Lancelot Hogben Scientific Humanist. An Unauthorised Autobiography

complete absence of glucocerebrosidase was caused by homozygosity for a null mutation.4 This report describes an unusual association of GD type 1 and JS in a nonconsanguineous family of Indonesian and white Dutch ancestry (fig 1). The proband, patient I.1, was born at term with a birth weight of 3550 g, a length of 54.5 cm, and a head circumference of 38.5 cm (>98th centile). The diagnosis ofJS was made by the presence of features including episodic hyperpnoea/apnoea, agenesis of the cerebellar vermis and corpus callosum, hydrocephalus, and chorioretinal colobomata. Severely delayed psychomotor development and generalised seizures were the major clinical features until death at the age of 4 years. Unexpectedly, lysosomal enzyme investigations showed a severe deficiency of glucocerebrosidase activity in cultured skin fibroblasts. Molecular studies showed compound heterozygosity N370S/L444P, the most common genotype in patients with GD type 1 in The Netherlands. Patient 11.3 was born after an uneventful pregnancy and delivery with normal weight, length, and head circumference. At the age of 2 years, she had retarded mental development and autistic behaviour. Magnetic resonance imaging of the brain was normal at 7 years. At this age, she had no clinical features of GD type 1, except for mild hepatosplenomegaly. Like her older brother, she appeared to have deficient glucocerebrosidase activity associated with compound heterozygosity for GD type 1. In the third patient (II.4) at 16 weeks of gestation hydrocephalus was detected by ultrasound. Birth weight was 3220 g and head circumference 39.3 cm (>98th centile). The patient fulfilled the diagnostic criteria for JS and he died at the age of 8 month; no material was available for analysis. This non-consanguineous family with two boys affected by JS and a girl with autistic behaviour was identified to have the most frequent genotype of GD type 1 in the Dutch population.6 The presence of a severe neurological disorder such as JS and autistic behaviour cannot be explained by the N370S/ L444P GD genotype alone. To address the possibility that the features ofGD type 1 have been masked by the early onset of severe manifestations of JS, we investigated eight additional patients with JS. In these patients, we found a normal glucocerebrosidase activity in fibroblasts. These results suggest that the JS and GD loci do not (simply) coincide. The most likely explanation for the coexistence of the two disorders in one person is the independent association ofGD and JS. In this case, our observation may be unique, since the statistical probability of this event is extremely small in a non-consanguineous and interracial relationship. The incidence of GD type 1 is estimated at 1:50 0003 and no more than 100 cases of JS have been reported.' Therefore, it may be worth considering other explanations. In this respect it is of interest to note the large clinical variability among GD patients with identical mutations, even within families.7 All patients with GD types 1, 2, and 3 have significant levels of residual glucocerebrosidase activity (3-8% ofthose in controls; Kleijer and Aerts, unpublished data), with the exception of a neonatal variant of GD with prenatal onset of fetal hydrops.4 Although our patients have a residual activity of 3-8% in fibroblasts, it remains possible that a complete knock out of glucocerebrosidase activity is present in some tissues, for example, the central nervous system in patient II.1 (but not in patient II.3) by an as yet unknown factor, interacting with the transcription or translation of the gene or with the enzyme activity.