Isolated dihydroxyacetonephosphate acyltransferase deficiency presenting with developmental delay

J. Inher. Metab. Dis. 17 (1994) 533-540 © SSIEM and KluwerAcademicPublishers. Printed in the Netherlands

Isolated Dihydroxyacetonephosphate Acyltransferase Deficiency Presenting with Developmental Delay P. T. CLAYTON1'2., S. ECKHARDT1. J. WILSON 1, C. M. HALL 1. Y. YOUSUF 1, R. J. A. WANDERS3 a n d R. B. H. SCHUTGENS3 1Hospital for Sick Children, Great Ormond Street, London; 2Metabolic Disease Unit, Institute of Child Health, London, UK," 3Department of Pediatrics and Clinical Biochemistry, University of Amsterdam, Amsterdam, The Netherlands

Summary: A boy aged 21 months who was being investigated for developmental delay and failure to thrive was found to have punctate epiphyseal calcification, He had no evidence of rhizomelic shortening of the limbs or cataracts. Investigation revealed defective plasmalogen synthesis due to isolated deficiency of dihydroxyacetonephosphate acyltransferase (DHAP-AT). The parents were consanguineous and a sister was similarly affected, suggesting autosomal recessive inheritance. Hitherto, recessively inherited isolated DHAP-AT deficiency has only been described in patients with a phenotype similar to that of rhizomelic chondrodysplasia punctata. This report indicates that the same biochemical disorder can be associated with a less severe phenotype.

Dihydroxyacetonephosphate acyltransferase (DHAP-AT) is a peroxisomal enzyme that catalyses the first step in the synthesis of ether phospholipids including plasmalogens. A deficiency of this enzyme can occur as a result of a Group A, a Group B or a Group C peroxisomal disorder (Wanders et al 1988; Lazarow and Moser 1989). In patients with a Group A peroxisomal disorder (e.g. Zellweger syndrome), DHAP-AT activity is deficient because the peroxisomes are absent or greatly reduced in number. The DHAP-AT deficiency is associated with deficiency of many other peroxisomal enzymes (e.g. those involved in/?-oxidation of very longchain fatty acids). Rhizomelic chondrodysplasia punctata (RCDP) is an example of a Group B disorder. Peroxisomes are present, at least in fibroblasts, but they are deficient in more than one enzyme. Thus DHAP-AT deficiency occurs in association with reduced phytanic acid c~-oxidation and, although oxidation of very long-chain fatty acids is normal, peroxisomal thiolase is present in its precursor form (Hoefler et al 1988). Finally, isolated deficiency of DHAP-AT has recently been described *Correspondence: Metabolic Disease Unit, Division of Biochemistry and Genetics, Institute of Child Health, 30 Guilford Street, London WC1N 1EH, UK MS received 4.2.94 Accepted 18.4.94 533

534

C l a y t o n et al.

(Wanders et al 1992; Barr et al 1993). Because this defect involves reduced activity of only one peroxisomal enzyme, it is an example of a Group C disorder. The two patients with isolated DHAP-AT deficiency who have been described had a phenotype essentially identical with classical RCDP. Dysmorphic features (craniofacial abnormalities, cataracts and pronounced rhizomelic shortening, particularly of the upper .limbs) were apparent at birth and the infants were profoundly hypotonic. In this report we describe a boy with isolated D H A P - A T deficiency who had only subtle dysmorphic features and no rhizomelic limb shortening and who presented at 5 months of age with developmental delay. The lack of major dysmorphic features and rhizomelic limb shortening seen in this case was also seen in the case of mild R C D P described by Poll-The et al (1991). CASE REPORT The patient was born at term to consanguineous parents after a normal pregnancy and delivery. His birth weight (3.46 kg), length (56 cm) and head circumference (34 cm) were all between the 25th and 50th centiles. He remained well until the age of 5 months, when he developed cellulitis of his left knee. Examination revealed that he was developmentally delayed, although his initial development has seemingly been unremarkable; he had smiled at 4 weeks of age and at 5 months he had learnt to roll over and was grasping objects with a palmar grasp. Vision and hearing was considered to be normal. In the following few months he developed recurrent respiratory infections and eczema. During hospitalization a CT-scan of his head was performed and showed mild generalized cerebral atrophy. He continued to develop rather poorly 'and failed to gain weight. He was fully assessed for the first time at the age of 21 months. His weight on admission (5.85 kg), his length (61.7 cm) and his head circumference (41.7 cm) were all well below the 3rd centiles. He had anteverted nostrils but no other dysmorphic features, and he showed moderate generalized muscle wasting. Vision and hearing were clinically normal, and there was a full range of normal eye movements. There were n o cataracts and his fundi looked healthy. He was able to babble. Upper and lower limb tone was reduced, but power was normal. There was a full range of normal movements at all joints and no deformity was present. A palmar grasp was present but no ability to transfer objects between hands. He was able to sit supported with a round back and good head control. Tendon reflexes were generally exaggerated but there were no pathological reflexes or clonus present. Landau and downward parachute reflexes were absent. He had gross gastro-oesophageal reflux for which he required a Nissen fundoplication during the admission. A skeletal survey was performed and showed the following abnormalities. There was irregularity of the vertebral end plates of the entire spine, with abnormal ossification in many of the vertebral end plates. There was a mild scoliosis, but the vertebral body height was well preserved. There was punctate calcification in the proximal humeral epiphyses (Figure 1). There was a stippled appearance of both femoral capital epiphyses and both knee epiphyses, which were also rather flattened and of irregular appearance. Around the knee joints there was soft-tissue punctate calcification. The distal femoral epiphyses also showed some irregularity in shape J. lnher. Metab. Dis. 17 (1994)

Developmental Delay Presentation of DHAP-ATdefect

535

Figure 1 Stippled calcification in the proximal humeral epiphysis (Figure 2). Generalized osteopenia and soft-tissue wasting was noted. The appearances were consistent with chondrodysplasia calcificans punctata. At this time the 3.5-year-old sister of the patient was also assessed. This child had been born at full term but was small for dates, with height (49cm), weight (2.66 kg) and head circumference (32 cm) below the third centiles. She showed severe developmental delay and failure to thrive from early life. She looked remarkably similar to her younger brother and unlike either parent. There was no speech apart from cooing, but she had smiled for the first time at 4 months. Vision and hearing were clinically normal and there was a full range of normal eye movements. There was no interest in reaching or grasping toys. She had suffered from two convulsive episodes at 16 months and 21 months but was now off any medication. She was unable to sit unsupported and was only able to roll over from supine to prone position. She showed evidence of bulbar palsy and there was marked truncal hypotonia but slight limb hypertonia. She had flexion contractures at both knees with equinus deformity of both feet. Her skeletal survey revealed the following abnormal features. There was irregularity of the vertebral end plates with the impression of residual coronal clefts. The proximal humeral epiphyses were of irregular appearance. Both femoral capital epiphyses (the left more than the right) were asymetrically flattened and also irregular in shape. These appearances were also consistent with chondrodysplasia calcifans punctata.

J. lnher. Metab. Dis. 17 (1994)

536

C l a y t o n et al.

Figure 2 Femoral capital epiphyses and knee epiphyses with stippled calcification

METHODS

Plasma concentrations of very long-chain fatty acids (VLCFA) and phytanic acid were measured by capillary gas-chromatography mass spectrometry (GC-MS) using deuterated internal standards. Fibroblasts from a skin biopsy were cultured using standard procedures. Established methods were used to measure the following in the fibroblasts: acyl-CoA dihydroxyacetonephosphate acyltransferase (DHAP-AT) (Schutgens et al 1984), plasmalogen levels (Bj6rkhem et al 1986), de novo plasmalogen biosynthesis (Schrakamp et al 1988) and the very long-chain fatty acid profile (Schutgens et al 1993). Immunoblot analysis of peroxisomal fl-oxidation enzymes was performed as described by Wanders et al (1991). RESULTS

Routine haematology and biochemistry were unremarkable, except a slightly elevated AST of 84U/L (normal < 50). Chrornosomal analysis was normal. The plasma VLCFA profile and phytanate were normal (see Table 1). The fibroblast VLCFA profile was also normal. Fibroblast fatty acid oxidation studies showed a normal J. Inher. Metab. Dis. 17 (1994)

537

Developmental Delay Presentation o f D H A P - A T defect

Table 1 Plasma VLCFA and phytanic acid concentration and fibroblast VLCFA in our patient compared to normal controls Patient

Normal range

Plasma VLCFA

C26:o (%mg/ml)

0.37 0.93 0.03

C24/C22 C26/C22 Plasma phytanic acid (pmol/L) Fibroblast VLCFA C26:o (#g/rag)

11.18

< 15

0.05 1.14 0.02

C24/C22 C26/C22

0.12 0.49 0.65-1.01 0.007-0.34

0.02-0.1 1.17-1.83 0.02 0.05

activity of phytanic acid oxidase of 132pmol/24h per mg protein (normal range of 85-154). Western blots performed using anti-peroxisomal thiolase and anti-acyl-CoA oxidase sera and a protein extract from the patient's cells were normal. The thiolase was present as the mature 41 kDa protein and not in the 4 4 k D a precursor form as found in classical RCDP. Analysis of the plasmalogens in the patient's fibroblasts showed a deficiency of both C16:o and C18:o plasmalogens (see Table 2). De novo plasmalogen biosynthesis in the fibroblasts was also found to be subnormal in comparison with the controls (Table 2). The activity of D H A P - A T in the patient's fibroblasts was considerably lower than in standardized controls (Table 2). All of these results were consistent with a deficiency of D H A P - A T in the patient's cells.

Table 2 Indices of plasmalogen synthesis in fibroblasts from our patient, from controls and from patients with classical rhizomelie chondrodysplasia punetata Control Patient (mean (5-95% range))

Plasmalogen level (% of control value) C16:o plm C18:o plm

59 25

100 100

RCDP

(Mean (range)) 27.2 (20-39) 5.7 (0-12)

De novo plasmalogen synthesis

%pPE in PE

44.6

%pPC in PC

1.0

DHAP-AT activity (nmol/2 h per mg protein)

0.1

89.2 (83-92)" (n = 59) 7.0 (3.3-13.6) (n = 59) 7.8 (3.8-11.8) (n = 59)

3.0 (0.9-10.3)" (n = 13) 0.4 (0.3-0.7) (n = 13) 1.2 (0.8-2.4)

aData from Schutgens et al (1992) plm: plasmalogen; pPE: plasmalogen phosphatidyl ethanolamine; PE, phosphatidyl ethanolamine; pPC: plasmalogen phosphatidyl choline; PC: phosphatidyl choline J. lnher. Metab. Dis. 17 (1994)

538

Clayton et al.

DISCUSSION The major presenting feature in our patient was developmental delay associated with hypotonia. He, as well as his sister, showed failure to thrive, with short stature and microcephaly and subtle dysmorphic features (anteverted nares). They were not rhizomelic, and only the girl had joint contractures, but both had marked punctate calcification of cartilages. The latter provided the important diagnostic clue: there is a very striking association between defective plasmalogen synthesis and punctate epiphyseal calcification. Our investigations confirmed that plasmalogen synthesis was impaired due to isolated deficiency of DHAP-AT. The normal very long-chain fatty acid metabolism ruled out a Group A peroxisomal disorder and RCDP (Group B) was excluded by normal values for plasma phytanic acid, fibroblast phytanic acid oxidation and fibroblast peroxisomal thiolase immunoblotting. Wanders et al (1992) and Barret al (1993) have described isolated deficiency of DHAP-AT causing a phenotype indistinguishable from rhizomelic chondrodysplasia punctata. Their patients had pronounced rhizomelic shortening, obvious dysmorphic features, cataracts and extreme hypotonia, and the patient described by Wanders et al (1992) died by the age of 6 months. The parents of this infant were normal but consanguineous, suggesting autosomal recessive inheritance. Our patient also had a pedigree suggesting autosomal recessive inheritance and he had three features in common with the phenotype of RCDP (microcephaly, developmental delay and the typical pattern of punctate calcification affecting the long bones symmetrically), but he lacked the classical rhizomelic shortening of the limbs. I t seems likely that our patient has a mutation of the same gene as the patient of Wanders et al (1992); complementation studies could be used to confirm this. (Complementation studies have already been used to show that the genetic defect in the patient of Wanders et al (1992) is different from that in classical RCDP (Heikoop et al 1992).) Our patient had just detectable DHAP-AT activity (0.1 nmol/2 h per mg protein); the patient of Wanders et al (1992) had undetectable activity - - this may explain why our patient was less severely affected. It is also clear from Table 2 that overall plasmalogen synthesis in our patient was less disrupted than is the case in most patients with RCDP; this again may explain why our patient had a less severe disease than patients with classical RCDP. Reduced activity of DHAP-AT has been described in association with two dominantly inherited disorders that can produce punctate calcification of cartilages. Clayton et al (1989) found reduced activity of DHAP-AT in the fibroblasts of a girl with X-linked dominant chondrodysplasia punctata (McKusick 302960). However, following further culture of the cells the enzyme activity became normal. The reason for this is not clear. It did not occur in the patient described in this report and, of course, X-linked dominant chondrodysplasia punctata can be excluded from the pedigree of our patient and his clinical features. Deficiency of DHAP-AT was also reported by Holmes et al (1987) in a mother and daughter whom they described as having the Conradi-Hiinermann (CH) syndrome (autosomal dominant; McKusick 118650) but who, from the description, may have had X-linked dominant chondrodysplasia punctata. Schutgens et al (1988) have been unable to confirm the finding of reduced activity of DHAP-AT in the CH syndrome. Our patient had several features J. lnher. Metab. Dis. 17 (1994)

Developmental Delay Presentation o f D H A P - A T d e f e c t

539

(apart from the pedigree) that made the diagnosis of CH syndrome untenable, microcephaly is rarely seen in children with the CH syndrome and CH patients usually have asymmetrical bone changes on radiography. On the basis of the available evidence it seems probable that the patient described by Wanders et al (1992) and the siblings described in this paper have a mutation in the D H A P - A T gene and that this is inherited in an autosomal recessive fashion. Other disorders producing stippled calcification of cartilages may do so by indirectly affecting D H A P - A T activity, but reduced activity of DHAP-AT cannot consistently be demonstrated in cultured skin fibroblasts. The mild form of isolated D H A P - A T deficiency described in this paper shares some clinical features with a patient with a mild form of R C D P described by PollThe et al (1991). In common with our patient, their case had no rhizomelic shortening or major dysmorphie features. On the other hand, unlike our patient, their patient had evidence of restricted movements and pain from birth that highlighted the skeletal problem and led to the detection of the underlying biochemical abnormalities. The major presenting features in our patient with D H A P - A T deficiency were developmental delay, microcephaly and failure to thrive. Many centres use a plasma very long-chain fatty acid assay to screen for peroxisomal disorders when an infant presents with developmental delay and hypotonia. This case report shows that such an approach will fail to detect at least one peroxisomal disorder that can produce these features. The case also shows once again that punctate epiphyseal calcification is an important pointer to deranged plasmalogen synthesis (Clayton and Thompson 1988). Wanders et al (1994) have recently shown that as well as occurring in isolated D H A P - A T deficiency (a defect in the first step in plasmalogen synthesis, Schrakamp et al 1988), punctate epiphyseal calcification can occur in isolated deficiency of the enzyme alkyldihydroxyacetonephosphate synthease (which catalyses the second step in plasmalogen synthesis).

REFERENCES

Barr DGD, Kirk JM, A1 Howasi M, Wanders RJA, Schutgens RBH (1993) Rhizomelic chondrodysplasia punctata with isolated DHAP-AT deficiency. Arch Dis Child 68: 415-417. Bj6rkhem I, Sisfontes L, Bostr6m B, Kase BF, Blomstrand R (1986) Simple diagnosis of the Zellweger syndrome by gas liquid chromatography of dimethylacetals. J Lipid Res 27: 786-791. Clayton PT, Thompson M (1988) Dysmorphic syndromes with demonstrable biochemical abnormalities. J Med Genet 25: 463-472. Clayton PT, Kalter C, Atheron D J, Besley GTN, Broadhead DM (1989) Peroxisomal enzyme deficiency in X-linked dominant Conradi Hiinermann syndrome. J Inher Metab Dis 12 (supplement 2): 358-360. Heikoop JC, Wanders RJ, Strijland A, Purvis R, Schutgens RBH, Tager JM (1992) Genetic and biochemical heterogeneity in patients with the rhizomelic form of chondrodysplasia punctata. Hum Genet 89: 439-444. Hoefler G, Hoefler S, Watkins PA et al (1988) Biochemical abnormalities in rhizomelic chondrodysplasia punctata. J Pediatr 112: 726-733. Holmes RD, Wilson GN, Hajra AK (1987) Peroxisomal enzyme deficiency in the ConradiH/inermann form of chondrodysplasia punctata. N Engl J Med 316: 1608. J. lnher. Metab. Dis. 17 (1994)

540

Clayton et al.

Lazarow PB, Moser HW (1989) Disorders of peroxisomes biogenesis. In Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The Metabolic Basis of Inherited Disease, 6th edn. New York: McGraw-Hill. Poll-The BT, Maroteaux P, Narcy C et al (1991) A new type of chondrodysplasia punctata associated with peroxisomal dysfunction. J Inher Metab Dis 14" 361-363. Schrakamp G, Schalwijk CG, Schutgens RB, Wanders RJ, Tager JM, Van den Bosch (1988) Plasmalogen biosynthesis in peroxisomal disorders: Fatty alcohol versus alkylglycerol precursors. J Lipid Res 29: 325-334. Schutgens RBH, Romeyn GJ, Wanders RJA, van den Bosch H, Schrakamp G, Heymans HSA (1984) Deficiency of acyl-CoA dihydroxyacetone phosphate acyltransferase in fibroblasts from patients with Zellweger (cerebrohepatorenal) syndrome. Biochem Biophys Res Commun 120: 179-184. Schutgens RBH, Heymans HSA, Wanders RJA et al (1988) Multiple peroxisomal enzyme deficiencies in rhizomelic chondrodysplasia punctata. Comparison with Zellweger syndrome, Conradi-Hiinermann syndrome and X-linked dominant type of chondrodysplasia punctata. Adv Clin Enzymol 6: 57-65. Schutgens RBH, Nijenhuis AA, Wanders RJA, Fruman MEJ (1992) Peroxisomal disorders. Recent observations in the metabolism of very long chain fatty acids and plasmalogens. Int Pediatr 7: 294-303. Schutgens RBH, Bouman IW, Nijenhuis AA, Wanders RJA, Fruman MEJ (1993) Profiles of very long chain fatty acids in plasma, fibroblasts, and blood cells in Zellweger syndrome, X-linked ALD, and rhizomelic chondrodysplasia punctata. Clin Chem 39: 1632-1637. Wanders RJA, Heymans HSA, Schutgens RBH, Barth PG, van den Bosch H, Tager JM (1988) Peroxisomal disorders in neurology. J Neurol Sci 88:1 39. Wanders RJA, van Roermund CWT, Griffioen M, Cohen L (1991) Peroxisomal enzyme activities in the human cell line HepG2 as compared to human liver. Biochim Biophys Acta 1115: 54-59. Wanders RJA, Schumacher H, Heikoop J, Schutgens RBH, Tager JM (1992) Human dihydroxyacetonephosphate acyltransferase deficiency: A new peroxisomal disorder. J Inher Metab Dis 15: 389-391. Wanders RJA, Dekker C, Horvath A et al (1994) Human alkylhydroxyacetonephosphate synthase deficiency: a new peroxisomal disorder. J Inher Metab Dis 17:315-318.

J. Inher. Metab. Dis. 17 (1994)