Continuous delivery of rhBMP2 and rhVEGF165 at a certain ratio enhances bone formation in mandibular defects over the delivery of rhBMP2 alone — An experimental study in rats

Journal of Controlled Release 220 (2015) 201–209

Contents lists available at ScienceDirect

Journal of Controlled Release journal homepage: www.elsevier.com/locate/jconrel

Continuous delivery of rhBMP2 and rhVEGF165 at a certain ratio enhances bone formation in mandibular defects over the delivery of rhBMP2 alone — An experimental study in rats N. Lohse a, N. Moser a, S. Backhaus b, T. Annen b, M. Epple b, H. Schliephake a,⁎ a b

Dept. of Oral and Maxillofacial Surgery, George-Augusta-University, Göttingen, Germany Inorganic Chemistry and Center for Nanointegration Duisburg-Essen (CeNIDE), University of Duisburg-Essen, Universitaetsstr. 5–7, 45117 Essen, Germany

a r t i c l e

i n f o

Article history: Received 15 July 2015 Received in revised form 3 October 2015 Accepted 16 October 2015 Available online 17 October 2015 Keywords: Bone morphogenic protein Vascular endothelial growth factor Polymer Retarded delivery Bone formation Angiogenesis

a b s t r a c t The aim of the present study was to test the hypothesis that different amounts of vascular endothelial growth factor and bone morphogenic protein differentially affect bone formation when applied for repair of non-healing defects in the rat mandible. Porous composite PDLLA/CaCO3 carriers were fabricated as slow release carriers and loaded with rhBMP2 and rhVEGF165 in 10 different dosage combinations using gas foaming with supercritical carbon dioxide. They were implanted in non-healing defects of the mandibles of 132 adult Wistar rats with additional lateral augmentation. Bone formation was assessed both radiographically (bone volume) and by histomorphometry (bone density). The use of carriers with a ratio of delivery of VEGF/BMP between 0.7 and 1.2 was significantly related to the occurrence of significant increases in radiographic bone volume and/or histologic bone density compared to the use of carriers with a ratio of delivery of ≤ 0.5 when all intervals and all outcome parameters were considered. Moreover, simultaneous delivery at this ratio helped to “save” rhBMP2 as both bone volume and bone density after 13 weeks were reached/surpassed using half the dosage required for rhBMP2 alone. It is concluded, that the combined delivery of rhVEGF165 and rhBMP2 for repair of critical size mandibular defects can significantly enhance volume and density of bone formation over delivery of rhBMP2 alone. It appears from the present results that continuous simultaneous delivery of rhVEGF165 and rhBMP2 at a ratio of approximately 1 is favourable for the enhancement of bone formation. © 2015 Published by Elsevier B.V.

1. Introduction Bone formation depends on the activity of a delicately orchestrated sequence of growth factors that coordinate angiogenesis, mesenchymal proliferation and osteogenic differentiation. A large body of research has assessed various dosages and temporal patterns of release of growth factors for the enhancement of bone repair [1–5]. Retarded delivery of either angiogenic or osteogenic growth factors has been in focus in the majority of these efforts in different experimental models [6–9]. In vivo applications of single growth factors such as bone morphogenic proteins (BMPs) have used high dosages which have been associated with considerable side effects such as implant loosening, swelling and inflammation [10–14]. In order to enhance the efficacy of growth factor delivery in bone repair and reduce the dosage of single growth factor applications, more recent strategies are directed towards the release

⁎ Corresponding author at: Dept. of Oral and Maxillofacial Surgery, George-AugustaUniversity, Robert-Koch-Str. 40, 37075 Göttingen, Germany. E-mail address: [email protected] (H. Schliephake).

http://dx.doi.org/10.1016/j.jconrel.2015.10.032 0168-3659/© 2015 Published by Elsevier B.V.

of more than just one growth factor [15–17]. Following the idea that angiogenic stimulation of proliferation of vessels into a bone defect precedes osteogenic differentiation of undifferentiated mesenchymal perivascular cells as sources for bone formation, most of these approaches have used the combined delivery of angiogenic and osteogenic signals at various dosage levels [18–20] with a concept of an increased early delivery of angiogenic signals followed by an increased release of osteogenic growth factors. So far, the results of these approaches have been inconclusive with respect to the role of angiogenic and osteogenic stimulation in a combined release fashion [20–24]. It was therefore felt desirable to use different dosages and different ratios of vascular endothelial growth factor (VEGF) and BMP in a combined delivery approach to test the hypothesis that different amounts of VEGF and BMP differentially affect bone formation when applied for repair of non-healing defects in the rat mandible. Moreover, the hypothesis was tested that the use of VEGF can reduce the amount of BMP to induce comparable bone volumes and density. For this purpose, rhVEGF165 and rhBMP2 were incorporated into porous composite poly-DL-lactic acid/CaCO3 carriers in ten different combinations and evaluated for their ability to enhance bone formation in a mandibular defect model.

202

N. Lohse et al. / Journal of Controlled Release 220 (2015) 201–209

2. Material & methods

2.2. Evaluation of growth factor release from the carriers

2.1. Preparation of retarded delivery devices

The carriers were evaluated in vitro for their release profile. Previous tests had shown that delivery occurred with an initial increase during the first 3 days with a subsequent gradual decline in delivery [25] which is considered to be based on superficial degradation/erosion of the polymer matrix. Blank carriers and carriers containing the a.m. amounts of rhBMP2/rhVEGF165 were immersed into cell culture medium (DMEM, 1 g Glc, Gibco, Invitrogen, www.invitrogen.de) in BSA coated 24 well plates. The medium was collected daily. Release after 1 day, 3 days, 7 days, and weekly thereafter until the 5th week was assessed by pooling the respective aliquots. The content of rhBMP2 was assessed using a custom made commercially produced sandwich ELISA. The minimum detectable amount was 100 pg/ml rhBMP2 (Dr. Mark Hennies, Euskirchen, Germany). rhVEGF165 was measured using a commercially available ELISA (Reliatech, Braunschweig, Germany) according to the instructions of the manufacturer. For all measurements, three carriers were evaluated at each interval. Measurements were performed twice on each carrier. In the Groups with combined growth factor loading, the ratio between released rhVEGF165 and rhBMP2 was calculated for each time point.

Poly-DL-lactic acid/CaCO3 composite granules were used for the fabrication of carriers for retarded release of growth factors. Gas foaming of this composite material in conjunction with lyophilized rhBMP2 had previously shown to produce porous carriers that provide effective retarded delivery of active rhBMP2 for 2 weeks [25]. At the same time, a drop in pH due to the occurrence of acidic degradation products during resorption that had been described before for the use of pure gas foamed poly-DL-lactide [26] was avoided by the addition of CaCO3. Sixteen grams of granular powder of amorphous poly-DL-lactic acid (PDLLA) (Resomer R 208, inherent viscosity: 1.8 dl/g, Boehringer, Ingelheim, Germany) and 4 g of spherulite-shaped precipitated calcium carbonate with the calcite crystal structure (average particle size 12 μm, Schaefer Kalk GmbH & Co. KG, Diez, Germany) were dry processed in an impact mill (NHS-0, Tokyo 143, Japan) at 5000 rpm and subsequently at 16.000 rpm [25]. The resulting granular powder (grain size: 200– 400 μm) was mixed with aqueous solutions of rhVEGF165 with 4, 25, and 100 μg/g composite (Groups 1A through 1C) and of rhBMP-2 at concentrations of 400, 800 and 1600 μg/g composite (Groups 2A through 2C). A third group received both growth factors at a lower dosage (400 μg rhBMP-2 and 25 μg rhVEGF165, respectively) and a higher dosage (800 μg rhBMP-2 and 100 μg rhVEGF165, respectively) in a cross over design (Groups 3A through 3D). The dosage levels of rhBMP2 were chosen based on previous experiments with pure PDLLA carriers showing that in vivo induction of bone tissue occurred at dosages of between 400µg per g polymer and the maximum applicable dosage was 1600 μg per g polymer [27,28]. The ratio of minimum dosages of rhBMP2 and rhVEGF165 was determined from the ratio of appr. 1/100 of in vitro levels of activity (2 ng/ml for rhVEGF165 and 300 ng/ml for rhBMP2 (data not shown)). The dosage levels for rhVEGF165 were varied to provide a moderate (6.25 fold) and a large (25 fold) dosage increase. The dosage levels for the cross over designs were restricted to two levels per growth factor in order to limit the number of experimental groups to a feasible extent. For rhBMP2, the lower two dosages (400 μg and 800 μg) were chosen to evaluate the potential of combined delivery with rhVEGF165 to induce a level of bone formation comparable to the maximum BMP2 dosage of 1600 μg per g polymer. This maximum dosage of rhBMP2 was used as a positive control. For rhVEGF165, the high and medium dosages (100 μg and 6 μg) were chosen to maximize possible effects of combined delivery as evaluation of delivery of VEGF alone during the course of the experiment showed that the release from medium dosage carriers approximated that of the low dosage carriers after day 3 (see below). Growth factors were purchased from ReliaTech (Braunschweig, Germany). All mixtures were subsequently lyophilized overnight. For each carrier, 0.06 g of the growth factor loaded granular powder was filled into custom made PTFE moulds for carriers of 8 mm diameter and 3 mm thickness as described previously [26] and was submitted to supercritical carbon dioxide (CO2 pressure: 100 bar for 2 h at 37 °C, filling time 20 min, soaking time 120 min, venting time: 20 min (5 bar/min)). The amounts of growth factors contained in the carriers of the respective groups are listed in Table 1. Blank PDLLA/CaCO3-carriers served as controls.

2.3. Surgical procedures/implantation of carriers The ability of the carriers to induce bone formation was assessed in a non-healing defect model in the rat mandible as described previously [27]. Briefly, carriers of 5 mm diameter were punched out of the 8 mm carrier discs. The carriers were inserted press fit into full thickness defects of 5 mm diameter in the ascending ramus of the mandibles of 120 adult male Wistar rats (weight range 330–680 g). The remaining carrier volume of the 8 mm discs was minced and used to augment the lateral side of the inserted carriers. Twelve additional animals served as controls with blank carriers being inserted on one side of the mandibles in an identical manner and empty defects on the opposite side. Each of the 10 combinations of growth factors was inserted unilaterally into the mandibular defects in 12 animals resulting in 12 carriers available for the evaluation of each growth factor dosage/dosage combination. Six carriers of each growth factor combination as well as 6 control animals were evaluated after 4 and 13 weeks each. At the end of each observation period, the mandibles were removed together with the surrounding soft tissue and fixated immediately in 4% buffered formalin. 2.4. Evaluation of bone formation Bone formation was assessed both radiographically for evaluation of bone volume (bone area) and by histomorphometry for assessment of bone quality (bone density). For radiographic evaluation of newly formed bone, the mandibles were split in the midline and each half mandible containing a carrier was submitted to volume computed tomography (VCT) (Orange Dental). Field of view was 50 × 50 mm, focus size was 500 μm at a maximum voltage/current of 120 kV/8 mA. The maximum voxel resolution was 80 μm. Radiographic analysis was performed on axial scans perpendicular to the defect and parallel to the lower border of the mandible. One scan through the center of the defect and two scans at a distance of 1/2 the radius above and below the center were used for radiographic evaluation of each defect.

Table 1 Growth factor content of carriers.

rhVEGF165 rhBMP2

Blank

Group 1A

Group 1B

Group 1C

Group 2A

Group 2B

Group 2C

Group 3A

Group 3B

Group 3C

Group 3D

0 0

0.24 μg 0

1.5 μg 0

6 μg 0

0 24 μg

0 48 μg

0 96 μg

1.5 μg 24 μg

6 μg 24 μg

1.5 μg 48 μg

6 μg 48 μg

N. Lohse et al. / Journal of Controlled Release 220 (2015) 201–209

Newly formed bone was clearly distinguishable from pristine bone and threshold values for counting pixels of newly formed bone were adjusted for every image under evaluation using visual control (Fiji (http://fiji. sc/) [29]. After calibration of the pixel size (0.016 mm2), results were expressed in mm2 and the values of each group of implants were averaged for each interval. For histomorphometric evaluation, non-decalcified thick sections were produced from the carriers after embedding into Technovit® according to Donath [30]. The specimens were surface stained with Alizarin-Methylene Blue and the density of the newly formed bone was evaluated using histomorphometric analysis of cross sections through the center of each defect. The specimens were mounted on a scanning table (Olympus dotSlide®, Olympus Life Science, Hamburg, Germany) and the micrographic images were recorded serially through a video camera (Sony 3CCD; Germany) at 100× magnification, digitized and mounted together into one picture. Pixel sizes were calibrated and the area occupied by newly formed bone was assessed by pixel counting using commercially available software (Adobe Photoshop CS5, Adobe Systems GmbH, München). Histological bone formation was assessed as bone density (percentage of bone area covering the image frame area). 2.5. Statistics Repeated measures analyses of variance (ANOVA) (IBM SPSS Statistics 21.0, http://support.spss.com) were used to compare the ratios of VEGF/BMP delivery, bone volume and bone density. The association between the ratio of delivery and significant increase in bone volume or bone density at the two intervals was tested using Chi-square test. The significance level was set to p b 0.05. 3. Results 3.1. Growth factor release The carriers loaded with rhVEGF165 only (Groups 1A through C) exhibited an initial dose dependent delivery of growth factor during the first three days with a high initial release of 250 ng/ml in the high dosage group (6 μg rhVEGF165), 60 ng/ml in the medium dosage group (1.5 μg rhVEGF165) and b10 ng/ml in the low dosage group (0.24 μg rhVEGF165). Thereafter, the delivery from carriers with 1.5 μg approximated that of carriers with 0.24 μg and arrived at a constant low level of release b10 ng/ml per week already at the end of the first week, whereas the high dosage carriers maintained a higher level of release of approx. 50 ng/ml per week until the end of week 5 (Fig. 1A). The carriers loaded with rhBMP2 alone (Groups 2A through C) exhibited a concentration dependent delivery that was characterized by an initial release between 500 and 700 ng/ml during the first 24 h with an increase in delivery during the first week in the high dosage group (96 μg rhBMP2) and a slowly decreasing delivery to 150 ng/ml until the end of the observation period. The carriers with 48 μg rhBMP2 reached this point after 4 weeks, carriers from the low dosage group (24 μg/carrier) arrived at 150 ng/ml per week at the end of the third week (Fig. 1B). The delivery of rhVEGF165 from the carriers with combined loading followed this pattern, too, however, the amount of rhVEGF165 delivered was approximately four times higher at the individual intervals in both VEGF dosage groups than in the carriers with rhVEGF165 only loading (Fig. 1C). The profile of BMP delivery from the carriers with combined loading was similar with a dose dependent level of release that slowly decreased towards weeks 3 and 4 with a release of 100–150 ng/ml per week (Fig. 1D). The ratio of the amounts of rhVEGF165 and rhBMP2 released from the carriers with combined loading showed a clear distinction between the high (6 μg) and the low (1.5 μg) VEGF dosage groups. The carriers from the former groups exhibited an initial ratio of 2.5 to 3 at day 1 and rather

203

constant value of 0.7–1.2 thereafter. Carriers with the low VEGF dosage showed a ratio of rhVEGF165/rhBMP2 delivery between 0.7 and 0.5 at day 1 and a decreasing ratio of 0.5 to 0.04 thereafter to the end of the observation period (Fig. 1E). Ratios of delivery from the two high VEGF dosage groups were not significantly different between the individual intervals (p = 0.300) after day 1, whereas the decrease in the ratio of delivery over time in the low VEGF dosage groups was significant (p b 0.001). Consequently, the ratios of delivery from the high VEGF groups was significantly different from the ratios of the low VEGF dosage carriers (p = 0.003). 3.2. Radiographic bone formation Surface rendering of the mandibles exhibited hardly any bone formation in empty defects (Fig. 2A). Only minor bone formation was observed at the margins of defects filled with blank carriers (Fig. 2B). There was incomplete bone formation in a few defects filled with rhVEGF165 only loaded carriers (Fig. 2C through E). Defects filled with rhBMP2 only exhibited complete bone fill of the defects and additional bone formation originating from the minced carrier material used for augmentation on the lateral side of the filled defect (Fig. 2F through H). Carriers with combined loading with rhBMP2 and rhVEGF165 exhibited higher volumes of bone formation (Fig. 2I through M) that surpassed that of carriers loaded with the corresponding dosage of rhBMP2 alone in Group 3D (Fig. 2M). 3.3. Fig. 2 exhibiting surface rendering of radiographic bone formation after 4 weeks Bone formation in the empty defects had filled an area of 2.25 mm2 (SD 0.59) on average after 4 weeks and 2.26 mm2 (SD 0.99) after 13 weeks in the sections evaluated. Blank carriers exhibited 0.93 mm2 (SD 0.31) of bone formation after 4 weeks and 2.93 mm2 (SD 1.39) after 13 week (Fig. 3). There was no significant difference between the two groups at neither interval. Carriers with rhVEGF165 loading only showed an average area of bone formation between 7.70 and 21.09 mm2 after 4 weeks and between 8.83 to 10.78 after 13 weeks. After 4 weeks, the area of newly formed bone was significantly larger in Groups 1A and 1C (p b 0.001 and p = 0.028, respectively) compared to blank carriers and larger than empty defects in Group 1A (p = 0.001). After 13 weeks, mean area of newly formed bone in Group 1A was significantly larger than empty defects and blank carriers (p = 0.012 and 0.024, respectively), but not in the other dosage groups. Significant differences were found between the Group 1A and the other two VEGF dosage groups after 4 weeks, whereas after 13 weeks there was no difference between the carriers with of rhVEGF165 loading only. Carriers loaded with rhBMP2 alone had induced bone formation after 4 weeks, covering an average area between 60.91 mm2 (SD 15.41) in the low dosage carriers (24 μg rhBMP2) and 77.46 mm2 (SD 5.38) in the high dosage carriers (96 μg rhBMP2). This was significantly larger than both the blank control and the empty control group (p b 0.001). No significant difference was found between the different dosages of rhBMP2 at this interval. After 13 weeks, again a significantly larger mean area of bone formation was found in all three dosage groups compared to the controls (p b 0.001). Moreover, the mean area of radiographic bone formation after 13 weeks was significantly larger in the medium and high dosage groups (48 μg and 96 μg rhBMP2) than in the low dosage carriers (24 μg rhBMP2) (p = 0.010 and 0.021). The area of bone formation in Groups 3A and B (combined loading of 24 μg rhBMP2 with 1.5 μg and 6 μg rhVEGF165, respectively) was not significantly larger after 4 weeks than in the carriers loaded with 24 μg rhBMP2 alone. After 13 weeks, a significant increase (p = 0.049) was found for 24 μg rhBMP2 combined with 6 μg rhVEGF165 compared to 24 μg rhBMP2 alone. In Groups 3C and D (48 μg rhBMP2 combined with 1.5 μg and 6 μg rhVEGF165, respectively) bone formation was

204

N. Lohse et al. / Journal of Controlled Release 220 (2015) 201–209

Fig. 1. In vitro release of growth factors. A: rhVEGF165 release from carriers with rhVEGF165 loading only (Groups 1A through C); B: rhBMP2 release from carriers with rhBMP2 loading only (Groups 2A through C); C: rhVEGF165 release from combined loading with rhVEGF165 and rhBMP2 (Groups 3A through D); D: rhBMP2 release from combined loading with rhVEGF165 and rhBMP2 (Groups 3A through D); E: Ratio of rhVEGF165/rhBMP2 delivery.

enhanced after 4 weeks in both groups compared to 48 μg rhBMP2 alone, where the addition of 6 μg rhVEGF165 resulted in a significant increase (p = 0.002), which was even significantly higher than the area induced by 96 μg rhBMP2 (p = 0.022). After 13 weeks, this effect had levelled off and no significant differences were found between the Groups 3C/D and the respective rhBMP dosage alone.

3.4. Histologic and histometric bone density in vivo Histologic features of bone formation after 4 weeks are displayed in Fig. 4A through M and after 13 weeks in Fig. 5A through M. Lower density of early bone formation is visible in the specimens after 4 weeks with large areas covered particularly in Group 3C and D. After 13 weeks, denser bone formation with thicker bone trabecula and a smoother outer contour in all specimens indicated a remodelled state of bone formation. Quantitative assessment revealed low bone density values in the control carriers and carriers loaded with rhVEGF165 alone reflecting the low degree of radiographic bone formation. Empty defects and blank carriers showed bone density values between 0.59 and 1.8% after 4 and 13 weeks, respectively (Fig. 6).

Carriers with rhVEGF165 loading exhibited between 4.4 and 7.3% after 4 weeks and 2.7 to 4.6% after 13 weeks depending on the amount of VEGF incorporated. The difference between blank carriers and rhVEGF165 loaded carriers was significant after 4 weeks only for carriers loaded with 1.5 μg rhVEGF165 (p = 0.005). After 13 weeks no significant difference in bone density was found for any of the VEGF groups. At neither interval was there a difference between the three dosage levels of rhVEGF165. In carriers with rhBMP2 loading alone, bone density exhibited a highly significant increase compared to empty defects and blank carriers at both intervals (p b 0.001). There was a dose dependent increase after 4 weeks from 17.4% (24 μg rhBMP2) to 23.5% (96 μg rhBMP2), with significant differences between 96 μg vs. 48 μg (p = 0.0204) as well as 96 μg vs. 24 μg rhBMP2 (p = 0.0316). After 13 weeks, dose dependent differences in bone density were not present anymore between the carriers loaded with rhBMP2 only (Fig. 6). In carriers with combined loading, a significantly enhanced mean density value compared to the corresponding rhBMP2 dosage alone was found after 4 weeks in Group 3B (6 μg rhVEGF165/24 μg rhBMP2) (p = 0.0157)). After 13 weeks, three groups of carriers (6 μg rhVEGF165/24 μg rhBMP2, 1.5 μg rhVEGF165/48 μg and 6 μg rhVEGF165/ 48 μg rh BMP2) showed significantly increased bone density over the respective rhBMP2 dosage alone (p = 0.0391, 0.0350, 0.0146, respectively).

N. Lohse et al. / Journal of Controlled Release 220 (2015) 201–209

205

Fig. 2. Radiographic assessment of bone formation after 4 weeks. Mandibles were placed upright with the occlusal plane of mandibular teeth plane oriented horizontally (arrow heads). A lateral view to the defect area (arrows) in the mandibular angle area and to the augmentation overlying the defect is chosen, Scale bar: 500 μm; A: Empty defect B: Blank carriers, C: Carrier with 0.24 μg rhVEGF165 (Group 1A) D: Carriers with 1.5 μg rhVEGF165 (Group 1B); E: Carrier with 6.0 μg rhVEGF165 (Group 1C); F: Carrier with 24 μg rhBMP2 (Group 2A); G: Carrier with 48 μg rhBMP2 (Group 2B); H: Carrier with 96 μg rhBMP2 (Group 2C); I: Carrier with 1.5 μg rhEVGF165 and 24 μg rhBMP2 (Group 3A); K: Carrier with 6.0 μg rhEVGF165 and 24 μg rhBMP2 (Group 3B); L: Carrier with 1.5 μg rhEVGF165 and 48 μg rhBMP2 (Group 3C); M: Carrier with 6.0 μg rhEVGF165 and 48 μg rhBMP2 (Group 3D);

The use of carriers with a ratio of delivery of VEGF/BMP between 0.7 and 1.2 (6 μg rhVEGF165 in conjunction with lower (24 μg) and higher (48 μg) dosages) was significantly related to the occurrence of significant increases in radiographic bone volume and/or histologic bone density (p = 0.039) compared to the use of carriers with a ratio of delivery of VEGF/BMP ≤ 0.5 when all intervals and all outcome parameters were considered.

4. Discussion The present study has used different amounts of rhVEGF165 and rhBMP2 as well as combinations of both growth factors to assess the effect of combined VEGF/BMP delivery in comparison to the delivery of these growth factors alone. Carriers with 0.4 μg and 6 μg rhVEGF165 alone had resulted in a small but significant increase in radiographic

Fig. 3. Radiographic evaluation of bone volume (bone area); * p b 0.05 in comparison between carriers with combined loading of rhVEGF165/rhBMP2 and corresponding rhBMP2 loading alone.

206

N. Lohse et al. / Journal of Controlled Release 220 (2015) 201–209

Fig. 4. Micrographs of cross sections through the area of bone formation in the mandible after 4 weeks; Alizarine Red/Methylene Blue stain. Bar 200 μm. A: Empty defect: lack of bone fill B: Blank carriers: Very little bone formation is seen, emerging from the defect margin; C: Carriers with 0,24 μg rhVEGF165 (Group 1A): Very little bone formation at the defect margins; D: Carriers with 1.5 μg rhVEGF165 (Group 1B): Osteoconductive bone formation along the surface of the carrier; E: Carrier with 6 μg rhVEGF165 (Group 1C): Negligible bone formation at defect margins; F: Carrier with 24 μg rhBMP2 (Group 2A): Young immature bone formation surrounds the carrier parts; G: Carrier with 48 μg rhBMP2 (Group 2B): Formation of young immature bone around and between carrier parts, starting to invade the carrier pores; Note bone formation between minced carrier material used for lateral augmentation. H: Carriers with 96 μg rhBMP2 (Group 2C): Loose bone formation in greater distance to the carrier; I: Carriers with 1.5 μg rhVEGF165 and 24 μg rhBMP2 (Group 3A): Loose bone formation in the closer vicinity of the carrier; K: Carriers with 6 μg rhVEGF165 and 24 μg rhBMP2 (Group 3B): Marked increase in bone formation also at larger distance to the carrier surrounding the augmented area L: Carriers with 1.5 μg rhVEGF165 and 48 μg rhBMP2 (Group 3C) and M: Carriers with 6 μg rhVEGF165 and 48 μg rhBMP2 (Group 3D): Formation of immature bone surrounding the carrier but also at greater distance to the carrier parts; note extensive soft tissue penetration of the carrier parts. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

bone volume compared to blank carriers and empty defects. Histologic bone density was only increased in one group after 4 weeks. RhBMP2 alone had elicited a much stronger highly significantly increase in both volume and density of newly formed bone compared to empty defects and blank controls in all dosage groups at both intervals. Compared to previous results reported for the use of gas foamed pure PDLLA carriers [25], the use of CaCO3-PDLLA scaffolds in the present study has improved the rate of bone formation at the lower dosage levels. The decreased rate of bone formation in pure PDLLA carriers has been considered to be accounted for resorption of newly formed bone due to the decrease in periimplant pH during degradation previously reported for the use of pure PDLLA scaffolds loaded with rhBMP2 [26]. In vitro evaluations of the CaCO3-PDLLA scaffolds had shown that this decrease did not occur in the composite scaffolds used in the present study [27]. The quantitative results of bone formation for rhVGEF165 and rhBMP2 moreover suggest that rhBMP2 would exert a major effect in enhancement of bone formation compared to VEGF, when considered as effects of isolated application of one factor alone. In the present model, different dosages of rhBMP2 did not have a differential effect on the volume of newly formed bone at early stages after

4 weeks but significant differences between the low dosage (24 μg) and medium (48 μg)/high (96 μg) dosages were seen later after 13 weeks. In contrast, bone density had shown an early dose dependent increase after 4 weeks with high amounts of rhBMP2 (96 μg) resulting in significantly increased values but dose dependent effects of rhBMP2 on bone density were not visible after 13 weeks from BMP delivery alone. When bone volume and bone density resulting from the application of rhBMP2 alone were compared to the values resulting from the combined delivery of rhBMP2 and rhVEGF165, it appeared that the addition of 6 μg rhVEGF165 to rhBMP2 had an enhancing effect on both the amount and the density of newly formed bone. When comparing the results of combined VEGF/BMP delivery to the corresponding BMP dosage, carriers with a dosage of 6 μg of rhVEGF165 exhibited a significant increase in bone volume after 4 weeks in combination with 48 μg of rhBMP2 and after 13 weeks in combination with 24 μg rhBMP2. Moreover, bone density around carriers with 6 μg of rhVEGF165 were significantly increased after 4 weeks in combination with 24 μg rhBMP2 and after 13 weeks in combination with both 24 μg and 48 μg rhBMP2 when compared to the corresponding BMP dosage alone. The magnitude of increase exceeded the level that would have resulted from a

N. Lohse et al. / Journal of Controlled Release 220 (2015) 201–209

207

Fig. 5. Micrographs of cross sections through the area of bone formation in the mandible after 13 weeks; Alizarine Red/Methylene Blue stain. Bar: 200 μm. A: Empty defect: Negligible bone formation; B: Blank carriers (Control Group): Negligible bone formation along the carrier surface is visible; C: Carriers with 0,24 μg rhVEGF165 (Group 1A): Very little bone formation with rounded defect margins; D: Carriers with 1.5 μg rhVEGF165 (Group 1B): Remodelled bone islands along the carrier surface; E: Carriers with 6 μg rhVEGF165 (Group 1C): Bone formation at the surface of the carrier with commencing penetration into the carrier from the anterior margin; F: Carriers with 24 μg rhBMp2 (Group 2A): Dense bone contour and thorough penetration of the carrier parts by ingrowing bone; G: Carriers with 48 μg rhBMP2 (Group 2B): Integration of carrier parts into the newly formed bone and remodelling of denser bone irrespective of carrier parts. H: Carriers with 96 μg rhBMP2 (Group 2C): Dense bone formation engulfing all carrier parts in greater distance to the carrier; I: Carriers with 1.5 μg rhVEGF165 and 24 μg rhBMP2 (Group 3A): Shrinkage in bone contour with formation of a thicker outer bone layer and dense penetration of carrier pores; K: Carriers with 6 μg rhVEGF165 and 24 μg rhBMP2 (Group 3B); L: Carriers with 1.5 μg rhVEGF165 and 48 μg rhBMP2 (Group 3C): Thick bone formation in parts with smoother outer contour of bone formation; commencing degradation of the carrier M: Carriers with 6 μg rhVEGF165 and 48 μg rhBMP2 (Group 3D): Well defined outer contour with individual carrier parts protruding through the contour; mature bone trabecula are visible throughout the carrier parts. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 6. Histometric evaluation of bone quality (bone density); * p b 0.05 in comparison between carriers with combined loading of rhVEGF165/rhBMP2 and corresponding rhBMP2 loading alone.

208

N. Lohse et al. / Journal of Controlled Release 220 (2015) 201–209

simple “addition” of the delivery of BMP and VEGF alone suggesting that the combination of both growth factors is synergistic and the function of VEGF is not only limited to the initial enhancement of angiogenesis but may also be contributory for osteogenesis in the cross-talk between VEGF and BMP signalling pathways as it has been hypothesized by Cui and coworkers [31]. It is also interesting to note that the addition of 1.5 μg rhVEGF165 to the two dosages of rhBMP2 did not have a significant effect on bone volume formation at any interval nor on bone density after 4 weeks. Only after 13 weeks did the combination of 1.5 μg rhVEGF165 with 48 μg rhBMP2 result in increased bone density. The effect of different combinations of growth factors can be assumed to be based on the magnitude and the quantitative ratio of their delivery. As the release curves of rhBMP2 have shown comparable characteristics in the groups with BMP only and with combined loading, differences in bone formation resulting from carriers with combined growth factor loading would be most likely based on the additional delivery of VEGF. The two dosage levels of rhVEGF165 that had been combined in a cross over design with two dosages of rhBMP2 have resulted in rather constant and distinct ratios of in vitro delivered amounts of growth factors after day 1. The ratio of in vitro delivery of rhVEGF165/ rhBMP2 after day 1 was approximately 1 (0.7–1.2) for the carriers with 6 μg rhVEGF165 whereas carriers with 1.5 μg rhVEGF165 delivered the two growth factors at a ratio of ≤ 0.5 (0.5–0.04) rhVEGF165/ rhBMP2. Under the assumption that the characteristics of in vivo delivery are comparable to the in vitro findings, the results suggest that a continuous ratio of VEGF/BMP delivery of about 1 is beneficial for the volume and the density of newly formed bone compared to a ratio ≤ 0.5 at the dosage levels used. In early osteogenesis after 4 weeks, the effect of the ratio of 1 on bone volume appears to occur in conjunction with higher amounts of BMP (48 μg), whereas at later stages, lower dosages (24 μg) may benefit from the addition of rhVEGF165 reaching a level comparable to that achieved by twice as high amounts of rhBMP2 alone. Hence, considerable amounts of BMP could potentially be saved by addition of rhVEGF165 at an appropriate ratio of delivery. A possible reason for the difference in bone volume at early vs. late stages of bone formation may be that in early stages, the high native signalling level in regenerating tissue may require higher levels of additional signals to induce a significant effect in bone volume, whereas the conditions after 13 weeks may have arrived at a more stable level where volume gain has reached a maximum with the delivery of 48 μg rhBMP2 alone that is not surpassed by even higher dosages of BMP alone (96 μg) but can be reached by addition of VEGF to a lower much dosage (24 μg) of rhBMP2. The results of bone density also point into this direction suggesting that tissue programming at later stages is directed towards remodelling and that the addition of rhVEGF165 to rhBMP2 increased bone density even to a greater extent than the increase in bone volume. There may also be a threshold level in growth factor concentrations that induces denser bone formation and that under these conditions even lower amounts of VEGF may induce an increase in bone density, as it has been the case for the combination of 1.5 μg rhVEGF165 and 48 μg rhBMP2. This does not contradict the fact that in all other combination groups and intervals only the carriers with a ratio of delivery of 0.7– 1.2 had induced significant increases in bone formation. The present results on the one hand are partially in accordance with those of Patel et al. [24] and Hernandez et al. [32] reporting a synergistic effect of simultaneous delivery of VEGF and BMP2 after 4 weeks on bone volume in intramedullary femur defects [32] and bone density in calvarial defects [24]. In particular Hernandez et al. had shown that a slow release of VEGF in conjunction with higher dosages of BMP2 may be beneficial. On the other hand, the results of the present study are in contrast to other previous studies reporting that the combined use of angiogenic growth factors and BMP has failed to improve bone formation in orthotopic sites [22,23,33]. Wang and coworkers even reported an inhibitory effect of the use of bFGF as an angiogenic growth factor in combination with BMP2 on bone formation [19].

The reasons for contradictory results may be different intervals of evaluation and differences in release characteristics of the carriers used. Young et al., who had used the same model and carrier material as Patel et al. [24] with very similar growth factor amounts, had looked at bone formation after 12 weeks, which may have been too late to see differences in pure bone volume. In the study of Kempen et al. on femoral defects in rats, an early short termed VEGF release compared to the present study had not increased vessel density in orthotopic sites indicating that a more retarded release may be required to elicit a biological reaction in orthotopic sites, which is also supported by the results of Hernandez et al. [32]. The same holds true for the combination of bFGF and BMP2, as Su et al. [18] have reported enhanced bone formation after simultaneous slow release of both growth factors. It thus appears that a continuous delivery of angiogenic and osteogenic growth factors is required to enhance bone formation in orthotopic sites rather than an early high delivery of angiogenic signals followed by high osteogenic signal release. However, it appears that there are site specific differences as both Geuze et al. [22] and Kempen et al. [23] have found synergistic effects of the combined use of VEGF and BMP2 in ectopic sites but not in orthotopic sites with identical dosages and patterns of release of growth factors used in both sites. Different patterns of cell populations that require different patterns of delivery may explain why different sites (ectopic vs. orthotopic) have shown contradictory results with the same temporal pattern of growth factor release. In conclusion, the present study has shown that the combined delivery of rhVEGF165 and rhBMP2 for repair of critical size mandibular defects can significantly enhance volume and density of early and late bone formation over delivery of rhBMP2 alone. It appears from the present results that continuous simultaneous delivery of rhVEGF165 and rhBMP2 at a ratio of approximately 1 is favourable for the enhancement of bone formation. Moreover, the results suggest that the simultaneous delivery at this ratio can help to “save” rhBMP2 as both bone volume and bone density after 13 weeks were reached/surpassed using half the dosage required for rhBMP2 alone. Acknowledgements The authors greatly value the help of Mrs. Jutta Schulz during the laboratory experiments. This work has been funded by a Grant from the Federal Ministry of Education and Research (BMBF) (No. 13N10003). References [1] A. Lochmann, H. Nitzsche, S. von Einem, E. Schwarz, K. Mäder, The influence of covalently linked and free polyethylene glycol on the structural and release properties of rhBMP-2 loaded microspheres, J. Control. Release 147 (2010) 92–100. [2] M. Ye, S. Kim, K. Park, Issues in long-term protein delivery using biodegradable microparticles, J. Control. Release 146 (2010) 241–260. [3] J. Hou, J. Wang, L. Cao, X. Qian, W. Xing, J. Lu, C. Liu, Segmental bone regeneration using rhBMP-2-loaded collagen/chitosan microspheres. Composite scaffold in a rabbit model, Biomed. Mater. 7 (3) (2012 Jun) 035002. [4] R. Tan, Z. She, M. Wang, X. Yu, H. Jin, Q. Feng, Repair of rat calvarial bone defects by controlled release of rhBMP-2 from an injectable bone regeneration composite, J. Tissue Eng. Regen. Med. 6 (2012) 614–621. [5] N. Fujita, T. Matsushita, K. Ishida, K. Sasaki, S. Kubo, T. Matsumoto, M. Kurosaka, Y. Tabata, R. Kuroda, An analysis of bone regeneration at a segmental bone defect by controlled release of bone morphogenetic protein 2 from a biodegradable sponge composed of gelatin and β-tricalcium phosphate, J. Tissue Eng. Regen. Med. 6 (2012) 291–298. [6] N. Kalaji, A. Deloge, N. Sheibat-Othmann, O. Boyron, I. About, H. Fessi, Controlled release carriers of growth factors FGF-2 and TGFbeta1: synthesis, characterization and kinetic modelling, J. Biomed. Nanotechnol. 6 (2010) 106–116. [7] Z.Y. Lin, Z.X. Duan, X.D. Guo, J.F. Li, H.W. Lu, Q.X. Zheng, D.P. Quan, S.H. Yang, Bone induction by biomimetic PLGA-(PEG-ASP)n copolymer loaded with a novel synthetic BMP-2 related peptide in vitro and in vivo, J. Control. Release 144 (2010) 190–195. [8] J.M. Kanczler, J. Barry, P. Ginty, S.M. Howdle, K.M. Shakesheff, R.O. Oreffo, Supercritical carbon dioxide generated vascular endothelial growth factor encapsulated poly(DL-lactic acid) scaffolds induce angiogenesis in vitro, Biochem. Biophys. Res. Commun. 352 (1) (5 2007) 135–141. [9] B. De la Riva, C. Nowak, E. Sanchez, A. Hernandez, M. Schulz-Siegmund, M.K. Pec, A. Delgado, C. Ecora, VEGF controlled release within a bone defect from alginate/ chitosan/PLA-H scaffolds, Eur. J. Pharm. Biopharm. 73 (2009) 50–58.

N. Lohse et al. / Journal of Controlled Release 220 (2015) 201–209 [10] K.N. Leknes, J. Yang, M. Qahash, G. Polimeni, C. Susin, U.M. Wikesjö, Alveolar ridge augmentation using implants coated with recombinant human bone morphogenetic protein-2: radiographic observations, Clin. Oral Implants Res. 19 (2008) 1027–1033. [11] O.E. Beidas, J.C. Deschamps-Braly, A.M. Morgan, M.C. Workman, C.D. Knotts, A.D. Denny, C.A. El Amm, Safety and efficacy of recombinant human bone morphogenetic protein 2 on cranial defect closure in the pediatric population, J. Craniomaxillofac. Surg. 24 (2013) 917–922. [12] F. Shweikeh, G. Hanna, L. Bloom, E.T. Sayegh, J. Liu, F.L. Acosta, D. Drazin, Assessment of outcome following the use of recombinant human bone morphogeneticprotein-2 for spinal fusion in the elderly population, J. Neurosurg. Sci. (2014) (Epub ahead of print). [13] M.T. Merrick, K.D. Hamilton, S.S. Russo, Acute epidural lipedema: a novel entity and potential complication of bonemorphogenetic protein use in lumbar spine fusion, Spine J. 13 (2013) e15–e19. [14] R.D. Muchow, W.K. Hsu, P.A. Anderson, Histopathologic inflammatory response induced by recombinant bone morphogeneticprotein-2 causing radiculopathy after transforaminal lumbar interbody fusion, Spine J. 10 (9) (2010 Sep) e1–e6. [15] M. Farokhi, F. Mottaghitalab, J. Ai, M.A. Shokrgozar, Sustained release of plateletderived growth factor and vascular endothelial growth factor from silk/calcium phosphate/PLGA based nanocomposite scaffold, Int. J. Pharm. 454 (2013) 216–225. [16] X. Sun, Y. Kang, J. Bao, Y. Zhang, Y. Yang, X. Zhou, Modeling vascularized bone regeneration within a porous biodegradable CaP scaffold loaded with growth factors, Biomaterials 34 (21) (Jul 2013) 4971–4981. [17] B. De la Riva, E. Sánchez, A. Hernández, R. Reyes, F. Tamimi, E. López-Cabarcos, A. Delgado, C. Evora, Local controlled release of VEGF and PDGF from a combined brushite–chitosan system enhances bone regeneration, J. Control. Release 143 (1) (2010) 45–52. [18] J. Su, H. Xu, J. Sun, X. Gong, H. Zhao, Dual delivery of BMP-2 and bFGF from a new nano-composite scaffolds, loaded with vascular stents for large size mandibular defect regeneration, Int. J. Mol. Sci. 14 (2013) 12714–12728. [19] H. Wang, Q. Zou, O.C. Boerman, A.W. Nijhuis, J.A. Jansen, Y. Li, S.C. Leeuwenburgh, Combined delivery of BMP-2 and bFGF from nanostructured colloidal gelatin gels and its effect on bone regeneration in vivo, J. Control. Release 166 (2) (Mar 10 2013) 172–181. [20] R. Reyes, B. De la Riva, A. Delgado, A. Hernández, E. Sánchez, C. Évora, Effect of triple growth factor controlled delivery by a brushite–PLGA system on a bone defect, Injury 43 (3) (Mar 2012) 334–342. [21] C. Schmitt, R. Lutz, H. Doering, M. Lell, J. Ratky, K.A. Schlegel, Bio-Oss® blocks combined with BMP-2 and VEGF for the regeneration of bony defects and vertical augmentation, Clin. Oral Implants Res. 24 (4) (2013 Apr) 450–460.

209

[22] R.E. Geuze, L.F. Theyse, D.H. Kempen, H.A. Hazewinkel, H.Y. Kraak, F.C. Oner, W.J. Dhert, J. Alblas, A differential effect of bone morphogenetic protein-2 and vascular endothelial growth factor release timing on osteogenesis at ectopic and orthotopic sites in a large-animal model, Tissue Eng. A 18 (2012) 2052–2062. [23] D.H. Kempen, L. Lu, A. Heijink, T.E. Hefferan, L.B. Creemers, A. Maran, M.J. Yaszemski, W.J. Dhert, Effect of local sequential VEGF and BMP-2 delivery on ectopic and orthotopic bone regeneration, Biomaterials 30 (2009) 2816–2825. [24] Z.S. Patel, S. Young, Y. Tabata, J.A. Jansen, M.E. Wong, A.G. Mikos, Dual delivery of an angiogenic and an osteogenic growth factor for bone regeneration in a critical size defect model, Bone 43 (2008) 931–940. [25] H. Schliephake, J. Boven, S. Backhaus, T. Annen, M. Epple, Solvent free production of porous PDLLA/calcium carbonate composite scaffolds for controlled release of rhBMP2 and rhVEGF165, Oral Maxillofac. Surg. 19 (2015) 133–141. [26] H. Schliephake, H. Weich, J. Schulz, H. Gruber, In-vitro characterization of a slow release system of polylactid acid and rhBMP-2, J. Biomed. Mater. Res. 83 (2007) 455–462. [27] H. Schliephake, H. Weich, C. Dullin, H. Gruber, S. Frahse, Mandibular reconstruction by implantation of rhBMP in a slow release system of polylactic acid, Biomaterials 29 (2008) 103–110. [28] R.M. Gruber, H. Weich, H. Schliephake, Ectopic bone formation after implantation of a slow release system of polylactid acid and rhBMP2, Clin. Oral Implants Res. 20 (1) (2009) 24–30. [29] J. Schindelin, I. Arganda-Carreras, E. Frise, V. Kaynig, M. Longair, T. Pietzsch, S. Preibisch, C. Rueden, S. Saalfeld, B. Schmid, J.Y. Tinevez, D.J. White, V. Hartenstein, K. Eliceiri, P. Tomancak, A. Cardona, Fiji: an open-source platform for biologicalimage analysis, Nat. Methods 9 (7) (Jun 28 2012) 676–682. [30] K. Donath, The diagnostic value of the new method for the study of undecalcified bones and teeth with attached soft tissue (Sage–Schliff (sawing and grinding) technique), Pathol. Res. Pract. 179 (1985) 631–633. [31] Q. Cui, A.S. Dighe, J.N. Irvine Jr., Combined angiogenic and osteogenic factor delivery for bone regenerative engineering, Curr. Pharm. Des. 19 (19) (2013) 3374–3383. [32] A. Hernandez, R. Reyes, E. Sanchez, M. Rodriguez-Evora, A. Delgado, C. Evora, In vivo osteogenic response to different ratios of BMP-2 and VEGF released from a biodegradable porous system, J. Biomed. Mater. Res. A 100A (2012) 2382–2391. [33] S. Young, Z.S. Patel, J.D. Kretlow, M.B. Murphy, P.M. Mountziaris, L.S. Baggett, H. Ueda, Y. Tabata, J.A. Jansen, M. Wong, A.G. Mikos, Dose effect of dual delivery of vascular endothelial growth factor and bone morphogenetic protein-2 on bone regeneration in a rat critical-size defect model, Tissue Eng. A 15 (2009) 2347–2362.