doi:**


231Letters to the Editor

molecular studies in the Prader-Willi and Angelman syn- usually first evident in the pelvic girdle and then spread-
drome: an overview. Am J Med Genet 46:2 – 6 ing to the upper limbs while sparing facial muscles. On-

Kuwano A, Mutirangura A, Dittrich B, Buiting K, Horsthemke set of symptoms is variable (mean age 9 years), and
B, Saitoh S, Niikawa N, et al (1992) Molecular dissection creatine kinase (CK) levels are elevated from early in-
of the Prader-Willi/Angelman syndrome region (15q11-13) fancy and remain elevated until the individual is well
by YAC cloning and FISH analysis. Hum Mol Genet 1:417 –

past this age (Jackson and Strehler 1968). Affected indi-
425

viduals are often wheelchair-bound 20 – 30 years afterLakich D, Kazazian HH, Antonarakis S, Gitschier J (1993)
the onset of symptoms. There is variability in the age ofInversions disrupting the factor VIII gene are a common
death, and most individuals die in middle age.cause of severe haemophilia A. Nat Genet 5:236 – 241

The gene for LGMD2A was first linked to chromo-Lehrman MA, Goldstein JL, Russel DW, Brown MS (1987)
Duplication of seven exons in LDL receptor gene caused by some 15 by Beckmann et al. (1991). Allamand et al.
Alu-Alu recombinatiion in a subject with familial hypercho- (1995) narrowed the region to 15q15.1-q15.3, using
lesterolemia. Cell 48:827 – 835 large kindreds from the Isle of La Réunion and the

Marcus S, Hellegren D, Lambert B, Fallstrom SP, Wahlstrom northern Indiana Amish. The muscle-specific calcium-
J (1993) Duplication in the hypoxanthine phosphoribosyl- activated neutral protease 3 or calpain 3 (CANP3) gene,
transferase gene caused by Alu-Alu recombination in a pa- a possible candidate gene in the 15q15.1-q15.3 region,
tient with Lesch Nyhan syndrome. Hum Genet 90:477 – 482

was examined by Richard et al. (1995). Fifteen different
Nathans J, Piantanida TP, Eddy RL, Shows TB, Hogness DS

mutations, including missense, splice-site, frameshift,(1986) Molecular genetics of inherited variation in human
and nonsense mutations, were identified in LGMD2Acolor vision. Science 232:203 – 210
patients, and many others have subsequently been iden-Pentao L, Wise CA, Chinault AC, Patel PI, Lupski JR (1992)
tified. Since the affected patients in La Réunion belongCharcot-Marie-Tooth type 1A duplication appears to arise

from recombination at repeat sequences flanking the 1.5 Mb to a genetic isolate presumed to derive from a single
monomer unit. Nat Genet 2:292 – 300 ancestor who immigrated to the island during the late

Reiter LT, Murakami T, Koeuth T, Pentao L, Muzny D, Gibbs 1670s, it was expected that all affected patients from
RA, Lupski JR (1996) A recombination hotspot responsible La Réunion would have the same LGMD2A mutation.
for two inherited peripheral neuropathies is located near a Paradoxically, six different mutations were identified.
mariner transposon-like element. Nat Genet 12:288 – 297 This paradox led the investigators to propose digenic

Robinson WP, Bottani A, Yagang X, Balakrishnan J, Binkert
inheritance, in which the founder effect is due to an

F, Mächler M, Prader A, et al (1991) Molecular, cytogenetic,
as-yet-unidentified modulating gene (either nuclear orand clinical investigation of Prader-Willi syndrome patients.
mitochondrial) that permits mutations in CANP3 to ex-Am J Hum Genet 49:1219 – 1234
press LGMD2A. This hypothesis does not require theRobinson WP, Lalande M (1995) Sex-specific meiotic recom-
presence of multiple mutations, since the genetic princi-bination in the Prader-Willi/Angelman syndrome imprinted

region. Hum Mol Genet 4:801 – 806 ples of digenic inheritance should apply to all popula-
Rossiter JP, Young M, Kimberland ML, Hutter P, Ketterling tions with LGMD caused by calpain-3 mutations.

RP, Gitschier J, Horst J, et al (1994) Factor VIII gene inver- In the Amish of northern Indiana, Richard et al.
sions causing severe hemophilia A originate almost exclu- (1995) identified a single mutation in CANP3
sively in male germ cells. Hum Mol Genet 3:1035 – 1039 (CGGrCAG, R769Q) in a homozygous state in affected

Weatherall DJ, Clegg JB, Higgs DR, Wood WG (1995) The patients. The authors speculated that the complete pene-
hemoglobinopathies. In: Scriver CR, Beaudet AL, Sly WS,

trance of this disease in the Amish and in the other
Valle D (eds) The metabolic and molecular bases of inherited

LGMD2A pedigrees might also be under the control of adisease, 7th ed. McGraw-Hill, New York, pp 3417 – 3484
second locus. One expectation of the digenic hypothesis

Address for correspondence and reprints: Dr. David H. Ledbetter, Center for would be that some individuals homozygous for the mu-
Medical Genetics, The University of Chicago, 924 East 57th Street, Room R110,

tation would be clinically unaffected (i.e., CK is normalChicago, IL 60637-1470. E-mail: dhl@genetics.uchicago.edu
� 1997 by The American Society of Human Genetics. All rights reserved. and there are no physical findings suggestive of LGMD).
0002-9297/97/6101-0030$02.00 Because of the possible implications in genetic testing

and counseling, we analyzed 580 DNA samples from
Amish individuals in one northern Indiana county forAm. J. Hum. Genet. 61:231 – 233, 1997
the presence of the R769Q mutation, looking for evi-

DNA Studies of Limb-Girdle Muscular Dystrophy dence of phenotypically normal R769Q homozygotes.
Type 2A in the Amish Exclude a Modifying We initiated the countywide screen by first identifying
Mitochondrial Gene and Show No Evidence for a carrier couples. Appropriate informed consent was ob-
Modifying Nuclear Gene tained from all individuals. In order to identify R769Q
To the Editor: carriers in this population, we specifically approached

members of 16 previously studied nuclear LGMD2ALimb-girdle muscular dystrophy type 2A (LGMD2A) is
characterized by slowly progressive muscle weakness, families from this county. We obtained blood samples

/ 9a2d$$jy27 07-09-97 14:20:50 ajhga UC-AJHG



232 Letters to the Editor

Table 1from spouses of already tested siblings of affected indi-
viduals and from siblings of known carriers and their

Results of R769Q-Mutation Screening in Northern Indiana
siblings’ spouses, because within this group there was Amish Couples
an increased likelihood of finding couples in which both

Matingsa No. of Couplesmembers were carriers. When carriers were identified
outside this group, their siblings and their siblings’

/// and /// 73bspouses were genotyped. For couples in which both indi- /// and spouse not tested 133c
viduals were heterozygous for R769Q, blood samples /// and //0 118
were requested from their children, for DNA studies (fig. //0 and spouse not tested (11 deceased) 12

//0 and //0 91) and CK analysis. Previous studies (C. E. Jackson,
Obligate //0 and //0 10dunpublished data) have shown that nonpaternity is rare
/// and 0/0 1in the Amish. All children of couples in which both

parents were homozygous wild type were assumed to a A plus sign (/) denotes wild type; and a minus sign (0) denotes
be homozygous wild type. The results are summarized R769Q mutation.

b In 6 couples the carrier status of both partners was presumedin table 1. An exact gene frequency in this particular
to be ///, on the basis of the /// status of both parents of eachAmish isolate is not known at this time.
individual.In addition, we searched for evidence of possible mod-

c In 107 individuals the carrier status of one of the partners was
ifying mitochondrial effects by studying maternal lin- presumed to be ///, on the basis of the /// status of both parents
eages and mtDNA haplotypes. Genealogical analyses of of that individual.

d Original 10 couples with known affected children.LGMD patients with the R769Q mutation embedded in
the same haplotype revealed 11 maternal lineages for the
23 northern Indiana and the three Pennsylvania Amish
sibships (traceable to ancestors born 150 – 275 years vania lineages and three of the northern Indiana lineages

covering 15 (58%) of 26 sibships, the mtDNA analysesago). This suggests that the mutation came from the
same founder. Samples from 5 of these 11 lineages were revealed four separate previously known northern Euro-

pean haplogroups. These four haplogroups are widelyscreened for mtDNA restriction-site polymorphisms
characteristic of previously identified Caucasian groups dispersed over the Caucasian phylogenetic tree and are

differentiated from each other. Hence, any mtDNA mu-of related haplotypes (haplogroups). The restriction
analyses were conducted on mtDNA PCR products, as tation of relevance to the expression of the phenotype

would have to have arisen independently in each of thedescribed by Torroni et al. (1994). In two of the Pennsyl-
relevant maternal lineages, an unlikely event. These re-
sults strongly suggest that the mitochondrial genetic
background does not play a major role in the type of
LGMD in this population.

The failure to identify any R769Q-homozygous indi-
vidual who was phenotypically normal (i.e., in whom
CK is normal) is evidence against a segregating modi-
fying gene in this population. Since the ratio of affected
to unaffected siblings is as expected within nuclear fami-
lies, the presence of a modifying mitochondrial gene was
considered. Since all individuals within a sibship would
inherit the same mtDNA, the presence of a shared ‘‘per-
missive’’ mitochondrial gene would cause that disorder
to appear as a simple recessive disorder. Our data do
not exclude a fixed second nuclear locus in the northern
Indiana Amish population, nor do they exclude the pos-
sibility of a second locus for digenic inheritance in the
Réunion population. A widespread permissive geneFigure 1 Representative family of five offspring from two carrier
could be present in the Amish population but not inparents in which R769Q-mutation analysis by allele-specific oligonu-

cleotide assay was performed. The first three children are heterozygous the Réunion population. Similar studies in the Réunion
for R769Q. The fourth child is homozygous for R769Q, and the fifth population could help to resolve the question of digenic
child is homozygous wild type. Genomic amplification of exon 22 of inheritance.
CANP3 was performed as described by Richard et al. (1995). The

Since digenic inheritance is unlikely in the Amish pop-allele-specific oligonucleotide primers for R769 and Q769 are 5�-TTA-
ulation, the Réunion paradox still needs to be resolved.CCATGCGGTACGCAGA-3� and 5�-TTACCATGCAGTACG-

CAGA-3�, respectively (the variant nucleotide is underlined). Zlotogora et al. (1996) observed several different muta-

/ 9a2d$$jy27 07-09-97 14:20:50 ajhga UC-AJHG



233Letters to the Editor

Richard I, Broux O, Allamand V, Fougerousse F, Chiannilkul-tions in the genes for metachromatic leukodystrophy
chai N, Bourg N, Brenguier L, et al (1995) Mutations inand Hurler disease in a small geographic location, Lower
the proteolytic enzyme calpain 3 cause limb-girdle muscularGalilee. They suggested that a high mutation rate and
dystrophy type 2A. Cell 81:27 – 40selective advantage of the carriers were responsible for

Torroni A, Lott MT, Cabell MF, Chen Y-S, Lavergne L, Wal-the multiple mutations, and they proposed that a similar
lace DC (1994) mtDNA and the origin of Caucasians: identi-

event may have occurred in the Réunion population.
fication of ancient Caucasian-specific haplogroups, one of

Perhaps there are multiple founders despite the evidence which is prone to a recurrent somatic duplication in the D-
of a single common ancestor in the Réunion families, loop region. Am J Hum Genet 55:760 – 776
or, less likely, perhaps there is in the CANP3 gene a Zlotogora J, Gieselmann V, Bach G (1996) Multiple mutations
mutation-rate increase due to exogenous (differential ex- in a specific gene in a small geographic area: a common

phenomenon? Am J Hum Genet 58:241 – 243posure to mutagens) or endogenous (unequal distribu-
tion of mutator genes) factors. Finally, perhaps there are

Address for correspondence and reprints: Dr. Gerald L. Feldman, Medicalunknown environmental or genetic factors that influence
Genetics and Birth Defects Center, 2799 West Grand Boulevard-CFP4, Detroit,the manifestations of mutations in the gene in the Ré-
MI 48202. E-mail: glfeldman@pol.net

union population. The search for possible modifying *Present affiliation: Department of Pediatrics, Meharry Medical College,
Nashville.genetic or environmental factors should continue, since
� 1997 by The American Society of Human Genetics. All rights reserved.it could disclose both the mechanism of action of mu-
0002-9297/97/6101-0031$02.00

tated CANP3 in LGMD2A and, possibly, factors of
therapeutic importance (Beckmann 1996). Currently,
we are offering this Amish population carrier testing
and genetic counseling, on the basis of both the R769Q
mutation analysis and monogenic autosomal recessive

Am. J. Hum. Genet. 61:233 – 238, 1997
inheritance for LGMD2A.

V. M. PRATT,1,* C. E. JACKSON,1 D. C. WALLACE,2
A C2055T Transition in Exon 8 of the ATP7A Gene IsD. S. GURLEY,2 A. FEIT,1 AND G. L. FELDMAN1
Associated with Exon Skipping in an Occipital Horn1Henry Ford Hospital, Detroit; and 2Emory University
Syndrome FamilySchool of Medicine, Atlanta

To the Editor:
Mutations associated with abnormal splicing accountAcknowledgement
for 10% – 20% of all gene mutations (Krawczak et al.

The authors wish to thank the Amish families for their coop- 1992). Mutations leading to abnormal splicing usually
eration. Dr. Delbert L. Gratz, of Bluffton, OH, provided valu-

are located within the donor or acceptor splice sites.
able genealogical data to the project. We thank J. Double and

However, exonic consensus sequences that are impliedG. Taylor for their field studies, and we thank K. Dziubek for
in the splicing process have been described outside theher technical assistance. This work was supported by a grant
splice sites (Ligtenberg et al. 1990; Steingrimsdottir etfrom the Association Française contre les Myopathies (to
al. 1992). Direct study of the genomic DNA may notG.L.F. and C.E.J.) and by NIH grants HL45572 and NS21328
easily reveal the splicing mutations. However, reverseand a grant from the Muscular Dystrophy Association (all to

D.C.W.). transcription (RT) followed by PCR analysis is likely to
detect such mutations.

In the ATP7A gene, which encodes a copper-trans-
References porting P-type ATPase (Chelly et al. 1993; Mercer et al.

1993; Vulpe et al. 1993), splicing mutations are frequentAllamand V, Broux O, Richard I, Fougerousse F, Chiannilkul-
chai N, Bourg N, Brenguier L, et al (1995) Preferential local- in patients with Menkes disease (Das et al. 1994; Kaler
ization of the limb-girdle muscular dystrophy type 2A gene et al. 1994, 1995; Tümer et al. 1997) and also are de-
in the proximal part of a 1-cM 15q15.1-q15.3 interval. Am scribed in patients with occipital horn syndrome (OHS)
J Hum Genet 56:1417 – 1430 (Kaler et al. 1994; Das et al. 1995). OHS, previously

Beckmann JS (1996) The Réunion paradox and the digenic known as ‘‘X-linked cutis laxa’’ (MIM 304150), is a
model. Am J Hum Genet 59:1400 – 1402 connective-tissue disorder characterized by skin laxity,

Beckmann JS, Richard I, Hillaire D, Broux O, Antignac C,
hyperextensible joints, skeletal anomalies, including oc-

Bois E, Cann H, et al (1991) A gene for limb-girdle muscular
cipital exostoses, and inconstant mild mental retarda-dystrophy maps to chromosome 15 by linkage. CR Acad
tion (Lazoff et al. 1975; Tsukahara et al. 1994). Here,Sci III 312:141 – 148
we report an unusual splice mutation, which wasJackson CE, Strehler DA (1968) Limb-girdle muscular dystro-
detected by RT-PCR, in the ATP7A gene of an OHSphy: clinical manifestations and detection of preclinical dis-

ease. Pediatrics 41:495 – 502 family.

/ 9a2d$$jy27 07-09-97 14:20:50 ajhga UC-AJHG