| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Division of Medical Genetics (D.A.S., J.C.C., J.L.B.B.), Department of Pediatrics, and Division of Maternal-Fetal Medicine (J.L.B.B.), Department of Obstetrics and Gynecology, University of Utah, Salt Lake City, Utah 84132; Shriners Hospital for Children Intermountain (D.A.S., J.C.C.), Salt Lake City, Utah 84103; Department of Chemistry (S.P.C., K.L.E.), Indiana University-Purdue University, Fort Wayne, Indiana 46805-1499; Division of Bone and Mineral Diseases (S.M., M.P.W.), Washington University School of Medicine, St. Louis, Missouri 63110; and Center for Metabolic Bone Disease and Molecular Research (S.M., M.P.W.), Shriners Hospitals for Children, St. Louis, Missouri 63131
Address all correspondence to: David A. Stevenson, M.D., Department of Pediatrics, Division of Medical Genetics, 2C412 SOM, University of Utah Health Sciences Center, Salt Lake City, Utah 84132. E-mail: david.stevenson{at}hsc.utah.edu.
| Abstract |
|---|
|
|
|---|
Objective: Our objective was to identify TNSALP mutations and characterize the inheritance pattern of a family with clinically variable HPP with one child manifesting in utero with long bone deformity but showing spontaneous prenatal and postnatal improvement.
Design: TNSALP enzyme and substrate analysis and TNSALP mutation analysis were performed on all family members.
Patients: A boy with HPP showing long bone deformity that spontaneously improved in utero and after birth is described. His older brother has the childhood form of HPP without findings until after infancy. His parents and twin sister are clinically unaffected.
Results: Both boys are compound heterozygotes for the same missense mutations in TNSALP, documenting autosomal recessive inheritance for their HPP. The parents each carry one defective allele.
Conclusions: The patient is an autosomal recessive case of HPP with prenatal long bone deformity but with spontaneous prenatal and postnatal improvement. Thus, prenatal detection by sonography of bowing of long bones from HPP, even with autosomal recessive inheritance, does not necessarily predict lethality but can represent variable expressivity or the effects of modifiers on the TNSALP defect(s).
| Introduction |
|---|
|
|
|---|
Genotype-phenotype correlations have been described for HPP (7, 8), yet management and counseling for HPP remain based upon the clinical nosology. The perinatal form of HPP, which can be detected in utero by sonography, has been considered nearly always lethal (4, 12, 13). However, there have been reports of HPP detected prenatally by sonography due to long bone deformity but demonstrating considerable, spontaneous, postnatal improvement (14, 15, 16, 17, 18, 19). Interestingly, several of these families manifested autosomal dominant transmission patterns for HPP [the genetic basis was reported for only two of the probands (15)], whereas others were compound heterozygous Japanese patients with F310L mutations in exon 9 of the TNSALP gene (17, 18, 19). Pauli et al. (14) proposed that this "bent, but not broken," phenotype in utero be considered a sixth form of HPP.
This report describes a boy with prenatal presentation of HPP manifesting as long bone deformity detected by sonography who had spontaneous in utero as well as postnatal improvement, yet due to autosomal recessive inheritance of two TNSALP missense mutations, neither of which was the F310L mutation (now called p.Phe327Leu).
| Patients and Methods |
|---|
|
|
|---|
Sib no. 1 presented in utero at 18-wk gestation with sharp angulation of the midshaft of the right humerus, radius, and ulna detected by sonography performed for twin gestation (Fig. 1
). Dichorionic, diamniotic twin gestation was previously documented at 14 wk by sonography performed for the establishment of gestational age in a mother taking clomiphene. At 18 wk, all major long bones were subjectively of slightly narrow diameter and poorly ossified. The right humerus, radius, and ulna were sharply angulated and 2.5–4 SD values below the mean for normal length, and the right arm was held in a fixed, bent position. The long bones on the left side otherwise appeared normal. His femurs were 1.5 SD values below the mean for normal length. At 22 wk, sonography showed improved growth of the right humerus, radius, ulna, and femurs, although the angulations in the upper and lower segments of the right arm persisted, and the right femur appeared curved. At 26-wk gestation, there was a mildly shortened and angulated right radius (Fig. 1C
), ulna, and humerus, with note of progressive improvement of the angulations. Movement of the right shoulder was observed.
|
Sib no. 1 was a dizygotic twin, and his twin sister (sib no. 2) showed no sonographic abnormalities throughout the pregnancy and was born without apparent skeletal deformities. Furthermore, she demonstrated no clinical or biochemical evidence of HPP. Her serum alkaline phosphatase (ALP) activity soon after birth was 145 IU/liter (normal 110–300). However, lethality for sib no. 1 was a concern given his prenatal findings and a diagnosis of childhood HPP for his older brother.
At birth, sib no. 1s APGAR scores were nine at 1 min and nine at 5 min. There was bowing of his right arm with dimpling of his skin near the apex of the bowing, but his physical examination was otherwise unremarkable. Length and head circumference were both at the 50th centile, but his weight was at the 10th centile. Radiographs showed bowing of his right humerus, radius, and ulna with callus formation suggestive of healing fractures (Fig. 2
A, B). Metaphyses were osteopenic and irregular, but the growth plates were unremarkable. There was also mild bowing and irregularity of his left femur, and mild bowing of his left tibia (Fig. 2A
). Radiographs of his skull were reported as normal. Serum ALP activity soon after birth was less than 20 IU/liter, and he was diagnosed with HPP.
|
At age 17 months, he was evaluated for premature loss of two deciduous teeth starting at age 16 months (Fig. 3B
). He had mild bowing of his right arm where the previous in utero humerus bowing was noted, but bowing and range of motion had improved spontaneously since the perinatal evaluation (Fig. 3A
). Extensive skeletal radiographs were not taken because his postnatal clinical deformities were minimal. Other clinical findings included short stature (<3rd centile) with relative macrocephaly (75th centile) and genu valgum. His weight was at the 10th centile.
|
Family history
His 4-yr-old brother has HPP (sib no. 3). His fetal sonography at 20-wk gestation was reported as normal, but the images were not available for review. He was first evaluated at age 19 months for the premature loss of three teeth starting at age 18 months. He was reevaluated at 2.5 yr with mild delays in gross motor skills. No bowing of his extremities was noted. He too had short stature (5th centile) with macrocephaly (98th centile) and a scaphocephalic skull with a high forehead (Fig. 4
). Radiographs, first taken at age 18 months, showed flared ribs and radiolucencies of his humeri that projected from the growth plates into the metaphyses characteristic of HPP (4).
|
TNSALP mutation analysis
Patient, parent, and sib blood leukocyte DNA was obtained together with blood and urine for routine biochemical studies after informed written consent approved by the Human Studies Committee, Washington University School of Medicine, St. Louis, MO.
Genomic DNA was extracted using the PUREGENE DNA Purification Kit (Gentra Systems, Minneapolis, MN). For both affected brothers, all coding exons (nos. 2–12) and adjacent mRNA splice junctions of TNSALP were amplified by PCR and sequenced in both directions using previously described methods and primers (20). The DNA sequence was examined visually and using VectorNTI AlignX software (Invitrogen Corp., Carlsbad, CA). Only TNSALP exons 6, 8, and 12 were sequenced for both parents and the apparently healthy twin sister.
TNSALP activity and substrate analyses
Serum total and bone-specific ALP (BALP) activities were measured at the Center for Metabolic Bone Disease and Molecular Research, Shriners Hospital for Children, and compared with age-matched control values established at that facility (21). Plasma PLP levels were quantitated using published HPLC techniques (22, 23) in the Department of Chemistry, Indiana-Purdue University, and compared with values from the same control specimens (21).
| Results |
|---|
|
|
|---|
A summary of the familys biochemical findings is found in Table 1
.
|
Sib no. 1s older affected brother (sib no. 3) had a serum total ALP activity assayed at the University of Utah that was low at 33 IU/liter at 17 months of age. At 2.5 yr of age, his serum total ALP and BALP activity (measured at the Shriners Hospital for Children) was low at 36 and 30 IU/liter, respectively. At 6 yr of age, his serum ALP and BALP activity was 63 and 36 IU/liter, respectively. His plasma PLP concentration was markedly elevated at 1332 nmol/liter at 2.5 yr of age and 4034 nmol/liter at 6 yr of age (Table 1
).
Sib no. 1s unaffected twin sister (sib no. 2) had normal serum total ALP activity soon after birth (145 IU/liter). At 4 yr of age, serum total ALP activity of 308 IU/liter and BALP of 144 IU/liter were normal, and plasma PLP concentration of 117 nmol/liter was minimally elevated (Table 1
).
The mothers serum total ALP and BALP activity were normal (32 and 11 IU/liter, respectively). The fathers serum total ALP and BALP activity were normal (55 and 21 IU/liter, respectively). However, both parents had elevated plasma PLP levels (mother = 258 nmol/liter; father = 191 nmol/liter) (Table 1
) (4).
TNSALP gene analysis
All coding exons and adjacent mRNA splice sites of TNSALP were sequenced for both affected brothers. Both boys were compound heterozygotes for loss-of-function TNSALP defects, revealing missense mutations in both exons 6 and 8 [exon 6 (c.526G>A, p.Ala176Thr); exon 8 (c.814C>T, p.Arg272Cys)]. Therefore, autosomal recessive inheritance accounted for their HPP. The mother carried the exon 8 mutation (c.814C>T, p.Arg272Cys), which is rare; it has not been detected in a cohort of approximately 150 HPP probands (unpublished data), but it has been reported with another TNSALP mutation once in an individual described as having perinatal HPP (24). The father carried the exon 6 mutation (c.526G>A, p.Ala176Thr), which is very common in a HPP patient cohort (unpublished data), and has been reported elsewhere (25).
The apparently healthy twin sister had neither TNSALP mutation.
| Discussion |
|---|
|
|
|---|
The biochemical markers of HPP quantitated in sib no. 1s family members generally reflected what was revealed by TNSALP mutation analysis. The clinically unaffected twin (sib no. 2) had normal serum total ALP and BALP activities and minimally elevated plasma PLP levels and harbored no TNSALP mutation. The mothers serum total ALP activity of 32 IU/liter is at the bottom of the normal range, and lies between values that have been encountered in adult controls and those with adult HPP. The fathers serum ALP is within the range of adult controls (4). Both parents had elevated plasma PLP levels. The mother had taken just one prenatal vitamin tablet 2 wk before the PLP measurement. Therefore, this was unlikely to have increased her plasma PLP level. These biochemical findings are consistent with both parents carrying a TNSALP mutation (Table 1
). Sib no. 1s older affected brothers serum total ALP activity was also low at 33 IU/liter at 17 months of age, and was higher at 63 IU/liter at 6 yr of age, and now within the range encountered in childhood HPP (4). At 6 yr of age, the older brother also had a distinctly elevated plasma PLP concentration (4034 nM), in keeping with levels observed in perinatal and infantile HPP. However, differences in the clinical and biochemical findings between sib nos. 1 and 3 occurred despite the fact that both were compound heterozygotes for the same two missense mutations within TNSALP.
Interfamilial variability of HPP is wide ranging, and largely explained by the patterns of inheritance and the variety of causal TNSALP mutations (4). Yet, most publications describe little intrafamilial variability for HPP and indicate that affected sibs are phenotypically similar. Occasionally, unexplained variations have been noted among affected sibs (26, 27, 28, 29). There are three published reports of sibs with HPP in which one died perinatally, and the other had HPP with later onset (27, 28, 29). Of interest, for two of these reports, the more severely affected case was one of twins, with the other twin being unaffected (27, 28). The third such case presented with oligohydramnios (29). Although in the family reported herein the brother diagnosed prenatally (sib no. 1) did not have a poor postnatal outcome, he seemed more severely affected than his older brother (sib no. 3), and was also a twin to an unaffected sister (sib no. 2). Perhaps twin gestation can modify HPP expression in utero, but how an unaffected twin could cause more severe HPP during gestation is an enigma. It may be that fetal crowding, either from twin gestation or in other cases from oligohydramnios, could be an explanation. However, upon serial sonography a force acting on the upper-right side of sib no. 1, where the majority of his deformities were located, could not readily be identified. In addition, oligohydramnios was not present.
Sib no. 1s clinical course resembled what has been detailed in a few cases previously with skeletal manifestations of HPP detected in utero by sonography yet with spontaneous postnatal improvement (14, 15, 16, 17, 18, 19). Several of these cases inherited HPP due to an autosomal dominant transmission pattern. The genetic etiology explored by TNSALP mutation analysis in the cases with autosomal dominant transmission patterns has been reported for two cases: the child described from the second family of Moore et al. (15) and the child reported by Comstock et al. (16). The child reported by Moore et al. (15) had a c.1133A>T, p.Asp378Val mutation, and the child reported by Comstock et al. (16) had a c.1144G>A, p.Val382Ile sequence variant. The autosomal dominant transmission patterns suggest possible genetic heterogeneity. However, as described before, two of the four previously reported families were found to have heterozygous mutations in the TNSALP gene (15, 16), suggesting that a heterozygous mutation in the TNSALP gene alone or in combination with other genetic or environmental modifiers could result in the described phenotype. In addition, several Japanese cases, which were compound heterozygotes, share an identical c.979T>C, p.Phe327Leu mutation in the TNSALP gene (17, 18, 19). Nevertheless, the two brothers with HPP (sib nos. 1 and 3) inherited HPP as an autosomal recessive trait, as shown by TNSALP mutation analysis, but without the p.Phe327Leu mutation. The fact that both parents appeared clinically unaffected by HPP gives further evidence to an autosomal recessive trait for in utero long bone deformities with spontaneous postnatal improvement in HPP. Notably, sib no. 1 displayed prenatal improvement.
Sib no. 1 represents a detailed report of an autosomal recessive case of this relatively benign clinical form of HPP. He carries two mutations of TNSALP (without a p.Phe327Leu mutation) manifesting in utero with deformity of long bones, yet with spontaneous prenatal and postnatal improvement. Thus, accumulating experience shows that prenatal or perinatal skeletal manifestations of HPP, even with autosomal recessive inheritance, are not uniformly lethal and can represent the variable expressivity or the effects of modifiers on the underlying TNSALP allelic defect(s).
| Acknowledgments |
|---|
| Footnotes |
|---|
This work was supported in part by Grant no. M01-RR00064 from the National Center for Research Resources, research Grant no. K23 NS052500 from the National Institute of Neurological Disorders and Stroke, the Childrens Health Research Center and Clinical Genetics Research Program at the University of Utah, Shriners Hospitals for Children, The Clark and Mildred Cox Inherited Metabolic Bone Disease Research Fund, The Hypophosphatasia Research Fund, and The Barnes-Jewish Hospital Foundation.
Presented in part at The 2006 David W. Smith Workshop on Malformations and Morphogenesis (Stevenson et al. Proc Greenwood Genet Center 2007;26:153–154 abstract), and the 30th Annual Meeting of the American Society for Bone and Mineral Research, Montreal, Canada, September 12–16, 2008 (J Bone Miner Res, in press).
Disclosure Statement: Dr. Whyte is a consultant for Enobia Pharma, Montreal, Canada.
First Published Online June 17, 2008
Abbreviations: ALP, Alkaline phosphatase; BALP, bone-specific alkaline phosphatase; HPP, hypophosphatasia; PLP, pyridoxal-5'-phosphate; TNSALP, tissue nonspecific alkaline phosphatase.
Received February 11, 2008.
Accepted June 9, 2008.
| References |
|---|
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |