A Comparison of Tenocytes and

Mesenchymal Stem Cells for Use in Flexor

Tendon Tissue Engineering
Gil S. Kryger, MD, Alphonsus K. S. Chong, MD, Melinda Costa, MD,
Hung Pham, BS, Steven J. Bates, MD, James Chang, MD

From the Division of Plastic Surgery, Stanford University Medical Center, Palo Alto, CA; Veteran’s
Administration Palo Alto Healthcare System, Palo Alto, CA; University of Pennsylvania School of Medicine,
Philadelphia, PA; and the Division of Plastic Surgery, Veteran’s Administration Palo Alto Healthcare System,
Palo Alto, CA.

Purpose: Tissue-engineered tendon grafts will meet an important clinical need. To engineer
tendons, we used acellularized allogeneic tendon as scaffold material. To determine the ideal
cell type to seed the scaffolds, we studied in vitro characteristics of epitenon tenocytes,
tendon sheath fibroblasts, bone marrow– derived mesenchymal stem cells (BMSCs), and
adipoderived mesenchymal stem cells (ASCs). Subsequently, we implanted reseeded acellu-
larized tendons in vivo as flexor tendon grafts.
Methods: Tenocytes, sheath fibroblasts, BMSCs, and ASCs were obtained from adult rabbits.
For all cell lines, collagen 1, 2, and 3 immunocytochemistry was performed, and proliferation
was assessed by hemacytometry and senescence by ␤-galactosidase staining. Flexor tendons
were acellularized after harvest. Tendons were assessed by histology after in vitro reseeding
with each of the cell types after 1, 4, and 8 weeks. Finally, reseeded tendons and controls
were implanted in a flexor profundus tendon defect. After 6 weeks, the reseeded tendons
were harvested and assessed by histology. Statistical analysis for cell proliferation was
performed using analysis of variance and t-tests with Bonferroni correction.
Results: All cell types had similar collagen expression. Cell proliferation was higher in ASCs
in late passage compared with early passage and in ASCs compared with epitenon tenocytes
at late passage. The other cell types were similar in growth characteristics. No senescence
was detected. In vitro assessment of reseeded constructs showed the presence of cells on the
construct surface. In vivo assessment after implantation showed viable cells seen within the
tendon architecture in all cell types.
Conclusions: This study suggests that the four cell types may be successfully used to
engineer tendons. Adipoderived mesenchymal stem cells proliferate faster in cell culture,
but the cell types were similar in other respects. All could be used to successfully
repopulate acellularized tendon in vivo as flexor tendon grafts. (J Hand Surg 2007;32A:
597– 605. Copyright © 2007 by the American Society for Surgery of the Hand.)
Key words: Flexor tendon, stem cells, tendon grafting, tendon repair, tissue engineering.

lexor tendon loss requiring tendon grafting is a come.2,3 However, this knowledge has not translated

F common clinical problem. The most common

tendons used for grafting are the palmaris,
plantaris, and foot extensor tendons.1 These donor
tendons are extrasynovial in origin and differ from
intrasynovial flexor tendons by the absence of an
epitenon layer. There is compelling evidence that
intrasynovial tendon grafts are superior to extrasyno-
vial tendons in terms of healing and ultimate out-
to wider clinical application because of a lack of
suitable intrasynovial tendon donor grafts.4
The application of tissue engineering techniques to
develop intrasynovial tendon grafts for tendon recon-
struction is a logical approach to overcome the lack
of donor tendons. A variety of scaffolds, biological
and synthetic, have been used in musculoskeletal
tissue engineering.5– 8 However, available scaffolds

The Journal of Hand Surgery 597

598 The Journal of Hand Surgery / Vol. 32A No. 5 May–June 2007
have inferior biological and material properties com-
pared with the tissue they are replacing.9 Much is
being done to improve biomaterials to match biolog-
ical needs. However, another approach involves us-
ing acellularized allogeneic tendon repopulated with
autologous cells. In this method, the material prop-
erties of the tendon are maintained, and the construct
best resembles native tendon. Experience with heart
valves and skin have shown that allogeneic acellu-
larized tissue repopulated with autogenous cells pro-
duces a reliable tissue-engineered construct.10,11 Ap-
plying this to tendons, we have found that it is
possible to produce acellularized tendon segments
for tissue engineering applications.
One major limitation in tissue engineering is the
time interval necessary to grow cells to populate
scaffolds.9 This is a concern particularly when using
primary differentiated cell lines. A more proliferative
cell line would allow a shorter time interval between
cell harvest and construct implantation in vivo.
We studied four candidate cell types for use in
reseeding acellularized tendon constructs. Specifi-
cally, we compared epitenon tenocytes, tendon
sheath fibroblasts, bone marrow– derived mesenchy-
mal stem cells (BMSCs), and adipoderived mesen-
chymal stem cells (ASCs) with respect to growth
characteristics, senescence, collagen production, and
viability of reseeded constructs in vitro. We also
studied the viability of reseeded tendon constructs in
vitro and after in vivo implantation in a clinically
relevant model of rabbit flexor tendon grafting.
Materials and Methods
Adult male New Zealand white rabbits (4.0 – 4.5 kg)
(n ⫽ 5 as cell and tendon source and n ⫽ 30 for in
vivo experiments) were used. All animal experiments
followed protocols approved by our institutional an-
imal ethics committee.
Cell Line Isolation, Culture, and Analysis
The rabbits were killed by an intravenous injection of
Euthasol (Pentobarbital 390 mg/Phentoin) (VIRBAC
AH Inc., Fort Worth, TX). Using sterile technique
under loupe magnification, the forepaw and hindpaw
flexor digitorum profundus (FDP) equivalents were
dissected and 2-cm portions of the intrasynovial dis-
tal FDP tendon and overlying flexor sheath were
excised. Specimens were placed into buffered
Hank’s saline solution with penicillin 100 IU/mL and
streptomycin 100 ␮g/mL.
Tenocytes and sheath fibroblasts were obtained
using a previously established protocol.12 Harvested
tendons were treated with 0.5% collagenase type IA
(Sigma, St. Louis, MO) for 10 minutes at room
temperature to obtain tenocytes. Tendon sheath fi-
broblasts were obtained from sheath tissue in the
same manner. The released cells were plated and
cultured in Ham’s F12 medium (Gibco, Rockville,
MD) with 10% fetal bovine serum (Gibco).
Bone marrow-derived stem cells were obtained
using established techniques.13,14 Rabbit bone mar-
row was aspirated via iliac crest puncture immedi-
ately after sacrifice. The aspirate was resuspended in
Ham’s F12 media with 10% rabbit serum and peni-
cillin/streptomycin. Cellular debris was removed by
straining through a 100-␮m nylon cell strainer (BD
Falcon, Franklin Lakes, NJ). A cell pellet was ob-
tained after centrifugation at 1,000 rpm for 5 minutes
at 4°C and resuspended. Cells were counted and
plated at a density of 5 ⫻ 106 nucleated cells/100
mm3 culture dish. On the following day, unattached
cells were removed by washing.
Adipoderived mesenchymal stem cells were ob-
tained from inguinal fat pads using a technique pub-
lished by Zuk et al.15,16 The fat was minced with
scissors and treated with 0.075% collagenase II in a
shaking water bath at 37°C for 10 minutes. Collage-
nase was then inactivated using fresh media. The cell
suspensions underwent centrifugation at 1,000 rpm
for 5 minutes at 4°C, and the resulting cell pellet was
resuspended and filtered through a 100-␮m nylon
cell strainer. After final centrifugation and resuspen-
sion, the cells were counted and incubated at a den-
sity of 5 ⫻ 106 nucleated cells/100 mm3 tissue cul-
ture dish in Ham’s F12 media.
All cells were grown at 37°C in a humidified
standard incubator with 5% CO2. Culture media was
changed every 3 days.
Immunocytochemistry
Each cell type was cultured on sterile glass coverslips
for 3 days. After washing with phosphate buffered
saline (PBS), the cells were fixed in 2% formalin in
PBS for 10 minutes. Hydrogen peroxide 0.3% was
used to quench endogenous peroxidase activity. The
following primary antibodies and dilutions were
used: anti-collagen 1, 1:100 (Oncogene Research
Products, Schwalbach, Germany); anti-collagen 2,
1:100 (Chemicon International, Temucula, CA); and
anti-collagen 3, 1:100 (Chemicon International) for 1
hour. Biotin-labeled secondary antibodies were used
at dilution 1:100 for 30 minutes. Detection was per-
formed using the Vectastain Elite ABC kit (Vector
Laboratories, Burlingame, CA) and VIP peroxidase
 Kryger et al / Tenocytes and Stem Cells for Tendon Engineering
Table 1. Summary of In Vivo Experimental Groups
Condition Description
Experiment 1 Implantation of epitenon
tenocyte-seeded
acellularized tendon
Experiment 2 Implantation of sheath
fibroblast-seeded
acellularized tendon
Experiment 3 Implantation of BMSC-seeded
acellularized tendon
Experiment 4 Implantation of ASC-seeded
acellularized tendon
Control 1 Implantation of autologous
tendon
Control 2 Implantation of allogeneic
tendon
Control 3 Implantation of acellularized
tendon
substrate (Vector Laboratories) according to the
manufacturer’s protocol. Negative controls were ob-
tained by omitting the primary antibody.
Cell Proliferation and Senescence
Viable cells were counted daily using a hemacytom-
eter on multiple representative plate regions. The
mean number of cells was calculated and plotted
against culture time to generate a growth curve. Dou-
bling times were calculated by dividing the total
number of hours in culture by the number of dou-
blings. Cell counts were performed at low passages
(4 to 9) and high passage (18 –21) to compare pro-
liferation rates.
Cell senescence was determined using the ␤-ga-
lactosidase activity staining assay.17 ␤-galactosidase
is found in senescent but not actively proliferating or
quiescent cells. Early passage cells were plated at a
density of 2 ⫻ 103 cells/cm2 and cultured for 2
weeks. After fixation for 5 minutes in 2% formalde-
hyde-glutaraldehyde, the ␤-galactosidase reaction
buffer (Sigma) containing 1 mg/mL X-gal 40 mmol
was incubated in a citric acid and sodium phosphate
buffer (pH 6.0) with 5 mmol potassium ferrous cya-
nide, 5 mmol potassium ferricyanide, 150 mmol so-
dium chloride, and 2 mmol magnesium chloride for 2
hours. Senescent cells stained blue. The percentage
of blue cells was calculated and compared between
cell types. This assay was repeated with cells at late
passage.
Acellularization of Flexor Tendons
Tendons for acellularization were harvested as de-
scribed above, washed with PBS, and subsequently
599
Time Analyzed (n) Purpose
6 weeks (n ⫽ 3) Experimental group
6 weeks (n ⫽ 3) Experimental group
6 weeks (n ⫽ 3) Experimental group
6 weeks (n ⫽ 3) Experimental group
4 weeks (n ⫽ 3), Assess baseline inflammation associated
8 weeks (n ⫽ 3) with tendon grafting
4 weeks (n ⫽ 3), Assess inflammation associated with
8 weeks (n ⫽ 3) allogeneic tendon grafting
4 weeks (n ⫽ 3), Assess inflammation associated with
8 weeks (n ⫽ 3) allogeneic acellularized tendon
frozen at –70°C. Frozen tendons were thawed to
room temperature and then placed into trypsin
0.05%/Ethylenediaminetetraaceticacid (EDTA) for
24 hours at 37°C followed by Triton X-100 0.5% for
24 hours at room temperature. Representative sam-
ples of tendon were obtained before and after treat-
ment for evaluation by light microscopy. Evaluation
of the samples confirmed that the protocol resulted in
complete acellularization of flexor tendons.
Reseeding of Acellularized Tendons In Vitro
Acellular tendon scaffolds (1.5 cm long) were re-
seeded by immersion of the tendons in media with
either epitenon tenocytes, tendon sheath fibroblasts,
BMSCs, or ASCs at a density of 2 ⫻ 106 cells/mL.
The cell-scaffold constructs were then incubated at
37°C in a humidified tissue culture chamber with 5%
CO2 to allow attachment of cells. The culture me-
dium was changed every other day during the incu-
bation period. Representative samples of tendon con-
structs were obtained at 1-, 4-, and 8-week intervals
for light microscopy and evaluation of construct vi-
ability (n ⫽ 3 each group). Cells from passage 2– 6
were used for reseeding of tendon scaffolds.
In Vivo Implantation of Reseeded Tendon
Constructs
The experimental conditions used are outlined in
Table 1. The rabbits were anesthetized with an intra-
muscular injection of acepromazine (0.01 mg/kg),
xylazine (5 mg/kg), and ketamine (50 mg/kg) and
maintained with inhalational isoflurane delivered via
face mask.
Under sterile conditions, a longitudinal incision
600 The Journal of Hand Surgery / Vol. 32A No. 5 May–June 2007
was made on the volar surface of the middle digit of
the right forelimb, between the metacarpophalangeal
and proximal interphalangeal joints. Tissues were
carefully dissected under loupe magnification until
the flexor sheath was identified. The sheath and over-
lying pulleys were sharply divided. The FDP tendon
was isolated, and a 1.5-cm-long section of tendon
was sharply excised. An immediate interposition ten-
don graft consisting of reseeded tendon constructs or
controls (see later) was sutured proximally and dis-
tally using 5-0 polypropylene (Prolene; Ethicon,
Somerville, NJ) using the Kessler technique. The
FDP tendon was then isolated proximally in the
forepaw and sharply transected to prevent loading of
the repaired tendon. Next, the tendon pulleys were
repaired with interrupted 5-0 polypropylene (Pro-
lene; Ethicon) stitches. The tourniquet was released,
and the skin was repaired using a running 5-0
polypropylene (Prolene; Ethicon) suture. A soft com-
pressive dressing was then applied to the right fore-
limb.
The surgical procedure was modified for the three
control groups. Control 1 was designed to assess the
inflammatory response associated with tendon au-
tografts. These rabbits underwent transection and
repair of the same tendon. Control 2 was designed to
assess the inflammatory response associated with
tendon allografts. Previously harvested tendons kept
in PBS for less than 2 hours were grafted into the
control animals. Control 3 was designed to assess the
inflammatory response and viability associated with
nonreseeded acellular tendon grafts. Previously har-
vested and acellularized tendon was implanted as
described earlier. There were six animals in each
control group.
Animals were fed standard lab chow and allowed
to roam unrestricted in their cages. The experimental
tendons were harvested at 6 weeks and the control
tendons at 4 and 8 weeks after surgery. The rabbits
were killed by intravenous injection of Euthasol. The
right forepaw incision was reopened, and the FDP
tendon and interpositional graft were dissected free
of surrounding tissues. The graft and adjoining prox-
imal and distal tendon were sharply transected en
bloc and preserved for histologic evaluation.
Histologic Analysis of Tendons
Harvested tissue was embedded with OCT Com-
pound (Tissue-Tek, Sakura Finetek, Japan) and
stored at – 80°C until use. Histologic sections were
cut at 10 ␮m with a cryostat (Leica CM 1800, Wet-
zlar, Germany), fixed with acetone for 10 minutes,
Figure 1. All four cell types demonstrate similar morphology,
including spindle-shaped cells with elongated fibroblast-like
cell bodies: (A) epitenon tenocytes, (B) sheath fibroblasts, (C)
BMSCs, (D) ASCs. Phase-Contrast with Methylene Blue,
100⫻.
and stored at –20°C until use. Hematoxylin and eosin
staining was performed in the standard manner. His-
tologic analysis was performed for tendon architec-
ture with specific emphasis on intact collagen pat-
tern, amount and nature of inflammatory response,
and tenocyte viability. Inflammatory response was
assessed by subjective histologic observation by an
experienced histotechnologist.
Statistical Analysis
Statistical analysis of the cell proliferation rates were
performed as follows. The area under the curve for
each cell proliferation experiment was calculated.
These were compared across cell types with one-way
analysis of variance (ANOVA). Subsequently, pair-
wise comparisons were performed between the cell
types using unpaired two-tailed Student’s t-tests with
Bonferroni correction. Early and late passage results
for each cell type were compared using unpaired
two-tailed Student’s t-tests. One-way ANOVA and
t-tests were performed using the Data Analysis Pack
in Microsoft Excel 9.0 (Microsoft Corporation, Red-
mond, WA) and Bonferroni correction performed with
R version 2.3.1 (http://www.R-project.org).
Results
Cell Culture and Analysis
All cell types were successfully grown in culture. All
four cell types (epitenon tenocytes, tendon sheath
fibroblasts, BMSCs, and ASCs) exhibited elongated
nuclei and spindle-shaped cytoplasm, characteristic
of a fibroblast-like morphology (Fig. 1).
 Kryger et al / Tenocytes and Stem Cells for Tendon Engineering
Figure 2. All four cell types demonstrate similar immunocy-
tochemical staining for collagen 1: (A) epitenon tenocytes,
(B) sheath fibroblasts, (C) BMSCs, (D) ASCs. Vector VIP Sub-
strate, 100⫻.
Immunocytochemistry
All cell types stained strongly for collagen 1 and 3 in
culture. Staining was strongest in the cytoplasm with
light staining in the extracellular space. As expected,
there was weak staining for collagen 2. There was no
appreciable difference between the cell types with
respect to staining (Figs. 2– 4).
Cell Proliferation and Senescence
The four cell types grew exponentially in culture and
reached confluence by day 6. Proliferation rates were
similar among the cell types except that ASCs
showed greater proliferation at later passage com-
pared with earlier passage (area under curve 6.8 ⫻
105 vs 1.5 ⫻ 105, p ⫽ .005) and was higher than
Figure 3. All four cell types demonstrate similar immunocy-
tochemical staining for collagen 2: (A) epitenon tenocytes,
(B) sheath fibroblasts, (C) BMSCs, (D) ASCs. Vector VIP Sub-
strate, 100⫻.
601
Figure 4. All four cell types demonstrate similar immunocy-
tochemical staining for collagen 3: (A) epitenon tenocytes,
(B) sheath fibroblasts, (C) BMSCs, (D) ASCs. Vector VIP Sub-
strate, 100⫻.
epitenon tenocytes at late passage (area under curve
6.8 ⫻ 105 vs 1.99 ⫻ 105, p ⫽ .031) (Fig. 5).
For each cell type, no notable senescence was
observed as late as passage 21. On average, 1% to
5% of cells stained positively for ␤-galactosidase at
early passage (Fig. 6). No difference in the staining
pattern was observed when these cells were exam-
ined at late passage (Fig. 7).
Reseeding of Acellularized Tendon In Vitro
All constructs appeared viable after 1 week in vitro.
Hematoxylin and eosin staining demonstrated a pres-
ervation of the collagen framework as well as clumps
of cells on the surface of the acellular construct.
There was no penetration of cells into the center of
the collagen scaffold at 1, 4, or 8 weeks. All con-
structs appeared similar in morphology (Fig. 8).
Tendon Graft Viability in Vivo
The seeded tendon grafts were histologically indis-
tinguishable between the different experimental
groups (tenocytes, sheath fibroblasts, BMSCs, and
ASCs). All grafts retained their collagen architecture.
The cells demonstrated a spindle-shaped morphology
that was similar to tenocytes located on both the
surface and the interior of the tendons (Fig. 9). The
inflammatory reaction was similar to the untreated
tendon autograft controls (control group 1).
All control specimens demonstrated a moderate
inflammatory response surrounding the graft sim-
ilar to the cell-seeded tendons. There was no dif-
ference between the inflammatory reactions in the
different control groups. The inflammatory re-
sponse subsided from week 4 to 8. The inflamma-
602 The Journal of Hand Surgery / Vol. 32A No. 5 May–June 2007
Figure 5. Cell proliferation curves (as measured by hemacytometry) of all cell types.
tory infiltrate was limited to the surface of the
tendon graft and did not penetrate the collagen
framework. Control group 2 was designed to as-
sess the immune reaction when tendon allografts
(untreated tendons) were implanted into outbred
animals. Tendon architecture was retained and
tenocytes appeared viable. Control group 3 con-
Figure 6. ␤-galactosidase staining at low passage. Roughly 1
in 100 cells stain positively, indicating senescence. (A) Ep-
itenon tenocytes, (B) sheath fibroblasts, (C) BMSCs, (D) ASCs.
Staining solution: 1 mg/ml X-gal, 40 nM citric acid, sodium
phosphate buffer at pH 6.0, 5 mM Potassium ferrous cyanide,
5 mM potassium ferric cyanide, 150 mM NaCl, 2 Mm Mag-
nesium chloride. 100⫻
sisted of acellular scaffolds that were not reseeded.
This group demonstrated a similar inflammatory
reaction to that of groups 1 and 2 and showed no
penetration of cells into the dense collagen frame-
work at either 4 or 8 weeks (Fig. 10).
Figure 7. ␤-galactosidase staining at high passage. Roughly 1
in 100 cells stain positively (indicated by black arrow), indi-
cating senescence; which is unchanged from passage 4. (A)
Epitenon tenocytes, (B) sheath fibroblasts, (C) BMSCs, (D)
ASCs. Staining solution: 1 mg/ml X-gal, 40 nM citric acid,
sodium phosphate buffer at pH 6.0, 5 mM Potassium ferrous
cyanide, 5 mM potassium ferric cyanide, 150 mM NaCl, 2
Mm Magnesium chloride. 100⫻
 Kryger et al / Tenocytes and Stem Cells for Tendon Engineering
Figure 8. Reseeded scaffolds after 1 week in vitro. Note
clumping of cells on surface of scaffold without penetration
into center of scaffold. Collagen framework is preserved. (A)
Epitenon tenocytes, (B) sheath fibroblasts, (C) BMSCs, (D)
ASCs. Arrows indicate scaffold surface with attached cells.
Hematoxylin and Eosin, 40⫻.
Discussion
Our results show that epitenon tenocytes, tendon
sheath cells, bone marrow and adipoderived stem
cells have similar growth characteristics and can be
used to successfully reseed acellularized tendon
grafts. Constructs using the four cell types were
successfully implanted in vivo and showed viability
at 6 weeks. This contrasts with the control group of
acellularized tendons where the construct remained
unpopulated by cells at 4 and 8 weeks.
A variety of scaffolds and cells have been used in
Figure 9. Bioartificial tendons after 6 weeks in vivo. Note
preservation of collagen framework, single epitenon-like
layer on surface (individual cells indicated by gray arrows),
and distribution of endotenon-like cells in center of grafts
(individual cells indicated by black arrows). (A) Epitenon
tenocytes, (B) sheath fibroblasts, (C) BMSCs, (D) ASCs. He-
matoxylin and Eosin, 40⫻.
603
Figure 10. Controls. (A) Acellular control not reseeded with
any cell type. Note inflammatory infiltrate and lack of penetra-
tion of cells into center of tendon. (B) Tendon autograft was
performed to assess the inflammatory reaction due to surgery
alone. (C) Tendon allograft from an unrelated animal demon-
strates the same degree of inflammation as seen in autogenous
tendon. Example area of inflammation is demonstrated by an
arrow in (B) and (C). Hematoxylin and Eosin, 100⫻.
tendon tissue engineering. The two cell types most
common used are fibroblasts (usually from teno-
cytes) and stem cells of mesenchymal origin.6 – 8 One
study suggests that in vitro-cultured rabbit tenocytes
continue to grow without signs of senescence.18 In
addition, tendon sheath fibroblasts have been shown
to proliferate faster than tenocytes and have been
used in tissue engineering applications.12,19 Bone
marrow-derived mesenchymal stem cells and ASCs
isolated using the techniques described have been
shown to have multipotency (ie, they can give rise to
various tissues of mesenchymal origin). Mesenchy-
mal stem cells have been maintained to high passage
in vitro and retain their proliferative and multipotent
abilities.14 –16,20 When seeded into tendon constructs
and exposed to the appropriate environment and me-
chanical forces, the cells may be driven toward teno-
cyte differentiation.4,21 These stem cells also produce
collagen, as shown in our immunocytochemistry
analysis and the work of others,22 suggesting that
they would contribute to tendon matrix remodeling in
vivo. However, there has been little prior data di-
rectly comparing important characteristics such as
growth potential, senescence, and collagen produc-
tion between the candidate cell types.
In this study, all four cell types continued to pro-
liferate and did not undergo senescence. Adipo-
derived mesenchymal stem cells showed higher pro-
liferation at later passage when compared with
epitenon tenocytes. The low proliferation rate of the
differentiated epitenon tenocyte suggests that ex vivo
604 The Journal of Hand Surgery / Vol. 32A No. 5 May–June 2007
expansion using this cell line will be slower. In
addition, given that it is easier to harvest large
amounts of cells from bone marrow or fat, BMSCs
and ASCs have a practical advantage compared with
epitenon tenocytes and sheath fibroblasts.
Scaffold materials used have ranged from collagen
gels to bioabsorbable synthetic material like poly-
lactide-co-glycolide. There is good evidence that
cell-seeded constructs have better material properties
compared with unseeded scaffolds.6 – 8 Nevertheless,
these constructs are inferior compared with native
tendon. This has major implications on the ability to
perform rehabilitation and ultimate functional out-
come after tendon grafting clinically.
Gross histologic observation of the reseeded ten-
dons showed no differences in inflammation between
the seeded constructs and autologous tendon con-
trols. This supports our expectation that acellularized
tendon reseeded with autologous cells will behave
immunologically like autologous tendon grafts. One
interesting finding in our in vivo experiment was the
finding of viable allogeneic tendon at 4 and 8 weeks.
There was no increased inflammation noted, and
viable tenocytes were seen. Although no blinding
was performed during the histologic assessment, our
findings are consistent with previously published lit-
erature demonstrating no increase in inflammatory
response with use of allogeneic cells.7,23,24 It is un-
clear whether the tenocytes seen are allogeneic teno-
cytes or infiltrated autogenic cells. Further work us-
ing X/Y fluorescent in situ hybridization will determine
the source of these tenocytes.
The rabbit model of flexor tendon repair was cho-
sen because the tendon has an intrasynovial compo-
nent like human flexor tendons and is large enough
for surgical repair. Nevertheless, it is possible that
differences exist in healing biology between rabbits
and humans such that the constructs will have to be
evaluated in a higher species after optimization in the
rabbit model. We were also unable to assess the
effect on early mobilization on tendon graft viability
and inflammation because the tendon was divided
proximally to unload and protect the implanted ten-
don graft. Although not quantitative, immunohisto-
chemistry for collagen demonstrated that the various
cell lines could produce collagen. This suggests the
potential usefulness of these cells in tendon tissue
engineering applications.
In vitro reseeding of tendon constructs showed
good surface attachment of cells but no penetration
of cells into the center of the tendon scaffold. In the
in vivo part of the study, we showed that cell-seeded
constructs demonstrate good repopulation in the cen-
ter of the scaffold within 6 weeks of implantation.
These findings in vivo suggest that the lack of pen-
etration in vitro may not be an issue for clinical
applications.
In summary, these four cell lines— epitenon teno-
cytes, sheath fibroblasts, BMSCs, and ASCs—are all
possible candidates for use in flexor tendon tissue
engineering. These cells, combined with scaffolds
such as acellularized tendon, offer the possibility of
expanding the source of tendon grafts for hand re-
construction. Further research will focus on testing
the biomechanical properties and inhibiting adhesion
formation in these novel constructs.
Received for publication October 18, 2006; accepted in revised form
February 23, 2007.
Part of this work was presented at the Adrian E. Flatt Residents and
Fellows Conference 2005 and awarded the Joseph H. Boyes award.
No benefits in any form have been received or will be received from
a commercial party related directly or indirectly to the subject of this
article.
Supported by VA Merit Review grants, American Society for Surgery
of the Hand, American Association of Hand Surgery, and the Thuss
Family Grant.
Corresponding author: James Chang, MD, Division of Plastic Surgery,
770 Welch Road, Suite 400, Palo Alto, CA 94304; e-mail: changhand@
aol.com.
Copyright © 2007 by the American Society for Surgery of the Hand
0363-5023/07/32A05-0002$32.00/0
doi:10.1016/j.jhsa.2007.02.018
References
1. White WL. Tendon grafts: a consideration of their source,
procurement and suitability. Surg Clin North Am 1960;40:
403– 413.
2. Leversedge FJ, Zelouf D, Williams C, Gelberman RH, Seiler
JG III. Flexor tendon grafting to the hand: an assessment of
the intrasynovial donor tendon-A preliminary single-cohort
study. J Hand Surg 2000;25A:721–730.
3. Seiler JG III, Chu CR, Amiel D, Woo SL, Gelberman RH.
The Marshall R. Urist Young Investigator Award. Autoge-
nous flexor tendon grafts. Biologic mechanisms for incor-
poration. Clin Orthop Relat Res 1997;345:239 –247.
4. Zhang AY, Chang J. Tissue engineering of flexor tendons.
Clin Plast Surg 2003;30:565–572.
5. Sharma B, Elisseeff JH. Engineering structurally organized
cartilage and bone tissues. Ann Biomed Eng 2004;32:148 –
159.
6. Cao Y, Liu Y, Liu W, Shan Q, Buonocore SD, Cui L.
Bridging tendon defects using autologous tenocyte engi-
neered tendon in a hen model. Plast Reconstr Surg 2002;
110:1280 –1289.
7. Ouyang HW, Goh JC, Mo XM, Teoh SH, Lee EH. The
efficacy of bone marrow stromal cell-seeded knitted PLGA
fiber scaffold for Achilles tendon repair. Ann N Y Acad Sci
2002;961:126 –129.
8. Young RG, Butler DL, Weber W, Caplan AI, Gordon SL,
Fink DJ. Use of mesenchymal stem cells in a collagen matrix
for Achilles tendon repair. J Orthop Res 1998;16:406 – 413.
 Kryger et al / Tenocytes and Stem Cells for Tendon Engineering
9. Chong AK, Chang J. Tissue engineering for the hand
surgeon: a clinical perspective. J Hand Surg 2006;31A:349 –
358.
10. Bader A, Schilling T, Teebken OE, Brandes G, Herden T,
Steinhoff G, et al. Tissue engineering of heart valves—
human endothelial cell seeding of detergent acellularized
porcine valves. Eur J Cardiothorac Surg 1998;14:279 –284.
11. Livesey SA, Herndon DN, Hollyoak MA, Atkinson YH, Nag
A. Transplanted acellular allograft dermal matrix. Potential
as a template for the reconstruction of viable dermis. Trans-
plantation 1995;60:1–9.
12. Klein MB, Pham H, Yalamanchi N, Chang J. Flexor tendon
wound healing in vitro: the effect of lactate on tendon cell
proliferation and collagen production. J Hand Surg [Am]
2001;26A:847– 854.
13. Colter DC, Class R, DiGirolamo CM, Prockop DJ. Rapid
expansion of recycling stem cells in cultures of plastic-
adherent cells from human bone marrow. Proc Natl Acad Sci
U S A 2000;97:3213–3218.
14. Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas
R, Mosca JD, et al. Multilineage potential of adult human
mesenchymal stem cells. Science 1999;284:143–147.
15. De Ugarte DA, Morizono K, Elbarbary A, Alfonso Z, Zuk
PA, Zhu M, et al. Comparison of multi-lineage cells from
human adipose tissue and bone marrow. Cells Tissues Or-
gans 2003;174:101–109.
16. Zuk PA, Zhu M, Mizuno H, Huang J, Futrell JW, Katz AJ, et
605
al. Multilineage cells from human adipose tissue: implications
for cell-based therapies. Tissue Eng 2001;7:211–228.
17. Dimri GP, Lee X, Basile G, Acosta M, Scott G, Roskelley C,
et al. A biomarker that identifies senescent human cells in
culture and in aging skin in vivo. Proc Natl Acad Sci U S A
1995;92:9363–9367.
18. Bernard-Beaubois K, Hecquet C, Houcine O, Hayem G,
Adolphe M. Culture and characterization of juvenile rabbit
tenocytes. Cell Biol Toxicol 1997;13:103–113.
19. Khan U, Occleston NL, Khaw PT, McGrouther DA. Differ-
ences in proliferative rate and collagen lattice contraction
between endotenon and synovial fibroblasts. J Hand Surg
1998;23A:266 –273.
20. Kassem M, Kristiansen M, Abdallah BM. Mesenchymal
stem cells: cell biology and potential use in therapy. Basic
Clin Pharmacol Toxicol 2004;95:209 –214.
21. Caplan AI. The mesengenic process. Clin Plast Surg 1994;
21:429 – 435.
22. Ge Z, Goh JC, Lee EH. Selection of cell source for
ligament tissue engineering. Cell Transplant 2005;14:
573–583.
23. Chong AK, Ang AD, Goh JC, Hui JH, Lim AY, Lee EH, et
al. Bone marrow-derived mesenchymal stem cells influence
early tendon-healing in a rabbit achilles tendon model.
J Bone Joint Surg 2007;89A:74 – 81.
24. Ryan JM, Barry FP, Murphy JM, Mahon BP. Mesenchymal
stem cells avoid allogeneic rejection. J Inflamm (Lond)
2005;2:8.