Performance of polyoxymethylene plastic (POM) as a
component of a tissue engineering bioreactor
Kitsie J. Penick,1 Luis A. Solchaga,2 Jim A. Berilla,1 Jean F. Welter1
1
Department of Biology (Skeletal Research Center), Case Western Reserve University, Cleveland, Ohio
2
Department of Orthopaedics, Case Western Reserve University, Cleveland, Ohio
Received 26 January 2004; revised 27 December 2004; accepted 3 March 2005
Published online 28 July 2005 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jbm.a.30351
Abstract: Polyoxymethylene (POM, acetal homopolymer,
polyacetal), commercialized as Delrin威 by DuPont, is an engi-
neering resin with mechanical properties that make it useful
for the prototyping and manufacture of laboratory apparatus.
These properties include excellent, “metal-like,” machining
characteristics and dimensional stability, as well as thermal
stability, which allows steam sterilization. Historically, POM
has been used widely, including as a surgical implant material.
For these reasons, we have used this plastic as a media-wetted
component in a tissue-engineering bioreactor, with good re-
sults. However, a study by LaIuppa et al.5 suggested that POM
is unsuitable for use in a cell culture environment (LaIuppa et
al. J Biomed Mater Res 1997;36:347–359). POM is based on the
polymerization of formaldehyde, and, in addition, contains
stabilizers and/or fillers. All of these could potentially be re-
leased into the medium, e.g., as formaldehyde or other thermal
breakdown products, especially upon repeated autoclaving.
The cited report thus appeared plausible, although contrary to
our observations. In this study, we specifically assessed
whether media conditioned by long-term exposure to ma-
chined white POM had a negative effect on the proliferation
and chondrogenic differentiation of human mesenchymal stem
INTRODUCTION
In our cartilage tissue-engineering studies, we re-
cently adopted a perfusion bioreactor system in which
we grow cell-scaffold composite constructs.1–3 These
constructs are based on human bone marrow– derived
mesenchymal stem cells (MSCs), which are induced to
differentiate into cartilage by the use of a chondro-
genic medium. Several media-wetted components in
this bioreactor (Fig. 1) are fashioned from the poly-
Correspondence to: Jean F. Welter, MD, PhD, Department of
Biology, Case Western Reserve University, 2080 Adelbert
Road, Millis Science Center, Room 113D, Cleveland, OH
44106; e-mail: jean.welter@case.edu
Contract grant sponsor: Case Western Reserve Univer-
sity/The Ohio Board of Regents
Contract grant sponsor: Arthritis Foundation
© 2005 Wiley Periodicals, Inc.
cells (MSCs). We selected this cell system, as cartilage tissue
engineering is the primary application of our bioreactor sys-
tem. The POM samples were steam-autoclaved 1 to 20 times, to
assess the possibility of any toxic thermal breakdown product
release into the media. We found that MSCs did not attach
directly to machined POM. Because cells that escape from the
tissue construct cannot colonize the reactor and compete for
nutrients, this is a desirable characteristic of a material used in
a tissue-engineering bioreactor. Furthermore, the use of POM-
conditioned media had no detectable impact on the prolifera-
tion rate of MSCs measured over a one-week period; nor was
any effect on chondrogenic differentiation observed at up to 3
weeks in culture. In summary, the use of POM as a culture
medium-wetted component appears to be innocuous, at least
for human MSCs. The contrast of these findings to those of
LaIuppa et al.5 may reflect a cell-type specific sensitivity, or
may be due to different handling of the material. © 2005 Wiley
Periodicals, Inc. J Biomed Mater Res 75A: 168 –174, 2005
Key words: polyacetal; delrin; mesenchymal stem cells;
chondrocytes
ether plastic polyoxymethylene homopolymer (POM,
acetal homopolymer, polyacetal, Delrin威; Delrin is a
trademark of E. I. DuPont de Nemours & Co). POM
was introduced in 1959 by DuPont (Wilmington, DE)
under the brand Delrin威; it is an engineering thermo-
plastic based on chain-growth polymerization of
formaldehyde (H2C⫽O). As the polymer itself is un-
stable, and reverts to the monomer upon heating to
120°C, the commercial product is reacted with acetic
anhydride to cap the ends of the polymer chains with
acetate groups.
We adopted this material because of its exceptional
balance of machining characteristics, tensile properties
(76 MPa ultimate at 23°C), shear strength, stiffness,
toughness, and thermal properties, but also because
POM has a history of use as a biocompatible material.
POM can be drilled and tapped and holds threads
well, has good impact strength, low friction, and it is
a good electrical insulator. It is reasonably wear-resis-
USE OF POM IN A TISSUE-ENGINEERING BIOREACTOR
Figure 1. Bioreactor chamber, partially disassembled. The white section (arrow) is machined from white POM sheet
material, and is in direct contact with the bioreactor contents. [Color figure can be viewed in the online issue, which is
available at www.interscience.wiley.com.]
tant, dimensionally stable, and chemically resistant to
solvents. Its melting point is 175°C, well above usual
steam autoclave temperatures; it is resistant to inter-
mittent high temperatures, and it retains its shape and
physical integrity at elevated temperatures. White
POM has negligible porosity, low moisture absorption
(0.9% at saturation), and is deemed safe for food-
contact use (FDA 21CFR177.2480).4 The low moisture
absorption also results in excellent dimensional stabil-
ity for close-tolerance machined parts in an aqueous
environment. Taken together, these characteristics
make it an excellent material from which to prototype
and fabricate small laboratory apparatus.
Recently, LaIuppa et al. studied the effects of a wide
range of plastic materials on the in vitro expansion of
hematopoietic progenitor cells.5 They concluded that
POM inhibited both colony formation and growth of
hematopoietic stem cells, whether it was in direct
contact with the cells or in indirect contact by condi-
tioning the media. They suggested leaching of toxic
compounds from the plastic material as the likely
mechanism for these effects. This is a reasonable in-
terpretation, as many polymeric materials may release
volatile substances that either have been trapped in
the polymeric matrix or can be formed as a result of
slow decomposition of the material.
However, POM also has a long history of use in
animal studies and as a long-term implant material in
a variety of medical applications. These include car-
169
diac valve prostheses (e.g., the Björk-Shiley valve tilt-
ing-disk style prosthesis6,7), dental implants8 and
prosthetics (Dental-D), and orthopedic implants9 –11 or
joint replacement components.9 POM performed well
in these roles, except as a load-bearing material in total
joint replacement where it exhibited excessive wear.12
In general, its use has decreased, and it has been
replaced by materials with better long-term wear
properties. POM has also been used in cell culture
environments,13,14 and we have been using this mate-
rial in our laboratories in a perfusion bioreactor sys-
tem for MSC-based cartilage tissue engineering re-
search (Fig. 1) for some time without noticing any
adverse effects. Previous studies have shown that
MSCs are exquisitely sensitive to small variations in
media composition.15 The present, brief study was
performed because, if an effect similar to that observed
by LaIuppa et al.5 was occurring in our system, we might
expect a significant improvement in the tissue-engi-
neered constructs by eliminating the POM components.
MATERIALS AND METHODS
Materials
White POM (Delrin威 trademark) was obtained as
12⬙⫻12⬙⫻41⬙ sheet material or as 41⬙ rods from a hardware
170
Figure 2. POM samples used for media-conditioning ex-
periments. [Color figure can be viewed in the online issue,
which is available at www.interscience.wiley.com.]
retailer (McMaster-Carr, Cleveland, OH). Pieces 22.9 mm
long were cut from the rod stock using a lathe (Fig. 2). These
weigh 1 g each, with a surface area of 520 mm2. For attach-
ment assays, a 25.4⫻76.2 mm (1⫻3⬙) well was milled into a
block of POM. All surfaces were machined with solid car-
bide cutters lubricated with ethanol. There was no detect-
able wear on the cutters, nor was there any detectable car-
bide contamination on the surfaces after machining.
Our standard cleaning protocol for the preparation of
machined parts for cell culture applications was applied as
follows: the material was scrubbed with a detergent solution
to remove machining residue, abundantly rinsed with dis-
tilled water, sonicated in 100% ethanol for 30 min, rinsed
with distilled water, dried, placed in a Pyrex glass beaker,
and then steam sterilized at 121°C for 30 min.
Cell culture media
Growth medium was Dulbecco’s Modified Eagle’s Me-
dium (DMEM) with 1 g/L glucose supplemented with 10%
fetal bovine serum (FBS, Invitrogen, Carlsbad, CA).
Chondrogenic medium was essentially as described by
Johnstone et al.16: DMEM with 4.5 g/L glucose supple-
mented with 1% ITS⫹Premix™ (BD Biosciences, San Jose,
CA), 100 ␮M ascorbate-2-phosphate (Wako Chemicals, Rich-
mond, VA), sodium pyruvate (Invitrogen), 10-7 M dexa-
methasone (Sigma Chemical Co, St. Louis, MO), and 10
ng/mL TGF-␤1 (R&D, Minneapolis, MN).
Repeated autoclaving
Tissue culture implements must be sterilized. Because it
seemed plausible that repeated autoclaving could either de-
plete the bulk material of any leachable volatile compounds,
PENICK ET AL.
or conversely might degrade the bulk material, increasing
the amount of leachables, we tested whether autoclaving the
samples repeatedly had an effect on our cell cultures. The
POM samples were subjected to 1, 5, 10, 15, or 20 autoclav-
ing cycles over a 5-day period. Between individual autoclav-
ing cycles (but not after the final one), the pieces were
allowed to cool and were rinsed with water.
Media conditioning
Aliquots of cell culture medium were POM-conditioned
as follows. Three autoclaved pieces of POM, for a total mass
of 3 g, and a total surface area of ⬃1500 mm2, were placed in
T-75 tissue culture flasks and soaked in 100 mL of complete
growth or chondrogenic cell culture medium for 10 days at
37°C in a cell culture incubator on a rocking platform. Con-
trol medium was incubated in parallel, but without the
addition of POM fragments.
Cells
Because the primary application of our bioreactor is stem
cell– based cartilage tissue engineering, human bone marrow–
derived MSCs were used in these experiments. These cells
have previously been shown to be exquisitely sensitive to
medium composition,15,17 and it was felt that they would pro-
vide the most relevant results for our study. The MSCs were
isolated as described by Haynesworth et al.18 Briefly, 10 –15 mL
of human bone marrow was harvested from the posterior iliac
crest of healthy adult volunteer donors through a Jamshidi
bone marrow needle. The marrow aspiration procedure was
reviewed and approved by the University Hospitals of Cleve-
land Institutional Review Board; informed consent was ob-
tained from all donors. The marrow sample was combined
with growth medium and centrifuged at 500 ⫻ g for 5 min. The
resulting pellet was resuspended in 5 mL of growth medium,
and layered over 35 mL of 63% (v/v) Percoll in Tyrode’s salt
solution (Sigma Chemical Co), adjusted to a final NaCl con-
centration of 0.1 M. After centrifugation at 460g, the top 25% of
the gradient (pooled density 1.03 g/mL) was transferred to a
50-mL tube. The volume was increased to 50 mL with growth
medium, the tube was centrifuged at 500g, and the resulting
pellet was resuspended in 7 mL of growth medium. These cells
were seeded onto 10-cm-diameter plates at 1.8 ⫻ 105 nucleated
cells/cm2. After 4 days, non-adherent cells were removed by
changing the medium, leaving only the adherent cells. The
medium was changed every 3 days thereafter. After 2 weeks in
primary culture, and prior to reaching confluence, the cells
were passaged using trypsin, and seeded at 5 ⫻ 103 cells/cm2.
Subsequently the cells were passaged at 70% confluence.
Assays
Attachment
Cells were plated directly into a well milled in a POM
block, or onto a glass microscope slide placed at the bottom
USE OF POM IN A TISSUE-ENGINEERING BIOREACTOR
of the well. The glass slide was autoclaved after cleaning by
immersion for 1 h in concentrated nitric acid followed by a
1-h wash in running deionized water. Twenty-four hours
later, the well and/or the slide were rinsed gently with PBS
and adherent cells were ethanol-fixed, stained with crystal
violet, and counted.
Proliferative potential
Passaged cells were plated at 80,000 cells per 60-mm di-
ameter dish, in 3 mL of medium. Twenty-four hours after
plating, the medium was replaced with fresh growth me-
dium consisting of a 1:1 mix of fresh and POM-conditioned
medium. Growth curves were determined by harvesting a
subset of the cells on each subsequent day and counting
them using a hemacytometer. A minimum of 300 cells were
counted for each of three triplicate samples in each treat-
ment group at each time point.
Chondrogenic potential
Chondrogenic potential was assayed by pellet culture as
described by Ballock and Reddi.19 Human MSCs have been
shown to undergo chondrogenic differentiation in vitro in
this culture system,16,20,21 and this is a documented assay for
the chondrogenic potential of cell populations.15,22–24 Pas-
saged cells were resuspended in a 1:1 mixture of fresh and
POM-conditioned chondrogenic media, and 0.5-mL aliquots
containing 250,000 cells were centrifuged at 500g in 15 mL
polypropylene conical tubes (Falcon, BD Biosciences).19 The
pelleted cells were then incubated for up to 3 weeks, with 3
weekly medium changes.
Chondrogenic differentiation of MSCs is extremely sensi-
tive to environmental conditions and disruption of this pro-
cess is immediately apparent upon histological evaluation.
We, therefore, performed a series of standard histological
and immunohistochemical assays on the pellets, including
Toluidine blue staining for GAG content and immunohisto-
chemistry for collagens type I and II. Primary antibodies
were from Sigma Chemical Co. for the anti-Type I and from
the Developmental Studies Hybridoma Bank for the anti-
Type II; a FITC-labeled secondary (Chemicon, Temecula,
CA) was used. The sections were photographed at 10⫻
magnification using a SpotRT camera (Diagnostic Instru-
ments, Sterling Heights, MI), and mosaic composites were
assembled using Adobe Photoshop.25
RESULTS
POM is made by the polymerization of formalde-
hyde. It, therefore, seems possible that residual form-
aldehyde might be leaching from the material into the
cell culture medium. Commercial POM (Delrin威) con-
tains 97% POM polymer, 3% stabilizer, and less than
0.005% free formaldehyde, according to the relevant
DuPont material safety data sheet. Excess heat is
known to cause thermal breakdown of POM; products
171
include formaldehyde, carbon monoxide, and carbon
dioxide. Because of the toxicity of the first two com-
pounds, it would, therefore, appear advisable to avoid
overheating the material. In our application, this
proved not to be an issue. With care, sharp tools, and
suitable lubrication/cooling, intricate machining can
be performed without overheating the material. The
melting (167–177°C) and thermal breakdown (230°C)
temperatures are also well above normal steam auto-
claving temperatures (121°C).
Effects of POM-conditioned medium on monolayer
MSCs
Attachment
Human MSCs did not attach to machined POM sur-
faces at all, even after 24 h. They did, however, within
6 h, adhere to a glass slide placed at the bottom of the
same well with the same plating efficiency as on a similar
slide placed in a tissue culture plastic dish (not shown).
Proliferation
As shown in Figure 3, the proliferative potential of
MSCs and chondrocytes was not adversely affected by
culturing in medium conditioned by long-term expo-
sure to machined POM surfaces. Subjecting the POM
samples to up to 20 autoclaving cycles prior to condi-
tioning the medium had no effect either.
Effects of POM-conditioned medium on MSC
chondrogenic potential
There was no difference in the timing of pellet for-
mation, which occurred between 12 and 24 h, between
cells grown in control or POM-conditioned medium.
As evaluated by collagen immunoreactivity and pro-
teoglycan expression patterns at 2 and 3 weeks, there
was no perceptible difference in the chondrogenic dif-
ferentiation of human MSCs whether grown in control
or POM-conditioned medium (Figs. 4, 5). If the differen-
tiation program was disrupted, proteoglycan and type II
collagen synthesis, both markers of chondrogenic differ-
entiation, would be sharply diminished or absent, while
type I collagen synthesis would be expected.
DISCUSSION
POM is generally viewed as a fairly innocuous ma-
terial. It is used in the food industry, and has a con-
172 PENICK ET AL.

Figure 3. Growth curves for human MSCs. The cells were grown in the presence of media conditioned (see text) in either
tissue culture plastic alone (TCP) or by exposure to POM rods autoclaved from 1 to 20⫻. [Color figure can be viewed in the
online issue, which is available at www.interscience.wiley.com.]

siderable history of use as a biocompatible implant ma- mately led to the withdrawal of these prostheses, but the
terial. It was used as a component of tilting disk cardiac failures occurred in the metal components of the im-
valve prostheses with considerable success,6 surviving in plant, and were not related to the POM material. It has
at least one case for 23 years.26 High failure rates ulti- also been used, albeit with poor results, as an orthopae-

Figure 4. Human MSC aggregate cultures grown in control medium or in the presence of medium conditioned with POM
rods as described in Materials and Methods. Toluidine Blue O Stain. [Color figure can be viewed in the online issue, which
is available at www.interscience.wiley.com.]
USE OF POM IN A TISSUE-ENGINEERING BIOREACTOR 173

Figure 5. Immunohistochemical staining for Collagen types I and II on aggregate cultures grown under control conditions
or in the presence of media conditioned by POM rods. Primary antibodies were from Sigma Chemical Co. for the anti-Type
I and from the Developmental Studies Hybridoma Bank for the anti-Type II, FITC-labeled secondary (Chemicon). [Color
figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

dic implant material: as a weight-bearing material, POM
components wore and failed at an unacceptable rate.12 It
has, thus, been completely replaced in this role by ma-
terials with better wear characteristics.
A study by LaIuppa et al. suggested that POM had
a significant negative effect on cell growth and colony-
forming ability of hematopoietic stem cells, whether it
was used as a growth substrate or as a medium-wetted
component.5 The present study was undertaken be-
cause, although we use a different cell population (hu-
man bone marrow– derived MSCs), these findings were
worrisome as the use of POM could be negatively affect-
ing cell proliferation and differentiation.
We, however, failed to detect any impact on the
proliferation or chondrogenic differentiation of hu-
man MSCs by media exposed to POM for extended
periods. Chondrogenic differentiation of MSCs is ex-
tremely sensitive to environmental conditions and
disruption of this process is immediately apparent
upon histological evaluation. In addition, an earlier
version of our bioreactor system did not make use of
POM components, and we have observed no appar-
ent negative impact after implementing a new bio-
reactor design with culture medium-wetted POM
components. On the contrary, cells appeared to sur-
vive and differentiate better in the newer chambers.
Finally, in addition to testing the POM material with
human bone marrow– derived MSCs, we have per-
formed similar experiments with human articular
chondrocytes (not shown) with similar results, i.e.,
no adverse effects were observed with either of
these cell types. Thus, although previous studies
have shown that MSCs, and their chondrogenic dif-
ferentiation, are exquisitely sensitive to small vari-
ations in media composition,15 POM was essentially
inert in our system.
We do not feel that the lack of attachment of our
cells to raw POM surfaces necessarily indicates a toxic
effect, as many plastics must be surface treated, usu-
ally by electron beam, to encourage cell attachment. In
the case of our bioreactor chamber, the fact that cells
do not adhere to POM directly is indeed an unantici-
pated advantage, as it does not permit cells that have
fallen out of the tissue-engineered constructs to attach
to the walls of the bioreactor chamber, proliferate, and
compete for nutrients.
CONCLUSIONS
The use of POM material as a culture-medium wet-
ted component appears to be innocuous, at least with
respect to human MSC-based tissue engineering, even
after machining and repeated autoclaving. The con-
trast of these observations to those of LaIuppa et al.5
may indicate a cell-type-specific sensitivity to POM or
its breakdown products, or may be due to different
handling of the material.
174
This study was funded in part by grants from Case West-
ern Reserve University/The Ohio Board of Regents (J.F.W./
L.A.S.), and the Arthritis Foundation (J.F.W.).
References
1. Baskaran H, Solchaga L, Penick K, Berilla J, Welter J. Substrate
mass transport in tissue engineered cartilage. In: Hoffner L,
editor. Proceedings of the Arthritis Research Conference; Ar-
thritis Foundation, Atlanta, GA, 2003. p 2.32.
2. Solchaga L, Tognana E, Goldberg V, Caplan A, Welter J. Tissue
engineered cartilage implants: in vitro chondrogenic pre-con-
ditioning. Proceedings of the 4th Symposium of the Interna-
tional Cartilage Repair Society 2003:271.
3. Welter J, Penick K, Berilla J, Caplan A, Goldberg VM, Solchaga
L. MSC-based Tissue engineered cartilage: Effects of construct
assembly method and perfusion rates. In: Hoffner L, editor.
Proceedings of the Arthritis Research Conference; Arthritis
Foundation, Altanta, GA 2003. p 2.33.
4. USFDA. 21CFR177.2480 Polyoxymethylene homopolymer.
Code Fed Regul 2002;3:331–332.
5. LaIuppa JA, McAdams TA, Papoutsakis ET, Miller WM. Cul-
ture materials affect ex vivo expansion of hematopoietic pro-
genitor cells. J Biomed Mater Res 1997;36:347–359.
6. Björk VO. Delrin as implant material for valve occluders.
Scand J Thorac Cardiovasc Surg 1972;6:103–107.
7. Björk VO. A new tilting disc valve prosthesis. Scand J Thorac
Cardiovasc Surg 1969;3:1–10.
8. MacAfee KA, Quinn PD. Total temporomandibular joint re-
construction with a Delrin titanium implant. J Craniofac Surg
1992;3:160 –169.
9. Fister JS, Memoli VA, Galante JO, Rostoker W, Urban RM.
Biocompatibility of Delrin 150: a creep-resistant polymer for
total joint prostheses. J Biomed Mater Res 1985;19:519 –533.
10. Brown SA, Mayor MB. The biocompatibility of materials for
internal fixation of fractures. J Biomed Mater Res 1978;12:67– 82.
11. Buch F, Albrektsson T, Herbst E. Effects of direct currents on
bone growth into delrin implants. Scand J Plast Reconstr Surg
1985;19:223–230.
12. Havelin LI, Gjerdet NR, Lunde OD, Rait M, Sudmann E. Wear
of the Christiansen hip prosthesis. Acta Orthopaed Scand 1986;
57:419 – 422.
13. Agnihotri N, Kisaalita WS, Keith CH. Micro-perfusion flow cell
for imaging cultured cells. Biotechniques 1999;27:722–726, 728.
14. Abdullah LH, Davis SW, Burch L, Yamauchi M, Randell SH,
Nettesheim P, Davis CW. P2u purinoceptor regulation of mu-
PENICK ET AL.
cin secretion in SPOC1 cells, a goblet cell line from the airways.
Biochem J 1996;316:943–951.
15. Lennon DP, Haynesworth SE, Young RG, Dennis JE, Caplan AI.
A chemically defined medium supports in vitro proliferation and
maintains the osteochondral potential of rat marrow-derived
mesenchymal stem cells. Exp Cell Res 1995;219:211–222.
16. Johnstone B, Hering TM, Caplan AI, Goldberg VM, Yoo JU. In
vitro chondrogenesis of bone marrow-derived mesenchymal
progenitor cells. Exp Cell Res 1998;238:265–272.
17. Lennon DP, Haynesworth SE, Bruder SP, Jaiswal N, Caplan AI.
Human and animal mesenchymal progenitor cells from bone
marrow: Identification of serum for optimal selection and pro-
liferation. In Vitro Cell Dev Biol 1996;32:602– 611.
18. Haynesworth SE, Goshima J, Goldberg VM, Caplan AI. Char-
acterization of cells with osteogenic potential from human
marrow. Bone 1992;13:81– 88.
19. Ballock RT, Reddi AH. Thyroxine is the serum factor that
regulates morphogenesis of columnar cartilage from isolated
chondrocytes in chemically defined medium. J Cell Biol 1994;
126:1311–1318.
20. Yoo JU, Barthel TS, Nishimura K, Solchaga L, Caplan AI,
Goldberg VM, Johnstone B. The chondrogenic potential of
human bone-marrow-derived mesenchymal progenitor cells.
J Bone Joint Surg Am 1998;80:1745–1757.
21. Allay JA, Dennis JE, Haynesworth SE, Majumdar MK, Clapp
DW, Shultz LD, Caplan AI, Gerson SL. LacZ and interleukin-3
expression in vivo after retroviral transduction of marrow-
derived osteogenic mesenchymal progenitors. Hum Gene Ther
1997;8:1417–1427.
22. Lennon DP, Haynesworth SE, Arm DM, Baber MA, Caplan AI.
Dilution of human mesenchymal stem cells with dermal fibro-
blasts and the effects on in vitro and in vivo osteochondrogen-
esis. Dev Dyn 2000;219:50 – 62.
23. Tsutsumi S, Shimazu A, Miyazaki K, Pan H, Koike C, Yoshida E,
Takagishi K, Kato Y. Retention of multilineage differentiation
potential of mesenchymal cells during proliferation in response to
FGF. Biochem Biophys Res Commun 2001;288:413– 419.
24. Mastrogiacomo M, Cancedda R, Quarto R. Effect of different
growth factors on the chondrogenic potential of human bone
marrow stromal cells. Osteoarthritis and Cartilage 2001;
9(Suppl A):S36 – 40.
25. Solchaga L, Penick K, Welter J. A “manual” mosaicking ap-
proach to generating large, high-resolution digital images of
histological sections. Proc Royal Microsc Soc 2004;39:313–320.
26. Lijoi A, Parodi E, Passerone GC. An old Delrin tilting disc:
eroded but strong. Tex Heart Inst J 2001;28:232.