Identification and functional characterization of a SUZ12 frameshift variant in a Chinese family with Imagawa-Matsumoto syndrome Authors: Xin Xu, Hong Xu, Songming Huang, Yikang He, Ling Zhang Published: 2026 ABSTRACT AbstractBackgroundImagawa-Matsumoto syndrome (IMMAS) is a rare overgrowth disorder caused by heterozygous loss-of-function variants in SUZ12, which encodes a core subunit of the Polycomb Repressive Complex 2 (PRC2). To date, fewer than 20 cases have been reported, and the molecular mechanisms underlying PRC2 dysfunction in IMMAS remain incompletely understood.MethodsTrio-based whole-exome sequencing (Trio-WES) was performed in a Chinese family presenting with overgrowth and dysmorphic features. The candidate variant was validated by Sanger sequencing and segregation analysis. AlphaFold 3 was used to predict the structural consequences of the variant. In vitro overexpression and protein assays were conducted to evaluate SUZ12 protein expression and subcellular localization.ResultsThe proband exhibited global developmental delay, macrocephaly, hypertrichosis, and distinctive craniofacial dysmorphism. Trio-WES identified a heterozygous frameshift variant in SUZ12 (c.1244_1248del; p.Glu415GlyfsTer5). Computational modeling predicted that the truncating variant removes the ZnF and VEFS domains and may affect the predicted SUZ12–EZH2 interface. In vitro expression of the mutant construct confirmed the production of a truncated SUZ12 protein. Furthermore, peripheral blood mononuclear cells (PBMCs) from both the proband and her mother showed a substantial global reduction in H3K27me3 levels, consistent with impaired PRC2-mediated histone methylation in vivo.ConclusionsWe report a Chinese family with Imagawa-Matsumoto syndrome harboring a SUZ12 frameshift variant (c.1244_1248del; p.Glu415GlyfsTer5). Our study expands the phenotypic spectrum associated with this variant and provides evidence consistent with impaired PRC2-mediated histone methylation.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12920-026-02431-5.Keywords: Imagawa-Matsumoto syndrome, SUZ12, PRC2, H3K27me3, Overgrowth Introduction Imagawa-Matsumoto syndrome (IMMAS; MIM: 618786) is a rare overgrowth disorder caused by heterozygous pathogenic variants in SUZ12, a gene located on chromosome 17q11.2. SUZ12 encodes a core subunit of the Polycomb Repressive Complex 2 (PRC2), which catalyzes trimethylation of histone H3 at lysine 27 (H3K27me3), a key repressive histone mark involved in chromatin-mediated gene silencing during embryonic and postnatal development [1–3]. Loss of SUZ12 function has been shown to impair PRC2 stability and disrupt transcriptional repression of developmental genes [4, 5]. IMMAS was recognized as a distinct overgrowth disorder following the identification of germline SUZ12 variants. The main clinical features include postnatal overgrowth, macrocephaly, characteristic facial dysmorphism, and skeletal abnormalities [6, 7]. Subsequent reports expanded the phenotypic spectrum to include developmental delay and/or intellectual disability (DD/ID), together with a broader range of facial features, including a round face, broad forehead or increased bifrontal diameter, hypertelorism, downslanting palpebral fissures, and a broad or low nasal bridge [7–10]. Other variable findings include clinodactyly, advanced bone age, and structural brain abnormalities [7, 9–11]. Marked phenotypic variability has been observed, even among affected members of the same family carrying the same SUZ12 variant [6, 9–11]. Although fewer than 20 affected individuals have been reported to date, available evidence supports SUZ12 haploinsufficiency as the main disease mechanism [7, 9–11, 13]. Since the first description of IMMAS in a patient who inherited a SUZ12 variant from a mosaic father [6], additional de novo and familial cases have expanded the phenotypic spectrum and supported autosomal dominant inheritance with variable expressivity [7, 9–11]. Furthermore, reported cases include mosaic parental transmission and large copy‑number deletions encompassing SUZ12, further highlighting the complexity of genotype–phenotype correlations [6, 10, 12, 13]. In this study, we describe a Chinese family with Imagawa-Matsumoto syndrome carrying a SUZ12 frameshift variant, c.1244_1248del (p.Glu415GlyfsTer5). We performed structural and functional analyses to establish the molecular diagnosis and to further investigate the pathogenic consequences of this variant in the family. Our findings expand the clinical and molecular spectrum of IMMAS and provide additional functional evidence consistent with impaired PRC2-mediated histone methylation. Materials and methods Patients and ethical approval A Chinese family with clinical features suggestive of Imagawa-Matsumoto syndrome (IMMAS) was enrolled in this study. The proband, a girl aged 2 years and 3 months, was evaluated at the Children’s Hospital of Nanjing Medical University in 2024. The study was approved by the Institutional Review Board of the Children’s Hospital of Nanjing Medical University (Approval No. 202510036-1). Written informed consent was obtained from the parents for participation, publication of clinical and genetic data, and use of identifiable photographs. Peripheral blood samples were collected from the proband and her parents for genetic and functional studies. Whole-exome sequencing and variant analysis Genomic DNA was extracted from peripheral blood samples obtained from the proband and both parents using the TIANamp Genomic DNA Kit (Tiangen, China). Trio-based whole-exome sequencing was performed in the proband and her parents. Exome capture was carried out using the xGen Exome Research Panel v2.0, and 150 bp paired-end sequencing was performed on the Illumina NovaSeq 6000 platform, with a mean coverage depth of > 100×. Reads were aligned to the human reference genome (GRCh37/hg19) using Burrows-Wheeler Aligner, and variant calling was performed according to the Genome Analysis Toolkit best-practice workflow. Variants were annotated using ANNOVAR and InterVar and filtered against population databases, including gnomAD, ExAC, and the 1000 Genomes Project. Variant interpretation was performed according to the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP) guidelines [14]. The SUZ12 variant identified by whole-exome sequencing was confirmed by Sanger sequencing. Structural analysis The amino acid sequence of human SUZ12 (UniProt ID: Q15022) was obtained from UniProt. AlphaFold 3 was used to predict the structures of wild-type SUZ12 and the p.Glu415GlyfsTer5 truncated protein. Protein–protein interaction modeling of SUZ12 and EZH2 was performed using AlphaFold-Multimer, and the top-ranked model was selected for visualization. Structural models were analyzed using PyMOL version 3.0.3. The N-terminal portion retained in the mutant protein, the Glu415 frameshift site, the ZnF domain (residues 448–471), and the VEFS domain (residues 563–639) were annotated. The models were used to illustrate the predicted structural consequences of the variant and were not considered direct experimental evidence of impaired SUZ12–EZH2 binding or PRC2 assembly. To assess the consistency of the predicted model with experimentally determined structures, the AlphaFold 3-predicted full-length wild-type SUZ12 model was compared with the experimentally resolved SUZ12 regions in PDB 6NQ3 (chain B, residues 76–545) and PDB 4W2R (chain T, residues 545–725). The corresponding structured regions were superimposed using the “align” command in PyMOL. Only experimentally resolved SUZ12 regions were included in the comparison, whereas unresolved or intrinsically disordered regions were excluded. PBMC isolation Peripheral blood mononuclear cells (PBMCs) were isolated from 5 mL of whole blood obtained from the proband, her affected mother, and her unaffected father by Ficoll density-gradient centrifugation. After centrifugation at 1200 × g for 15 min at 20 °C, the mononuclear cell layer was collected and washed twice with RPMI-1640 containing 10% fetal bovine serum. Cells were then pelleted at 750 × g for 5 min at 4 °C and resuspended for further analysis. Plasmid construction and cell culture The full-length coding sequence of human SUZ12 (NM_015355.4) was cloned into pcDNA3.1(+) to generate an N-terminal FLAG-tagged construct. The mutant construct carrying c.1244_1248del (p.Glu415GlyfsTer5) was generated by site-directed mutagenesis and confirmed by Sanger sequencing. HEK293 cells were cultured in Dulbecco’s modified Eagle medium supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin at 37 °C in 5% CO₂. Transient transfection was performed using Lipofectamine 3000 according to the manufacturer’s instructions. Cells were harvested 48 h after transfection. Western blotting Total protein was extracted in RIPA buffer containing protease inhibitors. Protein concentration was measured using a bicinchoninic acid assay. Equal amounts of protein were separated by 10% SDS-PAGE and transferred to polyvinylidene difluoride membranes. After blocking with 5% non-fat milk, membranes were incubated overnight at 4 °C with primary antibodies against FLAG (Sigma-Aldrich, F3165; 1:1000), H3K27me3 (Cell Signaling Technology, 9733; 1:1000), histone H3 (Abcam, ab1791; 1:1000), and GAPDH (Proteintech, 10494-1-AP; 1:5000). After incubation with horseradish peroxidase-conjugated secondary antibodies, signals were detected by enhanced chemiluminescence. Band intensity was quantified using ImageJ. For PBMC-derived H3K27me3 analysis, Western blotting was repeated three times using extracts from the same family members. Immunofluorescence Transfected HEK293 cells were fixed in 4% paraformaldehyde for 15 min, permeabilized with 0.2% Triton X-100, and blocked with 5% bovine serum albumin. Cells were incubated overnight at 4 °C with anti-FLAG antibody (Sigma-Aldrich, F3165; 1:200), followed by Alexa Fluor-conjugated secondary antibodies. Nuclei were counterstained with DAPI. Images were acquired using a Leica STELLARIS 5 confocal microscope under identical imaging settings. Statistical analysis Experiments were performed with at least three independent replicates unless otherwise stated. Data are presented as mean ± SD. Comparisons between two groups were made using two-tailed Student’s t tests. A P value < 0.05 was considered statistically significant. Statistical analyses were performed using GraphPad Prism 9. Results Case presentation The proband was a girl aged 2 years and 3 months who was referred for evaluation of global developmental delay. She was born at term by cesarean section due to suspected fetal macrosomia after an otherwise uncomplicated pregnancy, with a birth weight of 3.98 kg (> 90th percentile). Developmental milestones were delayed. She achieved independent walking at 14 months. Language development was also affected, and at 2 years of age her expressive speech was limited to monosyllabic words. She was unable to identify colors or follow simple verbal commands. Formal developmental assessment at 27 months confirmed global developmental delay, with gross motor skills at the 24-month level, social adaptability at the 18-month level, and cognitive performance at the 18-month level. The overall developmental quotient was 66, and the mental index was 57. On examination, she had tall stature (height 98.0 cm, + 2.67 SD) and macrocephaly (head circumference 51.5 cm, + 2.80 SD). Craniofacial features included a round face, bitemporal narrowing, hypertelorism, prominent supraorbital ridges, thick laterally flared eyebrows with sparse lateral thirds, long palpebral fissures, a horizontal chin crease, a broad, depressed nasal bridge, and large, prominent ears (Fig. 1B, C). Hypertrichosis was most evident over the forehead and limbs, with lower-limb hypertrichosis shown in Fig. 1D. Mild camptodactyly was observed in both fifth fingers (Fig. 1E). The toes were normal, with no obvious camptodactyly, clinodactyly, or overlapping toes. Apart from mild camptodactyly of both fifth fingers, there was no evidence of other limb contractures, scoliosis, or joint hypermobility. Brain magnetic resonance imaging showed bilateral periventricular T2 hyperintensities without major structural abnormalities (Fig. 1F). Clinical features of the affected mother and proband with Imagawa-Matsumoto syndrome. A Frontal facial photograph of the affected mother, showing hypertelorism, a broad forehead, flat nasal bridge, midfacial retrusion, and coarse facial appearance. B, C Frontal and lateral facial photographs of the proband, showing a round face, bitemporal narrowing, hypertelorism, a broad, depressed nasal bridge, and large, prominent ears. D Hypertrichosis over the lower limbs. E Mild camptodactyly of both fifth fingers. F Brain MRI showing bilateral periventricular T2 hyperintensities without major structural abnormalities The proband’s mother was also tall and showed craniofacial features overlapping with those of the proband, including a broad forehead, hypertelorism, a flat nasal bridge, midfacial retrusion, and coarse facial appearance (Fig. 1A). No formal neurodevelopmental or cognitive assessment was available for the mother. The father had no remarkable clinical features. The maternal grandfather was reportedly tall (185 cm) but died before genetic evaluation could be performed. No other family members were known to have developmental delay, overgrowth, or congenital anomalies. Molecular diagnosis Trio-WES identified a heterozygous frameshift variant in SUZ12 (NM_015355.4), c.1244_1248del (p.Glu415GlyfsTer5), in exon 10 (Fig. 2A). No other pathogenic or likely pathogenic variants associated with overgrowth or neurodevelopmental disorders were identified. Sanger sequencing confirmed maternal inheritance of the variant (Fig. 2B). The origin of the variant in the mother could not be determined because samples from the maternal grandparents were not available. The variant was absent from population databases, including gnomAD, ExAC, and the 1000 Genomes Project. It has been submitted to ClinVar (SCV001999392.3; RCV001765755.3), where it is currently classified as likely pathogenic. According to the ACMG/AMP criteria, the variant met PVS1, PM2, PP1_Moderate, and PP4, supporting its classification as pathogenic in this family. Mapping of previously reported pathogenic SUZ12 variants onto the canonical protein structure showed that c.1244_1248del (p.Glu415GlyfsTer5) lies upstream of the VEFS domain, within the region in which pathogenic truncating variants associated with IMMAS are concentrated (Fig. 3). Genetic findings in the family with Imagawa-Matsumoto syndrome. A Pedigree showing maternal inheritance of the SUZ12 c.1244_1248del variant. The arrow indicates the proband; filled symbols indicate affected individuals; and a diagonal slash indicates a deceased individual. B Sanger sequencing chromatograms showing the heterozygous SUZ12 variant, c.1244_1248del (p.Glu415GlyfsTer5), in the proband and her mother. The father carries the wild-type sequence. The arrow indicates the variant site Distribution of IMMAS-associated SUZ12 variants. Schematic representation of the SUZ12 protein showing previously reported pathogenic or likely pathogenic variants associated with Imagawa-Matsumoto syndrome and the variant identified in the present study. The ZnF domain and VEFS domain are indicated according to the canonical SUZ12 protein sequence. Previously reported variants summarized in Table 1 are shown in black, and the variant identified in this study, p.Glu415GlyfsTer5, is highlighted in red. Large copy-number deletions are not shown. This figure was modified from Imagawa et al. (2023) [10] Structural effect of the SUZ12 variant To examine the predicted structural consequences of p.Glu415GlyfsTer5, we performed computational modeling using AlphaFold 3 and AlphaFold-Multimer. To evaluate the reliability of the predicted structured regions, we compared the AlphaFold 3-predicted full-length wild-type SUZ12 model with experimentally resolved SUZ12 regions in PDB 6NQ3 and 4W2R. The experimentally resolved SUZ12 regions showed overall agreement with the corresponding AlphaFold 3-predicted regions, with minor local deviations near flexible loop regions (Supplementary Fig. S1). In the wild-type SUZ12 model, the N-terminal portion, Glu415, the ZnF domain (residues 448–471), and the VEFS domain (residues 563–639) were annotated. In contrast, the p.Glu415GlyfsTer5 model retained the N-terminal portion preceding the frameshift site but lacked the downstream ZnF and VEFS domains because of premature truncation (Fig. 4A). AlphaFold-Multimer modeling of SUZ12 and EZH2 suggested that the VEFS domain may contribute to the predicted SUZ12–EZH2 interface (Fig. 4B). The N-terminal portion of SUZ12, Glu415, the ZnF domain, and the VEFS domain were highlighted in the interaction model. These computational findings illustrate the predicted structural consequences of the truncation but do not directly demonstrate impaired SUZ12–EZH2 binding or disrupted PRC2 assembly. Structural modeling of the SUZ12 p.Glu415GlyfsTer5 variant. A AlphaFold 3-predicted structures of wild-type and mutant SUZ12. The N-terminal region, Glu415, ZnF domain (448–471), and VEFS domain (563–639) are indicated. The mutant protein retains the N-terminal region but lacks the ZnF and VEFS domains. B AlphaFold-Multimer model of the SUZ12–EZH2 complex. The N-terminal region, Glu415, ZnF domain, and VEFS domain are highlighted. The model suggests that the VEFS domain may contribute to the predicted SUZ12–EZH2 interface but does not directly demonstrate impaired binding Functional characterization of the SUZ12 variant To assess the functional effect of c.1244_1248del (p.Glu415GlyfsTer5), wild-type and mutant SUZ12 constructs were transiently expressed in HEK293 cells. Western blotting detected a lower-molecular-weight SUZ12 band only in cells expressing the mutant construct, confirming the production of a truncated protein (Fig. 5A). Immunofluorescence showed that both wild-type and mutant SUZ12 proteins were predominantly nuclear, with no obvious differences in subcellular localization (Fig. 5B). PBMCs from the proband and her mother showed markedly reduced global H3K27me3 levels, consistent with impaired PRC2-mediated histone methylation in vivo (Fig. 6A, B). Together, these findings indicate that c.1244_1248del produces a truncated SUZ12 protein lacking the downstream ZnF and VEFS domains and support a deleterious effect of this variant on PRC2-mediated histone methylation. Functional characterization of the SUZ12 p.Glu415GlyfsTer5 variant. A Western blot analysis of HEK293 cells transiently expressing FLAG-tagged wild-type (WT) SUZ12 or the p.Glu415GlyfsTer5 mutant construct. The mutant construct produced a lower-molecular-weight band consistent with a truncated protein. GAPDH was used as a loading control. B Immunofluorescence staining of FLAG-SUZ12 (green) in transfected HEK293 cells. Nuclei were counterstained with DAPI (blue). Both WT and mutant SUZ12 showed predominantly nuclear localization. Scale bar, 25 μm Reduced H3K27me3 levels in patient-derived cells. A Western blot analysis of PBMC protein extracts obtained from the proband, her affected mother, and her unaffected father. H3K27me3 levels were reduced in both affected individuals compared with the unaffected father. Total histone H3 was used as a loading control. B Quantification of H3K27me3 normalized to total H3. Quantification was based on three repeated Western blot experiments using extracts from the same family members. Data are presented as mean ± SD. Statistical significance was assessed using a two-tailed Student’s t test (****P < 0.0001) Discussion In this study, we identified a heterozygous SUZ12 frameshift variant, c.1244_1248del (p.Glu415GlyfsTer5), in a Chinese family with clinical features consistent with Imagawa-Matsumoto syndrome (IMMAS). This variant is predicted to produce a truncated protein lacking the downstream ZnF and VEFS domains and was classified as pathogenic according to ACMG/AMP criteria. Taken together, the genetic, structural, and functional findings support a deleterious effect of this variant on PRC2-mediated H3K27 methylation and further expand the clinical and molecular spectrum of IMMAS. To date, fewer than 20 individuals with pathogenic SUZ12 variants and well-characterized IMMAS phenotypes have been reported [6–12]. Review of these previously reported cases suggests that IMMAS has a recognizable but variable phenotype (Table 1). Common features include postnatal overgrowth, macrocephaly, and mild-to-moderate developmental delay or intellectual disability [11]. Craniofacial features such as a broad forehead, hypertelorism, and a broad or low nasal bridge are also common and were present in both the proband and her mother. Hypertrichosis, which was observed in both affected individuals in this family and has also been described in previous reports, may represent a useful clinical clue to SUZ12-related overgrowth. Taken together, the proband’s clinical features are in keeping with the clinical spectrum of previously reported SUZ12-related IMMAS cases, including overgrowth, macrocephaly, developmental delay, hypertelorism, a low and broad nasal bridge, hypertrichosis, and digital anomalies. The present family further contributes to the clinical and molecular characterization of SUZ12-related IMMAS by documenting a maternally inherited SUZ12 frameshift variant in a Chinese family, together with detailed craniofacial, digital, and neuroimaging findings. Table 1Clinical and molecular features of previously reported patients with Imagawa-Matsumoto syndrome associated with SUZ12 variantsPatientAge/SexVariant(NM_015355.4)VarianttypeInheritanceMacrocephaly(HC ≥ + 2 SD)Overgrowth(height ≥ + 2 SD)DDIDDysmorphicfeaturesSource (year)P111y/Fc.1829 A > T; p.Glu610ValMissensePaternal (mosaic)+ (+ 5.5 SD)+ (+ 4.5 SD)+ModerateRound face, increased bifrontal diameter, hypertelorism, plagiocephalyImagawa et al. 2017 [6]P29y/Fc.844_845del;p.Ala282GlnfsTer7FrameshiftDe novo+ (+ 2.0 SD)+ (+ 2.5 SD)−−Round face, increased bifrontal diameterImagawa et al. 2018 [7]P319y/Mc.1797 A > C;p.Gln599HisMissenseUnknown+ (+ 5.8 SD)+ (+ 3.5 SD)+−Round face, increased bifrontal diameter, hypertelorism, downslanting palpebral fissuresImagawa et al. 2018 [7]P427y/Mc.1807T > C;p.Phe603LeuMissenseMaternal (mosaic)+ (+ 6.9 SD)+ (+ 5.9 SD)−−Round face, brachycephaly, downslanting palpebral fissures, hypertelorism, low/broad nasal bridgeCyrus et al. 2019 [8, 9]P518y/Mc.1878delT;p.Phe626LeufsTer7FrameshiftDe novo+ ( > + 2.0 SD)+−−PresentCyrus et al. 2019 [8, 9]P611y2m/Fc.1715_1716insCA; p.Leu572PhefsTer11FrameshiftDe novo− (+ 1.01 SD)−+MildIncreased bifrontal diameter, downslanting palpebral fissures, hypertelorism, low/broad nasal bridgeCyrus et al. 2019 [8, 9]P77y7m/Mc.1960 C > T;p.Arg654TerNonsenseDe novo− (+ 0.26 SD)−+MildRound face, hypertelorism, plagiocephaly, low/broad nasal bridgeCyrus et al. 2019 [8, 9]P85y1m/Mc.1826dupG;p.Glu610ArgfsTer20FrameshiftDe novo− (+ 1.88 SD)−+MildRound face, low/broad nasal bridge, hypertelorism, downslanting palpebral fissuresCyrus et al. 2019 [8, 9]P911y/Fc.348_351del;p.Tyr117CysfsTer24FrameshiftUnknown− (+ 0.14 SD)−+BorderlineRound face, hypertelorism, low/broad nasal bridgeCyrus et al. 2019 [8, 9]P1015y/Mc.1794 + 1G > ASpliceDe novo+ (+ 4.0 SD)+ (+ 2.6 SD)+−Downslanting palpebral fissures, hypertelorismCyrus et al. 2019 [8, 9]P1119y/Mc.275–5 C > GSpliceMaternal+ (+ 4.1 SD)−NAMildIncreased bifrontal diameter, brachycephaly, downslanting palpebral fissures, hypertelorism, low/broad nasal bridgeCyrus et al. 2019 [8, 9]P1211y1m/Fc.1604G > A;p.Arg535GlnMissenseUnknown− (− 1.59 SD)−+SevereLow/broad nasal bridgeCyrus et al. 2019 [8, 9]P1310y/Fc.1023 + 1G > CSpliceDe novo− (− 2.9 SD)−+MildRound face, low/broad nasal bridge, hypertelorism, downslanting palpebral fissuresImagawa et al. 2023 [10]P1419y/F17q11.2 deletionCNV deletionDe novo+ (+ 2 SD)−++Prominent forehead, round face, broad nasal ridge, flat feet, large hands, camptodactyly, clinodactylyYücel et al. 2024 [11]Abbreviations: NA not assessed or not known, y years, m months, DD developmental delay, ID intellectual disability, F female, M male, SD standard deviation, HC head circumference Clinical and molecular features of previously reported patients with Imagawa-Matsumoto syndrome associated with SUZ12 variants c.844_845del; p.Ala282GlnfsTer7 c.1797 A > C; p.Gln599His c.1807T > C; p.Phe603Leu c.1878delT; p.Phe626LeufsTer7 c.1960 C > T; p.Arg654Ter c.1826dupG; p.Glu610ArgfsTer20 c.348_351del; p.Tyr117CysfsTer24 c.1604G > A; p.Arg535Gln Abbreviations: NA not assessed or not known, y years, m months, DD developmental delay, ID intellectual disability, F female, M male, SD standard deviation, HC head circumference Most reported pathogenic SUZ12 variants are truncating or splice-altering variants located upstream of or within the VEFS domain [6–13], underscoring the importance of the C-terminal VEFS-containing region for PRC2 assembly and function [15, 16]. The c.1244_1248del variant identified in this family introduces a premature termination codon upstream of both the ZnF and VEFS domains and is therefore likely to trigger nonsense-mediated mRNA decay, supporting SUZ12 haploinsufficiency as the major pathogenic mechanism in vivo [10]. Structural modeling was used to illustrate the predicted loss of the downstream ZnF and C-terminal VEFS domains and the potential impact of VEFS loss on the predicted SUZ12–EZH2 interface, rather than to demonstrate impaired SUZ12–EZH2 binding or prove the endogenous production of a stable truncated protein. Comparison with available experimental SUZ12 structures showed that the AlphaFold 3-predicted structured regions were generally consistent with experimentally resolved regions. However, the available experimental structures cover only selected portions of SUZ12 and do not provide continuous atomic coordinates for the full-length protein. AlphaFold 3 therefore provides a complementary representation of regions that are unresolved experimentally. Nevertheless, flexible or intrinsically disordered regions should be interpreted cautiously because they remain computational predictions rather than experimentally validated stable conformations. If a stable truncated protein is produced, its retained N-terminal portion may preserve some PRC2-associated function through accessory components, potentially contributing to the relatively mild and variable phenotype of SUZ12-associated IMMAS. Consistent with impaired PRC2-mediated histone methylation, PBMCs from the proband and her mother showed markedly reduced global H3K27me3 levels. Together, these findings support SUZ12 haploinsufficiency and reduced PRC2-mediated H3K27me3 deposition as the likely disease mechanism in this family. IMMAS shares clinical features with other PRC2-related overgrowth disorders, including Weaver syndrome caused by EZH2 variants and Cohen-Gibson syndrome caused by EED variants [17–19]. These disorders underscore the central role of PRC2 in developmental gene regulation and postnatal growth [20, 21]. Genotype–phenotype relationships may differ according to the affected PRC2 subunit and variant type. EZH2 encodes the catalytic subunit of PRC2, whereas EED is involved in PRC2 regulation and H3K27me3 propagation. In contrast, SUZ12-related IMMAS is mainly associated with truncating variants, splice variants, or deletions that are predicted to cause haploinsufficiency or affect the VEFS domain important for PRC2 integrity. Compared with EZH2- and EED-related disorders, SUZ12-associated disease appears to show variable and often milder somatic overgrowth, generally mild-to-moderate neurodevelopmental involvement, and recurrent hypertrichosis. These observations suggest gene-specific effects within PRC2, although genotype–phenotype correlations remain limited by the small number of reported SUZ12 cases. Although PRC2 dysfunction and SUZ12 alterations have been implicated in cancer biology [22], whether germline SUZ12 variants in IMMAS confer an increased cancer risk remains unclear because of the limited number of reported cases. Long-term follow-up of additional patients will be needed to clarify this issue. SUZ12 is required for PRC2-mediated repression of developmental regulators. In mice, complete loss of Suz12 causes early embryonic lethality, underscoring the essential role of PRC2 in development [4]. SUZ12 deficiency has also been shown to disrupt bivalent chromatin domains, shifting the balance from repressive H3K27me3 toward the active H3K4me3 and thereby promoting premature activation of lineage-specific genes [5]. The marked reduction in H3K27me3 observed in cells from our patients provides functional support for impaired PRC2 activity and links epigenetic dysregulation to the overgrowth and neurodevelopmental features of IMMAS. One limitation of this study is that endogenous SUZ12 protein levels in patient-derived PBMCs were not directly assessed because of the limited availability of patient-derived material. Therefore, we focused on global H3K27me3 levels as a downstream functional readout of PRC2-mediated histone methylation. Future studies using co-immunoprecipitation or other PRC2-complex assembly assays will be required to directly determine whether the p.Glu415GlyfsTer5 variant impairs SUZ12–EZH2 interaction or PRC2 assembly. Conclusion We describe a Chinese family with IMMAS carrying a SUZ12 frameshift variant, c.1244_1248del (p.Glu415GlyfsTer5). This variant produced a truncated protein in vitro and was associated with markedly reduced H3K27me3 levels in patient-derived cells. These findings support SUZ12 haploinsufficiency as a likely mechanism contributing to impaired PRC2-mediated histone methylation in this disorder. Our study further expands the clinical and molecular spectrum of IMMAS and highlights reduced H3K27me3 deposition as a key feature of SUZ12-related overgrowth. Further studies are needed to define how PRC2 dysfunction leads to the specific developmental phenotype. Supplementary Information Supplementary Material 1: Supplementary Figure S1. Comparison of the AlphaFold 3-predicted SUZ12 model with experimentally resolved SUZ12 structures. (A) Superposition of the AlphaFold 3-predicted full-length wild-type SUZ12 model with the experimentally resolved SUZ12 region in PDB 6NQ3 (chain B, residues 76–545). (B) Superposition of the AlphaFold 3-predicted full-length wild-type SUZ12 model with the experimentally resolved SUZ12 region in PDB 4W2R (chain T, residues 545–725). Only experimentally resolved SUZ12 regions were included in the comparisons. Supplementary Material 2: Supplementary Figure S2. Full-length uncropped blot images for Figures 5A and 6A. Acknowledgements We thank the family members for their participation in and support of this study. Abbreviations Imagawa-Matsumoto syndrome Whole-exome sequencing Suppressor of zeste 12 homolog Polycomb repressive complex 2 Enhancer of zeste homolog 2 Histone H3 lysine 27 trimethylation Peripheral blood mononuclear cells Authors’ contributions XX collected clinical data and drafted the manuscript. XX and HX performed the experiments. YH performed data analysis. SH and LZ conceived and supervised the study, critically reviewed the manuscript. All authors contributed to the acquisition and interpretation of data, revised the manuscript, and approved the final version for submission. Funding This work was supported by a grant from the Science and Technology Development Fund of Nanjing Medical University (No. NMUB2020091). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Data availability The datasets generated and/or analysed during the current study are not publicly available because they contain information that could compromise participant privacy. Raw de-identified data may be made available from the corresponding authors upon reasonable request and subject to approval by the institutional ethics committee. The SUZ12 variant has been deposited in ClinVar (Accession SCV001999392.3). Declarations Ethics approval and consent to participate This study was approved by the Ethics Committee of the Children’s Hospital of Nanjing Medical University (No. 202510036-1) and was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from both adult participants and from the parents/legal guardians of the minor participant. Consent for publication Written informed consent for publication of clinical information, genetic findings, and images was obtained from both adult participants and from the parents/legal guardians of the minor participant. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Songming Huang, Email: smhuang@njmu.edu.cn. Yikang He, Email: Hantanger9162@163.com. Ling Zhang, Email: hdyxyzl@126.com. References Associated Data Supplementary Materials Supplementary Material 1: Supplementary Figure S1. Comparison of the AlphaFold 3-predicted SUZ12 model with experimentally resolved SUZ12 structures. (A) Superposition of the AlphaFold 3-predicted full-length wild-type SUZ12 model with the experimentally resolved SUZ12 region in PDB 6NQ3 (chain B, residues 76–545). (B) Superposition of the AlphaFold 3-predicted full-length wild-type SUZ12 model with the experimentally resolved SUZ12 region in PDB 4W2R (chain T, residues 545–725). Only experimentally resolved SUZ12 regions were included in the comparisons. Supplementary Material 2: Supplementary Figure S2. Full-length uncropped blot images for Figures 5A and 6A. Data Availability Statement The datasets generated and/or analysed during the current study are not publicly available because they contain information that could compromise participant privacy. Raw de-identified data may be made available from the corresponding authors upon reasonable request and subject to approval by the institutional ethics committee. The SUZ12 variant has been deposited in ClinVar (Accession SCV001999392.3).