ARTICLE IN PRESS

www.elsevier.com/locate/yexcr

YEXCR-06733; No. of pages: 8; 4C: 4

DTD 5

Experimental Cell Researc

Functional studies of mesenchymal stem cells derived from

adult human adipose tissue

Andrea Dickera,1, Katarina Le Blancb,c,1, Gaby Astroma, Vanessa van Harmelena,

Cecilia Gotherstromb, Lennart Blomqvista, Peter Arnera, Mikael Rydena,*

aDepartment of Medicine, M63, Karolinska Institutet, Karolinska University Hospital, Huddinge, 141 86 Stockholm, SwedenbDepartment of Laboratory Medicine, Division of Immunology, Karolinska Institutet, Karolinska University Hospital, Huddinge, 141 86 Stockholm, Sweden

cCenter for allogeneic stem cell transplantation, Karolinska University Hospital, Huddinge, 141 86 Stockholm, Sweden

Received 4 February 2005, revised version received 20 April 2005

Abstract

Recent evidence suggests that cells with the properties of human mesenchymal stem cells (hMSCs) can be derived from adult peripheral

tissues, including adipose tissue, muscle and dermis. We isolated hMSCs from the stromal–vascular portion of subcutaneous adipose tissue

from seven adult subjects. These cells could be readily differentiated into cells of the chondrocyte, osteocyte and adipocyte lineage

demonstrating their multipotency. We studied the functional properties of hMSCs-derived adipocytes and compared them with adipocytes

differentiated from hMSCs obtained from bone marrow (BM-hMSC). The two cell types displayed similar lipolytic capacity upon stimulation

with catecholamines, including a pronounced antilipolytic effect mediated through a2A-adrenoceptors, a typical trait in human but not rodent

fat cells. Furthermore, both cell types secreted the fat cell-specific factors leptin and adiponectin in comparable amounts per time unit. The fat

tissue-derived hMSCs retained their differentiation capacity up to at least fifteen passages. We conclude that hMSCs derived from adult

human adipose tissue can be differentiated into fully functional adipocytes with a similar, if not identical, phenotype as that observed in cells

derived from BM-hMSCs. Human adipose-tissue-derived MSCs could therefore constitute an efficient and easily obtainable renewable

cellular source for studies of adipocyte biology.

D 2005 Elsevier Inc. All rights reserved.

Keywords: Mesenchymal stem cell; Adipocyte; Lipolysis; Leptin; Adiponectin

Introduction

Human mesenchymal stem cells (hMSCs) display multi-

potency with the ability under the correct conditions to

differentiate into lineages of mesenchymal tissues including

muscle, bone, cartilage and fat [1]. This property renders

them interesting as a model system for studies of differ-

entiation pathways and potentially useful for cell and gene

therapy. MSCs were first characterized in bone marrow, but

0014-4827/$ - see front matter D 2005 Elsevier Inc. All rights reserved.

doi:10.1016/j.yexcr.2005.04.029

Abbreviations: hMSC, human mesenchymal stem cell; TG, triglyceride;

AR, Adrenoceptor; Adr, adrenaline; Y, Yohimbine; BM, bone marrow.

* Corresponding author. Fax: +46 8 585 838 50.

E-mail address: mikael.ryden@medhs.ki.se (M. Ryden).1 These two authors contributed equally to the work presented herein.

a wealth of studies have demonstrated the presence of

uncommitted MSCs progenitor cells in the connective tissue

of several organs including muscle, adipose tissue and

trabecular bone (for a review see [2]). The role of these cells

is not entirely clear, but they are generally believed to

constitute a reserve cellular fraction for tissue maintenance

and repair. Although several authors have shown that

adipose tissue hMSC can differentiate into morphologically

clearly discernable cell types [3], few studies have

demonstrated that the functional properties of hMSCs

differentiated in vitro are equivalent to those of primary

cell cultures from the same organ. This is partly due to the

fact that there are no well-established methods to assess the

functional properties of chondrocyte/osteocyte/myocyte

function in vitro. In this respect, differentiation into the

h xx (2005) xxx – xxx

ARTICLE IN PRESSA. Dicker et al. / Experimental Cell Research xx (2005) xxx–xxx2

adipocyte lineage has a clear advantage since the functional

properties of these cells can be easily investigated. Mammal

adipocytes display at least two unique traits compared to

non-fat cells. They store energy in the form of triglycerides

(TGs) which are released via lipolysis in the form of

glycerol and free fatty acids and they secrete fat-specific

proteins, in particular leptin and adiponectin into the

circulation [4].

In experimental fat cell research, immortalized cell lines

are of paramount importance. Although several murine fat

cell lines are available, to date, no human white fat cell line

has been described. Since there are important species-

specific differences in adipocyte function, there is a strong

need for a human white fat cell line. It is quite possible that

hMSCs may be of value in this respect, providing they can

retain their capacity to differentiate into fat cells that

maintain human-specific adipocyte characteristics. In a

recent paper, we could in fact demonstrate that hMSCs

derived from fetal liver and adult human bone marrow

could be differentiated into functional adipocytes with

characteristics similar to those observed in primary cultures

of human preadipocytes (which are lineage-committed

progenitor cells) [5]. Due to ethical considerations related

to the use of fetal tissues and the somewhat laborious and

painful procedure associated with bone marrow aspiration,

it would be ideal to have an easily accessible source of

hMSCs in adult subjects. Investigators have recently shown

that hMSC with retained pluripotentiality can be derived

from subcutaneous adipose tissue [6–12]. It is not known

whether hMSCs present in adult fat are also able to

differentiate into functional cells. Since subcutaneous lip-

oaspirates are very easy to perform and associated with

minimal risks, adipose tissue could constitute an important

source for hMSCs. However, this warrants a detailed

analysis and comparison with hMSCs derived from bone

marrow. We determined whether subcutaneous adipose

tissue of adult humans contains hMSCs and whether these

cells display functional characteristics of adipocytes that

can be maintained for numerous passages. Furthermore, we

compared the phenotype of these cells with adipocytes

derived from BM-hMSCs.

Table 1

Mesenchymal stem cells in adult human adipose tissue

Subject number

(fat tissue donor)

Age

(years)

BMI

(kg/m2)

Gender Subject number

(BM donor)

Age

(years)

1 24 29.8 F 1 25

2 27 21.9 F 2 25

3 50 61.5 M 3 57

4 28 24.2 M 4 61

5 28 23.5 F 5 69

6 38 21.6 F

7 32 28.4 F

To the left is a summary of age and BMI of the different subjects from

which fat stroma was isolated for subsequent preparation of adipose-tissue-

derived hMSC. To the right are the ages of the bone-marrow donors (no

anthropometric data available).

Materials and methods

Human MSC culture

Subcutaneous fat was obtained from seven subjects (two

men, five women) undergoing cosmetic lipoaspiration. They

were aged between 24 and 50 years, with a body mass index

(BMI) ranging from 21.6 to 61.5 kg/m2. Although

adipocytes derived from the subject with the highest BMI

displayed somewhat lower levels of GPDH, in this limited

material, there was no apparent effect of BMI on either cell

differentiation or adipocyte function (data not shown). Bone

marrow aspirates were obtained from five normal-weight

subjects, aged between 25 and 69 years. Cells from these

subjects have been examined in detail in other studies from

our groups. Unfortunately, no anthropometric data were

obtained from these donors. The individual patient data are

detailed in Table 1. The hospital’s committee of ethics

approved this study, and informed consent was obtained

from all subjects. Mature adipocytes were separated out

according to Rodbell et al. [13]. The remaining stromal–

vascular portion was used for subsequent isolation of MSC

as described previously [5]. Cells were plated at a density of

160,000/cm2 and allowed to proliferate in Dulbecco’s

Modified Eagle’s Medium–Low Glucose (DMEM-LG)

(Life Technologies, Gaithersburg, MD), supplemented with

10% fetal bovine serum (FBS) (Sigma, St. Louis, MO,

USA) and 1% antibiotic–antimycotic solution (Life Tech-

nologies, Gaithersburg, MD). The serum lot used was

selected on the basis of optimal MSC proliferation. In

continued passages, cells were plated at 4000/cm2. Cultures

were maintained at 37-C in a humidified atmosphere

containing 5% CO2 and passaged at 60–70% confluency

for at least fifteen passages. Isolated MSCs were analyzed

by flow cytometry. Detached cells were washed and re-

suspended in PBS. Approximately 105 cells were incubated

on ice for 30 min with conjugated, monoclonal antibodies

against CD34, CD14 and CD45 (Becton Dickinson, San

Jose, CA, USA), CD44 (Immunotech, Marseilles, France),

CD29 (Coulter, Miami, FL, USA), CD 166 and CD105

(Ansell, Bayport, MN, USA) and CD73 (PharMingen,

Uppsala, Sweden). Nonspecific fluorescence was deter-

mined, using equal aliquots of cell preparation incubated

with anti-mouse monoclonal antibodies (Becton Dickinson).

Finally, the cells were assayed in a flow cytometer

(FACSort, Beckton Dickinson), and the data analyzed with

Cellquest software (Becton Dickinson).

Cell lineage differentiation

Both types of hMSCs were differentiated into the

adipogenic, osteogenic and chondrogenic lineage as

described previously [5]. To promote chondrogenic differ-

ARTICLE IN PRESSA. Dicker et al. / Experimental Cell Research xx (2005) xxx–xxx 3

entiation, 7.5 � 105 mesenchymal cells were pelleted in a

15-mL polypropylene tube and cultured in Dulbecco’s

Modified Eagles Medium–High Glucose (DMEM-HG)

(Life Technologies) with insulin (6.25 Ag/mL), transferrin

(6.25 Ag/mL), linolenic acid (5,33 Ag/ml), BSA (1.25 mg/

ml), pyruvate (1 mmol/L), ascorbate-2-phosphate (0.17

mmol/L), dexamethasone (0,1 Amol/L), proline (0.35

mmol/L) and selenous acid (6.25 ng/mL). Differentiation

was induced by the addition of TGF-h3 (0.01 Ag/mL) (R&D

Systems, Minneapolis, MN). The medium was replaced

every 3–4 days for 21 days. Pellets were formalin fixed,

embedded in paraffin, examined morphologically and

stained for the presence of glucosaminoglycans with

Alacian green. For osteogenic differentiation, 3.1 � 103

cells per cm2 were grown in MSC medium supplemented

with dexamethasone (0.1 Amol/L), ascorbic acid (0.05

mmol/L) and glycerophosphate (10 mmol/L). The medium

was replaced every 3–4 days for 21 days. Osteocytes were

stained with Alizarin Red. Differentiation of preadipocytes

from both bone marrow and adult adipose tissue was

performed as described previously [5] and assessed by

determining glycerol-3-phosphate-dehydrogenase (GPDH)

activity and staining with the triglyceride-specific dye Oil

Red O [5].

Lipolysis

Lipolysis experiments were performed on differentiated

adipocytes as described previously [14]. In brief, cells were

washed with DMEM/NUT.MIX.F-12 medium and then

incubated in duplicates for 3 h at 37-C with DMEM/

NUT.MIX.F-12 medium containing 20 g/L BSA. The

following concentrations were used for each agent: 0.1

mM adrenaline and 0.1 mM of the a2-adrenoceptor blocker

yohimbine. Incubation without drugs was made to deter-

mine basal lipolysis. Following incubation, medium was

removed and kept at �20-C for subsequent measurement of

glycerol concentration (an index of lipolysis) using a

bioluminescence method [15]. For protein measurements,

cells were lysed in a pre-chilled protein lysis buffer

containing 50 mM Tris pH 7.6, 150 mM NaCl, 1% Triton

X-100, 0.1 M PMSF and protease inhibitors as described

previously [16]. Protein content was assayed using BCA

Protein Assay Reagent Kit (PIERCE, Rockford, IL,USA) on

96-well microtiter plates with BSA as a standard. Measure-

ments were performed on a spectrophotometer in accord-

ance to the guidelines given in the kit. GPDH and protein

concentrations were almost identical in the different wells,

therefore glycerol release could be expressed in absolute

terms.

Adipokine release from differentiated adipocytes

Adipokine release was measured in conditioned media

from adipocytes of several passages. Medium was collected

after a 48-h incubation, and release was expressed per 24 h.

Leptin was measured according to the manufacturer’s

protocol using the Quantikine Human Leptin Immunoassay

(R&D Systems, Minneapolis, MM, USA). Adiponectin

release was determined using a human adiponectin RIA-

kit (Linco Research, Inc, St Charles, MO).

Statistical analysis

Values are given as mean T standard error (SE). They werecompared with the non-parametric test Kruskal–Wallis.

Drugs and chemicals

Bovine serum albumin (BSA) fraction V (A-9418),

glucose, glycerol kinase, adrenaline, Oil Red O and

yohimbine were obtained from Sigma Chemical (Sigma,

St. Louis, MO, USA). All chemicals used were of the

highest grade of purity that was commercially available.

Results

Phenotype of MSC derived from adipose tissue and bone

marrow

Cultured MSCs from both sources were strongly positive

for the endoglin receptor CD105, CD73, CD44, CD90 and

h-1-integrin 29 (Fig. 1). Flow cytometry was negative for

the endothelial marker CD31, CD80, the lipopolysacharide

receptor CD14 and the leukocyte common antigen CD45.

While bone-marrow-derived MSCs were negative for

hematopoietic stem cell marker CD34, a slight positivity

(15–20%) was observed in adipocyte-derived MSC which

is in agreement with findings in a recent report [17].

Adipose-tissue-derived hMSCs display multipotent

differentiation

The multipotency of the MSCs isolated from adult

subcutaneous adipose tissue was demonstrated by incubat-

ing the cells in media promoting differentiation into the

osteogenic, chondrogenic and adipogenic lineage, respec-

tively. Incubation in the presence of dexamethasone, h-glycerophosphate and ascorbate induced cell aggregation

and matrix production, which stained positive with the

calcium-specific marker Alizarin Red (Fig. 2A). Pelleted

cells cultured with TGF-h developed typical morphological

features of chondrocytes and a multilayered proteoglycan-

rich matrix which stained positive with Alacian green (Fig.

2B). Finally, MSCs incubated with insulin and the PPAR-g

agonist Roziglitazone differentiated into adipocyte-like

cells, which stained positive with the triglyceride specific

dye Oil Red O (Fig. 2C). The MSC-derived adipocytes

were also analyzed for the activity of the fat cell specific

enzyme glycerol-3-phosphate dehydrogenase (GPDH). The

activity levels were similar to that observed in human

ARTICLE IN PRESS

Fig. 2. Differentiation of hMSC derived from adult s.c. fat tissue. Cells were

incubated in the presence of specific differentiation agents. (A) Alizarin

staining shows the presence of bone tissue and osteocytes. (B) Staining with

Alacian green demonstrates the deposition of proteoglycans and the

morphological characteristics of chondrocytes. (C) Differentiation into the

adipocyte lineage was demonstrated by staining with Oil Red O.

Fig. 1. Epitope analysis of hMSC. Representative flow cytometric analysis

of cultured MSCs derived from adipose tissue (left panel) and bone marrow

(right panel) with monoclonal antibodies against CD45 (A), CD34 (B),

CD73 (C), CD90 (D) and CD105 (E). Dashed lines indicate ISO type-

matched mouse IgG antibody controls staining. All MSCs were negative for

CD45 and positive for CD73, CD90 and CD105. Adipose-tissue-derived

MSCs were positive for CD34, whereas bone-marrow-derived MSCs were

negative.

A. Dicker et al. / Experimental Cell Research xx (2005) xxx–xxx4

preadipocytes and adipocytes derived from bone-marrow-

derived MSCs (data not shown).

Adipose-tissue-derived hMSC adipocytes display lipolysis

similar to that of bone-marrow-derived hMSCs

A hallmark of adipocyte differentiation is the ability to

release stored TG through lipolysis. The lipolytic capacity

of adipose-tissue-derived hMSC was determined by assess-

ing glycerol release in the medium after stimulation with

lipolytic agents and expressing it relative to that observed in

medium alone (basal lipolysis), i.e. responsiveness. A 3-h

ARTICLE IN PRESS

Fig. 4. Secretion of adipocyte specific proteins. Leptin (A) and adiponectin

(B) release was measured and corrected for the incubation time in

conditioned media from adipose-tissue-derived hMSC adipocytes (r) and

BM-derived hMSC-adipocytes (o).

A. Dicker et al. / Experimental Cell Research xx (2005) xxx–xxx 5

incubation with 0.1 mM adrenalin (Adr) resulted in a small

but non-significant increase in lipolysis compared with

control cells (167 T 33% of basal lipolysis, Fig. 3A). The

weak effect of the endogenous catecholamine appeared to

be due to a potent stimulation of the antilipolytic a2A-

adrenoceptor (AR) since co-incubation with the a2-AR

antagonist yohimbine (Y) (0.1 mM) increased lipolysis

substantially (1561 T 1039%, P = 0.0148, Fig. 3A). The

lipolytic capacity of these cells was compared with that of

adipocytes differentiated from hMSCs of adult bone marrow

(BM). In agreement with previous studies [5], the a2-AR

effect was even more pronounced in these cells since

adrenalin reduced lipolysis (51.4 T 7.6%, Fig. 3B). Co-

incubation with yohimbine resulted in a significant increase

in lipolysis (449 T 64%, P = 0.0073, Fig. 3B). To correct for

differences in cell number and differentiation, the lipolytic

rate was also corrected for the activity of the adipogenic

enzyme glycerol-3-phosphate dehydrogenase. However, this

did not significantly alter the results.

Adipose-tissue-derived hMSC adipocytes secrete leptin and

adiponectin

The secretion of the fat cell-specific proteins leptin and

adiponectin was assessed in various passages from differ-

entiated adipocytes of adipose-tissue-derived hMSCs and

compared with that observed in adipocytes from BM-

Fig. 3. Lipolysis in differentiated adipocytes from hMSCs. Glycerol release

in the medium was measured in differentiated adipocytes from adipose-

tissue-derived hMSCs (A) or BM-derived hMSCs (B). Basal lipolysis was

set to 100% for each experiment, which was repeated at least four times

with cells from different donors. Note that adrenaline-induced lipolysis is

significantly augmented in the presence of yohimbine due to the strong

antilipolytic effect of a2A-AR. (*statistically significant over basal lipolysis

P < 0.05).

derived hMSCs. Both proteins were released into the

medium in a time-dependent manner (Fig. 4). Secretion

was expressed as release per 24 h. Adipose-tissue-derived

hMSCs released leptin at levels 2–4 times higher than that

of bone-marrow-derived hMSCs. In contrast, adiponectin

release was similar in the two cell types.

Fig. 5. Adipocyte differentiation of hMSC derived from adipose tissue is

constant through multiple passages. (A) GPDH activity was measured from

cells at different passages. Experiments were performed on cells from the

same subject, and a representative example from one of the hMSC lines is

depicted. (B) Glycerol release was assessed after stimulation with adrena-

line (N) or adrenaline + yohimbine (n) and expressed relative to that of

unstimulated cells (basal) incubated in parallel. Experiments were

performed on cells from the same subject at the indicated passages.

ARTICLE IN PRESSA. Dicker et al. / Experimental Cell Research xx (2005) xxx–xxx6

Adipose-tissue-derived hMSCs retain adipocyte

differentiation capacity through multiple passages

The functional differentiation capacity of hMSCs ob-

tained from adipose tissue was tested by analyzing GPDH

activity and lipolytic capacity in cells from several passages.

A representative example is depicted in Fig. 5, demonstrat-

ing that GPDH activity (Fig. 5A) and stimulated lipolysis

(Fig. 5B, expressed as glycerol release in adrenaline- or

adrenaline+yohimbine-treated cells, respectively over con-

trol cells) were similar for at least fifteen passages. Thus,

adrenaline, a natural catecholamine stimulated spontaneous

(basal) lipolysis approximately 1- to 3-fold in cells from

each passage. The addition of the a2-adrenoceptor inhibitor

yohimbine (which counteracts the antilipolytic effect of

catecholamines) augmented adrenaline-stimulated lipolysis

2- to 4-fold in each experiment. It is well-established that a

pronounced a2-adrenoceptor activity is characteristic of

human but not murine adipocytes [5].

Discussion

In this work, we have assessed the differentiation

capacity of hMSCs derived from the stromal compartment

of adult adipose tissue. In accordance with recent studies

from other groups, we show that these cells display

multipotency and can be differentiated into several different

lineages. However, in contrast to other reports, we have also

determined the function of these cells induced into the

adipogenic lineage. We chose to study the function of

adipocytes since, in contrast to osteocytes and chondrocytes,

there are well-established assays to study these cells and

determine their functional capacity in vitro. We demonstrate

that fat cells developed from these immature precursors are

fully functional and virtually indistinguishable from adipo-

cytes obtained from bone-marrow-derived hMSCs. More-

over, their functional differentiation capacity is constant

through multiple passages. This would suggest that hMSCs

in adult adipose tissue of humans are functionally identical

to those obtained from bone marrow aspirates. This implies

that adipose tissue could constitute an easily accessible

source to harvest hMSC and that these cells constitute an

efficient and virtually endless source of in vitro differ-

entiated adipocytes from a single donor.

The presence of immature precursors in adult human

tissue is interesting from several points of view and could

benefit several areas of research. In particular, human

adipocyte research is hampered by the lack of human fat

cell lines, and most studies are therefore performed on

immortalized rodent cell lines such as 3T3-L1 or primary

cultures of human preadipocytes. The former are murine

cells that display several distinct metabolic and signal

transduction differences compared with human cells [16].

Preadipocytes are fibroblast-like cells isolated from the

stromal portion of human fat tissue which are differentiated

in vitro into cells displaying functional traits identical to

those of mature fat cells [14, 18]. However, these cultures

are set up by plating the entire stromal compartment and are

therefore often hampered by contamination by other cell

types (endothelial cells, fibroblasts, smooth muscle cells,

etc.). Furthermore, since these cells are immediately differ-

entiated (in order to minimize the proliferation of contam-

inating cells), they cannot be expanded and display only a

limited life span. Adipocyte research is now focusing on

molecular and cellular mechanisms underlying obesity and

its consequences. A better understanding of these processes

requires large amounts of cells with a metabolism displaying

high homology to mature human adipocytes. Human MSCs

provide us with an optimal renewable system to study

mature adipocyte phenotypes as well as cell differentiation

steps including the possibility to assess points of differential

cell lineage determination. We show that fat-tissue-derived

hMSCs retain their capacity to differentiate into fat cells

(GPDH activity) for numerous passages and that the newly

developed adipocytes retain their human-specific lipolytic

properties (characterized by a pronounced a2-adrenoceptor

effect that can be counteracted by the receptor-specific

inhibitor yohimbine).

Fat cell lipolysis is a hallmark of developed adipocytes,

and lipolytic capacity in vitro is predictive of in vivo

lipolysis [19]. Lipolysis is dysregulated in several common

disorders of insulin resistance in man, including obesity,

dyslipidemia, type 2 diabetes, the metabolic syndrome and

polycystic ovarian syndrome [20]. Studies in the last

decades have shown that the close relationship between

lipolysis and insulin resistance is due to the elevated release

of free fatty acids (FFA) to the circulation. FFA have

multiple effects, visceral lipolysis leads to FFA release into

the draining portal vein with direct effects on the liver

leading to increased output of glucose and VLDL trigly-

cerides. Increased FFA release from subcutaneous adipose

tissue affects muscle (decreased glucose uptake and insulin

sensitivity) and probably also negatively affects insulin

release from the pancreas (for a review see [20]). The

pathogenic processes that influence fat cell lipolysis in these

conditions are not understood, and it has been difficult to

determine whether the observed lipolytic dysregulation is of

primary or secondary origin. The fibrous stroma which

surrounds fat cells has recently attracted much attention. In

obesity, the stromal compartment is lined with inflammatory

cells that interact with and influence adipocyte function

[21]. Moreover, obesity (as well as other insulin resistant

conditions) develops slowly in humans, and prospective

studies of fat cell function are therefore not feasible. In order

to circumvent these issues, we have previously set up

primary preadipocyte cultures from obese and non-obese

subjects. These cells have not yet been adipocytes in vivo,

and they are differentiated in vitro under standardized

conditions. By comparing these in vitro generated adipo-

cytes with freshly isolated mature fat cells from the same

subject, we have been able to differentiate between primary

ARTICLE IN PRESSA. Dicker et al. / Experimental Cell Research xx (2005) xxx–xxx 7

and secondary defects in obese subjects [14]. However, in

order to obtain reproducible results, extensive cell culture

work has to be performed. This is mainly due to the

interindividual differences between cells from different

subjects and the varying cellular contamination discussed

previously. Other disadvantages of this primary cell culture

system are that cell numbers cannot be increased effectively

by proliferation, thereby limiting the expandability of the

relatively small tissue samples and that differentiated cells

cannot be frozen for subsequent follow-up studies. In this

respect, fat-derived hMSCs have important advantages since

they constitute a homogenous proliferating cell population

that can be propagated and cryopreserved. This enables

researchers to study hMSC-derived adipocytes and build up

fat cell banks from subjects with different phenotypic

alterations and then compare these processes in vitro.

Hopefully, this will enable scientists in the metabolic field

to determine whether defects in lipolysis (and possibly other

traits such as lipogenesis and glucose uptake) are of primary

or secondary origin in several common conditions.

The presence of a small proportion of multipotent

progenitor cells in the stroma of peripheral organs suggests

that a cellular reserve exists which can be activated upon

tissue damage. However, although animal models have

shown that MSCs can be recruited in different damage

models, to date, there are few reports in humans demonstrat-

ing that these progenitor cells are of clinical value, although

in a recent case report, adipose-tissue-derived MSCs were

successfully used in the treatment of a rectovaginal fistula in

a patient with Crohn’s disease [22]. The role of these cells is

therefore not entirely understood, are they dormant remains

from earlier developmental stages or do they continue to

proliferate at a slow pace, replacing a proportion of

malfunctioning or degenerating cells? Further studies to

elucidate the importance of these cells are warranted.

In summary, we have demonstrated that the MSCs

present in adult adipose tissue can be differentiated into

fat cells which appear to be functionally indistinguishable

from adipocytes derived from bone marrow harvested

hMSCs. Since these adipose-tissue-derived hMSCs retain

their ability to differentiate into fat cells with intact cell-

specific characteristics for numerous passages, they con-

stitute an almost endless source of adipocytes for exper-

imental research on cells from a single donor. Given the

increasing research on white adipose tissue and its role in

development of insulin resistance and other common

disorders, this cellular system is a valuable addition to

researchers in the experimental metabolic field.

Acknowledgments

This work was supported by the foundations of Ake

Wiberg, Tore Nilsson, Signe and Olof Wallenius, Magnus

Bergvall and Novo Nordisk, the Swedish Research Council,

the Swedish Diabetes Association, the Tobias Foundation,

the Swedish Cancer Society, the Children’s Cancer Foun-

dation, the Stockholm Cancer Society, the Swedish Society

of Medicine, and Karolinska Institutet.

References

[1] M.F. Pittenger, A.M. Mackay, S.C. Beck, R.K. Jaiswal, R. Douglas,

J.D. Mosca, M.A. Moorman, D.W. Simonetti, S. Craig, D.R. Marshak,

Multilineage potential of adult human mesenchymal stem cells,

Science 284 (1999) 143–147.

[2] F.P. Barry, J.M. Murphy, Mesenchymal stem cells: clinical applica-

tions and biological characterization, Int. J. Biochem. Cell Biol. 36

(2004) 568–584.

[3] P.A. Zuk, M. Zhu, H. Mizuno, J. Huang, J.W. Futrell, A.J. Katz, P.

Benhaim, H.P. Lorenz, M.H. Hedrick, Multilineage cells from human

adipose tissue: implications for cell-based therapies, Tissue Eng. 7

(2001) 211–228.

[4] P. Arner, The adipocyte in insulin resistance: key molecules and the

impact of the thiazolidinediones, Trends Endocrinol. Metab. 14 (2003)

137–145.

[5] M. Ryden, A. Dicker, C. Gotherstrom, G. Astrom, C. Tammik, P.

Arner, K. Le Blanc, Functional characterization of human mesenchy-

mal stem cell-derived adipocytes, Biochem. Biophys. Res. Commun.

311 (2003) 391–397.

[6] D.A. De Ugarte, K. Morizono, A. Elbarbary, Z. Alfonso, P.A. Zuk, M.

Zhu, J.L. Dragoo, P. Ashjian, B. Thomas, P. Benhaim, I. Chen, J.

Fraser, M.H. Hedrick, Comparison of multi-lineage cells from human

adipose tissue and bone marrow, Cells Tissues Organs 174 (2003)

101–109.

[7] F. Guilak, H.A. Awad, B. Fermor, H.A. Leddy, J.M. Gimble, Adipose-

derived adult stem cells for cartilage tissue engineering, Biorheology

41 (2004) 389–399.

[8] Y.D. Halvorsen, A. Bond, A. Sen, D.M. Franklin, Y.R. Lea-Currie, D.

Sujkowski, P.N. Ellis, W.O. Wilkison, J.M. Gimble, Thiazolidine-

diones and glucocorticoids synergistically induce differentiation of

human adipose tissue stromal cells: biochemical, cellular, and

molecular analysis, Metabolism 50 (2001) 407–413.

[9] H. Hattori, M. Sato, K. Masuoka, M. Ishihara, T. Kikuchi, T.

Matsui, B. Takase, T. Ishizuka, M. Kikuchi, K. Fujikawa,

Osteogenic potential of human adipose tissue-derived stromal cells

as an alternative stem cell source, Cells Tissues Organs 178 (2004)

2–12.

[10] R.H. Lee, B. Kim, I. Choi, H. Kim, H.S. Choi, K. Suh, Y.C. Bae, J.S.

Jung, Characterization and expression analysis of mesenchymal stem

cells from human bone marrow and adipose tissue, Cell. Physiol.

Biochem. 14 (2004) 311–324.

[11] A.-M. Rodriguez, C. Elabd, F. Delteil, J. Astier, C. Vernochet, P.

Saint-Marc, J. Guesnet, A. Guezennec, E.-Z. Amri, C. Dani, G.

Ailhaud, Adipocyte differentiation of multipotent cells established

from human adipose tissue, Biochem. Biophys. Res. Commun. 315

(2004) 255–263.

[12] A. Sen, Y.R. Lea-Currie, D. Sujkowska, D.M. Franklin, W.O.

Wilkison, Y.D. Halvorsen, J.M. Gimble, Adipogenic potential of

human adipose derived stromal cells from multiple donors is

heterogeneous, J. Cell Biochem. 81 (2001) 312–319.

[13] M. Rodbell, Metabolism of isolated fat cells: I. Effects of hormones

on glucose metabolism and lipolysis, J. Biol. Chem. 239 (1964)

375–380.

[14] V. van Harmelen, A. Dicker, M. Ryden, H. Hauner, F. Lonnqvist, E.

Naslund, P. Arner, Increased lipolysis and decreased leptin production

by human omental as compared with subcutaneous preadipocytes,

Diabetes 51 (2002) 2029–2036.

[15] J. Hellmer, C. Marcus, T. Sonnenfeld, P. Arner, Mechanisms for

differences in lipolysis between human subcutaneous and omental fat

cells, J. Clin. Endocrinol. Metab. 75 (1992) 15–20.

ARTICLE IN PRESSA. Dicker et al. / Experimental Cell Research xx (2005) xxx–xxx8

[16] M. Ryden, A. Dicker, V. van Harmelen, H. Hauner, M. Brunnberg, L.

Perbeck, F. Lonnqvist, P. Arner, Mapping of early signaling events in

tumor necrosis factor-alpha-mediated lipolysis in human fat cells,

J. Biol. Chem. 277 (2002) 1085–1091.

[17] A.C. Boquest, A. Shahdadfar, K. Fronsdal, O. Sigurjonsson, S.H.

Tunheim, P. Collas, J.E. Brinchmann, Isolation and transcription

profiling of purified uncultured human stromal stem cells: alteration of

gene expression after in vitro cell culture, Mol. Biol. Cell 16 (2005)

1131–1141.

[18] H. Hauner, G. Entenmann, M. Wabitsch, D. Gaillard, G. Ailhaud, R.

Negrel, E.F. Pfeiffer, Promoting effect of glucocorticoids on the

differentiation of human adipocyte precursor cells cultured in a

chemically defined medium, J. Clin. Invest. 84 (1989) 1663–1670.

[19] M. Kolehmainen, J.J. Ohisalo, J.M. Kaartinen, V. Tuononen, M.

Paakkonen, E. Poikolainen, E. Alhava, M.I. Uusitupa, Concordance of

in vivo microdialysis and in vitro techniques in the studies of adipose

tissue metabolism, Int. J. Obes. Relat. Metab. Disord. 24 (2000)

1426–1432.

[20] P. Arner, Insulin resistance in type 2 diabetes: role of fatty acids,

Diabetes/Metab. Res. Rev. 18 (Suppl. 2) (2002) S5–S9.

[21] H. Xu, G.T. Barnes, Q. Yang, G. Tan, D. Yang, C.J. Chou, J. Sole, A.

Nichols, J.S. Ross, L.A. Tartaglia, H. Chen, Chronic inflammation in

fat plays a crucial role in the development of obesity-related insulin

resistance, J. Clin. Invest. 112 (2003) 1821–1830.

[22] D. Garcia-Olmo, M. Garcia-Arranz, L.G. Garcia, E.S. Cuellar,

I.F. Blanco, L.A. Prianes, J.A. Montes, F.L. Pinto, D.H.

Marcos, L. Garcia-Sancho, Autologous stem cell transplantation

for treatment of rectovaginal fistula in perianal Crohn’s disease:

a new cell-based therapy, Int. J. Colorectal Dis. 18 (2003)

451–454.