Biomimetic Strontium Substituted Calcium Phosphate Coating for Bone Regeneration

Cellulose acetate (CA)/strontium phosphate (SrP) hybrid coating has been proposed as an effective strategy to build up novel bone-like structures for bone healing since CA is soluble in most organic solvents. Strontium (Sr2+) has been reported as a potential agent to treat degenerative bone diseases due to its osteopromotive and antibacterial effects. Herein, bioactive hybrid composite SrP-based coatings (CASrP) were successfully produced for the first time. CASrP was synthesized via a modified biomimetic method (for 7—CA7dSrP, and 14 days—CA14dSrP), in which the metal ion Sr2+ was used in place of Ca2+ in the simulated body fluid. Energy-dispersive X-ray (EDX) and Fourier transform infrared spectroscopy (FTIR) analysis confirmed the SrP incorporation chemically in the CASrP samples. Atomic absorption spectroscopy (AAS) supported EDX data, showing Sr2+ adsorption into CA, and its significant increase with the augmentation of time of treatment (ca. 92%—CA7dSrP and 96%—CA14dSrP). An increment in coating porosity and the formation of SrP crystals were evidenced by scanning electron microscopy (SEM) images. X-ray diffraction (XRD) evidenced a greater crystallinity than CA membranes and a destabilization of CA14dSrP structure compared to CA7dSrP. The composites were extremely biocompatible for fibroblast and osteoblast cells. Cell viability (%) was higher either for CA7dSrP (48 h: ca. 92% and 115%) and CA14dSrP (48 h: ca. 88% and 107%) compared to CA (48 h: ca. 70% and 51%) due to SrP formation and Sr2+ presence in its optimal dose in the culture media (4.6–9 mg·L−1). In conclusion, the findings elucidated here evidence the remarkable potential of CA7dSrP and CA14dSrP as bioactive coatings on the development of implant devices for inducing bone regeneration.


Introduction
Bone fractures have been treated mainly by bone grafting; however, their effectiveness remains not well understood [1]. Bone tissue engineering is a promising approach to develop bioceramic implants for bone healing [2]. Bone is mainly composed of HA nanocrystals along collagenous fibers comprising around 90%, while other biomolecules (non-collagenous proteins, proteoglycans, and lipids), 10% [3,4]. Based on this hybrid composition, materials formed by inorganic and organic phases have been synthesized, like scaffolds, to mimic the natural bone structure [5]. These scaffolds must attend to biological functions by the successful design for guided bone regeneration (GBR) [6]. The hybrid composites (two phases) exhibit superior mechanical and biological properties compared to the single phases. The organic phase increases flexibility, biodegradability, biocompatibility, and reduces compaction, while the inorganic, in addition to the non-toxicity, favors For CASrP production, initially, a 2% (w·v −1 ) CA solution was prepared by dissolving CA powder in 100 mL acetone. Then, to facilitate the dissolution and homogenate completely, it was kept under the agitation of 500 rpm for 1 h (Drehzahl electronic, IKAMAG REO, Staufen, Germany) and sonicated for 30 min in an ultrasonic bath (Unique, USC 1450, Indaiatuba, SP, Brazil). Next, the final solution was verted in Petri dishes to allow acetone evaporation at room temperature for about 30 min, followed by rinsing with distilled water to interrupt the evaporation process. Finally, the membrane was washed with distilled water and stored. Then, to incorporate the Sr 2+ substituted hydroxyapatite, the biomimetic mineralized CASrP were fabricated as follows: (1) m-SBF solution (an equal mixture of solution A: 6.213 g·L −1 NaCl, 5.948 g·L −1 NaKHCO 3 , 0.450 g·L −1 KCl, 0.462 g·L −1 K 2 HPO 4 ·3H 2 O, 0.622 g·L −1 MgCl 2 ·6H 2 O, and 0.144 g·L −1 Na 2 SO 4 ; with solution B: 6.213 g·L −1 NaCl and 0.584 g·L −1 SrCl 2 ) was prepared based on a modified Bohner and Lemaître method [54], and (2) after that, the membranes of CA were immersed in m-SBF solution under stirring at pH 7.4 and temperature of 36.5 • C for 7 and 14 days using a Dubnoff shaking water bath (Quimis, Q226M, Diadema, SP, Brazil). Then, the CA7dSrP and CA14dSrP membranes were obtained, washed with distilled water, and freeze-dried.

Scanning Electron Microscopy (SEM)
The samples were covered with a thin layer of gold (10 nm) using a sputter coater (K650, Emitech, Lorraine, France) and observed using a scanning electron microscope DSM 940 model (accelerating voltage 15 kV, Zeiss, Jena, Germany).

Fourier Transform Infrared Spectroscopy (FTIR)
FTIR analyses were performed using the potassium bromide (KBR) method in a spectrophotometer (Perkin Elmer, Waltham, MA, USA) with a resolution of 4 cm −1 and 48 scans, and a reading range of 4000-400 cm −1 .

Bioactivity
To evaluate the potential of the CA to favor apatite formation during the synthesis of CASrP, the liquid absorbing capacity of CA was analyzed in SBF and m-SBF. The CA samples were maintained in the media during a fixed time interval (0, 2, 4, 6, 8, 10, 20, 30, and 40 min) at 25 • C. The excess of liquid was removed with a filter paper (Quanty, 8 µm) Coatings 2021, 11, 908 4 of 18 and the swollen material was weighed at the established times. The measurements were performed in triplicate. The swelling degree was calculated according to Equation (1): where D s is the ratio of liquid-absorbing capacity; W s is the wet weight sample; and W d is the dry weight sample.
In addition, to verify whether the biomimetic methodology employed induced apatite nucleation, energy-dispersive x-ray, operating at higher electron voltage (eV), provided elemental mapping to determine the presence of Sr and P elements in CA7dSrP and CA14dSrP.

In Vitro Degradability Test
The CA, CA7dSrP, and CA14dSrP samples were cut into cubes of 10 mm × 10 mm and weighed, followed by immersion in a 50 mL centrifuge tube containing 10 mL of phosphate-buffered saline (PBS), and incubated in a bath at 37 • C for 60 days under an agitated condition. After each incubation period, the sample was rinsed with distilled water and freeze-dried. The dry weights were measured. The percentual of degradation was calculated following Equation (2): where W i corresponds to the initial weight and W f refers to the final weight.

Sr 2+ Release
Atomic absorption spectroscopy (AAS) was used to determine the Sr 2+ concentration in the culture media after exposure to the medium alone in the indirect method according to the International Organization for Standardization (ISO 10993-12). The amount of Sr released from the samples was quantified using a calibration curve for standards of known concentrations of Sr (0, 0.5, 1, 3, 5, 7, 10, and 20 mg·L −1 ). The AAS measurements were performed in duplicate for each system. The cell culture medium was used as the control sample.

Cell Cytotoxicity
The cytotoxicity of the materials was evaluated through an indirect method, in which cell viability is measured after exposure to the sample extracts. To prepare the extracts, initially, the samples were cut into cubes 10 mm × 10 mm and sterilized by autoclaving at 121 • C for 20 min. Then, samples of each material were aseptically placed in a 24-well plate with 1 mL of the supplemented DMEM or α-MEM culture medium supplemented with 10% FBS and 1% penicillin-streptomycin and incubated at 37 • C for 24 h according to ISO 10993-5:2009. After the incubation period, the extracts were collected for the assay.
L-929 cells (mouse fibroblasts) were seeded in supplemented DMEM medium and MC3T3-E1 (mouse osteoblasts) in supplemented α-MEM culture medium at a density of 6 × 10 3 cells/well into the 96-well polystyrene plate, followed by incubation at 37 • C (5% CO 2 and 95% of humidity) for 24 h. After that, the culture medium was removed from the wells and 100 µL of the sample extract was added to each well and the plate was again incubated at 37 • C for 24 and 48 h. After this period, the extract was removed from the wells and 120 µL of supplemented culture medium containing the resazurin reagent (final concentration of 25 mg·L −1 ) was added. The plate was incubated for 4 h under standard culture conditions, then 100 µL of the medium containing the metabolized resazurin was transferred to a new 96-well plate and the absorbance was measured in a microplate reader (SpectraMax i3x, Molecular Device, Sunnyvale, CA, USA) in fluorescence mode (λ excitation = 560 nm e λ emission = 590 nm). The positive control was performed by exposing the cells to dimethylsulfoxide at 40% (diluted in the corresponding supplemented culture medium), highly toxic to the cells. The negative control group was performed by exposing the cells only to the corresponding supplemented culture medium. The viability of the cells of the negative control group was adjusted to 100% for the calculation of mean values and standard deviation (n = 5). The cell viability was calculated following Equation (3): where V c represents the percentual of living cells; F S corresponds to the intensity of the fluorescence of cultivated cells exposed to the extract of the sample; and F C refers to the intensity of the fluorescence of cultivated cells in the presence of the culture media.

Results and Discussion
Several approaches have been performed to improve scaffold properties for bone repair. A porous structure is essential to mass transfer such as oxygen, nutrients, growth factors, waste products, and bone cell ingrowth. To simulate the extracellular matrix of the natural bone and support the bone formation, the CA polymer was chosen for this work due to its superior degradation capacity, natural origin, and non-toxicity compared to other cellulose derivatives. CA membranes were synthesized using phase inversion by immersion precipitation to obtain porous 3D structures [55] easily.
Initially, CA was wholly dissolved in acetone and placed in a Petri dish so that the volatile solvent started to evaporate, forming a membrane and detaching from the plate. After 30 min, the polymer was rinsed with distilled water and the exchange between the solvent and non-solvent caused the precipitation and formation of pores during membrane synthesis. Since drying and precipitation are processes not entirely controlled, the inversion method produces heterogeneous membranes. Figure 1(a1,a2) shows that the CA membrane exhibited a smooth, porous, and non-uniform arrangement [56].
To mimic the inorganic phase of mineralized bone, SrP was incorporated through the biomimetic method, and to improve osteogenesis properties, Sr 2+ was used in place of Ca 2+ . In Figure 1(b1,b2), after immersion in the m-SBF solution for seven days to nucleate SrP into CA (CA7dSrP), the membrane exhibited a more significant number of pores and a decrease in pore size compared to CA. The increased permeability due to higher porosity contributed to the delivery of Sr 2+ metal ions, favoring bone formation. In addition, different patterns of crystal growth were evidenced in the CA-based scaffolds.
The hybrid material CA14dSrP displayed a superior Sr 2+ amount trapped into the matrix, inducing a homogeneous interaction between the sites of this metal ion and phosphate in the media during the treatment with m-SBF. Increased nucleation sites were imparted on forming SrP crystals (see Figure 1(c1,c2)) [33,42,43]. Spherical-like SrHA particles can be seen covering the CA membrane and the pores in the layer considered for SEM appeared to have been filled by SrP. As a result, the pores were not well determined.

XRD
X-ray diffractograms of CA, CA7dSrP, and CA14dSrP are shown in Figure 2. Th results display differences in the crystallinity of CA after the incorporation of SrP. Th semi-crystalline structure was verified by the broader characteristic peak of CA around 2θ = 20°. Since the measurement was performed in a range from 10° to 85°, the peak at 2θ = 9° observed by [57] was not evident. However, an elevation may be seen at the beginning of the graph in Figure 2a. CA7dSrP, on the other hand, besides the peak of CA at 20° wa less intense; peaks corresponding to 2θ = 29.5°, 32.3°, 37.2°, 42.5°, and 53.3° are also pre sented in Figure 2b. These peaks were observed by [36], referring to SrP. Like CA7dSrP XRD patterns were obtained for CA14dSrP; however, at 29.5°, 37.2°, and 42.5° less intens peaks were shown as well as an increase in the intensity of the 53° peak, as can be seen in Figure 2c. Other peaks appeared at about 66.5° and 78°. Even though the rise of new peak could indicate an increase in crystallinity, the intensity of some of them was reduced in CA14dSrP (see Figure 2c). A longer time in contact with the solution of m-SBF containing Sr 2+ may have improved the absorption of this metal ion by the CA matrix because of th large radius of this metal ion (0.118 nm) [41]. Previously, it was shown that with the aug mentation of Sr 2+ content, broader peaks were observed in XRD. Likewise, in the bacteria cellulose structure, Luz et al. mentioned that a destabilization and enhancement of its sol ubility [36] occurred following [58]. Consequently, after the formation of SrP into CA, th presence of new peaks resulted in an improved structural organization (as shown previ ously in Figure 1(c1,c2)) and thus crystallinity. A superior time of CA in contact with th m-SBF solution contributed to the SrP nucleation and growth, and the increased amoun of Sr 2+ adsorbed into the membrane.

XRD
X-ray diffractograms of CA, CA7dSrP, and CA14dSrP are shown in Figure 2. The results display differences in the crystallinity of CA after the incorporation of SrP. The semi-crystalline structure was verified by the broader characteristic peak of CA around 2θ = 20 • . Since the measurement was performed in a range from 10 • to 85 • , the peak at 2θ = 9 • observed by [57] was not evident. However, an elevation may be seen at the beginning of the graph in Figure 2a. CA7dSrP, on the other hand, besides the peak of CA at 20 • was less intense; peaks corresponding to 2θ = 29.5 • , 32.3 • , 37.2 • , 42.5 • , and 53.3 • are also presented in Figure 2b. These peaks were observed by [36], referring to SrP. Like CA7dSrP, XRD patterns were obtained for CA14dSrP; however, at 29.5 • , 37.2 • , and 42.5 • less intense peaks were shown as well as an increase in the intensity of the 53 • peak, as can be seen in Figure 2c. Other peaks appeared at about 66.5 • and 78 • . Even though the rise of new peaks could indicate an increase in crystallinity, the intensity of some of them was reduced in CA14dSrP (see Figure 2c). A longer time in contact with the solution of m-SBF containing Sr 2+ may have improved the absorption of this metal ion by the CA matrix because of the large radius of this metal ion (0.118 nm) [41]. Previously, it was shown that with the augmentation of Sr 2+ content, broader peaks were observed in XRD. Likewise, in the bacterial cellulose structure, Luz et al. mentioned that a destabilization and enhancement of its solubility [36] occurred following [58]. Consequently, after the formation of SrP into CA, the presence of new peaks resulted in an improved structural organization (as shown previously in Figure 1(c1,c2)) and thus crystallinity. A superior time of CA in contact with the m-SBF solution contributed to the SrP nucleation and growth, and the increased amount of Sr 2+ adsorbed into the membrane.  Figure 3 depicts the vibrational spectra of CA, CA7dSrP, and CA14dSrP, indicating the interaction of SrP with CA upon its incorporation by a modified biomimetic methodology. The vibration band between 3050 and 3700 cm −1 was present in all the samples referring to O-H stretching from unacetylated groups of cellulose, however, there was a small band for CA membranes [13]. This is possibly because water molecules tend to have weaker interactions with the solvent than with the water itself [59]. Additionally, two bands at 2975 and 2850 cm −1 related to symmetrical CH and asymmetrical CH2, a band at 750 cm −1 for C=O stretching, and two bands at 1239 and 1050 cm −1 due to C-O are depicted in Figure 3a [55,59]. In Figure 3b,c, the CA7dSrP and CA14dSrP samples, besides the characteristic spectrum of CA, showed the 863 cm −1 band referring to P-O(H) stretching [36,60]. CO3 2-gives rise to weaker peaks around 600 cm −1 and more intense ones at 1568 and 1434 cm −1 [60,61]. The previous 2850 cm −1 band corresponding to the C-H stretch, 1374 cm −1 to the C-H methyl rock. The increase near to 3600 cm −1 means an augmentation of the strength of hydrogen bonding (adsorbed water) compared to CA only, being related to the HA spectrum [59]. FTIR measurements affirmed the band assignments and additional functional group bands observed. The results indicate the presence of carbonated HA and SrHA. Despite a long period of contact between the CA interface and solution of m-SBF in CA14dSrP, its spectrum coincided mostly with that of CA7dSrP.  Figure 3 depicts the vibrational spectra of CA, CA7dSrP, and CA14dSrP, indicating the interaction of SrP with CA upon its incorporation by a modified biomimetic methodology. The vibration band between 3050 and 3700 cm −1 was present in all the samples referring to O-H stretching from unacetylated groups of cellulose, however, there was a small band for CA membranes [13]. This is possibly because water molecules tend to have weaker interactions with the solvent than with the water itself [59]. Additionally, two bands at 2975 and 2850 cm −1 related to symmetrical CH and asymmetrical CH 2 , a band at 750 cm −1 for C=O stretching, and two bands at 1239 and 1050 cm −1 due to C-O are depicted in Figure 3a [55,59]. In Figure 3b,c, the CA7dSrP and CA14dSrP samples, besides the characteristic spectrum of CA, showed the 863 cm −1 band referring to P-O(H) stretching [36,60]. CO 3 2− gives rise to weaker peaks around 600 cm −1 and more intense ones at 1568 and 1434 cm −1 [60,61]. The previous 2850 cm −1 band corresponding to the C-H stretch, 1374 cm −1 to the C-H methyl rock. The increase near to 3600 cm −1 means an augmentation of the strength of hydrogen bonding (adsorbed water) compared to CA only, being related to the HA spectrum [59]. FTIR measurements affirmed the band assignments and additional functional group bands observed. The results indicate the presence of carbonated HA and SrHA. Despite a long period of contact between the CA interface and solution of m-SBF in CA14dSrP, its spectrum coincided mostly with that of CA7dSrP.

Bioactivity
The liquid-absorbing ability (in %) for the CA was studied using SBF and m solvents to investigate the interaction of the media with the membrane before an submitting the membrane to a solution with a higher ion concentration. Since in thi SBF modified with SrCl2 instead of CaCl2 was used for synthesizing SrP, SBF and were utilized for understanding the absorbing behavior. The systems reached equi after about 2 min in SBF and m-SBF since it no significant difference overtime f system was observed according to the Kruskal-Wallis test. The CA in m-SBF show highest absorption capacity with an average ratio of around 3.33, in contrast to SB which is considered statistically significant. The evaluation of liquid uptake gav mation on the effect of ionic strength.
The membrane in SBF containing Sr 2+ showed a dramatic improvement of liq sorbing ability if this ability is compared when using SBF, as shown in Figure 4. sorption was enhanced due to the greater concentration of polar constituents metal ions, which led to an increase in osmotic pressure between the membrane swelling solution [62]. Their presence and the existence of hydrophilic groups (-CA enable that, in addition to hydrogen bonds, electrostatic and van der Waals tions are also created in the liquid-solid interface. As a result, regarding the ele properties of the cations in m-SBF, the interplay between these forces can be favor Ca 2+ , for example, is usually present in SBF composition, which is used for produc in the biomimetic method; however, in m-SBF, it was replaced by Sr 2+ . In our stu stead of Ca 2+ , Sr 2+ was used because of its interesting properties for bone repair an similarity in ionic radii and valence (2+), and common biochemical route that wou tribute to keeping the adsorbent capability and biological function. The m-SBF s was used to induce SrP nucleation and growth into the CA membrane after con seven days and 14 days. Pan et al. showed that as the Sr 2+ content was augmented

Bioactivity
The liquid-absorbing ability (in %) for the CA was studied using SBF and m-SBF as solvents to investigate the interaction of the media with the membrane before and after submitting the membrane to a solution with a higher ion concentration. Since in this work, SBF modified with SrCl 2 instead of CaCl 2 was used for synthesizing SrP, SBF and m-SBF were utilized for understanding the absorbing behavior. The systems reached equilibrium after about 2 min in SBF and m-SBF since it no significant difference overtime for each system was observed according to the Kruskal-Wallis test. The CA in m-SBF showed the highest absorption capacity with an average ratio of around 3.33, in contrast to SBF, 1.83, which is considered statistically significant. The evaluation of liquid uptake gave information on the effect of ionic strength.
The membrane in SBF containing Sr 2+ showed a dramatic improvement of liquidabsorbing ability if this ability is compared when using SBF, as shown in Figure 4. The absorption was enhanced due to the greater concentration of polar constituents such as metal ions, which led to an increase in osmotic pressure between the membrane and the swelling solution [62]. Their presence and the existence of hydrophilic groups (-OH) in CA enable that, in addition to hydrogen bonds, electrostatic and van der Waals interactions are also created in the liquid-solid interface. As a result, regarding the electronic properties of the cations in m-SBF, the interplay between these forces can be favored [14]. Ca 2+ , for example, is usually present in SBF composition, which is used for producing SrP in the biomimetic method; however, in m-SBF, it was replaced by Sr 2+ . In our study, instead of Ca 2+ , Sr 2+ was used because of its interesting properties for bone repair and their similarity in ionic radii and valence (2+), and common biochemical route that would contribute to keeping the adsorbent capability and biological function. The m-SBF solution was used to induce SrP nucleation and growth into the CA membrane after contact for seven days and 14 days. Pan et al. showed that as the Sr 2+ content was augmented in HA, so did the solubility of the hybrid [58]. Conversely, later, Luz et al.'s study suggested that Sr 2+ presence contributed to increasing the solubility of the scaffold, and thus, its swelling degree [36].  EDX chemically verified the ability of the material to induce apatite formatio surface when immersed in m-SBF after seven and 14 days [54]. Figure 5a shows th ical element mapping of CA, where, as expected, no percentage of the Sr elem found, and pores were evident. After the treatment with m-SBF for seven CA7dSrP, 1.7% of Sr and 0.7% of P elements were identified in EDX (see Figur  CA14dSrP, a percentage of about 1.5% of Sr and 0.5% of P was observed (see Figu Unlike CA and CA7dSrP (Figure 5a,b), clusters of SrP were evident in CA14 high internalization by the CA membrane and the coating with other salts of mthe characterized layer level and the low penetration of the EDX technique may h resented an obstacle in detecting the Sr element in the CA7dSrP and CA14dSrP s A lower percentage of Sr was observed in CA7dSrP and CA14dSrP (Figure 5 still, the element was identifiable. To confirm its presence through other techniq quots of the reaction solution were analyzed by atomic absorption spectroscop during the synthesis of the hybrids after seven and 14 days corresponding to C and CA14dSrP, respectively. Figure 6 displays the amount of Sr 2+ adsorbed on the CA membrane in both m at the original concentration of Sr 2+ in the m-SBF solution. The hybrid membra sented an average percentage adsorption of about 92.3% and 96.1%, showing a sta significant difference between them, considering a p-value of less than 0.05. There suppose that the EDX technique in this work was not sensitive enough to discrim reach layers covered by a greater quantity of Sr 2+ metal ions and AAS successfully its adsorption to the CA matrix. EDX chemically verified the ability of the material to induce apatite formation on its surface when immersed in m-SBF after seven and 14 days [54]. Figure 5a shows the chemical element mapping of CA, where, as expected, no percentage of the Sr element was found, and pores were evident. After the treatment with m-SBF for seven days, in CA7dSrP, 1.7% of Sr and 0.7% of P elements were identified in EDX (see Figure 5b). In CA14dSrP, a percentage of about 1.5% of Sr and 0.5% of P was observed (see Figure 5c).
Unlike CA and CA7dSrP (Figure 5a,b), clusters of SrP were evident in CA14dSrP. A high internalization by the CA membrane and the coating with other salts of m-SBF and the characterized layer level and the low penetration of the EDX technique may have represented an obstacle in detecting the Sr element in the CA7dSrP and CA14dSrP samples.
A lower percentage of Sr was observed in CA7dSrP and CA14dSrP (Figure 5b,c), but still, the element was identifiable. To confirm its presence through other techniques, aliquots of the reaction solution were analyzed by atomic absorption spectroscopy (AAS) during the synthesis of the hybrids after seven and 14 days corresponding to CA7dSrP and CA14dSrP, respectively. Figure 6 displays the amount of Sr 2+ adsorbed on the CA membrane in both methods at the original concentration of Sr 2+ in the m-SBF solution. The hybrid membranes presented an average percentage adsorption of about 92.3% and 96.1%, showing a statistically significant difference between them, considering a p-value of less than 0.05. Therefore, we suppose that the EDX technique in this work was not sensitive enough to discriminate or reach layers covered by a greater quantity of Sr 2+ metal ions and AAS successfully proved its adsorption to the CA matrix.

In Vitro Degradability
A long-term physical degradation study was performed to evaluate the stability of the coating in stressing conditions. This parameter was estimated by measuring the weight loss (in %) and indicates possible changes in the structure after SrP incorporation compared to CA. CA has a partial hydrophobic nature that confers specific stability for its structure. The molecular weight of CA is related to the number of ester bonds in the acetyl group and, as it increases, it becomes more difficult to degrade because more groups need to be cleaved. As a result, cellulose monoacetate with the lowest acetyl content (about 29%) was used in this work [11]. The insertion of metal ions into the structure, like Sr 2+ and other constituents of m-SBF, through adsorption, can lead to water molecules surrounding the solid-liquid interface interacting and hydrating these metals forming complexes. This hydration can increase strength loss, loss of mass integrity, solubilization of the polymeric matrix, and degradability [63,64]. The degradation process can occur by several mechanisms, besides the physical already described, hydrolysis, oxidation, and enzymatic reactions can also occur, releasing wear particles in the biological media [20]. Therefore, a first biocompatible material can become toxic as fragmented particles can elicit an unwanted immunological process in the body. The materials produced in this work showed a similar degradation behavior for all the samples according to the first derivative calculated through the average value of degradability ( Figure 7). 29%) was used in this work [11]. The insertion of metal ions into the structure, like Sr 2+ and other constituents of m-SBF, through adsorption, can lead to water molecules surrounding the solid-liquid interface interacting and hydrating these metals forming complexes. This hydration can increase strength loss, loss of mass integrity, solubilization of the polymeric matrix, and degradability [63,64]. The degradation process can occur by several mechanisms, besides the physical already described, hydrolysis, oxidation, and enzymatic reactions can also occur, releasing wear particles in the biological media [20]. Therefore, a first biocompatible material can become toxic as fragmented particles can elicit an unwanted immunological process in the body. The materials produced in this work showed a similar degradation behavior for all the samples according to the first derivative calculated through the average value of degradability (Figure 7).
Nevertheless, after one, 10, and 30 days of exposure to the stressful environment, it was shown that the CA was significantly different from the CASrP samples, showing raised values of material loss. Additionally, at two, three, and 45 days, CA14dSrP was considered different statistically from CA, but close to CA7dSrP. In general, the percentage of mass loss was higher for CA concerning the CASrP samples and a higher degradation rate was observed mainly in the initial period, reaching a certain stability up to 30 days. Ge et al. reported a material close to ours, prepared by the biomimetic method and Sr 2+ -doped, but different because it used poly (L-lactic acid) as the polymer component. The authors obtained a higher hydrophilicity after nucleation of SrP in the structure and diminished the degradability percentage [32]. The CA7dSrP and CA14dSrP presented a similar degradation profile. Non-significant differences between the bulk values were observed at all times. Although a modification of crystallinity was verified because of the higher quantity of Sr 2+ in CA14dSrP (Figure 2a,b), this change had no significant effect on the degradation behavior of CA7dSrP and CA14dSrP. Nevertheless, after one, 10, and 30 days of exposure to the stressful environment, it was shown that the CA was significantly different from the CASrP samples, showing raised values of material loss. Additionally, at two, three, and 45 days, CA14dSrP was considered different statistically from CA, but close to CA7dSrP. In general, the percentage of mass loss was higher for CA concerning the CASrP samples and a higher degradation rate was observed mainly in the initial period, reaching a certain stability up to 30 days. Ge et al. reported a material close to ours, prepared by the biomimetic method and Sr 2+ -doped, but different because it used poly (L-lactic acid) as the polymer component. The authors obtained a higher hydrophilicity after nucleation of SrP in the structure and diminished the degradability percentage [32]. The CA7dSrP and CA14dSrP presented a similar degradation profile. Non-significant differences between the bulk values were observed at all times. Although a modification of crystallinity was verified because of the higher quantity of Sr 2+ in CA14dSrP (Figure 2a,b), this change had no significant effect on the degradation behavior of CA7dSrP and CA14dSrP. Figure 8 displays a graph of the amount of Sr dissolved into the cell culture medium over a 24 h period in the absence (indirect method) of cells. Although in vivo conditions are more complex than in vitro, the study of degradation behavior in culture media conditions containing fetal bovine serum (FBS) gives insights into the degradability in real conditions, and thus the Sr 2+ release resulting from this process [65]. Furthermore, the presence of FBS is considered to slow this process since it simulates the scenario in the biological fluid with proteins that can adsorb into the surface of the material [66,67].  Figure 8 displays a graph of the amount of Sr dissolved into the cell culture medium over a 24 h period in the absence (indirect method) of cells. Although in vivo conditions are more complex than in vitro, the study of degradation behavior in culture media conditions containing fetal bovine serum (FBS) gives insights into the degradability in real conditions, and thus the Sr 2+ release resulting from this process [65]. Furthermore, the presence of FBS is considered to slow this process since it simulates the scenario in the biological fluid with proteins that can adsorb into the surface of the material [66,67]. The quantity of metal ions released into the cell culture medium was 4.63 and 6.52 mg·L −1 for CA7dSrP and CA14dSrP exposed to the DMEM corresponding to L-929 cells, while it was 9 and 8 mg·L −1 for CA7dSrP and CA14dSrP in the α-MEM media referred to MC3T3-E1 cells, respectively. Thus, a slightly increased release was observed in the culture medium of osteoblasts. The presence of a superior quantity of amino acids [68] and glucose [69] in the composition of DMEM in comparison to α-MEM media may have driven the interaction between these components with the material and reduced the degradation in DMEM, as mentioned previously in the case of FBS addition in the test media. This is possibly because proteins from serum media may physically adsorb through electrostatic interaction between their amino groups with groups in the CA coating and hydrophobic interactions, helping to retain the membrane structure [66,70]. Additionally, the high glucose concentration in DMEM may influence Sr 2+ availability through interaction between the glucose and Sr 2+ metal ions. Zuberek et al. showed that the difference in glucose concentration in the culture medium for cell maintenance negatively affected silver nanoparticle toxicity and oxidative stress [69].

Cell Cytotoxicity
GBR therapy involves stimulating the reconstitution of bone and occurs mainly using membranes to isolate the damaged area from other tissues that can invade and interfere with the repair process. The membrane serves as a support, simulating the extracellular matrix found in bone composition and interaction between biomolecules and the scaffold [71]. Fibroblasts and osteoblasts found in the body play a crucial role in the homeostasis of this structure, synthesizing components of the innate immune system and inducing bone healing, respectively. To favor this process, a coating for bone implants must trigger an appropriate acute inflammatory response and, ideally, ensure bone formation while The quantity of metal ions released into the cell culture medium was 4.63 and 6.52 mg·L −1 for CA7dSrP and CA14dSrP exposed to the DMEM corresponding to L-929 cells, while it was 9 and 8 mg·L −1 for CA7dSrP and CA14dSrP in the α-MEM media referred to MC3T3-E1 cells, respectively. Thus, a slightly increased release was observed in the culture medium of osteoblasts. The presence of a superior quantity of amino acids [68] and glucose [69] in the composition of DMEM in comparison to α-MEM media may have driven the interaction between these components with the material and reduced the degradation in DMEM, as mentioned previously in the case of FBS addition in the test media. This is possibly because proteins from serum media may physically adsorb through electrostatic interaction between their amino groups with groups in the CA coating and hydrophobic interactions, helping to retain the membrane structure [66,70]. Additionally, the high glucose concentration in DMEM may influence Sr 2+ availability through interaction between the glucose and Sr 2+ metal ions. Zuberek et al. showed that the difference in glucose concentration in the culture medium for cell maintenance negatively affected silver nanoparticle toxicity and oxidative stress [69].

Cell Cytotoxicity
GBR therapy involves stimulating the reconstitution of bone and occurs mainly using membranes to isolate the damaged area from other tissues that can invade and interfere with the repair process. The membrane serves as a support, simulating the extracellular matrix found in bone composition and interaction between biomolecules and the scaffold [71]. Fibroblasts and osteoblasts found in the body play a crucial role in the homeostasis of this structure, synthesizing components of the innate immune system and inducing bone healing, respectively. To favor this process, a coating for bone implants must trigger an appropriate acute inflammatory response and, ideally, ensure bone formation while attending to the exclusion principle during the bone healing process. In addition to biocompatibility, implant degradability is an imperative factor that must be considered when designing a scaffold for bone repair, as it would avoid the need for a second surgery to remove this implant after tissue repair. In addition, however, it is necessary to ensure that particles from implant degradation are neither toxic nor inappropriately activating the innate immune system.
In our work, we evaluated the cytotoxicity of the materials by the indirect assay in which fibroblasts and osteoblasts cells were exposed to the sample extracts containing some particles and Sr 2+ released from the materials during the extraction time. As a result, the cellular response was elicited not by the produced composites themselves, but as a consequence of their soluble degradation products after incubation at 37 • C for 24 h [65]. Figure 9 shows the cellular viability (in %) results from L-929 mouse fibroblasts and mouse osteoblasts after their exposure to CA and CASrP composites to the control (cells exposed only to the supplemented medium). The L-929 viability was higher for CASrP materials than CA, obtaining an average percentage of 75%, 86%, and 83% for 24 h and 70%, 92%, and 88% for 48 h after exposure to the media containing extracts of CA, CA7dSrP, and CA14dSrP, respectively (Figure 9a). Although cell viability increased after 48 h for CASrP composites, this difference was not considered statistically significant. In addition to L-929 cells, the MC3T3-E1 cells enabled us to analyze the cell viability of a bone cell line related to the formation of bone tissue. After incubation with the extracts of the CA, CA7dSrP, and CA14dSrP for 24 h, the average percentage of MC3T3-E1 cells obtained was 89%, 105%, and 106%, respectively, while for 48 h, it was 51%, 115%, and 107%, respectively (Figure 9b). The cell viability of CA decreased significantly after 48 h of exposure, while the incorporation of SrP on CA improved the biocompatibility of the polymeric matrix for both fibroblasts and osteoblasts. Since the HA produced was similar to the biological apatite, the inorganic component may have contributed to improved compatibility, while Sr 2+ presence in the structure increased its solubility and the release of Sr 2+ , which may lead to the proliferation of osteoblasts [72]. Several studies have demonstrated the osteogenic potential of SrHA doped composites [32,51,52,72]. Ge et al. reported increased bone volume and ALP activity, which is an important biomarker referred to in osteoblast differentiation, after the addition of HA into poly(L-lactic acid), and higher protein adsorption, bone volume, and ALP activity with the addition of Sr 2+ into HA/poly(L-lactic acid) [32]. Likewise, Zhang et al. evidenced a higher ALP activity after incorporating SrHA into a polymeric structure [52]. In another work, it was shown that depending on the concentration of Sr 2+ in the composite, a similar response with the HA of concern can be obtained. Increasing the Sr 2+ concentration proportion in the composite highly improved ALP activity [72]. Nevertheless, the quantity of Sr 2+ obtained for both samples after seven and 14 days of treatment in m-SBF for synthesizing CA7dSrP and CA14dSrP was not significantly different concerning chemical composition and degradability behavior. On the other hand, Sr 2+ absorption in CASrP production and Sr 2+ release in the culture media pointed out increased adsorption in m-SBF and metal ion release in DMEM for CA14dSrP.
Overall, CA is considered as a biocompatible polymer [19,73], presenting cell viability above 70%, referring to the required minimum value according to the international standard ISO 10993-5:2009. ISO 10993-5:2009 comprises the evaluation of the cytocompatibility of medical devices. However, the CA produced in our work resulted in 51% and 70% of living osteoblasts and fibroblasts after 48 h of exposure to the extracts, respectively. The greater and variation of the cytotoxicity of CA, despite this, were not considered statistically different at 24 h and 48 h [19,73], may have resulted from its degradation in the initial time (24 h), producing fragments of CA, acetic acid, and cellulose [74,75]; the composition media [68,69] as well as resulting from the lack of osteogenic activity associated with this material, while the presence of SrP may have favored increased compatibility because of its similarity to HA and Sr 2+ release [32,35,42,43]. It is known that Sr 2+ toxicity is dosedependent [42,72], on the other hand, HA is a known biocompatible constituent that also improves the physical properties of composites.
In contrast to organisms, cell cultures are a more sensitive model, and, naturally, as the material and the culture media are maintained in contact for a specific time, products of membrane degradation are released in the culture medium and the extract containing these degraded materials may intensify the toxicity, as shown by Figure 9 [76]. However, the release of Sr 2+ in the culture media improved the biocompatibility of CA itself [43,72]. Additionally, the superior concentration of Sr 2+ obtained in the extract after 24 h in α-MEM followed the high cell viability observed for CA7dSrP and CA14dSrP in osteoblasts to fibroblasts (see Figure 8). The enhanced cell viability, even superior to the control, may indicate cell proliferation or increase in metabolic activity of the osteoblasts [43,72].
is dose-dependent [42,72], on the other hand, HA is a known biocompatible constituent that also improves the physical properties of composites.
In contrast to organisms, cell cultures are a more sensitive model, and, naturally, as the material and the culture media are maintained in contact for a specific time, products of membrane degradation are released in the culture medium and the extract containing these degraded materials may intensify the toxicity, as shown by Figure 9 [76]. However, the release of Sr 2+ in the culture media improved the biocompatibility of CA itself [43,72]. Additionally, the superior concentration of Sr 2+ obtained in the extract after 24 h in α-MEM followed the high cell viability observed for CA7dSrP and CA14dSrP in osteoblasts to fibroblasts (see Figure 8). The enhanced cell viability, even superior to the control, may indicate cell proliferation or increase in metabolic activity of the osteoblasts [43,72]. Figure 9. L-929 mouse fibroblast (a) and MC3T3-E1 osteoblast (b) viability of CA, CA7dSrP, and CA14dSrP after 24 and 48 h. Letters inside the graphs indicate homogeneous groups (treatments with non-significant differences). Two-way ANOVA followed by Tukey's post-hoc multiple comparison tests were performed for statistical analysis of fibroblasts data, while the Kruskal-Wallis test was performed for the osteoblast data. A p-value of <0.05 was considered significant.

Conclusions
In this work, we report on the incorporation of SrP into CA membranes for the first time, effectively producing two promising composites for GBR (CA7dSrP and CA14dSrP). CA is highly relevant as a support to SrP crystal nucleation and scaffold design. In addition, the coating effectiveness was investigated by evaluating its bioactivity and biocompatibility for implant applications.
The SrP formation modified the morphology and physicochemical properties of the hybrid coating, leading to the formation of SrP sphere-like crystals, a slight decrease in the long-term degradation process, and an augmentation in crystallinity. Additionally, this inorganic component played a significant role in biocompatibility in comparison to CA alone.
Finally, even though no statistical difference was found in the properties of CA7dSrP and CA14dSrP and the studies were performed in vitro indirectly, the Sr 2+ release study allowed us to quantify the concentration of Sr 2+ in the exposure media, fairly well representing the body conditions, and showing that it was in an optimal dose for bone regeneration since the osteoblast viability (>100%) was boosted, instigating future research.  Letters inside the graphs indicate homogeneous groups (treatments with non-significant differences). Two-way ANOVA followed by Tukey's post-hoc multiple comparison tests were performed for statistical analysis of fibroblasts data, while the Kruskal-Wallis test was performed for the osteoblast data. A p-value of <0.05 was considered significant.

Conclusions
In this work, we report on the incorporation of SrP into CA membranes for the first time, effectively producing two promising composites for GBR (CA7dSrP and CA14dSrP). CA is highly relevant as a support to SrP crystal nucleation and scaffold design. In addition, the coating effectiveness was investigated by evaluating its bioactivity and biocompatibility for implant applications.
The SrP formation modified the morphology and physicochemical properties of the hybrid coating, leading to the formation of SrP sphere-like crystals, a slight decrease in the long-term degradation process, and an augmentation in crystallinity. Additionally, this inorganic component played a significant role in biocompatibility in comparison to CA alone.
Finally, even though no statistical difference was found in the properties of CA7dSrP and CA14dSrP and the studies were performed in vitro indirectly, the Sr 2+ release study allowed us to quantify the concentration of Sr 2+ in the exposure media, fairly well representing the body conditions, and showing that it was in an optimal dose for bone regeneration since the osteoblast viability (>100%) was boosted, instigating future research.