3.1 Implant stability Quotient measurement
ISQ was quantitative analysis of the degree of fixation to the bone after implant placement and was measured using an Osstell Mentor. There were 6 types of implants placed in rabbit tibia (3wks each group n=8, and 6wks each group n=8), and the measured value average were distributed between 74 and 79 (Fig. 3A, Additional file 1: Figure S1). Statistical significance of ISQ measurements in rabbit was observed in immy MS and 6wks MS (p<0.001), immy ME and 3wks MS (p<0.05), immy ME and 6wks MS (p<0.05), immy MA and 3wks MS (p<0.05), immy MA and 6wks MS (p<0.01), 3wks MA and 6wks MS (p<0.01), 3wks MA and 3wks MS (p<0.01), 3wks ME and 6wks MS (p<0.001), 3wks MS and 6wks MS (p<0.001), 6wks MA and 6wks MS (p<0.001), and 6wks MS and 6wks ME (p<0.01) (Fig. 3A).
There were a total of 4 types of implants placed in the beagle dog (6wks each group n=8, 12wks each group n=8), and the measured value average were distributed between 65 and 81 (Fig. 3B). Compared with immy MA, Statistical significance of ISQ measurements was observed with immy MS, immy ME, 6wks MA, 6wks MS, 6wks ME, 12wks MS and 12wks ME (p<0.001), and 12wks MA (p<0.01). Compared with immy MS, it was observed in 6wks MA, 6wks ME, 12wks MA and 12wks ME (p<0.001), 12wks MS (p<0.01), and 6wks MS (p<0.05). Compared with immy ME, it was observed in 6wks MA, 12wks MA and 12wks ME (p<0.001), 6wks ME and 12wks MS (p<0.01), and 6wks MS (p<0.05). Compared with 6wks MA, it was observed in 6wks MS, 6wks ME, 12wks MA, 12wks MS and 12wks ME (p<0.001). Compared with 12wks MA, it was observed in 6wks MA and 6wks ME (p<0.001) (Fig. 3B).
3.2 In vitro cellular study
The effects of ENF surface on the growth and proliferation of hMSCs were evaluated. The cell proliferation was assessed by CCK-8 assay, which determines cell population by quantification of reaction between metabolic products of cells and the assay solution. During 7 days of incubation, every group did not hinder the proliferation of hMSCs compared to the TCP group. Until the 4 days of incubation, the proliferation rates of MA, MR, MS, and their ENF counterparts were comparable to that of the TCP group. After 7 days of incubation, cell proliferation of ME, MRE, and MSE groups were significantly (p < 0.05) increased compared to their non-ENF counterparts, while there were no significant differences between ENF groups (Fig. 4). It was observed that the surface treatment through a single technique (MA, MR, and MS alone) did not facilitate cell proliferation compared to TCPs, while the additional ENF coating was significantly enhanced the cell proliferation after 7 days of culture. These results suggest that the synergistic effect of ENF with conventional functionalization methods can enhance the cytocompatibility of implant surfaces. These findings are mostly in accordance with those of previous studies demonstrating nanopatterned surface facilitates cell proliferation and cytocompatibility [19–22]. This is mainly because the transcriptional response of stem cells concerning cell proliferation is affected by cytoskeletal rearrangement by nanotopography sensing [23–24]. Furthermore, nanopattern can mediate the regulation of microRNA for maintenance of population growth and phenotype of stem cells [25]. Because the adhesion and early proliferation of surrounding cells on implant surface are essential factors for successful osteogenesis, ENF dual-modification is considered to be advantageous for osseointegration of implant at early post-transplant state [26–28].
For further investigation of the osteogenic capability of ENF surface, ALP activities of hMSCs on MA, MR, MS, and their ENF counterparts were assessed. The ALP activity is known as a marker for osteogenesis and is expressed in an early stage of osteogenic differentiation of stem cells [29]. On the first day of incubation, there was no significant difference between groups. On 7 and 14 days of incubation, ALP activities of ME, MRE, and MSE groups were significantly increased compared to non-ENF and TCP groups. Especially, the ME group significantly facilitated ALP activity of hMSCs at 7 and 14 days indicating the topography of ME is advantageous for early osteogenesis of hMSCs (Fig. 5). In addition to ALP activity, mineralized calcium nodule deposition of hMSCs, regarded as a marker for the late stage of osteogenesis and bone regeneration, was stained with ARS solution and quantified. On the first day of culture, MA group showed a slight decrease of mineralization nodule while MS, ME, and MSE groups showed a modest increase of that. On 14 and 21 days of culture, mineralization nodules of MS, ME, and SLA groups significantly increased while MA groups still maintained a similar level to the TCP group. Especially, ME induced the highest degree of mineralization nodule formation of hMSCs at 14 and 21 days of culture, suggesting the osteogenic differentiation and bone regeneration of hMSCs are most promoted on ME group (Fig. 6A, B). Considering the previous CCK-8 and ALP assay, the ENF double coating can promote the proliferation and osteogenesis of hMSCs. Especially, the ME group did not hamper proliferation rate and significantly increased early and late stage of osteogenesis of hMSCs compared to MRE and MSE groups.
Many studies emphasize that surface roughness and morphology have crucial effects on cell-matrix that regulate integrin-mediated signal cascade [30–32]. In a mechanotransductional interpretation, osteoblasts have larger cell morphology than other lineages, including adipocytes, fibroblasts, and MSCs, hence need large adhesion to support the tensile cytoskeletal scaffolding [33]. Interestingly, the cytoskeletal tension of MSCs can regulate differentiation lineages. McBeath et al. demonstrated that small fibronectin microcontact (1,000 µm2) led MSCs to differentiate into adipose cells, while large fibronectin contact (10,000 µm2) promoted spreading of MSCs leading to differentiation into osteoblasts [34]. Therefore, ENF-functionalization supported nanotopographical cues on hMSCs that increase the area of focal adhesion and tensile stress of cytoskeletons, hence, could have facilitated osteogenic differentiation of hMSCs. Moreover, microporous surfaces are known to promote cells to secret osteotropic factors including 1α,25(OH)2D3, PGE2, and TGF-β1 [35]. On the other hand, nanopatterned substrates often enhance protein adsorption and intracellular protein delivery by forming ionic bonding and electrical conductivity [36]. These lead to the anchoring of proteins in osteogenic media including dexamethasone, β-glycerolphosphate, and ascorbic acid. It is known that dexamethasone could upregulate many proteins and enzyme levels concerning osteogenesis, hence enhaces calcium deposition [37]. Dexamethasone synergistically acts with β-glycerolphosphate to enhance the ALP activity level in the cells and ascorbic acid favorably affects the maturation of osteoblasts [38]. Therefore, concerning in vitro cellular assays, it is suggested that ME is considered to be the most effective method for coating the implant surface to promote osseointegration and bone regeneration.
3.3 Histomorphometric analysis
After implant placement in rabbit and beagle dog, hematoxyline and eosin (H&E) staining was performed after sacrificed at each experimental schedule. Based on this, BV, new bone area (NB), and BIC was quantitatively analyzed.
As a result of H&E analysis of implant experiments in rabbits, in 3wks MA, bone changes were not evident and inflammatory tissue and connective tissue were observed at the interface, and in 6wks MA, most of the bone defect area was filled with new bone and partial bone remodeling was observed (Fig. 7A, D). In 3wks MS, osteoblasts and osteoclasts were distributed, and continuous bone realignment and differentiation were observed. In 6wks MS, new bone formation was observed to be active, but the contact of the implant interface was observed to be low (Fig. 7B, E). In 3wks ME, osteoblasts were abundant, and bone resorption and bone remodeling of existing bone were observed. In 6wks ME, osteoblasts were aligned and attached between the implant and bone, and most of the bone remodeling was completed, and the appearance of mature bone was also observed (Fig. 7C, F). In 3wks MR, bone resorption proceeded extensively at the implant interface and partial new bone formation was observed. New bone formation was well performed but bone remodeling was not started. In 6wks MR, bone remodeling was mostly completed (Additional file 1: Figure S2A, D). In 3wks MSE, new bone was being formed and contact with the implant interface was observed as an osteoblast lining. Osteogenesis was active but bone density was relatively low. In 6wks MSE, bone remodeling was mostly completed and contact with the implant interface was high (Additional file 1: Figure. S2B, E). In 3wks MRE, new bone formation was actively performed, but bone contact at the implant interface was low, and new bone formation occurred actively but bone density was low. In 6wks MRE, new bone formation and bone remodeling were in progress, and soft tissue were observed at the implant interface and many immature bones were observed (Additional file 1: Figure. S2C, F).
The implant placed in rabbits were observed separately in the initial period (3wks) and in the recovery period (6wks). Compared with 3wks ME, Statistical significance of BV measurements was observed with 3wks MA, 3wks MS, 6wks MA and 6wks MS (p<0.001). Compared with 6wks MS, it was observed with 3wks MA (p<0.01), and 3wks MS and 6wks MA (p<0.05). Compared with 6wks ME, it was observed with 3wks MA, 3wks MS, 6wks MA and 6wks MS (p<0.001) (Fig. 8A, Additional file 1: Figure S3A).
BIC was measured the contact rate between the implant thread and the regenerated bone. Compared with 3wks ME, Statistical significance of BIC measurements was observed with 3wks MA (p<0.01), and 6wks MA, 6wks MS and 6wks ME (p<0.001). Compared with 3wks MS, it was observed with 3wks MA, 6wks MA, 6wks MS and 6wks ME (p<0.001). Compared with 3wks MA, it was observed with 6wks MA and 6wks MS (p<0.001), and 6wks ME (p<0.01). Compared with 6wks ME, it observed with 6wks MA and 6wks MS (p<0.001). Statistical significance was observed because of comparison with 6wks MS and 6wks MA (p<0.001). As a result of BIC comparison between implants at each week, statistical significance was observed at both 3wks and 6wks (Fig. 8B, Additional file 1: Figure S3B).
NB measured the new bone area within the ROI after setting the ROI between the implant screw thread and the thread. Compared with 6wks ME, Statistical significance of NB measurements was observed with 3wks MA and 3wks MS (p<0.01), and 6wks MS, 6wks MA and 3wks ME (p<0.01). As a result of comparing NB between implants at each week, 3wks were not statistically significant, and 6wks were observed to be statistically significant (Fig, 8C, Additional file 1: Figure S3C).
As a result of H&E analysis of implant experiments in a beagle dog, new bone formation was observed in all groups at 6wks, but bone remodeling was not observed clearly, and the 12wks tissue showed the most active bone remodeling of ME (Fig. 9)
Compared with 12wks MA, Statistical significance of BV measurement was observed with 12wks MS (p<0.01), and 6wks MA, 6wks MS, 6wks ME and 12wks ME (p<0.001) (Fig. 10A).
Compared with 6wks MS, Statistical significance of BIC measurement was observed with 6wks MA and 6wks ME (p<0.01). Compared with 12wks MA, it was observed with 6wks MA and 6wks ME (p<0.001). Compared with 12wks MS, it was observed with 6wks MA and 6wks ME (p<0.001). Compared with 12wks ME, it was observed with 6wks MA and 6wks ME (p<0.001), 6wks MS and 12wks MA (p<0.05), and 12wks ME (p<0.01) (Fig. 10B).
Compared with 12wks MA, Statistical significance of NB measurement was observed with 6wks ME and 6wks MS (p<0.001), and 6wks MA (p<0.01). Compared with 12wks MS, it was observed with 6wks MA, 6wks MS, 6wks ME, 12wks MA and 12wks ME (p<0.001). Compared with 12wks ME, it was observed with 6wks ME, 6wks MS and 12wks MA (p<0.001), and 6wks MA (p<0.01) (Fig. 10C).
RT (12wks each group n=4) was measured at 12wks to mechanically check the stability of the implant-bone bonding. MA 32.25 ± 1.91 N/cm3; MS 116.77 ± 24.05 N/cm3; ME 54.60 ± 10.02 N/cm3. MA had the lowest RT among all implant groups, and ME was measured to be about 54% higher than MA but was measured to be about 50% lower than MS (Fig. 10D).