Accelerated Bone Healing by Tissue Progenitor Recruitment Inducing Hydrogels

Zurich Open Repository and Archive University of Zurich Main Library Strickhofstrasse 39 CH-8057 Zurich www.zora.uzh.ch

Year: 2017

Accelerated bone healing by tissue progenitor recruitment inducing hydrogels

Papageorgiou, Panagiota

Posted at the Zurich Open Repository and Archive, University of Zurich ZORA URL: https://doi.org/10.5167/uzh-144317 Dissertation Published Version

Originally published at: Papageorgiou, Panagiota. Accelerated bone healing by tissue progenitor recruitment inducing hydrogels. 2017, University of Zurich, Faculty of Science. Accelerated Bone Healing by Tissue Progenitor Recruitment Inducing Hydrogels

Dissertation

zur

(Dr. sc. nat.) Erlangung der naturwissenschaftlichen Doktorwürde vorgelegt der

Mathematisch-

naturwissenschaftlichender Fakultät

Universität Zürich von Panagiota Papageorgiou

aus Griechenland

Promotionskommission

Prof. Dr. Franz E. Weber (Vorsitz) PD Dr. Martin Ehrbar (Leitung der Dissertation) Prof. Dr. Christian Grimm PD Dr. Arnaud Scherberich Prof. Dr. Marcy Zenobi-Wong

Zürich, 2017

Abstract

The healing of large bone defects caused by injury or disease remains a clinically significant problem, often requiring surgical intervention, as bone is the second most commonly transplanted material in the human body after blood transfusion. Therefore, there is increasing need for the establishment of alternative treatments such as tissue engineered constructs or materials that promote bone healing. To date, for bone engineering applications growth factors such as bone morphogenetic protein (BMP) have been demonstrated to significantly promote bone regeneration in human patients and mesenchymal stem/progenitor cells are very useful tools for enhancing bone regeneration. Combinations thereof have shown promising results, though currently they remain difficult to be implemented.

The aim of this thesis was the development of novel strategies to promote the healing of bone by recruitment and manipulation of endogenous bone stem and progenitor cells.

This work relied on previously developed biomimetic hydrogel materials with rationally designed, fully defined properties that were optimized towards the culture and transplantation of cells as well as the release of growth factors. We designed and applied a feasible strategy for precise treatment of molecular modulators in tissue progenitor/stem cell populations to improve skeletal manifestations during injury. We aimed at dissecting the role of endogenous Mesenchymal Stem cells (MSCs) and the recently described Skeletal Stem (SSCs) and Bone Cartilage Stromal Progenitor (BCSP) cells combined with osteogenic cues in biomaterials-assisted bone regeneration.

We demonstrated that these provisional hydrogels, when implanted into bone defects serve as traps for cells participating in the process of bone regeneration. We were able to

1 harvest entrapped cells from explanted hydrogels and used them for the characterization of healing associated cell populations. We tested different freshly isolated or culture expanded cell populations towards their capacity to promote or to participate in bone formation and regeneration. We conclude that all the studied subpopulations, MSCs, SSCs and BCSPs can be easily accessible cell sources that allow the conception of novel bone healing regimen under controlled in vitro and in vivo conditions.

Taken together, we provide new stem/progenitor cell-based bioengineering tools to study factors that promote the healing of bone. These tools will also enable the development and testing of smart bone healing implants that are designed to exploit the

by controlling the dynamic and function of tissue progenitor/stembody’s intrinsic capacitycells. We tosuggest heal that in the future this approach will likely enable the development of novel treatment strategies.

Zusammenfassung

Durch Verletzungen oder Krankheiten hervorgerufene Knochendefekte sind ein immenses klinisches Problem und erfordern häufig operative Behandlungen. Nach

Bluttransfusionen ist Knochen das am zweithäufigsten transplantierte Gewebe. Um die

Entnahme von körpereigenem Knochengewebe zu vermeiden und die Knochenheilung zu unterstützen wurden osteogene Wachstumsfaktoren wie zum Beispiel bone morphogenetic protein (BMP) und mesenchymale Stammzellen/Vorläuferzellen sehr vielversprechende Hilfsmittel. In den vergangenen Jahren wurde an der Entwicklung von künstlichen Knochengeweben mittels «Tissue Engineering» gearbeitet. Obwohl beide

Faktoren alleine oder in Kombination durchaus vielversprechende Ergebnisse erbracht haben, bleibt die Implementierung dieser im klinischen Alltag weiterhin problematisch.

In der vorliegenden Arbeit beschreiben wir eine neue Strategie, in welcher

Knochenheilung durch die Rekrutierung und Manipulation von endogenen mesenchymalen Stammzellen/Vorläuferzellen unterstützt wird. Wir entwickeln eine biomimetische Methode, die darauf abzielt, die natürliche Knochenheilung nachzuahmen und zu unterstützen. Da diese Methode auf der Nutzung endogener Zellen beruht, werden potentielle endogene Vorläuferzellen untersucht. Folglich untersuchen wir klassische mesenchymale Stammzellen (MSCs), sowie kürzlich beschriebene Skelettstammzellen

(SSCs) und Knochen-Knorpel-Stromal-Vorläuferzellen (BCSP). Wir kombinieren diese

Zellen jeweils mit osteogenen Wachstumsfaktoren in Biomaterialien und analysieren die

Knochenheilung. In dieser Arbeit zeigen wir, dass solche Biomaterialien/Hydrogele, wenn sie in Knochendefekte implantiert werden, als Fallen für Zellen, die im

Knochenheilungsprozess eine wichtige Rolle spielen, dienen können. In diesem

Zusammenhang konnten wir rekrutierte Zellen aus explantierten Hydrogelen erfolgreich isolieren und folglich charakterisieren. Verglichen haben wir frisch isolierte und

3 expandierte Zellen hinsichtlich ihrer Fähigkeit Knochenheilung zu unterstützen oder zu begünstigen. Ausserdem haben wir gezeigt, dass MSCs, SSCs und BCSPs gut und einfach zu erhalten sind und daher jeweils einen vielversprechenden Zelltyp darstellen, um neue

Ansätze in der Knochenheilung unter kontrollierten in vitro und in vivo Bedingungen zu untersuchen.

Zusammengefasst legen die Ergebnisse dieser Arbeit nahe, dass diese Methode eine

Möglichkeit für neue und innovative Knochenimplantate schafft. Wichtig ist dabei vor allem, dass die Methode auf der intrinsischen Regenrationsfähigkeit des Körpers beruht.

Durch die Nutzung endogener Regenerationsprozesse können so neue Strategien für die

Knochenheilung entwickelt werden.

Acknowledgements

I would like to express my deepest gratitude to all the people who supported me during the years of my PhD studies. Without their advice and support, this thesis would not have been possible.

I would like to thank PD Dr. Martin Ehrbar for accepting me in his laboratory as a graduate student.

Thank you for giving me the opportunity to increase my scientific knowledge. Prof. Dr. Franz E.

Weber for accepting to be my thesis director. My thesis committee members, Prof. Dr. Christian

Grimm, Prof. Dr. Marcy Zenobi and PD Dr. Arnaud Scherberich thank you for the annual meetings, the diverse discussions with you and for accepting and sharing your time to evaluate my thesis.

All current and former members of the Ehrbar Lab for their support and the pleasant working environment: Dr. Stéphanie Metzger, Dr. Anna-Sofia Kiveliö, Dr. Vincent Milleret, Yannick Devaud,

Ulrich Blache and Queralt Vallmajó-Martin. I would also like to thank Dr. Ali Mirsaidi and Dr.

Anahita Rafiei for their support during the last period of my thesis and for all the advice and enjoyable moments they shared with me.

I am also very grateful to Esther Kleiner, without her I would not have achieved all this work, for all her help and faith in establishing new protocols resulting in beautiful histologies, but also for her support during my hard times. I also thank Yvonne Eisenegger for her precious help with administration and Corina von Arx for always having a bright smile on her face, supporting me with administration inside and outside of the University and always offering open ears to listen and help me with my struggles.

I would like to thank Dr. Malgorzata Kisielow, from the flow cytometry core facility at ETHZ, for helping me with FACS analysis, all the long sorting experiments and for always taking time to answer my questions and believing in my project. Dr. Chafik Ghayor, from Oral Biotechnology &

Bioengineering, for sharing materials and experience in order to conduct my experiments. Flora

Nicholls and the staff from the Animal Facility at USZ for providing advice and help with animal

5 experimentation. Dr. Paolo Cinelli from the division of trauma surgery at USZ, for the support with micro computed tomography.

Ms Heidi Preisig, the coordinator of the imMed PhD program, for all her precious help with administration and for always having a bright smile on her face.

All the fellows of the iTERM training network, for all the interesting meetings, the long discussions and the fun and inspiring moments we shared from the beginning of my PhD till the very end.

The organizing team of Vision 2020 seminars, Marek, Julia, Emilia and Darya, for including me in this team and giving me the opportunity to coordinate these seminar series at the Life Science

Graduate School of the University of Zurich, for further developing my interpersonal skills and having enjoyable dinners.

I also thank Christina, Dimitra, Andriana, Evi, Stella, Gianna, Despoina, Carmen, Asvin and all of my friends who have been part of my life for so many years and have helped me to go through the good and the bad times. My family, which have inspired and supported me all these years, I would like to thank my parents for helping me pursue my dreams ever since I left my country and for their constant support encouraging me to reach the end, without all these long phone-calls I

e made it. My brother for all the strength he has shown, you have been my example whenwouldn’t struggling. hav

Table of Contents

Chapter I Introduction ...... 8

Bone ...... 9

Bone repair ...... 11

Clinical approach ...... 16

Molecular mechanisms mediating bone repair ...... 18

Cellular Modulators during bone formation ...... 21

Bone Engineering ...... 23

PEG (poly (ethylene) glycol) hydrogel for bone tissue engineering...... 26

Motivation and overall goal ...... 28

Specific aims ...... 29

References ...... 31

Chapter III The role of biomaterials delivered SDF-1α in skeletal stem cell homing and bone regeneration...... 37

Abstract ...... 38

Introduction ...... 39

Results ...... 40

Discussion...... 49

Conclusions ...... 51

Materials and Methods ...... 52

Acknowledgments ...... 54

References ...... 55

Chapter V General discussion and future perspectives ...... 57

General discussion ...... 58

Future Perspectives ...... 64

References ...... 66

Chapter I Introduction

Bone

Bone is an essential mineralized tissue, which demonstrates a crucial role in maintaining mechanical and metabolic function of the body. It has the capacity to adapt to its functional environment so that its morphology is developed for the mechanical demand. Physiological bone turnover can be classified into two phases: modeling, which occurs during development, and remodeling, a process involving tissue renewal throughout life (1).

Bone integrity and function are maintained by an exquisite balance between two major cell types involved in the remodeling process: (i) the osteoblasts, arise from mesenchymal stem cells, (ii) the osteoclasts, differentiated from hematopoietic stem cells (2). Osteoblasts are bone-forming cells, while osteoclasts resorb or break down bone. It is essential to maintain a balance between osteoblasts and osteoclasts within the bone tissue.

There are two types of bone tissue: (i) compact and (ii) spongy. Compact bone consists of closely packed osteons or haversian systems. The osteon, which consists of a central canal termed as osteonic (haversian) canal, is encompassed by concentric rings (lamellae) of matrix. Between the rings of matrix, the mature bone cells (osteocytes) are located in spaces called lacunae. Small channels (canaliculi) diverge from the lacunae to the osteonic (haversian) canal in order to provide alleys throughout the hard matrix. In compact bone, the haversian systems are tightly organized to structure the solid mass. Throughout the osteonic canals, blood vessels are aligned in parallel to the long axis of the bone. These blood vessels are interconnected, by perforated canals, with vessels on the surface of the bone. Spongy (cancellous) bone is lighter and less dense than compact bone. Spongy bone consists of plates (trabeculae) and bars of bone alongside to small, irregular cavities that accommodate red bone marrow. To receive blood the canaliculi connect to the alongside cavities, instead of a central medullary canal. The trabeculae bone is arranged in a way to provide maximum strength, the trabeculae of spongy bone is aligned with the lines of stress and can be realigned according to the direction of stress forces (3).

Normally bone development occurs through two distinct mechanisms: (i) endochondral ossification, in which a cartilage model is gradually replaced by bone, and (ii) intramembranous ossification, in which bones are formed directly from condensations of

9 mesenchymal cells without an intermediate cartilage (Fig. 1) (3). Endochondral ossification is the process by which the embryonic cartilaginous model of long bones contributes to longitudinal growth and is gradually being replaced by bone. During endochondral ossification, chondrocytes proliferate, undergo hypertrophy and die; the cartilage extracellular matrix which they construct is then invaded by blood vessels, osteoclasts, bone marrow cells and osteoblasts, the last of these deposit bone on remnants of cartilage matrix. Unlike endochondral ossification, intramembranous ossification is the embryonic development of flat bones, without the participation of cartilaginous tissue. Bone formation in fetal stages begins with the aggregation of mesenchyme stem cells to form condensations. Within the mesenchymal condensation core, cells differentiate into either chondrocytes in endochondral ossification or directly into osteoblasts in the intramembranous bone formation pathway.

Bone is continuously remodeled throughout life and a malfunction between bone resorption and bone deposition of this process can emanate bone disease. Osteoblasts and osteocytes connected by gap junctions are organized within a cellular network permitting the infiltration from the bone surface to the mineral matrix. Osteoclasts, which originate from hematopoietic precursors, are modulators for bone resorption correlated to bone remodeling. (4).

Figure 1: Representation of bone development during: A) intramembranous ossification B) endochondral ossification. A. i) Formation of ossification center by clustering of centrally located mesenchymal cells and differentiation into osteoblast. ii) Osteoblasts begin to secrete osteoid and the entrapped differentiate into osteocytes. iii) Accumulating osteoid settle between embryonic vessels, resulting in the formation of trabeculae bone. iv) Trabeculae is restricted in the inner part as the periosteum thickens, forming the woven bone. B. i) Formation of bone collar around the hyaline cartilage. ii) Newly derived osteoblasts cover the shaft of the cartilage in a thin layer of bone. iii) Blood vessels penetrate the cartilage as new osteoblasts form the spongy bone. iv) The bone thickens and cartilage is restricted at the edges, while the secondary center of ossification is formed. v) Remodeling occurs as growth continues, creating the marrow cavity from the medullary canal. Hyaline cartilage remains now only in the epiphyseal plates (round edges of the long bone) as articular cartilage. (Images adjusted from: http://www.medbullets.com/step1-msk/12021/bone- formation)

Bone repair

Fracture healing is a complex, yet well-orchestrated, regenerative process initiated after injury, resulting in optimal skeletal repair and restoration of skeletal function. Bone injury is typically correlated with disruption of the integrity of the local soft and vascular tissue. During the repair process, the pathway of normal embryonic development is recapitulated by the coordinated participation of several cell types (5). Fracture repair is a highly complex process that is regulated by physiological, cellular, and molecular /genetic factors. The continuing fracture callus remodeling is crucial during the process of repair. This includes remodeling of the soft callus (non-specific catabolism), which is a largely osteoclast-independent process, and remodeling of the hard callus, which is an osteoclast-dependent process (specific catabolism).

Healing can occur either via direct intramembranous bone formation, if fractures are within a proximity range (<25 mm) and well stabilized, or via endochondral bone formation if the fracture is non-stabilized or large defects need to be bridged. This damage activates of non-specific wound healing pathways that accompany non-skeletal injuries. Within the large defects, bleeding is induced in the injury site that evolves into a hematoma, which coagulates between and around the injured bone edges forming a fibrinous clot (6). Shortly after the first hours of injury, the first cells recruited are polymorphonuclear neutrophils, attracted by dead cells and debris and rapidly accumulate. Polymorphonuclear neutrophils secrete several chemokines [IL-12+, CC- +] that attract longer lived monocytes chemokineand macrophages ligand 3 ( 7(CCL3;). The also resident known macrophage as MIP1α) cell population osteomacs, which is localized in the endosteal and periosteal surfaces in close proximity to healthy unfractured bone, seem to be essential for intramembranous bone formation during

11 fracture healing (8). Early during the inflammatory phase a large number of proinflammatory cytokines (IL-6, IL-1, TNF, macrophage colony-stimulating factor 1,

receptorgrowth factor activator (TGF)- of nuclear factorily (bone κB ligand morphogenetic [RANKL]) and protein members [BMP]-2, of the BMP-4, transforming BMP-5, BMP-6) are released β( 9,superfam 10). Concomitantly, platelet-derived growth factors (PDGFs) are secreted from the platelets, from endothelial cells, vascular cells, and macrophages. In addition, angiogenic factors (angiopoetin-1, and, later, vascular endothelial growth factor) are released, consequently to the hypoxic conditions generated by the disrupted vascularization (11). Revascularization is essential for fracture healing, and angiogenesis is required to restore normoxic conditions, abolish debris and supply the injury site with cells and mediators (10). Endothelial cells migrate from periosteal vessels to the bone edges and into the hematoma to generate new blood vessels. Blood vessels also allow access to an essential source of osteoprogenitor cells, which are suggested to derive from pericytes (12). Subsequently, the hematoma is steadily replaced by a granulation tissue, which is rich in collagen fibers, cells and invading capillaries.

Most of bone injuries retain a certain level of mechanical instability and are bridged via endochondral ossification. Chondrocytes and fibroblasts form a cartilaginous callus, which then undergoes mineralization, resorption and is ultimately replaced with bone (13). The callus forms during the late inflammatory phase and intramembranous bone formation starts 3-7 days after injury (7). The cells responsible for forming the soft callus arise from mesenchymal stem/progenitor cells (MSCs) that proliferate and generate the cartilaginous matrix. It is not fully understood where these cells derive, but MSCs most likely originate from different sources including periosteum (13, 14), trabecular bone (15), surrounding tissues (16), bone marrow (17), and circulation (18). Several cytokines and growth factors can induce the mobilization of MSCs to the site of injury. For example, stromal cell-derived factor-1 (SDF-1) and its receptor CXCR-4 form an axis (SDF-1/CXCR- 4) which regulates the recruitment of MSCs to the site of injury (19). Moreover, PDGF-BB, which is secreted by platelets and preosteoclasts, elevates the proliferation and migration of different mesenchymal cell types such as MSCs, pericytes and fibroblasts (20). Hypoxia, as well, is an essential factor for mobilization of progenitor cells (21). Chondrocytes derived from MSCs proliferate and synthesize a cartilaginous matrix composed of collagen-I and -II until the fibrinous/granulation tissue is replaced by cartilage. When cartilage can not be further produced, fibroblasts aggregate within the region generating

12 fibrous tissue. Fibroblast proliferation, chondrocyte differentiation, and matrix production are orchestrated by the sequential expression of growth factors including TGF- and 2, PDGF-BB, FGFs, insulin-like growth factor (IGF), and growth/differentiationβ3 –β factor 5 (GDF-5) (22, 23). Notably, members of the BMP family including BMP-2, -4, -5, -6, -7 also boost cell proliferation and chondrogenesis (24). TGF- .

Whileβs and responding BMPs are mainly to these secreted signals, bychondrocytes mesenchymal are cells,able toosteoblasts, generate significant and chondrocytes amounts of collagen II (or collagen X for hypertrophic chondrocytes) (22).

Following the regeneration process, the primary soft cartilaginous callus is gradually resorbed and replaced by a hard bony callus, which allows mechanical rigidness, due to its solid interface. At the site of bridging, the soft callus is gradually absorbed, and the concomitant revascularization allows the new hard callus to arise. During hard callus formation, a process which is typically irregular and under-remodeling, chondrocytes undergo apoptosis via hypertrophic formation and calcium release while the newly recruited monocytes to the sites differentiate into osteoclast-like cells in order to resorb the calcified cartilage (7). The resorption of this mineralized cartilage is organized via the signaling cascade comprising M-

(RANKL), osteoprotegerin (OPG), andCSF, TNF- receptor(22 ). activator At the same of nuclear time, MSCs factor differentiate κB ligand into osteoblasts and concomitantly replaceα the connective and granulation tissues through intramembranous bone formation. This process is mainly driven by BMPs including BMP-2, -3, -4, -5, -6, -7, and -8 (25), which are secreted by hypertrophic chondrocytes, osteoblasts and osteocytes (24, 25). Interestingly, this step recapitulates the embryological bone development to some extent, with a combination of cellular proliferation and differentiation, increasing cellular volume and matrix deposition (22).

During the final stage of bone repair, the hard callus is replaced by bone tissue with the original cortical and/or trabecular architecture. This phase has been termed as secondary bone formation (9). Initially, this process involves the converting of irregular woven bone callus into lamellar bone, while the standard cortical configuration is ultimately restored. A coupled process of orderly bone resorption followed by the formation of lamellar bone, drives the remodeling phase while maintaining the process of neo-vascularization. Osteoclasts are the key cells involved in the resorption of mineralized bone, adapt a polarized morphology and adhere to a mineralized surface. The processes by which

13 osteoclasts resorb bone are acidification and proteolysis of the bone matrix and the deposition hydroxyapatite crystals, which are encapsulated within the surrounding zone of attachment. The initial process during bone matrix resorption is aggregation of the hydroxyapatite crystals by absorption of their ligand to collagen. Then, the residual collagen fibers are digested by cathepsins or activated collagenases and the digested residues are internalized or transported across the cell and released at the basolateral domain. Bone resorption by osteoclasts creates erosive cavities on the bone surface (26). Osteoclast function is regulated by both locally acting knowncytokines as “Howship’sand systemic lacuna” hormones (27). The degradation products are removed through vesicles generated from the ruffled border towards the functional secretory domain, while osteoclasts are either apoptosed, or restored to the initial non-resorbing form.

Figure 2: Representation of the four phases during bone regeneration, including the cellular mechanisms and key regulators cytokines and growth factors. A) Inflammatory phase, B) soft callus formation, C) mineralization and resorption of the soft callus, and D) bone remodeling. During each phase cytokines and growth factors secreted by different cell types. Revascularization and angiogenesis is an ongoing process through the inflammatory phase and until the resorption phase (adapted from Mikaël M. Martino (6)). BMP = bone morphogenetic protein, FGF = fibroblast growth factor, GDF-5 = growth/differentiation factor 5, IGF-1= insulin-like growth factor 1, M-CSF = macrophage colony-stimulating factor, OPG = osteoprotegerin, PDGF = platelet-derived growth factor, PlGF=placental growth factor, PTH=parathyroid

-1=stromal cell-derived factor 1, TGF- facto - hormone, RANKL=receptor activator of nuclear factor κB ligand, SDF β=transforming growth r β, TNF α = tumor necrosis factor α, VEGF= vascular endothelial growth factor. Clinical approach

Large bone defects associated with trauma, tumor, or infection frequently require surgical intervention (5, 28). Non-union is the absence of bone union tissue, following a fracture that depends on additional treatment/ fixation before healing can occur. The exact time length needed to determine whether a fracture is a non-union is not well defined, nevertheless, it was suggested at nine months by the United States Food and Drug Administration (29). Autologous bone grafts have been used in the treatment of nonunions defects and currently autograft remains the most commonly used bone graft (28). It can originate from a variety of different bone tissue sites, such as the iliac crest, distal femur, proximal tibia, fibula, distal radius, and olecranon. However, the clinical most common source of autologous bone graft remains the aspiration of cancellous bone originated from the iliac crest. Iliac crest bone graft provides optimal osteoconductive and possibly osteoinductive properties to the non-union site. Typically, bone tissue can be harvested from the iliac crest preserving cell viability and the trabecular architecture. Even though the graft cannot provide mechanical stability, it is rapidly incorporated into the host healing fracture site (30). Aspirated bone marrow from the iliac crest encompasses progenitor cells which can be utilized to induce the osteogenic response of the implanted allografts or to restore a non-union defect when applied percutaneously. The fibrous tissue interposed between the bone edges ossifies following the injection of the bone marrow both in the non-union gap and around the bones. It is yet not fully understood whether the bone marrow is responsible to transform the fibrous tissue into bone or whether the interposed tissue is converted into bone tissue after the initiation of callus formation (obtained from the surrounded bone graft), which blocks micromotion at the non-union site and allows union of the gap (28).

Currently, the transplantation of autologous iliac crest bone graft remains the only clinically available graft source that fulfils the criteria of osteogenicity, osteoinductivity, and osteoconductivity, while containing viable precursor cells. Its use has resulted in reliable union in many clinical studies. However, the risks of infection, rejection, and donor site morbidity as well as its limited supply undermine its clinical application and remain a serious problem that must be resolved (28, 29). It has been suggested that 5%

16 to 10% of all long bone defects have delayed union, or developed into a non-union (31). Apparently, these patients lack or show reduced levels in one or more of the essential factors required in fracture healing. Early identification and treatment of impaired fracture healing could prevent significant negative impacts, such as prolonged therapy, associated with extensive treatment costs for patients (32).

Bone morphogenetic proteins (BMPs) have been shown to promote the healing of large bone defects. The growth factor BMP-2 has been approved by the US Food and Drug Administration (FDA) since 2002 and has been clinically established for lumbar spine fusion and repair of open tibia fractures (33). In a bone cortex, osteoprogenitor cells, osteoblasts, chondrocytes and platelets produce BMPs. After their release, BMPs are stored temporary in the extracellular matrix. The regulatory effects of BMPs are tightly associated to the target cell type, the stage of the cell differentiation, the local concentration of BMPs, along with the interactions with other released proteins. BMPs orchestrate a sequence of evolving events resulting in chondrogenesis, osteogenesis, angiogenesis and controlled secretion of extracellular matrix (33). A controlled and local delivery of BMPs is required in order exert their osteogenic potency. Furthermore, for clinical utilization, the short half-life of BMPs needs to be taken into consideration. To overcome this limitation several delivery systems have been advanced (22, 34). For clinical use of rhBMP-2 (dibotermin alfa), established under the product names InductOs® (UK) and InFUSE (US), the active factor is supplied within a bovine collagen sponge carrier, allowing relative slow release over time, a combination which was FDA approved in 2004 (35). The other clinically used BMP, rhBMP-7, under the product names Osigraft® (UK) and OP-1 Putty (US), is administrated in granular form supplemented by a bovine collagen carrier. Administration of rhBMP-7 demonstrated more efficient bone formation and better bridging the bone edges of defects in comparison to controls patients (36).

Several side effects have been implied with the use of rhBMPs during fracture healing. These are associated to the supra-physiological doses, resulting in ectopic bone formation and stimulation of cancer cells are being extensively studied. It has been demonstrated that RhBMP-2 can induce the formation of pancreatic cancer in a retrospective cohort study (37, 38). Side effects of BMP-2 such as ectopic bone formation in non-union treatment and acute soft tissue swelling for cervical spine fusions have also been

17 described. However, these effects were associated with the use of very high BMP doses in animal studies and varied among species. Ultimately remodeling to the normal bone architecture occurred. Apart from the side effects, the costs and the application of rhBMPs are also debatable. BMPs are delicate to handle, expensive to manufacture and used in supra-physiological doses, resulting in high costs (32, 35). Nevertheless, several studies indicate that, the clinical use of both BMP-2 and BMP-7 is still acclaimed, despite these costs. Cost reductions for hospitals are mainly associated to shorten the time of surgery, as a result of avoiding the bone grafting procedure and faster discharge of the patient. (39). Delivering superphysiological doses of BMPs with the aim for enhancement of the osteogenic potency might result in the opposite outcome. Recent studies using sophisticated biomaterials as BMP carriers show more efficient healing with reduced amounts, corresponding more with physiological concentrations (40, 41). Nonetheless, to date, BMPs remain a robust treatment to enhance fracture healing with low risk of adverse events if properly used.

Molecular mechanisms mediating bone repair

Bone injury is associated with destruction of the vasculature resulting in incomplete bone healing. Many growth factors, that are expressed during blood vessels formation and induced in response to angiogenesis during injury, are believed to play also a significant role in the process of bone repair (11). These include members of the FGF, TGF- , BMP,

IGF, PDGF, VEGF, hepatocyte growth factor (HGF), erythropoietin, growth hormone,β as well as parathyroid hormone.

Recent studies have been conducted to improve understanding of FGF signaling in bone formation. FGF-2 has been suggested to be a key-player of cartilage and bone cell formation (42, 43). FGF signals through fibroblast growth factor receptors (FGFRs), which are important regulators of osteoblastogenesis. Osteoblasts mainly express FGFR1 and FGFR2, which are the most crucial receptors in bone-associated cells during the postnatal life. Previous studies on humans and mice have indicated the consequences of FGFR dysregulation on bone formation (44-47). FGFRs are also important regulators of bone growth and development and are expressed differentially in bone and cartilage (45, 48). The importance of endogenous FGF2 on bone formation has been emphasized by the deletion of FGF2 that leads to decreased osteoblastogenesis associated with an osteopenic

18 phenotype in mice (49). Xiao et al. suggested that endogenous FGF2 drives mesenchymal bone marrow stromal cell differentiation into adipocytes or osteoblasts (50). In this murine model, endogenous FGF2 was suggested to participate in bone formation both in basal conditions and under induction by parathormone (51). Other members of the FGF family may also be significant modulators of osteogenesis in vivo. Furthermore, delayed ossification has been demonstrated in mice lacking expression of FGF18 indicating that FGF18 regulates osteogenesis (52).

The TGF- sed of over forty members, including TGF-

Activin, andβ superfamily bone morphogenetic is compri proteins (BMPs) (53, 54) . TGF- βs, Nodal, expressed in most cell types and are suggested to be involved in βtissue family morphogenesis members are and differentiation. TGF- such as epithelial and hematopoietic cells, and itsβ inhibits signaling proliferation regulates tumorigenesis of many cell types, (55, 56 ). The cells response to TGF- is dependent on the cell morphology and physiological environment and differs betweenβ transcriptional and nontranscriptional regulation . TGF- been demonstrated to couple bone construction by osteoblast and bone βs resorption have by osteoclast (57, 58). TGF- mesenchymal stem/progenitorβ is cells stimulated and plays by an autocrineimportant and/orrole in their paracrine maintenance effects and of expansion (59). Bone and cartilage comprise extensive amounts of TGF- for TGF- Moreover, TGF- induces osteoprogenitorβ and target cells proliferation,β activity.early differentiation, and differentiationβ signaling alsotowards the osteoblastic lineage’s via the selective MAPKs and Smad2/3 pathways. The cooperation between TGF-

PTH, Wnt, BMP, as well as FGF signaling has been studied. TGF- motes matrixβ and production and osteoblast differentiation and inhibits the osteoblasticβ1 pro secretion of the osteoclast differentiation factor RANKL. TGF- osteoclast formation and induces bone mass β1 (60 ). thereby TGF- indirectly restricts further closure by promoting osteoprogenitor proliferation throughβ2 stimulates the MKK cranialERK signaling suture (61), whereas combined delivery of TGF-’s calcium-phosphate– matrix at the tendon-bone repair site is associatedβ3 with withan injectable new bone formation, increased fibrocartilage, and improved collagen organization at the healing tendon-bone interface (62).

BMP is currently the most widely utilized inductive factor in bone tissue engineering as the US FDA has already approved BMP-2 and BMP-7 for application in fracture and non-

19 union management. BMP has been involved in a variety of functions, mainly acting as a differentiation modulator of mesenchymal stem cells during endochondral ossification (63). The osteoinductive potency of local BMP application have been widely studied in various animal species (murine, bovine, and primate). The efficacy of BMP upon bone formation have often been correlated with the stimulation of BMP during angiogenesis, specifically within bone. BMP secreted in the proximity of the vascularization site has a beneficial effect on inducing cells to produce angiogenic factors (64). However, the mechanism of BMP augmenting angiogenesis is not fully characterized. Deckers et al. showed that BMP induced angiogenesis via the secretion of VEGF from osteoblasts (65), whereas other studies revealed that the BMP/Smad signaling pathway is involved in the regulation of VEGF transcription, affecting endothelial cells proliferation, migration and tube formation (66).

Growth hormone (GH) and insulin-like growth factor (IGF-1) are essential during skeletal growth, puberty and bone health throughout life. GH increases tissue formation by acting directly and indirectly on target cells whereas IGF-1 is a critical mediator of bone growth. Clinical studies demonstrating the use of GH and IGF-1 in osteoporosis and bone fracture healing are depicted (67). GH deficiency in childhood decreases bone mineral density (BMD), while GH treatment leads to increases in bone growth and strength in defected children (68), IGF-1 is considered essential for longitudinal bone growth, skeletal maturation, and bone mass acquisition during development and in the maintenance of bone during adult life (69). Osteoblasts are stimulated by IGF secretion resulting in enhanced expression of extracellular matrix such as type-I-collagen (70, 71).

PDGF, which activates angiogenic pathways, is another important osteogenic growth factor in fracture healing (72). PDGF is a glycolytic protein released by platelets and other cell subpopulations during injury. It stimulates the growth of tissue progenitors in bone cartilage, vascular tissue, and connective tissue (72). PDGF-BB/PDGFR- constitutes the principal pathway responsible for pericyte recruitment and attachmentβ signaling to the vasculature, their subsequent maturation and potential destabilization and detachment (73, 74). Recently, PDGF-BB has gained great interest for bone regeneration. It appears that PDGF-BB could play an essential role contributing to the osteoblast differentiation process (75). PDGF-BB, in addition to its ability to promote angiogenesis, has been reported at sites of injury to promote and mobilize the proliferation of MSCs,

20 osteoblast progenitors and pericytes. In this way, PDGF-BB has been shown to both contribute to the osteogenic lineage differentiation and to support stabilization of newly forming blood vessels that act to arouse the multistep and multicomponent process of new bone formation (75). As a potent mitogen and chemokine, PDGF-BB facilitates tissue regeneration by stimulating mesenchymal originated cells such as fibroblasts, osteoblasts and chondrocytes. It is also an angiogenic factor, which can upregulate VEGF expression (76).

VEGF is established as a major modulator of angiogenesis, which stimulates endothelial cells proliferation, migration and formation of capillary-like structures (77). Although the primary function of VEGF is related to vasculature, strong evidence indicates that VEGF is also implicated in the survival, activity and recruitment of bone forming cells. Previous studies have shown that VEGF directly promotes migration and differentiation in primary human osteoblasts (78). VEGF is expressed before blood vessels formation in developing bones and its expression is associated with osteoblasts. Furthermore, Peng and colleagues showed that the synergistic interaction between BMP-2 and VEGF is partially generated by enhanced angiogenesis in the area of bone formation, which leads to increased cartilage resorption and augmented mineralized bone formation. These data indicated that VEGF plays also an important role in bone development and bone repair beyond its angiogenic function (79). Additionally, it was shown that the inhibition of VEGF activity disrupts repair of murine femoral fractures and cortical bone defects (80), decreases blood flow leading to non-unions in rabbit radial fractures (81). Thus, VEGF activity is not only essential for normal angiogenesis, but also for blood vessel invasion which is important during tissue repair, suggesting the tight connections between angiogenesis and osteogenesis in bone repair may occur through VEGF production (79, 81)

Cellular Modulators during bone formation

Extensive investigation has been reported over the previous years to identity an Mesenchymal stemendogenous cells (MSCs) ‘‘skeletal have stem been cell’’, widely but used it remains in stem not cell fullytransplantation, comprehensive. tissue engineering and immunotherapy throughout the years. Scadden determined MSC niches as cell pools that can facilitate homeostasis while sustaining MSC populations (82-84). Niches for MSCs of different origins can be placed neighboring the vessel walls, on the endosteal regions

21 of trabecular bone, within the interfibrillary areas, or perivascular (85). MSCs migrate to injury sites in different animal models, but the underlying mechanisms remain unclear. The term MSC was originally conceived in relevance to a common progenitor of a wide aematopoietic, nonepithelial, and mesodermal) tissues range(86). Focusing of “mesenchymal” on a combination (nonh of in vitro assays and cells surface markers expression (87), MSCs exhibit a broad spectrum of lineage potentialities (extending to chondrocytes, osteoblast, adipocytes and muscle cells).

In the past years, bone marrow extracts have been vastly used for treating large bone defects in orthopedic surgery (17, 82). Periosteum, as well, has been utilized as a source of MSCs (14, 88). Other studies have investigated the role of pluripotent mesenchymal cells from vessel walls, pericytes, during bone healing, since ineffective angiogenesis marks non-union defects (89, 90). They demonstrated that pericytes can be induced towards expression of chondrogenic and adipogenic markers under defined culture conditions. Muscle has also been utilized as cell source for bone repair (91), but harvesting is limited and the cells show restricted capacity during multilineage differentiation. The hypothesis of circulating MSCs during bone healing remains questionable (92). Sources of MSCs, previously used, include adipose tissue, umbilical cord blood, dental tissues, tendon and skin (15, 93-97). Numerous studies have demonstrated that MSCs undergo osteogenic lineage by differentiation and can form mature osteoblasts (12, 98). In addition, many other studies have reported that the delivery of recruitment factors together with a sufficient quantity of MSCs within the microenvironment of the injury site is effective for healing (99-101). The recruitment of the stem cells which are involved in callus condensation is related to the extent and the type of defect (102-104). Another parameter that should be taken into consideration is stability during fixation and immobilization of the damaged tissue, which will also influence the lineage restriction of skeletogenic stem cell to differentiate into either chondrocytes or osteoblasts, with more-extensive cartilage tissue formation associated with less stability for chondrocytes, and increased stability due to bone tissue in favor of osteoblasts (104).

However, despite the fact that the MSC populations have been extensively employed in defining immunophenotypes and in vitro clonogenic and multilineage potential (87), their in vivo function and anatomical localization has not been rigorously characterized (105). Only recently, it was shown that MSCs in multiple human tissues reside in perivascular

22 niches and that prospectively isolated CD146+ human bone marrow-derived SSCs give rise to bone when transplanted in the absence of osteogenic signals (12, 106, 107). These cells can self-renew upon serial transplantation, fulfilling a key criterion of bona fide stem cells. Analogues to CD146+ human bone marrow SSCs, murine SSCs have been isolated from the bone marrow by selecting for Sca1 (108), Nestin (109), Mx1 (82), or Leptin receptor expression (110). Furthermore, by lineage mapping of developing long bones, based on the expression of AlphaV ( V integrin) for the osteogenic lineage, a mouse SSC population has been identified, whichα when transplanted beneath the renal capsule, can give rise to osteogenic, chondrogenic and stromal tissue (111). Chan et al. suggested that skeletogenesis may proceed through lineage-restricted progenitors selected upon developmental hierarchy a process similar to hematopoiesis. The SSCs (CD45-TER119- Tie2-AlphaV+Thy-6C3-CD105-CD200+) initiate skeletogenesis by producing a hierarchy of increasingly fate restricted progenitors, such as Bone Cartilage Stromal Progenitors (BCSP) (CD45-TER119-Tie2-AlphaV+Thy-6C3-CD105+) (111). His research indicates that the skeletogenic progenitors are diverse, with specific cell-surface profile markers and skeletal tissue fates, resembling the hematopoietic progenitor cells that differentiate into the various blood cells.

Collectively, the ideal stem cell source for fracture repair should be easily accessible, isolation should be conducted without being time and labour-intense and cells should be rapidly expandable by in vitro culture, while survival and integration within the host bone tissue should be adequate, showing no tumourigenicity or site morbidity effects.

Bone Engineering

Bone has the capacity to regenerate without the formation of scar tissue, but only if certain conditions are met. These conditions are related to: (i) the size of the defect, (ii) the location and vicinity to an efficient vascular bed to allow vasculature and (iii) the mechanical stability of the defect. The combination of osteoconductive biomaterials with osteoinductive factors and stem/progenitor cells holds a great promise. Although, we still do not fully understand optimal combinations, doses and release kinetics, making them hard to be implemented in safe and clinically effective approaches.

Bone tissue engineering is a promising alternative strategy for inducing tissue regeneration and has thus attracted considerable attention (112, 113). The capacity of bone to self-renew has been an advantageous paradigm of nature to assist and inform tissue engineering strategies towards generating new options of treatment. Conventional tissue engineering combines of biomaterials, appropriate bioactive molecules, and cells, which may be limited due to long-term in vitro culture for the formation of new bones, the high cost of cell isolation, proliferation and maturation or differentiation, as well as the risk of pathogens contamination (114).

Alternatively, the principle of in situ bone tissue regeneration involves target-specific scaffolds incorporated with bioactive molecules and implanted in vivo. The sustained release of the bioactive cues deciphers

Consecutively, this enhances the mobilizationthe of body’s tissue-specific own regenerative host stem/progenitor capability. cells, promotes proliferation and differentiation of these recruited cells into the targeted cell types and reconstructs functional tissues allowing the damaged tissue to be regenerated without the need for cell transplantation. Biomaterials and scaffolds specific for bone, either alone or in combination with bioactive osteogenic modulators, are directly introduced to the defect regions, to recruit circulating host stem cells in situ, aiming to induce the osteogenic differentiation of tissue progenitors towards bone regeneration (115). Biomaterials for in situ tissue regeneration (BITR) must possess some basic characteristics such as: (i) suitable microstructures and mechanical properties, (ii) proper surface topography and chemistry, (iii) biodegradability and non- cytotoxic degradation products, and (iv) simple and cost effective manufacturing technology (116).

Biomaterials comprises a key element in bone tissue engineering and provides a three- dimensional (3D) scaffold to support cell growth and the formation of the extracellular matrix (ECM) (116). Scaffolds typically consist of a solid support structure with an interconnected pore network, while matrices are often hydrogels containing encapsulated cells. Both forms must support cell colonization, proliferation, differentiation and migration. Additionally, they should maintain appropriate physicochemical properties (such as stiffness, strength, biodegradability, surface configuration) that are essential for tissue formation and have an application in sustaining and responding to mechanical stresses (117). The site of implantation along with the size

24 and shape of the implant can affect in vivo degradation. Assessment of the biodegradability of engineered constructs in vivo is still its infancy, with the main focus being on the local effect of dissolution products from scaffold degradation on osseointegration and bone formation. Part of the problem is the lack of long-term follow- up studies on these materials when used in vivo in clinical settings (117). Most of these biomaterials are not in widespread clinical use, reducing the power of any follow-up study to determine the rate of degradation in humans.

Synthetic and natural polymers are popular biomaterials due to the vast diversity of their properties and their bioactivity (118). Natural polymers were among the first biodegradable scaffold materials to be used clinically, cause of their preferred overall interactions with various cell types, and lack of an immune response. Natural polymers -tricalcium phosphate are suchcurrently as fibrin, used collagen, both alone gelatin, and combined alginate, hydroxyapatite(119). While these and biomaterials β have shown to possess to certain extent osteogenic capacities, often they are not sufficient to promote bone healing. Indeed, optimal bone regeneration not only depends on an osteoconductive matrix and on mechanical stability, but also on osteoinductive modulators and osteogenic cells (120, 121). However, synthetic polymers despite the potential for an immune reaction or toxicity when combined with certain polymers, were later recognized to have lower cost and with the advantage to be tailored (122). Among the synthetic polymers, poly(glycolic acid) (PGA), poly(lactic-co-glycolic) acid (PLGA), poly(L-lactic acid) (PLLA) and poly(caprolactone) (PCL) are currently the most widespread for the creation of 3D scaffolds (123). These polymers have also been utilized in combination with natural polymers to improve adverse effects related to hydrophilicity, cell attachment, and biodegradability. These synthetic scaffolds offer a minimalist approach to develop models in a fully controlled and reproducible manner, enabling the study of biological and cellular ver, the scaffold can be functionalized using ligand processesspecific domains on a “blank such asslate”. protein Moreo modulators that assist in enhancing cellular responses. Therefore, common strategies to response into better or faster healing are currently issued by delivering osteoinductive growth factors and/or stem/progenitor cells through osteoconductive biomaterials.

PEG (poly (ethylene) glycol) hydrogel for bone tissue engineering

Many laboratories are using synthetic materials such as PEG to mimic stem cell environments. Previously in our lab, an enzymatically crosslinked PEG-based hydrogel termed TG-PEG was developed. TG-PEG is crosslinked by the transglutaminase factor XIII (FXIII) recapitulating the end of the blood coagulation cascade. In TG-PEG, 8-arm-PEG- vinylsulfone molecules are functionalized with two peptides via a Michael-type addition reaction. These peptides, containing lysine or glutamine residues, act as substrates for FXIIIa crosslinking (124). PEG-based hydrogels are inert and highly reproducible. Their mild crosslinking conditions maintain the viability of encapsulated cells and allow for ECM deposition. This hydrogel has proven its potential to allow the recreation of environments similar to those encountered by cells in nature.

The ECM-mimicking structure does not only give an adequate network for cell survival but also enables stiffness and degradability variation for optimal cell behavior. Furthermore, its chemical structure permits the addition of cell instructive signals to its backbone, allowing the flexible and independent variation of multiple parameters towards the reconstruction of a native-like 3D microenvironment (125). Simultaneously with the hydrogel crosslinking, growth factors, adhesion peptides and other biological entities can be incorporated in a highly controlled manner (126). Adhesion sites such as RGD are added following the same principles. This amino acid sequence combined with a MMP-sensitive domain, present in the linker-peptide, is essential for allowing cell survival, spreading and migration. Lastly, these hydrogels allow for facile tailoring of properties such as stiffness, degradability, and cell-adhesion (127, 128).

Figure 3: Formation of TG-PEG hydrogels by crosslinking two multiarm PEG peptide conjugates (n-PEG- MMP sensitive P-Lys and n-PEG-Gln) in combination with a cell adhesion peptide (TG-Gln-RGD) catalyzed by FXIII transglutamination.

Motivation and overall goal

Previously described clinical studies, based on strategies promoting the inherent (limited) capacity of bone to regenerate, rely on the delivery of osteoinductive factors such as BMPs (33) or on the transplantation of autologous bone marrow-derived stem cells BM-MSCs (28). MSCs were shown to promote bone regeneration in preclinical trials, however achieving statistically significant efficacy in clinical trials has been challenging (129). This is most likely due to variation in stem/progenitor cell selection criteria and because(2). the cells’ regenerative capability cannot readily be controlled once transplanted

While it is well known that bone has an inherent capacity to regenerate, this potential has not been fully explored to date. This is mostly due to the still incomplete knowledge on cellular and molecular mechanism modulating bone regeneration (104). One of these missing gaps until very recently has been the lacking information on both the phenotypic properties of skeletal stem cells and the anatomic localization of their niche (105, 107). Only very recently cell populations, which likely comprise skeletal stem and progenitor cells have been characterized and isolated (109-111). However, this newly established knowhow so far has not been implemented into therapeutic approaches, which would

Asmagnify a basis the for body’s this project, inherent we regenerative employed previously capacity. developed biomimetic PEG-hydrogels, which can be tailored towards specific biological applications by the independent variation of physical and biological properties (124, 127). In previous studies, these provisional materials were optimized toward the induction of bone regeneration in a murine craniotomy model (128, 130). Such materials which contained or did not contain growth factors were implanted in murine calvarial defects. Our data indicated that tissue stem/progenitor cells were able to infiltrate BMP-2 releasing hydrogels where they differentiated into bone forming cells and participated in bone healing. The overall goal of this thesis was to develop treatments that by mobilization or transplantation of skeletal stem and progenitor cells could promote the BMP-2-mediated murine calvarial bone regeneration.

Specific aims

Specific aim 1: To establish a model for recruitment, migration and differentiation of tissue stem/progenitor cells during orthotopic bone healing. The aim of this study was to identify healing associated stem/progenitor populations and to determine factors that by promoting their mobilization will enhance BMP-2 mediated bone regeneration. We focused on establishing an in vivo model, which allows the trapping of cells that participate in the healing murine calvarial bone defects. Applying this biomimetic cell- trap , we aimed at identifying cell populations that fulfil both in vitro and in vivo criteria“ of endogenous” skeletal stem/progenitor cells and factors that promote bone regeneration by the recruitment of identified skeletal stem/progenitor cells.

Specific aim 2: To investigate the role of SDF-

Figure 4: Conceptual schematic of herein thesis.

References

1. J. Rubin, C. Rubin, C.R. Jacobs, Molecular pathways mediating mechanical signaling in bone, Gene 367 (2006) 1-16. 2. L. Duplomb, M. Dagouassat, P. Jourdon, D. Heymann, Concise review: embryonic stem cells: a new tool to study osteoblast and osteoclast differentiation, Stem Cells 25(3) (2007) 544-52. 3. C.S. Soltanoff, W. Chen, S. Yang, Y.-P. Li, Signaling Networks that Control the Lineage Commitment and Differentiation of Bone Cells, Crit Rev Eukaryot Gene Expr. 19(1) (2009) 1- 46. 4. W. Huang, S. Yang, J. Shao, Y.-P. Li, Signaling and transcriptional regulation in osteoblast commitment and differentiation, Front Biosci. 12 (2013) 3068-3092. 5. R. Dimitriou, E. Tsiridis, P.V. Giannoudis, Current concepts of molecular aspects of bone healing, Injury 36(12) (2005) 1392-404. 6. M.M. Martino, P.S. Briquez, K. Maruyama, J.A. Hubbell, Extracellular matrix-inspired growth factor delivery systems for bone regeneration, Adv Drug Deliv Rev 94 (2015) 41-52. 7. L. Claes, S. Recknagel, A. Ignatius, Fracture healing under healthy and inflammatory conditions, Nat Rev Rheumatol 8(3) (2012) 133-43. 8. K.A. Alexander, M.K. Chang, E.R. Maylin, T. Kohler, R. Muller, A.C. Wu, N. Van Rooijen, M.J. Sweet, D.A. Hume, L.J. Raggatt, A.R. Pettit, Osteal macrophages promote in vivo intramembranous bone healing in a mouse tibial injury model, J Bone Miner Res 26(7) (2011) 1517-32. 9. L.C. Gerstenfeld, D.M. Cullinane, G.L. Barnes, D.T. Graves, T.A. Einhorn, Fracture healing as a post-natal developmental process: molecular, spatial, and temporal aspects of its regulation, J Cell Biochem 88(5) (2003) 873-84. 10. K. Schmidt-Bleek, H. Schell, N. Schulz, P. Hoff, C. Perka, F. Buttgereit, H.D. Volk, J. Lienau, G.N. Duda, Inflammatory phase of bone healing initiates the regenerative healing cascade, Cell Tissue Res 347(3) (2012) 567-73. 11. Z.S. Ai-Aql, A.S. Alagl, D.T. Graves, L.C. Gerstenfeld, T.A. Einhorn, Molecular Mechanisms Controlling Bone Formation during Fracture Healing and Distraction Osteogenesis, Journal of dental research 87(2) (2008) 107-118. 12. B. Sacchetti, A. Funari, S. Michienzi, S. Di Cesare, S. Piersanti, I. Saggio, E. Tagliafico, S. Ferrari, P.G. Robey, M. Riminucci, P. Bianco, Self-renewing osteoprogenitors in bone marrow sinusoids can organize a hematopoietic microenvironment, Cell 131(2) (2007) 324-36. 13. R. Marsell, T.A. Einhorn, The biology of fracture healing, Injury 42(6) (2011) 551-5. 14. S. Matsushima, N. Isogai, R. Jacquet, E. Lowder, T. Tokui, W.J. Landis, The nature and role of periosteum in bone and cartilage regeneration, Cells Tissues Organs 194(2-4) (2011) 320-5. 15. N.J.Y. Lin Song, Nicole E. Webb, and Dr. Rocky S. Tuan. , Origin and Characterization of Multipotential Mesenchymal Stem Cells Derived from Adult Human Trabecular Bone, Stem Cells and Development 14(6) (2006) 712-721. 16. G.E. Glass, J.K. Chan, A. Freidin, M. Feldmann, N.J. Horwood, J. Nanchahal, TNF-alpha promotes fracture repair by augmenting the recruitment and differentiation of muscle-derived stromal cells, Proc Natl Acad Sci U S A 108(4) (2011) 1585-90. 17. C. Colnot, S. Huang, J. Helms, Analyzing the cellular contribution of bone marrow to fracture healing using bone marrow transplantation in mice, Biochem Biophys Res Commun 350(3) (2006) 557-61. 18. X. Wang, Y. Wang, W. Gou, Q. Lu, J. Peng, S. Lu, Role of mesenchymal stem cells in bone regeneration and fracture repair: a review, Int Orthop 37(12) (2013) 2491-8. 19. T. Kitaori, H. Ito, E.M. Schwarz, R. Tsutsumi, H. Yoshitomi, S. Oishi, M. Nakano, N. Fujii, T. Nagasawa, T. Nakamura, Stromal cell-derived factor 1/CXCR4 signaling is critical for the recruitment of mesenchymal stem cells to the fracture site during skeletal repair in a mouse model, Arthritis Rheum 60(3) (2009) 813-23. 20. H. Xie, Z. Cui, L. Wang, Z. Xia, Y. Hu, L. Xian, C. Li, L. Xie, J. Crane, M. Wan, G. Zhen, Q. Bian, B. Yu, W. Chang, T. Qiu, M. Pickarski, L.T. Duong, J.J. Windle, X. Luo, E. Liao, X. Cao, PDGF-BB secreted by preosteoclasts induces angiogenesis during coupling with osteogenesis, Nat Med 20(11) (2014) 1270-8. 21. C. Yellowley, CXCL12/CXCR4 signaling and other recruitment and homing pathways in fracture repair, Bonekey Rep 2 (2013) 300. 22. A. Schindeler, M.M. McDonald, P. Bokko, D.G. Little, Bone remodeling during fracture repair: The cellular picture, Semin Cell Dev Biol 19(5) (2008) 459-66.

23. E.J. Mackie, Y.A. Ahmed, L. Tatarczuch, K.S. Chen, M. Mirams, Endochondral ossification: how cartilage is converted into bone in the developing skeleton, Int J Biochem Cell Biol 40(1) (2008) 46-62. 24. Y.Y. Yu, S. Lieu, C. Lu, T. Miclau, R.S. Marcucio, C. Colnot, Immunolocalization of BMPs, BMP antagonists, receptors, and effectors during fracture repair, Bone 46(3) (2010) 841-51. 25. L.J. van Baardewijk, J. van der Ende, S. Lissenberg-Thunnissen, L.M. Romijn, L.J. Hawinkels, C.F. Sier, I.B. Schipper, Circulating bone morphogenetic protein levels and delayed fracture healing, Int Orthop 37(3) (2013) 523-7. 26. D.J. Hadjidakis, Androulakis, II, Bone remodeling, Ann N Y Acad Sci 1092 (2006) 385-96. 27. S.L. Teitelbaum, Bone Resorption by Osteoclasts, Science 289(5484) (2000) 1504-1508. 28. P. Megas, Classification of non-union, Injury 36 Suppl 4 (2005) S30-7. 29. M.K. Sen, T. Miclau, Autologous iliac crest bone graft: should it still be the gold standard for treating nonunions?, Injury 38 Suppl 1 (2007) S75-80. 30. S.N.M.C. Khan, Frank P. Jr MD; Sandhu, Harvinder S. MD; Diwan, Ashish D. MD, PhD; Girardi, Federico P. MD; Lane, Joseph M. MD, The biology of bone grafting, Journal of the American Academy of Orthopaedic Surgeons. 13(1) (2005) 77-86. 31. J. Borrelli, Jr., W.D. Prickett, W.M. Ricci, Treatment of nonunions and osseous defects with bone graft and calcium sulfate, Clin Orthop Relat Res (411) (2003) 245-54. 32. N.K. Kanakaris, P.V. Giannoudis, The health economics of the treatment of long-bone nonunions, Injury 38 (2007) S77-S84. 33. E. Tsiridis, N. Upadhyay, P. Giannoudis, Molecular aspects of fracture healing: which are the important molecules?, Injury 38 Suppl 1 (2007) S11-25. 34. J.D. Boerckel, Y.M. Kolambkar, K.M. Dupont, B.A. Uhrig, E.A. Phelps, H.Y. Stevens, A.J. Garcia, R.E. Guldberg, Effects of protein dose and delivery system on BMP-mediated bone regeneration, Biomaterials 32(22) (2011) 5241-51. 35. S.N. Lissenberg-Thunnissen, D.J. de Gorter, C.F. Sier, I.B. Schipper, Use and efficacy of bone morphogenetic proteins in fracture healing, Int Orthop 35(9) (2011) 1271-80. 36. A.P. White, A.R. Vaccaro, J.A. Hall, P.G. Whang, B.C. Friel, M.D. McKee, Clinical applications of BMP-7/OP-1 in fractures, nonunions and spinal fusion, International Orthopaedics 31(6) (2007) 735-741. 37. D. Mines, Y. Gu, T.D. Kou, G.S. Cooper, Recombinant human bone morphogenetic protein-2 and pancreatic cancer: a retrospective cohort study, Pharmacoepidemiol Drug Saf 20(2) (2011) 111-8. 38. M. Chapellier, V. Maguer-Satta, BMP2, a key to uncover luminal breast cancer origin linked to pollutant effects on epithelial stem cells niche, Mol Cell Oncol 3(3) (2016) e1026527. 39. S.D. KR Garrison, J Ryder, I Shemilt, M Mugford, I Harvey, and F Song., Clinical effectiveness and cost-effectiveness of bone morphogenetic proteins in the non-healing of fractures and spinal fusion: a systematic review, Health Technol Assess 11(30) (2007). 40. W. Kang, D.-S. Lee, J.-H. Jang, Evaluation of Sustained BMP-2 Release Profiles Using a Novel Fluorescence-Based Retention Assay, PLOS ONE 10(4) (2015) e0123402. 41. M.T. Poldervaart, H. Wang, J. van der Stok, H. Weinans, S.C.G. Leeuwenburgh, F.C. Öner, W.J.A. Dhert, J. Alblas, Sustained Release of BMP-2 in Bioprinted Alginate for Osteogenicity in Mice and Rats, PLOS ONE 8(8) (2013) e72610. 42. M.M. Hurley, K. Marcello, C. Abreu, C.E. Brinckerhoff, C.C. Bowik, M.S. Hibbs, Transcriptional Regulation of the Collagenase Gene by Basic Fibroblast Growth Factor in Osteoblastic MC3T3- E1 Cells, Biochemical and Biophysical Research Communications 214(2) (1995) 331-339. 43. M.M. Hurley, K. Marcello, C. Abreu, M. Kessler, Signal transduction by basic fibroblast growth factor in rat osteoblastic py1a cells, Journal of Bone and Mineral Research 11(9) (1996) 1256- 1263. 44. A. Mansukhani, P. Bellosta, M. Sahni, C. Basilico, Signaling by Fibroblast Growth Factors (Fgf) and Fibroblast Growth Factor Receptor 2 (Fgfr2)–Activating Mutations Blocks Mineralization and Induces Apoptosis in Osteoblasts, The Journal of Cell Biology 149(6) (2000) 1297-1308. 45. C.J. Marie PJ, Hurley MM. , FGF and FGFR signaling in chondrodysplasias and craniosynostosis, Journal of cellular biochemistry 96(5) (2005) 888. 46. W. AO., Bad bones, absent smell, selfish testes: The pleiotropic consequences of human FGF receptor mutations, Cytokine & growth factor reviews 16(2) (2005) 187. 47. P.J. Marie, H. Miraoui, N. Severe, FGF/FGFR signaling in bone formation: progress and perspectives, Growth Factors 30(2) (2012) 117-23. 48. B.O. Wohrle S, Beluch N, Gaulis S, Stamm C, Scheibler M, Muller M, Kinzel B, Thuery A, Brueggen J, Hynes NE, Sellers WR, Hofmann F, Graus-Porta D. , FGF receptors control vitamin

D and phosphate homeostasis by mediating renal FGF-23 signaling and regulating FGF-23 expression in bone, Journal of bone and mineral research 26(10) (2011) 2486. 49. A. Montero, Y. Okada, M. Tomita, M. Ito, H. Tsurukami, T. Nakamura, T. Doetschman, J.D. Coffin, M. Hurley, Disruption of the fibroblast growth factor-2 gene results in decreased bone mass and bone formation, Journal of Clinical Investigation 105(8) (2000) 1085-1093. 50. S.T. Xiao L, Esliger A, Kronenberg MS, Coffin JD, Doetschman T, Hurley MM., Disruption of the Fgf2 gene activates the adipogenic and suppresses the osteogenic program in mesenchymal marrow stromal stem cells, Bone 47(2) (2010) 360. 51. A.D. Sabbieti MG, Xiao L, Marchetti L, Coffin JD, Doetschman T, Hurley MM. , Endogenous FGF-2 is critically important in PTH anabolic effects on bone, Journal of cellular physiology 219(1) (2009) 143. 52. S.M. Behr B, Manu A, Lehnhardt M, Longaker MT, Quarto N., Fgf-18 is required for osteogenesis but not angiogenesis during long bone repair, Tissue Engineering Part A 17(15– 16) (2011) 2061. 53. G. Chen, C. Deng, Y.-P. Li, TGF-β and BMP Signaling in Osteoblast Differentiation and Bone Formation, International Journal of Biological Sciences 8(2) (2012) 272-288. 54. X. Guo, X.-F. Wang, Signaling cross-talk between TGF-β/BMP and other pathways, Cell research 19(1) (2009) 71-88. 55. X.-H.F.a.R. Derynck, SPECIFICITY AND VERSATILITY IN TGF-β SIGNALING THROUGH SMADS, Annual Review of Cell and Developmental Biology 21(1) (2005) 659-693. 56. C. Bernabeu, J.M. Lopez-Novoa, M. Quintanilla, The emerging role of TGF-β superfamily coreceptors in cancer, Biochimica et Biophysica Acta (BBA) - Molecular Basis of Disease 1792(10) (2009) 954-973. 57. J.L. Crane, L. Xian, X. Cao, Role of TGF-β Signaling in Coupling Bone Remodeling, in: X.-H. Feng, P. Xu, X. Lin (Eds.), TGF-β Signaling: Methods and Protocols, Springer New York, New York, NY, 2016, pp. 287-300. 58. Y. Tang, X. Wu, W. Lei, L. Pang, C. Wan, Z. Shi, L. Zhao, T.R. Nagy, X. Peng, J. Hu, X. Feng, W. Van Hul, M. Wan, X. Cao, TGF-β1-induced Migration of Bone Mesenchymal Stem Cells Couples Bone Resorption and Formation, Nature medicine 15(7) (2009) 757-765. 59. R. Derynck, R.J. Akhurst, Differentiation plasticity regulated by TGF-[beta] family proteins in development and disease, Nat Cell Biol 9(9) (2007) 1000-1004. 60. T. Yasui, Y. Kadono, M. Nakamura, Y. Oshima, T. Matsumoto, H. Masuda, J. Hirose, Y. Omata, H. Yasuda, T. Imamura, K. Nakamura, S. Tanaka, Regulation of RANKL-induced osteoclastogenesis by TGF-β through molecular interaction between Smad3 and Traf6, Journal of Bone and Mineral Research 26(7) (2011) 1447-1456. 61. C.-F. Lai, S.-L. Cheng, Signal Transductions Induced by Bone Morphogenetic Protein-2 and Transforming Growth Factor-β in Normal Human Osteoblastic Cells, Journal of Biological Chemistry 277(18) (2002) 15514-15522. 62. D. Kovacevic, A.J. Fox, A. Bedi, L. Ying, X.-H. Deng, R.F. Warren, S.A. Rodeo, Calcium- Phosphate Matrix With or Without TGF-β3 Improves Tendon-Bone Healing After Rotator Cuff Repair, The American Journal of Sports Medicine 39(4) (2011) 811-819. 63. M. Beederman, J.D. Lamplot, G. Nan, J. Wang, X. Liu, L. Yin, R. Li, W. Shui, H. Zhang, S.H. Kim, W. Zhang, J. Zhang, Y. Kong, S. Denduluri, M.R. Rogers, A. Pratt, R.C. Haydon, H.H. Luu, J. Angeles, L.L. Shi, T.-C. He, BMP signaling in mesenchymal stem cell differentiation and bone formation, Journal of biomedical science and engineering 6(8A) (2013) 32-52. 64. C.A. Kirker-Head, Potential applications and delivery strategies for bone morphogenetic proteins, Advanced Drug Delivery Reviews 43(1) (2000) 65-92. 65. M.M.L. Deckers, R.L. van Bezooijen, G. van der Horst, J. Hoogendam, C. van der Bent, S.E. Papapoulos, C.W.G.M. Löwik, Bone Morphogenetic Proteins Stimulate Angiogenesis through Osteoblast-Derived Vascular Endothelial Growth Factor A, Endocrinology 143(4) (2002) 1545- 1553. 66. C. He, X. Chen, Transcription regulation of the vegf gene by the BMP/Smad pathway in the angioblast of zebrafish embryos, Biochemical and Biophysical Research Communications 329(1) (2005) 324-330. 67. R. Bouillon, A. Prodonova, Growth hormone deficiency and peak bone mass, Journal of Pediatric Endocrinology and Metabolism 13(SUPPL. 6) (2000) 1327-1336. 68. A.M. Boot, M.A.M.J. Engels, G.J.M. Boerma, E.P. Krenning, S.M.P.F. De Muinck Keizer- Schrama, Changes in bone mineral density, body composition, and lipid metabolism during growth hormone (GH) treatment in children with GH deficiency, Journal of Clinical Endocrinology and Metabolism 82(8) (1997) 2423-2428.

69 A. Giustina, G. Mazziotti, E. Canalis, Growth hormone, insulin-like growth factors, and the skeleton, Endocrine Reviews 29(5) (2008) 535-559. 70. J. Fiedler, C. Brill, W.F. Blum, R.E. Brenner, IGF-I and IGF-II stimulate directed cell migration of bone-marrow-derived human mesenchymal progenitor cells, Biochemical and Biophysical Research Communications 345(3) (2006) 1177-1183. 71. A. Lagumdzija, G. Ou, M. Petersson, E. Bucht, A. Gonon, Y. Pernow, Inhibited anabolic effect of insulin-like growth factor-I on stromal bone marrow cells in endothelial nitric oxide synthase- knockout mice, Acta Physiologica Scandinavica 182(1) (2004) 29-35. 72. M. Uutela, M. Wirzenius, K. Paavonen, I. Rajantie, Y. He, T. Karpanen, M. Lohela, H. Wiig, P. Salven, K. Pajusola, U. Eriksson, K. Alitalo, PDGF-D induces macrophage recruitment, increased interstitial pressure, and blood vessel maturation during angiogenesis, Blood 104(10) (2004) 3198-204. 73. H. Gerhardt, H. Semb, Pericytes: gatekeepers in tumour cell metastasis?, Journal of Molecular Medicine 86(2) (2008) 135-144. 74. A. Armulik, A. Abramsson, C. Betsholtz, Endothelial/Pericyte Interactions, Circulation Research 97(6) (2005) 512. 75. A.I. Caplan, D. Correa, PDGF in bone formation and regeneration: New insights into a novel mechanism involving MSCs, Journal of Orthopaedic Research 29(12) (2011) 1795-1803. 76. J. Hollinger, P. Dds, C. Hart, S. Hirsch, M. Bes, S. Lynch, D. Dmd, G. Friedlaender, Recombinant Human Platelet-Derived Growth Factor: Biology and Clinical Applications, Journal of Bone and Joint Surgery 90(1) (2008) 48-54. 77. F.G. Rocha, C.A. Sundback, N.J. Krebs, J.K. Leach, D.J. Mooney, S.W. Ashley, J.P. Vacanti, E.E. Whang, The effect of sustained delivery of vascular endothelial growth factor on angiogenesis in tissue-engineered intestine, Biomaterials 29(19) (2008) 2884-90. 78. M. Orlandini, A. Spreafico, M. Bardelli, M. Rocchigiani, A. Salameh, S. Nucciotti, C. Capperucci, B. Frediani, S. Oliviero, Vascular endothelial growth factor-D activates VEGFR-3 expressed in osteoblasts inducing their differentiation, J Biol Chem 281(26) (2006) 17961-7. 79. H. Peng, A. Usas, A. Olshanski, A.M. Ho, B. Gearhart, G.M. Cooper, J. Huard, VEGF improves, whereas sFlt1 inhibits, BMP2-induced bone formation and bone healing through modulation of angiogenesis, J Bone Miner Res 20(11) (2005) 2017-27. 80. J. Street, M. Bao, L. deGuzman, S. Bunting, F.V. Peale, N. Ferrara, H. Steinmetz, J. Hoeffel, J.L. Cleland, A. Daugherty, N. van Bruggen, H.P. Redmond, R.A.D. Carano, E.H. Filvaroff, Vascular endothelial growth factor stimulates bone repair by promoting angiogenesis and bone turnover, Proceedings of the National Academy of Sciences 99(15) (2002) 9656-9661. 81. C. Lu, E. Hansen, A. Sapozhnikova, D. Hu, T. Miclau, R.S. Marcucio, Effect of age on vascularization during fracture repair, J Orthop Res 26(10) (2008) 1384-9. 82. D. Park, J.A. Spencer, B.I. Koh, T. Kobayashi, J. Fujisaki, T.L. Clemens, C.P. Lin, H.M. Kronenberg, D.T. Scadden, Endogenous bone marrow MSCs are dynamic, fate-restricted participants in bone maintenance and regeneration, Cell Stem Cell 10(3) (2012) 259-72. 83. Y. Kfoury, D.T. Scadden, Mesenchymal cell contributions to the stem cell niche, Cell Stem Cell 16(3) (2015) 239-53. 84. D.T. Scadden, The stem-cell niche as an entity of action, Nature 441(7097) (2006) 1075-9. 85. A.I. Caplan, Adult Mesenchymal Stem Cells: When, Where, and How, Stem Cells Int 2015 (2015) 628767. 86. A.I. Caplan, Mesenchymal stem cells, Journal of Orthopaedic Research 9(5) (1991) 641-650. 87. M. Dominici, K. Le Blanc, I. Mueller, I. Slaper-Cortenbach, F. Marini, D. Krause, R. Deans, A. Keating, D. Prockop, E. Horwitz, Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement, Cytotherapy 8(4) (2006) 315-7. 88. K.N. Malizos, L.K. Papatheodorou, The healing potential of the periosteum molecular aspects, Injury 36 Suppl 3 (2005) S13-9. 89. C. Farrington-Rock, N.J. Crofts, M.J. Doherty, B.A. Ashton, C. Griffin-Jones, A.E. Canfield, Chondrogenic and Adipogenic Potential of Microvascular Pericytes, Circulation 110(15) (2004) 2226-2232. 90. Y. Tintut, Z. Alfonso, T. Saini, K. Radcliff, K. Watson, K. Boström, L.L. Demer, Multilineage Potential of Cells From the Artery Wall, Circulation 108(20) (2003) 2505-2510. 91. A. Usas, J. Huard, Muscle-derived stem cells for tissue engineering and regenerative therapy, Biomaterials 28(36) (2007) 5401-5406.

92. R. Tasso, V. Ulivi, D. Reverberi, C. Lo Sicco, F. Descalzi, R. Cancedda, In vivo implanted bone marrow-derived mesenchymal stem cells trigger a cascade of cellular events leading to the formation of an ectopic bone regenerative niche, Stem Cells Dev 22(24) (2013) 3178-91. 93. Y. Bi, D. Ehirchiou, T.M. Kilts, C.A. Inkson, M.C. Embree, W. Sonoyama, L. Li, A.I. Leet, B.-M. Seo, L. Zhang, S. Shi, M.F. Young, Identification of tendon stem/progenitor cells and the role of the extracellular matrix in their niche, Nat Med 13(10) (2007) 1219-1227. 94. S.-H. Kim, Y.-S. Kim, S.-Y. Lee, K.-H. Kim, Y.-M. Lee, W.-K. Kim, Y.-K. Lee, Gene expression profile in mesenchymal stem cells derived from dental tissues and bone marrow, Journal of Periodontal & Implant Science 41(4) (2011) 192-200. 95. D.R. Lemos, C. Eisner, C.I. Hopkins, F.M.V. Rossi, Skeletal muscle-resident MSCs and bone formation, Bone 80 (2015) 19-23. 96. D. Kouroupis, S.M. Churchman, A. English, P. Emery, P.V. Giannoudis, D. McGonagle, E.A. Jones, Assessment of umbilical cord tissue as a source of mesenchymal stem cell/endothelial cell mixtures for bone regeneration, Regenerative Medicine 8(5) (2013) 569-581. 97. P.A. Zuk, M. Zhu, P. Ashjian, D.A. De Ugarte, J.I. Huang, H. Mizuno, Z.C. Alfonso, J.K. Fraser, P. Benhaim, M.H. Hedrick, Human Adipose Tissue Is a Source of Multipotent Stem Cells, Molecular Biology of the Cell 13(12) (2002) 4279-4295. 98. D.L. Worthley, M. Churchill, J.T. Compton, Y. Tailor, M. Rao, Y. Si, D. Levin, M.G. Schwartz, A. Uygur, Y. Hayakawa, S. Gross, B.W. Renz, W. Setlik, A.N. Martinez, X. Chen, S. Nizami, H.G. Lee, H.P. Kang, J.M. Caldwell, S. Asfaha, C.B. Westphalen, T. Graham, G. Jin, K. Nagar, H. Wang, M.A. Kheirbek, A. Kolhe, J. Carpenter, M. Glaire, A. Nair, S. Renders, N. Manieri, S. Muthupalani, J.G. Fox, M. Reichert, A.S. Giraud, R.F. Schwabe, J.P. Pradere, K. Walton, A. Prakash, D. Gumucio, A.K. Rustgi, T.S. Stappenbeck, R.A. Friedman, M.D. Gershon, P. Sims, T. Grikscheit, F.Y. Lee, G. Karsenty, S. Mukherjee, T.C. Wang, Gremlin 1 identifies a skeletal stem cell with bone, cartilage, and reticular stromal potential, Cell 160(1-2) (2015) 269-84. 99. T. Aghaloo, C.M. Cowan, X. Zhang, E. Freymiller, C. Soo, B. Wu, K. Ting, Z. Zhang, The effect of NELL1 and bone morphogenetic protein-2 on calvarial bone regeneration, J Oral Maxillofac Surg 68(2) (2010) 300-8. 100. M.-T. Tsai, D.-J. Lin, S. Huang, H.-T. Lin, W.H. Chang, Osteogenic differentiation is synergistically influenced by osteoinductive treatment and direct cell–cell contact between murine osteoblasts and mesenchymal stem cells, International Orthopaedics 36(1) (2012) 199- 205. 101. L. Zhi, C. Chen, X. Pang, H. Uludag, H. Jiang, Synergistic effect of recombinant human bone morphogenic protein-7 and osteogenic differentiation medium on human bone-marrow-derived mesenchymal stem cells in vitro, International Orthopaedics 35(12) (2011) 1889-1895. 102. T.N. Gardner, T. Stoll, L. Marks, S. Mishra, M. Knothe Tate, The influence of mechanical stimulus on the pattern of tissue differentiation in a long bone fracture — an FEM study, Journal of Biomechanics 33(4) (2000) 415-425. 103. L.E. Claes, C.A. Heigele, Magnitudes of local stress and strain along bony surfaces predict the course and type of fracture healing, Journal of Biomechanics 32(3) (1999) 255-266. 104. E.F. Morgan, K.T. Salisbury Palomares, R.E. Gleason, D.L. Bellin, K.B. Chien, G.U. Unnikrishnan, P.L. Leong, Correlations between Local Strains and Tissue Phenotypes in an Experimental Model of Skeletal Healing, Journal of biomechanics 43(12) (2010) 2418-2424. 105. P. Bianco, P.G. Robey, P.J. Simmons, Mesenchymal Stem Cells: Revisiting History, Concepts, and Assays, Cell Stem Cell 2(4) (2008) 313-319. 106. M. Crisan, S. Yap, L. Casteilla, C.-W. Chen, M. Corselli, T.S. Park, G. Andriolo, B. Sun, B. Zheng, L. Zhang, C. Norotte, P.-N. Teng, J. Traas, R. Schugar, B.M. Deasy, S. Badylak, H.-J. Bűhring, J.-P. Giacobino, L. Lazzari, J. Huard, B. Péault, A Perivascular Origin for Mesenchymal Stem Cells in Multiple Human Organs, Cell Stem Cell 3(3) (2008) 301-313. 107. P. Bianco, X. Cao, P.S. Frenette, J.J. Mao, P.G. Robey, P.J. Simmons, C.Y. Wang, The meaning, the sense and the significance: translating the science of mesenchymal stem cells into medicine, Nat Med 19(1) (2013) 35-42. 108. Y. Nakamura, F. Arai, H. Iwasaki, K. Hosokawa, I. Kobayashi, Y. Gomei, Y. Matsumoto, H. Yoshihara, T. Suda, Isolation and characterization of endosteal niche cell populations that regulate hematopoietic stem cells, Blood 116(9) (2010) 1422-32. 109. S. Mendez-Ferrer, T.V. Michurina, F. Ferraro, A.R. Mazloom, B.D. Macarthur, S.A. Lira, D.T. Scadden, A. Ma'ayan, G.N. Enikolopov, P.S. Frenette, Mesenchymal and haematopoietic stem cells form a unique bone marrow niche, Nature 466(7308) (2010) 829-34.

110. Bo O. Zhou, R. Yue, Malea M. Murphy, J.G. Peyer, Sean J. Morrison, Leptin-Receptor- Expressing Mesenchymal Stromal Cells Represent the Main Source of Bone Formed by Adult Bone Marrow, Cell Stem Cell 15(2) 154-168. 111. C.K. Chan, E.Y. Seo, J.Y. Chen, D. Lo, A. McArdle, R. Sinha, R. Tevlin, J. Seita, J. Vincent- Tompkins, T. Wearda, W.J. Lu, K. Senarath-Yapa, M.T. Chung, O. Marecic, M. Tran, K.S. Yan, R. Upton, G.G. Walmsley, A.S. Lee, D. Sahoo, C.J. Kuo, I.L. Weissman, M.T. Longaker, Identification and specification of the mouse skeletal stem cell, Cell 160(1-2) (2015) 285-98. 112. D. Tang, R.S. Tare, L.Y. Yang, D.F. Williams, K.L. Ou, R.O. Oreffo, Biofabrication of bone tissue: approaches, challenges and translation for bone regeneration, Biomaterials 83 (2016) 363-82. 113. M. Farokhi, F. Mottaghitalab, M.A. Shokrgozar, K.-L. Ou, C. Mao, H. Hosseinkhani, Importance of dual delivery systems for bone tissue engineering, Journal of Controlled Release 225 (2016) 152-169. 114. R. Langer, Perspectives and Challenges in Tissue Engineering and Regenerative Medicine, Advanced Materials 21(32-33) (2009) 3235-3236. 115. I.K. Ko, S.J. Lee, A. Atala, J.J. Yoo, In situ tissue regeneration through host stem cell recruitment, Exp Mol Med 45 (2013) e57. 116. Q. Li, L. Ma, C. Gao, Biomaterials for in situ tissue regeneration: development and perspectives, Journal of Materials Chemistry B 3(46) (2015) 8921-8938. 117. S. Bose, M. Roy, A. Bandyopadhyay, Recent advances in bone tissue engineering scaffolds, Trends in Biotechnology 30(10) (2012) 546-554. 118. E. Piskin, Biodegradable polymers as biomaterials, Journal of Biomaterials Science, Polymer Edition 6(9) (1995) 775-795. 119. P.V. Giannoudis, H. Dinopoulos, E. Tsiridis, Bone substitutes: an update, Injury 36 Suppl 3 (2005) S20-7. 120. P.V. Giannoudis, T.A. Einhorn, D. Marsh, Fracture healing: The diamond concept, Injury 38 (2007) S3-S6. 121. P.V. Giannoudis, T.A. Einhorn, G. Schmidmaier, D. Marsh, The diamond concept – open questions, Injury 39 (2008) S5-S8. 122. Y.Y. Brahatheeswaran Dhandayuthapani, Toru Maekawa, and D. Sakthi Kumar., Polymeric Scaffolds in Tissue Engineering Application: A Review, International Journal of Polymer Science 2011 (2011). 123. S. Stratton, N.B. Shelke, K. Hoshino, S. Rudraiah, S.G. Kumbar, Bioactive polymeric scaffolds for tissue engineering, Bioactive Materials 1(2) (2016) 93-108. 124. M. Ehrbar, S.C. Rizzi, R. Hlushchuk, V. Djonov, A.H. Zisch, J.A. Hubbell, F.E. Weber, M.P. Lutolf, Enzymatic formation of modular cell-instructive fibrin analogs for tissue engineering, Biomaterials 28(26) (2007) 3856-66. 125. M.P. Lutolf, P.M. Gilbert, H.M. Blau, Designing materials to direct stem-cell fate, Nature 462(7272) (2009) 433-441. 126. S. Metzger, P.S. Lienemann, C. Ghayor, W. Weber, I. Martin, F.E. Weber, M. Ehrbar, Modular poly(ethylene glycol) matrices for the controlled 3D-localized osteogenic differentiation of mesenchymal stem cells, Adv Healthc Mater 4(4) (2015) 550-8. 127. M. Ehrbar, S.C. Rizzi, R.G. Schoenmakers, B.S. Miguel, J.A. Hubbell, F.E. Weber, M.P. Lutolf, Biomolecular hydrogels formed and degraded via site-specific enzymatic reactions, Biomacromolecules 8(10) (2007) 3000-7. 128. M. Ehrbar, A. Sala, P. Lienemann, A. Ranga, K. Mosiewicz, A. Bittermann, S.C. Rizzi, F.E. Weber, M.P. Lutolf, Elucidating the role of matrix stiffness in 3D cell migration and remodeling, Biophys J 100(2) (2011) 284-93. 129. E. Ratcliffe, K.E. Glen, M.W. Naing, D.J. Williams, Current status and perspectives on stem cell-based therapies undergoing clinical trials for regenerative medicine: case studies, Br Med Bull 108 (2013) 73-94. 130. P.S. Lienemann, S. Metzger, A.S. Kivelio, A. Blanc, P. Papageorgiou, A. Astolfo, B.R. Pinzer, P. Cinelli, F.E. Weber, R. Schibli, M. Behe, M. Ehrbar, Longitudinal in vivo evaluation of bone regeneration by combined measurement of multi-pinhole SPECT and micro-CT for tissue engineering, Sci Rep 5 (2015) 10238.

Chapter III The role of biomaterials delivered SDF-1α in skeletal stem cell homing and bone regeneration

Abstract

SDF-1 signals to induce the migration of stem and progenitorα expression cells to is sites considered of tissue as repair one of andthe regeneration. In this study, we aimed at determining if biomaterials delivery of SDF-1 and BMP-2 can facilitate recruitment of tissue stem/progenitor cells and bone formation.α Our results indicated that PEG- hydrogels-based dual delivery of BMP-2 and SDF- bone more efficiently than delivery of the individual1α factors. induced This the resulted augmentation in induced of native bone augmentation, demonstrating a potential osteogenic differentiation stimulus upon the periosteal cell layer. Furthermore, our findings indicated that PEG-hydrogel delivered

SDF- -union bone defects of the murine skull is unable to promote BMP-2- mediated1α in bone non regeneration as well as recruitment of Sca1+ SSCs. Taken together, our research suggests that during bone healing SDF- might have distinct functions in the recruitment, proliferation and differentiations of different1α cellular fractions.

Introduction

Bone morphogenetic protein (BMP) has been shown to be a growth factor that can promote the formation of bone. However, as sole treatment BMP is required in supraphysiological dose, restricting its clinical applicability. The mobilization of skeletal stem and progenitor cells is considered as an important concomitant parameter influencing the efficiency of bone regeneration and could be critical in the regeneration of non-union bone defects (1). Current data suggest that stromal cell-derived factor-1 (SDF- 1 or CXCL12) and its G-protein-coupled main receptor CXCR-4 form an essential axis (SDF-1/CXCR-4) for the recruitment of MSCs expressing CXCR-4 during endochondral fracture healing (2) and the acute phase of bone repair (3). Consistent with this, treatment with an anti-SDF-1 antagonist or genetic depletion of SDF-1 or CXCR-4 resulted in impaired fracture healing (4). Additionally, hypoxia has been shown to enhance CXCR4 expression in association with HIF-1 (5, 6), which might be a possible mechanism leading to the upregulated SDF-1 expressionα in the periosteum and at the boarders of fracture sites (7). Furthermore, perturbing the SDF-1/CXCR4 signaling axis in primary human and mouse bone marrow derived MSCs interrupt their BMP2-mediated differentiation towards osteogenic lineage (3). Together these recent findings indicate that SDF-1 in addition to mobilizing MSCs to sites of tissue regeneration could have a role in modulating their lineage specific differentiation.

Applying the chemoattractant and osteogenic properties of SDF-1 and BMP-2, respectively, several studies have already investigated their combination towards bone regeneration (8-13). Hwang and colleagues demonstrated that treatments with collagen scaffold sequentially releasing SDF-1 followed by BMP-2 exhibited more robust bone regeneration as compared to simultaneous treatment with both BMP-2 and SDF-1 (9). These results indicate that SDF-1 might improve the mobilization and local enrichment of MSCs, which can subsequently undergo osteogenic differentiation in response to low doses of BMP-2. They further suggest that the SDF-1/CXCR4 axis is of minor importance in promoting osteogenic signaling. However, in another study, the osteogenic induction by suboptimal concentrations of BMP-2 was potentiated in a dose-dependent manner by SDF- and inhibited by blocking of CXCR4/SDF-1 signaling axis, suggesting a role of SDF-

in1β BMP-2 osteogenic signaling (8). 1β

Novel fibroin-nanohydroxyapatite scaffolds comprising silk fibroin microspheres with encapsulated BMP-2 and physically adsorbed SDF-1 resulted in enhanced bone regeneration when compared to BMP-2 or SDF-1 only (12). Finally, in vivo experiments by Higashino et al. suggested that SDF-1 enhanced BMP-2-induced mobilization of circulating bone marrow-derived osteoprogenitor cells resulting in increased osteoprogenitor differentiation and ectopic bone formation in collagen discs. Unfortunately, in this study neither direct effect of SDF-1, SDF-1 release kinetics after implantation nor recruitment of local skeletal stem/progenitor cells were evaluated (10).

Therefore, the aim of this study was to determine if biomaterials-delivered SDF-1 can stimulate the recruitment of local bone stem/progenitor cells and by this promote the BMP-2-mediated regeneration of non-union bone defects. To dissect biomaterials and growth factor-mediated effects during the regeneration of bone we have earlier employed synthetic poly (ethylene)glycol (PEG)-based hydrogels that comprise the cell adhesion site RGD as well as MMP-1 sites. These provisional PEG-based materials have previously been optimized towards bone regeneration and have shown excellent properties when used for the delivery of BMP-2 (14, 15).

Herein we show evidence that combined fast delivery of SDF- and BMP-2 from PEG- hydrogel promotes the augmentation of native bone more efficiently1α than BMP-2 or SDF- monotherapy. Interestingly, this effect could not be reproduced in a non-union bone

1αdefect model using PEG-hydrogels or collagen sponges as growth factor delivery tools. Additionally, while BMP-2 induced the mobilization of skeletal stem/progenitor cells, SDF- had not detectable effect on these cells. Collectively, our data indicate that SDF-

; instead1α of promoting the mobilization of skeletal stem/progenitor cells, supports the 1αinduction of periosteal cells and enhances their osteogenic potential.

Results

SDF-

40 dose BMP-2 (0.2 µg) or a combination thereof were placed on top of the defect covering fibrous tissue layer. 4 weeks later the calvarial bones were harvested and bone formation was assessed by micro-CT and histology (Figure 1A, B). We show that neither the delivery of the individual factors nor the combined delivery promoted the regeneration of non- union defects (Figure 1B). Since the implantation of the scaffolds was achieved with a 2- step operation and the connective tissue generated within the defects was not removed, this led to an instable positioning and shifting of the hydrogels after implantation onto the native bone. Interestingly, the native bone was augmented by treatment with hydrogels released SDF- and BMP-2, while solo treatment with BMP-2 and SDF- had no impact in bone regeneration1α . SDF- ng hydrogels were fully degraded,1α while control hydrogels and BMP-2 releasing1α hydrogels releasi remained intact. (Figure 1B, indicated by dash line).

Figure 1. Assessment of dual delivery of BMP-2 and SDF-

42 generated bone volume, connectivity and % coverage of the defected area. Data is depicted as mean ± SD, n = 6. One- PEG-based hydrogels loaded with BMP-2 and SDF- way ANOVA Bonferroni’s post hoc test, no significance.

Figure 2. Assessment of dual delivery of BMP-2 and SDF- in 2-step operations. A) 8-weeks post-craniotomy and 4-weeks post-implantation bone formation was assed ex situ. Representative top views of 3D surface rendered micro-CT1α inmeasurements orthotopic bone (sca healingle bar: 1 mm, left panel). B) H&E staining of coronal cross sections, asterisks indicating the remaining volume of the hydrogels (scale

C) Quantitative assessment of generated bone volume, connectivity and % coverage. Data is depicted as mean ± SD, n = 4. One- bar: 200 μm). Effect of dual delivery BMP-2way ANOVA and Bonferroni’s SDF- post hoc test, no significance.

Figure 3. Assessment of dual delivery of BMP-2 and SDF- A) 4-weeks post- implantation bone formation was assed ex situ. B) Representative coronal cross sections of 3D surface rendered micro-CT measurements (scale bar: 1 mm, left 1αpanel) upon C) murine Quantitative calvarial assessment bone. of generated bone volume, connectivity and % coverage. Data is depicted as mean ± SD, n = 4. One-way ANOVA Bonferro * p < 0.05.

ni’s post hoc test, 46

Onlay bone augmentation of murine calvarial bone after synergetic effect of BMP-2 and SDF-

Injury-associated Sca1+SSCs recruitment under SDF-

Discussion

In this study, we present data indicating that SDF- -2 mediated augmentation of native, uninjured bone. Our current1α data can suggestspromote thatthe BMPthis effect is due to the induction of periosteum-derived stem/progenitor cells rather than the recruitment of Sca1+ SSCs from bone marrow or perivascular origin.

We employed an earlier established (Chapter II) PEG-hydrogel cell-trap approach for determining the role of SDF- -2 on the recruitment of healing associated Sca1+

SSCs and their potential contribution1α and BMP to the healing of non-union bone defects. The used soft (250Pa) biocompatible hydrogel materials were designed to be biodegradable, and to provide the cell adhesion site RGD, which are minimal criteria to allow the infiltration of healing-associated cells. Due to their hydrophilic and non-fouling properties, these hydrogels, in absence of engineered growth factor affinity sites, do not significantly bind and retain admixed native growth factors. These hydrogels therefore offer the advantage of evaluating cytokine functions under fully controlled environmental conditions and enable the dissection of individual experimental parameters.

Non-union bone defects remain one of the clinical challenges that needs to be overcome. In such defects a fibrous tissue is formed, which under normal circumstances will persist and even prevent a later initiation of natural bone healing (1). In order to imitate non- union defects, herein we established a critical-sized defect in the parietal bone of the skull and allowed the healing to occur for four weeks, enabling the formation of a fibrous tissue across the bone defect and the demission of the operation induced inflammatory response. We reason that in this model the microenvironmental conditions are not in favor of bone formation and that the initiation of bone formation will critically depend on the recruitment or induction of skeletal stem/progenitor cells. In this model, we showed that the dual delivery of BMP-2 and SDF- -based matrices could not induce bone healing. In contrast, when hydrogels1α containing from PEG low concentrations of BMP-2 and SDF- upon the periosteum of intact bones, new bone structures were initiated1α were on top placed of this mature bones. Our data indicate that the formation of new bone was more pronounced in presence of co-treatments compared to solo BMP-2 treatment, suggesting that during bone healing SDF-1 ive role on the

α might have a support

49 recruitment, proliferation and differentiation of osteogenic cellular fractions upon treatment of native calvarial bone.

In contrast to PEG hydrogels, collagen, one of the main classes of structural extracellular matrix (ECM) proteins exhibits multiple inherent biological properties such as variable cell-binding sites, biodegradability making it an excellent substrate for the delivery of growth factors and cytokines (20, 21). In this study, we demonstrated that the delivery of both SDF- -2 from collagen matrices was superior to their delivery from PEG hydrogels.1α A differenceand BMP which can be due to multiple reasons such as a change in cellular infiltration, selective integrin-dependent infiltration of cell populations, rate of growth factor delivery, rate of biomaterial degradation or changes in growth factor integrin co- signaling. As especially in collagen matrices these effects are highly interrelated they cannot be easily dissected and would clearly need to be evaluated in more detail.

BMP-2 has long been established as a factor that induces bone formation by the differentiation of mesenchymal stem as well as bone progenitor cells (1). However, very recent studies now provide evidence that BMP-2 could also promote the induction and mobilization of SSCs (22). In contrast, the SDF- involved in the recruitment of multiple stem1α/CXCR4 and progenitor axis has been cell typesdescribed including to be hematopoietic stem cells, endothelial progenitor cells, and healing associated stromal cells (19, 23). Only recently it has been suggested that SDF- improve the recruitment of MSCs and consequently to boost the1α regeneration could be employed of bone toin presence of BMP-2. In first preclinical trials, data regarding the role of biomaterials delivered SDF- -2-mediated healing of bone defects were contradictory. While in some studies1α SDF- in BMP -2 induced bone formation, in others studies the effect was not significant1α promoted or even inhibitory BMP (8, 9). To further explore a potential in vivo role of SDF- Sca1+ SSCs to healing bones, here fast releasing SDF-1α in mediating the migration of calvarial critical-sized defects, employed to 1α entrap hydrogels all healing were implanted participating in murine cells and characterize various cell populations by FACS at 8 days of treatment. Our data indicate that consistent with the literature BMP-2 stimulated, while SDF- t in the number of Sca1+ SSCs present in the healing bone defects. Interestingly,1α had no after effec 8 weeks of implantation SDF- releasing hydrogels were fully degraded and replaced by recruited cells, while control1α hydrogels and BMP-2 releasing hydrogels remained intact. These data indicate that the

50 mobilization and proliferation of cells distinct from SSCs were induced by SDF- treatment. 1α

In this study, the focus was on examining SDF- Sca1+ SSCs recruitment and bone formation at constant concentrations, therefore1α effects concentration on dependent effects cannot be excluded. In fact, a very recent study showing that improved bone healing was only obtained if SDF- -2 were delivered in a controlled and sequential manner implies that attention1α should and BMP be paid to the exact mode of delivery (12, 17). Although herein employed hydrogel can be engineered towards controlled delivery of growth factors and cytokines both factors SDF- -2 were simply admixed to the hydrogel, resulting in passive diffusion. Therefore,1α and BMPto fast release kinetics of SDF- rapid loss of the stem cell recruitment activity could be a reason for the failure1α and of combined treatment to improve healing of bone defects. Further evaluations of growth factor release rates under in vivo conditions, as well as using refined growth factor release properties would be required. Additionally, to fully understand SDF- -mediated cellular and molecular mechanisms contributing to BMP-mediated bone regeneration1α further investigation should be conducted.

Conclusions

While the role of SDF- in vivo bone formation remains to be fully understood,1α we signaling demonstrated in the complexitiesthat dual delivery of of BMP-2 and SDF- lead to bone augmentation upon treatment of native calvarial bone via PEG-based1α could hydrogel application. In this study, the release of growth factors has not been engineered and in vivo release kinetics have not been determined. Therefore, it remains unknown if dual delivery of BMP-2 and SDF- was ineffective due to the mode of delivery or due to the inability of healing associated1α to stem/progenitors respond to the provided signals.

Materials and Methods

Formation of scaffolds: TG-PEG hydrogels were formed by FXIIIa cross-linking of equimolar blends of two eight-arm PEG macromers (8-PEG-MMPsensitive-Lys and 8-PEG- Gln) as previously described (24, 25). Briefly, FXIII (200 U mL 1, Fibrogammin P, CSL Behring, Switzerland) was activated with thrombin (2 U − mL 1, Sigma Aldrich, − – Switzerland)solutions of 8-PEG- for 30Gln min and at 8-PEG- 37 °CMMP andsensitive stored-Lys at were −80 prepared °C. Stoichiometrically in Tris buffer balanced(50 mM, pH 7.6) and calcium chloride (50 mM). The cross-linking reaction was initiated by addition of 10 U mL 1 FXIIIa followed by vigorous mixing. Disc-shaped hydrogels were obtained by cross-linking− reaction for 30 min at 37 °C. Absorbable collagen sponges (ACS, Lyostypt®) were shaped in discs by 3 mm biopsy punches (Kai Medical), which after absorption resulted in 4 mm diameter discs.

Animals:Experiments were performed using 6-7-week-old female C57BL/6JOlaHsd. All animal research procedures were approved by the Animal Experimentation Committee of the Veterinary Office of the Canton of Zurich, Switzerland and followed the guidelines of the Swiss Federal Veterinary Office for the use and care of laboratory animals.

Orthotopic bone formation in calvarial defect model:Two craniotomies (4mm diameter) in both parietal bones of the skull on each side of the sagittal suture were created. 4-weeks after craniotomy, 11µl TG-PEG hydrogel discs were pre-formed in presence of 0, 0.2 µg recombinant human BMP-2 (CHO cell derived, PeproTech) supplemented with or without 1µg recombinant murine SDF-1 (E.coli derived,

PeproTech) and applied upon the critical-sized defects. The biomimeticα hydrogel discs were released from one glass side and stored in humidified atmosphere until transplantation. Immobilization of the hydrogels upon the defects was accomplished with fibrin as a provisional gluing material. After craniotomy, the skin was closed with 6.0 Vicryl sutures (Ethicon) and mice were sacrificed for ex vivo evaluations by µCT and histology after 8 weeks.

Micro-CT analysis of mouse calvarial specimens:Calvaria were excised and fixed in 4% Formalin. µCT scans were performed in a Micro-CT40 (Scanco Medical AG) the X-ray tube operating at an energy of 70 kVp and an intensity of 114 . Three-dimensional images were reconstructed with an isotropic voxel size of μA 10

52 μm. A global threshold corresponding to 9.8 % of the maximum grey values was used to separate bone from surrounding soft tissues. Regions of interest for the evaluation were selected by placing a cylinder of 4 mm diameter in the center of bone defects. Bone volume, connectivity, and trabecular thickness within this mask were measured using the ImageJ plugin BoneJ (26). Bone Coverage was measured in a dorso-ventral projection of the cylindrical mask.

Histological staining and immunohistochemistry of orthotopic bone formation: Samples were fixed in 4% (v/v) formalin, decalcified for 10 days with 10% w/v EDTA (pH

7.14)were stainedand embedded with Hematoxylin in paraffin. and For Eosin histological (H&E, Sigma-Aldrich). evaluations, 4 μm thin tissue sections

Identification of Sca1+ SSCs by fluorescence activated cell sorting (FACS):PEG- hydrogel based cell-traps were harvested after 8 days of implantation and infiltrated cells were released by mechanical and enzymatic dissociation. Specifically, the cell-traps were placed in collagenase A digestion buffer supplemented with DNase (Qiagen) and incubated at 37 °C for 40 min under gentle agitation. After collagenase digestion and neutralization, undigested materials were gently triturated by repeated pipetting. Resulting cells suspensions were filtered through 100 µm nylon mesh, pelleted at 400 g at 4 °C and re-suspended in FACS buffer (2% fetal bovine serum in PBS pH 7.2, 1 mM EDTA). Red blood cells were lysed by incubation with RBC lysis (Biolegend, cat.no 420301) buffer for 5 min on ice before cells were washed and re-suspended in FACS buffer. Cells were stained with fluorochrome-conjugated antibodies against CD45 (Biolegend, cat.no 103105), Tie2 (Biolegend, cat.no 124007), CD90.2 (Biolegend, cat.no. 140309), 6C3 (Biolegend, cat.no. 108307) and Sca1 (Biolegend, cat.no. 108111) for 30 min at 4 °C and analyzed by FACS. Sytox Red dead cell stain (ThermoFisher, cat.no. S34859) was used to evaluate cell viability, according to manufacturer recommendations. Flow cytometric experiments were performed with a LSR Fortessa (BD Biosciences) in the Flow Cytometry Core Facility, ETH Zurich. Gates were defined according to the fluorescence intensity of the isotype control and data was analyzed using FlowJo Software (FlowJo LLC).

Statistical Analysis: All statistical analysis was performed in GraphPad Prism (GraphPad Software). Data analysis was executed using one-

t-test assuming two-tailed distributionway ANOVA and and unequal post variances. Bonferroni’s In correction or Student’s 53 all cases, a p-value of <0.05 was considered statistically significant, and all data is depicted as mean ± SD.

Acknowledgments

We thank Esther Kleiner for superb technical assistance, Dr. Paolo Cinelli for support with micro-CT, Flora Nicholls for help with animal experiments and the Flow Cytometry Core Facility at ETH Zurich where all the FACS experiments were conducted. The research leading to these results has received funding from the People Programme (Marie Curie Actions) of the European Union's Seventh Framework Programme FP7/2007-2013/ under REA grant agreement No. 607868 (iTERM).

References

1. R. C. Rennert, M. Sorkin, R. K. Garg, G. C. Gurtner, Stem cell recruitment after injury: lessons for regenerative medicine. Regen Med 7, 833-850 (2012). 2. T. Kitaori, H. Ito, E. M. Schwarz, R. Tsutsumi, H. Yoshitomi, S. Oishi, M. Nakano, N. Fujii, T. Nagasawa, T. Nakamura, Stromal cell-derived factor 1/CXCR4 signaling is critical for the recruitment of mesenchymal stem cells to the fracture site during skeletal repair in a mouse model. Arthritis Rheum 60, 813-823 (2009). 3. N. Hosogane, Z. Huang, B. A. Rawlins, X. Liu, O. Boachie-Adjei, A. L. Boskey, W. Zhu, Stromal derived factor-1 regulates bone morphogenetic protein 2-induced osteogenic differentiation of primary mesenchymal stem cells. Int J Biochem Cell Biol 42, 1132-1141 (2010). 4. A. Faber, C. Roderburg, F. Wein, R. Saffrich, A. Seckinger, K. Horsch, A. Diehlmann, D. Wong, G. Bridger, V. Eckstein, A. D. Ho, W. Wagner, The many facets of SDF-1alpha, CXCR4 agonists and antagonists on hematopoietic progenitor cells. J Biomed Biotechnol 2007, 26065 (2007). 5. C. Wan, S. R. Gilbert, Y. Wang, X. Cao, X. Shen, G. Ramaswamy, K. A. Jacobsen, Z. S. Alaql, A. W. Eberhardt, L. C. Gerstenfeld, T. A. Einhorn, L. Deng, T. L. Clemens, Activation of the hypoxia-inducible factor-1α pathway accelerates bone regeneration. Proceedings of the National Academy of Sciences 105, 686-691 (2008). 6. D. J. Ceradini, A. R. Kulkarni, M. J. Callaghan, O. M. Tepper, N. Bastidas, M. E. Kleinman, J. M. Capla, R. D. Galiano, J. P. Levine, G. C. Gurtner, Progenitor cell trafficking is regulated by hypoxic gradients through HIF-1 induction of SDF-1. Nat Med 10, 858-864 (2004). 7. C. Yellowley, CXCL12/CXCR4 signaling and other recruitment and homing pathways in fracture repair. Bonekey Rep 2, 300 (2013). 8. S. Herberg, C. Susin, M. Pelaez, R. N. Howie, R. Moreno de Freitas, J. Lee, J. J. Cray, Jr., M. H. Johnson, M. E. Elsalanty, M. W. Hamrick, C. M. Isales, U. M. Wikesjo, W. D. Hill, Low-dose bone morphogenetic protein-2/stromal cell-derived factor-1beta cotherapy induces bone regeneration in critical-size rat calvarial defects. Tissue Eng Part A 20, 1444-1453 (2014). 9. H. D. Hwang, J. T. Lee, J. T. Koh, H. M. Jung, H. J. Lee, T. G. Kwon, Sequential Treatment with SDF-1 and BMP-2 Potentiates Bone Formation in Calvarial Defects. Tissue Eng Part A 21, 2125-2135 (2015). 10. K. Higashino, M. Viggeswarapu, M. Bargouti, H. Liu, L. Titus, S. D. Boden, Stromal cell-derived factor-1 potentiates bone morphogenetic protein-2 induced bone formation. Tissue Eng Part A 17, 523-530 (2011). 11. J. L. Holloway, H. Ma, R. Rai, K. D. Hankenson, J. A. Burdick, Synergistic Effects of SDF-1alpha and BMP-2 Delivery from Proteolytically Degradable Hyaluronic Acid Hydrogels for Bone Repair. Macromol Biosci 15, 1218-1223 (2015). 12. X. Shen, Y. Zhang, Y. Gu, Y. Xu, Y. Liu, B. Li, L. Chen, Sequential and sustained release of SDF-1 and BMP-2 from silk fibroin-nanohydroxyapatite scaffold for the enhancement of bone regeneration. Biomaterials 106, 205-216 (2016). 13. J. K. Wise, D. R. Sumner, A. S. Virdi, Modulation of stromal cell-derived factor-1/CXC chemokine receptor 4 axis enhances rhBMP-2-induced ectopic bone formation. Tissue Eng Part A 18, 860-869 (2012). 14. M. P. Lutolf, F. E. Weber, H. G. Schmoekel, J. C. Schense, T. Kohler, R. Muller, J. A. Hubbell, Repair of bone defects using synthetic mimetics of collagenous extracellular matrices. Nat Biotechnol 21, 513-518 (2003). 15. P. S. Lienemann, S. Metzger, A. S. Kivelio, A. Blanc, P. Papageorgiou, A. Astolfo, B. R. Pinzer, P. Cinelli, F. E. Weber, R. Schibli, M. Behe, M. Ehrbar, Longitudinal in vivo evaluation of bone regeneration by combined measurement of multi-pinhole SPECT and micro-CT for tissue engineering. Sci Rep 5, 10238 (2015). 16. Q. Jin, W. V. Giannobile, SDF-1 enhances wound healing of critical-sized calvarial defects beyond self-repair capacity. PLoS One 9, e97035 (2014). 17. T. Takayama, J. Dai, K. Tachi, R. Shohara, H. Kasai, K. Imamura, S. Yamano, The potential of stromal cell-derived factor-1 delivery using a collagen membrane for bone regeneration. J Biomater Appl 31, 1049-1061 (2017). 18. Y. Wang, W. Fu, S. Zhang, X. He, Z. a. Liu, D. Gao, T. Xu, CXCR-7 receptor promotes SDF- 1α-induced migration of bone marrow mesenchymal stem cells in the transient cerebral ischemia/reperfusion rat hippocampus. Brain Research 1575, 78-86 (2014).

19. X. Liu, B. Duan, Z. Cheng, X. Jia, L. Mao, H. Fu, Y. Che, L. Ou, L. Liu, D. Kong, SDF-1/CXCR4 axis modulates bone marrow mesenchymal stem cell apoptosis, migration and cytokine secretion. Protein Cell 2, 845-854 (2011). 20. W. Friess, Collagen – biomaterial for drug delivery1. European Journal of Pharmaceutics and Biopharmaceutics 45, 113-136 (1998). 21. C. H. Lee, A. Singla, Y. Lee, Biomedical applications of collagen. International Journal of Pharmaceutics 221, 1-22 (2001). 22. C. K. Chan, E. Y. Seo, J. Y. Chen, D. Lo, A. McArdle, R. Sinha, R. Tevlin, J. Seita, J. Vincent- Tompkins, T. Wearda, W. J. Lu, K. Senarath-Yapa, M. T. Chung, O. Marecic, M. Tran, K. S. Yan, R. Upton, G. G. Walmsley, A. S. Lee, D. Sahoo, C. J. Kuo, I. L. Weissman, M. T. Longaker, Identification and specification of the mouse skeletal stem cell. Cell 160, 285-298 (2015). 23. C. Cencioni, M. C. Capogrossi, M. Napolitano, The SDF-1/CXCR4 axis in stem cell preconditioning. Cardiovascular Research 94, 400-407 (2012). 24. M. Ehrbar, S. C. Rizzi, R. Hlushchuk, V. Djonov, A. H. Zisch, J. A. Hubbell, F. E. Weber, M. P. Lutolf, Enzymatic formation of modular cell-instructive fibrin analogs for tissue engineering. Biomaterials 28, 3856-3866 (2007). 25. M. Ehrbar, S. C. Rizzi, R. G. Schoenmakers, B. S. Miguel, J. A. Hubbell, F. E. Weber, M. P. Lutolf, Biomolecular hydrogels formed and degraded via site-specific enzymatic reactions. Biomacromolecules 8, 3000-3007 (2007). 26. M. Doube, M. M. Klosowski, I. Arganda-Carreras, F. P. Cordelieres, R. P. Dougherty, J. S. Jackson, B. Schmid, J. R. Hutchinson, S. J. Shefelbine, BoneJ: Free and extensible bone image analysis in ImageJ. Bone 47, 1076-1079 (2010).

Chapter V General discussion and future perspectives

General discussion

The overall aim of this thesis was to promote bone regeneration, a unique challenge for both clinicians and scientists by means of tissue engineering. The focus was to establish cell instructive microenvironments which by allowing cell infiltration and biomolecules signals could be utilized as sophisticated biomimetic hydrogel materials for bone regeneration. In the second chapter, we reported on a biomimetic cell-trap, which by permitting the homing of tissue progenitor cells to bone defects was employed to further characterize the entrapped healing-associated cellular participants. In the third chapter, a systematic in vivo approach was assumed, utilizing TG-PEG hydrogels to investigate the dual effect of BMP-2 and the chemokine SDF- at bone healing and skeletal stem cells recruitment. The fourth chapter aimed on understanding1α the recent identified skeletal stem and progenitor cells at the microenvironment of healing fracture site.

58 differentiate into osteoblasts and osteocytes or localize to the perivascular regions of sinusoids, being identified as bona fide stem cells for skeletal tissues (2).

Biomolecule factors for bone regeneration

A fundamental value of the tissue engineering field is to mimic the natural extracellular matrix and structure of the native milieu. A currently widely used approach is to incorporate biomolecules and various growth factors into biomaterials for controlled release (3). Growth factors such as BMPs and PDGF-BB have proven critical for proper wound repair and tissue regeneration (4, 5). Consistent with improved bone healing, we reported that defects treated with PDGF-BB releasing hydrogels contained significantly increased numbers of Sca1+ SSCs cells compared to both BMP-2 and control samples. These Sca1+ SSCs cells of the ventral part mostly co-localized with CD31+ endothelial cells, indicating their co-infiltration with vascular structures. This suggests that perivascular Sca1+ SSCs detected on the ventral side are crucial for bone regeneration. This is consistent with the current research showing that MSCs in multiple human tissues reside in perivascular niches (6) and that prospectively isolated CD146+ human bone marrow- derived SSCs give rise to bone when transplanted in absence of osteogenic cues (7). Notably, Caplan A. recently suggested that most, if not all, MSCs arise from pericytes, arising a new picture to describe their positioning and functionality in situ (8). Accordingly, future research could be addressed by locating of MSCs to injury sites, which are initiated by mobilization of the circulating peripheral blood.

Summarizing, in the second chapter we followed a stepwise bio-inspired engineering approach that allowed us to identify and characterize promising molecular and cellular key players in bone healing. We applied the knowledge gained from in vitro and in vivo experiments in creating cell permissive artificial biomimetic microenvironments for enhanced homing of Sca1+ SSCs to bone defects.

PEG-based hydrogels, concomitant delivery of SDF-

59 formation. To our experience, the concomitant release of these biomolecules did not result in bridging the edges of bone defects.

Different efforts to justify the failure of our system to boost bone formation were presented during the discussion including: i) natural scaffolds providing the appropriate adhesion sites directing the cellular processes during injury, ii) affinity bind sites for these bioactive molecules within natural scaffolds that lead to tissue formation, iii) high solubility of the biomolecules in vivo and rapid dispersion after implantation, and iv) necessity to incorporate and release of the biomolecules in a controlled manner.

The rapid release of SDF-1 promoted Sca1+ SSCs induction whereas; as well, fast release

+ of BMP-2 facilitated significantα increase of Sca1 SSCs within the microenvironment of bone injury. The application of PEG-based hydrogels scaffolds may comprise a powerful platform in bone tissue engineering, but in respect to underlie SDF-1 , we speculate that the chemoattractant effect of SDF- further augmented. We assumα ed that the fast release of the active biomolecule from1α wasour PEG-based hydrogel system after implantation resulted in rapid cell recruitment which was followed by swift degradation of our scaffold.

Skeletal stem and progenitor cells during bone injury and regeneration

In the fourth chapter, we focused on promoting bone healing by exploring the bone formation capacity of SSCs and BCSPs for the generation of novel bone healing inducing biomaterials. Chan and colleagues identified and introduced SSCs as cells that can commence progenitors, similar to the hematopoietic system. These cells are spatially restricted and locally resident of the bone/bone marrow organ and can generate in vivo, tissues associated to the skeletal system. Moreover, these skeletal stem cells are stimulated by BMP signaling, and instinctive to the skeletal system, when subcutaneous BMP-2 collagen-discs resulted in recruitment of SSCs in these matrices (14). We reported that biomimetic-based matrices incorporated with low-dose BMP-2 could significantly improve the mobilization of SSCs and BCSPs as compared to control treatments, while sole treatment of BMP-2 was sub-critical to promote bone regeneration. Additionally, transplantations of culture expanded SSCs and BCSPs to the bone-healing microenvironment confirmed that both cell populations maintained their ability to spontaneously form osteogenic cells and support the healing of critical-sized bone defects in presence of low-dose BMP-2.

These data indicate that expanded SSCs and BCSPs represent specific endogenous healing associated stem cell populations. Therefore, they could be highly relevant cells for the study of healing dynamics, cellular and molecular functions both in vitro and in vivo.

Orthotopic healing of flat bones through endochondral ossification

Fracture healing and bone repair are postnatal processes that resemble many of the developmental events taking place during embryogenesis of the skeleton and have been extensively reviewed (15, 16). The recapitulation of these ontological processes is strongly suggesting that fracture healing, which is one of the few postnatal developmental mechanisms, is indeed regenerative. This results in the reconstruction of the damaged skeletal tissue into the pre-injury cellular composition, structure and biomechanical function (17).

We showed that in vicinity of healing bones of both expanded or prospectively isolated SSCs and BCSPs in presence of sub-critical dose of BMP-2 spontaneously differentiated into osteogenic cells. Careful examination of these specimen demonstrated that healing occurs through endochondral ossification. In contrast to the embryonic development where flat bones are generated though mesenchyme condensation, a process known as intramembranous ossification, here we show that the healing of the flat bone is at least partially recapitulated through the process of endochondral ossification (Figure 1). It is known that long bones develop through endochondral ossification with cartilaginous formation. During this stage chondrocytes and fibroblasts are vastly outnumbered.

It is expected that parameters for healing are interrelated as paracrine and/or autocrine signaling among SSCs and their progeny and may positively regulate their own expansion, shifting from a stem population to a progenitor state, or participate in lineage specification (14) responsible to generate bone, cartilage and stromal cells within the wound site.

Figure1: Fracture repair in flat bone through endochondral ossification. Movat pentachrome stained sections of coronal cr ). Contribution of chondrocytes in fracture repair in an orthotopic bone model (indicated in blue/green). Complete remodeled mineralized bone is indicated in yellow and stromal oss sections (scale bar: 200μm Resorption of cortical bone and formation of new woven bone were both increased in all different treatments A-C (black and red dashed line inlets) Specific stains the healing ofcells flat bonein red (calvarial) (inlets scale is recapitulated bar: 20 μm). rather through endochondral ossification and not through intramembranous ossification as during development (asterisk indicating the remaining volume of the used cell-trap).

Sca1+ SSCs distribution within the neonatal SSC and BCSP populations

The ideal source of stem cells for bone fracture repair should be easily accessible and cells should sustain viability to allow rapid in vitro culture expansion. Survival and integration within the host bone tissue should be enabled, while the cell source should not show tumorigenicity. In order to correlate Sca1+ SSCs (as described in chapter II) with the latest identified SSCs (facilitated in chapter IV) we explored their overlap FACS plots distribution within the SSCs and BCSPs populations based on surface expression markers by FACS (Figure 2).

These results suggest that Sca1+ SSCs disseminate in favor of the progeny of SSCs particularly to BCSPs. Consequently, the bone nodules formation within the defects treated with Sca1+ SSCs could be mirrored to the ones of prospectively isolated BCSPs.

It is worth to note that here we demonstrated the differential expression of Sca1+ cells secreted from neonatal limbs digestion, where previously (chapter II) the facilitated Sca1+ cells were wound-derived, indicating that the different source of the harvested cells could potentially influence their differentiation fate. Collectively, it would be essential in the future to investigate the potential role of wound-derived SSCs and BCSPs in bone regeneration, as it was reported with Sca1+ SSCs. Furthermore, we could achieve the comparison between wound-derived Sca1+ SSCs and wound-derived SSCs and BCSPs.

Future Perspectives

Most adult stem cells are suggested to be quiescent and reside in a specialized microenvironment, which is termed as (18). The stem cell niche hypothesis provided an initial core in which“stem the cell cell niche” types and factors involved in determining stem cell behavior is still to be defined. Advancements in techniques such as lineage-tracing, single cell isolation, manipulation and imaging will broaden our knowledge in the nature of stem cell niche-cell interactions (19). These stem cells are activated after tissue injury and participate in the healing process. In addition, bone marrow-derived stem cells have been identified as crucial cell sources with regenerative capacity that could contribute to other tissues. The bone marrow pools home stem/progenitor cells that contains niches for hematopoietic stem cells (HSCs), mesenchymal stem cells (MSCs) and endothelial progenitor cells (EPCs). HSCs are regulating the secretion of all circulating blood cells. A crucial aspect of the HSC population for tissue regeneration is to support with paracrine bioactive factors to cells with regenerative capacities and further differentiate into desired tissue-specific lineages

(20). Up to now, the origin of stem/progenitor cells within the bone wound remains a question as well as their way of accessing the wound.

In this thesis, with the objective of bone repair in a manner which recapitulates natural bone healing process, both cell-based (Chapter II and IV) and free strategies (Chapter II and III) have been employed. Both showed to be advantageous, but currently cell-based therapeutic approaches are the status quo in pre-clinical applications. We believe that future research should be directed towards investigating cellular processes of MSCs and SSCs / BCSPs injury-derived, such as migration, induction, proliferation and differentiation in order to elucidate their role in bone healing. Future implants should therefore be designed such that they first release a chemoattractant that recruits healing- associated stem and progenitor cells, second deliver a mitogen stimulus to promote cell proliferation and third exert osteogenic cues to induce differentiation.

While the perfect biomimetic scaffold does not currently exist, constant advancements in tissue engineering research have resulted in increasing numbers of commercially available products and promising future therapies. As research in regenerative medicine continues to grow, we envision novel scaffold technology, which includes

autologous “off 64

-based scaffolds and combination of established strategies involving stem thecells, shelf” gene cell therapy, small biomolecules, and mechanical forces. We believe that the application of a biomaterial-based cell-trap presented herein, is an appealing approach for reducing the number of isolated stem or progenitor cells participating in bone regeneration in order to be applied in vivo in a healing fracture site. Due to its synthetic and bioinert nature the presented biomimetic cell-trap allows the occurrence of the natural healing cascade, which is essential for the creation of more efficient clinically safe bone healing implants. We see a tremendous clinical potential for smart cell-instructive implants that aim at substituting or accelerating naturally occurring signals to enhance the healing process in deficient tissues.

The engineering of systems that specifically control biological processes by spatiotemporally controlling the release of multiple factors will enable the study of healing mechanism and the design of next generation smart healing materials. The herein used synthetic material allows the study of the natural healing cascade in absence of uncontrollable confounding signals present in naturally derived biomaterials. Moreover, by tailoring the cell-trap degradation rate to the natural healing cascade the spatial distribution of elements“ participating” in bone healing can be conserved. With respect to the here intended manipulation of SSCs participating in bone regeneration, precisely balancing their migration, proliferation and differentiation will be essential. To fully exploit the body's healing capacity, selected elements of the natural extracellular matrix such as engineered integrin binding sites acting synergistically with adjacent to growth factor binding sites will be crucial (5). Besides, another point of consideration is the release kinetics in vivo of bioactive molecules. While in vitro release profiles have been evaluated (5, 21), in vivo release kinetics remains unknown. Furthermore, the potential cross-talks influenced by doses, release kinetics and release sequence of growth factors during combined treatments (22) have not been determined. Finally, the correlation between the here reported murine SSCs and human healing associated SSCs will be important for the creation of customized artificial microenvironments that augment the natural healing cascade, in order to be employed in defected sites for in situ tissue engineering applications.

References

1. T. G. Kim, H. Shin, D. W. Lim, Biomimetic Scaffolds for Tissue Engineering. Advanced Functional Materials 22, 2446-2468 (2012). 2. P. Bianco, X. Cao, P. S. Frenette, J. J. Mao, P. G. Robey, P. J. Simmons, C. Y. Wang, The meaning, the sense and the significance: translating the science of mesenchymal stem cells into medicine. Nat Med 19, 35-42 (2013). 3. K. Lee, E. A. Silva, D. J. Mooney, Growth factor delivery-based tissue engineering: general approaches and a review of recent developments. Journal of The Royal Society Interface 8, 153-170 (2011). 4. S. Barrientos, O. Stojadinovic, M. S. Golinko, H. Brem, M. Tomic-Canic, Growth factors and cytokines in wound healing. Wound Repair Regen 16, 585-601 (2008). 5. M. M. Martino, F. Tortelli, M. Mochizuki, S. Traub, D. Ben-David, G. A. Kuhn, R. Muller, E. Livne, S. A. Eming, J. A. Hubbell, Engineering the growth factor microenvironment with fibronectin domains to promote wound and bone tissue healing. Sci Transl Med 3, 100ra189 (2011). 6. M. Crisan, S. Yap, L. Casteilla, C.-W. Chen, M. Corselli, T. S. Park, G. Andriolo, B. Sun, B. Zheng, L. Zhang, C. Norotte, P.-N. Teng, J. Traas, R. Schugar, B. M. Deasy, S. Badylak, H.-J. Bűhring, J.-P. Giacobino, L. Lazzari, J. Huard, B. Péault, A Perivascular Origin for Mesenchymal Stem Cells in Multiple Human Organs. Cell Stem Cell 3, 301-313 (2008). 7. B. Sacchetti, A. Funari, S. Michienzi, S. Di Cesare, S. Piersanti, I. Saggio, E. Tagliafico, S. Ferrari, P. G. Robey, M. Riminucci, P. Bianco, Self-renewing osteoprogenitors in bone marrow sinusoids can organize a hematopoietic microenvironment. Cell 131, 324-336 (2007). 8. A. I. Caplan, The new MSC: MSCs as pericytes are sentinels and gatekeepers. J Orthop Res, (2017). 9. C. H. Lee, M. U. Jin, H. M. Jung, J. T. Lee, T. G. Kwon, Effect of dual treatment with SDF-1 and BMP-2 on ectopic and orthotopic bone formation. PLoS One 10, e0120051 (2015). 10. S. C. Lo, K. C. Li, Y. H. Chang, M. N. Hsu, L. Y. Sung, T. A. Vu, Y. C. Hu, Enhanced critical- size calvarial bone healing by ASCs engineered with Cre/loxP-based hybrid baculovirus. Biomaterials 124, 1-11 (2017). 11. Q. Jin, W. V. Giannobile, SDF-1 enhances wound healing of critical-sized calvarial defects beyond self-repair capacity. PLoS One 9, e97035 (2014). 12. J. L. Holloway, H. Ma, R. Rai, K. D. Hankenson, J. A. Burdick, Synergistic Effects of SDF-1alpha and BMP-2 Delivery from Proteolytically Degradable Hyaluronic Acid Hydrogels for Bone Repair. Macromol Biosci 15, 1218-1223 (2015). 13. X. Shen, Y. Zhang, Y. Gu, Y. Xu, Y. Liu, B. Li, L. Chen, Sequential and sustained release of SDF-1 and BMP-2 from silk fibroin-nanohydroxyapatite scaffold for the enhancement of bone regeneration. Biomaterials 106, 205-216 (2016). 14. C. K. Chan, E. Y. Seo, J. Y. Chen, D. Lo, A. McArdle, R. Sinha, R. Tevlin, J. Seita, J. Vincent- Tompkins, T. Wearda, W. J. Lu, K. Senarath-Yapa, M. T. Chung, O. Marecic, M. Tran, K. S. Yan, R. Upton, G. G. Walmsley, A. S. Lee, D. Sahoo, C. J. Kuo, I. L. Weissman, M. T. Longaker, Identification and specification of the mouse skeletal stem cell. Cell 160, 285-298 (2015). 15. C. Ferguson, E. Alpern, T. Miclau, J. A. Helms, Does adult fracture repair recapitulate embryonic skeletal formation? Mechanisms of Development 87, 57-66 (1999). 16. L. C. Gerstenfeld, D. M. Cullinane, G. L. Barnes, D. T. Graves, T. A. Einhorn, Fracture healing as a post-natal developmental process: Molecular, spatial, and temporal aspects of its regulation. Journal of Cellular Biochemistry 88, 873-884 (2003). 17. T. A. Einhorn, L. C. Gerstenfeld, Fracture healing: mechanisms and interventions. Nat Rev Rheumatol 11, 45-54 (2015). 18. D. T. Scadden, The stem-cell niche as an entity of action. Nature 441, 1075-1079 (2006). 19. J. Voog, D. L. Jones, Stem cells and the niche: a dynamic duo. Cell Stem Cell 6, 103-115 (2010). 20. A. Dahlberg, C. Delaney, I. D. Bernstein, Ex vivo expansion of human hematopoietic stem and progenitor cells. Blood 117, 6083-6090 (2011). 21. S. Metzger, P. S. Lienemann, C. Ghayor, W. Weber, I. Martin, F. E. Weber, M. Ehrbar, Modular poly(ethylene glycol) matrices for the controlled 3D-localized osteogenic differentiation of mesenchymal stem cells. Adv Healthc Mater 4, 550-558 (2015). 22. R. Mishra, R. S. Sefcik, T. J. Bishop, S. M. Montelone, N. Crouser, J. F. Welter, A. I. Caplan, D. Dean, Growth Factor Dose Tuning for Bone Progenitor Cell Proliferation and Differentiation on Resorbable Poly(propylene fumarate) Scaffolds. Tissue Eng Part C-Me 22, 904-913 (2016).