturbance, and increased apoptosis. These findings provide cellular observations that may help generate hypotheses regarding how MCM8 dysfunction could contribute to ovarian insufficiency. Introduction Primary ovarian insufficiency (POI) is a clinical condition characterized by the cessation of ovarian function before the age of 40 and presents as amenorrhea, decreased estrogen levels, and elevated gonadotropin concentrations (1, 2). The diagnostic criteria for POI, according to the guidelines of the European Society of Human Reproduction and Embryology (ESHRE) (3, 4) and the National Institute for Health and Care Excellence (NICE) (5), include menstrual disturbance (such as primary amenorrhea, oligomenorrhea, or secondary amenorrhea (SA) for at least 4 months) and elevated follicle-stimulating hormone (FSH) levels exceeding 25 IU/L or 30 IU/L, respectively, measured on 2 different occasions at least 4 weeks apart. The same diagnostic principles are applied in adolescents (6). The etiology of pediatric POI is multifactorial and complex, encompassing genetic predispositions, autoimmune disorders, iatrogenic factors (such as chemotherapy and radiation therapy), and environmental influences (1, 7–9). Genome-wide association studies (GWASs) have revealed genetic variants associated with POI, which may manifest as ovarian dysfunction as early as childhood (10, 11). The clinical manifestations of pediatric POI are complex and heterogeneous and often exhibit considerable variation because of individual differences. Early identification and accurate diagnosis are critical for improving long-term prognostic outcomes in affected patients. Minichromosome maintenance complex component 8 (MCM8), a member of the minichromosome maintenance (MCM) protein family, is associated with POI and plays a critical role in ovarian development during prenatal fetal ovarian development (12, 13). However, the underlying molecular mechanisms remain poorly understood. MCM8 is located on chromosome 20p12.3 and comprises 19 exons that encode a protein of 840 amino acids (14). This protein contains an N-terminal DNA-binding domain, including a zinc finger motif, as well as a C-terminal AAA+ core domain, and functions as a DNA helicase involved in DNA replication, meiosis, and homologous recombination (HR) (15, 16). In particular, MCM8 plays a crucial role in meiotic prophase during prenatal fetal ovarian development (17, 18). MCM8 not only is involved in DNA replication but also serves as a novel regulator of germ cell survival. Upon the occurrence of a DNA double-strand break (DSB), MCM8 activates the cellular DNA damage response (DDR) pathway, leading to rapid phosphorylation of histone H2AX (γ-H2AX) (19, 20). Simultaneously, upon the induction of double-strand breaks (DSBs), MCM8 expression is promptly upregulated, and the protein is recruited to DNA damage sites. It associates with MCM9 to form a complex demonstrating intrinsic DNA helicase activity, which synergistically modulates MCM9 stability and enables the recruitment of the MRE11–RAD50–NBS1 (MRN) complex to the break site during HR (21, 22). This complex subsequently orchestrates the assembly of downstream effector proteins, including recombinases, thereby increasing RAD51 loading onto single-stranded DNA and contributing to the stabilization and protection of genomic integrity (23, 24). MCM6 is a core component of the classic MCM2–7 complex, forming the catalytic core of the helicase for DNA replication in eukaryotes. It is loaded onto the replication origin during the G1 phase and drives DNA synthesis in the S phase (25). The knockdown of MCM6 can lead to delayed progression in the S/G2 phase, inhibiting cell proliferation, migration and invasion (26). The replication process driven by MCM6 may rely on the repair mechanism mediated by MCM8/9 to maintain genomic integrity (24). In our previous study, we reported a Chinese family with pediatric-onset POI that harbors two compound heterozygous variants in MCM8 gene (NM_001281522.1: c.724T>C [p.Cys242Arg] and c.1334C>A [p.Ser445*]). These variants were classified as pathogenic according to ACMG criteria. Because MCM8 is implicated in DNA repair and gonadal maintenance, these variants may be clinically relevant to the POI phenotype. However, the cellular consequences of this specific variant combination remain insufficiently characterized. In the present study, we describe the clinical and histological findings of the affected patients and explore the in vitro cellular effects using an artificial cis double-mutant MCM8 construct harboring both p.C242R and p.S445*. This in vitro model was used to assess the combined effects of the two variants under controlled experimental conditions and was not intended to reproduce the patients’ physiological trans compound heterozygous genotype. This study may provide cellular observations which could help generate hypotheses regarding how MCM8 dysfunction could contribute to ovarian insufficiency. Materials and methods Participants In this study, a Chinese family with two compound heterozygous variants in the MCM8 gene was recruited. The variants were identified by whole-exome sequencing (WES) and were confirmed in her family members and matched controls by Sanger sequencing. The proband was a 13-year-old girl who presented with delayed puberty, ovarian insufficiency, short stature, significantly elevated levels of luteinizing hormone (LH) and follicle-stimulating hormone (FSH)—consistent with hypergonadotropic hypogonadism—and low anti-Müllerian hormone (AMH) levels. The proband’s younger sister was 6 years and 7 months of age and was assessed by genetic testing when the uterine and ovarian volumes and the FSH level were normal, but unexpectedly, at the age of 7 years and 10 months, ultrasound revealed that the bilateral ovaries gradually progressed from a normal size to an unclear display, accompanied by a decrease in the FSH level. At follow-up at 11 years of age, the FSH levels progressively manifested as hypergonadotrophic hypogonadism. All tissue acquisitions were conducted during surgical diagnosis, and informed consent was obtained from the guardians of the patients and the subjects in the control group. Control ovarian tissue was obtained from adolescent girls who had undergone surgical removal due to ovarian torsion. Ovarian tissue from the proband was subjected to histopathological examination, immunofluorescence staining, and TUNEL-based apoptosis analysis. This study received approval from the Ethics Committee of Shanghai Children’s Hospital (2024R051-E01), and written informed consent was obtained from the individual and legal guardian. Cell culture HeLa cells, a human cervical carcinoma cell line, were obtained from the Cell Bank of the Chinese Academy of Sciences. The cells were maintained in DMEM supplemented with 10% fetal bovine serum, 100 U/mL penicillin, and 100 μg/mL streptomycin. The cultures were incubated at 37 °C under a humidified atmosphere of 5% CO2. Western blotting Proteins were extracted using lysis buffer (Thermo, MI, USA) supplemented with protease inhibitor (Roche, Japan). After quantification with a BCA Protein Assay Kit (Thermo, MI, USA), 20 μg of protein lysate was loaded and separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE). The proteins were then transferred to polyvinylidene fluoride (PVDF) (Millipore, USA) membranes and incubated with primary antibodies overnight at 4 °C. Following washing, the membranes were incubated with HRP-conjugated secondary antibodies for 1 hour at room temperature. Protein signals were detected using an enhanced chemiluminescence (ECL) system, and protein intensities were quantified using ImageJ software. A complete list of antibodies used in this study is provided in Supplementary Table 1. Plasmid construction and cell transfection Wild-type and mutant eukaryotic expression vectors for the MCM8 gene were constructed and packaged by OBiO Technology (Shanghai, China). A single MCM8 expression plasmid harboring both the c.724T>C (p.C242R) and c.1334C>A (p.S445*) variants was constructed. All plasmids were confirmed by Sanger sequencing. The plasmids were amplified and purified using a plasmid extraction kit (TIANGEN, China). To evaluate the functional impact of the MCM8 variants, cells were seeded into 6-well plates at a density of 5 × 105 cells per well. Cells were transiently transfected with plasmids encoding the wild-type MCM8, the p.Cys242Arg (C242R) variant, the p.Ser445* (S445*) variant, or both the c.724T>C (p.C242R) and c.1334C>A (p.S445*) variants using Lipofectamine™ 3000 reagent (Thermo Fisher, USA) according to the manufacturer’s instructions and incubated for 48 hours post-transfection. At 12 hours after transfection, the culture medium was replaced. To induce DNA double-strand breaks (DSBs), transfected cells were treated with 600 nM mitomycin C (MMC) (GLPBIO, USA) for 21 hours. Following MMC treatment, the cells were either harvested immediately or allowed to recover in fresh complete medium at 37 °C for 2 hours prior to collection for downstream analyses. Immunofluorescence staining Immunofluorescence staining was performed to assess the expression levels of MCM8 and to evaluate apoptosis in MCM8-mutant ovarian tissue from patients with POI. The tissues were incubated with primary antibodies for 24 hours at 4 °C. Following the washing steps, the corresponding fluorescently labeled secondary antibodies were applied and incubated for 1 hour in the dark. 4’,6-Diamidino-2-phenylindole (DAPI) and anti-fade mounting medium were subsequently added to the samples. Finally, fluorescence images were captured and analyzed using a fluorescence microscope (Leica, Germany). Detailed information on the antibodies used in the study is provided in Supplementary Table 1. Immunohistochemical staining Formalin-fixed, paraffin-embedded (FFPE) tissue sections (4 µm thick) were subjected to antigen retrieval in citrate buffer at 95 °C for 20 minutes. Endogenous peroxidase activity was blocked by incubation with 3% H2O2 for 15 minutes at room temperature. Following blocking with 5% normal serum for 1 hour at room temperature, the sections were incubated overnight at 4 °C with primary antibodies. Horseradish peroxidase (HRP)-conjugated secondary antibodies were then applied for 1 hour at room temperature. Immunoreactivity was visualized using 3,3′-diaminobenzidine (DAB) as the chromogen, followed by hematoxylin counterstaining. Slides were imaged using an Olympus BX53 microscope and independently scored by two pathologists who were blinded to the experimental conditions on the basis of a semiquantitative scoring system (0–3 scale). Coimmunoprecipitation Cells were lysed in NP-40 lysis buffer (a nonionic detergent) supplemented with protease inhibitors. The lysates were precleared using Protein A/G beads for 1 hour at 4 °C, followed by incubation of 500 µg of protein with 2 µg of specific antibody or control IgG overnight at 4 °C. Immune complexes were then captured with Protein A/G beads for 4 hours, after which they were washed four times with lysis buffer and eluted in Laemmli buffer. Coprecipitated proteins were subsequently analyzed by immunoblotting and mass spectrometry. Cell cycle analysis To analyze the cell cycle, the cells were fixed in ice-cold 70% ethanol for at least 2 hours. Following fixation, the cells were treated with RNase A (100 μg/mL; Thermo Fisher, USA) to digest the RNA and then stained with propidium iodide (50 μg/mL; Thermo Fisher, USA) for DNA content analysis. The samples were analyzed using a BD FACS flow cytometer (BD Biosciences, USA), and cell doublets were excluded by plotting the forward scatter width (FSC-W) against the forward scatter area (FSC-A). The proportions of cells in the G0/G1, S, and G2/M phases were determined using FlowJo software, and the sub-G1 peak was used as an indicator of apoptotic cells. Cleavage under targets and tagmentation assay To map the genome-wide chromatin occupancy of MCM8, we performed a Cut&Tag assay (GenSeq® Cut&Tag Kit, GenSeq Inc.) according to established protocols with minor modifications. In brief, nuclei from HeLa cells expressing wild-type (WT) MCM8 and double-mutant (C242R-S445*) MCM8 were isolated, permeabilized, and immobilized on concanavalin A-coated magnetic beads. The chromatin was then sequentially incubated with a primary anti-MCM8 antibody, a secondary antibody, and a protein A-Tn5 transposase fusion protein preloaded with sequencing adapters. After activation with Mg²+ to initiate targeted tagmentation, DNA was purified to generate sequencing libraries using GenSeq® 2× HiFi PCR Mix (GenSeq Inc.). A control sample using rabbit IgG was processed in parallel. The amplified libraries were quality-checked by Agilent 2100 Bioanalyzer and then subjected to high-throughput sequencing on Illumina NovaSeq sequencing platform (Illumina Novaseq 6000). Using bowtie2 software (v2.2.4) align high-quality reads onto the reference genome. MACS software (v1.4.2) was used for Peak Calling, diffReps software (v1.55.4) was used to identify differentially enriched regions, and UCSC RefSeq database was used for annotation to link peak and gene information. Differential binding analysis was performed to identify sites with significantly altered MCM8 occupancy in mutant versus WT cells. Statistical analysis All the quantitative data are expressed as the mean ± standard deviation (SD) of a minimum of three biological replicates. Statistical significance between experimental groups was determined by applying an unpaired, two-tailed Student’s t test using IBM SPSS Statistics (version 13.0). The data met the assumptions of the t test (normal distribution and homogeneity of variance). One-way ANOVA with Tukey’s post hoc test was used for multigroup comparisons. The associated candidate genes detected by Cut&Tag sequencing (fold change > 1) and the differentially expressed genes detected by mRNA-seq (fold change >= 1.5 and adjusted P-value [FDR] < 0.05) were intersected to identify genes associated with both altered chromatin binding and transcriptomic changes after expression of double-mutant MCM8 construct. Differences were considered statistically significant at P < 0.05. Graphical representations of the data were created with GraphPad Prism (version 8.0.1). Results Clinical and histological findings in patients carrying compound heterozygous MCM8 variants Pedigree analysis showed that the father carried a heterozygous MCM8 p.C242R variant, whereas the mother carried a heterozygous MCM8 p.S445* variant. Both affected daughters inherited the two variants and carried compound heterozygous MCM8 variants, p.C242R and p.S445* (Figure 1A). The proband was diagnosed with POI at 13 years of age, and her younger sister was diagnosed with POI at 6 years and 7 months of age. These findings support the clinical association between biallelic MCM8 variants and pediatric-onset POI in this family. Compared with control ovarian tissue, patient-derived streak gonadal tissue showed reduced MCM8 staining (Figure 1B). Quantitative analysis showed that the percentage of MCM8-positive cells was lower in the proband’s tissue (33.75%) than in the control (99.07%) (Figure 1C). TUNEL staining showed increased apoptosis in patient-derived tissue, with TUNEL-positive cells accounting for 42.75% of cells in the proband’s tissue and 0.17% in control (Figure 1D). These findings indicate reduced MCM8 expression and increased apoptosis in patient-derived streak gonadal tissue, and they are presented as descriptive histological observations. Clinical and histological findings in patients carrying compound heterozygous MCM8 variants. (A) Pedigree of the family carrying MCM8 variants. The proband is Subject II-1 (arrow). (B) Immunohistochemical staining of MCM8 in ovarian tissue from the proband and control ovarian tissue. Hematoxylin stains the cell nucleus blue, and positive expression of MCM8 appears as brownish yellow. (C) Immunofluorescence staining and quantification of MCM8-positive cells in ovarian tissue. DAPI-stained cell nuclei appear blue, and positive expression of MCM8 with corresponding fluorophore labeling emits a red light. (D) Fluorescence images and positive rates of TUNEL staining in the ovarian tissues of patients with MCM8 variants. NC: normal ovarian tissue. Patient: Patient with MCM8 mutation. DAPI stains the cell nuclei blue. TUNEL fluorescein labeling appears as a green light. DNA damage and repair abnormalities associated with double-mutant MCM8 construct To assess cellular changes associated with the combined presence of the two variants in vitro, HeLa cells were transfected with an artificial cis double-mutant MCM8 overexpression construct harboring p.C242R and p.S445*. Following MMC treatment, all cells showed increased γ-H2AX levels, indicating activation of the DNA damage response and accumulation of DNA damage-associated signals. After 2 hours of recovery, γ-H2AX levels decreased in WT cells, suggesting resolution of DNA damage-associated signaling, but remained elevated in double-mutant construct. The C242R-S445* double mutant showed the strongest γ-H2AX induction post-MMC and paradoxical increase during recovery (Figure 2A). This indicates delayed resolution of DNA damage in cells expressing the artificial double-mutant MCM8 construct. We then analyzed the dynamics of key HR repair proteins. In WT cells, the recruitment of DMC1 and RAD51 was transient, peaking after damage and returning to near-baseline levels following recovery. This normal resolution of the HR machinery was abrogated in cells expressing the C242R-S445* mutant, in which elevated levels of both DMC1 and RAD51 persisted throughout the recovery period, with no significant decrease relative to the peak damage phase (Figure 2B). Based on these findings, which suggest delayed resolution of DNA damage-associated signaling and altered HR-related factor dynamics associated with double-mutant MCM8, the C242R-S445* construct was selected for subsequent functional investigations. Effects of MCM8 mutation on DNA damage and repair. (A) Differences in the DNA damage repair response between the S445* mutation and the compound mutant C242R-S445* of MCM8. (B) Role of compound mutant MCM8 in the DMC1–RAD51 homologous recombination repair pathway. WT, overexpression of wild-type MCM8; -0, DNA damage; -2, DNA repair. Cell-cycle and apoptosis changes associated with double-mutant MCM8 We further analyzed the effects of the double-mutant MCM8 on the cell cycle and apoptosis. After DNA damage induction, no obvious differences in the expression patterns of cyclin D1, cyclin A2, and cyclin B1 were observed between double-mutant cells (C242R-S445*-0) and wild-type cells (WT-0) (Figure 3A). During the recovery period, the expression of these three cyclins was downregulated in the wild-type cells (WT-2), whereas only cyclin D1 was downregulated in the double-mutant cells (C242R-S445*-2), and the expression of cyclin A2 and cyclin B1 remained unchanged (Figure 3A). Flow cytometry analysis revealed that the proportion of cells in the G1 phase increased in all groups after damage. After recovery, the proportion of G2 phase cells among the wild-type cells significantly increased, whereas the double-mutant cells exhibited a decrease in G1 phase cells and a significant increase in S phase cells (Figure 3B). Regarding apoptosis-related proteins, the expression of Bax, caspase-3, and Bcl-2 increased in both wild-type and double-mutant cells after damage induction (Figure 3C). After recovery, the expression of Bax in the wild-type cells significantly decreased, whereas the expression of Bcl-2 in the double-mutant cells did not decrease significantly, and the downregulation of expression of Bax and caspase-3 was delayed in the double-mutant cells (Figure 3C). Flow cytometry revealed that the total apoptosis rate of the double-mutant cells was significantly greater than that of the wild-type cells after damage; after recovery, the apoptosis rate of the double-mutant cells decreased, while the apoptosis rate of the wild-type cells did not significantly differ at different time points (Figure 3D). Together, these results indicate that expression of double-mutant MCM8 construct was associated with altered cell-cycle distribution and increased DNA damage-associated apoptotic signals in this overexpression model. Effects of MCM8 mutation on the cell cycle and apoptosis. (A) Effects of a mutant MCM8 on cyclin expression during the DNA damage and repair phases. (B) Differential analysis of the effects of a mutant MCM8 on the cell cycle during the DNA damage and repair phases. NC: empty vector; WT: wild-type; MT: double-mutant C242R-S445* type; -0: DNA damage; -2: DNA repair. (C) Effect of mutant MCM8 on the expression of Bcl-2/Bax/caspase-3 apoptotic proteins during DNA damage repair. (D) A mutant MCM8 promotes apoptosis during DNA damage and repair. NC: empty vector; WT: wild-type MCM8; MT: double-mutant MCM8 (C242R-S445*); -0: DNA damage; -2: DNA repair. UR (upper right): Late apoptotic and necrotic cells (annexin V+/PI+). LL (lower left): Live cells (annexin V-/PI-). LR (lower right): Early apoptotic cells (annexin V+/PI-). *** p<0.001. CoIP–MS analysis of proteins associated with wild-type and double-mutant MCM8 To identify candidate proteins associated with MCM8, we performed CoIP–MS analysis. Western blotting confirmed that the bait protein was successfully expressed and enriched in both wild-type and double-mutant MCM8 cells (Figure 4A), and silver staining revealed that the protein quantity met the requirements for mass spectrometry (Figure 4B). A total of 1768 proteins were identified by mass spectrometry. A total of 159 differentially enriched MCM8-associated proteins were identified between wild-type and double-mutant MCM8 (98 increased and 59 decreased) (Table 1). Notably, although MCM9 is the canonical binding partner of MCM8, it was not identified among these 159 differentially enriched proteins. Therefore, we did not detect evidence for a major change in MCM8–MCM9 co-enrichment under these experimental conditions. Further analysis revealed 14 common and 51 specific differentially enriched proteins (Table 2), among which RAI14, MCM6, and ETFB were detected across replicates (Figure 4C). GO functional annotation revealed that the differentially enriched MCM8-associated proteins were involved mainly in cell processes, metabolism, biological regulation, stress response, and protein complex assembly (Table 3). The significantly enriched terms closely related to the function of MCM8 included cation channel complex, ion channel complex, DNA packaging complex, protein-containing complex, and MCM complex (Figure 4D). KEGG pathway enrichment analysis showed that the differentially enriched MCM8-associated proteins were mainly related to cellular senescence, the calcium signaling pathway, oocyte meiosis, apoptosis, and endocrine resistance (Figure 4E). The protein interaction network analysis included 82 proteins, and the interaction score between MCM8 and MCM6 was the highest (0.948). RAI14 interacted with PPP1CC (0.532) and PPP1CB (0.538), whereas ETFB interacted with UQCRC2 (0.559) (Figure 4F, Table 4). Further validation revealed that MCM6 expression was upregulated in both wild-type and double-mutant cells during the DNA damage period (0 hours); during the repair period (2 hours), its expression decreased in wild-type cells but increased in double-mutant cells (Figure 4G). Bidirectional CoIP experiments supported the association between MCM8 and MCM6. In this assay, MCM6 showed greater co-immunoprecipitation with the double-mutant MCM8 construct than with wild-type MCM8 (Figures 4H, I). Detection of differentially interacting proteins of MCM8 by LC–MS/MS. (A) CoIP of wild-type and compound-mutant MCM8 in DNA repair. (B) Silver-stained gel images of wild-type and compound-mutant MCM8 in DNA repair. 1: empty vector; 2. wild-type MCM8; 3. double-mutant MCM8 (C242R-S445*). (C) Mass spectra of identified differentially expressed proteins. (D) Gene ontology (GO) analysis of differentially expressed proteins (DEPs) in both wild-type and double-mutant MCM8 cells on the basis of the CoIP/MS results. (E) Kyoto encyclopedia of genes and genomes (KEGG) analysis of differentially expressed proteins (DEPs) in both wild-type and double-mutant MCM8 cells on the basis of the CoIP/MS results. (F) String protein interaction network analysis. (G) Expression of MCM6 during the DNA damage repair period. (H) CoIP validation of candidate MCM8-interacting proteins using wild-type/mutant MCM8-flag as bait protein. (I) Reciprocal CoIP assay using MCM6 as the bait protein to validate its interaction with MCM8. Table 1Mass spectrum results of common differential proteins.Uniprot IDNames of differentially expressed protein genesThe number of peptides identified for the proteinThe number of unique peptides identified for the main proteinSequence coverage [%]Protein molecular weight [kDa]Full-length proteinQ-valueProtein scoreMutation/wild trendP06733ENO1202040.647.1684340187.3DOWNQ8N257H2BU14135.713.9081260.0053642.0877DOWNQ9BYG3NIFK6624.934.222293021.598DOWN P13804 MCM8 11 10 17.9 93.696 840 0 30.145 DOWN Q9UJA3Q9UJA3 MCM8-DM 3 3 8.1 49.788 444 0 30.145 UP Q9NUQ6SPATS2L335.761.72855809.7807DOWNP27816MAP4333.91211152010.046DOWNP38117ETFB101035.727.843255031.802UPP13804ETFA6627.635.079333020.439UP Q14566 MCM6 11 11 16.3 92.888 821 0 81.51 UP P11498PC121212.6129.631178078.231UPP0DP91ERCC6444.4119.49106109.2532UP Q9P0K7 RAI14 2 2 3.4 110.04 980 0 7.6174 UP P51531SMARCA2432.8181.281590013.04UPThe bold values represent proteins that we have verified in this study using Western blot and COIP experiments, as well as proteins associated with ovarian function. Mass spectrum results of common differential proteins. The bold values represent proteins that we have verified in this study using Western blot and COIP experiments, as well as proteins associated with ovarian function. Table 2Mass spectrum results of unique differential proteins.Uniprot IDNames of differentially expressed protein genesThe number of peptides identified for the proteinThe number of unique peptides identified for the main proteinSequence coverage[%]Protein molecular weight [kDa]Full-length proteinQ-valueProtein scoreQ96C19EFHD2161557.526.6972400110.62Q9NYL9TMOD3191951.739.594352090.013Q71UM5RPS27L3138.19.47718407.2983Q9UHB6LIMA1202035.785.225759074.773P38117 ETFB* 101035.727.843255031.802O00159MYO1C363532.9121.6810630130.16P12814ACTN1241632.3103.06892062.295P00558PGK19931.944.614417081.387P62937PPIA7730.918.012165041.061O95400CD2BP2992937.646341023.082P13804 ETFA 6627.635.079333020.439Q96IX5ATP5MK2227.66.457558013.55P49368CCT3101026.460.5335450128.74Q15459SF3A1202025.588.885793068.851P16402H1-39124.422.352210.00432.1922O00165HAX14421.131.62279015.904P42167TMPO6220.550.6745403.4101Q96EL2 MRPS24 331819.01516704.2992Q12797ASPH8817.985.862758025.037MCM8MCM8111018.988.7793030.145MCM8-DMMCM8338.149.788444030.145P41091EIF2S36617.451.109472090.795Q9BYD6 MRPL1 4416.336.908325010.609Q14566 MCM6 111116.392.888821081.51Q9NR30DDX218812.887.343783042.591P23511NFYA2212.136.876347019.957Q96D15RCN33311.637.493328018.399O15355PPM1G4411.559.271546016.43Q9H5V9STEEP12211.325.62422204.2999P05186ALPL6611.157.304524017.588P45880 VDAC2 3310.231.56629405.1019Q5TBB1RNASEH2B229.935.138312016.555Q8NE86 MCU 339.439.866351011.311P84098RPL19229.223.46619605.6146Q9Y580RBM722930.50326603.9184Q9UN86G3BP2328.754.1248207.3792Q13162PRDX4318.530.5427103.3313Q9Y2X3NOP58337.459.57852906.7345O60762DPM1226.929.63426004.6036P31153MAT2A335.843.66395013.237P22695 UQCRC2 225.748.442453034.246Q99848EBNA1BP2225.634.85230604.4183P0DP91ERCC6444.4119.49106109.2532P16278GLB1224.376.07467706.9092P48681NES663.5177.441621015.075Q9ULX6AKAP8L223.471.63964605.149Q9P0K7 RAI14 223.4110.0498007.6174P51531SMARCA2432.8181.281590013.04Q9Y490TLN1442.2269.76254108.2563Q9ULV0MYO5B322.1213.67184807.951O96028NSD2332.1152.261365012.799 Mass spectrum results of unique differential proteins. Table 3GO secondary classification results.OntologynameGo IDCountbiological_process cellular process GO:0050875154biological_process metabolic process GO:0044710125biological_process biological regulation GO:0065007101biological_processresponse to stimulusGO:005186971biological_processmulticellular organismal processGO:005087463biological_processlocalizationGO:190257854biological_process reproduction GO:005087611biological_process reproductive process GO:004470211cellular_componentcellular anatomical entityGO:0110165158cellular_component protein-containing complex GO:0043234105molecular_functionbindingGO:0005488151molecular_functioncatalytic activityGO:000382457molecular_function ATP-dependent activity GO:014065718 GO secondary classification results. Table 4Protein interaction PPi information.accession_node1accession_node2#node1node2combined_scoreQ14573Q8NE86ITPR3MCU0.898Q14573P45880ITPR3VDAC20.745 Q14566 MCM8;MCM8-amphimutation MCM6 MCM8 0.948 Q8NE86P45880MCUVDAC20.625P22695P45880UQCRC2VDAC20.846Q14566Q9Y6C9MCM6MTCH20.434P20700Q14566LMNB1MCM60.58Q14566O60264MCM6SMARCA50.69Q14566Q5TBB1MCM6RNASEH2B0.429 P62140 Q9P0K7 PPP1CB RAI14 0.538 P36873 Q9P0K7 PPP1CC RAI14 0.532 P38117 P22695 ETFB UQCRC2 0.559 Q14566Q15459MCM6SF3A10.45The bold values represent proteins that we have verified in this study using Western blot and COIP experiments, as well as proteins associated with ovarian function. Protein interaction PPi information. The bold values represent proteins that we have verified in this study using Western blot and COIP experiments, as well as proteins associated with ovarian function. Cut&Tag–mRNA-seq analysis of MCM8-associated chromatin-binding and transcriptomic changes To elucidate the function of double-mutant MCM8 in genomic binding, we performed the Cut&Tag assay and RNA-seq analyses on cells carrying the wild-type plasmid, the double-mutant MCM8 plasmid, or empty vector during the DNA repair period. The Cut&Tag assay mapped the chromatin binding profile of MCM8 across the entire genome (Figure 5A), with binding peaks distributed mainly in intergenic regions (Figure 5B). The double-mutant MCM8 had 7379 differential binding sites, among which 2074 showed increased binding and 5305 showed decreased binding. GO functional analysis indicated that the differentially genes were enriched mainly in mitochondrial transmembrane transport, mitochondrial membrane organization, and the cellular nitrogen compound metabolic process (Figure 5C). KEGG enrichment analysis revealed that the differentially genes bound to MCM8 were enriched mainly in ribosome biogenesis in eukaryotes, ubiquinone and other terpenoid-quinone biosynthesis, and the PPAR signaling pathway (Figure 5D). RNA-seq analysis revealed 760 differentially expressed mRNAs, including 637 upregulated and 123 downregulated mRNAs (Figures 5E, F). GO functional analysis indicated that the differentially expressed genes were enriched mainly in multicellular organism development, the development process, system development, cell development and the cellular developmental process (Figure 5G). KEGG enrichment analysis of the differentially expressed genes showed enrichment mainly in cell adhesion molecules and the PI3K–Akt signaling pathway (Figure 5H). By integrating the Cut&Tag binding sites and differentially expressed genes identified by RNA-seq, we identified 50 candidate genes showing both altered MCM8-associated binding and altered transcript levels (Figure 5I). Further combined analyses highlighted transcriptomic changes involving genes related to the PI3K/AKT signaling pathway in the double-mutant group (Figure 5J), and PPP2R5B expression increased in double-mutant group (Figure 5K). To verify the differentially expressed candidate genes of the PI3K/AKT signaling pathway, the relative mRNA expression levels were assessed using qRT–PCR. As shown in Supplementary Figure 1, the mRNA levels of PTEN, PIK3AP1, PPP2R5B, and CREB5 were significantly higher in double-mutant MCM8 group (DM) than in the control. These results indicate that double-mutant MCM8 was associated with chromatin-binding-related and transcriptomic changes involving candidate genes related to PI3K–AKT signaling. Identification of genes associated with altered MCM8 chromatin binding and differential expression by Cut&Tag–mRNA-seq analysis. (A) Genome-wide chromatin binding profile of MCM8. (B) Genomic mapping analysis of MCM8 binding sites. (C) Gene Ontology (GO) analysis of differentially expressed genes (DEGs) in both wild-type and double-mutant MCM8 cells on basis of the Cut&Tag assay results. (D) Kyoto encyclopedia of genes and genomes (KEGG) analysis of differentially expressed genes (DEGs) in both wild-type and double-mutant MCM8 cells on the basis of the Cut&Tag assay results. (E) Volcano plot showing differentially expressed genes (DEGs) between wild-type and double-mutant MCM8 cells based on RNA-seq analysis. (F) Heatmap showing differentially expressed genes (DEGs) between wild-type and double-mutant MCM8 cells based on RNA-seq analysis. (G) Gene Ontology (GO) analysis of differentially expressed genes (DEGs) in both wild-type and double-mutant MCM8 cells on the basis of the RNA-Seq results. (H) Kyoto encyclopedia of genes and genomes (KEGG) analysis of differentially expressed genes (DEGs) in both wild-type and double-mutant MCM8 cells on the basis of the RNA-Seq results. (I) Venn diagram of key genes in the Cut&Tag–RNA-seq cross-linking analysis. (J) Mutant MCM8 is enriched in POI-related signaling pathways according to Cut&Tag–RNA-seq cross-linking analysis. (K) Mutant MCM8 is enriched in POI-related genes according to Cut&Tag–RNA-seq cross-linking analysis. (n=3 biological replicates). Discussion In our previous clinical study, we reported a Chinese family that presented with pediatric-onset POI harboring two compound heterozygous variants (NM_001281522.1: c.724T>C [p.Cys242Arg] and c.1334C>A [p.Ser445*]) in the MCM8 gene (27). Clinically, the compound heterozygous variants presented as progressive ovarian atrophy in two affected sisters, marked by follicular depletion, hypergonadotropic hypogonadism, and premature apoptosis—features consistent with autosomal recessive POI, as previously documented (23, 27–30). Notably, the proband’s younger sister had normal uterine and ovarian volumes and FSH levels at 6 years and 7 months of age. However, at the age of 7 years and 10 months, ultrasound revealed that the bilateral ovaries gradually progressed from a normal size to an unclear display, accompanied by a decrease in the FSH level, and at follow-up at 11 years of age, the FSH level progressively manifested as hypergonadotrophic hypogonadism. These data indicated the age-dependent penetrance of MCM8-related defects and reinforced the importance of early genetic screening in families with a history of POI. Building upon these clinical findings, in the present study, we aimed to provide model-based in vitro observations regarding cellular changes associated with double-mutant MCM8. Our integrated cellular and multi-omics analyses using an artificial cis double-mutant overexpression model suggest that the co-occurrence of p.C242R and p.S445* in the same MCM8 construct is associated with delayed DNA damage resolution, altered recruitment of HR-related factors, changes in MCM8-associated protein interactions, and transcriptomic alterations. These in vitro observations may generate hypotheses regarding cellular processes potentially affected by MCM8 dysfunction, but they do not directly establish the physiological pathogenesis of POI in patients carrying compound heterozygous variants. The canonical cellular function of MCM8, which functions primarily as a heterodimer with MCM9, is to act as a DNA helicase that unwinds double-stranded DNA during homologous recombination. This process is indispensable for repairing DNA double-strand breaks (DSBs) and maintaining meiotic fidelity in gonadal tissues (18, 22). This observations are consistent with previous studies showing that MCM8 deficiency in mice causes defects in primordial germ cell proliferation and meiosis, ultimately leading to premature ovarian failure (31, 32). Structural modeling and functional assays suggested that the p.S445* truncation variant in the AAA domain may induce greater structural perturbation than the p.C242R variant in the zinc finger domain, which is consistent with a stronger impact on the DNA damage response in our model. Our in vitro assays revealed that cells expressing double-mutant MCM8 exhibited delayed resolution of DNA damage response. Persistent γ-H2AX accumulation suggests impaired or delayed DNA damage repair, while altered recruitment of recombination-associated factors, including RAD51 and DMC1, indicates disruption of homologous recombination-related processes. This unresolved DNA damage may contribute to S-phase accumulation and cell apoptosis. While MCM8 is canonically known to form a functional heterodimer with MCM9, our CoIP-MS/MS screening did not identify MCM9 as a significantly altered interactor between the wild-type and double-mutant MCM8. Because the canonical MCM8-MCM9 interaction was not significantly disrupted according to our differential screening criteria, our interactome data unexpectedly directed our focus toward a profound shift in the interaction between the mutant MCM8 and MCM6. MCM6 is an essential core subunit of the MCM2–7 replicative helicase complex, which is strictly required for DNA replication origin licensing and replication fork progression (33). The altered MCM6 association observed in our interactome data led us to hypothesize that mutant MCM8 may affect MCM2–7 complex dynamics indirectly or through aberrant protein association. Such altered interactions could potentially affect replication-associated processes and contribute to replication stress in this overexpression system. Consistent with this possibility, changes in apoptosis-related markers were observed. However, we acknowledge that the precise biophysical consequences of this aberrant MCM8–MCM6 binding remain highly speculative. In the current experimental design, no direct causal link between this disrupted MCM8-MCM6 interaction and the observed apoptotic phenotype has been strictly established. Whether altered MCM6 association directly contributes to replication stress and apoptosis in these cells, or instead represents a parallel consequence of the altered conformation of the double-mutant protein, remains unknown. Moreover, the lack of targeted validation of the MCM8–MCM9 interaction in the present study limits a comprehensive interpretation of the homologous recombination-related phenotype. Beyond its role in DNA repair, our Cut&Tag and RNA-seq analyses revealed that the MCM8 mutations were associated with altered chromatin binding profiles and transcriptomic changes. It is crucial to clarify that our data do not establish MCM8 as a direct, classical transcription factor. Rather, we hypothesize that the altered chromatin occupancy of the mutant MCM8 protein may indirectly influence transcriptomic regulation. Notably, the candidate genes identified in our multi-omics intersection analysis, including PIK3AP1, CREB5, PPP2R5B, and PTEN, were related to the PI3K/AKT signaling pathway. The PI3K/AKT cascade is an important regulator of primordial follicle activation, survival, and metabolic homeostasis. Despite providing valuable functional annotations for these pathogenic variants, our findings must be interpreted with caution because of several substantial methodological limitations. First, our cellular model relied on HeLa cells, a highly aneuploid cervical cancer cell line whose baseline transcriptomic networks, cell cycle checkpoints, and chromatin landscapes differ from those of primary human ovarian granulosa cells. Extrapolating these in vitro findings to human ovarian biology is inherently limited. Second, the mutant constructs were transiently overexpressed in the presence of endogenous wild-type MCM8, which may introduce spurious protein–protein interactions because of altered stoichiometry. Third, our in vitro 33experimental design used a single expression construct harboring both variants (in cis), which inherently generates an artificial double-mutant protein. We acknowledge that this experimental strategy does not reproduce the endogenous compound heterozygous configuration (in trans) observed on patients’ homologous chromosomes. We recognize that accurately modeling true biallelic compound heterozygosity within standard in vitro cell models remains a technical challenge. While co-transfecting two separate single-mutant plasmids was considered as an alternative to mimic the trans configuration, such transient expression systems are limited by stochastic plasmid uptake. Due to these current technical limitations, such systems cannot reliably reproduce physiological 1:1 stoichiometric expression of both mutant alleles at the single-cell level, nor can they replicate endogenous chromosomal regulatory mechanisms. Therefore, our in vitro system does not model the patient’s precise physiological phenotype but rather serves as a related but distinct artificial model to explore the combined functional consequences of these specific variant combinations in an overexpression setting. Fourth, the lack of assessment of the MCM8–MCM9 heterodimer leaves a critical gap in our mechanistic understanding. Finally, comparing apoptotic markers between patient-derived streak gonads (which are highly fibrotic tissue devoid of functional follicles) and functional normal ovarian margins is subject to inherent biological bias. In conclusion, this study reports the clinical association of pediatric-onset POI with compound heterozygous MCM8 variants, p.C242R and p.S445*, and provides an in vitro functional characterization using an artificial double-mutant MCM8 overexpression construct. Rather than defining the physiological pathogenesis of the patients, our findings show that expression of this double-mutant construct is associated with several cellular abnormalities in vitro. First, cells expressing the double-mutant construct showed delayed resolution of DNA damage, persistent γ-H2AX accumulation, and altered recruitment of HR-related factors, together with S-phase accumulation. Second, altered interaction between double-mutant MCM8 and MCM6 may reflect changes in replication-associated protein complexes and may be related to replication stress in this overexpression system. Third, integrated Cut&Tag and RNA-seq analyses identified MCM8-associated chromatin-binding changes and transcriptomic alterations involving candidate genes in PI3K/AKT signaling pathway (Figure 6). Collectively, these findings provide model-based cellular insights that may help generate hypotheses regarding how MCM8 dysfunction affects DNA damage repair, cell-cycle progression and apoptosis. To validate these molecular mechanisms and accurately model compound heterozygosity in a physiological context, future studies employing more physiological models, such as biallelic CRISPR/Cas9-mediated knock-in systems or granulosa-like cells, will be required to validate the physiological relevance of these findings. Schematic model of cellular abnormalities associated with expression of double-mutant MCM8 construct in vitro. Expression of the artificial cis double-mutant MCM8 construct harboring p.C242R and p.S445* was associated with altered MCM8–MCM6 interaction, persistent γ-H2AX accumulation, impaired recruitment of homologous recombination-related factors, including DMC1 and RAD51, delayed DNA damage resolution, S-phase accumulation/cell-cycle arrest, and increased apoptotic signaling in vitro. These cellular abnormalities may provide model-based clues regarding the potential effects of MCM8 dysfunction on DNA repair and cell survival.