New Horizon Workshop: Scaffolds, Bioreactors
and Biologics in Orthopaedic Research
(Organized by the Tissue Engineering and
Regenerative Medicine International Society
(TERMIS) and ORS)
Organizers:
Dimitrios Zeugolis, PhD
Mauro Alini, PhD
Speakers:
Martin Stoddart, PhD
Christopher Evans, PhD
Alicia El Haj, PhD
Bioreactors for musculoskeletal research
Martin J. Stoddart, PhD.
AO Research Institute Davos, Davos Platz, Switzerland
Institute for Science and Technology in Medicine, Keele University, Stoke-on-Trent, UK
Abstract
Musculoskeletal research has made great advances over a number of years. The advent of molecular
biological techniques has allowed for detailed analysis and understanding of complex pathways.
Within the musculoskeletal system it is clear that the mechanical forces experienced by cells
contained within the tissue have a major influence on tissue homeostasis, pathogenesis and
regeneration. These forces are often lacking during in vitro or ex vivo culture and yet the interplay
between mechanical forces and other exogenous signals is likely to result in novel or synergistic
effects. This has led to the development of various bioreactors that are able to apply one or more
stimuli to the cultured tissue in order to further investigate these combined effects. Within this
presentation the influence of mechanical stimulation on the differentiation and maturation of
musculoskeletal tissues will be highlighted. The focus will be on cartilage and the behaviour of
chondroprogenitors and chondrocytes, however, bone and intervertebral disc will also be discussed.
Introduction
During kinematic motion cells experience a number of different mechanical stimuli, hydrostatic
pressure, tension, shear, compression or a combination of these. Maintenance of tissue homeostasis
requires a basal level of stimulation to maintain the health of the tissue, with long term unloaded
conditions such as bed rest known to have a detrimental effect. Regenerative medicine studies are
frequently performed in the absence of these stimuli potentially making clinically relevant
conclusions difficult. Incorporation of a bioreactor system into the study allows the composite effect
of physical and soluble stimuli to be established. In vitro studies frequently rely on the application of
exogenous growth factors. While this is a highly successful approach for mechanistic studies, it does
not investigate the endogenous source of these factors during normal healing.
Bone marrow derived mesenchymal stem cells (BMSCs) are frequently used as a source material for
cell based cartilage repair strategies. Whereas the articulating joint provides a unique, multiaxial
load environment, in vitro studies are classically performed under static conditions, or using uniaxial
load alone. Using a complex, multiaxial load bioreactor (Figure 1), we have demonstrated that
superficial shear, superimposed over uniaxial load, can provide a chondrogenic signal in the absence
of exogenous growth factors, namely TGF-β (Kupcsik et al., 2010; Li et al., 2010). This response is due
to an increase in the production of endogenous TGF-β by the mechanically stimulated cells. Crucially,
shear is major driver of the response observed (Schatti et al., 2011).
Using this device, we have also demonstrated that asymmetrical seeding of the construct, with a
greater percent of the total cells being deposited in the superficial zone, leads to increased cartilage
matrix deposition when using the same number of total cells (Gardner et al., 2016c). Deposition of
both glycosaminoglycan and collagen II are increased in asymmetrically seeded scaffolds when
compared to homogenously seeded scaffolds. This induced anisotropy is an interesting example of
naturally induced changes induced by physical loads. The increase in endogenous TGFb leads to an
increase in the latent form of the protein. Multiaxial load alone is capable of activating latent TGFβ
by removing the non-covalently bound latency associated peptide, a critical step in the functional
activity of endogenous TGFβ (Gardner et al., 2016b). This is possible even in the absence of cells.
Such results are clinically relevant, and yet they could not be obtained under static conditions. This
provides a new insight into mechanically induced chondrogenesis and offers an experimental test
bed for clinical therapies. It allows for the identification of novel markers and clinically relevant
targets that are only modified during articulation (Gardner et al., 2016a). In addition, the therapy in
its entirety, including the effect of the rehabilitation protocol, can be investigated using human
derived cells. This is leading to a new field of regenerative rehabilitation (Perez-Terzic and Childers,
2014).
The use of human cells and a more physiologically relevant loading environment, leads to more
clinically relevant studies being performed which should increase the potential translation into the
clinic.
Figure 1. Articular motion is a combination of complex multiaxial load (Left). This can be mimicked
using a ceramic hip ball bioreactor system (middle) that is able to recapitulate the articulating
motion (right).
References
Gardner OF, Fahy N, Alini M, Stoddart MJ (2016a) Differences in human mesenchymal stem
cell secretomes during chondrogenic induction. Eur Cell Mater 31: 221-235.
Gardner OF, Fahy N, Alini M, Stoddart MJ (2016b) Joint mimicking mechanical load activates
TGFbeta1 in fibrin-poly(ester-urethane) scaffolds seeded with mesenchymal stem cells. J Tissue Eng
Regen Med.
Gardner OFW, Musumeci G, Neumann AJ, Eglin D, Archer CW, Alini M, Stoddart MJ (2016c)
Asymmetrical seeding of MSCs into fibrin-poly(ester-urethane) scaffolds and its effect on
mechanically induced chondrogenesis. J Tissue Eng Regen Med In Press.
Kupcsik L, Stoddart MJ, Li Z, Benneker LM, Alini M (2010) Improving chondrogenesis:
potential and limitations of SOX9 gene transfer and mechanical stimulation for cartilage tissue
engineering. Tissue Eng Part A 16: 1845-1855.
Li Z, Kupcsik L, Yao SJ, Alini M, Stoddart MJ (2010) Mechanical Load Modulates
Chondrogenesis of Human Mesenchymal Stem Cells through the TGF-beta Pathway. J Cell Mol Med
14: 1338-1346.
Perez-Terzic C, Childers MK (2014) Regenerative rehabilitation: a new future? Am J Phys
Med Rehabil 93: S73-78.
Schatti O, Grad S, Goldhahn J, Salzmann G, Li Z, Alini M, Stoddart MJ (2011) A combination of
shear and dynamic compression leads to mechanically induced chondrogenesis of human
mesenchymal stem cells. Eur Cell Mater 22: 214-225.
Tissue engineering tools for orthopedic applications: Gene Therapy
C. H. Evans PhD
Rehabilitation Medicine Research Center, Mayo Clinic, Rochester, MN, USA.
Abstract
Gene transfer offers unique capabilities to regenerative medicine by delivering
morphogenetic stimuli to sites of tissue damage in a local and sustained fashion. In its
simplest form, the gene (usually a cDNA) encodes a product that stimulates cell division, cell
migration, differentiation or matrix production. Unlike their recombinant counterparts,
proteins synthesized under these conditions are likely to have undergone authentic posttranslational modification and, being nascent proteins, may have higher specific activity.
Gene transfer is also an advantage when delivering gene products with intra-cellular sites of
action, such as species of non-coding RNA and transcription factors. The use of inducible
promoters or other molecular control mechanisms offers opportunities for regulated gene
expression. Genes can be delivered by ex vivo or in vivo means using viral or non-viral
vectors. These principles have been applied successfully in animal models of cartilage repair,
the healing of bone, tendon, ligament, muscle and intervertebral disk degeneration. Clinical
translation is complicated by issues of safety, cost and long time lines. However, one
cartilage repair protocol has proceeded to a human clinical trial and a related product for
the gene therapy of osteoarthritis is under review for a biologics license application filed
with the Korean FDA.
Introduction
Among the challenges confronting the field of tissue engineering and regenerative medicine
(TERM) is the need to deliver morphogens and other such factors locally, in a sustained
fashion, to sites of healing. The use of scaffolds for this purpose has been widely explored,
with varying success; gene transfer offers an alternative approach1. The concept is quite
straightforward. Genes of interest or, more usually, their cDNA equivalents are delivered to
cells within the defect. Genes are transferred to cells by means of vectors, which may be
derived from viruses or be of non-viral origin. Unlike the treatment of genetic disease that
require life-time transgene expression, or periodic readministration, TERM will probably
only need transgene expression for a limited time, which is less demanding in terms of
vector properties. Vectors may be introduced directly into sites of tissue damage (in vivo
delivery), or introduced into cells outside the body for later implantation of the genetically
modified cells (ex vivo delivery). Both strategies have been explored experimentally in
orthopaedic contexts. In some in vivo protocols, vectors are associated with matrices to
form a “gene activated matrix” (GAM) prior to implantation. Although most experimental
studies use strong constitutive promoters to drive transgene expression, the use of
inducible promoters offers possibilities for regulating the level of gene expression and its
timing. It is also possible to deliver multiple genes and express them at different times,
although this has not been explored in the present context. Ex vivo gene delivery raises the
question of which cells to use. Although differentiated cells of the target tissue can be used
for this purpose, there is an increasing tendency to use progenitor cells. However, it remains
unclear whether they remain at the defect site and form new tissue directly, or whether
they serve as transient sources of gene product and then disappear.
Cartilage Repair
Two general strategies have emerged. One of these uses standard ex vivo gene delivery with
genetically modified chondrocytes or, more commonly, chondroprogenitor cells that are
implanted into defects in cartilage. A variety of transgenes have been explored, including
those encoding FGF-2, IGF-1, BMP-7, TGF- and the transcription factor Sox9 2,3. Adenovirus
and, more recently, adeno-associated virus (AAV) have been the most commonly used
vectors. A second approach seeks to improve the clinical outcome when using
microfracture. To this end, Madry and Cucchiarini 4 are developing technologies where AAV
vectors are applied directly to marrow as it emerges after penetration of the sub-chondral
plate. In a related strategy, marrow is recovered and mixed with vector as it clots. The
resulting “gene plug” is then press fit into full-thickness defects 5. This approach has shown
promise in rabbit models using adenovirus vectors encoding BMP-2 and Indian Hedgehog 6.
It has also been evaluated in chondral defects in sheep 7. Genetically modified, allograft
chondrocytes persist within defects in rabbit cartilage after implantation 8 and a clinical trial
using allograft human chondrocytes that express high levels of TGF- has been successfully
completed (ClinicalTrials.gov Identifier: NCT01825811).
Bone Healing
Using gene transfer to enhance bone healing has been the subject of considerable research
(reviewed in reference 9) that has utilized most of the strategies available to the field.
Considerable progress has been made with the approach of Lieberman’s group using the ex
vivo viral transduction of cells with a BMP-2 cDNA. Initially they used adenovirus to
transduce bone marrow stromal cells 10. This approach, however, is time-consuming and
would be expensive clinically because of the need to cultivate autologous cells under GMP
conditions. To expedite matters, their more recent work uses lentivirus to transduce freshly
isolated marrow buffy coat cells 11. Because of safety concerns with lentivirus, a “suicide
gene” can also be included to allow deletion of the genetically modified cells if necessary 12.
Another expedited approach seeks to transduce freshly harvested biopsies of muscle, fat or
marrow and then to immediately implant the genetically modified tissue into osseous
lesions 13. Results in rat models are promising, but data from large animals are lacking. A
different approach, which is an adaptation of the GAM strategy, coats allograft bone with
AAV vectors encoding osteogenic products. 14 This produces “allograft revitalization”,
leading to remodeling of the implant and its replacement with newly synthesized host bone.
So far, this has only been demonstrated in mice. There have, however, been a number of
studies using a variety of other approaches in large animal models that show promising
results 9.
Tendon, Ligament, Muscle and Intervertebral Disk
Repair of all of these tissues has been investigated from the point of view of gene therapy.
Most of the studies are preliminary but encouraging. With tendons and ligaments there is
much interest in using BMPs-12, -13, -14 and scleraxis, which have tenogenic properties 15.
Strategies for healing skeletal muscle include the delivery of genes encoding growth factors
such as IGF-1 in conjunction with an angiogenic stimulus, such as VEGF, while inhibiting scar
formation by blocking TGF-16. The intervertebral disc is interesting because resident cells
express transgenes for extended periods of time, even when the vector or transgene are
highly antigenic. This reflects the avascular nature of the disk and its dense matrix
surrounding cells in the nucleus pulposus. Kang and colleagues have been able to maintain
disk height in rabbit models of intervertebral disk degeneration by the intra-diskal injection
of AAV vectors 17.
Clinical Translation
Orthopaedic gene therapy has been discussed for over 20 years 18 without generating a
clinically approved product. In this it reflects the field of gene therapy as a whole, which is
only now developing drugs approved for clinical use. Clinical translation is difficult in the
TERM orthopaedic area because the target conditions are neither genetic nor lethal, leading
to heightened concerns about safety. Industry funding is sparse, because companies see
long-time lines and questionable return on investment. The field of gene therapy as a whole
has had major mood swings, but is now again on an upward trajectory which should provide
collateral benefit to TERM protocols in orthopaedics . One product (Invossa™) for treating
osteoarthritis, a related area, comprises an allograft chondrocyte line transduced to express
TGF-. It has completed Phase III trials in Korea (ClinicalTrials.gov Identifier:
NCT02072070), where an application for biologics license approval has been submitted to
the Korean FDA. This product is about to start Phase III trials in the US.
References
1.
Evans CH, Huard J. 2015. Gene therapy approaches to regenerating the musculoskeletal
system. Nat Rev Rheumatol 11:234-242.
2.
Goodrich LR, Hidaka C, Robbins PD, et al. 2007. Genetic modification of chondrocytes with
insulin-like growth factor-1 enhances cartilage healing in an equine model. J Bone Joint Surg Br
89:672-685.
3.
Ortved KF, Begum L, Mohammed HO, Nixon AJ. 2015. Implantation of rAAV5-IGF-I
transduced autologous chondrocytes improves cartilage repair in full-thickness defects in the equine
model. Mol Ther 23:363-373.
4.
Cucchiarini M, Orth P, Madry H. 2013. Direct rAAV SOX9 administration for durable articular
cartilage repair with delayed terminal differentiation and hypertrophy in vivo. J Mol Med (Berl)
91:625-636.
5.
Pascher A, Palmer GD, Steinert A, et al. 2004. Gene delivery to cartilage defects using
coagulated bone marrow aspirate. Gene Ther 11:133-141.
6.
Sieker JT, Kunz M, Weissenberger M, et al. 2015. Direct bone morphogenetic protein 2 and
Indian hedgehog gene transfer for articular cartilage repair using bone marrow coagulates.
Osteoarthritis Cartilage 23:433-442.
7.
Ivkovic A, Pascher A, Hudetz D, et al. 2010. Articular cartilage repair by genetically modified
bone marrow aspirate in sheep. Gene Ther 17:779-789.
8.
Kang R, Marui T, Ghivizzani SC, et al. 1997. Ex vivo gene transfer to chondrocytes in fullthickness articular cartilage defects: a feasibility study. Osteoarthritis Cartilage 5:139-143.
9.
Evans CH. 2010. Gene therapy for bone healing. Expert Rev Mol Med 12:e18.
10.
Lieberman JR, Daluiski A, Stevenson S, et al. 1999. The effect of regional gene therapy with
bone morphogenetic protein-2-producing bone-marrow cells on the repair of segmental femoral
defects in rats. J Bone Joint Surg Am 81:905-917.
11.
Virk MS, Sugiyama O, Park SH, et al. 2011. "Same day" ex-vivo regional gene therapy: a novel
strategy to enhance bone repair. Mol Ther 19:960-968.
12.
Alaee F, Sugiyama O, Virk MS, et al. 2014. Suicide gene approach using a dual-expression
lentiviral vector to enhance the safety of ex vivo gene therapy for bone repair. Gene Ther 21:139147.
13.
Evans CH, Liu FJ, Glatt V, et al. 2009. Use of genetically modified muscle and fat grafts to
repair defects in bone and cartilage. Eur Cell Mater 18:96-111.
14.
Ito H, Koefoed M, Tiyapatanaputi P, et al. 2005. Remodeling of cortical bone allografts
mediated by adherent rAAV-RANKL and VEGF gene therapy. Nat Med 11:291-297.
15.
Docheva D, Muller SA, Majewski M, Evans CH. 2015. Biologics for tendon repair. Adv Drug
Deliv Rev 84:222-239.
16.
Deasy BM, Huard J. 2002. Gene therapy and tissue engineering based on muscle-derived
stem cells. Curr Opin Mol Ther 4:382-389.
17.
Woods BI, Vo N, Sowa G, Kang JD. 2011. Gene therapy for intervertebral disk degeneration.
Orthop Clin North Am 42:563-574, ix.
18.
Evans CH, Robbins PD. 1995. Possible orthopaedic applications of gene therapy. J Bone Joint
Surg Am 77:1103-1114.
Bioscaffolds for Musculoskeletal Indications
Alicia El Haj 1, Molly Loundes2, Mike Rotherham1, Joshua Price1,2 and
Shukry Habib 2
1
2
Institute for Science and Technology in Medicine, Keele University, Keele, UK. Centre for Stem
Cells and Regenerative Medicine, King’s College London, London, UK.
A key unresolved element in existing musculoskeletal cell therapy approaches
is the formation of the normal 3D structural organisation of cartilage and the
integration with the surrounding cartilage and subchondral bone. This 3D structure
involves hierarchies of collagen families organised and tailored to provide the cell
niche. Mimicking these structures for regenerative medicine approaches has relied
on creating polymer or natural scaffolds which can represent the in vivo
characteristics. One example of our work is the use of collagen scaffolds which can
be processed to provide complex structures using different formulations ( Price et al
2015). These structures also include a complex cell community with
maintained progenitor cell populations as well as intermediary and mature
phenotypes present in cartilage and bone.
Tissue engineering can provide more physiologically relevant 3D models for
simulating this structure by combining biology with material chemistry which has the
potential to provide next generation therapeutic approaches to musculoskeletal
repair. What is lacking is the maintenance of the complex cell community which can
tailor a biologically relevant niche and capturing the biological signal processes
occurring during development and adult growth in a controlled manner.
We have been using tethered biological molecules on substrates such as
glass and polymers of varying degradation rates to guide cells in 3D models to
generate complex structures. In this presentation, we demonstrate this principle with
an example where we control the timing, location and level of Wnt signalling; a key
component which is highly regulated during embryonic development and for the
maintenance of adult tissues. The ability to provide a defined and directed source of
Wnt proteins is crucial to fully understanding its role in tissue development and to
mimic its activity in vivo. In addition, the Wnt signalling family is thought to be
involved in joint pathology and the pathways may be involved in the progression of
OA and bone degeneration (Lories et al 2013). In this workshop, our recent work as
part of the UK MRC UK Regenerative Medicine hub in Engineering the Stem Cell
Niche will be presented. We have investigated using the Wnt3A cue to maintain stem
cell populations alongside establishing a zonal architecture of differentiating
osteogenic cells within a 3D model (Lowndes et al 2016). This work ultimately aims
to provide a complex hierarchy of bone and cartilage cells within an implant which
can be used for improved treatment of regeneration of bone in multiple clinical
applications.
Our findings suggest that Wnt3a surfaces can act as a stem cell niche,
maintaining Stro1 expression at the base and increasing the number of migratory
cells that switch expression towards an osteogenic lineage. In this way we can
recapitulate the cascade of cell types prior to treatments and create a 3 D structure
tailored to the cell types present within. We are currently testing varying
polymers based Wnt bandages for in ex vivo cranial repair models to test the
endogenous responses to implants and their integration.
Highlights:





One step aldehyde based chemistry to covalently immobilize hydrophobic Wnt
proteins on glass or polymer scaffold materials.
Long-term storage and continued activation of Wnt signaling over days.
Wnt-platform can enrich and maintain adult/embryonic stem cells in 2D
cultures.
Basal Wnt can direct 3D-multicellularity for engineering tissues.
Immobilized Wnt3a surfaces can be adapted to 3D-culture to direct human
mesenchymal stem cell differentiation
References
Price, J and El Haj, A.J. ( in press) Liquid crystalline ordered collagen substrates for
applications in tissue engineering Royal Society of Chemistry Advances
Lories RJ, Corr M, Lane NE.To Wnt or not to Wnt: the bone and joint health
dilemma. Nat Rev Rheumatol. 2013 Jun;9(6):328-39
Lowndes et al ( 2016) Immobilized Wnt Proteins Act as a Stem Cell Niche for Tissue
Engineering, Stem Cell Reports 12;7(1):126-37
Taken from : Lowndes et al ( 2016) Stem Cell Reports
12;7(1):126-37 Immobilized Wnt3a surfaces can be adapted to 3D-culture to control
human mesenchymal stem cell differentiation.(A) hMSCs were seeded onto immobilized
Wnt3a ± DTT to form a confluent monolayer before overlaying a collagen gel. After 7 days
in culture cells were fixed and stained with DAPI to mark individual cells. The percent of
cells per layer of the collagen gel (lower level: up to 72µm from the gel base, middle: 72132µm and upper: 132-179µm) normalized to number of cells at the base. A representative
bottom-up max projection with DAPI (white) marking each nucleus; example cells in each
layer marked with arrows. (n=3 biological replicates, mean ± SEM). The scale bar
represents 10 μm. (B and C) hMSC in collagen gels were fixed after 7 days and
immunostained for Stro1 (B) and Osteocalcin (C). Expression levels were compared
between immobilized Wnt3a and DTT treated. Quantification of normalized (subtracted
background) image pixel intensity relative to cell number was plotted. (n=3 biological
replicates, mean ± SEM; statistical significance between groups determined by Post-hoc
Mann-Whitney tests; p values correspond to *<0.05). (D) Representative histological
staining of hMSC gels for calcium deposition (Alizarin red staining) after 7 days of culture.
The scale bar represents 100 μm. (E) hMSCs were grown on immobilized Wnt3a ± DTT,
BSA ± soluble Wnt3a (50ng/mL). Migrating cells reported as the percent of total cells in
each defined layer of the gel (lower level: up to 80 µm from the gel base, middle: 80-140 µm
and upper: 140-200 µm). (n=3 biological replicates, mean ± SEM) (F) The percent migratory
cells in each layer when grown on immobilized Wnt3a surfaces ± IWR treatment. (n=3
biological replicates, mean ± SEM)(G) After 7 days hMSCs grown on BSA with soluble
Wnt3a, immobilized Wnt3a or immobilized Wnt3a with IWR treatment were fixed and
immunostained for Stro1. Staining across the middle of the well at the base layer (4x mag)
was quantified. (n=3 biological replicates, mean ± SEM; statistical significance determined
by a one-way ANOVA test; p values correspond to *<0.05, **<0.01 and ***<0.001 ) (for
representative confocal images see Figure S4C). (H) After 7 days hMSCs grown on
immobilized Wnt3a ± IWR treatment were fixed and immunostained for Osteocalcin. A
representative image of an Osteocalcin expressing cell within the collagen gel (left) and the
quantification of Osteocalcin in the layers of the collagen gel (right). (n=3 biological
replicates, mean ± SEM; p values correspond to *<0.05) The scale bar represents 50 M.