The current bone autograft procedure for cleft palate repair presents several disadvantages such as limited availability, additional invasive surgery, and donor site morbidity. The present preliminary study evaluates the mineralization potential of electrospun polydioxanone:nano-hydroxyapatite : fibrinogen (PDO : nHA : Fg) blended scaffolds in different simulated body fluids (SBF). Scaffolds were fabricated by blending PDO : nHA : Fg in the following percent by weight ratios: 100 : 0 : 0, 50 : 25 : 25, 50 : 50 : 0, 50 : 0 : 50, 0 : 0 : 100, and 0 : 50 : 50. Samples were immersed in (conventional (c), revised (r), ionic (i), and modified (m)) SBF for 5 and 14 days to induce mineralization. Scaffolds were characterized before and after mineralization via scanning electron microscopy, Alizarin Red-based assay, and modified burnout test. The addition of Fg resulted in scaffolds with smaller fiber diameters. Fg containing scaffolds also induced sheet-like mineralization while individual fiber mineralization was noticed in its absence. Mineralized electrospun Fg scaffolds without PDO were not mechanically stable after 5 days in SBF, but had superior mineralization capabilities which produced a thick bone-like mineral (BLM) layer throughout the scaffolds. 50 : 50 : 0 scaffolds incubated in either r-SBF for 5 days or c-SBF for 14 days produced scaffolds with high mineral content and individual-mineralized fibers. These mineralized scaffolds were still porous and will be further optimized as an effective bone substitute in future studies.
Observed in approximately 1 in 700 live births, cleft lip (with or without cleft palate) is the most common congenital craniofacial birth defect in humans [
Bone is a natural composite of collagenous organic matrix reinforced by an inorganic mineral phase of hydroxyapatite (HA) whose structure is ultimately responsible for its functional properties. Other components of bone include calcium phosphates, water, proteins, and so forth [
A bone-like mineral (BLM) layer formed on the surface of biomaterials is an essential requirement for the material to bond to the living bone and enhance osteoconductivity. Simulated body fluid (SBF) has been used previously to induce mineral nucleation creating a BLM layer on the surface of materials. SBF has been widely used for biomimetic BLM coating on bioinert materials to directly mimic the process of mineralization in native bone and to predict the
Natural polymers attract special interest in tissue engineering since they are biocompatible, biodegradable, and natural substrates where cells can attach, proliferate, and function [
Electrospun scaffolds were fabricated with varying ratios of polydioxanone (PDO, Ethicon, Inc.), Fg (Fraction 1, Type 1-S from bovine plasma, Sigma-Aldrich, Co.), and HA nanopowder (particle size < 200 nm (BET), Sigma-Aldrich). PDO and Fg were dissolved in 1,1,1,3,3,3 hexafluoro-2-propanol (HFP, TCI America) at 100 mg/mL. nHA was added as a wt% of the polymer and sonicated for 10 minutes on pulse mode (on: 50 s, off: 10 s) at 38% maximum amplitude using a Cole-Palmer Ultrasonic Processor sonicator (model CPX 750). Sonicating was necessary to ensure proper dispersal of the mineral since nHA sedimented in HFP. To this sonicated solution, a known amount of polymer was added to attain the final concentration. It has been shown that by using this method of preparing polymer-nHA solutions, nHA is successfully incorporated within the composite scaffold [
Electrospinning parameters.
Composition (PDO : nHA : Fg) | Dispense rate (mL/hr) | Air gap distance (cm) | Voltage (kV) | Needle gauge |
---|---|---|---|---|
100 : 0 : 0 | 3.3 | 20 | 26 | 16 |
50 : 25 : 25 | 2 | 15 | 29 | 18 |
50 : 50 : 0 | 3.3 | 20 | 26 | 16 |
50 : 0 : 50 | 2 | 13 | 29 | 18 |
0 : 0 : 100 | 2 | 11 | 30 | 18 |
0 : 50 : 50 | 2 | 11 | 30 | 18 |
The ionic concentrations of the commonly used simulated body fluid, conventional SBF (c-SBF), are not exactly equal to those of blood plasma. Oyane et al. made revisions to c-SBF and created three new SBFs (revised (r), ionic (i), and modified (m)) that have ionic concentrations equal to, or closer to, those of blood plasma (Table
Ionic concentrations of human blood plasma. Modified from [
Ion | Concentration (mM) | ||||
---|---|---|---|---|---|
Total blood plasma | c-SBF | r-SBF | i-SBF | m-SBF | |
|
142.0 | 142.0 | 142.0 | 142.0 | 142.0 |
|
103.0 | 147.8 | 103.0 | 103.0 | 103.0 |
|
27.0 | 4.2 | 27.0 | 27.0 | 10.0 |
|
5.0 | 5.0 | 5.0 | 5.0 | 5.0 |
|
2.5 | 2.5 | 2.5 | 1.6 | 1.5 |
|
1.5 | 1.5 | 1.5 | 1.0 | 1.5 |
|
1.0 | 1.0 | 1.0 | 1.0 | 1.0 |
|
0.5 | 0.5 | 0.5 | 0.5 | 0.5 |
At the end of the experiment, scaffolds were removed and rinsed with DI water to wash off any minerals that were not bound to the scaffold. To visually inspect surface mineralization, one scaffold disc was dehydrated and used for scanning electron micrographs (SEM). For mineral quantification, three scaffolds were used for alizarin red S (ARS) staining and the remaining four scaffolds were analyzed using the burnout test to calculate the percent mineral composition of the scaffolds.
SEM was performed in order to evaluate the scaffold and fiber surface characteristics prior to and following mineralization. Samples of electrospun scaffolds for SEM were dehydrated, mounted on aluminum stubs, sputter coated in gold for 90 seconds (Electron Microscope Sciences model 550), and examined using a Zeiss EVO 50 XVP scanning electron microscope. Fiber diameters were calculated from the SEM images using UTHSCSA Image Tool 3.0 software.
ARS is a dye that selectively binds to calcium salts. ARS staining was used to quantify mineralization by modifying a published protocol [
The mineral content in the scaffolds was quantified by modifying a published burnout test protocol [
To determine burning times, the original nonmineralized scaffolds were burned at 500°C in a platinum crucible (Engelhard-Clal, item 201-20CC) until the scaffolds were completely disintegrated. Original electrospun scaffolds that did not contain nHA were burned until nothing remained in the crucible. Scaffolds containing nHA were burned until only the dry inorganic nHA powder remained in the crucible. By weighing the remaining nHA powder in the crucible after burning we were able to mathematically determine the efficiency of incorporating HA into electrospun scaffolds as well as determine how much HA is lost during the electrospinning process. The required burn times for each scaffold type were recorded (Table
Duration of burn-out test for each scaffold.
Time (hours) | Scaffold composition |
---|---|
1 | 100 : 0 : 0, 50 : 50 : 0 |
2 | 50 : 25 : 25, 0 : 0 : 100, 0 : 50 : 50 |
3 | 50 : 0 : 50 |
After mineralizing for 5 or 14 days, scaffolds were rinsed, air dried for 24 hours, and then placed in a crucible of a known weight. Scaffolds were weighed (
Control experiments were conducted in order to verify that the HA was not burned off though the burnout test. For the first control experiment, 0.2 grams of the original powdered nHA was placed in the crucible and burned for three hours. In the second experiment 0.25 grams of powder nHA was sonicated in 5 mL of HFP. After sonication the solution was left uncapped under a fume hood for 24 hours for the HFP to evaporate leaving only the dry sonicated nHA. This nHA was scraped off into the crucible and also burned for three hours. A burning duration of three hours was chosen because this was the maximum burn time that the scaffolds were be subjected to. After three hours in the muffle furnace, no loss or gain in weight to either of the HA samples should be observed. This confirms that any inorganic components left in the crucible after burning the original scaffolds are attributed to the incorporated electrospun nHA.
Statistical analysis was performed using JMP in 4 statistical software (SAS Institute) to determine significant differences between fiber diameters as well as ARS absorbance values. Analysis of the data was based on a Kruskal-Wallis one-way analysis of variance on ranks and a Tukey-Kramer pairwise multiple comparison procedure. The results are presented in mean ± SD.
SEM images of PDO : nHA : Fg original electrospun scaffolds are illustrated in Figure
Original electrospun PDO : nHA : Fg scaffolds.
Fiber diameters of PDO : nHA : Fg scaffolds.
Figures
PDO : nHA : Fg Scaffolds Incubated in c-SBF for 5 days (left 2 columns) and 14 days (right 2 columns), scale bar = 10
PDO : nHA : Fg scaffolds incubated in r-SBF for 5 days (left 2 columns) and 14 days (right 2 columns), scale bar = 10
PDO : nHA : Fg scaffolds incubated in i-SBF for 5 days (left 2 columns) and 14 days (right 2 columns), scale bar = 10
PDO : nHA : Fg scaffolds incubated in m-SBF for 5 days (left 2 columns) and 14 days (right 2 columns), scale bar = 10
Fg scaffolds without PDO (0 : 0 : 100 and 0 : 50 : 50) were not analyzed for mineral content via ARS because these scaffolds lost mechanical integrity after 5 days incubation in SBFs and could not be salvaged for staining. As expected, original electrospun scaffolds containing nHA had higher ARS absorbance values than mineralized scaffolds: 100 : 0 : 0 (0.0426 ± 0.0004), 50 : 0 : 50 (0.0547 ± 0.0012), 50 : 25 : 25 (0.1889 ± 0.0122), and 50 : 50 : 0 (0.2304 ± 0.0397). When scaffolds are mineralized it is possible for different apatites to form (carbonated apatite, Ca-P, HA, etc.). It is important to note that the original nonmineralized scaffolds either contained no nHA or pure nHA. ARS may have a different binding affinity to these various apatite structures. The environment in which the scaffolds were incubated is also an important consideration, specifically the 5% CO2. An increase in the CO2 concentration of the SBF will lower the pH of the mineralizing solution. HA becomes more soluble at a lower pH [
It is evident from SEM images that each scaffold nucleates apatites differently. Scaffolds without nHA most likely attract Ca-P apatites during primary nucleation since these are the beginning steps to forming bone. Scaffolds containing nHA may attract different apatites to their surface since nHA is a higher form of Ca-P. Considering the above possibilities, statistics were performed to determine differences of SBF treatment on each scaffold type.
The absorbance values of the mineralized scaffolds and the effect of different SBFs on each scaffold composition is displayed in Figure
ARS data and statistics for mineralized scaffolds. *Denotes statistical significance (
The same statistical analysis of ARS data was performed on scaffolds incubated in SBF for 14 days. Absorbance readings for 100 : 0 : 0 scaffolds in different SBFs were not significantly different from each other. An average-low value of 0.051 suggests that even after 14 days 100 : 0 : 0 scaffolds induced minimal mineralization. For 50 : 25 : 25 scaffolds m-SBF had the highest value (0.158) and was only significantly different (
The same statistical analysis was used to compare scaffolds compositions for 5 and 14 days incubation in SBF. 100 : 0 : 0 scaffolds showed statistical difference between day 5 and 14. For 50 : 25 : 25 scaffolds, 5 days in m-SBF was only significantly different (
For both 5 and 14 days, higher absorbance readings were recorded for 50 : 50 : 0 scaffolds showing a visible pattern of increased absorbance values with increased nHA composition. The ARS data suggests that 5 days incubation of 50 : 50 : 0 in SBFs is sufficient to induce mineralization while the other scaffold compositions showed no noticeable difference between 5 and 14 days.
As previously mentioned, Fg (0 : 0 : 100) and Fg-nHA (0 : 50 : 50) scaffolds degraded after 5 days incubation in SBFs and were not salvaged for mineral quantification via burnout test. The remaining scaffolds were intact and analyzed.
Figure
Mineralized scaffolds before and after burnout test. 100 : 0 : 0 scaffolds incubated for 5 days in m-SBF (a) before and (b) after burn. 50 : 50 : 0 scaffolds incubated for 5 days in i-SBF (c) before and (d) after burnout test.
Percent mineral composition of the scaffolds was calculated by weighing scaffolds before and after burning. All PDO : nHA : Fg scaffold compositions were analyzed for mineral content before and after incubation in different SBFs for 5 and 14 days under static conditions (Figure
Percent mineral composition of original and mineralized PDO : nHA : Fg scaffolds in different SBFs for 5 and 14 days.
The percent mineral composition of the original electrospun scaffolds prior to mineralization served as the control. The burnout test also proved as a method to measure the effectiveness of incorporating nHA within electrospun scaffolds by determining how much nHA was actually incorporated versus how much was lost during the electrospinning process. Polymer solutions that were prepared without nHA (100 : 0 : 0 and 50 : 0 : 50) yielded a value of 0% mineral. This control verifies that all the PDO and Fg fibers are completely burned. Solutions containing 25% (50 : 25 : 25) and 50% nHA (50 : 50 : 0) yielded values of 16.3% and 41.7%, respectively. This suggests that 34.8% and 16.6% nHA was lost, respectively, during the electrospinning process.
After 5 days incubation in different SBFs all PDO (100 : 0 : 0) scaffolds mineralized to some degree compared to original scaffold mineral composition of 0%. PDO scaffolds incubated in c-, i-, and m-SBFs each were comprised of approximately 3% mineral. Scaffolds incubated in r-SBF had a mineral composition of 9.6%. All 50 : 25 : 25 scaffolds increased in mineral content when compared to the original mineral composition (16%). Scaffolds in c-, r-, i-, and m-SBF contained 22.8%, 20.4%, 24.7%, and 18.4%, respectively. 50 : 50 : 0 scaffolds incubated in c-, i-, and m-SBFs resulted in a slightly lower mineral composition (38%) when compared to the original (41.7%). The highest percent mineral content was observed in 50 : 50 : 0 scaffolds in r-SBF (56.2%). 50 : 0 : 50 scaffolds incubated in all SBFs increased in mineral content from 0% to about 10%.
Overall the largest percent increase after 5 days was noticed in 50 : 50 : 0 scaffolds in r-SBF. These scaffolds in r-SBF also showed the highest percent mineral composition. By comparing scaffold compositions, it is evident that 50 : 50 : 0 scaffolds contained the highest mineral content before and after mineralization. This was expected since more nHA was incorporated within these scaffolds during the electrospinning process.
For each scaffold composition, incubation in c-SBF for 14 days resulted in the highest percent mineral content and highest mineral increase due to SBF treatment, suggesting that c-SBF induces the most mineralization independent of scaffold composition. After 14 days, 100 : 0 : 0 scaffolds in c-SBF increased in mineral content from 0% to 18.6% while all other SBFs only resulted in an increase from 0% to 3%. All mineralized 50 : 25 : 25 scaffolds increased in mineral content from 16.3% to 39.8% (c-SBF), 21.1% (r-SBF), 25.7% (i-SBF), and 33.3% (m-SBF). 50 : 50 : 0 scaffolds incubated in i- and m-SBF had no effect on mineral content (41%). Incubation in c- and i-SBF resulted in 57.1% and 29.1% mineral composition, respectively. The difference between c- and i-SBF suggest that 50 : 50 : 0 scaffolds mineralize differently depending on the SBF composition. After 14 days, 50 : 0 : 50 scaffolds increased in mineral content from 0% to various values depending on SBF type. 50 : 0 : 50 scaffolds incubated in c-, r-, i-, and m-SBF resulted in 18.4%, 9.6%, 10%, and 3.7% mineral content, respectively. The difference between c- and m-SBF suggest that 50 : 0 : 50 scaffolds mineralize differently depending on the SBF type. Once again 50 : 50 : 0 scaffolds contained higher percentages of mineral composition before and after mineralization since nHA was originally incorporated within the scaffolds. However, the highest overall mineral increase due to SBF treatment was in 50 : 25 : 25 scaffolds incubated in c-SBF for 14 days. Percent mineral content increased from 16.3% to 39.8% suggesting that 50 : 25 : 25 scaffolds in c-SBF for 14 days induce the most mineralization.
Kim et al. showed that formed apatites can be different in structure and composition depending on the ionic concentration of SBF used in their formation [
Overall visual progression of mineralization is observed when increasing incubation period from 5 to 14 days. PDO scaffolds without Fg (100 : 0 : 0 and 50 : 50 : 0) generally mineralized along individual fibers. 100 : 0 : 0 scaffolds have nucleation sites that are more distant from each other since it is a purely synthetic polymer. Depending on the SBF, this could cause continuous mineral growth at sparse sites where calcium phosphate CaP grows on itself forming a bead-like structure at each site. Scaffolds containing nHA (50 : 50 : 0) have more nucleation sites and beads of mineral do not form. HA containing scaffolds induce a more uniform dense individual fiber mineralization. PDO and Fg scaffolds (50 : 25 : 25 and 50 : 0 : 50) showed thin sheet-like mineral deposition throughout. Fg scaffolds not containing PDO (0 : 0 : 100 and 0 : 50 : 50) were entirely mineralized even after 5 days suggesting 5 days immersion in 1x SBF is sufficient to grow a thick BLM layer on the scaffolds. Since Fg has the capability to bind to a variety of molecules, we would expect to see this high degree of mineralization with Fg containing scaffolds.
The addition of Fg did not have an effect on the ARS results when analyzing mineralized scaffolds. Considering the concerns regarding different apatite formation and pH effects, ARS may not be the most accurate method of quantifying mineralization when Fg is incorporated in the scaffold which lead to the preliminary burnout test for mineral quantification. This burnout method will be independent of composition and will be the most accurate for measuring mineralization as long as the sample size is appropriate. If Fg is not present, the ARS method is a more streamlined assay for determining mineral content of the electrospun scaffolds.
From the preliminary burnout test data, some values increased while some decreased from 5 to 14 days incubation in SBF. The increases of mineral content can be attributed to the deposition of minerals on the electrospun scaffolds. The decreases may result from the solubility of nHA at a lower pH as previously described. Scaffolds without nHA (100 : 0 : 0 and 50 : 0 : 50) never surpassed 20% mineral content while 50 : 25 : 25 and 50 : 50 : 0 scaffolds reached up to 40% and 57%, respectively. 50 : 50 : 0 scaffolds in r-SBF for 5 days and 50 : 50 : 0 scaffolds in c-SBF for 14 days contained similar percent mineral compositions (~57%). Since 50 : 50 : 0 scaffolds in r-SBF for 5 days reached a high mineral content just after 5 days, we can assume that these scaffolds will bond to the host bone faster than the same scaffolds in c-SBF for 14 days.
ARS and burnout test data did not support each regarding which SBF type has the highest potential to induce mineralization. There are many variables to consider when quantifying mineralization via ARS which make it difficult to compare the different scaffold compositions. The burnout test is a more reliable method to quantify the mineral content of scaffolds.
This study has demonstrated an easy, cost effective approach using synthetic polymers, natural polymers, and inorganic apatites to produce a nanofibrous scaffold for bone tissue engineering applications. By combining the electrospinning of natural and synthetic materials (PDO, Fg, and nHA) and mineralization (SBF) methods, scaffolds with mineralized nanofibers were produced. Also, a reliable method for quantifying mineral content of 3D porous scaffolds before and after mineralization was developed. The degree and type of mineralization was dependent on the scaffold composition, type of SBF, and duration of incubation.
The addition of Fg resulted in smaller fiber scaffolds with thin sheet-like deposition of minerals unlike individual fiber mineralization seen in PDO and PDO-nHA scaffolds. Mineralized electrospun Fg scaffolds without PDO were not mechanically stable after 5 days, but had superior mineralization capabilities which produced a thick BLM layer throughout the scaffolds. Mineral quantification revealed that overall, 50 : 50 : 0 scaffolds had the highest mineral content.
This preliminary study focused on developing a mineralized porous nanofibrous scaffold intended for cleft palate repair. Results show that Fg containing scaffolds have superior mineralization potential but tend to mineralize as sheets decreasing the porosity of the scaffold. 50 : 50 : 0 scaffolds incubated in either r-SBF for 5 days or c-SBF for 14 days produce scaffolds with high mineral content and individual-mineralized fibers. These mineralized scaffolds were still porous and can potentially serve as effective substrates to induce three-dimensional bone formation. Most importantly, this study demonstrated the high mineralization potential of electrospun Fg scaffolds, which in future work will be studied in more detail to further characterize its mineralization capabilities (e.g., FTIR and X-ray diffraction) within shorter time periods.