Activity-based anorexia is associated with reduced hippocampal cell proliferation in adolescent female rats

Abstract

Activity-based anorexia (ABA) is an animal model of anorexia nervosa that mimics core features of the clinical psychiatric disorder, including severe food restriction, weight loss, and hyperactivity. The ABA model is currently being used to study starvation-induced changes in the brain. Here, we examined hippocampal cell proliferation in animals with ABA (or the appropriate control conditions). Adolescent female Sprague-Dawley rats were assigned to 4 groups: control (24h/day food access), food-restricted (1h/day food access), exercise (24h/day food and wheel access), and ABA (1h/day food access, 24h/day wheel access). After 3 days of ABA, 5-bromo-2'-deoxyuridine (BrdU; 200mg/kg, i.p.) was injected and the rats were perfused 2h later. Brains were removed and subsequently processed for BrdU and Ki67 immunohistochemistry. The acute induction of ABA reduced cell proliferation in the dentate gyrus. This effect was significant in the hilus region of the dentate gyrus, but not in the subgranular zone, where adult neurogenesis occurs. Marked decreases in cell proliferation were also observed in the surrounding dorsal hippocampus and in the corpus callosum. These results indicate a primary effect on gliogenesis rather than neurogenesis following 3 days of ABA. For each brain region studied (except SGZ), there was a strong positive correlation between the level of cell proliferation and body weight/food intake. Future studies should examine whether these changes are maintained following long-term weight restoration and whether alterations in neurogenesis occur following longer exposures to ABA.

Full text

Behavioural
Brain
Research
236 (2013) 251–
257
Contents
lists
available
at
SciVerse
ScienceDirect
Behavioural
Brain
Research
j
ourna
l
ho
mepage:
www.elsevier.com/locate/bbr
Research
report
Activity-based
anorexia
is
associated
with
reduced
hippocampal
cell
proliferation
in
adolescent
female
rats
Nicole
C.
Barbarich-Marstellera,b,∗,
Casimir
A.
Fornalc,
Luiz
F.
Takasec,
Miriam
E.
Bocarslyc,
Candice
Arnerc,
B.
Timothy
Walsha,b,
Bartley
G.
Hoebelc,
Barry
L.
Jacobsc
aDepartment
of
Psychiatry,
College
of
Physicians
and
Surgeons
of
Columbia
University,
1051
Riverside
Drive,
Unit
98,
New
York,
NY
10032,
United
States
bNew
York
State
Psychiatric
Institute,
1051
Riverside
Drive,
Unit
98,
New
York,
NY
10032,
United
States
cPrinceton
Neuroscience
Institute,
Princeton
University,
Princeton,
NJ
08544,
United
States
h
i
g
h
l
i
g
h
t
s
Activity-based
anorexia
is
an
animal
model
of
anorexia
nervosa.
This
model
is
being
used
to
study
starvation-induced
brain
changes.
We
report
reduced
hippocampal
cell
proliferation
following
induction
of
the
model.
The
results
suggest
a
primary
effect
on
gliogenesis
rather
than
neurogenesis.
a
r
t
i
c
l
e
i
n
f
o
Article
history:
Received
10
August
2011
Received
in
revised
form
22
August
2012
Accepted
28
August
2012
Available online 4 September 2012
Keywords:
Anorexia
nervosa
Activity-based
anorexia
Food
restriction
Hyperactivity
Self-starvation
Rat
Cell
proliferation
BrdU
Ki67
a
b
s
t
r
a
c
t
Activity-based
anorexia
(ABA)
is
an
animal
model
of
anorexia
nervosa
that
mimics
core
features
of
the
clinical
psychiatric
disorder,
including
severe
food
restriction,
weight
loss,
and
hyperactivity.
The
ABA
model
is
currently
being
used
to
study
starvation-induced
changes
in
the
brain.
Here,
we
examined
hippocampal
cell
proliferation
in
animals
with
ABA
(or
the
appropriate
control
conditions).
Adolescent
female
Sprague-Dawley
rats
were
assigned
to
4
groups:
control
(24
h/day
food
access),
food-restricted
(1
h/day
food
access),
exercise
(24
h/day
food
and
wheel
access),
and
ABA
(1
h/day
food
access,
24
h/day
wheel
access).
After
3
days
of
ABA,
5-bromo-2-deoxyuridine
(BrdU;
200
mg/kg,
i.p.)
was
injected
and
the
rats
were
perfused
2
h
later.
Brains
were
removed
and
subsequently
processed
for
BrdU
and
Ki67
immunohistochemistry.
The
acute
induction
of
ABA
reduced
cell
proliferation
in
the
dentate
gyrus.
This
effect
was
significant
in
the
hilus
region
of
the
dentate
gyrus,
but
not
in
the
subgranular
zone,
where
adult
neurogenesis
occurs.
Marked
decreases
in
cell
proliferation
were
also
observed
in
the
surrounding
dorsal
hippocampus
and
in
the
corpus
callosum.
These
results
indicate
a
primary
effect
on
gliogenesis
rather
than
neurogenesis
following
3
days
of
ABA.
For
each
brain
region
studied
(except
SGZ),
there
was
a
strong
positive
correlation
between
the
level
of
cell
proliferation
and
body
weight/food
intake.
Future
studies
should
examine
whether
these
changes
are
maintained
following
long-term
weight
restoration
and
whether
alterations
in
neurogenesis
occur
following
longer
exposures
to
ABA.
© 2012 Elsevier B.V. All rights reserved.
1.
Introduction
Anorexia
nervosa
is
a
life-threatening
psychiatric
disorder
with
an
initial
onset
that
occurs
primarily
during
adolescence
in
young
women
[1].
The
disorder
is
characterized,
in
part,
by
unre-
lenting
food
restriction,
severe
weight
loss,
and
in
many
cases
∗Corresponding
author
at:
College
of
Physicians
and
Surgeons
of
Columbia
Uni-
versity,
NYSPI,
1051
Riverside
Drive,
Unit
98,
New
York,
NY
10032,
United
States.
Tel.:
+1
212
543
5197;
fax:
+1
212
543
5607.
E-mail
addresses:
nbarbarich@yahoo.com,
nb2299@columbia.edu
(N.C.
Barbarich-Marsteller).
hyperactivity
[1–3].
The
lack
of
effective
treatments
[4,5]
and
high
mortality
rate
[6,7]
provides
strong
justification
for
utilizing
ani-
mal
models
to
identify
neurobiological
mechanisms
that
may
play
a
role
in
perpetuating
self-starvation
and
hyperactivity.
Activity-based
anorexia
is
a
translational
model
of
anorexia
nervosa
that
combines
unlimited
access
to
a
running
wheel
with
limited
access
to
food
(1
h/day),
resulting
in
significant
weight
loss,
hyperactivity,
and
a
failure
to
adapt
food
intake
to
increasing
energy
demands
[8–13].
Remarkably,
animals
continue
to
run
throughout
the
period
of
food
access,
thereby
promoting
self-starvation
and,
if
not
stopped,
death.
The
severity
of
weight
loss
and
hyperactivity
in
ABA
rats
escalates
rapidly
and
can
be
used
to
study
starvation-
induced
changes
in
the
brain.
0166-4328/$
–
see
front
matter ©
2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.bbr.2012.08.047

252 N.C.
Barbarich-Marsteller
et
al.
/
Behavioural
Brain
Research
236 (2013) 251–
257
The
neurobiology
driving
the
maladaptive
cycle
of
self-
starvation
and
hyperactivity
in
ABA
rats
is
not
well
understood,
yet
our
recent
work
indicates
increased
expression
of
␣4
and
␦
subunits
of
␣4␤␦
GABAAreceptors
in
the
hippocampus
following
the
initial
onset
of
ABA
[12].
In
the
current
study,
we
extended
our
analysis
of
the
hippocampus
to
cell
proliferation
(the
initial
step
towards
neurogenesis)
in
adolescent
female
rats
following
3
days
of
ABA.
Proliferation
levels
were
compared
to
three
control
groups
(cage
control,
food-restricted
control,
and
exercise
con-
trol),
thus
allowing
each
component
of
the
model
to
be
studied
separately.
2.
Materials
and
methods
2.1.
Subjects
All
animal
procedures
followed
the
principles
of
laboratory
animal
care
(NIH
publication
No.
86-23,
revised
1985)
and
were
approved
by
the
Institutional
Ani-
mal
Care
and
Use
Committee
at
Princeton
University.
Thirty-two
adolescent
female
Sprague-Dawley
rats
were
obtained
from
Taconic
Farms,
Inc.
(Germantown,
NY)
and
individually
housed
in
standard
home
cages.
Adolescent
females
(37–39
days
old)
were
chosen
in
order
to
closely
mimic
the
stereotypical
onset
of
anorexia
nervosa
in
adolescent
girls
[1].
Animals
were
maintained
on
a
reversed
12
h
light/12
h
dark
cycle
(lights
off
at
0600
h
or
1000
h)
and
were
fed
Purina
Rodent
Chow
5001.
For
animals
with
running
wheel
access,
8
rat
activity
wheel
cages
that
consisted
of
a
home
cage
with
a
running
wheel
attached
were
used
(Med
Associates,
Inc.,
St.
Albans,
VT;
ENV-046).
Wheel
turns
were
automatically
mon-
itored
by
computer
(Med-PC
IV,
wheel
counter
program)
and
a
back-up
manual
counter.
2.2.
Study
design
At
the
start
of
the
study,
rats
were
assigned
to
four
experimental
groups
(n
=
8/group)
that
were
matched
for
baseline
body
weight:
(1)
cage
control
(24
h/day
food
access),
(2)
exercise
control
(24
h/day
food
and
wheel
access),
(3)
food-
restricted
control
(1
h/day
food
access)
and
(4)
activity-based
anorexia
(ABA;
1
h/day
food
access,
24
h/day
wheel
access).
Body
weight
and
food
intake
were
recorded
daily
∼15
min
prior
to
the
start
of
the
dark
cycle.
On
days
−1
and
0,
the
exercise
and
ABA
groups
were
given
24
h/day
access
to
both
food
and
the
running
wheel
in
order
to
record
baseline
wheel
running
activity.
On
days
1–3,
the
food-restricted
and
ABA
groups
were
given
unlimited
access
to
food
for
1
h/day
at
the
onset
of
the
dark
cycle
in
order
to
mimic
the
typical
time
ad
libitum
fed
animals
begin
food
intake.
At
the
start
of
food
access,
food
pellets
were
weighed
and
placed
on
top
of
the
cage
for
1
h/day;
pellets
were
then
removed
for
the
remaining
23
h/day
and
weighed
again
to
determine
food
intake.
The
cage
was
checked
for
any
pieces
of
food
that
may
have
been
hidden
by
the
animal.
Individual
housing
was
also
used
to
increase
the
accuracy
of
individual
food
intake
measurements.
Wheel
access
was
not
restricted
during
the
period
of
food
access.
At
the
end
of
the
light
cycle
on
day
3,
rats
received
a
single
i.p.
injection
of
200
mg/kg
of
the
thymidine
analog
5-bromo-2-deoxyuridine
(BrdU)
to
label
pro-
liferating
cells
in
the
hippocampus.
This
dose
represents
a
near-maximal
dose
for
labeling
cells
and
does
not
induce
any
discernible
signs
of
toxicity
in
the
animal
[14].
The
BrdU
(Sigma–Aldrich,
St.
Louis,
MO)
was
dissolved
in
sterile
0.9%
(w/v)
saline
(containing
0.007
N
NaOH)
and
given
in
a
volume
of
10
ml/kg
of
body
weight.
Immediately
after
the
injection,
food
was
removed
from
the
cages
(or
withheld
in
the
case
of
the
food-restricted
animals),
and
all
running
wheels
were
locked.
This
was
done
in
order
to
prevent
any
direct
effects
of
eating
and/or
exercise
on
the
sys-
temic
absorption
of
BrdU
and
its
uptake
into
the
brain.
Two
hours
after
the
BrdU
injection,
animals
were
deeply
anesthetized
with
chloral
hydrate
(∼1.75
g/kg,
i.p.)
and
were
perfused
transcardially
with
cold
physiological
saline
(containing
10
IU
heparin/ml),
followed
by
paraformaldehyde
(4%
in
0.1
M
phosphate
buffer,
pH
7.4).
Brains
were
removed,
postfixed
in
paraformaldehyde
for
24
h
at
4◦C,
transferred
to
sucrose
(30%
in
0.1
M
phosphate-buffered
saline,
PBS)
for
2–3
days
until
equilibrated
and
then
sectioned
with
a
microtome.
2.3.
BrdU
and
Ki67
immunohistochemistry
Frozen
coronal
sections
(40-␮m
thick)
were
cut
throughout
the
entire
rostral-
caudal
extent
of
the
hippocampus
(bregma,
approximately
−1.80
mm
to
−6.80
mm
[15])
and
a
1-in-12
series
of
tissue
was
then
processed
for
BrdU
or
the
intrin-
sic
mitotic
marker
Ki67
using
a
slide-mounted
immunoperoxidase
technique,
as
described
previously
[16].
The
latter
marker
was
used
to
rule
out
the
possibility
that
any
observed
changes
in
BrdU
labeling
were
due
to
alterations
in
BrdU
bioavailabil-
ity,
as
a
result
of
rapid
changes
in
body
weight
and/or
composition.
Furthermore,
unlike
BrdU
which
labels
cells
only
during
DNA
synthesis
(S-phase),
Ki67
labels
pro-
liferating
cells
during
all
phases
of
mitosis
[17],
and
therefore
can
be
used
to
confirm
the
general
findings
obtained
with
BrdU.
For
BrdU
staining,
sections
were
heated
in
citric
acid
(0.01
M,
pH
6.0),
digested
with
trypsin
(0.1%
in
0.1
M
Tris
buffer,
pH
7.5,
containing
0.1%
CaCl2),
denatured
with
2.4
M
hydrochloric
acid,
and
then
incubated
with
a
mouse
monoclonal
antibody
raised
against
BrdU
(1:200
in
PBS
containing
0.5%
Tween
20;
NCL-BrdU;
Novocas-
tra
Laboratories
Ltd,
Newcastle
upon
Tyne,
UK)
for
48
h
at
4◦C.
For
Ki67
staining,
sections
were
heated
in
citric
acid
(0.01
M,
pH
6.0),
and
then
incubated
with
a
mouse
monoclonal
Ki67
antibody
(1:200
in
PBS
containing
0.5%
Tween
20;
NCL-
Ki-67-MM1;
Novocastra
Laboratories
Ltd.)
for
48
h
at
4◦C.
Following
the
primary
antibody
incubation,
sections
were
incubated
with
a
biotinylated
horse
anti-mouse
IgG
(1:200
in
PBS;
Vector
Laboratories,
Burlingame,
CA)
and
with
avidin-biotin
complex
(1:100
in
PBS;
Vectastain®Elite
ABC
kit,
Vector
Laboratories)
for
60
min,
and
then
reacted
with
3,3-diaminobenzidine
(DAB),
to
visualize
labeled
cells.
Sec-
tions
were
then
counterstained
with
cresyl
violet,
dehydrated
and
coverslipped
with
DPX.
2.4.
Tissue
analysis
All
slides
were
analyzed
blind
with
respect
to
treatment
using
an
Olympus
BX-60
light
microscope
(Olympus
America
Inc.,
Melville,
NY,
USA).
Immunopositive
cells
were
identified
by
their
distinct
dark-brown
nuclear
staining.
In
every
12th
sec-
tion,
BrdU/Ki67
labeled
cells
were
counted
bilaterally
in
the
dentate
gyrus
(DG),
the
dorsal
hippocampus
(minus
the
DG),
and
in
the
medial
corpus
callosum
at
high
mag-
nification
(400×
or
600×).
Cell
counts
per
region
were
summed
across
all
sections
for
each
animal
and
then
multiplied
by
12
(the
inverse
of
the
sampling
interval)
to
obtain
an
estimate
of
the
total
number
of
labeled
cells
in
each
region.
In
addi-
tion,
the
DG
was
divided
into
anterior
(dorsal)
and
posterior
(ventral)
portions,
as
in
Guzmán-Marin
et
al.
[18].
The
boundary
separating
the
anterior
and
posterior
hip-
pocampus
corresponded
to
the
region
where
the
CA2
and
CA3
pyramidal
cell
layers
coalesce
into
a
continuous
cell
layer
in
the
coronal
plane
(approximately
−4.5
mm
from
bregma,
according
to
the
atlas
of
Paxinos
and
Watson
[15]).
Typically,
there
were
6
anterior
and
4
posterior
sections
for
each
animal.
Labeled
cells
in
the
den-
tate
gyrus
were
also
counted
separately
in
the
subgranular
zone
(SGZ)
and
in
the
hilus.
Cells
located
within
two
cell-body
widths
of
the
granular
cell
layer
were
con-
sidered
to
be
in
the
SGZ;
cells
located
more
distally
were
considered
to
be
in
the
hilus.
2.5.
Statistical
analysis
The
behavioral
data
were
analyzed
using
a
two-way
repeated
measures
anal-
ysis
of
variance
(ANOVA)
with
time
as
the
within
subjects
factor
and
treatment
as
the
between
subjects
factor.
For
statistical
analysis
of
the
cell
proliferation
data
(BrdU
and
Ki67)
in
each
brain
region,
a
one-way
ANOVA
was
used.
When
a
signifi-
cant
F
value
was
calculated,
post
hoc
comparisons
between
groups
were
made
using
Bonferroni
multiple
comparison
test.
Correlations
between
BrdU
and
Ki67,
or
BrdU
and
the
various
behavioral/physiological
measures
were
analyzed
using
Pearson
correlation
coefficient.
The
statistical
analyses
were
carried
out
using
Prism
ver-
sion
5.0c
for
Mac
OS
(GraphPad
Software,
San
Diego,
CA).
All
data
are
expressed
as
means
±
standard
error
of
the
mean
(SEM).
In
all
cases,
a
probability
value
p
<
0.05
was
taken
as
statistically
significant.
3.
Results
3.1.
Behavior
Fig.
1
displays
changes
in
wheel
running
activity,
food
intake,
and
body
weight
over
the
course
of
the
study
for
the
different
treatment
groups.
For
wheel
running
activity
(top
panel),
ANOVA
revealed
significant
main
effects
of
both
treatment
(F(1,42)
=
11.82,
p
<
0.005)
and
time
(F(3,42)
=
16.98,
p
<
0.0001).
The
ABA
group
ran
significantly
more
than
the
exercise
group
on
days
2
(+129.6%,
p
<
0.05)
and
3
(+227.7%,
p
<
0.0001).
For
food
intake
(middle
panel),
an
ANOVA
beginning
on
day
0
indicated
significant
main
effects
of
both
treatment
(F(3,84)
=
243.5,
p
<
0.0001)
and
time
(F(3,84)
=
34.80,
p
<
0.0001).
Compared
to
the
control
group,
the
food-restricted
group
consumed
significantly
less
food
on
days
1
(−93.1%,
p
<
0.0001),
2
(−75.7%,
p
<
0.0001),
and
3
(−67.1%,
p
<
0.0001),
and
the
ABA
group
also
consumed
significantly
less
food
on
days
1
(−91.7%,
p
<
0.0001),
2
(−76.4%,
p
<
0.0001),
and
3
(−75.2%,
p
<
0.0001).
Compared
to
the
exercise
group,
the
food-restricted
and
ABA
groups
consumed
signifi-
cantly
less
food
on
days
1
(−91.8%
and
−90.2%,
p
<
0.0001),
2
(−72.4%
and
−73.2%,
p
<
0.0001),
and
3
(−62.9%
and
−72.1%,
p
<
0.0001).

N.C.
Barbarich-Marsteller
et
al.
/
Behavioural
Brain
Research
236 (2013) 251–
257 253
Fig.
1.
Behavioral/physiological
measures
in
adolescent
female
rats
under
4
condi-
tions:
control
(24
h/day
food
access),
food-restricted
(1
h/day
food
access),
exercise
(24
h/day
food
and
wheel
access),
and
activity-based
anorexia
(ABA;
1
h/day
food
access,
24
h/day
wheel
access).
Top
panel:
wheel
running
activity;
middle
panel:
food
intake;
bottom
panel:
body
weight.
Time
0
=
onset
of
restricted
food
access.
Values
are
means
±
SEM;
n
=
8/group.
Baseline
body
weights
(g)
at
the
start
of
the
experiment
were:
control,
152.1
±
2.2;
exercise,
153.2
±
5.7;
food
restriction,
153.4
±
5.2;
and
ABA,
151.5
±
4.2.
*Significantly
different
from
control; †significantly
different
from
exercise; §significantly
different
from
food
restriction;
by
two-way
repeated
measures
ANOVA
and
Bonferroni
posttest.
See
text
for
additional
details
regarding
statistical
analysis
and
levels
of
significance.
ABA
induced
significant
weight
loss,
hyperactivity,
and
a
failure
to
adapt
food
intake
to
increasing
energy
demands.
For
body
weight
(bottom
panel),
an
ANOVA
beginning
on
day
0
revealed
significant
main
effects
of
both
treatment
(F(3,84)
=
6.769,
p
<
0.005)
and
time
(F(3,84)
=
41.95,
p
<
0.0001).
Compared
to
the
control
group,
the
food-restricted
group
weighed
significantly
less
on
day
3
(−15.9%,
p
<
0.01),
whereas
the
ABA
group
weighed
significantly
less
on
days
2
(−20.2%,
p
<
0.0001)
and
3
(−30.1%,
p
<
0.0001).
Compared
to
the
exercise
group,
the
ABA
group
weighed
significantly
less
on
days
2
(−16.4%,
p
<
0.01)
and
3
(−26.4%,
p
<
0.0001).
The
ABA
group
also
weighed
significantly
less
than
the
food-restricted
group
on
day
3
(−16.9%,
p
<
0.01),
with
a
total
mean
weight
loss
of
−21.1
±
1.5%
compared
to
−10.4
±
1.2%
of
body
weight
at
the
start
of
restricted
food
access.
By
contrast,
the
control
and
exercise
groups
gained
7.4
±
1.0%
and
5.3
±
1.0%,
respectively,
of
body
weight
over
the
same
time
interval.
3.2.
Cell
proliferation
Cell
proliferation
levels
in
the
hippocampal
DG
for
the
dif-
ferent
treatment
groups
are
shown
in
Figs.
2
and
3.
For
BrdU
(top
panel),
an
ANOVA
revealed
significant
differences
between
groups
in
the
DG
as
a
whole
(F(3,28)
=
11.30,
p
<
0.0001)
and
in
the
hilus
(F(3,28)
=
23.14,
p
<
0.0001).
In
the
DG,
there
was
a
significant
decrease
in
the
number
of
BrdU+
cells
in
the
food-restricted
group
compared
to
the
exercise
group
(−29.3%,
p
<
0.01),
and
in
the
ABA
group
compared
to
both
the
control
group
(−35.2%,
p
<
0.01)
and
the
exercise
group
(−40.1%,
p
<
0.001).
Although
the
SGZ
showed
no
sig-
nificant
between
group
differences
(F(3,28)
=
2.711,
p
>
0.05),
five
group
differences
were
found
in
the
hilus:
there
was
a
significant
decrease
in
BrdU+
cells
in
the
food-restricted
group
compared
to
the
control
(−42.1%,
p
<
0.01)
and
exercise
(−41.7%,
p
<
0.01)
groups,
and
in
the
ABA
group
compared
to
the
control
(−74.9%,
p
<
0.001),
exercise
(−74.7%,
p
<
0.001),
and
food-restricted
(−56.5%,
p
<
0.05)
groups.
To
validate
the
BrdU
results,
we
used
the
intrinsic
proliferation
marker
Ki67.
As
expected,
there
was
a
highly
significant
positive
correlation
between
the
two
markers
(r
(27)
=
0.696;
p
<
0.0001),
with
Ki67
labeling
approximately
twice
as
many
proliferating
cells
in
the
DG
as
BrdU,
presumably
because
it
captures
a
greater
portion
of
the
active
cell
cycle
[17].
As
shown
in
Fig.
2
and
in
the
bottom
panel
of
Fig.
3,
similar
results
were
obtained
with
Ki67,
with
the
largest
reduction
in
labeling
occurring
in
the
ABA
group.
Significant
differences
between
groups
were
found
in
the
DG
as
a
whole
(F(3,28)
=
3.487,
p
<
0.05)
and
in
the
hilus
subregion
(F(3,28)
=
12.10,
p
<
0.0001).
In
the
DG,
the
num-
ber
of
Ki67+
cells
was
significantly
reduced
in
the
ABA
group
compared
to
the
control
group
(−26.1%,
p
<
0.05).
No
significant
differences
between
groups
were
noted
in
the
SGZ
(F(3,28)
=
0.292,
p
>
0.05).
In
the
hilus,
however,
the
number
of
Ki67+
cells
was
significantly
lower
in
the
food-restricted
(−49.9%,
p
<
0.001)
and
ABA
(−62.5%,
p
<
0.001)
groups
compared
to
the
control
group,
and
in
the
ABA
group
compared
to
the
exercise
group
(−50.9%,
p
<
0.05).
Interestingly,
for
both
BrdU
and
Ki67,
the
effects
of
ABA
on
cell
proliferation
were
much
more
pro-
nounced
in
the
ventral
vs.
dorsal
portions
of
the
DG
(data
not
shown).
Since
it
is
generally
believed
that
proliferating
cells
in
the
hilus
and
in
the
SGZ
give
rise
primarily
to
glia
and
neurons,
respectively
[19],
the
present
findings
suggest
a
possible
influence
of
ABA
on
gliogenesis
rather
than
neurogenesis.
To
determine
whether
this
effect
is
specific
to
the
hilus,
in
the
same
experimental
subjects,
we
examined
cell
proliferation
in
two
non-neurogenic
brain
regions:
the
corpus
callosum
(a
fiber
tract
region)
and
the
surrounding
dor-
sal
hippocampus
(another
neuronal
region).
This
analysis
revealed
a
marked
suppression
of
cell
proliferation
in
both
regions,
com-
parable
in
magnitude
to
that
observed
in
the
hilus
(Figs.
4
and
5).
Significant
differences
between
groups
were
observed
in
the
corpus
callosum
(F(3,28)
=
24.14,
p
<
0.0001)
and
in
the
dorsal
hippocam-
pus
(F(3,28)
=
35.40,
p
<
0.0001).
In
the
corpus
callosum,
there
was
a
significant
decrease
in
the
number
of
BrdU+
cells
in
the
ABA
group
compared
to
the
control
(−77.9%,
p
<
0.001),
exercise
(−77.1%,
p
<
0.001),
and
food-restricted
groups
(−72.5%,
p
<
0.001).
In
the
dorsal
hippocampus,
the
number
of
BrdU+
cells
was
significantly
reduced
in
both
the
food-restricted
(−45.7%,
p
<
0.001)
and
ABA
(−78.0%,
p
<
0.001)
groups
compared
to
the
control
group,
in
both
the
food-restricted
(−43.1%,
p
<
0.01)
and
ABA
(−77.0%,
p
<
0.001)
groups
compared
to
the
exercise
group,
and
in
the
ABA
group
(−59.5%,
p
<
0.01)
compared
to
the
food-restricted
group.
Similar

254 N.C.
Barbarich-Marsteller
et
al.
/
Behavioural
Brain
Research
236 (2013) 251–
257
Fig.
2.
Photomicrographs
(400×
magnification)
showing
BrdU
and
Ki67
labeling
in
the
dentate
gyrus
of
cage
control
and
activity-based
anorexia
(ABA)
adolescent
female
rats.
Newly
generated
BrdU-positive
and
Ki67-positive
cells
can
be
seen
(often
in
clusters)
at
the
border
of
the
granule
cell
layer
(GCL)
and
the
hilus,
in
the
subgranular
zone
(SGZ).
The
acute
induction
of
ABA
significantly
reduced
cell
proliferation
relative
to
the
experimental
controls
in
the
hilus,
but
not
in
the
SGZ.
Neurogenesis
in
the
rat
hippocampus
is
restricted
to
the
SGZ
of
the
dentate
gyrus.
Proliferating
cells
in
other
regions
(including
the
hilus)
are
non-neuronal
cells.
Scale
bar
=
50
␮m.
Fig.
3.
Cell
proliferation
in
the
hippocampal
dentate
gyrus
of
adolescent
female
rats
under
4
conditions:
control
(24
h/day
food
access),
food-restricted
(1
h/day
food
access),
exercise
(24
h/day
food
and
wheel
access),
and
activity-based
anorexia
(ABA;
1
h/day
food
access,
24
h/day
wheel
access).
Top
panel:
BrdU
incorporation;
bottom
panel:
Ki67
expression.
Values
are
means
±
SEM.
For
BrdU,
n
=
6
for
control;
n
=
7
for
exercise;
n
=
8
for
both
food
restriction
and
ABA.
For
Ki67,
n
=
8/group.
*Sig-
nificantly
different
from
control; †significantly
different
from
exercise; §significantly
different
from
food
restriction;
by
one-way
ANOVA
and
Bonferroni
posttest.
See
text
for
additional
details
regarding
statistical
analysis
and
levels
of
significance.
ABA
markedly
reduced
cell
proliferation
in
the
dentate
gyrus
(DG),
specifically
in
the
hilus,
but
not
in
the
subgranular
zone
(SGZ).
findings
were
also
obtained
in
both
regions
using
Ki67
(data
not
shown).
3.3.
Relationship
between
cell
proliferation
and
behavior
The
relationship
between
BrdU
cell
proliferation
and
behav-
ioral/physiological
measures
in
the
ABA
paradigm
was
examined
using
Pearson
correlation
coefficient.
BrdU
cell
counts
for
all
treatment
groups
combined
were
positively
correlated
with
body
weight
(measured
on
the
last
day
of
the
study
and
expressed
as
a
percentage
of
baseline
weight)
in
all
the
brain
regions
studied
(DG:
r27 =
0.669,
p
<
0.0001;
hilus:
r27 =
0.862,
p
<
0.0001;
corpus
callosum:
r27 =
0.844,
p
<
0.0001;
dorsal
hippocampus:
r27 =
0.906,
p
<
0.0001),
except
the
SGZ
(r27 =
0.314,
p
=
0.097,
NS).
This
is
shown
for
the
hilus
(Fig.
6).
Significant
positive
correlations
were
also
observed
between
BrdU
cell
proliferation
and
food
intake
(measured
on
the
last
day)
in
all
the
brain
regions
exam-
ined
(DG:
r27 =
0.761,
p
<
0.0001;
SGZ:
r27 =
0.499,
p
<
0.01;
hilus:
r27 =
0.807,
p
<
0.0001;
corpus
callosum:
r27 =
0.709,
p
<
0.0001;
dor-
sal
hippocampus:
r27 =
0.881,
p
<
0.0001).
Furthermore,
levels
of
cell
proliferation
were
negatively
correlated
with
wheel
running
activity
(last
day,
all
groups
combined)
in
the
hiIus
(r27 =
−0.485,
p
<
0.01),
corpus
callosum
(r27 =
−0.721,
p
<
0.0001),
and
dorsal
hip-
pocampus
(r27 =
−0.544,
p
<
0.005),
but
not
in
the
SGZ
(r27 =
−0.031,
p
=
0.871,
NS)
or
DG
as
a
whole
(r27 =
−0.281,
p
=
0.140,
NS).
Virtu-
ally
all
of
the
above
correlations
remained
significant,
even
after
controlling
for
the
influence
of
food
restriction
or
exercise
(data
not
shown).
Interestingly,
wheel
running
activity
(averaged
over
the
last
two
days
of
the
study)
was
positively
correlated
with
cell
proliferation
in
the
exercise/control
(no
exercise)
groups
combined
in
the
SGZ
(r11 =
0.593,
p
<
0.05),
but
was
negatively
correlated
with
cell
proliferation
in
the
ABA/food-restricted
(no
exercise)
groups
combined
in
the
hilus
(r13 =
−0.685,
p
<
0.005),
corpus
callosum
(r13 =
−0.797,
p
<
0.0005)
and
dorsal
hippocampus
(r13 =
−0.672,
p<
0.01).
These
latter
findings
suggest
a
facilitatory
effect
of
exercise

N.C.
Barbarich-Marsteller
et
al.
/
Behavioural
Brain
Research
236 (2013) 251–
257 255
Fig.
4.
Photomicrographs
(400×
magnification)
showing
BrdU
and
Ki67
labeling
in
the
corpus
callosum
of
cage
control
and
activity-based
anorexia
(ABA)
adolescent
female
rats.
Newly
generated
BrdU-positive
and
Ki67-positive
cells
can
be
seen
scattered
throughout
the
fiber
tracts
in
the
corpus
callosum.
The
acute
induction
of
ABA
significantly
reduced
cell
proliferation
relative
to
the
experimental
controls.
Scale
bar
=
50
␮m.
in
control
animals
and
a
suppressive
effect
of
exercise
in
food-
restricted
animals.
4.
Discussion
This
study
examined
hippocampal
cell
proliferation
in
an
animal
model
of
anorexia
nervosa
known
as
activity-based
anorexia.
As
expected,
animals
subjected
to
the
ABA
paradigm
developed
rapid
and
severe
weight
loss
and
hyperactivity;
moreover,
ABA
animals
did
not
adequately
increase
food
intake
during
the
1
h/day
period
of
food
access
to
compensate
for
the
increase
in
energy
expendi-
ture.
This
acute
induction
of
ABA
in
adolescent
female
rats
over
3
days
was
associated
with
a
profound
reduction
in
cell
prolifera-
tion
in
the
DG
(specifically
in
the
hilus),
as
well
as
the
surrounding
dorsal
hippocampus
and
the
corpus
callosum.
These
findings
were
obtained
using
two
distinct
proliferation
markers,
adding
to
their
validity.
Fig.
5.
BrdU
cell
proliferation
in
two
non-neurogenic
brain
regions
(the
medial
corpus
callosum
and
the
dorsal
hippocampus,
minus
dentate
gyrus)
of
adoles-
cent
female
rats
under
4
conditions:
control
(24
h/day
food
access),
food-restricted
(1
h/day
food
access),
exercise
(24
h/day
food
and
wheel
access),
and
activity-
based
anorexia
(ABA;
1
h/day
food
access,
24
h/day
wheel
access).
Values
are
means
±
SEM;
n
=
6
for
control;
n
=
7
for
exercise;
n
=
8
for
both
food
restriction
and
ABA. *Significantly
different
from
control; †significantly
different
from
exercise;
§significantly
different
from
food
restriction;
by
one-way
ANOVA
and
Bonferroni
posttest.
See
text
for
additional
details
regarding
statistical
analysis
and
levels
of
significance.
ABA
markedly
reduced
cell
proliferation
in
both
brain
regions.
Whether
certain
particular
features
of
the
ABA
model
contribute
more
substantially
than
others
to
this
suppressive
effect
on
cell
pro-
liferation
are
unknown,
but
merits
consideration.
Previous
studies
have
examined
the
influence
of
various
factors
relevant
to
ABA
on
neurogenesis,
including
food
restriction,
exercise,
and
sleep
depri-
vation.
While
an
in
depth
review
of
the
effects
of
experimental
manipulations
in
these
areas
is
beyond
the
scope
of
this
paper,
reviews
of
these
topics
have
been
cited
below.
Moreover,
it
should
be
considered
that
most
studies
have
utilized
adult
rats
or
mice
and
often
males,
whereas
the
current
study
utilized
adolescent
female
rats
to
better
mimic
the
clinical
population
of
anorexia
nervosa.
The
species,
strain,
age,
sex,
and
severity
of
the
experimental
manipula-
tions
may
also
influence
the
differences
in
findings
between
studies
that
are
discussed
below.
Mild
food
restriction
has
been
shown
to
either
increase
or
have
no
effect
on
hippocampal
neurogenesis
(for
review
see
Ref.
Fig.
6.
Relationship
between
BrdU
cell
proliferation
in
the
hilus
and
body
weight
of
adolescent
female
rats
under
4
conditions:
control
(24
h/day
food
access),
food-
restricted
(1
h/day
food
access),
exercise
(24
h/day
food
and
wheel
access),
and
activity-based
anorexia
(ABA;
1
h/day
food
access,
24
h/day
wheel
access).
Body
weight
at
the
end
of
the
study
was
expressed
as
a
percentage
of
initial
body
weight.
Pearson
correlation
analysis
revealed
a
significant
positive
correlation
(r27 =
0.8615;
p
<
0.0001)
between
the
number
of
BrdU+
cells
in
the
hilus
and
body
weight.
This
correlation
remained
significant
even
after
controlling
for
the
influence
of
food
restriction
(r14 =
0.7775;
p
<
0.0005)
and
exercise
(r13 =
0.8897;
p
<
0.0001)
in
the
ABA
group.

256 N.C.
Barbarich-Marsteller
et
al.
/
Behavioural
Brain
Research
236 (2013) 251–
257
[20]).
Food
restriction-induced
increases
in
neurogenesis
have
been
attributed
to
a
decrease
in
the
death
of
newly
produced
cells,
rather
than
from
an
increase
in
cell
proliferation
[21].
Although
these
findings
are
in
contrast
to
the
current
study
which
found
that
food
restriction
decreased
cell
proliferation,
increased
neuro-
genesis
in
other
studies
occurred
following
mild
dietary
restriction
(usually
resulting
in
∼10%
weight
loss)
over
a
long
period
of
time
(typically
several
months),
which
is
quite
different
from
the
severe
level
of
weight
loss
(−21%)
seen
in
ABA
rats
over
a
3-day
period.
Thus,
while
mild
dietary
restriction
may
enhance
cell
proliferation,
the
more
severe
levels
seen
in
ABA
appear
to
be
associated
with
suppressive
effects.
Voluntary
access
to
exercise
has
also
been
shown
to
increase
hippocampal
neurogenesis
(for
review
of
studies,
see
Ref.
[22]).
Although
we
did
not
see
an
increase
in
proliferation
in
our
exer-
cise
control
group
(ad
lib.
food
and
running
wheel
access),
3
days
may
not
be
sufficient
to
see
an
enhancement
of
cell
proliferation
in
adolescent
female
rats.
There
was,
however,
a
significant
positive
correlation
between
wheel
running
activity
and
cell
proliferation
observed
in
the
exercise/control
groups.
Moreover,
our
prelimi-
nary
data
do
show
an
increase
in
the
number
of
proliferating
Ki67+
cells
in
adolescent
female
rats
with
ad
lib.
food
and
running
wheel
access
after
2
weeks
(interestingly,
no
differences
were
seen
in
the
number
of
surviving
BrdU+
cells;
unpublished
data).
The
exercise-
induced
reduction
in
cell
proliferation
seen
in
ABA
rats
is
consistent
with
the
findings
from
a
study
that
reported
reduced
neurogenesis
in
spontaneously
hypertensive
rats
that
maintained
high
levels
of
running
(20
km/day;
[23]).
When
rats
in
this
study
were
restricted
to
6
km
of
running
per
day,
exercise
did
not
inhibit
proliferation
of
hippocampal
progenitor
cells.
Together,
these
studies
suggest
that
when
exercise
becomes
excessive,
there
may
be
a
shift
from
enhancement
to
suppression
of
cell
proliferation.
Sleep
deprivation
has
also
been
shown
to
dramatically
suppress
hippocampal
proliferation
and
neurogenesis.
Short-term
sleep
deprivation
decreased
neurogenesis
associated
with
hippocampal-
dependent
learning,
whereas
more
severe
sleep
deprivation
also
reduced
basal
levels
of
cell
proliferation
(for
review,
see
Ref.
[24]).
Cell
proliferation
in
the
hilus
appears
to
be
more
sensitive
to
the
disruptive
effect
of
sleep
deprivation
than
cell
proliferation
in
the
SGZ
[25].
The
question
of
whether
sleep
deprivation
plays
a
role
in
the
decreased
cell
proliferation
associated
with
ABA
is
intrigu-
ing.
There
is
clear
evidence
that
wheel
running
activity
increases
during
the
hours
preceding
food
access
in
adult
female
ABA
rats
(termed
food
anticipatory
activity
[26])
and
in
most
studies
this
time
frame
occurs
during
the
light
cycle
when
control
rats
are
typically
asleep.
Burden
et
al.
[27]
reported
that
52%
of
wheel
run-
ning
activity
in
female
ABA
rats
occurred
during
the
light
cycle,
however
the
amount
of
wheel
running
related
to
food
anticipa-
tory
activity
was
not
specified.
Interestingly,
recent
work
in
our
lab
suggests
that
hyperactivity
in
adolescent
female
ABA
rats
occurs
throughout
the
light
cycle
and
beyond
the
time
frame
considered
to
be
related
to
food
anticipatory
activity
(Barbarich-Marsteller,
unpublished
data).
When
wheel
running
activity
was
binned
into
10
min
intervals,
animals
with
the
most
severe
hyperactivity
dur-
ing
a
24
h
period
ran
almost
continuously
throughout
the
light
and
dark
cycles,
with
very
few
10
min
periods
in
which
the
animals
were
not
running
(Barbarich-Marsteller,
unpublished
data).
The
degree
to
which
this
continuous
running
may
contribute
to
sleep
deprivation
or
an
alteration
in
circadian
rhythm
and
the
poten-
tial
consequences
on
neurobiology
in
this
model
clearly
warrants
further
investigation.
Finally,
increases
in
plasma
glucocorticoids,
such
as
those
asso-
ciated
with
stress,
are
known
to
have
a
powerful
suppressive
effect
on
cell
proliferation
(for
review,
see
Ref.
[28]).
Individu-
als
with
anorexia
nervosa
have
increased
levels
of
cortisol
that
are
associated
with
physical
activity
and
commitment
to
exercise
[29].
Moreover,
ABA
rats
also
exhibit
elevated
corticosterone
lev-
els
(beyond
that
seen
with
food-restriction
alone)
[27],
suggesting
that
decreased
cell
proliferation
may
be
related
to
stress-induced
activation
of
the
HPA
axis.
The
finding
that
ABA
(and
to
a
lesser
extent
food
restriction)
is
associated
with
a
profound
suppression
of
cell
proliferation
in
the
hilus
(and
two
non-neurogenic
brain
regions),
as
compared
to
the
SGZ,
suggests
a
more
powerful
effect
on
gliogenesis
rather
than
neurogenesis
(predominately
in
the
SGZ).
The
functional
impli-
cations
of
this
suppression
remain
to
be
determined.
Whereas
neurons
are
responsible
for
information
processing
and
trans-
mission,
glia
cells
play
a
supportive
role
and
are
crucial
for
the
proper
functioning
of
the
central
nervous
system.
Besides
providing
structural
support
and
electrical
insulation,
these
specialized
cells
supply
metabolic
nutrients
to
neurons,
maintain
homeostasis,
and
participate
in
synaptic
modulation,
among
other
functions
[30].
Of
clinical
relevance,
decreases
in
glia
cell
number
have
been
linked
to
depression
[31],
and
individuals
suffering
from
anorexia
ner-
vosa
often
have
comorbid
major
depression
or
exhibit
symptoms
of
depression
[32],
which
may
be
related
to
a
loss
of
glia
in
the
brain.
Thus,
disruption
of
gliogenesis
could
potentially
have
significant
and
far-reaching
consequences
on
brain
function,
and
may
possibly
play
a
role
in
the
neuropsychopathology
of
anorexia
and
starvation.
It
should
be
noted
that
there
are
several
limitations
to
studying
neurogenesis
in
the
ABA
model
in
general
and
the
current
study
in
particular.
The
severity
of
the
model
and
the
rapidity
with
which
the
maladaptive
cycle
of
weight
loss
and
hyperactivity
develops
in
adolescent
female
rats
severely
limits
the
length
of
time
the
animals
can
be
studied
(usually
less
than
1
week),
and
consequently
what
can
be
measured
in
terms
of
cell
maturation
and
survival.
More-
over,
the
food-restricted
and
exercise
control
groups
display
much
more
modest
levels
of
weight
loss
and
activity
compared
to
the
ABA
group;
thus,
there
is
no
way
to
mimic
the
same
level
of
severity
in
dietary
restriction
or
exercise
as
seen
in
ABA.
Finally,
as
this
was
a
preliminary
study,
it
is
not
yet
known
whether
these
changes
per-
sist
following
weight
restoration
and
recovery
from
the
model
or
whether
alterations
in
neurogenesis
occur
following
longer
expo-
sures
to
ABA.
Despite
these
limitations,
the
ABA
model
provides
the
most
valid
animal
model
of
anorexia
nervosa
to
date
and
has
the
potential
to
provide
critical
insights
into
starvation-induced
changes
in
neurobiological
processes
and
functions.
This
is
critical
given
that
anorexia
nervosa
has
one
of
the
highest
mortality
rates
among
psychiatric
disorders
with
limited
effective
treatments.
Acknowledgement
Funding
for
this
study
was
provided
by
a
grant
to
Dr.
Nicole
Barbarich-Marsteller
from
the
National
Eating
Disorders
Associa-
tion.
References
[1] American
Psychiatric
Association.
Diagnostic
and
statistical
manual
of
mental
disorders.
4th
ed.
Washington,
DC:
American
Psychiatric
Association;
2000,
text
revision.
[2]
Kron
L,
Katz
JL,
Gorzynski
G,
Weiner
H.
Hyperactivity
in
anorexia
nervosa:
a
fundamental
clinical
feature.
Comprehensive
Psychiatry
1978;19:433–40.
[3]
Casper
RC.
The
‘drive
for
activity’
and
restlessness
in
anorexia
nervosa:
poten-
tial
pathways.
Journal
of
Affective
Disorders
2006;92:99–107.
[4]
Aigner
M,
Treasure
J,
Kaye
W,
Kasper
S.
World
Federation
of
Societies
of
Biologi-
cal
Psychiatry
(WFSBP)
Guidelines
for
the
pharmacological
treatment
of
eating
disorders.
World
Journal
of
Biological
Psychiatry
2011;12:400–43.
[5]
Barbarich-Marsteller
NC,
Laurino
K,
Colacino
AR.
Pharmacological
treatments
for
anorexia
nervosa.
In:
Barbarich-Marsteller
NC,
editor.
Anorexia
nervosa:
symptoms,
treatment,
and
neurobiology.
New
York:
Nova
Science
Publishers,
Inc.;
2012.
[6]
Sullivan
PF.
Mortality
in
anorexia
nervosa.
American
Journal
of
Psychiatry
1995;152:1073–4.
[7] Birmingham
CL,
Su
J,
Hlynsky
JA,
Goldner
EM,
Gao
M.
The
mortality
rate
from
anorexia
nervosa.
International
Journal
of
Eating
Disorders
2005;38:143–6.

N.C.
Barbarich-Marsteller
et
al.
/
Behavioural
Brain
Research
236 (2013) 251–
257 257
[8]
Hall
JF,
Hanford
PV.
Activity
as
a
function
of
a
restricted
feeding
schedule.
Journal
of
Comparative
and
Physiological
Psychology
1954;47:362–3.
[9]
Routtenberg
A,
Kuznesof
AW.
Self-starvation
of
rats
living
in
activity
wheels
on
a
restricted
feeding
schedule.
Journal
of
Comparative
and
Physiological
Psychology
1967;64:414–21.
[10]
Dwyer
DM,
Boakes
RA.
Activity-based
anorexia
in
rats
as
failure
to
adapt
to
a
feeding
schedule.
Behavioral
Neuroscience
1997;111:195–205.
[11] Dixon
DP,
Ackert
AM,
Eckel
LA.
Development
of,
and
recovery
from,
activity-
based
anorexia
in
female
rats.
Physiology
and
Behavior
2003;80:273–9.
[12] Aoki
C,
Sabaliauskas
N,
Chowdhury
T,
Min
JY,
Colacino
AR,
Laurino
K,
et
al.
Ado-
lescent
female
rats
exhibiting
activity-based
anorexia
express
elevated
levels
of
GABA(A)
receptor
alpha4
and
delta
subunits
at
the
plasma
membrane
of
hippocampal
CA1
spines.
Synapse
2012;66:391–407.
[13] Barbarich-Marsteller
NC.
Activity-based
anorexia
in
the
rat.
In:
Avena
NM,
editor.
Animal
models
of
eating
disorders.
New
York:
Humana
Press;
2012.
[14] Cameron
HA,
McKay
RD.
Adult
neurogenesis
produces
a
large
pool
of
new
granule
cells
in
the
dentate
gyrus.
Journal
of
Comparative
Neurology
2001;435:406–17.
[15]
Paxinos
G,
Watson
C.
The
rat
brain
in
stereotaxic
coordinates.
New
York:
Aca-
demic
Press;
1986.
[16] Fornal
CA,
Stevens
J,
Barson
JR,
Blakley
GG,
Patterson-Buckendahl
P,
Jacobs
BL.
Delayed
suppression
of
hippocampal
cell
proliferation
in
rats
following
inescapable
shocks.
Brain
Research
2007;1130:48–53.
[17]
Kee
N,
Sivalingam
S,
Boonstra
R,
Wojtowicz
JM.
The
utility
of
Ki-67
and
BrdU
as
proliferative
markers
of
adult
neurogenesis.
Journal
of
Neuroscience
Methods
2002;115:97–105.
[18]
Guzman-Marin
R,
Suntsova
N,
Stewart
DR,
Gong
H,
Szymusiak
R,
McGinty
D.
Sleep
deprivation
reduces
proliferation
of
cells
in
the
dentate
gyrus
of
the
hippocampus
in
rats.
The
Journal
of
Physiology
2003;549:563–71.
[19]
Bondolfi
L,
Ermini
F,
Long
JM,
Ingram
DK,
Jucker
M.
Impact
of
age
and
caloric
restriction
on
neurogenesis
in
the
dentate
gyrus
of
C57BL/6
mice.
Neurobiology
of
Aging
2004;25:333–40.
[20]
Park
HR,
Lee
J.
Neurogenic
contributions
made
by
dietary
regulation
to
hippocampal
neurogenesis.
Annals
of
the
New
York
Academy
of
Sciences
2011;1229:23–8.
[21]
Lee
J,
Duan
W,
Long
JM,
Ingram
DK,
Mattson
MP.
Dietary
restriction
increases
the
number
of
newly
generated
neural
cells,
and
induces
BDNF
expression,
in
the
dentate
gyrus
of
rats.
Journal
of
Molecular
Neuroscience
2000;15:
99–108.
[22]
van
Praag
H.
Neurogenesis
and
exercise:
past
and
future
directions.
Neuro-
molecular
Medicine
2008;10:128–40.
[23]
Naylor
AS,
Persson
AI,
Eriksson
PS,
Jonsdottir
IH,
Thorlin
T.
Extended
voluntary
running
inhibits
exercise-induced
adult
hippocampal
progenitor
prolifer-
ation
in
the
spontaneously
hypertensive
rat.
Journal
of
Neurophysiology
2005;93:2406–14.
[24]
Meerlo
P,
Mistlberger
RE,
Jacobs
BL,
Heller
HC,
McGinty
D.
New
neurons
in
the
adult
brain:
the
role
of
sleep
and
consequences
of
sleep
loss.
Sleep
Medicine
Reviews
2009;13:187–94.
[25] Kochman
LJ,
Fornal
CA,
Jacobs
BL.
Suppression
of
hippocampal
cell
proliferation
by
short-term
stimulant
drug
administration
in
adult
rats.
European
Journal
of
Neuroscience
2009;29:2157–65.
[26]
Scheurink
AJ,
Boersma
GJ,
Nergardh
R,
Sodersten
P.
Neurobiology
of
hyperac-
tivity
and
reward:
agreeable
restlessness
in
anorexia
nervosa.
Physiology
and
Behavior
2010;100:490–5.
[27] Burden
VR,
White
BD,
Dean
RG,
Martin
RJ.
Activity
of
the
hypothalamic–pituitary–adrenal
axis
is
elevated
in
rats
with
activity-based
anorexia.
Journal
of
Nutrition
1993;123:1217–25.
[28]
Schoenfeld
TJ,
Gould
E.
Stress,
stress
hormones,
and
adult
neurogenesis.
Exper-
imental
Neurology
2012;233:12–21.
[29]
Klein
DA,
Mayer
LE,
Schebendach
JE,
Walsh
BT.
Physical
activity
and
cortisol
in
anorexia
nervosa.
Psychoneuroendocrinology
2007;32:539–47.
[30]
Barres
BA.
The
mystery
and
magic
of
glia:
a
perspective
on
their
roles
in
health
and
disease.
Neuron
2008;60:430–40.
[31]
Banasr
M,
Dwyer
JM,
Duman
RS.
Cell
atrophy
and
loss
in
depression:
rever-
sal
by
antidepressant
treatment.
Current
Opinion
in
Cell
Biology
2011;23:
730–7.
[32] Fernandez-Aranda
F,
Pinheiro
AP,
Tozzi
F,
Thornton
LM,
Fichter
MM,
Halmi
KA,
et
al.
Symptom
profile
of
major
depressive
disorder
in
women
with
eat-
ing
disorders.
Australian
and
New
Zealand
Journal
of
Psychiatry
2007;41:
24–31.