Neuroscience 275 (2014) 12–21

DEMYELINATING EVIDENCES IN CMS RAT MODEL OF DEPRESSION:A DTI STUDY AT 7 T

B. S. HEMANTH KUMAR, S. K. MISHRA, R. TRIVEDI,S. SINGH, P. RANA AND S. KHUSHU *

NMR Research Centre, Institute of Nuclear Medicine and

Allied Sciences (INMAS), Brig. SK Mazumdar Marg, Timarpur,

Delhi 110054, India

Abstract—Depression is among the most debilitating dis-

eases worldwide. Long-term exposure to stressors plays a

major role in development of human depression. Chronic

mild stress (CMS) seems to be a valid animal model for

depression. Diffusion tensor imaging (DTI) is capable of

inferring microstructural abnormalities of the white matter

and has shown to serve as non-invasive marker of specific

pathology. We developed a CMS rat model of depression

and validated with behavioral experiments. We measured

the diffusion indices (mean diffusivity (MD), fractional

anisotropy (FA), axial (kk) and radial (k\) diffusivity) to

investigate the changes in CMS rat brain during depression

onset. Diffusion indices have shown to be useful to discrim-

inate myelin damage from axon loss. DTI was performed in

both control and CMS rats (n= 10, in each group) and maps

of FA, MD, kk and k\ diffusivity values were generated using

in-house built software. The diffusion indices were calcu-

lated by region of interest (ROI) analysis in different brain

regions like the frontal cortex, hippocampus, hypothalamus,

cingulum, thalamus, caudate putamen, corpus callosum,

cerebral peduncle and sensory motor cortex. The results

showed signs of demyelination, reflected by increased MD,

decreased FA and increased k\. The results also suggest

a possible role of edema or inflammation concerning the

brain morphology in CMS rats. The overall finding using

DTI suggests there might be a major role of loss of myelin

sheath, which leads to disrupted connectivity between

the limbic area and the prefrontal cortex during the onset

of depression. Our findings indicate that interpretation

of these indices may provide crucial information about

the type and severity of mood disorders.

� 2014 IBRO. Published by Elsevier Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.neuroscience.2014.05.0370306-4522/� 2014 IBRO. Published by Elsevier Ltd. All rights reserved.

*Corresponding author. Address: NMR Research Centre, Institute ofNuclear Medicine and Allied Sciences (INMAS), DRDO, LucknowRoad, Timarpur, Delhi, India. Tel: +91-11-23905313; fax: +91-11-23919509.

E-mail address: skhushu@yahoo.com (S. Khushu).Abbreviations: CC, corpus callosum; Cing, cingulum; CMS, chronicmild stress; CON, control group; CP, cerebral peduncle; CUP, caudateputamen; DTI, diffusion tensor imaging; FA, fractional anisotropy; FC,frontal cortex; FST, forced swim test; Hip, hippocampus; HTh,hypothalamus; MD, mean diffusivity; MRS, magnetic resonancespectroscopy; NAA, N-acetylaspartate; OFT, open field test; ROI,region of interest; SCT, sucrose consumption test; SMC, sensorymotor cortex; Th, thalamus.

12

Key words: diffusion tensor imaging, chronic mild stress,

fractional anisotropy, mean diffusivity, axial diffusivity, radial

diffusivity.

INTRODUCTION

Major depression is among the most debilitating diseases

worldwide (Falagas et al., 2007). It is reported that,

long-term exposure to multiple stressor leads to the

development of human depression (Anisman and

Zacharko, 1982; Brown and Harris, 1993). For better

understanding about depression, different animal models

have been developed to mimic these conditions. Out of

which, chronic mild stress (CMS) procedure seems to

be a valid animal model for depression, based on its

resemblance with several depressive symptoms

(Willner, 1997). The CMS model of depression has good

face validity, construct validity and predictive validity,

which makes it suitable for investigating depression-

induced alterations (Willner, 1997). Many studies have

been carried out on different brain regions associated with

depression and have documented the involvement of the

limbic system (Davidson et al., 2002). Studies have

reported abnormalities in the dorsolateral prefrontal cor-

tex (DLPFC) (Pause et al., 2003), anterior cingulate cor-

tex (ACC) (Gonul et al., 2004; Mayberg et al., 2005) in

depressed human patients. Therefore, CMS model could

be considered as best suitable model for studying these

specific mechanisms. The development of neuroimaging

technology that permits in vivo characterization of ana-

tomical, physiological and neurochemical correlates of

depression has enabled significant advances toward illu-

minating the pathophysiology of these conditions. Diffu-

sion tensor imaging (DTI) measures the diffusion of

water molecules, which reflect the microstructural organi-

zation of the tissues of interest. Most commonly used indi-

ces in DTI are mean diffusivity (MD) and fractional

anisotropy (FA) (Basser and Pierpaoli, 1996). MD is a

measure of the directionally averaged magnitude of diffu-

sion and is related to cell density, size, and parenchyma

permeability. FA represents the degree of diffusion anisot-

ropy, and reflects the degree of alignment of cellular

structure (Basser and Pierpaoli, 1996). They are known

to characterize alterations in myelin sheath and cell mem-

brane integrity, but not specific to demyelination or axonal

loss. MD has been widely investigated in psychiatric dis-

orders (Regenold et al., 2006; Shin et al., 2006). De Lisi

et al. (2006) have suggested that MD may be more

B. S. Hemanth Kumar et al. / Neuroscience 275 (2014) 12–21 13

sensitive to brain abnormalities compared to volume

assessments. Apart from MD and FA, an increasing body

of evidence from animal studies has shown that axial (kk)and radial (k\) diffusivity derived from DTI hold promise

as biomarkers of axonal damage and demyelination

(Sun et al., 2006). Axial diffusivity (kk) measures water dif-

fusion parallel to the fibers, whereas, in radial diffusivity

(k\) water diffusion is perpendicular to the fibers. These

directional diffusivities derived from DTI have shown to

reflect changes in white matter pathology in several ani-

mal models (Song et al., 2002; Schwartz et al., 2005).

Clinical and pathologic studies have found that axial diffu-

sivity tends to reflect axonal integrity and radial diffusivity

corresponds more closely to myelin integrity (Song et al.,

2002, 2003; Kim et al., 2006). Any changes in their values

are believed to represent axon loss or demyelination

(Song et al., 2005; Wheeler-Kingshott and Cercignani,

2009). Therefore, the MD and FA values along with kkand k\ may be of great significance in understanding

the mechanism underlying depression.

Given this background, in the present study, we

performed DTI studies on control and CMS rats to

establish the underlying pathology in depression.

Diffusion indices (FA, MD, kk and k\) in the CMS rat

brain were studied in brain regions like the frontal cortex

(FC), hippocampus (Hip), hypothalamus (HTh),

cingulum (Cing), thalamus (Th), caudate putamen

(CUP), corpus callosum (CC), cerebral peduncle (CP)

and sensory motor cortex (SMC). The goal of this study

was to investigate gray and white matter abnormalities

and water diffusivity, as reflected by FA, MD, kk and

k\, using DTI in the CMS animal model of depression.

EXPERIMENTAL PROCEDURES

Animals

Studies were carried out on 20 healthy male Sprague–

Dawley (SD) rats weighing 200–300 g (about

3–4 months of age). The animals were randomly divided

into two main groups, control group (CON) and CMS

group, (n= 10) in each group and were housed in a

group of five, at an average room temperature of 23 �Cand humidity of 50–60%, proper day and night cycle

from 8 AM to 8 PM was maintained regularly. Food and

water were given ad libitum, unless, otherwise stated

and their food intake was visually inspected. All rats

were acclimatized to the room condition for 2 weeks

Table 1. Time and length of stressors used in the chronic mild stress procedu

Chronic mild stress procedure

Day 1 Day 2 Day 3

Water deprivation 1200 ? 1200

Empty bottle 0900–1100

Food deprivation 1700

Restricted food access

Overnight illumination 1700 ? 1000

Cage tilt 1100–1800

Soiled cage

Isolation

prior to the beginning of experiments. The study was

approved by the animal ethics committee of our institute.

CMS regimen

The CMS procedure was performed as described

elsewhere (Kumar et al., 2012). Briefly, the CMS group

rats were exposed to a variety of stressors in a random

order for a period of 6 weeks. The details of the study

procedure, including variety, time, and length of activity

are as shown in Table 1. The experimental setup includ-

ing the CMS regime and the stressors applied is shown

in Fig. 1. Control animals were given ordinary daily care

and housed separately in an undisturbed, safe and calm

environment. The animal body weight was monitored on

regular basis throughout the experiment.

Behavioral experiments

Behavioral experiments like open field test (OFT), forced

swim test (FST) and sucrose consumption test (SCT)

were performed on control group and CMS group rats

as described in detail elsewhere (Kumar et al., 2012).

Briefly, in OFT the rats were placed in the center of the

apparatus and then allowed to behave and move freely

for 5 min. At the same time, rat’s behavior was recorded

and taped for further analysis. Following this, open field

activity was scored manually by technical personnel,

who were blinded to whether the animals were in the con-

trol or CMS group. Between each test, the arena was

thoroughly cleaned with a 5% ethanol solution.

In FST procedure the rats were made to swim

forcefully by placing rats in individual glass cylinders

(50 cm tall � 25 cm in diameter) containing 23–25 �Cwater and 30 cm deep. The swim test was conducted in

two sessions, one session for 15 min (trial) and next,

the actual session which lasted for 5 min. During actual

session, rat’s activity inside the water was recorded for

further analysis. Activity like swimming, immobility,

climbing and diving was observed and scored during the

session. Between each test, the glass cylinder was

thoroughly cleaned.

SCT was performed after 1 week of initial habituation

to the laboratory environment. Rats were exposed to two

standard drinking bottles, one bottle containing 1%

sucrose and the other normal tap water, for one week.

After this preliminary phase, rats were water deprived

over night and exposed to the sucrose solution and tap

water for one hour in the morning. The intake baseline

re

Day 4 Day 5 Day 6 Day 7

0900–1800 None

None

? 1000 None

0900–1000 None

1700 ? 1000 None

None

1400 ? 1000 None

1700 ? 1000 None

Fig. 1. Time schedule of procedures used in the present study. The numbers indicate the weeks. Sucrose preference tests were carried out on end

of each week. Open-field tests were conducted on week 0 and week 6 of the chronic mild stress procedure. Forced swim test was conducted on

week 6 of the chronic mild stress procedure.

14 B. S. Hemanth Kumar et al. / Neuroscience 275 (2014) 12–21

for the sucrose solution was established, which

corresponded to the average of three consecutive

measurements. The bottles were weighed before and

after the test and the value was subtracted which gives

the amount of sucrose consumption.

Behavioral experiments like OFT and FST were

conducted on CMS group rats before and after the CMS

regime. SCT was conducted on both Control and CMS

group rats at the end of each week of CMS regime.

In vivo DTI data acquisition and processing

All MRI experiments were carried out on a 7T horizontal

bore animal MRI scanner (Bruker Biospec USR 70/30,

AVANCE III). The scanner is equipped with a BGA12S

gradient system capable of producing 400 mT/m pulse

gradients in each of the three orthogonal axes and

interfaced to a Bruker Paravision 5.1 console.

Anesthesia was induced by i.p. injection of a mixture of

xylazine (10 mg/kg BW) and ketamine (80 mg/kg BW).

Anaesthetized animals (both CON and CMS) were

placed in prone position on an animal bed and then slid

into the center of the magnet bore. Radio frequency

(RF) excitation was accomplished using 72-mm inner

diameter (ID) linear birdcage coil and 30 mm receive

only surface coil was used for signal reception. The

first- and second-order localized shimming was

performed using field map based shimming (MAPSHIM),

a full-width half-maximum line width of water signal of

612 Hz was achieved. MRI protocol included turbo

RARE T2-weighted (TR/TE = 2500 ms/13 ms, slice

thickness 1 mm, FOV 2.5 � 2.5 cm, matrix size

256 � 256) and DTI sequence. DTI images were

acquired using a multi-slice, multiple-shot echo planar

imaging (EPI) sequence with the following parameters:

TR/ TE = 3800 ms / 32 ms, slice thickness 1 mm, FOV

Fig. 2. Representative figure showing image of anatomical T2, FA map and c

figure legend, the reader is referred to the web version of this article.)

4 � 4 cm, matrix size 128 � 128 with a spatial resolution

of 0.312 mm/pixel. Number of gradient directions 46,

b-value 700 s/mm2, diffusion gradient duration 7.5 ms

and a gradient separation of 12 ms was used. The DTI

data were processed as described in detail elsewhere

(Saksena et al., 2008). Briefly, Maps of FA and MD

were generated using in-house built JAVA-based

software (Saksena et al., 2008) and the axial (kk)and radial (k\) diffusivity values were manually calculated

from the eigenvalues obtained from the software. The

data were distortion corrected for shear, scale, rotation

and translation using the Automated Image and Registra-

tion (AIR) package (Woods et al., 1993). The distortion

corrected data were then interpolated to attain isotropic

voxels and decoded to obtain the tensor field for each

voxel. The tensor field data were then diagonalized using

the analytical diagonalization method to obtain the eigen-

values (k1, k2 and k3) and the three orthonormal eigen-

vectors (e1, e2, and e3). The tensor field data were

then used to compute the DTI metrics, MD (Eq. (1)), FA

(Eq. (2)), kk (Eq. (3)) and k\ (Eq. (4)) diffusivity values

for each voxel.

MD ¼ k1þ k2þ k33

ð1Þ

FA ¼ 1ffiffiffi2p

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiðk1� k2Þ2 þ ðk2� k3Þ2 þ ðk1� k3Þ2

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffik12 þ k22 þ k32

p ð2Þ

kðkÞ ¼ k1 ð3Þ

kð?Þ ¼ k2þ k32

ð4Þ

Fig. 2, shows the representative image of anatomical

T2, FA map and color-coded FA map. The FA, MD, kkand k\ values were calculated by placing region of

olor-coded FA map. (For interpretation of the references to color in this

B. S. Hemanth Kumar et al. / Neuroscience 275 (2014) 12–21 15

interest (ROI) on FC, Hip, HTh, Cing, Th, CUP, CC, CP

and SMC bilaterally as illustrated in Fig. 3. The ROIs

were manually drawn on the above mentioned regions

according to paxinos stereotactic coordinates (Paxinos

and Watson, 2007) and the size of ROI varied from

3 � 3 to 8 � 4 pixels. The shape and size of ROI chosen

in each region was based on the best fit to the structure of

interest.

Statistical analysis

The statistical analysis was carried out using Sigmaplot

11.0 (Systat, USA). Data were expressed as

mean ± SEM values. Significant differences between

groups Control and CMS (p 6 0.05) were identified by a

MANOVA, with all p-values corrected for multiple

Fig. 3. Representative figure showing color-coded FA (A, B, C, G, H and I) an

Control images showing placement of ROIs [right frontal cortex (1), caudate

hippocampus (5), corpus callosum (7), hypothalamus (8) and cerebral pe

anisotropy in corpus callosum in the CMS rat (arrow) compared with control r

this figure legend, the reader is referred to the web version of this article.)

comparisons by Bonferroni corrections. A correlation

analysis was performed between OFT and diffusivity

values using sigmaplot 11.0. A linear regression was

applied for body weight comparison and the sucrose

consumption test was analyzed using a repeated

measure one-way ANOVA followed with a pair wise

multiple comparison procedure (Bonferroni) post hoc

t-test.

RESULTS

Food intake and body weight

Food intake and growth rates of control (n= 10, CON)

and CMS (n= 10) animals were similar during the

acclimatization period. The food intake reduced

d FA maps (D, E, F, J, K and L) from control (A–F) and CMS (G–L) rat.

-putamen (2), cingulum (3), thalamus (4), sensory motor cortex (5),

duncle (9)] for DTI indices calculation. From FA maps, decreased

at is well appreciated. (For interpretation of the references to colour in

16 B. S. Hemanth Kumar et al. / Neuroscience 275 (2014) 12–21

gradually with time in the CMS group as compared to

CON group. Body weight was measured at regular

intervals from week 0 to week 6 of the chronic mild

stress regime. At the beginning of the experiment, the

average body weight of CON and CMS group rats was

around 237 ± 5 and 227 ± 5 g respectively. The body

weight of the CMS group was lower than the CON

group during the CMS period (342 ± 2 g for CON and

305 ± 1.6 g for CMS, p-value < 0.05). In the course of

the 6-week experiment, control rats gained a total of

105 g in body weight whereas stressed rats gained only

78 g. The regression data showed a good linear fit along

CMS and CON groups with an R2 value of 0.973 for

CON and 0.969 in CMS animals (Fig. 4).

Fig. 5. Bar diagram showing the total sucrose consumption during

each week of the chronic mild stress regime in both control and

chronic mild stress rats.

Sucrose consumption test

At the baseline, relative sucrose intake did not differ

significantly between the two groups; but, there was a

gradual reduction in the relative intake of sucrose in the

CMS group as compared to CON. Tests of the main

effects for the period of CMS procedure (weeks 1–5)

showed a difference between the two groups for relative

sucrose intake. The one-way ANOVA followed by

Bonferroni t- test indicated that there was a significant

effect of CMS on reduction in sucrose consump-

tion (85 ± 0.4 in CON and 63 ± 0.6 in CMS,

p-value < 0.01). The pairwise multiple comparison

procedure yielded a significant difference of p< 0.05

between all weeks’ points except between week 5 and

week 3. CMS rats showed significantly reduced sucrose

consumption, at weeks 2, 4 and 5 (Fig. 5).

Open field test (OFT)

There was an overall decreased activity observed in post

CMS (after CMS regime) rats in the CMS group during the

5 min open field test, however, pre CMS (before CMS

regime) rats in the same group, showed enhanced

locomotor activity during the 1st minute. Furthermore,

there was a tendency of enhanced activity in the center

Fig. 4. Graph depicting body weight variation and linear regression curve fit

mild stress rats.

squares and reduced rearing in the post CMS rats as

compared to the pre CMS rats in CMS group. A paired

t-test demonstrated a significant decrease in grooming

(6.5 ± 0.7 in pre and 3.2 ± 0.6 in post-CMS rats,

p< 0.01), climbing (24 ± 3 in pre and 23 ± 3.3 in post-

CMS rats, p< 0.05), peripheral (67 ± 4.6 in pre and

519 ± 3.5 in post-CMS rats, p< 0.01), 1st minute

activity (22 ± 1.8 in pre and 9 ± 2 in post-CMS rats,

p< 0.01) and overall activity (72.8 ± 5 in pre and

60.2 ± 3 in post-CMS rats, p< 0.04) in post CMS rats

as compared to pre CMS rats. But, there was an

increased central activity (6 ± 0.8 in pre and 12 ± 0.8

in post-CMS rats, p< 0.01) observed in post CMS rats

as compared to pre CMS rats. Graph depicting the

difference in OFT activities observed between pre and

post CMS is shown in Fig. 6.

Forced swim test (FST)

Paired t-test yielded an overall significant difference in the

rats exposed to FST in the post-CMS (after CMS regime)

ting during the chronic mild stress regime in both control and chronic

Fig. 6. Graph showing the difference in open field test activities (A, overall, B, peripheral, C, 1st minute, D, central, E, grooming and F, climbing)

observed between pre and post chronic mild stress rats.

B. S. Hemanth Kumar et al. / Neuroscience 275 (2014) 12–21 17

group as compared to pre-CMS (before CMS regime)

group. In post-CMS rats, there was a significant

decrease observed in swimming (35 ± 0.5 in pre and

26 ± 1.3 in post-CMS rats, p< 0.01) and climbing

activity (7 ± 3.6 in pre and 4 ± 0.5 in post-CMS rats,

p< 0.01). Compared to the pre CMS rats, the post

CMS rats showed an increase in the immobility period

(18.3 ± 0.6 in pre and 28 ± 1.8 in post CMS rats,

p< 0.05). Diving was observed only in few post-CMS

rats. Graph depicting the difference in FST activities

observed between pre and post CMS is shown in Fig. 7.

There were also differences observed in defecation

(number of the fecal boli) between pre and post CMS rats.

Diffusion tensor imaging (DTI)

DTI was used to identify the diffusion changes in different

brain regions. All MR images were reviewed for

detectable lesions or image abnormalities by our

radiologist. No lesions or any obvious differences of

image intensities were observed in the T2-weighted

images in any of the rat brain in both groups. The

results explained the MD, FA, kk and k\ changes

observed in different regions of the brain (Tables 2 and 3).

Mean diffusivity (MD) showed increasing trend in most

of the brain regions of CMS rats as compared to control

rats. However a significant increase in the MD was

observed in bilateral FC, CC, left Hip, right CP and left

HTh. We also observed a significant decrease in

bilateral Cing and right Th region. Fractional anisotropy

(FA) showed a decreasing trend in most of the brain

regions wherein significant differences were observed in

bilateral FC, bilateral HTh, and CC. We also observed a

significant increase in bilateral Th region.

Axial diffusivity (kk) showed not much significant

difference in brain regions except bilateral FC and left

Hip where an increase in kk values was observed in

CMS rats as compared to control. Radial diffusivity (k\)

showed overall increasing trend in which statistical

significance was observed in bilateral FC, bilateral HTh,

CC, left Hip and left CUP.

Correlation studies revealed a significant negative

correlation between 1st minute activity (OFT) and kkvalues of Th region �0.898 (p= 0.015). In addition a

positive correlation was observed between 1st minute

activity and k\ of FC 0.884 (p= 0.047), Significant

negative correlation was observed in central activity and

k\ of Hip region �0.827 (p= 0.084). We did not

observe many differences with other brain regions and

OFT parameters.

DISCUSSION

It is known that chronic stress plays a key role in the

development and acceleration of affective disorders like

depression (Sheline, 2000; Lee et al., 2002). In this

regard, an animal model of CMS-induced depression

has been developed to simulate the pathogenesis of

depression in humans. Several studies suggest that

CMS can induce behavioral and physiological changes

resembling symptoms of clinical depression (Li et al.,

2007; Luo et al., 2008) which can be evaluated through

Fig. 7. Graph showing the difference in forced swim test activities observed between pre and post chronic mild stress.

Table 2. Table showing the values of mean diffusivity (MD) and fractional anisotropy (FA) in different brain regions from control and CMS rats

Controls MD (10�3 mm2/s) FA

Control CMS p-Value Control CMS p-Value

RFC 0.69 ± 0.02 0.78 ± 0.04 <0.001 0.22 ± 0.02 0.20 ± 0.02 0.02

LFC 0.70 ± 0.02 0.81 ± 0.07 0.003 0.22 ± 0.01 0.19 ± 0.02 0.01

CC 0.78 ± 0.04 0.82 ± 0.03 0.041 0.63 ± 0.02 0.57 ± 0.03 <0.001

RHip 0.78 ± 0.03 0.79 ± 0.02 0.58 0.12 ± 0.01 0.12 ± 0.02 0.82

LHip 0.77 ± 0.02 0.79 ± 0.02 0.051 0.12 ± 0.01 0.11 ± 0.02 0.44

RCP 0.74 ± 0.03 0.78 ± 0.02 0.003 0.60 ± 0.03 0.59 ± 0.04 0.67

LCP 0.75 ± 0.005 0.76 ± 0.02 0.27 0.63 ± 0.02 0.61 ± 0.02 0.12

RCing 0.73 ± 0.03 0.71 ± 0.03 0.06 0.34 ± 0.02 0.35 ± 0.02 0.55

LCing 0.72 ± 0.03 0.69 ± 0.03 0.06 0.35 ± 0.02 0.36 ± 0.02 0.37

RHTh 0.75 ± 0.06 0.78 ± 0.04 0.15 0.32 ± 0.05 0.21 ± 0.02 <0.001

LHTh 0.73 ± 0.03 0.82 ± 0.07 0.01 0.32 ± 0.04 0.19 ± 0.02 <0.001

RTh 0.72 ± 0.007 0.69 ± 0.01 0.002 0.23 ± 0.06 0.29 ± 0.03 0.031

LTh 0.71 ± 0.03 0.68 ± 0.03 0.12 0.23 ± 0.05 0.28 ± 0.02 0.034

RSMC 0.72 ± 0.03 0.75 ± 0.05 0.16 0.24 ± 0.01 0.23 ± 0.04 0.30

LSMC 0.72 ± 0.03 0.75 ± 0.04 0.10 0.24 ± 0.02 0.24 ± 0.02 0.40

RCUP 0.71 ± 0.04 0.72 ± 0.04 0.58 0.21 ± 0.02 0.19 ± 0.04 0.13

LCUP 0.70 ± 0.04 0.72 ± 0.03 0.22 0.21 ± 0.04 0.21 ± 0.04 0.68

p-Values mentioned are Bonferroni corrected.

RFC and LFC – right and left frontal cortex; SMC – sensory motor cortex; Hip – hippocampus; CUP – caudate putamen; HTh – hypothalamus; Th – thalamus;

Cing – cingulum; CC – corpus callosum; CP – cerebral peduncle.

18 B. S. Hemanth Kumar et al. / Neuroscience 275 (2014) 12–21

behavioral tests like, FST, OFT and SCT (Papp et al.,

2002; Zhao et al., 2008). The results from the present

study suggest that prolonged exposure to stress might

also have unfavorable effects on brain activity. Using Dif-

fusion imaging, we have documented the changes in the

diffusion pattern using the diffusion indices, in the CMS

rat model of depression.

In the present study, a decrease in sucrose intake and

body weight was observed in CMS rats as compared to

control rats. The sucrose consumption data in the

present study shows a largest reduction of sucrose

consumption in the CMS rats obtained after 3 weeks,

and this trend was attenuated afterward, possibly

because of an adaptation effect to the stressors by the

animals (Katz et al., 1981). In OFT, the activities of a rat

placed in an open field can be taken as an indicator of

its emotional and motivational states (Li et al., 2007). Pre-

vious studies suggest that the effects of CMS on open

field behavior are different and might be influenced by a

number of potentially confounding factors involved (Hu

et al., 2010). An open field apparatus itself can be consid-

ered as a stressor which evokes passive reaction if the

stressor is inescapable, which lead to decreased activity

and immobility (Hu et al., 2010). This could be the possi-

ble reason of increased central activity and immobility

observed in our study. The FST shows the swimming

pattern of the rat in the cylinder which can be related to

the stressed state of the animal. In the present study,

there was an increase in the immobility of the rats in the

CMS group and a decrease in the swimming observed,

this could be because of the anhedonic behavior and

might also be due to the helplessness or the inescapable

environment.

DTI has widely been used in a broad spectrum of CNS

applications, including both neurological and psychiatric

disorders; it has potential to reveal microstructural

abnormalities in gray and white matter tracts. This is

because water diffusion in tissue is highly sensitive to

microstructural architecture (Alexander et al., 2007).

Various DTI metrics can be derived from the imaging data

to provide information about the orientation and architec-

ture of tissue microstructure at the voxel level. Our pres-

ent study demonstrates altered DTI indices (MD, FA, kkand k\) in various brain regions of CMS rats. The

Table 3. Table showing the values of axial (kk) and radial (k\) diffusivity in different brain regions from control and CMS rats

Region kk (10�3 mm2/s) k\ (10�3 mm2/s)

Control CMS p-Value Control CMS p-Value

RFC 0.89 ± 0.02 0.95 ± 0.06 0.02 0.61 ± 0.03 0.69 ± 0.06 0.004

LFC 0.89 ± 0.03 0.95 ± 0.07 0.02 0.62 ± 0.02 0.70 ± 0.06 0.005

CC 1.45 ± 0.11 1.43 ± 0.09 0.63 0.46 ± 0.04 0.52 ± 0.05 0.03

RHip 0.86 ± 0.03 0.87 ± 0.03 0.48 0.72 ± 0.03 0.73 ± 0.03 0.85

LHip 0.84 ± 0.02 0.87 ± 0.02 0.02 0.71 ± 0.03 0.74 ± 0.03 0.07

RCP 1.37 ± 0.08 1.36 ± 0.05 0.85 0.45 ± 0.04 0.47 ± 0.04 0.38

LCP 1.38 ± 0.06 1.35 ± 0.03 0.15 0.45 ± 0.02 0.45 ± 0.022 0.98

RCing 1.00 ± 0.06 0.98 ± 0.06 0.58 0.58 ± 0.04 0.58 ± 0.05 0.84

LCing 0.99 ± 0.05 0.97 ± 0.04 0.42 0.57 ± 0.04 0.57 ± 0.04 0.91

RHTh 0.98 ± 0.09 0.91 ± 0.08 0.12 0.60 ± 0.06 0.68 ± 0.06 0.02

LHTh 0.97 ± 0.07 0.96 ± 0.11 0.91 0.63 ± 0.05 0.72 ± 0.07 0.01

RTh 0.86 ± 0.06 0.89 ± 0.03 0.21 0.61 ± 0.06 0.59 ± 0.03 0.38

LTh 0.86 ± 0.04 0.87 ± 0.05 0.56 0.61 ± 0.05 0.58 ± 0.04 0.22

RSMC 0.99 ± 0.05 0.98 ± 0.05 0.42 0.57 ± 0.04 0.58 ± 0.04 0.90

LSMC 1.37 ± 0.08 1.37 ± 0.05 0.86 0.45 ± 0.05 0.47 ± 0.04 0.39

RCUP 1.38 ± 0.06 1.35 ± 0.03 0.18 0.45 ± 0.02 0.45 ± 0.02 0.99

LCUP 1.46 ± 0.12 1.43 ± 0.10 0.64 0.47 ± 0.04 0.52 ± 0.05 0.03

p-Values mentioned are Bonferroni corrected.

RFC and LFC – right and left frontal cortex; SMC – sensory motor cortex; Hip – hippocampus; CUP – caudate putamen; HTh – hypothalamus; Th – thalamus;

Cing – cingulum; CC – corpus callosum; CP – cerebral peduncle.

B. S. Hemanth Kumar et al. / Neuroscience 275 (2014) 12–21 19

pathophysiology underlying differences in MD and FA in

subjects with neuropsychiatric disorder is yet unclear

and could involve a number of processes, including

altered glial physiology, changes in the density of axons,

axonal diameter, myelination, the coherence of the fiber

tract and localized water content.

In the present study we observed elevated MD

values in the bilateral frontal cortex, hippocampus,

hypothalamus, CP and CC in CMS rats as compared to

control animals. It is well known that the frontal cortex,

hippocampus and hypothalamus are reported to play an

important role in the pathophysiology of depression

(Friedlander and Desrocher, 2006). Increased MD

observed in these regions suggests conditions of reduced

membrane density (Sen and Basser, 2005) such as tissue

degeneration after injury (Concha et al., 2006). The

prefrontal cortex exerts a potent regulatory influence over

the subcortical systems involved in the regulation of

affective states (Grace, 2006; Wager et al., 2008). Fron-

tal-subcortical circuits, formed by connections between

the orbital and medial prefrontal cortex, amygdala, hippo-

campal subiculum, ventromedial striatum, mediodorsal,

and midline thalamic nuclei, and ventral pallidum, are con-

sidered to underlie emotional regulation (Ongur et al.,

2003). These circuits can provide forebrain modulation

over visceral control structures in the hypothalamus and

brainstem, and their dysfunction can regulate the distur-

bances in autonomic regulation and neuro-endocrine

responses that are associated with mood disorders

(Seminowicz et al., 2004; Drevets et al., 2008 Moreover,

in our recent proton magnetic resonance spectroscopy

(MRS) study we found significant lower N-acetylaspartate

(NAA) levels in the prefrontal cortex and hippocampus of

CMS rats (Kumar et al., 2012). Irwan et al., 2005, had

reported that NAA determined by proton MRS and the dif-

fusivity value from DTI had a significant negative correla-

tion in healthy volunteers. Therefore, our results showing

an increased MD in the frontal cortex and hippocampus

are consistent with proton MRS data showing a decrease

in NAA in those regions. Increased diffusivity may also

reflect neuronal arborization and widening of the extracel-

lular space (ECS), which would facilitate extracellular

water diffusion, and neuronal arborization and/or gliosis

are also seen in the affected regions of the brain

(Londono et al., 2003). We also found decreased mean

diffusivity in bilateral cingulum and thalamic region of

CMS rats. The significance of decreased mean diffusivity,

as found here, remains uncertain, it may again reflect the

increased activity and connectivity in this region, which is

thought to be present in patients with obsessive-

compulsive disorder (OCD).

In our study we also observed a decreased FA values

in corpus callosum, bilateral frontal cortex, and bilateral

hypothalamus of CMS rats as compared to control rats.

FA is a widely accepted indicator of white matter

organization within fiber tracts. It is believed that FA

values increase with increasing tissue organization.

Decreased FA has been described in tissues with

demyelination, edema, gliosis and inflammation.

Therefore, decrease in FA might suggest a loss of

‘‘bundle coherence’’ in these regions. A study in

treatment resistive depression and MDD patients using

DTI analysis found reduced FA in the frontal lobes,

cingulate cortex and corpus callosum, which are

believed to play an important role in emotional

regulation (Kieseppa et al., 2010; Peng et al., 2013). In

an earlier study, greater white matter concentration was

found in the anterior genu of the corpus callosum in med-

ication-naive pediatric patients with OCD (MacMaster

et al., 1999). These are consistent with a range of findings

implicating cortico–striato–thalamic pathways in the path-

ophysiology of OCD. Our findings of decreased FA values

in the above mentioned regions suggest demyelination,

reduced neuronal fiber density, neuroinflammation or

directional coherence in these brain regions. Further, an

increasing body of evidence from animal studies has

20 B. S. Hemanth Kumar et al. / Neuroscience 275 (2014) 12–21

shown that kk and k\ derived from DTI hold promise as

biomarkers of axonal damage and demyelination, respec-

tively (Sun et al., 2006). A decrease in kk has shown to be

indicative of axonal damage, whereas an increase in k\has shown to be a noninvasive marker of demyelination

or dysmyelination. Both increase in k\ or decrease in

kk will result in decreased FA. In our study the FA

decrease observed was primarily driven by increase in

k\. This increase in k\ may reflect the degree of

demyelination while increase in kk may represent signs

of axonal loss (Song et al., 2002, 2005), which might be

due to the change in the dominant cell type contributing

to the signal, with axonal bundles being replaced by astro-

cytes and/or microglia, with increased diffusion in all

directions (Newcombe et al., 2011). This further comple-

ments our previous MRS finding where we observed a

decrease in NAA level in CMS rats (Kumar et al., 2012).

Similarly, the destruction of neurofibrils might be the rea-

son for increase in kk in the bilateral frontal cortex and left

hippocampus (Kinoshita et al., 1999). Therefore, it is likely

that changes in kk reflect complex interactions of multiple

biological factors that drive it in different directions.

Indeed, a recent study also had demonstrated that radial

diffusivity is increased and axial diffusivity unchanged in

the corpus callosum of patients with OCD compared with

controls (Bora et al., 2011). This supports our findings that

a possible demyelination process occurs in the commis-

sural fibers in the CMS rats. Among the four DTI indices,

significant differences were much more widespread in MD

and k\, than FA and kk. Significant regions of FA

decrease largely overlapped with that of k\, highlighting

increased k\ as the primary factor driving FA reduction.

However, there were additional significant regions of

increased k\ which was associated with decreased kk,therefore negating their effect on FA. The discrepancy

between our study and that of Delgado y Palacios et al.

(2011) could be attributed to the different types of stress-

ors that were applied. The physiological variability can

also influence the diffusion properties. Therefore, all phys-

iological parameters were kept within a narrow range. It

might also be due to the ROI analysis, which is operant-

dependant and might be influenced by partial volume

effects. Though, in our study, the ROI analysis was con-

ducted with strict criteria to minimize any potential partial

volume effect. In this study, we report successful in vivoapplication of quantitative DTI in different regions of the

CMS rat brain. Our present study reports a decrease in

axial diffusivity and an increase in radial diffusivities (thus

a decreased FA) after CMS regime in rats. However,

multiple factors including inflammation, myelin loss, and

axonal damage likely all contribute to the observed diffu-

sion changes after CMS regime. Overall, our findings indi-

cate that analysis of diffusion indices has the potential to

be a surrogate marker in living organisms for the type and

severity of mood disorders and furthermore, we speculate

that processes of demyelination take place during the

onset of depression.

Limitations of current study

There are some limitations of this study. Firstly, the direct

histological correlation between DTI indices and specific

tissue morphological characteristics was not attempted.

Secondly, the CMS rats were not longitudinally studied

because of the likely adverse effects of repetitive

anesthesia during MRI. Nevertheless, the main focus of

the current study was to investigate the gray and white

matter abnormalities and water diffusivity, as reflected

by FA, MD, kk and k\, using DTI in the CMS animal

model of depression.

Acknowledgments—The authors are grateful for the financial

support from Defence Research and Development Organization

(DRDO), Ministry of Defence, India. This work was performed

as a part of the DRDO sponsored R&D project INM 311(4.1).

REFERENCES

Alexander AL, Lee JE, et al (2007) Diffusion tensor imaging of the

brain. Neurotherapeutics 4(3):318–329.

Anisman H, Zacharko RM (1982) Stimulus change influences escape

performance: deficits induced by uncontrollable stress and by

haloperidol. Pharmacol Biochem Behav 17(2):263–269.

Basser PJ, Pierpaoli C (1996) Microstructural and physiological

features of tissues elucidated by quantitative diffusion-tensor MRI.

J Magn Reson B 111(3):209–219.

Bora E, Harrison BJ, Fornito A (2011) White matter microstructure in

patients with obsessive–compulsive disorder. J Psychiatry

Neurosci 36:42–46.

Brown GW, Harris TO (1993) Aetiology of anxiety and depressive

disorders in an innercity population. 1. Early adversity. Psychol

Med 23:143–154.

Concha L, Gross DW, et al (2006) Diffusion tensor imaging of time-

dependent axonal and myelin degradation after corpus

callosotomy in epilepsy patients. Neuroimage 32:1090–1099.

Davidson RJ, Pizzagalli D, et al (2002) Depression: perspectives from

affective neuroscience. Annu Rev Psychol 53:545–574.

De Lisi LE, Szulc KU, et al (2006) Early detection of schizophrenia by

diffusion weighted imaging. Psychiatry Res 148:61–66.

Delgado y Palacios R, Campo A, et al (2011) Magnetic resonance

imaging and spectroscopy reveal differential hippocampal

changes in anhedonic and resilient subtypes of the chronic mild

stress rat model. AJNR Am J Neuroradiol 70:449–457.

Drevets WC, Price JL, Furey ML (2008) Brain structural and

functional abnormalities in mood disorders: implications for

neurocircuitry models of depression. Brain Struct Funct 213(1–

2):93–118.

Falagas ME, Vardakas KZ, et al (2007) Under-diagnosis of common

chronic diseases: prevalence and impact on human health. Int J

Clin Pract 61:1569–1579.

Friedlander L, Desrocher M (2006) Neuroimaging studies of

obsessive–compulsive disorder in adults and children. Clin

Psychol Rev 26:32–49.

Gonul AS, Kula M, et al (2004) The regional cerebral blood flow

changes in major depressive disorder with and without psychotic

features. Prog Neuropsychopharmacol Biol Psychiatry

28:1015–1021.

Grace AA (2006) Disruption of cortical-limbic interaction as a

substrate for comorbidity. Neurotox Res 10(2):93–101.

Hu H, Su L, et al (2010) Behavioural and [F-18] fluorodeoxyglucose

micro positron emission tomography imaging study in a rat

chronic mild stress model of depression. Neuroscience

169:171–181.

Irwan R, Sijens PE, et al (2005) Correlation of proton MR

spectroscopy and diffusion tensor imaging. Magn Reson

Imaging 23:851–858.

Katz RJ, Roth KA, Carroll BJ (1981) Acute and chronic stress effects

on open-field activity in the rat – implications for a model of

depression. Pharmacol Biochem Behav 5:247–251.

B. S. Hemanth Kumar et al. / Neuroscience 275 (2014) 12–21 21

Kieseppa T, Eerola M, et al (2010) Major depressive disorder and

white matter abnormalities: a diffusion tensor imaging study with

tract-based spatial statistics. J Affect Disorder 120:240–244.

Kim JH, Budde MD, Liang HF, et al (2006) Detecting axon damage in

spinal cord from a mouse model of multiple sclerosis. Neurobiol

Disorder 21:626–632.

Kinoshita Y, Ohnishi A, et al (1999) Apparent diffusion coefficient on

rat brain and nerves intoxicated with methylmercury. Environ Res

80:348–354.

Kumar BSH, Mishra SK, et al (2012) Neurodegenerative evidences

during early onset of depression in CMS rats as detected by

proton magnetic resonance spectroscopy at 7 T. Behav Brain Res

232:53–59.

Lee AL, Ogle WO, Sapolsky RM (2002) Stress and depression:

possible links to neuron death in the hippocampus. Bipolar

Disorder 4:117–128.

Li S, Wang C, et al (2007) Antidepressant like effects of piperine in

chronic mild stress treated mice and its possible mechanisms. Life

Sci 15:1373–1381.

Londono A, Castillo M, et al (2003) Apparent diffusion coefficient

measurements in the hippocampi in patients with temporal lobe

seizures. AJNR Am J Neuroradiol 24:1582–1586.

Luo DD, An SC, Zhang X (2008) Involvement of hippocampal

serotonin and neuropeptide Y in depression induced by chronic

unpredicted mild stress. Brain Res Bull 77:8–12.

MacMaster FP, Keshavan MS, Dick EL (1999) Corpus callosal signal

intensity in treatment-naive pediatric obsessive compulsive

disorders. Prog Neuropsychopharmacol Biol Psychiatry

23:601–612.

Mayberg HS, Lozano AM, et al (2005) Deep brain stimulation for

treatment-resistant depression. Neuron 45:651–660.

Newcombe V, Chatfield D, Outtrim J, Vowler S, et al (2011) Mapping

traumatic axonal injury using diffusion tensor imaging: correlations

with functional outcome. PLoS ONE 6(5):e19214.

Ongur D, Ferry AT, Price JL (2003) Architectonic subdivision of the

human orbital and medial prefrontal cortex. J Comp Neurol

460(3):425–449.

Papp M, Vassout A, Gentsch C (2002) The NK1-receptor antagonist

NKP608 has an antidepressant-like effect in the chronic mild

stress model of depression in rats. Behav Brain Res 115:19–23.

Pause BM, Raack N, et al (2003) Convergent and divergent effects of

odors and emotions in depression. Psychophysiology

40:209–225.

Paxinos G, Watson C (2007) The rat brain in stereotaxic coordinates.

6th ed. San Diego: Elsevier Academic Press.

Peng HJ, Zheng HR, et al (2013) Abnormalities of cortical-limbic-

cerebellar white matter networks may contribute to treatment-

resistant depression: a diffusion tensor imaging study. BMC

Psychiatry 13:72.

Regenold WT, D’Agostino CA, et al (2006) Diffusion weighted

magnetic resonance imaging of white matter in bipolar disorder:

a pilot study. Bipolar Disord 8:188–195.

Saksena S, Rai V, Saraswat VA, et al (2008) Cerebral diffusion tensor

imaging and in vivo proton magnetic resonance spectroscopy in

patients with fulminant hepatic failure. J Gastroenterol Hepatol

23:e111–e119.

Schwartz ED, Chin CL, et al (2005) Apparent diffusion coefficients in

spinal cord transplants and surrounding white matter correlate

with degree of axonal dieback after injury in rats. AJNR Am J

Neuroradiol 26(1):7–18.

Seminowicz D, Mayberg H, et al (2004) Limbic–frontal circuitry in

major depression: a path modelling metanalysis. Neuroimage

22:409–418.

Sen PN, Basser PJ (2005) A model for diffusion in white matter in the

brain. Biophys J 89:2927–2938.

Sheline YI (2000) 3D MRI studies of neuroanatomic changes in

unipolar major depression: the role of stress and medical

comorbidity. Prog Neuropsychopharmacol Biol Psychiatry

48:791–800.

Shin YW, Kwon JS, et al (2006) Increased water diffusivity in the

frontal and temporal cortices of schizophrenic patients.

Neuroimage 30:1285–1291.

Song SK, Sun SW, et al (2002) Dysmyelination revealed through MRI

as increased radial (but unchanged axial) diffusion of water.

Neuroimage 17(3):1429–1436.

Song SK, Sun SW, Ju WK, Lin SJ, Cross AH, Neufeld AH (2003)

Diffusion tensor imaging detects and differentiates axon and

myelin degeneration in mouse optic nerve after retinal ischemia.

Neuroimage 20:1714–1722.

Song SK, Yoshino J, Le TQ, Lin SJ, Sun SW, Cross AH (2005)

Demyelination increases radial diffusivity in corpus callosum of

mouse brain. Neuroimage 26(1):132–140.

Sun SW, Liang HF, et al (2006) Noninvasive detection of cuprizone

induced axonal damage and demyelination in the mouse corpus

callosum. Magn Reson Med 55(2):302–308.

Wager TD, Davidson ML, et al (2008) Prefrontalsubcortical pathways

mediating successful emotion regulation. Neuron

59(6):1037–1050.

Wheeler-Kingshott CAM, Cercignani M (2009) About ‘‘axial’’ and

‘‘radial’’ diffusivities. Magn Reson Med 61(5):1255–1260.

Willner P (1997) Validity, reliability and utility of the chronic mild stress

model of depression: a 10-year review and evaluation.

Psychopharmacology 134:319–329.

Woods RP, Mazziotta JC, Cherry SR (1993) MRI-PET registration

with automated algorithm. J Comput Assist Tomogr 17:536–546.

Zhao Z, Wang W, Guo H, Zhou D (2008) Antidepressant-like effect of

liquiritin from Glycyrrhiza uralensis in chronic variable stress

induced depression model rats. Behav Brain Res 194:108–113.

(Accepted 16 May 2014)(Available online 29 May 2014)