International Journal of Obesity (1999) 23, 595±602
ß 1999 Stockton Press All rights reserved 0307±0565/99 $12.00
http://www.stockton-press.co.uk/ijo
Dehydroepiandrosterone alters the growth of
stromal vascular cells from human adipose
tissue
MK McIntosh1*, YR Lea-Currie1, C Geigerman1 and L Patseavouras2
1
Department of Nutrition and Foodservice Systems, School of Environmental Health Sciences, University of North Carolina at
Greensboro, Greensboro, NC 27402 ± 6170, USA; and 2Patseavouras Center for Facial Plastic and Laser Surgery, 530 North Elam Ave
Greensboro, NC 27403, USA
OBJECTIVE: The purpose of this study was to determine if the antiobesity actions of dehydroepiandrosterone (DHEA)
observed in vivo are due to an in¯uence on the proliferation and differentiation of primary cultures of stromal-vascular
(SV) cells isolated from human adipose tissue.
DESIGN: SV cells were isolated from subcutaneous adipose tissue obtained from a young adult female undergoing
elective liposuction. For the proliferation assay (Experiment 1), cultures were fed proliferation media containing 0, 5,
25 or 100 mM DHEA for 3 d. At the end of this treatment period, cultures were either prepared for counting or for
determining their metabolic activity using the Alamar Blue staining procedure. For the differentiation assays
(Experiment 2), cultures were fed differentiation media containing 0, 25 or 50 mM DHEA for 20 d. At the end of this
treatment period, cultures were either prepared for lipid staining using Oil Red O or for marker enzyme analysis
(glycerol-3-phosphate dehydrogenase activity; GPDH). To determine if the stimulatory effects of DHEA on SV cell
differentiation were dependent on the presence of thiazolidinediones (Experiment 3), cultures of differentiating SV
cells were incubated in the presence and absence of BRL 49653 and either 0, 25 or 50 mM DHEA.
RESULTS: In Experiment 1, cultures treated with 25 and 100 mM DHEA had fewer cells than cultures treated with either
0 or 5 mM DHEA. Alamar Blue staining decreased as the level of DHEA in the cultures increased. In Experiment 2,
cultures treated with DHEA had more lipid and GPDH activity than control cultures. In Experiment 3, cultures treated
with BRL 49653 had more triglyceride than cultures treated without BRL 49653. Likewise, cultures treated with DHEA
had more triglyceride than their non-DHEA controls. Regardless of the BRL status, cultures supplemented with DHEA
had more triglyceride than control cultures.
CONCLUSION: These data suggest that in cultures of SV cells from human adipose tissue, DHEA supplementation
attenuates proliferation and enhances differentiation. These data support the hypothesis that DHEA directly
attenuates preadipocyte proliferation in humans as we previously demonstrated in primary cultures of pig and rat
SV cells and in cultures of 3T3-L1 preadipocytes. In contrast, DHEA stimulated the differentiation of human
preadipocytes, which is contrary to its actions in differentiating cultures of preadipocytes from animals.
Keywords: dehydroepiandrosterone; preadipocyte proliferation and differentiation; human stromal vascular cells
Introduction
Dehydroepiandrosterone (DHEA) and DHEA-sulfate
(DHEAS) are the most abundant steroids in human
plasma.1,2 Plasma concentrations of DHEA(S)
decrease steadily as adults age.1,2 In contrast, aging
is associated with increased adiposity and decreased
lean body mass. Therefore, maintaining `youthful' or
higher levels of DHEA(S) may prevent the development of obesity. Alternatively, treating obese individuals with DHEA(S) may attenuate adiposity,
returning their body fat levels back to `normal'.
Indeed, many studies have shown that administration
*Correspondence: Michael K McIntosh, Department of Nutrition
and Foodservice Systems, University of North Carolina at
Greensboro, PO Box 26170, Greensboro, NC 27402-6170, USA.
E-mail: [email protected]
Received 15 September 1998; revised 28 December 1998;
accepted 14 January 1999
of DHEA, the intracellular form of the steroid, to
rodents reduces adipose tissue mass.2 ± 11 Moreover,
several clinical trials have demonstrated that DHEA
treatment reduces body fat12,13 and increases both lean
body mass13 and muscle strength14,15 in mature adults.
Furthermore, recent studies by our group indicate that
DHEAS is more effective than DHEA in reducing
adipose tissue mass and cellularity in rats.16,17
We have preliminary data suggesting that DHEA
directly alters the growth of adipose tissue. We found
that treating 3T3-L1 preadipocytes18 and stromalvascular (SV) cells from adipose tissue of pigs and
rats19 with DHEA attenuated cellular proliferation and
differentiation. The 3T3-L1 preadipocytes are nontransformed cell line of preadipocytes of embryonic
murine origin that is capable of differentiating into
adipose-like cells. Furthermore, treatment of differentiating preadipocytes with DHEA decreased the
expression of CCAAT=enhancer binding protein
alpha (C=EBPa), a `master' regulator of adipogenesis.
Dehydroepiandrosterone alters human preadipocyte growth
MK McIntosh et al
596
Moreover, there was a dose-dependent increase in the
expression of an unknown protein that may be a
repressor of C=EBPa and adipogenesis. In contrast
to the antiadipogenic actions of DHEA, DHEAS had
no impact on 3T3-L1 preadipocyte proliferation or
differentiation.
Our data demonstrate that DHEA, or a downstream
metabolite, attenuates murine, rat and pig preadipocyte growth. Furthermore, they suggest that DHEA
attenuates the growth of preadipocytes from animals
by impairing cell proliferation (for example, apoptosis
or growth arrest) and attenuating differentiation (for
example, decreased C=EBPa expression). However, it
is not known if DHEA attenuates the growth of human
preadipocytes from adults as it does in animals.
Furthermore, the antiadipogenic mechanism of
action of DHEA is not known. Based on these data
concerning DHEA, obesity and adipose tissue growth,
the speci®c aim of this study was to investigate if
DHEA attenuates the proliferation and differentiation
of human preadipocytes from an adult female.
Materials and methods
Isolation and culture of SV cells from human adipose
tissue
Abdominal adipose tissue was obtained from a female
donor aged 34 y with a body mass index (BMI) of
31.9 kg=m during liposuction. SV cells were isolated
and cultured using modi®ed procedures described by
Entenmann and Hauner.20 Brie¯y, adipose tissue was
enzymatically digested for 45 min in a Krebs-Ringer
buffer containing 1 mg=ml collagenase (CLS-1,
Worthington Biochemical, Freehold, NJ), 15 mg=ml
BSA, 5 mM glucose and 100 mM HEPES. The ratio of
digestion solution to adipose tissue mass was 5 ml=1 g.
The digesta was then ®ltered through 200- and 60micron mesh and pelleted at 600 g for 5 min. The SV
cells were resuspended in a RBC lysis buffer (0.154 M
ammonium chloride, 10 mM potassium bicarbonate,
0.1 mM EDTA) for 10 min and then ®ltered and
recentrifuged to remove contaminating endothelial
cells. Cultures of SV cells were grown in proliferation
medium containing 90% DMEM=Hams F-12 (1:1),
10% FBS, and penicillin=streptomycin=gentamicin=
fungizone (50 U=ml; 50 mg=ml; 25 mg=ml; 0.5 mg=
ml). Cultures were incubated at 37 C in a humidi®ed
O2:CO2 (95:5%) atmosphere. Initially, freshly isolated
SV cells were grown to 80% con¯uency and then
frozen in liquid nitrogen in aliquots (1106 cells=ml).
Aliquots were subsequently thawed and seeded for
these studies. After two passages, cells were plated for
the proliferation study (Experiment 1) or the differentiation studies (Experiment 2 and Experiment 3).
Induction of cell proliferation. Freshly split SV cells
were seeded (5103=cm2) in 12-well Falcon dishes
and cultured in basal proliferation medium containing
90% DMEM=Hams F-12 (1:1), 10% FBS, and antibiotics=antimycotics. Cultures were incubated at 37 C
in a humidi®ed O2:CO2 (95:5%) atmosphere. Medium
was changed every 2 ± 3 d. At this seeding density,
cells reached con¯uency in about 8 d.
Induction of cell differentiation. Freshly split SV
cells were seeded (4104=cm2) in 12-well Falcon
dishes and allowed to attach for 18 ± 24 h in
DMEM=F-12 containing 10% FBS and antibiotics=antimycotics. Following attachment, cultures
were grown in differentiation media for the next 3 d
containing DMEM=F-12 Ham's (1:1), 10% FBS,
antibiotics=antimycotics, 33 mM biotin, 17 mM pantothenate, 100 nM human insulin, 1 mM dexamethasone, 0.5 mM isobutylmethaxanthine (IBMX), and
1 mM BRL 49653 at 37 C and incubated in humidi®ed
O2:CO2 (95:5%) atmosphere for 3 d. Thereafter, cultures were exposed to an adipocyte media containing
97% DMEM=F-12 Ham's (1:1), 3% FBS, antibiotics=antimycotics, 33 mM biotin, 17 mM pantothenate,
100 nM human insulin, 1 mM dexamethasone, and
1 mM hydrocortisone. Under these growth conditions,
oil droplets began to appear in the cells by day 7 and
maximum lipid accumulation and glycerol-3-phosphate dehydrogenase (GPDH) activity occurred by
day 15 ± day 20.
Experiment 1. Impact of DHEA on SV cell proliferation
Proliferation of cultures of precon¯uent SV cells
treated with either 0, 5, 25 or 100 mM DHEA for 3 d
was determined by counting cells on a haemocytometer and measuring the metabolic activity of the
cells using the AlamarBlue procedure.21 Freshly split
SV cells were seeded at 5103=cm2 in 12-well dishes
for measuring cell proliferation via direct counting
(n ˆ 6=trt) and by the AlamarBlue procedure
(n ˆ 6=trt). To access metabolic activity, AlamarBlue
(AccuMed International, Inc, Westlake, OH) was
added to the test wells on day 3 of treatment, at a
concentration of 1% of the total well volume. The
medium was then transferred to a 96 well dish and the
absorbency was quanti®ed at 570 nm on a microtiter
plate reader (Tecan-SLM, Research Triangle Park,
NC, USA). This procedure incorporates a colorimetric
oxidation-reduction indicator that changes colour in
response to cell metabolic activity and growth. This
indicator is not toxic to cells within the ®rst 24 h of
use. We have found a good correlation between cell
numbers and metabolic activity using this indicator
with SV cells from human adipose tissue.
Experiment 2. Impact of DHEA on SV cell
differentiation
The degree of differentiation of cultures of SV cells
exposed to either 0, 25 or 50 mM DHEA was determined on day 20 by standard lipid staining procedures
Dehydroepiandrosterone alters human preadipocyte growth
MK McIntosh et al
(n ˆ 3=trt) and by measuring the activity of GPDH
(n ˆ 3=trt) of the cultures.
determined at 492 nm on a microtiter plate reader
(Tecan-SLM).
Measuring marker enzyme activity. Cultures were
rinsed twice with HBSS and scraped into an ice cold
sucrose buffer (29 mM sucrose, 73 mM Tris, 24 mM
EDTA, 0.02% b-mercaptoethanol) and stored at
7 20 C. The resulting extract was sonicated with
two 5 s bursts and centrifuged for 30 min at 16 000 g
at 4 C. GPDH activity was determined spectrophotometrically by measuring the oxidation of NADH at
340 nm using dihydroxyacetone phosphate as the
substrate, as previously described.19 Protein content
of the supernatant was determined using a commercially available kit (Bio-Rad Labs, Richmond, CA).
Statistical analyses
Staining lipid droplets. The presence of lipid within
adipocytes was visualized and quanti®ed by staining
with Oil Red O. Cultures were washed twice with 1 ml
HBSS, then ®xed for 1 h in a 10% formalin solution
(10% formalin, 4% calcium chloride, deionized water)
at 4 C. The cells were then washed twice with
distilled water and stained using a 0.3% oil red O in
isopropanol for 15 min at room temperature. Following staining, cells were rinsed several times using
deionized water to remove any undissolved stain. The
cells were then observed under an Olympus IMT-2
inverted microscope (Mellville, NY, USA) and photographs of representative cells taken for documentation. The amount of Oil Red O staining was quanti®ed
according to the method of Phillips et al.22 Brie¯y, the
stain was solubilized in 0.5 ml of a 4% Nonidet P-40
(NP-40) solution (in isopropanol) for 5 min. The
resulting fraction was transferred to a 1 ml polystyrene
cuvette and the absorbance measured at 540 nm on a
Beckman DU64 spectrophotometer (Beckman Instruments, Palo Alto, CA, USA). The absorbance at
540 nm is proportional to the amount of stainable
lipid accumulated in the cell monolayers.22
Data were analysed by the Least Squares ANOVA
General Linear Models Procedures (PROC GLM) of
SAS (SAS Institute, Cary, NC, USA). For Experiment
1 and Experiment 2, one-way ANOVA were conducted.23 For Experiment 3, the main effects of BRL
(n ˆ 2), DHEA (n ˆ 3) and the BRL by DHEA interactions (n ˆ 6) were compared for signi®cance at the
P < 0.05 level.23 The means s.e.m. of the treatment
interactions and their statistical differences are presented in the ®gures.
Results
Experiment 1. Impact of DHEA on SV cell proliferation
SV cell numbers decreased as the level of DHEA
increased above 5 mM (Figure 1). The metabolic
activity of proliferating cultures of human SV cells
decreased as the level of DHEA in the cultures
increased (Figure 2).
Experiment 2. Impact of DHEA on SV cell
differentiation
The activity of the adipocyte enzyme marker GPDH
was higher in cultures treated with 25 and 50 mM
DHEA compared to the control cultures (Figure 3).
Likewise, the amount of stainable lipid in the cultures
Experiment 3. Impact of DHEA on SV cell
differentiation in the presence and absence of
BRL49653.
Cultures of SV cells were grown in the presence or
absence of BRL 49653 and treated with either 0, 25 or
50 mM DHEA in this 25 factorial design. The degree
of differentiation of SV cells was determined on day
12 by measuring the triglyceride content (n ˆ 8=trt)
and the activity of GPDH (n ˆ 8=trt).
Measuring triglyceride content. Cellular triglyceride
content was determined by using a colormetric kit
(TG-INT #336, Sigma Chemical Co., St Louis, MO,
USA). Brie¯y, cells were harvested in an isotonic
sucrose buffer (29 mM sucrose, 73 mM Tris, 24 mM
EDTA) and sonicated at 20% output for 10 s. Aliquots
of this crude cell sonicate were incubated for 2 h with
the triglyceride reagent and then the absorbance
Figure 1 The effect of dehydroepiandrosterone (DHEA) treatment for 3 d on cell number in replicating cultures of stromalvascular (SV) cells from human adipose tissue. Values represent
the least square means s.e.m. Means not sharing a common
superscript (a, b, c) are signi®cantly (P < 0.05) different.
597
Dehydroepiandrosterone alters human preadipocyte growth
MK McIntosh et al
598
Figure 2 The effect of dehydroepiandrosterone (DHEA) treatment for 3 d on metabolic activity in replicating cultures of
stromal-vascular (SV) cells from human adipose tissue. Values
represent the least square means s.e.m. Means not sharing a
common superscript (a, b, c) are signi®cantly (P < 0.05) different.
Figure 4 The effect of dehydroepiandrosterone (DHEA) treatment for 20 d on Oil Red O-stained lipid content in differentiating
cultures of stromal-vascular (SV) cells from human adipose
tissue. Values represent the least square means s.e.m. Means
not sharing a common superscript (a, b, c) are signi®cantly
(P < 0.05) different.
Experiment 3. Impact of DHEA on SV cell
differentiation in the presence and absence of
BRL49653
BRL supplemented cultures had more triglyceride
than non-BRL supplemented cultures (main effect
BRL; P ˆ 0.001). Likewise, SV cultures treated with
DHEA had more triglyceride than non-DHEA treated
cultures (main effect DHEA; P ˆ 0.001). Regardless
of the BRL status of the cultures, cultures treated with
25 and 50 mM DHEA had more triglyceride than nonDHEA treated cultures (Figure 6). Treatment effects
on the activity of GPDH were similar to those of the
cellular triglyceride data and therefore these data are
not shown.
Discussion
Figure 3 The effect of dehydroepiandrosterone (DHEA) treatment for 20 d on the activity of glycerol-3-phosphate dehydrogenase (GPDH), an enzyme marker of late differentiation, in
differentiating cultures of stromal-vascular (SV) cells from
human adipose tissue. Values represent the least square means s.e.m. Means not sharing a common superscript (a, b, c) are
signi®cantly (P < 0.05) different.
treated with 25 and 50 mM DHEA was higher than that
of the non-DHEA treated cultures (Figure 4). As seen
in Figure 5, there are more clusters of larger lipid
droplets in the cultures treated with DHEA compared
to the controls.
The present study provides direct evidence that
DHEA treatment attenuates proliferation and stimulates differentiation in primary cultures of SV cells
isolated from human adipose tissue. To our knowledge this is the ®rst time the antiproliferative yet proadipogenic actions of DHEA have been demonstrated
in primary cultures of SV cells isolated from human
abdominal adipose tissue. The antiproliferative effects
of DHEA in SV cells from human adipose tissue
corresponds with previous data from our laboratory,
demonstrating the antiproliferative effects of DHEA
in pig and rat SV preadipocytes19 and in the 3T3-L1
preadipocyte cell line.18 Furthermore, several clinical
trials have demonstrated that treatment with DHEA
Dehydroepiandrosterone alters human preadipocyte growth
MK McIntosh et al
599
Figure 6 The effect of BRL 49653 and dehydroepiandrosterone
(DHEA) treatment for 12 d on the triglyceride content in differentiating cultures of stromal-vascular (SV) cells from human
adipose tissue. Values represent the least square means s.e.m.
Means not sharing a common superscript (a, b, c, d) are signi®cantly (P < 0.05) different.
Figure 5 Micrographs (10x) of cultures of stromal-vascular (SV)
cells from human adipose tissue treated with dehydroepiandrosterone (DHEA) for days and stained with Oil Red O. Arrowheads
indicate adipocytes loaded with Oil Red O-stained droplets.
reduces body fat12,13 and increases lean body mass
and muscle strength14,15 in mature adults. Moreover,
low levels of serum DHEA(S) have been correlated
with high levels of body fat in both men24,25 and
women.26 ± 28 Therefore, DHEA may antagonize adiposity by reducing preadipocyte proliferation.
In support of this concept, it is well documented
that the rate of conversion of preadipocytes to mature
adipocytes directly affects the development of obesity.29 The observed DHEA-mediated reduction in
proliferation of the preadipocyte pool would limit
the number of cells that could be induced to differentiate into mature lipid-®lled adipocytes. Therefore,
we speculate that the decline in DHEA levels that
occurs with age may in¯uence the size of the preadipocyte pool, by allowing greater proliferation of
cells capable of accumulating signi®cant amounts of
lipid as evidenced in obese subjects. These data
suggest that the antiobesity effects of DHEA observed
in humans may be due to the direct attenuation of
preadipocyte proliferation by DHEA or down-stream
metabolites. However, since we did not distinguish
between the proliferation of preadipocytes and
other supporting SV cells such as (pre)myocytes
or (pre)chondrocytes in our cultures, we must con®rm
these results by directly measuring preadipocyte
proliferation.
The mechanism by which DHEA attenuates preadipocyte proliferation is not known. DHEA may be
reducing proliferation by: 1) antagonizing the actions
of a gene whose activation is essential for cell growth;
2) limiting substrate availability for growth; 3) causing growth arrest; or 4) being cytotoxic. Schultz et
al 30 demonstrated that DHEA treatment of HT-29 SF
cells (a human colonic adenocarcinoma cell line)
inhibited cell proliferation and arrested the cells in
the G1 phase of the cell cycle. Supplementation with
mevalonic acid and=or isoprenoids reversed the
DHEA-mediated growth arrest. Mevalonic acid is an
essential substrate for the biosynthesis of cholesterol
and the precursor for farnesyl-pyrophosphate, an
Dehydroepiandrosterone alters human preadipocyte growth
MK McIntosh et al
600
important substrate for the post-translational modi®cation and activation of p21ras=G-protein.31 Moreover, mevalonic acid provides isoprene units for the
post-translational modi®cation of proteins integral to
proliferation and differentiation.30 These data suggest
that DHEA may reduce isoprenoid synthesis, thereby
limiting the availability of compounds essential for
cell proliferation.30 Furthermore, 3-hydroxy-3methylglutaryl coenzyme A (HMG-CoA) reductase,
the rate limiting microsomal enzyme in de novo
cholesterol and isoprenoid biosynthesis, is inhibited
by DHEA treatment.32 Moreover, the HMG-CoA
reductase inhibitor lovastatin has been shown to
inhibit 3T3-L1 preadipocyte differentiation by limiting the farnesylation of p21ras.33 Therefore, DHEA
may alter preadipocyte proliferation by interfering
with cholesterol and isoprene biosynthesis.
In contrast to the antiproliferative effects of DHEA,
treatment of cultures of human SV cells with DHEA
enhanced preadipocyte differentiation. The pro-adipogenic actions of DHEA in human SV cells are in
contrast to our previous studies demonstrating that
DHEA attenuated the differentiation of cultures of
3T3-L1 preadipocytes18 and primary cultures of rat
and pig SV preadipocytes.19 Several other groups
have also reported that DHEA reduced the rate of
differentiation of 3T3-L1 cells into mature, lipid-®lled
adipocytes by as much as 50%.9,10 The reason for this
differential effect of DHEA on preadipocyte differentiation between cultures of animal and human preadipocytes is not known. However, DHEA's in¯uence
of preadipocyte growth may be affected by: 1) the
culturing conditions of the cells; 2) the age, body mass
index (BMI); and gender of the subject; and 3) the site
from which the adipose tissue was collected.
The culturing conditions for preadipocyte differentiation of humans cells differed from that of animal
cells in that the thiazolidinedione BRL 49653 was
supplemented during the ®rst 5 d of the human differentiation protocol. The thiazolidinediones have insulin
sensitizing and adipogenic properties33,34 and are used
to enhance adipogenesis in cultures of human preadipocytes. DHEA stimulated differentiation in the presence and absence of the BRL 49653 (Figure 3, Figure
4 and Figure 6). The further stimulation of lipid
accumulation in the presence of BRL 49653 may be
due to the fact that BRL 49653 is a potent ligand for
peroxisome proliferator activated receptor gamma
(PPAR-g2).34 The expression of PPAR-g2 has been
shown to be essential for terminal adipocyte differentiation.34 ± 36 This transcription factor binds to the
promoter regions of genes involved in adipogenesis
such as the fatty acid binding protein aP2, lipoprotein
lipase (LPL), and acyl CoA synthase.35,36,37 DHEA
and DHEAS have been shown to enhance the size and
numbers of peroxisomes4,38 and peroxisomal betaoxidation in rodent liver.4,39,40 Therefore, it is possible
that DHEA stimulates adipogenesis in cultures of
human SV cells by activating PPAR-g2, thereby
stimulating differentiation by a mechanism similar
to that of BRL 49653. In the current study, we
observed an additive, but not synergistic, effect of
BRL 49653 in cultures treated with DHEA (Figure 6).
Future studies will examine whether DHEA increases
the expression of PPAR-g2.
Another possible reason for the differential effect of
DHEA between human and animal preadipocytes may
be due to the age and the degree of adiposity of the
subjects. The 3T3-L1 preadipocytes we used18 are of
embryonic origin and the animal SV cells were from
pigs aged 7 d and rats aged 3 weeks.19 In contrast, the
human SV cells were obtained from a female aged
34 y. It is therefore possible that DHEA may either
attenuate or stimulate differentiation, depending on
the age-related expression of genes that control adipogenesis. In support of this concept, it is well
documented that certain hormone receptors are
expressed maximally during speci®c windows of
time during development.41 Failure of these receptors
to be activated by their respective ligands may result
in permanent aberrations in growth and development,
regardless of hormone supplementation outside these
time periods. For example, cretinism is a permanent
neurological and skeletal retardation resulting from
inadequate thyroid hormone during fetal and neonatal
life.41 Whereas iodine or thyroid hormone supplementation gestationally or neonatally prevents cretinism,
post-natal supplementation does not reverse retardation.41 Therefore, future studies will examine whether
the age or the BMI of the donor affects DHEA's
in¯uence of adipogenesis.
The speci®c site from which the SV cells were
obtained, may also in¯uence the actions of DHEA.
For the primary culture of animal cells, we collected
dorsal fat from the pigs and inguinal fat from the
rats.19 In contrast, we obtained abdominal fat from the
human subject. Regional differences in fat depot
metabolism have been observed in animals42 and
humans.43 Future studies need to address the tissue
speci®c effects of DHEA to determine whether
DHEA's in¯uence on adipogenesis is dependent on
the site of adipose tissue.
Lastly, gender may in¯uence DHEA's actions on
adipogenesis. We obtained animal SV cells from
males whereas the human SV cells were obtained
from a female. Gender differences in adipose tissue
distribution and metabolism are well documented.44
Furthermore, steroid hormone pro®les differ between
genders. Therefore, it is possible that the differential
in¯uence of DHEA we found between animal and
human SV cells may be due to gender differences.
Future studies will address this gender issue.
Conclusion
In summary, this study demonstrates that DHEA
possesses antiproliferative effects in cultures of
Dehydroepiandrosterone alters human preadipocyte growth
MK McIntosh et al
human SV cells from adipose tissue as it does in rat,
pig and 3T3-L1 preadipocytes. In contrast, DHEA
stimulated human preadipocyte differentiation. Future
studies will determine if DHEA increases PPAR-g2
expression in cultures of SV cells from human adipose
tissue and whether the age, gender, BMI or site of
adipose tissue of the subject in¯uences DHEA's
impact on adipogenesis.
Acknowledgements
This research was supported by the North Carolina
Agricultural Research Service Project No. 05773 and
the UNC Institute of Nutrition. We thank Beth Alexander for technical assistance with helping establish
the preadipocyte cultures and the proliferation assays.
References
1 Orentreich N, Brind J, Rizer R, Vogelman J. Age changes and
sex differences in serum DHEA-sulfate concentrations
throughout adulthood. J Clin Endocrinol Metab 1985; 59:
551 ± 555.
2 Orentreich N, Brind J, Vogelman J, Andres R, Baldwin H.
Long term longitudinal measurements of plasma DHEA
sulfate in normal men. J Clin Endocrinol Metab 1992; 75:
1002 ± 1004.
3 Nestler J, Clore J, Blackard W. Dehydroepiandrosterone: The
missing link between hyperinsulinemia and atherosclerosis.
FASEB J 1992; 6: 3073 ± 3075.
4 Berdanier C, Parente J, McIntosh M. Is DHEA an antiobesity
agent? FASEB J 1993; 7: 414 ± 419.
5 Coleman DL. Therapeutic effects of DHEA and its metabolites in diabetes-obesity mutants. In: Lardy H, Stratman F
(eds). Hormones, Thermogenesis, and Obesity. Elsevier: New
York, 1989, pp 377 ± 384.
6 Cleary M. The antiobesity effects of DHEA in rats. Proc Soc
Exp Biol Med 1991; 196: 8 ± 16.
7 Mohan P, Ihnen J, Levin B, Cleary M. Effect of DHEA
treatments with diet-induced obesity. J Nutr 1990; 120:
1103 ± 1114.
8 Tagliaferro A, Ronan A, Payne J, Meeker L, Tse S. Increased
lipolysis to beta-adrenergic stimulation after DHEA treatment
in rats. Am J Physiol 1995; 268: R1374 ± R1380.
9 Gordon G, Newitt J, Shantz L, Weng D, Talalay P. Inhibition
of the conversion of 3T3-L1 ®broblast clones to adipocytes by
DHEA and related anticarcinogenic steroids. Cancer Res
1986; 46: 3389 ± 3395.
10 Shantz L, Talalay P, Gordon G. Mechanism of inhibition of
growth of 3T3-L1 ®broblasts and their differentiation to
adipocytes by DHEA and related steroids: role of G6PD.
PNAS 1989; 86: 3852 ± 3856.
11 Cleary M, Hood S, Chando C, Hansen C, Billheimer J.
Response of sucrose-fed rats to DHEA. Nutr Res 1984; 4:
485 ± 494.
12 Nestler J, Barlascini C, Clore J, Blackard W. DHEA reduces
serum LDL levels and body fat but does not alter insulin
sensitivity in older men. J Clin Endocrinol Metab 1988; 66:
57 ± 61.
13 Jakubowicz J, Beer N, Rengifo R. Effect of DHEA on c-GMP
and body composition in men of advancing age. Ann NY Acad
Sci 1995; 774: 312 ± 315.
14 Morales A, Nolan J, Nelson J, Yen S. Effects of replacement
dose of DHEA in men and women of advancing age. J Clin
Endocrinol Metab 1994; 78: 1360 ± 1367.
15 Yen S, Morales A, Khorram O. Replacement of DHEA in
aging men and women. Potential remedial effects. Ann NY
Acad Sci 1995; 774: 128 ± 142.
16 Lea-Currie R, Wu S, McIntosh M. Effects of acute administration of dehydroepiandrosterone sulfate on adipose tissue mass
and cellularity in male rats. Int J Obes 1997; 21: 147 ± 154.
17 Lea-Currie YR, Wen P, McIntosh M. Dehydroepiandrosterone
sulfate attenuates adipocyte hyperplasia associated with
feeding rats a high fat diet. Int J Obes 1997; 21:
1058 ± 1064.
18 Wen P, Lea-Currie R, Alexander E, McIntosh M. Dehydroepiandrosterone reduces proliferation and differentiation of
3T3-L1 preadipocytes. Biochem Biophys Res Commun 1998;
248: 497 ± 504.
19 McIntosh M, Hausman D, Martin R, Hausman G. Dehydroepiandrosterone (DHEA) attenuates preadipocyte proliferation
and differentiation in primary cultures of stromalvascular cells. Am J Physiol (Endocrinol Metab) 1998; 275:
E285 ± E293.
20 Entermann G, Hauner H. Relationship between replication
and differentiation in cultured human adipocyte precurser
cells. Am J Physiol (Cell Physiol) 1996; 270: C1011 ±
1016.
21 Goegan P, Johnson G, Vincent R. Effects of serum protein and
colloid on the AlamarBlue Assay in cell cultures. Toxic In
Vitro 1995; 9: 257 ± 266.
22 Phillips M, Enam E, Liu P, Matsumura F. Inhibition of 3T3-L1
adipose differentiation by 2,3,7,8-tetrachlorodibenzo-p-dioxin.
J Cell Sci 1995; 108: 395 ± 402.
23 Steele R, Torrie J. Principles and Procedures of Statistics.
McGraw-Hill: New York, 1980.
24 Haffner S, Valdez R. Decreased testosterone and DHEAS
concentrations are associated with increased insulin and glucose concentrations in nondiabetic men. Metabolism 1994; 43:
599 ± 603.
25 Lopez S, Krehl W. A possible relationship between glucose-6phosphate dehydrogenase and DHEA in obesity. Lancet 1967;
2: 485 ± 487.
26 DePergola G, Giagulli V, Garruti G, Cospite M, Giorgino F,
Cignarelli M, Giorgino R. Low DHEA levels in premenopausal obese women with very high body mass index. Metabolism
1991; 40: 187 ± 190.
27 DePergola G, Triggiani V, Giorgino F, Cospite M, Garruti G,
Cignarelli, M, Guastamacchia E, Giorgino R. The free testosterone to DHEAS molar ratio as a marker of visceral fat
accumulation in premenopausal obese women. Int J Obes
1994; 18: 659 ± 664.
28 Williams D, Boyden T, Pamenter R, Lohman T, Going S.
Relationship of body fat percentage and fat distribution with
dehydroepiandrosterone sulfate in premenopausal women.
J Clin Endocrinol Metab 1993; 77: 80 ± 85.
29 Martin R, Hausman G, Hausman D. Regulation of adipose cell
development in utero. Proc Soc Exp Biol Med 1998; 219:
200 ± 210
30 Schultz S, Klann R, Schonfeld S, Nyce J. Mechanism of cell
growth and inhibition and cell cycle arrest in human colonic
adenocarcinoma cells by dehydroepiandrosterone: Role of
isoprenoid biosynthesis. Cancer Res 1992; 52: 1372 ± 1376.
31 Maltese W. Covalent modi®cation of proteins by farnesol and
the control of cell proliferation. In: Columanp A, Feo F,
Pascale R, Pani P (eds). Chemical Carcinogenesis: Modulating Factors Vol. II. Plenum Press: New York, 1991, pp 327 ±
335.
32 Pascale R, Simile M, De Miglio M, Nufris A, Seddaiu M.
Inhibition of 3-hydroxy-3-methylglutaryl-CoA reductase
activity and gene expression by dehydroepiandrosterone
in prenoplastic liver nodules. Carcinogenesis 1995; 16:
1537 ± 1542.
33 Nishio E, Tomiyama K, Nakata H, Watanabe Y. 3-Hydroxy-3methylglutaryl A reductase inhibitor impairs cell differentiation in cultures adipogenic cells (3T3-L1). Eur J Pharmacol
1996; 301: 203 ± 206.
34 Chawla A, Lazar M. Peroxisome proliferator and retinoid
signaling pathways co-regulate preadipocyte phenotype and
survival. Proc Natl Acad Sci USA 1994; 92: 1786 ± 1790.
601
Dehydroepiandrosterone alters human preadipocyte growth
MK McIntosh et al
602
35 Lehmann J, Moore L, Smith-Oliver T, Wilkison W, Wilson T.
An antidiabetic thiazolidinedione is a high af®nity ligand for
peroxisome proliferator-activated receptor g (PPAR-g). J Biol
Chem 1995; 270: 12953 ± 12956.
36 Tontonoz P, Hu E, Spiegelman B. Stimulation of adipogenesis
in ®broblasts by PPARg2, a lipid-activated transcription factor.
Cell 1994; 79: 1147 ± 1156.
37 Tontonoz P, Hu E, Graves R, Budavari A, Spiegelman B.
mPPARg2: tissue-speci®c regulator of an adipocyte enhancer.
Genes Dev 1994; 8: 1224 ± 1234.
38 Rao M, Subbarau V, Yeldandi A, Reddy J. Hepatocarcinogenicity of dehydroepiandrosterone in the rat. Cancer Res
1992; 52: 2977 ± 2979.
39 Prasanna H, Hart R, Magee P. Recent studies on the effects of
DHEA on the metabolism of carcinogens in vivo. J Toxicol
Toxin Rev 1989; 8: 121 ± 131.
40 Yamada J, Sakuma, M, Suga T. Induction of peroxisomal boxidation enzymes by dehydroepiandrosterone and its sulfate
in primary cultures of rat hepatocytes. Biochim Biophys Acta
1992; 1137: 231 ± 236.
41 Catt K. Molecular mechanisms of hormone action: Control of
target cell function by peptide, steroid and thyroid hormones.
In: Felig P, Baxter J, Broadus A, Frohman L (eds). Endocrinology and Metabolism (2nd edn). McGraw-Hill: New York,
1987, 82 ± 160.
42 Santos R, Sztalryd C, Reavan G. Effect of anatomical site on
insulin action and insulin receptor phosphorylation in isolated
rat adipocytes. Int J Obes 1991; 15: 755 ± 762.
43 Bolinder J, Kager L, Ostman J, Arneer P. Differences at the
receptor and postreceptor levels between human omental and
subcutaneous adipose tissue in the action of insulin on lipolysis. Diabetes 1983; 32: 117 ± 123.
44 Bouchard C, DespreÂs J-P, Maureige P. Genetic and nongenetic determinants of regional fat distribution. Endocr Rev
1993; 14: 72 ± 93.