Pignata et al. Clinical Epigenetics (2022) 14:71 https://doi.org/10.1186/s13148-022-01292-w RESEARCH Open Access Novel genetic variants of KHDC3L and other members of the subcortical maternal complex associated with Beckwith–Wiedemann syndrome or Pseudohypoparathyroidism 1B and multi‑locus imprinting disturbances Laura Pignata1†, Francesco Cecere1,2†, Ankit Verma2, Bruno Hay Mele3, Maria Monticelli3, Basilia Acurzio2, Carlo Giaccari2, Angela Sparago1, Jose Ramon Hernandez Mora4, Ana Monteagudo‑Sánchez4, Manel Esteller5,6,7,8, Arrate Pereda9, Jair Tenorio‑Castano10,11,12, Orazio Palumbo13, Massimo Carella13, Paolo Prontera14, Carmelo Piscopo15, Maria Accadia16, Pablo Lapunzina10,11,12, Maria Vittoria Cubellis3, Guiomar Perez de Nanclares9, David Monk4,17, Andrea Riccio1,2* and Flavia Cerrato1*    Abstract Background: Beckwith–Wiedemann syndrome (BWS) and Pseudohypoparathyroidism type 1B (PHP1B) are imprint‑ ing disorders (ID) caused by deregulation of the imprinted gene clusters located at 11p15.5 and 20q13.32, respec‑ tively. In both of these diseases a subset of the patients is affected by multi-locus imprinting disturbances (MLID). In several families, MLID is associated with damaging variants of maternal-effect genes encoding protein components of the subcortical maternal complex (SCMC). However, frequency, penetrance and recurrence risks of these variants are still undefined. In this study, we screened two cohorts of BWS patients and one cohort of PHP1B patients for the presence of MLID, and analysed the positive cases for the presence of maternal variants in the SCMC genes by whole exome-sequencing and in silico functional studies. Results: We identified 10 new cases of MLID associated with the clinical features of either BWS or PHP1B, in which segregate 13 maternal putatively damaging missense variants of the SCMC genes. The affected genes also included KHDC3L that has not been associated with MLID to date. Moreover, we highlight the possible relevance of relatively common variants in the aetiology of MLID. Conclusion: Our data further add to the list of the SCMC components and maternal variants that are involved in MLID, as well as of the associated clinical phenotypes. Also, we propose that in addition to rare variants, common † Laura Pignata and Francesco Cecere have contributed equally *Correspondence: andrea.riccio@unicampania.it; flavia.cerrato@unicampania.it 1 Department of Environmental Biological and Pharmaceutical Sciences and Technologies (DiSTABiF), Università Degli Studi Della Campania “Luigi Vanvitelli”, Caserta, Italy Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​ mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Pignata et al. Clinical Epigenetics (2022) 14:71 Page 2 of 15 variants may play a role in the aetiology of MLID and imprinting disorders by exerting an additive effect in combina‑ tion with rarer putatively damaging variants. These findings provide useful information for the molecular diagnosis and recurrence risk evaluation of MLID-associated IDs in genetic counselling. Keywords: Multi-locus imprinting disturbance, Beckwith–Wiedemann syndrome, Pseudohypoparathyroidism, Genomic imprinting, DNA methylation, Maternal-effect variants, Subcortical maternal complex, Recurrent pregnancy loss, Infertility Background (AHO) with or without obesity. PHP is associated with Imprinting disorders (IDs) are a group of human diseases pathogenic variants and/or methylation defects within affecting growth, metabolism and neuro-behavioural the imprinted GNAS cluster on 20q13.32 [3]. Among the functions that are caused by deregulation of genes with subtypes of PHP, PHP1B (OMIM#603,233) (or iPPSD3, monoallelic and parent-of-origin dependent expres- according to the current nomenclature) is clinically sion, known as imprinted genes. Most imprinted genes characterized by isolated renal PTH resistance, in some are organized in clusters in which gene expression is cases by thyroid-stimulating hormone (TSH) resistance, regulated by genomic regions showing differential DNA and very rarely by some features of AHO. All patients methylation (Differentially methylated regions, DMRs) with PHP1B have methylation defects at the GNAS A/B between the two parental alleles. Loss of differential maternally methylated DMR, and possibly aberrations methylation of the DMRs is a hallmark of most imprint- at other DMRs in the GNAS locus. These methylation ing disorders [1]. Methylation abnormalities can affect a abnormalities are supposed to result in decreased expres- single or multiple DMRs (multi-locus imprinting distur- sion of the GNAS-Gs transcript [4]. bances, MLID) and can be associated with genetic vari- Approximately, 15% of the PHP1B cases are inher- ants acting in cis or in trans. ited and caused by microdeletions affecting the adja- Beckwith–Wiedemann syndrome (BWS, OMIM cent STX16 locus and causing loss of methylation of #130,650, prevalence of 10,500 live births) is a human the GNAS A/B:TSS-DMR [5]. Within the sporadic cases, condition that is part of a clinical spectrum (BWSp) characterized by methylation defects encompassing all characterized by more specific or “cardinal” features (e.g. the four DMRs of the locus, 10–25% are associated with macroglossia, exomphalos, lateralized overgrowth) and complete or segmental upd(20)pat [5], while the rest are less specific or “suggestive” features (e.g. neonatal mac- considered to present primary epimutations. MLID has rosomia, facial naevus flammeus, polyhydramnios, ear been detected in a subset of the cases with methylation creases or pits, abdominal wall defects). The imprinted abnormalities of the whole GNAS locus, although its inci- gene cluster associated with BWS is located on chromo- dence is still undefined due to limited investigation [6–9]. some 11p15.5 and is organized in two domains that are The molecular mechanism causing MLID in BWS controlled by the H19/ IGF2:IG-DMR (also known as and PHP1B is unknown. Its clinical consequences are IC1) and KCNQ1OT1:TSS-DMR (also known as IC2), uncertain although atypical phenotypes or phenotypes respectively. Gain of IC1 methylation and loss of IC2 overlapping multiple imprinting disorders have been methylation are found in 5–10% and 50% of the patients reported in several cases [10–13]. An accurate preva- with BWSp, respectively. In addition, uniparental pater- lence of MLID has not been determined yet because nal disomy (UPD) of chromosome 11p15.5 (upd(11)pat) there is still no international consensus for definition is reported in 20% of the cases, single nucleotide variants of MLID and recommendations for accurate molecular (SNVs) causing loss of function of CDKN1C in 5% and testing (number and genomic loci to analyse, and tis- copy number variants (CNVs) of chromosome 11p15.5 sue to analyse). Rare loss-of-function variants, affect- in 1–2%. About one third of the patients with loss of IC2 ing genes encoding either zygotic or oocytes-specific methylation display MLID [2]. trans-acting factors, have been associated with various Pseudohypoparathyroidism (PHP) is a heterogene- imprinting disorders and MLID. In particular, zygotic ous group of endocrine disorders characterized by renal biallelic variants of the ZFP57 gene have been found in resistance to parathyroid hormone (PTH), causing hypoc- case of Transient Neonatal Diabetes Mellitus (TNDM) alcaemia, hyperphosphataemia and elevated circulating [14], while biallelic or heterozygous variants affecting PTH levels. PHP is also characterized by several other some of the genes encoding the subcortical maternal clinical features such as brachydactyly, short stature, complex (SCMC) proteins have been identified in the stocky build, round face and subcutaneous ossification, healthy mothers of children affected by BWS or SRS also described as Albright hereditary osteodystrophy [13, 15–19]. Interestingly, while MLID associated with Pignata et al. Clinical Epigenetics (2022) 14:71 Page 3 of 15 ZFP57 variants consistently affect a limited number of Results imprinted loci, number and methylation levels of the Identification of MLID and DMR methylation profile imprinted loci affected by SCMC variants are variable. The patients 1–7 described in the present study received Moreover, SCMC variants appear to be associated with a clinical diagnosis of BWS. Three of them derived from increased risk of recurrence and reproductive prob- an Italian cohort of fifteen BWS patients with MLID, the lems, making their identification important for genetic other four from a Spanish cohort of nine patients. Patients counselling. However, due to the limited number of 8–10 received a clinical diagnosis of PHP1B. They derived studies performed so far, further research is needed to from a Spanish cohort of thirteen PHP1B patients with identify all the genes involved in MLID and their clini- MLID. The clinical features of the ten patients are sum- cal consequences. marized in Table 1. Further details and the clinical history The SCMC is composed of a group of proteins of their families are reported in Additional file 1: Families expressed in the oocytes and preimplantation embryos Information and Additional file 2: Fig. S1. of mammals. This complex plays a crucial role in MLID was identified in peripheral blood leukocyte (PBL) oocyte to embryo transition carrying out multiple bio- DNA by employing the multi-locus methylation-specific logical functions in the initial stages of embryogenesis, multiple ligation-dependent probe amplification (MS- including meiotic spindle formation and positioning, MLPA) assay ME034-B1 (MRC-Holland, Amsterdam, regulation of translation, organelle redistribution, The Netherlands) or the Illumina Infinium methylation and epigenetic reprogramming [20]. The proteins (HM450k, EPIC or custom) bead chip array. Altered meth- NLRP5, KHDC3L, TLE6 and OOEP have been iden- ylation levels of at least one germline DMR in addition tified as members of the human SCMC through pro- to that diagnostic for BWS or PHP1B (KCNQ1OT1:TSS- tein–protein interaction studies [21]. Other proteins DMR and GNAS A/B:TSS-DMR, respectively) were found (e.g. NLRP2, NLRP4, NLRP7, PADI6 and ZBED3) have in the probands of ten families (Additional file 3: Table S1). been proposed to be part of the complex on the basis The list of DMRs affected by methylation disturbances in of their co-localization or shared biological functions each proband is reported in Table 1 and the methylation with other SCMC components, or because they are values in Additional file 3: Table S1). orthologues of the mouse SCMC members [22–26]. The methylation defect was partial in most of the In addition to imprinting disorders, SCMC variants DMRs and the number of hypomethylated DMRs was have been associated with severe clinical conditions more prevalent compared to the hypermethylated DMRs. affecting reproduction such as female infertility, recur- The number and methylation levels of the affected DMRs rent biparental hydatidiform mole (RHM) and recur- were variable even between the probands sharing mater- rent miscarriages [20, 27]. However, the role of the nal effect variants in the same gene. In addition to the individual SCMC components in specific clinical con- diagnostic locus, the most frequently hypomethylated ditions and biological processes is not well defined yet. loci were GNAS (5/7) and PLAGL1 (5/7) in the patients For instance, KHDC3L and TLE6 variants have been affected by BWS, DIRAS (2/3), MEST (2/3) and PEG13 associated only with severe non-viable conditions, (2/3) in the patients affected by PHP1B. In two PHP1B such as RHM and zygotic lethality, so far [28–31]. Dif- patients, the proband of family 9 and the proband of ferently, variants of NLRP7, PADI6, NLRP2 and NLRP5 family 10, the GNAS DMRs were more severely hypo- have been associated with both severe reproductive methylated than in the BWS patients. In addition, the problems and imprinting disorders [17, 18, 32, 33]. KCNQ1OT1-TSS:DMR was found hypomethylated in Here, we report the results of genetic/epigenetic BWS but not in PHP1B, while PEG13 was hypomethyl- analyses performed on an Italian and a Spanish cohort ated only in PHP1B. Other imprinted loci were hypo- of patients with MLID and the clinical features of methylated in less than half of the patients. either BWSp or PHP1B. We have identified maternal Overall, the methylation data demonstrated a profile putatively damaging variants of several SCMC genes of apparently randomly disturbed methylation of the in ten pedigrees, including seven patients clinically imprinted DMRs in all the probands rather than a spe- affected by BWSp and three by PHP1B. The affected cific epi-signature for each single SCMC gene. genes also included KHDC3L that has not been associ- ated with MLID so far. Our findings further add to the list of the SCMC components and maternal variants Identification of the maternal variants of the SCMC genes that are involved in MLID, as well as of the associated Since maternal-effect variants of the SCMC genes have clinical phenotypes. been previously associated with MLID, we looked for Table 1 Summary of the clinical and molecular features of the probands and their families Family Proband sex Maternal effect variant Genotype Hypomethylated and Maternal reproductive Family Clinical features of proband and age (years) hypermethylated loci history history of note 1 Female, 24 KHDC3L M: hom Hypomethylated loci: KCNQ1OT1, None None Macroglossia, macrosomia, umbilical novel NM_001017361: P: het ZDBF2/GPR1, hernias, ear creases/pits, nevus flammeus c.296C > G; F: wt NAP1L5,PLAGL1, (eyelid), mild neonatal hypoglycaemia, p.Thr99Arg GRB10, PEG10, MEST, ZNF597, GNAS- pre-eclampsia, prognathism, maxillary AF: - XLas, GNAS A/B, GNAS-AS1 hypoplasia, seizures (once), strabismus GF: - Hypermethylated loci: DIRAS3:TSS, BWS score: 7 PolyPhen-2: GNAS-NESP, ZNF331-DMR1, ZNF331- Pignata et al. Clinical Epigenetics Possibly damaging DMR2 SIFT: Deleterious SDM: Destabilizing 2 Female, 39 PADI6 M: het Hypomethylated loci: KCNQ1OT1, PPIEL, Two miscarriages Second sister: three Macroglossia, polyhydramnios, novel P: het NAP1L5, miscarriages and one diastasis recti, neonatal hypoglycae‑ NM_207421: S: het PLAGL1, ZNF331-DMR1, ZNF331-DMR2, healthy daughter mia, ear creases/pits, nevus flammeus, c.356 T > C; F: wt GNAS-XLas, GNAS A/B, GNAS-AS1 Third sister: two healthy nephromegaly, enlarged pancreas, facies (2022) 14:71 p.Leu119Pro Hypermethylated loci: ZNF597, children grossolana, face asymmetry, maxillary AF: - GNAS-NESP hypoplasia, mild intellectual deficit, GF: - psychomotor retardation PolyPhen-2: BWS score: 8 Probably damaging SIFT: Deleterious SDM: Destabilizing 3 Female, 40 NLRP5 M: het Hypomethylated loci: KCNQ1OT1, Ovarian stimulation Proband: one healthy Macroglossia, macrosomia, lower limbs rs768443657 P: wt DIRAS3:TSS, PLAGL1, GRB10, Two miscarriages. One triplet preg‑ daughter one miscar‑ dysmetria, scoliosis, hypotonia NM_153447.4: F: na ERLIN2, GNAS-XLas, GNAS A/B, GNAS-AS1 nancy: one miscarriage, two children riage, one healthy son BWS score: 5 c.2615G > A; M: het Hypermethylated loci: ZNF331- born at 6 months of gestation and p.Arg872Lys P: het DMR1,ZNF331-DMR2, GNAS-NESP died few hours/days later. One healthy AF: 8.95E-06 F: na son GF: 0 M: het PolyPhen-2: P: het Possibly damaging F: na SIFT: Deleterious SDM: Destabilizing NLRP5 rs36118060 NM_153447.4: c.3584G > A; p.Arg1195Gln AF: 0.145 GF: 0.022 PolyPhen-2: Benign SIFT: Tolerated SDM: Destabilizing NLRP4 rs111284755 NM_134444.4: c.1279G > A; p.Ala427Thr AF: 0.006 GF: 2.00E-05 PolyPhen-2: Benign SIFT: Tolerated SDM: Destabilizing Page 4 of 15 Table 1 (continued) Pignata et al. Clinical Epigenetics Family Proband sex Maternal effect variant Genotype Hypomethylated and Maternal reproductive Family Clinical features of proband and age (years) hypermethylated loci history history of note 4 Male, 15 NLRP5 M: het Hypomethylated loci: KCNQ1OT1, Proband born by in vitro fertilization None Macroglossia, mild asymmetry, macroso‑ rs34175666 P: wt PLAGL1, IGF1R, GNAS-AS1 mia, neonatal hypoglycaemia, atopic NM_153447.4: F: wt Hypermethylated loci: none eczema, (2022) 14:71 c.1685G > A; M: hom gestational diabetes p.Arg562His P: het BWS score: 6 AF: 0.007 F: wt GF: 0 M: hom PolyPhen-2: P: het Possibly damaging F: wt SIFT: Deleterious SDM: Destabilizing NLRP5 rs12462795 NM_153447.4: c.3323C > G; p.Ser1108Cys AF: 0.145 GF: 0.022 PolyPhen-2: Probably damaging SIFT: Deleterious SDM: Stabilizing NLRP5 rs36118060 NM_153447.4: c.3584G > A; p.Arg1195Gln AF: 0.145 GF: 0.022 PolyPhen-2: Benign SIFT: Tolerated SDM: Destabilizing Page 5 of 15 Table 1 (continued) Family Proband sex Maternal effect variant Genotype Hypomethylated and Maternal reproductive Family Clinical features of proband and age (years) hypermethylated loci history history of note 5 Female, 22 NLRP2 M: het Hypomethylated loci: KCNQ1OT1, One miscarriage Two daughters Macroglossia, hyperinsulinism, hypogly‑ rs61735086 P: wt NAP1L5, NNAT, caemia, hepato/splenomegaly, nevus NM_017852.4: F: wt GNAS-XLas, GNAS A/B, GNAS-AS flammeus (head, neck), maxillary hypo‑ c.1681G > A; M: het Hypermethylated loci: ZNF331-DMR2, plasia, neonatal anaemia, respiratory p.Ala561Thr P: het GNAS-NESP distress, round face. BWS score: 6 AF: 0.000 F: wt Pignata et al. Clinical Epigenetics GF: 0 PolyPhen-2: Benign SIFT: Tolerated SDM: Destabilizing NLRP2 rs17699678 (2022) 14:71 NM_017852.4: c.662C > T; p.Thr221Met AF: 0.110 GF: 0.012 PolyPhen-2: Possibly damaging SIFT: Deleterious SDM: Stabilizing 6 Female, 14 NLRP2 M: het Hypomethylated loci: KCNQ1OT1, None None Macroglossia, exomphalos, perinatal rs17699678 P: het NAP1L5, PLAGL1, PEG10, MEST, GNAS- hypoglycaemia, macrosomia, anterior NM_017852.4: F: wt XLas, GNAS A/B, GNAS-AS1 creases in the ear, nevus flammeus c.662C > T; Hypermethylated loci: GNAS-NESP (neck), round face, haemangioma. BWS p.Thr221Met score: 7 AF: 0.110 GF: 0.012 PolyPhen-2: Possibly damaging SIFT: Deleterious SDM: Stabilizing 7 Male, 15 PADI6 M: het Hypomethylated loci: KCNQ1OT1, None None Macroglossia, macrosomia, round and rs74834315 P: het IGF1R, SNU13 coarse face with prominent forehead, NM_207421.4: Hypermethylated loci: none antimongoloid palpebral fissures and c.775G > A; later on advanced bone age. BWS p.Val259Ile score: 3 AF: 0.001 GF: 0 PolyPhen-2: Benign SIFT: Deleterious SDM: Destabilizing Page 6 of 15 Table 1 (continued) Pignata et al. Clinical Epigenetics Family Proband sex Maternal effect variant Genotype Hypomethylated and Maternal reproductive Family Clinical features of proband and age (years) hypermethylated loci history history of note 8 Female, 4 NLRP2 M: het Hypomethylated loci: GNAS-XLas, GNAS Proband born by in vitro fertilization None Early onset obesity, intellectual disability rs147585490 P: wt A/B, GNAS-AS, NM_017852.4: DIRAS3:Ex2, PLAGL1, MEST, ERLIN2, c.1055 T > G; PEG13, IGF1R, SNU13 (2022) 14:71 p.Ile352Ser Hypermethylated loci: GNAS-NESP, AF: 0.002 ZDBF2/GPR1 GF: 0 PolyPhen-2: Possibly damaging SIFT: Deleterious SDM: Destabilizing 9 Female, 35 NLRP2 M: het Hypomethylated loci: GNAS-XLas, GNAS None None Truncal obesity, hypothyroidism, rs61735077 P: wt A/B, GNAS-AS, DIRAS3, FAM50B, MEST, hypercholesterolemia, occasional high NM_017852.4: MCTS2P PTH levels c.1060A > G; Hypermethylated loci: GNAS-NESP, p.Ile354Val NDN, ZNF597 AF: 0.008 GF: 0.000 PolyPhen-2: Benign SIFT: Tolerated SDM: Destabilizing 10 Male, 8 NLRP2 M: het Hypomethylated loci: GNAS-XLas, GNAS One miscarriage None Short stature, mild global developmental rs4306647 P: wt A/B, GNAS-AS, PEG10, PEG13, ZNF331 delay, PTH resistance with hypocalcae‑ NM_017852.4: Hypermethylated loci: GNAS-NESP mia and hyperphosphataemia c.1091G > A; p.Arg364Lys AF:0.041 GF: 0.001 PolyPhen-2: Possibly damaging SIFT: Tolerated SDM: Destabilizing In third column AF: Allele frequency in European population; GF: frequency of the homozygous genotype of the variant in European population. In fourth column: M = mother, P = Proband, S = Sister, F = Father, na: not available Page 7 of 15 Pignata et al. Clinical Epigenetics (2022) 14:71 Page 8 of 15 variants of these genes in the mothers of the identified regions of heterozygosity mixed to long regions of loss probands, by either whole-exome sequencing (WES) or of heterozygosity along the entire chromosome 6 (Addi- SCMC gene targeted Sanger sequencing analyses (see tional file 2: Fig. S1c). This profile corresponds to a mixed also Methods section). We selected all rare (allele fre- heterodisomic/isodisomic UPD that is likely caused by quency < 1%) and common (allele frequency > 1%) vari- physiological meiotic recombination followed by non- ants that were predicted to be damaging/deleterious by disjunction and disomic gamete formation, and zygotic the bioinformatic prediction tools PolyPhen-2 [34, 35] trisomy rescue after fertilization [41]. The KHDC3L locus and SIFT [36] or destabilizing by SDM [37]. Overall, we was located in one of the three isodisomic intervals, thus identified thirteen putatively deleterious maternal mis- explaining the homozygous genotype of the mother. To sense variants in ten pedigrees. According to the cri- determine the parental origin of the UPD, we analysed teria of the American College of Medical Genetics and the DNA methylation of the imprinted PLAGL1 DMR Genomics (ACMG) [38] ten of these variants were classi- that is also located on chr 6. Because this DMR was fied as variants of uncertain significance (VUS), and three totally methylated while the DMRs located on different as benign because their frequency was higher than 5% in chromosomes showed a normal 50% methylation level, the general population. Possible limitations of the ACMG we concluded that the upd(6) was of maternal origin criteria for interpreting pathogenicity of maternal-effect (Additional file 2: Fig. S1d). Analysis of the grandparen- variants are described in the Discussion. tal DNAs by SNP-array confirmed maternal upd of the The genetic, epigenetic and phenotypic features of the entire chromosome 6 in the proband’s mother of family 1 families are summarized in Table 1, while all the charac- (Additional file 5: Table S3). teristics of the variants including ACMG classification Concerning the phenotype, no particular clinical sign are reported in Additional file 4: Table S2. The variants was observed in the mother. She was born at term from occurring in each family are described in more detail uneventful pregnancy, her birth length and weight were below. referred to be normal and her final adult height was Family 1. A novel variant of KHDC3L was found in 158 cm. She had normal cognitive and physical develop- homozygosity in the proband 1’s mother (Additional ment and she has not reported any significant diseases so file 2: Fig. S1a). This is a missense variant (chr 6, c.296 far. C > G; p.Thr99Arg) located within exon two and affecting Family 2. A novel variant was identified in PADI6 and the threonine in position 99 of the protein. This residue is found in heterozygosity in the proband 2’s mother. This is not evolutionarily conserved, but in silico analyses indi- a missense mutation (c.356C > T; p.Leu119Pro) in exon 3, cated its harmful effect. We observed that this position affecting the leucine in position 119 of the protein close is usually held by amino acids containing a methyl group, to the end of the N-terminal domain, and predicted to most frequently valine and threonine, two isosteric resi- be strongly destabilizing according to the score of SDM dues (Additional file 2: Fig. S1b, top). According to the prediction (Additional file 4: Table S2). The variant was AlphaFold [39] Protein Structure Database [40] model of inherited by the proband and her healthy second sister KHDC3L, Thr99 belongs to an alpha helix and entertains (II-2, Fig. 1a) who had a healthy daughter after three con- a hydrophobic interaction (via its methyl group) with res- secutive miscarriages (Table 1 and Additional file 1: sup- idue Ile103 and a hydrogen bond (via its hydroxyl group) plemental information). Methylation was tested in both with residue Ser70 (Fig. S1b, bottom). Both residues, as the proband’s sister II-2 (by multi-locus MS-MLPA and threonine, are needed to maintain multiple intramolecu- methylome EPIC 850 k array) and mother (by multi-locus lar interactions with other residues. The substitution of MS-MLPA) was found to be normal (data not shown). Thr99 with a charged amino acid, in our case arginine, Families 3–7. Rare missense variants affecting several likely disrupts these interactions, destabilizing the pro- genes of the SCMC and belonging to the NLRP family tein structure with harmful effects, as predicted by Poly- were found in heterozygosity in the probands’ mothers Phen-2, SIFT and SDM tools (Additional file 4: Table S2). of families 3–6 (Fig. 1b–e). In particular, p.Arg872Lys of The KHDC3L c.296C > G; p.Thr99Arg variant was NLRP5 was found in family 3, p.Arg562His of NLRP5 in present in heterozygosity in the proband and mater- family 4 and p.Ala561Thr of NLRP2 in family 5 (Table 1). nal grandmother and absent in the father and maternal These three women were also carriers of common vari- grandfather (Table 1 and Additional file 2: Fig. S1a). To ants affecting the same genes. P.Arg1195Gln of NLRP5 explain the homozygosity of this variant in the mother, was in heterozygosity in the proband’s mother of fam- a SNP-array analysis was performed to detect the pres- ily 3, p.Ser1108Cys and p.Arg1195Gln of NLRP5 were in ence of uniparental disomy or deletions on chromo- homozygosity in the proband’s mother of family 4 and some 6q. The results demonstrated a normal pattern on p.Thr221Met of NLRP2 was in heterozygous state in the most chromosomes, but revealed the presence of long proband’s mother of family 5. Segregation analysis in the Pignata et al. Clinical Epigenetics (2022) 14:71 Page 9 of 15 trios (Table 1 and Fig. 1) demonstrated that the rare and non-viable phenotypes, while the hypomorphic variants the common variants were present in compound het- result in MLID [17, 18]. The results obtained from the erozygosity in the mother of family 5 (Fig. 1d) and pos- BWS family 1 strongly support this hypothesis. In this sibly in the mother of family 3 (Fig. 1b). Indeed, only the case, the proband’s mother carries a hypomorphic vari- common variants were inherited by the probands and no ant of KHDC3L in homozygosity, while biallelic severely variant was present in the father of family 5. However, in inactivating pathogenic variants of this gene have been family 3 because of missing information on the father’s associated with RHM and recurrent pregnancy loss so far genotype, the possibility that the common variant was [28, 29, 42–45]. transmitted by the father to the proband cannot be ruled Some of the pedigrees reported in this study and segre- out. Furthermore, a rare variant of NLRP4 (p.Ala427Thr) gating maternal-effect SCMC missense variants include was found in heterozygosity in both the mother and recurrent miscarriages in addition to ID phenotypes in proband of family 3. The mothers of pedigrees 6–7 were the offspring. Intrafamilial heterogeneous reproductive heterozygous for single variants: p.Thr221Met of NLRP2 outcomes have been reported before and could be a con- in family 6 (Fig. 1e) and p.Val259Ile of PADI6 in family 7 sequence of the multiple cellular functions of the SCMC (Fig. 1f ). components [20, 27]. However, as no DNA from the mis- Families 8–10. Missense variants of NLRP2 were identi- carriages were available for genetic analysis, we cannot fied in heterozygosity in the mothers of the three patients rule out causes other than the SCMC gene variants for affected by PHP1B: p.Ile352Ser in family 8, p.Ile354Val in the pregnancy losses. family 9 and p.Arg364Lys in family 10 (Fig. 1g–i). Interestingly, two patients affected by PHP1B showed Overall, the sequencing data demonstrated that the ten also intellectual disability (proband of family 8) or mild mothers were carriers of at least one putatively damag- global developmental delay (proband of family 10), two ing variant affecting the SCMC genes. In some of them, clinical features already reported associated with several a rare variant was in homozygosity or in compound hete- cases of MLID [13, 14, 16–18]. rozygosity with another rare or common putatively dam- In all our families, the viable progeny shows a vari- aging variant, or variants affecting more than one SCMC able number of hypomethylated imprinted germline gene were present in the same woman. DMRs (Additional file 3: Table S1). Since the methyla- tion changes are often partial and likely in the mosaic form, it is possible that the clinical presentation of the Discussion affected individuals depends on which imprinted locus In this study, we report ten new cases of MLID and clini- is more strongly affected (epidominance hypothesis) [7, cal features of either BWS or PHP1B, in which segregate 17]. Interestingly, this methylation variability appears to putatively damaging maternal-effect missense variants of be more characteristic of the SCMC-MLID cases than the SCMC genes, further extending the list of genes and the ZFP57-TNDM-MLID cases in which some DMRs IDs associated with maternal variants. These variants are more consistently affected. This could be caused by were classified as VUS or benign according to the ACMG impairment of the different functions of these trans-act- criteria, but their pathogenic damaging effect could have ing factors: weaker binding of zygotic variants of ZFP57 been underestimated for the following reasons: i) mater- to the methylated allele at the imprinted DMRs or a nal-effect variants are present in the maternal genome general methylation establishment/maintenance defect but affect the offspring phenotype; ii) no effect on male caused by maternal SCMC variants. reproduction has been reported so far; iii) they are asso- Hypermethylation is also observed, mostly affecting ciated with variable expressivity and incomplete pen- paternally methylated secondary DMRs if their respec- etrance of the molecular (methylation disturbance) and tive maternally methylated germ-line DMRs are hypo- clinical phenotype of the offspring of carrier mothers; iv) methylated (e.g. GNAS-NESP, ZNF597, and ZDBF2/ all the previous points might explain why some SCMC GPR1). Mild hypermethylation affects also some mater- variants, although potentially damaging, have high fre- nal germ-line DMRs, likely as consequence of a defective quency in the general population. general mechanism of DNA methylation maintenance in The phenotypes. SCMC variants have been associated early embryogenesis or a secondary effect of the hypo- with phenotypes of different severity ranging from female methylation of other loci. infertility, RHM and recurrent miscarriages, to IDs with The genotype. If the SCMC variants act as dominant MLID in the offspring. It has been proposed that the or recessive mutation is debated. While the variants of severity of the reproductive outcome is dependent on the NLRP7, KHDC3L and PADI6 that were associated with impact of the variants on protein function, so that the RHM have consistently been found in homozygosity or variants completely inactivating the gene product cause compound heterozygosity, more than half of the SCMC Pignata et al. Clinical Epigenetics (2022) 14:71 Page 10 of 15 Fig. 1 Pedigrees of families 2–10. Black filled symbol represents the probands affected by BWS or PHP1B, black central dots the unaffected carrier mothers. Triangles: spontaneous miscarriage. Triangle with line: voluntary termination of pregnancy. The variants in bold are rare (AF < 0.01), those in regular font style are common (AF > 0.01). The asterisks indicate the family members whose DNA was not available for genetic analyses pathogenic variants reported in MLID cases have been maintenance. The involvement of KHDC3L in RHM, a found in heterozygosity in the probands’ mothers [16– gestational abnormality with broad loss of maternally but 18]. In the present study, putatively damaging maternal not paternally methylated imprints, implicates KHDC3L SCMC gene variants were found in either homozygosity in oocyte-specific methylation establishment. Also, or simple or compound heterozygosity. In particular, rare the methylation abnormalities of the family 1 proband heterozygous variants were found in families 2 and 6–10, appear to be restricted to the maternal germ-line DMRs a rare variant was present in homozygosity in family 1 suggesting an oocyte origin of the defect. However, and rare variants were in compound heterozygosity with because some DMRs are only partially hypomethylated common putatively damaging variants in families 3–5. in this patient, a role of KHDC3L in post-zygotic imprint This suggests that common variants may play a role in maintenance cannot be excluded. Consistent with this the aetiology of the MLID if potentially deleterious and hypothesis, also for NLRP7 a role in both imprint estab- present in combination with rarer variants. Alternatively, lishment and maintenance has been proposed, since heterozygous variants might act as dominant-negative maternal-effect variants of this gene were found in both mutants [17] or by exerting an additive effect in combi- RHM and MLID cases with hypomethylated paternal nation with variants affecting further SCMC components DMRs [17, 32]. or other unidentified genes. In this regard, we found het- Interestingly, in family 1 two rare events, homozygo- erozygous maternal variants of more than one SCMC sity for the rare putatively damaging KHDC3L variant gene in family 3. It is also worth mentioning that ours is and upd(6)mat, have occurred in the same individual, the the first report of SCMC variants associated with PHP1B. proband’s mother. Although UPD is a possible mecha- KHDC3L. The proband of family 1 represents the nism underlying recessive phenotypes of rare pathogenic first case of MLID associated with a maternal variant variants, the hypothesis that these two rare events are of KHDC3L. This finding raises a question concern- causally linked is plausible. Constitutional UPD of an ing the role of this protein in imprint establishment or entire chromosome is usually caused by trisomic rescue Pignata et al. Clinical Epigenetics (2022) 14:71 Page 11 of 15 of a zygote derived from fertilization of a disomic gamete required for cytoplasmic lattices formation and organelle [41, 46]. In this family, the hypermethylation of PLAGL1, distribution in oocytes [23, 26]. Notably, this mother had the segregation of the KHDC3L variant and the SNP- serious reproductive problems (Fig. 1b and Additional array data indicate that the disomic gamete has derived file 1: Supplemental information), and she was the only from the maternal grandmother. A number of evidences one carrying putatively damaging variants affecting more support the hypothesis that the SCMC members, and than one SCMC gene (NLRP4 and NLRP5). This suggests particularly KHDC3L, are involved in aneuploidies [27, an additive effect of the variants on the impairment of the 45, 47, 48]. Thus, it is possible that the KHDC3L variant SCMC function resulting in a more severe reproductive is responsible for the occurrence of gametic non-disjunc- outcome. tion and disomy in the maternal grandmother and upd(6) Further studies on larger cohort of MLID families and in the mother. Consistent with the finding that upd(6)mat animal models are needed to clarify the role of this com- is not necessarily associated with pathological conditions plex in human reproduction and the impact that the rare [47, 49] (http://​upd-​tl.​com/​upd.​html), no significant clin- and less rare missense variants of the SCMC genes have ical feature was reported for this woman. on methylation disturbance and imprinting disorders. PADI6. Similar to a maternal-effect variant reported by Begemann et al. [17], the p.Val259Ile variant found Conclusions in family 7 affects the second domain (the PAD middle In summary, we identified ten new cases of MLID, seven domain) of the PADI6 protein. In contrast, all the other clinically affected by BWSp and three by PHP1B. By MLID-associated pathogenic variants reported so far WES and SCMC gene targeted sequencing, we identi- fall within the third domain of this protein (the Protein fied in their mothers thirteen variants of genes encoding Arginase Deiminase domain) [18]. Differently, the novel components of the SCMC, including NLRP2, NLRP4, missense Leu119Pro variant found in family 2 affects the NLRP5, PADI6 and KHDC3L. Among the variants, two first domain (the PAD N-terminal) suggesting that also were novel, seven rare and four common, but all were this part of the protein might have a role in the establish- predicted to be harmful. Further, by methylome array ment/maintenance of the methylation at the imprinted we showed methylation profiles of the MLID patients loci. The p.Leu119Pro was predicted to strongly destabi- characterized by variable number and methylation level lize the protein and this might explain the recurrent preg- of affected DMRs, demonstrating lack of a specific epi- nancy loss of the family 2 (Fig. 1a). However, despite the signature for each single SCMC gene. Overall, these data inheritance of the variant and the recurrent pregnancy increase our knowledge on the link between the mater- loss, no methylation defect of the imprinted DMRs was nal-effect variants of the SCMC genes and the MLID- found in the proband’s sister II-2. This is consistent with associated IDs. All this information should be considered the incomplete penetrance of the epigenotype and phe- for genetic counselling of MLID families to improve notype that often characterizes the SCMC-MLID pedi- molecular diagnosis and prediction of the recurrence risk grees [17, 18], and that might be explained by a polygenic of the IDs. hypothesis, according to which, further unknown mater- nal or zygotic factors may be involved in MLID aetiology, Methods by acting as modifiers of the phenotype of the maternal Analyses of epigenetic and genetic variants SCMC genes. Genomic DNA was isolated from PBL by the salting-out NLRPs. Although several variants of NLRP2 and procedure [51]. NLRP5 have been found in MLID pedigrees, none of the MS-MLPA analysis was performed by SALSA MS- variants described in the present study have been previ- MLPA Kit (MRC-Holland, Amsterdam, The Netherlands) ously reported associated with IDs or reproductive con- according to the manufacturer’s protocol: Probemix ditions. Interestingly, the variant p.Thr221Met found in ME034 for multi-locus imprinting was applied to BWS pedigrees 5–6 was present at higher frequency than that cases (probands 1–7) [49]. Probemix ME031for GNAS reported in dbSNP in a group of 94 women with diagno- locus was applied to PHP1Bcases (probands 8–10). sis of unexplained infertility (Allele frequency: 0.313 vs SNP-array on DNA mother and maternal grandparents 0.061) [50]. of proband 1 was performed as previously reported [49]. The variant of NLRP4 found in the proband’s mother DNA Sequencing. Whole-exome sequencing was per- of family 3 represents the first maternal-effect variant of formed on DNA of mothers of families 1–6 at IGAT- this gene found associated with MLID. This finding sup- ech Service (Udine, Italy). Nextera Flex for Enrichment ports a role of NLRP4 as a member of the human SCMC. solution (Illumina, San Diego, CA) in combination with NLRP4 is orthologous to the mouse Nlrp4f that encodes a “SureSelect Human All Exon V7” probes (Agilent, Santa protein with the same cellular distribution of Nlrp5 and is Clara, CA) was used for library preparation and exome Pignata et al. Clinical Epigenetics (2022) 14:71 Page 12 of 15 enrichment, targeting 50 Mb of human exonic content. sequence homologies, which resulted in 365,359 retained The samples were quantified and quality tested using the probes, 658 mapping to imprinted DMRs common to Qubit 2.0 Fluorometer (Invitrogen, Carlsbad, CA) and both platforms. In-house bioinformatics R scripts were Agilent 2100 Bioanalyzer (Agilent Technologies, Santa utilized for statistical comparisons. Clara, CA). Libraries were sequenced on NovaSeq 6000 Probands 9 and 10 were analysed with a custom Illumina (Illumina, San Diego, CA) in 150 pair-end mode. The bio- Veracode GoldenGate methylation array and are already informatic analysis was performed as previously reported described in Court et al. [7] as PHP0081 and PHP0089, [18]. respectively. Twenty-eight DNA samples from phenotypi- Exon-PCR and Sanger sequencing of all known SCMC cally normal individuals aged between 1 and 35 years (17 genes (TLE6, NLRP2, NLRP7, NLRP5, OOEP, PADI6 and females, 11 males) were used as control for sample com- KHDC3L) was performed on PBL DNA from both moth- parison using the custom array platform, with methylation ers and probands of families 7–10. values of eight different complete chromosome UPDs (for Segregation analysis of variants in the family members chr:6, 7, 14, 15 and 20) and uniparental diploidy samples was performed by Sanger sequencing. In silico predic- employed to define extreme methylation profiles [7]. tion of variant pathogenicity by bioinformatic tools was To investigate the methylation profile at the imprinted performed as previously reported [18]. The variants were DMRs, we downloaded the coordinates from http://​ also classified following the ACMG standards and guide- www.​human​impri​nts.​net/, identifying every probe across lines [38]. the imprinted regions. After filtering out all the iDMRs Methylome array was performed using Illumina with low coverage (< 4 probes), we calculated the average Infinium Methylation BeadChip kit on PBL DNA of all methylation level for each region, considering as statis- patients. Probands 1–6 were analysed by the Epic 850 k tically significant those exceeding three standard devia- kit at BIODIVERSA srl Service (Milan, Italy) as previ- tions from the average of the controls. ously reported [18]. Briefly, DNA of probands and twelve non-affected individuals (7 females and 5 males, aged Protein bioinformatics between 2 and 32) used as was subjected to bisulfite con- To study Thr99 conservation, we used the NCBI blast version and methylation array processing by the EPIC service on the Swissprot database with the KHDC3L BeadChip array, following the manufacturer’s protocol. UniProt accession id (Q587J8) as the input sequence Array data were analysed in Rstudio (v4.1.0) using the identifier. We aligned the blast output with clustalW to Bioconductor package “ChAMP” (v.2.22.0). After import- produce a Multiple Sequence Alignment (MSA) and then ing the “idat” files in RStudio and filtering them using visualized the MSA using WebLogo. The effect of the del- “champ.load”, we obtained the Beta value matrix, which eterious variants was predicted using the sequence-based retains 713,288 probes. We then applied “BMIQ” nor- tools SIFT [34] and PolyPhen-2 [35, 36] and the SDM malization and batch correction using “champ.norm” and tool for the structure-based analysis of substitution toler- “champ.runCombat” functions, respectively. ance [37]. Chimera version 1.15 (build 42,258) was used The methylomes for patients 7 and 8 were generated to visualize the KHDC3L PDB from the AlphaFold Pro- using the Illumina Infinium Human Methylation 450 tein Structure Database [39, 40]. and EPIC 850 BeadChip arrays, respectively, and com- pared to 20 control PBL samples. Bisulphite conversion was performed according to the manufacturer’s recom- Abbreviations BWS: Beckwith–Wiedemann syndrome; DMRs: Differentially methylated mendations for the Illumina Infinium Assay (EZ DNA regions; ICs: Imprinting centres; IDs: Imprinting disorders; MLID: Multi-locus methylation kit, ZYMO, Orange, CA) and hybridiza- imprinting disturbances; PBL: Peripheral blood leukocytes; PHP: Pseudohy‑ tion following the Illumina Infinium HD methylation poparathyroidism; TNDM: Transient neonatal diabetes mellitus; SCMC: Sub- cortical maternal complex; UPD: Uniparental disomy; VUS: Variant of unknown protocol at genomic facilities of the Cancer Epigenetics significance; WES: Whole-exome sequencing. and Biology Program (Barcelona, Spain). Before analys- ing the data, we excluded possible sources of technical Supplementary Information biases that could influence results. We applied signal The online version contains supplementary material available at https://​doi.​ background subtraction, and interpolate variation was org/​10.​1186/​s13148-​022-​01292-w. normalized using default control probes in BeadStudio (version 2011.1_Infinium HD). We discarded probes with Additional file 1: Supplemental information. Clinical features of the patients described in the present study. a detection p value > 0.01, containing single nucleotide Additional file 2: Figure S1. Genetic analyses of family 1. (a) Pedigree. (b) polymorphisms (SNPs) within the interrogation or exten- top: WebLogo of KHDC3L orthologues MSA in the 70-103 position range. sion base, those mapping to sex chromosomes as well Y-axis: the probability of finding a specific residue in each position; dashed as those with potential cross-reaction due to multiple lines mark relevant model-based interaction between residuals (black: Pignata et al. Clinical Epigenetics (2022) 14:71 Page 13 of 15 Author details 1 T99-I103 hydrophobic, blue: T99 - S70 hydrogen bond); bottom: AlphaFold Department of Environmental Biological and Pharmaceutical Sciences Protein Structure Database model of KHDC3L showing T99 and the inter‑ and Technologies (DiSTABiF), Università Degli Studi Della Campania “Luigi actions: hydrophobic T99 - I103 via methyl group (black) and hydrogen Vanvitelli”, Caserta, Italy. 2 Institute of Genetics and Biophysics (IGB), “Adriano bond T99 - S70 via hydroxyl (blue). (c) SNP-array results of chromosome Buzzati-Traverso”, Consiglio Nazionale Delle Ricerche (CNR), Naples, Italy. 3 6. (d) MS-MLPA of the proband’s mother (II-2). The red arrow indicates the Department of Biology, Università Degli Studi Di Napoli “Federico II”, Naples, histogram of PLAGL1 DMR showing 100% of methylation. Italy. 4 Cancer Epigenetic and Biology Program (PEBC), Imprinting and Cancer Group, Institut d’Investigació Biomedica de Bellvitge (IDIBELL), Avinguda Gran‑ Additional file 3: Table S1. DNA methylation values of imprinted loci via, L’Hospitalet de Llobregat, Barcelona, Spain. 5 Josep Carreras Leukeamia obtained by methylome array. Research Institute, Can Ruti, Cami de les Escoles, Badalona, Barcelona, Spain. Additional file 4: Table S2. List of the SCMC variants identified in the 6 Centro de Investigacion Biomedica en Red Cancer (CIBERONC), Madrid, Spain. probands mothers. 7 Institucio Catalana de Recerca I Estudis Avançats (ICREA), Barcelona, Catalo‑ Additional file 5: Table S3. List of the SNP genotypes of the whole nia, Spain. 8 Physiological Sciences Department, School of Medicine and Health chromosome 6 obtained by SNP- array analysis on mother and maternal Sciences, University of Barcelona (UB), Barcelona, Catalonia, Spain. 9 Molecular grandparents of proband 1. (Epi)Genetics Laboratory, Rare Diseases Research Group, Bioaraba Health Research Institute, Araba University Hospital-Txagorritxu, C/Jose Atxotegi s/n, 01009 Vitoria‑Gasteiz, Spain. 10 CIBERER, Centro de Investigación Biomédica en Acknowledgements Red de Enfermedades Raras, Madrid, Spain. 11 Institute of Medical and Molecu‑ The authors are grateful to the patients and their family members for their lar Genetics, INGEMM-Idipaz, Madrid, Spain. 12 ITHACA, European Reference willingness to collaborate on this research project. Network, Brussels, Belgium. 13 Division of Medical Genetics, Fondazione IRCCS “Casa Sollievo Della Sofferenza”, 71013 San Giovanni Rotondo, FG, Italy. 14 Medi‑ Author contributions cal Genetics Unit, University and Hospital of Perugia, Perugia, Italy. 15 Medical LP, BA, CG, AS, JRHM, AMS, ME, PL, DM, FlCe and AR analysed and interpreted and Laboratory Genetics Unit, “Antonio Cardarelli” Hospital, 80131 Naples, Italy. 16 most of the molecular analyses of the families 1–7, AP and GPdN those of Medical Genetics Service, Hospital “Cardinale G. Panico”, 73039 Tricase, Lecce, families 8–10. BHM, MM and MVC carried out the exome-seq and in-silico Italy. 17 School of Biological Sciences, University of East Anglia, Norwich NR4 prediction analysis of the variants. FrCe and AV performed the bioinformatic 7TG, UK. analysis of the methylome array. OP and MC performed the SNP-array analysis. JT, AP, PP, CP and MA identified the families and made the clinical diagnosis. Received: 8 February 2022 Accepted: 16 May 2022 FlCe and AR wrote the manuscript and all authors commented to the draft. All authors read and approved the final manuscript. Funding The work was supported by the followings grants: INCIPIT H2020-MSCA- References COFUND grant agreement—N. 665403 project (AR and AV), Associazione 1. Monk D, Mackay DJG, Eggermann T, Maher ER, Riccio A. Genomic Italiana Ricerca sul Cancro IG 2020 N. 24405 (AR), “Progetti per la ricerca imprinting disorders: lessons on how genome, epigenome and environ‑ oncologica della Regione Campania” Grant: I-Cure (AR and FC), “Progetti ment interact. Nat Rev Genet. 2019;20(4):235–48. competitivi intraAteneo” Programma V:ALERE (VAnviteLli pEr la RicErca) 2019 – 2. Sanchez-Delgado M, Riccio A, Eggermann T, Maher ER, Lapunzina P, grant MIRIAM from Università degli Studi della Campania "Luigi Vanvitelli” (AR, Mackay D, Monk D. Causes and consequences of multi-locus imprinting FC and AS), Instituto de Salud Carlos III (ISCIIII) of the Ministry of Economy and disturbances in humans. Trends Genet. 2016;32(7):444–55. Competitiveness (Spain) (to GPdN and AP: PI20/00950), co-financed by the 3. Mantovani G, Bastepe M, Monk D, de Sanctis L, Thiele S, Usardi A, Ahmed European Regional Development Fund, 2019 research unit grant from ESPE SF, Bufo R, Choplin T, De Filippo G, et al. Diagnosis and management (to GdPN). of pseudohypoparathyroidism and related disorders: first international consensus statement. Nat Rev Endocrinol. 2018;14(8):476–500. Availability of data and materials 4. Thiele S, Mantovani G, Barlier A, Boldrin V, Bordogna P, De Sanctis L, Elli Methylation array data that support the findings of this study have been FM, Freson K, Garin I, Grybek V, et al. From pseudohypoparathyroidism to deposited under accession code GSE195873 and G19825 in the Gene Expres‑ inactivating PTH/PTHrP signalling disorder (iPPSD), a novel classification sion Omnibus repository. proposed by the EuroPHP network. Eur J Endocrinol. 2016;175(6):P1–17. 5. Elli FM, Linglart A, Garin I, de Sanctis L, Bordogna P, Grybek V, Pereda A, Giachero F, Verrua E, Hanna P, et al. The prevalence of GNAS deficiency- Declarations related diseases in a large cohort of patients characterized by the EuroPHP network. J Clin Endocrinol Metab. 2016;101(10):3657–68. Ethics approval and consent to participate 6. Perez-Nanclares G, Romanelli V, Mayo S, Garin I, Zazo Seco C, Fernandez- Written informed consent to genetic analyses was obtained from the patients Rebollo E, Martínez F, Lapunzina P, de Nanclares GP, Group SP. Detection or their parents. The ethical principles and the Italian and Spanish legislation of hypomethylation syndrome among patients with epigenetic altera‑ were applied in all steps of the research work that was approved by the Ethical tions at the GNAS locus. J Clin Endocrinol Metab. 2012;97(6):E1060-1067. Committees of the University of Campania “Luigi Vanvitelli” (approval number: 7. Court F, Martin-Trujillo A, Romanelli V, Garin I, Iglesias-Platas I, Salafsky 1135), the Hospital Universitario La Paz (CEIm, PI-659) and the Clinica del I, Guitart M, Perez de Nanclares G, Lapunzina P, Monk D. Genome-wide Hospital Universitari de Bellvitge (PR096/10), and the Basque Clinical Research allelic methylation analysis reveals disease-specific susceptibility to Ethics Committee (CEIC-E, PI2017018). multiple methylation defects in imprinting syndromes. Hum Mutat. 2013;34(4):595–602. Consent for publication 8. Maupetit-Méhouas S, Azzi S, Steunou V, Sakakini N, Silve C, Reynes C, Consent for publication of clinical and molecular data were obtained from all Perez de Nanclares G, Keren B, Chantot S, Barlier A, et al. Simultaneous participants involved in this study. hyper- and hypomethylation at imprinted loci in a subset of patients with GNAS epimutations underlies a complex and different mechanism Competing interests of multilocus methylation defect in pseudohypoparathyroidism type 1b. The authors declare that they have no potential competing interests to Hum Mutat. 2013;34(8):1172–80. disclose. The authors declare that they have no financial relationships relevant 9. Rochtus A, Martin-Trujillo A, Izzi B, Elli F, Garin I, Linglart A, Mantovani G, to this article to disclose. Perez de Nanclares G, Thiele S, Decallonne B, et al. Genome-wide DNA Pignata et al. Clinical Epigenetics (2022) 14:71 Page 14 of 15 28. Wang X, Song D, Mykytenko D, Kuang Y, Lv Q, Li B, Chen B, Mao X, Xu Y, methylation analysis of pseudohypoparathyroidism patients with GNAS Zukin V, et al. Novel mutations in genes encoding subcortical maternal imprinting defects. Clin Epigenet. 2016;8:10. complex proteins may cause human embryonic developmental arrest. 10. Bliek J, Verde G, Callaway J, Maas SM, De Crescenzo A, Sparago A, Reprod Biomed Online. 2018;36(6):698–704. Cerrato F, Russo S, Ferraiuolo S, Rinaldi MM, et al. Hypomethylation at 29. Rezaei M, Nguyen NM, Foroughinia L, Dash P, Ahmadpour F, Verma multiple maternally methylated imprinted regions including PLAGL1 IC, Slim R, Fardaei M. Two novel mutations in the KHDC3L gene in and GNAS loci in Beckwith-Wiedemann syndrome. Eur J Hum Genet. Asian patients with recurrent hydatidiform mole. Hum Genome Var. 2009;17(5):611–9. 2016;3:16027. 11. Bakker B, Sonneveld LJ, Woltering MC, Bikker H, Kant SG. A girl with 30. Alazami AM, Awad SM, Coskun S, Al-Hassan S, Hijazi H, Abdulwahab FM, Beckwith-Wiedemann syndrome and pseudohypoparathyroidism Poizat C, Alkuraya FS. TLE6 mutation causes the earliest known human type 1B due to multiple imprinting defects. J Clin Endocrinol Metab. embryonic lethality. Genome Biol. 2015;16:240. 2015;100(11):3963–6. 31. Lin J, Xu H, Chen B, Wang W, Wang L, Sun X, Sang Q. Expanding the 12. Sano S, Matsubara K, Nagasaki K, Kikuchi T, Nakabayashi K, Hata K, genetic and phenotypic spectrum of female infertility caused by TLE6 Fukami M, Kagami M, Ogata T. Beckwith-Wiedemann syndrome and mutations. J Assist Reprod Genet. 2020;37(2):437–42. pseudohypoparathyroidism type Ib in a patient with multilocus imprint‑ 32. Sanchez-Delgado M, Martin-Trujillo A, Tayama C, Vidal E, Esteller M, ing disturbance: a female-dominant phenomenon? J Hum Genet. Iglesias-Platas I, Deo N, Barney O, Maclean K, Hata K, et al. Absence of 2016;61(8):765–9. maternal methylation in biparental hydatidiform moles from women 13. Sparago A, Verma A, Patricelli MG, Pignata L, Russo S, Calzari L, De Franc‑ with NLRP7 maternal-effect mutations reveals widespread placenta- esco N, Del Prete R, Palumbo O, Carella M, et al. The phenotypic variations specific imprinting. PLoS Genet. 2015;11(11):e1005644. of multi-locus imprinting disturbances associated with maternal-effect 33. Rezaei M, Suresh B, Bereke E, Hadipour Z, Aguinaga M, Qian J, Bagga R, variants of NLRP5 range from overt imprinting disorder to apparently Fardaei M, Hemida R, Jagadeesh S, et al. Novel pathogenic variants in healthy phenotype. Clin Epigenet. 2019;11(1):190. NLRP7, NLRP5, and PADI6 in patients with recurrent hydatidiform moles 14. Mackay DJ, Callaway JL, Marks SM, White HE, Acerini CL, Boonen SE, and reproductive failure. Clin Genet. 2021;99(6):823–8. Dayanikli P, Firth HV, Goodship JA, Haemers AP, et al. Hypomethylation of 34. Adzhubei IA, Schmidt S, Peshkin L, Ramensky VE, Gerasimova A, Bork P, multiple imprinted loci in individuals with transient neonatal diabetes is Kondrashov AS, Sunyaev SR. A method and server for predicting damag‑ associated with mutations in ZFP57. Nat Genet. 2008;40(8):949–51. ing missense mutations. Nat Methods. 2010;7(4):248–9. 15. Meyer E, Lim D, Pasha S, Tee LJ, Rahman F, Yates JR, Woods CG, Reik 35. Cimmaruta C, Citro V, Andreotti G, Liguori L, Cubellis MV, Hay Mele B. W, Maher ER. Germline mutation in NLRP2 (NALP2) in a familial Challenging popular tools for the annotation of genetic variations with a imprinting disorder (Beckwith-Wiedemann Syndrome). PLoS Genet. real case, pathogenic mutations of lysosomal alpha-galactosidase. BMC 2009;5(3):e1000423. Bioinform. 2018;19(Suppl 15):433. 16. Docherty LE, Rezwan FI, Poole RL, Turner CL, Kivuva E, Maher ER, Smith‑ 36. Ng PC, Henikoff S. SIFT: predicting amino acid changes that affect protein son SF, Hamilton-Shield JP, Patalan M, Gizewska M, et al. Mutations in function. Nucleic Acids Res. 2003;31(13):3812–4. NLRP5 are associated with reproductive wastage and multilocus imprint‑ 37. Pandurangan AP, Ochoa-Montaño B, Ascher DB, Blundell TL. SDM: a server ing disorders in humans. Nat Commun. 2015;6:8086. for predicting effects of mutations on protein stability. Nucleic Acids Res. 17. Begemann M, Rezwan FI, Beygo J, Docherty LE, Kolarova J, Schroeder C, 2017;45(W1):W229–35. Buiting K, Chokkalingam K, Degenhardt F, Wakeling EL, et al. Maternal 38. Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, Grody WW, Hegde variants in NLRP and other maternal effect proteins are associated M, Lyon E, Spector E, et al. Standards and guidelines for the interpreta‑ with multilocus imprinting disturbance in offspring. J Med Genet. tion of sequence variants: a joint consensus recommendation of the 2018;55(7):497–504. American college of medical genetics and genomics and the association 18. Cubellis MV, Pignata L, Verma A, Sparago A, Del Prete R, Monticelli M, for molecular pathology. Genet Med. 2015;17(5):405–24. Calzari L, Antona V, Melis D, Tenconi R, et al. Loss-of-function maternal- 39. Jumper J, Evans R, Pritzel A, Green T, Figurnov M, Ronneberger O, Tunyas‑ effect mutations of PADI6 are associated with familial and sporadic uvunakool K, Bates R, Žídek A, Potapenko A, et al. Highly accurate protein Beckwith-Wiedemann syndrome with multi-locus imprinting distur‑ structure prediction with AlphaFold. Nature. 2021;596(7873):583–9. bance. Clin Epigenet. 2020;12(1):139. 40. Varadi M, Anyango S, Deshpande M, Nair S, Natassia C, Yordanova G, 19. Eggermann T, Kadgien G, Begemann M, Elbracht M. Biallelic PADI6 vari‑ Yuan D, Stroe O, Wood G, Laydon A, et al. AlphaFold Protein Structure ants cause multilocus imprinting disturbances and miscarriages in the Database: massively expanding the structural coverage of protein- same family. Eur J Hum Genet. 2021;29(4):575–80. sequence space with high-accuracy models. Nucleic Acids Res. 20. Bebbere D, Albertini DF, Coticchio G, Borini A, Ledda S. The subcortical 2022;50(D1):D439–44. maternal complex: emerging roles and novel perspectives. Mol Hum 41. Eggermann T, Soellner L, Buiting K, Kotzot D. Mosaicism and uniparental Reprod. 2021. https://​doi.​org/​10.​1093/​molehr/​gaab0​43. disomy in prenatal diagnosis. Trends Mol Med. 2015;21(2):77–87. 21. Zhu K, Yan L, Zhang X, Lu X, Wang T, Yan J, Liu X, Qiao J, Li L. Identifica‑ 42. Parry DA, Logan CV, Hayward BE, Shires M, Landolsi H, Diggle C, Carr I, tion of a human subcortical maternal complex. Mol Hum Reprod. Rittore C, Touitou I, Philibert L, et al. Mutations causing familial biparental 2015;21(4):320–9. hydatidiform mole implicate c6orf221 as a possible regulator of genomic 22. Qian J, Nguyen NMP, Rezaei M, Huang B, Tao Y, Zhang X, Cheng Q, Yang H, imprinting in the human oocyte. Am J Hum Genet. 2011;89(3):451–8. Asangla A, Majewski J, et al. Biallelic PADI6 variants linking infertility, mis‑ 43. Demond H, Anvar Z, Jahromi BN, Sparago A, Verma A, Davari M, Calzari carriages, and hydatidiform moles. Eur J Hum Genet. 2018;26(7):1007–13. L, Russo S, Jahromi MA, Monk D, et al. A KHDC3L mutation resulting in 23. Tian X, Pascal G, Monget P. Evolution and functional divergence of NLRP recurrent hydatidiform mole causes genome-wide DNA methylation loss genes in mammalian reproductive systems. BMC Evol Biol. 2009;9:202. in oocytes and persistent imprinting defects post-fertilisation. Genome 24. Gao Z, Zhang X, Yu X, Qin D, Xiao Y, Yu Y, Xiang Y, Nie X, Lu X, Liu W, et al. Med. 2019;11(1):84. Zbed3 participates in the subcortical maternal complex and regulates 44. Fatemi N, Ray PF, Ramezanali F, Shahani T, Amiri-Yekta A, Kherraf ZE, Cazin the distribution of organelles. J Mol Cell Biol. 2018;10(1):74–88. C, Almadani N, Varkiani M, Sarmadi S, et al. KH domain containing 3 like 25. Akoury E, Zhang L, Ao A, Slim R. NLRP7 and KHDC3L, the two maternal- (KHDC3L) frame-shift mutation causes both recurrent pregnancy loss and effect proteins responsible for recurrent hydatidiform moles, co-localize hydatidiform mole. Eur J Obstet Gynecol Reprod Biol. 2021;259:100–4. to the oocyte cytoskeleton. Hum Reprod. 2015;30(1):159–69. 45. Reddy R, Akoury E, Phuong Nguyen NM, Abdul-Rahman OA, Dery C, 26. Qin D, Gao Z, Xiao Y, Zhang X, Ma H, Yu X, Nie X, Fan N, Wang X, Ouyang Gupta N, Daley WP, Ao A, Landolsi H, Ann Fisher R, et al. Report of four Y, et al. The subcortical maternal complex protein Nlrp4f is involved in new patients with protein-truncating mutations in C6orf221/KHDC3L cytoplasmic lattice formation and organelle distribution. Development. and colocalization with NLRP7. Eur J Hum Genet. 2013;21(9):957–64. 2019. https://​doi.​org/​10.​1242/​dev.​183616. 46. Nakka P, Pattillo Smith S, O’Donnell-Luria AH, McManus KF, Mountain 27. Elbracht M, Mackay D, Begemann M, Kagan KO, Eggermann T. Disturbed JL, Ramachandran S, Sathirapongsasuti JF. Team aR: characterization genomic imprinting and its relevance for human reproduction: causes of prevalence and health consequences of uniparental disomy in four and clinical consequences. Hum Reprod Update. 2020;26(2):197–213. Pignata et al. Clinical Epigenetics (2022) 14:71 Page 15 of 15 million individuals from the general population. Am J Hum Genet. 2019;105(5):921–32. 47. Zheng P, Dean J. Role of Filia, a maternal effect gene, in maintaining euploidy during cleavage-stage mouse embryogenesis. Proc Natl Acad Sci U S A. 2009;106(18):7473–8. 48. Zheng P, Baibakov B, Wang XH, Dean J. PtdIns(3,4,5)P3 is constitutively synthesized and required for spindle translocation during meiosis in mouse oocytes. J Cell Sci. 2013;126(Pt 3):715–21. 49. Pignata L, Sparago A, Palumbo O, Andreucci E, Lapi E, Tenconi R, Carella M, Riccio A, Cerrato F. Mosaic segmental and whole-chromosome Upd(11)mat in silver-russell syndrome. Genes (Basel). 2021;12(4):581. 50. Aghajanova L, Mahadevan S, Altmäe S, Stavreus-Evers A, Regan L, Sebire N, Dixon P, Fisher RA, Van den Veyver IB. No evidence for mutations in NLRP7, NLRP2 or KHDC3L in women with unexplained recurrent preg‑ nancy loss or infertility. Hum Reprod. 2015;30(1):232–8. 51. Miller SA, Dykes DD, Polesky HF. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res. 1988;16(3):1215. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. Ready to submit your research ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations • maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions