© 1994 Oxford University Press
Human Molecular Genetics, 1994, Vol. 3, No. 6 861-866
Mucopolysaccharidosis type I: identification of 8 novel
mutations and determination of the frequency of the
two common a-L-iduronidase mutations (W402X and
Q70X) among European patients
Susanna Bunge*, Wim J.KIeijer1, Cordula Steglich, Michael Beck2, Cornelia Zuther, C.Phillip Morris3,
Eberhard Schwinger, John J.Hopwood3, Hamish S.Scott3 and Andreas Gal
Institut fur Humangenetik, Medizinische Universitat, Ratzeburger Allee 160, D-23538 Lubeck, Germany, department of Clinical Genetics, Erasmus
University, NL-3000 DR Rotterdam, The Netherlands, 2Kinderklinik der Universitat, D-55101 Mainz, Germany and 3Lysosomal Diseases Research
Unit, Department of Chemical Pathology, Adelaide Children's Hospital, North Adelaide, Australia
Received December 20, 1993; Revised and Accepted February 28, 1994
A group of 46 European patients with mucopoly-
saccharidosis type I (MPS I) was screened for mutations
of the a-L-iduronidase gene. The 2 common nonsense
mutations, W402X and Q70X, were identified in,
respectively, 37% and 35% of mutant alleles. Consider-
able differences were seen in the frequency of these
2 mutations in patients from North Europe (Norway and
Finland) and other European countries (mainly The
Netherlands and Germany). In Scandinavia, W402X and
Q70X account for 17% and 62% of the MPS I alleles,
respectively, while in other European countries W402X
is about 2.5 times more frequent (48%) than Q70X
(19%). Eight novel mutations are described including
4 missense mutations, 1 nonsense mutation, 1 inser-
tion of 2 base pairs, and 2 deletions of 1 and 12 base
Mucopolysaccharidosis type I (MPS I) is an autosomal recessive
condition caused by the deficiency of a-L-iduronidase (IDUA,
E.C. 18.104.22.168), an enzyme involved in the stepwise degradation
of glycosaminoglycans, dermatan and heparan sulfates in
lysosomes. Accumulation of partially degraded mucopoly-
saccharides leads to typical symptoms of a lysosomal storage
disorder including skeletal deformities, hepatosplenomegaly,
corneal clouding, and often mental retardation. Clinical pheno-
types may vary considerably so that originally 3 different forms,
severe (MPS IH, Hurler), mild (MPS IS, Scheie), and inter-
mediate (MPS IH/S, Hurler/Scheie) were defined. Patients show
greatly reduced IDUA activity in each of the three MPS I forms
and residual enzyme activity correlates only to a limited extent
with clinical severity (1). The phenotypic heterogeneity is thought
to be due to different (allelic) mutations of the iduronidase gene.
This suggestion was supported by the lack of genetic comple-
mentation of IDUA deficiency in heterokaryons obtained by
pairwise cell fusion of fibroblasts from patients with either of
the 3 different MPS I forms (2). Thus characterization of gene
mutations is an important prerequisite to analyse genotype/
phenotype correlation and select patients for therapy by such
approaches as bone marrow transplantation (3).
Both the cDNA encoding IDUA (4) and the exon-intron
structure of the iduronidase gene (5) have recently been
characterized. To date, two common nonsense point mutations
have been identified in a group of 71 MPS I patients, mainly
of Anglo-Saxon origin; W402X and Q70X were found in,
respectively, 31% and 15% of MPS I alleles (6,7). Three other
mutations, P533R, A75T, and 474-2a—g were present in,
respectively, 3%, 3.9% and 3% of mutant alleles (7,8). Several
rare mutations have also been identified (9 — 14).
We describe here the results of a molecular analysis of the
IDUA gene in a group of 47 patients with MPS I, 46 from
different European countries, and one from Egypt. All patients
had a typical, severe form of the disease (MPS IH) with marked
skeletal changes, hepatosplenomegaly, and mental retardation.
Pathogenic mutations, nonpathogenic sequence variants and
polymorphisms of the IDUA gene
Six overlapping PCR fragments of the IDUA cDNA, containing
the entire coding sequence, were screened for sequence alterations
by single strand conformation polymorphism (SSCP) analysis.
Direct sequencing was performed on fragments with mobility
shifts. All sequence alterations detected in cDNA fragments were
subsequently confirmed on genomic DNA. Table 1 summarizes
all mutations/gene alterations found. Of the 47 cases studied, only
3 patients were without identified mutations in both alleles.
For the two common nonsense mutations, Q70X and W402X,
the relevant exons (II and IX) were analysed by SSCP and
restriction digestion (not shown) as both mutations create an
additional Mael restriction site. Of the 46 European patients, 10
were homozygotes and 14 heterozygotes for W402X, i.e. this
mutation accounted for 37% of mutant alleles. Q70X was detected
in 11 patients in homozygous and in 10 in heterozygous form
(35% of mutant alleles). However, the relative frequency of
W402X and Q70X varies considerably in patients from
Scandinavia and other European countries (Table 2). A previously
described G-to-A transition predicting an alanine-75-threonine
change (8) was detected in one of our patients.
' To whom correspondence should be addressed
862 Human Molecular Genetics, 1994, Vol. 3, No. 6
A C G T
— " "
A C G T
621X< G C ™—
A T /
VT A J
3' A C G T A C G T . 3
CAA < 3 gt...ag gt gta gac gca gtg etc ecc egg ccc ag
ly Pha Lau U l
97 198 199 200..
OC TTC CTO AAC..
CAA G gt...ag gt gta gac gca gtg etc ccc
- ly pha Thr atop
97 198 199 200
OC TTC ACC TOA. .CCC ig
ly Val Aap Ala Val
OT OTA QAC OCA OTQ CTC CCC COB
Pi-a A M Pro Are Lau Hla Lau Aaa.
198 199 200.
CTT CA£ CTO AAC.
A C G T
W 134 del 12
Figure 1. Direct sequencing of IDUA gene mutations. A: Sequencing of the R621
point mutation. While in the genomic PCR fragment the patient shows a C-to-
transition (G-to-A in the noncoding strand shown here) in heterozygous forn
the mutation seems to be in homozygous form in the cDNA (see text for details
B: Sequencing of 964delC (right). An overlap of normal and mutant sequence
is seen from the position of the deletion. C: Sequencing of cDNA fragment 1
in a homozygote for 134dell2. W: wild-type sequence.
cDNA fragment 6 showed an unique band shift in one patient
(not shown). Sequencing revealed a C-to-T transition altering
codon 621 (CGA, arginine) into a stop codon (TGA). The
mutation seems to be homozygous in the cDNA while in the
patients' genomic DNA (exon XTV) both the wild type and mutant
sequences are present (Fig. 1A). We have shown that the patient
carries also the W402X mutation. Obviously, in this case, the
R621X transcript is more stable than the W402X transcript.
W402X was shown to be of maternal and R621X of paternal
origin (data not presented). R621X was not detected on 94 control
chromosomes (not shown).
One patient with W402X on one allele had also an SSCP
alteration in cDNA fragment 4. Sequencing revealed a 1-bp
deletion in exon VII (964delC), present in heterozygous form
(Fig. IB). Due to the frame shift, a stop signal appears 24 codons
ag gc Exon VI
Figure 2. Detection of alternative splicing in the IDUA gene. A: Direct sequencing
shows a 2-base pair insertion in the genomic DNA of the patient. Note that the
mutation leads to an overlap of the normal and mutant sequence in the heterozygous
mother. B: The insertion between codons 198 and 199 in exon VI (underlined)
predicts a frame shift and a premature stop codon (middle row). The arrowhead
points to a cryptic splice site. If this site is used, the alternatively spliced transcript
has a restored reading frame in the patient and encodes for 10 additional and
one altered amino acids (bottom row). C: Primers were designed to amplify only
alternatively (B—D) or only normally (C—D) spliced transcripts. Both kinds
of transcripts are present in the patient (?) and in controls (other lanes). However,
with primers (A—D) that amplify both cDNA types at the same time, only
alternatively spliced transcripts are detected in the patient.
later. In another patient, cDNA fragment 1 was smaller than
expected. Sequencing detected a deletion of 12 nucleotides
(134-145) in the cDNA fragment (Fig. 1C) predicting a loss
of 4 amino acids from the IDUA protein. The deletion was
confirmed and shown to be homozygous at the level of genomic
DNA (exon I). It includes the second of two direct repeats of
11 nucleotides and a T that separates the two repeats. The 12-base
Human Molecular Genetics, 1994, Vol. 3, No. 6 863
Table 1. Mutations identified in 47 patients with MPS I (Hurler syndrome)
insertion of two
deletion of 1
deletion of 12
fs, 24 altered aa.
deletion of 4 aa
CS, N (2), NL (3), YU, D (3)
D (2), NL, PL
F, NL (2)
SF (31, N (7), D
NL (2), D
A, D, N
F, N, NL
62% of alleles from SF/N
G116R (GGG-+AGG) on same
in addition to: V454I (GTC->
ATCI, and R489 (CGC->CGT)
G387 + 6c-»t on same allele
Numbers of nucleotides used to describe the insertion/deletions are according to reference 4.
A = Austria, CS = Czechoslovakia, D = Germany, E = Spain, F = France, N = Norway, NL = The Netherlands, PL = Poland,
SF = Finland and YU = Yugoslavia
Table 2. Frequency of common IDUA mutations among 46 European patients (% of mutant alleles)1
Region/Countries (No. of
L218P Other mutations
All European (46)
Other European (29):
The Netherlands (12)
'The patient from Egypt reported in the paper is not included in this table.
2Consists of 7 different mutations found in only one patient each, see Table 1.
pair deletion was not seen on 94 control chromosomes (not
In a severely affected child of a consanguineous Egyptian
couple, a homozygous insertion of two base pairs (682insAC)
was detected. Fig. 2A shows sequencing of the genomic fragment
in the patient and her mother. A missense point mutation (Gl 16R)
was also seen in the patient that was not detected in any other
MPS I patient studied here and may be a nonpathogenic sequence
Four novel point mutations predicting single amino acid
exchanges were also detected among the MPS I patients studied
here. One patient, heterozygote for W402X, carries a G-to-A
base change in codon 51 predicting a glycine-to-asparagine
alteration in the protein. G51D was also found on 4 alleles of
an independent group of 130 MPS I patients (H.S.Scott,
C.P.Morris, J.J.Hopwood, unpublished data) but not on 116
control chromosomes. In another patient, a G-to-C nucleotide
change was found in codon 489 (R—P). A T-to-C transition of
864 Human Molecular Genetics, 1994, Vol. 3, No. 6
Table 3. Primers used
name of fragment
cDNA fragment 1
Exon V +VI
primer B (see text)
primer C (see text)
in this study
see cDNA fragment 1
(used as primer A in Fig. 2)
(used as primer D in Fig. 2)
see cDNA fragment 3 (primer located in exon VII)
see cDNA fragment 6
'Numbers refer to nucleotide positions according to refs 4 and 5.
2Primers span the exon D/III boundary to prevent amplification of alternatively spliced transcripts without exon n.
the second nucleotide in codon 218 (L218P) was detected in
cDNA fragment 3 of 5 patients. This point mutation creates a
new EagI restriction site. One patient was homozygote for the
mutation, while 4 were compound heterozygotes, one with
W402X, and 3 with Q70X on the other allele. All 5 patients
carried also a previously described intronic alteration (387+6
c—t, ref. 8). Meanwhile, L218P has also been identified in the
patient in whom this nonpathogenic intronic base change was first
detected (L;A.Clarke, unpublished). In the W402X/L218P
compound heterozygote patient, both L218P and 387+6 c—t
were transmitted from the father, and W402X from the mother.
L218P was not present on 210 control chromosomes (data not
shown). In yet another patient, a band shift of cDNA fragment
4 was due to a G-to-C mutation in codon 327 (A—P). A327P
was shown to be present in heterozygous form in 9 patients from
an independent group of 130 MPS I cases (H.S.Scott,
C.P.Morris, J.J.Hopwood, unpublished data) but not on 116
control alleles. Although the 4 novel missense mutations described
above seem to represent the only alterations detected on the alleles
in question, and they have not been found on control
chromosomes, they should be subjected to in vitro expression
studies to demonstrate that they have the potential to cause MPS I.
In the patient with W402X and 964delC, 2 other sequence
alterations were also detected. One was a silent C-to-T transition
in codon 489, the other one a G-to-A in codon 454 predicting
a change of valine to isoleucine. Both alterations have not been
described yet by others. As the Val —He amino acid exchange
is conservative, has been detected on 8 alleles of 58 normal
controls, and appears on a most likely pathogenic MPS I allele,
it is very probably a polymorphism. Numerous different
polymorphisms were detected in the IDUA gene, some of them
altering amino acids (15). In our collective, we found the
following polymorphisms: A8 (C—A; detectable after digestion
with Eco47III) (15), Q33H (G-T; Nsp7524I) (15), R105Q
(G-A) (4), LI 18 (T-C, Kpnl) (4), N181 (T-C) (4) A314
(G-C) (15), A361T (G-A) (15), 3122 (t-c) (15).
Alternative splicing and instability of mRNA with frame shift
Transcripts with early stop codons are known to have a greatly
reduced stability and low abundance in the mRNA pool (for recent
references see 16). The 2-base pair insertion between codons 198
and 199 in exon VI found in one patient predicts a frame shift
and premature termination of translation (Fig. 2B). Surprisingly,
cDNA fragment 3 detected in the patient in routine assay was
about 30 bp larger than the wild type cDNA (Fig. 2C). Direct
sequencing showed that this was due to alternative splicing at
a cryptic splice site in intron 5 that resulted in integration of 28
intronic nucleotides. Interestingly, primers designed to amplify
exclusively one of the two possible messages (primers B and C)
detected both types of transcripts both in the patient and controls.
However, with primers that amplify both kinds of transcripts
(primers A and D), only normally spliced mRNA is seen in
controls, and only alternatively spliced mRNA in the patient. It
seems that the normally spliced transcript, with early stop codon
due to insertion and frame shift, has a considerably lower stability
in the patient than the alternatively spliced transcript, in which
the 2-base pair insertion in combination with alternative splicing
restores the reading frame. The IDUA protein that may result
from this latter transcript of the patient should contain 10
additional amino acids, and the amino acid following the
integration is also altered (Fig. 2B).
A screening for sequence alterations/mutations in the EDUA gene
was performed in a group of 47 nonrelated patients with MPS
I, 46 from different European countries, one from Egypt. Among
the European patients, the two common mutations W402X and
Q70X were present on 37% and 35% of the mutant alleles,
respectively. In addition, 8 other not yet described gene
alterations, most likely disease causing mutations, were identified
accounting for another 15% of the mutant alleles. It is not clear,
whether the remaining 13% of MPS I alleles have not been
Human Molecular Genetics, 1994, Vol. 3, No. 6 865
elucidated because of the insufficient sensitivity of the SSCP
screening or because some patients have large deletions, gross
rearrangements or promotor defects in the IDUA gene, that easily
escape detection by amplification of cDNA fragments and exons.
In addition to W402X and Q70X, L218P is the third frequent
mutation (6.5% of mutant alleles) in our patient group. As this
change has not been identified on 210 control chromosomes it
is unlikely that it represents a polymorphism. The patient,
homozygote for L218P was severely affected. Diagnosis was
reached at the age of 3 and he died at the age of 8. Interestingly,
on the L218P allele we have always found the nonpathogenic
intronic base change 387+6 c—t already described (8). The
presence of two different nucleotide exchanges on one allele, and
the fact that at least 3 of the six L218P alleles share the same
haplotype (Kpn-2, ref. 15, VNTR-1, ref. 17), suggest that this
mutation occurred only once. This is in contrast with other
common mutations which are associated with different haplotypes
Many polymorphisms were found in the IDUA gene, some
of them altering single amino acids (15). Therefore, all newly
identified sequence variants need further investigation whether
or not they are mutations causing MPS I. Of the 4 newly identified
missense point mutations, three (G51D, A327P, and R489P) alter
amino acid residues conserved between human and canine IDUA
(18), which share an overall homology of about 80%.
Nevertheless, to provide a definite proof that these missense
mutations are causative for the disease, in vitro expression studies
will be necessary to measure directly the enzyme activity of the
A homozygous 12-base pair deletion was found in one patient
that should lead to deletion of 4 amino acids (nos. 16-19) from
the signal sequence of the protein. It is possible that this alteration
prevents correct cotranslational transport of the protein into the
endoplasmic reticulum as well as its processing and transport to
the lysosome. The patient was severely affected, diagnosis being
reached at the age of 8 months, with typical and severe somatic
symptoms of the disease at the age of 4. The patient died aged
A nonsense mutation in codon 621 (R621X) should lead to the
lack of 33 amino acids at the amino terminus of the protein. The
patient (with W402X on the other allele) presented with a very
severe clinical phenotype. Facial dysmorphism was visible at
birth, cardiac insufficiency started at the age of 4 months, and
the patient died at the age of 7 years.
A homozygous 2-base pair insertion was detected in exon VI
of a child with typical, severe form of MPS I. The main transcript
derived from this MPS I allele was found to result from alternative
splicing at a cryptic splice site in intron 5. We have shown that
this cryptic site is also used in controls, but the alternatively
spliced transcript is present only in very low amount either
because the cryptic site is used less often than the regular one
or because the mRNA, with frame shift and early stop codon,
has a reduced stability. It is interesting to note that transcripts
spliced at the same cryptic site were also detected in a patient
with a mutation (678-7g —a) in intron 5 affecting splicing (13)
and in another patient with a 22 nucleotide deletion/10 nucleotide
insertion in exon VI (11).
The two common IDUA mutations, W402X and Q70X, were
detected on 72% of mutant alleles in the present study. In another
study on Anglo-Saxon patients, W402X and Q70X were seen
in 46% (31% and 15%) of MPS I alleles (6,7). In our collective
W402X was found in 37%, Q70X in 35% of the investigated
mutant alleles. However, while W402X is more frequent (48%)
in West and Central European countries (mainly The Netherlands
and Germany) and less common (17%) in Northern Europe
(Norway and Finland) the opposite holds for Q70X (62% in
Scandinavia, and 19% in West and Central Europe, Table 2).
The best known example for an uneven geographical distribution
of a common mutation in a recessive disease is the AF508
mutation of the CFTR gene, that is frequent (75-80%) in
Northern Europe while its frequency declines from Northwest
to Southeast (19). In the Turkish population, AF508 is found in
27%, in the Arabian population of Israel only in 21% of CF
alleles (20). Our data show that Q70X of the IDUA gene, that
has a relative frequency comparable to deltaF508 in Northern
Europe, seems also to decline in frequency towards Southeast.
In a small Italian collective (20), Q70X was identified in less
than 10% of mutant alleles while it has not been found yet in
Arab MPS I patients from Israel (9). W402X has not been
detected in this latter group either. W402X seems to have a high
frequency in Central Europe, but is present only with 12% in
Italian patients (21). The basis of such regional distribution of
mutations causative for recessive diseases is not clear. Founder
effects or genetic drift may play a role. A similar difference in
frequency of common mutations across Europe in two
independent genetic diseases like cystic fibrosis and MPS I may
reflect the relative genetic distance of the populations.
MATERIAL AND METHODS
Most of the skin fibroblast lines of MPS I patients used in this study are stored
in the European Human Cell Bank, Rotterdam (WJK). The cell lines of German
patients are from Kinderklinik Mainz (MB). IDUA deficiency was demonstrated
in all cases by enzyme assay (on fibroblasts) using 4-methylumbelliferyl CK-L-
iduronide as a substrate as described previously (22).
DNA and RNA were prepared from cultured fibroblasts according to standard
protocols. Reverse transcription of RNA, and direct sequencing of PCR products
were done as described (23).
The IDUA coding sequence was amplified from the cDNA in 6 overlapping
fragments (see Table 3). Primers for amplification of exons from genomic DNA
are also given in Table 3. Primers B and C were designed to distinguish transcripts
in which exons V and VI are connected (normal splicing) from those which contain
part of intron 5 due to the use of a cryptic splice site (cf. Fig. 2). Primer B is
composed of the last 14 nucleotides of exon V and the first 5 intronic nucleotides
that are integrated into the transcript when the cryptic site at position -28 with
respect to exon VI is used. Primer C spans the normal exon V/VI junction. Both
primers were used together with the 3' primer of cDNA fragment 3 (primer D
in Fig. 2).
PCR reactions were carried out in 10 mM Tris/HCl pH: 8.3, 1.5 mM MgCl2,
50 mM KC1,0.001 % gelatine, 10% DMSO, 0.2 mM dNTPs (0.4 mM for cDNA
fragments 1 and 6 and all exons but VAT), 0.4 ^M of each primer, and 1 U
Taq Polymerase. For cDNA amplification an aliquot of about 10% of the reverse
transcription reaction volume was used, for genomic fragments 200 ng genomic
DNA. After an initial denaturation step for 5 min at 94°C, 35 cycles of
amplification were performed: 45 sec at 94°C, 45 sec at annealing temperature,
90 sec (+ 2 sec in each cycle) at 72°C. Annealing temperatures were 50°C for
exon 1, 54°C for cDNA fragments 1-3, 6, and exon XTV, 58°C for cDNA
fragments 4, 5, and exons n, and V/VI-K.
Single strand conformation polymorphism analysis was performed as described
(22). Prior to SSCP analysis, cDNA fragments 3, 4, and 6 were digested with
Ddel, HphI, and PvuII, respectively. 8% acrylamide gels with 10% glycerol were
run overnight at 24 W for most fragments. Exons I and V/VI were separated
on 6% gels without glycerol and run for 7 hours at 24 W.
We acknowledge the contribution of many colleagues from various European
centres and in particular Professor Dr S.O.Lie, Oslo, for referring to us cell lines
for investigation. We thank Ms Xiao-Hui Guo for ASO screening of patient and
866 Human Molecular Genetics, 1994, Vol. 3, No. 6 Download full-text
control samples. This study was financially supported by the Deutsche
Forschungsgemeinschaft (SFB 367) and by a Program Grant from the National
Health and Medical Research Council of Australia. H.S.S. is a Raymond A.Bryan
IV Fellow in MPS Research.
1. Ashton.L.J., Brooks.D.A., McCourt,P.A.G., Muller,V.J., Clemems.P.R.
and Hopwood,J.J. (1992) Am. J. Hum. Genet. 50, 787-794.
2. Fortuin,J.J.H. and Kleijer.W.J. (1980) Hum. Genet. 53, 155-159.
3. Hopwood.J.J., Vellodi.A., Scott.H.S., Morris,C.P., Litjens.T.,
Clements,P.R., Brooks.D.A., Cooper.A. and Wraith.J.E. (1993) J. Inherit.
Metab. Dis. 16, 1024-1033.
4. Scott.H.S., Anson,D.S., Orsbom,A.M., Nelson.P.V., Clements,P.R.,
Morris,C.P. and Hopwood.J.J. (1991) Proc. Natl. Acad. Sri. USA 88,
5. Scott.H.S., Guo,X.-H., Hopwood,J.J. and Morris,C.P. (1992) Genomics
6. Scott.H.S., Litjens.T., Hopwood.J.J. and Morris.C.P. (1992) Hum. Mut.
7. Scott.H.S., Litjens.T., Nelson.P.V., Brooks.D.A., Hopwood.J.J. and
Morris.C.P. (1992) Hum. Mut. 1, 333-339.
8. Clarke.L.A., Nelson,P.V., Warrington.C, Morris.C.P., Hopwood.J.J. and
Scott.H. (.1994) Hum. Mut. 3, 275-282.
9. Bach.G., Moskowitz.S.M., Tieu.P.T., Matynia.A. and Neufeld.E.F. (1993)
Am. J. Hum. Genet. 53, 330-338.
10. Clarke.L.A. and Scott.H.S. (1993) Hum. Mol Genet. 2, 1311-1312.
11. Moskowitz.S.M., Tieu.P.T. and Neufeld.E.F. (1993) Hum. Mut. 2, 71-73.
12. Moskowitz.S.M., Tieu.P.T. and Neufeld.E.F. (1993) Hum. Mut. 2,
13. Scott.H.S., Litjens.T., Nelson.P.V., Thompson,P.R., Brooks.D.A.,
Hopwood,J.J. and Morris,C.P. (1993) Am. J. Hum. Genet. 53, 973-986.
14. Tieu.P.T., Matynia.A., Bach.G., Hwang.M., Dlott.B. and Neufeld.E.F.
(1993) Am. J. Hum. Genet. 53 (Suppl.), A42.
15. Scott,H.S., Nelson.P.V., Litjens.T., Hopwood.J.J. and Morris.C.P. (1993)
Hum. Mol. Genet. 2, 1471-1473.
16. McIntosh.I., Hamosh.A. and Dietz.H.C. (1993) Nature Genet. 4, 219.
17. Scott.H.S., Nelson,P.V., Hopwood.J.J. and Morris.C.P. (1991) Nucleic Acids
Res. 19, 6348.
18. Stoltzfus.L.J., Sosa-Pineda.B., Moskowitz.S.M., Meneon.K.P., Dlott.B.,
Hooper.L., Teplow.D.B., Shull.R.M. and Neufeld.E.F. (1992) J. Biol.
Chem. 267, 6570-6575.
19. The Cystic Fibrosis Genetic Analysis Consortium (1990) Am. J. Hum. Genet.
20. Lerer.I., Cohen.S., Chemke.M., Sanilevich.A., Rivlin.J., Golan.A.,
YahavJ., Friedman.A. and AbeUovich.D. (1990) Hum. Genet. 85, 416-417.
21. Tonlorenzi.R., Gatti,R., Scott,H.S., Nelson.P.V., Litjens.T., Morris.C.P.
and Hopwood.J.J. (1993) Book of Abstracts. 3rd International Symposium
on the Mucopolysaccharidoses and Related Diseases. May 28-31, 1993.
22. Kleijer,W.J., Thompson,E.J. and Niermeijer,M.F. (1983) Prenatal Diagnosis
23. Bunge.S., Steglich,C, Beck,M., Rosenkranz.W., Schwinger,E.,
Hopwood,J.J. and Gal.A. (1992) Hum. Mol. Genet. 1, 335-339.