| Background and purposes of the researchBone tissue engineering refers to a process to mimic physiological bone regeneration. The connection of molecular signal mechanism exists between blood system and skeleton, on which performing profound research helps to the further development in bone tissue engineering. G protein coupled receptors (GPCRs) represent the first switch not only to receive signals from outside, but also to transduct molecular signals to intracellular transcript factors which function to regulate the biological activities of various cells. G protein coupled receptor 48 (GPR48) gene knock-out fetal mice showed both anemia appearance and marked development retardation of skeleton, this phenotype suggested that some molecular connection existed between hematopoiesis and skeleton, but we haven't made it clear. Therefore, it is helpful to obtain profound acknowledge about bone regeneration under the research on the molecular connection between hematopoiesis and skeleton development through GPR48 signal pathway.In recent years, varied components in blood, especially platelet, have attracted more and more attention due to the important role they played in bone regeneration. Blood platelets function to regulate the proliferation and differentiation of varied cells including osteoblast and osteoclast by way of releasing various growth factors when activated by coagulate mechanism. The maintenance and activation of normal platelets' functions depends on the regulation by many signal pathways via GPCRs. Therefore blood platelet forms the bridge to connect blood system with bone repair through GPCRs pathway. Most resently, platelet lysate (PL), which has been reported to be derived from platelet and loaded with large amount of growth factors and less antigen, functions as a highly effective mitogen, all of which implies that PL might be a potential resource of growth factor substitutes applied into bone tissue engineering.The purpose of this reaseach aims: 1) To explore the molecular connection between blood components and skeleton tissue: the shared molecular signal pathway in erythropoiesis and skeleton development for GPR48 gene-traped fetal mice; 2) To evaluate the effects of PL as a resorce of growth factors on bone regeneration by way of bone tissue engineering strategies both in vitro and in vivo.Material and Methods1 Study on the shared molecular signal pathway functions both in fetal erythropoiesis and skeleton development1.1 Generation of GPR48 knocked out mice, determination of genotype and expression patternGpr48 knocked out mice of (129×C57BL/6) were generated via secretory-trap approach. Homozygous were obscured by intercrossing heterozygous mice. E12.5～14.5 fetal mice isolated from HE female pregnant mice were selected for this research program.Genotype of mice (wild-type +/+, heterozygous +/-, and homozygous -/-) was determined by PCR to distinguish the type of GPR48 genomic DNA extracted from the tails of fetal mice. LacZ histochemistry assay was carried out to detect the discrepancy of GPR48 expression in skeleton and liver between GPR48 wild-type and homozygous fetal mice.1.2 Grossic phenotypeTo observe the gross characteristics including size, color both for embryos and liver, the size of liver and action status of fetal mice obscured from E12.5～14.5 embryos.1.3 Analysis of erythrocytes and hemoglobin chains in peripheral blood Blood smears were prepared by wedge technique following Wright-Giemsa staining protocol for E12.5～14.5 embryos. For E13.5 fetal mice (8 WT and 7 HO) and E14.5(5WT, 4HO), 3 blood smear slides were randomly selected from each embryo in WT or HO, and 3 high power views of each slide were randomly chosen to count the number of nucleated erythrocytes and enucleated erythrocytes, and calculated the relative level of nucleated erythrocyte in total number of erythrocytes for comparisons.Real-time PCR was performed to detect the mRNA level ofβh1 andβhemoglobin chains in blood of E13.5 WT (6 embryos) and HO (6 embryos).1.4 ATF4 expression in fetal liver and skeletonThe total RNA and protein extracted from liver and bone (lib cage) of both E13.5 wild-type and homozygous embryos were detected by RT-PCR and Western-blotting approaches respectively to assay ATF4 expression in liver and skeleton (lib cage) at both of mRNA and protein level.1.5 Histology and Immunohistochemistry (IHC) assayFetal liver (E13.5) dissected from mice were fixed in 10% formalin and embedded in paraffin according to the standard techniques. Five micrometer-thick sections were stained with hematoxylin and eosin (H&E). For proliferation assays, the sections of both the liver (E13.5) and femur (E14.5) were stained with anti-PCNA using PCNA kit and then counterstained with hematoxylin. For ATF4 expression in liver and bone, the following antibodies were used: rabbit polyclonal anti-ATF4 (C-20) (1:100), hematoxylin was carried out as counterstain.1.6 Statistical analysisData were presented as mean±standard deviation and analyzed by 2-independed samples non-parameteric test with SPSS 11.5 software. For all analyses, P < 0.05 was considered statistically significant.2 Evaluation of the biological effects of PL via the strategies of tissue engineering2.1 Effects of PL on the proliferation and osteogenic differentiation of BMSCs. 2.1.1 Isolation, expansion and phenotypic analysis of BMSCsTotal 8 adult healthy Wistar rats with weight of 250-300g of clean grade were used in this section, whole bone marrow was cultured and expanded by changing medium in half every other day and passaged at a ratio of 1:3. The cell growth was observed by inverted microscope. The 5th passages were prepared to determine the expression of cell surface antigens CD45/CD90/CD29 which function as the specific marker of BMSCs by flow cytometry.2.1.2 Preparation of PL and ELISA assay for the determination of growth factorsPL was obtained through three times of centrifugations and repeated freeze-thaw for the venous blood aspirated from 16 rats and filtrated through 0.22μm sterilized filter membrane. ELISA assay was conducted following the manufacture's protocol to determine the concentration of growth factors PDGF, TGF-β1, IGF-1 and VEGF in PL samples from 6 rats.2.1.3 Effects of PL on cell growthPL at final concentration of 1% and 5% in basic medium (V/V) was prepared as conditioned medium. BMSCs of the 5th passages were classified into three groups of A1 (5% PL in basic medium) , Bl(l% PL in basic medium), and C1(no presence of PL in basic medium as controls) to expand BMSCs respectively for a series of 9 days so as to draw the cell growing curves through the assay by CASY cell analyzer. The living cell number of all groups was compared at 9d.2.1.4 Effects of PL on osteogenic differentiation of BMSCs2.1.4.1 The establishment of osteogenic differentiation systemThe cells at fourth passage were performed for osteogenic differentiation under induction environment in three groups of A (5% PL of final concentration in basic induction medium), B (1% PL of final concentration in basic induction medium), and C (no presence of PL in basic induction medium as controls). The morphological changes of cells were dynamically observed with inverted phase contrast microscope.2.1.4.2 ALP (Alkaline phosphatase) activity and mineral formation of BMSCs under inductive differentiation with different PL concentrationALP staining (7d) and ALP/TP (2, 8,12d) of the cells were detected to evaluate ALP activity for 3 samples of each group. At 20d, Von Kossa and Alizarin red staining were both carried out to observe extracellular mineral formation. For Alizarin red staining, 3 cases of each group were used to examine the mineral formation in extracellular martrix, randomly chose 3 views (×200) to obtain the number and area of mineralized nudes under microscope for each case with image analyis software.2.2 Study on the biocompatibility between PL/ADBG/CG and BMSCs2.2.1 Preparation of allogeneic decalcified bone granules (ADBG)10 Wistar rats were used to prepare ADBG, the limbs bone were processed through deep freeze, supersonic clean, grinding to obtain granules with diameter range of 300-500μm. The granules were further processed through 0. 6N HCl washing, lyophilization and radiation.2.2.2 The combination of Collgen I (CG) with ADBG via dural-phase precipitationThe CG resolution at the concentration of 0.25% (W/V) was prepared using 0.1M acetic acid as basic solvent and sterilized by radiation. ADBG was added into this CG resolution according to W/V fraction of 6% and mixed well to obtain a suspension, to which IN NaOH was added to adjust the PH value to 7.0 approximately, then CG and ADBG precipitated and adhered together to form a gel-like composite CG/ADBG (100mg/1g), which then was dried on the filter paper and stored at -20℃.2.2.3 Preparation of PL/ADBG/CG and co-culture with BMSCsPL/ADBG/CG was obtained by immersing 5g ADBG/CG with 5mL PL under aseptic condition at 4℃. BMSCs of fifth passage were coincubated with PL/ADBG/CG for 7d to observe the biocompatibility and adhensive growth with inverted microscope, and scanning electron microscope was conducted for the composite at 3d.2.3 Effects of PL/ADBG/CG transplantation on reconstruction of femoral condylar defect in vivo2.3.1 Establishment of femoral condylar defect model and graft transplantation30 Wistar rats as above description were devided into 3 groups (A, B, C) with 10 rats/20 condyles in each group. The bilateral femoral condylar defects (with 60-70% bone loss) were established for these animals and transplanted with 500mg of A(PL/ADBG/ CG), B(ADBG/CG) or C(CG) filled into defect respectively.2.3.2 Evaluation on bone defect repair after operation2.3.2.1 X-ray and bone density assayAfter operation, X-ray was performed to observe the new bone formation and repairing status. Image-Pro Plus 5.0 software was used to measure the gray level per unit area in repairing field of 8 femoral condyls in each group at postoprerative 2 and 4w, which represented the bone density for the evaluation on bone defect repair.2.3.2.2 Histology and morphological quantitative analysesAt 4 weeks, the femoral condyles of 3 rats in each group were cut off and processed by fixation, decalcified, dehydration, and other procedures to make paraffin sections with thickness of 5μrn. After regular HE staining, microscope was conducted to observe the bone formation, 3 sections were randomly selected and 5 views in each section under microscope (×200) were randomly chosen to measure the new bone area (pixels) per view (×200) with Image-Pro Plus 5.0 software.2.3.2.3 Biomechanical assay of anti-pressureAt 4 weeks, 6 condyles from 3 rats in each group were used for mechanical test together with 6 normal condyles of littermates as group D. Condyles were horizontally placed on the object stage, and pressed at the loading speed of 5mm/min. The destructive loading was recorded when displacement attained to 2mm and calculated the destructive intensity.2.3.2.4 Determination of T lymphocyte subsets in peripheral bloodBefore killing rats at 4w, 3 rats were randomly choosed from A,B and C ,and another 3 rats littermates were performed as controls. Three-color flow cytometry was conducted by using CD4-FITC, CD3-PE and CD8-TC to detect percentage of T lymphocyte subsets as well as the ratio of CD4/CD8.2.4 Statistical analysisData were presented as mean±standard deviation and analyzed by Factorial analysis for interaction effects, and One-way ANOVA was performed for one-factor effect. LSD was conducted for related multiple comparisons. For all analyses, P<0.05 was considered statistically significant. Results1 GPR48-CAMP-CREB-ATF4 signal transduction referred to the shared molecular signal pathway to regulate the development of both erythropoiesis and skeleton at embryonic stage1.1 Targeted inactivation of GPR48 and expression pattern in fetal liver and skeletonThe murine GPR 48 genomics was disrupted by randomly inserting a large secretory trap vector (11.98kb) to the intron 1. The insertion caused deletion of GPR48 in mouse genome, which was confirmed by PCR analysis which showed that GPR48-/- null mice didn't present the band of normal GPR48 gene at 450bp. LacZ staining showed that GRP48 was expressed both in E13.5 fetal liver and skeleton in blue color.1.2 Growth retardation and erythropoiesis phenotypeThe gross appearance of GPR48-/- null fetal mice from E12.5 to E14.5 showed growth retardation both in body size and liver size, limbs were shorter and both body and liver were markedly paler when compared with wild-type controls.The peripheral blood smear for E13.5～E14.5 indicated that HO had more nucleated erythrocytes in blood. E13.5 blood smear assay displayed the proportion of nucleated erythrocyte in total erythrocytes of GPR48-/- null fetal mice elevated about 2.7 folds of normal litters, which with significant difference between these two groups (HO 79.91±3.57, WT 30. 00±2. 17; z=-3.24, P<0.01). For E14.5 mice, there also was markedly difference between HO and WT (HO 54.99±1.93, WT 27.15±1.88; z=-2.45, P<0.02).Real-time PCR showed that relative mRNA level ofβh1 in E13.5 homozygous blood was significantly increased than that in normal littermates (z=-2.882, P<0.01). Otherwise, the relative mRNA level ofβin homozygous blood was markedly reduced than that in wild-type littermates (z=-2.882, P<0.01).1.3 ATF4 expression in GPR48-/- fetal liver and skeleton In RT-PCR, ATF4 expression in E13.5 GPR48-/- fetal liver and skeleton (lib cage) showed that HO presented a significantly weaker band at 400bp than WT. In Western-blotting, HO also presented a markedly weaker band at 40kD than WT. Immunohistochemistry assay indicated that ATF4 expression in both HO liver (E13.5) and skeleton (E14.5 femoral condyle) was weakerly in positive staining than normal controls.1.4 Histology and proliferation assayHistological examination and hematoxylin and eosin (H&E) staining showed that the gross morphologic features of homozygous fetal livers were normal in cellular architecture. However, fewer erythroid precursor or definitive progenitors and erythroid foci were visible in the homozygous fetal livers from E13.5 embryos compared with that in wild-type.Immunohistochemical assays (IHC) of PCNA for proliferation assay indicated that the thickness of femoral cortex in E14.5 HO was significantly thinner than that in WT control. The strength of proliferating stain was markedly weaker in the fetal livers and femoral condyles of homozygous Gpr48 null mice compared with their control littermates.2 Evaluation of the biological effects of PL via the strategies of tissue engineering2.1 Effects of PL on proliferation and osteogenic differentiation of BMSCs2.1.1 Isolation, expansion and phenotypic analysis of BMSCsUnder inverted microscope, cells at 5th passage presented the similar morphology of uniform long-spindle shape as fibroblasts and in regular arrangement of parallel aggregation. The cells expressed CD45~-CD90~+CD29~+ antigens by assay of flow cytometry which conformed to the characterized cell surface antigens of BMSCs.2.1.2 ELISA assay of the concentration of growth factors in PLELISA assay presented that the content of PDGF, TGF-β1, IGF-1 and VEGF in PL attained to 405±59.6, 140±26.65, 85.8±16.86, and 82.5±11.7 (pg/ml) respectively. 2.1.3 Effects of PL on cell proliferationUnder different conditioned medium, BMSCs in A1, Bl and C1 displayed nearly the same configuration with long shuttle-shape as fibroblasts. The growth curve displayed that cells grew faster from 3d and the living cells number at every timepoint presented ascending trend along with the prolongation of culture time. Cells in A1 grew fastest within these 3 groups from 3d into logarithmic growth phase which was 1 day earlier than that in Cl. The results of Factional analysis showed that different PL concentration had significant difference in effects on the cell proliferation (F=320.33, P<0.01), while different culture time also markedly affect the cell proliferation (F=406.04, P<0.01), and there was interaction effect between different PL concentration and cultural time (F=36.25, P<0.01) which certificated that cell proliferation presented ascending trend along with the prolongation of culture time in different PL concentration. Further LSD assay of the multiple comparations presented that cell proliferation of group A1 was higher that that of B1 (P <0.01) and C1(P <0.01), and that B1 was higher than C1(P <0.01), these differences were significantly different.At 9d, the living cell number in A1, B1 and C1 was 988436±15720.42, 616666.67±12423.1, and 345000±21931.71 respectively, One-way ANOVA showed there was significant difference within these groups (F=1064.11, P<0.01). Further multiple comparisons with LSD showed that the living cell number in A1 was significantly higher compared with that in B1 (P <0.01) or C1(P <0.01), and living cell number in B1 was higher than that in C1(P <0.01), there differences were significant within them. The living cell number in A1 reached about 3 folds of that in C1(P<0.01) at 9d.2.1.4 Effects of PL on the osteogenic differentiation of BMSCsFor differentiation analysis, morphological observation displayed BMSCs in group A2 showed slower shape-changes but higher proliferation than that in group B2 or C2. Cells in B2 showed higher proliveration than that in C2 but lower that that in A2. At 7d, the cells in group A2 showed smaller amount of granules with ALP positive staining in cytoplasm when compared with B2 or C2. Moreover, at the day of 20, the cells in group A2 still displayed much higher dense growth.At the day of 2, 8, and 12, Factorial assay of ALP activity showed there was marked diffence within these groups (F=45.698, P<0.01). Further multiple comparisons showed marked reduce of ALP activity in A2 compared with that in B2 (P <0.01) or C2 (P <0.01) and B2 displayed no significant decrease compared with C2 (P >0.05). Different days showed significant difference to affect the ALP activity (F=121, P<0.01). There was interaction effect between different days and different PL concentration (F=6.86, P<0.01) , the highest value of ALP activity occurred at 12d.At 20d, both Von Kossa and Alizarin red staining showed there were large amount of mineral deposits occurred in the excellular martrix of B2 and C2, but smaller amount in A2 compare with B2 and C2.In Alizarin red staining, the mineral nodes under the single view (×200) attained to A2 (7.67±1.10), B2 (14.0±2.23), C2 (14.97±1.88) respectively. With One-way ANOVA assay, there was significant difference in the nodes number under the single view within these groups (F=14.593, P<0.01), further LSD asaay showed significantly smaller amount of mineral deposits produced in the excellular martrix of A2 compared with B2 (P <0.01) or C2 (P <0.01), while no marked difference between B2 and C2 (P>0.05). The area of mineral nodes under the single view (x200) attained to A2 (161778.73±44550.80), B2 (337349.67±56083.24), C2 (415921.73±71725.39) pixels respectively. With One-way ANOVA assay, there was significant difference in the area of nodes under the single view within these groups (F=14.831, P<0.01), further LSD asaay showed significantly larger area of mineral deposits produced in the excellular martrix of A2 compared with B2 (P <0.01) or C2 (P<0.01) , while no marked difference between B2 and C2 (P>0.05).2.2 The biocompatibility of PL/ADBG/CG with BMSCsAfter 3d and 7d in coculture, BMSCs grew normally around or adhered to PL/ADBG/CG with no significant shape change under inverted microscope. Under scanning microscope at 3d, living cells with long processes tightly adhered to the surface and grew toward the deep inner of the composite.2.3 Effects of PL/ADBG/CG transplantation on reconstruction of femoral condylar defect in vivo2.3.1 Radiology and quantitative assay of bone densityIn group A, X-ray at 1w showed small amount of resistance projective image in repairing region occurred; at 2w, the density of resistance projective image was increased than before; at 3w, the image of resistance projective gradually tended dense and homogenous; at 4w, the image of bone density was close to normal bone nearby and the defect was nearly repaired. In group B, at 1w the center of defect region showed large amount of translucent region; at 2w, the image density was increased than before; at 3w, there was more irregular translucent region; at 4w, the defect center majored large amount of translucent region and no clues of defect repair.These observations were conformed by quantitative assay of bone density which showed that there was dramatic difference within A, B, and C (2w: F=25.08, P<0.01; 4w: F=29.048, P<0.01). Further multiple comparisons presented that at either 2 or 4 weeks, the bone density in A presented significantly higher than that in B (P <0.01) or C (P <0.01), and B was also significantly higher than C (P <0.05) in bone density.2.3.2 Histology and quantitative assayHistology assay at 4 weeks displayed that there were more osteoblast and new osteoid growing into bone granules, also the phenomenon of absorption was easy to be seen in the dead bone which surrounded by fibrous tissues with small amount of inflammatory cells infiltration. Group A showed large amount of new bone forming from normal bone nearby expanded and more endochondral ossification could be observed. Moreover, there were more osteoblastic and osteoclastic activities in A than in B or C. There were more dead bone pieces in disintegration and absorption and small amount of new bone formation occurred in defect region of group B. There were still large amount of fibrous tissues to fill the defect field in group C, many sites of normal bone walls showed less active in ossification.Quantitative assay displayed the newly reparative bone area within these 3 groups was dramatically different (F=122.398, P<0.01), and further LSD multiple comparisons showed that the newly formed bone area in A was significantly higher than that in B (P <0.01) or C (P <0.01), and B showed more bone formation than C (P 2.3.3 Anti-press biomechanical evaluationAnti-press mechanical measures showed that there was significant difference within A, B, C, and D (F=64. 469, P<0. 01). Further multiple comparisons were conducted and the results showed that either destructive load or destructive energy was higher in A than that in B (P <0.01) or C (P <0.01), but markedly lower than in D (P <0.01) which referred as normal controls, while C presented the worst result with statistic significance (P <0.01).2.3.4 T lymphocyte subsets determinationThe T lymphocyte subsets of CD3~+CD4~+CD8~-, CD3~+CD8~+CD4~- and the ratio of CD4/CD8 showed no significant difference within these 4 groups by One-Way ANOVA assay (P>0.05).CONCLUSIONBlood components incluling erythrocytes and platelets present the internal connection with skeleton tissues either in development or bone regeneration, and platelets functioning as a resorce of multiple factors may be applied to the strategies of bone tissue engineering.1 GPR48-cAMP-CREB-ATF4 signal transduction, which is one of the shared molecular signal pathways to regulate the development of both erythrocyte and skeleton in fetal mice, forms the molecular bridge to connect blood system with skeleton tissue.2 GPR48 signal pathway regulates the development of both erythrogenesis and skeleton via cell proliferation.3 PL is derived from platelet and presents to be a kind of system carrying various growth factors to regulate bone formation. PL functions as a kind of mitogen to promote the proliferation of BMSCs in a dose-dependent manner, while inhibits both of ALP activity and mineral formation of BMSCs also in dose-dependent manner under osteogenic induction environment. But this non-coordination effects of PL on proliferation and differentiation implies that PL may function to promote the bone formation at earlier stage.4 PL presents to be a nice substitute for growth factors and can be applied to the strategy of bone tissue engineering. PL/ADBG/CG, the recombinant decalcified allograft bone, which shows better compatibility with BMSCs, can promote bone defective repair at early stage.5 PL promotes the bone remodeling in repairing the femoral condylar defects, which may be related with the activation of coupling mechanism between bone absorption and formation via the synergistic effects of various factors.
|
PDF Full Text Request