Neonatal mucolipidosis type II alpha/beta due to compound heterozygosity for a known and novel GNPTAB mutation, and a concomitant heterozygous change in SERPINF1 inherited from the mother.

We report on a newborn with IUGR, rhizomelic dwarfism, and suspected chondrodysplasia punctata. At birth, OI was suspected; however, a skeletal survey suggested ML II alpha/beta. Sequencing revealed compound heterozygosity for a reported pathogenic and novel but expected pathogenic GNPTAB variant. Molecular testing for autosomal recessive OI identified a SERPINF1 variant.

Mucolipidosis type II alpha/beta (ML II alpha/beta), also known as inclusion‐cell (I‐cell) disease (OMIM 252500), is a rare autosomal recessive lysosomal storage disorder that has clinical onset at birth, and results in death most often in early childhood. ML II alpha/beta is characterized by limited postnatal growth, coarse facial features, thickened skin, generalized hypotonia, and hypertrophic gingiva. In addition, dysostosis multiplex, clubfeet, hip dislocation, movement limitation at the shoulder joints, thoracic deformity including kyphosis, and long bone deformities are present at birth 1. Premature death occurs during the first decade of life due to hardening and subsequent leaking of cardiac valves, progressive mucosal thickening of the airways, and stiffening of the thoracic cage 2. ML II alpha/beta is caused by a homozygous or compound heterozygous mutation of the GNPTAB (OMIM 607840) gene, located on chromosome 12 at 12q23.2 3. Mutations in GNPTAB result in a deficiency in N‐acetylglucosamine‐1‐phosphotransferase (GlcNAc‐1‐phosphotransferase), which is responsible for the trafficking of most lysosomal hydrolases and provides lysosomal enzymes with mannose‐6‐phosphate (M6P) in order for recognition by the M6P receptor to occur. This deficiency results in improper lysosomal enzyme phosphorylation, causing an accumulation of lysosomal substrates in several organs and tissues, specifically the skeletal system 4.

We report on a newborn infant that was born to a healthy 37‐year‐old African American father and a healthy, nonconsanguineous, 31‐year‐old G5P4L3 African American mother. Family evaluation revealed that the father is tall (6′6″) and macrocephalic and has a history of thumb subluxation. The mother was of slightly above average height (5′7″), had a history of ovarian cysts, presented hypertelorism, and had long fingers with extra finger flexion creases. The proposita was the product of the mother's fifth pregnancy but the first pregnancy for this couple. The mother's first four children share paternity, and one of the children died at age 2 weeks from sudden infant death syndrome (SIDS). Otherwise, family history of ML II alpha/beta or other conditions was negative. Prenatal ultrasound at 33 weeks of gestational age (WGA) and again at 366/7 WGA was significant for intrauterine growth retardation, and rhizomelia with possible intrauterine fractures and punctate calcifications. This raised concern for rhizomelic chondrodysplasia punctata; however, targeted mutation analysis for PEX7 and array comparative genomic hybridization were normal.

The proposita was born at 37 WGA via induced vaginal delivery. Her birthweight was 2154 g (3rd centile) and her birth length was 42.25 cm (<3rd centile), and OFC was 32.0 cm (10th centile). Apgar scores were 5, 7, and 8 at 1, 5, and 10 min, respectively. She received continuous positive airway pressure (CPAP) and oxygen due to mild respiratory distress. She was hypotonic, with a weak cry and cyanotic and dysmorphic features noted: She had coarse face and gingival hyperplasia, slightly posteriorly rotated ears, creases in her forehead, anteverted nostrils with short columella, rhizomelic shortening of all extremities, long fingers and toes with tendency to clinodactyly of the second and third toes, and camptodactyly of the third, fourth, and fifth fingers with some overriding fingers (Fig. 1A–E). Due to the prenatal skeletal findings, a skeletal survey was performed, revealing prominent cranium without evidence of wormian bones, thinning of the cortex of the long bones with some bowing, and deformity of the femurs and radii which were interpreted as possible fractures. These findings raised the possible diagnosis of osteogenesis imperfecta (OI).

Figure 1

(A–E) Proposita at age 1 day. (A–C) Hypertelorism, rhizomelic limbs, rotated ears, anteverted nostrils, micrognathia, and short columella were observed. (D) Long fingers. (E) Camptodactyly and overriding fingers. (F) Skull base is sclerotic with absence of wormian bones. Undermineralization of facial bones was observed. (G) Thin ribs with bulbous costochondral junctions. (H, I): Marked coarsening of trabecula with periosteal cloaking and metaphyseal cupping, flaring, and irregularity with bending metaphyseal fractures observed in long bones. Note the stippled calcification within the poorly ossified calcaneus bilaterally (arrows).

At this time, genetics was consulted. Initial consideration was an autosomal recessive form of OI. However, the patient was transferred to Children's Hospital NICU and additional radiographs were performed at 11 days of age, revealing a sclerotic skull base, undermineralization of the facial bones, and thin ribs with bulbous costochondral junctions (Fig. 1F and G) with marked coarsening of trabecula with periosteal cloaking, diaphyseal expansion in the bones of the hand with proximal and distal constriction, and metaphyseal cupping, flaring and irregularity with bending metaphyseal fractures and stippled calcifications around the calcaneus (Fig. 1H–K). These radiographic findings suggested a storage disorder. Laboratory workup revealed elevated alkaline phosphatase (>1000 U/L; NV = 90–421 U/L) and normal calcium levels (9.2 mg/dL; NV = 8.5–10.4 mg/dL), and PTH was not assessed. Given the high index of suspicion for a storage disorder, a blood smear was requested, identifying the presence of large vacuoles. A ML II plasma screen detected significant elevations in beta‐glucuronidase (1611.8 nmol/h/mL; NV = 4.8–110.8 nmol/h/mL), alpha‐fucosidase (4070.1 nmol/h/mL; NV = 50.4–742.9 nmol/h/mL), and beta‐hexosaminidase (12,737.3 nmol/h/mL; NV = 250–1568 nmol/h/mL). In patients with ML II alpha/beta, nearly all lysosomal hydrolases are 5‐ to 20‐fold higher in plasma and other body fluids than in normal controls due to improper targeting of lysosomal acid hydrolases to lysosomes 5.

These abnormal skeletal and serum biochemistry findings were consistent with the clinical diagnosis of ML II alpha/beta. GNPTAB gene sequencing revealed compound heterozygosity for a pathogenic sequence variant already reported in ML II alpha/beta (c.1399delG; p.Asp467Ilefs*33) 6 and also a novel but expected pathogenic sequence variant (c.1905_1908delAAGG; p.Glu637Aspfs*4). Both variants are predicted to result in a frameshift and premature termination of the protein GlcNAc‐1‐phosphotransferase subunits alpha/beta, therefore resulting in ML II alpha/beta 7. These two very severe mutations identified in the GNPTAB gene may explain the severe phenotype observed in the proposita. Ma et al. 8 also reported an infant with two compound heterozygous GNPTAB mutations, resulting in the diagnosis of ML II alpha/beta. The infant reported by 8 is of Han Chinese descent and presented a marked hair color change at age 2 months, dysostosis multiplex at age 2 months, and severe coarse facies at age 1 year. Septic shock with respiratory failure resulted in the death of this infant at age 14 months 8.

As mentioned, initial examination raised suspicion of one of the rare types of recessive OI. OI, also known as “brittle bone disease,” affects 6–7/100,000 people at birth and is associated with several gene mutations with varying grades of severity and expressivity 9, 10. The majority of cases are diagnosed as one of the four classical types of OI, which are caused by mutations in either the COL1A1 or COL1A2 gene 11, 12, 13. However, 10% of OI cases are due to causative variants which follow an autosomal recessive mode of inheritance 14. These rare forms of OI arise from mutations in several genes causing proteins to affect the biosynthesis of collagen type I 14. NGS panel for recessive OI was performed on the proposita, revealing heterozygous variant (c.1004A>G transition) in exon 8 of the SERPINF1 gene, resulting in the conversion of a codon for glutamine (CAA) to a codon for arginine (CGA). SERPINF1, located on chromosome 17 at 17p13.3 (OMIM 172860), encodes for pigment epithelium‐derived factor (PEDF), which is a protein that inhibits angiogenesis and induces neuronal differentiation in retinoblastoma cells 15. Becker et al. 9 proposed that a loss of PEDF due to a SERPINF1 mutation may result in an OI phenotype that is independent of collagen type I synthesis alterations. SERPINF1 loss‐of‐function mutations are responsible for the manifestation of OI type VI (OMIM 613982), which is one of the autosomal recessive forms of the disorder 9. OI type VI is an extremely rare form of the disorder, with only 30 reported cases up to date. OI type VI is distinguished by bone mineralization defects and moderate severity. In the 30 cases reported, 18 unique SERPINF1 mutations have been described, including frameshift, nonsense, in‐frame insertion, and in‐frame deletion mutations 16. The identification of several unique SERPINF1 mutations in the limited cohort of patients suggests that there may be additional unidentified SERPINF1 mutations that may cause OI type VI. To the best of our knowledge, this change has not been previously reported as either a mutation or polymorphism. The biological significance of this change is not known for certain, and clinical correlation is required.

Following the proposita's NGS results revealing the heterozygous variant in the SERPINF1 gene, parental studies were conducted. The father had no mutation; however, the mother presented an identical SERPINF1 mutation. It is not yet clear whether this variant is of biological significance or not. We suspect that the severe phenotype observed prenatally in the proposita may be due to complex interactions between the two variants identified in the GNPTAB gene.

In summary, we report on a newborn with severe intrauterine growth restriction, rhizomelic dwarfism with bowed femurs, and skeletal abnormalities detected prenatally in which the diagnosis of ML II alpha/beta (I‐cell disease) was confirmed postnatally. The patient was a compound heterozygote for a previously known pathogenic GNPTAB sequence variant (c.1399delG; p.Asp467Ilefs*33) as well as a novel but expected pathogenic GNPTAB sequence variant (c.1905_1908delAAGG; p.Glu637Aspfs*4). We would like to mention that as osteogenesis imperfecta was initially considered, molecular testing for rare AR forms of OI was performed, identifying a maternally inherited heterozygous variant in the SERPINF1 gene (c.1004A>G transition) related to OI type VI. The contribution and significance of this variant to the phenotype is unknown. We suspect that the identified compound heterozygosity for two GNPTAB variants is responsible for the severe ML II alpha/beta phenotype observed. Only when exome or genome sequencing is of daily use, clinicians may be able to recognize the importance of the interaction with the rest of the genome in the severity of clinical manifestations.

Authorship

KAW and YL: drafted the manuscript. KAW, RMZ, and YL: critically revised the manuscript. BJC, CA, MH, JS, and YL: performed clinical analysis, including pathological and radiological evaluations, and newborn management. SS and JCH: performed sequencing analysis.

Conflict of Interest

None declared.