Am J Physiol Endocrinol Metab 295: E313–E322, 2008.

First published May 20, 2008; doi:10.1152/ajpendo.90296.2008.

Macrophage infiltration into adipose tissue may promote angiogenesis

for adipose tissue remodeling in obesity
Can Pang,1 Zhanguo Gao,2 Jun Yin,2 Jin Zhang,2 Weiping Jia,1 and Jianping Ye2
1
Department of Endocrinology and Metabolism, Shanghai Jiaotong University Affiliated Sixth People’s Hospital, Shanghai
Diabetes Institute, Shanghai Clinical Center of Diabetes, Shanghai, China; and 2Pennington Biomedical Research Center,
Louisiana State University, Baton Rouge, Louisiana
Submitted 14 March 2008; accepted in final form 13 May 2008

Pang C, Gao Z, Yin J, Zhang J, Jia W, Ye J. Macrophage cell growth factor (PD-ECGF), transforming growth factor
infiltration into adipose tissue may promote angiogenesis for adi- (TGF)-␤, and angiopoietin. The functions of these factors are
pose tissue remodeling in obesity. Am J Physiol Endocrinol Metab demonstrated in experimental angiogenesis assays with condi-
295: E313–E322, 2008. First published May 20, 2008; tioned media obtained from preadipocytes and tissue homog-
doi:10.1152/ajpendo.90296.2008.—The biological role of macro-
enates of fat tissues (8, 17, 44). The factor that induces
phage infiltration into adipose tissue in obesity remains to be fully
understood. We hypothesize that macrophages may act to stimulate expression of the angiogenic factors or vascular compensation
angiogenesis in the adipose tissue. This possibility was examined by remains to be identified in adipose tissue. In a recent study, we
determining macrophage expression of angiogenic factor PDGF demonstrated hypoxia in the adipose tissue in ob/ob and dietary
(platelet-derived growth factor) and regulation of tube formation of obese mice (56). Adipose tissue hypoxia may play a role in the
endothelial cells by PDGF. The data suggest that endothelial cell induction of these angiogenic factors. Cross-talk between adi-
density was reduced in the adipose tissue of ob/ob mice. Expression pocytes and endothelial cells has been supported by many
of endothelial marker CD31 was decreased in protein and mRNA. The studies and is required for adipose tissue growth (8, 27, 34, 50).
reduction was associated with an increase in macrophage infiltration. However, not much is known about cross-talk between mac-
In the obese mice, PDGF concentration was elevated in the plasma, rophages and endothelial cells in adipose tissue in tissue
and its mRNA expression was increased in adipose tissue. Macro- growth or remodeling.
phages were found to be a major source of PDGF in adipose tissue, as
Macrophage infiltration into adipose tissue contributes to the
deletion of macrophages led to a significant reduction in PDGF
mRNA. In cell culture, PDGF expression was induced by hypoxia, increased expression of inflammatory cytokines in obesity (52,
and tube formation of endothelial cells was induced by PDGF. The 55). It remains to be investigated why macrophage infiltration
PDGF activity was dependent on S6K, as inhibition of S6K in is increased in adipose tissue. Some studies suggest that mac-
endothelial cells led to inhibition of the PDGF activity. We conclude rophage infiltration is for clearance of dead adipocytes in the
that, in response to the reduced vascular density, macrophages may adipose tissue (10, 45). Except for chronic inflammation and
express PDGF in adipose tissue to facilitate capillary formation in clearance of dead cells, the biological significance of macro-
obesity. Although the PDGF level is elevated in adipose tissue, its phage infiltration remains largely unknown in the adipose
activity in angiogenesis is dependent on the availability of sufficient tissue. In wound healing and tumor growth, macrophage infil-
endothelial cells. The study suggests a new function of macrophages tration contributes to the stimulation of angiogenesis (46).
in the adipose tissue in obesity. Such a role of macrophages remains to be established in
macrophage; platelet-derived growth factor; angiogenesis; ribosomal adipose tissue remodeling in obesity. We hypothesized that
protein S6 kinase; hypoxia; obesity macrophage infiltration might be involved in the stimulation of
angiogenesis during expansion of adipose tissue.
As a proangiogenic factor in serum, platelet-derived growth
TISSUE REMODELING IS INVOLVED in the expansion of adipose factor (PDGF) stimulates differentiation of endothelial cells
tissue during body weight gain. Angiogenesis is a critical event and migration of pericytes (21, 22, 29, 40). Although PDGF is
in tissue remodeling. Angiogenesis may be coupled with adi- secreted by many types of cells, including platelets, macro-
pogenesis during adipose tissue remodeling throughout life- phages, fibroblasts, and endothelial cells, macrophages are one
time (11, 28, 33, 34). The increase in adipocyte size is asso- of the major sources of PDGF in tissues (32, 43). Among the
ciated with compensation in the microcirculation, as has been three active isoforms of PDGF (AA, AB, and BB), PDGF-BB
reviewed (5, 11). Inhibition of angiogenesis prevents fat mass is able to activate all of the three PDGF receptors (␣␣, ␣␤, and
expansion in several rodent models of obesity (6, 41). In ␤␤), which are dimeric tyrosine kinases. The importance of
addition to the well-known adipokines (including adiponectin, PDGF in the regulation of vascular development and function
leptin, TNF-␣, and resistin), adipose tissue also produces a was demonstrated in gene knockout mice in which inactivation
variety of endothelial cell growth factors and angiogenic fac- of either PDGF-B or its receptor was embryonically lethal (29).
tors (11), such as fibroblast growth factor (FGF), vascular The mice died from hemorrhage, edema, and absence of kidney
endothelial growth factor (VEGF), platelet-derived endothelial glomerular mesangial cells. In addition to the regulation of
vascular development, PDGF is also involved in oncogenesis,
Address for reprint requests and other correspondence: J. Ye, Pennington atherosclerosis, lung fibrosis, kidney fibrosis, etc. (20). Al-
Biomedical Research Center, Louisiana State University, Baton Rouge, LA
70808 (e-mail: yej@pbrc.edu) or W. Jia, Dept. of Endocrinology and Metab-
olism, Shanghai Jiaotong University Affiliated Sixth People’s Hospital, Shang- The costs of publication of this article were defrayed in part by the payment
hai Diabetes Institute, Shanghai Clinical Center of Diabetes, Shanghai, China of page charges. The article must therefore be hereby marked “advertisement”
(e-mail: wpjia@yahoo.com). in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

http://www.ajpendo.org 0193-1849/08 $8.00 Copyright © 2008 the American Physiological Society E313
Downloaded from journals.physiology.org/journal/ajpendo at CAPES-USP (200.144.063.031) on August 23, 2021.
E314 REGULATION OF ADIPOSE TISSUE ANGIOGENESIS BY MACROPHAGES

though PDGF expression is elevated locally in tissue remod- liposome. Clodronate liposome was prepared and administration at
eling processes such as wound healing and tumor growth, it is 100 –150 mg/kg ip as described elsewhere (49). The macrophage
not clear whether PDGF expression is increased during adipose deletion was confirmed in adipose tissue at day 4 after injection.
tissue growth in obesity. As hypoxia induces PDGF expression Tube formation assay. Tube formation assay was conducted on the
(16), we propose that PDGF expression may be increased by Matrigel (35027, BD Biosciences), which was added in a volume of
50 ␮l/well to a 96-well plate (3585, Fisher Scientific) and allowed to
adipose tissue hypoxia and may be involved in stimulation of
polymerize at 37°C for 30 min. After polymerization, the endothelial
angiogenesis during adipose tissue remodeling in obesity. cells were plated on the Matrigel at 2 ⫻ 104 cells/well in 200 ␮l of
The angiogenic activities of PDGF are mediated by signal- serum-free medium with or without reagents. The cells were incu-
ing pathways of cell membrane receptors of PDGF (47). bated at 37°C with 95% humidity and 5% CO2. The tube formation
Ligand engagement leads to PDGF receptor phosphorylation was observed under an inverted microscope (Zeiss Axiovert 40 CFL)
and activation of several signaling pathways, including Src, after 10 h. Images were captured with a Zeiss AxioCam Hrc CCD
PI3K, and phospholipase C␥ (PLC␥) (47). The phophatidyli- camera attached to the microscope. The tube formation was quantified
nositol 3-kinase (PI3K)/Akt/mTOR (mammalian target of by measuring the long axis of the individual cells on Matrigel using
rapamycin) pathway is activated by PDGF (25, 58) and is NIH Image J (version 1.31). A mean value of total length in each
involved in the recruitment of pericytes (13). It remains to be sample was used to represent the tube formation.
tested whether the PI3K/Akt/mTOR pathway is involved in Protein extract and Western blot. SVEC4-10 cells (5 ⫻ 105/well)
tube formation of endothelial cells in response to PDGF. In were plated in a 12-well plate in DMEM supplemented with 10% FBS
for 24 h. Then the cells were starved overnight in serum-free medium
studies of other angiogenic factors, the PI3K/Akt/mTOR path-
and treated with various reagents for 30 min. Protein extraction and
way was reported to stimulate proliferation (51) and tube Western blot analysis were conducted for analysis of signaling activ-
formation of vascular endothelial cells (57). However, there is ities in cells, as described elsewhere (15). The intensity of Western
no direct evidence that this signaling pathway is required by blot signal was quantified with NIH Image J, and the signal was
PDGF in the stimulation of tube formation. normalized against loading control.
In this study, we demonstrated that vascular density was Generation of stable cell line with dominant-negative S6K mutant.
reduced in adipose tissue in ob/ob mice. PDGF expression was The SVEC4-10 endothelial cells were cotransfected with 3 ␮g of
elevated in adipose tissue and expressed in macrophages. In pRK7-HA-S6K1-KR plasmid for dominant-negative S6K (8985, Ad-
cell culture, PDGF stimulated tube formation of endothelial dgene) and 0.5 ␮g of pcDNA3.1 plasmid for neomycin resistance. In
cells, and the activity required ribosomal protein S6 kinase the control, the cells were transfected with pcDNA 3.1 plasmid. The
(S6K). These data suggest that macrophage PDGF may play an transfection was conducted with Lipofectamine 2000 in 4.5 ml of
important role in the stimulation of tube formation in the medium. The transfected cells were cultured in G418 (300 mg/ml)-
containing medium 36 h later for 24 –28 days. The stable positive cells
process of angiogenesis in adipose tissue. The study provides
were confirmed in Western blot with HA antibody and S6K rabbit
direct evidence for the role of the PI3K/mTOR/S6K signaling antibody. The positive cells were cultured in G418-free DMEM for
pathway in the PDGF-induced tube formation. two passages before experiment.
Cell viability (MTT) assay. SVEC4-10 (1 ⫻ 104 cells/well) were
EXPERIMENT PROCEDURES plated in a 96-well plate in DMEM supplemented with 10% FBS.
Then the cells were starved overnight in a serum-free medium and
Animals. Male C57BL/6J-Lepob, and C57BL/6J mice were pur-
treated with various reagents for 24 h. The MTT assay was conducted
chased from the Jackson Laboratory (Bar Harbor, ME) at 6 wk of age
and used in the study according to the animal protocol approved by by incubation of the cells in 20 ␮l of MTT solution (5 mg/ml in PBS)
the Institutional Animal Care and Use Committee at the Pennington for 3–5 h. Cell viability was determined by formazan (MTT metabolic
Biomedical Research Center, Louisiana State university (Baton product), which was quantified with optical density at 560 nm in 200
Rouge, LA). The mice were housed in regular cage with four mice per ␮l of DMSO and normalized over the subtract background at 670 nm.
cage with free access to water and standard chow unless noted Hypoxia treatment. Cells were treated with hypoxia (1% oxygen)
otherwise. The serum of the mice was collected from the tail vein. The in vitro, as described elsewhere (56). The control cells were main-
serum level of PDGF was quantified with a Mouse/Rat PDGF-BB tained in normoxic condition. After treatment, the cell culture super-
Immunoassay kit (MBB00; R&D Systems). natant was collected as conditioned medium and stored at ⫺80°C until
Cell lines and reagents. Murine endothelial cell line SVEC4-10 experiments (30).
(CRL-2181, ATCC), murine fibroblast cell line 3T3-L1 (CL-173, Quantitative RT-PCR. Total RNA was extracted from homoge-
ATCC), and the RAW 264.7 cell line (TIB-71) were maintained in nized fat pads or cells using Tri Reagent (T9424; Sigma, St. Louis,
DMEM with 10% fetal bovine serum in a CO2 (5%) incubator. MO). Quantitative real-time RT-PCR (qRT-PCR) was conducted
3T3-L1 preadipocytes were differentiated into adipocytes as described using the ABI 7900HT fast real-time PCR system (Applied Biosys-
elsewhere (15). Antibodies to phospho-p70 S6 kinase (Thr421/Ser424, tems, Foster City, CA). The following primers and probes were
9204) and phospho-Akt (Ser473, 9271) were obtained from Cell ordered from Applied Biosystems: PDGF (Mm00437304_m1), CD31
Signaling (Boston, MA). Antibodies to PDGF-BB (ab53716), phospho- (Mm01246167_m1), and F4/80 (Mm00802530_m1). The specific
glycogen synthase kinase (GSK)-3␤ (Ser9, ab30619), GSK-3␤ signal was normalized over 18S rRNA. A mean value of triplicates
(ab31366), S6K (ab9366), tubulin (ab7291), and actin (ab8227) were was used to express gene expression.
obtained from Abcam (Cambridge, MA). Antibodies to HA (sc-7392) Immunohistochemistry. The epididymal fat pads were isolated,
and Akt1 (sc-8312) were bought from Santa Cruz Biotechnoloty (Santa fixed in neutral buffered formalin, dehydrated, and embedded in
Cruz, CA). Rapamycin (A275-0001), LY-294002 (ST-420), and SP- paraffin. Thin tissue slides (5 ␮m) were deparaffinized, blocked, and
600125 (EI-305) were obtained from Biomol International (Plymouth incubated overnight at 4°C with a mouse anti-mouse CD31 antibody
Meeting, MA). Wortmannin (W1628) and SB-203580 (S8307) were (ab24590; Abcam, Cambridge, MA), which was followed by signal
obtained from Sigma-Aldrich (St. Louis, MO). These reagents were amplification using a VECTASTAIN Elite ABC Kit (PK-6102; Vec-
diluted and added to the cells at indicated working concentrations. tor Laboratories). The reaction was developed by addition of AEC
Deletion of macrophages in adipose tissue. Macrophages were chromogen substrate (AEC Staining Kit; Sigma-Aldrich). In fluores-
deleted in the adipose tissue by a single injection of clodronate cence analysis, the CD31 signal was determined with goat anti-mouse
AJP-Endocrinol Metab • VOL 295 • AUGUST 2008 • www.ajpendo.org
Downloaded from journals.physiology.org/journal/ajpendo at CAPES-USP (200.144.063.031) on August 23, 2021.
 REGULATION OF ADIPOSE TISSUE ANGIOGENESIS BY MACROPHAGES E315
IgG-FITC (sc-2010, Santa Cruz Biotechnology). Microphotographs PDGF expression in adipose tissue of ob/ob mice. To
were taken under a microscope (⫻20). understand the reduction in capillary density, we examined
Peritoneal macrophages. Primary peritoneal macrophages were PDGF activity in the adipose tissue of obese mice. As a
isolated from C57BL/6J mice, as described elsewhere (56). The
proangiogenic factor, PDGF is increased in human serum in
macrophages were transferred to 35-mm tissue culture dishes and
treated with hypoxia in serum-free medium 3 days later for collection response to inflammation and is involved in angiogenesis in
of supernatant. many tissues (1, 3). However, it is not clear whether blood
Statistical analysis. All of the experiments were conducted at least PDGF is elevated in an obese condition. To address this issue,
three times with consistent results. The gel or image from represen- the PDGF protein was compared in the serum of lean and ob/ob
tative experiment is presented. Values are means ⫾ SE of multiple mice. The PDGF protein was increased in obese mice, as
data points. Student’s t-test or one-way ANOVA was used in statis- indicated by the ELISA data (Fig. 2A). To determine the source
tical analysis of the data with a significance of P ⬍ 0.05. of PDGF protein, PDGF mRNA was determined in the adipose
tissue by qRT-PCR. The mRNA was increased in the ob/ob
RESULTS
mice as well (Fig. 2B). To determine the cell types for the
Decreased density of microvasculature in adipose tissue in source of increased PDGF, we examined PDGF mRNA in the
ob/ob mice. A decrease in capillary density has been proposed adipose tissue after deletion of macrophages in the obese mice.
as a mechanism of blood flow reduction in adipose tissue in The PDGF expression was reduced by macrophage deletion in
obesity. However, the reduction in capillary density is contro- a dose-dependent manner (Fig. 2C). These data suggest that
versial (7, 23). One possible reason for the discrepancy is the PDGF is elevated in the blood and that adipose tissue may
quantification method for capillary density, which was deter- contribute to the systemic increase in PDGF protein. Macro-
mined by immunostaining in the studies (7, 23). To resolve the phages are responsible for the PDGF increase in adipose tissue.
discrepancy, we compared capillary density in adipose tissues Expression of PDGF by macrophages and adipocytes. The
of lean mice and ob/ob mice via quantification of endothelial data above suggest that macrophages may be a major producer
cell marker CD31. The CD31 protein was examined by immu- of PDGF in the adipose tissue of obese mice. To test this
nostaining and Western blot. The CD31 mRNA was determined question, we compared macrophages and adipocytes for ex-
by qRT-PCR. In the immunohistostaining, the CD31 protein was pression of PDGF mRNA. The primary cells and cell lines
detected in the tissue with either colorimetric- or fluorescence- were used and their activities compared in the basal and
based staining. CD31, distributed in the extracellular matrix of hypoxia-treated conditions. In the primary cells, macrophages
adipocytes, was reduced in obese mice (Fig. 1, A and B). The expressed more PDGF than the adipocytes in the basal condi-
protein and mRNA of CD31 were both reduced in ob/ob mice tion (Fig. 3A). This pattern of expression was also observed in
(Fig. 1, C and D). These data suggest that neovascularization is the cell lines of macrophages (RAW cell) and adipocytes
reduced in white adipose tissue in obesity. (3T3-L1) (Fig. 3B). These data suggest that PDGF is expressed

Fig. 1. Vascular density decreased in adipose

tissue in ob/ob mice. A and B: sections of
epididymal fat pads were stained with CD31
antibody to reveal endothelial cell density.
Immunohistostaining was conducted for the
endothelial marker CD31. A: CD31 protein is
indicated in red from AEC staining. B: CD31
protein is indicated in green fluorescent. Blue
stands for nuclei stained by DAPI. Scale bar,
50 ␮m. C: CD31 protein was determined in
whole cell lysate of epididymal fat pads in
Western blot. Representative blot is shown
with an average signal strength in the bar
figure. D: CD31 mRNA was determined in
epididymal fat pads by qRT-PCR. In this
figure, each data point represents mean ⫾ SE
(n ⫽ 5). Experiments were repeated 3 times
with consistent results.

AJP-Endocrinol Metab • VOL 295 • AUGUST 2008 • www.ajpendo.org

Downloaded from journals.physiology.org/journal/ajpendo at CAPES-USP (200.144.063.031) on August 23, 2021.
E316 REGULATION OF ADIPOSE TISSUE ANGIOGENESIS BY MACROPHAGES

Fig. 2. Platelet-derived growth factor (PDGF) increased in adipose tissue in ob/ob mice. A: serum PDGF was determined by ELISA. B: PDGF mRNA was
determined in epididymal fat pads by qRT-PCR. C: PDGF mRNA in adipose tissue after macrophage deletion. Macrophages were reduced in adipose tissue by
single ip injection of clodronate liposome in ob/ob mice (6 wk old). PDGF was examined at day 4 after injection. F4/80 is a marker of macrophage. Each data
points represents mean ⫾ SE (n ⫽ 3).

in adipocytes, but the expression level is significantly lower remained to be tested. To test this, PDGF expression was
than that of macrophages. The impact of differentiation on determined in primary macrophages after exposure to ambient
PDGF expression was investigated in 3T3-L1 cells. After hypoxia. As expected, PDGF expression was increased by the
differentiation, PDGF expression was significantly reduced in hypoxia treatment (Fig. 3D). In the macrophage cell line
3T3-L1 cells (Fig. 3C). These data suggest that expression of (RAW), the same effect was observed for hypoxia (Fig. 3D).
PDGF is in the order of macrophage ⬎ preadipocytes ⬎ However, the response of RAW macrophages was stronger
mature adipocytes. than for the primary macrophages. These data further support
Induction of PDGF expression by hypoxia. Hypoxia in that macrophages are a the major producer of PDGF in the
adipose tissue may induce PDGF expression in obese mice, as adipose tissue of obese mice. Macrophage infiltration should
PDGF-B is a hypoxia response gene (16); this possibility contribute to the elevated PDGF expression in adipose tissue.

Fig. 3. PDGF expression in macrophages and adipocytes. A: PDGF mRNA in primary cells. The basal level of PDGF mRNA was determined in primary
macrophages and adipocytes of lean mice. B: basal level of PDGF mRNA was determined in macrophage cell lines (RAW cells) and 3T3-L1 adipocytes. C: basal
level of PDGF mRNA was determined in 3T3-L1 fibroblasts before and after differentiation into adipocytes. D: induction of PDGF mRNA expression by ambient
hypoxia in primary macrophages and RAW macrophages. E: expression of PDGF in adipose tissue of lean mice after macrophage deletion. A single injection
of clodronate liposome (150 mg/kg) was used to delete macrophages in lean C57BL/6J mice (8 wk old). Experiments were repeated 3 times with consistent
results. Each data point represents mean ⫾ SE (n ⫽ 3). *P ⬍ 0.05, treated vs. untreated in RAW cells; ⫹P ⬍ 0.05, treated vs. untreated in primary macrophages.

AJP-Endocrinol Metab • VOL 295 • AUGUST 2008 • www.ajpendo.org

Downloaded from journals.physiology.org/journal/ajpendo at CAPES-USP (200.144.063.031) on August 23, 2021.
 REGULATION OF ADIPOSE TISSUE ANGIOGENESIS BY MACROPHAGES E317
In lean mice, preadipocytes or stromal cells may be the primary PI3K/Akt pathway in pericytes that normally form part of the
source of PDGF, as deletion of macrophages did not lead to capillary wall. It remains to be tested whether the same path-
reduction in PDGF expression (Fig. 3E). way is involved in tube formation by endothelial cells. To
Function of PDGF in regulation of angiogenesis. Given the investigate the PI3K/Akt pathway in endothelial cells, we exam-
role of PDGF in angiogenesis, we proposed that macrophage ined PI3K inhibitors on tube formation. In the positive control,
PDGF may be involved in the stimulation of capillary growth insulin was used to induce tube formation (Fig. 5, A and B). In the
in adipose tissue under hypoxia, in which case the supernatant presence of PI3K inhibitor LY-294002 (100 ␮M) or wortmanin
of hypoxia-treated macrophages should be able to stimulate (1 ␮M), the insulin-induced tube formation was reduced sig-
angiogenesis. To test this possibility, the supernatant of RAW nificantly. Recombinant PDGF was used to stimulate tube
macrophages treated by hypoxia was examined in the tube formation in the mouse endothelial cells. As expected, the tube
formation assay. Such an assay was used to study PDGF formation was induced by the recombinant PDGF, and the
activity elsewhere (2, 12). The results suggest that the macro- effect was blocked by the PI3K inhibitors (Fig. 5, A and B). The data
phage supernatant induced more tube in the assay, and this suggest that PI3K activity is required for tube formation
activity increased in the supernatant of hypoxia-treated mac- induced by PDGF and insulin. The basal level of tube forma-
rophages (Fig. 4, A and B, control antibody). In the presence of tion was also reduced by the PI3K inhibitors (Fig. 5, A and B),
PDGF antibody, such an increase in tube formation was inhib- suggesting that PI3K activity is required for tube formation
ited (Fig. 4, A and B, PDGF antibody). induced by Matrigel, which contains proangiogenic factors.
PI3K/Akt pathway in PDGF-stimulated tube formation. The To verify activation of PI3K/Akt by PDGF, we examined
signaling pathway of PDGF receptor is involved in angiogen- the phosphorylation status of Akt and its downstream mole-
esis and is described in several pathological conditions (47). cules in a Western blot. Identical to those observed with insulin
The importance of the PI3K/Akt/mTOR pathway in PDGF- in the positive control, the activation markers of Akt, S6K, and
induced angiogenesis was demonstrated in vivo in rodent GSK3-␤ were all increased by PDGF (Fig. 5, C and D). These
models (13, 18, 25, 58). PDGF is known to activate the changes were consistently blocked by the PI3K inhibitor LY-
294002. We also tested the roles of MAPK kinases in PDGF-
induced tube formation with chemical inhibitors to ERK (PD-
098095 40 ␮M), JNK (SP-600125 50 ␮M), and p38 (SB-
203580 2 ␮M). Although these inhibitors could inhibit the
basal level of tube formation, they did not decrease the PDGF-
induced tube formation significantly (Supplemental Fig. S1,
online only). These data suggest that the PI3K/Akt pathway is
required for angiogenesis induced by PDGF.
mTOR in PDGF-induced tube formation. mTOR is a major
signaling molecule downstream of Akt. Inhibition of angiogen-
esis by rapamycin in mice suggests that mTOR may mediate
PI3K/Akt signaling in the stimulation of angiogenesis in mice
(18). In the current study, inhibition of mTOR led to a reduc-
tion in tube formation induced by PDGF as well as by insulin
(Fig. 6, A and B), suggesting that mTOR is required for the
PDGF signaling pathway. Activation of mTOR by PDGF and
its inhibition by rapamycin were examined by the phosphory-
lation status of S6K in Western blot. As was observed for
insulin, the phosphorylation was induced by PDGF and
blocked by rapamycin (Fig. 6, C and D). As inhibition of
protein synthesis by cycloheximide abolished the tube forma-
tion induced by PDGF or insulin (Fig. 6A), protein synthesis is
required for the tube formation. This observation suggests that
mTOR may act through S6K in the PDGF signaling pathway in
endothelial cells, as S6K controls gene translation.
S6K in PDGF-induced tube formation. S6K is one of the
major signaling molecules activated by mTOR and is involved
in regulation of protein synthesis. It remains to be tested
whether S6K mediates PDGF signal in endothelial cells for
tube formation (24, 38, 57). The S6K dominant-negative (S6K-
Fig. 4. Induction of tube formation by PDGF in macrophage supernatant. Tube
formation of endothelial cells on Matrigel was used to test proangiogenic
DN) mutant in stable transfection was generated to test S6K
activity of PDGF in supernatants of macrophages. Raw cells were treated with activity in PDGF-induced tube formation. The mutant S6K was
hypoxia for 4 h, and its supernatant was collected for induction of tube expressed in the SVEC4-10 cells, as indicated by detection of
formation in endothelial cells. PDGF antibody was used to determine activity HA-tag in Western blot (Fig. 7A). Inhibition of S6K function
of PDGF in the supernatant. A: representative images of tube formation of the
endothelial cells on Matrigel. B: quantification of tube formation. Experiments
was confirmed, with decreased phosphorylation of its substrate
were repeated 3 times with consistent results. Each data point represents S6 in the transfected cells (Fig. 7B). In response to PDGF, the
mean ⫾ SE (n ⫽ 3 experiments). stable cell line expressing S6K-DN exhibited much fewer tubes
AJP-Endocrinol Metab • VOL 295 • AUGUST 2008 • www.ajpendo.org
Downloaded from journals.physiology.org/journal/ajpendo at CAPES-USP (200.144.063.031) on August 23, 2021.
E318 REGULATION OF ADIPOSE TISSUE ANGIOGENESIS BY MACROPHAGES

Fig. 5. PI3K/Akt pathway in PDGF-induced tube formation. The role of PI3K in tube formation induced by PDGF was examined using PI3K inhibitors.
A: representative images of tube formation of the endothelial cells induced by PDGF. B: quantification of tube formation. C: representative Western blots of
phosphorylation of PI3K/Akt signaling molecules. D: quantification of Western blot signals. In this figure, all experiments were repeated 3 times with consistent
results. Each data point represents mean ⫾ SE (n ⫽ 3 experiments).

on the Matrigel (Fig. 7, C and D), suggesting that S6K is BMI ⬎25–30 kg/m2, a failure in angiogenic compensation may
required by PDGF in the signal transduction. occur and thus reduce the growth rate of fat tissues. A reduc-
Stimulation of tube formation by albumin. The data above tion in adipose tissue blood flow may be a result of such
suggest that the signal of tube formation is transduced through angiogenic failure (53). To explain adipose tissue hypoxia in
PI3K/Akt/mTOR/S6K under the PDGF receptor. It is not clear the obese condition (56), we proposed that angiogenic failure
whether activation of S6K in the absence of activation of may occur in adipose tissue of obese mice. In the current study,
upstream signaling molecules including PI3K and Akt is suf- this possibility is supported by the reduction of vascular den-
ficient to induce tube formation. To test this possibility, S6K sity in the adipose tissue. The CD31 data from immunohis-
was activated with albumin (Fig. 8A). The activation led to an tostaining, Western blot, and qRT-PCR consistently support
increase in tube formation, and the increase was blocked by the lack of endothelial cells in the adipose tissue of ob/ob mice
rapamycin (Fig. 8, B and C), which suppresses S6K by target- (Fig. 1). A lack of VEGF induction by hypoxia was observed
ing mTOR. The data suggest that activation of S6K by albumin in the adipose tissue in the same condition (56). The VEGF
is able to increase tube formation. Toxicity of rapamycin was non-response may contribute to the reduced endothelial cells,
examined in an MTT assay. Treatment of the cells with as VEGF is a primary growth factor for endothelia cells.
rapamycin for 24 h had no significant effect on cell viability in Macrophages may serve as a stimulator for angiogenesis in
presence or absence of PDGF (Fig. 8D). adipose tissue in obesity. Adipose tissue contains several types
of cells, such as preadipocytes (or stromal cells), adipocytes,
DISCUSSION
macrophages, and endothelial cells. PDGF is expressed in all
Our study suggests that vascular density is decreased in of these types of cells; however, the expression levels are
adipose tissue in ob/ob mice. Angiogenesis is required for the different. The current study suggests that preadipocytes ex-
development and growth of adipose tissue (7, 41). During press more PDGF than mature adipocytes. In obesity, preadi-
development of obesity, angiogenic activity will be increased pocyte number is reduced, as most of them are differentiated
to compensate for expansion of adipocyte tissue. This compen- into mature adipocytes. This change may lead to a decrease in
sation ensures quick growth in adipose tissue at the early stage local PDGF level, since mature adipocytes produce less PDGF
of weight gain. Once the weight gain reaches a level such as than the preadipocytes (Fig. 3C). To meet the demand for
AJP-Endocrinol Metab • VOL 295 • AUGUST 2008 • www.ajpendo.org
Downloaded from journals.physiology.org/journal/ajpendo at CAPES-USP (200.144.063.031) on August 23, 2021.
 REGULATION OF ADIPOSE TISSUE ANGIOGENESIS BY MACROPHAGES E319

Fig. 6. Mammalian target of rapamycin (mTOR) in PDGF-induced tube formation. The role of mTOR/S6K in PDGF-induced tube formation was tested with
mTOR inhibitor (rapamycin) and protein synthesis inhibitor cycloheximide (CHX). A: representative images of tube formation in the presence of rapamycin and
CHX. Insulin was used as a positive control in induction of mTOR activation. Rapamycin (200 nM) and CHX (20 ng/ml) were used to pretreat cells for 30 min,
followed by treatment of PDGF (50 ng/ml) or insulin (100 nM) for 10 h. B: quantification of tube formation. C: representative Western blots for mTOR/S6K
signaling pathway in endothelial cells. Phosphorylation of S6K was determined in SVEC4-10 cells after treatment with PDGF or insulin. D: quantification of
Western blot signals. In this figure, all experiments were repeated 3 times with consistent results. Each data point represents mean ⫾ SE (n ⫽ 3 experiments).

PDGF, macrophage infiltration into adipose tissue is increased or its receptor has been shown to promote GLUT4 transloca-
to compensate for the loss of preadipocytes for PDGF produc- tion in adipocytes (31, 36). PDGF has been reported to inhibit
tion (Fig. 2B). Our data support the idea that infiltration and adipocyte differentiation (19). In other organs, local PDGF has
activation of macrophages may contribute to the increased been suggested to play a role in pathogenesis of proliferative
PDGF expression in the adipose tissue of obese mice. This retinopathy and diabetic nephropathy (14, 26, 39). An increase
conclusion is supported by the observations that deletion of in PDGF has been found in the retinal membranes and vitreous
macrophages led to a significant reduction in PDGF expression fluid of patients with proliferative diabetic retinopathy (14, 39).
in adipose tissue of ob/ob mice. The reduction was observed in A higher PDGF level has also been reported in individuals with
obese mice but not in lean mice, which have abundant preadi- additional rubeosis iridis or ischemic nondiabetic retinopathy
pocytes (Figs. 2C and 3E). Macrophages are second next to (14). In the renal biopsy of patients with diabetic nephropathy,
platelets in expression of PDGF (32, 40, 43). We believe that, expression of PDGF-A and PDGF-B has been reported to be
in addition to its roles in chronic inflammation (52, 55), increased in mRNA and protein (26). In mice with type 1
macrophages may serve as an angiogenic stimulator in adipose diabetes, the PDGF elevation has been associated with macro-
tissue by expression of PDGF. To our knowledge, this may be phage infiltration into kidney with diabetic nephropathy (9).
the first study to support elevation and function of PDGF in These observations suggest a role of PDGF in the pathogenesis
adipose tissue of obesity. Hypoxia in adipose tissue is likely to of diabetic complications.
induce PDGF expression in macrophages (Fig. 3D). In the The current study suggests that the activity of PDGF in
current study, PDGF expression was determined mainly at the angiogenesis is dependent on VEGF activity. Angiogenesis
mRNA level. It remains to be tested whether PDGF protein is requires endothelial cell proliferation and tube formation. The
induced by hypoxia in macrophages. two events are equally important in the process of formation of
We did not find literature about PDGF expression in adipose capillary. Tube formation requires the presence of endothelial
tissue in obesity. Closely related literature includes PDGF cells; in the absence of endothelial cells, tube formation will
regulation of GLUT4 translocation, proliferation, or differen- not occur. Endothelial cell proliferation is dependent on proan-
tiation of adipocytes in cell culture (19, 31, 36). In vitro, PDGF giogenic factor VEGF, which stimulates cell proliferation
AJP-Endocrinol Metab • VOL 295 • AUGUST 2008 • www.ajpendo.org
Downloaded from journals.physiology.org/journal/ajpendo at CAPES-USP (200.144.063.031) on August 23, 2021.
E320 REGULATION OF ADIPOSE TISSUE ANGIOGENESIS BY MACROPHAGES

previous observation that VEGF expression was not increased

in adipose tissue by obesity in ob/ob mice (56). Lack of leptin
in the ob/ob mice may be related to the unresponsiveness of
VEGF to obesity or adipose tissue hypoxia. In the wild-type
mice, VEGF was increased by diet-induced obesity in the
adipose tissue (56). Without sufficient endothelial cells, PDGF
will not be able to increase angiogenesis in the adipose tissue
by itself.
In addition to stimulation of tube formation, PDGF was
reported to play a role in the recruitment of pericytes or the
proliferation of vessel smooth muscle cells in angiogenesis.
Recruitment of pericytes is required for maintenance of micro-
vascular stability and function (21, 22, 29). This function of
PDGF is established in a PDGF knockout study (29). Pericytes
are able to transdifferentiate into fibroblasts (stromal cells/
preadipocytes) and vessel smooth muscle cells (37). With these
activities, it is possible that PDGF is involved in recruitment of
preadipocytes into adipose tissue in obesity.
In the current study, we observed that insulin stimulated tube
formation, suggesting a role for insulin in the stimulation of
angiogenesis. This activity indicates that hyperinsulinemia
may serve to stimulate angiogenesis in adipose tissue in the
Fig. 7. S6K in PDGF-induced tube formation. A: expression of dominant- obese condition. Hyperinsulinemia is a prognosis factor for the
negative S6K (S6K-DN) in stable cells. B: inhibition of S6K activity by incidence of microvascular abnormalities in diabetes patients
S6K-DN mutant. C: tube formation in stable cells in response to PDGF. for retinopathy and nephropathy (4). In the present study,
D: quantification of tube formation. In this figure, all experiments were insulin exhibited a similar activity to PDGF in the induction of
repeated 3 times with consistent results. Each data point represents mean ⫾ SE
(n ⫽ 3 experiments). tube formation (Figs. 5 and 6). This observation suggests that
hyperinsulinemia may be involved in neovascularation in adi-
pose tissue during the development of obesity.
through VEGF receptor 2 in the endothelial cells. The lack of S6K is required for tube formation induced by PDGF and
endothelia cells was observed in adipose tissue of ob/ob mice insulin. S6K is activated by an array of mitogenic or nutrient
in this study (Fig. 1). The reduction in endothelial cells is stimuli, such as insulin, serum, phorbol ester, and PDGF (38,
supported by quantification of CD31 (endothelial cell marker) 48, 54). Knockout studies suggest that S6K is an important
in protein and mRNA. This reduction is consistent with our regulator of cell size and body growth in mice and Drosophila

Fig. 8. S6K in tube formation induced by al-

bumin. A: Induction of phosphorylation of S6K
by albumin. Phosphorylation of S6K was ex-
amined in whole cell lysate in Western blot.
B: BSA-stimulated tube formation and inhibi-
tion by rapamycin (200 nM). C: quantification
of tube formation. D: cell viability after rapa-
mycin treatment. Cell viability was determined
with MTT assay at 24 h after cell exposure to
rapamycin. In this figure, all experiments were
repeated 3 times with consistent results. Each
data points represents mean ⫾ SE (n ⫽ 3
experiments).

AJP-Endocrinol Metab • VOL 295 • AUGUST 2008 • www.ajpendo.org

Downloaded from journals.physiology.org/journal/ajpendo at CAPES-USP (200.144.063.031) on August 23, 2021.
 REGULATION OF ADIPOSE TISSUE ANGIOGENESIS BY MACROPHAGES E321
(35, 42). S6K1⫺/⫺ mice are protected against age- and high-fat 9. Chow FY, Nikolic-Paterson DJ, Atkins RC, Tesch GH. Macrophages in
diet-induced obesity and insulin resistance due to increased streptozotocin-induced diabetic nephropathy: potential role in renal fibro-
sis. Nephrol Dial Transplant 19: 2987–2996, 2004.
energy expenditure (48). In tube formation, S6K may act 10. Cinti S, Mitchell G, Barbatelli G, Murano I, Ceresi E, Faloia E, Wang
through induction of FGF-2 expression in endothelial cells S, Fortier M, Greenberg AS, Obin MS. Adipocyte death defines mac-
(57). In the present study, our data suggest that S6K mediates rophage localization and function in adipose tissue of obese mice and
signals of PDGF and insulin in the induction of tube formation. humans. J Lipid Res 46: 2347–2355, 2005.
This is supported by data that inhibition of S6K by dominant- 11. Crandall DL, Hausman GJ, Kral JG. A review of the microcirculation
of adipose tissue: anatomic, metabolic, and angiogenic perspectives.
negative mutant or chemical inhibitors (LY, wortmanin, and Microcirculation 4: 211–232, 1997.
rapamycin) lead to inhibition of PDGF and insulin activity 12. De Marchis F, Ribatti D, Giampietri C, Lentini A, Faraone D,
(Figs. 5–7). These data consistently support the notion that Scoccianti M, Capogrossi MC, Facchiano A. Platelet-derived growth
S6K is a critical kinase for endothelial differentiation and tube factor inhibits basic fibroblast growth factor angiogenic properties in vitro
formation. and in vivo through its alpha receptor. Blood 99: 2045–2053, 2002.
13. Dell S, Peters S, Muther P, Kociok N, Joussen AM. The role of PDGF
In summary, reduction in vascular density is likely a result receptor inhibitors and PI3-kinase signaling in the pathogenesis of corneal
of a lack of an endothelium growth factor (such as VEGF) in neovascularization. Invest Ophthalmol Vis Sci 47: 1928 –1937, 2006.
adipose tissue in ob/ob mice. This may contribute to the 14. Freyberger H, Brocker M, Yakut H, Hammer J, Effert R, Schiffer-
development of adipose tissue hypoxia and macrophage infil- decker E, Schatz H, Derwahl M. Increased levels of platelet-derived
tration. PDGF expression is increased in macrophages in re- growth factor in vitreous fluid of patients with proliferative diabetic
retinopathy. Exp Clin Endocrinol Diabetes 108: 106 –109, 2000.
sponse to the hypoxia. Although PDGF is able to stimulate 15. Gao Z, He Q, Peng B, Chiao PJ, Ye J. Regulation of nuclear translo-
tube formation, it may not stimulate angiogenesis in the ab- cation of hdac3 by I(kappa)B(alpha) is required for tumor necrosis factor
sence of sufficient endothelial cells or VEGF activity. The inhibition of peroxisome proliferator-activated receptor (gamma) function.
signaling pathway PI3K/Akt/mTOR/S6K is used by PDGF in J Biol Chem 281: 4540 – 4547, 2006.
the induction of tube formation. With insufficient endothelial 16. Gleadle JM, Ebert BL, Firth JD, Ratcliffe PJ. Regulation of angiogenic
growth factor expression by hypoxia, transition metals, and chelating
cells in the adipose tissue of ob/ob mice, macrophages may fail agents. Am J Physiol Cell Physiol 268: C1362–C1368, 1995.
to stimulate angiogenesis. These possibilities may support a 17. Goldsmith HS, Griffith AL, Kupferman A, Catsimpoolas N. Lipid
new function for macrophages in adipose tissue. Hyperinsulin- angiogenic factor from omentum. JAMA 252: 2034 –2036, 1984.
emia may also be involved in angiogenesis in adipose tissue in 18. Guba M, von Breitenbuch P, Steinbauer M, Koehl G, Flegel S,
obesity by stimulation of tube formation in endothelial cells. Hornung M, Bruns CJ, Zuelke C, Farkas S, Anthuber M, Jauch KW,
Geissler EK. Rapamycin inhibits primary and metastatic tumor growth by
ACKNOWLEDGMENTS antiangiogenesis: involvement of vascular endothelial growth factor. Nat
Med 8: 128 –135, 2002.
We thank Dr. David Burk in the Imaging Core Facility and Dr. Viktor Drel 19. Hauner H, Rohrig K, Petruschke T. Effects of epidermal growth factor
for their excellent technical support in the capture of images for tube formation (EGF), platelet-derived growth factor (PDGF) and fibroblast growth factor
and processing of tissue slides for immunohistostaining. We thank Hanjie (FGF) on human adipocyte development and function. Eur J Clin Invest
Zhang for technical support in the preparation of clodronate liposome. 25: 90 –96, 1995.
20. Heldin CH, Westermark B. Mechanism of action and in vivo role of
GRANTS
platelet-derived growth factor. Physiol Rev 79: 1283–1316, 1999.
This study is supported by National Institute of Diabetes and Digestive and 21. Hellstrom M, Gerhardt H, Kalen M, Li X, Eriksson U, Wolburg H,
Kidney Diseases (NIDDK) funds (DK-68036) and ADA research award Betsholtz C. Lack of pericytes leads to endothelial hyperplasia and
(7-07-RA-189) to J. Ye and the Key Project from the Science and Technology abnormal vascular morphogenesis. J Cell Biol 153: 543–553, 2001.
Commission of Shanghai municipality (04DZ19501, 01ZD002, and National 22. Hellstrom M, Kalen M, Lindahl P, Abramsson A, Betsholtz C. Role of
973 Program) to W. Jia. The qRT-PCR study was conducted in the genomic PDGF-B and PDGFR-beta in recruitment of vascular smooth muscle cells
core that is supported by the CNRU Grant (1P30 DK-072476) sponsored by and pericytes during embryonic blood vessel formation in the mouse.
NIDDK. Development 126: 3047–3055, 1999.
23. Kampf C, Bodin B, Kallskog O, Carlsson C, Jansson L. Marked
REFERENCES increase in white adipose tissue blood perfusion in the type 2 diabetic GK
rat. Diabetes 54: 2620 –2627, 2005.
1. Au PY, Martin N, Chau H, Moemeni B, Chia M, Liu FF, Minden M,
Yeh WC. The oncogene PDGF-B provides a key switch from cell death 24. Kanda S, Hodgkin MN, Woodfield RJ, Wakelam MJ, Thomas G,
to survival induced by TNF. Oncogene 24: 3196 –3205, 2005. Claesson-Welsh L. Phosphatidylinositol 3⬘-kinase-independent p70 S6
2. Battegay EJ, Rupp J, Iruela-Arispe L, Sage EH, Pech M. PDGF-BB kinase activation by fibroblast growth factor receptor-1 is important for
modulates endothelial proliferation and angiogenesis in vitro via PDGF proliferation but not differentiation of endothelial cells. J Biol Chem 272:
beta-receptors. J Cell Biol 125: 917–928, 1994. 23347–23353, 1997.
3. Betsholtz C. Role of platelet-derived growth factors in mouse develop- 25. Kratchmarova I, Blagoev B, Haack-Sorensen M, Kassem M, Mann M.
ment. Int J Dev Biol 39: 817– 825, 1995. Mechanism of divergent growth factor effects in mesenchymal stem cell
4. Bonadonna RC, Cucinotta D, Fedele D, Riccardi G, Tiengo A. The differentiation. Science 308: 1472–1477, 2005.
metabolic syndrome is a risk indicator of microvascular and macrovascu- 26. Langham RG, Kelly DJ, Maguire J, Dowling JP, Gilbert RE, Thom-
lar complications in diabetes: results from Metascreen, a multicenter son NM. Over-expression of platelet-derived growth factor in human
diabetes clinic-based survey. Diabetes Care 29: 2701–2707, 2006. diabetic nephropathy. Nephrol Dial Transplant 18: 1392–1396, 2003.
5. Bouloumie A, Lolmede K, Sengenes C, Galitzky J, Lafontan M. 27. Lau DC, Schillabeer G, Li ZH, Wong KL, Varzaneh FE, Tough SC.
Angiogenesis in adipose tissue. Ann Endocrinol (Paris) 63: 91–95, 2002. Paracrine interactions in adipose tissue development and growth. Int J
6. Brakenhielm E, Cao R, Gao B, Angelin B, Cannon B, Parini P, Cao Y. Obes Relat Metab Disord 20, Suppl 3: S16 –S25, 1996.
Angiogenesis Inhibitor, TNP-470, Prevents Diet-Induced and Genetic 28. Lijnen HR. Angiogenesis and obesity. Cardiovasc Res 2007.
Obesity in Mice. Circ Res 94: 1579 –1588, 2004. 29. Lindahl P, Johansson BR, Leveen P, Betsholtz C. Pericyte loss and
7. Cao R, Brakenhielm E, Wahlestedt C, Thyberg J, Cao Y. Leptin microaneurysm formation in PDGF-B-deficient mice. Science 277: 242–
induces vascular permeability and synergistically stimulates angiogenesis 245, 1997.
with FGF-2 and VEGF. Proc Natl Acad Sci USA 98: 6390 – 6395, 2001. 30. Lolmede K, Durand de Saint Front V, Galitzky J, Lafontan M,
8. Castellot JJ Jr, Karnovsky MJ, Spiegelman BM. Differentiation-de- Bouloumie A. Effects of hypoxia on the expression of proangiogenic
pendent stimulation of neovascularization and endothelial cell chemotaxis factors in differentiated 3T3-F442A adipocytes. Int J Obes Relat Metab
by 3T3 adipocytes. Proc Natl Acad Sci USA 79: 5597–5601, 1982. Disord 27: 1187–1195, 2003.

AJP-Endocrinol Metab • VOL 295 • AUGUST 2008 • www.ajpendo.org

Downloaded from journals.physiology.org/journal/ajpendo at CAPES-USP (200.144.063.031) on August 23, 2021.
E322 REGULATION OF ADIPOSE TISSUE ANGIOGENESIS BY MACROPHAGES

31. Marx M, Perlmutter RA, Madri JA. Modulation of platelet-derived 44. Silverman KJ, Lund DP, Zetter BR, Lainey LL, Shahood JA, Freiman
growth factor receptor expression in microvascular endothelial cells dur- DG, Folkman J, Barger AC. Angiogenic activity of adipose tissue.
ing in vitro angiogenesis. J Clin Invest 93: 131–139, 1994. Biochem Biophys Res Commun 153: 347–352, 1988.
32. Mornex JF, Martinet Y, Yamauchi K, Bitterman PB, Grotendorst 45. Strissel KJ, Stancheva Z, Miyoshi H, Perfield JW, 2nd DeFuria J, Jick
GR, Chytil-Weir A, Martin GR, Crystal RG. Spontaneous expression Z, Greenberg AS, Obin MS. Adipocyte death, adipose tissue remodeling,
of the c-sis gene and release of a platelet-derived growth factorlike and obesity complications. Diabetes 56: 2910 –2918, 2007.
molecule by human alveolar macrophages. J Clin Invest 78: 61– 66, 1986. 46. Sunderkotter C, Goebeler M, Schulze-Osthoff K, Bhardwaj R, Sorg
33. Neels JG, Thinnes T, Loskutoff DJ. Angiogenesis in an in vivo model of C. Macrophage-derived angiogenesis factors. Pharmacol Ther 51: 195–
adipose tissue development. FASEB J 18: 983–985, 2004. 216, 1991.
34. Nishimura S, Manabe I, Nagasaki M, Hosoya Y, Yamashita H, Fujita 47. Tallquist M, Kazlauskas A. PDGF signaling in cells and mice. Cytokine
H, Ohsugi M, Tobe K, Kadowaki T, Nagai R, Sugiura S. Adipogenesis Growth Factor Rev 15: 205–213, 2004.
in obesity requires close interplay between differentiating adipocytes, 48. Um SH, D’Alessio D, Thomas G. Nutrient overload, insulin resistance,
stromal cells, and blood vessels. Diabetes 56: 1517–1526, 2007. and ribosomal protein S6 kinase 1, S6K1. Cell Metab 3: 393– 402, 2006.
35. Pende M, Um SH, Mieulet V, Sticker M, Goss VL, Mestan J, Mueller 49. Van Rooijen N, Sanders A. Liposome mediated depletion of macro-
M, Fumagalli S, Kozma SC, Thomas G. S6K1(⫺/⫺)/S6K2(⫺/⫺) mice phages: mechanism of action, preparation of liposomes and applications.
exhibit perinatal lethality and rapamycin-sensitive 5⬘-terminal oligopyri- J Immunol Methods 174: 83–93, 1994.
midine mRNA translation and reveal a mitogen-activated protein kinase- 50. Varzaneh FE, Shillabeer G, Wong KL, Lau DC. Extracellular matrix
dependent S6 kinase pathway. Mol Cell Biol 24: 3112–3124, 2004. components secreted by microvascular endothelial cells stimulate preadi-
36. Quon MJ, Chen H, Lin CH, Zhou L, Ing BL, Zarnowski MJ, pocyte differentiation in vitro. Metabolism 43: 906 –912, 1994.
Klinghoffer R, Kazlauskas A, Cushman SW, Taylor SI. Effects of 51. Vinals F, Chambard JC, Pouyssegur J. p70 S6 kinase-mediated protein
synthesis is a critical step for vascular endothelial cell proliferation. J Biol
overexpressing wild-type and mutant PDGF receptors on translocation of
Chem 274: 26776 –26782, 1999.
GLUT4 in transfected rat adipose cells. Biochem Biophys Res Commun
52. Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, Fer-
226: 587–594, 1996.
rante AW Jr. Obesity is associated with macrophage accumulation in
37. Rajkumar VS, Howell K, Csiszar K, Denton CP, Black CM, Abraham
adipose tissue. J Clin Invest 112: 1796 –1808, 2003.
DJ. Shared expression of phenotypic markers in systemic sclerosis indi-
53. West DB, Prinz WA, Francendese AA, Greenwood MR. Adipocyte
cates a convergence of pericytes and fibroblasts to a myofibroblast lineage blood flow is decreased in obese Zucker rats. Am J Physiol Regul Integr
in fibrosis. Arthritis Res Ther 7: R1113–R1123, 2005. Comp Physiol 253: R228 –R233, 1987.
38. Rebholz H, Panasyuk G, Fenton T, Nemazanyy I, Valovka T, Flajolet 54. Woo SY, Kim DH, Jun CB, Kim YM, Vander Haar E, Lee SI, Hegg
M, Ronnstrand L, Stephens L, West A, Gout IT. Receptor association JW, Bandhakavi S, Griffin TJ, Kim DH. PRR5, a novel component of
and tyrosine phosphorylation of S6 kinases. FEBS J 273: 2023–2036, mTOR complex 2, regulates PDGFRbeta expression and signaling. J Biol
2006. Chem 282: 25604 –25612, 2007.
39. Robbins SG, Mixon RN, Wilson DJ, Hart CE, Robertson JE, Westra 55. Xu H, Barnes GT, Yang Q, Tan G, Yang D, Chou CJ, Sole J, Nichols
I, Planck SR, Rosenbaum JT. Platelet-derived growth factor ligands and A, Ross JS, Tartaglia LA, Chen H. Chronic inflammation in fat plays a
receptors immunolocalized in proliferative retinal diseases. Invest Oph- crucial role in the development of obesity-related insulin resistance. J Clin
thalmol Vis Sci 35: 3649 –3663, 1994. Invest 112: 1821–1830, 2003.
40. Ross R, Raines EW, Bowen-Pope DF. The biology of platelet-derived 56. Ye J, Gao Z, Yin J, He H. Hypoxia is a potential risk factor for chronic
growth factor. Cell 46: 155–169, 1986. inflammation and adiponectin reduction in adipose tissue of ob/ob and
41. Rupnick MA, Panigrahy D, Zhang CY, Dallabrida SM, Lowell BB, dietary obese mice. Am J Physiol Endocrinol Metab 293: E1118 –E1128,
Langer R, Folkman MJ. Adipose tissue mass can be regulated through 2007.
the vasculature. Proc Natl Acad Sci USA 99: 10730 –10735, 2002. 57. Zhang B, Cao H, Rao GN. 15(S)-hydroxyeicosatetraenoic acid induces
42. Shima H, Pende M, Chen Y, Fumagalli S, Thomas G, Kozma SC. angiogenesis via activation of PI3K-Akt-mTOR-S6K1 signaling. Cancer
Disruption of the p70(s6k)/p85(s6k) gene reveals a small mouse pheno- Res 65: 7283–7291, 2005.
type and a new functional S6 kinase. EMBO J 17: 6649 – 6659, 1998. 58. Zhang H, Bajraszewski N, Wu E, Wang H, Moseman AP, Dabora SL,
43. Shimokado K, Raines EW, Madtes DK, Barrett TB, Benditt EP, Ross Griffin JD, Kwiatkowski DJ. PDGFRs are critical for PI3K/Akt activa-
R. A significant part of macrophage-derived growth factor consists of at tion and negatively regulated by mTOR. J Clin Invest 117: 730 –738,
least two forms of PDGF. Cell 43: 277–286, 1985. 2007.

AJP-Endocrinol Metab • VOL 295 • AUGUST 2008 • www.ajpendo.org

Downloaded from journals.physiology.org/journal/ajpendo at CAPES-USP (200.144.063.031) on August 23, 2021.