Heat or radiofrequency plasma glow discharge treatment of a titanium alloy stimulates osteoblast gene expression in the MC3T3 osteoprogenitor cell line

Although the outcomes of dental and orthopedic implant procedures are usually successful, in many instances the long-term stability and functionality of the implant cannot be achieved. Despite the reported long-term predictability of dental implants [1, 2], failures do occur in 10% of cases within a 5-year period [3]. The survival rates decrease to 71 to 83.5% over a 5 to 6 year period for dental implants placed in previously failed implant sites [4, 5]. The quality and quantity of bone that forms at the implant-skeletal interface is generally believed to be one of the major determinants of implant success. Therefore, improving fixation by enhancing the attachment and function of osteoblastic cells at the implant surface is likely to substantially decrease the likelihood of failure, especially for implants placed in previously failed implant sites.

Commercially pure titanium (cpTi) and titanium alloys are widely employed in orthopedic and dental implants due to their biocompatibility [6, 7]. Titanium-aluminum-vanadium alloy is used in the fabrication of prosthetic joint replacements. The Ti6Al4V alloy has also been successfully utilized in the dental field [8]. Both titanium and its alloys form an active oxide layer that readily interacts with extracellular matrix proteins produced by the cells and cell surface proteins. It is this superficial oxide found on both metals, titanium dioxide (TiO₂) being the most abundant, that provides an interface that is biocompatible with peri-implant tissues [9]. The surface oxide of Ti6Al4V is similar to that of pure titanium except that it is enriched with aluminum-oxide when present in air [10]. Altering the surface of titanium implant materials has been shown to affect protein adsorption, cell-substrate interactions, and tissue development [11]. However, the mechanisms by which surface oxide properties modulate the bioactivity of bound osteogenic proteins to influence osteoblast behavior and osseointegration are poorly understood.

A variety of methods have been explored to prevent implant failure including the use of bioactive adhesive peptides or extracellular matrix proteins such as fibronectin (FN) to facilitate the attachment of osteogenic cells to the implant surface [12-16]. Our laboratory and others have studied extracellular matrix proteins such as FN or human bone sialoprotein (hBSP) or hBSP peptides following their non-covalent adsorption [12-16] to the implant surface oxide to increase the adhesion of osteoblasts. FN is also thought to play an important role in skeletal development by regulating osteoblast differentiation and mineralization [17]. Most importantly, since we have shown that FN binds rapidly and irreversibly to TiO₂ [15], the protein can be efficiently adsorbed to titanium materials without the use of intervening chemical coupling agents. However, the precise influence of the metal surface oxide on the capacity of FN to modulate osteoblast attachment and differentiation has yet to be fully elucidated.

When added to implant materials, exogenous FN is likely to manifest a capacity to enhance osteoblast attachment/function or integration that is not obscured or overcome by the effects of serum FN for a number of reasons. Histological analyses of immediately loaded titanium dental implants demonstrated a direct contact between titanium and bone over the whole surface area of the implant immediately after insertion. No substantial blood clotting or inflammatory repair process was observed in the implant-tissue interface preceding bone cell adhesion [18]. Other studies have shown that soluble FN in serum, which appears more inert than the insoluble fibrillar form [19], does not bind to tooth surfaces [20], and a number of serum and matrix proteins can compete with FN for binding to bone [21, 22]. Therefore, due to immediate direct implant-bone contact after insertion, the low intrinsic binding activity of serum FN, and competition from other proteins for binding to the implant surface, it is unlikely that endogenous FN is able to bind to implant materials in vivo (or in vitro). These findings suggest that the effects of exogenous FN on bone cells and integration, when it is adsorbed to an implant surface, will supercede those of endogenous FN, which are likely to be minimal or negligible.

Our earlier studies examined the effects of modifying the surface oxide properties, such as oxide chemistry or topography, of Ti6Al4V on adsorbed FN's bioactivity toward osteoblasts [10, 23-25]. These studies demonstrated that pretreating Ti6Al4V surface oxide with heat or radiofrequency plasma glow discharge (RFGD) increased the oxide's negative net charge and dramatically increased the number of osteoblastic cells that attached to surface adsorbed FN [23, 24]. In a later investigation of the mechanism for the effects of pretreatments on FN bioactivity, we showed that modified Ti6Al4V disks increased the exposure of FN's integrin binding domain to enhance its binding to α₅β₁ integrin receptors in osteoblasts [25]. These results suggested that heat or RFGD pretreatments of Ti6Al4V amplified FN's intrinsic biological activity toward osteoblast integrins by inducing conformational changes in the matrix protein upon its adsorption. More recently, our laboratory has shown that when FN is precoated on Ti6Al4V disks prior to the addition of osteoprogenitors, the peak expression of all of the osteoblast gene markers analyzed was increased modestly in comparison to the uncoated disks [26]. These latter results suggest that the coated FN has the capacity to stimulate osteoblast differentiation.

In view of the findings that heat and RFGD pretreatments increased FN's integrin binding activity [24, 25] and cell spreading [24], the purpose of this study was to determine if these pretreatments also augment FN-promoted osteoblast differentiation as well as baseline differentiation. In this study, we have tested the hypothesis that surface pretreatments of Ti6Al4V, which make the surface oxide more negatively charged [23], increase FN-promoted osteoblast differentiation. We have also tested the sub-hypothesis that engineering the surface oxide to become more negatively charged exerts a general stimulatory effect on the bioactivities of other serum and matrix proteins to increase osteoblastogenesis without a FN coating. These hypotheses were tested by using quantitative real-time polymerase chain reaction (qRT-PCR) to measure changes in the mRNA levels for several osteoblast genes over time.

A key finding in this study is that, beginning on day 0, heat and RFGD pretreatments increased osteoblast gene marker expression for FN-coated disks more than uncoated disks, suggesting a specific enhancement of FN's bioactivity. Similarly, by day 3, pretreatments also increased the expression of all of the osteoblast gene markers in the absence of adsorbed FN (control). These results suggest that osteoblastic cells began to differentiate earlier on pretreated disks whether they were FN-coated or uncoated. This conclusion is also supported by our findings that the pretreatments down-regulated the mRNA expression of Runx2 and osterix, markers for osteoprogenitors [28] and preosteoblasts [28], respectively, for both coated and uncoated disks. Findings that Runx2 and osterix transcript levels were lower for coated than uncoated disks on day 0 further suggest that both of these genes became down-regulated earlier on disks that were pretreated and coated with FN. All of these findings collectively suggest that our surface oxide pretreatments by RFGD or heat accelerated the maturation of osteogenic precursor cells into osteoblasts.

The heat or RFGD pretreatment of FN-coated or uncoated disks also increased gene marker transcript levels during late osteoblastogenesis (14 to 28 days), suggesting that more osteoprogenitors differentiated and attained peak levels of gene marker expression earlier on modified disks. The overall greatest effects were observed for pretreated samples coated with FN compared to pretreated and uncoated controls, supporting the specific stimulation of adsorbed FN's bioactivity. Our laboratory has reported that FN coated on untreated Ti6Al4V disks modestly increased the levels of osteoblast gene marker expression during the later stage (28 days) of development compared to uncoated disks [26]. In the current study, surface pretreatments combined with a FN coating increased osteoblast gene expression during the later phase of development (days 14 to 28) to levels that were far greater than those measured in the presence of FN alone (with no pretreatment). Therefore, these results suggest that our pretreated implant surfaces stimulate osteoblast differentiation by activating precoated FN and perhaps other adsorbed osteogenic serum and matrix proteins (even in the absence of a FN coating).

Our laboratory has investigated the influence of Ti6Al4V oxide physicochemical properties on the bioactivity of adsorbed FN [10, 23, 24]. Using atomic force microscopy, we previously showed that the preheated (600℃) surface oxide exhibited root mean square roughness values (12.8±1.7 nm) that were three-fold higher than those of untreated (4.1±1.1 nm) or RFGD-treated (3.6±0.9 nm) disks [23]. Heat pretreatment altered oxide topography by creating a pattern of oxide elevations that were approximately 50 to 100 nm in diameter [23]. In contrast, the RFGD pretreatment of Ti6Al4V disks did not alter the topography of the alloy surface (Fig. 1) since the RFGD-pretreated surface appeared to be at least as smooth as the control polished surface [23].

Both heat and RFGD pretreatments altered the Ti6Al4V oxide's net surface charge [23]. In these latter experiments, the electrostatic forces between the oxide surface and a negatively charged silicon nitride probe were used to estimate the net charge of the oxide [23]. Atomic force spectroscopy measurements demonstrated that both the RFGD and heat pretreatments made the surface oxide of Ti6Al4V more negatively charged at physiological pH [23]. The local metal oxide charge can be influenced by the acid-base balance of metal-hydroxo complexes [29]. The effects of the pretreatments on the net negative charge of the oxide at physiological pH are likely to have arisen from an elevation in the surface concentration of uncompensated negative charges from [M-O]^- groups. In addition, we have demonstrated that heat and RFGD pretreatment induced increases in adsorbed FN's cell attachment activity [24], the conformational exposure of its integrin-binding domain [25], and binding to α₅β₁ integrins [25]. Moreover, all of these parameters were more highly correlated with changes in oxide charge rather than roughness or topography [24, 25]. Taken together, these findings suggest that charged functional groups in the Ti6Al4V surface oxide can modulate the conformation of adsorbed FN, thereby increasing its capacity to bind to α₅β₁ integrins. In the current study, we showed that heat and RFGD pretreatments produced qualitatively similar effects on FN's capacity to induce a panel of osteoblast gene markers that were more highly correlated with increases in oxide charge than with oxide topography or roughness. These findings indicate that increases in oxide charge promoted by either pretreatment may enhance FN's capacity to promote osteoblast differentiation.

A number of studies have investigated the effects of non-metallic surface charge on FN structure and bioactivity. One study compared the effects of varying surface functionalities in self-assembled monolayers (SAMs) of alkanethiol on adsorbed FN's bioactivity toward MC3T3 cells [30]. The study found that FN displayed the greatest exposure of its integrin binding domain and binding to α₅β₁ integrins on a surface functionalized with hydroxyl groups (OH) with a negative charge compared to surfaces containing positively charged or hydrophobic functionalities [30]. It has been observed that a specific increase in FN binding to α₅β₁ integrins is associated with a switch in the cellular program from proliferation to differentiation [31]. Keselowsky et al. [32] showed that a OH-functionalized SAM surface also induced the highest levels of FN-stimulated focal adhesion kinase (FAK) signaling, which is mediated via integrin receptor activation. FAK can regulate a number of cell functions including differentiation [33]. In addition, OH-functionalized surfaces coated with FN have been shown to up regulate osteoblast-specific gene expression to a greater degree than other substrates [34]. In the latter study, the HFN7.1 monoclonal antibody, which blocks the binding of α₅β₁ integrins to the central integrin binding domain of FN, completely inhibited MC3T3 osteoblast differentiation and mineralization on FN-coated OH-functionalized surfaces [34]. We have also shown that the HFN7.1 antibody completely inhibited MC3T3 cell attachment to FN-coated heat or RFGD-treated disks [25], thereby complicating any measurements of osteoblast differentiation or mineralization. These studies suggest that [M-O]^- groups in the Ti6Al4V oxide modulate FN-mediated osteoblast differentiation by increasing the protein's binding to α₅β₁ integrins and activation of integrin-dependent cell signaling pathways. Whether or not negatively charged oxides also modulate the conformational bioactivities of other serum or matrix proteins awaits further investigation.

We have previously shown that MC3T3 osteoprogenitor cells cultured on a rigorously cleaned Ti6Al4V alloy were found to demonstrate a normal pattern of osteoblast differentiation [26] similar to that observed on a polystyrene substrate. Nevertheless, this and another study that employed surface cleaning methods to increase hydrophilicity [35] provide only a limited understanding of how surface oxide chemistry influences osteoblast gene expression. In comparison, the current study showed that RFGD pretreatment, which increased oxide charge without altering topography [23], was able to stimulate osteoblast differentiation even after surface contamination was removed by our rigorous cleaning protocol. By using RFGD treatment to selectively perturb oxide chemistry without affecting topography, our study has provided insight into the mechanism by which negatively charged Ti6Al4V oxide groups influence protein bioactivity and osteoblast differentiation. Our findings show that by engineering a surface oxide to become more negatively charged, an implant surface can be created that promotes osteoblast differentiation by enhancing the biological activities of adsorbed osteogenic proteins.

The quantitative differences between our two pretreatments (heat and RFGD) and their effects on the expression of osteoblast gene markers may have arisen from the unique physico-chemical characteristics of each resultant oxide surface. Heat pretreatment, for example, had greater effects on the mRNA expression of genes for PTH-rP, alkaline phosphatase, and osteocalcin, while RFGD pretreatment had greater effects on osteopontin and bone sialoprotein gene expression. These findings suggest that structural oxide properties such as roughness or topography that are altered by heat pretreatment may induce some osteoblast gene markers while decreasing the expression of others.

A number of studies have examined the influence of titanium surface topography on the capacity of osteoblastic cells to differentiate in vitro. TiO₂ surfaces containing nanofibers [36] or nanotubes [37] facilitated a higher cellular differentiation capacity than surfaces devoid of nanostructures. Zhao et al. [38] found that the microroughening of Ti6Al4V surfaces by grit-blasting with hydroxyapatite/beta tricalcium phosphate particles markedly increased the expression of osteocalcin but decreased alkaline phosphatase activity. In contrast, another study reported that microstructured titanium surfaces created by sandblasting/acid etching increased both alkaline phosphatase and osteocalcin expression in osteogenic cells compared to smoother machined surfaces [35]. The expression of alkaline phosphatase and osteocalcin was even further enhanced when the microstructured surfaces were made more hydrophilic by reducing surface contamination [35]. These findings suggest that changes in surface oxide chemistry, topography, and micro and/or nanoroughness interact in a complex manner to cooperatively alter the expression of individual osteoblast gene markers. Notably, variations in the expression of specific extracellular matrix genes such as those for type I collagen (α1), osteopontin, and bone sialoprotein, alter the protein composition of the matrix and the biochemical cues it provides for osteoblast gene marker expression and osteoblast function. Interactive effects of oxide chemistry and structure on protein conformational bioactivity, matrix protein synthesis/bioactivity, and osteoblast behavior may underlie the differences in gene marker expression observed in the current study between RFGD and heat-pretreated Ti6Al4V.

In summary, our results suggest that heat and RFGD pretreatments of the FN-coated Ti6Al4V surface oxide stimulated osteoblast differentiation through an enhancement of (a) FN's bioactivity and (b) the bioactivities of other serum and matrix proteins. These findings also suggest that negatively charged oxides can increase the capacity of coated FN, and perhaps other adsorbed osteogenic proteins, to promote osteoblast differentiation. Heat and RFGD pretreatments had different quantitative effects on osteoblast gene expression that may be related to possible effects of heat pretreatment-induced increases in oxide roughness. Our studies suggest that by combining FN coatings of Ti6Al4V with pretreatments that either alter surface oxide charge (RFGD) or charge and roughness (heat), an implantable surface can be developed that will more effectively support osteoblast differentiation. The investigation of how multiple surface oxide properties interact with selected matrix proteins or peptides to affect bone cell behavior may reveal novel strategies to improve implant fixation during the critical early healing phase.