- Email: [email protected]
Disease-causing mutations in cis with the common arylsulfatase A pseudodeficiency allele compound the difficulties in accurately identifying patients and carriers of metachromatic leukodystrophy
Molecular Genetics and Metabolism 79 (2003) 83–90 www.elsevier.com/locate/ymgme
Disease-causing mutations in cis with the common arylsulfatase A pseudodeficiency allele compound the difficulties in accurately identifying patients and carriers of metachromatic leukodystrophy Mohammad A. Rafi, Stephanie Coppola, Shu Ling Liu, Han Zhi Rao, and David A. Wenger* Lysosomal Diseases Testing Laboratory, Department of Neurology, Jefferson Medical College, 1020 Locust St., Room 394, Philadelphia, PA 19107, USA Received 28 February 2003; received in revised form 14 April 2003; accepted 16 April 2003
Abstract Metachromatic leukodystrophy (MLD) is a lysosomal storage disorder most often caused by mutations in the sulfatide sulfatase or arylsulfatase A (ASA) gene. This results in the storage of sulfatides in the peripheral and central nervous systems as well as in the kidneys. Patients with MLD exhibit a wide range of clinical features presenting from the late infantile period to adulthood. Testing for this disease is performed on a majority of the patient samples received for diagnostic testing in the authorsÕ laboratory. If low ASA activity is measured, additional testing is required to confirm the diagnosis due to several factors. ASA activity is also low in individuals with multiple sulfatase deficiency and in individuals with copies of the so-called pseudodeficiency (Pd) allele. Due to the high frequency of the common Pd allele, it is possible for individuals, both with and without neurologic disease, to have low ASA activity but not have MLD. Unfortunately, the finding of the Pd mutation by molecular analysis does not rule out a diagnosis of MLD. In a recent 25 month period, this laboratory diagnosed 52 patients with MLD, and of these, 13 patients from 10 families had one or two copies of the Pd mutation. Sequencing of the ASA gene in these 10 families revealed four new mutations in cis with the Pd allele (S43R, R84Q, R311X, and E329R) and two additional new mutations (R299W, C488R). Six patients had previously reported mutations on the Pd background. Thus, a total of 14 mutations have been found to occur in cis with the Pd allele. We estimate that 1–2% of Pd alleles will have a disease-causing mutation, and this complicates the identification of patients and the assignment of risk for a couple when a copy of the Pd allele is detected. Ó 2003 Elsevier Science (USA). All rights reserved. Keywords: Arylsulfatase A; Genotype–phenotype correlations; Lysosomal storage disease; Metachromatic leukodystrophy; Polymorphisms; Pseudodeficiency; Sulfatide
Introduction Metachromatic leukodystrophy (MLD) is an autosomal recessive lysosomal storage disorder that results from the accumulation of sulfatides in the central and peripheral nervous systems (reviewed in [1]). Patients exhibit a wide range of presenting symptoms, and their clinical courses can vary significantly. Late infantile,
* Corresponding author. Fax: 1-215-955-9554. E-mail address: [email protected] (D.A. Wenger).
juvenile and adult onset forms are generally recognized. MLD is usually caused by a deficiency of arylsulfatase A (ASA; sulfatide sulfatase; E.C. 3.1.6.1) activity which can be measured in several easily obtained tissue samples, including leukocytes isolated from whole blood or cultured skin fibroblasts [2]. In addition, some rare individuals store and excrete sulfatides due to a deficiency of saposin B, the required activator protein that facilitates the interaction between ASA and sulfatide [3]. The diagnosis of MLD is also complicated by several other factors. The high frequency of the so-called pseudodeficiency (Pd) allele results in low ASA activity
1096-7192/03/$ - see front matter Ó 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S1096-7192(03)00076-3
84
M.A. Rafi et al. / Molecular Genetics and Metabolism 79 (2003) 83–90
in individuals who do not have MLD and in carrier levels of ASA activity in individuals who are not carriers of MLD. This is due to a mutation in the polyadenylation signal that results in only about 10% of ASA messenger RNA being produced [4]. However, the finding of the Pd allele by molecular analysis does not rule out a diagnosis of MLD since there can be additional mutations on the Pd background [1,5–10]. In addition, patients with multiple sulfatase deficiency (MSD) will have low ASA activity as well as low levels of other sulfatases. These patients usually, but not always, have clinical features of both a mucopolysaccharidosis and a leukodystrophy. Because of the wide clinical spectrum observed in patients with MLD, ASA activity is measured in many individuals whose tissue samples are received in the Lysosomal Diseases Testing Laboratory. Since 1974 this laboratory has identified over 480 patients with MLD. In the 25 month period from November 2000 until December 2002, we diagnosed 52 individuals, including affected siblings, with MLD based on a combination of clinical, biochemical and magnetic resonance imaging (MRI) studies. With the cloning of the ASA cDNA [11] and gene [12], a large number of MLD-causing mutations and polymorphisms have been delineated (reviewed in [1]). In addition, the major cause of Pd has been determined [4]. While we normally do not attempt to identify all mutations in patients with low ASA activity, we do check for the mutation in the polyadenylation signal that results in Pd [4] and the common mutations associated with late infantile and adult MLD [13]. These mutations, G459 + 1A and P426L, were termed the ‘‘I’’ and ‘‘A’’ alleles, respectively, by Polten et al. [13]; however, it is obvious from this study and others that this designation only holds for patients homozygous for these mutations. In addition, urinary sulfatide excretion is examined in all individuals with low ASA activity. Thirteen patients, representing 10 families, diagnosed in the 25 month period of study were either heterozygous or homozygous for the common Pd mutation as determined by polymerase chain reaction (PCR)-based analysis and sequencing. While some MLD-causing mutations have previously been reported in cis with the common Pd mutation [1,5–10], it became obvious that four of the recent cases of MLD identified in this laboratory who had the Pd allele did not have any of the previously reported mutations. These individuals ranged clinically from late infantile to adult onset. A few of the patients who were heterozygous for the Pd allele were also heterozygous for either the common G459 + 1A or P426L mutation. Four new MLD-causing mutations on the Pd background were identified. In addition, two unreported disease-causing mutations not on the Pd allele were also found in these 10 families. Since about 1 in 7 individuals in the general population is heterozygous for the common Pd allele, these findings
have implications for patient and carrier identification within families and for individuals who marry into such a family.
Materials and methods Patients heterozygous or homozygous for the Pd allele included in this study Samples from the patients used in this study were sent here in an attempt to obtain a diagnosis to explain their neurologic and/or psychiatric symptoms. Initially, a panel of lysosomal enzyme activities, including ASA, was measured in sonicated leukocytes isolated from whole heparinized blood [2,14]. Since the ASA activity was within the range of patients previously confirmed to have MLD and other enzymes, including other sulfatases, were normal, a diagnosis of MLD was suspected. All patients with low ASA activity were initially tested for the Pd allele to aid in confirming the diagnosis and to determine if the low ASA activity could be explained by the presence of the Pd mutation. In addition, a first morning voiding of urine was requested for sulfatide extraction. A brief summary of the clinical and biochemical findings on the individuals studied is presented in Table 1. These patients ranged in age from 18 months to 48 years. While clinical information is requested with each specimen, the amount of information supplied varies greatly. While many of the younger patients had the typical features of late infantile MLD, the older patients tended to have motor and cognitive decline and psychiatric problems leading to dementia. When performed, MRI changes were consistent with a diagnosis of MLD. All had ASA activities in leukocytes within the range of patients with MLD, and excess sulfatides were clearly detected in urinary extracts (Fig. 1). While most of the original samples came from clinically affected individuals, younger asymptomatic siblings may be tested because of the diagnosis of MLD in the proband. It is critical that the parents of the proband be tested before or together with the siblings to prevent inaccurate interpretation of the results in the ‘‘normal’’ sibling. If a healthy parent is ‘‘pseudodeficient’’ (MLD/ Pd) and has not been tested, the measurement of low ASA activity in the sibling can result in an incorrect diagnosis. Urinary sulfatide excretion Ten ml of first morning voiding of urine was brought to pH 5.0 with a few drops of glacial acetic acid and placed at 4 °C overnight. The urine was then centrifuged at 1500g for 10 min. The precipitate was dissolved in 1 ml of distilled water, and 5 ml of chloroform–methanol (2–1, by vol) was added. After vortexing for 1 min, the
M.A. Rafi et al. / Molecular Genetics and Metabolism 79 (2003) 83–90
85
Table 1 Clinical and biochemical findings on patients heterozygous or homozygous for the Pd allele included in this study Family: patienta (Year of Birth)
Age of onset
Signs and symptoms
ASA activityb
Sulfatide excretionc
1: 1 (2000) 2: 1 (1999)
18 months 1–2 years
DD, hypotonia, esotropia, PR, and CSF protein 123 Clumsiness noted at 1 year with normal developmental milestones until about 24 months, PR obvious by 27 months, MRI c/w leukodystrophy None at time of diagnosis
4.2 10.0
+ ND
7.6
+
DD, spasticity, nystagmus, PR, and MRI c/w leukodystrophy PR, genu recurvatum, hypertonia, and NCV c/w leukodystrophy DD, nystagmus, spasticity, weakness, CSF protein 143, MRI c/w leukodystrophy, and decreased speech
3.1
+
3.7
+
4.6
+
PR, NCV and MRI c/w leukodystrophy, and CSF protein 79 Cognitive decline, psychiatric problems, and MRI c/w leukodystrophy None at time of diagnosis
12.3
+
4.6
+
0
+
N/A Cognitive decline, psychiatric problems, and CSF protein 150, MRI c/w leukodystrophy Cognitive decline, psychiatric problems, dementia, and NCV and MRI c/w leukodystrophy None at time of diagnosis
2.8 6.4
+ +
5.2
+
5.9
+
2 (2000) 3: 1 (1999)
Asymptomatic at 18 months 24 months
4: 1 (1999)
12–18 months
5: 1 (1999)
30 months
6: 1 (1994)
6 years
7: 1 (1991)
10 years
2 (1992) 8: 1 (1970) 9: 1 (1955)
Asymptomatic at 10 years N/A 42 years
10: 1 (1955)
42 years
2 (1960)
Asymptomatic at 40 years
Comments
African American/ Caucasian
Consanguineous parents from Bangladesh
Died following HSCT
Abbreviations: CSF, cerebrospinal fluid; c/w, consistent with; DD, developmental delay; HSCT, hematopoietic stem cell transplantation; N/A, not available; NCV, nerve conduction velocity; ND, not done; PR, psychomotor regression. a Multiple patients listed as one family are siblings. b nmol/h/mg protein measured in leukocytes. Normal mean standard deviation, 71.1 11.3. c ‘‘+’’ indicates the excretion of excess sulfatides clearly visible on the thin layer chromatography plate.
Fig. 1. Detail of the thin layer chromatography plate showing the sulfatide excretion from patients with MLD and a control. All samples were extracted as described in Materials and methods. The amount spotted was equivalent to 5 mg creatinine, except lane 2 which was equivalent to 1 mg creatinine. Lane 1, patient #1; lane 2, patient with late infantile MLD and no copies of the Pd allele; lane 3, patient #6; lane 4, patient #7; lane 5, sulfatide standard; lane 6, control; lane 7, patient #8; lane 8, patient #10.
mixture was centrifuged at 1500g for 5 min. The upper phase was removed and discarded, and the lower phase was evaporated with nitrogen. The residue was dissolved in 0.05 ml of chloroform–methanol (2–1, by vol), and half was spotted on a silica gel thin layer chromatography plate along with standard sulfatide. The plate was placed in a tank containing chloroform–methanol–water
(72–28–3.5, by vol), and the solvent was allowed to run until it was near the top of the plate. After drying, the plate was sprayed with orcinol reagent to detect sugarcontaining lipids. The sulfatide present in urine was not quantitated, but little, if any, sulfatide was excreted in individuals with no mutations in the ASA gene, individuals who are carriers of MLD-causing mutations or
86
M.A. Rafi et al. / Molecular Genetics and Metabolism 79 (2003) 83–90
the Pd mutation, individuals who are homozygous for the Pd mutation (without a disease-causing mutation), and compound heterozygotes who have one MLDcausing mutation and one Pd mutation (without an additional mutation). Attempts to equalize the amount spotted based on the creatinine concentration or the level of a lysosomal enzyme, such as b-hexosaminidase, were not entirely successful. If excess sulfatide is being excreted and other evidence indicates a leukodystrophy, MLD is almost surely the diagnosis. If the activities of other sulfatases are also low, a diagnosis of MSD must be considered. Molecular analysis of patients Molecular analysis was performed using DNA isolated from the remaining leukocyte sonicates [15], or an additional blood sample was requested, if necessary. Using the method of Chabas et al. [16], all samples were first tested for the mutation in the polyadenylation signal that results in lower than normal levels of ASA messenger RNA, the common cause of Pd [4]. This method will not detect the N350S polymorphism that is commonly found in cis with the polyadenylation signal mutation. The N350S mutation will eliminate a utilized N-glycosylation site but does not lower ASA activity. Initially, only patients with late infantile or juvenile onset were tested for the common G459 + 1A mutation reported previously [13] using the method of Ben-Yoseph and Mitchell [17]. However, after observing the results of the DNA sequencing, testing for this mutation was extended to patients of any age. If the patient had juvenile or adult onset, we tested for the common C1277T (P426L) mutation using the method of BenYoseph and Mitchell [17]. If the patients were considered for further analysis, the complete ASA gene was amplified by PCR using the GeneAmp XL PCR kit (Perkin Elmer, Norwalk, CT). An upstream primer, ASAa+ (50 -TATTTGGGTCCGGGGTCTCAG-30 ), located before the initiation codon and a downstream primer, ASAb) (50 -TCTTTCTGGCTGGGTGATCTT G-30 ), located after the first polyadenylation signal were used to produce a 2983 bp DNA fragment. This product was reamplified and purified using QIAquick PCR purification kit (Qiagen, Valencia, CA). Full length original PCR products were cloned in TA vector (Invitrogen, Carlsbad, CA) and transfected into TOP 10 competent Escherichia coli. Positive clones were grown in kanamycin containing Luria–Bertani medium, and DNA was extracted using QIAprep kit (Qiagen) for sequencing. Purified PCR products were either sequenced directly or after cloning in TA vector using eight internal primers (G11, 50 -AAGAGCCTGCTGAGGCCAAG-30 ; G13, 50 -CCACATTTCTGTCTGTCTCAG-30 ; G12, 50 CCAGCCCTCACGGCAGCTG-30 ; G3, 50 -CCATGGC TCATGAGCGCCTC-30 ; G31, 50 -AACTGAGTGACT
GACCAG-30 ; G4, 50 -CCAGGTATGTGCAGTGCTT G-30 ; G41, 50 -TTGCCCTGTGCACAGAATTG-30 ; and G43, 50 -TTGCTGCCATTGCCCAGAT-30 ). Cycle Sequencing kit (USB Corporation, Cleveland, OH) and/or ABI Prism 377 DNA automated sequencer with BigDye terminator reaction (Applied Biosystems, Foster City, CA) were employed. The sequences generated from the patients and carriers were compared to the ASA sequences obtained from the National Center of Biotechnology Information (NCBI) (Accession Nos. X52150, X52151). When a patient was determined to be heterozygous for the Pd allele, assignment of additional mutations to the correct allele was based either on published data from known mutations [5,8,9,13,18] or on studies of obligate carrier parents, siblings or offspring sent here for testing. As these individuals requested genetic testing, their DNA was sequenced using similar methods once the presence or absence of the Pd mutation was determined by the simple PCR-based method [16].
Results The results of the mutation analysis in these 10 families are shown in Table 2. Patient # refers to the family # since affected siblings had the identical genotype. Two patients were found to be homozygous for the Pd mutation in the polyadenylation signal, and eight were found to be heterozygous (Fig. 2). The common G459 + 1A and P426L mutations were identified in five individuals using the published methods [17]. The C287T (S96F) mutation, previously reported to occur on the Pd background [5], was found on 4 of 20 alleles. This included two late infantile patients (#1, #3) and one juvenile patient (#7). Of the two late infantile patients, one was homozygous for this mutation (#1), and one was also heterozygous for the common G459 + 1A mutation (#3). The juvenile patient was found to be heterozygous for T536G (I179S), a previously reported mutation that is considered mild [18]. One late infantile patient (#4) was found to be heterozygous for 1216del9 (2320del9), another mutation known to be on the Pd background [8], and the common G459 + 1A mutation. Patient #2, one of two affected siblings with late infantile MLD, was found to be heterozygous for the Pd allele and heterozygous for G985A (E329R) and C895T (R299W), two previously unreported mutations. Assignment of the E329R mutation to the Pd allele was accomplished by studies on the mother who also was heterozygous for the Pd allele. Patient #5 was found to be homozygous for the Pd allele and homozygous for C931T (R311X), an unreported mutation. The nine-year-old patient #6 was heterozygous for the Pd allele and heterozygous for two additional mutations, G251A (R84Q) and the common
M.A. Rafi et al. / Molecular Genetics and Metabolism 79 (2003) 83–90
87
Table 2 Summary of mutations in the 10 families with MLD and the Pd Allele Family: allele
Mutation position
Pd allele
Effect
Comments/references
cDNAa
Geneb
1 2
C287T C287T
436 436
+ +
S96F S96F
[5] [5]
1 2
G985A C895T
1724 1544
+ )
E329R R299W
New New
1 2
C287T G459 + 1A
436 609
+ )
S96F Splicing error
[5] [13]
1 2
1216del9 G459 + 1A
2320del9 609
+ )
3 amino acid del Splicing error
[8] [13]
1 2
C931T C931T
1580 1580
+ +
R311X R311X
New New
1
G251A
400
+
R84Q
2
C1277T
2381
)
P426L
Not previously reported on Pd allele [19] [13]
1 2
C287T T536G
436 799
+ )
S96F I179S
[5] [18]
1 2
G863A G459 + 1A
1512 609
+ )
R288H Splicing error
[9] [13]
1 2
G863A G459 + 1A
1512 609
+ )
R288H Splicing error
[9] [13]
C129A T1462C
129 2566
+ )
S43R C488R
New New
1:
2:
3:
4:
5:
6:
7:
8:
9:
10: 1 2
Nucleotide number 1 is the first base of the initiation codon. a Positions numbered according to NCBI entry X52151. b Position numbered according to NCBI entry X52150.
Fig. 2. Results of the PCR amplification of genomic DNA and digestion with RsaI to detect the polyadenylation signal mutation causing Pd. Lanes 1 and 14, 100 bp DNA ladder; lane 2, undigested PCR product; lane 3, MLD patient with no copies of Pd allele; lane 4, patient #1; lane 5, patient #2; lane 6, patient #3; lane 7, patient #4; lane 8, patient #5; lane 9, patient #6; lane 10, patient #7; lane 11, patient #8; lane 12, patient #9; lane 13, patient #10.
P426L mutation. R84Q was previously reported in lateonset patients [19]; however, it was not noted to occur on the Pd background. Studies on family members confirmed that this mutation occurred in cis with the Pd
mutation. Patients #8 and #9 are unrelated adult patients of Italian ancestry who share the same genotype. Both were found to be heterozygous for G863A (R288H), a previously reported mutation on the Pd
88
M.A. Rafi et al. / Molecular Genetics and Metabolism 79 (2003) 83–90
Table 3 Summary of all mutations found in cis with the Pd allele Number
Mutation position cDNAa
1: 2: 3: 4: 5: 6: 7: 8: 9: 10: 11: 12: 13: 14:
C129A G251A C287T C406T C464G G863A C931T G973C G985A C1130T G1144A 1192del3 1216del9
Effect
Comments/references
S43R R84Q S96F P136S P155R D281Y R288H R311X G325C E329R P377L E382K DF398 3 amino acid del
New Not previously reported on Pd allele [19] [5] [9] Lebanese [1] [10], nucleotide change not given [9] New Indian [1], loss of splice donor site New [7] [6] Polynesian [1] [8]
Geneb 129 400 436 555 727 1512 1580 1622 1724 2119 2133 2181del3 2320del9
Nucleotide number 1 is the first base of the initiation codon. Positions numbered according to NCBI entry X52151. b Position numbered according to NCBI entry X52150. a
allele [9], and the common G459 + 1A mutation found in many late infantile and some juvenile patients with MLD. Patient #10 was one of two affected siblings who had onset of symptoms, mainly psychiatric in nature, in the fifth decade. The patient was heterozygous for the Pd allele and heterozygous for two additional new mutations, C129A (S43R) and T1462C (C488R). The S43R mutation has been assigned to the Pd allele by studies in family members requesting carrier testing.
Discussion MLD is one of the most commonly tested for lysosomal storage disease. However, accurate diagnosis remains a serious problem due to the factors laid out in the Introduction, especially the high frequency of the Pd allele (with and without disease-causing mutations on the same allele) in the general population and in patients with neurologic disease. The difficulties related to ‘‘pseudodeficiency’’ of ASA activity and other lysosomal hydrolases have been raised previously [20,21]. Leukocyte ASA activities measured in the many samples received here for diagnostic studies break down into three categories with some overlap. Those individuals with ASA values between 50 and 120 nmol/h/mg protein most likely have two ASA genes without any mutations (Nl/ Nl). The activity range results from variable conditions during shipping of the samples, white cell differential, effects of other genes and other unknown factors. Those individuals with values between 25 and 45 nmol/h/mg protein are either carriers of MLD (MLD/Nl, PdMLD / Nl) or carriers of the Pd mutation (without an additional mutation on the Pd allele) (Pd/Nl). PdMLD denotes the Pd allele with an additional disease-causing
mutation. ASA values between 0 and 15 nmol/h/mg protein will be measured in individuals who have MLD (MLD/MLD, PdMLD /PdMLD , and MLD/PdMLD ) and in people who do not have MLD (MLD/Pd, Pd/Pd, Pd/ PdMLD , and MSD). All of the patients used in this study were confirmed to have MLD by a number of criteria, including clinical features, MRI findings, biochemical studies and mutation analysis. The patients ranged in age from 18 months to 48 years. Among the 10 families heterozygous or homozygous for the Pd allele, seven different mutations on the Pd background were detected. Three previously reported mutations on the Pd background were found in six patients. Four new mutations in cis with the polyadenylation signal mutation and two additional new mutations were identified. Five mutations on the Pd allele were missense, one was a nonsense mutation resulting in a stop at codon 311 and one caused a splicing error. The new missense mutations were found in exons 1, 2, 5, and 6, and the nonsense mutation occurred in exon 5. A total listing of the 14 mutations occurring on the Pd background is presented in Table 3. The other two new mutations were missense mutations occurring in exons 5 and 8. Transient expression studies were not performed on the newly identified mutations since those present on the Pd background would obviously show low ASA activity, and upon sequencing the entire ASA gene, only these mutations were found. The finding of disease causing-mutations on a polymorphic background, such as the Pd allele, is not unique to MLD. It is nearly impossible to speculate what the effect would be of the same mutation on the ‘‘normal’’ background. It is clear that the contribution of the other allele may be critical to producing a phenotype. It is possible that the same mutation on the ‘‘normal’’
M.A. Rafi et al. / Molecular Genetics and Metabolism 79 (2003) 83–90
background, when occurring with a mild mutation on the other allele, would not produce clinical disease but would cause disease if the second allele provided no residual ASA activity. This may be the case in patients #8 and #9 who are heterozygous for the G893A mutation on the Pd allele and the common G459 + 1A mutation which is considered severe. While the G459 + 1A mutation is most commonly found in late infantile and juvenile patients, it has previously been described in two additional adult patients, including one with the identical genotype to patients #8 and #9 [9] and one with a known mild mutation, I179S [22]. Therefore, we have extended the testing for the G459 + 1A mutation to patients of all ages. While some publications have stated that genotype–phenotype predictions for certain mutations causing MLD will be excellent, others have hinted that it is not straight-forward [23]. There have been a number of papers describing patients with a confirmed diagnosis of MLD or with neurologic or psychiatric symptoms who have ‘‘pseudodeficiency;’’ however, most of these patients were not analyzed for the presence of additional mutations on the Pd allele [22,24–34]. Therefore, there is confusion about whether low ASA activity caused by the presence of two Pd alleles (without additional mutations) or one Pd allele with an MLD-causing mutation on the second allele would result in clinical disease. Without searching for additional mutations on the Pd background, it is difficult to interpret those studies. Since the carrier rate for the Pd allele is about 1 in 7, about 1 in 14 parents of MLD patients would have a Pd allele in addition to an MLD-causing allele. A review of ASA values in hundreds of obligate carriers confirms this estimate. While we do not have followup on many of these carriers with low ASA activity, we have never heard of any of these individuals having symptoms that can be assigned to their low ASA activity. Due to the high frequency of the Pd allele in almost all ethnic groups, it is now clear that additional mutations can occur on this background (Table 3). Thomas [21] speculated that almost a fifth of all mutations causing MLD may be found on the Pd background. Of the approximately 100 alleles characterized from the patients diagnosed during the 25 month period, 13 disease-causing mutations were found in cis with the Pd allele. The identification of 14 mutations on the Pd background (Table 3) complicates the diagnosis of patients and the assignment of carrier status when molecular analysis detects the Pd allele. While most copies of the Pd allele do not have an additional mutation, we estimate that there is approximately a 1–2% chance that a disease causing-mutation may be present in cis with the Pd allele. This compares to the estimated carrier rate of 1 in 100 for MLD in the general population. The incidence of MLD is estimated to be approximately 1 in 40,000 births in the United States. Accurate
89
prenatal diagnosis of MLD for couples truly at risk or inconclusive because of the presence of the Pd allele can be accomplished by a combination of biochemical and molecular tests performed on chorionic villus samples or cultured amniotic fluid cells. In order to prevent problems with interpretation, it is ideal to have studies performed on the parents before analyzing fetal samples. At this time, the only treatment available is hematopoietic stem cell transplantation for those individuals who are pre-symptomatic or have only mild neurologic manifestations [35]. Although there are significant risks associated with the procedure, this treatment appears to slow the progression of the disease and improve the quality of life for the patient. Hopefully additional methods of treatment will be available in the near future.
Acknowledgments This research was supported in part by a grant from NIH (2 R01 DK38795). The authors thank the physicians who supplied pertinent clinical information with the samples from their patients.
References [1] K. von Figura, V. Gieselmann, J. Jaeken, Metachromatic leukodystrophy, in: C.R. Scriver, A.L. Beaudet, W.S. Sly, D. Valle (Eds.), The Metabolic and Molecular Bases of Inherited Disease, McGraw-Hill, New York, 2001, pp. 3695–3724. [2] D.A. Wenger, C. Williams, Screening for lysosomal disorders, in: F.A. Hommes (Ed.), Techniques in Diagnostic Human Biochemical Genetics. A Laboratory Manual, Wiley-Liss, New York, 1991, pp. 587–617. [3] K. Inui, M. Emmett, D.A. Wenger, Immunological evidence for a deficiency in an activator protein for sulfatide sulfatase in a variant form of metachromatic leukodystrophy, Proc. Natl. Acad. Sci. USA 80 (1983) 3074–3077. [4] V. Gieselmann, A. Polten, J. Kreysing, K. von Figura, Arylsulfatase A pseudodeficiency: loss of a polyadenylation signal and a N-glycosylation site, Proc. Natl. Acad. Sci. USA 86 (1989) 9436– 9440. [5] V. Gieselmann, A.L. Fluharty, T. Tonnesen, K. von Figura, Mutations in the arylsulfatase A pseudodeficiency allele causing metachromatic leukodystrophy, Am. J. Hum. Genet. 49 (1991) 407–413. [6] M.L. Barth, A. Fensom, A. Harris, Missense mutations in the arylsulphatase A genes of metachromatic leukodystrophy patients, Hum. Molec. Genet. 2 (1993) 2117–2121. [7] J. Zlotogora, G. Bach, C. Bosenberg, Y. Barak, K. von Figura, V. Gieselmann, Molecular basis of late infantile metachromatic leukodystrophy in the Habbanite jews, Hum. Mutat. 5 (1995) 137–143. [8] S. Regis, M. Filocamo, M. Stroppiano, F. Corsolini, F. Caroli, R. Gatti, A 9-bp deletion (2320del9) on the background of the arylsulfatase A pseudodeficiency allele in a metachromatic leukodystrophy patient and in a patient with nonprogressive neurological symptoms, Hum. Genet. 102 (1998) 50–53. [9] L. Gort, M.J. Coll, A. Chabas, Identification of 12 novel mutations and two new polymorphisms in the arylsulfatase A gene: haplotype and genotype–phenotype correlation studies in
90
[10]
[11]
[12] [13]
[14] [15]
[16]
[17]
[18]
[19]
[20] [21] [22]
[23]
M.A. Rafi et al. / Molecular Genetics and Metabolism 79 (2003) 83–90 Spanish metachromatic leukodystrophy patients, Hum. Mutat. 14 (1999) 240–248. D.J. Halsall, E.P. Halligan, T.S. Elsey, T.M. Cox, Metachromatic leukodystrophy: a newly identified mutation in arylsulfatase A, D281Y, found as a compound heterozygote with I179L in an adult onset case, Hum. Mutat. 14 (1999) 447. C. Stein, V. Gieselmann, J. Kreysing, B. Schmidt, R. Pohlmann, A. Waheed, H.E. Meyer, J.S. OÕBrien, K. von Figura, Cloning and expression of human arylsulfatase A, J. Biol. Chem. 264 (1989) 1252–1259. J. Kreysing, K. von Figura, V. Gieselmann, Structure of the arylsulfatase A gene, Eur. J. Biochem. 191 (1990) 627–631. A. Polten, A.L. Fluharty, C.B. Fluharty, J. Kappler, K. von Figura, V. Gieselmann, Molecular basis of different forms of metachromatic leukodystrophy, N. Engl. J. Med. 324 (1991) 18–22. H. Baum, K.S. Dodgson, B. Spencer, The assay of arylsulfatases A and B in human urine, Clin. Chim. Acta 4 (1959) 453–455. E. Louie, M.A. Rafi, D.A. Wenger, Leukocyte sonicates as a source for both enzyme assay and DNA amplification for mutation for analysis of certain lysosomal disorders, Clin. Chim. Acta 199 (1991) 7–16. A. Chabas, S. Castellvi, M. Bayes, S. Balcells, D. Grinberg, L.I. Vilageliu, G. Marfany, W. Lissens, R. Gonzalez-Duarte, Frequency of the arylsulphatase A pseudodeficiency allele in the Spanish population, Clin. Genet. 44 (1993) 320–323. Y. Ben-Yoseph, D.A. Mitchell, Rapid detection of common metachromatic leukodystrophy mutations by restriction analysis of arylsulfatase A gene amplimers, Clin. Chim. Acta 226 (1994) 77–82. A.L. Fluharty, C.B. Fluharty, W. Bohne, K. von Figura, V. Gieselmann, Two new arylsulfatase A (ARSA) mutations in a juvenile metachromatic leukodystrophy (MLD) patient, Am. J. Hum. Genet. 49 (1991) 1340–1350. J. Kappler, K. von Figura, V. Gieselmann, Late-onset metachromatic leukodystrophy: molecular pathology in two siblings, Ann. Neurol. 31 (1992) 256–261. D.A. Wenger, E. Louie, Pseudodeficiencies of arylsulfatase A and galactocerebrosidase activities, Dev. Neurosci. 13 (1991) 216–221. G. Thomas, ‘‘Pseudodeficiencies’’ of lysosomal hydrolases, Am. J. Hum. Genet. 54 (1994) 934–940. J. Berger, B. Loschl, H. Berheimer, A. Lugowska, A. TylkiSzymanska, V. Gieselmann, B. Molzer, Occurrence, distribution, and phenotype of arylsulfatase A mutations in patients with metachromatic leukodystrophy, Am. J. Med. Genet. 69 (1997) 335–340. M.L. Barth, C. Ward, A. Harris, A. Saad, A. Fensom, Frequency of arylsulphatase A pseudodeficiency associated mutations in a healthy population, J. Med. Genet. 31 (1994) 667–671.
[24] J. Peiffer, K. Harzer, W. Schlote, Diffuse-disseminated sclerosis combined with partial arylsulfatase A (ASA) deficiency. Mixed heterozygosity of ASA- and pseudo-ASA-deficiency? Neuropediatrics 15 (1984) 59–62. [25] K. Farrell, D.A. Applegarth, J.R. Toone, P.M. McLeod, A.V. Savage, Pseudoarylsulfatase-A deficiency in the neurologically impaired patient, Can. J. Neurol. Sci. 12 (1985) 274–277. [26] C. Hohenschutz, W. Friedl, K.H. Schlor, A. Waheed, E. Conzelmann, K. Sandhoff, P. Propping, Probable metachromatic leukodystrophy/pseudodeficiency compound heterozygote at the arylsulfatase A locus with neurological and psychiatric symptomatology, Am. J. Med. Genet. 31 (1988) 169–175. [27] C. Hohenschutz, P. Eich, W. Friedl, A. Waheed, E. Conzelmann, P. Propping, Pseudodeficiency of arylsulfatase A: a common polymorphism with possible disease implications, Hum. Genet. 82 (1989) 45–48. [28] S. Sangiorgi, A. Ferlini, A. Zanetti, M. Mochi, Reduced activity of arylsulfatase A and predisposition to neurological disorders: analysis of 140 pediatric patients, Am. J. Med. Genet. 40 (1991) 365–369. [29] J. Kappler, P. Leinekugel, E. Conzelmann, W.J. Kleijer, A. Kohlschutter, T. Tonnesen, M. Rochel, F. Freycon, P. Propping, Genotype–phenotype relationship in various degrees of arylsulfatase A deficiency, Hum. Genet. 86 (1991) 463–470. [30] M.L. Barth, A. Fensom, A. Harris, Prevalence of common mutations in the arylsulphatase A gene in metachromatic leukodystrophy patients diagnosed in Britain, Hum. Genet. 91 (1993) 73–77. [31] J.M. Penzien, J. Kappler, N. Herschkowitz, B. Schuknecht, P. Leinekugel, P. Propping, T. Tonnesen, H. Lou, H. Moser, S. Zierz, E. Conzelmann, V. Gieselmann, Compound heterozygosity for metachromatic leukodystrophy and arylsulfatase A pseudodeficiency alleles is not associated with progressive neurological disease, Am. J. Hum. Genet. 52 (1993) 557–564. [32] S. Leistner, E. Young, C. Meaney, B. Winchester, Pseudodeficiency of arylsulfatase A: strategy for clarification of genotype in families of subjects with low ASA activity and neurological symptoms, J. Inherit. Metab. Dis. 18 (1995) 710–716. [33] M.H. Ricketts, J.D. Amsterdam, D.S. Parks, R.-S. Yang, R.D. Poretz, X. Zhang, M. Fanale, A. Baddoo, P. Manowitz, A novel arylsulfatase A protain variant and genotype in two patients with major depression, J. Affect. Disord. 40 (1996) 137–147. [34] S. Emre, M. Topcu, M. Terziolu, Y. Renda, Arylsulfatase A pseudodeficiency incidence in Turkey, Turk. J. Pediatr. 42 (2000) 115–117. [35] W. Krivit, P. Aubourg, E. Shapiro, C. Peters, Bone marrow transplantation for globoid cell leukodystrophy, adrenoleukodystrophy, metachromatic leukodystrophy, and Hurler syndrome, Curr. Opin. Hematol. 6 (1999) 377–382.
Disease-causing mutations in cis with the common arylsulfatase A pseudodeficiency allele compound the difficulties in accurately identifying patients and carriers of metachromatic leukodystrophy Mohammad A. Rafi, Stephanie Coppola, Shu Ling Liu, Han Zhi Rao, and David A. Wenger* Lysosomal Diseases Testing Laboratory, Department of Neurology, Jefferson Medical College, 1020 Locust St., Room 394, Philadelphia, PA 19107, USA Received 28 February 2003; received in revised form 14 April 2003; accepted 16 April 2003
Abstract Metachromatic leukodystrophy (MLD) is a lysosomal storage disorder most often caused by mutations in the sulfatide sulfatase or arylsulfatase A (ASA) gene. This results in the storage of sulfatides in the peripheral and central nervous systems as well as in the kidneys. Patients with MLD exhibit a wide range of clinical features presenting from the late infantile period to adulthood. Testing for this disease is performed on a majority of the patient samples received for diagnostic testing in the authorsÕ laboratory. If low ASA activity is measured, additional testing is required to confirm the diagnosis due to several factors. ASA activity is also low in individuals with multiple sulfatase deficiency and in individuals with copies of the so-called pseudodeficiency (Pd) allele. Due to the high frequency of the common Pd allele, it is possible for individuals, both with and without neurologic disease, to have low ASA activity but not have MLD. Unfortunately, the finding of the Pd mutation by molecular analysis does not rule out a diagnosis of MLD. In a recent 25 month period, this laboratory diagnosed 52 patients with MLD, and of these, 13 patients from 10 families had one or two copies of the Pd mutation. Sequencing of the ASA gene in these 10 families revealed four new mutations in cis with the Pd allele (S43R, R84Q, R311X, and E329R) and two additional new mutations (R299W, C488R). Six patients had previously reported mutations on the Pd background. Thus, a total of 14 mutations have been found to occur in cis with the Pd allele. We estimate that 1–2% of Pd alleles will have a disease-causing mutation, and this complicates the identification of patients and the assignment of risk for a couple when a copy of the Pd allele is detected. Ó 2003 Elsevier Science (USA). All rights reserved. Keywords: Arylsulfatase A; Genotype–phenotype correlations; Lysosomal storage disease; Metachromatic leukodystrophy; Polymorphisms; Pseudodeficiency; Sulfatide
Introduction Metachromatic leukodystrophy (MLD) is an autosomal recessive lysosomal storage disorder that results from the accumulation of sulfatides in the central and peripheral nervous systems (reviewed in [1]). Patients exhibit a wide range of presenting symptoms, and their clinical courses can vary significantly. Late infantile,
* Corresponding author. Fax: 1-215-955-9554. E-mail address: [email protected] (D.A. Wenger).
juvenile and adult onset forms are generally recognized. MLD is usually caused by a deficiency of arylsulfatase A (ASA; sulfatide sulfatase; E.C. 3.1.6.1) activity which can be measured in several easily obtained tissue samples, including leukocytes isolated from whole blood or cultured skin fibroblasts [2]. In addition, some rare individuals store and excrete sulfatides due to a deficiency of saposin B, the required activator protein that facilitates the interaction between ASA and sulfatide [3]. The diagnosis of MLD is also complicated by several other factors. The high frequency of the so-called pseudodeficiency (Pd) allele results in low ASA activity
1096-7192/03/$ - see front matter Ó 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S1096-7192(03)00076-3
84
M.A. Rafi et al. / Molecular Genetics and Metabolism 79 (2003) 83–90
in individuals who do not have MLD and in carrier levels of ASA activity in individuals who are not carriers of MLD. This is due to a mutation in the polyadenylation signal that results in only about 10% of ASA messenger RNA being produced [4]. However, the finding of the Pd allele by molecular analysis does not rule out a diagnosis of MLD since there can be additional mutations on the Pd background [1,5–10]. In addition, patients with multiple sulfatase deficiency (MSD) will have low ASA activity as well as low levels of other sulfatases. These patients usually, but not always, have clinical features of both a mucopolysaccharidosis and a leukodystrophy. Because of the wide clinical spectrum observed in patients with MLD, ASA activity is measured in many individuals whose tissue samples are received in the Lysosomal Diseases Testing Laboratory. Since 1974 this laboratory has identified over 480 patients with MLD. In the 25 month period from November 2000 until December 2002, we diagnosed 52 individuals, including affected siblings, with MLD based on a combination of clinical, biochemical and magnetic resonance imaging (MRI) studies. With the cloning of the ASA cDNA [11] and gene [12], a large number of MLD-causing mutations and polymorphisms have been delineated (reviewed in [1]). In addition, the major cause of Pd has been determined [4]. While we normally do not attempt to identify all mutations in patients with low ASA activity, we do check for the mutation in the polyadenylation signal that results in Pd [4] and the common mutations associated with late infantile and adult MLD [13]. These mutations, G459 + 1A and P426L, were termed the ‘‘I’’ and ‘‘A’’ alleles, respectively, by Polten et al. [13]; however, it is obvious from this study and others that this designation only holds for patients homozygous for these mutations. In addition, urinary sulfatide excretion is examined in all individuals with low ASA activity. Thirteen patients, representing 10 families, diagnosed in the 25 month period of study were either heterozygous or homozygous for the common Pd mutation as determined by polymerase chain reaction (PCR)-based analysis and sequencing. While some MLD-causing mutations have previously been reported in cis with the common Pd mutation [1,5–10], it became obvious that four of the recent cases of MLD identified in this laboratory who had the Pd allele did not have any of the previously reported mutations. These individuals ranged clinically from late infantile to adult onset. A few of the patients who were heterozygous for the Pd allele were also heterozygous for either the common G459 + 1A or P426L mutation. Four new MLD-causing mutations on the Pd background were identified. In addition, two unreported disease-causing mutations not on the Pd allele were also found in these 10 families. Since about 1 in 7 individuals in the general population is heterozygous for the common Pd allele, these findings
have implications for patient and carrier identification within families and for individuals who marry into such a family.
Materials and methods Patients heterozygous or homozygous for the Pd allele included in this study Samples from the patients used in this study were sent here in an attempt to obtain a diagnosis to explain their neurologic and/or psychiatric symptoms. Initially, a panel of lysosomal enzyme activities, including ASA, was measured in sonicated leukocytes isolated from whole heparinized blood [2,14]. Since the ASA activity was within the range of patients previously confirmed to have MLD and other enzymes, including other sulfatases, were normal, a diagnosis of MLD was suspected. All patients with low ASA activity were initially tested for the Pd allele to aid in confirming the diagnosis and to determine if the low ASA activity could be explained by the presence of the Pd mutation. In addition, a first morning voiding of urine was requested for sulfatide extraction. A brief summary of the clinical and biochemical findings on the individuals studied is presented in Table 1. These patients ranged in age from 18 months to 48 years. While clinical information is requested with each specimen, the amount of information supplied varies greatly. While many of the younger patients had the typical features of late infantile MLD, the older patients tended to have motor and cognitive decline and psychiatric problems leading to dementia. When performed, MRI changes were consistent with a diagnosis of MLD. All had ASA activities in leukocytes within the range of patients with MLD, and excess sulfatides were clearly detected in urinary extracts (Fig. 1). While most of the original samples came from clinically affected individuals, younger asymptomatic siblings may be tested because of the diagnosis of MLD in the proband. It is critical that the parents of the proband be tested before or together with the siblings to prevent inaccurate interpretation of the results in the ‘‘normal’’ sibling. If a healthy parent is ‘‘pseudodeficient’’ (MLD/ Pd) and has not been tested, the measurement of low ASA activity in the sibling can result in an incorrect diagnosis. Urinary sulfatide excretion Ten ml of first morning voiding of urine was brought to pH 5.0 with a few drops of glacial acetic acid and placed at 4 °C overnight. The urine was then centrifuged at 1500g for 10 min. The precipitate was dissolved in 1 ml of distilled water, and 5 ml of chloroform–methanol (2–1, by vol) was added. After vortexing for 1 min, the
M.A. Rafi et al. / Molecular Genetics and Metabolism 79 (2003) 83–90
85
Table 1 Clinical and biochemical findings on patients heterozygous or homozygous for the Pd allele included in this study Family: patienta (Year of Birth)
Age of onset
Signs and symptoms
ASA activityb
Sulfatide excretionc
1: 1 (2000) 2: 1 (1999)
18 months 1–2 years
DD, hypotonia, esotropia, PR, and CSF protein 123 Clumsiness noted at 1 year with normal developmental milestones until about 24 months, PR obvious by 27 months, MRI c/w leukodystrophy None at time of diagnosis
4.2 10.0
+ ND
7.6
+
DD, spasticity, nystagmus, PR, and MRI c/w leukodystrophy PR, genu recurvatum, hypertonia, and NCV c/w leukodystrophy DD, nystagmus, spasticity, weakness, CSF protein 143, MRI c/w leukodystrophy, and decreased speech
3.1
+
3.7
+
4.6
+
PR, NCV and MRI c/w leukodystrophy, and CSF protein 79 Cognitive decline, psychiatric problems, and MRI c/w leukodystrophy None at time of diagnosis
12.3
+
4.6
+
0
+
N/A Cognitive decline, psychiatric problems, and CSF protein 150, MRI c/w leukodystrophy Cognitive decline, psychiatric problems, dementia, and NCV and MRI c/w leukodystrophy None at time of diagnosis
2.8 6.4
+ +
5.2
+
5.9
+
2 (2000) 3: 1 (1999)
Asymptomatic at 18 months 24 months
4: 1 (1999)
12–18 months
5: 1 (1999)
30 months
6: 1 (1994)
6 years
7: 1 (1991)
10 years
2 (1992) 8: 1 (1970) 9: 1 (1955)
Asymptomatic at 10 years N/A 42 years
10: 1 (1955)
42 years
2 (1960)
Asymptomatic at 40 years
Comments
African American/ Caucasian
Consanguineous parents from Bangladesh
Died following HSCT
Abbreviations: CSF, cerebrospinal fluid; c/w, consistent with; DD, developmental delay; HSCT, hematopoietic stem cell transplantation; N/A, not available; NCV, nerve conduction velocity; ND, not done; PR, psychomotor regression. a Multiple patients listed as one family are siblings. b nmol/h/mg protein measured in leukocytes. Normal mean standard deviation, 71.1 11.3. c ‘‘+’’ indicates the excretion of excess sulfatides clearly visible on the thin layer chromatography plate.
Fig. 1. Detail of the thin layer chromatography plate showing the sulfatide excretion from patients with MLD and a control. All samples were extracted as described in Materials and methods. The amount spotted was equivalent to 5 mg creatinine, except lane 2 which was equivalent to 1 mg creatinine. Lane 1, patient #1; lane 2, patient with late infantile MLD and no copies of the Pd allele; lane 3, patient #6; lane 4, patient #7; lane 5, sulfatide standard; lane 6, control; lane 7, patient #8; lane 8, patient #10.
mixture was centrifuged at 1500g for 5 min. The upper phase was removed and discarded, and the lower phase was evaporated with nitrogen. The residue was dissolved in 0.05 ml of chloroform–methanol (2–1, by vol), and half was spotted on a silica gel thin layer chromatography plate along with standard sulfatide. The plate was placed in a tank containing chloroform–methanol–water
(72–28–3.5, by vol), and the solvent was allowed to run until it was near the top of the plate. After drying, the plate was sprayed with orcinol reagent to detect sugarcontaining lipids. The sulfatide present in urine was not quantitated, but little, if any, sulfatide was excreted in individuals with no mutations in the ASA gene, individuals who are carriers of MLD-causing mutations or
86
M.A. Rafi et al. / Molecular Genetics and Metabolism 79 (2003) 83–90
the Pd mutation, individuals who are homozygous for the Pd mutation (without a disease-causing mutation), and compound heterozygotes who have one MLDcausing mutation and one Pd mutation (without an additional mutation). Attempts to equalize the amount spotted based on the creatinine concentration or the level of a lysosomal enzyme, such as b-hexosaminidase, were not entirely successful. If excess sulfatide is being excreted and other evidence indicates a leukodystrophy, MLD is almost surely the diagnosis. If the activities of other sulfatases are also low, a diagnosis of MSD must be considered. Molecular analysis of patients Molecular analysis was performed using DNA isolated from the remaining leukocyte sonicates [15], or an additional blood sample was requested, if necessary. Using the method of Chabas et al. [16], all samples were first tested for the mutation in the polyadenylation signal that results in lower than normal levels of ASA messenger RNA, the common cause of Pd [4]. This method will not detect the N350S polymorphism that is commonly found in cis with the polyadenylation signal mutation. The N350S mutation will eliminate a utilized N-glycosylation site but does not lower ASA activity. Initially, only patients with late infantile or juvenile onset were tested for the common G459 + 1A mutation reported previously [13] using the method of Ben-Yoseph and Mitchell [17]. However, after observing the results of the DNA sequencing, testing for this mutation was extended to patients of any age. If the patient had juvenile or adult onset, we tested for the common C1277T (P426L) mutation using the method of BenYoseph and Mitchell [17]. If the patients were considered for further analysis, the complete ASA gene was amplified by PCR using the GeneAmp XL PCR kit (Perkin Elmer, Norwalk, CT). An upstream primer, ASAa+ (50 -TATTTGGGTCCGGGGTCTCAG-30 ), located before the initiation codon and a downstream primer, ASAb) (50 -TCTTTCTGGCTGGGTGATCTT G-30 ), located after the first polyadenylation signal were used to produce a 2983 bp DNA fragment. This product was reamplified and purified using QIAquick PCR purification kit (Qiagen, Valencia, CA). Full length original PCR products were cloned in TA vector (Invitrogen, Carlsbad, CA) and transfected into TOP 10 competent Escherichia coli. Positive clones were grown in kanamycin containing Luria–Bertani medium, and DNA was extracted using QIAprep kit (Qiagen) for sequencing. Purified PCR products were either sequenced directly or after cloning in TA vector using eight internal primers (G11, 50 -AAGAGCCTGCTGAGGCCAAG-30 ; G13, 50 -CCACATTTCTGTCTGTCTCAG-30 ; G12, 50 CCAGCCCTCACGGCAGCTG-30 ; G3, 50 -CCATGGC TCATGAGCGCCTC-30 ; G31, 50 -AACTGAGTGACT
GACCAG-30 ; G4, 50 -CCAGGTATGTGCAGTGCTT G-30 ; G41, 50 -TTGCCCTGTGCACAGAATTG-30 ; and G43, 50 -TTGCTGCCATTGCCCAGAT-30 ). Cycle Sequencing kit (USB Corporation, Cleveland, OH) and/or ABI Prism 377 DNA automated sequencer with BigDye terminator reaction (Applied Biosystems, Foster City, CA) were employed. The sequences generated from the patients and carriers were compared to the ASA sequences obtained from the National Center of Biotechnology Information (NCBI) (Accession Nos. X52150, X52151). When a patient was determined to be heterozygous for the Pd allele, assignment of additional mutations to the correct allele was based either on published data from known mutations [5,8,9,13,18] or on studies of obligate carrier parents, siblings or offspring sent here for testing. As these individuals requested genetic testing, their DNA was sequenced using similar methods once the presence or absence of the Pd mutation was determined by the simple PCR-based method [16].
Results The results of the mutation analysis in these 10 families are shown in Table 2. Patient # refers to the family # since affected siblings had the identical genotype. Two patients were found to be homozygous for the Pd mutation in the polyadenylation signal, and eight were found to be heterozygous (Fig. 2). The common G459 + 1A and P426L mutations were identified in five individuals using the published methods [17]. The C287T (S96F) mutation, previously reported to occur on the Pd background [5], was found on 4 of 20 alleles. This included two late infantile patients (#1, #3) and one juvenile patient (#7). Of the two late infantile patients, one was homozygous for this mutation (#1), and one was also heterozygous for the common G459 + 1A mutation (#3). The juvenile patient was found to be heterozygous for T536G (I179S), a previously reported mutation that is considered mild [18]. One late infantile patient (#4) was found to be heterozygous for 1216del9 (2320del9), another mutation known to be on the Pd background [8], and the common G459 + 1A mutation. Patient #2, one of two affected siblings with late infantile MLD, was found to be heterozygous for the Pd allele and heterozygous for G985A (E329R) and C895T (R299W), two previously unreported mutations. Assignment of the E329R mutation to the Pd allele was accomplished by studies on the mother who also was heterozygous for the Pd allele. Patient #5 was found to be homozygous for the Pd allele and homozygous for C931T (R311X), an unreported mutation. The nine-year-old patient #6 was heterozygous for the Pd allele and heterozygous for two additional mutations, G251A (R84Q) and the common
M.A. Rafi et al. / Molecular Genetics and Metabolism 79 (2003) 83–90
87
Table 2 Summary of mutations in the 10 families with MLD and the Pd Allele Family: allele
Mutation position
Pd allele
Effect
Comments/references
cDNAa
Geneb
1 2
C287T C287T
436 436
+ +
S96F S96F
[5] [5]
1 2
G985A C895T
1724 1544
+ )
E329R R299W
New New
1 2
C287T G459 + 1A
436 609
+ )
S96F Splicing error
[5] [13]
1 2
1216del9 G459 + 1A
2320del9 609
+ )
3 amino acid del Splicing error
[8] [13]
1 2
C931T C931T
1580 1580
+ +
R311X R311X
New New
1
G251A
400
+
R84Q
2
C1277T
2381
)
P426L
Not previously reported on Pd allele [19] [13]
1 2
C287T T536G
436 799
+ )
S96F I179S
[5] [18]
1 2
G863A G459 + 1A
1512 609
+ )
R288H Splicing error
[9] [13]
1 2
G863A G459 + 1A
1512 609
+ )
R288H Splicing error
[9] [13]
C129A T1462C
129 2566
+ )
S43R C488R
New New
1:
2:
3:
4:
5:
6:
7:
8:
9:
10: 1 2
Nucleotide number 1 is the first base of the initiation codon. a Positions numbered according to NCBI entry X52151. b Position numbered according to NCBI entry X52150.
Fig. 2. Results of the PCR amplification of genomic DNA and digestion with RsaI to detect the polyadenylation signal mutation causing Pd. Lanes 1 and 14, 100 bp DNA ladder; lane 2, undigested PCR product; lane 3, MLD patient with no copies of Pd allele; lane 4, patient #1; lane 5, patient #2; lane 6, patient #3; lane 7, patient #4; lane 8, patient #5; lane 9, patient #6; lane 10, patient #7; lane 11, patient #8; lane 12, patient #9; lane 13, patient #10.
P426L mutation. R84Q was previously reported in lateonset patients [19]; however, it was not noted to occur on the Pd background. Studies on family members confirmed that this mutation occurred in cis with the Pd
mutation. Patients #8 and #9 are unrelated adult patients of Italian ancestry who share the same genotype. Both were found to be heterozygous for G863A (R288H), a previously reported mutation on the Pd
88
M.A. Rafi et al. / Molecular Genetics and Metabolism 79 (2003) 83–90
Table 3 Summary of all mutations found in cis with the Pd allele Number
Mutation position cDNAa
1: 2: 3: 4: 5: 6: 7: 8: 9: 10: 11: 12: 13: 14:
C129A G251A C287T C406T C464G G863A C931T G973C G985A C1130T G1144A 1192del3 1216del9
Effect
Comments/references
S43R R84Q S96F P136S P155R D281Y R288H R311X G325C E329R P377L E382K DF398 3 amino acid del
New Not previously reported on Pd allele [19] [5] [9] Lebanese [1] [10], nucleotide change not given [9] New Indian [1], loss of splice donor site New [7] [6] Polynesian [1] [8]
Geneb 129 400 436 555 727 1512 1580 1622 1724 2119 2133 2181del3 2320del9
Nucleotide number 1 is the first base of the initiation codon. Positions numbered according to NCBI entry X52151. b Position numbered according to NCBI entry X52150. a
allele [9], and the common G459 + 1A mutation found in many late infantile and some juvenile patients with MLD. Patient #10 was one of two affected siblings who had onset of symptoms, mainly psychiatric in nature, in the fifth decade. The patient was heterozygous for the Pd allele and heterozygous for two additional new mutations, C129A (S43R) and T1462C (C488R). The S43R mutation has been assigned to the Pd allele by studies in family members requesting carrier testing.
Discussion MLD is one of the most commonly tested for lysosomal storage disease. However, accurate diagnosis remains a serious problem due to the factors laid out in the Introduction, especially the high frequency of the Pd allele (with and without disease-causing mutations on the same allele) in the general population and in patients with neurologic disease. The difficulties related to ‘‘pseudodeficiency’’ of ASA activity and other lysosomal hydrolases have been raised previously [20,21]. Leukocyte ASA activities measured in the many samples received here for diagnostic studies break down into three categories with some overlap. Those individuals with ASA values between 50 and 120 nmol/h/mg protein most likely have two ASA genes without any mutations (Nl/ Nl). The activity range results from variable conditions during shipping of the samples, white cell differential, effects of other genes and other unknown factors. Those individuals with values between 25 and 45 nmol/h/mg protein are either carriers of MLD (MLD/Nl, PdMLD / Nl) or carriers of the Pd mutation (without an additional mutation on the Pd allele) (Pd/Nl). PdMLD denotes the Pd allele with an additional disease-causing
mutation. ASA values between 0 and 15 nmol/h/mg protein will be measured in individuals who have MLD (MLD/MLD, PdMLD /PdMLD , and MLD/PdMLD ) and in people who do not have MLD (MLD/Pd, Pd/Pd, Pd/ PdMLD , and MSD). All of the patients used in this study were confirmed to have MLD by a number of criteria, including clinical features, MRI findings, biochemical studies and mutation analysis. The patients ranged in age from 18 months to 48 years. Among the 10 families heterozygous or homozygous for the Pd allele, seven different mutations on the Pd background were detected. Three previously reported mutations on the Pd background were found in six patients. Four new mutations in cis with the polyadenylation signal mutation and two additional new mutations were identified. Five mutations on the Pd allele were missense, one was a nonsense mutation resulting in a stop at codon 311 and one caused a splicing error. The new missense mutations were found in exons 1, 2, 5, and 6, and the nonsense mutation occurred in exon 5. A total listing of the 14 mutations occurring on the Pd background is presented in Table 3. The other two new mutations were missense mutations occurring in exons 5 and 8. Transient expression studies were not performed on the newly identified mutations since those present on the Pd background would obviously show low ASA activity, and upon sequencing the entire ASA gene, only these mutations were found. The finding of disease causing-mutations on a polymorphic background, such as the Pd allele, is not unique to MLD. It is nearly impossible to speculate what the effect would be of the same mutation on the ‘‘normal’’ background. It is clear that the contribution of the other allele may be critical to producing a phenotype. It is possible that the same mutation on the ‘‘normal’’
M.A. Rafi et al. / Molecular Genetics and Metabolism 79 (2003) 83–90
background, when occurring with a mild mutation on the other allele, would not produce clinical disease but would cause disease if the second allele provided no residual ASA activity. This may be the case in patients #8 and #9 who are heterozygous for the G893A mutation on the Pd allele and the common G459 + 1A mutation which is considered severe. While the G459 + 1A mutation is most commonly found in late infantile and juvenile patients, it has previously been described in two additional adult patients, including one with the identical genotype to patients #8 and #9 [9] and one with a known mild mutation, I179S [22]. Therefore, we have extended the testing for the G459 + 1A mutation to patients of all ages. While some publications have stated that genotype–phenotype predictions for certain mutations causing MLD will be excellent, others have hinted that it is not straight-forward [23]. There have been a number of papers describing patients with a confirmed diagnosis of MLD or with neurologic or psychiatric symptoms who have ‘‘pseudodeficiency;’’ however, most of these patients were not analyzed for the presence of additional mutations on the Pd allele [22,24–34]. Therefore, there is confusion about whether low ASA activity caused by the presence of two Pd alleles (without additional mutations) or one Pd allele with an MLD-causing mutation on the second allele would result in clinical disease. Without searching for additional mutations on the Pd background, it is difficult to interpret those studies. Since the carrier rate for the Pd allele is about 1 in 7, about 1 in 14 parents of MLD patients would have a Pd allele in addition to an MLD-causing allele. A review of ASA values in hundreds of obligate carriers confirms this estimate. While we do not have followup on many of these carriers with low ASA activity, we have never heard of any of these individuals having symptoms that can be assigned to their low ASA activity. Due to the high frequency of the Pd allele in almost all ethnic groups, it is now clear that additional mutations can occur on this background (Table 3). Thomas [21] speculated that almost a fifth of all mutations causing MLD may be found on the Pd background. Of the approximately 100 alleles characterized from the patients diagnosed during the 25 month period, 13 disease-causing mutations were found in cis with the Pd allele. The identification of 14 mutations on the Pd background (Table 3) complicates the diagnosis of patients and the assignment of carrier status when molecular analysis detects the Pd allele. While most copies of the Pd allele do not have an additional mutation, we estimate that there is approximately a 1–2% chance that a disease causing-mutation may be present in cis with the Pd allele. This compares to the estimated carrier rate of 1 in 100 for MLD in the general population. The incidence of MLD is estimated to be approximately 1 in 40,000 births in the United States. Accurate
89
prenatal diagnosis of MLD for couples truly at risk or inconclusive because of the presence of the Pd allele can be accomplished by a combination of biochemical and molecular tests performed on chorionic villus samples or cultured amniotic fluid cells. In order to prevent problems with interpretation, it is ideal to have studies performed on the parents before analyzing fetal samples. At this time, the only treatment available is hematopoietic stem cell transplantation for those individuals who are pre-symptomatic or have only mild neurologic manifestations [35]. Although there are significant risks associated with the procedure, this treatment appears to slow the progression of the disease and improve the quality of life for the patient. Hopefully additional methods of treatment will be available in the near future.
Acknowledgments This research was supported in part by a grant from NIH (2 R01 DK38795). The authors thank the physicians who supplied pertinent clinical information with the samples from their patients.
References [1] K. von Figura, V. Gieselmann, J. Jaeken, Metachromatic leukodystrophy, in: C.R. Scriver, A.L. Beaudet, W.S. Sly, D. Valle (Eds.), The Metabolic and Molecular Bases of Inherited Disease, McGraw-Hill, New York, 2001, pp. 3695–3724. [2] D.A. Wenger, C. Williams, Screening for lysosomal disorders, in: F.A. Hommes (Ed.), Techniques in Diagnostic Human Biochemical Genetics. A Laboratory Manual, Wiley-Liss, New York, 1991, pp. 587–617. [3] K. Inui, M. Emmett, D.A. Wenger, Immunological evidence for a deficiency in an activator protein for sulfatide sulfatase in a variant form of metachromatic leukodystrophy, Proc. Natl. Acad. Sci. USA 80 (1983) 3074–3077. [4] V. Gieselmann, A. Polten, J. Kreysing, K. von Figura, Arylsulfatase A pseudodeficiency: loss of a polyadenylation signal and a N-glycosylation site, Proc. Natl. Acad. Sci. USA 86 (1989) 9436– 9440. [5] V. Gieselmann, A.L. Fluharty, T. Tonnesen, K. von Figura, Mutations in the arylsulfatase A pseudodeficiency allele causing metachromatic leukodystrophy, Am. J. Hum. Genet. 49 (1991) 407–413. [6] M.L. Barth, A. Fensom, A. Harris, Missense mutations in the arylsulphatase A genes of metachromatic leukodystrophy patients, Hum. Molec. Genet. 2 (1993) 2117–2121. [7] J. Zlotogora, G. Bach, C. Bosenberg, Y. Barak, K. von Figura, V. Gieselmann, Molecular basis of late infantile metachromatic leukodystrophy in the Habbanite jews, Hum. Mutat. 5 (1995) 137–143. [8] S. Regis, M. Filocamo, M. Stroppiano, F. Corsolini, F. Caroli, R. Gatti, A 9-bp deletion (2320del9) on the background of the arylsulfatase A pseudodeficiency allele in a metachromatic leukodystrophy patient and in a patient with nonprogressive neurological symptoms, Hum. Genet. 102 (1998) 50–53. [9] L. Gort, M.J. Coll, A. Chabas, Identification of 12 novel mutations and two new polymorphisms in the arylsulfatase A gene: haplotype and genotype–phenotype correlation studies in
90
[10]
[11]
[12] [13]
[14] [15]
[16]
[17]
[18]
[19]
[20] [21] [22]
[23]
M.A. Rafi et al. / Molecular Genetics and Metabolism 79 (2003) 83–90 Spanish metachromatic leukodystrophy patients, Hum. Mutat. 14 (1999) 240–248. D.J. Halsall, E.P. Halligan, T.S. Elsey, T.M. Cox, Metachromatic leukodystrophy: a newly identified mutation in arylsulfatase A, D281Y, found as a compound heterozygote with I179L in an adult onset case, Hum. Mutat. 14 (1999) 447. C. Stein, V. Gieselmann, J. Kreysing, B. Schmidt, R. Pohlmann, A. Waheed, H.E. Meyer, J.S. OÕBrien, K. von Figura, Cloning and expression of human arylsulfatase A, J. Biol. Chem. 264 (1989) 1252–1259. J. Kreysing, K. von Figura, V. Gieselmann, Structure of the arylsulfatase A gene, Eur. J. Biochem. 191 (1990) 627–631. A. Polten, A.L. Fluharty, C.B. Fluharty, J. Kappler, K. von Figura, V. Gieselmann, Molecular basis of different forms of metachromatic leukodystrophy, N. Engl. J. Med. 324 (1991) 18–22. H. Baum, K.S. Dodgson, B. Spencer, The assay of arylsulfatases A and B in human urine, Clin. Chim. Acta 4 (1959) 453–455. E. Louie, M.A. Rafi, D.A. Wenger, Leukocyte sonicates as a source for both enzyme assay and DNA amplification for mutation for analysis of certain lysosomal disorders, Clin. Chim. Acta 199 (1991) 7–16. A. Chabas, S. Castellvi, M. Bayes, S. Balcells, D. Grinberg, L.I. Vilageliu, G. Marfany, W. Lissens, R. Gonzalez-Duarte, Frequency of the arylsulphatase A pseudodeficiency allele in the Spanish population, Clin. Genet. 44 (1993) 320–323. Y. Ben-Yoseph, D.A. Mitchell, Rapid detection of common metachromatic leukodystrophy mutations by restriction analysis of arylsulfatase A gene amplimers, Clin. Chim. Acta 226 (1994) 77–82. A.L. Fluharty, C.B. Fluharty, W. Bohne, K. von Figura, V. Gieselmann, Two new arylsulfatase A (ARSA) mutations in a juvenile metachromatic leukodystrophy (MLD) patient, Am. J. Hum. Genet. 49 (1991) 1340–1350. J. Kappler, K. von Figura, V. Gieselmann, Late-onset metachromatic leukodystrophy: molecular pathology in two siblings, Ann. Neurol. 31 (1992) 256–261. D.A. Wenger, E. Louie, Pseudodeficiencies of arylsulfatase A and galactocerebrosidase activities, Dev. Neurosci. 13 (1991) 216–221. G. Thomas, ‘‘Pseudodeficiencies’’ of lysosomal hydrolases, Am. J. Hum. Genet. 54 (1994) 934–940. J. Berger, B. Loschl, H. Berheimer, A. Lugowska, A. TylkiSzymanska, V. Gieselmann, B. Molzer, Occurrence, distribution, and phenotype of arylsulfatase A mutations in patients with metachromatic leukodystrophy, Am. J. Med. Genet. 69 (1997) 335–340. M.L. Barth, C. Ward, A. Harris, A. Saad, A. Fensom, Frequency of arylsulphatase A pseudodeficiency associated mutations in a healthy population, J. Med. Genet. 31 (1994) 667–671.
[24] J. Peiffer, K. Harzer, W. Schlote, Diffuse-disseminated sclerosis combined with partial arylsulfatase A (ASA) deficiency. Mixed heterozygosity of ASA- and pseudo-ASA-deficiency? Neuropediatrics 15 (1984) 59–62. [25] K. Farrell, D.A. Applegarth, J.R. Toone, P.M. McLeod, A.V. Savage, Pseudoarylsulfatase-A deficiency in the neurologically impaired patient, Can. J. Neurol. Sci. 12 (1985) 274–277. [26] C. Hohenschutz, W. Friedl, K.H. Schlor, A. Waheed, E. Conzelmann, K. Sandhoff, P. Propping, Probable metachromatic leukodystrophy/pseudodeficiency compound heterozygote at the arylsulfatase A locus with neurological and psychiatric symptomatology, Am. J. Med. Genet. 31 (1988) 169–175. [27] C. Hohenschutz, P. Eich, W. Friedl, A. Waheed, E. Conzelmann, P. Propping, Pseudodeficiency of arylsulfatase A: a common polymorphism with possible disease implications, Hum. Genet. 82 (1989) 45–48. [28] S. Sangiorgi, A. Ferlini, A. Zanetti, M. Mochi, Reduced activity of arylsulfatase A and predisposition to neurological disorders: analysis of 140 pediatric patients, Am. J. Med. Genet. 40 (1991) 365–369. [29] J. Kappler, P. Leinekugel, E. Conzelmann, W.J. Kleijer, A. Kohlschutter, T. Tonnesen, M. Rochel, F. Freycon, P. Propping, Genotype–phenotype relationship in various degrees of arylsulfatase A deficiency, Hum. Genet. 86 (1991) 463–470. [30] M.L. Barth, A. Fensom, A. Harris, Prevalence of common mutations in the arylsulphatase A gene in metachromatic leukodystrophy patients diagnosed in Britain, Hum. Genet. 91 (1993) 73–77. [31] J.M. Penzien, J. Kappler, N. Herschkowitz, B. Schuknecht, P. Leinekugel, P. Propping, T. Tonnesen, H. Lou, H. Moser, S. Zierz, E. Conzelmann, V. Gieselmann, Compound heterozygosity for metachromatic leukodystrophy and arylsulfatase A pseudodeficiency alleles is not associated with progressive neurological disease, Am. J. Hum. Genet. 52 (1993) 557–564. [32] S. Leistner, E. Young, C. Meaney, B. Winchester, Pseudodeficiency of arylsulfatase A: strategy for clarification of genotype in families of subjects with low ASA activity and neurological symptoms, J. Inherit. Metab. Dis. 18 (1995) 710–716. [33] M.H. Ricketts, J.D. Amsterdam, D.S. Parks, R.-S. Yang, R.D. Poretz, X. Zhang, M. Fanale, A. Baddoo, P. Manowitz, A novel arylsulfatase A protain variant and genotype in two patients with major depression, J. Affect. Disord. 40 (1996) 137–147. [34] S. Emre, M. Topcu, M. Terziolu, Y. Renda, Arylsulfatase A pseudodeficiency incidence in Turkey, Turk. J. Pediatr. 42 (2000) 115–117. [35] W. Krivit, P. Aubourg, E. Shapiro, C. Peters, Bone marrow transplantation for globoid cell leukodystrophy, adrenoleukodystrophy, metachromatic leukodystrophy, and Hurler syndrome, Curr. Opin. Hematol. 6 (1999) 377–382.