Identification of a molecular defect in a stillborn fetus with perinatal lethal hypophosphatasia using a disease-associated genome sequencing approach

Introduction

Lethal skeletal disorders are an extremely heterogeneous group of genetic conditions, which pose a significant diagnostic challenge to the clinical geneticist, especially when resulting in early prenatal death of the fetus [1]. One example of such disease is hypophosphatasia (HPP), an inborn error of bone metabolism, historically classified as skeletal dysplasia. HPP is a rare genetic disorder characterized by decreased bone density and reduced activity of alkaline phosphatase, resulting from mutations in the liver/bone/kidney alkaline phosphatase gene – ALPL (MIM171760) – also referred to as tissue non-specific alkaline phosphatase (TNAP) [2, 3]. The gene is located on chromosome 1p36.1, composed of 12 exons, and encodes for a protein hydrolyzing phosphomonoesters with generation of inorganic phosphate [4, 5]. TNAP protein binds to the cell surface of osteoblasts, chondrocytes, and their matrix vesicles, playing an essential role in bone mineralization and formation of primary teeth [6, 7]. Loss of protein function leads to the accumulation of three substrates: inorganic pyrophosphate (PPi), which is a known inhibitor of calcification, pyridoxal 5’-phosphate (PLP), and phosphoethanolamine (PEA) [8].
Six clinical forms of HPP have been traditionally described in the literature, depending on the severity of symptoms and age of onset, i.e. perinatal lethal, prenatal benign, infantile (OMIM #241500), childhood (OMIM #241510), adult, and odontohypophosphatasia (OMIM #14630) [9]. All forms of HPP present with low serum alkaline phosphatase activity and variable bone involvement. The latter feature is however not observed in odontohypophosphatasia, which is characterized by dental abnormalities with no overt skeletal disease. HPP exhibits extreme clinical heterogeneity, ranging from lethal perinatal forms resulting in intrauterine or postnatal death to less severe phenotypes not influencing the lifespan. The broad clinical spectrum of HPP includes poor bone mineralization, thoracic and cranial deformities, long bone shortening and/or bowing, micromelia, and pathologic fractures [10, 11]. In the infantile form, apart from bone hypomineralization and fractures, premature loss of teeth, craniosynostosis, failure to thrive, hypotonia, and pulmonary insufficiency have been described. The milder types (childhood and adult) include delayed gross motor development, myopathy, low bone density, chronic pain of muscles and/or bones, premature loss of teeth, abnormal dentition prone to caries, osteomalacia or pseudogout [10, 11, 12, 13]. The most severe forms of HPP are transmitted in an autosomal recessive manner, whereas the milder types usually result from autosomal dominant mutations or less frequently recessive alterations. Genotype-phenotype correlations suggest that the most severe mutations are located in the active site and its vicinity, the active site valley, the homodimer interface, the crown domain, and the metal binding site of TNAP [7, 14, 15, 16].
The introduction of next generation sequencing (NGS) technologies into medical practice has led to unravelling many genes involved in the pathogenesis of human genetic disorders. Most of the NGS-based studies have used whole-exome sequencing (WES) or whole-genome sequencing (WGS) approaches, which are known to generate massive amounts of uninterpretable sequence variants with unclear clinical significance. In the diagnostic setting, gene panels, although providing less data, seem to be the optimal solution, as they usually allow for reliable interpretation of the detected variants. One interesting example of a comprehensive gene panel strategy was proposed in 2014 by Zemojtel et al. [17] and comprised selective targeting of the “disease-associated genome” (DAG), encompassing a total of 2741 known Mendelian genes (all known at that time).
In this report we applied NGS-based DAG panel sequencing in order to establish a diagnosis in a stillborn child affected by severe lethal skeletal dysplasia of unknown origin. With this approach, we detected compound heterozygous ALPL mutations demonstrating that the fetus presented with the perinatal type of HPP.

Clinical description

The index fetus affected by a prenatally unrecognizable skeletal disorder was stillborn at 22 weeks of a second gestation. The proband was conceived by a healthy, non-consanguineous couple (31-year-old mother and father), who already had a healthy daughter born from the first pregnancy. The family history revealed that the maternal grandmother of the index case had a single spontaneous abortion, while the sister of the maternal grandfather presented with short stature and rickets. Prenatal ultrasound scans performed in the index fetus at 18 weeks of gestation showed a 4-week delay in femoral length (FL) and humeral length (HL) in comparison with biparietal diameter (BPD), occipitofrontal diameter (OFD), and head circumference (HC). Amniocentesis performed at the same time allowed for conventional cytogenetic GTG banding and showed normal male karyotype (46,XY). Repeated ultrasound examination at 21+2 weeks of gestation revealed abnormal shape and significant shortening of the lower limbs (FL of 24.8 mm; Fig. 1A, B) and moderate shortening of the upper limbs (HL of 33.0 mm) in reference to BPD, OFD, and HC. In addition, the FL to AC ratio was recorded as 14%. Five days later (at 22 weeks of gestation), ultrasound examination revealed absence of the fetal heart rate, and the mother spontaneously delivered a stillborn male fetus, with a birth weight of 530 g. Autopsy examination confirmed that the fetus had severely shortened lower extremities and spontaneous intrauterine fracture of the right lower leg bones, with signs of bone healing. In addition, generalized ischaemia of the internal organs was noted.
The Institutional Review Board at the Poznan University of Medical Sciences approved the study, and written informed consent was obtained from all subjects or their legal guardians. Postnatal fetal photographs were unavailable for publication.

Methods

Next generation sequencing panel

Since no clinical diagnosis could be established in our proband, we decided to perform an NGS panel targeting all 2741 genes known to be associated with Mendelian disorders (the disease-associated genome – DAG) [17]. A chorionic sample for genetic testing was taken after placental delivery of the stillborn fetus. Next, genomic DNA was extracted according to standard procedures and subjected to next generation sequencing as described elsewhere [17]. The sample was sequenced using an Illumina HiSeq 1500 sequencer with a mean coverage of 360 reads, with 98% of the target region covered by at least 20 reads. In order to extract potentially causative variants from NGS data, we applied a computational algorithm termed Phenotypic Interpretation of eXomes (PhenIX). The software allows one to filter and rank the variants on the basis of population frequency, predicted pathogenicity as well as clinical relevance by calculating the match between observed symptoms, entered in the form of human phenotype ontology (HPO) terms, and detected gene alterations [17, 18]. In order to calculate the score for clinical relevance we entered the following HPO terms: lethal skeletal dysplasia (HP:0005716), rhizomelia (HP:0008905), and increased susceptibility to fractures (HP:0002659).

Sanger sequencing analysis

Genomic DNA of the parents was extracted from peripheral blood leukocytes using standard protocols. In order to validate the NGS results and test the parents we performed targeted Sanger sequencing. The reactions were performed with dye terminator chemistry (ABI Prism DigDye v3.1) and run on an automated sequencer: Applied Biosystems Prism 3700 DNA Analyzer. The sequencing results were visualized using Bioedit software.

Results

Because the fetal phenotype was strongly suggestive for skeletal disorder (probably skeletal dysplasia) and a postnatal baby-gram was not performed, we decided to apply NGS technology to establish the diagnosis at a molecular level. The NGS DAG panel revealed the compound heterozygous ALPL missense mutations c.1283G>C (p.Arg428Pro) in exon 10 and c.1363G>A (p.Gly455Ser) in exon 11 (NM_000478, NP_000469; Fig. 2A, B), which seemed to be the only variants relevant to the clinical phenotype. PhenIX used to interpret the NGS results ranked the two variants in 3rd place with a gene relevance score of 0.781 and a variant score of 1.00 (a total score of 0.891). Both mutations were further confirmed by Sanger sequencing in the index patient (Fig. 2C, D). Next, parental studies demonstrated that c.1283G>C (p.Arg428Pro) mutation was maternal (Fig. 2C), whereas c.1363G>A (p.Gly455Ser) was inherited from the father (Fig. 2D), corroborating “in trans” orientation of both alterations in our proband. To further assess the impact of both ALPL mutations on the functionality of alkaline phosphatase, we tested its activity in the serum of both parents. The mother, who was the carrier of the p.Arg428Pro variant, was shown to have enzyme activity of 16 U/l, while the father – a carrier of the p.Gly455Ser variant – had activity of 50 U/l (reference range: 40-150 U/l).

Discussion

Lethal skeletal disorders are an extremely heterogeneous group of genetic conditions, usually recognizable on the basis of prenatal ultrasound, autopsy examination, and post mortem radiography (baby-gram). Several studies have shown that a specific diagnosis can be achieved in 77% or 84% of prenatally suspected skeletal dysplasia cases, provided availability of detailed and accurate phenotypic data [19, 20]. In this report, we demonstrate an alternative approach that allowed us to identify a specific genetic diagnosis in a stillborn fetus with lethal skeletal disease, in whom correct post mortem examination (including X-ray) was not performed. Using NGS-based DAG panel sequencing, we were able to demonstrate a molecular diagnosis, which turned out to be perinatal lethal hypophosphatasia (pl-HPP). This severe form of HPP represents an inborn defect of ossification resulting in either stillbirth or early postnatal death. Clinical features comprise shortened lower limbs often with femoral and tibial bowing, multiple fractures, short ribs, narrow thorax, poorly ossified epiphyses, large fontanelles, osteochondral spurs, apnoea, and hypercalcaemia [20, 21, 23]. The fetal demise results mainly from generalized bone hypomineralization and respiratory failure due to thorax deformities and lung hypoplasia [9]. In general, the diagnosis of HPP is based on the prenatal ultrasound scan, postnatal X-ray imaging, and evaluation of serum ALP activity. Other laboratory findings of HPP include elevated plasma PLP, elevated serum PPi, and elevated serum or urinary PEA [10].
Upon repeated prenatal ultrasound scans, our patient did not show osteochondral spurs that are pathognomonic for pl-HPP, but presented with shortening of all limbs, lower limb deformities, and an FL to AC ratio of 14%, which falls within the range of lethal skeletal dysplasias [24]. Additionally, autopsy examination showed fractures of the right lower leg. Since stillbirth occurred at 22 weeks of gestation, laboratory blood testing could not be performed. Moreover, postnatal X-ray imaging (a baby-gram) was declined by the physician, so the clinical data were very limited and insufficient to attempt the diagnosis, which could involve at least several genetic conditions, including osteogenesis imperfecta type II, achondrogenesis, campomelic dysplasia, and others (Table I). Using NGS-based DAG panel sequencing, the index was demonstrated to harbour two pathogenic missense alterations in the ALPL gene, i.e. maternally inherited c.1283G>C(p.Arg428Pro) and paternally inherited c.1363G>A(p.Gly455Ser) mutations. Both variants have been previously described in the literature to be associated with hypophosphatasia [25, 26]. The mutation p.Arg428Pro was described by Spentchian et al. [26] in combination with c.997+2T>A (intron 9) splice site mutation in a male newborn presenting with pl-HPP, who died at day 14 after birth due to acute respiratory failure. Radiographic examination showed slender ribs, a small bell-shaped thorax, generalized poor bone mineralization, and erosive changes of the epiphyses of all long bones, while the serum ALP level was very low – 0.04 µmol/s*l (reference range: 0.4-4.17 µmol/s*l). The second pathogenic alteration (p.Gly455Ser) detected in our proband was previously described by Draguet et al. [25] in association with the c.341C>G+c.348_349insACCGTC (p.A114G+p.A116_Y117insTV) mutation in a 4-year-old girl affected by the childhood form of HPP. This proband presented with relatively mild symptoms of HPP comprising growth retardation, rickets, premature loss of teeth, bone deformities, and a reduced serum ALP level to 67 and 86 mU/ml (reference range: 100-720 mU/ml). Unlike the patient described by Draguet et al., our index manifested perinatal lethal HPP phenotype, suggesting possible severity of the p.Gly455Ser mutation in case of its association with the p.Arg428Pro allele. Alternatively, carrier state of the very severe p.Arg428Pro mutation in the mother may be of great clinical significance due to the deleterious in utero effect of maternal ALPL mutations on the baby [27]. We also tested ALPL in the serum of both carrier parents, demonstrating that heterozygosity for p.Arg428Pro and p.Gly455Ser variants was associated with enzymatic activity of 16 U/l and 50 U/l, respectively (reference range: 40-150 U/l), which further confirmed pathogenicity of the mutations. Finally, location of both mutations in the vicinity of the catalytically active site (position 455) or in the crown domain (position 428) supports their damaging effect on ALPL function [28, 29].
In conclusion, our paper describes a stillborn patient delivered at 22 weeks of gestation, presenting with pl-HPP. Mutations detected in our case, although previously demonstrated in other patients, have not been previously reported to co-occur in a single individual. Therefore, our paper may represent a foundation for the prognosis and genetic counselling of other individuals carrying identical mutations. Since the parents were declined access to an X-ray, the diagnosis in our index had to be established by means of NGS-based DAG panel sequencing, although it was apparently achievable with standard radiography. Thus, in our report we strongly emphasize the importance of routine X-ray examination, which should still be a first tier study in similar cases. Finally, we also highlight the emerging role of NGS strategies in the diagnostics of prenatally manifesting skeletal disorders, which should however be reserved only for cases in which clinical data are too sparse to allow for accurate clinical diagnosis and targeted gene analysis.

Acknowledgements

We are grateful to the patients for participating in this study.

The authors declare no conflicts of interest.
Accession numbers: NM_000478, NP_000469.
The research was performed at the Department of Medical Genetics, Poznan University of Medical Sciences, 8 Rokietnicka Street, 60-806 Poznan, Poland and Institute for Medical Genetics and Human Genetics, Charité Universitätsmedizin Berlin, 13353 Berlin, Germany.

Funding: This work was supported by a grant from the National Centre for Research and Development (LIDER/008/431/L-4/12/NCBR/2013) to AJ.

References

1. Krakow D, Lachman RS, Rimoin DL. Guidelines for the prenatal diagnosis of fetal skeletal dysplasias. Genet Med 2009; 11: 127-233.
2. Mornet E, Stura E, Lia-Baldini AS, et al. Structural evidence for a functional role of human tissue nonspecific alkaline phosphatase in bone mineralization. J Biol Chem 2001; 276: 31171-31178.
3. Warman ML, Cormier-Daire V, Hall C, et al. Nosology and classification of genetic skeletal disorders: 2010 revision. Am J Med Genet A 2011; 155A: 943-968.
4. Weiss MJ, Henthorn PS, Lafferty MA, et al. Isolation and characterization of a cDNA encoding a human liver/bone/kidney-type alkaline phosphatase. Proc Natl Acad Sci U S A 1986; 83: 7182-7186.
5. Greenberg CR, Evans JA, McKendry-Smith S, et al. Infantile hypophosphatasia: localization within chromosome region 1p36.1-34 and prenatal diagnosis using linked DNA markers. Am J Hum Genet 1990; 46: 286-292.
6. Whyte MP. Hypophosphatasia and the role of alkaline phosphatase in skeletal mineralization. Endocr Rev 1994; 15: 439-461.
7. Narisawa S, Frohlander N, Millan JL. Inactivation of two mouse alkaline phosphatase genes and establishment of a model of infantile hypophosphatasia. Dev Dyn 1997; 208: 432-446.
8. Whyte MP, Landt M, Ryan LM, et al. Alkaline phosphatase: placental and tissue-nonspecific isoenzymes hydrolyze phosphoethanolamine, inorganic pyrophosphate, and pyridoxal 5’-phosphate. Substrate accumulation in carriers of hypophosphatasia corrects during pregnancy. J Clin Invest 1995; 95: 1440-1445.
9. Moore CA, Curry CJ, Henthorn PS, et al. Mild autosomal dominant hypophosphatasia: in utero presentation in two families. Am J Med Genet 1999; 86: 410-415.
10. Mornet E. Hypophosphatasia. Orphanet J Rare Dis 2007; 2: 40.
11. Mornet E, Hofmann C, Bloch-Zupan A, et al. Clinical utility gene card for: hypophosphatasia – update 2013. Eur J Hum Genet 2014; 22. doi: 10.1038/ejhg.2013.177.
12. Brun-Heath I, Lia-Baldini AS, Maillard S, et al. Delayed transport of tissue-nonspecific alkaline phosphatase with missense mutations causing hypophosphatasia. Eur J Med Genet 2007; 50: 367-378.
13. Whyte MP, Leelawattana R, Reinus WR, et al. Acute severe hypercalcemia after traumatic fractures and immobilization in hypophosphatasia complicated by chronic renal failure. J Clin Endocrinol Metab 2013; 98: 4606-4612.
14. Shibata H, Fukushi M, Igarashi A, et al. Defective intracellular transport of tissue-nonspecific alkaline phosphatase with an Ala162Thr mutation associated with lethal hypophosphatasia. J Biochem 1998; 123: 968-977.
15. Di Mauro S, Manes T, Hessle L, et al. Kinetic characterization of hypophosphatasia mutations with physiological substrates. J Bone Miner Res 2002; 17: 1383-1391.
16. Rockman-Greenberg C. Hypophosphatasia. Pediatr Endocrinol Rev 2013; 10 Suppl 2: 380-388.
17. Zemojtel T, Köhler S, Mackenroth L, et al. Effective diagnosis of genetic disease by computational phenotype analysis of the disease-associated genome. Sci Transl Med 2014; 6: 252ra123.
18. Robinson PN, Kohler S, Bauer S, et al. The Human Phenotype Ontology: a tool for annotating and analyzing human hereditary disease. Am J Hum Genet 2008; 83: 610-615.
19. Gafnney G, Manning N, Boyd PA, et al. Prenatal sonographic diagnosis of skeletal dysplasias–a report of the diagnostic and prognostic accuracy in 35 cases. Prenat Diagn 1998; 18: 357-362.
20. Doray B, Favre R, Viville B, et al. Prenatal sonographic diagnosis of skeletal dysplasias. A report of 47 cases. Ann Genet 2000; 43: 163-169.
21. Nakamura-Utsunomiya A, Okada S, Hara K, et al. Clinical characteristics of perinatal lethal hypophosphatasia: a report of 6 cases. Clin Pediatr Endocrinol 2010; 19: 7-13.
22. Shohat M, Rimoin DL, Gruber HE, et al. Perinatal lethal hypophosphatasia; clinical, radiologic and morphologic findings. Pediatr Radiol 1991; 21: 421-427.
23. Mornet E, Stura E, Lia-Baldini AS, et al. Structural evidence for a functional role of human tissue nonspecific alkaline phosphatase in bone mineralization. J Biol Chem 2001; 276: 31171-31178.
24. Nelson DB, Dashe JS, Donald D, et al. Fetal skeletal dyspalsias Sonographic idices associated with adverse outcomes. J Ultrasound Med 2014; 33: 1085-1090.
25. Draguet C, Gillerot Y, Mornet E. Childhood hypophosphatasia: a case report due to a novel mutation. Arch Pediatr 2004; 11: 440-443.
26. Spentchian M, Brun-Heath I, Taillandier A, et al. Characterization of missense mutations and large deletions in the ALPL gene by sequencing and quantitative multiplex PCR of short fragments. Genet Test 2006; 10: 252-257.
27. Wenkert D, McAlister WH, Coburn SP, et al. Hypophosphatasia: Nonlethal disease despite skeletal presentation in utero (17 new cases and literature review). J Bone Miner Res 2011; 26: 2389-2398.
28. Hoylaerts MF, Manes T, Millán JL. Mammalian alkaline phosphatases are allosteric enzymes. J Biol Chem 1997; 272: 22781-22787.
29. Hoylaerts MF, Ding L, Narisawa S, et al. Mammalian alkaline phosphatase catalysis requires active site structure stabilization via the N-terminal amino acid microenvironment. Biochem 2006; 45: 9756-9766.

Address for correspondence

Aleksander Jamsheer, MD, PhD, MSc
Department of Medical Genetics
Poznan University of Medical Sciences
8 Rokietnicka 8
60-806 Poznan, Poland
tel. +48 61 854 76 18
fax +48 61 851 76 13
e-mail: jamsheer@wp.pl