| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Wyeth Research, Radnor, Pennsylvania 19087; and Department of Surgery, Downstate Medical Center (J.G.K.), Brooklyn, New York 11203
Address all correspondence and requests for reprints to: Dr. David L. Crandall, Wyeth Research, P.O. Box 42528, Philadelphia, Pennsylvania 19101. E-mail: crandad{at}war.wyeth.com
| Abstract |
|---|
|
|
|---|
Vß3, we
identified this receptor in human preadipocytes. Flow cytometric
analysis indicated that human preadipocytes express the vitronectin
receptor
Vß3 in a similar pattern as human
umbilical vein endothelial cells. Functional studies indicated that
active, but not latent, PAI-1 inhibited preadipocyte attachment to
vitronectin with an IC50 of 13.3 nmol/L, and preincubation
of vitronectin-coated Transwells with active PAI-1 prevented
preadipocyte migration. Vitronectin was identified in homogenates of
the stromal-vascular fraction of human adipose tissue, but was absent
from human adipocytes and cultured preadipocytes. These data indicate
that human preadipocyte migration is regulated through the
endogenous expression of PAI-1 and
Vß3
integrin, a novel autocrine mechanism for potentially regulating cell
cluster formation in adipogenesis. | Introduction |
|---|
|
|
|---|
More recent investigations have identified another role for PAI-1 in
cell migration and angiogenesis. PAI-1 has been shown to affect cell
migration by competing for a specific integrin-binding site on
vitronectin (7), and PAI-1 knockout mice are unable to vascularize
tumors (8). The integrin for which PAI-1 competes,
Vß3, has been shown to
be critical to developmental angiogenesis (9), and PAI-1 binding to the
extracellular matrix may function to regulate local pericellular
proteolysis central to tissue remodeling. Additionally, PAI-1 knockout
mice crossed with genetically obese mice weigh significantly less than
wild-type obese mice (10), although the mechanism of action of this
effect is unknown.
Obesity is a disease characterized by an increased fat mass, but because of the belief that the number of fat cells is fixed in adulthood (11), studies of adipose tissue growth have focussed almost exclusively on the enlargement of the adipocyte once the tissue matrix is established. Importantly, however, studies of the relationship between fat mass and adipocyte number in humans clearly indicate the potential for adipocyte hyperplasia in adulthood in the range of billions of fat cells (12, 13). Similar observations of adipose tissue growth through adipocyte hyperplasia are documented for animal models of obesity (14).
Investigations into the early stages of human adipose tissue development are primarily descriptive anatomical studies in human fetuses. The process of adipose tissue organogenesis is characterized by the formation of a vascularized primitive fat organ, followed by migration of preadipocytes into defined fat cell clusters (15), but the molecules regulating this process are unknown. The current study investigates novel aspects of cell physiology in human adipose tissue primary cultures, focussing specifically on identifying the various components required to establish PAI-1 as a regulatory protein in human preadipocyte migration, potentially contributing to adipose tissue growth.
| Materials and Methods |
|---|
|
|
|---|
Human preadipocytes were cultured from sc biopsies as previously described in detail (16). Cultures were maintained in a growth medium consisting of medium 199, 10% heat-inactivated FCS, and 1% antibiotic-antimycotic (Life Technologies, Inc., Grand Island, NY) at 37 C in 5%CO2. The human preadipocytes stain negative for factor VIII, and when stimulated to differentiate into the adipocyte phenotype, they express leptin mRNA and increased staining for triglyceride (16).
Human adipose tissue homogenates
Subcutaneous or mesenteric (visceral) adipose tissue was obtained by knife biopsy or suction-assisted sc lipectomy from men and women varying in age between 2275 yr, with body mass indexes (BMIs) of 1942 kg/m2. None of the individuals was diabetic or had undergone significant weight loss preoperatively. Samples were quickly frozen and maintained at -80 C until analysis. Both types of samples were thawed the day of the experiment and subjected to collagenase digestion and selective filtration (16). For liposuction samples, this procedure resulted predominantly in an adipocyte fraction only, whereas for knife biopsies, a significant stromal-vascular fraction was obtained as well. After washing with sterile phosphate-buffered saline (PBS), tissue fractions were homogenized briefly in a lysis buffer consisting of 1 x PBS, 1% Triton, and 1% Nonidet P-40 (Sigma, St. Louis, MO) and spun at 10,000 x g for 15 min, and the supernatant was used for determination of vitronectin content.
Identification of components of the PAI-1 migration regulatory pathway
Primary cultures of human preadipocytes were transferred to 24-well plates in passages 24, and after attachment, fresh medium containing 10% serum was added, with PAI-1 antigen measured in the conditioned medium 24 h later by enzyme-linked immunosorbent assay (American Diagnostica, Greenwich, CT).
Total RNA was prepared from cultured human preadipocytes using an adaptation of the guanidinium thiocyanate extraction method (RNeasy, QIAGEN, Chatsworth, CA). RT-PCR was performed for identification of PAI-1 mRNA (17), with RT-PCR of mRNA for human ß2-microglobulin (CLONTECH Laboratories, Inc., Palo Alto, CA) determined from the same complementary DNA. In addition to cultured preadipocytes, adipocyte RNA was obtained from the adipocyte fraction of human adipose tissue after collagenase digestion, and RT-PCR was employed for the identification of vitronectin mRNA using the procedure described by Seiffert et al. (18). Human liver RNA was obtained from CLONTECH Laboratories, Inc. and served as the standard for vitronectin mRNA identification.
Identification of the
Vß3 integrin on human
preadipocytes used flow cytometric analysis. For flow cytometry,
preadipocytes were detached using cell dissociation buffer (Life Technologies, Inc., Gaithersburg, MD). Cell suspensions were
pelleted, washed with 10 mL HBSS, and resuspended in PBS containing 2%
heat-inactivated FBS. Cell suspensions (50 µL; 5 x
105 cells) were stained using 50 µL
anti-
Vß3 monoclonal
antibody (CD51/61, PharMingen, San Diego, CA) diluted to
10 µg/mL, followed by staining with a 1:100 dilution of goat
antimouse secondary antibody directly conjugated to phycoerythrin
(Immunotech, Westbrook, ME). Cells were resuspended in 2
µg/mL propidium iodide/PBS immediately before flow cytometric
analysis. Collection criteria based on propidium iodide exclusion were
used to acquire data from 5000 vital cells/sample (FACSCalibur flow
cytometer, Becton Dickinson and Co., San Jose, CA). Human
preadipocytes stained with the isotypic (IgG1
) antibody
(Sigma) served as controls.
Identification and quantification of vitronectin
For identification of vitronectin in cultured cells and in adipose tissue biopsies we employed a capture antibody assay for native vitronectin. A 96-well plate was coated overnight at 4 C with 100 µL of a 10 µg/mL solution of vitronectin capture antibody MA-16A7 (Molecular Innovations, Inc., Royal Oak, MI). The plate was washed 3 times with Tris-buffered saline Triton (0.05%) (TBST) and blocked for 1 h with 300 µL blocking buffer (3% albumin/TBST). The blocking buffer was aspirated, and 25 µL sample (cultured preadipocyte or adipose tissue fraction homogenate) or standard were added to the plate (standards of human native vitronectin from 0.5200 ng/mL). The plates were incubated for 30 min at 25 C with gentle shaking, followed by washing 3 times in TBST. A 100-µL solution of rabbit antihuman vitronectin polyclonal antibody diluted 1:1,000 in blocking buffer was added, followed by incubation for 30 min with gentle shaking. After washing 3 times with TBST, 100 µL biotinylated goat anti-rabbit IgG (1:50,000 in Tris-buffered saline) was added to each well, then incubated for 30 min while shaking. After aspiration, 100 µL of a solution of avidin alkaline phosphatase (1:10,000 dilution in blocking buffer) were added, followed by a 30-min incubation. All wells were aspirated and washed 3 times with TBST, and 150 µL of a 0.1% PNPP solution were added. After incubation for 1 h, absorbance was measured at 405 nm. Fresh human citrated plasma was used to compare the vitronectin values obtained for adipose tissue to those in the circulation. In addition, human artery and vein samples were homogenized and used for determination of vitronectin in the vasculature.
Attachment and migration of human preadipocytes
Preadipocytes were trypsinized, washed in PBS, and resuspended in phenol red-free medium containing 0.5% heat-inactivated FCS. Preadipocytes (1 x 105/mL) were preincubated at room temperature for 15 min with varying concentrations (5100 nmol/L) of active human PAI-1 (Molecular Innovations) solubilized in 50 mmol/L phosphate buffer (pH 6.6). Cells were subsequently incubated for 1 h at 37 C on a vitronectin (Molecular Innovations)-coated (1 µg/mL) 96-well plate (7). At the end of the incubation period, the plate was rinsed with PBS, followed by assessment of cell attachment using a chromogenic reporter (CellTiter, Promega Corp., Madison, WI). The percentage of attached cells was determined, as was the concentration of PAI-1 required for 50% inhibition of attachment. In a separate set of experiments, preadipocyte migration assays were performed using a Transwell system with a 3-µm pore size in a 6.5-mm insert (Costar, Cambridge, MA). The insert was coated with human vitronectin as previously described by Stefansson and Lawrence (7). For determination of the effect of PAI-1 on cell attachment, either 1 µmol/L active or latent PAI-1 (pH 6.6) was added to the vitronectin-coated insert for 15 min at 37 C, followed by the addition of 50,000 human preadipocytes suspended in medium 199/0.5% FCS. The insert was placed in a tissue culture well containing medium 199/10% FCS, creating a chemotactic serum gradient compared to the insert. The culture was incubated at 37 C for 1 h, the insert was removed, and cells were fixed and stained with Hema 3 stain (Fisher Scientific, Pittsburgh, PA). After microscopic imaging, the number of cells that had migrated through the Transwell were quantitated using imaging software (Noesis Vision, Inc., Montréal, Canada).
Statistical analysis
Statistical analysis was performed using the Kruskal-Wallis analysis for multiple groups. Group means ± SEM were considered significantly different at P < 0.05.
| Results |
|---|
|
|
|---|
|
Vß3
integrin with the extracellular matrix component vitronectin, but
whether human preadipocytes or adipocytes express specific cellular
adhesion receptors or synthesize vitronectin is not known. Therefore,
it was necessary to identify these molecules in adipose tissue before
pursuing functional studies of PAI-1 in preadipocytes. Flow
cytometric data (Fig. 2
Vß3 integrin, a novel
finding with respect to adipose tissue physiology. The initial search
for vitronectin in adipose tissue employed assessment of vitronectin
mRNA from cultured preadipocytes and adipose tissue fractions using
RT-PCR. As shown in Fig. 3
|
|
|
Vß3 integrin, and
vitronectin in components of human adipose tissue, two separate assays
were performed to determine the effects of PAI-1 on preadipocyte
attachment and migration. In the first assay, human preadipocytes from
three separate cultures were preincubated with varying concentrations
of PAI-1, followed by seeding onto a vitronectin-coated plate. PAI-1
inhibited attachment in a concentration-dependent manner, with an
estimated IC50 of 13.3 nmol/L (Fig. 4
|
|
|
| Discussion |
|---|
|
|
|---|
Although experimental evidence suggests that human adipose tissue is a
significant source of PAI-1, the physiological relevance of the
synthesis of a fibrinolytic regulatory protein by adipose tissue
remains an enigma. Because PAI-1 affects cell migration in smooth
muscle, epithelial, and endothelial cells (7, 9, 25), we investigated
the role of PAI-1 in the motility of the human adipocyte precursor
cell, the preadipocyte. Our results clearly indicate that human
preadipocytes exhibit the ability to migrate, and that this process is
regulated by PAI-1. In addition, this cell type expresses additional
components required for autocrine regulation of migration, including
the membrane-associated
Vß3 integrin. The
mediation by PAI-1 of the interaction between the membrane integrin and
the matrix vitronectin is a high affinity process that is dependent
upon the structural conformation of the serpin, as indicated by an
IC50 in the low nanomolar range for active PAI-1,
and the inactivity of latent PAI-1. As PAI-1 has been reported to reach
local concentrations several thousand-fold greater than the
IC50 (26), the effect on preadipocyte migration
may have physiological relevance.
The final component required for PAI-1-mediated migration, vitronectin,
is an adhesive extracellular matrix glycoprotein that binds PAI-1 with
high affinity (27). The liver is the predominant organ for vitronectin
synthesis, whereas the involvement of adipose tissue is controversial,
as mRNA has been detected in murine adipose tissue by PCR, but not by
Northern analysis (28). It is important to note, however, that
vitronectin gene expression or protein identification has never been
performed in human adipose tissue or its cellular components. Using
PCR, we were unable to identify vitronectin mRNA in either cultured
preadipocytes or adipocytes isolated from tissue biopsies, although we
did observe gene expression in human liver. However, we identified
significant quantities of vitronectin in homogenates of the
stromal-vascular fraction of both visceral and sc adipose tissue. Using
plasma (29) and vessel samples for comparison, these studies are the
first to identify vitronectin in human adipose tissue, further
localizing it to the stromal-vascular compartment. It is a reasonable
assumption that vitronectin is transported to adipose tissue via the
circulation, where it is bound by its receptor,
Vß3 integrin, and
PAI-1.
In obesity, body fat mass increases through the enlargement of existing fat cells, but in gross obesity there is also an increase in cell number (12, 13). Classic studies on the development of human adipose tissue indicate that matrix deposition is followed by vascular growth within the matrix (15). Primitive fat organs, composed of mesenchymal cells, then localize and spread along established capillaries, eventually filling with lipid (15). Therefore, although preadipocyte migration is a prerequisite for adipose tissue development, the regulatory mechanisms have not been extensively studied. Our present findings indicate a functional role for PAI-1 in preadipocyte migration and expand upon previous studies from our laboratory that observed reduced preadipocyte PAI-1 synthesis once confluence was attained and active migration had ceased (23).
PAI-1 has been proposed to be involved in vascularizing murine tumors
(8), and the vitronectin receptor,
Vß3 integrin, has been
identified as important to angiogenesis, vascular remodeling, and
developmental neovascularization (30), albeit in tissues other than
adipose. PAI-1 can also affect cell surface proteolysis by inhibiting
urokinase plasminogen activator (31), but whether this role exists in
adipose tissue development has not been investigated. As adipose tissue
is one of the most highly vascularized tissues (32), and adipogenesis
proceeds along clearly established capillaries, the identification of
molecules in adipose tissue that regulate cell migration and
angiogenesis may be important for understanding the interaction of
adipocytes with the stromal-vascular component of adipose tissue.
Continued synthesis of PAI-1 by terminally differentiated adipocytes
and the elevated plasma PAI-1 found in obese humans may therefore
reflect the role of PAI-1 in adipose tissue development, which is
triggered in adult obesity as a response to adipose tissue expansion.
As the signals responsible for the early stages of adipogenesis remain
largely unknown, the production by primordial fat cells of significant
levels of a molecule capable of regulating migration, angiogenesis, and
local proteolysis warrants further investigation.
| Acknowledgments |
|---|
Received December 9, 1999.
Revised February 26, 2000.
Accepted March 9, 2000.
| References |
|---|
|
|
|---|
vß3 binding to
vitronectin. Nature. 383:441443.[CrossRef][Medline]
v integrins during angiogenesis: insights into
potential mechanisms of action and clinical development. J Clin
Invest. 103:12271230.[Medline]
This article has been cited by other articles:
![]() |
N. V. Gorlatova, J. M. Cale, H. Elokdah, D. Li, K. Fan, M. Warnock, D. L. Crandall, and D. A. Lawrence Mechanism of Inactivation of Plasminogen Activator Inhibitor-1 by a Small Molecule Inhibitor J. Biol. Chem., March 23, 2007; 282(12): 9288 - 9296. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. L. Crandall, E. M. Quinet, S. El Ayachi, A. L. Hreha, C. E. Leik, D. A. Savio, I. Juhan-Vague, and M.-C. Alessi Modulation of Adipose Tissue Development by Pharmacological Inhibition of PAI-1 Arterioscler. Thromb. Vasc. Biol., October 1, 2006; 26(10): 2209 - 2215. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. G. Gosman, H. I. Katcher, and R. S. Legro Obesity and the role of gut and adipose hormones in female reproduction Hum. Reprod. Update, September 1, 2006; 12(5): 585 - 601. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. J. Hausman, S. P. Poulos, R. L. Richardson, C. R. Barb, T. Andacht, H. C. Kirk, and R. L. Mynatt Secreted proteins and genes in fetal and neonatal pig adipose tissue and stromal-vascular cells J Anim Sci, July 1, 2006; 84(7): 1666 - 1681. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Chowers, Y. Kim, R. H. Farkas, T. L. Gunatilaka, A. S. Hackam, P. A. Campochiaro, S. C. Finnemann, and D. J. Zack Changes in Retinal Pigment Epithelial Gene Expression Induced by Rod Outer Segment Uptake Invest. Ophthalmol. Vis. Sci., July 1, 2004; 45(7): 2098 - 2106. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Urs, C. Smith, B. Campbell, A. M. Saxton, J. Taylor, B. Zhang, J. Snoddy, B. Jones Voy, and N. Moustaid-Moussa Gene Expression Profiling in Human Preadipocytes and Adipocytes by Microarray Analysis J. Nutr., April 1, 2004; 134(4): 762 - 770. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. J. Hausman and R. L. Richardson Adipose tissue angiogenesis J Anim Sci, March 1, 2004; 82(3): 925 - 934. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Yokota, C. S. R. Meka, T. Kouro, K. L. Medina, H. Igarashi, M. Takahashi, K. Oritani, T. Funahashi, Y. Tomiyama, Y. Matsuzawa, et al. Adiponectin, a Fat Cell Product, Influences the Earliest Lymphocyte Precursors in Bone Marrow Cultures by Activation of the Cyclooxygenase-Prostaglandin Pathway in Stromal Cells J. Immunol., November 15, 2003; 171(10): 5091 - 5099. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |