J. Biomedical Science and Engineering, 2011, 4, 684-691
doi:10.4236/jbise.2011.411085 Published Online November 2011 (http://www.SciRP.org/journal/jbise/
JBiSE
).
Published Online November 2011 in SciRes. http://www.scirp.org/journal/JBiSE
Compounds that inhibit triglyceride accumulation and TNFα
secretion in adipocytes
Naofumi Shiomi, Miwako Maeda, Minori Mimura
School of Human Sciences, Kobe College, Nishinomiya, Japan.
Received 19 September 2011; revised 17 October 2011; accepted 30 October 2011.
ABSTRACT
Obese subjects show both a fatty predisposition and a
higher risk of metabolic syndrome. This characteris-
tic depends on adipocytes. However, the roles of adi-
pocytes in metabolic syndrome have been insuffi-
ciently investigated, because few pure adipocyte cell
lines have been isolated. The present study had two
objectives: the isolation of a pure adipocyte cell line
and clarification of the differences between adipo-
cytes and preadipocytes, and screening for com-
pounds that can potentially prevent metabolic syn-
drome. We isolated a novel adipocyte cell line, 3T3-
L24. In the 3T3-L24 cells, the gene expression levels
(of C/EBPα and β, PPARγ and AP2) and the produc-
tion of triglyceride and TNFα were much higher than
those in the preadipocyte 3T3-L1 cells. We used the
3T3-L24 adipocytes to screen for compounds that
could inhibit triglyceride accumulation and TNFα
secretion. Fatty acids enhanced the triglyceride ac-
cumulation. Sodium carboxylate, taurine and car-
nitine not only inhibited triglyceride accumulation,
but also inhibited TNFα secretion. Therefore, these
compounds might be effective to decrease the risk of
metabolic syndrome in obese subjects.
Keywords: Adipocyte; Metabolic Syndrome; Triglyc-
eride; TNFα; Fatty Acid; Carboxylate; Taurine; Carnitine
1. INTRODUCTION
Obesity has been gradually increasing in Japan because
of the changes in eating habits, shortage of exercise and
increase in work-related stress [1]. The risk for hyper-
glycemia, diabetes, and heart disease in obese people is
approximately 2 - 4 times higher than that in normal-
weight persons. This trio of effects is well known as
“metabolic syndrome”. According to research performed
by the Ministry of Health, Labor and Welfare of Japan in
2006, half of males and one-fifth of females among
those over 40 years old show signs of metabolic syn-
drome, making the prevention (and treatment) of meta-
bolic syndrome an important subject in Japan.
The mechanism responsible for metabolic syndrome
has been studied since the 1990s. White adipose tissue
(WAT) works as an endocrine organ, and provides en-
ergy storage by the accumulation of triglycerides [2].
WAT secretes many adipocytokines [3-7], such as adi-
ponectin, leptin and tumor necrosis factor alpha (TNFα),
which can induce insulin resistance and decrease glucose
control, which are among the first symptoms of meta-
bolic syndrome. For example, a standard-weight person
generally has 5 - 10 μg/ml of adiponectin in the blood,
but the concentration in patients with type 2 diabetes or
ischemic heart disease is much lower [8]. The admini-
stration of adiponectin to these patients was effective for
reducing their symptoms. Leptin, a neuropeptide, works
as an inhibitor of appetite [6]. The administration of
leptin to diabetic patients led to recovery from insulin
resistance [9]. TNFα induces the expression of migration
inhibitory factor (MIF), which is also related to insulin
resistance [10]. Consequently, insulin resistance and the
unusual metabolism of glucose associated with meta-
bolic syndrome are caused by the induction of TNFα and
the repression of leptin and adiponectin in the WAT.
Recent studies have also suggested that hypertrophy
of the WAT is an important factor in metabolic syn-
drome. Hypertrophy of the WAT causes not only the
aging of adipocytes, but also inflammation by the oblit-
eration of the blood flow. This inflammation facilitates
the gathering of macrophage and secretion of adipocyto-
kines from the disrupted cells of the WAT [11]. In ex-
periments using rats, an increase in the expression of p53
and secretion of proinflammatory factors was found in
hypertrophic WAT [12].
The ratio of adipocytes to preadipocytes is very high
in obese in comparison with normal-weight people.
Therefore, to avoid the development of metabolic syn-
drome in obese subjects, it is important to fully under-

N. Shiomi et al. / J. Biomedical Science and Engineering 4 (2011) 684-691 685
stand the characteristics of adipocytes. However, no pure
adipocyte cell line has been isolated, and therefore, the
characteristics of adipocytes have not yet been fully elu-
cidated. In this study, we isolated a novel adipocyte cell
line, which we confirmed based on its characteristics of
triglyceride accumulation and TNFα secretion. We then
used this cell line to screen for potential inhibitors of
these functions. Our results may be useful for the pre-
venting the development of metabolic syndrome in
obese subjects.
2. MATERIALS AND METHODS
2.1. Cell Line and Culture Condition
The mouse preadipocyte line, 3T3-L1, which was origin-
nally generated from the Swiss albino mouse fibroblast
cell line, 3T3, by Green [13] was obtained from Dainip-
pon Sumitomo Pharma Corporation (Osaka, Japan). Dul-
becco’s modified Eagle’s medium (DMEM) containing
sodium bicarbonate (MP Biomedicals Inc., IIIkirch,
France) and fetal bovine serum (FBS), which was treated
at 56˚C for 30 min, were mixed at a ratio 9:1, and the
mixture (DMEM medium containing 10% FBS) was
used as a basic culture medium. The cells were cultured
in a CO2 incubator at 37˚C.
2.2. Isolation of Adipocytes
The following differentiation procedure was carried out
to obtain pure adipocytes based on a revision of a
method described in a previous study [14]. The 3T3-L1
preadipocytes (2.0 × 105 cells) were cultured in 15 ml of
DMEM medium containing 10% FBS for 2 days in 75
cm2 flasks, then the medium was replaced with DMEM
containing 10% FBS, 0.5 µM of 3-iisobutyl-1-methylx-
anthine (IBMX) and 4 µM of dexamethasone. The cells
were cultured for 2 additional days to differentiate 3T3-
L1 cells into adipocytes. The medium was replaced
again with DMEM containing 10% FBS and 10 µg/mL
of human insulin, and the cells were cultured for 2 addi-
tional days to mature the cells.
After three repeated differentiation steps to enrich the
ratio of adipocytes to preadipocytes, the cells were sus-
pended and diluted in DMEM containing 10% FBS. The
diluted cells were cultured to construct single colonies in
8 ml medium in 10 cm2 plates for 7 days. A stainless cup
(1 cm2 diameter) was used to covered each colony, then
the cells were removed from the culture using 0.1 µl of
0.25% trypsin, collected in a 1.5 ml microtube, and
washed with a DMEM medium containing 10% FBS.
The cells collected from each colony were cultured in 5
ml of medium in 2.5 cm2 plates for 7 days. Finally, a cell
line which contained more oil drops in comparison with
the preadipocyte 3T3-L1 line was isolated as a novel
adipocyte cell line, and was named 3T3-L24.
2.3. The Production of Triglycerides and TNFα
in Preadipocytes and the Newly-Isolated
Adipocytes
The amounts of triglyceride and TNFα produced by the
preadipocytes and adipocytes were examined. The prea-
dipocyte 3T3-L1 cell line and the adipocyte 3T3-L24
cell line (5 × 105 cells) were cultured in two 75 cm2
flasks containing 15 ml of DMEM medium for one day,
and 0.28 mM oleic acid was added to one of the flasks.
After another 3 days of culture, the culture broth was
removed for the analysis. The triglyceride level in the
cells was found to have gradually increased during the
cultivation, and the culture conditions (the lot number of
cells, pre-culture conditions, number of inoculated cells
and culture time) also led to differences in the triglyc-
eride level (data not shown). Thus, we elected to deter-
mine the triglyceride content when 90% - 95% of the
glucose in the medium was consumed. To measure the
concentration, the cells were washed once with PBS
solution containing 0.02 mM of EDTA, detached with
0.25% trypsin solution, and harvested. The concentra-
tions of protein, triglycerides and TNFα in cells were
assayed using the cells. A total of 3 - 4 independent ex-
periments were performed, and the average values,
standard error (S.E.) and p-value (by the t-test) were
calculated.
2.4. The Effects of Additives on the Triglyceride
and TNFα Production in Adipocytes
The 3T3-L24 adipocytes (5 × 105 cells), which were
pre-cultured in 8 ml of DMEM containing 10% FBS,
were cultured in several 75 cm2 flasks for one day, and
then were exposed to 20 mM glycerol, 0.28 mM long-
chain fatty acids (steric, oleic and linoleic acids), 0.5
mM of carboxylates (sodium acetate, sodium citrate,
sodium malate and sodium oleate), 0.5 mM of carnitine,
0.5 mM of taurine). The effective concentration of the
additive was determined by preliminary experiments
(data not shown). After another 3 days of culture (90% -
100% of the glucose was consumed), the culture broth
was removed, the cells were washed with 0.02 mM of
EDTA/PBS solution, detached with 0.25% trypsin solu-
tion, and harvested. The concentration of protein and
triglyceride in the cells and the concentration of TNFα in
the medium were analyzed. Three independent experi-
ments were performed, and the average, S.E. and p-val-
ues (by t-test) were calculated.
2.5. Assays
The triglyceride concentration in the cells was analyzed
by the following method developed in our laboratory:
C
opyright © 2011 SciRes. JBiSE

N. Shiomi et al. / J. Biomedical Science and Engineering 4 (2011) 684-691 686
the cells cultured in a 75 cm2 flask were collected in a
1.5 ml of microtube. The cell pellet was suspended in
100 µl of phosphate buffered saline (PBS) and disrupted
by using an ultrasonic disrupter (TOMY SEIKO Co. Ltd.,
Tokyo, Japan) for 30 s under a strength of 8. Then, 5 µl
of the solution was used for the analysis of total protein,
and the rest of the volume was used for the analysis of
the triglyceride concentration. Diethyl ether (400 µl) was
mixed with the rest of the solution and centrifuged at
14,000 ×g for 10 min. The upper layer (diethyl ether
containing oil) was transferred to a new 1.5 ml of mi-
crotube, and the diethyl ether was allowed to evaporate
at room temperature for 12 h. The oil drop that remained
after drying was dissolved in 10 µl isopropanol, and the
total amount of triglyceride was measured by the
Triglyceride E Test Wako kit (Wako Chem. Co. Ltd.,
Osaka, Japan). Using this method, the glycerol and lipids
initially contained in the cells were removed during the
extraction operation, and they did not affect the meas-
urement of the triglyceride. The total triglyceride level
could be measured to a concentration as low as 0.01
mg/ml.
The concentrations of TNFα were analyzed as follows:
the cells cultured in 75 cm2 flasks were collected in 1.5
ml microtubes, and the cell pellets were suspended in
200 µl of PBS and disrupted by using an ultrasonic dis-
rupter. Next, 5 µl of the solution was used for the analy-
sis of total protein, and the rest was used for the analysis
of TNFα. The amounts of TNFα present in the cells and
culture medium were determined using a highly sensi-
tive ELISA [15].
The concentration of total protein was determined by
a protein assay kit (Bio-Rad Laboratories, Inc., Tokyo,
Japan). The static error in cell number counted with a
cell counter was much higher than that of the total
amount of protein present in adipocytes. Therefore, we
utilized the value based on the total amount of protein
instead of the cell number for comparison.
2.6. Expression of mRNAs
The 3T3-L1 preadipocytes and the 3T3-L24 adipocytes
(5 × 105 cells) were cultured in 8 ml of DMEM contain-
ing 10% FBS in 25 cm2 flasks for 4 days. The preadipo-
cytes were also cultured in DMEM containing 10% FBS
and 4 µM dexamethasone and 0.5 µM 3-iisobutyl-1-
methyl xanthine (IBMX), and 10 μg/mL of human insu-
lin for 4 days. We named the differentiated cells 3T3-L2,
and the expression of genes in the 3T3-L1, 3T3-L2 and
3T3-L24 cell lines was compared by real-time PCR.
The total mRNA in the cells was purified, and the
cDNAs were synthesized by using an RNeasy Lipid
Tissue Mini kit and a QuantiTeck Reverse Transcription
kit. The reaction mixture for real-time PCR was pre-
pared with the Rotor-Gene SYBR Green PCR Kit. Quan-
Tech Primer Assays [Mm Aclb 2 SG (QT01136772),
Mm_Fabp4_1_SG (QT00091532), Mm_Cebpa_1_SG
(QT00311731), Mm_Cebpb_1_SG (QT00320313), Mm_
Ppag_1_SG (QT00100296)] were used as the primers
for β-actin, AP2, CCAAT/enhancer binding protein
(C/EBP)α, C/EBPβ, and PPARγ. These kits and primers
were obtained from Qiagen K. K. (Tokyo, Japan). Real-
time PCR was performed by the using Rotor-GeneTM
device (Qiagen K. K.). The reaction was performed for
70 cycles of treatment at 95˚C for 5 s and 65˚C for 10 s.
The threshold line and Ct values were determined by
using the Rotor-Gene 6000 series software program, and
the relative amount of mRNA in 3T3-L24 adipocytes or
the differentiated 3T3-L2 cells in comparison with that
in preadipocyte 3T3-L1 (control) was determined by the
ΔΔCt method, followed by calculating the ΔCt values by
using β-actin as a house-keeping gene. Three independ-
ent experiments were performed, and the average values,
S.E. and p-vales were calculated.
3. RESULTS
3.1. A Novel Adipocyte Cell Line, and Its
Expression of Signaling Genes
We tried to isolate a novel adipocyte cell line because a
suitable adipocyte line had not yet been isolated from
3T3-L1 preadipocytes. As shown in the Materials and
Methods section, dexamethasone, IBMX and insulin
were used to induce the differentiation of the mouse
preadipocytes into adipocytes. Following three round of
differentiation to enrich the ratio of adipocytes, colonies
were formed on the dishes. A cell line which contained
more oil drops in comparison with the 3T3-L1 preadi-
pocytes was finally isolated as a novel adipocyte cell
line, and was named 3T3-L24 (Figure 1(a) ).
Increased expression of the AP2, C/EBPα, C/EBPβ
and PPARγ genes are known to indicate the differentia-
tion to adipocytes [16-19]. Thus, we compared the ex-
pression levels of these genes in the 3T3-L24 cells with
those in the parental 3T3-L1 cells or in the differentiated
(a) (b)
Figure 1. Photographs of the 3T3-L24 adipocytes cultured
without (a) and with 0.28 mM oleic acid (b).
C
opyright © 2011 SciRes. JBiSE

N. Shiomi et al. / J. Biomedical Science and Engineering 4 (2011) 684-691 687
3.3. Compounds That Increase the Triglyceride
Level in the 3T3-L24 Adipocytes
3T3-L2 cells by using a real-time PCR method. The
3T3-L1 and 3T3-L24 cells were cultured in DMEM
containing 10% FBS, and the differentiated 3T3-L2 cells
were cultured in the same basic medium containing
dexamethasone, IBMX and insulin. Figure 2 shows the
ratios of mRNA expression in the 3T3-L24 adipocytes or
the differentiated 3T3-L2 cells to those of 3T3-L1
preadipocytes. As shown in Figure 2(a), the expression
of the AP2 and C/EBPβ genes in the 3T3-L2 cells were
1.5 times higher than those in the 3T3-L1 cells, but the
expression levels of C/EBPα and PPARγ were lower.
These results suggest that the differentiation in the
3T3-L2 cells was not complete. However, in the 3T3-
L24 adipocytes, the gene expression levels of AP2,
C/EBPα, C/EBPβ and PPARγ were 8.8, 23.2, 1.6 and 5.0
times higher than those in the 3T3-L1 cells (p < 0.01),
even though the 3T3-L24 cells were cultured without the
addition of dexamethasone, IBMX and insulin.
Triglyceride is synthesized from glycerol and long-chain
fatty acids. Thus, we investigated the effects of the addi-
tion of glycerol and long-chain fatty acids on the
triglyceride accumulation in the 3T3-L24 cells. Figure 3
shows the amount of triglyceride present in the cells
cultured in DMEM containing 10% FBS with the addi-
tion of glycerol or long-chain fatty acids. A small in-
crease in triglyceride accumulation was found when 22
mM of glycerol was added. On the other hand, many oil
droplets (Figure 1(b)) appeared when the cells were
cultured with 0.28 mM of long-chain fatty acids. The
average values of triglyceride in cultures with the addi-
tion of oleic acid and linolenic acid were 0.51 and 0.29
mg/mg protein, respectively (Figure 3(a)). These values
3.2. The Production of Triglycerides and TNFα
by the 3T3-L24 and 3T3-L1 Cells
We compared the amount of triglyceride present in
3T3-L24 adipocytes with that in 3T3-L1 preadipocytes.
Figure 3(a) shows the amount of triglyceride present in
cells cultured for 72 hr in the DMEM containing 10%
FBS. These average values in 3T3-L24 and 3T3-L1 cells
were 0.087 and 0.031 mg triglyceride/mg protein, re-
spectively. The production of TNFα was also compared.
Figure 3(b) shows the amount of TNFα in the cells cul-
tured under the same conditions as in Figure 3(a). The
average value of TNFα contained in the 3T3-L24 cells
was 0.68 ng/mg protein, but the value in the 3T3-L1
cells was only 0.007 ng/mg protein (the value of TNFα
in the medium was not detectable).
(a) (b)
Figure 2. The ratio of the mRNA expression in the 3T3-L2
differentiated cells (a) or 3T3-L24 adipocytes (b) per 3T3-L1
preadipocyte. The 3T3-L1 and 3T3-L24 cells were cultured in
DMEM containing 10% FBS (the basic medium) for 4 days,
and the 3T3-L2 cells were cultured in the basic medium con-
taining 4 µM dexamethasone, 0.5 µM IBMX, and 10 μg/mL of
human insulin. Bars, means ± S.E. of three independent experi-
ments. *p < 0.05 vs. control (3T3-L1), **p < 0.01.
(a) (b)
Figure 3. The amounts of triglyceride (a) and TNFα (b) present in the 3T3-L1 preadipocytes and
3T3-L24 adipocytes. The 3T3-L1 cells were cultured in the basic media without any additional
factors (control) or with 0.28 mM of oleic acid. The 3T3-L24 cells were cultured in the basic me-
dia without additional factors (control) or with 0.28 mM of oleic acid, sodium oleate, or linolenic
acid and 22 mM of glycerol. The 3T3-L1 and 3T3-L24 cells were cultured without additional
factors to examine the amount of TNFα present in the cells. Bars, means ± S.E. of the 3 inde-
pendent experiments. *p < 0.05 vs. control (without addition) and ** p < 0.01.
JBiSE

N. Shiomi et al. / J. Biomedical Science and Engineering 4 (2011) 684-691 688
were much higher than the value without the addition of
either of these factors (0.086 mg/mg protein). Further,
the amount of triglyceride accumulation in cells sub-
jected to culture media with the addition of sodium ole-
ate was only 32% in comparison with the addition of
oleic acid. Even in the 3T3-L1 preadipocytes, the amount
of triglyceride was also enhanced from 0.031 to 0.090
mg/mg protein by the addition of 0.28 mM oleic acid
(Figure 3(a)), but the value was much lower than that in
the 3T3-L24 adipocytes (0.51 mg triglyceride/mg pro-
tein).
3.4. Compounds That Inhibit Triglyceride and
TNFα Production in Adipocytes
We examined whether the triglyceride accumulation
could be inhibited by using sodium acetate, citrate, car-
nitine and taurine. As shown in Figure 4(a), the average
triglyceride present in cells treated with the addition of
sodium acetate, citrate, carnitine and taurine were de-
creased to 57%, 39%, 79% and 26% in comparison of the
cells treated without these factors (control) (all p < 0.01).
Additionally, the cells were treated with these agents in
combination with sodium oleate, which was used as a
representative free fatty acid, and the effects of these
compounds were examined again. As shown in Figure
4(b), the average triglyceride present in the cells treated
with sodium citrate, carnitine and taurine were decreased
to 44%, 33%, and 46% (all p < 0.01) of the control value,
even in the presence of sodium oleate. Therefore, the
additions of these compounds were effective to inhibit
the triglyceride accumulation.
The effects of triglyceride accumulation on TNFα se-
cretion were also examined. Figure 5 shows the amount
of TNFα present in the medium of cells cultured under
the same conditions (except acetate) as in Figure 4. The
concentration of TNFα in cells cultured with sodium
oleate was decreased to 68% in comparison with those
without (control). Moreover, when sodium citrate, car-
nitine and taurine were added, the TNFα secretion was
decreased to 90%, 38% and 43% in cells cultured with-
out sodium oleate and 67%, 9.4% and 33% in cases
where cells were also cultured with sodium oleate. Of
interest, the values after treatment with carnitine and
taurine were significantly different from the control (p <
0.01), although the value for sodium citrate was not sig-
nificantly different from the control (unless the cells
were cultured with sodium oleate).
Finally, the relationship between the triglyceride ac-
cumulation and TNFα expression was analyzed by cal-
culating the correlation coefficient for the covariance.
Figure 6 shows the covariance between the two factors
as determined by using the average values in Figures 4
and 5. The correlation coefficient (r) was 0.63, and the
value was increased to 0.87 when the values for sodium
citrate were excluded. Therefore, there was a positive
correlation between these two factors.
4. DISCUSSION
The current study had two objectives: to isolate a novel
adipocyte cell line and clarify the differences in the na-
tive characteristics between adipocytes and preadipo-
cytes, and to use the new-isolated adipocytes to screen
for compounds that can affect the triglyceride and TNFα
levels in the cells.
Under the first objective, we isolated a novel adipo-
cyte cell line by using the 3T3-L1 mouse preadipocytes.
(a) (b)
Figure 4. The effects of sodium carboxylates, carnitine and taurine on the triglyceride
level in 3T3-L24 cells. The 3T3-L24 cells were cultured in the basic media without addi-
tional factors (a) or with 0.28 mM sodium oleate; (b) A total of 0.5 mM of sodium acetate,
sodium citrate, carnitine and taurine were respectively added in these media to examine
their effects. The cells cultured in the media in (a) and (b) were used as control. Bars,
means ± S.E. of 3 independent experiments. *p < 0.05 vs. control (without addition) and
**p < 0.01.
C
opyright © 2011 SciRes. JBiSE

N. Shiomi et al. / J. Biomedical Science and Engineering 4 (2011) 684-691 689
(a) (b)
Figure 5. The effects of sodium carboxylates, carnitine and taurine on the TNFα
secretion by adipocytes. The 3T3-L24 cells were cultured under the same condi-
tions as in Figure 4. Bars, means ± S.E. of 3 independent experiments. **p <
0.05 and *p < 0.01 vs. control (without addition).
Figure 6. The relationship between the
triglyceride level and TNFα secretion in
the 3T3-L24 adipocytes. The average val-
ues of triglycerides and TNFα secretion in
Figures 4 and 5 were divided by these
control values for normalization (control:
1).
It is generally thought that the ratio of adipocytes to
preadipocytes is very high in obese subjects in compare-
son with normal-weight subjects. Therefore, a source of
adipocytes is needed to better understand metabolic syn-
drome. Many researchers [2-3] have used the preadipo-
cyte 3T3-L1 cells as a useful model for adipocytes, be-
cause they can be easily differentiated to 3T3-L2 adipo-
cyte-like cells by the addition of induction compounds,
such as dexamethasone, IBMX and a PPARγ antagonist
[15]. However, as shown in Figure 2(a), the differenti-
ated 3T3-L2 cells did not show the full characteristics of
adipocyte, because the cells not only still contained
many undifferentiated 3T3-L1 preadipocytes, but also
had a low ratio of adipocytes to preadipocytes, and the
ratio was not constant. Indeed, the newly-isolated
3T3-L24 cells, which were screened in this study, con-
tained few preadipocytes, making them a more appropri-
ate model for detailed in vitro research on adipocytes.
We compared the genetic and biochemical character-
istics between the 3T3-L24 and parental 3T3-L1 cells. In
the genetic analysis, the expression of the AP2, C/EBPα,
C/EBPβ and PPARγ genes were examined. These genes
were previously suggested to play important roles in the
differentiation of preadipocytes to adipocytes [16-19],
and the genes were not expressed or only weekly ex-
pressed in the 3T3-L1 preadipocytes as determined using
a RT-PCR method [14]. As shown in Figure 2(b), all
genes were highly expressed in the 3T3-L24 cells in the
present study, and the expression was maintained in the
absence of IBMX and other factors. Further, in a bio-
chemical examination, 3T3-L24 adipocytes showed a
higher capacity for triglyceride production than the
3T3-L1 preadipocytes, as shown in Figure 3(a). These
results are in agreement with the reports that obese indi-
viduals have a high ratio of adipocytes to preadipocytes
and have gained a predisposition toward obesity [20].
Under the second objective, we screened various
compounds for the ability to affect triglyceride produc-
tion and TNFα secretion. Hypertrophic WAT in an obese
subject causes inflammation, aging and obliteration of
the bloodflow, and finally leads to the secretion of an
undesirable amount of adipocytokines. Therefore, the
inhibition of hypertrophy and adipocytokine secretion
would be helpful to prevent the development of (or treat
existing) metabolic syndrome.
First, we examined the effects of glycerol and long-
chain fatty acids, the components of triglycerides. As
shown in Figure 3(a), the triglyceride levels were en-
hanced upon the addition of long chain fatty acids, but
were not enhanced by the addition of glycerol. Interest-
ingly, the amount of triglyceride in response to the addi-
tion of sodium oleate was lower than that in response to
the addition of oleic acid. These results suggest the fol-
lowing: there is already a sufficient amount of glycerol
in cells to synthesize triglyceride by lipoprotein lipase,
but long chain fatty acids are a limiting factor. The syn-
thesis rate by lipoprotein lipase is a rate-limiting reaction
when fatty acids are added to the culture, and the car-
C
opyright © 2011 SciRes. JBiSE

N. Shiomi et al. / J. Biomedical Science and Engineering 4 (2011) 684-691 690
boxyl group is the key factor involved in the synthesis.
Therefore, this suggests that some carboxylates might
act as a competitor of fatty acids in the reaction of lipo-
protein lipase, and a decrease in fatty acids by degrada-
tion might decrease the triglyceride level.
To address these possibilities, the effects of some
carboxylates were examined in the 3T3-L24 adipocytes
which were cultured in medium with or without sodium
oleate. As shown in Figure 4, the addition of a carboxy-
late, such as sodium acetate or citrate, inhibited the
triglyceride accumulation. Further, we examined the
effects of addition of carnitine and taurine. Carnitine
facilitates the transport of a fatty acid to an inner mem-
brane of the mitochondria, and taurine consumes the
acetyl CoA degraded from a fatty acid to synthesize
cholesterol. These molecules are generally used as die-
tary supplements and were previously pointed out to
have a relationship with metabolic syndrome [21-23]. In
adipocytes cultured in media containing additional car-
nitine or taurine, triglyceride accumulation was inhibited
(Figure 4(a)). Therefore, the enhancement of the degra-
dation of fatty acid metabolism is also effective to de-
crease the triglyceride accumulation.
Finally, we also examined whether the expression of
adipocytokines could be decreased by a decrease in
triglyceride accumulation. TNFα, one of the main factors
involved in type2 diabetes [9], was the focus of this
study. As shown in Figures 3(b), the amount of TNFα
produced and secreted by the adipocyte was much higher
than the levels of the preadipocytes. In fact, TNFα was
produced only in the adipocytes. This result suggests
that the enhancement of TNFα in an obese subject might
be caused by the increase in the ratio of adipocytes to
preadipocytes.
We also examined whether sodium citrate, carnitine
and taurine could decrease the TNFα secretion by 3T3-
L24 adipocytes under the same conditions. As shown in
Figure 5, TNFα secretion (as well as triglyceride accu-
mulation) could be decreased by the addition of sodium
citrate, carnitine and taurine. The levels were especially
decreased following the addition of carnitine or taurine.
Further, the relationship between triglyceride accumula-
tion and TNFα secretion in the adipocytes was analyzed
(Figure 6), and the results suggested that the amount of
triglyceride was related to the secretion of TNFα, be-
cause the correlation factor was high (0.87) when cells
were treated with carnitine or taurine.
Obese individuals have increased concentrations of
free fatty acids in their blood, which likely affects the
development and progression of metabolic syndrome
[24,25]. As shown in Figures 4(b) and 5(b), we exam-
ined the effects of free fatty acids by using sodium oleate
as a model compound. Although it was reported that
fatty acids regulate the production of TNFα and IL-10 in
3T3-L1 cells [26], the addition of sodium oleate affected
led to an increase in triglyceride accumulation but no
significant change in TNFα secretion. However, as shown
in Figures 4(b) and 5(b), even in cells cultured with so-
dium oleate, the triglyceride accumulation and TNFα
secretion could be inhibited by the addition of sodium
citrate, carnitine and taurine. These results suggest that
obese patients might be able to decrease the level of
triglycerides in the WAT and the serum TNFα level by
increasing their intake of carboxylates, including car-
nitine and/or taurine. However, obese subjects have in-
crease not only in free fatty acids, but also in insulin,
cholesterol and other adipocytokines. We did not exam-
ine the effects of all of these using the 3T3-24 adipo-
cytes. Further studies are currently underway, and the
results will be described in a subsequent manuscript.
5. ACKNOWLEDGEMENTS
Our group greatly thanks Ms. S. Aida, A. Fujiwara, K. Imanaka, Y.
Kanda, S. Nakayama, H. Shiotsu, A. Sogo, M. Tanaka, Y. Tanaka, M.
Yamashita and S. Sunami for their assistance with our research. This
work was supported by grant-in aids from the School of Human Sci-
ences and the laboratory at Kobe College.
REFERENCES
[1] Egusa, G., Murakami, F., Ito, C., Matsumoto, Y., Kado, S.,
Okamura, M., Mori, H., Yamane, K., Hara, H. and Ya-
makido, M. (1993) Westernized food habits and concen-
trations of serum lipids in the Japanese. A therosclerosis,
100, 249-255. doi:10.1016/0021-9150(93)90211-C
[2] Ahima, R.S. (2006) Adipose tissue as an endocrine organ.
Obesity, 14, 242-249. doi:10.1038/oby.2006.317
[3] Rosen, E.D. and Spiegelman, B.M. (2000) Molecular
regulation of adipogenesis. Annual Review of Cell Biol-
ogy, 16, 145- 171. doi:10.1146/annurev.cellbio.16.1.145
[4] Matuzawa, Y. (2006) The metabolic syndrome and adi-
pocytokines. FEBS Letters, 580, 2917-292.
doi:10.1016/j.febslet.206.04.028
[5] Maeda, K., Okubo, K., Shimomura, I., Funahashi T. and
Matsuzawa, K. (1996) cDNA cloning and expression of
novel adipose specific collagen-like factor apM1. Bio-
chemical and Biophysical Research Communications,
221, 286-289. doi:10.1006/bbrc.1996.0587
[6] Szkudelski, T. (2006) Intercellular mediators in regula-
tion of leptin secretion from adipocytes. Physiological
Research, 56, 503-512.
[7] Hotamisligi, G.S., Peraldi, P., Budavari, A., Ellis, R.,
White, M.F. and Spiegelman, B.M. (1996) IRS-1-medi-
ated inhibition of insulin receptor tyrosine kinase activity
in TNF-alpha- and obesity induced insulin resistance.
Science, 271, 665-668.
doi:10.1126/science.271.5249.665
[8] Yamaguchi, T., Kamo, J., Waki, H., Terauchi, Y., Kubota,
N., Hara, K., Mori, Y., Ide, T., Murakami, K., Tsuboyama-
Kasaoka, N., Ezaki, O., Akanuma, Y., Gavrilova, O.,
Vinson, C., Reitman, M.L., Kagechika, H., Shudo, K.,
C
opyright © 2011 SciRes. JBiSE

N. Shiomi et al. / J. Biomedical Science and Engineering 4 (2011) 684-691 691
JBiSE
Yoda, M., Nakano, Y., Tobe, K., Nagai, R., Kimura, S.,
Tomita, M., Froguel, P. and Kadowaki, T. (2001) The fat-
derived hormone adiponectin reverses insulin resistance
associated with both lipoatrophy and obesity, Nature
Medicine, 7, 941-946. doi:10.1038/090984
[9] Simomura, I., Hammer, R.E., Ikemoto, S., Brown, M.S.
and Gorldstein, J.L. (1999) Leptin reverses insulin resis-
tance and diabetes mellitus in mice with congenital li-
podystrophy. Nature, 401, 73-76. doi:10.1038/43448
[10] Hotamisligil, G.S., Sharg, N.S. and Spiegelman, B.M.
(1993) Adipose expression of tumor necrosis factor-α:
Direct role in obesity-linked insulin resistance. Science,
259, 87-91. doi:10.1126/science.7678183
[11] Bastard, J.P., Maachi, M., Lagathu, C., Kim, M.J., Caron,
M., Vidal, H., Capeau, J. and Feve, B. (2006) Recent ad-
vances in the relationship between obesity, inflammation,
and insulin resistance. European Cytokine Network, 17,
4-12.
[12] Minamimoto, T., Orimo, M., Shimizu, I., Kunieda, T.,
Yokoyama, M., Ito, T., Nojima, A., Nabetani, A., Oike, Y.,
Matsubara, H., Ishikawa, F. and Komuro, I. (2009) A
crucial role for adipose tissue p53 in the regulation of
insulin resistance. Nature Medicine, 15, 1082-1087.
doi:10.1038/nm.2014
[13] Green, H. and Kehinde, O. (1975) An established prea-
dipose cell line and its differentiation in culture. II. Fac-
tors affecting the adipose conversion. Cell, 5, 19-27.
doi:10.1016/0092-8674(75)90087-2
[14] Gregoire, F.M., Smas, C.M. and Sul, H.S. (1998) Under-
standing adipocyte differentiation. Physiologcal Review
78, 783-809.
[15] Mimura, M., Nabeshima, R., Maeda, M. and Shiomi, N.
(2008) A highly sensitive enzyme-linked immunosorbent
assay for quantification of adipocytokines secreted by
mouse adipocytes. Biochemical Engineering Journal, 43,
58-63. doi:10.1016/j.bej.2008.08.008
[16] Tang, Q.Q. and Lane, M.D. (2000) Role of C/EBP ho-
mologous protein (CHOP-10) in the programmed activa-
tion of CCAAT/enhancer binding protein-β during adi-
pogenesis. Proceeding of National Academy of Science in
USA, 97, 12446-12450. doi:10.1073/pnas.220425597
[17] Noon, L.A., Clark, A.K. and King, P.J. (2004) A perox-
isome proliferatore-response element in the murine
mc2-γ promoter regulates its transcriptional activation
during differentiation of 3T3-L1 adipocytes. Journal of
Biological Chemistry, 279, 22803-22808.
doi:10.1074/jbc.M401861200
[18] Prusty, D., Park, B.H., Davis, K.E. and Farmer, S.R.
(2002) Activation of MEK/ERK signaling promotes adi-
pogenesis by enhancing peroxisome proliferation-acti-
vated receptor γ (PPARγ) and C/EBPα gene expression
during the differentiation of 3T3-L1 preadipocytes. Jour-
nal of Biological Chemistry, 277, 46226-46232.
doi:10.1074/jbc.M207776200
[19] Wu, Z., Rosen, E.D., Brun, R., Hauser, S., Adelmat, G.,
Troy, A.E., Mckeon, C., Darlington, G.J. and Spiegelman,
B.M. (1999) Cross-regulation of C/EBP alpha and PPARγ
controls the transcriptional pathways of adipogenesis and
insulin sensitivity. Molecular Cell, 3, 151-158.
doi:10.1016/S1097-2765(00)80306-8
[20] Ruderman, N., Chisholm, D., Pi-Sunyer, X. and Schnei-
der, S. (1998) The metabolically obese, normal-weight
individual revisited. Diabetes, 47, 699-713.
doi:10.2337/diabetes.47.5.699
[21] Ishikawa, M., Arai, S., Takano, M., Hamada, A., Kuni-
masa, K. and Mon, M. (2010) Taurin’s health influence
on Japanese school girls. Journal of Biomedical Science,
17, S47. doi:10.1186/1423-0127-17-S1-S47
[22] Kim, H.M., Do, C.H. and Lee, D.H. (2010) Characteriza-
tion of taurine as anti-obesity agent in C. elegans. Jour-
nal of Biomedical Science, 17, S33.
doi:10.1186/1423-0127-17-S1-S33
[23] Derosa, G., Maffioli, P., Ferrari, I., D’Angelo, A., Fogari,
E., Palumbo, I., Randazzo, S. and Cicero, A.F. (2011)
Orlistat and L-carnitine compared to orlistat alone on in-
sulin resistance in obese diabetic patients. Endocrine
Journal, 57, 777-786. doi:10.1507/emdocrj.K10E-049
[24] Warensjö, E., Sundström, J., Lind, L. and Vessby, B.
(2006) Factor analysis of fatty acids in serum lipids as a
measure of dietary fat quality in relation to the metabolic
syndrome in men. American Journal of Clinical Nutri-
tion, 84, 442-448.
[25] Warensjö, E., Risèrus, U. and Vessby, B. (2005) Fatty
acid composition of serum lipids predicts the develop-
ment of the metabolic syndrome in men. Diabetologia,
48, 1999-2005. doi:10.1007/s00125-005-1897-x
[26] Badley, R.L., Fisher, F.M. and Maratos-Fisher, E. (2008)
Dietary acids differentially regulate production of TNF-
alpha and IL-10 by murine 3T3-L1 adipocytes. Obesity,
16, 938-944. doi:10.1038/oby.2008.39