Editors: Norio Fujiki, M.D. & Darryl R.J. Macer, Ph.D.
Vice Director, The Tokyo Metropolitan Institute of Medical Science, Tokyo, Japan
Gangliosides are characterized by a unique molecular structure containing sialic acid at the non-reducing terminal of the carbohydrate chain, and rich in the central nervous system. Two groups of inherited diseases are known to be due to abnormal neuronal storage of ganglioside, resulting in severe neurological manifestations in infancy: GM1-gangliosidosis and GM2-gangliosidosis. They have in fact been considered representative of classical neurodegenerative disease in infancy.
The carbohydrate moiety in the ganglioside molecule is degraded in order from its outer chain terminus by specific hydrolases. The terminal galactose in ganglioside GM1 is cleaved off by beta-galactosidase, producing ganglioside GM2. The latter in turn is further degraded by beta-hexosaminidase A, producing ganglioside GM3. Mutations of the gene coding for either of these enzymes are expressed clinically as a progressive neurological disease occurring mainly in infants, and rarely in children or even in adults. beta-Galactosidase deficiency results in GM1 accumulation (GM1-gangliosidosis) and deficiency in beta-Hexosaminidase A, GM2 accumulation (GM2-gangliosidosis of the classical type: Tay-Sachs disease). Another GM2-gangliosidosis in which globoside accumulates as well as GM2 is known: Sandhoff disease. In humans genetic information is encoded on different chromosomes for these three different gangliosidoses beta-galactosidase by a gene on chromosome 3, beta-hexosaminidase A by one on chromosome 15, and beta-hexosaminidase B by one on chromosome 5.
All known gangliosidoses are transmitted as autosomal recessive traits. In general, the parents of patients are heterozygous carriers of the disease. The biochemical and genetic backgrounds of the gangliosides have been extensively studied and a number of gene mutations have been reported in both disease groups.
In most cases, gangliosidoses which have been known for over a century are clinically manifested as neurodegenerative diseases in infancy and childhood. Macular cherry-red spots are a hallmark of them, and often the only specific physical finding for the diagnosis of GM2-gangliosidosis. Additional somatic changes are often observed in young GM1-gangliosidosis patients such as dysmorphism, bone dysplasias, and hepatosplenomegaly.
Recent enzymatic screening, however, has revealed rare and atypical cases of both diseases. Diagnosis of GM2-gangliosidosis has been established in patients with cerebellar ataxia, peripheral neuropathy, or other neurological manifestations in adults. Progressive dystonia has been observed as one of the characteristic clinical manifestations of adult GM1-gangliosidosis. These patients were diagnosed first as a result of extensive enzymatic screening using synthetic fluorogenic compounds as substrates for enzyme assays beta-galactosidase for diagnosis of GM1-gangliosidosis, and beta-hexosaminidase A and/or B for diagnosis of GM2-gangliosidoses.
beta-Galactosidase deficiency has been found not only in GM1-gangliosidosis but also in a rare systemic bone disease, Morquio B disease. These have now been confirmed as being caused by mutations of the same gene, and therefore we have recently proposed the term "beta-galactosidosis" for them (1) (Table 1).
Figure 1: beta-Galactosidase activity in GM1-gangliosidosis.
(A) Leukocyte enzyme activity (B) plasma enzyme activity
Lysosomal enzymes are expressed in all somatic cells except mature erythrocytes. We therefore use clinically available samples for diagnostic enzyme assays plasma/serum, leukocytes, or cultured skin fibroblasts. Diagnosis of a patient with GM2- or GM1-gangliosidosis is not difficult, as the enzyme activity is almost completely lost, but the diagnosis of heterozygous carriers is not always easy, as there is almost always an overlap in calculated activities between normal and carrier groups (Figure 1).
Serum or plasma is used for diagnosis of beta-hexosaminidase deficiency disorders (GM2-gangliosidosis), because of the high extracellular enzyme activity. Separation of A and B fractions used to be necessary for diagnosis of Tay-Sachs disease, but a substrate specific to beta-hexosaminidase A is currently available, and so now the diagnosis has been made much easier.
On the other hand, beta-galactosidase activity is low in the extracellular fluid, and we usually assay intracellular enzyme activity for conclusive diagnosis of GM1-gangliosidosis.
A number of mutations have been identified for both GM2-gangliosidosis and GM1-gangliosidosis. During the past few years, we have analyzed mutant genes for beta-galactosidase in GM1-gangliosidosis and Morquio B disease patients. Most mutations are single base substitutions resulting in single amino acid substitutions (2,3). Splicing defect, duplication, or insertion mutations have also been identified. Several different analytical methods are available for detection of these mutations, and the suitability depends on the exact conditions. Large deletions or duplication/insertions may be detected by Southern analysis, but we need special methods for the detection of small rearrangements or single base substitutions.
In my laboratory, restriction analysis, which results in changes in a restriction site in the region of a mutation, is often used in work on single base substitutions. The nucleotide sequence around the mutation is first amplified by polymerase chain reaction (PCR), and is then restriction-digested and subjected to agarose gel electrophoresis. The generation or elimination of a restriction site is identified by observation of the digested bands on the gel.
In Figure 2, an example of diagnosis of the R201C mutation is shown. It is a common mutation in Japanese patients with juvenile GM1-gangliosidosis. The 201Arg to Cys (CGC to TGC) substitution results in elimination of a site for the BspMI restriction enzyme. A 181-bp sequence comprising the mutation site is PCR-amplified, digested by the enzyme, and subjected to electrophoresis. A normal allele is completely digested to 128-bp and 53-bp fragments, but the R201C mutant allele becomes resistant to digestion, the 181-bp PCR product being undigested and appearing as a single band on the gel after BspMI treatment. Those cells containing both alleles, from a subject heterozygous for this mutation, produce all three fragments a normal 181-pb fragment, and two 128-bp and 53-bp digested fragments, simultaneously. Using this analytical procedure, the result is clear, and there is no overlap between a heterozygote and a normal homozygote, i.e. carriers and normal groups.
The region around a small deletion, insertion, or duplication, can be PCR-amplified in order to detect changes in size of the PCR product and so our duplication cases could be diagnosed directly by this simple procedure. However diagnosis is not always easy in the case of small deletions where large fragments have to be investigated by methods such as Southern blotting.
Genetic mutations do not always show up as such in the phenotype. To put it in another way, even for the same mutation there are cases where clinical severity and the symptoms presented differ. We have established a clear correlation between specific common mutations for late-onset clinical phenotypes of beta-galactosidosis; R201C mutation for juvenile GM1-gangliosidosis; (Japanese patients), I51T mutation for adult GM1-gangliosidosis (Japanese patients), and W273L mutation for Morquio B disease (Caucasian patients). These mutations were found in all patients we studied with these specific clinical phenotypes.
Among them, the I51T mutation was found in all Japanese adult GM1-gangliosidosis patients as homozygotes, except for one patient heterozygous for this mutation (Table 2); they all showed the phenotype I51T/I51T, but phenotypic expression was extremely heterogeneous, with regard to the age of onset and clinical course, although individual clinical signs were almost identical in every case.
The age of onset was variable even among siblings from the same parents. In general, a second younger sibling affected with the same inherited disease in a family is detected earlier than the proband. The age of onset in sibling cases in Table 2 (family 2, 8, 9, or 10), however, cannot be explained simply by the watchfulness of parents. We need more data on the pathogenesis of phenotypic expression in beta-galactosidosis.
This observation on sibling cases has raised another problem concerning the ethical aspect of late-onset inherited diseases. It is possible that a homozygous I51T mutation may be detected in a family member of a proband with this genotype, even if he or she is clinically symptomless, or in the preclinical state. Of course we can confirm the diagnosis by enzyme assay, but how should we handle this case? This is essentially the same problem we face with Huntington's disease or hereditary amyloidotic neuropathy in adults. This will be one of the major issues for future discussion and consideration.
*Age: years. #Percent of normal beta-galactosidase activity. Diag: diagnosis, Mental: mental disability, ExPyr: extrapyramidal signs and symptoms.
The cases in family 2, 8, 9, and 10 are siblings.