Accurate Mutation Analysis in 54 British and Irish Gauchers Patients


Return to contents page 


The Willink Unit at the Royal Manchester Children's Hospital has provided a diagnostic service for Gauchers disease and other lysosomal storage disorders since 1974. Over this period 90 patients with Gauchers disease have been identified. Dr Ed Wraith, director of the unit, is a consultant paediatrician with a special interest in lysosomal storage disorders and is in charge of the Gauchers Clinic which was given specific funding earlier this year by the Department of Health. Lorraine Burnett is the clinical nurse specialist responsible for patients on Cerezyme treatment. Dr Alan Cooper is the principal biochemist involved with these disorders since the diagnostic service was initiated and describes in this article some of his recent work following his talk at the EWGGD in Maastricht on 3 May 1997. Claire Hatton and Wendy Savage are responsible for molecular and biochemical analyses.

Gauchers disease is a lysosomal storage disorder which results in accumulation of the lipid glucocerebroside within cells. The accumulation is due to a deficiency of the enzyme glucocerebrosidase which normally breaks down this compound.
Measurement of this enzyme in white blood cells, under carefully controlled conditions allows a diagnosis of Gauchers disease Type 1, 2 or 3 to be made. It must be emphasised that considerable experience is needed both to accurately perform and interpret the results of such assays (tests).

Until recently, the three types of Gauchers disease could only be differentiated by clinical observation. In the last few years, molecular biology techniques have been devised which allow a laboratory such as ours to identify the exact mutations causing the disease in a patient.

There is relatively good correlation between genotype (mutations) and phenotype (severity of the disease) although this correlation is not absolute as will be discussed later. Thus if the genotype is known, it is possible to predict the likely outcome of the disease even in a young baby.

Gene Codes
The location of the gene for Gauchers disease is known and the gene like all others is made up of 4 bases (A, T, G and C) linked end to end. The gene carries the code needed to make the glucocerebrosidase enzyme.

It specifies which amino acids should be included and at what point. The bases are grouped in threes (codons): this provides enough unique codes for all the amino acids used to make enzyme proteins. For example the codon CCC is the code for the amino acid Proline whilst the codon AGT is the code for the amino acid Serine.

Mistakes which occur within the glucocerebrosidase code (mutations) are what are responsible for Gauchers disease. A number of laboratories around the world are now able to test for mutations in Gauchers patients. The most common technique used is called restriction digestion analysis.

A restriction enzyme cuts DNA when it encounters a specific sequence of bases. For example the restriction enzyme Nci I will cut a strand of DNA when it encounters the sequence CCGGG whilst the restriction enzyme Xho I cuts at the sequence CTCGAG.

As mutations generally substitute one base for another eg T for a G, the mutations often either destroy a site where a restriction enzyme cuts the normal sequence (so it won't be seen) or create a new cutting site for a restriction enzyme which is not present in the unmutated sequence.

For instance, one of the most common mutations in Gauchers disease called L444P changes the normal sequence of bases CCTGG to CCGGG (the T base is substituted by a G). As we already know, the restriction enzyme Nci I cuts DNA at the sequence CCGGG.

Thus by creating millions of copies of the portion of the glucocerebrosidase gene where the mutation occurs, using a technique called the polymerase chain reaction (PCR) and then digesting the amplified DNA with Nci I, the L444P mutation can be readily identified. DNA which contains the mutated sequence CCGGG is cut into two smaller fragments whilst the normal sequence which does not contain the mutation remains uncut, as can be seen by the illustration below:

1                    2                   3

1  Normal gene not cut by enzyme.
2  Carrier of L444P, half DNA cut into two  fragments and half not cut.
3 Homozygous (both genes) for L444P, all DNA cut to two smaller fragments.

How Mutations are Named
Point mutations (where one base is substituted for another) are often represented by a number and two letters. The number is the position of the amino acid in the glucocerebrosidase enzyme which is changed, the letter which precedes the number represents the amino acid in the normal enzyme protein whilst the letter which follows the number represents the amino acid which is substituted due to the mutation.

For example the L444P mutation changes the four hundred and forty fourth amino acid from a Leucine (L) to a Proline (P). Where a base is inserted into the gene this is indicated by the number of the base preceding the insertion followed by the base inserted. eg 84G indicates that a G has been inserted after the eighty fourth base of the gene. (This mutation has also been called 84GG in the past).

Where bases are deleted from the gene, they are represented by the number of the first base deleted followed by the number of bases deleted. eg. 1263Del 55 represents a deletion of fifty five bases starting at the one thousand two hundred and sixty third base of the gene.

The Pseudo-Gene
An almost exact copy of the glucocerebrosidase gene exists, this is known as the pseudo-gene. It has no known function and thus is free to mutate without any detrimental effect. This accounts for the number of point mutations present in the pseudo-gene sequence which are absent from the Glucocerebrosidase gene.

Occasionally DNA is exchanged between the glucocerebrosidase gene and the pseudo-gene, this is known as recombination. Such recombination can result in the incorporation of several mutations into the active glucocerebrosidase gene. There recombinant (Rec) alleles may result in Gauchers disease.

Mutation Analysis
Three mutations have been described as common in a number of countries most notably the United States. These mutations are N370S, L444P and 84G. We have found that in the UK that the R463C mutation is also common and several other mutations appear more frequently than described elsewhere.

This is why in our laboratory DNA from Gauchers patients is analysed for ten known mutations, 84G, N370S, L444P, R463C, R496H, IVS2+1, D409H, RecNciI, RecTL and 1263Del 55.

The point mutations are initially detected by restriction digest analysis, whilst recombinant mutations and the deletion are detected by sequencing and PCR respectively. Point mutations identified are further analysed by sequencing and a number of anomalies have been detected.

Anomalies Detected
Two patients thought to carry the R463C mutation using restriction digestion were subsequently shown to have the R463Q mutation and a new point mutation N462K. Seven patients who apparently carried the L444P mutation were shown to actually have the recombinant mutation RecNci I. This mutation arises by exchange of DNA between the functional and pseudo genes. The L444P mutation occurs naturally in the pseudo-gene and is transferred to the functional glucocerebrosidase gene by DNA exchange. However in RecNci I alleles, two further mutations which naturally occur in the pseudo-gene A456P and V460V are also transferred to the functional gene.

Two new recombinant alleles were also identified by this technique RecA456P and c1263Del 55+RecTL. The latter would have appeared homozygous L444P by restriction digestion. One patient thought to be homozygous R463C was later shown, by PCR, to carry one copy of the 55 base deletion 1263Del 55.

When one or more mutations remains unknown, the patients DNA is further analysed by sequencing of the coding region of the glucocerebrosidase gene. To date we have only analysed six of the eleven exons (coding regions). Several rare or unique mutations have been identified.

Genotype to Phenotype Correlation
Genotype         Adult 1  Child 1  Type 2   Type 3
N370S/N370S              4             0           0            0
N370S/L444P               4             0           0            0
N370S/RecNci I           4             0           0            0
N370S/L105R               1            0           0            0
N370S/R120W             1            0           0            0
N370S/?                        4            0          0            0
N370S/R463Q              0            1          0            0
N370S/RecA456P        0            1           0           0
L444P/R463C               2            1           0           0
L444P/?                        2            1          7            1
L444P/L444P               0            2           1            6
L444P/1263Del
              55+RecTL     0              0          1            0
R463C/R463C            1              0          0            0
R463C/RecNci I         0              3          0            0
R463C/IVS2+1           0              1          0            0
R463C/D409H           0               1          0           0
R463C/1263Del 55    0               1          0           0
1263Del 55/?              0               0         1           0
N462K/?                     0               0         1           0
?/?                               0              0          1           0
TOTAL                     23              12        12          7

As can be seen from the above table, there is reasonably good correlation between the mutations carried by a patient and the clinical course of that patient's disease.

Mild Disease
Patients with two copies of the N370S mutation invariably have relatively mild adult onset disease. The presence of one copy of the N370S mutation excludes Type 2 or 3 disease. Only two patients with one copy of this mutation had childhood onset disease and this probably reflects the severe nature of their second mutation.

We have identified the R463C mutation in the homozygous state for the first time. This mutation can now be classified as a relatively mild mutation as the homozygous patient had adult onset Type 1 disease. Furthermore all other patients with one copy of this mutation has childhood onset Type 1 disease despite carrying severe mutations as their second allele (gene).

More Severe Disease
The situation is less clear with the L444P mutation. The majority of patients with two copies of this mutation develop Type 3 disease. However one patient presented with severe Type 2 disease and two patients developed childhood onset Type I disease. The elder of these latter patients is now 30 years of age and confined to a wheel chair with severe bone disease. She has still shown no evidence of central nervous system involvement. In a paediatric centre such as ours, this apparent lack of correlation between genotype and phenotype for the L444P mutation is not problematic. Patients homozygous for this mutation always present in childhood with relatively severe disease regardless of whether the central nervous system is involved. We would always start such patients on high dose Cerezyme treatment unless there was strong clinical evidence of Type 2 disease.

Conclusions
Mutations resulting in Gauchers disease can now be readily identified by relatively simple techniques. This information is invaluable for making clinical decisions on the best form of treatment for patients and also for making some predictions about long term clinical outcome of individual patient's disease at an early stage.

Identification of Gauchers disease mutations provides the only reliable method of identifying carriers within a family when required. Theoretically mutation analysis could also be used for first trimester prenatal diagnosis but would offer little advantage over enzyme assay and would probably be less reliable.

Mutation analysis should not be undertaken by restriction digestion analysis alone as this has been shown to result in a significant number of incorrect genotypes. The results should be confirmed by sequencing of genomic DNA.
The method of choice would be automated fluorescent dye terminator cycle sequencing. This technique is rapid and involves little more work than restriction enzyme analysis once the site of the mutation is known.

In the Willink Unit we are happy to accept samples for mutation analysis from centres throughout the UK. If initial results have been obtained by restriction analysis, we would be able to confirm these results by sequencing. We also regularly receive samples from Ireland, India, Turkey, Taiwan, Saudi Arabia and several other countries around the world.

Mutation analysis is a powerful tool which has significantly increased knowledge about Gauchers disease. If done correctly it provides valuable clinical information for both doctors and families with this condition.


Return to contents page

Source: Gauchers News November 1997

© Copyright Gauchers Association 1997