होम Connective Tissue Research Freezing Adipose Tissue Grafts May Damage Their Ability to Integrate into the Host

Freezing Adipose Tissue Grafts May Damage Their Ability to Integrate into the Host

, , , , ,
यह पुस्तक आपको कितनी अच्छी लगी?
फ़ाइल की गुणवत्ता क्या है?
पुस्तक की गुणवत्ता का मूल्यांकन करने के लिए यह पुस्तक डाउनलोड करें
डाउनलोड की गई फ़ाइलों की गुणवत्ता क्या है?
खंड:
50
भाषा:
english
पत्रिका:
Connective Tissue Research
DOI:
10.1080/03008200802385981
Date:
January, 2009
फ़ाइल:
PDF, 8.58 MB
Conversion to is in progress
Conversion to is failed
0 comments
 

अपनी समीक्षा पोस्ट करने के लिए साइन इन करें या साइन अप करें
आप पुस्तक समीक्षा लिख सकते हैं और अपना अनुभव साझा कर सकते हैं. पढ़ूी हुई पुस्तकों के बारे में आपकी राय जानने में अन्य पाठकों को दिलचस्पी होगी. भले ही आपको किताब पसंद हो या न हो, अगर आप इसके बारे में ईमानदारी से और विस्तार से बताएँगे, तो लोग अपने लिए नई रुचिकर पुस्तकें खोज पाएँगे.
Connective Tissue Research, 50:14–28, 2009
c Informa Healthcare USA, Inc.
Copyright 
ISSN: 0300-8207 print / 1521-0456 online
DOI: 10.1080/03008200802385981

Freezing Adipose Tissue Grafts May Damage Their Ability
to Integrate into the Host
Navanjun Grewal, Luciana Yacomotti, and Vahe Melkonyan
Regenerative Bioengineering and Repair Laboratory, Division of Plastic Surgery,
Department of Surgery, David Geffen School of Medicine at UCLA, Los Angeles, California, USA
Connect Tissue Res Downloaded from informahealthcare.com by Universitat Autonoma Barcelona on 11/04/14
For personal use only.

Marga Massey
Division of Plastic Surgery, University of Utah, Salt Lake City, Utah, USA

James P. Bradley and Patricia A. Zuk
Regenerative Bioengineering and Repair Laboratory, Division of Plastic Surgery,
Department of Surgery, David Geffen School of Medicine at UCLA, Los Angeles, California, USA

Keywords
In this study, the effect of freezing on the morphology, viability,
and VEGF synthesis of human adipose tissue grafts is examined.
Currently, storage of adipose grafts involves freezing in simple
saline solutions. However, the effect of freezing on the morphology
and function of adipose tissue remains unclear. As a result, this
study attempts to determine whether freezing adipose grafts should
be considered prior to soft-tissue augmentation. In this study,
the freezing of adipose grafts in saline for only 24 hr resulted
in morphological changes in vivo and affected their ability to
synthesize VEGF. The use of a simple cryopreservation medium
containing sucrose appeared to maintain VEGF synthetic levels
by the grafts and improved both their morphology and retention
in vivo. However, the benefits of this cryopreservation medium
were directly linked to storage time as long-term storage did not
result in any noticeable benefit to graft retention. Finally, as an
alternative to freezing, adipose grafts were combined with human
adipose-derived stem cells (ASCs) to determine if their presence
could enhance in vivo graft struct; ure. The presence of ASCs did
appear to improve graft structure in vivo over the short term and
was also capable of improving tissue morphology when combined
with grafts frozen in PBS. In conclusion, the successful use of
adipose grafts may require a closer examination of the graft’s
storage conditions and time. Specifically, it now appears that the
practice of freezing in saline may not be advisable if graft viability,
activity, and structure are to be maintained in vivo.

Adipose Tissue Grafts, Adipose-Derived Stem Cells
(ASCs), Fat Storage, Cryostorage

INTRODUCTION
Autologous fat grafts are used extensively by cosmetic and
reconstructive surgeons to correct a variety of soft tissue defects
such wrinkles, scars and congenital soft tissue defects [1]. The
use of fat grafting as a method for soft-tissue augmentation
is straightforward as adipose tissue can be easily obtained
through suction-assisted lipoplasty (i.e., liposuction). However,
following injection, these grafts are easily resorbed over time
resulting in a reduction of volume anywhere from 20% to 70%
[2, 3]. Hence multiple injections are often needed to achieve the
desired clinical result. As a result, clinicians have pointed out
the need for the short-term storage of liposuctioned fat grafts
under conditions that would prevent damage to the graft. Such
storage would allow for multiple grafting sessions using only a
single harvest and would reduce patient discomfort, procedure
cost, and time spent by both physician and patient.
Current techniques for fat storage are primitive. Adipose
material, removed via simple syringe suction, is frequently
frozen within the syringe without the addition of any additional
reagents to ensure its integrity or is frozen in saline solutions.
Recent studies have suggested that the freezing of adipose tissue
can damage the tissue structurally, decreasing the viability of
resident adipocytes [4]. As such, freezing adipose grafts may
reduce the efficiency of the graft’s “take” to the host site.
Currently, there is little to no information available to plastic
surgeons regarding how storage of adipose tissue affects its
retention and integration in the host. Therefore, it is imperative
that an initial analysis of fat storage techniques is performed to

Received 18 April 2008; Accepted 31 July 2008; Revised 14 July
2008.
Address correspondence to Patricia Zuk, PhD, 72-131 Center for
Health Sciences, Department of Surgery, David Geffen School of
Medicine at UCLA, 10833 Leconte Ave., Los Angeles, CA 90095,
USA. E-mail: zukpat@yahoo.com

14

15

Connect Tissue Res Downloaded from informahealthcare.com by Universitat Autonoma Barcelona on 11/04/14
For personal use only.

FREEZING ADIPOSE TISSUE GRAFTS

assess the effect of freezing on the overall structure and function
of mature fat as a soft tissue filler. As such, we have performed
an initial analysis of the effects of freezing adipose tissue with
and without cryopreservation on the activity and structure of
adipose tissue.
Specifically, we have investigated the effect of freezing
adipose tissue on the overall morphology of the tissue and
ultrastructure of the adipocyte. Since retention of adipose grafts
is likely to be affected by its degree of vascularization, we also
investigated the effect of freezing on the expression of vascular
endothelial growth factor (VEGF). Finally, we have investigated
the expression of VEGF by a population of stem cells known
to reside in the adipose compartment (i.e., adipose-derived stem
cells or ASCs). From this data, we conclude that the long-term
storage of adipose tissue by freezing has a dramatic effect
on its function (i.e., VEGF synthesis) and morphology. As a
result, the storage condition must be taken into account when
considering fat grafting as a means of repairing soft tissue
defects.

MATERIALS AND METHODS
Adipose Tissue Harvest and ASC Isolation
Adipose tissue grafts for analysis and injection were obtained
from conventional suction-assisted lipectomy (SAL) procedures
performed under Institutional Review Board guidelines. Raw
human lipoaspirates were washed extensively in sterile 1× PBS
(phosphate-buffered saline) containing 50 µg/ml gentamicin
(Invitrogen, Carlsbad, CA, USA) and 25 ng/ml amphotericin
B and either processed to isolate the ASCs population or
prepared for implantation into athymic rats. ASCs were obtained
from raw lipoaspirates as previously described [5, 6]. Briefly,
raw lipoaspirates were washed extensively with sterile 1×
PBS and treated with 0.075% collagenase (type I bovine;
Sigma-Aldrich, St. Louis, MO, USA) for 30 min at 37◦ C with
gentle agitation. The collagenase was inactivated with an equal

volume of DMEM/10% FBS and the infranatant containing
the ASC population collected by low-speed centrifugation. The
pellet was resuspended in control medium (CM—DMEM—
4.5 g/l, with L-glutamine and sodium pyruvate [Mediatech
Cellgro, Herndon, VA, USA], 10% fetal bovine serum [Gemini
Biosciences, Woodland, CA, USA], 10000 U/ml penicillin,
streptomycin [Mediatech]) [5] and filtered through a 100-µm
mesh filter to remove debris. The filtrate was centrifuged
and plated onto conventional tissue culture plates in CM
and maintained under standard tissue culture conditions (5%
CO2 , 37◦ C). Fresh ASC cultures (passage 0) were maintained
in CM until 80% confluent and then passaged using 0.25%
trypsin/2.23 mM EDTA (Mediatech) to produce passage 1
cultures. First-passage cultures were used for all experiments.
Media changes were made based on the ELISA assays described
below.
Cryostorage of Adipose Tissue and ASC Cultures
To cryopreserve the harvested adipose tissue grafts for
implantation, 1 ml aliquots of washed adipose tissue were frozen
at −20◦ C for 24 hr, 7 days, or 8 weeks in either 10 ml sterile 1×
PBS or in 6 defined cryostorage media compositions (Table 1).
Following freezing, the adipose tissue grafts were washed once
in sterile 1× PBS and either injected into rats subcutaneously
or introduced into culture for analysis. For the cryostorage
of ASCs, cells were harvested in trypsin/EDTA and aliquots
of 250,000 cells were centrifuged at low speed (500–1000
rpm) to produce a pellet. The ASC pellets were resuspended
in a conventional cellular cryostorage medium (Opti-Freeze,
Fisher Scientific, Pittsburgh, PA, USA) and frozen overnight
in a controlled-rate freezing vessel at –80◦ C. ASC cultures
were maintained at –80◦ C for an additional 72 hr, thawed in
a 37◦ C water bath, and placed back into culture in CM for either
24 hr or 7 days without a media change. Media from these ASC
cultures were collected after this 24 hr or 7 day culture period
for ELISA analysis.

TABLE 1
Metabolic activity of frozen adipose tissue grafts—VEGF synthesis (ELISAs)
VEGF secretion—pg/ml (freezing time/culture time)
Cryostorage condition

1 week/day 1

1 week/day 7

2 weeks/day 1

2 weeks/day 7

Control—sterile PBS
Cryo#1—50% DMEM, 40% FBS,
10% DMSO
Cryo#2—10% DMEM, 80% FBS,
10% DMSO
Cryo#3—80% FBS, 20% DMSO
Cryo#4—30% sucrose in PBS
Cryo#5—80% glycerol
Cryo#6—100% DMSO

10.41 ± 5.80
8.52 ± 6.02

−4.51 ± 8.59
−6.04 ± 11.67

−14.91 ± 3.21
0.53 ± 13.81

−8.75 ± 15.04
−4.97 ± 7.23

−1.42 ± 4.18

−6.15 ± 9.32

−14.20 ± 6.35

−8.99 ± 8.60

−3.91 ± 9.66
18.82 ± 2.93
−7.81 ± 8.07
−3.20 ± 8.39

−9.47 ± 8.88
17.34 ± 3.96
−11.36 ± 4.49
−4.02 ± 10.26

−13.49 ± 4.10
−11.60 ± 4.53
−15.98 ± 15.05
−5.21 ± 6.46

−11.99 ± 11.12
15.86 ± 9.64
−6.39 ± 3.89
−6.53 ± 3.70

Numbers in bold indicate statistical significance (p < 0.05).

16

N. GREWAL ET AL.

Connect Tissue Res Downloaded from informahealthcare.com by Universitat Autonoma Barcelona on 11/04/14
For personal use only.

Autologous Fat Graft Implantation
Young 12-week old adult athymic rats (250–300 g) (NIH
Crl:rnu, Charles River Laboratories, USA) were used for all
experiments. Procedures involving animals and their care were
conducted in conformity with UCLA institutional guidelines.
Animals were lightly anesthetized with isofluorane after placement in a plastic restraint cone. The following 5 adipose grafts
were injected:
1. Fresh grafts—no storage.
2. Frozen grafts—adipose tissue frozen for up to 8 weeks at
−20◦ C in PBS.
3. Cryo#4 grafts—grafts frozen for up to 8 weeks in Cryo#4
cryopreservation medium.
4. Fresh + ASCs—fresh grafts combined with cultured ASCs.
5. Frozen + ASCs—grafts frozen for 1 week in PBS combined
with cultured ASCs.
Graft volumes of 1 ml were injected subcutaneously into the
dorsal panniculus carnosus by means of an 18-gauge needle and
a 1 ml sterile syringe. Four separate grafts were injected simultaneously into each rat. For the implantation of adipose grafts plus
cultured ASCs, a 1 ml syringe was loaded with the desired adipose graft. ASC pellets at a density of 500,000 cells were resuspended in 100 ul of PBS and the cell suspension loaded into the
same syringe containing the adipose tissue graft. Animals were
euthanized after 2, 4, or 8 weeks and the injected grafts recovered
for histologic examination. During all stages of the experiment,
animals were kept in standard cages at 25◦ C and provided with
natural lighting and pellet food and tap water ad libitum.
Analysis of Adipose Tissue Graft Structure, Retention,
and Viability
Fresh and frozen adipose grafts were either implanted into
athymic rats or placed into culture for in vitro analysis of viability and morphology. To assess the morphology of implanted
grafts, animals were sacrificed after 2, 4, and 8 weeks and the
grafts excised. Morphology was assessed using standard hematoxylin and eosin staining (H&E staining). The presence of human cells within the implanted grafts was confirmed using Alu
in situ hybridization. To calculate retention of these implanted
grafts, all grafts were weighed prior to implantation and immediately after harvest. Tissue retention in the harvested grafts
(i.e., retention level) was calculated as the harvested graft weight
expressed as percentage of the graft weight prior to implantation.
Average retention levels were expressed as the mean ± SD.
To assess graft morphology and viability ex vivo prior to
implantation, fresh adipose grafts were washed in 1× PBS and
placed into standard tissue culture conditions for 24 hr, 7 days,
or 14 days in CM without any media changes. For the frozen
grafts, the tissue was thawed for 15 min in a 37◦ C water bath,
washed in 1× PBS, and placed into culture for 24 hr, 7 days,
or 14 days prior to analysis. The grafts were removed from
culture and their morphology assessed histologically by H&E

staining. Viability was also assessed using commercial kits for
the assessment of apoptosis (TUNEL assay) and general cell
death (Live/Dead staining).
H&E Morphology
Adipose grafts were fixed overnight in 10% formalin and then
embedded in paraffin, followed by sectioning at a thickness of
5 µm. Sections were deparaffinized in three changes of 100%
xylene, followed by rehydration in decreasing grades of ethanol
(100%, 95%, and 70%, 5 min each). The sections were rinsed
in distilled water for 5 min followed by a 3-min staining in
hematoxylin (Harris hematoxylin, Sigma Diagnostics, St. Louis,
MO, USA). The sections were rinsed in fresh distilled water for
5 min and then stained for 2 min in eosin (Sigma Diagnostics).
The sections were washed again in water, dehydrated back
through increasing grades of ethanol, and incubated in 100%
xylene for 5 min. The sections were mounted with Cytoseal
(Fisher Scientific, Pittsburg, PA, USA), coverslipped, and
visualized under a Zeiss Axioskop microscope (Carl Zeiss,
Thornwood, NY, USA) fitted with a SPOT2 digital camera
(Diagnostic Instruments, Sterling Heights, MI, USA). Digital
images were acquired at 100× magnification unless specifically
noted otherwise.
Live/Dead Staining
Viability was assessed in the deparaffinized sections using a
Live/Dead viability stain according to the manufacturer (cat# L3224; Molecular Probes, Eugene, OR, USA). In this assay, ethidium homodimer-1 enters cells with damaged membranes (i.e.,
dying or dead cells) and undergoes a 40-fold enhancement of
fluorescence upon binding to nucleic acids, producing an intense
red fluorescent signal detectable using a conventional RITCfilter set. The intact plasma membrane of a viable cell completely
excludes the entrance of the ethidium homodimer and results in
a low level of background fluorescence due to the inactivity of
the homodimer in the absence of nucleic acid binding.
For this assay, fresh and frozen adipose grafts were removed
from culture and incubated at 37◦ C for 1 hr in CM containing
a mixture of 4 mM calcein and 2 mM ethidium homodimer-1.
The grafts were then fixed in 10% and embedded in paraffin as
above. To quantitate cell death, several sections were prepared
from fresh grafts maintained in culture and from grafts frozen in
PBS and processed using the Live/Dead stain described above.
The number of dead cells (red fluorescence) was quantified
in each image along with the total number of cells within the
section (DAPI stained nuclei). Cell death levels were determined
by expressing the total number of dead cells as a percentage of
total cells counted in each image. The level of cell death was
determined in each processed section and averaged for each
graft. Levels were expressed as the mean percentage ± SD.
TUNEL Assays
DNA fragmentation as an indication of apoptosis/cell death
was also assessed in deparaffinized sections using the Fragel

Connect Tissue Res Downloaded from informahealthcare.com by Universitat Autonoma Barcelona on 11/04/14
For personal use only.

FREEZING ADIPOSE TISSUE GRAFTS

DNA Fragmentation System to detect the presence of fragmented DNA that accompanies cell death (cat#QIA21-1EA;
EMD Biosciences, San Diego, CA, USA). For this, fresh and
frozen grafts were removed from culture, washed in 1× PBS,
and fixed in 10% formalin followed by paraffin sectioning. Dead
cells within the graft sections were detected fluorescently using
an RITC-filter set. For both the Live/Dead and TUNEL assays,
the processed sections were mounted with a DAPI-containing
mounting medium to minimize fluorescent quenching (Vectastain Hardmount, Vector Laboratories, Burlingame, CA, USA).
Fluorescent images at 100× magnification were acquired using
a SPOT2 camera. Cell death levels (green fluorescence) were
determined as for the Live/Dead studies.
Alu In Situ Hybridization
To confirm the presence of human cells within implanted
grafts, graft sections were processed by in situ hybridization
using a commercial kit for the detection of the human
genetic marker Alu. Grafts were embedded in paraffin and
multiple sections throughout the graft processed according to
the manufacturer (Biogenex Universal ISH kit, Biogenex, San
Ramon, CA, USA). An FITC-conjugated Alu oligonucleotide
(Biogenex) was hybridized to the sections and the hybridized
oligo detected using biotinylated anti-FITC antibodies according to the manufacturer. Endogenous rat adipose tissue and
sections processed without the Alu oligonucleotide were used
as negative controls (not shown). Human tissue was processed
as a positive control (not shown). Positive hybridization of the
oligo to the Alu sequence is shown as a brown staining within
the nuclei. All processed sections were counter stained with
hematoxylin. Images were acquired with a Spot2 Digital camera
at 200× and 400× magnification.
VEGF ELISA (Enzyme-Linked ImmunoSorbent Assay)
VEGF synthesis and secretion was quantitated in the
following groups: (1) fresh adipose tissue—in vitro culture,
(2) ASCs—in vitro culture, (3) frozen adipose tissue, frozen
in PBS—7 and 14 days, (4) frozen, cryopreserved adipose
tissue—7 and 14 days, and (5) cryopreserved ASC cultures.
For the fresh samples (i.e., adipose tissue and ASC cultures),
cells and tissue were placed into culture after processing and
CM was collected after 24 hr and 7 days continuous culture
(i.e., no media changes). For the cryopreserved samples, grafts
and cells were thawed, washed in 1× PBS, and placed back
in culture in CM. The media was collected after 24 hr and 7
days as for the fresh samples. VEGF was quantitated using
a commercial ELISA assay for the quantitation of the 165
amino acid isoform of VEGF (VEGF165) (Human VEGF
Quantikine ELISA Kits, R&D Systems, Minneapolis, MN,
USA). A standard curve for VEGF165 was created and used
to quantitate the amount of VEGF secreted. VEGF synthesis
and secretion were quantitated for each 1 ml adipose tissue
graft and expressed as pg VEGF/ml adipose tissue. For the ASC

17

cultures, VEGF was quantitated and expressed as pg VEGF/cell.
Each sample was analyzed for VEGF production a minimum of
three times and the VEGF level expressed as the average ± SD.
Adipose tissue samples and/or ASC cultures from 5 patients was
assessed.

RESULTS
Implantation of Adipose Tissue Grafts—Effect of Storage
at Subzero Temperatures
Adipose tissue harvested from several sites within the
athymic rat (e.g., inguinal, scapular) could be characterized
histologically as typical white adipose tissue (WAT) by the presence of multiple, small adipocytes of uniform size interspersed
with small amounts of fibrous tissue (Figure 1A). Human
adipose grafts injected subcutaneously in athymic rats without
storage (i.e., “fresh grafts” [Figure 1B]) showed a similar
morphology when harvested after 2 weeks with numerous
well-formed adipoctyes and limited fibrous tissue found within
the grafts. Despite excellent adipose morphology in these fresh
grafts, graft retention levels were significantly affected over this

FIG. 1. Freezing of adipose tissue grafts affects their morphology and
retention in vivo. (A) H&E morphologic assessment of endogenous rat adipose
tissue. (B through D) Fresh grafts: (B) Morphology of fresh human adipose
grafts implanted for 2 weeks; (C) 4 weeks, and (D) 8 weeks. (E through G)
Frozen grafts 1 week: Morphology of human adipose grafts frozen in PBS for 1
week prior to implantation. (E) Morphology after 2 weeks, (F) 4 weeks, and (G)
8 weeks implantation. (H and I) Frozen grafts 8 weeks: Morphology of human
adipose grafts frozen in PBS for 8 week prior to implantation. (H) Morphology
after 2 weeks and (I) 4 weeks implantation. No tissue was found after 8 weeks.
The presence of human cells using Alu in situ hybridization (brown nuclei—see
arrows) is shown for both 2 and 4 weeks (inset pictures). Magnification = 100×.

18

N. GREWAL ET AL.

TABLE 2
Retention levels upon cryostorage of adipose tissue grafts
Graft type
Frozen grafts—PBS∗
Fresh graft∗

1 week or less

2 weeks

51.33 ± 5.03%

38.50 ± 10.60%

4 weeks

49.33 ± 15.53%

44.50 ± 12.00%

8 weeks

Fibrotic tissue

No tissue

Connect Tissue Res Downloaded from informahealthcare.com by Universitat Autonoma Barcelona on 11/04/14
For personal use only.

Harvest time

∗

8 weeks
1/3 grafts—26.00%
2/3 grafts—no tissue
1/5 grafts—4.00%
4/5 grafts—no tissue
No tissue

Frozen grafts—CryoM#4∗
1 week or less
51.50 ± 21.92%
73.50 ± 20.5%
16.67 ± 11.22%

8 weeks
47.00 ± 21.37%
(all grafts retained)
27.23 ± 4.99%
1/5 grafts—no
tissue
No tissue

Graft volume retained in vivo as a percentage of original implanted graft volume.

2-week time point with an average retention volume of 51.00%
± 5.03 of the original implanted graft being measured upon harvest (Table 2). This rate of resorption did not appear to increase
with time as a similar level of graft retention was measured after
4 weeks implantation (Table 2—49.3% ± 15.53). However,
graft volumes were more variable at this 4-week time point
(range = 32% to 62% retention). Harvested 4-week fresh grafts
were characterized histologically by a significant reduction in
the number of small, well-formed adipocytes. Rather, these
grafts demonstrated an increase in larger adipocytes together
with an obvious increase in cellular infiltrate from the host
(Figure 1C).
To confirm that the adipose grafts retained within the rat were
of human origin, the expression of the human genetic marker
Alu was confirmed using in situ hybridization. As shown in
the inset picture, 4-week fresh grafts could be confirmed as
human in origin through their expression of the Alu marker
(brown-stained nuclei). Interestingly, the expression of the Alu
marker appeared to be more abundant within the fibrous portion
of the graft, suggesting a time-dependent conversion of adipose
tissue to fibrotic (Figure 1C, inset picture, arrowheads). The
Alu staining was specific to human tissue, as rat tissue failed
to show any Alu hybridization (not shown). In addition, the
Alu protocol exhibited low background levels, as negligible Alu
hybridization was observed in graft sections processed in the
absence of the Alu oligonucleotide (not shown). Increasing the
harvest time of these fresh grafts to 8 weeks resulted in few
grafts being retained with the majority completely resorbed into
the host. However, in a small number of animals, a small amount
of tissue could be harvested after 8 weeks. However, remnants of
these grafts were characterized by the presence of fibrous tissue
(Figure 1D)
To determine if graft longevity in vivo is affected by freezing
of the grafts, tissue was stored at −20◦ C in sterile saline (i.e.,
PBS) for two time periods: up to 1 week (i.e., 24 hr to 7 days)
or 2 months (i.e., 8 weeks). Grafts were frozen for 24 hr to
determine if the process of freezing itself damages the graft
while 1 week was chosen to study the effects of short-term

storage. Since fat grafts for clinical use are often stored frozen
for long periods of time, grafts were also frozen for 2 months.
When examined histologically, freezing in PBS appeared to
dramatically affect the structure of the graft. Specifically, 1week frozen grafts harvested after both 2 and 4 weeks exhibited
decreases in adipocyte number and increased fibrous tissue
located around the adipocytes (Figures 1E and 1F). Frozen grafts
harvested after 4 weeks also exhibited several adipocytes that
appeared to either enlarge or coalesce into “giant” adipocytes
(Figure 1F—arrows). As with the fresh grafts, the majority of
frozen grafts failed to persist at the 8-week time point. However,
some fibrotic tissue remnants could be found in a small percentage of animals (Figure 1G). Despite these obvious changes in
structure, freezing of grafts for 1 week in PBS only slightly
affected the level of graft resorption when compared to fresh
grafts.
To calculate graft resorption (i.e., retention level), grafts were
weighed immediately before implantation and immediately
after harvest. The harvested graft weight was expressed as
a percentage of the weight prior to implantation and this
value used to quantify retention levels. Using this method,
frozen grafts harvested after 2 and 4 weeks had an average
retention level of 38.5% ± 10.6 and 44.5% ± 12.0, respectively
(Table 2). In comparison, retention levels of fresh grafts
harvested at the same time points were 51.0% ± 10.6 and
49.3% ± 15.5, suggesting that while freezing in PBS may
affect the graft’s structure, overall graft volume may not be
affected over the short term. However, the time of storage
in PBS appeared to dramatically affect graft morphology and
retention. Grafts frozen in PBS for 2 months (i.e., 8-week frozen
grafts) and implanted for only 2 weeks exhibited a reduction in
adipocyte number, a loss of adipocyte structure (i.e., appearance
of “giant” adipocytes), and an obvious increase in fibrotic tissue
in comparison to 1-week frozen grafts harvested at the same
time point (compare Figures 1H to 1E).
More important, the morphology of these 8-week frozen
grafts after 2 weeks in vivo was similar to that observed in
1-week frozen grafts harvested after 4 weeks (compare Figure

Connect Tissue Res Downloaded from informahealthcare.com by Universitat Autonoma Barcelona on 11/04/14
For personal use only.

FREEZING ADIPOSE TISSUE GRAFTS

1H and 1F), suggesting that long-term storage in PBS can
significantly damage the graft’s structure and accelerate its
fibrotic conversion. Consistent with this, storing these grafts in
PBS for 2 months was also capable of affecting graft retention,
accelerating their loss within the host. When quantitated at
the 4-week harvest time point, only 1 of 5 animals implanted
with 8-week frozen grafts showed any tissue retention. In the
one graft that did persist, an approximate 95% reduction in
graft weight was measured (40 mg graft retained of 1000 mg
injected graft). Moreover, the remnant graft was characterized
histologically by an almost complete absence of adipocytes
and a significant amount of fibrotic tissue within the graft
(Figure 1I). In contrast, each animal injected with a 1-week
frozen graft retained some portion of the adipose tissue graft
after 4 weeks (see Figure 1F). Not surprisingly, little to no
tissue graft remained after 8-week frozen grafts were harvested
at the 8-week time point.
In light of these histologic results, fresh and frozen adipose
grafts were examined morphologically prior to injection to
assess the level of damage possibly imposed by ex vivo
maintenance or storage. For this, two storage conditions were
examined: maintenance ex vivo in a nutritive environment
to maintain graft viability—in vitro culture at 37◦ C in a
standard tissue culture growth medium (i.e., control medium)
and maintenance ex vivo after freezing in PBS. Using H&E
staining to assess morphology, fresh adipose tissue grafts
maintained in vitro for 24 hr in CM were found to exhibit
a morphology typical of adipose tissue with well-formed

19

adipocytes of consistent size (Figure 2A—top left). Qualitative
assessment of cell death within these freshly harvested grafts
using a conventional TUNEL assay (middle row), in addition to
an immunofluorescent protocol that detects dead cells in general
(i.e., Live/Dead stain−bottom row), confirmed a relatively low
level of apoptotic/dead cells. However, small areas of apoptotic
cells observed via TUNEL assay could be observed in the
freshly isolated grafts (see white arrows), suggesting that despite
maintenance in a nutritive environment, the simple act of
removing the graft from the donor tissue may be sufficient to
induce some cell death.
Consistent with this, maintenance of the grafts for increasing
time periods ex vivo (i.e., 7 and 14 days in CM) appeared to
increase overall tissue degradation. Specifically, in both the 7and 14-day grafts, slight changes in adipocyte shape and size
were observed with the adipocytes appearing to lose some of
their plasma membrane integrity and well-ordered arrangement
(Figure 2A, top row—see black arrowheads). In agreement with
this loss of morphology, the 7 and 14 days fresh adipose grafts
also appeared to show increased levels of dead cells within
the tissue (white arrows). When quantitated, the 7- and 14day grafts contained significantly more dead cells (7 day =
46.97 ± 14.68% dead cells; 14 days = 38.98 ± 8.81% dead
cells) when compared with fresh grafts maintained for 24 hr
in vitro (24 hr = 2.15 ± 1.34% dead cells; 7 days vs. 24 hr p =
0.06; 14 day vs. 24 hr p = 0.05), suggesting that adipose grafts
can be maintained for a limited amount of time ex vivo before
cellular death increases.

FIG. 2. Maintenance of adipose grafts ex vivo affects their morphology. (A) Top row: H&E staining of fresh adipose grafts maintained ex vivo under standard
tissue culture conditions for 24 hr (Fresh—24 hr), 1 week (Fresh—1 wk) and 2 weeks (Fresh—2 wks). Middle row: Assessment of apoptosis using a fluorescent
TUNEL assay (TUNEL—apoptotic cells shown as green, white arrows). Bottom row: Assessment of cell death using a Live/Dead fluorescent stain (L/D—dead
cells shown in red, white arrowheads). (B) Top row: H&E staining of adipose grafts frozen in PBS for (Frozen—24 hr), 1 week (Frozen—1 wk) and 2 weeks
(Frozen—2 wk) and placed into culture for 24 hr prior to analysis. Significant changes to adipocyte morphology upon storage at subzero temperatures is shown
as black arrows. Middle row: Assessment of apoptosis using a fluorescent TUNEL assay (TUNEL—green, white arrows). Bottom row: Assessment of cell death
using a Live/Dead fluorescent stain (L/D—red, white arrowheads). Magnification = 100×.

Connect Tissue Res Downloaded from informahealthcare.com by Universitat Autonoma Barcelona on 11/04/14
For personal use only.

20

N. GREWAL ET AL.

While the maintenance of adipose grafts for 24 hr in vitro in
CM appeared to affect the graft’s structure minimally, freezing
of grafts in PBS for as little as 24 hr significantly affected overall
adipocyte morphology as shown by H&E staining, suggesting
that freezing can damage the adipose graft. Moreover, grafts
frozen for 24 hr, 7 days, or 14 days all appeared to be similar
histologically, showing an increased loss of adipocyte integrity
and number (Figure 2B—top row, see black arrows), suggesting
that the tissue damage is incurred at the cellular level very early
in the freezing process.
As measured in the fresh adipose grafts, increasing amounts
of cell death could be measured between the 24-hr and 7/14-day
frozen grafts (24 hr = 5.55 ± 3.61% dead cells; 7 days =
34.50 ± 9.33% dead cells; 14 days = 77.60 ± 10.65% dead
cells). Interestingly, no significant difference in cell death could
be measured between the 24-hr and 7-day frozen grafts and
those fresh grafts maintained for similar time points in culture
(24 hr; p = 0.25; 7 days; p = 0.34), suggesting that, over the
short-term, storage time itself may induce cell death not the
storage condition (i.e., fresh vs. frozen). However, a moderate
increase in cell death was measured as grafts were frozen
for 2 weeks (compare fresh 14 days = 38.98 ± 8.81% dead
cells and frozen 14 days = 77.60 ± 10.65% dead cells, p
= 0.04), indicating that storage condition may play a role in
graft structure over longer time periods. Taken together, the data
indicate that the storage condition can dramatically affect graft
morphology and, as such, may affect overall retention levels
over time within the recipient.

VEGF Secretion by Fresh and Frozen Adipose Grafts
Increased adipose graft retention is likely to depend upon the
extent of vascularization from the host. Numerous studies have
characterized the role of VEGF in the angiogenesis of tissues
and suggest that it is critical for the retention of donor grafts
[7–9]. Based on this, the amount of VEGF synthesized and
secreted by adipose grafts was measured by ELISA analysis.
Fresh adipose grafts maintained for 24 hr in CM (i.e., control

FIG. 3. Freezing of adipose grafts dramatically decreases their VEGF
synthetic ability. Synthesis and secretion of VEGF into culture medium by
fresh and frozen adipose grafts. VEGF secretion by 1 ml adipose grafts was
quantitated by ELISA analysis and expressed as pg VEGF secreted per ml
culture medium (pg/ml). VEGF secretion by fresh adipose grafts upon 24 hr
culture in CM (Fresh 24 hr culture) and after 7 days culture (Fresh 7 days
culture) is shown along with VEGF secretion by adipose grafts frozen for
1 week in PBS and placed into culture for 24 hr (Frozen 24 hr culture). No
VEGF secretion could be measured in frozen grafts culture for an additional 6
days ∗ Frozen 7 days culture.

grafts) secreted an average of 37.60 pg/ml VEGF into the
culture medium (Table 3—fresh adipose grafts). Increasing the
culture time of these fresh grafts from 1 day to 7 days resulted
in a corresponding increase in VEGF secretion into the CM
(Figure 3 and Table 3—approximately 40-fold increase).
However, the metabolic activity of adipose grafts maintained
ex vivo appeared to be limited as no significant difference in
the VEGF levels measured at the 7-day and 14-day time points
could be discerned (14-day data not shown).
In contrast to fresh adipose grafts, storage of adipose tissue
frozen in PBS for 1 week (Figure 3 and Table 3—frozen
adipose grafts) decreased their VEGF synthetic ability, reducing
secretion over 3.6-fold when placed back into culture for 24 hr
(fresh graft = 37.60 pg/ml ± 2.06 vs. frozen graft = 10.41 pg/
ml ± 5.80). Moreover, freezing of the tissue appeared to
significantly damage the VEGF synthetic machinery as VEGF

TABLE 3
Metabolic activity of adipose tissue grafts—VEGF synthesis
VEGF secretion—pg/ml
(culture time)

Fresh adipose grafts

Day 1

Day 7

37.60 ± 2.06

1260.31 ± 310.86

VEGF secretion—pg/ml
(freezing time/culture time)

Frozen adipose grafts (PBS)

1 week/day 1

1 week/day 7

2 weeks/day 1

2 weeks/day 7

10.41 ± 5.80

−4.51 ± 8.59

−14.91 ± 3.21

−8.75 ± 15.04

21

FREEZING ADIPOSE TISSUE GRAFTS

TABLE 4
VEGF synthetic levels by fresh and cryopreserved adipose-derived stem cells (ASCs)
Culture condition/ culture time (ELISA)

Connect Tissue Res Downloaded from informahealthcare.com by Universitat Autonoma Barcelona on 11/04/14
For personal use only.

VEGF secretion
(pg/ml) per ASC
Average ASC # per ELISA
-fold change VEGF
(frozen vs. fresh ASC)

Fresh
ASCs—day 1

Fresh
ASCs—day 7

Frozen
ASCs—day 1

Frozen
ASCs—day 7

2.19 ×103 ±
8.4 ×10-4
20,000

1.45 ×10-4 ±
1.24 ×10-5
900,000

2.16 ×10-4 ±
4.0 ×10-6
250,000

7.44 ×10-4 ±
1.39 ×10-4
1.80 ×106

−9.43 ± 0.017

5.36 ± 2.00

secretion was lost completely upon continued culturing of the
frozen grafts for an additional 6 days in vitro (7 days total). Total
freezing time also appeared to damage graft viability as storage
of frozen grafts in PBS for 2 weeks resulted in a complete
inhibition of VEGF secretion after reintroduction to culture for
either 24 hr or 7 days.
This loss of VEGF secretion by frozen grafts is likely due to
damage at the cellular level. To assess this, the amount of VEGF
synthesized by the ASC population within the adipose graft
was also measured. ASC cultures maintained for 24 hr in CM
secreted an average of 0.0022 pg/ml VEGF per individual ASC
(Table 4), a level comparable to that previously reported by
Rehman and colleagues (0.0012 pg/ml VEGF per cell) [10].
Because the harvesting of cultures with trypsin has been shown
to affect their activity [11], VEGF secretion was also measured
after an additional 6 days in culture (i.e., 7 days total). The
amount of VEGF secreted per cell by ASC cultures at this
stage decreased an average of 14.3-fold (i.e., 14.35 ± 4.27-fold
decrease) when compared with the 24-hour values (see Table 4).
Since VEGF secretion was normalized to ASC cell number, this
decrease may be due to the increased levels of ASC proliferation
that occurred during the additional 6 days of culture (average
cell number day 1 = 20,000, average cell number day 7 =
900,000) and would suggest that proliferating cultures of ASCs
decrease their synthetic capacity.
Consistent with the freezing of adipose grafts, cryopreservation of ASC cultures followed by their reintroduction into
culture for 24 hr resulted in a significant reduction of VEGF
synthesis per ASC (9.43-fold decrease). However, this decrease
was short-lived as continued culturing of ASCs resulted in
a dramatic improvement in VEGF synthesis per ASC. This
period of increased VEGF synthesis was also associated with
a significant level of ASC proliferation resulting in a 5.4-fold
increase in VEGF vs. fresh ASCs upon normalization to ASC
number. The reason for the increase in VEGF synthesis 7 days
after cryostorage is unclear but was observed in many cryostored
populations (n = 4).
It is possible that reintroducing cryostored ASCs to culture
is followed by an intensive period of biosynthetic capacity

(e.g., growth factor synthesis) that precedes an equally intensive
period of proliferation. If this biosynthetic period is strong
enough, it could produce sufficient levels of VEGF such that an
increase in normalized VEGF production would be measured.
Taken together, the data suggest that freezing can affect the
structure and viability of adipose tissue grafts and its resident
ASC population. Therefore, use of conventional saline solutions
may not be an appropriate cryopreservation media if cellular
metabolic activity is to be preserved.

Cryopreservation of Adipose Grafts
Based on these data, adipose grafts were frozen for either
1 or 2 weeks in several cryostorage conditions to try and
preserve their VEGF secretory activity. As shown in Table 1, the
majority of cryostorage solutions were unable to preserve the
VEGF secretory activity of adipose grafts, with most grafts
losing their ability to secrete VEGF consistently after only
1 week storage. However, Cryostorage Media #1 (Cryo#1
= 50% DMEM, 40% FBS, 10% DMSO) did appear to
preserve some VEGF secretory ability after 1-week storage,
with those grafts secreting comparable amounts of VEGF as
that measured in the control groups (PBS only). However, this
ability was lost upon continued culturing in vitro. Moreover,
storage in Cryo#1 for 2 weeks also appeared to eliminate any
benefit this media provided with no VEGF production being
measured.
A more consistent cryopreservation of the grafts was
observed for Cyro#4 (30% sucrose in PBS). Grafts stored in
this media condition for 1 week secreted comparable amounts
of VEGF after 24 hr in culture as that measured in the control
groups (PBS). However, in contrast to control grafts, the Cryo#4
grafts continued to secrete VEGF into the culture medium for
an additional 6 days in culture (i.e., 7 days total). Moreover,
this media appeared to allow for longer storage of the grafts as
VEGF secretion could still be measured after 2 weeks freezing
and 7 days in culture. In fact, of the 6 storage conditions tested,
Cryo#4 was the only condition to preserve significant levels of
VEGF secretion after 2 weeks cryostorage. As such, the use

Connect Tissue Res Downloaded from informahealthcare.com by Universitat Autonoma Barcelona on 11/04/14
For personal use only.

22

N. GREWAL ET AL.

FIG. 4. Cryostorage of adipose grafts may improve their retention within the host. Morphologic analysis of implanted adipose grafts with and without cryostorage.
H&E analysis of adipose grafts stored for 24 hr or 1 week in PBS (PBS 24 hr, PBS 1 week, respectively) and harvested after 2, 4, or 8 weeks. Analysis of grafts
stored in PBS for 8 weeks yielded no tissue upon implantation at 2, 4, and 8 weeks and is not shown. H&E analysis of adipose grafts stored for 24 hr, 1 week, or
8 weeks in Cryo#4 (Cryo#4 24 hr, Cryo#4 1 week and Cryo#4 8 weeks, respectively) and harvested after 2, 4, or 8 weeks. Morphology from two representative
Cryo#4 grafts after 8 weeks shows a range of fibrous tissue infiltrate (K. and inset picture). Bottom row: H&E analysis of fresh human adipose grafts before
implantation is shown as a positive control (Fresh fat) together with tissue formed upon injection of Cryo#4 only (Cryo#4). The presence of blood vessels within
Cryo#4 grafts is shown (black arrows). Magnification = 100×.

of a cryopreservation media like 30% sucrose may enhance or
prolong the ability of stored adipose grafts to secrete VEGF over
conventionally used saline solutions.
Prolonging VEGF secretory ability by fat grafts through
the use of cryopreservation media may translate in vivo into
increased vascularization and enhanced survival and retention
clinically. To assess this, fat grafts cryopreserved for up to
2 months in Cryo#4 were injected subcutaneously into athymic
rats and fat pad formation assessed morphologically. As shown
in Figure 4, storage of adipose grafts in Cryo#4 appeared to
improve graft retention and morphology. While no discernible
difference in graft retention levels after 2 weeks in vivo could
be discerned between grafts frozen for short time periods (i.e.,
less than 1 week) in PBS or Cryo#4 (see Table 2), a distinct
advantage in using Cryo#4 became evident as implantation time
increased. Freezing of adipose grafts for 1 week in PBS resulted
in an average retention level of approximately 45% after 4
weeks in vivo (Table 2). However, the storage of grafts for 1
week in Cryo#4 improved retention levels to 73.5% ± 20.5
when measured at this same time point. A more pronounced
retention effect by Cryo#4 was seen after 8 weeks in vivo. At
this harvest point, no tissue could be found upon implantation
of grafts frozen for up to 1 week in PBS. However, a small

amount of graft retention (16.67 ± 11.72%) was measured
upon implantation of grafts frozen in Cryo#4. Consistent with
these data, short-term storage in Cryo#4 (i.e., 24 hr and 1
week) also appeared to improve graft morphology, with Cryo#4
appearing to lessen the formation of fibrotic tissue within the
graft after both 2 and 4 weeks in vivo and improving overall
adipose structure when compared with storage in PBS (Figure 4,
compare 4B/4G—Cryo#4 24 hr to 4A/4F—PBS 24 hr, compare
4D/4I Cryo#4 1 week to 4C/4H—PBS 1 week).
Moreover, grafts stored in Cryo#4 and harvested after
4 weeks were positive for the presence of small blood vessels
(see arrows), suggesting that these cryostored grafts may become vascularized with time. Short-term storage in Cryo#4 also
appeared to improve the morphology of grafts harvested after
2 months with small amounts of adipose tissue still being
retained within these long-term grafts (Figure 4K). The presence
of adipose tissue in Cryo#4 grafts after 8 weeks in vivo was in
direct contrast to animals implanted with grafts frozen for up
to 1 week in PBS. In these animals, the grafts were completely
resorbed after 8 weeks in vivo with no discernible tissue present
at the implantation site (Figure 4—no tissue found), suggesting
that Cryo#4 may be able to improve long term graft retention in
comparison to conventional saline.

Connect Tissue Res Downloaded from informahealthcare.com by Universitat Autonoma Barcelona on 11/04/14
For personal use only.

FREEZING ADIPOSE TISSUE GRAFTS

However, while Cryo#4 improved the morphology and
retention levels of frozen adipose grafts in comparison to PBS,
a time-dependent storage effect on graft morphology became
apparent for this cryopreservation medium when grafts stored
for 8 weeks were compared with grafts frozen for 1 week or less.
This storage effect was especially evident for PBS. Specifically,
storage of grafts in PBS for 8 weeks resulted in the resorption
of the majority of grafts at both the 2- and 4-week harvest
point (Table 2—2 out of 3 grafts, 4 out of 5 grafts resorbed,
respectively). In those animals where small remnants of graft
remained, the harvested tissue showed no evidence of adipocytes
and was completely fibrous in nature (data not shown). Grafts
frozen in PBS for 8 weeks and harvested after 2 months were
completely resorbed into the host. Such results suggest that PBS
may not be sufficient for long-term storage of adipose tissue. In
contrast, grafts stored for 8 weeks in Cryo#4 were retained at
a level of 47.00 ± 21.37% of the original graft after 2 weeks
and preserved at an average level of 27.23 ± 4.99% in 4 out of
5 implanted grafts after 4 weeks. However, when weights from
grafts stored for less than 1 week in Cryo#4 (73.5% retention)
were compared after 4 weeks to those from grafts stored for
2 months at this same time point (27.0% retention), a dramatic
decrease in retention in vivo was observed indicating that storage
time in Cryo#4 can affect grafts in vivo.
Finally, the effect of storage time was also apparent when
graft morphology was assessed histologically, with a dramatic
loss of adipocyte morphology and a significant increase in

23

fibrotic tissue being observed in 8-week Cryo#4 grafts implanted
for 2 and 4 weeks (Figures 4E and 4J, respectively). Taken
together, these data suggest that while cryopreservation of
adipose grafts in Cryo#4 may improve the overall morphology
of the graft and increase its retention in vivo, the time of storage
in this medium must be carefully considered if adipose structure
is to be maintained.
ASCs may Improve Retention of Fresh Adipose Grafts
In Vivo
If retention of adipose grafts is thought to be improved
through vascularization, an alternative to cryopreservation of
adipose grafts may be the introduction of ASCs with their
enhanced production of VEGF to the fresh adipose graft
immediately upon implantation. Injection of ASCs alone was
capable of producing adipose pads at both 2 and 4 weeks
time with excellent adipose morphology (Figure 5—ASCs).
However, the formation of these adipose pockets was extremely
inconsistent and observed in only a small fraction of the rat
population (less than 10% of total rats). In addition, the resulting
fat pads were very small, making the use of ASCs alone in
soft tissue enhancement limited. However, the combination
of ASCs with fresh adipose grafts (Figure 5—ASC + fresh)
did appear to improve the resulting graft morphology in vivo.
Specifically, the number of adipocytes and their arrangement
appeared to be enhanced in ASC + fresh grafts after 2 weeks
when compared with implantation of fresh grafts alone (Figure

FIG. 5. Adipose-derived stem cells (ASCs) may improve the morphology of implanted adipose grafts. Morphologic analysis of adipose tissue after 2 or 4
weeks resulting from the injection of ASCs or implantation of adipose grafts in combination with ASCs. Adipose tissue resulting from injection of ASCs only
(ASCs), fresh adipose grafts (Fresh fat), fresh adipose grafts combined with ASCs (Fresh + ASCs), grafts frozen in PBS (Frozen), and frozen grafts combined
with ASCs (Frozen + ASCs). Blood vessels are shown (back arrows). The presence of human ASCs within the harvested adipose tissue is confirmed using an
immunofluorescent DiI lipid stain (DiI—red fluorescence, DAPI—blue nuclei, see white arrows) or in situ hybridization for the Alu sequence (Alu—brown, black
arrowheads). Magnification = 100× unless specified.

Connect Tissue Res Downloaded from informahealthcare.com by Universitat Autonoma Barcelona on 11/04/14
For personal use only.

24

N. GREWAL ET AL.

5—fresh). This improvement was also noted in 4-week grafts
in which the presence of ASCs decreased the total amount
of fibrotic tissue within the graft and appeared to increase
the number of adipocytes when compared with fresh graft
control. To confirm retention of ASCs within these grafts
in vivo, injected ASCs were first labeled with the fluorescent
dye DiI prior to combination with the adipose graft. Consistent
with their retention within the graft, the presence of ASCs
could be observed throughout both the 2 and 4 weeks grafts
but were specifically observed in areas containing well-formed
adipocytes (see arrows). Such results could suggest that ASCs
could be used as an alternative to cryopreservation.
Alternatively, ASCs could be used in combination with
cryopreservation. In addition to improving the structure of
fresh adipose grafts, the presence of ASCs also appeared to
improve the structure of adipose grafts frozen in PBS (Figure
5—ASC + frozen). Specifically, the presence of ASCs within
the frozen graft lessened the formation of fibrotic tissue within
the graft after 4-week implantation. However, despite being an
excellent source of VEGF in vitro, the presence of ASCs within
these frozen or fresh “combination” grafts did not result in any
observable increases in graft vascularization when compared to
the fresh and frozen graft controls. Such a lack of vascularization
may indicate the eventual failure of these grafts to integrate
within the host tissue.
Taken together, the data in this study suggest that the
freezing of adipose grafts in conventional saline solutions may
damage their structure and retention in vivo over the long term.
While this damage may be ameliorated through the use of a
cryopreservation medium that retains the VEGF synthetic ability
of the graft, storage time was an important factor to consider,
with long-term storage in either PBS or Cryo#4 producing poor
retention levels and adipose morphology. As an alternative,
the structure and retention of fresh or frozen adipose grafts
may be improved through their combination with cultured
ASCs—an excellent source of VEGF. However, while ASCs did
synthesize extensive amounts of VEGF in vitro, no improvement
of vascularization could be observed in these “combination”
grafts suggesting that their viability and retention in vivo will
likely be limited.

DISCUSSION
This study attempts to determine if the freezing of adipose
tissue grafts is a viable option for long-term storage of
adipose tissue. The successful storage of adipose tissue in a
relatively undamaged state is a desirable outcome for both the
reconstructive and aesthetic plastic surgeon as adipose tissue
represents the optimal soft tissue filler. Although there is not
complete agreement in the literature, certain grafting techniques
such as the Coleman procedure [12–14], involving the injection
of small amounts of fat (i.e., less than 0.1 ml per injection) into
a defect using multiple injections to fill the defect, are more
successful than others. The use of smaller volumes of fat for

injection such as that used in the Coleman procedure may expose
the grafted tissue to a larger surface area for nutrient exchange,
thus increasing its chances of retention within the defect.
However, the filling of larger defects requiring larger volumes
of fat may not be amenable to procedures such as this. Current
surgical therapies for the correction of soft-tissue defects using
larger volume fat grafts involves immediate injection of fat
harvested by simple syringe aspiration into the defect followed
by a course of repeated injections at defined intervals over the
course of weeks or months. Such repeated injections are required
to ameliorate the loss of graft volume that results from poor
nutrient exchange (i.e., vascularization) and ischemia-induced
necrosis [15, 16]. To avoid the need for multiple lipectomy
procedures, surgeons elect to remove sufficient amounts of
adipose tissue for multiple procecures and store them frozen
often in simple saline solutions.
However, the proper technique for the freezing of adipose
tissue remains unclear. The majority of studies performed
to date have stressed the need for some type of cryopreservation to maintain adipocyte viability and graft integration
[4, 20–22]. For example, Moscatello et al. [19] concluded
that adipocytes isolated from frozen adipose samples are not
viable unless protected using cryopreservatives such as 10%
DMSO (dimethylsulfoxide), 7.5% PVP (polyvinylpyrrolidone),
or 10% glycerol. Cryopresevation of adipose grafts using DMSO
has also been described by Cui and colleagues [21], with
controlled freezing in this cryoprotectant resulting in decreased
graft resorption and well-preserved adipose architecture after
implantation. However, a few studies have documented viable
adipose grafts frozen without any preservation treatments [17,
18]. The reasons for the conflicting results in the current
literature remain unclear and may be the result of differences
in the rates of graft freezing and thawing, in addition to the
actual harvest site of the adipose grafts. As such, the effects
of freezing adipose grafts still remain unknown. Therefore, this
study attempted to determine if freezing adipose tissue should
be considered a storage option for future applications such as fat
grafting. However, as shown in this study, the long-term storage
of adipose tissue at subzero temperatures without imposing
structural and metabolic damage may not be possible.
Injection of fresh adipose grafts (i.e., no storage) represents
the ideal option, as the grafts would experience the least
level of damage and cellular trauma. Consistent with this
ideal, well-preserved adipose tissue morphology was evident in
fresh adipose grafts implanted for short periods of time (i.e.,
2 weeks). However, a complicating factor with fresh grafts
remains resorption and a loss of structure. Despite numerous
clinical studies presented in the literature [for reviews see 3,
15, 16, 23], a consensus on the sustainability of transplanted
fat grafts and the proper methods to obtain and store these
grafts has yet to be reached. Those animal studies that have
been presented are also conflicting in their results, with good
retention being reported in rodents [24] yet significant resorption
being observed in rabbit models [25, 26]. Consistent with these

Connect Tissue Res Downloaded from informahealthcare.com by Universitat Autonoma Barcelona on 11/04/14
For personal use only.

FREEZING ADIPOSE TISSUE GRAFTS

rabbit studies, injection of fresh grafts performed in this study
was accompanied by loss of total weight over time together
with a definitive loss in adipose tissue structure as the number
of adipocytes decreased and the amount of fibrous tissue within
these fresh grafts increased. While some injected fresh grafts
could be maintained in vivo for up to 8 weeks, most grafts were
completely resorbed into the animal host, resulting in remnants
of fibrotic tissue or no tissue at all. This loss of graft volume and
structure over time would suggest that fresh fat grafts possess a
limited time span in vivo and that this loss may be an unavoidable
result of the harvest itself.
Consistent with this, short-term maintenance of adipose
grafts under conventional tissue culture conditions (up to 2
weeks) was accompanied by steady increases in cell death due
to apoptosis and/or necrosis and a definitive loss of overall tissue
morphology. Observable levels of cell death were apparent after
only 24 hr ex vivo culture, suggesting that removal of the graft
from the patient is sufficient to damage the graft. While the
time period from initial harvest in the operating suite to in
vitro culture in this study was minimized (under 2 hr), it is
apparent that the removal of graft tissue did initiate a series of
events leading to tissue damage and cell death. However, since
augmentation with adipose grafts is normally done within the
operating suite and involves little processing time, damage to
fresh adipose grafts will likely remain negligible if they are
implanted immediately upon harvest. This may not be the case
if long-term storage is required.
While storage methods of fat grafts vary, the generally
accepted method is freezing in a sterile saline solution. However,
saline use may not be warranted as freezing grafts in PBS
appeared to damage the grafts structurally. The observed
deleterious changes to the grafts were apparent as early as
24 hr freezing and did not appear to change significantly when
freezing time was increased to 7 or 14 days, suggesting that even
short-term freezing in PBS (i.e., 24 hr) is sufficient to damage the
tissue. Consistent with this damage, the morphology of frozen
adipose grafts harvested after 2 and 4 weeks implantation was
significantly compromised when compared with fresh grafts,
with frozen grafts showing poor adipocyte structure and the
formation of fibrotic tissue around individual adipocytes. When
graft weights were quantitated after 2 and 4 weeks implantation,
no significant difference could be measured between grafts
frozen in PBS for 1 week and fresh adipose grafts. However,
it is important to note that despite maintaining their weights,
an obvious loss of adipose tissue structure was seen in the
frozen PBS grafts together with significant amounts of fibrotic
tissue. It is likely that this fibrotic tissue contributed to overall
harvested graft. This fibrotic tissue did not appear to have any
reconstructive benefit as these grafts were either eventually
resorbed or were greatly reduced after 8 weeks implantation.
Finally, while retention levels of 40 to 50% could be observed
upon the injection of grafts frozen for short periods in PBS (i.e.,
less than 1 week), increasing the storage time in PBS from
1 week to 8 week appeared to accelerate the loss of adipose

25

structure and increase the formation of fibrotic tissue within
the remaining graft. Increasing freezing time in PBS prior to
implantation to 8 weeks increased graft resorption over time
with the majority of grafts showing resorption into the host
as early as 4 weeks. Therefore, the use of saline solutions for
subzero storage of fat grafts, together with the time span of
storage, must be carefully considered prior to implantation.
The long-term survival of fat grafts—either freshly isolated
or frozen—is likely to be enhanced by improving their
vascularization within the recipient. Studies by Rupnick et al.
(2002) show that adipose mass can be regulated through the
vasculature [27]. Similarly, inhibition of angiogenesis in mice
can dramatically reduce the vascularization and retention of
transplanted fat grafts [9]. Based on these studies it becomes
apparent that by increasing vascularization, one may increase
the success of each graft injection thus reducing the total number
and volume of subsequent implantations. Early studies have
shown that the successful implantation of murine preadipocytes
and their formation into adipose tissue are linked to angiogenesis
[28], possibly through a reciprocal signaling mechanism involving multiple proadipogenic and proangiogenic factors, including
VEGF [29]. VEGF is secreted by endothelial precursors and
endothelial cells, and numerous studies have detailed its role
in angiogenesis in many tissues, both normal and neoplastic.
For example, topical application of recombinant VEGF to
irradiated surgical sites can increase vascularization of the
transplanted tissue [30] and skin flap survival can be increased
through adenoviral mediated VEGF therapy [31]. Studies on
preadipocytes confirm that these cells are excellent sources
of VEGF [28] and implantation of preadipocytes is associated
with increases in both VEGF and several adipogenic markers,
suggesting a relationship between fat development and this
angiogenic factor [32].
Also, circulating VEGF levels positively correlate with fat
mass in humans [27]. Despite this, the role of VEGF in the
angiogenesis of adipose tissue remains unclear, although the
association between VEGF and adipose tissue retention is
strengthening. For example, a study showing that the addition
of human endothelial precursors to fat grafts can improve their
survival in nude mice has recently been presented [33]. Similar
to this, work by several groups now suggests that treatment of
transplanted fat pads with VEGF can significantly increase graft
weight, volume, and microvessel density [34, 35]. Therefore it
is possible that VEGF secretion within a transplanted fat graft
may be critical to its vascularity and survival in vivo.
Based on these works, we examined the VEGF synthetic
ability of fresh and frozen adipose grafts. Freshly isolated grafts
maintained ex vivo for up to 2 weeks maintained their VEGF
synthetic ability, secreting significant amounts of this growth
factor into the culture medium. However, consistent with the
loss of morphology upon freezing, grafts stored for either 7 or
14 days in PBS completely lost their VEGF synthetic activity.
However, VEGF secretion levels could be preserved—albeit at
lower levels when compared with fresh grafts—when grafts

Connect Tissue Res Downloaded from informahealthcare.com by Universitat Autonoma Barcelona on 11/04/14
For personal use only.

26

N. GREWAL ET AL.

were frozen in cryopreservation media designed to maintain
cellular viability. In the cryopreservation of cells, many media
conditions have been developed to prevent the death of the
cell once reintroduced into culture. Low temperature storage of
bacteria is accomplished in solutions containing high levels of
glycerol. Media solutions containing low levels of DMSO are
used to prevent the formation of damaging ice crystals within the
cell and their lysis upon thawing. Based on this theory, several
cryopreservation media conditions were created in this study to
maintain the viability of adipose grafts upon reimplantation.
While the majority of these solutions, including those with
DMSO, were unable to preserve graft VEGF synthesis, a
cryostorage medium containing 30% sucrose in PBS (Cryo#4)
was the only medium condition that allowed VEGF synthesis
once the graft was reintroduced to culture. Embedding tissues in
30% sucrose is used prior to cryosectioning to protect the tissue
structure by extracting excess water. It is possible that a similar
effect was achieved in the adipose grafts of this study.
Accompanying this ability to preserve VEGF synthesis
was an enhanced effect upon graft morphology and volume.
Specifically, grafts preserved for up to 1 week in Cryo#4 showed
a higher level of retention within the host when compared with
control grafts frozen in PBS, with increased graft volumes being
measured in these Cryo#4 grafts after both 2 and 4 weeks
implantation. When grafts were stored in Cryo#4 for 8 weeks,
the retention benefit became clearer with only 1 of the 5 injected
grafts resorbing into the host by the 4 weeks harvest point, as
opposed to 4 of 5 injected PBS grafts resorbing within the
host. However, the increased retention measured in these 8
week Cryo#4 grafts did not translate into acceptable adipose
morphology.
In fact, the time-dependent effects on morphology seen for
storage in PBS also appeared to apply to storage in Cryo#4.
When stored in Cryo#4 for up to 1 week, the resulting graft
tissue appeared to be superior to PBS controls, with wellformed adipocytes and decreased fibrotic tissue within the graft.
Consistent with its ability to retain some level of VEGF synthesis
and secretion, several of these Cryo#4 grafts contained small
blood vessels, suggesting that this cryopreservation medium
does not, at the very least, prevent vascularization. However,
the apparent benefit to adipose morphology was lost completely
if grafts were cryopreserved in Cryo#4 for 8 weeks with large,
malformed adipocytes being seen with significant amounts of
fibrous tissue within the remaining graft. Therefore, the use of
cryopreservation media like Cryo#4 may only prove beneficial
for short-term storage of adipose tissue.
The synthesis and secretion of VEGF by animal and human
“pre-adipocytes” has been extensively documented [28, 36].
However, with the recent description of a population of stem
cells within adipose tissue depots—the ASC (published previously as PLA cells [5, 6])—these results could be reinterpreted
to indicate that the ASC population is capable of secreting
large amounts of VEGF. Moreover, several groups have recently
quantified the amount of VEGF specifically made by the ASC

population [10, 37, 38]. In agreement with these works, the
isolated ASC populations examined in this study were also
found to secrete significant amounts of VEGF. Based on these
findings, it becomes logical to assume that a simple combination
of ASCs with frozen adipose grafts could improve their retention
by increasing VEGF-induced vascularization. In support of this,
numerous studies document the use of preadipocytes or ASCs
in combination with various growth factors or scaffolds in the
production of adipose tissue in vitro and fat pads in vivo [39–41].
Recent work by Ramon and colleagues [42] also suggests
that simple concentration of fat cells and their combination
with adipose tissue can improve the “take” of injected grafts.
Consistent with these studies, preliminary results presented in
this study suggest that the in vivo morphology of fresh adipose
grafts and those frozen in PBS can be improved through their
combination with ASCs. However, it is important to note that
VEGF synthesis by the ASC can also be dramatically affected
by storage condition. Specifically, cryopreservation of ASC
populations did result in declining levels of VEGF. Therefore,
improved retention of adipose grafts would require the use of
“fresh” (i.e., nonpreserved) ASC populations.

CONCLUSION
This study finds that the conditions used to freeze adipose
tissue grafts and/or the use of ASCs as a component of an
implanted adipose tissue graft may dramatically affect the
structure of the graft and possibly its successful integration
into the host. Currently, long-term storage of adipose grafts,
used for repeated injections, involves the freezing of the graft
in simple saline solutions if anything at all. Unfortunately, little
thought has been given to the effect of storage condition on
the morphology and function of the adipose tissue itself. As
a result, this study attempts to determine at a preliminary level
whether freezing of the adipose tissue graft should be considered
as a storage option. The results presented in this study show
significant changes to the morphology of the adipose graft upon
freezing in PBS.
Moreover, significant reductions in the metabolic activity
of the graft—as assessed by measuring VEGF synthetic
ability—can also be measured. This reduction in VEGF
synthesis is likely to have significant effects on the successful
integration of the graft as VEGF synthesis is thought to improve
graft vascularization. The use of a simple cryopreservation
medium containing 30% sucrose was capable of retaining not
only VEGF synthetic ability upon freezing but was capable
of improving the retention and structure of frozen grafts
in vivo. However, this benefit was directly linked to storage
time with longer storage times in this cryopreservation medium
showing no clear advantage over PBS. Finally, improvements
in both fresh and frozen graft retention were observed upon the
combination of these grafts with a cellular source of VEGF—in
this case, adipose graft-derived ASCs.

Connect Tissue Res Downloaded from informahealthcare.com by Universitat Autonoma Barcelona on 11/04/14
For personal use only.

FREEZING ADIPOSE TISSUE GRAFTS

In conclusion, it appears that methods to preserve or augment
VEGF levels within an adipose graft either through their
combination with ASCs or storage in defined cryopreservation
media represent a viable approach to improving graft retention
clinically. However, it is important to note that long-term
storage of grafts, in either PBS or in a defined cryopreservation
medium, was not conducive to the retention of well-structured
adipose tissue grafts within an animal host. Therefore, additional
studies of how adipose grafts can be stored without damaging
their structure and potential retentive ability must be examined
carefully to determine how such grafts can be used most
efficiently in future cosmetic and soft-tissue reconstructive
applications.
ACKNOWLEDGMENT
A portion of the data presented in this study was supported
by the Bernard Sarnat Fellowship for Craniofacial Research.
Declaration of Interest: The authors report no conflicts of
interest. The authors alone are responsible for the content and
writing of the paper.
REFERENCES
1. Dasiou-Plakida, D. (2003). Fat injections for facial rejuvenation: 17 years
experience in 1720 patients. J. Cosmet. Dermatol., 2, 119–125.
2. Horl, H.W., Feller, A.M., and Biemer, E. (1991). Technique for liposuction
fat reimplantation and long-term volume evaluation by magnetic resonance
imaging. Ann. Plast. Surg., 26, 248–258.
3. Sommer, B., and Sattler, G. (2000). Current concepts of fat graft survival:
Histology of aspirated adipose tissue and review of the literature. Dermatol.
Surg., 26, 1159–1166.
4. Erdim, M., Tezel, E., Numanoglu, A., and Sav, A. (in press). The effects
of the size of liposuction cannula on adipocyte survival and the optimum
temperature for fat graft storage: An experimental study. J. Plast. Reconstr.
Aesthet. Surg.
5. Zuk, P.A., Zhu, M., Mizuno, H., Huang, J.I., Futrell, W.J, Katz, A.J.,
Benhaim, P., Lorenz, H. P., and Hedrick, M. H. (2001). Multi-lineage cells
from human adipose tissue: Implications for cell-based therapies. Tissue
Eng., 7, 211–226.
6. Zuk, P.A., Zhu M., Ashjian, P., De Ugarte, D.A., Huang, J.I., Mizuno, H.,
Alfonso, Z.C., Fraser, J.K., Benhaim, P., and Hedrick, M.H. (2002). Human
adipose tissue is a source of multipotent stem cells. Mol. Biol. Cell., 13,
4279–4295.
7. Hopkins, S.P., Bulgrin, J.P., Sims, R.L., Bowman, B., Donovan, D.L., and
Schmidt, S.P. (1998). Controlled delivery of vascular endothelial growth
factor promotes neovascularization and maintains limb function in a rabbit
model of ischemia. J. Vasc. Surg., 27, 886–894; discussion 895.
8. Rocha, F.G., Sundback, C.A., Krebs, N.J., Leach, J.K., Mooney, D.J.,
Ashley, S.W., Vacanti, J.P., and Whang, E.E. (2008). The effect of
sustained delivery of vascular endothelial growth factor on angiogenesis
in tissue-engineered intestine. Biomaterials, 29, 2884–2890.
9. Yamaguchi, M., Matsumoto, F., Bujo, H., Shibasaki, M., Takahashi, K.,
Yoshimoto, S., Ichinose, M., and Saito, Y. (2005). Revascularization
determines volume retention and gene expression by fat grafts in mice.
Exp. Biol. Med. (Maywood)., 230, 742–748.
10. Rehman, J., Traktuev, D., Li, J., Merfeld-Clauss, S., Temm-Grove, C.J.,
Bovenkerk, J.E., Pell, C.L., Johnstone, B.H., Considine, R.V., and March,
K.L. (2004). Secretion of angiogenic and antiapoptotic factors by human
adipose stromal cells. Circulation, 109, 1292–1298.

27

11. Mutin, M., George, F., Lesaule, G., and Sampol, J. (1996). Reevaluation
of trypsin-EDTA for endothelial cell detachment before flow cytometry
analysis. Endothelium, 4, 289–295.
12. Coleman, S.R. (1995). Long-term survival of fat transplants: Controlled
demonstrations. Aesthet. Plast. Surg., 19, 421–425.
13. Coleman, S.R. (2006). Structural fat grafting: More than a permanent filler.
Plast. Reconstr. Surg., 118, 108S–120S.
14. Coleman, W.P., 3rd.3rd. (1999). Fat transplantation. Dermatol. Clin., 17,
891–898, viii.
15. Kaufman, M.R., Bradley, J.P., Dickinson, B., Heller, J.B., Wasson, K.,
O’Hara, C., Huang, C., Gabbay, J., Ghadjar, K., and Miller, T.A. (2007).
Autologous fat transfer national consensus survey: Trends in techniques for
harvest, preparation, and application, and perception of short- and long-term
results. Plast. Reconstr. Surg., 119, 323–331.
16. Kaufman, M.R., Miller, T.A., Huang, C., Roostaien, J., Wasson, K.L.,
Ashley, R.K., and Bradley, J.P. (2007). Autologous fat transfer for facial
recontouring: Is there science behind the art? Plast. Reconstr. Surg., 119,
2287–2296.
17. MacRae, J.W., Tholpady, S.S., Ogle, R.C., and Morgan, R.F. (2004). Ex
vivo fat graft preservation: Effects and implications of cryopreservation.
Ann. Plast. Surg., 52, 281–283.
18. Shoshani, O., Ullmann, Y., Shupak, A., Ramon, Y., Gilhar, A., Kehat, I., and
Peled, I.J. (2001). The role of frozen storage in preserving adipose tissue
obtained by suction-assisted lipectomy for repeated fat injection procedures.
Dermatol. Surg., 27, 645–647.
19. Moscatello, D.K., Dougherty, M., Narins, R.S., and Lawrence, N. (2005).
Cryopreservation of human fat for soft tissue augmentation: Viability
requires use of cryoprotectant and controlled freezing and storage.
Dermatol. Surg., 31, 1506–1510.
20. Atik, B., Ozturk, G., Erdogan, E., and Tan, O. (2006). Comparison of
techniques for long-term storage of fat grafts: An experimental study. Plast.
Reconstr. Surg., 118, 1533–1537.
21. Cui, X.D., Gao, D.Y., Fink, B.F., Vasconez, H.C., and Pu, L.L. (2007).
Cryopreservation of human adipose tissues. Cryobiology, 55, 269–
278.
22. Lidagoster, M.I., Cinelli, P.B., Levee, E.M., and Sian, C.S. (2000).
Comparison of autologous fat transfer in fresh, refrigerated, and
frozen specimens: An animal model. Ann. Plast. Surg., 44, 512–
515.
23. Locke, M.B., and de Chalain, T.M. (2008). Current practice in autologous
fat transplantation: Suggested clinical guidelines based on a review of recent
literature. Ann. Plast. Surg., 60, 98–102.
24. Smith, P., Adams, W.P., Jr., Lipschitz, A.H., Chau, B., Sorokin, E., Rohrich,
R.J., and Brown, S.A. (2006). Autologous human fat grafting: Effect of
harvesting and preparation techniques on adipocyte graft survival. Plast.
Reconstr. Surg., 117, 1836–1844.
25. Baran, C.N., Celebioglu, S., Sensoz, O., Ulusoy, G., Civelek, B., and Ortak,
T. (2002). The behavior of fat grafts in recipient areas with enhanced
vascularity. Plast. Reconstr. Surg., 109, 1646–1651; 1652.
26. Kononas, T.C., Bucky, L.P., Hurley, C., and May, J.W., Jr. (1993). The
fate of suctioned and surgically removed fat after reimplantation for softtissue augmentation: A volumetric and histologic study in the rabbit. Plast.
Reconstr. Surg., 91, 763–768.
27. Rupnick, M.A., Panigrahy, D., Zhang, C.Y., Dallabrida, S.M., Lowell,
B.B., Langer, R., and Folkman, M.J. (2002). Adipose tissue mass can be
regulated through the vasculature. Proc. Natl. Acad. Sci. USA, 99, 10730–
10735.
28. Hausman, G.J., and Richardson, R.L. (2004). Adipose tissue angiogenesis.
J. Anim. Sci., 82, 925–934.
29. Fukumura, D., Ushiyama, A., Duda, D.G., Xu, L., Tam, J., Krishna, V.,
Chatterjee, K., Garkavtsev, I., and Jain, R.K. (2003). Paracrine regulation
of angiogenesis and adipocyte differentiation during in vivo adipogenesis.
Circ. Res., 93, e88–e97.
30. Schultze-Mosgau, S., Wehrhan, F., Rodel, F., Amann, K., Radespiel-Troger,
M., and Grabenbauer, G.G. (2003). Improved free vascular graft survival

28

31.

32.

33.

Connect Tissue Res Downloaded from informahealthcare.com by Universitat Autonoma Barcelona on 11/04/14
For personal use only.

34.

35.

36.

N. GREWAL ET AL.
in an irradiated surgical site following topical application of rVEGF. Int. J.
Radiat. Oncol. Biol. Phys., 57, 803–812.
Lubiatowski, P., Goldman, C.K., Gurunluoglu, R., Carnevale, K., and
Siemionow, M. (2002). Enhancement of epigastric skin flap survival by
adenovirus-mediated VEGF gene therapy. Plast. Reconstr. Surg., 109,
1986–1993.
Neels, J.G., Thinnes, T., and Loskutoff, D.J. (2004). Angiogenesis in
an in vivo model of adipose tissue development. FASEB J., 18, 983–
985.
Yi, C., Pan, Y., Zhen, Y., Zhang, L., Zhang, X., Shu, M., Han, Y., and Guo,
S. (2006). Enhancement of viability of fat grafts in nude mice by endothelial
progenitor cells. Dermatol. Surg., 32, 1437–1443.
Lei, M., Liu, S.Q., Peng, H., and Liu, Y.L. (2008). Effect of rhVEGF
gene transfection on survival of grafts after autologous free granular fat
transplantation in rats. Chin. J. Traumatol., 11, 49–53.
Yi, C.G., Xia, W., Zhang, L.X., Zhen, Y., Shu, M.G., Han, Y., and Guo,
S.Z. (2007). VEGF gene therapy for the survival of transplanted fat tissue
in nude mice. J. Plast. Reconstr. Aesthet. Surg., 60, 272–278.
Cao, Y. (2007). Angiogenesis modulates adipogenesis and obesity. J. Clin.
Invest., 117, 2362–2368.

37. Li, B., Zeng, Q., Wang, H., and Mao, X. (2006). Effect of cytokines secreted
by human adipose stromal cells on endothelial cells. J. Huazhong Univ. Sci.
Technol. Med. Sci., 26, 396–398.
38. Nakagami, H., Morishita, R., Maeda, K., Kikuchi, Y., Ogihara, T., and
Kaneda, Y. (2006). Adipose tissue-derived stromal cells as a novel option
for regenerative cell therapy. J. Atheroscler. Thromb., 13, 77–81.
39. Eppley, B.L., Snyders, R.V., Jr., Winkelmann, T., and Delfino, J.J.
(1992). Autologous facial fat transplantation: Improved graft maintenance
by microbead bioactivation. J. Oral. Maxillofac. Surg., 50, 477–482;
discussion 482–483.
40. Rubin, J.P., Bennett, J.M., Doctor, J.S., Tebbets, B.M., and Marra, K.G.
(2007). Collagenous microbeads as a scaffold for tissue engineering with
adipose-derived stem cells. Plast. Reconstr. Surg., 120, 414–424.
41. Torio-Padron, N., Baerlecken, N., Momeni, A., Stark, G.B., and Borges, J.
(2007). Engineering of adipose tissue by injection of human preadipocytes
in fibrin. Aesthet. Plast. Surg., 31, 285–293.
42. Ramon, Y., Shoshani, O., Peled, I.J., Gilhar, A., Carmi, N., Fodor, L., Risin,
Y., and Ullmann, Y. (2005). Enhancing the take of injected adipose tissue
by a simple method for concentrating fat cells. Plast. Reconstr. Surg., 115,
197–203.