INFLUENCE OF MITOCHONDRIAL ENZYME DEFICIENCY
ON ADULT NEUROGENESIS IN MOUSE MODELS OF
N. Y. CALINGASAN,a* D. J. HO,a E. J. WILLE,a
M. V. CAMPAGNA,a J. RUAN,a M. DUMONT,a L. YANG,a
Q. SHI,b G. E. GIBSONb AND M. F. BEALa
aDepartment of Neurology and Neuroscience, Weill Medical College of
Cornell University, New York, NY 10065, USA
bDepartment of Neurology and Neuroscience, Weill Medical College of
Cornell University at Burke Medical Research Institute, White Plains,
NY 10605, USA
Abstract—Mitochondrial defects including reduction of a key
mitochondrial tricarboxylic acid cycle enzyme ?-ketoglutarate–
dehydrogenase complex (KGDHC) are characteristic of many
neurodegenerative diseases. KGDHC consists of ?-ketogluta-
rate dehydrogenase, dihydrolipoyl succinyltransferase (E2k),
and dihydrolipoamide dehydrogenase (Dld) subunits. We inves-
tigated whether Dld or E2k deficiency influences adult brain
neurogenesis using immunohistochemistry f o r t h e immature
neuron markers, doublecortin (Dcx) and polysialic acid–neu-
ral cell adhesion molecule, as well as a marker for prolifera-
tion, proliferating cell nuclear antigen (PCNA). Both Dld- and
E2k-deficient mice showed reduced Dcx-positive neuroblasts
in the subgranular zone (SGZ) of the hippocampal dentate
gyrus compared with wild-type mice. In the E2k knockout
mice, increased immunoreactivity for the lipid peroxidation
marker, malondialdehyde occurred in the SGZ. These alter-
ations did not occur in the subventricular zone (SVZ). PCNA
staining revealed decreased proliferation in the SGZ of E2k-
deficient mice. In a transgenic mouse model of Alzheimer’s
disease, Dcx-positive cells in the SGZ were also reduced
compared with wild type, but Dld deficiency did not exacer-
bate the reduction. In the malonate lesion model of Hunting-
ton’s disease, Dld deficiency did not alter the lesion-induced
increase and migration of Dcx-positive cells from the SVZ
into the ipsilateral striatum. Thus, the KGDHC subunit defi-
ciencies associated with elevated lipid peroxidation selec-
tively reduced the number of neuroblasts and proliferating
cells in the hippocampal neurogenic zone. However, these
mitochondrial defects neither exacerbated certain patholog-
ical conditions, such as amyloid precursor protein (APP)
mutation-induced reduction of SGZ neuroblasts, nor inhib-
ited malonate-induced migration of SVZ neuroblasts. Our
findings support the view that mitochondrial dysfunction can
influence the number of neural progenitor cells in the hip-
pocampus of adult mice. © 2008 IBRO. Published by Elsevier
Ltd. All rights reserved.
Key words: mitochondrial enzyme, neurogenesis, neuro-
blasts, doublecortin, polysialic acid–neural cell adhesion
molecule, lipid peroxidation, subgranular zone.
Reduction of the mitochondrial enzyme ?-ketoglutarate–
dehydrogenase complex (KGDHC) is implicated in the
pathogenesis of many neurodegenerative diseases includ-
ing Alzheimer’s and Huntington’s disease (Gibson et al.,
1988; Klivenyi et al., 2004). Mammalian KGDHC is com-
posed of ?-ketoglutarate dehydrogenase (E1k subunit,
EC22.214.171.124), dihydrolipoyl succinyltransferase (E2k subunit,
EC 126.96.36.199) and dihydrolipoylamide dehydrogenase (Dld)
(encoded by Dld gene; E3 subunit, EC 188.8.131.52). We have
previously reported that in models of Huntington’s and
Parkinson’s disease, Dld-deficient mice show increased
vulnerability to the mitochondrial toxins 1-methyl-4-phenyl-
1,2,3,6-tetrahydropyridine (MPTP), malonate and 3-nitro-
propionic acid (Klivenyi et al., 2004). These studies sup-
port the idea that mitochondrial defects may contribute to
the pathogenesis of neurodegenerative diseases. The im-
pact of mitochondrial KGDHC defects on the generation of
neuronal precursors (neurogenesis) has not been ex-
plored. In the current study, we tested whether a deficiency
of Dld or E2k influences adult neurogenesis in mice including
mouse models of Alzheimer’s and Huntington’s disease.
In normal adult mammals, persistent neurogenesis oc-
curs in the subgranular zone (SGZ) of the hippocampal den-
tate gyrus as well as the subventricular zone (SVZ) that lines
the lateral ventricles in the forebrain. Impairment of neuro-
genesis has been associated with increased oxidative stress.
For example, a previous study documented that oxidative
stress reduces the number of hippocampal neural precursor
cells both in vitro and in vivo, and that this reduction could be
reversed with the antioxidant ?-lipoic acid (Limoli et al., 2004).
Oxidative stress can also inhibit the proliferative potential of
neural precursor cells derived from rat hippocampus (Limoli
et al., 2006). We previously reported that Dld-deficient mice
have significantly increased levels of the lipid peroxidation
marker malondialdehyde (MDA) in the striatum (Klivenyi et
al., 2004) raising the possibility that neurogenesis may be
altered in these mice.
The current investigation focuses on the impact of
KGDHC subunit deficiencies on the numbers of neural
progenitor cells in neurogenic zones of adult mouse brains
in health and in disease models. The influence of KGDHC
subunit deficiencies on differentiation of progenitor cells
and integration of newborn neurons is beyond the scope of
the present study. The term neurogenesis used in this
*Corresponding author: Tel: ?1-212-746-4969; fax: ?1-212-746-8741.
E-mail address: email@example.com (N. Y. Calingasan).
Abbreviations: APP, amyloid precursor protein; BrdU, bromodeoxyuri-
dine; BSA, bovine serum albumin; Dcx, doublecortin; Dld, dihydrolipo-
amide dehydrogenase; E2k, dihydrolipoyl succinyltransferase; GFAP,
glial fibrillary acidic protein; HPLC, high performance liquid chroma-
tography; KGDHC, ?-ketoglutarate–dehydrogenase complex; MDA,
malondialdehyde; PBS, phosphate-buffered saline; PCNA, proliferat-
ing cell nuclear antigen; PSA-NCAM, polysialic acid–neural cell adhe-
sion molecule; SGZ, subgranular zone; SVZ, subventricular zone.
Neuroscience 153 (2008) 986–996
0306-4522/08$32.00?0.00 © 2008 IBRO. Published by Elsevier Ltd. All rights reserved.
report pertains mainly to the number of immature neurons
in the neurogenic zones of adult mouse brain.
Heterozygous Dld knockout mice (Dld?/?; C57BL/6) have
been developed and characterized (Johnson et al., 1997; ob-
tained from Dr. Mulchand S. Patel from the State University of
Buffalo, Buffalo, NY, USA). Deficient in murine Dld, these mice
have reduced brain KGDHC activity compared with controls.
Mice deficient in the E2k subunit (E2k?/?; C57BL/6 and
129SV/EV hybrid) were obtained from Lexicon Pharmaceuti-
cals (The Woodlands, TX, USA). E2k?/? (n?10; five male, five
female) and Dld?/? (n?6; three male, three female) mice and
their respective wild-type littermate controls (n?10 for E2k; five
male, five female; n?5 for Dld; three male, two female) were
used in these studies. Both the Dld?/? and E2k?/? mice have
no overt phenotype.
The Tg19959 mouse model of amyloid plaque formation sim-
ilar to Alzheimer’s disease was obtained from Dr. George A.
Carlson, McLaughlin Research Institute, Great Falls, MT, USA.
These mice express two amyloid precursor protein 695 (APP695)
mutations (KM670/671NL and V717F), under control of the PrP
gene promoter. To test the impact of Dld deficiency on neurogen-
esis in Alzheimer transgenic mice, Dld?/? mice were crossed
with Tg19959 mice to generate Alzheimer transgenic mice with
(n?5) or without (n?6) Dld. All mice were maintained under
constant temperature (70 °F), humidity (50%) and 12-h light/
dark cycle with food and water available ad libitum. Mice were
killed at 4 months as described below.
To determine the influence of KGDHC deficiency on the genera-
tion of doublecortin (Dcx) -positive immature neurons in a mouse
model of Huntington’s disease, we performed brain lesion exper-
iments using the mitochondrial complex II inhibitor, malonate. The
selective neuropathology in human Huntington’s disease brain
can be mimicked in laboratory animals by direct intrastriatal infu-
sion of malonate. Thus, malonate lesioning in rodents has been
widely used as a model of Huntington’s disease (Beal et al., 1993).
Four-month-old wild-type (n?11) and E2k?/? (n?10) mice were
anesthetized with isoflurane inhalation (1–3%). Mice received a
unilateral stereotaxically-guided injection of malonate (1.5 ?l in
1.0 ?l saline) into the left striatum using coordinates (anterior to
bregma ?0.5 mm; lateral ?1.5 mm; ventral from dura ?4 mm)
ascertained from the mouse brain atlas of Paxinos and Franklin
(2003). The needle was left in place for 2–5 min to prevent
backflow of infusate up the needle track. Mice were allowed to
recover, and on day 7 post-surgery, mice were killed.
Tissue preparation for histological analysis
Killing was performed under deep anesthesia with sodium pento-
barbital (120 mg/kg i.p.) by transcardial perfusion with 0.9% sa-
line, followed by 4% paraformaldehyde. The brains were removed,
post-fixed in the same fixative, cryoprotected in 30% sucrose and
sectioned coronally (50 ?m) using a cryostat. All animal proce-
dures were carried out in compliance with the National Institutes of
Health Guide for the Care and Use of Laboratory Animals and
were approved by the Institutional Animal Care and Use Commit-
tee of Cornell University. In addition, the investigators took all
steps to minimize the number of animals used and their suffering
in conducting these studies.
Dcx and polysialic acid-neural cell adhesion molecule (PSA-
NCAM) served as markers of immature neurons. Dcx is a 43–53
kDa microtubule-associated protein required for normal neural
migration, cortical layering and interneuron migration during de-
velopment. It is widely expressed by migrating neuronal precur-
sors of the developing CNS (Gleeson et al., 1998; Friocourt et al.,
2007) but not by cells expressing antigens specific to glia or
undifferentiated cells (Rao and Shetty, 2004). Although the Dcx-
expressing population in the developing and adult brain contains
multipotential precursors in addition to neuronal-lineage cells
(Walker et al., 2007), Dcx has become a suitable and widely used
immunohistological marker of migrating neuroblasts, and a means
to perform relative quantitation of the level of neurogenesis under
normal and pathological conditions (Couillard-Despres et al.,
2005). Furthermore, the number of Dcx-positive cells directly cor-
relates with cognitive function including spatial memory and dis-
crimination learning in aged canine brain (Siwak-Tapp et al.,
2007). PSA is a linear homopolymer of ?2-8-N acetylneuraminic
acid whose major carrier in vertebrates is NCAM. PSA on NCAM
plays a role in various forms of neural plasticity including adult
neurogenesis (Bonfanti, 2006). Thus, the use of Dcx and PSA-
NCAM immunostaining in studies of adult neurogenesis has be-
come increasingly popular. For evaluating cell proliferation, immu-
nostaining for proliferating cell nuclear antigen (PCNA) was used.
Although the conventional technique for studying proliferation is
by in vivo labeling with bromodeoxyuridine (BrdU), PCNA was
used in this study avoiding the disadvantages of BrdU such as
toxicity, requirement for careful and accurate control of dosage,
and nonspecific labeling (Gould and Gross, 2002).
For histological demonstration of lipid peroxidation and gen-
eration of immature neurons in neurogenic zones of the brain, a
modified avidin–biotin–peroxidase immunohistochemistry was
used. Free-floating sections were pretreated with 3% H2O2in
0.1 M sodium phosphate-buffered saline (PBS) for 30 min. The
sections were incubated sequentially in (a) 1% bovine serum
albumin (BSA) and 0.2% Triton X-100 in PBS for 30 min, (b)
affinity-purified goat polyclonal antibody against a peptide map-
ping at the C-terminus of Dcx (Santa Cruz Biotechnology, Santa
Cruz, CA, USA; 1:200), mouse monoclonal antibody against PSA-
NCAM (Chemicon, Temecula, CA, USA; 1:300), rabbit polyclonal
antibody against PCNA (Santa Cruz Biotechnology; 1:1000), poly-
clonal rabbit anti-MDA (provided by Dr. Craig E. Thomas from
Hoechst-Marion-Roussel; Hall et al., 1997; 1:1000), or polyclonal
rabbit anti–glial fibrillary acidic protein (GFAP; DAKO, Carpinteria,
CA, USA; 1:1000), (c) appropriate secondary antibody, either
biotinylated anti-rabbit IgG, anti-goat IgG or anti-mouse IgM (Vec-
tor Laboratories, Burlingame, CA, USA) diluted at 1:200 in PBS/
0.5% BSA for 18 h, and (d) avidin–biotin–peroxidase complex
(Vector) with both reagents A and B diluted at 1:200 in PBS for 1 h.
The immunoreaction product was visualized using 3,3=-diamino-
benzidine tetrahydrochloride dihydrate (DAB, Vector).
Methodological specificity control was carried out by replacing
the primary antibody with PBS/0.5% BSA. Since Dcx immuno-
staining was used predominantly, immunological specificity con-
trols were carried out for Dcx antibody by incubating the tissue
sections either in goat IgG or in antibody that was preincubated
with the Dcx peptide (Santa Cruz Biotechnology).
To determine whether degeneration of immature neurons in the
SGZ occurs before maturation, silver staining with the FD Neuro-
silver kit (FD Neurotechnologies, Inc., Baltimore, MD, USA) was
used. The principle of this staining technique is based on the
findings that some components of degenerating neurons, such as
lysosomes, axons, dendrites and terminals become particularly
argyrophilic. As a positive control, sections from mice with 3-ni-
N. Y. Calingasan et al. / Neuroscience 153 (2008) 986–996 987
tropropionic acid–induced striatal neurodegeneration from an-
other study were stained simultaneously with the tissues from the
For quantitation of Dcx-positive immature neurons and PCNA-
labeled cells, we used a stereological technique based upon
unbiased principles of systematic uniformly random sampling. The
optical fractionator method in the Stereo Investigator (v3.45) soft-
ware program (Microbrightfield, Burlington, VT, USA) was used to
obtain counts of Dcx- or PCNA-labeled cells in the SGZ. Slide
labels were concealed before the analysis so the experimenter
was blind to genotype until completion of the quantification. The
same measure was taken for non-quantifiable analysis of MDA
Fig. 1. Dcx immunoreactivity in the SGZ of wild-type control (A, A=) and E2k?/? (B, B=) mice showing a reduction of Dcx-labeled cells in the latter.
High magnification photos depict reductions in the number and dendritic branching of Dcx-labeled neurons in the E2k?/? mice (B=) compared with
wild-type control (A=). Stereological evaluation (C) revealed a significant loss of Dcx-positive cells in the SGZ of E2k?/? mice compared with wild-type
controls (*** P?0.001). The stereological cell counts represent the total number of cells in the SGZ within the specified brain region included in the
analysis (see Experimental Procedures).
N. Y. Calingasan et al. / Neuroscience 153 (2008) 986–996 988
staining intensity. For Dcx analysis, only intensely labeled imma-
ture neurons with distinct nuclei were counted. For PCNA, only
cells with robust nuclear immunoreactivity were considered posi-
tive. SGZ was defined as the region between the granule cell layer
and the hilus of the dentate gyrus. All labeled cells in the SGZ
were counted except those touching the granule cell layer on the
border of the molecular layer. For the SGZ analysis, cell counts
were made in four serial sections (250 ?m apart) per mouse,
beginning rostrally from the level of bregma ?1.7 mm through
bregma ?2.45. The size of the x–y sampling grid was 140 ?m.
The counting frame thickness was 14 ?m. In this study, the SVZ
was defined as the layer of cells lining the lateral ventricle covering
the entire dorsoventral distance. For the SVZ analysis, cell counts
were made in four serial sections (250 ?m apart) per mouse
beginning rostrally from the level of bregma 1.18 through bregma
0.43. The size of the x–y sampling grid was 140 ?m. The counting
frame thickness was 14 ?m. The stereological cell counts repre-
sent the total number of cells in the SGZ or SVZ within the
specified brain region analyzed.
Data are expressed as mean?standard error of the mean
(S.E.M.). Statistical analysis of the data was performed using
one-way analysis of variance followed by the Student-Newman-
Keuls post test or unpaired Student’s t-test, when appropriate,
using the GraphPad Instat software (San Diego, CA, USA). A P
value of ?0.05 was considered statistically significant.
The specificity of Dcx immunostaining was verified by (a)
performing competition experiments on adjacent brain sec-
tions from a wild-type control mouse, and (b) incubating
the sections in goat IgG. While intense Dcx immunoreac-
tivity occurred in neuroblasts of the SGZ as in published
reports, incubation of adjacent sections with antibody that
was preabsorbed with the Dcx peptide, or incubation with
goat IgG in place of the Dcx antibody abolished the stain-
ing (data not shown).
We first sought to determine the influence of E2k de-
ficiency on the number of immature neurons in the SGZ of
4 month-old mice. Dcx-immunoreactive neuroblasts oc-
curred in the SGZ of both wild-type controls (Fig. 1A, A=)
and E2k?/? mice (Fig. 1B, B=). Stereological quantitation
of the Dcx-positive cells revealed a striking difference in
the number of immature neurons in E2k?/? mice relative
to the wild-type controls (Fig. 1C). The number of Dcx-
positive neuroblasts in the SGZ of E2k?/? mice
(4234?336) was significantly reduced compared with wild-
type controls (7351?588; P?0.001). Morphologically,
Dcx-positive cells in the SGZ of control mice displayed
complex arborization of dendrites toward the molecular
layer of the dentate gyrus (Fig. 1A=). In E2k?/? mice, the
Dcx-labeled cells exhibited a reduced dendritic arboriza-
tion pattern (Fig. 1B=).
marker, PSA-NCAM revealed a similar pattern of staining
as Dcx in the SGZ of controls (Fig. 2A, A=) and E2k?/?
mice (Fig. 2B, B=). For this reason, quantitative analysis
and subsequent experiments were conducted using Dcx.
Dld deficiency also influenced the number of neuro-
blasts in the SGZ. Compared with wild-type controls
(8166?693; P?0.01; Fig. 3A, A=, E), Dld?/? mice (Fig.
3B, B=, E) showed significantly less Dcx-labeled cells
(5410?885). We next determined whether Dld deficiency
influences the reduction of hippocampal neurogenesis in
a mouse model of Alzheimer’s disease. Previous studies
have documented decreased hippocampal neurogen-
Fig. 2. PSA-NCAM immunoreactivity in the SGZ of wild-type control (A, A=) and E2k?/? (B= B=) mice revealed a similar pattern of staining as
N. Y. Calingasan et al. / Neuroscience 153 (2008) 986–996989
esis in Alzheimer transgenic mice (Donovan et al., 2006;
Zhang et al., 2006). In the present study, we crossed the
Tg19959 model of Alzheimer’s disease with Dld deficient
mice to generate Alzheimer mice with or without Dld
deficiency. Consistent with other models of Alzheimer’s
disease, the Tg19959 mice showed a striking reduction
in Dcx-immunoreactive cells in the SGZ (3274?483; Fig.
3C, C=, E) compared with wild-type controls (P?0.001).
Interestingly, in Dld-deficient Tg19959 mice (Fig. 3D, D=,
E), the number of Dcx-labeled neuroblasts (3586?419)
did not differ significantly from Tg19959 mice without Dld
deficiency (P?0.05). Thus, Dld deficiency neither exac-
erbated nor ameliorated the impairment of adult hip-
pocampal neurogenesis in a mouse model of Alzhei-
To test whether E2k deficiency also influences prolif-
eration in the SGZ, sections adjacent to those used for Dcx
staining were processed for PCNA immunohistochemistry
(wild type, Fig. 4A, A=; E2k?/?, 4B, B=) and stereological
analysis (Fig. 4C). PCNA immunoreactivity occurred as
irregular or oval shaped intensely stained nuclei (Fig. 4A,
A=; 4B, B=). PCNA-positive cells were distributed singly or
clustered throughout the rostrocaudal extent of the SGZ.
Quantification revealed that E2k deficiency significantly
diminished the number of PCNA-labeled cells (1124?77)
compared with wild-type mice (1796?128; P?0.001).
To test whether the reduction of Dcx-labeled immature
neurons in the SGZ of E2k-deficient mice is due to a shift
of the phenotypic fate of progenitors toward the astrocytic
lineage, GFAP immunostaining was performed. Fig. 5
shows that there was no significant difference between the
number of GFAP-labeled astrocytes in wild-type controls
(8274?214; Fig. 5A, C) and that of E2k?/? mice
(8217?386; P?0.05; Fig. 5B, C).
We also investigated whether increased immature
neuronal degeneration contributed to the reduction in the
number and dendritic arborization of Dcx-positive neurons
in E2k?/? and Dld?/? mice. Silver staining of SGZ sec-
tions from E2k?/? (Fig. 6B), Dld?/? (Fig. 6D) mice, and
their respective controls (Fig. 6A, C) did not reveal any
increased argyrophilic structures similar to those found in
3-nitropropionic acid–lesioned striatal sections (Fig. 6E)
Fig. 3. Dcx immunoreactivity in the SGZ of wild-type control (A, A=), Dld?/? (B, B=), an Alzheimer mouse model Tg19959 (C, C=) and Tg19959 without
Dld (D, D=). Stereological quantitation (E) shows a significant reduction of Dcx-positive cells in the Dld?/? and Tg19959 mice (** P?0.01, *** P?0.001
vs. wild type). Dld deficiency did not affect the number of Dcx-positive cells in Tg19959.
N. Y. Calingasan et al. / Neuroscience 153 (2008) 986–996990
which are known to exhibit darkly stained argyrophilic neu-
ron bodies and terminals.
Both the SGZ and SVZ were tested for in situ levels of
lipid peroxidation. Using high performance liquid chroma-
tography (HPLC), we previously showed that Dld-deficient
mice have significantly increased levels of the lipid peroxi-
dation marker, MDA in the striatum, as measured by
amounts of thiobarbituric acid–MDA adducts (Klivenyi et
al., 2004). In the current study, we localized oxidatively
damaged cells within the hippocampal formation by immu-
nohistochemistry using a polyclonal antibody to MDA-mod-
ified proteins. We employed this in situ technique to enable
direct comparison of the dentate gyrus of control and
E2k?/? mice. Consistent with the HPLC results, we found
elevated MDA immunoreactivity diffusely in the dentate
gyrus including the SGZ of E2k?/? (Fig. 7B) compared
with wild-type controls (Fig. 7A). Intensely stained cells
were also found in the SGZ layer of E2k?/? mice (Fig. 7B
inset). No MDA immunoreactivity occurred in the SVZ of
either the E2k?/? (Fig. 7D) or wild-type control (Fig. 7C)
While neurogenesis in the SGZ was reduced in both
the E2k?/? and Dld?/? mice, the generation of immature
neurons in the SVZ did not appear to be affected by the
mitochondrial defects in this study. Fig. 8 shows that the
numbers of Dcx-positive immature neurons in the SVZ of
E2k?/? (7811?460; Fig. 8B, C) and Dld?/? (8429?481;
Fig. 8E, F) mice did not significantly differ from their re-
spective wild-type controls (9103?668, wild type for E2k;
9114?1143, wild type for Dld; Fig. 8A, C, D, F; P?0.05 for
Finally, we investigated the neurogenic response of the
progenitor cells in the SVZ to neuron loss induced by
intrastriatal malonate, and tested the impact of E2k defi-
ciency on this response. The close proximity of the SVZ to
the adjacent striatum (caudate putamen) makes this model
ideal for these experiments. Intrastriatal malonate injec-
tions in rats and mice produce lesions that mimic the
neuropathology of Huntington’s disease (Beal et al., 1993;
Greene et al., 1993). Seven days following malonate in-
jection, a well-defined area of neuron loss occurred in the
ipsilateral hemisphere of wild-type control mice (lesion
Fig. 4. PCNA immunoreactivity in the SGZ of wild-type control (A, A=) and E2k?/? (B, B=) mice showing a reduction of PCNA-labeled progenitor cells
in the latter. High magnification photos (A=, B=) show labeled nuclei occurring singly or in clusters in the SGZ. Stereological evaluation (C) revealed
a significant loss of PCNA-immunoreactive cells in the SGZ of E2k?/? mice compared with wild-type controls (*** P?0.001).
N. Y. Calingasan et al. / Neuroscience 153 (2008) 986–996 991
volume?1.3?0.4 mm3). In the contralateral hemisphere of
wild-type controls, robust Dcx immunoreactivity was con-
fined to the SVZ lining the lateral ventricle, but virtually
absent in the striatum (Fig. 9). In the ipsilateral side how-
ever, Dcx labeling occurred in the SVZ as well as in many
cells scattered in the dorsomedial striatum, extending from
the SVZ into the site of the lesion. Some of the malonate-
generated Dcx-positive neuroblasts appeared fusiform
with elongated processes while others had several pro-
cesses extending in different directions.
To test whether E2k deficiency suppressed the mal-
onate-induced increases in neurogenesis, we examined
the lesioned striatum of E2k?/? mice (lesion volume?
2.6?0.5 mm3) in comparison with the lesioned hemisphere
of wild-type controls. The pattern of Dcx staining in the
E2k?/? mice was similar to that of controls (Fig. 9). Quan-
tification of the immature neurons in the lesioned hemi-
spheres outside the SVZ revealed that E2k deficiency did
not impair the malonate-induced increases in neurogen-
esis. The number of Dcx-positive immature neurons in the
ipsilateral striatum of wild-type mice (23.6?1 cells per
section) did not differ significantly from that of E2k?/?
(23.8?1 cells per section). As in wild-type mice, malonate
lesion also induced astrogliosis in E2k?/? mice (data not
shown). These studies demonstrate that malonate lesion
enhanced the generation of immature neurons in the dam-
aged striatum with or without E2k deficiency.
Our findings show that KGDHC subunit deficiencies selec-
tively reduced the number of Dcx-labeled immature neu-
rons as well as PCNA-immunoreactive proliferating cells in
the hippocampal SGZ. This study is the first demonstration
that mitochondrial enzyme defects can reduce the number
of neural progenitor cells in adult mice.
KGDHC deficiency is associated with increased lipid
peroxidation as evidenced by elevation of the lipid peroxi-
dation marker MDA in brain of Dld?/? mice as measured
by HPLC (Klivenyi et al., 2004) and by immunohistochem-
istry in the E2k?/? mice in the current study. In a rat model
of chronic alcoholism, ethanol, which can damage tissues
by lipid peroxidation (Montoliu et al., 1995; Mi et al., 2000),
selectively impairs hippocampal neurogenesis (Herrera et
al., 2003). This reduction of neurogenesis can be com-
pletely mitigated by the organo-selenium antioxidant eb-
selen, a drug that decreases oxidative stress through in-
hibition of peroxynitrite generation and lipid peroxidation
(Briviba et al., 1996; Herrera et al., 2003). In the D-galac-
tose model of neurotoxicity in mice, increased MDA levels
Fig. 5. GFAP immunoreactivity in representative sections through the dentate gyrus of wild-type control (A) and E2k?/? (B) mice. Stereological
counts of GFAP-positive astrocytes in the SGZ (C) revealed no statistically significant difference between the two groups (P?0.05).
N. Y. Calingasan et al. / Neuroscience 153 (2008) 986–996992
are associated with reduced numbers and migration of
new neurons in the SGZ (Zhang et al., 2005; Cui et al.,
2006). Furthermore, in vitro studies show that lipid peroxi-
dation is involved in the developmental impairment of neu-
ronal progenitor cells by amyloid ?1–40 (Mazur-Kolecka et
al., 2006). It is conceivable, therefore, that the reduced
Fig. 6. Silver staining of wild-type (A) and E2k?/? (B) mice, and wild-type (C) and Dld?/? (D) mice showing absence of degenerating immature
neuron profiles in mice with mitochondrial enzyme defects. A positive control section through 3-nitropropionic acid–lesioned caudate putamen (E)
displayed dense black silver grains in degenerating cell bodies and terminals.
Fig. 7. MDA immunoreactivity in the hippocampal dentate gyrus (A, B) and SVZ (C, D) of wild-type control (A, C) and E2k?/? (B, D) mice. Elevated
MDA immunostaining was found in the dentate gyrus but not the SVZ of E2k?/? mice compared with wild-type control. Although staining was diffuse
in the dentate gyrus, intense cellular localization of MDA staining occurred in the SGZ of E2k?/? mice (inset). CPu, caudate putamen.
N. Y. Calingasan et al. / Neuroscience 153 (2008) 986–996 993
neurogenesis in the SGZ of E2k?/? mice is due to the
increased lipid peroxidation in the hippocampal dentate
gyrus. This idea is supported by the absence of neurogen-
esis reduction in the SVZ of Dld?/? or E2k?/? where
MDA was virtually undetectable by immunohistochemistry
in the present study. In Alzheimer transgenic mice, the
reduced neurogenesis could also be attributed in part to
lipid peroxidation. A previous study showed that in the
Tg2576 Alzheimer mouse model, lipid peroxidation in the
hippocampus is increased compared with wild-type mice
(Pratico et al., 2001). Whether the increased lipid peroxi-
dation is a direct consequence of Dld or E2k deficiency is
unclear. It is possible that the increased lipid peroxidation
associated with KGDHC deficiency inhibited proliferation
of neural progenitor cells, as evidenced by the reduction of
PCNA-labeled cells, and also prevented the growth and
maturation of neuroblasts by causing cytoskeletal alter-
ations. Further studies are required to elucidate the exact
mechanisms involved in the reduction of immature neu-
rons in mice with KGHDC subunit deficiency. Neverthe-
less, the current findings are consistent with the view that
mitochondrial defects and oxidative stress can inhibit adult
While dendritic branches of Dcx-labeled neuroblasts
extended radially toward the molecular layer of the dentate
gyrus in the wild-type mice, dendritic branching was ap-
parently reduced in the SGZ of E2k?/? and Dld?/? mice.
Based on our silver staining results, this reduction did not
result from degeneration. Argyrophilic profiles suggestive
of degenerating neuroblast processes were not detected in
mice lacking either E2k or Dld. Thus, the reduced arboriza-
tion pattern most likely reflects inhibition of maturation of
neuroblasts, possibly due to oxidative stress.
GFAP staining analysis revealed that the number of
astrocytes in the SGZ was not elevated in mice lacking E2k
compared with wild-type controls. This finding suggests
that the reduction of neuroblasts in the E2k?/? mice was
not a consequence of a shift of progenitor cell differentia-
tion toward the astrocytic lineage.
The hippocampal SGZ is a brain region that is evi-
dently vulnerable to mitochondrial KGDHC deficiency. Our
observations are consistent with the recent demonstration
that thiamine deficiency simultaneously impairs hippocam-
pal neurogenesis and induces cognitive dysfunction in
mice (Zhao et al., 2008). KGDHC is a thiamine pyrophos-
phate–dependent enzyme of oxidative metabolism. It is
well established that thiamine deficiency is characterized
by reduced activity of KGDHC in brain (Gibson et al.,
In several mammalian species such as rat and vole,
adult hippocampal neurogenesis is modulated by both
gender and endogenous levels of estradiol (Galea et al.,
2006). However, a recent study documented that adult
mice do not have gender differences in hippocampal pro-
liferation or neurogenesis (Lagace et al., 2007). These
results validate the use of male and female mice in adult
neurogenesis studies. In the current experiments, male
and female mice were pooled for each group.
Owing to the reduction of immature neurons in the
Dld?/? and AD transgenic mice, one would expect an
Fig. 8. Dcx immunoreactivity in the SVZ of wild-type (A) and E2k?/? (B) mice, and wild-type (D) and Dld?/? (E) mice with high magnification
photomicrographs of labeled neurons (insets). Quantification of Dcx-immunoreactive immature neurons revealed no significant reduction in either
E2k?/? (C) or Dld?/? (F) mice.
N. Y. Calingasan et al. / Neuroscience 153 (2008) 986–996 994
exacerbation of the reduction in AD transgenic mice that
lack Dld. However, the level of Dcx immunoreactivity did
not differ significantly between AD transgenic mice with
and without Dld. Compensatory phenomena most likely
play a role in counteracting the mitochondrial enzyme def-
icit. Another possibility is that the pathological changes in
Dld deficiency and amyloid precursor protein (APP) muta-
tion are induced by the same factor such as lipid peroxi-
dation, and after reaching a certain level, these changes
do not go any further. These observations suggest that
even in the presence of a mitochondrial KGDHC defi-
ciency, AD transgenic mouse brains maintain a capacity
The results from our malonate lesion studies confirm
the view that pathological stimuli can induce endogenous
neurogenesis in regions such as the striatum where adult
neurogenesis is non-existent under physiological condi-
tions. Our findings agree with a previous study on rats
lesioned with quinolinic acid, an excitatory amino acid
agonist that has been extensively used as another animal
model of Huntington’s disease (Beal et al., 1986). Tatters-
field and colleagues (2004) demonstrated that quinolinic
acid-induced cell loss in rat striatum increases SVZ neu-
rogenesis and migration of neuroblasts to the damaged
areas. Here we present evidence that malonate lesions
also led to the generation of Dcx-labeled cells migrating
from the SVZ into the damaged striatum of wild-type con-
trol mice. Another possibility is that malonate enhanced
neuronal precursor migration toward the lesioned area
without change in new neuron generation. Surprisingly,
mice with E2k deficiency also exhibited a similar response.
It is possible that in the presence of an additional insult
such as an excitotoxic lesion, the brain can overcome the
compromised mitochondrial KGDHC function, and trigger
compensatory mechanisms to sustain endogenous neuro-
genesis. This may explain why the same level of neuro-
genesis occurred in lesioned wild-type and E2k?/? mice.
More studies are necessary to understand the mechanism
by assessing the rate of proliferation of neural progenitors
and tracking their phenotypic fate into neurons or glia.
In summary, both Dld and E2k deficiencies selectively
reduced the numbers of young neurons in the SGZ, but
these mitochondrial defects do not exacerbate certain
pathological conditions, such as APP mutation-induced
reduction of SGZ neuroblasts or malonate excitotoxicity-
Fig. 9. Dcx immunoreactivity in the caudate putamen (CPu; striatum) and the SVZ lining the lateral ventricle (LV) of wild-type and E2k?/? mice with
unilateral malonate lesion. Dcx-positive cells migrating from the SVZ to the lesion site (arrows) occurred in both wild-type and E2k?/? mice, and are shown
at high magnification in the lower panel. Labeled cells are virtually absent in the intact caudate putamen. cc, Corpus callosum; MS, medial septal nucleus.
N. Y. Calingasan et al. / Neuroscience 153 (2008) 986–996995
induced migration of SVZ neuroblasts. Our findings are Download full-text
consistent with the view that mitochondrial dysfunction can
influence adult neurogenesis.
Acknowledgments—This work was supported by a grant from the
National Institutes of Health/National Institute on Aging (AG
14930) to Dr. M. F. Beal.
Beal MF, Brouillet E, Jenkins BG, Henshaw R, Rosen BR, Hyman BT
(1993) Age-dependent striatal excitotoxic lesions produced by the
endogenous mitochondrial inhibitor malonate. J Neurochem
Beal MF, Kowall NW, Ellison DW, Mazurek MF, Schwartz KJ, Martin
JB (1986) Replication of the neurochemical characteristics of Hun-
tington’s disease by quinolinic acid. Nature 321:168–171.
Bonfanti L (2006) PSA-NCAM in mammalian structural plasticity and
neurogenesis. Prog Neurobiol 80:129–164.
Briviba K, Roussyn I, Sharov VS, Sies H (1996) Attenuation of oxida-
tion and nitration reactions of peroxynitrite by selenomethionine,
selenocystine and ebselen. Biochem J 319:13–15.
Couillard-Despres S, Winner B, Schaubeck S, Aigner R, Vroemen M,
Weidner N, Bogdahn U, Winkler J, Kuhn H-G, Aigner L (2005)
Doublecortin expression levels in adult brain reflect neurogenesis.
Eur J Neurosci 21:1–14.
Cui X, Zuo P, Zhang Q, Li X, Hu Y, Long J, Packer L, Liu J (2006)
Chronic systemic D-galactose exposure induces memory loss,
neurodegeneration, and oxidative damage in mice: protective ef-
fects of R-alpha-lipoic acid. J Neurosci Res 84:647–654.
Donovan MH, Yazdani U, Norris RD, Games D, German DC, Eisch AJ
(2006) Decreased adult hippocampal neurogenesis in the PDAPP
mouse model of Alzheimer’s disease. J Comp Neurol 495:70–83.
Friocourt G, Liu JS, Antypa M, Rakic S, Walsh CA, Parnavelas JG
(2007) Both doublecortin and doublecortin-like kinase play a role in
cortical interneuron migration. J Neurosci 27:3875–3883.
Galea LA, Spritzer MD, Barker JM, Pawluski JL (2006) Gonadal hor-
mone modulation of hippocampal neurogenesis in the adult. Hip-
Gibson GE, Ksiezak-Reding H, Sheu K-FR, Mykytyn V, Blass JP
(1984) Correlation of enzymatic, metabolic and behavioral deficits
in thiamine deficiency and its reversal. Neurochem Res 9:
Gibson GE, Sheu K-FR, Blass JP, Baker A, Carlson KC, Harding B,
Perrino P (1988) Reduced activities of thiamine-dependent en-
zymes in the brains and peripheral tissues of patients with Alzhei-
mer’s disease. Arch Neurol 45:836–840.
Gleeson JG, Allen KM, Fox JW, Lamperti ED, Berkovic S, Scheffer I,
Cooper EC, Dobyns WB, Minnerath SR, Ross ME, Walsh CA
(1998) Doublecortin, a brain-specific gene mutated in human X-
linked lissencephaly and double cortex syndrome, encodes a pu-
tative signaling protein. Cell 92:63–72.
Gould E, Gross CG (2002) Neurogenesis in adult mammals: some
progress and problems. J Neurosci 22:619–623.
Greene JG, Porter RH, Eller RV, Greenamyre JT (1993) Inhibition of
succinate dehydrogenase by malonic acid produces an excitotoxic
lesion in rat striatum. J Neurochem 61:1151–1154.
Hall ED, Oostveen JA, Andrus PK, Anderson DK, Thomas CE (1997)
Immunocytochemical method for investigating in vivo neuronal
oxygen radical-induced lipid peroxidation. J Neurosci Methods
Herrera DG, Yague AG, Johnsen-Soriano S, Bosch-Morell F, Collado-
Morente L, Muriach M, Romero FJ, Garcia-Verdugo JM (2003)
Selective impairment of hippocampal neurogenesis by chronic al-
coholism: protective effects of an antioxidant. Proc Natl Acad Sci
U S A 100:7919–7924.
Johnson MT, Yang H-S, Magnuson T, Patel MS (1997) Targeted
disruption of the murine dihydrolipoamide dehydrogenase gene
(Dld) results in perigastrulation lethality. Proc Natl Acad Sci U S A
Klivenyi P, Starkov AA, Calingasan NY, Gardian G, Browne SE, Yang
L, Bubber P, Gibson GE, Patel MS, Beal MF (2004) Mice deficient
in dihydrolipoamide dehydrogenase show increased vulnerability
to MPTP, malonate and 3-nitropropionic acid neurotoxicity. J Neu-
Lagace DC, Fischer SJ, Eisch AJ (2007) Gender and endogenous
levels of estradiol do not influence adult hippocampal neurogen-
esis in mice. Hippocampus 17:175–180.
Limoli CL, Rola R, Giedzinski E, Mantha S, Huang TT, Fike JR (2004)
Cell-density-dependent regulation of neural precursor cell function.
Proc Natl Acad Sci U S A 101:16052–16057.
Limoli CL, Giedzinski F, Baure J, Rola R, Fike JR (2006) Altered
growth and radiosensitivity in neural precursor cells subjected to
oxidative stress. Int J Radiat Biol 82:640–647.
Mazur-Kolecka B, Golabek A, Nowicki K, Flory M, Frackowiak J (2006)
Amyloid-beta impairs development of neuronal progenitor cells by
oxidative mechanisms. Neurobiol Aging 27:1181–1192.
Mi LJ, Mak KM, Lieber CS (2000) Attenuation of alcohol-induced
apoptosis of hepatocytes in rat livers by polyenylphosphatidylcho-
line (PPC). Alcohol Clin Exp Res 24:207–212.
Montoliu C, Sancho-Tello M, Azorin I, Burgal M, Valles S, Renau-
Piqueras J, Guerri C (1995) Ethanol increases cytochrome
P4502E1 and induces oxidative stress in astrocytes. J Neurochem
Paxinos G, Franklin KBJ (2003) The mouse brain in stereotaxic coor-
dinates, 2nd edition. New York: Academic Press.
Pratico D, Uryu K, Leight S, Trojanowski JQ, Lee VM (2001) Increased
lipid peroxidation precedes amyloid plaque formation in an animal
model of Alzheimer amyloidosis. J Neurosci 21:4183–4187.
Rao MS, Shetty AK (2004) Efficacy of doublecortin as a marker to
analyse the absolute number and dendritic growth of newly gen-
erated neurons in the adult dentate gyrus. Eur J Neurosci 19:
Siwak-Tapp CT, Head E, Muggenburg BA, Milgram NW, Cotman CW
(2007) Neurogenesis decreases with age in the canine hippocam-
pus and correlates with cognitive function. Neurobiol Learn Mem
Tattersfield AS, Croon RJ, Liu YW, Kells AP, Faull RLM, Connor B
(2004) Neurogenesis in the striatum of the quinolinic acid lesion
model of Huntington’s disease. Neuroscience 127:319–332.
Walker TL, Yasuda T, Adams DJ, Bartlett PF (2007) The doublecortin-
expressing population in the developing and adult brain contains
multipotential precursors in addition to neuronal-lineage cells.
J Neurosci 27:3734–3742.
Zhang C, McNeil E, Dressler L, Siman R (2006) Long-lasting impair-
ment in hippocampal neurogenesis associated with amyloid dep-
osition in a knock-in mouse model of familial Alzheimer’s disease.
Exp Neurol 204:77–87.
Zhang Q, Li X, Cui X, Zuo P (2005) D-galactose injured neurogenesis
in the hippocampus of adult mice. Neurol Res 27:552–556.
Zhao N, Zhong C, Wang Y, Zhao Y, Gong N, Zhou G, Xu T, Hong Z
(2008) Impaired hippocampal neurogenesis is involved in cognitive
dysfunction induced by thiamine deficiency at early pre-patholog-
ical lesion stage. Neurobiol Dis 29:176–185.
(Accepted 27 February 2008)
(Available online 18 March 2008)
N. Y. Calingasan et al. / Neuroscience 153 (2008) 986–996996