Physiology & Behavior 106 (2012) 185–192

Contents lists available at SciVerse ScienceDirect

Physiology & Behavior

journal homepage: www.elsevier.com/locate/phb

Low-carbohydrate high-fat diets in combination with daily exercise in rats: Effects on

body weight regulation, body composition and exercise capacity
Samantha J. Caton a,⁎, Maximilian Bielohuby a, Yinglong Bai a, Lothar J. Spangler a, Lukas Burget a,
Paul Pfluger c, Claudia Reinel b, Michael Czisch b, Martin Reincke a, Silvana Obici c, Ellen Kienzle d,
Matthias H. Tschöp c, Martin Bidlingmaier a
a
Endocrine Research Unit, Medizinische Klinik und Poliklinik IV, Klinikum der LMU, Munich, Germany
b
Neuroimaging research group, Max Planck Institute of Psychiatry, Munich, Germany
c
Obesity Research Centre, Metabolic Diseases Institute, Department of Internal Medicine, University of Cincinnati, Cincinnati, USA
d
Animal Nutrition and Dietetics, Ludwig Maximilians University Munich, Germany

a r t i c l e i n f o a b s t r a c t

Article history: Background: The aim of the current investigation was to examine the effects of consuming a low-
Received 5 September 2011 carbohydrate high-fat diet (LC-HFD) in combination with daily exercise on body weight, body composition,
Received in revised form 29 January 2012 endocrine control of the energy balance system and exercise capacity in adolescent and mature rats.
Accepted 1 February 2012 Method: Adolescent (n = 23) and mature rats (n = 16) were maintained on either a standard chow diet (CH)
or a LC-HFD for a period of ten days prior to daily exercise training for 21 days in forced running wheel sys-
Keywords:
tem. At the end of the 21 day training sessions all rats took part in an exercise performance test where time to
Low-carbohydrate high-fat diet
Exercise
exhaustion was measured.
Body composition Results: Rats maintained on the LC-HFD demonstrated a significant lack of body weight gain (p b 0.05) com-
pared to CH maintained rats, despite equicaloric intake and performing identical amounts of daily exercise.
Body composition was significantly altered in the LC-HFD rats (p b 0.05) with increased body fat (p b 0.01).
Leptin concentrations were higher (p b 0.05) and IGF-I concentrations were lower (p b 0.01) in the LC-HFD
fed rats. Exercise performance was not diminished in the LC-HFD group despite the higher fat mass. Both
groups irrespective of age performed equally as well in the time to exhaustion test (p > 0.05).
Conclusion: Maintenance on the LC-HFD in combination with forced daily exercise did not impact exercise ca-
pacity (total distance and meters per minute). Additionally consumption of an extreme LC-HFD in combina-
tion with daily exercise resulted in significantly less body weight gain but increased fat mass. When
combined with daily exercise this diet clearly had a negative impact on body composition, but did not affect
exercise capacity.
© 2012 Elsevier Inc. All rights reserved.

1. Introduction a combination of dietary control and increased physical activity is

Given the global obesity epidemic individuals are constantly seek-
ing effective weight management strategies. Nutritional approaches
include low-carbohydrate high-fat diets (LC-HFD), such as the
“Atkins diet”. Proponents of these diets suggest that limiting the in-
take of dietary fat is not the most successful approach to weight loss
and that limiting carbohydrate intake represents a more effective ap-
proach to controlling body weight. Such diets have previously been
demonstrated to be relatively effective at reducing body weight
[1–8]. Weight loss with caloric restriction alone, however, is often ac-
companied with a decline in muscle mass [9–11] and strength [12]
and a concomitant decline in energy expenditure [13,14]. Therefore
⁎ Corresponding author at: Institute of Psychological Sciences, University of Leeds,
Leeds, LS2 9JT, England. Tel.: + 44 113 343 6692.
E-mail address: s.caton@leeds.ac.uk (S.J. Caton).
recommended for sustainable weight management. Exercise not
only increases energy expenditure but also enhances lean body
mass [15]. Despite evidence suggesting that low-carbohydrate diets
are effective strategies for body weight management, no previous in-
vestigations have examined the effect of combining a LC-HFD with
daily “prescribed” exercise.
Low carbohydrate diets are typically high in dietary fat, a factor
that has been previously linked to increased adiposity [16,17]. Diets
high in fat have been reported to elevate somatostatin concentrations
[18], thus inhibiting growth hormone (GH) secretion and IGF-I (Insu-
lin-like growth factor-1) concentrations [19]. GH and IGF-I are impor-
tant for fat metabolism and lean body mass maintenance and
accretion [20]. Several rodent investigations have clearly demonstrat-
ed increased body fat as a consequence of consuming a LC-HFD
[21–23]. More recently we, and others demonstrated an increase in
fat mass in LC-HFD-fed rats and also a reduced activity of the GH/

0031-9384/$ – see front matter © 2012 Elsevier Inc. All rights reserved.
doi:10.1016/j.physbeh.2012.02.003
186 S.J. Caton et al. / Physiology & Behavior 106 (2012) 185–192
IGF-I system [24,25]. Interestingly GH deficient patients demonstrate
increased fat mass, decreased muscle mass [26] and reduced exercise
capacity [27]. In addition several investigations demonstrate that in-
dividuals consuming a LC-HFD report increased fatigue and tiredness
[28,29] as a result of consuming such diets. In line with those results,
Bielohuby et al. [30] recently demonstrated that, in the absence of
forced exercise, rats fed LC-HFD show decreased spontaneous loco-
motor activity. This might be a consequence of reduced muscle glyco-
gen stores due to extremely limited carbohydrate intake or possibly
due to alterations in body composition.
Collectively these strands of evidence suggest that that the effects
of a LC-HFD on body composition may not bode well when coupled
with increased physical activity. It is a possibility that consuming a
LC-HFD might have a detrimental effect on exercise performance
and so rendering LC-HFD in combination with exercise as an unsuita-
ble approach to successful weight management.
Previously we demonstrated that a LC-HFD resulted in body
weight loss in mature heavier animals and lack of body weight gain
in lighter adolescent rats, however, this occurred in the absence of
daily exercise [24]. More importantly, we also observed a significant
increase in fat mass in animals maintained on a LC-HFD. Increased ex-
ercise has been demonstrated to have positive effects on body com-
position [15]. In the current investigation we aim to examine
whether the inclusion of daily exercise is sufficient to offset the in-
crease in body fat gain observed in the absence of daily exercise
[24] as well as examining endocrine controls of energy balance in
both adolescent and mature rats. Current recommendation for body
weight management is a combination of dietary modification with
physical activity; however, increased fatigue has been reported in
adults consuming LC-HFD [28,29]. Due to the extremely low CHO con-
tent of the diet and previously reported effects on body composition
we would expect animals maintained on the LC-HFD to have im-
paired exercise performance. The second aim of the current investiga-
tion is to examine the effects of consuming a LC-HFD on exercise
capacity.
2. Materials and methods
2.1. Animals
In study 1 twenty-three male Wistar rats (Harlan, Borchen, Ger-
many, approximately 9 weeks old) were weight matched and divided
in to two groups, one group was maintained on a low-carbohydrate
high-fat diet (starting weight: LC-HFD; 286.3 ± 8.19 g (mean ± SD)
n = 11) the other group was maintained on standard laboratory
chow (CH; 277.5 ± 9.4 g, n = 12). In study 2, 16 male rats (approxi-
mately 15 weeks) were weight matched and divided into two groups,
one group was maintained on LC-HFD (392.8 ± 4.7 g, n = 8) the other
group was maintained on standard laboratory chow (CH; 391.5 ±
5.2 g, n = 8). For the purpose of the paper the animals will be referred
to as adolescent (9 weeks old) and mature (15 weeks old). Animals
were housed in individual cages (21.8 ± 0.32 °C: humidity 70 ± 1.0)
and maintained on a 12 hour light- dark cycle throughout the study
(Lights on at 02:00 and off at 14:00). All animals received ad libitum
access to standard laboratory chow for the first ten days following de-
livery to allow acclimation to the new environment. All procedures
were approved by Upper Bavarian Government's ethical committee
for animal experiments (AZ 55.2-1-54-2531-47-05).
2.2. Diets
Animals were maintained on either standard laboratory chow
(CH, (% of metabolizable energy, diet specifications as provided by
the manufacturer) 9% fat, 33% protein, 58% CHO, 3.04 kcal/g) and
the high-fat low-carbohydrate diet (LC-HFD, 94% fat, 4.2% protein
and 1.3% CHO, 7.5 kcal/g, Ssniff, Soest, Germany). In the CH diet, the
only source of fat is soy oil (total crude fat content: 3.3%). In the LC-
HFD, the total crude fat content amounts to 79.3%. 70% of this fat de-
rives from pork lard, 9.3% from butter fat.
LC-HFD-fed animals were pair-fed with weight-matched controls
exposed to ad libitum CH. Pair-feeding was carried out by calculating
ad libitum energy intake of the CH-fed group over 24 hours and feed-
ing the equivalent amount of energy to the LC-HFD group over the
following 24 hours. Pair-feeding ensures that any effects observed
are due to the macronutrient composition of the diet and not due to
differences in caloric intake between the two groups. On a daily
basis experimenters confirmed that all food administered to the LC-
HFD group was completely consumed. In the few cases where all
the LC-HDF was not consumed, LC-HFD food left-over was weighed
and replaced with the equivalent weight of fresh food added to
their daily allowance. Over all the LC-HFD-fed rats consumed the
same amount of calories consumed by the CH-fed group. All animals
were given ad libitum access to water throughout the experiments.
2.3. Analysis of the nitrogen balance
In a preceding study in the absence of exercise with 12-weeks old
male Wistar rats (n = 4), we analyzed the nitrogen balance of rats.
Therefore, rats were pair- fed the identical LC-HFD to a control
group that were fed the same chow diet. Daily 24 h nitrogen intake
was calculated based on 24 h food intake (and knowledge of the di-
etary protein contents). To analyze nitrogen excretion, rats were
maintained for 1 week on the diets and then placed on stable metal
grids in cages without bedding. Urine and faeces were collected
twice daily for 7 days and immediately stored at − 20 °C until analy-
sis. Determination of nitrogen in urine and faeces samples was
performed by the Kjeldahl method [31,32]. Total 24 h nitrogen ex-
cretion of each rat was calculated by multiplying the nitrogen con-
centration per gram of urine and faeces with the respective
weights of the 24 h urine and faeces excretion. To obtain a nitrogen
balance, mean 24 h nitrogen excretion was subtracted from mean
24 h nitrogen intake.
2.4. Procedure
Animals were maintained on their respective diets for 10 days
(days 0–9) before the onset of daily training for 21 days (days
10–31) and immediately after training animals were placed into
the metabolic chambers for the assessment of post-exercise energy
expenditure (kcal/kg*hr). In study 1 (adolescent rats) the last
group to be trained on each day was placed into the chamber for 2
hours. Due to limited equipment and the original design of the
study each group of rodents entered the calorimetry system total of
two times. Each day just two animals from each group were placed
in the chambers (n = 6 animals from each diet group every three
days). In study 2 (mature rats), the study design was slightly adapted
and improved. All animals were placed into the system everyday for
two hours immediately following training (n = 8 per diet group each
day). Metabolic assessments were made via indirect calorimetry
(CalsoSys, TSE Systems GmbH, Bad Homburg, Germany). At the end
of the training period all animals took part in a performance test
(time until exhaustion) (days 33–34). Half of the animals took part
in the exhaustion test on day 33, the other half on day 34 and the
groups were randomised to avoid any group effects. Performance
tests were split over two days due to limited equipment and time
constraints. At the end of the experiment (day 35) all animals were
given access to food for 1 h after lights out, they were then fasted
for 6 h and were killed under isoflurane anaesthesia. Trunk blood
was collected for hormone analysis, plasma and serum samples
were then stored at − 80 °C until analysis. In study one the animals
were rapidly frozen in preparation for MRI analysis at a later date.
In study 2, fat pads (subcutaneous, epidydimal and inguinal) were
 S.J. Caton et al. / Physiology & Behavior 106 (2012) 185–192
also collected. The left sides were extracted and weighed immediate-
ly for body fat quantification. Samples (brown adipose tissue (BAT)
and quadriceps femoris muscle) were taken and snap frozen imme-
diately in liquid nitrogen (entire procedure b 2 minutes) before stor-
age at − 80 °C.
2.5. Running wheel exercise
Daily training took part in a custom built forced running wheel
system (TSE Systems GmbH, Bad Homburg, Germany). A motor drives
the wheel and so the time and speed can be accurately controlled.
Every three days the time and/ or the speed of the wheel was in-
creased. Researchers made sure that the animals were able to run at
the speed set before increasing the speed and duration of the exer-
cise. On some occasions the speed of the wheel was increased and
time kept constant, others the time of the exercise was increased
only and some days both were increased according to the abilities
of the animals. In study 1, the starting velocity was 10 meters per mi-
nute for 10 minutes reaching 14 meters per minute for 25 minutes. In
study 2, the starting velocity was 10 meters per minute for 10 mi-
nutes reaching 12 meters per minute for 30 minutes. The speed of
the wheel during the performance test was the maximum speed
achieved during training. The animals were forced to run until they
could no longer keep up with the speed of the running wheel. Ex-
haustion was operationally defined as an inability of the animals to
keep up with the speed of the wheel, involving loss of front and
hind-leg co-ordination with an inability to adjust running style back
to normal. The test was terminated when the animal displayed 5
uncontrolled rotations in the wheel within 3 minutes. Two experi-
menters rated exhaustion to further increase validity of the test.
2.6. Body composition (MRI)
Body composition measurements in study 1 (adolescent rats)
were measured via magnetic resonance imaging (MRI). The animals
were measured ex-vivo on a 7 T BRUKER Biospec system (Ettlingen,
Germany), equipped with a volume resonator to allow for whole-
body imaging of the rat. The animals were positioned in a supine po-
sition, fixated with tape. A T1-weighted sequence was run (2D spin
echo, TR = 900 ms , TE = 11.6 ms, matrix size 512 x 256, slice thick-
ness = 1 mm, 34 slices, in plane field of view 4.6 cm x 4.6 cm, 8 aver-
ages). Using this sequence, MR-images provide a strong contrast
between fat and muscular/organ tissue. In addition, free liquids like
urine also display a low intensity and can therefore be clearly differ-
entiated from fat. To achieve full body coverage, two slice packages
were scanned subsequently after repositioning of the animal inside
the resonator, covering the complete abdomen from tail to liver.
Anatomical images were bias corrected using SPM2 (www.fil.ion.
ucl.ac.uk/spm) and further analyzed using the Paravision software
(BRUKER, Ettlingen, Germany). For each slice, regions of interests
were manually outlined including intra-abdominal tissue, and ex-
cluding subcutaneous, cutaneous and skeletal muscular compart-
ments. For each slice, an intensity histogram was created and the
average intensity of the non-fatty tissue was determined. This value
was multiplied by 1.8 to result in a conservative cutoff value to sepa-
rate fat and non-fat compartments. Visceral and subcutaneous fat
compartments were identified by summing all image pixels in all
slices, excluding non-fat intra-abdominal volumes using threshold-
ing. Total body volume was determined on the same slices without
thresholding.
2.7. Quantitative real time PCR
Quadriceps muscle and BAT tissue was isolated and powdered
under liquid nitrogen. Quadriceps RNA was extracted using Trizol
(Invitrogen,Carlsbad, CA) and a modified RNEasy Mini Kit protocol
187
(Qiagen, Valencia, CA). RNA from brown adipose tissue was extracted
by using the Rneasy lipid tissue mini kit (Biorad, Hercules, CA),
according to the manufacturers instructions. Total RNA integrity
was checked by agarose gel electrophoresis, and quantified using
the NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies,
Wilmington, DE). Four micrograms of total RNA from muscle sample
and 2 μg of RNA from BAT sample were subsequently reversely tran-
scribed to cDNA via oligo(DT) priming using the SuperScript III First
Strand Synthesis System for RT-PCR (Invitrogen,Carlsbad, CA) accord-
ing to the manufacturer's instructions.
Quantitative real-time PCR (qPCR) was performed for various
genes of interest and reference genes (β-actin; Acidic ribosomal
phosphoprotein PO, 36B4-Arbp, HPRT) on an iCycler iQ Real Time
PCR detection system using iQ Sybr Green Supermix (Biorad, Hercu-
les, CA) in 20 ul final reactions with 0.2uM of each primer. RT(−)
negative control reactions were also run to check for potential geno-
mic DNA contamination using both reference gene primers. No tem-
plate controls (NTC) using DEPC water in place of cDNA template
was also run on each plate with all samples. All primers were
designed for qPCR and to anneal only to either rat or rat/mouse
mRNA. Primer sequences utilized for each gene of interest, and the
reference genes, are listed in Table 1.
2.8. Assays
All hormones were measured using commercially available kits.
Serum was tested for, insulin (Sensitive Rat Insulin Ria Kit, LINCO,
Missouri, USA), leptin (ACTIVE® Murine Leptin ELISA, DSL, Texas,
USA), IGF-I (Rat/Mouse IGF-I, IDS, Boldon, UK). Albumin, total protein,
cholesterol and triacylglycerides (COBAS INTEGRA 800, Roche Diag-
nostics GmbH, Mannheim, Germany) were also measured.
2.9. Statistical analysis
Data are presented as means ± SEM and all analysis was per-
formed on SPSS for Windows (v14.0; SPSS Inc, Chicago, IL). Indepen-
dent group t-tests were used to examine differences in body weight
gain, body composition, hormonal parameters, gene expression and
performance test results in each investigation. Paired samples t-test
were used to examine ad libitum 24 hour caloric intake pre-exercise
and during exercise. Independent t-tests were carried out at the dif-
ferent time points to investigate post exercise energy expenditure in
study 1 due to the design of the investigation where different animals
were measured at different time points. ANOVA was also used to ex-
amine post-exercise energy expenditure in study 2, with day of train-
ing and diet as factors. The alpha value chosen was 0.05.
3. Results
3.1. Body weight and 24 hour caloric intake
Body weight development (Fig. 1: Upper panel) was examined to
investigate the effects of maintenance on a LC-HDF in combination
Table 1
Primer sequences used in study 2.
Gene Forward Primer Reverse Primer
Cox-II ACACACAAGCACAATAGACGCCC AATTCGTAGGGAGGGAAGGGC
UCP-1 GGGCCCTTGTAAACAACAAA GTCGGTCCTTCCTTGGTGTA
UCP-2 CATGGTTGGTTTCAAGGCCA GACGGAGGCAAAGCTCATC
UCP-3 GTGGATGTGGTAAAGACCCG CGCAGTACCTGGACTTTCAT
CPT1A TGGGAGAGAATTTCATCCACTTCC TGGCTTGTCTCAAGTGCTTCCC
CPT1B AACCACATCCGTCAAGCAC CCTTATGCCTGTGAACTGGC
HPRT TGCTCGAGATGTCATGAAGG TATGTCCCCCGTTGACTGAT
β-Actin CTGCCGCATCCTCTTCCTCC GATGCCACAGGATTCCATACCC
36B4 ATCCCTGACGCACCGCCGTG GCGCATCATGGTGTTCTTGC
188 S.J. Caton et al. / Physiology & Behavior 106 (2012) 185–192

Diet only Daily exercise P/T

feeding this reduction in energy intake was also applied to the LC-
460 HFD-fed groups. In adolescent rats, food intake was 81.4 ± 2.3 kcal
450
440 before training yet during training 24 hour food intake declined by
430
around 16% to 63.3 ± 1.8 kcal (p b 0.01). This was also confirmed in
body weight (grams)

420
410
400
390
mature rats, pre exercise ad libitum food intake was significantly
380
370 higher, 80.9 ± 1.7 kcal in comparison to ad libitum food intake during
360
350 daily training, 67.6 ± 3 kcal (p b 0.01) with a reduction of around 22%.
340
330
320
310
300 3.2. Nitrogen balance
290
280
270
260
250
Mean 24 h nitrogen intake in the LC-HFD group was significantly
lower when compared to chow (CH: 0.8 ± 0.0 g vs. LC-HFD: 0.1 ±
adolescent LC-HFD (study one) adolescent CH (study one) 0.0 g, p b 0.001). With respect to nitrogen excretion, mean 24 h nitrogen
mature LC-HFD (study two) mature CH (study two) excretion in faeces (CH: 0.3 ± 0.1 g vs. LC-HFD: 0.0139 ± 0.006 g,
p b 0.001) and urine (CH: 0.1 ± 0.1 g vs. LC-HFD: 0.03 ± 0.01 g,
140 ** p = 0.07) was lower in rats fed LC-HFD. Calculation of the 24 h nitrogen
balance has shown that rats fed both diets maintained a positive nitro-
120 gen balance, indicating that they were not catabolic (CH: 0.5 ± 0.1 g vs.
Delta body weight (grams)

LC-HFD: 0.1 ± 0.01 g, p b 0.001).

100

80 * 3.3. Body composition

60
40
20
0 were significantly increased in the LC-HFD animals (Fig. 3: Middle
adolescent (study one)
LC-HFD
•p < 0.05
Fig. 1. Body weight development in adolescent (study 1, n = 23) and mature rats
(study 2, n = 16) maintained via a pair-feeding schedule on either a LC-HFD or stan-
dard CH (pair-feeding reference group). Lower panel: Body weight change (g) from
the start to the end of the study in adolescent rats (study1) and mature rats (study
2). Data are shown as means ± SEM.
with daily exercise in comparison to maintenance on CH in combina-
tion with daily exercise. In both adolescent and mature rats consum-
ing the LC-HFD resulted in significantly less body weight gain (Fig. 1:
Lower panel). Adolescent rats maintained on the LC-HFD gained only
56% of the body weight gained by the CH-fed rats (p b 0.01). Similarly,
mature LC-HFD rats gained only 47.6% of the body weight gained by
mature rats maintained on CH (p b 0.05).
24 hour caloric intake (average 24 hour intake over the whole pre-
intervention and exercising periods) was reduced in ad libitum CH-
fed animals during exercise in comparison to ad libitum 24 hours ca-
loric intake before the exercise intervention (Fig. 2). Because of pair
Results from the MRI scans (see Fig. 3: Upper panel) reveal that,
whilst the animals maintained on the LC-HFD gained less body
weight over the entire experimental period, they did however, have
increased body fat relative to total body volume (23.8 ± 1.4% versus
15.6 ± 0.7%, p b 0.001). Both the visceral and the subcutaneous depots
mature (study two)
panel), (subcutaneous: p b 0.001, visceral: p b 0.01). Body volume did
CH not differ significantly between the two groups (p > 0.05). Fat pad ex-
** p < 0.01 traction in mature rats (Fig. 3: Lower panel) confirmed the findings of
study 1 (adolescent rats). Despite lower body weights, animals main-
tained on a LC-HFD were significantly fatter than those maintained on
CH. Fat pad weight expressed as a percentage of total body weight
was significantly higher (71.7%) in the LC-HFD animals in comparison
to the CH fed controls (p b 0.01). Inguinal fat (% body weight) was sig-
nificantly higher (64.5%) in the LC-HFD group (p b 0.01). Similarly,
epidydimal fat weight (% body weight) was higher (78.2%) in the
LC-HFD group (p b 0.05).
3.4. Exercise time until exhaustion
Consuming a LC-HFD did not impair exercise performance since
mean exercise time until exhaustion did not differ between groups.
Adolescent LC-HFD-fed animals ran for 117.8 ± 15.2 minutes and
the CH group for 129.5 ± 12.1 minutes (p > 0.05). In study 2, the ma-
ture LC-HFD animals ran for 131.6 ± 31.9 minutes and the CH animals
for 131.7 ± 25.9 minutes (p > 0.05).

3.5. Post exercise energy expenditure

Daily exercise

Energy expenditure (Kcal/kg*hr) was examined over three days

chow intake (kcal)

100
95 corresponding to changes in the daily exercise regime (Fig. 4: Upper
90
85 panel, adolescent rats (study 1)) Consuming a LC-HFD resulted
80
75 in an approximately 14% higher post-exercise energy expenditure fol-
70 lowing 14 days exposure to the diet (p b 0.05) compared with CH-fed
65
60 controls. These findings were confirmed in mature rats as post exercise
55
50 energy expenditure was elevated by 13.3% in the LC-HFD animals.
Given the slightly different design of study 2 (mature rats), ANOVA
(repeated measures) was performed on the data. A main effect of
days diet was found, whereby the rats maintained on the LC-HFD had ele-
adolescent rats (study one) mature rats (study two)
vated post exercise energy expenditure (Fig. 4: Lower panel, main ef-
Fig. 2. Twenty-four hour caloric intake in adolescent (study 1, n = 12) and mature rats
fect of diet, p b 0.05). No main effect of training day or diet x training
(study 2, n = 8) maintained on ad libitum chow (pair-feeding reference groups). Data day interactions was observed (p> 0.05). Baseline EE expenditure did
are shown as means ± SEM. not differ significantly between groups in either study (p> 0.05).
 S.J. Caton et al. / Physiology & Behavior 106 (2012) 185–192 189

A. LC-HFD B. CH
250000.00

200000.00
volume (mm3)

150000.00

100000.00
p<0.01
p<0.001
50000.00 p<0.01

0.00
subcu visc total fat total body
volume

3
fat (% of terminal body weight)

2.5 p<0.01

1.5 p<0.05
p<0.01
1

0.5

0
Epididymal Inguinal Total fat
LC-HFD CH

Fig. 3. Upper panel: Representative MRI scans of adolescent rats (study 1). A: LC-HFD-fed rat B: CH-fed rat. Middle panel: fat compartments (mm3) and total body volume in LC-
HFD-fed rats (n = 11) and CH-fed adolescent rats (n = 12) (study 1). Lower panel: fat pad weights (% terminal body weight) extracted from mature rats (study 2) maintained on
either LC-HFD (n = 8) or CH (n = 8). Data are shown as means ± SEM.

3.6. Effects on plasma lipids and endocrine parameters 3.7. Gene analysis: Quantitative real time PCR

Maintenance on a LC-HFD resulted in elevated plasma triacylgly-
ceride (TAG) concentrations in both adolescent (p b 0.01) and mature
rats (p b 0.05) whilst cholesterol was significantly decreased in the
LC-HFD groups (adolescent p b 0.01 and mature p b 0.05). Insulin
was consistently lower in the animals maintained on the LC-HFD (ad-
olescent p b 0.01and mature p b 0.05). As expected, due to increased
body fat, significantly higher leptin concentrations were found in an-
imals maintained on the LC-HFD (adolescent p b 0.05 and mature
p b 0.05). IGF-I was also significantly lower in the LC-HFD group (ado-
lescent p b 0.01 and mature p b 0.01). Plasma albumin was significant-
ly elevated in both groups of rats maintained on the LC-HFD
(adolescent p b 0.01and mature p b 0.01), whilst total plasma protein
did not differ between groups (adolescent p > 0.05 and mature
p > 0.05). (Table 2)
UCP-1 mRNA expression in brown adipose tissue did not differ be-
tween groups (LC-HFD 1 ± 0.47 and CH 1.1 ± 0.29) (p > 0.05). Fur-
thermore, in muscle tissue of rats fed either chow or LC-HFD, no
significant differences (p > 0.05) were found in mRNA expression of
COX II, UCP-2, UCP-3, CPT-1A, CPT 1B (data not shown).
4. Discussion
In the current investigations we demonstrate that maintenance on
a LC-HFD in combination with daily exercise in both mature and ado-
lescent rats results in lower terminal body weights and less body
weight gain in comparison to rats maintained on a standard chow
diet despite identical energetic intakes. However, despite this effect
on body weight, LC-HFD rats had significantly elevated fat mass,
190 S.J. Caton et al. / Physiology & Behavior 106 (2012) 185–192

* * * * * * between the findings could be due to a number of possible reasons:

9
composition of the diets, especially different protein to energy ratios
EE (kcal/(kg*h)

8
and the sources of macronutrients used, differences in determination
7 of metabolizable energy contents, mode of exercise (forced exercise
versus voluntary) or due to differences in sensitivity of methods
6
used to measure body fat. In the current study and others [22,24,35]
5 normal weight animals were used to investigate the impact of con-
suming LC-HFD on body weight. Given the established effects of LC-
4
d11-13 d14-16 d17-19 d20-22 d23-25 d26-28 d29-31 HFD on body weight in normal weight animals, future work should
days on respective diet investigate these effects in already obese animals.
We have previously shown that feeding rats with a ketogenic LC-
LC-HFD CH
HFD results in lower IGF-I concentrations [24,25]. In this study, we
9 confirmed the lower IGF-I levels in rats fed the LC-HFD. It is known
* that exercise, muscle strength and walking speed are correlated
EE (Kcal/(kg*h)

8 with IGF-I concentrations [36] and that physical activity may increase
7 circulating levels of IGF-I [37]. Interestingly, we have now shown that
6 not rescue the low IGF-I levels. Whilst GH concentrations were not
5
4 low IGF-I. However, we have recently demonstrated that a ketogenic
LC-HFD CH
* p < 0.05
Fig. 4. Post exercise energy expenditure (kcal/kg*hr) measured for 2 hours following
exercise. Upper panel: Adolescent rats (study 1) showing elevated post exercise energy
expenditure following 13 days exposure to the diet. Data are presented as Means ±
SEM over every three days, corresponding to increased in the speed and/ time of ex-
ercise training. Lower panel shows a main effect of diet on post exercise EE measured
for two hours in mature rats (study 2) (main effect of diet only). Data are shown as
means ± SEM.
suggesting that LC-HFD may not be a favourable strategy for weight
management. Interestingly, the LC-HFD in combination with daily ex-
ercise did not appear, in this instance to affect exercise capacity since
LC-HFD-fed rats performed similarly to their CH-fed counterparts in
the performance test (time to exhaustion) at the end of 21 days
training.
In both adolescent and mature male Wistar rats, consumption of
the LC-HFD in combination with daily exercise resulted in less body
weight gain and ultimately in lower terminal body weights in com-
parison to the CH-fed control groups. Taken alone these results dem-
onstrate that LC-HFD do have a significant impact upon body weight,
which has previously been demonstrated in both rodents [21] and
humans [1,2,4,8,33]. However, despite the lower body weight, the
LC-HFD rats had increased body fat [22,23,34]. Increased body fat
was demonstrated via MRI, fat pad extraction and was confirmed, as
expected with elevated leptin concentrations in the LC-HFD-fed rats.
Strikingly, forced daily exercise did not lead to reductions of body
fat. Our findings concur with those by Lapachet et al. [21] who report
that a high-fat diet in combination with daily exercise did not prevent
higher fat accumulation. More recently however, Kinzig et al. [35]
suggested that when rats are given access to running wheels, in-
creased fat deposition is prevented in rats maintained on a LC-HFD.
Yet, this was not the case in the current investigation. The differences
feeding of a LC-HFD in combination with daily physical exercise does
measured in the current investigations, GH and IGF-I are tightly reg-
ulated [38], and physiologically GH levels increase in response to
LC-HFD which was similar to the LC-HFD used in the current study,
impaired the central GH/IGF-I feedback mechanism, meaning that
rats fed this ketogenic diet failed to increase GH levels in response
to low IGF-I [39].
The effect of the LC-HFD on the GH/IGF system is two fold; first,
impaired growth hormone (GH) secretion has implications for lipoly-
sis. Secondly, the effects of reduced IGF-I are extremely profound in
terms of lean body mass (LBM) and overall metabolic health. GH con-
tributes to the accretion and/or maintenance of LBM by stimulating
IGF-I production in muscle tissue, thus preventing proteolysis and
possibly by increasing protein synthesis [40]. Interestingly, results of
the MRI scans demonstrate that not only did the rats maintained on
the LC-HFD have increased adiposity but also that they did not differ
in total body volume to the CH-fed rats, thus demonstrating a loss of
LBM. Whilst the longer-term effects of a LC-HFD on body composition
remain to be fully explored several pieces of human data also suggest
that LBM may be reduced as a function of consuming such diets
[1,41,42]. Part of the weight reducing properties of LC-HFD has been
attributed to the decline in insulin concentrations, thus increasing li-
polysis [33]. However, we demonstrate a decline in fasting insulin
concentrations yet increased fat mass. The current findings do not
support the concept from proponents of LC-HFD's that carbohydrate
ingestion (leading to elevated glucose and insulin) is the sole deter-
minant responsible for fat deposition.
This LC-HFD diet was also used in a previous investigation [24]
and it has much lower protein content than what would be usually
consumed by humans adhering to low-carbohydrate diets. However
the protein content of the current diet contains adequate protein and
amino acid profile to ensure normal growth [43]. Furthermore, anal-
ysis of the daily nitrogen intake and excretion has shown that also
rats fed this low protein LC-HFD in the absence of daily exercise
maintain a positive nitrogen balance. We therefore suggest that
loss of body weight with this LC-HFD is likely to be independent of
catabolic processes due to lack of dietary protein. Whilst dietary

Table 2
Circulating concentrations of insulin, leptin, IGF-I, cholesterol, TAG, albumin and total protein in adolescent (A) (n = 23) and mature rats (M) (n = 16) maintained on either a LC-
HFD or CH. Statistical analysis comparing LC-HFD or CH fed rats was carried out separately for adolescent and mature rats (* p b 0.05 ** b 0.01).

TAG (mg/dl) Cholesterol (mg/dl) Insulin (ng/ml) Leptin (ng/ml) IGF-I (ng/ml) Albumin (g/dl) Total protein (g/dl)

CH (A) 83.5 ± 4.4 96.3 ± 3.6 0.6 ± 0.1 2.5 ± 0.2 1663.7 ± 76.8 3.5 ± 0.1 6.3 ± 0.1
LC-HFD 224.1 ± 20.1** 80.6 ± 3.8** 0.3 ± 0.1** 4.4 ± 0.7* 1220.3 ± 100** 4 ± 0.1** 6.2 ± 0.8
(A)
CH (M) 68.6 ± 34.5 96.4 ± 5.8 0.6 ± 0.17 0.8 ± 0.1 1410.9 ± 53.5 3.6 ± 0.2 6.3 ± 0.1
LC-HFD (M) 154.9 ± 9.7* 79.4 ± 8.8* 0.1 ± 0.1* 1.5 ± 0.3* 953.4 ± 56.5** 4.0 ± 0.1 ** 6.1 ± 0.1
 S.J. Caton et al. / Physiology & Behavior 106 (2012) 185–192
protein was low in the LC-HFD we have demonstrated that it was
sufficient to avoid endogenous protein breakdown. However, it is
important to consider that these animals were not exercised and so
this warrants consideration. Furthermore, Bielohuby et al. [39]
reported lower IGF-1 concentrations in unexercised rats consuming
the lower-protein version of a LC-HFD. This highlights the impor-
tance of dietary protein in sustaining IGF-1 concentrations, therefore
the lower IGF-1 values and decreased LBM might be a product of the
lower protein content of the LC-HFD despite the fact that the diet is
adequate in protein content. Additionally it is important to consider
that since energy intake during the exercise period was reduced,
protein intake was further reduced in the LC-HFD group. In order to
further investigate the effect of the current diet in combination
with daily exercise future studies might benefit from providing ad
libitum access to the LC-HFD.
Very few investigations have measured the effects of both the en-
ergetic cost of exercise and changes in body composition during
maintenance on a LC-HFD. We observed an increase in absolute post
exercise energy expenditure in the LC-HFD rats, despite the animals
being exposed to exactly the same exercise program. In the current
studies, we found no difference in BAT UCP1, or COX II, CPT-2, CPT-
3, UCP-2 and UCP-3 mRNA expression in the quadriceps muscle. Ad-
ditionally, it is also a possibility that physical activity was increased
in the LC-HFD rats when they were placed in the metabolic chambers
following exercise, however, this is unlikely, since locomotor activity
is reported to be lower in animals maintained on LC-HFD's [30]. One
important consideration of these finding are that we did not normal-
ise energy expenditure to lean body mass because body composition
was not assessed until the end of the investigation. Ideally this should
have been carried out given the profound differences observed in
body composition between the groups.
Human studies have suggested that low-carbohydrate diets have a
negative impact on exercise performance [28,29] due to increased
perceived exhaustion. It was therefore expected that rats maintained
in the LC-HFD would not be able to perform as well on the perfor-
mance test (time until exhaustion). Supporting clinical evidence for
this assumption is derived from GH deficient patients, who have a
similar body composition (increased fat mass and decreased LBM)
to the rats maintained on this diet. GH deficient patients demonstrate
decreased exercise capacity [27]. Surprisingly, rats in the LC-HFD
groups did not demonstrate impaired ability to perform exercise, de-
spite extremely low CHO intake and apparent reduced lean body
mass. However, it is unclear from the current finding whether this re-
sult would persist under the conditions of further elevated fat mass
and further reductions to LBM.
In summary here we demonstrate that maintenance on LC-HFD in
combination with daily exercise results in a lack of body weight gain
in comparison to rats maintained on standard CH in a pair feeding
setting. This effect was independently observed in both adolescent
and mature rats. Body composition was significantly altered by the
LC-HFD, in that increased fat deposition was consistently observed
in both investigations via independent techniques and this was fur-
ther exemplified by elevated leptin concentrations. IGF-I concentra-
tions were lower in the LC-HFD animals thus contributing to our
understanding of the effect of the diet composition on body composi-
tion with particular focus on LBM.
5. Conclusion
Overall it appears that whilst LC-HFD results in less body weight,
increased fat deposition with possible reduction to LBM are the di-
rect consequences of consuming such diets. We demonstrated in
both investigations that combining the LC-HFD with daily physical
activity is not sufficient to offset increased fat deposition in the LC-
HFD groups. Furthermore, exercise capacity was not affected by the
191
LC-HFD in combination with daily exercise, a finding worthy of fu-
ture investigations.
References
[1] Brehm BJ, Seeley RJ, Daniels SR, D'Alessio DA. A randomized trial comparing a very
low carbohydrate diet and a calorie-restricted low fat diet on body weight and
cardiovascular risk factors in healthy women. J Clin Endocrinol Metab 2003;88:
1617–23.
[2] Brehm BJ, Spang SE, Lattin BL, Seeley RJ, Daniels SR, D'Alessio DA. The role of en-
ergy expenditure in the differential weight loss in obese women on low-fat and
low-carbohydrate diets. J Clin Endocrinol Metab 2005;90:1475–82.
[3] Foster GD, Wyatt HR, Hill JO, McGuckin BG, Brill C, Mohammed BS, et al. A ran-
domized trial of a low-carbohydrate diet for obesity. N Engl J Med 2003;22:
2082–90.
[4] Samaha FF, Iqbal N, Seshadri P, Chicano KL, Daily DA, McGrory J, et al. A low-
carbohydrate as compared with a low-fat diet in severe obesity. N Engl J Med
2003;348:2074–81.
[5] Sharman MJ, Gómez AL, Kraemer WJ, Volek JS. Very low-carbohydrate and low-fat
diets affect fasting lipids and postprandial lipemia differently in overweight men.
J Nutr 2004;134:880–5.
[6] Sondike SB, Copperman N, Jacobson MS. Effects of a low-carbohydrate diet on
weight loss and cardiovascular risk factor in overweight adolescents. J Pediatr
2003;142:253–8.
[7] Volek JS, Sharman MJ, Love DM, Avery NG, Gómez AL, Scheett TP, et al. Body com-
position and hormonal responses to a carbohydrate-restricted diet. Metabolism
2002;51:864–70.
[8] Yancy Jr WS, Olsen MK, Guyton JR, Bakst RP, Westman EC. A low-carbohydrate,
ketogenic diet versus a low-fat diet to treat obesity and hyperlipidemia: a ran-
domized, controlled trial. Ann Intern Med 2004;140:769–77.
[9] Janssen I, Ross R. Effects of sex on the change in visceral, subcutaneous adipose
tissue and skeletal muscle in response to weight loss. Int J Obes Relat Metab Dis-
ord 1999;23:1035–46.
[10] Janssen I, Fortier A, Hudson R, Ross R. Effects of an energy-restrictive diet with or
without exercise on abdominal fat, intermuscular fat, and metabolic risk factors in
obese women. Diabetes Care 2002;25:431–8.
[11] Johnson MJ, Friedl KE, Frykman PN, Moore RJ. Loss of muscle mass is poorly
reflected in grip strength performance in healthy young men. Med Sci Sports
Exerc 1994;26:235–40.
[12] Nindl BC, Friedl KE, Frykman PN, Marchitelli LJ, Shippee RL, Patton JF. Physical per-
formance and metabolic recovery among lean, healthy men following a prolonged
energy deficit. Int J Sports Med 1997;18:317–24.
[13] Dulloo AG, Calokatisa R. Adaptation to low calorie intake in obese mice: contribu-
tion of a metabolic component to diminished energy expenditures during and
after weight loss. Int J Obes 1991;15:7–16.
[14] Ravussin E, Burnand B, Schutz Y, Jequier E. Energy expenditure before and during
energy restriction in obese patients. Am J Clin Nutr 1984;41:753–9.
[15] Hill JO, Wyatt HR. Role of physical activity in preventing and treating obesity. J
Appl Physiol 2005;99:765–70.
[16] Macdiarmid JI, Cade JE, Blundell JE. High and low fat consumers, their macronutri-
ent intake and body mass index: further analysis of the National Diet and Nutri-
tion Survey of British Adults. Eur J Clin Nutr 1996;50:505–12.
[17] Goris AH, Westerterp KR. Physical activity, fat intake and body fat. Physiol Behav
2008;23:164–8.
[18] Penman E, Wass JA, Medbak S, Morgan L, Lewis JM, Besser GM, et al. Response of
circulating immunoreactive somatostatin to nutritional stimuli in normal sub-
jects. Gastroenterology 1981;81:692–9.
[19] Cappon JP, Ipp E, Brasel JA, Cooper DM. Acute effects of high fat and high glucose
meals on the growth hormone response to exercise. J Clin Endocrinol Metab
1993;76:1418–22.
[20] Poehlman ET, Copeland KC. Influence of physical activity on insulin-like growth
factor-I in healthy younger and older men. J Clin Endocrinol Metab 1990;71:
1468–73.
[21] Lapachet RA, Miller WC, Arnall DA. Body fat and exercise endurance in trained
rats adapted to a high-fat and/or high-carbohydrate diet. J Appl Physiol
1996;80:1173–9.
[22] Kinzig KP, Scott KA, Hyun J, Bi S, Moran TH. Altered hypothalamic signaling and
responses to food deprivation in rats fed a low-carbohydrate diet. Obes Res
2005;13:1672–82.
[23] McNeel RL, Mersmann HJ. Low- and high-carbohydrate diets: body composition
differences in rats. Obes Res 2005;13:1651–60.
[24] Caton SJ, Yinglong B, Burget L, Spangler LJ, Tschöp MH, Bidlingmaier M. Low-
carbohydrate high-fat diets: regulation of energy balance and body weight
regain in rats. Obesity 2009;17:83–9.
[25] Bielohuby M, Matsuura M, Herbach N, Kienzle E, Slawik M, Hoeflich A, et al. Short-
term exposure to low-carbohydrate, high-fat diets induces low bone mineral den-
sity and reduces bone formation in rats. J Bone Miner Res 2010;25:275–84.
[26] Mukherjee A, Murray RD, Shalet SM. Impact of growth hormone status on body
composition and the skeleton. Horm Res 2004;62:35–41.
[27] Hoffman AR. Treatment of the adult growth hormone deficiency syndrome: direc-
tions for future research. Growth Horm IGF Res 2005;15:48–52.
[28] Burke LM, Hawley JA, Angus DJ, Cox GR, Clark SA, Cummings NK, et al. Adapta-
tions to short-term high-fat diet persist during exercise despite high carbohy-
drate availability. Med Sci Sports Exerc 2002;34:83–91.
192 S.J. Caton et al. / Physiology & Behavior 106 (2012) 185–192
[29] Phinney SD. Ketogenic diets and physical performance. Nutr Metab 2004;17:2.
[30] Bielohuby M, Menhofer D, Kirchner H, Stoehr BJ, Müller TD, Stock P, et al. Induc-
tion of ketosis in rats fed low-carbohydrate, high-fat diets depends on the relative
abundance of dietary fat and protein. Am J Physiol Endocrinol Metab 2011;300:
65–76.
[31] Vickery HB. The early years of the Kjeldahl method to determine nitrogen. Yale J
Biol Med 1946;18:473–516.
[32] Lynch JM, Barbano DM. Kjeldahl nitrogen analysis as a reference method for pro-
tein determination in dairy products. J AOAC Int 1999;82(6):1389–98.
[33] Volek JS, Westman EC. Very-low-carbohydrate weight-loss diets revisited. Cleve
Clin J Med 2002;69:856–8.
[34] Axen KV, Axen K. Very low-carbohydrate versus isocaloric high-carbohydrate diet
in dietary obese rats. Obesity 2006;14:1344–52.
[35] Kinzig KP, Taylor RJ. Maintenance on a ketogenic diet: voluntary exercise, adipos-
ity and neuroendocrine effects. Int J Obes 2009;33:824–30.
[36] Gläser S, Friedrich N, Ewert R, Schäper C, Krebs A, Dörr M, et al. Association of cir-
culating IGF-I and IGFBP-3 concentrations and exercise capacity in healthy volun-
teers: results of the Study of Health in Pomerania. Growth Horm IGF Res 2010;20:
404–10.
[37] Nemet D, Eliakim A. Growth hormone-insulin-like growth factor-1 and inflamma-
tory response to a single exercise bout in children and adolescents. Med Sport Sci
2010;55:141–55.
[38] Moller N, Jorgensen JO. Effects of growth hormone on glucose, lipid, and protein
metabolism in human subjects. Endocr Rev 2009;30:152–77.
[39] Bielohuby M, Sawitzky M, Stoehr BJ, Stock P, Menhofer D, Ebensing S, et al. Lack of
dietary carbohydrates induces hepatic growth hormone (GH) resistance in rats.
Endocrinology 2011;152:1948–60.
[40] Rooyackers OE, Nair KS. Hormonal regulation of human muscle protein metabo-
lism. Annu Rev Nutr 1997;17:457–85.
[41] Meckling KA, O'Sullivan C, Saari D. Comparison of a low-fat diet to a low-
carbohydrate diet on weight loss, body composition, and risk factors for diabetes
and cardiovascular disease in free-living, overweight men and women. J Clin
Endocrinol Metab 2004;89:2717–23.
[42] Noakes M, Foster PR, Keogh JB, James AP, Mamo JC, Clifton PM. Comparison of isoca-
loric very low carbohydrate/high saturated fat and high carbohydrate/low saturated
fat diets on body composition and cardiovascular risk. Nutr Metab 2006;3:7–19.
[43] National Research Council. Nutritional Requirements of Laboratory Animals. 4th
ed. Washington DC: National Academy Press; 1995.