4.2.1. Preparation and Characterization of the Silica/Gelatin/β-TCP Sol-Gel Ink for Robocasting

A gelatin solution in 80 ◦C distilled water (100 mg mL−<sup>1</sup> ) was prepared under vigorous agitation for 10 min. After cooling at room temperature, functionalization of gelatin was promoted by reaction with GPTMS to give a molar ratio of 750 with respect to gelatin. The silica sol was obtained separately by mixing TEOS with a stoichiometric quantity of acidified water (HCl, 0.1 N) to obtain a transparent silica sol. Next, both gelatin-GPTMS and TEOS solutions were mixed and homogenized to produce silica/gelatin hybrid sol matrices, labeled as SGx, with x = 20, 40 and 60 wt% gelatin content. All chemical reactions were carried out at room conditions with mechanical stirring. The gelation time of SGx gels decreased from 60 to ~5 min with increasing gelatin content from 20 to 60 wt%, respectively. Therefore, given that the setting time and rheological properties are critical for the direct writing process, the SG40 hybrid was considered to exhibit the best value of gelation time desirable for successful printing (~30 min), while SG20 and SG60 with, respectively, longer or shorter gelation times were not used in the ink formulation. Next, three different microsized β-TCP high powder contents (40, 50, 60 wt%) were added to the selected SG40 hybrid sol, until three different solid suspensions were obtained, named as SG40TCP40, SG40TCP50 and SG40TCP60, which were placed in a planetary centrifugal mixer (ARE-250, Thinky Corp., Tokyo, Japan) for 5 min to obtain stable homogeneous solid suspensions. Next, the rheological characterization of the suspensions was performed at room temperature, and the best composition for the ink formulation was selected. The apparent viscosities of the three hybrid pastes prior to aging and gelation were measured with the Discovery HR10 rheometer (TA instruments, USA), using a cone-plate geometry and a gap of 50 µm, for shear rates ranging from 1 to 100 s−<sup>1</sup> at 25 ◦C. All measurements were conducted in triplicate.

#### 4.2.2. Fabrication of Silica/Gelatin/β-TCP Hybrid Scaffolds by Robocasting

The optimized ink (SG40/β-TCP 60 wt%, as detailed in Section 2.1) was introduced into the printing syringe and then placed on the three-axis motion stage of an A3200 robotic

deposition device (Aerotech/3D inks, Stillwater, OK, USA), which was controlled by a computer-aided direct-write program (Robocad 3.0, 3-D Inks, Stillwater, OK). Deposition of the ink was carried out by extrusion of filament rods layer-by-layer at room temperature with a speed of 20 mm s−<sup>1</sup> , using a conical nozzle with a tip diameter of 600 µm. The scaffolds were created from a CAD model using a control software (RoboCAD 4.1, 3D inks LLC, OK, USA) and consisted of a 10 <sup>×</sup> 10 mm<sup>2</sup> squared base, where the distance between filaments was 400 µm. The model comprises 5 layers, always rotated 90◦ from the previous one, with a distance between layers of 480 µm. Deposition was performed in a paraffin oil bath to prevent non-uniform drying during assembly, while the hybrid paste underwent chemical polymerization to produce solid composite filament rods. This way, a 3D scaffold was obtained, as described in Figure 13a. The resulting sample, labeled as SG40TCP60-RC, was dried in air at room temperature for 24 h and then at 100 ◦C for 48 h to evaporate organics. Given that the viscosity of the ink increased in time due to the sol-gel condensation and that gelation is the mechanism associated with the shape retention and solidification of the 3D-deposited structures, a 30-minute printing window interval, as found for the SG40 hybrid matrix, was considered long enough to perform the robocasting procedure. botic deposition device (Aerotech/3D inks, Stillwater, OK, USA), which was controlled by a computer-aided direct-write program (Robocad 3.0, 3-D Inks, Stillwater, OK). Deposition of the ink was carried out by extrusion of filament rods layer-by-layer at room temperature with a speed of 20 mm s−1, using a conical nozzle with a tip diameter of 600 μm. The scaffolds were created from a CAD model using a control software (RoboCAD 4.1, 3D inks LLC, OK, USA) and consisted of a 10 × 10 mm2 squared base, where the distance between filaments was 400 μm. The model comprises 5 layers, always rotated 90° from the previous one, with a distance between layers of 480 μm. Deposition was performed in a paraffin oil bath to prevent non-uniform drying during assembly, while the hybrid paste underwent chemical polymerization to produce solid composite filament rods. This way, a 3D scaffold was obtained, as described in Figure 13a. The resulting sample, labeled as SG40TCP60-RC, was dried in air at room temperature for 24 h and then at 100 °C for 48 h to evaporate organics. Given that the viscosity of the ink increased in time due to the solgel condensation and that gelation is the mechanism associated with the shape retention and solidification of the 3D-deposited structures, a 30-minute printing window interval, as found for the SG40 hybrid matrix, was considered long enough to perform the robocasting procedure.

The optimized ink (SG40/β-TCP 60 wt%, as detailed in Section 2.1) was introduced into the printing syringe and then placed on the three-axis motion stage of an A3200 ro-

*Gels* **2022**, *8*, 634 16 of 23

4.2.2. Fabrication of Silica/Gelatin/β-TCP Hybrid Scaffolds by Robocasting

**Figure 12.** Schematic showing the different processing methods used to prepare SG40TCP60 3D solgel-derived scaffolds by robocasting (RC) and laser micromachining (LM). The last-mentioned scaffold was fabricated by laser irradiation of the corresponding monolith composites, which in turn were obtained by solvent casting of SG40TCP60 ink, followed by evaporative drying, (y = 60 wt% represents the β-TCP content for the optimized ink composition according to rheological studies.) **Figure 12.** Schematic showing the different processing methods used to prepare SG40TCP60 3D sol-gel-derived scaffolds by robocasting (RC) and laser micromachining (LM). The last-mentioned scaffold was fabricated by laser irradiation of the corresponding monolith composites, which in turn were obtained by solvent casting of SG40TCP60 ink, followed by evaporative drying, (y = 60 wt% represents the β-TCP content for the optimized ink composition according to rheological studies.)

#### 4.2.3. Fabrication of Silica/Gelatin/β-TCP Hybrid Scaffolds by Laser Micromachining 4.2.3. Fabrication of Silica/Gelatin/β-TCP Hybrid Scaffolds by Laser Micromachining

The sol-gel slurry was allowed to polymerize at room temperature in hermetically closed plastic containers, where gelation took place in about 30 min. The resultant gel was aged for 1 week and then dried by evaporation at 50 °C for another 48 h to obtain the related SG40TCP60 monolith composite to be used as the solid target for laser irradiation. Next, laser micromachining of the composite surface was conducted in the lasing pro-The sol-gel slurry was allowed to polymerize at room temperature in hermetically closed plastic containers, where gelation took place in about 30 min. The resultant gel was aged for 1 week and then dried by evaporation at 50 ◦C for another 48 h to obtain the related SG40TCP60 monolith composite to be used as the solid target for laser irradiation. Next, laser micromachining of the composite surface was conducted in the lasing processing laboratory of the Condensed Matter Physics Department of the University of Cadiz by using a commercial high repetition rate pulsed laser model Carbide CB3-40W of light conversion (nominal pulse energy beam (Ep) of 38 µJ, changeable pulse width (τp) from 190 fs to 10 ps, an adjustable repetition rate from one single pulse to 1 MHz) equipped with a Carbide

harmonics module to generate ultrashort laser pulses with variable wavelength in the range between 1035 nm and 343 nm. This laser system was equipped with galvanometric mirrors, which allowed the movement of the laser radiation along the surface of the target; a mechanical Z-axis that allowed the position of the laser beam focus to be modified with respect to the surface of the sample. This laser device was controlled with a CAD-like LS-PRO software, which allows a precise control of the laser working parameters. In these laser micromachining, the laser beam was incident perpendicular to the longitudinal section of a 10.0 mm × 10.0 mm × 2.0 mm composite rectangular prism in order to generate a pattern of cylinders 350 µm in diameter, whose centers were separated by 1050 µm. To perform this, the laser working parameters were established at: 343 nm (UV) for the pulse laser wavelength, 38 µJ for the average laser pulse energy (0.76 W of beam power), 20 kHz for the pulse repetition rate, 220 fs for the pulse width and 10 mm·s −1 for the scanning speed. In order to drill through the whole thickness of the rectangular prism, the pattern was repeated 3 times in the same spot, and then, the Z-axis was lowered 150 µm. A geometrical model of the perforated monoliths, labeled as SG40TCP60-LM, is shown in Figure 13b. iable wavelength in the range between 1035 nm and 343 nm. This laser system was equipped with galvanometric mirrors, which allowed the movement of the laser radiation along the surface of the target; a mechanical Z-axis that allowed the position of the laser beam focus to be modified with respect to the surface of the sample. This laser device was controlled with a CAD-like LS-PRO software, which allows a precise control of the laser working parameters. In these laser micromachining, the laser beam was incident perpendicular to the longitudinal section of a 10.0 mm × 10.0 mm × 2.0 mm composite rectangular prism in order to generate a pattern of cylinders 350 μm in diameter, whose centers were separated by 1050 μm. To perform this, the laser working parameters were established at: 343 nm (UV) for the pulse laser wavelength, 38 μJ for the average laser pulse energy (0.76 W of beam power), 20 kHz for the pulse repetition rate, 220 fs for the pulse width and 10 mm·s−1 for the scanning speed. In order to drill through the whole thickness of the rectangular prism, the pattern was repeated 3 times in the same spot, and then, the Z-axis was lowered 150 μm. A geometrical model of the perforated monoliths, labeled as SG40TCP60- LM, is shown in Figure 13b.

cessing laboratory of the Condensed Matter Physics Department of the University of Cadiz by using a commercial high repetition rate pulsed laser model Carbide CB3-40W of light conversion (nominal pulse energy beam (Ep) of 38 μJ, changeable pulse width (τp) from 190 fs to 10 ps, an adjustable repetition rate from one single pulse to 1 MHz) equipped with a Carbide harmonics module to generate ultrashort laser pulses with var-

*Gels* **2022**, *8*, 634 17 of 23

**Figure 13.** Design section views of the scaffold shapes and its corresponding pore structure: (**a**) layer-by-layer of the interconnected macroporous pattern planned by using robocasting; (**b**) arrangement of the cylindrical holes for a macroporous structure generated by applying laser ablation in a rectangular prism monolith composite. All dimensions are presented in mm. **Figure 13.** Design section views of the scaffold shapes and its corresponding pore structure: (**a**) layerby-layer of the interconnected macroporous pattern planned by using robocasting; (**b**) arrangement of the cylindrical holes for a macroporous structure generated by applying laser ablation in a rectangular prism monolith composite. All dimensions are presented in mm.

#### *4.3. Characterization Techniques 4.3. Characterization Techniques*

Physical, textural and mechanical properties as well as biodegradation and biological responses were investigated for the SG40TCP60 composition optimized for extrusion. Physical, textural and mechanical properties as well as biodegradation and biological responses were investigated for the SG40TCP60 composition optimized for extrusion.

The bulk density (ρ*Bulk*) was obtained by measuring the dimensions with a sliding caliper and the mass with a microbalance (precision of ± 0.1 mg). In addition, the skeleton density (ρ*Skel*) of the SG40TCP60 monolith composite was measured using helium pycnometry, according to method described by Weinberger et al. [81], so that the porosity of the scaffolds () could be estimated from Equation (1). Values were obtained from the average of three replicates. The bulk density (*ρBulk*) was obtained by measuring the dimensions with a sliding caliper and the mass with a microbalance (precision of ±0.1 mg). In addition, the skeleton density (*ρSkel*) of the SG40TCP60 monolith composite was measured using helium pycnometry, according to method described by Weinberger et al. [81], so that the porosity of the scaffolds (*ε*) could be estimated from Equation (1). Values were obtained from the average of three replicates.

$$
\varepsilon = \left(1 - \frac{\rho\_{Bulk}}{\rho\_{Skel}}\right) \times 100 \tag{1}
$$

ௌ Specific surface area, pore volume and pore size were investigated using nitrogen physisorption experiments, considering BET and BJH standard models for the analysis. A Micromeritics ASAP2010 working at 77 K and equipped with a pressure transducer with Specific surface area, pore volume and pore size were investigated using nitrogen physisorption experiments, considering BET and BJH standard models for the analysis. A Micromeritics ASAP2010 working at 77 K and equipped with a pressure transducer with resolution of 10−<sup>4</sup> mm Hg was used for the analysis.

resolution of 10−4 mm Hg was used for the analysis. The mechanical behavior was characterized through uniaxial compression tests in samples obtained by robocasting (SG40TCP60-RC) using as-printed right-angled prisms with <sup>10</sup> <sup>×</sup> <sup>10</sup> <sup>×</sup> 5 mm<sup>3</sup> dimensions as test specimens, according to ASTM D7012 standard [82]. Compressive loads were applied in the direction perpendicular and parallel to the printing plane of the RC samples. The compressive strength and maximum strain were computed from the maximum stress and deformation before the fracture of the sample, and the Young's modulus was obtained from the slope at the beginning of the stress–strain curve. For comparison, the compressive behavior of the SG40TCP60 cylinder-shaped monoliths (18.00-millimeter height and 10-millimeter diameter) was also examined. All the tests were conducted in air with a Shimadzu AG-I Autograph 5 kN using a load cell of 500 N within a ±1% of the indicated test force at 1/250 load cell rating and a constant crosshead speed of 0.1 mm min−<sup>1</sup> . For each testing condition, a minimum of 3 samples were examined.

#### *4.4. Biodegradation*

The in vitro biodegradation study of the SG40TCP60 material was performed in phosphate buffer saline (PBS; pH 7.4), an effective physiological buffer with pH, osmolarity and ion concentrations similar to human plasma [83–85]. The weight loss of the samples was determined gravimetrically at room temperature, using a microbalance (precision of ±0.1 mg). The pH time-dependence of the PBS-incubated solution was also monitored using a pH meter probe (HACH sensIONTM + pH = 3, with pH resolution of 0.01). Basically, printed samples were cut with a sterile scalpel knife and then immersed in PBS (10 mg mL−<sup>1</sup> ) under physiological conditions at 37 ◦C. In the next stage, samples were removed from buffer media weekly and then weighed after wiping off any liquid on the surface. Likewise, the pH of the incubated solution was recorded. Last, samples were immersed again to continue with the degradation test in a totally refreshed PBS solution. This procedure was repeated over 9 weeks of incubation period, and the weight loss time-dependence was calculated according to Equation (2), as follows:

$$\text{Weight loss } (\%) = \frac{(W\_{\text{i}} - W\_{\text{l}})}{W\_{\text{i}}} \times 100\tag{2}$$

where *W<sup>i</sup>* and *W<sup>t</sup>* are the weights of the samples before and after the degradation experiment, respectively, determined at regular intervals of 1 week.

In addition, to provide a more detailed picture of the biodegradation process, the concentrations of accumulated of Si, Ca and P ions in the incubated PBS solution were also measured. The corresponding ion release profile analysis was performed by inductively coupled plasma mass spectrometry with an ICP-MS mass spectrometer, Series X2, Thermo Elemental. The ion concentrations were quantified by collecting 4-milliliter aliquots of the scaffold/PBS-incubated solution (1 mg mL−<sup>1</sup> ), at intervals of 12 h, 24 h, 48 h, 72 h, 120 h and 168 h after starting the experiment and without refreshing solution. The aliquot parts were filtered with a 0.45-millimeter Millipore membrane filter, subsequently placed in plastic vials to prevent any type of contamination and stored at 4 ◦C until the ICP-MS analysis. Weight loss, pH and Si, Ca and P ion-released concentrations were calculated using the average of all triplicates of the samples.

#### *4.5. Cell Culture*

HOB® cells were seeded, under sterile conditions, on the preselected scaffolds. Osteoblasts were detached when the optimal confluence was reached, then counted to optimal cell density and studied for cell viability in an automated Luna® cell counter, Invitrogen. All experiments were performed with cells under ten population doublings. In order to achieve optimal sterilization of the biomaterials prior to cell seeding, a clinically standardized autoclave (under European standard DIN EN ISO 13060 recommendations for class B autoclaves) was employed. Sterilized samples were placed on Mattek® glass bottom wells in a laminar flow chamber. A drop of 50 µL of cell suspension containing 15,000 HOB® cells/cm<sup>2</sup> was then added to each sample and then kept for 30 min under incubation at 37 ◦C and 5%. Afterwards, wells were filled with supplemented OGM® (final concentration of 0.1 mL/mL of fetal calf serum). Experiments were performed at 37 ◦C and 5% CO2. The growth medium was changed every two days. Experimental groups included: SG40TCP60 sample and HOB® cells grown on glass under the conditions before mentioned, used as the control.

#### *4.6. Live/Dead Cell Assay*

In order to evaluate the viability/cytotoxicity of HOB cells grown on SG40TCP60 sample and also in the controls, live /dead cell assay was performed as follows: after being incubated for 48 h, the cell/scaffold constructs were twice PBS-rinsed and then exposed to calcein-AM (0.5 µL/mL) in PBS to display the live cells and then to ethidium homodimer-1 (EthD-1) (2 µL/mL) in PBS to display dead cells, correspondingly. The samples were then observed in the fluorescence and Nomarski modes of a Leica DMLI LED inverted microscope.

#### *4.7. Cell Morphology and Spreading*

Cell morphology, alignment, distribution and spreading of osteoblasts were daily assessed after examination with the phase-contrast microscope and at the end of any experimental period. Furthermore, the Nomarski mode of both fluorescence and confocal microscopes was combined with the fluorescence mode to assess the simultaneous imaging of both the material and growing cells.

#### *4.8. Actin Cytoskeletal Organization*

Rhodamine-phalloidin and vinculin immunolabeling was performed after incubation for 48 h, 72 h and 7 days in order to assess cytoskeletal changes and focal adhesion development. Briefly, cells were washed with prewarmed PBS, keeping pH to 7.4 and fixing the constructs with paraformaldehyde (3.7%) at room temperature, and then permeabilized with 0.1% Triton x-100. After twice washing, preincubation with 1% bovine serum albumin in PBS for 20 min was performed, prior to cell immunolabeling with rhodamine phalloidin for 20 min. Then, samples were rinsed twice with prewarmed PBS prior to Vectashield ® mounting. At least five samples were seeded and analyzed per experiment and experimental group: SG40TCP60 sample and control. HOB® cells grown on glass under the conditions above-mentioned were used in the last case.

#### *4.9. Confocal Examination*

An Olympus confocal microscope was employed to assess the surface influence on the following parameters: cell density, cytoskeletal features and organization, focal adhesion distribution and changes in cell morphology. A total of 50 cells per sample were analyzed at least in each group, for a time interval not higher than 5 min to avoid photobleaching using a pinhole of 1 Airy unit.

#### *4.10. Image Analysis*

Sample images were collected as frames obtained at 40x magnification at a resolution of 1024 × 1024 Image J software (NIH, http://rsb.info.nih.gov/ij, accesed on 9 October 2020) was used for image processing. Shape variables analyzed were: area, roundness, circularity, perimeter and aspect ratio. A minimum of 40 regions of interest (ROIs) were measured under the following criteria: clear identification of nucleus, well-defined limits and absence of intersection with neighboring cells. All experiments were repeated in triplicates. Data were analyzed with SPSS and expressed as the mean ± standard deviation. After confirmation of normality and homoscedasticity, a one-way analysis of variance, Brown–Forsythe and Games-Howell tests were employed to assess the difference between the mean values. Statistical significance was defined as *p* < 0.05.

**Author Contributions:** Conceptualization, M.V.R.-P., F.J.M.-V., E.F., Ó.B.-M., M.S. and M.P.; methodology, M.V.R.-P., F.J.M.-V., R.F.-M., M.d.M.M.-D., E.F., Ó.B.-M., N.D.l.R.-F., M.S. and M.P.; formal analysis, M.V.R.-P., M.d.M.M.-D., N.D.l.R.-F., R.F.-M., M.S. and M.P.; investigation, M.V.R.-P., F.J.M.-V., E.F., R.F.-M., M.S., R.A., N.D.l.R.-F. and M.P.; resources, F.J.M.-V., Ó.B.-M., J.I.V.-P., R.A., M.S., N.D.l.R.- F. and M.P.; data curation, M.V.R.-P., E.F., R.F.-M., Ó.B.-M., N.D.l.R.-F., M.S. and M.P.; writing—original draft preparation, M.V.R.-P., M.S., N.D.l.R.-F. and M.P.; writing—review and editing, M.V.R.-P., M.S. and M.P.; supervision, Ó.B.-M., M.S., N.D.l.R.-F. and M.P.; project administration, M.S.; funding acquisition, M.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was 80% supported by Andalucía FEDER/ITI 2014-2020 Grant for PI 013/017 and Junta de Andalucía TEP115 and CTS 253 PAIDI research groups (Spain). The work has also been co-financed by the 2014–2020 ERDF Operational Program and by the Department of Economy, Knowledge, Business and University of the Junta de Andalucía. FEDER-UCA18\_106598. OBM thanks the Project for Scientific Equipment Acquisition: "Sistema Láser de Generación de Nanomateriales (NANO-GLAS): Fabricación y procesado de materiales nanoestructurados y síntesis directa de dispersiones coloidales de nanopartículas funcionalizadas", from the Ministerio de Ciencia, Investigación y Universidad EQC2018-004979.

**Acknowledgments:** Authors acknowledge the use of instrumentation as well as the technical advice provided by the GEMA-Uex research group from Universidad de Extremadura (UNEX) with robocasting equipment, as well as SCCYT (UCA) for SEM, ICP and EA divisions as well as SCBM at the University of Cadiz. The authors would also like to thank. J. Vilches-Troya, retired Professor of Histology and Pathology of the University of Cadiz, for his expert advice and supervision, and Enrique Gallero-Rebollo for his assistance in figure design. All individuals included in this section have consented the acknowledgment.

**Conflicts of Interest:** The authors declare no conflict of interest.
