Can Arthroscopically Harvested Synovial Stem Cells Be
Preferentially Sorted Using Stage-Specific Embryonic Antigen
4 Antibody for Cartilage, Bone, and Adipose Regeneration?
Jingting Li, M.S., Douglas D. Campbell, B.S., George K. Bal, M.D., and Ming Pei, M.D., Ph.D.
Purpose: The aim of this study was to investigate the relation between stage-specific embryonic antigen 4 (SSEA4)
expression and synovium-derived stem cell (SDSC) lineage differentiation. Methods: Human SDSCs were collected
during arthroscopic surgery from 4 young patients with anterior cruciate ligament injuries. Passage 2 SDSCs were sorted
by fluorescence-activated cell sorting using phycoerythrin-conjugated monoclonal antibody against SSEA4 into 3 groups:
SSEA4(þ) cells, SSEA4() cells, and unsorted control cells. After 1 more passage, expanded cells from each group were
evaluated for SSEA4 expression by use of flow cytometry as well as multilineage differentiation capacities, including
chondrogenesis, adipogenesis, and osteogenesis, using biochemical analysis, histologic analysis, immunostaining, and realtime polymerase chain reaction. Results: After cell sorting, 1 more passage expansion decreased SSEA4(þ) cells from
99.8% to 79.2% and increased SSEA4() cells from 4.4% to 53.3% compared with 70.3% in the unsorted cell population. SSEA4() SDSCs with a lower cell proliferation exhibited higher chondrogenic potential (in terms of the ratio of
glycosaminoglycan to DNA [P < .001] and COL2A1 [type II collagen] messenger RNA [mRNA] [P < .001]) and adipogenic
potential (in terms of oil red O staining and quantitative assay [P ¼ .007], LPL [lipoprotein lipase] mRNA [P ¼ .005], and
CEBP [CCAAT/enhancer-binding protein alpha] mRNA [P ¼ .010]). In contrast, SSEA4(þ) SDSCs retained cell expansion
and enhanced osteogenic capacity, as evidenced by intense calcium deposition stained by alizarin red S and a significantly
elevated expression of OPN (osteopontin) mRNA (P ¼ .007). Conclusions: In this study, for the first time, we showed the
benefit of using the surface marker SSEA4 in SDSCs to preferentially sort a mixed population of cells. SSEA4(þ) SDSCs
indicated a strong potential for osteogenesis rather than chondrogenesis and adipogenesis. Clinical Relevance: SDSCbased mesenchymal tissue regeneration can be easily achieved by arthroscopic harvesting followed by quick cell sorting.
M
esenchymal stem cells (MSCs) are a unique population of adult stem cells that have the ability
to differentiate into multiple lineages, such as bone, fat,
and cartilage. Recent clinical trials using MSC cellular
From the Stem Cell and Tissue Engineering Laboratory, Department of
Orthopaedics (J.L., D.D.C., G.K.B., M.P.), Department of Exercise Physiology
(J.L., M.P.), and Department of Mechanical & Aerospace Engineering (M.P.),
West Virginia University, Morgantown, West Virginia, U.S.A.
This project was partially supported by research grants from the West
Virginia University Senate Research Grant Award (R-12-010), the AO
Foundation (S-12-19P), the Musculoskeletal Transplant Foundation, and the
National Institutes of Health R03 Program (1 R03 AR062763-01A1 and 5
R03 DE021433-02). The authors report that they have no conflicts of interest
in the authorship and publication of this article.
Received July 3, 2013; accepted December 16, 2013.
Address correspondence to Ming Pei, M.D., Ph.D., Stem Cell and Tissue
Engineering Laboratory, Department of Orthopaedics, West Virginia University, PO Box 9196, One Medical Center Drive, Morgantown, WV 265069196, U.S.A. E-mail: mpei@hsc.wvu.edu
Ó 2014 by the Arthroscopy Association of North America
0749-8063/13457/$36.00
http://dx.doi.org/10.1016/j.arthro.2013.12.009
352
therapies have shown considerable promise and a broad
therapeutic potential in regenerative treatment of many
common diseases.1 MSCs of synovial origin, referred to
as synovium-derived stem cells (SDSCs), have been
found to display an enhanced capability for chondrogenesis when compared with MSCs from other origins,
supporting the notion of a tissue-specific stem cell for
articular cartilage regeneration.2-6 SDSCs may be easily
acquired under arthroscopic observation with a small
punch biopsy providing sufficient cells for clinical use.7
Although MSCs may be harvested from a wide variety
of sources, such as bone marrow, adipose, periosteum,
muscle, perichondrium, and synovium, it has been challenging to isolate and purify them because of limited
selectivity of available markers.6,8 Specific surface markers
for identifying MSCs with high self-renewal and desired
lineage differentiation have potential for clinical applications. A recent article reviewed potential markers that may
serve to indicate enhanced chondrogenic differentiation
capabilities of SDSCs.9 One such surface marker that has
gathered interest is the glycolipid stage-specific embryonic
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 30, No 3 (March), 2014: pp 352-361
SSEA4-SORTED SDSC DIFFERENTIATION
antigen 4 (SSEA4). This antigen, previously used as a
marker for embryonic stem cell identification, has recently
been discovered to also serve as a marker for MSCs.10
Recent studies indicated that human synovium-derived
stem cells (hSDSCs) and human bone marrow stromal
cells (hBMSCs) expanded on decellularized stem cell
matrix (DSCM) exhibited an upregulation of SSEA4
expression with concomitantly enhanced self-renewal
and chondrogenic potential.11-13 These observations
motivated us to investigate the relation between the
surface marker SSEA4 expression and hSDSC lineage
differentiation. The aim of this study was to investigate
the relation between SSEA4 expression and SDSC lineage
differentiation. We hypothesized that SSEA4 as a surface
marker could preferentially sort hSDSC populations with
lineage differentiation potentials for cartilage, bone, and
adipose tissue regeneration.
Methods
Preparation of hSDSCs
Random biopsy specimens from the intimal layer of synovial tissue were obtained aseptically during arthroscopic
surgery from the knees of 4 young patients (3 men and 1
woman; mean age, 23 years) with anterior cruciate ligament injuries. The inclusion criteria were specified as patients aged between 18 and 30 years and undergoing knee
arthroscopy. The exclusion criteria included patients with
any history of inflammatory arthropathy, previous knee
surgery, immunodeficiency, connective tissue disorders or
septic arthritis, and chronic infection (bacterial, fungal, or
viral). This project was approved by the institutional review
board (protocol H-23564), and consent was obtained from
each patient before the deidentified sample was collected.
The method for digesting synovial tissue was described
previously.6 In brief, after temporary storage in culture
medium at 4 C, the synovial tissue was finely minced and
digested at 37 C for 30 minutes in phosphate-buffered
saline solution (PBS) containing 0.1% trypsin and then
for 2 hours in a 0.1% solution of collagenase P in alpha
modified eagle medium/10% fetal bovine serum (FBS).
The cell suspension was passed through a 70-mm nylon
filter, and the cells were collected from the filtrate by
centrifugation. Cells were cultured in growth medium
(alpha modified eagle medium/10% FBS, 100-U/mL
penicillin, 100-mg/mL streptomycin). Nonadherent cells
were removed by a PBS wash on days 2 and 4.
SSEA4-Based Cell Sorting
Passage 2 hSDSCs were labeled with phycoerythrinconjugated monoclonal antibody against SSEA4 (BioLegend, San Diego, CA) in PBS without Ca2þ and
Mg2þ supplemented with 2.5-mmol/L EDTA, 1% FBS,
and 25-mmol/L N-(2-hydroxyethyl)-piperazine-N0 -2ethanesulfonic acid (HEPES), followed by cell sorting
with FACSAria (BD Biosciences, San Jose, CA) into 3
353
groups: SSEA4(þ) cells, SSEA4() cells, and unsorted
cells serving as a control. Phycoerythrin was excited at
488 nm by an argon ion laser, and fluorescence was
detected by use of a 575/26 filter. Dead cells were
excluded by gating on forward and side scatter. Data were
analyzed by the FCS Express 3 software package (De
Novo Software, Los Angeles, CA). All sorted and unsorted
cells were expanded for 1 more passage and were then
evaluated for SSEA4 expression by use of flow cytometry
as well as multilineage differentiation capacities, including
chondrogenesis, adipogenesis, and osteogenesis, using
biochemical analysis, histologic analysis, immunostaining,
and real-time polymerase chain reaction (PCR).
Multilineage Induction and Differentiation Analysis
Chondrogenic Induction. As described in our previous
study,12 3 105 hSDSCs from each group were centrifuged
at 500g for 5 minutes in a 15-mL polypropylene tube to
form a pellet. After overnight incubation, the pellets were
transferred to a serum-free chondrogenic medium
consisting of high-glucose Dulbecco modified eagle
medium, 40-mg/mL proline, 107-mol/L dexamethasone,
100-U/mL penicillin, 100-mg/mL streptomycin, 0.1mmol/L ascorbic acid-2-phosphate, and 1 ITS Premix
(6.25-mg/mL insulin, 6.25-mg/mL transferrin, 6.25-mg/mL
selenous acid, 5.35-mg/mL linoleic acid, and 1.25-mg/mL
bovine serum albumin; BD Biosciences) with the
supplementation of 10-ng/mL transforming growth
factor b3 (PeproTech, Rocky Hill, NJ) in a 37 C, 5% CO2/
5% O2 incubator for up to 27 days. On days 0 and 27,
pellets from each group were collected for evaluation
of chondrogenic differentiation by histochemistry,
immunohistochemistry, biochemistry, and quantitative
real-time PCR.
Histochemistry and Immunohistochemistry. The pellets
(n ¼ 2) were fixed in 4% paraformaldehyde at 4 C
overnight, followed by dehydrating in a gradient
ethanol series, clearing with xylene, and embedding in
paraffin blocks. Five-micrometer sections underwent
histochemical staining with alcian blue (Sigma-Aldrich,
St. Louis, MO) (counterstained with fast red) for sulfated
glycosaminoglycans (GAGs). For immunostaining, the
sections were immunolabeled with primary antibodies
against type I collagen (Abcam, Cambridge, MA) and
type II collagen (II-II6B3; Developmental Studies
Hybridoma Bank, Iowa City, IA), followed by the
secondary antibody of biotinylated horse anti-mouse
immunoglobulin G (Vector Laboratories, Burlingame,
CA). Immunoactivity was detected by use of Vectastain
ABC reagent (Vector Laboratories) with 3,30 diaminobenzidine as a substrate.
Biochemical Analysis for DNA and GAG Content. The pellets (n ¼ 4) were digested at 60 C for 6 hours with 125mg/mL papain in PBE buffer (100-mmol/L sodium
phosphate buffer and 10-mmol/L EDTA, pH 6.5)
354
J. LI ET AL.
containing 10-mmol/L cysteine. To quantify cell density,
the amount of DNA in the papain digestion was
measured with the Quant-iT PicoGreen dsDNA Assay
Kit (Invitrogen, Carlsbad, CA) with a CytoFluor Series
4000 (Applied Biosystems, Foster City, CA). GAG was
measured by use of dimethylmethylene blue dye and a
Spectronic BioMate 3 Spectrophotometer (Thermo
Scientific, Milford, MA) with bovine chondroitin sulfate
(Sigma-Aldrich) as a standard.
Adipogenic Induction. Expanded hSDSCs were replated
at 10,000 cells/cm2 in T25 flasks. Once cells reached
100% confluence, the culture medium was switched to
adipogenic induction medium consisting of growth
medium supplemented with 1-mmol/L dexamethasone,
0.5-mmol/L isobutyl-1-methyxanthine, 200-mmol/L
indomethacin, 10-mg/mL insulin, and 1-nmol/L 3,30 50 triiodo-L-thyronine for an additional 21 days. Oil red O
(ORO) staining and quantitative assays were conducted
as described previously.12 In brief, adipogenically
induced hSDSCs (n ¼ 3) were fixed in 4%
paraformaldehyde for 60 minutes and stained with
0.3% ORO solution (Sigma-Aldrich) for 30 minutes.
After rinsing in distilled water, cells were photographed
with an AmScope Microscope Digital Camera (model
MD1900; iScope, Irvine, CA). ORO was extracted from
cells using 100% isopropanol, and the absorbance was
determined at 510 nm. For a blank control, we used
100% isopropanol. The ORO optical density value was
normalized by total DNA content. TaqMan real-time
PCR was used for quantification expression of
adipogenic marker genes.
Osteogenic Induction. Expanded hSDSCs were replated at
8,000 cells/cm2 in T25 flasks. Once cells reached 90%
confluence, the culture medium was switched to osteogenic induction medium consisting of growth medium
supplemented with 0.01-mmol/L dexamethasone, 10mmol/L b-glycerol phosphate, 50-mmol/L ascorbate-2phosphate, and 0.01-mmol/L 1,25-dihydroxyvitamin D3
for an additional 21 days. Osteogenic differentiation was
assessed by use of alizarin red S (ARS) staining
for calcium deposition and TaqMan real-time PCR
for quantification expression of osteogenic marker
genes.12 In brief, osteogenically induced hSDSCs
(n ¼ 3) were fixed with 70% ice-cold ethanol for
1 hour and then incubated in 40-mmol/L ARS at
pH 4.2 for 20 minutes at room temperature with
agitation on an orbital shaker (60 rpm). After 2 rinses
with deionized water, matrix mineral-bound staining
was photographed under a Nikon TE300 Inverted
Phase Contrast Microscope (Nikon, Tokyo, Japan).
Accumulated calcium was extracted using 0.5 mL of 0.5
N hydrochloric acid and quantified according to the
manufacturer’s instructions in a QuantiChrom Calcium
Assay Kit (BioAssay Systems, Hayward, CA). Total
calcium was calculated from standard solutions prepared
in parallel and normalized to the total protein content.
The values of blank controls were subtracted from the
corresponding samples.
TaqMan Quantitative PCR. As described previously,12
total RNA was extracted from the samples (n ¼ 4)
using a ribonuclease-free pestle in TRIzol (Invitrogen).
Two micrograms of messenger RNA (mRNA) was used
for reverse transcriptase with the High-Capacity cDNA
Archive Kit (Applied Biosystems) at 37 C for 120
minutes. Chondrogenic marker genes (COL1A1 [assay
ID Hs00164004_m1] and COL2A1 [assay ID
Hs00156568_m1]); adipogenic marker genes (LPL
[lipoprotein lipase] [assay ID Hs00173425_m1] and
CEBP [CCAAT/enhancer-binding protein alpha] [assay
ID Hs00269972_m1]); and an osteogenic marker gene
(OPN [osteopontin] [assay ID Hs01587814_g1]) were
purchased from Applied Biosystems. Eukaryotic 18S
ribosomal RNA (assay ID HS99999901_s1) was carried
out as the endogenous control gene. Real-time PCR
was performed with the iCycler iQ Multi-Color RealTime PCR Detection System (Biorad, Hercules, CA),
and the data were calculated with computer software
(PerkinElmer, Wellesley, MA). Relative transcript levels
were calculated as c ¼ 2DDCt, in which DDCt ¼ DE DC, DE ¼ Ctexp Ct18s, and DC ¼ Ctct1 Ct18s.
Power Analysis
Estimation of the sample size and power for this study
was performed with JMP software (SAS Institute, Cary,
NC). Previous results obtained in our laboratory indicated differences in both GAG amount (obtained by
biochemical analysis) and chondrogenic marker genes
(COL1A1, COL2A1, LPL, CEBP, and OPN, evaluated by
TaqMan quantitative PCR) of approximately 5 times
their within-group standard deviations (signal-to-noise
ratio). We had 4 samples per group, which provided
82% power.
Statistical Analysis
Numerical data are presented as the mean and standard error of the mean. The Mann-Whitney U test was
used for pair-wise comparison in biochemistry, ORO
assay, and real-time PCR data analysis. All statistical
analyses were performed with SPSS statistical software
(version 13.0; SPSS, Chicago, IL). P < .05 was considered statistically significant.
Results
Sorting Mixed Population of SDSCs Into Distinct
Populations With SSEA4
After cell sorting by flow cytometry, as shown in Fig 1,
the percentage of SSEA4-positive cells was 99.8% in
the sorted SSEA4(þ) group and 4.4% in the sorted
SSEA4() group, distinct from 73.4% in the unsorted
cells. After 1 more passage expansion from the sorted
SSEA4-SORTED SDSC DIFFERENTIATION
355
Fig 1. Flowchart of research design.
Passage 2 hSDSCs labeled with
phycoerythrin-conjugated SSEA4
were sorted by fluorescenceactivated cell sorting. The SSEA4
expression pattern showed 3 populations: SSEA4()/SSEA4(þ)1,
SSEA4(þ)2,
and
SSEA4(þ)3.
SSEA4() and SSEA4(þ)2 hSDSCs
were chosen to have cells with a
more uniform SSEA4 expression.
SSEA4 expression levels in both the
positive (99.8%) and negative
(4.4%) groups were confirmed after
sorting. Both sorted groups were
expanded for another passage with
unsorted cells serving as a control;
the SSEA4 percentages in the unsorted cells, SSEA4(þ) cells, and
SSEA4() cells were 70.3%, 79.2%,
and 53.3%, respectively. Expanded
cells from all 3 groups were induced
in chondrogenic, adipogenic, or
osteogenic medium before analysis.
(FL2-H: fluorescence intensityheight;
PE-A,
phycoerythrinabsorbance.)
cells, we found that, compared with 70.3% in the unsorted cells, SSEA4-positive cells decreased to 79.2% in
the SSEA4(þ) group and increased to 53.3% in the
SSEA4() group. Compared with a slight increase in the
SSEA4(þ) group, the cell number dramatically decreased
to 57.0% in the SSEA4() group, suggesting that SSEA4
is a positive indicator for stem cell proliferation.
SSEA4(D) Expression Does Not Favor hSDSC
Chondrogenesis
After 27 days of chondrogenic induction, cell pellets
from the SSEA4(þ), SSEA4(), and unsorted population groups were analyzed and compared for their
ability to grow cartilage. Morphologic analysis included
staining for sulfated GAG with alcian blue (Fig 2A), as
well as immunohistochemical staining for type II
collagen (Fig 2B) and type I collagen (Fig 2C). Size
differences were found to exist between the pellets,
with the SSEA4() population producing a visibly
larger pellet than the SSEA4(þ) population. This
finding suggests that enhanced chondrogenesis
occurred in the SSEA4() population of cells because of
the larger pellet formed. Biochemical analysis (Fig 2D)
allowing for the quantification of DNA and GAG to
assess the ratio of GAG to DNA, otherwise known as the
chondrogenic index, showed the amount of DNA present as well as the significantly elevated chondrogenic
index observed in the SSEA4() population compared
with the SSEA4(þ) population (1.493 0.086 ug/ug
v 1.173 0.015 ug/ug, n ¼ 4, P < .001). Real-time PCR
356
J. LI ET AL.
Fig 2. Chondrogenic induction and evaluation of SSEA4-sorted hSDSCs. After cell sorting, unsorted (control) and sorted
[SSEA4(þ) and SSEA4()] cells were expanded for 1 more passage. The expanded cells were cultured in a pellet system supplemented with a serum-free chondrogenic induction medium for 27 days. (A) Alcian blue (AB) staining was used for sulfated
GAG, (B) immunohistochemistry (IHC) was for type II collagen (Col II), and (C) type I collagen (Col I). (Original magnification
100.) Quantitative data included (D) biochemical analysis (DNA and GAG amount as well as ratio of GAG to DNA [i.e.,
chondrogenic index]), (E) real-time PCR analysis for COL2A1, and (F) COL1A1 mRNAs. Data are shown as mean standard
deviation (n ¼ 4). The asterisks indicate significant differences (P < .05).
data were consistent with the previously mentioned
histologic and biochemical data. COL2A1 mRNA was
found to have significantly elevated levels in the
SSEA4() population relative to the SSEA4(þ) population (4.280 0.124 v 1.001 0.048, n ¼ 4, P < .001)
(Fig 2E). Alternatively, the COL1A1 mRNA level was
found to be significantly elevated in the SSEA4(þ)
population compared with the SSEA4() population
(1.000 0.024 v 0.554 0.008, n ¼ 4, P < .001)
(Fig 2F).
SSEA4(D) Expression Does Not Favor hSDSC
Adipogenesis
After 21 days of adipogenic induction, the quantitative data from ORO staining (Fig 3A) showed that there
was significantly more ORO staining in the SSEA4()
population than in the SSEA4(þ) population (1.145 0.142 v 0.695 0.062, n ¼ 4, P ¼ .007) (Fig 3B). Realtime PCR data showed that both the LPL and CEBP
genes were found to be expressed at significantly higher
levels in the SSEA4() population relative to the
SSEA4(þ) population (1.230 0.049 v 1.001 0.053,
n ¼ 4, P ¼ .005, and 1.162 0.037 v 1.001 0.049,
n ¼ 4, P ¼ .010, respectively) (Fig 3C), supporting the
results observed with ORO staining.
SSEA4(D) Expression Benefits hSDSC
Osteogenesis
After 21 days of osteogenic induction, ARS staining
showed that SSEA4(þ) cells produced distinct bone
nodules with extensive calcium staining; very little, if
any, calcium staining was seen in the SSEA4() population (Fig 4A). The SSEA4(þ) population also displayed
a significantly elevated expression of OPN relative to the
SSEA4-SORTED SDSC DIFFERENTIATION
357
Fig 3. Adipogenic induction and
evaluation
of
SSEA4-sorted
hSDSCs. (A) After cell sorting,
unsorted (control) and sorted
[SSEA4(þ) and SSEA4()] cells
were expanded for 1 more passage. The expanded cells were
incubated in an adipogenic induction medium for 21 days. (B) ORO
staining was used for lipids. (Original magnification 200.) Quantitative data included biochemical
analysis (ORO optical density [OD]
adjusted by total DNA content).
(C) Real-time PCR analysis for LPL
and CEBP mRNAs (adipogenic
markers). Data are shown as mean
standard deviation (n ¼ 4). The
asterisks indicate significant differences (P < .05).
SSEA4() population (1.010 0.173 v 0.462 0.062,
n ¼ 4, P ¼ .007) (Fig 4B).
Discussion
In this study we hypothesized that SSEA4 as a surface
marker could preferentially sort hSDSC populations with
lineage differentiation potentials for cartilage, bone, and
adipose tissue regeneration. We found that SSEA4()
SDSCs had lower proliferation but higher chondrogenic
and adipogenic potential whereas SSEA4(þ) SDSCs
showed retained proliferative and enhanced osteogenic
abilities. The utility of the surface marker SSEA4 in
identification of chondrogenic, adipogenic, and osteogenic differentiation of SDSCs could help us better understand its role in mesenchymal differentiation and
potential clinical regenerative medicine applications.
We chose hSDSCs as target cells in this study rather
than hBMSCs because we expected to find an
appropriate adult stem cell for cartilage repair; SDSCs
are stem cells that are easily accessible and have a
sufficient cell number and favorable differentiation capacity. Despite the fact that BMSCs were the first stem
cells discovered14 with multilineage differentiation potentials,15 the harvesting process is an invasive technique and results in a low harvest rate. The colony
formation rate of primary BMSCs is approximately 1/
105 to 20/105, the lowest among MSCs from mesenchymal tissues.16 The tendency to develop hypertrophy
and endochondral ossification when induced to undergo chondrogenesis also discouraged its application in
cartilage regeneration.17 Unlike BMSCs, SDSCs are
harvested from synovial tissue, which can be conducted
by arthroscopy, a minimally invasive surgical procedure. According to Sakaguchi et al.,7 a mean of
21,000 cells/mg of synovium could be obtained, indicating that a small sample of synovium harvested with
358
J. LI ET AL.
Fig 4. Osteogenic induction and evaluation of SSEA4-sorted
hSDSCs. (A) After cell sorting, unsorted (control) and sorted
[SSEA4(þ) and SSEA4()] cells were expanded for 1 more
passage. The expanded cells were incubated in an osteogenic
induction medium for 21 days. ARS staining was used for calcium deposition. (Original magnification 200.) (B) Quantitative data included real-time PCR analysis for OPN mRNA
(osteogenic marker). Data are shown as mean standard
deviation (n ¼ 4). The asterisk indicates a significant difference
(P < .05).
punch biopsy would be sufficient to obtain SDSCs for
clinical treatment. The colony number for nucleated
cells derived from synovium was 100-fold higher than
that for cells from bone marrow. Those cells also achieved the highest colony-forming efficiency, fold increase, and growth kinetics among stem cells derived
from bone marrow, adipose, muscle, and periosteum.18
SDSCs are also considered a tissue-specific stem cell for
chondrogenesis, retaining stronger proliferative and
chondrogenic potential.3
Despite the advantages of SDSCs over BMSCs for cartilage regeneration, like other stem cells, donor age and
ex vivo expansion resulting in cell senescence are also
challenges for the application of SDSCs in cartilage regeneration.19 Our recent findings indicated that DSCM
could rejuvenate SDSCs during ex vivo cell propagation,
enhancing cell proliferation and chondrogenic potential.20
This cell expansion system has been tested in primary
chondrocytes and nucleus pulposus cells, as well as adult
stem cells.11,20 Most recently, we found that the unique
components and elasticity of DSCM deposited by fetal
hSDSCs might be responsible for the rejuvenation of adult
hSDSCs in terms of the lower apoptosis rate and higher
proliferative and chondrogenic potentials.21 The combination of this expansion system with a growth factor
(such as basic fibroblast growth factor [fibroblast growth
factor 2]) and low oxygen (5% O2) contributed to a
36-fold cell number increase in only a 6-day expansion
compared with the untreated control group without
compromising SDSC chondrogenic potential.12 The previously mentioned evidence suggested that DSCM, a
3-dimensional matrix that can be deposited by a tissuespecific stem cell from young and healthy donors, is
easily available through a decellularizing process, which
shows promise for providing a sufficient number of highquality SDSCs for clinical applications.
Recent findings showed that SSEA4 was expressed by
human MSCs isolated from a variety of sources,
including amniotic fluid,22 bone marrow,13,23 dermis,24
ligament,25 synovium,11,12 and umbilical cord blood.26
The existence of SSEA4 in many types of stem cells
raises the question of whether SSEA4 levels represent a
population in stem cells favoring a lineage-specific differentiation. Maddox et al.27 found that the level of
SSEA4 in either human breast tissue stem cells or human abdominal adipose tissue stem cells (hASCs)
remained relatively consistent with passaging. However, SSEA4-positive hASCs exhibited a higher potential
for differentiation toward osteogenic and adipogenic
cell lineages in vitro when compared with a mixed
population.27 In an independent study using hASCs,
Mihaila et al.28 found that SSEA4(þ) hASCs exhibited
an enhanced capacity to differentiate into endothelial
cells and osteoblasts. Consistent with these previous
reports, our data showed that SSEA4(þ) SDSCs exhibited an enhanced osteogenic capacity.
In this study we cannot explain the increase of SSEA4positive cells in the SSEA4() group after passaging; it
could be because of the presence of a group of highly
proliferative cells within the heterogeneous SDSC population, as in the findings of Karystinou et al.29 We also
found that SSEA4() hSDSCs exhibited enhanced
chondrogenic and adipogenic capacities. SSEA4(þ) did
not favor hSDSC chondrogenesis, which is corroborated
by a recent report that SSEA4 was not a marker for
chondrogenic potential in cultured human chondrocytes
and hBMSCs.30 An assumed shared pathway between
adipogenesis and chondrogenesis in SSEA4() hSDSCs
SSEA4-SORTED SDSC DIFFERENTIATION
359
Fig 5. Clinical perspectives using
surface marker SSEA4 sorting for
SDSC lineage differentiation. Synovial
tissue is harvested by arthroscopy,
followed by enzyme digestion to
release SDSCs. If more cells are
needed, cell expansion can be processed with the addition of growth
factors (GFs) such as fibroblast growth
factor 2 (FGF2) and/or DSCM.
Fluorescence-activated cell sorting
(FACS) can be used to preferentially
separate SSEA4(þ) and SSEA4()
cells for cartilage, bone, and adipose
regeneration.
was supported by a recent report that CD271(þ) SDSCs
had greatly enhanced chondrogenic capabilities, as well
as adipogenic potential.31
In BMSCs, various projects have shown that osteogenesis and adipogenesis appear to involve competing
pathways. Most evidence to support this supposition
comes from osteoporosis studies, in which increased
bone marrow adipose tissue is known to correlate with
decreased bone volume.32 More specifically, Li et al.33
showed the effects of overexpression of S100A16 resulting in a significant increase in ORO staining with a
decrease in ARS staining and, vice versa, with decreasing
expression of S100A16, suggesting direct and competing
effects of osteogenesis and adipogenesis. In addition,
hypoxic conditions have been found to affect MSC differentiation. Induction of the transcription factor
hypoxia-inducible factor 1 has been found to enhance
osteogenesis and inhibit adipogenesis in BMSCs.34 These
findings suggest that certain proteins in BMSCs have the
capability to simultaneously promote differentiation of 1
lineage fate while inhibiting differentiation of the competing lineage.
Our findings suggest a similar scenario in SDSCs. When
examining the amount of lipid staining in the sorted cell
populations, as well as the adipogenic mRNA levels, we
observed a significant elevation in the SSEA4() cells.
Alternatively, calcium staining and osteogenic mRNA
levels were elevated in the SSEA4(þ) cells. These data
support the theory of competing pathways involved in
adipogenic and osteogenic differentiation, despite the
stem cell source used.
Interestingly, it appears that the presence of the surface marker SSEA4 allows for a significant induction of
SDSCs into bone formation, despite previous evidence
that SDSCs have a reduced capability for osteogenesis.3
Our results are in line with previous reports showing
that SSEA4 could be a positive marker for selecting
BMSCs with higher proliferative and osteogenic potentials both in vitro and in vivo, which provide a useful
tool for enumerating BMSCs in vivo.23,35 These data
indicate the strong role of SSEA4 in osteogenesis
and provide evidence of the benefit of using SSEA4
as a potential marker for stem cellebased bone
regeneration.
As summarized in Fig 5, synovial tissue can be easily
obtained by minimally invasive arthroscopic procedures. Enzymatic digestion is able to release stem cells
from synovium. If more cells are needed, the initial
number of SDSCs can be magnified through the addition of growth factors such as fibroblast growth factor 2
and/or expansion using DSCM. Expanded cells can be
quickly sorted based on surface marker SSEA4 staining
360
J. LI ET AL.
using flow cytometry. Ultimately, these SSEA4(þ) and
SSEA4() cells could be applied in autologous cartilage,
bone, or fat regeneration.
Limitations
One limitation of this study was the unstable expression
of SSEA4 in the sorted groups after cell expansion. Because
of the insufficient cell number in the SSEA4() group
available for multilineage differentiation evaluation, all 3
groups were expanded for 1 more passage. Flow cytometry was used to evaluate SSEA4 expression in expanded
cells. We found that, compared with 70.3% in the unsorted cells, the SSEA4(þ) cells decreased the SSEA4
expression percentage from 99.8% to 79.2% whereas the
SSEA4() cells had an increased percentage from 4.4% to
53.3%. Despite the fact that the change in SSEA4
expression after cell expansion was not what we expected,
the difference of about 10% among the 3 groups
[SSEA4(þ) > unsorted > SSEA4()] did contribute a
significant difference in multilineage differentiation capacities. Given the enhanced osteogenic potential in the
SSEA4() SDSC population and considering the inherent
properties of SDSCs favoring chondrogenesis and BMSCs
benefiting osteogenesis, it will be interesting to investigate
whether this SSEA4-based cell sorting approach can
concentrate the BMSC population with higher osteogenic
capacity. The low number of patients tested in this study is
also a limitation of our report. Because the focus was on
the expression of SSEA4 in the overall SDSC population,
synovial samples collected from 4 young patients were
pooled for this study. The donor-to-donor variability will
be explored in a future investigation.
Conclusions
In this study, for the first time, we showed the benefit
of using the surface marker SSEA4 in SDSCs to preferentially sort a mixed population of cells. SSEA4(þ)
SDSCs indicated a strong potential for osteogenesis
rather than chondrogenesis and adipogenesis. This study
provides evidence showing that SDSC-based mesenchymal tissue regeneration can be easily achieved by
arthroscopic harvesting followed by quick cell sorting.
Acknowledgment
The authors thank Suzanne Danley for her help in
editing the manuscript.
References
1. Wagner J, Kean T, Young R, Dennis JE, Caplan AI.
Optimizing mesenchymal stem cell-based therapeutics.
Curr Opin Biotechnol 2009;20:531-536.
2. Bilgen B, Ren Y, Pei M, Aaron RK, Ciombor DM. CD14negative isolation enhances chondrogenesis in synovial
fibroblasts. Tissue Eng Part A 2009;15:3261-3270.
3. Jones BA, Pei M. Synovium-derived stem cells: A tissuespecific stem cell for cartilage engineering and regeneration. Tissue Eng Part B 2012;18:301-311.
4. Pei M, He F, Boyce BM, Kish VL. Repair of full-thickness
femoral condyle cartilage defects using allogeneic synovial
cell-engineered tissue constructs. Osteoarthritis Cartilage
2009;17:714-722.
5. Pei M, He F, Kish V, Vunjak-Novakovic G. Engineering of
functional cartilage tissue using stem cells from synovial
lining: A preliminary study. Clin Orthop Relat Res 2008;466:
1880-1889.
6. Pei M, He F, Vunjak-Novakovic G. Synovium-derived
stem cell-based chondrogenesis. Differentiation 2008;76:
1044-1056.
7. Sakaguchi Y, Sekiya I, Yagishita K, Muneta T. Comparison
of human stem cells derived from various mesenchymal
tissues: Superiority of synovium as a cell source. Arthritis
Rheum 2005;52:2521-2529.
8. Nöth U, Rackwitz L, Steinert AF, Tuan RS. Cell delivery
therapeutics for musculoskeletal regeneration. Adv Drug
Deliv Rev 2010;62:765-783.
9. Campbell D, Pei M. Surface markers for chondrogenic
determination: A highlight of synovium-derived stem
cells. Cells 2012;1:1107-1120.
10. Yanagisawa M. Stem cell glycolipids. Neurochem Res
2011;36:1623-1635.
11. Li J, He F, Pei M. Creation of an in vitro microenvironment to enhance human fetal synovium-derived stem cell
chondrogenesis. Cell Tissue Res 2011;345:357-365.
12. Li JT, Jones B, Zhang Y, Vinardell T, Pei M. Low density
expansion rescues human synovium-derived stem cells
from replicative senescence. Drug Deliv Transl Res 2012;2:
363-374.
13. Pei M, He F, Kish VL. Expansion on extracellular matrix
deposited by human bone marrow stromal cells facilitates
stem cell proliferation and tissue-specific lineage potential. Tissue Eng Part A 2011;17:3067-3076.
14. Friedenstein AJ. Precursor cells of mechanocytes. Int Rev
Cytol 1976;47:327-359.
15. Muraglia A, Cancedda R, Quarto R. Clonal mesenchymal
progenitors from human bone marrow differentiate in vitro
according to a hierarchical model. J Cell Sci 2000;113:
1161-1166.
16. Castro-Malaspina H, Ebell W, Wang S. Human bone
marrow fibroblast colony-forming units (CFU-F). Prog Clin
Biol Res 1984;154:209-236.
17. Pelttari K, Winter A, Steck E, et al. Premature induction of
hypertrophy during in vitro chondrogenesis of human
mesenchymal stem cells correlates with calcification and
vascular invasion after ectopic transplantation in SCID
mice. Arthritis Rheum 2006;54:3254-3266.
18. Yoshimura H, Muneta T, Nimura A, Yokoyama A, Koga H,
Sekiya I. Comparison of rat mesenchymal stem cells
derived from bone marrow, synovium, periosteum, adipose tissue, and muscle. Cell Tissue Res 2007;327:449-462.
19. Li JT, Pei M. Cell senescence: A challenge in cartilage engineering and regeneration. Tissue Eng Part B 2012;18:270-287.
20. Pei M, Li JT, Shoukry M, Zhang Y. A review of decellularized stem cell matrix: A novel cell expansion system for
cartilage tissue engineering. Eur Cell Mater 2011;22:333-343;
discussion 343.
SSEA4-SORTED SDSC DIFFERENTIATION
21. Li JT, Hansen K, Zhang Y, et al. Rejuvenation of chondrogenic potential by young stem cell microenvironment.
Biomaterials 2014;35:642-653.
22. De Coppi P, Bartsch G Jr, Siddiqui MM, et al. Isolation of
amniotic stem cell lines with potential for therapy. Nat
Biotechnol 2007;25:100-106.
23. Gang EJ, Bosnakovski D, Figueiredo CA, Visser JW,
Perlingeiro RC. SSEA-4 identifies mesenchymal stem cells
from bone marrow. Blood 2007;109:1743-1751.
24. Vaculik C, Schuster C, Bauer W, et al. Human dermis
harbors distinct mesenchymal stromal cell subsets. J Invest
Dermatol 2012;132:563-574.
25. Zhang J, Pan T, Im HJ, Fu FH, Wang JH. Differential properties of human ACL and MCL stem cells may be responsible
for their differential healing capacity. BMC Med 2011;9:68.
26. Suila H, Pitkänen V, Hirvonen T, et al. Are globoseries
glycosphingolipids SSEA-3 and -4 markers for stem cells
derived from human umbilical cord blood? J Mol Cell Biol
2011;3:99-107.
27. Maddox JR, Ludlow KD, Li F, Niyibizi C. Breast and
abdominal adipose multipotent mesenchymal stromal
cells and stage-specific embryonic antigen 4 expression.
Cells Tissues Organs 2012;196:107-116.
28. Mihaila SM, Frias AM, Pirraco RP, et al. Human adipose
tissue-derived SSEA-4 subpopulation multi-differentiation
potential towards the endothelial and osteogenic lineages.
Tissue Eng Part A 2013;19:235-246.
361
29. Karystinou A, Dell’Accio F, Kurth TB, et al. Distinct
mesenchymal progenitor cell subsets in the adult human
synovium. Rheumatology 2009;48:1057-1064.
30. Schrobback K, Wrobel J, Hutmacher DW, Woodfield TB,
Klein TJ. Stage-specific embryonic antigen-4 is not a
marker for chondrogenic and osteogenic potential in
cultured chondrocytes and mesenchymal progenitor cells.
Tissue Eng Part A 2013;19:1316-1326.
31. Arufe MC, De la Fuente A, Fuentes I, De Toro FJ,
Blanco FJ. Chondrogenic potential of subpopulations of
cells expressing mesenchymal stem cell markers derived
from human synovial membranes. J Cell Biochem 2010;111:
834-845.
32. Kawanabe N, Murata S, Fukushima H, et al. Stage-specific
embryonic antigen-4 identifies human dental pulp stem
cells. Exp Cell Res 2012;318:453-463.
33. Li D, Zhang R, Zhu W, et al. S100A16 inhibits osteogenesis but stimulates adipogenesis. Mol Biol Rep 2013;40:
3465-3473.
34. Wagegg M, Gaber T, Lohanatha FL, et al. Hypoxia promotes osteogenesis but suppresses adipogenesis of human
mesenchymal stromal cells in a hypoxia-inducible factor1 dependent manner. PLoS One 2012;7:e46483.
35. Rosu-Myles M, McCully J, Fair J, et al. The globoseries
glycosphingolipid SSEA-4 is a marker of bone marrowderived clonal multipotent stromal cells in vitro and
in vivo. Stem Cells Dev 2013;22:1387-1397.