J Mol Neurosci (2007) 32:72-79
DOI 10.1007/s 12031-007-0016-5
Decoding of Dopaminergic Mesolimbic Activity
and Depressive Behavior
Alexander Friedman i liana Deri i Yaakov Friedman i
Eliyahu Dremencov. Sophia Goutkin.
Elizabeth Kravchinsky i Matti Mintz i Dino Levi i
David H. Overstreet. Gal Yadid
Received: 4 February 2007/Accepted: 6 February 2007/Published online: 26 April 2007
y Humana Press Inc. 2007
Abstract Dopaminergic mesolimbic and mesocortical sys-
tems are involved in hedonia and motivation, two core
symptoms of depression. However, their role in the
pathophysiology of depression and their manipulation to
treat depression has received little attention. Previously, we
showed decreased limbic dopamine (DA) neurotransmis-
sion in an animal model of depression, Flinder sensitive
line (FSL) rats. Here we describe a high correlation
between phase-space algorithm of bursting-like activity of
DA cells in the ventral tegmental area (VTA) and efficiency
of DA release in the accumbens. This bursting-like activity
A. Friedman" G. Yadid
Leslie Susan Gonda (Goldschmied) Multidisciplinary Brain
Research Center, Bar-Ilan University,
Ramat-Gan 52900, Israel
I. Deft" E. Dremencov i G. Yadid ([~)
The Mina & Everard Goodman Faculty of Life Sciences,
Ramat-Gan 52900, Israel
Department of Mathematics, Jerusalem College of Technology,
Jerusalem 91160, Israel
E. Dremencov i S. Goutkin i E. Kravchinsky
Department of Bio-Informatics, Jerusalem College of Technology,
Jerusalem 91160, Israel
M. Mintz" D. Levi
Psychobiology Research Unit, Department of Psychology,
Tel-Aviv University, Tel-Aviv 69978, Israel
D. H. Overstreet
Department of Psychiatry, University of North Carolina,
Chapel Hill, NC 27515, USA
of VTA DA cells of FSL rats is characterized by a low
dimension complexity. Treatment with the antidepressant
desipramine affected both the dimension complexity of cell
firing in the VTA and rate of DA release in the accumbens,
as well as alleviating depressive-like behavior. Our findings
indicate the potential usefulness of monitoring limbic
dopaminergic dynamics in combination with non-linear
analysis. Decoding the functionality of the dopaminergic
system may help in development of future antidepressant
Keywords Dopamine. Electrophysiology-
Flinders Sensitive Line of rats i Nonlinear analysis-
Ventral tegmental area- Dimension analysis- Swim test-
For decades, clinical treatment of depression has usually
involved antidepressants that target noradrenergic and
serotonergic neurotransmission. Although currently used
antidepressants affect monoamine neurotransmission almost
immediately, behavioral benefits are usually only seen after
4 to 6 weeks of use (Ordway and Mann 2002).
Dopaminergic neurons label environmental stimuli with
appetitive value, as well as, predict rewards and motivating
events (Ikemoto and Panksepp 1999; Schultz et al. 2000).
Since anhedonia and loss of motivation are core
characteristics of depression (Kapur and Mann 1992),
altered limbic DA neurotransmission may be involved in
depression. In fact, the following suggest involvement of
the mesolimbic DA system in depression and response to
chronic antidepressant therapies. Levels of homovanillic
J Mol Neurosci (2007) 32:72-79 73
acid, the major metabolite of DA, are reduced in the
cerebrospinal fluid (CSF) of depressed patients (Kapur
and Mann 1992; Reddy et al. 1992). Pharmacological
interventions that block or decrease DA, are associated with
the induction and deepening of depression (Kapur and
Mann 1992; Ordway and Mann 2002). After acute
administration of sulpiride (D2-1ike receptor antagonist)
(Cervo and Samanin 1988; Willner et al. 2005), depressed
patients successfully treated with serotonin selective reup-
take inhibitors (SSRIs), display increased mood and
psychomotor symptoms associated with depression (Willner
et al. 2005). Conversely, DA agonists can mimic antide-
pressant effects (Willner et al. 2005).
Mesolimbic DA pathways originate in the ventral teg-
mental area (VTA) (Cooper 2002). VTA neuronal activity
consists of mixed bursting and single-spike firing modes
(Di Mascio et al. 1999; Grace and Bunney 1984a). Each
burst event can be characterized by its length (number of
spikes) or duration (s). A burst is defined as a sequence of
spikes started with an interspike interval (ISI) <80 ms and
ending with the concurrence of two spikes with an ISI
>160 ms (Di Mascio et al. 1999; Grace and Bunney 1984a,b;
Wang 1981). DA release at the terminal site due to burst-
ing activity of DA cells was found to increase greater than
that due to single-spiking mode (Cooper 2002). In ad-
dition, bursting activity corresponds to synaptic plasticity
in the VTA during reward-related learning (Schultz and
Phase-space dimensions were suggested by Grassberger
and Procaccia (1983) as a tool to distinguish stochastic
processes from deterministic chaotic time series (which
may be low-dimensional or high-dimensional). Phase-
space (or state space) dimensions describe the level of
complexity of the system and allow representation of the
behavior of a system in a geometric form. The number of
dimensions required for the phase space is a function of the
"degrees of freedom" of the system. A dynamical system
consists of two parts: the notions of a state (the essential
information about a system) and a dynamics (a rule that
describes how the state evolves with time). This evolution
can be visualized in a phase space.
The genetically selected Flinders Sensitive Line (FSL)
of rats was used. These rats exhibit behavioral features
characteristic of depression, such as reduced locomotion,
reduced activity in the swim test, increased anhedonia in
response to chronic mild stress, increased amount and re-
duced onset of REM (rapid eye movement) sleep, cognitive
difficulties, and reduced body weight (for review see
Overstreet 1993). Chronic, but not acute, treatment of the
FSL rats with antidepressants, such as desipramine, imipra-
mine, sertraline, and fluoxetine, almost abrogates their
behavioral manifestations of depression (for review see
Yadid et al. (Overstreet et al. 2005; Yadid et al. 2000).
Using in vivo microdialysis, we previously demonstrated
decreased extracellular DA levels in the nucleus accumbens
(NAc) of FSL rats (Dremencov et al. 2004b, 2005; Zangen
et al. 2001). However, conventional analysis of neuron-
firing time series showed to unable predicting abnormalities
in this model, whereas non-linear analysis of neuron-
spiking time-series did (Dremencov et al. 2004a; Friedman
et al. 2005).
Cell bursting-like activity is correlated with the func-
tional dynamics of the accumbal DA release. Therefore,
we analyzed, in the current study, the burst dynamics
recorded from the VTA of FSL and control rat, separately
from the whole firing time series (ISis). Moreover, a se-
lective DA reuptake inhibitor, GBR 12009 (Sarre et al.
2004), was administered into the accumbens to test if DA
accumulation is effected by altered DA reuptake or im-
paired tonic regulation (Grace 1991). Finally, we examined
the correlation between VTA-cell firing, accumbal DA re-
lease, and depressive-like behavior in an animal model for
Materials and Methods
Male Sprague-Dawley and FSL rats (an animal model of
depression (Dremencov et al. 2004b) (250-300 g) were
housed three per cage under conditions of constant
temperature (25~ and humidity (50%), with a 12:12 h
light/dark cycle. Food and water were provided ad libitum
Desipramine hydrochloride (Sigma Chemical CO.,
St. Louis, MO), a tricyclic antidepressant, was dissolved
in saline and administered i.p. daily for 14 days to FSL rats
(5 mg/kg). All animal procedures were approved by the
Bar-Ilan University Animal Care Committee and were
carried out in accordance with the National Institutes of
Health Guide for the Care and Use of Laboratory Animals.
In Vivo Electrophysiology
Rats were anesthetized with chloral hydrate (400 mg/kg,
i.p.) and a recording electrode (tungsten, 5 M[~) was ste-
reotaxically inserted into a tract in the VTA (4.2 mm anterior
to the interaural line, 1.0 mm lateral from the midline, and
7.8-8.8 mm ventral to lambda (Friedman et al. 2005). The
level of anesthesia was kept constant and a slow heart
rate (300 beats/min) was maintained monitoring continu-
ous electrocardiogram recordings (Friedman et al. 2005).
74 J Mol Neurosci (2007) 32:72-79
Single unit recordings from VTA neurons of Sprague-
Dawley, FSL rats, and desipramine-treated FSL rats treated
(n = 10 per each group), respectively, were performed using
amplitude discriminationi After the last recording from each
animal, the location of the electrode tip was determined by
electrolytic lesioning (0.5 mA of DC current, 15 s). After
transcardially perfusing the rats, their brains were removed
and post-fixed in 10% formaldehyde. Frozen sections
(50 ~tm) were cut and examined under a microscope to
verify the placement of the electrode tip in the VTA. Only
data sets recorded from the VTA were analyzed.
Measurements of Dopamine Release
Sprague-Dawley (n = 10), FSL (n = 10), and desipramine-
treated FSL rats (n=10) were anesthetized with chloral
hydrate (400 mg/kg, i.p.; Merck Chemicals Ltd.; Darmstadt,
Germany). A microdialysis probe (2 mm in length, 20 kD
cutoff value; CMA/10; Carnegie Medicine; Stockholm,
Sweden) was surgically implanted into the shell of the
nucleus aecumbens (1.4 mm anterior and 1.2 mm lateral to
bregma; 7.6 mm ventral to the dura) of each rat using a
stereotaxic apparatus (David-Kopf Instruments; Tujunga,
CA) and cemented to the skull. After surgery, rats were
habituated for 22-24 h to a cylindrical microdialysis
chamber (35 cm diameterx40 cm high) that was con-
structed in our laboratory at Bar-Ilan University. Teflon
microdialysis tubing (MF-5164; Bioanalytical Systems; W.
Lafayette, IN), protected with a metal spring (30 cm in
length), was then passed from the perfusion pump to the
microdialysis probe via a swivel (Instech; Plymouth
Meeting, PA) and positioned above the center of the cage
to allow free movement of the rat. Artificial CSF (aCSF;
145 mM NaC1, 1.2 mM CaC12, 2.7 mM KC1, 1.0 mM
MgCI2, pH 7.4) was pumped continuously (1 ~tl/min)
through the dialysis probe using a microinjection pump
(CMA 100; Carnegie Medicine), starting upon the insertion
of the probe into the brain. Microdialysis samples were
collected from non-anesthetized, freely-moving rats. The
dialysates were collected during 30 min intervals into
polyethylene tubes containing 15 p.1 of a 0.02% EDTA
and 1% ethanol solution, and stored at -70~
subjected to HPLC for monoamine analysis. After collect-
ing samples for determination of baseline levels (mean of
the first four fractions), the aCSF was switched to aCSF
containing 10 ~tM of GBR 12909, a selective DA reuptake
inhibitor, for 3 h (six samples). The mean DA levels in the
nucleus accumbens, which were calculated as a percentage
of the basal DA levels, were used to assess DA release in
the nucleus accumbens.
A modified forced swim test protocol was conducted in a
cylindrical tank (40 cm high and 18 cm in diameter; con-
structed at Bar-Ilan University), which contained enough
water at 25~ so that the rat could not touch the bottom
with its hind paws. A rat was considered to have stopped
swimming when both front paws were immobile. The
animals were given a single 5-min exposure to the swim
tank after 14 days. This protocol was previously effective in
detecting the anti-immobility effects of classical antidepres-
sants (Overstreet et al. 1995).
Phase Space Dimension Algorithm
In chaos theory the correlation dimension or phase space
dimension (denoted by u) is a measure of the dimension-
ality of the space occupied by a set of random points. For
example, if we have a set of random points on the real
number line between 0 and 1, the correlation dimension
will be u=l, while if they are distributed on say, a triangle
embedded 3-space (or m-space, for that matter), the
correlation dimension will be u=2. This is what we would
intuitively expect from a measure of dimension. The real
utility of the correlation dimension is in determining the
(possibly fractional) dimensions of fraetal objects. There
are other methods of measuring dimension (e.g. the
Hausdorff dimension, the box-counting dimension, and
the information dimension) but the correlation dimension
has the advantage of being straightforwardly and quickly
calculated, and is often in agreement with other calculations
If we have a set of N points in an m-dimensional space:
X(i) = [xl(i), x2(i), x3(i),...x,(i)] where i=1,2, U then
the correlation integral C(e) is calculated by: C(e)=
hm ~ where g is the total number of pairs of points
which have a distance between them that is less than or
equal to distance ~. As the number of points tends to
infinity, and the distance between them tends to zero, the
correlation integral, for small values of ~:, will take the
form: C(c)-c v.
If the number of points is sufficiently large, and evenly
distributed, a plot of the correlation integral versus ~: will
yield an estimate of u. This idea can be qualitatively
understood by realizing that for higher dimensional objects,
there will be more ways for points to be close to each other,
and so the number of pairs close to each other will rise
more rapidly for higher dimensions.
Grassberger and Procaceia (1983) is the main reference
for this technique, and gives the results of such estimates
for a number of fractal objects, as well as comparing the
J Mol Neurosci (2007) 32:72-79 75
values to other measures of fractal dimension. The
technique can be used to distinguish between chaotic and
truly random behavior.
In chaos theory the correlation integral is the mean
probability that the states at two different times are close:
C(E) = N--+o~lim ~-~-~ |
where N is the number of considered states ~(i), e is a
threshold distance, [1. II a norm (e.g. Euclidean norm) and
(9(.) the Heaviside step function. If only a time series is
available, the phase space can be reconstructed by using a
time delay embedding (see Takens' theorem):
2(i) : [u(i), u(i + r),.., u(i + l'(m - 1))]
where u(i) is the time series, m the embedding dimension
and ~-the time delay.
The correlation integral is used to estimate the correla-
An estimator of the correlation integral is the correlation
Discrete Form of Heaviside Step Function
We can also define an alternative form of the unit step as a
function of a discrete variable n:
[.0,n < 0
where n is an integer.
This function is the cumulative summation of the
u[n]-- ~ 6[k]
~[k] = &0
is the discrete unit impulse function.
All data are expressed as mean•
(SEM). Significance was determined by a one way
ANOVA and p<0.05 was considered significant. Variance
data are expressed as variance•
cance of variances was determined by F test, and p<0.05
was considered significant. Correlation of parameters was
determined by Pearson test.
error of the mean
ISis of DA VTA Cells
ISI activity was monitored from the VTA (Fig. lb~). The
absolute refractory period of VTA dopaminergic neurons is
SD FSL FSL
i ~ 1
1 200 300 400 500
._c -~. 3
200 300 400 500
Figure 1 DA concentrations in the extracellular space of the NAc and
DA cells activity in the VTA. a Basal and GBR-12909-mediated DA
accumulation in the extracellular space of the NAc. Dialysates from
the NAc were collected with and without GBR-12909 (10 ~tM). The
mean• of baseline and the peak level after perfusion of GBR-
12909 are depicted. (N= 10 per group), b Representative recording of
DA-like cell activity in control Sprague-Dawley rats. e Representative
recording of DA-like cell activity in FSL rats
76 J Mol Neurosci (2007) 32:72-79
2.5 ms (Grace and Bunney 1984a,b). The longest single inter-
spike intervals (ISI) we recorded from the VTA was about
3.5 s. Mean cell frequencies were around 4-5.5 Hz. Mean
frequencies did not differ significantly between the FSL,
FSL treated with desipramine, and Sprague-Dawley rats.
The burst was defined as a sequence of spikes starting
with an ISI <80 ms and ending with the concurrence of two
spikes with an ISI >160 ms (Di Mascio et al. 1999; Grace
and Bunney 1984a,b; Wang 1981). Accordingly, bursts
were separated from single spikes firing. Mean burst length
(spikes/burst) and duration did not differ significantly be-
tween the FSL, FSL treated with desipramine, and Sprague-
Dawley rats (Table 1A).
Phase-space Dimensions of ISis, Single Spikes, and Bursts
Since the ability of a burst to increase DA release from
nerve terminals depends both on burst length and on the
neuron-firing rate within a given burst (Cooper 2002), we
introduced a mathematical descriptive parameter, which is
defined as burst length (number of spikes) multiplied by the
firing rate within the burst (number of spikes/second). This
mathematical parameter may be also defined as the burst
influence. This Burst influence was calculated for each
individual burst. Burst influence series were analyzed by
phase-space dimensions algorithm. The stability of phase-
space dimensions for each animal at non-intersecting time
intervals was verified prior to further analysis. Since it was
found relatively stable (average variation <0.7), it was used
to measure the neuronal activity. The mean phase-space
value was calculated for the three types of time-series
(whole ISI, single-spike events, and sequence of the bursts)
for the three experimental groups.
The phase-space dimensions of entire VTA ISI time
series and of single spike events were larger (indicating
highly chaotic processes) than that of bursting-like activity
(indicating nonlinear determinism) (Table 1C). For FSL,
desipramine-treated FSL, and Sprague-Dawley rats, the
phase-space dimensions of the ISI time series and single
spike events, were similar. However, the phase-space
dimensions of the bursting-like activity of FSL and
Sprague-Dawley rats were significantly different. Treat-
ment of the FSL rats with desipramine for 14 day increased
their bursting-like activity to levels similar to those of the
control Sprague-Dawley rats (F(2,27)=3.35, P<0.001, One
Table 1 Characteristics of 1SI, DA release and behavioral parameters
Sprague-Dawley 3 FSL FSL + Desipramine i
A. Burst characteristics
Mean burst length (num. of spikes)
Mean burst duration (s)
GBR 12909-induced DA release
ISI (0.2~).4 s)
C. Phase-space dimensions m Full ISI time-series
D. Dialysate DA levels (riM)
G. GBR 12909-induced DA levels (nM)
E. DA release rate (% baseline per rain) m
E Immobility (s) TM
- 1,444.91 h
6.7• 1.24 a
34.3• 10.3 e
91 • 13 a,k
4.11 +1.51 c
3.9• 1 a'l
9.8• 1.04 a
39.1 a:7.1 g
83 • 20 a'k
aNumber of rats in group~10
b 10 rats, 10 cells
c 10 rats, 10 cells
d 10 rats, 10 cells
e 10 rats
f 10 rats
g 10 rats
h Confidence value
i5 mg/kg i.p. for 14 days
k Swim test performed 22-24 h after last injection of desipramine.
l Significant difference between FSL and other two groups p<0.05, one way ANOVA for mean and F test for variance.
m A linear correlation dependence between the phas~space dimension of bursting-like activity, immobility and Speed of DA release (% baseline
per rain) was observed (Pearson test r=0.9973).
J Mo| Neurosci (2007) 32:72-79 77
Basal and GBR 12909-induced DA Accumulation in the
Extracellular Space of the Nucleus Accumbens
The basal levels of DA in the extracellular spaces of the
NAc of the FSL (N = 10), desipramine-treated FSL (N = 10),
and Sprague-Dawley (control; n = 10) rats were determined
by microdialysis (Table 1D, Fig. la). The basal DA levels
thus assessed were lower in the FSL than in the Sprague-
Dawley rats. After chronic (14 days) treatment with
desipramine, the low basal levels of DA in the FSL rats
were elevated (Table 1G, Fig. la) (F(2,27)=2.35, p<0.05,
One way ANOVA).
By perfusing GBR 12909, an inhibitor of DA reuptake,
through the microdialysis probe into the accumbens shell,
we were able to monitor the rate of DA release by its
accumulation in the extracellular space. We detected a
significant decrease in the rate of DA release (accumula-
tion) in the extraceUular space of FSL rats compared to
Sprague-Dawley (Table 1E). Administration of desipra-
mine for 14 days caused increases in FSL variances to
levels similar to, or exceeding, those observed in the control
Sprague-Dawley rats (F(2,27)=4.05, p<0.05, One way
Correlation of the Phase-space Dimension of Bursting-like
Activity with Rate of GBR 12909-induced DA Release
The rate of GBR 12909-mediated DA accumulation in the
NAc and phase-space dimension of bursting-like activity of
dopaminergic neurons in the VTA was significantly lower
((F(2,27)=5.15, p<0.05, One way ANOVA); F(2,27)=
3.35, P<0.001, One way ANOVA) in FSL rats than in the
Sprague-Dawley controls. Chronic treatment of FSL rats
with desipramine for 14 days raised the values of these
parameters in FSL rats to levels comparable to those of the
control Sprague-Dawley rats. A linear correlation depen-
dence between the phase-space dimension of bursting-like
activity and rate of GBR 12909-induced DA accumulation
was observed (Pearson test; r=0.9973).
Assessment of Depressive Behavior by Immobility during a
Forced Swim Test
Depressive behavior can be assessed by immobility during
a forced swim test. When immobility during the swim test
was assessed for all three experimental groups, the FSL rats
displayed greater immobility than the Sprague-Dawley
(control) rats (F(2,27)=4.25, P<0.05, One way ANOVA).
After chronic (14 day) treatment with desipramine the FSL
rats demonstrated the same amount of immobility as the
Sprague-Dawley rats. Thus implying that the specific dose
and course of desipramine treatment can alleviate the be-
havioral manifestations of depression.
Figure 2 Correlation between Neuronal Firing, rate of DA Release,
and Depressive Behavior. Neuronal firing was determined by phase-
dimension of bursting-like activity in the VTA, rate of the GBR
12909-mediate DA release by microdialysis was determined in the
NAc, and depressive behavior by immobility in a swim test
Correlation between Neuronal Firing, DA Release Rate,
and Depressive Behavior
A-3 dimensional correlation between neuronal firing, DA
release, and depressive behavior in FSL, desipramine-
treated FSL, and Sprague-Dawley rats was determined
using phase-space dimensions of bursting-like activity,
microdialysis, and immobility during swim test data
(Fig. 2). A high correlation of these three variables was
observed between the three experimental groups.
The underlying hypothesis of this study is that the bursting-
like activity of VTA dopaminergic neurons is essential for a
beneficial response to currently used antidepressants. We
found that the bursting-like activity in the VTA is highly
deterministic, signal regulated by a relatively small number
of parameters, unlike the chaotic-like single-spiking activ-
ity, which has random features (Table 1C). The chaotic
character of the neuronal activity in the VTA can be
explained by highly complex network of auto- and cross-
regulation of dopaminergie neurons (Adell and Artigas
2004), as well as by stochastic processes in the neuron
membrane, sueh as channel noise (Steinmetz et al. 2000).
78 J Mol Neurosci (2007) 32:72 79
FSL rats displayed a decreased dimension complexity of
bursting-like activity of VTA dopaminergic cells when
compared to the Sprague-Dawley rats used as controls.
This parameter was restored by repetitive antidepressant.
Chronic administration of the antidepressant desipramine to
FSL rats elevated the dimensional complexity of this
activity to levels observed in Sprague-Dawley rats. We
also observed a correlation between firing of neurons in the
VTA, the dynamics of DA release in the accumbens and
FSL rats, an animal model of depression, are character-
ized by a deficit in basal levels of DA release in the NAc
(Dremencov et al. 2004b; Dremencov et al. 2005; Zangen
et al. 2001). This may indicate a possible lower phasic
activity of VTA cells. Moreover, lowering DA levels in
terminal sites may increase DA synthesis and abnormal
phasic responses to compensate for the decreased responses
of the DA system (Floresco et al. 2003). Indeed we
observed higher DA tissue content in the Nac of FSL rats
(Zangen et al. 2001).
The release of mesolimbic DA is regulated by a number
of factors. Synaptic (phasic) levels are mediated primarily
by bursting events of cell bodies and lead to much larger
release then if these neuron fire irregular single spike mode
(Cooper 2002; Floresco et al. 2003; Lisman 1997; Schultz
1998). Phasic burst firing induces massive synaptic DA
release (Cooper 2002; Floresco et al. 2003), accompanied
by rapidly DA removal by reuptake before escaping the
synaptic cleft, whereas increased input of other neurons
activity to the DA synapse modulates tonic extrasynaptic
DA levels, which are less affected by reuptake processes
(Floresco et al. 2003).
In fact, blocking DA reuptake at NAc synapses by GBR
12909 increased DA levels 8-fold over baseline. Thus,
phasic activity massively contributes to DA release in the
NAc, but is masked by the fast and efficient reuptake of
DA. Although presynaptic modulation cannot be excluded,
we assume that DA accumulation due to blocking of DA
reuptake characterizes dominantly the phasic release of DA
in the NAc. The rate of accumulation of this accumbal
DA can be compared with the DA cell fining in the VTA.
Similarly, when DA reuptake is blocked by amphet-
amines, release of DA is phasic, not tonic (Floresco
et al. 2003). Comparing the rate of VTA activity and the
rate of DA release in the NAc, reveal that the efficacy of the
mesolimbic DA pathway is decreased in FSL rats. It is
worth to note that bursting-like activity in the VTA of rats is
still observed under chloral-hydrate anesthesia (Di Mascio
et al. 1999; Dremencov et al. 2004a). This can be explained
by high sensitivity of dopaminergic cells to spontaneous
inputs arriving from the cortex (Cooper 2002). However,
this spontaneous bursting by intrinsic clock oscillations
may occur in the membrane of dopaminergic cells and be
facilitated by acetylcholine input from the pedunculopon-
tine nucleus (Kitai et al. 1999).
Previously, we showed that FSL rats are characterized by
decreased variance of 0.2-0.4 s ISis, a variance that is
normalized by repetitive administration of desipramine
(Friedman et al. 2005). Therefore, we postulated that
variance of ISis is relevant to the dynamics of DA release
in limbic areas of the brain. Indeed, in this study we found a
correlation between the variance of 0.2-0.4 s ISis and
variance of DA release (Table 1B). In addition, FSL rats
demonstrated monotonic firing of the DA VTA cells, with
ISis of 0.2~).4 s. Thus, the parameter of ISI variance seems
to be relevant to the dynamics of DA neurotransmission
and prediction of antidepressant effects.
Based on phase-space dimension values, we conclude
that FSL (depressed) rats are characterized by different type
of bursting-like dynamics than normal (non-depressed) rats.
The phase-space dimensions of bursting-like activity of
FSL rats indicate a fully predictable linear process, while
that of Sprague-Dawley rats can be characterized as a more
non-linear deterministic process (Baker et al. 1996). Our
findings indicate the potential usefulness of monitoring
limbic dopaminergic dynamics during development of
future antidepressant drugs.
Acknowledgements This study was supported in part by a grant
from The Israel Science Foundation to GY. We wish to thank
Professors Moshe Abeles and Uziel Sandier for critically reviewing
the manuscript and Dr. Barbara Schick for her comments.
Adell, A., & Artigas, F. (2004). The somatodendrific release of
dopamine in the ventral tegmental area and its regulation by
afferent transmitter systems. Neuroscience and Biobehavioral
Reviews, 28, 415~431.
Baker, G. L., Gollub, J. P., & Blackburn, J. A. (1996). Inverting chaos:
Extracting system parameters from experimental data. Chaos, 6,
Cervo, L., & Samanin, R. (1988). Repeated treatment with imip-
ramine and amitriptyline reduced the immobility of rats in the
swimming test by enhancing dopamine mechanisms in the nu-
cleus accumbens. Journal of Pharmacy and Pharmacology, 40,
Cooper, D. C. (2002). The significance of action potential bursting
in the brain reward circuit. Neurochemistry International, 41,
Di Mascio, M., Di Giovanni, G., Di, M. V., & Esposito, E. (1999).
Decreased chaos of midbrain dopaminergic neurons after
serotonin denervation. Neuroscience, 92, 237-243.
Dremencov, E., Gispan-Herman, I., Rosenstein, M., Mendelman, A.,
Overstreet, D. H., Zohar, J., et al. (2004a). The serotonin-
dopamine interaction is critical for fast-onset action of antide-
pressant treatment: In vivo studies in an animal model of
depression. Progress in Neuro-Psychopharmaeology & Biologi-
cal Psychiatry, 28, 141-147.
Dremencov, E., Nahshoni, E., Levy, D., Mintz, M., Overstreet, D. H.,
Weizman, A., et al. (2004b). Dimensional complexity of the
J Mol Neurosci (2007) 32:72-79 79
neuronal activity in a rat model of depression. Neuroreport, 15,
Dremencov, E., Newman, M. E., Kinor, N., Blatman-Jan, G.,
Schindler, C. J., Overstreet, D. H., et al. (2005). Hyperfunction-
ality of serotonin-2C receptor-mediated inhibition of accumbal
dopamine release in an animal model of depression is reversed by
antidepressant treatment. Neuropharmacology, 48, 34-42.
Floresco, S. B., West, A. R., Ash, B., Moore, H., & Grace, A. A.
(2003). Afferent modulation of dopamine neuron firing differen-
tially regulates tonic and phasic dopamine transmission. Nature
Neuroscience, 6, 968-973.
Friedman, A., Dremencov, E., Crown, H., Levy, D., Mintz, M.,
Overstreet, D. H., et al. (2005). Variability of the mesolimbic
neuronal activity in a rat model of depression. Neuroreport, 16,
Grace, A. A. (1991). Phasic versus tonic dopamine release and the
modulation of dopamine system responsivity: A hypothesis for
the etiology of schizophrenia. Neuroscience, 41, 1-24.
Grace, A. A., & Bunney, B. S. (1984a). The control of firing pattem in
nigral dopamine neurons: Burst firing. Journal of Neuroscience,
Grace, A. A., & Bunney, B. S. (1984b). The control of firing pattern in
nigral dopamine neurons: Single spike firing. Journal of
Neuroscience, 4, 2866-2876.
Grassberger, P., & Procaccia, I. (1983). Measuring the strangeness of
strange attractors. Physica D, 9, 189 208.
Ikemoto, S., & Panksepp, J. (1999). The role of nucleus accumbens
dopamine in motivated behavior: A unifying interpretation with
special reference to reward-seeking. Brain Research: Brain
Research Reviews, 31, 6-41.
Kapur, S., & Mann, J. J. (1992). Role of the dopaminergic system in
depression. Biological Psychiatry, 32, 1-17.
Kitai, S. T., Shepard, P. D., Callaway, J. C., & Scroggs, R. (1999).
Afferent modulation of dopamine neuron firing patterns. Current
Opinion in Neurobiology, 9, 690-697.
Lisman, J. E. (1997). Bursts as a unit of neural information: Making
unreliable synapses reliable. Trends Neuroscience, 20, 38-43.
Ordway, G. A., & Mann, J. J. (2002). Neurocircuitry of mood
disorders. In K. L. Davis, J. T. Coyle, & C. Nemeroff (Eds.),
Neuropsychopharmacology: The Fifth Generation of Progress
(pp. 1051-1064). Lippincott Williams & Wilkins: Philadelphia.
Overstreet D. H. (1993). The Flinders Sensitive Line rats: A genetic
animal model of depression. Neuroscience and Biobehavioral
Reviews, 17(1), 51-68 PMID: 8455816 [PubMed - indexed for
Overstreet, D. H., Friedman, E., Mathe, A. A., & Yadid, G. (2005).
The Flinders Sensitive Line rat: A selectively bred putative
animal model of depression. Neuroscience and Biobehavioral
Reviews, 29, 739-759.
Overstreet, D. H., Pucilowski, O., Rezvani, A. H., & Janowsky, D. S.
(1995). Administration of antidepressants, diazepam and psycho-
motor stimulants further confirms the utility of Flinders Sensitive
Line rats as an animal model of depression. Psychopharmacology
(Berl), 121, 27-37.
Reddy, E L., Khanna, S., Subhash, M. N., Channabasavanna, S. M., &
Rao, B. S. (1992). CSF amine metabolites in depression.
Biological Psychiatry, 31, 112-118.
Sarre, S., Yuan, H., Jonkers, N., Van Hemelrijek, A., Ebinger, G., &
Michotte, Y. (2004). In vivo characterization of somatodendritic
dopamine release in the substantia nigra of 6-hydroxydopamine-
lesioned rats. Journal of Neurochemistry, 90, 29-39.
Schultz, W. (1998). Predictive reward signal of dopamine neurons.
Journal of Neurophysiology, 80, 1-27.
Schultz, W., & Dickinson, A. (2000). Neuronal coding of prediction
errors. Annual Review of Neuroscience, 23, 473-500.
Schultz, W., Tremblay, L., & Hollerman, J. R. (2000). Reward
processing in primate orbitofrontal cortex and basal ganglia.
Cerebral Cortex, 10, 272-284.
Steinmetz, P. N., Manwani, A., Koch, C., London, M., & Segev, I.
(2000). Subthreshold voltage noise due to channel fluctuations in
active neuronal membranes. Journal of Computational Neuro-
science, 9, 133-148.
Wang, R. Y. (1981). Dopaminergic-neurons in the rat ventral
tegmental area. 1. Identification and characterization. Brain
Research Reviews, 3, 123-140.
Willner, P., Hale, A. S., & Argyropoulos, S. (2005). Dopaminergic
mechanism of antidepressant action in depressed patients.
Journal of Affective Disorders, 86, 37-45.
Yadid, G., Nakash, R., Deft, I., Tamer, G., Kinor, N., Gispan, I., et al.
(2000). Elucidation of the neurobiology of depression: Insights
from a novel genetic animal model. Progress in Neurobiology,
Zangen, A., Nakash, R., Overstreet, D. H., & Yadid, G. (2001).
Association between depressive behavior and absence of seroto-
nin~dopamine interaction in the nucleus accumbens. Psycho-
pharmacology (Berl), 155, 434-439.