HUMAN MUTATION 25:38^44 (2005)
RESEARCH ARTICLE
The Most Common Mutation in FKRP Causing
Limb Girdle Muscular Dystrophy Type 2I (LGMD2I)
May Have Occurred Only Once and Is Present in
Hutterites and Other Populations
Patrick Frosk,1 Cheryl R. Greenberg,1,2 Alysa A.P. Tennese,1 Ryan Lamont,1 Edward Nylen,1 Cheryl
Hirst,1 Danielle Frappier,4 Nicole M. Roslin,4 Michaela Zaik,6 Kate Bushby,5 Volker Straub,6 Mayana
Zatz,7 Flavia de Paula,7 Kenneth Morgan,3,4 T. Mary Fujiwara,3,4 and Klaus Wrogemann1,2n
1
Departments of Biochemistry and Medical Genetics, University of Manitoba, Winnipeg, Canada; 2Department of Pediatrics and Child Health,
University of Manitoba, Winnipeg, Canada; 3Departments of Human Genetics and Medicine, McGill University, Montreal, Canada; 4The
Research Institute of the McGill University Health Centre, Montreal, Canada; 5Institute of Human Genetics, University of Newcastle upon Tyne,
United Kingdom; 6Department of General Pediatrics and Neuropediatrics, University of Essen, Essen, Germany; 7Human Genome Research
Center, Departamento de Biologia, Instituto de Biociências, Universidade de São Paulo, São Paulo, Brazil
Communicated by Jacques Beckmann
Limb girdle muscular dystrophy (LGMD) is common in the Hutterite population of North America. We
previously identified a mutation in the TRIM32 gene in chromosome region 9q32, causing LGMD2H in
approximately two-thirds of the 60 Hutterite LGMD patients studied to date. A genomewide scan was
undertaken in five families who did not show linkage to the LGMD2H locus on chromosome 9. A second
LGMD locus, LGMD2I, was identified in chromosome region 19q13.3, and the causative mutation was
identified as c.826C4A (L276I), a missense mutation in the FKRP gene. A comparison of the clinical
characteristics of the two LGMD patient groups in this population reveals some differences. LGMD2I patients
generally have an earlier age at diagnosis, a more severe course, and higher serum creatine kinase (CK) levels.
In addition, some of these patients show calf hypertrophy, cardiac symptoms, and severe reactions to general
anesthesia. None of these features are present among LGMD2H patients. A single common haplotype
surrounding the FKRP gene was identified in the Hutterite LGMD2I patients. An identical core haplotype was
also identified in 19 other non-Hutterite LGMD2I patients from Europe, Canada, and Brazil. The occurrence of
this mutation on a common core haplotype suggests that L276I is a founder mutation that is dispersed among
populations of European origin. Hum Mutat 25:38–44, 2005. r 2004 Wiley-Liss, Inc.
KEY WORDS:
FKRP; TRIM32; limb girdle muscular dystrophy; LGMD; Hutterites; linkage mapping; linkage
disequilibrium; founder effect
DATABASES:
TRIM32– OMIM: 602290, 254110 (LGMD2H); GDB: 9957765; GenBank: NM_012210.2
FKRP – OMIM: 606596, 607155 (LGMD2I); GDB: 4589036; GenBank: NM_024301.2
www.dmd.nl (Leiden Muscular Dystrophy Pages)
www.ncbi.nlm.nih.gov/SNP/index.html (dbSNP)
INTRODUCTION
The limb girdle muscular dystrophies (LGMDs) are a clinically
and genetically heterogeneous group of disorders characterized by
weakness and wasting in the pelvic and shoulder girdles [Bushby,
1999a, 1999b]. The LGMDs are rare worldwide, with a prevalence
of about 40 per million [Emery, 1991]. LGMDs have been mapped
to 15 different loci; five LGMDs have an autosomal dominant
mode of inheritance and 10 have an autosomal recessive mode of
inheritance [Kaplan and Fontaine, 2004; Zatz et al., 2003].
Mutations in genes have been identified in three of the
dominantly inherited LGMDs and for all of the recessively
inherited LGMDs (see the Leiden muscular dystrophy website;
r2004 WILEY-LISS, INC.
Received 14 July 2003; accepted revised manuscript 9 June 2004.
n
Correspondence to: Dr. Klaus Wrogemann, Department of Biochemistry and Medical Genetics, University of Manitoba, 770 Bannatyne Ave.,Winnipeg, MB R3E 0W3, Canada.
E-mail: K [email protected]
Grant sponsor: German Research Society; Grant numbers: DFG Str
498/3^2; SPP 1086; Grant sponsors: Canadian Institutes of Health
Research; Muscular Dystrophy Association; Muscular Dystrophy Association of Canada; Children’s Hospital Foundation of Manitoba;
Manitoba Health Research Council; Canadian Genetic Diseases Network (Networks of Centres of Excellence Program); Krupp-Stiftung;
Muscular Dystrophy Campaign.
DOI 10.1002/humu.20110
Published online in Wiley InterScience (www.interscience.wiley.com).
A COMMON LGMD2I FOUNDER MUTATION
www.dmd.nl). Among the latter are LGMD2H (MIM# 254110)
and LGMD2I (MIM# 607155). LGMD2H is a relatively mild
form that was first described in the Manitoba Hutterites in 1976
[Shokeir and Kobrinsky, 1976] and has become known as the
‘‘Hutterite type’’ muscular dystrophy. We previously identified a
missense mutation, c.1459G4A (D487N), in TRIM32 as the
putative causative mutation, and all known Hutterite patients in
families that showed linkage to chromosome region 9q32 were
homozygous for this mutation [Frosk et al., 2002]. TRIM32 is a
member of the tripartite-motif family of proteins [Reymond et al.,
2001] and may be an E3 ubiquitin ligase due to the presence of a
RING–finger domain [Freemont, 2000; Horn et al., 2004]. To
date, mutations in this gene have not been found in any patients
outside the Hutterite population. LGMD2I was first described in a
large consanguineous Tunisian family and mapped to chromosome
region 19q13.3 [Driss et al., 2000]. Concurrent with our study,
another group [Brockington et al., 2001a, b] found mutations
within the fukutin-related protein gene (FKRP) in LGMD2I
families, as well as in families with a severe form of congenital
muscular dystrophy, MDC1C (MIM # 606612). FKRP is thought
to be a glycosyltransferase that may function in the O-linked
glycosylation of proteins such as a-dystroglycan [Esapa et al.,
2002; Hewitt and Grewal, 2003]. Recent work has shown that
LGMD2I is one of the most common forms of LGMD worldwide
[Bushby and Beckmann, 2003].
The Hutterites, also called Hutterite Brethren, live on farming
colonies located predominantly in the Prairie Provinces and Great
Plains of North America, and constitute a religious and genetic
isolate. They immigrated to North America from Europe in the
1870s and formed three subdivisions. There has been very little
intermarriage between subdivisions. The ancestry of the overwhelming majority of the Hutterites can be traced back to 89
ancestors [Nimgaonkar et al., 2000]. The history and social
structure of the Hutterite Brethren are described in Hostetler
[1985].
Here we report that the second type of LGMD in the Hutterite
population maps to chromosome 19q31–q33 and is due to
homozygosity for the L276I mutation in FKRP. We also present a
comparison of the clinical variability of LGMD2I and LGMD2H
and evidence that the L276I mutation observed in Hutterites and
other populations is inherited from a common ancestor.
MATERIALS AND METHODS
Patients, Families, and Controls
DNA samples from 38 Hutterites, including 12 affected with
LGMD, from five nuclear families, were included in a genomewide
scan. Subsequent to the genomewide scan, we collected DNA
from five other small Hutterite families with seven individuals
affected with LGMD. The LGMD2H families referred to in this
study are those reported in Frosk et al. [2002], along with one
newly ascertained patient referred by Dr. Keith Brownell,
University of Calgary. The clinical criteria for LGMD were
essentially the same as previously used [Weiler et al. 1998a].
DNA from 19 non-Hutterite LGMD2I patients was obtained. A
total of 12 are from the UK, five are from Brazil (four Caucasian,
one African-Brazilian), one is from Germany, and one is GermanCanadian. A total of 14 of these patients are homozygous for the
FKRP L276I mutation, and the remaining five are heterozygous for
this mutation. The second mutation in four of the five compound
heterozygotes is E310X, W279X, V300M, or Y307N, while no
other mutation was identified in the coding sequence of FKRP for
the remaining heterozygote. A control group of 111 healthy
individuals from Manitoba, not known to have LGMD, was used
to determine allele frequencies. Controls were obtained from the
39
Rh Laboratory, University of Manitoba, and are 90% Caucasian,
with the remainder being Aboriginal, Asian, and African. This
study was approved by the Health Research Ethics Board of the
University of Manitoba.
DNA Analysis
Protocols for DNA analysis have been published previously, as
follows: genotyping of microsatellite markers in candidate genes
and for fine mapping [Weiler et al., 1998a], genomewide scan at
the Montreal Genome Centre [Mira et al., 2003], and DNA
sequencing [Frosk et al., 2002]. PCR cloning was done using a
TOPO TA cloning kit (Invitrogen, www.invitrogen.com) as per
the manufacturer’s instructions. A total of 18 additional markers
used in this study were: D19S903, D19S918, D19S908, D19S219,
DM, D19S412, FKRP52 (dbSNP rs8179080), a C4G SNP in the
FKRP promoter (rs3810288), FKRP c.135C4T (rs2287717),
FMS2 (rs3138636), D19S540, D19S606, D19S902, D19S596,
D19S879, D19S550, D19S867, and D19S904. FKRP52 [Louhichi
et al., 2003], a CA/CAA repeat found within an intron of the
PRKD2 gene, was amplified by PCR with FKRP52_F (50 TCTCCAAAAAACAACAACAAC-30 ) and FKRP52_R (50 CTAGTGTTCTGGGACCTTT-30 ). FMS2, a CA repeat found
within the 30 untranslated region (UTR) of SLC1A5 [Jones et al.,
1994], was amplified with FMS2_F (50 -GGAGGGAATAGGGGATCTGG-30 )
and
FMS2_R
(50 -CACCATGCTGGT0
TATTTTGGC-3 ). FKRP c.135C4T is a silent variant within
codon 45 of FKRP. It was amplified in a 702-bp fragment using
FKRP_ex.4(1)F (50 -CTCAACCTTCTGGTCCTCTTC-30 ) and
FKRP_ex.4(1)R (50 -CCGAGAGGTTGAAGAGGT-30 ). The fragment was then digested with NgoMIV (New England Biolabs,
www.neb.com), the C allele yields two fragments (499 and 203 bp)
and the T allele remains uncut. Rs3810288 in the FKRP promoter
was amplified in a 118-bp fragment using rs3810288_F (50 TCCAACCTGCACCTGGCTAGG-30 ) and rs3810288_R (50 AGCTGGAGGGGTCTGGGAGAtCT-30 ). A mismatch (denoted
by lowercase t) creates an Hpy188I site (New England Biolabs),
the C allele yields two fragments when digested (96 and 22 bp)
and the G allele remains uncut. Samples for both SNPs,
c.135C4T and rs3810288, were analyzed by electrophoresis on
polyacrylamide gels (8 and 15%, respectively).
In addition to genotyping FKRP c.826C4A/L276I
(NM_024301.2) by BfaI digestion [Brockington et al., 2002], we
developed a nested allele-specific PCR assay. A 2-kb PCR product
was generated using primers FKRP_ex4(1)F (50 -CTGCCTTC
CCTTTCGTCC-30 ) and FKRP_ex4(3)R (50 -CCAAAACTCTG
CCCCTGC-30 ). This fragment contains most of exon 4 of the
FKRP gene and was used as template in a secondary allele-specific
amplification step using the reverse primers FKRP_826C (50 CCTTCCCAGCTCACTcG-30 ) and FKRP_826A (50 -CCTTCCC
AGCTCACTcT-30 ). Each primer has an intentional mismatch at
the second-to-last base pair (lowercase) to destabilize the resultant
DNA duplex, and the last base pair corresponds to either the
normal (C) or mutant (A) sequence. A common forward primer,
FKRP_ex4(2)F (50 -CCTGGACGGAGATGCTGT-30 ) was used
and amplification with the two reverse allele-specific primers was
done in separate PCR reactions. Samples were analyzed by
electrophoresis in 1.5% agarose.
Protein Analysis
Muscle proteins were extracted in treatment buffer containing
0.125 M Tris-HCl (pH 6.8), 4% SDS, 20% glycerol, 5%
mercaptoethanol, and 0.0001% bromphenol blue. Soluble proteins
were separated by SDS-PAGE on 3 to 10% linear gradient gels and
transferred to a nitrocellulose membrane. The membrane was
blocked in 3% milk powder in phosphate buffered saline (PBS),
treated with anti-glycosylated a-dystroglycan antibody (IIH6C4,
Upstate 1:1,000, www.upstate.com) or anti-b-dystroglycan antibody (Novocastra 1:50, www.novocastra.co.uk), and washed and
40
FROSK ET AL.
incubated with horseradish peroxidase (HRP)-conjugated antimouse secondary antibodies (Dianova, www.dianova.de). Immunoreactive bands were detected using a chemiluminescence
detection system (ECL, Amersham Biosciences, www.amersham
biosciences.com).
Skeletal muscle tissues were embedded in traganth and frozen in
liquid nitrogen-cooled isopentane. Cryosections (7 mm) were
immunostained with anti-glycosylated a-dystroglycan (IIH6C4;
Upstate) in TBS/1% bovine serum albumin (BSA) for 90 min.
After washing, sections were incubated with the appropriate Texas
Red dye-conjugated secondary antibody for 30 min. Sections were
observed under a Zeiss Axioplan fluorescence microscope
(www.zeiss.com).
RESULTS
Identi¢cation of the LGMD2I Locus in Hutterites
After the LGMD2H locus had been mapped, five Hutterite
families did not show linkage to chromosome region 9q31–q33,
two of which had been reported previously [Weiler et al., 1998b].
After excluding the known LGMD loci LGMD1A, 1B, 1D, and
2A–2G, we performed a genomewide scan, using 389 microsatellite markers with an average spacing of 9.1 cM, on 38
individuals, including 11 affected with LGMD. Single-point
parametric linkage analysis with a fully penetrant autosomal
recessive disease model was performed using GENEHUNTER 2.1
[Kruglyak et al., 1996; Markianos et al., 2001]. A maximum
logarithmic odds (lod) score of 1.50 at D19S587 was obtained. The
second highest score was 0.87 at the adjacent marker, D19S178,
and the third highest score was 0.67 at D2S407. These lod scores
were all at zero recombination. Using multipoint linkage analysis,
the maximum multipoint lod score was 3.18 at D19S178. The
region with the next highest multipoint lod score was 1.06 at
D2S407.
To define a candidate gene interval, families were genotyped for
18 additional markers in a 13-cM region between two genomescan markers, D19S178 and D19S246 (see Materials and Methods
section). Three markers, FMS2, FKRP52, and D19S902, in a 1.3cM interval, had single-point lod scores of 5.33 at zero
recombination. During our fine-mapping studies, Brockington
et al. [2001a] reported that a form of congenital muscular
dystrophy was caused by mutations in FKRP, a promising candidate
gene for LGMD2I. The entire coding region of FKRP was
sequenced in one Hutterite LGMD2I patient who was found to
be homozygous for a missense mutation, c.826C4A (L276I), that
results in the substitution of isoleucine for leucine. This same
mutation was reported by Brockington et al. [2001b], with the
patients from 15 of their 17 LGMD2I families being homozygous
(five families) or heterozygous (10 families). We found that every
Hutterite LGMD patient who was not homozygous for the
TRIM32 D487N mutation, was homozygous for the FKRP L276I
mutation. L276I was not found in our control group of 111
individuals.
The majority of Hutterites who are homozygous for the L276I
mutation are also homozygous for D19S412 (109 bp), FKRP52
(110 bp), rs3810288 (G allele), c.135C4T (T allele), FMS2 (142
bp), and D19S540 (184 bp) (Fig. 1). This indicates that a genomic
segment of about 0.5 Mb is shared among Hutterites carrying
L276I and is likely identical by descent from a common ancestor.
In comparison to the allele frequencies of our 111 controls (Table
1), there is a strong association of L276I in the non-Hutterite
patients with the 110-bp allele of FKRP52 (52 kb centromeric to
L276I), the G allele of rs3810288 (10 kb centromeric to L287I),
the T allele of c.135C4T (0.7 kb centromeric to L276I), and the
142-bp allele of FMS2 (19 kb telomeric to L276I). The association
of L276I with the 109-bp allele of D19S412 (250 kb centromeric
to L276I) and the 184 bp allele of D19S540 (250 kb telomeric to
L276I) does not appear as strong (Table 1). The markers flanking
this region (DM, 1 Mb centromeric and D19S606, 0.7 Mb
telomeric) show no association with L276I in the patients that we
studied (Table 1). Recombination appears to have occurred
between L276I and FMS2 on one chromosome (Patient
C11.975) and between rs3810288 and c.135C4T on another
(Patient NCL-10; Fig. 1). This results in a very small common core
haplotype consisting of the mutation itself and c.135C4T. The
likelihood that this set of associations has occurred by chance is
low and reflects strong linkage disequilibrium. Thus, L276I appears
to have arisen only once and is identical by descent in most, and
possibly all patients. The presence of a homozygous AfricanBrazilian individual showing these same associations (Patient
C10.882; Fig. 1) raises the possibility that the mutation is not
Genotypes for markers in a 1.7-Mb region £anking FKRP.The Hutterite haplotype corresponds to the most frequent haplotype in the Hutterites. Patients NCL-01 through 12 are Caucasian patients collected in the UK. Patients C6.625, C8.749, C11.975, and
C17.553 are Caucasian Brazilian patients and Patient C10.882 is African-Brazilian [de Paula et al., 2003]. Patient G-01 is German and
Patient G-04 is German-Canadian. Black regions indicate alleles that are consistent with the consensus Hutterite haplotype; gray
shading indicates alleles that are one repeat unit di¡erent and may have arisen by slipped mispairing during DNA replication; boxed
areas indicate alleles that were phased by PCR cloning. Alleles for microsatellite markers are designated by length in bp except for
DM, which is reported as the number of CAG repeats. For c.135C4Tand rs3810288, alleles are designated by nucleotide and for FKRP
L276I, alleles are designated by amino acid.
FIGURE 1.
A COMMON LGMD2I FOUNDER MUTATION
TABLE 1.
Marker
DM
D19S412
FKRP52
rs3810288
c.135C4T
FKRP L276I
FMS2
D19S540
D19S606
41
Frequency of Chromosome 19 MarkerAlleles Associated With the L276Ia Mutation in FKRP
Allele
Hutterite 276I
chromosomes (%)
Non-Hutterite 276I
chromosomes (%)
Frequency of allele in
controls chromosomes
(%) N=222
Physical distance from
the L276I mutation
(bp) b
22 repeats
109 bp
110 bp
G
T
Ile
142 bp
184 bp
184 bp
92 c
100
97 c
100
100
100
100
100
87 c
0
70
82
97
100
100
94
79
10
nd
36
2
34
14
0
18
31
nd
986,085
248,199
52,237
10,554
691
0
+18,767
+247,548
+714,207
Inferred from the c.826C4A change identi¢ed within FKRP (NM_024301.2).
UCSC Human Genome Assembly website (http://genome.ucsc.edu/index.html; April 2003 Assembly).
The other observed alleles were DM/5 and 13 repeats; FKRP52/112 bp; D19S606/180 and 182 bp.
a
b
c
specific to Caucasians. However, with our limited data this cannot
be conclusively determined, particularly in light of the extreme
amount of admixture amongst Brazilians [Parra et al., 2003].
Of note are the discrepancies present at the FKRP52 locus. A
possible mutation in FKRP52 due to slipped strand mispairing
during DNA replication is present on five of the nine Brazilian
chromosomes (110 bp4108 bp). This suggests a recent Brazilian
mutational event in a common ancestor of three of these five
Brazilian patients [de Paula et al., 2003]. In addition, within the
Hutterite population there are also individuals with a 112-bp allele
instead of a 110-bp allele on the same haplotype as the L276I
mutation. Through cloning and sequencing, we have found a large
amount of variation in FKRP52 (data not shown). There is
variation in both a dinucleotide (CA) and trinucleotide (CAA)
stretch between the FKRP52 primers, however, trinucleotide
variation is much less common. We have found that the range of
variation in the CAA stretch is limited to two possibilities (three or
four CAA repeats, four being rare), whereas the range of variation
in the CA stretch is much greater (13 to 31 CA repeats). The net
result is allele sizes in the range of 92–128 bp, with variation in the
CAA stretch showing up as odd-sized alleles within this range. A
total of 17 different alleles were found to be present in our 111
control samples at frequencies ranging from 0.5 to 26.1% (data not
shown). This hypervariability readily explains the discrepancies
that we have detected at this locus in our patient samples.
Phenotype^Genotype Correlations
To date we have identified 60 Hutterite individuals from 27
nuclear families who are homozygous for either TRIM32 D487N
(41 individuals from 17 nuclear families) or FKRP L276I (19
individuals from 10 nuclear families). Table 2 shows clinical details
on individuals who were homozygous or heterozygous for the
mutations. There is considerable clinical variability even among
siblings. Serum creatine kinase (CK) levels tended to be higher in
the LGMD2I than in the LGMD2H patients. Severe dilated
cardiomyopathy was the presenting symptom in one LGMD2I
patient and has subsequently been found in another three
patients. Calf hypertrophy with proximal muscle wasting reminiscent of that seen in patients with Becker muscular dystrophy has
been observed in the LGMD2I patients in our study, but is not a
constant feature. No other muscular hypertrophy, including
macroglossia, was noted. Two children presented with reactions
to inhalation anesthetics (succinylcholine/halothane) during
dental surgery, one with a masseter spasm and the other with
severe rhabdomyolysis. Both of these patients recovered post-
operatively. They showed persistently elevated resting CK levels
and muscle biopsies with dystrophic features. Both were subsequently found to be L276I homozygotes. Recently, a German
LGMD2I patient (compound heterozygote for L276I and V121E)
was reported to also have had a malignant hyperthermia-like
episode subsequent to inhalation anaesthetic [Walter et al., 2004].
This suggests that LGMD2I patients, in general, are at risk for
reactions to inhalation anaesthetic. Unlike LGMD2I, to our
knowledge, LGMD2H patients have shown none of the following
characteristics: muscle hypertrophy, reaction to general anesthetics, or development of cardiomyopathy. In addition, none of
the LGMD patients of either type that we have studied have
shown signs of facial weakness or any respiratory symptoms;
however, subtle respiratory difficulties cannot be ruled out
[Dohna-Schwake et al., 2004].
Among the Hutterite LGMD families that we studied, there was
one individual who was homozygous for FKRP L276I and
heterozygous for TRIM32 D487N. This individual has proximal
muscle weakness and a highly elevated CK level of 9,190 U/L.
There were also six individuals who were homozygous for TRIM32
D487N and heterozygous for FKRP L276I, representing two
families (a father and daughter from one family and a set of four
siblings from another family). The father is ambulatory but has
proximal muscle weakness, a dystrophic muscle biopsy, and a CK
level of 669 U/L. The daughter is asymptomatic at this time and
her CK level is 267 U/L. In the remaining family, all four siblings
have CK levels Z10 maximum normal (range 1,700–2,960 U/
L), one has had a clearly dystrophic muscle biopsy, and two show
proximal weakness but they are all ambulatory. Within both of
these families, there are eight individuals who are heterozygous for
both mutations. Seven of these individuals were available for study
and were clinically normal (as examined by C.R.G.). CK values for
these individuals ranged from 35–294 U/L.
DISCUSSION
We have demonstrated unexpected locus heterogeneity for
LGMD in the Hutterite population, and have identified patients
who are homozygous for a missense mutation (D487N) in TRIM32
or homozygous for a missense mutation (L276I) in FKRP. This
provides another example of genetic heterogeneity of an autosomal
recessive disease in a genetically isolated population. Both locus
and allelic heterogeneity were found for LGMD in the Amish,
another Anabaptist isolate [Duclos et al., 1998]. At this time,
there is no evidence for a third locus causing LGMD in this
population.
1
1
52^411
35^294
1
33 44^1,460
322^26,087
Six of these are also heterozygous for LGMD2I with no obviously di¡erent phenotype.
One of these is also heterozygous for LGMD2H with no obviously di¡erent phenotype.
b
a
23
8
Both
Heterozygote
Double
Heterozygote
na
na
Absent (23)
Absent (7),
not studied (1)
Not studied (49)
Dilated
cardiomyopathy (4),
normal (10),
not Studied (5)
Not studied (23)
Not Studied (8)
Absent (49)
Absent (8),
Borderline (2),
Present (9)
49
19b
LGMD2I/
FKRP L276I
Heterozygote
Homozygote
na
2^25
mean=12
10 81^5,556
Normal (13),
not studied (28)
41a
Homozygote
LGMD2H/
TRIM32 D487N
9^42
mean=24
Absent (41)
Mean Multiples
of Max Normal
Highest Resting
CK (U/litre)
Cardiac Status
Di¡erences Between LGMD2H and LGMD2I in the Hutterites
Calf
Hypertrophy
Status
Mutation
Age of Onset
(years)
Number of
Individuals
TABLE 2. Clinical
13 Asymptomatic (7^42),
4 Assisted ambulation (42^47),
3 Wheelchair bound (60^66),
21 Ambulatory (13^53)
49 Asymptomatic (6^77)
1 Deceased/congestive
heart failure (40),
1 Assisted ambulation (40),
17 Ambulatory (5^50)
23 Asymptomatic (6^70)
7 Asymptomatic (12^47),
Not studied/Deceased (1)
FROSK ET AL.
Current status
(age range in years)
42
LGMD2H appears to be more frequent in the Schmiedeleut
subdivision of the Hutterites, whereas LGMD2I appears to be
more frequent in the Dariusleut. Currently, we do not have an
accurate estimate of the relative frequencies of the two mutations
in each of the subdivisions. Overall, we have ascertained 58
Canadian Hutterite LGMD patients and the Canadian Hutterite
population is estimated to be 28,020 (Statistics Canada;
www.statcan.ca). Thus, the estimated prevalence of LGMD in
this population is at least 1 in 483. This is very much higher than
the highest prevalence reported to date of 1 in 14,493 (69 per
million) in the Guipúzcoa population in Spain [Urtasun et al.,
1998].
There is considerable clinical heterogeneity for both LGMD2H
and LGMD2I in spite of a uniform communal lifestyle and only
one mutation for each of the LGMDs. A wide spectrum in clinical
severity has been previously reported for LGMD2I and ascribed to
the various mutations found in compound heterozygotes [Mercuri
et al., 2003]. We have observed similarly large clinical variation,
although all our patients are homozygous for the FKRP L276I
mutation. This was also seen in a recent study by Walter et al.
[2004] in which 13 out of 20 patients from nonconsanguineous
matings were homozygous for this mutation and showed similar
clinical variability. Our impression is that patients with LGMD2I
present earlier, follow a more severe course with possible
cardiomyopathy (Poppe et al., submitted manuscript), and have
higher serum CK levels than LGMD2H patients, although we may
be underascertaining more mildly affected LGMD2I patients. A
comprehensive ascertainment of LGMD in the Hutterites is
needed to provide accurate information on the prevalence,
penetrance of the genotypes, and clinical variability of LGMD.
In addition, there is no indication of any interaction between the
two loci as individuals with mutations at both loci are
indistinguishable from those with mutations at only one locus
(Table 1). This is not surprising due to the apparently different
mechanisms by which these genes appear to cause muscular
dystrophy [Esapa et al., 2002; Frosk et al., 2002]. Presumably, the
only situation that might result in a more severe phenotype would
be a double mutation homozygote.
It appears that all Hutterite LGMDs are of either type 2H or
type 2I. This will make it possible to provide accurate noninvasive
diagnostic and carrier testing for LGMD in Hutterites. Specific
tests for these two mutations are currently being established in the
molecular diagnostic laboratory at the Health Sciences Centre,
Winnipeg, Manitoba. Such a DNA-based approach is not yet
practical for the non-Hutterite LGMD population because of the
marked locus and allelic heterogeneity. Given the high incidence
of LGMD2I carriers in the Hutterite population, and the risk of
cardiomyopathy and anaesthetic reactions in this group, we would
suggest that genetic testing of at-risk individuals even below the
normal age of consent for such testing should be considered and
discussed with the families.
The FKRP L276I mutation appears to be common worldwide,
with respect to LGMD-causing mutations. Carrier frequency is
estimated to be 1 out of 306 on the basis of controls typed for
L276I [Brockington et al., 2001b; de Paula et al., 2003; Walter
et al., 2004; this study]. In addition, patients homozygous for
L276I from 28 nonconsanguineous families have been reported
[Brockington et al., 2001b; Poppe et al., 2003; Walter et al., 2004].
Our analysis indicates that L276I may be a founder mutation, as
all Hutterite and non-Hutterite disease chromosomes tested to
date carry the low frequency T allele (14%) at an intragenic SNP
(c.135C4T) and the G allele at rs3810288 in the putative
A COMMON LGMD2I FOUNDER MUTATION
promoter of FKRP (with one exception). The L276I mutation with
the C allele at rs3810288 (Patient NCL-10; Fig. 1) is consistent
with being a recombinant chromosome but due to the lack of
phase information for most of the genotypes this cannot be firmly
established. The T allele of c.135C4T and the G allele of
rs3810288 have been shown to be associated with a further 26
L276I chromosomes from German patients [Walter et al., 2004],
strengthening the evidence for a founder mutation. Markers as far
away from the L276I mutation as 0.25 Mb in each direction show
readily detectable linkage disequilibrium; however, markers 0.75–
1.0 Mb away show very little linkage disequilibrium in the samples
used in this study. This is strong evidence that L276I has arisen
only once. The small common core haplotype is an indication that
the mutation may have occurred long ago. Further analysis of
SNPs in and around FKRP and additional patients will be needed
to confirm that most, if not all, copies of L276I in the
contemporary population are identical by descent and to
accurately estimate the age of the mutation.
The relative frequency of L276I in Caucasians, compared to
other LGMD-causing mutations, and the high likelihood that it is
a founder mutation, are readily explained by genetic drift.
However, given the postulated function of FKRP, it is tempting
to speculate that a selective advantage may also be contributing to
the maintenance of this allele. FKRP is thought to be a
glycosyltransferase, and its mutations affect the glycosylation of
a-dystroglycan, an essential component of muscle cell membranes
[Durbeej et al., 1998a; Esapa et al., 2002; Hewitt and Grewal,
2003]. An immunoblot of muscle from an L276I homozygote
shows a decreased level of fully glycosylated a-dystroglycan (Fig.
2). Two other groups have recently reported variably decreased
levels of a-dystroglycan in patients homozygous for L276I as well
[Brown et al., 2004; Walter et al., 2004]. a-Dystroglycan is known
to be expressed in numerous tissues and has been shown to be the
receptor for the entry of two different types of pathogens [Cao
et al., 1998; Durbeej et al., 1998b; Rambukkana et al., 1998]. It is
FIGURE 2. Immunodetection of glycosylated a-dystroglycan. Immuno£uorescence staining of skeletal muscle from a control individual (a) and from LGMD2I Patient G-01 (b) with antiglycosylated a-dystroglycan antibody (IIH6C4, Upstate). c: Reduction of a-dystroglycan expression in skeletal muscle of a patient (lane 2) compared to control muscle (lane 1) by
immunoblot. No reduction is apparent in b-dystroglycan. Equal
loading determined by Ponceau S staining (not shown).
43
possible that an FKRP L276I heterozygote with a mild defect in
glycosylation may have a partial resistance to these or other
pathogens. Over long periods this advantage would then increase
the prevalence of the FKRP L276I allele in areas where the
pathogen is endemic. On the basis of the data presented here, it
appears that the frequent occurrence of L276I is not the result of
multiple de novo mutations as was previously thought [Bushby
and Beckmann, 2003]. Instead, our findings suggest this mutation
has occurred once and became prevalent through either genetic
drift, selective advantage, or some combination of both.
ACKNOWLEDGMENTS
We are indebted to the patients and their families for their
participation in this study. We thank Drs. T. Bree, F. Booth, C.
Bourque, K. Brownell, A. Hoke, W. Ilse, T. Ladd, N. Lowry, C.
Toth, and S. Weizman for patient samples, and Dr. T. Zelinski for
control samples. We thank Dr. T. Hudson and A. Verner for
facilitating the genomewide scan in the Montreal Genome Centre
(McGill University and Genome Quebec Innovation Centre), J.
Crumley for maintaining the genealogical database for many years
and for the XBase computer programs, J. Loredo-Osti for helpful
discussions, and L. Cree and M. Buddles for mutation analysis in
the Newcastle patients. V. Straub and M. Zaik were supported by
the German Research Society (DFG Str 498/3–2, SPP 1086) and
the Krupp-Stiftung. The Newcastle Muscle Centre receives
financial support from the Muscular Dystrophy Campaign.
REFERENCES
Brockington M, Blake DJ, Prandini P, Brown SC, Torelli S, Benson
MA, Ponting CP, Estournet B, Romero NB, Mercuri E, Voit T,
Sewry CA, Guicheney P, Muntoni F. 2001a. Mutations in the
Fukutin-related protein gene (FKRP) cause a form of congenital
muscular dystrophy with secondary laminin alpha2 deficiency
and abnormal glycosylation of alpha-dystroglycan. Am J Hum
Genet 69:1198–1209.
Brockington M, Yuva Y, Prandini P, Brown SC, Torelli S, Benson
MA, Herrmann R, Anderson LVB, Bashir R, Burgunder JM,
Fallet S, Romero N, Fardeau M, Straub V, Storey G, Pollitt C,
Richard I, Sewry CA, Bushby K, Voit T, Blake DJ, Muntoni F.
2001b. Mutations in the fukutin-related protein gene (FKRP)
identify limb girdle muscular dystrophy 2I as a milder allelic
variant of congenital muscular dystrophy MDC1C. Hum Mol
Genet 10:2851–2859.
Brockington M, Blake DJ, Torelli S, Brown SC, Muntoni F. 2002.
The gene for a novel glycosyltransferase is mutated in congenital
muscular dystrophy MDC1C and limb girdle muscular dystrophy
2I. Neuromuscul Disord 12:233–234.
Brown SC, Torelli S, Brockington M, Yuva Y, Jimenez C, Feng L,
Anderson L, Ugo I, Kroger S, Bushby K, Voit T, Sewry C,
Muntoni F. 2004. Abnormalities in alpha-dystroglycan expression in MDC1C and LGMD2I muscular dystrophies. Am J
Pathol 164:727–737.
Bushby KM. 1999a. Making sense of the limb-girdle muscular
dystrophies. Brain 122:1403–1420.
Bushby KM. 1999b. The limb-girdle muscular dystrophies-multiple
genes, multiple mechanisms. Hum Mol Genet 8:1875–1882.
Bushby KM, Beckmann JS. 2003. The 105th ENMC sponsored
workshop: pathogenesis in the non-sarcoglycan limb-girdle
muscular dystrophies, Naarden, April 12-14, 2002. Neuromuscul Disord 13:80–90.
44
FROSK ET AL.
Cao W, Henry MD, Borrow P, Yamada H, Elder JH, Ravkov EV,
Nichol ST, Compans RW, Campbell KP, Oldstone MB. 1998.
Identification of alpha-dystroglycan as a receptor for lymphocytic choriomeningitis virus and Lassa fever virus. Science
282:2079–2081.
de Paula F, Vieira N, Starling A, Yamamoto LU, Lima B, de Cassia
PR, Vainzof M, Nigro V, Zatz M. 2003. Asymptomatic carriers
for homozygous novel mutations in the FKRP gene: the other
end of the spectrum. Eur J Hum Genet 11:923–930.
Dohna-Schwake C, Ragette R, Mellies U, Straub V, Teschler H,
Voit T. 2004. Respiratory function in congenital muscular
dystrophy and limb girdle muscular dystrophy 2I. Neurology
62:513–514.
Driss A, Amouri R, Ben Hamida C, Souilem S, Gouider-Khouja N,
Ben Hamida M, Hentati F. 2000. A new locus for autosomal
recessive limb-girdle muscular dystrophy in a large consanguineous Tunisian family maps to chromosome 19q13.3. Neuromuscul Disord 10:240–246.
Duclos F, Broux O, Bourg N, Straub V, Feldman GL, Sunada Y,
Lim LE, Piccolo F, Cutshall S, Gary F, Quetier F, Kaplan JC,
Jackson CE, Beckmann JS, Campbell KP. 1998. Beta-sarcoglycan: genomic analysis and identification of a novel missense
mutation in the LGMD2E Amish isolate. Neuromuscul Disord
8:30–38.
Durbeej M, Henry MD, Campbell KP. 1998a. Dystroglycan in
development and disease. Curr Opin Cell Biol 10:594–601.
Durbeej M, Henry MD, Ferletta M, Campbell KP, Ekblom P.
1998b. Distribution of dystroglycan in normal adult mouse
tissues. J Histochem Cytochem 46:449–457.
Emery AE. 1991. Population frequencies of inherited neuromuscular diseases—a world survey. Neuromuscul Disord
1:19–29.
Esapa CT, Benson MA, Schroder JE, Martin-Rendon E, Brockington M, Brown SC, Muntoni F, Kroger S, Blake DJ. 2002.
Functional requirements for fukutin-related protein in the Golgi
apparatus. Hum Mol Genet 11:3319–3331.
Freemont PS. 2000. RING for destruction? Curr Biol 10:R84–R87.
Frosk P, Weiler T, Nylen E, Sudha T, Greenberg CR, Morgan K,
Fujiwara TM, Wrogemann K. 2002. Limb-girdle muscular
dystrophy type 2H associated with mutation in TRIM32, a
putative E3-ubiquitin-ligase gene. Am J Hum Genet 70:663–672.
Hewitt JE, Grewal PK. 2003. Glycosylation defects in inherited
muscle disease. Cell Mol Life Sci 60:251–258.
Horn EJ, Albor A, Liu Y, El Hizawi S, Vanderbeek GE, Babcock M,
Bowden GT, Hennings H, Lozano G, Weinberg WC, KuleszMartin M. 2004. RING protein Trim32 associated with skin
carcinogenesis has anti-apoptotic and E3-ubiquitin ligase
properties. Carcinogenesis 25:157–167.
Hostetler JA. 1985. History and relevance of the Hutterite
population for genetic studies. Am J Med Genet 22:453–462.
Jones EM, Menzel S, Espinosa R, III, Le Beau MM, Bell GI, Takeda
J. 1994. Localization of the gene encoding a neutral amino acid
transporter-like protein to human chromosome band 19q13.3
and characterization of a simple sequence repeat DNA
polymorphism. Genomics 23:490–491.
Kaplan JC, Fontaine B. 2004. Neuromuscular disorders: gene
location. Neuromuscul Disord 14:85–106.
Kruglyak L, Daly MJ, Reeve-Daly MP, Lander ES. 1996.
Parametric and nonparametric linkage analysis: a unified
multipoint approach. Am J Hum Genet 58:1347–1363.
Louhichi N, Triki C, Quijano-Roy S, Richard P, Makri S, Meziou
M, Estournet B, Mrad S, Romero NB, Ayadi H, Guicheney P,
Fakhfakh F. 2003. New FKRP mutations causing congenital
muscular dystrophy associated with mental retardation and
central nervous system abnormalities. Identification of a founder
mutation in Tunisian families. Neurogenetics 5:27–34.
Markianos K, Daly MJ, Kruglyak L. 2001. Efficient multipoint
linkage analysis through reduction of inheritance space. Am J
Hum Genet 68:963–977.
Mercuri E, Brockington M, Straub V, Quijano-Roy S, Yuva Y,
Herrmann R, Brown SC, Torelli S, Dubowitz V, Blake DJ,
Romero NB, Estournet B, Sewry CA, Guicheney P, Voit T,
Muntoni F. 2003. Phenotypic spectrum associated with mutations in the fukutin-related protein gene. Ann Neurol 53:
537–542.
Mira MT, Alcais A, Van Thuc N, Thai VH, Huong NT, Ba NN,
Verner A, Hudson TJ, Abel L, Schurr E. 2003. Chromosome
6q25 is linked to susceptibility to leprosy in a Vietnamese
population. Nat Genet 33:412–415.
Nimgaonkar VL, Fujiwara TM, Dutta M, Wood J, Gentry K,
Maendel S, Morgan K, Eaton J. 2000. Low prevalence of
psychoses among the Hutterites, an isolated religious community. Am J Psychiatry 157:1065–1070.
Parra FC, Amado RC, Lambertucci JR, Rocha J, Antunes CM,
Pena SD. 2003. Color and genomic ancestry in Brazilians. Proc
Natl Acad Sci USA 100:177–182.
Poppe M, Cree L, Bourke J, Eagle M, Anderson LV, Birchall D,
Brockington M, Buddles M, Busby M, Muntoni F, Wills A,
Bushby K. 2003. The phenotype of limb-girdle muscular
dystrophy type 2I. Neurology 60:1246–1251.
Rambukkana A, Yamada H, Zanazzi G, Mathus T, Salzer JL,
Yurchenco PD, Campbell KP, Fischetti VA. 1998. Role of alphadystroglycan as a Schwann cell receptor for Mycobacterium
leprae. Science 282:2076–2079.
Reymond A, Meroni G, Fantozzi A, Merla G, Cairo S, Luzi L,
Riganelli D, Zanaria E, Messali S, Cainarca S, Guffanti A,
Minucci S, Pelicci PG, Ballabio A. 2001. The tripartite motif
family identifies cell compartments. EMBO J 20:2140–2151.
Shokeir MH, Kobrinsky NL. 1976. Autosomal recessive muscular
dystrophy in Manitoba Hutterites. Clin Genet 9:197–202.
Urtasun M, Saenz A, Roudaut C, Poza JJ, Urtizberea JA, Cobo
AM, Richard I, Garcia BF, Leturcq F, Kaplan JC, Marti Masso JF,
Beckmann JS, Lopez dM. 1998. Limb-girdle muscular dystrophy
in Guipuzcoa (Basque Country, Spain). Brain 121:1735–1747.
Walter MC, Petersen JA, Stucka R, Fischer D, Schroder R,
Vorgerd M, Schroers A, Schreiber H, Hanemann CO, Knirsch
U, Rosenbohm A, Huebner A, Barisic N, Horvath R, Komoly S,
Reilich P, Muller-Felber W, Pongratz D, Muller JS, Auerswald
EA, Lochmuller H. 2004. FKRP (826C4A) frequently causes
limb-girdle muscular dystrophy in German patients. J Med
Genet 41:e50.
Weiler T, Greenberg CR, Zelinski T, Nylen E, Coghlan G, Crumley
MJ, Fujiwara TM, Morgan K, Wrogemann K. 1998a. A gene for
autosomal recessive limb-girdle muscular dystrophy in Manitoba
Hutterites maps to chromosome region 9q31-q33: evidence for
another limb-girdle muscular dystrophy locus. Am J Hum Genet
63:140–147.
Weiler T, Sudha T, Brunham L, Nylen E, Patel L, Halliday W,
Greenberg CR, Wrogemann K. 1998b. Genetic heterogeneity of
limb girdle muscular dystrophy in Manitoba Hutterites. Am J
Hum Genet 63(Suppl):A392–A392.
Zatz M, de Paula F, Starling A, Vainzof M. 2003. The 10 autosomal
recessive limb-girdle muscular dystrophies. Neuromuscul Disord
13:532–544.