3D-Printed Poly(-caprolactone) Scaffold

Augmented With Mesenchymal Stem Cells

for Total Meniscal Substitution
A 12- and 24-Week Animal Study in a Rabbit Model
Zheng-Zheng Zhang,* MD, PhD, Shao-Jie Wang,*y MD, PhD, Ji-Ying Zhang,* BS,
Wen-Bo Jiang,z PhD, Ai-Bing Huang,* MD, PhD, Yan-Song Qi,* MD, PhD,
Jian-Xun Ding,§ PhD, Xue-Si Chen,§ PhD, Dong Jiang,*|| MD, and Jia-Kuo Yu,*|| MD, PhD
Investigation performed at the Institute of Sports Medicine of Peking University Third Hospital,
Beijing, China

Background: Total meniscectomy leads to knee osteoarthritis in the long term. The poly(e-caprolactone) (PCL) scaffold is a prom-
ising material for meniscal tissue regeneration, but cell-free scaffolds result in relatively poor tissue regeneration and lead to joint
degeneration.
Hypothesis: A novel, 3-dimensional (3D)–printed PCL scaffold augmented with mesenchymal stem cells (MSCs) would offer ben-
efits in meniscal regeneration and cartilage protection.
Study Design: Controlled laboratory study.
Methods: PCL meniscal scaffolds were 3D printed and seeded with bone marrow–derived MSCs. Seventy-two New Zealand
White rabbits were included and were divided into 4 groups: cell-seeded scaffold, cell-free scaffold, sham operation, and total
meniscectomy alone. The regeneration of the implanted tissue and the degeneration of articular cartilage were assessed by gross
and microscopic (histological and scanning electron microscope) analysis at 12 and 24 weeks postoperatively. The mechanical
properties of implants were also evaluated (tensile and compressive testing).
Results: Compared with the cell-free group, the cell-seeded scaffold showed notably better gross appearance, with a shiny white
color and a smooth surface. Fibrochondrocytes with extracellular collagen type I, II, and III and proteoglycans were found in both
seeded and cell-free scaffold implants at 12 and 24 weeks, while the results were significantly better for the cell-seeded group at
week 24. Furthermore, the cell-seeded group presented notably lower cartilage degeneration in both femur and tibia compared
with the cell-free or meniscectomy group. Both the tensile and compressive properties of the implants in the cell-seeded group
were significantly increased compared with those of the cell-free group.
Conclusion: Seeding MSCs in the PCL scaffold increased its fibrocartilaginous tissue regeneration and mechanical strength, pro-
viding a functional replacement to protect articular cartilage from damage after total meniscectomy.
Clinical Relevance: The study suggests the potential of the novel 3D PCL scaffold augmented with MSCs as an alternative me-
niscal substitution, although this approach requires further improvement before being used in clinical practice.
Keywords: knee; meniscus; stem cell therapy; tissue engineering

Lesions in the meniscus are the most frequently recorded
orthopaedic diagnosis.14 More than 1.5 million people across
the United States and Europe receive partial or total menis-
cectomy annually.37 However, the abnormal physiological
stress placed on articular cartilage postoperatively often
leads to knee osteoarthritis and related morbidity.15,26
Meniscal allograft transplant provides an alternative to
The American Journal of Sports Medicine, Vol. 45, No. 7
DOI: 10.1177/0363546517691513
Ó 2017 The Author(s)
meniscectomy, but long-term results of such transplants
are unsatisfactory given poor tissue remodeling and are con-
troversial due to issues of compatibility and risk of disease
transmission.18,19 Moreover, no total meniscal substitute
has been approved by the Food and Drug Administration.
Tissue engineering, which aims to regenerate damaged
tissue, offers a potential strategy for meniscal replacement.
Recently, numerous materials have been investigated as
scaffolds to construct meniscal implants. However, tissue-
based materials or extracellular matrix (ECM) compo-
nents, such as tendon,23 small intestinal submucosa,8
and collagen meniscal implants,43 have been shown to

1497
1498 Zhang et al The American Journal of Sports Medicine

possess unsatisfactory mechanical properties for in vivo Study Design

application.50 Synthetic polymers, such as polyurea,32,46
poly(e-caprolactone) (PCL),24,25,30 and polyglycolide,1 have
also been investigated for preparing meniscal scaffolds.
However, polyurea is thought to potentially release carci-
nogenic compounds upon degradation.45 Polyglycolide scaf-
fold with a porosity of 95% demonstrated successful
fibrochondrocyte seeding but had insufficient mechanical
properties after 7 weeks of culture.1 PCL is a promising
material for meniscal tissue engineering given its superior
mechanics, bioactivity, and material processability.33 In
addition, PCL loses its molecular weight in vivo consider-
ably more slowly than do other aliphatic polyesters,27
and the scaffold could maintain initial mechanical support
until adequate tissue ingrowth occurs.
As previously reported, a 3-dimensional (3D) PCL scaf-
fold fabricated by fused deposition modeling (FDM) was
used to create tissue-engineered meniscus.51 Although
cell-free scaffolds showed potential in fibrocartilaginous
tissue regeneration, joint degeneration occurred in the
early stage after implantation. Compared with strategies
that used seeding cells, cell-free augmentation resulted
in inadequate ECM secretion and insufficient mechanical
strength of implants, which might be the reason for joint
degeneration.21 Therefore, we determined that further ani-
mal studies were necessary to compare cell-seeded scaf-
folds with cell-free scaffolds in terms of fibrocartilaginous
tissue regeneration, biomechanical properties, and carti-
lage protection for total meniscal replacement.
In this context, the 3D-printed PCL scaffolds were seeded
with mesenchymal stem cells (MSCs); these are seed cells
used in meniscal tissue engineering that are promising for
their significant healing capacity in a meniscal defect model,
but they have not been used for total meniscal replacement.16
Next, total meniscal substitution was performed in a rabbit
model. The fibrocartilaginous formation (histology and immu-
nohistochemistry), the mechanical strength, and the chondro-
protective effect of the implants were evaluated and compared
with those of the cell-free scaffold and the native meniscus.
Our hypothesis was that the 3D-printed PCL scaffold seeded
with MSCs would offer some benefits in meniscal regenera-
tion, mechanical strength, and chondroprotection. The results
of the present study might offer an alternative for total menis-
cal substitution, providing further insight into clinical treat-
ment for meniscal lesions.
Seventy-two skeletally mature, male New Zealand White
rabbits weighing 3.0 kg were used for animal experiments.
Forty-eight animals underwent total medial meniscectomy
of the left knees and were divided into 2 groups. Twenty-
four of the left knees received implantation by means of
PCL scaffolds seeded with MSCs (cell-seeded group), and
24 others received cell-free scaffolds (cell-free group). As
a control, the remaining 24 rabbits underwent total medial
meniscectomy of the right knees without implantation
(Meni group). The left knees of the Meni group received
sham operation involving exposure of the medial joint
and closure in layers (sham group). Twelve rabbits were
randomly selected to be sacrificed at the 12-week and 24-
week time points. At each time point, 6 knees from each
group were collected for synovial fluid testing, followed
by histological evaluation for implant regeneration and
articular cartilage degeneration. Six other knees under-
went biomechanical testing of implants and microscopic
observation of cartilage surface by means of scanning elec-
tron microscope (SEM).
Fabrication of 3D PCL Scaffolds
Five skeletally mature, male New Zealand White rabbit
legs were scanned by magnetic resonance imaging (3.0-T
Signa HDxt; GE Healthcare) (512 3 512 pixels bitmap
image, 0.31-mm resolution). The magnetic resonance imag-
ing data were imported into a Mimics software system
(version 17.0; Materialise), and then a 3D reconstruction
model of the meniscus was generated to calculate its size,
including the posterior horn width, anterior horn width,
anterior horn length, posterior horn length, posterior
horn to anterior horn distance, peripheral horn thickness,
and peripheral horn width. For the rabbit medial meniscal
scaffold, a wedge-shaped arc disk was selected as a typical
model (Figure 1, A-C). As previously described regarding
the fabrication process, PCL (average molecular weight
74,600 g mol-1 and melting point 52.9°C) (Changchun Sino-
Biomaterials) was melted and extruded through a heated
metal nozzle, which was controlled by FDM software. For
the scaffold parameters, 215 6 23 mm pore size with
300 mm road width (diameter of the printed fiber) and
200 mm space was determined to be optimal.51

Harvest, Culture, and Seeding of MSCs

METHODS
Bone marrow–derived MSCs were isolated from 3-month-
All experimental protocols involving animals were approved old New Zealand White rabbits. Isolation, cultivation,
by the local institutional animal care and use committee. and trilineage differentiation potential assays of MSCs

||
Address correspondence to Jia-Kuo Yu, MD, PhD, or Dong Jiang, MD, Institute of Sports Medicine, Peking University Third Hospital, No. 49 North
Garden Road, Haidian, Beijing, China 100191 (email: yujiakuo@126.com; bysyjiangdong@126.com).
*Institute of Sports Medicine, Beijing Key Laboratory of Sports Injuries, Peking University Third Hospital, Beijing, P.R. China.
y
Department of Joint Surgery, Zhongshan Hospital of Xiamen University, Xiamen University, Xiamen, P.R. China.
z
Clinical Translational R&D Center of 3D Printing Technology, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine,
Shanghai, P.R. China.
§
Key Laboratory of Polymer Ecomaterials, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun, P.R. China.
One or more of the authors has declared the following potential conflict of interest or source of funding: This study was supported by the National Nat-
ural Scientific Foundation of China (Grant Nos. 51273004, 31200725, 31670982, 81630056).
AJSM Vol. 45, No. 7, 2017 Scaffold Augmented With MSCs for Total Meniscal Substitution 1499

Figure 1. (A) Anatomic reconstruction model of rabbit menisci in left knee. (B) A typical model of 3D medial meniscal scaffold. (C)
3D-printed poly(e-caprolactone) (PCL) scaffold seeded with mesenchymal stem cells (scale bar represents 10 mm). (D) PCL scaf-
fold (black arrow) implanted between femur and tibia, with medial collateral ligament (green arrow) reserved.

were performed as previously described.13 MSCs in the
third passage were then seeded on the scaffold (5 3 106
cells per scaffold) by means of a centrifugal method.52
Briefly, the scaffold was placed at the bottom of a 1.5-mL
centrifugal tube, and 100.0 mL of concentrated cell solution
was added. The tube was centrifuged at 500 rpm for 1 min-
ute and then turned over as 1 loop. The process was
repeated 3 consecutive times. The seeded scaffolds were
cultured under Dulbecco’s modified eagle medium (DMEM)
for 24 hours before implantation.
Surgical Procedure
Antibiotic prophylaxis was administered preoperatively by
means of intramuscular penicillin (400,000 U). The
implantation procedure was performed as previously
reported.20 Briefly, after anesthesia and routine prepara-
tion, the knee was approached through a medial parapatel-
lar incision. A total meniscectomy was performed by
resecting the medial meniscus sharply along the periphery
and detaching it from its anterior and posterior junction.
Care was taken not to injure the medial collateral liga-
ment, which is important for the postoperative stability
of the knee joint (Figure 1D). Both the anterior and poste-
rior horns and periphery of the scaffolds were reattached to
the respective root attachments and appropriate adjacent
synovium with absorbable No. 4-0 sutures (Ethicon). For
the posterior root, attachment of the medial meniscus is
adjacent to the posterior cruciate ligament.6 We used
a self-made threading apparatus and extracapsular knot-
ting technique to fix the posterior horn of the scaffold with
the ligamentous structures. The joint capsule, periarticular
tissue, and skin were closed with No. 3-0 Vicryl sutures (Ethi-
con). For the Meni group, total resection of the medial menis-
cus was performed only in the right knees; the left knees,
serving as sham controls, underwent a procedure involving
exposure of the medial joint and closure in layers. Animals
were allowed unlimited movement after the operation. The
antibiotic prophylaxis was continued for 3 days. At each
time point, the randomly selected animals were sacrificed
via pentobarbital sodium.
Synovial Fluid Collection and Analysis
At each time point, synovial fluid was collected via a 2-mL
syringe with an 18-gauge needle, as previously reported.17
The supernatants were assayed for interleukin 1 (IL-1) and
tumor necrosis factor a (TNF-a) by means of standard
enzyme-linked immunosorbent assay (ELISA) kits (Rabbit
IL-1 ELISA Kit, I079SC; Rabbit TNF-a ELISA Kit, T103SC;
Hermes Criterion Biotechnology).
Evaluation of Implants
Half of the knees of the experimental group were excised,
and then the exposed femoral condyles and the tibial sur-
face with the implants in place were photographed. Three
researchers who were blinded to the experimental groups
evaluated the menisci by using the Gross Evaluation of
Meniscus Implant Score.24 This score includes 9 parame-
ters: implant integration, implant position, horn position,
shape, presence of tears in the implant, implant surface,
implant size, tissue quality, and condition of the synovia.
In this system, each parameter is scored from 1 to 3 based
1500 Zhang et al
on the condition of the meniscal implants. The Gross Eval-
uation of Meniscus Implant Score is an evaluation system
for meniscal implants that is based on previous litera-
ture.20,24,25,51 For scoring, each meniscal implant was cut
at the level of the medial collateral ligament. The speci-
mens were cut to produce blocks exposing their wedge-
shaped profile, which showed outer and inner regions
and superior and inferior surfaces of the regenerated
menisci in histological sections. For each implant (total of
6 implants for each group), inner, intermediate, and outer
zones were evaluated. Blocks of meniscal implant were
scored for the presence of residual scaffold, foreign body
response, cellularity, blood vessel ingrowth, fibrosis, carti-
laginous matrix, integration of the implant with the joint
capsule, and inflammatory cell infiltrate. The above fea-
tures were scored as being either present or absent. Then
the implant was dissected from the tibia and fixed in
10% (vol/vol) neutral buffered formalin (Sigma Diagnos-
tics). All samples were then dehydrated and embedded in
paraffin. Next, 6-mm-thick cross sections were cut and
stained with hematoxylin and eosin (H&E) and toluidine
blue (TB, positive for proteoglycans). In addition, sections
were treated by an immunohistochemistry procedure
with labeling of collagen I and II (Calbiochem) and picro-
sirius red (PR, staining to distinguish collagen I and III).
The sections of implants were analyzed by blinded
researchers according to the meniscal histology scoring
system.24 Moreover, immunohistochemical analyses were
used to conduct semi-quantitative study of the differences
of collagen contents within the implants.53 Briefly, 10 dig-
ital images were captured by an Olympus BX-51 micro-
scope. Integrated optical density (IOD) value and area of
each microimage were measured with Image-Pro Plus 6.0
software (Media Cybernetics).47 The descriptors of relative
density (IOD per area) were measured to semi-quantify the
deposition of collagen I and II.
Evaluation of Joint Cartilages
The cartilages of the femur and tibia were macroscopically
evaluated according to the criteria of the International
Cartilage Repair Society (ICRS) cartilage lesion classifica-
tion.3 Three researchers who were blinded to the experi-
mental groups analyzed the cartilage of the medial
femoral condyle and the tibial plateau. The mean scores
of the researchers were then calculated.
For microscopic observation of the cartilage surface,
specimens were assessed by JSM5600LV SEM (JEOL
USA). Briefly, samples were trimmed without disturbing
the cartilage surface, fixed immediately in 10 mL of 25%
glutaraldehyde for 1 day at 4°C, dehydrated in a graded
ethanol series, and finally subjected to critical point drying
for complete dehydration. The surface of the cartilage was
coated with a 5-nm layer of gold and then viewed.
For histological evaluation, osteochondral specimens
were fixed in 10% (vol/vol) neutral buffered formalin and
decalcified in 10% ethylenediaminetetraacetic acid (Titri-
plex III; Merck). At 2 to 3 weeks, the osteochondral speci-
mens were sectioned in the coronal plane at the midpoint
of the tibial plateau and in the sagittal plane at the
The American Journal of Sports Medicine
midpoint of the medial femoral condyle, as described previ-
ously.20 The slides of the femur and tibia were stained with
H&E and TB and graded blindly according to the Mankin
grading system.34
Biomechanical Analyses
The biomechanical properties of specimens were assessed by
means of a materials testing machine (AG-IS; Shimadzu).31
The specimens were prepared as previously described.35,44
Briefly, in tensile testing, the samples (100 mm thick)
were cut into rectangular shapes from the peripheral edge
of each meniscus, leaving a 3-mm gauge length in the
approximate circumferential direction for testing. The sam-
ples could not be cut into ‘‘dogbone’’ shapes owing to their
small size. The sample was tested to failure at a rate of
0.06 mm/s. Multiple samples from each meniscus or implant
were tested. The elastic modulus was analyzed from the lin-
ear portion of the stress-strain curve.10 Given the small size
of the medial rabbit meniscus, only the posterior portion
could be cut for sampling in confined compression testing.
The samples were loaded into a custom-made jig.35 The
creep response of the specimen under the step force was
monitored until equilibrium (defined as slope \1 3
10-6 mm/s, at least 7200 s). Then the test force was automat-
ically removed and the recovery phase began until equilib-
rium (defined as slope \1 3 10-6 mm/s, at least 3600 s).
Overall, the creep indenter yielded the creep and recovery
deformation of each specimen in response to a 0.02-N step
load.44 The aggregate modulus and permeability were calcu-
lated by using Mow’s biphasic theory.38
Statistical Analyses
All statistical data were expressed as mean 6 standard
deviation (SD). Significance of the results was determined
with analysis of variance (ANOVA) or Mann-Whitney test
and repeated-measures tests with Bonferroni correction.
All data analyses were performed via SPSS statistical soft-
ware (version 15.0; SPSS Inc). P \ .05 was considered sta-
tistically significant, and P \ .01 and P \ .001 were
considered highly significant.
RESULTS
Gross Observation
All of the rabbits had regained their normal gait patterns
at 2 weeks postoperatively. No significant weight changes
or complications were seen. No signs of intra-articular
hemorrhage, swelling, or inflammation were seen in the
joints. The synovial fluid was clear postmortem.
At the 12-week time point, all the scaffolds formed
meniscal tissue in vivo (Figure 2). The cell-seeded scaffold
became better integrated to the joint, with no sign of tears
and gap formation. Most of the implants maintained their
original shape and size, while only 1 implant was slightly
smaller than the original construct. The new meniscal tis-
sue formed with the cell-free scaffold showed significantly
AJSM Vol. 45, No. 7, 2017 Scaffold Augmented With MSCs for Total Meniscal Substitution 1501
Figure 2. Macroscopic observations of joints at 12 and 24
weeks after operation. Meniscal tissue excised from the tibial
plateau is shown on the right. Scale bars represent 10 mm.
Meni, group that underwent total medial meniscectomy with-
out implantation of right knees.
less integration with the joint compared with the seeded
scaffold (P 5 .02; Table 1), although no significant difference
was found for total score between these 2 groups (P 5 .108).
After 24 weeks, the gross appearance of all meniscal tis-
sues from the cell-seeded or the cell-free groups was nor-
mal with a shiny white color and a smooth surface. No
obvious synovial hypertrophy or implant tears were found
in either group. The new meniscal tissue formed with the
cell-free scaffold had an inferior shape, especially in the
anterior horn (P 5 .025). Moreover, significant differences
were found between the 2 groups in total scores at week 24
(P 5 .013).
At each time point, the meniscus was intact in the sham
control group. Moreover, no sign of joint degradation was
observed. However, various degrees of cartilage damage
were observed in the other operated joints, especially in
the Meni group.
Histological Evaluation of Implants
The histological evaluation of implants (Table 2) revealed no
evidence of a significant inflammatory cell infiltration in the
implants. At week 12, a small number of lymphocytes were
found, mostly within new tissue, while foreign body reaction
(macrophages, foreign body giant cells) sometimes appeared
around the polymer scaffolds. The persistence of scaffold
residuals together with a consequent foreign body reaction
was present in both the cell-seeded and the cell-free implant
groups. Moreover, the foreign body reaction localized in
some areas of the implants, where hypocellularity was evi-
dent. Specially, the cell-seeded group showed a general
trend for a reduction of lymphocytes and foreign body giant
cells from 12 to 24 weeks with the absorption of the scaffold
and the new tissue regeneration. At each time point, all
implants showed extensive vascularization in the outer
regions of the new meniscal tissue.
The main differences observed between the cell-seeded
and cell-free groups were related to fibrosis and cartilage
metaplasia (Figure 3). At 12 weeks, new cartilaginous
matrix deposition in the tip region of the implant was
observed in 8 cases (5 cell-seeded and 3 cell-free), among
which 5 cases (3 cell-seeded and 2 cell-free) presented car-
tilaginous staining also at the central region of the new
meniscal tissue. At 24 weeks, there were more cases show-
ing cartilaginous matrix deposition in the cell-seeded
group, and most cases revealed staining at the tip and cen-
tral regions. Otherwise, cell phenotype appeared different
between groups. Interestingly, in the cell-seeded group,
a small number of round cells around the cartilage islands
presented in the inner region of the implant (Figure 3,
Aiii), which exhibited similar cell morphologic characteris-
tics based on the observation of native meniscus in the
sham group (Figure 3, Ciii). However, in the cell-free
group, a large number of spindle-shaped cells with elon-
gated nuclei were surrounded by fibrous tissue in each
region of the implants (Figure 3, Bi-Biii). At 24 weeks,
the various cell phenotypes within implants at different
zones were also observed in the cell-seeded group. It was
apparent that outer zone cells had fusiform-shaped fibro-
blast-like cells, while cells in the inner portion appeared
as circular fibrochondrocytic shapes that were embedded
in cartilage islands (Figure 3D), which was similar to the
native meniscus in the sham group (Figure 3F).
Immunohistochemical, PR, and TB
Evaluation of Implants
Immunohistochemistry and PR staining were carried out to
explore the collagen component and type of the implants. In
PR staining, red and birefringent fibers represented collagen
1502 Zhang et al The American Journal of Sports Medicine

TABLE 1
Gross Evaluation of Meniscal Implant Scorea

P Valuesb

Parameter Cell-Seeded, 12 wk Cell-Free, 12 wk Cell-Seeded, 24 wk Cell-Free, 24 wk a b

c
Integration 2.0 (1-3) 1.2 (1-2) 2.2 (2-3) 1.5 (1-2) .020 .067
Implant position 2.0 (1-3) 1.7 (1-2) 2.3 (2-3) 1.8 (1-3) .654 .397
Horn position 2.2 (1-3) 1.7 (1-3) 2.5 (2-3) 2.0 (1-3) .403 .403
Shape 2.0 (1-3) 1.7 (1-2) 2.7 (2-3) 1.7 (1-2) .606 .025c
Tears 2.7 (2-3) 2.2 (2-3) 2.8 (2-3) 2.3 (2-3) .149 .149
Surface 1.3 (1-2) 1.5 (1-2) 2.3 (2-3) 1.7 (1-2) .830 .078
Size 1.7 (1-2) 1.5 (1-2) 2.2 (2-3) 1.7 (1-3) .864 .292
Tissue 2.2 (1-3) 1.8 (1-3) 2.3 (2-3) 1.8 (1-3) .746 .524
Synovia 2.3 (2-3) 2.0 (1-3) 2.3 (2-3) 2.0 (1-3) .685 .685
Total score 18.3 (14-21) 14.7 (11-19) 21.7 (20-23) 16.2 (11-22) .108 .013c

a
Each parameter is scored from 1 to 3 based on the condition of the meniscal implant. Scores are provided as mean (range).
b
a represents cell-seeded 12 wk versus cell-free 12 wk; b represents cell-seeded 24 wk versus cell-free 24 wk.
c
Statistically significant (P \ .05).

TABLE 2
Histological Features of Implantsa

Feature Cell-Seeded, 12 wk Cell-Free, 12 wk Cell-Seeded, 24 wk Cell-Free, 24 wk

Residual scaffold 6 (100) 6 (100) 6 (100) 6 (100)

Foreign body reaction 6 (100) 6 (100) 4 (66.7) 6 (100)
Hypocellular areas 4 (66.7) 6 (100) 4 (66.7) 5 (83.3)
Blood vessels 6 (100) 6 (100) 6 (100) 6 (100)
Fibrosis 3 (50) 6 (100) 1 (16.7) 5 (83.3)
Cartilage metaplasia
Tip 5 (83.3) 3 (50) 6 (66.7) 3 (50)
Central 3 (50) 2 (33.3) 5 (83.3) 2 (33.3)
Integration
Good 5 (83.3) 3 (50) 6 (100) 4 (66.7)
Poor 1 (16.7) 3 (50) 0 (0) 2 (33.3)
Inflammatory infiltrate
Lymphocytes 6 (100) 6 (100) 4 (66.7) 6 (100)
Plasma 0 0 0 0
Neutrophils 0 0 0 0

a
The features were scored as being either present or absent for each implant (total of 6 implants for each group). Data are provided as n
(%).

I, and green and thin fibers with weak birefringence repre-
sented collagen III. The proteoglycan content of the tissues
was detected by means of TB staining. As shown by PR stain-
ing, the native menisci in the sham group showed strong
staining for collagen I (Figure 4). In the cell-seeded group,
the implants at week 12 formed a matrix with partial collagen
I staining. At 24 weeks, the cell-seeded and sham groups
showed large amounts of red fibers (collagen I), especially in
intermediate and outer regions, compared with the cell-free
group. Consistent with the PR staining, the cell-free group
revealed weaker staining than the other groups in immuno-
histochemical images, which was also indicated by the rela-
tive density of IOD per area value. The relative density of
IOD per area value was used for semi-quantitative analysis
of the collagen component in immunohistochemistry. The pro-
tein expression of collagen I in the seeded scaffold was lower
at week 12 compared with the native meniscus (P \ .001),
while the matrix increased at week 24 (Figure 6A). The value
of IOD per area was higher in the cell-seeded group than the
cell-free group at week 24 (P \ .001).
As shown in Figure 5, the native menisci in the sham
control group demonstrated strong TB staining in the
inner and central regions, whereas the implants in the
cell-free scaffolds at weeks 12 and 24 showed only slight
proteoglycan staining. In the cell-seeded group, TB stain-
ing appeared stronger at week 24 than at week 12. At
each time point, stronger collagen II staining was observed
in the cell-seeded group compared with the cell-free group,
which is presented by IOD per area value (Figure 6B).
Evaluation of Joint Cartilage Degradation
According to the histopathological observation of articular
cartilage and the scoring systems used (Figure 7), the
AJSM Vol. 45, No. 7, 2017 Scaffold Augmented With MSCs for Total Meniscal Substitution 1503

Figure 3. Macroscopic and histological images (general view and high-power magnification) of meniscus tissue. Gross views of
implants or native menisci are shown on the left. Insert boxes identify the region analyzed by histology. General views of outer,
intermediate, and inner zones of implants or native menisci (hematoxylin and eosin staining) are shown in the middle (scale bars 5
100 mm). High-power magnification images of meniscus tissue are shown on the right (scale bars 5 25 mm). Fibroblast-like cells
(black arrows) and chondrocyte-like cells (asterisks) were observed simultaneously in the implants of the cell-seeded group and
the sham group.

sham group showed the lowest grade of degenerative
changes, with a mean value of 0.17 to 0.5 for the Mankin
score, and some specimens showed only cartilage surface
irregularities. In contrast, the femoral condyle and tibial
plateau revealed complete disorganization of the cartilage
and severe reduction of TB staining in the total meniscec-
tomy group at 12 weeks. The highest mean Mankin and
ICRS scores at each time point were found in the Meni
group, and the cartilage damage progressed with time.
The cell-free scaffold groups revealed diffuse chondrocyte
cloning and clefts to the transitional or radial zone with
slight reduction of TB staining, which indicated less severe
cartilage degradation than that of the Meni group. How-
ever, the degree of cartilage damage increased with time.
Compared with the Meni and cell-free groups, the cell-
seeded group fared significantly better based on ICRS
and Mankin scores. Surface irregularities with diffuse
hypercellularity were displayed. Moreover, no significant
changes appeared at each time point.
SEM was used to assess microscopic changes in the sur-
faces of articular cartilage (Figure 8). Articular surfaces of
blocks taken from the sham group did not differ signifi-
cantly from normal cartilage.42 A relatively uniform area
without cracks was observed, and some of the residues
noted may be attributable to synovial fluid crystals or
bone fragments. In contrast, in the Meni group, the femoral
condyle and tibial plateau showed deep cracks and lacunae
on the surface enmeshed within frayed fibrils. The cell-free
scaffold groups also revealed superficial cracks and large
amounts of uniformly fibrous surfaces. In the cell-seeded
group, the articular cartilage surface appeared disrupted,
with a hillocky appearance, at 12 weeks. At 24 weeks, few
superficial cracks were noted in the cartilage surface.
To quantify inflammatory factors, synovial fluid was
collected to analyze IL-1 and TNF-a (Figure 9). Similar
to the tendency reported above, the greatest number of
cytokines was observed in the Meni group at 12 weeks.
Although IL-1 and TNF-a levels in the cell-seeded group
1504 Zhang et al
Figure 4. Representative picrosirius red (PR) and immunohistochemical staining for collagen I (Col I) of regenerated and native
menisci. Crassi red fiber represents Col I in PR-stained images. Scale bars represent 50 mm.
were identical to those of the sham group at 12 weeks,
these inflammatory factors increased progressively with
time. At 24 weeks postoperatively, the amount of either
cytokine in the cell-seeded group was greater than that
in the sham group while lower than that in the cell-free
or Meni group.
Evaluation of Biomechanical Properties of Implant
At 12 and 24 weeks, the biomechanical properties of native
meniscus were compared with those of the sham group,
and no differences were found. In tensile testing, the ten-
sile modulus of the native meniscus was significantly
higher than that of the cell-seeded or cell-free group at
each time point (Figure 10A). In addition, the value of
the tensile modulus in the cell-seeded group was higher
than that of the cell-free group at 24 weeks (P 5 .019).
The ultimate tensile strength of the implants in the cell-
seeded and cell-free groups at weeks 12 and 24 was approx-
imately half that of the native medial menisci (Figure
10C). The ultimate strength in the cell-seeded group was
The American Journal of Sports Medicine
larger than that of the cell-free group at week 24 (P 5
.0423).
Confined compressive creep showed that the implants in
both the cell-seeded and the cell-free groups were 20% to
40% as stiff in compression as the native meniscus (Figure
10B). Moreover, the value of aggregate modulus in the
cell-seeded group was higher than that of the cell-free group
at 24 weeks (P \ .0001). The permeability of the scaffolds
was significantly reduced after implantation (Figure 10D).
At 12 weeks, there were no differences between groups.
DISCUSSION
The purpose of this study was to investigate the fibrocarti-
laginous regeneration, mechanical properties, and carti-
lage protection capacity of a novel 3D-printed PCL
scaffold augmented with MSCs after implantation as
a full meniscal substitution in a rabbit model. To our
knowledge, this is the first study to examine the possibility
of regenerating the whole meniscus in a total meniscec-
tomy animal model by implanting MSCs-seeded scaffolds
AJSM Vol. 45, No. 7, 2017 Scaffold Augmented With MSCs for Total Meniscal Substitution 1505

Figure 5. Representative toluidine blue (TB) and immunohistochemical staining for collagen II (Col II) of regenerated and native
menisci. Scale bars represent 50 mm.

Figure 6. Immunohistochemical analyses of native meniscus and implants. Values for integrated optical density (IOD) per area of
(A) collagen I (Col I) and (B) collagen II (Col II) were larger in the cell-seeded group compared with the cell-free group, similar to the
native meniscus at 24 weeks (***P \ .001).

directly into the knee. Throughout 24 weeks, significantly
better tissue formation, lower cartilage degeneration, and
retained mechanical strength were found in the cell-seeded
scaffolds. The results indicated that the 3D-printed PCL
scaffold seeded with MSCs might be an alternative for
meniscal replacement.
In this study, a PCL scaffold fabricated by FDM was
used to create tissue-engineered meniscus. Compared
with traditional processes such as particle/salt leaching11
and electrospinning,2 FDM is a rapid prototyping tech-
nique that can produce a 3D scaffold with complete control
of geometric parameters such as pore size, porosity, and
1506 Zhang et al The American Journal of Sports Medicine

Figure 7. (A) Hematoxylin and eosin (H&E) and toluidine blue (TB) staining of articular cartilage surfaces in the femoral condyle
and tibial plateau (scale bars 5 100 mm). As indicated by (B) International Cartilage Repair Society (ICRS) and (C) Mankin scores,
the cell-seeded group presented lower cartilage degeneration in both the femur and tibia compared with cell-free group or Meni
group (*P \ .05, **P \ .01, ***P \ .001).
AJSM Vol. 45, No. 7, 2017 Scaffold Augmented With MSCs for Total Meniscal Substitution 1507

Figure 8. Scanning electron microscope (SEM) images of articular cartilage surfaces in femoral condyle and tibial plateau (scale
bars 5 5 mm).

Figure 9. Synovial fluid assessment. Levels of (A) interleukin 1 (IL-1) and (B) tumor necrosis factor a (TNF-a) in synovial fluid were
lower in the cell-seeded group compared with cell-free group or Meni group (*P \ .05, **P \ .01, ***P \ .001).

pore interconnection size.28,48,49 Moreover, PCL is a prom-
ising material because of its superior bioactivity and
mechanical characteristics.50 At week 24, better fibrocarti-
laginous formation coupled with lower foreign body reac-
tion was present in the cell-seeded group. These results
suggest the feasibility of this scaffold material as well as
the potential use of MSC augmentation in tissue engineer-
ing. MSCs have the ability to differentiate into chondro-
cytes, adipocytes, and osteoblasts, and they have immune
regulatory capacity. Although residual materials consist-
ing mainly of polymeric components caused foreign body
reaction, most of the implants showed new vessel ingrowth
and excellent integration with surrounding tissues. In the
present study, the response of the host tissue against more
or less degradation-resistant foreign macromolecular
material appeared as previously reported.36,41 We consider
that this low degree of lymphocyte infiltration combined
with foreign body reaction is not significant; a significant
response would include a large amount of neutrophil and
eosinophil infiltration and even biomaterial encapsulation
and failure. In addition, the degradation profiles of PCL
extended to 12 months in another study25; thus, the scaf-
fold can sustain loading until adequate tissue ingrowth
occurs with time.
The characterization of meniscus cells appears inconsistent
in the literature, and various terms are used (ie, fibrochondro-
cytes, meniscus cells, fibrocytes, fibroblasts, chondrocytes).39
Regardless of the term used, it is apparent that outer zone
1508 Zhang et al
Figure 10. Biomechanical properties. (A) Tensile modulus and (C) ultimate tensile strength of the scaffold (time 0) and implants
compared with the native medial meniscus. (B) Compressive aggregate modulus and (D) permeability of the scaffold (time 0) and
implants compared with the native posterior portion of medial meniscus (*P \ .05, ***P \ .001; n 5 6 per group).
cells have a fusiform shape and are similar to fibroblasts.
These fibroblast-like cells are surrounded with collagen I
(80% composition by dry weight).4 In contrast, round
chondrocyte-like cells in the inner region are embedded in
an ECM composed largely of collagen II and glycosamino-
glycans. In previous research regarding tissue-engineered
meniscus, Kon et al25 reported that a polycaprolactone/hya-
luronic acid scaffold seeded with autologous chondrocytes
was implanted into a sheep model. After 12 months, the his-
tological image of the implant showed cartilage metaplasia
with spherical and chondroid cells, while large amounts of
fibroblast-like cells and fibrous tissue deposition were pre-
sented in the cell-free scaffold group.25 Other investigators
also reported the presence of collagen I and II in implanted
meniscal scaffolds.21,22,36,41 In our study, the various cell
phenotypes within implants at different zones were
observed in the cell-seeded group at 12 weeks. Meanwhile,
ECM deposition, with histological evidence of collagen I
and II, occurred at a higher level in the cell-seeded group
compared with the cell-free group. In particular, the
fibroblast-like cells around the scaffold in the outer two-
thirds appeared elongated, suggesting that tensile strength
oriented circumferentially and radially might stimulate
MSCs to differentiate into fibroblasts and produce collagen
I.7 The rounded shape of cells in the inner margin implied
a response to compression, and research has demonstrated
that compressive loading induces MSC chondrogenesis.40
Although Patel et al41 also showed various distributions of
cellular phenotypes and collagens in a 1-year in vivo exper-
iment using an cell-free scaffold, a similar result was
observed in the current study at 24 weeks in the cell-seeded
group. These data confirm the benefit of seeding cells in
order to produce meniscal regeneration.
The American Journal of Sports Medicine
Several researchers have reported the development of
tissue-engineered meniscus using a biodegradable scaffold
seeded with native meniscus cells or chondrocytes.5,11,21
We believe that autologous cells will be safe for clinical
treatment. However, the protocol of using autologous cells
exhibited several limitations. Two surgical interventions
would be required. Moreover, given the scarcity and dedif-
ferentiation of autologous cells after culture and expansion,
as well as the possibility that autologous cells are in either
an age-related disease state or a degenerated state, their
application in tissue engineering is limited.9 The present
study is a primary exploration of the potential of allogenic
MSCs for meniscal tissue engineering. No reports are avail-
able describing measurable immune response when allo-
genic or xenogeneic cell sources are used, so for further
clinical translation, a 1-step strategy using a bioactive factor
for endogenous cell (ie, stem cell, progenitor cell) homing
will be investigated to overcome the limitations of allogenic
cell transplantation.30 As previously reported, MSCs can
escape immune recognition and exhibit an immune-
tolerance capacity.12,29 Moreover, the allogenic MSC aug-
mentation approach for meniscal defect or cartilage repair
has proven successful.13,16 In the present study, although
we noted no allogenic response of New Zealand White rab-
bits, we observed no obvious immunological rejection or sig-
nificant inflammatory reactions in histological and synovial
fluid assessment. Therefore, the present study using allo-
genic MSCs is a preliminary examination of a model that
requires further study. In regard to clinical application,
the presence and outcome of immune reaction should be
investigated. In the present study, we observed not only
increased quantity but also improved quality of the regener-
ated fibrocartilaginous tissue in knees implanted with MSC-
AJSM Vol. 45, No. 7, 2017 Scaffold Augmented With MSCs for Total Meniscal Substitution 1509
seeded scaffolds. Although these cell-free scaffolds showed
an IOD per area value of collagen I similar to that of the
cell-seeded group at 12 weeks (P 5 .063, Figure 6A), the
inferior cartilage metaplasia features of implants seen in
Table 2 illustrated that these scaffolds ultimately filled
with fibrous tissue as opposed to meniscus fibrocartilage.
In contrast, scaffolds treated with MSCs regenerated
with well-bonded tissue resembling the native meniscus in
matrix composition and cellularity, suggesting the potential
of MSCs in constructing a functional tissue-engineered
meniscus.
As tissue-engineered meniscus plays a key role in joint
load distribution, similar to the native meniscus, the
implant should become resistive to tensile and compressive
loads in vivo. Unfortunately, few reported scaffolds have
strength comparable to that of the native meniscus.30 As
anticipated in the current study, ultimate tensile strength
and tensile modulus were reduced after implantation, due
to polymer fiber degradation. However, from 12 to 24
weeks, tensile properties did not change significantly,
which suggested that regenerated tissue took over the
load that the scaffold could no longer bear in its degrading
state. Confined compression testing revealed a significant
decrease in permeability of the scaffolds in both the cell-
seeded and the cell-free groups from time zero to 24 weeks
after implantation. This decreased permeability may have
resulted from tissue infiltration into the scaffold. In addi-
tion, the value of aggregate modulus in implants was lower
than that of native meniscus, although the implants in the
cell-seeded group were significantly stiffer than unseeded
implants at week 24. In the cell-seeded group, the compres-
sive value at week 24 was double that of time-zero scaf-
folds, which might be due to the tissue deposition and
organization. To our knowledge, no tissue-engineered
meniscus has achieved greater compressive moduli than
native meniscus, and few in vivo studies have reported
these values in a confined creep model.25 Further research
is needed to increase the initial tensile and compressive
stiffness of the scaffold and deposition of ECM.
Although the chondroprotective effect of the cell-seeded
scaffold was demonstrated based on ICRS and Mankin
scores, cartilage degeneration was not completely pre-
vented. Significant differences were found between the
cell-seeded group and the sham group at each time point
in each scoring system. This could be attributed to several
reasons. One concern was the biochemical content of the
implant. Native meniscus is an anisotropic fibrocartilagi-
nous tissue, with various regions composed of different
types of collagen resistant to diverse loading.33 The trans-
planted tissue in our study mainly contained collagen I and
II, but the distribution and deposition of these collagens
did not provide the tissue with normal biological capacity.
One possible reason is that the distribution of new tissue
could have been affected by extrusion and/or thinning.
Further study is needed to create meniscal constructs
with appropriate fixation to replicate the biochemical prop-
erties of native tissue. As well, the scaffold degradation
and tissue ingrowth played key roles in protection of carti-
lage, as previously described. Therefore, the cell-seeded
meniscal tissue could not completely simulate the biochem-
ical and biomechanical properties of the native meniscus or
fully prevent cartilage degeneration. We acknowledge that
sham and total meniscectomy in the same rabbit might
have little effect on the results due to weight distribution
after surgery. However, the sham operation simply
involved exposure of the medial joint and closure in layers,
which had no effect on the menisci and articular cartilage.
As shown by the histological and SEM images of cartilage
and meniscus, the degeneration of cartilage in the Meni
group increased with time and few signs of joint degrada-
tion were observed in the sham control. Therefore, we con-
sidered that the bilateral surgery had little effect on the
results.
The major limitation of the present study is that we did
not follow the outcome of the implanted MSCs within the
knee joint. Although the distribution, migration, and out-
come of these transplanted cells are still unclear due to
the limited tracking methods, green fluorescent protein
or other markers might be needed to label MSCs for track-
ing progeny after implantation.16 However, the labeling
time was limited and the seeded MSCs would eventually
proliferate and differentiate into chondrocytes, fibrochon-
drocytes, or other phenotypes, which might increase the
difficulty of cell labeling and tracing. Other limitations of
translational assumptions in this study include those
that are inherent to the animal model. Rabbit knees may
have a weightbearing pattern that differs from that of
human knees, for the rabbit is a mostly sedentary animal
whose knees are flexed more than 90°. On the basis of
these results, a longer implantation time will be observed
in another model (ie, sheep, goat, or primate). Finally, after
the operation, animals were allowed unlimited movement,
which was not representative of the restrictions on move-
ment and loading after procedures in humans. In light of
this, caution should be used when extrapolating these
results to clinical treatment.
CONCLUSION
In this study, we demonstrated that a novel 3D-printed PCL
scaffold augmented with MSCs acted as a functional menis-
cal replacement and had a chondroprotective effect after 24
weeks of implantation. The cell-seeded scaffold showed
superiority in tissue regeneration, biomechanical strength,
and chondroprotection. The present study is also a primary
exploration of regenerating menisci with MSCs, which is
rarely reported but might be promising for meniscal tissue
engineering. In future, a 3D-printed, human meniscus–
shaped biomaterial scaffold will be investigated if tissue
organization and mechanics are comparable with the full
size of the human meniscus. Moreover, further research
will focus on a delivery approach for a human clinical trial
in which bioactive polypeptides or drugs are packaged in
a good laboratory practice (GLP)/good manufacturing prac-
tice (GMP) facility. Thus, rather than transplant an allo-
graft meniscus, surgeons could implant a ready-made
meniscal product using the existing arthroscopic procedure.
1510 Zhang et al
REFERENCES
1. Aufderheide AC, Athanasiou KA. Comparison of scaffolds and culture
conditions for tissue engineering of the knee meniscus. Tissue Eng.
2005;11(7-8):1095-1104.
2. Baker BM, Mauck RL. The effect of nanofiber alignment on the matura-
tion of engineered meniscus constructs. Biomaterials. 2007;28(11):1967-
1977.
3. Brittberg M, Winalski CS. Evaluation of cartilage injuries and repair.
J Bone Joint Surg Am. 2003;85(suppl 2):58-69.
4. Cheung HS. Distribution of type I, II, III and V in the pepsin solubilized
collagens in bovine menisci. Connect Tissue Res. 1987;16(4):343-
356.
5. Chiari C, Koller U, Dorotka R, et al. A tissue engineering approach to
meniscus regeneration in a sheep model. Osteoarthritis Cartilage.
2006;14(10):1056-1065.
6. Civitarese D, Donahue TL, LaPrade CM, et al. Qualitative and quan-
titative measurement of the anterior and posterior meniscal root
attachments of the New Zealand white rabbit. J Exp Orthop.
2016;3(1):10.
7. Connelly JT, Vanderploeg EJ, Mouw JK, Wilson CG, Levenston ME.
Tensile loading modulates bone marrow stromal cell differentiation
and the development of engineered fibrocartilage constructs. Tissue
Eng Part A. 2010;16(6):1913-1923.
8. Cook JL, Tomlinson JL, Kreeger JM, Cook CR. Induction of meniscal
regeneration in dogs using a novel biomaterial. Am J Sports Med.
1999;27(5):658-665.
9. Darling EM, Athanasiou KA. Rapid phenotypic changes in passaged
articular chondrocyte subpopulations. J Orthop Res. 2005;23(2):425-432.
10. Drury JL, Dennis RG, Mooney DJ. The tensile properties of alginate
hydrogels. Biomaterials. 2004;25(16):3187-3199.
11. Esposito AR, Moda M, Cattani SM et al. PLDLA/PCL-T scaffold for
meniscus tissue engineering. Biores Open Access. 2013;2(2):138-147.
12. Frank MH, Sayegh MH. Immunomodulatory functions of mesenchy-
mal stem cells. Lancet. 2004;363(9419):1411-1412.
13. Fu WL, Zhou CY, Yu JK. A new source of mesenchymal stem cells for
articular cartilage repair: MSCs derived from mobilized peripheral
blood share similar biological characteristics in vitro and chondro-
genesis in vivo as MSCs from bone marrow in a rabbit model. Am
J Sports Med. 2014;42(3):592-601.
14. Garrett WE Jr, Swiontkowski MF, Weinstein JN, et al. American
Board of Orthopaedic Surgery Practice of the Orthopaedic Surgeon:
Part II, certification examination case mix. J Bone Joint Surg Am.
2006;88(3):660-667.
15. Holzer LA, Leithner A, Holzer G. Surgery versus physical therapy for
meniscal tear and osteoarthritis. N Engl J Med. 2013;369(7):677.
16. Horie M, Driscoll MD, Sampson HW, et al. Implantation of allogenic
synovial stem cells promotes meniscal regeneration in a rabbit
meniscal defect model. J Bone Joint Surg Am. 2012;94(8):701-712.
17. Huang H, Zhang X, Hu, X et al. A functional biphasic biomaterial hom-
ing mesenchymal stem cells for in vivo cartilage regeneration. Bioma-
terials. 2014;35(36):9608-9619.
18. Jiang D, Ao YF, Gong X, Wang YJ, Zheng ZZ, Yu JK. Comparative
study on immediate versus delayed meniscus allograft transplanta-
tion: 4- to 6-year follow-up. Am J Sports Med. 2014;42(10):2329-
2337.
19. Jiang D, Yu JK. Immediate versus delayed meniscus allograft trans-
plantation: response. Am J Sports Med. 2015;43(5):NP9-10; discus-
sion NP11-12.
20. Jiang D, Zhao LH, Tian M, Zhang JY, Yu JK. Meniscus transplanta-
tion using treated xenogeneic meniscal tissue: viability and
chondroprotection study in rabbits. Arthroscopy. 2012;28(8):1147-
1159.
21. Kang SW, Son SM, Lee JS, et al. Regeneration of whole meniscus
using meniscal cells and polymer scaffolds in a rabbit total meniscec-
tomy model. J Biomed Mater Res A. 2006;77(4):659-671.
22. Klompmaker J, Jansen HW, Veth RP, Nielsen HK, de Groot JH, Pen-
nings AJ. Porous implants for knee joint meniscus reconstruction:
The American Journal of Sports Medicine
a preliminary study on the role of pore sizes in ingrowth and differen-
tiation of fibrocartilage. Clin Mater. 1993;14(1):1-11.
23. Kohn D, Wirth CJ, Reiss G, et al. Medial meniscus replacement by
a tendon autograft: experiments in sheep. J Bone Joint Surg Br.
1992;74(6):910-917.
24. Kon E, Chiari C, Marcacci M, et al. Tissue engineering for total menis-
cal substitution: animal study in sheep model. Tissue Eng Part A.
2008;14(6):1067-1080.
25. Kon E, Filardo G, Tschon M, et al. Tissue engineering for total menis-
cal substitution: animal study in sheep model—results at 12 months.
Tissue Eng Part A. 2012;18(15-16):1573-1582.
26. Krause W, Pope M, Johnson R, Wilder D. Mechanical changes in the
knee after meniscectomy. J Bone Joint Surg. 1976;58(5):599-604.
27. Kweon H, Yoo MK, Park IK, et al. A novel degradable polycaprolactone
networks for tissue engineering. Biomaterials. 2003;24(5):801-808.
28. Lantada AD, Morgado PL. Rapid prototyping for biomedical engi-
neering: current capabilities and challenges. Annu Rev Biomed
Eng. 2012;14:73-96.
29. Le Blanc K, Rasmusson I, Sundberg B, et al. Treatment of severe
acute graft-versus-host disease with third party haploidentical mes-
enchymal stem cells. Lancet. 2004;363(9419):1439-1441.
30. Lee CH, Rodeo SA, Fortier LA, Lu C, Erisken C, Mao JJ. Protein-
releasing polymeric scaffolds induce fibrochondrocytic differentiation
of endogenous cells for knee meniscus regeneration in sheep. Sci
Transl Med. 2014;6(266):266ra171.
31. Li P, Feng X, Jia X, Fan Y. Influences of tensile load on in vitro deg-
radation of an electrospun poly(L-lactide-co-glycolide) scaffold. Acta
Biomater. 2010;6(8):2991-2996.
32. Liu C, Abedian R, Meister R, et al. Influence of perfusion and com-
pression on the proliferation and differentiation of bone mesenchymal
stromal cells seeded on polyurethane scaffolds. Biomaterials.
2012;33(4):1052-1064.
33. Makris EA, Hadidi P, Athanasiou KA. The knee meniscus: structure-
function, pathophysiology, current repair techniques, and prospects
for regeneration. Biomaterials. 2011;32(30):7411-7431.
34. Mankin HJ, Dorfman H, Lippiello L, Zarins A. Biochemical and meta-
bolic abnormalities in articular cartilage from osteo-arthritic human
hips, II: correlation of morphology with biochemical and metabolic
data. J Bone Joint Surg Am. 1971;53(3):523-537.
35. Martin Seitz A, Galbusera F, Krais C, Ignatius A, Durselen L. Stress-
relaxation response of human menisci under confined compression
conditions. J Mech Behav Biomed Mater. 2013;26:68-80.
36. Merriam AR, Patel JM, Culp BM, Gatt CJ Jr, Dunn MG. Successful
total meniscus reconstruction using a novel fiber-reinforced scaffold:
a 16- and 32-week study in an ovine model. Am J Sports Med.
2015;43(10):2528-2537.
37. Moran CJ, Busilacchi A, Lee CA, Athanasiou KA, Verdonk PC. Bio-
logical augmentation and tissue engineering approaches in meniscus
surgery. Arthroscopy. 2015;31(5):944-955.
38. Mow VC, Kuei SC, Lai WM, Armstrong CG. Biphasic creep and
stress relaxation of articular cartilage in compression? Theory and
experiments. J Biomech Eng. 1980;102(1):73-84.
39. Nakata K, Shino K, Hamada M, et al. Human meniscus cell: charac-
terization of the primary culture and use for tissue engineering. Clin
Orthop Relat Res. 2001;391(suppl):S208-218.
40. O’Conor CJ, Case N, Guilak F. Mechanical regulation of chondrogen-
esis. Stem Cell Res Ther. 2013;4(4):61.
41. Patel JM, Merriam AR, Culp BM, Gatt CJ Jr, Dunn MG. One-year out-
comes of total meniscus reconstruction using a novel fiber-reinforced
scaffold in an ovine model. Am J Sports Med. 2016;44(4):898-907.
42. Sastre S, Suso S, Segur JM, et al. Hyaline cartilage surface study
with an environmental scanning electron microscope: an experimen-
tal study. J Mater Sci Mater Med. 2009;20(11):2181-2187.
43. Steadman JR, Rodkey WG. Tissue-engineered collagen meniscus
implants: 5- to 6-year feasibility study results. Arthroscopy. 2005;
21(5):515-525.
44. Sweigart MA, Athanasiou KA. Tensile and compressive properties of the
medial rabbit meniscus. Proc Inst Mech Eng H. 2005;219(5):337-347.
AJSM Vol. 45, No. 7, 2017 Scaffold Augmented With MSCs for Total Meniscal Substitution 1511

45. Szycher M. Biostability of polyurethane elastomers: a critical review.
J Biomater Appl. 1988;3(2):297-402.
46. Tienen TG, Heijkants RG, de Groot JH, et al. Replacement of the
knee meniscus by a porous polymer implant: a study in dogs. Am
J Sports Med. 2006;34(1):64-71.
47. Wang CJ, Zhou ZG, Holmqvist A, et al. Survivin expression quantified
by Image Pro-Plus compared with visual assessment. Appl Immuno-
histochem Mol Morphol. 2009;17(6):530-535.
48. Wang S-J, Zhang Z-Z, Jiang D, et al. Thermogel-coated poly(e-cap-
rolactone) composite scaffold for enhanced cartilage tissue engi-
neering. Polymers. 2016;8(5):200.
49. Zein I, Hutmacher DW, Tan KC, Teoh SH. Fused deposition modeling
of novel scaffold architectures for tissue engineering applications.
Biomaterials. 2002;23(4):1169-1185.
50. Zhang Z-Z, Jiang D, Wang S-J, et al. Scaffolds drive meniscus tissue
engineering. RSC Advances. 2015;5(95):77851-77859.
51. Zhang ZZ, Jiang D, Ding JX, et al. Role of scaffold mean pore size in
meniscus regeneration. Acta Biomater. 2016;43:314-326.
52. Zhang ZZ, Jiang D, Wang SJ, Qi YS, Zhang JY, Yu JK. Potential of
centrifugal seeding method in improving cells distribution and
proliferation on demineralized cancellous bone scaffolds for tissue-
engineered meniscus. ACS Appl Mater Interfaces. 2015;7(28):
15294-15302.
53. Zouein FA, Kurdi M, Booz GW, Fuseler JW. Applying fractal dimen-
sion and image analysis to quantify fibrotic collagen deposition and
organization in the normal and hypertensive heart. Microsc Micro-
anal. 2014;20(4):1134-1144.

For reprints and permission queries, please visit SAGE’s Web site at http://www.sagepub.com/journalsPermissions.nav.