All the animals used were from the Experimental Animal Center of Hebei Medical University. All procedures involving animals are in compliance with the "Animal Care and Use" guidelines and approved by the Animal Care and Experimental Committee of the Third Affiliated Hospital of Hebei Medical University. ID of the ethics approval is Z2018-005-1.
Prosthesis design
The shape and size of the upper tibia, including the medullary cavity, were observed and measured on the transverse, coronal and sagittal planes by dissecting the upper tibia of the rats. Then, a tibial knee replacement (hemiarthroplasty) was designed to replace the tibial articular surface. Based on the implant shape showed in human knee replacement, a press-fit titanium tibial implant comprised of an articular baseplate and intramedullary stem. The baseplate surface was oval in shape with a medial–lateral width of 7.0mm,anterior–posterior depth of 5.0mm and thickness of 1.5mm. The area of the baseplate was similar to that of the tibial plateau of rat, which formed joint with femoral condyle. The stem was approximately columnar with 5.0mm in length and 1.5mm in diameter . So as to make the prosthesis stable, one side of the stem was attached a small wing, which could not only prevent the prosthesis from rotating in the bone, but also better adapt to the shape of the medullary cavity of the upper tibia of rats. The 3D image of prosthesis was simulated by computer-aided design, and then the prosthesis was made by 3D powder printer (Arcam Q10, Sweden) with titanium powder. The powder was spherical in shape with a median size of approximately 40 µm (Fig. 1). The articular surface was polished using 1,200 grit silicon carbide grinding paper, and the stem surface was coated with close-packed 40-µm-diameter titanium spheres which made the surface rough and beneficial to bone ongrowth. Implants were degreased and cleaned by using Alconox detergent (Seebio, Xi'an, China) and sonication in deionized water. Before implantation, the grafts were sterilized by using a standard autoclave (121℃; humidity 100%; 30min).
Establishment of osteoporosis rat model
90 two-month-old female Sprague Dawley (SD) rats were obtained from Laboratory Animal Center of Hebei Medical University (Shijiazhuang, China) and housed in the Research Animal Facility. The facility is controlled by temperature, ventilation, and illumination. Rats were fed standard hard chow and water ad libitum. When the rats reached three months old, five rats were randomly designated as members of the sham group. The remaining rats were subjected to ovariectomies (OVX) to simulate osteoporosis models. Preoperative subcutaneous injections of gentamicin (5mg/kg) were administered to protect from infection. Anesthesia was successfully achieved by intraperitoneal injection of 1.5% phenobarbital sodium 40mg/kg (Schering-Plough, Belgium). One small incision (1.5cm) was made through the skin and the muscle wall on each side of the backbone in the dorsal aspect, and the bilateral ovaries were exposed and removed from OVX animals. In the sham animals, only a small amount of fat around the ovaries was removed. The wound was sutured in layers. After 3 months, five sham rats and five OVX rats were sacrificed, and the BMD of the fifth lumbar of each rat was measured by using dual-energy X-ray absorptiometry (DEXA; Lunar Corporation, USA). A reduction in bone mineral density of more than 20% was considered as a successful model of osteoporosis[27][28].
Prosthesis implantation surgery
After identifying osteoporosis in rats, a unilateral (left) tibial knee replacement was performed on each animal. The skin over the left knee was sterilized twice with alcohol. The fur was removed with a hair razor. Preoperative subcutaneous injections of gentamicin (5mg/kg), buprenorphine (0.03mg/kg) were administered. After successful anesthesia according to the methods mentioned above, the operation area was covered with sterile sheet leaving only the knee joint exposed. The knee was opened with a parapatellar medial incision and the tendon with the patella was dislocated laterally. After the joint cavity was exposed completely, the anterior cruciate ligament and menisci were resected, and the posterior cruciate ligament was reserved. Bone scissors were used to remove articular cartilage and proximal epiphysis to accommodate the implant. The thickness of the cut bone is about 1.5mm, which is equivalent to the prosthesis tray.A hand drill was used to drill a 1-mm wide and 5-mm deep hole in the tibia to fit the stem of the prosthesis. After washing bone surface with aseptic water, the prosthesis was implanted in a press-fit manner (cementless). Ethibond 4-0 suture was used to close the joint capsule and Vicryl 5-0 suture was used to close the skin (Fig. 2). After the operation, the knee joint is stable and can be extended and flexed as normal.
Delivery of DFO
Subsequently, the rats were randomly divided into DFO group (n = 40 ) and control group (n = 40 ). Starting on postoperative day 4, all rats in DFO group were injected with 200uL of 200umol DFO solution (Sigma, St. Louis, MO, USA) in normal saline directly into the knee joint cavities. The rats were mildly sedated before injection, and a 25-gauge needle was used to deliver the solution carefully (Fig. 2). The rats in control group were injected with 200uL of normal saline in an identical manner. The procedure was repeated every 48h for a total of six doses. The DFO dose and injection schedule were based on previous review of the literature[16][18][20].
Postoperative course
After the operation, rats were allowed free motion and unrestricted access to food and water. All rats were monitored daily for signs of inflammation, lethargy, and for general health by an experienced animal care technician. Two hours after the surgery, injections of buprenorphine at 0.01mg/kg (Schering-Plough, Belgium) were administrated and continued at 0.005 mg/kg for three days. After that, an X-ray examination was performed to confirm that the implant was in a satisfactory predetermined position. We found no knee dislocation or subluxation, which proved that the joint was stable. We also found there was no periprosthetic gap visible to the naked eye on radiographs, which meant the prosthesis and bone were in close contact (Fig. 2).
After 2 weeks of implantation, euthanasia was performed by overdosing 8 rats of each group with sodium thiopental at 90 mg/kg (Schering-Plough, Belgium) after an overdose of anesthaesia. Proximal tibia was explanted, the bones were stripped of surrounding soft tissue and the prosthesis was removed, then the bones were fixed in cold 4% formaldehyde for 1 day for immunohistochemical examination. Another 16 rats (8 rats in each group) were euthanized in the same way. The bone tissue was cut into about 0.5× 0.5 ×0.5 cm in size. The whole process was completed within 30 minutes. Afterwards the samples were put into the sterile cryopreservation tube, pre-cooled in liquid nitrogen for 5 minutes, and finally stored in liquid nitrogen tank(-80℃) for PCR. Three months after knee surgery, all rats from each group were sacrificed in the same way. All tibias containing the stabilized prosthesis were stripped of surrounding soft tissue, and subjected to micro-CT analysis, histopathology and biomechanical testing as detailed below.
Quantitative real-time reverse transcription PCR
The expressions of vascular endothelial growth factor (VEGF) and vascular endothelial cell marker CD31 were detected by RT-PCR. Total RNA was extracted by the TRIZOL protocol (Sigma-Aldrich, MO), then was concentrated and purified according to the instructions. RNA integrity was measured by using agarose gel electrophoresis (AGE). A Prime-Script First-Strand cDNA Synthesis Kit (Takara, Dalian, China) was used for reverse transcription. An SYBR® premix Ex TaqTM kit (Takara, Dalian, China) was used for PCR amplification. Real-time quantitative PCR was performed at 57°C for 30 cycles in the Opticon Continuous Fluorescent Detector by using IQTM SYBR green supermix (BioRad, USA). Each group of samples was repeatedly measured three times. 2-△△CT was used to analyze the relative expression of each group of genes.Rat 18S ribosomal RNA was used as internal reference. The following primers were used (Table 1).
Table 1. Primers for Real-Time PCR Analysis
Gene
|
Forward Premier
|
Reverse Premier
|
VEGF
|
AGAAAGCCCATGAAGTGGTGA
|
GCTGGCTTTGGTGAGGTTTG
|
CD31
|
TTGTGACCAGTCTCCGAAGC
|
TGGCTGTTGGTTTCCACACT
|
Rat-18sRNA
|
GTAACCCGTTGAACCCCATT
|
CCATCCAATCGGTAGTAGCG
|
Immunohistochemistry
At the same time, immunohistochemistry was also performed to assess the expression of HIF-1a,VEGF, CD31, as well as to assess the angiogenic potential of periprosthetic bone tissue. Briefly, the upper tibial bone tissue from which the prosthesis was removed was fixed, decalcified, embedded in paraffin, and sliced to a thickness of 5mm. The sections were deparaffinized and washed with PBS. The slice was immersed in 3% hydrogen peroxide for 10min to block endogenous peroxidase activity and then were rinsed several times in PBS. After being blocked with 1% BSA solution at room temperature for 1 hour, sections were treated with a primary antibody overnight at 4℃ and incubated with the biotinylated secondary antibody for 30min. Negative control sections were incubated with PBS solution. The following specific antibodies were used: HIF-1a, HIF-2a (Upstate Biotechnology, Lake Placid, New York, USA), VEGF (Upstate Biotechnology, Lake Placid, New York, USA), CD31 (Santa Cruz biotechnology, Dallas, Texas, USA). Next, the slides were visualized using a streptavidin-biotin staining technique that involves peroxidase labeling. The nuclei were counterstained with hematoxylin. The sections were observed by optical microscope (Olympus BH-2, Japan) and photographed by the high performance computer-aided image analysis system (Nikon H600l, Japan). The yellowcolored cells were considered antigen-positive. Then, the image analysis software (Image Pro Plus 6.0 software) was used for quantitative analysis of the positive staining images.Ten randomly fields in bone marrowcavities in each section were selected to quantify the positive staining with the accumulated optical density value. The ratio of the accumulated optical density value to the corresponding observation area was used as the final quantitative parameter and recorded as the mean optical density value (MOD). Eight sections for each protein from each group were analyzed. After immuohistochemical staining for CD31, the formation of microvessel could be observed by light microscopy. The observer was blind regarding the identity of two groups.
Micro-CT scan analysis
8 tibias were harvested and the attached soft tissues were excluded, and then the gross specimens were observed. After the prosthesis was removed gently, the bone around the prosthesis was examined by high-resolution micro CT (skyscan1176; skyscan, knotich, Belgium). The setting parameters of μCT were as follows: resolution ratio, 45 mm; volume, 50 kV; current, 500 UA. 2mm thick bone around the prosthesis was used as the volume of interest (VOI) (Fig.3). The change in the amount of cancellous bone depends not only on the distance from the joint surface, but also on the distance from the cartilage growth zone. The implantation was a standardized operation. When we implanted the prosthesis, the stem passed through the growth plate. The distal end of the prosthesis was 6.5mm under the joint surface and 4.5mm under the growth plate. Continuous scanning was performed from the proximal to distal tibia along the longitudinal axis of the tibia. The scanning length was 5 mm and the layer thickness was 18 um. Then 3D reconstruction of the trabecular bone was performed, bone mineral density (BMD), ratio of bone volume to tissue volume (BV/TV), trabecular thickness (Tb.Th), trabecular separation (Tb.Sp), and trabecular number (Tb.N) of the VOI were calculated by Mimics 19.0 software (Materialise, Leuven, Belgium).
Biomechanics
WWD-10A electronic universal testing machine controlled by microcomputer (Sanfeng Instrument Technology Co., Ltd. Changzhou, China) was used to pull out the prosthesis of two groups of samples. Following euthanasia, 8 tibias were harvested,and surrounding soft tissue was removed carefully. The proximal bone and prosthesis were exposed clearly. Two ends of tibia specimen and prosthesis were fixed with self-curing plastic(Methyl methacrylate) respectively. Two 1mm diameter steel wire ropes were respectively reserved in the self-curing plastic at both ends. The steel wire ropes were connected with the clamps at both ends of electronic universal testing machine to keep the longitudinal axis of tibia perpendicular to the horizontal line of the ground. The axial and lateral stress of the prosthesis must be avoided during the whole mechanical process. The load measurement accuracy of the tester is 0.01N, and the loading speed is 0.1mm/s. Pull-out test is carried out to check the intensity of osseointegration. The stress-time curve is described by the self-contained software, and its peak value is the maximum pull-out force.
Histopathology
Following euthanasia, 8 retrieved upper tibial bone including implants were processed for undecalcified histology. After fixation in 4% paraformaldehyde for 48h, the implants were rinsed in water for 24h, dehydrated in ethanol step by step, cleared in xylene for 12h, and embedded in polymethyl methacrylate resin (Leica Microsystems GmbH, Wetzlar, Germany). After sufficient polymerization, each implant was cut along its vertical axis giving 7-8 sections by using hard tissue slicer (SP1600 Leica Microsystems Concord, ON, Canada). The location of the histological section was 1.5mm from the distal end of the prosthesis. The thickness of each slice was 70um, and three most central sections from each specimen were kept for polishing, staining with 1% toluidine blue and histological analysis. Semi-automatic digital image analysis system of bone morphology was used for histomorphometric analysis. The system includes microscope (Olympus BX51, Japan), digital camera (Nikon H600L, Japan) connected with computer, and bone morphometry analysis software (Bioquant, USA). Bone-to-implant contact (BIC) value is the ratio of the length of direct contact between the surrounding bone tissue and the implant body to the total circumference of the implant body. BIC value can be calculated by software to indicate the degree of bone integration.
Statistical analysis
Before the investigation, the sample size was estimated by using bone-implant contact (BIC) as the primary variable. On the basis of our pilot studies , the standard deviation was assumed at 13% in both experiment and control groups, and an estimated difference of 20% was determined between the groups. A power calculation was performed with a confidence level of 95% (α=0.05) and power (1-β) of 80%. This yielded an estimated sample size of 7 knees per group. We would use N=8 for every examination at each time point to account for potential unexpected losses of animals during our experiments.
All the data were described as mean±standard deviation. Student’s t-test was used for comparison of means between the two groups (SPSS version 20.0), p<0.05 was determined to be significant.