(Received for publication, October 17, 1996, and in revised form, February 18, 1997)
From the A nuclear genome delivery system was developed to
deduce the modes of inheritance of the clinical phenotypes observed in
patients with mitochondrial diseases by transfer of pure nuclei from
normal cells to fibroblasts from the patients. The problem of possible contamination of the nuclei with a small amount of mtDNA was overcome by using mtDNA-less ( Numerous disorders, including mitochondrial encephalomyopathies,
cardiomyopathies, and aging, have been reported to be associated with
defects of mitochondrial respiratory function, which is under the
control of both nuclear and mitochondrial genomes (1-4). For example,
nuclear DNA encodes most mitochondrial proteins including factors
necessary for replication and expression of the mitochondrial genome.
On the other hand, mtDNA encodes 13 polypeptides, which are subunits of
respiratory chain complexes and oligomycin-sensitive mitochondrial
ATPase, and two rRNAs and 22 tRNAs necessary for their translation in
mitochondria. Therefore, mutations in both genomes could contribute to
mitochondrial dysfunction and to the pathogenesis of these
mitochondrial disorders.
Mitochondrial encephalomyopathies (1) were shown to be closely
associated with heteroplasmy of wild-type mtDNA and mutant mtDNA with
large scale mtDNA deletions including several tRNA genes (5), with
point mutations of the mitochondrial tRNALeu(UUR) gene (6),
and with point mutations of the mitochondrial tRNALys gene
(7), respectively. The intercellular co-transfer of the mutant mtDNA
and mitochondrial dysfunction into human cell lines lacking mtDNA
( In most mitochondrial encephalomyopathies, however, the candidate
mutations of mtDNA have not been identified. In these cases, even if a
patient's mtDNA and mitochondrial dysfunction are not co-transferred
to On the contrary, intercellular transfer of pure nuclei from normal
cells to cells from patients should be an effective system for
determining which genome, mitochondrial or nuclear, is responsible for
pathogenesis. Moreover, with this system it is possible to determine
whether a defect is inherited in a nuclear dominant or nuclear
recessive way. For example, if mitochondrial dysfunction in cells from
a patient is restored by introduction of normal nuclei, the defect can
be attributed to nuclear recessive mutations, whereas if it is not, the
defect should be due to mtDNA mutations or nuclear dominant mutations.
However, the problem with this system is the preparation of pure nuclei
completely free from mtDNA, since even a small amount of wild-type
mtDNA is known to compensate for the influence of a predominant
pathogenic mutant mtDNA and help in maintaining normal mitochondrial
translation activity (8-13). This problem can be overcome by using
In this study, HeLa nuclei completely free from mtDNA were introduced
into 553 fibroblasts from a patient with a fatal mitochondrial disease
by fusion of the fibroblasts with The human skin fibroblast line 553 was isolated from a 3-month-old patient with a fatal mitochondrial
disease causing congenital COX and SDH
deficiencies.2 A normal human skin
fibroblast line TIG3S (15) from a fetus was obtained from Dr. H. Kondo,
Department of Biology, Tokyo Metropolitan Institute of Gerontology. The
Intercellular transfer of HeLa nuclei to TIG3S or 553 fibroblasts was achieved by fusion of the fibroblasts with
Total DNA (1 µg)
extracted from 2 × 105 cells was digested with a
single cut restriction enzyme, BamHI, and the restriction fragments were separated by 0.6% agarose gel electrophoresis. After
blotting onto a NYTRAN membrane, the DNA fragments were hybridized with
[ Total
cellular RNA was extracted with an ISOGEN RNA isolation kit (Nippon
Gene, Toyama, Japan). Total denatured RNA (10 µg) was electrophoresed
in 1.1% agarose gel containing formaldehyde and then transferred to a
NYTRAN membrane. The membrane was hybridized with the
[ Mitochondrial translation products were labeled with
[35S]methionine as described (16) with slight
modifications. Briefly, semiconfluent cells in a dish were incubated in
methionine-free medium containing 10% bovine serum for 1 h at
37 °C. Then the cells were labeled with
[35S]methionine for 1 h in the presence of emetine
(0.15 mg/ml). The mitochondrial fraction (50 µg/lane) was separated
by SDS-urea-polyacrylamide gel electrophoresis. The dried gel was
exposed to an imaging plate for 6 h, and the labeled polypeptides
were located with a bioimaging analyzer, Fujix BAS 2000.
The activities of the mitochondrial respiratory
enzymes COX (complex IV), oligomycin-sensitive mitochondrial ATPase
(complex V), and SDH were measured as described in Refs. 17-19,
respectively.
The chromosome compositions of cybrid
clones were analyzed immediately after cloning using air-dried
chromosome preparations as described previously (20).
The human skin fibroblast line 553 was isolated from a 3-month-old
patient with a fatal mitochondrial disease. Since the patient showed
progressive COX and SDH deficiencies in skeletal muscles, we examined
whether 553 fibroblasts from the patient also expressed these clinical
phenotypes. Biochemical studies showed that COX and SDH activities of
553 fibroblasts were about 22 and 10%, respectively, of those of
normal skin TIG3S fibroblasts from a fetal subject (Fig.
1, a-b).
The observed COX deficiency in 553 fibroblasts could be due to
reduction of the amount of mtDNA or to down-regulation of
transcriptional or translational activity in mitochondria, because the
three COX subunits COI, COII, and COIII are encoded by mtDNA. To
examine these possibilities, we first compared the total amounts of
mtDNA in 553 and TIG3S fibroblasts by Southern blot analysis after
digestion of total DNA samples with a single-cut restriction enzyme,
XhoI. Results showed that the amounts of mtDNA in these two
fibroblast lines were similar (Fig. 2a).
On the contrary, Northern blot analysis revealed very different
transcription profiles of 553 fibroblasts from those of normal TIG3S
fibroblasts. Fig. 2b shows that the processings of
mitochondrial rRNA and mRNA in 553 fibroblasts were not sufficient,
resulting in the productions of increased amounts of unprocessed
mitochondrial transcripts and reduced amounts of mature 16 S rRNA
(68.5%) and COI mRNA (28.2%), whereas the amounts of mRNAs of
nuclear genome-encoded subunits of oligomycin-sensitive ATPase and
mtTFA were essentially normal.
Next we examined mitochondrial translation activity using
[35S]methionine incorporation in the presence of emetine
to determine whether the observed abnormality in processing of
mitochondrial transcripts in 553 fibroblasts affected the amounts of
newly synthesized polypeptides encoded by mtDNA. Fig. 3
shows that the incorporations of [35S]methionine into
most mtDNA-encoded polypeptides, including COI, COII, and COIII, in 553 fibroblasts were about 30% of those in TIG3S fibroblasts. Therefore,
the reduced COX activity in 553 fibroblasts might be due at least
partly to incomplete processing of mitochondrial transcripts, resulting
in reduced translation activity in the mitochondria.
The profile of mitochondrial translation in 553 fibroblasts also showed
another unexpected abnormality, namely the amounts of
[35S]methionine incorporation into ATP6 and ATP8
polypeptides were slightly increased over those in control TIG3S
fibroblasts (Fig. 3). Moreover, the amounts of their mature mRNAs
were also increased correspondingly (Fig. 2b). These
observations predicted slight increase of oligomycin-sensitive
mitochondrial ATPase activity in 553 fibroblasts, and the results of
biochemical analysis shown in Fig. 1c confirmed this
prediction.
Since subunits of COX and oligomycin-sensitive mitochondrial ATPase are
encoded by both nuclear and mitochondrial genomes, and since
mitochondrial protein synthesis is under the control of both genomes,
it was necessary to determine which genome was responsible for the
reduced COX activity and the elevated ATPase activity in the patient's
fibroblasts. Furthermore, although the subunits of SDH are all encoded
by the nuclear genome, it was necessary to determine whether the
pathogenic mutations causing low SDH activity were recessive or
dominant.
For these purposes, pure HeLa cell nuclei were introduced into the 553 fibroblasts and the control TIG3S cells, respectively. As
Table I.
Somatic cell genetic characteristics and chromosome complement of
parent cells and their nuclear hybrid clones
Institute of Biological Sciences,
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
0) human cells as nuclear
donors. In this study, intercellular transfer of pure nuclei was
carried out by simple fusion of
0 HeLa cells with 533 fibroblasts from a patient with a fatal mitochondrial disease, which
were deficient in cytochrome c oxidase and succinate dehydrogenase activities. The results showed that the cytochrome c oxidase and succinate dehydrogenase activities were
restored by the introduction of pure HeLa nuclei, suggesting that the
observed phenotypes of mitochondrial dysfunction were not due to mtDNA mutations but to nuclear, recessive mutations. Thus, our nuclear transfer system is effective for determining whether a mitochondrial or
nuclear genome of a patient is responsible for a disease and whether
deficiency of mitochondrial enzymes, including enzymes exclusively
encoded by nuclear genomes, is transmitted in a nuclear recessive or
nuclear dominant way, providing the parents of the patients with
valuable information for genetic counseling on the risk of
mitochondrial diseases in their next babies.
0 cell lines) provided unambiguous evidence that
accumulation of these mutant mtDNA molecules alone without defects in
the nuclear genome was sufficient to produce the pathogeneses leading
to the diseases (8-11).
0 cells, mtDNA mutations could be responsible for the
pathogeneses for the following reasons. First, during selection to
remove
0 cells using medium without pyruvate and
uridine, mtDNA-repopulating cells (cytoplasmic hybrids; cybrids)
expressing mitochondrial dysfunction due to pathogenic mtDNA mutations
might also be preferentially excluded together with
0
cells, so that the isolated cybrid clones would be limited to those
with normal mitochondrial respiratory function. Furthermore, the
possibility that the amount of mtDNA with unidentified pathogenic mutations is not sufficient to induce mitochondrial dysfunction in the
cybrids cannot be excluded. Therefore, intercellular transfer of mtDNA
from a patient to
0 cells is not appropriate for
determining the involvement of mtDNA mutations in the pathogenesis,
particularly when candidate pathogenic mutations of mtDNA in the
patient have not been identified.
0 cells as nuclear donors; simple fusion of cells from a
patient with
0 cells results in intercellular transfer
of pure normal nuclei uncontaminated by any mtDNA from the nuclear
donor cells.
0 HeLa cells (14)
without using any selective pressure upon mitochondrial respiratory
function to remove
0 HeLa cells. The 533 fibroblasts
showed very low COX1 and SDH activities,
and intercellular transfer of nuclei from
0 HeLa cells
to the 533 fibroblasts restored their reduced COX and SDH activities,
suggesting that these clinical phenotypes were not due to mtDNA
mutations but to nuclear recessive mutations.
Cells and Cell Culture
0 HeLa cells are resistant to 20 µM
6-thioguanine and 2 mM ouabain (Oua) (8). The fibroblast
lines and
0 HeLa cells were grown in normal medium (RPMI
1640 + pyruvate (0.1 mg/ml) + uridine (50 µg/ml) + 10% fetal bovine
serum).
0 HeLa cells using polyethylene glycol 1500 (Boehringer
Mannheim) as described previously (13) with slight modifications.
Briefly, cells in the fusion mixtures were cultivated in selective
medium (RPMI 1640 + pyruvate (0.1 mg/ml) + uridine (50 µg/ml) + 10%
fetal bovine serum + hypoxanthine/aminopterin/thymidine (HAT; Sigma) + 2 mM Oua). On day 10 after fusion, colonies grown in the
selection medium were cloned by the cylinder method, and the nuclear
hybrid clones were then cultivated in the normal medium.
-32P]dATP-labeled HeLa mtDNA. The radioactivity of
the fragments was measured with a bioimaging analyzer, Fujix BAS 2000 (Fuji Film, Tokyo).
-32P]dATP-labeled oligonucleotide probes COI
(nucleotide positions 5847-7746), ATP6 (8704-9020), and 16 S rRNA
(1740-3130) and with [
-32P]dATP-labeled cDNA
probes of mtTFA and ATPase F1a. The radioactivities of the
bands were measured a bioimaging analyzer, Fujix BAS 2000.
Fig. 1.
Expressions of clinical phenotypes in
fibroblasts from a patient with mitochondrial disease. The
mitochondrial respiratory enzyme activities of COX (a), SDH
(b), and oligomycin-sensitive mitochondrial ATPase
(c) in TIG3S fibroblasts from a normal subject (TIG) and 553 fibroblasts from a patient (553)
are compared.
[View Larger Version of this Image (11K GIF file)]
Fig. 2.
Comparison of the contents of mtDNA and its
transcripts in fibroblast line TIG3S derived from a normal subject
(TIG) and 553 from a patient with mitochondrial disease
(553). a, Southern blot analysis of a
BamHI restriction fragment of mtDNA. Total DNA (5 µg/lane)
extracted from fibroblasts was analyzed using [-32P]dATP-labeled HeLa mtDNA as a probe.
b, Northern blot analysis. Total RNA (10 µg/lane)
extracted from the fibroblasts was separated on agarose/formaldehyde
gel and blotted onto filters. Hybridization was carried out with
various probes: nuclear DNA-encoded mtTFA, the a subunit of
oligomycin-sensitive mitochondrial ATPase, mtDNA-encoded 16 S rRNA,
the COI subunit of COX and the ATP6 subunit of oligomycin-sensitive mitochondrial ATPase. The relative contents of the mature mRNAs (indicated by arrowheads) in 553 fibroblasts were 87.0, 105.1, 68.5, 28.2, and 57.9%, respectively.
[View Larger Version of this Image (42K GIF file)]
Fig. 3.
Analysis of mitochondrial translation
products in TIG3S fibroblasts from a normal subject (TIG)
and 553 fibroblasts from a patient with mitochondrial disease
(553). Protein synthesis in mitochondria was examined
after [35S]methionine-labeling of mitochondrial
translation products in the presence of emetine (0.2 mg/ml) to protect
translation in cytoplasm. Proteins of the mitochondrial fraction (50 mg/lane) were separated by SDS-polyacrylamide gel electrophoresis.
ND5, COI, ND4, Cytb (cytochrome b) ND2, ND1, COII,
COIII, ATP6, ND6, ND3, ATP8, and ND4L are polypeptides
assigned to mtDNA genes. TIG, TIG3S fibroblasts from a
normal subject; 553, 553 fibroblasts from a patient.
[View Larger Version of this Image (43K GIF file)]
0 HeLa cells were shown to have no mtDNA (8) and to be
resistant to both 6-thioguanine and Oua, HeLa nuclei free from mtDNA
were introduced into 553 fibroblasts simply by fusion of 553 fibroblasts with
0 HeLa cells, and then the cells were
cultivated in selective medium with Oua + HAT (Table I;
cf. "Materials and Methods") to exclude unfused parental
0 HeLa cells and parental fibroblasts by HAT and Oua,
respectively, with selective survival of nuclear hybrids (Table I). The
possibility that the nuclear hybrid clones we isolated were hybrids
between the fibroblasts and contaminating presumptive revertant
0 HeLa cells with recovered HeLa mtDNA was excluded by
the fact that no HeLa mtDNA was present in these nuclear hybrid clones. This was shown by HhaI digestion of the PCR products (data
not shown), a procedure that can distinguish HeLa mtDNA from other human mtDNAs, as described previously (21). Therefore, the nuclear genome of the nuclear hybrids was derived from both parental
0 HeLa cells and fibroblasts, while the mitochondrial
genome was exclusively from fibroblasts. Thus, pure HeLa nuclei were
transferred to 553 fibroblasts from the patient and control TIG3S
fibroblasts, respectively (Table I).
Cell line
Drug
resistancea
Selection
mtDNAb
Chromosome
numberc
Parent cell lines
TIG3S
(normal)
+
46
553
(patient)
+
46
0
HeLa(nuclear donor)
6-Thioguaniner,
Ouar
48
Nuclear hybrid clones
HTIG-1
(
0 HeLa × TIG3S)
Oua + HAT
+
78
HTIG-2 (
0 HeLa × TIG3S)
Oua + HAT
+
77
H553-1 (
0 HeLa × 553)
Oua + HAT
+
80
H553-2 (
0 HeLa × 553)
Oua + HAT
+
72
H553-3 (
0 HeLa × 553)
Oua + HAT
+
75
a
r, -resistant.
b
The absence of HeLa mtDNA in nuclear hybrid clones was
confirmed by HhaI digestion of the polymerase chain reaction
products.
c
Modal chromosome number was obtained from 25 metaphases.
Then we compared the activities of COX, SDH, and oligomycin-sensitive
mitochondrial ATPase, and the profiles of mitochondrial protein
synthesis in nuclear hybrid clones of 0 HeLa cells × 553 fibroblasts and
0 HeLa cells × TIG3S
fibroblasts. Results showed that all the abnormal clinical phenotypes
were reversed in nuclear hybrid clones of
0 HeLa
cells × 553 fibroblasts to levels comparable to those in control
nuclear hybrid clones of
0 HeLa cells × TIG3S
fibroblasts (Figs. 4 and 5). Moreover,
the slight overexpressions of ATP6 and ATP8 polypeptides in 553 fibroblasts (Fig. 3) were no longer observed in their nuclear hybrid
clones (Fig. 5). Since
0 HeLa cells have no mtDNA,
restoration of the normal clinical phenotypes must have been due to
cooperation between the imported HeLa nuclear genome and the
mitochondrial genome of the patient.
Therefore, mtDNA in the patient was not responsible for the pathogenesis, and all the clinical phenotypes related to mitochondrial dysfunction observed in 553 fibroblasts, i.e. reduced COX and SDH activities, enhanced oligomycin-sensitive mitochondrial ATPase activity, reduced translations of COX subunits (COI, COII, and COIII), and enhanced translations of ATP6 and ATP8 subunits, should be due to nuclear recessive mutations.
In this study, we developed a system for delivery of a normal
nuclear genome from 0 HeLa cells to fibroblasts from a
patient to determine whether a mitochondrial or nuclear genome of the
patient was responsible for expression of the clinical phenotypes of
mitochondrial dysfunction. Intercellular transfer of pure nuclei from
0 HeLa cells to 553 fibroblasts from the patient was
achieved by simple fusion of
0 HeLa cells with the 553 fibroblasts followed by HAT and Oua selection. In these nuclear hybrid
clones, the nuclear genome was derived from both parental 553 fibroblasts and
0 HeLa cells, whereas the mitochondrial
genome was exclusively from the 553 fibroblasts. Therefore, if
mitochondrial dysfunction was restored by the introduction of pure
0 HeLa nuclei, it should be attributed to nuclear
recessive mutations, and if not, it should be due to mtDNA mutations or
nuclear dominant mutations.
The 553 fibroblasts showed apparently diverse abnormalities of mitochondrial respiratory function; reduced COX and SDH activities slightly enhanced oligomycin-sensitive mitochondrial ATPase activity, incomplete processing of mitochondrial transcripts, and abnormal mitochondrial translation profiles. Transfer of the pure nuclear genome showed that all the mitochondrial disorders of 553 fibroblasts were reversed completely by the introduction of pure HeLa nuclei, suggesting that the clinical phenotypes were diagnosed to be due to nuclear recessive mutations and not to mtDNA mutations in the 553 fibroblasts. Accordingly, abnormalities of at least two nuclear encoded factors are involved in the expressions of the clinical phenotypes in the patient. One is a factor necessary for the processing of mitochondrial transcripts. Mutation of this factor could be responsible for insufficient processing of mitochondrial transcripts resulting in the productions of abnormal amounts of mitochondrial translation products followed by abnormal COX and ATPase activities. The other is a factor necessary for the expression of SDH activity. Recently, a point mutation in the nuclear encoded flavoprotein subunit gene of SDH, which might be responsible for the pathogenesis, was reported in two SDH-deficient siblings with Leigh's syndrome by Bougeron et al. (22), but it was not present in the patient reported here (data not shown). We are now investigating the regulation of expression of SDH subunits in 553 fibroblasts using Northern blot and Western blot analyses.
In our nuclear genome delivery system, the fate of 0
HeLa cell-derived mitochondria in the nuclear hybrids requires
consideration, since
0 HeLa cells still have these
organelles that have no mtDNA or respiratory enzyme activities (14,
23). We recently showed that when HeLa or fibroblast mitochondria were
introduced into
0 HeLa cells by their fusion with
cytoplasts of HeLa cells or fibroblasts, mtDNA from the imported
mitochondria rapidly spreads to all the
0 HeLa
mitochondria (24). Therefore, in our nuclear hybrids, all
0 HeLa cell-derived mitochondria come to possess mtDNA,
probably by intermitochondrial fusion with the imported mitochondria
containing mtDNA (25).
Previously, similar experiments using somatic hybrids isolated by the
fusion of HeLa cells with fibroblasts from a patient with Leigh's
syndrome were reported by Miranda et al. (26), and their
results showed that COX deficiency observed in fibroblasts from a
patient was due to nuclear DNA-encoded mutations. To eliminate HeLa
cell-derived mtDNA from the hybrids, they used a specific characteristic of HeLa mtDNA, a characteristic which we reported previously, that HeLa mtDNA preferentially segregates from both hybrids
and cybrids during prolonged cultivation (27). However, since they did
not use 0 HeLa cells as nuclear donors, they could not
exclude the possible contribution of a slight amount of remaining HeLa
mtDNA (about 5%) in maintaining COX activity in the hybrids. In fact,
it has been generally thought that a very small amount of wild-type
mtDNA (less than 5%) can compensate for the influence of predominant pathogenic mutant mtDNA with tRNALeu(UUR) 3243 (10, 11),
3271 (12), and tRNALys 8344 (9) in cybrids and help in
maintaining normal translation activity in mitochondria.
Recently, Tiranti et al. (28) reported the nuclear DNA
origin of reduced COX activity in fibroblasts from a patient with Leigh's syndrome based on the absence of cytoplasmic co-transfer of
the patient's mtDNA and COX deficiency to 0 cells. On
the other hand, replacement of mtDNA of fibroblasts from a patient by
mtDNA from normal cells did not restore the reduced COX activity. For
accomplishment of these experimental processes, however, an
SV40-transformed, neomycin-resistant (Neor), and
0 fibroblast line must be isolated from the patient's
fibroblasts by their SV40-Neor transfection to isolate
transformed Neor fibroblast lines followed by long term
ethidium bromide treatment to remove their own mtDNA completely.
Therefore, this mtDNA transfer system requires complicated experimental
processes, and it is possible that SV40-induced transformation and
ethidium bromide treatment of fibroblasts from the patient may have
side effects on some factors necessary to ensure normal COX activity.
Moreover, for isolation of mtDNA-repopulating cells (cybrids),
selective medium without uridine and pyruvate must be used for the
exclusion of parental
0 cells, a medium in which cybrids
with deficient COX activity might also be preferentially eliminated
together with the
0 cells, as reported previously (8,
12, 15), resulting in the possibly misleading conclusion that mtDNA
mutations are not involved in the pathogenesis.
In the present study, HeLa nuclei completely free from mtDNA were
introduced into fibroblasts from a patient simply by fusion of the
fibroblasts with 0 HeLa cells followed by HAT selection
without using any selective pressure upon mitochondrial respiratory
function to remove
0 HeLa cells. Therefore, this nuclear
genome delivery system using
0 HeLa cells can be used to
determine the localizations of pathogenic mutations and the modes of
inheritance of clinical phenotypes, although possibilities on the loss
of a chromosome with dominant allele or increased competition of
dominant allele with recessive alleles (1 versus 4 or
higher) must be considered carefully before conclusion. This can
provide the parents of the patients with valuable information for
genetic counseling on the risk of mitochondrial disorders in their next
babies.
We are grateful to Dr. S. Horai, National Institute of Genetics, Dr. Y.-I. Goto, National Center of Neurology and Psychiatry, and Dr. Y. Kagawa, Jichi Medical School, for providing the probes used in this study.