Page 1
Combined angiogenic and osteogenic factor delivery for bone regenerative engineering
Quanjun Cui, MD, Abhijit S. Dighe, PhD, and James N. Irvine Jr. MD.
Department of Orthopaedic Surgery, University of Virginia School of Medicine, Charlottesville, Virginia, USA
Running title: Combined angiogenic and osteogenic therapy for bone healing
Each author’s email address:
Quanjun Cui [email protected] S. Dighe [email protected] N. Irvine Jr. [email protected]
Corresponding authors name, address, phone, fax and email:
Quanjun Cui, MD, MSc.Department of Orthopaedic Surgery, P.O. Box 800159, University of Virginia, Charlottesville, VA 22908.Phone : 1-434-243-0236Fax: 1-434-243-0242Email: [email protected]
1
1
2
3
567
8
9
10
11
1213141516
17
18192021222324
12
Page 2
Abstract
Both osteogenesis and angiogenesis are integrated parts of bone growth and regeneration.
Combined delivery of osteogenic and angiogenic factors is a novel approach in bone
regenerative engineering. Exogenous addition of mesenchymal stem cells (MSCs), vascular
endothelial growth factor (VEGF) and bone morphogenetic proteins (BMPs) together with an
osteoconductive scaffold is a very promising method to enhance bone repair. This concept has
been incorporated into the development of new strategies for bone tissue engineering and
significant advancements have been made in last 10 years. In contrary to previous belief that
VEGF modulates bone repair only by enhancing angiogenesis in the proximity of bone injury,
recent evidence also suggests that cross-talk between VEGF and BMP signaling pathways in
MSCs promotes osteoblastic differentiation of MSCs which aids in fracture repair. Future
studies should focus on cross-talk between angiogenesis and osteogenesis, optimization of
VEGF/BMP ratios, selection of the most potent BMPs, and optimization of delivery methods for
VEGF and BMP. Recent discoveries from basic research including effective delivery of growth
factors and cells to the area of interest will help bring VEGF plus BMP for bone healing from the
bench to the patient’s bedside.
Key Words: angiogenesis, osteogenesis, vascular endothelial growth factor, bone morphogenetic
proteins, bone formation, bone tissue engineering, growth factor.
2
2526
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
34
Page 3
Introduction:
Orthopaedic surgery is a unique field in that the majority of patients presenting with bone
fractures, joint disease and pain are made better following their surgery. Roughly 10% of
fractures are subject to delayed union and nonunion . Conditions such as cancer and infection
require large segments of bone to be removed to treat the patient, and in trauma cases, the bone
may be missing on arrival to the emergency department or it may be debrided later in surgery for
contamination from injury. Spinal fusion presents an issue with bone-on-bone healing and
osteonecrosis (ON) is a disease of impaired osseous blood flow that is often seen in patients
with femoral neck fractures or hip dislocations . ON can also occur in association with non-
traumatic factors such as excessive use of steroids and alcohol, which contribute to ON in about
two-thirds of all cases .
Bone injuries to the postnatal skeleton are repaired through natural healing which is a complex,
well orchestrated process that essentially recapitulates the pathway of embryonic development. It
occurs in three successive steps – inflammation, osteogenesis (endochondral as well as
intramembraneous ossification) and bone remodeling and involves a variety of cell types and
signaling molecules . The inflammation phase is characterized by production of Interleukin-1
(IL-1), Interleukin-6 (IL-6) and tumor necrosis factor (TNF-α); osteogenesis phase involves
production of bone morphogenetic proteins (BMPs), transforming growth factor-beta (TGF-β),
insulin-like growth factors (IGFs), vascular endothelial growth factor (VEGF), angiopoietins,
platelet-derived growth factor (PDGF) , fibroblast growth factor (FGF), receptor activator of
nuclear factor kappa B ligand (RANKL), macrophage colony-stimulating factor (MCSF),
osteoprotegerin (OPG), and the bone remodeling phase is modulated by production of
metalloproteinases, RANKL, OPG, IL-1, IL-6, TNF-α . The cell types involved in the fracture
repair process includes platelets, macrophages (inflammatory phase); mesenchymal stem cells
3
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
56
Page 4
(MSCs), chondrocytes, endothelial cells, osteoblasts (osteogenesis phase), and osteoclasts (bone
remodeling phase) . Deficiency or impaired function of one or more cellular and non-cellular
components of this cascade may arise in pathophysiological conditions which may impair tissue
healing. Trauma can create critical sized defects in bone which are unable to heal naturally. In
these cases, therapeutic interventions must be available to the patient, which includes
administration of functional factors to allow bone repair. In vivo studies have revealed that
BMPs, TGF-β, IGF, FGF, VEGF and PDGF are all present during natural fracture healing .
Research has proven that the formation of new blood vessels is essential for fracture repair and
that expression of BMPs are downregulated in fracture non-unions . These findings would
suggest that the addition of agents enhancing angiogenesis and BMP-signaling pathway may
enhance bone repair. Although angiogenesis is modulated mainly by VEGF and osteogenesis is
mainly induced by BMPs there are several other growth factors that directly or indirectly
enhance these two processes during fracture repair.
Prior to the application of small molecule and cell based therapies for these cases of poor and
impaired bone healing as well as large segmental defects, these conditions were and still are
treated with bone grafts. The patient’s own bone (autograft) can be harvested from their iliac
crest but there is associated morbidity with the harvest and supply is limited. Another option is to
use bone from donors (allograft), however, there is risk of disease transmission and it largely
lacks osteoinductive potential due to the harshness of processing the tissue. This leaves allograft
primarily as an osteoconductive material to fill in defects and acts as a scaffold for native bone to
repair the site. Synthetic graft options have been sought after due to the limited availability and
morbidity associated with autograft and the risk of disease transmission and lack of
osteoinductive potential of allograft. While synthetic options are still being created and tested in
4
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
78
Page 5
the lab, an even newer generation of bone regeneration engineering is being sought out which
focuses on the delivery of osteogenic and angiogenic factors.
Since ON, fracture, and bone defects due to trauma and surgery are associated with disruption of
the local blood supply, the role of angiogenesis in repair and remodeling is of great significance.
Vascular endothelial growth factor is known for its ability to enhance angiogenesis as well as
osteogenesis. It has been studied in fields outside the musculoskeletal arena since its
identification in 1980 . Its characteristics make it an attractive option to help promote healing in
patients suffering with or at risk for ON. The physiology of how VEGF promotes healing lies in
its ability to allow new vessels to invade the area of insult and bring nutrients necessary for
repair, as well as a direct passageway for cells that aid in repair to gain access to the site of
injury. Aside from ON, delayed union or nonunion of bone may result in cases of deficient
vascularity and failure of sufficient angiogenesis to help return adequate blood supply and
nutrition necessary for healing . Fortunately, bone morphogenetic proteins have been discovered
and demonstrate the ability to promote osteogenesis in vitro, in vivo and clinically in humans,
with BMP-2 and -7 the only ones currently approved by the FDA for clinical use. Although they
have been shown to enhance fracture healing and to improve outcomes of ON treatment in
earlier studies, recent independent clinical trials showed they are not as effective as previously
demonstrated. Thus the search for alternative strategies in bone tissue engineering continues. In
the last few years, many studies have focused on combined delivery of osteogenic and
angiogenic growth factors and have shown that this novel approach may be an attractive option
for the aforementioned clinical conditions. This review aims to discuss the current status of the
combined angiogenic and osteogenic factor delivery for bone regenerative engineering.
5
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
910
Page 6
The need for alternative bone graft substitutes
The estimated direct and indirect costs of all fractures in the United States approach 20 billion
dollars with an associated 101.3 million days of restricted activity each year . Some fractures can
heal naturally when the bone is kept under restricted mobility, while some cannot and require
surgical intervention with bone grafting procedures. In 1996, bone grafts were used to
reconstruct bone defects and enhance bone formation in more than 425,000 procedures with an
estimated cost of up to two billion dollars . As mentioned earlier, autogenous bone is a
preferable graft material but supply is limited, with an overall major complication rate of (8.6%)
including infection (2.5%), prolonged wound drainage (0.8%), large hematomas (3.3%),
reoperation (3.8%), pain greater than 6 months (2.5%), sensory loss (1.2%), and unsightly scars.
Minor complications (20.6%) included superficial infection, minor wound problems, temporary
sensory loss, and mild or resolving pain. Allografts are an alternative but are not desirable
because of their potential immunogenicity and the risk of fatal disease transmission. Therefore,
alternative therapeutic strategies for bone regenerative engineering need to be investigated.
Successful use of graft materials is based on their osteoinductive and/or osteoconductive
properties that induce and provide a favorable environment for osteoprogenitor cells to grow and
differentiate into osteocytes . Unfortunately, the available bone graft substitutes and growth
factors are far from ideal . Research in tissue engineering geared at combined use of
bioresorpable scaffolds, osteogenic growth factors such as BMPs and MSCs to generate new
bone at targeted sites may provide new therapeutic options. BMPs induce an intracellular signal
following their binding to a cell-surface receptor which upregulates a cascade of events inside
6
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
1112
Page 7
the cell. These intracellular processes cause differentiation of progenitor cells into chondrocytes
and osteoblasts, which aid new bone formation by enhancing endochondral ossification .
Unfortunately, numerous studies have shown that BMPs failed to enhance stem cell directed
fracture healing and that stem cells failed to respond to BMPs in vitro , which raise concerns on
the efficacy of BMPs in bone repair. Although definitive mechanistic insights on BMP non-
responsive phenotype of MSCs are not available in the literature, it is possible that extracellular
matrix and different culture conditions modulate response of MSCs to BMPs . Recombinant
human BMP-2 and BMP-7 proteins were approved by FDA for clinical use in 2002, however,
the clinical trials to enhance fracture healing using BMPs did not yield desired outcomes .
Angiogenesis is a critical process in fracture repair and bone formation. The invading vessels bring
not only nutrients to the fracture site but also cells that participate in the repair process. Deficiencies
in vascularity and angiogenesis will lead to a delayed union or non-union . Thus, it is important to
include angiogenic factors in bone tissue engineering to optimize strategies for bone repair. In
recent investigations, this novel approach has been explored and has demonstrated the potential for
clinical application.
Angiogenic growth factor (VEGF) and bone formation
Angiogenesis, bone formation , and fracture healing are closely associated at the cellular level .
Angiogenesis involves several steps including degradation of the existing basement membrane and
migration of endothelial sprouts. These processes are controlled by angiogenic factors including
FGFs, angiogenin, TGF- and , TNF-, and VEGF. Because of the effects shown in several soft
tissue models, VEGF is of particular therapeutic interest . It is known that VEGF is one of the most
7
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
1314
Page 8
important mediators of physiologic and pathologic angiogenesis. Exogenous VEGF can induce
new blood vessel formation and increase perfusion in ischemic rabbit limbs. In addition to enhanced
angiogenesis and bone formation in healing osseous tissue, VEGF has also aided skeletal growth .
VEGF couples hypertrophic cartilage remodeling, ossification, and angiogenesis during
endochondral bone formation. Several articles published in recent years suggest that VEGF not only
enhances fracture repair but it is essential component of the bone healing cascade. In their in vivo
studies, Street et al found that the local administration of exogenous VEGF enhanced angiogenesis,
osteogenesis and bony bridging in femoral and tibial defects created in mice as well as in radial
defects created in the rabbits. In addition, when they blocked VEGF receptor signaling it inhibited
healing of tibial defects in mice . In another study, the VEGF inhibitor TNP-470 completely
prevented osteogenesis in a model of closed femoral fracture . As for the mechanism, VEGF can
increase osteogenesis through its action on endothelial cells which increases blood vessel density in
its vicinity. Thorough reviews on VEGF signaling have been published . VEGF has four isoforms
A, B, C and D but VEGF-A (VEGF165) is the protein responsible for inducing angiogenesis. There
isn’t a single universal receptor for all isoform types, instead, VEGFR1 (Flt-1), VEGFR2 (Flk-1),
and VEGFR3 exist. VEGF-A binds to VEGFR2 and activation of VEGFR2 controls angiogenesis
as well as endothelial cell functions. VEGFR2 is a tyrosine kinase type receptor and binding of
VEGF-A to VEGFR2 phosphorylates tyrosine residues. Several signaling molecules like PI3K,
Shb, and TSAd bind to phoshorylated tyrosines within VEGFR2 to activate downstream signaling.
VEGFRs are expressed on MSCs, osteoblasts, endothelial cells and osteoclasts. Therefore, VEGF
can directly bind to osteoprogenitor mesenchymal stem cells , osteoblasts ) and osteoclasts . VEGF
binding can enhance mineralization in osteoprogenitor cells, alkaline phosphatase activity and
capability of migration and differentiation of osteoblasts which may be useful in the prevention and
8
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
1516
Page 9
treatment of osteoporosis. One recent study demonstrated that reduced VEGF expression in MSCs
promoted adipogenic differentiation while inhibiting osteogenic differentiation of MSCs .
Additionally, conditional VEGF knockout mice with a VEGF deficiency in osteoblastic
precursor cells exhibited increased marrow fat and osteoporotic phenotype .
Synergistic effects of VEGF and BMPs on bone formation were also demonstrated in a skull defect
model and in distraction osteogenesis models . These findings suggest that in addition to its effects
on angiogenesis, VEGF plays an important role in bone growth and repair by interacting with other
growth factors.
Osteogenic growth factors (BMPs) and bone formation
BMPs produced by stem cells act in autocrine and paracrine fashion to stimulate differentiation of
stem cells first to cartilage and then to bone cells. The BMP family of cytokines consists of more
than 15 members identified to date . BMP-2 and BMP-7 have been tested extensively.
Recombinant human BMP-2 (rhBMP-2) and rhBMP-7 were approved by the FDA for clinical use
even though rhBMP-7 did not prove to be more useful than bone grafts . Additionally, the combined
administration of BMP-2 and BMP-7 was found to be more effective than the use of a single protein
. Effects of BMP-2, -4 and -6 on differentiation of bone marrow derived stromal cell lines were
compared and it was observed that the effects differed depending on the cell type exposed to these
proteins. One recent study used human stromal cells and compared the effects of BMP-2, -4, -6 and
-7 on the differentiation of these cells to osteoblasts. Among all BMPs tested, BMP-6 was the most
consistent and potent regulator of osteoblast differentiation and only BMP-6 gene expression was
detected prior to induction of their osteoblast differentiation of human stromal cells . Since reports
9
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
1718
Page 10
from other investigators indicated that MSCs did not respond to BMPs, which contradicted these
findings, it is important that the mechanism of response of MSCs to BMPs be further investigated.
Li et al compared the osteogenic activities of human BMPs – 2, 4, 6, 7 and 9 and produced results
similar to another study by Kang et al that compared 14 BMPs for their osteogenic potential . It was
clearly demonstrated in these studies that BMP-6 and BMP- 9 were the most effective osteogenic
proteins. Phylogenetic analysis revealed that BMPs 5, -6, -7, and -8 form one cluster that is
different than the tight cluster of BMP-2 and BMP-4 . BMP-2, -4 and BMP-6, -7 were found to use
different receptors on the surface of human bone marrow derived stromal cells and therefore utilized
different mechanisms to induce osteoblastic differentiation of primary hMSC . Based on those
findings, it is clear that many unanswered questions exist in terms of mechanisms of action and
effectiveness of individual BMPs. The most potent molecules in BMP family are still under
investigation in the pre-clinical stage. In addition, efficacy in fracture repair of clinically available
BMPs-2 and -7 are inconclusive; plus they are extremely expensive, limiting their (off label) uses in
other musculoskeletal conditions. The limited impact of BMP delivery on a successful outcome of
fracture repair strongly suggests that alternative approaches are necessary. Targeting mesenchymal
cells with a combination of factors that potentiate the osteogenic and angiogenic properties of the
cells may be an alternative strategy towards the development of a therapy for the effective
management of challenging clinical problems such as large bone defects, fracture non-unions and
osteonecrosis.
Combined angiogenic and osteogenic factor delivery
Combined delivery of osteogenic and angiogenic factors for bone regeneration is a novel approach
which aims to one day enhance treatment of musculoskeletal problems. The exogenous addition of
10
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
1920
Page 11
VEGF and BMP-2 or -4 has proven to be beneficial for fracture repair in various animal models
(Table 1). The recent review by Evans included a list of transgenes used in experimental models
of bone healing by gene transfer. Studies have demonstrated that VEGF and BMP-2 act
synergistically to enhance bone formation and healing and also note that specific ratios of the
two are important because detrimental healing effects were observed at improper ratios .
Excessive VEGF expression in MSCs has been shown to inhibit mineralization of MSCs in vitro
and bone formation in vivo (72, 112). Similar synergistic effects were also observed in combined
delivery of VEGF and BMP-6 . However, the combination of VEGF and BMP-7 did not enhance
osteogenesis . Dual growth factor (VEGF and BMP-2) delivery appears promising in critical
sized murine femur bone defect and subcutaneous implant models . In a subcutaneous implant
model concerted delivery of VEGF, BMP-4 and MSCs using poly (lactic-co-glycolic acid)
(PLGA) scaffold induced greater osteogenesis relative to any single factor or combination of two
factors. The combination of VEGF and BMP-2 enhanced bone bridging and union of the critical
size defect compared to delivery of BMP-2 alone. This study utilized a composite system of
gelatin microparticles and poly (propylene fumarate) (PPF) scaffold. Even more complex
scaffolds have recently been developed to quickly release VEGF to promote angiogenesis and
osteogenesis followed by BMP-2 being slowly released in hopes of more accurately mimicking
the sequential events underlying the repair process. These complex scaffolds have also been
implanted with human bone marrow cells and other progenitor cells which result in better repair
of difficult fractures.
Interestingly, many studies have demonstrated that VEGF/BMP ratio is critical in controlling the
outcomes of the combined therapy of VEGF and BMPs. The VEGF/BMP-4 ratios of 0.4 and 1.8
were found to be beneficial for fracture healing whereas a ratio of 9 and 1.3 were found to be
11
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
2122
Page 12
inhibitory for bone repair. In the case of BMP-2, a VEGF/BMP-2 ratio of 4 was found to be
inhibitory while ratios of 0.8 or less were beneficial for fracture repair . It is evident from these
studies that excessive VEGF can be harmful for bone regeneration and optimum ratios of VEGF
and BMP are necessary to benefit the host. Interestingly, during natural (without exogenous
addition of VEGF or BMPs) maturation of B6 mouse derived stromal cells to osteoblasts the
VEGF/BMP-2 ratios were quite high. VEGF was produced throughout the experiment with an
average value of 1500 pg. BMP-2 production increased gradually to reach 250 pg at day 12 and
then it decreased gradually to 100 pg on day 28. Thus, the VEGF/BMP-2 ratio was 25 through day
11, 6 on day 12, and 15 on day 13-28. As previously described, VEGF interacts differently with
BMP-2 than with BMP-4. VEGF interacted effectively with BMP-4 by enhancing cartilage
formation but failed to do so with BMP-2. Although BMP-6 and BMP-9 were found to be the most
effective BMPs, BMP-9 inhibited VEGF induced angiogenesis. At this point, it is difficult to
determine if BMP-9 can exhibit synergistic action with VEGF in bone repair.
The convoluting phenomenon of enhancement of osteogenesis at low VEGF/BMP ratios and
inhibition of osteogenesis at relatively high VEGF/BMP ratio remains elusive. A simplistic
explanation for in vivo inhibition of osteogenesis at higher VEGF/BMP ratios can be attributed to
increased osteoclast survival and recruitment, in the presence of excessive VEGF concentrations .
It is, however, difficult to explain inhibition of osteogenesis as the outcome of enhanced osteoclast
survival as VEGF also enhances osteoblast survival and differentiation . The findings of this
review indicate that the mechanism of VEGF induced enhancement of osteogenesis and fracture
repair is not limited to the increased vascular density which subsequently increases the availability
of nutrients and osteoprogenitor cells at the site of bone repair. A recent investigation clearly reveals
12
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
2324
Page 13
that VEGF has direct effect on stem cells and modulates BMP-4 induced osteogenic differentiation
of stem cells . We have found that VEGF enhances BMP-6 mediated osteogenic differentiation of
stem cells through RunX2 dependent mechanisms and it enhances LMP-1 induced osteogenesis
through osterix dependent mechanisms. All of these findings suggest that VEGF interacts directly
with stem cells and modulates BMP-induced osteogenic differentiation which deserves further
investigation .
Cross-talk between VEGF and BMP signaling pathways in mesenchymal stem cells.
Exogenous addition of mesenchymal stem cells, VEGF and BMP could enhance osteogenesis more
efficiently in comparison with that induced by delivery of any single factor or a combination of any
two factors. Although the number of blood vessels in the poly (lactic-co-glycolic acid) (PLAGA)
implants carrying MSCs (P+MSCs) (P=PLAGA) that expressed VEGF (P+MSCsVEGF,
P+MSCsVEGF+BMP-6 and P+MSCsVEGF+LMP-1 ) were almost similar, the bone volumes were significantly
greater in VEGF+BMP-6 and VEGF+LMP-1 groups. Similarly, the number of blood vessels was
significantly less in P+MSCsBMP-6 and P+MSCsLMP-1 group in comparison with that in P+MSCs-
VEGF group but the bone volumes were comparable. These findings suggested that enhanced
osteogenesis induced by combined delivery of MSCs, VEGF and BMP cannot be explained by
VEGF induced angiogenesis alone in the combination groups. When osteoblastic differentiation of
human adipose derived stem cells (hADSCs) was investigated in vitro in the presence of VEGF and
BMP-6 it was revealed that cross-talk between VEGF and BMP-6 signaling pathways enhanced
osteoblastic differentiation of hADSCs . The combination of VEGF plus BMP-6 significantly
enhanced expression of osteogenic genes ALP, Dlx5 and osterix in hADSCs as compared to
using VEGF or BMP-6 alone. The combination group of VEGF and BMP-6 also had an additive
13
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
2526
Page 14
effect on expression of COL1A1, Msx2 and runx2. Interestingly, there was no change in
expression of OCN gene suggesting that the cross-talk selectively enhanced expression of
specific osteogenic genes. While the mechanisms are not known, the cross-talk between VEGF
and BMP-6 signaling pathways promoted in vitro osteoblastic differentiation of hADSCs but
inhibited in vitro mineralization and in vivo osteogenesis induced by BMP-4 expressing C2C12
cells and NIH 3T3 cells . Since C2C12 and NIH 3T3 cells may not represent true MSCs these
findings suggest that enhancement of osteogenesis may depend on the cell type as well as the
specific BMP used in the study.
Modes of delivery of osteogenic and angiogenic factors for fracture repair
A number of methods have been reported in the literature , transient transfection of MSCs with
plasmids expressing these two genes , retroviral transduction of MSCs , adenoviral transduction
of MSCs , gene activated matrix (GAM) and pure proteins directly incorporated in the
scaffolds . Since very diverse delivery methods were employed by different investigators, the
concentration of VEGF and BMP made available for bone repair in those studies ranged greatly
from only a few picograms to a few micrograms. Similarly, use of MSCs isolated from various
sources, different VEGF/BMP ratios, use of different types of BMP: -2,-4,-6 or -7 and various
scaffolds makes it difficult to conclude what delivery methods were the best given the difficulty
in making direct comparisons amongst studies. However, despite these diverse approaches
employed for the delivery of VEGF and BMP by various groups, these studies collectively
suggest that the combination of MSCs, VEGF and BMP is a very promising method of improved
fracture healing and repair of large defects.
14
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
2728
Page 15
Conclusions:
Both osteogenesis and angiogenesis are integral parts of bone growth and regeneration.
Combined delivery of osteogenic and angiogenic factors is a novel approach in bone
regenerative engineering. Addition of mesenchymal stem cells (MSCs), vascular endothelial
growth factor (VEGF) and bone morphogenetic proteins (BMPs) together with osteoconductive
scaffolds is a very promising method to enhance bone repair. This concept has been incorporated
into the development of new strategies for bone tissue engineering and significant advancements
have been made in this area over the past 10 years. In contrary to previous belief that VEGF
modulates bone repair only by enhancing angiogenesis in the proximity of bone injury, recent
evidence also suggests that cross-talk between VEGF and BMP signaling pathways in MSCs
promotes osteoblastic differentiation of MSCs which aid fracture repair. However, challenges
still remain in the repair of large bone defects due to limitations of adequate blood vessel
network in the large tissue engineered bone graft substitute. Future studies should focus on cross-
talks between angiogenesis and osteogenesis, optimization of VEGF/BMP ratios, selection of the
most potent BMPs, and optimization of delivery methods for VEGF and BMPs. Recent
discoveries from basic research including effective delivery of growth factors and cells to the
area of interest will help bring VEGF plus BMP for bone healing from the bench to the patient’s
bedside.
Conflict of interest
15
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
2930
Page 16
Dr. Cui or an immediate family member has received research or institutional support from
Orthopaedic Research and Education Foundation, National Institute of Health, and Department
of Defense; received royalties from Elsevier. Neither of the following authors nor any immediate
family member has received anything of value from or owns stock in a commercial company or
institution related directly or indirectly to the subject of this article: Dr. Dighe and Dr. Irvine.
Acknowledgments
The research is supported by Orthopaedic Research and Education Foundation/Zachary B.
Friedenberg Clinician Scientist Award.
16
347
348
349
350
351
352
353
354
355
3132
Page 17
Literature cited Table 1: Angiogenic and osteogenic growth factors.
Growth Factor Receptors Cell producing
growth factor
Target cells Function
Bone Morphogenetic
Proteins (BMPs)
Alk1, Alk2,
Alk3, Alk6
(type I
receptors) and
ActRII,
ActRIIB,
BMPRII (type
II receptors)
MSCs,
Osteoblasts,
Chondrocytes
MSCs,
Osteoblasts
BMPs induce
differentiation
of progenitor
cells into
chondrocytes
and osteoblasts.
Growth Differentiation
Factors 5 (GDF-5)
Alk6 Cartilage, MSCs Chondrogenesis
Transforming Growth
Factor – β (TGF-β)
Alk1, Alk5
(type I
receptors) and
TβRII (type II
receptor)
Osteoblasts,
Platelets, Immune
cells,
Chondrocytes
MSCs,
Chondrocytes,
Osteoblasts
Mitogenic and
chemotactic for
osteoprogenitor
cells
Fibroblast Growth
Factor (FGF)
FGFRs 1-4. MSCs,
Osteoblasts,
Chondrocytes,
Fibroblasts,
Endothelial
cells, Smooth
Angiogenesis,
Mitogenic.
17
356357358359
3334
Page 18
Macrophages muscle cells,
MSCs,
Osteoblasts,
Chondrocytes
Platelet Derived
Growth Factor (PDGF)
PDGFR-αα,
PDGFR-ββ,
PDGFR-αβ.
Platelets, Smooth
muscle cells,
Endothelial cells,
Macrophages
MSCs,
Smooth
muscle cells,
Endothelial
cells,
Osteoblasts
Chemotactic,
Angiogenesis,
Mitogenic.
Insulin like Growth
Factor (IGF)
IGF1R, IGF2R Liver cells,
Endothelial cells,
Osteoblasts,
Chondrocytes,
MSCs.
Skeletal
muscle cells,
Chondrocytes,
MSCs,
Endothelial
cells,
Osteoblasts.
Induces protein
synthesis,
Mitogenic,
Chemotactic
Vascular Endothelial
Growth Factor (VEGF)
VEGFR1,
VEGFR2
Lung, kidney,
heart, adrenal
gland, liver,
gastric mucosa,
spleen, MSCs.
Endothelial
cells,
Ostoclasts,
Osteoblasts,
MSCs.
Master
regulator of
angiogenesis,
Mitogenic,
Survival signal
for osteoclasts
183536
Page 19
and osteoblasts.
Angiopoietin Tie-1, Tie-2 MSCs, Smooth
muscle cells
Endothelial
cells
Chemotactic,
Vessel
remodeling
during tissue
repair
19
360
3738
Page 20
Table 2: Summary of literature on combined use of VEGF and BMPs in osteogenesis/fracture repair models.
Animal model and defect Source and number of stem cells
Therapeutic gene/protein used and its concentration
Conclusion of the study
B6 mice.
Scaffold implanted in subcutaneous tissue
6 mm diameter in parietal bone.
Peng H, et al. (2002)
Isolated from muscles of 3 weeks old C57BL/10ScSn mdx/mdx mice.
0.6 million cells.
Coated on gelfoam disk.Coated on collagen disk.
Combined use of VEGF and BMP4.
Stem cells were transfected using a retroviral vector.
~250 ng VEGF/ 1 million stem cells.~140 ng BMP4/ 1 million stem cells
In 28 days, synergistic effects of combined therapy of VEGF and BMP4 were observed in terms of new bone formation.
VEGF/BMP-4 = 0.4 , 1.8 beneficialVEGF/BMP-4 = 9 inhibitory
SCID mice
Scaffolds were implanted in subcutaneous tissue to observe differentiation of stem cells into osteocytes.
Huang Y, et al. (2005)
Isolated from human bone marrow.
0.15 million cells
Seeded on PLGA
Combined use of VEGF and BMP4.
3 g VEGF pure protein incorporated in PLGA.200 g of plasmid expressing BMP4 incorporated in PLGA
In 105 days greater bone formation was observed in mice that received all three – cells, VEGF and BMP4 as compared to single factors or combination of two factors.
Mice (SCID)
Implants in the skeletalmuscle pocket of the gluteofemoral muscles
Li G, et al. (2009)
Cell lines obtained from ATCC. C2C12 and 3T3.
0.2 million cells
6 x 6 mm Gelatin foam (Gelfoam)
Combined use of VEGF and BMP-4
Cells transduced with retroviral vectors
~ 130 ng VEGF/ million cells ~ 100 ng BMP-4/million cells
Cells expressing only BMP-4 formed bones. Cells expressing VEGF and BMP-4 did not form bone.
VEGF/BMP-4 = 1.3 complete inhibitory
In vitro mineralization drastically affected at VEGF/BMP-4 = 0.4 or above.
B6 mice.
8 mm defect in femur
6 mm diameter defect in calvaria
Peng H, et al. (2005)
Isolated from muscles of 3 weeks old C57BL/10ScSn mdx/mdx mice.
0.6 million cells
Coated on gelfoam disk
Combined use of VEGF and BMP2.
Stem cells were transfected using a retroviral vector.
~200 ng VEGF / 1 million stem cells.
~250 ng BMP2 / 1 million stem cells
In 28 days, synergistic effects of combined therapy of VEGF and BMP2 were observed in terms of new bone formation.
VEGF/BMP-2 = 0.16, 0.8 beneficial
VEGF/BMP-2 = 4 inhibitory
Balb/c (nu/nu)
Scaffolds placed in thigh muscle pouch of hind legs
Isolated from periosteum of mandibles or maxillas of humans.
1.25 million cells
Combined use of VEGF and BMP2.
The cells were transfected with VEGF and BMP2
VEGF+BMP2 group showed more bone formation as compared to VEGF and BMP-2 groups.
20
361362
363364
3940
Page 21
Samee M, et al. (2008)
Porous beta-TCP scaffolds
plasmids.
~1 ng VEGF / 1 million cells
~ 700 ng BMP2 / 1 million cells
Fischer-344 rats
8 mm critical size cranial defect
Patel ZS, et al. (2008)
No cells used
Gelatin microparticles confined in polypropylene fumarate scaffolds.
Combined VEGF and BMP-2 use.
12 g VEGF; 2 g BMP-2.
VEGF did not enhance vessel formation at 4 weeks; VEGF improved bone formation in VEGF+BMP-2 group at 4 weeks and to some extend at 12 weeks.
Nude mice. Sub-cutaneous assays
Rabbits15 mm critical size defect in radius
Hou H, et al. (2009)
Isolated from bone marrow of rabbits.
Loaded on tricalcium phosphate scaffold
Cells mixed to have VEGF cells :BMP-2 cells = 1 :4
Combined use of VEGF, Angiopoietin and BMP-2
Stem cells transduced with adenovirus particles.
~ 3 ng VEGF/1 million cells~ 8 ng BMP-2/1 million cells
VEGF and angiopoietin showed additive effect on BMP-2 induced osteogenesis in 12 weeks in both models.
VEGF/BMP-2 = 0.1
Fischer 344 rats.
8 mm diameter cranial defect
Young S, et al. (2009)
No cells used
Gelatin micro particles and porous polypropylene fumarate scaffolds
Combined use of VEGF and BMP-2.
0.5, 1.0, 2.0 g BMP-2 ;and ; 6, 12 g VEGF incorporated into the scaffolds
An in vivo dose-dependent decrease in percentage of bone fill (PBF) was observed for BMP-2. The addition of VEGF was unable to reverse this decrease in PBF.
Sprague-Dawley Rats Ectopic assays
Sprague-Dawley Rats 5 mm segmental defect in femur
Kempen DH, et al. (2009)
No cells used
BMP-2 loaded PLAGA microspheres embedded in polypropylene. VEGF mixed gelatin hydrogel surrounded, BMP-2 scaffold, to sequencially release VEGF first, and then, BMP-2
Combined use of VEGF and BMP-2
2 g VEGF; 6.5 or 9.2 g BMP-2 loaded on respective scaffolds.
VEGF did not induce bone formation, did increase vascular network and in combination with BMP-2 significantly enhanced bone formation in ectopic assays. In femoral defects, it had no effect on vasculature and little effect on bone formation with BMP-2 after 8 weeks.
All VEGF released in first 3 days. Actual BMP-2 release ranged between 100 ng to 250 ng per day till 7 weeks.VEGF/BMP-2 = 0.01/200
New Zeland Rabbits
12 mm intra-orbital margin defect
Isolated from bone marrow of rabbits.
6 million cells
Combined use of VEGF and BMP-2.
Stem cells transduced with adenoviral particles and mixed as 1: 5 of VEGF:BMP-2 expressing cells.
Combined VEGF and BMP-2 delivery increased new bone deposition and formation compared to all other groups.
214142
Page 22
Xiao C, et al. (2011)
Loaded on natural coral scaffolds
~ 300 pg VEGF /million cells~300 pg BMP-2/million cells
VEGF/BMP-2 ratio = 1.
MF-1 nu/nu
5 mm critical size segmental defect in femur
Kanczler JM, et al. (2010)
Isolated from bone marrow of humans
0.2 million cells
Alginate PDLA composite scaffold
Combined use of VEGF and BMP-2.
20 µg VEGF and 20 µg BMP-2 added to scaffolds
Significant bone repair was observed only when VEGF and BMP-2 both were used with stem cells.
VEGF/BMP-2 ratio = 0.5.
B6 miceSub-cutaneous assays
Roldan JC, et al. (2010)
Isolated from bone marrow of B6 mice.
1 million cells
Loaded on calcium phosphate scaffolds.
Combined use of VEGF and BMP-7
VEGF and BMP-7 injected in the scaffolds
2 µg VEGF5 µg BMP-7
VEGF or stem cells did not increase BMP-7 induced osteogenesis.
VEGF/BMP-7 ratio = 0.4
22
365366
4344
