вход по аккаунту


Association of myopathy with large-scale mitochondrial dna duplications and deletions Which is pathogenic.

код для вставкиСкачать
Association of Myopathy with Large-Scale
Mitochondria1 DNA Duplications and
Deletions: Which Is Pathogenic?
Giovanni Manfredi,*t Tuan Vu,*t Eduardo Bonilla,*l- Eric A. Schon,*tf Salvatore DiMauro,*i
Enrica Arnaudo,*? Lee Zhang,*t Lewis P. Rowland,*? and Michio Hiram*?
We identified large-scale heteroplasmic mitochondrial DNA (mtDNA) rearrangements in a 50-year-old woman with an
adult-onset progressive myopathy. The predominant mtDNA abnormality was a 21.2-kb duplicated molecule. In addition, a small population of the corresponding partially deleted 4.6-kb molecule was detected. Skeletal muscle histology
revealed fibers that were negative for cytochrome c oxidase (COX) activity and had reduced mtDNA-encoded COX
subunits. By single-fiber polymerase chain reaction analysis, COX-negative fibers contained a low number of wild-type
or duplicated mtDNA molecules (ie, nondeleted). In situ hybridization demonstrated that the abnormal fibers contained
increased amounts of mtDNA compared with normal fibers and that most of the genomes were deleted. We concluded
that deleted mtDNA molecules were primarily responsible for the phenotype in this patient.
Manfredi G, Vu T, Bonilla E, Schon EA,DiMauro S, Arnaudo E, Zhang L, Rowland LP, Hirano M.
Association of rnyopathy with large-scale rnitochondrial DNA duplications and deletions:
which is pathogenic? Ann Neurol 1997;42:180-188
Large-scale mitochondrial DNA (mtDNA) rearrangements have been associated with disorders of mitochondrial metabolism affecting muscle and other tissues. Partial mtDNA deletions were first described in
1988 11-31 in patients with Kearns-Sayre syndrome
(KSS), a progressive multisystemic disease characterized
by onset before age 20, ptosis, ophthalmoplegia, pigmentary retinopathy, heart block, ataxia, elevated cerebrospinal fluid (CSF) proteins, and diabetes mellitus
[4, 51. In 1989, Poulton’s group first reported the association of KSS with heteroplasmic mtDNA tandem
duplications [6]. It was later demonstrated that duplications and deletions coexisted in a subset of KSS patients [7,81. Other investigators reported different
clinical phenotypes associated with mtDNA duplications, including renal tubulopathy, cerebellar ataxia,
and diabetes mellitus [9]; chronic progressive external
ophthalmoplegia (CPEO), myopathy, and diabetes
[ 101; and diabetes and deafness [ 1I , 121. Diabetes is a
frequent clinical feature in patients with mtDNA duplications.
Low amounts of a relatively small heteroplasmic
(260-bp) duplication located within the D-loop region
of the mtDNA have been reported in patients with
KSS who also harbored large-scale deletions [ 131.
Higher amounts of the same 260-bp duplication, without the corresponding deletion, were thought to be
pathogenic in a patient with myopathy [ 141. Because
this mtDNA rearrangement has also been reported to
be present in very low abundance in elderly individuals
[15] and in normal whites [16], the pathogenicity of
this particular mtDNA duplication remains unclear.
Unlike mtDNA deletions, which are generally sporadic, duplications are frequently maternally transmitted [9-11]; maternal inheritance was also present in
some of the pedigrees with the 260-bp duplication
[13]. The origin and the pathogenic significance of
large-scale mtDNA duplications also remain unclear.
Herein, we report the unusual observation of a largescale mtDNA tandem duplication and small amounts
of the corresponding deletion in a woman with a pure
myopathy. She did not show any symptoms or signs of
multisystemic involvement and did not have diabetes
mellitus. We also attempted to determine the pathogenic significance of the two species of mtDNA rearrangements.
From the *H. Houston Merritt Clinical Research Center for Muscular Dystrophy and Related Disorders, and Departments of tNeurology and $Generics and Dcvelopment, Columbia University College of Physicians and Surgeons, New York, NY.
Address correspondence to Dr Manfredi, Department of Neurology,
Room 4-431, Columbia University, 630 West 168th Street, Ncw
York, NY 10032.
Received Sep 26, 1996; and in revised form Jan 13, 1997. Accepted
for publication Jan 22, 1937.
Copyright 0 1997 by the American Neurological Association
Patients and Methods
The patient was normal in childhood and through adolescence. She was active in games and sports in school. At age
20, she first noted difficulty in climbing stairs, rising from
low chairs, and running. She complained of myalgias, which
were worse after exercise. These symptoms progressed slowly.
Presently, at age 51, she has difficulty getting in and out of
a car and reaching for objects above her head. She never
complained of dysphagia, dysarthria, ptosis, or double vision.
A neurological examination revealed mild facial and neck
flexor muscle weakness, decreased bulk and strength against
moderate resistance of the proximal arm and leg muscles.
There was no ptosis or ophthalmoparesis. Sensory examination was normal. Tendon reflexes were present. Babinski and
Hoffman signs were absent. Routine laboratory tests, including creatine kinase, blood glucose, glycohemoglobin, and a
glucose tolerance test, were normal. Standard audiologic evaluation was normal. Venous lactate at rest ranged from normal to 3.1 mM (control values, 0.5-2.2 mM). The patient
performed a cycle ergometry test. She attained a workload of
25 W for 58 seconds, before premature interruption of the
rest due to the onset of dizziness. At the exercise peak, heart
rate was 80% of predicted, blood pressure was 184/85 mm
Hg, and oxygen consumption was only 37% of predicted.
With this mild exercise, arterial lactate rose ninefold above
the baseline level. Nerve conduction studies revealed normal
motor and sensory amplitudes and conduction velocities.
Electromyography did not reveal spontaneous activity, and
recruitment patterns were nondiagnostic.
The patient has two sons and a daughter, all of whom are
normal in stature and build. They have no evidence of diabetes, hearing loss, myopathy, ptosis, or ophthalmoparesis.
The patient's two brothers (age, 54 and 57 years) are
healthy. The patient's mother died at age 83 without history
of muscle weakness. The patient's maternal aunt developed a
waddling gait, ptosis, and dysarthria 2 months before death
at age 84. The son of that aunt had a normal neurological
examination at age 62 and died 1 year later of a myocardial
A biopsy was obtained from the deltoid muscle for diagnostic
purposes, with the patient's informed consent. All studies
were conducted on discarded pathology samples under a
Columbia-Presbyterian Medical Center institutional review
board protocol.
Mitochondria1 enzyme activities in the patient's whole
muscle homogenate were measured as described previously
[ 171. Eight-micrometer-thick frozen sections were used for
histochemical analyses. Staining for cytochrome c oxidase
(COX) and succinate dehydrogenase (SDH) activities were
performed as described [ 18, 191. Indirect immunofluorescence staining to detect COX subunit I1 (COX 11) and COX
subunit IV (COX IV) was performed as described [20].
We used restriction fragment length polymorphism
(RFLP) screening of polymerase chain reaction (PCR) products to search for the most commonly encountered pathogenic mtDNA mutations in MELAS (mitochondria1 encephalopathy, lactic acidosis, and stroke-like episodes; mutations
in the tRNAL'"'UUK' gene at nucleotide [nt] 3,243 and nt
3,271), M E W (myoclonus epilepsy with ragged-red fibers
[RRFs]; mutations in the tRNALYsgene at nt 8,344 and nt
8,356), and NARP (neuropathy, ataxia, and retinitis pigmentosa; mutations in the ATPase 6 gene at nt 8,993). PCR
screening for the 260-bp D-loop duplication was performed
according to the protocol described previously 113, 141.
For Southern blot analysis, total DNA was extracted from
muscle and blood [3]. Five micrograms of total DNA was
digested with the restriction enzymes PvuII, BamHI, or
SnaBI. Gel electrophoresis and blotting were performed
as described [3]. Two random-primed j2P-labeled DNA
fragments corresponding to nt 1,227-2,896 and nt 11,68012,570 of the human mtDNA sequence [all were used as
probes 1 and 2, respectively. The relative proportions of the
radiolabeled PCR fragments were quantitated by scanning
the membrane in a PhosplhorImager Model SF by Molecular
Dynamics (Sunnyvale, CA). Quantification of muscle
mtDNA on Southern blots was performed by hybridizing
PvuII-digesred total DNA with "P-labeled probes of a
nuclear-encoded 18s ribosomal DNA fragment and of purified whole human mtDNA, as described [22].
Mapping of the duplication junction point was performed
by PCR amplification of the patient's muscle mtDNA. A
series of PCR reactions was performed using a single backward primer, located at nt 15,870-15,849 [21], together
with four different forward primers (located at nt 3,4703,492, 4,185-4,208, 5,2(;0-5,289, and 5,685-5,703). The
following PCR conditiom were used in all experiments: 25
cycles at 9 4 T , 1 minute; 55°C 1 minute; 7 2 ° C 2 minutes;
10-minute final extension at 72"C, in the presence of
For DNA sequencing, PCR products were isolated from
an agarose gel with a Wizard PCR-preps DNA purification
system (Promega Inc, Madison, WI) and direct sequencing
was performed using an fmole DNA sequencing kit (Promega Inc), according to the manufacturer's protocol.
Single-Fiber PCR
Based on their histochemical phenotype, muscle fibers were
classified as COX-positive (COX')
or COX-negative
(COX-). Dissection and lysis of single muscle fibers were
performed as described [23, 241. The PCR amplification was
conducted using primers P3 (nt 8,657-8,685) and P4 (nr
9,298-9,276), which yielded a 642-bp fragment from the
unduplicatedldeleted mtDNA region (see Fig 4). The amplification products obtained from a single fiber were electrophoresed through a 6% nondenaturing polyacrylamide gel
and subjected to autoradiography.
The amplified fragment was also gel purified from lowmelting agarose gel by using a Wizard PCR-preps DNA purification system and was inserted into a pGEM-t phagemid
bacterial vector from Promega Inc. Tenfold serial dilutions of
the plasmid (in amounts corresponding to 15, 1.5, 0.15, and
0.015 pg of mtDNA), were used to verify the linearity of the
PCR amplification in this range of template DNA concentrations and as standards to estimate the PCR yield. Total
DNA extracted from human cells devoid of mtDNA (po cells
[25]) was added in a 100:l mass ratio (nuclear DNAlplasniid
DNA) to supplement the system with a nuclear genomic
Manfredi et al: Large-Scale Mitochondria1 DNA Rearrangements
background. Before PCR amplification, the D N A standards
were submitted to the same procedures as those used for the
muscle fibers. Amplification and gel electrophorebis of
the standards were performed simultaneously and under the
same conditions as for the single fibers.
I% Situ Hybridization
In situ hybridization was performed as described (261 with
some modifications. In brief, 8-pm-thick muscle sections
were fixed in 4% paraformaldehyde for 45 minutes, washed
in distilled water, and dehydrated in graded alcohol solutions. After a wash in phosphate-bufferrd saline (PBS) containing 5 mM MgCI,, the sections were exposed to proteinase K (10 pgiml) for 45 minutes, PBS washed, acetylated,
and treated with DNase-free RNase from BoehringerMannheim (Indianapolis, IN). The sections were prehybridized for 2 hours at 42°C in a solution containing 50%)formamide, 0.6 mM NaCI, 20 mM Tris-HC1, pH 7.5, 0.02%
Ficoll, 0.02% polyvinylpyrrolidone, 0.12% bovine serum
albumin, 1 mM EDTA, 0.5 mgiml herring sperm DNA,
10% dextran sulfate, 0.5 mg/ml total yeast DNA, and 0.01
pg/ml yeast rRNA. Probes were prepared by PCR using two
pairs of oligonucleotide primers corresponding to mrDNA
positions nt 1,690-1,711 and 2,447-2,426 and to nt 9,1759,198 and 9,764-9,740 (probes 3 and 4, respectively, in Fig
4A). Digoxigenin-dUTP from Boehringer-Mannheim was
incorporated into the probes during PCK. The labeled PCR
products were purified from an agarose gel by using a Wizard PCR-preps DNA purification system and were quantitated by dot blot. Specificity of these probes was confirmed
by Southern blot, The probes were added to the sections and
denaturated at 92°C for 10 minutes. The sections were then
hybridized at 42°C overnight. After washing in 1X salinesodium citrate (SSC) twice, 0.2X SSC, and then in washing
buffer (0.3% Tween 20 in 0.1 M maleic acid buffer, p H
7.5), detection was performed using a Genius Kit from
Boehringer-Mannheim, according to the manufacturer's instructions. Control sections included nondcnatured tissues
and nondenatured probes. Images of sections were captured
with a Dage-MTI CCD-72 digital camera from MTI (Michigan City, IN) and analyzed using an N I H Image software
package (version 1.59). All images were analyzed in one session. Optical density was expressed in arbitrary units, relative
to a standardized gray scale. Serial muscle sections, immediately above and below those used for the in situ hybridization, were stained for COX and SDH activities. Fiber type
was determined from serial sections treated with ATPase
stain at p H 9.4.
Muscle Biochemistry, Morphology, and
Mitochondria1 respiratory chain enzyme activities on
whole muscle homogenate were normal (not shown).
However, the muscle biopsy revealed the presence of
numerous nonatrophic COX- fibers (Fig lB), a proportion of which stained very intensely for SDH (Fig
lA), indicating abnormal mitochondria1 proliferation
(ie, RRFs). We did not observe any COXt RRFs. The
overall percentage of COX- fibers was approximately
10% in the muscle sections examined, which is higher
than age-matched controls, who have < I % [27].
There was no muscle fiber atrophy, necrosis, or inflammation. Besides the mitochondria1 alterations, hematoxylin-eosin, nicotinamide adenine dinucleotidetetrazolium reductase, and modified Gomori trichrome
stains showed normal cytoarchitecture without a loss of
muscle fibers.
Serial muscle sections were stained for COX activity
and with anti-COX I1 (mtDNA-encoded) or antiCOX IV (nuclear DNA-encoded) antibodies. All
COX- fibers showed a strong signal for COX IV,
whereas immunoreactivity for COX I1 was reduced
(Fig 2A-C), indicating a selective decrease of the
mtDNA-encoded polypeptides.
DNA Studies
PCR-RFLP analysis failed to reveal the presence of
known pathogenic point mutations at nt 3,243, 3,271,
8,344, 8,356, or 8,993. The 260-bp D-loop duplication was also absent by PCR.
Southern blot analysis of muscle DNA digested with
PvuII and probed with a radiolabeled DNA fragment
corresponding to mtDNA positions 1,227 to 2,896
(probe 1) revealed two bands, one, 16.6 kb in size, corresponding to wild-type mtDNA, and the other, 4.6
kb in size, corresponding to a mtDNA with an ostensible 12.0-kb deletion (Fig 3). However, with BamHI
(not shown) and SnaBI digestions, the same probe revealed the 16.6-kb band and a 21.2-kb band, whereas
the 4.6-kb band was not detectable (see Fig 3 ) , demonstrating that the rearranged mtDNA was predominantly a 4.6-kb duplication, not a deletion. A band o f
Fig I . Examples of histochemical and in situ hybridization studies on muscle serial sectionr. (A) Succinate dehydrogenase (SOH)
staining illustrates three raged-red fibers (RRFs) (labeled with numbers 1-3). (B) Cytochrome c oxidase (COX) staining shows that
these RRFs are COX-negative (COX--). (C) In situ hybridization to detect mtDNA using probe 3 (I6S, nucleotide [nt] 1,6902,443, corresponding to a nondeleted region of mtDNA. (0)
In situ hybridization to detect mtDNA using probe 4 (COX Ill, nt
9,175-9,764), corresponding t o a deleted region of mtDNA. (E) Optical densities of the numberedfibers and of a sample of normal jbers (with standard errors), expressed as arbitraiy units relative t o a standard gray scale. (F) Example of polymerase chain
reaction (PCR) amplifcation of wild-ype rind duplicated mtDNA (nondeleted mtDNA) fiom COXt and COX- single fibers.
Serial dilutions of a plasmid containing tbt, 642-bp fiagment o f interest were used as standards. C, control PCR without template.
M , molecular weight marker of Haefll-digested X I 73 (sizes in bp at I&).
182 Annals of Neurology
Vol 42
No 2
August 1997
DNA standards, pg
M 15
.015 C
COX+ fibers
COX- fibers
Manfredi et al: Large-Scale Mitochondria1 DNA Rearrangements
probe 1
probe 2
Fig 3. Southern blot analysis of total DNA isolated fiom a
normal individual; muscle and from the patient j muscle,
digested with Pvu I1 and SnaBI, and probed with a radiolabeled DNA fiagment corresponding to nucleotide (nt) 1,2272,896 (probe I ) and to nt 11,680-12,570 (probe 2) of the
human mtDNA sequence. U, uncut DNA. M, molecular
weight marker of HindIII-digested phage h fiizes at left).
Sizes of wild-type and rearranged mtDNAs are indicated at
right, in kb. 166 kb = wild-type mtDNA; 21.2 kb = duplicated mtDNA; 4.6 kb = fragment derivedfiom PvuII
digestion of both duplicated and deleted mtDNA species. A
band migrating above the wild-type mtDNA in the patient;
U and SnaBI lanes is indicated (7.This band is detectable
after hybridization with probe 1, but not with probe 2, indicating that it corresponds to uncut deleted mtDNA monomers
or dimers.
Fig 2. Serial sections from the patient? muscle. (A) Cytochrome c oxidase (COX) staining (B) Imnzunostaining with
anti-COX IV subunit antibodies. (C) Imwcunostaining with
anti-COX II subunit antibodies. The indicated COX-negative
fibers (stars) reacted intensely with anti-COX IV antibodies,
but they were negative with anti-COX II antibodies.
184 Annals of Neurology
Vol 42
No 2
August 1997
weak intensity was observed above the wild-type and
the duplicated bands, Probably corresponding to an
uncut molecule (see Fig 3). The membrane was subsequently stripped and reprobed with a radiolabeled
DNA fragment corresponding to nt 11,680-12,570
(probe 2). This probe revealed the 16.6-kb wild-type
and the 21.2-kb duplicated molecules, whereas the
more slowly migrating bands were not detectable (see
Fig 3), indicating that they were deleted molecules,
possibly nicked circles or deletion dimers. Therefore,
the 4.6-kb bands, which was revealed after PvuII digestion and hybridization with probe 1, represented
mainly the portion of the duplicated region between
the two PvuII sites and corresponded to the size of the
duplication (Fig 4A). In addition, a small proportion
of this 4.6-kb band was derived from the linearized deleted molecules.
Of the patient’s muscle mtDNA, 40% was duplicated and the deleted molecules were approximately
1%. In the patient’s blood, the duplication was too low
for accurate quantitation and deletions were undetectable by Southern blot analysis (not shown). The total
mtDNA content in the patient’s muscle was normal
compared with aged-matched controls (not shown).
The proportion of the duplication was 78% in the patient’s son’s blood, but no deletion was detectable (not
shown). Blood mtDNA from the patient’s other two
offspring, her two brothers, and her maternal uncle did
not reveal any large-scale rearrangements by Southern
blot. We were unable to procure skeletal muscle from
these relatives.
PCR analysis using the backward primer at nt
15,870-15,849 (primer 2 [P2]) and three forward
primers located at nt 4,185-4,208, 5,260-5,289, and
5,685-5,709 failed to generate an amplification product, because the DNA segment between the forward
and the backward primers was too large to amplify
(1 1,685, 10,610, and 10,185 bp, respectively); however, using P2 in conjunction with a forward primer
located at nt 3,470-3,492 (primer 1 [Pl]), a 431-bp
DNA fragment was amplified, presumably from a rearranged mtDNA molecule (see Fig 4A). PCR did not
yield any product from the wild-type mtDNA, because
Fig 4. (A) Diagrams of the wild-iype and of the rearranged
mtDNA molecules. Shown are circular maps of the duplicated
mtDNA (Dup-mtDNA) and its corresponding deleted mtDNA
(A-mtDNA) a~ well as the wild-type mtDNA (WT-mtDNA).
The A-mtDNA harbors a 11,969-bp deletion ftviangle) that
encompasses the origin of light-strand replication (OJ. The
pie sections” in the Dup-mtDNA represent the 4,GOO-bp regions that are duplicated in tandem. Only the genes involved
in the rearrangement junctions (ie, subunit 1 of NADHdehydrogenase-coenzyme Q-reductase ( N D l ] and cytochrome b
(Cyt b]) are shown. Poiymerase chain reaction primers (PI,
nucleotide [nt] 3,470-3,492; P2, nt 15,870-15,849; P3, nt
8,657-8,685; P4, nt 9,298-9,276) are indicated. Pvuh’,
BamHI, and SnaBI restktion sites are shown. The probes
used for the Southern blots and in sitn hybridizations (numbers in boldfice) are also indicated. Map is not to scale.
(B) DNA sequence (G, A, T, C) of the duplication junction
region. The 10-bp sequence common to N D l and Cyt b
is boxed; it corresponds to nt 3,568-3,577 in the N D l
gene and to nt 15,537-15546 in the Cyt b gene.
- 3567
Manfredi et al: Large-Scale Mitochondria1 DNA Rearrangements
the DNA segment between P1 and P2 was too large
(12,130 bp) to amplify.
DNA sequencing of the 431-bp fragment showed
normal sequence until nt 3,577, at which point the
N D 1 gene sequence was followed by Cyt b gene sequences commencing at nt 15,547 (see Fig 4 B ) . The
rearrangement junction contained one of two copies of
a 10-bp perfect direct repeat located, in wild-type
mtDNA, at nt 3,568-3,577 in the ND1 gene and at
nt 15,537-15,546 in the Cyt b gene (see Fig 4B).
PCR screening using primers P1 t- P2 (see Fig 4A),
which were specific for the rearranged region, amplified the abnormal 431-bp fragment only from the
muscle and blood of the patient and from the blood of
her youngest son, but not from blood of the other two
offspring, two brothers, or maternal uncle of the patient.
Single-fiber PCR analysis was used to determine
whether there was any difference in the content of
wild-type plus duplicated mtDNA (nondeleted
mtDNA) between COX’ and COX- fibers. Results of
a typical single-fiber PCR experiment are shown in
Figure 1F. In the experiment shown in Figure 1, the
standard amplification curve obtained from plasmid
DNA containing the fragment of interest showed that,
for DNA amounts between 0.0 15 and 15 pg, the yield
of the PCR amplification was linearly correlated with
the template amount. All of the COXt fibers had between 0.15 and 1.5 pg of nondeleted mtDNA, whereas
the content of nondeleted mtDNA in the COX- fibers
was approximately 10-fold lower, between 0.0 15 and
0.15 pg. In most COX- fibers amplified (25 in total),
the content of nondeleted mtDNA was severalfold
lower than in the COXt fibers (total 16). In three
COX- fibers, the content o f nondeleted mtDNA was
similar to that in COXt fibers (not shown). In these
single-fiber analyses, the amount of deleted molecules
was not measured.
Hybridization of muscle sections with probe 3,
which recognized all mtDNA species (see Fig 4A),
showed an intense signal in all COX- fibers examined
(example shown in Fig 1C). The same fibers showed a
very weak hybridization signal with probe 4 (see Fig
ID), which corresponded to a deleted region of the
mtDNA (see Fig 4A). Normal C O X ’ fibers did not
show any significant hybridization difference with the
two probes. Densitometric analysis revealed that the intensity of the hybridization with probe 3 was two- to
threefold mote intense in rhe COX- fibers than in
both type 1 and type 2 normal fibers. Probe 4 showed
an inverse ratio; ie, COX- fibers weie two- to threefold less intense than in both type 1 and type 2 normal
fibers (see Fig 1E). These results indicated that the total amount of mtDNA was increased in the COXfibers and that most mtDNA molecules in these fibers
were deleted.
186 Annals of Neurology
Vol 42
No 2
A u g h t 1997
We identified a patient with clinical, biochemical, and
histological features of a mitochondrial myopathy who
harbored heteroplasmic mtDNA duplications and deletions.
This patient has unique clinical features compared
with others with mitochondrial disorders in which the
coexistence of duplicated mtDNA and its corresponding deletion has been reported [6-8, 11, 121. She has
a pure myopathy and does not have diabetes mellitus,
deafness, or other multisystemic manifestations, nor
does she have KSS. Diabetes mellitus was reported in
15 of 20 patients wirh mtDNA duplications and deletions [6, 9, 10-131, suggesting that pancreatic islets
might be especially vulnerable to the metabolic impairment associated with these mtDNA rearrangements.
The absence of diabetes in our patient could be explained by skewed mitotic segregation of the rearranged molecules during embryogenesis (ie, the pancreatic islets might have received an amount of abnormal
genomes insufficient to cause dysfunction).
An unbalanced distribution of the rearrangement in
different tissues of the proband was also suggested by
the very low abundance of duplicated genomes in
blood compared with muscle. This was another unusual finding in this patient, because most reported patients contained higher levels of duplicated genomes in
blood than in muscle [9, 10, 12, 131. We could not
assess whether the mtDNA rearrangement was inherited or arose spontaneously in our patient, because her
mother was already deceased at the time of this study.
Nevertheless, the duplication was found in the blood
of the 20-year-old son of the proband, demonstrating
that this mtDNA rearrangement can be transmitted
through the germline. Maternal inheritance of a mitochondrial disorder associated with mtDNA duplications has been described in two unrelated families [ l o 121. In contrast to these reports, our patient’s son was
asymptomatic, despite high levels of mtDNA duplication (78%) in blood. We could not establish whether
his lack of muscular symptoms was due to the absence
of deleted genomes in muscle, because that tissue was
not available. It is also possible that he is presymptomatic, because his mother had been free of symptoms
until age 20.
What is the relative contribution of these two genetic abnormalities (ie, deletions and duplications) in
the pathogenesis of mitochondrial disorders? The theory that large-scale deletions can be pathogenic in humans has been supported extensively both in vivo, by
in situ hybridization [28, 291 and single-fiber PCR experiments [24] in muscle, and in vitro, by experiments
on somatic cell hybrids [3O] and transmitochondrial
hybrids [31] derived from tissues from affected patients. It has been postulated that mtDNA deletions
impair mitochondria] protein synthesis due to the loss
of tRNA genes [32]. By contrast, the pathogenic significance of mtDNA duplications is still uncertain, and
the tRNA hypothesis does not apply to duplications, as
there is no loss of mtDNA genes and presumably no
mutation of tRNA sequences [33].
T o help understand the relative pathogenic roles of
mtDNA duplications and deletions in our patient, we
explored the relationship between the distribution of
the rearranged molecules and the histochemical phenotype in muscle fibers. We studied COX activity and
COX subunit immunoreactivities in combination with
single-fiber PCR analyses and in situ hybridization.
Single-fiber PCR analyses demonstrated that the content of wild-type and duplicated (nondeleted) mtDNA
was lower in the COX- fibers than in the COXf fibers. This result could be explained either by depletion
of mtDNA or by accumulation of deleted genomes in
the COX- fibers. However, by in situ hybridization,
we demonstrated that the absolute mtDNA content in
the histochemically abnormal fibers was increased and
that most mitochondrial genomes in these fibers were
deleted. Thus, it appears that the deletions were the
primary cause of the mitochondrial myopathy in this
patient. It was therefore surprising that the duplication,
which was, by far, the predominant mtDNA rearrangement found in the patient’s muscle (comprising 40%
of the total mtDNA) was apparently unrelated to the
phenotype, whereas the deleted genomes, which only
comprised 1% of the total, could apparently cause the
disease. Given the uneven distribution of the rearranged mtDNA molecules in different tissues and in
cells within a single tissue, as demonstrated by in situ
hybridization studies, it is likely that different muscles
or even different portions of a single muscle harbor
varying amounts of deleted and duplicated mtDNA.
Moreover, the segmentary nature of the morphological
alterations in skeletal muscle may underestimate the
overall number of fibers affected. In other words, the
proportion of histologically abnormal muscle fibers and
the low percentage of deleted mtDNA molecules found
in the muscle biopsy might not be fully representative
of the extent of the muscle involvement in this patient.
A hypothetical pathogenic role of the duplications
might be the generation of deleted molecules; because
the minimal size of a mtDNA molecule containing
both the origin of heavy-strand replication (0,)
the origin of light-strand replication (0,)
must be at
least 5.7 kb, a 4.6-kb deleted mtDNA, lacking 0,,
may be unable to replicate. This is supported by two
lines of evidence. First, the family described by us and
the one described by Ballinger and colleagues [I 1, 121
are the only ones in whom deleted mtDNA molecules
lacking 0, were detected, suggesting that this kind of
mtDNA rearrangement is extremely rare. Second, Ballinger and colleagues [I21 observed that Epstein-Barr
virus-transformed lymphoblastoid cells, which origi-
nally contained both deletions and duplications, lost
the deleted mtDNA during subsequent passages [ 121.
This observation reinforces the hypothesis that replication of deleted mtDNA molecules lacking 0, may be
severely disadvantaged, at least in rapidly dividing cells.
However, the deleted molecules detected in these patients might have arisen from the duplicated mtDNAs,
via further recombination events [7].In this way, duplicated genomes might provide a continuing source of
deletions, which could slowly accumulate in nonreplicating tissues, such as mature skeletal muscle. Thus, a
low rate of accumulation of deleted molecules could
explain the late onset and the relatively mild but progressive course of the muscle disorder in our patient.
In conclusion, we describe a patient with pure mitochondrial myopathy associated with a heteroplasmic
population of tandem duplications and deletions of
mtDNA. The proportion of the molecular rearrangements was higher in muscle than in blood, suggesting
that mitotic segregation occurred at an early stage during embryogenesis, sparing other tissues, such as the
pancreatic islets. An alternative explanation would be a
clonal selection of rapidly dividing stem cells harboring
normal mtDNA in the patient’s bone marrow. Although representing a small fraction of the rearrangements, the deletions appeared to have a disproportionate impact on the phenotype in this patient. To our
knowledge, this is the first investigation of the relative
pathogenic roles of duplicated and deleted mtDNAs.
Further studies are necessary to assess if the duplicated
mtDNA molecules play a more subtle role in the
pathogenic process, perhaps by contributing to the
genesis of the deletions.
This study was supported by grants from the Muscular Dystrophy Association, the National Institutes of Health (HD32062,
NS28828, NS11766, AG12131, HD32062, and NSO1617), Telethon Italy, the Myoclonus Research Foundation, the Procter &
Gamble Company, and the Dana Foundation. The Irving Clinical
Research Center also supported this project.
W e thank Drs Filippo M. Santorelli, Bernard Fromenty, and
Francesco Pallotti for their expert advice. We appreciate Dr Rochelle Goldsmith’s cardiopulmonary exercise evaluation.
1. Holt IJ, Harding AE, Morgan-Huges JA. Deletion of muscle
mitochondrial DNA in patients with mitochondrial myopathies. Nature 1988;331:717-71 7
2. Lestienne P, Ponsot G. Kearns-Sayre syndrome with muscle
mtDNA deletion. Lancet 1988;1:885 (Letter)
3. Zeviani M , Moraes C T , DiMauro S, et al. Deletions of mitochondrial DNA in Kearns-Sayre syndrome. Neurology 1988;38:
4. Kearns TP, Sayre GP. Retinitis pigmentosa, external ophthalmoplegia and complete heart block. Usual syndrome with histologic study in one of two cases. Arch Ophthalmol 1958;60:
5. Berenberg RA, Pellock JM, DiMauro S, et al. Lumping or split-
Manfredi et al: Large-Scale Mitochondria1 DNA Rearrangements
ring? Ophthalmoplegia-plus or Kearns-Sayre syndrome? Ann
Neurol 1976;1:37-54
6. Poulton J, Deadman ME, Gardiner RM. Duplications of mitochondrial DNA in mitochondrial myopathy. Lancet 1989;l:
7. Poulton J, Morten KJ, Weber K, et al. Are duplications of mitochondria] DNA characteristic of Kearns-Sayre syndrome?
H u m Mol Genet 1994;3:947-951
8. Brockington M, Alsanjari N , Sweeney M G , et al. Kearns-Sayre
syndrome associated with mitochondrial DNA deletion or
duplication: a molecular genetic and pathological study. J Neurol Sci 1995;131:78-87
9. Rotig A, Bessis JL, Romero N , et al. Maternally inherited duplication of the mitochondrial genome in a syndrome of proximal tubulopathy, diabetes mellitus, and cerebellar ataxia. Am J
Hum Genet 1992;50:364-370
10. Dunbar DR, Moonie PA, Swingler RJ, er al. Maternally transmitted partial direct tandem duplication of mitochondrial DNA
associated with diabetes inellitus. H u m Mol Genet 1993;2:
16 19-1624
11. Ballinger SW, Shoffner JM, Hedaya E, et al. Maternally transmitted diabetes and deafness associated with a 10.4 kb mitochondrial DNA deletion. Nature Genet 1992;l:l 1-15
12. Ballinger SW, Shoffner JM, Gebhart S, et al. Mitochondrial
diabetes revisited. Nature Genet 1994;7:458-459 (Letter)
13. Brockington M, Sweeney M , Hammans !;, et al. A tandem duplication in the D-loop of human mitochondrial DNA is associated with deletions in mitochondrial myopathies. S a t Genet
1993;4:67-7 1
14. Manfredi G , Servidei S, Bonilla E, et al. High levels of mitochondrial DNA with an unstable 260-bp duplication in a patient with a mitochondrial myopathy. Neurology 1995;45:762768
15. Lee H C , Pang CY, Hsu HS, Wei YH. Ageing-associated tandem duplications in the D-loop of mitochondrial DNA of human muscle. FEBS Lett 1994;354:73-8.1
16. Torroni A, Lort MT, Cabell MF, et al. Mitochondrial DNA
and the origin of Caucasians. Identification of ancient caucasianspecific haplogroups, one of which is prone to a recurrent somatic duplication in the D-loop region. Am J H u m Genet
17. DiMauro S, Servidei S, Zeviani M, et al. Cytochrome c oxidase
deficiency in Leigh syndrome. Ann Neurol 1987;22:498-506
18. Seligman AM, Karnovsky MJ, Wasscrl.:rug HL, Hanker JS.
Nondroplet ultrastructural demonstration of cytochrome oxidase activity with a polymerizing osmiophiiic reagent, diaminobenzidine (DAB). J Cell Biol 1968;38:1--14
19. Dubowirz UK, Brooke M H . Muscle biopsy: a modern approach. Philadelphia: WB Saunders, 1973
20. Tritschler HJ, Bonilla E, Lombes A, et al. Differential diagnosis
of fatal and benign cytochrome c oxidase-deficient myopathies
188 Annals of Neurology
Vol 42
’No 2
August 1997
of infancy: an irnmunohistochemical approach. Neurology
1991;4 1:300-305
21. Anderson S, Bankier AT, Barrel BG, et al. Sequence and organization of the human mitochondrial genome. Narure 1981;
22. Moraes C T , Shanske S, Tritschler HI, et al. MtDNA depletion
with variable tissue expression: a novel genetic abnormality in
mitochondrial diseases. Am J H u m Genet 1991;48:492-501
23. Moraes CT, Ricci E, Bonilla E, et al. The mitochondrial
tRNALeu(UUR) mutation in mitochondrial encephalomyopathy, lactic acidosis. and strokelike episodes (MELAS): genetic,
biochemical, and morphological correlations in skeletal muscle.
Am J Hum Genet 1992;50:934-949
24. Sciacco M, Bonilla E, Schon EA, et al. Distribution of wildtype and common deletion forms of mtDNA in normal and
respiration-deficient muscle fibers from patients with mitochondrial myopathy. H u m Mol Genet 1994;3:13-19
25. King MP, Attardi G. Human cells lacking m t D N A repopulation with exogenous mitochondria by complementation. Science 1989;246:500 -503
26. Muller-Hocker J, Seibel P, Schneiderbanger K, Kadenbach B.
Different in situ hybridization patterns of mitochondrial DNA
in cytochrome c oxidase-deficient extraocular muscle fibers in
the elderly. Virchows Arch [A] 1993;422:7-15
27. Schon EA, Sciacco M, Pallotti F, et al. Mitochondrial DNA
mutations and aging. In: Holbrook NI, Martin GR, Lockshin
RA, eds. Cellular aging and cell death. New York: Wiley-Liss,
28. Mita S, Schmidt B, Schon EA, et al. Detection of “deleted”
mitochondrial genomes in cytochrome c oxidase-deficient muscle fibers of a patient with Kearns-Sayre syndrome. Proc S a t 1
Acad Sci USA 1989;86:9509-9513
29. Shoubridge EA, Karpari G, Hastings KE. Deletion mutants are
functionally dominant over wild-type mitochondrial genomes
in skeletal muscle fiber segments in mitochondria1 disease. Cell
30. Sancho S, Moraes CT, Tanji K, Miranda AF. Structural and
functional mitochondrial abnormalities associated with high
levels of partially deleted mitochondrial DNAs in somatic cell
hybrids. Somat Cell Mol Genet 1992;18:431-442
31. Hayashi J-I, Ohta S, Kagawa Y, et al. Functional and morphological abnormalities of mitochondria in human cells containing
mitochondrial DNA with pathogenic point mutations in tRNA
genes. J Biol Chein 1994;269:19060-19066
32. Nakase H, Moraes CT, Rizzuto R, et al. Transcription and
translation of deleted mitochondrial genoines in Kearns-Sayre
syndrome: implications for pathogenesis. Am J H u m Genet
33. Lander ES, Lodish H . Mitochondrial diseases: gene mapping
and gene therapy. Cell 1990;61:925-926
Без категории
Размер файла
1 009 Кб
scala, myopathy, large, associations, pathogens, dna, deletion, mitochondria, duplication
Пожаловаться на содержимое документа