© 2002 Oxford University Press Human Molecular Genetics, 2002, Vol. 11, No.

4 431–438

Differentiation-specific effects of LHON mutations

introduced into neuronal NT2 cells
Alice Wong, Lucia Cavelier1, Heather E. Collins-Schramm2, Michael F. Seldin2,
Michael McGrogan3, Marja-Liisa Savontaus4 and Gino A. Cortopassi*
Department of Molecular Biosciences, One Shields Avenue, School of Veterinary Medicine, University of California,
Davis, CA 95616, USA, 1Section of Medical Genetics, Department of Genetics and Pathology, Rudbeck Laboratories,
Dag Hammarsjölds väg 20, 751 85, Uppsala, Sweden, 2Rowe Program in Human Genetics, Departments of
Biological Chemistry and Medicine, One Shields Avenue, School of Medicine, University of California,
Davis, CA 95616, USA, 3Layton Biosciences, 709 East Evelyn Avenue, Sunnyvale, CA 94086, USA and
4Departments of Medical Genetics and Biology, University of Turku, FIN-20500 Turku, Finland

Received October 18, 2001; Revised and Accepted December 5, 2001

Inheritance of one of three primary mutations at positions 11778, 3460 or 14484 of the mitochondrial genome
in subunits of Complex I causes Leber’s Hereditary Optic Neuropathy (LHON), a specific degeneration of the
optic nerve, resulting in bilateral blindness. It has been unclear why inheritance of a systemic mitochondrial
mutation would result in a specific neurodegeneration. To address the neuron-specific degenerative phenotype of
the LHON genotype, we have created cybrids using a neuronal precursor cell line, Ntera 2/D1 (NT2), containing
mitochondria from patient lymphoblasts bearing the most common LHON mutation (11778) and the most
severe LHON mutation (3460). The undifferentiated LHON-NT2 mutant cells were not significantly different from
the parental cell control in terms of mtDNA/nDNA ratio, mitochondrial membrane potential, reactive oxygen
species (ROS) production or the ability to reduce Alamar Blue. Differentiation of NT2s resulted in a neuronal
morphology and neuron-specific pattern of gene expression, and a 3-fold reduction in mtDNA/nDNA ratio in
both mutant and control cells; however, the differentiation protocol yielded significantly less LHON cells than
controls, by 30%, indicating either a decreased proliferative potential or increased cell death of the LHON-NT2
cells. Differentiation of the cells to the neuronal form also resulted in significant increases in ROS production
in the LHON-NT2 neurons versus controls, which is abolished by rotenone, a specific inhibitor of Complex I.
We infer that the LHON genotype requires a differentiated neuronal environment in order to induce increased
mitochondrial ROS, which may be the cause of the reduced NT2 yield; and suggest that the LHON degenerative
phenotype may be the result of an increase in mitochondrial superoxide which is caused by the LHON mutations,
possibly mediated through neuron-specific alterations in Complex I structure.

INTRODUCTION Thus, the defect in respiration observed in whole mitochondria

Leber’s Hereditary Optic Neuropathy (LHON) is caused by
inheritance of mutations at positions 11778, 3460 or 14484 of
the mitochondrial genome, each of which are components of
Complex I (1–4). The clinical phenotype of LHON is the
degeneration of retinal ganglion cells (RGCs), and a progressive
degeneration of the optic nerve. One puzzling feature of LHON is
the optic nerve specificity of the disease. Although the LHON
mutations are usually homoplasmic (i.e. occur systemically),
in most patients the only symptom is vision loss, the result of
degeneration of the optic nerve. Studies in cybrids have
demonstrated that the LHON mutations cause a decrease in
molecular oxygen consumption in whole mitochondria
supplied with Complex I substrates, but not Complex II
substrates (5,6). However, in studies of Complex I activity in
submitochondrial particles, estimates of the severity of an
enzymatic defect in Complex I vary considerably (7–10).
does not appear completely attributable to a simple enzymatic
defect of Complex I activity, and other hypotheses for LHON
pathophysiology have been suggested, including an apoptotic
hypothesis (11), and a reactive oxygen species (ROS) hypothesis,
which are non-exclusive of each other (8,10,12).
One possible explanation for the optic neuron-specific
phenotype of the LHON mitochondrial genotype is that the
LHON mutations only express their pathogenicity in the
context of a neuronal cell, perhaps because of differences in
mitochondrial structure or function in neuronal cells. In order
to address this possibility, we fused LHON mitochondria with
the neural Ntera-2/D1 (NT2) human teratocarincoma cell line,
which had been depleted of their mitochondria. The NT2 cell
line is perhaps the most ‘physiological’ human neuronal cell
line available. Precursor cells undergo differentiation into
neuronal cells following retinoic acid (RA) treatment. Upon
differentiation, the cells assume a neuronal morphology, have

*To whom correspondence should be addressed. Tel: +1 530 754 9665; Fax: +1 530 754 9342; Email: gacortopassi@ucdavis.edu
432 Human Molecular Genetics, 2002, Vol. 11, No. 4

Figure 1. (A) A schematic representation of the method used to generate LHON-NT2 cybrids with the 11778 or 3460 mutation. (B) PCR product analysis of the
mtDNA genome in LHON-NT2 cell lines bearing the 11778 mutation (lane 1). PCR products and restriction enzyme digests of the products were separated on a
2% agarose gel by electrophoresis. SfaNI only cleaves the wild-type, i.e. NT2 mtDNA, whereas the G→A mutation that creates the LHON 11778 mutation
abolishes the recognition site of the enzyme. MaeIII, which cleaves the newly established site created by the 11778 mutation, was used to confirm the presence of
the 11778 mutation in the two clones, 11778-1 and 11778-2. The mitochondrial donor cell line, 910555 was included as a control. To detect the 3460 mutation,
Hin1I was used to cleave the NT2 DNA, but does not cleave the 3460-1 DNA due to the 3460 G→A mutation (data not shown). L, 123 bp ladder; lane 1, undigested
PCR product; lanes 2–5, SfaNI; lanes 6–9, MaeIII digests of NT2, 11778-1, 11778-2 and L-910555 PCR products, respectively. (C) Microsatellite image of one
allelic marker, DXS6807, of the 70 used. The x-axis represents size in base pairs (bp) and y-axis represents fluorescence intensity.

a neuronal pattern of gene expression, become responsive to
neuron-specific stimuli and are electrophysiologically active
(13–15). We have fused mitochondria from patient lymphoblasts
bearing the most common LHON 11778 or 3460 mutations
with the NT2 cell line, to determine if the introduction of
LHON mutations into a neuronal environment causes cell-type
specific effects or differentiation-specific effects.
RESULTS
Construction of LHON-NT2 cybrids
LHON-NT2 cybrids were created as described in the schema
(Fig. 1A) and Materials and Methods. DNA from all clones
was subjected to PCR of the relevant mtDNA region and
restriction fragment length polymorphism (RFLP) analysis
was performed (Fig. 1B). Restriction enzymes were used to
cleave the 499 bp PCR fragment (Fig. 1B, lane 1). The patterns
of the fragments confirm the presence of the LHON mutations
in each respective cell line, thus these LHON-NT2 cells
contain pure (homoplasmic) LHON mtDNA.
Verification of NT2 nuclear background in LHON-NT2
cybrids
The LHON-NT2 cells which contain homoplasmic LHON
mtDNA also contain an NT2 nuclear genotype, as determined
by several criteria. The LHON-NT2 clones were selected on the
basis of geneticin (G418) resistance, which is conferred by the
NT2 nucleus that was transfected with a vector encoding G418
resistance (M.McGrogan, unpublished data). Also, under light
microscopy the 11778-1, 11778-2 and 3460-1 cell lines had an
indistinguishable morphology from the NT2 control cells, i.e.
cuboidal and attached (Fig. 2, left), which was different from the
lymphoblastoid and unattached morphology of the LHON
lymphoblast donor cells (not shown). To further confirm the
nuclear genotype of the LHON-NT2 cybrids, 70 microsatellite
markers representing chromosomes 1–20, 22 and X of the

Figure 2. Differentiation of NT2 and LHON-NT2 cells into neurons. Cells were
treated with 10 µM retinoic acid for 5 weeks and plated with mitotic inhibitors
for 1 week. Neurons were then harvested and plated on rat astrocytes. Nomarski
images of undifferentiated (left) and differentiated cells 2 weeks after completion of
RA treatment (right). 20× magnification; scale bar, 20 µm.
Marshfield screening set 8A (http://research.marshfieldclinic.org/
genetics/) were amplified by PCR from total DNA. Of the
70 markers, 65 and 59 were variable (i.e. informative) between
NT2 and L-910555 DNA (lymphoblasts with 11778 mutation)
and NT2 and L-910615 DNA (lymphoblasts with 3460 mutation),
respectively. An example of one marker, DXS6807, is shown
in Figure 1C. In any particular run, any alleles of the 70 that did
not amplify were excluded from the analysis. Cell lines 11778-1,
11778-2 and 3460-1 shared 65, 63 and 59 informative micros-
atellite alleles with the NT2 parent, respectively, and none with
the L-910555 or L-910615 parent. Since the amplified nuclear
microsatellite alleles were 100% identical to that of the
parental NT2 DNA, the morphology of the undifferentiated
cells is consistent with an NT2 nucleus, and the cells are
geneticin resistant, there is no support for nuclear contamination
from the LHON lymphoblast nuclei in the LHON-NT2 cybrids.
Figure 3. Comparison of mitochondrial DNA copies/nuclear DNA copies in NT2, LHON-NT2 and parental cell lines. Quantitative PCR was used to examine the
ratio of mitochondrial to nuclear genes in each sample. This was done by using pre-optimized standard curves to calculate the number of copies of mitochondrial
(in 0.1 ng total DNA) and nuclear (in 10 ng total DNA) genes per sample based on SYBR-green fluorescence. The cystic fibrosis and ND5 genes were amplified
to determine the ratio of nuclear and mitochondrial copies in the sample, respectively. The means of three experiments are shown, error bars represent 2 SEM.
Human Molecular Genetics, 2002, Vol. 11, No. 4 433
Differentiation of NT2 neural cell precursors to neuronal
form induces a 3-fold decrease in mtDNA/nDNA ratio
In order to determine whether the LHON-NT2 cells had a
normal mtDNA/nDNA ratio, quantitative PCR analysis was
performed. Differentiation of mutant and control precursor
cells to neuronal form caused an ∼3-fold reduction in the
mtDNA/nDNA ratio (Fig. 3). However, there was no significant
difference in the mtDNA/nDNA ratio between mutants and
controls in either the undifferentiated or differentiated form.
This is the first report to our knowledge demonstrating that
differentiation of a neuronal precursor cell reduces the
mtDNA/nDNA ratio; a differentiation-dependent decrease in
the number of mitochondria could be a contributor to the
general observation that mitochondrial defects preferentially
affect the neuronal cell type.
Differentiation of LHON and parental lines induces
neuronal genes
Differentiation of the NT2-LHON neuronal precursors into mature
neurons was carried out (16); differentiated LHON-NT2 cells had
a distinctly neuronal morphology that was indistinguishable
from the differentiated parental NT2 cells (Fig. 2). Although
we did not observe differences in neuronal morphology
between differentiated LHON-NT2 and NT2 cells, we also
assayed the expression of several genes in the undifferentiated
and differentiated NT2 cells, to address whether the presence
of the LHON mutations affected the qualitative gene expression
of these markers of differentiation. Glia, undifferentiated
neuronal cells, and differentiated neurons have a pattern of
gene expression which is cell-specific (14–17). We observed a
completely normal pattern of gene expression in the mutants
versus controls (Fig. 4). For example, both undifferentiated
and differentiated cells express vimentin, but do not express
glial fibrillary acidic protein (GFAP), a marker of glial lineage.
Also, differentiated neurons produced tau mRNA, which is
not found in undifferentiated cells (17). In addition, both
undifferentiated and differentiated cells express the intermediate
neurofilament NF-M, the levels of which increase over time
with RA treatment (17). Thus, the occurrence of the 11778 and
434 Human Molecular Genetics, 2002, Vol. 11, No. 4
Figure 4. Expression of tau, NFM, vimentin and GFAP in undifferentiated and differentiated control and LHON-NT2 cells bearing the 11778 mutation. RT–PCR
was performed to analyze the RNA expression of tau, neurofilament medium chain (NFM), vimentin and GFAP in undifferentiated and differentiated cells. L, ladder;
UC, undifferentiated control; UM, undifferentiated mutant; DC, differentiated control; DM, differentiated mutant. Although multiple cell-type dependent or differentiation
dependent differences are described in the text, there is no observable effect of the LHON mutation on these proteins. Similar results were seen with the LHON-NT2n
cell line bearing the 3460 mutation (data not shown).
Figure 5. Measurement of mitochondrial membrane potential and reduction by
Alamar blue. Undifferentiated cells were incubated with 50 nM TMRM and
mitochondrial membrane potential was measured using FACS analysis (A).
DNP (1 µM) was added as a control and a decrease in membrane potential is
shown for 34600-1. Alamar blue reduction is shown after 6 h incubation for
neurons and undifferentiated cells (B).
3460 mutations (not shown) exerted no observable effect on
the qualitative pattern of gene expression relative to controls.
No difference in mitochondrial membrane potential nor
reduction of Alamar blue in mutants relative to controls
Mitochondrial membrane potential was measured in undiffer-
entiated cells. No significant difference between LHON-NT2
cells and controls was observed (Fig. 5A). As a control,
dinitrophenol (DNP) was added to depolarize the membrane
potential, and a decrease in log fluorescence was observed for
all cells treated with DNP, an example is shown in Figure 5A.
Alamar blue is a dye whose reduction is correlated with
mitochondrial electron transport chain activity. No significant
differences in Alamar blue reduction were observed between
LHON-NT2 and control cells in the undifferentiated or
differentiated state (Fig. 5B). Thus, we did not observe any
significant effects of the LHON mutations on membrane
potential or the ability to reduce Alamar Blue.
Decreased yield of LHON-NT2 cells during differentiation
Although LHON-NT2 cells differentiate into neurons, we
observed a subtle difference in yield of total cells (i.e. differen-
tiated + undifferentiated) at 4 weeks from the start of
differentiation. Figure 6 shows the total number of cells after
differentiation in RA for 4 weeks. 2.5 × 107 cells were started
for each cell line and, after 4 weeks, the cell population for NT2
cells increased 50-fold, while the LHON-NT2 cell population
increased only 36-fold. Note that this yield includes both the
differentiated overlying neurons (NT2 neurons) and the under-
lying supportive cells of glial lineage. This difference in cell
number could be explained either by a decreased proliferation
of the LHON-NT2 cells during differentiation or an increased
cell death, or both.
Increased ROS production in LHON-NT2 neurons relative
to parental controls
One possible pathophysiological mechanism in mitochondrial
disease (18,19) and specifically in LHON (8,10,12), is
increased production of ROS in mutant cells. Superoxide and
peroxides are two examples of ROS produced by mitochondria
which can be assayed by the mitochondria-specific dye
dihydroethidium (DHE) and the cell-general dye dichloro-
fluorescein diacetate (DCFDA) assays (20–24). No differences
in ROS production were observed between undifferentiated
NT2 and LHON-NT2 cell lines (Fig. 7), nor between osteo-
sarcoma cell lines with and without these LHON mutations
(data not shown). In contrast, after differentiation to the
neuronal form, the LHON-NT2n cells produced about 2.5-fold
more cellular hydroperoxide, as determined by the DCFDA
assay, (P < 0.005, Fig. 7A). To identify the origin of increased
ROS in LHON neurons, we used an assay specific for mito-
chondrial superoxide, the DHE assay (22,23,25). An increased
production of mitochondrial superoxide was observed in
differentiated LHON neurons relative to controls, and neurons
 Human Molecular Genetics, 2002, Vol. 11, No. 4 435

Figure 6. Total number of cells after retinoic acid treatment. 2.5 × 107 cells
were used for differentiation. After 4 weeks of RA treatment, the total number
of cells was determined by the trypan blue exclusion assay. Error bars represent
2 SEM, *P < 0.05.

bearing the more severe 3460 mutation produced more super-

oxide (Fig. 7B). We investigated the specificity of the assay by
short treatments of the mutant and control neurons with
specific mitochondrial inhibitors, antimycin A and rotenone
(Fig. 7C). As expected, antimycin A, a specific inhibitor of
Complex III, increased DHE fluorescence in all three groups of
cells, demonstrating specificity of the assay for mitochondrial
superoxide. In contrast, a short treatment of the cells with the
Complex I-specific inhibitor rotenone, specifically inhibited
the excess superoxide produced in the mutant neurons, but
only to a small and not statistically significant extent in the
controls. The simplest interpretation of these data is that the
11778 and 3460 mutations specifically cause an increase in
mitochondrial superoxide in the differentiated neuronal
environment which is related to the severity of the disease, and
that the mutation-specific increase in superoxide is the direct
result of electron flow through Complex I.

DISCUSSION
In contrast to the pleiotropic phenotypes observed in other
mitochondrial diseases, in patients affected with LHON the
systemic inheritance of mtDNA mutation causes a strikingly
specific and singular phenotype, i.e. the degeneration of the
RGCs and optic nerve. The reason(s) for this specificity of
neurodegeneration have not been clear, and we have addressed
this problem by introducing LHON mutations into neural cells.
It was demonstrated that the 11778 and 3460 mutations from
lymphoblastoid mitochondria were introduced into NT2
cybrids by PCR and restriction enzyme analysis (Fig. 1). There
was no evidence of lymphoblastoid nuclear contamination of the
cybrids, by selection with geneticin, by cellular morphology, and
by microsatellite analysis.
LHON-NT2 cells differentiated into neurons, as demonstrated
by a neuronal morphology and a neuron-specific pattern of gene
expression (Figs 2 and 4). The differentiation process itself
caused a 3-fold reduction in mtDNA/nDNA ratio in both mutant
and control NT2 neurons (Fig. 3). This is to our knowledge the
first demonstration that neurons have a significantly lower
mtDNA/nDNA ratio than lymphoblasts, NT2 precursors and
other reference cells such as fibroblasts that have about
1000 mtDNAs/nDNA or more. The decreased mitochondrial
Figure 7. Measurement of ROS in differentiated NT2 neurons. Neurons and
undifferentiated cells were incubated with 2 µM DCFDA (A) or 2 µM DHE (B),
and fluorescent measurements were taken at 2 h. (C) DHE data were recorded
for neurons in PBS (clear bars), and in the presence of two mitochondrial
inhibitors, 10 µM rotenone (dark bars) or 10 µM antimycin A (gray bars). Data
represent the mean of three experiments performed in triplicate. Error bars
represent 2 SEM, *P < 0.05, **P < 0.005.
content of differentiated neurons could underlie the increased
sensitivity of neurons to mitochondrial defects.
We observed a subtle defect in cellular yield during the differen-
tiation of LHON-NT2 neurons versus controls, i.e. 9 × 108 cells
were produced from the LHON-NT2s, versus 1.3 × 109 in the
parental NT2s (Fig. 6). This defect could either be the result of
decreased proliferative capacity or increased cell death during
the differentiation process of the mutant cells, and we are
436 Human Molecular Genetics, 2002, Vol. 11, No. 4
currently investigating these possibilities. The analysis is
complicated by the fact that the differentiation takes 4 weeks
and is a mixture of cells destined to be neurons and glia.
Measurements of mitochondrial function could address the
differences in cellular yield that were observed during differ-
entiation. There was no strong support for a mitochondrial
bioenergetic defect conferred by LHON mutations in NT2 or
NT2 neurons, by the Alamar blue and TMRM assays (Fig. 5).
It has previously been suggested that perhaps in optic neurons
Complex I activity is limiting to the respiratory chain, and thus
a small defect in the enzymatic activity of Complex I limits
bioenergetic function, resulting in a specific bioenergetic
problem in these cells (9). However, at the level of sensitivity
of these assays, we do not observe a difference in bioenergetic
function.
An alternative to the bioenergetic hypothesis is that the
LHON mitochondria produce more ROS (26,27), and either
the optic neurons are more sensitive to ROS, or that the LHON
mitochondria produce more ROS specifically in the neuronal
environment (this work). We only observed increases in
mutants versus control cells in the differentiated neuronal state,
and not in the undfferentiated NT2 state, nor in comparison of
mutant and control osteosarcoma cybrids. We have observed
an increase in ROS measured with DCFDA, which is a general
assay of cellular peroxides, and also with DHE, which is a
more specific assay of mitochondrial superoxide (Fig. 7). We
observed that on average differentiated NT2 neuron controls
produced less peroxide and superoxide than the undifferentiated
NT2 neurons, but that the mutant differentiated neurons
produced more peroxide and mitochondrial superoxide than
the parental differentiated line. The fact that the LHON-NT2
neurons produced more superoxide than the undifferentiated
controls supports the notion that the LHON mutations are
causing a differentiation-dependent pathophysiological
change, rather than, the simple inhibition of differentiation,
and this is also consistent with the identical neuronal
morphology of LHON-NT2s (Fig. 2), and identical pattern of
neuron-specific gene expression between mutants and controls
observed in Figure 4.
The observation that antimycin A produced a large increase
in superoxide in the neurons supports the specificity of the
DHE assay, as antimycin A is known to increase mitochondrial
superoxide production. The inhibitability of ROS production
specifically by the Complex I inhibitor rotenone suggests that
the excess mitochondrial superoxide observed in mutants is
derived from Complex I. The simplest inference from these
data is that LHON mitochondria cause increased ROS
production through alterations of the function of Complex I,
specifically in a differentiated neuronal environment.
The observation of increased mitochondrial ROS in differentiated
LHON cybrids suggests two pathophysiological mechanistic
possibilities, but does not prove either one. The increased ROS
are presumably the result of the LHON mutations, and may
directly be the cause of increased cell death in the LHON
disease, or not. Conversely, the increased ROS may be the
consequence of some other process which is downstream of
the LHON mutations. The yield defect in differentiating
LHON-NT2 cells may allow the dissection of whether this is a
proliferative versus cell death problem, and may provide an
experimental means by which to test whether ROS are on the
pathway to the cell yield defect, for example with antioxidants.
To date, we have not observed significant differences in
LHON-NT2 versus control neuronal death provoked by bolus
addition of exogenous hydrogen peroxide, indicating that there
is not direct additivity of the mitochondrial superoxide we observe
with external H2O2 to produce neuronal death. This could either be
explained by frank ROS toxicity not being the proximal cause of
neuronal death in LHON, or by the non-physiological way
(i.e. bolus addition of exogenous hydrogen peroxide) in which
these neurons were exposed to ROS, and we are exploring
other schemes for exposure to oxidative stress.
The reason for the neuron-specific increase in ROS is not
known at this point, but two possibilities include: (i) that there
is a neuron-specific alteration of Complex I structure that leads
to increased ROS in the context of LHON mutations or (ii) that
the LHON mutations lead to other pathophysiological changes,
which then cause increased mitochondrial ROS. The first
hypothesis is more direct, and is thus more simply tested.
These results, to our knowledge, provide the first experimental
support to explain the cell specificity of the LHON degenerative
phenotype, and also the first experimental support for the
‘oxidative stress hypothesis for LHON’, and could be relevant
to LHON therapy, for example, with antioxidants. Another
appealing feature of an oxidative stress hypothesis for LHON is
that it has the potential to explain the 3-fold excess of LHON in
males versus females if, for example, there is increased
production of mitochondrial ROS in males versus female
humans, as has been observed by others in laboratory animals
(J.Vina, personal communication).
MATERIALS AND METHODS
Cell culture
NT2 cells transfected with the geneticin (G418) resistance gene
were maintained in DMEM supplemented with 10% FBS, 50 µg/ml
uridine, 1 µM sodium pyruvate, penicillin and streptomycin,
and 0.2 mg/ml G418. For differentiation, 2.5 × 106 cells were
seeded in 75 cm2 flasks and treated with 10 µM RA in freshly
prepared media twice a week for 5 weeks. The cells were then
replated at 50 × 106 cells/T75 cm2 flasks, and following 1 day
in culture without RA, the media was saved and new media
supplemented with mitotic inhibitors (1 µM cytosine arabino-
side and 10 µM fluorodeoxyuridine) was given for 1 week.
Neurons were harvested by gentle trypsinization and plated
onto poly-D-lysine-Matrigel or -astrocyte coated plates or
coverslips, and given media consisting of 50% NT2 and 50%
media saved from the 1 day in culture without RA. Differentiated
neuronal cell lines are indicated with a ‘n’ after the name, i.e. NT2n.
Astrocyte preparation
Rat astrocytes were generously provided by Dr Patty Wong-Yim
(University of California Davis), and were isolated according
to the procedures described by Goslin and Banker (28). Frozen
astrocytes were thawed and plated at a density of 12 000 cells/cm2
in MEM supplemented with 10% horse serum and 20% glucose.
Astrocytes reached confluency within 10–14 days of initial plating.
Production of LHON-NT2 cybrids
A modified version of cybrid production described by King
and Attardi (29) and Trounce and Wallace (30) was used to

produce the LHON-NT2 cell lines (Fig. 1). For enucleation of
donor cells, 107 lymphoblasts were resuspended in 3 ml 12.5%
Ficoll containing 10 µg/ml cytochalasin B, and this was
layered onto a discontinuous Ficoll gradient layer containing
10 µg/ml cytochalasin B. The gradient was centrifuged at
28 000 r.p.m. for 1 h at 37°C in a SW41 swinging bucket ultra-
centrifuge (Beckman Institute, Inc.), after which, the cytoplasts
were removed and washed. The recipient cells, NT2, were
treated 3 days prior to fusion with 1 µg/ml rhodamine-6-G
(R6G) and harvested for fusion. The recipient and donor cells
were mixed together and resuspended in 40–50% PEG for
1 min. Media was added and cells were plated overnight, and
the next day, diluted into 100 mm plates. Colonies were picked
3–4 weeks later and screened for the presence of the 11778 or
3460 mutation by PCR. Approximately 10–20 colonies were
screened for each fusion, two were homozygous for the 11778
mutation, 11778-1 and 11778-2, and one was homozygous for
the 3460 mutation, 3460-1.
LHON mutation detection
For mutation detection, the following primers were used to
amplify the region containing the 11778 mutation: forward
5 ′-CCCATCGCTGGGTCAATAGT-3′ and reverse 5′-ATTT-
GATCAGGAGAACGTGG-3′. PCR conditions were as follows:
30 s denaturation at 95°C, 30 s annealing at 58°C, and 1 min
NSextension at 72°C. The G→A mutation can be screened via
two restriction enzyme digestions, SfaNI cleaves wild-type but
not mutant (1), and MaeIII makes an additional cut in the
mutant (31). The 3460 amplification used the following primers
and PCR conditions: forward 5′-GCAGAGCCCGGTAATCG-
CATA-3′ and reverse 5′-AAGGTCGGGGCGGTGATGTAG-
3′, same PCR conditions as above, except annealing was per-
formed at 55°C. Hin1I cleaves the wild-type (32).
Microsatellite analysis
All samples were tested with 70 markers from the Marshfield
screening set 8A (http://research.marshfieldclinic.org/genetics/).
PCR consisted of 0.5 µl PCR buffer, 0.7 µl 2.5 mM dNTP mix
(Pharmacia and Upjohn), 0.05 µl cDNA polymerase (Clonetech
Advantage), 0.1 µl of 10 µM fluorescently labelled primer mix,
2.65 µl double distilled H20, and 1 µl DNA at ∼0.5 ng/µl. PCR
conditions were as described (http://research.marshfieldclinic.org/
genetics/). Approximately 0.5 µl PCR product (amount varied
depending on marker optimization) was loaded onto a 6%
denaturing polyacrylamide gel with Genescan-350 Tamra
standard (Perkin-Elmer) and run at 25 W on an ABI 373 DNA
sequencer. Gels were then analyzed with Genescan software
and imported into Genotyper software (Perkin-Elmer) for
electrophoretogram analysis. Genotyper’s automated analysis
was used to initially genotype samples, and all genotypes were
then verified by inspection.
Microscopy
Undifferentiated cells were grown on uncoated glass coverslips
and differentiated cells were plated onto glass coverslips
coated with astrocytes. The cells were imaged using an
Olympus AX70 epifluorescent Provis microscope equipped
with a SONY 3CCD video camera. Magnification and scale
bars are indicated in the figure legends.
Human Molecular Genetics, 2002, Vol. 11, No. 4 437
Mitochondrial and nuclear DNA quantitation
Quantitation of mtDNA and nDNA from control and mutant
NT2 cells and neurons was performed using the LightCycler
instrument (Roche). Cystic fibrosis primers were used to
amplify a 460 bp single copy nuclear gene using the following
primers: forward 5′-AGCAGAGTACCTGAAACAGGAA-3′
and reverse 5′-AGCTTACCCATAGAGGAAACATAA-3′.
Primers to the ND5 gene (13175–13501) were used to amplify
the mitochondrial genome (forward 5′-AGGCGCTATCAC-
CACTCTGTTCG-3′ and reverse 5′-AACCTGTGAGGAAA-
GGTATTCCTG-3′). PCR was performed under the following
conditions, after an initial 30 s denaturation at 95°C, 40 cycles
of the following was used: 0 s for 95°C, 5 s for 60°C and 13 s
for 72°C. Ten nanograms and 0.1 ng of DNA was added to
amplify nDNA and mtDNA genes, respectively. To establish a
standard curve for both nuclear and mitochondrial DNA,
K562 DNA (Life Technologies) was used at optimal concentrations
(0–100 ng for nuclear gene and 0–1 ng for mitochondrial
gene). A standard curve was performed during every PCR run.
Quantitation was determined by the cycle number at which
fluorescent values reached threshold. One nanogram of DNA was
assumed to have 150 nuclear copies and 300 000 mitochondrial
copies.
RT–PCR
Total RNA was prepared from cells using the SV total RNA
isolation system (Promega). To obtain cDNA, reverse transcriptase
reactions were performed using 1 µg of RNA. The following
primers were used to amplify tau, neurofilament medium chain
(NFM), vimentin and GFAP: tau, forward 5′-ACAAGCTGA-
CCTTCCGCGAG-3′ and reverse 5′-ACAAACCCTGCTTG-
GCCAGG-3′; medium neurofilament (NFM), forward 5′-TCT-
GTAACCGTCACTCAAAAGG-3′ and reverse 5′-CCGTTC-
TGTTTTGAAGCTGCC-3′; vimentin, forward 5′-GTCAGC-
AATATGAAAGTGTGGC-3′ and reverse 5′-GGTAGTTAG-
CAGCTTCAACGG-3′; and GFAP, forward 5′- GAGGGAC-
AATCTGGCACAGG-3′ and reverse 5′-CAGCTGCTCCTG-
GAGTTCCC-3′. PCR conditions were as follows, denaturation at
95C for 30 s, annealing at 54°C (NFM, vimentin and GFAP) or
at 58°C (tau) for 30 s, and extension at 72°C for 60 s.
Alamar blue assay
Alamar Blue reduction was measured in undifferentiated and
differentiated cells following the manufacturer’s protocol
(AccuMed International Companies, Westlake, OH). Briefly,
10 000 cells/well were plated in 96-well plates in 200 µl
volume and 20 µl of the Alamar blue reagent was added to
each well. Six hours later, fluorescence measurements were
taken at 560 nm excitation and 590 nm emission.
Measurement of mitochondrial membrane permeability
Mitochondrial membrane permeability was measured in digitonin
permeabilized cells using the assay described by Antonická
et al. (32). Briefly, cells were harvested and washed three
times in PBS, resuspended in KCl medium (80 mM KCl,
10mM Tris–HCl, 3 mM MgCl2, 1 mM EDTA, 5 mM KH2PO4
pH 7.4, 10 mM succinate and 1 µM rotenone, and incubated
with digitonin (0.2 mg/ml) for 5 min on ice. The cells were
washed once in KCl medium and incubated with 50 nM
438 Human Molecular Genetics, 2002, Vol. 11, No. 4
TMRM for 15 min at room temperature, after which, the cells
were washed once and resuspended in KCl medium. To
depolarize the membrane, 1 µM DNP was added to cells before
analysis. Analysis was performed on the FACSort flow cytometer
(Becton Dickinson, San Jose, CA), with a 488 nm argon laser,
and data was acquired.
ROS measurements
ROS measurements were performed following the procedure of
Degli Esposti and McLennan (24). Briefly, cells were harvested
and resuspended in PBS (0.5 × 106 cells/ml) supplemented with
20 mM glucose and 2 µM DCFDA or DHE, and incubated at
37°C. Cells treated with mitochondrial inhibitors received 10 µM
antimycin A or 10 µM rotenone. Fluorescence readings were
taken at 2 h (485–530 nm for DCFDA and 530–620 nm for
DHE). A standard curve of dichlorofluorescein was used for
quantification purposes.
ACKNOWLEDGEMENTS
We thank Drs Rebecca Hartley and Lupe Gonzalez for helpful
discussions. This work was supported by USPHS grants
EY12245,AG11967, AG16719, and core support by P30ES05707
to G.A.C.
REFERENCES
1. Wallace,D.C., Singh,G., Lott,M.T., Hodge,J.A., Schurr,T.G., Lezza,A.M.,
Elsas,L.J.d. and Nikoskelainen,E.K. (1988) Mitochondrial DNA mutation
associated with Leber’s hereditary optic neuropathy. Science, 242, 1427–1430.
2. Huoponen,K., Vilkki,J., Aula,P., Nikoskelainen,E.K. and Savontaus,M.L.
(1991) A new mtDNA mutation associated with Leber hereditary optic
neuroretinopathy. Am. J. Hum. Genet., 48, 1147–1153.
3. Howell,N., Bindoff,L.A., McCullough,D.A., Kubacka,I., Poulton,J.,
Mackey,D., Taylor,L. and Turnbull,D.M. (1991) Leber hereditary optic
neuropathy: identification of the same mitochondrial ND1 mutation in six
pedigrees. Am. J. Hum. Genet., 49, 939–950.
4. Johns,D.R., Neufeld,M.J. and Park,R.D. (1992) An ND-6 mitochondrial
DNA mutation associated with Leber hereditary optic neuropathy.
Biochem. Biophys. Res. Commun., 187, 1551–1557.
5. Hofhaus,G. and Attardi,G. (1995) Efficient selection and characterization
of mutants of a human cell line which are defective in mitochondrial
DNA-encoded subunits of respiratory NADH dehydrogenase.
Mol. Cell. Biol., 15, 964–974.
6. Hofhaus,G., Johns,D.R., Hurko,O., Attardi,G. and Chomyn,A. (1996)
Respiration and growth defects in transmitochondrial cell lines carrying
the 11778 mutation associated with Leber’s hereditary optic neuropathy.
J. Biol. Chem., 271, 13155–13161.
7. Hofhaus,G. and Attardi,G. (1993) Lack of assembly of mitochondrial
DNA-encoded subunits of respiratory NADH dehydrogenase and loss of
enzyme activity in a human cell mutant lacking the mitochondrial ND4
gene product [published erratum appears in EMBO J. (1994) 13, 5794].
EMBO J., 12, 3043–3048.
8. Carelli,V., Ghelli,A., Ratta,M., Bacchilega,E., Sangiorgi,S., Mancini,R.,
Leuzzi,V., Cortelli,P., Montagna,P., Lugaresi,E. et al. (1997) Leber’s hereditary
optic neuropathy: biochemical effect of 11778/ND4 and 3460/ND1 mutations
and correlation with the mitochondrial genotype. Neurology, 48, 1623–1632.
9. Majander,A., Huoponen,K., Savontaus,M.L., Nikoskelainen,E. and
Wikström,M. (1991) Electron transfer properties of NADH:ubiquinone
reductase in the ND1/3460 and the ND4/11778 mutations of the Leber
hereditary optic neuroretinopathy (LHON). FEBS Lett., 292, 289–292.
10. Brown,M.D., Trounce,I.A., Jun,A.S., Allen,J.C. and Wallace,D.C. (2000)
Functional analysis of lymphoblast and cybrid mitochondria containing
the 3460, 11778, or 14484 Leber’s hereditary optic neuropathy mitochondrial
DNA mutation. J. Biol. Chem., 275, 39831–39836.
11. Howell,N. (1998) Leber hereditary optic neuropathy: respiratory chain
dysfunction and degeneration of the optic nerve. Vision Res., 38, 1495–1504.
12. Degli Esposti,M., Carelli,V., Ghelli,A., Ratta,M., Crimi,M., Sangiorgi,S.,
Montagna,P., Lenaz,G., Lugaresi,E. and Cortelli,P. (1994) Functional
alterations of the mitochondrially encoded ND4 subunit associated with
Leber’s hereditary optic neuropathy. FEBS Lett., 352, 375–379.
13. Hartley,R.S., Margulis,M., Fishman,P.S., Lee,V.M. and Tang,C.M.
(1999) Functional synapses are formed between human NTera2 (NT2N,
hNT) neurons grown on astrocytes. J. Comp. Neurol., 407, 1–10.
14. Lee,V.M. and Andrews,P.W. (1986) Differentiation of NTERA-2 clonal
human embryonal carcinoma cells into neurons involves the induction of
all three neurofilament proteins. J. Neurosci., 6, 514–521.
15. Andrews,P.W. (1984) Retinoic acid induces neuronal differentiation of a cloned
human embryonal carcinoma cell line in vitro. Dev. Biol., 103, 285–293.
16. Pleasure,S.J., Page,C. and Lee,V.M. (1992) Pure, postmitotic, polarized human
neurons derived from NTera 2 cells provide a system for expressing exogenous
proteins in terminally differentiated neurons. J. Neurosci., 12, 1802–1815.
17. Pleasure,S.J. and Lee,V.M. (1993) NTera 2 cells: a human cell line which
displays characteristics expected of a human committed neuronal progenitor
cell. J. Neurosci. Res., 35, 585–602.
18. Rustin,P., von Kleist-Retzow,J.C., Vajo,Z., Rotig,A. and Munnich,A. (2000)
For debate: defective mitochondria, free radicals, cell death, aging-reality or
myth-ochondria? Mech. Ageing Dev., 114, 201–206.
19. Wong,A. and Cortopassi,G. (1997) mtDNA mutations confer cellular
sensitivity to oxidant stress that is partially rescued by calcium depletion
and cyclosporin A. Biochem. Biophys. Res. Commun., 239, 139–145.
20. LeBel,C.P., Ischiropoulos,H. and Bondy,S.C. (1992) Evaluation of the
probe 2′,7′-dichlorofluorescin as an indicator of reactive oxygen species
formation and oxidative stress. Chem. Res. Toxicol., 5, 227–231.
21. Rothe,G., Emmendörffer,A., Oser,A., Roesler,J. and Valet,G. (1991) Flow
cytometric measurement of the respiratory burst activity of phagocytes using
dihydrorhodamine 123. J. Immunol. Methods, 138, 133–135.
22. Carter,W.O., Narayanan,P.K. and Robinson,J.P. (1994) Intracellular
hydrogen peroxide and superoxide anion detection in endothelial cells.
J. Leukoc. Biol., 55, 253–258.
23. Keston,A.S. and Brandt,R. (1965) The fluorometric analysis of ultramicro
quantities of hydrogen peroxide. Anal. Biochem., 11, 1–5.
24. Degli Esposti,M. and McLennan,H. (1998) Mitochondria and cells
produce reactive oxygen species in virtual anaerobiosis: relevance to
ceramide-induced apoptosis. FEBS Lett., 430, 338–342.
25. Bindokas,V.P., Jordán,J., Lee,C.C. and Miller,R.J. (1996) Superoxide
production in rat hippocampal neurons: selective imaging with hydroethidine.
J. Neurosci., 16, 1324–1336.
26. Betarbet,R., Sherer,T.B., MacKenzie,G., Garcia-Osuna,M., Panov,A.V.
and Greenamyre,J.T. (2000) Chronic systemic pesticide exposure reproduces
features of Parkinson’s disease. Nat. Neurosci., 3, 1301–1306.
27. Li,Y., Huang,T.T., Carlson,E.J., Melov,S., Ursell,P.C., Olson,J.L.,
Noble,L.J., Yoshimura,M.P., Berger,C., Chan,P.H. et al. (1995) Dilated
cardiomyopathy and neonatal lethality in mutant mice lacking manganese
superoxide dismutase. Nat. Genet., 11, 376–381.
28. Goslin,K.A. and Banker,G. (1998) Culturing nerve cells. In Rat Hippocampal
Neurons in Low-density Culture. MIT Press, Cambridge, MA.
29. King,M.P. and Attardi,G. (1996) Isolation of human cell lines lacking
mitochondrial DNA. Methods Enzymol., 264, 304–313.
30. Trounce,I. and Wallace,D.C. (1996) Production of transmitochondrial
mouse cell lines by cybrid rescue of rhodamine-6G pre-treated L-cells.
Somat. Cell Mol. Genet., 22, 81–85.
31. Johns,D.R. (1990) Improved molecular-genetic diagnosis of Leber’s
hereditary optic neuropathy. New Eng. J. Med., 323, 1488–1489.
32. Antonická,H., Floryk,D., Klement,P., Stratilová,L., Hermanská,J.,
Houstková,H., Kalous,M., Drahota,Z., Zeman,J. and Houstek,J. (1999)
Defective kinetics of cytochrome c oxidase and alteration of mitochondrial
membrane potential in fibroblasts and cytoplasmic hybrid cells with the
mutation for myoclonus epilepsy with ragged-red fibres (‘MERRF’) at
position 8344 nt. Biochem. J., 342, 537–544.