Biofunctionalization of titanium surfaces with alendronate and albumin modulates osteoblast performance.

BACKGROUND: Biofunctionalization of titanium surfaces can improve host responses, especially considering the time for osteointegration and patient recovery. This prompted us to modify titanium surfaces with alendronate and albumin and to investigate the behavior of osteoblasts on these surfaces.
METHODS: The biofunctionalization of titanium surfaces was characterized using classical physicochemical approaches and later used to challenge pre-osteoblast cells up to 24 h. Then their viability and molecular behavior were investigated using mitochondrial dehydrogenase activity and RTq-PCR technologies, respectively. Potential stimulus of extracellular remodeling was also investigated by zymography.
RESULTS: Our data indicates a differential behavior of cells responding to the surfaces, considering the activity of mitochondrial dehydrogenases. Molecularly, the differential expression of genes related with cell adhesion highlighted the importance of Integrin-β1, Fak, and Src. These 3 genes were significantly decreased in response to titanium surfaces modified with alendronate, but this behavior was reverted when alendronate was associated with albumin. Alendronate-modified surfaces promoted a significant increase on ECM remodeling, as well as culminating with greater gene activity related to the osteogenic phenotype (Runx2, Alp, Bsp).
CONCLUSION: Altogether, our study found interesting osteogenic behavior of cells in response to alendronate and albumin surfaces, which indicates the need for in vivo analyses to better consider these surfaces before clinical trials within the biomedical field.

Introduction

Biomaterials play key roles in procedures within the engineering and medicine fields, in which their primary role worldwide is to benefit people's quality of lives with several applications in the dental, orthopedic, and cardiovascular areas [1]. Metal-based alloys are commonly used as biomaterials, especially for implants, and titanium is one of the metals most employed for these purposes. Commercially pure titanium (cp-Ti) and its alloys are promising metals for these biomaterials, due to relevant characteristics that determine the efficacy of the implant, such as biocompatibility and high resistance to corrosion [2, 3], as well as greater interaction with biological tissues than other metals [4, 5, 6, 7, 8, 9, 10].

Although titanium presents considerable interesting and already known properties, improving their biological properties and modifying their surfaces are a very interesting strategy. The surface of the biomaterial is responsible for a significant role with direct contact on the host tissue that ensures the osseointegration of the implant. The interaction with the peri-implant in the early steps of tissue healing determine the success of later stages as appositional tissue growth. Several modifications on the surfaces of the implants have been proposed to improve the interaction between the implant and host tissue [1, 11, 12].

The surface of the titanium is composed of a layer of oxide (TiO2) that forms promptly. This film is thin and unstable and may not withstand the mechanical stresses required for an implant. Therefore, this native layer must be thickened, since the properties of the film are a decisive factor that define the degree of corrosion for metal or alloys [13, 14]. In addition to surface treatments to increase TiO2 film thickness, the literature reports the efficiency of this layer in delivering and releasing specific drugs to optimize the osseointegration process [15, 16, 17, 18]. Several categories of drugs are used for this purpose, but currently attention is focused on the class of bisphosphonates to modulate bone response, by directly hampering the action of osteoclasts [15, 19]. Sodium alendronate is an important drug of interest because of studies that report increased bone density and formation, once it has been used for bone disorders such as osteoporosis [20, 21].

The reactional tissue close to the implant is important for circulating protein for bioactivity of implants, because the biological fluids containing proteins make first contact with the biomaterials, establishing a “biological cap”, which will be recognized later by the host cells [22, 23]. Among the circulating proteins with already known properties, albumin (BSA) has important and interesting roles especially considering bloodstream and osteoblast performance [24, 25].

This situation prompted us to evaluate the activity of pre-osteoblastic cells (MC3T3-E1, subclone 4) on the three proposed titanium surfaces, considering TiO2 thin film, bisphosphonate (alendronate) and albumin (BSA) coatings. Although some studies have look at these materials, further information is needed about the biological performance of osteoblast when in contacting with those surfaces. Better understanding about the biocompatibility and performance of new surfaces are important preclinical issues.

Materials and methods

TiO2 synthesis and surface modification methodology

Titanium dioxide synthesis. The titanium dioxide solution was synthesized using a sol-gel process, as previously reported by our group [26].

Substrate preparation. Commercial pure titanium discs (12.7 mm diameter and 3 mm thickness) were used as substrates. they were polished (meshes of 320, 400, 600, and 800) to make the surface uniform and to remove the natural oxide layer. The substrates were washed in an ultrasonic bath for 30 min, of which 10 min were immersed in deionized water, 10 min in isopropyl alcohol, and 10 min in deionized water. Then, the samples were hydroxylated in a mixture of 7:3 (v/v) 98% H2SO4 and 30% H2O2 for 2 h [26].

Oxide deposition. The deposition was performed by the spin coating technique (2000 rpm for 60 s) and this procedure was repeated three times for each sample to acquire thicker layers. Between the depositions, the samples were heated to 40 °C for 5 min. After that, the substrates coated with TiO2 oxide were annealed at 850 °C for 2 h with a heating rate of 1 °C min−1 to obtain a rutile crystalline polymorphic phase [26].

Deposition of the biomolecules

After growing TiO2 film on the surfaces, the samples were functionalized with sodium alendronate and albumin, separately. The functionalization of the discs occurred using an immersion method, as described in detail below.

Sodium alendronate. The samples were immersed in a solution containing deionized water and methanol at a 60:40 ratio, respectively. The sodium alendronate (99.99%, Sigma-Aldrich Corp., St. Louis, MO, USA) was then dissolved in the solution at a concentration of 0.5 mg/ml. Three titanium devices remained immersed in 5 ml of the solution for 12 h to maintain control of the samples. Then, the samples were washed with deionized water and heated to 100 °C for 2 h [27].

Albumin (BSA). The samples were immersed in a solution containing 0.4 mg/ml of albumin (BSA, 96%, Sigma-Aldrich Corp., St. Louis, MO, USA) in a phosphate buffer solution (PBS, Sigma-Aldrich Corp.). Three substrates remained immersed in 30 ml of the solution for 6 h to maintain control of the samples. Finally, the samples were washed with PBS and deionized water [28].

TiO surfaces modified with enriched medium. The TiO2 discs (n = 3) were conditioned in α-MEM medium without FBS up to 24 h at 37 °C, 5% CO2, and 95% humidity (ISO 10993:2016). This conditioned media contains molecules potentially released by the surfaces.

TiO2 surface characterization

The crystal structure of the TiO2 films was analyzed by X-ray diffraction (XRD, D/MAX-2100/PC, Rigaku) using Cu Kα radiation (λ = 1.54056 Å) coupled to a nickel filter, to eliminate the Cu Kβ radiation. The measurements were taken from 20° to 45°, with regular step of 0.02° min−1 and scan speed of 2° min−1, at 40 kV/20 mA. For phase identification, the PCPDF 65–3411, 44–1294, and 76–1949 cards were employed for Ti and TiO2, respectively.

The XPS measurements were performed on a Kratos spectrometer (AXIS-165) using a monochromatic Al Kα X-ray source. Each sample was analyzed at a 54.7° angle, which is defined as the emission angle relative to the surface. The energy resolution was 0.45 eV. The pass spectrum energy of the spectra was 80 eV and the individual high-resolution spectra was 20 eV. To correct for any charging effect, the binding energy of the Ti 2p3/2 peak was normalized to 458.8 eV. The CasaXPS computational package was used for the chemical study and analysis of the functionalized sample surfaces.

Surface roughness characterization was performed by Confocal Microscopy (Leica DCM3D) employing a 20x lens. Arithmetical surface roughness values (Ra) were determined by calculating the mean value from six randomly selected areas of each sample [29].

Biological analysis

Osteoblasts and culture conditions. MC3T3-E1 cells (pre-osteoblast cells, subclone 4) were cultured in α minimum essential medium (α-MEM) supplemented with 10% fetal bovine serum (FBS) and 1% antibiotics (penicillin and streptomycin) (NUTRICELL, Campinas, SP, Brazil) and maintained at 37 °C, 5% CO2, and 95% humidity. In all experiments, trypsinization of the cells was performed in the sub-confluent stage and with low passages (<20).

Cell viability assay. The pre-osteoblasts used were treated with the conditioned medium by the different surfaces modified during the 24-hour period, which were then trypsinized and seeded (1×104 cells/ml) into 96-well microplates. The viable cells were stained with the vital dye MTT (1 mg/ml). The insoluble blue dye that formed was resuspended appropriately in dimethylsulfoxide (DMSO) and the cell viability was estimated by measuring absorbance in a microplate reader (SYNERGY-HTX multi-mode reader, Biotek, USA) [30] at a 570 nm wavelength.

Cell adhesion assay. The cells were treated with the conditioned medium for 24 h, then they were trypsinized and seeded (1×104 cells/ml) in to 96-well microplates. After 24 h of seeding, cell adhesion was estimated by incorporating the crystal violet. The resulting absorbance was measured at 540 nm using a microplate reader (SYNERGY-HTX, multi-mode reader, Biotek, USA).

Extracellular remodeling stimulus was measured by matrix metalloproteinase activities. Once we had demonstrated the importance of ECM remodeling during the early behavior of osteoblast interacting with biomaterials surfaces, we addressed the proteolytic activities of both Mmp2 and Mmp9 by performing gelatin-based zymography technology. Briefly, the conditioned medium was collected, clarified by centrifugation at 13.200 xg for 15 min at 4 °C and stored at -20 °C. Protein quantification was performed by the Lowry method [31] and the samples were prepared to be applied in equal amounts (150 μg) in 10% (w/v) SDS-polyacrylamide gel and 4% (w/v) gelatin. MMP renaturation was performed in 2% (v/v) Triton X-100 for 40 min, followed by incubation in a proteolysis buffer [50 mM Tris-HCl and 10 mM CaCl2 (pH 7.4)] at 37 °C for 18 h. Subsequently, the gels were stained with 0.5% (w/v) coomassie blue G 250 for 30 min, washed in a 30% (v/v) methanol and 10% (v/v) acetic acid glacial and analyzed using the ImageJ Software [32].

Quantitative PCR assay (qPCR). After being treated with conditioned medium by the different modified titanium surfaces for 24 h, the total mRNA of the cells was extracted using the Ambion TRIzol Reagent (Life Sciences - Fisher Scientific Inc, Walthan, MA, USA) and then these samples were treated with DNase I (Invitrogen, Carls-band, CA, USA). The synthesis of the cDNA was performed using a high capacity cDNA reverse transcription kit (Applied Biosystems, Foster City, CA, USA) based on the manufacturer's instructions. A 10 μl of sample containing PowerUpTMSYBRTM Green Master Mix 2x (5 μl) (Applied Biosystems, Foster City, CA, USA), 0.4 μM of each primer, 50 ng of cDNA, and nuclease-free H2O was performed using qPCR. The results were expressed as relative quantities of each selected target gene, using the Gapdh gene as housekeeping with the cycle threshold (Ct) method. The analyzed primers and their associated details are described in Table 1.

Table 1

Expression primers sequences and PCR cycle conditions.

Gene (ID)	Primer	5′- 3′ Sequence	Reaction Condition
Gene expression
RUNX2 (12393)	Forward	GGACGAGGCAAGAGTTTCA	95 °C - 15 s; 60 °C - 30 s; 72 °C - 30 s
	Reverse	TGGTGCAGAGTTCAGGGAG
BSP (103993)	Forward	GTACCGGCCACGCTACTTTCT	95 °C - 15 s; 60 °C - 30 s; 72 °C - 30 s
	Reverse	GTTGACCGCCAGCTCGTTTT
ALP (11647)	Forward	GAAGTCCGTGGGCATCGT	95 °C - 15 s; 60 °C - 30 s; 72 °C - 30 s
	Reverse	CAGTGCGGTTCCAGACATAG
MMP9 (17395)	Forward	TGTGCCCTGGAACTCACACGAC	95 °C - 15 s; 60 °C - 30 s; 72 °C - 30 s
	Reverse	ACGTCGTCCACCTGGTTCACCT
MMP2 (17390)	Forward	AACTTTGAGAAGGATGGCAAGT	95 °C - 15 s; 60 °C - 30 s; 72 °C - 30 s
	Reverse	TGCCACCCATGGTAAACAA
FAK (14083)	Forward	TCCACCAAAGAAACCACCTC	95 °C - 15 s; 60 °C - 30 s; 72 °C - 30 s
	Reverse	ACGGCTTGACACCCTCATT
SRC (17977)	Forward	TCGTGAGGGAGAGTGAGAC	95 °C - 15 s; 60 °C - 30 s; 72 °C - 30 s
	Reverse	GCGGGAGGTGATGTAGAAAC
TIMP2 (21858)	Forward	GCAACAGGCGTTTTGCAATG	95 °C - 15 s; 60 °C - 30 s; 72 °C - 30 s
	Reverse	CGGAATCCACCTCCTTCTCG
INTEGRIN-Β1 (16412)	Forward	CTGATTGGCTGGAGGAATGT	95 °C - 15 s; 60 °C - 30 s; 72 °C - 30 s
	Reverse	TGAGCAATTGAAGGATAATCATAG
COFILIN (12631)	Forward	CAGACAAGGACTGCCGCTAT	95 °C - 15 s; 60 °C - 30 s; 72 °C - 30 s
	Reverse	TTGCTCTTGAGGGGTGCATT
PP2A (19052)	Forward	ATGGGCCTCTCTCCCATTCT	95 °C - 15 s; 60 °C - 30 s; 72 °C - 30 s
	Reverse	CATGCACAGGGAGTGACAGT
BGLAP2 (12097)	Forward	AGACTCCGGCGCTACCTT	95 °C - 15 s; 60 °C - 30 s; 72 °C - 30 s
	Reverse	CTCGTCACAAGCAGGGTTAAG
SPP1 (20750)	Forward	TTTGCTTTTGCCTGTTTGGC	95 °C - 15 s; 60 °C - 30 s; 72 °C - 30 s
	Reverse	CAGTCACTTTCACCGGGAGG
DMP1 (13406)	Forward	CCCAAAACAGGAAAGGATCA	95 °C - 15 s; 60 °C - 30 s; 72 °C - 30 s
	Reverse	ACCAGACTGCTTCTCCTCCA
GAPDH (14433)	Forward	AGGCCGGTGCTGAGTATGTC	95 °C - 15 s; 60 °C - 30 s; 72 °C - 30 s
	Reverse	TGCCTGCTTC ACCACCTTCT

Statistical analysis

All the experiments were performed at least three times for better statistics, and the results were plotted as mean ± standard deviation (SD). A one-way analysis of variance (ANOVA) was performed using the Tukey test for multiple comparisons with the GraphPad Prism7 software (GraphPad Software Inc., San Diego, CA, USA) to evaluate whether different treated surfaces of the biomaterials influenced the adhesion or cell differentiation.

Results and discussion

In the last few decades, interest has increased about developing a biomaterial that can biologically interact with host tissue assembling properties of autogenous bone tissue, to improve osseointegration process and reduce implant failures. An important factor to consider is the surface properties, which are the first part of the implanted devices in contact with the host tissues and is expected to later drive cell behavior. To address this issue, we produced different modification on the titanium based surfaces and then performed in vitro tests to define the biological behavior of osteoblast cells responding to them, considering cell adhesion, ECM remodeling, and osteoblastic phenotype of these cells, which are considered important outcomes to evaluate in response to titanium-based surfaces [5]. Considering the dynamism of the host biology in responding to implantable devices, we previously demonstrated that cells respond directly to the surfaces, but what about the indirect effects that trigger responses [8], especially considering neighboring cells.

Initially, the structural characterization of the TiO2 thin films was performed using X-ray diffraction (XRD). The obtained XRD results indicate that the formed TiO2 films had a rutile crystalline phase, as evidenced by the presence of a high-intensity diffraction peak (110) at 2θ = 27.4°, as shown in Figure 1.

Figure 1

Crystallographic characterization of the rutile phase TiO2 film.

Furthermore, Fourier transform infrared spectroscopy (FTIR) measurements confirm the effectiveness of the functionality, as shown in Figure 2. In the region between 570 and 648 cm−1 (1), observe the Ti–O bond band, relative to the TiO2 film. The presence of P–O bonds is confirmed by bands in the region between 1160 and 1172 cm−1 (2), while values between 1279 and 1314 cm−1 (3) confirmed the existence of P=O bonds, in the alendronate molecule. The region between 3032 and 3045 cm−1 (4) refers to the binding of NH2, which is present in both alendronate and albumin [33, 34, 35]. Thus, the functionalization of rutile TiO2 with alendronate and albumin was efficient, indicating that the prepared film was effectively functionalized.

Figure 2

FTIR spectrum for alendronate (ALN) and albumin (BSA) functionalized samples.

X-ray photoelectron spectroscopy (XPS) experiments were performed to better analyze the chemical compositions of the surfaces and the electronic states of the functionalized materials. The spectrum of TiO2 emphasizes the presence of the TiO2 film due to the higher O 1s peak intensity (Figure 3a). The Ti 2p1/2 (464.4 eV) and Ti 2p3/2 (458.7 eV) contributions [36] (Figure 3b) are two peaks associated with the contributions from the base metal and have an energy difference of 5.7 eV, which is consistent with the oxidation state Ti+4 presented in TiO2. The high-resolution spectrum for the O 1s peak (Figure 3c) is the convolution of three peaks. The two low energy peaks are related to the TiO2 metal oxide bonds [37]. The third contribution has the highest energy and is related to the oxygen of the Ti–OH bond [36].

Figure 3

XPS survey spectrum of the TiO2 (a), high-resolution Ti 2p spectrum (b), and high-resolution O 1s spectrum (c).

Figure 4a illustrates the presence of the O 1s, Ti 2p, N 1s, P 2p, and C 1s peaks for the alendronate functionalized surfaces (TiO2 + ALN). The presence of the N 1s and P 2p peaks in the survey spectrum is related to the elemental nitrogen and phosphorous from the alendronate, validating the presence of this molecule at the surface. The high-resolution spectrum (Figure 4b to 4e) indicates the functionalization of the surfaces. The O 1s spectrum, in Figure 4b shows a peak at 532.0 eV associated with the P–O/OH [38], which is also related to the alendronate. The Ti–O peak with a binding energy of 530.0 eV indicates the presence of the oxide layer [39].

Figure 4

XPS survey spectrum of the TiO2 + ALN (a), high-resolution O 1s spectrum (b), high-resolution C 1s spectrum (c), high-resolution N 1s spectrum (d), and high-resolution P 2p spectrum (e).

The three different contributions to the C 1s peak (Figure 4c) are at 285.3 eV (C–C), 286.7 eV (C–O/C–N) [40], and 289.3 eV (C-PO (OH)2) [41]. The presence of the C-N and C-PO (OH)2 bonds confirm the existence of alendronate on the surface. Finally, the N 1s spectrum (Figure 4d) confirms the functionalization due to the NH3+ [42] peak at 401.2 eV and the organic binding C–N at 399.2 eV [40]. The presence of NH3+ suggests that it could have been generated by the formation of a zwitterionic amine structure, which favors adsorption through the C-PO(OH)2 group [43, 44]. Figure 4e shows the high-resolution spectrum for P 2p, which contains two contributions, P 2p1/2 (134.5 eV) and P 2p3/2 (133.3 eV) [36], confirming the presence of alendronate on the TiO2 surface.

The evaluation of the albumin functionalized surfaces is presented in Figure 5. The TiO2 + BSA survey spectrum (Figure 5a) demonstrates increased binding energy for the N 1s and C 1s signals compared to the alendronate-functionalized surfaces, which may be due to the chemical composition of albumin. The O 1s spectrum in Figure 5b contains the OH (531.8 eV) and Ti–O (529.9 eV) [45] peaks for the titanium bonds. The three contributions present for the C 1s spectrum (Figure 5c) are all related to the albumin functionalization. The N 1s spectrum (Figure 5d) indicates that the peak at 401.5 eV [42] is related to the presence of NH3+ in the albumin molecule, as well as the high-intensity C–N peak.

Figure 5

XPS survey spectrum of the TiO2 + BSA (a), high-resolution O 1s spectrum (b), high-resolution C 1s spectrum (c), and high-resolution N 1s spectrum (d).

Additionally, the TiO2 + ALN + BSA survey spectrum in Figure 6a has peaks associated with the functionalization of the two compounds. In Figure 6b, the OH contributions (531.9 eV) and Ti–O (530.0 eV) peaks were nearly the same as the albumin-only functionalization. For the C 1s spectrum (Figure 6c), the three different contributions are the COOH/C-PO (OH)2, C–N, and C–C. The presence of the COOH groups are due to the albumin, while the C-PO (OH)2 groups are from alendronate. In the N 1s spectrum (Figure 6d), contributions from the NH3+ (401.2 eV) and C–N (400.3 eV) [41] found may be related to the active functionalization of both alendronate and albumin. In Table 2, a compositional analysis of high-resolution XPS peaks is provided.

Figure 6

XPS survey spectrum for the TiO2 + ALN + BSA (a), high-resolution O 1s spectrum (b), high-resolution C 1s spectrum (c), and high-resolution N 1s spectrum (d).

Table 2

Compositional analysis of high resolution XPS peaks.

	TiO2 (Figure 3)	TiO2 + ALN (Figure 4)	TiO2 + BSA (Figure 5)	TiO2 + ALN + BSA (Figure 6)
Ti 2p
O 1s	TiO2 – 77.89%TiO – 5.02%O–H – 17.09%	OH/P–O – 52.71%Ti–O – 47.29%	OH – 61.52%Ti–O – 38.48%	OH – 58.12%Ti–O – 41.88%
C 1s		C–N – 33.13%C–OH – 58.92%C – 7.95%	C–C – 45.34%COOH – 16.50%C–N – 38.16%	C–N – 50.34%COOH – 23.41%C – 26.24%
N 1s		NH3 – 72.09%C–N – 27.91%	C–N – 85.05%NH3 – 14.95%	NH3 – 34.30%N – 63.70%
P 2p		2p3/2–73.87%2p1/2–26.13%		2p3/2–65.76%2p1/2–26.35%Na3PO4 – 7.89%

The surface roughness of the samples was investigated using Confocal Optical Microscopy considering the arithmetic surface roughness (Ra). Figures 7, 8, 9, 10, and 11 are 2D and 3D images obtained from the surface roughness analysis of the samples.

Figure 7

Confocal Optical Microscopy image for the sanded Ti Cp4 sample in 2D (a) and 3D (b).

Figure 8

Confocal Optical Microscopy image for the sanded TiO2 sample in 2D (a) and 3D (b).

Figure 9

Confocal Optical Microscopy image for the sanded TiO2 + ALN sample in 2D (a) and 3D (b).

Figure 10

Confocal Optical Microscopy image for the sanded TiO2 + BSA sample in 2D (a) and 3D (b).

Figure 11

Confocal Optical Microscopy image for the sanded TiO2 + ALN + BSA sample in 2D (a) and 3D (b).

First, we verify whether the medium conditioned by the modified surfaces was cytotoxic, and the results showed a higher mitochondrial dehydrogenase activity in all modified groups, presenting more viable cells that the control (polished Ti), especially in TiO2 + BSA (Figure 12a). After the non-toxicity of the conditioned medium by the surfaces was identified, the cell adhesion was evaluated using violet crystal and a standardization of the number of adhered cells was observed in all groups (Figure 12b). These results suggest a differential metabolic behavior of the osteoblast cells interacting with the different titanium surfaces evaluated in this study, which prompted us to better address the molecular behavior of these cells.

Figure 12

Cell survival and adhesion behavior promoted by the different surfaces. Cytotoxic effects of TiO2-modified surfaces were measured by mitochondrial dehydrogenase activity (MTT assay) as represented by the percentage of cellular viability of the pre-osteoblasts (a). Cell adhesion was evaluated when pre-osteoblasts were treated up to 24 h with enriched medium from modified surfaces and the adherent cells were stained with violet crystals (b). A one-way ANOVA was performed with Tukey's test as a multiple comparison test to evaluate the effects of different treated surfaces on the cellular adhesion. The results are represented as the mean ± SD of three independent experiments. Differences were considered when ∗ for p < 0.05, ∗∗ for p < 0.001, and ∗∗∗∗ for p < 0.0001.

To better understand the intracellular mechanisms involved in adhesion events, we evaluated genes related to cell adhesion using qPCR technology. The results illustrate a repertory of genes related to cell adhesion. First, the gene expression of Integrin-β1 (Figure 13a) was differently required considering the biomaterials, with a significant decrease in response to TiO2 + ALN and an significant increase in response to TiO2 + BSA and TiO2 + ALN + BSA. Integrin has subunits (α and β) that are correlated with transduction of signals, connecting with ECM proteins and other signaling proteins included in cell proliferation and migration processes [46]. Integrin activation benefits the interaction with other key proteins involved in this cellular signaling cascade, such as Fak and Src, responsible for cell-surface interaction, which exhibited significant gene expression on remodeled surfaces [47, 48]. Therefore, we investigated the behavior of both Fak and Src genes. Figures 13b and 13c show a panorama of the expression very similar to Integrin-β1. Fak and Src were probably activated before 24 h when in contact with the different surfaces; however, these results highlight that both these genes are modulated during the phenotype changes of these cells.

Figure 13

Differential behavior of osteoblast responding to different titanium-based surfaces – a special look at cell adhesion mechanism. The total mRNA was collected and adhesion-related genes (Integrin, Fak, Src, Cofilin, and Pp2A) evaluated by utilizing qPCR technology (a–e). The results are represented as the mean ± SD of three independent experiments. A one-way ANOVA was performed with Tukey's test as a multiple comparison test to evaluate the effects of different treated surfaces on the cellular adhesion. Differences were considered when ∗ for p < 0.05, ∗∗ for p < 0.001, and ∗∗∗∗ for p < 0.0001.

Cytoskeleton rearrangement is an important event during cell adhesion; therefore, we also investigated genes related with this condition and found that the expression of Cofilin significantly decreased in response to TiO2 + ALN (Figure 13d). As Cofilin activity is modulated by phosphorylation at Ser, we evaluated Ser/Thr phosphatase. The activity of the Pp2A gene decreased in osteoblast responding to TiO2 + ALN and TiO2 + BSA (Figure 13e). The osteoblast clearly responded differently to the proposed surfaces, modifying their phenotype and requiring genes related to cell adhesion and cytoskeleton rearrangement. In addition, an important axis of intracellular signaling seems to require Pp2A to modulate the Cofilin activity through transient dephosphorylating the Ser03.

The metabolism, involved in these cells challenged with the different surface compositions, displayed an important dynamism in the phenotypic variations necessary for this adaptive process, significantly interfering in responsive cells indirectly, which indicates that this event is something dynamic in the peri-implant tissue (considering this material in the host). For this cell adhesion, supramolecular platforms explain this performance of adhesion as previously reported through behaviors of the Fak and Src genes. The importance was also highlighted of Src in adhesion processes due to osteoblast differentiation promoting important control in osteogenesis; which reflects a biphasic function of this protein in biological events to control the biology of these cells.

This interaction with these surfaces appears to be decisive for the subsequent events of cellular performance and might even contribute to important biological events of matrix rearrangement through the activity of metalloproteinases (MMPs) [5, 7, 49, 50, 51]. Therefore, we evaluated ECM-related genes by considering the behavior of both gene and gelatinolytic activities of MMPs by zymography technology. We obtained significantly increased expression of Timp2 gene (Figure 14a) in all modified groups, especially in TiO2 + ALN + BSA; while the Mmp2 gene (Figure 14b) decreased compared to the control (Ti). Figure 14c illustrates an over expression of Mmp9 for all the modified surfaces, suggesting a balance between Mmp2 and Mmp9 in those conditions. The zymography assay for the activity of the MMPs indicated a significant increase of both MMPs activities in response to TiO2 + ALN group (Figure 14d-j; Figure S1). We suggest that the dynamic ECM remodeling may be a prerequisite to late biomineralization events, where ECM rearrangement is necessary for the nucleation of hydroxyapatite, the major mineral constituent of bone.

Figure 14

Extracellular matrix (ECM) remodeling after indirect contact with osteoblast responding to different biofunctionalized surfaces. The profile expression of the genes involved in the rearrangement of the extracellular matrix (ECM) - Timp2 (a), Mmp2 (b) and Mmp9 (c) were analyzed using qPCR technology. In addition, MMP activities were measured by the zymography technology (d; Fig. S1) and the analysis of those activities of both Mmp2 (e, g, i) and Mmp9 (f, h, j) are shown. The results are represented as the mean ± standard deviation from three independent experiments. A one-way ANOVA was performed with Tukey's test as a multiple comparison to evaluate the effects of different treated surfaces in the extracellular matrix, where ∗ for p < 0.05, ∗∗ for p < 0.01, ∗∗∗ for p < 0.001, and ∗∗∗∗ for p < 0.0001.

The extracellular matrix (ECM) provides a scaffold for cells, mainly considering adhesion, proliferation, and differentiation, and their remodeling is described as a fundamental activity to improve these functions [52]. Matrix metalloproteinases (MMPs) are zinc endopeptidases responsible for the regulation of various mechanisms including the remodeling of the extracellular matrix rearrangement [53]. During osteoblastic differentiation, extracellular remodeling is strictly controlled through the balanced activity of MMPs and their tissue inhibitors, Timps, which were very dynamic in response to the surfaces. In particular, Timp2 plays an important role in osteoblastic differentiation, and its expression positively modulates MMPS activity. The expression of Mmp2 and Mmp9 together with the enzymatic activity indicated remodeling of the ECM in the different modified surfaces, which was also demonstrated by another study [54].

Subsequently, the mineralizing phenotype of these cells was evaluated respecting the biomarker analysis of the osteoblastic differentiation, by examining the following genes: Runx2, Alkaline Phosphatase (Alp), Bone Sialoprotein (Bsp), Osteopontin (Spp1), Osteocalcin (Bglap2), and the Dentin Matrix Protein (Dmp1) (Figure 15). Our results show that Runx2 expression increased about 15-fold in response to TiO2 + ALN compared to control (Figure 15a), suggesting a strong involvement of this treatment on osteoblastic phenotype. A similar behavior in response to TiO2 + ALN (~3-fold change) was observed considering the Alp gene, which is an essential gene for bone mineralization events [55] (Figure 15b). The Bsp gene is related to the deposition of hydroxyapatite in the bone matrix, and its expression significantly increased on the surfaces of TiO2 + ALN and TiO2 + ALN + BSA (Figure 15c). Other genes were found to be involved with in this process such as Spp1 (Figure 15d) and Bglap2 (Figure 15e). The expression of Dmp1, which is responsible for the maturation of osteoblasts [56], decreased in response to TiO2 + ALN considering this biological model (Figure 15f).

Figure 15

Biofunctionalized titanium surfaces differentially promote osteoblastic-related genes. The cells were harvested in TriZOL for RNA extraction and subsequent cDNA synthesis. The Runx (a), Alp (b), Bsp(c), Spp1 (d), Bglap2 (e), and DMP1 (f) genes were each analyzed, and the results are represented as the mean ± standard deviation of three independent experiments. A one-way ANOVA was performed with Tukey's test as a multiple comparison to evaluate the effects of different treated surfaces on the genes related to the process and regulation of osteoblastic differentiation, where ∗∗ for p < 0.01, ∗∗∗, and ∗∗∗∗ for p < 0.0001.

Conclusion

Finally, despite the limitations of an in vitro study and the short time of cell-surface interaction considered in this study, we observed interesting osteogenic behavior on alendronate and albumin surfaces. Now in vivo analyses in needed to better consider these surfaces for clinical trials within biomedical area.

Declarations

Author contribution statement

Carolina S. Albano: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Wrote the paper.

Anderson M. Gomes, Geórgia da Silva Feltran, Célio Junior da Costa Fernandes: Performed the experiments; Wrote the paper.

Luciana Daniele Trino: Performed the experiments; Analyzed and interpreted the data; Wrote the paper.

Willian Fernando Zambuzzi, Paulo Noronha Lisboa-Filho: Conceived and designed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Funding statement

This work was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP: 2014/22689-3; 2017/02366-3 and 2017/15035-5) and .

Competing interest statement

The authors declare no conflict of interest.

Additional information

The data that support the findings of this study are available from the corresponding author upon reasonable request.