Preparation of β-TCP scaffolds with multiple channels
Two geometries of interconnected porous β-TCP scaffolds were fabricated by using a template-casting method that was reported in the previous study22. One is porous β-TCP scaffold, the other is porous β-TCP scaffold with five channels. The diameter of each channel is 1 mm. The five channels are homogenously distributed in the porous β-TCP scaffold. Briefly, β-TCP slurry was prepared by stirring mixture of β-TCP nano-powder (Nanocerox, Inc, Ann Arbor, Michigan), carboxymethyl cellulose powder, dispersant (Darvan C), and surfactant (Surfonals) with distilled water (Fisher Scientific). After filling the paraffin beads into the two types of customized modes, β-TCP slurry was casted into the molds and solidified in ethanol for two days, followed with a gradient ethanol dehydration. The completely dried β-TCP green-bodies were then sintered for 3 h at 1250 °C. The fully sintered scaffolds were then used in the in vitro and in vivo studies.
Characterizations
The pore morphologies of the porous scaffold were observed by scanning electron microscopy (SEM). The dried scaffold was sputter-coated with gold for observation under a benchtop SEM (JEOL, JCM-6000Plus). The tortuosity of pores was observed by a MicroCT scanner (Imtek Micro CAT II, Knoxville, TN) at a resolution of 80 μm.
In vitro cell attachment quantification
To measure whether hBMSCs’ attachment efficiency was promoted by channels, six scaffolds of each type were placed in the wells of 24 ultra-low adherent well plates (Corning, NY). One scaffold per well was set. hBMSCs from Lonza were cultured with MSCGM medium (Lonza, Basel Switzerland) under a standard cell culture condition. Two milliliters of hBMSCs suspension with a concentration of 5 × 105 cells/mL were added into each well. To dynamically seed cells onto the scaffolds homogenously, the well-plates were then placed on a 3D platform rotator (Fisher scientific, 3D platform rotator, Hampton, NH) and rotated at 30 rpm in CO2 incubator at 37 °C. At 4 and 16 h, 3 wells with each type of scaffolds were used to determine the cell attachment efficiency according to our published method17. Briefly, each scaffold was taken out and washed with 1 × PBS three times. All washed PBS, the leftover media, and the trypsinized content from each well were collected into a 15 mL centrifuge tube. The cell number was counted from each tube. Final cell attachment efficiency was determined by measuring the cell number per scaffold surface area and followed the equation: Na = (Ns − Nc)/As, where Na stands for the number of attached cells per unit area, Ns stands for the initially seeded cell number, and Nc stands for the collected cell number in the tube, and As stands for the surface area of each type of scaffolds. The surface area of each scaffold was calculated through the dimensions and volumes of the scaffold and its pores.
hBMSCs proliferation and osteogenic differentiation
Channels’ function on hBMSCs proliferation property was investigated in both static and dynamic culturing conditions. One hundred microliters of 105 hBMSCs suspension was pre-seeded onto each scaffold and incubated for an hour at 37 °C, after which equal and enough MSCGM medium was added to each well to fully cover the scaffolds and kept culturing for another 24 h. The cells/scaffolds were kept culturing in 24-well plates with MSCGM medium for static condition measurement. A continuous 3-, 7-, and 14-days’ culture with medium changed every 2 days was followed. For dynamic condition measurement, after 24 h of initial culture, all the cells/scaffolds were transferred into a dynamic circulating bioreactor system where fresh MSCGM medium was circulated through the cells/scaffolds at a rate of 10 μL/min for 3, 7 and 14 days as well, according to the setting of the dynamic culture system we used in previous study17. At the end of each time point, samples from both static and dynamic culturing conditions were collected and rinsed with 1 × PBS twice and stored at -80 °C.
Fluorometric assay was utilized to measure the cell proliferation property quantitatively. Briefly, the cell lysate was collected by immersing all the stored cells/scaffolds samples with 0.2% Triton X-100 in 1 × TE buffer solution and followed by three freeze–thaw cycles. For each cycle, cells/scaffolds were frozen at − 80 °C for 20 min, and then thaw to 37 °C for another 20 min. After that, the content of dsDNA from each scaffold was measured by using a Quant-iT™ PicoGreen™ dsDNA Assay Kit (Invitrogen, Carlsbad, CA) according to the manufacturer’s instruction. All the samples were read through Spectra Max Gemini EM plate reader with 480/520 nm excitation/emission wavelengths43.
Alkaline phosphatase (ALP) activity was measured as one way to investigate the effect of channels on hBMSCs osteogenic differentiation. The cells/scaffolds were cultured in both static and dynamic systems as described above as well. However, after 24 h pre-seeding in MSCGM medium, the culturing media was changed to osteogenic differentiation media, which contained 10% FBS, 10 mM β-glycerophosphate, 10 nM dexamethasone, 50 mg/mL ascorbic acid, and 1% PSA, other than MSCBM. The cells were continuously cultured for 3, 7 and 14 days, and samples were rinsed with 1 × PBS, collected and stored in − 80 °C at each time points for further tests.
To determine the cell ALP activity quantitatively, the total protein was isolated through three times of 80 °C to 37 °C freeze–thaw cycles as well by immersing all the stored cells/scaffolds osteogenic differentiation samples within 0.2% Triton X-100 solutions. The total protein was measured through Pierce™ BCA Protein Assay Kit (Thermofisher Scientific), and read by Spectra Max 190 plate reader at the wavelength of 562 nm. The total ALP was determined through Quantitative Alkaline Phosphatase ES Characterization Kit (EDM Millipore, CA), and read by Spectra Max 190 plate reader at the wavelength of 405 nm. Final cell ALP activity was determined through the following equation: CAa = Ca/Cp, in which CAa stands for ALP activity, Ca stands for the total alkaline phosphatase amount and Cp stands for the amount of total protein of each sample44.
Real time PCR
Osteogenic differentiation related genes were run by real time PCR. Human BMSCs were cultured with osteogenic differentiation media in the same static and dynamic conditions as mentioned above. At the end of 7 and 14 days, total RNA of cell/scaffolds samples were extracted by using RNeasy Mini Kit (QIAGEN, Hilden, Germany). The concentration of RNA was measured through NanoDrop™ One/OneC Microvolume UV–Vis Spectrophotometer (Thermofisher Scientific). mRNA of all samples was then transcribed into cDNA as templates in real-time PCR by iScriptTM cDNA Synthesis Kit (Bio-Rad Laboratories, Hercules, CA). Quantitative real-time PCR was performed afterwards by using iQ™ SYBR® Green Supermix Kit (Bio-Rad Laboratories) on an AriaMx Real-Time PCR System (Agilent Technologies, Santa Clara, CA), and followed the instructions according to the manufacturers. Specific primers (expressed by italic, lower cases) including runt-related transcription factor 2 (Runx2), alkaline phosphatase (alp), bone sialoprotein (bsp), osteocalcin (oc), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) with the sequences shown in (Table 1) were purchased from Invitrogen. The (2^{{ – Delta Delta C_{T} }}) method was used to relatively quantify genes expression, where (Delta Delta C_{T} = left( {C_{T, Multiple} – C_{T,Non} } right)_{Target;gene} – left( {C_{T, Multiple} – C_{T,Non} } right)_{GAPDH}), in which expressed target genes were normalized to the expression level of a housekeeping gene GAPDH45,46.
Immunofluorescent staining
The samples were immuno-probed with FAK and YAP-1 antibodies by following the instruction of manufacturers to investigate whether the addition of multiple channels inside scaffolds initiates the markers along with mechanotransduction pathway of hBMSCs. Briefly, the immunofluorescent staining of FAK and YAP-1, both non-channeled and multiple-channeled scaffolds were incubated with primary anti-FAK (1:100) and anti-YAP-1 (1:50) antibodies (Abcam, Cambridge, UK) separately at 4 °C overnight, followed with a subsequently fluorescent conjugated secondary antibody Alexa Fluor 594 (Invitrogen) labeling for 2 h at room temperature. F-actin fiber of cells on the scaffolds was stained by a fluorescent Phalloidin kit (Cytoskeleton, Denver, CO). Briefly, Actin-stain 555 phalloidin solution was used for the incubation of two types of scaffolds at room temperature for 30 min by following the instruction of manufacturer. After that, all the samples were co-stained with 4′6-diamidino-2-phenylindole (DAPI). The stained scaffolds were then placed on a glass cover slide on an inverted fluorescent microscope. Stained cells on the scaffolds were observed and imaged through a Nikon TE-2000 fluorescent microscope. Three random regions of each type of scaffold from the bottom to an inner area of the scaffold that the lens reached were pictured. The number of positively stained cells by FAK, YAP-1, and F-actin, and the total cells of each image per type of scaffolds were semi-quantified by FIJI ImageJ (NIH). The ratio of positively stained cells to total cells were calculated.
In vivo implantation and fluorochrome labeling
Animals and ethical aspect
This study was carried out in compliance with the ARRIVE (Animal Research: Reporting In Vivo Experiments) guidelines. All the experiments were performed in compliance with the guidance and regulations of the Institutional Animal Care and Use Committee (IACUC) of Florida Atlantic University (FAU). All the experimental protocols were approved by the ethical committee of FAU IACUC, and the approved IACUC protocol number is A16-30. Nine female and nine male Wistar rats with the body weight 226–250 g were purchased from Charles River laboratories (Wilmington, MA).
In vivo implantation
To evaluate the function of the channel-pore architecture on bone regeneration, scaffolds were implanted in rat critical-sized calvarial bone defects. The animals were randomly divided to three groups, including non-channeled, multiple channeled scaffolds, and no-implant as a control. Six rats were used for each group (three female and three male rats). A circular 8 mm defect in diameter was created on each rat’s calvarial bone using a dental drill, and a scaffold with 8 mm in diameter and 1.5 mm in height was implanted. All the scaffold samples were shaped with sandpapers, washed with 70% ethanal and 1 × PBS for 3 times each, and autoclaved before utilization in the implantation surgery.
Fluorochrome labeling was performed postoperatively. Alizarin red and Calcein green were injected subcutaneously at week 6 and 10 weeks separately after the surgery. Both fluorochromes were purchased from Sigma-Aldrich (Munich, Germany), and solutions was prepared as shown in (Table 2). Before injections, both fluorochrome solutions were adjusted to pH 7.4 and sterilized through a 0.22 µm filter. The dosage was calculated in accordance with the body weight (BW) of each rat.
Tissue harvest, fixation and X-ray characterization
All the animals were euthanized with 5% CO2 after 3 months of surgery. The entire defect with implanted scaffolds, and adjacent native bone tissue were harvested, rinsed with 1 × PBS, and fixed with 10% buffered formalin solution (Thermofisher Scientific) for 48 h. All the samples were characterized by a SkyScan microCT machine (Bruker, Billerica, MA) associated with the digital image analysis software (CT Analyser Version 1.18.4.0, SkyScan, Bruker microCT) to evaluate the regeneration and osseointegration of new bone on different implants.
Histological and histomorphometric analysis
All harvested samples were processed for histological analysis by decalcifying with a rapid decalcifier solution (RDO Rapid Decalcifier, IL) for 7 days until the samples were softened. The samples were then embedded with paraffin and sectioned laterally and transversely. Hematoxylin and eosin (H&E) staining was carried out to evaluate newly formed bone density and height. For newly formed bone density analysis, each implant was analyzed from proximal, middle, and distal three transversal segments, and 0.5 mm height for each segment (bottom, middle, top). Three different regions of each segmental section were calculated, and three samples of each type of implant was considered. For all the other semi-quantitative evaluations, only central lateral segmental sections were considered for each implant.
Immunohistochemical characterization
The central lateral sections of each group sample were probed with collagen type I, bone sialoprotein (BSP), osteocalcin (OC), and angiogenic marker CD31. Briefly, the samples were retrieved with 1 × citrate buffer (Sigma-Aldrich, St. Louis, MO) for 30 min at 95–100 °C. Followed which, all sets of groups were incubated with primary anti-BSP (1:500), anti-collagen type I (1:250), anti-OC (1:100), and anti-CD31 (1:50) antibodies (Abcam, Cambridge, UK) separately in a humidity chamber at 4 °C for overnight, and were subsequently marked with anti-biotinylated binding IgG secondary antibodies for at room temperature for 30 min. All immunohistochemical images were taken from a Nikon TE-2000 microscope, and three regions of each section were captured for semi-quantitative analysis using FIJI ImageJ (NIH). For BSP, collagen type I, and OC staining, percentage of positively stained tissue area to the whole tissue area were utilized on osteogenesis analysis. For CD 31 staining, the amount of blood vessels that formed on tissues were calculated for angiogenesis analysis47,48. All the other chemicals which were used in the characterizations were purchased from Vector Laboratories (CA, US).
Statistical analysis
GraphPad Prism 6 (GraphPad, San Diego, CA) was used for conducting statistical analysis and plotting graphs. ANOVA with Tukey multiple comparison test was applied, and the results between groups are statistically significant when p < 0.05. All the experiments were performed in triplicates using one hBMSC donor.
