MN1 gene loss-of-function mutation causes cleft palate in a pedigree

We read with great interest the article entitled ‘MN1 C-terminal truncation syndrome is a novel neurodevelopmental and craniofacial disorder with partial rhombencephalosynapsis’ published in Brain by Mak et al. (2020). They reported 22 patients with CEBALID syndrome who showed severe symptoms including global developmental delay, craniofacial and brain abnormalities. All of these patients carried MN1 heterozygous C-terminal mutations which may have a dominant-negative or gain-of-function effect.

MN1 (MIM 156100) was initially identified to be disrupted by balanced chromosomal translocation and play a role in meningioma and myeloproliferative diseases (Buijs et al., 1995; Lekanne Deprez et al., 1995). There were no MN1-related phenotypes established in humans until Mak et al. first described the gene as a causative gene for MN1 C-terminal truncation (MCTT) syndrome-CEBALID syndrome (MIM 618774) in 2020. They found that these mutations lay in exon 2 or 3′ of exon 1 and were predicted to escape from the nonsense-mediated mRNA decay (NMD) (Mak et al., 2020). Miyake et al. (2020) also reported three patients carrying MN1 C-terminal mutations with similar severe clinical features. They proved that aberrant truncated MN1 proteins acted in a gain-of-function manner by increasing protein stability, inhibiting cell proliferation and enhancing MN1 nuclear aggregation (Miyake et al., 2020).

Although the mutations in the N-terminus of the gene were predicted to cause the distinct clinical features due to MN1 haploinsufficiency, functional studies were not performed on mechanisms underlying the milder MN1-related phenotype. In support and to supplement the message of this paper, our study identified a mutation located outside the C-terminal region of MN1 in a Chinese pedigree and performed expression experiments for the first time.

We identified a heterozygous frameshift deletion variant in two patients in a Chinese pedigree. Proband whole exome sequencing was conducted and variants were validated by Sanger sequencing in other members in the pedigree including the parents (Subjects II-1 and II-2), the aunt (Subject II-3), and the maternal grandparents (Subjects I-1 and I-2). A frameshift deletion located in the middle of MN1 exon 1 [NM_002430.3:c.2253del, p.G752Afs*12 (chr22:27798291-27798291)] was identified resulting in a complete co-segregation within the pedigree (Fig. 1A and B). All participants signed informed consent forms, and the study was approved by the ethics committee of the Maternal and Child Health Hospital of Hunan Province (2020-S003).

Figure 1.

Pedigree and Sanger sequencing. (A) Pedigree with MN1 loss-of-function variant. Individuals with heterozygous variants and without the variant are indicated by plus/minus (+/−) and minus/minus (−/−) symbols, respectively. (B) Sanger sequencing results of MN1 loss-of-function variant in family members. (C) Normal facial features of the proband (Subject III-1) and the mother of proband (Subject II-2). The arrows point to the cleft palate. (D) Systematic ultrasonography assessment of the foetus showed in some sections, the median palatal suture of foetal hard palate is slightly wider (0.17–0.21 cm).

The two patients showed similar clinical phenotypes as cleft palate and conductive hearing loss. No severe clinical phenotypes in CEBALID syndrome were observed including developmental delay, craniofacial features or characteristic brain imaging (Mak et al., 2020). The proband (Subject III-1) is a 9-year-old female, born at full term (38 weeks) following a normal pregnancy. Cleft palate was present at birth and at the age of 1.5 years, cleft palate repair surgery was carried out (Fig. 1C). The mother of the proband (Subject II-2) is a 30-year-old female, born at full term (40 weeks) following a normal pregnancy. Cleft palate was present at birth and repair surgery was carried out at 22 years of age (Fig. 1C).

Pure tone audiometry (PTA) (Supplementary Fig. 1), tympanometry (Supplementary Fig. 2) and acoustic reflex testing (Supplementary Fig. 3) of the two patients (Subjects III-1 and II-2) showed conductive hearing loss, suggesting the presence of external and middle ear lesions or functional abnormalities. Physical examination of the ear showed that external auditory canals were patent and tympanic membranes were intact in both patients. The proband and the mother of the proband were of normal intelligence with IQ scores of 96 and 93, respectively. Brain MRI was normal.

At 23 weeks of a second pregnancy the systematic ultrasonography assessment of the foetus showed sonographic changes of palatal suture in the foetal hard palate (Fig. 1D). No other dysmorphism or developmental delay was observed in the foetus. Invasive prenatal diagnosis by amniocentesis was performed and Sanger sequencing was performed to detect the variant. The same variant in MN1 was identified (Fig. 1B).

To confirm the pathogenesis of the variant in MN1, in vitro functional expression studies were performed on peripheral blood lymphocytes of patients and controls in the pedigree. Lymphocytes were separated from blood using lymphocyte isolation solution (Ficoll^® Paque Plus 17-1440-02). Quantitative real-time PCR (RT-PCR) was used to analyse the RNA expression of MN1. Briefly, total mRNA was prepared for RT-PCR using a RNeasy^® Mini Kit (Qiagen) and cDNA was synthesized with the AMV First Strand cDNA Synthesis Kit (SK2445). The sequences of primers were MN1-F, 5′-GCTTTCCGTTTGGTGCAG-3′; MN1-R, 5′-TGGAAATCAGGCTGCGG-3′; b-actin-F, 5′-TAGTTGCGTTACACCCTTTCTTG-3′; and b-actin-R , 5′-TCACCTTCACCGTTCCAGTTT-3′. RT-PCR was carried out using SYBR^TM Green PCR master mix (Applied Biosystems) according to the manufacturer’s instructions. PCR amplifications (40 cycles at 95°C for 15 s and 57°C for 20 s) were performed using an ABI 7900 (Applied Biosystems). The average Ct values calculated from quadruplicate PCR reactions were normalized to the average Ct values for β-actin (ACTB). The normalized values were used to calculate the expression of MN1 relative to the controls using the ΔΔCt method. RT-PCR analysis showed significantly reduced levels of mRNA expression in two patients (Subjects II-2 and III-1) compared with unaffected control subjects (Subjects I-1, I-2, II-1 and II-3) in the pedigree (Fig. 2A), suggesting NMD-mediated mRNA degradation of MN1 mRNA.

Figure 2 Expression level analyses.

(A) RT-PCR analysis showed decreased MN1 mRNA expression level in patients compared to controls in lymphocytes. β-Actin mRNA was used as an internal control for normalization and quantitative analysis of MN1 mRNA levels. P < 0.01, Student’s t-test. (B) Western blot analysis showing that MN1 protein levels were decreased in patients compared with control subjects.

Western blot was carried out to analyse MN1 protein. Total proteins were extracted by RIPA buffer (Biosesang) and subjected to the regular western blot procedure. Antibodies used for the determination of MN1 protein included anti-MN1 (1:1000, SAB4501908, Sigma), anti-MN1 (1:1000, PA5-38666, ThermoFisher), anti-β-actin (1:2000), and anti-Rabbit IgG (1:5000, Sigma-Aldrich). Enhanced chemiluminescence (ECL) plus western blotting substrate (Thermo Scientific) was used for detection of protein bands. The level of MN1 protein detected in two patients (Subjects II-2 and III-1) was significantly lower than in four unaffected individuals (Subjects I-1, I-2, II-1 and II-3) in the pedigree (Fig. 2B) suggesting an MN1 haploinsufficiency caused by MN1 loss-of-function mutation.

The MN1 gene comprises two exons and encodes a 1320 amino acid protein; exon 1 encodes amino acids 1–1260 and exon 2 encodes the remaining 60 C-terminal amino acids. Deletions, including the whole gene sequence of MN1, have been reported previously to be associated with variable clinical phenotypes including neurodevelopmental anomalies, facial dysmorphisms, congenital heart defects, etc. (Said et al., 2011; Davidson et al., 2012; Breckpot et al., 2016; Mak et al., 2020). However, these deletions contained many MIM genes and it is difficult to judge the gene contributions. Until 2020, the truncating variants in the C-terminal region of the gene were recognized to cause a specific disease—CEBALID syndrome—which was characterized by CNS and craniofacial skeletal development disorders (Mak et al., 2020). Most of the heterozygous variants associated with CEBALID syndrome were de novo mutations, except for one family in which two affected brothers inherited the mutation from a mildly affected father who carried a somatic mosaic mutation. In the same year, a functional study revealed that the C-terminal variants escaped the NMD system and increased the amount of mutant MN1 protein, which could explain the severe clinical manifestations of the syndrome (Miyake et al., 2020).

The molecular mechanism by which loss-of-function mutations of MN1 lead to mild clinical features in the patients is not known. In our study, we first identified a loss-of-function variant located in the middle of exon 1 in two patients in a pedigree and proved that MN1 causes a mild phenotype of disease by NMD leading to MN1 haploinsufficiency by expression experiments. The heterozygous Mn1 knockout mice showed an intermediate phenotype with hypoplastic membranous bone and incomplete penetrance of the cleft palate when compared to homozygous knockout Mn1 (Meester-Smoor et al., 2005). Therefore, combined with the former research on C-terminal MN1 variants, the clinical severity of patients carrying MN1 variants may be related to a dosage effect of the MN1 protein. The truncated MN1 protein is a transcriptional cofactor and the mutant protein could impair the binding with transcription factors and cause dysregulation of target genes (Meester-Smoor et al., 2005; Miyake et al., 2020). It is plausible that the distinct clinical phenotypes determined by different regions of MN1 were through regulation of corresponding downstream genes (Lai et al., 2014). It has been reported that MN1 may interplay with transcription factors regulating palate development genes such as Tbx22 (Liu et al., 2008). Our study adds to the growing evidence of genotype-phenotype correlations of MN1 and provides clues for future mechanism research on MN1 loss-of-function variants.

 

Data availability

The data that support the findings of this study are available from the corresponding author, upon reasonable request.

Competing interests

The authors report no competing interests.

Funding

This work was supported by the National Natural Science Foundation of China (grant number 81801136), Hunan Provincial Major Science and Technology Project (grant number 2019SK1010) and the China Postdoctoral Science Foundation (2019M662804).

Supplementary material

Supplementary material is available at Brain online.