Regenerating Critical Size Rat Segmental Bone Defects with a ...

RESEARCH ARTICLEwww.mbs-journal.de

Regenerating Critical Size Rat Segmental Bone Defects witha Self-Healing Hybrid Nanocomposite Hydrogel: Effect ofBone Condition and BMP-2 Incorporation

Claire I. A. van Houdt, Marianne K. E. Koolen, Paula M. Lopez-Perez,Dietmar J. O. Ulrich, John A. Jansen, Sander C. G. Leeuwenburgh, Harrie H. Weinans,and Jeroen J. J. P. van den Beucken*

The aim of the current study is to assess the biological performance ofself-healing hydrogels based on calcium phosphate (CaP) nanoparticles andbisphosphonate (BP) conjugated hyaluronan (HA) in a critical size segmentalfemoral bone defect model in rats. Additionally, these hydrogels are loadedwith bone morphogenetic protein 2 (BMP-2) and their performance iscompared in healthy and osteoporotic bone conditions. Treatment groupscomprise internal plate fixation and placement of a PTFE tube containinghydrogel (HABP-CaP) or hydrogel loaded with BMP-2 in two dosages(HABP-CaP-lowBMP2 or HABP-CaP-highBMP2). Twelve weeks after bonedefect surgery, bone formation is analyzed by X-ray examination, micro-CTanalysis, and histomorphometry. The data show that critical size, segmentalfemoral bone defects cannot be healed with HABP-CaP gel alone. Loading ofthe HABP-CaP gel with low dose BMP-2 significantly improve bone formationand resulted in defect bridging in 100% of the defects. Alternatively, high doseBMP-2 loading of the HABP-CaP gel does not improve bone formation withinthe defect area, but leads to excessive bone formation outside the defect area.Bone defect healing is not affected by osteoporotic bone conditions.

Dr. C. I. A. van HoudtBiomaterialsPlastic, Reconstructive and Hand SurgeryRadboudumcGeert Grooteplein Zuid 10, Nijmegen, Gelderland 6525 GA, TheNetherlands

The ORCID identification number(s) for the author(s) of this articlecan be found under https://doi.org/10.1002/mabi.202100088

© 2021 The Authors. Macromolecular Bioscience published byWiley-VCH GmbH. This is an open access article under the terms of theCreative Commons Attribution-NonCommercial-NoDerivs License,which permits use and distribution in any medium, provided the originalwork is properly cited, the use is non-commercial and no modificationsor adaptations are made.

DOI: 10.1002/mabi.202100088

1. Introduction

Severe trauma, bone tumors, congenitalmalformation, or extensive infection ofbone tissue can be the cause of critical sizebone defects. The term critical size bone de-fect is defined as an orthotopic intraosseouswound that does not spontaneously healwithout surgical intervention.[1] Urist con-cluded in 1954 that there is no limitedamount of time after which a pseudarthro-sis cannot unite anymore by prolongedimmobilization.[2] Urist further concludedthat a new proliferative process begins af-ter any surgical intervention and after theimplantation of any type of bone graft.[2]

In case of a critical size defect, bone graft-ing surgery is generally performed to obtainbone healing and reduce recovery time. Anestimated 500 000 bone grafting proceduresare performed annually in the United Statesand this number easily doubles on a globalbasis.[3]

Dr. M. K. E. Koolen, Prof. H. H. WeinansOrthopedicsUMC UtrechtHeidelberglaan 100, Utrecht, Utrecht 3584 CX, The NetherlandsDr. P. M. Lopez-Perez, Prof. J. A. Jansen, Prof. S. C. G. Leeuwenburgh,Dr. J. J. J. P. van den BeuckenBiomaterialsRadboudumcPhilips van Leijdenlaan 25, Nijmegen, Gelderland 6525 EX, TheNetherlandsE-mail: [email protected]. D. J. O. UlrichPlastic, Reconstructive and Hand SurgeryRadboudumcGeert Grooteplein Zuid 10, Nijmegen, Gelderland 6525 GA, The Nether-

lands

Macromol. Biosci. 2021, 2100088 2100088 (1 of 9) © 2021 The Authors. Macromolecular Bioscience published by Wiley-VCH GmbH

www.advancedsciencenews.com www.mbs-journal.de

Autologous bone grafting is the golden standard for treat-ing bone defects.[4–7] Drawbacks of autologous bone grafting arethe need for two surgical sites, donor site morbidity, increasedsurgery and recovery time, and shortage in the availability ofdonor tissue.[8] Over the next 50 years, the number of elderlypatients with the need for bone grafting is expected to increase,including their comorbidities. Especially the increase in osteo-porotic patients represents a major challenge for medical care.[9]

Osteoporosis compromises bone strength (both trabecular andcortical), hence increases fracture risks, and further negativelyinfluences the bone healing process.[10–13]

Over the past decades, different types of synthetic bone grafthave emerged, of which particularly calcium phosphate (CaP)based ceramics have shown large potential.[14,15] In view of in-creased demands for synthetic bone grafts with advantageousproperties, we previously designed an injectable, self-healing hy-drogel consisting of bisphosphonated (BP) hyaluronan (HA) andCaP nanoparticles.[16] The major biological advantage of this ma-terial was the observed ingrowth of newly-formed trabecular-likebone interspersed throughout the material after 4 weeks in afemoral condyle defect in rats.[16]

In view of the BP-moiety in the material and the use of BPs as adrug in the treatment of osteoporosis, in the current study boneregeneration was challenged using a rat critical-size segmentaldefect model, in which the effect of bone condition (healthy vsosteoporotic bone) on bone defect healing using HABP-CaP gelwas comparatively evaluated. Additionally, the effect of enrichingthis HABP-CaP gel with bone morphogenetic protein 2 (BMP-2)was explored. We hypothesized that i) an osteoporotic conditionwould negatively affect defect healing, and ii) that enrichment ofthe HABP-CaP gel with BMP-2 would improve defect healing inboth healthy and osteoporotic conditions.

2. Experimental Section

2.1. Materials

The HA with BP and calcium phosphate hydrogel (HABP-CaP)was prepared as previously described,[16] mixing equal volumesof HA- BP with a CaP suspension in a connected syringe.This resulted in a non-covalently cross-linked HABP-CaP hybridnanocomposite gel.

To further enhance bone healing, BMP-2 (InductOs,Medtronic BioPharma B.V., Heerlen, The Netherlands) wasadded to the liquid component of the gel in 3 (HABP-CaP-lowBMP2) and 30 µg (HABP-CaP-highBMP2) per implant froma 1.5 mg mL−1 solution.

In order to maintain the gel inside the defect, polytetrafluo-rethylene (PTFE) tubing of±7 mm in length, 4.2 mm in diameterof 0.25 mm ± 0.07 mm (Polyfluor Plastics BV, Breda, the Nether-lands) was used, as was used in, for example, vascular surgery. Toallow for fluid exchange, the tubing was perforated using a 1 mmbiopsy punch (Kai Europe GmbH, Solingen, Germany) creating4 holes in a row in 4 directions resulting in 16 holes per tube.

2.2. Animals

For this experiment, 56 healthy mature male Wistar rats (age:3 months; weight: ± 250 g) were used. All animals received ei-

ther orchidectomy to induce an osteoporotic bone condition orsham surgery, as described previously.[17] Six weeks after, all an-imals received 1 unilateral segmental defect in the right femur,which was treated with HABP-CaP (8 healthy and 8 osteoporoticanimals), HABP-CaP-lowBMP2 (8 healthy and 8 osteoporotic an-imals), HABP-CaP-highBMP2 (8 healthy and 8 osteoporotic ani-mals), empty PTFE tubing (4 healthy animals), or left empty (4healthy animals).

All experiments were conducted in accordance with institu-tional, national, and international guidelines for animal care andthe Dutch law concerning animal welfare. The studies were re-viewed and approved by the Experimental Animal Committee ofthe Radboud University (RUDEC 2013–206).

2.3. Surgical Procedure to Induce Osteoporotic Condition

Anesthesia was induced and maintained by Isoflurane inhalation(Rhodia Organique Fine Limited) combined with oxygen deliv-ered by mask. Pre-operative pain medication was provided byan injection of Carprofen (5 mg kg−1; Rimadyl, Pfizer AnimalHealth, New York, USA), given 15 min before surgery. After con-firmation of effective anesthesia, the animal was placed on anelectric heating blanket in supine position. The area disinfectionwas performed by application of a povidone-iodine solution. Askin incision was made over the scrotum, followed by blunt dis-section of subcutaneous connective tissue and creating a 5 mmopening in the tip of the scrotal sac. The gonadal tissue was pulledthrough, including the caudal epididymis, caput epididymis, vasdeferens, and spermatic blood vessels. A single ligature was per-formed around the blood vessels and the vas deferens using asurgical polyester suture (Terylene 2.0 undyed, Serag-WiessnerGmbH & Co, Naila, Germany) after testis and epididymis wereremoved. The remaining vas deferens and fat was replaced inthe sac. After hemostasis, the skin was closed with subcuta-neous absorbable sutures (Vicryl 3.0, Ethicon, Amersfoort, TheNetherlands). Finally, the animals were covered with paper orcloth in order to avoid hypothermia. The ORX animals receiveda low calcium diet for six weeks. The sham-operated animals un-derwent the same surgical procedure except for the actual re-moval of the gonads and normal food pellets were provided. Fora minimum duration of 2 days postoperatively, Carprofen wasgiven every 24 h. Additionally, to diminish postoperative pain,Buprenorfine (Temgesic, Reckitt Benckiser Health Care Limited,Schering-Plough, UK; 0.02 mg kg−1) was given every 12 h.

2.4. Surgical Procedure to Create Critical Size Segmental FemurDefect and Placement of Bone Substitute Material

A critical size segmental defect was created unilaterally in the fe-mur of the rats. A single dose of antibiotics (Enrofloxacin, 5 mgkg−1 body weight) was administered 1 h before surgery. To reduceintra-operative and post-operative pain, Temgesic (0.02 mg kg−1)was administered subcutaneously pre-operatively. All surgerieswere performed under general inhalation anesthesia with pre-operative pain medication and sterile conditions as mentionedabove (see Surgical procedure to induce osteoporotic conditions).The right leg was shaved, and the rat was placed in a left lateral



Figure 1. A–H) Photographs during surgery of creating the critical-sized segmental defect in male rats and implantation of the material (in chronologicalorder from left to right and top to bottom). I) A removed bone segment of 6 mm next to a polyurethane tube containing experimental HABP-CaP gel.

position on heating mat. The surgical area was disinfected withpovidone-iodine. Through an anterolateral approach, the vastuslateralis and biceps femoris were separated and elevated whilecautioning to keep the periosteum intact along the surface of thebone (Figure 1A,1B). First, a PEEK plate was inserted and fixedin the anterolateral plane (RatFix, RISystem AG, AOFoundation,Davos, Switzerland; Figure 1C–1E). Then, a segmental defect wascreated with the use of a Gigli saw with a saw guide to ensure the6 mm length of the femoral segment (both saw and guide werepurchased from RISystem) under constant cooling with saline(Figure 1F,1G). After removing the segment, the defect was leftempty or placed a tube of PTFE in the defect, filled with the ex-perimental material (HABP-CaP, HABP-CaP-lowBMP2, or HABP-CaP-highBMP2) or with an empty tube (Figure 1H,1I). Subse-quently, the muscles and fascia were closed with absorbable su-tures (Vicryl 4.0, Ethicon Products, Amersfoort, the Netherlands)and the skin with both absorbable sutures and agraves (Vicryl4.0 and agraves, Ethicon Products, Amersfoort, the Netherlands).Postoperatively, Temgesic (0.02 mg kg−1) and Carprofen (5.0 mgkg−1) were administered for 2 days post-operatively every 12 h.

2.5. Animal Euthanasia and Specimen Retrieval

Twelve weeks post-operatively, specimens were retrieved for exvivo X-ray imaging and micro-CT scanning, as well as histologicalassessment. All rats were euthanized by CO2-suffocation in ac-cordance with ISO standards and protocols of the Radboud Uni-versity Nijmegen Medical Center, The Netherlands. The femursincluding the implanted materials were harvested, stripped fromsoft tissues, and immediately fixed in 10% formalin solution. Af-ter fixation, specimens were stored in 70% ethanol.

2.6. Validation of Osteoporotic Condition by ELISA for SerumTRAP Analysis

During specimen retrieval, blood was taken from several rats ofboth SHAM and ORX (8 and 10 rats respectively) to validate os-teoporotic condition. This was evaluated by a serum analysis forTRAP enzyme activity performed with the RatTRAP Assay kit(Immunodiagnostic Systems GmbH, Frankfurt am Main).

2.7. X-Ray Imaging

After specimen retrieval and formalin fixation, X-ray images weremade of the femora using dental X-ray equipment and imagingplate (Planmeca ProX 2016 & Soredex DIGORA Optime 2015;Dental Union Plandent, Nieuwegein, the Netherlands).

2.8. Micro-CT Scanning

Micro-computed tomography (micro-CT) scanning of the femorawas performed to assess bone formation at the endpoint of12 weeks with a micro-CT imaging system (Quantum FX;PerkinElmer, Waltham, MA, USA). Three minutes of scan timewas required per bone for an isotropic voxel size of 42 µm (volt-age 90 kV, current 180 mA, field of view = 21 mm). To ana-lyze the results, all datasets were aligned. A region of interest(ROI) of 4.6×5×6 mm (bone volume inside the defect; BVi) and of6×8.4×8.4 mm (total bone volume; TBV) was selected. The ROIwas segmented with a global threshold. Bone volume was mea-sured in mm3 using image processing software (Image-J; Java,Redwood Shores, CA, USA).



Figure 2. Example of a histological image with all three analyzed regions (colors and lettering are only drawn to indicate the areas in this image). ROIIN– Region of Interest within defect area; ROIOUT – Region of interest outside the defect area.

For 3D imaging, one sample of each group was selected andscanned with a desktop X-ray micro-CT system (Skyscan-1072,TomoNT version 3N.5, SkyscanVR, Kontich, Belgium). The spec-imens were placed onto the sample holder with the long axis ofthe femur perpendicular to the scanning beam. The 3D recon-struction and imaging processing were performed using NReconV1.4 and Dataviewer V1.4.3.0 (SkyscanVR, Kontich, Belgium), re-spectively.

2.9. Histological Processing

After performing the X-ray imaging and micro-CT scanning, theobtained specimens were dehydrated in a graded series of ethanol(70–100%), and embedded in polymethylmethacrylate (PMMA).After polymerization, thin sections of the defect of ≈10 µm wereprepared in a plane parallel to the axis of the femur using a dia-mond blade microtome (Leica Microsystems SP 1600, Nussloch,Germany) as previously described.[17,18] At least three sections ofeach specimen were made and stained with methylene blue andbasic fuchsine.

2.10. Descriptive Histology and Histomorphometrical Analysis

After an implantation period of 12 weeks, PMMA histologicalsections (3 sections per specimen) were examined by light mi-croscopy (Leica Microsystems AG, Wetzlar, Germany) and digi-talized with an automatic scanning microscope (Panoramic250,3DHISTECH Ltd., Budapest, Hungary). Quantitative assessmentof the PMMA sections was performed using ImageJ computer-based image analysis software (Java ImageJ 1.47, Image process-ing and analysis in Java).[19] From digitalized images of the sec-tions (magnification: 5x), a rectangular ROI was created with di-mensions equal to the created defects (ROIIN). Second, a rectan-gular ROI above and below the implant was positioned as ROIOUT(Figure 2). Within all ROIs, the amount of newly formed boneand material remnants was measured using color discriminationand morphology for verification. Evaluation of the gel capacity toovercome the critical size defect included whether or not the de-fect was bridged. This was defined by bone formation from bothends of the defect that was connected in the middle, with visi-ble cortical bone connection, or by relatively thick trabecular-likebone dispersed throughout the whole defect.

2.11. Statistical Analysis

Data are presented as mean with SD for histomorphometric dataand CT-scan analysis. Statistical analysis of quantitative data wasperformed using one-way analysis of variance (ANOVA) with aTukey post-hoc test for comparison between the three differentgroups. Student’s t-tests were used for comparison between bothconditions for selected materials. GraphPad Software (PRISM,La Jolla CA, USA) was used to carry out the statistical analysis.Values of p<0.05 were considered statistically significant.

3. Results

3.1. Animals

All of the 56 animals recovered well after orchidectomy or shamsurgery. There were no wound problems and all but one ani-mal remained in good condition. Two days after surgical segmen-tal defect creation, one animal was found dead (sham-operated,HABP-CaP); obduction did not reveal issues in the operated leg,however, ascites and pleural effusion were observed. All 55 re-maining animals were healthy and recovered well, presentingnormal activity without signs of infection or other wound heal-ing problems.

3.2. Validation of Osteoporotic Condition by Serum TRAPAnalysis

The induced osteoporotic bone condition after orchidectomy re-sulted in significantly decreased serum TRAP enzyme activity(p<0.001) compared to healthy animals (Figure 3), with activityvalues of 1.2 ±0.5 and 2.5 ±0.6 U L−1 h−1, respectively.

3.3. X-Ray Imaging

Representative X-ray images are depicted in Figure 4. These im-ages showed substantial healing for HABP-CaP-lowBMP2 andHABP-CaP-highBMP2, but low radiopacity in the defect regionfor HABP-CaP and empty tube controls. In one of the four femurswith no tube, the screws were broken and in the second animalof this group the screws were loosened resulting in a 45° angle ofthe distal femur.



Figure 3. Serum TRAP enzyme activity (mean ±SD) comparing ORX andSHAM operated animals, ***) significant difference between ORX andSHAM (p<0.001).

3.4. Micro-CT Analysis

Representative micro-CT images of the segmental defects are de-picted in Figure 5A. Data of the micro-CT analysis are shownin Figure 5B. Overall bone formation in ROIIN demonstratedthat defects could not be bridged by the gel alone. Only twoout of eight defects in the ORX animals showed bridging. ForSHAM animals this material had the least amount of boneformation compared to other materials and none bridged thedefect.

Further specification within the different experimental groupsshowed bone formation to be highest for HABP-CaP-lowBMP2in both SHAM and ORX animals (38.5 and 37.0% respec-tively). Bone formation was significantly highest for HABP-CaP-lowBMP2 compared to all other groups for both conditions(p<0.001, p<0.001, and p<0.01 compared to, respectively, controlgroup, HABP-CaP and HABP-CaP-highBMP2). Bone formationwas not only seen within the original defect area, but also outsideof this area. Particularly for those groups with added BMP-2, sub-stantial bone tissue was formed outside the original defect area.Outside the ROI there was significantly less bone in SHAM ani-mals for HABP-CaP compared to HABP-CaP-lowBMP2 and HABP-CaP-highBMP2 (both p<0.001). In ORX animals the bone inROIOUT was significantly higher for HABP-CaP-lowBMP2 com-pared to HABP-CaP and HABP-CaP-highBMP2 (both p<0.001).Between conditions, there was a significant difference in the fa-vor of SHAM for HABP-CaP-highBMP2 both inside and outsidethe ROI (ROIIN p<0.05 and ROIOUT p<0.01).

3.5. Descriptive Histology

Representative histology images of the segmental defects are de-picted in Figure 6A. Histological images of both control groups(with no tube or empty tube) confirmed that the used segmentaldefect can be classified as a critical size defect.

In all three of the experimental groups (HABP-CaP, HABP-CaP-lowBMP2, and HABP-CaP-highBMP2) there was apparent bone

formation. There was a difference between bone formed withor without BMP-2. When using the gel only (HABP-CaP) boneregeneration was seen at both ends of the defect as a corticaledge growing towards the center. However, none of the sam-ples with HABP-CaP had fully bridged bone defects (Table 1).For the HABP-CaP-lowBMP2 samples, all had a complete fillingof bone within the defect areas for both ORX and SHAM. Thenewly formed bone in these samples was spongious-like and dis-persed throughout the ROIIN. In the HABP-CaP-highBMP2 sam-ples, bone defects were filled with bone but the bone formationwas cyst-like with a very thin shell. Also around the tube andboth sides of the fixation plate, an even thinner layer of bone wasformed.

Magnifications of the samples showed almost all material hasdegraded. Only in a few samples material remnants were de-tected, as small uncolored particles with irregularly shaped edges,entrapped within the newly formed bone (Figure 6A, black ar-rows indicate material remnants). Higher magnification showedthe cysts formed with a high dosage of BMP-2 were filled with adi-pose cells (Figure 6A). Defects bridged by bone with a low dose ofBMP-2 also contained more adipose cells within the defect areacompared to the defects without BMP-2.

3.6. Histomorphometry Analysis

Histomorphometrical data are depicted in Figure 6B. For SHAManimals this material had the least amount of bone forma-tion compared to other materials and none bridged the defect.Histomorphometry analysis between the different experimen-tal groups bone formation showed to be highest for HABP-CaP-lowBMP2 in both SHAM and ORX animals (34.2 and 38.5%respectively). As the already mentioned bone formation wasnot only seen within the original defect area, but also outsideof this area. Particularly for those groups with added BMP-2,substantial bone tissue was formed outside the original defectarea.

Statistical analysis for histomorphometric bone formation wassignificantly higher for HABP-CaP-lowBMP2 compared to othermaterial groups (p<0.05 compared to HABP-CaP in both condi-tions and p<0.01 compared to HABP-CaP-highBMP2 for SHAManimals). In ORX, HABP-CaP-highBMP2 showed significantlyless bone compared to HABP-CaP (p<0.01). Bone formation out-side the ROI was significantly lower for HABP-CaP compared toHABP-CaP-lowBMP2 and HABP-CaP-highBMP2 in both SHAMand ORX animals (all p<0.001).

Between the two bone conditions, there was more bone in fa-vor of SHAM for HABP-CaP-highBMP2 within the defect areacompared to osteoporotic bone conditions (p<0.05).

4. Discussion

In the current study, bone regeneration was examined using arat critical-sized femoral defect model, in which the effects ofbone condition (healthy vs osteoporotic bone) and BMP-2 incor-poration into the HABP-CaP hydrogel were comparatively eval-uated. We hypothesized that i) an osteoporotic condition wouldnegatively affect defect healing, and ii) that enrichment of the



Figure 4. Representative X-ray images of all material groups in SHAM and ORX animals, and of the SHAM animals with an empty tube control.

HABP-CaP gel with BMP-2 would improve defect healing in bothhealthy and osteoporotic conditions.

The osteoporotic model was validated by serum TRAP anal-ysis. In a previous study, we used in vivo micro-CT to analyzetrabecular bone morphometry after orchidectomy and found lesstrabecular bone compared to healthy rats.[17] Also, the segmentalfemoral bone defect model of 6 mm proved to be of critical sizeas none of the control groups showed bridging. This corroborateswith other studies using the RatFix system.[20–24]

For this study, the hydrogel was applied in a PTFE tube tocontain it within the defect area. The strong bond of the carbon-fluorine gives PTFE the durable properties, inertness, and resis-tance to chemical, biological and physical degradation.[25] PTFEas s biomaterial has been widely used in clinics for severalyears,[26–35] and is generally considered to be a well-tolerated ma-terial and useful when autologous tissue is not available. Onemajor drawback is the risk of infection. In our study, no in-fection was observed in any of the animals. To study the ef-fect of the tube on bone formation, two control groups were in-cluded, one with and one without the tube. The distal femur oftwo out of four defects without tube was found to be loosened.The other two femurs had intact plates and screws and showedsome bone formation on the plate side of the defect. The fourfemurs with an empty tube showed more equally distributed

bone formation starting from each side of the defect, suggest-ing a scaffolding effect of the tube. However, none of the controlgroups, with or without tube, showed complete bridging of thedefect.

In view of increased demands for synthetic bone grafts with ad-vantageous properties, we previously designed an injectable, hy-drogel consisting of BP HA and CaP nanoparticles.[16] The majorbiological advantage of this material at the time was the ingrowthof newly-formed trabecular-like bone interspersed throughoutthe material after 4 weeks in a femoral condyle defect in rats.[16]

Our results show that the hydrogel was almost all resorbed by 12weeks and did allow for bone ingrowth. However, the results ofthis study demonstrate that this material alone is insufficient toregenerate 6 mm rat critical-size segmental bone defects.

Additionally, the efficacy of BMP-2 loading in the hydrogelwas examined in this study and found to be effective anddose-dependent. Qualitative and quantitative analysis for boneformation by radiography, micro-CT, and histomorphometryshowed optimal healing for HABP-CaP-lowBMP2 in both healthyand osteoporotic bone conditions. Qualitative data showed thetype of bone formation was different depending on the materialused. When the tube was empty or filled with gel only morecortical bone was formed from the edges of the bone inwards.Whereas when BMP-2 was used bone formation was more



Figure 5. A) Representative micro-CT images (reproduced with DataViewer) of all 3 experimental groups of SHAM and ORX animals and of the SHAManimals with an empty tube control. Green boxed images are coronal cross-sections with the plate to the right side of the image, being the anterolateralside of the femur. The blue boxed (middle) images are of sagittal planes through the femur. The red boxes are transverse planes with the plate to the rightside of the image. The colors are a pre-selected setting in DataViewer (Color 2), which represents density. B) Quantitative analysis based on micro-CTimages of bone formation (percentage % bone area in mean data ±SD) within and outside the ROI. With the larger graph comparing between groupsand the smaller between ROIIN and ROIOUT and per material group; showing significant differences of *p<0.05, **p<0.01, and ***p<0.001.

spongious-like, especially for low dose BMP-2. A more cyst-likeformation was seen when a high dosage of BMP-2 was used.The quantitative data showed the most bone volume for the hy-drogel with a low dosage of BMP-2. However, the spongious-likebone was not tested for strength and more research should beperformed to further inform on the quality of the two differentbone formations with or without BMP-2.

Since the discovery of BMP-2 by Urist in 1965, many studiesshow the similar promising effect of BMP-2. In 2007 it was firstapproved by the FDA (Food and Drug Administration) for clinicalpurpose reporting “near perfect” safety.[36] In recent years, how-ever, several (pre-)clinical studies with the use of BMP-2 for boneregeneration do show several adverse effects.[36,37] These includegeneral inflammatory and wound complications, ectopic bone,osteoclast activation, osteolysis, and graft subsidence.[36] Morespecifically for spinal fusion, complications can be radiculopathy,urogenital events (retrograde ejaculation and bladder retention),and most concerning cervical spine swelling to the extent of beingpotentially life-threatening.[36,37] Remarkably, we observed signif-icantly less bone and lower quality of formed bone for HABP-CaP-highBMP2 compared to HABP-CaP-lowBMP2. Studies analyzingthe signaling of BMP-2 have found that BMP-2 induces the ex-pression of numerous inflammatory cytokines and chemokines,including TNF-𝛼 and interleukins.[38–40] Yamazaki et al. foundthat TNF-𝛼 represses BMP-2, which is increased by the use ofBMP-2, therefore not only resulting in inflammation but also ininhibition of osteoblastogenesis.[41] Liao et al. found that BMP-2not only affects MSCs by inducing both osteogenic and chondro-

genic differentiation through RUNX2 pathway, but also can pro-mote adipogenic differentiation.[36,42] The latter could be relatedto bone cyst formation with decreased overall bone quality.[36]

This was also seen in the study by Zara et al., who used the samesegmental defect as was used in this study, filled with collagensponges infused with several dosages of BMP-2.[43] Their studyshowed that high dosages of BMP-2 induce inflammation result-ing in structurally abnormal bone in vivo, with the formation ofcyst-like bony shells filled with adipose tissue.[43] This corrobo-rates with our results of high BMP-2 dosage showing bone for-mation outside the defect in a similar matter. Also, the increasedamount of adipose cells within the defect area for both high andlow-dose BMP-2 corresponds with these studies.[36,42,43] Zara et al.postulated that a high dosage of BMP-2 deregulates Wnt signal-ing and activate PPAR𝛾 leading to abnormal structure decreasingbone quality and probably influencing the mechanical properties.

As osteoporosis is increasingly becoming a major health prob-lem, we aimed to examine the bone regenerating capacity of abone substitute material in both healthy and osteoporotic con-ditions. However, very few differences were observed betweenthe healthy and osteoporotic conditions. Only for the HABP-CaP-highBMP2, superior regeneration in healthy conditions wasfound. This observation corroborates with one of our other stud-ies, in which CPC/PLGA and Bio-Oss were compared in healthyand osteoporotic conditions in a rat femoral bone defect model,and also no significant differences between the two conditionsin terms of bone regeneration were found.[44] As already men-tioned, a high dosage BMP-2 can have adverse effects, such as



Figure 6. A) Representative histological images and magnifications (orange = 20x and red 40x) of all 3 material groups of SHAM and ORX animals andof the SHAM animals with an empty tube. Green asterisks are placed below pieces of the tube. Black arrows indicate material remnants. Orange A’s areplaced in adipose tissue. B) Quantitative analysis based on histomorphometry of bone formation (percentage % bone area in mean data ±SD) withinand outside the ROI. With the larger graph comparing between groups and the smaller between ROIIN and ROIOUT and per material group; analysisshowing significant differences of *p<0.05, **p<0.01, and ***p<0.001.

Table 1. Segmental defect bridging.

SHAM* ORX

ROIIN ROIOUT ROIIN ROIOUT

HABP-CaP 1/7 0/7 2/8 0/8

HABP-CaP-lowBMP2 8/8 8/8 8/8 8/8

HABP-CaP-highBMP2 4/8 8/8 3/8 8/8

Empty tube 0/4 0/4 NA NA

∗Bridging was assessed histologically; numbers indicate the number of defects show-ing complete segmental defect bridging out of the total number of defects per exper-imental group.

general inflammatory and wound complications, ectopic bone,osteoclast activation, osteolysis, and graft subsidence. In thisstudy, these adverse effects of a high dosage of BMP-2 on bone re-generation seem less pronounced in osteoporotic conditions. Anexplanation for this could be found in the serum TRAP analysis.Our results show significantly lower amounts of serum TRAPin osteoporotic conditions, which is in concurrence with previ-ous studies.[45,46] We concluded that the lower amount of TRAPindicates a decreased absolute number of active osteoclast dueto a decreased amount of bone.[45,46] This absolute decreasedamount of active osteoclasts combined with less bone volumecan explain why BMP-2 has less effect in osteoporotic condi-tions. Furthermore, the high dosage of BMP-2 could result inmore bone resorption by further increasing the number of ac-

tive osteoclasts.[36,43] To confirm this hypothesis more researchwith gene expression analysis in osteoporotic conditions shouldbe performed.

5. Conclusion

The current study shows that critical size, segmental femoralbone defects cannot be healed with HABP-CaP gel alone. Loadingof the HABP-CaP gel with low dose BMP-2 significantly improvedbone formation and resulted in defect bridging in 100% of thedefects. Alternatively, high dose BMP-2 loading of the HABP-CaPgel did not improve bone formation within the defect area, but ledto excessive bone outside the defect area by cyst-like formation.Surprisingly, bone defect healing was not negatively affectedby osteoporotic bone conditions, except when using high doseBMP-2.

AcknowledgementsThe authors would like to acknowledge N. van Dijk for her assistance withthe histological sectioning and Dr. V. Cuijpers for his assistance on thehistological imaging and histomorphometrical analysis. The authors fur-ther acknowledge the department of pathology, especially M. Hermsen,for assistance with automated scanning of histological sections.

Conflict of InterestThe authors declare no conflict of interest.



Data Availability StatementResearch data are not shared.

Keywordsbone healing, bone morphogenetic protein 2, critical size defect, hydrogel,osteoporosis

Received: March 15, 2021Revised: May 14, 2021

Published online:

[1] P. P. Spicer, J. D. Kretlow, S. Young, J. A. Jansen, F. K. Kasper, A. G.Mikos, Nat. Protoc. 2012, 7, 1918.

[2] M. R. Urist, R. Mazet Jr., C. F. McLean, J. Bone Jt. Surg. 1954, 36, 931.[3] A. S. Greenwald, S. D. Boden, V. M. Goldberg, Y. Khan, T. Laurencin,

R. N. Rosier, J. Bone Jt. Surg. 2001, 83, 98.[4] A. Berggren, A. J. Weiland, H. Dorfman, Plast. Reconstr. Surg. 1982,

69, 19.[5] A. J. Weiland, T. W. Phillips, M. A. Randolph, Plast. Reconstr. Surg.

1984, 74, 368.[6] K. Muramatsu, A. T. Bishop, J. Orthop. Res. 2002, 20, 772.[7] M. A. Arata, M. B. Wood, W. P. Cooney 3rd., J. Reconstr. Microsurg.

1984, 1, 11.[8] R. Dimitriou, G. I. Mataliotakis, A. G. Angoules, N. K. Kanakaris, P. V.

Giannoudis, Injury 2011, 42, S3.[9] World Health Organ. Tech. Rep. Ser. 2003, 921, 1.

[10] F. Sanfilippo, A. E. Bianchi, Int. J. Periodontics Restorative Dent. 2003,23, 447.

[11] M. Fini, G. Giavaresi, P. Torricelli, V. Borsari, R. Giardino, A. Nicolini,A. Carpi, Biomed. Pharmacother. 2004, 58, 487.

[12] J. Finigan, Nurs. Stand. 2003, 17, 104.[13] S. M. Ott, Am. J. Nephrol. 2018, 47, 373.[14] M. Bongio, J. J. J. P. van den Beucken, S. C. G. Leeuwenburgh, J. A.

Jansen, J. Mater. Chem. 2010, 20, 8747.[15] L. L. Hench, J. Mater. Sci.: Mater. Med. 2006, 17, 967.[16] M. R. Nejadnik, X. Yang, M. Bongio, H. S. Alghamdi, J. J. J. P. van den

Beucken, M. C. Huysmans, J. A. Jansen, J. Hilborn, D. Ossipov, S. C.G. Leeuwenburgh, Biomaterials 2014, 35, 6918.

[17] H. S. Alghamdi, J. J. van den Beucken, J. A. Jansen, Tissue Eng., PartC 2014, 20, 493.

[18] H. B. van der Lubbe, C. P. Klein, K. de Groot, Stain Technol. 1988, 63,171.

[19] M. A. Lopez-Heredia, M. Bongio, V. M. Cuijpers, N. W. M. van Dijk, J.J. J. P. van den Beucken, J. G. C. Wolke, J. A. Jansen, Tissue Eng., PartC 2012, 18, 369.

[20] L. Poser, R. Matthys, P. Schawalder, S. Pearce, M. Alini, S. Zeiter,Biomed Res. Int. 2014, 2014, 348635.

[21] A. Chatterjea, J. van der Stok, C. B. Danoux, H. Yuan, P. Habibovic, C.van Blitterswijk, H. Weinans, J. de Boer, J. Biomed. Mater. Res., Part A2014, 102, 1399.

[22] J. van der Stok, M. K. Koolen, H. Jahr, N. Kops, J. H. Waarsing, H.Weinans, O. P. van der Jagt, Eur. Cells Mater. 2014, 27, 137.

[23] J. Van der Stok, O. P. Van der Jagt, S. Amin Yavari, M. F. de Haas, J.H. Waarsing, H. Jahr, E. M. van Lieshout, P. Patka, J. A. Verhaar, A. A.Zadpoor, H. Weinans, J. Orthop. Res. 2013, 31, 792.

[24] K. Sato, Y. Watanabe, N. Harada, S. Abe, T. Matsushita, K. Yamanaka,T. Kaneko, Y. Sakai, Tissue Eng., Part C 2014, 20, 1037.

[25] B. J. Henry, J. P. Carlin, J. A. Hammerschmidt, R. C. Buck, L. W. Bux-ton, H. Fiedler, J. Seed, O. Hernandez, Integr. Environ. Assess. Manage.2018, 14, 316.

[26] J. M. Bellon, J. Bujan, L. A. Contreras, A. Hernando, F. Jurado, J.Biomed. Mater. Res. 1996, 31, 1.

[27] F. A. Grimme, M. M. Reijnen, K. Pfister, J. M. Martens, P. Kasprzak,Eur. J. Vasc. Endovasc. Surg. 2014, 48, 545.

[28] L. D. Baker Jr., J. M. Johnson, D. Goldfarb, Trans. Am. Soc. Artif. Intern.Organs 1976, 22, 382.

[29] D. V. Feliciano, K. L. Mattox, J. M. Graham, C. G. Bitondo, J. TraumaAcute Care Surg. 1985, 25, 71.

[30] J. M. Bellon, L. A. Contreras, J. Bujan, A. Carrera-San Martin, J. Bio-mater. Appl. 1997, 12, 121.

[31] J. F. Gillion, G. F. Begin, C. Marecos, G. Fourtanier, Am. J. Surg. 1997,174, 16.

[32] A. Lukasiewicz, T. Drewa, Adv. Clin. Exp. Med. 2014, 23, 135.[33] P. D. Fotek, R. F. Neiva, H. L. Wang, J. Periodontol. 2009, 80, 776.[34] T. Jacob, O. J. LaCour, C. F. Burgoyne, P. K. LaFleur, E. Duzman, J.

Glaucoma 2001, 10, 115.[35] A. Sevy, M. Arriaga, Otolaryngol. Clin. North Am. 2018, 51, 393.[36] A. W. James, G. LaChaud, J. Shen, G. Asatrian, V. Nguyen, X. Zhang,

K. Ting, C. Soo, Tissue Eng., Part B 2016, 22, 284.[37] N. E. Epstein, Surg. Neurol. Int. 2013, 4, S343.[38] J. Liu, L. Chen, Y. Zhou, X. Liu, K. Tang, PLoS One 2014, 9, e85469.[39] M. J. Jeon, J. A. Kim, S. H. Kwon, S. W. Kim, K. Soo, K. S. Park, S. W.

Park, S. Y. Kim, C. S. Shin, J. Biol. Chem. 2003, 278, 23270.[40] T. Suda, N. Takahashi, N. Udagawa, E. Jimi, M. T. Gillespie, T. J. Mar-

tin, Endocr. Rev. 1999, 20, 345.[41] M. Yamazaki, H. Fukushima, M. Shin, T. Katagiri, T. Doi, T. Takahashi,

E. Jimi, J. Biol. Chem. 2009, 284, 35987.[42] X. Liao, L. Wu, M. Fu, D. He, Y. Gu, W. Chen, W. Yin, Zhongguo Xiu Fu

Chong Jian Wai Ke Za Zhi 2012, 26, 743.[43] J. N. Zara, R. K. Siu, X. Zhang, J. Shen, R. Ngo, M. Lee, M. Li, W. Li,

M. Chiang, J. Chung, J. Kwak, B. M. Wu, K. Ting, C. Soo, Tissue Eng.,Part A 2011, 17, 1389.

[44] C. I. A. van Houdt, D. J. O. Ulrich, J. A. Jansen, J. van den Beucken, J.Biomed. Mater. Res., Part B 2018, 106, 131.

[45] C. I. van Houdt, C. R. Tim, M. C. Crovace, E. D. Zanotto, O. Peitl, D.J. O. Ulrich, J. A. Jansen, N. A. Parizotto, A. C. Reno, J. J. J. P. van derBeucken, Biomed. Mater. 2015, 10, 035003.

[46] J. P. Rissanen, M. I. Suominen, Z. Peng, J. M. Halleen, Calcif. TissueInt. 2008, 82, 108.


Regenerating Critical Size Rat Segmental Bone Defects with a ...

Documents

Transcript of Regenerating Critical Size Rat Segmental Bone Defects with a ...