Effects of titanium(IV) ions on human monocyte-derived dendritic cells

Abstract

Orthopaedic metal implants composed of titanium are routinely used in bone fracture repair and for joint replacement therapies. A considerable fraction of implant recipients are unable to benefit due to implant failure resulting from aseptic loosening, while others may experience cutaneous sensitivity to titanium after implantation. An adaptive immune reactivity towards titanium ions, originating from the biocorrosion of the implants, could play a role. As an initiator of the adaptive immune response, dendritic cells (DC) were studied for uptake and characteristics after titanium exposure. Energy filtered transmission electron microscopy showed uptake of titanium(IV) (Ti(IV)) ions by DCsin vitro and co-localisation with phosphorus-rich cell structures of the DC membranes (phospholipids), cytoplasm (ribosomes and phosphorylated proteins) and the nucleus (DNA). DC maturation and function were investigated by measuring cell surface marker expression by flow cytometry. After exposure, DCs showed a decrease in MHC class II (HLA-DR), co-stimulatory molecules (CD40, CD80 & CD86) and chemokine receptors (CCR) 6 and CCR7 but an increase in CCR4 after Ti(IV) treatment. However, Ti(IV) treated DCs had an increased stimulatory capacity towards allogenic lymphocytes. A Ti(IV) concentration dependant increase of IL-12p70 was observed amidst decrease of the other measured cytokines (TGF-β1 and TGF-β2). Hence, Ti(IV) alters DC properties, resulting in an enhanced T lymphocyte reactivity and deviation towards a Th1 type immune response. This effect may be responsible for the inflammatory side effects of titanium implants seen in patients.

---

Introduction

The biocorrosion¹ of titanium-based metal implants used in orthopaedic treatments results in the release of both wear particles^2,3 and metal ions.^4,5 Biological degradation of metal implants by cells like osteoclasts,⁶ release sub-particle sized metal ions into the surrounding tissue and bloodstream, which facilitates their distribution and accumulation in distal organs like the liver, spleen, lungs and lymph nodes.^7,8 Elevated levels of titanium in the blood (67 ng ml⁻¹) and joint capsule (1540 ng ml⁻¹) have been detected in patients with a total joint arthroplasty and are elevated (19 173 ng ml⁻¹) in patients with failed implants.⁹ Specifically, Ti(IV) has been described as the most abundant species present and most frequently available ion after biocorrosion.¹⁰Titanium ions released into the peri-implant tissue can exist in the body as free ions but it is more likely for them to bind to proteins due to their unstable nature.^11,12 It is still unclear how such sub-toxic levels of Ti(IV) ions affect the biological activity of cells.

One of the main problems with metal implants is aseptic loosening, which affects 5–20% of implant recipients.¹³ Although the definitive cause of these implant side effects is uncertain, several possibilities have been identified. It has been suggested that an inflammatory reactivity contributes to this event and that metal hypersensitivity can also result after the implantation of titanium.¹⁴ It has been clinically observed and documented that titanemia¹⁵ and titanium hypersensitivity^16–18 occur in patients after implantation of titanium prostheses. Biopsies of tissues surrounding implant areas show increased levels of macrophages resulting in foreign body reactions as well as B and T lymphocyte infiltration.^19,20Hypersensitivity has been linked to the reactivity of Th1 cells through the measurement of patients’ lymphocyte production of inflammatory cytokines in culture.²¹ In most cases, the removal of titanium implants alleviates the hypersensitivity reactions and the patients do not experience further problems.²²

Because of the specificity of the reactivity, it indicates the role of the adaptive immune reactivity, which is often controlled by the dendritic cells (DC). DCs are central to the establishment of a specific, controlled, sustained and lifelong immunological memory.²³ Characteristically, DCs continually sample their microenvironment, selectively take up antigens according to other microenvironment stimuli (LPS, Danger signals), and act as reservoirs for antigens and present processed antigens to naïve T lymphocytes.^24,25 Mature inflammatory DCs are able to drive the proliferation of naïve T lymphocytes and their differentiation into inflammatory effectorT lymphocytes. While, on the opposite scale, steady state DCs would not cause T lymphocyte proliferation and would be involved in the development of peripheral metal tolerance through adaptive regulatory T lymphocytes (Treg).^26,27 Currently, not much is understood about the immunogenicity of titanium ions on the different components of the immune system. It is likely that a chronic or delayed immune reactivity observed in patients would need to be initiated by professional antigen presenting cells like dendritic cells (DCs) and effected by specialized T lymphocytes of CD4 lineage.^28,29 Both of these immune cell types have also been described to be present in the peri-implant tissue, which would suggest recruitment is a role in implant-related inflammation.³⁰

To investigate the role of the adaptive immune response in metal hypersensitivity leading to aseptic loosening, this study focuses on the uptake of Ti(IV) by human monocyte-derived DCs and the changes in their characteristics (antigen presentation, co-stimulation, migratory capability) and function (lymphocyte stimulatory capacity, cytokine and chemokine expression).

Experimental

Isolation of peripheral blood monocytes (PBMCs)

PBMCs were obtained from the Ficoll-Plaque (Amersham Biosciences, Uppsala, Sweden) separation of healthy human whole blood buffy coats (Australian Red Cross Blood Service, Perth, Western Australia). The monocyte layer was removed and washed in 0.1 M Phosphate Buffered Solution (PBS) pH 7.2 (Gibco/Invitrogen, Auckland New Zealand). Cells were placed in 25 cm³ culture flasks (Sarstedt, Nuembrecht, Germany) to allow for adherent cells to adhere to the flask for one hour. The non-adherent cells were removed and the adherent cells left in the flask were washed twice with PBS.

Culture of human monocyte-derived dendritic cells

Adherent cells were cultured in RPMI 1640 Glutamax (Gibco/Invitrogen), supplemented with 5% allogeneic human serum, antibiotics (10 000 units mL⁻¹Penicillin G sodium, 10 000 μg mL⁻¹Streptomycin sulfate and 25 μg mL⁻¹ Amphotericin B, Gibco/Invitrogen). Cytokines to differentiate monocytes into DCs were added; recombinant human granulocytemacrophage colony-stimulating factor (GM-CSF; 10 ng ml⁻¹; R&D Systems, Minneapolis, MN) and interleukin 4 (IL-4; 10 ng ml⁻¹; R&D System), per 5 ml of culture. After 5 to 7 days of incubation (37 °C, humidified, 5% CO₂), the monocyte-derived DCs were used for further treatments.

Exposure to metal ions

TiCl₄ (mass spectrometric purity, Fluka, Buchs, Switzerland) was used as the source of Ti(IV) ions. TiCl₄ was diluted (100–1000 μM) in human serum to allow for the formation of possible metal–protein complexes. There was no titanium precipitate seen. DCs were treated with the Ti(IV) ions or human serum only (negative control) up to 3 days. The treated DCs were then washed twice in PBS to remove excess metal ions, thereby preparing them for further analysis.

Viability of dendritic cells

The viability of DCs after exposure to various concentrations of Ti(IV) was assessed through cellular mitochondrial activity. Mitochondrial activity was assessed with the CellTiter96® Aqueous One Cell Proliferation Assay (Promega, Madison, WI). The colourimetric reaction was read at an absorbance of 490 nm with an ELISA reader (Labsystems Multiskan RC, Finland). One sample was measured in triplicate. Data was collected and analysed with Prism v4.0 (GraphPad Software, La Jolla, CA).

Energy-filtered transmission electron microscopy and electron energy loss spectrometry

After incubation with 1 mM Ti(IV) for 4 hours to 3 days, the treated DCs were washed twice in fresh culture medium and immediately fixed in culture medium containing 2.5% EM grade glutaraldehyde (EMS, Washington, PA). The fixation was carried out at room temperature for 15 min and no osmium post-fixation was used. After a brief wash in culture medium, the cells were transferred into 70% ethanol before being transferred into hard grade LR White resin (ProSciTech, Thuringowa, QLD, Australia) at room temperature for overnight incubation. The LR White was replaced by a fresh one and incubated for 1 hour at room temperature before being polymerized at 50 °C for at least 24 hours. 30 nm ultra-thin sections were cut on a Leica Ultracut UCT ultramicrotome, collected on 1000-mesh copper grids and coated with a carbon layer (5 nm) to stabilize the sections before they were analysed. The sections were examined using a 300 kV JEOL 3000F field-emission TEM equipped with a post-column Gatan Imaging Filter, a TV-rate retractable camera and Multi Scan digital camera (Gatan, Pleasanton, CA). A combination of 150 μm condenser and 60 μm objective apertures was used. The selection of the appropriate cellular region was achieved using the TV-rate retractable camera and dark field microscopy mode. For this purpose, the inelastic contrast imaging was done using settings of energy losses between 100 eV to 150 eV and slit width of 20 eV. A three windows method was used for the elemental mapping. The two background windows (pre-edge 1 and pre-edge 2), centred at E = 428 eV and E = 436 eV, respectively, have been selected carefully to avoid overlapping with the extended tail of the nitrogen-specific edge. The titanium-specific window (post-edge) was centred at E = 464 eV (L2,3-titanium ionization edge). An energy selecting slit width of 8 eV was used. To determine the intracellular distribution of titanium relative to the naturally existing cellular elements, the same cellular region that was mapped for titanium was also mapped for carbon, chlorine, nitrogen, oxygen, phosphorus and sulfur. To confirm the specificity of the titanium map, a titanium jump ratio map was obtained, simply by dividing the post-edge by pre-edge 2. All images (256 × 256 pixels) were recorded at a nominal magnification of 10 000× with a CCD camera using the viewing and imaging software DigitalMicrograph (Gatan). Electron energy loss spectra (EELS) were recorded from the same cellular region used for elemental mapping. The spectrum and energy dispersion were selected such that only nitrogen, titanium, and oxygen ionization edges were visible in the spectrum.

Proliferation assay

BrdU incorporation assays (Roche Diagnostics, Basel, Switzerland) were used to assess the proliferation of non-adherent PBMCs (naPBMC) exposed to titanium-treated DCs. Allogenic human lymphocytes from the non-adherent fraction were used containing more than 80% T lymphocytes. The assay was done on 96-well plates (Sarstedt). Maximal proliferation was achieved with the addition of 50 μg mL⁻¹ Phytohaemagglutinin (PHA, Sigma, St Louis, MO). After 6 days of incubation, absorbance of the colourimetric reaction was measured at 405 nm on an ELISA reader (Labsystems Multiskan RC). Data was collected and analysed with Prism v4.0 (GraphPad Software, La Jolla, CA, USA). Three replicates were measured and a single representative data shown. T-tests were used to determine significance when p < 0.05.

Flow cytometry and data analysis

The analysis of the DC surface molecules was made with flow cytometry.³³DCs were stained with fluorescently labeled mouse anti-human antibodies. HLA-DR-PE, CD45-PerCP, CD11c-PE, CD25-APC, CD80-PE, CD54-PECy5 (BD Pharmingen, San Jose, CA), CCR4-PE, CCR6-FITC, CCR7-APC (R&D Systems). Unlabelled mouse anti-human CD83 (Novocastra, Newcastle upon Tyne, UK) and CD86 (Exalpha Biological, Shirley, MA) antibodies were indirectly labeled with a secondary donkey anti-mouse antibody with Alexa488 fluorescent (Molecular Probes/Invitrogen). Unstained controls were used for the establishment of background fluorescence. For each stain, the antibodies were incubated for 15 minutes and washed with PBS. Cells were fixed in 3% paraformaldehyde + 2% sucrose before sample analysis on an FACSCanto II (BD Systems) and data acquisition on an FACSdiva (CMCA: Centre for Microscopy, Characterisation and Analysis). A total of 10 000 gated cells were captured for the analysis. Results are representative and cumulative of seven samples. Analysis of FACS data was done on a FlowJo v8.7 (TreeStar, Ashland, OR). DCs were gated from the forward (FSC) and side scatter (SSC) plots, followed by CD45 positive cells where possible. Representative histograms of several samples show the expression of individual surface markers.

Cytokine and chemokine measurements

Supernatants were collected from 8 separate samples and cytokine levels measured. IL-1β, IL-4, IL-6, IL-10, IL-12p70 and TNF were detected by Cytometric Bead Assay (BD Biosciences). The processed samples were then measured on the BD FACSCanto II flow cytometer (CMCA). The results were analysed with FCAP Array v1.0 Software (Softflow Technologies, New Brighton, MN). TGF-β1, 2, 3 were measured in duplicates on a Beadlyte Detection System (Upstate/Millipore, North Ryde, NSW, Australia) kit. The Luminex Plate Reader (Luminex, Austin. TX, USA) was used for detection and obtaining readouts. Chemokines CCL17 and CCL22 were measured (in duplicate) with Quantikine^® kits (R&D Systems) and read out at 450 nm on an ELISA reader (Labsystems Multiskan RC). All cytokine and chemokine concentrations were measured in pg mL⁻¹. All samples were prepared according to the manufacturer’s instructions.

Statistical analysis

Data was analysed for significance with t-tests and significant when p < 0.05. Graphs with error bars indicate the standard deviation (SD).

Results

Cell viability and working concentrations of metal ions

To establish a working concentration and in vitro environment mimicking the presence of Ti(IV), the DC viability was assessed after exposure to various concentrations of Ti(IV). After 18 hours of exposure to Ti(IV) for up to 1000 μM, viability of treated DC was comparable to untreated cells (Fig. 1). Hence, it was decided that concentrations up to 100 μM were suitable for this study.

Fig. 1 Dendritic cell viability in various concentrations of Ti(IV) measured through MTT. Triplicate measurements of one sample (n = 1) were done and error bars indicate standard deviation. Higher absorbance indicates higher viability of cells.

Detection of Ti(IV) in dendritic cells by EFTEM and EELS

To investigate Ti(IV) uptake and cellular distribution, DCs were incubated between 4 hours and 3 days with Ti(IV) before being processed to ultra-thin sections (about 30 nm) for energy-filtered transmission electron microscopy (EFTEM) as described above in the methods section. Using previously optimised microscopy settings, including an energy slit width of 8 eV and the titanium specific post-edge energy window (E = 464 eV), as well as 2 pre-edge background energy windows (E1 = 428 eV, E2 = 436 eV), titanium maps were collected. In addition, element maps of the same cellular area were taken, including carbon, nitrogen, oxygen, phosphorus and sulfur. The presence of titanium in the sample was confirmed through electron energy loss spectroscopy (EELS) of the same cellular area. One representative example is presented in Fig. 2E and F for Ti(IV). Element mapping of the cellular area of interest helped with orientation and localisation of subcellular structures. Carbon maps (Fig. 2A) clearly indicated cellular membrane structures, including the surface cell membrane, the nuclear membranes and membranes of cytoplasmic organelles. Nitrogen maps (Fig. 2B) showed mainly areas where proteins or nucleic acid containing structures were located. Phosphorus maps (Fig. 2C) indicated phospholipids containing membrane structures and phosphate groups containing DNA and RNA, showing the surface cellular membrane, the nuclear membranes, ribosomes and the nuclear heterochromatin. Consequently, the element mapping resulted in morphological information without having to stain with heavy metals as used for conventional transmission electron microscopy. Ti(IV) was homogenously co-localised with phosphorus in the membranes, as well as in RNA and DNA containing structures (Fig. 2D). Prolonged incubation time increased the amount of detected titanium, resulting in a stronger EFTEM signal and increased titanium peak in the EELS measurement, indicating that titanium accumulates over time in the cells.

Fig. 2 EFTEM and EELS of Ti(IV) treated dendritic cells. 2A–D show representative examples of elemental mapping of one selected area of a dendritic cell using EFTEM. Bright/white areas indicate the presence of the element of interest, i.e. carbon (2A), nitrogen (2B), phosphorus (2C) and titanium (2D). Arrows indicate co-localisation of phosphorus and titanium in the cell membrane and cytoplasmic structures, arrow heads in the nucleus. 2E and 2F show the electron energy loss spectrum corresponding to the EFTEM analysed cellular area, 2F being the extracted and normalised spectrum of 2E. In the spectrum, the presence of nitrogen, titanium and oxygen at their specific energies, as detected in the represented cellular area, are indicated with arrows. The edge size of the EFTEM images corresponds to 1.25 μm.

Stimulatory potential of dendritic cells

Ti(IV) treated DCs were used to stimulate freshly isolated allogenic naPBMC in a proliferation assay (Fig. 3) by assessing the BrdU incorporation. The assay was done at the ratio of 1 DC : 10 naPBMCs. The proliferation of the naPBMCs in the presence of Ti(IV) treatments was significantly higher (p = 0.0080) than the untreated DCs. The proliferation rate without PHA was equivalent to that seen when PHA was added to help the lymphocytes in the naPBMCs to achieve maximal proliferation.

Fig. 3 Mixed leukocyte reactions. Ti(IV) treated or non-treated DCs were cultured together with allogenic non-adherent PBMC in the presence or absence of PHA. Note the enhanced proliferation of the allogenic PBMC when cultured with Ti(IV) treated dendritic cells. Representative graph shown with error bars indicating standard deviation of a single sample. Higher absorbance indicates higher proliferation of lymphocytes. Significant changes are marked with an asterisk (*).

DC surface marker expression

After treatment with Ti(IV), DCs were analysed for their surface marker expression (Fig. 4). As one of the main functions of DCs is to present captured antigens on MHC class II molecules, the detection of HLA-DR was done. When measured, Ti(IV) treated DCs had lower expression of HLA-DR than untreated DCs. Besides antigen presentation, DCs provide the necessary combination of co-stimulatory signals to naïve T lymphocytes, enabling them to differentiate into their respective effector subsets. CD80/86 (B7 family) on DCs would conjugate with CD28 on T lymphocytes.^34–36Ti(IV) treated DCs down-regulated CD80, CD86 co-receptor and CD40 expression. CD54, an adhesion molecule which helps in the formation of tight immunological synapses between DCs and naïve T-cells, was reduced in Ti(IV) treated DCs. CD25 (IL-2 receptor-α) is important in regulating the production of IL-2 by DCs^37,38 as IL-2 is influential in the development of an inflammatory immune responsevia Th1 pathways.³⁹ However, there was no noticeable influence of titanium on CD25 expression, which remained elevated similar to the untreated DCs. Chemokine receptors were measured to give an indication of the maturation state and migratory potential of the Ti(IV) treated DCs. Thus, alterations in the expression of chemokine receptors indicate the chemokine responsiveness of the Ti(IV) treated DCs and their potential to migrate out or remain in the peri-implant tissue.^40–43 CCR4 expression on Ti(IV) treated DCs was slightly higher than untreated DCs. DCs treated with Ti(IV) showed reduced expression of CCR7 to non-detectable levels. Expression of CCR6 in the treatment was very low in all the treatments. Summarizing, titanium has peculiar effects on the expression of a variety of surface markers resulting in patterns that make it difficult to explain the enhanced T-cell proliferative capacity of titanium treated DCs.

Fig. 4 Flow cytometry. Representative histograms of key dendritic cell markers.

Cytokine and chemokine expression

Cytokines indicative of inflammation (IL-1β, IL-6, IL-12, TNFα), humoral immunity (IL-4, IL-10) and immune regulation (IL-10 and TGF-β1, 2, 3) were measured to define the influence of Ti(IV) treated DCs on the outcome of the immune response (Fig. 5). Production of the Th1 cytokine IL-12 increased significantly (p = 0.0151) in Ti(IV) treated DCs, while all other cytokines (IL-1β, IL-4, IL-6, TNFα, TGF-β3) measured were either not significantly influenced by Ti(IV) or significantly decreased (IL-10: p = 0.029; TGF-β1: p = 0.0102; TGF-β2: p = 0.0057). Variation among the inter-individual samples can be seen in the measurements. Chemokines specific for CCR4, CCL17/TARC and CCL22/MDC remained at high levels (CCL17 > 500 pg mL⁻¹ and CCL22 > 3000 pg mL⁻¹) across all samples and treatments, thus Ti(IV) treatment did not influence secretion of those chemokines by DCs (Data not shown).

Fig. 5 Cytokine secretion by Ti(IV) treated and untreated dendritic cells, as detected in culture supernatant. Significant changes are marked with an asterisk (*).

Discussion

Aseptic loosening and metal hypersensitivity are two major problems affecting patients with orthopaedic metal implants. Inflammatory and immune-related responses caused by titanium have been well documented in humans.^31,32 Previous studies have investigated mainly the effects of larger titanium particulates and little is known about the role of titanium ions in immune reactions. In that respect, DCs reacting to ions from the biocorrosion of titanium implants could play a significant role.³³ This study investigated the influence of Ti(IV) ions on DC phenotype and function after uptake.

Firstly, we were interested in investigating Ti(IV) ion uptake by DC and the sub-cellular distribution. More recently, significant improvements in energy filtered transmission electron microscopy (EFTEM) and electron energy loss spectrometry have become promising methods for elemental detection and mapping of metals, not only in materials science, but also in biology.^34–37 This is the first study investigating titanium ion uptake and distribution by human dendritic cellsin vitro using EFTEM. The results show that Ti(IV) is taken up by DCs and accumulates in the cells. Ti(IV) bound homogeneously to phosphorus-containing molecular structures of membranes, the cytoplasm and the nucleus, suggesting binding to phospholipids, RNA and DNA. In addition, titanium has a strong binding affinity for phosphorylated proteins,³⁸ as titanium dioxide (TiO₂) has recently been introduced as a new method for enrichment and further characterization of phosphorylated peptides and proteins using mass spectrometry.^39,40 Once within the cell, the Ti(IV) ions would bind to phosphorylated proteins involved in all manner of cellular activity ranging from gene regulation to intracellular signaling, all of which if altered by the binding of titanium can affect the metabolic processes of a cell.⁴¹ In the context of phosphorylated intra- and extracellularprotein complexes, they may well be processed by DCs towards titanium containing phosphorylated peptides with new antigenic properties. Those new complexes may be presented on surface MHC molecules similar to simple phosphorylated peptides as already described by Zarling et al.⁴²

The increase in proliferative capacity by Ti(IV) treated DCs would support such a possible route of an antigenic titanium complex loaded on surface MHC. However, the flow cytometry analysis of cell surface markers of Ti(IV) treated DCs showed a reduced level of MHC class II molecules, with reduced co-stimulatory molecules of CD80/CD86 and cell-to-cell adhesion (via CD54). This would indicate that Ti(IV) treated DCs would form weaker immunological synapses with T lymphocytes as well as providing decreased antigenic and co-stimulation signals, both necessary for classic T lymphocyte proliferation. However, Ti(IV) treated DC enhanced T lymphocyte proliferation in mixed leucocyte cultures. Thus, it is likely that other mechanisms exist to stimulate the proliferation of T lymphocytes. Enrichment of titanium ions in the cell membrane could cause the stiffening of the membrane, changing the fluidity of the DC membrane and affecting cell-to-cell interactions.⁴³

Ti(IV) treated DCs showed reduced CCR7 expression, suggesting a reduced ability of these DCs to mature and migrate towards the lymph nodes after exposure to Ti(IV) ions. In vivo, it could perhaps allow the Ti(IV) matured DC to remain in the tissue, awaiting titanium-specific lymphocytes to enter and be re-activated with a stronger signal from both the surrounding antigenic titanium–protein complexes and Ti(IV) challenged DCs. Most interestingly, CCR4 was highly expressed in Ti(IV) treated DCs. CCR4 is the receptor for CCL17 and CCL22,^44,45 both of which are produced by cells of the skin.⁴⁶ While CCR4 expression is commonly associated with skin homing T lymphocytes, the up-regulation of CCR4 on Ti(IV) treated DCs might suggest a migratory potential towards the skin. This would give support to the clinical observations of some patients experiencing skin sensitivity towards titanium after implantation of titanium-based implants.^47–51 This could also explain the variable results seen during patch testing. Patch tests are commonly used tools to assess the sensitivity of patients to metals before implantation.⁵ It might also lead to sensitivity to titanium ions applied on the skin via sun-block lotions, cosmetics or deodorants⁵² after metal implantation. Alternatively, CCR4 + DCs would also remain in the peri-implant tissue due to production of CCL22 by RANKL induced osteoclasts,⁵³ activated macrophages⁵⁴ and even by DCs themselves.⁵⁵ However, as the DCs are unable to move out of the peri-implant area, they must await the incoming of CCR4+ T lymphocytes. This is also supported by our data, as the production of CCL17 and CCL22 levels remained high in DCs after Ti(IV) treatment, which makes it a possibility. CCR4+ T lymphocytes may be continually drawn in to the implant site⁵⁶ resulting in already described lymphocytic tissue infiltration.^19,20,54,57 Currently, CCR4+ T lymphocytes are mainly known to occur during atopic dermatitis⁵⁸ and contact hypersensitivity.⁵⁹ However, they seem to play a major role in rheumatoid arthritis and related diseases.⁵⁷ Our recent research has also shown that exposure to Ti(IV) induces upregulation of CCR4 on human T lymphocytes.⁶⁰ Together with these results, it is tempting to postulate that CCR4 and its ligands CCL17 and CCL22 play a major role in the local titanium-related inflammatory reaction.

The type of immune response (inflammatory, humoral or regulatory) is determined by both the stimulatory support and the cytokines secreted by mature DCs. Untreated DCs would be an immature subtype. Their cytokine profile is different because of the biological variation and inter-individual differences commonly seen in human studies. The cytokine analysis showed a trend of increased secretion of IL-12 by Ti(IV) treated DCs. IL-12 is an important driving force for a Th1 type inflammatory response. Usually, the Th1 response is counter-regulated by IL-10, but Ti(IV) decreased IL-10 secretion in exposed DCs. Also, the decrease in the suppressive cytokine TGF-β1 and TGF-β2 suggests that regulatory mechanisms are blocked by Ti(IV) exposure. However, the Th1 response has been clearly shown to be predominant in patients with failed implants, where high levels of IFN-γ were detected, supporting our in vitro data.²¹

Conclusion

This study has shown that DCs take up Ti(IV) ionsand they bind specifically to phosphorus-containing molecules. The resulting phenotype and function of DCs is affected by the uptake of Ti(IV) ionsthrough alterations in the surface marker expression and its cytokine production. Currently, the pathways that titanium ions alter are unclear. Hence, titanium ions have an effect on the maturation and possibly migration of DCs and drive an inflammatory titanium-related Th1 immune response. This would indicate that DCs play a role in the development of metal hypersensitivity.

Acknowledgements

We thank Steve Parkinson, Steve Parry and John Murphy for their excellent technical support. Also the facilities, scientific and technical assistance of the Australian Microscopy & Microanalysis Research Facility at the Centre for Microscopy, Characterisation & Analysis, The University of Western Australia, a facility funded by The University, State and Commonwealth Governments. This work was supported by National Institutes of Health Grant GM072726 and the AO Foundation Grant 05Z34.

References

1. J. J. Jacobs, J. L. Gilbert and R. M. Urban, J Bone Jt Surg., Am. Vol., 1998, 80, 268–282.
2. J. J. Jacobs and N. J. Hallab, J Bone Jt Surg., Am. Vol., 2006, 88, 1171–1172.
3. E. Ingham and J. Fisher, Proc. Inst. Mech. Eng., Part H, 2000, 214, 21–37.
4. L. Büdinger and M. Hertl, Allergy, 2000, 55, 108–115.
5. N. Hallab, K. Merritt and J. J. Jacobs, J Bone Jt Surg., Am. Vol., 2001, 83, 428–436.
6. L. Filgueira, J. Meagher, P. Clode, M. Saunders and R. Zellweger, Eur. Cells Mater., 2007, 13, 30.
7. D. Olmedo, M. B. Guglielmotti and R. L. Cabrini, J. Mater. Sci.: Mater. Med., 2004, 13, 793–796.
8. R. M. Urban, J. J. Jacobs, M. J. Tomlinson, J. Gavrilovic, J. Black and M. Peoc'h, J Bone Jt Surg., Am. Vol., 2000, 82, 457–476.
9. N. J. Hallab, K. Mikecz, C. Vermes, A. Skipor and J. J. Jacobs, Mol. Cell. Biochem., 2001, 222, 127–136.
10. P. Boscolo, L. Di Giampaolo, M. Reale, M. L. Castellani, A. R. Volpe, M. Carmignani, J. Ponti, R. Paganelli, E. Sabbioni, P. Conti and M. Di Gioacchino, Ann. Clin. Lab. Sci., 2004, 34, 299.
11. A. D. Tinoco, E. V. Eames and A. M. Valentine, J. Am. Chem. Soc., 2008, 130, 2262–2270.
12. M. Guo, H. Sun, H. J. McArdle, L. Gambling and P. J. Sadler, Biochemistry, 2000, 39, 10023–10033.
13. P. Korovessis, G. Petsinis, M. Repanti and T. Repantis, J Bone Jt Surg., Am. Vol., 2006, 88, 1183–1191.
14. P. Drees, A. Eckardt, R. Gay, S. Gay and L. Huber, Nat. Clin. Pract. Rheumatol., 2007, 3, 165–171.
15. H. P. von Schroeder, D. C. Smith, A. E. Gross, R. M. Pilliar, R. A. Kandel, R. Chernecky and S. J. Lugowski, The Journal of Arthroplasty, 1996, 11, 620–625.
16. M. W. Elves, J. N. Wilson, J. T. Scales and H. B. Kemp, Br. Med. J., 1975, 4, 376–378.
17. K. E. Müller and E. Valentine-Thon, Neuroendocrinol. Lett., 2007, 27(Suppl 1), 31–35.
18. P. A. Lalor, P. A. Revell, A. B. Gray, S. Wright, G. T. Railton and M. A. Freeman, J. Bone Jt Surg., Br. Vol., 1991, 73, 25–28.
19. R. J. Langford and J. W. Frame, Journal of Cranio-Maxillo-Facial Surgery, Official Publication of the European Association for Cranio-Maxillo-Facial Surgery, 2002, 30, 103–107.
20. A. G. Gristina, Clin. Orthop. Relat. Res., 1994, 106–118.
21. N. J. Hallab, M. Caicedo, A. Finnegan and J. J. Jacobs, J. Orthop. Surg. Res., 2008, 3, 6.
22. V. Stejskal, R. Hudecek, J. Stejskal and I. Sterzl, Neuroendocrinol. Lett., 2007, 27(Suppl 1), 7–16.
23. K. Shortman and S. H. Naik, Nat. Rev. Immunol., 2007, 7, 19–30.
24. R. M. Steinman, Annu. Rev. Immunol., 1991, 9, 271–296.
25. J. Banchereau, F. Briere, C. Caux, J. Davoust, S. Lebecque, Y. J. Liu, B. Pulendran and K. Palucka, Annu. Rev. Immunol., 2000, 18, 767–811.
26. J. K. Tan and H. C. O'Neill, J. Leuk. Biol., 2005, 78, 319–324.
27. K. Roelofs-Haarhuis, X. Wu, M. Nowak, M. Fang, S. Artik and E. Gleichmann, J. Immunol., 2003, 171, 2863–2872.
28. A. Lanzavecchia and F. Sallusto, Curr. Opin. Immunol., 2001, 13, 291–298.
29. A. Langenkamp, M. Messi, A. Lanzavecchia and F. Sallusto, Nat. Immunol., 2000, 1, 311–316.
30. M. Thewes, R. Kretschmer, M. Gfesser, J. Rakoski, M. Nerlich, S. Borelli and J. Ring, Arch. Orthop. Trauma Surg., 2001, 121, 223–226.
31. W. A. High, R. A. Ayers, J. R. Adams, A. Chang and J. E. Fitzpatrick, J. Am. Acad. Dermatol., 2006, 55, 716–720.
32. M. Degidi, L. Artese, A. Scarano, V. Perrotti, P. Gehrke and A. Piattelli, J. Periodontology, 2006, 77, 73–80.
33. E. M. Evans, M. A. Freeman, A. J. Miller and B. Vernon-Roberts, J. Bone Jt Surg., Br. Vol., 1974, 56–B, 626–642.
34. Y. Z. Chen, N. Shah, F. E. Huggins and G. P. Huffman, Aerosol Sci. Technol., 2005, 39, 509–518.
35. N. Kapp, W. Kreyling, H. Schulz, V. Im Hof, P. Gehr, M. Semmler and M. Geiser, Microsc. Res. Tech., 2004, 63, 298–305.
36. M. Diociaiuti, J. Electron Spectrosc. Relat. Phenom., 2005, 189–203.
37. M. Geiser, B. Rothen-Rutishauser, N. Kapp, S. Schurch, W. Kreyling, H. Schulz, M. Semmler, V. I. Hof, J. Heyder and P. Gehr, Environ. Health Perspect., 2005, 113, 1555–1560.
38. M. Larsen, T. Thingholm, O. Jensen, P. Roepstorff and T. Jørgensen, Mol. Cell. Proteomics, 2005, 4, 873–886.
39. Y. Ishihama, F. Wei, K. Aoshima, T. Sato, J. Kuromitsu and Y. Oda, J. Proteome Res., 2007, 6, 1139–1144.
40. C. Klemm, S. Otto, C. Wolf, R. F. Haseloff, M. Beyermann and E. Krause, J. Mass Spectrom., 2006, 41, 1623–1632.
41. S. J. Kang, B. M. Kim, Y. J. Lee and H. W. Chung, Environ. Mol. Mutagen., 2008, 49(5), 399–405.
42. A. L. Zarling, S. B. Ficarro, F. M. White, J. Shabanowitz, D. F. Hunt and V. H. Engelhard, J. Exp. Med., 2000, 192, 1755–1762.
43. G. S. Schuster and G. B. Caughman, J. Biomed. Mater. Res., Part A, 2004, 70, 347–353.
44. T. Imai, M. Baba, M. Nishimura, M. Kakizaki, S. Takagi and O. Yoshie, J. Biol. Chem., 1997, 272, 15036–15042.
45. T. Imai, D. Chantry, C. J. Raport, C. L. Wood, M. Nishimura, R. Godiska, O. Yoshie and P. W. Gray, J. Biol. Chem., 1998, 273, 1764–1768.
46. S. Sozzani, P. Allavena, A. Vecchi and A. Mantovani, J. Leuk. Biol., 1999, 66, 1–9.
47. F. Nawaz and B. M. Wall, Am. J. Med. Sci., 2007, 334, 215–218.
48. P. Thomas, H. Gollwitzer, S. Maier and F. Rueff, Contact Dermatitis, 2006, 54, 222–225.
49. R. Kubba, J. S. Taylor and K. E. Marks, Arch. Dermatol. Res., 1981, 117, 554–560.
50. D. Granchi, E. Cenni, D. Tigani, G. Trisolino, N. Baldini and A. Giunti, Biomaterials, 2008, 29, 1494–1500.
51. D. Granchi, E. Cenni, G. Trisolino, A. Giunti and N. Baldini, J. Biomed. Mater. Res., Part B, 2005, 77, 257–264.
52. G. J. Nohynek, J. Lademann, C. Ribaud and M. S. Roberts, Crit. Rev. Toxicol., 2007, 37, 251–277.
53. E. S. Nakamura, K. Koizumi, M. Kobayashi, Y. Saitoh, Y. Arita, T. Nakayama, H. Sakurai, O. Yoshie and I. Saiki, Clin. Exp. Metastasis, 2006, 23, 9–18.
54. F. Katou, H. Ohtani, T. Nakayama, K. Ono, K. Matsushima, A. Saaristo, H. Nagura, O. Yoshie and K. Motegi, Am. J. Pathol., 2001, 158, 1263–1270.
55. M. Vulcano, C. Albanesi, A. Stoppacciaro, R. Bagnati, G. D'Amico, S. Struyf, P. Transidico, R. Bonecchi, A. Del Prete, P. Allavena, L. P. Ruco, C. Chiabrando, G. Girolomoni, A. Mantovani and S. Sozzani, Eur. J. Immunol., 2001, 31, 812–822.
56. D. P. Andrew, N. Ruffing, C. H. Kim, W. Miao, H. Heath, Y. Li, K. Murphy, J. J. Campbell, E. C. Butcher and L. Wu, J. Immunol., 2001, 166, 103–111.
57. P. T. Yang, H. Kasai, L. J. Zhao, W. G. Xiao, F. Tanabe and M. Ito, Clin. Exp. Immunol., 2004, 138, 342–347.
58. S. Hashimoto, K. Nakamura, N. Oyama, F. Kaneko, Y. Tsunemi, H. Saeki and K. Tamaki, J. Dermatol. Sci., 2006, 44, 93–99.
59. M. Kusumoto, B. Xu, M. Shi, T. Matsuyama, K. Aoyama and T. Takeuchi, J. Interferon Cytokine Res., 2007, 27, 901–910.
60. D. Cadosh & M. Sutanto, E. Chan and L. Filgueira, submitted for Publication, 2008.

---